--- Log opened Tue Oct 01 00:00:59 2019 2019-10-01T00:04:14 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 268 seconds] 2019-10-01T00:05:33 < zyp> machinehum, yes, that's what a function frame is 2019-10-01T00:06:37 < zyp> and since it's cooperative multitasking, it's up to your functions to hand over program control when they don't have anything to do 2019-10-01T00:10:02 < zyp> the advantage with coroutines over multithreading is that you don't have to create thread stacks, you only need one main stack for the normal functions 2019-10-01T00:11:09 < zyp> only coroutines can yield, so only coroutines can exist in parallel, so only coroutine frames needs to be allocated in parallel, all the other functions (even when called from a coroutine) can be stacked on the same stack 2019-10-01T00:11:33 < zyp> because when a coroutine calls a function, it can't yield before the function has returned 2019-10-01T00:13:04 < kakimir32> DAC fixed 2019-10-01T00:16:59 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-01T00:17:17 < karlp> I always find coroutine code hard to read/write, but maybe I'm just not pro enough 2019-10-01T00:17:44 < zyp> compared to what? 2019-10-01T00:18:18 < karlp> how i normally do things.... 2019-10-01T00:18:28 < zyp> how do you normally do async? 2019-10-01T00:18:38 < karlp> state machine 2019-10-01T00:18:48 < karlp> don't get me wrong, I hate them too, 2019-10-01T00:18:55 < karlp> I've _tried_ to use coroutines 2019-10-01T00:19:06 < karlp> I just can't get it to stick in my head well. 2019-10-01T00:19:22 < karlp> I should restate, coroutines I've found easy to read, once they're working 2019-10-01T00:19:27 < karlp> but hard to wirte until thhey are. 2019-10-01T00:19:54 -!- Luggi09 is now known as Lux 2019-10-01T00:20:00 < zyp> IMO the point of coroutines is that it writes like blocking code, hiding away all the complexity of doing async 2019-10-01T00:20:55 < zyp> you don't even have to think nonblocking except for the edges where it touches non-coroutine code 2019-10-01T00:21:20 < zyp> and the future/promise-concept simplifies a lot of that too 2019-10-01T00:21:36 < karlp> well, I'll try again sometime, I jsut haven't been as happy with it as I'd hoped. 2019-10-01T00:21:48 < karlp> I forced myself to do something using them, and was hoping it would enlightenme. 2019-10-01T00:22:11 < zyp> let me show you something cool 2019-10-01T00:25:41 < zyp> https://paste.jvnv.net/view/rOIHf <- first I made a little function decorator like this 2019-10-01T00:26:08 < zyp> https://paste.jvnv.net/view/ZP8xz <- which lets me write functions like this, and bind them directly to qt signals 2019-10-01T00:27:14 < zyp> https://paste.jvnv.net/view/7zZDW <- i.e. this 2019-10-01T00:29:21 -!- sterna [~Adium@c-f8b9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-01T00:30:49 < zyp> using pyqt for gui, quamash to tie asyncio into the qt event loop 2019-10-01T00:31:42 < Cracki> it's like monads. a proper explanation seems to be unoptainium 2019-10-01T00:32:15 < Cracki> recipes? sure, lots of them. but the principles behind it all? good luck 2019-10-01T00:32:22 < zyp> end result is that I can write functions as they were blocking and trigger them directly from qt signals, and still have them do their stuff without making the gui unresponsive 2019-10-01T00:33:03 < Cracki> all this stuff involves a kind of coroutine scheduler/trampoline that does some kind of message passing 2019-10-01T00:33:17 < jadew> how is that blue gunk that is put on screws and PCB pots to lock them in place? 2019-10-01T00:33:27 < zyp> loctite? 2019-10-01T00:33:35 < jadew> isn't that a company? 2019-10-01T00:33:46 < Cracki> like xerox isn't all copier machines 2019-10-01T00:33:58 < jadew> no... I mean they make a lot of stuff 2019-10-01T00:34:08 < jadew> including solder paste if I'm not mistaken 2019-10-01T00:34:21 < jadew> that gc 10 stuff, no? 2019-10-01T00:34:25 < Cracki> loctite is their original product 2019-10-01T00:34:38 < Cracki> "lock tight" 2019-10-01T00:34:50 < jadew> right, that makes sense 2019-10-01T00:35:25 < zyp> also, for completion, here's the other edge of the coroutine code: https://paste.jvnv.net/view/AEYwy 2019-10-01T00:35:29 < Cracki> >asyncio.get_event_loop() 2019-10-01T00:35:30 < Cracki> magic 2019-10-01T00:36:06 < jadew> it looks loctite it is 2019-10-01T00:36:09 < jadew> loctite 242 2019-10-01T00:36:18 < zyp> Cracki, they have a global event loop, set earlier by asyncio.set_event_loop() 2019-10-01T00:36:20 < jadew> 290 and 271 are also used but they're stronger 2019-10-01T00:36:36 < zyp> easier to just use that than pass around a reference to the actual loop object when I only have one anyway 2019-10-01T00:36:46 < Cracki> true 2019-10-01T00:37:00 < Cracki> in what thread does this event loop run 2019-10-01T00:37:03 < zyp> I figure the only reason you'd want to have more than one is if you were using threads as well, with multiple loops in different threads 2019-10-01T00:37:12 < Cracki> this feels equivalent to spawning a proper thread 2019-10-01T00:37:33 < Cracki> qt has an event loop per thread anyway 2019-10-01T00:37:40 < zyp> well, like I said I use quamash to integrate with the qt event loop, so I assume it runs in the main thread along with the gui 2019-10-01T00:38:18 < zyp> and yes, like I said, coroutines can do a lot of stuff that proper threads can, without all the hassle of spawning proper threads 2019-10-01T00:39:31 < Cracki> does python even have a traditional stack or are all stack frames living in the heap and it's really a kind of DAG 2019-10-01T00:39:51 < Cracki> a tree even 2019-10-01T00:40:08 < zyp> I believe it's using a traditional stack 2019-10-01T00:40:27 < zyp> coroutines are defined with the async keyword 2019-10-01T00:41:06 < Cracki> I looked into what exactly that keyword does, because python's stdlib async stuff is a layer on top of the async keyword 2019-10-01T00:41:14 < zyp> although it also has generators which are not defined with a special keyword, they are just autodetected during compilation because they use yield rather than return 2019-10-01T00:41:21 < Cracki> it's very hard to find a proper explanation of this mechanism 2019-10-01T00:41:39 < zyp> are you familiar with generators? 2019-10-01T00:41:42 < Cracki> yes 2019-10-01T00:41:49 < zyp> did you ever use tornado? 2019-10-01T00:41:53 < Cracki> and I read an explanation that talked about how this grew out of those 2019-10-01T00:41:55 < Cracki> hell no 2019-10-01T00:42:10 < zyp> tornado build a whole similar framework on python 2 on top of generators 2019-10-01T00:42:50 < zyp> the new stuff does the same, just with better syntax and support code in the standard library 2019-10-01T00:43:37 < Cracki> is this supposed to have any advantages to spawning threads, apart from using fewer resources? 2019-10-01T00:43:52 < Cracki> because I find spawning a thread and communicating between threads to be quite natural 2019-10-01T00:44:08 < Cracki> and this has the same behavior: asyncio.get_event_loop().create_task(coro) 2019-10-01T00:44:11 < Cracki> just implicitly 2019-10-01T00:44:21 < Cracki> and it also involves a scheduler 2019-10-01T00:44:22 < karlp> some people don't like the methods for synbchnchronisng data between threads 2019-10-01T00:45:14 < Cracki> I get the "fewer resources" argument, which is a major one when you have thousands or more of some concurrent thing 2019-10-01T00:45:59 < zyp> the way I see it, coroutines are essentially small lightweight threads, but the way you use them are different from traditional threads 2019-10-01T00:46:26 < Cracki> I'll bookmark this quamash thing. might come in handy. 2019-10-01T00:47:14 < zyp> e.g. most coroutines are started to do one thing and then (asynchronously) return a result 2019-10-01T00:47:54 < Cracki> promise/future, yes 2019-10-01T00:48:00 < zyp> create_task() is not how most coroutines are started, most are started like «x = await give_me_some_data()» 2019-10-01T00:48:21 < zyp> which in turn might invoke another coroutine to do something 2019-10-01T00:48:34 < zyp> and so on, until you hit another primitive in the other end of the async pipeline 2019-10-01T00:48:47 < zyp> like the recv_queue I got in the last paste 2019-10-01T00:48:49 < Cracki> the question I have is whether await does a blocking wait on some promise/future, or whether it causes the caller to be an async function too and be put on the scheduler's blocked queue 2019-10-01T00:49:29 < Cracki> the observable behavior is the same, but the mechanism is different 2019-10-01T00:49:32 < zyp> await can only be used in a coroutine, it suspends execution of the coroutine until the future has a result 2019-10-01T00:49:57 < zyp> allowing the scheduler to run other coroutines 2019-10-01T00:50:02 < Cracki> the notion of "suspend" is an abstraction I'd like to see into 2019-10-01T00:50:41 < Cracki> is await something like yield/yield-from? 2019-10-01T00:51:13 < zyp> or in my case, it allows the qudpsocket to fire its signal whose handler fills my recv_queue that my stack of coroutines are awaiting the result from 2019-10-01T00:51:16 < zyp> yes 2019-10-01T00:51:28 < zyp> await is basically syntactical sugar for yield from 2019-10-01T00:51:33 < Cracki> that's a mechanism I understand. the scheduler would be spinning these generators, and whatever they yield to the scheduler is something the scheduler has to test for it being runnable or complete 2019-10-01T00:51:46 < zyp> it works the same but await makes more sense when thinking async 2019-10-01T00:53:02 < Cracki> the usually awkward and insufficient explanations make me homicidal. same goes for fucking monads. 2019-10-01T00:53:03 < Cracki> >A monad is just a monoid in the category of endofunctors, what's the problem? 2019-10-01T00:53:15 < Cracki> only redditors say such things 2019-10-01T00:53:50 < Cracki> thank you for the await <> yield from point 2019-10-01T00:55:01 < zyp> https://paste.jvnv.net/view/dF0Cf <- here's some old tornado-based code I once wrote 2019-10-01T00:56:00 * karlp had to re-rad the log again. <> is != for me, for too long... :) 2019-10-01T00:57:31 < Cracki> all the usual explanations gloss over the "plumbing" (as someone put it on SO) that actually makes the thing work 2019-10-01T00:57:52 < Cracki> i.e. I'd probably have to dig into exactly what tornado, or python's async support module, actually do 2019-10-01T00:58:31 < Cracki> it's like driving school where they try to explain "clutch" without explaining the mechanics that happen behind the pedal 2019-10-01T01:00:45 < Cracki> thx for the tornado code. the first few lines (5-7) already make things MUCH clearer. it's obvious that the scheduler gets a future and only "resumes" (nexts on) when that future has a result 2019-10-01T01:00:47 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-01T01:06:24 < zyp> yep 2019-10-01T01:08:30 < zyp> https://paste.jvnv.net/view/FFPGb <- here I'm creating futures explicitly and awaiting them 2019-10-01T01:11:46 < zyp> hrm, I'm trying to revive some code I wrote a couple of years ago, and one of the libs I'm using has changed its API 2019-10-01T01:12:18 < karlp> progress == backwards incompatible changes.... 2019-10-01T01:12:39 < zyp> yeah, I don't mind that so much 2019-10-01T01:13:24 < zyp> the problem is that they've apparently removed a feature I'm using without a good replacement 2019-10-01T01:13:54 < Steffanx> Must have been bad code 2019-10-01T01:14:13 < zyp> https://construct.readthedocs.io/en/latest/api/core.html#construct.core.EmbeddedSwitch 2019-10-01T01:14:28 < zyp> «Macro that simulates embedding Switch, which under new embedding semantics is not possible.» 2019-10-01T01:14:43 < zyp> it's a poor simulation, it doesn't do what I want 2019-10-01T01:15:50 < zyp> https://paste.jvnv.net/view/802J8 <- my old zwave code consists of two levels of embedded switches, since all messages firt starts with a command class, followed by a command, followed by command-specific arguments 2019-10-01T01:19:59 < Cracki> i'd look up what they mean by "embedded" and what the alternatives are 2019-10-01T01:20:33 < Cracki> your use case sounds common. they should have something sane to handle that 2019-10-01T01:20:39 < zyp> embedded means that the structs get flattened 2019-10-01T01:21:27 < zyp> i.e. when decoding a message I get something like {'command_class': 'SWITCH_BINARY', 'command': 'SET', 'value': 1} 2019-10-01T01:22:27 < zyp> instead of something like {'command_class': 'SWITCH_BINARY', {'command': {'command': 'SET', 'arguments': {'value': 1}}} 2019-10-01T01:29:14 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-01T01:30:12 < zyp> hmm, I think I've figured out something that should work 2019-10-01T01:42:08 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 276 seconds] 2019-10-01T01:44:15 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Read error: Connection reset by peer] 2019-10-01T01:48:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-01T02:01:50 < kakimir32> is there a counter for user logins? 2019-10-01T02:01:55 < kakimir32> in linux 2019-10-01T02:03:48 < Cracki> never heard of such a thing. last login time, yes. maybe systemd has invented something, maybe you can grep/awk some logs? 2019-10-01T02:04:01 < kakimir32> sounds nasty 2019-10-01T02:09:46 < Cracki> systemd-logincounterd 2019-10-01T02:11:00 < Cracki> TIL: our audimax is still audimax even after the new lecture hall complex. they must have intentionally designed all the new halls to stay _slightly_ under the seat count of the old audimax 2019-10-01T02:11:21 < karlp> audimax? 2019-10-01T02:15:28 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T02:17:45 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T02:25:14 < dongs> Uni-ABC: Was ist das Audimax? | Hier die Erklärung 2019-10-01T02:25:26 < dongs> sounds like some german shit for max number of people in auditorium 2019-10-01T02:25:37 < dongs> oh wait no 2019-10-01T02:25:40 < dongs> its an actual location 2019-10-01T02:25:50 < dongs> fuckin germans and their shitty names 2019-10-01T02:34:16 < Cracki> auditorium maximum, largest lecture hall 2019-10-01T02:34:21 < Cracki> easier than naming it 2019-10-01T02:34:50 < Cracki> had no idea that it's only a thing in... LATIN SPEAKING COUNTRIES 2019-10-01T02:35:06 < Cracki> maybe muricans have a greek-derived name for it 2019-10-01T02:35:29 < zyp> my old uni also had an aud-max 2019-10-01T02:35:50 < Cracki> --d-m-x 2019-10-01T02:36:11 < zyp> not sure the new one got it 2019-10-01T02:36:13 < Cracki> purely coincidentally, audi the car maker has a presence here 2019-10-01T02:37:05 < Cracki> nvm, just institutes cooperating with them 2019-10-01T02:37:51 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T02:39:59 < zyp> hmm, the map for the new uni just says C2036 for the big auditorium 2019-10-01T02:40:12 < zyp> building C, floor 2, room 36 2019-10-01T02:42:30 < Cracki> lol 2019-10-01T02:42:49 < kakimir32> https://github.com/ReFirmLabs/binwalk 2019-10-01T02:43:00 < Cracki> ours has retarded numbers too. four digits for each building, with no particular scheme apart from them all looking like basic line numbers (xxx0 and xxx5) 2019-10-01T02:45:22 < Cracki> wtf it's a cult https://pbs.twimg.com/media/EFu4EaTUEAAlJ8m.jpg:orig 2019-10-01T02:45:49 < Cracki> (yes it's no credible source :P) 2019-10-01T02:46:17 < kakimir32> swedes 2019-10-01T02:48:09 < kakimir32> https://www.youtube.com/watch?v=GIU4yJn2-2A 2019-10-01T02:48:27 < kakimir32> interesting firmware reversing 101 2019-10-01T02:49:38 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T02:52:35 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-01T02:56:27 -!- Streaker [~Streaker@unaffiliated/streaker] has joined ##stm32 2019-10-01T02:56:38 < kakimir32> looks simple 2019-10-01T02:58:17 -!- Streake_ [~Streaker@unaffiliated/streaker] has quit [Ping timeout: 240 seconds] 2019-10-01T02:59:48 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-01T03:03:41 -!- [1]MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-01T03:04:41 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 276 seconds] 2019-10-01T03:05:07 < kakimir32> https://www.youtube.com/watch?v=4urMITJKQQs what is that codebrowser thing they use to generate code out of binary? 2019-10-01T03:05:35 < kakimir32> at 8:00 and most of the video too 2019-10-01T03:05:39 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-01T03:05:39 -!- [1]MrMobius is now known as MrMobius 2019-10-01T03:11:48 < rajkosto> its in the video title... 2019-10-01T03:11:49 < rajkosto> Ghidra 2019-10-01T03:11:50 < kakimir32> A software reverse engineering (SRE) suite of tools developed by NSA’s Research Directorate in support of the Cybersecurity mission. 2019-10-01T03:12:22 < rajkosto> its a worse version of something we had for ages: IDA pro 2019-10-01T03:12:27 < rajkosto> but its free and extensible 2019-10-01T03:13:09 < kakimir32> this all looks fun 2019-10-01T03:13:59 < kakimir32> how much is ida pro licence? 2019-10-01T03:14:07 < rajkosto> l o d s o e m o n e 2019-10-01T03:14:37 < kakimir32> a pile of moneys? 2019-10-01T03:16:56 < rajkosto> like 1500$ for a single processor type 2019-10-01T03:17:01 < rajkosto> theres a pricing sheet on their website 2019-10-01T03:18:11 < kakimir32> sweet 2019-10-01T03:30:28 < Cracki> tyvm kakimir32 good video 2019-10-01T03:31:42 < Cracki> codebrowser is https://www.zdnet.com/article/nsa-release-ghidra-a-free-software-reverse-engineering-toolkit/ 2019-10-01T03:31:54 < Cracki> oh, was discovered already :D 2019-10-01T03:32:18 < Cracki> nobody has ever bought an ida license 2019-10-01T03:37:19 -!- Laurenceb_ [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-01T03:38:00 < rajkosto> mah boi jiang yang has 2019-10-01T03:38:03 < rajkosto> god damn it jiang yang 2019-10-01T03:40:53 < kakimir32> by looking the icon theme and daily tips window this tool was not recent innovation of NSA 2019-10-01T03:45:43 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-01T03:48:01 < catphish> sometimes C is exhausting 2019-10-01T03:48:03 < catphish> floppy.c:188:58: error: '*ordered_sectors[n]' is a pointer; did you mean to use '->'? 2019-10-01T03:48:03 < catphish> usb_write_packet(0x82, ordered_sectors[n]->raw + m, 64); 2019-10-01T03:48:28 < catphish> yes, compiler, i did mean to use -> so i did 2019-10-01T03:50:05 < Laurenceb_> CockWombles!! 2019-10-01T03:50:15 < Laurenceb_> today I visited Reddit 2019-10-01T03:50:25 < Laurenceb_> BaD IdEa 2019-10-01T03:50:54 < Laurenceb_> its like the worst aspects of every other webshite without any of the positive bits 2019-10-01T03:51:27 < catphish> ah more levels of pointer there then i ment to make :) 2019-10-01T03:51:37 < catphish> i'm completely addicted to reddit :( 2019-10-01T03:53:04 < Laurenceb_> maybe there are some ok subreddits hiding 2019-10-01T03:53:19 < Laurenceb_> MaYbE 2019-10-01T03:54:00 < Cracki> redditor? remove yourself from society. 2019-10-01T03:55:17 < Laurenceb_> lmao 2019-10-01T04:10:37 < roomba> no need. #cancelLaurence 2019-10-01T04:10:44 < roomba> boom! now he can't buy or sell. 2019-10-01T04:25:38 -!- Laurenceb_ [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-01T04:49:12 -!- Streaker [~Streaker@unaffiliated/streaker] has quit [Ping timeout: 245 seconds] 2019-10-01T05:07:37 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-01T05:33:00 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Disconnected by services] 2019-10-01T05:33:02 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-01T06:03:58 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-01T06:11:42 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-01T06:12:47 < R2COM> sup 2019-10-01T06:18:27 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-01T06:31:44 < dongs> continuous innovation 2019-10-01T06:44:45 < R2COM> thats good 2019-10-01T06:44:53 -!- fc5dc9d4_ [~quassel@p5B0813FA.dip0.t-ipconnect.de] has joined ##stm32 2019-10-01T06:44:59 < R2COM> im almost ready to hook up finally BLDC circuit to my F0 2019-10-01T06:45:03 < R2COM> all timing looks proper 2019-10-01T06:48:47 -!- fc5dc9d4 [~quassel@p5B08148A.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 2019-10-01T06:50:32 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-01T07:07:32 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 245 seconds] 2019-10-01T07:41:01 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-01T07:44:50 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-01T07:44:50 -!- day__ is now known as day 2019-10-01T08:24:08 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T08:34:26 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T08:49:05 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T08:50:19 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T08:59:48 -!- emeryth [emeryth@2a0d:eb00:2137:2::10] has quit [Ping timeout: 252 seconds] 2019-10-01T09:11:08 -!- sterna [~Adium@c-f8b9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-01T09:13:13 -!- emeryth [emeryth@boston-packets.hackerspace.pl] has joined ##stm32 2019-10-01T09:14:06 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T09:18:17 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-01T09:20:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-01T09:22:06 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has left ##stm32 [] 2019-10-01T09:24:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-01T09:24:37 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-01T09:27:13 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T09:39:13 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T09:42:53 -!- sterna [~Adium@c-f8b9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-01T09:47:34 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T09:56:52 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-01T10:14:59 -!- sterna [~Adium@79.136.90.120] has joined ##stm32 2019-10-01T10:15:01 -!- sterna [~Adium@79.136.90.120] has quit [Client Quit] 2019-10-01T10:44:48 < Cracki> supercomputer that can compete in the top500 2019-10-01T10:44:51 < Cracki> just in case 2019-10-01T10:45:20 < Cracki> most expensive fpga you can find 2019-10-01T10:45:40 < Cracki> calculate 1000 code monkeys for 5 years 2019-10-01T10:46:09 < Cracki> there's got to be limits 2019-10-01T11:05:18 -!- jly [uid355225@gateway/web/irccloud.com/x-upvfrtxunzbtjjgp] has joined ##stm32 2019-10-01T11:30:56 < dongs> fucking libusb garbage 2019-10-01T11:31:21 < dongs> my device is composite bulk/hid and niggerusb tries to open HID one with no way to open bulk first 2019-10-01T11:32:16 < Steffanx> This is scary, jly 2019-10-01T11:32:34 < jly> sounds all fucked up 2019-10-01T11:32:37 < Steffanx> Yea 2019-10-01T11:39:54 < BrainDamage> if you don't know what's going to do then write software first then design the hardware 2019-10-01T11:40:14 < BrainDamage> software revisions are much easier compared to hw revisions 2019-10-01T11:40:42 < jly> You're the problem! You're the fucking problem you fucking Dr White honkin' jam-rag fucking spunk-bubble! I'm telling you Aitch you keep looking at me I'll put you in the fucking ground, promise you! 2019-10-01T11:46:08 < zyp> dongs, are composite devices showing up as more than one device in libusb? 2019-10-01T11:46:52 < dongs> no 2019-10-01T11:47:04 < zyp> then what's the problem? 2019-10-01T11:47:10 < dongs> it doesnt open. 2019-10-01T11:47:18 < dongs> unless I put both hid and bulk as winusb driver 2019-10-01T11:47:36 < dongs> im stepping through niggerusb sores right now to see what its fucking up 2019-10-01T11:48:24 < dongs> https://i.imgur.com/qXSmype.png 2019-10-01T11:48:26 < zyp> as in libusb_open() fails? 2019-10-01T11:48:29 < dongs> the #0 is the one i want 2019-10-01T11:48:29 < dongs> yes 2019-10-01T11:48:36 < dongs> but it tries to open #1 2019-10-01T11:49:09 < zyp> are you using libusb_open_device_with_vid_pid()? 2019-10-01T11:49:16 < zyp> or libusb_open()? 2019-10-01T11:49:19 < dongs> the latter. 2019-10-01T11:49:25 < dongs> because i need to do multiple devices. 2019-10-01T11:49:28 < dongs> (at some point later) 2019-10-01T11:50:01 < zyp> and the device only shows up once in libusb_get_device_list()? 2019-10-01T11:50:28 < dongs> yes 2019-10-01T11:50:39 < dongs> the libusb open failure error is hid-related 2019-10-01T11:50:46 < dongs> i dont haev it now cuz im stepping, lemme fail this oen and see 2019-10-01T11:51:51 < zyp> normally there's no restrictions on opening devices, but libusb_claim_interface() will fail if another driver is bound to it 2019-10-01T11:53:10 < dongs> if (available[SUB_API_MAX]) { // HID driver 2019-10-01T11:53:11 < dongs> found it 2019-10-01T11:53:15 < dongs> i think im gonna have to just trash this 2019-10-01T11:53:40 < dongs> libusb: warning [hid_open] could not open HID device in R/W mode (keyboard or mouse?) - trying without 2019-10-01T11:53:43 < dongs> libusb: error [hid_open] could not open device \\.\USB#VID_14AC&PID_0008#20190530#{A5DCBF10-6530-11D2-901F-00C04FB951ED} (interface 1): [2] The system cannot find the file specified. 2019-10-01T11:54:07 < dongs> // On Windows 10 version 1903 (OS Build 18362) and later Windows blocks attempts to 2019-10-01T11:54:10 < dongs> // open HID devices with a U2F usage unless running as administrator. We ignore this 2019-10-01T11:54:13 < dongs> // failure and proceed without the HID device opened. 2019-10-01T11:54:16 < dongs> what 2019-10-01T11:54:16 < dongs> "ignore" it doesnt actually lmao 2019-10-01T11:55:30 < dongs> switch (GetLastError()) { 2019-10-01T11:55:31 < dongs> case ERROR_FILE_NOT_FOUND: // The device was disconnected 2019-10-01T11:55:32 < dongs> yeah 2019-10-01T11:55:41 < dongs> it gets a file_not_found, because it does the check for access_denid later on 2019-10-01T11:55:51 < dongs> the 1st one fails wiht access denid, and it loses that error 2019-10-01T11:55:56 < dongs> so getlasterror returns garbage 2019-10-01T11:56:05 < dongs> fuckin opensores 2019-10-01T11:58:59 < zyp> good opportunity to submit a patch and troll the AUTHORS list 2019-10-01T11:59:17 < zyp> libusb/AUTHORS is still one of the top hits when I google my own name 2019-10-01T12:00:31 < Sadale> It's obviously a dumb idea to reuse a toaster/oven that had previous used for reflow soldering for food. However, I'm thinking if it'd be ok to reuse a hot plate. 2019-10-01T12:00:57 < dongs> libusb_open_device_with_vid_pid(devh) = 00BE05E8 2019-10-01T12:00:57 < dongs> nice 2019-10-01T12:00:58 < dongs> works 2019-10-01T12:01:12 < dongs> honestly i think the real fix is having a fucking option to NOT try opening HID 2019-10-01T12:01:23 < dongs> because what fucking use is there for acomposite device with hid and something else AND libusb? 2019-10-01T12:01:41 < dongs> its not like youre goinna be doing hid wiht libusb anyway 2019-10-01T12:35:55 < Steffanx> Ah he is. Lol. 2019-10-01T12:36:05 < Steffanx> He didnt even troll it 2019-10-01T12:36:42 < Steffanx> Zyp should only use his middle name imho 2019-10-01T12:36:55 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Read error: Connection reset by peer] 2019-10-01T12:38:57 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-01T12:43:30 < Steffanx> Awh 2019-10-01T12:44:45 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection] 2019-10-01T12:51:03 -!- Laurenceb_ [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-01T12:53:11 < Steffanx> What about the duck? 2019-10-01T12:53:33 < dongs> lol libusb still supports solaris 2019-10-01T12:53:38 < dongs> i bet they have a total of like 2 users 2019-10-01T12:53:43 < dongs> what the fuck retarded opensores 2019-10-01T12:54:08 < dongs> just like conigure checks if you're on a fuckign VAX or someshit 2019-10-01T12:54:11 < dongs> so much aids 2019-10-01T12:57:00 < Steffanx> I can do it for you 2019-10-01T12:59:26 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 276 seconds] 2019-10-01T13:00:01 -!- Mangy_Dog [~Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-01T13:01:28 < zyp> Haohmaru, «author» 2019-10-01T13:01:39 < zyp> I submitted a patch of a few lines once some years back 2019-10-01T13:01:43 -!- Laurenceb_ [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-01T13:02:12 < dongs> hmm 2019-10-01T13:02:21 < dongs> ok so the opening resolved 2019-10-01T13:02:24 < dongs> remaining shit is 2019-10-01T13:02:35 < dongs> bulk stops flowing unless I also run control transfers in anothe thread 2019-10-01T13:02:37 < dongs> fucking why? 2019-10-01T13:02:46 < dongs> never been an issue with actual winusb/cyapi 2019-10-01T13:03:30 < zyp> dongs, how are you making libusb handle events? 2019-10-01T13:03:39 < dongs> anothe thread 2019-10-01T13:04:10 < zyp> running libusb_handle_events()? 2019-10-01T13:04:14 < dongs> yes 2019-10-01T13:04:22 < dongs> while (run) libusb_handle_events(session->ctx); 2019-10-01T13:04:36 < dongs> i tried various combinations of it and they were all same 2019-10-01T13:05:59 < zyp> and you're using the async api? 2019-10-01T13:06:03 < dongs> yea 2019-10-01T13:06:11 < dongs> the onyl way to do > 100mbit transfers anyway 2019-10-01T13:06:15 < dongs> i queue 32 requests 2019-10-01T13:06:27 < dongs> (lol at libusb shitting itself over 60 of them or something, some "windows" select() limitation 2019-10-01T13:06:55 < dongs> if only 2019-10-01T13:08:04 < dongs> while (!completed) 2019-10-01T13:08:04 < dongs> libusb_handle_events_completed(session->ctx, &completed); 2019-10-01T13:08:08 < dongs> i tried this way too 2019-10-01T13:08:11 < dongs> that also failed 2019-10-01T13:08:34 < dongs> also set that threads priority to high 2019-10-01T13:11:29 < dongs> i don't realyl mind that I need to touch control, i could jsut run it in another thread, just seems kinda gay 2019-10-01T13:11:58 < dongs> especilaly since it was never a problem with cyapi or winusb direct 2019-10-01T13:13:32 < karlp> sweet package option: https://eenews.cdnartwhere.eu/sites/default/files/styles/inner_article/public/sites/default/files/images/2019-09-25-jh-silicon.png 2019-10-01T13:16:12 < dongs> wait waht 2019-10-01T13:16:16 < dongs> whats o specail about it 2019-10-01T13:16:27 < dongs> isnt that just a pcb with castellated vias 2019-10-01T13:16:37 < dongs> or perhaps a ceramic but I doubt it 2019-10-01T13:16:42 < englishman> oh great, a blackfin project 2019-10-01T13:17:17 < karlp> dongs: just liked the module being setup for both flat mount and socket/slot mount. 2019-10-01T13:18:16 < dongs> you mean the wider pads on the right side? 2019-10-01T13:18:21 < dongs> but i dont think it would be very useful in a socket 2019-10-01T13:18:27 < karlp> yeah, wit hthe cutout, it's for going into a board slot 2019-10-01T13:18:29 < dongs> cuz its castellated you'd only have 6 signals 2019-10-01T13:18:32 < karlp> I've seen esp modules like that too 2019-10-01T13:18:39 < karlp> it's enough for doing uart wifi shits. 2019-10-01T13:18:42 < karlp> 6 on each side remember 2019-10-01T13:18:43 < dongs> right 2019-10-01T13:18:45 < dongs> nope 2019-10-01T13:18:48 < dongs> both same 2019-10-01T13:18:57 < karlp> hrm, the castellations imply that right, 2019-10-01T13:18:58 < dongs> cuz its a via, the center shit connects top/bottom 2019-10-01T13:19:00 < dongs> yes 2019-10-01T13:19:01 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-01T13:19:06 < karlp> the espshits I've seen have been 6 on each side 2019-10-01T13:19:17 < karlp> enough for uart wifi crap still. 2019-10-01T13:19:24 < karlp> just not using any of the adc/periphs on it. 2019-10-01T13:19:39 < karlp> anyway, just liked the multi cut shape of it. 2019-10-01T13:19:44 < dongs> rite 2019-10-01T13:20:03 < dongs> but does anyone actualyl use this gecko shit in prodcution 2019-10-01T13:20:14 < karlp> isn't it what's in zyps ikea lights? 2019-10-01T13:20:14 < dongs> the only silabs products that arent aids are their USB serial shits 2019-10-01T13:20:50 < zyp> karlp, yup 2019-10-01T13:23:38 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-01T13:25:14 < englishman> hey, their lora stuff is great 2019-10-01T13:25:33 < karlp> just like lora.... 2019-10-01T13:30:34 < zyp> the ikea stuff is cool in that their modules aren't locked in any way, so I could just have reflashed it with a custom firmware 2019-10-01T13:30:37 < zyp> https://bin.jvnv.net/file/SdwDM.jpg 2019-10-01T13:31:24 < zyp> I was considering that, before I found out the radio registers aren't documented, and I'm not interested in spending time messing with a radio stack blob 2019-10-01T13:32:42 < zyp> on the other hand, dumping the flash over swd allowed me to find the network key of the zigbee network it was part of 2019-10-01T13:33:05 < zyp> wasn't hidden at all 2019-10-01T13:33:40 < dongs> looks like OSP coating 2019-10-01T13:33:58 < zyp> I think it's just the photo, the actual board looks like ENIG 2019-10-01T13:34:13 < dongs> wait why is zyp hacking shitty ikea lights 2019-10-01T13:34:33 < englishman> before he had his kid 2019-10-01T13:37:28 < zyp> dongs, not really hacking the lights, but I wanna make shit to talk to them 2019-10-01T13:37:53 < zyp> or rather, I wanna make shit that talks ZLL in general 2019-10-01T13:38:37 < zyp> thing is, the ikea lightbulbs are nice and cheap, so I got a bunch of them 2019-10-01T13:38:48 < zyp> but the remote controllers for them are not very fancy 2019-10-01T13:39:02 < zyp> so I wanna make some more flexible control devices 2019-10-01T13:42:19 < Steffanx> Dont you want to turn everything on at once? 2019-10-01T13:43:21 < zyp> well, the thing is that the ikea remotes don't do exactly that 2019-10-01T13:43:27 < zyp> they toggle everything at once 2019-10-01T13:43:57 < zyp> so if some lights are off and some are on in the same group (set via the gateway or whatever) then they all toggle if you hit the remote 2019-10-01T13:47:59 < Steffanx> I use home assistant mainly and the gateway 2019-10-01T13:54:05 < zyp> yeah, I mean in theory I could run the control devices on zwave or some other separate mesh, feed that into a central software that then controls the ZLL lights via the gateway 2019-10-01T13:55:15 < zyp> but then there would be two gateways and some central software between the lights and the control devices, i.e. several pieces that can fail and render me unable to control the lights 2019-10-01T13:55:48 < zyp> whereas ZLL control devices can talk directly to the lights without relying on anything central 2019-10-01T13:56:01 < zyp> I want that robustness 2019-10-01T14:22:14 < Steffanx> Do you even use it Haohmaru ? 2019-10-01T14:26:32 < zyp> Haohmaru, I also have normal light switches in most rooms 2019-10-01T14:26:42 < zyp> they are not very convenient 2019-10-01T14:27:53 < zyp> all the lamps here only came with led bulbs in the first place 2019-10-01T14:28:43 < zyp> but I'm planning to eventually replace them all with smart bulbs 2019-10-01T14:29:17 < zyp> partly for the dimming ability, partly for the remote control ability 2019-10-01T14:29:58 < zyp> so far I only put one in the bedroom, because that's where dimming is most useful 2019-10-01T14:30:39 < zyp> of course, I assume it's highfreq pwm in the led driver itself 2019-10-01T14:31:01 < zyp> wat 2019-10-01T14:31:22 < zyp> current control is a poor way of dimming leds 2019-10-01T14:31:50 < zyp> because when you vary led current, it'll affect the color it emits 2019-10-01T14:32:07 < zyp> and current/light output is not linear either 2019-10-01T14:32:34 < zyp> yeah, but why bother 2019-10-01T14:32:44 < zyp> also, how are you gonna vary current? 2019-10-01T14:32:58 < zyp> doing pwm and filtering to get DC? :p 2019-10-01T14:33:23 < zyp> pwm is simpler and better :p 2019-10-01T14:34:00 < zyp> oh, another benefit of the smart bulbs; they can control color/white balance 2019-10-01T14:34:08 < zyp> depending on which one you got 2019-10-01T14:34:48 < zyp> some do rgb, some have warmer and colder white leds, some only have one color of white 2019-10-01T14:35:01 < zyp> maybe 2019-10-01T14:35:13 < zyp> I've never really bothered with that 2019-10-01T14:36:02 < zyp> for most indoor photos, a decent flash takes out all lighting issues 2019-10-01T14:36:06 < englishman> the IKEA floalt panels look cool 2019-10-01T14:36:52 < zyp> why? by decent I mean one you can bounce off the ceiling 2019-10-01T14:37:19 < zyp> indoors? 2019-10-01T14:37:41 < zyp> so, uh, how are you setting up lighting without a flash then? 2019-10-01T14:38:08 < zyp> setup for a decent flash is basically «point it upwards» 2019-10-01T14:38:54 < zyp> power and stuff is measured automatically, and camera already knows the white balance of the flash 2019-10-01T14:39:16 < zyp> that's not a decent flash. 2019-10-01T14:40:38 < zyp> :p 2019-10-01T14:41:53 < zyp> haha 2019-10-01T14:42:56 < zyp> I have a nice collection of lenses, I just don't use them enough :( 2019-10-01T14:43:18 < zyp> it's always easier to pull the phone out of the pocket 2019-10-01T14:43:24 < zyp> heh 2019-10-01T14:43:30 < zyp> I don't even have a kit lens 2019-10-01T14:43:47 < zyp> which brand? 2019-10-01T14:44:08 < zyp> ah, not very familiar, I've got canon stuff 2019-10-01T14:44:48 < zyp> I got a 17-55/2.8, 70-200/2.8 and 60/2.8 macro 2019-10-01T14:45:03 < zyp> and some cheap shit I never bother using, like 50/1.8 2019-10-01T14:50:01 < zyp> the 60mm is great for technical photos 2019-10-01T14:50:05 < zyp> like https://bin.jvnv.net/file/E08pl.JPG 2019-10-01T15:02:24 < kakimir32> sweet 2019-10-01T15:21:12 < dongs> balllls 2019-10-01T15:21:47 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 276 seconds] 2019-10-01T15:31:36 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-01T15:38:53 < BrainDamage> correct 2019-10-01T15:39:19 < BrainDamage> mains is not a great waveguide, so if you want to go long distance your speed is going to suck 2019-10-01T15:44:53 -!- jly [uid355225@gateway/web/irccloud.com/x-upvfrtxunzbtjjgp] has quit [Quit: Connection closed for inactivity] 2019-10-01T16:03:05 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-01T16:06:04 < kakimir32> https://www.youtube.com/watch?v=Lmp51YN-7wc military parade 2019-10-01T16:18:09 < kakimir32> kakimortex 2019-10-01T16:19:37 < kakimir32> enhanced yes 2019-10-01T16:22:51 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-01T16:32:50 < kakimir32> poetic 2019-10-01T16:33:45 < Steffanx> How does muh wifi sound Haohmaru ? 2019-10-01T16:36:13 < roomba> https://www.youtube.com/watch?v=gsNaR6FRuO0 2019-10-01T16:44:42 -!- kow_ [~iccy@135.0.26.39] has quit [Read error: Connection reset by peer] 2019-10-01T16:50:00 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-01T16:54:32 < roomba> whoever got me addicted to canadian rednecks playing with old vehicles in the woods thanks 2019-10-01T16:54:42 < roomba> now i'm in the rare coins rabbit hole 2019-10-01T16:54:44 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-01T16:55:23 -!- kow_ [~iccy@135.0.26.39] has joined ##stm32 2019-10-01T16:59:14 < Steffanx> Hello Welcome 2019-10-01T17:00:41 < roomba> because i get addicted to anything and everything 2019-10-01T17:14:55 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-01T17:18:57 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T17:21:09 -!- jadew [~rcc@unaffiliated/jadew] has quit [Quit: WeeChat 2.5] 2019-10-01T17:31:18 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T17:31:42 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-01T17:35:06 < dongs> i gotta say 2019-10-01T17:35:42 < dongs> super impressed with mikrodik 2019-10-01T17:36:00 < karlp> gotta say, super unimpressed with mbed so far... 2019-10-01T17:36:03 < dongs> for the first time in years i tried to update firmware on my shit, remotely 2019-10-01T17:36:27 < dongs> lol mbed is a dumpster fire 2019-10-01T17:40:10 < kakimir32> yes 2019-10-01T17:40:18 < karlp> trying to work out how to run it on custom board, and it's just disaster 2019-10-01T17:40:34 < kakimir32> dongs: what microdik you have? 2019-10-01T17:40:39 < dongs> a bunch of them 2019-10-01T17:42:35 < dongs> mant19s, a couple SXT HG5 ac and a couple SXT SA5 2019-10-01T17:42:48 < Steffanx> Did upgrade to windows or what? 2019-10-01T17:58:31 < dongs> default system time on mikrodik is sept11... 2019-10-01T17:58:35 < dongs> they gotta be joking 2019-10-01T18:02:25 < zyp> haha 2019-10-01T18:04:08 < zyp> dongs, isn't remote firmware update supposed to just work? 2019-10-01T18:04:22 < dongs> for sure but it worked so well 2019-10-01T18:04:32 < dongs> also actual "update" is just as quick as reboot 2019-10-01T18:04:36 < zyp> I've never had any issues neither with mikrotik nor ubnt 2019-10-01T18:04:45 < dongs> shit was only down for like 10 seconds and back up with new version 2019-10-01T18:04:48 < dongs> crazy shit 2019-10-01T18:06:24 < zyp> Steffanx, was it you asking about my kitchen cupboards? https://bin.jvnv.net/file/cLASi.jpg 2019-10-01T18:07:36 < dongs> you need to fill that shit up with something useful man 2019-10-01T18:08:00 < zyp> that's the plan 2019-10-01T18:09:38 < zyp> or half the plan is to get all the useless shit off my desk so I can have useful stuff on my desk instead 2019-10-01T18:39:49 < karlp> like, even just all this shit, just to add it in the upstream is _nuts_ https://os.mbed.com/teams/ST/wiki/steps-to-create-a-new-STM32-platform 2019-10-01T18:46:11 < dongs> python??? json??? the actual fuck lol 2019-10-01T18:47:27 < emeb> these days if you don't do it with python/json/ then apparently it's not worth doing. 2019-10-01T18:47:31 < karlp> I ddidn't mind the little json snippet, except that none of it worked, 2019-10-01T18:47:49 < karlp> I should just use freertos and move on with life. 2019-10-01T18:48:08 < dongs> whty are you even using mbed? 2019-10-01T18:48:13 < dongs> for what feature of it 2019-10-01T19:04:28 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 245 seconds] 2019-10-01T19:04:38 < karlp> was considering it. 2019-10-01T19:04:47 < karlp> want to do a nice rtos based project fora new build. 2019-10-01T19:05:01 < karlp> was hoping to rip off all their flash storage and shit. 2019-10-01T19:05:13 < karlp> they're throwing shit tons of dev work on it, surely there must be something good in there. 2019-10-01T19:06:55 < emeb> hah. 2019-10-01T19:07:53 < karlp> apache2 licensed, no xFUCKYOU naming like freertos, all the lwip and usb shits all included these days, should be "no brainer" except it clearly isn't 2019-10-01T19:08:45 < dongs> well uh 2019-10-01T19:09:01 < dongs> you know they/re just using cmsis-rtos stuff right 2019-10-01T19:09:07 < dongs> you can just import that into your own shit directly 2019-10-01T19:09:25 < dongs> donno about flashfs and stuff, isnt fatfs + any sorta sd/sdio driver is basically plug and play on stm32 these days? 2019-10-01T19:09:46 < dongs> FUCK MBED; MBED SUX; MBED IS DYING; MBED IS DEAD TO ME; MBED HIT WTC 2019-10-01T19:10:15 * emeb has missed seeing that macro 2019-10-01T19:10:53 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-01T19:10:59 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-01T19:14:28 < veverak> karlp: they're throwing shit tons of dev work on it, surely there must be something good in there. || that is very naive reasoning 2019-10-01T19:14:52 < veverak> something can be good, but better question is if it is worth it 2019-10-01T19:15:01 < veverak> we used it in project. 2019-10-01T19:15:16 < veverak> it's nice that it is easy to use 2019-10-01T19:15:42 < karlp> well, it _isn't_ easy to use for anything that isn't running on an existing board with existing bootloader support is my problem :) 2019-10-01T19:16:01 < veverak> but we are switching to native sdk, as embed blocks us from doing whatwe want 2019-10-01T19:16:16 < veverak> too thick layer between our code and hw 2019-10-01T19:16:50 < veverak> (true, i am talkin about kinetis k66, butI suppose same holds for stm32) 2019-10-01T19:17:59 < emeb> thick layers are not good. 2019-10-01T19:18:07 < emeb> HAL is almost too much sometimes. 2019-10-01T19:18:51 < emeb> waste more time trying to work around the thick layer than it would take to just do it yourself. 2019-10-01T19:19:01 < veverak> exactly 2019-10-01T19:20:23 < veverak> => not worth it for serious projects 2019-10-01T19:21:26 * emeb is still waiting for that "serious" project. :P 2019-10-01T19:24:29 < englishman> not worth it for casual projects either 2019-10-01T19:27:37 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-01T19:27:50 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-01T19:32:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-01T19:33:52 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-01T19:52:31 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-01T19:55:12 < antto> zyp, https://i.imgur.com/KwRC2nr.jpg https://i.imgur.com/mMGUBpN.jpg 2019-10-01T19:55:27 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-01T19:57:15 < antto> https://i.imgur.com/7IUyfZ5.jpg 2019-10-01T19:57:29 < antto> https://i.imgur.com/4v3NqMZ.jpg 2019-10-01T19:58:43 < antto> https://i.imgur.com/1XUkN.jpg 2019-10-01T19:59:10 < antto> >:) 2019-10-01T20:05:48 < emeb> kitty 2019-10-01T20:06:10 < antto> predator 2019-10-01T20:06:42 < emeb> true. bugs and lizards lived in fear when we had cats. 2019-10-01T20:11:35 < zyp> I can also do usb connectors: https://bin.jvnv.net/file/LU8mZ.JPG 2019-10-01T20:12:06 < zyp> https://bin.jvnv.net/file/Kg4JE.JPG <- pins 1 and 4 doesn't connect properly 2019-10-01T20:12:12 < zyp> or are they 2 and 5, idk 2019-10-01T20:15:57 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-01T20:18:05 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-01T20:20:22 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-01T20:20:56 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-01T20:21:08 < kakimir32> it looks good though 2019-10-01T20:21:25 < kakimir32> oh wait 2019-10-01T20:21:30 < kakimir32> is it levitating? 2019-10-01T20:31:14 < Steffanx> yes, magic uh? 2019-10-01T20:36:46 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-01T20:41:09 < kakimir32> I did successful microusb soldering yesterday.. usually plastic inside melts but maybe this was better connector 2019-10-01T20:41:11 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-01T20:41:20 < kakimir32> it was not only successful 2019-10-01T20:41:23 < kakimir32> it was perfect 2019-10-01T20:41:33 < antto> 2019-10-01T20:41:52 < kakimir32> first reflowed them 5pins and then everything else with ts100 2019-10-01T20:49:30 < Steffanx> So what is kakimir32 up to? 2019-10-01T20:49:33 < Steffanx> New job? 2019-10-01T20:49:47 < kakimir32> yes I'm writing application at the moment 2019-10-01T20:51:34 < Steffanx> What will it be? Sw or hardware? 2019-10-01T20:51:35 < Steffanx> or both? 2019-10-01T20:54:16 < kakimir32> I applied for layout 2019-10-01T20:54:24 < kakimir32> they said they have enough layout 2019-10-01T20:54:34 < kakimir32> apply for hw sw 2019-10-01T20:54:40 < kakimir32> so I'm applying 2019-10-01T20:54:50 < kakimir32> firetrial number 2 2019-10-01T20:55:19 < kakimir32> *trial by fire 2019-10-01T20:55:45 < Steffanx> whats next? mech too? 2019-10-01T20:56:05 < kakimir32> I try to talk it down for some internship level at first. in finland interns get salary too 2019-10-01T20:56:44 < Steffanx> lol, dont do that 2019-10-01T20:57:04 < kakimir32> in and out then? 2019-10-01T20:57:46 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a92b:1d9d:aa2:d4e1] has joined ##stm32 2019-10-01T20:58:25 < kakimir32> you forgot I cannot do shit 2019-10-01T20:59:17 < BrainDamage> study microwave design and work for a connector company designing SMA connectors 2019-10-01T20:59:40 < kakimir32> I only use SMA for DC 2019-10-01T21:00:11 < kakimir32> someone once said anything under 2GHz is DC 2019-10-01T21:05:58 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-01T21:08:14 < rajkosto> oh so no proper routing needed for DDR 2019-10-01T21:08:18 < rajkosto> good to know 2019-10-01T21:08:54 < antto> zyp, https://i.imgur.com/KA1n5e2.jpg 2019-10-01T21:09:09 < kakimir32> rajkosto: just get the timings right 2019-10-01T21:09:26 < kakimir32> antto: is that rohs? 2019-10-01T21:10:13 < antto> RoHS? what is this, i don't even 2019-10-01T21:11:04 < Steffanx> i like your positive self-image kakimir32 2019-10-01T21:11:13 < Steffanx> Did you prev boss notice you cant do shit? 2019-10-01T21:11:46 < kakimir32> no 2019-10-01T21:12:13 < kakimir32> but they noticed that this guy is slow to shit 2019-10-01T21:12:43 < kakimir32> just fast enought but slow still 2019-10-01T21:12:55 < BrainDamage> long bathroom breaks frantically reading an electronics book? 2019-10-01T21:12:56 < kakimir32> *enough 2019-10-01T21:13:10 < kakimir32> circuit design 101 2019-10-01T21:13:31 < antto> muh collegue sh*ts like up to 4 times 2019-10-01T21:13:39 < kakimir32> a day? 2019-10-01T21:13:47 < Steffanx> damn, tell him to visit a doctor 2019-10-01T21:13:48 < antto> that is while at work, not counting sh*tting in his home 2019-10-01T21:13:48 < kakimir32> sorry day for highlight 2019-10-01T21:13:53 < Steffanx> are you sure he's shitting? 2019-10-01T21:14:02 < antto> yeah 2019-10-01T21:14:08 < Steffanx> sure? 2019-10-01T21:14:17 < antto> he is tall, not fat, and he eats tons 2019-10-01T21:14:31 < kakimir32> do noises and smells confirm your assumption? 2019-10-01T21:14:37 < antto> of course 2019-10-01T21:14:45 < Steffanx> antto: checks. 2019-10-01T21:14:49 < kakimir32> interesting 2019-10-01T21:15:28 < antto> and he plays an annoying and dumb starwars game on his dumbphone 2019-10-01T21:15:57 < antto> doesn't go to sh*t without his phone (coz he usually always gets it lost somewhere, but not when going to take a sh*t) 2019-10-01T21:16:13 < antto> and it takes like 30 minutes 2019-10-01T21:16:33 < antto> don't tell anyone :/ 2019-10-01T21:16:34 < rajkosto> gaming the system 2019-10-01T21:16:36 < rajkosto> gettin paid for shittin 2019-10-01T21:16:54 < kakimir32> livin a dream 2019-10-01T21:17:17 < Steffanx> You shouldn't be working harder than 80% of your full capacity anyway. 2019-10-01T21:17:28 < antto> i'm not ;P~ 2019-10-01T21:17:38 < Steffanx> when the time comes you can show how awesome you by adding the remaining 20% 2019-10-01T21:17:46 < antto> huhuhu 2019-10-01T21:17:57 < BrainDamage> and not burn out of fatigue 2019-10-01T21:18:04 < antto> those 20% is when i become the batman 2019-10-01T21:18:38 < antto> no one knows what it's like... to be the baaatmaaan.. 2019-10-01T21:18:40 < Steffanx> That too BrainDamage. I want 6 hour workdays. 2019-10-01T21:18:58 < kakimir32> Steffanx: when it's time to shine you burn through it at 115% power and getting one year older in one month 2019-10-01T21:19:19 < kakimir32> *get 2019-10-01T21:19:20 < Steffanx> i dont do that 2019-10-01T21:19:25 < antto> i got moar and moar white hair :/ 2019-10-01T21:20:17 < kakimir32> one hairstyle is the easiest style 2019-10-01T21:20:42 < Steffanx> You're old, antto 2019-10-01T21:20:49 < antto> i know 2019-10-01T21:20:54 * antto farts 2019-10-01T21:21:04 < antto> muh cat's got white hair too ;P~ 2019-10-01T21:21:27 < antto> dayum she's old as well 2019-10-01T21:22:07 < kakimir32> kitten required 2019-10-01T21:22:26 < antto> she's like 9 or 10 now, i don't know exactly 2019-10-01T21:22:54 < kakimir32> to replenish your cats 2019-10-01T21:23:10 < kakimir32> thats not even old 2019-10-01T21:23:41 < antto> it is.. quite old for a cat 2019-10-01T21:23:48 < Steffanx> no, its not :P 2019-10-01T21:23:59 < antto> they normally like like 5-ish years on teh streets 2019-10-01T21:24:01 < kakimir32> 10 years ago I thought the old cat was soon ded.. I have no idea how old it is 2019-10-01T21:24:04 < antto> * live 2019-10-01T21:24:05 < Steffanx> jly is 9 and shes not old 2019-10-01T21:24:10 < kakimir32> it's still kicking 2019-10-01T21:24:37 < antto> yeah they can grow very old like 20 or even 30? iirc 2019-10-01T21:24:44 < antto> but they are not supposed to, normally 2019-10-01T21:25:00 < antto> they start getting all kinds of "old man's" health issues 2019-10-01T21:25:05 < kakimir32> getting mice and terrorising birds.. I don't think it can see anymore 2019-10-01T21:25:24 < kakimir32> at least sharply 2019-10-01T21:26:27 < antto> mine's hunting me when i secretly OMNOM nice stuff like icecream or what not 2019-10-01T21:26:38 < Steffanx> our cats usually die around 15. 2019-10-01T21:26:43 < Steffanx> or they die younger when hit by a car :P 2019-10-01T21:27:00 < antto> my previous cat was 13 2019-10-01T21:27:01 < antto> :~( 2019-10-01T21:27:18 < antto> when i was 26 2019-10-01T21:27:35 < antto> half of my life 2019-10-01T21:27:57 < antto> oh well 2019-10-01T21:27:59 < Steffanx> You're not even 26 yet. 2019-10-01T21:28:21 < antto> eh? 2019-10-01T21:28:37 < antto> i'm furdy-free, brun 2019-10-01T21:28:51 < antto> dayum, so much typos today 2019-10-01T21:29:22 < antto> now i'm driving with teh instructor in teh morning 2019-10-01T21:29:32 < antto> this morning was muchos disaster0s 2019-10-01T21:30:30 < Steffanx> as in instructor to get your car drivers license? 2019-10-01T21:31:02 < antto> yeah i'm trying to get a loicens 2019-10-01T21:31:34 < Steffanx> doesnt a .bg guy just buy something like that? 2019-10-01T21:31:45 < Steffanx> or drive around on the parking 2 times and gets it 2019-10-01T21:32:48 < antto> well, there are many completely uneducated ghetto folks (but with moniez) who happen to have a loicens 2019-10-01T21:33:15 < antto> which means that you can buy it, but they are supposedly working on reducing corruption 2019-10-01T21:33:50 < BrainDamage> here driving license culture is so strongly rooted i find it weird hearing people without one 2019-10-01T21:33:55 < antto> otherwise you gotta be able to read.. at least.. to even pass the theory test 2019-10-01T21:34:03 < BrainDamage> even with pretty wide availability of public transport 2019-10-01T21:35:58 < Steffanx> same here. 2019-10-01T21:36:09 < Steffanx> i was a weird one when i didnt have one by 20 years old. 2019-10-01T21:37:34 < antto> https://www.youtube.com/watch?v=9DT-mWJWSSw 2019-10-01T21:38:10 < Steffanx> thats how imagine all of yo 2019-10-01T21:38:11 < Steffanx> u 2019-10-01T21:38:35 < antto> nah, that's a gypsy in some ghetto 2019-10-01T21:39:10 < antto> most of the people here are white people, in something that looks very ghetto but isn't technically a ghetto ;P~ 2019-10-01T21:40:34 < antto> well, soviet apartment blocks where each apartment might have (or not) varying degrees of modifications, so the whole block ends up looking kinda like.. ghetto 2019-10-01T21:41:08 < Steffanx> Will you buy a fiat 126p first? 2019-10-01T21:41:19 < Steffanx> aka maluch 2019-10-01T21:41:51 < Steffanx> ohno that's more a polish thing i think 2019-10-01T21:42:12 < antto> we don't have a car brand here 2019-10-01T21:42:40 < antto> we do manufacture GreatWall cars iirc tho.. or we assemble them? 2019-10-01T21:43:18 < antto> our prime minister was recently trying to seduce VW to make a factory here 2019-10-01T21:43:34 < antto> they were considering turkey.. then he went to seduce Hyundai >:) 2019-10-01T21:44:37 < Steffanx> you have Litex Motors 2019-10-01T21:44:40 < Steffanx> ask wikipedia 2019-10-01T21:45:04 < PaulFertser> Steffanx: here the exam for cars is relatively sane (except for the lack of highway practice/test) and reasonably hard, but one can often "buy" it. For motor bikes it's just horrible, you need to show just basic handling skills, no exposure to real world traffic at all. 2019-10-01T21:45:28 < antto> Steffanx well that's the GreatWall thing 2019-10-01T21:45:37 < antto> there ain't no .bg car brand 2019-10-01T21:46:13 < Steffanx> i see 2019-10-01T21:46:24 < antto> our theory exam is available online, i think it even has english translation 2019-10-01T21:46:54 < antto> afaik the "real" exam is basically, you are given a tablet with that thing running on it 2019-10-01T21:47:13 < antto> then there's the driving test too 2019-10-01T21:47:14 < Steffanx> aha PaulFertser. Do you also practice drunk? :P 2019-10-01T21:47:27 < Steffanx> *driving drunk 2019-10-01T21:47:56 < antto> https://www.avtoizpit.com/listovki#/sub-categories 2019-10-01T21:48:11 < PaulFertser> Steffanx: not funny, many people here die because of DUI... 2019-10-01T21:48:39 < antto> it'd be funny if i get higher score on this if i run it in english 2019-10-01T21:48:41 < antto> huhuhu 2019-10-01T21:48:42 < qyx> but why 2019-10-01T21:50:06 < Steffanx> What the hell antto. The first question i have "the visibility on the figure is". .. "limited" "reduced". 2019-10-01T21:50:42 < antto> yeah, there is a distinction between limited and reduced visibility 2019-10-01T21:50:52 < Steffanx> heh, yeah that's sad PaulFertser 2019-10-01T21:51:00 < BrainDamage> I remember some dumb semantics questions in the mock theory tests as well here 2019-10-01T21:51:09 < antto> limited == something is blocking it, reduced == iz like foggy as fugg 2019-10-01T21:51:12 < BrainDamage> but they were very few 2019-10-01T21:51:14 < antto> THICC FOG 2019-10-01T21:51:19 < Steffanx> yes, because im really going to think about that when i drive around on such road. 2019-10-01T21:51:54 < PaulFertser> Steffanx: I have to admit I was riding a push bike drunk, and not once... 2019-10-01T21:52:36 < antto> iirc you gotta get at least 87 points on this to pass it 2019-10-01T21:52:49 < antto> the max score is 97 2019-10-01T21:53:06 < PaulFertser> Steffanx: but that way I was endangering only myself. 2019-10-01T21:54:11 < antto> Steffanx after all, this driving loicens is valid within the EU (at minimum) they can't just give it away to anyone (that is, ignoring corruption) 2019-10-01T21:54:58 < PaulFertser> antto: do you have real traffic riding practice for motor bikes? 2019-10-01T21:55:18 < antto> huh, no idea 2019-10-01T21:56:03 < antto> i'm going for category B, which is for carz and vehicles under 3.5 tonz, and for up to 9 or so passangers 2019-10-01T21:56:14 < PaulFertser> 8 passengers most likely 2019-10-01T21:56:18 < antto> this category also lets me drive a "moped" 2019-10-01T21:56:28 < antto> 8 or 9, i forgot ;P~ 2019-10-01T21:56:46 < PaulFertser> I still do not like cars... 2019-10-01T21:57:00 < kakimir32> cars are boring 2019-10-01T21:57:06 < antto> but for akchual motor bikes, there are the A categories, i'm not familiar with themz 2019-10-01T21:57:27 < antto> iz not funny to be on a bike in teh winter here 2019-10-01T21:57:45 < antto> muchos cold, very snow, such ice 2019-10-01T21:58:01 < antto> and teh sea breeze 2019-10-01T22:00:27 < Steffanx> here you do have them, pretty serious ones. 2019-10-01T22:01:06 < antto> bikes? 2019-10-01T22:01:28 < kakimir32> moped 2019-10-01T22:01:46 < antto> we got bikes here too, anything from silly mopeds to even choppers o_O 2019-10-01T22:01:47 < Steffanx> does you moped move? 2019-10-01T22:02:23 < antto> i don't understand these folks.. we are not america here, we don't have long roads.. a chopper is a sofa on wheels 2019-10-01T22:02:39 < antto> i hate the noise some of them make, so loud 2019-10-01T22:02:46 < antto> UGH 2019-10-01T22:03:14 < kakimir32> like every motorcyclist I know wants to have or has harley davidson 2019-10-01T22:03:18 < kakimir32> it's a trend 2019-10-01T22:03:47 < antto> yeah.. and then you GOTTA have a stylized beard 2019-10-01T22:04:02 < antto> you know, designer beard, or however they call em 2019-10-01T22:04:24 < Steffanx> here they are only owned by biker gang members 2019-10-01T22:04:30 < Steffanx> and most gangs are forbidden 2019-10-01T22:05:14 < kakimir32> tried HD softtail 2019-10-01T22:05:33 < kakimir32> it was spinal rape 2019-10-01T22:05:53 < kakimir32> comfy - my ass 2019-10-01T22:07:24 < antto> our church here has had some spectacular moments: https://www.youtube.com/watch?v=8NWwuZEfHXA 2019-10-01T22:07:55 < Steffanx> "our" 2019-10-01T22:08:43 < antto> mitropolit (hower that translates) Kiril of Varna ;P~ 2019-10-01T22:09:09 < antto> had this lincoln when at the time it was a brand new model and perhaps only a few other people had such model 2019-10-01T22:09:16 < antto> Obama was one of them iirc 2019-10-01T22:09:24 < antto> >:) 2019-10-01T22:09:53 < kakimir32> where do you live antto? 2019-10-01T22:09:54 < antto> Insert coin for teh poor church 2019-10-01T22:10:00 < kakimir32> kazak? 2019-10-01T22:10:05 < antto> eh? 2019-10-01T22:10:20 < antto> Varna 2019-10-01T22:10:32 < antto> kazak is some russian thang 2019-10-01T22:10:34 < kakimir32> oh yes 2019-10-01T22:10:39 < kakimir32> makes sense now 2019-10-01T22:11:02 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-01T22:22:17 < Steffanx> i hate drunks in general, so i hate you too PaulFertser :P 2019-10-01T22:22:32 < PaulFertser> Steffanx: but I'm not drunk all the time 2019-10-01T22:22:44 < Steffanx> i hope so. 2019-10-01T22:23:42 < Steffanx> and hate is such big word. i dislike being around them. 2019-10-01T22:25:41 < PaulFertser> Steffanx: do you prefer being around stoners ;) 2019-10-01T22:25:57 < Steffanx> They tend to be less annoying, yeah. 2019-10-01T22:26:56 < Steffanx> and who needs any of those when you have ##stm32 \o/ 2019-10-01T22:45:08 < zyp> Steffanx, you don't drink yourself? 2019-10-01T22:45:50 < Steffanx> Sure i do, but im never drunk. EVER. 2019-10-01T22:46:14 < Steffanx> i dont even drive when i had a single beer (which is waaay below the limit) 2019-10-01T22:46:33 < zyp> uh, isn't that normal? 2019-10-01T22:46:44 < Steffanx> no, here way too many people still drive 2019-10-01T22:46:49 < Steffanx> because .. way below the limit 2019-10-01T22:47:01 < zyp> what's the limit? IIRC it's 0.2 here 2019-10-01T22:47:47 < zyp> I'm usually thinking about it the other way around, I don't drink if I'm planning to drive later 2019-10-01T22:49:01 < Steffanx> oh, its actually not "waaay below the limit", but under the limit still :P 2019-10-01T22:49:32 < zyp> there's been a few occasions where I've taken the car some hours after having a single beer, but generally I also don't drive after drinking before I've slept on it :p 2019-10-01T22:49:46 < Steffanx> its 0.5 Per mille here. not sure what kind of standard you use. 2019-10-01T22:49:56 < zyp> a few = two or three in my lifetime 2019-10-01T22:50:04 < zyp> yeah, 0.2 per mille 2019-10-01T22:50:20 < Steffanx> although we now use microgram/liter "ugl" 2019-10-01T22:50:56 < zyp> my understanding is that one beer may bring you above 0.2, but below again after a few hours 2019-10-01T22:51:29 < Steffanx> as a starter its 0.2 too here. 2019-10-01T22:52:11 < BrainDamage> here's 0.5per thousand, but a medium beer would still bring you over the limit 2019-10-01T22:52:23 < BrainDamage> only a small one wouldn't 2019-10-01T22:53:51 < Steffanx> Does italy beer have more alcohol or .. ? 2019-10-01T22:54:16 < BrainDamage> no idea, 5% ABV or so 2019-10-01T22:54:28 < Steffanx> unless a medium beer is ~0.5L or something in italy. 2019-10-01T22:54:36 < BrainDamage> yes, that's a medium beer here 2019-10-01T22:54:41 < BrainDamage> a small one is 0.33L 2019-10-01T22:54:56 < BrainDamage> while a large one is 1L 2019-10-01T22:55:05 < BrainDamage> a medium beer is also roughly a pint 2019-10-01T22:55:38 < PaulFertser> zyp: if your beer is imperial stout or barleywine your calculations are way off ;) 2019-10-01T22:56:08 < BrainDamage> belgian quadrupel can easily hit 10-12% 2019-10-01T22:56:21 < PaulFertser> Oh that yes. Or baltic porter. 2019-10-01T22:56:42 < Steffanx> Anyway, beer is overrated 2019-10-01T23:02:58 < catphish> i've not had any beer this year :( 2019-10-01T23:03:39 < Steffanx> just replaced it with stronger stuff uh? 2019-10-01T23:04:53 < catphish> actually ye 2019-10-01T23:04:57 < catphish> *yes 2019-10-01T23:05:36 < catphish> that wasn't my intention, but it's what's happened, no beer rule was ill conceived :) 2019-10-01T23:05:48 < catphish> but it did help me lose weight 2019-10-01T23:09:08 < Steffanx> lol 2019-10-01T23:09:53 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 268 seconds] 2019-10-01T23:21:38 < karlp> I've not had any beer in almost 24 hours.... 2019-10-01T23:31:34 < zyp> PaulFertser, by «a beer» I think of the ones we can buy in normal grocery stores, legal limit is 4.3% ABV 2019-10-01T23:32:48 < zyp> speaking of, I'm not sure I've had any beer this year either 2019-10-01T23:33:23 < PaulFertser> zyp: oh, if grocery stores do not sell anything abouv 4.3% that's pretty lame and limiting style options drastically. 2019-10-01T23:33:37 < zyp> I had a bunch of bottles that I put in a box when I moved in january, and they are still in that box 2019-10-01T23:34:24 < zyp> and I'm not sure I had any beer when I went for that wedding in poland, mostly just vodka 2019-10-01T23:34:47 < zyp> oh, I did have a couple before the party 2019-10-01T23:36:41 < zyp> PaulFertser, better than sweden and their 3.5% limit 2019-10-01T23:36:47 < Steffanx> next time it's just water for you. 2019-10-01T23:36:57 < PaulFertser> ЯР 2019-10-01T23:36:58 < PaulFertser> :D 2019-10-01T23:37:24 < zyp> but I mean, you can get higher percentages, just not in grocery stores 2019-10-01T23:38:02 * karlp bought this yesterday: http://www.borgbrugghus.is/thebeer/product/nr-68-karl 2019-10-01T23:38:24 < karlp> zyp: is sweden 3.5 by weight though? that's pretty similar to 4.low by volume. 2019-10-01T23:38:36 < karlp> some of the us states are 3.2 by weight too. 2019-10-01T23:40:52 < zyp> I think it's by volume 2019-10-01T23:41:22 < zyp> yeah, wikipedia confirms 2019-10-01T23:42:28 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-01T23:42:40 < karlp> yar. we have 2.25 by volume in any shop, anything over that is in the state liquor store. 2019-10-01T23:43:12 < zyp> oh, even lower 2019-10-01T23:43:12 < BrainDamage> here you can buy all sort of alcohol in supermarkets 2019-10-01T23:43:27 < PaulFertser> Here too 2019-10-01T23:43:32 < BrainDamage> but then again alcoholism rates are pretty low here 2019-10-01T23:43:56 < BrainDamage> iirc we're the lowest country in eu in terms of death by alcoholism and alcohol per capita 2019-10-01T23:43:57 < PaulFertser> Unlike here :( 2019-10-01T23:45:57 < karlp> I think it's been quite conclusively demonstrated taht on the whole, availability doesn't have any major impact. 2019-10-01T23:46:22 < karlp> individuals can fall off the wagon more easily when it's in the grocery stores (fuck em) but net societal impact is ~neglible 2019-10-01T23:46:39 < karlp> (we're having arguments here about privatising liquor sales, so this is all being brough tup again) 2019-10-01T23:49:53 < catphish> is it possible for software to jump (or reboot) to the built in bootloader? 2019-10-01T23:52:49 < catphish> looks like you can, but there's some prep work to do 2019-10-01T23:54:16 < catphish> it's tempting just to self-brick and reboot :) next time i'll remember to attach a button to boot0 2019-10-01T23:54:52 < zyp> why not roll your own bootloader? 2019-10-01T23:55:31 < catphish> seems like a lot of work for little benefit 2019-10-01T23:56:06 < catphish> i guess my app could do its own firmware updates though 2019-10-01T23:56:46 < catphish> oh, you can't just write over the running program, never mind :) 2019-10-01T23:57:36 < catphish> i see no benefit of using a custom bootloader over the system one, just need a way to get into it 2019-10-01T23:57:57 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has quit [Ping timeout: 245 seconds] 2019-10-01T23:58:04 < catphish> in the meantime i'll just leave my nucleo hanging off it --- Day changed Wed Oct 02 2019 2019-10-02T00:04:03 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a92b:1d9d:aa2:d4e1] has quit [Quit: Konversation terminated!] 2019-10-02T00:14:14 < zyp> well, you're hitting on one of the big benefits, you can do custom trigger conditions, e.g. with buttons you already have 2019-10-02T00:14:33 < zyp> like how on smartphones you hold one of the volume buttons during poweron 2019-10-02T00:17:20 < emeb_mac> +1 for rolling your own bootloader. 2019-10-02T00:18:40 < emeb_mac> mine flashes from SD card. Includes CRC and PRN scrambling of the binary to detect errors and obfuscate code. 2019-10-02T00:21:39 < karlp> being able to flash over an interface you already have instead of what the room bootloader supports too 2019-10-02T00:25:15 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 240 seconds] 2019-10-02T00:29:17 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-02T00:45:31 < kakimir32> I have ruined my lower back 2019-10-02T00:45:40 < kakimir32> before 30's 2019-10-02T00:45:46 < kakimir32> it's shitted 2019-10-02T00:45:50 < roomba> no problem 2019-10-02T00:46:34 < kakimir32> I think office work was the root cause 2019-10-02T00:47:03 < kakimir32> and those ergonomic chairs that wont allow you sit like shit 2019-10-02T00:47:59 < kakimir32> propper sitting position is when back arcs to C shape 2019-10-02T00:51:00 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has joined ##stm32 2019-10-02T00:58:14 < machinehum> If you see someone breaking into a building... should you call 911 or non-emergency? 2019-10-02T00:59:18 < karlp> depends whether it's your building or not :) 2019-10-02T00:59:38 < BrainDamage> 911 is pretty justified, but prepare to open yourself to a can of worms 2019-10-02T00:59:46 < BrainDamage> no good deed will remain unpunished 2019-10-02T01:05:04 < kakimir32> machinehum: you follow them inside 2019-10-02T01:13:44 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-02T01:13:49 < Laurenceb> autism horse 2019-10-02T01:19:04 < Laurenceb> https://hardware.slashdot.org/comments.pl?sid=14922966&cid=59258916 2019-10-02T01:26:01 < Laurenceb> https://pbs.twimg.com/media/EFnUTVPWsAAsloi?format=jpg&name=small 2019-10-02T01:26:27 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-02T01:51:37 < Laurenceb> Musk is getting fat 2019-10-02T01:52:40 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-02T01:59:16 < machinehum> kakimir32: k 2019-10-02T01:59:50 < machinehum> I just told some other guy at work and he called the cops 2019-10-02T02:01:45 < Laurenceb> >murcan cops 2019-10-02T02:01:49 < Laurenceb> u gunna get shot 2019-10-02T02:01:59 < Laurenceb> >britbong cops 2019-10-02T02:02:10 < Laurenceb> u gonna get arrested for bullshit 2019-10-02T02:02:59 < machinehum> Canada 2019-10-02T02:03:27 < machinehum> u gonna watch as they let the guy go 2019-10-02T02:08:16 < Laurenceb> argggg why did I try to work with nasaspaceflight loons 2019-10-02T02:08:27 < Laurenceb> they are still smarter than babbyshake 2019-10-02T02:08:55 < Laurenceb> >send email 2 weeks ago about plan of action, mention that simulation is going to be tricky 2019-10-02T02:09:18 < Laurenceb> >they are liek "yeah whatever, dude weed lmao" 2019-10-02T02:09:30 < Laurenceb> >send another email 1 week ago with more details 2019-10-02T02:09:34 < Laurenceb> >they dont care 2019-10-02T02:09:57 < Laurenceb> >suddenly they start email panicking 2019-10-02T02:25:37 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-02T02:26:18 -!- Mangy_Dog [~Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 246 seconds] 2019-10-02T02:33:14 < Laurenceb> how will the libtards recover from Trumps 5 dimensional hungry hungry hippos? 2019-10-02T02:38:54 < dongs> Join ST for a 1-hour webinar to discover a new motion sensor with Machine Learning that is disrupting the IoT and revolutionizing embedded sensor design. 2019-10-02T02:39:00 < dongs> oh for fucks sake 2019-10-02T02:39:09 < dongs> why the shit does a fucking accelerometer need dick learning 2019-10-02T02:43:14 < rajkosto> so you have to do more than minimal math to convert accelerometer data to a quaternion 2019-10-02T02:43:21 < rajkosto> waste some cpu cycles 2019-10-02T02:43:27 < rajkosto> madgwick is all you need 2019-10-02T02:44:56 < dongs> whats a ESD circuit in context of synaptics touchpads 2019-10-02T02:45:19 < dongs> is that just captouch stuff? 2019-10-02T02:50:20 < dongs> oh never mind. they literally just mean esd as in esd. 2019-10-02T02:52:26 < kakimir32> what to do with 10kiloeuros? 2019-10-02T02:52:53 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-02T02:53:34 < kakimir32> I'm free to do anything 2019-10-02T03:03:37 < Laurenceb> coke and hookers 2019-10-02T03:04:02 < kakimir32> okay 2019-10-02T03:04:03 < kakimir32> where? 2019-10-02T03:04:04 < Laurenceb> I'm not sure I can be bothered to email the nasaspaceflight crew 2019-10-02T03:04:09 < Laurenceb> britbongland 2019-10-02T03:04:34 < Laurenceb> there is a d e s i g n a t e d hooker pen near my house 2019-10-02T03:04:34 < kakimir32> when did uk leave eu? 2019-10-02T03:04:38 < Laurenceb> never 2019-10-02T03:04:44 < kakimir32> touche 2019-10-02T03:04:57 < Laurenceb> >youtube banned honker video 2019-10-02T03:05:00 < Laurenceb> my sides 2019-10-02T03:05:09 < kakimir32> do you visit often Laurenceb? 2019-10-02T03:05:23 < Laurenceb> youtube or the hooker pen? 2019-10-02T03:05:32 < kakimir32> ... 2019-10-02T03:05:39 < kakimir32> hookers 2019-10-02T03:05:44 < Laurenceb> nah 2019-10-02T03:06:16 < kakimir32> when your jubb ends lurence? 2019-10-02T03:07:06 < Laurenceb> next month 2019-10-02T03:07:07 < Laurenceb> https://www.youtube.com/watch?v=51IXwqDDLG4 2019-10-02T03:07:40 < kakimir32> you have next jubb in line? 2019-10-02T03:07:49 < Laurenceb> my plan was to team up with nasaspaceflight forum crew 2019-10-02T03:07:56 < Laurenceb> bad idea, they are too retarded 2019-10-02T03:07:59 < kakimir32> :o 2019-10-02T03:08:21 < kakimir32> there is grll you want to bang? 2019-10-02T03:08:34 < Laurenceb> nope 2019-10-02T03:08:55 < kakimir32> oh she was some 8chan mod.. 2019-10-02T03:09:05 < kakimir32> I mixed those up 2019-10-02T03:09:09 < Laurenceb> oh her lol 2019-10-02T03:09:17 < Laurenceb> she is out of a jerb now 2019-10-02T03:09:27 < kakimir32> did she go to jail? 2019-10-02T03:09:41 < Laurenceb> I dont think so 2019-10-02T03:09:55 < Laurenceb> the youtube nazi one got 8 years in jail tho 2019-10-02T03:10:43 < Laurenceb> but she was nasty and used to get 8channers b& from youtube for trolling her, so shrug 2019-10-02T03:11:36 < Laurenceb> Virginia Hale, that was the 8chan moderatess 2019-10-02T03:11:49 < BrainDamage> what did you need to team up for? 2019-10-02T03:12:04 < Laurenceb> google images are your friend, beware of the autism 2019-10-02T03:12:15 < Laurenceb> BrainDamage: muh space """runway"" 2019-10-02T03:12:30 < Laurenceb> we teamed up to get US gov funding 2019-10-02T03:12:37 < kakimir32> is nasaspaceflight something you get money from? 2019-10-02T03:12:44 < Laurenceb> maybe 2019-10-02T03:13:01 < BrainDamage> convince enough people and someone will throw money at you 2019-10-02T03:13:02 < Laurenceb> $1.25M if we succeed in getting the grant 2019-10-02T03:13:51 < Laurenceb> kakimir32: https://www.hopenothate.org.uk/wp-content/uploads/2017/10/virginia-hale-1-455x285.png 2019-10-02T03:13:59 < Laurenceb> ^grrrls of 8chan 2019-10-02T03:14:09 < kakimir32> cute 2019-10-02T03:14:14 < Laurenceb> lmao 2019-10-02T03:17:25 < Laurenceb> BrainDamage: https://forum.nasaspaceflight.com/index.php?topic=48773.0 2019-10-02T03:17:41 < Laurenceb> beware, ten pages of autism at the start 2019-10-02T03:19:22 < BrainDamage> >- No crazy physics, crazy materials or crazy engineering is required, is not a "megascale" engineering project 2019-10-02T03:19:22 < BrainDamage> >a long aluminum tube, 10-30km in length in LEO 2019-10-02T03:20:36 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 265 seconds] 2019-10-02T03:21:37 < aandrew> https://v.redd.it/w0jgpjci4wp31 2019-10-02T03:24:15 < Laurenceb> BrainDamage: its actually not that unfeasible 2019-10-02T03:24:37 < Laurenceb> >reddit 2019-10-02T03:24:52 < Laurenceb> BrainDamage: no gravity in orbit 2019-10-02T03:26:08 < BrainDamage> sure, but getting the materials up there and welding it doesn't exactly seems easy to me 2019-10-02T03:26:44 < BrainDamage> also how do you plan to generate the mag field on the vehicle 2019-10-02T03:26:58 < Laurenceb> yeah, but unlike all the other schemes its not actually provably impossible 2019-10-02T03:27:04 < Laurenceb> like space elevator etc 2019-10-02T03:27:18 < Laurenceb> using HTS superconducting tape coils 2019-10-02T03:28:17 < Laurenceb> about 0.5T 2019-10-02T03:28:26 < BrainDamage> so the strength constraints on the forces are more likely coming from the coil integrity rather than the tube's 2019-10-02T03:28:53 < Laurenceb> hmm I dont think so 2019-10-02T03:29:07 < Laurenceb> coil can be wrapped in fibreglass like the ITER stuff 2019-10-02T03:29:17 < BrainDamage> aren't most supercond brittle? 2019-10-02T03:29:23 < Laurenceb> but the tube gets really hot 2019-10-02T03:29:33 < Laurenceb> thats been solved in the last few years 2019-10-02T03:29:57 < Laurenceb> using laminated tapes, there are a few products on the market now that will work with LN2 2019-10-02T03:30:38 < Laurenceb> it looks like a stainless steel foil tube works 2019-10-02T03:31:14 < antto> loirens, do u got loicens to drive a shuttle? 2019-10-02T03:31:29 < Laurenceb> lol 2019-10-02T03:31:55 < antto> can you parallel park it? ;P~ 2019-10-02T03:33:35 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-02T03:42:16 < antto> didn't this musk guy work on some rockets that could land safely back on ground and be reused? 2019-10-02T03:43:55 < BrainDamage> are you sure about the insane heating? I've tried of a back of the envelope calc and I'm getting pretty low values 2019-10-02T03:44:09 < BrainDamage> 0.5 10ton (7km/s)^2 / 10 km / ( ((3m)^2·pi − (2.7m)^2·pi) · 2710kg/m^3 · 0.91 kJ/(kgK) ) gives me 0.2K 2019-10-02T03:45:34 < antto> aww noes, such much mathz 2019-10-02T04:21:42 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-02T04:40:27 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-02T05:07:49 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-02T05:34:27 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 240 seconds] 2019-10-02T05:52:34 < dongs> is there any practical difference between HITL and HILS or is that just the shit naming for same thing 2019-10-02T06:11:20 < aandrew> wtf is HITL/HILS 2019-10-02T06:13:19 < dongs> hardware in the loop / hardware in loop simulation 2019-10-02T06:13:32 < dongs> im wonderingf if the "simulation" part has some difference 2019-10-02T06:14:11 < dongs> for fucks sake. forgot to tent vias on this fucking QFN 2019-10-02T06:14:17 < dongs> chinapcb leaked straight through 2019-10-02T06:19:46 < dongs> sometimes i use chinapcb for throaway protos and i always wish i didnt after 2019-10-02T06:19:57 < aandrew> heh 2019-10-02T06:20:07 < aandrew> so what is simulation if it has hardware 2019-10-02T06:20:13 < aandrew> sounds like verification rather than simulation 2019-10-02T06:20:26 < dongs> no 2019-10-02T06:20:59 < dongs> the HITL stuff i know is, hardware runs the shit but environmental inputs are simulated 2019-10-02T06:21:30 < dongs> i.e. you have some embedded system that turns on pump depdning on water level, you run the board and fake water level meter to see how it acts 2019-10-02T06:21:46 < aandrew> right I get it 2019-10-02T06:21:55 < aandrew> you basically provide sensor input and measure system output 2019-10-02T06:22:08 < aandrew> but the system doing the computation/control is real 2019-10-02T06:22:26 < aandrew> never heard of it has HITL/HILS before 2019-10-02T06:43:41 -!- fc5dc9d4 [~quassel@p5B081358.dip0.t-ipconnect.de] has joined ##stm32 2019-10-02T06:47:32 -!- fc5dc9d4_ [~quassel@p5B0813FA.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 2019-10-02T07:01:37 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-02T07:01:49 -!- fbu [~fbu@2a01:e0a:11c:6bf0:129a:ddff:fe4f:4bc9] has quit [Quit: WeeChat 2.5] 2019-10-02T07:17:26 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-02T07:30:27 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-02T07:34:18 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-02T07:34:19 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-02T07:38:10 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-02T07:38:13 -!- tomeaton17_ [tomeaton17@unaffiliated/tomeaton17] has joined ##stm32 2019-10-02T07:38:22 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:22 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:22 -!- varesa [~varesa@ec2-52-49-18-111.eu-west-1.compute.amazonaws.com] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:47 -!- yaqwsx_ [znc@anna.fi.muni.cz] has joined ##stm32 2019-10-02T07:38:47 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:47 -!- scrts2 [~scrts@d27-96-211-8.nap.wideopenwest.com] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:47 -!- zyp [zyp@zyp.no] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:47 -!- yaqwsx [znc@anna.fi.muni.cz] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:47 -!- munki_ [munki@blogging.is.nigga.technology] has quit [Ping timeout: 245 seconds] 2019-10-02T07:38:51 -!- yaqwsx_ is now known as yaqwsx 2019-10-02T07:38:59 -!- munki_ [munki@blogging.is.nigga.technology] has joined ##stm32 2019-10-02T07:39:12 -!- effractur [~Erik@195.140.241.50] has quit [Ping timeout: 245 seconds] 2019-10-02T07:40:27 -!- inca [~inca@162.154.131.90] has quit [Ping timeout: 240 seconds] 2019-10-02T07:40:48 -!- zyp [zyp@zyp.no] has joined ##stm32 2019-10-02T07:40:51 -!- scrts [~scrts@d27-96-211-8.nap.wideopenwest.com] has quit [Ping timeout: 240 seconds] 2019-10-02T07:40:51 -!- Sadale [~Sadale@unaffiliated/sadale] has quit [Ping timeout: 240 seconds] 2019-10-02T07:40:52 -!- effractur [~Erik@195.140.241.50] has joined ##stm32 2019-10-02T07:43:23 -!- Sadale [~Sadale@unaffiliated/sadale] has joined ##stm32 2019-10-02T07:43:29 -!- inca [~inca@162.154.131.90] has joined ##stm32 2019-10-02T07:43:56 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-02T07:44:39 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-02T07:44:39 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-02T07:44:39 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-02T07:45:25 -!- day [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-02T07:46:46 -!- scrts [~scrts@d27-96-211-8.nap.wideopenwest.com] has joined ##stm32 2019-10-02T07:46:46 -!- scrts [~scrts@d27-96-211-8.nap.wideopenwest.com] has quit [Changing host] 2019-10-02T07:46:46 -!- scrts [~scrts@unaffiliated/scrts] has joined ##stm32 2019-10-02T07:47:17 -!- scrts2 [~scrts@d27-96-211-8.nap.wideopenwest.com] has joined ##stm32 2019-10-02T07:58:03 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-02T07:59:33 -!- varesa [~varesa@ec2-52-49-18-111.eu-west-1.compute.amazonaws.com] has joined ##stm32 2019-10-02T08:06:54 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32 2019-10-02T08:44:13 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-xatkedinvnrswiui] has quit [Quit: Connection closed for inactivity] 2019-10-02T09:14:08 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-02T10:00:14 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-02T10:12:26 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-02T10:13:17 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-02T10:38:13 < Steffanx> Welcome at work Haohmaru 2019-10-02T10:43:47 -!- irf21k [~irf21k@106.193.54.85] has joined ##stm32 2019-10-02T10:49:55 -!- irf21k [~irf21k@106.193.54.85] has quit [Remote host closed the connection] 2019-10-02T10:51:49 -!- jly [uid355225@gateway/web/irccloud.com/x-ulbyehdxgkbdnvwh] has joined ##stm32 2019-10-02T11:18:47 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 268 seconds] 2019-10-02T11:21:55 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-02T11:50:03 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-02T11:55:42 < jpa-> tcth: why are your nicks so weird nowadays? 2019-10-02T12:02:56 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-02T12:48:14 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-02T12:57:23 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-02T12:59:01 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-02T13:05:28 < Steffanx> What's weird about it jpa- 2019-10-02T13:05:42 < Steffanx> Now you know from what location he's tweeting 2019-10-02T13:06:08 < Steffanx> Your negativity does not help with that. 2019-10-02T13:06:32 < Steffanx> Or is it just an indication for your state of mind? :) 2019-10-02T13:19:05 < jpa-> oh, it encodes a location? 2019-10-02T13:19:18 < jpa-> h = holland, w = witzerland? 2019-10-02T13:20:54 < jpa-> Steffanx: thanks, that makes perfect sense, i just didn't realize it 2019-10-02T13:22:50 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 265 seconds] 2019-10-02T13:26:33 < Steffanx> Yeah. jpa- 2019-10-02T13:27:08 < Steffanx> Holland is home sweet home for him too, now 2019-10-02T13:27:26 < Steffanx> witzerland his work location. 2019-10-02T13:27:54 < Steffanx> So how is this - location today? 2019-10-02T13:35:33 < jpa-> rainy and windy 2019-10-02T13:35:49 < zyp> same here 2019-10-02T13:45:00 < englishman> days sure are short again 2019-10-02T13:51:23 -!- jly [uid355225@gateway/web/irccloud.com/x-ulbyehdxgkbdnvwh] has quit [Quit: Connection closed for inactivity] 2019-10-02T14:03:45 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-02T14:09:39 -!- inca [~inca@162.154.131.90] has quit [Ping timeout: 240 seconds] 2019-10-02T14:09:54 -!- inca [~inca@162.154.131.90] has joined ##stm32 2019-10-02T14:10:25 < englishman> zyp: tesla outsold all gas and diesel cars in Norway in September 2019-10-02T14:10:32 < englishman> that's pretty astounding 2019-10-02T14:10:45 < mawk> if you cut out the government subsidies you have 0 tesla sales 2019-10-02T14:10:59 < mawk> and I hope they cut it out, it's a shame to sponsor luxury cars with taxpayer's money 2019-10-02T14:11:15 < englishman> if you cut out depressive incels you have 0 lunix devs 2019-10-02T14:12:32 < zyp> englishman, did they have a wave of deliveries again? 2019-10-02T14:13:02 < englishman> guess so 2019-10-02T14:13:11 < zyp> hmm, not as big as the march one 2019-10-02T14:13:17 < zyp> https://elbilstatistikk.no 2019-10-02T14:13:35 < englishman> why all the hate for tesla, after 7 years there is still not a single bev available with the same range as a 2012 tesla 2019-10-02T14:13:37 < zyp> they delivered a ton of model 3s in march 2019-10-02T14:13:46 < englishman> that is not teslas fault 2019-10-02T14:14:37 < zyp> hmm, model 3 is catching up to other models pretty quick 2019-10-02T14:14:41 < englishman> 2/3 of Norway car sales are now bev or plug in hybrid 2019-10-02T14:18:11 < zyp> it's interesting that there's already more model 3 in norway than model X 2019-10-02T14:23:25 < englishman> model x is pretty expensive 2019-10-02T14:27:38 < zyp> not really that expensive compared to other cars that size, and EVs don't pay VAT in norway, so they are effectively getting a 20% discount vs other cars 2019-10-02T14:28:24 < sync> I wonder how the statistics were if there was no subsidy 2019-10-02T14:29:00 < zyp> presumably like all the other countries, it's the main reason norway stands out 2019-10-02T14:32:48 < englishman> here model x base model starts at more than 2x model 3, and about 80% more expensive than an Audi q7 2019-10-02T14:33:12 < englishman> it's a lot of dough for a car with no door handles 2019-10-02T14:47:42 < tcth> how is life? what are we complaining about today? :) 2019-10-02T14:53:40 < karlp> englishman: ddidn't know door handles were so close to your heart :) 2019-10-02T14:57:30 < Steffanx> About you tcth 2019-10-02T14:57:41 < tcth> Steffanx, how come? :o 2019-10-02T14:58:04 < Steffanx> What jpa- said a bit back in the logs 2019-10-02T14:58:53 < tcth> jpa-, they try to be informative 2019-10-02T15:25:52 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-02T15:35:46 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-02T15:44:43 < benishor> zyp: how come the same license plate is shared to different districts on different machines? 2019-10-02T15:44:57 < benishor> or is XX a placeholder for numbers? 2019-10-02T15:47:20 < zyp> yes 2019-10-02T15:47:42 < zyp> more like, the last two digits are hidden out of privacy concerns or something 2019-10-02T15:52:13 < benishor> I got it, thanks for clarifying 2019-10-02T16:08:47 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 245 seconds] 2019-10-02T16:10:05 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-02T16:19:41 < specing> Is it normal for the stm's calibration of the voltage reference to be waaaaaay off? 0.3V off 2019-10-02T16:20:47 < specing> ADC: 1936, factory calibrated: 1541 (@3v3), current 3v3 rail is 2.97V. 1541 * 3300mV / 1933 = 2.63V (!?) 2019-10-02T16:33:38 < jpa-> i'm not sure what you expect that calculation to give? 2019-10-02T16:34:05 < jpa-> 1541 * 3.3V/4096 = 1.24V, 1936 * 2.97V/4096 = 1.40V 2019-10-02T16:34:25 < jpa-> also check your sampling time, IIRC you can't sample the internal reference at fastest sample time 2019-10-02T16:34:40 < karlp> specing: calibration is at the voltage that it says on the datasheet, isn't it at 3V? not 3.3? 2019-10-02T16:35:37 < specing> karlp: 3v3 2019-10-02T16:35:44 < specing> +- 10mV 2019-10-02T16:35:55 < specing> at 30'C, +- 5'C 2019-10-02T16:35:55 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-02T16:37:27 < specing> jpa-: I'll check sampling time 2019-10-02T16:38:53 < specing> jpa-: I'm using slowest time psosibly, SMPR[2:0] = 111 2019-10-02T16:39:03 < specing> possible* 2019-10-02T16:40:08 < specing> 240 clock cycles at 48MHz/2 = 10 microseconds 2019-10-02T16:40:54 < specing> jpa-: that calculation is supposed to give the stm32 supply voltage 2019-10-02T16:43:14 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-02T16:45:23 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Quit: Leaving] 2019-10-02T16:46:54 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-02T16:47:29 < jpa-> specing: have you run the ADC calibration on boot? 2019-10-02T16:47:37 < jpa-> (depending on device family) 2019-10-02T16:48:00 < specing> yes, jpa- 2019-10-02T16:53:46 < specing> Maybe I should review the calibration/bootup code, it does not resemble the reference manual example 2019-10-02T16:54:07 < jpa-> which device is this? 2019-10-02T16:54:36 < specing> stm32f072rb 2019-10-02T16:55:44 < jpa-> embedded reference is specced at 1.23V, so the calibration value seems correct 2019-10-02T16:55:51 < jpa-> but your measured value seems too high 2019-10-02T16:56:16 < jpa-> did you wait for the ADC_IN17 buffer startup time? 2019-10-02T17:12:09 -!- tctw [~Tectu@82-197-160-105.init7.net] has quit [Read error: Connection reset by peer] 2019-10-02T17:28:07 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-02T17:28:44 < specing> jpa-: what do you mean by that? I can't find it in the reference manual 2019-10-02T17:29:30 < specing> I enable the voltage reference after configuring the ADC, and then go to sleep. The conversions are started by a 1Hz timer interrupt and the reference is never turned off 2019-10-02T17:29:48 < jpa-> specing: it's in datasheet, and reference manual example for temperature sensor mentions it also 2019-10-02T17:30:03 < jpa-> which sleep? 2019-10-02T17:31:09 < specing> Whatever is default without any specific configuration after the "wfi" instruction 2019-10-02T17:31:23 < specing> so, probably only the m0 core sleeps 2019-10-02T17:31:32 < jpa-> ok, that sounds like it should keep the buffer active 2019-10-02T17:32:34 < specing> The temperature readings are also totaly off (69'C) 2019-10-02T17:32:48 < specing> I'll try using one ancient firmware because I think it worked back then 2019-10-02T17:32:52 < jpa-> karlp needed a much longer sampling time than that to get reliable readings of the internal temperature sensor, so it might make sense to slow down your sampling also; your current 10µs is only a bit slower than the 4µs minimum 2019-10-02T17:33:50 < jpa-> also, are you really running the ADC at 24MHz? because the maximum specified in datasheet is 14 MHz 2019-10-02T17:34:50 < specing> No, its totaly busted there as well 2019-10-02T17:44:29 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-02T17:45:19 < bitmask> https://www.tme.com/Document/5e47640ba39fa492dbd4c0f4c8ae7b93/MR30PW%20SPEC.pdf 2019-10-02T17:45:41 < specing> jpa-: Hmm, If I'm understanding this RCC init code correctly, PCLK = HCLK = SYSCLK 2019-10-02T17:45:57 < specing> So it would seem that I'm indeed running ADC in turbo mode 2019-10-02T17:46:03 < bitmask> is the picture on the right, showing the holes as 1.7mm the footprint or the actual size 2019-10-02T17:46:41 < bitmask> so do I make the holes 1.7 or make them bigger to accommodate a leg of 1.7mm diameter 2019-10-02T17:46:49 < bitmask> I'm guessing the latter 2019-10-02T17:49:02 < jpa-> bitmask: heh, in the cad image they actual legs are drawn as 2mm 2019-10-02T17:49:27 < jpa-> ah but they thin out at pcb interface 2019-10-02T17:49:30 < bitmask> I think thats how far they stick out 2019-10-02T17:49:34 < bitmask> ? 2019-10-02T17:50:07 < jpa-> no i mean measuring from the image, no marking of the dimensions 2019-10-02T17:50:13 < bitmask> ohh 2019-10-02T17:50:16 < jpa-> 1.8mm on the lower left (pink) image 2019-10-02T17:50:24 < bitmask> I may have to order them and measure em myself 2019-10-02T17:50:30 < bitmask> why do they have to be so expensive 2019-10-02T17:51:29 < bitmask> they are pretty neat though, didnt know they had something like a 3 pin xt30 with a right angle for pcb mount 2019-10-02T17:51:52 < bitmask> do you solder the pins in that are there for support? 2019-10-02T17:51:57 < bitmask> not sure what they are made of 2019-10-02T17:58:59 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-02T18:00:37 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-02T18:11:17 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-02T18:26:02 -!- roomba [~npc@slipgate.logbook.pw] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-02T18:28:43 < specing> jpa-: still broken after switching to PCLK/4 2019-10-02T18:28:49 < specing> ah well 2019-10-02T18:30:05 < karlp> what part is this again? 2019-10-02T18:30:12 < specing> 072rb 2019-10-02T18:30:12 < karlp> (and what silicon rev?) 2019-10-02T18:30:23 < specing> How should I know that :DDD 2019-10-02T18:30:40 < karlp> first few revs of l1 for instance it was completely broken. "calibration data is not actually provided" (was garbage) 2019-10-02T18:30:55 -!- roomba [~npc@slipgate.logbook.pw] has joined ##stm32 2019-10-02T18:30:55 < karlp> there's markings on the chip itself you can line up with the datasheet, and also in the dbgmcu registers 2019-10-02T18:31:06 < karlp> (and then also check if it even is an errata for that part) 2019-10-02T18:31:12 < specing> calibration data seems okay, unless the voltage reference has become way off 2019-10-02T18:31:38 < karlp> "calibration is ok" and "after calibration, these numbers are bullshit" don't really agree with each other.... 2019-10-02T18:31:59 < specing> I'm speaking of factory calibration 2019-10-02T18:32:04 < karlp> yes, the ones in rom. 2019-10-02T18:32:16 < karlp> early revs of l1 had _values_ there, but they were not useful. 2019-10-02T18:32:29 < specing> Haha :D 2019-10-02T18:32:52 < karlp> this was on teh errata. have you looked at f0 errata? f072rb was on the first editions of disco baords of the first f0 iirc, 2019-10-02T18:33:22 < specing> I have not looked at the errata, but this is indeed a 072 discovery board 2019-10-02T18:44:32 -!- roomba [~npc@slipgate.logbook.pw] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-02T18:47:35 < aandrew> englishman: yeah I noticed that too, it seemed a lot more sudden this fall than most 2019-10-02T18:47:41 < aandrew> it's just all of a sudden dark 2019-10-02T18:48:05 -!- dexterlb [~dexterlb@qtrp.org] has joined ##stm32 2019-10-02T18:49:36 < karlp> new office right? 2019-10-02T18:49:39 -!- roomba [~npc@slipgate.logbook.pw] has joined ##stm32 2019-10-02T18:49:43 < karlp> different shadows outside 2019-10-02T18:53:52 < dexterlb> hello.. I'm trying to get PWM working (using HAL) on STM32F030F4 TIM1. Here's the code: https://github.com/DexterLB/led_controller/blob/master/stm/src/pwm.c 2019-10-02T18:53:52 < dexterlb> TIM3 and TIM14 work perfectly, but TIM1 doesn't output any signal. Although it is supposed to have more features, the HAL timer guide says nothing different about handling PWM. What am I missing? 2019-10-02T18:54:48 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Remote host closed the connection] 2019-10-02T19:01:00 < karlp> iirc there's some extra master mode bits that need to be set/cleared 2019-10-02T19:01:10 < karlp> istr others having issues hwere tim1 needed "more" 2019-10-02T19:04:17 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-02T19:13:32 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e855:edd:3495:3762] has joined ##stm32 2019-10-02T19:28:46 < dexterlb> hmm 2019-10-02T19:29:18 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-02T19:43:17 < srk> atim_bdtr_moe maybe, master ouput enable 2019-10-02T19:44:09 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-02T19:45:14 -!- boB_K7IQ [~boB_K7IQ@VIRGINIA-MA.ear2.Seattle1.Level3.net] has joined ##stm32 2019-10-02T19:56:30 < karlp> that's the one, i think 2019-10-02T20:06:54 -!- boB_K7IQ [~boB_K7IQ@VIRGINIA-MA.ear2.Seattle1.Level3.net] has quit [Ping timeout: 265 seconds] 2019-10-02T20:15:54 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-02T20:17:32 < kakimir32> LTE modem, real routed mode, gigabit ethernet, directional integrated antenna or external mimo 2019-10-02T20:18:11 < kakimir32> mikrotik routerboard + LTE card? 2019-10-02T20:32:19 < kakimir32> what pins in m.2 or ngff card are used for sim card(s)? 2019-10-02T20:32:22 < kakimir32> smbus? 2019-10-02T20:44:08 < kakimir32> why does LTE cards have GPS? 2019-10-02T20:46:19 < kakimir32> because it doesn't add cost that much if industry or automotive client needs global positioning too? 2019-10-02T20:46:46 < kakimir32> or accurate timecode 2019-10-02T20:48:04 < qyx> zyp: hows your stm32wb? did you try BLE? 2019-10-02T20:50:21 < specing> jpa-, karlp I figured out what the problem was 2019-10-02T20:50:23 < specing> I was reading temperature as if it was voltage reference and vice versa 2019-10-02T20:50:25 < specing> mixed up channel IDs 2019-10-02T20:50:42 < karlp> that would do it too :) 2019-10-02T20:50:53 < karlp> it's a pity they didn't get it stable across the whole family 2019-10-02T20:51:47 < karlp> F4 is even worse, f40x has it on 16, f42x has it on 18 2019-10-02T20:53:19 < karlp> https://zerobin.net/?8c5101dfae365a79#yauMeMpPcZRVSsUgjy1EYQbwtwCrsZMlAQ1e0tQplzg= 2019-10-02T20:54:20 < qyx> kakimir32: if you find a directional LTE antenna, let me know 2019-10-02T20:54:46 < qyx> you can get mikrotik +ith lte integrated and packaged as an outdoor unit 2019-10-02T20:55:09 < kakimir32> I have pair of iskra yagis outside 2019-10-02T20:55:21 < kakimir32> it's overkill 2019-10-02T20:55:24 < jpa-> specing: lol 2019-10-02T20:55:43 < specing> on a side note, now the temperature readings are making sense as well 2019-10-02T20:58:14 -!- banana is now known as baenana 2019-10-02T20:58:35 < kakimir32> qyx: https://www.eurodk.com/en/products/routerboards/routerboard-m11g perfect 2019-10-02T20:58:48 < kakimir32> nothing that I don't need 2019-10-02T20:59:15 < kakimir32> 1x gigabit ethernet and minipci-e 2019-10-02T21:02:27 < qyx> I have rbm33g 2019-10-02T21:02:33 < kakimir32> I see 2019-10-02T21:02:36 < kakimir32> m.2 slot 2019-10-02T21:02:42 < qyx> m.2? 2019-10-02T21:02:49 < qyx> in the m33g? 2019-10-02T21:03:04 < qyx> yeah but it is of no use 2019-10-02T21:03:36 < qyx> because there are no NVMes in that form factor 2019-10-02T21:03:36 < kakimir32> I mean I would preffer some m.2 modem 2019-10-02T21:04:06 < qyx> I have not seen any m.2 LTE modem so far 2019-10-02T21:04:24 < qyx> that doesn't mean they don't exist though 2019-10-02T21:07:07 < kakimir32> oh 2019-10-02T21:07:12 < kakimir32> I think in 5g 2019-10-02T21:07:20 < kakimir32> future 2019-10-02T21:08:32 < kakimir32> I think I would use sierra wireless airprime MC7455 2019-10-02T21:15:34 < kakimir32> what is up with them mikrotik licences 2019-10-02T21:15:41 < kakimir32> you need to buy the OS too? 2019-10-02T21:16:38 < kakimir32> router os controller? 2019-10-02T21:16:47 < kakimir32> some cloud thing 2019-10-02T21:16:48 < kakimir32> ? 2019-10-02T21:19:38 < kakimir32> https://forum.openwrt.org/t/mikrotik-rbm-33g-requires-sierra-mc7455-mpcie-modem-usb3-pins-to-be-taped/20073 2019-10-02T21:19:51 < kakimir32> they use openwrt too with mikrotik boards? 2019-10-02T21:20:09 < qyx> wat 2019-10-02T21:20:18 < qyx> you don't have to buy anything 2019-10-02T21:20:24 < qyx> he license is included in the baord price 2019-10-02T21:20:53 < qyx> the inportant thing you should check is if it is L3/L4 if you want to do wifi ap bridge (to multiple clients) 2019-10-02T21:21:04 < qyx> some boards are L3 only and don't support that 2019-10-02T21:21:12 < qyx> for your app it doesn't matter 2019-10-02T21:21:15 < qyx> and yes, you can use openwrt 2019-10-02T21:21:18 < qyx> I am using it too 2019-10-02T21:22:13 -!- baenana is now known as beenana 2019-10-02T21:23:38 < kakimir32> I want - mc7455 bridged mode to that ethernet port 2019-10-02T21:23:45 < kakimir32> dats dat 2019-10-02T21:25:53 < qyx> how are you expecting it to work? 2019-10-02T21:26:23 < qyx> idk if there is a mode which would allow you to do a bridge 2019-10-02T21:26:48 < qyx> also you said routed earlier 2019-10-02T21:27:13 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-02T21:27:54 -!- beenana is now known as banana 2019-10-02T21:29:05 < kakimir32> whoops 2019-10-02T21:29:11 < kakimir32> I meant real bridged mode 2019-10-02T21:32:47 < kakimir32> https://www.ebay.com/itm/Sierra-Wireless-EM7455-with-PCI-E-adapter-Same-as-MC7455-Generic-Ver/153240018267 EM7455 looks like m.2? 2019-10-02T21:33:33 < kakimir32> EM7455 costs like nothing compared to MC7455 2019-10-02T21:34:55 < kakimir32> at least from used markets 2019-10-02T21:38:34 -!- bvernoux1 [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-02T21:38:51 -!- bvernoux is now known as Guest40815 2019-10-02T21:39:41 -!- bvernoux1 [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Remote host closed the connection] 2019-10-02T21:40:04 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-02T21:41:39 -!- Guest40815 [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-02T21:49:30 < kakimir32> https://hackaday.com/2017/03/29/use-a-mini-pci-e-3g-card-with-usb-instead/ this is how you get internets 2019-10-02T22:02:46 < kakimir32> https://www.ebay.com/itm/WWAN-M-2-NGFF-B-Key-module-To-Mini-PCI-E-Adapter-Card-Converter-Module/264293745867 2019-10-02T22:05:30 < zyp> qyx, no, not yet 2019-10-02T22:05:44 < zyp> at the moment I wanna do zigbee, not BLE 2019-10-02T22:06:05 < qyx> for the ikea bulbs? 2019-10-02T22:07:23 < zyp> yeah 2019-10-02T22:09:55 < roomba> zigbee :( 2019-10-02T22:10:45 < roomba> zigbee could've been the kings of IoT now 2019-10-02T22:17:02 < zyp> how so and how isn't it? 2019-10-02T22:19:10 < roomba> s/IoT/diy IoT 2019-10-02T22:19:44 < kakimir32> is it a night of codings bros? 2019-10-02T22:23:17 < zyp> maybe 2019-10-02T22:24:18 < zyp> I'm recycling my old zwave code 2019-10-02T22:25:01 < zyp> partly because it's due for an overhaul, partly because one of the libraries I used broke their api so I need to rewrite how I use it anyway 2019-10-02T22:25:19 < Steffanx> So how did you extract that key again? Was it in ram or flash? 2019-10-02T22:25:59 < zyp> for the ikea stuff? it's stored in an eeprom emulation area in flash 2019-10-02T22:26:21 < Steffanx> Ah 2019-10-02T22:26:46 < Steffanx> And how did you find that out? That its eeprom emulation stuff? 2019-10-02T22:26:58 < zyp> I dumped the flash of a paired device, did a factory reset of it and dumped it again, then diffed the two dumps 2019-10-02T22:27:21 < Steffanx> Ah, the simple way. Not even real rev engineering :) 2019-10-02T22:27:53 < zyp> well, still had to figure out how the area was structured 2019-10-02T22:28:09 < Steffanx> Expected something way complicated. 2019-10-02T22:29:24 < Steffanx> Still cool though 2019-10-02T22:38:32 -!- jadew [~rcc@5-12-46-173.residential.rdsnet.ro] has joined ##stm32 2019-10-02T22:38:32 -!- jadew [~rcc@5-12-46-173.residential.rdsnet.ro] has quit [Changing host] 2019-10-02T22:38:32 -!- jadew [~rcc@unaffiliated/jadew] has joined ##stm32 2019-10-02T22:42:33 < Steffanx> Probably asked it before but what kind of zigbee usb thingy do you use zyp? 2019-10-02T22:42:56 < Steffanx> Or was it some development kit stick 2019-10-02T22:45:34 < zyp> zigbee or zwave? 2019-10-02T22:46:16 < zyp> for zigbee I'm mainly using the ikea gateway, nothing else is at a usable stage yet 2019-10-02T22:46:28 < zyp> for hacking I got some nrf52 board 2019-10-02T22:46:33 < zyp> (and the stm32wb nucleo) 2019-10-02T22:49:33 < Steffanx> I recall you sniffing some zigbee 2019-10-02T22:49:48 < Steffanx> Using wireshark 2019-10-02T22:50:06 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-02T22:50:28 < zyp> oh, yeah, that was with some spi module plugged on a rpi 2019-10-02T23:02:12 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-02T23:09:42 < qyx> I am surprised micropython + umqtt works 2019-10-02T23:10:04 < Steffanx> on what platform? stm32? 2019-10-02T23:10:22 < qyx> apparently those iot toys are becoming mature 2019-10-02T23:10:29 < Steffanx> or the fancy esp 2019-10-02T23:10:36 < qyx> no, *nix port 2019-10-02T23:10:40 < Steffanx> ah 2019-10-02T23:11:34 < qyx> I don't like the idea of handling all the hw things in micropython 2019-10-02T23:11:39 < qyx> like on esp 2019-10-02T23:11:52 < qyx> even the wifi connection 2019-10-02T23:12:06 < kakimir32> well that came out to be quite expensive modem 2019-10-02T23:12:17 < kakimir32> 130eur total 2019-10-02T23:12:33 < kakimir32> most expensive modem 2019-10-02T23:12:54 < qyx> many-G? 2019-10-02T23:13:04 < kakimir32> 4G for now 2019-10-02T23:13:08 < qyx> wut 2019-10-02T23:13:25 -!- nuxil_ [~nuxil_@185.47.251.249] has joined ##stm32 2019-10-02T23:13:26 < kakimir32> 5G cards not available yet 2019-10-02T23:13:30 < qyx> you can get 4G for 50e 2019-10-02T23:13:33 < kakimir32> but as samples maybe 2019-10-02T23:13:45 < qyx> did you see the one from mikrotik? 2019-10-02T23:13:50 < qyx> forget sierra 2019-10-02T23:14:08 < kakimir32> I ordered sierra and it was the same price that mikrodik 2019-10-02T23:14:15 < kakimir32> used ofc but CAT6 2019-10-02T23:14:27 < kakimir32> instead of CAT4 2019-10-02T23:14:27 < qyx> also check compatibility 2019-10-02T23:14:36 < kakimir32> fingers crossed 2019-10-02T23:16:24 < kakimir32> so 60eur the modem card itself, 10eur for mpcie m.2 b-keyed adapter, 30eur for routerboard, 10eur for case and 10eur for u.fl pigtails 2019-10-02T23:17:20 < qyx> speed is expensive 2019-10-02T23:17:20 -!- nuxil [~nuxil_@185.47.251.249] has quit [Ping timeout: 276 seconds] 2019-10-02T23:18:19 < zyp> what routerboard are you getting? 2019-10-02T23:18:27 < kakimir32> the cheapest 2019-10-02T23:18:36 < zyp> I'm not following context here 2019-10-02T23:19:03 < kakimir32> RBM11G 2019-10-02T23:19:14 < kakimir32> exactly what I need and nothing more 2019-10-02T23:19:24 < zyp> what are you trying to do? 2019-10-02T23:19:35 < kakimir32> just simple LTE modem 2019-10-02T23:19:51 < kakimir32> LTE in -> Gigabit ethernet out 2019-10-02T23:20:23 < zyp> why not get one of the LTE boxes? 2019-10-02T23:20:42 < Steffanx> hmm, i was curious how you did that again zyp. The module spi -> rpi -> wireshark, but or jpa wasnt around or i just cant find it in the logs :( 2019-10-02T23:20:42 < kakimir32> you mean mikrodik boxes? 2019-10-02T23:20:53 < zyp> https://mikrotik.com/product/wap_lte_kit <- like this shit 2019-10-02T23:21:00 < zyp> Steffanx, can look it up later 2019-10-02T23:21:30 < zyp> now is later 2019-10-02T23:21:30 < zyp> https://openlabs.co/store/Raspberry-Pi-802.15.4-radio 2019-10-02T23:24:09 < kakimir32> zyp: because I wanted sierra wireless card, no wlan and gigabit ethernet 2019-10-02T23:24:22 < kakimir32> and I have external antennas installed 2019-10-02T23:24:39 < kakimir32> just connect them smas 2019-10-02T23:24:53 < Steffanx> ah, that was it indeed. it seems it's not in jpa-logs. 2019-10-02T23:24:54 < Steffanx> ty zyp 2019-10-02T23:27:08 < kakimir32> my last modem had pings of 20ms but it rotted 2019-10-02T23:27:21 < kakimir32> current one has 50ms ping 2019-10-02T23:27:32 < zyp> what speeds are you getting over LTE? 2019-10-02T23:27:55 < kakimir32> 50/50 nowdays 2019-10-02T23:28:04 < zyp> then what do you need gigabit for? 2019-10-02T23:28:22 < kakimir32> for long time they forgot to choke the connection and it went all way to 140 or so 2019-10-02T23:28:28 < zyp> heh 2019-10-02T23:28:54 < kakimir32> latency zyp 2019-10-02T23:30:04 < qyx> I only have 10/5 LTE 2019-10-02T23:30:14 < qyx> it was 25/25 in the past 2019-10-02T23:34:02 < kakimir32> external antennas are just a measure to keep signals so high that even in rush hour it should get the most out of tight time slot it has 2019-10-02T23:36:34 < qyx> so did they improve anything? 2019-10-02T23:36:34 < kakimir32> I don't recall it ever going under 50/50 but when connection was not choked before I could see transfer speeds falling close to 50 in the evenings 2019-10-02T23:36:40 < qyx> also, are they legal? 2019-10-02T23:36:45 < kakimir32> before external antennas 2019-10-02T23:38:00 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-02T23:38:02 < kakimir32> idk. those are sold here freely and there is no legal texts on their pages 2019-10-02T23:38:21 < kakimir32> passive antennas after all 2019-10-02T23:39:29 < qyx> passive antennas have gain roo 2019-10-02T23:39:31 < qyx> roo 2019-10-02T23:39:34 < qyx> fuk, too 2019-10-02T23:40:08 < zyp> there's probably a legal limit on EIRP 2019-10-02T23:40:42 < zyp> as long as txpower is not too high, antenna should be legal 2019-10-02T23:41:00 < kakimir32> I would assume the system adjusts the power itself 2019-10-02T23:41:29 < zyp> system doesn't know antenna gain 2019-10-02T23:41:47 < zyp> EIRP is txpower multiplied by antenna gain 2019-10-02T23:41:56 < kakimir32> but it might know signal level at receiving end 2019-10-02T23:42:43 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-02T23:43:07 < qyx> so I shjould get two yagis for B3 and mount them at 90deg angles 2019-10-02T23:44:00 < kakimir32> there might be different limits for handheld tranceivers and fixed installations zyp 2019-10-02T23:44:21 < kakimir32> I don't really care 2019-10-02T23:44:59 < kakimir32> it's nothing that power these devices transmit 2019-10-02T23:45:24 < kakimir32> compare it to the cell tower 2019-10-02T23:46:22 < kakimir32> some basejump guys say that you feel warm if you go in front of those antennas 2019-10-02T23:47:56 < Steffanx> Placebo? 2019-10-02T23:49:31 < kakimir32> microwaves to your skin at hundreds or thousands watts of power? 2019-10-02T23:51:06 < qyx> lte is not thousands watts 2019-10-02T23:51:14 < qyx> https://www.dipol.sk/lte_mimo_antena_atk-log_lte_800-2170mhz_5m_kabel_sma_plug__A7056_5.htm 2019-10-02T23:51:25 < qyx> this one I like 2019-10-02T23:51:47 < qyx> actually now I have only a single antenna 2019-10-02T23:51:50 < qyx> so no MIMO 2019-10-02T23:56:49 < kakimir32> interesting mimo yagi 2019-10-02T23:56:53 < kakimir32> I have never seen such --- Day changed Thu Oct 03 2019 2019-10-03T00:01:49 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e855:edd:3495:3762] has quit [Quit: Konversation terminated!] 2019-10-03T00:10:00 < PaulFertser> What about this weird yagi: https://1000kanalov.ru/wa-data/public/shop/products/96/28/2896/images/3420/3420.970.jpg ? It's meant for DVB-T2 but why, why there're three of them? You're supposed to point it directly on the nearest transmitter and it's not like you can only turn it horizontally. 2019-10-03T00:11:35 < qyx> 1000kanalov :D 2019-10-03T00:13:08 < kakimir32> tv antenna 2019-10-03T00:13:34 < kakimir32> weirder it looks the better 2019-10-03T00:14:06 < PaulFertser> I know ~nothing about RF but I would guess those 3 antennas make it just less directional and more expensive. So it's like there's no sense at all in doing that? 2019-10-03T00:15:27 < kakimir32> signal path varies 2019-10-03T00:15:36 < kakimir32> with weather and season 2019-10-03T00:16:29 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 276 seconds] 2019-10-03T00:16:45 < kakimir32> but idk if it's about that 2019-10-03T00:23:11 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-03T00:35:33 < PaulFertser> I guess jadew knows for sure 2019-10-03T00:50:22 < catphish> i love how unambiguous this field definition is: "A RESPONSE DATA FORMAT field value of two indicates that the data shall be in the format defined in this manual. Response data format valuesless than two are obsolete. Response data format values greater than two are reserved." 2019-10-03T00:50:35 < catphish> i guess i'll go with 2 then :) 2019-10-03T00:55:57 < qyx> I would not be sure 2019-10-03T01:01:22 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Read error: Connection reset by peer] 2019-10-03T01:30:36 < mawk> I'm in Delft Steffanx 2019-10-03T01:30:42 < mawk> I got the interview today 2019-10-03T01:30:46 < mawk> they're happy it seems 2019-10-03T01:30:56 < roomba> they have a foosball table, right? 2019-10-03T01:31:14 < mawk> lol 2019-10-03T01:31:15 < mawk> didn't see 2019-10-03T01:31:18 < Steffanx> Aha cool cool mawk 2019-10-03T01:31:37 < roomba> better find that out 2019-10-03T01:31:40 < Steffanx> Did you use your bestest dutch? 2019-10-03T01:31:50 < mawk> no, english 2019-10-03T01:31:54 < mawk> but I told them I saw dutch tv 2019-10-03T01:32:04 < mawk> the show where you have criminal cases and you can call if you're a witness 2019-10-03T01:32:15 < Steffanx> Lol 2019-10-03T01:32:17 < roomba> oh god that sounds awesome 2019-10-03T01:32:26 < roomba> imagine if they had that show in the USA 2019-10-03T01:32:28 < mawk> they show security footage with the heads of the guys in clear view 2019-10-03T01:32:35 < Steffanx> Opsporing verzocht 2019-10-03T01:32:36 < mawk> 100% of the time it's "light tan" people 2019-10-03T01:32:38 < mawk> yes ! 2019-10-03T01:33:34 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T01:34:21 < Steffanx> You did pick the perfect time. 2019-10-03T01:34:37 < Steffanx> With the farmers demonstration and the awesome weather 2019-10-03T01:34:39 < mawk> lol 2019-10-03T01:34:40 < mawk> yeah 2019-10-03T01:34:45 < mawk> I saw it on tv 2019-10-03T01:34:58 < mawk> the green party guy getting booed and middle fingers, and geert wilders getting cheers 2019-10-03T01:35:17 < mawk> standing on a tractor 2019-10-03T01:35:18 < Steffanx> Heh yesh 2019-10-03T01:40:10 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-03T01:52:58 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-03T01:53:38 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T01:57:17 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Ping timeout: 240 seconds] 2019-10-03T02:14:57 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 240 seconds] 2019-10-03T02:19:51 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-03T02:47:00 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-03T03:55:09 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-03T04:07:22 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-03T04:26:41 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-03T04:28:14 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-03T04:30:51 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has quit [Ping timeout: 240 seconds] 2019-10-03T05:30:47 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Disconnected by services] 2019-10-03T05:30:49 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-03T06:32:04 < qyx> lol you have demonstrating farmers too Steffanx 2019-10-03T06:40:05 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-03T06:42:43 -!- fc5dc9d4_ [~quassel@p5B081376.dip0.t-ipconnect.de] has joined ##stm32 2019-10-03T06:47:08 -!- fc5dc9d4 [~quassel@p5B081358.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds] 2019-10-03T06:48:42 < R2COM> https://www.youtube.com/watch?v=0BvfVM0sBfI&feature=youtu.be 2019-10-03T07:39:12 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-03T07:43:02 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-03T07:43:02 -!- day__ is now known as day 2019-10-03T08:36:25 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-03T09:05:15 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-03T09:12:23 < jpa-> Steffanx: or maybe i just had zyp on ignore back then! ;) 2019-10-03T09:15:20 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-03T09:23:30 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Read error: Connection reset by peer] 2019-10-03T09:29:21 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-03T09:36:11 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-03T09:51:45 < Steffanx> Hah, I think that was it. jpa- 2019-10-03T09:56:17 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-03T09:56:43 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T10:04:52 -!- renn0xtk9 [~max@2a02:810d:1540:2448:6452:b718:6cc4:dcee] has joined ##stm32 2019-10-03T10:06:51 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-03T10:16:11 -!- jly [uid355225@gateway/web/irccloud.com/x-hnpiihnonukkckfd] has joined ##stm32 2019-10-03T10:20:51 -!- nuxil_ [~nuxil_@185.47.251.249] has quit [Ping timeout: 240 seconds] 2019-10-03T10:23:40 -!- tprrt [~tprrt@upc31-1-78-208-110-13.fbx.proxad.net] has joined ##stm32 2019-10-03T10:26:28 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-03T10:50:26 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-03T10:52:38 -!- User__ [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-03T10:54:36 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-03T10:55:15 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 240 seconds] 2019-10-03T10:57:02 < grummund> stm32 now in 8-pin package 2019-10-03T10:58:17 < grummund> https://www.st.com/content/st_com/en/products/microcontrollers-microprocessors/stm32-32-bit-arm-cortex-mcus/stm32-mainstream-mcus/stm32g0-series/stm32g0x0-value-line/stm32g030j6.html 2019-10-03T10:59:44 < jly> hmm just now? 2019-10-03T11:00:27 < jly> ready for IoT.................... 2019-10-03T11:01:35 < grummund> not sure when released but they are marketing it now 2019-10-03T11:02:19 -!- User__ [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-03T11:02:49 -!- User__ [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-03T11:08:31 < jpa-> i wonder how you'll enter debug mode on one of those if you disable NRST pin and enter powersave as soon as you boot :) 2019-10-03T11:11:22 < grummund> presumably you'd need to keep SWD pins unassigned 2019-10-03T11:12:03 < jpa-> that would be silly on 8-pin device though 2019-10-03T11:12:43 < grummund> yeah just don't write bugs :D 2019-10-03T11:16:23 < jpa-> hm, i can't actually find info how one can disable NRST there, even though they've provided alternate functions on the NRST pin so i assume it must be possible 2019-10-03T11:21:44 < jpa-> seems like it should be set by the option bits, and that it will always be NRST on power-on 2019-10-03T11:21:58 < jpa-> so to recover from bugs, one would hold reset down while booting, connect by SWD and then release reset 2019-10-03T11:22:54 < grummund> i.e. at power-on 2019-10-03T11:27:55 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Excess Flood] 2019-10-03T11:28:13 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-03T11:30:24 < zyp> aww, it doesn't have usb 2019-10-03T11:30:27 -!- User__ [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-03T11:30:56 -!- User__ [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-03T11:31:04 < jpa-> time to bitbang usb ls 2019-10-03T11:32:40 < grummund> zyp: g031 has usb 2019-10-03T11:33:09 < grummund> STM32G031Jx SO8N pinout 2019-10-03T11:33:58 < grummund> hmm, maybe 2019-10-03T11:34:15 < zyp> doesn't look like 2019-10-03T11:34:37 < grummund> yeah, i closed the link now but the main page said USB. 2019-10-03T11:36:40 < zyp> doesn't really matter 2019-10-03T11:37:35 < zyp> SO8N has few legs, but that in itself isn't a design goal, and area-wise it isn't really that small 2019-10-03T11:37:48 < jpa-> SO8 is huge anyway, many of the BGA/CSP packages are smaller 2019-10-03T11:37:56 < zyp> yeah, or even QFN 2019-10-03T11:38:06 < zyp> for something that's doable on standard specs 2019-10-03T11:38:32 < zyp> according to the datasheet there's a 4x4mm qfn28 2019-10-03T11:38:41 < zyp> SO8N is 4.9x6mm 2019-10-03T11:41:48 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-builfmjqrfvlnjva] has joined ##stm32 2019-10-03T11:42:00 < grummund> zyp: it's being marketed as competetive cost vs. 8-bit, not physical size. 2019-10-03T11:42:18 < grummund> that said i can't find any for sale 2019-10-03T11:42:33 < grummund> but should be pennies 2019-10-03T11:43:54 < zyp> stm32 has been at competitive cost for years 2019-10-03T11:44:17 < zyp> and a package with fewer pins doesn't really make it cheaper 2019-10-03T11:47:40 < grummund> 37 cents may be cheapest yet 2019-10-03T11:54:04 < grummund> looks like the usb parts are currently vapourware 2019-10-03T11:56:26 -!- Mangy_Dog [~Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-03T12:11:37 < dexterlb> srk, karlp: I tried setting MOE, it doesn't help: TIM1 still outputs nothing. Sorry for the late reply, I was away from my setup 2019-10-03T12:26:00 -!- renn0xtk9 [~max@2a02:810d:1540:2448:6452:b718:6cc4:dcee] has quit [Remote host closed the connection] 2019-10-03T12:27:02 -!- renn0xtk9 [~max@2a02:810d:1540:2448:4a:7f91:6d12:fd85] has joined ##stm32 2019-10-03T12:50:02 -!- User__ [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 240 seconds] 2019-10-03T13:04:30 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-03T13:19:06 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-03T13:22:30 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-03T13:22:32 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Excess Flood] 2019-10-03T13:22:54 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-03T13:47:51 < Thorn> soyuz ms-12 landing https://www.youtube.com/watch?v=21X5lGlDOfg 2019-10-03T13:57:43 -!- renn0xtk9 [~max@2a02:810d:1540:2448:4a:7f91:6d12:fd85] has quit [Remote host closed the connection] 2019-10-03T13:58:57 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a896:3587:78d5:24e8] has joined ##stm32 2019-10-03T14:01:25 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-builfmjqrfvlnjva] has quit [Quit: Connection closed for inactivity] 2019-10-03T14:15:49 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has joined ##stm32 2019-10-03T14:25:49 < englishman> https://cloudblogs.microsoft.com/opensource/2019/08/28/exfat-linux-kernel/ 2019-10-03T14:42:04 < karlp> I feel like I read that months ago? 2019-10-03T14:42:13 < Ultrasauce> Microsoft ♥ Linux – we say that a lot 2019-10-03T14:45:49 < jadew> so... paypal is full of lies 2019-10-03T14:46:29 < jadew> they said they're unable to send money to mastercard, yet I managed to do it last week 2019-10-03T14:46:55 < jadew> now the option is gone and I called them and they said it's impossible to do it 2019-10-03T14:47:01 < jadew> then they checked and said it was a glitch 2019-10-03T14:47:08 < jadew> so the glitch was that it worked... 2019-10-03T14:47:38 < jadew> and you can only withdraw in EUR 2019-10-03T14:48:01 < jadew> so if you get paid in any other currency, you have to pay them extra for the conversion 2019-10-03T14:48:06 < jadew> there's no way around it 2019-10-03T14:48:20 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a896:3587:78d5:24e8] has quit [Quit: Konversation terminated!] 2019-10-03T14:52:25 < karlp> paypal are greedy scammers, news at 9.... 2019-10-03T14:55:12 < jadew> karlp, I do feel scammed tbh 2019-10-03T14:55:19 < Steffanx> Scampal I heard 2019-10-03T14:55:36 < jadew> it's like they won't let you go with your money unless they take a bigger cut than you initially thought they would 2019-10-03T15:04:33 < jadew> I think I know what caused the glitch 2019-10-03T15:04:49 < jadew> I had a VISA card attached and I removed it the other day 2019-10-03T15:05:15 < jadew> anyway, this annoys me so much 2019-10-03T15:10:01 < dongs> y ou can always use your illicit scampal gains to simplyt order shit with paypal from other sites 2019-10-03T15:10:53 < Ultrasauce> the first iteration of muskscrip 2019-10-03T15:11:12 < jadew> dongs, because I have money in GBP and nobody wants GBP 2019-10-03T15:11:21 < jadew> which means I'll have to convert it to USD 2019-10-03T15:11:22 < dongs> well, obviously UK does? 2019-10-03T15:11:31 < dongs> order something from UK shopping sites. 2019-10-03T15:11:36 < dongs> before they brefit 2019-10-03T15:11:38 < dongs> prexit 2019-10-03T15:11:40 < dongs> brexit 2019-10-03T15:11:40 < dongs> wahtever 2019-10-03T15:11:55 < jadew> this is company money... I don't need anything from the UK 2019-10-03T15:14:00 < karlp> let me guess, the real problem is that you put your margins on your products too low.... 2019-10-03T15:14:37 < jadew> karlp, well, that is a different story 2019-10-03T15:14:57 < jadew> even if they were extremely high, I would still feel cheated by this 2019-10-03T15:15:42 < karlp> cost of doing business. 2019-10-03T15:15:53 < karlp> how do you think merchants feel about ccard fees? 2019-10-03T15:16:22 < karlp> why do you think merchants offer things like paypal, (if they can use paypal money) to try and avoid other fees. 2019-10-03T15:19:58 < jadew> karlp, don't know.. think it's higher than 8.7%? 2019-10-03T15:20:23 < jadew> paypal takes 5% when I make a sale and 3.7% when I withdraw (because it's forcing the currency conversion) 2019-10-03T15:21:38 < karlp> so, why are you using it again? 2019-10-03T15:21:47 < jadew> ebay 2019-10-03T15:21:58 < jadew> they don't give you another option 2019-10-03T15:22:24 < karlp> if there's no option, then cost of doing business on ebay is x 2019-10-03T15:22:34 < karlp> if you're not including that, and just being upset by it, then that's on you 2019-10-03T15:22:50 < jadew> well.. I am including it, but I'm still upset 2019-10-03T15:54:16 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-03T16:04:07 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-03T16:25:56 -!- jly [uid355225@gateway/web/irccloud.com/x-hnpiihnonukkckfd] has quit [Quit: Connection closed for inactivity] 2019-10-03T16:33:54 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-03T16:39:24 < dongs> https://github.com/ocornut/imgui/issues/2529#issuecomment-502853949 kikecad users rejoice 2019-10-03T16:40:39 < jadew> neat 2019-10-03T16:40:46 < jadew> how did you come across that? 2019-10-03T16:40:58 < dongs> updating my version of imgui 2019-10-03T16:41:01 < dongs> and clicked 'screenshots' thread 2019-10-03T16:41:06 < dongs> from latest release 2019-10-03T16:42:42 < dongs> https://forum.kicad.info/t/another-symbol-generator-for-kicad/17619/12 this is the thread 2019-10-03T16:43:36 < jadew> I'd probably do the symbols myself anyway 2019-10-03T16:43:38 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-03T16:43:52 < jadew> I like to group pins based on function 2019-10-03T16:44:01 < jadew> what I want to see in kicad is colored nets 2019-10-03T16:44:18 < dongs> same, but for initial import i use altium's smart paste thing from a xls which i generate from datasheet/massage a bit to my liking 2019-10-03T16:44:28 < dongs> then that gives me pin#, name, description, io type 2019-10-03T16:44:34 < dongs> and i can move those around as i want to make the symbol 2019-10-03T16:44:54 < jadew> yeah, that sounds faster 2019-10-03T16:46:28 < karlp> do you customize the symbol of say an stm32 for each project? moving pins to where they are used in that project and so on? 2019-10-03T16:46:37 < dongs> haha no 2019-10-03T16:46:37 < karlp> or do you have the same symbol for the part no matter the project 2019-10-03T16:46:41 < dongs> same symbol 2019-10-03T16:46:46 < dongs> by "move around" i mean 2019-10-03T16:46:53 < dongs> waht judew said, arrange pins by functio nor wahtever 2019-10-03T16:47:00 < dongs> i.e. for stm32 i might have them grouped by port 2019-10-03T16:47:02 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-03T16:47:04 < dongs> PortA PortB etc 2019-10-03T16:47:10 < dongs> for larger pincount stms 2019-10-03T16:47:20 < dongs> for shit like F103 its just a rectangle with 24 pins on 2 sides 2019-10-03T16:47:36 < dongs> other stuff would have like inputs on left, outputs on right, reset/i2c/etc on bottom left and such 2019-10-03T16:48:23 < dongs> grounds on bottom, vcc/whaever on top 2019-10-03T16:48:35 < dongs> i generally dislike symbols that are just 1:1 pinout of chip 2019-10-03T16:48:47 < dongs> cuz then ground/power stuff looks dumb 2019-10-03T16:59:53 < dongs> gui making thingy 2019-10-03T17:00:10 < jadew> looks nice 2019-10-03T17:00:27 < jadew> is that for embedded stuff? 2019-10-03T17:00:40 < dongs> nah desktop 2019-10-03T17:00:49 < dongs> its all c++ filth 2019-10-03T17:01:11 < zyp> I dicked around with imgui a few months ago 2019-10-03T17:01:18 < jadew> doesn't look nice enough for desktop tho... 2019-10-03T17:02:01 < zyp> it's originally made for devtools in games and stuff 2019-10-03T17:02:13 < jadew> haha Haohmaru 2019-10-03T17:02:25 < zyp> i.e. it can render shit on top of an opengl context 2019-10-03T17:02:29 < jadew> zyp, that makes sense 2019-10-03T17:02:38 < dongs> *directx context 2019-10-03T17:02:43 < dongs> nobody fucking wirtes games in opengl in 2019 2019-10-03T17:02:49 < dongs> last opengl game = tuxracer 2019-10-03T17:02:51 < zyp> either, also vulkan 2019-10-03T17:03:10 < dongs> literally everyone who is making money 2019-10-03T17:03:37 < zyp> isn't all the money in phone games nowadays, with opengl es? 2019-10-03T17:05:06 < jadew> kid of a friend made a game, earned like $400 from ads in the first day 2019-10-03T17:07:03 < jadew> I think the trick is to make something that doesn't make you tense when you're shitting 2019-10-03T17:07:18 < jadew> otherwise people won't play it 2019-10-03T17:07:52 < Steffanx> Tuxracer? 2019-10-03T17:08:12 < jadew> I tried that game... doesn't seem like a game 2019-10-03T17:08:25 < Steffanx> TIM 2019-10-03T17:11:27 < qyx> cs? 2019-10-03T17:24:29 < dongs> hmm 2019-10-03T17:24:29 < dongs> wtf 2019-10-03T17:53:05 < Steffanx> Ty Haohmaru 2019-10-03T18:27:04 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 264 seconds] 2019-10-03T18:52:57 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-03T18:54:16 -!- learningc [~learningc@121.121.98.53] has quit [Read error: Connection reset by peer] 2019-10-03T19:00:41 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-03T19:02:19 < antto> this game: https://www.youtube.com/watch?v=FBgk5kSjn5M 2019-10-03T19:03:11 < dongs> is that a lunix game 2019-10-03T19:03:17 < antto> yez 2019-10-03T19:03:22 < dongs> lul. 2019-10-03T19:03:49 < antto> mm, level6 looks moar like taking a sh*t 2019-10-03T19:06:13 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-03T19:07:54 < antto> hahaha, muh cat is watching a gish blob in a video 2019-10-03T19:14:52 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-03T19:21:58 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-03T19:23:02 < aandrew> https://imgur.com/a/W7bu9Tt *RAGE* 2019-10-03T19:23:41 < aandrew> who the fuck indents code like that 2019-10-03T19:26:07 < Steffanx> Sry 2019-10-03T19:26:27 < dongs> who teh fuck uses vs.choad 2019-10-03T19:26:33 < dongs> especially with such a disgusting ass color scheme 2019-10-03T19:27:09 < aandrew> who's using vs? 2019-10-03T19:27:45 < dongs> is this some ethernet trash 2019-10-03T19:27:50 < dongs> but yeah that indent is total aids 2019-10-03T19:29:20 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T19:34:48 < dongs> https://github.com/libusb/libusb/issues/86 haha opensores 2019-10-03T19:41:25 < kakimir32> oh wow 2019-10-03T19:41:59 < dongs> its happening(tm) 2019-10-03T19:42:39 < kakimir32> what? 2019-10-03T19:42:56 < dongs> hotplug detection on winduhs 2019-10-03T19:43:00 < dongs> and me going to sleep 2019-10-03T19:44:20 < kakimir32> ok 2019-10-03T19:44:27 < kakimir32> sleep tight 2019-10-03T19:50:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-03T19:54:37 < Steffanx> Welcome mr kakimir32 2019-10-03T19:54:56 < kakimir32> hello papa 2019-10-03T19:54:59 < Thorn> kakimirv7-m 2019-10-03T19:55:02 < Steffanx> papa is dead. 2019-10-03T19:55:13 < Steffanx> its just mr x now, maybe Steffanx 2019-10-03T19:55:14 < kakimir32> son 2019-10-03T20:01:29 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-03T20:03:42 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-03T20:06:37 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-03T20:16:08 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-03T20:18:50 < jpa-> https://github.com/gavanderhoorn/rep-ros-i/blob/a88596b8a12b95a1d3bfb0987fa23389c686199a/rep-ixxxx.rst#tri-states oh, how lovely, true = 0; just what i'd expect 2019-10-03T20:19:39 < zyp> just like the classic enum { True, False, FileNotFound }; 2019-10-03T20:24:05 < jpa-> :) 2019-10-03T20:24:38 < zyp> (https://thedailywtf.com/articles/What_Is_Truth_0x3f_) 2019-10-03T20:25:00 < Thorn> who needs that when you have sql 2019-10-03T20:29:09 < jpa-> zyp: yeah, back when dailywtf was still good :) 2019-10-03T20:34:02 < zyp> I don't know if it's good or bad nowadays, I haven't followed for years 2019-10-03T20:34:19 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-03T20:45:01 -!- Drzacek [~Drzacek@2001:16b8:1ced:6d00:cda9:d572:7e4:8488] has joined ##stm32 2019-10-03T21:14:14 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T21:16:28 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-03T21:21:56 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-03T21:35:17 < kakimir32> so I repaired NTFS 2019-10-03T21:35:38 < kakimir32> no files became visible 2019-10-03T21:36:09 < kakimir32> MTF has only header and then 156megabytes of zeros 2019-10-03T21:36:27 < kakimir32> same thing for logfile etc. 2019-10-03T21:36:48 < kakimir32> so now just photorec it 2019-10-03T21:38:01 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-03T21:38:20 < kakimir32> expecting to find like 200000files 2019-10-03T21:38:26 < kakimir32> oh boy 2019-10-03T21:39:05 < kakimir32> it's all just junk like GTA san andreas game files 2019-10-03T21:41:56 < kakimir32> there was MTF header in MTF mirror 2019-10-03T21:42:04 < kakimir32> MFT header 2019-10-03T21:42:13 < kakimir32> but I dont recall seeing file records after that 2019-10-03T21:42:36 < kakimir32> I need to check it again after photorec 2019-10-03T21:44:40 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Remote host closed the connection] 2019-10-03T21:45:11 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-03T21:47:13 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-03T21:47:33 < Steffanx> time to trash this shit kakimir32 2019-10-03T21:48:33 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds] 2019-10-03T21:59:29 < kakimir32> kkm32-filesystem 2019-10-03T21:59:56 < kakimir32> "who needs files anyway?" 2019-10-03T22:00:06 < Steffanx> yeah 2019-10-03T22:00:25 -!- boB_K7IQ [~boB_K7IQ@c-73-11-172-226.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T22:04:30 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-03T22:17:07 < mawk> I had stroopwafels at coffee time 2019-10-03T22:17:25 < Steffanx> ohno 2019-10-03T22:17:29 < Steffanx> did you have real one from a market 2019-10-03T22:17:36 < Steffanx> or kanjer stroopwafels? 2019-10-03T22:17:46 < mawk> it was from a restaurant 2019-10-03T22:18:03 < mawk> Hajé, in a highway rest area on the way to Drachten 2019-10-03T22:18:40 < Steffanx> why on earth are you going to drachten of all places? 2019-10-03T22:19:24 -!- Drzacek [~Drzacek@2001:16b8:1ced:6d00:cda9:d572:7e4:8488] has quit [Quit: Leaving] 2019-10-03T22:20:00 < mawk> lol 2019-10-03T22:20:05 < mawk> my family is there 2019-10-03T22:20:21 < mawk> oma mawk, pake mawk 2019-10-03T22:20:22 < Steffanx> i see 2019-10-03T22:20:29 < mawk> pake mawk is in a funerary urn tho 2019-10-03T22:20:35 < Steffanx> Real friezen. 2019-10-03T22:20:36 < mawk> but the oma is still alive 2019-10-03T22:20:40 < mawk> yes 2019-10-03T22:21:12 < Steffanx> Fryslan boppe. 2019-10-03T22:21:20 < mawk> the language is weird 2019-10-03T22:21:26 < mawk> they talk it among themselves 2019-10-03T22:21:30 < mawk> frysian 2019-10-03T22:21:30 < Steffanx> yeah 2019-10-03T22:22:11 < Steffanx> Yeah its weird 2019-10-03T22:23:33 < mawk> maybe one day I'll learn it, so that truly no-one can understand me 2019-10-03T22:23:41 < mawk> outside of friesland 2019-10-03T22:25:32 < Steffanx> You already speak french. Not many in dutchland speak french 2019-10-03T22:26:54 < Steffanx> All you have to do is visit dumpert.nl and install the dumpert app for the sound board 2019-10-03T22:27:00 < Steffanx> For the common well known dutch phrases 2019-10-03T22:27:22 < Steffanx> "Hedde drugs op" "dit is niet mijn winkel vriend" "blikje in die water" etc. 2019-10-03T22:27:57 -!- boB_K7IQ [~boB_K7IQ@c-73-11-172-226.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-03T22:28:51 < bitmask> someone make me some altium footprints, ok thanks 2019-10-03T22:29:04 < Steffanx> what kind of weird footprints you need? 2019-10-03T22:29:21 < bitmask> very weird, mr30pw, xt30pw 2019-10-03T22:33:06 < Steffanx> looks easy enough 2019-10-03T22:33:43 < bitmask> yea, im just not sure if the dimensions on the 'datasheet' are actual dimensions or footprint/hole dimensions 2019-10-03T22:34:00 < Steffanx> dont forget to add your 3d model https://www.3dcontentcentral.com/Search.aspx?arg=mr30pw&SortBy=match&PageSize=10 2019-10-03T22:34:18 < bitmask> nice, thanks 2019-10-03T22:36:06 < mawk> I repaired oma's iPad and another oma's document scanner 2019-10-03T22:36:18 < mawk> "oh mawk you're studying computers right ? can you fix mine ?" 2019-10-03T22:37:00 < Steffanx> repaired an ipad? 2019-10-03T22:37:16 < mawk> yes 2019-10-03T22:37:24 < Steffanx> as in ..installed windows? 2019-10-03T22:37:24 < mawk> icloud synchronization something something 2019-10-03T22:37:27 < Steffanx> oh 2019-10-03T22:37:28 < mawk> software repair 2019-10-03T22:37:29 < mawk> lol 2019-10-03T22:37:29 < mawk> no 2019-10-03T22:37:56 < mawk> if it were me I'd just throw it away, but the oma wanted to keep it 2019-10-03T22:38:20 < Steffanx> it works kinda well for older people.. 2019-10-03T22:50:45 < zyp> https://giphy.com/gifs/fYr43caEW3tRSbFflh/fullscreen bus stops in norway 2019-10-03T22:53:11 < Steffanx> hah 2019-10-03T22:59:14 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-03T23:00:23 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-03T23:12:16 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-03T23:14:31 -!- tprrt [~tprrt@upc31-1-78-208-110-13.fbx.proxad.net] has quit [Quit: leaving] 2019-10-03T23:31:17 -!- jon101249 [~jon1012@foresight/developer/jon1012] has quit [Ping timeout: 240 seconds] 2019-10-03T23:40:36 < aandrew> full screen bus stops? 2019-10-03T23:40:48 < aandrew> that's pretty much exactly what ours look like, minus the driving water 2019-10-03T23:43:06 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-03T23:46:43 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 268 seconds] --- Day changed Fri Oct 04 2019 2019-10-04T00:06:47 < aandrew> hm, something I've never bothered to try before -- with the USART devices is it really one transmit empty interrupt = one char can be written to TDR? 2019-10-04T00:06:55 < aandrew> there's no internal fifo that I can try to fill? 2019-10-04T00:07:16 < aandrew> so if I have a RAM buffer of 1k characters and I'm transmitting via interrupts, I generate 1k interrupts? 2019-10-04T00:10:18 < aandrew> I guess they want to encourage the use of DMA for transmitting 2019-10-04T00:12:58 < bitmask> fpc cable, clamshell vs up drawer vs down drawer.... GO! 2019-10-04T00:18:37 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-04T00:18:55 < karlp> what? 2019-10-04T00:20:46 < mawk> yes aandrew 2019-10-04T00:21:01 < mawk> even for receiving I guess, in a hackish way 2019-10-04T00:29:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-04T00:39:58 < aandrew> receiving with DMA is stupid unless you have well defined incoming packets and even then it's kind of shite becuase errors happen 2019-10-04T00:40:03 < aandrew> so you need a timeout mechanism anyway 2019-10-04T01:02:48 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 246 seconds] 2019-10-04T01:04:15 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-04T01:13:30 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-04T02:06:40 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 264 seconds] 2019-10-04T02:11:21 -!- Mangy_Dog [~Mangy_Dog@05449c7a.skybroadband.com] has quit [Remote host closed the connection] 2019-10-04T02:53:34 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-04T02:57:08 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-04T03:05:22 < karlp> yeah, it's a reallllly minor optimization for rx 2019-10-04T03:23:15 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-04T03:27:40 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-04T03:33:33 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-04T03:36:22 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-04T03:40:37 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 240 seconds] 2019-10-04T03:41:00 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-04T03:44:20 < Laurenceb> die cislunar scum 2019-10-04T04:01:41 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-04T04:18:26 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-04T04:26:10 -!- c4017_ [~c4017@S010664777dab66f3.vf.shawcable.net] has quit [Read error: Connection reset by peer] 2019-10-04T04:26:30 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Read error: Connection reset by peer] 2019-10-04T04:26:36 -!- c4017_ [~c4017@S010664777dab66f3.vf.shawcable.net] has joined ##stm32 2019-10-04T04:26:54 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-04T04:30:27 -!- Lux [~Luggi09@parabox.it-syndikat.org] has quit [Ping timeout: 245 seconds] 2019-10-04T04:30:52 -!- benishor [~benishor@95.85.48.123] has quit [Ping timeout: 245 seconds] 2019-10-04T04:31:36 -!- benishor [~benishor@95.85.48.123] has joined ##stm32 2019-10-04T04:33:34 -!- Lux [~Luggi09@parabox.it-syndikat.org] has joined ##stm32 2019-10-04T04:44:00 -!- turnip420 [~machinehu@184.71.172.142] has joined ##stm32 2019-10-04T04:44:13 < turnip420> Can you buy 18650 holders that put the cells in parallel 2019-10-04T05:07:57 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has quit [Ping timeout: 240 seconds] 2019-10-04T05:40:24 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 252 seconds] 2019-10-04T05:40:53 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-04T05:41:50 < Cracki> turnip420, sure. I got 4-slot ones where each cell has its own two legs. you can wire them up however you like. 2019-10-04T05:44:37 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-04T06:07:49 -!- turnip420 [~machinehu@184.71.172.142] has quit [Ping timeout: 268 seconds] 2019-10-04T06:17:14 -!- turnip420 [~machinehu@184.71.172.142] has joined ##stm32 2019-10-04T06:21:15 < dongs> https://tech.slashdot.org/story/19/10/03/1557215/stack-exchange-removes-moderator-for-preferred-pronouns-policy haha 2019-10-04T06:26:35 -!- irf21k [~irf21k@106.193.41.34] has joined ##stm32 2019-10-04T06:33:16 < Cracki> political purges 2019-10-04T06:33:22 < Cracki> so much fun, until the witches burn 2019-10-04T06:35:11 < Cracki> >our CoC is not up for debate 2019-10-04T06:35:25 < Cracki> fuck these commissars with a chainsaw 2019-10-04T06:38:10 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-04T06:39:57 < dongs> https://www.dailystar.co.uk/news/world-news/iphone-user-sues-apple-after-20392402 2019-10-04T06:41:54 < Cracki> white iphone means he's a bottom 2019-10-04T06:41:54 -!- fc5dc9d4 [~quassel@p5B08144E.dip0.t-ipconnect.de] has joined ##stm32 2019-10-04T06:42:16 < Cracki> pink iphone means he'll do your laundry too 2019-10-04T06:42:59 < Cracki> >GayCoins 2019-10-04T06:45:43 -!- fc5dc9d4_ [~quassel@p5B081376.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-04T06:49:50 < Cracki> shit like stackoverflow always happens when the ones adding value have no say and those who have a say add no value 2019-10-04T06:50:38 < dongs> blind leading teh blind? 2019-10-04T06:54:54 < Cracki> no, theft 2019-10-04T06:55:15 < Cracki> the SO owners have a cash cow and they're milking it right there 2019-10-04T06:55:27 < Cracki> and the cash cow gets raepd 2019-10-04T06:56:15 < Cracki> the total worth of SO is volunteer contributions. basically, responding on SO is equivalent to throwing your money/time out the window 2019-10-04T06:56:32 < Cracki> they control your account, your reputation/standing, everything 2019-10-04T06:56:49 < Cracki> you can't take it with you and go elsewhere, you can't fight them 2019-10-04T07:01:57 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-04T07:02:19 < Cracki> >But we hope all moderators know that we very much value and appreciate their contributions 2019-10-04T07:02:42 < Cracki> meaning "so long, suckers!" 2019-10-04T07:03:30 < Cracki> when people start thanking you, you KNOW you aren't getting what you're worth 2019-10-04T07:09:53 -!- turnip420 [~machinehu@184.71.172.142] has quit [Ping timeout: 265 seconds] 2019-10-04T07:19:16 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-04T07:22:26 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-04T07:25:32 < dongs> oh right 2019-10-04T07:25:36 < dongs> the chinese clone usb arrived 2019-10-04T07:25:41 < dongs> lets see how it works inside standard footprint 2019-10-04T07:29:29 < dongs> damn nice 2019-10-04T07:29:34 < dongs> it fits into the FCI footprint 2019-10-04T07:31:25 < Cracki> tell me more about FCI footprints. is that a company or some company's series or a standard by some standards body? 2019-10-04T07:32:07 < dongs> ops, i left altidumb running while updating nvidia driver 2019-10-04T07:32:55 < Cracki> >FCI Asia Pte Ltd 2019-10-04T07:33:04 < dongs> https://github.com/karlp/zypsnips/blob/master/parts.usb.txt 2019-10-04T07:33:06 < Cracki> so it's like jst, got it 2019-10-04T07:33:32 < Cracki> thx 2019-10-04T07:34:22 < dongs> however i think that needs to be updated 2019-10-04T07:34:25 < dongs> i think that fci is the old one 2019-10-04T07:34:28 < dongs> with all pins parallel 2019-10-04T07:34:50 < dongs> yeah 2019-10-04T07:34:53 < dongs> karlp: the new one is 10118194-0001LF 2019-10-04T07:35:22 < dongs> and chinaclone version is U254-051T-4BH23-S2B 2019-10-04T07:35:28 < dongs> on lcsc as https://lcsc.com/product-detail/USB-Connectors_XKB-Enterprise-U254-051T-4BH23-S2B_C319164.html 2019-10-04T07:36:21 < dongs> https://lcsc.com/product-detail/totally_not_scam_C10418.html and this is same, compatible footprint 2019-10-04T07:36:25 < Cracki> two micro B, with and without flange, but apparently different footprints https://www.amphenol-icc.com/micro-usb-101041100001lf.html https://www.amphenol-icc.com/micro-usb-101181920001lf.html 2019-10-04T07:36:26 < R2COM> yo 2019-10-04T07:36:31 < R2COM> fuck im so tired 2019-10-04T07:36:55 < dongs> cracki those are SMT 2019-10-04T07:37:13 < dongs> no throhg-hole pegs for stability 2019-10-04T07:37:43 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-04T07:37:50 < Cracki> right. just saying, I doubt there's "one" fci footprint. maybe for micro b with TH legs 2019-10-04T07:38:34 < dongs> there's only one fci footprint when its being discussed in this channel 2019-10-04T07:38:41 < Cracki> ic 2019-10-04T07:41:29 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-04T07:41:29 -!- day__ is now known as day 2019-10-04T07:42:32 -!- irf21k [~irf21k@106.193.41.34] has quit [Remote host closed the connection] 2019-10-04T08:20:29 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-ooudrtnnczumfwhy] has joined ##stm32 2019-10-04T08:27:19 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-04T08:45:06 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-04T09:16:16 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-04T09:27:01 -!- kow_ [~iccy@135.0.26.39] has quit [Read error: Connection reset by peer] 2019-10-04T09:35:51 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 265 seconds] 2019-10-04T09:36:02 -!- learningc [~learningc@2001:e68:5869:c400:acfb:af7c:b861:dfda] has joined ##stm32 2019-10-04T09:36:32 < Thorn> they bow have a wiki about mp1 https://wiki.st.com/stm32mpu/wiki/Getting_started/STM32MP1_boards 2019-10-04T09:45:40 -!- learningc [~learningc@2001:e68:5869:c400:acfb:af7c:b861:dfda] has quit [Ping timeout: 264 seconds] 2019-10-04T10:07:01 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-04T10:10:20 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-04T10:19:39 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-04T10:33:30 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-04T10:34:22 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has joined ##stm32 2019-10-04T10:44:17 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-04T10:45:12 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-04T11:05:23 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-04T11:34:19 -!- fc5dc9d4 [~quassel@p5B08144E.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-04T11:34:41 -!- fc5dc9d4 [~quassel@p5B08144E.dip0.t-ipconnect.de] has joined ##stm32 2019-10-04T11:36:53 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-04T11:36:57 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Remote host closed the connection] 2019-10-04T11:39:57 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 240 seconds] 2019-10-04T11:42:23 -!- onio [~onio@2a00:23c5:7a01:8600:7866:d9d8:a253:58fe] has joined ##stm32 2019-10-04T11:58:32 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-04T12:09:05 < zyp> shit takes forever, and then you get like one or two sentences 2019-10-04T12:09:22 < zyp> and you think «fuck man, just how slow are you typing?» 2019-10-04T12:30:22 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-04T12:39:32 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-04T12:39:33 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-04T12:51:35 -!- sterna [~Adium@2a02:aa1:101c:cb72:a138:b2c1:4c8d:d541] has joined ##stm32 2019-10-04T12:57:17 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-04T13:16:12 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Remote host closed the connection] 2019-10-04T13:19:46 < Steffanx> Do you Haohmarufi a sentence after you wrote it, Haohmaru ? 2019-10-04T13:19:58 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-04T13:34:56 -!- sterna [~Adium@2a02:aa1:101c:cb72:a138:b2c1:4c8d:d541] has quit [Quit: Leaving.] 2019-10-04T13:50:34 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Ping timeout: 265 seconds] 2019-10-04T13:51:06 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-04T14:43:50 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has joined ##stm32 2019-10-04T14:46:22 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-04T14:49:41 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-04T15:13:17 -!- catphish [~catphish@unaffiliated/catphish] has quit [Remote host closed the connection] 2019-10-04T15:13:46 < Jak_o_Shadows> Remember how I was talking about the usb-i2c bridge? I might be able to do it with the Cypress FX2 - which is the hardware in the knock-off SALAE 8 channel logic analyzers 2019-10-04T15:15:35 < Jak_o_Shadows> Probably not neccessarily an easier path, but it is kinda appealing in some ways 2019-10-04T15:18:49 < PaulFertser> Jak_o_Shadows: related: https://www.karosium.com/2016/08/smbusb-hacking-smart-batteries.html 2019-10-04T15:19:31 < PaulFertser> I used that project with salea clone and talked to my laptop battery just fine 2019-10-04T15:25:13 < Jak_o_Shadows> That is defintely related! Thanks 2019-10-04T15:38:53 < dongs> yeah nothing wrong wiht fx2 other than having to write 8051 shit 2019-10-04T15:41:51 < kakimir32> is there good tools available for writing shit for fx2? 2019-10-04T15:41:59 < kakimir32> slash development kit 2019-10-04T15:42:01 < dongs> keil 2019-10-04T15:42:22 < dongs> oh you can negro it up with sdcc 2019-10-04T15:42:22 < kakimir32> keil does everything? 2019-10-04T15:42:32 < dongs> keil started as 8051 compiler 2019-10-04T15:42:33 < dongs> so yes 2019-10-04T15:42:41 < dongs> and all chinese fx2 firmware is amde for keil 2019-10-04T15:42:52 < kakimir32> is keil eclipse+ 2019-10-04T15:42:56 < kakimir32> ? 2019-10-04T15:42:57 < dongs> no 2019-10-04T15:43:03 < dongs> keil is amazing closed sores for windows 2019-10-04T15:45:29 < kakimir32> is it lightweight? 2019-10-04T15:45:36 < kakimir32> aka. responsive 2019-10-04T15:47:40 < dongs> well if your reference is "eclipse" then yes, absolutely 2019-10-04T15:50:43 < Jak_o_Shadows> god I hate eclipse 2019-10-04T15:52:17 < Jak_o_Shadows> bugger knows why we semi-standardised on it at work 2019-10-04T16:03:49 < Steffanx> What's the alternative for a Jak_o_Shadows ? 2019-10-04T16:04:10 < Jak_o_Shadows> One guy uses visual studio code, that seems nice. 2019-10-04T16:04:36 < Jak_o_Shadows> My main beef with eclipse, apart from working poorly in a vertical monitor, is that the debugging window absolutely spews open files 2019-10-04T16:04:51 < Steffanx> Yeah, except that many extensions are half arsed. 2019-10-04T16:05:05 < Steffanx> And the json config stuff you often need, even for basic stuff 2019-10-04T16:05:37 < Steffanx> > GOOGLE: How do I do x y z in visual studio code 2019-10-04T16:06:51 < Jak_o_Shadows> I'm not the kind of person to go too crazy with learning all the shortcuts for an editor. So that doesn't bother me. 2019-10-04T16:44:17 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-04T16:53:31 -!- kow_ [~iccy@135.0.26.39] has joined ##stm32 2019-10-04T17:08:38 < roomba> windows ME 2019-10-04T17:24:58 < Thorn> received ugreen type c / type c cables (1m long), the CC line is 0.33Ω (I used delta mode to subtract multimeter leads) 2019-10-04T17:25:12 < Thorn> what kind of wire gauge is that 2019-10-04T17:25:32 < Thorn> vbus/gnd are <0.03Ω 2019-10-04T17:25:58 < Steffanx> Whaaaat fullhd for a Haohmaru ? 2019-10-04T17:31:29 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-04T17:32:28 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-04T17:33:11 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-04T17:37:13 < zyp> Thorn, according to my calculcator, 0.33 ohm at 1 meter is 0.05 mm2 of copper 2019-10-04T17:38:02 < Thorn> their A->C cable has a 56k resistor btw 2019-10-04T17:38:05 < zyp> about awg30 according to some random table google found 2019-10-04T17:38:32 < Thorn> which is 500/900 mA, lowest current setting 2019-10-04T17:39:23 < zyp> IIRC it's not 500/900, it's «legacy» 2019-10-04T17:39:46 < zyp> which is exactly what a cable with a legacy plug in one end should have 2019-10-04T17:40:31 < Thorn> so it actually means "use bcd1.2"? that would be cool 2019-10-04T17:40:51 < zyp> think of it this way; when the power comes from an A port, the CC line can't signal how much current the port is capable of 2019-10-04T17:40:55 < Thorn> sources that I studied seemed to tell a different story 2019-10-04T17:41:08 < Thorn> yeah that's very true 2019-10-04T17:41:12 < zyp> so yeah, devices would have to resort to BC or similar if they wanna draw more than 500/900 2019-10-04T17:42:19 < Thorn> but iirc what I've been reading type C overrides bcd, hopefully I'm wrong because otherwise there's no way to safely draw > 500mA if a type C device is plugged into a type A port 2019-10-04T17:43:14 < Thorn> need to find it in the actual spec 2019-10-04T17:43:25 < Thorn> it seems more readable than many blog posts out there 2019-10-04T17:43:57 < zyp> well, duh, C would override BCD if set to anything other than legacy 2019-10-04T17:44:54 < zyp> picture a flowchart; first you check the CC resistor; if it signals 3A or 1.5A you use that, otherwise you resort to BC or other legacy protos 2019-10-04T17:45:00 < zyp> if they also fail you resort to 500/900 2019-10-04T17:45:40 < Thorn> that's what I was going to do, it's good to know the standard agrees with me lol 2019-10-04T17:45:44 < zyp> disclaimer: this is only my interpretion, based on what is sane 2019-10-04T17:46:04 < Thorn> but I did read something that said "500/900 and no bcd" 2019-10-04T17:46:18 < Thorn> brb reading type c spec 2019-10-04T17:46:42 < zyp> the way I see it, that can't really make sense, it would make A-C-cables useless for charging 2019-10-04T17:47:26 < Thorn> useless or dangerous if the Chinese decide to put a low value resistor there to make customers' phones charge faster 2019-10-04T17:47:41 < zyp> Table 2-1 in the standard suggests I'm right 2019-10-04T17:47:52 < Thorn> and fry badly designed motherboards in the process (true story afaik) 2019-10-04T17:48:16 < zyp> https://bin.jvnv.net/file/UnsVs.png 2019-10-04T17:54:49 < zyp> «The following describes the behavior when a legacy host adapter that has an Rp to VBUS so as to mimic the behavior of a Source that is connected to a Sink. The value of Rp shall indicate an advertisement of Default USB Power (See Table 4-15), even though the cable itself can carry 3 A. This is because the cable has no knowledge of the capabilities of the power source, and any higher current is negotiated 2019-10-04T17:54:55 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-04T17:54:55 < zyp> via USB BC 1.2 or by proprietary means.» 2019-10-04T17:55:10 < zyp> «Default USB Power» is 56k 2019-10-04T17:56:58 < Thorn> that is cool 2019-10-04T17:57:54 < Thorn> btw does it mean that any non-marked cable (without PD chip) is supposed to support up to 3A? 2019-10-04T17:58:30 < zyp> yes 2019-10-04T17:58:48 < zyp> type-C mandates all sockets support 5A and all cable assemblies support 3A 2019-10-04T17:58:53 < zyp> minimum 2019-10-04T17:59:27 < Thorn> I need to run 3A through my cables and measure voltage drop 2019-10-04T18:01:09 < zyp> I need to get more type-c cables 2019-10-04T18:02:47 < zyp> so far I pretty much only have my phone that's using it 2019-10-04T18:03:16 < zyp> (and a powerbank, and a charger, but they are both involved in charging my phone) 2019-10-04T18:04:03 < Thorn> there's a ugreen charger on aliexpress that seems to support most PD modes (up to 60W), would be interesting for testing PD negotiation 2019-10-04T18:04:37 < zyp> the charger I use for my phone also supports PD, up to 30W or so IIRC 2019-10-04T18:05:31 < zyp> bought that three years ago, still don't have any PD sinks :) 2019-10-04T18:16:29 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Quit: Leaving] 2019-10-04T18:17:56 < carl0s-> I guess there isn't a room/channel for gdb is there? I have a question on debugging in general, although it's with an old PowerPC mpc562, with a BDI2000 background debug mode debugger. I'm just wondering, can I use this to halt the running target, even if I haven't run/loaded the executable via the debugger? I do have a download of the ROM from the device, but, I just want to halt it running as it does normally, and then dump 2019-10-04T18:17:57 < carl0s-> the RAM.. without uploading/runnning an executable via the debugger. Is that even possible? 2019-10-04T18:18:14 < carl0s-> the BDI2000 implements the gdb remote protocol 2019-10-04T18:21:32 < carl0s-> ok so there is a #gdb channel ;) this matrix riot chat thing isn't so good for finding channels. 2019-10-04T18:25:59 < Thorn> carl0s-: it's very target-specific and depends a lot on your gdbserver. for example with arm coresight and proper tool support you can do it 2019-10-04T18:26:09 < Thorn> it's not really about gdb itself 2019-10-04T18:26:43 < carl0s-> ah right. my gdbserver I guess is the BDI2000 box.. it is motorola/freescale BDM to gdb protocol over ethernet 2019-10-04T18:26:54 < carl0s-> I should just get on and have a play I suppose.. 2019-10-04T19:12:36 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-04T19:17:06 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-04T19:21:16 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2019-10-04T19:22:52 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has quit [Remote host closed the connection] 2019-10-04T19:23:35 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-04T19:24:45 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-04T19:30:08 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-04T19:30:53 < Thorn> why no type C to micro B cables (except 3.0 micro B) 2019-10-04T19:37:13 < jpa-> does anyone know if there exists a kind of "spacer pin" or something that could be soldered between two pcbs to keep them together? ideally a pin that would be narrower at the ends 2019-10-04T19:45:38 < jpa-> https://www.digikey.com/product-detail/en/mill-max-manufacturing-corp/3125-2-00-80-00-00-08-0/ED1299-ND/4440733 hmm i guess this is quite close 2019-10-04T19:46:31 < zyp> Thorn, what do you mean why? I'm pretty sure I've got one somewhere 2019-10-04T19:51:02 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-04T19:56:02 < Thorn> ugreen and other aliexpress brands don't seem to have any 2019-10-04T19:56:53 < Thorn> there are only 3.0 noname cables which can't be used to charge phones etc. because of the ugly 3.0 micro B plug 2019-10-04T19:57:23 < Thorn> https://mobile.twitter.com/gamepopper/status/1180037076701601793 2019-10-04T20:01:46 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c80:2731:538c:8191] has joined ##stm32 2019-10-04T20:02:01 < englishman> bc1.2 is optional anyway 2019-10-04T20:02:16 < englishman> plenty of chargers supply 3a without any way to know it 2019-10-04T20:08:20 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-04T20:10:09 < zyp> Thorn, idk, here's first hit when I searched aliexpress: https://www.aliexpress.com/item/32991012275.html 2019-10-04T20:12:57 < Thorn> not in my search results wtf 2019-10-04T20:13:08 < zyp> search better 2019-10-04T20:13:38 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-04T20:20:24 < BrainDamage> search butter 2019-10-04T20:24:49 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-04T20:32:58 < Thorn> an A-C cable that came with my satechi keyboard doesn't even have d+/d-. the cc resistor is still 56kΩ 2019-10-04T20:33:03 < aandrew> hm, STM32F756, which peripheral clock is TIM13 hanigng off of? 2019-10-04T20:34:39 < zyp> check the RM? 2019-10-04T20:35:15 < aandrew> that's what I'm in, can't seem to find it 2019-10-04T20:35:37 < aandrew> it just says "TIMxCLK = 2xPCLKx" 2019-10-04T20:35:40 < aandrew> gotta dig a little deeper 2019-10-04T20:35:50 < zyp> what's the RM number? 2019-10-04T20:35:53 < aandrew> 0385 2019-10-04T20:36:03 < Thorn> aandrew: check the clock page in cube 2019-10-04T20:36:25 < zyp> APB1 2019-10-04T20:36:30 < Thorn> they're usually pretty comprehensive afaict 2019-10-04T20:36:36 < zyp> so PCLK1 2019-10-04T20:36:48 < aandrew> where are you finding that 2019-10-04T20:36:54 < zyp> page 72 2019-10-04T20:37:09 < zyp> (rev8) 2019-10-04T20:37:24 < aandrew> ah yes 2019-10-04T20:37:28 < aandrew> I was looking for a diagram 2019-10-04T20:37:42 < aandrew> yeah 2,3,4,5,6,7,12,13,14 are all APB1 which is PCLK1 2019-10-04T20:37:59 < aandrew> 1,8 are APB2 which is PCLK2 2019-10-04T20:38:09 < aandrew> hm so why is this stupid timer running fast 2019-10-04T20:44:51 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-04T20:45:26 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-04T20:48:05 < Cracki> by how much 2019-10-04T20:48:32 < Cracki> what clock does it have if the timer's registers are to be believed? 2019-10-04T20:50:42 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-04T20:55:23 < aandrew> moment I may have found it 2019-10-04T20:56:01 < aandrew> HAL_IncTick and osSysTickHandler() were being called 2019-10-04T20:56:55 < aandrew> yep that was all it was 2019-10-04T21:03:53 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-04T21:30:11 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-04T21:47:48 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-04T21:49:35 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-04T21:57:10 < Thorn> https://www.aliexpress.com/item/32892378633.html 2019-10-04T22:03:39 -!- ekaologik [~quassel@p5DE86412.dip0.t-ipconnect.de] has joined ##stm32 2019-10-04T22:11:37 < aandrew> tinned wire has reduced electrical conductivity compared to copper? 2019-10-04T22:11:42 < aandrew> sounds dubious 2019-10-04T22:13:19 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has quit [Quit: Leaving] 2019-10-04T22:21:55 < qyx> heh 2019-10-04T22:22:23 < qyx> but I found out most high tempoerature cables have tinned copper conductors 2019-10-04T22:30:16 < kakimir32> fools 2019-10-04T22:30:23 < kakimir32> are you looking for propper usb cables? 2019-10-04T22:31:17 < kakimir32> juizebitz cables from uk 2019-10-04T22:32:04 < BrainDamage> pure copper vs same diameter of copper + tin, the tinned version has less conducibility because some copper is replaced by tin 2019-10-04T22:32:24 < BrainDamage> but same diamater of copper AND tin, the tin one has less resistance 2019-10-04T22:34:33 < Cracki> why would they tin them? I can only imagine that it's to keep the copper from corroding 2019-10-04T22:35:42 < BrainDamage> yes and it makes it easier to solder 2019-10-04T22:36:58 < BrainDamage> tinning pcb traces is also a known practice in power circuits to increase current carrying capability 2019-10-04T22:37:28 < BrainDamage> because you can slather a thick amount of tin paste on top 2019-10-04T22:42:08 < Cracki> soldering, sure. I was puzzled as to why they'd use tinned copper in a usb cable tho 2019-10-04T23:04:58 < carl0s-> Would this work? Mipi dbi type c option 1 display expects 9 bit spi where bit9 is data/command bit. The Linux tiny DRM mipi driver has this comment for spi controllers that won't do 9 bit: https://dpaste.de/rNnA 2019-10-04T23:05:49 < carl0s-> Would you expect the target to ignore all the 0s? If each bit is sent as it's own 8 bit packet padded with zeros? 2019-10-04T23:13:35 -!- tkoskine [tkoskine@kapsi.fi] has quit [Ping timeout: 276 seconds] 2019-10-04T23:14:02 -!- ekaologik [~quassel@p5DE86412.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-04T23:14:19 -!- tkoskine [tkoskine@kapsi.fi] has joined ##stm32 2019-10-04T23:17:50 -!- XTL [xtl@kapsi.fi] has quit [Ping timeout: 244 seconds] 2019-10-04T23:27:58 -!- XTL [xtl@kapsi.fi] has joined ##stm32 2019-10-04T23:36:04 < englishman> kakimirV 2019-10-04T23:36:04 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-04T23:38:04 < roomba> kakimir64 2019-10-04T23:38:16 < kakimir32> I have been summoned 2019-10-04T23:38:36 < englishman> riskimirV 2019-10-04T23:38:41 < kakimir32> yes 2019-10-04T23:38:54 < kakimir32> how is works englishman? 2019-10-04T23:39:04 < englishman> ok 2019-10-04T23:42:28 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-04T23:43:45 < kakimir32> first snow today 2019-10-04T23:44:13 < kakimir32> kinda cool 2019-10-04T23:46:23 < qyx> where and how much 2019-10-04T23:48:20 < karlp> dongs: yeah, had the new usb micro part locally, haven't pushed it yet, was also waiting for you to actually get them :) 2019-10-04T23:53:45 < kakimir32> funland ofc 2019-10-04T23:53:55 < kakimir32> 1cm 2019-10-04T23:59:04 < karlp> aandrew: bit late, but I find it sometimes to easier to find which rcc register teh peripheral enable bit is in, rather than finding it on the diagrams and descriptions. --- Day changed Sat Oct 05 2019 2019-10-05T00:00:18 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-05T00:11:00 < PaulFertser> kakimir32: are you going to ride for the whole winter? 2019-10-05T00:11:19 < kakimir32> no 2019-10-05T00:12:38 < PaulFertser> What's your plan? 2019-10-05T00:15:01 < kakimir32> for my life? 2019-10-05T00:15:09 < kakimir32> riding? 2019-10-05T00:16:51 < kakimir32> move to laurences basement? 2019-10-05T00:22:24 < PaulFertser> kakimir32: for riding. I'm kinda obsessed with riding so all my questions are about riding y'know. 2019-10-05T00:22:54 < kakimir32> if I had changed my bike I would probs do it for lulz 2019-10-05T00:23:20 < kakimir32> bike with fall protection pipe things 2019-10-05T00:23:56 < kakimir32> but I don't feel like putting any money to current one as it's a piece of junk that keeps moving somehow 2019-10-05T00:24:43 < kakimir32> spike tyres and falling protection, heated stuff, windshield and the price of my bike would be doubled 2019-10-05T00:24:52 < kakimir32> for me, but not for resale 2019-10-05T00:25:38 < kakimir32> but I will ride a month or so when there is no ice 2019-10-05T00:27:05 < kakimir32> I remember when I was younger -5C was not a problem 2019-10-05T00:27:36 < PaulFertser> I plan to ride till there can be ice, but not for any long distances. 2019-10-05T00:27:57 < karlp> I'm not even sure I'll switch to studs this winter. 2019-10-05T00:28:14 < karlp> new office is just in town, wayyyyy less chance of ice on that route. 2019-10-05T00:28:27 < karlp> should probably do it becaus I have them though I guess. 2019-10-05T00:28:32 < kakimir32> PaulFertser: do you need to ride in early hours? 2019-10-05T00:29:02 < PaulFertser> kakimir32: no, and yes, I'm aware the probability to find ice on the roads is much higher in the morning. 2019-10-05T00:29:38 < PaulFertser> karlp: regular summer tires are no good in subzero conditions I guess. 2019-10-05T00:29:42 < kakimir32> friend had a bike with spike tyres 2019-10-05T00:30:01 < kakimir32> insane how it went on flat ice 2019-10-05T00:30:32 < kakimir32> it was like on dry road 2019-10-05T00:30:33 < karlp> spike tires are AWESOME on ice. it's unbelievable how much differency they make 2019-10-05T00:31:18 < kakimir32> spikes are illegal on road 2019-10-05T00:31:40 < kakimir32> and only useful when there is 100% ice 2019-10-05T00:32:29 < kakimir32> 100% ice or slippery generally 2019-10-05T00:35:47 < kakimir32> or snowing condition 2019-10-05T00:39:12 < karlp> we're allowed spikes from nov to april on cars, whenever you feel like on your (bicycle) bike 2019-10-05T00:41:38 < kakimir32> studs 2019-10-05T00:41:55 < kakimir32> spikes are like 1cm long, 2cm long 2019-10-05T00:41:57 < kakimir32> or even longer 2019-10-05T00:42:01 < karlp> wat 2019-10-05T00:42:09 < karlp> that sounds insane. 2019-10-05T00:43:55 < englishman> getting to try out lecroy 2019-10-05T00:44:05 < englishman> to see if it's better than the tek 6 2019-10-05T00:46:30 < kakimir32> https://youtu.be/xJpmgYx_X-4?t=202 I would not ride on summertires anything that isn't lightweight and low height 2019-10-05T00:46:46 < kakimir32> on snowing conditions or on ice 2019-10-05T00:59:27 < PaulFertser> kakimir32: heh, it's like he's done that on purpose, rolled the throttle hard 2019-10-05T00:59:59 < PaulFertser> (leading to sudden engine braking) 2019-10-05T01:00:00 < kakimir32> anything for youtubes 2019-10-05T01:02:32 < kakimir32> it's not even fun to ride like that 2019-10-05T01:04:29 < oz4ga> only an idiot rides a motor cycle on snow 2019-10-05T01:04:35 < PaulFertser> I ride my push bike though. 2019-10-05T01:08:54 < kakimir32> it's doable though oz4ga 2019-10-05T01:09:11 < kakimir32> https://www.youtube.com/watch?v=QjV-d1844TU 2019-10-05T01:09:15 < oz4ga> yes but damn precarious 2019-10-05T01:09:33 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-05T01:12:02 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-05T01:14:00 < kakimir32> https://www.youtube.com/watch?v=AaP6gI6NHPI 2019-10-05T01:16:03 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T01:20:59 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T01:23:46 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T01:27:56 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-05T01:30:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-05T01:45:18 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-05T01:50:38 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c80:2731:538c:8191] has quit [Quit: Konversation terminated!] 2019-10-05T01:57:28 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T02:03:37 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 240 seconds] 2019-10-05T02:08:26 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 276 seconds] 2019-10-05T02:11:40 -!- onio [~onio@2a00:23c5:7a01:8600:7866:d9d8:a253:58fe] has quit [Ping timeout: 252 seconds] 2019-10-05T02:16:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T02:32:42 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T02:42:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T02:43:19 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T02:51:39 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T02:52:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T03:00:54 < Thorn> http://www.freedsp.cc/ 2019-10-05T03:10:27 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-05T03:16:52 < machinehum> Hello weiners 2019-10-05T03:18:24 < machinehum> Does this seem like a reasonable way to sample this MIC https://github.com/Machine-Hum/FinalProject/blob/master/birdEar/plot/birdEar.pdf 2019-10-05T03:18:48 < machinehum> This is a private repo. 2019-10-05T03:19:40 < machinehum> https://drive.google.com/file/d/1Ad-SvGXlVou9D8EI10rDuH0gfys4tGsn/view?usp=sharing 2019-10-05T03:21:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T03:22:58 < Laurenceb> hello, Steve Dawson here from Ashford in Kent 2019-10-05T03:24:05 < Laurenceb> https://twitter.com/stevedawson0972/status/895008318544130052?lang=en 2019-10-05T03:24:44 < Laurenceb> https://www.youtube.com/watch?v=3wOyqiThZ40 2019-10-05T03:25:21 < machinehum> Hello Steve 2019-10-05T03:25:56 < englishman> hi, im Laurence blaxter posting poor quality memes from my basement in Leeds. and I'm not a pedo 2019-10-05T03:26:03 < Laurenceb> keeek 2019-10-05T03:26:14 < Laurenceb> englishman confirmed as Ben Garrison fan 2019-10-05T03:27:03 < Laurenceb> he always draws trolls as living in a basement in Leeds 2019-10-05T03:27:17 < Thorn> musics https://www.youtube.com/watch?v=t7Gz8jJOLAk 2019-10-05T03:27:32 < mawk> I ported the wireguard vpn to android 10 2019-10-05T03:27:38 < mawk> I'm a celebrity of open source now 2019-10-05T03:27:51 < emeb_mac> we're not worthy 2019-10-05T03:30:02 < dongs> machinehum: what the fuck. just buy a fucking PDM mic breakout from china 2019-10-05T03:30:09 < dongs> and you can skip all teh analog cancer 2019-10-05T03:32:40 < specing> > analog cancer 2019-10-05T03:36:18 < machinehum> pulse density modulation 2019-10-05T03:36:24 < machinehum> Go fuck yourself 2019-10-05T03:37:06 < roomba> go listen to somafm suburbs of goa 2019-10-05T03:42:13 < machinehum> on it 2019-10-05T03:49:29 < machinehum> Pretty good not going to lie 2019-10-05T03:53:58 < dongs> machinehum: or I2S or any other fucking digital mic that doesnt require aids 2019-10-05T03:56:54 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T03:58:23 < machinehum> Got aids though? 2019-10-05T03:58:51 < machinehum> Other mic's not compatable with aids people, or only non-aids people 2019-10-05T03:59:25 < machinehum> Using it with dnn, need to be able to control sample rate and bit depth 2019-10-05T03:59:48 < machinehum> If you know of a non-aids mic that will do that please tell 2019-10-05T04:00:21 < machinehum> Or else it's back to Suburbs of GOA 2019-10-05T04:04:52 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-05T04:16:04 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2019-10-05T04:20:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T04:33:32 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-05T04:49:09 < karlp> "dnn" ? 2019-10-05T05:10:18 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Read error: Connection reset by peer] 2019-10-05T05:27:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-05T05:35:57 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-05T05:40:03 < aandrew> karlp: that's not a bad idea, squirrelling that away for next time 2019-10-05T05:47:28 -!- qyx_ [~qyx@84.245.121.45] has joined ##stm32 2019-10-05T05:47:49 < aandrew> machinehum: I don't like the use of 1M resistors everywhere, you'll have lots of noise pickup 2019-10-05T05:48:24 < aandrew> but it looks reasonable otherwise, keep in mind that your gain is 1+rf/ri since it's a non-inverting opamp 2019-10-05T05:49:17 -!- qyx [~qyx@gw2.krtko.org] has quit [Ping timeout: 240 seconds] 2019-10-05T05:55:19 < aandrew> also generally you should tie - and out together and ground the + through 1k or something on unused opamps to keep them from oscillating 2019-10-05T05:55:55 < aandrew> your loudness comparator has no hysteresis, might want to play with that a little 2019-10-05T05:56:13 < aandrew> generally speaking I have input resistors on all my opamp inputs so I can tweak shit without cutting traces 2019-10-05T05:56:51 < aandrew> I also don't see any antialias filter on the mic output, you could incorporate that into the amplifier pretty easily 2019-10-05T05:57:35 < aandrew> I strongly suggest using a standard 2x5 0.050" header for the debug connector with standard pinout 2019-10-05T05:58:22 < aandrew> PB1/2 should have pullups and ideally some debounce but you can do it with internal pullup and rely completely on digital debounce 2019-10-05T05:58:25 < aandrew> I'm just oldschool 2019-10-05T05:59:08 < aandrew> clock and D0-D3 should have 33 ohm series resistors 2019-10-05T05:59:30 < aandrew> I'd also put a pullup on the detect 2019-10-05T05:59:55 < aandrew> you have the room, connect SWO to the debug connector as well 2019-10-05T06:00:10 < aandrew> if you connect the "loud" to PA0 you can wake up from deep sleep rather than just EXTI 2019-10-05T06:00:38 < aandrew> machinehum: any particular reason you're using such a high pin count STM32 for such a low pin count task? 2019-10-05T06:01:14 < aandrew> I'd use 100nF not 1u for all your bypass, add a 1u/10u for bulk only 2019-10-05T06:01:26 < aandrew> also nRST should have a pullup (and ideally a 100nF to ground as well) 2019-10-05T06:01:33 < aandrew> boot0 should not be floating 2019-10-05T06:01:52 < aandrew> that 2.2uF on VCAP1/2 isn't right, each pin needs a cap to ground 2019-10-05T06:02:31 < aandrew> UART_RX should have a pullup, also suggest using a 1x6 header with standard pinning so you can easily connect serial without screwing around 2019-10-05T06:03:12 < aandrew> since your'e actually using the ADC you might want to use a BLM21 and separate bypass for the VDDA and keep your analog and digital relatively away from each other 2019-10-05T06:03:34 < aandrew> check the docs, SWDIO and SWDCK need pullup/down (can't remember which needs which) 2019-10-05T06:04:41 < aandrew> no charging circuit? no battery monitor so the stm32 can do something when the voltage reaches low threshold? it can be as simple as a voltage divider going into an analog input 2019-10-05T06:32:53 < dongs> aandrew: who teh hell are you talking to>? 2019-10-05T06:32:58 < dongs> all i see is your monologue 2019-10-05T06:33:04 < dongs> oh, reviewing his dumb schematic 2019-10-05T06:33:09 < dongs> that i said to throw in the trash and use a PDM mic 2019-10-05T06:33:23 < dongs> why is he even MAKING a schematic, STM32F7 evalboard has a pair of PDM mics on it already 2019-10-05T06:33:29 < dongs> the -discovery shit 2019-10-05T06:39:23 < aandrew> yeah a digital mic would probably be cheaper and higher quality in the long run 2019-10-05T06:40:43 -!- fc5dc9d4_ [~quassel@p57A32E60.dip0.t-ipconnect.de] has joined ##stm32 2019-10-05T06:44:27 -!- fc5dc9d4 [~quassel@p5B08144E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-05T06:49:32 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-05T07:11:25 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-05T07:14:24 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-05T07:14:52 < carl0s-> A 2019-10-05T07:17:59 < R2COM> yo 2019-10-05T07:37:16 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-05T07:40:03 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-05T07:40:03 -!- day__ is now known as day 2019-10-05T08:02:07 < R2COM> what is the best earset for phone which can stream from 2 devices (switch between them) 2019-10-05T08:30:27 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [Ping timeout: 240 seconds] 2019-10-05T08:31:05 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-05T08:31:48 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-05T08:44:56 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-05T09:20:02 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 276 seconds] 2019-10-05T09:39:41 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-05T09:56:56 < qyx_> that electret mic with a 2k2 bias from digital 3v3 vdd is not going to be usable 2019-10-05T09:57:07 < qyx_> also, use pdm mic 2019-10-05T09:57:19 < qyx_> machinehum: ^ 2019-10-05T09:59:37 < qyx_> the rest what aandrew said is all legit 2019-10-05T10:12:38 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-05T10:16:33 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has joined ##stm32 2019-10-05T10:18:25 < dongs> R2COM: pretty much any non-garbage BT headphone will be able to connect to several sources? 2019-10-05T10:18:42 < dongs> i have my QC35s paired to my laptop, phone, and cars 2019-10-05T10:22:35 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Ping timeout: 265 seconds] 2019-10-05T10:26:54 -!- sterna [~Adium@2a02:aa1:1013:10f6:75a0:6564:e06:b051] has joined ##stm32 2019-10-05T10:31:06 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has quit [Ping timeout: 246 seconds] 2019-10-05T10:34:04 < Cracki> how does that work, does it *connect* to multiple sources and mix their audio in the headset? 2019-10-05T10:34:08 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-05T10:34:24 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-05T10:34:55 < jpa-> the ones i have can just pair with multiple sources, but only one can be connected at a time 2019-10-05T10:39:41 < Cracki> https://www.howtogeek.com/297281/how-to-connect-your-bluetooth-headphones-to-more-than-one-device-at-a-time/ 2019-10-05T10:40:01 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T10:41:57 < Cracki> they even have "advanced" multipoint, but it all still sounds as if they take a single active source at once https://www.jabra.com/blog/multipoint-explained-one-headset-for-all-devices/ 2019-10-05T10:43:17 < Cracki> still, it's a step up from having to pair every time you wanna switch 2019-10-05T10:50:08 -!- qyx [~qyx@84.245.120.8] has joined ##stm32 2019-10-05T10:53:38 -!- qyx_ [~qyx@84.245.121.45] has quit [Ping timeout: 276 seconds] 2019-10-05T10:56:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-05T11:13:41 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-05T11:20:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T11:28:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T11:31:30 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T11:51:01 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has joined ##stm32 2019-10-05T11:51:33 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 245 seconds] 2019-10-05T11:57:13 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T11:58:03 -!- sterna1 [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-05T12:00:00 < jpa-> i still wish there was bluetooth headsets that could just walkie-talkie to each other 2019-10-05T12:00:35 -!- sterna [~Adium@2a02:aa1:1013:10f6:75a0:6564:e06:b051] has quit [Ping timeout: 276 seconds] 2019-10-05T12:06:26 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T12:08:50 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T12:15:25 < Steffanx> Apple Earbutts is all you need right? 2019-10-05T12:24:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 264 seconds] 2019-10-05T12:33:22 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T12:37:12 < zyp> Apple Earcigs 2019-10-05T12:42:49 < dongs> Cracki: it connects to them all, and whatever plays takes priority. 2019-10-05T12:43:05 < dongs> for example if im listning to muzask from laptop and phone gets notification, that goes through 2019-10-05T12:43:17 < dongs> temporarily muting laptop 2019-10-05T12:43:37 < zyp> you have your headset paired to your car? 2019-10-05T12:44:49 < zyp> how does that work? so you can listen to radio and shit even when your passengers don't want to? 2019-10-05T12:50:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T12:51:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T12:56:13 < Steffanx> That why he can listen to Louis rossman without ruining his kids 2019-10-05T12:56:22 < Steffanx> Way* 2019-10-05T13:03:45 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-05T13:08:47 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-05T13:15:09 -!- ekaologik [~quassel@p5DE86C60.dip0.t-ipconnect.de] has joined ##stm32 2019-10-05T13:23:47 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has quit [Ping timeout: 276 seconds] 2019-10-05T13:27:41 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T13:28:03 < Cracki> some cars have phone 2019-10-05T13:28:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T13:28:39 < Cracki> not sute why one wouldn't use the car's stereo for calls then 2019-10-05T13:29:41 < Cracki> hm wait, might be the other way around. what I saw could have been car paired to smartphone, and car is "keyboard" 2019-10-05T13:35:23 < Steffanx> mine works like that. 2019-10-05T13:57:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-05T14:00:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T14:17:38 < zyp> most do 2019-10-05T14:27:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T14:29:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T14:55:11 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 265 seconds] 2019-10-05T15:05:36 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T15:06:29 < karlp> aandrew: what's "standard pinning" for 1x6 ttl serial? I've seen quite a few variants :) 2019-10-05T15:07:02 < karlp> swdio and swdclk don't need shit pulls either :) 2019-10-05T15:08:52 < zyp> karlp, didn't «everybody» copy this? https://www.ftdichip.com/Support/Documents/DataSheets/Cables/DS_TTL-232R_CABLES.pdf 2019-10-05T15:09:06 < karlp> hah no 2019-10-05T15:09:20 < zyp> karlp, speaking of, I meant to ask you if there were a «standard» pinout for RS485 on RJ45 2019-10-05T15:09:31 < zyp> but when I googled it I found a few variants 2019-10-05T15:09:34 < karlp> I've seen variants there too 2019-10-05T15:09:41 < zyp> yeah 2019-10-05T15:09:43 < karlp> we don't use it, rj45 isd too big. 2019-10-05T15:09:53 < karlp> plug and play is nice, but whatever. 2019-10-05T15:09:54 < zyp> so I just ordered a couple of usb to rj45 cables and are gonna go with that 2019-10-05T15:10:00 < karlp> china custom makes cables :) 2019-10-05T15:11:11 < zyp> I don't need RJ45 on the equipment side, but I wanna run it through a patch panel 2019-10-05T15:11:56 < zyp> well, not really patch panel, but I'm gonna run a cable from the outlet box in top here to the ventilation system: https://bin.jvnv.net/file/OV2am.jpg 2019-10-05T15:12:38 < zyp> then stuff a rpi in there with a usb to rs485-rj45 cable 2019-10-05T15:18:23 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has joined ##stm32 2019-10-05T15:22:20 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-05T15:22:29 < kakimir32> yes hello 2019-10-05T15:32:29 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T15:35:15 < kakimir32> I'm on desktop now 2019-10-05T15:35:27 < kakimir32> time to fire up some projects 2019-10-05T15:35:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T15:43:55 < kakimir32> where was the real openocd git? 2019-10-05T15:46:07 -!- sterna1 [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-05T16:02:57 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-05T16:05:03 < kakimir32> best text editor for linux? 2019-10-05T16:18:33 < kakimir32> no emacs vim battle? 2019-10-05T16:24:48 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-05T16:25:48 < benishor> no battle whatsoever, everybody agrees vim is the best 2019-10-05T16:29:23 < specing> both are the best 2019-10-05T16:29:30 < specing> nano sucks 2019-10-05T16:29:48 < Steffanx> Is there an editor written in Ada? 2019-10-05T16:32:06 < specing> no idea 2019-10-05T16:33:30 < catphish> nano is best 2019-10-05T16:46:05 -!- renn0xtk9 [~max@2a02:810d:1540:2448:8115:1cfd:2335:1121] has quit [Quit: Konversation terminated!] 2019-10-05T16:50:09 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 246 seconds] 2019-10-05T16:55:03 -!- ekaologik [~quassel@p5DE86C60.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-05T16:55:25 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-05T16:56:34 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T17:03:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 245 seconds] 2019-10-05T17:04:28 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-05T17:05:36 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T17:08:11 < Cracki> I'm with nano if it has to run in a terminal. not weird, explains itself. 2019-10-05T17:08:53 < Cracki> sublime text is reasonably tolerable :P 2019-10-05T17:10:53 < PaulFertser> nano is the most weird editor. It's non-modal but doesn't support the most common Readline/Emacs non-modal keybindings without any good reason. It's just silly. 2019-10-05T17:14:14 < roomba> i just use EDIT.COM in dosbox 2019-10-05T17:14:32 < roomba> nano is a poser 2019-10-05T17:17:47 -!- kow_ [~iccy@135.0.26.39] has quit [Read error: Connection reset by peer] 2019-10-05T17:19:12 < aandrew> karlp: sorry I meant the ftdi "smart cables", seems pretty standard 2019-10-05T17:20:11 < aandrew> Cracki: most bluetooth headsets I've used can connect to multiple devices. it's one at a time, sometimes stupidly so 2019-10-05T17:20:27 < aandrew> qyx: how would you bias an electret? 2019-10-05T17:22:14 < aandrew> karlp: I tend to not trust internal pulls, but you're right 2019-10-05T17:26:32 < kakimir32> what is ARMv7-M "Code" area? 2019-10-05T17:26:37 < kakimir32> context: SRAM 2019-10-05T17:27:13 < kakimir32> is it just an area that can be executed by cpu? 2019-10-05T17:35:08 < Cracki> core coupled memory maybe, faster than 'regular' sram 2019-10-05T17:35:49 < Cracki> I'm probably wrong. I did read about code area somewhere 2019-10-05T17:36:14 < Cracki> where do you read this? 2019-10-05T17:37:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T17:37:20 < Cracki> ah, openocd... 2019-10-05T17:37:52 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T17:56:52 < Cracki> some assembly listings equate it to the start of .text 2019-10-05T17:57:13 < Cracki> the constant for lpcxxx was 0x10000000 iirc 2019-10-05T17:59:18 < Cracki> some rust code labels that constant as core coupled ram for stack, 0x20000000 as just ram 2019-10-05T18:07:11 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T18:16:38 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-05T18:22:31 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T18:32:40 -!- qyx [~qyx@84.245.120.8] has quit [Ping timeout: 268 seconds] 2019-10-05T18:39:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-05T18:44:05 -!- ekaologik [~quassel@p5DE86C60.dip0.t-ipconnect.de] has joined ##stm32 2019-10-05T18:45:06 < catphish> Cracki: sublime is my main editor for work 2019-10-05T18:45:33 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T18:45:44 -!- kow_ [~iccy@135.0.26.39] has joined ##stm32 2019-10-05T18:48:24 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-05T19:10:26 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-05T19:19:47 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-05T19:25:09 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-05T19:52:36 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-05T19:54:57 < srk> https://gist.github.com/sorki/c4ae0314a084cec54498ffc843972176 2019-10-05T19:55:18 < srk> such diffs 2019-10-05T19:59:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-05T20:00:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T20:01:59 -!- ekaologik [~quassel@p5DE86C60.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-05T20:34:03 < R2COM> alright just hooked the unknown 3 phase motor with hall sensors to my crafted shit with F0 and it seems to work 2019-10-05T20:34:11 < R2COM> it spins 2019-10-05T20:34:33 < R2COM> but signal from those hall sensors is very low.. not sure if its cuz i powered it from 5V 2019-10-05T20:34:37 < R2COM> maybe it needs 12V 2019-10-05T20:34:42 < R2COM> but afraid to burn it 2019-10-05T20:34:47 < R2COM> dont have DS for that motor 2019-10-05T20:34:51 < R2COM> forgot where it came from 2019-10-05T20:43:33 < zyp> dickspin.gif 2019-10-05T20:45:30 < R2COM> ideally one supposed to measure the 3 wires toggling from hall sensor, but for now im tricking it by providing that toggle to inputs of F0 (as if they were to come from motor) 2019-10-05T20:45:56 < R2COM> and i found that around 100hz commutation rate makes the motor spin properly 2019-10-05T20:47:06 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-05T20:56:04 < Steffanx> Damn, zyp. That was so unzyp. 2019-10-05T21:07:59 < R2COM> now if course with know BLDC motor ill use its sensors for commutation, and write probnably some PID loop for speed control 2019-10-05T21:08:05 < R2COM> known* 2019-10-05T21:08:25 < jpa-> R2COM: some times hall sensors need external pull-ups, do you have that? 2019-10-05T21:11:00 < R2COM> https://www.youtube.com/watch?v=60-COGwe-Hk 2019-10-05T21:11:03 < R2COM> hmmm 2019-10-05T21:11:07 < R2COM> actually no... 2019-10-05T21:11:19 < R2COM> oh that might be true 2019-10-05T21:11:25 < R2COM> ill try pulling it up to 5v 2019-10-05T21:20:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T21:25:35 < R2COM> yeah...i pulled it to 5V now hall sensors give proper values out 2019-10-05T21:25:40 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T21:34:07 < catphish> i know i asked this last week, but figured i'd try again, does anyone have any idea how to use he double-buffered USB bulk feature? 2019-10-05T21:35:53 -!- renn0xtk9 [~max@2a02:810d:1540:2448:15ce:6cc1:23ca:afa2] has joined ##stm32 2019-10-05T21:37:31 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-05T21:38:24 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 252 seconds] 2019-10-05T21:39:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T21:54:30 < Thorn> catphish: it's on my todo list 2019-10-05T21:54:38 < Thorn> for 2+ years now lol 2019-10-05T22:00:58 < catphish> i'm going to make a simple test case and see if i can get it right 2019-10-05T22:02:29 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-05T22:05:53 < jpa-> R2COM: are you going for simple 6-step control or do you have current sensing for FOC / DTC? 2019-10-05T22:06:24 < R2COM> jpa- what is foc/dtc? 2019-10-05T22:06:33 < R2COM> oh wait 2019-10-05T22:06:34 < R2COM> no 2019-10-05T22:06:43 < R2COM> if it has hall sensor, you dont need current sensing right 2019-10-05T22:06:52 < jpa-> field oriented control / direct torque control; for having more than 60° steps in the control waveform 2019-10-05T22:06:54 < R2COM> cuz difference between hall sensor toggles can be related to speed 2019-10-05T22:07:17 < jpa-> yeah, with hall sensors you can drive the motor fets even without MCU 2019-10-05T22:07:42 < R2COM> those RC BLDCs powerfull ones, which come without those sensors, they need that control you mentioned 2019-10-05T22:07:46 < R2COM> you talking about measuring its BEMF? 2019-10-05T22:07:53 < jpa-> but FOC/DTC gives a bit better efficiency and a bit higher torque, because with 60° control the magnetic field can be up to 30° misaligned 2019-10-05T22:08:37 < jpa-> no, BEMF is voltage measurement to detect motor position without hall sensors; FOC/DTC uses current sensing on the motor phases to more accurately control the magnetic field direction 2019-10-05T22:08:57 < R2COM> so FOC/DTC is much better than BEMF? 2019-10-05T22:09:18 < jpa-> more like complementary - you need to know the motor orientation anyway, you can do it either with hall sensors or with BEMF 2019-10-05T22:09:55 < R2COM> so... if one is doing BEMF measurement, does he still need FOC/DTC? 2019-10-05T22:10:25 < jpa-> FOC/DTC helps equally with BEMF, though it is more difficult to do accurately with BEMF than it is with hall sensors 2019-10-05T22:11:03 < R2COM> from what i learned from multiple docs: BEMF good for very high rate rotational applications 2019-10-05T22:11:04 < jpa-> but FOC/DTC is alternative to simple 6-step control - whereas with 6-step control you just pick the FET states based on motor orientation, while FOC/DTC picks that based on motor orientation + motor phase current - to automatically keep the magnetic field in optimal orientation for torque 2019-10-05T22:11:08 < R2COM> and doesnt work well with low speed ones 2019-10-05T22:11:16 < R2COM> but Hall is working well for low rotation speed 2019-10-05T22:11:41 < jpa-> IMO hall sensors are superior for both high and low speed, but BEMF is only good for high speed 2019-10-05T22:11:54 < R2COM> my application will not be RC (im not going to spin with crazy speed, its actually gonna be like 350-500 RPM max rotation...soo...dont think that BEMF will work for me) 2019-10-05T22:12:07 < R2COM> yes thats what i learned 2019-10-05T22:12:30 < R2COM> but if you noticed all those RC heli/aircraft motors dont come with hall sensors 2019-10-05T22:12:31 < Cracki> consider using a magnetic encoder if you want it very precise 2019-10-05T22:12:31 < R2COM> not sure why 2019-10-05T22:12:41 < jpa-> because it's cheaper without, and works well enough 2019-10-05T22:12:51 < R2COM> i guess yes 2019-10-05T22:13:00 < R2COM> and noone spins them with like 200rpm 2019-10-05T22:13:01 < Cracki> the point of bemf, hall sensors, encoders is to know where the rotor is 2019-10-05T22:13:28 < jpa-> Cracki: yeah, that summarises that well 2019-10-05T22:13:28 < Cracki> the point of FOC/DTC is similar to microstepping, you apply a field that changes smoothly 2019-10-05T22:14:15 < jpa-> FOC/DTC does that (sinusoidal control), but it does more: it monitors the magnetic field orientation, instead of just approximating like 6-step control does 2019-10-05T22:14:27 < Cracki> if you wanna look at fancy things, look up VESC (by some guy named vedder) 2019-10-05T22:14:34 < R2COM> so...if for this relatively simple application all i need is to keep it constant 200-500RPM i guess just HS is enough for me 2019-10-05T22:14:55 < jpa-> at different speeds the magnetic field will change at different delay due to coil inductance 2019-10-05T22:14:57 < R2COM> so... FOC/DTC i assume is especially useful in something like...motor cars? 2019-10-05T22:15:03 < Cracki> those RC bldcs have hall sensors, but multiple poles, so one 60 degree *electrical* step is still very small mechanically 2019-10-05T22:15:12 < jpa-> R2COM: yeah, the higher power the more efficiency matters 2019-10-05T22:15:13 < R2COM> cuz thats where gaining extra efficiency is important 2019-10-05T22:15:18 < Cracki> foc/dtc is useful when you don't want the motor to vibrate or whine 2019-10-05T22:15:19 < BrainDamage> you get extra torque too 2019-10-05T22:15:23 < BrainDamage> and less torque ripple 2019-10-05T22:15:30 < R2COM> also it might be used in electric superchargers too? 2019-10-05T22:15:34 < jpa-> R2COM: though also if you need smooth and silent rotation, FOC/DTC is great 2019-10-05T22:15:48 < Cracki> foc/dtc can keep torque constant. without it, just dumb commutation, you get torque ripple, hence vibration 2019-10-05T22:16:07 < BrainDamage> a cnc machine will likely use sinusoidal drive as well 2019-10-05T22:16:14 < BrainDamage> since it allows accurate control 2019-10-05T22:16:22 < R2COM> so...FOC/DTC still needs hall sensors too for system to know when to commutate? 2019-10-05T22:16:38 < Cracki> the meaning of FOC is that you move the direction (and magnitude) of the magnetic field around on a unit circle, and you want it 90 degrees ahead of the rotor for best torque 2019-10-05T22:16:50 < jpa-> you can use either hall or BEMF for FOC/DTC, though with the same limits that BEMF only works for high speed 2019-10-05T22:16:53 < Cracki> it needs *something* to know rotor pos 2019-10-05T22:17:13 < Cracki> some applications use (magnetic/optical) rotary encoers 2019-10-05T22:17:40 < R2COM> i think im missing something... but...if i use FOC/DTC why i need hall/BEMF? 2019-10-05T22:17:46 < Cracki> blind commutation would mean open loop, almost like a typical stepper motor application 2019-10-05T22:17:47 < R2COM> cant i just use info from 3 current sensors? 2019-10-05T22:17:51 < R2COM> to know when to commutate? 2019-10-05T22:18:07 < Cracki> foc/dtc is OUTPUT, hall/bemf/encoders is INPUT 2019-10-05T22:18:08 < jpa-> R2COM: no, because phase current only tells you the magnetic field orientation, it doesn't tell anything about rotor orientation 2019-10-05T22:18:17 < R2COM> ah right 2019-10-05T22:18:23 < R2COM> oritntation still needs to be known 2019-10-05T22:18:35 < R2COM> FOC/DTC = estentsion of hall sensing 2019-10-05T22:18:38 < R2COM> extension* 2019-10-05T22:18:49 < jpa-> more like extension of the 6-step control of fets 2019-10-05T22:18:57 < BrainDamage> you want the rotor to always be orthogonal to the mag field 2019-10-05T22:19:03 < BrainDamage> for best torque 2019-10-05T22:19:06 < Cracki> no, it's not an extension of hall sensing 2019-10-05T22:19:14 < Cracki> it's how you shape the coil currents 2019-10-05T22:19:17 < BrainDamage> it's an extension of trapezoidal drive 2019-10-05T22:19:21 < R2COM> and with current that can be achieved 2019-10-05T22:19:26 < R2COM> current snesing* 2019-10-05T22:19:40 < Cracki> distinguish between sensing and acting 2019-10-05T22:19:44 < Cracki> */actuating 2019-10-05T22:19:44 < R2COM> so it would run more silent too you said? 2019-10-05T22:19:49 < Cracki> yes 2019-10-05T22:20:10 < Cracki> but if your PWM (for coil current control) is in the audible range, you'll hear it whine 2019-10-05T22:20:21 < Cracki> it won't vibrate like trapezoidal tho 2019-10-05T22:20:22 < jpa-> https://performance-motion-devices.s3.amazonaws.com/uploads%2Fbadcca93-5c33-4c3f-b031-fe068df58275%2FF5_Commutation+Schemes+-+final.jpg sinusoidal vs. 6-step, sinusoidal is much quieter - though you can do sinusoidal driving *without* using FOC/DTC also, you just don't get the torque and efficiency benefits 2019-10-05T22:21:13 < R2COM> okkay 2019-10-05T22:21:35 < R2COM> so FOC/DTC = run quiter without compromising efficienty 2019-10-05T22:22:12 < R2COM> lets see what would one need then... 3 extra precision resistors to stick between each phase 2019-10-05T22:22:19 < R2COM> but 2019-10-05T22:22:29 < R2COM> if one goes with that, then his sensing needs to be relatively low noise too 2019-10-05T22:22:46 < R2COM> soo... not sure if using STM's internal ADC would make it 2019-10-05T22:22:48 < R2COM> maybe it would 2019-10-05T22:22:59 < R2COM> 3 resistors + amplifier most likely to signal condition it 2019-10-05T22:23:00 < R2COM> and ADC 2019-10-05T22:23:14 < srk> yup and some filtering 2019-10-05T22:23:31 < Cracki> there are chips that take care of all that 2019-10-05T22:23:54 < srk> drv8306 drv8323 2019-10-05T22:23:57 < jpa-> R2COM: the minimum for current sensing is one resistor between the FET ground and the real ground - then you can switch fets on and off to see each phase current 2019-10-05T22:24:11 < Cracki> the current control and mosfet driving at least. i see that you want to have fun with commutation 2019-10-05T22:24:14 < R2COM> and now the speed control loop will get as input the sensed current, and as output PWM period 2019-10-05T22:24:44 < Cracki> not quite 2019-10-05T22:24:45 < jpa-> alternative is to use something like ACS712 isolated current sensor to stick it between the motor wires and FETs 2019-10-05T22:25:01 < srk> voltage, current and angle :) 2019-10-05T22:25:44 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-05T22:26:29 < R2COM> drv8306 is nice 2019-10-05T22:26:40 < jpa-> https://github.com/PetteriAimonen/ebike-controller/blob/master/src/motor_control.c#L82 here's my control loop, it takes target current (i.e. what torque user wants), the measured current, and the measured motor orientation 2019-10-05T22:26:47 < R2COM> i used ACS sensors for some other stuff 2019-10-05T22:29:17 < R2COM> ill be looking later into current sensing 2019-10-05T22:29:23 < srk> R2COM: meant drv8301, 8306 looks too smart 2019-10-05T22:29:36 < R2COM> but for this application most likely will go with just hall sensing, since i can do all that with <$1 2019-10-05T22:29:52 < R2COM> otherwise its MCU + sensors/amplifiers 2019-10-05T22:30:09 < R2COM> STM32F030C6 in qty costs $0.8 or less 2019-10-05T22:30:43 < jpa-> R2COM: yeah, and even with hall sensing if noise becomes a problem you can do sinusoidal driving in software, and if you need better torque / efficiency, you can tune phase timing manually if the RPM range is somewhat constant 2019-10-05T22:31:57 < R2COM> doing "sinusoidal driving in software" means that during the ON period of any step, increas/decrease PWM dynamically, right? 2019-10-05T22:32:16 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-05T22:33:53 < jpa-> R2COM: yeah, basically use timer to measure how long it takes between hall sensor pulses, and then divide that into equal portions and lookup sinewave from table and use that for the pwm values 2019-10-05T22:34:28 < R2COM> hmmm okay i definitely would wanna do that 2019-10-05T22:34:31 < R2COM> by the way 2019-10-05T22:35:01 < R2COM> jpa- i heard that new thing now is some new type of sensing "MR" magnetically resonant sensors which are currently used in advanced systems like electric superchargers 2019-10-05T22:35:13 < R2COM> i wonder if they are much more efficient than regular hall sensors 2019-10-05T22:35:20 < R2COM> and dont require DTC/FOC 2019-10-05T22:35:39 < R2COM> https://www.analog.com/en/technical-articles/motor-control-performance.html# 2019-10-05T22:37:00 < jpa-> R2COM: yeah, that's used in the magnetic encoders Cracki mentioned 2019-10-05T22:37:36 < jpa-> it's nice in that you get high-resolution rotor angle with simple hardware - but many magnetic angle sensors are quite slow, so it's not great for high-rpm motors yet 2019-10-05T22:38:06 < R2COM> i wonder what max RPM motor is used in cars like Tesla? 2019-10-05T22:38:39 < jpa-> google says 18krpm, quite high actually 2019-10-05T22:38:44 < Cracki> it's not direct drive but something like 1:10, take the average wheel circumference 2019-10-05T22:39:08 < R2COM> i wonder why cuz its not gonna spin that high in the final system 2019-10-05T22:39:38 < jpa-> probably optimized for efficiency 2019-10-05T22:40:17 < jpa-> IIRC at higher rpm you get less resistive losses but more magnetic losses 2019-10-05T22:40:28 < jpa-> and there is a sweet spot somewhere in between 2019-10-05T22:41:28 < jpa-> looks like that ADA4571 can do 30krpm at 0.8deg error, and it's not even expensive (just $3) 2019-10-05T22:41:40 < Cracki> 21" diameter, 90 mph, 1440 rpm 2019-10-05T22:41:51 < Cracki> so it *might* be direct drive 2019-10-05T22:42:22 < Cracki> or 1:10 ¯\_(ツ)_/¯ 2019-10-05T22:46:30 < R2COM> yeah! 2019-10-05T22:56:08 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 252 seconds] 2019-10-05T22:57:29 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-05T23:04:47 -!- syn0 [hoofman@sdf-eu.org] has joined ##stm32 2019-10-05T23:07:03 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-05T23:12:27 < R2COM> but that sensor is more like for those who gonna be making motors 2019-10-05T23:12:29 < R2COM> i guess 2019-10-05T23:12:39 < R2COM> i mean it has to be stuck there inside somehow 2019-10-05T23:13:13 < R2COM> its definitely not for simple applications where you buy cheapo assembed motor 2019-10-05T23:14:24 < jpa-> you can just glue a magnet at the end of shaft and stick the sensor on pcb there 2019-10-05T23:14:37 < Cracki> find me a bldc without hall sensors 2019-10-05T23:14:39 < jpa-> https://tropical-labs.com/mechaduino/ like this retrofit for stepper motors does 2019-10-05T23:14:43 < Cracki> they'd be the cheapest shit you can have 2019-10-05T23:14:58 < Cracki> it's trivial to get one with hall sensors at least 2019-10-05T23:17:27 < Cracki> I was wondering what you meant by magnetic resonance... >anisotropic magnetoresistive (AMR) technology 2019-10-05T23:18:10 -!- syn0 [hoofman@sdf-eu.org] has quit [Quit: leaving] 2019-10-05T23:19:53 < Cracki> AMS has the AS5048A and a bunch more, melexis has MLX903xx, ... projects like VESC popularize particular parts 2019-10-05T23:34:23 < karlp> PaulFertser: emacs is most weird, it's non-modal, but doesn't support the most common pico/nano non-modal keybindings without any good reason, it's just silly.... :) 2019-10-05T23:35:56 < BrainDamage> emacs and vim makes hard things less hard, but also make simple things harder 2019-10-05T23:40:39 < karlp> aandrew: (et al) none of the cp2104 cheap dongles i have use the ftdi cable layout, and the ch34x dongles I have are in a box, but iirc, they had a third way. 2019-10-05T23:50:19 < aandrew> karlp: *nods* 2019-10-05T23:51:22 < karlp> and none of those match what's on those goddamn bluepills iirc, though I've not confirmed 2019-10-05T23:51:35 < karlp> srk: what is your "diff" trying to say 2019-10-05T23:52:00 < karlp> and why the fuck are you using your own style of formatting diffs if you wnat a diff 2019-10-05T23:52:22 < karlp> sort and use diff directly, get a common form that people are used to reading 2019-10-05T23:53:00 < karlp> does your complaint basically come down to "OMG, THEY ADDED FUNCTIONALITY TO THE UART IN NEWER PARTS" ? 2019-10-05T23:54:03 -!- renn0xtk9 [~max@2a02:810d:1540:2448:15ce:6cc1:23ca:afa2] has quit [Quit: Konversation terminated!] 2019-10-05T23:55:11 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 268 seconds] --- Day changed Sun Oct 06 2019 2019-10-06T00:10:37 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 240 seconds] 2019-10-06T00:12:08 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-06T00:12:45 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-06T00:17:46 < PaulFertser> karlp: how are pico/nano keybindings common? Readline is really common, all shells by default use similar bindings, and plenty of other interactive programs. 2019-10-06T00:18:38 < karlp> you're assuming people know readline bindings who aren't emacs people 2019-10-06T00:18:54 < karlp> pico/nano say the bindings on the screen. those win. 2019-10-06T00:19:08 < karlp> only emacs people expect the bindings to be the same as their shell. 2019-10-06T00:25:07 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-06T01:11:55 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-06T01:13:53 < zyp> catphish, just like the single buffered one, except you have two buffers that you alternate between writing to 2019-10-06T01:14:23 < zyp> full disclosure: I haven't added support for it myself, but it seems really straight forward 2019-10-06T01:17:13 < catphish> zyp: the part that has confused me is when you switch the buffers, i think the answer is that you do it immediately when the interrupt for the previous transfer complete fires (assuming the next buffer is ready) 2019-10-06T01:18:03 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-06T01:18:06 < catphish> so one needs an extra variable to say that the current software facing buffer has been filled, and if that's set, swap the buffers on the TX complete interrupt 2019-10-06T01:18:26 < catphish> if not, swap them, as soon as that buffer is full 2019-10-06T01:18:29 < zyp> what do you mean by switch? you've got two buffers, either can be either free or pending 2019-10-06T01:19:29 < catphish> there are 2 buffers, one is being accessed by hardware 2019-10-06T01:20:54 < zyp> by the way, are you sending or receiving? I wanna use the right terminology 2019-10-06T01:21:43 < catphish> i'm sending from the STM32 to the host, "IN" in USB terminology 2019-10-06T01:21:57 < zyp> good, that's what I assumed 2019-10-06T01:22:25 < catphish> at some point you have to toggle the "SW_BUF" flag to mark the current sofware-facing buffer as avaiable to the hardware to send 2019-10-06T01:24:14 < catphish> but (because there are only 2 buffers), has the side effect of marking the current hardware-facing buffer as software-facing, so you shouldn't do it until the hardware has finished the current operation 2019-10-06T01:24:53 < catphish> so it seems to me that the correct approach is to do this "switch" at the moment when the previous operation finishes 2019-10-06T01:25:12 < zyp> I feel like you're either overthinking something or going the wrong way about something, but let me finish reading up on what the RM says 2019-10-06T01:25:32 < catphish> please do 2019-10-06T01:25:43 < catphish> i couldn't make any sense of the RM 2019-10-06T01:27:44 < zyp> hmm, I'm looking at RM0376 rev1, there's a typo in the buffers usage table, not sure it's the same in other RMs 2019-10-06T01:27:58 < zyp> which had me confused for a moment 2019-10-06T01:27:58 < catphish> i'm reading RM0394 2019-10-06T01:29:07 < zyp> anyway, you're right, here's my take: 2019-10-06T01:29:48 < zyp> first of all, you don't have to care about the normal flow control mechanism (STAT=NAK/VALID), it remains VALID after a transfer 2019-10-06T01:30:19 < zyp> flow control is instead implemented by the DTOG/SW_BUF mechanisms, it's NAK when they are equal, VALID when they are not 2019-10-06T01:30:21 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-06T01:31:12 < catphish> my questions are: when can you populate the software facing buffer (as defined by SW_BUF), and when should you toggle SW_BUF 2019-10-06T01:31:16 < zyp> so initially you fill the first buffer and get the transfer started, that sets up DTOG=0, SW_BUF=1, and meanwhile you can then fill the second buffer 2019-10-06T01:31:53 < zyp> once the first packet is transferred, DTOG switches to 1, and the endpoint will NAK until you toggle SW_BUF=0 2019-10-06T01:32:16 < zyp> which means that that's when you should toggle SW_BUF 2019-10-06T01:33:21 < catphish> so you populate TX_0, then you set DTOG=0, SW_BUF=1, then you start the transfer, at which point, you're also free to populate TX_1 2019-10-06T01:33:33 < zyp> correct 2019-10-06T01:33:55 < catphish> then... you wait for a "TX complete" interrupt? 2019-10-06T01:33:58 < zyp> and when DTOG=1 by hardware and you're done populating, you do SW_BUF=0 2019-10-06T01:34:12 < zyp> and yes, waiting for TX complete is the natural way 2019-10-06T01:35:13 < zyp> this means you still have to hit SW_BUF at a low latency to not cause flow control to NAK 2019-10-06T01:35:15 < catphish> so when you receive the "TX complete" interrupt, you toggle SW_BUF, but only if you've finished populating TX_! 2019-10-06T01:35:18 < catphish> *TX_1 2019-10-06T01:35:21 < zyp> yes 2019-10-06T01:35:44 < zyp> this reconfirms why I think double buffering is pretty useless 2019-10-06T01:36:32 < catphish> so essentially you have to toggle SW_BUF *extremely* fast when that interrupt occurs 2019-10-06T01:36:43 < zyp> in a single buffered setup, you have to copy 64 bytes and set STAT=VALID at each TX event 2019-10-06T01:36:56 < zyp> in a double buffered setup you have to toggle SW_BUF 2019-10-06T01:36:56 < catphish> so that the next bulk transfer (which essentially happens instantly) is ready to go 2019-10-06T01:37:55 < catphish> i had it in my head that it should be possible to do this pre-emptively, ie toggle SW_BUF as soon as you've finished populating the buffer 2019-10-06T01:38:08 < zyp> copying 64 bytes is actually not a big operation, interrupt entry/exit is probably around the same order of magnitude 2019-10-06T01:38:29 < zyp> so I don't think double buffering helps much at all 2019-10-06T01:38:54 < catphish> the problem is that the time between bulk transfers is essentially zero 2019-10-06T01:39:11 < catphish> each one immediately follows the previous one 2019-10-06T01:39:33 < zyp> how do you tell? 2019-10-06T01:39:34 < catphish> so this approach seems dodgy 2019-10-06T01:40:21 < catphish> double buffering would be great *if* you could populate the inactive buffer and pre-emptively mark it as ready 2019-10-06T01:40:42 < catphish> which i believe is possible, but i don't know when you would do the toggle 2019-10-06T01:40:46 < zyp> have you actually tested a single buffered solution and confirmed that the device NAKs? 2019-10-06T01:41:40 < catphish> zyp: unfortunately no, i don't have that data, but what i do know is that my throughput is a little under 4Mbps, which is about 50% of my expectation 2019-10-06T01:42:05 < zyp> but you don't know where your bottleneck is 2019-10-06T01:44:51 < karlp> are you on using an ULPI phy for HS? 2019-10-06T01:44:57 < catphish> that's correct 2019-10-06T01:45:09 < catphish> karlp: i'm using FS 2019-10-06T01:45:25 < zyp> catphish, https://bin.jvnv.net/file/5pyzn.png 2019-10-06T01:45:37 < zyp> IIRC that's a trace of this code: https://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_msc 2019-10-06T01:46:02 < zyp> it's mass storage, fetching data from an SD card 2019-10-06T01:46:06 < catphish> how do you make such a trace> 2019-10-06T01:46:34 < catphish> linux usb pcap is vague :( 2019-10-06T01:46:37 < zyp> SD card reads are 512B, so flow control are invoked ahead of every eight usb packet 2019-10-06T01:46:55 < zyp> yeah, host side capture is useless, you don't get any of the interesting details 2019-10-06T01:47:07 < zyp> that trace was taken with a beagle 2019-10-06T01:48:01 < zyp> in case my point is unclear, the [x POLL] notation is how many times the host had to ask before it got data 2019-10-06T01:48:09 < catphish> so you were doing back to back transmits with no NAK? 2019-10-06T01:48:10 < zyp> the packets without that all replied on the first poll 2019-10-06T01:48:29 < zyp> well, yeah, the 7 packets in between each SD card fetch 2019-10-06T01:48:35 < catphish> the datasheet implies it's impossible to do that 2019-10-06T01:48:58 < zyp> also, this was with dwc_otg, not st_usbfs, but still 2019-10-06T01:49:08 < catphish> if that's possible that there's no point in me doing double buffering 2019-10-06T01:49:18 < zyp> I told you that before :) 2019-10-06T01:49:37 < catphish> i wonder where my bottleneck is then 2019-10-06T01:49:48 < zyp> hmm 2019-10-06T01:50:31 < catphish> i wonder if i should write a test that just dumps 64 bytes of \0 as fast as possible and benchmark it 2019-10-06T01:50:50 < zyp> yes, do that 2019-10-06T01:50:52 < catphish> i wouldn't even need to write the data at all 2019-10-06T01:51:02 < zyp> also, which chip were you using again? 2019-10-06T01:51:09 < catphish> l433 2019-10-06T01:51:25 < zyp> that should be reasonably fast, no? 2019-10-06T01:51:40 < catphish> yes, 80MHz core 2019-10-06T01:52:25 < karlp> i'm getting ~499KB/sec just now running theh locm3 perf tests, but i've seen ~800 or so in the past. 2019-10-06T01:52:35 < karlp> I would have to shuffle what's plugged in where.. 2019-10-06T01:52:42 * karlp lsusb -t's and disconnects things 2019-10-06T01:52:51 < zyp> hmm 2019-10-06T01:53:10 < zyp> on my trace, the packets go out at 70us intervals, it seems 2019-10-06T01:53:44 < zyp> at 12Mb/s, that equals about 105 bytes of bus time, to transfer 64 bytes of payload plus some overhead 2019-10-06T01:54:13 < zyp> I'm not sure how many cycles it takes to fill the buffer, maybe a couple hundred, 2019-10-06T01:54:30 < zyp> which at 80 MHz is a few us 2019-10-06T01:54:41 < catphish> i need 4Mbps for my application 2019-10-06T01:55:06 < catphish> i'm getting just marginally less right now (it works sometimes but fails sometimes) 2019-10-06T01:55:25 < karlp> hrm, 515/505, not sure, I'm pretty sure I've seen a lot higher in the past, 2019-10-06T01:55:46 < zyp> hmm 2019-10-06T01:56:17 < catphish> "on my trace, the packets go out at 70us intervals, it seems" << the problem is that you don't have the period from the start of one packet to the start of the next, you have the period *between* packets 2019-10-06T01:56:21 < zyp> 64B/70us is around 7.3 Mb/s 2019-10-06T01:56:37 < zyp> yes 2019-10-06T01:57:10 < zyp> but there's surely a couple of microseconds to spare there 2019-10-06T01:57:15 < catphish> ie the period betwen the "TX complete" interrupt, and the start of the next request 2019-10-06T01:57:31 < catphish> i wonder how long that is 2019-10-06T01:57:43 < karlp> ah right, ok. read 566KB/sec write 790KB/sec. 2019-10-06T01:57:51 < karlp> by removing the delays from the gadget0 stuff 2019-10-06T01:59:12 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-06T01:59:31 < zyp> catphish, anyway, if you really want sane double buffering, switch to a chip with dwc_otg 2019-10-06T01:59:42 < karlp> ok, 590KB/s and 805KB/s by dropping all delays and moving to an isolated port on the root hub. 2019-10-06T02:00:08 < catphish> 590KB/s isn't much more than i'm achieving 2019-10-06T02:00:11 < zyp> it's using a FIFO that you can just keep feeding as long as there's room in it, and it will base flow control on whether there's enough in the FIFO to put together a whole packet 2019-10-06T02:00:24 < karlp> catphish: yes, but that's completely naive single buffered. 2019-10-06T02:00:31 < karlp> that's 6.6MBps, so... 50% more than you? 2019-10-06T02:01:06 < catphish> no, 590 * 8 = 4720 2019-10-06T02:01:06 < karlp> no, sorry, used the 805 tiing, not the 590 timing. 2019-10-06T02:01:09 < zyp> if I find time, I could try doing some benchmarks for you 2019-10-06T02:01:24 < karlp> copied the wrong line when converting from the gadget0 output to your numbering goals .) 2019-10-06T02:01:28 < zyp> on some st_usbfs chip 2019-10-06T02:02:12 < karlp> (my numbers are from su_usbfs) 2019-10-06T02:02:17 < karlp> let me plug in a dwc_otg board 2019-10-06T02:03:28 < catphish> my worst case requirement is 500,000 bytes per second 2019-10-06T02:04:09 < catphish> and in my real world tests, that is sometimes (but not always) achieveable] 2019-10-06T02:04:35 < zyp> I still don't know why you're not just getting enough memory to buffer the entire track so you don't have to deal with real time transfers 2019-10-06T02:04:36 < karlp> 872/805 for the same setup. 2019-10-06T02:04:56 < catphish> zyp: how would you suggest doing so? 2019-10-06T02:05:08 < catphish> i'm entirely happy with that idea 2019-10-06T02:05:11 < zyp> sdram? 2019-10-06T02:05:17 < zyp> I forgot how much you need 2019-10-06T02:05:38 < zyp> doesn't some of the highend chips come with like 1M of sram now? 2019-10-06T02:05:53 < catphish> right now my BOM is just an stm32, a voltage regulator, and 7 capacitors, it's insanely simple 2019-10-06T02:06:13 < karlp> zyp: https://imgflip.com/i/3chjud :) 2019-10-06T02:06:14 < catphish> but if i can find a way to hook up some ram, i'd love to 2019-10-06T02:06:34 < zyp> catphish, how much do you need? 2019-10-06T02:06:52 < karlp> so yeah, dumb naive f4 instead of f3 is ~twice as fast on read (devce->host) and ~same on write (host->device) 2019-10-06T02:07:00 < zyp> f7 got 512k, h7 got 1M 2019-10-06T02:07:02 * catphish calculates 2019-10-06T02:07:17 < zyp> and that's just with a plain mcu that only needs power and some caps 2019-10-06T02:07:29 < zyp> karlp, that's interesting 2019-10-06T02:07:38 < zyp> didn't think st_usbfs would be that bad 2019-10-06T02:08:00 < catphish> zyp: minimum 256KB, ideally 1MB 2019-10-06T02:08:08 < karlp> I don't remember stusbfs being so async for the two directions, but it's been a while since I ran the perf tests. 2019-10-06T02:08:08 < zyp> grab a h7 then 2019-10-06T02:09:00 < karlp> f413 and f469 both have 384k ram too 2019-10-06T02:09:20 < catphish> the h7 is a big boy :( 2019-10-06T02:09:22 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-06T02:09:56 < karlp> l4+ are all 640k ram 2019-10-06T02:10:18 < catphish> karlp: that might work then 2019-10-06T02:10:18 < karlp> that's just staying in st 2019-10-06T02:10:34 < karlp> l496 is 320k 2019-10-06T02:11:41 < karlp> mp1 has 448 for the m4 core... 2019-10-06T02:11:57 < catphish> one track of DD floppy data is 0.2s @ 500,000 bps (100KB), i sample for 0.5s for redundancy, so 250KB 2019-10-06T02:12:07 < karlp> anyway, this is all distracting me from my problems .) 2019-10-06T02:12:42 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-06T02:12:43 -!- day__ is now known as day 2019-10-06T02:12:47 < catphish> 640k would work perfectly, and also allow me to work with HD disks (1Mbps), or 500KB for 0.5s 2019-10-06T02:13:05 < karlp> how long does it take to read a track? 2019-10-06T02:13:13 < catphish> 0.2s 2019-10-06T02:13:14 < karlp> could you just spin a few more times anyway? 2019-10-06T02:13:43 < catphish> (i read 0.5s for redundancy, but the actual track length is 0.2s) 2019-10-06T02:14:03 < karlp> can you do the redundancy on the host instead? 2019-10-06T02:14:15 < karlp> have them read 0.3s 3 times for instance? 2019-10-06T02:14:17 < karlp> let them sort it out? 2019-10-06T02:14:22 < catphish> yes 2019-10-06T02:14:27 < catphish> that would work 2019-10-06T02:14:45 < karlp> move the ram constraint somewhere where it stops being a constraint 2019-10-06T02:15:16 < catphish> i prefer to read 2n+1, that guarentees that the start and end of the track are continuously read 2019-10-06T02:15:35 < catphish> ie 0.400000001 seconds 2019-10-06T02:16:31 < catphish> if i can get a chip with 500kB+ this is moot 2019-10-06T02:17:13 < karlp> man zephy has halted on a fault, but I can't find docs on how to decode it 2019-10-06T02:17:31 < karlp> well, 500 is much mor elimiting thatn 384 for instance :) 2019-10-06T02:18:27 < catphish> ah yes i see there are l4 with 640KB of RAM 2019-10-06T02:18:41 < catphish> that would be amazing here 2019-10-06T02:19:05 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-06T02:20:05 < catphish> STM32L4R5VG looks to be the cheapest, 100 pins, 640KB RAM, 1MB flash, nice, $5.67 official price 2019-10-06T02:20:44 < catphish> cheaper than adding external RAM 2019-10-06T02:21:10 < catphish> and 120MHz, a pleasant improvement 2019-10-06T02:22:11 < catphish> thanks humans 2019-10-06T02:23:30 < Laurenceb> tfw garbage human 2019-10-06T02:24:05 < karlp> 512k sram is only $1.32, but yeah, sure, less placement and sw hassle 2019-10-06T02:24:17 < karlp> 512KB even, not Kb. 2019-10-06T02:25:21 < karlp> tfw lolrence arrives, ignores whatevers goign on, and just rants on bullshit 2019-10-06T02:26:20 < Laurenceb> correct 2019-10-06T02:31:33 < antto> 2019-10-06T02:32:38 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2019-10-06T02:33:08 < catphish> karlp: what part is that? 2019-10-06T02:33:36 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-06T02:33:39 < karlp> dunno, close theh tab, some cypress 2019-10-06T02:33:44 < catphish> 512KB of SRAM would be a sensible idea if the interconnect was simple 2019-10-06T02:33:54 < karlp> just digikey, memory, include sram and psram, filter size, order by price 2019-10-06T02:34:24 < karlp> don't think you'll find "simple" ones big enough for your ideas 2019-10-06T02:35:08 < catphish> i just checked mouser, i see 512KB 16-bit SRAM for £1.50 2019-10-06T02:35:34 < catphish> but i'd need to find a 16 bit contiguous bus :( 2019-10-06T02:35:50 < karlp> "but yeah, sure, less placement and sw hassle" 2019-10-06T02:36:02 < catphish> yeah i see 2019-10-06T02:36:03 < karlp> you'r enot doign enough of these for the BoM cost to matter much are you? 2019-10-06T02:36:43 < catphish> aiming for a batch of 100, it's not critical 2019-10-06T02:37:11 < catphish> serial sram seem really small :( 2019-10-06T02:37:22 < karlp> yeah, they're legacy shits 2019-10-06T02:37:40 < karlp> like, "I want to have some ram" on my hc11 with 512bytes of ram. 2019-10-06T02:37:50 < catphish> i find it very odd that nobody makes a large cheap serial RAM 2019-10-06T02:38:00 < catphish> i assume there's a reason 2019-10-06T02:38:01 < karlp> why? 2019-10-06T02:38:04 < karlp> most people don't need them 2019-10-06T02:38:13 < karlp> even you don't really want it, you just want a chip with enough on board 2019-10-06T02:38:16 < karlp> you don't want the hassle 2019-10-06T02:38:27 < karlp> and if you need that much, you presumably need it fast enough to be useful. 2019-10-06T02:38:36 < karlp> pretty small intersection 2019-10-06T02:39:14 < catphish> well the issue is that most MCUs don't have 1MB of RAM, but 128MB of DRAM is cheap, surely people want a few MB of DRAM for their arduino project 2019-10-06T02:39:45 < catphish> just seems odd that people don't sell DRAM + serializer on a chip 2019-10-06T02:40:12 < karlp> "pretty small intersection" 2019-10-06T02:40:45 < catphish> i assume you're correct 2019-10-06T02:41:01 < catphish> despite my repeate desire for such an IC 2019-10-06T02:41:40 < catphish> i was looking at making an audio looper device, same bottleneck 2019-10-06T02:50:22 < catphish> chips like this are ideal (apart from the cost) https://www.mouser.co.uk/ProductDetail/Cypress-Semiconductor/CY15B104Q-SXI?qs=2De2tSDgB01Uyxkl7oDfXA%3D%3D 2019-10-06T02:55:31 < kakimir32> 151years 2019-10-06T02:57:56 < antto> https://i0.kym-cdn.com/photos/images/newsfeed/000/281/983/f6c.jpg 2019-10-06T03:00:01 * karlp eats more horse instead 2019-10-06T03:11:46 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-06T03:21:14 < Laurenceb> >people are protesting at Joker film 2019-10-06T03:21:57 < Laurenceb> https://patents.google.com/patent/US10144532B2/en 2019-10-06T03:23:07 < Laurenceb> US gov just went full emdrive 2019-10-06T03:23:13 < Laurenceb> never go full emdrive 2019-10-06T03:23:36 < Laurenceb> >Matter, energy, and spacetime are all emergent constructs which arise out of the fundamental framework that is the vacuum energy state. 2019-10-06T03:23:49 < Laurenceb> citation required 2019-10-06T03:38:10 < mawk> lol 2019-10-06T03:41:45 < specing> catphish: for a few MBs you can get such chips, no? 2019-10-06T03:42:17 < specing> My guess is that the reason is in the process technology 2019-10-06T03:42:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-06T03:54:14 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 252 seconds] 2019-10-06T04:13:35 < kakimir32> Laurenceb: https://www.youtube.com/watch?v=GFcQevflStk 2019-10-06T04:15:55 < Thorn> catphish: https://lcsc.com/product-detail/RAM_Lyontek-Inc-LY68L6400SLIT_C261881.html 2019-10-06T04:18:31 < kakimir32> what is psuedo-ram? 2019-10-06T04:18:38 < kakimir32> psuedo-sram* 2019-10-06T04:18:59 < Thorn> pseudo-static means it's dram with a sram-like interface 2019-10-06T04:19:01 < karlp> dram with onmboard refresh circuitry so it looks like sram 2019-10-06T04:19:24 < Thorn> in this case it's more like serial flash than sram though 2019-10-06T04:19:25 < karlp> Thorn: thanks, I tried searching for psram on lcsc first and didn't find anything 2019-10-06T04:19:37 < kakimir32> yes 2019-10-06T04:19:45 < kakimir32> that is why it's cheap? 2019-10-06T04:20:04 < Thorn> it's cheap because China 2019-10-06T04:21:21 < Thorn> I believe these chips are in esp32-wrover modules for example 2019-10-06T04:23:36 < karlp> so... l1 has SPI_CR2_FRF, and cube LL driver has an #ifdef SPI2_CR2_FRF around LL_SPI_SetStandard 2019-10-06T04:23:43 < karlp> so far so good, should be avialabel on l1. 2019-10-06T04:24:13 < karlp> zephyr says it's undefined, and their modules/hal/ seems to have a soc/stm32lxxxx for ... some variants. 2019-10-06T04:24:17 < karlp> this must be a zephyr bug. 2019-10-06T04:30:10 < karlp> hah 2019-10-06T04:30:11 < karlp> it's a cube bug 2019-10-06T04:30:17 < karlp> mother fuckers 2019-10-06T04:30:27 < karlp> cube exported project doesn't have it either 2019-10-06T04:31:43 < karlp> I don't get why Drivers/CMSIS/Device/ST/STM32L1xx/Include/ doesn't have the xxx6 and xxx8 flash size files either... 2019-10-06T04:34:15 < karlp> ok, that's just because they define the bigest for each, so xx6 and xx8 are just xxxb for them. 2019-10-06T04:34:49 < karlp> still a cube bug that FRF isn't defined in the headers 2019-10-06T04:35:24 < karlp> it's defined in the xxxc files, but not the xxxb files. 2019-10-06T04:36:37 -!- fenugrec [~fenugrec@98.159.33.226] has joined ##stm32 2019-10-06T04:56:42 < karlp> and even with it defined it crashes on something "illegal" 2019-10-06T04:56:50 < karlp> fucking hal bullshits why did I start this 2019-10-06T05:11:29 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 276 seconds] 2019-10-06T05:33:37 -!- fenugrec [~fenugrec@98.159.33.226] has quit [Ping timeout: 240 seconds] 2019-10-06T05:58:54 < aandrew> what's zephyr 2019-10-06T06:10:46 -!- smvoss [~smvoss@207.191.220.92] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-06T06:31:57 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-06T06:37:47 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Bye!] 2019-10-06T06:39:17 -!- fc5dc9d4 [~quassel@p57A320C7.dip0.t-ipconnect.de] has joined ##stm32 2019-10-06T06:44:02 -!- fc5dc9d4_ [~quassel@p57A32E60.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-06T06:52:40 -!- ohsix [~ohsix@bc175210.bendcable.com] has quit [Ping timeout: 265 seconds] 2019-10-06T07:03:03 -!- ohsix [~ohsix@bc175210.bendcable.com] has joined ##stm32 2019-10-06T07:15:38 -!- ohsix [~ohsix@bc175210.bendcable.com] has quit [Ping timeout: 276 seconds] 2019-10-06T07:21:35 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-06T07:22:21 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 250 seconds] 2019-10-06T07:22:22 -!- BrainDamage_ is now known as BrainDamage 2019-10-06T07:24:36 -!- ohsix [~ohsix@bc175210.bendcable.com] has joined ##stm32 2019-10-06T07:35:38 < R2COM> dafuq 2019-10-06T07:35:48 < R2COM> dongs purchased Destiny 2 game? 2019-10-06T07:38:17 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-06T07:57:59 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-06T08:00:51 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-06T08:00:51 -!- day__ is now known as day 2019-10-06T08:04:59 -!- fenugrec [~fenugrec@98.159.33.226] has joined ##stm32 2019-10-06T08:11:27 -!- fenugrec [~fenugrec@98.159.33.226] has quit [Ping timeout: 265 seconds] 2019-10-06T08:49:12 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-06T09:16:11 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-06T09:27:11 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-06T09:46:14 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-06T10:16:58 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-06T10:17:51 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-06T10:52:37 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-06T11:00:03 -!- ekaologik [~quassel@p5DE86A6C.dip0.t-ipconnect.de] has joined ##stm32 2019-10-06T11:01:59 < dongs> R2COM: its free lol 2019-10-06T11:02:05 < dongs> why would i pay for it 2019-10-06T11:02:08 < dongs> i spent 5 mins in it and deletd 2019-10-06T11:02:15 < dongs> its like a dumb chink RPG with guns 2019-10-06T11:04:42 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-06T11:24:23 < dongs> god damn this fucking PCBA job 2019-10-06T11:24:28 < dongs> fuckers made it in kikecad 2019-10-06T11:24:47 < dongs> just got the PCBs and there's 2 blank spots for oscillator, which of course wasn't on the shitty BOM 2019-10-06T11:25:11 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-06T11:27:22 -!- sterna [~Adium@2a02:aa1:1013:10f6:a92a:702a:b603:f90] has joined ##stm32 2019-10-06T12:01:30 < catphish> i think i'll send my PCBA jobs to dongs :) he sounds like he enjoys untangling shitty data files 2019-10-06T12:11:26 < PaulFertser> I'd expect him to charge extra for any project involving kicad :) 2019-10-06T12:12:39 < specing> *kikecad 2019-10-06T12:12:49 < catphish> i should actually figure how how to cost effectively do assembly soon 2019-10-06T12:14:24 < specing> catphish: jlcpcb assembly? 2019-10-06T12:14:38 < specing> catphish: jlcpcb partial* assembly 2019-10-06T12:15:01 < catphish> i looked at that, but it seemed like they'd basicaly just do the passives 2019-10-06T12:15:43 < catphish> i need at least the stm32 assembling 2019-10-06T12:16:41 < catphish> i'll have another look at https://jlcpcb.com/client/index.html#/parts 2019-10-06T12:17:51 < catphish> looks like they do have some stm32 2019-10-06T12:20:32 < specing> some? a lot of them! 2019-10-06T12:29:17 -!- grindhold [~quassel@84.200.43.162] has quit [Ping timeout: 240 seconds] 2019-10-06T12:52:32 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-06T13:01:56 < antto> allpcb also offer assembly since some time now.. no idea if it's gud or not 2019-10-06T13:07:55 -!- sterna1 [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-06T13:10:15 -!- sterna [~Adium@2a02:aa1:1013:10f6:a92a:702a:b603:f90] has quit [Ping timeout: 246 seconds] 2019-10-06T13:13:50 < dongs> if you want cost effective assembly, definitely don't do it with me 2019-10-06T13:14:13 < dongs> and yes, i charge like $400 minimum for protos made in kikecad 2019-10-06T13:15:24 < dongs> and these oscs are next to RF chips so they're surely not gonna work with the shit missing 2019-10-06T13:16:04 < zyp> not your problem if you followed the spec you got 2019-10-06T13:19:11 < dongs> yeah 2019-10-06T13:19:15 < dongs> its the kikecad .pos file 2019-10-06T13:19:26 < zyp> fitting name 2019-10-06T13:19:27 < dongs> which for some reason decided not to incldue these. 2019-10-06T13:19:45 < dongs> also he's got 2 RF chips which are linked to same antenna im not sure why they're not just sharing one xtal 2019-10-06T13:19:56 < dongs> hes got 2 oscs going through a cap to xtal_in 2019-10-06T13:20:00 < dongs> seems like a waste 2019-10-06T13:20:24 < dongs> o well i warned that if shit doens't work its not my problem in 1st email 2019-10-06T13:21:01 < qyx> some freelancer startup thing? 2019-10-06T13:21:34 < dongs> some jap startup i think 2019-10-06T13:21:42 < dongs> retarded marine tracking shit for AIS or wahtever 2019-10-06T13:31:42 -!- fenugrec [~fenugrec@98.159.33.226] has joined ##stm32 2019-10-06T13:35:09 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-06T13:54:22 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-06T14:12:26 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-06T14:13:44 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-06T14:18:02 -!- fenugrec [~fenugrec@98.159.33.226] has quit [Ping timeout: 240 seconds] 2019-10-06T14:42:14 < Thorn> spacewalk https://www.youtube.com/watch?v=z0Ut8daLD6I 2019-10-06T15:02:03 < dongs> fucking opensores 2019-10-06T15:02:09 < dongs> why teh fuck did gcc invent typeof() shit 2019-10-06T15:02:14 < dongs> when vc doesnt support it? 2019-10-06T15:03:30 < Steffanx> Such weird question :P 2019-10-06T15:03:53 < Steffanx> How can you invent something that already exists 2019-10-06T15:04:32 < jadew> dongs, probably because they needed the functionality :) 2019-10-06T15:05:49 < jadew> dongs, are you writing C or C++? 2019-10-06T15:05:54 < dongs> C of course 2019-10-06T15:05:56 < dongs> i am not writing 2019-10-06T15:05:59 < dongs> im compiling opensores 2019-10-06T15:06:03 < dongs> but im about to just shift-del it 2019-10-06T15:06:06 < jadew> then you're out of luck 2019-10-06T15:06:16 < dongs> i tried to crosscompile using mingw on a lunixvm and of course that broke 2019-10-06T15:06:22 < dongs> loaded the shit into vstudio and its even worse 2019-10-06T15:06:27 < dongs> fucking nigger used retarded gccisms all over 2019-10-06T15:06:35 < dongs> and he has a fucking linked list thing made with fucking #defines 2019-10-06T15:06:36 < dongs> that uses typeof 2019-10-06T15:06:38 < dongs> for "safety" 2019-10-06T15:06:47 < dongs> fucking nigger 2019-10-06T15:06:50 < jadew> you could replace all that 2019-10-06T15:06:56 < jadew> with the actual type 2019-10-06T15:06:56 < dongs> yes but my time isnt worthless 2019-10-06T15:07:22 < jadew> and if you can compile it as C++, without getting even more errors, then you can replace typeof with decltype 2019-10-06T15:07:36 < dongs> how cna i force the shit to cpp 2019-10-06T15:07:42 < Steffanx> Best GNU extension is the one that allows you to write " switch () { case 1 ... 8: break ; }" 2019-10-06T15:07:44 < jadew> you have to rename the files 2019-10-06T15:07:48 < dongs> fuck outta here 2019-10-06T15:08:46 < jadew> Steffanx, what does that do? 2019-10-06T15:08:48 < PaulFertser> Hey jadew :) Can you please take a look at https://1000kanalov.ru/wa-data/public/shop/products/96/28/2896/images/3420/3420.970.jpg and explain what are the possible benefits of this design? 2019-10-06T15:09:05 < dongs> looks like a yagi or wahtever 2019-10-06T15:09:08 < Steffanx> its just a "range" thing. instead of case 1: case 2: etc. 2019-10-06T15:09:22 < PaulFertser> dongs: like 3 yagis combined in a seemingly meaningless way! 2019-10-06T15:09:29 < dongs> Compile as C++ Code (/TP) 2019-10-06T15:09:30 < dongs> found 2019-10-06T15:09:30 < jadew> yeah, looks like a yagi 2019-10-06T15:09:45 < PaulFertser> dongs: I would also suggest msys2 as that installs GCC compiler that can build native executables for windows (without cross-compiling etc) 2019-10-06T15:10:00 < PaulFertser> dongs: I can also try to cross-compile the shit for you on my machine. 2019-10-06T15:10:47 < jadew> PaulFertser, I don't know much about antennas, but I would have to assume the various parts are separated by some meaningful degree 2019-10-06T15:10:57 < dongs> PaulFertser: go ahead, libzc at shithub 2019-10-06T15:10:57 < PaulFertser> kakimir32: just a had a short ride. No fun at all, it's about 2-3C and raining (or rather snowing) heavily. 2019-10-06T15:11:01 < dongs> i gave up, it doesnt work 2019-10-06T15:11:07 < dongs> https://github.com/mferland/libzc 2019-10-06T15:11:22 < jadew> 90/180 and when picked up, the signal combines constructively to give a couple of extra dBs 2019-10-06T15:11:47 < jadew> maybe they're in phase.. who knows 2019-10-06T15:11:52 < Steffanx> just install ubuntu/whateverlinuxdistribution on your windows pc from the marketplace and apt install gcc 2019-10-06T15:12:12 < PaulFertser> jadew: hm, I'd expect combining yagis like that to give the same sensitivity but less direction selectivity in the vertical plane and just that... 2019-10-06T15:12:26 < BrainDamage> it's a log periodic 2019-10-06T15:12:32 < PaulFertser> Steffanx: but it won't compile for windows, it'll still compile for glibc. 2019-10-06T15:12:40 < Steffanx> yeah true 2019-10-06T15:13:06 < Steffanx> but somehow i got the idea dongs just needs the tool for himself 2019-10-06T15:13:22 < PaulFertser> dongs: looking 2019-10-06T15:14:36 < BrainDamage> it has the same bandwith as a single yagi, but more vertical aperture 2019-10-06T15:14:57 < PaulFertser> BrainDamage: and how is that useful? 2019-10-06T15:15:19 < BrainDamage> it makes pointing easier 2019-10-06T15:15:20 -!- tkoskine [tkoskine@kapsi.fi] has quit [Ping timeout: 276 seconds] 2019-10-06T15:15:48 < dongs> i wonder what kinda fuckerface writes this kinda nigger code in macros 2019-10-06T15:15:50 < BrainDamage> the more your antenna is directional, the more it's blind to signals not coming from its max gain 2019-10-06T15:15:55 < dongs> thats completelly undebuggable and fucking unreadable 2019-10-06T15:16:00 -!- tkoskine [tkoskine@kapsi.fi] has joined ##stm32 2019-10-06T15:16:11 < dongs> like, you cant even fucking see what the result code is because its all hacked together in preprocessor 2019-10-06T15:16:16 < BrainDamage> remember that tv antennas are mounted on vertical pole with little to no adjustment of azimuth 2019-10-06T15:16:29 < dongs> jadew: replacing typeof->declrtype directly doenst work 2019-10-06T15:16:31 < jadew> BrainDamage, does that mean that it has less gain than a regular yagi? 2019-10-06T15:16:40 < dongs> for (pos = list_entry((head)->next, typeof(*pos), member);\ 2019-10-06T15:16:42 < dongs> this is how they do ti 2019-10-06T15:16:50 < antto> aww 2019-10-06T15:16:57 < jadew> dongs, you also have to specify which c++ you're using, decltype was introduced in c++11 IIRC 2019-10-06T15:17:03 < dongs> yeah its over that 2019-10-06T15:17:10 < dongs> i mean, VC supports up to c++17 2019-10-06T15:17:15 < BrainDamage> jadew: in the horizontal plane yes 2019-10-06T15:18:31 < dongs> error C2528: 'abstract declarator': pointer to reference is illegal 2019-10-06T15:18:32 < antto> dongs the oscs were maybe custom made footprints, and they've forgot to adjust their settings, if it's "THT" then it won't go into the .pos file 2019-10-06T15:18:32 < dongs> same shit tho 2019-10-06T15:18:48 < dongs> antto: fuckin nice 2019-10-06T15:18:54 < dongs> well, i'm gonna call them tomorow morning and bitch 2019-10-06T15:19:02 < antto> a matter of RTFM 2019-10-06T15:19:04 < dongs> they dont look custom, shit's just a regular 3225 osc 2019-10-06T15:19:09 < antto> they should check their stuff 2019-10-06T15:19:13 < dongs> for sure 2019-10-06T15:19:16 < antto> oh? 2019-10-06T15:19:20 < PaulFertser> BrainDamage: hm, so your point is that it's practical to turn it around but not up/down? Then it would make sense indeed. 2019-10-06T15:19:28 < dongs> i mean footprint looks just regular 3225 thing 2019-10-06T15:19:48 < antto> there are different kinds of "3225" 4-pin footprints 2019-10-06T15:20:05 < dongs> there's a standard pinout for like 95% of oscillators in that package. 2019-10-06T15:20:19 < antto> pinout - yes, but "footprint" .. 2019-10-06T15:21:09 < antto> plus, that setting can be adjusted afterwards per instance, so they might have accidentally switched it to THT or virtual, in both cases it's not gonna apear in the .pos afaik 2019-10-06T15:24:25 < dongs> #define list_entry(ptr, type, member) \ ((type *)((char *)(ptr)-(unsigned long)(&((type *)0)->member))) 2019-10-06T15:24:29 < dongs> this is utterly disgusting 2019-10-06T15:24:35 < dongs> i think the problem is here, not in decltype shit 2019-10-06T15:25:06 < dongs> oh its fucking shit stolen from freebsd 2019-10-06T15:25:07 < dongs> no wonder 2019-10-06T15:26:23 -!- ekaologik [~quassel@p5DE86A6C.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-06T15:30:53 < Steffanx> it seems it was in the lunix kernel as well. in the linked list implementation 2019-10-06T15:34:44 < PaulFertser> dongs: that library depends heavily on pthreads... 2019-10-06T15:35:27 < dongs> and that doesnt work on windows? 2019-10-06T15:35:43 < PaulFertser> Works but requires additional libraries. 2019-10-06T15:36:18 < PaulFertser> I'm still trying... 2019-10-06T15:39:37 < englishman> dongs, why and how would you pair bt headphones to a car 2019-10-06T15:40:13 < dongs> my shitty nissan paired TO headphones as audio source for the voice control thing 2019-10-06T15:40:16 < dongs> idk why 2019-10-06T15:40:27 < dongs> when you press a button and say shit like "dial nigger" 2019-10-06T15:40:34 < dongs> i only used it like once to test and never again 2019-10-06T15:40:52 < englishman> lol I have never used my Nissan's voice shit the Bluetooth barely works anyway 2019-10-06T15:40:58 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-06T15:41:17 -!- XTL [xtl@kapsi.fi] has quit [Ping timeout: 240 seconds] 2019-10-06T15:41:47 < englishman> but p funny that you can pair them up just for that 2019-10-06T15:43:40 < dongs> oh fucking god 2019-10-06T15:43:46 < dongs> im trying to find this bug right 2019-10-06T15:43:51 < dongs> these fucking CLOWNS 2019-10-06T15:43:58 < dongs> #define info() some niggershit for logging 2019-10-06T15:44:02 < dongs> and then 2019-10-06T15:44:07 < dongs> struct zc_info *info; 2019-10-06T15:44:15 < dongs> GUESS WHAT THE PROPER COMPILER DOES 2019-10-06T15:44:33 < dongs> fuck me, opensores niggers lever learn 2019-10-06T15:44:35 < dongs> never 2019-10-06T15:44:45 < mawk> that's just how defines work 2019-10-06T15:44:51 < mawk> it's on windows too 2019-10-06T15:44:53 -!- XTL [xtl@kapsi.fi] has joined ##stm32 2019-10-06T15:44:56 < dongs> uh 2019-10-06T15:45:01 < dongs> well it BUILDS on lunix/gcc 2019-10-06T15:45:02 < dongs> so 2019-10-06T15:45:05 < dongs> GUESS NOT 2019-10-06T15:45:20 < mawk> put the define in uppercase 2019-10-06T15:45:29 < mawk> that's what people do to avoid this 2019-10-06T15:45:33 < dongs> no shit 2019-10-06T15:45:42 < mawk> or push pop kind of stuff 2019-10-06T15:45:42 < dongs> but im not gonna chnge 10+ files vs 1 file 2019-10-06T15:45:55 < mawk> a little sed command 2019-10-06T15:45:59 < mawk> thanks to linux 2019-10-06T15:46:11 < PaulFertser> dongs: oh, and it also uses mmap... 2019-10-06T15:46:53 < dongs> oh, rip then 2019-10-06T15:46:58 < PaulFertser> dongs: I think the best bet is to just compile it on windows with cygwin 2019-10-06T15:47:25 < mawk> windows has a mmap equivalent no ? 2019-10-06T15:47:28 < PaulFertser> Native compilation (of the whole suit including the binaries that are calling library for cracking) doesn't seem to be possible. 2019-10-06T15:47:31 < mawk> you can port that easily 2019-10-06T15:47:47 < mawk> weak mmap() symbol with the windows wrapper 2019-10-06T15:47:54 < mawk> no code to change 2019-10-06T15:48:15 < dongs> absolutely disgusting 2019-10-06T15:48:40 < PaulFertser> dongs: yes, windows is disgusting, why did they decide to not adhere to POSIX, why?! 2019-10-06T15:49:03 < dongs> because posix is retarded shit stuck in 1986 2019-10-06T15:49:13 < dongs> that nobody gives a fuck about in 29019 2019-10-06T15:49:14 < dongs> 2019 2019-10-06T15:49:22 < mawk> some version of windows received a Posix certification 2019-10-06T15:49:24 -!- grindhold [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-06T15:49:28 < mawk> iirc 2019-10-06T15:50:04 < PaulFertser> dongs: if you're using para-virtualisation you should be able to run this bruteforcing app with almost zero overhead in VM. 2019-10-06T15:50:27 < PaulFertser> (if you are not willing to install Cygwin to try it building there) 2019-10-06T15:50:46 < mawk> or WSL ? 2019-10-06T15:51:05 < mawk> that thing works most of the time, impressively 2019-10-06T15:51:09 < PaulFertser> Oh that, yes. 2019-10-06T15:52:33 < BrainDamage> windows takes retrocompatibility to insane levels, it still has code for 16 bit executables and the clock is stuck in localtime in 2019 2019-10-06T15:52:57 < BrainDamage> so whatever poor decision they made years ago, they are stuck with it 2019-10-06T16:02:34 < PaulFertser> Heh, I managed to compile it :) 2019-10-06T16:04:48 < dongs> on what? cygwin? 2019-10-06T16:06:30 < PaulFertser> dongs: nope, natively with mmap wrapper and pthread library 2019-10-06T16:06:36 < PaulFertser> dongs: https://paulfertser.info/yazc.zip 2019-10-06T16:07:30 < PaulFertser> mawk: indeed, got the wrapper from https://github.com/troydhanson/tpl/tree/master/src/win 2019-10-06T16:09:51 < dongs> newp fails 2019-10-06T16:10:06 < Laurenceb> the clock is in localtime in the current year 2019-10-06T16:10:22 < dongs> Error: zc_crk_bforce_init() failed! 2019-10-06T16:11:58 < PaulFertser> Without any clarifications? Too bad :( But it did build and link... 2019-10-06T16:12:06 < dongs> yeah 2019-10-06T16:12:13 < dongs> im looking why would it fail and i dotn get it either 2019-10-06T16:12:18 < dongs> every fail in _init_ should print some other shit 2019-10-06T16:12:49 < PaulFertser> dongs: why do you need this app? 2019-10-06T16:13:22 < dongs> cuz i dont wanna run shitty communist elcomsoft single-threaded zip cracker for days 2019-10-06T16:13:32 < dongs> when i have a 32core threadnigger machine i can use 2019-10-06T16:14:39 < dongs> ohh 2019-10-06T16:14:41 < dongs> i think 2019-10-06T16:14:42 < dongs> err() macro 2019-10-06T16:14:45 < dongs> is off by default or soemthing 2019-10-06T16:14:49 < dongs> or it goes to fucking syslog 2019-10-06T16:15:01 < dongs> thats why i dont see the explanation 2019-10-06T16:15:03 < antto> what kind of nissan? 2019-10-06T16:15:16 < dongs> truck. LE something, i forget 2019-10-06T16:15:25 < englishman> japputruck 2019-10-06T16:15:30 < antto> leaf? ;P~ 2019-10-06T16:15:32 < englishman> not white man truck 2019-10-06T16:15:34 < antto> skyline? 2019-10-06T16:15:40 < dongs> no, a fucking truck you moron 2019-10-06T16:15:44 < dongs> skyline is a fucking coupe or whatever 2019-10-06T16:15:48 < antto> aww 2019-10-06T16:15:52 < englishman> only japs call those little things trucks 2019-10-06T16:15:55 < dongs> frontier 2019-10-06T16:15:58 < dongs> yea that 2019-10-06T16:16:09 < englishman> oh it is a truck 2019-10-06T16:16:11 < BrainDamage> brute forcing zip pw is parallelizable to the extreme, maybe there's a gpu cracker 2019-10-06T16:16:15 < antto> wasn't frontier an opel 2019-10-06T16:16:27 < dongs> BrainDamage: no doubt, but im failing to find anything for windows that actualyl works 2019-10-06T16:16:35 < englishman> antto, I have a leaf 2019-10-06T16:16:38 < dongs> and absolutely parallizeable cuz they all have "start at xx" thing 2019-10-06T16:16:57 < dongs> i mean i could prolly open like 10 elcomsoft pieces of shit and start them all from like different codes but fuck that shit 2019-10-06T16:17:02 < antto> one of these leafs was gonna run over me on a quiet street 2019-10-06T16:17:07 < antto> "came outta nowhere" 2019-10-06T16:17:21 < dongs> http://www.crark.net/ maybe ill tr this 2019-10-06T16:17:36 < dongs> has a fucking windows. exe 2019-10-06T16:17:37 < englishman> surely multithreaded zip cracking is better done in python 2019-10-06T16:17:40 < PaulFertser> dongs: I've rebuilt it with gnu printf and re-uploaded. No idea if that might anyhow affect the result though, doesn't seem to be related. 2019-10-06T16:17:47 < dongs> ok lemme see 2019-10-06T16:17:58 < englishman> antto: get off the road then 2019-10-06T16:18:11 < antto> inposhiburu 2019-10-06T16:18:21 < dongs> nope same shit, and i sdont see a -debuglevel like thing either 2019-10-06T16:18:36 < antto> teh sidewalk iz terriburu 2019-10-06T16:19:13 < dongs> wait, crark only works for fucking .rar 2019-10-06T16:19:35 < PaulFertser> Oh shit, err indeed goes to syslog, that's stupid 2019-10-06T16:19:43 < dongs> PaulFertser: lol, retarded 2019-10-06T16:20:17 < PaulFertser> I'll try to recompile with logging enabled :) 2019-10-06T16:21:53 < karlp> aandrew: another big industry backed rtos, https://www.zephyrproject.org/ 2019-10-06T16:22:16 < karlp> so far I feel like it's ok if you use i2c and supported boards, but everything else is feeling a bit tricky at the moment 2019-10-06T16:24:40 < dongs> arduino photo on front page 2019-10-06T16:24:41 < dongs> closed 2019-10-06T16:26:55 < Thorn> >150+ boards supported 2019-10-06T16:26:57 < Thorn> in their world nobody designs their own hardware? 2019-10-06T16:27:24 < karlp> yeah, it has a much better out of tree board file story than mbed does. 2019-10-06T16:27:33 < karlp> that's why it's getting me trying it 2019-10-06T16:27:49 < karlp> but I'm on teh edge of "it's sunk cost! ignore it!" and still "but I must be close!" 2019-10-06T16:27:51 < BrainDamage> dongs: hashcat uses opencl now, which means it'll use gpu and cpu 2019-10-06T16:28:20 < BrainDamage> use zip2john to extract the hash from the zip file, then run hashcat over it 2019-10-06T16:28:26 < BrainDamage> both should be available in windows 2019-10-06T16:33:02 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Excess Flood] 2019-10-06T16:36:00 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-06T16:41:49 < kakimir32> PaulFertser: any idea why flashing leaves 180bytes hole to end of flash bank? 2019-10-06T16:42:39 < qyx> nah zephyr 2019-10-06T16:42:42 < qyx> much hype 2019-10-06T16:43:09 < Steffanx> your binary = flash size - 180 bytes? kakimir32 :P 2019-10-06T16:43:18 < kakimir32> no 2019-10-06T16:43:37 < kakimir32> I need to check 2019-10-06T16:43:40 < Steffanx> so what is the most hot rtos, qyx? 2019-10-06T16:43:45 < dongs> https://hashcat.net/forum/archive/index.php?thread-8342.html hahahah 2019-10-06T16:44:28 < dongs> fucking opensores 2019-10-06T16:44:31 < dongs> it doesnt work, BrainDamage 2019-10-06T16:44:58 < PaulFertser> dongs: recompiled with logging and reuploaded. 2019-10-06T16:45:08 < dongs> nice 2019-10-06T16:45:09 < dongs> lets try 2019-10-06T16:46:50 < dongs> ok, i guess it just doesnt like that zip file format 2019-10-06T16:47:12 < dongs> libzc: read_crypt_data: fread() read past eof. File corrupted? 2019-10-06T16:47:13 < dongs> libzc: zc_crk_bforce_init: failed to read cipher datae 2019-10-06T16:47:13 < dongs> etc 2019-10-06T16:47:23 < PaulFertser> You have that on GNU/Linux VM compiled the normal way, is it "working" the same? 2019-10-06T16:47:45 < dongs> yeah 2019-10-06T16:47:47 < dongs> oh wait right 2019-10-06T16:47:49 < dongs> it runs on lunix 2019-10-06T16:48:12 < dongs> oh 2019-10-06T16:48:14 < dongs> it uses fopen() 2019-10-06T16:48:18 < dongs> isnt that non-binary by default on windows 2019-10-06T16:48:21 < PaulFertser> kakimir32: not yet, haven't checked it. 2019-10-06T16:48:25 < dongs> it uses "r" 2019-10-06T16:48:26 < dongs> not "rb" 2019-10-06T16:48:30 < PaulFertser> Shit 2019-10-06T16:48:40 < dongs> stream = fopen(file->filename, "r"); 2019-10-06T16:48:41 < Cracki> the merge is march 15, the forum post march 13, so I'd give the pull request the benefit of the doubt 2019-10-06T16:48:42 < kakimir32> do you have any eduncated guess? 2019-10-06T16:48:49 < dongs> there's 2 fopens in entire project 2019-10-06T16:48:58 < Cracki> hahahahaha r not rb lol 2019-10-06T16:48:58 < dongs> one in _file and one in _dict 2019-10-06T16:49:23 < dongs> PaulFertser: i hope youre collecting those fixes and opening a PR to that asshole even tho he doesn't deserve it 2019-10-06T16:50:36 < PaulFertser> dongs: reuploaded with fopen fixed 2019-10-06T16:51:26 < dongs> bingo 2019-10-06T16:51:27 < PaulFertser> kakimir32: please upload a -d3 log of your attempt at flashing the whole flash 2019-10-06T16:51:34 < dongs> now we're talking 2019-10-06T16:51:35 < dongs> 100% cpu 2019-10-06T16:52:08 < PaulFertser> You need to force thread number, I had to disable core number autodetection. 2019-10-06T16:52:17 < dongs> yea, thats a required param i htink anyway 2019-10-06T16:52:24 < dongs> i'm using -t8 or whatever 2019-10-06T16:52:30 < dongs> as test on home desktop 2019-10-06T16:52:32 < dongs> gonna move to threadnigger 2019-10-06T16:53:20 < qyx> Steffanx: idk, I'll wait for karlp's result 2019-10-06T16:53:34 < dongs> thanks, this is pretty great 2019-10-06T16:54:00 < dongs> way faster than elcomshit 2019-10-06T16:54:54 < PaulFertser> Wouldn't have been possible without having the source. 2019-10-06T16:55:05 < dongs> if there was no source, it would just fuckin work 2019-10-06T16:55:14 < dongs> cuz it would be a one-click .exe that simply did the job :) 2019-10-06T16:55:56 < karlp> hahaha 2019-10-06T16:56:07 * karlp laughs at qyx. or cries. not sure. 2019-10-06T16:56:23 < dongs> karlp: did yuo update usb-connector zypsnip 2019-10-06T16:56:42 < karlp> did you check? 2019-10-06T16:56:46 < PaulFertser> dongs: it is already a one-click executable file, it just runs on standard OSes, not windows, heh. 2019-10-06T16:56:49 < dongs> oo, 2 days ago 2019-10-06T16:56:57 < karlp> I'll do more tidyup there one day and try and make it a bit cleaner 2019-10-06T16:57:04 < karlp> but got the shit for people who can read at least now 2019-10-06T16:57:23 < karlp> did you hear I got inside non website info from ampehnol on some of their product codes? 2019-10-06T16:57:34 < dongs> ya? no, waht does it do 2019-10-06T16:57:36 < karlp> they have different types of fuckign nylon in some of their connectors 2019-10-06T16:57:47 < dongs> like, hightemp vs normal or whatever? 2019-10-06T16:57:50 < dongs> thermoplastic 2019-10-06T16:57:54 < karlp> and their website only lists one of them, and digikey carries both 2019-10-06T16:57:58 < dongs> cool 2019-10-06T16:58:02 < qyx> karlp: O_o why 2019-10-06T16:58:09 < karlp> nylon 46 vs nylon 66 2019-10-06T16:58:15 < dongs> amphenol site(s) are a fucking mess anyway 2019-10-06T16:58:23 < dongs> there's different COUNTRY sites and they have shit other sites dont 2019-10-06T16:58:30 < karlp> seems to be "designed for wave soldering" vs "designed for reflow soldering" 2019-10-06T16:58:31 < dongs> like, i remember i had to go to amphenol-canada to find some fucking socket 2019-10-06T16:58:43 < karlp> yeah, it's messy 2019-10-06T16:58:46 < dongs> usa site had no info on it whatsoever 2019-10-06T16:59:03 < karlp> the last page of the footprint sometimes has more ordering code info, like, shit they should just put on the top level product page 2019-10-06T16:59:10 < karlp> not the footprint pdf of a particlar version 2019-10-06T17:00:19 < karlp> qyx: big loss for zephry already is that (at least for st) it's a hal on top of cube LL. 2019-10-06T17:00:32 < karlp> so never going to be all that compact. 2019-10-06T17:01:15 < Steffanx> you have to use the HAL? 2019-10-06T17:01:47 < Steffanx> ll by itself is often not that much more than a wrapper around the registers et all. 2019-10-06T17:02:33 < Laurenceb> https://endchan.net/.media/t_ec32ef23a7a9ecdc33787dc18894412e-imagepng 2019-10-06T17:03:04 < karlp> Steffanx: the zephyr apis use LL internally 2019-10-06T17:06:47 < Steffanx> So you would want them to use the registers directly karlp? 2019-10-06T17:08:51 < qyx> I am also using libopencm3 under my HAL 2019-10-06T17:10:01 < PaulFertser> dongs: please let me know how the cracking went after it's finished. 2019-10-06T17:10:26 < Steffanx> Im not convinced locm3 is doing much bettert, code size wise, than the ll libs. but i might be totally wrong on that. 2019-10-06T17:11:04 < dongs> PaulFertser: running now for a while, i haev no idea on the pasword restrictions tho 2019-10-06T17:11:41 < BrainDamage> you could test it first with a self made zip with a trivial pw such as 'a' 2019-10-06T17:11:49 < dongs> yes, it works for that 2019-10-06T17:15:33 < PaulFertser> dongs: I wrote the author listing all of the issues uncovered so far, if he's interested then I'll send him patches. 2019-10-06T17:16:02 < dongs> cool 2019-10-06T17:16:13 < karlp> Steffanx: I just feel if you're doing a big vendor supported thing you maybe didn't have to start out by importing an existing pile of bugs 2019-10-06T17:16:15 * karlp shrugs 2019-10-06T17:16:42 < karlp> locm3 doesn't really target a real HAL, and yeah, code size isn't anything it's really explicitly targettign either 2019-10-06T17:17:12 < karlp> my only exposure to the LL libs was basically last night finding that they have errors in them for l1. on the first time I tried to use them 2019-10-06T17:17:23 < karlp> so not feelign very excited about them right now 2019-10-06T17:17:53 < srk> karlp: I'm just diffing registers from parsed svd files to see if peripherals are compatible, not complaining if they add anything, ideally if they preserve functionality :) 2019-10-06T17:18:42 < Steffanx> errors like what karlp? 2019-10-06T17:18:47 < Steffanx> i never run into issues with ll, yet. 2019-10-06T17:19:13 < srk> there are errors even in svd files 2019-10-06T17:19:19 < PaulFertser> Thanks mawk for making me search for mmap wrapper. 2019-10-06T17:22:00 < PaulFertser> BrainDamage: I somehow assumed that when one installs a directional TV antenna it's possible to equally easy to point it to any direction, guess I was wrong then. Thanks for the clarification. 2019-10-06T17:24:28 < BrainDamage> it's also a matter of tolerance 2019-10-06T17:24:43 < BrainDamage> high gain means that you can have little tolerance 2019-10-06T17:25:03 < BrainDamage> so the higher the gain, the more accurate your pointing has to be 2019-10-06T17:25:33 < BrainDamage> tv antenna pointing is generally done as 'point in the general direction of the repeater/transmitter without any instrumentation whatsoever' 2019-10-06T17:26:11 < BrainDamage> while eg tv sat installation assumes you'll need precise pointing from the get-go so they'll have instrumentation to measure signal strength, crosspolarization bleeding, etc 2019-10-06T17:28:55 < PaulFertser> BrainDamage: well yes, but if you choose a highly directional (in the horizontal plane) then I'd expect same sensitivity would make sense in the vertical plane. 2019-10-06T17:29:57 < PaulFertser> Here we have DVB-T2 broadcasts everywhere now, and so people are looking at maps to find the nearest transmitter and can probably use magnetic compasses in smartphones for directions. 2019-10-06T17:29:57 < BrainDamage> a standard yagi is decently well selective in horizontal, but /extremely/ selective in the vertical 2019-10-06T17:30:17 < PaulFertser> Aha, that's the point I was missing! Thanks a lot! 2019-10-06T17:32:20 < catphish> gay, my wireless headphones are broken, ebay seller advertised expensive moisture resistant product, sent cheaper non resistant version, they didn't like my sweat 2019-10-06T17:43:25 < Cracki> >buying non-disposable shit on ebay 2019-10-06T17:43:49 < Cracki> made the same mistake with a 10 bucks trackpad replacement. it works, but synaptics control panel won't show up at all 2019-10-06T17:44:15 < Cracki> and the alternative driver (lenovo uses ALPS sometimes too) doesn't even recognize the device's pnp id (LEN0039) 2019-10-06T17:45:40 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 264 seconds] 2019-10-06T17:50:28 < Cracki> again same mistake with a 10 bucks replacement battery for my phone. lasted a year, now at 50% capacity 2019-10-06T17:51:05 -!- kakimir64 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-06T17:51:24 < kakimir64> where to paste bigger texts 2019-10-06T17:51:31 < kakimir64> like megabytes of them 2019-10-06T17:52:08 < PaulFertser> Hm, https://en.wikipedia.org/wiki/Yagi%E2%80%93Uda_antenna says that design "three stacked Yagis fed in phase in order to increase gain in the horizontal direction (by cancelling power radiated toward the ground or sky)" 2019-10-06T17:56:26 < kakimir64> paste.ee wins with 1MB limit 2019-10-06T17:57:00 < kakimir64> PaulFertser: https://paste.ee/p/Er3YM 2019-10-06T17:57:29 < kakimir64> next plan is to write smaller sections of flash and see if every write has that 180byte hole 2019-10-06T17:57:29 < Cracki> megabytes? that's whole books. split up maybe. 2019-10-06T17:58:57 < kakimir64> megabytes ain't nothign nowdays 2019-10-06T18:00:34 < englishman> PaulFertser: neato 2019-10-06T18:00:46 < englishman> plenty of goog results for stacked yagi 2019-10-06T18:00:49 < englishman> https://www.qsl.net/dk7zb/Stacking/stacking.htm 2019-10-06T18:06:52 < kakimir64> indeed. I write 4kB to the chip with program command and downloaded image tells that last 180bytes or so before 4kB boundary is 0x00 2019-10-06T18:07:08 < kakimir64> after that 0xff as erased flash should be 2019-10-06T18:10:22 < PaulFertser> englishman: so stacking should be increasing the gain... 2019-10-06T18:11:35 < englishman> sounds like it yes 2019-10-06T18:11:46 < englishman> what freq was the one you linked earlier 2019-10-06T18:17:52 < PaulFertser> englishman: 470—862 MHz I think 2019-10-06T18:18:43 < BrainDamage> PaulFertser: the link's correct and my memory's wrong, sorry 2019-10-06T18:18:47 < englishman> hmm they seem too close to be 0.5-1 wavelength apart maybe it's a) another effect b) pointless 2019-10-06T18:18:47 < BrainDamage> I checked my notes 2019-10-06T18:19:23 < englishman> does it say anything about phasing the elements 2019-10-06T18:19:45 < englishman> it's some china shit right? 2019-10-06T18:20:30 < BrainDamage> elements so close to each other it means that they are coupled by near-field 2019-10-06T18:20:40 < BrainDamage> so you cannot analyse the elements separately 2019-10-06T18:21:02 < BrainDamage> so you cannot in principle use the '3 yagis' analysis for the v shaped one 2019-10-06T18:21:06 < PaulFertser> englishman: not sure if china or not, I just noticed design like that became popular recently. 2019-10-06T18:22:18 < BrainDamage> unfortunately, the only way to be sure would be to slap this into nec2 2019-10-06T18:22:38 < BrainDamage> because of the near field coupling 2019-10-06T18:25:01 < BrainDamage> consider that at the highest freq involved, c/900MHz / 4 = 8 cm, there's about lambda/4 spacing between the outermosts elements, but that's not true for frequencies below it 2019-10-06T18:26:04 < karlp> Steffanx: https://github.com/zephyrproject-rtos/hal_stm32/pull/22 2019-10-06T18:26:26 < BrainDamage> at 400MHz you could even assimilate the 3 vertical dipoles as a single 'fat' dipole 2019-10-06T18:26:58 < englishman> feko or nothing 2019-10-06T18:27:49 < englishman> yeah sounds to me like each segment will resonate well below lam/4 2019-10-06T18:28:36 < englishman> maybe it is a cheap way to make a square yagi 2019-10-06T18:30:34 < englishman> https://1000kanalov.ru/wa-data/public/shop/products/96/28/2896/images/3420/3420.970.jpg 2019-10-06T18:31:07 < englishman> you can see the dipole feed, maybe there are phasing coaxial elements but it's also weird that they claim such a wide band 2019-10-06T18:32:02 < englishman> hmm it doesn't have the same number of elements? 2019-10-06T18:34:03 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-06T18:36:05 < karlp> "Japanese intelligence officers did not even recognise that Yagi was a Japanese name in this context." 2019-10-06T18:36:12 < karlp> I always assumed it was russian... 2019-10-06T18:37:52 < MrMobius> Baba Yagi? 2019-10-06T18:38:28 < PaulFertser> They claim 20dB https://1000kanalov.ru/product/goldmaster-gm-510/ and another 22dB https://1000kanalov.ru/product/goldmaster-gm-500/ 2019-10-06T18:41:55 < qyx> lol baba jaga 2019-10-06T18:44:05 < PaulFertser> Every russian village has plenty of yagis :) The most common japanese thing in russia I guess. 2019-10-06T18:48:18 < dongs> mr miyagi 2019-10-06T18:54:51 < englishman> is this just for TV or what 2019-10-06T18:55:35 < dongs> looks like it 2019-10-06T18:55:37 < dongs> DVB_T2 = tv 2019-10-06T18:58:09 < BrainDamage> yes, dvb-t is terrestrial tv 2019-10-06T18:58:15 < englishman> ppl still watch tv 2019-10-06T18:58:25 < BrainDamage> here's ubiquious as well, every house's roof is littered with tv antennas 2019-10-06T18:58:47 < englishman> oh I didn't read anything on that page 2019-10-06T18:59:10 < BrainDamage> tv is still plenty popular here 2019-10-06T19:04:17 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-06T19:07:45 < srk> /win 62 2019-10-06T19:08:12 < zyp> I have the impression it's mostly iptv here now 2019-10-06T19:08:50 < zyp> but then again I'm living in a new area where everybody got fiber 2019-10-06T19:23:48 < emeb> sounds like a medical condition 2019-10-06T19:40:04 < kakimir64> how do I debug a debugger? 2019-10-06T19:41:03 < kakimir64> I'm dedicated to waste my time 2019-10-06T19:48:45 < kakimir64> interesting 2019-10-06T19:48:54 < kakimir64> I tried flash write_image 2019-10-06T19:49:46 < kakimir64> https://paste.ee/p/ixFVD it should be filled with 0xAD so now even the begin of the flash has gone into shits 2019-10-06T19:50:14 < kakimir64> flash write_image* instead of program 2019-10-06T19:59:09 < kakimir64> i did read that flash for 2 times and it matches.. that crap is actually written into flash 2019-10-06T19:59:31 < dongs> http://www.everanalog.com/Product/ProductEA3036DetailInfo.aspx retweetomg 2019-10-06T19:59:54 < dongs> https://lcsc.com/product-detail/PMIC-AC-DC-Converters_Everanalog-IC-design-EA3036QDR_C111587.html aww fuk its discontinued 2019-10-06T20:00:31 < dongs> (4ch version was 40c in singles 2019-10-06T20:01:38 < kakimir64> is there mechanism for writing support for chips yourself to jlink dongs? 2019-10-06T20:01:45 < dongs> uh like what 2019-10-06T20:01:47 < dongs> and no there isnt 2019-10-06T20:02:00 < kakimir64> scripts 2019-10-06T20:02:06 < dongs> what is a "chip"? like right now RV64 support is broken in jewlink and you cant fix it until they do 2019-10-06T20:02:16 < kakimir64> oh 2019-10-06T20:02:22 < kakimir64> you just need to sit it 2019-10-06T20:02:27 < dongs> what do you want to "support" 2019-10-06T20:02:46 < kakimir64> I just wonder if you need to sit it if something is fukd 2019-10-06T20:02:57 < dongs> depends on whats fucked 2019-10-06T20:03:34 < qyx> the jlink. 2019-10-06T20:04:30 < kakimir64> jewlink 2019-10-06T20:04:36 < kakimir64> how much it did cost 2019-10-06T20:04:57 < dongs> well im sure you bought a chinaclone 2019-10-06T20:05:08 < kakimir64> yes 2019-10-06T20:05:26 < kakimir64> I have piles of them 2019-10-06T20:05:42 < kakimir64> those do not come without their issues 2019-10-06T20:06:50 < kakimir64> I don't use them with jlink software 2019-10-06T20:10:03 < kakimir64> PaulFertser: https://paste.ee/p/Er3YM first text I linked to you.. search "freed 180" 2019-10-06T20:11:33 < kakimir64> working area size seems to be.. 180. Flash hole size is 180 2019-10-06T20:11:38 < kakimir64> I'm getting there 2019-10-06T20:11:48 < dongs> < kakimir64> I don't use them with jlink software 2019-10-06T20:11:50 < dongs> this is your mistake 2019-10-06T20:11:55 < kakimir64> yes 2019-10-06T20:12:00 < dongs> non-jlink (openocd) just uses it as a dumb bit wiggler 2019-10-06T20:12:02 < dongs> a slow one at that 2019-10-06T20:12:09 < kakimir64> I'm concidering to buy in if I some day need to get something done 2019-10-06T20:13:01 < kakimir64> I have not had performance issues yet dongs but things like uart or tracing do not work 2019-10-06T20:16:48 < PaulFertser> dongs: SWD protocol was designed specifically to provide high speed with "dumb wigglers" (that can queue many exchanges) like that. 2019-10-06T20:18:26 < kakimir64> Debug: 1394 314 target.c:1862 target_alloc_working_area_try(): allocated new working area of 180 bytes at address 0x02000000Debug: 1395 314 target.c:1724 print_wa_layout(): * 0x02000000-0x020000b3 (180 bytes)Debug: 1396 314 target.c:1724 print_wa_layout(): 0x020000b4-0x02002fdf (12076 bytes) 2019-10-06T20:18:47 < kakimir64> what is that 12076 bytes? unused ram? 2019-10-06T20:19:06 < PaulFertser> kakimir64: why is working area so small? 2019-10-06T20:19:29 < kakimir64> 180bytes? 2019-10-06T20:19:47 < PaulFertser> Hm, it's not, you specify enough 2019-10-06T20:21:07 < kakimir64> I have not set any IAP working area stuff 2019-10-06T20:21:20 < PaulFertser> kakimir64: yes, initially it allocates just 180 bytes as it's enough for the first IAP call, then frees. 2019-10-06T20:22:21 < PaulFertser> Then it again allocates 180 for the IAP and 4096 for the data. 2019-10-06T20:22:58 < kakimir64> is that 180 stack? 2019-10-06T20:24:22 < PaulFertser> kakimir64: yes, stack for IAP call. 2019-10-06T20:24:43 < PaulFertser> kakimir64: and the flash driver is writing in 4096 blocks, probably the erase sector size for this target, right? 2019-10-06T20:25:04 < kakimir64> yes and size of foo_4k.bin 2019-10-06T20:25:13 < kakimir64> oh wait 2019-10-06T20:25:28 < kakimir64> I get ahead of things 2019-10-06T20:25:36 -!- renn0xtk9 [~max@2a02:810d:1540:2448:65ff:fa3f:56e0:44c8] has joined ##stm32 2019-10-06T20:25:55 < kakimir64> foo.bin is 64kB 2019-10-06T20:25:58 < kakimir64> 16secotrs 2019-10-06T20:26:01 < kakimir64> sectors 2019-10-06T20:26:24 -!- tomeaton17_ [tomeaton17@unaffiliated/tomeaton17] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-06T20:26:35 < PaulFertser> kakimir64: and jest the very last gets written wrong? 2019-10-06T20:26:49 < kakimir64> last 180bytes 2019-10-06T20:26:53 < kakimir64> indeed 2019-10-06T20:27:09 < kakimir64> same thing with foo_4k.bin last 180bytes 2019-10-06T20:28:43 < kakimir64> is there any interrupts related to iap calls? 2019-10-06T20:29:25 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has joined ##stm32 2019-10-06T20:30:52 < PaulFertser> kakimir64: no, shouldn't be, and since program does "reset init" all the IRQs should be disabled. 2019-10-06T20:31:29 < PaulFertser> kakimir64: and with 8k , same last 180 bytes fail? 2019-10-06T20:31:40 < kakimir64> just a second 2019-10-06T20:32:12 -!- fenugrec [~fenugrec@tsrvpq3101w-lp130-02-76-68-230-159.dsl.bell.ca] has joined ##stm32 2019-10-06T20:34:10 < kakimir64> yes 2019-10-06T20:34:35 < kakimir64> https://paste.ee/p/iITpR 2019-10-06T20:34:40 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-06T20:35:35 < kakimir64> do you want to try odd sizes? 2019-10-06T20:40:23 < srk> karlp: u tried reporting svd bugs to st? 2019-10-06T20:40:44 < srk> wondering if it's worth the shot 2019-10-06T20:49:46 < kakimir64> PaulFertser: 3 parameters passed.. cmd50 uses only 2 Debug: 5953 510 lpc2000.c:864 lpc2000_iap_call(): IAP command = 50 (0x00000000, 0x0000000f, 0x00002ee0, 0x00000000, 0x00000000) completed with result = 00000000 2019-10-06T20:50:08 < kakimir64> start from sector 0x0 and end to 0xF 2019-10-06T20:50:40 < Laurenceb> stm64 2019-10-06T20:56:27 < kakimir64> CMD51: Param3: NULL 2019-10-06T21:09:20 < kakimir64> Param3 is overwritten by the fixed value of 12 MHz, which is the IRC referenceclock used by the flash controller. 2019-10-06T21:13:03 -!- fenugrec [~fenugrec@tsrvpq3101w-lp130-02-76-68-230-159.dsl.bell.ca] has quit [Ping timeout: 250 seconds] 2019-10-06T21:31:34 < srk> haha, another one 2019-10-06T21:31:34 < srk> , rcc_apb1enr1_sp3en :: Bit -- SPI3 clock enable 2019-10-06T21:31:35 < srk> , rcc_apb1enr1_spi2en :: Bit -- SPI2 clock enable 2019-10-06T21:32:57 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 240 seconds] 2019-10-06T21:34:47 < R2COM> if someone considers using github should he download some soft on windows too? 2019-10-06T21:34:52 < R2COM> or just web interface is enough 2019-10-06T21:35:25 < srk> just command line is enough 2019-10-06T21:35:49 < srk> install WSL, install git, learn basics 2019-10-06T21:36:42 < qyx> if you want click-click, maybe try sourcetree 2019-10-06T21:36:50 < qyx> or tortoisegit for '90 experience 2019-10-06T21:37:26 < R2COM> but isnt it possible just to get all done through web interface? 2019-10-06T21:37:35 < qyx> no 2019-10-06T21:37:58 < qyx> iirc you are able to upload files, download, make PRs and stuff 2019-10-06T21:38:20 < srk> you can even edit 2019-10-06T21:38:41 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-06T21:38:41 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-06T21:38:41 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-06T21:38:54 < R2COM> my mains tuff is in visual studio (for micros) 2019-10-06T21:39:01 < R2COM> so i found visualstudio tools for git 2019-10-06T21:39:04 < R2COM> curious to see.. 2019-10-06T21:39:13 < R2COM> seems to be a plugin 2019-10-06T21:39:17 < kakimir64> howe do I generate a file filled with running byte value? 2019-10-06T21:39:32 < R2COM> i wonder if upon commit it would upload in new branch ALL changed files 2019-10-06T21:39:35 < R2COM> i.e. create 100 2019-10-06T21:39:38 < R2COM> % duplicate 2019-10-06T21:39:57 < Steffanx> kakimir64. write a 10 line c app? 2019-10-06T21:40:06 < qyx> or write a 2 line python one 2019-10-06T21:40:10 < kakimir64> that calls linux commands 2019-10-06T21:40:36 < srk> R2COM: no 2019-10-06T21:40:57 < srk> R2COM: it uses diffs and only tracks changes 2019-10-06T21:41:09 < R2COM> lets say in myu project in visual c i made changes to few files, main.c , xxx._it.c, yyy.c and now i wanna commit WHOLE thing not just three changed files 2019-10-06T21:41:11 < R2COM> in other words 2019-10-06T21:41:22 < R2COM> i wanna create a new branch with all stuff replicated including new changes 2019-10-06T21:41:30 < kakimir64> seq command 2019-10-06T21:41:33 < srk> git checkout -b newbranch 2019-10-06T21:41:37 < srk> git add 2019-10-06T21:41:48 < srk> git commit 2019-10-06T21:42:19 < R2COM> srk so what i described above not possible? 2019-10-06T21:42:24 < srk> git commit -a will commit every change, you need to call git add to add new files 2019-10-06T21:42:25 < qyx> it is 2019-10-06T21:42:31 < qyx> you have to learn how git manages things 2019-10-06T21:43:05 < srk> full session might look like 2019-10-06T21:43:10 < srk> git checkout -b git_newb 2019-10-06T21:43:24 < srk> git add newstuff.c 2019-10-06T21:43:32 < srk> git commit -a -m "commit message is for suckers" 2019-10-06T21:43:42 < srk> git push origin git_newb 2019-10-06T21:44:21 < srk> creates new branch, one commit in there and pushes it to origin remote (where you cloned it from) 2019-10-06T21:44:29 < R2COM> so it adds new file newstuff.c, but also it transfers all the changed files to new branch as well 2019-10-06T21:44:30 < R2COM> ? 2019-10-06T21:45:14 < srk> commit -a will include all the changes made to already tracked files 2019-10-06T21:45:19 < srk> in the new commit 2019-10-06T21:45:37 < R2COM> also you said install WSL? why? 2019-10-06T21:45:41 < R2COM> i work only in win10 2019-10-06T21:45:42 < srk> you can also add files (or even parts of them) separately via git add or git add -p 2019-10-06T21:45:56 < srk> well for win10 you can easily install WSL ;) 2019-10-06T21:46:18 < srk> imho better option than cygwin 2019-10-06T21:46:19 < R2COM> but why i need wsl? 2019-10-06T21:46:30 < srk> for bash and git ;) 2019-10-06T21:46:54 < Steffanx> or just google git for windows. 2019-10-06T21:46:59 < Steffanx> and find something like https://git-scm.com/download/win 2019-10-06T21:47:42 < R2COM> just found it actually 2019-10-06T21:47:47 < srk> yeah send us some autodownload thingie 2019-10-06T21:48:10 < Steffanx> poor srk 2019-10-06T21:48:12 < srk> do you know there's a package manager for windows? 2019-10-06T21:48:34 < R2COM> dunno 2019-10-06T21:48:35 < R2COM> maybe 2019-10-06T21:49:04 < Steffanx> ask dongs first what he thinks about it. Then you can decide to get it or not. 2019-10-06T21:49:06 < specing> you can always install GNU+Linux 2019-10-06T21:49:27 < srk> yeah there is, it's called chocolatey :) 2019-10-06T21:49:38 < srk> specing: even from windows app store 2019-10-06T21:49:47 < srk> if you're not a total computer noob 2019-10-06T21:51:40 < Steffanx> Arent well all? 2019-10-06T21:51:43 < Steffanx> *we all 2019-10-06T21:52:56 < srk> especially ST ppl hand managing SVD files 2019-10-06T21:53:10 < srk> if they ever heard of git 2019-10-06T21:53:37 < Steffanx> then they would just ignore your pull requests. 2019-10-06T21:53:44 < srk> I cant even pull request xD 2019-10-06T21:53:52 < srk> can post to community forum apparently 2019-10-06T21:54:09 < Steffanx> i wonder if they would listen to the ST clive 2019-10-06T21:54:17 < R2COM> wait but after downloading and dicking with commands it creates it locally on my disk 2019-10-06T21:54:26 < R2COM> but then i transfer it all to github somehow right? 2019-10-06T21:54:40 < srk> wish there's totally open RISC-V MCU where everything is generated from one repository or so 2019-10-06T21:55:23 < srk> R2COM: git remote add origin git@github.com ... ; git push origin 2019-10-06T21:55:23 < R2COM> so if i got o my existing stm32 MSVC project folder and do "git bash here" 2019-10-06T21:55:28 < R2COM> and 2019-10-06T21:55:29 < R2COM> ah 2019-10-06T21:55:37 < srk> github shows you the steps 2019-10-06T21:55:43 < srk> after you create the repo 2019-10-06T21:55:58 < R2COM> github showed in initial "hello world" tutorial steps through their web window 2019-10-06T21:56:00 < srk> both for creating new repo and adding new remote 2019-10-06T21:56:11 < R2COM> ok 2019-10-06T21:57:41 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-06T21:58:14 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-06T21:59:36 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-06T22:03:51 < Steffanx> BTW srk. Strictly seen SP3EN is right. As that's how it's in the reference manual as well ;) 2019-10-06T22:04:35 < Steffanx> oh but also as SPI3EN -_- 2019-10-06T22:06:52 < Laurenceb> https://www.telegraph.co.uk/education/11187063/Professor-suspended-from-top-university-for-giving-off-negative-vibes.html 2019-10-06T22:08:19 < srk> Steffanx: haha lol, there's a bug in the table but before it's correct xD 2019-10-06T22:08:37 < Steffanx> i just downloaded the latest 433 ref man. its still in there. 2019-10-06T22:08:41 < Steffanx> *L433 that is 2019-10-06T22:08:43 < srk> now I'm wondering where is their source for that 2019-10-06T22:09:20 < srk> this is for G474 2019-10-06T22:09:24 < Steffanx> oh 2019-10-06T22:09:44 < Steffanx> it's spreading. 2019-10-06T22:10:12 < srk> some more https://github.com/HaskellEmbedded/data-stm32/blob/ac747f51bfb945c423ae71c556a5359b22a53938/app/genstm/Coerce.hs#L273 2019-10-06T22:21:21 < srk> ST ppl: lets copypaste L4xx to G4xx, rename, add HRTIM, replace CAN with FDCAN, done 2019-10-06T22:22:40 < Steffanx> hah awesome. 2019-10-06T22:33:13 < R2COM> im having trouble connecting local directory to github account 2019-10-06T22:33:17 < R2COM> i created an rsa key 2019-10-06T22:33:23 < R2COM> and pasted it in github settings 2019-10-06T22:33:27 < R2COM> (create SSH key) 2019-10-06T22:33:34 < R2COM> then you do what? 2019-10-06T22:33:41 < R2COM> ssh -T git@github.com 2019-10-06T22:33:42 < R2COM> ? 2019-10-06T22:42:00 < Laurenceb> haxor the server 2019-10-06T22:48:28 < roomba> Cloning into 'data-stm32'... 2019-10-06T22:48:30 < roomba> remote:Try GitHub with Microsoft Edge now and receive 10 free commit tokens. 2019-10-06T22:48:32 < roomba> Segmentation fault (core dumped) 2019-10-06T22:48:49 < Steffanx> That's not how a BSOD works. 2019-10-06T22:53:06 < englishman> R2COM: if you use Windows like a pro, just use sourcetree 2019-10-06T22:53:16 < Cracki> wat r u doing that causes git to segfault 2019-10-06T22:53:33 < englishman> you should never touch a commandline 2019-10-06T22:54:10 < Cracki> ah, roomba is being a roomba... 2019-10-06T22:54:57 < Cracki> R2COM, command line, $ git clone 2019-10-06T22:55:29 < Cracki> if it's windows, same works too if you have git, OR pick a gui client. my personal favorite is smartgit 2019-10-06T22:56:11 < srk> there are two different urls, one for https and one for ssh 2019-10-06T22:58:24 < Cracki> both work for clone 2019-10-06T22:58:45 < Cracki> if you consider contributing, fork it into your own account first, then push to there and make pull requests *on* github 2019-10-06T23:00:36 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-06T23:01:21 < Cracki> huff some owl https://youtu.be/S8BN6ySgFcU?t=447 2019-10-06T23:02:55 < R2COM> lwell first i created a key 2019-10-06T23:02:58 < R2COM> key.pub 2019-10-06T23:03:13 < R2COM> then i copied its contents into settings--->new SSH key in github settings 2019-10-06T23:03:27 < R2COM> (received email notifying about it as well) 2019-10-06T23:03:30 < R2COM> now 2019-10-06T23:03:33 < R2COM> after its done 2019-10-06T23:03:36 < R2COM> i suppose to do: 2019-10-06T23:03:52 < R2COM> ssh -T git@github.com 2019-10-06T23:04:11 < R2COM> but it returns:L 2019-10-06T23:04:14 < R2COM> git@github.com: Permission denied (publickey). 2019-10-06T23:04:23 < R2COM> what am i missing? 2019-10-06T23:04:36 < qyx> your ssh client is not using the key? 2019-10-06T23:04:36 < Cracki> NO ssh 2019-10-06T23:04:41 < Cracki> git does the ssh 2019-10-06T23:04:45 < Cracki> no explicit ssh 2019-10-06T23:04:47 < qyx> also, why are you complicating you life 2019-10-06T23:04:49 < qyx> your 2019-10-06T23:04:52 < Cracki> he's doing it wrong 2019-10-06T23:04:59 < qyx> just use https 2019-10-06T23:05:02 < qyx> if you get it working 2019-10-06T23:05:06 < qyx> move on to ssh 2019-10-06T23:05:10 < Cracki> that just circumvents his misunderstanding 2019-10-06T23:05:32 < Cracki> you DO NOT do any $ ssh something 2019-10-06T23:05:36 < Cracki> you do $ git clone 2019-10-06T23:06:36 < Cracki> or just get tortoisegit, if you're familiar with tortoiseCVS or maybe even tortoiseSVN 2019-10-06T23:07:09 < R2COM> so what would be the command now to "connect" to my remote? 2019-10-06T23:07:15 < R2COM> git remote add origin ? 2019-10-06T23:07:40 < R2COM> (i see examples where they all say first create folder manually in web interface in github account) 2019-10-06T23:07:47 < R2COM> but i wanna be able to just connect with github account 2019-10-06T23:07:50 < R2COM> and transfer folder 2019-10-06T23:07:54 < R2COM> and have it auytomatically created there 2019-10-06T23:08:16 < qyx> ok, maybe it is the right time to read some git basics tutorial 2019-10-06T23:11:19 < Cracki> you need a git tutorial 2019-10-06T23:11:31 < Cracki> these are beginner questions from absolute misconceptions 2019-10-06T23:11:59 < Cracki> have you ever used svn before, or any other source control stuff? 2019-10-06T23:12:16 < Cracki> there is no "connecting" 2019-10-06T23:12:40 < qyx> you need to understand what cloning means, how to initialize a repo 2019-10-06T23:12:47 < qyx> what is the repo 2019-10-06T23:12:53 < qyx> how does pull/push/merge work 2019-10-06T23:13:10 < qyx> how index works, how to add files for versioning control 2019-10-06T23:13:35 < Cracki> yeah, and maybe stay away from the command line interface 2019-10-06T23:15:20 < englishman> fully lunix'd 2019-10-06T23:15:35 < englishman> why the fuck would you type things into a console in 2019 2019-10-06T23:15:43 < specing> why not 2019-10-06T23:15:58 < specing> it is the most efficient means of doing certain tasks 2019-10-06T23:16:54 < Cracki> rm -rf / 2019-10-06T23:17:31 < MrFahrenheit> --no-preserve-root 2019-10-06T23:18:04 < Cracki> srsly a gui would be easier to deal with for him... but AFTER learning what a revision control system does 2019-10-06T23:19:36 < PaulFertser> Cracki: well, in case of git I'd say the idea of "connecting" repositories makes sense. You have a local repository already, and you want to start pulling and pushing to a remote repository. Isn't that connecting? 2019-10-06T23:19:49 < Cracki> no 2019-10-06T23:20:05 < Cracki> because his desire to "connect" is because that's the first action/verb he thinks this thing supports 2019-10-06T23:20:25 < PaulFertser> But adding a remote does connect local repo to a remote one in a way. 2019-10-06T23:20:29 < Cracki> this thing has entirely unfamiliar actions/verbs and "winging it" isn't gonna be good enough 2019-10-06T23:20:54 < Cracki> please don't explain to him how he can do what he says he does, but what he should want 2019-10-06T23:21:10 < Cracki> adding a fucking remote is WAY TOO ADVANCED an action at this stage 2019-10-06T23:21:24 < Cracki> are you seriously suggesting that's what he needs to be explained now? 2019-10-06T23:21:43 < Cracki> he saw an url beginning with "ssh" and thought he could ssh to it 2019-10-06T23:21:56 < PaulFertser> Eh why? You worked on a project, of course you used a local git repo for it, then you decided it's time to share with the world, so you create a github repo and add a remote, isn't that the way it's usually done? 2019-10-06T23:22:12 < Cracki> look at his situation 2019-10-06T23:22:15 < Cracki> he has no local repo 2019-10-06T23:22:20 < Cracki> he has something remote he wants to clone 2019-10-06T23:22:28 < Cracki> which is why several people tell him to clone the repo 2019-10-06T23:23:09 < Cracki> I don't care what possible scenarios there could be where knowing how to add a remote would be good to know, HIS situation is what's relevant at the moment 2019-10-06T23:23:50 < Cracki> and unless I'm completely mistaken, he is trying to clone a repo from github 2019-10-06T23:23:52 < PaulFertser> TBH I just couldn't parse his sentences. Some folders, some "github account"... I assumed he was doing something reasonable. 2019-10-06T23:24:13 < Cracki> hahaha yeah that's what one might think 2019-10-06T23:24:23 < Cracki> the problem is assumptions 2019-10-06T23:25:11 < PaulFertser> R2COM: is your "local directory" a git repository already, did you ever "git init" in there and added some commits? Then indeed adding a remote is the way to go, then you just "git push" and all your relevant commits are transferred to the server. 2019-10-06T23:25:39 < Cracki> hm, indeed he posed a hypothetical that sounds like he did some local changes and now wants it pushed to github... but I'm not sure if that's hypothetical or real 2019-10-06T23:26:59 < Cracki> oh hell, sorry PaulFertser, I misunderstood the situation. it seems indeed to be how you described... and now I'm horrified that anyone should explain to him how to do this 2019-10-06T23:28:19 < PaulFertser> git init in the directory, then "git add remote origin url" and that's it... 2019-10-06T23:28:36 < Cracki> with url being the url, all else literal 2019-10-06T23:28:39 < PaulFertser> I think github even shows exact commands to do that in their "repo created" page. 2019-10-06T23:36:10 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-06T23:44:19 < karlp> srk: no idea if it's in the svd, not using them, I bothered reporting it to zephyr as it's all st and linaro people so hopefully they can get it fixed in the right places themselves. 2019-10-06T23:45:58 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-06T23:46:14 < karlp> srk: the amount of errors I've seen referred to in srd files implies to me taht they're not the source that st's actualyl using for anything, so I ignore them 2019-10-06T23:58:38 < kakimir64> https://www.youtube.com/watch?v=SiAuAJBZuGs musics then 2019-10-06T23:59:00 < kakimir64> kaki midnight youtube musics --- Day changed Mon Oct 07 2019 2019-10-07T00:02:48 < zyp> karlp, all this «just autogenerate the headers from svds» sounds impractical 2019-10-07T00:03:24 < zyp> that means you've got source material with errors, and you can't easily get fixes back in upstream so you get to maintain your own fork 2019-10-07T00:03:40 < karlp> yeah exactly, I don't buy it as a good idea 2019-10-07T00:04:02 < zyp> might as well just maintain the headers directly, that way you don't have to worry about shit that won't translate well from svd to usable header 2019-10-07T00:04:04 < karlp> until I can submit changes or at least see the code taht st's obviously using to generate shit, I'm not trying to fix any of it. 2019-10-07T00:04:23 < karlp> so that's locm3 then, all oru own headers, no attempt to reuse vendor stuff, but lots of work upfront. 2019-10-07T00:04:35 < karlp> but also means no-one else did any implemenations on that we could have used. 2019-10-07T00:04:42 < zyp> also laks 2019-10-07T00:05:03 < karlp> indeed. 2019-10-07T00:05:15 < karlp> laks has a lot of appeal right now. 2019-10-07T00:05:18 < zyp> but yeah, I guess some community-maintained svd forks could take care of the upstreaming issue 2019-10-07T00:09:26 < zyp> speaking of code generation, tomorrow I'm gonna write some shit to fill some C structs from yaml data 2019-10-07T00:14:54 < karlp> fatal error, error code 0. 2019-10-07T00:15:00 < karlp> like, wtf is that meant to mean. 2019-10-07T00:15:37 < karlp> Generic CPU exception, not covered by other codes. nice. 2019-10-07T00:15:58 < zyp> fun 2019-10-07T00:16:28 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-07T00:17:19 < karlp> I'm more and more inclined that it' sjust broken shit in zephyr. 2019-10-07T00:17:41 < karlp> should hve a) stuck with 1.14 lts, instead of 2.0, and b) shoudl have got all periphs working on existing boards before adding l1 that wasn't supported at all. 2019-10-07T00:17:50 < karlp> fuck it. beer thyme instead 2019-10-07T00:18:16 < karlp> it's also fun to note taht the zephy channel is _completely_ silent on the weekends outside one or two newcomers (like me) asking for help and getting crickets. 2019-10-07T00:39:54 -!- sterna1 [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-07T00:44:21 < kakimir64> https://community.arm.com/developer/tools-software/tools/f/keil-forum/34284/lpc1517-swd-flash-programming-failure it's not always roses in keil land 2019-10-07T00:47:28 < Cracki> says to update his 4.7 to 4.74 2019-10-07T00:47:39 < Cracki> outrageous 2019-10-07T00:49:23 < Steffanx> karlp: some errors in the svds do show up in the datasheet/ref manual. 2019-10-07T00:49:30 < Steffanx> Like the SP3EN typo 2019-10-07T00:50:43 < Cracki> lpc15xx user manual, first revision early 2014, forum post early 2015, mdk 4.70 was released in early 2013, so that's 2-year-old software 2019-10-07T00:51:50 < Cracki> 4.74 is from april 2014, nearly synchronous with lpc15xx's first manual 2019-10-07T00:54:17 < kakimir64> what version of jlink is for pros? 2019-10-07T00:54:21 < kakimir64> dongs? 2019-10-07T00:54:47 < Thorn> https://hackaday.io/project/164932-axiom-100kw-motor-controller 2019-10-07T00:55:14 < Thorn> opentesla soon 2019-10-07T00:55:50 < kakimir64> is there 100kw tesla model? 2019-10-07T00:56:11 < kakimir64> ofc multiple inverter/motor solution 2019-10-07T00:56:19 < specing> you'll need a bit more than 100kW for an opentesla 2019-10-07T00:58:22 < kakimir64> 100kw is more in nissan leaf category 2019-10-07T00:58:43 < specing> its non-tesla territory ;p 2019-10-07T01:02:14 < Laurenceb> Musk is an asshole tbh 2019-10-07T01:02:28 < Laurenceb> like a Victorian businessman 2019-10-07T01:02:59 < Laurenceb> working for him is probably worse than babbyshake 2019-10-07T01:04:56 < specing> yeah 2019-10-07T01:05:07 < specing> can't deny that he gets things done, though 2019-10-07T01:05:32 < Laurenceb> xactly 2019-10-07T01:06:29 < Laurenceb> I'm surprised nobody has been killed in his factories, working conditions look pretty poor 2019-10-07T01:06:45 < Laurenceb> maybe he just pays out to keep people quite if they are injured 2019-10-07T01:17:38 -!- renn0xtk9 [~max@2a02:810d:1540:2448:65ff:fa3f:56e0:44c8] has quit [Quit: Konversation terminated!] 2019-10-07T01:20:50 < Cracki> 100 kw on each wheel, then you're going somewhere 2019-10-07T01:21:18 < Cracki> 100 kw is 136 horses, still quite decent 2019-10-07T01:33:32 < kakimir64> it's nissan leaf 2019-10-07T01:35:28 < BrainDamage> 100kW divided on all wheels and it's still better than most commuter vehicles 2019-10-07T01:36:03 < Cracki> and put this in an "e-bike" and it'll be overkill 2019-10-07T01:37:18 < Thorn> it's 10kW for 1 motor, so you'd need a controller for each drive wheel anyway 2019-10-07T01:38:34 < Thorn> *100kW 2019-10-07T01:39:38 < Cracki> some EV designs have one motor for two wheels, and 2WD overall 2019-10-07T01:40:27 < BrainDamage> at some point differential gears is a source of problems in the control loop vs individual direct drive wheels 2019-10-07T01:40:40 < Cracki> the vesc guy already has driver board for almost comparable power: https://vesc-project.com/Hardware 2019-10-07T01:40:52 < BrainDamage> the elasticity introduces a delay and a complex conugate poles which limit your control margin 2019-10-07T01:41:04 < Laurenceb> ur mums dildo 2019-10-07T01:41:38 < BrainDamage> vs just throwing more power ™ at it 2019-10-07T01:41:46 < Cracki> not saying it's smart to just take a gas car and swap the motor 2019-10-07T01:41:59 < Cracki> lots of gain from throwing out all the gears 2019-10-07T01:43:13 < BrainDamage> the main reason to go on a single motor is economics, rather than quality 2019-10-07T01:44:06 < Cracki> economics meaning it's cheaper than two/four individual motors? because I don't see that taking less power 2019-10-07T01:44:29 < BrainDamage> it's cheaper to have a single power train source 2019-10-07T01:45:00 < BrainDamage> not because engines are cheap, but because gearing and power train connection is mass produced 2019-10-07T01:45:48 < BrainDamage> so your solution clashes against an existing market 2019-10-07T01:46:23 < BrainDamage> ... for marginal gains, except for racing performance 2019-10-07T01:54:49 < kakimir32> how to pick a real jlink 2019-10-07T01:55:09 < kakimir32> there is jlinks from multiple vendors in mouser 2019-10-07T01:55:16 < kakimir32> segger, renesas 2019-10-07T01:56:41 < Thorn> https://hackaday.io/project/164932-axiom-100kw-motor-controller/log/168791-making-hardware-the-hard-way 2019-10-07T01:59:04 < kakimir32> argentina 2019-10-07T01:59:28 < Laurenceb> argenitalia 2019-10-07T02:00:04 < kakimir32> how is uk argentina relations today? 2019-10-07T02:00:34 < Laurenceb> dunno 2019-10-07T02:12:33 < Cracki> kakimir32, segger makes jlink, only segger 2019-10-07T02:12:46 < Cracki> everything else is maybe some custom dongle with a jlink OB firmware in it 2019-10-07T02:13:25 < kakimir32> so which model to get? 2019-10-07T02:13:32 < kakimir32> money ain't really a problem 2019-10-07T02:13:39 < Cracki> ¯\_(ツ)_/¯ segger.com pick one 2019-10-07T02:13:50 < Cracki> I do wonder why the fuck mouser lists "jlink" by renesas and others 2019-10-07T02:13:54 < kakimir32> wait 2019-10-07T02:14:01 < Cracki> it's prolly because they're resellers 2019-10-07T02:14:03 < kakimir32> should I get keil 2019-10-07T02:14:12 < Cracki> jlinks work with anything 2019-10-07T02:14:27 < Cracki> if you want something that ONLY works with keil, that's an "ulink" 2019-10-07T02:17:58 < kakimir32> jlink works with keil? 2019-10-07T02:18:19 < Cracki> yes 2019-10-07T02:18:23 < Cracki> everything works with jlink 2019-10-07T02:21:32 < Cracki> seger made their jlink OB firmware to run even on stlink and lpc link/xpresso 2019-10-07T02:22:23 < Cracki> feature/price comparison https://www.segger.com/products/debug-probes/j-link/models/model-overview/ 2019-10-07T02:22:32 < Cracki> (oh, no price) 2019-10-07T02:28:14 < Cracki> jlink v8 (clones) exist on aliex if you wanna be cheap about it 2019-10-07T02:32:24 < kakimir32> I have a pile of them 2019-10-07T02:33:16 < kakimir32> ultra+ would be 1000eur with vat 2019-10-07T02:36:18 < kakimir32> but seriously at the moment even EDU would do 2019-10-07T02:36:41 < kakimir32> it just costs nothing 2019-10-07T02:36:59 < kakimir32> if I need more features I can go crazy then 2019-10-07T02:40:07 < kakimir32> while I'm at it.. what did I need from mouser? 2019-10-07T02:44:05 < catphish> i tested my usb throughput, without double buffering it maxes out at 800,000 B/s, that's 12.5 x 64 byte packets per 1ms frame 2019-10-07T02:44:21 < catphish> i'm having trouble finding out what the theoretical maximum is 2019-10-07T02:46:02 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 240 seconds] 2019-10-07T02:46:51 < Laurenceb> catphish: interesting 2019-10-07T02:46:55 < Laurenceb> what usb class? 2019-10-07T02:47:28 < catphish> just a totally bare endpoint, class is set to "vendor defined" 2019-10-07T02:47:35 < Laurenceb> ah 2019-10-07T02:47:40 < Laurenceb> seems quite slow 2019-10-07T02:47:45 < catphish> literally just blasting bits at fast as possible 2019-10-07T02:47:57 < Laurenceb> to windozer? 2019-10-07T02:48:03 < catphish> lunicks 2019-10-07T02:48:09 < Laurenceb> hmm 2019-10-07T02:48:17 < catphish> i'm having trouble working out what the theoretical maximum is though 2019-10-07T02:48:18 < Laurenceb> libusb or something? 2019-10-07T02:48:25 < catphish> yeah linbusb 2019-10-07T02:48:29 < catphish> *libusb 2019-10-07T02:48:38 < Laurenceb> well I get 925kB/s to mass storage class using F103 2019-10-07T02:49:10 < Laurenceb> so that seems a bit slow 2019-10-07T02:49:10 < catphish> that's also interesting 2019-10-07T02:49:30 < catphish> not dramatically more, but my application is so intentioanlly dumb it should be maxing out the bus 2019-10-07T02:49:35 < Laurenceb> are you just using control packets? 2019-10-07T02:49:51 < catphish> bulk data packets 2019-10-07T02:50:02 < Laurenceb> ah ok, same as mass storage 2019-10-07T02:50:13 < Laurenceb> whats the cpu clock speed? 2019-10-07T02:50:37 < catphish> 80MHz 2019-10-07T02:50:57 < Laurenceb> ah plenty fast, F103 was at 72MHz 2019-10-07T02:51:34 < R2COM> so which debug device i should update to 2019-10-07T02:51:49 < R2COM> is jlink really a thing? 2019-10-07T02:51:59 < catphish> i suspect the problem is that the pc is having to wait between each packet while i populate the buffer, i don't know how often it retries 2019-10-07T02:52:36 < catphish> i'll try to implement double buffering now and see if that improves it or not 2019-10-07T02:52:39 < Laurenceb> ok 2019-10-07T02:52:46 < catphish> zyp had some compelling evidence that it shouldn't 2019-10-07T02:52:50 < Laurenceb> yeah my mass storage was double buffered 2019-10-07T02:53:05 < Laurenceb> but I never tried it without double buffering 2019-10-07T02:53:40 < catphish> sadly i have no way to observe the bus directly 2019-10-07T02:53:57 < Laurenceb> there are quite useful tools built into lunix 2019-10-07T02:54:13 < Laurenceb> usbmon and the various gui front ends 2019-10-07T02:54:29 < dongs> found what Laurenceb has been busy with 2019-10-07T02:54:31 < dongs> https://www.kickstarter.com/projects/twostepsahead/this-girl-a-random-generator-for-beautiful-models 2019-10-07T02:54:40 < Laurenceb> fapfapfap 2019-10-07T02:54:51 < Laurenceb> actually I was wurking on muh space runway 2019-10-07T02:59:29 < kakimir32> jlink ordered 2019-10-07T03:00:09 -!- codyps [~codyps@richard.einic.org] has quit [Remote host closed the connection] 2019-10-07T03:00:14 < kakimir32> robot sex when 2019-10-07T03:00:19 < karlp> you could also just stop using lpc shit dude 2019-10-07T03:00:25 < dongs> lol lpc 2019-10-07T03:00:34 < dongs> doesn't that shit require some fucking flash checksum 2019-10-07T03:00:37 < dongs> before it will boot your code 2019-10-07T03:00:40 < kakimir32> karlp: shut up 2019-10-07T03:00:45 < dongs> i would never use garbageware like this 2019-10-07T03:00:50 < karlp> that's why he's trying to buy a segger, hoping it will fix the flashing 2019-10-07T03:01:02 < karlp> other people have amanged with openocd, kaks is having issues, 2019-10-07T03:01:10 < kakimir32> it needs ISR checksum dongs and openocd did that automatically if not correctly set 2019-10-07T03:01:24 < dongs> and? 2019-10-07T03:01:29 < dongs> so why isnt it working for you 2019-10-07T03:01:46 < karlp> he's still never really justified using lpc honestly :) 2019-10-07T03:01:52 < karlp> other htan finnish masochism 2019-10-07T03:02:23 < kakimir32> dongs: why do you expect it's that checksum? 2019-10-07T03:02:24 < Thorn> I've worked with lpc successfully using jchink 2019-10-07T03:02:41 < kakimir32> Thorn: which models? 2019-10-07T03:02:54 < Thorn> lpc824 2019-10-07T03:02:55 < dongs> yeah i mean, kikemir cant be possibly using anything super modern LPC that wouldnt be supported by even ancient versions of segger shit 2019-10-07T03:02:59 < dongs> is it some dualcore shit? 2019-10-07T03:03:30 < kakimir32> I has flash with 0x00 hole 2019-10-07T03:03:40 < kakimir32> exactly size of iap working memory 2019-10-07T03:03:58 < kakimir32> other chip variant with more flash and usb works with openocd 2019-10-07T03:05:45 < kakimir32> 180bytes hole left after last lpc2000_write_flash() of the whole write operation 2019-10-07T03:07:57 < karlp> fucking zephyr talks abotu "topologies supported" but gives no guidance on how to actually set up the topologies supported. https://docs.zephyrproject.org/latest/guides/west/repo-tool.html#topologies-supported 2019-10-07T03:08:15 < dongs> you still trying to avoid mbed? 2019-10-07T03:08:25 < R2COM> oh shit im still trying to catchup with this new git shit 2019-10-07T03:08:26 < karlp> nah, still trying to avoid laks+freertos 2019-10-07T03:08:34 < dongs> The following three source code topologies supported by west: 2019-10-07T03:08:34 < karlp> or freertos+locm3 2019-10-07T03:08:36 < dongs> west?????????????? 2019-10-07T03:08:38 < kakimir32> Thorn: popular chip 2019-10-07T03:08:41 < R2COM> whatever described in tutorials and their outputs completely different from what i get 2019-10-07T03:08:54 < R2COM> like if i do git pull it gives: 2019-10-07T03:08:56 < R2COM> fatal: refusing to merge unrelated histories 2019-10-07T03:09:09 < karlp> mbed got written off quick, supporting out of tree boards is ~vapour ware. 2019-10-07T03:09:17 < R2COM> and if after remote connecting i push i get: 2019-10-07T03:09:20 < R2COM> Updates were rejected because the tip of your current branch is behind 2019-10-07T03:09:20 < R2COM> hint: its remote counterpart. Integrate the remote changes (e.g. 2019-10-07T03:09:20 < R2COM> hint: 'git pull 2019-10-07T03:09:29 < karlp> git pull --rebase might be enough 2019-10-07T03:10:40 < R2COM> this git command source control shit already starts looking like lunix shit 2019-10-07T03:10:49 < Cracki> >stretch goals 2019-10-07T03:11:11 < karlp> I don't know hwere you started or with what in the begnning commie, *shrugs* 2019-10-07T03:11:18 < R2COM> and if you pick any tutorial and follow it precisely your output is fucken complete different from what it looks there 2019-10-07T03:11:26 < R2COM> well started easily 2019-10-07T03:11:27 < R2COM> here 2019-10-07T03:11:32 < R2COM> created simple folder with few files 2019-10-07T03:11:34 < R2COM> then: 2019-10-07T03:11:42 < R2COM> git init 2019-10-07T03:11:42 < R2COM> git add . 2019-10-07T03:11:42 < R2COM> git commit -m "commit #1" 2019-10-07T03:11:44 < R2COM> ==== 2019-10-07T03:11:57 < karlp> ok.... so what's the problem? 2019-10-07T03:12:04 < R2COM> and now, at this point, all i wanna know is: "how can i transfer this into my github account" 2019-10-07T03:12:08 < karlp> (apart from the shitty commit message) 2019-10-07T03:12:08 < R2COM> ^ 2019-10-07T03:12:21 < karlp> go to github, press "create new repo" follow thhe fucking instructions 2019-10-07T03:12:27 < R2COM> but 2019-10-07T03:12:30 < R2COM> i wanted to ask is 2019-10-07T03:12:44 < R2COM> is there a way to do it without going to github and manually creating repo? 2019-10-07T03:12:50 < R2COM> or one does have to do it manually? 2019-10-07T03:13:54 < R2COM> so the mechanizm is: 1) you do what i described in above commands, 2) you go to github.com and create repo manually with SAME name as your folder you created here, 3) connect it together and after manipulate with commands from scm bash console? 2019-10-07T03:13:55 < Cracki> you haven't googled your error messages, have you 2019-10-07T03:14:18 < karlp> name doesn't matter at all. 2019-10-07T03:14:20 < Cracki> they mean that you didn't follow some tutorial exactly, but winged it 2019-10-07T03:14:25 < R2COM> so lets say i did step #2, now i have a problem... 2019-10-07T03:14:38 < R2COM> it does not transfer my files from local folder to online account 2019-10-07T03:14:41 < Cracki> they mean that you tried to mate two repos that don't contain the same history 2019-10-07T03:15:03 < karlp> https://imgur.com/a/jCklh10 2019-10-07T03:15:15 < karlp> you already did the git init; git add, blah lcoally you now have a repo 2019-10-07T03:15:23 < karlp> the "name" of the repo is just where github will present it. 2019-10-07T03:15:25 < karlp> nothing more 2019-10-07T03:15:41 < R2COM> k trying.. 2019-10-07T03:15:45 < dongs> fucking hell 2019-10-07T03:15:51 < Cracki> try #git 2019-10-07T03:15:52 < dongs> this retard wants a 30mhz osc 2019-10-07T03:15:54 < dongs> no local stock 2019-10-07T03:15:56 < dongs> of course 2019-10-07T03:16:02 < Cracki> thirty? what wants that 2019-10-07T03:16:08 < dongs> some silabs RF trash 2019-10-07T03:16:13 < Cracki> uh 2019-10-07T03:16:25 < Cracki> wat for, some nice round 900 MHz or other? 2019-10-07T03:16:34 < dongs> well tbh last i looked at datasheet it wanted anything around 24-32mhz or someshit 2019-10-07T03:16:42 < R2COM> ok it doesnt work 2019-10-07T03:16:44 < R2COM> here: 2019-10-07T03:16:45 < R2COM> ! [rejected] master -> master (non-fast-forward) 2019-10-07T03:17:05 < dongs> R2COM: when git doesnt work i just shift-del entire pr oject dir and start over 2019-10-07T03:17:07 < R2COM> happened after: git push -u origin master 2019-10-07T03:17:18 < dongs> also holy shit just install sourcetree 2019-10-07T03:17:22 < dongs> stop fucking with git on command line 2019-10-07T03:17:29 < karlp> when you created teh repo on github, ther was a tickbox "initialize this with a READMe? (skip this if you have an existing repo) 2019-10-07T03:17:33 < karlp> you must have ticked it 2019-10-07T03:17:35 < Cracki> 30... that could be 8 *15/4 or 24 *5/4 2019-10-07T03:17:39 < karlp> so git pull --rebase should be ok. 2019-10-07T03:17:39 < R2COM> umm 2019-10-07T03:17:43 < R2COM> lemme try again 2019-10-07T03:17:45 < dongs> AIS transponders and receivers use two VHF radio frequencies: 161.975 MHz (AIS1, or channel 87B) and 162.025 MHz (AIS2, or channel 88B). 2019-10-07T03:17:48 < dongs> ^ Cracki 2019-10-07T03:17:56 < karlp> it's saying you have commits on the "remote" (github) and you don't have them locally, 2019-10-07T03:18:04 < karlp> unlees you're just trolling... 2019-10-07T03:18:04 < catphish> R2COM: that means you are trying to push a commit that is not a descendant of the current remote commit 2019-10-07T03:18:05 < Cracki> those are some nice round frequencies 2019-10-07T03:18:26 < Cracki> the remote repo must have NO commits, not even an initial commit 2019-10-07T03:18:28 < catphish> R2COM: usually happens if you do something weird like push, then change your last commit, and try to push again 2019-10-07T03:18:42 < Cracki> when in doubt, push --force 2019-10-07T03:18:44 < catphish> or try to push when someone else has pushed in the meantime 2019-10-07T03:18:52 < karlp> that's the worst advice ever dude :) 2019-10-07T03:19:00 < Cracki> be wary of force pushing. people will behead you if you fuck up their repo 2019-10-07T03:19:01 < R2COM> git pull --rebase gave me this: 2019-10-07T03:19:02 < dongs> when in doubt, delete git folder 2019-10-07T03:19:11 < Cracki> don't just run commands without understanding 2019-10-07T03:19:16 < kakimir32> you cannot always have convinience of the most popular chip type and variation.. and I want to get to writing application. That's why jlink dongs 2019-10-07T03:19:18 < Cracki> rebase has a specific meaning and does a specific thing 2019-10-07T03:19:25 < karlp> that's the complete opposite of what you just said "when in doutb, -f" 2019-10-07T03:19:26 < catphish> git pull --rebase is usually the solution if someone has pushed in between 2019-10-07T03:19:39 < R2COM> https://justpaste.it/3l2n9 2019-10-07T03:19:42 < Cracki> and that doesn't mean "magically fix it", it means juggling changes around 2019-10-07T03:19:46 < kakimir32> I might even go to even more obscure mcus 2019-10-07T03:19:59 < R2COM> here is output of git pull --rebase 2019-10-07T03:20:03 < R2COM> not sure if its a success 2019-10-07T03:20:12 < catphish> R2COM: you don't have your master branch linked to the remote master 2019-10-07T03:20:28 < karlp> what exactly did you do with the github repo after you made it? 2019-10-07T03:20:35 < catphish> you've been pushing it manually with git push origin master 2019-10-07T03:20:35 < karlp> you can do what they're telling you to link upstream 2019-10-07T03:21:07 < R2COM> nothing its sitting there 2019-10-07T03:21:09 < catphish> first things first, link your master to the remote master: git push -u origin master 2019-10-07T03:21:13 < R2COM> with default readme file 2019-10-07T03:21:22 < catphish> the -u tells it to link those 2 branched together "tracking" 2019-10-07T03:21:26 < Cracki> just try your luck in #git 2019-10-07T03:21:31 < dongs> R2COM: did you download sourcetree already 2019-10-07T03:21:46 < catphish> then it knows that remote changes to master should be merged back into your master 2019-10-07T03:21:55 < Cracki> >default readme 2019-10-07T03:21:57 < Cracki> *facepalm* 2019-10-07T03:22:03 < Cracki> that means there's a commit 2019-10-07T03:22:11 < Cracki> just force push over it this one time 2019-10-07T03:22:14 < catphish> ooh 2019-10-07T03:22:21 < R2COM> dongs downloading, but in meantime wanna see wtf is wrong with this stuff 2019-10-07T03:22:30 < catphish> yeah, your commit isn't a descendant of the commit that made the readme! 2019-10-07T03:22:41 < dongs> jsut give up, git was written by lunix toreballs, NOBODY knows what the fuck is going on with it 2019-10-07T03:23:01 < R2COM> wait...sourcetree works with what system?? 2019-10-07T03:23:12 < catphish> R2COM: essentially you have totally separate repos, the one with the readme in, and yours 2019-10-07T03:23:23 < Cracki> git starts making sense when you graduate from the command line. the command line doesn't show you that it's all "just" trees and operating on them 2019-10-07T03:23:25 < dongs> R2COM: ? windows? 2019-10-07T03:23:27 < R2COM> catphish yes, 2019-10-07T03:23:28 < dongs> what "system"? 2019-10-07T03:23:35 < Cracki> c:\windows\system32 2019-10-07T03:23:38 < dongs> is your code on shithub? or you want to put it on shithub whatever 2019-10-07T03:23:38 < Cracki> deltree that 2019-10-07T03:23:43 < R2COM> dongs i mean... if i wanna store code online in sourcetree where it holds it? 2019-10-07T03:23:48 < dongs> it doesnt 2019-10-07T03:23:51 < catphish> R2COM: you're trying to push one over the other, this won't work, on this *one* occasion, you should git push -f -u origin master 2019-10-07T03:23:51 < Cracki> sourcetree is a client, like chrome 2019-10-07T03:23:54 < dongs> you dont need git to store anything online 2019-10-07T03:23:56 < Cracki> or firefox 2019-10-07T03:24:06 < catphish> R2COM: that will obliterate the readme and replace the whole repo with yours 2019-10-07T03:24:06 < karlp> that default readme is what I said is what you must have added whhen you _didn't read what it said on screen_ with "skip this if you have an existing repo" 2019-10-07T03:24:07 < dongs> you can store your shit locally and never send it to any server 2019-10-07T03:24:07 < Cracki> except sourcetree is for git repos, and web browsers are for web sites 2019-10-07T03:24:16 < karlp> and it's not even the fucking default, you had to expolicitly check the box to do that. 2019-10-07T03:24:19 < karlp> so... whatevs 2019-10-07T03:24:35 < R2COM> ok! 2019-10-07T03:24:36 < R2COM> git push -f -u origin master 2019-10-07T03:24:38 < R2COM> ^ 2019-10-07T03:24:39 < R2COM> that worked 2019-10-07T03:24:46 < R2COM> now i see my shit uploaded to github 2019-10-07T03:24:52 < catphish> from now on, you just need "git push" and it will happily push :) 2019-10-07T03:25:11 < catphish> it knows which branch to push, and as long as nobody else pushes in the meantime, your commits will be accepted 2019-10-07T03:25:19 < R2COM> so... from what i udnerstand, for very first time one needs to do "git push -f -u origin master" but just 1 time during creation and thats it, correct? 2019-10-07T03:25:20 < karlp> just don't ever go "oh, it didn't work, just needs -f" cos that's a sure fire way to throw away shit 2019-10-07T03:25:24 < catphish> if someone *does* push in the meantime, then you need to pull first to merge 2019-10-07T03:25:25 < karlp> no, you never needed to 2019-10-07T03:25:30 < dongs> Cracki: https://github.com/astuder/dAISy/blob/master/KiCad/schematic.pdf other kikecaders seems to be using 30mhz for this application as well 2019-10-07T03:25:31 < Cracki> remember, THIS ONE TIME only 2019-10-07T03:25:49 < karlp> you deliberately checked a box that added files to your github verison, despite it saying, "skip this if you have a repo already" 2019-10-07T03:25:49 < R2COM> karlp but all other commands failed except this one 2019-10-07T03:25:53 < Cracki> can that thing take non-30 MHz at all? 2019-10-07T03:25:59 < R2COM> ahhhhh 2019-10-07T03:26:01 < R2COM> karlp ok 2019-10-07T03:26:10 < karlp> no, there wewre methods about setting the upstream that we could have done to make pull --rebase work, for instnace 2019-10-07T03:26:13 < R2COM> ill try another test now where i wont check this and try regular ways 2019-10-07T03:26:14 < catphish> R2COM: the only time you should ever use push -f is if you want to obliterate the repo on the server and replace it with your local one 2019-10-07T03:26:19 < karlp> push -f just says, "yar, nvm, jsut trash that shit" 2019-10-07T03:26:29 < catphish> R2COM: there may be other times to do it, but it's very very rarely a good idea 2019-10-07T03:26:47 < R2COM> hmm yea cuz now that default readme file gone 2019-10-07T03:26:53 < catphish> right 2019-10-07T03:27:24 < catphish> and that's fine this time, but if that readme was something you committed from another computer, you'd be unhappy you lost it 2019-10-07T03:27:43 < Cracki> dongs, > The default crystal is 30 MHz, but the circuit is designed to handle any XTAL from 25 to 32 MHz 2019-10-07T03:27:56 < Cracki> > If a crystal different than 30 MHz is used, the POWER_UP API boot command must be modified. 2019-10-07T03:27:58 < R2COM> i have 1 more question 2019-10-07T03:28:04 < catphish> so if you ever get a push rejecected in future, instead you should do a pull first, to merge the remote changes into yours before you push 2019-10-07T03:28:22 < R2COM> so..is it possible with some command to create repo? or creation of repo first time must be done only from github website? 2019-10-07T03:28:23 < Cracki> this feels like explaining how to program a VCR to my parents 2019-10-07T03:28:47 < R2COM> catphish ok noted about pull first 2019-10-07T03:28:52 < Cracki> github has an api, and maybe sourcetree can use *that* to create repos on github 2019-10-07T03:28:57 < catphish> R2COM: on github, you use the web interface to make repos 2019-10-07T03:29:10 < R2COM> ok just that needs web interface, all other stuff from bash commands 2019-10-07T03:29:12 < Cracki> but whatever api github might have is beyond git, and git itself won't be able to create repos 2019-10-07T03:29:13 < catphish> oh yeah there is indeed an API too 2019-10-07T03:29:14 < Ultrasauce> sweating-button-press.jpg [making fun of r2com] [making fun of git ux] 2019-10-07T03:29:16 < Cracki> (on github) 2019-10-07T03:29:56 < kakimir32> I need to get the ultimate debugging tool though 2019-10-07T03:30:10 < catphish> remember that git is distributed, you can push as many copies as you like to different remotes, like your own server, as a backup 2019-10-07T03:30:50 < catphish> and your local copy is also a full backup when you pull, it's not like some version control systems where you only have the current revision 2019-10-07T03:30:57 < karlp> R2COM: "git init" did the repo create 2019-10-07T03:31:18 < karlp> github just needs something too to setup urls for you to be able to push them to and so on. 2019-10-07T03:31:31 < R2COM> got it now 2019-10-07T03:31:40 < catphish> well git init makes an empty repo, github run the same thing at their end 2019-10-07T03:31:43 < karlp> you can _also_ skip doing git init repo, create on github first, and then _clone_ the github one first, and add files that way. 2019-10-07T03:31:57 < R2COM> hmm okay\ 2019-10-07T03:32:01 < catphish> so there's an empty repo in both places, then you can start pushing stuff from one to the other 2019-10-07T03:32:23 < dongs> R2COM: idk i always just make a blank shithub repo then point sorestere to it 2019-10-07T03:32:28 < catphish> yeah it would have been simpler for you just to clone the repo with the readme in it 2019-10-07T03:32:31 < dongs> the adding of readme shit is definitely an extra checkbox you clicked 2019-10-07T03:32:37 < dongs> i've never had any files in NEW stuff that i make 2019-10-07T03:32:54 < catphish> but meh, this way worked too 2019-10-07T03:33:23 < dongs> Initialize this repository with a README 2019-10-07T03:33:24 < dongs> This will let you immediately clone the repository to your computer. 2019-10-07T03:33:24 < dongs> yeah 2019-10-07T03:33:25 < dongs> see? 2019-10-07T03:33:33 < dongs> what you were SUPPOSED to do after checking that is 2019-10-07T03:33:36 < dongs> clone it first 2019-10-07T03:33:38 < dongs> tehn add your sores 2019-10-07T03:33:39 < karlp> and it's not checked by default 2019-10-07T03:33:40 < dongs> then commit 2019-10-07T03:33:41 < dongs> yes, right 2019-10-07T03:33:46 < karlp> and it says, "skip this if you have one already.." 2019-10-07T03:33:53 < karlp> but whatevers, -f for "fixing" :) 2019-10-07T03:33:55 < dongs> R2PRO can't read 2019-10-07T03:34:03 < karlp> hence my repo name if you saw :) 2019-10-07T03:34:45 < dongs> https://twitter.com/flaplette/status/1180531938156830722?s=19 2019-10-07T03:34:47 < dongs> lel 2019-10-07T03:35:11 < Cracki> that's a communist furry 2019-10-07T03:35:20 < dongs> I wouldnt know 2019-10-07T03:35:22 < dongs> im just retweeting 2019-10-07T03:35:23 < Cracki> icon 2019-10-07T03:35:23 < R2COM> yeah i figured it now 2019-10-07T03:35:26 < dongs> i dont follow any of these furfags 2019-10-07T03:35:35 < Cracki> icon says it all 2019-10-07T03:36:09 < karlp> best comment, "I feel this. I wanted to draw yesterday, but my $100 pencil needed to charge. I was recharging my book at the same time." 2019-10-07T03:36:17 * antto farts 2019-10-07T03:36:27 < dongs> lmao fucking applefags 2019-10-07T03:36:51 < R2COM> uh true, but my ipad pro is my main tool 2019-10-07T03:36:56 < antto> do you mean crApple? 2019-10-07T03:36:57 < R2COM> for notes and sketches 2019-10-07T03:37:13 < antto> eye pad 2019-10-07T03:37:15 < R2COM> i have by now over 1500 pages with schematics and other stuff for work etc 2019-10-07T03:37:16 < dongs> no wonder you dont know how to compute 2019-10-07T03:37:17 -!- learningc [~learningc@2001:e68:5869:c400:5c28:70ca:81f3:8407] has joined ##stm32 2019-10-07T03:37:25 < dongs> youre using a fucking apple shitpad 2019-10-07T03:37:34 < Cracki> hormone therapy? https://twitter.com/flaplette/status/1180256562851565568 2019-10-07T03:37:36 < R2COM> no, its better than pencil or pen actually 2019-10-07T03:37:38 < antto> crApple sh*tpad 2019-10-07T03:37:43 < karlp> Cracki: you're getting distracted 2019-10-07T03:38:20 < antto> real h4x0rz sketch things in ascii art 2019-10-07T03:39:06 < Cracki> scratch it into a PCB 2019-10-07T03:39:21 < Cracki> macbook mainboard perhaps 2019-10-07T03:39:49 < R2COM> alright i exercised whole thing from scratch it seems working fine now 2019-10-07T03:40:01 < R2COM> ok now lemme see what that sourcetree is dongs was talking about 2019-10-07T03:41:09 < R2COM> wtf it asks me to register with some bitbnucket 2019-10-07T03:41:15 < dongs> just ignore that 2019-10-07T03:41:25 -!- User_ [~learningc@2001:e68:5869:c400:6d5a:ac0d:7ee:6a01] has joined ##stm32 2019-10-07T03:41:45 < dongs> oh, you might need a fucking atlassianaccount or whatever thing but just sign into that shit with shithub and done 2019-10-07T03:41:48 < R2COM> it requires onetime registration 2019-10-07T03:41:57 < dongs> yeah right, sorry ive been using it for years and forgot 2019-10-07T03:42:19 < antto> iz iz tiem to https://i.imgur.com/2392QEh.jpg 2019-10-07T03:42:57 -!- User_ [~learningc@2001:e68:5869:c400:6d5a:ac0d:7ee:6a01] has quit [Read error: Connection reset by peer] 2019-10-07T03:45:13 -!- learningc [~learningc@2001:e68:5869:c400:5c28:70ca:81f3:8407] has quit [Ping timeout: 250 seconds] 2019-10-07T03:46:50 < R2COM> dongs so i regged there, now i linked my github to it and can view it from sourcetree app 2019-10-07T03:46:58 < R2COM> dongs now what, means i can do all that with gui stuff? 2019-10-07T03:47:34 < antto> https://i.imgur.com/NOxdyic.png ;P~ 2019-10-07T03:48:07 < antto> muh innovation 2019-10-07T03:48:24 < dongs> R2COM: yeah, edit sores, select files to commit, write message, commit, push, etc. 2019-10-07T03:48:28 < dongs> no need to touch cliaids 2019-10-07T03:48:43 < dongs> plus you can see diff visually and fix up whitespace or other shit if needed before committing 2019-10-07T03:48:52 < R2COM> hmmmkay 2019-10-07T03:49:06 < dongs> branching works and looks not bad etc 2019-10-07T03:49:52 < Laurenceb> http://www.stratosolar.com/gravity-energy-storage.html 2019-10-07T03:49:54 < Laurenceb> kek wut 2019-10-07T03:51:40 < R2COM> antto atmega+74hc logic on single pcb, perfect 2019-10-07T03:51:50 < antto> it's not atmega 2019-10-07T03:54:12 < antto> gotta give me a bonus point at least for the STA120DJ 2019-10-07T03:56:53 < Cracki> Laurenceb, wat is that, flat blimps? 2019-10-07T03:57:39 < Cracki> they could hydrolyze the water to make hydrogen. nothing can go wrong. 2019-10-07T03:58:35 < antto> i think it's a perfect scheme, it involves pumping.. water 2019-10-07T03:59:08 < Laurenceb> Cracki: yeah giant blimps 2019-10-07T04:00:21 < antto> is that a huge solar panel up in the stratosphere, with electrical cables going down to urf? 2019-10-07T04:02:20 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 276 seconds] 2019-10-07T04:02:44 < Laurenceb> antto: yeah 2019-10-07T04:02:52 < Laurenceb> it kind of makes sense 2019-10-07T04:02:59 < antto> okay, i'm a bit skeptical 2019-10-07T04:03:04 < Laurenceb> solar panels work crazy well at 20km 2019-10-07T04:03:11 < antto> sure but 2019-10-07T04:03:13 < catphish> i implemented double buffering, increased throughput from 800,000 to 830,000 2019-10-07T04:03:27 < Laurenceb> catphish: interesting 2019-10-07T04:03:42 < Laurenceb> maybe my lunix hdd test utility was off somehow 2019-10-07T04:03:43 < antto> wut'cha buffering? 2019-10-07T04:03:49 < Laurenceb> thats how I got my results 2019-10-07T04:04:11 < catphish> Laurenceb: i'm literally just counting received bytes in the libusb app 2019-10-07T04:04:15 < R2COM> dongs do you use sourcetree with your mcu projects as well, the ones which have 1 folder and all various shit inside including sourcecode? 2019-10-07T04:04:16 < Laurenceb> ok 2019-10-07T04:04:30 < antto> why not put a large fresnel lens instead of those panels 2019-10-07T04:04:48 < Laurenceb> antto: it sort of makes sense, but it needs to be huge, like at least a cubic kilometer 2019-10-07T04:04:57 < antto> focus teh light onto a very efficient(TM) solar panel on earth 2019-10-07T04:04:58 < Laurenceb> antto: because the sun changes position 2019-10-07T04:05:10 < Laurenceb> solar panels are cheap now 2019-10-07T04:05:23 < antto> okay, build 3 big arms to hold the sun steady ;P~ 2019-10-07T04:05:28 < Laurenceb> especially when they are operating at 20km, no clouds, nice low temperatures 2019-10-07T04:05:30 < catphish> i actually only need 500,000 bps for my application, so this whole thing is moot :) 2019-10-07T04:05:49 < catphish> quite clearly with my optimizations, it's theoretically sufficient 2019-10-07T04:05:56 < antto> or create a smol sun here on earth 2019-10-07T04:05:59 < antto> put it in a jar 2019-10-07T04:06:29 < antto> pocket sun (R) 2019-10-07T04:06:51 < Laurenceb> even more deranged version: just stick reflective foil on the top 2019-10-07T04:06:52 < antto> gonna sell like hot cakes in scandinavia 2019-10-07T04:06:56 < Laurenceb> to cool the planet 2019-10-07T04:07:13 < catphish> must now sleep 2019-10-07T04:07:16 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-07T04:12:18 < antto> i should put a wind turbine on muh a$$ right now 2019-10-07T04:12:39 < antto> i drank a lot of milk, i fart like a factory 2019-10-07T04:12:57 < antto> i should also put a metane collector 2019-10-07T04:14:37 < antto> enough internetz for today 2019-10-07T04:14:42 < R2COM> hmm nice 2019-10-07T04:14:54 < R2COM> man this sourcetree looks really neat 2019-10-07T04:18:17 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-07T04:18:31 < dongs> R2COM: yeah of course 2019-10-07T04:18:35 < dongs> i use it for whatever sores 2019-10-07T04:19:30 < Laurenceb> Herpes sores 2019-10-07T04:21:26 < Cracki> can it show side-by-side diffs instead of that +- cancer? 2019-10-07T04:21:34 < Cracki> i.e. old version left, new version right 2019-10-07T04:21:51 < dongs> idk, it color codes -/+ i think, i never looked if tehre was a better mode 2019-10-07T04:22:18 < Cracki> winmerge/meld/smartgit have side by side diff, it's the superior way 2019-10-07T04:22:55 < dongs> welp because the shithead fucked up and didnt order crystals i have to get them from mouser again 2019-10-07T04:23:03 < dongs> any new ST evalboard i should buy 2019-10-07T04:23:03 < Cracki> crap, supports external tools but I want it integrated properly 2019-10-07T04:23:18 < dongs> hmm maybe the new stdick with F7? 2019-10-07T04:23:30 < Cracki> got all the WBsomething and G4? 2019-10-07T04:23:40 < Cracki> got the motor control dev board? 2019-10-07T04:23:51 < dongs> dont need those 2019-10-07T04:23:58 < dongs> is that new 8pin g0 thing available? 2019-10-07T04:24:13 < dongs> > 46 on order 2019-10-07T04:24:14 < dongs> sux 2019-10-07T04:24:33 < dongs> Estimated Ship Date 2019-10-07T04:24:33 < dongs> 4404-Nov-19 2019-10-07T04:24:49 < Cracki> they don't actually make DIPs do they? what I read was some legfree package 2019-10-07T04:25:06 < dongs> its SOP-8 2019-10-07T04:25:18 < Cracki> is that the 4404th of november 2019, or the 19th november of 4404 2019-10-07T04:25:20 < dongs> i mean, one of the packages, im sure you can get it in wlcsp or something 2019-10-07T04:25:26 < dongs> no, 44 pieces on nov4 2019-10-07T04:25:32 < Cracki> sop8 is cute 2019-10-07T04:26:11 < dongs> https://www.mouser.jp/datasheet/2/389/stlink-v3mini-1634136.pdf cute 2019-10-07T04:26:12 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 246 seconds] 2019-10-07T04:26:35 < R2COM> wow 2019-10-07T04:26:37 < R2COM> stlink v3 2019-10-07T04:26:40 < R2COM> what special about it? 2019-10-07T04:26:46 < dongs> v3 is old, was releaserd months ago 2019-10-07T04:26:54 < dongs> its usb2 highspeed, uses stm32f7 or someshit instead of F103 2019-10-07T04:27:03 < dongs> and faster swo viewer and bla 2019-10-07T04:27:13 < R2COM> which device would you recommend now for debugging in custom boards? 2019-10-07T04:27:15 < R2COM> jlink? 2019-10-07T04:27:18 < dongs> jlink of course yeah 2019-10-07T04:27:23 < R2COM> which specifically ? 2019-10-07T04:27:24 < dongs> cuz ozone is >>>> any vendor shit 2019-10-07T04:27:31 < Cracki> I tried ordering that as a consumer... nobody sold it, or they had stock and still wouldn't sell it 2019-10-07T04:27:35 < dongs> ozone is best fucking debugger ever, even better than keil 2019-10-07T04:27:43 < dongs> Cracki: ordering waht, stlink v3? 2019-10-07T04:27:46 < Cracki> yes 2019-10-07T04:27:58 < dongs> why? is it coz it has F7 with crypto? lol 2019-10-07T04:28:12 < dongs> mouser has like 450 in stock 2019-10-07T04:28:19 < Cracki> nah, just dumb sellers 2019-10-07T04:28:52 < Cracki> not sure why I didn't check mouser... might have been shipping cost, or they didn't have it then 2019-10-07T04:28:55 < R2COM> jlink pro or plus? 2019-10-07T04:28:57 < R2COM> or which one? 2019-10-07T04:29:02 < R2COM> https://www.segger.com/products/debug-probes/j-link/models/model-overview/ 2019-10-07T04:29:06 < Cracki> which costs more 2019-10-07T04:29:33 < dongs> anything below ultra you can just buy -EDU 2019-10-07T04:29:42 < dongs> because you don't get any advantage over any of those 4 lesser models 2019-10-07T04:29:47 < Cracki> base looks equal to plus, and edu is base 2019-10-07T04:29:50 < dongs> yeah 2019-10-07T04:29:51 < Cracki> might differ in licenses 2019-10-07T04:29:54 < R2COM> so ultra+ is minimum one to get? 2019-10-07T04:30:10 < Cracki> _depends_ on what you wanna do 2019-10-07T04:30:25 < Cracki> edu is at-cost base 2019-10-07T04:30:36 < Cracki> and plus is base+speshul licenses 2019-10-07T04:31:04 < Cracki> see the next table 2019-10-07T04:31:14 < R2COM> also...hmm 2019-10-07T04:31:21 < R2COM> it will work fine with VisualGDB right? 2019-10-07T04:31:27 < dongs> yeah their gdbserver works great 2019-10-07T04:31:41 < dongs> but if thats waht youre using just keep using cheap shit 2019-10-07T04:31:50 < dongs> the only reason to get jlink is fast flash programing and ozone for debugging 2019-10-07T04:32:31 < R2COM> but can that ozone be mated with my VisualGDB? 2019-10-07T04:32:35 < R2COM> or proly not right 2019-10-07T04:32:39 < R2COM> cuz its gdb shit 2019-10-07T04:32:59 < dongs> its just a standalone debugger that takes keil or gcc or rhwatever-built .elf files with debugging info 2019-10-07T04:33:07 < dongs> and it does source level debugging and shit 2019-10-07T04:34:13 < R2COM> but i mean it somehow needs to be synced with my MSVC where i have code for that elf 2019-10-07T04:34:18 < R2COM> to step through code etc 2019-10-07T04:34:28 < dongs> .elf has this info 2019-10-07T04:34:37 < dongs> it will open your existing sores to step through and whatever 2019-10-07T04:34:43 < R2COM> hmm 2019-10-07T04:34:56 < R2COM> ozone purchased separately? 2019-10-07T04:35:00 < dongs> free 2019-10-07T04:35:20 < Cracki> ozone is a gui, it's not something you can plug into visual studio 2019-10-07T04:35:23 < dongs> it works on any jlink, i think non-trial mode requires RDI license but I don't think there's any limitation (time or otherwise) to trial mode so you can just use it with -EDU or whatever same 2019-10-07T04:35:38 < Cracki> the whole point of ozone is that it's a gui that does things better than other programs 2019-10-07T04:35:44 < dongs> yeah 2019-10-07T04:36:03 < dongs> its fucking amazing, combined with partial flash updates doing incremental changes and debugging them is so fast 2019-10-07T04:36:12 < Cracki> partial flash update? neat 2019-10-07T04:36:19 < Cracki> how does it do that, read before write? 2019-10-07T04:36:22 < dongs> yes 2019-10-07T04:36:27 < dongs> read, compare, write only changes 2019-10-07T04:36:35 < Cracki> that would help the flash live longer indeed 2019-10-07T04:36:37 < dongs> small updates like constants or wahtever are nearly instant 2019-10-07T04:36:37 -!- learningc [~learningc@2001:e68:5869:c400:3447:c534:b135:6b5] has joined ##stm32 2019-10-07T04:36:39 < R2COM> ultra and pro differ only in that ultra doesnt have ethernet...does one ever utilize it? 2019-10-07T04:36:54 < dongs> my pal flew a jlink pro on a shitcopter connected to ethernet>wifi adapter 2019-10-07T04:36:57 < dongs> and logged shit live 2019-10-07T04:37:24 < Cracki> oh, instant too? does SWD allow device-side reading and checksumming, does it upload a little program to handle that? or is reading just that much faster than writing? 2019-10-07T04:37:32 < Cracki> hah 2019-10-07T04:37:42 < Cracki> jlink pro has ethernet, right 2019-10-07T04:37:47 < R2COM> hah 2019-10-07T04:37:51 < dongs> Cracki: its nearly instant, thre's progress bars but they flash away so fast I have no idea. i think it does stick some stub in ram or something, but who cares how its done 2019-10-07T04:38:00 < dongs> Cracki: yea has ethernet, that was my point 2019-10-07T04:38:03 < Cracki> lovely 2019-10-07T04:38:22 < dongs> and it auto-reloads your elf when you tab into it 2019-10-07T04:38:24 < Cracki> I wasnt sure at what level they give these things ethernet 2019-10-07T04:38:29 < dongs> so lately when im debugging i dont even bother keil debugger 2019-10-07T04:38:33 < Cracki> hah 2019-10-07T04:38:38 < R2COM> dongs and of course segger jlink pro supports absolutely all ST's micros rite? 2019-10-07T04:38:39 < Cracki> I really gotta give it a try 2019-10-07T04:38:42 < dongs> i just have keil/ozone open side by side, rebuild in keil and alt-tab to ozone to debug 2019-10-07T04:39:00 < dongs> R2COM all jlinks have same supported mcus list, at least for arm 2019-10-07T04:39:01 < Cracki> that applies to all jlinks, not just the pro 2019-10-07T04:39:05 < dongs> yeah 2019-10-07T04:39:15 < dongs> some more expensive shit support renesas rx or wahtever trash that noboey uses 2019-10-07T04:39:19 < dongs> but for arm its all same 2019-10-07T04:39:20 < Cracki> so long as it has SWD, the rest is software support 2019-10-07T04:39:50 < R2COM> $1k 2019-10-07T04:39:53 < Cracki> well firmware mebbeh 2019-10-07T04:39:56 < R2COM> but well if its worth it 2019-10-07T04:39:58 < Cracki> get the base/whatever 2019-10-07T04:40:16 < Cracki> you upsell yourself, why 2019-10-07T04:40:22 < dongs> just get -edu for dicking or a chinaclone for even less 2019-10-07T04:40:40 < R2COM> chinaclone segger pro? 2019-10-07T04:40:48 < dongs> i used -edu for a while but once i started making proper money with STM32 i upgraded to ultra+ 2019-10-07T04:40:52 < dongs> i dont think there's chinaclone of -pro 2019-10-07T04:40:53 < dongs> or ultra 2019-10-07T04:40:57 < dongs> they have fpga and shit 2019-10-07T04:41:05 < dongs> all chink clones are of -base + licenses 2019-10-07T04:41:23 < dongs> so i guess its jlinkplus maybe equivalent 2019-10-07T04:41:25 < Cracki> what proof do they want for an edu? university email enough? 2019-10-07T04:41:44 < dongs> none 2019-10-07T04:41:47 < dongs> you can jsut buy -edu from mouser 2019-10-07T04:41:49 < Cracki> okeh 2019-10-07T04:41:50 < dongs> as anyone 2019-10-07T04:41:57 < dongs> or digikey or wahtever 2019-10-07T04:42:04 < Cracki> the only china clones I've found so far are of the jlink v8, which is old firmware and old hardware 2019-10-07T04:42:16 < dongs> there's definitely v9 stuf 2019-10-07T04:42:35 < Cracki> right 2019-10-07T04:42:41 < dongs> or even 'V10' haha 2019-10-07T04:42:50 < dongs> V8s are dirt chieap now 2019-10-07T04:42:57 < dongs> like <$10 2019-10-07T04:43:06 < Cracki> oh, now they have new ones. good. 2019-10-07T04:44:24 < Cracki> v10 clone for 40 bucks on ali 2019-10-07T04:44:39 < dongs> yes sounds about right 2019-10-07T04:44:45 < dongs> they're 250-260 rmb on taobao 2019-10-07T04:45:02 < dongs> at 40 bucks you gotta decide if you just wanna get -edu and be somewhat legit or just fuckit and go china 2019-10-07T04:45:07 < dongs> cuz -edu is like $50 or something 2019-10-07T04:45:12 < Cracki> soem forum says v10 is high speed usb 2019-10-07T04:45:39 < Cracki> yes, tiny difference in price 2019-10-07T04:45:57 < dongs> https://forum.segger.com/index.php/Thread/3270-SOLVED-What-s-the-difference-between-J-Link-V9-and-V10/ 2019-10-07T04:46:48 < R2COM> or...just continue using stlink v2 debugger from dev boards :P 2019-10-07T04:47:00 < Cracki> hah jlink edu mini, looks like stlink v3 mini 2019-10-07T04:47:02 < R2COM> if no hard need arised yet 2019-10-07T04:47:11 < Cracki> I wonder which of them designed it, and who copied it 2019-10-07T04:47:39 < dongs> not really it doesnt 2019-10-07T04:47:45 < dongs> stlink mini is way better spec 2019-10-07T04:47:53 < dongs> jlink shit is like 2 layer pcb with a F103 on it 2019-10-07T04:48:02 < dongs> stlink is BGA F7 on 6layer board 2019-10-07T04:48:03 < R2COM> except pro 2019-10-07T04:48:15 < dongs> im comparing -edu mini 2019-10-07T04:48:22 < dongs> i never took mine apart to see whats inside 2019-10-07T04:48:31 < Cracki> are you sure you haven't got them mixed up? I'd expect segger to do 6layer and bga 2019-10-07T04:48:32 < R2COM> so st link v3 would be > than jlinkn -edu 2019-10-07T04:48:51 < dongs> https://www.segger.com/fileadmin/images/products/J-Link/J-Link-Edu-Mini/J-Link_EDU_Mini_500x307.png 2019-10-07T04:48:59 < Cracki> ok that's no bga 2019-10-07T04:49:09 < Cracki> cute, a fagconnect 2019-10-07T04:49:50 < Cracki> I haven't been able to explain to my boss that he should just use the 5x2 cortex header because those tagconnects aren't very convenient for developing, only for flashing 2019-10-07T04:50:04 < Cracki> and for flashing I've already told him to make a bed of nails 2019-10-07T04:50:26 < R2COM> my visual gdb dopesnt have stlink v3 in drop down 2019-10-07T04:50:30 < R2COM> just stlink v2 and v2.1 2019-10-07T04:50:38 < Cracki> file a bug report 2019-10-07T04:50:41 < R2COM> or v1 2019-10-07T04:50:47 < Cracki> nobody has a v1 2019-10-07T04:51:51 < R2COM> v2 on digikey costs $22 2019-10-07T04:51:53 < dongs> https://www.segger.com/products/debug-probes/j-link/models/other-j-links/st-link-on-board/ 2019-10-07T04:51:55 < Cracki> chinapics say the jlink v10 has an lpc4337 on it. m4+m0 https://www.nxp.com/products/processors-and-microcontrollers/arm-microcontrollers/general-purpose-mcus/lpc4300-cortex-m4-m0/32-bit-arm-cortex-m4-m0-mcu-up-to-1-mb-flash-and-136-kb-sram-ethernet-two-high-speed-usb-lcd-emc:LPC4337FET256 2019-10-07T04:51:58 < dongs> this is the best thing you can do with a stlink v2 2019-10-07T04:52:02 < dongs> is to turn it into jlink 2019-10-07T04:52:17 < R2COM> wow 2019-10-07T04:52:26 < R2COM> and use ozone?? 2019-10-07T04:52:28 < dongs> yeah 2019-10-07T04:52:33 < R2COM> hmm 2019-10-07T04:52:44 < dongs> except it runs at around teh speed of j-link edu mini 2019-10-07T04:53:07 < dongs> so 4mhz target interfce speed 2019-10-07T04:53:12 < R2COM> its ok 2019-10-07T04:53:12 < dongs> but it works at least 2019-10-07T04:54:51 < R2COM> one confusing thing... for example in my VisualGDB when i create MCU project i can at the end select debugger, and in drop down i see segger jlink 2019-10-07T04:54:52 < R2COM> but 2019-10-07T04:54:57 < R2COM> does that assume ozone? 2019-10-07T04:55:00 < R2COM> or most likely no? 2019-10-07T04:55:01 < dongs> no 2019-10-07T04:55:04 < R2COM> ozone is separate shit rite? 2019-10-07T04:55:06 < dongs> that has nothing to do with that 2019-10-07T04:55:14 < Cracki> separate, yes 2019-10-07T04:55:16 < dongs> its completely separate you just load your binary into it and debug 2019-10-07T04:55:22 < R2COM> yeah 2019-10-07T04:55:29 < R2COM> so ozone is different standalone debugger 2019-10-07T04:56:16 < dongs> you can continue iusing it with visualgdb since it does have a gdbserver etc. 2019-10-07T04:56:28 < R2COM> ozone has gdbserver? 2019-10-07T04:56:38 < dongs> no, the stlink you convert to jlink with that thing 2019-10-07T04:56:42 < R2COM> yeah 2019-10-07T04:57:07 < R2COM> because so far this visualgdb works flawlessly 2019-10-07T04:57:27 < dongs> yera other than stlink being a complete shit. 2019-10-07T04:57:32 < dongs> like reflashin entire chip on debug 2019-10-07T04:57:58 < R2COM> jlink does not reflash every time you press debug? 2019-10-07T04:57:58 < dongs> i even did separate flash pages for readonly resources because it made debug with ozone so much faster 2019-10-07T04:58:09 < dongs> pin R/O stuff like assets/bitmaps/etc in higher flash blocks 2019-10-07T04:58:19 < Cracki> with linker script for fixed addresses? 2019-10-07T04:58:20 < dongs> and then link code from start->begin of RO block 2019-10-07T04:58:33 < dongs> this way it will never need to update the assets while debugging just one time only 2019-10-07T04:58:43 < dongs> Cracki: well, with keil scatter file but yeah 2019-10-07T04:58:47 < Cracki> ic 2019-10-07T04:58:54 < dongs> and #pragma to put resources in that flash region 2019-10-07T04:59:37 < Cracki> *notes* 2019-10-07T05:00:48 < dongs> with stdick that doesnt matter cuz it will erase entire flash anyway :D 2019-10-07T05:01:31 < Cracki> jlink ob, can that replace stlinkv3 firmware already? 2019-10-07T05:01:39 < dongs> nope 2019-10-07T05:01:41 < dongs> v1/v2 only 2019-10-07T05:01:51 < dongs> just the shit on -nucleo / disco boards 2019-10-07T05:02:07 < dongs> i think they technically support standalone stlink but i never owned one to try 2019-10-07T05:03:21 < Cracki> I might try that with a china clone. someone here had some success with that. 2019-10-07T05:03:31 < dongs> stlink chinaclone? 2019-10-07T05:03:33 < Cracki> yes 2019-10-07T05:03:35 < dongs> those $1 things in pink metal case? 2019-10-07T05:03:36 < dongs> lo 2019-10-07T05:03:39 < Cracki> which accidentally had 128k flash 2019-10-07T05:03:51 < Cracki> yes, 1.60 eurobucks depending on forex 2019-10-07T05:04:09 < Cracki> contains stm32f103 or equivalent CS32/GD32 2019-10-07T05:04:28 < Cracki> but it needs 128k flash which isn't what they're specd as 2019-10-07T05:04:32 < dongs> J-Link-OB-STM32F103-Cortex-M 2.25 MHz 2019-10-07T05:04:36 < dongs> ya pretty gey 2019-10-07T05:04:45 < dongs> even worse than -edu 2019-10-07T05:04:51 < Cracki> heh 2019-10-07T05:04:53 < dongs> EDU uses some NXP K22 shit 2019-10-07T05:04:56 < dongs> at 4mhz 2019-10-07T05:05:29 < Cracki> J-Link-OB-STM32F103-Cortex-A <- do they mean A as a target? 2019-10-07T05:05:34 < Cracki> weird table 2019-10-07T05:19:05 < dongs> yea i think so 2019-10-07T05:19:24 < dongs> as in what cores it can support 2019-10-07T05:19:32 < dongs> with that chip 2019-10-07T05:25:14 < Cracki> hm I thought of the wrong firmware replacement. the stlink v2.1 needs 128k chip variants, the jlink ob (for v2.0) works with standard 64k of course 2019-10-07T05:25:52 < Cracki> oh no the jlink ob lists an 103TB which is also 128k 2019-10-07T05:25:59 < dongs> TB? 2019-10-07T05:26:01 < dongs> the hell is that 2019-10-07T05:26:05 < Cracki> *f103tb 2019-10-07T05:26:28 < dongs> ohh huh they came up with a 36pin qfn package for that? 2019-10-07T05:26:30 < dongs> that must be new 2019-10-07T05:28:51 < Cracki> they support the f4-disc1, which has an f103c8, which is 64k, so it must fit in 64k despite what they list in that weird table 2019-10-07T05:33:29 < dongs> well that or they'er abusing 128k on C8 anyway 2019-10-07T05:33:39 < dongs> you can actually see all the jlink firmwares inside JlinkArmDLL 2019-10-07T05:33:43 < dongs> they're just attached as binary blobs 2019-10-07T05:33:50 < dongs> im pretty sure none of them are < 64k 2019-10-07T05:34:01 < dongs> nice 2019-10-07T05:39:57 -!- learningc [~learningc@2001:e68:5869:c400:3447:c534:b135:6b5] has quit [Read error: Connection reset by peer] 2019-10-07T05:43:06 < Cracki> yeah I'm gonna try that 2019-10-07T05:43:24 < Cracki> didn't find any info on how fast a normal stlinkv2 is with its swd, only the 4.5 MHz for the jlink ob 2019-10-07T05:43:28 < Cracki> (on f103) 2019-10-07T05:44:46 < dongs> its bout same, around 4mhz 2019-10-07T05:45:32 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-07T05:45:33 < dongs> or 2, i forget 2019-10-07T05:45:36 < dongs> but it was something pretty crap 2019-10-07T05:47:10 -!- learningc [~learningc@2001:e68:5869:c400:287a:6481:96e6:568c] has joined ##stm32 2019-10-07T05:47:14 < Cracki> 72/4.5 = 16, that sounds like it's not using any clock to sample the data signal 2019-10-07T05:47:33 < Cracki> or the bitbanging uses so many instructions 2019-10-07T05:47:58 < dongs> the swd stuff is on timers 2019-10-07T05:48:05 < dongs> im pretty sure they're not THAT dumb 2019-10-07T05:48:10 < dongs> they're at least using some hardware bits to bang it 2019-10-07T05:51:18 < Cracki> you never know how awful things are hooked up 2019-10-07T05:52:20 < dongs> https://ssd.userbenchmark.com/Compare/Crucial-MX500-1TB-vs-Samsung-860-Evo-500GB/m406099vsm428560 will i regret this purchase decision for a throwaway data drive using amazon gift card 2019-10-07T05:52:43 < Cracki> my boss, when he switched out an atmega for an stm32 in an existing design, just hooked slow/gpio signals up randomly (PWM he did sensibly). I had to go through it and assign by function and try to keep pins grouped by peripheral so they could serve alternate functions 2019-10-07T05:53:09 < Cracki> I saw generally favorable things about the mx300 and mx500 2019-10-07T05:53:25 < Cracki> same about samsung tho :P 2019-10-07T05:59:20 < Cracki> hm... stlink v3... no jlink ob support jet, so no ozone with that 2019-10-07T06:01:53 < Cracki> but keil would support it 2019-10-07T06:01:56 < dongs> making proper -ob support on such good hardware would jeopradize their real jlink sales 2019-10-07T06:02:13 < dongs> beacuse highspeed usb and proper mcu 2019-10-07T06:02:18 < Cracki> yeh 2019-10-07T06:02:25 < dongs> so i doubt they're in ah urry 2019-10-07T06:03:10 -!- learningc [~learningc@2001:e68:5869:c400:287a:6481:96e6:568c] has quit [Read error: Connection reset by peer] 2019-10-07T06:08:27 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 240 seconds] 2019-10-07T06:13:52 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-07T06:13:53 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-07T06:13:53 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-07T06:38:21 -!- fc5dc9d4_ [~quassel@p5B081662.dip0.t-ipconnect.de] has joined ##stm32 2019-10-07T06:41:39 -!- fc5dc9d4 [~quassel@p57A320C7.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-07T07:40:04 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-07T07:56:28 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-07T07:59:39 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-07T07:59:39 -!- day__ is now known as day 2019-10-07T08:26:10 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-07T08:29:31 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-07T08:33:20 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-07T09:24:39 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-07T09:30:59 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-07T09:39:25 -!- m4ssi [~massi@82.62.130.138] has joined ##stm32 2019-10-07T09:48:40 -!- learningc [~learningc@2001:e68:5869:c400:598f:599d:27de:c838] has joined ##stm32 2019-10-07T09:52:58 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-07T09:55:31 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-07T10:01:35 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-07T10:05:12 -!- jly [uid355225@gateway/web/irccloud.com/x-ymjnucgtrhyklako] has joined ##stm32 2019-10-07T10:05:31 < Steffanx> Wlcm 2019-10-07T10:05:46 < jly> If you have a chance to come to Japan, you can enjoy watching Japanese gardens, shrines, and watching anime.　　🗾⛩🌍 2019-10-07T10:20:10 < jly> no i'm not 2019-10-07T10:20:18 < jly> i don't like any anime tbh 2019-10-07T10:21:08 < jly> yeah 2019-10-07T10:22:32 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-07T10:50:12 < dongs> all anime is fucking garbage 2019-10-07T10:50:12 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-07T10:50:14 < dongs> > not very fat 2019-10-07T10:50:17 < dongs> > not very far 2019-10-07T10:50:22 < dongs> its like a 8 hours flight still 2019-10-07T10:50:54 < jly> guess it's quicker than blaxterland 2019-10-07T10:51:32 < Steffanx> You dont wanna visit that shithole.. 2019-10-07T10:54:11 < PaulFertser> dongs: the libzc author says it builds and runs just fine under Cygwin. Why were you opposed to the idea? Why should anyone spend extra time trying to build it fully natively? 2019-10-07T10:54:29 < PaulFertser> (still few real bugs uncovered along the way, those will patched properly) 2019-10-07T10:54:38 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-07T11:03:24 < dongs> PaulFertser: well, cygwin requires installation on windows, its not something I would want to deal with. cross compile seemed like something that would make more sense 2019-10-07T11:03:50 < PaulFertser> dongs: I think it just requires a single additional DLL to run, not any specific installation? 2019-10-07T11:04:01 < dongs> it requires the toolchain isntalled ON windows 2019-10-07T11:04:05 < dongs> which is aids 2019-10-07T11:04:08 < dongs> have yo useen cygwin installer? 2019-10-07T11:04:12 < dongs> it hasnt been updated for like 10 years 2019-10-07T11:04:15 < dongs> it ships with some ancient gcc etc 2019-10-07T11:04:21 < dongs> i just dont want that shit on my system 2019-10-07T11:04:42 < dongs> hmm 2019-10-07T11:04:46 < srk> just use WSL :) 2019-10-07T11:04:47 < dongs> is page size on F401 really like 128k?? 2019-10-07T11:04:50 < dongs> at the bottom of flash 2019-10-07T11:05:07 < PaulFertser> I see 2019-10-07T11:08:17 < qyx> wut, no iirc 2019-10-07T11:08:27 < qyx> it was something like 16, 16, 32, 64, 128, etc. 2019-10-07T11:10:23 < dongs> qyx, yes, last page is 128k 2019-10-07T11:10:47 < dongs> trying to use as eeprom lol 2019-10-07T11:10:48 < dongs> fuck this 2019-10-07T11:35:36 < dongs> question for RF pros how is C/N measurement done by using "out of band noise point" whatever the fck that is 2019-10-07T11:35:51 < dongs> http://www.micronix-jp.com/english/note/application/C_N.html oh is it this? 2019-10-07T11:36:49 < dongs> ahh, okay cool thats very easy to follow 2019-10-07T11:45:05 < jadew> are you writing documentation? 2019-10-07T11:45:48 < dongs> no 2019-10-07T11:45:58 < dongs> why would I be? 2019-10-07T11:46:06 < jadew> was wondering why you needed that 2019-10-07T11:46:11 < dongs> oh, to measure the cn 2019-10-07T12:03:31 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-07T12:03:56 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-07T12:27:57 < jpa-> so, what do people think, is there any point putting a 100nF cap next to cpu if there is already regulator's 1µF cap right next to it? 2019-10-07T12:32:17 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-07T12:40:38 < jadew> depends on the CPU 2019-10-07T12:48:35 < jadew> sometimes that's good enough 2019-10-07T12:51:50 < jadew> they're not necessarily decoupling caps :) 2019-10-07T12:56:10 < karlp> we even have a collection of them in zypsnips 2019-10-07T12:56:32 < karlp> jpa-: in teh words of dongus maximums, "remove for great justice" 2019-10-07T12:57:01 < karlp> though 1uF+100nF are for different freqs too, not just size... 2019-10-07T12:57:47 < zyp> 1uF and 100nF aren't that different in ESR though 2019-10-07T12:58:42 < jadew> you short it 2019-10-07T12:59:03 < zyp> it's ESR, not EPR 2019-10-07T12:59:12 < PaulFertser> 1uF electrolytic compared to ceramic 100nF? 2019-10-07T12:59:24 < zyp> both ceramic, I assume 2019-10-07T12:59:51 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-07T12:59:55 < PaulFertser> jpa-: did you mean ceramic? 2019-10-07T13:02:47 < jadew> I'm sure he meant ceramic 2019-10-07T13:02:57 < jadew> he's not dumb 2019-10-07T13:03:00 < karlp> what's antoo putting 47uF on the usb vbus for? by spec, you're only meant to be putting 10uF 2019-10-07T13:03:47 -!- learningc [~learningc@2001:e68:5869:c400:598f:599d:27de:c838] has quit [Ping timeout: 250 seconds] 2019-10-07T13:05:29 < jadew> I guess that's what they came up with after analyzing the garbage that comes out of PC PSUs 2019-10-07T13:06:41 < jadew> it was either "ok... it looks like 10 uF works best." or "10 uF for good measure." 2019-10-07T13:07:27 -!- mitrax [mitrax@lfbn-ncy-1-393-156.w83-196.abo.wanadoo.fr] has joined ##stm32 2019-10-07T13:09:01 < jadew> sleep time, bye 2019-10-07T13:09:27 < karlp> Cracki: even git diff on the cli supports external tools for people who "need" winmerge 2019-10-07T13:09:42 < jadew> Haohmaru, I woke up at midnight :/ 2019-10-07T13:16:08 < karlp> Haohmaru:it might work, but why did you even think you needed a 47uF electrolytic in addition to the 10uf input cap you already had there? 2019-10-07T13:16:32 < mitrax> is it common for screen driver/controllers to have a D/C (data/command) pin that's required in spi mode? i have an oled screen with an SSD1305Z controller (that supports parallel/i2c/spi ), from the screen datasheet the D/C pin is not specified as used in spi mode, though the SSD1305 datasheet seems to imply that pin is necessary to write to GDDRAM 2019-10-07T13:16:42 < karlp> jadew: it's to limit inrush / to limit outrush required on mobos.... 2019-10-07T13:17:01 < karlp> it's 10uF/50uC. 2019-10-07T13:19:37 < karlp> because? 2019-10-07T13:20:46 < karlp> like, yuou go all matched trace lenghts for some stuff, then just "let's just make shit up y0l0" for others... 2019-10-07T13:21:16 < zyp> don't we all? 2019-10-07T13:21:50 < zyp> I think it's pretty common to overthink some stuff and handwave other stuff :p 2019-10-07T13:23:18 < PaulFertser> mitrax: iirc with ssd13xx controllers D/C (called RS in my projects for whatever reason) was always needed in SPI, yes. 2019-10-07T13:24:01 < PaulFertser> But despite being called RS it's switching between "data" and "control". 2019-10-07T13:24:26 < zyp> yeah 2019-10-07T13:24:33 < zyp> https://paste.jvnv.net/view/SlArb <- here's my ssd1331 driver 2019-10-07T13:24:33 < mitrax> PaulFertser: ok, so as soon as you want to write to the graphic display memory you need to turn it high? (i.e data vs command) 2019-10-07T13:25:08 < zyp> mitrax, correct, just like the parallel interface 2019-10-07T13:25:18 < PaulFertser> mitrax: yes. There must be some reason why it's called RS in my sources though... 2019-10-07T13:25:54 < mitrax> zyp/PaulFerster: alright, thanks :) 2019-10-07T13:27:58 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-07T13:28:04 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-07T13:32:00 < jpa-> PaulFertser: yeah, ceramic sure 2019-10-07T13:32:41 < zyp> jpa-, I'd skip the 100n if I wanted to save the area 2019-10-07T13:33:07 < zyp> but if there's no benefit from removing it, might as well put it there, 100n costs nothing 2019-10-07T13:34:01 < jpa-> yep, that's my feeling also that it doesn't really add much benefit 2019-10-07T13:34:33 < karlp> time for basic? https://www.kickstarter.com/projects/703885653/basic-compiler-for-arduino-zero-boards/ 2019-10-07T13:34:42 < zyp> no thanks 2019-10-07T13:36:55 < Steffanx> BASCOM was tha bomb. 2019-10-07T14:01:53 < dongs> is that a dickstarter by the bascocks dude? 2019-10-07T14:02:12 < dongs> ah no 2019-10-07T14:02:18 < dongs> oh well, but who cares about basic in 2019? 2019-10-07T14:04:16 < karlp> https://www.eenewseurope.com/news/avnet-loses-tis-19bn-distribution-business?news_id=122018 2019-10-07T14:07:31 < dongs> not even TI wants to deal waith avnet 2019-10-07T14:07:39 < dongs> isnt avnet one of those dickstarter scam enablers 2019-10-07T14:07:49 < dongs> that "checks" technical projects and vets that they're not "hot garbage" 2019-10-07T14:07:59 < karlp> owns farnell and silica and a few other distirbuts 2019-10-07T14:08:09 < dongs> https://ir.avnet.com/news-releases/news-release-details/avnet-dragon-innovation-and-kickstarter-launch-hardware-studio 2019-10-07T14:08:12 < dongs> lol 2019-10-07T14:08:20 < dongs> fucking wankers 2019-10-07T14:08:33 < dongs> yeah 2019-10-07T14:08:39 < dongs> but who teh fuck acutgally buys from avnet 2019-10-07T14:08:44 < dongs> they don't evne haev free shipping 2019-10-07T14:09:07 < dongs> one time I tried to buy some xilinx shit that was ONLY at avnet 2019-10-07T14:09:12 < dongs> it was like $100 of parts and $100 shipping 2019-10-07T14:09:14 < dongs> fuck outta here 2019-10-07T14:21:11 < qyx> I said it 2019-10-07T14:21:19 < qyx> farnell is a no-go since it is owned by avnet 2019-10-07T14:21:44 < qyx> it has an ugly green web 2019-10-07T14:22:47 < dongs> one of my customers keeps using retarded TI RF shit 2019-10-07T14:22:51 < dongs> they're so fucking expensive 2019-10-07T14:23:03 < qyx> I am using mostly ti parts 2019-10-07T14:23:10 < dongs> im using chinese parts 2019-10-07T14:23:12 < dongs> and so should you 2019-10-07T14:23:12 < qyx> except when reasonable to use something else 2019-10-07T14:23:20 < dongs> why would you use ti vregs 2019-10-07T14:23:26 < qyx> why not 2019-10-07T14:23:28 < dongs> there's literally hundreds of china compabitles for fraction of cost 2019-10-07T14:23:29 < dongs> because 2019-10-07T14:23:36 < dongs> qyx, gimme example of wat ti shit youre using 2019-10-07T14:23:47 < qyx> all dc/dc for example 2019-10-07T14:23:57 < dongs> silly goose 2019-10-07T14:24:27 < qyx> it has a nice webench, clicky-clicky, parts selected, sch done 2019-10-07T14:24:42 < dongs> or you could use parts that dont need a fucking webench 2019-10-07T14:24:44 < dongs> and just work 2019-10-07T14:25:42 < qyx> wut? 2019-10-07T14:26:53 < qyx> in the past, yes 2019-10-07T14:27:05 < Steffanx> some people still live in the past. 2019-10-07T14:54:51 -!- jly [uid355225@gateway/web/irccloud.com/x-ymjnucgtrhyklako] has quit [Quit: Connection closed for inactivity] 2019-10-07T15:29:28 < Ecco> Hi :) 2019-10-07T15:29:46 < Ecco> I'm looking at using an USB-C connector to power an STM32 board 2019-10-07T15:30:05 < Ecco> nothing fancy, just low-amp 5V power input, plus good-old D+/D- 2019-10-07T15:30:21 < Ecco> micro-usb would be easier, but these days people like USB-C better 2019-10-07T15:30:25 -!- fc5dc9d4_ is now known as fc5dc9d4 2019-10-07T15:30:32 < Ecco> I know the people at Rpi fucked up when they switched to USB-C 2019-10-07T15:30:47 < Ecco> question -> Is there a simple reference schematics I could follow for this (I think very common) use case? 2019-10-07T15:31:03 -!- fc5dc9d4 [~quassel@p5B081662.dip0.t-ipconnect.de] has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.] 2019-10-07T15:32:07 < Ecco> Ha 2019-10-07T15:32:09 < Ecco> Apprently https://medium.com/@leung.benson/how-to-design-a-proper-usb-c-power-sink-hint-not-the-way-raspberry-pi-4-did-it-f470d7a5910 2019-10-07T15:43:54 < karlp> Ecco: just use https://lcsc.com/product-detail/USB-Connectors_XKB-Enterprise-U262-16XN-4BVC11_C319148.html 2019-10-07T15:44:05 < karlp> usb c.... but just what you needed from it... 2019-10-07T15:50:10 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-07T15:50:35 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-07T15:55:25 < zyp> Ecco, if you want a usb-c connector and don't care about any of the new features and just wanna use it as a direct swap for a micro-b-connector, all you need to add is two pulldowns 2019-10-07T15:55:33 < zyp> or maybe they were pullups, I forgot 2019-10-07T15:56:31 < zyp> oh, that article is interesting 2019-10-07T15:57:11 < zyp> I did some consulting a few months ago, reviewing a design, and one of the things I commented on was that they had tied CC1 and CC2 together 2019-10-07T15:57:53 < zyp> I was like «I believe you can't do it this way, but I'm not 100% sure, so please double check the spec» 2019-10-07T15:58:23 < zyp> I hope they fixed that 2019-10-07T16:13:39 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-07T16:14:10 < Ecco> :-) 2019-10-07T16:14:19 < Ecco> So, pretty much just what the article says ? 2019-10-07T16:14:31 < Ecco> One 5.1K pull-down per CCi 2019-10-07T16:14:43 < Ecco> And then tie D+/D-/Vbus just like before 2019-10-07T16:14:48 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds] 2019-10-07T16:16:26 < zyp> yep, if your application ignored ID and only hooked up vbus/d+/d-/gnd, you can keep hooking up them like before and then only add a pulldown on each CC 2019-10-07T16:17:33 < zyp> alternatively if you did OTG and therefore need ID, you can get a chip that translates the CC lines to ID and hook the rest up as before 2019-10-07T16:19:02 < zyp> and if you wanna do usb3 you can get a chip that does the plug orientation muxing for you 2019-10-07T16:19:44 < zyp> that should cover all micro-usb can do 2019-10-07T16:20:37 < zyp> usb2 doesn't need muxing, you just hook both d+ and both d- orientations together 2019-10-07T16:44:32 < Ecco> ok, great :) 2019-10-07T16:44:41 < Ecco> I was just doing the dumbest thing anyway :) 2019-10-07T16:44:45 < Ecco> Thanks zyp :) 2019-10-07T16:44:59 < Ecco> Now I'm trying to figure out how this connector is supposed to be routed 2019-10-07T16:45:26 < Ecco> the "B" line of pads (B1-B12) is blocked below by the edge of the PCB 2019-10-07T16:45:31 < Ecco> and above by the A line of pads 2019-10-07T16:45:45 < Ecco> And they're super close to each other 2019-10-07T16:46:09 < Ecco> I wonder how you're supposed to fan this out 2019-10-07T16:47:47 -!- kow__ [~iccy@135.0.26.39] has joined ##stm32 2019-10-07T16:48:12 < Ecco> Ha, apparently there are 12 or 16 contacts ports 2019-10-07T16:48:20 < Ecco> I'd better grab one of those I guess 2019-10-07T16:48:36 < karlp> that's the one I linked, drops the highspeed pairs to give you less contacts to deal with 2019-10-07T16:49:15 < Ecco> ok :) 2019-10-07T16:51:04 -!- kow_ [~iccy@135.0.26.39] has quit [Ping timeout: 264 seconds] 2019-10-07T16:52:51 < jadew> karlp, I know what decoupling caps are for :) 2019-10-07T16:53:40 < Steffanx> for decoupling right? 2019-10-07T16:53:50 < jadew> right 2019-10-07T16:54:35 < zyp> are you sure? 2019-10-07T16:55:10 < Steffanx> only when you are 2019-10-07T16:55:50 < jadew> I use them for other stuff too, if the price is better 2019-10-07T16:56:00 < jadew> 2019-10-07T17:15:33 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-07T17:19:59 < kakimir64> muh jlink has left the warehouse 2019-10-07T17:24:55 < dongs> wat did you get 2019-10-07T17:25:03 < dongs> chinklink? 2019-10-07T17:25:06 < kakimir64> EDU for now 2019-10-07T17:25:46 < kakimir64> for now even getting gdb load to do uncorrupted flashing would be a victory 2019-10-07T17:26:12 < dongs> ah your LPC trash lol 2019-10-07T17:28:48 < kakimir64> idk what makes it trash 2019-10-07T17:29:01 < kakimir64> fairly modern variant etcetc. 2019-10-07T17:29:16 < dongs> nope 2019-10-07T17:29:20 < dongs> their peripherals are tarsh 2019-10-07T17:29:24 < dongs> their stdlib is nonexistent 2019-10-07T17:29:33 < dongs> nobody actually uses their shit so there's no support/etc 2019-10-07T17:29:35 < karlp> jadew: I was referring to the spec limits 2019-10-07T17:30:04 < kakimir64> dongs: on what chip models you base your accusation of tarsh peripherals? 2019-10-07T17:30:11 < dongs> they're all same??? 2019-10-07T17:30:16 < kakimir64> nope 2019-10-07T17:30:25 < dongs> all lpc peripherals are from like thier 8bit retarded things from 10 years ago 2019-10-07T17:31:38 < kakimir64> and even so.. what is the problem 2019-10-07T17:32:26 < kakimir64> newer lpcs have different peripherals than your pre cortex era and early cortex era.. at least some of them 2019-10-07T17:33:53 < karlp> what's the future path with both kinetis and lpc though? 2019-10-07T17:33:58 < karlp> I was surprised to new lpc parts honestly. 2019-10-07T17:34:37 < kakimir64> idk 2019-10-07T17:36:58 < zyp> IME lpcs have some cool peripherals and some that are trash 2019-10-07T17:37:06 < zyp> problem are you can't get the good without the bad 2019-10-07T17:37:22 < Steffanx> did you even try real segger tools kakimir64? 2019-10-07T17:37:51 < kakimir64> not with segger jlink ever Steffanx 2019-10-07T17:37:52 < zyp> flash peripheral sounds like one of the bad ones :p 2019-10-07T17:37:58 < dongs> haha 2019-10-07T17:38:00 < kakimir64> just with integrated programmers 2019-10-07T17:38:12 < dongs> kikemir why not reflash a stlink to jlink and try your shit now 2019-10-07T17:39:35 < Steffanx> i thought you said you were already using a jlink kakimir64 2019-10-07T17:40:22 < kakimir64> stay confused Steffanx 2019-10-07T17:40:40 < Steffanx> Try to unconfuse me 2019-10-07T17:41:12 < kakimir64> I have used jlink software with only jlink OB's 2019-10-07T17:41:36 < kakimir64> not that EDU would have any more features or performance 2019-10-07T17:43:16 < Steffanx> but did you even TRY your jlink with real segger-ware to see if it has the same issue as your openocd, kakimir64? 2019-10-07T17:43:28 < kakimir64> oh 2019-10-07T17:43:32 < kakimir64> with this chip 2019-10-07T17:43:34 < dongs> edu IS slightly better perf 2019-10-07T17:43:34 < kakimir64> no 2019-10-07T17:43:47 < dongs> than OB 2019-10-07T17:43:59 < qyx> and functional-wise? 2019-10-07T17:44:03 < dongs> OB is 4mhz max, edu is 15mhz interface speed 2019-10-07T17:44:05 < dongs> function-wise same 2019-10-07T17:45:47 < Steffanx> You should try ozone kakimir64. 2019-10-07T17:45:50 < Steffanx> Best think about ozone is the speed of it. How fast it (partially) reloads firmware. 2019-10-07T17:45:55 < Steffanx> its pretty damn fast. 2019-10-07T17:46:12 < Steffanx> *try = use 2019-10-07T17:47:38 < Steffanx> Im not sure if it's an improvement. 2019-10-07T17:48:27 < kakimir64> it's kakimir odesktop version 2019-10-07T17:48:36 < Steffanx> We should put him back in his Poké Ball 2019-10-07T17:48:40 < kakimir64> craptop version is still 32bit 2019-10-07T17:49:02 < kakimir32> poop 2019-10-07T17:52:55 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-07T17:54:22 < zyp> who cares? stm32 is better, this is ##stm32 2019-10-07T17:54:49 < kakimir64> yes Haohmaru.. use what you want but remember the best is always the best 2019-10-07T17:55:07 < Steffanx> Until you have a look at the SERCOM, must be the most complicated peripheral/pin config ever. 2019-10-07T17:55:22 < kakimir64> SAMs look like good combination of performance and pricing in some cases 2019-10-07T18:08:53 < srk> haha, if uart + spi + i2c weren't complicated enough you put them all together! 2019-10-07T18:10:58 < srk> crossbar sounds like a better idea 2019-10-07T18:16:33 < Steffanx> perhaps it's not that complicated at all. but for example the pin config missed a proper overview. 2019-10-07T18:16:47 < Steffanx> *misses 2019-10-07T18:17:20 < Steffanx> and the i2c part cant be worse than what stm32 manages to do for years now 2019-10-07T18:23:45 < srk> v2 i2c is quite nice, you just set tx/rx lengths and feed t/rxdata regs on ISR 2019-10-07T18:23:59 < srk> unlike massive state machine of v1 2019-10-07T18:25:13 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-07T18:25:41 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-07T18:43:46 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-07T18:49:02 -!- kakimir64 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-07T18:58:41 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-07T18:58:45 -!- yaqwsx [znc@anna.fi.muni.cz] has quit [Ping timeout: 246 seconds] 2019-10-07T19:00:03 -!- yaqwsx [znc@anna.fi.muni.cz] has joined ##stm32 2019-10-07T19:01:41 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-07T19:01:54 < bitmask> https://youtu.be/-MMJBvFomaM 2019-10-07T19:02:49 < aandrew> that's quite the camera rig, wow 2019-10-07T19:03:22 < aandrew> using the xbox controller is probably the most asian thing eer 2019-10-07T19:03:24 < aandrew> ever 2019-10-07T19:03:48 < aandrew> SHORYUKEN! 2019-10-07T19:06:10 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-07T19:07:11 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-07T19:12:48 -!- renn0xtk9 [~max@2a02:810d:1540:2448:f9ef:95e5:ae1f:a3fc] has joined ##stm32 2019-10-07T19:13:51 < bitmask> any place to learn common pcb design stuff? like things you should always do? decoupling caps and stuff 2019-10-07T19:14:05 < bitmask> I dont have enough experience and I want this board to be good 2019-10-07T19:15:05 < jpa-> copy dev boards 2019-10-07T19:22:29 -!- renn0xtk9 [~max@2a02:810d:1540:2448:f9ef:95e5:ae1f:a3fc] has quit [Quit: Konversation terminated!] 2019-10-07T19:24:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds] 2019-10-07T19:47:48 -!- yaqwsx [znc@anna.fi.muni.cz] has quit [Ping timeout: 245 seconds] 2019-10-07T19:48:41 -!- yaqwsx [znc@anna.fi.muni.cz] has joined ##stm32 2019-10-07T20:19:25 < antto> yeah, don't skip PCB layout recommendations in datasheets, look for example PCBs from the manufacturer (like dev boards, demo boards, etc..) 2019-10-07T20:20:09 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-07T20:20:11 < antto> replace your smile with a bored expression, and u gon look pr0 2019-10-07T20:20:16 < Steffanx> Where did you take the Haohmaru guy, antto? 2019-10-07T20:20:22 < Steffanx> is he alright? 2019-10-07T20:20:25 < Steffanx> Do you know? 2019-10-07T20:20:37 < antto> i haven't.. 2019-10-07T20:22:21 < Steffanx> That's what i would say 2019-10-07T20:26:39 < antto> >:/ 2019-10-07T20:26:43 * antto looking pr0 2019-10-07T20:29:56 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-07T20:30:04 < Steffanx> Will try again tomorrow 2019-10-07T20:31:18 < antto> wut'cha gon try? 2019-10-07T20:34:44 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-07T20:36:02 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-07T20:41:42 < brdb> bitmask: EEVBlog has a few videos up on PCB design, provided you can stand Dave droning on and on, there are a lot of things to learn from that along with the other recommendations made above 2019-10-07T20:41:57 < brdb> at the end of the day it really depends on your application. best way to learn is try and fail 2019-10-07T20:42:12 < bitmask> oh people responded, thanks guys 2019-10-07T20:42:26 < antto> "HOY DERR, i'm DAYV" 2019-10-07T20:42:53 < brdb> dont turn it on, TAKE IT APA'HT 2019-10-07T20:43:02 < antto> if you like steve erwin you'd love dave 2019-10-07T20:43:12 < brdb> dave's a bit more whiny, but yeah 2019-10-07T20:43:32 < antto> are all australians like that? 2019-10-07T20:43:32 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-07T20:43:40 < brdb> i doubt it 2019-10-07T20:43:49 < brdb> but there's a chance 2019-10-07T20:46:18 < brdb> bitmask: oh also if you want to look like a pro, add test pads/points. these are also great if you mess something up and have to cut and reroute a connection using bodge wire 2019-10-07T20:46:26 < brdb> the more bodge wires you have, the more pro you are 2019-10-07T20:49:15 < brdb> that begs the question: do most people include test points in their schematic to auto populate all the required test points (for a given footprint) when doing pcb layout or add them in as simple pads after the fact? 2019-10-07T20:51:21 < Ultrasauce> they are of pretty limited usefulness if you cant crossreference them to the schematic 2019-10-07T20:54:33 < qyx> I add them as a symbol in the schematic 2019-10-07T20:59:09 < antto> i don't like adding sh*t that matters to the PCB but not to the schematic 2019-10-07T21:13:45 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has quit [Read error: Connection reset by peer] 2019-10-07T21:17:41 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-07T21:34:35 < zyp> what matters is relative 2019-10-07T21:35:12 < zyp> I don't like adding mechanical shit to the schematic 2019-10-07T21:35:16 < zyp> like mounting holes 2019-10-07T21:36:05 < zyp> the way I see it, schematic specs logical design, layout specs physical design, and the mounting holes are part of the latter 2019-10-07T21:36:26 < zyp> test points on the other hand is also part of the logical design, since they connect to the signals 2019-10-07T21:36:37 < zyp> so I would absolutely put them in the schematic 2019-10-07T21:38:33 < Steffanx> even if your mounting holes happen to be connected to ground? 2019-10-07T21:38:53 < zyp> yes 2019-10-07T21:39:40 < zyp> I would reconsider if I were gonna connect them to anything other than ground 2019-10-07T21:39:43 -!- kakimir64 [575d220c@87-93-34-12.bb.dnainternet.fi] has joined ##stm32 2019-10-07T21:40:39 -!- m4ssi [~massi@82.62.130.138] has quit [Remote host closed the connection] 2019-10-07T21:59:57 < Steffanx> hm 2019-10-07T22:19:00 < kakimir64> i need wired lazör mouse 2019-10-07T22:19:25 < kakimir64> my mx master is constantly trying to wall from my tilted keyboard desk 2019-10-07T22:19:29 < kakimir64> fall* 2019-10-07T22:19:43 < kakimir64> also laggy in lunix for some reason 2019-10-07T22:34:45 < aandrew> https://www.digikey.ca/product-detail/en/stmicroelectronics/STEVAL-USBC2DP/497-18037-ND/8593717 2019-10-07T22:34:48 < aandrew> interesting 2019-10-07T22:37:57 < kakimir64> 144hz coding monitor 2019-10-07T22:41:18 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-07T22:43:35 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 276 seconds] 2019-10-07T22:43:56 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2019-10-07T22:45:55 < aandrew> https://www.tindie.com/products/clarahobbs/pd-buddy-sink/ that's really fucking expensive for what it is 2019-10-07T22:48:04 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-07T22:51:21 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Read error: No route to host] 2019-10-07T22:58:48 < kakimir64> heap_useNewlib.c 2019-10-07T22:58:58 < kakimir64> where should I find this 2019-10-07T23:05:01 < kakimir64> it's in linked resources and URI is just garbage 2019-10-07T23:05:13 < kakimir64> anything i try to do causes java vommit 2019-10-07T23:05:14 < zyp> aandrew, eh, $30 isn't that bad for a small volume project 2019-10-07T23:05:28 < aandrew> I guess, but he wants aother $12 to ship 2019-10-07T23:05:56 < aandrew> now we're up to ~50, which is in the "eeh maybe not" territory 2019-10-07T23:06:54 < kakimir64> only for reversing the design 2019-10-07T23:07:20 < zyp> don't even need to reverse it, it's pretty obvious just looking at what it can do 2019-10-07T23:07:27 < aandrew> yes 2019-10-07T23:07:44 < aandrew> does G0 have PD circuitry in it? everyone seems to use G0 for this 2019-10-07T23:08:31 < zyp> PD is just highlevel protocol, it runs on top of whatever the digital CC protocol is called 2019-10-07T23:08:39 < zyp> and yes, G0 can talk CC 2019-10-07T23:11:44 < zyp> hmm, ST calls the periperal UCPD though 2019-10-07T23:20:18 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Ping timeout: 245 seconds] 2019-10-07T23:20:57 < aandrew> right that's what I meant, there's some peripheral that does the physicla layer 2019-10-07T23:23:04 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-07T23:33:35 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-07T23:34:51 < karlp> aandrew: is anythin on tindie cheap/fair priced? 2019-10-07T23:37:53 < qyx> kakimir64: freertos? 2019-10-07T23:38:02 < kakimir64> itf 2019-10-07T23:38:14 < kakimir64> yes 2019-10-07T23:38:24 < kakimir64> project is fucked 2019-10-07T23:38:25 < qyx> memmang/gcc/heap3.c 2019-10-07T23:38:28 < qyx> or so 2019-10-07T23:38:38 < kakimir64> eclipse project* 2019-10-07T23:39:05 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-07T23:39:11 < qyx> portable/MemMang/heap_3.c 2019-10-07T23:39:13 < qyx> exact 2019-10-07T23:43:17 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-07T23:49:09 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 --- Day changed Tue Oct 08 2019 2019-10-08T00:08:10 < Cracki> kakimir64, so you discovered that ST has fucked up newlib+freertos? 2019-10-08T00:08:30 < Cracki> http://www.nadler.com/embedded/newlibAndFreeRTOS.html 2019-10-08T00:09:27 < Cracki> corresponding discussion: https://community.st.com/s/question/0D70X000007Oohv/bug-cubemx-freertos-projects-corrupt-memory 2019-10-08T00:09:27 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-08T00:13:05 < Cracki> cubemx may overwrite syscalls.c in some situations, so be sure to have that under revision control 2019-10-08T00:14:55 < englishman> aandrew: why the fuck are you waiting time on tindie 2019-10-08T00:15:11 < englishman> cypress' BCR devboard is cheaper than that 2019-10-08T00:15:42 < englishman> *wasting 2019-10-08T00:17:41 < kakimir64> Cracki: I moved workspace from much older windows tool to newer linux tool 2019-10-08T00:17:46 < kakimir64> eclipse life 2019-10-08T00:17:52 < kakimir64> everything is fucked now 2019-10-08T00:19:06 < kakimir64> I need to go throught all settings 2019-10-08T00:23:41 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-08T00:24:51 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-08T00:27:36 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-08T00:39:43 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 265 seconds] 2019-10-08T00:50:26 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-08T01:29:37 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 240 seconds] 2019-10-08T01:35:33 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-08T01:41:59 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-08T01:42:41 < Lux> aandrew: there is also that stuff: https://www.aliexpress.com/item/4000056182909.html 2019-10-08T01:42:53 < Lux> no idea how well it works though 2019-10-08T01:45:24 < aandrew> hah they're using G0 as well 2019-10-08T01:45:40 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-08T01:45:52 -!- con3 [~kvirc@146.232.65.240] has quit [Client Quit] 2019-10-08T01:46:35 < aandrew> actually that looks like F030 2019-10-08T01:50:17 < Lux> definetly 2019-10-08T01:51:18 < Lux> it probably uses that st example https://github.com/st-one/X-CUBE-USB-PD 2019-10-08T01:51:55 < mawk> Steffanx: I watched Chateau Meiland 2019-10-08T01:51:57 < mawk> you like it ? 2019-10-08T01:52:57 < mawk> there are too much commercials on dutch tv I hate it 2019-10-08T01:56:22 < Lux> but tbh i have no idea about how usb-pd works :) 2019-10-08T01:56:56 < Lux> is it also just some strange line toggling sequence like quick charge ? 2019-10-08T02:00:34 < mawk> no 2019-10-08T02:00:38 < mawk> it's usb c 2019-10-08T02:00:41 < mawk> with the additional data lines 2019-10-08T02:00:50 < mawk> no ugly pullup hacks like QC 2019-10-08T02:05:33 -!- kakimir64 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-08T02:14:28 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 264 seconds] 2019-10-08T02:43:32 < englishman> it uses the cc wire 2019-10-08T02:50:59 < Lux> finally found it: https://www.embedded.com/usb-type-c-and-power-delivery-101-power-delivery-protocol/ 2019-10-08T02:51:15 < Lux> yeah seems pretty sophisticated 2019-10-08T02:54:35 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-08T02:54:38 < Laurenceb> who here looking forward to 8kun? 2019-10-08T02:57:48 < Laurenceb> hopefully I'll have somewhere to shitpost after I got b& from endchan this morning https://endchan.net/.media/t_65e4824cfea6c9e16a9a15c70d7d935e-imagejpeg 2019-10-08T03:06:38 < englishman> it is rather complicated 2019-10-08T03:06:49 < englishman> you can download the pd spec and see for yourself 2019-10-08T03:11:54 < Cracki> ozone's flash download is even fast on some jlink v8 china clone. very nice. 2019-10-08T03:12:47 < Cracki> wish I'd have tried ozone earlier. it took the elf and didn't even ask where the sources are 2019-10-08T03:16:30 < dongs> yep 2019-10-08T03:16:36 < dongs> shits amazing 2019-10-08T03:18:47 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has left ##stm32 [] 2019-10-08T03:31:58 -!- learningc [~learningc@2001:e68:5869:c400:5cfa:d5f1:c85c:9769] has joined ##stm32 2019-10-08T03:33:10 -!- User_ [~learningc@2001:e68:5869:c400:598f:599d:27de:c838] has joined ##stm32 2019-10-08T03:33:23 < Cracki> swo viewer is glitching but it's updating isntantly, not with that 1 fucking second delay truestudio/cubeide does 2019-10-08T03:34:08 < Cracki> now it shows what I print but several times the debug session didn't show it. might be my shit wiring, or the china clone has loose solder joints or whatever 2019-10-08T03:34:30 < dongs> wait till you discover teh graphing stuff in watch window 2019-10-08T03:34:36 < Cracki> now I'm not so sure anymore what the point would be in an stlinkv3 2019-10-08T03:34:39 < dongs> you can refresh it at like 10khz 2019-10-08T03:34:47 < Cracki> yeh I saw the power metering 2019-10-08T03:34:56 < dongs> not power metering 2019-10-08T03:34:56 < Cracki> variable plotting is nice too :) 2019-10-08T03:34:59 < dongs> you can watch a var and plot yeah 2019-10-08T03:35:09 < dongs> i dont think power metering shit would work on chinalink. 2019-10-08T03:36:22 < Cracki> it shows some wiggly milliamps 2019-10-08T03:36:50 < Cracki> but now I think I crashed it because the jlink (hostside software) has a web interface to configure power on that pin 2019-10-08T03:36:57 -!- learningc [~learningc@2001:e68:5869:c400:5cfa:d5f1:c85c:9769] has quit [Ping timeout: 250 seconds] 2019-10-08T03:38:27 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-08T03:39:20 < Laurenceb> keeek bbc news 2019-10-08T03:39:28 < Laurenceb> >is environmentalism fascist 2019-10-08T03:39:41 < mawk> where 2019-10-08T03:40:22 < Laurenceb> was on bbc 1 2019-10-08T03:41:18 < Cracki> it doesn't detect target freq. I need to tell it that in the trace settings... 2019-10-08T03:41:30 < Cracki> protecting nature is as fascist as it gets 2019-10-08T03:42:01 < Cracki> all these right wingers want to protect nature so much, I keep wondering why commies are for it too 2019-10-08T03:42:59 < Cracki> hell, what's the point of getting an stlink v3 when ozone only works with jlinks, and it works with this cheapass 10 bucks china clone already 2019-10-08T03:44:08 < dongs> yeep 2019-10-08T03:45:15 < Cracki> I looked, the jewlink edu non-mini would be 60 bucks. that's less than anyone here makes in an hour, no-brainer to buy 2019-10-08T03:50:15 < jadew> google is the best at extracting excerpts from wikipedia: "Arianny Celeste is an American ring girl and model. She is best known for being a ring girl for the UFC. She is also a co-host on TV show Overhaulin' and a model featured in magazine Playboy." 2019-10-08T03:50:23 < jadew> that's all I needed to know 2019-10-08T03:52:27 < dongs> pretty sure you didnt need to know this to jerk off, and thats all you do with women 2019-10-08T03:53:29 < jadew> it's not for that, I'm just curious 2019-10-08T03:54:43 < jadew> fake boobs 2019-10-08T03:56:32 < Cracki> hah segger is "investigating" the stink v3 https://forum.segger.com/index.php/Thread/6397-SOLVED-STLink-conversion-utility-for-STLink-v3/ 2019-10-08T03:57:15 < dongs> oh wow 2019-10-08T03:57:17 < dongs> buying that board. 2019-10-08T03:57:24 < dongs> for my mouser order 2019-10-08T03:57:33 < Cracki> boss is gonna think I'm nuts, now I have to tell him to DO buy it 2019-10-08T04:00:25 < dongs> Cracki: nice 2019-10-08T04:00:29 < dongs> added the nucleo to my order 2019-10-08T04:00:32 < dongs> and now i have free shipping 2019-10-08T04:00:46 < Cracki> why you buying a nucleo? 2019-10-08T04:00:46 < jadew> what board? 2019-10-08T04:00:56 < dongs> jadew: the one in that thread 2019-10-08T04:01:00 < dongs> with stdick v3 2019-10-08T04:01:10 < Cracki> and g4 target 2019-10-08T04:01:11 < dongs> NUCLEO-G474RE 2019-10-08T04:01:36 < Cracki> anyway, you can order stinkv3 mini as standalone of course, if you don't care for the rest of the nucleo 2019-10-08T04:01:58 < dongs> im totally fine with the nucleo 2019-10-08T04:02:27 < Cracki> manual for the nucleo would probably include schematics for the v3 too... 2019-10-08T04:02:45 < dongs> there's no point having schematics, that F7 bga is $15 by itself 2019-10-08T04:02:51 < dongs> nobody is gonna try to clone it 2019-10-08T04:03:04 < Cracki> sure but I'm thinking that's how segger is gonna have to work with it 2019-10-08T04:03:09 < dongs> UPS Worldwide Express Saver $0.00 2019-10-08T04:03:13 < dongs> oh, right 2019-10-08T04:03:25 < Cracki> but only them, and maybe the chinese 2019-10-08T04:03:38 < dongs> https://www.st.com/resource/en/user_manual/dm00556337.pdf 2019-10-08T04:03:42 < Cracki> I do wonder if the chinese can beat that price... st must be selling this at a loss 2019-10-08T04:04:06 < Cracki> or making insane quantities to drive cost down 2019-10-08T04:05:03 < dongs> whoa 2019-10-08T04:05:06 < dongs> there's altium files 2019-10-08T04:05:08 < dongs> for the nucleo 2019-10-08T04:05:08 < Cracki> no schematic I can see 2019-10-08T04:05:14 < Cracki> oooh 2019-10-08T04:06:44 < dongs> https://i.imgur.com/VEs66Ol.png 2019-10-08T04:06:51 < dongs> theres your stdick v3 2019-10-08T04:07:05 < Cracki> STM32F723IEK6, 7.08 EUR @1k 2019-10-08T04:07:23 < Cracki> saved 2019-10-08T04:07:25 < Cracki> thx 2019-10-08T04:07:35 < dongs> 4L board 2019-10-08T04:07:37 < dongs> on the nucleo 2019-10-08T04:08:19 < dongs> didnt need 4L at least for the BGA part 2019-10-08T04:08:23 < dongs> most pins are unused anyway heh 2019-10-08T04:09:21 < Cracki> ah, they have pdf exports of altium under resources for the nucleo 2019-10-08T04:09:36 < dongs> ya? idk, i exported pdf but i was too lazy to find some filehost to throw it on 2019-10-08T04:10:26 < Cracki> didn't find the altium files themselves yet, prolly the next item in the table 2019-10-08T04:10:46 < Cracki> hah, first and I looked at the third 2019-10-08T04:10:47 < dongs> o yeah schematic pdf is 3rd under hw resources 2019-10-08T04:10:50 < dongs> yeah 2019-10-08T04:10:54 < dongs> boardf design project files 2019-10-08T04:11:23 < Cracki> I don't quite trust their ST-LINKV3E_SWD symbol. that might be just an abbreviated F7, or contain more stuff 2019-10-08T04:11:43 < Cracki> I'm sure the altium files contain the whole truth 2019-10-08T04:11:47 < dongs> no its whole thing 2019-10-08T04:12:00 < dongs> literally almost nothing is connected 2019-10-08T04:13:04 < dongs> blob:https://imgur.com/eb92d6c1-0fd6-4315-80c2-4492f3bc4dc9 2019-10-08T04:13:06 < dongs> fuck 2019-10-08T04:13:09 < dongs> https://i.imgur.com/4ECJSBe.png 2019-10-08T04:13:29 < Cracki> relaxed 2019-10-08T04:13:54 < dongs> o that thing is 0.65mm pitch 2019-10-08T04:13:55 < dongs> not bad 2019-10-08T04:14:30 < Laurenceb> >Sweet but lonely Seth spends his days working in an animal shelter. In a hopeless daze, he has a chance encounter with beautiful young waitress Holly, who awakens something within him. Obsessed, he tries everything to win her over. Time and again, she rejects him 2019-10-08T04:14:52 < Laurenceb> wtf google I wanted Polyethylene terephthalate 2019-10-08T04:14:54 < Cracki> then he turns into a werewolf and knots her 2019-10-08T04:15:13 < Laurenceb> duckduckgo ftw 2019-10-08T04:16:05 < Cracki> duckdickgo 2019-10-08T04:20:48 < kakimir32> 420 2019-10-08T04:28:19 < Cracki> maaan ozone's plotting is so much nicer and hasslefree than cubeshit ide 2019-10-08T04:28:33 < Cracki> I think I'm gonna have to try keil now 2019-10-08T04:30:00 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T04:31:57 < englishman> arent there always altidumb files for nucleo 2019-10-08T04:34:08 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-08T04:54:32 < qyx> whats modern today for DIN rail mounted devices - 7segment LEDs or OLED displays 2019-10-08T04:58:13 < Cracki> you want it to last 10 years? then maybe not oled, they age 2019-10-08T04:58:50 < Cracki> 7seg sounds a little old school tho quite serviceable. if you need graphics, I'd pick good old LCD 2019-10-08T04:59:48 < Cracki> a customer sent in pieces of some fitness/training device because it's from around 2006 and the oled screens had worn out 2019-10-08T05:00:36 < Cracki> he wanted to know if we could find replacements; we couldn't. we offered to reverse engineer the display protocol (no docs ofc) and adapt to available displays. 2019-10-08T05:13:39 < Cracki> must have been early oled displays... they all claim now that their shit lasts for half a century 2019-10-08T05:15:05 < Cracki> i'd still go for something that can be seen in factory lighting or daylight, so either bright led 7seg, or lcd with backlight 2019-10-08T05:17:05 < Cracki> https://en.wikipedia.org/wiki/OLED#Lifespan 2019-10-08T05:35:39 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-08T05:38:16 < R2COM> sup niggas 2019-10-08T05:43:51 < brdb> hard R dude 2019-10-08T05:47:29 < brdb> i might be having a brain fart here, but using libopencm3 on the stm32f0 discovery board i'm trying to figure out how to have timer 3 (leds LD3 & LD4 for GPIOC pins 8 & 9) tie to those pins using the alternate function (TIM3 CH3 && CH4) 2019-10-08T05:47:48 < brdb> forgetting to set something here for somehow having the gpio tie to the compare 2019-10-08T05:55:04 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-08T06:28:13 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 245 seconds] 2019-10-08T06:33:07 < dongs> did you set AF to timer? 2019-10-08T06:33:14 < dongs> did you configure output compare channels in timer? 2019-10-08T06:33:33 < dongs> did you enable timer to control PWM (i dont remember if tim3 needs that or not, but i always do it just in case anyway (BDTR register) 2019-10-08T06:34:30 < dongs> https://business.kioxia.com/en-us/news/2019/corp-20191001-1.html ol wut 2019-10-08T06:58:40 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-08T07:08:21 < brdb> dongs: yeah forgot to set the output compare to "output" 2019-10-08T07:08:25 < brdb> i said that completely wrong, but yes 2019-10-08T07:08:33 < brdb> the BDTR is such a beginner's gotcha 2019-10-08T07:08:48 < brdb> first time around that really had me chasing ghosts for the longest time 2019-10-08T07:11:10 -!- User_ [~learningc@2001:e68:5869:c400:598f:599d:27de:c838] has quit [Read error: Connection reset by peer] 2019-10-08T07:20:26 < R2COM> dongs i found another disco F4 board, maybe i can try upload jlink on that one 2019-10-08T07:20:36 < R2COM> i wonder if jlink cares about what device it is on 2019-10-08T07:44:23 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 276 seconds] 2019-10-08T07:45:22 < bitmask> anyone up?? 2019-10-08T07:45:46 < bitmask> how do you do the land pattern for the tab of a TO263-5 2019-10-08T07:47:53 < bitmask> TI's datasheet shows a weird T shape, others show a rectangle. If you do a rectangle how does that work with solder paste for the area that is black/epoxy 2019-10-08T07:56:04 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-08T07:59:33 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 265 seconds] 2019-10-08T07:59:33 -!- day__ is now known as day 2019-10-08T08:09:32 < rajkosto> it stays on the pcb only 2019-10-08T08:15:10 < R2COM> so am i supposed to use gitignore.io to add ignore files to source control for VisualStudio? 2019-10-08T08:15:18 < R2COM> seems like it would result in less mess 2019-10-08T08:17:07 < R2COM> umm yes i do 2019-10-08T08:24:49 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Read error: Connection reset by peer] 2019-10-08T08:27:58 < dongs> brdb: hehe yeah 2019-10-08T08:28:29 < dongs> the fuck is gitignore.io 2019-10-08T08:29:01 < dongs> i just add .vs, Debug/* *.vcxproj.user or whatevr to ignore 2019-10-08T08:29:15 < dongs> you can just rightclick that shit in sourcetree and 'add to ignore' then commit the stuff. 2019-10-08T08:29:49 < R2COM> well, you go to gitignore.io, then enter name of your IDE, it generates a file, then you save that file with name ".gitignore", this way, sourcetreee will ignore the trash which does not need to be logged or saved (some auto generated files and shit for most IDEs) 2019-10-08T08:30:20 < R2COM> its just this is big database for most IDEs and now it covers more you just do this step and dont add manually 2019-10-08T08:30:57 < R2COM> by the way this sourcetree is awesome! it works so well, im getting more used to it now 2019-10-08T08:30:58 < dongs> looks completely uslesss 2019-10-08T08:31:09 < R2COM> that command line source control is really for niggers 2019-10-08T08:31:25 < dongs> http://bcas.tv/paste/results/Fygqw721.html 2019-10-08T08:31:32 < dongs> this is literally my entire .gitignore for a vs project 2019-10-08T08:31:57 < dongs> i didnt bother adding *.vcxproj.user to that one but it can also go 2019-10-08T08:32:10 < R2COM> http://gitignore.io/api/visualstudio 2019-10-08T08:32:23 < dongs> yeah waaaaaaaaaaaaaaaay too much unnecessar shit 2019-10-08T08:32:37 < dongs> I've never seen 95% of those files 2019-10-08T08:32:40 < dongs> in any of my porjects 2019-10-08T08:32:56 < R2COM> you ignore Debug and Release? 2019-10-08T08:32:59 < R2COM> actualkly makes sense 2019-10-08T08:33:08 < dongs> why wouldnt i, those are all binary builds 2019-10-08T08:33:08 < R2COM> this file doesnt cover those folders lol 2019-10-08T08:33:14 < dongs> uit does 2019-10-08T08:33:18 < dongs> # Build results 2019-10-08T08:33:19 < dongs> [Dd]ebug/ 2019-10-08T08:33:30 < dongs> it uses some case match shti but I use windows so this is not a problem 2019-10-08T08:33:35 < R2COM> oh 2019-10-08T08:33:36 < R2COM> there you go 2019-10-08T08:44:54 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T08:54:40 < dongs> apparently that site is jsut stealing from https://github.com/github/gitignore 2019-10-08T08:59:11 < R2COM> do you first "Create" in sourcetreee a folder and then in that folder create files with IDE? 2019-10-08T08:59:14 < R2COM> or the other way 2019-10-08T08:59:30 < R2COM> cuz sourcetree cant create or add on existing IDE's full of files folder 2019-10-08T09:00:19 < Steffanx> Lol never heard of if mawk 2019-10-08T09:00:41 < mawk> really ?? 2019-10-08T09:00:47 < mawk> the guy in it is famous 2019-10-08T09:00:56 < dongs> R2PRO, 'creating' a repository in existing folder with files absolutely works 2019-10-08T09:01:05 < mawk> they even talk of them on the radio 2019-10-08T09:01:07 < dongs> it asks you "folder not empty, create anyway"? but you say yes adn continue. 2019-10-08T09:01:30 < dongs> and then you can right click stuff to make .gitignore, select all teh shit you want to commit, and make initial commit with all that stuff in it. 2019-10-08T09:03:26 < R2COM> no 2019-10-08T09:03:29 < R2COM> doesnt work 2019-10-08T09:03:32 < R2COM> doensnt offer such option 2019-10-08T09:03:33 < dongs> of course it does 2019-10-08T09:03:33 < R2COM> https://i.imgur.com/jr3jhH2.jpg 2019-10-08T09:03:45 < dongs> "clone" 2019-10-08T09:03:50 < dongs> thats not waht im doing 2019-10-08T09:04:03 < dongs> i dont bother linking shit to github for 95% of my projects 2019-10-08T09:04:06 < dongs> they're just local git repositories. 2019-10-08T09:04:18 < R2COM> ahh 2019-10-08T09:04:57 < dongs> so i use the create button 2019-10-08T09:05:04 < dongs> IF i want to later post the stuff on shithub 2019-10-08T09:05:14 < R2COM> i used create button too, but checked the checkbox "create repository on account" 2019-10-08T09:05:19 < dongs> i click the 'repository from existing code' in shithub gui and follow instructions 2019-10-08T09:05:25 < dongs> yeah, ididnt bother weith that 2019-10-08T09:05:29 < R2COM> k\ 2019-10-08T09:07:09 < Steffanx> Just looked it up, from the photo there seems to be no man in it, mawk :p 2019-10-08T09:11:52 < R2COM> ok i just created into existing locally only 2019-10-08T09:12:07 < R2COM> but then all i did is created on github website, and linked to it 2019-10-08T09:12:09 < R2COM> works! 2019-10-08T09:12:19 < dongs> yeah. 2019-10-08T09:17:13 < R2COM> dongs do you also use this shit with your PCB soft? or it doesnt work well there? 2019-10-08T09:17:23 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 245 seconds] 2019-10-08T09:17:35 < dongs> nah, too much binary shit there. no point 2019-10-08T09:17:49 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T09:17:53 < dongs> when I make PCB revisions i just copy shit, rename .proj to rev1 / rev2/whatever and keep working on that 2019-10-08T09:18:19 < R2COM> thats what i do now 2019-10-08T09:18:41 < dongs> altium has some version control shit but its fuckign SVN 2019-10-08T09:18:43 < dongs> so hell no 2019-10-08T09:30:20 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-08T09:34:20 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-08T09:56:47 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-08T10:03:54 < jpa-> zyp: digikey doesn't stock the no legs version of this tag connect cable, do you think i can just snap off the legs to make it compatible with a no-legs pcb? https://www.digikey.com/product-detail/en/tag-connect-llc/TC2030-CTX/TC2030-CTX-ND/5023324 2019-10-08T10:07:08 < R2COM> win10 tries to delete the Segger jlink utility for flashint stlink 2019-10-08T10:07:11 < R2COM> annoying piece of shit 2019-10-08T10:07:15 < dongs> wat lol 2019-10-08T10:07:17 < R2COM> how do i prevent it from doing it 2019-10-08T10:07:21 < dongs> it thinks its a virus? 2019-10-08T10:07:25 < R2COM> im starting that segger reflasher 2019-10-08T10:07:25 < R2COM> yes 2019-10-08T10:07:28 < dongs> lmaoo 2019-10-08T10:07:39 < dongs> goto settings/dickfender and temporarily disable it i guess 2019-10-08T10:10:42 -!- jly [uid355225@gateway/web/irccloud.com/x-cksgkwzfknvqyitc] has joined ##stm32 2019-10-08T10:11:54 < R2COM> Identifying ST-LINK variant...O.K.: ST-LINK/V2 2019-10-08T10:11:54 < R2COM> Performing firmware update...O.K. 2019-10-08T10:11:59 < R2COM> hmm 2019-10-08T10:12:15 < dongs> sounds like it worked? 2019-10-08T10:12:36 < R2COM> yeah 2019-10-08T10:12:41 < R2COM> so i download ozone now? 2019-10-08T10:13:01 < R2COM> hmm 2019-10-08T10:13:07 < dongs> yes 2019-10-08T10:13:39 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-08T10:15:16 < R2COM> so without connecting to target device can i poke ozone and make sure it now identifies newly flashed disco board as jlink device? 2019-10-08T10:15:55 < dongs> just open any jlink thing 2019-10-08T10:16:07 < dongs> like j-link configurator 2019-10-08T10:16:51 < jly> kike`d 2019-10-08T10:17:09 < R2COM> i see now: 2019-10-08T10:17:15 < R2COM> "emulators connected via jlink" 2019-10-08T10:17:19 < R2COM> and says stlink jlink 2019-10-08T10:17:19 < R2COM> v2 2019-10-08T10:17:23 < dongs> vua USB yeah 2019-10-08T10:17:37 < R2COM> yes via usb 2019-10-08T10:17:48 < R2COM> soo...technically good to go then? 2019-10-08T10:18:08 < R2COM> hmm 2019-10-08T10:18:25 < R2COM> okay 2019-10-08T10:18:29 < R2COM> the rest will test tomorrow 2019-10-08T10:18:30 < dongs> yeah open ozone, select proper target proc, and point it to your .elf file from build 2019-10-08T10:18:36 < R2COM> but 2019-10-08T10:18:46 < R2COM> for that i need to have electrical connection right 2019-10-08T10:18:57 < dongs> well yeah 2019-10-08T10:20:35 < R2COM> or wait, it can run gdbserver too? 2019-10-08T10:20:57 < dongs> yeah 2019-10-08T10:21:09 < R2COM> but then you wont be using features of ozone right? 2019-10-08T10:21:10 < R2COM> or? 2019-10-08T10:21:19 < dongs> unrelated 2019-10-08T10:23:48 < R2COM> oh darn its past midnight now 2019-10-08T10:37:10 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-08T10:55:02 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has joined ##stm32 2019-10-08T11:10:11 < jadew> ffs... chinese companies are not below scamming you 2019-10-08T11:10:43 < jadew> I ordered $150 worth of connectors last month 2019-10-08T11:11:14 < jadew> lead time was 5-10 days and apparently they didn't have any of them on stock (there were just a couple of connectors) 2019-10-08T11:11:30 < jadew> weird for a company that makes connector not to have samples available on demand 2019-10-08T11:12:01 < jadew> anyway, 20+ days later they still didn't ship them and according to them, still not in production 2019-10-08T11:12:06 < jadew> so I asked for a refund 2019-10-08T11:12:21 < jadew> 10+ days later guess what I receive, after contacting them again? 2019-10-08T11:12:54 < day> open a twitter acc in their name and spam pooh memes 2019-10-08T11:13:30 < jadew> a canceled "Bill From" thing that shows like a payment in my paypal (for less money than I sent - they subtracted the paypal fee) 2019-10-08T11:13:46 < jadew> and a payment for products for the same ammount 2019-10-08T11:14:24 < jadew> which looked like they paid me twice that smaller amount, when I checked the activity log 2019-10-08T11:14:44 < jadew> so the bottom line was this: 2019-10-08T11:15:13 < jadew> in my account, if I didn't pay attention, it appeared like I received more money than I sent 2019-10-08T11:15:43 < jadew> but in reality, I received a payment for products, for less money, for which I paid the paypal fee too, and which could have been disputed 3 months from now 2019-10-08T11:15:52 < jadew> and they would have won (confirmed by paypal on the phone) 2019-10-08T11:18:35 < jadew> so... it's either that they're extremely incompetent, to the point that they produce high frequency/high quality connectors but don't have ANY on stock, and the financial department is equally incompetent to the point that they can't issue a refund 2019-10-08T11:18:53 < jadew> or that the whole thing was a scam from the get go 2019-10-08T11:20:14 < jadew> I mean... why would they have done that thing that looked like they paid me, when they didn't? 2019-10-08T11:28:00 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-08T11:30:27 < jadew> it feels like it 2019-10-08T11:36:52 -!- dexterlb [~dexterlb@qtrp.org] has quit [Quit: Boing] 2019-10-08T11:37:09 -!- dexterlb [~dexterlb@qtrp.org] has joined ##stm32 2019-10-08T11:50:34 < Cracki> get a chinagirl maybe 2019-10-08T11:51:53 < jadew> Haohmaru, https://asianbride.me/countries/chinese-brides 2019-10-08T11:53:10 < jadew> where did you see that? 2019-10-08T11:53:40 < jadew> ah 2019-10-08T11:53:49 < jadew> some of those chicks are hot 2019-10-08T11:54:10 < jadew> I'd "marry" them once or twice 2019-10-08T11:56:38 -!- dexterlb [~dexterlb@qtrp.org] has quit [Quit: Boing] 2019-10-08T11:57:47 < jadew> https://www.bbc.com/news/world-asia-china-47615966 2019-10-08T11:58:08 < jadew> "China's community of goths is coming together in protest online after a woman was made to remove her make-up before being allowed to enter a busy subway." 2019-10-08T11:58:10 < jadew> lol 2019-10-08T12:01:29 < qyx> heh 2019-10-08T12:06:43 < dongs> so i got a chink lcd with SPI/PAR pin 2019-10-08T12:06:51 < dongs> any gueses whci direction makes it SPI? 2019-10-08T12:06:52 < dongs> high or low 2019-10-08T12:06:57 < dongs> fucking retarded dataasheet doenst mention of course 2019-10-08T12:07:10 < jadew> heh 2019-10-08T12:07:12 < dongs> controller has the usual IM0..3 thing but no bit patterns match there that would turn it into parallel with just one bit change 2019-10-08T12:11:06 < Cracki> fuzz it 2019-10-08T12:11:37 < Cracki> tardweeno, port = rand() 2019-10-08T12:12:10 -!- dexterlb [~dexterlb@qtrp.org] has joined ##stm32 2019-10-08T12:54:36 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-08T13:09:27 < karlp> Cracki: got anything newer on oled lifespan? 2019-10-08T13:11:10 < karlp> fuck me, gitignore.io is a solution looking for a problem. 2019-10-08T13:30:26 -!- jly [uid355225@gateway/web/irccloud.com/x-cksgkwzfknvqyitc] has quit [Quit: Connection closed for inactivity] 2019-10-08T13:39:03 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 245 seconds] 2019-10-08T13:52:12 < mitrax> i was going to use an oled as a replacement for a lcd in a product, but the screen i tested exhibits a sort of afterglow for some of the pixel columns when you turn all the pixels on for a few minutes then turn them off, instead of a black background you can see what looks like a faint barcode, it's not a burn in, not sure if it's a common phenomenon but it doesn't make very comfortable with 2019-10-08T13:52:12 < mitrax> going oled, doesn't seem like the technology is reliable/mature enough 2019-10-08T14:05:08 < Thorn> how do I power a backlight for a display that has leds in parallel rather than in series? the voltage drop is 3.2...3.4V, power comes from a li-ion battery 2019-10-08T14:05:46 < Thorn> are there any buck-boost regulators with low Vfb on lcsc attn dongs 2019-10-08T14:07:29 < zyp> jpa-, yeah, I think that'd work out fine 2019-10-08T14:07:54 < dongs> karlp: right? who the fuck needs that shit automated 2019-10-08T14:08:07 < dongs> like, if youer working on a proj y ou know EXACTLY what needs to be ignored 2019-10-08T14:08:15 < dongs> you fucking made the shit 2019-10-08T14:08:56 < dongs> well then you might as well goto gitignore.io and copypaste some useless shit 2019-10-08T14:09:02 < zyp> haha 2019-10-08T14:26:30 < PaulFertser> mitrax: that might be related with the settings you used for OLED, have you tried different values for some of those parameters you can tweak? I think I remember seeing something odd like that. 2019-10-08T14:34:29 < mitrax> PaulFertser: i only messed with the contrast, i didn't try messing with the pre-charge periode / display clock yet 2019-10-08T14:34:31 < dongs> http://www.lcm-zsxy.com/en/50-hd-ips_52067.html what the fuck 2019-10-08T14:34:38 < dongs> lol @ 'detailed introduction' 2019-10-08T14:35:17 < zyp> heh 2019-10-08T14:37:55 < dongs> blah chinese version is same 2019-10-08T14:38:19 < dongs> i think whole site is a scam 2019-10-08T14:38:23 < dongs> none of hte links actually work 2019-10-08T14:38:36 < dongs> ill just email this yangxiangguowahtevershit@chink.com 2019-10-08T14:38:37 < dongs> and see 2019-10-08T14:38:47 < mitrax> PaulFertser: http://download.unfiltered.net/oled.png the problem is exagerated on the picture (i brightened it) 2019-10-08T14:39:03 < dongs> whats the problem 2019-10-08T14:39:44 < PaulFertser> mitrax: I see, yes. Not sure why you're getting it, I do not remember any after-glow on several OLED displays I used. 2019-10-08T14:42:17 < mitrax> worst thing is that's a taiwanese screen (winstar), and rather expensive for what it is... should have went with a cheapo chinese brand :) 2019-10-08T14:42:45 < jpa-> mitrax: does it have any relation to what was previously displayed? 2019-10-08T14:43:05 < mitrax> jpa-: no, that's the weird part 2019-10-08T14:43:24 < Thorn> very cheap current shunt amp from lcsc plz 2019-10-08T14:43:29 < jpa-> does it appear on power on & reset display to black, or only after displaying content? 2019-10-08T14:44:12 < mitrax> on power on & reset, it didn't at first, then i displayed a pattern with almost all pixels turned on for two hours or so, and it exhibits the problem since 2019-10-08T14:44:22 < jpa-> it could be the oled "pixel imbalance" thing, some higher end displays have per-pixel or per-column compensation of brightness - but i've never seen that on simple black & white displays 2019-10-08T14:45:16 < jpa-> zyp: thanks :) 2019-10-08T14:45:47 -!- onio [~onio@2a00:23c5:7a01:8600:d8fa:6dd8:1daf:f6a8] has joined ##stm32 2019-10-08T14:46:45 < mitrax> i'll email Winstar and see what they say 2019-10-08T15:35:22 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-08T15:40:31 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-08T15:47:46 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-08T15:56:36 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-08T16:00:48 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-08T16:01:03 < karlp> thorn, istr one of the ti INA182/184 maybe was available in china? 2019-10-08T16:01:10 < karlp> but I'm happy to hear of another one :) 2019-10-08T16:01:58 < dongs> Thorn: no such thing, if you go to opamp category on lcsc and filter out all ';murican makers you only get a dozen chink ones 2019-10-08T16:02:01 < dongs> adn they're all pretty shit 2019-10-08T16:03:05 < dongs> so use something murican and cheap 2019-10-08T16:03:57 < dongs> https://lcsc.com/product-detail/Others_Texas-Instruments_INA180A2IDBVR_Texas-Instruments-TI-INA180A2IDBVR_C192764.html this is cheapest in current sense category 2019-10-08T16:07:17 < emeryth> INA199 is cheaper actually 2019-10-08T16:08:34 < karlp> what are my chances of a 40V+ fet that can be switched sanely at 3.3v? ao3400 is lyfe, but only goes to 30V, which is uncomfortably close to 24V... 2019-10-08T16:08:53 < dongs> you mean direct gate? 2019-10-08T16:08:57 < karlp> yeah. 2019-10-08T16:08:59 < dongs> why not use a cheap ass gate driver 2019-10-08T16:09:08 < karlp> fuck that. 2019-10-08T16:09:12 < karlp> this is just a power on /off switch 2019-10-08T16:09:20 < dongs> oh, like for whole system? 2019-10-08T16:09:31 < karlp> I can robably find a 5v pin easier than I'll find a gate driver 2019-10-08T16:09:33 < dongs> soft-power-off 2019-10-08T16:09:41 < karlp> yeah, for connected systems that are powered by a bus cable from me. 2019-10-08T16:09:59 < dongs> i think in that case 30V fet would be just fine 2019-10-08T16:10:13 < karlp> _probably_ yes, 2019-10-08T16:10:17 < dongs> not like youre pwming it into coils and getting huge high current pulses through it 2019-10-08T16:13:25 < dongs> huh, *single* logic level fet category on digikey is onyl 3 parts. 2019-10-08T16:13:25 < karlp> "High density cell design for extremely low R DS(ON)" 2019-10-08T16:13:28 < dongs> dual are 100s... 2019-10-08T16:13:33 < karlp> (R DS(ON) = 10Ω @ V GS = 4.5 V) 2019-10-08T16:13:38 < karlp> _TEN_ fucking whole ohms! 2019-10-08T16:13:40 < dongs> haha 10ohm 2019-10-08T16:14:44 < dongs> AO3422 ? 2019-10-08T16:15:24 < englishman> why not use a power on off switch 2019-10-08T16:15:37 < karlp> englishman: show me a part so I know what you're talking about? 2019-10-08T16:15:42 < englishman> dps1133 2019-10-08T16:16:08 < karlp> how'd you find ao3422 dongs? ao's website was a trainwreck for me. 2019-10-08T16:16:16 < dongs> digikey filtering 2019-10-08T16:16:29 < dongs> vgs <= 3V, vds >= 40V, a&o manufacturer 2019-10-08T16:16:32 < dongs> that was like 1st result 2019-10-08T16:16:57 < dongs> you didnt specify any other parameters so I just pasted that one 2019-10-08T16:17:13 < karlp> yeah, I was looking directly on aosemi... 2019-10-08T16:18:18 -!- sandeepkr_ [~sandeepkr@2a03:b0c0:2:d0::cac:7001] has joined ##stm32 2019-10-08T16:22:02 < karlp> ao3422 looks perfect. 2019-10-08T16:22:21 < dongs> shit's like 8c on lcsc in qty 50, your work is done 2019-10-08T16:23:03 < dongs> weird, its listed as "international brands" 2019-10-08T16:23:09 < dongs> i was prety sure ao was chink 2019-10-08T16:24:37 < karlp> dps1133 looks kinda nice englishman, and cheaper than I was expecting, but way not what I want really. 2019-10-08T16:36:16 < karlp> Si2356 also cool. 2019-10-08T16:36:24 < karlp> man, there's some garbage parts there in the same list. 2019-10-08T16:38:00 < karlp> 2n7002 for instance 2019-10-08T16:38:39 < karlp> it's a fet, but yeah.. 2019-10-08T16:41:24 < dongs> 2n7002 is truly shit yeah 2019-10-08T16:41:28 < dongs> but its like $0.01 2019-10-08T17:06:55 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-08T17:11:51 -!- kakimir32 [575d220c@87-93-34-12.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-08T17:28:13 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-08T17:32:26 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-08T17:47:59 < karlp> I don't think I can use this anyway, Vgs max is 12V, so can't actyually drive it off 3.3v/0 anyway, when Vs is 24V. 2019-10-08T17:51:36 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-08T17:51:44 < bitmask> heyo 2019-10-08T17:51:55 < bitmask> I asked last night, dunno if anyone answered so i'll ask again... 2019-10-08T17:52:03 < bitmask> how do you do the land pattern for the tab of a TO263-5 2019-10-08T17:52:07 < bitmask> TI's datasheet shows a weird T shape, others show a rectangle. If you do a rectangle how does that work with solder paste for the area that is black/epoxy 2019-10-08T17:53:03 < bitmask> I think I wanna do the T shape but I'm not sure how in altium 2019-10-08T17:53:37 < bitmask> and I read the paste layer should have only 1/2 the area uncovered 2019-10-08T17:53:47 < bitmask> to prevent tombstoning 2019-10-08T17:55:39 < fenugrec> karlp, NCE40P05Y is pmos but its complementary might work 2019-10-08T17:58:05 < karlp> pmos is probably easier for my brane anyway. I feel like this is "easy" but I'm not doing something right on falstad for modelling it as a pass transistor anwyay 2019-10-08T18:02:00 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T18:14:17 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-08T18:15:16 < bitmask> hubbily bubbily 2019-10-08T18:15:26 < bitmask> ohhhhhhhh what am i doing 2019-10-08T18:17:15 < bitmask> Is 1-3mA enough to turn on a smd led? 2019-10-08T18:24:37 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-08T18:32:00 < karlp> normally 2019-10-08T18:33:11 < bitmask> i'll have to test these 2019-10-08T18:33:20 < bitmask> I gotta figure out this landing pad 2019-10-08T18:37:01 < Thorn> even 1µA is visible in low ambient light 2019-10-08T18:37:27 < Thorn> (depending on the led) 2019-10-08T18:38:01 < bitmask> any idea on the T shaped TO263-5 Tab landing pattern? 2019-10-08T18:38:46 < zyp> bitmask, read a datasheet, follow the recommended pattern? 2019-10-08T18:39:48 < bitmask> there are two different recommendations (depending on manufacturer), one is to make it the T shape, the other is to make it a rectangle that covers almost the whole back, including parts of the black epoxy covered bits 2019-10-08T18:39:52 < bitmask> thats what im not sure about 2019-10-08T18:40:16 < bitmask> is it ok to have solder between the copper of the board and the epoxy of the IC 2019-10-08T18:40:20 < zyp> cool, then just pick whichever you prefer because they both obviously will work then 2019-10-08T18:40:22 < bitmask> will it flow right? 2019-10-08T18:40:33 < bitmask> I want to understand how it works 2019-10-08T18:41:29 < zyp> solder doesn't adhere to epoxy 2019-10-08T18:41:39 < bitmask> but it does to the copper underneath... 2019-10-08T18:41:57 < zyp> sure 2019-10-08T18:42:03 < bitmask> https://www.onsemi.com/pub/Collateral/LM2596-D.PDF 2019-10-08T18:42:10 < bitmask> the last page shows just a rectangle 2019-10-08T18:42:26 < zyp> looks reasonable 2019-10-08T18:42:28 < bitmask> so thats what im asking, its ok to just let the solder cover the whole area, even the area with black epoxy above? 2019-10-08T18:42:44 < zyp> of course it's ok, otherwise they wouldn't tell you to 2019-10-08T18:42:58 < bitmask> haha ok, just wanted to verify 2019-10-08T18:43:05 < bitmask> seems odd is all 2019-10-08T18:43:34 < zyp> surface tension will pull most of the solder to where it can adhere both above and below 2019-10-08T18:43:38 < karlp> I can't pull up the gate high enough to turn it off without having that voltage on the gpio pin. 2019-10-08T18:43:47 < bitmask> I see 2019-10-08T18:43:48 < zyp> karlp, what are you trying to do? 2019-10-08T18:43:50 * karlp has overcomplicated life again and gotten distracted 2019-10-08T18:44:11 < karlp> want to switch a 24V power line from a 3.3v gpio 2019-10-08T18:44:16 < zyp> karlp, I haven't followed your entire problem, but it sounds like you wanna do high side switching 2019-10-08T18:44:19 < karlp> I do. 2019-10-08T18:44:37 < zyp> and high side switches are too expensive? 2019-10-08T18:44:58 < Ultrasauce> just drive the gate with an nfet and a pull up 2019-10-08T18:44:59 < karlp> I was under the impression that this should have been "just" a fet and a resistor/diode, but... 2019-10-08T18:45:22 < karlp> yeah, everything seems to just use a second nfet to drive it. 2019-10-08T18:45:29 < zyp> I mean, the ghetto solution is a pfet with a pullup to turn it off, then a nfet to pull it down to turn it on 2019-10-08T18:46:11 < qyx> wut, use a high side smart switch 2019-10-08T18:46:14 < zyp> but I would look into just using a power switch 2019-10-08T18:46:31 < qyx> or are you targetting china market with dongs priced china parts? 2019-10-08T18:46:39 < zyp> they exist and my impression is that they can be affordable enough that you don't have to overthink it 2019-10-08T18:51:32 < karlp> NCV7450 is ~1ohm, and uses ~1mA, and is ~$1. I mean, yeah, it's _ok_ but doesn't really sound awesome. 2019-10-08T18:52:58 < zyp> https://www.onsemi.com/pub/Collateral/NCV7450-D.PDF <- this? 2019-10-08T18:53:50 < karlp> yeah, I mean, it would work I guess. 2019-10-08T18:54:06 < karlp> neft+pfet+R works now that I've used the nfet as well, 2019-10-08T18:54:16 < karlp> was trying to fiddle it with just pfet 2019-10-08T18:54:19 < zyp> well, uh, that shit is a lot more than a load switch 2019-10-08T18:54:38 < zyp> no wonder it doesn't sound awesome if you're not gonna make use of all the other shit 2019-10-08T18:54:40 < karlp> well, it's the first one I found in the load switches category 2019-10-08T18:54:45 < karlp> 8450 sorry, not 7450 2019-10-08T18:55:04 < zyp> ah, that looks more reasonable 2019-10-08T18:55:07 < karlp> yeah, 7450 looks nuts :) 2019-10-08T18:56:34 < zyp> by the way, what's the typical Rds_ON of a typical pfet nowadays? 2019-10-08T18:56:56 < qyx> karlp: http://www.ti.com/product/TPS1H000-Q1 no gud? 2019-10-08T18:57:07 < qyx> 40V, 1R, <$1 2019-10-08T18:57:20 < karlp> zyp: ~100mOhm 2019-10-08T18:57:58 < qyx> with adjustable current limit 2019-10-08T18:58:00 < qyx> and fault output 2019-10-08T18:58:33 < zyp> what voltage do you need? 2019-10-08T18:58:38 < zyp> and current? 2019-10-08T18:58:54 < bitmask> would you order an HGSEMI LM2595 for 74 cents or the TI for $2.55 2019-10-08T19:00:54 < karlp> qyx: that part is ~similar to the onsemi one really. 2019-10-08T19:01:27 < karlp> zyp: 12-24V plus margins, so like >30, and ~3-5W 2019-10-08T19:01:30 < zyp> hmm, all the cheap stuff is just 5V 2019-10-08T19:01:44 < qyx> single cnahhel only? 2019-10-08T19:01:45 < qyx> channel 2019-10-08T19:01:51 < karlp> eyah, shit heaps of 5V switches from usb shits 2019-10-08T19:01:54 < karlp> yeah, signle channel 2019-10-08T19:02:04 < karlp> two fets and an resistor is looking good. 2019-10-08T19:02:15 < karlp> it' 2019-10-08T19:02:36 < karlp> s not even needing any sort of response time, just "be able to power cycle things maybe, might be nice" 2019-10-08T19:03:12 < qyx> check hotswap controllers too 2019-10-08T19:03:20 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has quit [Remote host closed the connection] 2019-10-08T19:03:27 < qyx> they have a charge pump to drive n-fet directly 2019-10-08T19:03:32 < qyx> and some sort of overcurrent and stuff 2019-10-08T19:04:58 < Ultrasauce> you need a second resistor to limit Vgs on the pfet, no? 2019-10-08T19:05:32 < qyx> that too 2019-10-08T19:05:34 < qyx> or a zener 2019-10-08T19:05:50 < qyx> or select a fet with >~30V Vgs 2019-10-08T19:05:56 < qyx> if there are any 2019-10-08T19:10:32 < karlp> Ultrasauce: where does that R go? 2019-10-08T19:11:26 < qyx> if you pull the p-fet gate to GND with a n-fet 2019-10-08T19:11:42 < qyx> you will get 24V Vgs, which is not so good 2019-10-08T19:11:55 < karlp> yeah, that's what I've got now. 2019-10-08T19:12:17 < karlp> ah, got it 2019-10-08T19:12:20 < qyx> so place a resistor between p-fet gate and it's source 2019-10-08T19:12:23 < karlp> resistor on that ground path 2019-10-08T19:12:26 < qyx> and a second one from the source to the n-fet 2019-10-08T19:12:38 < qyx> to form a 1:1 divider for example 2019-10-08T19:12:44 < qyx> to get 12V on the gate 2019-10-08T19:12:45 < karlp> yeah, that's limited my Vgs below the threshold 2019-10-08T19:13:09 < zyp> I searched for load switched and sorted by price, and TBD62783AFG came out pretty cheap, despite being a 8ch device 2019-10-08T19:13:36 < zyp> internal structure is apparently just a pfet controlled by a nfet with some resistors and shit :) 2019-10-08T19:15:21 < karlp> heh. yeah, same price as that other one I linked. 2019-10-08T19:15:44 < zyp> yeah, that one came out top in my search too 2019-10-08T19:16:11 < zyp> NCV8461 costs a bit more but has lower Rds_ON 2019-10-08T19:17:48 < zyp> fuckit, with these options I'd also just ghetto it 2019-10-08T19:18:23 < karlp> yeah. 2019-10-08T19:18:41 < qyx> import fuckit 2019-10-08T19:19:32 < qyx> (https://pypi.org/project/fuckit/) 2019-10-08T19:19:40 < zyp> qyx, that looks pretty cool 2019-10-08T19:21:21 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-08T19:24:39 < brdb> for linux, any recommendations on quick and dirty analog simulations? i vaguely recall KiCAD having some beta functionality for this built in, but can't remember 2019-10-08T19:24:47 < brdb> ngspice, perhaps? 2019-10-08T19:26:53 < Ultrasauce> falstad for the real quick and dirty 2019-10-08T19:28:39 < karlp> http://tinyurl.com/y2bxt5mx for falstad url for those playing at home :) 2019-10-08T19:29:27 < Ultrasauce> wtf 2019-10-08T19:29:39 < Ultrasauce> put the nfet below the resistor 2019-10-08T19:29:58 < karlp> what difference does it make? 2019-10-08T19:30:50 < qyx> I would say it will be always turned off? 2019-10-08T19:30:52 < zyp> resistor moves S up from 0V, so V_GS is lower than 3.3V 2019-10-08T19:31:09 < brdb> hey falstad is just what i needed nice 2019-10-08T19:31:12 < karlp> ~same in falstad , just looks prettier 2019-10-08T19:32:55 < karlp> just changes the nfet into linear instead of sat region? 2019-10-08T19:33:24 < Ultrasauce> when its off, Vgs on the nfet is rather large 2019-10-08T19:33:28 < qyx> no, your H=3V3 will be always under Vgate 2019-10-08T19:33:44 < karlp> Ultrasauce:no, look at what falstad calculated 2019-10-08T19:34:24 < Ultrasauce> okay the real reason is it assaults my greater sensibilities 2019-10-08T19:34:36 < Ultrasauce> please revise the aesthetic quality of your driver thanks 2019-10-08T19:35:03 < karlp> falstad doesn't even seem to care whether nfet portion is d/s or s/d 2019-10-08T19:35:26 < karlp> ahh no. it's magically flipping when you rotate the part 2019-10-08T19:35:27 < karlp> hah 2019-10-08T19:36:16 < karlp> honestly, fet below resistor results in 1.2mA through the fet, fet above resistor is ~0. 2019-10-08T19:37:10 < karlp> Ultrasauce: happier? http://tinyurl.com/y3wjcvoc 2019-10-08T19:37:28 < Ultrasauce> no the topological aesthetic 2019-10-08T19:37:59 < zyp> haha 2019-10-08T19:38:10 < karlp> switching R and the nfet does dramatically reduce Vds on the fet, but 20V is well below what the ao3400 cheap nfet can do. 2019-10-08T19:38:17 < zyp> really, resistor below doesn't make any sense, even if it works 2019-10-08T19:38:37 < karlp> zyp: why? falstad says above just means you get current through the neft 2019-10-08T19:38:42 < Ultrasauce> okay look at Vgs on the pfet 2019-10-08T19:38:53 < Ultrasauce> you definitely want more than that 2019-10-08T19:39:00 < zyp> karlp, that's the point of the nfet 2019-10-08T19:39:17 < karlp> Ultrasauce: what's wrong with -1.8? 2019-10-08T19:39:29 < zyp> karlp, if you want less current through the nfet, you reduce the pullup strength 2019-10-08T19:39:37 < Ultrasauce> what pfet do you plan to use? 2019-10-08T19:39:51 < karlp> ok, there we go, got what you were getting at. 2019-10-08T19:39:57 < karlp> threshold was too low on the falstad model. 2019-10-08T19:40:19 < zyp> karlp, the required current through the nfet is given by the pullup size and the required V_GS on the pfet 2019-10-08T19:40:42 < Ultrasauce> you could use 1M resistors just fine, only tradeoff will be switching time 2019-10-08T19:41:09 < Ultrasauce> which for this application (long duty cycle switching) is surely tolerable 2019-10-08T19:41:30 < karlp> oh, 1.2mA was "fine" I just wasn't seeing the problems 2019-10-08T19:41:54 < zyp> decide on a V_GS from the pfet datasheet, decide on a pullup size by pfet gate capacitance, then calculate the other resistor to give desired V_GS and hook that to the drain of the nfet, problem solved 2019-10-08T19:43:03 < qyx> wut zyp lol 2019-10-08T19:43:14 < qyx> pick the first two resistors you have, use them 2019-10-08T19:43:38 < karlp> so I don't even need Vds > 24V for this at all then. 2019-10-08T19:43:48 < zyp> qyx, well, yeah, but somebody cared about the current through the nfet 2019-10-08T19:43:48 < karlp> doh, no of course, for the off state. 2019-10-08T19:43:58 < qyx> no 2019-10-08T19:44:19 < qyx> most fets are reasonably opened with ~5V, most will tolerate ~20V 2019-10-08T19:44:33 < qyx> so you are safe from 10 to 40V with 1:1 div 2019-10-08T19:45:04 < qyx> or check and calculate as zyp suggested, but meh 2019-10-08T19:45:31 < karlp> still need margin on Vds for the offstate. 2019-10-08T19:45:43 < zyp> calculations are not so hard, just need to ballpark it 2019-10-08T19:45:44 < karlp> Vds of 30 is _probably_ ok. 2019-10-08T19:46:08 < qyx> whats te problem with offstate? 2019-10-08T19:46:13 < zyp> there's V_DS limits and then there's V_GS limits 2019-10-08T19:46:22 < zyp> nfet drain will be at 24V when off 2019-10-08T19:46:23 < qyx> you are driving it with 0V, n-fet is surely off 2019-10-08T19:46:34 < brdb> bah falstad wants to input the beta value for nfets >:( 2019-10-08T19:46:37 < brdb> what is this, the 1980s 2019-10-08T19:47:04 < qyx> yeah so Vgs of p-fet is 0V in all cases then 2019-10-08T19:47:12 < qyx> regardless of the voltage 2019-10-08T19:47:16 < zyp> qyx, sure, V_GS is not a problem 2019-10-08T19:47:25 < zyp> but V_DS needs to tolerate >24V 2019-10-08T19:47:37 < zyp> which also shouldn't be a problem 2019-10-08T19:47:43 < qyx> oh yes 2019-10-08T19:48:02 < qyx> but I considered it obvious 2019-10-08T19:48:10 -!- Jibz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-08T19:48:15 < Ultrasauce> you definitely want some good margins on Vds, the current and inductance of the supply are probably significant enough for the odd transient 2019-10-08T19:48:26 -!- Jibz [~jibz@91-166-99-132.subs.proxad.net] has quit [Client Quit] 2019-10-08T19:48:43 -!- Jibz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-08T19:48:50 < aandrew> ip KVMs are super expensive, even just one port ones 2019-10-08T19:48:51 < aandrew> and there are so many "cheap" one port ones but they're just transmitters to a multi-thousand-$ "head" 2019-10-08T19:48:52 < karlp> https://lcsc.com/product-detail/MOSFET_Alpha-Omega-Semicon_AO4614BL_Alpha-Omega-Semicon-AOS-AO4614BL_C68416.html both in the same package, 40V 2019-10-08T19:49:06 < Ultrasauce> nice 2019-10-08T19:49:49 < zyp> aandrew, don't everybody just buy motherboards with IPMI nowadays instead of bothering with an external KVM? 2019-10-08T19:50:00 < qyx> I can send you some 0201 resistors to complete your circuit 2019-10-08T19:50:06 < qyx> there are some spares.. 2019-10-08T19:50:26 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-08T19:50:54 -!- Jibz [~jibz@91-166-99-132.subs.proxad.net] has quit [Client Quit] 2019-10-08T19:51:45 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-08T19:56:54 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-08T20:02:29 -!- renn0xtk9 [~max@2a02:810d:1540:2448:74d6:d85b:5ce2:9b8e] has joined ##stm32 2019-10-08T20:05:51 < aandrew> zyp: yeah unfortunately this is not one of those systems 2019-10-08T20:06:49 < aandrew> my blade server is all IPMI but still has no ipkvm, just serial console 2019-10-08T20:06:53 < aandrew> well IPMI console 2019-10-08T20:07:38 < zyp> aww 2019-10-08T20:08:08 < zyp> serial console helps though 2019-10-08T20:14:15 < aandrew> yes I love serial console but this is particular one is a win10 pc. I use rdp but when something fucks with the networking it's less helpful :-) 2019-10-08T20:15:22 < zyp> better run it in a vm with serial console to the hypervisor :) 2019-10-08T20:15:49 < aandrew> I'd like to but unfortunately this is very specific hardware 2019-10-08T20:15:55 < aandrew> 99.9% of my stuff is virtualized 2019-10-08T20:16:10 < aandrew> I keep meaning to pick up Pebble's qemu'd stm32 for doing virtual stm32 too 2019-10-08T20:19:39 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has quit [Ping timeout: 240 seconds] 2019-10-08T20:43:51 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-08T20:52:51 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T20:57:39 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-08T20:59:25 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-08T21:08:13 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-08T21:16:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-08T21:19:48 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection] 2019-10-08T21:20:42 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-08T21:22:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-08T21:41:09 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-08T21:47:21 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection] 2019-10-08T21:58:39 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-08T21:58:45 < bitmask> yeaaaa boiii 2019-10-08T22:17:42 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-08T22:48:25 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection] 2019-10-08T22:49:22 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 265 seconds] 2019-10-08T22:53:00 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-08T22:57:04 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 264 seconds] 2019-10-08T22:58:10 < Steffanx> ohno, you too bitmask 2019-10-08T22:58:18 < bitmask> me too what 2019-10-08T22:59:15 < Steffanx> what you said like an hour ago 2019-10-08T23:00:00 < bitmask> I was just being obnoxious, someone else says that? 2019-10-08T23:01:37 -!- CheBuzz [~CheBuzz@unaffiliated/chebuzz] has quit [Ping timeout: 244 seconds] 2019-10-08T23:02:22 < bitmask> fo shiz my niz 2019-10-08T23:04:46 < Ultrasauce> https://www.youtube.com/watch?v=u2X7GCUhZJs 2019-10-08T23:04:57 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-08T23:20:20 -!- onio [~onio@2a00:23c5:7a01:8600:d8fa:6dd8:1daf:f6a8] has quit [Quit: Leaving] 2019-10-08T23:20:22 -!- CheBuzz [~CheBuzz@204.77.3.219] has joined ##stm32 2019-10-08T23:20:22 -!- CheBuzz [~CheBuzz@204.77.3.219] has quit [Changing host] 2019-10-08T23:20:22 -!- CheBuzz [~CheBuzz@unaffiliated/chebuzz] has joined ##stm32 2019-10-08T23:26:16 -!- banana is now known as not_banana 2019-10-08T23:28:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-08T23:29:41 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-08T23:41:05 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 265 seconds] 2019-10-08T23:42:11 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-08T23:44:09 < specing> It doesen't seem to say anything in datasheet, but writing what value to BSRR will bridge an open drain pin to gnd? 2019-10-08T23:44:50 < specing> If I'm understanding correctly, writing 0 will do this. The datasheet says that when it is configured as open drain, only the lower nmos is controllable 2019-10-08T23:45:06 < specing> and it makes sense conceptually 2019-10-08T23:45:46 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-08T23:46:23 -!- renn0xtk9 [~max@2a02:810d:1540:2448:74d6:d85b:5ce2:9b8e] has quit [Ping timeout: 276 seconds] 2019-10-08T23:46:42 -!- renn0xtk9 [~max@ip5f5bfcf2.dynamic.kabel-deutschland.de] has joined ##stm32 2019-10-08T23:46:59 -!- renn0xtk9 [~max@ip5f5bfcf2.dynamic.kabel-deutschland.de] has quit [Client Quit] 2019-10-08T23:48:22 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-08T23:49:53 < kakimir32> yes hello 2019-10-08T23:52:03 < Steffanx> Hello welcome 2019-10-08T23:52:06 < Steffanx> Have room --- Day changed Wed Oct 09 2019 2019-10-09T00:01:41 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-09T00:01:59 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 276 seconds] 2019-10-09T00:04:35 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 250 seconds] 2019-10-09T00:48:14 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-09T00:49:26 -!- not_banana is now known as baenana 2019-10-09T00:50:44 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-09T00:51:38 < Laurenceb> arggg 2019-10-09T00:51:46 * Laurenceb just got goatsied 2019-10-09T00:51:50 < Laurenceb> its been a while 2019-10-09T01:09:00 < karlp> thanks for sharing 2019-10-09T01:14:03 < Steffanx> 2019-10-09T01:14:10 < Steffanx> Whops 2019-10-09T01:16:00 < Steffanx> Laurenceb gets goatsied. Joins ##stm32. Not sure if I should feel honoured or depressed. 2019-10-09T01:16:33 < Laurenceb> I was trying to find the new 8chan 2019-10-09T01:17:51 < antto> wut does goetsied mean? 2019-10-09T01:18:09 < antto> will i regret asking? 2019-10-09T01:19:30 < Laurenceb> yes u will 2019-10-09T01:19:52 < Laurenceb> looks like 8kun is a smokescreen 2019-10-09T01:26:47 < Steffanx> Yes you will. YES YOU WILL, antto 2019-10-09T01:27:12 < Steffanx> I wouldn't even show it to Haohmaru 2019-10-09T01:27:54 < antto> if it's some mental filth - keep it away from me 2019-10-09T01:28:48 < Steffanx> Yessir 2019-10-09T01:30:06 < Thorn> https://www.youtube.com/watch?v=hzKb_fV7UB8 2019-10-09T01:30:30 < antto> i ain't clicking, uh-uh 2019-10-09T01:30:47 < antto> nnnnnnnnnnnope 2019-10-09T01:31:23 < Steffanx> Its safu 2019-10-09T01:31:30 < Laurenceb> goatkcd.com 2019-10-09T01:32:03 < antto> enuf internetz for today 2019-10-09T01:33:50 < kakimir32> how do arm machines boot? 2019-10-09T01:34:12 < kakimir32> is there uefi? 2019-10-09T01:35:50 -!- grindhold [~quassel@mail.skarphed.org] has quit [Read error: Connection reset by peer] 2019-10-09T01:36:35 < karlp> can be done. other bootlaoders too 2019-10-09T01:37:04 < Laurenceb> stratosolar actually makes sense 2019-10-09T01:37:25 < Laurenceb> also stratosolar guy is the guy behind transmeta 2019-10-09T01:40:38 < karlp> you're not concerned about the weight of the cables for the weights at all? 2019-10-09T01:41:02 < Laurenceb> that part seems to make sense 2019-10-09T01:44:19 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-09T02:31:29 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 276 seconds] 2019-10-09T02:34:55 < Laurenceb> https://www.researchgate.net/publication/334987450_A_sceptical_analysis_of_Quantized_Inertia 2019-10-09T02:55:33 < mawk> We've picked up some unusual traffic from your network and have temporarily blocked access from your IP address. 2019-10-09T02:55:35 < mawk> fdsoixfhdsiofhdsuifhdsuif 2019-10-09T02:55:54 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-09T03:04:34 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-09T03:10:56 < kakimir32> CP 2019-10-09T03:11:32 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-09T03:14:08 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-09T03:24:48 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-09T03:28:49 -!- fenugrec [~fenugrec@104.221.51.116] has joined ##stm32 2019-10-09T03:39:05 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 276 seconds] 2019-10-09T04:27:11 -!- qyx [~qyx@gw2.krtko.org] has quit [Ping timeout: 250 seconds] 2019-10-09T04:27:39 -!- tsprlng [~tsprlng@cpc99580-brnt1-2-0-cust501.4-2.cable.virginm.net] has quit [Ping timeout: 240 seconds] 2019-10-09T04:28:06 -!- tsprlng [~tsprlng@cpc99580-brnt1-2-0-cust501.4-2.cable.virginm.net] has joined ##stm32 2019-10-09T04:55:08 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 276 seconds] 2019-10-09T04:55:14 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-09T04:55:40 < kakimir32> https://www.youtube.com/watch?v=EK32jo7i5LQ 2019-10-09T05:18:42 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Quit: ZNC 1.6.1 - http://znc.in] 2019-10-09T05:19:05 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-09T05:19:05 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-09T05:19:05 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-09T05:19:30 -!- grindhold [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-09T05:29:43 -!- dark_machinehum [~machinehu@ip-142-232-165-0.ptr.bcit.ca] has joined ##stm32 2019-10-09T05:29:51 < dark_machinehum> Hello weiners 2019-10-09T05:50:23 -!- dark_machinehum [~machinehu@ip-142-232-165-0.ptr.bcit.ca] has quit [Ping timeout: 276 seconds] 2019-10-09T06:06:09 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-09T06:41:58 -!- fenugrec [~fenugrec@104.221.51.116] has quit [Ping timeout: 245 seconds] 2019-10-09T06:50:25 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-09T07:27:39 -!- qyx [~qyx@84.245.120.190] has joined ##stm32 2019-10-09T07:32:32 -!- qyx_ [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-09T07:33:44 -!- qyx [~qyx@84.245.120.190] has quit [Ping timeout: 268 seconds] 2019-10-09T07:42:33 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-09T07:53:54 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-09T07:57:39 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-09T07:57:39 -!- day__ is now known as day 2019-10-09T08:07:08 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 265 seconds] 2019-10-09T08:22:11 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-09T08:25:21 < R2COM> sup 2019-10-09T08:25:31 < R2COM> so fucken tired 2019-10-09T08:25:42 < R2COM> been innovating all day at work from 8 to 6:30 2019-10-09T08:41:58 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-09T08:42:43 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-09T09:07:27 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-09T09:24:51 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-09T09:26:25 -!- jly [uid355225@gateway/web/irccloud.com/x-uprypmmfexegitga] has joined ##stm32 2019-10-09T09:26:45 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-09T09:37:30 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-09T09:49:39 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-09T09:49:48 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-09T09:51:40 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-09T09:51:54 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-09T09:55:18 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-09T10:05:00 -!- tprrt [~tprrt@217.114.204.178] has quit [Ping timeout: 252 seconds] 2019-10-09T10:06:36 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-09T10:11:13 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-09T10:11:19 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-09T10:36:05 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-09T11:20:06 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Read error: Connection reset by peer] 2019-10-09T11:22:36 -!- m4ssi [~massi@82.62.130.138] has joined ##stm32 2019-10-09T11:45:09 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-09T11:54:55 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-09T11:54:55 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Read error: Connection reset by peer] 2019-10-09T12:02:55 < Thorn> https://twitter.com/SaschaNaske/status/1181830339183992833 2019-10-09T12:21:00 < jly> why not jlc pcb it 2019-10-09T12:21:15 < jly> Haohmaru 2019-10-09T12:22:38 < dongs> uh dude 2019-10-09T12:22:39 < dongs> china silk 2019-10-09T12:22:40 < dongs> lol 2019-10-09T12:22:43 < dongs> don't even start 2019-10-09T12:22:53 < jly> yeah 2019-10-09T12:23:12 < jly> i don't know who's blocked who 2019-10-09T12:24:23 < jly> surprised the boss hasn't pulled the plug on my 3352nd nick here 2019-10-09T12:30:46 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-09T12:30:47 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Read error: Connection reset by peer] 2019-10-09T12:33:51 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has joined ##stm32 2019-10-09T12:39:20 < jly> Steffanx is a fucking prick 2019-10-09T12:41:21 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-09T12:43:52 < dongs> its almost as if stvn is trying to get bant 2019-10-09T12:44:08 < dongs> pumping without a pause 2019-10-09T12:44:34 < jly> worth a few seconds 2019-10-09T12:44:46 < jly> off to others things now 2019-10-09T12:45:04 < jly> i learned pumping from the king 2019-10-09T12:46:53 -!- qyx_ is now known as qyx 2019-10-09T12:48:59 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-09T13:01:17 < PaulFertser> jly: your king has tits https://en.wikipedia.org/wiki/Elizabeth_II 2019-10-09T13:01:20 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 276 seconds] 2019-10-09T13:03:28 < jly> hi paul 2019-10-09T13:05:33 < PaulFertser> Haohmaru: yes, I hope jly is proud to have such a great leadership. 2019-10-09T13:06:07 < jly> good in the oven 2019-10-09T13:09:20 < Thorn> https://www.youtube.com/watch?v=84sI8GQ8o34 2019-10-09T13:12:27 < PaulFertser> jly: btw, I remember you used to fix professional sound equipment. How can one possibly find schematics for SPL Vitalizer MK2-T ? 2019-10-09T13:12:48 < effractur> /win 141 2019-10-09T13:13:06 < jly> well 2019-10-09T13:13:11 < jly> is SPL a brand? 2019-10-09T13:13:51 < jly> if you tell me the fault i can pretend i have one on the bench 2019-10-09T13:14:23 < jly> most of the ppl on the 'private' forum know i'm an EE 2019-10-09T13:14:43 < PaulFertser> jly: yes, something like this https://vintageking.com/spl-stereo-vitalizer-mk2-t . The problem is channel imbalance, one is few dB lower than the other. 2019-10-09T13:15:43 < jly> you're on the magic hidden place i am on too yea? 2019-10-09T13:16:01 < PaulFertser> Eh, no 2019-10-09T13:16:05 < jly> ask senpai 2019-10-09T13:16:12 < jly> it's just easier if we chat there 2019-10-09T13:16:45 < PaulFertser> I have no idea what you're talking about :) 2019-10-09T13:17:19 < jly> is that some kind of cat? 2019-10-09T13:20:16 < Steffanx> Ty jly 2019-10-09T13:20:57 < jly> i'd start with a spec sheet 2019-10-09T13:21:08 < jly> few dBs may be normal who knows 2019-10-09T13:21:18 < PaulFertser> Haohmaru: analague 2019-10-09T13:23:08 < jly> it's always fun to have a 'good working one' on the bench too 2019-10-09T13:23:22 < jly> profile what 'a good one' does/measures 2019-10-09T13:23:44 < PaulFertser> Haohmaru: bought used 2019-10-09T13:23:53 < jly> hah they're the good ones 2019-10-09T13:23:55 < PaulFertser> New tubes installed 2019-10-09T13:24:07 < jly> oh wow toobs!! 2019-10-09T13:24:19 < jly> just like my guitar amplifier 2019-10-09T13:24:45 < jly> best thing is when the paper board conducts and raises the control grids to over 0v 2019-10-09T13:24:56 < jly> 500v B+ next door :) 2019-10-09T13:25:24 < PaulFertser> My brother was doing "in the box" mixing for years, and he's quite good in it. And now they're trying to add some analogue equipment and he says it's mostly feels nice, often does the job faster than what it takes him to configure all the VST plugins etc. 2019-10-09T13:25:53 < jly> and here I was thinking about noisy plate resistors :# 2019-10-09T13:27:22 < jly> i was going to build a choob amp for fun once upon a time 2019-10-09T13:29:27 < jly> dsl100 was good for me 2019-10-09T13:33:50 < PaulFertser> "He began to cultivate his passion for music from age twelve. At age fourteen, he intensified his commitment to the art, and began DJing professionally at school parties. " 2019-10-09T13:34:16 < PaulFertser> Yep 2019-10-09T13:34:39 < PaulFertser> So gents intensify your commitments before it's too late to become pro. 2019-10-09T13:35:31 < jly> I was never serious about gtr 2019-10-09T13:35:36 < jly> good fun however 2019-10-09T13:35:46 < jly> easier to play with voltages 2019-10-09T13:36:24 < PaulFertser> Too bad jly, you won't be able to play at school parties then. 2019-10-09T13:39:11 < jly> i'm too old.ex 2019-10-09T13:39:30 < jly> e 2019-10-09T13:49:40 < Steffanx> Grandpa jly 2019-10-09T13:50:02 < jly> hi sir 2019-10-09T13:50:18 < jly> i'm glad you could come 2019-10-09T13:50:44 < jly> with all this new telegramming and instagramming.... 2019-10-09T13:50:55 < jly> nobody remembers the old ir C 2019-10-09T13:52:19 < jly> isn't it https://twitter.com/RealTimeWWII/ 2019-10-09T14:12:33 < PaulFertser> In russia some IT motherfuckers are trying to block telegram and as the result they blocked plenty (millions) of IP addresses, and they're still on it. What I was surprised with is that during a recent "meeting" of the motherfuckers' leader with the students nobody punched him in the face, and apparently nobody even said a word to him. "Every nation gets the government it deserves." indeed! 2019-10-09T14:20:57 < Cracki> define "some IT motherfuckers" 2019-10-09T14:25:13 < PaulFertser> Cracki: special goverment agency 2019-10-09T14:25:46 < PaulFertser> Cracki: https://media.ccc.de/v/35c3-9653-russia_vs_telegram_technical_notes_on_the_battle 2019-10-09T14:27:30 < Thorn> MPS has a complete clone of tps63030 including package and pinout 2019-10-09T14:34:03 -!- ekaologik [~quassel@p4FF16842.dip0.t-ipconnect.de] has joined ##stm32 2019-10-09T14:42:23 < Steffanx> Da da. Na tochku! 2019-10-09T14:43:36 < Steffanx> My gaming russian is best. 2019-10-09T14:45:11 < dongs> PaulFertser: is this now? my pal in uk is having trouble with telegram today 2019-10-09T14:45:17 < dongs> are they blocking shit? 2019-10-09T14:45:52 < PaulFertser> dongs: the UK does internet censorship on dns level only. roskomnadzor is unlikely to affect any UK users. 2019-10-09T14:52:17 < dongs> instafail maybe, i dont know. it sucks so much and is so irrelevant i dont ever need to mention about it 2019-10-09T14:57:23 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Ping timeout: 245 seconds] 2019-10-09T15:07:04 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-09T15:20:34 < kakimir64> product: USB to PCIE Bridge 2019-10-09T15:20:44 < kakimir64> product: USB to PCIE Bridge 2019-10-09T15:21:36 < qyx> does it make PCI-e from USB 2019-10-09T15:21:42 < kakimir64> serial: 0123456789ABCDEF 2019-10-09T15:22:08 < kakimir64> I thought they would use legit chips 2019-10-09T15:22:22 < dongs> http://www.frescologic.com/zh/product/single/fl6000/ im suposed to be getting evalboard of this sometime this year 2019-10-09T15:22:33 < kakimir64> works like no stinking clone though so no problemo 2019-10-09T15:22:37 < dongs> they claimed it does work with * as opposed to that VLI shit 2019-10-09T15:22:52 < dongs> this thing shows up as XHCI controller or something 2019-10-09T15:25:13 < kakimir64> usb controller connected via usb? 2019-10-09T15:25:19 < dongs> yes 2019-10-09T15:25:21 < dongs> yo dawg. 2019-10-09T15:25:41 < kakimir64> yo dawg indeed 2019-10-09T15:26:10 < kakimir64> what it's needed for? 2019-10-09T15:26:51 < dongs> full bandwidth usb 2.0 hub -> usb 3.0 uplink 2019-10-09T15:27:30 < kakimir64> doesn't usb3.0 hubs works? 2019-10-09T15:27:35 < dongs> no 2019-10-09T15:27:40 < kakimir64> connecting multiple usb2.0 devices 2019-10-09T15:27:41 < dongs> USB3 hub doesn't do anytrhing with usb 2.0 2019-10-09T15:27:42 < dongs> no 2019-10-09T15:27:50 < dongs> they go over same sahred 480mbit usb2.0 link inside USB3 plug. 2019-10-09T15:28:00 < kakimir64> oh 2019-10-09T15:28:02 < dongs> *shared 2019-10-09T15:28:14 < kakimir64> so this is hub on steroids 2019-10-09T15:29:43 < kakimir64> or hub that works as you expect rather 2019-10-09T15:32:30 < kakimir64> is this originally chinese innovation? 2019-10-09T15:35:36 < dongs> yea 2019-10-09T15:40:37 < kakimir64> I got my first mikrodik product today 2019-10-09T15:40:54 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-09T15:41:39 < kakimir64> it's cute 2019-10-09T15:48:30 < kakimir64> came with all the compliance stickers and stickers for MAC address etc. 2019-10-09T15:48:48 < kakimir64> and ofc. recycle bin sticker 2019-10-09T15:50:44 < kakimir64> the board itself is all business 2019-10-09T15:52:25 < qyx> they are super-pro, using them since 2006 2019-10-09T15:58:41 < dongs> yes microdik is amazing 2019-10-09T16:01:36 < kakimir64> where do i put these stickers 2019-10-09T16:04:53 < kakimir64> I did professionally place them in the back of the box 2019-10-09T16:05:46 < kakimir64> yes 2019-10-09T16:06:09 -!- jly [uid355225@gateway/web/irccloud.com/x-uprypmmfexegitga] has quit [Quit: Connection closed for inactivity] 2019-10-09T16:11:44 < day> you arent supposed to read books 2019-10-09T16:11:54 < day> just grind the questions :P 2019-10-09T16:13:07 < day> thats what i mean totally pointless 2019-10-09T16:13:24 < kakimir64> you don have loicence Haohmaru? 2019-10-09T16:13:32 < day> it's insane how the driving tests are made to literally get everyone to pass it X: 2019-10-09T16:14:27 < day> thats normal. some questions are just to specific 2019-10-09T16:15:28 < day> i made it twice. the first time i was gullable and attended to the mandatory lessons etc. 2019-10-09T16:16:59 < day> the 2nd time a few years later (motorcycle) i made clear i want to get through this quickly, no mandatory theory lessons. just driving lessons. 2019-10-09T16:28:37 < Cracki> they aren't mandatory if you can get around them 2019-10-09T16:29:32 < Cracki> when the dentist sells you on deluxe tooth tickles with dodo ass hair for just 999.95, it's not mandatory 2019-10-09T16:29:50 < Cracki> *feathers 2019-10-09T16:30:42 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-09T16:32:59 < day> they are mandatory here 2019-10-09T16:33:29 < karlp> so you did it then.... because it was mandatory... 2019-10-09T16:33:35 < day> but faking some attendence signatures isnt exactly difficult 2019-10-09T16:33:38 < day> nah 2019-10-09T16:33:53 < Cracki> ic, greasing some palms 2019-10-09T16:34:18 < day> i mean this shit costs 1k+EUR if i pay this kind of money i choose the music 2019-10-09T16:34:25 < Cracki> heh 2019-10-09T16:39:11 < Cracki> mandatory lecture feels weird once you're used to university lectures where you can just work with a book or presentation slides 2019-10-09T16:39:31 < Cracki> ... or lecture videos you can speed up because it's mostly fluff anyway 2019-10-09T16:42:27 < karlp> check out this quality product: https://www.digikey.com/product-detail/en/infineon-technologies/KITLGCAPBOM005TOBO1/KITLGCAPBOM005TOBO1-ND/10279173 2019-10-09T16:46:15 < kakimir64> is that oxidation? 2019-10-09T16:46:54 < kakimir64> okay now how do I get started with this routerboard thing? 2019-10-09T16:47:34 < zyp> put it in the box? 2019-10-09T16:47:48 < kakimir64> check 2019-10-09T16:47:50 < kakimir64> then 2019-10-09T16:48:00 < zyp> throw out the box 2019-10-09T16:48:05 < kakimir64> ok 2019-10-09T16:48:29 < qyx> download quality winbox software 2019-10-09T16:48:39 < qyx> no aids os supported 2019-10-09T16:48:42 < qyx> so get windows 2019-10-09T16:48:52 < qyx> or wine 2019-10-09T16:49:02 < zyp> what do you need winbox for? 2019-10-09T16:49:03 < PaulFertser> aids os meaning macOS? 2019-10-09T16:49:05 < qyx> click click, press search, click on the router 2019-10-09T16:49:24 < qyx> to set it up in a more sane way than with the web interface 2019-10-09T16:49:40 < qyx> also for L2 connection 2019-10-09T16:50:01 < PaulFertser> I thought one is supposed to ssh to RB and use those fancy coloured CLI to set it up, no 2019-10-09T16:50:04 < PaulFertser> ? 2019-10-09T16:50:10 < qyx> that too 2019-10-09T16:50:21 < qyx> but for all those things you need to have at least some default configuration 2019-10-09T16:50:31 < kakimir64> how do I configure the box after it has been set to bridging mode? 2019-10-09T16:50:50 < qyx> see 2019-10-09T16:51:08 < qyx> if you forgot to set an ip on the bridge, no web, no ssh 2019-10-09T16:51:21 < zyp> fun 2019-10-09T16:52:17 < kakimir64> in huawei you connect laptop to free lan port excl. lan4 that is bridging port and set ip for machine in 198.168.1.x address range 2019-10-09T16:52:30 < qyx> so get winbox, connect with ethernet to whatever port you have free in the same subnet 2019-10-09T16:53:02 < qyx> it will do a broadcast search and you are able to connect even to a router with no valid config 2019-10-09T16:53:11 -!- effractu1 [~Erik@195.140.241.50] has joined ##stm32 2019-10-09T16:53:22 < qyx> even to an unconfigured port 2019-10-09T16:53:31 < qyx> (if it is not disabled) 2019-10-09T16:53:36 < kakimir64> let's see that winbox thing 2019-10-09T16:53:45 < kakimir64> what is the dude and netinstall? 2019-10-09T16:53:50 -!- markus-k_ [~markus@server01.comtime-it.eu] has joined ##stm32 2019-10-09T16:54:36 -!- marble_visions_ [~user@68.183.79.8] has joined ##stm32 2019-10-09T16:54:44 -!- dan3wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-09T16:54:44 -!- dan3wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-09T16:54:45 -!- dan3wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-09T16:55:15 -!- fest [~fest@static.170.38.201.195.clients.your-server.de] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:15 -!- effractur [~Erik@195.140.241.50] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:15 -!- markus-k [~markus@server01.comtime-it.eu] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- marble_visions [~user@68.183.79.8] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- Alexer [~alexer@alexer.net] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- mrus [~mrus@35.211.88.77] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- ub|k [~ubikuitou@indico/developer/ubik] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:16 -!- grindhold [~quassel@mail.skarphed.org] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:17 -!- canton7 [~canton7@about/csharp/regular/canton7] has quit [Ping timeout: 268 seconds] 2019-10-09T16:55:17 -!- Alexer [~alexer@alexer.net] has joined ##stm32 2019-10-09T16:55:18 -!- dan3wik is now known as dan2wik 2019-10-09T16:55:21 -!- canton7_ [~canton7@about/csharp/regular/canton7] has joined ##stm32 2019-10-09T16:55:29 -!- mrus [~mrus@35.211.88.77] has joined ##stm32 2019-10-09T16:55:38 -!- day [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-09T16:55:40 -!- canton7_ is now known as canton7 2019-10-09T16:56:08 -!- grindhold [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-09T16:56:08 -!- ub|k [~ubikuitou@indico/developer/ubik] has joined ##stm32 2019-10-09T16:56:49 -!- fest [~fest@static.170.38.201.195.clients.your-server.de] has joined ##stm32 2019-10-09T16:57:45 < Cracki> tv series recommendation: Fremvandrerne (Beforeigners) 2019-10-09T16:58:03 -!- antto [~pewpew@antonsavov.net] has quit [Ping timeout: 263 seconds] 2019-10-09T16:58:41 < Cracki> they equate ancient vikings in modern norway to africans in europe 2019-10-09T16:59:25 < Cracki> for some reason it feels like satire 2019-10-09T17:00:17 -!- antto [~pewpew@antonsavov.net] has joined ##stm32 2019-10-09T17:02:51 < qyx> kakimir64: dude is mass configurator/management 2019-10-09T17:03:06 < qyx> and netinstall is dhcp/tftp server to install routerboards 2019-10-09T17:03:35 < qyx> if you manage to brick the router somehow 2019-10-09T17:10:00 < kakimir64> yes tftp 2019-10-09T17:10:16 < kakimir64> does it works with other than routerboard too? 2019-10-09T17:16:20 < qyx> idk 2019-10-09T17:27:46 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-09T17:37:32 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-09T17:48:15 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-09T17:53:14 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-09T18:34:23 < bitmask> maruhaoh 2019-10-09T18:46:54 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-09T18:49:12 -!- m4ssi [~massi@82.62.130.138] has quit [Remote host closed the connection] 2019-10-09T19:12:43 < bitmask> for connectors that have through hole support pins do you connect those to ground? 2019-10-09T19:12:51 < bitmask> or isolate them 2019-10-09T19:13:06 < Steffanx> Mitbask. Lol. 2019-10-09T19:14:57 < bitmask> Steffanx would you ground them? 2019-10-09T19:17:29 < jpa-> bitmask: are they electrically connected to something on the connector? 2019-10-09T19:17:38 < bitmask> no 2019-10-09T19:17:52 < jpa-> then sure, why not 2019-10-09T19:18:03 < jpa-> they'll be more solid if they're soldered to ground plane 2019-10-09T19:18:34 < bitmask> its like those to220 heatsinks that have supports, this is for a xt30pw and mr30pw though 2019-10-09T19:28:25 -!- ekaologik [~quassel@p4FF16842.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-09T19:31:22 -!- sterna [~Adium@m5-241-62-173.cust.tele2.se] has joined ##stm32 2019-10-09T19:41:41 * qyx going to breadboard some LM2575 buck 2019-10-09T19:42:01 < qyx> for a home arduino project 2019-10-09T19:44:31 < roomba> those things are terrible 2019-10-09T19:44:49 < roomba> i spent 2 weeks reading datasheets and those were one of the worst 2019-10-09T19:45:08 < roomba> didn't find anything better either lol 2019-10-09T19:45:18 < roomba> even ones that look good. just paste in here and get shot down. 2019-10-09T19:46:41 < qyx> what do you mean, LM2575? 2019-10-09T19:48:51 < bitmask> whats the difference between 2575 and 2595, I plan on using 2595 for my project 2019-10-09T19:49:07 < bitmask> 2595 is higher switching freq maybe? 2019-10-09T19:58:24 < qyx> idk, I just found some 10y+ old parts 2019-10-09T19:59:18 < qyx> and I need to arduino up a 24->5V for a nucleo to control a pump 2019-10-09T19:59:52 < karlp> qyx: pcie board fits nicely. no chamfered edges or hard gold, just the stock kicad footprint for it. 2019-10-09T20:00:35 < qyx> \o/ 2019-10-09T20:00:40 < qyx> hows aisler? 2019-10-09T20:00:55 < karlp> pcbs are ok, stencil went well, first time with that, but I don't like the process 2019-10-09T20:01:04 < karlp> shipping was a train wreck 2019-10-09T20:01:57 < bitmask> if an IC has a tab and is internally connected to another pin, do you use the same designator/pin number or not 2019-10-09T20:08:26 < zyp> sometimes, usually not 2019-10-09T20:08:32 < bitmask> k 2019-10-09T20:12:02 < jpa-> i use the same labels as in the datasheet, unless kicad already has the footprint in which case i use that numbering 2019-10-09T20:12:46 < antto> i obey teh KLC 2019-10-09T20:12:58 < antto> usually 2019-10-09T20:13:21 < qyx> same as jpa- here 2019-10-09T20:13:25 < bitmask> https://pdf1.alldatasheet.es/datasheet-pdf/view/189106/MPS/MP1593.html PAGE 2 2019-10-09T20:13:33 < bitmask> so I added a pin 9 and named it GND 2019-10-09T20:13:35 < jpa-> klc = kentucky lied chicken? 2019-10-09T20:13:40 < bitmask> not really happy with that but not sure what else to do 2019-10-09T20:13:48 < antto> jpa- kicad library convention 2019-10-09T20:13:54 < karlp> so, how am I mean tto clean this stencil for reuse? 2019-10-09T20:14:09 < zyp> IPA 2019-10-09T20:14:11 < antto> karlp afaic with speshul crap 2019-10-09T20:14:26 < jpa-> IPA works for me 2019-10-09T20:14:40 < antto> or you just had to be jentle 2019-10-09T20:14:50 < antto> * gentle 2019-10-09T20:15:07 < jpa-> well rinse with IPA, not pressure wash with IPA 2019-10-09T20:17:03 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-09T20:17:34 < zyp> those both sounds wrong 2019-10-09T20:17:37 < zyp> I wipe with IPA 2019-10-09T20:18:44 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Excess Flood] 2019-10-09T20:19:05 < antto> i also thought you'd wipe it with sumfin 2019-10-09T20:19:25 * karlp wonders if there's any IPA in the office 2019-10-09T20:19:40 < antto> someone drank it 2019-10-09T20:19:44 < zyp> just beware of paper lint sticking to the edges of the apertures 2019-10-09T20:20:28 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-09T20:23:15 < kakimir64> nice moped ride 2019-10-09T20:23:21 < kakimir64> no other moped riders 2019-10-09T20:23:50 < kakimir64> 110km total 2019-10-09T20:23:58 < antto> isn't it kinda cold up there kakimir64? 2019-10-09T20:24:23 < kakimir64> on my way back it started snowing 2019-10-09T20:24:35 < antto> r u crazy o_O 2019-10-09T20:24:59 < antto> well, perhaps it could be warm-ish 2019-10-09T20:25:00 < kakimir64> my decision was based on careful judgement 2019-10-09T20:25:31 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-09T20:25:45 < antto> you collected all teh factors and ran statistical analysis and predictive estimations? 2019-10-09T20:26:17 < antto> you tossed a coin? 2019-10-09T20:26:50 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-09T20:27:19 < kakimir64> I felt the air outside and it was pretty cold but it didn't feel like icing condition yet 2019-10-09T20:27:38 < antto> ah, i also do that 2019-10-09T20:27:50 < antto> i open teh window slightly and stick muh nose out 2019-10-09T20:28:09 < antto> if muh snot freezes - it's too cold 2019-10-09T20:28:41 < antto> or if i get hit in teh face by an ice kewb 2019-10-09T20:33:31 < PaulFertser> kakimir64: how do you keep your hands warm? 2019-10-09T20:34:17 < PaulFertser> And what speed? 2019-10-09T20:34:18 < kakimir64> one stop to get coffee indoors and another to heat my hands with engine block 2019-10-09T20:34:38 < kakimir64> leather riding hand things 2019-10-09T20:34:47 < sync> PaulFertser: heated grips ftw 2019-10-09T20:34:57 < kakimir64> no heated grips 2019-10-09T20:35:13 < PaulFertser> sync: heh, I think about installing those for about few weeks before winter and few weeks after. 2019-10-09T20:35:24 < sync> yes 2019-10-09T20:36:14 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-09T20:36:18 < PaulFertser> kakimir64: what was your speed? I notice every extra 10 km/h adds considerably to the perceived cold. 2019-10-09T20:37:14 < kakimir64> 100 2019-10-09T20:37:53 < PaulFertser> That's plenty when it's close to 0C... 2019-10-09T20:38:08 < kakimir64> 30km @ 130 2019-10-09T20:39:02 < antto> kakimir64 gloves 2019-10-09T20:39:06 < kakimir64> yes 2019-10-09T20:39:20 < antto> they fit like a gluv 2019-10-09T20:39:27 < kakimir64> no airflow through leather 2019-10-09T20:39:29 < kakimir64> zero 2019-10-09T20:39:44 < antto> same goes for plastic bags 2019-10-09T20:39:56 < antto> ghetto leather ;] 2019-10-09T20:40:13 < PaulFertser> I have warm dainese dry-fit, they do not let air in as well but it's still cold :/ 2019-10-09T20:40:51 < kakimir64> install gloves to handlebar 2019-10-09T20:40:57 < PaulFertser> d-dry 2019-10-09T20:41:04 < PaulFertser> kakimir64: do you have those? 2019-10-09T20:41:24 < kakimir64> I have but I have never used them in motorized vehicles :o 2019-10-09T20:41:46 < kakimir64> (intended use case for handlebar gloves) 2019-10-09T20:42:40 < kakimir64> I have neopren skimask under helmet 2019-10-09T20:44:58 < antto> kakineo 2019-10-09T20:45:21 < antto> equipimir 2019-10-09T20:51:10 < kakimir64> actually if my hands didn't get cold it would have been not cold at all 2019-10-09T20:51:17 < kakimir64> maybe install them 2019-10-09T20:52:15 < Cracki> kek boss wants to get into youtubing because blogs are so last decade 2019-10-09T20:56:59 < antto> much kek, very lol 2019-10-09T21:03:04 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-09T21:07:20 < kakimir64> he wants you to youtube? 2019-10-09T21:11:44 < Cracki> he thinks he would need me to edit the videos 2019-10-09T21:12:04 < Cracki> that real estate guy who works for louis rossmann... he has no video editing AT ALL 2019-10-09T21:12:26 < Cracki> just presses the button on his phone, maybe adjusts an in/out-slider youtube gives him 2019-10-09T22:09:25 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-09T22:15:02 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3980:648d:d875:5209] has joined ##stm32 2019-10-09T22:21:16 < benishor> try kinemaster on android 2019-10-09T22:21:26 < benishor> no need for fancy schmancy shit 2019-10-09T22:21:32 < benishor> that gots all you needz 2019-10-09T22:21:45 < benishor> just use your phone and fire away 2019-10-09T22:23:03 < benishor> https://www.kinemaster.com/ 2019-10-09T22:23:10 < benishor> awesome set of features 2019-10-09T22:23:57 < Cracki> thx for the rec 2019-10-09T22:24:27 < Cracki> I would not expect him to edit on a phone tho 2019-10-09T22:24:49 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-09T22:26:30 < Cracki> I'd also want him to have something that shows the audio waveform. NOBODY who does serious editing pays much attention to the thumbnail strip, apart from identifying what source it is 2019-10-09T22:26:36 < catphish> morning retards 2019-10-09T22:26:56 < Cracki> moinmoin 2019-10-09T22:27:39 < Cracki> biggest "issue" rn is how do I, who has no youtubing experience either, help him make decent videos, apart from saying to JFGI 2019-10-09T22:28:48 < Cracki> he was like "I'm gonna need to make a script of what to say" and omg I hope not, reading a speech makes for unbearable audio 2019-10-09T22:29:15 < catphish> i've never really understood the secret of making good videos, the obvious thing is to not read a script 2019-10-09T22:29:44 < catphish> rambling, being dumb, just forcing yourself to be yourself and get into it wholeheartedly seems the answer 2019-10-09T22:29:58 < catphish> and maybe not expecting it to be good straight away 2019-10-09T22:29:59 < Cracki> tending towards having him just talk, repeat and reformulate sections however often he likes, then removing stuff... like writing a letter on paper, where you can strike out stuff without having to actually do a clean copy 2019-10-09T22:30:12 < catphish> that makes sense 2019-10-09T22:30:14 < Cracki> camera angle and setting/stage is important too 2019-10-09T22:30:28 < Cracki> eye level, no confrontational pose 2019-10-09T22:30:36 < benishor> scripts are good 2019-10-09T22:30:42 < benishor> if you learn them before hand 2019-10-09T22:30:52 < Cracki> no plan/script survives contact with the enemy/camera 2019-10-09T22:30:54 < benishor> in any case, it's good to have an idea on what you're about to say 2019-10-09T22:31:05 < benishor> it's fucking annoying to watch videos filled with "erm", "ahemm" 2019-10-09T22:31:13 < Cracki> you want to write down the points you wanna make, but the rest is ok to wing 2019-10-09T22:31:16 < benishor> there needs to be a direction 2019-10-09T22:31:22 < Cracki> that's what editing is for 2019-10-09T22:31:32 < Cracki> repeat a section until there are no uhms anymore 2019-10-09T22:31:40 < catphish> ^ this 2019-10-09T22:31:55 < Cracki> literally you can strike out text, it's like a typewriter 2019-10-09T22:31:57 < catphish> practice until it comes out coherantly 2019-10-09T22:32:06 < Cracki> not even practice, just repeat right there 2019-10-09T22:32:23 < mitrax> cracki: what will be the videos about? 2019-10-09T22:32:26 < Cracki> basically, go ramble, then summarize, summarize again if needed 2019-10-09T22:32:47 < Cracki> his side job of about two decades 2019-10-09T22:32:54 < catphish> the most important think i've learned about speaking is not to be afraid of silence, it's not like a meeting where someone might butt in, teach yourself to fill gaps with nothing 2019-10-09T22:33:06 < Cracki> he has a good main job in the same field, but the side job is basically hobby 2019-10-09T22:33:10 < Cracki> electronics 2019-10-09T22:33:28 < Cracki> silence is awesome to edit, you see it in the audio waveform 2019-10-09T22:33:36 < catphish> i missed the context here anyway, what are these videos? 2019-10-09T22:33:50 < Cracki> boss wants to get int youtubing for hobby/side job 2019-10-09T22:34:01 < Cracki> his blog is getting long in the tooth 2019-10-09T22:34:11 < Cracki> and all his family and friends nag him to get with the times 2019-10-09T22:34:17 < catphish> makes sense 2019-10-09T22:34:22 < Cracki> he wants to showcase his work, basically 2019-10-09T22:34:36 < Cracki> what he has to sell, what custom jobs he can do for a customer, etc 2019-10-09T22:34:51 < catphish> my partner really wants to make videos, but he's never got into it, doesn't really know what to say 2019-10-09T22:34:53 < Cracki> some technical videos too, like mini lectures 2019-10-09T22:35:04 < Cracki> it's like selfies 2019-10-09T22:35:10 < Cracki> you don't point shoot and it's a masterpiece 2019-10-09T22:35:14 < mitrax> Cracki: will the editing be done during your work hours? or is it a side thing for you too? 2019-10-09T22:35:16 < catphish> i love selfies :) 2019-10-09T22:35:19 < Cracki> you take dozens, hundreds, discard almost all of them 2019-10-09T22:35:35 < Cracki> it's like painting, you gotta practice until you get it right instantly 2019-10-09T22:36:03 < Cracki> he's paying me, nothing for free 2019-10-09T22:36:17 < mitrax> ok good... cause that's super time consuming 2019-10-09T22:36:17 < Cracki> practice is probably the best advice 2019-10-09T22:36:22 < Cracki> eh, I'm quick with editing 2019-10-09T22:36:37 < mitrax> still 2019-10-09T22:36:47 < benishor> also audio needs to be quite good 2019-10-09T22:36:48 < mitrax> plus he'll probably require that you'll be there during filming etc won't he? 2019-10-09T22:37:05 < benishor> that means a good mike, proper speech compressor settings 2019-10-09T22:37:20 < benishor> no popping 2019-10-09T22:37:25 < Cracki> the biggest time sink is finding WHERE to edit 2019-10-09T22:37:30 < mitrax> bah a lavalier mic should do 2019-10-09T22:37:36 < Cracki> when I record lectures, I place markers in the audio track 2019-10-09T22:37:43 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3980:648d:d875:5209] has quit [Quit: Konversation terminated!] 2019-10-09T22:37:44 < Cracki> those I see in the editing program, quick work 2019-10-09T22:37:51 < Cracki> fuck filters 2019-10-09T22:37:55 < Cracki> good mike is enough 2019-10-09T22:38:06 < benishor> that kinemaster thinggie has a nice feature: it allows you to do voice overs 2019-10-09T22:38:17 < benishor> compressor is not a filter 2019-10-09T22:38:22 < Cracki> lavalier should do indeed 2019-10-09T22:38:22 < benishor> well, sort of 2019-10-09T22:38:30 < Cracki> or maybe the phone has good enough quality 2019-10-09T22:38:37 < benishor> it's very important to use one to get the speech clear 2019-10-09T22:38:39 < Cracki> anything that changes input is a filter 2019-10-09T22:38:51 < benishor> nice sharp attack 2019-10-09T22:39:02 < benishor> not all of us have a natural born speaker voice 2019-10-09T22:39:04 < Cracki> do you distinguish between noise gate and compressor? 2019-10-09T22:39:10 < Cracki> not sure what you mean 2019-10-09T22:39:19 < benishor> noise gate is where you cut off the silence 2019-10-09T22:39:23 < Cracki> I know 2019-10-09T22:39:25 -!- fc5dc9d4 [~quassel@p5B08131E.dip0.t-ipconnect.de] has joined ##stm32 2019-10-09T22:39:32 < benishor> compressor is where you change the attack/decay of wording 2019-10-09T22:39:46 < Cracki> for the voice tracks I edit, I throw on some speech leveler, that's including a compressor I think 2019-10-09T22:39:47 < benishor> you want some punchiness when wording begins 2019-10-09T22:39:59 < Cracki> but I only do that because the lecturer varies in volume 2019-10-09T22:40:02 < benishor> because some people just speak softly 2019-10-09T22:40:08 < benishor> and it sounds like shit 2019-10-09T22:40:40 < benishor> but yeah, in the end it's all about practicing 2019-10-09T22:40:46 < benishor> and getting a feel of what works and what doesn't 2019-10-09T22:40:58 < benishor> don't expect to have brilliance from the first try 2019-10-09T22:41:30 < Cracki> I think I'll have him start "practicing the selfie" so he gets comfortable speaking to a camera 2019-10-09T22:42:14 < Cracki> making videos is similar but slightly different from public speaking 2019-10-09T22:42:23 < benishor> I wanted to start some video tutorials on how to build a 2d game platformer from scratch 2019-10-09T22:42:47 < benishor> I gave up after I fucked up an entire day doing and redoing voice overs for 5 min worth of speech 2019-10-09T22:42:59 < benishor> babbling does not help either 2019-10-09T22:43:03 < Cracki> sped up video for the boring typetypetype 2019-10-09T22:43:24 < Cracki> tbh screw voiceovers. just talk and record 2019-10-09T22:43:31 < Cracki> you can edit the video to fit the audio later 2019-10-09T22:43:33 < benishor> how does the practicing the selfie work? 2019-10-09T22:44:07 < Cracki> I'll have him record himself talking, "repeating the takes", until he finds his own face and voice bearable in the recordings 2019-10-09T22:44:32 < Cracki> that helps immensely because his perception of self is something I can't affect in editing 2019-10-09T22:44:56 < benishor> funny how most of us hate our voices and video representations 2019-10-09T22:45:02 < benishor> I wonder what is that 2019-10-09T22:45:06 < benishor> s/what/why 2019-10-09T22:45:07 -!- renn0xtk9 [~max@2a02:810d:1540:2448:ac15:1de5:ad7:e69b] has joined ##stm32 2019-10-09T22:45:35 < Cracki> one aspect is asymmetry: you find yourself bearable in the mirror, but take a photo (not mirrored) and it looks weird 2019-10-09T22:45:36 < benishor> probably because we never hear ourselves from the "outside". we always hear our voice from inside 2019-10-09T22:45:39 < benishor> and it sounds different 2019-10-09T22:45:44 < mitrax> because it doesn't match the way you picture yourself and hear your voice 2019-10-09T22:46:07 < benishor> but yeah, we see outselves in the mirror quite a lot 2019-10-09T22:46:12 < Cracki> you only know yourself from mirror reflections and from hearing your voice WHILE you speak, which is more bass-heavy because sound travels through your bones and flesh 2019-10-09T22:46:12 < mitrax> you never see yourself from certain angles 2019-10-09T22:47:23 < mitrax> you rarely see your head from the side :) plus everybody makes a special face when looking in the mirror, it's actually funny to watch 2019-10-09T22:47:58 < Cracki> brb googling cats seeing themselves in a mirror 2019-10-09T23:31:18 -!- Mangy_Dog [Mangy_Dog@05449c7a.skybroadband.com] has quit [Ping timeout: 252 seconds] 2019-10-09T23:51:15 -!- renn0xtk9 [~max@2a02:810d:1540:2448:ac15:1de5:ad7:e69b] has quit [Quit: Konversation terminated!] 2019-10-09T23:58:27 -!- sterna [~Adium@m5-241-62-173.cust.tele2.se] has quit [Quit: Leaving.] 2019-10-09T23:59:25 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-09T23:59:37 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] --- Day changed Thu Oct 10 2019 2019-10-10T00:10:18 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has quit [Quit: ZNC 1.7.5 - https://znc.in] 2019-10-10T00:11:10 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has joined ##stm32 2019-10-10T00:13:25 < jadew> Cracki, I agree about it being similar to public speaking 2019-10-10T00:13:43 < jadew> I also agree that doing it without a script is best, but only if you know what you're doing 2019-10-10T00:13:58 < Cracki> cheatsheet, not for the cheating but for the preparation 2019-10-10T00:14:09 < jadew> if you do that at first it most likely be a disaster, especially if it's not in your native language 2019-10-10T00:14:36 < Cracki> yeah language barrier is huge, even for lecturers who've been at it for over a decade 2019-10-10T00:14:37 < karlp> if you'r ebrand new, make sure you're videos are fucking short. 2019-10-10T00:15:05 < jadew> I shot about a week of video for a 5 minutes presentation for my product 2019-10-10T00:15:34 < Cracki> define "a week", 24*7 hours? 2019-10-10T00:15:35 < jadew> I tried the no-script thing in a previous presentation, I came out goofy, didn't like it 2019-10-10T00:15:59 < jadew> Cracki, no... like Monday to Sunday 2019-10-10T00:16:09 < jadew> that's what I did that week, "the video" 2019-10-10T00:16:14 < Cracki> ah 2019-10-10T00:16:29 < jadew> after the first one, I kinda new what was wrong and what I wanted from the new one 2019-10-10T00:16:37 < jadew> but it was extremely difficult to get there 2019-10-10T00:16:39 < Cracki> working on _things_ to present too, like scope views, curves, whatnot, I assume 2019-10-10T00:16:47 < jadew> I wanted a different voice, a different tone, different speed 2019-10-10T00:17:03 < jadew> Cracki, yeah, that part didn't take long tho 2019-10-10T00:17:24 < jadew> the time consuming bit was saying the same thing 100 times and not liking how it sounds 2019-10-10T00:17:27 < Cracki> different person maybe 2019-10-10T00:17:37 < Cracki> you can hire people to do voiceovers for those dickstarter videos 2019-10-10T00:17:46 < jadew> I didn't want that either 2019-10-10T00:17:50 < jadew> I wanted to be me 2019-10-10T00:17:52 < Cracki> literally make a script with some timing info and they'll read it 2019-10-10T00:18:00 < Cracki> you aren't you when it's not your voice 2019-10-10T00:18:03 < jadew> but me sounds squeeky by default 2019-10-10T00:18:36 < Cracki> voice filter, makes you sound like a boxing match announcer 2019-10-10T00:18:38 < jadew> thing is, the clips I used for the final video were eventually shot in ~2 hours 2019-10-10T00:18:41 < jadew> if not less 2019-10-10T00:18:57 < jadew> but that was at the end of the week, because I finally realized what I had to do 2019-10-10T00:19:02 < Cracki> :) 2019-10-10T00:19:23 < jadew> I chilled out 2019-10-10T00:19:42 < Steffanx> Did you take a good nap? 2019-10-10T00:20:15 < jadew> there were lots of good naps :) 2019-10-10T00:20:27 -!- notSimon-- [~sim@2606:6a00:0:28:5604:a6ff:fe02:702b] has quit [Quit: desk move] 2019-10-10T00:21:30 < jadew> I think that if you do this many times, eventually you can do it without a script, but at first I don't think you can do can just start filming 2019-10-10T00:22:40 < Cracki> that's for sure 2019-10-10T00:22:54 < jadew> weirdly enough, the no-script one got more likes/view 2019-10-10T00:23:06 < karlp> wow, zephyr really knows how to treat new users. 2019-10-10T00:23:07 < jadew> but I think they were petty likes 2019-10-10T00:23:22 < jadew> pitty 2019-10-10T00:23:33 < Cracki> I was about to ask 2019-10-10T00:46:40 < zyp> karlp, what's up? 2019-10-10T00:57:29 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-10T00:57:36 < Laurenceb> muh 2019-10-10T01:05:12 < englishman> happy yom kippur jewrenceb 2019-10-10T01:05:22 < Laurenceb> thankyou 2019-10-10T01:05:42 < Laurenceb> muh nephew has had his benis ritually chopped 2019-10-10T01:08:41 < jadew> :( 2019-10-10T01:08:47 < jadew> poor kid 2019-10-10T01:09:01 < Cracki> did a robed figure suck the blood too? 2019-10-10T01:09:37 < Cracki> I hear that's how many of these poor children get herpes 2019-10-10T01:09:52 < jadew> heard that too, they actually bite the foreskin off 2019-10-10T01:10:07 < jadew> many died 2019-10-10T01:10:26 < Cracki> show me your peepee right now 2019-10-10T01:11:24 < jadew> mine doesn't have parts missing 2019-10-10T01:11:38 < Cracki> I'll believe that when I see it 2019-10-10T01:11:47 < jadew> hah 2019-10-10T01:12:22 < Laurenceb> no homo 2019-10-10T01:12:30 < Cracki> just look up "circumcision neonatal herpes" 2019-10-10T01:13:11 < Cracki> buncha writeups about single incidents buuut the population isn't that large to begin with 2019-10-10T01:13:14 < Laurenceb> and take a seat over there 2019-10-10T01:13:20 < kakimir64> NAME STATE READ WRITE CKSUM freenas-boot ONLINE 0 0 0 ada0p2 ONLINE 1 4 0 2019-10-10T01:13:21 < jadew> I was even against my son's baptism (we did it anyway, because parents) 2019-10-10T01:13:25 < BrainDamage> I imagine a priest isn't exactly the authority over medical procedures 2019-10-10T01:13:33 < Cracki> baptism is where they drown the infant? 2019-10-10T01:13:35 < kakimir64> what are theres read 1 and write 4 values 2019-10-10T01:13:42 < kakimir64> zpool status -v 2019-10-10T01:13:42 < jadew> Cracki, yeah 2019-10-10T01:13:43 < BrainDamage> starting from basic sterilization of the tools 2019-10-10T01:13:46 < catphish> is it possible on stm32's ARM for an interrupt to yield to other interrupts then re-enter? 2019-10-10T01:13:57 < Laurenceb> catphish: yes 2019-10-10T01:14:04 < Cracki> catphish, yes, the nested part of the NVIC 2019-10-10T01:14:11 < Laurenceb> you can configure this with NVIC 2019-10-10T01:14:11 < jadew> BrainDamage, they think holy water is sterile 2019-10-10T01:14:14 < catphish> thanks, i'll have a look 2019-10-10T01:14:30 < Cracki> pee is probably cleaner than random holy water 2019-10-10T01:14:32 < Laurenceb> you can also generate interrupts from software and use them for simple multitasking 2019-10-10T01:14:35 < catphish> it seeme like shitty design to have a long running interrupt really, but it's a simple approach here 2019-10-10T01:14:36 < jadew> Cracki, yep 2019-10-10T01:14:54 < Cracki> because pee has been water in your bloodstream before it was pee, so an immune system has been watching over it 2019-10-10T01:15:06 < Laurenceb> catphish: not really, if the hardware allows it 2019-10-10T01:15:08 < catphish> i have a USB RX interrupt, it needs to get data out of the USB buffer ASAP, but other interrupts are more important 2019-10-10T01:15:11 < BrainDamage> pee is actually reasonably sterile in most individuals 2019-10-10T01:15:45 < catphish> just changing the priority of the interrupts would be good enough 2019-10-10T01:15:49 < BrainDamage> as opposed to a basin of water recycled over and over 2019-10-10T01:16:11 < Cracki> or the mouth of an old man 2019-10-10T01:16:28 < Laurenceb> catphish: sounds like a perfect case for NVIC 2019-10-10T01:16:37 < BrainDamage> mouths are pretty filthy in terms of bacterial load 2019-10-10T01:16:57 < jadew> when we had a priest come over for some other ritual (called by my mother in law), I had him wash his hands with surgical grade hand sanitizer 2019-10-10T01:17:04 < catphish> i believe i'm using the NVIC in the first place 2019-10-10T01:17:09 < BrainDamage> and the worst part is that it's pathogens fit for human life ... being harbord in a human 2019-10-10T01:17:24 < catphish> bit i don't touch it yet, except to enable interrupts 2019-10-10T01:17:29 < Cracki> you might be using the nvic, but you might not have configured any actual nesting/priorities 2019-10-10T01:17:44 < catphish> Cracki: indeed i haven't 2019-10-10T01:17:52 < Cracki> so... having a dog lick the blood off the penis would be a better idea? :P 2019-10-10T01:17:56 < catphish> i'll have a read of the NVIC documentation now 2019-10-10T01:17:59 < Laurenceb> tho I dont have this issue with babbyshake usb 2019-10-10T01:18:14 < Laurenceb> I just double buffered the outgoing usb data to solve this issue 2019-10-10T01:18:20 < Cracki> I understand that pathogens have preferred hosts and dissimilar hosts mean poor transmission between hosts 2019-10-10T01:18:56 < Laurenceb> nephewb is sucking on the khazar milkers 2019-10-10T01:18:57 < Cracki> let's make that a lizard because dogs are still mammals and also because lizard people 2019-10-10T01:19:15 < Cracki> also because lizard licks the lizard 2019-10-10T01:19:28 < Laurenceb> >""lizard"" 2019-10-10T01:19:29 < BrainDamage> yes, the more different the biology the lower the chance for a random pathogen, but that's not always the case 2019-10-10T01:19:37 < BrainDamage> there's notable exceptions 2019-10-10T01:20:04 < BrainDamage> the plague, sars, etc etc 2019-10-10T01:20:15 < BrainDamage> but yes, most pathogens don't cross species 2019-10-10T01:20:30 < Cracki> oh I had to explain to someone why surgeons hold their hands up while and after washing, rather than down. he thought the dripping water should go away from dry skin towards cleaned skin, which is wrong tho 2019-10-10T01:20:33 < jadew> but those that do seem to be devastatic 2019-10-10T01:20:36 < jadew> *devastating 2019-10-10T01:20:43 < jadew> like ebola 2019-10-10T01:20:55 < Cracki> bird flu, foot and mouth, ... 2019-10-10T01:21:12 < jadew> I had foot and mouth a couple of years back 2019-10-10T01:21:13 < Cracki> they always show up when "the host" is two species living together 2019-10-10T01:21:23 < jadew> that disease didn't exist when I was a kid 2019-10-10T01:21:25 < jadew> not in here anyway 2019-10-10T01:21:58 < kakimir64> ebolasong.mp3 2019-10-10T01:22:02 < Cracki> apropos wrong species: https://twitter.com/liangweihan4 2019-10-10T01:22:35 < kakimir64> https://www.youtube.com/watch?v=XGltVAJ4JCk ebolasong 2019-10-10T01:22:39 < BrainDamage> jadew: part of the reason is that the pathogen had all the time to evolve in the the other species and then when it bridges the species is already one that has been refined through centuries of selective breeding 2019-10-10T01:23:15 < jadew> BrainDamage, that sounds right, yeah 2019-10-10T01:23:25 < karlp> zyp: just the way they treated a bug report. I'd come to conclusion last night that it wasn't going to work out for me at least, I could never resolve crashes, and ahd spent too much time on it to get not-nearly far enough. 2019-10-10T01:24:07 < BrainDamage> also, there's a bit of observation bias 2019-10-10T01:24:27 < BrainDamage> if a pathogen only causes mild problems, it won't propagate or make a scandal 2019-10-10T01:29:48 < BrainDamage> meanwhile tonight at the pub I had to discuss to an idiot that said that hiv doesn't produce aids, because gary fucking mullis said so 2019-10-10T01:32:32 < Cracki> let's discuss this. hiv is the pathogen, aids is "the immune system" attacking itself (pathogen doing it). hiv leads to aids, UNLESS virus load can be kept low enough that it the pathogen can't do anything 2019-10-10T01:34:21 < BrainDamage> sure, hiv can not lead to aids, but he claims that hiv virus wasn't found in all aids victims ( or the antigen ), which is a straight lie 2019-10-10T01:35:54 < Cracki> now i'd need to ask whether "no hiv was found in any/all aids patients" or "it's not the case that hiv can be found in any/all aids patients" 2019-10-10T01:36:19 < Cracki> I'd have to dust off my formal logic to express this ambiguity of language more precisely 2019-10-10T01:36:54 < Cracki> not (forall (aids => hiv)) vs. forall (not (aids => hiv)) 2019-10-10T01:37:26 < Cracki> eh, ignore that hair splitting 2019-10-10T01:37:47 < Cracki> I'm sure there are ways to present with AIDS that stem from some other pathogen 2019-10-10T01:39:21 < BrainDamage> pretty sure that his argument is that aids isn't caused by hiv at all 2019-10-10T01:39:34 < BrainDamage> rather than some [most] cases are caused by it 2019-10-10T01:39:58 < kakimir64> any zpool pros around? 2019-10-10T01:40:14 < zyp> sup? 2019-10-10T01:40:20 < kakimir64> suup 2019-10-10T01:40:25 < Cracki> I choose to think that people making _such_ claims don't really believe in what they say, but they believe in something contradictory from which MUST follow such an assertion 2019-10-10T01:41:01 < Cracki> have you tried btrfs? I hear it's butter at fucking your system 2019-10-10T01:41:06 < kakimir64> zyp: can anything sensible be done out of 2 x 0.5T, 2 x 1T, 2 x 2T 2019-10-10T01:41:27 < zyp> three mirror vdevs, 3.5T usable total? 2019-10-10T01:41:31 < Cracki> mirror of jbod? 2019-10-10T01:41:39 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-10T01:41:53 < kakimir64> any high efficiency high reduntancy zpool configuration? 2019-10-10T01:42:14 < Cracki> sell the 0.5T 2019-10-10T01:42:23 < BrainDamage> high redundancy automatically implies low efficiency 2019-10-10T01:42:30 < kakimir64> sell to who? 2019-10-10T01:42:35 < zyp> well, you could do raidz2 for 2T usable total 2019-10-10T01:42:37 < Cracki> data recovery shops maybe :P 2019-10-10T01:42:50 < Cracki> they pay whatever because the customer pays the bill 2019-10-10T01:42:55 < kakimir64> zyp: which drives? 2019-10-10T01:42:59 < kakimir64> all of them? 2019-10-10T01:43:09 < zyp> yeah, essentially 6x0.5 2019-10-10T01:43:46 < kakimir64> that is 3T 2019-10-10T01:43:53 < zyp> I originally built my raid as 16x1.5T despite half being 3T drives 2019-10-10T01:44:05 < zyp> then capacity doubled once I replaced the last 1.5T drive 2019-10-10T01:44:11 < Cracki> z2 means two parity, so effective 4x0.5 2019-10-10T01:44:14 < Cracki> which is 2t 2019-10-10T01:44:18 < zyp> what Cracki said 2019-10-10T01:44:44 < kakimir64> ok 2019-10-10T01:44:50 < zyp> but you should consider your future plans for the raid 2019-10-10T01:44:54 < kakimir64> if I replace 500GB disks 2019-10-10T01:45:10 < Cracki> where "parity" can be whatever raid6 does and beyond 2019-10-10T01:45:10 < kakimir64> do I go up to 4TB then? 2019-10-10T01:45:23 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Bye!] 2019-10-10T01:45:42 < Cracki> 1+1+2+2 in z2 is (4-2)*1 TB is still 2 TB 2019-10-10T01:46:13 < zyp> the disadvantage with zfs is that it doesn't do reshaping, so you're stuck with the vdev layout you initially choose, all you can do later is add vdevs 2019-10-10T01:46:34 < zyp> so you wanna think through how you want it to be in the future before deciding on what you want to do today 2019-10-10T01:46:41 < kakimir64> fuu fuu 2019-10-10T01:46:42 < Cracki> how does your utility function weigh net storage and reliability? 2019-10-10T01:46:57 < kakimir64> both should be good 2019-10-10T01:47:08 < kakimir64> I think I will buy 2 x 2T more 2019-10-10T01:47:22 < Cracki> for raid levels you can calculate probability of failure from individual probabilities 2019-10-10T01:47:32 < zyp> mirror vdevs are recommended for performance and that's what your current setup is best suited for 2019-10-10T01:47:50 < zyp> but that won't give you more than 50% net capacity 2019-10-10T01:48:18 < zyp> my 16-drive raidz3 got 81% net capacity 2019-10-10T01:48:31 < BrainDamage> remember that raid is mostly about availability rather than resilience 2019-10-10T01:48:31 < kakimir64> I don't need performance 2019-10-10T01:48:36 < kakimir64> my lines are gigabit max 2019-10-10T01:48:48 < BrainDamage> an offline backup is more resilient than most / all raid configs 2019-10-10T01:49:17 < kakimir64> nobody ain't doing offline backups 2019-10-10T01:49:17 < zyp> yeah, raid is not backup 2019-10-10T01:49:27 < kakimir64> it kinda is 2019-10-10T01:49:33 < zyp> no, it's not 2019-10-10T01:49:36 < zyp> it's redundancy 2019-10-10T01:49:38 < kakimir64> when you have the stuff in 2 places 2019-10-10T01:49:51 < Cracki> when you can delete stuff in both places with a single command 2019-10-10T01:50:08 < Cracki> accidental failure is one thing, accidental deletion another 2019-10-10T01:50:25 < zyp> Cracki, but to be fair, you can do snapshots 2019-10-10T01:50:27 < Cracki> intentional failure would be taking a drill to it 2019-10-10T01:50:32 < Cracki> snapshots are awesome 2019-10-10T01:50:33 < zyp> delete by accident, restore from snapshot 2019-10-10T01:50:40 < Cracki> everyone should use snapshots 2019-10-10T01:50:44 < kakimir64> snapshots for important shit 2019-10-10T01:56:08 < kakimir64> ZFS is only as good as the hardware it is put on. Even ZFS can corrupt your data or lose it, if placed on inferior components. Examples of things you don't want to do if you want to keep your data intact include using non-ECC RAM, using non-enterprise disks, using SATA disks behind SAS expanders, using non-enterprise class motherboards, using a 2019-10-10T01:56:08 < kakimir64> RAID card (especially one without a battery), putting the server in a poor environment for a server to be in, etc 2019-10-10T01:56:24 < kakimir64> I check all the boxes 2019-10-10T01:59:17 < kakimir64> literally every box 2019-10-10T02:00:00 < catphish> kakimir64: i just googled SATA SAS expander 2019-10-10T02:00:00 < Cracki> then it's a good thing you are at least doing a raid 2019-10-10T02:00:09 < catphish> i'm struggling to see why it would matter 2019-10-10T02:00:22 < Cracki> sas controllers can talk to sata disks, the connectors are even compatible 2019-10-10T02:00:23 < catphish> but people have written about it, specifically with ZFS 2019-10-10T02:00:27 < kakimir64> can I split dataset any other way than creating directories inside it and then attaching those directories to different shares? 2019-10-10T02:01:29 < zyp> what do you mean? 2019-10-10T02:02:52 < kakimir64> I mean one pool but multiple shares and permissions and quotas 2019-10-10T02:03:18 < kakimir64> like home directories in desktop computer 2019-10-10T02:03:44 < kakimir64> can I create dataset inside dataset? 2019-10-10T02:03:59 < catphish> i'm not sure i understand 2019-10-10T02:04:07 < catphish> but zfs has nested subvolumes 2019-10-10T02:04:13 < zyp> a pool gets a root filesystem and then you can create more filesystems under it, which may or may not be mounted as a subdir 2019-10-10T02:04:31 < zyp> and you can also create block devices under it 2019-10-10T02:04:36 < catphish> which are basically just directories, but can have their own snashots 2019-10-10T02:04:51 < zyp> catphish, and settings 2019-10-10T02:04:54 < catphish> and of course with linux you can mount a totally different filesystem inside 2019-10-10T02:04:56 < zyp> like compression, etc… 2019-10-10T02:04:58 < kakimir64> pool != dataset 2019-10-10T02:05:04 < kakimir64> ? 2019-10-10T02:05:08 < catphish> i don't know what you mean by dataset 2019-10-10T02:05:20 < kakimir64> hey.. me neather 2019-10-10T02:05:31 < zyp> hmm 2019-10-10T02:05:39 < kakimir64> dataset has a mount point at leat 2019-10-10T02:05:42 < kakimir64> least 2019-10-10T02:05:54 < catphish> i need a really badass server project to work on soon 2019-10-10T02:06:11 < catphish> when i'm done with my hardware stuff 2019-10-10T02:06:20 < zyp> apparently on my server, I have one zpool containing 453 filesystems and 4 block devices 2019-10-10T02:06:42 < zyp> kakimir64, so by dataset you mean filesystem 2019-10-10T02:07:52 < kakimir64> 453 filesystems? 2019-10-10T02:07:58 < zyp> yes 2019-10-10T02:08:09 < kakimir64> explain 2019-10-10T02:08:26 < catphish> what on earth is a filesystem in zfs? 2019-10-10T02:08:36 < catphish> i treat one zfs as a filesystem 2019-10-10T02:08:44 < kakimir64> https://www.youtube.com/watch?v=JYqjcHYTQgQ 2019-10-10T02:08:49 < zyp> catphish, you called it subvolume 2019-10-10T02:09:06 < zyp> root@utsuho:~# zfs list | wc -l 2019-10-10T02:09:06 < zyp> 458 2019-10-10T02:09:12 < zyp> first line is header, so 457 2019-10-10T02:09:21 < catphish> zyp: oh yeah, i called it a subvolume :) 2019-10-10T02:10:26 < catphish> oh yeah, my scripts make them with "zfs create", never knew they were called filesystems 2019-10-10T02:10:37 < zyp> so anyway, of the 457, four are block devices containing VM images 2019-10-10T02:10:41 < catphish> my comment says "create a volume" 2019-10-10T02:11:07 < catphish> "block device" is a fixed size file in the root of the pool? 2019-10-10T02:11:38 < catphish> like a lvm volume? 2019-10-10T02:11:59 < catphish> i use zfs but only explored the specific bits i needed 2019-10-10T02:12:11 < zyp> it's not a file, it's a proper block devixe 2019-10-10T02:12:59 < catphish> the difference is a little moot, but sure 2019-10-10T02:13:12 < zyp> root@utsuho:~# ls -l /dev/zvol/zp0/vm/fpgabuild 2019-10-10T02:13:12 < zyp> lrwxrwxrwx 1 root root 12 Oct 9 19:59 /dev/zvol/zp0/vm/fpgabuild -> ../../../zd0 2019-10-10T02:13:15 < kakimir64> no compression and shit zyp? 2019-10-10T02:13:15 < zyp> root@utsuho:~# ls -l /dev/zd0 2019-10-10T02:13:18 < zyp> brw-rw---- 1 root disk 230, 0 Oct 9 19:59 /dev/zd0 2019-10-10T02:13:25 < catphish> it's presented as a block device rather than over VFS 2019-10-10T02:13:38 < zyp> IIRC I have lz4 compression enabled on the root, so all sub-filesystems inherits it 2019-10-10T02:13:38 < catphish> that's pretty cool 2019-10-10T02:13:58 < catphish> i might start using zfs more, mostly for its snapshots 2019-10-10T02:14:25 < catphish> right now i use it dor rsync backup targets 2019-10-10T02:14:29 < catphish> *for 2019-10-10T02:14:38 < zyp> the reason I have a ton of filesystems is because I'm running docker on top of it, both images and volumes 2019-10-10T02:14:44 < catphish> rsync->snapshot->rsync->snapshot 2019-10-10T02:15:29 < zyp> so I have 23 docker-volumes and 422 docker images 2019-10-10T02:15:32 < catphish> i'm also looking at using it for block based backups, same thing, copy blocks, snapshot, repeat 2019-10-10T02:15:46 < zyp> each docker layer gets its own filesystem 2019-10-10T02:15:57 < catphish> zyp: sounds good 2019-10-10T02:16:18 < zyp> I think it does some tricks with snapshots as well, but I haven't looked into the details 2019-10-10T02:16:26 < catphish> does everyone here secretly have a devops dayjob? 2019-10-10T02:16:41 < zyp> I don't, I just do devops as hobby :p 2019-10-10T02:16:55 < catphish> i did, they people kept giving me money 2019-10-10T02:17:02 < catphish> now it's a job :( 2019-10-10T02:17:07 < catphish> *then 2019-10-10T02:17:09 < zyp> aww 2019-10-10T02:17:30 < zyp> I have a hobby kubernetes cluster I've been fucking around with too 2019-10-10T02:17:56 < catphish> hopefully one day i'll go back to doing it for fun once the work settles down 2019-10-10T02:18:53 < catphish> my usb stuff works great now by the way, though i gave up bothering with a SCSI layer 2019-10-10T02:19:14 < kakimir64> Exec on or off? 2019-10-10T02:19:26 < kakimir64> if process can be executed from this dataset 2019-10-10T02:19:38 < kakimir64> executed on what? nas server? 2019-10-10T02:19:44 < catphish> yes 2019-10-10T02:19:54 < kakimir64> would I want that 2019-10-10T02:20:20 < catphish> will you put binaries on it that need to run on that host? 2019-10-10T02:20:31 < kakimir64> most likelly not 2019-10-10T02:20:54 < catphish> personally i've never disabled it, but if its purely a fileserver i see no reason to allow it 2019-10-10T02:20:56 < kakimir64> it's only purpose would be to act as nas 2019-10-10T02:21:21 < catphish> might as well turn it off, pretty sure it won't impact the ability of the NAS clients executing stuff off it 2019-10-10T02:21:28 < catphish> just stops things executing locally from it 2019-10-10T02:21:29 < kakimir64> what is copies - number of copies on this dataset 2019-10-10T02:21:38 < catphish> redundancy i think 2019-10-10T02:22:07 < catphish> ie the number of different disks each block of data is written to 2019-10-10T02:22:35 < kakimir64> oh 2019-10-10T02:22:41 < kakimir64> 1? 2019-10-10T02:22:45 < kakimir64> default: 1 2019-10-10T02:23:26 < catphish> https://jrs-s.net/2016/05/02/zfs-copies-equals-n/ 2019-10-10T02:25:10 < catphish> "In summary, copies is useful for specifying different redundancy policies for datasets, but it is not a replacement for proper mirroring or raidz" 2019-10-10T02:25:23 < catphish> so yeah, use proper mirrroring and ignore that 2019-10-10T02:28:00 < kakimir64> poor man's raid 2019-10-10T02:35:57 < kakimir64> add zvol button adds block devices? 2019-10-10T02:40:53 < kakimir64> probs 2019-10-10T02:46:31 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-10T02:47:10 < kakimir64> do you guise use NFS? 2019-10-10T02:52:54 < kakimir64> these settings :o 2019-10-10T02:53:20 < kakimir64> samba is easy 2019-10-10T02:57:53 -!- ABLomas [abl@78-58-248-227.static.zebra.lt] has quit [Ping timeout: 276 seconds] 2019-10-10T02:58:13 -!- ABLomas [abl@78-58-248-227.static.zebra.lt] has joined ##stm32 2019-10-10T03:01:33 < Laurenceb> I used to use samba for work 2019-10-10T03:01:41 < Laurenceb> now we use office 360 O_o 2019-10-10T03:04:18 < kakimir64> fun 2019-10-10T03:04:30 < Laurenceb> babbyshake 2019-10-10T03:04:53 < Laurenceb> only one more month 2019-10-10T03:09:01 -!- Simon-- [~sim@2606:6a00:0:28:5604:a6ff:fe02:702b] has joined ##stm32 2019-10-10T03:11:30 < Laurenceb> then I can regain my sanity 2019-10-10T03:20:20 < Laurenceb> maybe I could make a joker spinoff where I turn into a crazy clown due to babbyshake 2019-10-10T03:22:01 < Laurenceb> "Shaker" 2019-10-10T03:22:05 < Laurenceb> I like this plan 2019-10-10T03:24:15 -!- grindhold [~quassel@mail.skarphed.org] has quit [Quit: No Ping reply in 180 seconds.] 2019-10-10T03:25:35 -!- grindhold [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-10T03:27:57 < Laurenceb> muhahahaha App APP... BIG DATA hahaha muhahaha Crowd Sourcing Deep learning Database Hadoop 2019-10-10T03:28:23 < Laurenceb> are you enjoying the internet of things yet??? 2019-10-10T03:52:45 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-10T03:53:03 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-10T03:53:47 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-10T03:57:45 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-10T04:06:27 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 268 seconds] 2019-10-10T04:38:29 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T06:05:39 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Quit: Leaving] 2019-10-10T06:06:14 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-10T06:28:37 < Cracki> scheduled email sending is nice. I can hide my fucked up sleep cycle while sending out invoices. 2019-10-10T06:34:56 -!- fc5dc9d4_ [~quassel@p5B0812C4.dip0.t-ipconnect.de] has joined ##stm32 2019-10-10T06:38:37 -!- fc5dc9d4 [~quassel@p5B08131E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-10T07:11:06 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Read error: Connection reset by peer] 2019-10-10T07:20:15 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has joined ##stm32 2019-10-10T07:21:02 < aandrew> hm, the output of this GPSDO is a 10MHz sine wave 2019-10-10T07:22:40 < aandrew> pk-pk is 3.6V, but the scope is claiming cycle RMS is 1.24V? 3.6 / sqrt(2) is 2.546, so I guess cycle RMS is half that? 2019-10-10T07:23:38 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-10T07:25:08 < aandrew> no, cycle rms is supposed to just be measured for the first complete cycle on the screen 2019-10-10T07:25:46 < aandrew> either way it's way too hot for the input for this AD9959 2019-10-10T07:27:14 < aandrew> hm, wont' matter, AD9959 minimum refclk is 20MHz anyway 2019-10-10T07:53:45 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-10T07:57:05 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-10T07:57:06 -!- day__ is now known as day 2019-10-10T08:22:35 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-10T08:23:18 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-10T08:31:57 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-10T08:47:48 -!- jly [uid355225@gateway/web/irccloud.com/x-vgcaqxgssjfblrsv] has joined ##stm32 2019-10-10T08:52:11 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-10T08:57:31 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-10T08:59:40 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-10T09:00:20 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T09:21:42 < jpa-> aandrew: yeah, RMS is 1/sqrt(2) times the amplitude, and amplitude is half of peak-to-peak voltage 2019-10-10T09:22:44 < aandrew> I must be misremembering. I thought (for a sine wave) RMS was .707*pk-pk. .707 is 1/sqrt(2) like you said 2019-10-10T09:23:17 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-10T09:23:30 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T09:31:57 < jly> ok 2019-10-10T09:32:43 < jly> i just remember .707 or 1.414 (depending on which way you're going) 2019-10-10T09:36:13 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-10T09:37:48 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-10T10:03:05 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-10T10:03:42 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-10T10:18:15 < jly> hi Haohmaru 2019-10-10T10:22:38 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Read error: Connection reset by peer] 2019-10-10T11:15:55 < englishman> square rekt 2019-10-10T11:36:31 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-10T11:45:47 < englishman> https://leahneukirchen.org/blog/archive/2019/10/ken-thompson-s-unix-password.html 2019-10-10T12:27:35 -!- jly [uid355225@gateway/web/irccloud.com/x-vgcaqxgssjfblrsv] has quit [Quit: Connection closed for inactivity] 2019-10-10T12:38:28 -!- grummund [~grummund@unaffiliated/grummund] has quit [Remote host closed the connection] 2019-10-10T12:55:04 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-10T13:00:29 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 268 seconds] 2019-10-10T13:09:07 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 268 seconds] 2019-10-10T13:18:19 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-10T13:26:35 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-10T13:27:09 < zyp> aandrew, if you have 3.6V peak to peak and rectify it (ignoring diode losses), you flip up the negative halfs to get 1.8V peak to peak 2019-10-10T13:27:47 < zyp> then picture you cut off the tops of the peaks and use that to fill up the valleys and you've got RMS 2019-10-10T13:29:35 < jadew> aandrew, then maybe you don't have a sine wave? 2019-10-10T13:29:40 < jadew> how distorted is the output? 2019-10-10T13:31:25 < jadew> for a perfect square wave the RMS value will be 1/2 of the pk-pk one 2019-10-10T13:39:02 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 265 seconds] 2019-10-10T13:52:29 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-10T14:44:22 < qyx> or just imagine icecream melting 2019-10-10T15:03:10 < zyp> that doesn't have anything to do with what was said 2019-10-10T15:09:53 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Quit: quit has dan2wik!] 2019-10-10T15:10:59 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-10T15:10:59 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-10T15:10:59 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-10T15:14:37 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-10T15:36:14 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-10T15:40:39 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-10T15:40:54 < kakimir32> yes 2019-10-10T15:41:49 < kakimir32> I wonder if my legit jlink has same pinout than chinklink 2019-10-10T15:49:07 < kakimir32> looks like matching pinout ye 2019-10-10T15:49:32 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-10T15:51:26 < jadew> amazing Whitney Huston remix (1990): https://www.youtube.com/watch?v=JR49dyo-y0E 2019-10-10T15:51:40 < jadew> I'm surprised how fresh they made it sound 2019-10-10T15:51:57 < jadew> (she's dead btw) 2019-10-10T15:52:43 < kakimir32> it's been on radio for months 2019-10-10T15:52:48 < kakimir32> but yes 2019-10-10T15:52:58 < jadew> I just discovered it 2019-10-10T15:55:26 < kakimir32> yes 2019-10-10T15:55:56 < kakimir32> aka. pop music 2019-10-10T15:56:09 < kakimir32> sometimes it's rather good 2019-10-10T15:56:43 < jadew> I like pop 2019-10-10T15:57:25 < jadew> I like any song that sounds good 2019-10-10T15:57:53 < jadew> if you're ignoring nice things based on ideology or fixed ideas, you're missing out 2019-10-10T15:58:50 < jadew> here's some with just vocals: https://www.youtube.com/watch?v=gJ_71rROxcs 2019-10-10T15:59:37 < jadew> Karma Police by Knauskoret 2019-10-10T16:00:24 < jadew> I think they're finnish, but I'm not sure 2019-10-10T16:00:47 < jadew> norwegian 2019-10-10T16:02:05 < kakimir32> never heard 2019-10-10T16:03:55 < dongs> hello jews 2019-10-10T16:05:19 < kakimir32> hello dongs 2019-10-10T16:05:34 < kakimir32> when you go to alaska? 2019-10-10T16:06:32 < kakimir32> idk. 2019-10-10T16:06:34 < dongs> https://vocaroo.com/i/s0NFHIKmj468 can one of you fags identify waht language this is 2019-10-10T16:06:51 < dongs> (you will obviosuly need working audio and speakers, so lunix users prolly dont need apply 2019-10-10T16:07:19 < kakimir32> sounds eastern european definitelly 2019-10-10T16:09:01 < kakimir32> dongs: do you know the answer? 2019-10-10T16:10:20 < kakimir32> is there a prize for right answer? 2019-10-10T16:11:04 < dongs> no, i want to know the language and translation 2019-10-10T16:11:27 < kakimir32> why 2019-10-10T16:11:33 < dongs> idk, i wanna know what it says 2019-10-10T16:11:38 < dongs> maybe he wants to shootup another mosque 2019-10-10T16:12:32 < sync> no 2019-10-10T16:12:43 < sync> I think it is kakimir-istan or one one of his neighbors 2019-10-10T16:13:04 < kakimir32> not nordic language at least 2019-10-10T16:13:27 < dongs> it almost sounded korean to me 2019-10-10T16:13:32 < dongs> but my gook pal said no 2019-10-10T16:13:57 < kakimir32> maybe greenlandic 2019-10-10T16:13:58 < dongs> the beginning intonation and shit sounds very korean but i donno 2019-10-10T16:14:01 < kakimir32> inuit language 2019-10-10T16:14:44 -!- comptroller [~comptroll@47-213-222-253.paolcmtc01.res.dyn.suddenlink.net] has quit [Ping timeout: 268 seconds] 2019-10-10T16:15:09 < dongs> how would you even do that 2019-10-10T16:15:14 < dongs> lunix can't play and record audio at same time 2019-10-10T16:15:32 < dongs> for dicatation? 2019-10-10T16:15:36 < dongs> you gotta talk super slowly and shit 2019-10-10T16:16:29 < dongs> no 2019-10-10T16:16:32 < dongs> windows 10 1903 of course 2019-10-10T16:16:57 < kakimir32> obsession 2019-10-10T16:20:38 < kakimir32> first half second I thought it might be finnish, second half second I thought it might be estonian but then I realized it is not 2019-10-10T16:21:24 < kakimir32> https://en.wikipedia.org/wiki/Uralic_languages 2019-10-10T16:22:31 < dongs> is it one of those? 2019-10-10T16:23:09 < kakimir32> not the ones I know 2019-10-10T16:27:08 < PaulFertser> It seems to have sounds similar to Ukrainian but it's definitely not it. Probably jadew knows? 2019-10-10T16:28:52 < kakimir32> first half sounds like some ahmed language (unclear babbling) and in second half sounds are similar than in finnish language 2019-10-10T16:29:18 < kakimir32> maybe look into balkan area 2019-10-10T16:30:26 < kakimir32> maybe jadew language? 2019-10-10T16:31:37 < kakimir32> *ahmed language babbling without excessive throat sounds 2019-10-10T16:34:18 < jadew> PaulFertser, no, I don't know 2019-10-10T16:34:29 < jadew> but yeah, it sounds like something from around here 2019-10-10T16:39:29 -!- mirage335 [~mirage335@2001:470:8ede:0:216:3eff:fe97:ac6d] has quit [Ping timeout: 276 seconds] 2019-10-10T16:39:55 < kakimir32> voice in that sample sounds like a white man 2019-10-10T16:40:53 < kakimir32> it just sounds european somehow 2019-10-10T16:41:19 < srk> haha, my lunex currently handles 4 soundcards - hdmi, spdif dac, usb soundblaster and usb focusrite, none of them required installing any drivers :D 2019-10-10T16:42:04 < srk> instead of switching jacks I'm switching cards in pulseaudio 2019-10-10T16:42:28 < jadew> must be really old stuff 2019-10-10T16:42:39 < jadew> either that or it's not handling them well 2019-10-10T16:43:06 < srk> none of them is old, you can still buy any of them 2019-10-10T16:43:19 < srk> it's handling them with no issues at all 2019-10-10T16:43:44 < jadew> that is surprising 2019-10-10T16:44:08 < srk> if looking for pro cards supported everywhere focusrite is quite nice 2019-10-10T16:44:59 < srk> asus xonar as well 2019-10-10T16:47:11 -!- mirage335 [~mirage335@2001:470:8ede:0:216:3eff:fe97:ac6d] has joined ##stm32 2019-10-10T16:47:50 < sync> only that unfortunately the fuckusrite is a nice distortion generator 2019-10-10T16:52:26 < srk> I wanted to buy older ESI PCI rack mounted soundcard from ebay to have enough inputs but wasn't sure about driver support 2019-10-10T16:53:36 < srk> this one iirc https://www.esi-audio.com/products/esp1010/ 2019-10-10T16:53:37 < srk> not sure 2019-10-10T16:53:38 < srk> aha 2019-10-10T16:53:43 < srk> yeah 2019-10-10T17:04:50 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-10T17:05:41 < kakimir32> dongs: where did you get the audio? 2019-10-10T17:09:23 < kakimir32> what was I doing before this.. 2019-10-10T17:09:38 < dongs> prolly your mom 2019-10-10T17:11:05 < kakimir32> nope 2019-10-10T17:11:21 < kakimir32> jlink 2019-10-10T17:11:32 < kakimir32> howe is innovation dongbro? 2019-10-10T17:11:47 < kakimir32> any cool projects? 2019-10-10T17:16:17 < dongs> nah just landed in taiwan 2019-10-10T17:18:58 < kakimir64> cool 2019-10-10T17:24:52 < kakimir64> jlink propper flashing confirmed 2019-10-10T17:25:12 < kakimir64> no problems 2019-10-10T17:25:26 < aandrew> jadew: no, it's sine. zyp's right, I'm just failing at remembering my basic electricity theory 2019-10-10T17:26:08 < aandrew> kakimir64: that would drive me mental if a clone had a different pinout. lol 2019-10-10T17:26:24 < kakimir64> pinout was the same 2019-10-10T17:26:51 < kakimir64> I just pulled cable from clone and jammed it into jlink 2019-10-10T17:27:01 < aandrew> so no worries then 2019-10-10T17:27:12 < kakimir64> anyways it's ready now for the codez 2019-10-10T17:27:14 < aandrew> I think I have a half dozen JLink base 8.x clones 2019-10-10T17:27:19 < aandrew> they all seem to work reasonably well 2019-10-10T17:27:21 < kakimir64> me too 2019-10-10T17:27:30 < aandrew> SET US UP THE BOMB 2019-10-10T17:28:10 < kakimir64> 00001ff0: adad adad adad adad adad adad adad adad ................00002000: ffff ffff ffff ffff ffff ffff ffff ffff ................ 2019-10-10T17:28:17 < kakimir64> sweet 2019-10-10T17:28:47 < kakimir64> now I just need to unfuk eclipse workspace and project 2019-10-10T17:29:19 < aandrew> > eclipse 2019-10-10T17:29:21 < aandrew> you're already fucked 2019-10-10T17:30:03 < kakimir64> some javaruntimeexprillegal 2019-10-10T17:30:09 < kakimir64> build failed 2019-10-10T17:30:18 < kakimir64> CDT related problemk 2019-10-10T17:30:44 < kakimir64> I see build actually not failing 2019-10-10T17:32:05 < kakimir64> https://paste.ee/p/vkGhA importing dictionary.txt caused fun relative path thing here 2019-10-10T17:33:20 < kakimir64> interesting thing that there is files that can be poked by user in /usr/local 2019-10-10T17:34:35 < kakimir64> I took old lpcxpresso workspace from windowzer machine 2019-10-10T17:35:06 -!- renn0xtk9 [~max@2a02:810d:1540:2448:f0bb:33e:99e3:d860] has joined ##stm32 2019-10-10T17:35:27 < aandrew> this is the top reason why I hate all IDEs. your work is fine but because some hidden fucking detail is not perfect the IDE tells you to get fucked 2019-10-10T17:35:56 < kakimir64> there is multiple details 2019-10-10T17:35:57 < aandrew> I've got enough shit to do without trying to appease the C gods, the STM32 gods *and* the fucking ecli..javanullpointerexception 2019-10-10T17:36:09 < kakimir64> multiple relative paths not set correctly etc. 2019-10-10T17:36:22 < kakimir64> some absolute paths 2019-10-10T17:36:39 < srk> aandrew: lol! 2019-10-10T17:37:46 < emeb> aandrew: editor/make/gcc/gdb for me! 2019-10-10T17:38:02 < Ecco> hi :) 2019-10-10T17:38:14 < Ecco> GCC says "switch -mcpu=cortex-m7 conflicts with -march=armv7e-m" 2019-10-10T17:38:18 < Ecco> Does that make any sense? 2019-10-10T17:38:46 < emeb> I've worked with various eclipse-based IDEs and keeping track of all the configs in those tools is impossible. 2019-10-10T17:38:58 < PaulFertser> Ecco: Hi :) I'd suggest gcc -print-multi-lib and just use one of the sets shown. 2019-10-10T17:39:20 < Ecco> huh 2019-10-10T17:39:23 < Ecco> Thanks PaulFertser 2019-10-10T17:39:26 < Ecco> Not sure what that does though 2019-10-10T17:39:59 < Ecco> I think the line I'd be interested in would be "thumb/v7e-m+fp/hard;@mthumb@march=armv7e-m+fp@mfloat-abi=hard 2019-10-10T17:40:39 < PaulFertser> Ecco: when gcc is built in a way to allow cross-compilation for cortex-m based microcontrollers all of the libraries (libgcc and libstdc++ etc) are built multiple times, for each possible target. 2019-10-10T17:41:02 < PaulFertser> Ecco: so I'd use -mthumb -march=armv7e-m+fp -mfloat-abi=hard 2019-10-10T17:41:33 < Ecco> yes indeed 2019-10-10T17:41:47 < Ecco> that's pretty much what I did 2019-10-10T17:41:54 < Ecco> but I also added -mcpu=cortex-m7 2019-10-10T17:42:02 < PaulFertser> Just do not do that :) 2019-10-10T17:42:09 < Ecco> I don't know if that's needed or makes any sense 2019-10-10T17:42:30 < Ecco> But I think that GCC says that specifying mcpu allows it to make finer optimization specifically for the given CPU 2019-10-10T17:42:58 < PaulFertser> But if multilib doesn't have that... 2019-10-10T17:43:21 < Ecco> well, then I would expect a link-time failure? 2019-10-10T17:43:33 < Ecco> Also, code emitted with or w/o mcpu should be able to interop 2019-10-10T17:43:46 < Ecco> IIRC, the problem with mcpu is that code might be less efficient on a different cpu 2019-10-10T17:44:29 < PaulFertser> I understand your reasoning, and I have no more arguments to support my idea :) 2019-10-10T17:44:33 < Ecco> :-D 2019-10-10T17:44:41 < Ecco> https://gcc.gnu.org/onlinedocs/gcc/ARM-Options.html 2019-10-10T17:45:22 < Ecco> yeah maybe 2019-10-10T17:45:35 < aandrew> emeb: same. 2019-10-10T17:45:43 < PaulFertser> Hm, probably you should ask on Linaro channel Ecco . 2019-10-10T17:45:54 < Ecco> https://stackoverflow.com/questions/41934764/gnu-arm-warning-mcpu-cortex-r5-conflicts-with-march-armv7-r-switch 2019-10-10T17:46:13 < Ecco> So there's mtune too :) 2019-10-10T17:46:30 < Ecco> ok, apparently, mcpu = generate code that has to work *only* on given CPU 2019-10-10T17:46:38 < Ecco> so in a way, march is less restrictive than mcpu 2019-10-10T17:47:03 < Ecco> of course you need the appropriate libgcc, but I think it's ok if libgcc has been built using march while the rest of the code has been built using mcpu 2019-10-10T17:47:20 < emeb> When I know the hardware is cortex m7, why not use mcpu? 2019-10-10T17:47:34 < Ecco> it's exactly the use case 2019-10-10T17:47:44 < Ecco> if you're 100% sure your code will run on a given CPU, use mcpu 2019-10-10T17:47:56 < Ecco> if you want the same binary to run on a whole class of CPUs, use march 2019-10-10T17:47:57 < PaulFertser> avrgcc multilib uses -mmcu :) 2019-10-10T17:48:39 < PaulFertser> emeb: the question is whether it'll link to the proper multilib. 2019-10-10T17:49:12 < emeb> PaulFertser: Ah. Well, it works so I guess I'm OK. :) 2019-10-10T17:49:33 * emeb whistles in blissful ignorance. 2019-10-10T17:51:29 < PaulFertser> Ecco: it would be nice if you asked linaro folks to enlighten us properly :) 2019-10-10T17:51:32 * emeb does -print-multi-lib. gawks at all the options. 2019-10-10T18:12:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T18:21:47 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-10T18:54:16 < Ecco> < rth> Ecco, you are correct, -mcpu specifies more than -march 2019-10-10T18:54:39 < Ecco> < rth> Ecco, the conflict is that -mcpu=cortex-m7 uses "-march=armv7e-m+fp.dp" 2019-10-10T19:01:38 < karlp> read https://launchpadlibrarian.net/287100883/readme.txt 2019-10-10T19:01:45 < karlp> (it's the readme.txt in your toolchain 2019-10-10T19:03:50 -!- learningc [~learningc@121.121.98.53] has quit [Ping timeout: 265 seconds] 2019-10-10T19:26:45 < emeb> whew - based on that table I'm doing the right thing. 2019-10-10T19:27:32 < emeb> but I do wonder what the difference is between fpv5-sp-d16 and fpv5-d16. I assume the 2nd form just enables double precision? 2019-10-10T19:31:31 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 250 seconds] 2019-10-10T19:32:08 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-10T19:34:35 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-10T19:38:36 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-10T19:38:38 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 265 seconds] 2019-10-10T19:39:57 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-10T19:41:48 -!- syn0 [hoofman@sdf-eu.org] has joined ##stm32 2019-10-10T19:41:54 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-10T19:43:28 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 265 seconds] 2019-10-10T19:45:38 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 268 seconds] 2019-10-10T19:51:00 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-10T19:54:21 < antto> doesn't sound slavic tbh 2019-10-10T20:11:42 < karlp> emeb: I'm 99% sure that fpv5-sp-d16 is an alias for fpv5-d16 2019-10-10T20:11:46 < karlp> and fpv5-d16 is now legacy 2019-10-10T20:11:59 < karlp> after they introduced f7s with both single and double precision 2019-10-10T20:12:09 < karlp> the double one is fpv5-dp-d16 2019-10-10T20:12:37 < emeb> karlp: interesting 2019-10-10T20:13:08 < emeb> so does fpv5-sp-d16 and fpv5-d16 prevent using double precision? 2019-10-10T20:13:08 < karlp> hrm. 2019-10-10T20:13:32 < karlp> unless they really do mean that fpv5-d16 is double precision and fpv5-sp-d16 is single. 2019-10-10T20:13:36 < karlp> that would be inconsistent and lame 2019-10-10T20:13:51 < karlp> re-reading it though, I think they must. 2019-10-10T20:14:00 < karlp> boo 2019-10-10T20:14:05 < emeb> that's what I was thinking. 2019-10-10T20:38:48 < srk> is this legit? https://github.com/HaskellEmbedded/data-stm32/blob/master/src/Data/STM32/Core.hs#L49 2019-10-10T20:39:10 < srk> or is better to use fpv5-d16? 2019-10-10T20:39:14 < srk> or 4 2019-10-10T20:43:18 < srk> looks like sp for m4 https://www.mikrocontroller.net/articles/ARM_GCC#Compiler_.26_Linker_Flags 2019-10-10T20:44:32 < srk> ah, only some f7s are double precision :( https://www.st.com/en/microcontrollers-microprocessors/stm32f7-series.html 2019-10-10T20:45:00 < srk> \o/ 765 I've picked is double 2019-10-10T21:08:58 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-10T21:09:02 < Cracki> does it say how sp/dp performance compares? 2019-10-10T21:17:10 < srk> not really 2019-10-10T21:30:06 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-10T21:32:57 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-10T21:43:47 < Cracki> casting couch https://imgur.com/gzOnLpn 2019-10-10T21:51:06 < kakimir64> I ran out of snus 2019-10-10T21:51:10 < kakimir64> situation bad 2019-10-10T21:51:28 < kakimir64> Cracki :o 2019-10-10T21:57:13 < kakimir64> problem in eclipse is in .metadata slash plugins 2019-10-10T21:57:16 < kakimir64> somewhere 2019-10-10T21:57:27 < kakimir64> can I just nuke this dir? 2019-10-10T21:57:38 < kakimir64> as in plugins folder 2019-10-10T21:57:43 < Cracki> just rename maybe 2019-10-10T21:58:03 < Cracki> it's ok to nuke but you might lose some settings nobody ever cares about 2019-10-10T21:58:09 < kakimir64> I don't do sensible things cracki 2019-10-10T21:58:17 < kakimir64> it's all full power always 2019-10-10T21:58:21 < kakimir64> or not at all 2019-10-10T21:58:40 < kakimir64> it's rm -rf with sudos ofc. 2019-10-10T22:02:27 < kakimir64> oh wow mr. robot 4th season is now 2019-10-10T22:02:33 < kakimir64> in us at least 2019-10-10T22:11:08 < Cracki> funded by evil corp 2019-10-10T22:12:06 -!- renn0xtk9 [~max@2a02:810d:1540:2448:f0bb:33e:99e3:d860] has quit [Quit: Konversation terminated!] 2019-10-10T22:15:01 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3d96:fe58:e1d4:e3d5] has joined ##stm32 2019-10-10T22:26:36 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T22:33:14 -!- grindhold_ [~quassel@185.163.117.141] has joined ##stm32 2019-10-10T22:46:49 -!- jly [uid355225@gateway/web/irccloud.com/x-exvwhadlisomponx] has joined ##stm32 2019-10-10T22:48:32 -!- baenana is now known as beanana 2019-10-10T22:48:58 -!- beanana is now known as beanienana 2019-10-10T22:53:53 < kakimir64> I fixed eclipse workspace 2019-10-10T22:53:58 < kakimir64> time to fix projects then 2019-10-10T22:54:47 < kakimir64> build fails but no java bullshit poppin up 2019-10-10T22:54:59 < kakimir64> situation normal 2019-10-10T22:55:56 < kakimir64> never move workspace 2019-10-10T22:56:11 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-10T22:56:12 < Cracki> have you tried keil 2019-10-10T22:56:19 < kakimir64> you can use it with program you want but when you move the workspace it's fukd 2019-10-10T22:56:33 < kakimir64> no 2019-10-10T22:56:52 < Cracki> java is very enterprise 2019-10-10T22:56:54 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-10T22:57:14 < Cracki> it assumes that you program java, which is why everything about eclipse feels foreign 2019-10-10T23:01:52 -!- oz4ga [~tim@unaffiliated/oz4ga] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-10T23:08:21 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3d96:fe58:e1d4:e3d5] has quit [Ping timeout: 246 seconds] 2019-10-10T23:11:22 < kakimir64> I had to trick it to regenerate makefiles 2019-10-10T23:12:26 < kakimir64> by adding POOPOO to symbols 2019-10-10T23:22:56 -!- oz4ga [~tim@bestie.edr.dk] has joined ##stm32 2019-10-10T23:22:57 -!- oz4ga [~tim@bestie.edr.dk] has quit [Changing host] 2019-10-10T23:22:57 -!- oz4ga [~tim@unaffiliated/oz4ga] has joined ##stm32 2019-10-10T23:30:58 < kakimir64> okay it doesn't build all things i want to link 2019-10-10T23:31:35 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-10T23:43:52 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-10T23:52:44 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3d96:fe58:e1d4:e3d5] has joined ##stm32 2019-10-10T23:52:59 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 --- Day changed Fri Oct 11 2019 2019-10-11T00:00:44 -!- renn0xtk9 [~max@2a02:810d:1540:2448:3d96:fe58:e1d4:e3d5] has quit [Quit: Konversation terminated!] 2019-10-11T00:10:17 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-11T00:22:45 -!- grindhold_ [~quassel@185.163.117.141] has quit [Read error: Connection reset by peer] 2019-10-11T00:22:59 -!- grindhold_ [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-11T00:28:26 -!- grindhold_ [~quassel@mail.skarphed.org] has quit [Ping timeout: 240 seconds] 2019-10-11T00:30:05 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-11T00:35:12 < kakimir64> https://stackoverflow.com/questions/7728878/project-dependency-in-eclipse-cdt eclipse pro hack 2019-10-11T00:38:31 < Steffanx> Bad hack. 2019-10-11T00:39:45 < karlp> so, stenciling went, pretty well at least. 2019-10-11T00:39:56 < Steffanx> I'm not 100% convinced this is still relevant, kakimir32 2019-10-11T00:40:19 < karlp> qfp 0.5mm pitch got a lot of bridges though, I'm guessing, is that from the stencil not being firmly down enough first? 2019-10-11T00:40:31 < karlp> so it got more through that it should have? 2019-10-11T00:41:03 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-11T00:41:23 < kakimir64> Steffanx: it isn't 2019-10-11T00:41:49 < kakimir64> fatal error: reent.h: No such file or directory 2019-10-11T00:41:53 < kakimir64> now I'm coding 2019-10-11T00:44:21 < karlp> I jsut cut a hole in the cardboard it came with to use as a pcb frame, and taped the stencil down, so.... 2019-10-11T00:44:39 < karlp> it certainly took a few passes of squeegee, and iirc, it's meant to be "one pass! no more!" 2019-10-11T00:44:58 < karlp> is my other option, "more paste upfront, so youonly need one pass" ? 2019-10-11T00:45:31 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-11T00:46:31 < englishman> how did the paste look before reflow 2019-10-11T00:46:49 < englishman> it should like perfect little brick shaped turds 2019-10-11T00:47:42 < englishman> https://pbs.twimg.com/media/D9BemJYUEAARVRt.jpg 2019-10-11T00:47:55 < englishman> example turds 2019-10-11T00:48:19 < englishman> and yeah one pass 2019-10-11T00:48:54 < englishman> then lift the stencil up perfectly, if you do it by hand it helps to make a tape hinge 2019-10-11T00:50:02 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-11T00:52:07 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-11T00:54:02 -!- kow__ [~iccy@135.0.26.39] has quit [Ping timeout: 240 seconds] 2019-10-11T01:14:32 < karlp> paste _looked_ nice, but I'm presuming it wsa a little high maybe? 2019-10-11T01:14:38 < karlp> yeah, it was tape hinged 2019-10-11T01:14:44 < karlp> I'm trying to find the video of what I copied 2019-10-11T01:14:57 < karlp> I definitely didn't put enough paste across theh top of the stencil then I guess. 2019-10-11T01:16:25 -!- jly [uid355225@gateway/web/irccloud.com/x-exvwhadlisomponx] has quit [Quit: Connection closed for inactivity] 2019-10-11T01:20:45 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-11T01:32:50 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-11T01:33:26 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-11T01:35:04 -!- kow_ [~iccy@135.0.26.39] has joined ##stm32 2019-10-11T01:36:35 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-11T01:40:04 -!- grindhold_ [~quassel@mail.skarphed.org] has joined ##stm32 2019-10-11T01:47:30 < zyp> karlp, common problem 2019-10-11T01:47:43 < zyp> I usually need to remove some bridges on qfp 0.5 too 2019-10-11T01:51:31 < karlp> well, I wouldn't mention it, but qfp0.5 was the _best_ bit of just dragging a line from the syringe across the each side and being 99% 2019-10-11T01:51:47 < karlp> I got... more, definitely more bridges from the stencil. 2019-10-11T01:52:05 < karlp> still, was _super_ nice for all thhe rest, much prettyier, much nicer looking 2019-10-11T01:52:20 < karlp> had to make sure I seated parts in the paste though, had a few blow around a bit. 2019-10-11T01:52:31 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-11T01:53:47 < karlp> https://nc.beeroclock.net/s/m5Xnttt4ei5LEeF is the simple board. 2019-10-11T01:54:04 < karlp> no pictures of the other one, (simple too, just not _as_ simple) 2019-10-11T01:56:30 < zyp> cute 2019-10-11T01:56:57 < zyp> I've considered making a base board with a usb hub and a bunch of pcie slots with usb routed to each 2019-10-11T01:57:08 < karlp> https://nc.beeroclock.net/s/XfBj9Y894qTf3dy was the setup 2019-10-11T01:57:39 < karlp> that's a dual purpose board too, decided I could bolt extra functionality on. 2019-10-11T01:57:50 < karlp> only one "omission" on that one so far. 2019-10-11T01:58:00 < karlp> the other half of it has a few more niggles, but nothing showstopper so far. 2019-10-11T01:59:52 < zyp> I figure a base board with some cortex-a module and a bunch of plug in cards that hooks up to usb would make for a convenient iot platform 2019-10-11T02:01:04 < zyp> I mean, for a lot of shit you can grab a rpi and a hat for whatever IO you wanna do, but as soon as you want to do more than one kind, it gets more complicated 2019-10-11T02:01:16 < zyp> since most don't stack well 2019-10-11T02:01:36 < karlp> yeah, this is the preliminary card with an stm32, to do the basic bringup 2019-10-11T02:01:49 < aandrew> pciexkthxbye lol 2019-10-11T02:02:01 < karlp> plan (so far) is an allwinner h3 system there instead. 2019-10-11T02:02:15 < zyp> oh, I didn't realize your stuff was also designed to be modular 2019-10-11T02:02:25 < aandrew> that setup link is password protected 2019-10-11T02:02:25 < karlp> "base" for me is the mains metering and mains offline supplies 2019-10-11T02:02:30 < zyp> thought it was just to join a horizontal and a vertical board 2019-10-11T02:02:49 < karlp> well, yeah, but designed to be useful for using different cards for testing and so on. 2019-10-11T02:03:00 < karlp> I mean, it's not going to be "build your own" shit, but... 2019-10-11T02:03:07 < karlp> aandrew: hrm, one tick 2019-10-11T02:03:10 < zyp> kinda like my EV charger shit then 2019-10-11T02:03:20 < zyp> except I'm using flat flex 2019-10-11T02:03:48 < karlp> https://nc.beeroclock.net/s/XfBj9Y894qTf3dy fixed 2019-10-11T02:03:54 < karlp> sorry, must have fat fclicked. 2019-10-11T02:04:15 < zyp> karlp, does the height match exactly? 2019-10-11T02:04:23 < karlp> which height? 2019-10-11T02:04:29 < zyp> pcb and cardboard 2019-10-11T02:04:32 < karlp> oh, the pcb in the hole 2019-10-11T02:04:37 < karlp> umm, seems so? 2019-10-11T02:04:47 < karlp> they recommended using the cardboard it came on to do this. 2019-10-11T02:04:58 < aandrew> karlp: just a piece of cardboard cut out to give the uniform flatness across the stencil? 2019-10-11T02:04:58 < karlp> but... cutting thick solid card leaves edges and shit and burrs so... 2019-10-11T02:05:01 < zyp> when I did stencilling at home, I used scrap pcbs around the target 2019-10-11T02:05:17 < aandrew> zyp: me too, but I stencil very little. my paste game is weak 2019-10-11T02:05:27 < karlp> yeah, these pcbs all have shitty mousebites from the panelling, so not nice laser sharp edges 2019-10-11T02:05:29 < zyp> same, I haven't reflowed at home for years 2019-10-11T02:05:32 < aandrew> I wouldn't have used paste on that board at all personally, just hand solder 2019-10-11T02:05:48 < zyp> eh 2019-10-11T02:05:55 < zyp> those usb connectors are a bitch to hand solder 2019-10-11T02:05:59 < aandrew> nah 2019-10-11T02:06:05 < karlp> aandrew: well, I've been syringing paste and ovening before, and it's been... eh, better than trying to tin and shit, so this was first go 2019-10-11T02:06:21 < karlp> it was fine really, much nicer end product than syringing paste 2019-10-11T02:06:29 < aandrew> yeah I'm *really* good at tinning and hand placing with a fine tip iron but I know that's a losing game 2019-10-11T02:06:32 < karlp> but I still had to cleanup the qfp 0.5mm 2019-10-11T02:07:12 < karlp> I think next time I try this I'll use _wayyyyyy_ more paste in a line across the top, and try and get a nice single pass 2019-10-11T02:07:19 < karlp> instead of little bits and re-applying 2019-10-11T02:07:23 < zyp> I've gotten good at paying people to assemble shit for me so I don't have to 2019-10-11T02:07:36 < aandrew> for those USB connectors (we all use the dongs-approved one now I think) I tin those two ground pads near the edge, place the part, flux+touch the 5 pins, reflow with a bit of hot air and then finally do the four pins 2019-10-11T02:07:46 < karlp> yeah, I actually quoted it, and it was.... a bit too much. 2019-10-11T02:07:53 < karlp> or, at least, I felt it was. 2019-10-11T02:07:59 < karlp> maybe it wasn't really, work hours wise :) 2019-10-11T02:08:22 < karlp> aandrew: are you not up to date?! dongs approved is dead. there's a new dongs approved 2019-10-11T02:08:55 < zyp> karlp, how much did you get quoted? 2019-10-11T02:09:24 < karlp> eh, I don't remember a number solidly enough to repeat it now. 2019-10-11T02:09:40 < karlp> it was 100-150 or so 2019-10-11T02:09:54 < karlp> so yeah, don't do the icelandic wage comparisons or anything 2019-10-11T02:10:16 < karlp> I'll tell myself it's still valuable to know what I did wrong on things that were hard to assemble to make the next stuff better. 2019-10-11T02:11:56 < karlp> kakimir32: this if for you: https://www.youtube.com/watch?v=QKwK-ZhTurg 2019-10-11T02:13:18 < zyp> karlp, no, I'm just curious 2019-10-11T02:13:34 < karlp> I only quoted it on seeeeeeeeed 2019-10-11T02:13:59 < karlp> I even had to tweak the bom a bit to use parts they had readily available 2019-10-11T02:14:10 < karlp> it was reasonable, just not... quite... 2019-10-11T02:15:02 < zyp> I forgot what I paid dongs to assemble some protos, but it wasn't outrageous 2019-10-11T02:15:20 < zyp> and then I paid him to assemble another 200 boards once I got the protos to work, so… 2019-10-11T02:15:30 < karlp> yeah, 200boards for sure 2019-10-11T02:15:37 < karlp> this is just proper prototype stage 2019-10-11T02:15:46 < karlp> the amortization didn't really work out in my head. 2019-10-11T02:16:03 < zyp> true, but the protos were more like 5 2019-10-11T02:16:06 < karlp> I might not even build a second pcb of these, depending on what gets proved out correct on first go. 2019-10-11T02:30:15 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-11T02:31:03 < Laurenceb> keeek Dyson 2019-10-11T02:31:21 < Laurenceb> I know ppl who were trying to get jerbs at his car company 2019-10-11T02:32:50 < Laurenceb> I met him irl once and he was very pleasant tbh 2019-10-11T02:34:15 < Laurenceb> I got caught "stealing" Dyson prototype shit out of a skip at the back of the NEC in Birmingham 2019-10-11T02:35:37 < kakimir32> karlp: 5/5 2019-10-11T02:45:40 < englishman> he wanted to make cars? 2019-10-11T02:45:58 < englishman> didn't he shill for brexit then move his company to singapoor 2019-10-11T02:51:17 < Laurenceb> yeah 2019-10-11T02:51:49 < Laurenceb> he had two Ukrainian hookers with him when I met him 2019-10-11T02:55:40 < specing> lol 2019-10-11T02:59:16 < Laurenceb> they wanted to "get back to the hotel room" 2019-10-11T02:59:51 < Laurenceb> basically I got caught by the security guards and they rang him to ask what to do with me, so he came to find us 2019-10-11T03:02:02 < Laurenceb> they seemed to think I was a vacuum cleaner spy, but he let me take as much shit as I wanted 2019-10-11T03:05:46 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T03:11:29 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Ping timeout: 268 seconds] 2019-10-11T03:12:25 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T03:35:50 < jadew> tired, busy and hungry... what a crazy combination 2019-10-11T03:36:17 < jadew> if you ever want to get anxiety, this is how it's done 2019-10-11T03:45:21 < sync> I have a F746 and am trying to talk to some ksz8081 phy but the HAL is timeouting when it sets the SWR bit, the fuck is going on 2019-10-11T04:27:09 < aandrew> karlp: it's USBc now? 2019-10-11T04:35:08 < karlp> dongs love? yes but now. he has found both a luvverly usbc with "just the bits of usbc you need" but also ... 2019-10-11T04:35:10 < karlp> https://github.com/karlp/zypsnips/blob/master/parts.usb.txt#L64-L78 2019-10-11T04:36:57 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-11T04:43:41 < karlp> kakimir32: ok, next up: https://www.youtube.com/watch?v=G77xXmA5tII 2019-10-11T04:44:04 < karlp> (let me know if that was too far, we'll take it in steps...) 2019-10-11T04:44:38 < englishman> Laurenceb: https://www.bbc.com/news/business-50004184 2019-10-11T04:44:44 < englishman> how topical 2019-10-11T04:45:13 < karlp> fantastic 2019-10-11T04:45:17 < karlp> best electric car ever 2019-10-11T04:45:28 < karlp> but the media/enemies x hate it 2019-10-11T04:45:37 < karlp> so no. we'r ekeeping it to ourselves. 2019-10-11T04:48:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-11T04:49:36 < karlp> dongs: did you ever figure out the language? my gueses were turkik, or maybe north west african? 2019-10-11T04:49:43 < karlp> but I'm curious! 2019-10-11T04:55:02 < karlp> ok, kaks, last call: https://www.youtube.com/watch?v=ZRsHUCK-Yno 2019-10-11T04:57:47 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-11T05:04:22 < Ultrasauce> oh musicspam 2019-10-11T05:04:23 < Ultrasauce> https://tomorrowthecure.bandcamp.com/album/live-from-the-spare-room-etc 2019-10-11T05:22:29 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-rrttdwbsdtzhtuzc] has joined ##stm32 2019-10-11T05:22:35 < Tordek> hi 2019-10-11T05:23:47 < Tordek> I'm using STM32CubeIDE and I want to .incbin a file in an assembly file; how do I set the directories it should search? 2019-10-11T05:25:38 < Tordek> I added Inc under Properties > C/C++ General > Preprocessor Include Path > Assembly > CDT User Setting, but I get a File Not Found error 2019-10-11T05:26:47 < Cracki> incbin? 2019-10-11T05:26:53 < Cracki> ah, assembly 2019-10-11T05:27:02 < Cracki> stm32cubeide is eclipse + gcc 2019-10-11T05:27:49 < Cracki> so that's what you can search for on google. I think eclipse has some gui/dialog/settings for "the assembler" in particular, beside "the C compiler" 2019-10-11T05:27:56 < Cracki> project specific settings 2019-10-11T05:28:50 < Cracki> file not found eh? 2019-10-11T05:29:15 < Tordek> ah, found it, it was under "Path" instead of "Include" 2019-10-11T05:29:23 < Cracki> excellent 2019-10-11T05:29:27 < Tordek> thanks! 2019-10-11T05:29:34 < Cracki> dindu nuffin 2019-10-11T05:29:49 < Tordek> bender.avi 2019-10-11T05:47:21 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-11T06:32:12 -!- beanienana is now known as beanana 2019-10-11T06:33:25 -!- fc5dc9d4 [~quassel@p5B0816AF.dip0.t-ipconnect.de] has joined ##stm32 2019-10-11T06:36:57 -!- fc5dc9d4_ [~quassel@p5B0812C4.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-11T06:52:10 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-11T06:58:58 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T06:59:29 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-11T07:00:32 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T07:03:41 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-11T07:16:05 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 265 seconds] 2019-10-11T07:26:20 -!- qyx_ [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-11T07:26:20 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-11T07:50:40 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-11T07:51:06 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T07:52:17 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-11T07:55:15 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-11T07:55:15 -!- day__ is now known as day 2019-10-11T08:05:28 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-11T08:17:28 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-11T08:19:48 -!- jly [uid355225@gateway/web/irccloud.com/x-wjeitwerqxhqvmey] has joined ##stm32 2019-10-11T08:22:26 -!- learningc [~learningc@mti-37-145.tm.net.my] has joined ##stm32 2019-10-11T08:35:17 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-11T08:38:33 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-11T08:54:35 -!- Getty [getty@stardestroyer.de] has quit [Quit: changing servers] 2019-10-11T08:54:35 -!- Amun_Ra [~amun-ra@retro.rocks] has quit [Ping timeout: 245 seconds] 2019-10-11T08:54:49 < jly> i hear this real prick named steffan hangs around 2019-10-11T08:54:55 -!- Amun_Ra_ [~amun-ra@retro.rocks] has joined ##stm32 2019-10-11T09:04:03 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-11T09:04:07 -!- Amun_Ra_ is now known as Amun_Ra 2019-10-11T09:04:26 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-11T09:10:04 < jadew> movie recommendations? 2019-10-11T09:10:22 < jly> pump up the volume 2019-10-11T09:10:26 < jly> 1990 2019-10-11T09:10:50 < jly> bit too 'teen angst' for anytime now 2019-10-11T09:11:52 < jadew> I probably saw it 2019-10-11T09:12:19 -!- dobson` [~dobson@static.38.6.217.95.clients.your-server.de] has joined ##stm32 2019-10-11T09:12:26 < jly> somehow i did too 2019-10-11T09:12:47 < jadew> the images on imdb seem familiar 2019-10-11T09:13:20 -!- dobson [~dobson@static.38.6.217.95.clients.your-server.de] has quit [Ping timeout: 276 seconds] 2019-10-11T09:13:22 < jly> the soundtrack is cool 2019-10-11T09:14:02 < jly> my autism has metastasized so I got the soundtrack on LP for a giggle 2019-10-11T09:15:13 < jly> that's when cancer gets really_fucking_serious 2019-10-11T09:19:10 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Read error: Connection reset by peer] 2019-10-11T09:21:38 < jadew> I'm not sure what happened... they're either not making good movies anymore, or they're still making them, but in the past there were more good movies that were already made 2019-10-11T09:21:52 < jly> yeah who knows 2019-10-11T09:22:10 < jadew> or... my tastes changed so much that I can't enjoy them anymore 2019-10-11T09:22:36 < jly> i completely switched off, other than curing boredom with the occasional movie at the cinema 2019-10-11T09:23:06 < jly> as for not at the cinema.... sexy beast was lol 2019-10-11T09:23:12 < jadew> that's not a bad idea, maybe I'll go to the cinema later today 2019-10-11T09:23:19 < jly> it's worth a go 2019-10-11T09:23:26 < jly> better experience 2019-10-11T09:23:50 < jly> don't have to download, and i'm sure we can all afford $10 2019-10-11T09:24:35 < jadew> and they're in 3d 2019-10-11T09:24:46 < jly> if needed 2019-10-11T09:24:55 < jly> haven't seen that mentioned in 3 or 4 years 2019-10-11T09:24:59 < jly> never participated either 2019-10-11T09:25:12 < jly> as i've never owned a surround sound either 2019-10-11T09:25:16 < jly> same shit 2019-10-11T09:25:28 < jadew> you've never seen a movie in 3d? 2019-10-11T09:25:37 < jly> kind of..... 2019-10-11T09:25:41 < jly> didn't really thrill me 2019-10-11T09:25:44 < jly> i have IRL 2019-10-11T09:25:47 < jly> don't need to pay 2019-10-11T09:26:30 < jadew> action movies are great in 3d 2019-10-11T09:26:45 < jly> it does sound like a good idea for 3d 2019-10-11T09:27:11 < jly> just honestly don't care now 2019-10-11T09:27:32 < jly> not to take it away from those interested 2019-10-11T09:27:48 < jadew> ever tried VR? 2019-10-11T09:28:15 < jly> yeah in 1995 2019-10-11T09:28:20 < jadew> heh 2019-10-11T09:28:22 < jly> fx-1 or something 2019-10-11T09:29:12 < jly> it was some extremely autistic dos era thing 2019-10-11T09:29:26 < jly> and yes i tried a thing on your android phone whatever 2019-10-11T09:30:01 < jadew> with an android cardboard kind of thing? 2019-10-11T09:30:29 < jadew> *google cardboard 2019-10-11T09:30:32 < jly> yep 2019-10-11T09:30:36 < jly> a chunk of cardboard 2019-10-11T09:30:41 < jly> you see 2019-10-11T09:30:48 < jly> brenton tarrant wasn't happy with it either 2019-10-11T09:30:51 < jly> so he decided to go IRL 2019-10-11T09:31:16 < jly> killed 50 musos.... then SHIT i forgot not VR 2019-10-11T09:31:38 < jadew> it's mostly used for porn 2019-10-11T09:32:01 < jadew> like... 99.99% of the use 2019-10-11T09:32:08 < jly> yeah 2019-10-11T09:32:22 < jly> i kinda feel sorry for those elliot rodgers out there.... 2019-10-11T09:32:41 < jly> dad and mum 2019-10-11T09:32:45 < jly> and you Haohmaru ? 2019-10-11T09:33:13 < jly> well, stm32? 2019-10-11T09:33:26 < jly> tell me about what you need the awesome power for? 2019-10-11T09:33:34 < jly> i've regressed to stm8 2019-10-11T09:34:00 < jly> it's better than people abusing my for using microchip 2019-10-11T09:34:04 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-11T09:34:04 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-11T09:34:04 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-11T09:34:05 < jly> *me 2019-10-11T09:35:22 < jly> how would they fuck things up? 2019-10-11T09:35:44 < jly> are we talking microchip or microsoft? 2019-10-11T09:36:01 < jly> hhaha i love it 2019-10-11T09:36:20 < jly> an extended middle finger logo too 2019-10-11T09:36:35 < jly> they're all just cogs 2019-10-11T09:36:53 < jly> pick (NO PUN INTENDED) whatever fits...................... 2019-10-11T09:36:54 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-11T09:37:05 < jadew> last time I checked they've made a mess from the documentation 2019-10-11T09:37:10 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-11T09:37:47 < jly> mplab was horrific tbh 2019-10-11T09:37:58 < jly> i was going to use a pic16 2019-10-11T09:38:07 < jly> then jumped 2019-10-11T09:38:08 < jadew> I was upset when I heard about the acquisition 2019-10-11T09:39:03 < jly> yeah i was happy with the options being open 2019-10-11T09:39:12 < jadew> everything about atmel was top notch 2019-10-11T09:40:05 < jadew> and they got bought by a company that exudes incompetence and "good enough to be passable" attitude from everything they ever made 2019-10-11T09:40:19 < jly> makes stm32 all the more.... attractive 2019-10-11T09:40:26 < jadew> indeed 2019-10-11T09:40:47 < jly> i see people using it in a LOT of places where it absolutely wasn't necessary 2019-10-11T09:41:02 < jadew> stm32? 2019-10-11T09:42:04 < jly> yeah 2019-10-11T09:42:15 < jly> any 8bit whatever no probs 2019-10-11T09:42:21 < jly> time to market or something idk 2019-10-11T09:42:36 < jly> no longer about bom 2019-10-11T09:43:04 < jadew> well, 8bit stuff is easier to get going tho 2019-10-11T09:43:15 < jadew> there's less complexity 2019-10-11T09:43:40 < jly> turnkey bob has stm32 solution 2019-10-11T09:43:44 < jly> whatever 2019-10-11T09:43:46 < jadew> HorizonBreak, probably because it's used in the traduino 2019-10-11T09:44:58 < jadew> it's well made and well documented 2019-10-11T09:46:59 < jly> well 2019-10-11T09:47:13 < jly> i can tell you the really fucking easy way 2019-10-11T09:47:17 < jly> sigmadsp ADI 2019-10-11T09:47:35 < jly> point and click audio dsp 2019-10-11T09:47:43 < jly> no R&D reqd 2019-10-11T09:47:59 < jly> no need to wait it's here 2019-10-11T09:48:10 < jly> same here 2019-10-11T09:48:29 < jly> good with dynamics processing? 2019-10-11T09:48:57 < jly> iir biquads are childs play 2019-10-11T09:49:21 < jly> well yea...... ... 2019-10-11T09:50:10 < jly> yeah well they all do! 2019-10-11T09:50:41 < jly> with your codec DNR? 2019-10-11T09:50:44 < jly> any idea? 2019-10-11T09:51:02 < jly> dynamic range 2019-10-11T09:51:14 < jly> it's like SNR without squelch hacks thrown in 2019-10-11T09:52:40 < jly> perhaps https://www.akm.com/content/dam/documents/products/audio/audio-codec/ak4621ef/ak4621ef-en-datasheet.pdf 2019-10-11T09:53:11 < jly> couldn't tell you, from anything audio it's up to the DSP 2019-10-11T09:53:35 < jly> and yes for sharc i have some 3rd party on trial 2019-10-11T09:54:08 < jly> these guys have been at DSP for 20+ years 2019-10-11T09:54:15 < jly> it's not RF 2019-10-11T09:54:17 < jly> afaik 2019-10-11T09:55:03 < jly> 21261 is a common audio sharc 2019-10-11T09:55:07 < Amun_Ra> Haohmaru: iirc all M4s have 32-bit fpu 2019-10-11T10:10:20 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-11T10:15:18 -!- tprrt [~tprrt@217.114.204.178] has quit [Ping timeout: 245 seconds] 2019-10-11T10:16:11 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-11T11:06:39 < Ecco> I'd like to have an interruption called every milisecond 2019-10-11T11:06:47 < Ecco> So far I'm using systick and it works really well 2019-10-11T11:06:55 < Ecco> now I'd like to change CPU frequency 2019-10-11T11:07:33 < Ecco> I assume I'll need to update setRELOAD accordingly 2019-10-11T11:07:40 < Ecco> Is there any gotcha to be aware of when doing so? 2019-10-11T11:14:36 < PaulFertser> Ecco: systick can be fed from different clock sources 2019-10-11T11:15:04 < Ecco> PaulFertser: You sure? 2019-10-11T11:15:06 < PaulFertser> So I'd check the clock distribution diagram and your systick configuration code with that regard. 2019-10-11T11:15:07 < Ecco> ARM doc says 2019-10-11T11:15:12 < Ecco> "The SysTick counter runs on the processor clock." 2019-10-11T11:15:28 < Ecco> Actually I've found something odd 2019-10-11T11:15:39 < Ecco> when changing frequency, I'd like to adjust two things : 2019-10-11T11:15:58 < Ecco> 1 - The number of cycles for a tick, so that it fires on a 1ms period 2019-10-11T11:16:14 < Ecco> 2 - The number of remaining cycles to the next tick, so the current tick period remains 1ms 2019-10-11T11:16:31 < PaulFertser> Ecco: check STK_CSR register description 2019-10-11T11:16:31 < Ecco> I'm having issues with #2, apparently you cannot do this 2019-10-11T11:17:02 < Ecco> PaulFertser: Oh you're right 2019-10-11T11:17:08 < Ecco> SYST_CSR.CLKSOURCE 2019-10-11T11:17:19 < PaulFertser> And btw, libopencm3 has systick_set_frequency() that calculates the appropriate reload value and optionally enables AHB/8 prescaler automatically. 2019-10-11T11:18:17 < PaulFertser> Ecco: I can't understand your #2, as systick is periodic. 2019-10-11T11:18:50 < Ecco> I want systick to fire exactly every 1ms 2019-10-11T11:19:01 < Ecco> at some point I change the CPU frequency, let's say I divide it by 2 2019-10-11T11:19:17 < Ecco> systick works by decrementing a counter every CPU cycle 2019-10-11T11:19:31 < Ecco> and when counter reaches 0 it's reset to RELOAD 2019-10-11T11:19:33 < PaulFertser> Ecco: it's either AHB frequency or AHB/8 (if not on m0) for you to choose from. 2019-10-11T11:19:42 < Ecco> and the exception is run 2019-10-11T11:19:46 < Ecco> if I change the CPU freq 2019-10-11T11:19:51 < Ecco> I need to ajust the RELOAD value 2019-10-11T11:19:56 < Ecco> *and* the value of the counter 2019-10-11T11:20:15 < Ecco> otherwise one systick period will be too long (or too short) 2019-10-11T11:20:33 < Ecco> but apparently you can only set the current value of the counter to zero 2019-10-11T11:20:52 < PaulFertser> I see now 2019-10-11T11:20:56 < Ecco> :) 2019-10-11T11:21:10 < Ecco> and libopencm3 doesn't do that 2019-10-11T11:21:23 < Ecco> but that's kinda ok because the function is called systick_set_frequency 2019-10-11T11:21:55 < Ecco> the problem arises if you dynamically change AHB freq 2019-10-11T11:22:09 < Ecco> then in an ideal world you'd like to take care of systick 2019-10-11T11:22:15 < PaulFertser> Right, you might need to do it in two steps then... 2019-10-11T11:24:35 < PaulFertser> Or use another timer that doesn't depend on AHB if one is available on your hardware... 2019-10-11T11:24:54 < Ecco> indeed 2019-10-11T11:25:36 < Ecco> Well systick seemed easy to set up, but actually all I want to do is count ellapsed miliseconds 2019-10-11T11:25:56 < Ecco> all the interruption does is "globalMiliSecondsCounter++" 2019-10-11T11:27:04 < Ecco> is systick the right timer to do this? 2019-10-11T11:37:19 -!- sterna [~Adium@cvis.oal.lindholmen.se] has joined ##stm32 2019-10-11T11:38:02 -!- learningc [~learningc@mti-37-145.tm.net.my] has quit [Remote host closed the connection] 2019-10-11T11:51:10 < Tordek> in order to place some values into ROM in assembly, do I use .section rodata? or is there a different name? 2019-10-11T11:52:35 < Tordek> and if so, how do I access that data? I tried reading through a pointer but it points to 0x00000000 2019-10-11T11:59:03 < PaulFertser> Tordek: why can't you declare const global variables? 2019-10-11T11:59:21 < Amun_Ra> I'm not much of an arm asm user, but I'd use .section .rodata and access the data in the section via a label 2019-10-11T12:00:27 < Tordek> PaulFertser: I'm including some binary blobs with .incbin 2019-10-11T12:00:54 < PaulFertser> Tordek: but some externally-visible symbol is assigned to it somehow, right? 2019-10-11T12:01:06 < PaulFertser> Tordek: what's .incbin, where is it explained? 2019-10-11T12:01:19 < Tordek> in the gcc assembler manual 2019-10-11T12:01:30 < Tordek> Amun_Ra: ah, it seems it was silly; .rodata not rodata 2019-10-11T12:01:50 < Amun_Ra> :) 2019-10-11T12:01:50 < Tordek> thanks! 2019-10-11T12:02:38 < PaulFertser> Tordek: so you just add a label in S file, then .incbin and it gets there in section? Nice. 2019-10-11T12:03:03 < Tordek> PaulFertser: I have a .h that tells my C where those values are located, but they were misplaced because of the name :P 2019-10-11T12:03:12 < Tordek> and, yeah, that's exactly it 2019-10-11T12:03:37 < Tordek> just a bunch of "foo: .incbin "foo.bin"" 2019-10-11T12:04:12 < Tordek> I've got a 6502 emulator and I needed a way to load the ROM without the hassle of escaping everything into a C string 2019-10-11T12:04:45 < Tordek> (sorry, you also need "global foo") 2019-10-11T12:04:58 < PaulFertser> Tordek: when you say "where" it sounds like you're hardcoding the address, but you should be able to count on the linker to give you any address? 2019-10-11T12:05:18 < PaulFertser> And then just extern uint8_t *foo; in the header. 2019-10-11T12:05:25 < Tordek> aye, that's what I'm doing 2019-10-11T12:05:44 < Tordek> but I point it to .section .rodata so it's stored in ROM instead of taking up RAM 2019-10-11T12:35:28 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 264 seconds] 2019-10-11T12:36:58 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has joined ##stm32 2019-10-11T12:38:13 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-11T12:38:15 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-11T12:38:15 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-11T12:50:16 < Ecco> I'm reading a PCB footprint, and I don't understand some notations 2019-10-11T12:50:59 < Ecco> https://i.imgur.com/MI5JFmp.png 2019-10-11T12:51:59 < jly> ty 2019-10-11T12:53:13 < Ecco> Apprently the "2-" should just be ignored 2019-10-11T12:53:14 < jpa-> Ecco: "2 holes of 0.50 wide, +- 0.05" 2019-10-11T12:53:20 < Ecco> ha 2019-10-11T12:53:29 < Ecco> indeed!! 2019-10-11T12:53:30 < Ecco> thanks jpa- 2019-10-11T12:54:14 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-11T12:55:08 < jpa-> i think the standard notation would be "2x 0.50" which would be kind of clearer 2019-10-11T12:55:15 < jly> yea 2019-10-11T12:55:17 < jly> hi jpa 2019-10-11T12:55:27 < jpa-> hello 2019-10-11T12:55:56 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T12:59:31 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-11T13:02:09 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-11T13:02:54 < karlp> jadew: I donno about your microchip attitude man, I mean, pics are their own world, but the rest of their parts are great. super cheap, good specs. 2019-10-11T13:04:23 < karlp> hah, Haohmaru 32bit float isn't enough for your HIGH FIDELLLITY? you want doubles? 2019-10-11T13:04:35 < jpa-> even for pics, microchip is great in that they keep ooooold parts available for decades 2019-10-11T13:06:55 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T13:09:57 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 240 seconds] 2019-10-11T13:12:41 < jly> IIR biquads make me happy 2019-10-11T13:13:15 < jly> https://wikimedia.org/api/rest_v1/media/math/render/svg/0998a31bbbc6c35128556e666e308cdc1ef027b0 2019-10-11T13:14:52 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-11T13:18:42 < jly> Haohmaru: do you think the new AKM 768kHz ADC will have better tone/ 2019-10-11T13:28:20 < zyp> are you a dog? 2019-10-11T13:28:48 -!- sterna [~Adium@cvis.oal.lindholmen.se] has quit [Quit: Leaving.] 2019-10-11T13:38:12 < qyx_> a dong 2019-10-11T13:38:14 -!- qyx_ is now known as qyx 2019-10-11T13:51:03 < jly> well just the law of diminishing returns coming into play 2019-10-11T13:55:21 < karlp> sounds about right for audiophool :) 2019-10-11T13:57:54 < jly> welkon 2019-10-11T13:58:03 < jly> 48khz suck it off 2019-10-11T13:59:05 < jly> whats your fucking bit clock mate 2019-10-11T14:00:15 < jly> tell me about your oversampling ratio 2019-10-11T14:01:26 < jly> don't give me your 1 bit shit 2019-10-11T14:01:34 < jly> ah fuck off 2019-10-11T14:01:37 < jly> you're drunk 2019-10-11T14:02:11 < jly> so 2019-10-11T14:02:15 < jly> so's my dad 2019-10-11T14:02:35 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 250 seconds] 2019-10-11T14:02:48 < jly> that's great.... 2019-10-11T14:03:30 < jly> other than curing my perpetual boredom 2019-10-11T14:03:33 < jly> where was this going? 2019-10-11T14:03:45 < jly> steffan stopped sending me pics you see 2019-10-11T14:04:14 < jly> there were dogs and cats 2019-10-11T14:04:20 < jly> it was all quite jolly 2019-10-11T14:04:33 < jly> jly you might say. 2019-10-11T14:04:53 < jly> are you abusing your animals or something?? 2019-10-11T14:05:40 < jly> so i've heard 2019-10-11T14:06:36 < jly> wht 2019-10-11T14:06:39 < jly> why g othere 2019-10-11T14:06:49 < jly> you need to find tactile switches 2019-10-11T14:06:57 < jly> ffs 2019-10-11T14:07:12 < jly> surely theres something that can be shipped in a day from your local 2019-10-11T14:07:37 < jly> i can't remember the japanese brand i used to buyt 2019-10-11T14:07:43 < jly> omron? 2019-10-11T14:07:54 < jly> pioneer dj trash in the masses 2019-10-11T14:08:45 < jly> yeah i never bothered 2019-10-11T14:08:54 < jly> that used to be the dread of living for me 2019-10-11T14:09:07 < jly> all types of midi / lighting control / dj / turd / etc 2019-10-11T14:09:21 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-11T14:09:30 < jly> m 2019-10-11T14:09:51 < jly> and then each had 10 different sizes of actuator 2019-10-11T14:10:12 < jly> i can't remember the one lighting controller that used cherrys 2019-10-11T14:10:14 < jly> chamsys 2019-10-11T14:10:18 < jly> that's the one 2019-10-11T14:10:24 -!- grindhold [~quassel@mail.skarphed.org] has quit [Read error: Connection reset by peer] 2019-10-11T14:10:29 < jly> the 'non tactile' ones 2019-10-11T14:10:33 < jly> blues i think 2019-10-11T14:10:38 < jly> yes 2019-10-11T14:10:40 < jly> o 2019-10-11T14:10:41 < jly> that 2019-10-11T14:10:51 < jly> no tactile notch 2019-10-11T14:10:52 < jly> ye 2019-10-11T14:10:59 < jly> i have brown on my kb 2019-10-11T14:11:02 < jly> tact but no noise 2019-10-11T14:11:03 < jly> yes? 2019-10-11T14:11:14 < jly> just a da 2019-10-11T14:11:16 < jly> das 2019-10-11T14:11:23 < jly> yea 2019-10-11T14:11:27 < jly> tactile no noise 2019-10-11T14:12:11 < jly> they're good too 2019-10-11T14:12:24 < jly> it's just when they get too cheap 2019-10-11T14:12:40 < jly> honestely 2019-10-11T14:12:55 < jly> some IBM one were really nice 2019-10-11T14:13:07 < jly> i still use a few 2019-10-11T14:13:16 < jly> and i'm not talking model M 2019-10-11T14:14:26 < jly> most actually 2019-10-11T14:14:50 < jly> that's where the problem was realised 2019-10-11T14:15:44 < jly> i never had any issue personally 2019-10-11T14:16:00 < jly> after realizing something was crap 2019-10-11T14:16:14 < jly> pay $20 for a kb 2019-10-11T14:16:16 < jly> or $200 2019-10-11T14:16:28 < jly> the price doesn't shake me either way 2019-10-11T14:16:52 < jly> i've paid more for a good bump of pure charie tbh 2019-10-11T14:17:00 < jly> once a year.... 2019-10-11T14:17:21 < jly> ebay? 2019-10-11T14:17:26 < jly> does it talk to you? 2019-10-11T14:18:36 < jly> just like me and gibson les paul guitars 2019-10-11T14:18:57 < jly> the 60's scale neck is completely different to the 50's scale neck 2019-10-11T14:19:06 < jly> one was made for my hand 2019-10-11T14:19:15 < jly> other one not even close 2019-10-11T14:19:29 < jly> a guitar neck 2019-10-11T14:19:39 < jly> thing where your left hand goes 2019-10-11T14:19:53 < jly> just for something to compare 2019-10-11T14:20:17 < jly> your 'resolution' goes down into the hundreds of nanometers quite easily 2019-10-11T14:22:26 < jly> https://www.ebay.com.au/itm/1995-GIBSON-USA-LES-PAUL-EBONY-CUSTOM-BLACK-BEAUTY-w-HSC-JAPAN-YAMANO-EXPORT/202794858096 2019-10-11T14:22:33 < jly> i got mine for $2500aud 2019-10-11T14:22:51 < jly> 1994 is my fave, the centennial year 2019-10-11T14:22:57 * karlp facepalms. reads the note, "can be left floating" and... left it floating. ignored the "can be left floating because it has an internal pulldown" 2019-10-11T14:23:02 < karlp> moronic move. 2019-10-11T14:23:07 < Amun_Ra> I have mine dblneck chibson, plays fine 2019-10-11T14:23:22 < jly> fuck that's a beautiful piece of ebony 2019-10-11T14:23:54 < jly> Haohmaru: i've seen people pay much more for shit that won't appreciate in value 2019-10-11T14:24:15 < jly> if i don't pick this up i'll forward it onto the other guy 2019-10-11T14:24:24 < jly> not really 2019-10-11T14:24:43 < jly> i'm among people in the club of $1,000,000 cars 2019-10-11T14:24:51 < jly> i'm in the baby's game 2019-10-11T14:26:22 < jly> cars mean nothing to me 2019-10-11T14:26:28 < jly> accelerator brake 2019-10-11T14:26:35 < jly> manual/auto doesn't matter 2019-10-11T14:26:43 < jly> glad i learned on a stick 2019-10-11T14:27:15 < jly> learning? 2019-10-11T14:27:21 -!- grindhold_ is now known as grindhold 2019-10-11T14:27:25 < jly> should say loving 2019-10-11T14:27:30 < jly> wekk 2019-10-11T14:27:31 < jly> well 2019-10-11T14:27:32 < jly> yeah 2019-10-11T14:27:50 < jly> it's amazing how easy it is to fry the tyres 2019-10-11T14:27:57 < jly> even on an itty bitty 4cyl 2019-10-11T14:28:00 < jly> HAHAHHAHAHA 2019-10-11T14:28:19 < jly> feed it too much, relaease the clutch 2019-10-11T14:28:20 < jly> HHAHAHAH 2019-10-11T14:28:37 < jly> i couldn't do that on a 6cyl 4.0 six 2019-10-11T14:28:59 < jly> oh man 2019-10-11T14:29:18 < jly> that was on my 1st lesson 2019-10-11T14:29:28 < jly> i had to pull over and laugh for 10 mins 2019-10-11T14:30:24 < jly> but yeah honestly without the torque converter in the chain a totally different feel 2019-10-11T14:30:27 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T14:30:28 < jly> so touchy 2019-10-11T14:31:00 < jly> didn't even know there was a fluid coupling in the drivechain 2019-10-11T14:32:37 < PaulFertser> I've seen somebody playing a "car mechanic" game. Totally unrealistic. 2019-10-11T14:33:59 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 265 seconds] 2019-10-11T14:34:06 * karlp loved being a "mechanic" in https://www.retrogames.cz/play_1145-DOS.php?language=EN 2019-10-11T14:34:36 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T14:37:50 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 268 seconds] 2019-10-11T14:38:36 < PaulFertser> Haohmaru: hm do you know if you watched some youtube videos explaining how the clutch works internally it would help you to drive? 2019-10-11T14:40:46 < PaulFertser> Haohmaru: I think the only important thing regarding the clutch handling is that you depress it slowly and then when you feel the car starts moving you are NOT letting it go immediately and to the contrary you either stop depressing it altogether or continue doing much slower. 2019-10-11T14:42:04 < PaulFertser> (with old Ford model T you could just do it as fast as you wanted apparently, it was more "user-friendly") 2019-10-11T14:43:18 < PaulFertser> Haohmaru: I meant not about handling but about internals. My idea is that for a smart person knowing internals should help but I might be wrong. 2019-10-11T14:47:03 < PaulFertser> I once had a funny clutch malfunction which required pressing the pedal pretty fast, whenever I did it a bit slower, the clutch didn't disconnect. 2019-10-11T14:50:32 < PaulFertser> In fact, that was one of the most pleasant repair experiences in my life. It was summer, no raining, I just bought a new clutch master cylinder (it was readily available and really cheap too), replaced it in like 30 minutes without any issues, bled the system, done. Was really a breeze. 2019-10-11T14:50:54 < karlp> feck, can only find usb-c and usb micro cables. I've been excessive with the mini-b purge 2019-10-11T14:51:09 < zyp> common problem 2019-10-11T14:51:30 < zyp> I mean, I haven't even been throwing away mini-b-cables 2019-10-11T14:51:34 < zyp> they have just disappeared 2019-10-11T14:52:31 < zyp> then again, I feel like I've got both a bunch of micro-b and a bunch of usb-c cables that have disappeared too 2019-10-11T14:52:52 < zyp> I need to go through old boxes some time and organize stuff 2019-10-11T14:53:28 < zyp> all the stuff I got in boxes now is mostly organized by when it was put in a box, not what it is 2019-10-11T15:01:55 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-11T15:02:35 -!- kakimir32 is now known as kakimir64 2019-10-11T15:18:55 < Thorn> spacewalk in progress https://www.youtube.com/watch?v=NCOGg8sIngk 2019-10-11T15:42:43 -!- sterna [~Adium@cvis.oal.lindholmen.se] has joined ##stm32 2019-10-11T15:42:54 < karlp> those are the boxes you can just throw away next time you move :) 2019-10-11T15:48:45 < karlp> heh, renesas has moved to cortex-m: https://www.eenewsembedded.com/news/ra-32-bit-arm-cortex-m-mcus-provide-superior-performance?news_id=122144 2019-10-11T15:49:39 < jpa-> zyp: re: tagconnect, it fits fine and between the cut legs, i'm-too-cheap alternative to TC2030-CLIP and the adapter to discovery board, it's suitably ghetto for my workbench 2019-10-11T15:49:42 < jpa-> https://jpa.kapsi.fi/stuff/pix/IMG_20191011_153552.jpg 2019-10-11T15:54:53 < karlp> well, renesas was one of the "others" until now. 2019-10-11T15:57:04 < englishman> should have gone to riscV now they are 15 years behind 2019-10-11T16:05:01 < kakimir64> maybe I should update my projects freertos version 2019-10-11T16:09:02 < kakimir64> as it's not even working 2019-10-11T16:09:33 < kakimir64> currently 2019-10-11T16:15:36 -!- c10ud_ [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-11T16:16:45 < zyp> jpa-, haha, nice 2019-10-11T16:17:04 < zyp> I lost the first clip I bought, so I picked up another three and I haven't lost one since :p 2019-10-11T16:17:30 < zyp> assuming I can still find the two that I don't keep attached to the cable when not in use :p 2019-10-11T16:20:45 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-11T16:35:28 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-11T16:40:01 < englishman> what's a popular site for chinese junk these days, banggood, gearbest etc? 2019-10-11T16:46:18 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-11T16:47:00 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-11T16:50:23 < Adluc> hello guys and trolls , any ideas how to make non-retard and aestetically pleasing break-tabs (between the mikroBus parts) http://91.148.0.109/junk/pcbshit.png ? 2019-10-11T16:51:03 < Adluc> two teeth on right side are meant to be holders for the board in enclosure, as well as when you break it, moar teeth come up 2019-10-11T16:51:55 < Adluc> should any of you have ideas/opinions, let me know, the slot for card is M2 (NGFF) E-slot with M2.COM pinout 2019-10-11T16:52:10 < PaulFertser> englishman: I had reasonably good experience with banggood. 2019-10-11T16:52:40 < englishman> same 2019-10-11T16:53:57 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-11T17:25:51 -!- onio [~onio@2a00:23c5:7a01:8600:a161:f997:855f:a492] has joined ##stm32 2019-10-11T17:33:47 < kakimir64> Adluc: cool.. what it's for? 2019-10-11T17:37:14 < karlp> Ultrasauce: this live from the spare room is great htanks 2019-10-11T17:38:46 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-11T17:39:27 < Adluc> kakimir64: diploma thesis, includes M2 boards with STM32L4+lora, and also for a company for current probes for remote AC measuring, as well as other bullshits 2019-10-11T17:39:59 < Adluc> and I quite like this way where you can swap the module containing MCU+modem 2019-10-11T17:40:46 < Adluc> on bottom is supply LTC3129-1 (2-15V input -> 3V3), including auto shutdown when module powers 3.3v rail 2019-10-11T17:43:07 < qyx> oh LT smps vreg, much pro 2019-10-11T17:43:12 < qyx> also $$ 2019-10-11T17:43:38 < Adluc> qyx: quite difficult to find similar buck-boost with such range 2019-10-11T17:43:52 < karlp> why do you need theb oost portion? 2019-10-11T17:43:52 < Adluc> price as end-product justifies the price :) 2019-10-11T17:44:11 < Adluc> want to run it from lifepo4 single cell 2019-10-11T17:44:21 < karlp> so what on earth do you need 15V for then? 2019-10-11T17:44:32 < Adluc> so it can from 12V adapter 2019-10-11T17:44:37 < Adluc> without change 2019-10-11T17:46:58 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-11T17:47:06 < Adluc> lets assume +/- 5$ doesnt matter anyway :D 2019-10-11T17:50:04 -!- learningc [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-11T17:50:25 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T17:53:09 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-11T17:54:08 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 252 seconds] 2019-10-11T17:59:27 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T18:00:49 < kakimir64> recommend coding displays 2019-10-11T18:01:00 < kakimir64> dual or extra wide 2019-10-11T18:01:51 < kakimir64> yes 2019-10-11T18:02:32 < kakimir64> it's connected to dual mount 2019-10-11T18:02:37 < kakimir64> vesa100 2019-10-11T18:03:03 < kakimir64> I mean 2019-10-11T18:03:11 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 268 seconds] 2019-10-11T18:03:16 < kakimir64> all pro monitors tilt 2019-10-11T18:03:23 < kakimir64> and rotate 2019-10-11T18:03:25 < kakimir64> and shit 2019-10-11T18:03:53 < kakimir64> also height adjustment with springloading 2019-10-11T18:04:19 < kakimir64> moves with little force both up and down 2019-10-11T18:04:42 < Adluc> http://91.148.0.109/junk/pcbshit2.png 2019-10-11T18:04:44 < kakimir64> like elevator 2019-10-11T18:05:04 < kakimir64> nice logo Adluc 2019-10-11T18:05:36 < Adluc> kakimir64: thanks ^_^ https://www.dafont.com/solaria.font should you need 2019-10-11T18:06:18 < kakimir64> yes dafont 2019-10-11T18:06:35 < kakimir64> maybe I just get size identical monitor next to this sx2262w 2019-10-11T18:06:44 < kakimir64> monitor is so close that 22" is ok 2019-10-11T18:08:09 -!- sterna [~Adium@cvis.oal.lindholmen.se] has quit [Quit: Leaving.] 2019-10-11T18:09:18 < karlp> solaria looks like the iceland airwaves font 2019-10-11T18:09:33 < kakimir64> I have another eizo 24" but it was some basic line when it was new 2019-10-11T18:09:43 < kakimir64> just garbage image quality 2019-10-11T18:12:02 < kakimir64> there is little availibility for 1920x1200 22" monitors anymore 2019-10-11T18:13:11 < mitrax> i've got 25" Dells but tilted 90° they're too tall to be looked at at close distance, the top of the screen ends up being too further away than the middle of the screen 2019-10-11T18:13:20 < mitrax> so it feels weird 2019-10-11T18:14:23 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-11T18:15:00 < kakimir64> maybe one curved monitor would do 2019-10-11T18:16:28 < mitrax> friend of mine got a 30" curved Dell, but i don't like it 2019-10-11T18:16:37 < mitrax> when you do CAD stuff things looks distored 2019-10-11T18:17:04 < mitrax> err distorded 2019-10-11T18:20:12 < kakimir64> howe does it distort? 2019-10-11T18:23:15 < mitrax> kakimir64: horizontal straight lines look curvy if they stretch across all the width of the screen, it's not really a huge issue 2019-10-11T18:24:29 < kakimir64> unless straigth at eye level 2019-10-11T18:24:38 < mitrax> kakimir64: or at least it's not obvious when you look right in the center of the screen, but from the side or if you're standing above you notice it immediately 2019-10-11T18:24:40 < mitrax> yup 2019-10-11T18:28:55 < kakimir64> shoud reeRTOScommonhooks 2019-10-11T18:29:27 < kakimir64> FreeRTOScommonhooks* be placed in project/ or project/freertos 2019-10-11T18:29:33 < kakimir64> like it would matter but 2019-10-11T18:30:18 < kakimir64> I have kaki'ed it with header in project/inc and source in project/freertos/src 2019-10-11T18:32:14 < kakimir64> wtf is happening in here.. in project dir there is no src/ 2019-10-11T18:32:26 < kakimir64> but under Debug and Release there is src/ 2019-10-11T18:32:59 < kakimir64> wait 2019-10-11T18:33:04 < kakimir64> forget 2019-10-11T18:33:07 < kakimir64> it's there 2019-10-11T18:39:40 < kakimir64> Am I correct to assume that heap_3.c and heap_useNewlib.c are both implementing pvPortMalloc() and vPortFree() and I should only have one of them? 2019-10-11T18:40:26 < kakimir64> https://github.com/ErichStyger/mcuoneclipse/blob/master/Examples/MCUXpresso/FRDM-K64F/FRDM-K64F_Zork/Generated_Code/heap_useNewlib.c 2019-10-11T18:40:55 < kakimir64> I have forgot why heap_useNewlib.c is needed instead of ones provided 2019-10-11T18:41:48 < kakimir64> maybe it was something about sbrk 2019-10-11T18:58:05 -!- ekaologik [~quassel@p5DE865AE.dip0.t-ipconnect.de] has joined ##stm32 2019-10-11T19:08:12 < kakimir64> port.c:727: undefined reference to `__SRAM_segment_start__' 2019-10-11T19:08:35 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has quit [Remote host closed the connection] 2019-10-11T19:09:00 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-11T19:09:03 -!- jly [uid355225@gateway/web/irccloud.com/x-wjeitwerqxhqvmey] has quit [Quit: Connection closed for inactivity] 2019-10-11T19:15:23 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has joined ##stm32 2019-10-11T19:17:04 < kakimir64> I wonder if __SRAM_segment_start__ is something that I should set myself to refer to some linker value 2019-10-11T19:20:27 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T19:22:23 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-11T19:23:19 < karlp> nice one. freertos provides xTaskGetTickCount but it's only 32bit, with no helpers for handling rollovers 2019-10-11T19:23:44 < kakimir64> what happens then 2019-10-11T19:23:48 < kakimir64> when it overflows 2019-10-11T19:24:21 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-11T19:24:35 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 268 seconds] 2019-10-11T19:25:40 < karlp> it's handled internally properly, 2019-10-11T19:26:09 < karlp> but the user api "uptime", which is what xTaskGetTickCount actually is, jsut rolls over. 2019-10-11T19:26:40 < kakimir64> https://www.freertos.org/FreeRTOS_Support_Forum_Archive/February_2012/freertos_Tick_count_overflow_5005076.html read whining here 2019-10-11T19:26:55 < fenugrec> you should design your system to crash/reset periodically to avoid an overflow then 2019-10-11T19:26:57 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 240 seconds] 2019-10-11T19:27:39 < kakimir64> is this how shit goes wrong? 2019-10-11T19:27:45 < karlp> fenugrec: harhar 2019-10-11T19:28:40 < kakimir64> developer thinks that yeah nobody will keep this airliner APU unit controller powered for 270days straight or so 2019-10-11T19:28:47 < kakimir64> then it goes to shit 2019-10-11T19:29:08 < kakimir64> because freertos XTask 2019-10-11T19:29:42 < karlp> well, it's only my own "uptime" reporting that rolls over. 2019-10-11T19:29:46 < fenugrec> xNumOfOverflows ? 2019-10-11T19:30:50 < kakimir64> Freemaker linker script generator.. is it fun? 2019-10-11T19:31:04 < kakimir64> it seem I need to get into templating 2019-10-11T19:31:04 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Quit: Leaving] 2019-10-11T19:34:09 < kakimir64> have you used freemaker? 2019-10-11T19:35:44 -!- fenugrec_ [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-11T19:37:06 < kakimir64> actually I could use portable files for CM3 instead of CM3_MPU couldn't I? 2019-10-11T19:37:26 < kakimir64> I don't feel like I need MPU 2019-10-11T19:38:38 < kakimir64> it's not really the purpose of this excercise 2019-10-11T19:39:55 < Thorn> kakimir64: if you're talking freertos and if you aren't using floating point then you use the CM3 port (even on m4) 2019-10-11T19:40:21 < kakimir64> oh 2019-10-11T19:40:23 < Ecco> http://www.ti.com/product/TPD2S300 2019-10-11T19:40:38 < Ecco> -> Is this of any use in the "dumb" case of a replacement of an USB-B port? 2019-10-11T19:40:40 < kakimir64> I have CM3 MPU in use Thorn 2019-10-11T19:40:48 < Ecco> i.e. CC1 and CC2 are just pulled down to GND 2019-10-11T19:40:59 < kakimir64> but what floating points have to do with it Thorn? 2019-10-11T19:41:09 < Ecco> Can an ESD destroy a pull-down? 2019-10-11T19:41:39 < kakimir64> what pull-down? 2019-10-11T19:41:53 < Thorn> the MPU port saves floating point registers on context switch 2019-10-11T19:42:03 < Thorn> iirc 2019-10-11T19:42:04 < kakimir64> resistor? 2019-10-11T19:43:07 < kakimir64> oh 2019-10-11T19:43:47 < kakimir64> I thought it was for memory protection 2019-10-11T19:44:24 < kakimir64> any other takes on this issue guise? 2019-10-11T19:46:07 < kakimir64> Using a Memory Protection Unit (MPU) can protect applications from a number of potential errors, ranging from undetected programming errors to errors introduced by system or hardware failures. FreeRTOS-MPU can be used to protect the RTOS kernel itself from invalid execution by tasks and protect data from corruption. It can also protect system 2019-10-11T19:46:07 < kakimir64> peripherals from unintended modification by tasks and guarantee the detection of task stack overflows. 2019-10-11T19:47:03 < kakimir64> it sound like it would be quite useful in development phase 2019-10-11T19:47:17 < kakimir64> https://www.freertos.org/FreeRTOS-MPU-memory-protection-unit.html 2019-10-11T19:48:47 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-11T19:49:09 < kakimir64> it doesn't say anything about floating points Thorn. Can you elaborate? 2019-10-11T19:49:24 < karlp> fenugrec: yeah, xNumOfOverflows would be fine, but it's "private" internal. patching it is boring... 2019-10-11T19:50:56 < kakimir64> it sounds liek a case where patching is required 2019-10-11T19:51:24 < kakimir64> you should see in internets how other people have done it 2019-10-11T19:57:29 -!- fenugrec_ [~fenugrec@24.105.71.66] has quit [Ping timeout: 250 seconds] 2019-10-11T19:58:54 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has quit [Ping timeout: 246 seconds] 2019-10-11T20:08:35 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-11T20:17:54 < kakimir64> why stack pointer must be double word spacing? 2019-10-11T20:18:09 < kakimir64> embedded 101 2019-10-11T20:18:43 < kakimir64> double word boundary* 2019-10-11T20:19:35 < kakimir64> not that I have had any problems related to it but I just want to know 2019-10-11T20:37:15 -!- ekaologik [~quassel@p5DE865AE.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-11T20:46:26 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-11T20:46:39 < jpa-> kakimir64: because some ARM processors have native 8-byte datatypes, so the ABI specifies 8-byte alignment 2019-10-11T20:47:17 < jpa-> cortex-m will work fine even with 4-byte alignment, though some library functions and gdb traceback might break 2019-10-11T21:03:19 < kakimir64> okay 2019-10-11T21:03:31 < kakimir64> thanks for proinfo 2019-10-11T21:03:50 < kakimir64> also what you say about ARM_CM3 or ARM_CM3_MPU? 2019-10-11T21:03:59 < kakimir64> both should work on my chip 2019-10-11T21:04:59 < kakimir64> what was that floating point thing about Thorn said? 2019-10-11T21:24:35 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has quit [Ping timeout: 250 seconds] 2019-10-11T21:55:57 < jpa-> that would be FPU support, not MPU support 2019-10-11T22:02:04 < kakimir64> indeed 2019-10-11T22:07:17 < Adluc> kakimir64: http://91.148.0.109/junk/sickshit3.png 2019-10-11T22:07:32 < Adluc> lickable 2019-10-11T22:07:42 < kakimir64> yes 2019-10-11T22:07:53 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-11T22:14:39 < catphish> sorry, advice request: i'd like to stream the time between pulses over a 1-byte serial connection, this means that if a pulse doesn't arrive for 255 cycles, i need to send 0xff to the serial line, and restart the timer from 0, then when a pulse *does* arrive, i need to send the current timer value, and also reset it 2019-10-11T22:15:12 < Cracki> uh 2019-10-11T22:15:28 < Cracki> step back, consider the range of intervals you're likely to see 2019-10-11T22:16:19 < Cracki> consider encoding the situation as an "end marker" 2019-10-11T22:17:03 < Cracki> and encode the restart (first pulse) as a beginning 2019-10-11T22:29:37 -!- c4017_ [~c4017@S010664777dab66f3.vf.shawcable.net] has quit [Ping timeout: 240 seconds] 2019-10-11T22:34:13 < kakimir64> it builds 2019-10-11T22:34:21 < kakimir64> it compiels 2019-10-11T22:35:42 < kakimir64> now priviledged_functions : 2019-10-11T22:36:21 < Cracki> old white male functions 2019-10-11T22:36:51 < kakimir64> triggered 2019-10-11T22:36:54 < kakimir64> .exe 2019-10-11T22:37:11 < specing> check your execution privilege 2019-10-11T22:37:25 < Cracki> only way to wash yourself of the original sin is to dye your hair a neon color, cut off your peepee, and iron out your wrinkles 2019-10-11T22:39:13 < roomba> don't forget to gain 100lbs and wear a bernie t-shirt 2019-10-11T22:45:02 < catphish> Cracki: some background is needed here - the serial line is not fast enough to post a 2 byte value if the period between pulses is < 255, so a variable length encoding is needed 2019-10-11T22:45:40 < Cracki> encode end of train as 255, encode first pulse as 0 2019-10-11T22:46:33 < Cracki> do you really care about the precise length of intervals longer than 255? 2019-10-11T22:46:38 < catphish> yes 2019-10-11T22:46:45 < catphish> up to any length 2019-10-11T22:46:54 < Cracki> then send 255 as "this isn't a pulse, it's a gap, add to it" 2019-10-11T22:47:09 < catphish> right, and that's exactly what i want to do 2019-10-11T22:47:12 < Cracki> so 254 is still an actual interval, but 255 is speshul 2019-10-11T22:47:18 < catphish> so if the distance between 2 pulses is 260... 2019-10-11T22:47:28 < catphish> i'd send 0xff, 0x05 2019-10-11T22:47:41 < catphish> the question is how to configure the timer to permit this 2019-10-11T22:47:53 < Cracki> for dist 255 it's FF+00 by that rule, so be careful if 00 should mean anything special 2019-10-11T22:48:09 < Cracki> the timer? you don't 2019-10-11T22:48:14 < Cracki> you run the timer in 16 or 32 bits 2019-10-11T22:48:27 < Cracki> and figure out what to send when it's fired 2019-10-11T22:48:40 < catphish> Cracki: that's what i do right not, but there's a problem... 2019-10-11T22:48:51 < Cracki> *maybe* limit it to 255 and catch the overflow interrupt 2019-10-11T22:49:11 < Cracki> that would nicely trigger so you can keep sending FF values 2019-10-11T22:49:20 < catphish> if there's no pulse for (for example) 10,000 cycles, i suddenly have to send 40 bytes at once 2019-10-11T22:49:47 < Cracki> k then, period 255/256, overflow interrupt handler 2019-10-11T22:50:06 < catphish> this was my thought process, period 255, with an overflow interrupt 2019-10-11T22:50:17 < Cracki> be sure to distinguish overflow (->0) from input capture (->0) 2019-10-11T22:50:21 < catphish> but then when the timer actually fires, that also needs to reset the timer. right 2019-10-11T22:50:37 < Cracki> timer can reset itself after capturing counter value 2019-10-11T22:50:46 < Cracki> in hw of course 2019-10-11T22:50:49 < catphish> can it do that? 2019-10-11T22:51:13 < catphish> i think perhaps it can on some channels but not others 2019-10-11T22:51:58 < catphish> anyway, what you're suggesting now is the conclusion i reached when i thought about it, i just wanted to check this for sanity 2019-10-11T22:52:24 < catphish> frankly, i wish my usb line were just a little faster and i could sent 32 bit values without worrying 2019-10-11T22:52:58 < catphish> but it's not fast enough to send 32 bytes at 2us intervals :( 2019-10-11T22:59:30 < catphish> yeah looks like my current input line can't reset the timer when capturing (CC3) 2019-10-11T22:59:39 < catphish> my own dumb hardware design choice 2019-10-11T23:01:39 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 268 seconds] 2019-10-11T23:11:21 < Cracki> pcb surgery 2019-10-11T23:11:44 < catphish> i'm just trying to think of simpler software solutions to this 2019-10-11T23:13:30 < catphish> since timings will never be less than about 50 cycles, i can safely use low values of the first byte for special purposes (most likely to indicate that a 16 or 32 bit value follows), and higher values as actual 1-byte values 2019-10-11T23:15:54 < catphish> for example [0x10-ff] == 16-255, [0x1, n] == 256+n, [0x2, n] == 512+n ... [0xf,n] == 15x256+n, [0x10, n,n,n,n] == 32 bit value 2019-10-11T23:16:13 < Cracki> maybe it's time to use some actual compression 2019-10-11T23:16:23 < catphish> *[0x11-ff] == 17-255 2019-10-11T23:17:35 < catphish> the scheme above should allow the duration value of any period to fit into bytes that can be transmitted in that period 2019-10-11T23:18:00 < catphish> i don't know a vast amount about proper compression 2019-10-11T23:19:12 < Cracki> fix the hardware maybe :P 2019-10-11T23:19:22 < Cracki> rewire the pins 2019-10-11T23:20:19 < catphish> i could redo the board or wire the pin to CC1 of the timer 2019-10-11T23:20:39 < catphish> but i actually prefer the idea of compressing the values 2019-10-11T23:21:06 < Cracki> a dirt simply scheme is what UTF-8 does. 2019-10-11T23:21:08 < catphish> because it means i don't have to do constant running addition of 0xff values 2019-10-11T23:21:31 < catphish> yeah what i proposed above isn't dissimilar to UTF8 actually 2019-10-11T23:22:15 < catphish> it abuses the fact that some bit of the 1-byte values will never be used (in my case, low values, in the case of ASCII it's the high values) 2019-10-11T23:23:25 < kakimir64> should vtable be a part of priviledged data? 2019-10-11T23:37:37 < specing> What does one do when your calculated reading from a large resistor divider are way off? 2019-10-11T23:39:03 < specing> I have 36.45V (average of two multimeters going through 100K-6600ohm divider, giving me 34274 mV reading 2019-10-11T23:39:27 < qyx> pray. 2019-10-11T23:39:39 < qyx> how large are the resistors? 2019-10-11T23:39:46 < qyx> and how long sampling time are you using? 2019-10-11T23:40:02 < qyx> search for AN "accuracy of stm32 adc" 2019-10-11T23:40:32 < specing> 100K-6.6K as stated, sampling time is slowest (240 clocks) and 12 MHz ADC peripheral clock 2019-10-11T23:40:57 < qyx> if you don't need fast sampling, a quick fix is to add a random cap parallel to the bottom resistor 2019-10-11T23:41:04 < specing> multimeter measured Vdd is 2.97V, ADC-measured Vdd is 3.007V, so reasonably accurate 2019-10-11T23:41:06 < qyx> say, 1n 2019-10-11T23:41:18 < qyx> oh you are getting larger value? 2019-10-11T23:41:29 < specing> no, lower 2019-10-11T23:41:38 < specing> the setup is on a perfboard so there should be plenty stray capacitance 2019-10-11T23:42:07 < qyx> 6k6 is too much 2019-10-11T23:42:13 < Cracki> specing, sample longer, change divider to have A LOT LOWER RESISTANCE 2019-10-11T23:42:13 < specing> why 2019-10-11T23:42:16 < Cracki> 100k is awful 2019-10-11T23:42:21 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-11T23:42:27 < specing> Cracki: why a lot lower? 2019-10-11T23:42:43 < Cracki> because stm32 adc need a certain amount of current to charge the sampling cap 2019-10-11T23:42:51 < specing> okay, so I'll add a capacitor 2019-10-11T23:42:52 < qyx> it is all described in the said AN 2019-10-11T23:42:55 < Cracki> so the higher impedance the signal, the longer you have to sample 2019-10-11T23:42:56 < specing> ok 2019-10-11T23:43:05 < Cracki> a cap isn't gonna help you 2019-10-11T23:43:10 < qyx> even with equations to calculate the error 2019-10-11T23:43:10 < Cracki> each sampling sucks current 2019-10-11T23:43:15 < specing> ok 2019-10-11T23:43:18 < Cracki> what qyx said 2019-10-11T23:43:20 < qyx> cap helps Cracki 2019-10-11T23:43:38 < Cracki> in some situations, I agree 2019-10-11T23:43:45 < qyx> I am regularly using ~1M dividers with 100n cap to measure Vbat 2019-10-11T23:43:52 < specing> " How to get the best ADC accuracy in STM32 microcontrollers" <- This? 2019-10-11T23:43:56 < qyx> yes 2019-10-11T23:43:57 < Cracki> if he samples for a microsecond every minute, a cap helps. if he continuously adcs as fast as possible, not so much 2019-10-11T23:44:16 < specing> Cracki: I'm sampling with 0.3Hz 2019-10-11T23:44:27 < qyx> thats ok 2019-10-11T23:44:49 < qyx> just try to piggyback some cap on the resistor 2019-10-11T23:44:59 < qyx> youll se if it helps 2019-10-11T23:52:56 < kakimir64> linker script wants everything to be in the first RAM block 2019-10-11T23:59:25 < kakimir64> 3 x 4K ram blocks 2019-10-11T23:59:40 < kakimir64> I need to think about it --- Day changed Sat Oct 12 2019 2019-10-12T00:31:08 < specing> qyx, Cracki: According to this simple (and probably wrong) calculation, no extra capacitors should be necessary: t0= t0 = 240 * (1/12e6); R_lo=6.6e3; R_hi = 1e5; U_divider = 36*R_lo/(R_lo+R_hi); C_sampl = 10e-12; tau = R_hi * C_sampl; U_sampl = U_divider * (1 - exp(-t0 /tau )) 2019-10-12T00:32:11 < specing> (I don't know which voltage to take for the capacitor time-to-fill calculation) 2019-10-12T00:33:26 < specing> Q = C_sampl*36*R_lo/(R_lo+R_hi), I = (36/R_hi); t = Q/I 2019-10-12T00:33:47 < specing> 61 ns to fill S&H capacitor 2019-10-12T00:34:05 < specing> well, I guess this equation is not taking into account the inductance of wires 2019-10-12T00:37:11 < qyx> 61ns to what percentagw? 2019-10-12T00:37:40 < mawk> Q/I is for it is a constant current source specing 2019-10-12T00:37:50 < mawk> but you have a constant voltage source 2019-10-12T00:38:00 < mawk> you just do 3tau for 95% loading 2019-10-12T00:39:01 < specing> mawk: doesen't a resistive voltage divider behave like a constant current source? 2019-10-12T00:39:09 < specing> disregarding inductance 2019-10-12T00:39:23 < Cracki> current ~ voltage across resistor 2019-10-12T00:39:24 < mawk> if the resistance is very high yes, approximately 2019-10-12T00:39:34 < specing> it is very high 2019-10-12T00:39:34 < Cracki> and since cap's voltage changes, so does the v across r 2019-10-12T00:39:36 < specing> 100k 2019-10-12T00:39:37 < BrainDamage> no, even if the resistance is high 2019-10-12T00:39:43 < mawk> ah 2019-10-12T00:39:48 < mawk> I meant if the voltage is very high 2019-10-12T00:39:48 < mawk> sorry 2019-10-12T00:40:01 < BrainDamage> because on a transient the cap will be a short and sink all the current 2019-10-12T00:40:15 < mawk> if the voltage is very high, and passing through a resistance with the right value you will have a ~constant current source specing 2019-10-12T00:40:22 < mawk> but it wastes a lot of power in the resistance, we don't do that 2019-10-12T00:40:34 < englishman> can anyone recommend some books on how to write bioses, system supervisors, power domain controllers or other shit like that 2019-10-12T00:40:48 < mawk> I read the Intel SDM for that englishman 2019-10-12T00:40:51 < mawk> it's very complete 2019-10-12T00:41:01 < mawk> the system programming volume, IIIa I think 2019-10-12T00:41:30 < BrainDamage> the condition to have a constant current, is that the voltage drop across the resistor on the head of the divider's constant 2019-10-12T00:41:46 < BrainDamage> which means that the ratio of the divider must be pretty large 2019-10-12T00:42:16 < BrainDamage> so even if the voltage on R2//C changes, the voltage on R1 changes little 2019-10-12T00:42:30 < BrainDamage> so the condition is R2 << R1 2019-10-12T00:42:48 < BrainDamage> but that also means that the final voltage will be a tiny fraction 2019-10-12T00:42:59 < englishman> hmm, thanks mawk, although very interesting, what i'm looking for is some guidelines on how to supervise and control disparate systems that might integrate multiple processors 2019-10-12T00:43:09 < BrainDamage> eg if you set a ratio of 1/100, it should work 2019-10-12T00:43:23 < mawk> ah I see 2019-10-12T00:43:41 < englishman> and ensure system settings are updated etc. over these domains 2019-10-12T00:48:19 -!- sterna [~Adium@c-d0e3e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-12T00:56:48 < specing> Cracki: qyx ok, this is much better now 2019-10-12T00:57:07 < specing> now the error is 0.5V, instead of -2.2V 2019-10-12T01:19:56 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-12T01:32:44 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-12T01:32:44 -!- Alexer [~alexer@alexer.net] has quit [Ping timeout: 268 seconds] 2019-10-12T01:33:08 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-12T01:44:21 -!- Alexer [~alexer@alexer.net] has joined ##stm32 2019-10-12T01:47:51 -!- Alexer [~alexer@alexer.net] has quit [Client Quit] 2019-10-12T01:48:12 -!- Alexer [~alexer@alexer.net] has joined ##stm32 2019-10-12T01:52:16 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds] 2019-10-12T02:14:03 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 250 seconds] 2019-10-12T02:15:09 < Ultrasauce> goodness me look at the time. it's music spam o'clock, UTC 2019-10-12T02:15:10 < Ultrasauce> https://dreamcatalogue.bandcamp.com/album/corrupt 2019-10-12T02:27:37 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-12T03:22:34 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-12T04:21:53 < Tordek> I don't understand, which clock do the timers use as "internal clock"? HSI? APB1? APB2? 2019-10-12T04:22:48 -!- User_ [~learningc@121.121.98.53] has joined ##stm32 2019-10-12T04:24:06 < Thorn> Tordek: look at the clock tab in cube 2019-10-12T04:24:41 < Thorn> and/or clock tree in the RM 2019-10-12T04:24:46 < Tordek> I'm there, yeah 2019-10-12T04:34:38 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-12T04:39:35 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has quit [Ping timeout: 268 seconds] 2019-10-12T05:02:15 < mawk> Tordek: APB1 and APB2 2019-10-12T05:02:20 < mawk> SysTick uses a different one tho 2019-10-12T05:02:25 < mawk> don't remember which 2019-10-12T05:02:32 < mawk> check the RM for which TIM uses which APB 2019-10-12T05:03:49 -!- onio [~onio@2a00:23c5:7a01:8600:a161:f997:855f:a492] has quit [Quit: Leaving] 2019-10-12T05:06:30 < Tordek> thanks mawk 2019-10-12T05:08:22 -!- learningc [~pi@121.121.98.53] has joined ##stm32 2019-10-12T05:19:47 < Cracki> chopped liver 2019-10-12T05:20:57 < Cracki> where fuck is the global warming we were promised? it's been dark and rainy for weeks and I'm getting grumpy 2019-10-12T05:21:10 < mawk> lol 2019-10-12T05:21:40 < Cracki> also curious: all the "historical" climate charts seem to start at around 1900 (like terrorism casualties start after 9/11/2001) 2019-10-12T05:22:24 < Cracki> before that shit was warmer but let's not mention that... Inconvenient Truth 2019-10-12T05:30:17 < mawk> well you have the projections 2019-10-12T05:30:24 < mawk> and even the most gentle lower estimates don't look good for us 2019-10-12T05:30:37 < mawk> it's not about what it is now, it's about what could happen 2019-10-12T05:30:53 < mawk> with respect to temperature rise, and then mechanically see level rise, and then climate change 2019-10-12T05:31:18 < mawk> if the models predict that with enough confidence I don't see a reason for rejection 2019-10-12T05:31:24 < mawk> it's not like we're doing 15th century maths 2019-10-12T05:31:34 < mawk> we do things to a very high degree of confidence in modern physics 2019-10-12T05:31:51 < mawk> at least in the more theoretical stuff, I don't know about climatology 2019-10-12T05:51:44 < Cracki> >projections 2019-10-12T05:52:03 < Cracki> for half a century, doomsday predictions have turned out to be wrong 2019-10-12T05:54:15 < Cracki> the only confidence I can find is that doomsday predictions are false, and it's no stretch to assume that they're made with ill intent 2019-10-12T05:54:36 < mawk> I read the contrary 2019-10-12T05:54:37 < mawk> well 2019-10-12T05:54:41 < mawk> that projections were too gentle 2019-10-12T05:54:45 < Cracki> the predictions that predicted Now most accurately decades ago come from oil companies and other energy companies 2019-10-12T05:54:45 < mawk> from the last centuries 2019-10-12T05:54:53 < mawk> so yeah they were wrong, but in a bad way 2019-10-12T05:54:59 < mawk> at least for temperature rise 2019-10-12T05:55:04 < mawk> I don't know about the rest 2019-10-12T05:57:24 < Cracki> there are some assumptions that shouldn't be blindly accepted 2019-10-12T05:57:29 < Cracki> such as how numbers are presented 2019-10-12T05:57:41 < Cracki> and what they really indicate 2019-10-12T05:57:58 < Cracki> I just did a quick google (yeah yeah) on polar ice caps 2019-10-12T05:58:17 < Cracki> what I find is UNITLESS plots 2019-10-12T05:58:21 < mawk> lol 2019-10-12T05:58:25 < mawk> yeah media like to do that 2019-10-12T05:58:33 < Cracki> "anomaly" expressed in standard deviations relative to some arbitrary time span 2019-10-12T05:58:34 < mawk> or cut axis 2019-10-12T05:58:53 < Cracki> so if the standard deviation were ten square centimeters, that plot would be worth shit 2019-10-12T05:59:41 < Cracki> the safest assumption is that we're being lied to. can't give trust that hasn't been earned, let alone destroyed. 2019-10-12T06:01:16 < Cracki> the top handful of plots for one search (polar ice cap plot) all start in 1978 2019-10-12T06:04:24 < Cracki> >50 years of failed doomsday, eco-pocalyptic predictions 2019-10-12T06:04:50 < Cracki> those might be nutjob websites but that doesn't prevent anyone from checking that those claims were actually made and that they all missed the mark 2019-10-12T06:05:31 < Cracki> doomsday cults have always been a thing. only today do we deny their existence. 2019-10-12T06:06:25 < Cracki> >if the current [...] trend continues 2019-10-12T06:21:59 < jadew> sup? 2019-10-12T06:22:15 < jadew> doomsday cults... hmm that puts me in the mood for a horror movie 2019-10-12T06:32:26 -!- fc5dc9d4_ [~quassel@p57A324E8.dip0.t-ipconnect.de] has joined ##stm32 2019-10-12T06:36:02 -!- fc5dc9d4 [~quassel@p5B0816AF.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-12T06:41:25 < Cracki> is Idiocracy enough horror? 2019-10-12T06:42:27 < Cracki> I want y2k hysteria back 2019-10-12T07:08:31 < jadew> saw that too many times 2019-10-12T07:31:32 < Cracki> ex machina movie maybe 2019-10-12T07:38:05 < aandrew> ex machina was such a fucking stupid movie 2019-10-12T07:38:13 < aandrew> so disappointed with it 2019-10-12T07:38:48 < jadew> what do you recommend? 2019-10-12T07:39:49 < jadew> I saw ex machina 2019-10-12T07:50:40 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-12T07:53:38 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-12T07:53:39 -!- day__ is now known as day 2019-10-12T07:54:10 < aandrew> recommendations are hard 2019-10-12T07:55:15 < jadew> maybe I'll take the family to the movies this morning 2019-10-12T07:55:27 < jadew> I better see what's on today 2019-10-12T07:56:23 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 250 seconds] 2019-10-12T07:57:27 < jadew> nothing worth seeing 2019-10-12T07:59:17 < aandrew> you wouldn't take the kids to see ex machina 2019-10-12T07:59:37 < aandrew> I don't know what's new that's worthwhile, it's all shitty "live action" remakes of animated movies these days 2019-10-12T07:59:40 < jadew> actually, for my kid I'd like something with fighting (think ninjas) 2019-10-12T07:59:49 < aandrew> kung fu panda 2019-10-12T07:59:54 < jadew> didn't like that 2019-10-12T07:59:58 < aandrew> fine. 2019-10-12T07:59:59 < aandrew> matrix 2019-10-12T08:00:08 < jadew> hmm, that's not bad 2019-10-12T08:00:14 < jadew> I might see it with him 2019-10-12T08:00:20 < aandrew> I liked mulan, but I'm a sucker for disney movies for kids 2019-10-12T08:00:22 < aandrew> how old is he? 2019-10-12T08:00:26 < jadew> 6 2019-10-12T08:00:37 < aandrew> heh I'd not take a 6yo to matrix 2019-10-12T08:00:39 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-12T08:00:44 < aandrew> but I coddle my kids 2019-10-12T08:00:49 < aandrew> something I'm working on 2019-10-12T08:00:53 < jadew> he can take it 2019-10-12T08:00:54 < aandrew> but they have their entire lives to be adults 2019-10-12T08:01:05 < aandrew> prefer to keep them innocent as long as I can 2019-10-12T08:01:07 < jadew> he saw a UFC fight with me the other day, when one of the guys got mauled really bad 2019-10-12T08:01:15 < jadew> I was afraid it might affect him in a wrong way 2019-10-12T08:01:18 < aandrew> heh 2019-10-12T08:01:24 < jadew> instead he said he liked it and started punching like a pro 2019-10-12T08:02:20 < jadew> I'm going to get some training gear this weekend 2019-10-12T08:02:49 < jadew> he only has boxing gloves and I can take those, but it seems he likes kicking too 2019-10-12T08:04:09 < jadew> oh, and we saw oblivion together 2019-10-12T08:04:33 < jadew> he loved that one too and most of the movie is war scenes 2019-10-12T08:05:16 < jadew> oh, sorry, not oblivion 2019-10-12T08:05:26 < jadew> the other one with tom cruise 2019-10-12T08:05:48 < jadew> edge of tomorrow 2019-10-12T08:07:21 < jadew> aandrew, they're not maturing faster if you give them access to info 2019-10-12T08:09:59 < aandrew> I disagree quite strongly 2019-10-12T08:10:08 < aandrew> there's a difference between education and maturity 2019-10-12T08:10:24 < jadew> I think the only bit of info I kept from him so far was how I put my cell into his mom 2019-10-12T08:10:34 < jadew> other than that, I've been giving him every bit of information he asked for 2019-10-12T08:10:44 < aandrew> it's the type of information and how their mind can process it based on age and their own developmental levels 2019-10-12T08:11:33 < aandrew> I share quite a bit with my kids too, we talk of science and math and all kinds of tech shit, but I work quite hard to limit their exposure to violence or "stupidity" 2019-10-12T08:11:44 < jadew> I don't know... after they learn that their parents are going to die and that they are going to die, I don't think there's much more you can tell them that would be difficult to process 2019-10-12T08:11:47 < aandrew> e.g. vice.com type crap online and people doing stupid things for attention 2019-10-12T08:12:00 < jadew> oh, right 2019-10-12T08:12:05 < jadew> I don't allow him to watch crap either 2019-10-12T08:12:20 < aandrew> it's so easy for them to get sucked in, and then they think that's how to be funny/cool 2019-10-12T08:12:24 < jadew> but that's not really information, that's stupid entertainment 2019-10-12T08:12:28 < aandrew> so so so so easy 2019-10-12T08:12:37 < jadew> and they can associate it with likable behaviour 2019-10-12T08:12:38 < aandrew> well what did I read recently 2019-10-12T08:13:01 < aandrew> something like the #1 "job" kindergarden (i.e. beofre grade 1) students wanted to be when they grew up 2019-10-12T08:13:09 < aandrew> "youtube star" and "influencer" 2019-10-12T08:13:12 < aandrew> what. the. fuck. 2019-10-12T08:13:34 < jadew> yeah, that's silly 2019-10-12T08:13:39 < aandrew> not silly, scary 2019-10-12T08:14:18 < aandrew> my kids have almost zero exposure to tv commercials, it was very ... wonderful? to see them get upset at a friend's house becuase they were watching tv and a commercial came on. they had no concept of that 2019-10-12T08:14:20 < jadew> but violence is not stupid tho 2019-10-12T08:14:25 < jadew> it's very real 2019-10-12T08:14:26 < Cracki> TIL: crisp desktop video capture on windows, even if that fucking Game Bar refuses to work: https://github.com/MathewSachin/Captura 2019-10-12T08:14:37 < aandrew> I think a young mind has trouble understanding the place for violence 2019-10-12T08:14:43 < Cracki> (perfect for presentations, you can even tell it to use the qtrle codec) 2019-10-12T08:14:54 < Cracki> I hear violence? 2019-10-12T08:15:00 < jadew> aandrew, if you don't take the time to explain it, sure 2019-10-12T08:15:14 < jadew> but it's difficult to explain a war, without understanding violence 2019-10-12T08:15:18 < aandrew> agreed 2019-10-12T08:15:52 < aandrew> my 7yo loves to play tank games, knows all about the specific capabilities of different types of artillery, understands active armor etc etc 2019-10-12T08:15:59 < jadew> also, sport violence is very different from random violence 2019-10-12T08:16:05 < jadew> especially when the guys hug at the end 2019-10-12T08:16:06 < aandrew> but has zero mature concept of killing 2019-10-12T08:16:20 < jadew> it shows that it's just for sports and there is no malevolence there 2019-10-12T08:17:34 < aandrew> my 10yo knows more about the technical aspects of flying, can name all the major systems in a modern jetliner and how airbus/boeing are different in how they approach design, can recite specific details of probably a hundred different actual accidents including the flight numbers, but does not really understand what a crash does to everyone involved, directly and indirectly 2019-10-12T08:17:41 < Cracki> tank games are like chess 2019-10-12T08:17:46 < Cracki> in chess you kill pieces too 2019-10-12T08:17:47 < aandrew> that's what I mean by innocence/maturity 2019-10-12T08:17:55 < aandrew> yes very similar, exactly 2019-10-12T08:17:55 < Cracki> it's abstract 2019-10-12T08:18:09 < aandrew> hell, most idiot gangbangers have no idea what shooting a person actually is like 2019-10-12T08:18:13 < aandrew> it's all glorified "bang bang" shit 2019-10-12T08:18:16 < jadew> aandrew, how did you explain to them that you're going to die? 2019-10-12T08:18:25 < jadew> didn't they deal with that already? 2019-10-12T08:18:26 < Cracki> teach him about the easter bunny and santa and religions and cults 2019-10-12T08:18:30 < aandrew> they understand death to an extent 2019-10-12T08:18:44 < Cracki> and by that I mean teach people's need to believe in something 2019-10-12T08:18:45 < aandrew> these kids have never believed in santa or any of that shit 2019-10-12T08:18:56 < Cracki> and how that makes people act insane 2019-10-12T08:19:06 < aandrew> when I was younger I had no idea what the purpose of faith and religion was 2019-10-12T08:19:18 < Cracki> because your IQ isn't double digit 2019-10-12T08:19:28 < Cracki> teach them about the reality of IQ 2019-10-12T08:19:44 < aandrew> I do now and it's not as jaded as I used to think 2019-10-12T08:19:53 < Cracki> and the difference between thinking and acting 2019-10-12T08:20:07 < jadew> aandrew, after seeing how my kid dealt with the concept of death, I realized there's nothing he can't process 2019-10-12T08:20:26 < jadew> it took a while for him to work through it, because it's painful 2019-10-12T08:20:27 < aandrew> the 7yo is an actual certified genius, member of mensa etc etc and the 10yo isn't far behind him in actual IQ score 2019-10-12T08:20:31 < aandrew> but that's not what I'm talking about 2019-10-12T08:20:46 < Cracki> kids still need guidance. show him one of those drag queen story hours they do in murica, with the naked demons 2019-10-12T08:20:48 < aandrew> these kids have never seen/known death at a personal level 2019-10-12T08:20:54 < jadew> but once it sank in and he got over it, he was tough enough to learn about other people dying (who he doesn't care about) 2019-10-12T08:20:58 < Cracki> or that godawful tv show where they talk about children and sex 2019-10-12T08:21:01 < aandrew> in fact I clearly remember my own first experience with death 2019-10-12T08:21:03 < Cracki> that might fuck him up 2019-10-12T08:21:29 < Cracki> the death of a pet is probably the earliest experience kids have of death in general 2019-10-12T08:21:29 < aandrew> it was a friend of the family, I barely knew her 2019-10-12T08:21:39 < jadew> aandrew, me too, but I'm not talking about real experience, just understanding how things work 2019-10-12T08:21:39 < aandrew> did the whole funeral and wake, I must have been 7 or so 2019-10-12T08:22:06 < jadew> my grandma used to take me to funerals since I was younger than that 2019-10-12T08:22:11 < Cracki> the awkwardness is mostly because kids sense that it's a touchy subject to adults but they need to know how to act, but adults are too pussy to sit them down and lay it on the table 2019-10-12T08:22:22 < jadew> also saw a family member die when I was about that age (6-7 yo) 2019-10-12T08:22:28 < jadew> didn't affect me much 2019-10-12T08:22:28 < aandrew> it was completely lost on me until bedtime, when I suddenly realized that that was an actual, once-alive person lying in that coffin and I fell apart. Like I said I barely knew her and she was not in any way shape or form some part of my life 2019-10-12T08:23:10 < aandrew> actually my nephew in .ro committed suicide when he was 16. that was like 6 or 7 years ago now 2019-10-12T08:23:14 < Cracki> what annoyed me about adults was their mysterious rules for what's ok and what's not, and instead of plainly explaining when I've stepped into some unseen landmine, they FLIP THE FUCK OUT and rip my head off 2019-10-12T08:23:32 < aandrew> his little sister (5 at the time) actually saw him do it 2019-10-12T08:23:47 < jadew> aandrew, that sucks, puberty is dangerous 2019-10-12T08:23:58 < jadew> aandrew, that's awful... 2019-10-12T08:24:10 < Cracki> what kids need to know about suicide is that the person suffered before it, and now they don't 2019-10-12T08:24:12 < aandrew> that girl is *brilliant*, I pay for her school in .ro (special math school) and she comes over roughly every year, brilliant and funny and kind in every way 2019-10-12T08:24:21 < Cracki> and maybe what (who) caused them to suffer 2019-10-12T08:24:21 < aandrew> but she is completely *fucked* from an emotional place 2019-10-12T08:24:41 < jadew> Cracki, yeah, we're explaining everything (including why he's not allowed to do something) 2019-10-12T08:24:46 < aandrew> it's taken her until JUST this summer to tell anyone what she knew and saw about her brother's suicide 2019-10-12T08:24:59 < aandrew> the guilt she felt evne though she had absolutely no control or power over the situation 2019-10-12T08:25:00 < jadew> even if the explanation is "because it's annoying me and if it keeps going on, I'm going to get angry" 2019-10-12T08:25:19 < Cracki> maybe she has no role models that show her how to deal with (talk about) such things 2019-10-12T08:25:30 < jadew> aandrew, yeah, going through something like that must be awful 2019-10-12T08:25:41 < Cracki> when I grew up, it was an environment where hugging never happened, not in the family, not at school, not at class mates' homes 2019-10-12T08:25:44 < jadew> but it's not the same as teaching a kid about the ways of the world 2019-10-12T08:25:58 < aandrew> it's a long road for her but she is slowly working through everything she has had to deal with since that event 2019-10-12T08:26:10 < aandrew> yes, it's different for sure 2019-10-12T08:26:36 < jadew> also, let me ask you something 2019-10-12T08:26:45 < aandrew> but what I am telling the story for is to try to describe how it is very typical for small children to internalize things in ways that adults simply never concieve of 2019-10-12T08:26:53 < jadew> when you were young, 6-7 yo, didn't you like action movies? 2019-10-12T08:27:11 < jadew> there was so much killing in them, it was crazy 2019-10-12T08:27:14 < aandrew> I'm not saying that I'm right and you're wrong or anything like that, but it's often hard to judge exactly what a child is pulling out of a situation 2019-10-12T08:27:31 < jadew> all the kids I knew saw them and it only led to us pretending we're ninjas 2019-10-12T08:27:33 < aandrew> jadew: it's hard for me to remember my own childhood, but yes, action type movies are always fun and popular 2019-10-12T08:27:37 < jadew> and enacting fighting scenes 2019-10-12T08:28:15 < aandrew> it was only last year that I watched the back to the future series with them, and then moved into indiana jones, league of extraordinary gentlemen and sky cpatain and the world of tomorrow 2019-10-12T08:28:30 < aandrew> but those are all very very mild and .. idealized violence 2019-10-12T08:28:57 < aandrew> I want to get into iron man with them becuase it's probably one of my favourite movies (don't ask me about the other two iron men movies, they don't exist in my mind) 2019-10-12T08:29:20 < aandrew> they've never seen star wars, not because I think it's too violent but because I never had any ambition to see them so have no ambition to sit down with them now to see it 2019-10-12T08:29:30 < jadew> I liked edge of tomorrow because the enemy was not human 2019-10-12T08:29:36 < aandrew> jumanji actually freaked my then-9yo out 2019-10-12T08:29:43 < aandrew> I don't remember edge of tomorrow 2019-10-12T08:29:48 < jadew> so it didn't have that layer of malevolence and hate that other war movies have 2019-10-12T08:29:52 < Cracki> when I was little and I saw action movies, let's say Last Action Hero (where they satirize it and act like real people die during movie making), it was obvious to me that they're movies and nobody gets hurt 2019-10-12T08:30:11 < jadew> Cracki, yeah, same here 2019-10-12T08:30:17 < aandrew> yes, and we always talk about how movies and fiction books are not real and they understand 2019-10-12T08:30:27 < aandrew> but it's very hard for htem to separate something they *see* as not real sometiems 2019-10-12T08:30:36 < aandrew> particuarly if it's really compelling 2019-10-12T08:30:57 < jadew> nah, not in my experience 2019-10-12T08:31:14 < jadew> he would rather remember a story I told him, if I mentioned it was real 2019-10-12T08:31:19 < Cracki> I liked edge of tomorrow because it's a brainier movie than the ordinary action movie where violence IS the plot 2019-10-12T08:31:22 < jadew> and it had some important information there 2019-10-12T08:31:55 < jadew> like... if I tell him about something that happened when someone got hurt, he might come with a question about that even even 2 months later 2019-10-12T08:31:59 < Tordek> I want to modify several pins on a port and I see a mention of GPIOx_BSRR; if I'm modifying pins PEx I should be writing to GPIOE_BSRR right? But there's no such variable, at least according to autocomplete? 2019-10-12T08:32:02 < jadew> never happens for movies 2019-10-12T08:32:50 < Cracki> Tordek, hal/ll or libopencm3 or something else? 2019-10-12T08:32:56 < Tordek> HAL 2019-10-12T08:33:07 < Cracki> then you use hal's functions for touching gpios 2019-10-12T08:33:17 < Cracki> no touching plain registers 2019-10-12T08:33:30 < Tordek> is there one for changing several at once? 2019-10-12T08:33:36 < aandrew> this is actually one of the reasons i hate this whole Greta Thunberg shit 2019-10-12T08:33:51 < Cracki> uhhhh you'll wanna look into hal docs. they come with cubemx, are probably some chm files 2019-10-12T08:33:56 < aandrew> it's weaponizing the kid's natural instinct to feel responsible for things outside their control 2019-10-12T08:33:57 < Thorn> https://lcsc.com/product-detail/DC-DC-Converters_RT6150AGQW_C147971.html https://lcsc.com/product-detail/DC-DC-Converters_TI_TPS63030DSKR_TPS63030DSKR_C12462.html https://lcsc.com/product-detail/DC-DC-Converters_MPS_MP2155GQ-Z_MP2155GQ-Z_C110397.html 2019-10-12T08:33:59 < Tordek> thanks 2019-10-12T08:34:03 < jadew> aandrew, they live in an era when they're bombarded with entertainment, I think their brains are used to discard it 2019-10-12T08:34:04 < Cracki> child soldiers 2019-10-12T08:34:08 < jadew> *discarding it 2019-10-12T08:34:09 < Cracki> children as shields 2019-10-12T08:34:27 < jadew> I don't know what that is about 2019-10-12T08:34:37 < jadew> what do I google? Greta Thunberg? 2019-10-12T08:34:41 < Cracki> exploiting children's NEED to feel accepted and validated 2019-10-12T08:34:56 < aandrew> Tordek: you can use the HAL_GPIO_Init() with multiple pins ont he same port 2019-10-12T08:34:57 < Cracki> fridays for future, extinction rebellion, the doomsday cults 2019-10-12T08:35:16 < aandrew> Tordek: e.g. io.Pin = GPIO_PIN_1 | GPIO_PIN_5 | GPIO_PIN_6 | GPIO_PIN_10 2019-10-12T08:35:27 < aandrew> (if you want them all to have the same direction/function) 2019-10-12T08:35:30 < jadew> "teenage environmental activist" LOL 2019-10-12T08:35:48 < Tordek> WritePin(..., Pin1 | Pin2 | ...)? 2019-10-12T08:35:48 < aandrew> jadew: teenage is bad enough but they're making kids your son's age worry about it 2019-10-12T08:35:51 < jadew> like anyone is going to listen to a kid on any serious matter 2019-10-12T08:35:57 < Cracki> their message is: humans are born evil (we destroy the planet), so now you children must flagellate yourselves and hate yourselves and loudly proclaim to hate everyone,a nd then people will think you're morally superior 2019-10-12T08:36:10 < aandrew> Tordek: I think so; have to look at the hal_gpio.c file to make sure 2019-10-12T08:36:16 < aandrew> the pins are bit positions 2019-10-12T08:36:18 < Tordek> thanks 2019-10-12T08:36:38 < Cracki> it's political recruitment. they offer various levels of engagement, from tweeting about it upto throwing molotov cocktails at cops 2019-10-12T08:36:48 < aandrew> jadew: certainly you've heard about the swedish teenager who is lecturing UN on environmentalism 2019-10-12T08:36:57 < Cracki> it's action drills. it's literally drills. 2019-10-12T08:36:58 < jadew> not until now 2019-10-12T08:37:06 < aandrew> jadew: wow, I am envious 2019-10-12T08:37:07 < jadew> it sounds so cringe... 2019-10-12T08:37:07 < Tordek> ah, Write takes SET/RESET, that's great, thanks! 2019-10-12T08:37:36 < aandrew> jadew: I get the message and yes, believe we need to take way better care of the planet, but the way she's doing it is just so... ugh 2019-10-12T08:37:43 < aandrew> she's being manipulated 2019-10-12T08:38:11 < jadew> obviously 2019-10-12T08:38:19 < jadew> everyone under 25 is stupid 2019-10-12T08:38:28 < Cracki> they want to lower the voting age to 16 2019-10-12T08:38:33 < jadew> they don't have a wide enough view on how the world works 2019-10-12T08:38:34 < Cracki> it's all to get votes 2019-10-12T08:38:41 < jadew> or the human nature to begin with 2019-10-12T08:38:44 < aandrew> she's got aspergers and a host of anxiety issues and nobody can put 2 and 2 together and think that her going around clutching her pearls about the environment ahs anything to do with her mental issues 2019-10-12T08:38:58 < Cracki> even 25 is debatable for western europeans who are held in delayed puberty for longer than the rest of the world 2019-10-12T08:39:09 < aandrew> yeah I think if you're old enough to bleed for your country you are old enough to vote 2019-10-12T08:39:12 < Cracki> her parents play a part in it 2019-10-12T08:39:33 < aandrew> you either bring the recruitment/conscription age up to voting age or bring voting age down 2019-10-12T08:39:36 < aandrew> personally I prefer the former 2019-10-12T08:39:38 < Cracki> child soldiers can bleed as soon as they can pick up a gun 2019-10-12T08:39:42 < Cracki> and those are child soldiers 2019-10-12T08:40:00 < aandrew> because it's too easy to fill empty and hormone-flooded young minds with propaganda 2019-10-12T08:40:12 < jadew> wait... what's the voting age? 2019-10-12T08:40:13 < aandrew> personally I think basic trainign shoudl be part of the requirements for high school education though 2019-10-12T08:40:28 < jadew> in here, it's 18 for both 2019-10-12T08:40:34 < Cracki> same 2019-10-12T08:40:49 < Cracki> and we don't have the draft no more 2019-10-12T08:41:18 < jadew> we neither 2019-10-12T08:41:31 < jadew> I was the last generation to get drafted 2019-10-12T08:41:34 < jadew> but they didn't take me 2019-10-12T08:41:45 < Cracki> I'd disconnect (and elevate) granting a political voice from granting general competency to act in society 2019-10-12T08:41:58 < jadew> I think I nailed the IQ test and got to speak with the psychologist 2019-10-12T08:41:59 < Cracki> becuase in the latter you're responsible for yourself 2019-10-12T08:42:06 < Cracki> in the former you make decisions for your whole people 2019-10-12T08:42:31 < Cracki> you weren't dumb enough to follow simple orders? 2019-10-12T08:42:38 < aandrew> Cracki: if you are asked to die for your country you should have a say in its policies which send you off to die 2019-10-12T08:42:53 < jadew> we had a chat and I think she got me out 2019-10-12T08:43:03 < Cracki> let's not draft or accept udner-25 year olds for military service then 2019-10-12T08:43:04 < jadew> I basically told her I don't want to be there and I find the army stupid anyway 2019-10-12T08:43:11 < jadew> also let her know I'm not good at following orders 2019-10-12T08:43:23 < Cracki> let's only allow people into the military after they've had kids 2019-10-12T08:43:31 < Cracki> because *then* they have a stake in the future of their nation 2019-10-12T08:43:43 < jadew> that's actually not a bad idea 2019-10-12T08:44:03 < Cracki> only give the right to vote to those who have a stable job 2019-10-12T08:44:10 < jadew> but I think wars work differently 2019-10-12T08:44:11 < Cracki> or at least not live on welfare 2019-10-12T08:44:20 < Cracki> military service is peace time 2019-10-12T08:44:22 < jadew> while someone older could be a better fighter (smarter) 2019-10-12T08:44:31 < Cracki> when war breaks out, you shoot those that don't fight 2019-10-12T08:44:32 < jadew> they probably just need bodies to push the front line 2019-10-12T08:44:38 < aandrew> oh ffs 2019-10-12T08:44:38 < aandrew> MemFault MMSR:0x00000082 MMAR:0x00000004 $pc:0x00258bdc $lr:0x00258bd5 r[0x00000019 0x12000000 0x00000000 0x00000000] r12:0x00000000 2019-10-12T08:44:40 < jadew> at least I think that was the case historically 2019-10-12T08:44:50 < Cracki> fridays for future ALSO needs dumb child bodies for cannon fodder 2019-10-12T08:45:16 < aandrew> Cracki: I feel that the people who declare war should require their kids to go to the front lines first 2019-10-12T08:45:22 < Cracki> absolutely 2019-10-12T08:45:28 < aandrew> Cracki: that would make those people think twice about this shit 2019-10-12T08:45:34 < jadew> completely agree 2019-10-12T08:45:53 < Cracki> the requirement to have kids should be for all politicians 2019-10-12T08:46:00 < jadew> it's easy to want war when you have nothing to lose 2019-10-12T08:47:01 < jadew> on the other hand, if it's voluntary... 2019-10-12T08:47:07 < jadew> those people knew what they're signing up for 2019-10-12T08:47:22 < jadew> so... yeah 2019-10-12T08:48:37 < jadew> why do people willingly enroll in the army anyway? 2019-10-12T08:49:03 < aandrew> Kurt Vonnegut wrote abou tit 2019-10-12T08:49:21 < aandrew> there *are* good reasons to defend your country 2019-10-12T08:49:53 < aandrew> he even said, after being a prisoner of war in WW2, that if it were needed again, and it were a just war, he would like to volunteer again 2019-10-12T08:50:25 < aandrew> my maternal grandfather was polish and fled to Canada to avoid the nazis 2019-10-12T08:50:53 < aandrew> my paternal grandfather was more valuable in the nickel mines in Sudbury than as a solider so they told him to stay there 2019-10-12T08:51:07 < aandrew> I don't have any familial experience with war 2019-10-12T08:53:03 < Cracki> ebay needs an option to laugh at listings. someone is selling an item for 250 bucks, with 200 bucks shipping to my country. 2019-10-12T08:53:19 < aandrew> lol 2019-10-12T08:53:24 < aandrew> I like that idea 2019-10-12T08:53:29 < aandrew> a "lolwut" button 2019-10-12T08:54:00 < jadew> is it a heavy item? 2019-10-12T08:54:26 < jadew> if it's heavy and they're only offering express shipping, then that could make sense 2019-10-12T08:54:39 < jadew> I recently stopped offering standard shipping outside of the EU 2019-10-12T08:54:55 < jadew> only DHL for non-EU, because it's less of a hassle 2019-10-12T08:55:14 < jadew> if it's a similar scenario and it's also heavy, then it might make sense 2019-10-12T08:56:24 < Cracki> item weighs as much as a bar of chocolate 2019-10-12T08:56:35 < Cracki> (magewell hdmi to usb3 thingy) 2019-10-12T08:56:37 < jadew> then it's a ripoff 2019-10-12T08:56:41 < Cracki> obv :P 2019-10-12T08:56:57 < Cracki> must be super fast instant shitting 2019-10-12T08:57:16 < Cracki> they'll send a courier over the atlantic 2019-10-12T08:57:36 < Cracki> oh god how much I hate dhl express 2019-10-12T08:57:56 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-12T08:58:00 < jadew> I like it for shipping with them 2019-10-12T08:58:10 < Cracki> thankfully never experienced it myself but they seem to rip off everyone with extra processing fees and import tax and whatnot 2019-10-12T08:58:43 < Cracki> there must be no law forcing them to be up front about it all 2019-10-12T08:58:56 < jadew> oh, there's a way here in which you can dodge that 2019-10-12T08:59:13 < jadew> once I figured out how to prepare all the documentation for them, it all went smoothly 2019-10-12T08:59:24 < Cracki> wizardry 2019-10-12T08:59:55 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-12T09:00:08 < Cracki> I wish it were legal to Go Postal on these clowns for when they fuck you over 2019-10-12T09:01:26 < jadew> what I'd like is to get the UPS rates that farnell is getting 2019-10-12T09:01:40 < jadew> for $5 they ship overnight from the UK to Romania 2019-10-12T09:03:22 < jadew> I'm starting to become indifferent to taxes and all that 2019-10-12T09:03:27 < jadew> I just factor them in 2019-10-12T09:04:07 < aandrew> I've only had good experiences with DHL 2019-10-12T09:04:10 < aandrew> I'm in Canada 2019-10-12T09:04:17 < aandrew> although today UPS pissed me off something terrible 2019-10-12T09:04:29 < aandrew> waiting on a shipment that i paid overnight for 2019-10-12T09:05:14 < aandrew> at 10:29am I check the site and see "delivery attempt at 10:25am" -- no sticker on the door saying they were here, and the security camera shows there was no fucking truck even drove past the street all morning 2019-10-12T09:05:30 < aandrew> I call them and complain, they say they'll contact dispatch with an urgent message and get it taken care of 2019-10-12T09:05:44 < jadew> I had that happen too 2019-10-12T09:05:55 < aandrew> 1.5h later I call and they say that the message was not sent urgent which is why there was no call, and then say they'll send it urgent this time 2019-10-12T09:06:12 < aandrew> I said no fucking dice, get me this box that I paid for and you guys didn't even ATTEMPT to fucking deliver 2019-10-12T09:06:47 < aandrew> "oh it says it was delivered" wtf the site says it coudln't be delivered because I wasn't home and now you're telling me it was delivered?" 2019-10-12T09:06:58 < aandrew> got a supervisor who actually kept me on hold while she talked to dispatch 2019-10-12T09:07:16 < aandrew> 45m on hold and the driver shows up to deliver the package and has this look of "what, I did nothing wrong" 2019-10-12T09:07:22 < aandrew> wanted to punch this fucker right in the fucking mouth 2019-10-12T09:07:41 < aandrew> lady on the phone comes back and says she just got finished talking to dispatch who had her on hold while talking to the driver 2019-10-12T09:07:49 < aandrew> so I told her while I was on hold he just gave me the package 2019-10-12T09:08:21 < aandrew> before the supervisor the CSR was like "do you want us to open an investigation?" fuck no I want my fucking delivery which is what you were paid to do 2019-10-12T09:08:32 < aandrew> the supervisor said they'll open an investigation into what happened 2019-10-12T09:08:49 < aandrew> after this driver's attitude I told them I'll happily share my security footage for the morning if they like 2019-10-12T09:08:52 < aandrew> fuck him 2019-10-12T09:08:54 < aandrew> some young punk 2019-10-12T09:09:04 < aandrew> I was expecting an old fucker who didn't give a shit 2019-10-12T09:09:08 < aandrew> not a young guy, but there you have it 2019-10-12T09:09:40 < aandrew> you want to play "delivery attempt" games don't do it with people with surveillance on their homes 2019-10-12T09:10:39 < aandrew> anyway 2019-10-12T09:10:40 < Cracki> lol 2019-10-12T09:10:42 < aandrew> this memfault is weird 2019-10-12T09:10:45 < Cracki> no mercy 2019-10-12T09:11:00 < aandrew> it's always in the middle of a printf() but the printf() works hundreds of times until the memfault 2019-10-12T09:11:12 < aandrew> wonder if I have a memory leak or stack corruption 2019-10-12T09:11:18 < Cracki> define memfault 2019-10-12T09:11:22 < Cracki> got any interrupts? 2019-10-12T09:11:28 < aandrew> MemFault MMSR:0x00000082 MMAR:0x00000004 $pc:0x00258b9c $lr:0x00258b95 r[0x00000019 0x12000000 0x00000000 0x00000000] r12:0x00000000 2019-10-12T09:11:28 < Cracki> might be stack/heap corruption 2019-10-12T09:11:31 < aandrew> that 2019-10-12T09:11:34 < Cracki> ah 2019-10-12T09:11:44 < aandrew> gotta look up what MMSR/MMAR are again 2019-10-12T09:12:01 < Cracki> sometimes I wish segger would implement "backwards stepping" 2019-10-12T09:12:19 < Cracki> with enough checkpointing it could be done 2019-10-12T09:12:27 < aandrew> yeah I'd love that 2019-10-12T09:12:43 < Cracki> I think they even have something almost like it 2019-10-12T09:12:51 < aandrew> TIPU should help with that if you ahve trace pins brought out and expensive equipment to log it 2019-10-12T09:12:54 < Cracki> you can restore the DUT's state from a saved debug session 2019-10-12T09:13:23 < Cracki> so all it needs is regular checkpoints, and then you can bisect on the program counter 2019-10-12T09:13:40 < aandrew> 0x82 = valid fault data, not stacking fault, not unstacking fault, DACCVIOL = invalid data access 2019-10-12T09:13:46 < aandrew> (memory did not allow the access type 2019-10-12T09:13:57 < Cracki> oh yes tracing but that'd mean buying their four figures devices 2019-10-12T09:14:40 < aandrew> wait 2019-10-12T09:14:57 < aandrew> MMAR=0x00000004.. that means it tried to access 0x00000004 2019-10-12T09:15:03 < aandrew> that's almost a null pointer deref 2019-10-12T09:17:44 < aandrew> interesting 2019-10-12T09:17:45 < aandrew> https://pastebin.com/hjJsc7n2 2019-10-12T09:17:58 < aandrew> it faulted on the ldr r3, [r7, #4] 2019-10-12T09:18:07 < aandrew> R7 must be 0x00000000 2019-10-12T09:18:14 < aandrew> but the LR looks fucked 2019-10-12T09:18:29 < aandrew> 0x2588b95 is the "ldr r3, [r4, #36]" instruction 2019-10-12T09:18:46 < aandrew> that is the insn after the bl printf 2019-10-12T09:19:07 < aandrew> oh wait that's not fucked 2019-10-12T09:19:14 < aandrew> becuase that was just the most recent call 2019-10-12T09:19:16 < aandrew> that makes sense 2019-10-12T09:19:32 < aandrew> but that means it's something right *after* the printf that failed 2019-10-12T09:21:26 < aandrew> no 2019-10-12T09:21:27 < aandrew> #define DBGV(...) { TimeInternal tmpTime; getTime(&tmpTime); printf("(d %ld.%09ld) ", tmpTime.seconds, tmpTime.nanoseconds); printf(__VA_ARGS__); } 2019-10-12T09:21:42 < aandrew> the first printf() worked, it's the printf(__VA_ARGS__) that is crashing 2019-10-12T09:22:18 < aandrew> so one of those pointers is goofy 2019-10-12T09:22:22 < aandrew> but only sometimes 2019-10-12T09:22:27 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-12T09:24:34 < aandrew> oh wait 2019-10-12T09:24:37 < aandrew> I think I fucked this up 2019-10-12T09:24:48 < aandrew> I think I call that function with one of the parameters NULL as a hack 2019-10-12T09:24:51 < aandrew> fuck a duck 2019-10-12T09:25:35 < aandrew> updateDelay(ptpClock, &ptpClock->timestamp_delayReqSend, &ptpClock->timestamp_delayReqRecieve, NULL); 2019-10-12T09:25:38 < aandrew> I'm a fucking loser 2019-10-12T09:27:40 < aandrew> don't y'all jump in to console me at the same time... 2019-10-12T09:27:42 < aandrew> fuckers. :-) 2019-10-12T09:31:37 < jadew> if you used c++, you could have avoided that 2019-10-12T09:31:48 < aandrew> yeah but then I'd be completely insane 2019-10-12T09:32:11 < jadew> with variadic templates, you can have type checked printf 2019-10-12T09:32:24 < aandrew> *shrugs* 2019-10-12T09:32:28 < jadew> it's actually one of the first examples that were given for their use 2019-10-12T09:32:33 < aandrew> this is specifically why I enabled the MPU 2019-10-12T09:32:38 < aandrew> to catch null pointer derefs 2019-10-12T09:34:25 < aandrew> https://i.imgur.com/0Dk2Cz4.jpg 2019-10-12T09:34:27 < aandrew> not bad 2019-10-12T09:34:52 < jadew> what are you measuring there? jitter? 2019-10-12T09:35:16 < aandrew> looking at a little more than 250ns of jitter between two nodes synchronized to the same ptpd grandmaster 2019-10-12T09:35:45 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-12T09:54:40 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-12T09:55:59 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 276 seconds] 2019-10-12T09:56:17 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-12T09:57:04 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-12T10:23:37 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-12T10:47:29 -!- User_ [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-12T10:47:56 -!- User_ [~learningc@121.121.98.53] has joined ##stm32 2019-10-12T11:00:04 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 268 seconds] 2019-10-12T11:06:39 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-12T11:11:38 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-12T11:13:17 < jadew> this youtube comment made me feel a little bad: "plot twist : this guy was really alive in the 80s because he's actually a vampire." 2019-10-12T11:51:57 < jadew> aandrew, if you decide to see edge of tomorrow with any of the kids, be warned that there's a scene with a melting face 2019-10-12T11:52:03 < jadew> I didn't let my son see that 2019-10-12T11:52:40 < jadew> there's more questionable stuff there, so you should see it by yourself first 2019-10-12T11:53:31 < PaulFertser> How about the whole Terminator melting, did you also leave that out? 2019-10-12T11:53:48 < jadew> you mean in the terminator movie? 2019-10-12T11:53:52 < PaulFertser> Yes, the second. 2019-10-12T11:54:43 < jadew> that's a little different, but that's a good idea for movie night 2019-10-12T12:01:48 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-12T12:05:23 < jadew> "By 2010, the franchise has generated $3 billion in revenue." 2019-10-12T12:07:12 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-12T12:07:48 < PaulFertser> I remember Terminator 2 has some scenes with riding motor bikes without a helmet. Probably those should be censored as well. 2019-10-12T12:08:01 < jadew> hehe 2019-10-12T12:09:08 < PaulFertser> I watched "The Autopsy of Jane Doe" yesterday. That I wouldn't recommend seeing with the kids. 2019-10-12T12:09:27 < jadew> I wouldn't recommend seeing that alone, at night 2019-10-12T12:11:25 < jadew> but it was good 2019-10-12T12:54:59 -!- User_ [~learningc@121.121.98.53] has quit [Remote host closed the connection] 2019-10-12T13:04:26 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-12T13:27:27 -!- learningc [~pi@121.121.98.53] has quit [Ping timeout: 250 seconds] 2019-10-12T13:41:37 -!- learningc [~pi@121.121.98.53] has joined ##stm32 2019-10-12T15:01:10 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-12T15:29:38 < kakimir32> anyone have nice freertos-mpu linker script? 2019-10-12T15:40:50 < kakimir32> I just need inspiration for my own 2019-10-12T15:41:18 < kakimir32> I have done privileged_functions but not yet privileged_data 2019-10-12T15:44:25 < kakimir32> actually I'm doing it with altenated freemarker templates because why do it easy 2019-10-12T15:45:55 < kakimir32> when you can do it right 2019-10-12T15:47:12 < kakimir32> if I have 3 x 4kilobytes ram blocks 2019-10-12T15:48:01 < kakimir32> would it be even overkill to have like one of them blocks for heap only 2019-10-12T15:50:47 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-12T16:08:38 -!- Laurenceb [~laurence@140.141.208.46.dyn.plus.net] has joined ##stm32 2019-10-12T16:09:28 < Laurenceb> https://hardware.slashdot.org/comments.pl?sid=14994098&cid=59298556 2019-10-12T16:13:10 < Laurenceb> emdrive is from the future man 2019-10-12T16:24:41 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-12T16:47:45 -!- Jan- [~IceChat9@87.114.135.182] has joined ##stm32 2019-10-12T16:47:53 < Jan-> crikey there's an actual stm32 channel 2019-10-12T16:48:01 < Jan-> hihi stm32 people 2019-10-12T16:48:10 < Jan-> I'm having my first stm32 adventure having moved up from arduino :D 2019-10-12T16:51:59 -!- Jan- [~IceChat9@87.114.135.182] has quit [Read error: Connection reset by peer] 2019-10-12T17:00:38 < Laurenceb> well he didnt last mong 2019-10-12T17:02:39 < Laurenceb> https://endchan.net/.media/t_65e4824cfea6c9e16a9a15c70d7d935e-imagejpeg 2019-10-12T17:06:07 -!- mode/##stm32 [+o englishman] by ChanServ 2019-10-12T17:06:08 -!- mode/##stm32 [+b *!*@140.141.208.46.dyn.plus.net] by englishman 2019-10-12T17:06:08 -!- Laurenceb was kicked from ##stm32 by englishman [Laurenceb] 2019-10-12T17:06:30 -!- superbia [~user@unaffiliated/superbia] has joined ##stm32 2019-10-12T17:06:35 < karlp> thanks. 2019-10-12T17:06:39 < superbia> hallo 2019-10-12T17:20:37 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-12T17:28:04 < kakimir32> should I open lurencelink? 2019-10-12T17:28:56 < kakimir32> switch to desktop> 2019-10-12T17:31:18 < kakimir64> is there a way to one click refresh all build related files in eclipse cdt 2019-10-12T17:31:57 < kakimir64> when I do linker script templates I need to go into symbols and add or remove symbol POOPOO and save config 2019-10-12T17:32:08 < kakimir64> in order to create new .ld 2019-10-12T17:32:49 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-12T17:35:17 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-12T17:46:29 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-12T17:48:55 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Client Quit] 2019-10-12T17:54:13 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-12T17:59:46 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-12T18:04:50 < kakimir64> 3 ram blocks make linker script interesting 2019-10-12T18:05:05 < bitmask> with switching regulators do you use electrolytic or tantalum 2019-10-12T18:06:01 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Ping timeout: 265 seconds] 2019-10-12T18:06:12 < kakimir64> mlcc 2019-10-12T18:06:35 < kakimir64> and electrolytic 2019-10-12T18:06:38 < bitmask> ok 2019-10-12T18:07:43 < bitmask> almost got all the component footprints/schematic symbols done, if only I had access to the vault 2019-10-12T18:08:18 < bitmask> still gotta choose inductors 2019-10-12T18:14:03 -!- Jan- [~IceChat9@87.114.135.182] has joined ##stm32 2019-10-12T18:14:08 < Jan-> hihi 2019-10-12T18:14:47 < Jan-> I have here a black circuit board that is about three by two and a half inches marked STM32F4XX and STM32_F4VE. In platformio we see an entry for board type called "Black STM32F406VE." Is that the right one? 2019-10-12T18:17:51 < superbia> Jan-: who is "we" 2019-10-12T18:25:55 < kakimir64> I doubt that many people here use platformIO 2019-10-12T18:26:13 < kakimir64> never heard a mention about it 2019-10-12T18:27:33 < kakimir64> see the config files of that Black STM32F406VE 2019-10-12T18:27:39 < kakimir64> and if those match to what you have 2019-10-12T18:28:47 <@englishman> lol very descriptive 2019-10-12T18:29:47 < kakimir64> tool is open source so I'd imagine finding some nice cleartext confs 2019-10-12T18:29:54 < Jan-> Someone told me platformio was good but I'm not enjoying it much 2019-10-12T18:30:17 < Jan-> I'll happily take any other recommendations (though I do like the way visual studio works with intellisense and so on) 2019-10-12T18:31:16 < kakimir64> visual studio embedded? 2019-10-12T18:31:19 <@englishman> does visualgdb work with ozone? 2019-10-12T18:31:32 <@englishman> idk why it wouldnt 2019-10-12T18:31:50 <@englishman> maybe time to ditch keil, just keep the compiler 2019-10-12T18:32:33 < kakimir64> you move to visual studio embedded? 2019-10-12T18:32:46 < Jan-> are you talking to me 2019-10-12T18:32:52 < kakimir64> englishman 2019-10-12T18:33:24 < Jan-> oh 2019-10-12T18:33:30 <@englishman> i tried visualgdb for like 5 minutes it's still pretty raw but worked 2019-10-12T18:33:50 <@englishman> it didn't install at first cuz it was missing some pack but just errored out instead of telling me 2019-10-12T18:34:34 < kakimir64> Jan-: have you tried ST's studio and cubemx? 2019-10-12T18:35:52 < roomba> cubemx + vim + libopencm3 = heaven 2019-10-12T18:37:27 < roomba> cubemx more of a parts search tool and exploring the pinouts/clock config/functions 2019-10-12T18:37:27 < Jan-> I don't know what almost any of those words mean 2019-10-12T18:38:16 < superbia> without vim 2019-10-12T18:38:22 < superbia> and without libopencm3 2019-10-12T18:38:26 < superbia> and without cubeshit 2019-10-12T18:39:04 < Jan-> I haven't tried anything 2019-10-12T18:39:11 < Jan-> we're literally plugging it in for the first time ever 2019-10-12T18:39:47 < kakimir64> well cubemx you need to check 2019-10-12T18:39:51 < roomba> is that one of those stm32duino blue pill boards? 2019-10-12T18:40:12 < Jan-> we have this https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-12T18:41:09 < Jan-> and we have an st-link connected to the correct pins on the connector on the end 2019-10-12T18:41:49 < Jan-> do we have to supply power to the usb socket AND to the stlink 2019-10-12T18:42:51 < BrainDamage> if you connected vdd target in the swd, then the stlink can power the target board 2019-10-12T18:43:13 < BrainDamage> otherwise you have to power them separately 2019-10-12T18:43:13 < Jan-> for that I guess we would have to hook up the power lines from the st-link 2019-10-12T18:43:23 < Jan-> and I'm not sure how to do that 2019-10-12T18:45:01 < Jan-> I mean the st-link dongle has 5v and 3.3v outputs marked but there's no indication of where to connect them on the board. 2019-10-12T18:51:54 < kakimir64> https://www.st.com/content/ccc/resource/technical/document/user_manual/98/2e/fa/4b/e0/82/43/b7/DM00105823.pdf/files/DM00105823.pdf/jcr:content/translations/en.DM00105823.pdf 2019-10-12T18:52:21 < Jan-> is that aimed at me 2019-10-12T18:53:03 < kakimir64> yes 2019-10-12T18:53:05 < kakimir64> page 20 2019-10-12T18:53:22 < kakimir64> maybe save the pdf 2019-10-12T18:53:49 < Jan-> that's about the nucleo board, right 2019-10-12T18:54:42 < kakimir64> oh it's not the same board 2019-10-12T18:54:46 < Jan-> no. 2019-10-12T18:55:03 < kakimir64> anyway the logic is the same - get pdfs 2019-10-12T18:55:05 < Jan-> I am increasingly aware that I should have got a nucleo board because that's the one that everyone seems to understand. 2019-10-12T18:55:17 < Jan-> But someone said "just get this one it's cheap" so I did :( 2019-10-12T18:56:06 < Jan-> we have the pinout but it doesn't say if any of the power pins can be inputs. 2019-10-12T18:57:19 < kakimir64> do you have schematic? 2019-10-12T18:57:36 < Jan-> this is what we have: https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/ 2019-10-12T18:57:43 < BrainDamage> nucleo and the discovery boards are excellent for the price 2019-10-12T18:57:45 -!- superbia1 [~user@unaffiliated/superbia] has joined ##stm32 2019-10-12T18:57:56 < BrainDamage> a couple of euro difference is not worth the hassle 2019-10-12T18:58:06 < BrainDamage> and they come with the programmer too 2019-10-12T18:58:13 < Jan-> yeah well. 2019-10-12T18:58:17 < kakimir64> and code examples that work 2019-10-12T18:58:27 < Jan-> I did question this guy as I know he's quite open source happy and that tends to mean he is OK with lots of screwing around. 2019-10-12T18:58:29 < kakimir64> Schematic link returns 404 2019-10-12T18:58:31 < Jan-> But he promised. 2019-10-12T18:58:39 < Jan-> yeah I noticed that kakimir32. 2019-10-12T18:58:43 < Jan-> welcome to foss world :( 2019-10-12T18:58:53 < kakimir64> not good 2019-10-12T18:59:05 < kakimir64> for development board 2019-10-12T18:59:11 < BrainDamage> there's more open source support for discovery than weird boards ... 2019-10-12T19:00:06 < Jan-> is this helpful https://rusefi.com/forum/viewtopic.php?f=4&t=1489#p31542 2019-10-12T19:00:51 < BrainDamage> yes, it means it's a standard 20 pin jtag/swd interface 2019-10-12T19:01:07 -!- superbia [~user@unaffiliated/superbia] has quit [Ping timeout: 265 seconds] 2019-10-12T19:01:26 < Jan-> is that good? 2019-10-12T19:01:27 < Cracki> I have that board too. it 1:1 fits jlink v8 (china clone) 2019-10-12T19:01:29 < Jan-> can I power it from that? 2019-10-12T19:01:34 < BrainDamage> yes and yes 2019-10-12T19:01:55 < Jan-> so the VCC on pin one there is an input? 2019-10-12T19:02:04 < Jan-> is that 3.3 or 5 2019-10-12T19:02:04 < BrainDamage> yes 2019-10-12T19:02:13 < BrainDamage> 3.3 2019-10-12T19:02:28 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-12T19:02:40 < Jan-> so there's no weirdness with feeding back into the output of some 3.3v regulator? 2019-10-12T19:02:41 < BrainDamage> stm32 are 3.3V devices with some pins 5V tolerant 2019-10-12T19:02:55 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-12T19:03:00 < BrainDamage> there might be a jumper you'll have to disconnect 2019-10-12T19:03:15 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-12T19:03:19 < BrainDamage> to disconnect the regulator 2019-10-12T19:03:19 < Cracki> grab a bench supply, feed 3.3v into it, see if anything starts smoking 2019-10-12T19:03:33 < BrainDamage> they might also have a simple diode to prevent that 2019-10-12T19:03:46 < BrainDamage> you need the schematic to know how safe it is 2019-10-12T19:03:54 < Jan-> ah crap 2019-10-12T19:04:02 < Jan-> remind me to yell at the asshat who told me this was a very standard thing 2019-10-12T19:04:04 < Cracki> it's a simple 1117 on the board. I see a Diode D1 right next to it. 2019-10-12T19:04:20 < Cracki> the asshat is open source friendly, of course he loves to fuck around 2019-10-12T19:04:30 < Cracki> don't worry tho, this is workable 2019-10-12T19:04:46 < Jan-> I'm a windows person 2019-10-12T19:04:48 < kakimir64> well how complicated simple mcu devboard can be 2019-10-12T19:04:49 < Jan-> I'm results oriented 2019-10-12T19:05:09 < BrainDamage> there's open source rants here daily, you'll fit in 2019-10-12T19:05:19 <@englishman> you'll fit right in 2019-10-12T19:05:23 <@englishman> welcome to the pro-est channel 2019-10-12T19:05:25 < Cracki> oh wait that's not "D1", it's an "R1" with its legs sawed off 2019-10-12T19:05:53 < Jan-> so there's an ams1117 on the board 2019-10-12T19:06:08 < Jan-> its centre pin is connected to the 3.3v pin on the 20 pin connector 2019-10-12T19:06:19 <@englishman> if you are results oriented tho, why not buy a $10 stm32 nucleo devboard with onboard debugger and, you know, documentation 2019-10-12T19:06:20 < Jan-> so we don't reeeeally want to stick 3.3v into that, do we? 2019-10-12T19:06:32 < Jan-> englishman: I took advice from the wrong guy what can I say. 2019-10-12T19:06:42 <@englishman> ah, lesson learned then 2019-10-12T19:06:50 < Jan-> well not really how was I supposed to know :/ 2019-10-12T19:07:07 <@englishman> what do you want to do just dick around with stm32 make a led blink? 2019-10-12T19:07:18 < Jan-> I ran out of performance in an AVR project. 2019-10-12T19:07:35 < Jan-> So I said "what's bigger than an atmega328p, goes way faster, does more math, and is more whizzy." 2019-10-12T19:07:41 < BrainDamage> is 10 bucks really that much an investiment though? 2019-10-12T19:07:43 < Cracki> if you don't feel comfortable supplying 3.3v over the debug header, only connect ground there, and give the board power through the mini usb 2019-10-12T19:07:54 < Jan-> Cracki: yeah. for that we really need another usb isolator. 2019-10-12T19:08:02 < Cracki> you don't have two usb ports? 2019-10-12T19:08:05 < Jan-> BrainDamage I didn't even know it existed. 2019-10-12T19:08:17 < Jan-> Cracki I do but I prefer not to connect tiny helpless dev board to huge pc power supply directly. 2019-10-12T19:08:28 < Jan-> that is bad 2019-10-12T19:08:34 < Jan-> that way leads fire and dead pcs 2019-10-12T19:08:38 < Cracki> it's not getting more than 100 ma anyway, and those 5v go through the 1117 2019-10-12T19:08:47 < BrainDamage> the regulator protects both ways 2019-10-12T19:08:49 < Cracki> maybe get a nucleo 2019-10-12T19:08:56 < BrainDamage> and usb ports have inbuilt protection too 2019-10-12T19:09:18 < Jan-> yeah and sometimes usb port power fuses are not resettable 2019-10-12T19:09:18 < BrainDamage> you can short a usb port with no consequence, there's power supply negotiation and current limiting 2019-10-12T19:09:36 < kakimir64> implementation might vary 2019-10-12T19:09:51 < Jan-> Can we just solder a wire onto the input side of the 1117 2019-10-12T19:09:53 < Jan-> is that a solution 2019-10-12T19:10:15 < BrainDamage> just get a wall wart usb charger 2019-10-12T19:10:26 < BrainDamage> no soldering necessary 2019-10-12T19:10:46 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-12T19:10:48 < Jan-> oh hey well 2019-10-12T19:10:58 < BrainDamage> communication to the target board will happen through the programming header 2019-10-12T19:11:04 < Jan-> one of the pins on the 1117 is connected directly to the 5V pins on the board 2019-10-12T19:11:11 < Jan-> according to multimeter continuity check 2019-10-12T19:12:03 < Cracki> I just gave my board 3.3v through the header. it's ok, drawing around 60-70 mA 2019-10-12T19:12:07 < BrainDamage> i do wholeheartly recommend you grab the schematic, one way or another 2019-10-12T19:12:18 < Jan-> dude if we can find a schematic great 2019-10-12T19:12:26 < BrainDamage> because you'll have several questions in the future too 2019-10-12T19:12:35 < Jan-> no kidding 2019-10-12T19:12:46 < Jan-> we have a pinout but not schematic 2019-10-12T19:13:22 < Cracki> "stm32_F4VE" on the back side 2019-10-12T19:13:30 < Jan-> that's the one 2019-10-12T19:13:39 < Cracki> random result: https://wikipeida.deadbsd.org/view/STM32F407VET6 2019-10-12T19:13:43 -!- superbia1 [~user@unaffiliated/superbia] has left ##stm32 ["WeeChat 2.4"] 2019-10-12T19:13:55 < Cracki> links a schematic at the bottom 2019-10-12T19:14:26 < Cracki> ok their idea of "schematic" is connectors and ICs 2019-10-12T19:15:17 < Jan-> ok so the input side of the 1117 is connected to the bus called 5V 2019-10-12T19:15:21 < Jan-> and there's nothing else called 5V 2019-10-12T19:15:45 < Jan-> so can we assume those 5V pins that seem connected to it directly... are? 2019-10-12T19:15:57 < BrainDamage> yes, same name = same wire 2019-10-12T19:16:10 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-12T19:16:20 < Jan-> So we can treat the 5V pins as an input. 2019-10-12T19:16:24 < BrainDamage> yes 2019-10-12T19:16:38 < Jan-> kickass 2019-10-12T19:16:41 * Jan- waves her arms 2019-10-12T19:16:50 < BrainDamage> you can also see the usb port 2019-10-12T19:17:13 < BrainDamage> it's connected to the 5V rail through R25, which is marked 500mA 2019-10-12T19:17:57 < Jan-> a 500 milliamp resistor? 2019-10-12T19:18:00 * Jan- scratches her head 2019-10-12T19:18:04 < Cracki> I'm skimming the ams1117 sheet, haven't seen anything talking about supplying output while leaving input floating 2019-10-12T19:18:06 < BrainDamage> chinese schematic 2019-10-12T19:18:10 < BrainDamage> if you check the datashit for the 1117 regulator, you should be able to 2019-10-12T19:18:16 < BrainDamage> what he said 2019-10-12T19:18:29 < Cracki> that being neither positive nor negative 2019-10-12T19:18:32 < Jan-> we're about to plug in with the 5v output from the st-link connected to one of the 5v pins 2019-10-12T19:18:35 < Jan-> wish us luck 2019-10-12T19:18:40 < karlp> Jan-: this board? https://github.com/libopencm3/libopencm3-miniblink/commit/0b85958064ab35e402dc64b68830cb2dcb5673ab 2019-10-12T19:18:42 < BrainDamage> that'll work fine 2019-10-12T19:20:12 <@englishman> wtf 2019-10-12T19:20:20 <@englishman> why did google remove the fucking url from search results 2019-10-12T19:20:21 < Jan-> we have a heartbeat! 2019-10-12T19:20:22 <@englishman> holy FUCK 2019-10-12T19:20:44 < Jan-> an LED is flashing 2019-10-12T19:20:50 < Jan-> is this the stm32 equivalent of "blink" 2019-10-12T19:20:50 <@englishman> 100% switching to cuckcuckgo 2019-10-12T19:21:20 < kakimir64> is there viewers for .map file? 2019-10-12T19:21:27 < Jan-> OK so now what's the least possible software I can install to flash something else into it 2019-10-12T19:21:28 < Jan-> to prove we can 2019-10-12T19:21:38 < kakimir64> make the bloats of data more readable 2019-10-12T19:22:05 <@englishman> you can download st-link utility 2019-10-12T19:22:15 <@englishman> which is the flash and option byte tool 2019-10-12T19:22:35 < Cracki> what something else would you like 2019-10-12T19:22:52 < Cracki> I'm sure there are ready binaries/elfs for that chip or even that board 2019-10-12T19:23:48 < Jan-> I uhoh 2019-10-12T19:23:48 < Jan-> er 2019-10-12T19:24:01 < Jan-> on avr I just tend to make it flash faster 2019-10-12T19:24:04 < karlp> Thorn: what was with the three smps links you pasted last night, I never saw any context. 2019-10-12T19:24:12 < Cracki> on avr you have an IDE or arduino 2019-10-12T19:24:17 < Jan-> exactly 2019-10-12T19:24:26 < Jan-> I tend to use the arduino ide (but not the library) 2019-10-12T19:24:34 < Jan-> presumably there is something similar for this 2019-10-12T19:24:35 < Cracki> that stm32f4 is probably supported by stm32 for arduino, and maybe platformio 2019-10-12T19:24:45 < karlp> heh, library is way better than the ide IMO :) 2019-10-12T19:24:56 < Jan-> as a priority I would like to get some serial communications working too. 2019-10-12T19:25:02 < Jan-> so I can send debug etc 2019-10-12T19:25:08 < BrainDamage> there's 2 leds, PA6 and PA7 2019-10-12T19:25:16 < Cracki> as a priority you should pick something that's not arduino 2019-10-12T19:25:20 < BrainDamage> have them blink alternately and you'll know 2019-10-12T19:25:45 < Jan-> well right now I can't do anything as I don't have any way of writing, compiling or uploading code. 2019-10-12T19:25:49 < Cracki> stm32 have uarts. the usb is complex, you'll want the necessary code *generated* by things like cubemx 2019-10-12T19:26:00 < Jan-> yeah I don't need usb 2019-10-12T19:26:08 < Cracki> replicate a blinky maybe 2019-10-12T19:26:09 < Jan-> maybe in the future I guess there's a library 2019-10-12T19:26:47 < Cracki> well, try getting platformio or arduino for stm32 to workw ith it 2019-10-12T19:26:57 < Jan-> I got platformio and installed the thing 2019-10-12T19:27:03 < Cracki> you seem familiar with that 2019-10-12T19:27:06 < Jan-> I'm not 2019-10-12T19:27:09 < Jan-> First time today 2019-10-12T19:27:12 < Cracki> what are you familiar with 2019-10-12T19:27:13 < Jan-> it seems horrible to be honest 2019-10-12T19:27:16 < BrainDamage> you can also use semihosting on the debugger to get data from/to the board 2019-10-12T19:27:17 < Cracki> then discard it 2019-10-12T19:27:20 < Jan-> er, arduino and notepad++ :) 2019-10-12T19:27:26 < Cracki> then take arduino 2019-10-12T19:27:39 < Jan-> not sure how that works, it would need to be a different compiler? 2019-10-12T19:27:41 < Cracki> "stm32duino" 2019-10-12T19:27:48 < Cracki> it's a board support thingy 2019-10-12T19:27:50 < Jan-> also I went to the stm32 for more performance 2019-10-12T19:27:56 < Jan-> not so I could clog it up with some huge ugly library 2019-10-12T19:27:58 < Cracki> also pulls in the right compiler 2019-10-12T19:27:58 < BrainDamage> yes, you need a compiler that can target a stm32 2019-10-12T19:28:33 < Cracki> stm32 is a much more complex beast compared to avr 2019-10-12T19:28:41 < Jan-> I'm aware. 2019-10-12T19:28:46 < Cracki> you can poke registers, sure 2019-10-12T19:29:03 < Jan-> I looked into this. 2019-10-12T19:29:10 < Jan-> mostly people seem to use a hardware abstraction layer 2019-10-12T19:29:14 < Cracki> even arduino does a bunch of setup you don't see 2019-10-12T19:29:17 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-12T19:29:26 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Ping timeout: 268 seconds] 2019-10-12T19:29:29 < Jan-> well it actually doesn't as I tend to write my own main() but I know what you mean :) 2019-10-12T19:29:34 < Cracki> and stm32 need a bit more setup than avr... look up "stm32 clock tree" to get an idea 2019-10-12T19:29:43 < Cracki> yes, HAL is what ST supplies 2019-10-12T19:29:43 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-12T19:29:48 < Cracki> along with tools such as cubemx 2019-10-12T19:29:59 < Cracki> there is stuff before the main() 2019-10-12T19:30:14 < Cracki> main() is run after vital things such as the clock tree are set up 2019-10-12T19:30:24 < Jan-> Oh sure but that's more compiler magic than arduino 2019-10-12T19:30:28 < Jan-> I don't know how much they futz with it 2019-10-12T19:30:40 < BrainDamage> one thing you'll have to drill yourself in mind is that stm32 everything is off by default, unlike avr 2019-10-12T19:30:45 < Cracki> stm32duino will allow you to use arduino to program this F4 board 2019-10-12T19:30:46 < BrainDamage> no clock provided, no worky 2019-10-12T19:31:00 < BrainDamage> so prepare to stare at the clock tree a lot 2019-10-12T19:31:16 < Jan-> yeah well I guess I'll just have to figure it out 2019-10-12T19:31:40 < Cracki> if you get fed up with stm32duino, take a look at stm32cubeIDE 2019-10-12T19:31:42 < Jan-> I just wanted something with an FPU :/ 2019-10-12T19:32:03 < Cracki> eclipse isn't the nicest there is, but it's tolerable and the official tool by ST 2019-10-12T19:33:37 < Jan-> we're playing with platformio right now and can't even figure out how to get to the point of having a code file we can type "main" into 2019-10-12T19:35:04 < mawk> you finally did it Jan- 2019-10-12T19:35:07 < mawk> you quit the avr stuff 2019-10-12T19:35:48 < Jan-> not yet I can't even code for this stm32 2019-10-12T19:36:49 < mawk> you used "platformio" you choose the black magic yourself 2019-10-12T19:36:58 < Jan-> I uhoh about "used" 2019-10-12T19:37:00 < Jan-> we installed it 2019-10-12T19:37:01 < mawk> if you used cube you'd have a bare compiler to compile stuff 2019-10-12T19:37:01 < mawk> ah 2019-10-12T19:37:07 < mawk> go on with it then 2019-10-12T19:37:10 < Jan-> it is complicated to the point of being sort of hilarious 2019-10-12T19:37:25 < mawk> you don't even need cube, you just need a compiler that's all 2019-10-12T19:37:29 -!- jon101249 [~jon1012@pha75-18-88-178-181-173.fbx.proxad.net] has joined ##stm32 2019-10-12T19:37:50 < mawk> all the rest is icing on the cake 2019-10-12T19:37:57 < Jan-> doesn't st have something 2019-10-12T19:37:59 < Cracki> just get cubeide, it contains everything 2019-10-12T19:38:28 < mawk> Jan-: you use the compiler from ARM, and then you use the set of libs you want, there is libopencm3, or there is the st stuff (stm32cube) 2019-10-12T19:38:36 < mawk> if you want all in one package use what Cracki said 2019-10-12T19:38:54 < Cracki> arm gcc isn't from arm, armcc is from arm, just saying :P 2019-10-12T19:39:20 < mawk> lol 2019-10-12T19:39:25 < mawk> yeah I mean I download it from the ARM website 2019-10-12T19:39:32 < BrainDamage> 3 things you need: compiler, debugger/programmer, and a hardware library 2019-10-12T19:39:41 < mawk> https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads 2019-10-12T19:39:48 < Cracki> one thing to notice: everyone here is crazy about open source 2019-10-12T19:39:53 < BrainDamage> some things give you all 3 in a single package, but you're free to mix and match 2019-10-12T19:40:02 < Cracki> that's why they all recommend that you buy your wheels and your engine and your chassis separately 2019-10-12T19:40:06 < mawk> lol 2019-10-12T19:40:10 < mawk> cubeide works fine 2019-10-12T19:40:19 < Cracki> there is no reason for a newbie to do that. get cubeide. 2019-10-12T19:40:32 < Jan-> this is one of those nerd religion discussions isn't it 2019-10-12T19:40:39 < BrainDamage> it is 2019-10-12T19:40:54 < mawk> get cubeide if you want to be handguided Jan- 2019-10-12T19:40:58 * Jan- holds her head in her hands 2019-10-12T19:41:02 < mawk> it's just click click click 2019-10-12T19:41:05 < Jan-> auurrrngh 2019-10-12T19:41:14 * Jan- repeats mantra under breath 2019-10-12T19:41:15 < mawk> it packages the programmer software, the compiler, the software library, the code generator for your target board 2019-10-12T19:41:16 < Cracki> cubeide needs little handholding. all the other stuff needs messing with so much... 2019-10-12T19:41:20 < Jan-> you love nerds. you love nerds. you love nerds. 2019-10-12T19:41:45 < mawk> it even packages a RTOS should you need one 2019-10-12T19:41:52 < Cracki> ignore that 2019-10-12T19:41:52 < BrainDamage> stm32 has a lot of options for development, this can be overwhelming if you're not sure what you want 2019-10-12T19:41:54 < Jan-> I did wonder about that 2019-10-12T19:41:57 < mawk> and all sort of libs like for touchscreen or bluetooth or anything 2019-10-12T19:42:01 < Jan-> I was thinking, 168mhz this thing is faster than an Amiga 2019-10-12T19:42:12 < BrainDamage> just go for a complete package and switch to another package if it doesn't satisfy you 2019-10-12T19:42:31 < BrainDamage> use a custom setup only after you tried everything and want a personal touch 2019-10-12T19:42:31 < mawk> Jan-: https://www.st.com/en/development-tools/stm32cubeide.html 2019-10-12T19:42:42 < Jan-> I'm downloading it 2019-10-12T19:42:44 < mawk> st is one of these sites for which you need to register to download, they're stuck in 1990 2019-10-12T19:42:45 < Jan-> we had to register and everything 2019-10-12T19:42:46 < mawk> yeah 2019-10-12T19:43:03 < Jan-> yeah they wouldn't want it to be easy for people to develop for their parts would they. 2019-10-12T19:43:05 < Jan-> who'd want that 2019-10-12T19:43:10 < Jan-> newfangled whippersnappers 2019-10-12T19:43:18 < mawk> through registration they sent me invites for a workshop 2019-10-12T19:43:24 < mawk> if I didn't forget to go there it would've been cool 2019-10-12T19:43:59 < Jan-> jesus it's nearly a 2gb install 2019-10-12T19:44:07 < mawk> lol 2019-10-12T19:44:11 < mawk> you wanted the whole thing 2019-10-12T19:44:14 < Jan-> the fuck st 2019-10-12T19:44:17 < Jan-> arduino is like 650k 2019-10-12T19:44:20 < mawk> eclipse is heavy 2019-10-12T19:44:32 < Cracki> arduino isn't 650k. that's just the installer. 2019-10-12T19:44:35 < BrainDamage> the compiler is several tens/hundred mb 2019-10-12T19:44:40 < BrainDamage> this for arduino too 2019-10-12T19:44:43 < Cracki> the toolchain it pulls in is orders of magnitude more 2019-10-12T19:45:04 < mawk> the toolchain is 500MiB uncompressed, then the peripheral libs are pretty heavy too 2019-10-12T19:45:13 < mawk> but it's not packaged, you download them after lol 2019-10-12T19:45:24 < kakimir64> https://raw.githubusercontent.com/jpgvandijk/linker-map-viewer/master/examples/example.png anyone know tool like this that works? 2019-10-12T19:45:30 < Cracki> mind you, cubeIDE isn't the pinnacle of user friendliness, but it keeps you from having to fuck around with command line bullshit 2019-10-12T19:46:04 < Jan-> I don't mind a bit of command line 2019-10-12T19:46:17 < Jan-> but honestly with most open source stuff it's just laziness that they can't be bothered to write a UI 2019-10-12T19:46:35 < BrainDamage> that's correct 2019-10-12T19:46:42 < kakimir64> are you planning to use rtos Jan-? 2019-10-12T19:46:57 < Jan-> I don't think so. 2019-10-12T19:47:07 < Jan-> OK so the installer said "Extract: error writing to file libstdc++.a" 2019-10-12T19:47:12 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-12T19:47:14 < Cracki> disk full? 2019-10-12T19:47:40 < mawk> lol 2019-10-12T19:47:50 < Cracki> you're on windows, the installer should automatically ask for admin privleges to be able to write to wherever it needs to 2019-10-12T19:47:55 < Jan-> it did 2019-10-12T19:48:02 < mawk> UI isn't magic Jan- , for a certain level of complexity that doesn't cut it 2019-10-12T19:48:11 < mawk> or, rather, the complexity to use the UI overcomes that to use the CLI 2019-10-12T19:48:13 < Cracki> UI gives you options 2019-10-12T19:48:19 < Cracki> such as which fucking stm32 chip you wanna use 2019-10-12T19:48:29 < Cracki> commandline? you need to know the exact argument, and switch, and whatnot 2019-10-12T19:48:34 < mawk> --help 2019-10-12T19:48:50 < Cracki> that's more work than being presented a decent ui 2019-10-12T19:48:55 < mawk> although the fact you need to know the flash start address isn't very cool 2019-10-12T19:48:57 < Cracki> try ozone, that's how it should be 2019-10-12T19:49:28 < Cracki> see, that kinda bs should be assumed by the tool, unless you want something else, which you don't in 99% of cases 2019-10-12T19:49:48 < BrainDamage> both have their own benefits, ui is self documenting without digging into manuals, but command line allows to chain tools together and create automation 2019-10-12T19:50:07 < BrainDamage> so a program ideally should have both 2019-10-12T19:50:31 < BrainDamage> an easy to use frontend, and for all the cases you cannot cram into a gui, a command line interface to every single bit 2019-10-12T19:53:48 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-12T20:13:20 -!- via [~via@vtluug/member/via] has joined ##stm32 2019-10-12T20:14:32 < via> on an stm32f4, each of TIM2's capture channels has a dma1 stream that it can use (5, 6, 1, and 7 respectively), but for some reason ch4 can also be used on the same stream as ch2 (stream 6) 2019-10-12T20:15:07 < via> as far as i can tell, this makes it impossible to use all 4 capture channels with dma, since ch4's dma's will go to ch2/stream6, even if i am trying to use it with stream 7 2019-10-12T20:15:15 < via> am i missing something? i can't imagine why they would do this 2019-10-12T20:18:27 < kakimir64> what are symbols - 2019-10-12T20:18:44 < kakimir64> _data and _edata used for exactly? 2019-10-12T20:22:12 < jpa-> for copying initial values from flash to ram 2019-10-12T20:23:06 < jpa-> via: what are you trying to do? if you just want to update the values of all the CCR channels, you can do that with a single dma stream 2019-10-12T20:23:45 < jpa-> via: but each dma stream has a channel selection, if you select the TIM2CCR2 channel, it won't react to TIM2CCR4 triggers 2019-10-12T20:23:47 < kakimir64> does priviledged_data need to be in that jpa? 2019-10-12T20:23:52 < kakimir64> freertos-mpu 2019-10-12T20:24:36 < jpa-> kakimir64: maybe, or maybe it needs separate initialization - copy from some example 2019-10-12T20:27:24 < via> jpa-: i'm using each channel to trigger a dma from that channel's compare register to a buffer in memory dedicated to that channel 2019-10-12T20:27:46 < kakimir64> some have it in _bss _ebss range 2019-10-12T20:27:58 < kakimir64> some don't 2019-10-12T20:28:41 < kakimir64> https://paste.ee/p/QsTp2 lines 173 and 176 are my problem now 2019-10-12T20:28:48 < qyx> via: are you triggering them at different times? 2019-10-12T20:28:51 < via> jpa-: so i have all four streams enabled, but that seems to cause tim2 ch4's events to use dma1 ch3 stream 6 instead of dma1 ch3 stream 7 (since both are enabled) 2019-10-12T20:28:53 < Steffanx> so how is the kakimir64 nas saga going? 2019-10-12T20:28:54 < via> yes 2019-10-12T20:28:59 < via> they are independent 2019-10-12T20:29:12 < kakimir64> Steffanx: it's off now. need to buy 2 x 2T 2019-10-12T20:29:34 < via> basically, is there any way to force ch4 to only have its events go to stream7, not stream6 2019-10-12T20:29:42 < jpa-> via: ah, so you are trying to use the same dma channel for multiple streams? 2019-10-12T20:29:56 < Thorn> >low power op amp 2019-10-12T20:29:58 < Thorn> >0.5mA typical supply current at 5V 2019-10-12T20:30:02 < Thorn> what 2019-10-12T20:30:02 < jpa-> via: err hm 2019-10-12T20:30:09 < via> jpa-: i mean yeah, but thats not the issue 2019-10-12T20:30:18 < jpa-> yeah, i mixed up channel vs. stream 2019-10-12T20:30:28 < via> yeah, its confusing that the timer ICs are channels 2019-10-12T20:30:32 < qyx> uh is Jan- a true she or I misunderstood something 2019-10-12T20:30:39 < qyx> Thorn: do you need one? 2019-10-12T20:30:42 < Steffanx> also got myself a 9211-8i, kakimir64. That thing gets quite warm 2019-10-12T20:30:54 < kakimir64> it hase heatsinks 2019-10-12T20:30:57 < Thorn> the lowest I found is 60µA. (does this also apply when both inputs are grounded?) 2019-10-12T20:31:08 < kakimir64> Steffanx: from same url that I gave you 2019-10-12T20:31:10 < kakimir64> ? 2019-10-12T20:31:28 < kakimir64> it actually has PPC processor in there running at some 900mhz or so 2019-10-12T20:31:53 < via> if tim2_ch4 only went to one stream, this would work perfectly, but it goes two streams, one of which is for another IC channel on the same timer, it just feels like a silly problem 2019-10-12T20:31:53 < Thorn> qyx: I'm innovating a constant current backlight supply circuit with a buck-boost and op amp to amplify a current shunt 2019-10-12T20:31:57 < Steffanx> no, from a dutch guy 2019-10-12T20:32:08 < kakimir64> it mode firmware? 2019-10-12T20:32:23 < jpa-> via: indeed it's quite weird that there is both TIM2_CH2 and TIM2_CH4 for DMA1_Stream6_Channel3 2019-10-12T20:32:24 < Steffanx> yeah 2019-10-12T20:32:40 < kakimir64> with cables and shiet Steffanx? 2019-10-12T20:32:47 < jpa-> via: for most there is only single peripheral per dma stream/channel combo 2019-10-12T20:32:51 < kakimir64> harnesses almost 2019-10-12T20:32:58 < Steffanx> just 1 cable for now. 2019-10-12T20:33:01 < Steffanx> need more soon 2019-10-12T20:33:11 < via> jpa-: yeah 2019-10-12T20:33:21 < Steffanx> 1x 4x sata that is ofcourse, kakimir64 2019-10-12T20:33:23 < via> so i'm not missing anything, can you tihnk of any way around this? other than using like, tim5 instead 2019-10-12T20:33:31 < kakimir64> yes stef 2019-10-12T20:33:39 < kakimir64> help me with linker script stef 2019-10-12T20:34:05 < Steffanx> Sorry, not in the mood :P 2019-10-12T20:34:27 < jpa-> via: or using interrupts.. hmm 2019-10-12T20:34:52 < via> i'm actually already using interrupts too, but for diffrent reasons 2019-10-12T20:35:31 < via> i just mean, there's no way to disable a peripheral->dma channel linkage is there? 2019-10-12T20:35:51 < jpa-> TIM2->DIER does that, but then it disables the link to both streams 2019-10-12T20:36:00 < via> yeah -_- 2019-10-12T20:36:18 < qyx> Thorn: TLV8811 2019-10-12T20:36:28 < qyx> maybe 2019-10-12T20:37:40 < jpa-> via: i guess you could copy all of the CCR channels in a single dma trigger, but then it will be annoying to try to figure out which channel it really was from 2019-10-12T20:38:20 < via> yeah, i don't think i can do that for this use case 2019-10-12T20:38:30 < via> well bummer, i'll try to see if i can free up tim5 2019-10-12T20:39:29 < Thorn> qyx: 6kHz only, I've no idea if it will keep the feedback loop stable 2019-10-12T20:39:42 < qyx> me neither 2019-10-12T20:39:49 < qyx> but iirc it is low power 2019-10-12T20:40:12 < Thorn> yes very, half a µA 2019-10-12T20:40:13 < jpa-> via: hmm, how are your stream priorities set? if you make stream 7 higher priority, maybe it would catch the TIM2_CH4 trigger before stream 6 has time to acknowledge it? 2019-10-12T20:40:29 < Thorn> I guess I should simply switch the op amp off with a mosfet then 2019-10-12T20:40:52 < via> jpa-: i will try that, though is there any confidence that if that works it works consistently? 2019-10-12T20:41:23 < jpa-> via: yeah, see arbiter chapter in DMA ref man; by default it goes by stream number if all the priorities are equal 2019-10-12T20:41:32 < Jan-> OK so now I have stm32 cube ide 2019-10-12T20:43:38 < Jan-> and it is complex 2019-10-12T20:43:53 < Jan-> "target selection" 2019-10-12T20:43:58 < Jan-> there's a slider for price? 2019-10-12T20:44:01 * Jan- scratches her head 2019-10-12T20:44:55 < Steffanx> time to go back to avr :P 2019-10-12T20:44:58 < via> jpa-: this is bizarre, it seems stream7 gets it now, but doesn't get ack'd so it only works once...? still lookin 2019-10-12T20:45:43 < Jan-> "This kind of project is associated with the STM32cubeMx perspective. Do you want to open this perspective now?" 2019-10-12T20:45:48 < Jan-> errr 2019-10-12T20:45:49 < Jan-> do I? 2019-10-12T20:46:00 < kakimir64> try stuff 2019-10-12T20:46:33 < Jan-> it's downloading stm32cube_fw_f4_v1240.zip which is 667 megabytes. 2019-10-12T20:46:37 < qyx> I would not know too 2019-10-12T20:46:46 < qyx> lol 2019-10-12T20:46:53 -!- fooman2011 [~IceChat9@157.206.119.80.rev.sfr.net] has joined ##stm32 2019-10-12T20:46:57 < fooman2011> Hello. I would like to use this display https://www.alibaba.com/product-detail/3-4-Inch-800x800-Round-TFT_60815628362.html?spm=a2700.7724857.discountZoneStyleB_top.3.64c37504VZYKUa that is sold to be used by a raspberry pi. But I would like to use a microcontroller, is it possible ? 2019-10-12T20:47:00 < fooman2011> the interface is MIPI 3 lane, and a driver MIPI to HDMI can be used too 2019-10-12T20:47:02 < fooman2011> In fact I need a round display of at leat 3.1" that can be used by a micro controler 2019-10-12T20:47:42 < qyx> see STM32 F7/H7, it has MIPI DSI 2019-10-12T20:47:51 < qyx> iirc 2019-10-12T20:48:16 < Jan-> mipi is the cellphone thing, right? 2019-10-12T20:48:17 < fooman2011> qyx: I have read on forum that stm32 mipi dsi is a mipi 2 lane 2019-10-12T20:48:28 < fooman2011> not 3 lane 2019-10-12T20:48:42 < qyx> I am not into lanes much 2019-10-12T20:49:08 < qyx> couldn't you get one with the righ number of lanes? :X 2019-10-12T20:49:12 < fooman2011> (I don't know what is a MIPI lane :p ) 2019-10-12T20:50:27 < Jan-> I think mipi is a cellphone industry thing 2019-10-12T20:51:30 < jpa-> fooman2011: maybe ILI9881C can be configured to use only 2 lanes? 2019-10-12T20:57:15 < jpa-> fooman2011: https://electronics.stackexchange.com/questions/392873/can-i-operate-a-4-lane-mipi-dsi-display-from-just-one-lane-connected suggests that there are converter ICs between different lane counts 2019-10-12T20:59:59 < Jan-> so this "blink" application 2019-10-12T21:00:09 < Jan-> that comes with the cube ide 2019-10-12T21:00:18 < Jan-> it doesn't actually blink an LED because the while loop in main is empty. 2019-10-12T21:00:25 < Jan-> so it's not really a blink application. or am I missing something. 2019-10-12T21:03:18 < mawk> did you try to run it ? 2019-10-12T21:05:06 < Jan-> I don't know how 2019-10-12T21:05:12 < Cracki> you hit "debug" 2019-10-12T21:05:32 < jpa-> fooman2011: looks like ILI9881C has the lane count setting on the IM0/IM1/IM2 pins, but the datasheet doesn't specify how they are connected on that module 2019-10-12T21:05:45 < Cracki> there is no "write to device and let it run" in eclipse because ST haven't figured out how to make it do that 2019-10-12T21:05:56 < Jan-> it wants to update the firmware in the st-link 2019-10-12T21:06:53 < Jan-> ok well 2019-10-12T21:07:01 < Jan-> we updated the st-link and now the LED doesn't flash any longer 2019-10-12T21:07:10 < Cracki> good :P 2019-10-12T21:07:11 < Jan-> so it looks like "blink" is in fact "not blink" 2019-10-12T21:07:16 < Jan-> which is mad 2019-10-12T21:07:36 < fooman2011> jpa-: I'm so noob at this. All I want is a round display larger than 3.1" with a interface that can be used by a microcontroller :/ 2019-10-12T21:08:38 < Cracki> anyway, I don't know what "blinky example" you have there and what gpio it uses, or what gpio you should use that has an led on this board... but to reset a pin you use something like HAL_GPIO_WritePin(GPIOC, GPIO_PIN_13, GPIO_PIN_RESET); 2019-10-12T21:08:53 < Cracki> and then HAL_Delay would be a good idea 2019-10-12T21:09:19 < Cracki> maybe that example uses a timer... but then it would make assumptions about what pin to blink 2019-10-12T21:09:23 < Jan-> jesus that's how you set a pin? 2019-10-12T21:09:34 < jpa-> fooman2011: here would be easy 3 inch one https://www.alibaba.com/product-detail/3-inch-round-LCD-display-round_62193124835.html?spm=a2700.7724857.normalList.2.44383750auHHin&s=p 2019-10-12T21:09:48 < Jan-> it's called "blinky" and it comes with cube ide 2019-10-12T21:10:11 < Cracki> that's not much different from a digitalWrite(whatpin, whatstate) 2019-10-12T21:10:27 < Cracki> stm32 have lots more pins, so you describe them with port (A-Z) and number (0-15) 2019-10-12T21:10:44 < Jan-> I'd usually just PORTD &= 1 << n; 2019-10-12T21:11:12 < Jan-> I appreciate that is actually a macro 2019-10-12T21:11:15 < Jan-> but still 2019-10-12T21:12:32 < Cracki> nah, that's barely a macro. portd is just a memory location 2019-10-12T21:12:35 < Cracki> and the same you can do for stm32 2019-10-12T21:12:50 < Jan-> You'd think but in most avr things if you look it up, "portd" is actually a macro. 2019-10-12T21:12:57 < Cracki> BUT nice descriptively named hal functions should beat cryptic magic numbers and bitshifts 2019-10-12T21:13:04 < jpa-> fooman2011: M0420HAI also if you can get that somewhere 2019-10-12T21:13:04 < Cracki> sure, it's a macro 2019-10-12T21:13:12 < Cracki> a macro defining a location in memory so you can access it 2019-10-12T21:13:46 < Jan-> as is TCCR0A and so on. 2019-10-12T21:14:01 < Cracki> so really just a define 2019-10-12T21:14:05 < Jan-> So there's this huge graphical configurator for clock sources for everything. 2019-10-12T21:14:22 < Jan-> I guess that's actually adding lines to void SystemClock_Config automagically? 2019-10-12T21:15:18 < Cracki> you'll see clock sources like HSE, LSE, HSI, ... those are high/low speed external/internal. your board has an 8 MHz crystal, so you use HSE 2019-10-12T21:16:09 < Cracki> the rest is juggling multiplicators, which cubemx can do halfway successfully 2019-10-12T21:16:26 < Jan-> I had a feeling this might be a bit beyond me 2019-10-12T21:16:27 < Jan-> but ok 2019-10-12T21:16:37 < fooman2011> M0420HAI is very interesting 2019-10-12T21:16:55 < fooman2011> It uses a TTL interface ? 2019-10-12T21:18:17 < jpa-> nothing uses TTL in this day and age 2019-10-12T21:18:41 < jpa-> but it uses 3.3V rgb or spi interface, apparently 2019-10-12T21:22:44 < mawk> it's automatic Jan- , just set the frequencies you want and hit "automatic solver" 2019-10-12T21:22:49 < mawk> and then it will autowrite in the location you said 2019-10-12T21:23:04 < Jan-> I don't think I need to futz with that anyway 2019-10-12T21:23:14 < mawk> well you need a clock 2019-10-12T21:23:18 < mawk> but I guess the default values work 2019-10-12T21:23:19 < Jan-> need to figure out which pins are connected to what here 2019-10-12T21:23:23 < mawk> even though it's not the highest freq 2019-10-12T21:23:25 < Jan-> so we can make it blink the led 2019-10-12T21:23:35 < mawk> that's in the datasheet for your board 2019-10-12T21:23:43 < mawk> but it should be on the silkscreen of the board too I think 2019-10-12T21:23:54 < fooman2011> jpa-: Where you can see that SPI interface can be used ? 2019-10-12T21:25:58 < mawk> but blink should blink Jan- , I don't know what you did 2019-10-12T21:26:01 < mawk> let me try 2019-10-12T21:27:15 < Jan-> PA6 and PA7 apparently 2019-10-12T21:27:17 < mawk> I think I uninstalled it lol 2019-10-12T21:27:19 < mawk> yeah try that 2019-10-12T21:27:21 < mawk> like Cracki said 2019-10-12T21:27:22 < Jan-> traced it out 2019-10-12T21:27:28 < jpa-> fooman2011: pinout 2019-10-12T21:27:30 < Jan-> well I can't see how it can blink 2019-10-12T21:27:37 < Jan-> its while(1) in main is empty 2019-10-12T21:27:45 < mawk> ah 2019-10-12T21:27:48 < mawk> maybe 2019-10-12T21:27:59 < jpa-> fooman2011: but that might be just configuration interface, some controllers can only receive display data by the parallel interface 2019-10-12T21:28:11 < mawk> HAL_GPIO_WritePin(GPIOA, GPIO_Pin_6, GPIO_PIN_SET); HAL_Delay(500); HAL_GPIO_WritePin(GPIOA, GPIO_Pin_6, GPIO_PIN_RESET); 2019-10-12T21:28:14 < mawk> something like this Jan- 2019-10-12T21:28:16 < jpa-> fooman2011: but without datasheet hard to know anything 2019-10-12T21:28:50 < Jan-> ok let's try 2019-10-12T21:28:58 < Jan-> is that PA6 2019-10-12T21:29:08 < mawk> yes 2019-10-12T21:29:11 < mawk> gpio bank A, number 6 2019-10-12T21:29:29 < Jan-> hang on 2019-10-12T21:29:34 < Jan-> it'll need another delay after the reset 2019-10-12T21:29:41 < mawk> ah 2019-10-12T21:29:44 < mawk> yes sure 2019-10-12T21:29:50 < mawk> or use a variable for toggling 2019-10-12T21:30:06 < Jan-> [07:29.50] or use a variable for toggling 2019-10-12T21:30:10 < Jan-> wait what 2019-10-12T21:30:12 < Jan-> sorry 2019-10-12T21:30:19 < Jan-> ../Startup/Src/main.c:98:29: error: 'GPIO_Pin_6' undeclared (first use in this function); did you mean 'GPIO_PIN_6'? 2019-10-12T21:30:34 < mawk> int state = 0; for (;;) { HAL_GPIO_WritePin(GPIOA, GPIO_Pin_6, (state ^= 1)); HAL_Delay(500); } 2019-10-12T21:30:36 < mawk> something like this 2019-10-12T21:30:38 < mawk> yeah use uppercasez 2019-10-12T21:30:46 < mawk> I was writing from memory 2019-10-12T21:31:05 < Jan-> ok so it says "download verified successfully" 2019-10-12T21:31:08 < Jan-> but no blinking 2019-10-12T21:31:28 < mawk> you hit debug ? 2019-10-12T21:31:29 < jpa-> fooman2011: FG040210DSSWMG01 also 2019-10-12T21:31:37 < Jan-> I hit F11 which is debug? 2019-10-12T21:32:06 < mawk> right 2019-10-12T21:32:10 < mawk> is it single stepping maybe ? 2019-10-12T21:32:14 < mawk> you have a breakpoint in Reset by default 2019-10-12T21:32:15 < mawk> hit continue 2019-10-12T21:32:26 < Jan-> er 2019-10-12T21:32:27 < Jan-> where 2019-10-12T21:32:29 < mawk> but you should know how to do that right, it's the same as for any program 2019-10-12T21:33:16 < Jan-> Thread #1 [main] [core:0] (Running) 2019-10-12T21:33:24 < Jan-> does this have relevance 2019-10-12T21:33:54 < mawk> I guess it's running then 2019-10-12T21:34:12 < Jan-> it did say "suspended: breakpoint" to begin with 2019-10-12T21:34:24 < mawk> yeah you have a "resume" button 2019-10-12T21:34:25 < jpa-> fooman2011: this is probably same as M0420HAI and has specs available https://www.usmicroproducts.com/product/USMP-T042-680680NAV-A0 2019-10-12T21:34:26 < Jan-> then I hit F8 for Resume 2019-10-12T21:34:29 < mawk> for when it's in a breakpoint 2019-10-12T21:34:29 < mawk> good 2019-10-12T21:34:39 < Jan-> nice curve ball thanks st 2019-10-12T21:34:45 < Jan-> default is not to run the software 2019-10-12T21:34:49 < Jan-> but still no blinking 2019-10-12T21:35:26 < mawk> let me try the blink project on my cubeIDE 2019-10-12T21:36:36 < fooman2011> jpa- yeah it seems to be 2019-10-12T21:38:01 < Jan-> I put the code in the while(1) loop in main. 2019-10-12T21:39:43 < mawk> you initialized all peripherals in default mode ? 2019-10-12T21:39:46 < mawk> there was a prompt at some point 2019-10-12T21:40:09 < Jan-> um, er 2019-10-12T21:40:14 < Jan-> not that I remember 2019-10-12T21:40:16 < mawk> I have no idea where you found this blinky thing 2019-10-12T21:40:26 < mawk> I just opened cubeide, hit "new STM32 project", that's all 2019-10-12T21:40:27 < Jan-> I think it just popped up by default 2019-10-12T21:40:43 < Jan-> ok if we do that we get the target selector 2019-10-12T21:40:58 < Jan-> select stm32f407 2019-10-12T21:41:03 < Jan-> next 2019-10-12T21:41:19 < Jan-> target language C, executable, project type stm32cube. 2019-10-12T21:41:23 < mawk> it's C++ 2019-10-12T21:41:27 < mawk> well, C works too 2019-10-12T21:41:38 < mawk> but if you want to be part of the 21th century you should select C++ 2019-10-12T21:41:41 < mawk> otherwise you go to hell 2019-10-12T21:41:46 < Jan-> "This kind of project is associated with the STM32CubeMx perspective. Do you want to open this perspective now?" 2019-10-12T21:41:50 < mawk> yeah 2019-10-12T21:41:55 < mawk> then you configure the pins you want to use 2019-10-12T21:41:58 < mawk> select as GPIO_Output 2019-10-12T21:42:03 < mawk> just click on them, select that, nothing more to do 2019-10-12T21:42:07 < Jan-> select what as GPIO_Output 2019-10-12T21:42:52 < mawk> the leds 2019-10-12T21:42:57 < mawk> if they are not select like this before 2019-10-12T21:43:01 < mawk> you select your board right ? 2019-10-12T21:43:16 < mawk> if you select the core you have to put every pin in the position you want, but if you select the board it's already done 2019-10-12T21:43:27 < mawk> but you told me you selected stm32f407, so go back and select the nucleo board you have (if it is indeed nucleo) 2019-10-12T21:43:34 < mawk> selecting a bare chip is for custom designs 2019-10-12T21:43:49 < Jan-> it's not 2019-10-12T21:43:55 < Jan-> it's some rando board that some rando guy on irc told me to get 2019-10-12T21:43:58 < Jan-> so I did :( 2019-10-12T21:43:59 < mawk> ah 2019-10-12T21:44:01 < mawk> then it's custom 2019-10-12T21:44:04 < mawk> don't worry it's easy 2019-10-12T21:44:16 < mawk> you just have to select the pins you want, and set the oscillator frequency like it should be 2019-10-12T21:44:22 < Jan-> ok hang on 2019-10-12T21:44:29 < Jan-> PA6 and PA7 wasn't it 2019-10-12T21:44:32 < mawk> yes 2019-10-12T21:44:34 < mawk> open the cubeMX perspective you talked about 2019-10-12T21:44:57 < Jan-> ok set the pins 2019-10-12T21:45:09 < mawk> the most important one is the clock, then everything will run smoothly 2019-10-12T21:45:12 < Jan-> what does that actually do, write some automatic setup code? 2019-10-12T21:45:16 < mawk> so, I guess you have like a 8MHz crystal on the board ? 2019-10-12T21:45:18 < mawk> yes 2019-10-12T21:45:21 < Jan-> yes we noticed that 2019-10-12T21:45:30 < Jan-> I thought it was a 168mhz chip but the crystal is 8 2019-10-12T21:45:54 < mawk> yeah 2019-10-12T21:45:56 < mawk> that's not a problem 2019-10-12T21:45:59 < Jan-> is this under "clock configuratioN" 2019-10-12T21:46:01 < mawk> you have frequency multipliers inside the chip 2019-10-12T21:46:04 < mawk> yes 2019-10-12T21:46:10 < mawk> the crystal you see on the board is named HSE 2019-10-12T21:46:25 < Jan-> mmhmm 2019-10-12T21:46:44 < mawk> so in clock configuration you see a box named HSE 2019-10-12T21:46:49 < Jan-> sorry if I'm slow I have a buddy helping me 2019-10-12T21:46:50 < mawk> just left of that you can input its frequency, set it to 8 2019-10-12T21:46:54 < mawk> yeah don't worry 2019-10-12T21:47:03 < Jan-> right now it says "input frequency 25" 2019-10-12T21:47:07 < Jan-> but it's greyed out 2019-10-12T21:47:21 < mawk> ah 2019-10-12T21:47:32 < mawk> in the pinout & configuration tab, look at the RCC peripheral 2019-10-12T21:47:33 < mawk> in system core 2019-10-12T21:47:36 < Jan-> won't that just mean the whole thing runs a bit too slowly 2019-10-12T21:47:36 < mawk> tick the HSE box 2019-10-12T21:47:45 < mawk> now it won't be greyed out anymore 2019-10-12T21:47:49 < mawk> then you can set its freq to 8MHz 2019-10-12T21:48:08 < mawk> well bad clock will either not run, or run too slow or too fast yes 2019-10-12T21:48:23 < Jan-> hang on a second 2019-10-12T21:48:30 < Jan-> I have "high speed clock (HSE)" 2019-10-12T21:48:38 < Jan-> disable, bypass or ceramic resonator 2019-10-12T21:48:41 < Jan-> right now it's at disable 2019-10-12T21:48:53 < jpa-> (though if you configure too fast clock and the flash wait states are not correct, it will just crash) 2019-10-12T21:49:03 < Jan-> oo brutal 2019-10-12T21:49:06 < mawk> Jan-: https://pix.watch/Co3LoC/ujD_NE.png 2019-10-12T21:49:08 < mawk> like this ? 2019-10-12T21:49:20 < mawk> let me use the same board as you to see how it looks like 2019-10-12T21:49:32 < mawk> which stm32f407 you selected ? 2019-10-12T21:49:56 < Jan-> vet6 2019-10-12T21:50:10 < Jan-> because that's what's engraved on the chip 2019-10-12T21:50:23 < mawk> thanks 2019-10-12T21:50:23 < mawk> yeah 2019-10-12T21:50:24 < Jan-> it doesn't look like that 2019-10-12T21:50:43 < Jan-> we have 2 dropdowns called high speed clock and low speed clock 2019-10-12T21:50:55 < Jan-> and three checkboxes, master clock output 1, master clock output 2 and audio clock input 2019-10-12T21:51:02 < mawk> only HSE is useful to us for now 2019-10-12T21:51:04 < PaulFertser> So much handholding wow 2019-10-12T21:51:07 < mawk> lol 2019-10-12T21:51:32 < Jan-> PaulFertser: I have a blink program that doesn't include any commands that would make any LEDs blink, gimme a fricken break. 2019-10-12T21:51:32 < mawk> I was very confused in Jan- shoes when I first used cubemx too PaulFertser 2019-10-12T21:51:38 < mawk> (but I figured it out myself by trying things) 2019-10-12T21:51:59 < PaulFertser> Seriously, isn't one supposed to take a look at the clock distribution diagram in the family manual regardless of the software used to configure the peripherals? 2019-10-12T21:52:01 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-12T21:52:34 < Jan-> we have the diagram here at least I think we do 2019-10-12T21:52:42 < PaulFertser> mawk: that UI doesn't seem to be useful without looking and actually understanding the diagram. 2019-10-12T21:52:46 < Jan-> we need apparently to change the 25 to and 8 but 2019-10-12T21:52:50 < Jan-> it's disabled 2019-10-12T21:53:07 < mawk> yes you set the HSE to crystal 2019-10-12T21:53:17 < mawk> in the RCC peripheral 2019-10-12T21:53:25 < srk> PaulFertser: clock tree is just a tab away anyway :) 2019-10-12T21:53:28 < mawk> I'm downloading your software package to have the same view as you 2019-10-12T21:53:33 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-12T21:53:59 < PaulFertser> And also, for getting a simple blink as well as many other purposes one doesn't really need HSE, as HSI is enough. Why risk enabling HSE, it might be not properly soldered or defective etc, why not get it working with HSI first? 2019-10-12T21:54:23 < mawk> ok good, yeah set it to "crystal/ceramic resonator" Jan- , the other option "bypass" means you get a clock signal from a master device (which is what happens on Nucleo boards for instance, which gets the HSE clock signal from the embedded programmer) 2019-10-12T21:54:41 < Jan-> OK done 2019-10-12T21:54:57 < mawk> yes to begin you can just select the internal clock, it should be the default configuration anyway 2019-10-12T21:55:14 < mawk> if you want to go further with that board that's what you'd have to do to use the discrete crystal 2019-10-12T21:55:23 < mawk> (for instance for USB) 2019-10-12T21:56:51 < Jan-> Still no blinking. 2019-10-12T21:57:03 < mawk> yes you need to select the pins as GPIO outputs now 2019-10-12T21:57:10 < Jan-> have done 2019-10-12T21:57:11 < PaulFertser> mawk: I'm actually surprised you're trying to walk the person through the configuration because I know you have serious skills digging math. And if math requires one to read and think to understand the concepts why do you think an "easy start" can be beneficial in embedded development? 2019-10-12T21:57:13 < Jan-> they're green 2019-10-12T21:58:08 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-12T21:58:17 < Jan-> PaulFertser I'm not quite sure what you expect here. Are we somehow supposed to have automatically known to go into "pinout and configuration" and select "RCC" then set "HSE" to "crystal/ceramic resonator" then to into "clock configuration" and set "HSE" to 8? 2019-10-12T21:58:33 < Jan-> I mean obviously a five year old would know that right. That's completely obvious. 2019-10-12T21:59:37 < mawk> he means if you read the reference manual you would know all the clock names and so on, and know that the external crystal will be named HSE, but that you also have an internal high speed clock named HSI which is good for most purposes, and so on 2019-10-12T21:59:46 < PaulFertser> Jan-: please take no offense but IMHO that should be about obvious for someone who read the family reference manual, yes. It's not necessary to read all of it, but Clocks/RCC is the obviously essential part. And then yes, it should be obvious that you need to look for those options. 2019-10-12T21:59:49 < Jan-> OK right what we have now are both LEDs illuminated, which suggests that it is at least setting the gpio up. 2019-10-12T21:59:49 < bitmask> to measure a 12v battery on a 5v system can I use a 470K + 100K Voltage divider and a 0.1uF cap? 2019-10-12T22:00:25 < mawk> PaulFertser: I know that Jan- is already developing on embedded but not on stm32 for now, otherwise I wouldn't be going all that way like that I guess indeed 2019-10-12T22:00:34 < bitmask> I was gonna use a mosfet to reduce power consumption but at those resistances I really don't mind having it on all the time 2019-10-12T22:00:34 < mawk> I trust Jan- to be able to take on from there after the setup 2019-10-12T22:00:50 < Jan-> we have a heartbeat :D 2019-10-12T22:01:06 < Jan-> I would suggest that the defaults are 2019-10-12T22:01:08 < Jan-> unhelpful 2019-10-12T22:01:11 < PaulFertser> mawk: doesn't the same apply to AVR? They also have clock diagrams, internal and external sources etc. 2019-10-12T22:01:11 < mawk> HSE: https://pix.watch/QMyx4c/Ilshpb.png ; HSI: https://pix.watch/4Fz-3I/62E1Yv.png 2019-10-12T22:01:20 < mawk> well if you had a nucleo boards the defaults work great Jan- , but you have a custom board 2019-10-12T22:01:23 < Jan-> PaulFertser: to be honest not to anything like the same degree. 2019-10-12T22:01:27 < mawk> yes indeed PaulFertser 2019-10-12T22:01:37 < Jan-> mawk: I guess. But to breakpoint before running any code is evil 2019-10-12T22:01:45 < mawk> it's to detect if the board is resetting 2019-10-12T22:01:51 < mawk> but yeah it's maybe not great to have that as default 2019-10-12T22:01:57 < Jan-> PaulFertser: AVRs don't have clock configuration like that. 2019-10-12T22:02:18 < Jan-> You have an F_CPU define that tends to be set and if it is wrong then things like delays will be wrong, but nothing like that. 2019-10-12T22:02:29 < Jan-> I guess you can set the clock source and dividers for the SPI and other peripherals. 2019-10-12T22:03:14 < PaulFertser> Jan-: well, yes, stm32 manual is like 5 times bigger because it's more capable and its peripherals are more flexible. But all the basic principles are the same, clocks, plls, dividers, GPIO modules, so IMHO it's odd you were confused by the added complexity. 2019-10-12T22:03:25 < qyx> you shall not using exact nop delays on cortex-m 2019-10-12T22:03:47 < qyx> +be 2019-10-12T22:03:53 < Jan-> I'm not confused by the added complexity, I'm confused by the fact that "clock" is called HSI and the controls for it are in an option called "RCC" in a tab called "Pinout." 2019-10-12T22:04:02 < Jan-> Which is, and let's be fair to ST here, bullshit. 2019-10-12T22:04:37 < Jan-> qyx if I wouldn't use them on 168mhz cores I'm not going to use them on 16mhz cores :D 2019-10-12T22:04:38 < srk> it means High Speed Internal 2019-10-12T22:04:40 < PaulFertser> Jan-: with AVR you can select either internal RC or external resonator or just clock input via "fuse" bytes, same configuration just more complicated because it's a special programming step and not tweakable at runtime. 2019-10-12T22:04:44 < srk> or External 2019-10-12T22:05:05 < srk> qyx: freertos delays aren't nop based anyway 2019-10-12T22:05:19 < PaulFertser> Jan-: and F_CPU is the same meaning as this "8" number you enter for HSE value. If you get it wrong then you'll have your software make wrong assumptions about the frequency but that's about it. 2019-10-12T22:05:28 < Jan-> I'm slightly confused as to why the clock divider being wrong stopped it working. 2019-10-12T22:05:30 < mawk> it's just reference manual lingo Jan- 2019-10-12T22:05:36 < Jan-> or the clock multiplier, more to the point 2019-10-12T22:05:38 < mawk> they won't call is "high speed external clock" all throughout 2019-10-12T22:05:49 < mawk> or "reset and clock configuration peripheral" 2019-10-12T22:05:54 < Jan-> I would have thought it would just run about a third normal speed. 2019-10-12T22:05:55 < mawk> HSE and RCC are more concise 2019-10-12T22:05:59 < Jan-> instead it seemed not to start up at all. 2019-10-12T22:06:05 < mawk> it's maybe just the GPIO lacking 2019-10-12T22:06:14 < mawk> the default conf is with HSI, it should work 2019-10-12T22:06:30 < fooman2011> ...mmmh I will forget to output display from stm32 using MIPI interface.... Is there a way to display HDMI using an stm32 ? 2019-10-12T22:07:00 < jpa-> you can get a MIPI-to-HDMI converter :) 2019-10-12T22:07:18 < jpa-> (though you can also get a RGB-to-HDMI converter for that matter) 2019-10-12T22:07:35 < Jan-> so I guess all the stuff we just did is actually reflected in MX_GPIO_Init and SystemClock_Config? 2019-10-12T22:07:51 < mawk> yes 2019-10-12T22:08:00 < mawk> especially MX_GPIO_Init yes 2019-10-12T22:08:06 < Jan-> I can't find an 8 anywhere 2019-10-12T22:08:16 < mawk> if you modify the code outside of the tags that say you can, it will be lost 2019-10-12T22:08:28 < PaulFertser> Jan-: the level of "hate" on this channel is usually outstanding, you're really lucky this evening to get none of it so far :) 2019-10-12T22:08:38 < Jan-> it's IRC whaddayawant 2019-10-12T22:08:43 < Jan-> happily I like nerds 2019-10-12T22:08:50 * Jan- offers PaulFertser a little cake 2019-10-12T22:09:38 < PaulFertser> Thank you Jan- . I'll be happy to help you with OpenOCD when you finally decide to use Free Software tools ;) 2019-10-12T22:09:48 < Jan-> I'm not loving cubemx 2019-10-12T22:09:55 < Jan-> it seems set up for way bigger projects than the one I intend to do 2019-10-12T22:09:55 < fooman2011> ok so I will ask my question differently. What is the easiest way to ake an stm32 output display to a MIPI 3 lane LCD ? 2019-10-12T22:10:09 < mawk> you hit generate code right Jan- 2019-10-12T22:10:12 < mawk> it's in #define HSE_VALUE ((uint32_t)8000000U) /*!< Value of the External oscillator in Hz */ 2019-10-12T22:10:14 < Jan-> but I could not possibly care less if it is free software or not, it is absolutely meaningless to me. 2019-10-12T22:10:28 < Jan-> I thought "generate code" might be where it would turn all the settings into code? 2019-10-12T22:10:34 < mawk> yes indeed 2019-10-12T22:10:42 < fooman2011> It's for an embedded project 2019-10-12T22:10:44 < mawk> alt-k 2019-10-12T22:11:03 < jpa-> fooman2011: it will probably be quite complex in any case 2019-10-12T22:11:17 < fooman2011> So the "less complex" 2019-10-12T22:11:17 < mawk> PaulFertser: given that cubeIDE can't flash to a board Jan- will probably need it someday 2019-10-12T22:11:41 < Jan-> what do you mean can't flash to a board 2019-10-12T22:11:43 < Jan-> what's it doing right now 2019-10-12T22:12:04 < mawk> it's debugging 2019-10-12T22:12:14 < Jan-> excuse me if I say 2019-10-12T22:12:17 < Jan-> huh? 2019-10-12T22:12:42 < mawk> I mean there is no button to just flash the image to device 2019-10-12T22:12:51 < PaulFertser> mawk: cubeIDE can't be used in batch mode for mass-production? Hah, fancy. 2019-10-12T22:12:52 < Jan-> we were wondering about that 2019-10-12T22:13:12 < Jan-> where's the upload button 2019-10-12T22:13:24 < jpa-> fooman2011: set mode 1100 on the controller chip pins and use 2-lane MPI DSI? 2019-10-12T22:15:07 < fooman2011> jpa-: You mean the controller chip on the lcd ? 2019-10-12T22:15:13 < jpa-> yeah 2019-10-12T22:15:50 < jpa-> with the poor specs it is not clear whether those IMx pins are lurking somewhere among the NC pins on the connector and/or are they available as solder bridges on the pcb 2019-10-12T22:15:53 < fooman2011> MIPI 3 lane LCD is compatible with MIPI 2 lane stm32 ? 2019-10-12T22:16:22 < fooman2011> lane seems to be link to the data transfert speed 2019-10-12T22:16:23 < jpa-> no, but that ILI9881C controller supports 2-4 lanes 2019-10-12T22:16:27 < mawk> but it can use openOCD for debugging PaulFertser 2019-10-12T22:16:27 < PaulFertser> Jan-: btw, I'd say one of the big advantages of using stm32 is that you have proper on chip debug without any expensive equipment. So when something goes wrong you can actually debug it, not just make little amendments and flash again and again looking at prints over uart or leds blinking. 2019-10-12T22:16:36 < fooman2011> jpa-: ok understood 2019-10-12T22:17:17 < mawk> https://pix.watch/UNr8PR/XXHT6u.png 2019-10-12T22:17:38 < PaulFertser> That's eclipse 2019-10-12T22:18:00 < mawk> ah, I thought ST added support for that themselves 2019-10-12T22:18:08 < mawk> or truestudio guys maybe 2019-10-12T22:18:21 < mawk> it's a screenshot of STM32CubeIDE 2019-10-12T22:19:31 < PaulFertser> So they just took upstream Eclipse, GNU toolchain, OpenOCD, and just added some nice peripheral configurator and that's it? 2019-10-12T22:19:44 < mawk> they bought truestudio and put their name on it 2019-10-12T22:19:48 < mawk> and integrated cubeMX into it also 2019-10-12T22:19:53 < PaulFertser> What about that other eclipse-based IDE that ST sponsored? 2019-10-12T22:20:05 < mawk> deprecated now I think 2019-10-12T22:20:10 < mawk> since they're pushing for this new thing now 2019-10-12T22:21:16 < PaulFertser> What are professionals like dongs supposed to be using? Keil + jlinkserver + cubeMX? 2019-10-12T22:21:24 < mawk> lol 2019-10-12T22:22:07 < kakimir64> yes 2019-10-12T22:23:08 < PaulFertser> mawk: so if it's eclipse does it have the same "feature" of not allowing to move a created eclipse project to another directory without pain? 2019-10-12T22:23:23 < mawk> I think yes 2019-10-12T22:23:31 < mawk> last time I tried to move things around I had to modify config files manually 2019-10-12T22:23:55 < PaulFertser> Nuts 2019-10-12T22:24:14 < kakimir64> eclipse = pain 2019-10-12T22:24:29 < Jan-> so er 2019-10-12T22:24:49 < kakimir64> PaulFertser: are you freertos pro? 2019-10-12T22:24:57 < Jan-> am I to understand that I have an IDE here that is for developing software to be uploaded to microcontrollers, but which can't uplaod software to microcontrollers. 2019-10-12T22:25:07 < Jan-> Forgive me if I find that a trifle 2019-10-12T22:25:10 < Jan-> odd 2019-10-12T22:25:14 < PaulFertser> kakimir64: I used to be but I guess I'm not up to date with the current developments now. 2019-10-12T22:25:33 < kakimir64> Jan-: there are so many programmers 2019-10-12T22:26:06 < Jan-> but cube mx ain't one. 2019-10-12T22:26:20 < kakimir64> cubemx is code generator 2019-10-12T22:26:35 < kakimir64> PaulFertser: did you ever use freertos-mpu? 2019-10-12T22:26:36 < PaulFertser> Jan-: I think upload is a bit confusing because you do not need any special support to upload something to any cortex-m target's RAM but you certainly need special target-specific support to alter its Flash. 2019-10-12T22:26:46 < PaulFertser> kakimir64: no, alas 2019-10-12T22:27:13 < Jan-> so it doesn't actually run the program from flash, then 2019-10-12T22:27:18 * Jan- scratches her head 2019-10-12T22:27:48 < Cracki> what 2019-10-12T22:27:49 < PaulFertser> Jan-: I would guess that it does and somehow you managed to configure it already the way it flashes the target on start of a debugging session. 2019-10-12T22:27:56 < PaulFertser> Without noticing it. 2019-10-12T22:27:57 < Cracki> ignore all the talk about "uploading to ram" 2019-10-12T22:28:06 < Jan-> I was going to say 2019-10-12T22:28:11 < Jan-> it doesn't load the entire program into ram to run it surely 2019-10-12T22:28:17 < kakimir64> writing flash is not identical across different arm cortex-m core devices 2019-10-12T22:28:33 < jpa-> which is ridiculous, IMO 2019-10-12T22:28:35 < kakimir64> vendors have their own routines 2019-10-12T22:28:40 < Cracki> I would like to state that "hello world" doesn't need to imply "blinky" 2019-10-12T22:28:53 < Cracki> also a blank generated project doesn't need to mean it's "ready" 2019-10-12T22:28:54 < PaulFertser> Cracki: my point is that the word "upload" is ambigous in this context and hence shouldn't be used. Isn't that a sound point? 2019-10-12T22:29:04 < Cracki> certainly 2019-10-12T22:29:10 < Cracki> but we have a newbie there 2019-10-12T22:29:11 < kakimir64> jpa-: freertos-mpu.. have you played with it? 2019-10-12T22:29:13 < Cracki> with cubeIDE 2019-10-12T22:29:18 < jpa-> kakimir64: nope 2019-10-12T22:29:26 < Cracki> cubeIDE likely has no concept of uploading to ram 2019-10-12T22:29:44 < Cracki> and if it does, it's buried in some settings 2019-10-12T22:29:45 < Jan-> if this thing isn't uploading the code to the stm32 chip and running it, wtf is it doing 2019-10-12T22:29:48 * Jan- gesticulates wildly 2019-10-12T22:29:54 < Cracki> ignore the nerds 2019-10-12T22:30:03 < PaulFertser> Cracki: if it's just Eclipse with HW debugging plugin then I expect it to "load" the project just fine if it's linked to be executed from RAM. 2019-10-12T22:30:12 < PaulFertser> Without any settings. 2019-10-12T22:30:30 < Cracki> nowhere is it implied that this thing would upload or execute from ram 2019-10-12T22:30:36 < Cracki> it's a standard cubeide situation 2019-10-12T22:30:56 < Cracki> and that thing will certainly know how to program the target's flash 2019-10-12T22:30:57 < PaulFertser> I would guess it'll do that automatically if your linker script points to RAM. 2019-10-12T22:31:07 < Jan-> er 2019-10-12T22:31:19 < Cracki> you realize cubeide abstracts that away from you? you never have to see a linker script when you use it 2019-10-12T22:31:19 < Jan-> how can that be true considering how much program flash it has, and how much ram it has 2019-10-12T22:31:37 < PaulFertser> According to this screenshot https://pix.watch/UNr8PR/XXHT6u.png that thing has no idea how to flash target memory, it offloads the work to OpenOCD. 2019-10-12T22:31:40 < Cracki> there are situations when SOME code can be run from ram 2019-10-12T22:31:42 < Cracki> such as interrupt handlers 2019-10-12T22:31:48 < Cracki> or small enough programs 2019-10-12T22:32:04 < jpa-> it would be silly if cubeide didn't ship with stm32 flashing support 2019-10-12T22:32:18 * Jan- now has slightly less of a clue than she had half an hour ago 2019-10-12T22:33:17 < Cracki> look around for an ".ioc" file with a blueish square icon in the project files. doubleclick it. that'll open cubemx and let you configure gpios 2019-10-12T22:33:19 < jpa-> i bet people are confusing cubeide with the non-ST stm32 eclipse combos, and/or with the plain cubemx code generator 2019-10-12T22:33:30 < Cracki> you want that to make sure the gpios you WANT to use are configured right (i.e. output) 2019-10-12T22:33:54 < Cracki> you want to use cubemx for this because it also generates the proper code to initialize these I/O ports (clock, power) 2019-10-12T22:35:01 < Cracki> the setup code itself is rather trivial (a few calls, some structures filled), but instead of explaining the specifics of that, just use cubemc 2019-10-12T22:35:03 < Cracki> mx* 2019-10-12T22:36:01 < Cracki> https://i0.wp.com/www.breiteneder.me/wp-content/uploads/2016/08/05_GPIO_Output.png?resize=1110%2C699 2019-10-12T22:36:02 < PaulFertser> Those initialisation structures were an awful idea, a really lame and confusing and error-prone API. 2019-10-12T22:37:00 < PaulFertser> libopencm3's approach is probably the sanest it can get with C language. 2019-10-12T22:37:13 < Cracki> this is what you should see: https://i.ytimg.com/vi/LKiHViMFTKM/maxresdefault.jpg 2019-10-12T22:37:21 < Cracki> left panel, last item is an .ioc file 2019-10-12T22:37:37 < Cracki> the center tabl with "test-F031.ioc" is cubemx inside the IDE 2019-10-12T22:38:27 < Cracki> the pinout view lets you click/rightclick on pins, set their purpose, such as "GPIO output" 2019-10-12T22:38:59 < PaulFertser> So that's just a pre-configured Eclipse... And everybody was raging here about how they are mad at Eclipse and Java language and runtime and now it's like the best solution, heh? 2019-10-12T22:39:04 < Cracki> you can also do that using the left side panel titled "categories" where it has, in the first collapsible thingy, DMA, GPIO, ... 2019-10-12T22:39:38 < Cracki> "just" preconfigured, "just" 2019-10-12T22:39:47 < kakimir64> https://paste.ee/p/Jayg5 how do I change this so that .uninit_RESERVED and .data do not come right after variables end in privileged_data but start after ORIGIN( SRAM1 ) + _Privileged_Data_Region_Size; jpa-? 2019-10-12T22:39:52 < Cracki> I'm sorry that you don't see the value in that 2019-10-12T22:40:28 < PaulFertser> kakimir64: they're all put in sections in order 2019-10-12T22:40:55 < PaulFertser> kakimir64: what's specified first goes first 2019-10-12T22:41:07 < kakimir64> I figure 2019-10-12T22:41:15 < kakimir64> but https://paste.ee/p/QsTp2 2019-10-12T22:41:35 < kakimir64> lines 173 176 2019-10-12T22:42:08 < kakimir64> .data starts right where variables end in privileged_data 2019-10-12T22:42:38 < jpa-> kakimir64: maybe try to put that line 8 into inside the .uninit_RESERVED {}, like the ALIGN() commands are 2019-10-12T22:43:19 < jpa-> because i think the > SRAM1 might otherwise override the . from the outer context 2019-10-12T22:43:25 < PaulFertser> Cracki: several years ago I tried to configure Eclipse from scratch and successfully made it to debug target, including semi-hosting input and output, without much effort. And that was before GNU MCU Eclipse became available. 2019-10-12T22:43:38 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-12T22:44:02 < kakimir64> jpa-: makes sense 2019-10-12T22:44:10 < kakimir64> I need to go deeper in template 2019-10-12T22:44:14 < PaulFertser> BTW, Cracki, does this ST Eclipse offer full decoded view into peripherals? I wonder how they did that. And does it support SWO/SWV ? 2019-10-12T22:45:32 < PaulFertser> kakimir64: aha, so your line 8 is wrong, dot doesn't reserve any memory. 2019-10-12T22:45:52 < kakimir64> i know 2019-10-12T22:45:57 < Jan-> did we decide if I can actually upload code to this stm32 with this software 2019-10-12T22:45:58 < Jan-> or what 2019-10-12T22:46:07 < kakimir64> but I thought it would like.. continue from it 2019-10-12T22:46:17 < PaulFertser> Jan-: you saw leds blinking right? 2019-10-12T22:46:18 < kakimir64> lets try 2019-10-12T22:46:45 < Thorn> I *almost* came up with the same circuit, I promise http://www.ti.com/lit/ug/slvu391a/slvu391a.pdf 2019-10-12T22:47:39 < Cracki> Jan-, you know that it does that because when the board was fresh it had some code on it that made leds blink. after you flashed the empty while loop on it, it stopped doing that and it will continue to not do that even after a reset (so the new code is in flash, not just in ram) 2019-10-12T22:48:54 < Jan-> right so what was that guy saying earlier on 2019-10-12T22:49:24 < Jan-> I do notice the only option I have is "debug" 2019-10-12T22:49:27 < Jan-> not "make final version" or whatever 2019-10-12T22:49:28 < PaulFertser> Jan-: that "upload" is an unsuitable word to describe the process. 2019-10-12T22:49:33 < kakimir64> what is uninit_RESERVED anyway.. nothing is stored in it 2019-10-12T22:49:41 < Jan-> PaulFertser: er... flash? 2019-10-12T22:49:46 < Jan-> program? 2019-10-12T22:50:01 < PaulFertser> Jan-: yes, if you're talking specifically about storing to flash 2019-10-12T22:50:19 < Cracki> he's making a fine distinction you don't need to concern yourself with at this point 2019-10-12T22:50:27 < Jan-> Oh right. 2019-10-12T22:50:34 < Jan-> I'm aware that's a nerd thing. 2019-10-12T22:50:38 * Jan- patpats PaulFertser 2019-10-12T22:50:40 < Cracki> because the tools do the thing you expect them to 2019-10-12T22:50:43 < Jan-> good neckbeard :) 2019-10-12T22:51:00 < Jan-> on the upside, it works :D 2019-10-12T22:52:37 < Jan-> I would ask how I'm likely to have figured out all that stuff without asking someone. 2019-10-12T22:53:20 < PaulFertser> In fact, I know some engineers not concerned with the "fine distinctions" and other nit-picking. Surprisingly, they can be very productive, but usually only to a certain point where they get stuck for real. It's often beneficial to have people like that in the team as they can do a lot in a short time. 2019-10-12T22:53:34 < Cracki> >at this point 2019-10-12T22:53:52 < Cracki> when you're on the way to the grocery store, you don't stop to pick flowers 2019-10-12T22:54:39 < Cracki> pick flowers after the groceries are home 2019-10-12T22:54:58 < PaulFertser> Then you'll have to return back to the flowers, that's suboptimal time-wise. 2019-10-12T22:55:33 < Cracki> there's a right time to explain everything, and one hour after a newbie has unboxed their first stm32 board, explaining what uploading to ram is... is not the right time 2019-10-12T22:56:19 < Cracki> especially when you notice that it's information overload and distracting from the problem at hand, which is getting a blinky to work 2019-10-12T22:56:34 < PaulFertser> How is stm32 different from just about any other microcontroller or microprocessor systems we used to have (like Intel 8080)? 2019-10-12T22:56:45 < Cracki> there was no reason to suspect that cubeide would upload anything to ram in its newly installed state 2019-10-12T22:57:02 < Jan-> Cracki I think I'm going to step back from this discussion now :) 2019-10-12T22:57:28 < Cracki> blinky works? good :) 2019-10-12T22:57:37 < PaulFertser> Cracki: you're right indeed. It's just that I consider "getting a blinky to work" to be a useless target. 2019-10-12T22:57:59 < kakimir64> PaulFertser: it did it 2019-10-12T22:58:03 < kakimir64> jpa- 2019-10-12T22:58:07 < Cracki> because you have had years of experience already 2019-10-12T22:58:49 < Cracki> it may be a useless target for you specifically, but in general it's a good method to verify that the tools do what they should, and that the operator uses them right 2019-10-12T22:58:56 < PaulFertser> kakimir64: I checked the info manual for ld and your non-working example looks like almost verbatim taken from there :) 2019-10-12T22:59:22 < kakimir64> I need to check english word verbatim.. a moment 2019-10-12T22:59:42 < kakimir64> okay 2019-10-12T22:59:42 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has joined ##stm32 2019-10-12T22:59:49 < Cracki> it's like driving school, where on the first day, before you understand which pedal is gas and which is brake, you're told to drive through the city 2019-10-12T23:00:06 < kakimir64> a "don't do this" example from manual? 2019-10-12T23:00:27 < Cracki> the general point is undisputed, i'm just amazed that you use yourself as a yard stick 2019-10-12T23:01:16 < PaulFertser> Cracki: I'm afraid that certain people can be steered on the side of doing things the easy way (basically, cargo culting, copy-pasting etc, what you usually find in amateur PHP or arduino projects) too easily. So I think a deeper understanding approach should be encouraged generally. 2019-10-12T23:01:58 < PaulFertser> Cracki: funny you mention it but I was actually told to drive through the (calm parts) of the city on my first day at a driving school, yes. 2019-10-12T23:02:16 < PaulFertser> kakimir64: yeah 2019-10-12T23:02:48 < PaulFertser> kakimir64: "The Location Counter" section, and it explains how orphaned sections can later bite you. 2019-10-12T23:02:54 < Cracki> I hope they explained the clutch and made you practice a few times before sending you into traffic 2019-10-12T23:03:11 < specing> nobody has any time to gain a deeper understanding when they spend 5h/day on facebook 2019-10-12T23:03:59 < PaulFertser> Cracki: probably, I do not remember, I knew how to use clutch because I had A license already. And yes, they made me drive forward/backwards for a while, then we went on public road. 2019-10-12T23:06:59 < PaulFertser> Cracki: I taught programming (and microcontrollers a bit) at a local uni. My impression is that the best (wrt engineering skills) students mostly agreed with my approach. 2019-10-12T23:07:35 < kakimir64> now this is dumb.. this project cannot automatically share data across ram blocks. Tried to increase heap and stack until there was no more room and indeed 111% of first block is used 2019-10-12T23:07:41 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has quit [Ping timeout: 276 seconds] 2019-10-12T23:07:44 -!- fooman2011 [~IceChat9@157.206.119.80.rev.sfr.net] has quit [Ping timeout: 268 seconds] 2019-10-12T23:08:09 < kakimir64> is it expected that I set variables in other banks with attributes 2019-10-12T23:08:19 < Cracki> that's neither here nor there. in this situation you have someone who's a newbie and may _not_ be "the best", so you modulate your approach to suit. 2019-10-12T23:08:24 < PaulFertser> Cracki: sometimes I was literally sitting near somebody and had him or her read the book and explain it sentence by sentence. Because apparently reading and understanding text is not such a common thing and needs to be trained specifically. People tend to glimpse over the words. 2019-10-12T23:09:04 < Cracki> we don't have a text book here. 2019-10-12T23:09:11 < srk> :)) 2019-10-12T23:09:18 < PaulFertser> We do, the family reference manual. 2019-10-12T23:09:24 < Cracki> what we could have had were docs for some nucleo, if she had a nucleo 2019-10-12T23:09:33 < PaulFertser> And unknown words (like PLL) can be looked up elsewhere. 2019-10-12T23:09:34 < Cracki> that thing isn't a text book, it's a reference 2019-10-12T23:10:06 < srk> PaulFertser: mcu programming is pretty advanced topic and ppl just want to do *stuff*, not realizing how deep the rabbit hole goes 2019-10-12T23:10:30 < Cracki> the mark of a text book is that it prioritizes and serializes information into an order such that you learn what you need at the specific point you're at 2019-10-12T23:11:44 < Cracki> let's say we wouldn't have gone with cubeide. then *someone* would have had to explain what flags to pass the compiler and the linker so it uses machine instructions instead of soft float, and what calling convention to use, ... 2019-10-12T23:11:49 < PaulFertser> My first microcontroller was some 8051 atmel device, I soldered a PLCC socket to a board, attached max232 and wrote a tool to interact with its ROM bootloader. All by reading the offical datasheet. 2019-10-12T23:12:07 < Cracki> which is sure good to know, but with an integrated environment those things are (usually) already set up correctly 2019-10-12T23:12:07 < srk> Cracki: most toolchains will do that for you 2019-10-12T23:12:22 < Cracki> so you can turn your attention to these things _when_ you want, not before you've got anything running at all 2019-10-12T23:12:36 < Cracki> srk, know any that don't? 2019-10-12T23:12:43 < PaulFertser> David A. Patterson, John L. Hennesy “Computer Organization and Design. The Hardware/Software Interface” <-- that's a nice textbook 2019-10-12T23:13:35 < srk> Cracki: not really, depends how well is the concept of platforms / mcus done 2019-10-12T23:13:46 < Cracki> take plain arm gcc 2019-10-12T23:13:55 < srk> learn ld 2019-10-12T23:13:57 < srk> good to go! 2019-10-12T23:13:57 < Cracki> with your own handwritten makefile, in case anyone does it 2019-10-12T23:14:05 < Cracki> see? 2019-10-12T23:14:24 < srk> or just flash micropython a be done 2019-10-12T23:14:25 < karlp> fuck me you guys went on a bit this afternoon. 2019-10-12T23:14:31 < PaulFertser> Cracki: I think the stm32 family reference manuals are organised in a textbook-like fashion, prioritising stuff until it comes to sections describing peripherals. 2019-10-12T23:14:33 < srk> *and 2019-10-12T23:14:35 < karlp> what a distraction with "not uploading" 2019-10-12T23:14:36 < Cracki> I think I'm being trolled. it feels like you're saying "before you can use a phone, you gotta know how GSM works" 2019-10-12T23:15:01 < Jan-> I think you are being trolled 2019-10-12T23:15:01 < karlp> mawk: that' 2019-10-12T23:15:04 < karlp> s on you, 2019-10-12T23:15:07 < Jan-> that's why I stepped out 2019-10-12T23:15:32 < karlp> it flashes as part of debug, just doesn't give you the standalone button you desire that _only_ flashes 2019-10-12T23:15:34 < PaulFertser> Cracki: haha, before I could boot my phone I had to fix its bootloader to boot from my new uSD card :) 2019-10-12T23:15:35 < srk> Cracki: depends, e.g. my toolchain abstracts most of the gore details :) 2019-10-12T23:15:47 < Cracki> quite wise, I shall step away now too 2019-10-12T23:17:10 < srk> sometimes people just need a setup and for(;;) loop with blocking calls 2019-10-12T23:17:30 < srk> interrupts and DMA are too scary 2019-10-12T23:17:50 < karlp> PaulFertser: I'm generallyu on your side with learning more first, but there is a lot of value in "is my setup actually functional" and some cargo cult simplicity can be very useful there. 2019-10-12T23:18:08 < karlp> hence https://github.com/libopencm3/libopencm3-miniblink :) 2019-10-12T23:18:14 < kakimir64> https://www.youtube.com/watch?v=h7qTI6Njp9g musics then 2019-10-12T23:18:38 < karlp> speaking of, time to get to a concert 2019-10-12T23:18:38 < PaulFertser> karlp: I probably had a feeling that I spotted some Arduino-like attitude in Jan- , and _that_ I didn't like. 2019-10-12T23:18:40 < srk> hehe, can't get blink running can be pretty demotivating 2019-10-12T23:20:38 < Jan-> One of the best things about the world is that it doesn't care what PaulFertser likes. 2019-10-12T23:20:58 < PaulFertser> Cracki: btw, I appreciate your input much, and when I'm wrong please tell me I'm wrong, I'm willing to learn. 2019-10-12T23:20:59 * srk is looking for some reading materials about digital synths / wavetable synthesis 2019-10-12T23:21:31 < PaulFertser> Jan-: haha, you never know. Ain't many of us thinking the world would be a better place if we were making the crucial decisions? :) 2019-10-12T23:23:04 < PaulFertser> Jan-: I probably like digging because that's what I do so I'm certainly biased. But that doesn't necessarily mean I'm wrong. 2019-10-12T23:23:26 < Jan-> no just needlessly pedantic, distracting, and unhelpful in a way you should know you are. 2019-10-12T23:25:34 < PaulFertser> Jan-: I am today in this specific conversation, not necessarily in general. I've fixed many bugs and gave plenty of helpful advices just talking to people on IRC, so I do not think I'm generally exactly like you put it. 2019-10-12T23:27:35 < Thorn> musics https://www.youtube.com/watch?v=jdSeufSZaGc 2019-10-12T23:28:26 < Steffanx> i think kakimir64 will give this a 1/5 Thorn 2019-10-12T23:28:59 < Thorn> he lacks computing power necessary to appreciate this masterpiece 2019-10-12T23:29:01 < Cracki> not enough cats in it? 2019-10-12T23:29:23 < Steffanx> i like it. 2019-10-12T23:29:23 < Cracki> what audio playback got to do with computing power 2019-10-12T23:29:33 < Cracki> except if you mean wetware crunch 2019-10-12T23:29:36 < Jan-> is there some advantage to using a version of the software that's not in debug mode once I'm happy I've got it all right 2019-10-12T23:30:09 < Cracki> debug/release determines how much optimization the compiler does. the more it optimizes, the weirder it's running when you debug it 2019-10-12T23:30:17 < BrainDamage> it'll run marginally faster and with less resources, 2019-10-12T23:30:31 < Cracki> so for debugging you generally want just slight or no optimization, so the code executes exactly as you write it 2019-10-12T23:30:33 < BrainDamage> but if it already accomplishes the task you've set yourself for, no real reason 2019-10-12T23:30:50 < Jan-> I just can't find the debug/final switch 2019-10-12T23:30:55 < Jan-> the only button is "debug" 2019-10-12T23:31:18 < Cracki> some optimizations may reorder code. you'll notice that when singlestepping through it, which can be a headache 2019-10-12T23:31:33 < Cracki> ah, that debug is actually running the debugger and being able to singlestep 2019-10-12T23:31:44 < Cracki> there is a "debug" build option, as opposed to a "release" option 2019-10-12T23:31:51 < kakimir64> eclipse Debug Release button? 2019-10-12T23:32:20 < kakimir64> in mine it's right next to save all button 2019-10-12T23:32:25 < kakimir64> kinda dial looking 2019-10-12T23:32:45 < kakimir64> "Manage configurations for current project" 2019-10-12T23:33:33 < kakimir64> it has little arrow for dropdown that reveals list: release, debug 2019-10-12T23:34:05 < Cracki> arduino only knows how to write a program to flash (and then take its hands off to let it run). cubeIDE only knows how to (program flash and) run the target under debugger control, but not how to _merely_ write a program to flash 2019-10-12T23:34:45 < kakimir64> sometimes there is flash tool button in these IDEs 2019-10-12T23:34:57 < kakimir64> that works only with real jlink or so 2019-10-12T23:35:15 < srk> make flash ftw 2019-10-12T23:35:31 < kakimir64> makefile flasher? 2019-10-12T23:35:34 < kakimir64> :o 2019-10-12T23:35:43 < Cracki> make great flash again 2019-10-12T23:35:53 < srk> make flash great again! 2019-10-12T23:36:10 < Jan-> er 2019-10-12T23:36:10 < srk> like gpu accelerated 2019-10-12T23:36:27 < Cracki> for great epilepsy 2019-10-12T23:37:05 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-12T23:37:07 < srk> I'm using arm-none-eabi-gdb flasher bith black magic 2019-10-12T23:37:17 < srk> sounds way better than jlink! 2019-10-12T23:37:38 < srk> *w 2019-10-12T23:37:39 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Quit: ZNC - https://znc.in] 2019-10-12T23:37:44 < Cracki> try ozone, you'll only use jlink from then on 2019-10-12T23:37:59 < srk> how come? 2019-10-12T23:38:20 < Cracki> because ozone is excellent, and because they only support their own devices 2019-10-12T23:38:40 < srk> excellent how? :) 2019-10-12T23:38:43 < srk> faster? 2019-10-12T23:38:46 < Cracki> yes 2019-10-12T23:38:54 < Jan-> I'm still not really clear whether my code is actually flashed into this dev board in a way that's correct. 2019-10-12T23:39:04 < Cracki> I don't have much to compare it with, but it's eggselent 2019-10-12T23:39:04 < srk> way faster? usb3 fast? 2019-10-12T23:39:16 < Cracki> it doesn't upload the whole program, just changed bits 2019-10-12T23:39:19 < Cracki> *bytes/parts 2019-10-12T23:39:24 < Cracki> so that speeds shit up a lot 2019-10-12T23:39:51 < Cracki> Jan-, if it works, that's correct enough 2019-10-12T23:40:13 < srk> I still spend most of the time during compilation or actually thinking about what to write, upload is not really a bottleneck 2019-10-12T23:40:50 < srk> often I don't even upload when my code compiles, saves a lot of time :D 2019-10-12T23:40:59 < Cracki> that's true 2019-10-12T23:41:04 < srk> but diffs are cool! 2019-10-12T23:41:15 < Cracki> something that's almost taken for granted now is plotting variables while the target is running 2019-10-12T23:41:15 < srk> gues it could be done with BMP or OOCD as well 2019-10-12T23:41:23 < Cracki> ozone does it, and does it better than eclipse shit 2019-10-12T23:41:30 < srk> via SWO? 2019-10-12T23:41:43 < Cracki> not sure, either that or normal SWD 2019-10-12T23:41:53 < Cracki> same with SWV, updates isntantly, not once a sec like eclipse plugin does 2019-10-12T23:42:26 < srk> once I've tried to install eclipse with some arm plugin to get register viewer 2019-10-12T23:42:32 < srk> gave up on third package manager 2019-10-12T23:42:37 < Cracki> :) 2019-10-12T23:43:08 < Cracki> that's why I don't bother setting it up manually. I expect ST to figure it out and make it work, and if they fuck it up, I start looking. 2019-10-12T23:43:42 < Cracki> and ozone seems to have the attention to detail that's required to make things run smoothly 2019-10-12T23:44:06 < srk> what's SWV? 2019-10-12T23:44:31 < srk> ST can't even get their SVD files right! :D 2019-10-12T23:44:34 < Cracki> SWV is the text you dump over SWO channel 0 I think 2019-10-12T23:44:53 < srk> ah, ok 2019-10-12T23:45:07 < Cracki> SWO is the pin, it's carrying a level or two of encoded data 2019-10-12T23:45:31 < srk> reminds me I wanted to write ITM print support 2019-10-12T23:46:01 < Cracki> ITM_SendChar? 2019-10-12T23:46:14 < Cracki> that's what you must call in _write in syscalls.c 2019-10-12T23:47:05 < srk> need registers :) I don't have ITM_SendChar 2019-10-12T23:47:18 < Jan-> what's an .elf file anyway 2019-10-12T23:50:49 < Cracki> machine code, but packaged so it ends up in the right memory locations 2019-10-12T23:51:06 < Cracki> roughly comparable to a .exe 2019-10-12T23:57:46 < PaulFertser> Cracki: for your projects, are you usually debugging -O0 code? I tend to enable full optimisations right from the beginning as it "unhides" bugs as I make them, not several months later when the project "just doesn't fit anymore". 2019-10-12T23:59:04 < Cracki> I don't think I ever use O0. O2 standard, I haven't touched O3 in a while, dropping to O1 if code debugs strangely --- Day changed Sun Oct 13 2019 2019-10-13T00:00:16 < PaulFertser> LTO is a bitch though, really makes debugging problematic in my experience. -O2 is kinda weird but you soon get used to it, if some var is not yet available just check it after few more steps and it's there. 2019-10-13T00:00:31 < Steffanx> And you never have to debug code that get optimised in such way you cant even watch half of the variables you'd like to see? 2019-10-13T00:01:09 < Steffanx> Like "optimized out" crap 2019-10-13T00:01:48 < Cracki> "not yet available" I've gotten used to that. if I really need to see that variable, I make it volatile while I scrutinize that part 2019-10-13T00:02:00 < Cracki> only works for some classes of bug obv 2019-10-13T00:02:35 < PaulFertser> Steffanx: it usually becomes available a bit later, just before it's actually needed for some other expression or function call. 2019-10-13T00:03:00 < PaulFertser> I guess srk doesn't need step-by-step debugging mostly :) 2019-10-13T00:03:52 < Steffanx> Haskell code is bugfree 2019-10-13T00:04:47 < PaulFertser> I guess some bugs happen but they're of different nature and hence need different treatment. 2019-10-13T00:07:00 < PaulFertser> IvoryTower is a DSL so you do not program controllers in Haskell actually but rather in IvoryTower? 2019-10-13T00:08:32 < kakimir64> debugging session is running 2019-10-13T00:08:48 < kakimir64> dumdum didn't start jlink server itself so I started it separatelly 2019-10-13T00:09:09 < kakimir64> it kept asking me to give a location of jlinkserver executeable 2019-10-13T00:11:53 -!- fooman2011 [~IceChat9@157.206.119.80.rev.sfr.net] has joined ##stm32 2019-10-13T00:11:54 < PaulFertser> Cracki: I think SWV is not a text, it's a special ITM event generated by watchpoint/breakpoint unit and the host is decoding and showing it, it can work alongside the ITM prints automatically. 2019-10-13T00:12:04 < Cracki> yes 2019-10-13T00:12:38 < Cracki> whatever they say in the arm coresight docs 2019-10-13T00:14:02 < PaulFertser> I mean I have an impression you can still use ITM stimulus 0 port for arbitrary text, and those variable values are transferred independently. I admit I do not remember it clearly, but I thought it's a completely different time of ITM event. 2019-10-13T00:31:14 -!- fooman2011 [~IceChat9@157.206.119.80.rev.sfr.net] has quit [Quit: It's a dud! It's a dud! It's a du...] 2019-10-13T00:34:26 < kakimir64> jlink server just closes itself when gdb is dumped 2019-10-13T00:34:40 < kakimir64> meeh 2019-10-13T00:36:56 < kakimir64> PLL fixed 2019-10-13T00:37:08 < kakimir64> mainloop and while(1) reached 2019-10-13T00:39:04 < kakimir64> oh while(1) in startup file 2019-10-13T00:40:14 < Steffanx> Ohno 2019-10-13T00:41:19 < kakimir64> it blew right pass the vTaskStartScheduler(); //Launch RTOS scheduler 2019-10-13T00:46:43 < kakimir64> actually it didn't 2019-10-13T00:46:49 < kakimir64> main returned before that 2019-10-13T00:47:24 < kakimir64> oh I had breakpoints disabled :o 2019-10-13T00:49:38 < PaulFertser> Cracki: btw, did you know ELF files basically have a Turing-complete language in them? https://media.ccc.de/v/29c3-5195-en-executable_metadata_h264 2019-10-13T00:51:05 < kakimir64> are we going to hear pro level stuff now? 2019-10-13T00:54:41 < PaulFertser> I'd say that video is pro level, yes. 2019-10-13T01:05:29 < Thorn> which one(s) do you like more https://xfmotospace.com/gallery/admin/images/J0300802X/J0300802X.jpg 2019-10-13T01:06:54 < kakimir64> pen skins 2019-10-13T01:07:43 < kakimir64> you need to get them all 2019-10-13T01:07:49 < Thorn> https://i.redd.it/5z37dbu0nu911.jpg 2019-10-13T01:08:14 < kakimir64> one on right 2019-10-13T01:08:23 < kakimir64> is this like a hobby of sorts? 2019-10-13T01:08:26 < Cracki> writing propagates raep culture. the paper is the feminine, the pen is the phallic, defiling the paper. 2019-10-13T01:09:13 < kakimir64> you are like lurencer 2019-10-13T01:09:18 < kakimir64> but better at it 2019-10-13T01:09:28 < kakimir64> it's a complement 2019-10-13T01:09:42 < Cracki> I'm not sure I want to be better at it 2019-10-13T01:10:22 < kakimir64> it seems I finally need to start using asserts and provide an output for them 2019-10-13T01:10:42 < kakimir64> going in rabbithole with debugger.. nope 2019-10-13T01:11:18 < Cracki> order a jlink clone if you don't have one already 2019-10-13T01:11:32 < kakimir64> I have a pile of them 2019-10-13T01:11:34 < Cracki> wait you do, all the talk of jlink gdb servers 2019-10-13T01:11:48 < kakimir64> but I have real jlink now 2019-10-13T01:12:03 < Cracki> moneyed westerner 2019-10-13T01:12:11 < kakimir64> I reffer freertos as the rabbithole 2019-10-13T01:12:39 < Cracki> 1 link is like 2 cups of coffee at starcucks 2019-10-13T01:12:51 < kakimir64> it's terribly inefficient to do it all with debugger stepping 2019-10-13T01:12:58 < kakimir64> and breakpoints 2019-10-13T01:12:59 < Cracki> data watpoints 2019-10-13T01:13:12 < Cracki> trigger when hatefacts are detected 2019-10-13T01:13:12 < kakimir64> wat 2019-10-13T01:13:21 < Cracki> ch 2019-10-13T01:13:31 < kakimir64> lurencer did broke you 2019-10-13T01:14:02 < Cracki> bongland attacked germoney and got trashed so they had to call in the burgers 2019-10-13T01:14:47 < kakimir64> I think I'm done for the day with rtos 2019-10-13T01:14:57 < kakimir64> or am I? 2019-10-13T01:15:06 < Cracki> dog in water https://youtu.be/XITHDuemkHs?t=1370 2019-10-13T01:21:16 < Mangy_Dog> bark bark bark bark bark 2019-10-13T01:31:39 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-13T01:32:16 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-13T01:34:23 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-13T01:34:30 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 268 seconds] 2019-10-13T01:37:17 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-13T01:46:07 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-13T02:21:51 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-13T02:25:04 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-13T02:26:16 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 265 seconds] 2019-10-13T03:07:23 < Cracki> detecting deep fake videos https://www.youtube.com/watch?v=RoGHVI-w9bE 2019-10-13T03:15:36 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 246 seconds] 2019-10-13T03:24:27 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-13T03:25:38 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-13T03:28:07 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 250 seconds] 2019-10-13T03:50:22 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-13T03:57:16 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-13T04:05:54 < Tordek> is it possible to, I guess, mark 2 breakpoints and count how many times each is reached, without pausing execution? 2019-10-13T04:08:52 < kakimir64> not that I know of 2019-10-13T04:09:16 < kakimir64> how about just counter++; 2019-10-13T04:11:33 < Cracki> pedestrian way is to insert "counter++" at those places, and if you want it to break after X counts, do a data watchpoint with that value 2019-10-13T04:11:42 < Cracki> you want a profiler, don't you 2019-10-13T04:12:54 < Cracki> some IDEs support that from trace output, either instruction level if chip and debugger hw support it, or function entry/exit, or even program counter sampling 2019-10-13T04:14:03 < Cracki> I think I've seen "break point on x-th hit" in _some_ IDE that may not have been for embedded programming at all 2019-10-13T04:21:26 < Tordek> thing is I have some code running inside a timer interruption and I wanted to check if it's not taking too long 2019-10-13T04:21:52 < Tordek> but yea, just using two counters serves as a solution I guess 2019-10-13T04:22:35 < Thorn> Tordek: may be possible in hardware using ITM and/or DWT http://www2.keil.com/coresight/ 2019-10-13T04:22:38 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-13T04:23:07 < Thorn> but probably only on cortex-m4/m7 2019-10-13T04:24:32 < Tordek> the two counters solutin was good enough; unfortunately it seems my bug isn't there 2019-10-13T04:24:38 < Tordek> thanks 2019-10-13T04:28:10 < Tordek> https://gist.github.com/Tordek/efcabf3f54ab62f8dab8410cf16d6ee2 I have that code making a clock that _should_ be .985mhz (ish, not critical), but it's not giving me a steady square wave 2019-10-13T04:28:20 < Cracki> if you want to measure execution time, inside the handler you can take the cycle count and store its difference in a variable you can watch 2019-10-13T04:28:40 < Cracki> not steady looks like what? 2019-10-13T04:28:58 < Cracki> got any other interrupt handlers enabled? 2019-10-13T04:29:06 < Cracki> you might wanna use a timer to generate signals 2019-10-13T04:29:25 < kakimir64> Tordek: toggle GPIOs at certain points 2019-10-13T04:29:29 < Cracki> and I mean have the timer set gpios 2019-10-13T04:29:38 < kakimir64> connect oscope or logic analyzer 2019-10-13T04:29:46 < kakimir64> see where the jitter happens 2019-10-13T04:29:53 < Cracki> that. 2019-10-13T04:30:09 < Cracki> and you might have to raise priority of that handler 2019-10-13T04:30:15 < kakimir64> Tordek: do you have only one interrupt in use? 2019-10-13T04:31:01 < Tordek> that I set, yes; well, those 2, tim2 and tim3, but the other one is at 50hz 2019-10-13T04:31:26 < Cracki> pics or gtfo 2019-10-13T04:31:32 < Cracki> scope/LA 2019-10-13T04:32:15 < Tordek> I don't have it on hand, but I'll try again later 2019-10-13T04:35:40 < Thorn> >board almost done 2019-10-13T04:35:55 < Thorn> >I should replace linear charger with a switched mode one 2019-10-13T04:36:30 < Thorn> >board is a week away again 2019-10-13T04:36:40 < kakimir64> I wonder if the jitter happens every 20ms 2019-10-13T04:37:09 < Cracki> >_> 2019-10-13T04:37:52 < kakimir64> lets sleep 2019-10-13T04:37:59 < kakimir64> or so 2019-10-13T04:38:01 < kakimir64> > 2019-10-13T04:38:42 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 265 seconds] 2019-10-13T04:40:00 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-13T04:44:20 < Tordek> kakimir64: you mean the 50hz? I'll try disabling that next, but it shouldn't be noticeable against the1MHz clock, should it? 2019-10-13T04:45:23 < Tordek> https://i.imgur.com/0HiYE6d.jpg this is my oscope 2019-10-13T04:45:41 < Tordek> and my freq meter shows like... 115khz? 2019-10-13T04:45:54 < Tordek> heading out, bbl 2019-10-13T04:52:59 < Cracki> divide, is it 8:1 2019-10-13T04:53:52 < Cracki> hm no, 985/115=8.56, no nice round number. you must have fucked up in a more complicated way 2019-10-13T04:54:25 < Tordek> indeed 2019-10-13T04:54:39 < Cracki> how much clock goes to the timer 2019-10-13T04:54:45 < Cracki> I see period 42+1 2019-10-13T04:56:30 < Tordek> 84MHz 2019-10-13T04:57:31 < Cracki> use a period of 41, it counts 0, 1, ..., 41 which is 42 ticks 2019-10-13T04:57:39 < Cracki> if you want exactly 1 or 2 MHz 2019-10-13T04:58:11 < Cracki> 42+1 gives you .977 MHz or 1.953 MHz 2019-10-13T04:58:36 < Tordek> nevermind the exact freq becaus i can ultimately fiddle with it 2019-10-13T04:58:41 < Cracki> right 2019-10-13T04:58:50 < Tordek> the jitter is the issue :/ 2019-10-13T04:59:00 < Cracki> the scope view could be sharper. I see smears 2019-10-13T04:59:31 < Cracki> single shot would be interesting to see 2019-10-13T04:59:52 < Cracki> with time division around 1-10 us 2019-10-13T05:00:40 < Cracki> single shot or LA record 2019-10-13T05:01:32 < Tordek> i'm at dinner, I'll have to try later 2019-10-13T05:03:40 < kakimir32> have you tried turning off that 50hz interrupt? 2019-10-13T05:04:01 < kakimir32> it could be done in mainloop without interrupts 2019-10-13T05:05:10 < kakimir32> if no drift is required then utilize some timer without interrupt 2019-10-13T05:05:26 < Cracki> flip a flag in interrupt so main loop knows when to handle exactly 2019-10-13T05:05:40 < Cracki> same principle as interrupts, except you get another priority :> 2019-10-13T05:06:51 < Tordek> oh, maybe Im handling only one of the ints when both trigger together? 2019-10-13T05:07:26 < Tordek> but ok, I'll verify that when I get back 2019-10-13T05:27:26 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-13T05:35:18 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-13T05:52:13 < jadew> Tordek, nice screenshot 2019-10-13T05:53:01 < jadew> if you toggle the IOs in code, it's very difficult to get rid of the jitter 2019-10-13T05:53:59 < jadew> Cracki, why do you never sleep at this hour? 2019-10-13T05:54:07 < Cracki> shits fucked 2019-10-13T05:54:43 < Cracki> I'll sleep soon, wanna catch the last sun of the year tomorrow 2019-10-13T05:55:06 < jadew> what last sun? 2019-10-13T05:55:46 < Cracki> weather is gonna be overcast forever 2019-10-13T05:56:06 < Cracki> I hate this place, fewer sunshine days than anywhere else 2019-10-13T05:57:27 < jadew> today was sunny here 2019-10-13T05:57:56 < Cracki> romania made a pact with god to keep dracula away 2019-10-13T05:59:16 < jadew> he was actually loved by the people 2019-10-13T05:59:29 < jadew> at least that's what the history books say 2019-10-13T05:59:33 < Cracki> right, he skewered the kebabs 2019-10-13T05:59:48 < jadew> and the thieves 2019-10-13T06:00:50 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-13T06:02:34 < jadew> his family name was literally "The Devil" 2019-10-13T06:03:15 < jadew> talk about predestination... 2019-10-13T06:10:02 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 240 seconds] 2019-10-13T06:13:46 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-13T06:20:04 -!- deltab [~deltab@95.154.230.49] has joined ##stm32 2019-10-13T06:22:40 < jadew> do you guys know of some cheap test clips that don't melt away? 2019-10-13T06:31:25 -!- fc5dc9d4 [~quassel@p57A32AA0.dip0.t-ipconnect.de] has joined ##stm32 2019-10-13T06:31:49 < Cracki> 0.1" or finer? I bought some 0.1" suitable ones but they're plastic and did soften at some temperature 2019-10-13T06:32:16 < Cracki> cheap but durable would imply metal sheath for the pincers 2019-10-13T06:33:06 < jadew> 0.1" would be fine 2019-10-13T06:33:22 < jadew> yeah, I have the plastic kind too and they melt 2019-10-13T06:34:51 -!- fc5dc9d4_ [~quassel@p57A324E8.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-13T06:34:51 < Cracki> I have no idea what those are but they're metal tipped https://www.aliexpress.com/item/4000077479667.html 2019-10-13T06:35:41 < jadew> I want something like this: https://www.aliexpress.com/item/32844943593.html 2019-10-13T06:35:52 < jadew> but not for $50 or some silly amount 2019-10-13T06:36:21 < Cracki> or meltable 2019-10-13T06:36:31 < jadew> yes 2019-10-13T06:36:38 < Cracki> something like that? but _that_ is just a buck a piece 2019-10-13T06:36:39 < jadew> I'd pay $20 for some good ones 2019-10-13T06:36:59 < jadew> no... $20 would seem like a fair price 2019-10-13T06:37:09 < jadew> my DMM leads were $40 I think 2019-10-13T06:37:13 < Cracki> well check those out https://www.aliexpress.com/item/4000077479667.html 2019-10-13T06:37:27 < Cracki> the "instructions" show some fingers for scale 2019-10-13T06:37:40 < Cracki> they appear fully metal so they won't melt right away 2019-10-13T06:37:59 < jadew> yeah, but they look like they could short easily 2019-10-13T06:38:21 < jadew> there have to be some made with silicone or something 2019-10-13T06:38:43 < Cracki> what temperatures do you anticipate 2019-10-13T06:39:37 < jadew> don't know... it's just that I have melted (accidentally) the ones I have several times 2019-10-13T06:40:21 < jadew> can't be more than 150 °C... 2019-10-13T06:41:02 < Cracki> I guess everyone just solders on some magnet wire 2019-10-13T06:41:44 < jadew> well, if I knew it got so hot, maybe I'd do that too, but this happens when I don't expect it 2019-10-13T06:42:01 < Cracki> I'd go the disposable route then :> 2019-10-13T06:42:07 < jadew> sometimes I do, but then the temperature is above the melting point of solder 2019-10-13T06:42:31 < Cracki> k then microwelding 2019-10-13T06:42:59 < Cracki> or metal construction but lacquered for isolation 2019-10-13T06:42:59 < jadew> I just use crocodile connectors for that 2019-10-13T06:42:59 < Tordek> https://i.imgur.com/AmFKHay.jpg ok, I got my dad's osc... 2019-10-13T06:43:10 < Cracki> nice 2019-10-13T06:43:13 < Cracki> I see varying widths 2019-10-13T06:43:18 < Tordek> yeh 2019-10-13T06:43:57 -!- learningc [~pi@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-13T06:44:00 < Cracki> so this means some interrupt delays *that* one by 1-3 microseconds, eyeballing it here 2019-10-13T06:44:14 < Tordek> also 119khz 2019-10-13T06:44:18 < Cracki> tell the nvic to give it higher prio 2019-10-13T06:44:20 < Tordek> instead of 985 2019-10-13T06:44:42 < Cracki> verify clock tree, then verify timer period 2019-10-13T06:44:54 < Cracki> it was 84 MHz? 2019-10-13T06:45:37 < Tordek> https://i.imgur.com/uSZA9rK.png 2019-10-13T06:45:58 < Tordek> and period is 42, should be 1mhz, right 2019-10-13T06:46:07 < Cracki> hm 2019-10-13T06:46:16 < Cracki> may not 2019-10-13T06:46:39 < Cracki> oh, apb2 timer clocks is 168 mhz 2019-10-13T06:47:08 < Cracki> I'm not getting calculations making accidental sense 2019-10-13T06:47:10 < Tordek> but tim2/tim3 are apb1? 2019-10-13T06:48:08 < Cracki> 119e3 * 2 * 42 is ~ 10 MHz 2019-10-13T06:48:15 -!- learningc [~pi@121.121.98.53] has joined ##stm32 2019-10-13T06:48:25 < Cracki> that's the roundest I can come up with 2019-10-13T06:49:22 < Cracki> that i2c clock is severe 2019-10-13T06:49:25 < Cracki> x192 2019-10-13T06:49:33 < Cracki> but beside the point 2019-10-13T06:50:00 < Cracki> uncouple all that off your timer, check that it works right and you understand the basic operation 2019-10-13T06:50:31 < Cracki> set up output compare for a dumb function test 2019-10-13T06:51:05 < Cracki> wait... why is your signal not having a duty cycle of 50:50? 2019-10-13T06:51:10 < Cracki> it should be toggling! 2019-10-13T06:51:46 < Cracki> unless I have taken too brief a glance at your code 2019-10-13T06:52:29 < Tordek> yeah, the duty cycle is wrong too 2019-10-13T06:52:43 < Tordek> even though that pin is only set/unset through that clock 2019-10-13T06:52:50 < Tordek> like, at that interrupt 2019-10-13T06:53:25 < Cracki> check that the timer's counter is reloaded right 2019-10-13T06:53:38 < Tordek> unless... interrupts are queued? i.e., it's taking too long to handle step6502() and the "off" interrupt happens at the wrong time? 2019-10-13T06:53:51 < Cracki> that is likely 2019-10-13T06:54:03 < Cracki> strip everything out of that interrupt, then add in piece by piece 2019-10-13T06:54:17 < Tordek> kk 2019-10-13T06:54:38 < Cracki> always break search space/problem up, "clear the house room by room" 2019-10-13T06:55:02 < Cracki> throwing bacon on the roof isn't gonna make the bug come out 2019-10-13T06:58:49 < Cracki> what does that step6502() do? I assumed it only toggles some other clock line to an actual 6502 2019-10-13T06:59:03 < Cracki> or does it run a whole bunch of code... 2019-10-13T07:19:23 < Tordek> it does run a bunch of code 2019-10-13T07:19:33 < Tordek> it's a 6502 emu 2019-10-13T07:20:21 < Cracki> consider doing that in the main loop 2019-10-13T07:20:35 < Cracki> and just set a flag in the handler so the main loop knows to run it 2019-10-13T07:21:09 < Cracki> that, or upgrade to an RTOS and give that part of the code its own task that sleeps until deadline, then deadline += 50ms 2019-10-13T07:22:14 < Cracki> freertos and I'm sure others have apis where you put in the last deadline, the interval, and it makes sure that even if multiple interval times have passed, it's not gonna "catch up" but keep a steady pace 2019-10-13T07:24:13 < jadew> look what I found: https://www.aliexpress.com/item/32913653633.html 2019-10-13T07:24:39 < jadew> it's a korad tho 2019-10-13T07:24:46 < jadew> so it's probably shit 2019-10-13T07:26:22 -!- qyx_ [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-13T07:26:22 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-13T07:40:48 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 268 seconds] 2019-10-13T07:41:32 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-13T07:49:03 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-13T07:50:11 < Cracki> pressure washer porn https://www.youtube.com/watch?v=t6o7m-sCwBk 2019-10-13T07:51:57 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-13T07:51:57 -!- day__ is now known as day 2019-10-13T07:56:34 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-13T08:11:34 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-13T08:33:48 < Tordek> welp, after the simplification the clock signal became stable... at 186khz 2019-10-13T08:34:36 < Tordek> which, times 43 (the autoreload value) is 8MHz... so, alright, you'd think I just misread the clock source and it's 8mhz not 48 2019-10-13T08:34:58 < Tordek> but I set the AR value to 7 and the clock remains at 186 2019-10-13T08:35:04 < Tordek> so, 1: what the fuck 2019-10-13T08:35:10 < Tordek> 2: back to the drawing board 2019-10-13T09:07:22 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-13T09:14:27 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-13T09:42:46 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-13T10:03:40 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-13T10:09:18 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Ping timeout: 265 seconds] 2019-10-13T10:09:35 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-13T10:52:37 -!- qyx_ [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-13T10:57:22 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-13T11:00:13 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-13T11:02:23 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-13T11:05:38 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-13T11:05:39 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-13T11:06:42 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-13T11:07:23 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-13T11:09:39 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-13T11:12:23 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-13T11:38:35 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has joined ##stm32 2019-10-13T11:39:01 -!- renn0xtk9 [~max@2a02:810d:1540:2448:84ca:776d:822f:e933] has quit [Client Quit] 2019-10-13T11:40:36 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c55e:f2ce:57de:5521] has joined ##stm32 2019-10-13T11:50:17 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-13T11:54:21 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c55e:f2ce:57de:5521] has quit [Quit: Konversation terminated!] 2019-10-13T12:37:03 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-13T13:11:28 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-13T13:13:38 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 245 seconds] 2019-10-13T13:13:38 -!- BrainDamage_ is now known as BrainDamage 2019-10-13T13:30:50 < Tordek> also if I set the counter to 84 million I do get a led to blink at 1hz, so yea, the plot chickens 2019-10-13T13:32:18 < Steffanx> dumdum dumdum dumdumdum 2019-10-13T13:35:09 < Steffanx> John Williams did a better job 2019-10-13T13:47:27 < catphish> what's the purpose of PWM mode 2? it looks to me just to be a reverse polarity of PWM mode 1, am i missing someghing? 2019-10-13T13:53:38 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 276 seconds] 2019-10-13T13:55:09 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-13T14:28:11 -!- User__ [~learningc@121.121.98.53] has joined ##stm32 2019-10-13T14:35:39 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds] 2019-10-13T14:37:02 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2019-10-13T15:11:46 -!- learningc [~pi@121.121.98.53] has quit [Quit: WeeChat 2.6] 2019-10-13T15:12:42 -!- learningc [~pi@121.121.98.53] has joined ##stm32 2019-10-13T15:19:07 -!- User__ [~learningc@121.121.98.53] has quit [Ping timeout: 265 seconds] 2019-10-13T15:20:13 -!- learningc [~pi@121.121.98.53] has quit [Ping timeout: 268 seconds] 2019-10-13T15:21:48 -!- learningc [~learningc@121.121.98.53] has joined ##stm32 2019-10-13T15:22:01 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 265 seconds] 2019-10-13T15:24:30 -!- learning1 [~pi@121.121.98.53] has joined ##stm32 2019-10-13T16:28:56 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-13T16:34:02 -!- learning1 [~pi@121.121.98.53] has quit [Quit: WeeChat 2.6] 2019-10-13T16:42:37 -!- learning1 [~pi@121.121.98.53] has joined ##stm32 2019-10-13T16:42:44 -!- learningc [~learningc@121.121.98.53] has quit [Quit: Leaving] 2019-10-13T16:45:14 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-13T16:45:23 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-13T17:04:27 -!- Jan- [~IceChat9@87.114.135.182] has quit [Ping timeout: 240 seconds] 2019-10-13T17:11:15 -!- Jan- [~IceChat9@87.114.135.182] has joined ##stm32 2019-10-13T17:22:17 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-13T17:25:10 < jadew> what's the hole tolerance for through hole pads at reputable PCB houses? 2019-10-13T17:27:12 < jadew> karlp, how were your ailser boards? 2019-10-13T17:30:02 < Jan-> is there any way to program stm32 boards other than cubemx or platformio 2019-10-13T17:41:06 < specing> most stm32 board you buy from STM have built-in programmer/debugger 2019-10-13T17:41:23 < Jan-> I'm thinking of the software 2019-10-13T17:41:25 < Jan-> not the hardware 2019-10-13T17:41:26 < specing> ah 2019-10-13T17:41:40 < veverak> wat 2019-10-13T17:41:45 < veverak> Jan-: you make binary and flash it 2019-10-13T17:41:45 < specing> you can use GPS + GNAT + bb-runtimes + Ada_Drivers_Library 2019-10-13T17:41:49 < Jan-> I can't make platformio work and the hardware access layer in cube is horrific. 2019-10-13T17:41:51 < veverak> that is not related to cubemx/platformio 2019-10-13T17:41:58 < Jan-> so I'm thinking what else would there be if anything 2019-10-13T17:42:33 < veverak> you just need to fill registers at proper memory addresses 2019-10-13T17:42:35 < veverak> ;) 2019-10-13T17:44:24 < Jan-> no kidding 2019-10-13T17:44:48 < specing> Jan-: you can program stm32s in Ada instead of C :) 2019-10-13T17:44:55 < Jan-> I don't mind C 2019-10-13T17:45:16 < specing> then enjoy the horrors that come with it :) 2019-10-13T17:46:18 < Jan-> I just don't like having to type 50 characters to set a pin 2019-10-13T17:46:53 < specing> GPIOC.BSRR.BS.Arr(9) := 1; 2019-10-13T17:47:01 < specing> This is too many for you? 2019-10-13T17:47:21 < Jan-> what about PORTB &= (1 << n); 2019-10-13T17:48:25 < jpa-> jadew: you mean for the holes? 2019-10-13T17:48:38 < specing> jpa-: GPIOC.BSRR.BR.Arr(n) := 1; 2019-10-13T17:48:42 < specing> Jan-: GPIOC.BSRR.BR.Arr(n) := 1; 2019-10-13T17:48:51 < specing> Jan-: GPIOB.BSRR.BR.Arr(n) := 1; 2019-10-13T17:49:01 < jpa-> jadew: jlcpcb specifies +-0.08mm for hole size tolerance 2019-10-13T17:49:57 < Jan-> specing: once more in english? 2019-10-13T17:50:25 < specing> Jan-: the last line, GPIOB.BSRR.BR.Arr(n) := 1; 2019-10-13T17:50:37 < Jan-> oh. 2019-10-13T17:50:40 < Jan-> I have no idea what that means. 2019-10-13T17:50:42 < specing> BR is the bit-reset register, check datasheet 2019-10-13T17:50:53 < specing> for atomic access to ports 2019-10-13T17:51:03 < Jan-> what's wrong with PORTB &= (1< Jan-: a) it looks ugly, b) it does a read-modify-write 2019-10-13T17:51:35 < Jan-> it has to do that whatever 2019-10-13T17:51:56 < specing> if the hardware supports atomic pin access, then it doesen't have to 2019-10-13T17:52:08 < specing> and stm32 does support that 2019-10-13T17:53:10 < Jan-> well ok. 2019-10-13T17:53:16 < Jan-> but that doesn't solve my problem right now. 2019-10-13T17:53:18 * Jan- headscratch 2019-10-13T17:55:00 < Jan-> can you use some other hardware access layer with cube? 2019-10-13T18:02:28 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-13T18:06:25 < qyx> Jan-: do you !mean some other environments? 2019-10-13T18:06:52 < qyx> zyp has nice laks library for hw access 2019-10-13T18:06:59 < qyx> there is also libopencm3 2019-10-13T18:07:09 < qyx> you can start with libopencm3-examples 2019-10-13T18:07:22 < qyx> you can flash the chip with openocd for example 2019-10-13T18:07:31 < qyx> or jlink or whatever link 2019-10-13T18:07:40 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-13T18:07:42 < qyx> or with the builtin bootloader 2019-10-13T18:11:18 <@englishman> doesn't f4 still have stdperiphlib 2019-10-13T18:18:46 < Jan-> I think openocd is the problem 2019-10-13T18:18:54 < Jan-> is that the flash utility 2019-10-13T18:27:35 < Jan-> ok yeah that's the problem 2019-10-13T18:27:47 < Jan-> the utility that probes for st-link programmers detects mine OK 2019-10-13T18:27:52 < Jan-> but openocd can't find it 2019-10-13T18:30:18 < PaulFertser> Jan-: what's the actual error? 2019-10-13T18:30:33 < PaulFertser> Jan-: and what are the openocd command line arguments or config? 2019-10-13T18:36:57 < jadew> jpa-, ah, that makes sense 2019-10-13T18:37:10 < jadew> I measured 0.8mm on a 0.7mm hole 2019-10-13T18:38:51 < PaulFertser> Jan-: most common error with openocd not detecting stlink is due to v2/v2-1 mismatch. Got me several times (enough to make me fix autodetection but you're probably using old release, not current git master). 2019-10-13T18:44:01 < Jan-> when I used it with cube it did want to update the firmware 2019-10-13T18:44:08 < Jan-> but I don't know what the command line is as I'm using platformio 2019-10-13T18:47:11 < PaulFertser> Jan-: -f interface/stlink-v2-1.cfg -f target/stm32f4x.cfg -c "program path/to/your.elf verify reset exit" 2019-10-13T18:47:17 < PaulFertser> (amend to your liking) 2019-10-13T18:47:43 < Jan-> I don't know what almost any of those things mean. 2019-10-13T18:47:58 < Jan-> I don't have any choice anyway 2019-10-13T18:48:04 < PaulFertser> OpenOCD ships with README and an (almost) extensive official user manual. 2019-10-13T18:48:15 < Jan-> I don't specifically have openocd, I have something called platformio 2019-10-13T18:48:27 < Jan-> to make it build and run I have to type "pio run something something" 2019-10-13T18:48:31 < Jan-> can't even remember what now 2019-10-13T18:48:38 < PaulFertser> You said openocd can't find it. I'm asking about the details trying to help you. 2019-10-13T18:48:47 < Jan-> yeah hang on I'm looking 2019-10-13T18:50:22 < Thorn> what is "NVDC-1" in TI (and other) battery chargers 2019-10-13T18:50:22 < Jan-> this is what I get https://pastebin.com/kANMJExv 2019-10-13T18:51:15 < Jan-> if it's not fixable don't worry I don't want it to be a big deal 2019-10-13T18:51:20 < Jan-> there are things other than stm32 etc 2019-10-13T18:53:05 < catphish> Jan-: i program with either st-flash (for SWD) or dfu-util (for USB DFU) on linux 2019-10-13T18:53:06 < PaulFertser> Jan-: open failed means it doesn't see stlink. It might be because your config tells it to look for v2 when you have v2-1 or vice versa. 2019-10-13T18:53:56 < Jan-> if I knew how to type an ascii shrug smiley, I would 2019-10-13T18:54:01 < Jan-> is that fixable 2019-10-13T18:56:26 < PaulFertser> I'm not a platformio user, so I have no idea how exactly to do that but I would bet it's easily fixable, yes. 2019-10-13T18:58:04 < Jan-> has this happened because I let cube update the st-link 2019-10-13T18:58:48 < PaulFertser> Hm, might be but I doubt it changed the USB PID with new firmware. 2019-10-13T18:58:58 < Jan-> it didn't I checked 2019-10-13T18:59:22 < PaulFertser> I'd need to see openocd config or command line and to know the current PID to tell how to fix that. 2019-10-13T18:59:49 < Jan-> right now it says USB\Vid_0483&Pid_3748&Rev_0100 2019-10-13T19:01:35 < srk> or reflash stlink with blackmagic and use that 2019-10-13T19:01:54 < Jan-> whatever works dudes 2019-10-13T19:03:18 < srk> choose your poison https://wiki.base48.cz/STM32#Debugging 2019-10-13T19:03:28 < PaulFertser> Jan-: 3748 is regular v2, not v1. It's strange you get old v2 on nucleo. 2019-10-13T19:03:36 < PaulFertser> v2, not v2-1 2019-10-13T19:03:46 < PaulFertser> Jan-: so with that pid you need to choose v2 as the debug adapter. 2019-10-13T19:04:02 < Jan-> er 2019-10-13T19:04:16 < PaulFertser> Probably you downgraded its firmware from v2-1 to v2? 2019-10-13T19:04:58 < Jan-> I uh. 2019-10-13T19:04:59 < Jan-> er. 2019-10-13T19:05:02 < Jan-> I uhoh. 2019-10-13T19:06:16 < Jan-> why would the official ide downgrade... 2019-10-13T19:08:43 < PaulFertser> Why do not you run openocd manually to check? 2019-10-13T19:08:47 < PaulFertser> I gave you full command line. 2019-10-13T19:09:00 < Jan-> well sure but I don't have the elf file 2019-10-13T19:09:44 < PaulFertser> What's .pio\build\black_f407ve\firmware.elf then? 2019-10-13T19:09:54 < PaulFertser> Is it getting auto-erased after a flashing attempt? 2019-10-13T19:10:38 < Jan-> no 2019-10-13T19:11:35 < Jan-> now I need to find openocd 2019-10-13T19:11:41 < Jan-> which I assume is around somewhere 2019-10-13T19:16:12 < Jan-> the last line says: invalid command name ".piuillack_f407ve♀irmware.elf" 2019-10-13T19:17:11 < PaulFertser> Looks like lack of slashes 2019-10-13T19:17:27 < Jan-> no it wants slashes the wrong way 2019-10-13T19:17:32 < PaulFertser> Are you using backslashes as path separators? Then you need to escape them. 2019-10-13T19:18:12 < Jan-> now it just says invalid command name ".pio/build/black_f407ve/firmware.elf" 2019-10-13T19:20:38 < Jan-> did screw up the format of the command or something 2019-10-13T19:20:53 < PaulFertser> Please paste the full output along with the invocation command line. 2019-10-13T19:22:30 < Jan-> https://pastebin.com/kE4xazmD 2019-10-13T19:23:40 < PaulFertser> Jan-: you left out "program" before the path 2019-10-13T19:25:47 < Jan-> OK now I get basically the same failure that platformio pushed through 2019-10-13T19:26:27 < Jan-> Info : Clock speed 2000 kHzError : open failedin procedure 'program' ** OpenOCD init failed **shutdown command invoked 2019-10-13T19:27:03 < PaulFertser> Jan-: now that I know that you're using a new enough version we can proceed. 2019-10-13T19:27:45 < PaulFertser> Jan-: I suggest to reinstall WinUSB for the stlink device with Zadig (since you're using a weird non-standard OS that doesn't have an API to access USB devices from userspace without tricks) 2019-10-13T19:28:28 < Jan-> sorry? 2019-10-13T19:28:37 < Jan-> I'm not sure I've ever installed anything called winusb. 2019-10-13T19:30:34 < PaulFertser> stlink software installs it automatically but somehow the version it installs doesn't always work with libusb (used by OpenOCD to access hw interfaces). 2019-10-13T19:32:34 < PaulFertser> OpenOCD version you run is new enough that v2/v2-1 differences are not important, it autodetects the device properly so that's not an issue. 2019-10-13T19:34:40 < Jan-> well all I can do is run a search for "winusb" on c: and see if anything pops up 2019-10-13T19:35:06 < PaulFertser> Jan-: no, I'm suggesting to download Zadig from the internet, to run it, to make it install WinUSB driver on stlink device. 2019-10-13T19:35:31 < Jan-> like this zadig? https://zadig.akeo.ie/ 2019-10-13T19:36:09 < PaulFertser> Yes 2019-10-13T19:36:18 < Jan-> ok got that 2019-10-13T19:36:33 < Jan-> am I creating new device 2019-10-13T19:37:48 < Jan-> there's nothing in the dropdown box, this feels wrong 2019-10-13T19:37:59 < Jan-> oh ok "list all devices" 2019-10-13T19:38:31 < PaulFertser> Yes, as it already has some driver bound to it. But it needs to be reinstalled with WinUSB as offered by Zadig. 2019-10-13T19:38:37 < Jan-> I can install driver, upgrade driver or extract files 2019-10-13T19:39:34 < PaulFertser> Install is appropriate iirc 2019-10-13T19:40:06 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Remote host closed the connection] 2019-10-13T19:40:19 < Jan-> mmm I think it's just locked up it's gone all "not responding" 2019-10-13T19:40:43 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-13T19:41:03 < Jan-> the existing driver is "WinUSB (v2.1.0.0)" and I assume it's offering me "WinUSB v6.1.7600.16385" 2019-10-13T19:41:57 < Jan-> oh okay fine installed successfully 2019-10-13T19:42:05 < PaulFertser> Now please rerun openocd 2019-10-13T19:42:13 < PaulFertser> And pastebin the output. 2019-10-13T19:42:17 < Jan-> Info : Clock speed 2000 kHzError : open failedin procedure 'program' ** OpenOCD init failed **shutdown command invoked 2019-10-13T19:43:14 < Jan-> https://pastebin.com/WgUMk3rg 2019-10-13T19:43:17 < PaulFertser> Hm, still open failed? It certainly means it can't talk to the debug adapter... 2019-10-13T19:44:39 < PaulFertser> It just can't talk to the adapter, doesn't even detect it even though it should have the list of vid/pids. 2019-10-13T19:44:53 < kakimir32> cannot really tell without -d3 2019-10-13T19:44:55 < PaulFertser> You can see it's trying different pids with -d3 2019-10-13T19:45:05 < Jan-> there is some utility that lists all the stlink adapters 2019-10-13T19:45:08 < PaulFertser> Right kakimir32, that should help 2019-10-13T19:45:09 < Jan-> but I can't remember what it's called 2019-10-13T19:45:42 < Jan-> it scans for them and gives you the firmware version and serial number 2019-10-13T19:46:18 < jadew> I think google chrome stopped showing results from history so you get to search for them 2019-10-13T19:47:02 < PaulFertser> Jan-: please rerun with -d3 appended to the parameters 2019-10-13T19:48:18 < Jan-> ok so I think that's more output than will fit in the cmd window 2019-10-13T19:48:29 < BrainDamage> jadew: in the url bar? you can get them back using extensions 2019-10-13T19:48:45 < jadew> BrainDamage, still... kind of a dick move on google 2019-10-13T19:48:57 < BrainDamage> what's more concerning is that google pulled off ublock origin from the store, and put it back only massive outcries 2019-10-13T19:49:12 < PaulFertser> Jan-: last 100 lines should be enough 2019-10-13T19:49:25 < PaulFertser> Where stlink_usb_open() is mentioned 2019-10-13T19:49:34 < BrainDamage> jadew: https://github.com/uBlockOrigin/uBlock-issues/issues/745 2019-10-13T19:49:36 < jadew> BrainDamage, really? didn't know about that 2019-10-13T19:50:00 < Jan-> https://pastebin.com/jSSqvSaf 2019-10-13T19:50:39 < PaulFertser> Jan-: you see, 0x3748 is mentioned there. 2019-10-13T19:51:14 < Jan-> okay 2019-10-13T19:51:28 < PaulFertser> Zadig should have installed it properly. But I've heard some windows version require you to enable some additional option in zadig menu to show "hidden" devices and to install WinUSB for the composite device parent. 2019-10-13T19:52:52 < jadew> BrainDamage, is that what you're using for ad blocking? 2019-10-13T19:53:07 < BrainDamage> that, among other tools 2019-10-13T19:53:13 < Jan-> I can uncheck "Ignore hubs or composite parents" 2019-10-13T19:54:14 < Jan-> that reveals a bunch of stuff 2019-10-13T19:54:36 < Jan-> but I don't want to start randomly installing drivers here 2019-10-13T19:56:01 < PaulFertser> Jan-: I'm not talking about randomly, I'm talking about stlink composite device parent. It should have that vid/pid you know. 2019-10-13T19:56:34 < Jan-> so I'm looking for other product id 3748? 2019-10-13T19:57:08 < Jan-> all I really have are the root hubs on the motherboard, the keyboard and mouse, and the stlink 2019-10-13T19:57:26 < Jan-> at least I assume it's the mouse it's a "receiver" 2019-10-13T19:59:02 < Jan-> I guess we're at the point of giving up on this? 2019-10-13T20:01:44 < jadew> BrainDamage, I like it, I think I'll use that instead of adblock plus 2019-10-13T20:03:15 < jadew> wow, it actually works better 2019-10-13T20:03:32 < Jan-> cripes 2019-10-13T20:03:35 < Jan-> is your name actually jade 2019-10-13T20:03:51 < jadew> no, my nickname is based on a movie 2019-10-13T20:03:59 < Jan-> d'oh 2019-10-13T20:04:55 < jadew> https://www.imdb.com/title/tt0416871/reference 2019-10-13T20:05:20 < jadew> it's about a love story that transcends time and death 2019-10-13T20:05:43 < Jan-> I was hoping it'd be about not being the only girl in 145 users :( 2019-10-13T20:06:15 < jadew> you might not be 2019-10-13T20:06:35 < Jan-> I'm not normally very political about this but jesus internet you male-dominated 2019-10-13T20:07:04 < jadew> you know males... we like to dominate 2019-10-13T20:07:08 < specing> :D 2019-10-13T20:07:41 < catphish> oh hey Jan-, are you normally here, or are you visiting from ##electronics? 2019-10-13T20:09:03 < PaulFertser> Jan-: so there's just one device, its PID/VID matches what OpenOCD prints in -d3 log, it's connected, WinUSB installed for it, and yet openocd gives "open failed"? 2019-10-13T20:09:29 < Tordek> is therer a way to count cycles between two breakpoints in stm32cubeide? 2019-10-13T20:09:29 < Jan-> catphish: well visiting but you know. doing some stm32 stuff. 2019-10-13T20:09:42 < Jan-> PaulFertser: I think so? 2019-10-13T20:09:55 < Jan-> let me re run zadig 2019-10-13T20:09:57 < catphish> Jan-: good choice :) people here are weird but they know stm32 :) 2019-10-13T20:09:59 < Tordek> or at least a general cycle count I can take the difference for 2019-10-13T20:10:12 < Jan-> winusb 6.1.7600.16385 2019-10-13T20:10:16 < Jan-> stmicroelectronics stlink dongle 2019-10-13T20:10:21 < catphish> then stop using windows :) 2019-10-13T20:10:24 < Jan-> 0483 3748 2019-10-13T20:11:26 < PaulFertser> Jan-: hm, is it possible some windows update happened between you had it working and now? 2019-10-13T20:11:36 < Jan-> pretty sure not 2019-10-13T20:11:40 < Jan-> it worked in cube mx 2019-10-13T20:11:41 < PaulFertser> I just can't explain how it's possible to get what you're seeing. 2019-10-13T20:11:46 < mawk> in cube IDE* 2019-10-13T20:11:53 < mawk> cube mx doesn't flash, cube mx is just the configurator tool 2019-10-13T20:11:56 < Jan-> people call it various things 2019-10-13T20:12:10 < Jan-> that thing 2019-10-13T20:12:13 < Jan-> the official IDE 2019-10-13T20:12:16 < mawk> yes 2019-10-13T20:12:20 < mawk> STM32CubeIDE 2019-10-13T20:12:40 < mawk> all their stuff starts with STM32Cube, STM32Cube is the peripheral library, STM32CubeIDE is the IDE, STM32CubeMX is the configurator 2019-10-13T20:12:55 < Jan-> I call him Mister C:\ST\STM32CubeIDE_1.0.2\STM32CubeIDE\stm32cubeide.exe 2019-10-13T20:13:00 < mawk> lol 2019-10-13T20:13:03 < mawk> but you're on windows Jan- 2019-10-13T20:13:13 < mawk> did you know PaulFertser ? 2019-10-13T20:13:13 < Jan-> yeah because it's sane 2019-10-13T20:13:15 < mawk> lol 2019-10-13T20:13:19 < catphish> is it ok to go to dutchland without being able to speak a word of dutch? 2019-10-13T20:13:23 < mawk> me I call it /opt/st/stm32cubeide_1.0.0/stm32cubeide 2019-10-13T20:13:27 < mawk> yes catphish 2019-10-13T20:13:30 < mawk> they all speak good english 2019-10-13T20:13:31 < Jan-> catphish: everyone in amsterdam speaks better english than most english people. 2019-10-13T20:13:33 < PaulFertser> mawk: did I know what? 2019-10-13T20:13:38 < mawk> that Jan- is on windows 2019-10-13T20:13:46 < catphish> Jan-: yeah i'm just going to amsterdam, i won't worry :) 2019-10-13T20:13:50 < mawk> you're going to live there catphish ? 2019-10-13T20:13:52 < mawk> ah, just vacation 2019-10-13T20:14:03 < catphish> just a week :) 2019-10-13T20:14:04 < mawk> coffeeshop is "coffeeshop" in dutch don't worry 2019-10-13T20:14:04 < PaulFertser> mawk: ah, yes, I wouldn't be suggesting installing WinUSB with Zadig otherwise. Was pretty obvious from backslashes used as path separators too. 2019-10-13T20:14:24 < catphish> doesn't coffeeshop mean something different there? 2019-10-13T20:14:32 < mawk> well you can get coffee in them too 2019-10-13T20:14:37 < catphish> ieal 2019-10-13T20:14:40 < Jan-> go to rembrandtplein and the coffeeshop smokey :D 2019-10-13T20:14:40 < catphish> *ideal 2019-10-13T20:14:51 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 240 seconds] 2019-10-13T20:14:54 < mawk> I'm going to live there, catphish 2019-10-13T20:14:59 < mawk> not in amsterdam tho, I'm not that rich 2019-10-13T20:15:33 < catphish> mawk: cool, i always thought BE and NL seemed like really nice places 2019-10-13T20:15:40 < catphish> haven't been in a while 2019-10-13T20:15:51 < mawk> yeah 2019-10-13T20:15:53 < mawk> don't forget LU 2019-10-13T20:15:56 < mawk> the holy trinity 2019-10-13T20:16:01 < Jan-> the verbose output doesn't really tell us much more than the not so verbose output 2019-10-13T20:16:02 < catphish> this time just taking mrs catphish to AMS 2019-10-13T20:16:05 < Jan-> it lists all the devices then says "nup" 2019-10-13T20:16:53 < PaulFertser> But you can be sure it tries those vid/pids 2019-10-13T20:17:01 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-13T20:17:47 < Jan-> I don't know anything about openocd 2019-10-13T20:17:53 < Jan-> I'm completely helpless here 2019-10-13T20:18:25 < mawk> you're in luck Jan- , PaulFertser knows a bit about it 2019-10-13T20:18:44 < mawk> don't abandon stm32 for a few petty difficulties 2019-10-13T20:19:07 < Jan-> I'm just really aware of soaking up time 2019-10-13T20:19:25 < Jan-> to be honest I'm starting to just not have any confidence in it now 2019-10-13T20:19:32 < Jan-> even if we fix this tomorrow it can break again 2019-10-13T20:19:32 < PaulFertser> mawk: but I have no idea how to diagnose this apparently windows-specific issue. Someone else should chime in. 2019-10-13T20:19:37 < mawk> well I don't know how you managed to break it Jan- honestly 2019-10-13T20:19:51 < mawk> I'm on linux since I'm 9, and one time I had to set up openocd on windows 2019-10-13T20:20:01 < mawk> I built the latest git snapshot, ran it, perfect from the first time 2019-10-13T20:20:11 < mawk> under MSYS 2019-10-13T20:21:18 < Jan-> I broke it? I didn't even know I had it 2019-10-13T20:21:28 < Jan-> please don't do the standard open source "blame the user" thing 2019-10-13T20:21:31 < mawk> lol 2019-10-13T20:21:34 < mawk> it was just a joke 2019-10-13T20:21:46 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 268 seconds] 2019-10-13T20:21:59 < mawk> could you recap where you're at now ? still using openocd through platformio ? which programmer you have ? 2019-10-13T20:22:10 < Jan-> it's an stlink or presumably a chinese clone thereof 2019-10-13T20:22:14 < mawk> if yes to the first question I'd advise trying standalone openocd 2019-10-13T20:22:16 < mawk> ok good 2019-10-13T20:22:22 < mawk> yeah if you bought it for $2 it's a clone 2019-10-13T20:22:37 < PaulFertser> mawk: https://pastebin.com/jSSqvSaf 2019-10-13T20:23:03 < PaulFertser> It's apparently stlink-v2 2019-10-13T20:23:19 < PaulFertser> And Jan- says vid/pid match the one expected by OpenOCD. 2019-10-13T20:23:39 < PaulFertser> Jan-: probably it'd be the right idea to reboot windows? Used to help with many issues in win95 times. 2019-10-13T20:25:05 < Jan-> that was a while go 2019-10-13T20:25:12 < Jan-> I can try plugging the thing into a different usb port? 2019-10-13T20:25:22 < mawk> it's the only stlink you have ? 2019-10-13T20:25:27 < mawk> are you 100% sure of the wiring ? 2019-10-13T20:25:33 < mawk> aren't the cables wiggly ? 2019-10-13T20:25:34 < Jan-> yes it worked in cube 2019-10-13T20:25:36 < mawk> ah 2019-10-13T20:25:42 < Jan-> anyway wouldn't that be a idfferent error 2019-10-13T20:25:42 < mawk> and you didn't knock it off or spilled coffee on it since 2019-10-13T20:25:45 < PaulFertser> Jan-: I guess , and windows usually installs drivers per-port so it might make a difference 2019-10-13T20:25:52 < mawk> yeah probably 2019-10-13T20:25:53 < PaulFertser> Jan-: also try using a non-usb-3.0 port. 2019-10-13T20:26:07 < PaulFertser> mawk: open failed means libusb can't see the device. 2019-10-13T20:26:08 < antto> muh latest innovation: https://i.imgur.com/VMf46L5.png 2019-10-13T20:26:34 < mawk> lol antto 2019-10-13T20:26:38 < BrainDamage> literally fitting square pegs in round holes 2019-10-13T20:26:44 < antto> now now 2019-10-13T20:26:44 < Jan-> err 2019-10-13T20:26:50 < Jan-> ok I think 2019-10-13T20:26:53 < Jan-> er did that work? 2019-10-13T20:26:55 < antto> ignore the model, imma use precidip headers 2019-10-13T20:27:11 < mawk> if openocd is waiting for commands and not failed then it works Jan- 2019-10-13T20:27:15 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-13T20:27:16 < mawk> it's waiting for commands on the telnet port 2019-10-13T20:27:23 < mawk> depending on what you told it 2019-10-13T20:27:59 < mawk> it also opened a remote gdb port for use with eclipse or whatever, or CLI gdb 2019-10-13T20:28:12 < Jan-> is this success: https://pastebin.com/qGgjkf46 2019-10-13T20:28:40 < mawk> I guess 2019-10-13T20:28:44 < mawk> can you remind me of the command line ? 2019-10-13T20:28:51 < mawk> how did you call openocd 2019-10-13T20:28:53 < PaulFertser> Jan-: looks like it, you can rerun without -d3 2019-10-13T20:29:07 < PaulFertser> Jan-: so was it in a usb-3.0 port? Was it in it when it was working before? 2019-10-13T20:29:38 < Jan-> I honestly don't know 2019-10-13T20:29:42 < Jan-> there's two usb extender cables on the desk here 2019-10-13T20:29:45 < Jan-> I just tried the other 2019-10-13T20:30:41 < PaulFertser> Probably related then: https://marc.info/?l=libusb-devel&m=130934450111495&w=2 2019-10-13T20:31:10 < PaulFertser> Mentioned here: https://github.com/libusb/libusb/wiki/Windows 2019-10-13T20:31:29 < mawk> DNS_PROBE_FINISHED_NXDOMAIN :( my dns resolver is behaving again 2019-10-13T20:31:59 < PaulFertser> Jan-: trying a different usb port was a nice idea indeed. I've forgotten about that usb-3.0 quirk with certain windows drivers. 2019-10-13T20:33:18 < BrainDamage> hell will freeze over before web mailing list viewers will present a nice interface 2019-10-13T20:33:39 < mawk> oh you can resolve it BrainDamage ? 2019-10-13T20:33:42 < BrainDamage> why can't they just show a sidebar with the tree of the conversation 2019-10-13T20:33:44 < BrainDamage> yes, I can 2019-10-13T20:33:46 < mawk> I'm curious why this domain never works with me 2019-10-13T20:35:34 < mawk> ok, I can't resolve ns[35].korelogic.com which is the authority for marc.info 2019-10-13T20:35:36 < BrainDamage> drill marc.ifo | ix-> http://ix.io/1YwM 2019-10-13T20:35:41 < mawk> my unbound says SERVFAIL 2019-10-13T20:36:03 < BrainDamage> try to reach the ip 2019-10-13T20:36:37 < mawk> ah 2019-10-13T20:36:41 < mawk> the ip doesn't answer to me 2019-10-13T20:36:53 < mawk> if I use my home ip address it works 2019-10-13T20:37:09 < mawk> what a broken config that "korelogic.com" has, probably ip-racism 2019-10-13T20:37:12 < mawk> since I have a server ip block 2019-10-13T20:37:52 < antto> ur IP address, if interpreted as a color, looks too black? 2019-10-13T20:37:57 < mawk> yes 2019-10-13T20:38:00 < antto> aww 2019-10-13T20:38:02 < mawk> many sites block me, or captcha me 2019-10-13T20:38:11 < antto> racism confirmed 2019-10-13T20:38:23 < BrainDamage> the 4 blocks of the ip range are RGBA 2019-10-13T20:38:30 < mawk> let me see the color of that 2019-10-13T20:38:44 < antto> or you're too transparent ;P~ 2019-10-13T20:39:12 < antto> or maybe they only let a certain hue in 2019-10-13T20:39:32 < mawk> I'm a nice blue 2019-10-13T20:39:42 < antto> aww, i would block blue ones 2019-10-13T20:39:49 < mawk> rgba(62, 210, 190, 71) 2019-10-13T20:39:55 < antto> cuz of eiffel 69 2019-10-13T20:40:02 < antto> or wutever it was 2019-10-13T20:40:08 < BrainDamage> eiffel69 are italians 2019-10-13T20:40:14 < BrainDamage> not french 2019-10-13T20:40:19 < BrainDamage> and it was horrible here 2019-10-13T20:40:21 < antto> so? 2019-10-13T20:40:41 < antto> i meant whoever made that "i'm blue dabadee dabadaa" sh*t song 2019-10-13T20:41:07 < BrainDamage> yes, it was horrible here with that song, just saying that they didn't block blue colours because of that :p 2019-10-13T20:41:38 < antto> i won't tolerate it 2019-10-13T20:42:39 < Tordek> is therer a way to count cycles between two breakpoints in stm32cubeide? 2019-10-13T20:42:44 < Tordek> or at least a general cycle count I can take the difference for 2019-10-13T20:43:19 < BrainDamage> start a timer? 2019-10-13T20:43:43 < mawk> yeah start a timer at the first, stop it at the second 2019-10-13T20:43:47 < mawk> it shouldn't waste much cycles 2019-10-13T20:44:06 < mawk> or just look at your pocket watch when hitting "continue" at the first breakpoint 2019-10-13T20:44:29 < mawk> for the general cycle count well you have the HAL ticks, easy to read 2019-10-13T20:45:03 < Tordek> where do I see the hal ticks? 2019-10-13T20:45:10 < mawk> well are you using HAL ? 2019-10-13T20:45:16 < Tordek> yes 2019-10-13T20:46:17 < mawk> uint32_t HAL_GetTick(void); 2019-10-13T20:46:27 < Tordek> thanks! 2019-10-13T20:46:47 < mawk> but I'm not extra sure about if this is really a real clock 2019-10-13T20:46:50 < mawk> like, maybe it can be paused 2019-10-13T20:46:57 < mawk> better use a timer of your own if you want a real thing 2019-10-13T20:48:54 < Tordek> yeah, it doesn't look like it does what I need; it's stuck at the same value when the timer ticks 2019-10-13T20:49:48 < mawk> it counts in milliseconds 2019-10-13T20:49:55 < mawk> so if your stuff takes less than a ms it's totally normal 2019-10-13T20:50:23 < mawk> for better accuracy use a timer 2019-10-13T20:50:28 < mawk> you know how to do it ? 2019-10-13T20:51:10 < Jan-> OK all this seems to working 2019-10-13T20:51:21 < Jan-> to be working. to working be. worker bee! bzzz. 2019-10-13T20:51:49 < Jan-> for some reason this test code has a setup and a loop like arduino. 2019-10-13T20:52:00 < Jan-> I don't usually use the arduino library but apparently the stm32 implementation of it is okay. 2019-10-13T20:52:08 < mawk> I don't like it 2019-10-13T20:52:15 < mawk> it may look like arduino yeah 2019-10-13T20:52:22 < mawk> but it's not as handholdy 2019-10-13T20:52:30 < mawk> it's a real main function, not a "setup()" and "loop()" 2019-10-13T20:52:44 < Jan-> well there *is* setup and loop 2019-10-13T20:52:56 < mawk> really ? 2019-10-13T20:53:00 < mawk> you're using stm32duino then ? 2019-10-13T20:53:05 < mawk> I thought you were using cubeIDE with HAL 2019-10-13T20:53:34 < Jan-> I started off doing that 2019-10-13T20:53:40 < Jan-> but it seemed sort of overcomplicated 2019-10-13T20:53:43 < PaulFertser> It's platformio today :) 2019-10-13T20:53:46 < mawk> ah lol 2019-10-13T20:53:46 < PaulFertser> From bad to worse 2019-10-13T20:53:58 < mawk> why overcomplicated Jan- ? you were blinking 2019-10-13T20:54:10 < Jan-> I think I probably have the same reluctance to use arduino stuff as you guys would 2019-10-13T20:54:19 < mawk> I told you all the "complication" you encountered was only because you have a custom board 2019-10-13T20:54:25 < Jan-> but I have asked a lot about it and apparently the stuff like digitalwrite is single cycle 2019-10-13T20:54:29 < Jan-> so it doesn't suck ass like real arduino 2019-10-13T20:54:31 < mawk> you can't blame it on the software not knowing what the board looks like all by itself 2019-10-13T20:55:51 < Jan-> compare and contrast https://pastebin.com/y0Ti2Vhu 2019-10-13T20:57:13 < mawk> lol 2019-10-13T20:57:15 < mawk> that proves nothing 2019-10-13T20:57:40 < mawk> first, you can ask cube to split the main.c file into subfiles for better accuracy 2019-10-13T20:57:56 < mawk> and even with that it's just a bunch of comments and empty things for cube 2019-10-13T20:58:04 < mawk> /* USER CODE BEGIN PTD */ 2019-10-13T20:58:36 < mawk> and even if cube looks verbose it just means arduino is hiding stuff below the carpet 2019-10-13T20:59:04 < BrainDamage> just look at the asm 2019-10-13T20:59:05 < Jan-> fine that's what computers are for :) 2019-10-13T20:59:13 < BrainDamage> then you'll know what it does 2019-10-13T20:59:18 < mawk> with cube you have cubeMX to configure graphically, not with arduino 2019-10-13T20:59:21 < mawk> I thought you loved UIs 2019-10-13T20:59:39 < mawk> for better readability 2019-10-13T20:59:40 < mawk> * 2019-10-13T21:00:38 < Jan-> I think what really pushed me over the edge was the discovery that it's 50 characters to set a pin 2019-10-13T21:02:07 < mawk> lol 2019-10-13T21:02:10 < mawk> so what 2019-10-13T21:02:12 < mawk> make an alias 2019-10-13T21:02:29 < mawk> HAL_GPIO_WritePin(LD2_GPIO_Port, LD2_GPIO_Pin, 1); 2019-10-13T21:02:31 < mawk> it's not that long 2019-10-13T21:03:05 < Jan-> it's longer than digitalWrite(PA7, HIGH); 2019-10-13T21:03:10 < mawk> lol 2019-10-13T21:03:23 < mawk> that's not exactly how you compare software quality 2019-10-13T21:03:23 < Jan-> and even that is longer than PORTA &= 1 < 7 2019-10-13T21:03:28 < specing> Not a good start, Jan- 2019-10-13T21:04:29 < Jan-> what isn't 2019-10-13T21:04:42 < Jan-> apparently you can do it that way on stm32 anyway. 2019-10-13T21:04:44 < mawk> hitting periph registers directly with bit manipulation isn't exactly the best thing to do 2019-10-13T21:04:49 < mawk> yeah, you can 2019-10-13T21:04:52 < mawk> but you shouldn't 2019-10-13T21:04:54 < Jan-> volatile uint32_t* MODER_B = (volatile uint32_t*) 0x40020400; *MODER_B &= ~(0b11<<26); 2019-10-13T21:05:14 < Jan-> I was always surprised they weren't pointers like that on avr 2019-10-13T21:05:18 < Jan-> on avr they're macros 2019-10-13T21:05:24 < BrainDamage> remember one thing, you need to evaluate a lib how it handles easy tasks, but also how it handles hard tasks 2019-10-13T21:05:38 < BrainDamage> arduino libs makes simple things simple, but hard things impossible 2019-10-13T21:05:39 < Jan-> I'm too stupid to do hard tasks anyway. 2019-10-13T21:05:47 < BrainDamage> hard is relative 2019-10-13T21:05:50 < specing> llol 2019-10-13T21:06:17 < Jan-> to be fair though BrainDamage if it comes to it you can always just start using normal C as far as I know 2019-10-13T21:06:34 < Jan-> that's what I do in arduino, I never use the library, I write my own main function and go from there. 2019-10-13T21:06:36 < BrainDamage> sure, but then you're not using the arduino "platform" 2019-10-13T21:06:45 < Jan-> so? 2019-10-13T21:06:56 < BrainDamage> my biggest gripe against arduino is the libs indeed 2019-10-13T21:07:13 < BrainDamage> and I thought it was what was being discussed right now 2019-10-13T21:07:19 < Jan-> Seriously, I looked at this and went "well I'm using stm32 for more performance why would I use arduino libraries" 2019-10-13T21:07:27 < Jan-> and I was repeatedly promised that the stm32 libraries don't suck like the avr ones. 2019-10-13T21:07:33 < jpa-> BrainDamage: meh, IMO arduino libs make simple things simple, and for hard things you can copy the code and extend because it's a good enough starting point 2019-10-13T21:08:12 < jpa-> with stdperiph and cube, hard things are confusing and the code is confusing and not a good starting point either 2019-10-13T21:09:00 < jpa-> Jan-: no, stm32 libs suck in all new ways, but fortunately there are so many of them you can always switch to another one and get new funny flaws! 2019-10-13T21:09:57 < mawk> stop scaring Jan- away :( 2019-10-13T21:11:03 < mawk> if you're never going to use arduino why are you using arduino Jan- ? I think cube or libopencm3 to be more suited to that 2019-10-13T21:11:15 < mawk> you get a main function and you go on 2019-10-13T21:11:23 < mawk> if you want to configure pins or whatever, it's in the UI, very easy 2019-10-13T21:11:40 < mawk> change clock freq, add a timer, use DMA? all of this is through the GUI 2019-10-13T21:13:41 < Tordek> sorry, I was afk 2019-10-13T21:14:13 < Tordek> I added a timer, but I don't know how to read its value 2019-10-13T21:17:26 < Jan-> so which is the library I get in cube 2019-10-13T21:17:40 < Jan-> if I had to say I'd say it seems enterprisey 2019-10-13T21:18:02 < jpa-> Jan-: it's "cube", and it is layered on top of something called "low-level hal" 2019-10-13T21:18:14 < jpa-> all ST code feels enterprisey 2019-10-13T21:18:20 < Jan-> it's not just me then :/ 2019-10-13T21:18:38 < Tordek> I tried reading TIM3->CNT (that's the timer I configured) but it remains at 0 2019-10-13T21:18:48 < Tordek> it is in count up mode 2019-10-13T21:25:15 < Jan-> so what does code written with libopencm3 look like 2019-10-13T21:26:47 < mawk> yes CNT Tordek 2019-10-13T21:27:01 < jpa-> Tordek: have you remembered to enable the clock in CCR->APBxENR? 2019-10-13T21:27:11 < mawk> and also start the clock 2019-10-13T21:27:32 < jpa-> Tordek: (check if TIM3->CR1 has correct value when you read it back, if it is always zero, it is most probably the CCR enable that is missing) 2019-10-13T21:28:26 < mawk> you configured the clock using cubemx Tordek ? 2019-10-13T21:28:30 < mawk> or manally 2019-10-13T21:29:40 < kakimir64> should we fire up codes for today? 2019-10-13T21:32:07 < mawk> yes 2019-10-13T21:37:07 < kakimir64> kinda slept the whole day 2019-10-13T21:38:00 < mawk> me too 2019-10-13T21:38:07 < mawk> my head extremely hurts 2019-10-13T21:38:35 < kakimir64> oh fellow uncontrollable sleep syndrome 2019-10-13T21:39:04 < roomba> stop doing drugs 2019-10-13T21:39:08 < roomba> drugs r bad 2019-10-13T21:40:58 < Steffanx> bitmask did you have one of those fancy ikea lack tables? 2019-10-13T21:41:01 < Steffanx> *didnt 2019-10-13T21:41:15 < bitmask> yea 2019-10-13T21:41:21 < bitmask> two of em stacked 2019-10-13T21:41:46 < Steffanx> Whats the height of the top? 2019-10-13T21:41:54 < Steffanx> ~5cm? 2019-10-13T21:42:18 < bitmask> yea exactly 2019-10-13T21:42:25 < Steffanx> ah thanks 2019-10-13T21:43:31 < Steffanx> You have to assembly it yourself right? As in.. attach the legs to the top. 2019-10-13T21:43:47 < Steffanx> thanks a yes :P 2019-10-13T21:44:17 < Tordek> yes mawk 2019-10-13T21:44:22 < bitmask> yea, just screw em in 2019-10-13T21:44:36 < Tordek> i started the clock too 2019-10-13T21:45:03 < Tordek> ill check ccr 2019-10-13T21:46:51 < Tordek> sorry, by yes I mean "i used cubemx" 2019-10-13T21:48:04 < mawk> ah ok 2019-10-13T21:48:21 < mawk> then you start-and-reset the clock on first breakpoint, and stop it on the second 2019-10-13T21:48:29 < mawk> if you want you can screenshot the clock conf so I take a look 2019-10-13T21:52:18 < kakimir64> I think I need to define configASSERT 2019-10-13T21:52:41 < Tordek> the code or the cube config? 2019-10-13T21:57:15 < mawk> the cube config 2019-10-13T21:57:20 < mawk> since you dind't modify the code 2019-10-13T21:57:31 < mawk> to track down a bug kakimir64 ? 2019-10-13T21:57:40 < kakimir64> yes 2019-10-13T21:57:52 < kakimir64> possibly a ton of them 2019-10-13T21:59:34 < Tordek> https://i.imgur.com/Uaa5lHo.png 2019-10-13T21:59:56 < Tordek> and I start it with HAL_TIM_Base_Start_IT(&htim3); 2019-10-13T22:01:09 < bitmask> how do you make a schematic? :P 2019-10-13T22:01:48 < Tordek> I like kicad 2019-10-13T22:02:25 < Jan-> can I even use an alternate library in platformio? 2019-10-13T22:02:30 < bitmask> I'm using altium, I just mean in general though, I have no experience so I don't do things correctly 2019-10-13T22:02:36 < Jan-> right now I'm like "#include " 2019-10-13T22:02:40 < Jan-> presumably I could use something else 2019-10-13T22:02:45 < bitmask> like naming and organizing and stuff 2019-10-13T22:02:55 < Tordek> Jan-: you can choose one when creating the project 2019-10-13T22:03:03 < bitmask> when to split up things and just use net names 2019-10-13T22:03:28 < mawk> don't start it in IT mode Tordek 2019-10-13T22:03:30 < mawk> if you're just counting 2019-10-13T22:03:35 < mawk> start it normally 2019-10-13T22:03:55 < mawk> like, libopencm3 instead Jan- ? 2019-10-13T22:04:04 < mawk> you'd just use the compiler and that's all 2019-10-13T22:04:13 < mawk> there are examples in libopencm3 code, it's pretty easy 2019-10-13T22:04:29 < Tordek> what's IT mode, I thought it was Internal Timer ;[ 2019-10-13T22:04:33 < mawk> maybe if your board is popular enough it's known by libopencm3 too 2019-10-13T22:04:36 < mawk> IT is interrupt Tordek 2019-10-13T22:05:08 < Jan-> well 2019-10-13T22:05:12 < Jan-> I'm using platformio 2019-10-13T22:05:15 < Jan-> I don't know if it's an option here 2019-10-13T22:05:16 < Tordek> oic 2019-10-13T22:05:28 < Tordek> well, I changed it to `HAL_TIM_Base_Start(&htim3);` and it still remains at 0 2019-10-13T22:05:44 < mawk> you need to start the counter also 2019-10-13T22:05:45 < mawk> let me see 2019-10-13T22:06:59 < mawk> hmm first to get the counter you do __HAL_TIM_GET_COUNTER(TIM1) for instance 2019-10-13T22:07:29 < Tordek> __HAL_TIM_GET_COUNTER(TIM1) 2019-10-13T22:07:32 < Tordek> er 2019-10-13T22:07:38 < Tordek> yay copypaste 2019-10-13T22:08:16 < mawk> then to make it count you do __HAL_TIM_ENABLE(TIM3) 2019-10-13T22:08:22 < mawk> yours is TIM3 sorry, not TIM1 2019-10-13T22:09:15 < mawk> you better reset the value to 0 at the first breakpoint, otherwise you have to mess with loop around at stuff 2019-10-13T22:09:40 < mawk> then, you'll need to set up prescaler and period all accordingly to the relevant peripheral clock to know exactly the frequency at which it counts, of course 2019-10-13T22:09:44 < mawk> otherwise the value is meaningless 2019-10-13T22:10:38 < Tordek> htim3->Instance->CNT 2019-10-13T22:11:06 < mawk> don't use that directly 2019-10-13T22:11:08 < mawk> use the macro 2019-10-13T22:11:23 < Tordek> ...ok, copypaste screwed me over 2019-10-13T22:11:30 < Tordek> __HAL_TIM_GET_COUNTER(&htim3); 2019-10-13T22:11:32 < Tordek> there 2019-10-13T22:11:37 < mawk> yes 2019-10-13T22:11:39 < mawk> you can put TIM3 in it too 2019-10-13T22:11:45 < mawk> it's a macro with directly the address 2019-10-13T22:11:52 < Tordek> no, it gave me a type error with TIM3 2019-10-13T22:11:53 < mawk> hmm I think 2019-10-13T22:11:56 < mawk> yeah, &htim3 sorry 2019-10-13T22:11:59 < mawk> ok 2019-10-13T22:12:30 < Tordek> what I'm trying to count is how many cycles it takes between each execution of the interruption 2019-10-13T22:12:36 < mawk> so, you enable the timer, on first breakpoint you set the value to 0, on second you fetch the value 2019-10-13T22:12:47 < Tordek> because it _should_ be running at 1mhz, but it's not 2019-10-13T22:12:52 < mawk> the interrupt ? 2019-10-13T22:12:54 < mawk> ah 2019-10-13T22:13:00 < mawk> maybe you didn't set up clock right ? 2019-10-13T22:13:02 < Tordek> a different timer that I'm testing 2019-10-13T22:13:12 < mawk> ah 2019-10-13T22:13:24 < mawk> you have a custom board ? or a nucleo board 2019-10-13T22:13:33 < Tordek> discovery f4 2019-10-13T22:13:36 < mawk> can you show the RCC peripheral, and the clock configuration tab ? 2019-10-13T22:13:37 < mawk> ah good 2019-10-13T22:13:44 < mawk> so it should be set up correctly then 2019-10-13T22:13:53 < Tordek> that was my first guess; but the clock does run at 84mhz, because I set a counter to 84 million and it does blink 1/2 2019-10-13T22:13:54 < mawk> maybe you didn't compute the prescaler and so on correctly for the bad timer then 2019-10-13T22:13:57 < Tordek> 1/s 2019-10-13T22:13:58 < mawk> ah 2019-10-13T22:14:24 < Tordek> but I then put it to 42 (should be getting around 2mhz), but I get 186khz instead 2019-10-13T22:15:08 < mawk> odd 2019-10-13T22:15:08 < Tordek> and all I have in the interrupt is some ifs and 3 HAL_GPIO_WRITEPIN 2019-10-13T22:15:15 < mawk> can you show clock configuration tab ? 2019-10-13T22:15:17 < mawk> ah 2019-10-13T22:15:21 < mawk> well 2019-10-13T22:15:30 < mawk> maybe that's the reason ? 2019-10-13T22:15:41 < mawk> but I guess it's ok to write GPIO from ISR 2019-10-13T22:15:45 < kakimir64> is configTOTAL_HEAP_SIZE the size of heap I have defined in linker script? 2019-10-13T22:15:46 < Tordek> is writepin THAT slow? 2019-10-13T22:16:00 < mawk> I don't know, you should ask someone more expert 2019-10-13T22:16:25 < mawk> yes kakimir64 that value will be put in the linker script normally 2019-10-13T22:16:27 < mawk> iirc 2019-10-13T22:16:40 < kakimir64> writing pins can be quick or slow depending on chip 2019-10-13T22:17:07 < kakimir64> it's called commonly "GPIO toggle rate" 2019-10-13T22:17:12 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 265 seconds] 2019-10-13T22:18:44 < kakimir64> when you move out of AVR single cycle world it might come as suprise that things can take cycles and cycles to be performed 2019-10-13T22:20:20 < kakimir64> but what I have understood is that most of stm32s have exeptionally good gpio rates 2019-10-13T22:25:47 < kakimir64> oh there is 6th heap scheme for freertos 2019-10-13T22:25:56 < kakimir64> heap_6.c to be used with newlib 2019-10-13T22:27:14 < Tordek> yeah, htim3 remains at 0 still 2019-10-13T22:32:53 < mawk> you need to set up the prescaler and everything for it 2019-10-13T22:33:16 < Tordek> https://gist.github.com/Tordek/ecc017cbc70325b71aed118bf9e5ae8f 2019-10-13T22:34:01 < mawk> move line 12 and 13 inside the if at line 14 2019-10-13T22:34:06 < mawk> just for safety 2019-10-13T22:34:32 < Tordek> updated to show tim3_init 2019-10-13T22:34:43 < mawk> hmm also you're doing a clock manually ? like using TIM2 as master 2019-10-13T22:34:53 < mawk> you can do that with no code, only configuring TIM2 2019-10-13T22:35:40 < Tordek> sorry, what do you mean? 2019-10-13T22:36:34 < mawk> you're toggling a GPIO using TIM2 as a master 2019-10-13T22:36:45 < mawk> and I mean that you can configure TIM2 to do that automatically, instead of using an interrupt 2019-10-13T22:36:50 < kakimir64> pros@freertos - why heap_6.c is not included in freertos source ? 2019-10-13T22:37:27 < mawk> it's not an official thing 2019-10-13T22:37:30 < Tordek> you mean CLK_Pin? Sure, but that's only one of the things going on 2019-10-13T22:37:30 < mawk> https://mcuoneclipse.com/2017/07/02/using-freertos-with-newlib-and-newlib-nano/ 2019-10-13T22:37:37 < kakimir64> heap_6.c is made by Real Time engineers ltd. though 2019-10-13T22:37:42 < mawk> ah 2019-10-13T22:38:05 < qyx> Tordek: are you trying to do interrupts at 2MHz? 2019-10-13T22:38:11 < Tordek> yes 2019-10-13T22:38:21 < qyx> with "only" some ifs and gpio writing? 2019-10-13T22:38:25 < mawk> https://github.com/ErichStyger/McuOnEclipse_PEx it's here 2019-10-13T22:38:43 < Tordek> yeah, that's all that's happening inside the interrupr 2019-10-13T22:38:51 < qyx> your are maybe an order of magnitude off between your expectations and reality 2019-10-13T22:39:13 < qyx> e. you are doing it wrong 2019-10-13T22:40:00 < mawk> 0 as autoreload won't work Tordek 2019-10-13T22:40:08 < Tordek> I assumed since it's running at 168mhz it would be fast enough, but I was expecting that 2019-10-13T22:40:15 < mawk> aka period 2019-10-13T22:40:21 < Tordek> mawk: autoreload is disabled for tim3 2019-10-13T22:40:38 < mawk> doesn't matter 2019-10-13T22:40:45 < mawk> it's also used as the period of that timer 2019-10-13T22:41:07 < mawk> what is disabled is autoreload *preload*, it's something else that you think of 2019-10-13T22:41:47 < kakimir64> #ifndef configLINKER_HEAP_SIZE_SYMBOL #define configLINKER_HEAP_SIZE_SYMBOL _HeapSize 2019-10-13T22:42:03 < kakimir64> there is no _HeapSize so dis must be the problem 2019-10-13T22:42:15 < mawk> it's defined in the linker script no ? 2019-10-13T22:42:19 < mawk> or somewhere lese 2019-10-13T22:42:24 < mawk> but you don't have to do this yourself 2019-10-13T22:42:29 < mawk> just change the config #define and that's all 2019-10-13T22:42:30 < kakimir64> not in mcuxpresso 2019-10-13T22:42:36 < mawk> a 2019-10-13T22:42:37 < mawk> broken thing 2019-10-13T22:42:52 < kakimir64> I could change the linker template to include it 2019-10-13T22:43:12 < mawk> ah 2019-10-13T22:43:15 < mawk> just do #define configLINKER_HEAP_SIZE_SYMBOL 2019-10-13T22:43:19 < mawk> and try again 2019-10-13T22:43:33 < kakimir64> or could I just _pvHeapLimit-_pvHeapStart in that define? 2019-10-13T22:43:43 < mawk> ugly 2019-10-13T22:43:51 < Tordek> alright, there we go 2019-10-13T22:43:58 < Tordek> thanks! 2019-10-13T22:44:06 < kakimir64> mawk: rather do it in linker script 2019-10-13T22:44:07 < kakimir64> ? 2019-10-13T22:44:23 < mawk> as I remember it you don't have to mess with linker script at all kakimir64 , so I don't know what's going on 2019-10-13T22:44:37 < mawk> or maybe you do ? and cubeMX was doing it automatically for me 2019-10-13T22:44:51 < mawk> just look at the linker script, is the value hardcoded ? 2019-10-13T22:44:56 < mawk> if yes, change it both places and that's all 2019-10-13T22:44:58 < kakimir64> cubeworld != nxpworld 2019-10-13T22:45:09 < mawk> the linker scripts will be similar anyway 2019-10-13T22:45:19 < qyx> Tordek: was the ARR value wrong? 2019-10-13T22:45:27 < qyx> did you get 2MHz? 2019-10-13T22:45:48 < Cracki> maybe the interrupt just takes too long 2019-10-13T22:46:08 < Tordek> qyx: yes and no; my Period was at 0 so the counter easn't counting 2019-10-13T22:46:19 < Cracki> he is aiming for 2 MHz interrupt frequency, that leaves 84 clock cycles (at 168 mhz) per execution 2019-10-13T22:46:22 < qyx> idk, I was getting around 150K on F1 at 72MHz 2019-10-13T22:46:25 < Tordek> but apparently there's 817 instructions between each interrupr call 2019-10-13T22:46:35 < Cracki> hm 2019-10-13T22:46:37 < kakimir64> mawk: oh linker script has _HeapSize so.. I trusted internets saying mcuxpresso doesn't have such symbol in linker script 2019-10-13T22:46:51 < kakimir64> it's not that then 2019-10-13T22:48:39 < mawk> TIM counts time, not instructions, Tordek 2019-10-13T22:48:54 < Tordek> er, yes, sorry 2019-10-13T22:49:00 < Tordek> which is even worse 2019-10-13T22:49:23 < mawk> to know exactly the period of the timer you need to know the periph clock frequency it's using 2019-10-13T22:49:32 < Tordek> mhz 2019-10-13T22:49:36 < Tordek> 84 2019-10-13T22:49:49 < Tordek> (yay numpad not sending keypresses to putty) 2019-10-13T22:54:19 < Tordek> ok, after a few runs (I guess it was being interrupted by something else) it stabilizes at 214/284 2019-10-13T22:55:01 < mawk> you have a 16bit timer, so with a 65535 autoreload value you can count up to around 780µs 2019-10-13T22:55:06 < mawk> it's not that much 2019-10-13T22:55:16 < mawk> you better want to use a prescaler, or a 32bit timer 2019-10-13T22:55:57 < Tordek> sure but I'm only counting time between interruptions, it can't possibly be above 65k 2019-10-13T22:56:23 < mawk> if you have more than 780µs between them sure 2019-10-13T22:56:40 < mawk> but I guess it's okay in your case 2019-10-13T22:57:03 < Tordek> I get 170 ticks of the clock when doing nothing other than counting 2019-10-13T22:57:20 < Tordek> so I guess that's the lowest limit for counting 2019-10-13T22:58:09 < mawk> you mean between two consecutive reads of the counter ? 2019-10-13T22:58:13 < mawk> that's 2µs, that's odd 2019-10-13T22:58:30 < mawk> your core is running slow 2019-10-13T22:59:12 < Tordek> when reading the counter at the start of the interruption, yeah 2019-10-13T22:59:51 < Tordek> https://gist.github.com/Tordek/b2dee7aeda3aa46428128f054f1cbd69 2019-10-13T23:00:16 < Tordek> t is 170 when I put a breakpoint at line 5 2019-10-13T23:01:52 < R2COM> darn i forgot which source file by default has interrupt dummy functions? 2019-10-13T23:01:56 < R2COM> like TIM1_IRQHandler() 2019-10-13T23:02:21 < R2COM> which file has those blank functions? (the ones which i fill in and program in my _it.c files) 2019-10-13T23:04:39 < Tordek> R2COM: HAL_Driver/Src/whatever.c? 2019-10-13T23:05:24 < R2COM> yes so what is that whatever.c file 2019-10-13T23:06:37 < Tordek> there's several files, but you're looking for __weak functions 2019-10-13T23:06:42 < Tordek> e.g. __weak void HAL_TIM_PeriodElapsedCallback(TIM_HandleTypeDef *htim) 2019-10-13T23:07:06 < R2COM> i know what im looking for 2019-10-13T23:07:12 < R2COM> im asking which file has that shit 2019-10-13T23:07:20 < Tordek> depends on the shit 2019-10-13T23:07:26 < R2COM> i just showed 2019-10-13T23:07:29 < R2COM> TIM1_IRQHandler 2019-10-13T23:07:30 < R2COM> ^ 2019-10-13T23:07:34 < Tordek> tim funcions are on hal_tim 2019-10-13T23:07:36 < R2COM> which file has that weak blank function 2019-10-13T23:07:41 < R2COM> tim files dont have it 2019-10-13T23:07:58 < Tordek> void HAL_TIM_PeriodElapsedCallback(TIM_HandleTypeDef *htim) <- is this not the funciton you're looking for? 2019-10-13T23:08:59 < R2COM> thats callback functions 2019-10-13T23:09:06 < R2COM> im looking for is interrupt related 2019-10-13T23:09:30 < R2COM> IRQ handler blank functions 2019-10-13T23:10:03 < Tordek> interruptions are handled via callbacks, though? 2019-10-13T23:10:22 < R2COM> ... 2019-10-13T23:12:05 < Tordek> you mean src/_it.c? 2019-10-13T23:12:21 < Tordek> void TIM2_IRQHandler(void) 2019-10-13T23:13:12 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Ping timeout: 268 seconds] 2019-10-13T23:13:24 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-13T23:17:42 < kakimir64> is there a way in linux to autocomplete with mismatching case? 2019-10-13T23:18:08 < kakimir64> I mean some file nimes.. XxxxXXXXxxxXXX 2019-10-13T23:18:11 < Tordek> bash? vim? 2019-10-13T23:18:15 < kakimir64> bash 2019-10-13T23:18:39 < Tordek> set completion-ignore-case On 2019-10-13T23:18:39 < kakimir64> I mean it should take a hint when I keep smashing tabulator 2019-10-13T23:19:02 < kakimir64> oh wow 2019-10-13T23:19:08 < kakimir64> bashrc? 2019-10-13T23:19:28 < Tordek> https://askubuntu.com/questions/87061/can-i-make-tab-auto-completion-case-insensitive-in-bash this answer says inputrc 2019-10-13T23:19:37 < Tordek> but I don't use bash so :shrug: 2019-10-13T23:21:13 < kakimir64> sweet! 2019-10-13T23:21:19 < kakimir64> thanks Tordek 2019-10-13T23:22:20 < kakimir64> it's like learning to walk 2019-10-13T23:23:27 < Steffanx> Did you ask google how to walk? 2019-10-13T23:23:31 < Steffanx> or was it irc as well? 2019-10-13T23:24:33 < kakimir64> irc is the best 2019-10-13T23:27:22 < kakimir64> literally the first time I'm using assert 2019-10-13T23:27:26 < kakimir64> assert() 2019-10-13T23:27:56 < Steffanx> dont forget http://www.robertgamble.net/2012/01/c11-static-assertions.html 2019-10-13T23:28:09 < Steffanx> C11 moite. 2019-10-13T23:29:04 < Tordek> mawk/cracki/qyx well, R2COM's option is also possible, if I avoid the HAL and handle the IRQ manually, I can go down to 14 cycles 2019-10-13T23:29:13 < Tordek> well, ticks 2019-10-13T23:29:26 < kakimir64> Steffanx is moite 2019-10-13T23:29:36 < Cracki> apropos HAL, Jan-, under hal is LL and that might be more to your liking 2019-10-13T23:29:53 < Steffanx> those function names are pretty long as well Cracki 2019-10-13T23:30:03 < Cracki> have you considered generating those damned clock pulses autonomously, with output compare function of the timer? 2019-10-13T23:30:22 < Cracki> interrupts and bitbanging is just not sensible at these frequencies 2019-10-13T23:30:31 < kakimir64> longest function name I have made was like 200char 2019-10-13T23:30:52 < kakimir64> something like full screen width 2019-10-13T23:31:20 * Steffanx bans kakimir64 from coding ever again 2019-10-13T23:31:56 < Steffanx> i mean bans kakimir64 from coding. forever. 2019-10-13T23:31:57 < kakimir64> I know I did something wrong but don't know exactly why 2019-10-13T23:32:06 < Steffanx> did it also have 25 parameters? 2019-10-13T23:32:17 < Steffanx> param1, param2, param3, param5. etc? 2019-10-13T23:32:24 < kakimir64> I have not had so many ever 2019-10-13T23:32:27 < kakimir64> not even ten ever 2019-10-13T23:32:35 < Steffanx> (yes, param4 was left out intentionally) 2019-10-13T23:32:50 < kakimir64> it's left out for lulz 2019-10-13T23:33:40 < kakimir64> so first time I use assert and the __file__ is apparently a number? 2019-10-13T23:33:47 < kakimir64> or am I failing 2019-10-13T23:33:52 < Cracki> ura fail 2019-10-13T23:33:55 < Cracki> __file__ is a string 2019-10-13T23:34:00 < Cracki> compiler inserts that 2019-10-13T23:34:09 < Cracki> __line__ is a number, same 2019-10-13T23:34:24 < Cracki> there's also __func_ or something like that 2019-10-13T23:34:45 < Cracki> a proper assert already formats the message with those things, and also the expression that failed 2019-10-13T23:35:04 < Cracki> assert(condition that should be true but may not be) 2019-10-13T23:35:23 < kakimir64> there is no string in that location 2019-10-13T23:35:33 < Cracki> *jedi wave* 2019-10-13T23:35:55 < kakimir64> oh well shit 2019-10-13T23:36:10 < kakimir64> assert define and function have different parameter order 2019-10-13T23:39:53 < kakimir64> certainly I have used __FILE__ before 2019-10-13T23:48:00 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-13T23:52:21 < Tordek> also it's typically used as assert(something && "Explanation string"); 2019-10-13T23:54:02 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] --- Day changed Mon Oct 14 2019 2019-10-14T00:00:46 < bitmask> do you capitalize net names 2019-10-14T00:01:00 < bitmask> as in all caps 2019-10-14T00:01:02 < bitmask> or no 2019-10-14T00:01:16 < kakimir64> I apitalize all names 2019-10-14T00:01:34 < bitmask> k 2019-10-14T00:24:09 < mawk> lowercase __file__ __func__ and so on are from C99 2019-10-14T00:24:18 < mawk> if you're not C99 you want the GNU extensions which are uppercase 2019-10-14T00:25:03 < Cracki> always remember, taking a breath is impldef 2019-10-14T00:25:08 < mawk> lol 2019-10-14T00:28:42 < mawk> I got my name in the changelog of WireGuard 2019-10-14T00:28:47 < mawk> I fixed it for android 10 2019-10-14T00:30:05 < mawk> I will put it on my grave 2019-10-14T00:35:30 < kakimir64> you have been mentioned in changelog 2019-10-14T00:35:35 < kakimir64> your life is complete 2019-10-14T00:36:51 < kakimir64> I wonder when will the android version be like firefox version now 2019-10-14T00:57:12 < jadew> man... I'm freaking stuck with paying the stupid paypal conversion fee 2019-10-14T00:57:41 < jadew> ebay won't allow to change my listings from GBP to EUR 2019-10-14T00:57:59 < jadew> not without losing the listing history, watchers, etc 2019-10-14T00:58:14 < BrainDamage> maybe consider another selling platform? 2019-10-14T00:58:24 < BrainDamage> how much would it cost to be an amazon store? 2019-10-14T00:58:38 < jadew> I tried to do that, but it didn't work for some reason 2019-10-14T00:58:47 < jadew> don't know what they didn't like about my card 2019-10-14T00:59:15 < jadew> I have an online shop, but I think most people search on eBay 2019-10-14T00:59:37 < jadew> they're making really good use of their position, that's for sure 2019-10-14T00:59:54 < jadew> eBay is a sister company to PayPal, so it makes sense that they wouldn't let you change currencies 2019-10-14T01:00:26 < jadew> they have everything to gain from this 2019-10-14T01:01:37 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-14T01:02:04 < jadew> there's actually a way of reducing that fee to a fixed 1.5 GBP, but transferwise doesn't support it (it shows up in the log, but they reject the transfer for "unknown error") 2019-10-14T01:02:34 < jadew> you can withdraw to card, and that way you can withdraw to the card's currency 2019-10-14T01:03:01 < jadew> so if you have a card with the same currency you're withdrawing, you get away without the conversion fee 2019-10-14T01:03:13 < jadew> if you withdraw to a bank account, it will convert to EUR no matter what 2019-10-14T01:03:47 < jadew> so if you have USD in paypal, it converts that to EUR and pays you EUR, then your bank charges you again, in case you don't have a EUR account (like me) 2019-10-14T01:14:25 < Cracki> just set the offers to sold out, and note in the description (that you can change) that you've created new items 2019-10-14T01:16:45 < jadew> yeah, I think that's the only way 2019-10-14T01:17:45 < jadew> and check this out, paypal has removed the currency conversion option 2019-10-14T01:18:11 < jadew> not sure why, but I bet they're not losing money on that move 2019-10-14T01:18:15 < Cracki> they're screwing everyone over. it's time to boycott them. 2019-10-14T01:20:31 -!- [1]MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-14T01:21:39 < karlp> jadew: boards are great, shipping and shit was non ideal. they promised batching, but sent them separately. there's no cart or anything. 2019-10-14T01:21:51 < karlp> paid 9k in feeds for one, 2k in fees on the other. 2019-10-14T01:23:17 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-14T01:23:17 -!- [1]MrMobius is now known as MrMobius 2019-10-14T01:23:53 < jadew> karlp, and how much did you end up paying per cm^2? 2019-10-14T01:24:00 < jadew> the price they list on that github account? 2019-10-14T01:24:22 < jadew> https://github.com/AislerHQ/aisler-wiki/blob/master/submitting-orders/pcb-price.md 2019-10-14T01:25:00 < Jan-> so you guys 2019-10-14T01:25:31 < jadew> I'm getting increasingly unhappy with the chinese fabs, one run the boards are OK, the next it looks like they pulled them through their ass crack before shipping them to me 2019-10-14T01:31:25 < Jan-> we've had some pretty variable results 2019-10-14T01:31:36 < Jan-> some of the stuff we've had done has needed to look presentable as we are making front panels and suchlike 2019-10-14T01:31:44 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 276 seconds] 2019-10-14T01:31:45 < Jan-> the cheaper places are less good as you would expect. 2019-10-14T01:32:00 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-14T01:32:37 < Jan-> seed is better than dirty pcbs 2019-10-14T01:32:44 < Jan-> seed's been pretty good 2019-10-14T01:33:32 < jadew> yeah, I had good results with seed too, but IIRC they don't allow you to hide that little number 2019-10-14T01:33:41 < jadew> and they place it randomly 2019-10-14T01:34:18 < Jan-> true but then we've always been doing stuff on snap out frames and they always seem to put it on the frame 2019-10-14T01:34:26 < Jan-> soooo 2019-10-14T01:34:32 < Jan-> ascii shrug smiley? 2019-10-14T01:35:21 < jadew> I guess I could consider getting them on a panel 2019-10-14T01:35:29 < jadew> I like individual boards better 2019-10-14T01:35:37 < Jan-> well we often wanted round corners and stuff 2019-10-14T01:35:52 < Jan-> we had white solder mask and left gold metal visible 2019-10-14T01:36:02 < Jan-> white and gold is class 2019-10-14T01:36:12 < jadew> heh 2019-10-14T01:36:19 < Jan-> cost extra. 2019-10-14T01:36:21 < Jan-> but worth it. 2019-10-14T01:37:04 < jadew> can you get matte white? 2019-10-14T01:37:14 < Jan-> can't remember 2019-10-14T01:37:17 < Jan-> probably somewhere? 2019-10-14T01:37:33 < Jan-> the problem is that the edges look crappy 2019-10-14T01:37:37 < jadew> I'm torn between front panel options at the moment 2019-10-14T01:37:58 < Jan-> honestly for routing, cutouts, and silk screen you are not going to find a way to do it cheaper 2019-10-14T01:38:02 < jadew> yeah, I figured, not sure I'd be able to live with that 2019-10-14T01:38:08 < BrainDamage> white pcbs stain easily with rosin flux 2019-10-14T01:38:15 < BrainDamage> the stains look so fugly 2019-10-14T01:38:20 < Jan-> well, we just stuck them to masking tape and sprayed 2019-10-14T01:38:27 < Jan-> the other thing is you can make them screened 2019-10-14T01:38:31 < jadew> BrainDamage, there wouldn't be any soldering on that board tho 2019-10-14T01:38:39 < Jan-> and if you are crafty you can actually make a solder-together enclosure where the solder is on the inside 2019-10-14T01:39:00 < Jan-> much cleverness is possible 2019-10-14T01:40:37 < Cracki> huh, enclosure made of pcb... I can imagine same patterns as for MDF laser cutting, and then solder like gluing 2019-10-14T01:41:15 < Jan-> we didn't do tab and slot if that's what you were thinking 2019-10-14T01:41:21 < Jan-> you can just butt joint the corners and solder 2019-10-14T01:41:36 < Jan-> put dashes in the exposed metal so you don't have to try to heat up the entire corner joint at once 2019-10-14T01:41:37 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-14T01:41:45 < Cracki> good idea 2019-10-14T01:42:01 < Jan-> tape it together while you solder you know? 2019-10-14T01:42:35 < Jan-> if you are really crafty you can solder brass standoffs inside so you can have a screw on lid 2019-10-14T01:43:03 < Cracki> apropos laser cut... laser cut plexiglass will crack when exposed to any kind of solvent, even plain alcohol. surfaces that got heated have tension in them. best CNC mill the contours you need to glue 2019-10-14T01:43:21 < Jan-> if we do laser cut boxes we do stacks 2019-10-14T01:43:32 < Jan-> then it's just holes through the whole lot and long screws 2019-10-14T01:43:43 < Cracki> I have some laser cut pieces that are rectangular on the outside, so I can easily cut back a few mm to make them glueable 2019-10-14T01:43:45 < Jan-> just be really cautious with a counter sinking bit on laser cut 2019-10-14T01:43:53 < Jan-> it cracks easily 2019-10-14T01:44:15 < Jan-> pcb is a lot tougher 2019-10-14T01:44:20 < Cracki> indeed 2019-10-14T01:44:31 < Jan-> I mean basically you're making a fiber glass box which is pretty solid 2019-10-14T01:44:41 < Cracki> not transparent tho :P 2019-10-14T01:45:03 < BrainDamage> Im sure Jan- won't see the difference :p 2019-10-14T01:45:22 < Cracki> blind? 2019-10-14T01:45:45 * Jan- takes bow 2019-10-14T01:45:50 < Cracki> ic 2019-10-14T01:45:57 < Jan-> don't worry 2019-10-14T01:45:59 < Jan-> it's not catching 2019-10-14T01:46:11 < Cracki> oh the jokes 2019-10-14T01:46:45 * Jan- is here all week 2019-10-14T01:47:49 < Jan-> well that killed the conversation 2019-10-14T01:47:51 < Jan-> sorry to be a downer 2019-10-14T01:47:57 < BrainDamage> sorry, my fault 2019-10-14T01:48:31 < BrainDamage> wrt laser cutting, can't you apply heat to the edge? 2019-10-14T01:48:36 < Jan-> oh yeah it is your fault 2019-10-14T01:48:36 < Cracki> that makes it worse 2019-10-14T01:48:46 * Jan- gouges out BrainDamage's eyes with a tire iron 2019-10-14T01:48:53 < Jan-> preciousssss 2019-10-14T01:48:55 < BrainDamage> it should anneal a bit the tension 2019-10-14T01:48:56 < Cracki> heat for making it melt, maybe 2019-10-14T01:49:13 < BrainDamage> yes, it has to reach almost melting point 2019-10-14T01:49:21 < Jan-> you can flame polish acrylic 2019-10-14T01:49:32 < Cracki> it's extruded acrylic, so any heat caues stress. people recommended annealing it between proper glass sheets to maintain the shape 2019-10-14T01:49:36 < Cracki> (in oven) 2019-10-14T01:49:59 < Cracki> so instead of doing that, I'll just cnc mill a bit off the edges and try gluing that 2019-10-14T01:50:04 < jadew> can't you cook it? 2019-10-14T01:50:11 < Cracki> you can cook it, it'll go soft like noodles 2019-10-14T01:50:17 < Cracki> I think 2019-10-14T01:50:26 < jadew> sorry, didn't get to that line :) 2019-10-14T01:50:29 < Cracki> it'll definitely go soft in the oven, which is why glass to keep it flat 2019-10-14T01:50:34 < jadew> yeah, annealing it would probably solve the problem 2019-10-14T01:51:05 < Cracki> besides, their laser cutter's beam isn't perfectly perpendicular 2019-10-14T01:51:09 < jadew> you don't have to soften it too much 2019-10-14T01:51:15 < Jan-> mill everything from solid mahahaha 2019-10-14T01:51:22 < Cracki> apart from the general cone/wedge shape it has, it's tilted, so my edges aren't straight 2019-10-14T01:51:58 < Cracki> they're practiced with their laser cutter but I don't know if anyone there is confident enough about cnc milling 2019-10-14T01:52:28 < Cracki> a laser cutter can't crash like a cnc mill can 2019-10-14T01:53:08 < BrainDamage> depending on the collimation distance, it can drill a hole below though >_> 2019-10-14T01:53:14 < Cracki> and last time I got to handle a cnc mill (portal type) was 2016 and they didn't let me get to know the machine or the tools, just "click this, turn that, hit the button if it goes bad" 2019-10-14T01:53:24 < Cracki> which is SHIT instruction 2019-10-14T01:53:38 < BrainDamage> most people teach and learn through recipes 2019-10-14T01:53:45 < BrainDamage> rather than first principles 2019-10-14T01:53:55 < BrainDamage> so no real understanding, just copy 2019-10-14T01:54:03 < Cracki> their laser is so weak, and the mirrors so smoked up, it maybe has half the norminal power 2019-10-14T01:54:14 < Cracki> I had to run it very slowly, TWICE, to get through 4mm acrylic 2019-10-14T01:54:30 < Cracki> that contributes to heat into the material 2019-10-14T01:54:42 < Cracki> quick powerful cuts would have been better 2019-10-14T01:54:52 < Jan-> so what libopencm3 hasn't been updated since 2015? 2019-10-14T01:55:03 < Cracki> either way I'm taught, I need some "alone time" with the machine 2019-10-14T01:55:19 < Cracki> to build my own model of its weirdness 2019-10-14T01:55:26 < Cracki> it's ded jim 2019-10-14T01:55:42 < Cracki> maybe you found an old fork of it 2019-10-14T01:55:53 < Jan-> what's dead 2019-10-14T01:55:55 < BrainDamage> Jan-: last commit 2 days ago 2019-10-14T01:56:09 < Jan-> I'm trying to figure out if I can use it via platformio 2019-10-14T01:56:10 < Cracki> "it's dead jim" is a star trek quote 2019-10-14T01:56:23 < Jan-> yeah I know it's a star trek quote I just don't know what's dead :A) 2019-10-14T01:56:47 < BrainDamage> https://github.com/libopencm3 2019-10-14T01:56:58 < Jan-> yeah I avoid github 2019-10-14T01:57:23 < Jan-> getting from "it's on github" to "it's on my computer and set up and working properly" is often a long and arduous journey. 2019-10-14T01:57:41 < BrainDamage> it's also where it's documented 2019-10-14T01:58:23 < Jan-> I don't think there's a platformio preset for it anyway 2019-10-14T01:58:26 < Jan-> not for the board I have 2019-10-14T01:58:33 < Cracki> it can't be the git clone step because that's always the same, so you're making a statement about the general quality of projects attracted by the largest platform 2019-10-14T02:00:12 < Jan-> I'm just sitting here thinking 2019-10-14T02:00:20 < Jan-> do I really want to do a fairly big project in basically arduino code 2019-10-14T02:00:24 < Jan-> and the answer is no it'd be embarrassing 2019-10-14T02:00:43 < BrainDamage> because it's embarassing or because it's unmanageable? 2019-10-14T02:01:00 < BrainDamage> you'll be the only one that sees the code, that's one of the benefit and cons of embedded 2019-10-14T02:01:08 < Jan-> depends what you mean by manageable I guess 2019-10-14T02:01:13 < BrainDamage> so you only have to account to yourself for shit code 2019-10-14T02:02:34 < BrainDamage> manageable in the sense that you can add more code without fighting the current code and you can take it back after few weeks and understand it properly 2019-10-14T02:02:47 < Jan-> I think that's always a challenge 2019-10-14T02:03:12 < BrainDamage> it is, but if using arduino style lets you achieve it, don't overcomplicate your life 2019-10-14T02:03:46 < BrainDamage> your personal insight will have to be that you have to guesstimate the complexity of it 2019-10-14T02:04:23 < Jan-> I guess I'd just like to evaluate all the options 2019-10-14T02:04:50 < BrainDamage> since you're approaching the platform just now, it might make sense to try a few small projects first 2019-10-14T02:05:19 < Jan-> what I'm doing here is more or less porting (but really rewriting a bit) an existing project that basically ran out of oomph on an avr 2019-10-14T02:07:49 < Cracki> "fountain poo" https://www.youtube.com/watch?v=1HiPhkF8pnY 2019-10-14T02:09:52 < Jan-> that sounds lovely 2019-10-14T02:10:09 < Cracki> my god that dog gets gourmet cooking 2019-10-14T02:10:30 < Cracki> if more people would get that instead of junkfood, the world would be better 2019-10-14T02:26:05 < Cracki> liberal drone strikes for justice https://www.youtube.com/watch?v=9Yx2QHQnn7w 2019-10-14T03:05:34 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Read error: Connection reset by peer] 2019-10-14T03:05:50 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-14T04:12:52 < emeb_mac> haha 2019-10-14T04:19:39 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 240 seconds] 2019-10-14T04:26:03 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 240 seconds] 2019-10-14T04:28:16 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-14T04:40:57 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 240 seconds] 2019-10-14T04:59:15 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-14T05:04:03 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-14T05:04:42 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-14T05:04:50 < kakimir32> buttcoin is high again 2019-10-14T05:05:02 < kakimir32> relativelly high 2019-10-14T05:05:37 -!- Datz [~datz@unaffiliated/datz] has quit [Ping timeout: 265 seconds] 2019-10-14T05:06:31 -!- Datz [~datz@cpe-24-209-176-183.wi.res.rr.com] has joined ##stm32 2019-10-14T05:06:31 -!- Datz [~datz@cpe-24-209-176-183.wi.res.rr.com] has quit [Changing host] 2019-10-14T05:06:31 -!- Datz [~datz@unaffiliated/datz] has joined ##stm32 2019-10-14T05:19:34 < Cracki> log plot is the way 2019-10-14T05:20:15 < Cracki> someone fitted a reasonably sane curve to the log plot. it stayed within about an order of magnitude for its whole history 2019-10-14T05:21:07 < Cracki> curve was (in log space) a second order polynomial I think, maybe a decaying exponential 2019-10-14T05:23:47 < Cracki> https://medium.com/max-exchange/bitcoin-the-logarithmic-growth-curve-by-dave-the-wave-a2ef93b02df1 2019-10-14T05:24:32 < Cracki> according to their predictions from mid 2018, it should be near a minimum soon 2019-10-14T05:25:42 < Cracki> it's like reading coffee grounds 2019-10-14T05:39:57 -!- effractu1 is now known as effractur 2019-10-14T06:28:55 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-14T06:29:59 -!- fc5dc9d4_ [~quassel@p5B0816D6.dip0.t-ipconnect.de] has joined ##stm32 2019-10-14T06:34:44 -!- fc5dc9d4 [~quassel@p57A32AA0.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-14T07:46:31 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-14T07:48:26 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-14T07:51:49 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-14T07:51:49 -!- day__ is now known as day 2019-10-14T08:15:05 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-14T08:27:52 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-14T08:34:26 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-14T08:35:57 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-14T08:59:32 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-14T09:26:26 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-14T09:31:03 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-14T09:38:26 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-14T09:43:43 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-14T10:00:36 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-14T10:06:13 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-14T10:14:15 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-14T10:26:54 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has joined ##stm32 2019-10-14T10:28:46 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-14T10:34:00 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-14T10:34:57 -!- catphish [~catphish@unaffiliated/catphish] has quit [Ping timeout: 240 seconds] 2019-10-14T10:36:56 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-14T10:38:47 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-14T10:41:03 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 265 seconds] 2019-10-14T11:02:37 -!- specing [~specing@unaffiliated/specing] has quit [Ping timeout: 240 seconds] 2019-10-14T11:11:12 -!- specing [~specing@unaffiliated/specing] has joined ##stm32 2019-10-14T11:20:05 < Cracki> kek https://www.dailymail.co.uk/news/article-7566515/Extinction-Rebellion-files-reveal-climate-zealots-paid-400-week.html 2019-10-14T11:20:26 < Cracki> 400/wk is about 1740/month 2019-10-14T11:20:30 < Cracki> not shabby 2019-10-14T11:21:10 < Cracki> you can make a vacation out of it 2019-10-14T11:24:21 < Cracki> he's... vacationing 2019-10-14T11:24:46 < Cracki> I just found it funny that in german communist circles they vehemently deny that there's money in "activism" 2019-10-14T11:34:15 -!- fc5dc9d4_ [~quassel@p5B0816D6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-14T11:34:28 -!- fc5dc9d4 [~quassel@p5B0816D6.dip0.t-ipconnect.de] has joined ##stm32 2019-10-14T11:37:44 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-14T11:38:08 -!- catphish [~catphish@unaffiliated/catphish] has quit [Client Quit] 2019-10-14T11:48:16 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-14T11:50:36 < Cracki> https://pics.me.me/you-guys-are-getting-paid-42900594.png 2019-10-14T11:51:22 < Cracki> they're like that because they're good sheeple, much self flagellation and moral superiority but no gain, even though they propagate redistribution of wealth 2019-10-14T12:17:36 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-14T12:18:57 -!- jon101249 [~jon1012@pha75-18-88-178-181-173.fbx.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-14T12:22:47 -!- jon101249 [~jon1012@pha75-18-88-178-181-173.fbx.proxad.net] has joined ##stm32 2019-10-14T13:18:11 < karlp> jadew: I uploaded my design and paid what it said. the fact that that link of prices is on github instead of their website is the sort of thing I had a really bad taste in my mouth about 2019-10-14T13:18:20 < karlp> it's all too airy and "everything's rad dudes" 2019-10-14T13:18:33 < karlp> right downt o the "trust me, I'm a maker" stickers that came in the parcel 2019-10-14T13:19:19 < karlp> you don't get perfectly milled board edges either, they've got panel bites on them. not as rough as oshpark has/did, but still real. 2019-10-14T13:27:30 < jadew> karlp, oh, thanks about that info! 2019-10-14T13:27:40 < jadew> the mouse bites are a deal breaker for me 2019-10-14T13:28:05 < jadew> the boards are sold without an enclosure, so they have to look good 2019-10-14T13:28:37 < karlp> I'll send a pic. oe tick 2019-10-14T13:31:17 < karlp> https://nc.beeroclock.net/s/Jnr5objDiHZFjHc 2019-10-14T13:32:27 < jadew> thanks 2019-10-14T13:32:33 < jadew> it's asking for a password tho :) 2019-10-14T13:32:58 < jadew> works now 2019-10-14T13:33:12 < karlp> sorry, they changed it to default on, which is probably right, but I don't always notice it. 2019-10-14T13:33:33 < jadew> the edges are not fully routed? 2019-10-14T13:33:38 < jadew> everything else looks really good tho 2019-10-14T13:34:24 < jadew> oh, that section next to JP1 is full of mouse bites, isn't it? 2019-10-14T13:34:32 < jadew> and they're probably sanding them off 2019-10-14T13:34:58 < jadew> doesn't look that bad 2019-10-14T13:36:34 < karlp> just the little bit by r13 is the panel bites 2019-10-14T13:37:03 < jadew> and that waviness next to JP1? 2019-10-14T13:37:05 < karlp> I hadn't even noticed the burr at jp1. 2019-10-14T13:37:29 < karlp> that culd feasibly be my shitty file, or a wavy cut on their end. 2019-10-14T13:37:49 < jadew> right, could be the router being pushed too fast 2019-10-14T13:40:17 < karlp> it's not my file at least, at least this time that edge is a single line in the edge cuts. 2019-10-14T13:51:09 < karlp> re 32bit ticks in freertos, I can use tickhook to do my own, but at least from the api, that won't work with tickless idle. you just get called once, not with the number of ticks since last time. 2019-10-14T13:57:05 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-14T13:58:22 < kakimir64> yes hello 2019-10-14T14:01:03 < kakimir64> I cannot select the saleae clone with pulseview. It finds it. I select it and press ok but device field doesn't update 2019-10-14T14:01:43 < kakimir64> I have them plugdev rules set 2019-10-14T14:05:06 < kakimir64> yes 2019-10-14T14:05:11 < kakimir64> I remember now 2019-10-14T14:05:29 < kakimir64> I need to launch from terminal so I can see all the shit 2019-10-14T14:06:10 < kakimir64> ftw. it has Saleae Logic selected now right from start 2019-10-14T14:06:39 < kakimir64> Setting up sigrok-firmware-fx2lafw (0.1.6-1) ... 2019-10-14T14:06:42 < kakimir64> maybe it was this 2019-10-14T14:12:28 < kakimir64> I wonder if I set sampling number to gigasamples if it uses my plentyful ram 2019-10-14T14:16:39 < rajkosto> what else would it use ? 2019-10-14T14:16:44 < rajkosto> the saleae have no ram on them 2019-10-14T14:25:53 -!- tprrt [~tprrt@217.114.204.178] has quit [Ping timeout: 276 seconds] 2019-10-14T14:33:10 < kakimir64> harddrive allocation 2019-10-14T14:33:16 < kakimir64> with ram buffer 2019-10-14T14:33:30 < kakimir64> there is an option for 1terasamples 2019-10-14T14:43:15 < kakimir64> nothing is working 2019-10-14T14:43:35 < kakimir64> dis good 2019-10-14T14:47:36 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-14T14:50:56 -!- sk_tandt [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 265 seconds] 2019-10-14T14:59:11 < kakimir64> trying to fix my freertos mess 2019-10-14T14:59:53 < kakimir64> I got configASSERT from xQueueGiveFromISR 2019-10-14T15:00:02 < karlp> great. enable stack checking = 2 in freertos, crashes in memset trying to paint ram. leave it off, works fine. 2019-10-14T15:13:47 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-14T15:37:56 < srk> \o/ standby with sram2 retention and iwdg reset working \o/ 2019-10-14T15:49:48 < kakimir64> karlp: how does it crash? 2019-10-14T15:50:13 < kakimir64> I wonder what should I set to that when I have freertos-mpu 2019-10-14T15:52:32 < kakimir64> void __malloc_lock(struct _reent *x) { vTaskSuspendAll(); };void __malloc_unlock(struct _reent *x) { (void)xTaskResumeAll(); }; 2019-10-14T15:52:47 < kakimir64> is this ok at all? 2019-10-14T15:53:02 < karlp> I dunno. I'm not getting anywhere. busfault impreceise leading to hardfault, always at the same thing. 2019-10-14T15:53:21 < karlp> inside memset, and a pendsv comes in, when it's alreday in one. 2019-10-14T15:53:54 < kakimir64> what heap you use karlp? 2019-10-14T15:57:36 < karlp> heap_1 at the moment. 2019-10-14T15:58:03 < karlp> it's a bit boring that it doesn't fail if I dont' use stack checking = 2. stack checking = 1, works fine. 2019-10-14T16:00:50 < kakimir64> why heap_1? 2019-10-14T16:04:28 < srk> why another one? you need malloc and free? 2019-10-14T16:21:58 -!- roomba is now known as doomba 2019-10-14T16:24:06 < kakimir64> I wonder why it takes so long for PLL to be locked 2019-10-14T16:24:25 < kakimir64> it gets locked eventually 2019-10-14T16:24:30 < kakimir64> but it's like 10seconds 2019-10-14T16:26:27 < kakimir64> can it even take so long? 2019-10-14T16:28:35 -!- talsit [foobar@gromit.mixdown.ca] has joined ##stm32 2019-10-14T16:38:19 -!- grummund [~grummund@unaffiliated/grummund] has joined ##stm32 2019-10-14T16:38:37 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 240 seconds] 2019-10-14T16:41:15 < karlp> https://zerobin.net/?a427c512546933df#b2sSoVl68Vs91cpMjzA30sxFQ3geMLM99vd8BesnyTQ= just says that psr was 6100600e, so it was PendSV irq that was inside it. makes no sense. 2019-10-14T16:41:51 < kakimir64> maybe I have higher frequency crystal that it should be 2019-10-14T16:42:13 < kakimir64> and I'm running the chip at mad 75Mhz 2019-10-14T16:42:20 < kakimir64> instead of 72Mhz 2019-10-14T16:42:34 < mawk> if you don't input the good frequencies it will have a hard time I guess 2019-10-14T16:44:54 < kakimir64> http://www.nadler.com/embedded/newlibAndFreeRTOS.html I'll use this guys heap implementation 2019-10-14T16:45:12 < kakimir64> time to update it too. I have 2017 version 2019-10-14T16:45:30 < mawk> if you plan on using dynamic memory sure 2019-10-14T16:45:42 < mawk> but do you really need it ? 2019-10-14T16:45:50 < mawk> you can get a small printf implem that doesn't use malloc and so on 2019-10-14T16:52:54 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has quit [Quit: ZNC 1.7.5 - https://znc.in] 2019-10-14T16:53:54 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has joined ##stm32 2019-10-14T16:55:21 < karlp> hrm, fixed by making systick be at a lower priority. 2019-10-14T17:02:15 < kakimir64> sounds like "fixed" 2019-10-14T17:05:10 < karlp> not really, it gets done automatically by the xPortStartScheduler() call 2019-10-14T17:05:19 < karlp> problem is more that I had already enabled teh interrupt 2019-10-14T17:09:29 < karlp> dropping systick setup from my own code also fixes it, and is less code, so win overall. 2019-10-14T17:12:01 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-14T17:12:57 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 240 seconds] 2019-10-14T17:12:58 -!- BrainDamage_ is now known as BrainDamage 2019-10-14T17:13:28 < karlp> yep. always best when deleting code is the answer 2019-10-14T17:14:48 -!- jon101249 [~jon1012@pha75-18-88-178-181-173.fbx.proxad.net] has quit [Ping timeout: 246 seconds] 2019-10-14T17:15:15 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-14T17:17:43 < kakimir64> I have freertos running now apparently 2019-10-14T17:17:48 < kakimir64> doesn't get to assert 2019-10-14T17:18:20 < kakimir64> 258% heap usage looks bad though 2019-10-14T17:18:50 < karlp> counted how? 2019-10-14T17:19:01 * karlp notes to update his demo code. 2019-10-14T17:20:17 < Ecco> heya 2019-10-14T17:21:23 < kakimir64> dis MCUxpresso shows freertos heap usage in such window thing 2019-10-14T17:21:30 < kakimir64> viewer 2019-10-14T17:21:37 < kakimir64> idk. how it works 2019-10-14T17:23:31 < kakimir64> I can view tasks here 2019-10-14T17:23:35 < kakimir64> IDLE and uptime task 2019-10-14T17:24:42 < kakimir64> uptime task is not counting 2019-10-14T17:24:47 < kakimir64> it's blocked 2019-10-14T17:24:55 < Ecco> I'd like to use an USB Type C connector 2019-10-14T17:25:00 < Ecco> I need to connect CC1 and CC2 to ground 2019-10-14T17:25:14 < Ecco> quesiton : do I need something special to handle ESD on those resistors? 2019-10-14T17:25:36 < kakimir64> surge resistors 2019-10-14T17:25:40 < kakimir64> if you fancy 2019-10-14T17:25:48 < Ecco> what's that? 2019-10-14T17:25:58 < kakimir64> resistor that is more resistant to surging 2019-10-14T17:26:08 < kakimir64> and derate less when surged 2019-10-14T17:26:34 < Ecco> oh, ok 2019-10-14T17:26:40 < kakimir64> usually can take powers orders of magnitudes higher than normal ones 2019-10-14T17:27:05 < Ecco> mind giving me an example on digikey or mouser? 2019-10-14T17:27:20 < kakimir64> and voltages multiple times higher but then as you need to use something really small you run into maximum voltages 2019-10-14T17:27:59 < Ecco> :-D 2019-10-14T17:33:28 < kakimir64> I don't think I have tick running 2019-10-14T17:35:26 < kakimir64> configUSE_TIMERS do I need this? 2019-10-14T17:40:41 < kakimir64> is IDLE task stack size fixed during operation pros@freertos? 2019-10-14T17:42:33 < kakimir64> this says it has used 100/504Bytes and I have set #define configMINIMAL_STACK_SIZE ( (uint16_t) 128 ) 2019-10-14T17:42:52 < kakimir64> where does that 504 come from? 2019-10-14T17:43:26 < kakimir64> sorry IDLE task 68/504Bytes 2019-10-14T17:43:41 < kakimir64> uptime task 100 2019-10-14T17:44:45 < kakimir64> maybe these tools are not combatible with my garbage 2019-10-14T17:45:11 < kakimir64> will see when I get uart task back up and some stats out 2019-10-14T17:45:39 < kakimir64> #define configOVERRIDE_DEFAULT_TICK_CONFIGURATION 1 whoops 2019-10-14T17:46:10 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-14T17:46:30 < karlp> stack size is in words, not bytes. 2019-10-14T17:47:27 < karlp> you need USE_TIEMRS if you're ... using timers :) (like xTimerCreate() methods) 2019-10-14T17:47:57 < kakimir64> yes 2019-10-14T17:48:14 * karlp ponders when to decide "i'm alive" for iwdg resetting in a freertos world. 2019-10-14T17:49:29 < kakimir64> so will my IDLE task stack change runtime? 2019-10-14T17:49:42 < kakimir64> usage* 2019-10-14T17:50:20 < kakimir64> why its 504 and not 512? 2019-10-14T17:51:15 < kakimir64> it stores some 2word header in there? 2019-10-14T17:52:38 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has joined ##stm32 2019-10-14T17:54:05 < kakimir64> my uptime task and IDLE task have same task number 0xa5a5a5a5 2019-10-14T17:54:33 < kakimir64> according to this viewer 2019-10-14T17:55:37 -!- sk_tandt_ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Ping timeout: 240 seconds] 2019-10-14T17:57:46 < jadew> Cracki, I just looked into annealing acrylic and it seems pretty easy, you just have to heat it to 80 °C, and hold it there for an hour for each mm of thickness 2019-10-14T17:57:46 < karlp> your viewer is broken then. 2019-10-14T17:57:57 < karlp> a5a5a is what they paint the stack with if you turn on stack overflow checking 2019-10-14T17:58:07 < karlp> what version of freertos are you using? 2019-10-14T17:58:18 < karlp> "easy" 2019-10-14T17:58:41 < jadew> then cooling to below 60 °C has to be done at 15 minutes per mm of thickness 2019-10-14T17:59:16 < jadew> karlp, in the sense that you put it in the oven and you take it out several hours later :P 2019-10-14T18:00:10 < jadew> a snail made of entropy 2019-10-14T18:08:42 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has quit [Remote host closed the connection] 2019-10-14T18:11:26 < jadew> https://www.aliexpress.com/item/1000005913175.html 2019-10-14T18:14:07 < jadew> https://www.aliexpress.com/item/4000118464175.html 2019-10-14T18:14:13 < jadew> I swear I was looking for electronics 2019-10-14T18:15:52 < jadew> NSFW btw 2019-10-14T18:17:03 < emeb> anything is SFW if properly overlaid w/ fruit icons. 2019-10-14T18:17:50 < jadew> indeed 2019-10-14T18:18:47 < kakimir64> I enlarged heap to 4kb and now sbrk asserts 2019-10-14T18:18:48 < rajkosto> https://images.sshnuke.net/2019-10-14_17-18-45_lkjBPAwG0.png 2019-10-14T18:18:56 < kakimir64> moved stack to other bank 2019-10-14T18:19:10 < kakimir64> *block 2019-10-14T18:28:24 < kakimir64> oh sweet 2019-10-14T18:28:46 < kakimir64> this heap implementation assumes that stack and heap are next to each other 2019-10-14T18:29:20 < kakimir64> char* limit = (xTaskGetSchedulerState()==taskSCHEDULER_NOT_STARTED) ? stack_ptr : // Before scheduler is started, limit is stack pointer (risky!) &__HeapLimit-ISR_STACK_LENGTH_BYTES; // Once running, OK to reuse all remaining RAM except ISR stack (MSP) stack 2019-10-14T18:37:47 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-14T18:40:02 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 265 seconds] 2019-10-14T18:40:02 -!- BrainDamage_ is now known as BrainDamage 2019-10-14T18:48:07 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-14T18:57:49 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Ping timeout: 268 seconds] 2019-10-14T18:58:17 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-14T19:04:20 < bitmask> what do you do with unconnected pins on a mcu 2019-10-14T19:04:24 < bitmask> tie em to ground? 2019-10-14T19:04:30 < zyp> clip them off 2019-10-14T19:04:43 < Ecco> well I"d say #1 avoid having them in the first place 2019-10-14T19:04:47 < Ecco> order the smallest package that fits 2019-10-14T19:04:51 < Ecco> it's cheaper and smaller 2019-10-14T19:04:53 < zyp> bullshit 2019-10-14T19:05:05 < zyp> smaller doesn't imply cheaper 2019-10-14T19:05:15 < Ecco> well, from my measurements 2019-10-14T19:05:17 < Ecco> it did 2019-10-14T19:05:26 < zyp> also qfp32 takes the same area as qfp48 2019-10-14T19:05:31 < bitmask> this is the smallest I can do, theres only 3 pins, maybe 2 2019-10-14T19:05:33 < Ecco> but again, I haven't had the opportunity to work with to many 2019-10-14T19:06:03 < Ecco> Well I'd say just leave them floating 2019-10-14T19:06:21 < bitmask> its not like they make mcus with every pin count :) 2019-10-14T19:06:25 < Ecco> sure 2019-10-14T19:06:31 < rajkosto> bitmask, tie them to something, gnd or vccio 2019-10-14T19:06:40 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-14T19:06:42 < rajkosto> if you dont intend to solder to them afterwards 2019-10-14T19:06:47 < rajkosto> in a board fixup 2019-10-14T19:07:07 < zyp> that sounds pointless 2019-10-14T19:07:13 < zyp> serious answer: just leave them floating 2019-10-14T19:07:25 < bitmask> ok 2019-10-14T19:07:29 < bitmask> 2 to 1 2019-10-14T19:07:35 < bitmask> floating wins 2019-10-14T19:07:42 < rajkosto> if you do that, dont enable the input gpio buffer on those pins 2019-10-14T19:08:02 < zyp> also true, but I don't usually bother 2019-10-14T19:08:16 < Ecco> problem with non-floating : if you accidentally set the GPIO to output, you risk a short circuit 2019-10-14T19:08:17 < rajkosto> if you tie them to gnd and also set them as gpio output to gnd 2019-10-14T19:08:21 < rajkosto> they ground your chip more :O 2019-10-14T19:08:33 < rajkosto> same with vcc 2019-10-14T19:08:38 < rajkosto> but it doesnt matter 2019-10-14T19:08:57 < rajkosto> it also prevents the pad from getting scraped off with rough soldering 2019-10-14T19:09:05 < bitmask> this whole schematic only took me a day, I'm surprised, it took so much longer to make the individual component symbols and footprints 2019-10-14T19:09:10 < Ecco> Maybe from a mechanical standpoint, if you leave the pins floating, then you have vey tiny pads on the PCB 2019-10-14T19:09:26 < bitmask> laying out the components will take much longer though 2019-10-14T19:09:38 < Ecco> bitmask: what are you building? 2019-10-14T19:09:41 < rajkosto> yeah if i leave the tiny pads and resolder the board many times or try to get the chip off, they can leave 2019-10-14T19:09:49 < zyp> who cares about scraping off unconnected pads anyway 2019-10-14T19:09:58 < zyp> not like you're gonna miss them in the event that happens 2019-10-14T19:09:59 < rajkosto> they can move around and short other pads 2019-10-14T19:10:07 < Ecco> zyp: yeah, but those could get in the way 2019-10-14T19:10:11 < rajkosto> then i cant get them out of the spot they are shorting other pads from 2019-10-14T19:10:41 < bitmask> heated jacket v2, this is my first time using a bare mcu and not just soldering breakouts to a pcb 2019-10-14T19:10:48 < zyp> idk, I haven't had any pads falling off in the first place 2019-10-14T19:11:05 < Ecco> one reason to leave the pins floating : you can solder a wire straight on the PCB if you need to patch something 2019-10-14T19:11:12 < zyp> yep 2019-10-14T19:11:17 < zyp> that I've done 2019-10-14T19:11:19 < Ecco> zyp: probably because you know how to handle an iron :) 2019-10-14T19:11:20 < rajkosto> thats what i meant by board fixup 2019-10-14T19:14:09 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-14T19:14:27 -!- learning1 [~pi@121.121.98.53] has quit [Ping timeout: 240 seconds] 2019-10-14T19:15:50 < bitmask> I still need to figure out some stuff, the toughest thing is choosing input capacitor for this MP1593 buck. The datasheet isn't very helpful for that. 2019-10-14T19:18:32 < Ecco> What are SBU pins for in USB type C? 2019-10-14T19:19:06 -!- learning1 [~pi@121.121.98.53] has joined ##stm32 2019-10-14T19:19:15 < zyp> sideband use 2019-10-14T19:19:21 < Ecco> yeah, but what does that mean? 2019-10-14T19:19:33 < Ecco> Would it be safe to use those as UART RX/TX? 2019-10-14T19:20:02 < zyp> I believe function is subject to negotiation 2019-10-14T19:20:47 < zyp> so I figure it'd be safe if you negotiated using them for UART before enabling the output 2019-10-14T19:21:01 < Ecco> interesting 2019-10-14T19:24:10 < zyp> also you need to keep sorted out which of SBU1 and SBU2 are TX and RX, since that depends on plug orientation :) 2019-10-14T19:25:09 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-14T19:40:21 -!- sk_tandt__ [~sk_tandt@net-5-88-141-17.cust.vodafonedsl.it] has quit [Read error: Connection reset by peer] 2019-10-14T19:42:33 < karlp> nah, just turn it over if it doesn't work :) 2019-10-14T19:47:10 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-14T20:01:54 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Remote host closed the connection] 2019-10-14T20:14:07 < bitmask> https://i.imgur.com/rvcOHDV.png 2019-10-14T20:14:28 < bitmask> Did I do a decent job laying things out and breaking things up? go easy on me 2019-10-14T20:16:43 < Steffanx> AVR bitmask ? 2019-10-14T20:16:47 < bitmask> yea 2019-10-14T20:16:59 < bitmask> no need for stm for this 2019-10-14T20:17:30 < bitmask> and its my first time using a bare mcu, figured avr would be easier 2019-10-14T20:18:54 < bitmask> Theres 3 or so holes where inductors will go, I havent picked out what I'm using yet so I haven't made the footprints/symbols 2019-10-14T20:20:38 < PaulFertser> Bare stm32 is easier than avr. First, you do not need to connect resonator at all to get it started, second, UART is enough to flash it, third, if something goes wrong (like on avr when you use wrong "fuse" bytes) it's easier to recover by reprogramming. Eh? 2019-10-14T20:21:03 < bitmask> fair enough 2019-10-14T20:21:12 < bitmask> too late though :) 2019-10-14T20:23:57 < bitmask> hows the schematic look for a noob 2019-10-14T20:25:08 < srk> altium makes it look pro :D 2019-10-14T20:26:21 < PaulFertser> 5V UART :/ 2019-10-14T20:26:58 < bitmask> ? 2019-10-14T20:27:12 < PaulFertser> bitmask: I remember some megas needed SPI pins for ISP and some others used TDI/TDO, have you checked that you route the right pins to the ISP connector? 2019-10-14T20:27:34 < PaulFertser> bitmask: you have your avr powered by 5V Vcc right, so it's uart will be 5V too. 2019-10-14T20:29:08 < bitmask> I thought tdi and tdo were for jtag? 2019-10-14T20:31:26 < bitmask> and should I use a voltage divider for serial? didnt realize 5v was a problem 2019-10-14T20:33:36 < kakimir64> PROs: https://paste.ee/p/g127k: why the limit before taskScheduler is started is different than after it? Line 12. 2019-10-14T20:39:30 < zyp> presumably because after the scheduler is started, all threads have their own stack 2019-10-14T20:39:49 < zyp> before the scheduler is started, you're running a single thread on the main stack 2019-10-14T20:41:08 < kakimir64> I had Heap at start of RAM3 and stack in end of RAM3 2019-10-14T20:41:40 < kakimir64> I moved stack to RAM2 2019-10-14T20:42:17 < kakimir64> and this happened so.. what should I do? 2019-10-14T20:45:23 < kakimir64> what has the stack to do with this anyway? 2019-10-14T20:45:41 < zyp> well, duh 2019-10-14T20:46:13 < zyp> when you have the stack and the heap growing towards each other in the same area, the maximum area available to the heap is all area not consumed by the stack 2019-10-14T20:46:17 < zyp> only that. 2019-10-14T20:46:48 < zyp> if you have a dedicated area for the heap, the limit would be the end of that area 2019-10-14T20:46:56 < kakimir64> I have 2019-10-14T20:46:59 < zyp> so either hack that sbrk or write your own 2019-10-14T20:47:22 < kakimir64> I mean.. why wouldn't you have dedicated area? 2019-10-14T20:48:08 < zyp> because the mcu I use only has one contiguous area that I can throw everything into? 2019-10-14T20:49:26 < kakimir64> is there an actual benefit? 2019-10-14T20:49:36 < zyp> to which approach? 2019-10-14T20:49:51 < kakimir64> to have shared stack/heap? 2019-10-14T20:50:22 < kakimir64> main stack is not used after freertos scheduler is up? 2019-10-14T20:50:28 < zyp> well, yeah, if I got 64k of ram I don't have to care whether I need 16k of stack and 48k of heap or 48k of stack and 16k of heap 2019-10-14T20:50:39 < zyp> as long as I've got room for both I'm good 2019-10-14T20:50:54 < kakimir64> hmm 2019-10-14T20:51:26 < kakimir64> isn't there mechanisms that require size for those and check overflow? 2019-10-14T20:51:45 < zyp> dude, that's what you're looking at 2019-10-14T20:52:06 < kakimir64> indeed 2019-10-14T20:52:12 < kakimir64> that is the only one? 2019-10-14T20:52:19 < zyp> idk 2019-10-14T20:52:45 < kakimir64> I think all heap related is in this file though 2019-10-14T20:52:50 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-14T20:56:10 < kakimir64> what is that ISR_STACK_LENGTH for? 2019-10-14T20:56:33 < zyp> what do you think? 2019-10-14T20:56:49 < kakimir64> is stack grows beyond what is was after scheduler was started? 2019-10-14T20:57:22 < zyp> doesn't parse 2019-10-14T20:57:49 < qyx> did you deink? 2019-10-14T20:59:19 -!- renn0xtk9 [~max@2a02:810d:1540:2448:175:e601:a692:3a55] has joined ##stm32 2019-10-14T21:00:17 < kakimir64> Pushes xPSR, PC, r0, r1, r2, r3, r12, and LR on selected stack. 2019-10-14T21:01:33 < kakimir64> I have never had other than main stack 2019-10-14T21:03:43 < kakimir64> I take it that I need to keep that in the code 2019-10-14T21:04:02 < kakimir64> but just overwrite the limit variable to be heaplimit-isr_stack... 2019-10-14T21:06:12 < kakimir64> it works 2019-10-14T21:06:16 < kakimir64> whatever 2019-10-14T21:06:23 < kakimir64> bbl> 2019-10-14T21:07:51 < zyp> … 2019-10-14T21:08:21 < zyp> if you're not sharing area between stack and heap, you don't need any stack bullshit in your sbrk 2019-10-14T21:26:48 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-14T21:39:04 < PaulFertser> bitmask: some parts (e.g. atmega128) use PDI/PDO for ISP, some (e.g. atmega8) use MOSI/MISO. 2019-10-14T21:39:24 < PaulFertser> bitmask: re voltage divider, depends on what you plan to attach to that serial. 2019-10-14T21:39:41 < PaulFertser> bitmask: btw, what battery do you plan to power the device from? 2019-10-14T21:39:46 < bitmask> I checked, it uses mosi/miso for serial and tdi/tdo for jtag 2019-10-14T21:39:54 < bitmask> 3s lipo 2019-10-14T21:40:26 < bitmask> the serial is for debug 2019-10-14T21:40:42 < bitmask> so some usb serial thingamajig 2019-10-14T21:40:42 < PaulFertser> bitmask: there's not much difference between using a 3.3V LDO and a 5V LDO so why not do everything in 3.3V? 2019-10-14T21:41:23 < PaulFertser> Beware the maximum frequency though, I remember some parts need it lower if 3.3V power is used. 2019-10-14T21:41:25 < bitmask> didn't think it mattered, figured avr stuff is usually 5v 2019-10-14T21:41:56 < PaulFertser> Well you want it to be future-proof right :) 2019-10-14T21:43:40 < bitmask> why does 3.3v make it future proof 2019-10-14T21:44:15 < PaulFertser> To connect to 3.3V peripherals. 2019-10-14T21:44:38 < PaulFertser> Probably to a 3.3V UART WiFi or bluetooth bridge or something like that. 2019-10-14T21:44:39 < bitmask> oh yea 2019-10-14T21:44:52 < PaulFertser> Or to later rebuild it using stm32 or stm8. 2019-10-14T21:45:00 < bitmask> I really dont think I need to worry about it for this project but i'll consider it 2019-10-14T21:45:22 < bitmask> I planned on only using a 16MHz xtal anyway so I think that is fine for 3.3v 2019-10-14T21:46:59 < PaulFertser> I noticed 20MHz that looked weird 2019-10-14T21:47:12 < bitmask> why weird 2019-10-14T21:47:25 < PaulFertser> I remember some parts were limited to 8MHz for 3.3 and 16MHz for 5V. But meh, avr. 2019-10-14T21:47:59 < bitmask> this chip can do 20 2019-10-14T21:58:41 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-14T22:07:17 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Remote host closed the connection] 2019-10-14T22:15:42 < Steffanx> no way atmega128 uses that PaulFertser. the atmega128 is VERY old. xmega does. 2019-10-14T22:15:59 < Steffanx> and newer avrs use some sort of 1 pin version of that 2019-10-14T22:18:18 < PaulFertser> Steffanx: you want me to download the datasheet to check? You can try it yourself, get the datasheet, scroll to ISP section. 2019-10-14T22:19:14 < PaulFertser> Steffanx: uses what? PDI/PDO? That's the names for the JTAG signals. 2019-10-14T22:19:35 < PaulFertser> No idea why they called it that way. But I have some idea why they used it for SPI ISP. 2019-10-14T22:22:22 < PaulFertser> "SPI Serial Programming Pin Mapping" 2019-10-14T22:22:45 < Steffanx> Am I confused now with the stuff they call PDI? 2019-10-14T22:23:34 < PaulFertser> Yes :) 2019-10-14T22:24:06 < bitmask> pdi is two line interface on xmega 2019-10-14T22:24:10 < Steffanx> Why did they reuse that name -_- 2019-10-14T22:24:32 < PaulFertser> Because nice acronym 2019-10-14T22:25:05 < PaulFertser> Hm, and JTAG interface is called properly, those are another pins, now I checked. 2019-10-14T22:25:16 < bitmask> yea 2019-10-14T22:25:19 < bitmask> tdi/tdo 2019-10-14T22:25:28 < Steffanx> :D 2019-10-14T22:30:48 < bitmask> you guys have a goto brand for inductors? 2019-10-14T22:44:27 < kakimir64> is it normal to have 3.36kilobyte heap usage with only idle and uptime tasks? 2019-10-14T22:47:39 < kakimir64> it's time to put it on diet I think 2019-10-14T22:51:20 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-14T22:53:23 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-14T22:56:26 < qyx> there are some in zypsnips/inductors 2019-10-14T23:00:20 -!- renn0xtk9 [~max@2a02:810d:1540:2448:175:e601:a692:3a55] has quit [Remote host closed the connection] 2019-10-14T23:01:48 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e17b:11c4:a3b3:6064] has joined ##stm32 2019-10-14T23:03:26 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-14T23:08:37 < karlp> there are? 2019-10-14T23:09:13 < karlp> oh, yeah, but it's more about how they work and shti not like a particular "just use this" 2019-10-14T23:09:45 < karlp> more than I though, neat. 2019-10-14T23:16:10 < kakimir64> I enabled vApplicationIdleTask and I think mcu went to permanent sleep 2019-10-14T23:16:20 < kakimir64> WFI 2019-10-14T23:17:34 < catphish> all your linux are belong to us https://thehackernews.com/2019/10/linux-sudo-run-as-root-flaw.html 2019-10-14T23:18:56 < kakimir64> sudo had update today 2019-10-14T23:19:44 < catphish> well that makes sense 2019-10-14T23:20:12 < catphish> the description says: even when the "sudoers configuration" explicitly disallows the root access 2019-10-14T23:20:53 < catphish> but my reading of the report implies that in fact explicitly disallowing root is precisely the *cause* of the problem 2019-10-14T23:21:17 < catphish> oh, i think it allows a bypass of restricted comands too 2019-10-14T23:21:24 < catphish> even when root access is allowed 2019-10-14T23:21:50 < kakimir64> that privilege elevation looks absolutelly trivial 2019-10-14T23:22:02 < kakimir64> even I could do it 2019-10-14T23:23:41 < kakimir64> E: The repository 'http://packages.linuxmint.com tara Release' is no longer signed.N: Updating from such a repository can't be done securely, and is therefore disabled by default. 2019-10-14T23:24:40 < kakimir64> but yeah update came from ubuntu-security 2019-10-14T23:25:05 < catphish> basically, if you're allowed to sudo to UID "-1" then you can run anything 2019-10-14T23:25:17 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-14T23:25:22 < kakimir64> basically there is no limits 2019-10-14T23:25:42 < catphish> that means if you have access to run limited commands as any user, or as "!root" which includes -1, then you can run anything as root 2019-10-14T23:26:09 < kakimir64> yes privilege elevation 2019-10-14T23:28:54 < catphish> yes, but very specifically what i described 2019-10-14T23:29:04 < catphish> you have to have sudo permission to run commands as UID -1 2019-10-14T23:37:37 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-14T23:38:03 < kakimir64> how do I see what is blocking my task? 2019-10-14T23:40:54 < kakimir64> it's not lack of tick blocking it 2019-10-14T23:43:05 < catphish> what OS (not that i know the answer) 2019-10-14T23:45:10 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Read error: Connection reset by peer] 2019-10-14T23:45:45 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-14T23:47:08 < kakimir64> freertos 2019-10-14T23:49:15 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-14T23:51:09 < antto> bitcoin_task() 2019-10-14T23:52:21 < kakimir64> my board is not mining if I cannot get simple uptime going 2019-10-14T23:54:48 < antto> it's too busy being idle 2019-10-14T23:54:58 < antto> i told u to kill teh idle task ;P~ 2019-10-14T23:55:05 < antto> iz iz evil 2019-10-14T23:57:50 < kakimir64> it needs to get full busy 2019-10-14T23:59:17 < Steffanx> Hello antto. How was your day --- Day changed Tue Oct 15 2019 2019-10-15T00:01:15 < antto> still soldering 2019-10-15T00:01:37 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-15T00:01:44 < antto> i think i finally found PCB connectors for muh 10mm sandwich 2019-10-15T00:02:06 < Steffanx> Leaded sodder? 2019-10-15T00:02:17 < antto> que? 2019-10-15T00:06:54 < Steffanx> Si si. 2019-10-15T00:09:17 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-15T00:21:31 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-15T00:24:06 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-15T00:29:10 < kakimir64> sometimes it just doesn't work at all 2019-10-15T00:29:14 < kakimir64> it hardfaults 2019-10-15T00:29:27 < kakimir64> I click restart and then it's going again 2019-10-15T00:39:11 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-15T00:47:39 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-15T00:50:59 < catphish> ffs, youtube is killing my USB bandwidth 2019-10-15T00:51:18 < catphish> i can't prove or explain it, but i can't stream data when it's playing 2019-10-15T01:01:53 < catphish> seems to be a cpu bottleneck, poor show intel 2019-10-15T01:02:03 < catphish> cpu in performance mode, problem gone 2019-10-15T01:05:29 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e17b:11c4:a3b3:6064] has quit [Ping timeout: 276 seconds] 2019-10-15T01:13:14 < catphish> oh, libusb can receive more than one packet at once! 2019-10-15T01:13:18 < catphish> that'll help 2019-10-15T01:18:13 < catphish> all that time i spent trying to optimize, usb the bottleneck was probably my host! 2019-10-15T01:31:05 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-15T01:31:09 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-15T01:33:51 < Jan-> so what do you guys think about the stm32 arduino esque library 2019-10-15T01:36:07 < catphish> personally i love arduino, what i do tends to prefer low level control and masochism, so it rarely suits me, but if it does what you need, then awesome 2019-10-15T01:36:21 < catphish> afaik stm32 supports actual arduino 2019-10-15T01:36:23 < Jan-> yeah that's what worries me 2019-10-15T01:36:45 < catphish> but figure out your requirements first 2019-10-15T01:37:47 < catphish> if you're doing mad things with accurate timings and DMA, arduino is probably not for you, if you're using a serial port to control some traffic lights, go for it :) 2019-10-15T01:38:49 < Jan-> the thing is as far as I know every way of writing stm32 code uses some sort of intermediate layer 2019-10-15T01:39:01 < catphish> not the way i do it 2019-10-15T01:39:16 < catphish> but most sane people, yes 2019-10-15T01:39:31 < catphish> and again, it probably depends on requirment 2019-10-15T01:39:35 < Jan-> so you just hit the registers 2019-10-15T01:39:38 < catphish> yes 2019-10-15T01:39:53 < Jan-> doesn't that basically mean you end up sort of writing your own HAL by default 2019-10-15T01:40:05 < catphish> depends what you want to do 2019-10-15T01:40:30 < catphish> but you write functions to abstract register tasks, sure 2019-10-15T01:41:34 < catphish> ST provides nice header files with pointers to all the registers, so it's not *that* hard to drive them directly if you're comfortable with the reference manual 2019-10-15T01:41:43 < Jan-> that's how I write avr code 2019-10-15T01:42:39 < Jan-> I did look at the stm32 register definitions 2019-10-15T01:42:45 < Jan-> I mean there's a lot of it but that's whatever 2019-10-15T01:43:16 < catphish> the document you'd want is the reference manual, it's specific to the particular stm32 you're using 2019-10-15T01:43:27 < catphish> and it's extremely well written and helpful 2019-10-15T01:43:52 < catphish> if you're comfortable with hitting registers, it's a nice way to go, but not for everyone 2019-10-15T01:44:07 < Jan-> there's nothing wrong with a combined approach I guess 2019-10-15T01:44:35 < Jan-> if you don't use the hal stuff it won't get compiled in 2019-10-15T01:44:37 < catphish> sure 2019-10-15T01:44:39 < Jan-> I guess? 2019-10-15T01:44:51 < Jan-> so if you hit some special thing you need to do you can always hit the registers 2019-10-15T01:44:52 < catphish> i actually have no idea, i've never used it 2019-10-15T01:44:56 < catphish> but i assume it's efficient enough 2019-10-15T01:45:13 < catphish> but afaik yes, you can always hit registers if you want 2019-10-15T01:45:18 < catphish> it's not like it can stop you 2019-10-15T01:46:26 < Jan-> in arduino if you replace their main() then you switch off most of it 2019-10-15T01:46:31 < Jan-> well all of it really 2019-10-15T01:46:59 < catphish> yeah, i used to do AVR by putting my own register code in an arduino main() loo 2019-10-15T01:47:02 < catphish> *loop 2019-10-15T01:47:24 < catphish> mostly because i couldn't be bothered to work out how to program it without the IDE 2019-10-15T01:47:41 < Jan-> I like the way in platformio I can just use notepad to write the code 2019-10-15T01:47:43 < Jan-> that's good 2019-10-15T01:48:10 < catphish> i've not looked at that 2019-10-15T01:48:20 < catphish> i just went the hardcore route and i'm happy with it 2019-10-15T01:48:42 < catphish> although trying to learn the USB peripheral nearly killed me 2019-10-15T01:49:34 * Jan- doesn't need that 2019-10-15T01:49:41 * Jan- ...until the asshats demand usb 2019-10-15T01:50:01 < catphish> lol 2019-10-15T01:50:48 < catphish> well if you have questions, i have answers :) 2019-10-15T01:51:42 < Jan-> I guess the first thing I need is serial output 2019-10-15T01:51:49 < Jan-> just for debug 2019-10-15T01:52:03 < Jan-> but I guess I should just use the arduino stuff 2019-10-15T01:56:07 < qyx> you can always just use a notepad 2019-10-15T01:56:26 < Jan-> a whatpad? 2019-10-15T02:03:14 < Jan-> so um where even is the serial output on this thing 2019-10-15T02:11:36 < qyx> what thing exactly 2019-10-15T02:12:02 < qyx> it is on the selected UART/USART 2019-10-15T02:12:51 < Jan-> Sooo we need to connect a usb to serial widget to some of the pins 2019-10-15T02:13:03 < Jan-> I think the term for it is "Black STM32F407VE" 2019-10-15T02:13:35 < qyx> yeah you can do it that way 2019-10-15T02:15:12 < qyx> some of the pins = see the datasheet, there is an alternate function tzble 2019-10-15T02:15:27 < qyx> select the suitab 2019-10-15T02:15:30 < qyx> e port 2019-10-15T02:19:00 < qyx> meh android 2019-10-15T02:22:22 < mawk> Jan-: you have a tty to usb adapter ? 2019-10-15T02:22:26 < mawk> uart to usb 2019-10-15T02:25:17 < Jan-> I think so 2019-10-15T02:25:20 < Jan-> have to check 2019-10-15T02:31:34 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-15T02:59:20 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-15T03:14:51 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-15T03:17:08 <@englishman> zyp, these ikea lights are pretty trick. what problems did you have with it that made you want to add on your own stuff? 2019-10-15T03:26:58 -!- smvoss [~smvoss@207.191.220.92] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-15T03:27:25 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-15T03:27:51 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-15T03:30:47 -!- smvoss [~smvoss@207.191.220.92] has quit [Client Quit] 2019-10-15T03:31:15 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-15T03:33:34 -!- smvoss [~smvoss@207.191.220.92] has quit [Client Quit] 2019-10-15T03:33:59 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-15T03:45:02 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-15T03:45:44 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-15T04:09:57 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Ping timeout: 240 seconds] 2019-10-15T04:19:00 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-15T04:33:08 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has joined ##stm32 2019-10-15T04:35:54 < ac_slater> hey guys. I have a weird issue that's hard to debug. I have a library that puts structures (packed) in FLASH via `__attribute__((section(foo))`. On Linux, I can iterate this 'section' easily as GCC spits out `__start_foo` and `__stop_foo` extern variables I can use to iterate/inspect at the section 2019-10-15T04:37:05 < ac_slater> In linux, I can use this library no issue and the contents of the section are fine. I can iterate simply. On my STM32F4 and F7 (nucleo dev board), The second iteration always goes to an address starting in 0xffee 2019-10-15T04:37:30 < ac_slater> which makes me thing something about iterating a FLASH section is causing an issue. I haven't tried moving it to RAM yet. 2019-10-15T04:37:53 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-15T04:38:06 < ac_slater> My question is really: 1) should there be a difference accessing 'section' mappings between linux and stm32? I feel like this answer is always "yes" 2019-10-15T04:43:12 < ac_slater> I was thinking it could be alignment issue 2019-10-15T04:53:37 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-15T04:59:59 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-15T05:03:29 < ac_slater> objdump shows my created section 'foo' has alignment of 2**3 ... 2019-10-15T05:24:34 < ac_slater> I think i figured it out. My structure definition and my section mappings had different alignments 2019-10-15T05:33:36 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 268 seconds] 2019-10-15T05:35:09 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-15T06:05:57 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-15T06:28:59 -!- fc5dc9d4_ [~quassel@p5B081211.dip0.t-ipconnect.de] has joined ##stm32 2019-10-15T06:31:32 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-15T06:32:57 -!- fc5dc9d4 [~quassel@p5B0816D6.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-15T06:46:37 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has quit [Ping timeout: 240 seconds] 2019-10-15T07:22:40 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-15T07:23:55 -!- fenugrec_ [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-15T07:41:16 -!- Kitlith [~Kitlith@theobromine.kitl.pw] has quit [Quit: ZNC 1.6.5 - http://znc.in] 2019-10-15T07:46:53 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-15T07:50:13 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-15T07:51:44 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-15T07:51:44 -!- day__ is now known as day 2019-10-15T07:53:11 < R2COM> sup 2019-10-15T07:55:45 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-15T08:04:00 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-15T09:00:56 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-15T09:12:02 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-15T09:13:37 -!- fenugrec_ [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-15T09:23:17 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 240 seconds] 2019-10-15T09:30:36 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-15T09:39:49 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-15T10:04:34 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-15T10:07:49 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-15T10:09:00 -!- ohama [ohama@cicolina.org] has quit [Ping timeout: 265 seconds] 2019-10-15T10:16:46 -!- ohama [ohama@cicolina.org] has joined ##stm32 2019-10-15T10:21:47 * day observes that it appears to run better now 2019-10-15T10:24:41 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has joined ##stm32 2019-10-15T10:29:17 -!- ohama [ohama@cicolina.org] has quit [Disconnected by services] 2019-10-15T10:29:36 -!- ohama [ohama@cicolina.org] has joined ##stm32 2019-10-15T10:32:48 -!- m4ssi [~massi@host138-130-static.62-82-b.business.telecomitalia.it] has quit [Quit: Leaving] 2019-10-15T10:37:37 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-15T10:38:28 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-15T11:18:23 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-15T11:50:07 -!- grevaillot [~moot@zed.f00.be] has joined ##stm32 2019-10-15T12:11:58 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-15T12:15:34 * Ecco is puzzled by the SBU pins on USB-C cable 2019-10-15T12:15:43 < Ecco> So there are two SBU pins, SBU1 and SBU2 2019-10-15T12:15:53 < Ecco> but the cable is reversible 2019-10-15T12:16:00 -!- tprrt [~tprrt@217.114.204.178] has quit [Client Quit] 2019-10-15T12:16:10 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-15T12:16:12 < Ecco> so can we use both pins? Or is just one of them guaranteed to be connected? 2019-10-15T12:16:28 -!- tprrt [~tprrt@217.114.204.178] has quit [Client Quit] 2019-10-15T12:16:44 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-15T12:18:31 < zyp> they will both be connected, but you'll have to mux them for plug orientation just like the SS* signals 2019-10-15T12:18:46 < Ecco> oh, ok, makes sense 2019-10-15T12:18:57 < Ecco> so extra circuitry is needed to make sure they're "reversed" if needed 2019-10-15T12:18:59 < Ecco> is that correct? 2019-10-15T12:19:47 < zyp> yes, in both ends 2019-10-15T12:21:16 < zyp> each end only knows the orientation of the plug in its own socket, so they are both responsible for reversing the signals 2019-10-15T12:21:32 < zyp> Haohmaru, better suggest that for USB-D 2019-10-15T13:28:25 < qyx> also great for FO 2019-10-15T13:28:35 < qyx> a single FO in the middle of TRRRRRS 2019-10-15T13:29:54 < qyx> also, offtopic, zyp-blogging about the new home 2019-10-15T13:30:08 < qyx> I ordered some 30m LC-LC SM optic patchcords 2019-10-15T13:30:24 < qyx> to run them from the roof to the basement 2019-10-15T13:31:13 < qyx> the fuk outher sheath is made of polyurethane 2019-10-15T13:31:55 < qyx> it was insanely hard to get it into a conduit 2019-10-15T13:32:08 < qyx> fiberptics 2019-10-15T13:52:08 < zyp> probably shouldn't buy PU-coated cables for pulling through conduit 2019-10-15T13:52:23 < zyp> I should probably check that before I buy the ones I'm gonna pull 2019-10-15T13:54:28 < qyx> yeah I didn't check 2019-10-15T13:54:57 < qyx> they were those used for hanging FTTH in the air 2019-10-15T13:55:15 < qyx> 3mm outer dia, simplex SM fiber 2019-10-15T13:56:02 -!- catphish [~charlie@unaffiliated/catphish] has joined ##stm32 2019-10-15T13:56:04 < zyp> the shit I've looked at is PVC-coated 2019-10-15T13:56:04 < catphish> morning 2019-10-15T13:56:09 < catphish> kinky 2019-10-15T13:56:15 < zyp> yup 2019-10-15T13:57:07 < catphish> i ordered some nice outdoor cable yesterday, no pvc nonsense for me 2019-10-15T13:58:34 < catphish> https://www.kenable.co.uk/en/networking/cables-patch-and-reels/network-cables-reels/8535-external-cat5e-outdoor-use-copper-ethernet-network-cable-reel-utp-50m-black-008535-5055383485352.html 2019-10-15T14:00:04 < zyp> I hate that full copper has become something you need to advertise, not something that is to be expected 2019-10-15T14:00:24 < catphish> that company offers both clearly labelled 2019-10-15T14:00:51 < catphish> i always get copper, but i guess there's no harm in having CCA if it works 2019-10-15T14:03:23 < zyp> well, duh, I figure the issue is it doesn't :p 2019-10-15T14:23:28 < karlp> wat. gdb's using 100% cpu sitting there doing nothing after loading. 2019-10-15T14:23:49 < karlp> run, ctrl-c, back to ~0% 2019-10-15T14:37:57 -!- ekaologik [~quassel@p4FF16B3A.dip0.t-ipconnect.de] has joined ##stm32 2019-10-15T14:44:55 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-15T15:01:36 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-15T15:37:52 -!- talsit [foobar@gromit.mixdown.ca] has left ##stm32 [] 2019-10-15T15:41:03 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 265 seconds] 2019-10-15T15:48:08 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has joined ##stm32 2019-10-15T15:48:11 -!- con3 [~kvirc@146.232.65.240] has quit [Ping timeout: 276 seconds] 2019-10-15T15:55:23 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has quit [Quit: WeeChat 2.6] 2019-10-15T16:19:50 < day> zyp: but is it oxygen free copper? 2019-10-15T16:20:53 < zyp> dunno, not gonna breathe it 2019-10-15T17:04:41 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-15T17:05:22 < con3> guys need some help.. 2019-10-15T17:05:42 < con3> the irq handlers are for a range of interrupts , i.e. EXTI4_15_IRQHandler 2019-10-15T17:06:19 < con3> so as far as i remember all i need to do is have an if within the handler to check whether the interrupt was triggered 2019-10-15T17:06:28 < con3> i.e. if(EXTI->PR & (1<<5)) 2019-10-15T17:06:47 < con3> and then reset the pending bit within the if statement :EXTI->PR &= ~(1 << 5); 2019-10-15T17:07:32 < con3> This doesnt seem to work though.. The pending bit doesnt seem to reset correctly and it ends up retriggering continously 2019-10-15T17:08:31 < con3> Haohmaru: what i said? 2019-10-15T17:09:03 < con3> Haohmaru: i just remember someone on the stm32 community telling me to do it this way months ago 2019-10-15T17:09:11 < con3> how else would i do it? 2019-10-15T17:11:04 < qyx> con3: isnt there a dedivated reset register? 2019-10-15T17:11:09 < qyx> dedicated 2019-10-15T17:11:21 < con3> qyx: not that ive seen, let me quickly recheck 2019-10-15T17:13:39 < con3> Haohmaru: heh looks like its the same. I missed that 2019-10-15T17:14:01 < con3> one sec 2019-10-15T17:14:11 < qyx> con3: reset by writing 1 2019-10-15T17:14:25 < qyx> it is in the refman 2019-10-15T17:14:52 < qyx> right in the EXTI_PR register description 2019-10-15T17:14:53 < con3> qyx: yeah i missed that in the ref manual. might be lacking sleep. just weird because when the bit is 1 it means it's set? 2019-10-15T17:14:54 < grevaillot> con3: afair you have to write the bit, not to clear it :) 2019-10-15T17:14:59 < grevaillot> yeah 2019-10-15T17:15:13 < con3> This bit is set when the selected edge event arrives on the external interrupt line. This bit is cleared by writing a 1 to the bit. 2019-10-15T17:15:20 < con3> 1: selected trigger request occurred 2019-10-15T17:15:22 < qyx> yeash 2019-10-15T17:15:41 < con3> so if i write one to the register it'll reset to 0? 2019-10-15T17:15:47 < con3> or am i going mad ? :/ 2019-10-15T17:15:48 < grevaillot> yep 2019-10-15T17:16:13 < con3> ok cool, thanks guys, appreciate it. completely missed it 2019-10-15T17:16:38 < con3> yeah Haohmaru, i just missed it in the ref manual. remember reading the ref manual yesterday but lack of sleep is getting to me 2019-10-15T17:16:51 < con3> thanks for the help and sorry about that, should've read better 2019-10-15T17:19:20 < qyx> such a speshul Haohmaru 2019-10-15T17:21:14 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-15T17:34:23 -!- fenugrec_ [~fenugrec@104.221.51.116] has joined ##stm32 2019-10-15T17:38:05 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-15T17:42:43 -!- talsit [foobar@gromit.mixdown.ca] has joined ##stm32 2019-10-15T17:44:03 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-15T17:44:09 < bitmask> hey 2019-10-15T17:44:21 < bitmask> what do you use for decoupling caps 2019-10-15T17:44:33 < bitmask> 0.1uF or 1uF 2019-10-15T17:45:08 < qyx> yeah 2019-10-15T17:45:27 < qyx> for stm32, 100n for every VDD/VSS pair 2019-10-15T17:45:49 < qyx> and one 1u for a group 2019-10-15T17:46:02 < qyx> VDDA the same 2019-10-15T17:46:06 < bitmask> datasheet I believe shows 100n but arduino schematic shows 1u 2019-10-15T17:47:32 < qyx> for avr? 2019-10-15T17:48:02 < bitmask> yup 2019-10-15T17:48:16 < bitmask> I'm thinking 100n on each pair and a 4.7 tantalum 2019-10-15T17:48:19 < qyx> but avr is slower 2019-10-15T17:48:32 < bitmask> I dont need speed for this 2019-10-15T17:48:55 < qyx> the caps depend 2019-10-15T17:49:24 < bitmask> not sure where the 1u came from, i'm gonna switch em to 100n 2019-10-15T17:49:48 < qyx> just clear them 2019-10-15T17:50:02 < qyx> decoupling caps are overrated 2019-10-15T17:50:33 < bitmask> heh 2019-10-15T17:51:06 < bitmask> the other shit show I'm dealing with is inductors, thats even harder to get answers than caps 2019-10-15T17:51:26 < bitmask> I think I'm gonna do a ferrite bead on AVcc but nothing on Vcc 2019-10-15T17:51:32 < bitmask> and just use the inductors on the switcher side 2019-10-15T17:52:33 < qyx> build, measure, check 2019-10-15T17:52:42 < qyx> throw away 2019-10-15T17:52:45 < qyx> and then aghain 2019-10-15T17:52:46 < bitmask> I dont have a scope unfortunately 2019-10-15T17:52:51 < bitmask> that would make this easier 2019-10-15T17:52:57 < qyx> then nah 2019-10-15T18:11:43 < bitmask> its not, just curious about the discrepancy on something like that 2019-10-15T18:13:41 < bitmask> 10uF for analog? Thought it was just the 100nF 2019-10-15T18:14:18 < emeb> heh - buddy just noticed that prices for STM32F103CBT6 at usual distributors is going thru the roof. eg ~$5-7/ea 2019-10-15T18:14:29 < emeb> meanwhile they're $1/ea @ aliexpress. 2019-10-15T18:14:59 < bitmask> right 2019-10-15T18:17:19 < emeb> and prerelease pricing of H750 parts is also ~$5 at distys. Is ST trying to send a message? 2019-10-15T18:21:03 < emeb> I've often wondered at what kind of cost shifting is involved in keeping the "China Price" so low. 2019-10-15T18:21:50 < emeb> Kinda like how big pharma charges US patients kilobux for drugs that they sell for pennies in foreign countries. 2019-10-15T18:25:27 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-15T18:30:24 -!- Streaker [~Streaker@unaffiliated/streaker] has joined ##stm32 2019-10-15T19:00:23 -!- learningc [~pi@121.122.105.195] has joined ##stm32 2019-10-15T19:02:40 -!- learning1 [~pi@121.121.98.53] has quit [Ping timeout: 268 seconds] 2019-10-15T19:11:19 < specing> I don't think you'd die from eating a non-RoHS one, either 2019-10-15T19:11:31 < specing> pretty sure you'd poop it out intact 2019-10-15T19:13:51 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-15T19:25:39 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-15T19:38:49 -!- catphish [~charlie@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-15T19:39:25 < antto> will its price drop then? 2019-10-15T19:40:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-15T19:54:37 < Jan-> hello :) 2019-10-15T19:54:47 < antto> hai derr 2019-10-15T19:55:48 < antto> vast ist kookin'? 2019-10-15T19:56:08 < Jan-> instant noodles 2019-10-15T19:56:16 < Jan-> I'd be living a bachelor lifestyle this week, if I was a bachelor. Or a guy. 2019-10-15T19:56:16 < antto> yuck 2019-10-15T19:56:20 < qyx> a vi-fon? 2019-10-15T19:56:30 < qyx> or vin-shu 2019-10-15T19:56:45 < antto> oh wait, aren't u miss assembler? 2019-10-15T19:57:03 < Jan-> am I wha? 2019-10-15T19:57:13 < qyx> if you miss the assembler 2019-10-15T19:57:30 < BrainDamage> Jan-: you can considerably improve instant noodles by replacing the powder with proper sauce / ingredients 2019-10-15T19:57:37 < antto> there was supposedly some chick on #avrs struggling with assembler x_x 2019-10-15T19:57:49 < Jan-> But I like the combination of minimum-cost chilli powder and thickening agents. 2019-10-15T19:58:04 < Jan-> antto: I am some chick, but I have never in my life written assembler. 2019-10-15T19:58:38 < antto> yeah, i wouldn't recommend it even to muh enemy 2019-10-15T19:58:52 < Jan-> I did once copy and paste some assembly code into an avr program 2019-10-15T19:59:00 < antto> who hasn't 2019-10-15T19:59:12 < Jan-> my cat 2019-10-15T19:59:29 < Jan-> she's more interested in where the next salmon dinner is coming from 2019-10-15T19:59:30 < antto> not.. yet 2019-10-15T19:59:38 < srk> I've did __asm volatile ("wfi") recently :D 2019-10-15T20:00:13 < antto> mine's blocking muh other keyboard 2019-10-15T20:00:23 < antto> i can't kicad like this 2019-10-15T20:00:43 < emeb> Quick way to add some veg to your diet: Cook the instant ramen, then dump a bag of shredded cabbage on top & let it steam for a few before stirring in and eating. 2019-10-15T20:00:47 < emeb> Yum. 2019-10-15T20:01:04 < Jan-> but that's REAL food 2019-10-15T20:01:04 < antto> i got beans 2019-10-15T20:01:40 < emeb> only 1/2 real food. 2019-10-15T20:01:54 < antto> i think i have a pic of teh beans 2019-10-15T20:02:49 < antto> https://i.imgur.com/UbddN.jpg 2019-10-15T20:02:53 < Jan-> I did find some seriously outdated instant ramen the other day 2019-10-15T20:03:05 < Jan-> I'm not sure the ramen really goes off but I think the sauce in the little packet is probably a bit far gone 2019-10-15T20:03:34 < Jan-> maybe I can repurpose the noodles 2019-10-15T20:03:57 < antto> when i haven't got wut to eat, i can just get some biscuits or peanutz or chipz or crackerz 2019-10-15T20:04:16 < Jan-> that's really bad for you 2019-10-15T20:04:20 < Jan-> all carbs 2019-10-15T20:04:26 < antto> wut's gonna happen to me? 2019-10-15T20:04:36 * Jan- pokes antto's belly 2019-10-15T20:04:49 < antto> i'm probably 59kg 2019-10-15T20:05:32 < Jan-> that's OK unless you're four feet tall 2019-10-15T20:05:43 < antto> feet? 2019-10-15T20:06:09 < Jan-> how tall are you 2019-10-15T20:06:20 < antto> i'm about 1.78m i think 2019-10-15T20:06:37 < qyx> and how many feet do you have? 2019-10-15T20:06:45 < antto> two.. wait 2019-10-15T20:06:52 < antto> yeah two, confirmed 2019-10-15T20:07:09 < Jan-> I'm five feet one and 105lb 2019-10-15T20:07:15 < antto> pls 2019-10-15T20:07:23 < antto> wut r these units 2019-10-15T20:08:06 < antto> another option is assembling together some sandwichez 2019-10-15T20:08:14 < Jan-> 152cm, 46kg 2019-10-15T20:08:43 < antto> tiny 2019-10-15T20:09:06 < Jan-> hey. 2019-10-15T20:09:08 < Jan-> I'm not tiny. 2019-10-15T20:09:19 < antto> i've been 63kg-ish max.. when i worked @ teh warehauz 2019-10-15T20:09:21 < Jan-> who am I kidding I shop in the hobbit section :( 2019-10-15T20:09:28 < antto> hahaah 2019-10-15T20:15:48 < Jan-> do not mock 2019-10-15T20:16:04 * Jan- fetches the stepladder 2019-10-15T20:16:08 * Jan- climbs up 2019-10-15T20:16:11 * Jan- bops antto on the nose 2019-10-15T20:16:15 * Jan- puts the stepladder away 2019-10-15T20:16:20 < Jan-> man that's a process 2019-10-15T20:19:26 < BrainDamage> you can always learn to walk in stilts 2019-10-15T20:19:44 < Jan-> I just thank god for japanese shoe fashion 2019-10-15T20:19:58 < Jan-> they singlehandedly brought into style shoes with incredibly gigantic soles 2019-10-15T20:20:09 < Jan-> so they and I can pass among humans 2019-10-15T20:20:17 * Jan- kids 2019-10-15T20:20:20 * Jan- wears sneakers a lot 2019-10-15T20:21:13 * Jan- is five two in fila disruptors :D 2019-10-15T20:22:19 * karlp has tom66d this board, but it works finally. 2019-10-15T20:22:29 < karlp> fuck me. nothing went right today :) 2019-10-15T20:22:47 < Jan-> I know what you mean 2019-10-15T20:22:53 * Jan- offers karlp a dip in the cookie jar 2019-10-15T20:23:16 < karlp> who the fuck though 1Mhz spi was a suitable max speed too hey? atmel? w.t.f 2019-10-15T20:23:31 < karlp> "good for EMC/EMI" my arse. 2019-10-15T20:23:44 < antto> https://i.imgur.com/7kwMjmX.mp4 2019-10-15T20:24:31 < qyx> just relay 2019-10-15T20:24:47 < qyx> relax even 2019-10-15T20:24:57 < antto> karlp eh? which chip is that? 2019-10-15T20:24:59 < qyx> karlp: any good .is metal? 2019-10-15T20:25:02 < karlp> yah man, it works, I can go back to a nice desk now. 2019-10-15T20:25:08 < karlp> and stop by atvr.is on the way home :) 2019-10-15T20:25:20 < karlp> qyx: hrm, what sort of metal? 2019-10-15T20:25:46 < karlp> dimma? solstafir? or un misere? what direction from those three and I can probably find you some more. 2019-10-15T20:25:53 < karlp> "muck" as well. 2019-10-15T20:26:05 < antto> https://i.imgur.com/AcaL0aY.jpg 2019-10-15T20:26:14 < karlp> antto: atm90e32 2019-10-15T20:26:31 < antto> at90 or atm90? 2019-10-15T20:26:50 < karlp> what I said. 2019-10-15T20:26:58 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-15T20:27:21 < qyx> karlp: more like viking/folk, eluveitie, xandria, leaves eyes 2019-10-15T20:27:21 < antto> i can't even check that cuz muh cat.. >:( 2019-10-15T20:28:14 < antto> oh, dafuq is that chip about 2019-10-15T20:28:34 < karlp> qyx: will try and find some. can't think of any off the top of my head though. 2019-10-15T20:29:31 < qyx> I'll check yours 2019-10-15T20:38:47 * qyx @ youtube - the winter in norway 2019-10-15T20:39:46 < antto> https://i.imgur.com/nYcPFMm.jpg 2019-10-15T20:40:30 < kakimir64> have unboxings gone too far 2019-10-15T20:40:44 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-15T20:42:12 < BrainDamage> hellraiser franchise is the original unboxing video 2019-10-15T20:42:52 < Jan-> that's funny :) 2019-10-15T20:44:02 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-15T20:52:31 < bitmask> if adc input is above the reference voltage will it damage it or just read maxxed out, say 2V with a ref of 1.1V 2019-10-15T20:53:51 < kakimir64> are freertos task priorities related at all to interrupt priorities? 2019-10-15T20:54:25 < kakimir64> ie. order of priorities low-high or high-low 2019-10-15T20:55:27 < qyx> no 2019-10-15T20:56:11 < qyx> bitmask: no to you too 2019-10-15T20:56:15 < qyx> see abs max ratings 2019-10-15T20:56:35 < bitmask> no what, I asked an or question 2019-10-15T20:56:38 < bitmask> :) 2019-10-15T20:57:44 < qyx> it will not damage it 2019-10-15T20:57:48 < bitmask> ok awesome 2019-10-15T20:57:52 < bitmask> makes things much easier 2019-10-15T21:05:17 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-15T21:12:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-15T21:23:43 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-15T21:38:23 < kakimir64> freertos pros: semaphore from ISR causes task1 to continue running, task2 with vTaskDelay never continues running 2019-10-15T21:38:43 < kakimir64> tick interrupt is running 2019-10-15T21:39:32 < qyx> do you use FromISR API? 2019-10-15T21:40:00 < kakimir64> #define INCLUDE_xSemaphoreGiveFromISR 1 ? 2019-10-15T21:40:08 < qyx> yes 2019-10-15T21:40:13 < kakimir64> I have that 2019-10-15T21:40:23 < qyx> I mean 2019-10-15T21:40:32 < qyx> how do you signal the semaphore 2019-10-15T21:40:40 < kakimir64> from uart interrupt 2019-10-15T21:40:51 < qyx> kaki pls 2019-10-15T21:40:55 < qyx> paste the line 2019-10-15T21:41:06 < qyx> are you using xSemaphoreGiveFromISR? 2019-10-15T21:41:37 < qyx> I am not asking if you defined that macro to be 1 2019-10-15T21:43:27 < kakimir64> static void uart_rx_waitCallback(uint32_t err_code, uint32_t n __attribute__((unused))){ if(err_code == 0){ uart_rx_gotCallback = true; xSemaphoreGiveFromISR(uart_rx_semaphore, NULL); //Give semaphore so receiver task can continue with received data }} 2019-10-15T21:43:44 < kakimir64> sweet lines 2019-10-15T21:44:15 < qyx> wait what 2019-10-15T21:44:19 < qyx> thats not an isr 2019-10-15T21:44:33 < qyx> when is the function called 2019-10-15T21:44:43 < qyx> shouldn't you call uart_rx_wait or whatever 2019-10-15T21:44:47 < kakimir64> from rom 2019-10-15T21:44:57 < qyx> and when a character is received, the callback gets called 2019-10-15T21:45:02 < kakimir64> true 2019-10-15T21:45:24 < kakimir64> it gives semaphore and task starts running happily 2019-10-15T21:46:09 < kakimir64> am I getting somewhere? 2019-10-15T21:46:57 < kakimir64> this is interrupt related problem that vTaskDelay doesn't return but semaphorefromISR causes task to unblock right? 2019-10-15T21:47:29 < qyx> this may be anything 2019-10-15T21:47:42 < kakimir64> sweet 2019-10-15T21:47:58 < qyx> why are you using such unknown rom code? 2019-10-15T21:48:14 < qyx> how do you how does it behave with freertos 2019-10-15T21:48:33 < kakimir64> i dont know and I dont know 2019-10-15T21:48:50 < kakimir64> it had problems I cannot remember 2019-10-15T21:48:58 < qyx> sweet :> 2019-10-15T21:49:40 < kakimir64> for some reason I have skecthed some context saving code for that uart interrupt 2019-10-15T21:49:47 < kakimir64> and commented it out 2019-10-15T21:49:53 < kakimir64> cannot remember what it was 2019-10-15T21:49:53 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has joined ##stm32 2019-10-15T21:50:08 < kakimir64> I found the handler 2019-10-15T21:50:11 < kakimir64> void UART0_IRQHandler(void){ LPC_UARTD_API->uart_isr(uartHandle);} 2019-10-15T21:51:00 < kakimir64> uart code is currently totally disabled until I get uptime going 2019-10-15T22:07:13 < antto> https://i.imgur.com/Q15CT4o.mp4 2019-10-15T22:08:17 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-15T22:09:00 < srk> I've managed to generate freertos callgraph for one of my test apps http://48.io/~rmarko/random/callgraph2.png .. with -fdump-rtl-expand and https://www.gson.org/egypt/egypt.html 2019-10-15T22:09:36 < srk> https://mcuoneclipse.com/2015/08/21/gnu-static-stack-usage-analysis/ 2019-10-15T22:09:57 < srk> -fstack-usage ! 2019-10-15T22:10:05 < zyp> looks similar to the stuff I were dicking around with some years ago 2019-10-15T22:10:39 < zyp> https://bin.jvnv.net/file/j4TBZ.png <- except my firmware were a lot smaller 2019-10-15T22:11:14 < srk> that's mostly generated code 2019-10-15T22:11:56 < kakimir64> I don't see rtos zyp 2019-10-15T22:12:29 < zyp> you don't see what? 2019-10-15T22:12:58 < srk> zyp: this is a slightly higher-level representation of the same app http://48.io/~rmarko/random/tower.png 2019-10-15T22:16:47 < kakimir64> will try that avstack srk thanks for the tip 2019-10-15T22:17:53 < zyp> kakimir64, are you trying to statically calculate how big each thread stack needs to be? 2019-10-15T22:17:57 < sync> emeb: the H750 looks seksi 2019-10-15T22:18:28 < emeb> sync: right? all except for that 128kB flash thing 2019-10-15T22:18:33 < sync> yeah 2019-10-15T22:19:02 < emeb> guess you could hang a QSPI on it and XIP 2019-10-15T22:19:05 < emeb> but still... 2019-10-15T22:19:08 < sync> ram is 8 times bigger than flash :DD 2019-10-15T22:19:32 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-15T22:19:38 < emeb> we live in strange times 2019-10-15T22:19:42 < Steffanx> but at least its value line. 2019-10-15T22:19:45 < catphish> morning 2019-10-15T22:19:53 < qyx> yes, it is written somewhere 2019-10-15T22:19:57 < qyx> the F750 too 2019-10-15T22:20:06 < qyx> it is meant to be running from a qspi flash 2019-10-15T22:20:18 < emeb> interesting. 2019-10-15T22:20:34 < qyx> there were some AN benchmarking performance between sram, internal flas and XIP from QSPI 2019-10-15T22:20:46 < catphish> i have a complaint, why does the f103 not have USB DFU firmware 2019-10-15T22:20:50 < qyx> there was only a couple of % difference 2019-10-15T22:20:59 < zyp> catphish, because it's old as fuck 2019-10-15T22:21:17 < zyp> if you want new features, use a new chip 2019-10-15T22:21:22 < catphish> zyp: idk, other devices in the series have it 2019-10-15T22:21:32 < zyp> what series? 2019-10-15T22:21:32 < emeb> catphish: zyp is right 2019-10-15T22:21:36 < Steffanx> its still old as fuck :P 2019-10-15T22:21:46 < srk> or load cm3 bootloader on it 2019-10-15T22:21:47 < qyx> so what new chip should I use for 240x160px lcd, RS485, CAN and some audio output 2019-10-15T22:22:11 < srk> F7xx 2019-10-15T22:22:20 < zyp> qyx, f401? 2019-10-15T22:22:21 < srk> or G4 2019-10-15T22:22:27 < srk> too old :D 2019-10-15T22:23:17 < srk> G4 has no bxCAN anymore but FDCAN 2019-10-15T22:23:24 < catphish> zyp: specifically, the RM for the F103 says "In connectivity line devices the boot loader can be activated through one of the following interfaces: USART1, USART2 (remapped), CAN2 (remapped) or USB OTG FS in Device mode (DFU: device firmware upgrade)." 2019-10-15T22:23:54 < catphish> so clearly the concept of USB DFU existed, and the part has USB, but they decided to exclude it 2019-10-15T22:23:55 < qyx> zyp: I would use it but it has no canz, F411/412/413 do have 2019-10-15T22:24:14 < catphish> i'm sure there's a reason, it's just annoying 2019-10-15T22:24:15 < zyp> catphish, more like, it was only supported on dwc_otg devices, not st_usbfs 2019-10-15T22:24:26 < catphish> zyp: oh, that might be why 2019-10-15T22:24:27 < zyp> qyx, then those 2019-10-15T22:25:11 < catphish> zyp: i'm using the l433 for now, was thinking i could cheap out a bit, but seemingly not worth it 2019-10-15T22:25:19 < catphish> the l433 is surprisingly cheap anyway 2019-10-15T22:25:24 < catphish> but it's no f103 :) 2019-10-15T22:25:47 < qyx> I am using L432 regularly, not too fast though 2019-10-15T22:26:00 < kakimir64> hmm my project space doesn't have CMSIS 2019-10-15T22:26:07 < kakimir64> hoe does this even work 2019-10-15T22:26:12 < zyp> catphish, even if you wanna pick something cheaper, you don't have to pick the oldest shit of them all 2019-10-15T22:26:40 < qyx> you can get F0 for cheap and it is still newer than F1 2019-10-15T22:26:43 < kakimir64> I wonder of lacking CMSIS effect on my project 2019-10-15T22:26:54 < qyx> which reminds me 2019-10-15T22:27:14 < qyx> wheres jpa-'s stm32s prices link 2019-10-15T22:27:46 < kakimir64> oh it's in chip_lpc15xx 2019-10-15T22:28:03 < kakimir64> *lpc_chip_15xx 2019-10-15T22:28:13 < zyp> catphish, what are your prices? 2019-10-15T22:29:18 < catphish> zyp: L433 is £3 GPB 2019-10-15T22:29:24 < catphish> *GBP 2019-10-15T22:30:08 < zyp> and F103? 2019-10-15T22:31:06 < catphish> the f103 is a weird part, there must have been some insanely high sales volume in china because they're available at less than their ST cost price 2019-10-15T22:31:33 < catphish> i don't have a price to hand, i'll check 2019-10-15T22:32:19 < zyp> speaking of, I still haven't gotten rid of that pile of 64-pin l053 2019-10-15T22:32:22 < catphish> stupidly the F103 is *more expensive* in the uk 2019-10-15T22:32:59 < catphish> i guess i should ignore the silly "blue pill" pricing, and stick to the official pricing 2019-10-15T22:33:15 < srk> prize is the only reason why it's worth supporting f103 2019-10-15T22:33:24 < qyx> zyp: stm32l053r8tx? 2019-10-15T22:33:39 < catphish> aliexpress has STM32F103C8 for ~ 0.70 2019-10-15T22:33:41 < srk> bluepills might have defective chips or clones, they die easily 2019-10-15T22:34:00 < srk> good for testing / development tho 2019-10-15T22:34:07 < srk> brushing hair 2019-10-15T22:34:22 < zyp> sorry, l052, not 3 2019-10-15T22:34:29 < qyx> without lcd? 2019-10-15T22:34:30 < zyp> STM32L052R8T6 2019-10-15T22:34:50 < catphish> i went with the l433 because it's got crystal-less USB 2019-10-15T22:35:07 < catphish> but it might be worth looking at other USB capable parts, crystals are cheap and more accurate 2019-10-15T22:35:16 < catphish> maybe even an l0 2019-10-15T22:35:19 < kakimir64> #define __MPU_PRESENT 0 2019-10-15T22:35:31 < PaulFertser> qyx: in /topic : http://essentialscrap.com/tips/stm32prices/ 2019-10-15T22:35:32 < zyp> qyx, yeah, I think the lcd interface is the only difference between l052 and l053 2019-10-15T22:35:41 < catphish> zyp: by the way, i discovered that my USB bottleneck was likely my host all along 2019-10-15T22:35:41 < qyx> :( 2019-10-15T22:36:19 < catphish> zyp: i was running libusb with my cpu in power saving mode and 64-byte reads, turns out it can buffer much larger reads in the kernel or library 2019-10-15T22:36:30 < zyp> heh 2019-10-15T22:36:31 < qyx> but even if they were L053, I would want only around 10pcs 2019-10-15T22:36:44 < kakimir64> UM.pdf: Cortex-M3 configuration options: A Memory Protection Unit (MPU) is included. 2019-10-15T22:37:26 < qyx> PaulFertser: O-o too long topic and too narrow FOV 2019-10-15T22:37:36 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-15T22:38:24 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-15T22:42:44 < qyx> stm32l451 is newer and cooler than f401, also cheaper 2019-10-15T22:42:49 -!- ekaologik [~quassel@p4FF16B3A.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-15T22:42:50 < qyx> it seems 2019-10-15T22:45:04 < kakimir64> https://gitlab.com/late347/xyplotter/blob/master/lpc_chip_15xx/inc/cmsis.h#L166 when chip has MPU and cmsis says nah.. why? 2019-10-15T22:46:08 < kakimir64> have you seen such before? 2019-10-15T22:49:34 < Cracki> mpu is part of cortex m spec? 2019-10-15T22:49:41 < Cracki> maybe they fucked up 2019-10-15T22:50:07 < Cracki> maybe that's how they set it to not use that, rather than something better named 2019-10-15T22:50:14 < kakimir64> it's in user manual right under Cortex-M3 configuration 2019-10-15T22:50:31 < kakimir64> MPU included 2019-10-15T22:50:37 < Cracki> that's a random gitlab project 2019-10-15T22:50:55 < kakimir64> indeed but it has same old lpc lib I use 2019-10-15T22:50:57 < Cracki> don't assume that the headers are unchanged 2019-10-15T22:51:33 < kakimir64> I have latest 2019-10-15T23:07:46 < kakimir64> should there be prvSetupTimerInterrupt() in port.c? 2019-10-15T23:12:07 < kakimir64> interesting when scheduler is started it throws MSP to start of the stack? 2019-10-15T23:18:42 < Cracki> you are asking tough questions 2019-10-15T23:22:27 < zyp> kakimir64, why wouldn't it? 2019-10-15T23:22:49 < kakimir64> no I'm just amazed of how smart stuff is 2019-10-15T23:23:07 < zyp> in freertos, once the scheduler is started, the original thread that started the scheduler doesn't exist anymore 2019-10-15T23:23:18 < zyp> so its stack is not needed anymore either 2019-10-15T23:23:19 < kakimir64> I should do that combined heap stack thing 2019-10-15T23:23:30 < zyp> which means it can be emptied 2019-10-15T23:23:37 < kakimir64> I don't know how does it like it across the ram blocks 2019-10-15T23:24:01 < zyp> are they contiguous or not? 2019-10-15T23:24:36 < kakimir64> it's continuous ram address space so is there anything to stop me? 2019-10-15T23:24:56 -!- ekaologik [~quassel@p4FF16B3A.dip0.t-ipconnect.de] has joined ##stm32 2019-10-15T23:25:01 < zyp> if you've only got contigous ram blocks, I'd just treat it as one big block 2019-10-15T23:25:09 -!- ekaologik [~quassel@p4FF16B3A.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-15T23:25:18 < kakimir64> oh now I understand 2019-10-15T23:25:25 < kakimir64> why it's block based 2019-10-15T23:25:30 < kakimir64> you can shut off some of them 2019-10-15T23:25:45 < zyp> yeah, and if you have multiple masters, they can be accessed in parallel 2019-10-15T23:26:13 < kakimir64> so you place stuff that you can generate after returning from sleep to blocks you would otherwise turn off 2019-10-15T23:26:13 < zyp> so in a multi-cpu or a DMA-heavy system, you might want to pay attention to that 2019-10-15T23:26:46 < zyp> I mean, in your case you could write a sbrk that keeps blocks off until you need them 2019-10-15T23:26:58 < kakimir64> I wouldn't 2019-10-15T23:27:01 < zyp> although that seems fairly pointless :p 2019-10-15T23:27:03 < kakimir64> this is powered from mains 2019-10-15T23:27:30 < kakimir64> it means full steam 2019-10-15T23:34:32 < Tordek> oh, btw 2019-10-15T23:34:40 < Tordek> I managed to get 1mhz clock 2019-10-15T23:34:49 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-15T23:34:54 < Tordek> by using the interrupts and removing as much of the HAL as possible 2019-10-15T23:35:31 < Tordek> write straight to BSRR instead of calling HAL_GPIO_Write 2019-10-15T23:35:40 < zyp> Tordek, what are you doing? 2019-10-15T23:35:55 < Tordek> ~~dumb things~~ a SID player 2019-10-15T23:36:07 < kakimir64> USE_PROCESS_STACK symbol is not defined in port.c 2019-10-15T23:36:12 < kakimir64> is this a problem? 2019-10-15T23:36:20 < zyp> who knows 2019-10-15T23:37:37 < kakimir64> https://paste.ee/p/VNY5l 2019-10-15T23:39:37 < kakimir64> I don't know what it means but it does something with the main stack.. I'll need to decode it after night snack 2019-10-15T23:39:51 < zyp> it's just an optimization 2019-10-15T23:40:29 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-15T23:41:05 < zyp> when entering an interrupt, lr will contain a magic value called EXC_RETURN, one of the bits in the value signals if the stacked context resided on msp or psp 2019-10-15T23:42:00 < kakimir64> oh 2019-10-15T23:42:07 < zyp> if you know all stacks are running on psp, you define USE_PROCESS_STACK to skip checking that bit to shave a few instructions off the context switch 2019-10-15T23:42:14 < zyp> all threads, I mean 2019-10-15T23:43:12 < kakimir64> amazing 2019-10-15T23:43:20 < zyp> also, why are you fucking around with this shit? 2019-10-15T23:43:39 < kakimir64> just to set breakpoints 2019-10-15T23:43:44 < zyp> wat 2019-10-15T23:43:44 < kakimir64> to dem handlers 2019-10-15T23:43:56 < zyp> why? 2019-10-15T23:44:03 < kakimir64> chill.. I just check the handlers 2019-10-15T23:44:44 < kakimir64> if those are called 2019-10-15T23:45:08 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-15T23:45:11 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Quit: The Lounge - https://thelounge.github.io] 2019-10-15T23:45:22 < kakimir64> I mean internal mechanisms require those in order to switch tasks? 2019-10-15T23:49:35 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-15T23:51:03 < kakimir64> also I'm learning --- Day changed Wed Oct 16 2019 2019-10-16T00:02:33 < Jan-> ok so 2019-10-16T00:02:38 < Jan-> now I have some time to play with libopencm3 2019-10-16T00:03:51 < Jan-> it is sort of funny that it's really obvious that it's a library and open source 2019-10-16T00:03:55 < Jan-> but not at all obvious what it does 2019-10-16T00:16:03 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-16T00:21:18 < karlp> possibly just skip it then. 2019-10-16T00:21:25 < karlp> no need to play with things that don't appeal to you... 2019-10-16T00:22:21 < qyx> :D 2019-10-16T00:24:27 < Jan-> can anyone suggest why this is failing https://pastebin.com/FwVVQWRN 2019-10-16T00:24:36 < Jan-> am I using the wrong pio command? I'm not sure 2019-10-16T00:25:02 < karlp> qyx: perhaps skálmöld might be suitable for you? 2019-10-16T00:31:45 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-16T00:35:08 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-16T00:36:27 < qyx> karlp: thanks, much better! 2019-10-16T00:36:43 < karlp> goodo. 2019-10-16T00:37:03 < karlp> Jan-: if you're goign to try libopencm3, I'd suggest platformio being one of the worst ways of doing it. 2019-10-16T00:39:15 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has quit [Ping timeout: 240 seconds] 2019-10-16T00:39:53 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-16T00:41:45 < karlp> more joy with zephyr https://github.com/zephyrproject-rtos/zephyr/pull/19808 2019-10-16T00:41:50 < karlp> perhaðs I'm being thin-skinned. 2019-10-16T00:42:15 < karlp> openwrt all over again. 2019-10-16T00:46:54 < qyx> the live concert with the icelandic symphonic orchestra is great 2019-10-16T00:47:40 < karlp> yeah, we went to one of them. 2019-10-16T00:47:54 < karlp> they were stuck on the tip of my tongue earlier, couldn't think of them. 2019-10-16T00:49:02 < karlp> misþyrming are probably to far out in the dark abyss for you, but well regarded 2019-10-16T00:49:53 < zyp> karlp, what part of it is it you take an issue with? 2019-10-16T00:50:27 < zyp> all the duplicated shit? 2019-10-16T00:51:37 < zyp> or people submitting an incomplete PR, or people arguing it's incomplete? 2019-10-16T00:51:48 < zyp> or people approving it? 2019-10-16T00:52:17 < karlp> I guess it's hard to see, but it's the classic, "some people can submit whatever half baked garbage they like and it's approved, others are held to much much much higher standards" 2019-10-16T00:52:49 < zyp> well, yeah, I'm only looking at this PR 2019-10-16T00:53:27 < karlp> yeah, nvm then. 2019-10-16T00:53:38 < zyp> but sure, I get your point 2019-10-16T00:53:41 < karlp> I should be used to it. especially given the corporate names on it. 2019-10-16T00:54:04 < karlp> (mine's the PR linked at the bottom, which got blocked until I'd done what this person had half baked) 2019-10-16T00:54:21 < zyp> on the other hand, isn't it kinda natural that a regular contributor gets more lenient reviews than a new one? 2019-10-16T00:54:51 < karlp> to a point, sure. 2019-10-16T00:54:59 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-16T00:55:18 < zyp> the former can be expected to stick around and clear up their own mess later, the latter probably not so much 2019-10-16T00:55:19 < karlp> but there were _two_ separate people, who pointed out incomplete nature of that one I linked, and it still just got blindly approved by two other people later. 2019-10-16T00:55:52 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-16T00:55:53 < karlp> you also don't _get_ new contributors if you require them to make changes to things that didn't exist at the time. 2019-10-16T00:56:09 < zyp> true 2019-10-16T00:56:27 * karlp shrugs. not doing anything withh zephyr once this lands anyway. had long come to that conclusion, just pushing what I did that was useful 2019-10-16T00:56:44 < con3> welp the pending bit thing is sorted, but my pwm isnt doing jack 2019-10-16T00:56:45 < zyp> what were you looking at using zephyr for? 2019-10-16T00:57:05 < karlp> replacement for freertos+[locm3|laks|cmsis] 2019-10-16T00:57:26 < karlp> same as when I tried out mbed 2019-10-16T00:57:38 < zyp> does it need replacing? 2019-10-16T00:57:59 < karlp> eh, I'd be fine with dropping locm3. 2019-10-16T00:58:18 < karlp> and I'd be fine with getting all the freebies in zephyr stack space vs freertos, withhout getting into amazon's iot world 2019-10-16T00:58:27 < karlp> freertos keeps coming up winner though 2019-10-16T00:58:54 < karlp> I'm having a bit of a struggle switching some stuff that's all grown up as big superloop into nice tasks and notifications/queues though :) 2019-10-16T00:59:11 < zyp> hehe 2019-10-16T00:59:34 < karlp> I'm telling myself it will save me time down the road, and make it simpler to extend things, but... :) 2019-10-16T01:00:14 < zyp> my workstuff is not using freertos optimally either 2019-10-16T01:00:30 < karlp> well, I'm not using freertos _at all_ in workstuff yet, but am planning on it :) 2019-10-16T01:00:42 < zyp> I'm not using freertos in anything but 2019-10-16T01:01:02 < karlp> qyx: Kontinuum too perhaps? 2019-10-16T01:01:15 < zyp> workstuff is a nice mix of nice code and hacked up shit now 2019-10-16T01:01:31 < karlp> auðn and endless dark, and then I'm out :) 2019-10-16T01:01:54 < zyp> last week I added code generation from yaml files 2019-10-16T01:02:04 < zyp> for system parameters 2019-10-16T01:02:04 * karlp hates yaml 2019-10-16T01:02:10 < zyp> I used to 2019-10-16T01:02:20 < zyp> then I realized it's more pleasant than json 2019-10-16T01:02:26 < karlp> "it's readable" "as long as you know _exactly_ the way _we_ intended to do the whitespace" 2019-10-16T01:02:39 < karlp> yeah, it's easier to write if you get it into your mind. 2019-10-16T01:02:39 < zyp> yeah 2019-10-16T01:02:50 < karlp> but if you're not working with it often, it's a fucking train wreck 2019-10-16T01:02:59 < zyp> I used to hate it because I found it unintuitive 2019-10-16T01:03:03 < karlp> "does this get a new line, an indent, or a -. or a :... I'm not sure" 2019-10-16T01:03:19 < karlp> is yaml built into py3 yet? 2019-10-16T01:03:37 < karlp> that was always a big plus for json. it was built into py2.5+ or something 2019-10-16T01:03:49 < zyp> and then I had to dick around with it when I were messing around with some kubernetes shit a few months back, and after I got into the syntax I actually found it quite neat 2019-10-16T01:04:04 < zyp> no, that's the disadvantage, but pyyaml is in pypi 2019-10-16T01:04:21 < karlp> yeah, like I said, perfectly fine if you do enough of it. 2019-10-16T01:04:43 < karlp> I got it enough for doing raml stuff for ws apis, but it's all gone again. I had to edit some for zephyr and it was nightmares all over again. 2019-10-16T01:05:13 < zyp> I find it's more source control friendly than json 2019-10-16T01:05:33 < karlp> because of trailing , stuff? 2019-10-16T01:05:38 < zyp> yes 2019-10-16T01:05:41 < karlp> heh :) 2019-10-16T01:05:41 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-16T01:05:45 < karlp> I fucking hate that about json. 2019-10-16T01:05:49 < zyp> yes 2019-10-16T01:05:51 < karlp> only IE and vim refuse it. 2019-10-16T01:05:56 < karlp> use lua instead :=) 2019-10-16T01:06:01 < zyp> haha 2019-10-16T01:06:02 < karlp> lua tables are great for config files. 2019-10-16T01:06:29 < karlp> all the nice bits of json 2019-10-16T01:07:06 < zyp> as for yaml, not having to mess around with closing braces is also nice 2019-10-16T01:08:21 < zyp> and the extendability is cute 2019-10-16T01:08:46 < zyp> I added an !include thing so I can split stuff into multiple files 2019-10-16T01:09:26 < zyp> https://gist.github.com/joshbode/569627ced3076931b02f <- this thing 2019-10-16T01:09:43 < karlp> everytime I look at kubernetes I'm like, "you want me to write _how_ many fucking files?!" 2019-10-16T01:09:57 < karlp> and then I remember who and what it's fore, and why I don't need it, and I step away again :) 2019-10-16T01:10:02 < zyp> hehe 2019-10-16T01:10:33 < karlp> or maybe I do, and it would make my teeth whiter after all 2019-10-16T01:10:37 < zyp> I just like the fact everything is defined in a file, can be updated by editing the file and can be deleted and recreated again from the file 2019-10-16T01:10:47 < karlp> oh yeah, that concept is great. 2019-10-16T01:11:00 < karlp> I saw it refered to as "cattle vs pets" 2019-10-16T01:11:55 < karlp> I moved some stuff from a local hosted $$$$ server setup to an aws setup, and making it all ansible based was _very_ relaxing. 2019-10-16T01:12:15 < karlp> instead of "where did previous employee x decide to implement this on the server?!" 2019-10-16T01:12:32 < qyx> karlp: Ill check 2019-10-16T01:12:56 < kakimir64> xTickCount is really slow 2019-10-16T01:13:16 < zyp> I wrote an ansible playbook to provision some servers on hetzner cloud and bring up a whole kubernetes cluster on them 2019-10-16T01:14:06 < zyp> now I've been paying for running them a couple of months, but I still haven't had time to put anything useful on it :) 2019-10-16T01:14:14 < karlp> heh. 2019-10-16T01:14:34 < qyx> kubernetes is the nowadays openvz? 2019-10-16T01:14:45 < zyp> not really 2019-10-16T01:14:57 < karlp> I tried to use a container on aws to deploy something tiny, because a) I could and b) that shold be the best right? and it was more expensive than firing up a t2.micro and just apt-installing the same daemon. 2019-10-16T01:15:22 < kakimir64> actually my uptime task is working 2019-10-16T01:15:26 < zyp> sure 2019-10-16T01:15:31 < karlp> if I already had the kubernetes cluster on my own aws ec2 infra, it would have been lighter/cheaper, but useless for 1. 2019-10-16T01:15:32 < kakimir64> just clocks are totally shitted 2019-10-16T01:15:43 < zyp> containers are great 2019-10-16T01:16:05 < zyp> I run plain docker on my home server 2019-10-16T01:16:38 < qyx> I still run libvirt with kvm :X 2019-10-16T01:17:41 < zyp> managing vms are annoying 2019-10-16T01:17:48 < qyx> speaking of home servers 2019-10-16T01:18:12 < qyx> I need to get a mikrotik with 2 SFP 2019-10-16T01:18:40 < zyp> if I wanna run some random shit in docker I can just find a prebuilt image on docker hub and spin it up in half a minute or whatever 2019-10-16T01:18:51 < zyp> for what? 2019-10-16T01:19:09 < karlp> zyp: I use both vms _and_ containers :) 2019-10-16T01:19:16 < qyx> as a home switch/router 2019-10-16T01:19:31 < qyx> fiber is going to the roof for lte stuff 2019-10-16T01:20:03 < qyx> maybe some 2xxx red soho one 2019-10-16T01:25:41 < zyp> two SFPs to the roof? 2019-10-16T01:27:26 < zyp> I don't really like routeros that much, so I rather prefer ubnt 2019-10-16T01:27:45 < zyp> if I needed two SFPs on the router, I'd probably be looking at the ER-12 2019-10-16T01:27:46 < qyx> no, one to the roof and one to some otjer place outside 2019-10-16T01:27:55 < kakimir64> my tick is running at 0.6hz 2019-10-16T01:28:01 < zyp> since I only need one, I'm planning to get the ER-4 2019-10-16T01:28:01 < kakimir64> 1000hz requested 2019-10-16T01:28:14 < kakimir64> time to see how my core cycle count is going 2019-10-16T01:30:17 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-16T01:30:19 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-16T01:46:10 < con3> hmm the pwm passes the pulse value and reloads but no change in the output 2019-10-16T01:47:56 < kakimir64> time to change to internal oscillator 2019-10-16T01:56:27 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has quit [Ping timeout: 240 seconds] 2019-10-16T01:57:53 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-16T01:58:24 < kakimir64> it works 2019-10-16T01:58:27 < kakimir64> uptime works 2019-10-16T01:58:31 < con3> kakimir64: yay 2019-10-16T01:59:02 < kakimir64> I mean I wonder how didn't the chip crash more often 2019-10-16T01:59:20 < kakimir64> I probed mainclk with logic analyzer 2019-10-16T01:59:27 < kakimir64> it was a mess 2019-10-16T01:59:51 < kakimir64> totally mess I thought it must be uart with random break conditions 2019-10-16T02:00:17 < kakimir64> I wonder how did PLL get a lock 2019-10-16T02:01:51 < con3> /me kicks pwm 2019-10-16T02:02:16 < kakimir64> Logc Analyzer says 71,99Mhz 2019-10-16T02:04:59 < kakimir64> how bad idea it is to call freertos functions before scheduler has started? 2019-10-16T02:05:19 < kakimir64> example: xSemaphoreGiveFromISR 2019-10-16T02:06:12 < kakimir64> if I want to use same uart facilities without any mechanisms that check if we are yet in scheduler 2019-10-16T02:06:41 < kakimir64> to say print init strings etc. diagnostical 2019-10-16T02:09:24 < kakimir64> anyway time to turn uart task back on 2019-10-16T02:11:15 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 246 seconds] 2019-10-16T02:12:46 -!- grummund [~grummund@unaffiliated/grummund] has quit [Ping timeout: 265 seconds] 2019-10-16T02:13:33 < con3> wtf 2019-10-16T02:16:55 < catphish> i just ran into a whole new fun challenge of USB, timeout / unexpected going away of the host application :( 2019-10-16T02:17:26 < catphish> like any serial application, the cost can leave the hardware in any number of unknown state by disapearing mid operation 2019-10-16T02:17:56 < Cracki> https://www.youtube.com/watch?v=BrH8bClnlzQ 2019-10-16T02:21:03 < con3> the capture compar interrypt flag does get triggered but the output remains high 2019-10-16T02:21:30 < kakimir64> Cracki: guyes go there with 10k of bullets 2019-10-16T02:21:33 < kakimir64> or more 2019-10-16T02:21:40 < kakimir64> then belt them all 2019-10-16T02:21:45 < Cracki> the "condition" may trigger an interrupt, but if you want it to control a pin, there are a few other bits to set 2019-10-16T02:21:58 < Cracki> kakimir64, ABC news sold this as "turks smoking the kurds" 2019-10-16T02:22:08 < kakimir64> really? 2019-10-16T02:22:10 < Cracki> yes. 2019-10-16T02:22:12 < kakimir64> fake news 2019-10-16T02:22:16 < Cracki> ikr 2019-10-16T02:22:18 < kakimir64> trump was right 2019-10-16T02:22:30 < Cracki> who owns the media? hello 2019-10-16T02:22:52 < con3> Cracki: all i can think of is the capture compare pulse, output polarity, cc1E and finally just ensuring the correct pin is set as alternate function 2019-10-16T02:23:10 < Cracki> hmhm 2019-10-16T02:24:16 < con3> like the pin does get triggered when the pwm activates, it goes high which is right, but when the counter hits the pulse value it does nada 2019-10-16T02:24:36 < Cracki> ah, and you expect it to flip back down again there 2019-10-16T02:24:42 < con3> Cracki: yep 2019-10-16T02:24:47 < con3> and then back up at the arr 2019-10-16T02:24:49 < Cracki> yeah, the upflip and downflip can be independent 2019-10-16T02:24:50 < con3> value 2019-10-16T02:24:53 -!- grummund [~grummund@unaffiliated/grummund] has joined ##stm32 2019-10-16T02:25:00 < con3> Cracki: independent? 2019-10-16T02:25:02 < Cracki> yes 2019-10-16T02:25:19 < Cracki> you can have it do both, or one but not the other, or the other but not the one, ... 2019-10-16T02:25:19 < con3> im literally completely lost as to why it's not flipping 2019-10-16T02:25:39 < con3> wait...how does that get set? 2019-10-16T02:26:01 < Cracki> the right bits :> 2019-10-16T02:26:19 < Cracki> I'll assume you did skim the RM for all bits in all registers related to the whole timer? 2019-10-16T02:26:25 < con3> like i havent seen that anywhere for pwm? what would the point be of not having it do both? 2019-10-16T02:26:36 < con3> ill check again 2019-10-16T02:27:08 < Cracki> good q. point could be that something else also toggles the pin (by poking the right bit in the timer registers), but the timer-flipped edge is important 2019-10-16T02:27:37 < con3> ive checked the ref manual and tutorials but havent seen this. rechecking the rm now 2019-10-16T02:27:52 < Cracki> I think there's also a mode where the event *flips* the level, instead of setting it high or low always 2019-10-16T02:28:19 < kakimir64> my code has hash based command parser 2019-10-16T02:28:22 < con3> maybe i should say this is an stm32f0 2019-10-16T02:28:28 < Cracki> (i know, "event" is a specific thing beside interrupts, but I mean the general idea) 2019-10-16T02:28:52 < Cracki> I am so sorry but this KNOB (!) Creek Gun Range footage is just so hypnotizing 2019-10-16T02:29:17 < kakimir64> it generates hash out of command and then stores it in array with command number macro 2019-10-16T02:30:16 < kakimir64> when command comes in it hashes it and bada bing bada boom 2019-10-16T02:30:40 < kakimir64> it's simple comparison of integers 2019-10-16T02:30:49 < kakimir64> 8o 2019-10-16T02:30:53 < con3> Cracki: this would be in the timer registers right? 2019-10-16T02:30:59 < Cracki> yes 2019-10-16T02:32:19 < Cracki> I found at least 2 rms for F0: F0x1/2/8, and F030x4/6/8/C, F070x6/B 2019-10-16T02:32:32 < con3> https://www.st.com/content/ccc/resource/technical/document/reference_manual/c2/f8/8a/f2/18/e6/43/96/DM00031936.pdf/files/DM00031936.pdf/jcr:content/translations/en.DM00031936.pdf 2019-10-16T02:32:38 < con3> busy looking through this one 2019-10-16T02:33:28 < Cracki> output compare modes 2019-10-16T02:33:45 < Cracki> random bit: OC1M 2019-10-16T02:33:53 < Cracki> (has 3 bits) 2019-10-16T02:34:26 < Cracki> nada, set high/low on match, toggle on match, forced high/low, PWM behavior 2019-10-16T02:34:30 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-16T02:34:51 < con3> Cracki: thank you, checking now 2019-10-16T02:35:09 < Cracki> may not be what you want, but probably close 2019-10-16T02:37:58 < con3> Cracki: yeah it's set to pwm1. 2019-10-16T02:38:06 < con3> something is hiding somewhere 2019-10-16T02:38:14 < con3> i just dont know where 2019-10-16T02:38:38 < Cracki> so you got the period set up, and the compare register set to something between that and 0, yes? 2019-10-16T02:39:19 < con3> yep period is 1200 and the ccr1 is 300 2019-10-16T02:39:53 < con3> the counter increments, hit 300 and in the sr the cc1IF gets set 2019-10-16T02:39:59 < con3> pin remains in the same state though 2019-10-16T02:40:24 < Cracki> upon reload the pin _does_ get set, yes? 2019-10-16T02:40:43 < con3> yeah, when it gets reloaded I can here it get set 2019-10-16T02:40:48 < con3> it just never goes low 2019-10-16T02:41:08 < con3> constant high except when the arr value hits then it does a quick toggle 2019-10-16T02:41:15 < Cracki> does it reach 1200 and then reset, or does it keep running until 2^16? 2019-10-16T02:41:21 < Cracki> hm 2019-10-16T02:41:24 < Cracki> quick toggle? 2019-10-16T02:41:26 < con3> 1200 and then a reset 2019-10-16T02:41:33 < Cracki> like, actual pulse on the pin? 2019-10-16T02:41:39 < con3> pulse on the pin 2019-10-16T02:41:42 < Cracki> !!! 2019-10-16T02:41:44 < Cracki> that's something 2019-10-16T02:41:49 < con3> im so fuckn confused 2019-10-16T02:41:56 < Cracki> I would say it's using the wrong compare register 2019-10-16T02:42:00 < Cracki> one that's still at 0 2019-10-16T02:42:04 < Cracki> or 1 or whatev 2019-10-16T02:42:09 < con3> omg 2019-10-16T02:42:15 < con3> let me check 2019-10-16T02:42:20 < Cracki> dun dun dun 2019-10-16T02:44:08 < con3> nope 2019-10-16T02:44:14 < Cracki> impossibru 2019-10-16T02:44:28 < con3> the thing is the status register does trigger when the counter passes the ccr1 value] 2019-10-16T02:45:32 < Cracki> can you replicate this behavior (or the desired one) on a different time in the same chip? on another chip? 2019-10-16T02:45:37 < Cracki> *timer 2019-10-16T02:45:48 < Cracki> or even just a different channel of the same timer 2019-10-16T02:46:16 < con3> Im trying to do this on three channels, happens on all of them :( 2019-10-16T02:46:37 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-16T02:47:26 < con3> the thing that really gets me is the little toggle it does at the end of the period 2019-10-16T02:47:27 < Cracki> that at least excluded hardware failure 2019-10-16T02:47:44 < Cracki> yeah that's probably what its bits are telling it to do 2019-10-16T02:48:13 < Cracki> is this hal/ll, stiperiph, locm3, ...? 2019-10-16T02:48:18 < con3> hal 2019-10-16T02:48:34 < jadew> I pay for aliexpress standard shipping, I get china post :/ 2019-10-16T02:48:43 < Cracki> maybe it's time to question the magic numbers and their given names 2019-10-16T02:49:10 < Cracki> so... you just want it to generate a standard PWM, right? 2019-10-16T02:49:32 < con3> Cracki: yep,i have no idea why im stuggling this much 2019-10-16T02:50:26 < Cracki> break it in a debugger, check that the bits in the timer's registers are exactly what you expect 2019-10-16T02:50:36 < Cracki> maybe some HAL call got the wrong magic number 2019-10-16T02:50:52 < con3> that's what im doing, so far everything looks fine. recheking rn 2019-10-16T02:51:01 < Cracki> maybe something overwrites or resets the compare register 2019-10-16T02:51:29 < con3> not sure whether this is weird but once trigerred the SR remains high 2019-10-16T02:51:58 < Cracki> I'd bet that CCR is near 0 or near ARR 2019-10-16T02:52:52 < con3> ccr1 is at 300, arr is 1200. 2019-10-16T02:53:42 < Cracki> debugger confirms? what are the CCRMx bits? 2019-10-16T02:53:48 < Cracki> *CCMRx 2019-10-16T02:54:19 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-16T02:54:26 < Cracki> it talks here about enabling preload register and auto-reload preload register 2019-10-16T02:54:50 < con3> yep debugger says the same, 0c1M is 110 2019-10-16T02:55:45 < Cracki> it also talks about shadow registers, which may only get an update when you generate an update event... setting the UG bit in EGR reg 2019-10-16T02:57:06 < con3> Cracki: one sec 2019-10-16T02:57:06 < Cracki> F0 hal docs: https://www.st.com/content/ccc/resource/technical/document/user_manual/2f/77/25/0f/5c/38/48/80/DM00122015.pdf/files/DM00122015.pdf/jcr:content/translations/en.DM00122015.pdf 2019-10-16T02:58:02 < Cracki> hal funcs distinguish between PWM and output compare... 2019-10-16T02:59:45 < con3> Cracki: where did you read about the shadow registers? 2019-10-16T02:59:52 < Cracki> refman 2019-10-16T02:59:59 < con3> checking 2019-10-16T03:00:59 < Cracki> it uses those to update at reload. atmegas do that so the comparison being equal isn't accidentally missed when you set registers to new values that jump over the equality 2019-10-16T03:01:29 < con3> wait is that why its toggling at reload 2019-10-16T03:01:38 < Cracki> unlikely 2019-10-16T03:01:45 < con3> damnit 2019-10-16T03:01:59 < Cracki> I think it's doing that because its shadow register still operates with the default value from system reset 2019-10-16T03:02:05 < Cracki> just guessing tho 2019-10-16T03:02:28 < Cracki> try this: crank up the prescaler, make it slow as hell 2019-10-16T03:02:33 < Cracki> measure the width of that pulse 2019-10-16T03:02:42 < con3> will do 2019-10-16T03:02:45 < Cracki> if it's a single timer count, or two counts, that's suspicious 2019-10-16T03:02:54 < bitmask> mouser or digikey 2019-10-16T03:03:35 < bitmask> If I recall correctly, digikey was more friendly to smaller orders 2019-10-16T03:03:40 < Cracki> ~10 us per count is probably good enough to be resolved well on a cheap LA 2019-10-16T03:03:52 < Cracki> friendly as in free shipping? 2019-10-16T03:04:28 < bitmask> I dont think its free, but I thought they accepted smaller quantity orders, like 1 or 10 and mouser didnt and maybe $5-10 for shipping, I could be wrong as I havent checked in forever 2019-10-16T03:04:30 < Cracki> small as in BGA grains of sand, or small as in reels of cheap Rs? 2019-10-16T03:04:51 < Cracki> me neither 2019-10-16T03:05:01 < Cracki> I don't remember either of them having remarkable MOQ 2019-10-16T03:05:14 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-16T03:05:35 < bitmask> I was gonna go with LCSC but not sure its worth it at this point 2019-10-16T03:11:07 < con3> Cracki: looks like about 10 ticks long 2019-10-16T03:11:23 < Cracki> okay that's very odd 2019-10-16T03:11:53 < Cracki> varies with prescaler? whole timer period in seconds matches calculation? 2019-10-16T03:14:03 -!- qyx [~qyx@gw2.krtko.org] has quit [Ping timeout: 245 seconds] 2019-10-16T03:15:15 < Cracki> I'd see if there are any source examples for pwm on F1, and whether they "just work" 2019-10-16T03:15:30 < karlp> bitmask: different countries have better things for digikey/mouser 2019-10-16T03:16:01 < karlp> mouser is free shipping over 60€ or so for me, whereas digikey is flat rate 30€ until 100€ or something. 2019-10-16T03:16:15 < kakimir64> memmanagehandler invoked 2019-10-16T03:16:20 < bitmask> I see 2019-10-16T03:20:27 < jadew> man... aliexpress are fucking crooks 2019-10-16T03:20:45 < jadew> they don't allow you to read negative feedback, not easily anyway 2019-10-16T03:20:59 < jadew> if you go to a stores feedback and click on the negative one, it shows you ALL feedback 2019-10-16T03:21:07 < jadew> and the sort by: thing, doesn't work 2019-10-16T03:21:19 < jadew> in the HTML, it has the class selector: link-fake-selector 2019-10-16T03:24:34 < Cracki> lol 2019-10-16T03:24:51 < Cracki> time to greasemonkey it 2019-10-16T03:25:11 < jadew> I asked the seller why he shipped via china post and not aliexpress standard shipping like I paid for (it's more expensive) 2019-10-16T03:25:24 < jadew> and he said ali shipping also sends by china post sometimes 2019-10-16T03:25:26 < jadew> ... 2019-10-16T03:25:38 < con3> so refund? 2019-10-16T03:25:54 < jadew> con3, there used to be a dispute section for wrong shipping method, but they removed it 2019-10-16T03:26:02 < con3> ffs 2019-10-16T03:27:14 < jadew> they said next time I should let them know 2019-10-16T03:27:26 < jadew> I said "I thought that paying extra for shipping, was self explanatory" 2019-10-16T03:40:16 -!- fenugrec_ is now known as fenugrec 2019-10-16T03:49:46 < jadew> I must be still sleeping or something... 2019-10-16T03:50:18 < jadew> ebay is saying my balance is -0.00 USD and I have to pay that before editing a listing :/ 2019-10-16T03:53:02 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-16T03:56:34 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-16T03:58:07 < kakimir32> strlen("string") causes Mem Management Handler 2019-10-16T03:59:26 < kakimir32> I suspect that linker script is still broken 2019-10-16T04:05:05 < kakimir32> where are the strings stored :o 2019-10-16T04:05:11 < kakimir32> dem have no name 2019-10-16T04:15:54 < kakimir32> cant see them in map 2019-10-16T04:17:30 < kakimir32> I wonder if there is such test suite for freertos that I just apply the code to my freertos platform and run the test task and it goes through everything that have been enabled 2019-10-16T04:17:53 < kakimir32> and tries to crash something 2019-10-16T04:23:48 -!- con3 [~kvirc@154.119.40.174] has joined ##stm32 2019-10-16T04:24:52 < con3> hmm..maybe im just using a pin that isnt referenced to the pwm alternate function 2019-10-16T04:25:18 < Cracki> there are tables that should state precisely which pins get that AF 2019-10-16T04:25:32 < con3> yeah im going to check quick 2019-10-16T04:25:39 < con3> if its not that I have no idea whats going on 2019-10-16T04:25:50 < Cracki> would be funny tho because you did see action on the pin that correlates to the timer's state 2019-10-16T04:26:51 < con3> i honestly have no idea what wrong. this shit is breaking my mind. it's just a freaking pwm signal. the code looks exactly the same to multiple tutorials, etc ive seen. So i can only think its the pin 2019-10-16T04:26:52 < con3> *shrug 2019-10-16T04:28:35 * con3 puts away all his embedded stuff and looks for mcdonalds jobs 2019-10-16T04:29:35 -!- gnom [~aleksande@178.150.7.153] has quit [Read error: Connection reset by peer] 2019-10-16T04:30:01 < mawk> sloppy wiring 2019-10-16T04:30:16 < con3> mawk: you called? 2019-10-16T04:30:30 < mawk> lol 2019-10-16T04:33:38 < Cracki> I'm betting it's a silly programming mistake, usage of some api or argument or whatever 2019-10-16T04:33:54 < Cracki> srsly, go find an example project for F0 pwm with hal 2019-10-16T04:34:19 < Cracki> cubemx generates almost everything, except a few calls to actually run the timer 2019-10-16T04:36:01 < con3> Cracki: im going to. just a kick in the nuts that i cant get this going 2019-10-16T04:36:25 < Cracki> sleep solves many problems 2019-10-16T04:36:53 < con3> cant sleep though.. really should get some sleeping pills 2019-10-16T04:39:01 < con3> Cracki: how long have you been working on embedded systems? 2019-10-16T04:39:38 -!- fenugrec [~fenugrec@104.221.51.116] has quit [Ping timeout: 240 seconds] 2019-10-16T04:40:03 < Cracki> uh, first contact in 2008 in class, fucked with it for a while then, began doing it for money in 2015 2019-10-16T04:43:43 < con3> how long did it take you till you felt comfortable? Have any mentors? 2019-10-16T04:44:07 < Cracki> valuable traits: playing with things, turning frustration into a puzzle, keeping at it until it works, finding different ways to look at it, breaking it down 2019-10-16T04:44:13 < Cracki> lol mentors 2019-10-16T04:44:26 < Cracki> I wish 2019-10-16T04:44:43 < Cracki> what we have is professors and TAs and so on 2019-10-16T04:45:43 < Cracki> I had a class mate I did the project with. I was good at coding already, he was the studious type, so we did pair programming. I did my best to make the code so obvious that it smacks you in the face 2019-10-16T04:45:48 < con3> Cracki: but you're currently working as an embedded engineer or are you still in academia? 2019-10-16T04:45:55 < Cracki> half and half 2019-10-16T04:46:09 < Cracki> finishing up the master's at a snail's pace 2019-10-16T04:46:46 < con3> im just trying to figure out how you got where you are and how i can make this road easier :/ 2019-10-16T04:46:47 < Cracki> not working _that_ much either, mostly just being fucking depressed and PTSDd from psychopaths and assholes 2019-10-16T04:46:52 * con3 feels stupid af 2019-10-16T04:46:58 < Cracki> love procrastinating? 2019-10-16T04:47:06 < con3> ayyy depression as well 2019-10-16T04:47:10 < Cracki> do it by absorbing useful material 2019-10-16T04:47:27 < Cracki> you know why I hang around here? I see others bang their heads against the wall just like I would 2019-10-16T04:47:45 < Cracki> BUT it's not my ass at stake, so I can stay calm about trouble 2019-10-16T04:47:52 < con3> Cracki: hate procrastination, live learning and work for hours. but i just feel like i keep freaking not progressing fast enough 2019-10-16T04:47:53 < Cracki> emotions kill thought 2019-10-16T04:48:23 < doomba> con3: that's life 2019-10-16T04:48:32 < con3> yeah i need to pop in here more and join the chats. most people around me are either using 8 bit atmels or arduinos 2019-10-16T04:48:42 < Cracki> the history of the combine: 2019-10-16T04:48:55 < Cracki> eh nvm 2019-10-16T04:49:03 < jadew> con3, that's great if you have people near you doing electronics 2019-10-16T04:49:15 < doomba> also tutorials... for stm32. 2019-10-16T04:49:18 < doomba> nope 2019-10-16T04:49:32 < Cracki> think of it like this: you absorb an area either by racing over it in thin stripes, or by crawling over it with broad sweeps 2019-10-16T04:49:36 < doomba> read the real documentation 2019-10-16T04:49:52 < Cracki> informational gluttony helps 2019-10-16T04:49:59 < con3> jadew: yeah but its not quite the same. they just pop in some arduino library and run lora.send() and boom done. I want to understand this shit 2019-10-16T04:50:05 < Cracki> if you don't have 50+ tabs open, you are failing at it 2019-10-16T04:50:11 < Cracki> skim everything 2019-10-16T04:50:19 < doomba> tutorials are stack overflow copypasta for blog cred 2019-10-16T04:50:26 < Cracki> those libs you can dig into, see how it's done 2019-10-16T04:50:30 < con3> Cracki: wait i should be skimming? 2019-10-16T04:50:33 < jadew> con3, ah, arduweenies are useless 2019-10-16T04:50:45 < Cracki> and then you see that they're only cooking with piss 2019-10-16T04:50:52 < jadew> but old school electronics guys are not 2019-10-16T04:51:06 < jadew> you need to find friends like that 2019-10-16T04:51:19 < con3> god i would pay with a kidney if i could have an old school embedded person teach me. 2019-10-16T04:51:23 < con3> id give my liver too 2019-10-16T04:51:28 < Cracki> broad view helps. you know where you are, and what is around you, and what might be further away 2019-10-16T04:51:36 < Cracki> renaissance man 2019-10-16T04:51:47 < Cracki> wat is old school 2019-10-16T04:52:07 < jadew> before make:r shit 2019-10-16T04:52:11 < Cracki> hooking up uv erasable proms to hand cranked 6502? 2019-10-16T04:52:42 < Cracki> all that isn't magic, it's just lost craft 2019-10-16T04:53:19 < con3> i just feel like i need guidance from an "old scool 2019-10-16T04:53:21 < Cracki> every single one of these people who dazzle you with having done that in middle school... either had a mentor/teacher/dad or a decent book that explains it in small enough steps that it's basically a recipe 2019-10-16T04:53:25 < jadew> old schoolers will have a better grasp of the analog side 2019-10-16T04:53:41 < con3> " guy or someone with experience to just get slaps in the right direction 2019-10-16T04:54:01 < Cracki> if you have the iq points, you can put them to use with the right tools 2019-10-16T04:54:04 < doomba> by hand cranked do you mean they wire wrapped that shit from old-style to-72 transistors? 2019-10-16T04:54:17 < Cracki> your eyes are the highest bandwidth input you have, so any tool that *shows* you things is valuable 2019-10-16T04:54:48 < con3> like honestly is it just me that feels the more i dig the more i feel there is to learn. like being competent or comfortable is a cruel joke 2019-10-16T04:54:55 < Cracki> any kind of scope, anything that shows pictures or moving pictures, anything that produces obscene amounts of data you can sift through 2019-10-16T04:55:08 < jadew> con3, if you really want to understand stuff, do bare metal with no libraries 2019-10-16T04:55:12 < Cracki> with a quadrature encoder and a LA you can record sound :> 2019-10-16T04:55:27 < Cracki> at least the sound the motor shaft makes 2019-10-16T04:55:43 < jadew> con3, I feel the same way 2019-10-16T04:56:04 < con3> jadew: i did start doing it at a point, but i still felt like i was doing it in an odd way 2019-10-16T04:56:17 < Cracki> if your knowledge is a circle, its area increases by the square while its perimeter increases linearly... so it's not that bad :> 2019-10-16T04:56:19 < con3> jadew: good to know im not the only one 2019-10-16T04:56:25 < jadew> con3, that may be the case, but you at least get a feel of what's going on 2019-10-16T04:57:19 < Cracki> a little bare bones helps understanding for sure... but that's all you get on atmegas :> atmel hadn't dared piss their base off with anything HALish 2019-10-16T04:57:43 < con3> wait is atmel just baremetal? 2019-10-16T04:57:52 < Cracki> no, comparable to stdperiph or ll 2019-10-16T04:58:14 < jadew> and that holds true only for the AVRs, they have ARMs too 2019-10-16T04:58:23 < Cracki> same with stm32, all the peripheral registers are mapped to ram, and you get their names 2019-10-16T04:58:29 < con3> stdperiph was prior to my time :p cant remember why i had issues with ll 2019-10-16T04:58:46 < Cracki> and you can do someting like DDRB |= (1<< DDRB5) to set B5 direction to output 2019-10-16T04:58:54 < Cracki> stm32 bare metal is similar 2019-10-16T04:59:13 < con3> sweet god, bitwise operations. 2019-10-16T04:59:14 < jadew> I've only done stm32 bare metal, with no lib 2019-10-16T04:59:20 < Cracki> with some smarter ways that let you avoid read-modify-write... by simply writing to it and it does the ORing in hardware 2019-10-16T04:59:30 < con3> yeah i thinkk i used cmsis and the ref manual when i tried bare metal 2019-10-16T04:59:31 < jadew> well, I wrote the libs myself, but I guess that qualifies 2019-10-16T04:59:48 < Cracki> if you don't have a lib, you go like #define DDRB (*(uint8_t*)(0x1234)) 2019-10-16T04:59:53 < Cracki> then it's like a variable 2019-10-16T04:59:58 < con3> the one thing i hated about doing things without cube was the clock tree 2019-10-16T05:00:06 < Cracki> yeah clock tree! 2019-10-16T05:00:13 < Cracki> atmegas don't have that, trivial 2019-10-16T05:00:39 < con3> Cracki: sweet god, they setup the clock baremetalish as well? 2019-10-16T05:00:59 < Cracki> heh, I chatted to the (undergrad) TAs at the hardware lab course yesterday. the biggest problem in class is buttons bouncing. 2019-10-16T05:01:21 < Cracki> and hardly anybody seems to understand what the fuck is going on because NOBODY dares to ask to touch the scopes they have there 2019-10-16T05:01:29 < jadew> lol 2019-10-16T05:01:29 < Cracki> atmega has no clock tree 2019-10-16T05:01:42 < Cracki> you get "the clock" that's the crystal, on everything 2019-10-16T05:01:53 < con3> Cracki: dear god 2019-10-16T05:01:55 < Cracki> everything clocky has a prescaler 2019-10-16T05:01:56 < con3> no pll's? 2019-10-16T05:01:59 < Cracki> timers, uart, i2c, etc 2019-10-16T05:02:03 < Cracki> nah pll is fancy 2019-10-16T05:02:09 < jadew> they have a fuse to divide it by 8 2019-10-16T05:02:13 < Cracki> you find a 20 MHz crystal and hook it up 2019-10-16T05:02:15 < Cracki> ah yes 2019-10-16T05:02:19 -!- Streake_ [~Streaker@unaffiliated/streaker] has joined ##stm32 2019-10-16T05:02:26 < con3> i cant tell if you guys are messing with me 2019-10-16T05:02:40 < Cracki> srsly, look at an arduino nano 2019-10-16T05:02:45 < Cracki> 20 mhz crystal on it 2019-10-16T05:02:52 < con3> ok i believe you 2019-10-16T05:02:52 < jadew> we're not, but that's part of what makes them appealing 2019-10-16T05:02:58 < con3> just been trolled before :p 2019-10-16T05:03:03 < jadew> didn't you just say your biggest problem is the clock tree? 2019-10-16T05:03:05 < Cracki> they come with internal... 8 mhz or something but that's off by percents not ppms 2019-10-16T05:03:07 < jadew> AVR = problem solved 2019-10-16T05:03:14 < con3> jadew: i can imagine, the clock tree is absolute agony 2019-10-16T05:03:33 < jadew> Cracki, yeah, 8 MHz RC oscillator 2019-10-16T05:03:37 < jadew> but it can be tuned 2019-10-16T05:03:59 < con3> jadew: while not my biggest problem in that sense, just its crappy setting it up bare metal 2019-10-16T05:04:37 < Cracki> https://vivonomicon.com/category/stm32_baremetal_examples/ 2019-10-16T05:04:45 < Cracki> skim ;) it 2019-10-16T05:05:09 < Cracki> absorb the general feel, maybe some details, know that the information exists 2019-10-16T05:05:32 < Cracki> https://vivonomicon.com/2018/04/02/bare-metal-stm32-programming-part-1-hello-arm/ 2019-10-16T05:05:37 < Cracki> I think that's the entry point 2019-10-16T05:05:42 < jadew> I don't even look for examples when I start doing something 2019-10-16T05:05:48 -!- Streaker [~Streaker@unaffiliated/streaker] has quit [Ping timeout: 265 seconds] 2019-10-16T05:05:48 < jadew> I found that most of them are wrong 2019-10-16T05:05:49 < Cracki> looky they use f031 2019-10-16T05:06:14 < con3> Cracki: thank you, really 2019-10-16T05:06:24 < con3> i used the stm32 snippets as examples for bare metal 2019-10-16T05:06:28 < jadew> for me, reading the manual seems like the best approach 2019-10-16T05:06:30 < con3> this site will help a lot 2019-10-16T05:06:31 < Cracki> forums are bad, blogs less bad, books better 2019-10-16T05:06:37 < jadew> you get the most complete information out of there 2019-10-16T05:06:41 < Cracki> proprtional to how much effort someone needs to expend 2019-10-16T05:06:45 < con3> there's just so many books out there 2019-10-16T05:06:48 < jadew> and you can decide on weather enabling something or not really makes sense 2019-10-16T05:06:54 < Cracki> fuck all the books 2019-10-16T05:07:21 < Cracki> find easily digestible stuff, augment with official docs 2019-10-16T05:07:31 < Cracki> I tried looking for a book on WPF with C# 2019-10-16T05:07:33 < Cracki> forget it 2019-10-16T05:07:47 < con3> ok yeah i think i need to change my approach to learning 2019-10-16T05:07:48 < jadew> Cracki, MS has lots of videos on that 2019-10-16T05:07:55 < Cracki> now I consume video tutorials off youtube. I don't have to read, I can listen 2019-10-16T05:08:00 < con3> that might be whats keeping me back 2019-10-16T05:08:10 < Cracki> if I want facts, I reach for the official docs 2019-10-16T05:08:22 < Cracki> if I want "a mentor guiding me" yt vids are the next best thing 2019-10-16T05:08:34 < Cracki> books mostly exist because someone wants to have written one 2019-10-16T05:08:48 < Cracki> every tech publishing house has 1 or more books on the same topic 2019-10-16T05:08:49 < jadew> except you won't find anyone on YT spilling out the really good information 2019-10-16T05:09:05 < Cracki> yeh yt is entertainment, a gateway drug 2019-10-16T05:09:33 < con3> i havent found a lot of people with good embedded stuff on yt 2019-10-16T05:09:53 < jadew> probably because it's so overcrowded 2019-10-16T05:10:05 < Cracki> mathworks (matlab) do it too. tasty little videos showcasing their stuff. if you want more, you can pay for a real life course in a city near you, or more online material that may be free but no handholding 2019-10-16T05:10:15 < jadew> it's not worth to even bother doing something interesting 2019-10-16T05:10:31 < con3> you guys see the siraj controversy? 2019-10-16T05:10:32 < jadew> if you do, you'll get copied by 1000 shit shoveling channels 2019-10-16T05:10:37 < Cracki> the wat 2019-10-16T05:11:01 < Cracki> "thinkfluencer" "plagiarized" 2019-10-16T05:11:07 < Cracki> show bob and vagene? 2019-10-16T05:11:21 < con3> plagiarized guy 2019-10-16T05:11:29 < con3> fucked up bad 2019-10-16T05:11:33 < Cracki> yeah don't pay the scammer 2019-10-16T05:12:01 < Cracki> anything with block chain, deep learning, internet of shit, ... in the name is disease ridden 2019-10-16T05:12:44 < jadew> most channels are made by people who are actually shoveling content for money 2019-10-16T05:12:44 < con3> i always feel like there's a lot more resources and well put down shit out there for most things, but embedded is a hard find 2019-10-16T05:12:48 < jadew> they don't care about the content 2019-10-16T05:13:04 < jadew> that great scott channel is a very prominent example 2019-10-16T05:13:06 < Cracki> that barebutt stm32 blog shows you a linker script and explains it... that's exactly what you wanna know (about) to have a proper model of reality 2019-10-16T05:13:26 < con3> Cracki: yeah i wish i found this site earlier 2019-10-16T05:13:53 < con3> if you have any other resources you could recommend please pop me a link or name. would really appreciate it 2019-10-16T05:15:56 < con3> and its 4am 2019-10-16T05:16:59 < Cracki> uh... I watch the macbook repair guy. he sometimes has to explain a few basics such as what a buck converter does, or that there's a 1wire bus in the charge port, and how these things have power rails sequenced in order, etc 2019-10-16T05:17:36 < jadew> louis? 2019-10-16T05:17:41 < Cracki> shows me enough that I know what to do and what to ask for if someone wants me to do the same, not that I'd have any practice :> 2019-10-16T05:17:45 < Cracki> the rossmann 2019-10-16T05:18:07 < jadew> he doesn't know much about this stuff tho 2019-10-16T05:18:18 < Cracki> probably 2019-10-16T05:18:21 < jadew> he only knows what to do to repair them, and some of the basics 2019-10-16T05:18:25 < jadew> but he's not like us 2019-10-16T05:18:33 < Cracki> half his work is "see corrosion, replace part" 2019-10-16T05:18:46 < con3> anyone have any book recommendations on computer architecture? 2019-10-16T05:18:51 < Cracki> he has no need to program anything, of course 2019-10-16T05:18:56 < jadew> he doesn't make stuff 2019-10-16T05:18:58 < Cracki> knuth maybe 2019-10-16T05:19:02 < jadew> yeah 2019-10-16T05:19:17 < Cracki> he's european, so his computer science comes from math, not from electrical engineering 2019-10-16T05:19:29 < Cracki> but he's rigorous, no hand waving 2019-10-16T05:19:31 < con3> whose this Cracki? 2019-10-16T05:19:49 < Cracki> knuth, wrote some thicc tomes 2019-10-16T05:19:58 * con3 googles 2019-10-16T05:20:33 < Cracki> if you want another rabbit hole to hell, "structure and interpretation of computer programs"... it's about as practical as learning latin 2019-10-16T05:21:01 < Cracki> will teach you nothing about computers, but a bit about the math underlying 2019-10-16T05:21:13 < con3> im just looking for something that goes indepth and explains well 2019-10-16T05:21:33 < con3> our comp arch course was a joke 2019-10-16T05:21:41 < jadew> ours too 2019-10-16T05:22:02 < Cracki> "There are only two hard things in Computer Science: cache invalidation and naming things." 2019-10-16T05:22:11 < Cracki> so look up cache invalidation if you want to suffer 2019-10-16T05:22:26 < con3> ill leave naming things for just before i retire 2019-10-16T05:22:47 < Cracki> https://en.wikipedia.org/wiki/MESI_protocol 2019-10-16T05:22:48 < con3> think the clean code book went off on it 2019-10-16T05:22:55 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-16T05:23:28 < con3> Cracki: what was your undergrad major? 2019-10-16T05:23:32 < Cracki> cs 2019-10-16T05:23:37 < Cracki> same as the master's 2019-10-16T05:23:40 < con3> ah 2019-10-16T05:23:47 < con3> fuck i wish i did cs 2019-10-16T05:23:56 < jadew> why? 2019-10-16T05:24:03 < Cracki> I wish I had taken something closer to EE 2019-10-16T05:24:17 < con3> well maybe compe tbh 2019-10-16T05:24:31 < Cracki> this place is worth it but dry 2019-10-16T05:25:01 < con3> i did power systems, so yaaaay 2019-10-16T05:27:37 < Cracki> naming things: http://photos1.blogger.com/blogger2/1715/1669/1600/larson-oct-1987.gif 2019-10-16T05:28:02 < Cracki> (article for it https://steve-yegge.blogspot.com/2006/10/egomania-itself.html?m=1 ) 2019-10-16T05:28:11 < Cracki> power systems are nice 2019-10-16T05:28:24 < Cracki> you can do things that are dangerous :P 2019-10-16T05:28:31 < con3> yeah i wish i rather went the embedded route 2019-10-16T05:29:36 < Cracki> I'd tell freshmen to go into VLSI, chip design 2019-10-16T05:30:07 < Cracki> not sure the demand will increase that much but it's a hard field at the edge of interdisciplinary research 2019-10-16T05:30:25 < Cracki> that or flipping burgers, burgers never go out of style 2019-10-16T05:30:26 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 276 seconds] 2019-10-16T05:31:20 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-16T05:31:40 < jadew> I tell people to go into business 2019-10-16T05:32:09 < jadew> no matter the field, there will be businesses 2019-10-16T05:32:25 < jadew> and you don't have to understand everything and you don't have to do the ground work 2019-10-16T05:32:43 < jadew> and the pay is better 2019-10-16T05:32:49 < Cracki> "do business" is like "do breathing exercises". every type of sport requires good breathing 2019-10-16T05:33:06 < Cracki> anthing that isn't a business isn't keeping you afloat 2019-10-16T05:33:17 -!- Streaker [~Streaker@unaffiliated/streaker] has joined ##stm32 2019-10-16T05:33:33 < jadew> someone with a business degree will almost always earn more than the engineers working at the same company 2019-10-16T05:33:41 < Cracki> that's changing 2019-10-16T05:34:00 < jadew> nah, it used to be different for a brief moment, with software engineering 2019-10-16T05:34:02 < jadew> but not anymore 2019-10-16T05:34:07 < jadew> everyone is a software engineer now 2019-10-16T05:34:07 < con3> yeah i just wouldnt enjoy doing it 2019-10-16T05:34:11 < jadew> and everyone does electronics 2019-10-16T05:34:21 < jadew> con3, let me tell you something about money 2019-10-16T05:34:27 < jadew> money is great 2019-10-16T05:34:42 < jadew> if you have money, you can do whatever you enjoy doing 2019-10-16T05:35:10 < con3> jadew: i agree,but god doing an 8-5 minimum with something you hate is agonizing 2019-10-16T05:35:25 < jadew> if you only prepare for one job, then you're stuck with it, even if you like it or not 2019-10-16T05:35:43 < jadew> con3, that's true, but you misunderstand something 2019-10-16T05:35:43 < Cracki> transferable skills 2019-10-16T05:35:46 -!- Streake_ [~Streaker@unaffiliated/streaker] has quit [Ping timeout: 265 seconds] 2019-10-16T05:35:51 < Cracki> the biggest skill is your brain 2019-10-16T05:35:56 < jadew> people with business degrees don't work the same as engineering degrees 2019-10-16T05:36:01 < Cracki> but also the slowest to apply 2019-10-16T05:36:12 < con3> jadew: how so? 2019-10-16T05:36:26 < Cracki> a business degree is an accountant on steroids 2019-10-16T05:36:38 < jadew> con3, because the engineers have to work 9 to 5 to make something difficult happen 2019-10-16T05:36:50 < Cracki> I get to audit some econ lectures. they're all about rational decision making support, so it's all numbers and math 2019-10-16T05:36:53 < jadew> business degrees go out to launch with customers and give them a good time 2019-10-16T05:37:01 < jadew> and then they tell you what to do 2019-10-16T05:37:04 < Cracki> but they need nothing more than high school arithmetic and plotting functions 2019-10-16T05:37:16 < Cracki> the rest is *knowing* things 2019-10-16T05:37:38 < Cracki> you have marketing and sales and customer relations 2019-10-16T05:39:14 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has left ##stm32 [] 2019-10-16T05:39:23 < jadew> Cracki, all of that is easier than engineering 2019-10-16T05:39:38 < Cracki> to normies 2019-10-16T05:39:45 < jadew> and the first two would also earn you more 2019-10-16T05:39:46 < Cracki> it's all normies can do 2019-10-16T05:39:56 < Cracki> maybe, maybe not 2019-10-16T05:40:03 < jadew> 100% 2019-10-16T05:40:21 < jadew> business owners see engineering as an expense 2019-10-16T05:40:28 < jadew> while sales, as something that brings money in 2019-10-16T05:40:33 < Cracki> an engineer can have skills to leverage too 2019-10-16T05:40:33 < jadew> (marketing too) 2019-10-16T05:40:43 < jadew> which is why they are likely to get paid more 2019-10-16T05:40:45 < Cracki> it's just that engineers are seldom the type to press their advantage 2019-10-16T05:41:10 < Cracki> if you have the taste for bargaining, you do that for a living 2019-10-16T05:41:37 < jadew> my point is that there are better things to do than engineering 2019-10-16T05:41:50 < Cracki> if a business owner sees engineering as an expense, they're selling hot air 2019-10-16T05:41:52 < jadew> and more fulfilling in the long run 2019-10-16T05:42:01 < jadew> Cracki, they all see it like that 2019-10-16T05:42:04 < jadew> no exception 2019-10-16T05:42:32 -!- Streake_ [~Streaker@unaffiliated/streaker] has joined ##stm32 2019-10-16T05:42:33 < jadew> even you, if you started a business, you'd know that you have to pay those engineers some money in order to make you YOUR product 2019-10-16T05:42:46 < jadew> so they're an expense to get YOUR product done 2019-10-16T05:42:53 < jadew> it's like ordering a custom t-shirt 2019-10-16T05:42:56 < jadew> it's an expense 2019-10-16T05:43:10 -!- con3 [~kvirc@154.119.40.174] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-16T05:43:12 < jadew> then you have to sell it and you can't do it by yourself, so you need aid in bringing in money 2019-10-16T05:43:16 < Cracki> ... I was wondering where I was shedding all this time... it's all in the keyboard! 2019-10-16T05:43:16 < jadew> that's the sales department 2019-10-16T05:43:19 < jadew> those are your friends 2019-10-16T05:43:23 < jadew> the other ones are your tools 2019-10-16T05:43:36 < jadew> Cracki, haha, yeah 2019-10-16T05:43:39 < Cracki> engineers are an investment. 2019-10-16T05:43:54 < Cracki> money in, value out 2019-10-16T05:43:56 < jadew> sure, but still... a cost center 2019-10-16T05:44:14 < jadew> I'm not saying that's what it should be, just that that's how businesses view things 2019-10-16T05:44:19 < jadew> and it makes sense 2019-10-16T05:44:31 < jadew> it's just unfortunate for engineers 2019-10-16T05:44:57 -!- Streaker [~Streaker@unaffiliated/streaker] has quit [Ping timeout: 265 seconds] 2019-10-16T05:45:20 < jadew> however, the hard reality is that there are lots of engineers available now and you can contract them from anywhere in the world 2019-10-16T05:45:37 < jadew> and a product without the rest of the business around it, is just something nobody buys 2019-10-16T05:46:27 < jadew> and once the whole business is set up, sales and marketing comes ahead, simply because they're perceived as the people who bring money in 2019-10-16T05:46:40 < jadew> despite the fact that buyers may buy that product because engineering did a good job 2019-10-16T05:46:42 < Cracki> indeed any time a shortage is announced, it's only to drive down wages 2019-10-16T05:47:04 < Cracki> i.e. to increase supply 2019-10-16T05:47:04 < jadew> the boss will still consider that it was marketing who brought that detail up and made the buyers aware of it 2019-10-16T05:52:20 < Cracki> one key delinted, 100 others still fluffy and cushioned 2019-10-16T05:53:51 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-16T05:54:11 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-16T05:55:27 < Cracki> https://en.wikipedia.org/wiki/MMIX an implementation of that thing might help you follow exactly what happens in an MCU, also when interrupts happen etc 2019-10-16T05:55:44 < Cracki> same as any assembly, it has the notion of "first program location" 2019-10-16T05:56:20 < Cracki> because ahead of that is other stuff that's special, not normal execution 2019-10-16T06:24:41 -!- Streake_ [~Streaker@unaffiliated/streaker] has quit [Remote host closed the connection] 2019-10-16T06:27:58 -!- fc5dc9d4 [~quassel@p5B0810BE.dip0.t-ipconnect.de] has joined ##stm32 2019-10-16T06:31:39 -!- fc5dc9d4_ [~quassel@p5B081211.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-16T07:29:29 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has joined ##stm32 2019-10-16T07:34:21 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-16T07:46:19 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-16T07:49:14 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-16T07:49:15 -!- day__ is now known as day 2019-10-16T08:10:02 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-16T08:21:25 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-16T09:01:59 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-16T09:09:39 -!- fc5dc9d4 [~quassel@p5B0810BE.dip0.t-ipconnect.de] has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.] 2019-10-16T09:10:53 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-16T09:12:37 -!- fc5dc9d4 [~quassel@p5B0810BE.dip0.t-ipconnect.de] has joined ##stm32 2019-10-16T09:17:06 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-16T09:28:53 -!- qyx [~qyx@84.245.120.156] has joined ##stm32 2019-10-16T09:30:32 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-16T09:33:44 -!- qyx_ [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-16T09:34:52 -!- qyx [~qyx@84.245.120.156] has quit [Ping timeout: 264 seconds] 2019-10-16T09:59:43 -!- Test12 [50bb6e01@tmo-110-1.customers.d1-online.com] has joined ##stm32 2019-10-16T10:01:51 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-16T10:17:22 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-16T10:44:47 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-16T10:57:51 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Quit: quit has dan2wik!] 2019-10-16T10:58:27 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-16T10:58:27 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-16T10:58:28 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-16T11:26:39 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-16T11:45:01 -!- talsit [foobar@gromit.mixdown.ca] has left ##stm32 [] 2019-10-16T12:00:47 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-16T12:08:47 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away] 2019-10-16T12:10:55 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-16T12:28:18 -!- jly [uid355225@gateway/web/irccloud.com/x-hkufssfawyixhulv] has joined ##stm32 2019-10-16T12:41:15 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-16T12:48:19 -!- Test12 [50bb6e01@tmo-110-1.customers.d1-online.com] has quit [Ping timeout: 260 seconds] 2019-10-16T13:00:03 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-16T13:08:02 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 276 seconds] 2019-10-16T13:12:46 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-16T13:20:37 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-16T13:28:05 -!- ekaologik [~quassel@p4FF16AEE.dip0.t-ipconnect.de] has joined ##stm32 2019-10-16T14:15:19 -!- qyx_ is now known as qyx 2019-10-16T14:32:12 < kakimir64> back to codez 2019-10-16T14:40:09 < kakimir64> yesterday we left at strlen("string") 2019-10-16T14:43:06 < kakimir64> at that point vPortResetPrivilege was invoked idk. why and immediatelly MemManage_Handler 2019-10-16T14:50:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-16T14:51:03 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-16T15:05:24 < kakimir64> how do I locate "string" in map file? 2019-10-16T15:17:33 < kakimir64> how do I get into breakpoint from halted state from another breakpoint? 2019-10-16T15:18:31 < kakimir64> I mean I set breakpoint to my task an plan to figure out what is going on 2019-10-16T15:20:16 < kakimir64> it's halted there and 3seconds later it has jumped to "hardstop on malloc failure" trap in sbrk 2019-10-16T15:30:26 < kakimir64> okay it has something to do with uart rom driver 2019-10-16T15:30:27 < karlp> heh, sw chip select via gpio is faster than hwardware chipselect via enable/disable spi, at least with the current locm3 apis. 2019-10-16T15:30:46 < rajkosto> isnt hardware chipselect only useful when doing spi slave ? 2019-10-16T15:30:59 < karlp> well, by all accounts, critically important there, 2019-10-16T15:31:12 < karlp> but you can have the stm32 drive the line too, 2019-10-16T15:31:31 < rajkosto> it doesnt even work on f103 when master 2019-10-16T15:31:42 < rajkosto> probably fixed in spi2 2019-10-16T15:31:48 < karlp> https://zerobin.net/?e70a0eb131aa3778#6tqHOfmToBpLPPoJUFDVr9Yik8N38ILIWsaL5kbUPos= 2019-10-16T15:32:01 < karlp> well, i'm on spi 1 on l1, and yes, it does work 2019-10-16T15:32:24 < rajkosto> you have to do spi_disable ? isnt there a bit to set inside the spi controller to set cs 2019-10-16T15:32:28 < rajkosto> thats what i tried and thats what didnt work 2019-10-16T15:32:37 < karlp> well then you didn't read the ref man properly :) 2019-10-16T15:35:29 < karlp> I think I need to get thread profiling working before I bother doing the thredaing work queues I was looking at.... 2019-10-16T15:37:28 -!- talsit [foobar@gromit.mixdown.ca] has joined ##stm32 2019-10-16T15:49:13 < kakimir64> I might need to use portSAVE_CONTEXT(); in my handler that calls ROM based handler 2019-10-16T15:53:52 < kakimir64> I see why I had tried to make that asm mess to store registers 2019-10-16T15:54:30 < kakimir64> there should be facilities in freertos somewhere to do the exact thing - better than my crappy codez 2019-10-16T15:57:02 < kakimir64> yes I don't need to anymore but I would like to know how I find "string" 2019-10-16T16:05:21 < kakimir64> yes but which one is my "string" 2019-10-16T16:05:44 < kakimir64> let's do hard things that are marginal importance! 2019-10-16T16:12:29 < kakimir64> now I'm thinking what kind of wrappings my interrupt requires 2019-10-16T16:14:06 < kakimir64> https://paste.ee/p/dWOyd I think also that line 16 might not work as I intended 2019-10-16T16:15:01 < kakimir64> 0x0000000000002f70 __vectors_start__ = ABSOLUTE (.) 0x0000000000004000 . = (__privileged_functions_start__ + _Privileged_Functions_Region_Size) 2019-10-16T16:15:15 < kakimir64> vectors start from 0x0 2019-10-16T16:16:23 < kakimir64> les fix this 2019-10-16T16:16:49 < kakimir64> I move line 16 to line 7 2019-10-16T16:17:30 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-16T16:19:29 < kakimir64> I have code red startup code and mcuxpresso linker templates playing that for me 2019-10-16T16:20:52 < kakimir64> it's eggz-m3 2019-10-16T16:22:07 < kakimir64> https://paste.ee/p/Aw9hb ahh much better now 2019-10-16T16:22:42 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-16T16:23:35 < kakimir64> i doubt __vectors_start__ do anything important but better have it right 2019-10-16T16:28:38 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-16T16:38:21 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] 2019-10-16T16:42:58 < kakimir64> portEND_SWITCHING_ISR(xHigherPriorityTaskWoken); I need this to my handler rigth? 2019-10-16T16:45:09 < mawk> you're doing something shady kakimir64 2019-10-16T16:45:18 < kakimir64> always 2019-10-16T16:45:41 < kakimir64> I forgot if you aere pro@freertos? 2019-10-16T16:45:58 < mawk> me ? not really 2019-10-16T16:47:14 < mawk> but all the "port" stuff is depending on your MCU, it's not exactly freertods 2019-10-16T16:47:15 < mawk> -d 2019-10-16T16:47:25 < mawk> so you may have luck in your mcu's RM 2019-10-16T16:48:02 -!- jly [uid355225@gateway/web/irccloud.com/x-hkufssfawyixhulv] has quit [Quit: Connection closed for inactivity] 2019-10-16T16:50:30 < kakimir64> no man 2019-10-16T16:50:34 < kakimir64> no manuals man 2019-10-16T16:51:12 < kakimir64> port are system port not the exact chip vendor model stuff 2019-10-16T16:52:53 < kakimir64> in this case M3 + MPU 2019-10-16T16:55:37 < mawk> a* 2019-10-16T16:55:44 < mawk> then it's in the M3 manual 2019-10-16T17:12:55 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-16T17:13:35 < kakimir64> (*rollseyes*) 2019-10-16T17:14:32 < kakimir64> generated linker script doesn't do everything 2019-10-16T17:16:47 < kakimir64> it has limited options and cannot bend to every usecase 2019-10-16T17:18:16 < con3> is the alternate function pin mapping for the timer oc mode in the rm? I cant see it. not sure if im missing it 2019-10-16T17:20:21 < qyx> in the ds 2019-10-16T17:20:36 < qyx> right after the pinout diagrams and table 2019-10-16T17:23:16 < kakimir64> I have multiple callbacks from rom based interrupt handler 2019-10-16T17:23:30 < kakimir64> each gives semaphore 2019-10-16T17:24:23 < kakimir64> I assume not only one callback can be performed in one handler call 2019-10-16T17:26:03 < kakimir64> how do I know if higherprioritytaskwoken? 2019-10-16T17:26:49 < kakimir64> do I pass the same pointer to flag to all of givesemaphorefromISRs? 2019-10-16T17:28:02 < kakimir64> are fromISR procedures made so that it can only set but not clear flag? 2019-10-16T17:29:19 < kakimir64> bbl> 2019-10-16T17:50:26 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-16T17:50:41 < bitmask> ugh, i gotta finalize my parts list and order this shit 2019-10-16T17:50:45 < bitmask> still stuck on a few components 2019-10-16T17:53:55 < mawk> what does the autogenerated linker script don't do kakimir64 ? 2019-10-16T17:55:10 < con3> qyx: thankk you! checking now 2019-10-16T18:00:12 < mawk> you know the company ASML Steffanx ? 2019-10-16T18:00:21 < mawk> apparently it's a famous dutch semiconducteur supplier 2019-10-16T18:02:18 -!- fenugrec_ [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-16T18:03:24 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-16T18:04:33 < bitmask> hmmm I was designing this smps for 1A but I really only need 500mA 2019-10-16T18:04:35 < bitmask> maybe I should choose new components 2019-10-16T18:04:47 < mawk> planning for a bit of slack is good 2019-10-16T18:04:54 < bitmask> double though? 2019-10-16T18:05:01 < bitmask> even 500 is too much probably 2019-10-16T18:05:26 < bitmask> ehh maybe i'll keep it 2019-10-16T18:06:33 < bitmask> :) 2019-10-16T18:09:05 < bitmask> i just added it up and its 260mA 2019-10-16T18:10:03 < bitmask> how do you choose an input filter capacitor for a smps 2019-10-16T18:10:10 < bitmask> this datasheet is no help 2019-10-16T18:10:18 < mawk> big 2019-10-16T18:10:36 < mawk> and failing in the right mode 2019-10-16T18:11:06 < bitmask> for the 12v to 5v 1A its 220uF/25V, not sure for this 12V to 5v 3A at a higher frequency 2019-10-16T18:11:29 < bitmask> the 1A was the lm2595, for the 3A I'm using some cheap MP1593 2019-10-16T18:12:01 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-16T18:12:18 < bitmask> yea I was thinking of doing that but this one is a higher frequency and seems to use smaller components, or its just chinese recommendations 2019-10-16T18:12:41 < bitmask> 385k 2019-10-16T18:13:00 < bitmask> the 2595 is 150k 2019-10-16T18:13:05 < bitmask> not sure if the 6 is the same 2019-10-16T18:13:26 < bitmask> ? no 2019-10-16T18:13:52 < bitmask> 150k was the 2595 and the mp1593 is 385k 2019-10-16T18:14:33 < bitmask> I'm using the 2595 for the mcu and the mp1593 as a phone charger 2019-10-16T18:16:05 < bitmask> i'm using the 5V one 2019-10-16T18:16:59 < bitmask> what does that mean hahaz 2019-10-16T18:17:21 < bitmask> yea what do you mean you had to vomit it 2019-10-16T18:17:34 < bitmask> i see 2019-10-16T18:17:47 < bitmask> gotcha 2019-10-16T18:18:16 < bitmask> :P 2019-10-16T18:29:21 < Ecco> Heya :) 2019-10-16T18:29:32 < Ecco> I'm a bit puzzled regarding this USB2 ESD protection 2019-10-16T18:29:33 < Ecco> https://www.st.com/resource/en/datasheet/usblc6-2.pdf 2019-10-16T18:30:00 < Ecco> For D+ and D-, it's easy : the IC has an input and an output 2019-10-16T18:30:02 < Ecco> so far, so good 2019-10-16T18:30:08 < Ecco> but what about vbus and gnd? 2019-10-16T18:30:27 < Ecco> what if an ESD happens on the VBUS pin, and I want to connect that pin to my MCU? 2019-10-16T18:33:54 < karlp> what do you mean? it says right ther eit protects vbus as well. 2019-10-16T18:34:24 < Ecco> yeah 2019-10-16T18:34:25 < karlp> d+/d- aren't inputs and outputs on that part 2019-10-16T18:34:29 < Ecco> But I wonder how that works 2019-10-16T18:34:29 < karlp> it's just for layout convenience 2019-10-16T18:34:34 < karlp> look at the diagram of what's inside 2019-10-16T18:34:35 < Ecco> huh 2019-10-16T18:34:40 < Ecco> ok, /me is clueless 2019-10-16T18:34:57 < Ecco> Acutally, here's my real question : how am I supposed to route this shit? 2019-10-16T18:35:23 < karlp> run d+/d- straight throuh io1/io2, and probably vias for teh vbus/ground. 2019-10-16T18:35:44 < karlp> "everytings four layers" when you're talking to people who give a shit abotu esd :) 2019-10-16T18:36:19 < Ecco> huh 2019-10-16T18:36:31 < Ecco> I was thinking of this part as some kind of a fuse 2019-10-16T18:36:34 < Ecco> which you want to put in series 2019-10-16T18:36:39 < Ecco> apparently I'm wrong :) 2019-10-16T18:36:43 < karlp> you're wrong :) 2019-10-16T18:36:56 < karlp> you can get them too, but that's not an esd suppressor 2019-10-16T18:37:00 < Ecco> ok 2019-10-16T18:37:50 < Ecco> but how does it work then? 2019-10-16T18:38:07 < karlp> that's not a topic for here :) 2019-10-16T18:38:13 < Ecco> imagine a layout like this (ascii art incoming) 2019-10-16T18:38:13 < karlp> but also, it has layout guides in the doc... 2019-10-16T18:38:24 < Ecco> yeah it has but the doc is really really short 2019-10-16T18:38:41 < Ecco> USB connector (dangerous outside world) ----+---- MCU 2019-10-16T18:38:51 < Ecco> |_ ESD chip 2019-10-16T18:39:15 < Ecco> if a discharge happens on the USB connector, why will it "go" into the ESD chip and not the MCU? 2019-10-16T18:41:12 < Ecco> :-D 2019-10-16T18:41:20 < Ecco> Here's a schematic by ST with an ESD chip 2019-10-16T18:41:45 < karlp> you can see from the internal diagram that's a diode right? 2019-10-16T18:41:47 < Ecco> https://i.imgur.com/kxeoqTC.png 2019-10-16T18:41:51 < Ecco> yes indeed 2019-10-16T18:41:59 < karlp> go read about tvs diodes if you want to understand what's inside it. 2019-10-16T18:42:03 < Ecco> ok 2019-10-16T18:42:22 < karlp> basically diode does nothing until a certain voltage is reached, then it becomes conductive. 2019-10-16T18:42:27 < Ecco> ok 2019-10-16T18:42:56 < Ecco> but then, more conductive than, say, an STM32? 2019-10-16T18:43:06 < Ecco> I mean, at 30kV, everything is pretty much conductive :-D 2019-10-16T18:43:13 < Ecco> :-D 2019-10-16T18:43:20 < karlp> Ecco: so go read about tvs diode theory then. 2019-10-16T18:43:37 < Ecco> ok :) 2019-10-16T19:16:51 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-16T19:17:39 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-16T19:17:41 -!- day__ is now known as day 2019-10-16T19:26:26 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-16T19:31:39 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-16T19:39:26 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-16T19:41:23 * karlp balks at how much machinery needs to be layered on to get tracing nicely. 2019-10-16T20:02:50 < antto> tracing as in debugging? 2019-10-16T20:08:49 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-16T20:08:51 < kakimir64> bakc> 2019-10-16T20:10:42 < kakimir64> is it true that interrupt save api can only set pdTrue to higherPriorityTaskWoken flag? if I use it chained inside isr? 2019-10-16T20:11:14 < kakimir64> it cannot clear it to pdFalse? 2019-10-16T20:11:44 < kakimir64> I assume so because you need to initialize variable with pdFalse before using it as parameter in isr safe api 2019-10-16T20:22:16 < karlp> antto: using swo to get timing of task switching. 2019-10-16T20:22:42 < karlp> instead of timing it on the target, like https://www.freertos.org/rtos-run-time-stats.html does. 2019-10-16T20:25:32 < antto> fancy 2019-10-16T20:26:23 < antto> not sure about them tasks, but i've used LED toggling to see how long my main polling "tasks" take on muh simple sh*t 2019-10-16T20:26:27 < antto> no OS 2019-10-16T20:28:11 < karlp> looks like i'm overflowing itm anyway by turning on timestamping of it all. 2019-10-16T20:28:35 < karlp> I wonder how close you could go by just yoloing it and timestamping it host side. 2019-10-16T20:33:01 < karlp> gek, the amount of shitty host sw to write for that gets gross an unfun quickly. 2019-10-16T20:37:56 < kakimir64> infreertos do you need to use some freertos method to control interrupts 2019-10-16T20:38:21 < karlp> what? 2019-10-16T20:38:34 < kakimir64> NVIC_ClearPendingIRQ causes hardfault 2019-10-16T20:39:15 < kakimir64> NVIC_ClearPendingIRQ(UART0_IRQn); inside a task 2019-10-16T20:39:29 < karlp> why would you be clearing irq flags from tasks anyway? 2019-10-16T20:39:34 < karlp> you clear irq flags i irqs 2019-10-16T20:40:24 < kakimir64> hmm 2019-10-16T20:40:55 < kakimir64> well I don't want uart to generate any interrupts before vUartReceiverTask 2019-10-16T20:41:07 < karlp> so don't turn it on until then. 2019-10-16T20:41:13 < kakimir64> or - before scheduler is up 2019-10-16T20:41:27 < karlp> just setup your uart in the start of your uart task? 2019-10-16T20:41:42 < kakimir64> but turn interrupts on before initializing uart? 2019-10-16T20:41:49 < kakimir64> or *uart interrupt 2019-10-16T20:41:56 < karlp> why would you do that? (with or without freertos) 2019-10-16T20:42:27 < kakimir64> do what? 2019-10-16T20:42:48 < kakimir64> interrupt is required 2019-10-16T20:43:10 < kakimir64> turning it on in task causes hardfault at clear pending phase 2019-10-16T20:45:18 -!- ekaologik [~quassel@p4FF16AEE.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 2019-10-16T20:50:42 -!- ekaologik [~quassel@p4FF16AEE.dip0.t-ipconnect.de] has joined ##stm32 2019-10-16T20:52:23 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-16T20:54:54 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-16T21:02:57 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-16T21:03:29 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-16T21:04:55 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Read error: Connection reset by peer] 2019-10-16T21:07:09 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 268 seconds] 2019-10-16T21:14:11 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-16T21:23:43 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-16T21:25:43 -!- tcth [~tcth@adsl-130-227.dsl.init7.net] has quit [Ping timeout: 250 seconds] 2019-10-16T21:31:35 < kakimir64> I see in disassembly there is thumb and non-thumb instructions mixed in standard functions 2019-10-16T21:32:00 < kakimir64> this might be a problem? 2019-10-16T21:32:31 < mawk> how did you come up to doing that ? 2019-10-16T21:32:35 < mawk> did you change gcc options ? 2019-10-16T21:32:43 < Cracki> disassembly bug 2019-10-16T21:32:45 < mawk> you should have -mthumb on the command line itself 2019-10-16T21:32:46 < mawk> yes 2019-10-16T21:33:20 < Cracki> what it *should* do is start disassembling at the same location the program would be executed, and follow all branches and jumps 2019-10-16T21:33:35 < Cracki> what it tends to do is disassemble everything assuming alignment 2019-10-16T21:33:41 < mawk> ah 2019-10-16T21:33:42 < Cracki> and that catches data 2019-10-16T21:33:48 < mawk> yeah 2019-10-16T21:33:59 < mawk> yeah if you have debug info handy it's way easier to not mistake data for code 2019-10-16T21:55:41 < Cracki> hell yeah! they wanna turn Magneto into a palestinian! 2019-10-16T21:59:30 < kakimir64> I need to reboot I fkd bluetooth somehow 2019-10-16T21:59:33 < kakimir64> > 2019-10-16T22:00:00 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-16T22:05:58 -!- kakimir3271 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-16T22:06:05 -!- kakimir3271 is now known as kakimir64 2019-10-16T22:08:53 < kakimir32> https://drive.google.com/file/d/1Dyk8gHLz4u1x_J3I6YRltga_ALwS1Ops/view?usp=sharing mawk 2019-10-16T22:09:15 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e17b:11c4:a3b3:6064] has joined ##stm32 2019-10-16T22:09:32 < kakimir32> stepped in code editor that view comes right after strlen.. not sure what to think of it 2019-10-16T22:10:22 < kakimir32> and if I F5 once more it can be found from memmanage handler trap when halted 2019-10-16T22:14:43 < Steffanx> so is mawk going dutch? 2019-10-16T22:16:13 -!- fenugrec_ [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-16T22:23:10 < kakimir32> Steffanbox - tell me what I can see in my IDE view? I don't totally understand 2019-10-16T22:24:13 -!- fenugrec_ [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-16T22:26:42 < Steffanx> idk. is it macro using strlen? 2019-10-16T22:29:38 < kakimir32> nope 2019-10-16T22:30:07 < kakimir32> note mismatch in editor and disassembly view 2019-10-16T22:31:10 < kakimir32> note instruction suffixes in disassebly view 2019-10-16T22:33:19 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-16T22:33:59 < Steffanx> do you expect me to have a clue? 2019-10-16T22:36:35 < kakimir32> (shrug) 2019-10-16T22:38:13 < kakimir32> anyone have strlen disassembly to show for comparison? 2019-10-16T22:46:45 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 250 seconds] 2019-10-16T22:48:33 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-16T22:50:40 < catphish> seems like a pretty trivial function no? 2019-10-16T22:51:27 < steve> Hello, if anyone knows FreeRTOS... It appears that osThreadCreate, ie. xTaskCreate disables interrupts until the scheduler is started? Is that right? 2019-10-16T22:53:55 < Mangy_Dog> tbh i dont know 2019-10-16T22:54:04 < Mangy_Dog> i thought nothing could really disable hardware interupts 2019-10-16T23:00:49 < steve> What do you mean? Of course you can disable interrupts. Anyway I call osThreadCreate (which is the cmsis wrapper) , and then do a HAL_Delay (STM HAL code that blocks until ticks count up) before the scheduler is started, and it blocks forever. If I do HAL_Delay inside a task after scheduler is started, it works fine. If I do HAL_Delay before I create the task it also works fnie. That's super odd to me. 2019-10-16T23:00:55 < steve> The weirdest thing is I'm commenting out code in xTaskCreate and it seems like the line that causes the behavior is pvPortMalloc, allocating the stack for the task. Makes zero sense 2019-10-16T23:02:06 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-16T23:06:47 -!- Jybz [~jibz@91-166-99-132.subs.proxad.net] has quit [Ping timeout: 268 seconds] 2019-10-16T23:14:02 < steve> aww *insert swear word* the uxCriticalNesting count is not initialized until the scheduler is started so any critical sections that happen won't get the interrupts re-enabled until xPortStartScheduler is called. Why..... 2019-10-16T23:14:49 -!- ekaologik [~quassel@p4FF16AEE.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-16T23:19:24 < steve> wasted a day "discovering" that because I thought it made sense to add my initialization code to a STM32 demo project just before the scheduler is started 2019-10-16T23:26:38 < qyx> TIL: do not share AC-adapter-detect voltage divider with a STM32 analog input to sense the adapter voltage 2019-10-16T23:27:06 < qyx> if there is no adapter present, STM32 VDD=0V, the power supply is not started 2019-10-16T23:27:35 < qyx> when an adapter is connected, AC-adapter-detect cannot reach the desired threshold to start the power supply 2019-10-16T23:27:47 < qyx> because current from the divider leaks to the STM32 2019-10-16T23:28:15 < qyx> (still unpowered at that time) 2019-10-16T23:40:16 -!- gnom [~aleksande@178.150.7.153] has joined ##stm32 2019-10-16T23:40:51 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-16T23:42:05 < karlp> steve: st's freertos integration does "special" things with systick.... 2019-10-16T23:46:10 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-16T23:47:21 -!- renn0xtk9 [~max@2a02:810d:1540:2448:e17b:11c4:a3b3:6064] has quit [Quit: Konversation terminated!] --- Day changed Thu Oct 17 2019 2019-10-17T00:08:03 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-17T00:11:08 < steve> karlp what specifically? What I described is in FreeRTOS proper 2019-10-17T00:14:10 < karlp> right, but hal_delay expects systick to be high priority. freertos expects it to be low priority 2019-10-17T00:14:15 < karlp> ==> hanging crap 2019-10-17T00:14:29 < karlp> and/or the need to have a wrapper handler that feeds them both. 2019-10-17T00:15:12 < mawk> cubemx specifically addresses this by asking you don't use systick in HAL 2019-10-17T00:15:16 < mawk> but another timer 2019-10-17T00:15:24 < mawk> just needs 1ms time base 2019-10-17T00:16:10 < karlp> yeah, that's one way of doing it :) 2019-10-17T00:16:26 < karlp> I had some good links for it open yesterday, but, not using hal, so closed them again. 2019-10-17T00:18:08 < steve> OK haven't ran into the priority issue yet but I'll keep that in mind. Yes, the demo project I have does HAL_IncTick with a timer and ticks FreeRTOS with SYSTick 2019-10-17T00:19:12 < steve> speaking of not using HAL, am I correct that CubeMX uses HAL? What are the alternatives? Mbed? 2019-10-17T00:19:20 < mawk> cubemx makes you use HAL yes, or LL 2019-10-17T00:19:28 < mawk> alternative is libopencm3 yes for instance 2019-10-17T00:19:37 < mawk> you have mbed also indeed but nobody seriously uses that I think 2019-10-17T00:19:41 < mawk> you have arduino libs also 2019-10-17T00:21:11 < steve> cool thanks.. Looking at those will help me in trash talking HAL/Cube as much as possible 2019-10-17T00:21:26 < mawk> lol 2019-10-17T00:26:38 -!- c10ud^^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 276 seconds] 2019-10-17T00:28:26 < mawk> yes Steffanx 2019-10-17T00:33:47 < karlp> zephyr, chibios, nuttx, laks, these all have hal layers too. 2019-10-17T00:41:31 < bitmask> BOM: 108 components 2019-10-17T00:41:46 < bitmask> now I just gotta look for protective stuff I might need? 2019-10-17T00:46:52 -!- con3|2 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-17T00:47:27 < steve> thanks karlp 2019-10-17T00:48:22 < Jan-> maybe I should ask a more general question 2019-10-17T00:48:27 < Jan-> this may be a religious question 2019-10-17T00:48:41 < Jan-> which of the basic libraries should I use? libopencm3? arduino? 2019-10-17T00:48:50 < mawk> I use HAL 2019-10-17T00:48:53 < mawk> some here like libopencm3 2019-10-17T00:49:06 < Mangy_Dog> https://www.bbc.co.uk/news/technology-50073102 finally some sense... This shitty good for nothing government has actually done something right 2019-10-17T00:49:08 < mawk> some others use nothing at all I think, bare registers or maybe with a sprinkle of CMSIS 2019-10-17T00:49:14 < bitmask> stdperiph 2019-10-17T00:49:19 < mawk> stdperiph is the old thing 2019-10-17T00:49:22 < Jan-> I'm not sure how to even try libopencm3 2019-10-17T00:49:28 < Jan-> I have platformio but it's not supported with the board I have 2019-10-17T00:49:37 < bitmask> I liked stdperiph better than hal 2019-10-17T00:49:51 < bitmask> but I didnt really give hal a chance 2019-10-17T00:49:57 < mawk> libopencm3 is very simple Jan- , more than ST stuff 2019-10-17T00:50:05 -!- con3 [~kvirc@146.232.65.240] has quit [Ping timeout: 264 seconds] 2019-10-17T00:50:08 < mawk> just go into the directory, build an example with make 2019-10-17T00:50:09 < Jan-> I loooked at the ST library in cube 2019-10-17T00:50:14 < mawk> provided you have the ARM GCC in your path 2019-10-17T00:50:15 < Jan-> it seemed sort of wordy 2019-10-17T00:50:16 < mawk> that's all 2019-10-17T00:50:23 < mawk> well it's well documented I'd say 2019-10-17T00:50:36 < mawk> or at least well delimited with tons of tags START XXX CODE, END XXX CODE 2019-10-17T00:50:44 < Jan-> yes I got that 2019-10-17T00:50:48 < mawk> but it's not that wordy, it's just exhaustive in its setup of the board 2019-10-17T00:50:51 < Jan-> I thought it'd be better to delete all that 2019-10-17T00:50:52 < mawk> which is a good thing 2019-10-17T00:50:58 < mawk> you don't want to do that 2019-10-17T00:51:05 < mawk> else, you wouldn't be able to run cube MX ever again after that 2019-10-17T00:51:05 < Jan-> er ok 2019-10-17T00:51:09 < Jan-> oh. 2019-10-17T00:51:16 < Jan-> well I wasn't planning to run cube mx ever again anyway 2019-10-17T00:51:17 < mawk> or you could but it would mess up horribly your code 2019-10-17T00:51:20 < karlp> back to trolling again hey. 2019-10-17T00:51:20 < mawk> lol 2019-10-17T00:51:28 -!- con3|2 [~kvirc@146.232.65.240] has quit [Ping timeout: 252 seconds] 2019-10-17T00:51:30 < mawk> if you want to add more GPIO pins or whatever you'll need cubemx Jan- 2019-10-17T00:51:34 < mawk> who karlp ? Jan- ? 2019-10-17T00:51:37 < karlp> yar 2019-10-17T00:51:47 < karlp> asking the same questions over and over, getting the same answers... 2019-10-17T00:51:59 < qyx> wtf is even platformio 2019-10-17T00:52:06 < mawk> I don't think Jan- is trolling personnally 2019-10-17T00:52:06 < Cracki> nothing of relevance 2019-10-17T00:52:08 < Jan-> qyx: someone recommended it. 2019-10-17T00:52:10 < mawk> some hipster stuff I think 2019-10-17T00:52:12 < Jan-> beyond that I have no idea. 2019-10-17T00:52:17 < qyx> Jan-: so let it go 2019-10-17T00:52:19 < Jan-> it does make things quite simple. 2019-10-17T00:52:20 < qyx> arduino too 2019-10-17T00:52:25 < mawk> « PlatformIO is an open source ecosystem for IoT development » 2019-10-17T00:52:35 < Cracki> simple is worthless if it doesn't do what you want 2019-10-17T00:52:37 < qyx> too many buzzwords in a single sentence 2019-10-17T00:52:37 < bitmask> someone pro-itize my schematic for me 2019-10-17T00:52:42 < Jan-> well I mean it works. 2019-10-17T00:52:49 < Cracki> "ecosystem" yeah they're trying to sell you something 2019-10-17T00:53:03 < Cracki> bitmask, autorouter, then it's pro 2019-10-17T00:53:12 < qyx> simply download libopencm3 examples, some blinky-led one 2019-10-17T00:53:13 < bitmask> I havent started board layout yet 2019-10-17T00:53:15 < qyx> do make 2019-10-17T00:53:17 < qyx> flash 2019-10-17T00:53:22 < qyx> enjoy blinky led 2019-10-17T00:53:25 < qyx> then modify and play 2019-10-17T00:53:27 < mawk> yeah it's very easy to get started with libopencm3 2019-10-17T00:53:32 < Jan-> I think getting that set up under windows is likely to be a bit of a PITA 2019-10-17T00:53:41 < mawk> no 2019-10-17T00:53:43 < qyx> why so 2019-10-17T00:53:51 < mawk> you have ARM GCC distributions for windows, pre-compiledf 2019-10-17T00:53:59 < mawk> download, unzip somewhere, add to %PATH% 2019-10-17T00:55:01 < Jan-> but I need more than gcc 2019-10-17T00:55:05 < qyx> now when windows even have gnu/linux tools 2019-10-17T00:55:13 < qyx> I am expecting it to add apt/rpm soon too 2019-10-17T00:55:34 < Cracki> https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads 2019-10-17T00:55:35 < kakimir64> is it possible that I don't have newlib linked to my project? 2019-10-17T00:55:35 < karlp> qyx: they have pacman in the msys2 installs... 2019-10-17T00:55:36 < Cracki> get it 2019-10-17T00:55:46 < kakimir64> only headers 2019-10-17T00:55:47 < Cracki> kakimir64, why would you think that 2019-10-17T00:55:48 < mawk> what more Jan- ? 2019-10-17T00:55:51 < karlp> Jan-: or, you know, just keep using platformio, as it's already working. 2019-10-17T00:55:52 < mawk> you need gcc and make, you're right 2019-10-17T00:56:00 < Cracki> that download has everything you want 2019-10-17T00:56:03 < karlp> unless you have explicit things you want differentyly, you can't be helped much. 2019-10-17T00:56:04 < Jan-> karlp: that was my thought. 2019-10-17T00:56:15 < mawk> you already have openocd, otherwise you have windows programs for flashing ST from ST themselves 2019-10-17T00:56:21 < Jan-> I'm just rather committed to using the arduino library that way. 2019-10-17T00:56:27 < Jan-> And it seemed a bit. I uhoh. Training wheels. 2019-10-17T00:56:37 < karlp> you sound like you need them anyway, so... poerfect. 2019-10-17T00:56:55 < Jan-> well, I don't use the arduino libraries on AVR, because they suck. 2019-10-17T00:57:39 < kakimir64> Cracki: in project settings I cannot find any reference to the source 2019-10-17T00:57:42 < kakimir64> only headers 2019-10-17T00:58:30 < Cracki> does it link 2019-10-17T00:58:48 < Cracki> it's not building newlib from source, it's likely using prebuilt 2019-10-17T01:00:42 < kakimir64> yes 2019-10-17T01:00:55 < kakimir64> it doesn't probs build 2019-10-17T01:02:22 < kakimir64> I mean would my freertos project even run without newlib linked to it? 2019-10-17T01:02:23 < Jan-> mawk I guess in theory I have openocd 2019-10-17T01:02:28 < Jan-> I'm not sure where it actually is or how to use it 2019-10-17T01:02:30 < kakimir64> by just having headers only 2019-10-17T01:02:36 < Jan-> the platformio scripts seem to run it 2019-10-17T01:02:41 < Cracki> freertos dgaf about newlib 2019-10-17T01:02:52 < kakimir64> yes 2019-10-17T01:02:59 < Cracki> newlib is an implementation of C's std lib 2019-10-17T01:03:11 < kakimir64> Cracki: are you familiar with Eclipse CDT? 2019-10-17T01:03:19 < Cracki> hm... well I may be wrong, it prolly needs some standard c upport 2019-10-17T01:03:19 < mawk> you have it Jan- , and it works 2019-10-17T01:03:25 < mawk> you can flash 2019-10-17T01:03:30 < Cracki> fuck my s key 2019-10-17T01:03:41 -!- grummund_ [~grummund@unaffiliated/grummund] has joined ##stm32 2019-10-17T01:03:48 < mawk> it's not extremely intuitive to use but we can help you, just ask when you're at it Jan- 2019-10-17T01:03:53 < mawk> Paul is the openocd expert 2019-10-17T01:04:15 < kakimir64> Cracki: get a new kb 2019-10-17T01:04:26 -!- grummund [~grummund@unaffiliated/grummund] has quit [Ping timeout: 240 seconds] 2019-10-17T01:04:30 < kakimir64> or wash the keyboard 2019-10-17T01:04:32 < Cracki> ssssssssss 2019-10-17T01:04:35 < Cracki> no 2019-10-17T01:04:51 < kakimir64> pull keys and then open the case and you find 2 membranes 2019-10-17T01:04:58 < Cracki> the patina is already declared an endangered species 2019-10-17T01:04:58 < qyx> Jan-: you are approaching the problem from the wrong side 2019-10-17T01:05:08 < Jan-> I am? 2019-10-17T01:05:10 < qyx> how can you say you don't know what openocd is and how to use it 2019-10-17T01:05:15 < qyx> or libopencm3 or whatever 2019-10-17T01:05:37 < kakimir64> I don't know what libopencm3 is 2019-10-17T01:05:44 < kakimir64> and frankly I don't care 2019-10-17T01:05:47 < qyx> I would expect you to find solutions to a problem you have 2019-10-17T01:05:49 < qyx> you need a hal? 2019-10-17T01:05:52 < qyx> so search for one 2019-10-17T01:05:57 < qyx> oh, you find libopencm3 2019-10-17T01:06:07 < qyx> not you know that libopencm3 is basically a hal 2019-10-17T01:06:09 < Jan-> well I found a bunch. 2019-10-17T01:06:36 < kakimir64> there are manufacturer provided hals i would use as first option 2019-10-17T01:06:42 < kakimir64> unless totally crap 2019-10-17T01:06:44 < Jan-> I did. 2019-10-17T01:06:50 < Jan-> That's the first one I looked at. 2019-10-17T01:06:51 < karlp> qyx: they also heard that libopencm3's default position was "here's a makefile" and tehy said, "well, walking away then" so.... maybe just tell them to use keil+cubemx.... 2019-10-17T01:06:53 < qyx> idk, I see you are trying to collect keywords from different people 2019-10-17T01:06:55 < Jan-> It's sort of super wordy. 2019-10-17T01:06:58 < qyx> and then don't know what to do with them 2019-10-17T01:07:11 < karlp> qyx: aka, trolling 2019-10-17T01:07:41 < kakimir64> it's just helpless beginer syndrome 2019-10-17T01:07:41 < qyx> karlp: heh 2019-10-17T01:07:51 < qyx> Jan-: and did you try to run some examples? 2019-10-17T01:07:54 < qyx> follow some tutorial? 2019-10-17T01:07:55 < kakimir64> some may have it for years and years 2019-10-17T01:08:49 < qyx> because now you are just like, oh I need gcc, oh windows, oh fuckit 2019-10-17T01:09:09 < qyx> not an iption. 2019-10-17T01:09:15 < kakimir64> Cracki: I don't see anything newlib related in linker command 2019-10-17T01:09:38 < Cracki> ssssssay what 2019-10-17T01:09:40 < qyx> like dongs, oh this shit uses 6 copies of java JRE, I am not gonna use it 2019-10-17T01:09:59 < Cracki> what _do_ you see your linker doing 2019-10-17T01:10:27 < Cracki> this isn't a game of battleship 2019-10-17T01:10:43 < kakimir64> -L"/home/janne/Documents/MCUXpresso_11.0.1_2563/workspace_lpc15xx/lpc_chip_15xx/Debug" -L"/home/janne/Documents/MCUXpresso_11.0.1_2563/workspace_lpc15xx/CMSIS_DSPLIB_CM3/lib" -Xlinker -Map="ovencontrol.map" -Xlinker --gc-sections -Xlinker -print-memory-usage -mcpu=cortex-m3 -mthumb -T "ovencontrol_Debug.ld" 2019-10-17T01:10:59 < Cracki> hmmm 2019-10-17T01:11:05 < Jan-> I guess I'm just a bit aware that gcc takes one file and turns it into one lump of compiled code. 2019-10-17T01:11:27 < Jan-> whereas to get to an .elf file involves a bit more than that 2019-10-17T01:11:27 < Cracki> at least that line announces library directories 2019-10-17T01:12:13 < Cracki> open those .ld files, kakimir64 2019-10-17T01:12:13 < kakimir64> library path for newlib should be in that command right? 2019-10-17T01:12:27 < Cracki> the last one is probably the most interesting, "ovencontrol_Debug.ld" 2019-10-17T01:12:34 < kakimir64> btw. I just removed -nostdlib 2019-10-17T01:12:37 < Cracki> yes 2019-10-17T01:13:02 < Cracki> either that, or gcc magically knows from some other mechanism where to look 2019-10-17T01:13:09 < kakimir64> does it? 2019-10-17T01:13:10 < Cracki> which is expected 2019-10-17T01:13:17 < steve> I thought newlib or nanolib was default in arm-none-eabi-gcc? One of those 2019-10-17T01:13:31 < Cracki> normal desktop gcc compilation and linkage also doesn't need to know where libc can be found, it knows already 2019-10-17T01:13:34 < mawk> yes steve 2019-10-17T01:13:36 < qyx> Jan-: it may or may not, you can write a .bat file to compile your whole project 2019-10-17T01:13:37 < steve> you can see the libs where arm-none-eabi-gcc is installed 2019-10-17T01:13:39 < mawk> not nanolib, it's called newlib 2019-10-17T01:13:43 < mawk> and there is a nano flavor 2019-10-17T01:13:53 < kakimir64> includes have multiple arm-none-eabi directories 2019-10-17T01:13:55 < qyx> Jan-: but it is not done that way, there are build systems to take care of it 2019-10-17T01:13:55 < Cracki> look into that ld file 2019-10-17T01:14:01 < kakimir64> Cracki: ok 2019-10-17T01:14:15 < Cracki> that ld file is the next interesting thing 2019-10-17T01:14:15 < Jan-> qyx: as I understand it that's sort of what platformio is doing 2019-10-17T01:14:18 < mawk> Jan-: gcc produces elf binaries directly, what do you mean ? 2019-10-17T01:14:38 < mawk> you can ask gcc to produce intermediate files you link into an elf (still with gcc), or produce the elf directly 2019-10-17T01:14:52 < Jan-> okay 2019-10-17T01:15:32 < qyx> Jan-: from a quick google it seems platformio is just using SCons build system 2019-10-17T01:15:47 < Cracki> that's one of these weird ones 2019-10-17T01:15:52 < qyx> I like it 2019-10-17T01:16:02 < Cracki> you like haskell 2019-10-17T01:16:03 < mawk> abscons 2019-10-17T01:16:13 < mawk> haskell is on my list of langs to learn 2019-10-17T01:16:39 < Cracki> if you want something a little more immediately useful, look into forth 2019-10-17T01:16:54 * qyx pouring a boiling water into a magic vifon noodles 2019-10-17T01:17:05 < Cracki> that's quite useful for interpreters in embedded systems 2019-10-17T01:17:45 < kakimir64> Cracki: in *library.ld i see a group() of .a files: "libgcc.a", "libc_nano.a", "libm.a", "libcr_newlib_none.a" 2019-10-17T01:17:47 < karlp> qyx: it's scons, plus a _mountain_ of wrapper shit. 2019-10-17T01:17:51 < Cracki> bingo 2019-10-17T01:17:52 < kakimir64> and my build failed now 2019-10-17T01:17:59 < karlp> it's a long way from "just a scons build" 2019-10-17T01:18:05 < kakimir64> freertos/src/croutine.d:1: *** multiple target patterns. Stop. 2019-10-17T01:18:11 < kakimir64> I need to figure this out 2019-10-17T01:18:48 < Cracki> "libcr_newlib_none.a" yields people suffering from mcuxpresso 2019-10-17T01:19:15 < mawk> lol 2019-10-17T01:19:38 < karlp> kaks has spent the last 3-4 years suffering from mcuxpresso and the lpc world 2019-10-17T01:19:43 < mawk> .d files are dependancy files kakimir64 2019-10-17T01:19:44 < Cracki> this has screenshots https://www.silabs.com/community/software/simplicity-studio/forum.topic.html/force_gcc_to_linkl-DMyz 2019-10-17T01:19:47 < mawk> just delete all of them and try again 2019-10-17T01:19:58 < kakimir64> make clean? 2019-10-17T01:20:06 < mawk> well for make at least, but I guess it's the same for your build system 2019-10-17T01:20:15 < mawk> yeah try that at first, check if the .d go away 2019-10-17T01:20:56 < Cracki> https://i.kym-cdn.com/photos/images/newsfeed/001/392/890/953.jpg 2019-10-17T01:21:17 < Cracki> curbstomp those files if they don't get the hint 2019-10-17T01:21:32 < kakimir64> should I set to linker -nostartfiles as I have my own? 2019-10-17T01:22:18 < mawk> you link with crt1.o and so on ? 2019-10-17T01:22:22 < Cracki> you can do that but then you need to tell it those "implied" libraries explicitly 2019-10-17T01:22:26 < mawk> if no, no, don't use -nostartfiles 2019-10-17T01:22:33 < mawk> otherwise C/C++ constructors don't run and so on 2019-10-17T01:22:39 < mawk> start files are harmless 2019-10-17T01:23:02 < kakimir64> ok I figure 2019-10-17T01:23:27 < Cracki> some docs: https://gcc.gnu.org/onlinedocs/gcc/Link-Options.html 2019-10-17T01:24:10 < mawk> you need ld documentation also 2019-10-17T01:25:46 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-17T01:27:17 < kakimir64> how do I make eclipse build it's lib dependancy same time with the project itself? 2019-10-17T01:28:02 < kakimir64> something is broken but I can get it to build somehow 2019-10-17T01:29:59 < kakimir64> yes by selecting same build configuration for both and then build 2019-10-17T01:31:24 < kakimir64> region `FLASH' overflowed by 11532 bytes 2019-10-17T01:32:37 < mawk> big overflow 2019-10-17T01:33:09 < kakimir64> oh wait I need to manage debug configuration for that linked lpc lib 2019-10-17T01:33:16 < kakimir64> -O0 there 2019-10-17T01:33:35 < kakimir64> I thought these things would inherit the configuration from project that requires the lib 2019-10-17T01:34:23 < kakimir64> region `FLASH' overflowed by 10196 bytes 2019-10-17T01:34:47 < mawk> try -Os 2019-10-17T01:34:52 < mawk> optimize for size 2019-10-17T01:35:14 < kakimir64> minor change 2019-10-17T01:35:22 < mawk> :( 2019-10-17T01:35:37 < kakimir64> yes 2019-10-17T01:35:39 < kakimir64> I did 2019-10-17T01:37:07 < kakimir64> -nostdlib back on 2019-10-17T01:37:08 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-17T01:37:51 < kakimir64> size is over 20kilos smaller and it all works until strlen 2019-10-17T01:39:11 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-17T01:40:00 < mawk> your strlen is very simple 2019-10-17T01:41:02 < kakimir64> I don't knew I was missing anything before that 2019-10-17T01:41:23 < kakimir64> *didn't 2019-10-17T01:43:03 < kakimir64> I think literally 2 standard functions I use are strlen and strcpy_s 2019-10-17T01:44:42 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-17T01:46:03 < Jan-> So I found something called stm32-base, does anyone have any experience with it? 2019-10-17T01:46:35 < kakimir64> in what context? 2019-10-17T01:46:41 < kakimir64> hal? 2019-10-17T01:46:55 < Jan-> I guess it's just that if 300 people immediately warned me off it, I'd probably not use it. 2019-10-17T01:47:20 < kakimir64> never heard 2019-10-17T01:47:25 < Jan-> fair enough 2019-10-17T01:47:59 < con3> Cracki: i somewhat found the issue 2019-10-17T01:48:07 < Cracki> do tell 2019-10-17T01:48:25 < con3> eh i dont have a solution yet but it's cause i tried chaining timers 2019-10-17T01:48:38 < con3> i still have no idea why exactly it's happening 2019-10-17T01:48:41 < Cracki> eh that MAY have been useful to know :> 2019-10-17T01:49:00 < con3> Cracki: sorry i was awake for an ungodly amount of time 2019-10-17T01:49:08 < Cracki> sleep solves many problems 2019-10-17T01:49:20 < con3> literally slept, woke up and got the fact that it was doing this 2019-10-17T01:50:17 < con3> so the one timer im just using to scale down the clock even more, cause i need a very slow pulse, so it has a prescaler, feeds another clock and each time the master timer produces an update event the slave timers counter incremements 2019-10-17T01:50:20 < con3> so that works 2019-10-17T01:50:54 < con3> but running the pwm from the internal clock instead of having it as a slave and the pin toggles, running it as a slave and no toggle 2019-10-17T01:51:37 < Cracki> must be some very slow pwm if you need a prepositioned timer 2019-10-17T01:51:54 < con3> like its got a period of 15 minutes 2019-10-17T01:52:02 < con3> so very slow 2019-10-17T01:52:09 < Cracki> nice 2019-10-17T01:53:16 < con3> just need to figure out why it toggles on the one case and not the other 2019-10-17T01:53:18 * con3 debugs 2019-10-17T01:55:15 < kakimir64> what are those C constructors? 2019-10-17T01:55:42 < mawk> __attribute__((__constructor__)) void func(void) { puts("I am a constructor"); } 2019-10-17T01:55:47 < mawk> this, kakimir64 2019-10-17T01:55:51 < mawk> a function that will run before main() 2019-10-17T01:55:52 < Jan-> what's the difference between an stm32f407vet6 and one that ends vgt6 2019-10-17T01:55:54 < Jan-> I did google 2019-10-17T01:56:11 < Jan-> apparently it's a big enough difference that they need programming differently 2019-10-17T01:56:14 < con3> crap im gonna need to interconnected timers 2019-10-17T01:56:14 < mawk> let me look 2019-10-17T01:56:17 < Cracki> open data sheet or RM 2019-10-17T01:56:19 < kakimir64> mawk: resethandler? 2019-10-17T01:56:20 < mawk> I don't think they're programmed differently 2019-10-17T01:56:21 < Cracki> may be packaging 2019-10-17T01:56:25 < mawk> no kakimir64 2019-10-17T01:56:35 < Jan-> well, they have different board files in platformio 2019-10-17T01:56:35 < mawk> it's initialized alongside libc stuff kakimir64 , it doesn't run as soon as reset 2019-10-17T01:56:53 < Cracki> there may be no reason for them to have diff board files 2019-10-17T01:57:03 < Cracki> even tho the board files differ, it may be arbitrary 2019-10-17T01:57:13 < Jan-> hm I guess 2019-10-17T01:57:18 < mawk> VE is 512K flash, VG is 1M flash 2019-10-17T01:57:23 < Cracki> hm 2019-10-17T01:57:34 < Jan-> aha 2019-10-17T01:58:15 < kakimir64> mawk: any web page or document I should see for this? 2019-10-17T01:58:17 < Cracki> V=100 pins 2019-10-17T01:58:25 < Cracki> table 99 ordering stuff https://www.st.com/resource/en/datasheet/stm32f407ve.pdf 2019-10-17T01:58:43 < Cracki> T=lqfp as opposed to bga and whatnot 2019-10-17T01:59:31 < mawk> kakimir64: __attribute__((__constructor__)) places the function's address in a special section called .init_array, and the libc start files (the things that -nostartfiles disables) will go through all of them and call them all 2019-10-17T02:00:01 < kakimir64> interestings 2019-10-17T02:00:13 < mawk> https://gcc.gnu.org/onlinedocs/gcc-4.7.2/gcc/Function-Attributes.html#index-g_t_0040code_007bconstructor_007d-function-attribute-2509 2019-10-17T02:00:22 < kakimir64> at what point? before entering reset handler code? 2019-10-17T02:00:26 < mawk> no, no 2019-10-17T02:00:29 < mawk> just before main() 2019-10-17T02:00:40 < kakimir64> I see 2019-10-17T02:00:48 < mawk> the reset handler is called by the processor, it's fundamental 2019-10-17T02:00:54 < mawk> while constructor functions are not fundamental 2019-10-17T02:01:03 < karlp> do you _really_ have a crt0 that you can use? that's pretty special. 2019-10-17T02:01:11 < kakimir64> I thought name of main function was not important 2019-10-17T02:01:24 < kakimir64> but I assume it triggers that init array 2019-10-17T02:02:23 < mawk> well if you had no libc the main function doesn't have to be named like this 2019-10-17T02:02:40 < mawk> but the libc litterally calls something like exit(main(argc, argv, envp)); 2019-10-17T02:02:41 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-17T02:05:02 < kakimir64> I need to go deeper 2019-10-17T02:08:17 < mawk> why ? 2019-10-17T02:08:23 < mawk> what do you want to know kakimir64 ? 2019-10-17T02:08:38 < kakimir64> what is inside init array 2019-10-17T02:09:12 < kakimir64> maybe I could build with startup files and see myself from map 2019-10-17T02:09:12 < mawk> a bunch of pointers to functions that must be called before main 2019-10-17T02:09:18 < mawk> there are stuff from the libc in it 2019-10-17T02:09:28 < mawk> for instance to initialize the objects stdin, stdout, stderr 2019-10-17T02:10:10 < kakimir64> I don't use them 2019-10-17T02:10:17 < kakimir64> http://static.grumpycoder.net/pixel/uC-sdk-doc/stdio_8c_source.html this? 2019-10-17T02:10:25 < mawk> it was just an example 2019-10-17T02:10:29 < mawk> with newlib it's not the caser 2019-10-17T02:10:30 < mawk> -r 2019-10-17T02:10:35 < mawk> why do you want to disable it ? 2019-10-17T02:10:56 < kakimir64> lemme see the size of it 2019-10-17T02:11:02 < mawk> of what 2019-10-17T02:11:06 < mawk> the start files ? 2019-10-17T02:11:10 < kakimir64> yes 2019-10-17T02:11:17 < kakimir64> I mean I will see myself 2019-10-17T02:11:27 < mawk> it's crtbegin.o crt1.o crti.o crtn.o crtend.o 2019-10-17T02:11:31 < mawk> all of these 2019-10-17T02:11:41 < mawk> just compare with and without -nostartfiles 2019-10-17T02:11:47 < mawk> but then some parts of libc won't function correctly 2019-10-17T02:12:05 < mawk> lookup for all __attribute__((__constructor__)) in the libc code 2019-10-17T02:12:06 < kakimir64> few kilos 2019-10-17T02:12:18 < kakimir64> there is multiple of them? 2019-10-17T02:12:21 < kakimir64> oh 2019-10-17T02:12:44 < kakimir64> I mean I'm within budget with this startupthing 2019-10-17T02:13:14 < mawk> try with the final object instead of adding all the .o together 2019-10-17T02:13:38 < kakimir64> I did 2019-10-17T02:14:11 < kakimir64> I think I need to download libc source? 2019-10-17T02:14:24 < mawk> why ? 2019-10-17T02:14:26 < mawk> ah 2019-10-17T02:14:31 < mawk> to check where the constructors are used ? 2019-10-17T02:14:38 < kakimir64> I'm interested 2019-10-17T02:14:41 < mawk> well the start files aren't just the constructors 2019-10-17T02:14:51 < mawk> it's some other stuff too 2019-10-17T02:14:52 < mawk> yeah 2019-10-17T02:14:58 < mawk> you can run nm on the start files to see what's in them 2019-10-17T02:15:10 < mawk> nm $(gcc -print-file-name=crtbegin.o) 2019-10-17T02:15:10 < kakimir64> thats better idea 2019-10-17T02:15:11 < mawk> and so on 2019-10-17T02:15:27 < mawk> with your embedded gcc tho, not your linux one 2019-10-17T02:17:48 < karlp> do you need to try using this stuff? 2019-10-17T02:17:59 < karlp> do you have a reset ahndler in your own code? 2019-10-17T02:19:07 < kakimir64> sure 2019-10-17T02:19:13 < kakimir64> in startup.c 2019-10-17T02:19:30 < karlp> then probably stop trying to use libc provided start files as well? 2019-10-17T02:20:04 < kakimir64> it's libc startup 2019-10-17T02:20:31 < kakimir64> I try to figure out what it includes 2019-10-17T02:20:45 < kakimir64> I did pretty well without it though 2019-10-17T02:25:48 < kakimir64> does startup do anything memory allocation related? 2019-10-17T02:33:28 < kakimir64> maybe I just solder a chip with more mem to my board 2019-10-17T02:33:38 < kakimir64> turn everything on then 2019-10-17T02:34:14 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: No route to host] 2019-10-17T02:34:55 < kakimir64> newlib seems to have useful facilities 2019-10-17T02:36:24 < karlp> you can use newlib without using their crt0 though... 2019-10-17T02:37:24 < kakimir64> what is there? 2019-10-17T02:39:02 < kakimir64> ah 2019-10-17T02:39:15 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-17T02:40:38 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-17T02:42:00 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-17T02:43:57 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-17T02:49:51 < Cracki> moo https://www.youtube.com/watch?v=6vwzaC4ArTs 2019-10-17T02:50:09 < kakimir64> waitaminute 2019-10-17T02:50:26 < kakimir64> how did malloc work without standard libs 2019-10-17T02:50:47 < kakimir64> void *pvPortMalloc( size_t xSize ) PRIVILEGED_FUNCTION { 2019-10-17T02:50:56 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-17T02:51:01 < kakimir64> void *p = malloc(xSize); 2019-10-17T02:54:42 < kakimir64> yeah time to move away from this for a while> 2019-10-17T02:58:10 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-17T03:04:37 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-17T03:06:45 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-17T03:24:23 < mawk> malloc is in the libc kakimir64 2019-10-17T03:24:28 < mawk> libc is linked in by default 2019-10-17T03:24:31 < mawk> malloc always works 2019-10-17T03:24:37 < kakimir32> oh 2019-10-17T03:24:41 < mawk> is always present rather, but it doesn't play nicely with freertos 2019-10-17T03:24:56 < kakimir32> I got that covered 2019-10-17T03:25:06 < kakimir32> with heap implementation 2019-10-17T03:25:15 < mawk> then don't use libc's malloc 2019-10-17T03:25:22 < mawk> change your heap_N.c 2019-10-17T03:26:54 < kakimir32> why I wouldn't use it? 2019-10-17T03:27:27 < mawk> you said you have your own malloc 2019-10-17T03:27:32 < mawk> for freertos 2019-10-17T03:27:39 < mawk> then tell freertos to use it instead of the libc malloc 2019-10-17T03:27:43 < mawk> otherwise what's the point 2019-10-17T03:28:00 < kakimir32> no I have sbrk 2019-10-17T03:28:09 < mawk> sbrk you made yourself ? 2019-10-17T03:28:20 < kakimir32> that I inspected myself 2019-10-17T03:28:23 < mawk> ah 2019-10-17T03:28:32 < kakimir32> and seems to work 2019-10-17T03:28:33 < mawk> sbrk is easy to write, but it's hard to get right with proper reentrancy support 2019-10-17T03:28:42 < mawk> so at first it seems to work but provokes all kind of unexpected bugs 2019-10-17T03:28:50 < mawk> that's why you have to patch out a few things 2019-10-17T03:29:08 < mawk> http://www.nadler.com/embedded/NXP_newlibAndFreeRTOS.html 2019-10-17T03:29:11 < mawk> like in this 2019-10-17T03:29:37 < kakimir32> lemme see 2019-10-17T03:29:45 < kakimir32> I was reading another page of nadler 2019-10-17T03:29:48 < kakimir32> I use his heap 2019-10-17T03:29:51 < mawk> the one for stm32 ? 2019-10-17T03:29:52 < mawk> yeha 2019-10-17T03:29:55 < mawk> there is a page for nxp too 2019-10-17T03:29:59 < mawk> that I just linked 2019-10-17T03:30:05 < mawk> if you use his implem then good 2019-10-17T03:30:10 < mawk> as long as it's not the default one 2019-10-17T03:31:21 < kakimir32> is there malloc in newlib that overrides libc malloc? 2019-10-17T03:34:49 < kakimir32> just random questions 2019-10-17T03:34:53 -!- mirage335 [~mirage335@2001:470:8ede:0:216:3eff:fe97:ac6d] has quit [Ping timeout: 245 seconds] 2019-10-17T03:36:23 < kakimir32> that nxp page is same than stm32 only heap related symbols are different 2019-10-17T03:41:44 -!- grummund_ is now known as grummund 2019-10-17T03:48:00 -!- mirage335 [~mirage335@204.141.172.74] has joined ##stm32 2019-10-17T04:00:43 -!- con3 [~kvirc@154.119.40.174] has joined ##stm32 2019-10-17T04:05:52 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-17T04:24:49 < bitmask> hmm, how should I add a power switch for my project? It can draw up to 20A, should I get a 20A switch or a small switch and a relay? 2019-10-17T04:31:06 < kakimir32> insufficient info 2019-10-17T04:40:10 < kakimir32> 20A of what from and to where? 2019-10-17T04:41:55 < Cracki> 20A is almost welding range, hell I could make two copper wires stick together with a mere 2-3 amps 2019-10-17T05:04:42 -!- con3 [~kvirc@154.119.40.174] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-17T05:07:57 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-17T05:47:07 -!- jackson [b694245b@182.148.36.91] has joined ##stm32 2019-10-17T06:12:34 < mawk> newlib is libc kakimir32 2019-10-17T06:12:43 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-17T06:12:52 < mawk> newlib is the name of redhat's tiny libc implementation 2019-10-17T06:27:55 -!- fc5dc9d4_ [~quassel@p57A32EB8.dip0.t-ipconnect.de] has joined ##stm32 2019-10-17T06:30:18 -!- fc5dc9d4 [~quassel@p5B0810BE.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds] 2019-10-17T06:32:58 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-17T07:12:19 < Thorn> who bought all the stock of stm32l476rct6 on lcsc :/ 2019-10-17T07:37:43 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-17T07:40:37 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-17T07:40:37 -!- day__ is now known as day 2019-10-17T07:48:06 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-17T07:57:05 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 276 seconds] 2019-10-17T08:34:10 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-17T08:37:41 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-17T08:43:27 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 250 seconds] 2019-10-17T08:57:28 -!- jackson [b694245b@182.148.36.91] has quit [Ping timeout: 260 seconds] 2019-10-17T08:59:44 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-17T09:13:25 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-17T09:15:18 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-17T09:49:07 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-17T09:51:55 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-17T09:52:12 -!- con3 [~kvirc@146.232.65.240] has joined ##stm32 2019-10-17T10:23:30 -!- sterna [~Adium@cvis.oal.lindholmen.se] has joined ##stm32 2019-10-17T10:31:17 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-17T10:31:49 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-17T10:32:09 -!- jly [uid355225@gateway/web/irccloud.com/x-adxmvxxydncrsrah] has joined ##stm32 2019-10-17T10:40:11 < Ecco> BTW, you may want to have a look at picolibc 2019-10-17T10:40:11 < Ecco> https://keithp.com/blog/ 2019-10-17T10:40:14 < Ecco> It's a fork of newlib 2019-10-17T10:40:16 < Ecco> that's not GPL 2019-10-17T10:58:55 < jly> good mroning Haohmaru 2019-10-17T10:59:23 < jly> i was thinking about diodes again 2019-10-17T11:00:35 < con3> jly: did the diodes thoughts keep you up? 2019-10-17T11:01:03 < jly> not that I know of 2019-10-17T11:01:14 < con3> just me then 2019-10-17T11:01:29 < con3> i hope so 2019-10-17T11:02:37 < Jan-> If I was completely new to stm32 (which I am) and I wanted a development board based around an stm32f407 which had good compatibility with as wide a variety of IDEs as possible, which would you recommend? 2019-10-17T11:02:42 < jly> https://usercontent.irccloud-cdn.com/file/pYohOr9J/yep.png 2019-10-17T11:02:42 -!- markus-k_ [~markus@server01.comtime-it.eu] has quit [Read error: Connection reset by peer] 2019-10-17T11:02:59 < jly> haha hi Jan- you're from ##electronics yea? 2019-10-17T11:03:25 < Jan-> er... I'm in that channel yes? 2019-10-17T11:03:35 < jly> there was this channel on freenode that probably imploded on itself 2019-10-17T11:04:37 < jly> very good though 2019-10-17T11:05:31 < zyp> mawk, uh, does sbrk really need to be reentrant? 2019-10-17T11:06:04 < Jan-> I hope it's a reasonable question anyway. I got two boards at the recommendation of some random dude on IRC and possibly I should have asked here first :) 2019-10-17T11:06:07 < zyp> I'd expect malloc to handle that and ensure sbrk is only called from one thread at a time 2019-10-17T11:06:29 < jly> why not get boards, boards are good 2019-10-17T11:07:13 < Jan-> well, the one I have isn't compatible with the libopencm3 framework in the IDE-like-thing I have 2019-10-17T11:07:25 < Jan-> I think it could be made to be but I have no idea how to do that 2019-10-17T11:08:13 < zyp> Jan-, why does it matter to your IDE what board you have? 2019-10-17T11:08:24 < zyp> what are you gonna do the day you make your own board? 2019-10-17T11:08:50 < zyp> or are you planning to only use some premade toy board forever? 2019-10-17T11:09:12 < Jan-> I guess. 2019-10-17T11:10:51 < Ecco> well, it's a libc 2019-10-17T11:10:59 < Ecco> targeted at microcontrollers 2019-10-17T11:11:05 < Jan-> I think in theory I would need a version of the platformio.ini file that was set up right. 2019-10-17T11:11:23 < Jan-> But I don't know if there would be other changes required. 2019-10-17T11:11:38 < Ecco> Haohmaru: it definitely works on cortex cores 2019-10-17T11:11:48 < zyp> Haohmaru, why wouldn't it? libc is platform independent 2019-10-17T11:11:49 < Ecco> I mean, aren't they the default anyway? 2019-10-17T11:12:59 < zyp> Jan-, maybe you should find a platformio-specific channel, I don't think many people here use it 2019-10-17T11:13:10 < Jan-> I'm happy to not use it. 2019-10-17T11:13:19 < zyp> yeah? 2019-10-17T11:13:20 < Jan-> I'm just not really very sure what the alternatives might be. 2019-10-17T11:13:50 < Ecco> well, to make a led blink you don't need no libc 2019-10-17T11:13:50 < Jan-> I guess in theory you can but I wouldn't know where to start. 2019-10-17T11:14:02 < Ecco> Haohmaru: libc is not a HAL 2019-10-17T11:14:07 < zyp> Haohmaru, libc doesn't have anything to do with hardware drivers 2019-10-17T11:14:17 < zyp> avrlibc is not just a libc 2019-10-17T11:14:41 < Ecco> libc does things like malloc, sin, uint8_t 2019-10-17T11:14:53 < Ecco> Haohmaru: well yes, a LED driver :-) 2019-10-17T11:15:01 < zyp> Haohmaru, not strictly, you can poke the memory addrs directly if you know which 2019-10-17T11:15:10 < Ecco> but then you've written a LED driver :-p 2019-10-17T11:15:36 < zyp> well, a register definition is the simplest form of a hardware driver 2019-10-17T11:16:09 < zyp> to put it another way: libc does not contain register definitions 2019-10-17T11:17:02 < zyp> and it doesn't make sense for it to either 2019-10-17T11:17:46 < zyp> I mean, why do you want a lib that contains both portd.whatever and strlen()? the former is common for anywhere, the latter is platform independent 2019-10-17T11:17:59 < Ecco> well maybe avrlibc's name sucks? 2019-10-17T11:18:11 < Ecco> then you want libopencm3 + picolibc ? 2019-10-17T11:18:40 < zyp> picolibc/newlib/whatever + cube/cmsis/libopencm3/whatever 2019-10-17T11:18:43 < zyp> problem solved 2019-10-17T11:20:50 < zyp> think of it another way; why would HAL authors bother implementing their own libc functions when newlib exists? 2019-10-17T11:24:36 < Jan-> I guess all this is why I'm asking about which dev boards to get 2019-10-17T11:24:54 < zyp> Jan-, it shouldn't matter 2019-10-17T11:25:23 < zyp> all decent HALs are focused on the chip, not the board it's mounted on, because most people intend to make their own boards anyway 2019-10-17T11:25:27 < Jan-> in theory it doesn't but I think that some of them make it easier than others 2019-10-17T11:25:48 < Jan-> otherwise they'd all be the same :) 2019-10-17T11:26:17 < zyp> the way I see it, HALs that bother with board definition bullshit makes it harder to support new boards, not easier 2019-10-17T11:26:32 < zyp> so you might as well not bother 2019-10-17T11:27:14 < Jan-> well I didn't design it :/ 2019-10-17T11:27:46 < Jan-> I mean it needs to know what sort of chip is on the board and so on 2019-10-17T11:27:55 < zyp> when I'm shopping for devboards or select a devboard, I'm picking the one that lets me test the features I want the easiest 2019-10-17T11:28:11 < Jan-> I guess my requirements are pretty simple in that regard. 2019-10-17T11:28:27 < zyp> then I prototype up some code with the devboard, learn some stuff, then make my own board 2019-10-17T11:29:03 < Jan-> that's sort of the stage I'm at 2019-10-17T11:29:16 < Jan-> Right now I have two ways I can program it 2019-10-17T11:29:19 < zyp> Jan-, so what's your project? 2019-10-17T11:29:23 < Jan-> I can load up cube and do it that way, or I can use platformio 2019-10-17T11:29:25 < zyp> what are you gonna make? 2019-10-17T11:29:38 < Jan-> It's a controller for switching power supplies to do colour mixing with LEDs. 2019-10-17T11:30:00 < zyp> Jan-, how is that gonna work? 2019-10-17T11:30:06 < Jan-> Haohmaru: well, it seems to automate all of the gcc/make/openocd stuff. 2019-10-17T11:30:21 < Jan-> Which is good because I have no idea how any of that works and there is no documentation. 2019-10-17T11:31:26 < Jan-> basically I just have to put a main.cpp in a folder called "src" and connect the board and type something like "pio run -t upload" and it just does its thing. 2019-10-17T11:31:59 < Jan-> the only thing is for this board it only supports the arduino framework 2019-10-17T11:32:02 < Jan-> so I uhoh. 2019-10-17T11:32:23 < Jan-> for some boards it supports libopencm3 but not this one 2019-10-17T11:32:40 < Jan-> which confuses me a bit because people say "oh they're all the same" well clearly they're not! 2019-10-17T11:33:17 < Jan-> yeah I think we all know where that will lead :) 2019-10-17T11:33:30 < Jan-> well yes the alternative way is sort of what I'm asking about. 2019-10-17T11:33:36 < zyp> Jan-, well, if you insist on using platformio, just buy a board supported by platformio 2019-10-17T11:33:49 < Jan-> I don't insist on using platformio at all I'm happy with whatever 2019-10-17T11:34:14 < Jan-> nobody ever seems to have heard of it which makes me like it less 2019-10-17T11:34:14 < zyp> then go grab libopencm3 or something and run it on any board? 2019-10-17T11:34:28 < Jan-> zyp well sure but... how? 2019-10-17T11:34:49 < zyp> I've heard of platformio, it's just that I haven't heard much good about it :) 2019-10-17T11:35:12 < zyp> Jan-, which board do you have? 2019-10-17T11:35:37 < Jan-> this one: https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-17T11:36:58 < zyp> umm 2019-10-17T11:37:37 < zyp> have you considered using mbed, as per that page? 2019-10-17T11:37:38 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-17T11:37:47 < Jan-> errr yeah I guess? 2019-10-17T11:37:49 < zyp> I believe mbed is supposed to be similar to platformio, just better :p 2019-10-17T11:38:14 < Jan-> it also mentions something called "seed arch max" 2019-10-17T11:38:59 < zyp> anyway, that board looks like a pretty plain stm32f407 breakout 2019-10-17T11:39:11 < Jan-> that's what I thought 2019-10-17T11:39:35 < Jan-> there is an spi flash chip and some other basic stuff on it but nothing that should alter anything unless I specifically want to use it. 2019-10-17T11:39:51 < zyp> yeah, which is true for most boards 2019-10-17T11:40:32 < Jan-> actually that really is it 2019-10-17T11:40:39 < Jan-> there's a header for an nrf24l01 2019-10-17T11:40:45 < Jan-> flash chip, a few leds and buttons 2019-10-17T11:41:11 < Jan-> 3.3v regulator and for some strange reason an sd card slot. 2019-10-17T11:41:18 < Jan-> not sure what you would do with the sd card slot but hey 2019-10-17T11:43:11 < zyp> duh, read and write sd cards of course 2019-10-17T11:43:44 < Jan-> so does this mbed studio thing have its own library 2019-10-17T11:43:54 < zyp> yes 2019-10-17T11:43:58 < zyp> mbed has its own hal 2019-10-17T11:44:01 < Jan-> I mean I don't need a whole RTOS 2019-10-17T11:44:04 < Jan-> that's way overkill 2019-10-17T11:44:16 < zyp> anyway, if you wanna use libopencm3, I figure you should start with these examples: https://github.com/libopencm3/libopencm3-examples/tree/master/examples/stm32/f4/stm32f4-discovery 2019-10-17T11:44:41 < Jan-> yes zyp I found lots of git hub repositories with code in them I just have absolutely no idea how to use any of it 2019-10-17T11:44:46 < Jan-> I mean I can download the c code but then what 2019-10-17T11:45:02 < zyp> build it and flash it? 2019-10-17T11:45:31 < Jan-> well sure but how 2019-10-17T11:45:36 < zyp> building is probably as easy as running «make» from the example folder 2019-10-17T11:45:42 < Jan-> Presumably it should involve gcc 2019-10-17T11:45:43 < Jan-> somehow 2019-10-17T11:45:59 < zyp> flashing depends on what hardware you have, but it'll probably be some openocd incantation 2019-10-17T11:46:26 < Jan-> well exactly 2019-10-17T11:46:34 < zyp> the example makefile should invoke gcc correctly, assuming you have it installed 2019-10-17T11:48:35 < Jan-> theoretically I have make installed but I find it very unreliable 2019-10-17T11:48:42 < Jan-> most makefiles assume they're running under linux 2019-10-17T11:49:25 < zyp> oh, you're trying to do this on windows? 2019-10-17T11:49:31 < zyp> good luck :) 2019-10-17T11:50:18 < Jan-> I really hate github it is the most unfriendly thing ever 2019-10-17T11:50:36 < zyp> haha 2019-10-17T11:50:42 < Jan-> also where it says "stm32f4-discovery" that's a specific board 2019-10-17T11:50:42 < zyp> how so? 2019-10-17T11:50:46 < Jan-> which is not the one I Have 2019-10-17T11:50:54 < zyp> I'm aware 2019-10-17T11:50:57 < zyp> it doesn't matter 2019-10-17T11:51:10 < Jan-> it's hard even to just download a file 2019-10-17T11:51:22 < Jan-> I find myself copying and pasting code out of the github site 2019-10-17T11:51:24 < zyp> uh, are you using git? 2019-10-17T11:51:59 < Jan-> I'm pointing a browser at github.com 2019-10-17T11:52:02 < zyp> if you're trying to get code from github without using git, you're doing it wrong :) 2019-10-17T11:52:37 < zyp> I mean, you can download a zip file of a repo, but that's kinda pointless since that'll leave out all the history and metadata 2019-10-17T11:53:15 < Jan-> hang on I've almost got it it's only 3 files 2019-10-17T11:53:23 < Jan-> let's see how spectacularly this fails 2019-10-17T11:53:29 < Jan-> well it might work it's super simple 2019-10-17T11:53:33 < Jan-> but let's see 2019-10-17T11:53:35 < zyp> also, the libopencm3-examples repo is referencing the libopencm3 repo as a submodule, and you're gonna need both to build 2019-10-17T11:53:53 < zyp> haha, no, it's not gonna work to just grab three files 2019-10-17T11:54:11 < Jan-> No, it isn't. 2019-10-17T11:54:22 < zyp> no wonder you're having struggles when you're not using the tools like they are supposed to 2019-10-17T11:54:23 < Jan-> my experience of this is that even if you get all the files make never really works 2019-10-17T11:54:30 < zyp> tell me, what is your programming experience so far? 2019-10-17T11:54:32 < Jan-> even on linux it fails more often than it works 2019-10-17T11:55:17 < Jan-> well I did a bit of javascript at university, lots of c# a few years ago, then some AVR C 2019-10-17T11:55:42 < zyp> have you ever used a revision control system, like git or svn or whatever? 2019-10-17T11:55:48 < Jan-> I try to avoid it 2019-10-17T11:55:54 < Jan-> more trouble than it's worth when it's just you 2019-10-17T11:55:54 < zyp> why? 2019-10-17T11:57:00 < zyp> is it really? 2019-10-17T11:57:17 < Jan-> yes. 2019-10-17T11:58:15 < zyp> have you ever reflected over the fact that you think that, while copypasting files from github? 2019-10-17T11:58:27 < Jan-> oh yes. 2019-10-17T11:58:59 < Jan-> usually my conclusion is that only the linuxy open source world could find a way of making it complicated to make a backup copy of your files every so often :) 2019-10-17T11:59:08 < zyp> do you think there could be a slight chance that everything would get a lot easier if you just learned how to use git? 2019-10-17T11:59:21 < Jan-> very slight. 2019-10-17T11:59:36 < zyp> oh, yeah, git is only for backups :) 2019-10-17T11:59:46 < Jan-> I think it's much more likely that the deeper I go down the rabbit hole of all this open source stuff the more layers of problems and complexity I will hit. 2019-10-17T12:00:08 < srk> it's actually about being able to manage the complexity 2019-10-17T12:00:15 < srk> which comes with embedded devel 2019-10-17T12:00:36 < srk> git is handy even during development to be able to look back and find why things break 2019-10-17T12:01:37 < Jan-> I think to do what you're proposing I need to be one of those people who lives permanently in linux and recompiles their kernel for fun every lunchtime and I am never going to be that person. 2019-10-17T12:02:00 < srk> he? 2019-10-17T12:03:14 < srk> why would anyone do that? we have buildsystems that do that for us and we mostly install updates, unless you're a kernel dev who needs to rebuild/reboot stuff often (except for working on modules which you can un/load dynamically) 2019-10-17T12:03:35 < Jan-> on the upside I now have this mbed thing 2019-10-17T12:03:37 < zyp> he's confusing linux fanboys with productive devs 2019-10-17T12:03:39 < Jan-> which I am installing 2019-10-17T12:03:52 < Jan-> who is 2019-10-17T12:03:58 < zyp> you :) 2019-10-17T12:03:59 < srk> any idea how come that variable initialization (with __attribute__(section..)) works during first boot but persists its value with sram2 retention? 2019-10-17T12:04:06 < Jan-> oh 2019-10-17T12:04:10 < Jan-> it's short for Janine :) 2019-10-17T12:04:15 < srk> tower_state_monitor_th.h:extern bool firstBoot __attribute__((section(".sram2Section"))); 2019-10-17T12:04:18 < srk> tower_state_monitor_th.c:bool firstBoot = true; 2019-10-17T12:04:24 -!- sterna [~Adium@cvis.oal.lindholmen.se] has quit [Ping timeout: 268 seconds] 2019-10-17T12:04:40 < srk> this works as expected but I would like to understand why (when using stand-by with sram2 retention) 2019-10-17T12:04:51 < Jan-> I appreciate Jan is a very serviceable guy name in many parts of the world but not on this occasion. 2019-10-17T12:04:54 < zyp> srk, presumably because your startup doesn't clear it like it clears .bss 2019-10-17T12:05:21 < srk> but how come it's set for the first time correctly? 2019-10-17T12:05:37 < Jan-> so do I have to use their OS thing if I use mbed 2019-10-17T12:05:40 < srk> this was helpful https://mcuoneclipse.com/2014/04/19/gnu-linker-can-you-not-initialize-my-variable/ 2019-10-17T12:05:46 < srk> but doesn't really explain this one 2019-10-17T12:06:13 < zyp> Jan-, my apologies, but my point still stands :) 2019-10-17T12:07:08 < zyp> I haven't touched mbed in 8-9 years or so, and I don't think they had an OS at the time, so I can't say 2019-10-17T12:07:08 < Jan-> yeah well. 2019-10-17T12:07:19 < Jan-> wander around the internet for long enough and you get a certain impression of linux people. 2019-10-17T12:07:32 < Jan-> I really really don't want to end up doing development like that 2019-10-17T12:08:23 < srk> yeah, it's terribly complicated to just be able to run 'make flash' to build & flash your code to mcu 2019-10-17T12:08:28 < zyp> I made the mistake of installing archlinux on a worklaptop once 2019-10-17T12:08:54 < zyp> that was pretty terrible 2019-10-17T12:09:09 < Jan-> srk I think the issue for me is that getting to that point will require 500 years of special setup techniques. 2019-10-17T12:09:47 < zyp> Jan-, anyway, your stereotypes are not helpful, FWIW I'm not even using linux 2019-10-17T12:09:54 < srk> well if you quit being unconstructive we can guide you thru that easily 2019-10-17T12:10:03 < srk> ^^ 2019-10-17T12:10:16 < Jan-> Let's see how we do with mbed 2019-10-17T12:10:20 < Jan-> it's installing right now 2019-10-17T12:10:28 < Jan-> it's sort of huge 2019-10-17T12:10:57 < srk> zyp: my startup clears bss but not sram2 which makes sense, but still how come sram2 value gets initialized on first boot 2019-10-17T12:11:19 < srk> major brainfuck 2019-10-17T12:11:19 < zyp> initialized to what? which chip is this? 2019-10-17T12:11:58 < srk> to true in this case, it's literally called firstBoot .. stm32l475 2019-10-17T12:13:50 < zyp> does STM32L have ECC on the RAM? maybe it reads 0 on ECC mismatch, and with random values out of reset it would mismatch 2019-10-17T12:14:23 < zyp> or maybe it even comes out as 0 out of power on reset 2019-10-17T12:15:13 < srk> it does have ECC 2019-10-17T12:15:35 < srk> interesting, that's why I'm wondering if I'm just lucky that this works or it is expected behavior 2019-10-17T12:16:05 < srk> guess the better way is to look at status register to see if we woke up from standby 2019-10-17T12:16:12 < zyp> in most cases, uninitialized ram content is undefined 2019-10-17T12:16:30 < Jan-> well the installer froze partway through at like 75% progress 2019-10-17T12:16:34 < Jan-> but the program seems to run... 2019-10-17T12:16:41 < srk> yep, that works correctly with the other few vars that don't have initializers 2019-10-17T12:16:51 < Jan-> oh god it wants me to have an "arm mbed" account just to use it 2019-10-17T12:17:03 < Jan-> I need more accounts on different websites like I need a butthole on my forehead 2019-10-17T12:19:24 < Jan-> oh yikes yes it is all based around their OS 2019-10-17T12:20:41 -!- con3 [~kvirc@146.232.65.240] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-17T12:20:54 < PaulFertser> Are you a MAKE:R Jan- ? 2019-10-17T12:20:54 < srk> at least you can install WSL ;) 2019-10-17T12:21:15 < zyp> makefiles are terrible 2019-10-17T12:21:22 < Jan-> am I a what PaulFertser? 2019-10-17T12:21:29 < PaulFertser> Jan-: https://en.wikipedia.org/wiki/Maker_movement 2019-10-17T12:21:42 < zyp> well, that's even worse 2019-10-17T12:22:16 < zyp> makefiles as a concept are good, but the implementation is dated and terrible 2019-10-17T12:22:18 < Jan-> PaulFertser: I'm just someone who ran out of performance on an AVR 2019-10-17T12:22:23 < zyp> modern replacements are good 2019-10-17T12:22:50 < zyp> Jan-, how did you build your AVR software? arduino? 2019-10-17T12:23:06 < Jan-> zyp: yes, although I never used the arduino library, I just used it as a way to write C code. 2019-10-17T12:23:42 < Jan-> Haohmaru: I tend to agree, this open source dev stuff either works perfectly first time or it falls apart so spectacularly. 2019-10-17T12:24:09 < PaulFertser> Jan-: so you are not identifying yourself as a "maker", I see. Your attitude to the topic looks a bit odd for somebody who studies or plans to participate in electronics projects "professionally", just a sidenote. 2019-10-17T12:24:17 < zyp> doesn't that go for anything? 2019-10-17T12:24:39 < Jan-> PaulFertser: yeah well. takes all sorts I guess. 2019-10-17T12:25:20 < zyp> shit works or it doesn't 2019-10-17T12:25:43 < Jan-> See theoretically mbed is supposed to support this board, as it says here: https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-17T12:26:06 < zyp> Jan-, that's not how it reads to me 2019-10-17T12:26:37 < zyp> to me it reads like «here's a recipe on how to get this board to work with mbed», written by some random mbed user 2019-10-17T12:26:43 < Jan-> well sure OK 2019-10-17T12:26:47 < PaulFertser> Jan-: what makes your board special? What else does it feature but the MCU? And why are you looking for "board" support and not "peripherals" support? 2019-10-17T12:26:55 < Jan-> I'm at the point where it says "if you would like to use an st-link v2 usb dongle..." 2019-10-17T12:27:11 < Jan-> it says "click on the compile button," I can't find a "compile" button. 2019-10-17T12:27:13 < zyp> Jan-, so what debugger hardware do you have? 2019-10-17T12:27:58 * Jan- should have got a nucleo board :/ 2019-10-17T12:28:07 < Jan-> everything supports those 2019-10-17T12:28:07 < karlp> has jan decided to do anything yet, or still just asking for generic answers 2019-10-17T12:28:13 < zyp> PaulFertser, already went over that before, but I'm not sure my point got through 2019-10-17T12:28:24 < karlp> everything supports yours too. only difference is the led is on a different pin.... 2019-10-17T12:28:41 < Jan-> so what do I select for "target" 2019-10-17T12:28:58 < zyp> karlp, something about controlling SMPSes to run RGB LEDs 2019-10-17T12:29:04 * karlp lols 2019-10-17T12:29:07 < Jan-> I uhoh like 200 of them Haohmaru 2019-10-17T12:29:46 < Jan-> it says here to select "seeed arch max" so I have 2019-10-17T12:29:50 < PaulFertser> Jan-, I'm not sure this is not an elaborate channel trolling attempt at this point. Can you share a link to your github or something? 2019-10-17T12:30:00 < zyp> haha 2019-10-17T12:30:06 < day> kek 2019-10-17T12:30:19 < karlp> kakimir32: newlib _is_ libc. 2019-10-17T12:30:39 < day> no pls please 2019-10-17T12:30:53 < srk> Initialized global variables have their initialization values copied from flash to RAM at this step. 2019-10-17T12:30:56 < srk> but how? 2019-10-17T12:31:21 < PaulFertser> srk: startup code copies a whole section from flash to RAM, defined by the ld script load address etc. 2019-10-17T12:31:30 < zyp> srk, startup code is responsible for both copying .data and clearing .bss 2019-10-17T12:31:30 < Jan-> does anyone have any idea what this means: https://pastebin.com/KY8x2krx 2019-10-17T12:31:52 < srk> hmm, that's /* relocate sidata to sdata */ 2019-10-17T12:32:01 < zyp> Jan-, yes 2019-10-17T12:32:47 < zyp> apparently your python is missing some lib 2019-10-17T12:33:02 < zyp> possible this is because your installation only got to 75% 2019-10-17T12:33:15 < karlp> Ecco: newlib isn't gpl either.... (picolibc is much simpler, and they got rid of some of the lgpl parts, that only affected you if you were targetting linux anyway) https://sourceware.org/newlib/COPYING.NEWLIB 2019-10-17T12:33:24 < Jan-> zyp: right. 2019-10-17T12:34:18 < Jan-> oh great and the uninstaller doesn't even work 2019-10-17T12:34:21 < Jan-> grr 2019-10-17T12:34:23 < Jan-> that is annoying 2019-10-17T12:34:39 < Jan-> oh yeah like apt has never done this 2019-10-17T12:35:01 < Jan-> pfft, I say again pfft 2019-10-17T12:35:06 < zyp> does it fucking matter, it's not a pissing contest 2019-10-17T12:35:32 < Jan-> ok so mbed is no go 2019-10-17T12:35:39 < zyp> Haohmaru, we're well aware you're full of piss 2019-10-17T12:36:16 < Jan-> that's sort of what I'm doing Haohmaru 2019-10-17T12:36:21 < Jan-> Looking around for options 2019-10-17T12:36:40 < Jan-> really didn't like cube 2019-10-17T12:36:55 < Jan-> platformio works but I'm stuck with arduino style coding which I'm not a huge fan of 2019-10-17T12:36:58 < zyp> nobody likes cube 2019-10-17T12:37:00 < Jan-> mbed doesn't work 2019-10-17T12:37:32 < Jan-> I can get that but I have no way of using it 2019-10-17T12:37:47 < zyp> fuck picolibc, just use the newlib bundled with the compiler 2019-10-17T12:37:59 < Jan-> yes I'd like to at least try doing that but the only way seems to be using a sort of super basic approach with manually running make and so on 2019-10-17T12:38:05 < karlp> ok, after reading the backlog, I'm again convinced jan is just trolling. 2019-10-17T12:38:10 < Jan-> which on windows is just never going to happen. 2019-10-17T12:38:12 < zyp> libc is not the problem, the problem is which HAL to pick 2019-10-17T12:38:44 < srk> Haohmaru: yep, or use virtualbox 2019-10-17T12:38:46 < zyp> isn't that what this is about? 2019-10-17T12:38:54 < srk> oops, scrolledup 2019-10-17T12:38:59 < Jan-> I guess I need *something* right? 2019-10-17T12:39:21 < zyp> nah, you could write your own register definitions from scratch 2019-10-17T12:39:23 < zyp> I did 2019-10-17T12:39:38 < Jan-> I think not zyp there's millions of them :) 2019-10-17T12:39:46 < Jan-> I don't even do that on AVR 2019-10-17T12:39:54 < karlp> Haohmaru: if you want to just "include chipheader.h, use registers in your code" that's the cmsis headers. 2019-10-17T12:40:14 < karlp> one way tot get them is from cube, and just toss all the code they generated. 2019-10-17T12:40:31 < Jan-> I guess the other approach is, what is there other than stm32 2019-10-17T12:40:41 < Jan-> because the development environment situation with stm32 seems to suuuuuck. 2019-10-17T12:41:36 < zyp> my impression is there's not really that much difference between vendors, the vendor HALs all suck 2019-10-17T12:41:40 < PaulFertser> Jan-, have you ever encountered professional equipment that doesn't require putting appropriate effort into learning it to be used effectively? 2019-10-17T12:42:36 < Jan-> I don't mind C 2019-10-17T12:42:38 < PaulFertser> Equipment that has a wide range of uses. 2019-10-17T12:42:38 < karlp> Haohmaru: no, the vendor provides the headers for the periphs too 2019-10-17T12:42:48 < Jan-> I'm not even at the point of being able to write C at this point 2019-10-17T12:42:57 < zyp> Jan-, you could try laks, I just suspect you don't have the patience to get that going either 2019-10-17T12:43:22 < zyp> Haohmaru, no. 2019-10-17T12:43:27 < Jan-> it's not really patience zyp I've just learned that this open source stuff tends to either work or not and if it doesn't work pretty quickly it's likely never going to. 2019-10-17T12:43:29 < zyp> libopencm3 or vendor headers, not both 2019-10-17T12:44:35 < PaulFertser> Jan-: you've learnt it wrong then. All programs have bugs and omissions. With "open source stuff" you can often make it work for your own usecases with reasonably humble effort if you have the skills. 2019-10-17T12:44:42 < srk> 'this open source stuff' is pretty big generalization as not every project is equal in terms of quality :) 2019-10-17T12:44:46 < zyp> Jan-, well, laks is my lib, and I can say for a fact that it works wonderfully for everything I've ever needed it to 2019-10-17T12:45:35 < PaulFertser> Haohmaru: it's just that libopencm3 ships their own headers, startup codes, everything needed. You can enchance it as you wish if support for some peripheral is missing. 2019-10-17T12:45:52 < Jan-> zyp I guess I'm at the same point though, I still need some way to actually compile code and send it to the stm32. 2019-10-17T12:45:57 < srk> my extern bool firstBoot __attribute__((section(".sram2Section"))); firstBoot = true might really be a coincidence .. 2019-10-17T12:46:42 < Jan-> there's all kinds of libraries I could in theory use 2019-10-17T12:46:58 < karlp> Haohmaru: you can get the "pack" from https://developer.arm.com/embedded/cmsis/cmsis-packs/devices/STMicroelectronics/ 2019-10-17T12:47:21 < karlp> it's just a zip file, and you can open it up and grab device/include/*.h and have al the avrlibc style headers you want. 2019-10-17T12:47:46 < Jan-> and zyp "laks" does not google very well :) 2019-10-17T12:48:05 < Jan-> are we talking about https://github.com/zyp/laks 2019-10-17T12:48:15 < karlp> no, the other zyp, with the other laks. 2019-10-17T12:48:31 < zyp> Jan-, yeah, but the github copy is probably outdated, upstream is here: https://cgit.jvnv.net/laks 2019-10-17T12:48:46 < zyp> examples are here: https://cgit.jvnv.net/laks_demo/ 2019-10-17T12:48:47 < Jan-> I guess I still have no actual way to use it. 2019-10-17T12:49:01 < karlp> Haohmaru: remind me again, you wanted to have just a header with register definitions, but you're complaining that things are in C? 2019-10-17T12:49:10 < karlp> you're trolling too :) 2019-10-17T12:49:21 < zyp> Jan-, well, you need the compiler toolchain installed first, and scons 2019-10-17T12:49:38 < zyp> and you'll want git also, for your sanity 2019-10-17T12:50:37 < karlp> zyp:nah, git is just linux neckbeard loser shit for people who can't manage backups 2019-10-17T12:50:51 < PaulFertser> Haohmaru: if you mind C you can as well take a look at stm32plus library :) 2019-10-17T12:50:52 * Jan- snaps fingers 2019-10-17T12:50:59 < Jan-> that's the word I was looking for! 2019-10-17T12:51:01 < Jan-> neckbeard! 2019-10-17T12:51:04 < Jan-> Can I be a neckbeard? 2019-10-17T12:51:23 < Jan-> I mean you know, no beard, but.. :) 2019-10-17T12:51:30 < PaulFertser> Jan-: easy, shoot some testosterone for few months 2019-10-17T12:51:39 < Jan-> owww. 2019-10-17T12:51:49 < Jan-> do I have to have the beard or can it be a notional beard 2019-10-17T12:52:02 < Jan-> Haohmaru: scarf. requires less shaving. 2019-10-17T12:52:04 < karlp> Haohmaru: those pack files from arm even have svd files, and all the docs, both the datasheets, erratas, ref mans, and the arm core guides. 2019-10-17T12:52:36 < PaulFertser> Haohmaru: actually, not too good, it tends to get covered in ice outside, then you go in and it's all wet and water drips. 2019-10-17T12:53:44 < srk> what kind of DSP? 2019-10-17T12:54:01 < karlp> (just remember, when you look at the arm startup files, that they don't _need_ to be assembly. that's just vendor carryover of their old startup from arm9 shits. you can do it all in c: https://github.com/libopencm3/libopencm3/blob/master/lib/cm3/vector.c#L63-L100 2019-10-17T12:54:16 < jly> the audio devices 2019-10-17T12:54:28 < srk> Haohmaru: well sure but it's pretty broad term, for music synthesis or just processing? 2019-10-17T12:54:54 < jly> Haohmaru: stick this on it ak4621 2019-10-17T12:55:22 < srk> had the "pleasure" with analog devices adau something DSP recently, awful 2019-10-17T12:55:57 < srk> yep 2019-10-17T12:56:07 < srk> I've started with some building blocks like waveform generators few days ago 2019-10-17T12:56:27 < srk> then got lost in dsp literature 2019-10-17T12:56:31 < jly> srk: what did you dislike about sigmaDSPs 2019-10-17T12:56:43 < srk> everything, especially windows GUI and EZUSB 2019-10-17T12:56:50 < Jan-> ok forgive the abstract question but, would it be reasonable to say that to get access to almost all of the stm32 software that exists, I'm going to have to go down the route of manually writing makefiles and figuring out gcc command lines? 2019-10-17T12:56:54 < jly> haha yeah that's.... fun 2019-10-17T12:56:55 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-17T12:57:14 < srk> like if someone did his homework he would choose another adau chip which can be loaded over network with rPi 2019-10-17T12:57:19 < srk> as it's just i2c 2019-10-17T12:57:25 < PaulFertser> Jan-: some folks here say the Keil is the best tool. Why do not you just use it? 2019-10-17T12:57:26 < jly> well yeah 2019-10-17T12:57:47 < srk> but no, first you need to figure out how to flash ez-usb to get a frikin usb->i2c "programmer" 2019-10-17T12:58:05 < Jan-> PaulFertser: because until you mentioned it there I don't think I'd ever heard of it! 2019-10-17T12:58:22 < srk> and load it every single time before actually flashing the DSP which works like 20% of time 2019-10-17T12:58:27 < zyp> keil uvision is supposedly nice, but also $$$ 2019-10-17T12:58:44 < zyp> but there's a trial version limited to 32k or so, might be enough? 2019-10-17T12:59:08 < zyp> and I think you still need a HAL 2019-10-17T12:59:36 < karlp> I suggested keil two days ago given how much you hated all this "open source bullshit" 2019-10-17T13:00:11 < Jan-> oh okay sorry my bad 2019-10-17T13:00:14 < Jan-> we talked about a lot of stuff 2019-10-17T13:00:26 < jly> yeah keil is awesome if you can pirate it or buy it 2019-10-17T13:00:29 < srk> Haohmaru: have floats and doubles, PCA8591 8bit i2c adc/dac for testing (and a builtin ones I guess), don't have the DSP skillz :D 2019-10-17T13:00:29 < Jan-> okay yeah it's $1500 for the unrestricted version I think they're not going to do that 2019-10-17T13:00:30 < PaulFertser> Regarding makefiles and IDEs, there's another option that some might like, QtCreator. 2019-10-17T13:00:41 < PaulFertser> Jan-: why do not you download and use it "illegally"? 2019-10-17T13:01:05 < Jan-> oh god that isn't even a perpetual license 2019-10-17T13:01:21 < Jan-> that's some expensive software 2019-10-17T13:01:26 < srk> Haohmaru: this is cool tho https://www.dsprelated.com/freebooks.php 2019-10-17T13:01:43 < PaulFertser> If you can't afford software, warez it. 2019-10-17T13:01:50 < zyp> Jan-, that's the price of closed source :p 2019-10-17T13:01:52 < Jan-> PaulFertser: isn't that the qt gui tool thing? 2019-10-17T13:02:12 < jly> AKM's new website is fking annoying 2019-10-17T13:02:30 < PaulFertser> Jan-: it's an IDE with debugging and all, doesn't require one to write makefiles, pretty lean, familiar for those doing Qt apps etc. Why not use it for embedded? 2019-10-17T13:02:46 < srk> I know how to IIR as well but not the theory 2019-10-17T13:03:09 < Jan-> PaulFertser: well I guess I could set up visual studio to do it if that was the approach 2019-10-17T13:03:10 < Jan-> but yikes 2019-10-17T13:03:18 < Jan-> that would require lots of very low level knowledge 2019-10-17T13:03:21 < PaulFertser> Jan-: yes, visualstudio if you like it, plus visualgdb. 2019-10-17T13:03:28 < jly> what was that piece of math to prototype IIR of continuous time filters... something bilinear blah blah blah 2019-10-17T13:03:41 < Jan-> I'm not that bothered about debugging, you don't get fancy debugging on avr 2019-10-17T13:03:47 < Jan-> I wouldn't know what to do with it 2019-10-17T13:03:48 < PaulFertser> Jan-: or is visualgdb too expensive for you too? 2019-10-17T13:04:33 < Jan-> it's not so much if it's too expensive for me, I'm not the only person involved 2019-10-17T13:05:11 < Steffanx> On fancy avrs you do get debugging... 2019-10-17T13:05:23 < Steffanx> Even on older less fancy ones 2019-10-17T13:05:24 < PaulFertser> If you do not need debugging (which is a silly idea IMHO especially if you plan to write in C) then you can use you favourite Visual Studio to build projects, then flash with the windows vendor stlink utility. 2019-10-17T13:05:30 < jly> hi steffan 2019-10-17T13:05:35 < Steffanx> Debugwire has been around for ages 2019-10-17T13:05:41 < Steffanx> Its not optimal, but it works 2019-10-17T13:05:44 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-17T13:05:56 < Steffanx> And now you have pdi/updi 2019-10-17T13:06:01 < PaulFertser> Steffanx: yep, I've seen it in action with "avr dragon" and avrstudio. It kinda works fine, just a bit slow. Surprisingly. 2019-10-17T13:06:02 < Steffanx> Hello jly. 2019-10-17T13:06:02 < Jan-> PaulFertser: I know in theory that is true but I suspect if I knew how to get visual studio to build stm32 projects, I would know enough to get gcc and make to do it. 2019-10-17T13:06:05 < srk> Haohmaru: sometimes I wish I went studying electronics instead of CS 2019-10-17T13:06:19 < Jan-> the only thing I've ever used visual studio for is c# 2019-10-17T13:06:25 < Jan-> and that comes set up for it. 2019-10-17T13:06:51 < PaulFertser> Jan-: follow some visualgdb cortex-m tutorial, just do not install visualgdb. 2019-10-17T13:07:02 < jly> I started playing assembler with 8 bit 2019-10-17T13:07:04 < Steffanx> Ohno Haohmaru 2019-10-17T13:07:11 < Jan-> avr is fun it's like writing code for a zx spectrum 2019-10-17T13:07:15 < Steffanx> And them you got into eggz 2019-10-17T13:07:16 < jly> 6502/z80 2019-10-17T13:07:20 < srk> cool 2019-10-17T13:07:35 < Jan-> I sort of hoped stm32 could be done in the same way :/ 2019-10-17T13:07:42 < day> it can? 2019-10-17T13:08:07 < srk> yeah treat it like a blackbox and only use stm32flash 2019-10-17T13:08:09 < PaulFertser> Jan-: and you already had toolchain working with platformio and with cubemxide. You're hunting ghosts actually, you already have everything you say you need. 2019-10-17T13:08:12 < jly> it was part of my college degree to play asm with stm32 2019-10-17T13:08:21 < day> yes there are more things to take care of, but in essence theres not a big difference 2019-10-17T13:08:24 < Jan-> PaulFertser: in theory I guess I do 2019-10-17T13:08:34 < jly> I remember the literal pool.... that's about it 2019-10-17T13:08:34 < Jan-> but it's all buried in the platformio setup 2019-10-17T13:08:59 < Steffanx> I like how you never changed over the years Jan- . In #avr, before ##stm32 even existed, you also had those "I dont wanna do this"-complaints :P 2019-10-17T13:09:11 < jly> at least she's not leeloominai 2019-10-17T13:09:18 < Jan-> theoretically I do have make (if I type "make" into a command prompt it says "no targets specified and no makefile found") 2019-10-17T13:09:18 < PaulFertser> It doesn't prevent you from writing the code any way you like. Same as that "Arduino IDE" btw, it also doesn't really make anybody use their weird classes to implement anything. 2019-10-17T13:09:20 < Steffanx> Leeloo didnt do that 2019-10-17T13:09:21 < jly> you have to at least give credit for that 2019-10-17T13:09:26 < Jan-> but it won't ever really work properly under windos 2019-10-17T13:09:33 < Jan-> I don't know really why they make windows ports of stuff like make 2019-10-17T13:09:40 < Jan-> maybe if you developed the whole project under windows? 2019-10-17T13:09:50 < Steffanx> Leeloo wasnt too bad. Except for the screenshotted code snippets, jly 2019-10-17T13:10:08 < jly> fucking endlessly annoying is what I remember 2019-10-17T13:10:12 < Jan-> PaulFertser: well I don't use the arduino library when I write avr code, I just use the IDE and the boards as a simple way to write and upload code. 2019-10-17T13:10:28 < jly> so I didn't mind mr d hurling abuse at he/her 2019-10-17T13:10:32 < PaulFertser> Jan-: so why do not you stop complaining and start doing the same with platformio? 2019-10-17T13:10:36 < Steffanx> Who was capacitor or crt, jly? 2019-10-17T13:10:45 < jly> some pumper 2019-10-17T13:10:50 < Jan-> Uh. Can I do that? 2019-10-17T13:11:10 < day> you can even do it without platformio 2019-10-17T13:11:11 < PaulFertser> Jan-: what makes you think you can't? 2019-10-17T13:11:16 < Jan-> welll 2019-10-17T13:11:22 < Jan-> the fact that this main.cpp doesn't even have a main() 2019-10-17T13:11:34 < Jan-> it has a setup() and a loop() 2019-10-17T13:11:37 < PaulFertser> Treat setup as your main. 2019-10-17T13:12:04 < Jan-> in arduino I just write another main() and it seems to overwrite the existing one. 2019-10-17T13:12:09 < srk> Haohmaru: how do you replicate something like tb-303? was it reverse engineered? 2019-10-17T13:12:18 < PaulFertser> Just write all of your code in setup(), what's the problem Jan- ? 2019-10-17T13:12:36 < Jan-> I guess I'd be a bit concerned that their main() was doing other freaky stuff 2019-10-17T13:12:46 < jly> this tb-303 looks pretty 2019-10-17T13:12:58 < PaulFertser> Jan-: it can't be doing it concurrently so while your setup() is working, "their" main() has no say in what's happening. 2019-10-17T13:13:17 < Jan-> I suppose I could just not #include :D 2019-10-17T13:13:29 < Jan-> but then I guess I'd be down to like, raw register access. 2019-10-17T13:13:47 < Jan-> unless their makefile includes other stuff which I guess it does? 2019-10-17T13:14:03 < srk> Haohmaru: oO, x0xb0x, seen that before 2019-10-17T13:14:09 < PaulFertser> For functions it doesn't matter what's #included, it matters what's linked. 2019-10-17T13:14:22 < Jan-> yeah see I have no way of knowing what else gets linked in. 2019-10-17T13:14:31 < PaulFertser> Jan-: just write your code in setup() 2019-10-17T13:14:40 < jly> I want to hear this tb-303 now 2019-10-17T13:14:47 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 265 seconds] 2019-10-17T13:14:53 < srk> Haohmaru: I like modular synths, started to build one recently but didn't get too far (I made some noize from attiny osc tho :)) 2019-10-17T13:15:03 < srk> s/I/it/g 2019-10-17T13:15:24 < jly> aren't modular synths usually blocks of analog circuits? 2019-10-17T13:15:27 < PaulFertser> Jan-: if their defaults are sane everything that's linked but not used is stripped down from the target binary anyway. 2019-10-17T13:15:31 < jly> (honestly I don't know) 2019-10-17T13:15:58 < Jan-> see when I run "pio run -t upload" it goes through a huge list of .o files 2019-10-17T13:16:16 < PaulFertser> Jan-: unneeded stuff is garbage collected 2019-10-17T13:16:19 < srk> you can mix analog/digital as well unless you want it pure analog for extra clock cycles :D 2019-10-17T13:16:51 < Jan-> apparently I can just write a main() 2019-10-17T13:16:56 < Jan-> it works fine 2019-10-17T13:17:02 < karlp> leeloo wass't a troll though, big difference. 2019-10-17T13:17:14 < Jan-> I thought it would scream about redefinition 2019-10-17T13:17:16 < Jan-> but it doesn't 2019-10-17T13:17:20 < srk> some stuff like oscillator sync is easier with analog than digital and there are no speed limits so you can glitch it beyond repair 2019-10-17T13:17:40 < PaulFertser> GNU ld allows redefinition by default. Sometimes leading to funny results. 2019-10-17T13:18:03 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-17T13:18:33 * PaulFertser wonders if Jan- ever read the K&R C book ... 2019-10-17T13:18:36 < karlp> Haohmaru: nearly got myself a tb303 a few years ago on a junk sale from a muso. 2019-10-17T13:18:50 < Jan-> I can deal with C 2019-10-17T13:18:50 < karlp> Haohmaru: when I picked it up and asked how much, he was like, "Oh, shit, that's not meant to be here. sorry" 2019-10-17T13:18:52 < Jan-> C is always the same 2019-10-17T13:18:58 < jly> karlp: hahaha 2019-10-17T13:19:19 < Jan-> it's just the process of getting from foobar.c to the microchip concerned that is a bit of a nightmare 2019-10-17T13:19:19 < jly> I can honestly say something similar has happened to me 2019-10-17T13:19:43 < jly> along the lines of them pulling a $5 price tag away and never returning :) 2019-10-17T13:19:44 < srk> Haohmaru: some i2s codec and DMAs 2019-10-17T13:20:01 < jly> literally hit the eject button on me when they realised 2019-10-17T13:20:12 < jly> ad1939 2019-10-17T13:20:19 < Jan-> so when this default file has "#include Arduino.h" in it, where do I actually find Arduino.h 2019-10-17T13:20:33 < jly> Haohmaru: 4in 8out I think 2019-10-17T13:20:48 < karlp> pity too, I bet it had some stories. it was this guys: https://www.youtube.com/watch?v=XGWQNQs5iOc 2019-10-17T13:21:20 < srk> jly: that's pretty cool 2019-10-17T13:21:41 < jly> good DNR compared to most of ADIs shit 2019-10-17T13:22:14 < Jan-> oh found it. 2019-10-17T13:22:18 < Jan-> how does it know where to find that... 2019-10-17T13:23:27 < srk> need to figure out something simpler like phatDAC with single i2s first 2019-10-17T13:26:16 < jly> pcm5102 looks alright 2019-10-17T13:26:40 < jly> TI probably inherited most of the IP from burr brown 2019-10-17T13:28:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-17T13:28:38 < jly> pcm1789 is one I've played with 2019-10-17T13:30:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Excess Flood] 2019-10-17T13:34:55 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-17T13:36:03 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-17T14:05:24 < Jan-> should I assume that behind the scenes, platformio is automating a bunch of stuff involving gcc and make. 2019-10-17T14:05:31 < Jan-> and maybe other stuff? 2019-10-17T14:07:19 < karlp> well, duh 2019-10-17T14:08:07 < Jan-> is that a yes or a no 2019-10-17T14:09:03 < karlp> glofo's getting out of the process race, curious to see how that pans out: https://www.elektroniknet.de/international/ceo-draws-the-line-at-12-nm-170233.html 2019-10-17T14:09:12 < Jan-> I'm not being funny but it's not like that is actually written down anywhere, there's no documentation 2019-10-17T14:09:18 < karlp> that's the platformio way. 2019-10-17T14:09:26 < karlp> "just run this platform io command, and.... magic!" 2019-10-17T14:09:52 < Jan-> it's the open source way but I digress 2019-10-17T14:10:02 < Jan-> I'm just wondering 2019-10-17T14:10:09 < Jan-> if that's the case, can I sort of reverse engineer what it is doing 2019-10-17T14:11:35 < PaulFertser> Jan-: if you need something simple and clean and you want to code in C, just use libopencm3, it has all the makefiles written for you already. 2019-10-17T14:11:43 < PaulFertser> If not, why are you asking... 2019-10-17T14:12:05 < Jan-> that's sort of the idea paul 2019-10-17T14:12:24 < Jan-> but the only way I have of compiling code for stm32 is via platformio right now and it doesn't support libopencm3 on this board. 2019-10-17T14:13:41 < Jan-> I just thought if I could go through what it's doing in detail then I could maybe persuade things to do what I want, I don't really know. 2019-10-17T14:14:03 < karlp> cm4f at 0.4v, and only 7µW/MHz.. sexy... 2019-10-17T14:16:59 < Jan-> I guess I could get another board, which might be easiest. 2019-10-17T14:19:00 < jly> i'd buy the stm32f4 discovery 2019-10-17T14:19:42 < PaulFertser> Jan-: it doesn't depend on the board. 2019-10-17T14:19:58 < karlp> Haohmaru: yarmon. (referred in the link) 2019-10-17T14:20:02 < PaulFertser> Jan-: you can just git clone libopencm3-examples and that's it, you only need the official toolchain from ARM and make. 2019-10-17T14:20:30 < karlp> (there's no premade example in that repo for their exact board, but hey...) 2019-10-17T14:20:44 < Jan-> well I can't just git clone anything right now. 2019-10-17T14:20:47 < karlp> that 2019-10-17T14:20:52 < karlp> 's because you're a troll. 2019-10-17T14:22:15 < Jan-> yes it does :/ 2019-10-17T14:22:25 < karlp> speaking of, where is senpai? 2019-10-17T14:22:53 < karlp> he'd have a good answer though, "just use fucking keil and move on with your life" 2019-10-17T14:26:02 < Jan-> OK so I got one of the command lines it is using from the --verbose output 2019-10-17T14:26:09 < Jan-> it is huge 2019-10-17T14:26:10 < Jan-> https://pastebin.com/fAyBakuu 2019-10-17T14:28:09 < Jan-> well I only actually wrote one file, main.cpp! 2019-10-17T14:28:17 < Jan-> is that all one command? yikes. 2019-10-17T14:28:48 < zyp> no, but if all you needed were main.cpp, you wouldn't be looking for libraries and shit 2019-10-17T14:29:23 < Jan-> so both no AND yes, uhhuh :) 2019-10-17T14:30:43 < Jan-> anyway that's sort of why I'm keen not to get into it at that level 2019-10-17T14:30:48 < karlp> hah, I wonder who at pio thought they needed to explicitly set max-inline-insns-single 2019-10-17T14:30:52 < karlp> madmen 2019-10-17T14:31:03 < Jan-> it's not very practical for me to figure all that out 2019-10-17T14:31:05 < Jan-> it'd take months 2019-10-17T14:32:07 < karlp> so don't. keep using platformio and stop wasting everyone's time. 2019-10-17T14:32:22 < Jan-> I guess the question is how does literally anyone ever figure that out 2019-10-17T14:32:36 < Jan-> well that's sort of my approach, figure it out 2019-10-17T14:32:42 < Jan-> but this is just, I mean, ridiculous 2019-10-17T14:32:44 < zoobab> platformio is cloud dependent? 2019-10-17T14:32:57 < oz4ga> Jan-: All that command you posted id compiling you main and include headers formstm32arduino. Platformino is just a wrapper. You could equally well just use the arduino ide 2019-10-17T14:33:00 < karlp> zoobab: no? 2019-10-17T14:33:23 < Jan-> oz4ga: sure but someone had to write it 2019-10-17T14:33:27 < oz4ga> you gain nothing using platfomino 2019-10-17T14:33:32 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-17T14:34:09 < oz4ga> worst thing about platfomino is that you have to subscribe to use the debugger :( 2019-10-17T14:34:50 < Jan-> see right now the only way I have of compiling stuff is literally to manually type that command line in. 2019-10-17T14:34:57 < Jan-> which I assume would work 2019-10-17T14:35:07 < Jan-> but it doesn't strike me as very practical 2019-10-17T14:35:21 < karlp> that's because you're willfully disregarding everythign everyone says.... 2019-10-17T14:35:51 < Jan-> which ide is that 2019-10-17T14:36:43 < Jan-> well sure but I mean, I literally have no way of doing it right now 2019-10-17T14:36:48 < Jan-> I have platformio and cube, that's it 2019-10-17T14:37:53 < Jan-> I need to go and get a computer science degree :( 2019-10-17T14:38:29 < Jan-> jesus, that's the best way? 2019-10-17T14:38:39 < Jan-> that's sort of what I'm attempting here I guess. 2019-10-17T14:38:47 < Jan-> let platformio do it, observe what it does, then modify 2019-10-17T14:39:12 < Jan-> but the complexity is just so enormous that I don't know if that's possible. 2019-10-17T14:39:27 < Jan-> I mean it's possible theoretically but yikes. 2019-10-17T14:40:18 < Jan-> that basically tends to confirm what I'm doing here 2019-10-17T14:40:19 < Jan-> I guess 2019-10-17T14:40:26 < Jan-> but I can't believe it's so crazy 2019-10-17T14:42:10 < Jan-> I'm not even at the stage of being able to do programming 2019-10-17T14:42:27 < karlp> Haohmaru: we don't hve an opinion dude, you're just the only one here using it. 2019-10-17T14:42:59 < Jan-> to be honest at this point it's not the arm chip that's the problem 2019-10-17T14:43:05 < BrainDamage> I've used c::b for a long time, but that was more than 7 years ago 2019-10-17T14:43:07 < Jan-> I'm not even using it 2019-10-17T14:43:55 < BrainDamage> some things were awkward to set, but not impossible, and no weird performance problems like some other infamous ides 2019-10-17T14:44:23 < PaulFertser> If understanding GCC options is an enourmous complexity for you, how do you plan to implement anything useful ever... 2019-10-17T14:44:42 < Jan-> C code is at least documented paul. 2019-10-17T14:44:58 < PaulFertser> Jan-: so are GCC options, in the official Info manual, all documented, clearly and sanesly. 2019-10-17T14:45:06 < Jan-> whereas this is just "oh download this stuff from github and somehow there's some sort of procedure you have to run." 2019-10-17T14:45:57 < zyp> Jan-, so do you want to skip over the details or not? 2019-10-17T14:47:29 <@englishman> zyp, these Ikea lights are super cool, what project did you want to make for them? 2019-10-17T14:47:48 < zyp> englishman, better controllers 2019-10-17T14:48:13 < oz4ga> my guess is that Jan-'s head is going to explode if we expose him to CMSIS 2019-10-17T14:48:22 -!- mode/##stm32 [-b *!*@host31-49-125-0.range31-49.btcentralplus.com] by englishman 2019-10-17T14:48:39 -!- mode/##stm32 [-b *!*@140.141.208.46.dyn.plus.net] by englishman 2019-10-17T14:48:41 <@englishman> better in what way 2019-10-17T14:49:02 < zyp> englishman, well, I assume you've used the ikea ones 2019-10-17T14:49:13 <@englishman> yes 2019-10-17T14:49:32 -!- sterna [~Adium@cvis.oal.lindholmen.se] has joined ##stm32 2019-10-17T14:49:48 < zyp> the round ones or the new square ones? 2019-10-17T14:50:12 < zyp> the new square ones fixed one thing that always annoyed me with the old ones 2019-10-17T14:50:13 <@englishman> the switches? both the round one that changes colour temp and the smaller, cheaper square ones that don't 2019-10-17T14:50:38 < zyp> the round ones only have one button that toggles, the new one have dedicated on and off buttons 2019-10-17T14:51:13 < zyp> toggling is a bit inconvenient, if you have multiple lights in a group and one happens to be off, toggling won't let you turn off the entire group 2019-10-17T14:51:21 < zyp> so that was one annoyance 2019-10-17T14:51:30 <@englishman> yes I found that immediately annoying 2019-10-17T14:51:40 <@englishman> I prefer the smaller cheaper switch by far 2019-10-17T14:52:27 < zyp> secondly, they don't have any concept of scenes, and you can't control less than a whole group at once except through the gateway 2019-10-17T14:52:49 < zyp> I have the whole kitchen/living room in one group, but I don't always want all the lights at the same level 2019-10-17T14:53:37 <@englishman> yes, I think this is not just a cost cutting measure but might not be their focus as a way to limit features and increase simplicity, scenes and stuff I've done through Google assistant 2019-10-17T14:53:53 <@englishman> which works way better than I expected 2019-10-17T14:54:18 < zyp> thirdly, related to scenes, I want to be able to push a button and not have it trigger lights directly, but send a command out through the gateway to trigger scripts on a central controller 2019-10-17T14:54:51 <@englishman> this could be done with a NFC tag and tasker 2019-10-17T14:55:01 <@englishman> or some other thing related to google and not to ikea 2019-10-17T14:55:03 < zyp> and lastly I want something that can be mounted in standard confirmation frames like regular switches 2019-10-17T14:55:23 <@englishman> that would be cool 2019-10-17T14:55:49 < zyp> well, yeah, I can get everything I want if I just choose something that don't talk to the bulbs at all, but instead talks to something that talks to the gateway 2019-10-17T14:56:04 < zyp> but I also want to be able to control the lights directly 2019-10-17T14:56:16 <@englishman> imo all of the automation functionality is through google 2019-10-17T14:56:48 < zyp> google assistant depends on the cloud, it's pretty fucking annoying to not be able to turn on and off your lights when your internet connection is down 2019-10-17T14:56:55 <@englishman> tho it could be added to the zwave part it sounds like a lot of work and would limit possible features 2019-10-17T14:58:38 <@englishman> ha yeah 2019-10-17T14:58:51 <@englishman> tho the switches cover that still 2019-10-17T14:59:08 < zyp> exactly 2019-10-17T14:59:19 <@englishman> if the internet was down, for some unbearable amount of time 2019-10-17T14:59:31 < zyp> but I want them integrated as well 2019-10-17T14:59:41 < zyp> so I might as well make something that does everything I want 2019-10-17T15:00:48 < zyp> if I only wanted them integrated, I guess I could steal the modules from the new square switches and put them in the integrated switches :p 2019-10-17T15:01:23 < zyp> but meh 2019-10-17T15:01:56 < zyp> I figure as long as I manage to talk zigbee at all, getting in all the features I want shouldn't be too hard 2019-10-17T16:18:53 < Steffanx> Scenes. Works fine with my homeassistent setup :P 2019-10-17T16:41:40 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-17T16:46:09 -!- brdb [~basdb@2601:18c:8500:7f5b::9bb] has quit [Read error: Connection reset by peer] 2019-10-17T16:51:23 -!- brdb [~basdb@2601:18c:8500:7f5b:21e:6ff:fe30:9762] has joined ##stm32 2019-10-17T16:51:48 -!- jly [uid355225@gateway/web/irccloud.com/x-adxmvxxydncrsrah] has quit [Quit: Connection closed for inactivity] 2019-10-17T16:58:43 < Jan-> what does "landed" mean 2019-10-17T16:58:53 < Jan-> as in "duplication of all the copies of the linker script templates has also been eliminated and landed." 2019-10-17T17:00:32 < oz4ga> the Eagle has landed 2019-10-17T17:01:42 < oz4ga> you should only have one linkerscript per project. It often has .ld extension 2019-10-17T17:02:10 < Jan-> I have no idea what a linker script is, I was just using that as an example 2019-10-17T17:02:53 < Jan-> I believe you. 2019-10-17T17:03:55 < BrainDamage> a linker script is a script that controls the linker 2019-10-17T17:04:15 < Jan-> that's a very microsoft answer 2019-10-17T17:04:17 < Jan-> :D 2019-10-17T17:04:23 < BrainDamage> which is the 2nd stage of a compilation process 2019-10-17T17:04:39 < BrainDamage> first the compiler takes your code for each file and compiles it to an object 2019-10-17T17:04:49 < BrainDamage> then the linker takes the object and puts them together 2019-10-17T17:04:50 < Jan-> is this where it makes all the .o files and turns them into the final binary, be that .exe or .elf or whatever 2019-10-17T17:04:56 < BrainDamage> yes 2019-10-17T17:05:21 < Jan-> I guess that's mainly a case of just joining it all together and updating any offsets in it? 2019-10-17T17:05:35 < Jan-> I mean there'll be a lot of subtleties but mainly... 2019-10-17T17:05:39 < BrainDamage> yes, however, you have a micro here 2019-10-17T17:05:45 < BrainDamage> your code is not position independent 2019-10-17T17:05:58 < BrainDamage> mostly because you don't have an os to calculate the shifts 2019-10-17T17:06:06 < Jan-> well I mean a binary loader in an OS like windows might have to go and update a bunch of offsets 2019-10-17T17:06:10 < BrainDamage> so you have to enter parameters where eg flash memory starts 2019-10-17T17:06:24 < Jan-> yeah exactly 2019-10-17T17:06:40 < Jan-> so um. 2019-10-17T17:07:39 < Jan-> right now I have to understand c, including the preprocessor and all that stuff, I have to understand platformio which is written mainly in python, also makefiles and linker scripts which are both their own private formats, and then the API itself and the microcontroller underneath it. 2019-10-17T17:08:06 * Jan- scratches head 2019-10-17T17:08:07 < Jan-> Whew? 2019-10-17T17:08:30 < BrainDamage> it's not necessary to understand linker scripts if the buildsystem you use targets your platform 2019-10-17T17:08:41 < karlp> remind me again how this is any different than avr? 2019-10-17T17:08:43 < BrainDamage> because the params are already set for you 2019-10-17T17:08:55 < karlp> it just hides it from you behind -mcpu 2019-10-17T17:09:02 < Jan-> BrainDamage: I think we're sort of establishing that there isn't a build system that targets my platform, at least not one that understands libopencm3. 2019-10-17T17:09:13 < karlp> there is, your platform ist stm32lf4 2019-10-17T17:09:23 < karlp> ht eonly difference is what pin the led is on. 2019-10-17T17:09:42 < Jan-> STM32F407VET6, apparently. 2019-10-17T17:09:51 < Jan-> I don't know how much of that matters. 2019-10-17T17:10:10 < karlp> also, the fact that you're quoting commit log entries from libopencm3 is just more evidence that you're trolling 2019-10-17T17:10:29 < Jan-> karl honestly I have no idea what you're on about seriously 2019-10-17T17:11:04 < Jan-> OK, on this board the LEDs are on the pins that the arduino framework calls PA6 and PA7 and they are active low 2019-10-17T17:11:41 < Jan-> it's a framework for stm32 that sort of makes it work like arduino 2019-10-17T17:11:50 < Jan-> so you can just digitalWrite(PA6, HIGH); etc 2019-10-17T17:12:00 < Jan-> like you my reaction was "ehh do I really want that" 2019-10-17T17:12:25 < Jan-> port A pin 6 and port A pin 7. 2019-10-17T17:12:27 < Jan-> we know that. 2019-10-17T17:13:00 < Jan-> I take your point but this code works: https://pastebin.com/1Y9bWr0M 2019-10-17T17:13:25 < Jan-> we did have to look up with a meter where they were connected. 2019-10-17T17:13:31 < Jan-> but that is where they are connected, that at least we do know. 2019-10-17T17:14:38 < Jan-> I thought that :) 2019-10-17T17:14:53 < karlp> no, because arduino uses numeric pins, not PxN 2019-10-17T17:15:02 < karlp> and 13.set_mode(OUTPUT) wouldn't have worked :) 2019-10-17T17:15:03 < BrainDamage> OOP adds [some] overhead fwiw 2019-10-17T17:15:08 < Jan-> I am told that the arduino implementation for STM32 is a lot better than the one for AVR. 2019-10-17T17:15:20 < Jan-> But I don't know. 2019-10-17T17:15:39 < BrainDamage> Haohmaru: indirect calls and memory overhead 2019-10-17T17:15:45 < Jan-> well it must do, it has to do lookups at some point 2019-10-17T17:15:48 -!- fenugrec_ [~fenugrec@24.105.71.66] has quit [Remote host closed the connection] 2019-10-17T17:15:59 < BrainDamage> so to do the same things, you add few more function calls and more memory 2019-10-17T17:16:03 < BrainDamage> it may matter, or may not 2019-10-17T17:16:11 < BrainDamage> depends how tight your micro budget is 2019-10-17T17:16:12 < Jan-> it's true you do, I tried it 2019-10-17T17:16:34 < BrainDamage> normally it doesn't 2019-10-17T17:16:36 < Jan-> Haohmaru the problem is that on an AVR, a toggle is a read-modify-write. 2019-10-17T17:16:55 < Jan-> I think (confirm this someone?) that an stm32 has hardware for toggling. 2019-10-17T17:17:01 < Jan-> but an AVR does not. 2019-10-17T17:17:20 < Jan-> It depends what you're doing Haohmaru. 2019-10-17T17:17:37 < Jan-> If you end up writing something non-oop that sort of reimplements oop, then probably doing it in oop in the first place doesn't cost you anything. 2019-10-17T17:17:48 < Jan-> and that could just be an array of structs. 2019-10-17T17:19:36 < Jan-> bear in mind if it comes down to that, then the compiler will optimise it to that regardless of how oop the language is. 2019-10-17T17:19:48 < Jan-> but it would depend very much exactly what you asked it to do 2019-10-17T17:21:32 < BrainDamage> templates are resolved by the compiler 2019-10-17T17:21:41 < BrainDamage> so it can optimize it to memory write 2019-10-17T17:21:52 < BrainDamage> while oop can be instantiated at runtime 2019-10-17T17:22:05 < Jan-> I just thought if you only ever created one of an object it might pretty much just get compiled down to something very direct. 2019-10-17T17:22:09 < Jan-> It would depend on the specifics. 2019-10-17T17:22:18 < BrainDamage> unless you make your classes static, you cannot skip the function call 2019-10-17T17:23:22 < BrainDamage> sure, just not through oop 2019-10-17T17:23:23 < Jan-> Haohmaru: like BrainDamage said, abuse "static" :D 2019-10-17T17:25:48 < Jan-> my main background in this is in writing c# and javascript where we are usually on platforms that don't care so much about a few k of ram either way 2019-10-17T17:26:57 < Jan-> I guess... is there an IDE (ideally free or cheap) that comes set up with libopencm3 2019-10-17T17:27:39 < Jan-> also I'm not wedded to libopencm3 2019-10-17T17:27:53 < Jan-> I only know about three. The one that cube comes with (which is ick), arduino and libopencm3. 2019-10-17T17:29:57 < mawk> just use a text editor Jan- 2019-10-17T17:30:03 < mawk> what do you need an IDE for anyway 2019-10-17T17:30:07 < mawk> real pros use emacs 2019-10-17T17:30:28 < mawk> a text editor, a terminal tab for building and flashing, a terminal tab for debugging 2019-10-17T17:30:32 < Jan-> mawk I'm totally happy to use notepad++ 2019-10-17T17:30:36 < mawk> good 2019-10-17T17:30:47 < Jan-> in an ideal world I'd do it in visual studio and have intellisense which is awesome 2019-10-17T17:30:56 < Jan-> but notepad++ is better than the arduino editor 2019-10-17T17:31:16 < mawk> if you mess 10 minutes with your IDE you can even add a macro to flash with openocd 2019-10-17T17:31:19 < Jan-> but then I'm back to okay I've written my main.cpp, what now 2019-10-17T17:31:32 < mawk> add a terminal view for the serial output of the mcu, etc 2019-10-17T17:31:35 < Jan-> at some point I have to type gcc [something] 2019-10-17T17:31:38 < mawk> you don't need a gas factory 2019-10-17T17:31:47 < mawk> at least write a makefile lol 2019-10-17T17:31:53 < mawk> you need a bunch of arguments to gcc to make it right 2019-10-17T17:31:58 < Jan-> exactly 2019-10-17T17:32:05 < Jan-> I don't even know where to start 2019-10-17T17:32:13 < BrainDamage> that's where examples help 2019-10-17T17:32:14 < mawk> you can take a look in the makefile CubeMX generates 2019-10-17T17:32:27 < Jan-> I was trying to look at what platformio does as that actually seems to work 2019-10-17T17:32:29 < mawk> you don't need much, you're creating a regular binary after all, nothing fancy 2019-10-17T17:34:30 < Jan-> there's only two files called "Makefile" in platformio 2019-10-17T17:34:33 < mawk> for instance for a cortex M3 with no FPU you compile like this: gcc -g -mcpu=cortex-m3 -mthumb source.c -o executable -lc -lm -lnosys -specs=nano.specs -T script.ld 2019-10-17T17:34:46 < mawk> I just copy pasted this from CubeMX makefile 2019-10-17T17:35:15 < mawk> it's the general spirit 2019-10-17T17:35:33 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-17T17:35:57 < Jan-> the concern I have is that usually make just doesn't work on windows 2019-10-17T17:36:01 < mawk> but anyway you have example makefiles in libopencm3 2019-10-17T17:36:04 < mawk> why ? 2019-10-17T17:36:09 < mawk> make is a simple program 2019-10-17T17:36:18 < Jan-> usually the makefiles are set up to expect linuxy folder structures 2019-10-17T17:36:26 < Jan-> and it just explodes 2019-10-17T17:36:27 < kakimir64> Jan-: use linux 2019-10-17T17:36:29 < mawk> lol 2019-10-17T17:36:43 < mawk> if you don't use absolute paths it'll be fine 2019-10-17T17:37:03 < kakimir64> also if you use some ide they have some make system that you can adjust through graphical user interface 2019-10-17T17:37:53 < Jan-> so there's a makefile in .platformio\packages\framework-libopencm3\lib\stm32\f4 2019-10-17T17:37:57 < kakimir64> mawk: how about / \ 2019-10-17T17:38:13 < Jan-> kakimir64: windows understands slashes that go either way usually. 2019-10-17T17:38:19 < mawk> kakimir64: windows will translate / into \ don't worry 2019-10-17T17:38:32 < kakimir64> how nice 2019-10-17T17:39:26 < Jan-> so let's say for example I just wanted to do the build that platformio does 2019-10-17T17:39:29 < Jan-> just for the sake of argument 2019-10-17T17:39:52 < Jan-> there's nothing called "Makefile" in my project folder or any subfolder 2019-10-17T17:40:29 < kakimir64> yes 2019-10-17T17:40:29 < mawk> you can use a makefile that's in another directory 2019-10-17T17:40:41 < mawk> make -f path/to/Makefile ... 2019-10-17T17:40:49 < Jan-> the verbose output doesn't specify where it is getting the makefile 2019-10-17T17:41:06 < mawk> you found it already, you said in .platformio\packages\framework-libopencm3\lib\stm32\f4 2019-10-17T17:41:12 < mawk> ah but platformio is using libopencm3 ? 2019-10-17T17:41:16 < mawk> I don't get it 2019-10-17T17:41:17 < Jan-> but that's libopencm3 2019-10-17T17:41:21 < kakimir64> the ide invokes required programs by project settings 2019-10-17T17:41:22 < Jan-> I'm using arduino apparently 2019-10-17T17:41:26 < mawk> it's in a directory called .platformio 2019-10-17T17:41:27 < mawk> ah 2019-10-17T17:41:53 < mawk> I don't think it's a useful exercise to try to replicate the build process of platformio 2019-10-17T17:42:06 < mawk> if you want to do something similar download libopencm3 and use the packaged examples with their own makefiles 2019-10-17T17:42:09 < Jan-> it's just that it's a known working example 2019-10-17T17:42:14 < mawk> libopencm3 works too 2019-10-17T17:42:20 < mawk> it even supports custom boards 2019-10-17T17:42:26 < mawk> like "bluepill" 2019-10-17T17:42:42 < Jan-> there is a "Makefile" in .platformio\packages\tool-stm32duino\src\stm32flash_serial\src 2019-10-17T17:42:52 < mawk> I don't think that's it 2019-10-17T17:42:59 < mawk> this looks like the makefile to build the tool to flash over UART 2019-10-17T17:43:07 < Jan-> What about .platformio\packages\tool-stm32duino\src\stm32flash_serial\src\parsers 2019-10-17T17:43:11 < BrainDamage> Jan-: you should also try to be less opinionated if you're not informed on the subject, like the past chat demonstrated 2019-10-17T17:44:06 < Jan-> BrainDamage I don't want an argument with anyone. 2019-10-17T17:44:51 < Jan-> I wonder if it is creating the makefile on the fly, there are python source files called "makefile.py" etc. 2019-10-17T17:45:24 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-17T17:46:08 < mawk> maybe 2019-10-17T17:46:18 < Jan-> OK so I got the libopencm3-master from github. 2019-10-17T17:46:35 < Jan-> I guess like 2019-10-17T17:46:44 < Jan-> you said there were examples? 2019-10-17T17:46:46 < mawk> https://github.com/libopencm3/libopencm3/archive/master.zip 2019-10-17T17:47:02 < Jan-> Yes I have a folder called libopencm3-master 2019-10-17T17:47:05 < Jan-> unpacked from the zip 2019-10-17T17:47:33 < mawk> https://github.com/libopencm3/libopencm3-miniblink 2019-10-17T17:47:46 < mawk> you have this as a blink project, here's the zip: https://github.com/libopencm3/libopencm3-miniblink/archive/master.zip 2019-10-17T17:48:22 < Jan-> I put that alongside the libopencm3-master folder. OK. 2019-10-17T17:48:44 < mawk> now, you have a custom board so I don't know if it's referenced or not 2019-10-17T17:48:59 < Jan-> I don't really know what that means 2019-10-17T17:49:27 < mawk> so you must understand that it's not libopencm3's or anyone's fault that there is maybe no blink project available, because the project can't possibly guess on which pin the board designer put the LEDs 2019-10-17T17:49:30 < Jan-> in "boards stm32.mk" there is reference to STM32F4 2019-10-17T17:49:39 < mawk> yeah, but stm32f4 is the bare chip, the LED can be anywhere 2019-10-17T17:49:40 < kakimir64> https://www.youtube.com/watch?v=01y6bR6ETpA 2019-10-17T17:49:48 < Jan-> I get that I was expecting to have to change the C source 2019-10-17T17:49:49 < mawk> so I expect you have to modify a file or two to indicate where is the LED, but that's all 2019-10-17T17:49:53 < Jan-> sure 2019-10-17T17:50:11 < Jan-> there is no main.c here? 2019-10-17T17:50:50 < Jan-> so like template_stm32.c I guess 2019-10-17T17:50:57 < day> mawk: "you must understand..." haha 2019-10-17T17:51:06 < mawk> just edit boards.stm32.mk Jan- , and add a definition for your board if it's not already present 2019-10-17T17:51:10 < mawk> does your board have a name ? 2019-10-17T17:51:23 < mawk> maybe it's already present 2019-10-17T17:51:29 < kakimir64> main.c is not required name Jan- 2019-10-17T17:51:43 < Jan-> most places call it "Black STM32F407" or "Black STM32F407VET6" 2019-10-17T17:51:49 < mawk> ah 2019-10-17T17:51:52 < mawk> so it's already present ! perfect 2019-10-17T17:51:54 < mawk> just type make 2019-10-17T17:52:05 < mawk> "black-stm32f407ve-v2.0" is already defined in the makefile 2019-10-17T17:52:09 < Jan-> like a lot of this chinese stuff it's sort of unclear what it really is called or who makes it, it says "STM32_F4VE" on the back 2019-10-17T17:52:29 < Jan-> how will it know which to use 2019-10-17T17:52:35 < mawk> it builds everything 2019-10-17T17:52:46 < mawk> then you select the right elfr 2019-10-17T17:52:49 < mawk> elf * 2019-10-17T17:53:22 < Jan-> nrrrk https://pastebin.com/i6E6kJR9 2019-10-17T17:53:54 < mawk> install git 2019-10-17T17:53:59 < Jan-> no. 2019-10-17T17:54:01 < Jan-> :) 2019-10-17T17:54:07 < mawk> then you can't make 2019-10-17T17:54:17 < Jan-> crap :( 2019-10-17T17:54:29 < Jan-> ohKAY *heavy sigh* where do I get it 2019-10-17T17:54:40 < kakimir64> from google 2019-10-17T17:54:42 < mawk> there's a windows installer on their website, just make sure to tick "install to path" 2019-10-17T17:55:11 < Jan-> what like: https://github.com/git-for-windows 2019-10-17T17:55:16 < mawk> https://github.com/git-for-windows/git/releases/download/v2.23.0.windows.1/Git-2.23.0-64-bit.exe 2019-10-17T17:55:18 < mawk> this 2019-10-17T17:55:18 < Jan-> there's like 400 things 2019-10-17T17:55:26 < Jan-> aaa why don't they put that 2019-10-17T17:55:36 < mawk> I got it from the website, not the github: https://git-scm.com/download/win 2019-10-17T17:55:39 < Jan-> by the way 2019-10-17T17:55:44 * Jan- hands Mawk a beer 2019-10-17T17:55:51 < mawk> lol 2019-10-17T17:55:56 < Jan-> and a cookie 2019-10-17T17:55:58 < Jan-> and cake 2019-10-17T17:55:59 < mawk> thanks 2019-10-17T17:56:14 < Jan-> you're the only person who's actually gone "ok, bunky, put this folder here and type this" 2019-10-17T17:56:37 < mawk> yeah well I saw you code already so I know it's not for nothing 2019-10-17T17:56:41 < Jan-> it's times like this we need my as yet unrealised website idea, it's called interbeer 2019-10-17T17:56:48 < Jan-> like interflora but for beer 2019-10-17T17:56:54 < Jan-> more appropriate for neckbeard world. 2019-10-17T17:57:36 < Steffanx> All he need is stroopwafels 2019-10-17T17:58:05 < Jan-> stroop waffles are AWESOME 2019-10-17T17:58:14 < Jan-> I was in amsterdam in early september and I had loads. 2019-10-17T17:58:37 < mawk> on the delft market I got stroopwafel crumbs for cheaper, but it was a scam there was no stroop in them 2019-10-17T17:58:44 < Steffanx> Haha 2019-10-17T17:58:56 < Jan-> you can get them here but I have to go to a special shop 2019-10-17T17:59:00 < Jan-> which is good 2019-10-17T17:59:07 < Jan-> otherwise I would be... neckbeard shaped :/ 2019-10-17T17:59:25 < Jan-> ok so I now have git but "make" gives the same error and if I just type "git" it has never heard of it 2019-10-17T18:00:08 < Steffanx> Did add git to your path? Did you restart your command prompt? 2019-10-17T18:02:38 < Jan-> Um 2019-10-17T18:02:38 < Jan-> Er 2019-10-17T18:02:44 < mawk> restart the prompt yes 2019-10-17T18:02:59 < Jan-> so I have C:\Program Files\Git\cmd\git.exe 2019-10-17T18:03:51 < Jan-> now it says fatal: not a git repository (or any of the parent directories): .git 2019-10-17T18:04:46 < PaulFertser> What's your favourite VCS Jan- ? 2019-10-17T18:05:04 * karlp laughs 2019-10-17T18:05:17 < Jan-> as far as I know a VCS is an Atari games machine 2019-10-17T18:05:20 < PaulFertser> (the error is due to downloading an archive instead of doing a regular git clone) 2019-10-17T18:05:48 < Jan-> *sigh* only linus torvalds could come up with a special linuxified way of downloading something 2019-10-17T18:06:10 < PaulFertser> Jan-: so what's your non-special way? What's your favourite? 2019-10-17T18:06:22 < Jan-> Er, like, just... downloading it? 2019-10-17T18:06:24 < Jan-> Like I did? 2019-10-17T18:06:32 < PaulFertser> Visual SourceSafe? 2019-10-17T18:07:17 < BrainDamage> remember when I said to not be opinionated on certain subjects? 2019-10-17T18:07:19 < Jan-> Never used it. 2019-10-17T18:07:31 < BrainDamage> VCS stands for version control system 2019-10-17T18:07:44 < Jan-> last time I came across that everyone was using svn 2019-10-17T18:07:45 < BrainDamage> it doesn't store just your current version, it has all the history 2019-10-17T18:07:59 < BrainDamage> yes, svn also has its own transfer method 2019-10-17T18:08:04 < BrainDamage> so you cannot pin it on git 2019-10-17T18:08:08 < BrainDamage> in fact all of them do 2019-10-17T18:08:14 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-17T18:08:36 < Jan-> I don't know about that, if you want to import a library into a visual studio project you just import the DLL 2019-10-17T18:08:47 < Jan-> doesn't matter where it came from it's just a file 2019-10-17T18:08:55 < BrainDamage> that's a compiled library 2019-10-17T18:09:06 < BrainDamage> here we're talking about source code 2019-10-17T18:09:08 < kakimir64> is +20kilos acceptable size for newlib nano? 2019-10-17T18:09:19 < kakimir64> in final image 2019-10-17T18:09:26 < BrainDamage> and having all the history of a project is a good thing 2019-10-17T18:09:26 < PaulFertser> kakimir64: is that using floating point printf? 2019-10-17T18:09:32 < BrainDamage> you can chose any version you like 2019-10-17T18:09:32 < kakimir64> no 2019-10-17T18:09:40 < kakimir64> it shouldn't be 2019-10-17T18:09:46 < PaulFertser> kakimir64: you can check the map file to see what got included 2019-10-17T18:09:52 < Jan-> I'd like to choose whatever version it is I have here 2019-10-17T18:10:06 < BrainDamage> you can with a simple command 2019-10-17T18:10:32 < BrainDamage> also, git has a good gui client on windows, iirc tortoisegit 2019-10-17T18:10:47 < Jan-> am I right in thinking that I have all the code, the library underlying it, I have make, I have gcc, and all the stuff I'd need. 2019-10-17T18:10:55 <@englishman> linker scripts were invented by RMS to ensure opensores toolchains were convoluted enough to prevent anyone from using them in real projects 2019-10-17T18:10:58 < Jan-> But make won't make it because git doesn't think it's been downloaded in the right way? 2019-10-17T18:11:07 < BrainDamage> no 2019-10-17T18:11:11 < Jan-> englishman: I was beginning to draw that conclusion. 2019-10-17T18:11:13 < BrainDamage> git is just a transfer method 2019-10-17T18:11:23 < BrainDamage> if you have the code, then make will work 2019-10-17T18:11:49 < PaulFertser> BrainDamage: in fact, I guess this time Jan- describes it properly, it can happen if something in Makefile is calling git to embed the version or something like that. 2019-10-17T18:12:09 < BrainDamage> that'd be a horrible build system 2019-10-17T18:12:17 < Jan-> well the code is right here, I have it in an editor, but make says "initializing libopencm3 module" then "fatal: not a git repository (or any of the parent directories): .git" 2019-10-17T18:12:23 <@englishman> isn't that how node.js works 2019-10-17T18:12:50 < BrainDamage> ok, then it does what PaulFertser said 2019-10-17T18:12:53 <@englishman> it's certainly how go works 2019-10-17T18:12:59 < Jan-> where should I be when I type make 2019-10-17T18:13:05 < PaulFertser> BrainDamage: at least in https://pastebin.com/i6E6kJR9 paste it's obvious that libopencm3 git submodule wasn't downloaded, which make sense. 2019-10-17T18:13:08 < BrainDamage> the base project directory 2019-10-17T18:13:10 < Jan-> in libopencm3-miniblink-master? 2019-10-17T18:13:20 < Jan-> or libopencm3-master 2019-10-17T18:13:30 < mawk> in miniblink 2019-10-17T18:13:31 < mawk> to start with 2019-10-17T18:13:34 < BrainDamage> miniblink 2019-10-17T18:13:35 < mawk> it will build all examples 2019-10-17T18:13:42 < PaulFertser> Jan-: libopencm3-master should be named "libopencm3" and located in the root directory of libopencm3-examples 2019-10-17T18:13:58 < mawk> but the make already does that PaulFertser no ? 2019-10-17T18:13:59 < mawk> as submodule 2019-10-17T18:14:05 < mawk> https://github.com/libopencm3/libopencm3-miniblink/blob/master/.gitmodules 2019-10-17T18:14:07 < PaulFertser> Or just clone libopencm3-examples properly dammit Jan- 2019-10-17T18:14:15 < Jan-> I just get https://pastebin.com/WnAb23KW 2019-10-17T18:14:23 < BrainDamage> Jan-: you only downloaded the example without the library 2019-10-17T18:14:25 < mawk> ah right 2019-10-17T18:14:27 < PaulFertser> mawk: it should if libopencm3-examples is a git repo with .gitmodules file etc, then make will do the right thing. 2019-10-17T18:14:30 < Jan-> I have the library 2019-10-17T18:14:35 < BrainDamage> hence the buildsystem is trying to be smart and download the missing lib for you 2019-10-17T18:14:37 < BrainDamage> and failing 2019-10-17T18:14:45 < mawk> delete the miniblink directory, and type git clone https://github.com/libopencm3/libopencm3-miniblink.git 2019-10-17T18:14:48 < Jan-> I did ask where I should put it 2019-10-17T18:14:50 < BrainDamage> Jan-: not where the buildsystem expects it 2019-10-17T18:15:07 < Jan-> ...I did ask 2019-10-17T18:15:17 < karlp> PaulFertser: does git clone do recursive by default these days? it didn't use to 2019-10-17T18:15:25 < BrainDamage> it doesn't 2019-10-17T18:15:41 < karlp> anyway, check out my freertos profiling: https://zerobin.net/?deaeb0a0792aa86d#yakQJdcrPSS8xKWw4/rEV4OjYwoZ8+0xERRPPQqzW1w= 2019-10-17T18:15:55 < karlp> not exactly tracelayzer, but.... 2019-10-17T18:16:27 < Jan-> now it says "cloning into '/libopencm3-miniblink.git'... remote: Not Found fatal: repository 'https://github.com/libopencm3' not found 2019-10-17T18:16:35 < BrainDamage> Jan-: see PaulFertser's note 2019-10-17T18:16:39 < BrainDamage> where to put it 2019-10-17T18:17:12 -!- sterna [~Adium@cvis.oal.lindholmen.se] has quit [Quit: Leaving.] 2019-10-17T18:17:22 < karlp> englishman: are linker scripts any different than keil scatter files? really? 2019-10-17T18:17:28 < BrainDamage> also, when asking for help you should put on a pastebin both your command and the whole output 2019-10-17T18:17:47 < BrainDamage> this saves headaches on both sides 2019-10-17T18:17:57 < mawk> I don't know what you typed exactly Jan- 2019-10-17T18:18:05 < mawk> but apparently a strange thing 2019-10-17T18:18:15 < mawk> copy-paste the following: git clone https://github.com/libopencm3/libopencm3-miniblink.git 2019-10-17T18:18:36 < Jan-> oh ok what's happening is that it gets a space in it when I copy it out of the irc iwndow 2019-10-17T18:18:45 < mawk> ah, I see 2019-10-17T18:19:24 < Jan-> holy shitcakes batman, make is doing something useful 2019-10-17T18:19:31 * Jan- runs around in circles, waving her arms 2019-10-17T18:20:06 < Jan-> ok so it's going through like every model of microcontroller ever made by ST. 2019-10-17T18:20:19 < Jan-> Can we maybe ask it not to do that 2019-10-17T18:20:40 < karlp> no. it's to get you a basic point and it's fast. 2019-10-17T18:20:42 < BrainDamage> you can ask for a specific target, but in this case it's better, less things to go wrong for you 2019-10-17T18:20:48 < karlp> when you're ready yo can chop things out. 2019-10-17T18:21:04 < Jan-> it's not that fast for a blink program it's still cranking through it :D 2019-10-17T18:21:10 < karlp> "make -j5" if you're on a half decent computer. 2019-10-17T18:21:22 < karlp> it will take a bit the first time as it builds the library 2019-10-17T18:21:42 < BrainDamage> control-c to stop current build, then do what karlp said for multithreaded compile 2019-10-17T18:21:59 < karlp> or just wait 2019-10-17T18:22:51 < Jan-> "Intel (R) Xeon(TM) CPU X5680 @ 3.33GHz" 2019-10-17T18:22:55 < Jan-> half decent? 2019-10-17T18:23:04 < Jan-> it's oldish 2019-10-17T18:23:22 < BrainDamage> oh and one thing, NEVER EVER omit a number for -j in make, unless you like to kill your OS 2019-10-17T18:23:32 < BrainDamage> it has this mega idiotic behaviour 2019-10-17T18:23:47 < BrainDamage> where it keeps spawning processes as much as it can 2019-10-17T18:23:51 < BrainDamage> until your os dies 2019-10-17T18:24:44 < karlp> yeah, the idioicy of that not defaulting to NPROC is just bullet to the head batshit 2019-10-17T18:26:53 < Jan-> I thought that open source was awesome! 2019-10-17T18:26:54 * Jan- ducks 2019-10-17T18:27:21 < BrainDamage> idiotic behaviour in programs is irrelevant to being open source or not 2019-10-17T18:27:23 < Jan-> it's still going! 2019-10-17T18:27:40 < karlp> yeah, well, I just added another target this morning, sorry :) 2019-10-17T18:27:43 * Jan- decides now is the time for a coffee 2019-10-17T18:28:05 < karlp> though, miniblink won't have pulled that in yet. 2019-10-17T18:28:10 < Jan-> it keeps saying BUILD /lib/stm32/f0 2019-10-17T18:28:19 < Jan-> shouldn't that be changing 2019-10-17T18:29:00 < karlp> it's just completely stuck on that line? 2019-10-17T18:30:06 < mawk> you have a 1990 processor Jan- you lied to us 2019-10-17T18:30:13 < Jan-> it's sorta hard to tell: https://pastebin.com/YKmrsWbd 2019-10-17T18:30:37 < mawk> it looks like it's going like it should 2019-10-17T18:30:55 < mawk> or maybe not 2019-10-17T18:31:04 < mawk> looks like it's looping 2019-10-17T18:31:10 < Jan-> I can't quite tell if it's just repeating the same thing 2019-10-17T18:31:25 < karlp> that looks kinda busted. 2019-10-17T18:31:27 < mawk> yeah it's repeating 2019-10-17T18:31:38 < karlp> you're on windows? 2019-10-17T18:31:53 < Jan-> uhhuh 2019-10-17T18:32:00 < karlp> it's meant to work there, can't say I've tried in a while, 2019-10-17T18:32:09 < karlp> we got some patches to make the build work from windows users. so *shrugs* 2019-10-17T18:33:05 < Jan-> so I shouldn't keep waiting 2019-10-17T18:33:09 < mawk> indeed 2019-10-17T18:33:27 < mawk> I'm sure WSL's make and git would work fine and have no chance of failing like that 2019-10-17T18:33:45 < mawk> it's a single box to tick in windows settings, then you download Debian from the windows store 2019-10-17T18:33:48 < Ecco> Hmm 2019-10-17T18:34:02 < Jan-> well if it is any consolation you have joined a long line of open source code that doesn't work very reliably :) 2019-10-17T18:34:03 < mawk> and then you can't ever say again "I'm on windows it doesn't work" Jan- because you'll have a nice linux system on hand 2019-10-17T18:34:09 < Ecco> I'm using some C code that pulls emutls_alloc from libgcc 2019-10-17T18:34:17 < Ecco> how can I know which code eventually calls emults_alloc? 2019-10-17T18:34:20 < mawk> "open source", "windows" the problem is here Jan- 2019-10-17T18:34:35 < karlp> are you in an msys2 shell or what anway? 2019-10-17T18:34:40 < Jan-> I'm in cmd.exe 2019-10-17T18:34:47 < karlp> I've got ~zero interest in trying to get it work in cmd.exe, but "patches welcome" 2019-10-17T18:34:59 < Jan-> anyway fine OK we found something that doesn't work that's still information 2019-10-17T18:35:05 < karlp> but hey, nice new failure mode. 2019-10-17T18:36:17 < Jan-> I've got cygwin somewhere 2019-10-17T18:36:26 < mawk> WSL ! 2019-10-17T18:36:49 < mawk> it will be useful for the rest of your life 2019-10-17T18:37:18 < Jan-> that would require windows 10 I think 2019-10-17T18:37:26 < mawk> yes 2019-10-17T18:37:41 < mawk> too bad 2019-10-17T18:38:04 < Jan-> on top of which that's basically "use linux" 2019-10-17T18:38:08 < Jan-> which I am desperate to avoid 2019-10-17T18:38:20 < Jan-> If I wanted to do that we have a laptop that has ubuntu on it for real emergencies 2019-10-17T18:38:27 < Jan-> but that is ONLY for real emergencies 2019-10-17T18:38:30 < karlp> so, Keil it is. 2019-10-17T18:38:36 < Jan-> Euuugh 2019-10-17T18:38:43 < Jan-> I'd use the arduino thing before I did that 2019-10-17T18:39:25 < Jan-> I can try mingw if you think it'll help 2019-10-17T18:40:44 < mawk> yeah try something that looks the most like a regular make 2019-10-17T18:41:02 < mawk> but cygwin or msys looked like a good choiec 2019-10-17T18:41:15 < mawk> I don't think you should try standalone mingw 2019-10-17T18:42:28 < BrainDamage> linux stuff expects a certain toolset to be available, this toolset is not installed on windows by default 2019-10-17T18:42:47 < BrainDamage> installing the toolset one by one will be an exercise in frustration 2019-10-17T18:42:51 < mawk> yes that's the reason 2019-10-17T18:43:12 < BrainDamage> hence things like wsl, mingw and cygwin, etc exist 2019-10-17T18:43:20 < BrainDamage> err, mysys 2019-10-17T18:52:19 < Jan-> so why did it fail 2019-10-17T18:52:27 < Jan-> I mean, it wasn't putting out error messages 2019-10-17T18:52:35 < Jan-> just doing the same thing again and again 2019-10-17T18:52:35 < karlp> dunnot, and unless you're going to patch it, it's not interesting either. 2019-10-17T18:53:59 < mawk> because some tool was missing, misbehaving, whatever, Jan- 2019-10-17T18:54:20 < Jan-> well, the mingw installer claims mingw is installed but it isn't 2019-10-17T18:54:54 < karlp> probably something here: https://github.com/libopencm3/libopencm3/blob/master/Makefile#L50 but again, unless you want to work on that, not exciting. 2019-10-17T18:55:56 < mawk> but why did you pick mingw Jan- ? 2019-10-17T18:56:08 < mawk> you have distributions like MSYS or cygwin that include everything 2019-10-17T18:57:13 < mawk> but you can try like that if you want, just make sure it's in your PATH then restart your prompt 2019-10-17T18:58:15 < Jan-> I have it, it's in c:\mingw 2019-10-17T18:58:26 < Jan-> I've always had it apparently 2019-10-17T18:58:28 < Jan-> hasn't helped 2019-10-17T19:00:08 < Jan-> if I run msys.bat I get basically a command prompt that says JanH@CobaltBox ~$ 2019-10-17T19:00:45 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-17T19:01:54 < BrainDamage> then try to use that, your mileage may vary 2019-10-17T19:02:01 < BrainDamage> cd into the dir, and run make 2019-10-17T19:02:16 < Jan-> I'm not sure i can even do that, does it understand what a g: drive is 2019-10-17T19:02:25 < BrainDamage> it uses linux paths 2019-10-17T19:02:33 < BrainDamage> so it should be /g/ 2019-10-17T19:03:16 < Jan-> ok here we go 2019-10-17T19:04:10 < Jan-> https://pastebin.com/KwMNJBhw 2019-10-17T19:05:50 < Jan-> nopd 2019-10-17T19:07:30 < karlp> well, you can fix that one right? 2019-10-17T19:08:06 < karlp> in other news, hammering my modbus task makes the task usage of the modbus task go _down_ not up :) 2019-10-17T19:11:33 < BrainDamage> locked in wait? 2019-10-17T19:12:22 < karlp> well, the modbus task was blocking on serial receive yes. 2019-10-17T19:12:48 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds] 2019-10-17T19:13:01 < karlp> I still think the instrumentation should handle this better though... 2019-10-17T19:17:09 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-17T19:17:28 < karlp> hah no, I had task ids wrong. 2019-10-17T19:21:42 < Jan-> er, can I fix that one? 2019-10-17T19:22:04 < mawk> put arm gcc toolchain in PATH 2019-10-17T19:22:07 < mawk> then you're good to go 2019-10-17T19:22:17 < Jan-> it kept doing the same thing a lot and ended with "Failure building" and a list of folder names 2019-10-17T19:29:48 < kakimir64> any magic command to quickly see why my bluetooth audio keeps disconnecting from my linux box? 2019-10-17T19:30:46 < kakimir64> only problems with linux box 2019-10-17T19:32:04 < kakimir64> 5minutes musics and sound mutes and status light starts blinking in the audiobox 2019-10-17T19:39:26 < kakimir64> PaulFertser: https://paste.ee/p/22QBd interesting how "nano" is omited from object file name when I go with full std options 2019-10-17T19:42:44 < Jan-> so I installed "gcc-arm-none-eabi-8-2019-q3-update-win32" 2019-10-17T19:42:53 < Jan-> and it fails in the same way 2019-10-17T19:43:19 < karlp> whatever "install" means then, it didn't put it on your path. 2019-10-17T19:43:42 < Jan-> I ran "gcc-arm-none-eabi-8-2019-q3-update-win32.exe" 2019-10-17T19:43:52 < kakimir64> iirc it doesn't even have option to path it 2019-10-17T19:44:27 < kakimir64> the installer 2019-10-17T19:44:34 < Jan-> arm-none-eabi-gcc.exe is in C:\Program Files (x86)\GNU Tools ARM Embedded\8 2019-q3-update\bin 2019-10-17T19:44:42 < kakimir64> yes 2019-10-17T19:44:58 < kakimir64> did the installer have option to add it to $PATH? 2019-10-17T19:45:31 < kakimir64> I don't remember very well it was like a year ago when I last time installed it to windows 2019-10-17T19:46:28 < kakimir64> your build environment should have an parameter to place that full path of arm-none-eabi-gcc.exe though 2019-10-17T19:46:56 < karlp> anyone got a stlnk3 yet? how fast can you capture trace? 2019-10-17T19:47:06 < kakimir64> stink3 2019-10-17T19:51:13 < Jan-> ok now it goes GENHDR include/libopencm3/swm050/irq.jsonBUILD lib/stm32/f0CC can.c 2019-10-17T19:51:21 < Jan-> is that what we would expect 2019-10-17T19:51:48 < Jan-> previously after the build line it would crap out 2019-10-17T19:52:54 < kakimir64> https://paste.ee/ << please 2019-10-17T19:53:05 < kakimir64> baste it 2019-10-17T19:53:07 < Jan-> well it's still working and some of that stuff is no longer in the buffer 2019-10-17T19:54:11 < Jan-> interestingly it still works better under mingw than it does in cmd 2019-10-17T19:58:47 < PaulFertser> kakimir64: but that's not a map file 2019-10-17T19:58:57 < PaulFertser> (not full map file) 2019-10-17T19:59:46 < Jan-> well it finished 2019-10-17T19:59:51 < Jan-> with no obvious error messages 2019-10-17T20:00:45 < Jan-> in the bin folder I have black-stm32f407ve-v2.0.bin, and also elf and hex files with the same name 2019-10-17T20:01:46 < Jan-> presumably I now have to use openocd to flash it onto the board? 2019-10-17T20:01:52 < Jan-> it's all connected up via an stlink 2019-10-17T20:02:42 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-17T20:02:47 < karlp> https://github.com/libopencm3/libopencm3-miniblink#flashing 2019-10-17T20:03:17 < karlp> this is the bit that can also differ a lot based on your board. "is it connected via jtag or swd" "does it even have that connected" "do you need to use the rom bootloader" 2019-10-17T20:03:42 < Jan-> I can't just type "openocd" anyway as the file is in C:\Users\JanH\.platformio\packages\tool-openocd\bin 2019-10-17T20:03:50 < Jan-> I guess I can use that one 2019-10-17T20:06:45 < Jan-> I'm not sure I have an "stlink.cfg" assuming that's a file 2019-10-17T20:08:00 -!- onio [~onio@2a00:23c5:7a01:8600:593a:de40:c8e0:3324] has joined ##stm32 2019-10-17T20:09:27 < karlp> you do. 2019-10-17T20:09:33 < karlp> it comes with openocd. 2019-10-17T20:10:05 < karlp> (unless platformio is hideously out of date) 2019-10-17T20:10:31 < onio> Hello having a bit of a problem and was wondering if anyone here could spot what I am doing wrong. 2019-10-17T20:10:43 < onio> https://paste.ubuntu.com/p/MgDCdT99C3/ 2019-10-17T20:11:35 < antto> srk, https://www.youtube.com/watch?v=pnBEACajFtg 2019-10-17T20:11:49 < onio> Datasheet to ADE7878 http://www.autex.spb.su/ad/ADE7878.pdf 2019-10-17T20:13:30 < onio> search "Figure 75. Connecting ADE7878 HSDC with an SPI" 2019-10-17T20:15:06 < Jan-> I'm sort of guessing at openocd.exe -f interface/stlink.cfg -f target/stm32f4x.cfg -c "program myfile.elf verify reset exit" 2019-10-17T20:15:08 < karlp> why link to spb.su? why not to analog devices? 2019-10-17T20:15:15 < srk> antto: pretty cool 2019-10-17T20:15:45 < kakimir64> PaulFertser: yes but it's something 2019-10-17T20:16:16 < Jan-> if it says "verified OK" is that a good indication it's worked 2019-10-17T20:16:19 < karlp> onio: do you have pullups on things? 2019-10-17T20:16:35 < karlp> Jan-:should be blinking then... 2019-10-17T20:16:51 < Jan-> Actually I can't tell whether it's blinking or not. 2019-10-17T20:17:03 < Jan-> But this all sounds pretty positive 2019-10-17T20:17:05 * Jan- two thumbs up 2019-10-17T20:17:23 < karlp> you're vision impaired right? or whatever the current wording for that should be? 2019-10-17T20:17:46 < karlp> the miniblink examples unfortunately use blinking to demo "working" 2019-10-17T20:17:50 < Jan-> I uhoh. 2019-10-17T20:17:51 < onio> karlp: not sure if I need pull up this is SPI and not I2C 2019-10-17T20:18:02 < Jan-> I say "blind" I'm not into the politics 2019-10-17T20:18:25 < karlp> onio: what cpol/cpha do you have? 2019-10-17T20:18:27 < antto> srk basically, it was a long iterative process.. first i began just by inspiration, started collecting audio recordings of the thing, then i found a guy on teh internetz who had two 303s and was gonna record specific clips for me, that helped.. and afterwards i obtained a x0xb0x, that helped tons, i kept it open for years and poked it.. 2019-10-17T20:18:28 < Jan-> my other half should be back really late tonight 2019-10-17T20:18:52 < karlp> onio:thoughyou say you're getting an entire byte?word? shifted? 2019-10-17T20:18:58 < kakimir64> Jan-: you guessed openocd command right 2019-10-17T20:19:00 < antto> and also i got help from a pile of clever people 2019-10-17T20:19:03 < karlp> you might just be dealing with fifos on the spi 2019-10-17T20:19:14 < karlp> and/or the write sizes/ read sizes to the databuffer. 2019-10-17T20:20:08 < onio> kind of the spi is set for 32-bit reads 2019-10-17T20:20:49 < onio> it seems like the rx buffer in the stm32 needs clearing or I need a another flag 2019-10-17T20:21:20 < karlp> how did you pick that part by the way? it's super pric 2019-10-17T20:21:58 < onio> no i didnt 2019-10-17T20:24:52 < Jan-> so okay right now I can compile what's in template_stm32.c 2019-10-17T20:24:59 < Jan-> what if I want to create my own project 2019-10-17T20:27:02 < kakimir64> PaulFertser: can you tell me why full config uses libc.a and without standard library I see libc_nano.a? 2019-10-17T20:27:14 < karlp> github.com/libopencm3/libopencm3-template 2019-10-17T20:27:24 < karlp> miniblink isn't a great base for making your own projects, as it says in the readme. 2019-10-17T20:27:41 < karlp> it's goal is to help people verify that their tools and hardware all work. 2019-10-17T20:27:56 < kakimir64> aka. "hello world" 2019-10-17T20:28:19 < Jan-> so hang on karl, who's behind libopencm3 other than you 2019-10-17T20:28:48 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-17T20:29:11 < catphish> so how should i spend my final 2 weeks of EU citizenship before they eject me? 2019-10-17T20:29:34 < BrainDamage> by glance at least 6 more main contributors 2019-10-17T20:29:56 < karlp> onio: what exactly is hsdc and how is it different to spi? 2019-10-17T20:30:06 < kakimir64> catphish: are you saying it's happening? 2019-10-17T20:30:20 < karlp> I just maintain it. others wrote some/most of it. 2019-10-17T20:30:24 < catphish> kakimir64: i am 2019-10-17T20:30:28 < karlp> depending on target. 2019-10-17T20:30:53 < kakimir64> visit steff 2019-10-17T20:31:09 < BrainDamage> how brexit works is that it happens if they don't do anything 2019-10-17T20:31:10 < kakimir64> and stack up stroopwaffles 2019-10-17T20:31:16 < BrainDamage> so they had to actively delay over and over 2019-10-17T20:31:38 < BrainDamage> they could / can still cancel it, but it's an hilarious situation in the first place 2019-10-17T20:31:40 < karlp> I wonder what they actually changed in that ade7878 vs the prior gens. 2019-10-17T20:31:46 < onio> From the micro-controller point of view it is simply spi working in receive mode only. Analog electronics are the people who decided to use fancy name 2019-10-17T20:32:19 < karlp> so you have to use i2c + hsdc, isntead of just using spi? 2019-10-17T20:32:20 < karlp> why? 2019-10-17T20:33:09 < karlp> why not just use the spi interface? 2019-10-17T20:33:51 < Jan-> er karl what is api.c for 2019-10-17T20:34:01 < PaulFertser> kakimir64: sorry, I do not understand how you switch between "configs". 2019-10-17T20:34:03 < Jan-> other than calculating 3x 2019-10-17T20:34:08 < onio> yes the, i2c is for configuring and then the device start sending data to the micro 2019-10-17T20:34:36 < kakimir64> PaulFertser: I build and see the bottom of .map that has debug_frame:s 2019-10-17T20:35:06 < kakimir64> I change eclipse CDT setting from project properties and from linker page 2019-10-17T20:35:11 < kakimir64> then build 2019-10-17T20:35:18 < karlp> it's just a dummy project Jan-, it doesn't do anything. 2019-10-17T20:35:26 < karlp> just file placeholders and the makefiles 2019-10-17T20:35:36 < onio> the hsdc sends instantaneous data values which you can guarantee with simple spi 2019-10-17T20:35:47 < PaulFertser> kakimir64: debug frames do not go into flash, those do not matter. Whatever settings you're changing, I have no idea about, so can't tell what they do. 2019-10-17T20:36:03 < Jan-> so I just delete everything and put my c code in "my-project" 2019-10-17T20:36:08 < Jan-> or do I leave the Makefile 2019-10-17T20:36:30 < kakimir64> PaulFertser: it has checkboxes -nostartfiles -nostdlib etc. 2019-10-17T20:36:48 < kakimir64> full is without those -no* checkboxes 2019-10-17T20:37:54 < kakimir64> also: -nodefaultlibs 2019-10-17T20:38:09 < PaulFertser> kakimir64: nostdlib should be without newlib but you still see it linked, something's fishy there. 2019-10-17T20:39:15 < karlp> fucking freertos. https://www.freertos.org/a00021.html#vTaskGetRunTimeStats doesn't mention that it also requires a heap implementation that implements free. dipshits 2019-10-17T20:40:07 < qyx> I am not using it 2019-10-17T20:40:11 < qyx> it is totally useless 2019-10-17T20:40:21 < karlp> just wanted to compare it to what I've got 2019-10-17T20:40:24 < Jan-> what does "cm3" mean anyway 2019-10-17T20:40:25 < karlp> I didn't _want_ to use it. 2019-10-17T20:40:32 < qyx> oh 2019-10-17T20:40:35 < karlp> cortex-m3, and no, we're not changing ti. 2019-10-17T20:40:48 < Jan-> karlp: But this is a - 2019-10-17T20:40:49 < Jan-> Oh. 2019-10-17T20:40:54 * Jan- subsides :/ 2019-10-17T20:41:18 < karlp> it used to be called libopenstm32, so this is "better" 2019-10-17T20:41:29 < Jan-> that would have made more objective sense. 2019-10-17T20:41:38 < qyx> yeah 2019-10-17T20:41:39 < karlp> except it supports quite a bit more than stm32, so no. 2019-10-17T20:41:42 < qyx> it should be libopenall 2019-10-17T20:41:55 < karlp> meh. 2019-10-17T20:42:02 < karlp> there's a cortex-a target in there already 2019-10-17T20:42:07 < karlp> and people want to add riscv 2019-10-17T20:42:11 < Jan-> I'm not sure if I'd have called it "libopen" anything 2019-10-17T20:42:17 < qyx> such names are cool 2019-10-17T20:42:23 < qyx> opencl, opengl, openal, openall 2019-10-17T20:42:23 < karlp> Jan-:well, fortunately, we're passed that point :) 2019-10-17T20:42:32 < Jan-> I mean obviously it's a library 2019-10-17T20:42:36 < Jan-> Obviously it's open source 2019-10-17T20:42:37 < qyx> ok, back to work 2019-10-17T20:42:40 < Jan-> and "cm3" tells me nothing 2019-10-17T20:42:51 < karlp> you don't need it to tell you anything. 2019-10-17T20:42:54 < karlp> it's a name. 2019-10-17T20:43:01 < Jan-> names should be descriptive 2019-10-17T20:43:01 < kakimir64> PaulFertser: are you more interested of text section? 2019-10-17T20:43:06 < karlp> like Jan? 2019-10-17T20:43:10 < karlp> how is that descriptive? 2019-10-17T20:43:36 < karlp> if you want to be upset by things like this, you're going to have a bad time. and probbly a stressful life. 2019-10-17T20:43:43 < karlp> or, just trolling 2019-10-17T20:43:52 < Jan-> I suppose I should have gone with "shortgirl" :( 2019-10-17T20:43:54 < qyx> that would be the better case 2019-10-17T20:44:00 < PaulFertser> kakimir64: by inspecting full map file you can see what functions are getting pulled in why. 2019-10-17T20:44:05 < qyx> what is a shortgirl 2019-10-17T20:44:11 * Jan- points at self :( 2019-10-17T20:44:23 < qyx> noice 2019-10-17T20:44:30 * Jan- is five feet one in platform sneakers 2019-10-17T20:44:32 < qyx> ok, so, back to work 2019-10-17T20:44:35 < kakimir64> PaulFertser: what configuration you want it with? 2019-10-17T20:44:58 < PaulFertser> Jan- is also using imperial units :/ From bad to worse indeed. 2019-10-17T20:45:08 < Jan-> gimme a break I was born in california 2019-10-17T20:45:11 < PaulFertser> kakimir64: idk, the one with nano I guess 2019-10-17T20:45:12 < Jan-> I'm like 152cm 2019-10-17T20:46:01 < BrainDamage> for the record, cm3 stands for cortex-m3 2019-10-17T20:46:17 < BrainDamage> which is the cpu type many micros run, including stm32 ... 2019-10-17T20:46:27 < karlp> already told them, they already got upset abotu it 2019-10-17T20:47:30 < Jan-> is an stm32f407 a cortex m3 2019-10-17T20:47:40 < BrainDamage> cortex m4 2019-10-17T20:47:49 < BrainDamage> which is cortex m3 compatible 2019-10-17T20:48:08 < BrainDamage> just like i686 is compatible with i586 2019-10-17T20:48:09 < kakimir64> PaulFertser: what functions should I see in .text with -nostdlib? 2019-10-17T20:49:55 < PaulFertser> kakimir64: you should see all the functions that were included in the binary. And for each function you see what other function needed it. 2019-10-17T20:51:55 < kakimir64> I surely see what is included but I don't see a part where it says what function needs what 2019-10-17T20:53:55 < kakimir64> https://paste.ee/p/7yTzo you wanted to see this part? 2019-10-17T20:54:03 < kakimir64> -nostdlib 2019-10-17T20:56:29 < kakimir64> I assumed -nostdlib implies at least no printf 2019-10-17T20:56:39 < qyx> whats wrong? https://i.imgur.com/nLSDXUZ.jpg 2019-10-17T20:56:59 < kakimir64> I need to get those removed from my build wasting my flash 2019-10-17T20:57:20 < kakimir64> qyx: everything 2019-10-17T20:57:26 < BrainDamage> neutral shorted to ground 2019-10-17T20:57:31 < PaulFertser> kakimir64: my example: http://paste.debian.net/1107761/ first line says: fflush lib function was included because main.o references fflush 2019-10-17T20:57:45 < BrainDamage> and live plugged into the neutral ... 2019-10-17T20:58:18 < PaulFertser> kakimir64: do you have -ffunction-sections and -Wl,--gc-sections ? 2019-10-17T21:02:13 < BrainDamage> the hilarious thing is that that socket will work, but the case of something plugged in it will be a murder machine depending on how well the differential works 2019-10-17T21:02:31 < kakimir64> oh wow 2019-10-17T21:02:44 < kakimir64> PaulFertser: linker has only -gc-sections 2019-10-17T21:03:45 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-17T21:04:32 < PaulFertser> kakimir64: have you checked my paste? You should have similar section in your map file. 2019-10-17T21:05:22 < kakimir64> oh 2019-10-17T21:05:24 < kakimir64> yes 2019-10-17T21:05:32 < kakimir64> I see it it's first 2019-10-17T21:06:20 < kakimir64> I need to go bbl> 2019-10-17T21:08:15 < PaulFertser> Hm, I thought people who can't see much tend to use command line interfaces because they're much more predictable and manageable without visual feedback. 2019-10-17T21:09:06 < BrainDamage> text based, rather than command line 2019-10-17T21:09:19 < BrainDamage> tui is not command line 2019-10-17T21:09:59 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-17T21:42:23 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-17T21:43:40 < PaulFertser> BrainDamage: well, type command -> hear response feels like the most natural way to communicate with computers, an "ncurses" interfaces would be harder to use I guess? 2019-10-17T21:48:26 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-17T21:57:01 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-17T22:17:30 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-17T22:19:30 < kakimir64> can UPS wake up computer after mains has returned? 2019-10-17T22:19:40 < kakimir64> via usb 2019-10-17T22:20:25 < kakimir64> I have never used UPS but I would like to server to gracefully suspend after a period without mains 2019-10-17T22:20:34 < kakimir64> but keep backup power output on 2019-10-17T22:20:55 < Steffanx> i dont know about such system 2019-10-17T22:21:00 < Steffanx> mine cant do that 2019-10-17T22:21:07 < kakimir64> what can it do? 2019-10-17T22:21:29 < Steffanx> boot it up. it can only signal it, its running on the battery 2019-10-17T22:21:36 < Steffanx> *it cant boot it up 2019-10-17T22:21:47 < kakimir64> yes certainly 2019-10-17T22:21:52 < Steffanx> that's all mine can do 2019-10-17T22:22:22 < kakimir64> that is why I figured suspend would allow some wakeup from usb 2019-10-17T22:22:32 < Steffanx> this reminds me i probably didnt reconnect the usb connector after i moved my NAS around. It's no longer on the floor. 2019-10-17T22:23:49 < Steffanx> opened it up a week or so ago. tooo much dust in there 2019-10-17T22:25:08 < kakimir64> I turned of all case fans 2019-10-17T22:25:11 < kakimir64> that should do it 2019-10-17T22:25:17 < kakimir64> off 2019-10-17T22:25:40 < kakimir64> dust should not be that much of a problem 2019-10-17T22:26:15 < Steffanx> i have em running 24/7 2019-10-17T22:26:19 < Steffanx> at a pretty low speed 2019-10-17T22:26:45 < Steffanx> especially the lsi card get pretty damn warm, so i have a fan blowing some air at it 2019-10-17T22:32:37 -!- beanana [~banandana@cassini.whatbox.ca] has quit [Quit: WeeChat 2.5] 2019-10-17T22:34:56 < Steffanx> you dont even cool it kakimir64? 2019-10-17T22:35:07 < kakimir64> yeah 2019-10-17T22:35:21 < kakimir64> I mean I have not measured the temperature 2019-10-17T22:35:41 < kakimir64> I assume it pulls like 10watts 2019-10-17T22:35:55 < kakimir64> idle* 2019-10-17T22:36:27 < kakimir64> that is the state of card in my use 2019-10-17T22:36:31 < kakimir64> 99.999% of time 2019-10-17T22:41:23 < Steffanx> hm 2019-10-17T22:42:34 < Steffanx> must be -10C in finland all the time :) 2019-10-17T22:45:44 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 268 seconds] 2019-10-17T22:46:43 < kakimir64> I need to measure the temperatuire of LSI card when it's relevant 2019-10-17T22:49:52 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-17T22:49:52 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-17T22:49:52 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-17T22:49:59 < Steffanx> damn i suck so hard. Spend part of my yesterday evening on getting homeassistant to autostart (again) using systemd and a user called homeassistant. All fine. But i couldnt make it work 2019-10-17T22:50:12 < Steffanx> only just now i found out why. I wrote "homeassistent" all the time -_- 2019-10-17T22:51:23 < zyp> I just run it in docker 2019-10-17T22:52:08 < Steffanx> i have it running on some rpi 2019-10-17T22:52:33 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-17T22:52:57 < qyx> what is it 2019-10-17T22:53:04 < qyx> lazy to google 2019-10-17T22:53:06 < Steffanx> just some home automation crap 2019-10-17T22:53:13 < qyx> ok, not being kaki 2019-10-17T22:53:46 < Steffanx> it seemed pretty light weight at the time. 2019-10-17T22:53:50 < Steffanx> but fuck yaml. 2019-10-17T22:54:05 < kakimir64> it would be fun to make those ai assistants to work for you 2019-10-17T22:54:08 < kakimir64> trick them 2019-10-17T22:55:01 < kakimir64> I mean they have set some limitations to resources per user 2019-10-17T22:56:27 < kakimir64> what other register should be set on startup than SP? 2019-10-17T22:57:15 < kakimir64> and even that is done by hardware 2019-10-17T23:01:01 < bitmask> this board's components come out to like $13, why did I spend 60 2019-10-17T23:01:08 < Steffanx> Yes why 2019-10-17T23:02:27 < bitmask> quantities 2019-10-17T23:02:31 < bitmask> and shipping 2019-10-17T23:02:47 < karlp> qyx: that's not the normal place N gets tied to earth :) 2019-10-17T23:02:58 < karlp> qyx: was that tripping an rcd? 2019-10-17T23:04:41 < qyx> L was tied to earth :> 2019-10-17T23:04:56 < karlp> well, I don't trust that anyonegets the colours right. 2019-10-17T23:04:58 < qyx> it was a friend's outlet 2019-10-17T23:07:02 < kakimir64> PaulFertser: https://paste.ee/p/Nlm7U strlen doesn't have .text section? 2019-10-17T23:07:48 < kakimir64> in binary every other str* function has it's own .text.str* section 2019-10-17T23:07:52 < PaulFertser> kakimir64: yes, but it's part of an archive and it shouldn't get included if it's not needed. 2019-10-17T23:08:23 < kakimir64> in binary I mean the included part with addresses in image and stuff 2019-10-17T23:09:35 < kakimir64> every other libc_nano.a function has it's own .text.(name) section 2019-10-17T23:10:29 < PaulFertser> I am not sure how to explain that. But strlen can't take much anyway. 2019-10-17T23:11:20 < kakimir64> 0x10 2019-10-17T23:11:27 < kakimir64> https://paste.ee/p/7yTzo 2019-10-17T23:14:09 < kakimir64> see that .text that sticks out like a sore thumb? 2019-10-17T23:14:45 < zyp> it's a question of whether it was built with -ffunction-sections or not 2019-10-17T23:16:04 < kakimir64> yes it is build with 2019-10-17T23:17:05 < kakimir64> and -fdata-sections 2019-10-17T23:17:16 < zyp> what is? strlen? 2019-10-17T23:17:55 < kakimir64> in my project properties I think newlib is prebuilt 2019-10-17T23:18:34 < kakimir64> I think I need to check this newlib version 2019-10-17T23:19:02 < PaulFertser> kakimir64: I suggest to not bother about it. 2019-10-17T23:19:13 < PaulFertser> kakimir64: what makes you worry, 16 extra bytes really? 2019-10-17T23:19:28 < kakimir64> strlen causing mem manage handler 2019-10-17T23:21:06 < kakimir64> could strlen fail because of no libc start? 2019-10-17T23:21:57 < kakimir64> did you see the google drive shared screenshot PaulFertser? 2019-10-17T23:22:05 < kakimir64> like 2 days ago 2019-10-17T23:25:59 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-17T23:28:28 < PaulFertser> kakimir64: no, haven't noticed it 2019-10-17T23:28:52 < PaulFertser> kakimir64: strlen should be very simple, I'd just go through it with stepi. 2019-10-17T23:28:58 < PaulFertser> To see why it fails. 2019-10-17T23:29:17 < kakimir64> oh yes 2019-10-17T23:29:23 < kakimir64> I got a bit sidetracked 2019-10-17T23:29:30 < kakimir64> like 2 days ago 2019-10-17T23:30:22 -!- con3 [~kvirc@154.119.40.174] has joined ##stm32 2019-10-17T23:31:35 < canton7> kakimir64, you haven't set it up to fault on unaligned access, have you? Have you enabled the MPU? 2019-10-17T23:31:50 < antto> em pea you 2019-10-17T23:32:09 < kakimir64> MPU should be enabled by freertos I assume 2019-10-17T23:32:34 < antto> 2019-10-17T23:32:51 < canton7> I've seen people accidentally configure a region as device memory (or something?), which disallowed unaligned access 2019-10-17T23:33:02 < kakimir64> I have MPU port of freertos 2019-10-17T23:33:03 < con3> hmm... guys ive got a question. i've been calibrating adc's for a bit, but hows this done in production. How do they calibrate stuff? Do they have like a custom calibration jig that they build that feeds references and they calibrate automatically according to that? 2019-10-17T23:33:03 < karlp> well, he was takling about using the mpu version of freertos 2019-10-17T23:33:34 < kakimir64> I can actually view MPU regions in IDE 2019-10-17T23:33:49 < kakimir64> at least what freertos would like to be regions 2019-10-17T23:35:44 < antto> con3 no idea on your chipz, but for eggzample in xmega, there are factory calibration values stored in some place in the flash which you can load into the periph 2019-10-17T23:35:57 < kakimir64> well shit.. I have been debugging the wrong configuration for a while 2019-10-17T23:36:04 < sync> rekt 2019-10-17T23:36:05 < antto> huhuhu 2019-10-17T23:36:12 < kakimir64> I started with release and then moved to debug configuration 2019-10-17T23:36:20 < antto> каким0р pls 2019-10-17T23:36:20 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 276 seconds] 2019-10-17T23:36:32 < kakimir64> but the debug button didn't realize it 2019-10-17T23:36:34 < antto> i've done that.. very long ago 2019-10-17T23:37:09 < antto> iz iz like adjusting the volume/eq/pan/fx of a muted audio channel 2019-10-17T23:37:46 < antto> "ah, that's too quiet now, too bassy, too much to the left.. okay that's much better" 2019-10-17T23:37:48 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-17T23:38:40 < kakimir64> I realized as I build Release with stdlib and it was too big binary 2019-10-17T23:38:56 < kakimir64> it complained about build error when I pressed debug button 2019-10-17T23:39:04 < con3> antto: i take it not all mcu's have factory calibrated adc's or there might be an external high precision adc being used. how do they mass calibrate these things :/ 2019-10-17T23:39:41 < antto> no idea 2019-10-17T23:39:44 < kakimir64> I once made code that linearized internal voltage reference in function of internal temperature 2019-10-17T23:39:56 < antto> but factories surely have some advanced toyz 2019-10-17T23:39:58 < sync> huh con3? 2019-10-17T23:40:14 < sync> they just inject reference voltages like you speculated 2019-10-17T23:40:26 < sync> and then burn the values during their test 2019-10-17T23:40:34 < sync> if they are factory calibrated 2019-10-17T23:41:51 < kakimir64> how de fug I stepi in eclipse? 2019-10-17T23:42:07 < con3> sync: ah just wondering. I've just been thinking about how production stuff are done and while calibrating an adc wondered if they just build a precision jig to do what im doing. Was just a thought as there are various sensors that I looked at like carbon monoxide, etc and the thought of calibration for those also popped in. But then again, at that scale they have nice toys 2019-10-17T23:43:06 < sync> they have to test the chip anyway 2019-10-17T23:43:09 < antto> kakimir64 put yourself in the shoes of a java guy.. now what would you do to step? 2019-10-17T23:43:31 -!- onio [~onio@2a00:23c5:7a01:8600:593a:de40:c8e0:3324] has quit [Quit: Leaving] 2019-10-17T23:43:35 < kakimir64> I know absolutelly nothing about java and I wish not to 2019-10-17T23:44:19 < antto> i mean 2019-10-17T23:44:32 < antto> if you wanna find teh kill0r, you gotta think like teh kill0r 2019-10-17T23:44:50 < antto> that's how they do it in teh muviez 2019-10-17T23:44:54 < kakimir64> now I found the stepi 2019-10-17T23:45:05 < antto> javimir64 2019-10-17T23:45:11 < kakimir64> little i button in debug tab 2019-10-17T23:45:18 < kakimir64> not in toolbar but debug tab 2019-10-17T23:45:40 < antto> stepping is not a "tool" tbh ;P~ 2019-10-17T23:47:05 < kakimir64> how do I step backwards 2019-10-17T23:47:08 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-17T23:47:19 < kakimir64> actually stepi backwards 2019-10-17T23:47:27 < antto> unclick the stepi butten? 2019-10-17T23:47:32 < kakimir64> joke joke 2019-10-17T23:47:55 < antto> put teh mouse on the under side of the desk, and click THEN 2019-10-17T23:48:14 < antto> i don't know, improvise 2019-10-17T23:48:31 < Cracki> backstepping is an advanced feature, requires snapshotting system state somehow 2019-10-17T23:48:47 < antto> wait, stepping back during debug? 2019-10-17T23:49:04 < antto> wut tha hell kakimir64, where have you seen such fancyness? 2019-10-17T23:49:10 < Cracki> yes, in java it's possible because java is specified enough that you can easily snapshot 2019-10-17T23:49:27 < Cracki> other IDEs for other languages have gained the ability too 2019-10-17T23:49:47 < antto> i don't believe this sh*t 2019-10-17T23:50:11 < antto> step back from a crash ;P~ 2019-10-17T23:50:23 < Cracki> a common approach is to recalculate all deterministically following changes, and only snapshot things that act on outside input 2019-10-17T23:50:33 < Cracki> saves storage 2019-10-17T23:50:34 < karlp> just read the gdb manual 2019-10-17T23:51:00 < antto> 2019-10-17T23:51:04 < karlp> https://www.gnu.org/software/gdb/news/reversible.html 2019-10-17T23:53:25 < karlp> lauterbach (at least) supports it on arm-cortex too. 2019-10-17T23:54:03 -!- c10ud^ [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-17T23:55:14 < qyx> ok, I made the battery gauge working 2019-10-17T23:55:20 < qyx> I don't know how though 2019-10-17T23:55:37 < qyx> interesting that the charge and discharge FETs are off by default 2019-10-17T23:56:43 < qyx> also whoever invented smbus should try to actually use it 2019-10-17T23:57:13 < qyx> so to set a flag in the dataflash you have to invoke a command 2019-10-17T23:57:14 < qyx> huh 2019-10-17T23:57:30 < qyx> a command can be run by writing to a manufacturerdata register 2019-10-17T23:57:51 < qyx> a response? ok, read manufacturer data length register 2019-10-17T23:58:00 < qyx> then read manufacturer data data register 2019-10-17T23:58:24 < kakimir64> no use, only invent 2019-10-17T23:58:30 < Jan-> So I have this "black stm32f407" board. It has an stm32f407, a flash chip, a couple of leds and buttons and an SD card slot 2019-10-17T23:58:36 -!- sterna [~Adium@c-ddb9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-17T23:58:37 < qyx> it would be too easy to simply write 0xf00 to address 0x4000 2019-10-17T23:58:37 < Jan-> There isn't a programmer on. 2019-10-17T23:58:43 < Jan-> I just use an stlink. 2019-10-17T23:58:48 < qyx> yes 2019-10-17T23:58:57 < Jan-> So how does this differ from any other board using that same microcontroller? 2019-10-17T23:59:46 < PaulFertser> Jan-: it doesn't, that's the point you've been told several times. --- Day changed Fri Oct 18 2019 2019-10-18T00:00:46 * con3 needs to figure out version control on altium 2019-10-18T00:01:00 < kakimir64> https://imgflip.com/i/3dm0z7 2019-10-18T00:01:26 < Jan-> PaulFertser: well sure I get that but in that case why is there so much screwing around with board definitions. 2019-10-18T00:01:27 < kakimir64> so strlen counts okay 2019-10-18T00:01:32 < kakimir64> returns then 2019-10-18T00:01:48 < kakimir64> then shit goes shit 2019-10-18T00:01:57 < con3> ah it supports git 2019-10-18T00:04:07 < kakimir64> is my disassembly view ok if I see .w here and there? 2019-10-18T00:04:16 < qyx> because there may be a memory on a I2C1 port, or there may not be 2019-10-18T00:04:21 < qyx> or there is a LED on a PA15 2019-10-18T00:04:21 < PaulFertser> Jan-: I'm not sure, haven't seen any screwing. 2019-10-18T00:04:39 < qyx> or a SD card may be on a sdmmc1 or no 2019-10-18T00:04:46 < qyx> t 2019-10-18T00:06:22 < antto> electric eel on ADC3 2019-10-18T00:06:36 < Jan-> qyx: I'm happy to deal with all that 2019-10-18T00:06:49 < Jan-> there's really not much on these dev boards anyway 2019-10-18T00:06:58 -!- banana [~banandana@cassini.whatbox.ca] has joined ##stm32 2019-10-18T00:06:59 < Jan-> I mean all that screwing around so I can use "LED1" rather than "PA6" 2019-10-18T00:07:57 < Jan-> oh and yeah the upload did work 2019-10-18T00:07:59 < Jan-> it was flashing 2019-10-18T00:08:16 < qyx> the screwing around is to allow you to write beautiful codez 2019-10-18T00:08:45 < antto> LED1.kindly_blink_pls(); 2019-10-18T00:09:11 < Jan-> I can do that with a one line define 2019-10-18T00:09:20 < Jan-> it's almost like I could just lie to it 2019-10-18T00:09:25 < Jan-> and say it was any board with that chip 2019-10-18T00:09:44 < antto> lying is not gud 2019-10-18T00:09:55 < antto> sooner or later u gon get entangled in teh lies 2019-10-18T00:12:08 < Jan-> but it'd work 2019-10-18T00:12:09 < Jan-> wouldn't it? 2019-10-18T00:13:51 < qyx> you will find out in the future 2019-10-18T00:16:13 < qyx> I was tempted to be impolite and ask you ASL 2019-10-18T00:16:23 < qyx> are you a uni student? 2019-10-18T00:17:12 < antto> at teh university of lies and fraud 2019-10-18T00:17:30 < qyx> I mean, it takes years to start enjoying writing kawaii code 2019-10-18T00:17:46 < qyx> it must be a pleasure to look at 2019-10-18T00:18:07 < qyx> sometimes the look and feel of the code is way more than its functionality 2019-10-18T00:18:14 < antto> yeah 2019-10-18T00:18:17 < qyx> if we return to the libopencm3 2019-10-18T00:18:20 < Cracki> >trap code 2019-10-18T00:18:29 < qyx> you can always use direct register access 2019-10-18T00:18:36 < antto> Cracki don't ruin teh moment 2019-10-18T00:18:37 < qyx> or #0xdeadbeaf = 0x24; 2019-10-18T00:18:44 < oz4ga> you could try your code on a F1 or a board where the LED is connected to another GPIO 2019-10-18T00:18:45 < Cracki> that's not consensual 2019-10-18T00:19:25 < Cracki> >programming socks 2019-10-18T00:19:32 < qyx> so theres a ton of different "ecosystems", platforms, tools, build systems and things 2019-10-18T00:19:51 < qyx> because everyone things they can write some code better 2019-10-18T00:19:58 < qyx> to use better algo 2019-10-18T00:20:13 < qyx> to actually name something they came up with (huh patterns) 2019-10-18T00:20:14 < oz4ga> ofcause we can 2019-10-18T00:20:58 < qyx> see, like building better and better antennas 2019-10-18T00:21:35 < karlp> we even have a list of c++ ones in zypsnips 2019-10-18T00:22:08 < kakimir64> arm-none-eabi-objdump --disassembler-options=force-thumb -x -d "${FileName}" maybe force-thumb should make sense to disassembler view 2019-10-18T00:24:56 < Cracki> force thumb o.o 2019-10-18T00:25:27 < kakimir64> force dump 2019-10-18T00:27:22 < kakimir64> oh 2019-10-18T00:27:29 < kakimir64> me stupid 2019-10-18T00:27:41 < kakimir64> forget what I said 2019-10-18T00:28:06 < qyx> I already did 2019-10-18T00:29:51 < Cracki> https://i.imgflip.com/3dm4jp.jpg 2019-10-18T00:31:39 < kakimir64> it's forced 2019-10-18T00:32:09 < Cracki> say what *g* 2019-10-18T00:40:30 -!- brdb [~basdb@2601:18c:8500:7f5b:21e:6ff:fe30:9762] has quit [Read error: Connection reset by peer] 2019-10-18T00:41:48 < kakimir64> janne@HP-Compaq-Pro-6300-MT:/opt/SEGGER/JLink/GDBServer$ lsRTOSPlugin_embOS.so RTOSPlugin_FreeRTOS.so 2019-10-18T00:41:51 < kakimir64> what dat? 2019-10-18T00:43:39 < kakimir64> https://dzone.com/articles/adding-freertos-thread-awareness-to-gdb-andeclipse 2019-10-18T00:43:41 < kakimir64> sweet 2019-10-18T00:43:53 < kakimir64> this is pro stuff 2019-10-18T00:44:50 < karlp> works in openocd too. 2019-10-18T00:45:16 < karlp> just needs http://openocd.zylin.com/#/c/5273/ 2019-10-18T00:45:29 < karlp> (depending on freertos versions....) 2019-10-18T00:45:35 < karlp> (I'm using that patch right now and it works. 2019-10-18T00:45:46 < karlp> though threadids are kinda gross, they're just using the address of the TCB 2019-10-18T00:54:36 < Jan-> karlp: you're a developer on libopencm3, right? 2019-10-18T00:55:10 < kakimir64> oh wow what a turd.. mcuxpresso bundled with some 4.x version of jlink and it cannot even launch GDBServer from ide because some change in jlink binary causes plugin not to detect it as jlinkgdbserver 2019-10-18T00:56:32 < kakimir64> current version is like 6.52e 2019-10-18T00:58:07 < kakimir64> actually I this has 6.48 bundled not bad.. my mistake yet again 2019-10-18T00:58:17 < karlp> Jan-: maintainer mostly, but yes. 2019-10-18T01:06:57 -!- con3 [~kvirc@154.119.40.174] has quit [Read error: Connection reset by peer] 2019-10-18T01:10:42 < Jan-> karlp: how long has it been in existence? has it been used on any big projects? 2019-10-18T01:13:47 < karlp> git log goes back to 2009. 2019-10-18T01:13:59 < karlp> I know of a few people using it commercially, and bunch of other projects. 2019-10-18T01:14:11 < karlp> there's no requirement of anyone to notify anyone about using it. 2019-10-18T01:14:25 < karlp> it's lgpl mind you, so it's not for everyone 2019-10-18T01:15:35 < karlp> https://github.com/libopencm3/libopencm3/commit/1b708b68b gives an example of the different, purer times it lived in back then... 2019-10-18T01:17:47 < karlp> thhe things you find reading the arch ref man that you wouldn't normally think about, "An unaligned or multi-word access that crosses a 0.5GB address boundary is UNPREDICTABLE ." 2019-10-18T01:26:23 < Jan-> does that mean I can use it and sell the widget or what 2019-10-18T01:31:39 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 240 seconds] 2019-10-18T01:33:40 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-18T01:37:44 < karlp> you have to comply with the terms of the lgpl :) 2019-10-18T01:37:49 < karlp> and yeah, fuck that shit. 2019-10-18T01:38:43 < qyx> lol https 2019-10-18T01:38:52 < karlp> the general consensus is that you "must" provide "a mechanism" that allows your endusers the ability to replace the libopencm3 portion with their own version of libopencm3. 2019-10-18T01:39:22 < karlp> this can be, but is not restricted to, providing a makefile and some .o files for your private portions. 2019-10-18T01:39:32 < karlp> but beyond that, speak to your legal counsel. 2019-10-18T01:39:57 < karlp> alternatively, simply make your project lgpl or stronger itself. 2019-10-18T01:40:06 < Jan-> but that's completely mad in our industry, it would only confuse people 2019-10-18T01:40:11 < Jan-> they wouldn't have the tiniest idea what we were on about 2019-10-18T01:40:23 < qyx> just provide all the sources and you are safe 2019-10-18T01:40:24 < karlp> no-one said they had to use it. 2019-10-18T01:40:33 * qyx is using it commercially 2019-10-18T01:40:37 < karlp> you just have to have it available "on demand" and noted somewhere in your shrinkwrap 2019-10-18T01:40:41 * karlp too 2019-10-18T01:40:46 < qyx> https://imgur.com/7wf6Uhj 2019-10-18T01:41:05 < qyx> finally I am able to extract the charging current from the mighty battery gauge 2019-10-18T01:41:08 < karlp> qyx: is that 3.5mm or 3.81mm? :) 2019-10-18T01:41:18 < qyx> 3.5 2019-10-18T01:41:23 * karlp laughs 2019-10-18T01:41:32 < qyx> why 2019-10-18T01:41:33 < karlp> have you ever gotten the wrong one? 2019-10-18T01:41:38 < Jan-> so how the hell would anyone ever replace libopencm3 in the presence of .o files for my code. 2019-10-18T01:41:39 < qyx> sure :D 2019-10-18T01:41:41 < Jan-> that's absolutely crazy 2019-10-18T01:42:23 < karlp> gcc -o newversion.elf [list of private.os] [rebuilt "fixed" libopencm3.a] 2019-10-18T01:42:36 < qyx> btw it has 12V + rs485 2019-10-18T01:42:41 < qyx> bay fit your etactica 2019-10-18T01:42:48 < Jan-> karlp: do you think it is even vaguely plausible that anyone will ever do that 2019-10-18T01:43:13 < karlp> Jan-:doens't matter, lgpl requires that the possibility is available. 2019-10-18T01:43:32 < karlp> it generally only becomes an issue when the original vendor is no longer around. 2019-10-18T01:43:51 < Jan-> I long ago concluded that gpl was nuts 2019-10-18T01:43:53 < karlp> I'm not forcing you to use it, just saying that it is something you should at least be aware of. 2019-10-18T01:44:05 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-18T01:44:10 < karlp> well, the project _used_ to be gpl :) so .... 2019-10-18T01:44:26 < karlp> qyx: modbus? 2019-10-18T01:44:31 < qyx> not yet 2019-10-18T01:44:39 < karlp> I can write plugins for anything modbus. 2019-10-18T01:44:42 < karlp> want help? :) 2019-10-18T01:44:56 < qyx> it is basically a 2S li-ion backup psu 2019-10-18T01:44:57 < kakimir32> https://drive.google.com/file/d/1EQ1eVcTH4r-runFZOdEEYCga5Ds3zgdh/view?usp=sharing one stepi and shit goes into shit 2019-10-18T01:45:01 < karlp> the original freemodbus duder just put it all on github, as v1.6, 2019-10-18T01:45:04 < Jan-> is the arduino style hal the same? 2019-10-18T01:45:11 < karlp> so I sent him a bunch of patches, but it's been radio silence 2019-10-18T01:45:14 < kakimir32> is it trying to do what I think it's trying to do? 2019-10-18T01:45:18 < karlp> Jan-:do your own reading, I've no idea. 2019-10-18T01:45:27 < kakimir32> load to R2 value that is in address of R2? 2019-10-18T01:45:35 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T01:45:47 < kakimir32> *in address stored in R2? 2019-10-18T01:46:03 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has joined ##stm32 2019-10-18T01:46:05 < qyx> for this particular application I will not be using modbus 2019-10-18T01:48:01 < kakimir32> address 0x3000200 is no no 2019-10-18T02:03:15 < zyp> Jan-, I've bought at least two commercial devices containing libopencm3 2019-10-18T02:04:16 < zyp> I don't know how you qualify big, and I don't know the sales numbers either 2019-10-18T02:04:48 < Jan-> is it just me or is this one of those things that will never actually be a thing 2019-10-18T02:04:58 < zyp> «this»? 2019-10-18T02:05:13 < Jan-> the whole lgpl thing 2019-10-18T02:05:32 < zyp> in what sense? 2019-10-18T02:05:33 < Jan-> I mean nobody but nobody is ever going to go "Ooh, can I have the uncompiled source of your stm32 hardware access layer and binaries of your code." 2019-10-18T02:05:45 < Jan-> there's no way you'd release the code anyway 2019-10-18T02:05:50 < Jan-> binary or not 2019-10-18T02:05:55 < Jan-> you'd get ripped off 2019-10-18T02:06:34 < zyp> idk, both the products I've got in mind released their source 2019-10-18T02:06:45 < Jan-> anyway it looks like stm32duino is a strange mix of licenses 2019-10-18T02:07:12 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-18T02:09:16 < zyp> most stuff is 2019-10-18T02:09:52 < Jan-> it looks like all this is a no go anyway then 2019-10-18T02:10:23 < deltab> "This software or any part thereof, including modifications and/or derivative works of this software, must be used and execute solely and exclusively on or in combination with a microcontroller or microprocessor device manufactured by or for STMicroelectronics." 2019-10-18T02:10:41 < Jan-> what does that mean 2019-10-18T02:10:50 < Jan-> what else are you gonna do with it :) 2019-10-18T02:10:53 < zyp> I used to work on hardware drivers for android 2019-10-18T02:10:57 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-18T02:11:02 < deltab> the USB code can only be used on ST's hardware 2019-10-18T02:11:06 < zyp> android has a pretty fucked up mix of licenses 2019-10-18T02:11:26 < karlp> zyp: what products if I may ask? 2019-10-18T02:11:29 < karlp> bmp? and? 2019-10-18T02:11:31 < Jan-> I get the feeling that most of this open source stuff only works out in practice because nobody ever actually wants the source. 2019-10-18T02:11:35 < zyp> karlp, trezor 2019-10-18T02:11:38 < karlp> ah. 2019-10-18T02:11:48 < karlp> they've abandoned it now fwiw. 2019-10-18T02:11:52 < zyp> yeah 2019-10-18T02:11:57 < karlp> usb man 2019-10-18T02:11:59 < zyp> that's my understanding too 2019-10-18T02:12:00 * karlp flips the table 2019-10-18T02:12:19 < zyp> but I figure they had quite a few thousand devices out with libopencm3 at some point 2019-10-18T02:12:21 * karlp has a surprisingly short set of antiloop locally. 2019-10-18T02:12:30 < karlp> that's my understanding yes, 2019-10-18T02:12:37 < karlp> at least, from their own presentations of it. 2019-10-18T02:13:21 < zyp> I mean, there's over a thousand devices running laks in the wild now, and trezor is quite some orders of magnitude more known 2019-10-18T02:13:49 < karlp> indeed. 2019-10-18T02:14:03 < karlp> still, visibility is weird. 2019-10-18T02:14:22 < karlp> openssl was used _everywhere_ and had nothing behind it, as we all know :) 2019-10-18T02:14:59 < zyp> lack of better alternatives at the time? 2019-10-18T02:15:41 < Jan-> so does this mean that every commercial device that uses the arduino libraries (for instance) is legally obliged to give me all their firmware on demand? 2019-10-18T02:15:57 < karlp> Jan-:no, and we never said that. 2019-10-18T02:16:06 < zyp> what does the arduino license say? 2019-10-18T02:16:07 < Jan-> sure you did 2019-10-18T02:16:10 < karlp> gpl would, but arduino isn't that, and nore is libopencm3 2019-10-18T02:16:17 < Jan-> because that's nuts 2019-10-18T02:16:24 < karlp> well, sure, I read firmware source into your statement 2019-10-18T02:16:36 < karlp> lgpl requires firmware .o's or equivalent 2019-10-18T02:16:45 < Jan-> it hardly matters it's what you need to build the software 2019-10-18T02:16:49 < karlp> that's no different that just hooking up an stlink to you rproduct and dumping rom. 2019-10-18T02:17:02 < zyp> lgpl is kinda bullshit for embedded though 2019-10-18T02:17:03 < Jan-> sure, but you can turn that off. at least you can on avr, I assume you can on stm32. 2019-10-18T02:17:03 < karlp> yeah, but you can't do anything with it other than replace the lgpl portions 2019-10-18T02:17:29 < Jan-> requiring people to give away their firmware is insane, nobody is ever going to go for that 2019-10-18T02:17:30 < zyp> lgpl was written for platforms with shared libs, where you can just replace the .so/.dll files 2019-10-18T02:17:44 < zyp> not projects that statically link 2019-10-18T02:17:48 < Jan-> your stuff will be cloned in china before you can say "intellectual property" 2019-10-18T02:18:22 < karlp> Jan-: if your product has no value without your support, then it deserved it. 2019-10-18T02:18:39 < Jan-> we're talking about embedded hardware here 2019-10-18T02:18:42 < Jan-> it could be anything 2019-10-18T02:18:48 < karlp> and if that's your concern, then just dumping the rom and flashing on another part is the same 2019-10-18T02:18:55 < karlp> lgpl doesn't make that better or wose. 2019-10-18T02:19:02 < zyp> Jan-, how many embedded hardware products are you selling? 2019-10-18T02:19:22 < Jan-> is there no way on stm32 to switch off the ability to read the code back from it? I bet there is. 2019-10-18T02:19:27 < zyp> sure there is 2019-10-18T02:19:34 < Jan-> right 2019-10-18T02:19:54 < deltab> like your power supplies 2019-10-18T02:20:16 < Jan-> exactly 2019-10-18T02:20:19 < zyp> I mean, I've made two products so far that I'm selling 2019-10-18T02:20:21 < deltab> you can't change the firmware on those, only replace it 2019-10-18T02:20:27 < zyp> I'm giving away the source for both 2019-10-18T02:20:29 < Jan-> so why on earth would anyone want to give up their firmware 2019-10-18T02:20:37 < zyp> why not? 2019-10-18T02:20:46 < Jan-> like I say 2019-10-18T02:20:58 < Jan-> and actually, for the sort of stuff we're doing for this project, it's actually a totally plausible thing to get ripped off. 2019-10-18T02:21:01 < deltab> to save time or money 2019-10-18T02:21:06 < zyp> I like giving my customers the ability to customize the firmware if the stock one isn't to their liking 2019-10-18T02:21:16 < Jan-> I mean literally the only reason anyone would ever ask for the firmware is to rip us off. 2019-10-18T02:21:21 < zyp> at this point there's a couple of third party forks of my firmware 2019-10-18T02:21:29 < zyp> with features I didn't write myself 2019-10-18T02:21:46 < Jan-> yeah, um, this is the whole "linux" thing which assumes everyone in the world is a computer scientist. 2019-10-18T02:21:48 < zyp> I'm happy with that, leads to more happy customers and thus more sales 2019-10-18T02:22:06 < deltab> Jan-: do you see yourself as ripping off the DPS makers? 2019-10-18T02:22:07 -!- con3 [~kvirc@154.119.40.174] has joined ##stm32 2019-10-18T02:22:20 < Jan-> the sort of people who would be using these things are not even going to understand what firmware is let alone want to hack it about or be able to. 2019-10-18T02:22:27 < zyp> also, I'm pretty sure all my customers are using the devices with windows 2019-10-18T02:22:29 < Jan-> deltab: no, since we're not even doing that anymore :) 2019-10-18T02:22:36 < zyp> I distribute firmware updates as self-executing .exe files 2019-10-18T02:23:25 < Jan-> the thing is for the most part, forcing people to give away the firmware is totally pointless 2019-10-18T02:23:31 < deltab> Jan-: but you changed the firmware, and you would've backed up the existing firmware if you'd been able to 2019-10-18T02:23:34 < Jan-> the only reason anyone would want it is to rip you off, regardless. 2019-10-18T02:23:37 < zyp> Jan-, I agree with that 2019-10-18T02:23:51 < zyp> I'm not forced to give away my firmware 2019-10-18T02:24:25 < Jan-> well, you are... aren't you? 2019-10-18T02:24:31 < deltab> right, you only make it available as part of your product 2019-10-18T02:24:53 < karlp> deltab: what's DPS here? 2019-10-18T02:24:56 < Jan-> and then anyone can go "oh, lgpl, you have to give it to me" 2019-10-18T02:25:02 < zyp> apart for maybe some libc functions which are unencumbered by licensing, my firmware contains only code I wrote myself, including laks, so I'm free to use it however I please 2019-10-18T02:25:20 < deltab> the thing the LGPL changes about that is that the purchaser can change part of the firmware in the device they bought 2019-10-18T02:25:27 < zyp> also, laks is BSD licensed, not LGPL, so nobody else using laks would have to either 2019-10-18T02:25:57 < zyp> in other words, I'm giving away the source because I think it's beneficial, not because I'm forced to 2019-10-18T02:26:01 < Jan-> deltab: and to do that you have to give them the rest of the firmware it's ... wha? 2019-10-18T02:26:27 < deltab> karlp: DPS5005 and similar programmable power supplies; OpenDPS provides replacement firmware 2019-10-18T02:27:02 < karlp> deltab: you obviously know jan from somewhere else? 2019-10-18T02:27:04 < deltab> but the built-in firmware is not extractable 2019-10-18T02:27:11 < deltab> ##electronics 2019-10-18T02:27:15 < karlp> hah, "not extractable" 2019-10-18T02:27:17 < Jan-> I sort of wish we could get the original firmware and put it back, but we knew what we were doing when we redid it. 2019-10-18T02:27:31 < zyp> Jan-, consider this; is the complexity of your firmware gonna be so high that it matters? 2019-10-18T02:27:43 < Jan-> on this? yes. 2019-10-18T02:27:57 < Jan-> I ran out of performance on an avr. 2019-10-18T02:28:08 < zyp> complexity, as in work to implement 2019-10-18T02:28:16 < Jan-> Yes. 2019-10-18T02:28:19 < Jan-> No question. 2019-10-18T02:28:34 < zyp> I mean, why would the chinese bother copying your code if they can hire a dude to hammer out a replacement in a few days? 2019-10-18T02:28:42 < Jan-> They can't, I promise you. 2019-10-18T02:29:22 < zyp> well, the firmware for my products are that simple 2019-10-18T02:31:03 < karlp> Jan-: it turns out that a lot of customers just want to get their problem solved, and if they can buy it for X, from a company that can supporit it, they just will 2019-10-18T02:31:16 < karlp> even if X-30% is available from somone that can't/won't support it. 2019-10-18T02:31:37 < karlp> and if they weren't going to buy it for X, and will only buy it for X-30%, then it's not even a lost sale. 2019-10-18T02:31:43 < Jan-> probably your firmware isn't going to involve five dimensional lookup tables designed to do calibrated colour control. 2019-10-18T02:31:49 < zyp> karlp, and then you have X-90% like the saleae analyzers :) 2019-10-18T02:31:52 < Jan-> based on extensive measurement of real world cameras and lights 2019-10-18T02:33:23 < karlp> zyp: same thing, I'm _never_ going to buy a salea logic 8. 2019-10-18T02:33:27 < karlp> I'm not a lost sale. 2019-10-18T02:33:45 < karlp> OMG did you say FIVE DIMENSIOANL!!!!"!"#!"#!!!!!!111!1!!!1!!? 2019-10-18T02:33:53 < deltab> Jan-: maybe you can put that code/data in locked memory and allow the rest to be read and changed 2019-10-18T02:34:18 < deltab> including any LGPL code you rely on 2019-10-18T02:34:18 < karlp> (especially when the old logic 8 was the cypress ref design....) 2019-10-18T02:34:25 < Jan-> karlp: the lights have five channels of LEDs. 2019-10-18T02:34:39 < zyp> karlp, I did buy one, still happy with it :) 2019-10-18T02:35:09 < karlp> I'm sure, but I'm also super happy with my "cypress ref design fx2 based logic analyuser" 2019-10-18T02:35:19 < karlp> that I use with sigrok firmware with sigwork host software 2019-10-18T02:35:48 < karlp> I largely resent the implication that I'm somehow "ripping off" saleae that is prevalent 2019-10-18T02:36:06 < zyp> I wouldn't imply that 2019-10-18T02:36:21 < zyp> you would be if you used the saleae software 2019-10-18T02:36:47 < zyp> since you're not, it's a moot point 2019-10-18T02:37:50 < Jan-> One question. 2019-10-18T02:37:56 < Jan-> If I just used libopencm3 and didn't tell anyone 2019-10-18T02:38:00 < Jan-> how the hell would you prove I had 2019-10-18T02:38:10 < Jan-> rather than just use a scratch built HAL 2019-10-18T02:38:24 < karlp> that's between you, your lawyer, and your risk assessment of your customers 2019-10-18T02:38:46 < karlp> zyp: no you, no, but it comes up from some of the haters sometimes .) 2019-10-18T02:39:10 < zyp> Jan-, plenty of companies violate gpl or lgpl that way 2019-10-18T02:39:19 < zyp> or other licenses for that matter 2019-10-18T02:39:35 < deltab> including commercial 2019-10-18T02:39:53 < zyp> I mean, there's a whole list here: https://gpl-violations.org 2019-10-18T02:40:13 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-18T02:40:59 < Jan-> well they're claimed violations 2019-10-18T02:41:02 < Jan-> I've heard of that 2019-10-18T02:41:04 < Jan-> some of it is bullshit 2019-10-18T02:43:16 < bitmask> hmm, how do you add components to rooms in altium 2019-10-18T02:43:46 < zyp> doesn't each schematic sheet usually make a room with components from that sheet? 2019-10-18T02:44:02 < zyp> idk, I tend to avoid using rooms 2019-10-18T02:44:37 < zyp> my understanding is that rooms are mostly useful if you're doing hierarchical designs with repeated instances 2019-10-18T02:44:45 < zyp> and I haven't had a need to do that yet 2019-10-18T02:45:13 < bitmask> ahh, yea I deleted the default room 2019-10-18T02:45:28 < bitmask> I thought it would be useful to hold components for specific things 2019-10-18T02:45:34 < zyp> I usually turn off the «generate rooms» setting 2019-10-18T02:45:45 < bitmask> like one room for the 5V supply, one room for the usb supply and one for the rest 2019-10-18T02:46:13 < zyp> maybe, I personally don't find that useful 2019-10-18T02:46:18 < bitmask> k 2019-10-18T02:47:15 < bitmask> ahh I get it now 2019-10-18T02:47:33 < bitmask> you use queries, now to decide if its worth it 2019-10-18T02:47:47 < karlp> so, did senpai drown in the last .jp floods or what? 2019-10-18T02:48:32 < zyp> probably just busy again 2019-10-18T02:49:01 < karlp> I guess, but been a while longer, and I don't remember a "going to .tw to meet these maker bois" like the last few times. 2019-10-18T02:49:14 < karlp> maybe got involved in freedom or something bad. 2019-10-18T02:51:22 * Jan- puts The Who on the channel stereo at volume = 11 2019-10-18T02:51:24 * Jan- rocks out 2019-10-18T02:51:52 < zyp> karlp, last active about a week ago, not that long 2019-10-18T02:51:56 * karlp goes for tall paul vs billie.... bringing those 90s back one more time.... 2019-10-18T02:53:00 < karlp> zyp: feels like like that's longer than in my memory, but yeah, we're not talking https://imgur.com/gallery/d1FvY long or anything... 2019-10-18T02:54:48 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has quit [Ping timeout: 246 seconds] 2019-10-18T02:55:28 * Jan- won't get fooled again 2019-10-18T02:55:36 * Jan- grabs karlp and bops around the channel with him 2019-10-18T02:57:31 < zyp> karlp, doesn't seem like the typhoon hit his area anyway 2019-10-18T02:58:52 < karlp> zyp: yeah, not as I understand it no, was just trying to be suave and demonstr4ate that I knew they had a typhoon in the area at all :) 2019-10-18T03:04:59 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-18T03:08:11 < jadew> uhmm.. euro symbol looks bad next to prices :/ 2019-10-18T03:09:03 < jadew> it looks like clutter, while the USD one looks like money 2019-10-18T03:09:33 < karlp> wat? 2019-10-18T03:10:31 < jadew> consider this: (+$17) vs (+€17) 2019-10-18T03:10:45 < Tordek> is there anything special about port e? Because I have this: 2019-10-18T03:10:45 < Tordek> HAL_GPIO_TogglePin(LD6_GPIO_Port, LD6_Pin); 2019-10-18T03:10:46 < Tordek> HAL_GPIO_TogglePin(CE_GPIO_Port, CE_Pin); 2019-10-18T03:11:04 < Tordek> one's in B (the led) and one's in E, and the E one won't toggle 2019-10-18T03:11:34 < jadew> maybe it's not set up to be an output? 2019-10-18T03:11:58 < Jan-> is anyone aware of a HAL for stm32 that doesn't require giving away the firmware on demand? 2019-10-18T03:12:00 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-18T03:12:04 < Jan-> I assume the cube one doesn't 2019-10-18T03:12:35 < karlp> sure, cube, mbed, laks, zephyr, nuttx, 2019-10-18T03:12:37 < jadew> lol, the joys of GNU crapware 2019-10-18T03:12:42 < karlp> some versions of chibi 2019-10-18T03:12:55 < Jan-> see mbed looked really good but I just couldn't get it to work 2019-10-18T03:13:16 < Tordek> jadew: it is set as an ouput in cubemx 2019-10-18T03:13:17 < karlp> and again, locm3 doesn't require it either, it just requires "a way to relink the libopencm3 portion" 2019-10-18T03:13:50 < Jan-> I have to ask... why? 2019-10-18T03:15:47 < karlp> because that's what the lgpl actually asks for. 2019-10-18T03:15:58 < Jan-> sure, but why is that useful to anyone? 2019-10-18T03:16:18 < jadew> Jan-, it's a commie thing 2019-10-18T03:16:33 < jadew> you wouldn't understand, because it doesn't make sense 2019-10-18T03:17:07 < Jan-> I was forming that impression. 2019-10-18T03:17:23 < karlp> well.... 2019-10-18T03:17:34 < jadew> but the bottom line is that that's what the original devs wanted, and you have to obey that 2019-10-18T03:17:41 < karlp> imagine that there was a bug in the libopencm3 code 2019-10-18T03:17:50 < Jan-> heaven forfend. 2019-10-18T03:17:53 < karlp> and your vendor didn't offer you a patch 2019-10-18T03:18:00 < karlp> for whatever reason 2019-10-18T03:18:23 < karlp> lgpl jus tsays that you "must" be allowed to relink against your own version of libopencm3 to fix things 2019-10-18T03:19:14 < Jan-> I can absolutely one thousand per cent guarantee you that none of our customers would even understand that concept let alone know how to do it. 2019-10-18T03:19:51 < jadew> if anyone asks, you can just say "hold my beer" and never reply back, because "you're still trying to make the code compile" 2019-10-18T03:20:01 < jadew> is there a time frame for that requirement? 2019-10-18T03:20:21 < Jan-> more than that what on earth is anyone going to do about enforcing that 2019-10-18T03:20:33 < Jan-> the only way you could enforce it is to take someone to court privately 2019-10-18T03:20:45 < Jan-> which is I suspect going to cost more money than most open source projects have. 2019-10-18T03:20:53 < jadew> I think there are organizations that harass people who use GNU code and don't follow the licenses to the letter 2019-10-18T03:20:59 < karlp> Jan-: I don't give a flying _FUCK_ what you abnd your customers think. 2019-10-18T03:21:16 < karlp> the lgpl says one thing, you can choose to comply or not, based on your own legal advice 2019-10-18T03:21:24 < jadew> ^ 2019-10-18T03:21:37 < Jan-> doesn't that strike you as a bit counter productive 2019-10-18T03:21:40 < karlp> wheter that suits you or your business model is compeltely and entirely up to you 2019-10-18T03:21:57 < jadew> personally, I avoid using anything that's not MIT/BSD/etc 2019-10-18T03:22:04 < karlp> but I have negative interest in hearin you whine about shit. 2019-10-18T03:22:16 < Jan-> I'm just asking questions. 2019-10-18T03:22:28 < karlp> no, you'r ejust moaning. 2019-10-18T03:23:07 < Jan-> Look I don't want to argue with you but you have constantly been really unpleasant to me so I will be blunt. If someone asking questions makes you that uncomfortable then you might want to look at your own behaviour. 2019-10-18T03:23:42 < karlp> you bet :) 2019-10-18T03:23:53 < specing> I can confirm that jan is just moaning 2019-10-18T03:24:11 < karlp> Jan-: do remember that I brought this up to try and help you, 2019-10-18T03:24:24 < karlp> if this is something you and your business feel is insurmountable jsut move on. 2019-10-18T03:24:35 < Jan-> In the end it's not really my call. 2019-10-18T03:24:36 < karlp> locm3 can't be changed, an dno amount of moaning can or will change that. 2019-10-18T03:24:56 < karlp> hell, _I_ hate locm3's licensing. and I maintain it. 2019-10-18T03:25:09 < karlp> but pretending that I'm the problem here is precious :) 2019-10-18T03:25:11 < Jan-> If it were me, I'd have a paid licensing option. 2019-10-18T03:25:46 < karlp> you've clearly never worked with a large multi contributor open source project :) 2019-10-18T03:26:12 < Jan-> that sounds like a tacit admission that this wasn't ever really thought about very hard but just sort of happened. 2019-10-18T03:26:39 < jadew> that's how communism works 2019-10-18T03:26:46 < karlp> for the project it was very clearly thought about 2019-10-18T03:26:49 < jadew> it sort of just happens 2019-10-18T03:26:55 < karlp> why do I keep maintaining it? that's a good question. 2019-10-18T03:27:10 < karlp> most of the original contributors wanted it to stay gpl. 2019-10-18T03:27:37 < Jan-> are any of those people professional software engineers 2019-10-18T03:27:58 < jadew> Jan-, GPL kind of makes sense from a programmer's point of view 2019-10-18T03:28:02 < jadew> pro or not 2019-10-18T03:28:25 < jadew> it's a way of saying "If I don't get to make money from it, you shouldn't either!" 2019-10-18T03:28:36 < jadew> "not unless I get something back in return" 2019-10-18T03:28:44 < Jan-> it's also a pretty effective way of saying "nobody gets to make anything from it." 2019-10-18T03:29:05 < specing> GPL is license slang for "request a quote" 2019-10-18T03:29:17 < jadew> yes, personally I don't release much opensource code, but when I do, I usually make it public domain 2019-10-18T03:29:25 < Jan-> jadew: you took the words out of my mouth 2019-10-18T03:29:32 < specing> And there are companies out there making money off GPL code 2019-10-18T03:29:35 < Jan-> I have never ever written any code that went to anyone else that I felt the need to put a license on 2019-10-18T03:29:47 < specing> like AdaCore, who maintains the GNU Ada toolchain 2019-10-18T03:30:02 < Jan-> oh no wait there's some code of mine in an audio framework for c# which is probably NeckbeardLicense 1.0 or something 2019-10-18T03:30:11 < Jan-> but I just don't care 2019-10-18T03:30:20 < Jan-> if someone wants to use it, just let them use it, jeez. 2019-10-18T03:31:49 < jadew> I think you just have to think about it differently, if your business model doesn't involve opensource, then you should see GNU code as off limits 2019-10-18T03:32:08 < Jan-> oh I know that 2019-10-18T03:32:11 < jadew> or... in the case of LGPL libraries, link to them dynamically 2019-10-18T03:32:31 < Jan-> I don't think you can really do that in a microcontroller. 2019-10-18T03:32:33 < jadew> not possible for embedded stuff usually 2019-10-18T03:32:35 < jadew> yeah 2019-10-18T03:33:01 < karlp> Jan-: you keep confusing your position with other people' sposito as beiung the only one. 2019-10-18T03:33:10 < Jan-> so do you :) 2019-10-18T03:33:20 < Jan-> I guess you could do something freaky with a bootloader that linked the code. 2019-10-18T03:33:28 < Jan-> it'd be quite the task 2019-10-18T03:33:31 < karlp> if you say so. 2019-10-18T03:33:44 < karlp> remember, I'm not trying to tell you you have to do anything 2019-10-18T03:34:02 < jadew> karlp, if you dislike the license, is there no way to switch to something that works for embedded stuff? 2019-10-18T03:34:04 < karlp> I said, btw, lcom3 is lgpl, you should be aware of that before you go anywwhere 2019-10-18T03:34:16 < Jan-> so apparently are at least parts of arduino 2019-10-18T03:34:20 < Jan-> and that is a well recognised problem 2019-10-18T03:34:30 < karlp> jadew: you forget, I'm just defacto maintainer, I can't go blindly relicensing things. 2019-10-18T03:34:35 < karlp> large portions are not written by me. 2019-10-18T03:34:36 < jadew> like... do a poll or something and then query all the developers to see how much code has to be rewritten? 2019-10-18T03:34:54 -!- machinehum [~machinehu@184.71.172.142] has joined ##stm32 2019-10-18T03:35:00 < Jan-> what scares me slightly is that there's some very important code that has this problem 2019-10-18T03:35:16 < jadew> probably not worth the effort, but I guess this could be a way to go for projects wanting to adopt more permissive licenses 2019-10-18T03:35:18 < Jan-> basically all the world's video encoders are based on libavformat and libavcodec 2019-10-18T03:35:33 < jadew> tbh, I find LGPL ok for non-embedded programs 2019-10-18T03:35:54 < Jan-> are the really low level parts of stm32 also lgpl? 2019-10-18T03:35:58 < Jan-> I'm thinking of all the register definitions 2019-10-18T03:40:56 < jadew> Jan-, http://marktarver.com/problems.html 2019-10-18T03:41:14 < deltab> if they're in the HAL code from STM, that seems to be covered by 3-clause BSD 2019-10-18T03:41:59 < Jan-> jadew: I know it well, I have expressed basically that opinion lots and it has made me really unpopular :/ 2019-10-18T03:42:11 < jadew> welcome to the club 2019-10-18T03:42:41 < jadew> when I found that article I couldn't believe someone else saw things the way I did 2019-10-18T03:42:57 < Jan-> there are really good modern examples of open source software that everyone bigs up on the basis they are great 2019-10-18T03:43:03 < Jan-> things like blender 2019-10-18T03:43:08 < Jan-> and of course android 2019-10-18T03:43:22 < Jan-> which were either basically given to the open source world free, or basically aren't open source. 2019-10-18T03:43:40 < specing> https://web.archive.org/web/20191009224744/https://dev.to/degoodmanwilson/open-source-is-broken-g60 2019-10-18T03:43:42 < Jan-> I don't think the open source development model has ever once created what I would call a "big project." 2019-10-18T03:43:47 < jadew> I think this phrase pretty much sums opensource: "In fact FOSS is often a recipe for half-baked freebies that displace good software because they are free, but which then go on to soak up everybody’s time by being full of bugs that require ‘support’ (i.e. your time or money) to heal." 2019-10-18T03:44:05 < Jan-> about the only counter example is gimp which was always an open source project, though again you can give it partial credit for being written by funded students. 2019-10-18T03:44:22 < specing> Jan-, jadew "open source" is different to "free software" 2019-10-18T03:44:26 < specing> keep that in mind 2019-10-18T03:44:46 < Jan-> bah it's all the same neckbeards. 2019-10-18T03:44:48 < specing> most open source is also free software, and vice versa. however the politics are different 2019-10-18T03:44:50 < Jan-> and I say that as a wannabe neckbeard. 2019-10-18T03:44:52 < jadew> specing, sure, there's also good proprietary free software 2019-10-18T03:45:08 < jadew> but yeah, I'm aware of the difference 2019-10-18T03:45:27 < deltab> and half-baked freebie proprietary software 2019-10-18T03:45:34 < specing> jadew: proprietary free software? ... what? 2019-10-18T03:45:36 < Jan-> I mean look at all the shit I have been through today. Because there is literally no documentation that says "you need git, gcc, make and openocd. check this out, type this, type make, done." 2019-10-18T03:45:48 < jadew> specing, there's lots of that around 2019-10-18T03:45:49 < Jan-> it is just not documented anywhere and it hardly works right when you get it, and the debug output is miserable. 2019-10-18T03:46:15 < Jan-> I mean the software is almost good but it's just barely usable. 2019-10-18T03:46:18 < jadew> deltab, yeah, but shit software that's proprietary, doesn't get adopted by anyone 2019-10-18T03:46:24 < jadew> so it's not a displacement force 2019-10-18T03:46:47 < jadew> opensource one does, and then those people end up supporting crap, instead of paying a few bucks and getting great software 2019-10-18T03:46:57 < deltab> right, we tend not to hear about it 2019-10-18T03:47:10 < Jan-> I am very interested that the document you linked mentions ubuntu 2019-10-18T03:47:19 < Jan-> which is always pushed as "the linux distro that actually works." 2019-10-18T03:47:24 < Jan-> try putting it on a laptop, even now 2019-10-18T03:47:26 < specing> > shit software that's proprietary, doesn't get adopted by anyone 2019-10-18T03:47:28 < specing> Hahahaha 2019-10-18T03:47:33 < specing> if only that would be true 2019-10-18T03:47:47 < Jan-> five seconds later some neck beard will be telling you that you have to pick your laptop carefully so that the evil closed source drivers aren't a problem. 2019-10-18T03:48:02 * Jan- bonks her head against the monitor screen *thunk* *thunk* 2019-10-18T03:48:30 < specing> Jan-: there are distros with paid support and those will work 2019-10-18T03:48:45 < specing> e.g. RedHat desktop and Suse desktop 2019-10-18T03:48:56 < deltab> Mint has a reputation for good hardware support 2019-10-18T03:49:14 < Jan-> the whole of open source software has a good reputation for various things at various times., 2019-10-18T03:49:21 < Jan-> I still have no idea why. 2019-10-18T03:49:28 < jadew> specing, and the people working on those have 0 incentive to make things easy to use :) 2019-10-18T03:49:37 < jadew> because then people wouldn't pay for support 2019-10-18T03:49:46 < jadew> you don't see windows users paying for support to use windows 2019-10-18T03:50:20 * Jan- points at jadew 2019-10-18T03:50:22 < Jan-> What he said. 2019-10-18T03:50:37 < Jan-> I don't actually think there's that much wrong with most versions of linux. 2019-10-18T03:50:55 < Jan-> The very problem is that there are 101 of them and they're just different enough to be a massive hurt in the hiney. 2019-10-18T03:51:04 < Jan-> This is really really really stupid. 2019-10-18T03:51:12 < jadew> linux is good for many things, but the way in which it works makes it difficult to set up 2019-10-18T03:51:34 < deltab> Jan-: but you get the same thing in, for example, electronics 2019-10-18T03:51:36 < Jan-> I think it'd be fine if there was one version 2019-10-18T03:51:46 < Jan-> and they actually cared about not fucking everyone over every version with massive changes 2019-10-18T03:51:51 < deltab> lots of boards that you could use, all slightly different 2019-10-18T03:52:19 < Jan-> I have said this a lot but I will say it again here: you cannot defeat linux as a learning exercise because it will be different next time. 2019-10-18T03:53:38 -!- talsit [foobar@gromit.mixdown.ca] has left ##stm32 [] 2019-10-18T03:53:56 < Jan-> deltab don't I know it, I should have got a damn nucleo board 2019-10-18T03:54:12 < Jan-> and guess why I didn't, because I listened to a neckbeard who promised it would all be easy. 2019-10-18T03:54:22 < Jan-> and it was a standard open source flustercuck. 2019-10-18T03:55:18 < jadew> you know.. if it's for personal use, nobody will know 2019-10-18T03:55:29 < Jan-> nobody will know what 2019-10-18T03:55:35 < jadew> that you used GPL code 2019-10-18T03:55:40 < jadew> and in what manner 2019-10-18T03:55:47 < jadew> or if you added insults in the comments 2019-10-18T03:56:09 < deltab> you shouldn't pay too much heed to promises like that 2019-10-18T03:56:16 < Jan-> *sigh* the reality is that we are actually hoping this will be a product. 2019-10-18T03:56:24 < deltab> it's not something that can be realistically promised 2019-10-18T03:56:34 < Jan-> and one would hate to go through the staggering amount of bullshit that creating a product will entail, then have a licensing problem. 2019-10-18T03:57:01 < deltab> and there's the human tendency to discount problems we've had when we're eventually successful 2019-10-18T03:57:23 < Jan-> it is all quite annoying given that all the lgpl does for an stm32 hardware access layer is make it really hard for legitimate users 2019-10-18T03:57:32 < Jan-> while leaving the door wide open to troublemakers 2019-10-18T03:57:47 < Jan-> with rules that the project can't easily detect if they're broken, or possibly afford to enforce 2019-10-18T03:57:48 < Jan-> gah. 2019-10-18T03:58:03 < deltab> yeah 2019-10-18T03:58:04 < Jan-> tell me my interpretation here is wrong somehow. 2019-10-18T03:58:09 < kakimir32> take it or leave it 2019-10-18T03:58:45 < deltab> it's idealism versus pragmatism 2019-10-18T03:59:36 * Jan- is very depressed 2019-10-18T04:00:30 < deltab> if your ideal is that anyone can tinker with the things they have (or get someone else to), then you'll want to make things as open as possible, with 'right-to-repair', standard connectors, etc. 2019-10-18T04:01:21 < Jan-> That's a nice enough idea but it falls at the hurdle of who cares 2019-10-18T04:01:22 < deltab> and the GPL and LGPL are one aspect of that 2019-10-18T04:01:25 < Jan-> say we do actually make these lights 2019-10-18T04:01:49 < Jan-> am I going to walk onto a film set and tell a 55 year old gaffer who has never written more than an email on a computer 2019-10-18T04:02:02 < Jan-> "do you want the object files so you can relink the hardware access layer on the microcontroller in this light?" 2019-10-18T04:02:15 < deltab> right, another ideal is that you don't want to be bothered, you just want to maximize how useful your code can be: then you make it public domain or BSD or the like 2019-10-18T04:02:15 < Jan-> it's totally absurd 2019-10-18T04:02:36 < Jan-> the stallman approach requires that everyone on the planet is an expert software engineer 2019-10-18T04:02:43 < Jan-> and that is neither possible nor desirable. 2019-10-18T04:03:01 < deltab> should he be able to take the device to his tech wizard and ask for it to be changed? 2019-10-18T04:03:46 < Jan-> But what are the requirements for that? Said tech wizard needs the build system it just took me a whole afternoon to set up, knowledge of the hardware they don't have, knowledge of my code they don't have 2019-10-18T04:04:16 < Jan-> I mean sure yeah OK I want the source code to our microwave so that I can make it bleep exactly once for every "increase time" button press 2019-10-18T04:04:21 < Jan-> but ffs it's totally impractical 2019-10-18T04:04:49 < Jan-> what microcontroller does it use, is it one time programmable, are the programming headers exposed, is there a development suite for it. 2019-10-18T04:04:51 < Jan-> it's crazy. 2019-10-18T04:04:54 < deltab> that's the ideal; you're giving the pragmatic counterpoint 2019-10-18T04:05:18 < Jan-> I know what you mean but I think there is not much room between that being an ideal, and that being an absurdist fantasy. 2019-10-18T04:07:35 < deltab> that's in the nature of ideals, yes 2019-10-18T04:08:09 < Jan-> If it were me I hope I would be able to look at the current situation and say "hmm this is not working very well." 2019-10-18T04:09:03 < deltab> sure, like a fair amount of the world 2019-10-18T04:09:25 < jadew> Jan-, I'm sure karlp won't come after you if you don't comply with that stuff, he's a nice guy 2019-10-18T04:09:26 < karlp> (late, things happend) jadew: we've done it once in the past, not to relicense, but to at least go to lgpl+linking exception, but there's just a lot of people, and not all of them have the same opinion. a sizeable group felt that even lgpl was a "concession too far" and that it shouldall be gpl still. 2019-10-18T04:09:37 < specing> The moaning continues 2019-10-18T04:09:53 < Jan-> Phil was having a conversation about this with a buddy of his from the post production side of TV 2019-10-18T04:10:04 < karlp> it's owrht pointing out that when libopenstm32/lipbopencm3 started, the cmsis headers were _not_ licensed in a way that permitted reuse 2019-10-18T04:10:06 < Jan-> phil is a cameraman, buddy i san editor 2019-10-18T04:10:10 < karlp> that situation has changed. 2019-10-18T04:10:18 < jadew> karlp, I see, I would have expected the number of people against it to be very small 2019-10-18T04:10:29 < karlp> for many of us, the _premise_ of libopencm3 has dissappeared 2019-10-18T04:10:35 < Jan-> and the editor was saying "hmm I use linux with resolve" (which is an editor) and they discussed how there simply is not a usable finished video editor on linux 2019-10-18T04:10:42 < Jan-> not an open source one. 2019-10-18T04:11:01 < Jan-> there are a few projects to create one... some of them have been going for... fifteen years? and still nothing. 2019-10-18T04:11:16 < Jan-> you have to say guys, I respect the effort, but something is amiss here. 2019-10-18T04:11:27 < deltab> well, something 2019-10-18T04:11:56 < deltab> it's not like there's a deviding line where something is usable or not 2019-10-18T04:12:17 < Jan-> no there's a big grey area but I raise the video editor thing because there is nothing anywher near. 2019-10-18T04:12:59 < Jan-> what is frustrating is that libavcodec and libavformat are probably a better and more flexible codec library than anything that exists in the proprietary world but open source just cannot get it together. it is a crying shame. 2019-10-18T04:13:00 < aandrew> Jan-: I think the arguemnt is not that you must go and offer the gaffer the source, only that you make the source available to the indie group that *does* do interesting things with their lighting and could use the source 2019-10-18T04:13:31 < jadew> aandrew, it's not even about the source 2019-10-18T04:13:43 < karlp> "jadew | you don't see windows users paying for support to use windows" actually, you fucking do 2019-10-18T04:13:46 < jadew> LGPL is meant to be a lot more permissive than GPL and allow closed source stuff to use the library 2019-10-18T04:14:17 < jadew> but in the case of microcontrollers, it's a bit more limiting... 2019-10-18T04:14:18 < aandrew> right, which is why I thought it odd that jan was railing against it, although I admit I haven't read earlier than about 10m of convo 2019-10-18T04:14:30 < aandrew> I disagree, although the points are well made 2019-10-18T04:14:44 < aandrew> sure that micro may be OTP but I can always get another and replace it if I really want to 2019-10-18T04:15:01 < Jan-> I really don't want to come off as an unqualified critic. the whole open source thing is a great idea. I really wish it could succeed more than it does. 2019-10-18T04:15:04 < karlp> Jan-: your gaffer example is you setting up straawmen, 2019-10-18T04:15:12 < aandrew> it's not about the practicality, it's about preventing exactly the shit we are in the middle of these days iwth closed source, welded-shut boxes and a 'fuck you' attitude from the vendor 2019-10-18T04:15:15 < Jan-> I just think that it needs... management, or funding, or both. 2019-10-18T04:15:29 < karlp> you didnt' sell you product to the gaffer 2019-10-18T04:15:31 < aandrew> that's just it though... it doesn't 2019-10-18T04:15:37 < jadew> karlp, ok you do, but not for the same stuff 2019-10-18T04:15:39 < karlp> and of course the gaffer doesn't give a shit 2019-10-18T04:15:53 < aandrew> if there's some weird bug in the lights and you've gone off the reservation, I can take the code and hire a tech wizard to get shit figured out 2019-10-18T04:15:53 < karlp> but the solution vendor or the equipment rental agency does, 2019-10-18T04:15:59 < jadew> karlp, for windows you get companies that hire other companies to manage all their PCs, at once 2019-10-18T04:16:02 < aandrew> hell I'm quite well paid to figure that shit out for people within their own company 2019-10-18T04:16:10 < deltab> Jan-: sure, funding would help, and for a rounded product you need a mix of different skills 2019-10-18T04:16:31 < Jan-> What you need is someone who's willing to say "no, we're not doing that, decision made here's what we're doing and we will stay that course a while." 2019-10-18T04:16:42 < Jan-> I think that's what a lot of open source lacks 2019-10-18T04:16:43 < Jan-> direction 2019-10-18T04:16:46 < aandrew> btw sam adams oktoberfest beer sucks 2019-10-18T04:16:54 < aandrew> that is very, very true 2019-10-18T04:17:09 < aandrew> OSS lacks direction because direction means everyone pushing to a common goal instead of saying "fuck you, I'm forking" 2019-10-18T04:17:18 < Jan-> aandrew: exactly. 2019-10-18T04:17:33 < Jan-> Or just people coming and going and working on it for a short while, and then someone else pulling in another direction etc. 2019-10-18T04:17:50 < aandrew> the linux kernel has the benevolent dictator approach 2019-10-18T04:18:05 < jadew> in other words, open source lacks money 2019-10-18T04:18:09 < aandrew> most successful projects do. leadership is important 2019-10-18T04:18:13 < deltab> companies have people who set goals and enforce them 2019-10-18T04:18:51 < jadew> money is the only thing people don't say "screw you" and go watch TV instead of working on what you tell them to 2019-10-18T04:18:56 < karlp> jadew: closed source lacks money too biatch, look at your own business ;) 2019-10-18T04:18:59 < jadew> *reason 2019-10-18T04:19:05 < jadew> haha karlp 2019-10-18T04:19:12 < karlp> why arenð't you rich selling all you private shhit? :) 2019-10-18T04:19:23 < jadew> I haven't started doing it yet 2019-10-18T04:19:39 < karlp> man, all I ever did was say, "btw, locm3 is lgpl, you should check if that matters for you" 2019-10-18T04:20:08 < aandrew> karlp: oh so YOU started this 2019-10-18T04:20:14 * karlp almost wishes for a fucking feline memem. 2019-10-18T04:20:23 < karlp> no, fucking jan started it by being a fucking troll 2019-10-18T04:20:27 < Jan-> karl I think we're allowed to discuss this like grownups eh :) 2019-10-18T04:20:46 < karlp> no, you'r ejust ranting that certain things aren' tthe way you want them to be. 2019-10-18T04:21:02 < karlp> x is Y. no amount of complaining that Y doesn't suit your business will change X. 2019-10-18T04:21:14 < Jan-> I don't think I'm doing that and nobody else has started being personally unpleasant so I think maybe you could... try to calm down? 2019-10-18T04:21:42 < karlp> and no amount of declaring that Z is the only trutht in business will make anyone give a shit about your opinion on it. 2019-10-18T04:22:25 < Jan-> OK look we're all having a perfectly fine discussion about it and nobody is getting upset except you karl. I've tried to be polite, but please stop cursing at me, I don't deserve it. 2019-10-18T04:22:36 < jadew> you know, there's this idea that things you know to be true, that others don't think their true, are an advantage in business 2019-10-18T04:22:53 < jadew> *they're 2019-10-18T04:23:22 < jadew> there's actually a very clear way of making money off of this 2019-10-18T04:23:36 < Jan-> do tell :) 2019-10-18T04:23:39 < jadew> start making opensource libraries that can't be used without a license 2019-10-18T04:23:43 * karlp wonders why jade isn't rich yet..... 2019-10-18T04:23:48 < jadew> and let the money pour in 2019-10-18T04:23:55 < deltab> ideals have their place too: they give us somewhere to aim for, even if we can never actually arrive there 2019-10-18T04:24:13 < jadew> karlp, I wonder about that too 2019-10-18T04:24:24 < karlp> jadew: so... original bsd style? you can have the code, it offers functyionality you can't get elsewhere, but it's so arcane stil that you need a support contract? 2019-10-18T04:24:33 < karlp> I'm not convinced you can play that card these days. 2019-10-18T04:24:40 < karlp> market pressures and all that. 2019-10-18T04:24:42 < Jan-> I think you can but only in very specialist areas. 2019-10-18T04:24:49 < karlp> what she said :) 2019-10-18T04:24:52 < jadew> no, it's so good you want it and you can have it, if you pay us $5 2019-10-18T04:25:06 < karlp> also, see AGPL.... 2019-10-18T04:25:08 < Jan-> That's actually something that I have thought open source projects should have. 2019-10-18T04:25:16 < jadew> and you don't have to open your source code 2019-10-18T04:25:25 < Jan-> We once omitted a big feature from something because we just couldn't figure out of we were legal or not. 2019-10-18T04:25:38 < Jan-> If we had the opportunity to pay a smallish fee to have someone go "yes you're OK" that would have been great! 2019-10-18T04:26:13 < Jan-> but no, it's just "gpl or the highway" and gpl is not particularly well written nor has it actually been tested in court that much. 2019-10-18T04:26:42 < jadew> karlp, hmm, the wiki article about the AGPL thing is interesting 2019-10-18T04:27:13 < jadew> is it considered that when you use LGPL code in a MCU, are you using the code or are you distributing it? 2019-10-18T04:27:20 < jadew> even tho, you're just distributing the MCU 2019-10-18T04:27:30 < jadew> I guess it applies... 2019-10-18T04:27:40 < deltab> you're distirbuting the code with it, otherwise it won't do anything 2019-10-18T04:28:09 < Jan-> the question we had was about rental equipment 2019-10-18T04:28:13 < Jan-> the gear was being rented 2019-10-18T04:28:17 < Jan-> is that "distribution?" 2019-10-18T04:28:37 < Jan-> nobody has the faintest idea because that use was never foreseen by the 1970s neckbears who dreamt all this up. 2019-10-18T04:29:23 < deltab> yeah, it's somewhat like the situation with cloud services 2019-10-18T04:29:47 < Jan-> there's something else about that situation 2019-10-18T04:30:06 < Jan-> we were using an ffmpeg executable and driving it as such 2019-10-18T04:30:14 < deltab> Amazon etc. can take GPL software and run it on their servers and sell the services to us 2019-10-18T04:30:17 < Jan-> because if we'd linked it we'd have become subject to bits of GPL. 2019-10-18T04:30:26 < Jan-> so you can use it, you can implement the feature 2019-10-18T04:30:31 < Jan-> so long as you do it really really badly. 2019-10-18T04:35:33 < karlp> jadew: re lgpl on mcus, it's gross and "consult with counsel" 2019-10-18T04:37:09 < karlp> deltab: amazone and services is what agpl is "for" 2019-10-18T04:37:14 < Jan-> I guess it's possible in principle to write code that would link existing code with newly downloaded code. 2019-10-18T04:37:21 < karlp> it's not really in ther same direction at all as lgpl on embedded. 2019-10-18T04:37:34 < Jan-> Which would be an insane bit of engineering but possible. And then you'd be able to avoid giving away the firmware. 2019-10-18T04:37:35 < karlp> Jan-: not just "in principle" 2019-10-18T04:38:03 < karlp> you just say, "here's some .os, and a makefile, it's completely on you to make sure that all apis are the same" 2019-10-18T04:38:19 < Jan-> well what you're trying to do is avoid giving way the objects. 2019-10-18T04:38:41 < karlp> it really is "gcc -o newelf.o [private.o's] [public_rebuilt_lib.a]" 2019-10-18T04:38:56 < jadew> Jan-, problem is you can't guarantee that thing is going to be available forever 2019-10-18T04:39:14 < Jan-> well sure but after we've gone bust who're they gonna sue 2019-10-18T04:39:15 < karlp> you can get nerdily into whether your .o is just a .bin loader, that's irrelevant. 2019-10-18T04:39:35 < karlp> all lpgl says is, "somehow, allow the lgpl portions to be relinked" 2019-10-18T04:39:49 < Jan-> anyway I need to sleep 2019-10-18T04:39:55 < karlp> it doesn't say it has to be _easy_, it just stays it has to be _pssible_ 2019-10-18T04:40:11 < Jan-> earlier on I did talk to someone who said that all his professional work just hit the registers directly 2019-10-18T04:40:15 < Jan-> just like you would writing AVR C 2019-10-18T04:40:18 < Jan-> so I guess I'll just have to do that 2019-10-18T04:40:30 < Jan-> which means setting up *another* build environment to do that :/ 2019-10-18T04:40:34 < karlp> and we've told you (two days ago iirc, perhaps 3) exactly how to do that. 2019-10-18T04:41:00 < Jan-> grr 2019-10-18T04:41:01 < Jan-> makefiles 2019-10-18T04:42:24 < karlp> no-one said makefiles. 2019-10-18T04:42:28 < Jan-> well it'll need one somehow :) 2019-10-18T04:42:29 < karlp> cmsis is a set of .h files. 2019-10-18T04:42:35 < karlp> you can write your own, 2019-10-18T04:42:42 < karlp> and no-one says you hve to use make. 2019-10-18T04:42:49 < Jan-> no karl YOU can write your own :) 2019-10-18T04:43:04 < karlp> meson/ninja/cmake/gradle/gyp/blah 2019-10-18T04:43:24 < karlp> scons too, probably msbuild, ant, pick your poison 2019-10-18T04:43:32 < karlp> they can all be used to build stuff if you want 2019-10-18T04:43:42 < karlp> you've just got to decide what you want to spend your available hours on. 2019-10-18T04:44:03 < Jan-> none of this would worry me if it was documented anywhere. 2019-10-18T04:44:22 < Jan-> it's the thousand character gcc command lines that alarm me 2019-10-18T04:44:35 < karlp> so read the gcc docs 2019-10-18T04:44:42 < karlp> stop moaning that they're not documented 2019-10-18T04:44:44 < karlp> they are. 2019-10-18T04:45:01 < karlp> and stop pretending that platformio is the only way it can be done 2019-10-18T04:45:12 < Jan-> Who on earth is pretending that? 2019-10-18T04:45:16 < karlp> you? 2019-10-18T04:45:16 < Jan-> I just did it without. 2019-10-18T04:45:28 < Jan-> What on earth are you talking about? 2019-10-18T04:45:50 < karlp> you're using long platformio command lines as "proof" that "everything needs thousand char gcc command lines" 2019-10-18T04:45:59 < Jan-> am I? 2019-10-18T04:46:17 < karlp> apparentlt only to me :) 2019-10-18T04:46:28 < Jan-> I think it's possible you're getting a bit carried away with your ideas there. 2019-10-18T04:46:35 < deltab> Jan-: 80% of that command was listing directories to search for include files 2019-10-18T04:46:57 < Jan-> well it still has to be there... 2019-10-18T04:47:13 < deltab> 10% of it was your name in those directory names :-) 2019-10-18T04:47:33 < deltab> right, which is why everyone uses make or the like 2019-10-18T04:47:49 < deltab> put that stuff in a makefile and then you don't have to keep seeing it 2019-10-18T04:47:51 < Jan-> well that has to be set up. 2019-10-18T04:48:02 < deltab> sure, like everything 2019-10-18T04:48:14 < Jan-> I guess I can google around to see if anyone has some predefined stuff for doing "just cmsis" stm32 work. 2019-10-18T04:48:45 < deltab> it'll be predefined for what they're doing, which won't quite match what you need 2019-10-18T04:48:59 < Jan-> sure I guess but I got close enough with libopencm3 2019-10-18T04:49:05 < Jan-> that it worked at least 2019-10-18T04:49:14 < karlp> platform io worked too... 2019-10-18T04:49:22 < karlp> your threshold seems low :) 2019-10-18T04:49:33 < Jan-> I suppose I could just use the build system I have and just not refer to any of the library stuff. 2019-10-18T04:49:43 < Jan-> though I would be betting that it didn't compile any of it in anyway. 2019-10-18T04:49:49 < Jan-> which would be tough to prove. 2019-10-18T04:49:51 < karlp> again, it would be nice ifyou could clarify what you'd _like_ so we could give you better answers 2019-10-18T04:49:58 < deltab> yes; it just had the disadvantage that only arduino worked out of the box 2019-10-18T04:50:28 < Jan-> anyway I really need to sleep I'm working in a shockingly small number of hours 2019-10-18T04:50:56 < deltab> which I believe is mainly a failing in platformio's board-to-framework mapping table 2019-10-18T04:51:11 < Jan-> what platformio doesn't seem to have is a framework called "cmsis" 2019-10-18T04:51:18 < Jan-> so you just get the register definitions and not much else 2019-10-18T04:51:29 < Jan-> that doesn't seem to be an option 2019-10-18T04:52:02 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 268 seconds] 2019-10-18T04:52:06 < karlp> (that wouldn'ðt help platformio justify it's existance) 2019-10-18T04:52:13 < deltab> as I understand it, platformio just has a big list of boards and frameworks and what compiler settings they need 2019-10-18T04:52:37 < Jan-> I don't get why it needs to exist really if these boards are all as similar as people say, but whatever. 2019-10-18T04:53:11 < Jan-> anyway I'm going to idle later 2019-10-18T04:53:12 < deltab> you make a config file that states the board and framework, run a short standard platformio command, and it looks up what stuff is needed to build 2019-10-18T04:53:26 < deltab> because everything has differences 2019-10-18T04:53:38 < deltab> different compilers, different include files, different programmers 2019-10-18T04:54:16 < deltab> make a big database of the differences and you can in theory save most people from having to learn them 2019-10-18T04:54:58 < deltab> the downside is they don't know how to fix it when the database doesn't have an entry for their exact situation 2019-10-18T04:59:34 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-18T06:08:42 -!- machinehum [~machinehu@184.71.172.142] has quit [Ping timeout: 246 seconds] 2019-10-18T06:11:04 < jadew> do you know any sites (other than mouser) that allow you to change the currency? 2019-10-18T06:11:33 < jadew> I'm looking for layout ideas (not sure where to put the button/link/whatever) 2019-10-18T06:16:39 < deltab> top right? 2019-10-18T06:17:03 < deltab> with language or country selection, if you have those 2019-10-18T06:17:09 < jadew> yeah, that's what I'm trying now, but I'm not convinced 2019-10-18T06:17:16 < deltab> or near the first price 2019-10-18T06:17:22 < jadew> just english 2019-10-18T06:18:07 < jadew> it's probably not a bad idea to emulate mouser, since people will be used to it 2019-10-18T06:18:17 < deltab> I checked for a Nielsen Norman Group article, but didn't find one 2019-10-18T06:18:22 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T06:18:24 < jadew> but I'm not sure how it will work, especially in mobile layout 2019-10-18T06:19:17 < jadew> interesting, someone made a business out of it 2019-10-18T06:19:23 -!- rue_mohr [~rue_mohr@d50-92-152-244.bchsia.telus.net] has joined ##stm32 2019-10-18T06:19:27 < rue_mohr> hey 2019-10-18T06:19:40 < deltab> ideally, try it with a mix of people and see how they get on 2019-10-18T06:19:51 < jadew> and I was only able to charge $50 last week, for a UX report 2019-10-18T06:20:25 < rue_mohr> hello 2019-10-18T06:20:30 < jadew> hey rue 2019-10-18T06:20:37 < jadew> deltab, yeah, I don't know that many people 2019-10-18T06:20:44 < rue_mohr> so, I have an issue, trying to compile a multi-source project for stm32 2019-10-18T06:21:03 < rue_mohr> I'm using the "new" oooo makefile 2019-10-18T06:21:23 < rue_mohr> http://paste.debian.net/1107875/ 2019-10-18T06:21:26 < rue_mohr> as per^^ 2019-10-18T06:21:38 < rue_mohr> I think were using the linker wrong 2019-10-18T06:21:53 < deltab> jadew: you could try A/B testing, if you have enough traffic 2019-10-18T06:21:58 < rue_mohr> usr/bin/arm-none-eabi-objcopy: error: the input file 'program.elf' has no sections 2019-10-18T06:22:22 < jadew> deltab, I don't 2019-10-18T06:22:57 < rue_mohr> hmm, well this worked with a 1 file project... 2019-10-18T06:23:34 < rue_mohr> http://paste.debian.net/1107877/ 2019-10-18T06:23:53 < rue_mohr> can anyone see something obvious? 2019-10-18T06:25:14 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-18T06:25:58 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T06:26:53 -!- fc5dc9d4 [~quassel@p5B081616.dip0.t-ipconnect.de] has joined ##stm32 2019-10-18T06:28:02 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-18T06:29:46 < deltab> jadew: probably just go with what mouser does, then 2019-10-18T06:31:14 < rue_mohr> ok, the only difference I see between the old computer that works, and the new one 2019-10-18T06:31:18 < rue_mohr> -D_ROM=64K -D_RAM=20K -DSTM32F1 -D_ROM_OFF=0x08000000 -D_RAM_OFF=0x20000000 2019-10-18T06:31:19 -!- fc5dc9d4_ [~quassel@p57A32EB8.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-18T06:31:20 < rue_mohr> vs 2019-10-18T06:31:26 < rue_mohr> -DSTM32F1 2019-10-18T06:31:27 < deltab> jadew: oh, to find more sites try searching for "select currency" and the like 2019-10-18T06:31:47 < rue_mohr> thoughts? 2019-10-18T06:33:46 < deltab> rue_mohr: examine the program.elf file: is it in the right format? 2019-10-18T06:34:16 < rue_mohr> zippo:/morfiles/programming/c/stm32/filterprof# cat program.elf 2019-10-18T06:34:16 < rue_mohr> ELF.... 2019-10-18T06:34:35 < rue_mohr> whats the thing for getting the details? 2019-10-18T06:35:02 < deltab> objdump? 2019-10-18T06:35:47 < rue_mohr> program.elf: file format elf32-little 2019-10-18T06:35:47 < rue_mohr> objdump: can't disassemble for architecture UNKNOWN! 2019-10-18T06:36:20 < deltab> hmm, those -D options look like they'd set up the memory map 2019-10-18T06:36:41 < rue_mohr> the new linker script isn't addign them 2019-10-18T06:38:01 < rue_mohr> oh, it might have been an issue with -rw-rw-rw- 1 1005 1005 0 Oct 17 20:07 stm32f103c8t6.ld 2019-10-18T06:38:19 < rue_mohr> yup 2019-10-18T06:38:21 < rue_mohr> hmm 2019-10-18T06:39:11 < rue_mohr> hmm 2019-10-18T06:39:27 < rue_mohr> the new makefile isn't properly copying? that file 2019-10-18T06:43:11 < rue_mohr> one of the linker scripts seems to be breaking 2019-10-18T06:53:19 < rue_mohr> I have no idea where genlink.py comes from 2019-10-18T06:53:31 < rue_mohr> I think I'm dealing with an out of date debian package here 2019-10-18T07:09:02 < rue_mohr> ok, opencm3 update fixed it 2019-10-18T07:09:08 < rue_mohr> how amny bits on this adc? 2019-10-18T07:10:59 < rue_mohr> 12 ok 2019-10-18T07:11:16 < rue_mohr> and as I'm the only person awake in this channel, farewell! 2019-10-18T07:11:18 -!- rue_mohr [~rue_mohr@d50-92-152-244.bchsia.telus.net] has left ##stm32 ["Leaving"] 2019-10-18T07:36:25 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-18T07:39:06 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-18T07:39:41 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 265 seconds] 2019-10-18T07:39:41 -!- day__ is now known as day 2019-10-18T07:53:49 < antto> oh wow, mr rue_m0hr o_O 2019-10-18T08:24:46 < qyx> who is mr rue mohr? 2019-10-18T08:29:33 < antto> imma tell u lat0r, if ya really wanna know 2019-10-18T08:29:49 < antto> but.. u don't 2019-10-18T09:15:41 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-18T09:18:46 < emeb_mac> murders in the rue morgue 2019-10-18T09:19:19 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-18T09:39:14 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has joined ##stm32 2019-10-18T09:48:39 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-18T09:49:03 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T10:01:16 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-18T10:08:31 -!- con3 [~kvirc@154.119.40.174] has quit [Ping timeout: 252 seconds] 2019-10-18T10:16:23 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-18T10:18:52 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-18T10:25:45 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-18T10:59:45 < dongs> is there such thing a PCIe m.2 ssd that is NOT nbme? 2019-10-18T10:59:48 < dongs> nvme 2019-10-18T11:15:37 < dongs> anyway, looks liek my china nvme>usb3 thing actually worked 2019-10-18T11:15:47 < dongs> and this ancient samsung i tried in it is not compatible 2019-10-18T11:21:15 < dongs> of course/ 2019-10-18T11:22:09 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-18T11:23:28 < dongs> i was at this IdiOT expo, they had something called "Intelligent Racket" 2019-10-18T11:29:15 < Steffanx> Can you juggle with those Haohmaru ? 2019-10-18T11:37:59 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Read error: Connection reset by peer] 2019-10-18T11:40:38 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-18T11:46:23 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-18T11:46:23 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-18T11:46:23 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-18T11:47:14 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Ping timeout: 240 seconds] 2019-10-18T11:49:14 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-18T12:06:29 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Ping timeout: 265 seconds] 2019-10-18T12:09:49 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Quit: ZNC 1.6.1 - http://znc.in] 2019-10-18T12:10:03 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has quit [Quit: Leaving] 2019-10-18T12:10:27 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-18T12:10:27 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-18T12:10:28 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-18T12:48:44 < karlp> dongs: what would an m2.ssd that wasn't nvme look like? 2019-10-18T12:48:49 < karlp> also, welcome bac k:) 2019-10-18T12:48:52 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-18T12:57:39 -!- via [~via@vtluug/member/via] has quit [Ping timeout: 240 seconds] 2019-10-18T12:58:42 -!- via [~via@vtluug/member/via] has joined ##stm32 2019-10-18T13:03:15 < Steffanx> Haohmaru: fyi I can :) 2019-10-18T13:05:27 < Steffanx> Dongs wants a pcie to m.2 sata card? 2019-10-18T13:23:31 < dongs> karlp: apparently, AHCI 2019-10-18T13:23:34 < dongs> karlp: and PCIe 2019-10-18T13:23:37 < dongs> found review for that samsung model 2019-10-18T13:23:40 < dongs> thats what it is 2019-10-18T13:23:53 < karlp> what's the difference between pcie and nvme? 2019-10-18T13:24:01 < dongs> ahci and nbme 2019-10-18T13:24:05 < dongs> protocol i guess 2019-10-18T13:24:19 < dongs> https://www.anandtech.com/show/8006/samsung-ssd-xp941-review-the-pcie-era-is-here 2019-10-18T13:24:36 < karlp> ahci and nvme both being on top of pcie though? 2019-10-18T13:24:36 < day> i've got that one 2019-10-18T13:24:42 < day> yeah 2019-10-18T13:25:06 < dongs> karlp: yes, but my usb3 bridge only supports NVMe as the only drives that did AHCI were very old ones (like this asmsung) and it didnt last long 2019-10-18T13:25:43 < dongs> "In terms of the software interface, the XP941 is still AHCI based but Samsung does have an NVMe based SSD for the enterprise market. I would say that AHCI is a better solution for the consumer market because the state of NVMe drivers is still a bit of a question" 2019-10-18T13:26:22 < karlp> is that not just sata? 2019-10-18T13:26:26 < dongs> no 2019-10-18T13:26:38 < dongs> its pcie 4x with AHCI being the protocol 2019-10-18T13:27:02 < dongs> https://www.akitio.com/blog/articles/nvme-vs-ahci-pcie-ssd apparently it was a thing until like 2016 or so while they still made ahci versions 2019-10-18T13:27:56 < karlp> I just remember my sata controller having options of, "legacy" and "ahci" 2019-10-18T13:28:08 < dongs> thats beacuse ahci covers everything disk related 2019-10-18T13:28:14 < dongs> spinning drives were ahci too 2019-10-18T13:28:16 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 264 seconds] 2019-10-18T13:28:17 < karlp> right 2019-10-18T13:28:30 < karlp> so for a while there was ahci on m2/ngff cards then, like you got. 2019-10-18T13:28:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-18T13:28:39 < karlp> ok, good to know I guess :) 2019-10-18T13:28:56 < dongs> completely irrelevant now, I dont think there's any pcie drives made anymore 2019-10-18T13:29:01 < dongs> pcie + ahci 2019-10-18T13:29:22 < dongs> i just got this usb3>nvme adapter thingy from chinagirl about a year ago and was gonna try it but the only m2 drive i had was that samsung 2019-10-18T13:29:42 < dongs> plugged it in, nothing happened, gave up thought the adapter was trash 2019-10-18T13:30:00 < dongs> pal gave me a 128gb nvme drive and that worked right away 2019-10-18T14:00:17 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-18T14:19:52 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-18T14:19:58 < karlp> well this is lame, freertos internal stats don't match my stats from their trace 2019-10-18T14:24:30 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Remote host closed the connection] 2019-10-18T14:37:45 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-18T14:41:16 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-18T14:48:17 -!- Sadale [~Sadale@unaffiliated/sadale] has quit [Ping timeout: 276 seconds] 2019-10-18T14:48:50 -!- Sadale [~Sadale@unaffiliated/sadale] has joined ##stm32 2019-10-18T15:44:21 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-18T15:57:48 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-18T16:04:15 < kakimir32> proTalk 2019-10-18T16:07:13 < doomba> Title: kakimir32.exe has been detected. Risk: High Status: Fully Resolved Action: [Fix / *Quarantine / Remove] 2019-10-18T16:07:43 < kakimir32> installing troijan dowloader 2019-10-18T16:10:49 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-18T16:20:22 < dongs> karlp: you ended up going with failRTOS after all? 2019-10-18T16:20:28 < dongs> how can you deal wiht thier fucktastic naming 2019-10-18T16:20:30 < dongs> xFuckOFF 2019-10-18T16:21:18 < kakimir32> https://drive.google.com/file/d/1EQ1eVcTH4r-runFZOdEEYCga5Ds3zgdh/view?usp=sharing what is happening with R2? it loads value from 0x7490. That value is 0x3000200 then it tries to load value from 0x3000200 and mem man handler 2019-10-18T16:21:32 < dongs> > mcuxpresso lol 2019-10-18T16:21:37 < dongs> is this what tehy call exclipse now? 2019-10-18T16:21:47 < kakimir32> nice 2019-10-18T16:22:02 < karlp> dongs: well, it's more pleasant build and infrastructure wise than cmsis-rtos2 is, 2019-10-18T16:22:26 < dongs> kakimir32: your indentation/style is fucking atrocious 2019-10-18T16:22:33 < dongs> you are bad and you should feel bad 2019-10-18T16:22:45 < dongs> Haohmaru: it says right in teh title MCUXPRESSO some NXP clone of eclipse i guess 2019-10-18T16:23:10 < dongs> oh lul, is this lunix? 2019-10-18T16:23:11 < dongs> aids. 2019-10-18T16:23:24 < dongs> lol @ using guis on lunix 2019-10-18T16:23:41 < dongs> kikemir, im gonna guess by the time that function is called the pointer to send_data is gone 2019-10-18T16:23:50 < kakimir32> dongs: what you see in editor is pretty much straight copy from uart rom api example code 2019-10-18T16:23:51 < dongs> so strlen fails 2019-10-18T16:24:05 < kakimir32> dongs: it actually didn't 2019-10-18T16:24:28 < kakimir32> value in R0 is what it returned 2019-10-18T16:25:33 < dongs> so what do you thin kis the problem 2019-10-18T16:26:22 < dongs> thats mr. kikemir 2019-10-18T16:27:18 < karlp> pay attention then :) 2019-10-18T16:27:22 < kakimir32> dongs: the value it loads from 0x7490 to R2 2019-10-18T16:29:00 < dongs> at which point does it fail? 2019-10-18T16:29:10 < dongs> does UART_PARAM_T have anything otehr than those 2 members? 2019-10-18T16:29:11 < kakimir32> one stepi from this current view 2019-10-18T16:29:35 < kakimir32> yes 2019-10-18T16:30:29 < BrainDamage> jadew: you're mistaken on GNU licenses wrt personal usage, you're not forced to make the source available if you use it yourself, you're forced to provide the code if you distribute your program 2019-10-18T16:30:42 < BrainDamage> jadew: so self usage is perfectly fine 2019-10-18T16:30:53 < dongs> i would start off by zeroing the struct just in case, _T param = { 0, }; 2019-10-18T16:31:15 < dongs> does it fail on 1st call or soemtime down the line 2019-10-18T16:31:26 < kakimir32> first 2019-10-18T16:31:40 < dongs> are you out of stack lol 2019-10-18T16:31:47 < dongs> tho that struct is only like 10 bytes 2019-10-18T16:32:20 < dongs> well, 16-ish but close enough 2019-10-18T16:32:40 < kakimir32> not likelly dongs. see the tasks on bottom right 2019-10-18T16:33:25 < kakimir32> but those gauges show whatever they want 2019-10-18T16:34:05 < jadew> BrainDamage, I didn't claim otherwise 2019-10-18T16:34:08 < dongs> ovencontrol.c reporting you to hitler right now 2019-10-18T16:34:11 < jadew> in fact, that's what I suggested too 2019-10-18T16:34:45 < BrainDamage> jadew: ah, I interpreted what you said as "it's illegal but cannot be checked", while it's legal, even if it can't be checked 2019-10-18T16:37:13 < jadew> actually, I didn't know it's legal 2019-10-18T16:37:27 < jadew> you understood right 2019-10-18T16:37:38 < karlp> (l)gpl stuff only triggers on distribution 2019-10-18T16:37:44 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-18T16:38:18 < karlp> it's why they had to and get angry and make agpl things when it was discoverd that running it on your own hardware and offering it as a service wass't distribution 2019-10-18T16:38:43 < jadew> ha 2019-10-18T16:48:06 -!- con3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-18T16:52:20 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Quit: Textual IRC Client: www.textualapp.com] 2019-10-18T16:52:48 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2019-10-18T16:56:04 < Thorn> has anyone tried stm32mp1 for real yet? how horrible is the lunix? is it better than rpi overall? 2019-10-18T16:58:45 < zyp> No. Guess it depends which lunix you put on it. What metric do you use for better? 2019-10-18T16:59:30 < Thorn> can you put anything other than the ST distro on it? 2019-10-18T16:59:43 < zyp> I sure hope so 2019-10-18T17:02:16 < zyp> it's also kinda weird to compare rpi to a SoC 2019-10-18T17:03:04 < zyp> the modern rpis also have a lot more cpu power, doesn't mp1 only have a single A7 core? 2019-10-18T17:04:58 < Thorn> 1 or 2 a7 cores afaik. not sure if the dual core version is available 2019-10-18T17:10:12 < Thorn> I don't see any SoM for sale anywhere though so there is no realistic way to compare it to the rpi som yet 2019-10-18T17:12:03 < aandrew> the best buys here in the US stock 2GB rpi4s 2019-10-18T17:13:48 < BrainDamage> rpi also exist mostly through a contract with broadcom 2019-10-18T17:13:55 < aandrew> yep 2019-10-18T17:14:12 < jadew> are you guys aware of something like a business channel? I have noob questions 2019-10-18T17:14:30 < aandrew> hm? 2019-10-18T17:15:23 < jadew> customer asked me for schematics, I refused, then he asked me for a design detail that I'm not really sure he needs to know about 2019-10-18T17:15:33 < jadew> not sure how to answer to an inquiry like that 2019-10-18T17:16:10 < zyp> what's the product? 2019-10-18T17:16:32 < jadew> an 8 GHz prescaler 2019-10-18T17:17:03 < zyp> a what? 2019-10-18T17:17:04 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-18T17:17:15 < jadew> https://cojotech.com/pmo-080-8-ghz-replacement-kit-input-c 2019-10-18T17:17:19 < BrainDamage> a device that divides a signal in frequency 2019-10-18T17:17:29 < BrainDamage> typically used in frac-n pll 2019-10-18T17:17:55 < BrainDamage> or in that case, a freq counter 2019-10-18T17:18:13 < zyp> ah, right 2019-10-18T17:19:39 < BrainDamage> jadew: nice site, I dig the aesthetics 2019-10-18T17:19:53 < bitmask> hello 2019-10-18T17:19:55 < zyp> but I mean, if customers are being unreasonable, you say no 2019-10-18T17:19:56 < jadew> BrainDamage, thanks, it was all me :) 2019-10-18T17:20:03 < zyp> if they don't accept a no, you ignore them 2019-10-18T17:21:04 < sync> jadew: tbh one can always lift the can and see what is going on 2019-10-18T17:21:38 < jadew> sync, I go to great lengths to make that difficult 2019-10-18T17:21:51 < jadew> tbh, I wouldn't mind customers taking a peak 2019-10-18T17:21:57 < jadew> or even sharing the schematics with them 2019-10-18T17:22:24 < jadew> but I know that if they post a picture on some blog, in 2 weeks there'll be a chinese clone selling for $20 less 2019-10-18T17:22:30 < bitmask> jadew, you are cojotech? 2019-10-18T17:22:35 < jadew> bitmask, yes 2019-10-18T17:22:38 < bitmask> neat site 2019-10-18T17:22:38 < aandrew> jadew: what specifically are they asking? 2019-10-18T17:22:49 < sync> the chinacloners just can buy one and clone it themselves, if they really wanted 2019-10-18T17:22:55 < kakimir64> #define LPC_ROM_API_BASE_LOC 0x03000200UL 2019-10-18T17:23:10 < jadew> sync, I don't ship to China tho 2019-10-18T17:23:15 < jadew> bitmask, thanks 2019-10-18T17:23:55 < zyp> jadew, I don't think that would stop any determined cloner 2019-10-18T17:24:20 < aandrew> yeah that definitely won't stop someone who wants to clone 2019-10-18T17:24:27 < bitmask> do you guys have a rule of thumb when placing components like how far from the mcu should res/caps be and stuff? I'm not sure on my spacing 2019-10-18T17:24:37 < aandrew> but let's face it, the market for frequency counter front ends isn't high 2019-10-18T17:24:41 < jadew> zyp, sure, but the clones for the other ones that are there, were only possible because the schematic and pictures were available 2019-10-18T17:24:49 < aandrew> so the risk is minimal, but taking precautions is always good 2019-10-18T17:25:36 < jadew> bitmask, you place decoupling caps as close as possible, in the order of small to big 2019-10-18T17:25:49 < bitmask> what is close as possible though 2019-10-18T17:26:10 < bitmask> you gotta have some space between the mcu pins and the caps 2019-10-18T17:26:11 < aandrew> bitmask: I put 100nf caps directly under most ICs, and if there are sensitive analog bits and I'm using 1nf/10nf they go closest, then 100nf 2019-10-18T17:26:16 < jadew> bitmask, whatever distance allows you to manufacture to board easily 2019-10-18T17:26:29 -!- kakimir6486 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-18T17:26:32 < jadew> *the 2019-10-18T17:26:41 < Thorn> bitmask: leave just enough space for your soldering tip to fit 2019-10-18T17:26:42 < aandrew> generally speaking you keep high impedance shit away from noise 2019-10-18T17:26:46 < bitmask> under the board, hmm I wasnt thinking about that 2019-10-18T17:26:52 < bitmask> I was keeping everything on top 2019-10-18T17:27:30 < aandrew> bitmask: most of my designs are 4+L and ViP which keeps inductance low (and there's a gigantic (size) capcitor between the top and bottom layers 2019-10-18T17:27:32 < Thorn> bitmask: you should probably keep them on top unless you're routing BGAs 2019-10-18T17:27:47 < aandrew> Thorn: hm? 2019-10-18T17:27:55 < bitmask> this is my first bare mcu project so im not really sure what im doing but this is what I came up with just putting stuff in the general area: https://i.imgur.com/L7G9Olj.png 2019-10-18T17:28:17 < aandrew> the power supplies all drill down to one of the inner layers anyway, stick a cap across the vias on the bottom for lowest inductance and shortest path 2019-10-18T17:29:03 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-18T17:29:20 < aandrew> bitmask: you're doing a lot of effort on the silk before it's time. not bad otherwise though 2019-10-18T17:29:25 < aandrew> is there any reason to keep the bottom clean? 2019-10-18T17:29:31 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-18T17:29:41 < bitmask> no reason 2019-10-18T17:30:04 < aandrew> assembly is cheaper/faster if nothing on bottom, but that may not be an issue 2019-10-18T17:30:11 < kakimir6486> I narrowed it down to rom api call 2019-10-18T17:30:19 < bitmask> nah its a one off, just for me 2019-10-18T17:30:26 < kakimir6486> maybe my MPU port is actually working 2019-10-18T17:30:39 < bitmask> I just havent really seen a reason not to keep it all on top, placement seems to work so far 2019-10-18T17:31:18 < bitmask> havent started on the two smps though, that could cause problems 2019-10-18T17:31:34 < bitmask> thats what im most worried about 2019-10-18T17:31:49 < Thorn> bitmask: stm32 pinouts let you palce 0.1µF decoupling caps in the corners of the package while keeping them as close to Vdd pins as possible, e.g. https://i.imgur.com/bPHzH0K.png 2019-10-18T17:32:50 < bitmask> ahh the corners are nice, at least the atmega has the vcc/gnd pins next to each other 2019-10-18T17:33:05 < aandrew> bitmask: yeah placement seems pretty clean so far 2019-10-18T17:33:30 < bitmask> thanks 2019-10-18T17:33:31 < aandrew> I probably spend the most time on placment in power supplies 2019-10-18T17:34:22 < karlp> "doh, why isn't my bit set: trace_send8(STIM, 1<<8 | last_tid);" 2019-10-18T17:34:27 * karlp lurns to count again. 2019-10-18T17:34:58 * karlp thought shift overflow would handle that, but no... 2019-10-18T17:35:10 < karlp> sorry, default warnings on shift overflows 2019-10-18T17:35:28 < jadew> are there shift overflow warnings? 2019-10-18T17:35:49 < jadew> I don't remember seeing one 2019-10-18T17:35:50 < bitmask> the most ridiculous thing I cant figure out is a power switch... the board can draw up to 20A and so I'm not sure if I want to go with a relay and a small rocker switch or just a 20A rocker switch. I'm trying to keep this board/case as small as possible, or at least as thin as possible, and a 20A switch is neither of those 2019-10-18T17:36:27 < BrainDamage> if you don't need isolation, a mosfet may fit the bill 2019-10-18T17:36:34 < BrainDamage> without the large size of a relay 2019-10-18T17:37:04 < BrainDamage> and unlike a relay, it's pretty flush 2019-10-18T17:37:16 < bitmask> hmm 2019-10-18T17:37:33 < karlp> yeah, even with Wshift-overflow=2 it doesn't cach it. 2019-10-18T17:38:00 < BrainDamage> bonus point also, it's not through hole 2019-10-18T17:38:10 < BrainDamage> unlike ~all relays 2019-10-18T17:38:21 < jadew> but it will need some heat sinking 2019-10-18T17:38:30 < BrainDamage> that's what ground planes are for 2019-10-18T17:39:41 < BrainDamage> also, a 10mOhm RDSon mos at 20A will have 4W most, so you should be able to do thermal management fine 2019-10-18T17:39:53 < BrainDamage> even easier for lower RDSon 2019-10-18T17:39:56 < bitmask> ok im liking this idea thanks 2019-10-18T17:41:08 < bitmask> if something on the board shorts, I dont have to worry about power finding a way through the mosfet ? do I need any kind of protection? 2019-10-18T17:41:33 < jadew> add a fuse 2019-10-18T17:41:53 < dongs> welp one of my SXT ac's died 2019-10-18T17:41:56 < bitmask> yes ive been wanting to, these protection things are what im not sure about, where do you add it 2019-10-18T17:41:58 < BrainDamage> you can add a crowbar circuit that shuts off the mosfet 2019-10-18T17:42:08 < dongs> time to order audience and a new replacement 2019-10-18T17:42:48 < BrainDamage> a small sense resistor + diff amp which triggers a transistor which shuts off the mos 2019-10-18T17:43:44 < jadew> if you don't expect this to happen often due to user interaction, then a fuse is the simpler/quicker way 2019-10-18T17:43:45 < BrainDamage> if you can get your hands on a double-gate mos, you don't even need a sense resistor, but those are rate outside ic design 2019-10-18T17:44:18 < BrainDamage> rare* 2019-10-18T17:44:35 < jadew> BrainDamage, how does that work? 2019-10-18T17:46:05 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-18T17:46:10 < BrainDamage> you cut a section of the gate plate, and add another terminal 2019-10-18T17:46:53 < BrainDamage> due to kirkoff laws the plate gets induced a voltage proportional to the current underneath, using the same geometries of the mosfet, with the exception of the gate area 2019-10-18T17:47:12 < BrainDamage> so you read the voltage, and you have voltage proportional to the current 2019-10-18T17:47:21 < jadew> interesting 2019-10-18T17:47:32 < aandrew> stuff like https://imgur.com/a/4imCWDk 2019-10-18T17:47:34 < BrainDamage> ic design tries to get away the most possible without resistors, because resistors suck 2019-10-18T17:49:02 < bitmask> whats that andrew 2019-10-18T17:49:07 < bitmask> aandrew 2019-10-18T17:49:19 < jadew> looks like a DC-DC + something else 2019-10-18T17:50:18 < jadew> ah, maybe 2 DC-DC supplies? 2019-10-18T17:50:44 < aandrew> they'r'e just different switcher layouts 2019-10-18T17:50:53 < bitmask> ahh 2019-10-18T17:51:01 < aandrew> but I was trying to show how careful placement can minimize high current loop area 2019-10-18T17:51:16 < aandrew> interestingly, almost all the components there are on the same side of the board 2019-10-18T17:51:24 < aandrew> but the geometry worked out nicely 2019-10-18T17:51:32 < aandrew> sometimes it works out better on the other side 2019-10-18T17:52:29 < aandrew> that's altium. I have earlier stuff with eagle but don't have a way to view them presently 2019-10-18T17:59:55 < kakimir6486> pro tools look ugli 2019-10-18T17:59:59 < Steffanx> Bye Haohmaru 2019-10-18T18:00:33 < Steffanx> I want antto 2019-10-18T18:00:52 < jadew> he's gaining numbers 2019-10-18T18:02:06 < kakimir6486> how do I call rom api from user mode task? 2019-10-18T18:02:12 < kakimir6486> or should I change the task 2019-10-18T18:02:18 < kakimir6486> to be privileged mode 2019-10-18T18:05:01 < zyp> haha, good luck 2019-10-18T18:05:51 < BrainDamage> btw jadew: your business name will get funny reaction from italian customers 2019-10-18T18:06:01 < jadew> BrainDamage, how come? :) 2019-10-18T18:06:14 < jadew> in here it sounds like testicaltech lol 2019-10-18T18:06:18 < BrainDamage> exactly 2019-10-18T18:06:20 < jadew> haha 2019-10-18T18:06:56 < jadew> it's based on my name, which has nothing to do with testicles :) 2019-10-18T18:07:09 < zyp> nothing at all ;) 2019-10-18T18:07:14 < jadew> heh 2019-10-18T18:07:19 < jadew> I figured it sounds ok in english 2019-10-18T18:07:49 < zyp> dongs would approve 2019-10-18T18:08:00 < aandrew> testicletech HAHAHA 2019-10-18T18:08:18 < aandrew> "our products have the balls to power your projects" 2019-10-18T18:09:19 < jadew> aandrew, had the same reaction when I revealed it to my friends 2019-10-18T18:09:51 < jadew> I was aware of this when I chose the name :) 2019-10-18T18:10:15 < jadew> my wife was making fun of me, saying that if I took my mother's maiden name, it would have been called pubestech 2019-10-18T18:10:30 < aandrew> lol 2019-10-18T18:11:36 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-18T18:13:42 < jadew> Haohmaru, cojotech 2019-10-18T18:14:16 < jadew> exactly 2019-10-18T18:15:02 < kakimir6486> jadew: you should use propper IDC connectors in your products 2019-10-18T18:15:43 < jadew> kakimir6486, what's wrong with that one? 2019-10-18T18:16:16 < kakimir6486> it's not glass filled hi-quality one 2019-10-18T18:16:47 < kakimir6486> but if you get them soldered without deforming pin arrangement there is no actual problem 2019-10-18T18:17:42 < Thorn> pro altiuming https://i.imgur.com/H1udXyl.png 2019-10-18T18:17:48 < jadew> yeah, I have no issue soldering them 2019-10-18T18:18:42 < Thorn> industry grade polygon pour 2019-10-18T18:20:55 < jadew> kakimir6486, also.. they go through extensive testing after they're assembled, so if something would be wrong with the connector, I'd notice 2019-10-18T18:21:30 < kakimir6486> i just doesn't give those quality fibes 2019-10-18T18:21:51 < jadew> I'll see if I can find nicer ones, thanks 2019-10-18T18:22:39 < zyp> kakimir6486, isn't it a SMT vs PTH thing? 2019-10-18T18:22:59 < zyp> SMT stuff is made of nice plastics that can handle reflow, PTH doesn't need to be 2019-10-18T18:23:44 < kakimir6486> but sure is nice to have GF plastic 2019-10-18T18:23:59 < kakimir6486> it gives the fibes 2019-10-18T18:24:06 < jadew> thing is... I kinda went for a color coded thing too 2019-10-18T18:24:10 < jadew> so it would still have to be black 2019-10-18T18:24:17 < jadew> https://cojotech.com/media/datasheets/HPO-080%20-%20Installation%20instructions.pdf 2019-10-18T18:24:19 < jadew> see page 5 2019-10-18T18:24:57 < kakimir6486> connector looks glass filled in this manual 2019-10-18T18:25:17 < kakimir6486> definitelly 2019-10-18T18:25:34 < jadew> first customer plugged the cable the other way around and before talking to me, thought I did a mistake with the cable by adding that tab 2019-10-18T18:25:41 < jadew> kakimir6486, the one on the left is on the mother board 2019-10-18T18:25:47 < jadew> the one on the right is the one on my board 2019-10-18T18:25:55 < jadew> the idea was that white goes to white, black to black 2019-10-18T18:26:19 < jadew> because one has the top cover, the other doesn't 2019-10-18T18:26:26 < kakimir6486> but in your front page it has warped chinesium IDC in product photos 2019-10-18T18:26:38 < jadew> and it matters how you plug them in, because if you plug in the one with the hat in the connector with the clips, it won't go all the way 2019-10-18T18:26:50 < jadew> kakimir6486, they're all like that 2019-10-18T18:27:25 < kakimir6486> oh maybe 2019-10-18T18:28:23 < kakimir6486> it looks glass filled and warped at same time :o 2019-10-18T18:28:44 < jadew> I think they're amphenol connectors, let me check the datasheet 2019-10-18T18:29:17 < jadew> high temperature thermoplastic 2019-10-18T18:29:27 < jadew> nothing about glass fiber 2019-10-18T18:30:04 < jadew> I do get what you're saying tho 2019-10-18T18:30:14 < jadew> I'll see if I can find something better 2019-10-18T18:30:24 < kakimir6486> well if it's not chinesium then it's good 2019-10-18T18:31:38 < jadew> the glass fiber ones are 60c vs 30c for the regular ones 2019-10-18T18:31:55 < mawk> dutchland sent me a contract Steffanx 2019-10-18T18:31:59 < jadew> guess I could give them a try, but then I have to get new female connectors too (they're different colors) 2019-10-18T18:33:09 < kakimir6486> how many you build slash order at one time? 2019-10-18T18:33:18 < kakimir6486> boards 2019-10-18T18:33:44 < jadew> I build about 10 boards at a time 2019-10-18T18:33:51 < jadew> but I order for more usually 2019-10-18T18:35:27 < kakimir6486> do you have new products comming? 2019-10-18T18:35:34 < jadew> yeah 2019-10-18T18:36:28 < kakimir6486> have you considered fluorence display replacements with oled modules? 2019-10-18T18:37:01 < kakimir6486> I wonder if anyone already makes them 2019-10-18T18:37:19 < jadew> don't know if people would be interested in that 2019-10-18T18:37:39 < kakimir6486> they would be interested but not interested to pay 2019-10-18T18:37:43 < jadew> hehe 2019-10-18T18:37:57 < jadew> I mean, they'd probably prefer keeping the original feel 2019-10-18T18:38:01 < jadew> but yeah, I thought about that too 2019-10-18T18:38:10 < jadew> I think there's a guy who did it several years ago too 2019-10-18T18:39:13 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-18T18:39:55 < kakimir6486> tip for text labels on silkscreen: make the thickness narrower so those text look more pro 2019-10-18T18:40:36 < kakimir6486> I can see from board it's made with kicad 2019-10-18T18:42:48 < jadew> those are very small boards, thinner text would be more difficult to see 2019-10-18T18:42:58 < jadew> I'm not fond of the font either 2019-10-18T18:43:39 < kakimir6486> I disagree 2019-10-18T18:43:52 < kakimir6486> try it out 2019-10-18T18:44:33 < jadew> I'll give it a try, thanks 2019-10-18T18:44:38 -!- kakimir6486 is now known as kakimir64 2019-10-18T18:44:48 < jadew> but if it's going to look better, I'll hate taking all those photos again :) 2019-10-18T18:45:45 < kakimir64> what I did I made them a lot narrower and then added some height 2019-10-18T18:46:31 < kakimir64> by default kicad text aspect ratio makes them quite packed in height direction 2019-10-18T18:47:39 < kakimir64> downside is to keep consistency across projects 2019-10-18T18:47:45 < kakimir64> slash products 2019-10-18T18:48:01 < jadew> yeah 2019-10-18T18:48:04 < kakimir64> calculating aspect ratios and size to width ratios 2019-10-18T18:48:07 < jadew> I wish it allowed custom fonts 2019-10-18T18:49:24 < jadew> I think that if it looked like the one where it says X GHz Option, it would have been a lot nicer 2019-10-18T18:49:38 < jadew> but that one was a SVG converted to a footprint 2019-10-18T18:49:48 < jadew> not exactly something you'd want to do too often 2019-10-18T18:51:35 < kakimir64> it's nice looking but it's more of a product label text 2019-10-18T18:54:58 < kakimir64> technical text needs to be looking like nothing 2019-10-18T18:55:00 < kakimir64> flavourless 2019-10-18T18:55:11 < jadew> kinda looks like that, isn't it? 2019-10-18T18:55:11 < kakimir64> tasteless 2019-10-18T18:56:32 < kakimir64> X Ghx Option looks as fancy as any text ever should on pcb appart from logo 2019-10-18T18:57:54 < kakimir64> kicad default text looks like comic sans until you make it into sensible line width and aspect ratio 2019-10-18T18:59:10 < kakimir64> *comic sans bold 2019-10-18T19:02:56 < kakimir64> kinda 2019-10-18T19:07:11 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-18T19:08:59 < jadew> yeah, it looks a bit goofy 2019-10-18T19:10:21 < kakimir64> and it associates the board with mental image of "diy" 2019-10-18T19:11:54 < jadew> ok, you've convinced me :) 2019-10-18T19:12:06 < jadew> I'll have to find a way of using different fonts 2019-10-18T19:12:27 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-18T19:12:56 < kakimir64> it costs nothing but effort 2019-10-18T19:17:01 < Thorn> >The connection between the SAMR21E18A and the QCA4531 is mostly done through a USB to Serial chip, a PL2303SA 2019-10-18T19:17:05 * Thorn vomits 2019-10-18T19:17:45 < karlp> https://zerobin.net/?ebb7a8461ad6ed4d#1R8fWdAB84OdnRKauhpfkCwm5qKdQrJP7lgY1UbAzNw= 2019-10-18T19:17:51 < jadew> sort by cheap 2019-10-18T19:18:54 < jadew> blinky is resource intensive 2019-10-18T19:20:14 < karlp> deliberately. 2019-10-18T19:20:27 < karlp> wanted to ahve some things doing busy waits and shit so I could see the preemption and timing changes. 2019-10-18T19:22:39 < kakimir64> I wonder if I should make seperate task for uart that uses standard input and output 2019-10-18T19:23:44 < kakimir64> that is used by vuart_receiver_task instead of using directly putLineUart etc. 2019-10-18T19:24:30 < kakimir64> calling rom api requires privileged task 2019-10-18T19:27:07 < kakimir64> *thread 2019-10-18T19:31:26 < Thorn> Station, this is President Donald Trump, doo you hear me? https://www.youtube.com/watch?v=Iji5hTQ3CUo 2019-10-18T19:32:27 < jadew> is that his wife? 2019-10-18T19:32:34 < kakimir64> yes 2019-10-18T19:32:59 < kakimir64> almost 2019-10-18T19:35:21 < doomba> yep we're launching from the moon 2019-10-18T19:35:46 < doomba> our savior elon will take us to the moon and then to mars! we don't need to fix our problems here first. 2019-10-18T19:35:57 < jadew> lol 2019-10-18T19:35:59 < jadew> awkward silence 2019-10-18T19:36:21 < doomba> inb4 all the npc news reports trump said moon base 2020 2019-10-18T19:38:43 -!- mentar [~quassel@ec2-18-224-19-6.us-east-2.compute.amazonaws.com] has quit [Remote host closed the connection] 2019-10-18T19:44:55 < kakimir64> how bad it is using rtos to have command parser to write received parameters to a variable and then give a semaphore for other task to read those values and update operation per those variables? 2019-10-18T19:45:13 < kakimir64> you shouldn't do that? 2019-10-18T19:46:19 < kakimir64> actually the updater task should give a semaphore first to command parser to allow writing to those variables 2019-10-18T19:48:41 < kakimir64> or something like that 2019-10-18T20:01:04 -!- sterna [~Adium@2a02:aa1:1619:8e7f:61bd:3ea9:7986:513e] has joined ##stm32 2019-10-18T20:03:28 < kakimir64> mpu region restrictions would make that interesting if variables should be accessed strictly on need basis 2019-10-18T20:18:11 < catphish> ooo spacewalk 2019-10-18T20:18:26 < catphish> looks terrifying 2019-10-18T20:23:11 -!- Jybz [~jibz@91-169-131-97.subs.proxad.net] has joined ##stm32 2019-10-18T20:24:02 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-18T20:28:01 -!- Troupal [~Troupal@imac194.ups-tlse.fr] has joined ##stm32 2019-10-18T20:43:08 < karlp> heh, still all sorts of fun converting some master loop stuff into freertos tasks and timers. 2019-10-18T20:43:30 < mawk> so you use freertos finally ? not keil rtx ? 2019-10-18T20:44:59 < Cracki> I saw that movie. they're racing to the moon for its helium 3 2019-10-18T20:45:06 < karlp> nah, tried it ages ago, didn't really like it. 2019-10-18T20:45:17 < karlp> though largely ~interchangeable 2019-10-18T20:45:51 < karlp> they're both just a generic OS with OS utils, and no hal. 2019-10-18T20:46:21 < mawk> [18:44:55] <kakimir64> how bad it is using rtos to have command parser to write received parameters to a variable and then give a semaphore for other task to read those values and update operation per those variables? 2019-10-18T20:46:40 < mawk> if you want yes, although a semaphore is more than a bare lock 2019-10-18T20:52:07 < kakimir64> I need to figure out inter-task signaling when I get there those things I have never had problems with 2019-10-18T20:52:21 < kakimir64> I need to thing the architecture 2019-10-18T20:52:23 < kakimir64> think 2019-10-18T20:53:08 -!- sterna1 [~Adium@2.65.27.93.mobile.tre.se] has joined ##stm32 2019-10-18T20:55:00 < kakimir64> if I want to access certain mem areas I don't have access to I need to do counter-intuitive thing and call xTaskCreateRestricted instead of xTaskCreate? 2019-10-18T20:55:15 -!- sterna [~Adium@2a02:aa1:1619:8e7f:61bd:3ea9:7986:513e] has quit [Ping timeout: 246 seconds] 2019-10-18T20:56:22 -!- sterna [~Adium@2a02:aa1:1619:8e7f:698b:2579:7c3e:cf39] has joined ##stm32 2019-10-18T20:56:24 < mawk> I never used MPU soryr 2019-10-18T20:56:42 -!- sterna1 [~Adium@2.65.27.93.mobile.tre.se] has quit [Read error: Connection reset by peer] 2019-10-18T20:56:44 < kakimir64> if I get this going 2019-10-18T20:57:14 < kakimir64> I'm highest level pro then 2019-10-18T20:57:21 < mawk> lol 2019-10-18T20:58:06 < kakimir64> it would be just easy to change port files from CM3 non-mpu and carry on 2019-10-18T21:00:35 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T21:05:10 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-18T21:07:49 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-18T21:08:18 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T21:15:19 < jadew> kakimir64, what do you have to say about the just deployed USD | EUR switch at the top? 2019-10-18T21:15:38 -!- sterna1 [~Adium@2.65.27.93.mobile.tre.se] has joined ##stm32 2019-10-18T21:16:08 < kakimir64> I say it needs ex.VAT inc.VAT switch too 2019-10-18T21:16:18 < jadew> it's with VAT 2019-10-18T21:16:23 < jadew> all prices are with VAT 2019-10-18T21:16:28 < jadew> I guess I should specify that somewhere 2019-10-18T21:16:45 < kakimir64> it should say it 2019-10-18T21:16:49 < jadew> in fact, there's no distinction, it is what it is 2019-10-18T21:17:00 < kakimir64> *All prices with VATXX% 2019-10-18T21:17:23 < kakimir64> right next to eur usd switch 2019-10-18T21:17:51 < jadew> thing is, I'm not VAT registered, so I'm not sure how that works 2019-10-18T21:18:05 < jadew> if I say there's VAT included, they might assume they can recover the VAT 2019-10-18T21:18:12 < jadew> and for that to happen I have to become VAT registered 2019-10-18T21:18:24 < kakimir64> okay maybe that is good 2019-10-18T21:18:45 < kakimir64> prices may or may not be VAT 2019-10-18T21:18:48 < kakimir64> * 2019-10-18T21:18:52 < jadew> hehe 2019-10-18T21:19:09 < kakimir64> it depends of contect 2019-10-18T21:19:12 < kakimir64> context 2019-10-18T21:19:15 < jadew> technically, it's considered that the VAT is included, but it cannot be recovered 2019-10-18T21:19:33 < kakimir64> are your client companies? 2019-10-18T21:19:35 -!- sterna [~Adium@2a02:aa1:1619:8e7f:698b:2579:7c3e:cf39] has quit [Ping timeout: 276 seconds] 2019-10-18T21:19:53 < jadew> kakimir64, not for this stuff 2019-10-18T21:19:59 < jadew> but I offer services to companies 2019-10-18T21:20:03 < jadew> and the same thing applies 2019-10-18T21:20:12 < kakimir64> don't stress then 2019-10-18T21:20:12 < jadew> they have to write it off as an expense 2019-10-18T21:23:05 < jadew> it's not oddly placed, is it? 2019-10-18T21:23:41 < aandrew> my use of the RTOS primitivies is very sparse 2019-10-18T21:23:56 < aandrew> I should really spend some time getting to know them 2019-10-18T21:30:42 < qyx> fopencookie() 2019-10-18T21:30:43 < qyx> wow 2019-10-18T21:31:22 < mawk> first time I hear of that qyx 2019-10-18T21:31:32 < qyx> me too 2019-10-18T21:32:36 < qyx> ok seems overly complicated, I just want to redirect printf to an USART 2019-10-18T21:33:16 < mawk> using newlib ? 2019-10-18T21:33:22 < mawk> for that you implement the write() syscall stub 2019-10-18T21:33:38 < qyx> https://github.com/libopencm3/libopencm3-examples/blob/master/examples/stm32/f4/nucleo-f411re/usart-stdio/usart-stdio.c 2019-10-18T21:33:42 < mawk> you can implement read() too 2019-10-18T21:33:47 < qyx> this is what I wanted 2019-10-18T21:33:52 < mawk> yes that's the thing 2019-10-18T21:33:57 < mawk> they're implementing the stubs in that file 2019-10-18T21:34:09 < mawk> int _write(int fd, char *ptr, int len); see there is a comment just above 2019-10-18T21:34:25 < aandrew> qyx: hm, I usually just take over __io_putchar() 2019-10-18T21:34:47 < mawk> with _write() you write whole buffers, but char by char works too if you reimplement __io_putchar() yes 2019-10-18T21:35:28 < aandrew> I also use setvbuf() to set up a buffer or else vsnprintf() will try to allocate 1k on the stack 2019-10-18T21:35:54 < aandrew> setvbuf(stdout, NULL, _IOLBF, 8) 2019-10-18T21:36:02 < mawk> yes 2019-10-18T21:36:19 < mawk> or even disable the buffer in there if you don't want \n buffering 2019-10-18T21:36:21 < aandrew> I need to take a closer look at the SWO channels, there is some neat stuff there 2019-10-18T21:36:29 < mawk> yeah it's easy to use, even 2019-10-18T21:36:35 < mawk> it'd be cooler than UART debugging 2019-10-18T21:36:41 < aandrew> no, if you disable the buffer vfprintf() (not vsnprintf()) will create a buffer of about 1k on the stack 2019-10-18T21:36:57 < mawk> with _IONBF ? 2019-10-18T21:37:00 < aandrew> yep 2019-10-18T21:37:13 < mawk> ah, I didn't know 2019-10-18T21:37:36 < aandrew> I learned the hard way which is the only reason I know this 2019-10-18T21:37:59 < aandrew> /* 2019-10-18T21:37:59 < aandrew> * Note: stdout must be buffered. We use _IOLBF so that 2019-10-18T21:37:59 < aandrew> * printf will flush on a newline character. Attempting to 2019-10-18T21:37:59 < aandrew> * use _IONBF (unbuffered stream) will result in vfprintf() 2019-10-18T21:38:00 < aandrew> * calling __sbprintf() which will place a working buffer of 2019-10-18T21:38:01 < aandrew> * almost 1K onto the stack. 2019-10-18T21:38:04 < aandrew> */ 2019-10-18T21:38:06 < aandrew> setvbuf(stdout, NULL, _IOLBF, 8); 2019-10-18T21:47:41 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-18T21:53:57 < kakimir64> how would I round upwards to closest 2^? 2019-10-18T21:54:20 < kakimir64> and this should be done for align attribute 2019-10-18T21:54:43 < mawk> good question 2019-10-18T21:55:02 < mawk> test if power of 2 with x & (x-1), if it's the case don't touch x 2019-10-18T21:55:10 < mawk> otherwise get msb, multiply it by 2 2019-10-18T21:55:22 < mawk> there must be a better way surely, with lookup tables and stuff 2019-10-18T21:55:41 < mawk> on x86 there are instructions for that, arm I don't think so 2019-10-18T21:56:46 < qyx> kakimir64: pow = 0; while (a & (1 << 31)) { a >>= 1; pow++; } 2019-10-18T21:56:51 < qyx> or something like that? 2019-10-18T21:57:12 < kakimir64> doesn't look like compile time 2019-10-18T21:57:13 < mawk> he wants ceil(log2(x)) 2019-10-18T21:57:16 < qyx> yeah iirc there is a instruction to count bits to first 1 2019-10-18T21:57:19 < mawk> yes it's not compile time obviously 2019-10-18T21:57:30 < kakimir64> it goes to aligned attribute 2019-10-18T21:57:34 < mawk> ah 2019-10-18T21:57:39 < qyx> oh 2019-10-18T21:58:23 < Cracki> https://graphics.stanford.edu/~seander/bithacks.html 2019-10-18T21:58:30 < Cracki> recipes! 2019-10-18T21:58:49 < mawk> it's a hard operation by nature kakimir64 , you need to scan the number to get the msb 2019-10-18T21:59:02 < kakimir64> I just hardcode it 2019-10-18T21:59:03 < mawk> if you find something the compiler deems compile-time it will be a long and ugly expression 2019-10-18T21:59:18 < Cracki> for (int bits = 0; number; bits+=1, number>>=1); 2019-10-18T21:59:24 < Cracki> *int bits before that :P 2019-10-18T21:59:28 < mawk> if it's 32 bits well you can test each bit in turn and that'll be compile time 2019-10-18T21:59:52 < mawk> (x & 0x80000000) ? 31 : x (& 0x40000000) ? 30 : ... 2019-10-18T21:59:53 < Cracki> not sure why 1<<31 (high bit) is tested 2019-10-18T21:59:53 < mawk> and so on 2019-10-18T22:00:06 < kakimir64> I can make a macro that switches line per sizeof(struct) 2019-10-18T22:00:14 < mawk> switch isn't compile time 2019-10-18T22:00:15 < Cracki> for 32 bit numbers that should be the case after the first shift 2019-10-18T22:00:20 < mawk> C macros are very limited 2019-10-18T22:00:31 < kakimir64> mawk: if else if 2019-10-18T22:00:37 < mawk> what do you mean ? 2019-10-18T22:00:43 < mawk> #if ? 2019-10-18T22:00:46 < kakimir64> yes 2019-10-18T22:00:50 < mawk> I don't think that will work 2019-10-18T22:01:02 < Cracki> time to C++ constexpr? 2019-10-18T22:01:05 < kakimir64> ok 2019-10-18T22:01:07 < mawk> the C preprocessor runs first thing, before the compiler, it has absolutely no notion of what sizeof() means 2019-10-18T22:01:15 < mawk> "sizeof" is just a string with regards to the preprocessor 2019-10-18T22:02:02 < Cracki> compiler might optimize anything that depends on the value of a sizeof because it's constant 2019-10-18T22:02:36 < Cracki> there are c preprocessor fuckeries where you can actually iterate! 2019-10-18T22:02:44 < mawk> a limited number of times 2019-10-18T22:02:56 < mawk> you can code something ugly that works like for 16 iterations, then you hit the recursion limit 2019-10-18T22:02:56 < Cracki> so if you need rounded up to powers of two... for a LIMITED range of values, that may be possible 2019-10-18T22:03:11 < mawk> but you can use the assembly preprocessor which is much more cool and powerful 2019-10-18T22:03:17 < mawk> it has actual loops, if, and so on 2019-10-18T22:03:26 < mawk> __asm__(".macro here"); 2019-10-18T22:04:09 < mawk> kakimir64: maybe your best chance is that: ((__sizeof__ (x)) & 0x80000000) ? 31 : ((__sizeof__ (x)) & 0x40000000) ? 30 : ((__sizeof__ (x)) & 0x20000000) ? 29 : ((__sizeof__ (x)) & 0x10000000) ? 28 : ((__sizeof__ (x)) & 0x08000000) ? 27 : ... 2019-10-18T22:04:15 < mawk> write a python script to finish the line 2019-10-18T22:04:30 < Cracki> https://github.com/pfultz2/Cloak 2019-10-18T22:04:41 < Cracki> >for production... 2019-10-18T22:04:42 < kakimir64> mawk: :o 2019-10-18T22:05:12 < mawk> that's floor not ceil though, you can modify it easily to get ceil 2019-10-18T22:05:18 < Cracki> you can limit that to values upto 256 maybe 2019-10-18T22:05:24 < mawk> add +1 to all the 31, 30, 29..., and add a special case for already power of 2 2019-10-18T22:05:25 < mawk> yeah 2019-10-18T22:05:36 < mawk> to test for power of 2 you do (x & (x-1)) == 0 2019-10-18T22:12:44 < mawk> kakimir64: https://paste.serveur.io/raw/PlS3NePJ 2019-10-18T22:12:47 < mawk> I didn't test it tho 2019-10-18T22:13:05 < mawk> I don't know about your platform maybe I need to add UL at the end of each number to not have overflows 2019-10-18T22:14:17 < mawk> like this: https://paste.serveur.io/raw/fjZVFGeI 2019-10-18T22:15:22 < mawk> this is the ceil version 2019-10-18T22:16:06 < kakimir64> what would you call it? 2019-10-18T22:16:25 < mawk> #define CEIL2SIZEOF(x) ... 2019-10-18T22:16:25 -!- Troupal [~Troupal@imac194.ups-tlse.fr] has quit [Read error: Connection reset by peer] 2019-10-18T22:16:26 < mawk> something like this 2019-10-18T22:17:02 < bitmask> is 16V Vgs (max) enough wiggle room for a max of 12.6V actual gate voltage 2019-10-18T22:17:18 < bitmask> mosfet 2019-10-18T22:17:18 -!- boB_K7IQ [~boB_K7IQ@c-73-11-172-226.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T22:19:15 < mawk> it's in the absolute maximum section ? 2019-10-18T22:19:18 < mawk> or just in the common ratings 2019-10-18T22:19:25 < mawk> if it's common I guess it's good 2019-10-18T22:20:32 < bitmask> max 2019-10-18T22:22:17 < bitmask> do I need to go up to 20? 2019-10-18T22:24:04 < qyx> I would 2019-10-18T22:25:46 < bitmask> k 2019-10-18T22:29:34 -!- sandeepkr_ [~sandeepkr@2a03:b0c0:2:d0::cac:7001] has quit [Read error: Connection reset by peer] 2019-10-18T22:29:34 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has quit [Read error: Connection reset by peer] 2019-10-18T22:30:39 -!- boB_K7IQ [~boB_K7IQ@c-73-11-172-226.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-18T22:40:37 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Ping timeout: 240 seconds] 2019-10-18T22:46:43 -!- sterna1 [~Adium@2.65.27.93.mobile.tre.se] has quit [Read error: Connection reset by peer] 2019-10-18T22:52:19 -!- smvoss [~smvoss@207.191.220.92] has quit [Ping timeout: 250 seconds] 2019-10-18T22:52:42 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-18T22:57:05 < kakimir64> mawk: IDE evaluates right value 2019-10-18T22:57:10 < kakimir64> 256 2019-10-18T22:57:20 < kakimir64> but I try to set it to variable 2019-10-18T22:57:21 < mawk> good ! 2019-10-18T22:57:29 < mawk> yeah that will work too 2019-10-18T22:57:35 < mawk> variable can be const 2019-10-18T22:57:56 < kakimir64> 33564928 2019-10-18T22:58:29 < kakimir64> Name : CEIL2SIZEOF(shared_variables_struct) Details:256 2019-10-18T22:58:29 < mawk> sign error 2019-10-18T22:58:40 < kakimir64> oh yes 2019-10-18T22:58:41 < mawk> variable should probably be unsigned int or bigger 2019-10-18T22:59:21 < kakimir64> const uint32_t __attribute__((unused)) shared_variables_struct_ceil2sizeof = CEIL2SIZEOF(shared_variables_struct); 2019-10-18T22:59:42 < kakimir64> let's try bigger 2019-10-18T23:01:48 < kakimir64> let's try -E 2019-10-18T23:02:44 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-18T23:08:12 < mawk> uint32_t is enough 2019-10-18T23:08:33 < mawk> normally 2019-10-18T23:08:44 < mawk> if indeed UL refers to that, not sure 2019-10-18T23:09:04 < mawk> try to print sizeof(1UL) and confirm it's >= 4 maybe 2019-10-18T23:09:58 < kakimir64> less see 2019-10-18T23:10:31 < mawk> it should be 4 on all sane platforms 2019-10-18T23:10:43 < mawk> and maybe 8 on amd64 2019-10-18T23:10:55 < specing> bitmask: I'd go atleast twice 13V 2019-10-18T23:11:34 < bitmask> ehh I'm fine with 20 2019-10-18T23:12:10 < bitmask> I'm seriously under "amping" them 2019-10-18T23:12:59 < bitmask> im gonna parallel two 40-50A for a max draw of 20A 2019-10-18T23:13:12 < bitmask> and that max will be rare 2019-10-18T23:13:59 < mawk> parallel mosfet ? is that a thing ? 2019-10-18T23:14:26 < mawk> mosfets have ample dispersion of characteristics, you'd need to like handpick similar units 2019-10-18T23:15:03 < kakimir64> const uint32_t __attribute__((unused)) ul_sizeof = sizeof(1UL); 2019-10-18T23:15:26 < mawk> yes 2019-10-18T23:15:29 < mawk> is it 4 ? 2019-10-18T23:15:34 < kakimir64> 33564928 2019-10-18T23:15:37 < mawk> wtf 2019-10-18T23:15:42 < qyx> mawk: just see china-made battery protection pcbs 2019-10-18T23:16:10 < mawk> lol 2019-10-18T23:16:26 < qyx> usually there are tens of mosfets for the high-amperage ones 2019-10-18T23:16:29 < bitmask> I thought it was a common thing? guess not 2019-10-18T23:16:55 < mawk> yeah it's maybe common, it was just surprising 2019-10-18T23:17:06 < mawk> if you plan for that dispersion it's fine 2019-10-18T23:17:18 < mawk> like, one mosfet could take like 30% more current than the others or whatever 2019-10-18T23:17:27 < qyx> bitmask: for static conditions select mosfets depending on Rdson at a given Vgs 2019-10-18T23:17:59 < qyx> you can easily go under 1mohm 2019-10-18T23:18:05 < qyx> which is much lower than your wires 2019-10-18T23:18:07 < specing> mawk: it will heat up, its RDSon will rise and they'll balance out 2019-10-18T23:18:26 < bitmask> yea what he said 2019-10-18T23:18:28 < specing> mawk: there is no problem in paralleling them 2019-10-18T23:18:40 < bitmask> I was looking at 5 mOhm Rds on, maybe 7 max 2019-10-18T23:18:53 < bitmask> then it will be 3W max between the two 2019-10-18T23:19:17 < qyx> such low expectations 2019-10-18T23:19:24 < mawk> ah I see specing , mother nature has china's back 2019-10-18T23:19:27 < specing> bitmask: might get 10 fets so you disperse the dissipation and get discounted rates 2019-10-18T23:19:34 < specing> mawk: yes :D 2019-10-18T23:19:50 < qyx> bitmask: n-channel? 2019-10-18T23:19:52 < bitmask> yes 2019-10-18T23:20:01 < qyx> smd? 2019-10-18T23:20:03 < kakimir64> mawk: value I see in expression monitor is _vStackTop 2019-10-18T23:20:07 < bitmask> yes 2019-10-18T23:20:15 < qyx> psmn1R2 not an option? 2019-10-18T23:20:16 < kakimir64> 0x20002900 2019-10-18T23:20:55 < kakimir64> let's validate something with conditional trap then 2019-10-18T23:21:51 < qyx> bitmask: or just select something from the psmn series, they are super-low-Rdson 2019-10-18T23:22:07 < qyx> they are going from 0.7mohm up 2019-10-18T23:22:14 < qyx> and not very expensive 2019-10-18T23:22:15 < bitmask> I see, look good, so just need one of em then? 2019-10-18T23:22:23 < bitmask> wow they are cheap 2019-10-18T23:22:24 < qyx> depends on your pcb 2019-10-18T23:22:29 < specing> huh, 0.7 mohm? 2019-10-18T23:22:30 < bitmask> i'll throw two on 2019-10-18T23:23:06 < qyx> I am using them as load switches, battery disconnects, e-fuses, etc. 2019-10-18T23:23:22 < qyx> ie. for slow switching applications 2019-10-18T23:23:24 < bitmask> yea im using them as battery disconnects 2019-10-18T23:23:37 < bitmask> thanks for the tip, these are nice 2019-10-18T23:24:33 < kakimir64> mawk: expression viewer just lags 2019-10-18T23:24:47 < kakimir64> these views can show absolute garbage sometimes 2019-10-18T23:26:40 < mawk> ah that's good news I guess 2019-10-18T23:27:01 < mawk> it's TUI ? 2019-10-18T23:27:14 < mawk> you can hit ^L for refresh maybe 2019-10-18T23:32:37 < kakimir64> TUI? 2019-10-18T23:32:49 < kakimir64> I mean registers update badly 2019-10-18T23:33:04 < kakimir64> disassembly viewer doesn't sync the address line when scrolling 2019-10-18T23:35:34 < kakimir64> sometimes it jolts in sync but it's lost immediately 2019-10-18T23:35:49 < kakimir64> trust nothing 2019-10-18T23:38:49 < karlp> aandrew: stdio via swo is just "same" https://github.com/libopencm3/libopencm3/blob/master/tests/shared/trace_stdio.c 2019-10-18T23:41:50 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-18T23:45:11 < kakimir64> mawk: no TUI for me thanks 2019-10-18T23:45:33 < mawk> TUI is UI in terminal 2019-10-18T23:47:49 < aandrew> karlp: yes but swo has different channels so you can mux different data without it getting mixed up 2019-10-18T23:48:05 < karlp> indeed. 2019-10-18T23:48:32 < karlp> I've got 0 for "printf" 31 for rtos task tracing, and everything else for "stuff" --- Day changed Sat Oct 19 2019 2019-10-19T00:05:10 < mawk> I received a love letter https://pix.watch/DbBMML/k1I7_u.jpeg 2019-10-19T00:10:49 < jadew> speeding ticket? 2019-10-19T00:11:25 < jadew> throw it in the trash and say you didn't get it 2019-10-19T00:11:57 < mawk> it's an invitation to the department of locative expellings 2019-10-19T00:12:18 < jadew> ah, right 2019-10-19T00:12:20 < mawk> I wonder what happens if I don't go 2019-10-19T00:12:30 < jadew> jail 2019-10-19T00:12:45 < jadew> or fine and if you don't pay that, jail 2019-10-19T00:15:11 < doomba> stop being poor and just pay the ticket. it's easy. don't you have a trust fund? cmon now. 2019-10-19T00:22:35 < kakimir64> I wonder if I can just add MPU region for ROM to https://paste.ee/p/ARdhL 2019-10-19T00:27:06 < kakimir64> if I would go with 2 configurable regions 2019-10-19T00:27:24 < kakimir64> instead of 3 2019-10-19T00:28:09 < aandrew> hm, is there some tool I can feed a stream of numbers into and it'll just graph them, constantly updating, maybe with a few config buttons to control scale/auto or reset the accumulated graph data? 2019-10-19T00:28:35 < zyp> aandrew, if you find one, let me know, I've been looking for the same 2019-10-19T00:35:24 < karlp> I use kst2 for that. 2019-10-19T00:35:39 < karlp> it's not quite as barebones that, but it works. 2019-10-19T00:35:51 < karlp> upgrdman's telemetry viewer is meant to be that too 2019-10-19T00:35:58 < qyx> there was some kde thing 2019-10-19T00:36:02 < qyx> was it kst? 2019-10-19T00:36:26 < qyx> also, gnuplot with few lines in bash 2019-10-19T00:36:41 < aandrew> yeah I've dicked with gnuplot in bash before, it's not as nice as I'd have hoped 2019-10-19T00:36:54 < qyx> fair enuff 2019-10-19T00:37:11 < qyx> it wasnct much pleasant to work with 2019-10-19T00:37:49 < aandrew> ok. if I have two ptp slaves syncing to a master and they seem to "sit" with a 300us offset between them (i.e. the rising edge of slave 1 pps to rising edge of slave 2 pps hovers around 300us that must be because I have some internal delay/latency that I'm not accounting for 2019-10-19T00:38:03 < kakimir64> I extended the peripherals region to basically across the addresspace down to 0x0300000 where the rom is and removed execute never flag 2019-10-19T00:38:25 < kakimir64> my serial command interface is now workings 2019-10-19T00:39:04 < jadew> aandrew, how do you want to stream the data? 2019-10-19T00:39:18 < jadew> does it have to be real time or do you want to just take a peak from time to time? 2019-10-19T00:39:37 < aandrew> jadew: it's coming in over stdout on a shell prompt, I'd have three or four sources I'd just feed into listening pipes I guess 2019-10-19T00:39:42 < aandrew> no it's gotat be realtime 2019-10-19T00:39:58 < aandrew> I want to have the gfx output onscreen and screw with my telnet interface to adjust things and see how it affects the output 2019-10-19T00:40:26 < Thorn> anyone used this soc? https://www.cypress.com/file/298236/download 2019-10-19T00:40:40 < Thorn> cortex-r4, 2MB SRAM, wi-fi 2019-10-19T00:40:45 < Thorn> murata modules available 2019-10-19T00:41:02 < qyx> such sram 2019-10-19T00:41:16 < qyx> aandrew: did you check iot tools? 2019-10-19T00:41:25 < qyx> aren't thjere some mqtt grapjera? 2019-10-19T00:41:30 < qyx> graphers 2019-10-19T00:42:04 < karlp> haven't seen many that are really goood at higher speed stuff 2019-10-19T00:42:25 < aandrew> qyx: that's a good idea... something like graphite or grafaxa or whatever its called 2019-10-19T00:42:34 < aandrew> this isn't super fast, updates are 1-4 times per second 2019-10-19T00:42:35 < Thorn> grafana 2019-10-19T00:43:10 < qyx> I am using grafana 2019-10-19T00:43:25 < qyx> biut there are others with multiple sources 2019-10-19T00:43:29 < karlp> should be fine then. statsd might be even easier than mqtt 2019-10-19T00:47:59 < kakimir64> bestest serial terminal for linux? 2019-10-19T00:48:07 < karlp> picocom 2019-10-19T00:48:15 < specing> karlp: xfce4-terminal 2019-10-19T00:48:20 < specing> kakimir64:^ 2019-10-19T00:48:38 < specing> oh, serial terminal? minicom 2019-10-19T00:49:46 < qyx> screen 2019-10-19T00:51:04 < qyx> Thorn: it is even abgn o_O 2019-10-19T00:51:34 < qyx> zyp: how does it feel to work with R4? 2019-10-19T00:52:12 -!- Jybz [~jibz@91-169-131-97.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-19T00:53:33 < karlp> abgn? 2019-10-19T00:53:41 < karlp> nvm, misread logs 2019-10-19T00:53:50 < qyx> haha it supports sdio but they screwed it up apparently 2019-10-19T00:54:12 < qyx> it is deprecated from the datasheet 2019-10-19T00:54:22 < qyx> also rtc is not functioning 2019-10-19T00:54:41 < qyx> max spi clock is 26MHz for reliable operation 2019-10-19T00:54:49 < qyx> huh 2019-10-19T00:54:49 < kakimir64> http://der-hammer.info/pages/terminal.html this looks more like it 2019-10-19T00:55:02 < Thorn> this is what they used in philips hue bridge 2019-10-19T00:56:30 < qyx> 37.4MHz xtal 2019-10-19T00:56:38 < qyx> were they on drugs or what 2019-10-19T01:00:15 < Thorn> uh no. hue bridge uses something from qualcomm. this one is in ikea 2019-10-19T01:04:57 < Thorn> has anyone used gowin fpgas 2019-10-19T01:05:49 < aandrew> I have not 2019-10-19T01:10:23 -!- con3 [~kvirc@154.119.40.228] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-19T01:19:01 < karlp> saw them on cnxsoft refently, that counts right? :) 2019-10-19T01:19:08 < karlp> they have one with a hard m3 right? 2019-10-19T01:23:30 < kakimir64> https://www.freertos.org/xTaskCreateRestricted.html it wants statically allocated stack or to application writer to provide pvPortMallocAligned() 2019-10-19T01:24:27 < karlp> the Restricted stuff there is for static alloc, not for MPU restrictions... 2019-10-19T01:24:38 < karlp> same with all the xxxxCreateRestricted stuff 2019-10-19T01:24:56 < kakimir64> but what are the versions with static in suffix? 2019-10-19T01:25:07 < karlp> you're right 2019-10-19T01:25:10 < karlp> I mixed them up in my head 2019-10-19T01:25:25 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-19T01:26:00 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-19T01:31:34 < kakimir64> #define pvPortMallocAligned( x, puxStackBuffer ) ( ( ( puxStackBuffer ) == NULL ) ? ( pvPortMalloc( ( x ) ) ) : ( puxStackBuffer ) ) this is like only one I have seen 2019-10-19T01:31:44 < kakimir64> as an example 2019-10-19T01:32:15 < kakimir64> I don't see how it's going to align anything 2019-10-19T01:32:28 < aandrew> hm ok grafana up, influxdb up, it looks like I can just squirt data into it willy-nilly 2019-10-19T01:32:32 < aandrew> now to see the best way to do that 2019-10-19T01:33:03 < Thorn> aandrew just went full iot 2019-10-19T01:33:08 < Thorn> never go full iot 2019-10-19T01:33:29 < aandrew> yeah that is actually exactly what I was thinking 2019-10-19T01:33:36 < aandrew> holy fuck I even learned what an ansible was 2019-10-19T01:33:47 < aandrew> excuse me while I shard 2019-10-19T01:34:30 < mawk> yeah me neither kakimir64 lol 2019-10-19T01:34:35 < mawk> I just expect pvPortMalloc to be already aligning 2019-10-19T01:34:38 < mawk> hence this macro is a stub 2019-10-19T01:34:57 < kakimir64> is it? 2019-10-19T01:35:06 < mawk> yeah, it just calles pvPortMalloc 2019-10-19T01:35:32 < kakimir64> oh yes pvPortMalloc 2019-10-19T01:35:45 < kakimir64> mine just calls malloc 2019-10-19T01:39:17 < aandrew> haha I just realized I can shit out UDP packets to influxdb from the stm32 devices. let the fuckers report themselves 2019-10-19T01:39:34 < karlp> yeah, statsd style and friends ar epretty handy 2019-10-19T01:39:41 < karlp> do you even need influxdb? 2019-10-19T01:39:52 < karlp> isn't grafana itself enough? 2019-10-19T01:40:26 < aandrew> karlp: grafana needs a data source 2019-10-19T01:41:08 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-19T01:42:45 < aandrew> good god though, how many shitty cloud services are out there all doing almost identical things 2019-10-19T01:44:11 < aandrew> that and eveery single one of these fuckers wants some docker instance, it's no fucking wonder people scream for i9 CPUs and 64GB of RAM with terabytes of drive space 2019-10-19T01:44:19 < aandrew> it's all layers of abstraction, all the way down 2019-10-19T01:45:40 < aandrew> I don't need a full fucking db, just some rrd would have been plenty but that's harder to set up and graph than this 2019-10-19T01:46:37 < BrainDamage> the reason why docker is so popular is that people can forget about the underlying os when deploying shit 2019-10-19T01:46:55 < BrainDamage> it's super useful with the cloud because you can abstract away the cloud's os implementation 2019-10-19T01:47:01 < aandrew> sure I get that 2019-10-19T01:47:06 < aandrew> but the waste must be enormous 2019-10-19T01:47:16 < aandrew> it's like chroot on steroids 2019-10-19T01:47:20 < BrainDamage> yes, even more if you have what's called microservices 2019-10-19T01:47:30 < BrainDamage> where every fucking thing gets its own container 2019-10-19T01:47:32 < mawk> well you can assume the OS is linux, people can make a little effort 2019-10-19T01:47:43 < mawk> and VMs are not for dogs anyway 2019-10-19T01:47:59 < mawk> docker is a new and special kind of laziness 2019-10-19T01:48:03 < aandrew> I was going to use graphite but it's awful to set up on osx 2019-10-19T01:48:18 < zyp> qyx, old-fashioned 2019-10-19T01:48:46 < aandrew> yeah I have an esxi blade at home, I have some win7 shits for altium and other shit and a few linux shits for other stuff, it's handy and actually utilizes the hardware better than separate machines 2019-10-19T01:49:15 < aandrew> I see docker as a step "down" from that in the sense that instead of virtualizing the machine they virutalize/abstract the OS 2019-10-19T01:49:34 < aandrew> but a docker instance for each tool that would run on the same machine anyway? I don't get it 2019-10-19T01:49:40 < aandrew> compartmentalization sure but that's just crazy 2019-10-19T01:49:41 < karlp> aandrew: hrm, I just ran a grafana docker and it had something sufficient for protos internally, didn't need anything else 2019-10-19T01:49:57 < aandrew> mind you we have go as well which is just statically linked everything all over again 2019-10-19T01:49:57 < doomba> saves time too because you don't have to deal with editing config files and upgrading packages and circular dependencies and lol this is centos not debian oh wait it's centos now? hrm i need to tune percona erm which resource limits do i need for that jvm? omg 2019-10-19T01:49:58 < zyp> qyx, for most of what I do I don't really notice the difference, but stuff like the interrupt model is much less convenient than cortex-m's NVIC 2019-10-19T01:50:14 < aandrew> I wonder if the OS can correctly COW all the common shit in go apps 2019-10-19T01:50:14 < mawk> aandrew: docker runs on the same machine sure, it's not VM 2019-10-19T01:50:22 < mawk> they keep the same kernel, and virtualize everything(*) else 2019-10-19T01:50:35 < mawk> everything could include a whole OS like with VMs, but you usually don't do that, you start just what you need in the container 2019-10-19T01:50:42 < mawk> eg no or very limited PID 1, for instance 2019-10-19T01:50:49 < aandrew> mawk: right, it's liek chroot on steroids 2019-10-19T01:50:52 < mawk> yes 2019-10-19T01:50:54 < mawk> exactly 2019-10-19T01:51:06 < doomba> no. it's network and pid namespaces, cgroups, and chroot with fancy layered fs bullshit 2019-10-19T01:51:09 < mawk> a chroot that compartimentizes not just the filesystem but also PIDs, mount points, network, cgroups, UIDs, etc 2019-10-19T01:51:17 < aandrew> karlp: weird. what datasources does your docker instance say it has? 2019-10-19T01:51:35 < BrainDamage> assigning every service a container makes sense only if you plan to scale massively your system 2019-10-19T01:51:47 < BrainDamage> having hundred thousand of clients, etc 2019-10-19T01:51:55 < BrainDamage> not necessarily now, but in the future 2019-10-19T01:51:57 < doomba> even for small things docker is great because it lets you be super lazy 2019-10-19T01:52:05 < doomba> write one yaml file and have a whole stack of services running 2019-10-19T01:52:05 < aandrew> BrainDamage: sure, but that breaks one of my primary rules: get it working THEN get it good 2019-10-19T01:52:09 < BrainDamage> so you design from ground up to be fragmented in small parts 2019-10-19T01:52:15 < aandrew> don't write your shitty web app to scale from the start 2019-10-19T01:52:19 < aandrew> it's stupid and wasteful 2019-10-19T01:52:35 < BrainDamage> I agree there's a bit of premature optimization there 2019-10-19T01:52:36 < karlp> aandrew: on a computer at work, nfi sorry 2019-10-19T01:52:42 < aandrew> I'm saying all this as someone who's never even been to the docker or kubernetes site 2019-10-19T01:52:54 < aandrew> you badass, ircing from work 2019-10-19T01:52:58 < aandrew> that's what I'm doing toop 2019-10-19T01:52:59 < BrainDamage> oh and there's business incentive too on this method 2019-10-19T01:53:01 < aandrew> ALL DAY LONG 2019-10-19T01:53:01 < doomba> aandrew: just getting it working and not thinking about scale is why there is such a demand for people who can "dockerize" and re-rig for "scale" lol 2019-10-19T01:53:02 < karlp> no, I mean, I'm at home 2019-10-19T01:53:07 < karlp> and the grafan thing is at work. 2019-10-19T01:53:22 < aandrew> oh I take it back, you're no badass 2019-10-19T01:53:32 < karlp> well, I irc all day anyway:) 2019-10-19T01:53:45 < aandrew> heh 2019-10-19T01:53:45 < BrainDamage> you can slap each service to one person in a team, and reduce the level of cross competency required 2019-10-19T01:53:50 < doomba> because if you let devs just write the code and get it working, it's going to be a busted ass PHP and jquery spaghetti clusterfuck that only works on Habib's ubuntu 14.04 box in india. 2019-10-19T01:54:03 < aandrew> doomba: that is a fair point 2019-10-19T01:54:06 < BrainDamage> basically what doomba said 2019-10-19T01:54:43 < doomba> then they want it to be put into "a kubernetes" 2019-10-19T01:54:53 < aandrew> but now you've got a service that needs eleventy-seven layers of shit just to give you a "Hello World" 2019-10-19T01:55:00 < doomba> and you have to tell them "your shit wasn't built to work in A KUBERDOOBER" 2019-10-19T01:55:03 < BrainDamage> welcome to the web stack 2019-10-19T01:55:44 < BrainDamage> don't forget node.js and npm where leftpad is an external package 2019-10-19T01:55:49 < doomba> so many startups want their code that wasn't optimised for running in such an environment put into said environment - and will pay people to do it - without consulting devs at all. 2019-10-19T01:55:57 < mawk> lol BrainDamage 2019-10-19T01:56:02 < mawk> I've seen an "invert" external package 2019-10-19T01:56:04 < mawk> to do f(x) = -x 2019-10-19T01:56:10 < mawk> I beat you 2019-10-19T01:56:45 < doomba> because they want to be able to tell their investors "it's a kubernetes" to try to squeeze a little higher evaluation from them to keep the foosball tables packed a little longer. 2019-10-19T01:57:03 < BrainDamage> well, it also has actual value for them 2019-10-19T01:57:12 < BrainDamage> it's a modular system where you can swap out parts 2019-10-19T01:57:27 < BrainDamage> so if one doesn't satisfy them, they recognize the value to be able to throw it out 2019-10-19T01:57:31 < doomba> not when the code is written for 2003 "lol it's the dev server. just use WinSCP bro" architecture. 2019-10-19T01:57:47 < BrainDamage> yes, because they have no direct familiarity with the concept 2019-10-19T01:57:55 < BrainDamage> but approach it from a layperson 2019-10-19T01:58:10 < BrainDamage> one is modular but inefficient, the other is efficient but closely knit together 2019-10-19T01:58:26 < doomba> a layperson doesn't care about if the code fits the environment. they just want "a kuberneeeders" 2019-10-19T01:58:27 < kakimir64> mawk: does memalign do what I would like to have? 2019-10-19T01:58:43 < doomba> i deal with these dumbfucks 24/7/365 2019-10-19T01:59:12 < mawk> what do you want to have kakimir64 ? 2019-10-19T01:59:20 < mawk> you said you wanted to align structures 2019-10-19T01:59:32 < mawk> you first wanted to do it compile time with some padding I think, but that's ugly 2019-10-19T01:59:34 < kakimir64> now I want dynamic allocation 2019-10-19T01:59:36 < mawk> you can do it at alloc time 2019-10-19T01:59:37 < mawk> good 2019-10-19T01:59:41 < mawk> see, we're code-soulmates 2019-10-19T01:59:56 < mawk> then it's much easier, you just malloc structures on the boundary of a power of 2 2019-10-19T02:00:22 < mawk> then you need an aligned malloc, or just alloc 15 bytes extra if you don't have aligned malloc 2019-10-19T02:00:26 < mawk> (for a 16 bytes alignment) 2019-10-19T02:00:54 < mawk> you just ceil16 the pointer 2019-10-19T02:01:12 < kakimir64> do you know why I'm doind this? 2019-10-19T02:01:23 < mawk> void *p = malloc(sz + 15); p = (void *)(16*(((uintptr_t)p - 1)/16 + 1)); 2019-10-19T02:01:25 < mawk> something like this 2019-10-19T02:01:27 < mawk> no, why ? 2019-10-19T02:01:36 < zyp> aandrew, you view containers as chroot on steroids, I view containers as lightweight VMs 2019-10-19T02:01:41 < kakimir64> MPU regions 2019-10-19T02:01:50 < mawk> but they *really* are chroot + stuff zyp 2019-10-19T02:01:54 < kakimir64> now I want task stacks 2019-10-19T02:01:58 < kakimir64> dynamically allocated 2019-10-19T02:02:09 < mawk> to make a basic container you need a couple namespaces, then you move $rootdir to /, you chroot into it, and that's all 2019-10-19T02:02:20 < mawk> what's the align requirement on MPU regions kakimir64 ? 2019-10-19T02:02:26 < mawk> 256 bytes I guess ? 2019-10-19T02:02:30 < kakimir64> nope 2019-10-19T02:02:32 < mawk> aka flash sections or something 2019-10-19T02:02:33 < mawk> ah 2019-10-19T02:02:34 < kakimir64> size of the region 2019-10-19T02:02:37 < mawk> yes 2019-10-19T02:02:45 < mawk> what's the requirement ? 2019-10-19T02:02:47 < BrainDamage> I think what zyp means is that the alternative for abstracting away the os is a vm 2019-10-19T02:02:53 < kakimir64> and size of the region must be 2^ 2019-10-19T02:02:53 < mawk> ah 2019-10-19T02:02:56 < BrainDamage> which is considerably heavier 2019-10-19T02:03:05 < mawk> yeah 2019-10-19T02:03:12 < zyp> mawk, it gets it's own network interface too 2019-10-19T02:03:19 < mawk> yeah you can do it 2019-10-19T02:03:28 < mawk> but for instance systemd-containers doesn't separate the network by default 2019-10-19T02:03:36 < zyp> docker does 2019-10-19T02:03:38 < mawk> yes 2019-10-19T02:03:43 < zyp> and you're nitpicking 2019-10-19T02:04:20 < zyp> my point is that I'm using containers now for stuff I used to run in dedicated VMs before 2019-10-19T02:04:31 < mawk> ah 2019-10-19T02:04:35 < mawk> yes 2019-10-19T02:04:42 < zyp> so the way I see it, they save resources over the alternative 2019-10-19T02:04:47 < mawk> but you use them like vms ? like, with a full init systme 2019-10-19T02:04:58 < zyp> no, why would I? 2019-10-19T02:04:59 < mawk> or in the new docker way as the only PID running 2019-10-19T02:05:38 < mawk> well you have a middle view between docker containers and VMs which is using lxc for instance 2019-10-19T02:06:04 < mawk> where you run a full system in them, since maybe one does not have in the start a very clear idea or which services would run into the container, or if they require further servicing in the future or not 2019-10-19T02:06:10 < zyp> true, it's a whole scale 2019-10-19T02:06:30 < zyp> no, what I used to do before was to create a VM and run a single thing in it 2019-10-19T02:06:30 < karlp> has anyone ever actually tried design spark? (RS's cad tool) 2019-10-19T02:10:21 < mawk> I did a docker clone in bash, but I lost the pastebin 2019-10-19T02:11:12 < mawk> ah here https://paste.serveur.io/suat46i8.sh 2019-10-19T02:13:12 < karlp> and that's why people use docker... 2019-10-19T02:13:19 < karlp> because fuck using the machinery by hand. 2019-10-19T02:13:21 < mawk> lol 2019-10-19T02:13:40 < aandrew> holy fuck the collectd website is garbage 2019-10-19T02:13:48 < karlp> collectd is history man. 2019-10-19T02:14:00 < karlp> I keep looking at it, and then going, "nope, no thanks" 2019-10-19T02:14:10 < karlp> idea is kinda nice, 2019-10-19T02:14:11 < aandrew> I just need the fucking protocol 2019-10-19T02:14:22 < karlp> which one :) 2019-10-19T02:16:59 < aandrew> oh is this just a TLV shit? 2019-10-19T02:30:12 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 265 seconds] 2019-10-19T02:33:55 < Thorn> oracle just killed solaris btw 2019-10-19T02:34:14 < mawk> really ? 2019-10-19T02:34:23 < mawk> the bastards 2019-10-19T02:34:28 < Thorn> fired almost the entire team 2019-10-19T02:36:21 < BrainDamage> basically, oracle business model is buy a semi-successful product, milk off its fame while driving it into the ground, kill it, and move to the next thing 2019-10-19T02:36:32 < BrainDamage> almost like an economic parasite 2019-10-19T02:36:55 < specing> terminal stage of capitalism 2019-10-19T02:36:57 < Thorn> is java next 2019-10-19T02:36:59 < specing> the hedge fund stage? 2019-10-19T02:37:13 < mawk> don't forget the lawsuits and the outrageous licensing terms BrainDamage 2019-10-19T02:37:37 < mawk> yeah Thorn they're trying to kill java EE now or something 2019-10-19T02:38:44 < Thorn> spring + kotlin will live 2019-10-19T02:48:01 < doomba> good. kill java plz. 2019-10-19T02:48:45 < doomba> isn't kotlin just syntax sugar on top of java? if so it will die too. 2019-10-19T02:49:13 < BrainDamage> kotlin is another language that targets the jvm 2019-10-19T02:49:21 < mawk> maybe in its infancy yes doomba 2019-10-19T02:49:27 < mawk> but now it compiles to jvm bytecode 2019-10-19T02:49:36 < mawk> but it's mostly compatible I think 2019-10-19T02:49:43 < mawk> at least in the java → kotlin direction 2019-10-19T02:50:13 < BrainDamage> and java won't die, not with the most popular platform ( android ) using it 2019-10-19T02:50:26 < BrainDamage> just desktop / server java maybe 2019-10-19T03:04:57 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T03:30:58 < catphish> does usb have any way to mark the end of a multi-packet bulk transfer? 2019-10-19T03:31:14 < catphish> or does one simply have to know the length of the data in advance? 2019-10-19T03:34:09 < catphish> got it - "A bulk transfer is complete when the endpoint does one of the following: Transfers a packet with a payload size less than wMaxPacketSize or transfers a zero-length packet" 2019-10-19T03:43:59 < dongs> yeah 2019-10-19T03:45:34 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has joined ##stm32 2019-10-19T03:47:11 < dongs> < Thorn> spring + kotlin will live 2019-10-19T03:47:25 < dongs> implying both of thse filthy shits arent just fucking literally preprocessor on top of java 2019-10-19T03:51:49 < dongs> fuck kotlin in particular 2019-10-19T03:51:51 < dongs> lazy nigger shit 2019-10-19T03:52:12 < dongs> i have some fagdroid app i use which functionality requires literally maybe 10k of code 2019-10-19T03:52:22 < dongs> but its 10megs installed because shithead used kotlin 2019-10-19T03:53:32 < dongs> ooh and they just added unreadableroutines to it 2019-10-19T03:53:39 < dongs> where tyou have same faggotry as javascript 2019-10-19T03:53:54 < dongs> func nigger() = fuckoff { nigger nigger nigger nigger nigger nigger } } } } } } 2019-10-19T03:54:18 < dongs> i think niggerterm for that is lambdas or wahtever 2019-10-19T03:54:24 < dongs> fuck any langauge that supports that shit 2019-10-19T04:06:35 < mawk> lol 2019-10-19T04:06:41 < mawk> but it's functional programming 2019-10-19T04:06:44 < mawk> like all the cool kids do 2019-10-19T04:07:02 < aandrew> how does collectd associate data types (by name) to their id? types.db doesn't seem to contain that 2019-10-19T04:11:36 < dongs> oh and kotlin looks like some commecial faggotry pushed by fucking jetbrainz 2019-10-19T04:11:53 < dongs> once they get a bunch of dumb niggers onboard they'll start charging per line of code compiled 2019-10-19T04:12:41 < doomba> that's kind of the idea with all things dongs 2019-10-19T04:13:21 < mawk> jetbrains is cool 2019-10-19T04:13:24 < mawk> they're eastern europe 2019-10-19T04:13:30 < dongs> i dont care 2019-10-19T04:13:33 < doomba> a tiny subset of billionaires at the top collecting subscription fees all the way down 2019-10-19T04:13:33 < dongs> they deal with javashit 2019-10-19T04:13:36 < dongs> so tehy can get fucked 2019-10-19T04:19:47 < aandrew> ah I see. so I'd have to send something like [type=0,length=blah,"host_string\0"][type=1,length=12,64b_time][type=4,length=blah,"type_name\0"][type=6,len=whatever,datatype=0,64b_value] 2019-10-19T04:19:51 < aandrew> for each entry I want to log 2019-10-19T04:20:02 < aandrew> that's kind of shitty for a binary protocol 2019-10-19T04:20:14 < aandrew> may as well snprintf the text protocol instead 2019-10-19T04:22:33 < dongs> what 2019-10-19T04:22:36 < dongs> are ytou talking to ohsix or something 2019-10-19T04:22:44 < dongs> or just autisming 2019-10-19T04:22:46 < aandrew> I haven't seen ohsix say anything in a long time 2019-10-19T04:22:49 < dongs> good 2019-10-19T04:22:50 < aandrew> did you cut out his tongue or something 2019-10-19T04:22:54 < dongs> hopefluly 2019-10-19T04:23:00 < dongs> what the fuck is that shit you just pasted 2019-10-19T04:23:10 < dongs> strings with nulls in the middle? 2019-10-19T04:23:16 < aandrew> nah I'm thinking of using the collectd protocol to spew data from a bunch of stm32s over to influxdb so I can graph the shit 2019-10-19T04:23:26 < aandrew> no not strings with nulls in the middle, a string with a null term 2019-10-19T04:23:51 < dongs> collectd sounds too similar to systemd for me to even care about researching it 2019-10-19T04:23:56 < aandrew> haven't you read the backlog? I've full out kubernetes'd myself into an iot docker wonderfuckingland 2019-10-19T04:24:11 < dongs> fucking troll 2019-10-19T04:24:20 < aandrew> coming from you I take that as high praise 2019-10-19T04:24:34 < aandrew> anyway 2019-10-19T04:24:52 < aandrew> I think I have the format figured out, now gotta code up a bunch of shitty routines to make the bits and shit them out ethernet 2019-10-19T05:09:40 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 264 seconds] 2019-10-19T05:17:40 -!- splud [~noneya.bi@unaffiliated/splud] has quit [Ping timeout: 246 seconds] 2019-10-19T05:24:16 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-19T05:24:28 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-19T06:10:27 -!- splud [~noneya.bi@unaffiliated/splud] has joined ##stm32 2019-10-19T06:12:54 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T06:25:50 -!- fc5dc9d4_ [~quassel@p57A3271A.dip0.t-ipconnect.de] has joined ##stm32 2019-10-19T06:29:46 < bitmask> what do you do for something like a mosfet that has 4 pins and a tab, do you still just use a 3 pin schematic symbol and just remember to connect it correctly? 2019-10-19T06:29:56 -!- fc5dc9d4 [~quassel@p5B081616.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-19T06:36:28 < jadew> if you don't have a symbol for it, make one 2019-10-19T06:49:49 < bitmask> how, a mosfet with 3 source pins? 2019-10-19T06:49:55 < bitmask> I guess that works 2019-10-19T06:50:48 < jadew> if they all have the same function, then you only need a footprint 2019-10-19T06:50:55 < Cracki> uh, normal mosfet symbol, but footprint accordingly 2019-10-19T06:50:58 < jadew> and have the same pin multiple times 2019-10-19T06:51:38 < Cracki> footprint with 2+ pads of same net, sure 2019-10-19T07:02:43 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-19T07:02:47 < jadew> https://www.youtube.com/watch?v=7JRyy2MM-rI 2019-10-19T07:03:12 < jadew> interesting points of view 2019-10-19T07:06:53 < Cracki> >play time 40:05 2019-10-19T07:07:26 < Cracki> >thiel 2019-10-19T07:07:28 < Cracki> >immigrant 2019-10-19T07:07:29 < Cracki> LOL 2019-10-19T07:07:39 < Cracki> the politically correct term is "settler" 2019-10-19T07:11:43 < Cracki> tldw https://medium.com/@bonniekavoussi/notes-from-peter-thiels-speech-at-the-national-conservatism-conference-on-july-14-2019-6a51b26b202 2019-10-19T07:12:50 < Cracki> >Is it because Google is so infiltrated that they have decided to work with the Chinese military and not the U.S. military? 2019-10-19T07:15:20 < Cracki> a bunch of sensible points there 2019-10-19T07:15:31 < Cracki> >The lies we should worry about the most are the ones that are so big that no one calls them out. 2019-10-19T07:18:16 < Cracki> >There’s a crazy conspiracy theory on the left that we were in Iraq to get the oil. It’s actually crazy that it wasn’t true. Shouldn’t American oil companies have gotten the oil? 2019-10-19T07:18:20 < Cracki> who owns the oil? hello 2019-10-19T07:20:18 < jadew> also saw a couple of Douglas Murray interviews lately, what do you think of him? 2019-10-19T07:22:07 < jadew> judging by what you see online, it almost looks like all gays are conservatives and the other letters of the LGBTQ are left wing 2019-10-19T07:22:26 < Cracki> that's half true 2019-10-19T07:22:38 < Cracki> all the other letters are tankies, but half the gays are leftists too 2019-10-19T07:23:25 < Cracki> tho it'd be more accurate to call those eunuchs because they hate themselves for having a peepee 2019-10-19T07:23:33 < jadew> really? I can't name one leftist gay, but I can name 3 right wing ones from the top of my head 2019-10-19T07:23:56 < Cracki> you can include me in the second number :P 2019-10-19T07:24:03 < jadew> right, so... 4 2019-10-19T07:24:28 < Cracki> not sure about online, but offline I see a bunch of neon colored gays 2019-10-19T07:24:52 < Cracki> they buy into the bullshit. 2019-10-19T07:25:30 < Cracki> those that aren't leftist probably realize who would throw them off buildings, and who lets those in the country 2019-10-19T07:26:05 < jadew> heh, I think "conservative" has shifted 2019-10-19T07:26:18 < jadew> is it possible that some of them still think it means what it used to mean 50 years ago? 2019-10-19T07:26:35 < Cracki> everything is shifting. 2019-10-19T07:26:48 < Cracki> new terms every year, in every country different ones 2019-10-19T07:27:13 < Cracki> for example, I can't make sense of french politics that goes beyond macron vs. le pen. 2019-10-19T07:27:44 < Cracki> they have their own philosophers that are all but unknown outside the nation 2019-10-19T07:27:57 < Cracki> that goes for every nation 2019-10-19T07:28:44 < Cracki> I don't identify with the label "conservatism" 2019-10-19T07:29:06 < jadew> yeah, I don't identify with any of the US labels 2019-10-19T07:29:13 < jadew> probably libertarian? 2019-10-19T07:29:27 < Cracki> I don't even identify with my own nation's popular right wing party because they are too much of a bunch of pansies in some aspects 2019-10-19T07:29:40 < Cracki> at heart, I'm too liberal for my own good. 2019-10-19T07:29:51 < Cracki> I give pieces of shit too much leeway. 2019-10-19T07:32:48 < Cracki> basically, german politics is fucked. the only way to fix it is to nuke the population centers (because the rot congregates there) and then reeducate (brainwash) the weakness out of who's remaining for the next 70 years 2019-10-19T07:33:14 < Cracki> take control of the media and education system 2019-10-19T07:33:27 < jadew> I remember an interview I saw a couple of days ago, Douglas Murray and... not sure who, and an interesting remark was made (almost jokingly) 2019-10-19T07:33:40 < Cracki> and persecute anyone who questions the narrative 2019-10-19T07:33:41 < jadew> that maybe conservativism and maturity are the same thing 2019-10-19T07:33:53 < Cracki> they probably are 2019-10-19T07:34:03 < jadew> that's what I thought 2019-10-19T07:34:06 < Cracki> young people haven't worked for anything. they don't know the value of anything. 2019-10-19T07:34:23 < Cracki> I see that in my volunteer work at uni. 2019-10-19T07:34:44 < Cracki> you can bust your ass to build something, but then pieces of shit come along, haven't done shit for it, and take it all away. 2019-10-19T07:34:46 < jadew> it makes sense and it also makes sense that with people staying immature for longer, you would get an increase in whatever the opposite of that is 2019-10-19T07:35:01 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-19T07:35:54 < Cracki> they campaign to let 16 year olds vote. that adds two years worth of gullible idiots to the voting herd 2019-10-19T07:36:05 < jadew> hah 2019-10-19T07:37:06 < jadew> that's hilarious 2019-10-19T07:38:27 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-19T07:38:28 -!- day__ is now known as day 2019-10-19T07:40:10 < Cracki> that one new dyed hair dude who's very leftist talked of "tradition" when he complimented our work... I think he's taking the piss 2019-10-19T07:48:24 < bitmask> can you assign a net to a region? 2019-10-19T07:51:34 < jadew> what software? 2019-10-19T07:51:40 < bitmask> altium 2019-10-19T07:51:44 < jadew> but the answer is probably yes 2019-10-19T07:51:45 < dongs> of course? 2019-10-19T07:51:46 < dongs> why wouldnt you 2019-10-19T07:51:54 < bitmask> I dont see an option 2019-10-19T07:52:19 < dongs> the fuck you mean 2019-10-19T07:52:23 < dongs> properties, net 2019-10-19T07:52:29 < dongs> its literally one of like 3 options you can set on a region 2019-10-19T07:52:41 < dongs> net/layer/type of region (copper, cutout etc 2019-10-19T07:52:44 < dongs> make sure its set to copper? 2019-10-19T07:52:59 < bitmask> its set to copper, top layer, but no option for net 2019-10-19T07:53:03 < dongs> regions is how i make small copper pours around shit like dc/dc coils etc 2019-10-19T07:53:09 < dongs> yeah thats right above that 2019-10-19T07:53:32 < dongs> https://i.imgur.com/jyOv46R.png 2019-10-19T07:54:17 < bitmask> :( 2019-10-19T08:04:27 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has joined ##stm32 2019-10-19T08:06:37 < ronox> hey, could someone help me figure out how to set up the ADC on an stm32f4, to work the same as for an F3? https://github.com/cnlohr/colorchord/blob/master/embeddedstm32f303/adc.c 2019-10-19T08:07:35 < ronox> they use completely different libraries and i cant make sense of what the actual functions do, best ive managed is with the few functions where they are described as doing the same thing, but i still dont get how certain stuff is cryptically spoken 2019-10-19T08:08:51 < ronox> anyway my goal is to rewrite the initADC() function for the F3 version of colorchord, for the F4 version of colorchord 2019-10-19T08:11:11 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-19T08:12:04 < dongs> looks very simple to me 2019-10-19T08:12:18 < dongs> sets up a channel, sets up a timer which starts conversion on each timer update 2019-10-19T08:12:36 < ronox> it isnt to me, im a beginner to C and arduino, so, hardware specific programming is abit beyond me 2019-10-19T08:12:50 < ronox> it doesnt help i dont really know how ADCs work either 2019-10-19T08:13:06 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-19T08:13:07 < dongs> blame ST for coming up wiht the fucking HAL shit 2019-10-19T08:13:20 < ronox> that HAL stuff automates it now doesnt it 2019-10-19T08:13:48 < dongs> also i'm almost certain you can configure ADC to actually automatically trigger off timers 2019-10-19T08:14:10 < dongs> so the shit they're doing there is kinda dumb 2019-10-19T08:14:52 < ronox> in the library im trying to work with? 2019-10-19T08:15:08 < dongs> i donno what the library is, im commenting about teh particular code you linked. 2019-10-19T08:15:35 < ronox> yeah that 2019-10-19T08:15:36 < dongs> isnt cnlorh some youtube retard 2019-10-19T08:15:43 < ronox> dunno 2019-10-19T08:15:53 < ronox> i dont think he has much youtube stuff 2019-10-19T08:16:26 < ronox> anyway as im porting stuff, i cant find like certain equivalents or understand why i get errors 2019-10-19T08:18:09 < ronox> for example why ADC_CommonInitStructure.ADC_DMAMode = ADC_DMAMode_OneShot; gives an error 2019-10-19T08:19:46 < ronox> i dont know what "oneshot" even means, i cant find any mention of it in the F4s peripheral library 2019-10-19T08:20:10 < ronox> so, i have no idea what code to use for the F4 to create the same result 2019-10-19T08:20:59 < ronox> here is the work i have done so far https://pastebin.com/iXCYqPDz 2019-10-19T08:21:22 < ronox> i have commented //XX next to each function i found an equivalent for in the F4 library 2019-10-19T08:21:41 < ronox> and commented //no equivalent , when i couldnt find an equivalent at all 2019-10-19T08:22:06 < ronox> the plainly commented out lines i cant figure out why they dont work 2019-10-19T08:22:59 < ronox> everything below the /* is the old initADC function 2019-10-19T08:24:17 < dongs> oh this is not even hal 2019-10-19T08:24:19 < dongs> wtf dude 2019-10-19T08:24:26 < dongs> what fuckign IDE are y using 2019-10-19T08:25:06 < dongs> also for the love of fuck never fucking ever comment out more than one line of code with /* 2019-10-19T08:25:13 < dongs> use fucking #if 0 / #endif 2019-10-19T08:26:48 < Cracki> ronox, first assumption you need to throw overboard is that the adcs in f3 and f4 work exactly the same 2019-10-19T08:27:00 < ronox> im not assuming that at all 2019-10-19T08:27:03 < Cracki> good 2019-10-19T08:27:13 < dongs> they're not, F3 is slightly newer/better 2019-10-19T08:27:15 < Cracki> same for timers 2019-10-19T08:27:30 < ronox> also as for IDE, its not IDE specific, its just text 2019-10-19T08:27:31 < Cracki> so what you do when something is unknown is that you google it 2019-10-19T08:27:37 < Cracki> in particular this oneshot stuff 2019-10-19T08:27:43 < dongs> ronox: the fuck? yes its IDE specific because you can use shit like working code complete 2019-10-19T08:27:50 < dongs> to cross reference headers/etc 2019-10-19T08:27:56 < dongs> i mean if youre using retarduino "IDE" then yeah 2019-10-19T08:28:20 < ronox> i use vscode but, thats not much more useful than notepad besides the autoformatting 2019-10-19T08:28:31 < ronox> code-wise it doesnt help all that much 2019-10-19T08:28:52 < dongs> anyway origian code doens't use DMA so there's nothing to configure thats DMA related anyway. 2019-10-19T08:28:59 < ronox> ok thats good 2019-10-19T08:29:20 < Cracki> wait so why do you introduce DMA then? 2019-10-19T08:29:28 < dongs> there's also XXX_StructInit() calls 2019-10-19T08:29:44 < dongs> you should do that before assigning shit to peripheral structs if you have no idea what youre doing 2019-10-19T08:29:46 < Cracki> >ADC_CommonInitStructure.ADC_DMAAccessMode = ADC_DMAAccessMode_Disabled; 2019-10-19T08:29:49 < ronox> my education in C only breifly touched structures 2019-10-19T08:29:50 < Cracki> good. 2019-10-19T08:30:15 < dongs> so ADC_StructInit(&blah); and then blah.ADC_Whatever = foo; 2019-10-19T08:30:17 < ronox> do what? assign all the structures first? 2019-10-19T08:30:41 < dongs> no, to initialize them so that you ONLY set the stuff that you need and not end up with wrong values due to random shit on stack 2019-10-19T08:31:20 < dongs> vschoad is absolutely disgusting 2019-10-19T08:31:37 < dongs> literally a glorified notepad.exe that also happens to require like 200megs of diskspace 2019-10-19T08:32:02 < ronox> its identical on windows and linux so, was useful for my programming class 2019-10-19T08:32:10 < ronox> they wanted everyone to program in nano 2019-10-19T08:32:20 < dongs> you should have literally walked out of that class 2019-10-19T08:32:24 < dongs> if that was a requirement 2019-10-19T08:32:36 < ronox> it wasnt a requirement, just a suggestion 2019-10-19T08:32:39 < dongs> if a school can't afford fucking free visual studio community edition 2019-10-19T08:32:41 < ronox> the only suggestion though 2019-10-19T08:32:43 < dongs> then i dont know 2019-10-19T08:33:10 < ronox> along the way people realized they could do the same stuff in notepad or vim or something-leaf-pad 2019-10-19T08:33:42 < ronox> or that they can open 2 instances of terminal so they dont have to close nano every time they wanna try compiling 2019-10-19T08:34:02 < ronox> anyway no arguments here it wasnt that great a programming class 2019-10-19T08:35:18 < ronox> trying to set up the ADC the same, i just have no idea what im doing, so best i can do is go line by line and try to find the equivalent functions 2019-10-19T08:35:49 < ronox> i cant tell if certain stuff even is neccesary or not 2019-10-19T08:39:02 < ronox> anyway, maybe best i come bck once ive converted as much as i can find, then to ask if im missing anything 2019-10-19T08:39:21 < Cracki> make sure to run it and see if it works 2019-10-19T08:39:51 < Cracki> it looks like the author also just took the code from some recipes 2019-10-19T08:40:05 < dongs> this is just super simple adc setup which is then manually triggered by timer. 2019-10-19T08:40:12 < Cracki> those apis appear to be from stdperiph, the predecessor of st's hal/ll code 2019-10-19T08:40:12 < dongs> take a look at F4 stdperiphlib Examples\ADC 2019-10-19T08:40:19 < dongs> there/s a few setups of doing just that. 2019-10-19T08:40:22 < dongs> yeah 2019-10-19T08:40:42 < dongs> STM32F4xx_DSP_StdPeriph_Lib_V1.8.0\Project\STM32F4xx_StdPeriph_Examples\ADC 2019-10-19T08:40:44 < dongs> tons of shit there 2019-10-19T08:41:26 < dongs> i think tehre might be some adc examples mixed in elsewhere too 2019-10-19T08:41:47 < dongs> or else just google shit like ADC_CommonInit and look at other peoples shit doing same thing 2019-10-19T08:43:13 < Cracki> grab the same example for F3, then diff the code. F3 example is likely based on F4 example 2019-10-19T08:43:37 < Cracki> "meld" is a graphical diff tool that has proper visualization 2019-10-19T08:43:49 < Cracki> (i.e. not this +- bullshit but side by side) 2019-10-19T08:45:14 < Cracki> average = ((average*1023) + (value*1024))/1024; 2019-10-19T08:45:15 < Cracki> lol 2019-10-19T08:46:12 < ronox> ty, ill do that, ill come back once im good and proper stuck and not making progress 2019-10-19T08:47:01 < Cracki> also demand to speak to the manager of that college course. they either assume prerequisites that aren't taught or they just plain suck 2019-10-19T08:47:02 < ronox> so, what your saying is, this is more or less just a typical ADC setup 2019-10-19T08:47:08 < Cracki> it looks typical 2019-10-19T08:47:21 < Cracki> line 99, timer handler 2019-10-19T08:47:23 < ronox> it just has a shitload of unneccesary stuff defined 2019-10-19T08:47:35 < Cracki> I wouldn't say that 2019-10-19T08:47:55 < Cracki> but wat dongs said, structure init can be simplified 2019-10-19T08:48:16 < ronox> line 99 from what i can tell reads the data 2019-10-19T08:48:25 < ronox> is the function for that i mean 2019-10-19T08:48:30 < Cracki> the way that looks is not unusual. you wanna make sure all fields of a struct have been assigned a known value 2019-10-19T08:48:52 < Cracki> tim2 irq handler is what gets run periodically from the timer 2019-10-19T08:49:08 < Cracki> and all the adc stuff is just 107,108 and that's all 2019-10-19T08:49:47 < ronox> 107,108? 2019-10-19T08:49:50 < ronox> oh right 2019-10-19T08:49:51 < Cracki> line 2019-10-19T08:49:56 < ronox> sorry :P 2019-10-19T08:50:08 < dongs> if it was me doing it i'd just setup automatically triggered DMA conversion with a timer and then slurp the numbers during DMA_TC or DMA_HT and average them 2019-10-19T08:50:09 < Cracki> all the init wankery is verbose but boring 2019-10-19T08:50:14 < dongs> this way entire thing is basically hands-off 2019-10-19T08:50:37 < dongs> you can set your averaging window by the size of DMA buffer 2019-10-19T08:50:47 < ronox> DMA is beyond me at the moment 2019-10-19T08:50:52 < Cracki> that may be too advanced to arduino folks. they don't know DMA 2019-10-19T08:51:07 < dongs> is tehre node.js for retarduino yet 2019-10-19T08:51:09 < Cracki> also assumes you have a bit of ram to write into :> 2019-10-19T08:51:13 < dongs> so you can copypaste shit like left_pad() 2019-10-19T08:51:33 < Cracki> there is node.js on satellites http://johnny-five.io/ 2019-10-19T08:51:35 < ronox> anyway so, my understanding then is, i should be optimistic about not being able to find all the same function member enables and stuff 2019-10-19T08:52:02 < ronox> because its just enabling stuff already enabled by default anyway 2019-10-19T08:52:09 < Cracki> you should be optimistic about finding the f4 adc example he gave the path to, and basically that's it 2019-10-19T08:52:20 < Cracki> you'll need a timers example too, but those should also be mostly identical to f3 2019-10-19T08:52:53 < Cracki> google each function in initADC(), see what it's supposed to do 2019-10-19T08:53:04 < Cracki> you see clock config. stm32 have a clock tree 2019-10-19T08:53:14 < Cracki> clocks to timer, adc, and I/O ports 2019-10-19T08:53:33 < Cracki> also that code has some actual comments, which isn't usual 2019-10-19T08:54:11 < ronox> in the stm32f4xx_adc.c everything is well annotated 2019-10-19T08:54:22 < ronox> thats what ive been going off, the descriptions in there 2019-10-19T08:54:29 < ronox> and the f3 version 2019-10-19T08:54:53 < ronox> anyway presetly requesting the full peripheral library with its examples so i can look at it 2019-10-19T08:55:32 < Cracki> I do wonder why the f3 code puts the adc in continuous conversion mode, but in the Interrupt handler it calls ADC_StartConversion 2019-10-19T08:55:38 < dongs> 'requesting'? 2019-10-19T08:55:41 < dongs> its free download 2019-10-19T08:56:15 < ronox> yeah it takes a while for the download to start 2019-10-19T08:56:24 < ronox> have to sign shit to get to the download page 2019-10-19T08:56:32 < ronox> anyway i got it now and unzipping 2019-10-19T08:56:45 < Cracki> "login/register" is all you need https://www.st.com/en/embedded-software/stsw-stm32065.html#get-software 2019-10-19T08:57:06 < ronox> guess i was already logged in 2019-10-19T08:57:37 < ronox> anyway, from what your saying, im starting to think that this has all been one giant unneccesary copypaste, right? 2019-10-19T08:57:48 < ronox> that the author did in the F3 adc file 2019-10-19T08:58:10 < dongs> he' 2019-10-19T08:58:18 < Cracki> I'm guessing the adc_startconversion is only needed once but he still calls it every time 2019-10-19T08:58:19 < ronox> that they probably didnt individually write out each function as needed 2019-10-19T08:58:23 < dongs> s initializing unnecessary struct bits to confuse you isntead of using StructInit(), for sure 2019-10-19T08:58:37 < ronox> pardon? 2019-10-19T08:58:40 < Cracki> ugh that code is weird 2019-10-19T08:58:51 < dongs> Cracki: just looks like retarded copypasted shit 2019-10-19T08:58:57 < dongs> typical of a retarduino lib i guess 2019-10-19T08:59:01 < Cracki> yes, structinit() is what you want 2019-10-19T08:59:10 < ronox> you mean structinit() can be called to just initialize all of it rather than calling each component too? 2019-10-19T08:59:12 < dongs> steal bits of shit here and tehre and randomly throwe bits at a wall until something sticks 2019-10-19T08:59:20 < Cracki> adc continuous conv instead of timer-triggered... 2019-10-19T08:59:36 < Cracki> structinit() gives you good defaults 2019-10-19T08:59:42 < dongs> ronox, do you know C? struct faggot foo; > contents of foo are now random thanks to stack allocation. 2019-10-19T08:59:42 < Cracki> it's (almost) like zeroing the whole struct 2019-10-19T08:59:51 < Cracki> after that you only set what you actually mean to set 2019-10-19T09:00:07 < dongs> so you could either do struct faggot foo = { 0, } OR use provided StructInit() to have a good known state 2019-10-19T09:00:14 < ronox> i know C but i only did 1 lesson in structures 2019-10-19T09:00:21 < ronox> basic stuff for organisation purposes 2019-10-19T09:00:24 < Cracki> then you're "new to C" 2019-10-19T09:00:28 < ronox> yeah 2019-10-19T09:01:03 < ronox> ive know C longer than i did the class for however, but that class had me go into more advanced stuff than i had ever done before 2019-10-19T09:01:36 < Cracki> if that class is the first time you've _really_ touched structures, you've "heard of" C before 2019-10-19T09:02:05 < Cracki> anyway, stay at it 2019-10-19T09:02:12 < ronox> mostly its experience in arduino and this special custom C language written for lego NXT firmware hacking 2019-10-19T09:02:22 < ronox> i do plan to, C is useful AF 2019-10-19T09:02:58 < ronox> the lego one may have been C#, i forget 2019-10-19T09:03:26 < ronox> ah, things done unzipping, time to take a peek 2019-10-19T09:05:34 < ronox> aha 2019-10-19T09:05:39 < ronox> ADC_CommonInit(&ADC_CommonInitStructure); 2019-10-19T09:05:46 < ronox> this one was stumping me 2019-10-19T09:06:00 < ronox> in the F4, the ADC its used for simply is not specified 2019-10-19T09:06:06 < Cracki> the underscoring may be confusing because it's not grouping things sensibly 2019-10-19T09:06:21 < ronox> ADC_CommonInit(ADC1, &ADC_CommonInitStructure); is what it is for the F3 2019-10-19T09:07:09 < ronox> also, reading how the structure is set up in the peripheral library adc.c, was a bit too confusing for me to see that it doesnt take the ADC as an argument 2019-10-19T09:08:50 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T09:33:24 < ronox> ok, i made a ton of progress but im stuck at thiis because its definitely required 2019-10-19T09:33:25 < ronox> while (!ADC_GetFlagStatus(ADC1, ADC_FLAG_RDY)); 2019-10-19T09:33:40 < ronox> what exactly does this do? 2019-10-19T09:34:21 < ronox> in the f3 periph lib it says "ADC_FLAG_RDY: ADC Ready flag" 2019-10-19T09:34:36 < ronox> i cannot find anything similar for the F4 as a "ready" flag 2019-10-19T09:38:48 < dongs> after waht stuff does it call that 2019-10-19T09:40:17 < dongs> ronox: you are fucking bad and should feel bad 2019-10-19T09:40:26 < ronox> ? 2019-10-19T09:40:32 < ronox> it does it after the adc is enabled 2019-10-19T09:40:42 < dongs> < ronox> here is the work i have done so far https://pastebin.com/iXCYqPDz 2019-10-19T09:40:52 < dongs> each time i clicke that link i get triggered into the fucking stratosphere 2019-10-19T09:41:01 < ronox> ? 2019-10-19T09:42:21 < Cracki> he's like a singer and that code is like a howling dog 2019-10-19T09:42:40 < ronox> do you mean because of the /* being used like //? 2019-10-19T09:42:54 < Cracki> the loop is waiting for the ready flag to come on 2019-10-19T09:43:21 < Cracki> nah, he's complaining about the general arduino ecosystem because that's godawful 2019-10-19T09:43:41 < Cracki> so what you wanna look for is *some* mention of ready/rdy for the f4 adc 2019-10-19T09:43:46 < Cracki> or "complete" 2019-10-19T09:43:51 < Cracki> or general status flags 2019-10-19T09:43:53 < dongs> there isn't one, its just a shit in SR register 2019-10-19T09:44:10 < dongs> just remove that shit or stick a couple us delay in tehre if datasheet says it needs some startup time 2019-10-19T09:44:14 < ronox> what about the endofconversion flag? would that be suitable? 2019-10-19T09:44:20 < dongs> no 2019-10-19T09:44:23 < dongs> cuz youre not converting anything 2019-10-19T09:44:36 < Cracki> his f3 code already has a delay for startup 2019-10-19T09:44:43 < dongs> then just delete it 2019-10-19T09:44:53 < Cracki> ye 2019-10-19T09:45:35 < Cracki> not sure why that's even there. initializing nvic and timer and first run of the handler will take enough time 2019-10-19T09:45:55 < dongs> < dongs> Cracki: just looks like retarded copypasted shit 2019-10-19T09:45:58 < ronox> so, i dont need that while then? 2019-10-19T09:46:07 < dongs> neither does the origianl code 2019-10-19T09:46:09 < Cracki> comment it out, see if it catches fire 2019-10-19T09:46:31 < Cracki> it is copypasted ¯\_(ツ)_/¯ 2019-10-19T09:47:06 < ronox> otherwise, here is my choice of flags https://pastebin.com/HjcMbR5Z 2019-10-19T09:47:18 < Cracki> hm well, someone may have exactly followed the stdperiph docs because the setup ceremony is described there in that exact sequence 2019-10-19T09:47:31 < Cracki> ye remove 2019-10-19T09:47:40 < ronox> ok then 2019-10-19T09:47:48 < ronox> what exactly is that while loop doing btw? 2019-10-19T09:48:09 < Cracki> [19.10. 08:42:54] the loop is waiting for the ready flag to come on 2019-10-19T09:48:25 < ronox> yes, i mean, for what purpose, what triggers the flag 2019-10-19T09:48:32 < Cracki> the peripheral itself 2019-10-19T09:48:40 < ronox> assume i dont know how an ADC works, because i dont 2019-10-19T09:48:43 < Cracki> when it's "ready", whatever that implies 2019-10-19T09:48:54 < Cracki> that code told it some stuff 2019-10-19T09:49:02 < Cracki> then you say enable or run or whatever it was 2019-10-19T09:49:10 < Cracki> ADC_Cmd( ADC4, ENABLE ); 2019-10-19T09:49:13 < ronox> so, it halts the code because there is some time it takes between being enabled, and being ready then? 2019-10-19T09:49:26 < Cracki> yes, likely 2019-10-19T09:49:41 < Cracki> it may be just a few cycles 2019-10-19T09:50:14 < Cracki> if the docs say to wait, you wait. if the docs don't say that, no need to. 2019-10-19T09:50:33 < Cracki> if you get a bug, that's one place to revisit 2019-10-19T09:50:34 < ronox> the consequence of this would be that, the first few times it tries reading the ADC the values would be bad right 2019-10-19T09:50:59 < ronox> or would the issue persist 2019-10-19T09:51:36 < ronox> my assumption for that idea is that initADC only runs the one time 2019-10-19T09:52:11 < Cracki> no need to speculate. grab docs. 2019-10-19T09:53:25 < ronox> which ones? 2019-10-19T09:53:35 < ronox> oh wait i should check the periph lib 2019-10-19T09:54:11 < Cracki> look for a huge pdf that has literally this line in it: "Activate the ADC peripheral using ADC_Cmd() function." 2019-10-19T09:54:16 < Cracki> and see what comes after it 2019-10-19T09:54:41 < ronox> found it, was a compiled html file 2019-10-19T09:55:47 < Cracki> compare to the same docs but for F3 2019-10-19T09:55:47 < Jak_o_Shadows> Hmm. Whats a good way of connecting 9x16 digital inputs? 2019-10-19T09:56:01 < Cracki> what's "9x16 digital inputs" 2019-10-19T09:56:14 < Jak_o_Shadows> I found an old bit setting board - It has 9 rows of 16 toggle switches 2019-10-19T09:56:27 < Cracki> ah, literally that many digital inputs 2019-10-19T09:56:31 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-19T09:56:38 < Jak_o_Shadows> Each connected to a parallel port. 2019-10-19T09:56:39 < Cracki> grab a handful of shift registers 2019-10-19T09:56:47 < Jak_o_Shadows> All common ground annoyingly 2019-10-19T09:56:50 < Cracki> then hook up to spi peripheral 2019-10-19T09:57:30 < Jak_o_Shadows> Hmm. 2019-10-19T09:57:30 < Cracki> parallel in type 2019-10-19T09:57:32 < Cracki> not out 2019-10-19T09:57:37 < Jak_o_Shadows> It'd be nice to have a 16 bit one 2019-10-19T09:57:41 < Cracki> 74 165 for example, I think 2019-10-19T09:57:48 < Cracki> 18 eight bit ones 2019-10-19T09:57:55 < Cracki> or look for 16 bit regs 2019-10-19T09:58:15 < Cracki> there are 64 bit shift register ics, it appears 2019-10-19T09:58:36 < Jak_o_Shadows> I'm either looking for through hole or prebuilt modules tbh 2019-10-19T09:58:48 < Cracki> bag of 74 165 then, they're 8 bit 2019-10-19T09:58:57 < ronox> took me a while but finally found the section about ADC_cmd 2019-10-19T09:59:01 < ronox> reading it now 2019-10-19T09:59:39 < BrainDamage> many shift registers allow them to be easily chained serially, so 2 for each bank, then 9 rows of them 2019-10-19T09:59:42 < ronox> ok, it doesnt say anything other than what it does 2019-10-19T10:00:01 < Cracki> ronox, does it say to wait for ready or anything? 2019-10-19T10:00:07 < ronox> nope 2019-10-19T10:00:16 < Jak_o_Shadows> Yeah. 2019-10-19T10:00:18 < Jak_o_Shadows> That'll work 2019-10-19T10:00:25 < ronox> just says "Enables or disables the specified ADC peripheral." 2019-10-19T10:00:33 < Cracki> then adc_cmd() is the last step 2019-10-19T10:00:38 < Jak_o_Shadows> Now I just need a source of short DB25 cables 2019-10-19T10:00:43 < Cracki> look for the list where it's mentioned 2019-10-19T10:00:53 < Cracki> not the docs specific to the function itself 2019-10-19T10:02:42 < Cracki> my god that chm file is shit 2019-10-19T10:02:55 < Cracki> the doxygen html has more pages than are linked in the chm tree 2019-10-19T10:03:11 < Cracki> use the html's own navigation, not the chm tree 2019-10-19T10:03:41 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-19T10:04:29 < Cracki> ##### How to use this driver ##### 2019-10-19T10:04:37 < Cracki> that's what you're looking for 2019-10-19T10:05:53 < ronox> searching the chm is quite difficult 2019-10-19T10:06:59 < Cracki> anyway, even the f3 docs don't mention any waiting for that flag 2019-10-19T10:08:06 < Cracki> so maybe it failed and someone actually looked into the chip's reference manual 2019-10-19T10:08:45 < ronox> so i guess just keep going then and come back to it if it errors 2019-10-19T10:09:34 < Cracki> right 2019-10-19T10:10:50 < Cracki> the f3 example has a ADC_Config() that uses it... so that's where it must have come from 2019-10-19T10:11:40 < Cracki> f3 example has none of that 2019-10-19T10:12:04 < ronox> i see. is there any reason why the initadc function should be called every cycle? 2019-10-19T10:12:10 < Cracki> wat 2019-10-19T10:12:17 < ronox> i just noticed that its called in the main loop 2019-10-19T10:12:22 < Cracki> what gave you that impression 2019-10-19T10:12:39 < Cracki> show that code 2019-10-19T10:12:45 < Cracki> your paste only shows the function 2019-10-19T10:14:01 < ronox> https://github.com/cnlohr/colorchord/blob/master/embeddedstm32f303/main.c 2019-10-19T10:14:11 < Cracki> ok, let's debug your understanding 2019-10-19T10:14:15 < Cracki> because the code is right 2019-10-19T10:14:23 < ronox> upon further inspection it seems all the update code is executed in a while loop 2019-10-19T10:14:36 < Cracki> you said "main loop" 2019-10-19T10:14:45 < Cracki> the main function isn't a loop 2019-10-19T10:14:48 < Cracki> it's a function 2019-10-19T10:15:02 < Cracki> "a" main loop is any loop you consider to be central 2019-10-19T10:15:12 < Cracki> and initadc is only called once in main, and main only runs once 2019-10-19T10:15:13 < ronox> ah right, i forgot main runs through just the one time 2019-10-19T10:15:19 < Cracki> not arduino 2019-10-19T10:15:34 < Cracki> even in arduino, it's the loop function that loops, the init doesn't do that 2019-10-19T10:15:48 < ronox> yeah, sorry, i dont even know how i made that mistake 2019-10-19T10:15:51 < Cracki> and if you give arduino a literal main function instead of those two, that's also an option 2019-10-19T10:15:53 < ronox> anyway good to know 2019-10-19T10:16:20 < ronox> that its possible to just use a delay function in place of waiting for a flag, should a flag be needed to wait- for 2019-10-19T10:17:10 < Cracki> what specific f4 do you have 2019-10-19T10:17:36 < Cracki> for f3, take this exemplary RM and open 15.3.9 to find out what the RDY flag does: https://www.st.com/content/ccc/resource/technical/document/reference_manual/4a/19/6e/18/9d/92/43/32/DM00043574.pdf/files/DM00043574.pdf/jcr:content/translations/en.DM00043574.pdf 2019-10-19T10:19:42 -!- renn0xtk9 [~max@2a02:810d:1540:2448:293a:54a6:8926:6d85] has joined ##stm32 2019-10-19T10:20:13 < Cracki> this is one RM for some F4, section 13.3.1 has on-off control https://www.st.com/content/ccc/resource/technical/document/reference_manual/3d/6d/5a/66/b4/99/40/d4/DM00031020.pdf/files/DM00031020.pdf/jcr:content/translations/en.DM00031020.pdf 2019-10-19T10:20:28 < Cracki> compare, you see one calls it ADEN, the other ADON 2019-10-19T10:21:26 < Cracki> and the f3 has a voltage regulator that probably accounts for the startup, while the f4 is simpler and doesn't mention any startup time right there (although it likely needs some) 2019-10-19T10:21:35 < ronox> stm32f407vet6 2019-10-19T10:21:53 < Cracki> good, that random RM is coincidentally for 407 2019-10-19T10:22:15 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Read error: Connection reset by peer] 2019-10-19T10:23:02 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-19T10:23:52 < Cracki> good, that random RM is coincidentally for 407 2019-10-19T10:25:06 < ronox> woo, finished, time to compile and watch it do nothing haha 2019-10-19T10:25:53 < ronox> do i need to do anything in particular to be able to watch it do stuff with audio input, over serial monitor? 2019-10-19T10:26:13 < ronox> like, printing to serial voltages or timing or whatever so i can confirm its actually reading the input 2019-10-19T10:26:50 < ronox> by in particular i mean, without lagging the shit out of the board 2019-10-19T10:27:05 < ronox> i dont know how much resources is involved in serial output 2019-10-19T10:27:22 < Cracki> uh, nobody knows what that code actually does 2019-10-19T10:28:12 < Cracki> you might wanna invest in a cheap (but decent) multimeter, and a cheap logic analyzer (fx2lp based, "24M 8CH" for search term) 2019-10-19T10:28:13 < ronox> my understanding is using some mystery math, it reads an audio signal and then works out the octavess 2019-10-19T10:28:20 < ronox> colorchord is a visualizer 2019-10-19T10:28:23 < Cracki> ah 2019-10-19T10:28:35 < ronox> it visualizes over some RBGLED strips 2019-10-19T10:29:01 < Cracki> serial output can be cheap but don't do it from interrupt handlers 2019-10-19T10:29:02 < ronox> anyway, most of the reason why i didnt half ass my own sound reactive driver is because colorchord is insanely fast 2019-10-19T10:29:14 < ronox> with the driving side 2019-10-19T10:29:36 < ronox> the fact its a perfect octave visualizer / equalizer display is just gravy 2019-10-19T10:30:30 < ronox> to drive these rbg strips you have to send them information really fast, because the commands get passed through each LED, they have a series dataline, each reads, enterprets, and forwards to the next in a chain 2019-10-19T10:31:15 < ronox> so, the ability to pack a massive amount of information in each frame is paramount 2019-10-19T10:32:16 < ronox> oh wait i just realized something 2019-10-19T10:32:20 < Cracki> their fuckery is legendary around here 2019-10-19T10:32:31 < Cracki> waiting x microseconds to "latch" and all that 2019-10-19T10:32:32 < ronox> it doesnt matter if i slow it down because im just debugging 2019-10-19T10:32:51 < Cracki> the worst it can do is catch fire 2019-10-19T10:32:56 < ronox> XD 2019-10-19T10:33:19 < ronox> i only need to do this kind of debugging if its not reading anything at all 2019-10-19T10:33:29 < ronox> anywho, fingers crossed now as is, its just plugnplay 2019-10-19T10:33:43 < ronox> and doesnt require further modification to work 2019-10-19T10:57:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T10:59:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T11:08:27 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T11:09:23 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T11:16:44 < Jan-> happy weekend stm32 guys :) 2019-10-19T11:34:45 < Cracki> https://en.wikipedia.org/wiki/Dab_(dance)#Illegality_in_Saudi_Arabia 2019-10-19T11:35:01 < Cracki> public whipping maybe 2019-10-19T11:36:58 < Jan-> saudi arabia :( 2019-10-19T11:37:08 < Jan-> a friend of mine works there, she went there a few days ago 2019-10-19T11:37:12 < Jan-> doesn't like it 2019-10-19T11:41:54 < Cracki> the wippings shall continue until morality improves 2019-10-19T11:42:07 < Cracki> europe needs more islam 2019-10-19T11:42:20 < Jan-> katie says it's basically like working for the Empire from star wars 2019-10-19T11:42:38 < Cracki> rawr force choke 2019-10-19T11:42:53 < Jan-> it's like you know you're doing evil 2019-10-19T11:42:56 < Cracki> lots of stooges in shining armor 2019-10-19T11:42:57 < Jan-> but the money is quite good :) 2019-10-19T11:43:04 < specing> mohammad bin skywalker 2019-10-19T11:43:07 < Cracki> eh what evil do they even really do 2019-10-19T11:43:17 < Jan-> well they're awful to women 2019-10-19T11:43:47 < specing> Jan-: I think your (female?) friend is nuts if she went to SA 2019-10-19T11:43:47 < Cracki> is that because they're saudis or because they're muslim? 2019-10-19T11:43:55 < specing> you don't go to SA if you are female 2019-10-19T11:43:56 < specing> ever 2019-10-19T11:44:00 < Cracki> if you talk with a liberal about this, you get beheaded for your bigotry 2019-10-19T11:44:27 < Cracki> there's a reason so many were triggered by the "islam is right about women" flyers that got taped up at some colleges a few weeks ago 2019-10-19T11:44:46 < Cracki> you go there if you want a husband who's into beating you 2019-10-19T11:45:11 < Cracki> I wouldn't set foot in that country 2019-10-19T11:45:38 < Jan-> you would for what she's getting paid. 2019-10-19T11:45:43 < Cracki> they would escort me to the roof of the nearest building I'm sure of it 2019-10-19T11:46:04 < Cracki> and no amount of money will make me risk that 2019-10-19T11:46:56 < Jan-> if you want to hear the stories 2019-10-19T11:47:14 < Jan-> Katie is a pilot, the second time she went to saudi she flew an aircraft down there 2019-10-19T11:47:18 < Cracki> I have heard about the things saudis pay women to do 2019-10-19T11:47:19 < Jan-> landed at a saudi military airfield 2019-10-19T11:47:23 < Jan-> then wasn't allowed to drive to the hotel. 2019-10-19T11:47:26 < Jan-> :D 2019-10-19T11:47:27 < Cracki> lol 2019-10-19T11:47:37 < Cracki> so she got driven? 2019-10-19T11:47:43 < Jan-> went with a buddy. 2019-10-19T11:47:45 < Cracki> I know they're quite new to the letting women drive thing 2019-10-19T11:48:08 < Jan-> anyway I have to go I'll catch you guys later ;) 2019-10-19T11:48:10 < Cracki> yes, you always need a husband or brother or someone else to hold your leash 2019-10-19T11:48:43 < Cracki> don't steal their sand. the punishment is getting it shoved up some orifices. 2019-10-19T11:49:48 -!- renn0xtk9 [~max@2a02:810d:1540:2448:293a:54a6:8926:6d85] has quit [Quit: Konversation terminated!] 2019-10-19T12:04:03 < doomba> https://www.thewrap.com/game-of-thrones-who-is-that-young-girl-running-house-mormont/ 2019-10-19T12:04:19 < doomba> she killed an ice giant 2019-10-19T12:13:32 < doomba> https://en.wikipedia.org/wiki/Greta_Thunberg 2019-10-19T12:13:55 < doomba> she killed grand solar minimum 2019-10-19T12:14:37 < Cracki> plants start to die under 200 ppm CO2 2019-10-19T12:15:06 < Cracki> >a minimum of 150 ppm to survive [stanford.edu] 2019-10-19T12:15:26 < doomba> yes. there's too much CO2. we need to eat the babies. 2019-10-19T12:15:28 < Cracki> so... they basically hate plants 2019-10-19T12:15:36 < Cracki> yes they hate the babies too 2019-10-19T12:16:02 < Cracki> https://www.youtube.com/watch?v=epwUTVUwB7A 2019-10-19T12:16:27 < Cracki> swedish professor says to eat babies. SWEDEN fuck yeah 2019-10-19T12:16:41 < doomba> is that who she is? 2019-10-19T12:16:51 < Cracki> no, she just wants to eat babies 2019-10-19T12:17:26 < doomba> either she is legit batshit or it was a top-tier troll 2019-10-19T12:17:27 < Cracki> https://www.foxnews.com/world/swedish-scientist-eat-human-flesh-climate-change 2019-10-19T12:17:29 < doomba> either way i want to marry her 2019-10-19T12:18:03 < Cracki> Soylent(R) Homo, the new flavor 2019-10-19T12:18:23 < Cracki> man, the jokes write themselves. it's fun to be right. 2019-10-19T12:19:48 < doomba> trust in greta 2019-10-19T12:20:01 < doomba> she's the only one who can save us 2019-10-19T12:20:02 < Cracki> https://i1.wp.com/stonetoss.com/wp-content/uploads/2019/10/capitalism-and-fascism-comic.png 2019-10-19T12:38:44 < antto> that woman is hungry, feed her some babies so she sh*ts up and calms down 2019-10-19T12:39:03 < antto> muh cat would approve 2019-10-19T12:43:02 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-19T12:48:47 < Cracki> I'm still tickled how those xr doomstay cultists got owned by london subway commuters 2019-10-19T12:48:53 < Cracki> *day 2019-10-19T12:49:50 < doomba> doomstayers 2019-10-19T12:49:52 < doomba> they stay in the doom 2019-10-19T12:52:06 < Cracki> that would be splendid 2019-10-19T12:53:59 < Cracki> oh, nobody talked about the farmers in NL the past few weeks 2019-10-19T12:59:41 < Cracki> music https://pr0gramm.com/top/wiedermisch/3197113 2019-10-19T13:27:32 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 265 seconds] 2019-10-19T13:27:46 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-19T14:26:35 -!- RocketScientist [~User@31.192.14.195] has joined ##stm32 2019-10-19T14:26:59 < RocketScientist> Hi 2019-10-19T14:27:44 < RocketScientist> Developing software and firmware for quadcopter ground tests 2019-10-19T14:27:49 < RocketScientist> Hardware: 2019-10-19T14:28:50 < RocketScientist> <---USB -> MPU9250 -> STM32F130C8T6 -> 4xESC ->MOTORS 2019-10-19T14:29:02 < RocketScientist> Software: 2019-10-19T14:29:56 < RocketScientist> Qt/Cpp written UI for controlling everything with that four motors and gyroscope, PID settings, realtime AHRS, etc.. 2019-10-19T14:30:02 < RocketScientist> I have a question 2019-10-19T14:30:51 < RocketScientist> What to name it? "Aerial vehicle bench test kit" is ok? English is not my native language. 2019-10-19T14:31:42 < catphish> that sounds cool 2019-10-19T14:31:53 < catphish> UAVBench 2019-10-19T14:31:59 < catphish> DroneBench 2019-10-19T14:32:51 < catphish> AerialBenchKit, DroneBenchKit, UAVBenchKit 2019-10-19T14:33:43 < RocketScientist> catphish: Thanks, I was just wondering if "bench" in "Aerial vehicle bench test kit" was representing my idea correctly. 2019-10-19T14:35:58 < RocketScientist> Kit's main purpose is to provide easy testing way for heavy lift quadcopters, for example when you are trying to test something like this: https://cms.qz.com/wp-content/uploads/2015/06/screen-shot-2015-06-26-at-3-12-27-pm.png?w=857 2019-10-19T14:39:15 < RocketScientist> When you are focused on motors/weight/performance design, to get rid of RC hobby electronics problems. Only single STM32 BluePill + MPU9250 module and you will easily test lift power to weight/size relationship. 2019-10-19T14:43:18 < antto> i'd call it BrrrrrrWATCHOUT! 2019-10-19T14:58:32 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has joined ##stm32 2019-10-19T14:59:08 -!- mitrax [mitrax@lfbn-ncy-1-393-156.w83-196.abo.wanadoo.fr] has quit [] 2019-10-19T15:04:22 < zyp> RocketScientist, how are you planning to test it? sounds similar to what I were doing years ago 2019-10-19T15:05:31 < antto> deliver unwanted "presents" to people from a black list? 2019-10-19T15:05:44 < zyp> https://bin.jvnv.net/file/uhket.jpg for instance 2019-10-19T15:07:02 < Steffanx> dont forget to post the "after" photos zyp :P 2019-10-19T15:07:20 < Steffanx> or are those of before you went for this setup 2019-10-19T15:07:41 < antto> oh wait, is that attached to the whole box? 2019-10-19T15:09:54 < Jak_o_Shadows> Is the NSS1 pin worth using for SPI on the F1s? 2019-10-19T15:15:41 < RocketScientist> zyp: pasteall.org/pic/show.php?id=a0f7cf9a03349992d067e694a3361114 2019-10-19T15:22:01 < RocketScientist> zyp: https://streamable.com/rb368 2019-10-19T15:23:21 < RocketScientist> It will have all the safety options, to keep quadcopter alive if something goes wrong. So will save people from chaining quadcopters on the ground. 2019-10-19T15:24:04 < RocketScientist> if imu detects unusual movement it will just stop immediately, with detailed explanations why it stopped. 2019-10-19T15:27:56 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has quit [Quit: Leaving] 2019-10-19T16:03:39 <@englishman> hasn't all this quadcopter stuff been done to death around 2012 2019-10-19T16:04:15 < antto> hai there breggzitman 2019-10-19T16:04:27 <@englishman> like half the people in this channel have sold stm32 quadcopter equipment and stopped because noone cares about drones anymore 2019-10-19T16:04:50 < Jan-> I think lots of people care about drones 2019-10-19T16:04:58 < Jan-> it's just that there's a billion people making them as it's quite asy 2019-10-19T16:08:03 < doomba> the draconian laws pretty much killed it 2019-10-19T16:08:06 * Jan- attempts to figure out how CMSIS works 2019-10-19T16:08:08 * Jan- sobs 2019-10-19T16:09:02 < antto> a mountain of text creates magical connections in a symphony of chaos 2019-10-19T16:09:09 < antto> sorta 2019-10-19T16:09:18 < Jan-> yeah 2019-10-19T16:10:09 < Jan-> as I understand it there's not much restriction on distributing it 2019-10-19T16:10:37 < Jan-> so I'm trying to figure out what it takes to say, get serial ports and a to d converters working and stuff. 2019-10-19T16:10:47 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-19T16:10:58 < antto> reading the datasheets? 2019-10-19T16:10:59 < Jan-> If it's like a thousand lines of crap just to get those things set up then I'm screwed, but if not, maybe it's doable. 2019-10-19T16:11:36 < antto> well, first you gotta get to a working main(); with the chip all started, clocks and sh*t configured properly 2019-10-19T16:11:49 < Jan-> yeah I'm aware 2019-10-19T16:12:00 * Jan- didn't really want to do this :/ 2019-10-19T16:12:34 < Jan-> I mean even before that I've got to get an actual working build environment set up 2019-10-19T16:12:43 < antto> of course 2019-10-19T16:12:46 < antto> how else 2019-10-19T16:12:53 < Jan-> with the bits of cmsis that I need for an stm32f407, makefiles, all that jazz 2019-10-19T16:13:35 < antto> and you're on crapdows 2019-10-19T16:13:48 < Jan-> I'm not sure that makes much difference to some extent 2019-10-19T16:14:24 < antto> wellz.. 2019-10-19T16:14:38 < Jan-> Apparently I need something called STSW-STM32065, which is the "stm32f4xx standard peripherals library" 2019-10-19T16:15:30 < Jan-> Though apparently ST consider it superseded by cube. 2019-10-19T16:29:17 -!- jadew [~rcc@unaffiliated/jadew] has quit [Ping timeout: 240 seconds] 2019-10-19T16:39:28 < RocketScientist> englishman: if people sold quadcopter equipment as you say, that means someone bought that equipment, isn't it? 2019-10-19T16:50:24 < Jan-> sorry if this is a stupid question, but I understand that cmsis is a set of headers underlying all the HALs that we talk about 2019-10-19T16:50:32 < Jan-> is there somewhere I can actually get that stuff 2019-10-19T16:50:51 < Jan-> I started at https://developer.arm.com/tools-and-software/embedded/cmsis 2019-10-19T16:53:52 < Jan-> eventually ended up here: https://developer.arm.com/embedded/cmsis/cmsis-packs/devices/STMicroelectronics/STM32F407VETx 2019-10-19T16:54:12 < Jan-> It wants to give me a .pack file. 2019-10-19T16:56:16 < Thorn> Jan-: open cube, select 'manage packages', select the package for your mcu family, let cube download it 2019-10-19T16:57:57 < Jan-> I was trying to avoid using cube 2019-10-19T16:58:04 < Jan-> the HAL is incredibly wordy 2019-10-19T16:59:16 < Thorn> cmsis (complete with stm32-specific definitions) will also be in that package 2019-10-19T16:59:32 < Thorn> that's how you get it 2019-10-19T16:59:52 < Thorn> you can download the package manually, search for stm32cubef4 (or whatever family) 2019-10-19T17:00:32 < Jan-> it turns out the .pack file is actually just a zip 2019-10-19T17:00:57 < Jan-> I'm just sort of sanity checking this to see if (for instance) firing up a UART is going to require 1500 lines of code. 2019-10-19T17:25:57 < Steffanx> No it does not. 2019-10-19T17:27:05 < Jan-> that would be good. 2019-10-19T17:32:13 < Thorn> https://www.youtube.com/playlist?list=PLnMKNibPkDnGzoPVm2BJJS27DPec84shH&fbclid=IwAR2crepT96fjeyDbq8xzqhffZB3EOFJvmgzNOeKmS5cA5py1G-ieLlwGETs 2019-10-19T18:06:09 < kakimir64> anyone has played with newlibs mallocs etc. in pro level? 2019-10-19T18:07:21 < kakimir64> I mean know them inside out 2019-10-19T18:08:32 < specing> Jan-: it'll take 150_000 lines of code 2019-10-19T18:08:37 < specing> at the bare minimum 2019-10-19T18:20:30 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-19T18:21:33 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-19T18:27:14 < srk> find . -name '*.hs' | xargs wc -l 2019-10-19T18:27:15 < srk> 156709 total 2019-10-19T18:27:19 < srk> not bad estimate! 2019-10-19T18:50:03 < Steffanx> But with haskell you can be sure it works and safe. 2019-10-19T18:50:52 < Jan-> I'm not even at that stage yet 2019-10-19T18:50:56 < karlp> you can be sure it executes your model perfectly... 2019-10-19T18:51:00 < Jan-> I have no idea what a basic development environment for cmsis even looks like 2019-10-19T18:51:01 < karlp> not that y our model is useful :) 2019-10-19T18:53:57 < Jan-> god I hope specing wasn't serious :/ 2019-10-19T18:58:49 < mawk> HAL isn't that wordy, you just have long lines 2019-10-19T18:58:57 < mawk> just set up the UART in the cube MX view 2019-10-19T18:59:09 < mawk> but, why are you talking about hal, weren't you on libopencm3 yesterday or the day before ? 2019-10-19T19:03:56 < Jan-> I'm looking around for the best way to do it. 2019-10-19T19:04:13 < Jan-> libopencm3 seems OK, but both it and the arduino thing are LGPL so you have to give away your firmware. 2019-10-19T19:04:15 < bitmask> can you use an n channel mosfet as a high side switch? 2019-10-19T19:04:16 < Jan-> Can't really do that. 2019-10-19T19:04:25 < mawk> LGPL doesn't mean you give away your firmware 2019-10-19T19:04:29 < mawk> the L in LGPL isn't here for nothing 2019-10-19T19:04:45 < Jan-> Actually it does, but I'm not really here to argue about it 2019-10-19T19:04:52 < mawk> lol 2019-10-19T19:05:09 < mawk> you're free to link with the library as long as you provide a way to swap out the library 2019-10-19T19:05:17 < mawk> eg, provide the .o files and linker scripts and whatever 2019-10-19T19:05:26 < Jan-> Right and then you've given away your firmware. 2019-10-19T19:05:31 < mawk> ? 2019-10-19T19:05:32 < Jan-> On top of which I have no idea how to do anyt of that, 2019-10-19T19:05:43 < mawk> .o isn't your firmware, it's compiled code 2019-10-19T19:05:58 < Jan-> What's your point. 2019-10-19T19:06:07 < Jan-> You have then given away your firmware. 2019-10-19T19:06:18 < mawk> in compiled form 2019-10-19T19:06:32 < Jan-> Yes? 2019-10-19T19:06:39 < mawk> so what's your problem with that 2019-10-19T19:06:49 < Jan-> Because you've then given away the firmware. 2019-10-19T19:06:51 < mawk> lol 2019-10-19T19:06:52 < mawk> so what 2019-10-19T19:06:59 < mawk> it's not source code, you're protected 2019-10-19T19:07:05 < Jan-> Well, I have no idea if this thing will be successful or not. 2019-10-19T19:07:17 < mawk> if it requires any action looking more or like disassembly, every copyright law under the sun protects you 2019-10-19T19:07:20 < Jan-> But if we create a situation where we have to give away the firmware, it becomes stupidly easy to clone us. 2019-10-19T19:07:26 < zyp> it's pretty easy to dig out Jan-'s super secret five-dimensional tables from a firmware binary 2019-10-19T19:07:29 < Thorn> bitmask: yes but you'll have to bootstrap it or use something like a transformer driver 2019-10-19T19:07:47 < Jan-> LGPL was not designed for firmware on embedded systems. 2019-10-19T19:07:53 < bitmask> can I just use it as a low side switch? arent low side frowned upon? 2019-10-19T19:07:54 < Jan-> and as such it doesn't work very well :/ 2019-10-19T19:08:16 < bitmask> using mosfet as a battery disconnect 2019-10-19T19:08:32 < Jan-> bitmask I think using mosfets as a low side switch is pretty normal. 2019-10-19T19:08:40 < bitmask> oh ok, 2019-10-19T19:08:44 < bitmask> that makes things easier 2019-10-19T19:09:02 < Thorn> bitmask: that's how dw01 works for example 2019-10-19T19:09:05 < Jan-> N channel mosfets have better performance than P channel ones. 2019-10-19T19:09:14 < Jan-> though if you have some reason to use a high side switch, you can get a P-fet and use that. 2019-10-19T19:09:17 < bitmask> dw01? the battery charger? 2019-10-19T19:09:49 < zyp> Jan-, to be fair, you could do a dynamic linker on the target, keeping the LGPL part separate so it can be replaced without touching the rest of the firmware 2019-10-19T19:09:50 < mawk> you underestimate the chinese ability to extract firmware, Jan- 2019-10-19T19:09:57 < mawk> and yeah like zyp says 2019-10-19T19:10:04 < mawk> you could provide an alternate way of swapping out libopencm3.a 2019-10-19T19:10:07 < Jan-> zyp: I thought that! But yikes it would be tricksy. 2019-10-19T19:10:16 < zyp> yeah 2019-10-19T19:10:24 < mawk> not so much tricky, I did a dynamic linker in a few days for my kernel project 2019-10-19T19:10:33 < mawk> it's just about parsing ELF and filling up relocations 2019-10-19T19:10:39 < Jan-> It would also be complicated to the point of complete nuts-ness to try to create a compatible cm3 object. 2019-10-19T19:10:42 < mawk> and you don't have to do the funky relocs we have on x86 2019-10-19T19:10:45 < mawk> no PLT/GOT and so on 2019-10-19T19:11:11 < zyp> mawk, dynamic linkers are easy when everything is running from ram and can be rewritten 2019-10-19T19:11:19 < mawk> yeah 2019-10-19T19:11:21 < zyp> bit harder when you've got XIP flash 2019-10-19T19:11:32 < Jan-> I'm not sure you could do it when you don't have as much ram as you have flash 2019-10-19T19:11:40 < Jan-> maybe you could do it page by page but it would be horribly complicetd 2019-10-19T19:11:42 < mawk> yeah I can see the complication 2019-10-19T19:11:52 < mawk> then a jump table would be nice, but not very feasible on embedded 2019-10-19T19:12:04 < mawk> that's what the PLT is for on x86, all the .so are in read-only pages since they're shared 2019-10-19T19:12:20 < Jan-> anyway the point is that the lgpl stuff isn't really usable for commerical projects for the same reason people say don't use arduino on AVR for commercial projects. 2019-10-19T19:12:23 < mawk> so you have a jump table for all the symbols you require from the .so that the dynamic linker fills up at start (or on demand, without the lazy attribute) 2019-10-19T19:12:45 < Jan-> I think the mbed stuff isn't lgpl (haven't checked) but their IDE just didn't work for us. 2019-10-19T19:12:47 < mawk> I don't know what the performance penalty of that would be, but at least it takes no extra space 2019-10-19T19:12:58 < mawk> you can just get the mbed libs without the IDE, Jan- 2019-10-19T19:13:14 < mawk> in the IDE export one sample project, you get the libs locally 2019-10-19T19:13:21 < Jan-> yeah, but I don't even know how to do that 2019-10-19T19:13:51 < Jan-> right now I am working on the idea of just starting with literally nothing, just an empty folder, and getting the appropriate cmsis files and working up from there. But that is a nightmare. 2019-10-19T19:14:00 < BrainDamage> mawk: since the addresses are hardcoded at compile time, at worst one extra unconditional jump, at best no extra jump at all 2019-10-19T19:14:14 < BrainDamage> as in, one single jump to function call 2019-10-19T19:14:44 < mawk> yes 2019-10-19T19:14:47 < BrainDamage> best case is if the compiler optimizes the double useless jump 2019-10-19T19:15:10 < mawk> but we're talking about a dynamic linking 2019-10-19T19:15:18 < mawk> since Jan- doesn't want people to have the .o on their hands 2019-10-19T19:15:42 < mawk> so we have to have a PLT-like system 2019-10-19T19:15:44 < Jan-> I don't think I was proposing to actually do this guys :) 2019-10-19T19:16:00 < mawk> yes, don't worry 2019-10-19T19:16:05 < mawk> you'll invent something for humanity 2019-10-19T19:16:12 < mawk> LGPL for all corporations that are afraid of their users 2019-10-19T19:17:13 < Jan-> it's not really their users 2019-10-19T19:17:16 < Jan-> anyone can just demand the code 2019-10-19T19:19:27 < mawk> really ? grsecurity used to give code only to customers 2019-10-19T19:19:51 < mawk> and stamped the code to be able to punish customers that leaked it or something 2019-10-19T19:21:09 < mawk> or another option Jan- give karlp a truckload of cash for private licensing 2019-10-19T19:21:39 < BrainDamage> not just him, but all other contribs too 2019-10-19T19:21:44 < mawk> yeah 2019-10-19T19:21:48 < mawk> let the cash trickle 2019-10-19T19:21:52 * mawk quickly pushes something to libopencm3 2019-10-19T19:22:10 < Jan-> that's a big mistake that a lot of open source projects make 2019-10-19T19:22:23 < Jan-> they literally don't have any license other than gpl 2019-10-19T19:22:41 < BrainDamage> it's not a mistake if it's intentional 2019-10-19T19:23:03 < BrainDamage> allowing to change a license easily can be a loophole to bypass the gpl terms 2019-10-19T19:23:12 < Jan-> that's sort of the idea 2019-10-19T19:23:22 < BrainDamage> the gpl works as in 'if I don 2019-10-19T19:23:32 < mawk> all my code is MIT, hire me 2019-10-19T19:23:35 < BrainDamage> if I don't want to make money out of this, neither should you 2019-10-19T19:23:58 < BrainDamage> ( granted, it doesn't cover every case, but I think it's the spirit ) 2019-10-19T19:24:08 < Jan-> what's wrong with everyone making money 2019-10-19T19:24:11 < Jan-> anyway 2019-10-19T19:24:20 < mawk> lol 2019-10-19T19:24:23 < Jan-> are there any HALs that I can actually use for something that might become a commercial project 2019-10-19T19:24:24 < mawk> is that a serious question ? 2019-10-19T19:24:41 < mawk> you just said you don't want your users to hack on your devices for a vaporous "clone" fear 2019-10-19T19:24:42 < srk> my stack is BSD :D 2019-10-19T19:24:44 < mawk> that's one of the reasons right here 2019-10-19T19:25:02 < mawk> chinese will clone you even if you don't provide your firmware, don't worry 2019-10-19T19:25:09 < Jan-> possibly they will 2019-10-19T19:25:14 < Jan-> but it does make them work a bit harder 2019-10-19T19:25:22 < Jan-> and possibly their clone will be shitty, in a market that does care a bit. 2019-10-19T19:26:24 < mawk> you bought a stlink clone 2019-10-19T19:26:33 < srk> you can trigger flips in flash protection bits and read firmware that way :) 2019-10-19T19:27:12 < srk> what's wrong with stlink clones? 2019-10-19T19:27:31 < Jan-> mawk st are selling chips not programmers 2019-10-19T19:27:40 < Jan-> I think the issue with mbed is that it's a whole operating system. 2019-10-19T19:27:42 < BrainDamage> they are selling programmers too 2019-10-19T19:27:44 < Jan-> and I thought it was completely overkill. 2019-10-19T19:27:45 < mawk> st is selling stlink 2019-10-19T19:27:55 < mawk> mbed includes a RTOS too yeah, but you don't have to use it 2019-10-19T19:29:54 < Jan-> I downloaded the blinky example 2019-10-19T19:30:59 < Jan-> so now I have a bunch of files 2019-10-19T19:31:10 < Jan-> no makefile involved 2019-10-19T19:32:12 < Jan-> so how do I even 2019-10-19T19:36:41 < Jan-> according to the readme I do need mbed installed to use it 2019-10-19T19:36:43 < Jan-> so that's no go 2019-10-19T19:40:30 < bitmask> hmmm, what size pads (without a through hole) would you use to solder wires to for a rocker switch that isn't switching high/any current 2019-10-19T19:41:03 < mawk> lol Jan- 2019-10-19T19:41:05 < mawk> are you serious 2019-10-19T19:41:17 < mawk> you're trying to do a commercial product but a small difficulty like something to download makes you say you won't use mbed 2019-10-19T19:41:25 < mawk> there are very good reasons to not use mbed, but this one isn't one of them 2019-10-19T19:41:25 < Jan-> apparently I have to "mbed import mbed-os-example-blinky" 2019-10-19T19:41:34 < Jan-> for that I'd need a command called "mbed" 2019-10-19T19:41:37 < Jan-> which I don't have. 2019-10-19T19:41:44 < mawk> [18:41:03] .:mawk:. lol Jan- 2019-10-19T19:41:44 < mawk> [18:41:04] .:mawk:. are you serious 2019-10-19T19:41:44 < mawk> [18:41:17] .:mawk:. you're trying to do a commercial product but a small difficulty like something to download makes you say you won't use mbed 2019-10-19T19:41:46 < mawk> sorry 2019-10-19T19:41:48 < mawk> https://github.com/ARMmbed/mbed-cli 2019-10-19T19:42:09 < Jan-> is that not what we installed the other day 2019-10-19T19:42:21 < mawk> I don't know, I wasn't here maybe 2019-10-19T19:42:28 < mawk> here's the installer: https://github.com/ARMmbed/mbed-cli-windows-installer/releases/download/v0.4.10/Mbed_installer_v0.4.10.exe 2019-10-19T19:42:30 < mawk> do you remember it ? 2019-10-19T19:42:38 < Jan-> no not that 2019-10-19T19:42:48 < Jan-> well hang on 2019-10-19T19:43:03 < Jan-> I had MbedStudio-0.7.0.exe 2019-10-19T19:43:36 < srk> https://github.com/0xc0170/mbed_gcc_makefile 2019-10-19T19:43:39 < srk> :D 2019-10-19T19:43:58 < srk> too old, but you get the idea 2019-10-19T19:44:33 < Jan-> I do? 2019-10-19T19:44:40 < Jan-> can I actually use that and build the project with make? 2019-10-19T19:44:45 < srk> there's even arduino-makefile 2019-10-19T19:45:08 < srk> I've tried searching for 'mbed makefile' 2019-10-19T19:45:31 < Jan-> no I can't 2019-10-19T19:45:32 < Jan-> makefile:31: Platforms: No such file or directory 2019-10-19T19:45:33 < srk> or you can use cli to "export" to makefile 2019-10-19T19:45:46 < mawk> the link I gave is specifically mbed for CLI 2019-10-19T19:45:53 < Jan-> I'm getting it mawk 2019-10-19T19:45:58 < oz4ga> mbed-cli can export to make and gcc 2019-10-19T19:46:00 < Jan-> see if it makes any difference 2019-10-19T19:46:08 < Jan-> the mbed install I have right now is so broken it won't even uninstall 2019-10-19T19:47:15 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T19:47:50 < Jan-> during installation it says "the driver could not be installed. no mbed microcontrollers were found". 2019-10-19T19:47:56 < Jan-> I have an stlink plugged in, will that not do? 2019-10-19T19:48:01 < oz4ga> mbed-cli does not need an uninstaller. you just remove it. Of cause if you didn't user virtuels-env-win you have a lot of python modules left over in you global python installation 2019-10-19T19:50:28 < Jan-> anyway it failed at the installing-serial-driver part 2019-10-19T19:50:42 < Jan-> I do apparently have an mbed command now though 2019-10-19T19:50:53 < Jan-> so it failed and yet didn't remove the partial installation again.... 2019-10-19T19:51:16 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-19T19:51:36 < srk> you expect windows installers to be atomic? :D 2019-10-19T19:52:19 < Jan-> they should be 2019-10-19T19:52:32 < Jan-> they completely CAN be, if whoever writes them does it right 2019-10-19T19:59:01 < aandrew> https://pastebin.com/cmx1jY4L 2019-10-19T19:59:13 -!- con3|2 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-19T19:59:16 < aandrew> my shitty take on a collectd binary protocol emitter 2019-10-19T20:00:41 < aandrew> haven't tested multiple values (where the "value" is [type][type][type][value][value][value]) but it should work 2019-10-19T20:01:05 < aandrew> 0: 00 00 00 0d 66 6f 6f 5f 68 6f 73 74 00 00 01 00 ....foo_host.... 2019-10-19T20:01:08 < aandrew> 16: 0c 00 00 00 00 5d ab 39 86 00 04 00 0e 74 65 73 .....].9.....tes 2019-10-19T20:01:11 < aandrew> 32: 74 5f 74 79 70 65 00 00 06 00 0f 00 01 00 00 00 t_type.......... 2019-10-19T20:01:14 < aandrew> 48: 00 04 d2 00 00 00 ...... 2019-10-19T20:02:16 -!- con3|3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-19T20:02:58 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 252 seconds] 2019-10-19T20:04:04 -!- con3|2 [~kvirc@154.119.40.228] has quit [Ping timeout: 252 seconds] 2019-10-19T20:05:00 -!- con3|3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-19T20:06:09 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-19T20:12:50 -!- ekaologik [~quassel@p5DE86603.dip0.t-ipconnect.de] has joined ##stm32 2019-10-19T20:14:43 < bitmask> does a low side switch raise the gnd reference to slightly above 0V? 2019-10-19T20:15:35 < BrainDamage> yes 2019-10-19T20:15:54 < BrainDamage> and an high side switch lowers vdd slowly below the input 2019-10-19T20:15:56 < bitmask> is that a problem? 2019-10-19T20:16:05 < BrainDamage> depends how much 2019-10-19T20:16:13 < mawk> the gnd reference of the switched device 2019-10-19T20:16:38 < BrainDamage> for your case: 10mOhm*20A=0.2V 2019-10-19T20:16:40 < mawk> so you have to see if the drop in voltage is acceptable, and if the device is referenced to ground elsewhere 2019-10-19T20:17:20 < bitmask> what does that do? will it affect adc readings and stuff? 2019-10-19T20:17:34 < mawk> what are you switching ? 2019-10-19T20:17:44 < bitmask> everything, im using it as a battery disconnect 2019-10-19T20:17:49 < Jan-> what's the difference between a .bin and a .elf 2019-10-19T20:17:58 < BrainDamage> then yes 2019-10-19T20:18:00 < mawk> .bin is raw code, elf is elf headers 2019-10-19T20:18:07 < Jan-> ah. 2019-10-19T20:18:08 < bitmask> crap 2019-10-19T20:18:11 < Jan-> can I convert bin to elf somehow? 2019-10-19T20:18:25 < mawk> no Jan- , or it a way that loses information maybe 2019-10-19T20:18:30 < BrainDamage> you may be able to put the adc ref directly to the bat though 2019-10-19T20:18:31 < mawk> if you know the target MCU 2019-10-19T20:18:38 < BrainDamage> check vdd to vref tolerances 2019-10-19T20:18:40 < Jan-> can openocd send the .bin or does it need .elf 2019-10-19T20:18:53 < BrainDamage> or just use the internal ref 2019-10-19T20:19:00 < mawk> BrainDamage or put the measured signal with respect to the shifted gnd ? 2019-10-19T20:19:01 < BrainDamage> which is independent from the supply 2019-10-19T20:19:08 < BrainDamage> or that 2019-10-19T20:19:10 < mawk> then the shift will cancel out 2019-10-19T20:19:28 < bitmask> I'm using the internal ref 2019-10-19T20:19:41 < mawk> then it's tied to GND of the mcu 2019-10-19T20:19:53 < mawk> you have no AVSS pin ? 2019-10-19T20:20:03 < bitmask> I do 2019-10-19T20:20:10 < BrainDamage> what are you measuring? 2019-10-19T20:20:16 < bitmask> thermistor 2019-10-19T20:20:20 < BrainDamage> are you measuring something powered by the same bat? 2019-10-19T20:20:28 < BrainDamage> if yes, you'll have no problems 2019-10-19T20:20:50 < bitmask> yea its powered by the 5v regulator that the mcu uses 2019-10-19T20:21:05 < BrainDamage> because both the mcu and the thermistor will have the same floating gnd 2019-10-19T20:21:16 < bitmask> I see 2019-10-19T20:21:21 < BrainDamage> it'll eat away some dynamic 2019-10-19T20:21:25 < mawk> both work Jan- 2019-10-19T20:21:29 < Jan-> okay 2019-10-19T20:21:35 < Jan-> final question before I can try this 2019-10-19T20:21:45 < Jan-> apparently I have to enter "mbed compile -m K64F -t ARM" 2019-10-19T20:21:53 < Jan-> but ARM and K64F might have to be different. 2019-10-19T20:22:06 < Jan-> any idea what they might need to be? 2019-10-19T20:22:07 < BrainDamage> as, you should not let the adc's voltage to reach vdd-0.2/vdd-0.3v 2019-10-19T20:22:11 < BrainDamage> but that's easy 2019-10-19T20:22:26 < BrainDamage> and since you're using the internal vref, chances are you are already doing that 2019-10-19T20:22:36 < mawk> if you dump project from mbed web page it gives the command you need, otherwise you need to read manuals 2019-10-19T20:22:39 < bitmask> right, im using a 1.1v ref 2019-10-19T20:23:03 < bitmask> so not even close 2019-10-19T20:26:15 < Jan-> so k64f is a development board, the FRDM-K64F by NXP. 2019-10-19T20:26:27 < BrainDamage> oh and for the record, mosfets often have lower resistance than mechanical switches 2019-10-19T20:26:33 < Jan-> or at least presumably it is 2019-10-19T20:26:47 < BrainDamage> so don't think that a switch will necessarily let you ignore the drop either 2019-10-19T20:27:02 < mawk> you have a custom board Jan- , I forgot 2019-10-19T20:27:11 < mawk> you have to define stuff manually here and there now 2019-10-19T20:27:19 < Jan-> how 2019-10-19T20:27:21 < Jan-> Happy to 2019-10-19T20:27:23 < Jan-> just how? 2019-10-19T20:27:45 < bitmask> k 2019-10-19T20:27:46 < mawk> I don't know lol 2019-10-19T20:27:53 < mawk> look around in mbed-cli docs 2019-10-19T20:28:04 < Jan-> there is actually a reference to the board I have on the mbed site, https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-19T20:28:07 < mawk> you'll have to define some pins, clock speed 2019-10-19T20:28:28 < mawk> that's an user page 2019-10-19T20:28:33 < mawk> but yes follow that 2019-10-19T20:29:02 < mawk> you have a clone of seeed arch max apparently 2019-10-19T20:29:14 < Jan-> yes I found that part 2019-10-19T20:29:41 < Jan-> presumably there's a code to go after -m that means "seed arch max" 2019-10-19T20:30:07 < Jan-> that page only mentions using the online compiler 2019-10-19T20:34:14 < mawk> use the online compiler and download the project 2019-10-19T20:34:32 < Jan-> docs say "To find your board's target name, use --supported." 2019-10-19T20:34:35 < mawk> I think that's quick to do 2019-10-19T20:34:38 < mawk> ah 2019-10-19T20:34:40 < Jan-> but "mbed --supported" doesn't do much 2019-10-19T20:34:51 < Jan-> "too few arguments" 2019-10-19T20:35:00 < mawk> add --help 2019-10-19T20:35:02 < mawk> or -h 2019-10-19T20:35:31 < Jan-> what like "mbed --supported --help" 2019-10-19T20:37:31 < mawk> yes 2019-10-19T20:37:33 < antto> aww, command line interface 2019-10-19T20:38:40 < Jan-> "mbed --supported --help" does the same thing as just "mbed --help" 2019-10-19T20:39:07 * antto hacks into Jan-'s comput0r, downloads and burns debian10-netinst.iso onto one of the attached USB sticks, and puts "format C:\" in autoexec.bat 2019-10-19T20:39:40 < Jan-> how would that even help, it'd be the same software :) 2019-10-19T20:40:02 < antto> it gon work smoother 2019-10-19T20:41:03 < antto> as smooth as senpai chorus insult rhapsody 2019-10-19T20:41:44 < Jan-> oh "target" not "targets" 2019-10-19T20:41:51 < Jan-> no wait that's wrong too 2019-10-19T20:46:27 < Jan-> so I tried "mbed target fnord --global" in the hope that it might give me a list of possible targets 2019-10-19T20:46:33 < Jan-> but no, it just set the target to "fnord" quite happily 2019-10-19T20:49:38 < aandrew> hm, maybe I wrote that collectd shit for nothing, influxdb has [[udp]] which is just a text line protocol I could write to with sprintf() 2019-10-19T20:55:19 < RocketScientist> Anyone from Germany, please send me Sugar Free CocaCola ingredients label photo, I want to compare with local what is sold here, after 5+ years of avoiding it I wanted to buy one and on label was some cheap sweeteners mixed with aspartame (I wonder no one is carrying about our health, reason is only money, however I don't say aspartame is a good thing), and there was no sucralose on label also. So I want to figure out, it was not with us 2019-10-19T20:55:19 < RocketScientist> because it's more expensive or it's not any more in CocaCola at all. 2019-10-19T20:56:30 < antto> u drink coca cola? 2019-10-19T20:57:03 < specing> drinking cocacola is bad for your health 2019-10-19T20:57:06 < specing> more news at 11 2019-10-19T20:57:11 < antto> yeah, drink pepsi ;P~ 2019-10-19T20:57:27 < antto> pepsi is noicer 2019-10-19T21:01:04 < Jan-> all examples for mbed command line give "mbed compile --target K64F", am I wrong for assuming it should be something else? 2019-10-19T21:02:54 < Cracki> all those assume a kinetis k64f 2019-10-19T21:03:05 < Jan-> OK so what's the code for "seed arch max" 2019-10-19T21:03:14 < Cracki> wat 2019-10-19T21:03:34 < Jan-> apparently I need a differen target code for this board (which is reportedly compatible with seed arch max) 2019-10-19T21:03:41 < Jan-> the target codes are not listed anywhere I can find 2019-10-19T21:04:29 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 264 seconds] 2019-10-19T21:04:56 < Jan-> I tried "mbed detect" but it just says no targets found, I'm not sure what it's looking for 2019-10-19T21:05:18 < Cracki> mbed compile -S https://os.mbed.com/teams/Mbed-IoT-workshop/wiki/2-Getting-started-with-Mbed-CLI- 2019-10-19T21:06:24 < Cracki> this also looks related https://os.mbed.com/docs/mbed-os/v5.14/reference/configuration.html 2019-10-19T21:07:28 < Cracki> here someone also complains about no way to list targets: https://github.com/ARMmbed/mbed-cli/issues/244 2019-10-19T21:07:50 < Jan-> yes I tried that 2019-10-19T21:08:19 < Cracki> aha! THERE they mention the string "seed arch max" https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-19T21:08:29 < Jan-> that's the board I have 2019-10-19T21:09:02 < Cracki> this says "arch_max" is the target https://os.mbed.com/platforms/Seeed-Arch-Max/ 2019-10-19T21:09:21 < Cracki> you need to put yourself on the waiting list for a neuralink 2019-10-19T21:10:35 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-19T21:12:45 < Jan-> also I have the gcc arm variant somewhere 2019-10-19T21:12:52 < Jan-> so should I maybe put GCC_ARM for -t 2019-10-19T21:14:43 -!- con3 [~kvirc@154.119.40.228] has quit [Client Quit] 2019-10-19T21:16:51 < Jan-> wow this is taking an incredible amount of time to compile for a blinky 2019-10-19T21:24:31 < Cracki> if it's as retarded as arduino, it removes rebuilds everything every time because it doesn't keep the build directory around 2019-10-19T21:27:14 < Jan-> it errored ou anyway 2019-10-19T21:28:12 < Jan-> any ideas? https://pastebin.com/HKzHXPp4 2019-10-19T21:28:18 < qyx> aandrew: yeah the text protocol is ~easy~ 2019-10-19T21:28:55 < bitmask> I feel retarded but I'm not sure how to tell what wattage resistor I'd need for a mosfet gate. If its 12V the gate can pull a lot of current really quickly until the cap is charged. if I used a normal 150 ohm resistor wouldnt that be a watt? 2019-10-19T21:30:41 < BrainDamage> bitmask: it's a watt only for few nanoseconds 2019-10-19T21:30:55 < bitmask> so an 0805 resistor say would be fine? 2019-10-19T21:31:04 < qyx> yes 2019-10-19T21:31:19 < bitmask> interesting, ok, wasnt sure how long/short of a burst it could handle 2019-10-19T21:31:57 < bitmask> what value would you use? 2019-10-19T21:32:09 < qyx> it depends on your requirements 2019-10-19T21:32:23 < BrainDamage> couple hundred ohms, you don't have speed requirements 2019-10-19T21:32:25 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d98:fc9:760f:f6fc] has joined ##stm32 2019-10-19T21:32:28 < bitmask> its a battery disconnect so its not switching at all 2019-10-19T21:32:36 < bitmask> ok 2019-10-19T21:32:37 < qyx> for a battery disconnect, higher value one is better (slower dV/dt) 2019-10-19T21:32:37 < BrainDamage> just not blow up 2019-10-19T21:33:08 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d98:fc9:760f:f6fc] has quit [Client Quit] 2019-10-19T21:33:15 < bitmask> 1K would even be fine right? 2019-10-19T21:33:17 < qyx> some batt protectors even recommend a cap from the gate 2019-10-19T21:33:31 < qyx> to slow it down even more 2019-10-19T21:34:05 < bitmask> I think i'll go with 1K so that I dont have to order another value resistor 2019-10-19T21:34:13 < qyx> I was about to write that 2019-10-19T21:34:19 < qyx> use some generic value 2019-10-19T21:35:50 < aandrew> hm, can anyone explain in simpler terms the difference between PBUF_RAM and PBUF_POOL? 2019-10-19T21:36:01 < aandrew> isn't one a fragmented payload and i have to be careful when filling it? 2019-10-19T21:37:28 < PaulFertser> Jan-: it tells you none of the files included in the project have main() function. 2019-10-19T21:38:58 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T21:40:34 < Jan-> PaulFertser: then why didn't it say so. 2019-10-19T21:40:44 < PaulFertser> Jan-: it said just that 2019-10-19T21:40:45 -!- renn0xtk9 [~max@2a02:810d:1540:2448:926:f056:1d3d:222f] has joined ##stm32 2019-10-19T21:40:57 < aandrew> Jan-: what does "m.c:77: undefined reference to `main' 2019-10-19T21:40:59 < aandrew> mean to you? 2019-10-19T21:41:24 -!- renn0xtk9 [~max@2a02:810d:1540:2448:926:f056:1d3d:222f] has quit [Client Quit] 2019-10-19T21:41:30 < Jan-> is m.c a file with a single character name 2019-10-19T21:42:06 < PaulFertser> Jan-: no, it's a long filename with path starting on previous line 2019-10-19T21:42:10 < PaulFertser> Doesn't matter anyway. 2019-10-19T21:42:19 < PaulFertser> You need to write main() 2019-10-19T21:42:31 < Jan-> I shouldn't do, this is a test project 2019-10-19T21:43:03 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-19T21:43:04 < PaulFertser> Then you failed your test :) 2019-10-19T21:43:44 < Jan-> okay. 2019-10-19T21:43:58 < Jan-> I don't think it would have worked anyway because "mbed detect" didn't detect any boards 2019-10-19T21:44:05 < PaulFertser> Haha 2019-10-19T21:44:09 < Jan-> though I don't know that it's supposed to be able to detect an st-link 2019-10-19T21:44:13 < Jan-> ...is it? 2019-10-19T21:44:23 < PaulFertser> Compiling and linking doesn't really need a physical board. 2019-10-19T21:44:29 < Jan-> well sure 2019-10-19T21:44:42 < Jan-> but as I say even without this it wouldn't have actually worked 2019-10-19T21:44:59 < PaulFertser> mbed works with stlink-v2-1 afair, so probably stlink-v2 is not something it supports out of the box. But it doesn't matter, you can flash by other means. 2019-10-19T21:45:58 < Jan-> that's what I was thinking 2019-10-19T21:46:04 < Jan-> the problem is that it produces .bin files, apparently 2019-10-19T21:46:08 < Jan-> whereas I've mainly been using .elf 2019-10-19T21:46:21 < PaulFertser> It can't produce bin without producing an elf first, and I doubt it deletes the elfs. 2019-10-19T21:46:40 < Jan-> but usually I'd give openocd an elf 2019-10-19T21:46:46 < PaulFertser> bin files are meant to be copying to the "mbed emulated usb mass storage" of stlink-v2-1 2019-10-19T21:46:59 < PaulFertser> I'm sure it produces and elf first and keeps it. 2019-10-19T21:47:03 < Jan-> the main thing is, that's only a blinky sketch 2019-10-19T21:47:08 < Jan-> and it took 35 minutes to compile 2019-10-19T21:47:14 < PaulFertser> Even Keil produces ELFs (it just calls them .axf). 2019-10-19T21:47:29 < Jan-> it's like mbed is an entire realtime os 2019-10-19T21:47:47 < PaulFertser> Jan-: what didn't you like about libopencm3 btw? 2019-10-19T21:47:54 < Jan-> it's lgpl 2019-10-19T21:48:00 < Jan-> I'm not sure about mbed 2019-10-19T21:48:11 < Jan-> I think it's Apache 2019-10-19T21:48:36 < veverak> why is LGPL problem? 2019-10-19T21:48:44 < veverak> it means that only the original source has to stay open... 2019-10-19T21:48:50 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-19T21:49:03 < Jan-> well you have to give away your firmware... 2019-10-19T21:49:13 < PaulFertser> So far the only way you managed to program your board was requiring the stinky "open source" and everything else didn't make it? lol 2019-10-19T21:49:30 < Jan-> well, cube actually worked first time. 2019-10-19T21:49:32 < Jan-> it's the only thing that did 2019-10-19T21:49:43 < PaulFertser> Jan-: you can give away obfuscated sources or sometimes even just an object file, that's not really "giving away", it's not GPL. 2019-10-19T21:50:11 < Jan-> You're still giving away the firmware. I mean I have no idea how successful this might be, but I'm not going to basically ruin it at stage 1. 2019-10-19T21:50:18 < veverak> wat 2019-10-19T21:50:25 < veverak> how are you giving away the firmware? 2019-10-19T21:50:48 < veverak> you are only using something that has to stay open, (hence it's allready available to others) 2019-10-19T21:50:51 < Jan-> because you have to send it to anyone who asks... 2019-10-19T21:50:59 < Jan-> OK so here's what mbed now says: https://pastebin.com/LQFGTt0e 2019-10-19T21:51:05 < Cracki> the diff between lgpl and gpl is just this issue 2019-10-19T21:51:10 < veverak> the libopencm3 you use? yes 2019-10-19T21:51:12 < veverak> your code? no 2019-10-19T21:51:14 < Cracki> and where does it say that for LGPL you have to give anything away? 2019-10-19T21:51:22 < Cracki> what he said 2019-10-19T21:51:36 < veverak> yeah 2019-10-19T21:51:45 < veverak> LGPL does not force other code to be open 2019-10-19T21:51:46 < veverak> GPL does 2019-10-19T21:51:50 < PaulFertser> I guess Jan- is afraid of cl0n3rs. 2019-10-19T21:52:23 < Jan-> Honestly I think it'd be unlikely this became sufficiently popular for that to be an issue. 2019-10-19T21:52:31 < Jan-> But like I said it's not a smart way to start out is it. 2019-10-19T21:52:34 < PaulFertser> Her firmware will be so cool that the evil China people would lust even for the firmware binaries. 2019-10-19T21:52:47 < Jan-> I take your point paul 2019-10-19T21:52:52 < veverak> wat 2019-10-19T21:52:55 * veverak confused 2019-10-19T21:52:58 < Jan-> but honestly who'd start out by removing any possibility that you could get really successful. 2019-10-19T21:53:05 < Jan-> it's just not smart 2019-10-19T21:53:22 < veverak> Jan-: but how do you remove possibillity? 2019-10-19T21:53:25 < veverak> I don't understand 2019-10-19T21:53:48 < Jan-> You don't send anyone the firmware. OK fine you can imagine they might magically reverse engineer it out of the chip somehow but at least you made it hard. 2019-10-19T21:53:53 < PaulFertser> Jan-: do you really think anyone in this channel considers what you're doing for the last few days to be actually smart? 2019-10-19T21:54:07 < Jan-> PaulFertserl: No, but I'm used to being shit on by neckbeards. 2019-10-19T21:54:18 < veverak> Jan-: WAT 2019-10-19T21:54:49 < BrainDamage> you routinely sound pretty aggressive, it doesn't surprise me 2019-10-19T21:54:52 < veverak> I think you are just afraid of libopencm3 because you don't understand what the LGPL part means, not because you actually have valid reason 2019-10-19T21:54:54 < veverak> but nah 2019-10-19T21:55:16 < Jan-> veverak: if I use lgpl code, I have to send people the firmware on demand, that is pretty simple. 2019-10-19T21:55:55 < veverak> nope, you don't 2019-10-19T21:55:57 < veverak> next 2019-10-19T21:56:10 < veverak> or, to be specific 2019-10-19T21:56:12 < Jan-> veverak what you're saying isn't accurate, I'm not sure what you want me to say. 2019-10-19T21:56:21 < veverak> only the part that contains the libopencm3 have to be shared 2019-10-19T21:56:21 < Cracki> again, LGPL doesn't force you to open your own source, only the source you used. 2019-10-19T21:56:31 < veverak> but that is shared anyway 2019-10-19T21:56:36 < veverak> your part of code does not have to be shared 2019-10-19T21:56:42 < Jan-> My understanding is that you would have to hand out at least compiled object files for your stuff. 2019-10-19T21:56:54 < veverak> it's like using library -> you have to share the library, not the actuall code 2019-10-19T21:57:29 < Cracki> who says you have to hand out object files of your own stuff? 2019-10-19T21:57:32 < PaulFertser> Jan-: I think some people here are honestly trying to help you. I haven't seen so much hand-holding here for years. But let me put it blunt: your efforts look nothing like something to be expected from an engineer (or a smart student). You're going in circles actively refusing to learn the essence of what you're trying to achieve. 2019-10-19T21:57:38 < Cracki> lgpl was *made* for your situation 2019-10-19T21:57:55 < Jan-> PaulFertser: I have never ever claimed to be an engineer. But we all have to start somewhere. 2019-10-19T21:58:18 < PaulFertser> Cracki: yes, LGPL requires the copyright holder to make everybody able to relink the stuff with another version of the LGPL'd portion. 2019-10-19T21:58:24 < PaulFertser> veverak: ^ 2019-10-19T21:58:35 < Cracki> uh 2019-10-19T21:59:29 < Cracki> that may be an interpretation of the license 2019-10-19T21:59:30 < PaulFertser> Jan-: to start you need to dig. That's how any real knowledge is gained. 2019-10-19T21:59:47 < Jan-> what on earth do you think I've been doing. 2019-10-19T22:00:30 < veverak> Cracki: PaulFertser: based on what I've found, it's perfectly fine in case of dynamic libraries to not share anything 2019-10-19T22:00:42 < veverak> in case of static, it may be better to share linkable object 2019-10-19T22:00:53 < PaulFertser> "Use a suitable shared library mechanism for linking with the Library. A suitable mechanism is one that (a) uses at run time a copy of the Library already present on the user's computer system, and (b) will operate properly with a modified version of the Library that is interface-compatible with the Linked Version." 2019-10-19T22:01:08 < Jan-> veverak I think that's the root cause of the issue. the lgpl was designed to work with dynamically linked libraries on computers, not firmware. 2019-10-19T22:01:22 < Jan-> it's not a very smart choice for embedded code as it basically just turns back into almost the GPL 2019-10-19T22:01:42 < Cracki> an explanation/interpretation of the license is not part of the license's text. they may advise you to follow that but that's all 2019-10-19T22:01:48 < PaulFertser> Jan-: believe me or not, you're doing basically nothing, just randomly "trying things"... That's just my opinion, the others will correct me if I'm wrong. 2019-10-19T22:01:53 < veverak> Jan-: wat 2019-10-19T22:02:01 < PaulFertser> Cracki: I cited a part of the license itself. 2019-10-19T22:02:03 < Jan-> you want me to NOT try things? 2019-10-19T22:02:18 < Cracki> anyway, if you don't like lgpl, screw that and pick something that uses a less copyleftist license 2019-10-19T22:02:29 < Jan-> cracki that's exactly what I've done. 2019-10-19T22:02:44 < Cracki> indeed, damn that bs is part of the license 2019-10-19T22:02:47 < PaulFertser> Jan-: yes, to learn something you need to actually understand what and why you're doing, it's a matter of digging, not trying. 2019-10-19T22:03:06 < Cracki> I like MIT, it's so nice and short 2019-10-19T22:03:10 < Jan-> for what it's worth it's come up with a whole new error message. Now it says "[ERROR] Library name 'storage_tdb_internal' is not unique" 2019-10-19T22:03:33 < PaulFertser> Jan-: then it likely means you included some object file twice. Or some incompatible objects which both provide that symbol. 2019-10-19T22:03:44 < Jan-> I didn't include any object file. 2019-10-19T22:03:47 < Jan-> It's a test. 2019-10-19T22:03:53 < Cracki> some older discussion of lgpl in embedded situations https://lwn.net/Articles/510820/ 2019-10-19T22:03:54 < PaulFertser> Jan-: you say you know C but you get confused by the error messages like that. And you refuse to read a textbook on C... 2019-10-19T22:04:18 < Jan-> It's the results of "mbed import mbed-os-example-blinky" 2019-10-19T22:04:42 < Jan-> that doesn't compile on its own, it then says "warning could not find mbed program in current path... you can fix this by calling mbed new . in the root of your program" 2019-10-19T22:04:51 < Jan-> so I did that, and now it's got two copies of the whole mbed os library tree. 2019-10-19T22:04:56 < Jan-> hence that error. 2019-10-19T22:05:08 < Jan-> I'm just following instructions here. 2019-10-19T22:05:56 < PaulFertser> Cracki: I think with GPLv3 the comments about "not requiring to be able to run" are no longer relevant as GPLv3 prohibits tivoization and LGPLv3 is based on that. 2019-10-19T22:06:07 < Cracki> all gpl is cancer 2019-10-19T22:06:18 < PaulFertser> It was designed to be that way, yes. 2019-10-19T22:06:23 < Cracki> best to shun it like the plague it is 2019-10-19T22:06:43 < Cracki> it's almost like software patents 2019-10-19T22:06:59 < Cracki> similar diseases 2019-10-19T22:07:02 < Jan-> paul for what it's worth I have no doubt that given enough time and effort most of these things could be made to work but like I say, I'm just following instructions. 2019-10-19T22:07:10 < PaulFertser> Jan-: so probably you imported the project twice. Delete the whole directory and start afresh. 2019-10-19T22:07:23 < Jan-> that's how I ended up there in the first place 2019-10-19T22:07:32 < Jan-> I can "mbed import mbed-os-example-blinky" 2019-10-19T22:07:48 < Jan-> then "mbed compile -m arch_max -t GCC_ARM" 2019-10-19T22:07:54 < Jan-> then it'll produce the error about "could not find mbed program" 2019-10-19T22:08:03 < PaulFertser> Are you reading the official mbed documentation from the beginning and not omitting any chapters? 2019-10-19T22:08:29 < Jan-> I'm reading "README.md" which is titled "# Getting started example for Mbed OS" 2019-10-19T22:08:56 < Jan-> It tells me to install mbed CLI, which I have done. 2019-10-19T22:09:21 < Cracki> I find texts of a supposed gplv3 that says "this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM)." 2019-10-19T22:09:21 < PaulFertser> Jan-: do you have any degree in something? Have you already mastered some skills? 2019-10-19T22:09:34 < Jan-> I have a degree in history if it helps. 2019-10-19T22:09:49 < Jan-> I've written thousands and thousands of lines of code in c#, but then that's in a proper IDE. 2019-10-19T22:10:01 < PaulFertser> Cracki: aha, interesting, so installing in ROM is an option to tivoize device, thanks. That kinda makes sense. 2019-10-19T22:10:10 < veverak> oooh 2019-10-19T22:10:12 < veverak> cool 2019-10-19T22:10:12 < Cracki> note that's gplv3, not lgpl 2019-10-19T22:10:18 < Cracki> I have no idea why lgpl doesn't say this 2019-10-19T22:10:22 < PaulFertser> LGPLv3 is based upon GPLv3 2019-10-19T22:10:30 < Cracki> not sure that's lawyer-proof 2019-10-19T22:10:37 < PaulFertser> It says that it's GPLv3 with certain specific changes. 2019-10-19T22:10:51 < Cracki> ¯\_(ツ)_/¯ still infectious and leftist 2019-10-19T22:11:16 < Jan-> I think if people want to write code and make it useful to others just make it public domain. If someone is good enough to contribute great, if they can't what's the harm. 2019-10-19T22:11:27 < PaulFertser> Jan-: so history, cool, thanks for answering. You should be used to reading long books then I guess. 2019-10-19T22:11:36 < Jan-> Most of the people who complain about this stuff could never afford to enforce the license anyway. 2019-10-19T22:12:10 < jpa-> Jan-: public domain is poorly defined in many places, e.g. in finland it is apparently not possible to really public domain anything, so a specific license is actually better 2019-10-19T22:12:20 < jpa-> though one can also use cc-zero or similar 2019-10-19T22:12:31 < Jan-> jpa-: people often put clauses like that, "in case this isn't legally possible anyone's granted a license to do anything" etc 2019-10-19T22:12:41 < jpa-> yeah 2019-10-19T22:12:45 < Jan-> anyway I don't much care about finland. 2019-10-19T22:12:48 < Jan-> :) 2019-10-19T22:13:18 < Jan-> Apparently the missing link in getting that thing to compile was I had to type "cd mbed-os-example-blinky/mbed-os-example-blinky 2019-10-19T22:13:23 < Jan-> ...obviously. 2019-10-19T22:14:12 < PaulFertser> It says something like that here: https://os.mbed.com/docs/mbed-os/v5.14/quick-start/compiling-the-code.html 2019-10-19T22:14:21 < PaulFertser> That you're supposed to cd into it. 2019-10-19T22:14:26 < Jan-> yes paul *like* that 2019-10-19T22:14:27 < Jan-> but not THAT. 2019-10-19T22:15:04 < Jan-> OK yes we do have an mbed-os-example-blinky.elf in the BUILD/ARCH_MAX/GCC_ARM folder. 2019-10-19T22:23:37 < Jan-> Is ST's programming software likely to work with this 2019-10-19T22:23:42 < Jan-> or is it going to want something made by cube? 2019-10-19T22:26:28 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away] 2019-10-19T22:29:46 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-19T22:30:21 < mawk> st programming software will likely take a .bin 2019-10-19T22:30:27 < Jan-> it did 2019-10-19T22:30:30 < Jan-> just uploaded it 2019-10-19T22:30:34 < Jan-> no flashing though 2019-10-19T22:30:40 < Jan-> blinking 2019-10-19T22:30:41 < mawk> you uploaded the .elf ? 2019-10-19T22:30:46 < mawk> or the .bin 2019-10-19T22:30:48 < Jan-> no st's software wanted the bin 2019-10-19T22:30:54 < mawk> yeah 2019-10-19T22:31:02 < mawk> did you move around jumpers on the board ? 2019-10-19T22:31:09 < mawk> is BOOT0 in the 0 position 2019-10-19T22:31:25 < Jan-> no jumpers on it 2019-10-19T22:32:31 < mawk> and gdb says something about it ? if you start openocd then run gdb on that 2019-10-19T22:32:55 < mawk> or maybe you don't have a setup like that 2019-10-19T22:32:58 < Jan-> no, I don't. 2019-10-19T22:33:48 < Jan-> it's possible it's just blinking the wrong pins 2019-10-19T22:33:48 < mawk> just run openocd with no specific command and it will start a server, then you can connect your gdb to it and know what the mcu is doing 2019-10-19T22:33:51 < mawk> ah 2019-10-19T22:33:52 < mawk> yeah 2019-10-19T22:34:04 < mawk> you can put the pins by hand in the blinking example 2019-10-19T22:34:19 < Jan-> this is one of the things that is annoying, all these IDEs are going through a lot of work to try and automate something where I'd much rather just put the port and pin numbers in by hand. 2019-10-19T22:34:21 < Jan-> it'd be so much easier. 2019-10-19T22:34:41 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-19T22:34:56 < Jan-> It just seems to define it as "DigitalOut led1(LED1);" 2019-10-19T22:35:06 < Jan-> I don't know what LED1 points to 2019-10-19T22:36:14 < kakimir64> well shit 2019-10-19T22:36:54 < kakimir64> find the definition of led1 2019-10-19T22:37:04 < Jan-> I'm looking 2019-10-19T22:37:12 < kakimir64> there is a shortcut that finds selected keyword 2019-10-19T22:37:19 < kakimir64> from the whole projecty 2019-10-19T22:37:47 < Jan-> it's just... somewhere in mbed, which is like nearly a gigabyte 2019-10-19T22:37:55 < Jan-> 969MB 2019-10-19T22:38:27 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T22:38:50 < Jan-> the other thing that springs to mind is, just practically, if it takes like fifteen or twenty minutes to compile something, how practical is that really. 2019-10-19T22:38:54 < Jan-> iteration is going to be super slow 2019-10-19T22:39:52 < PaulFertser> Jan-: it's just for the first time 2019-10-19T22:40:46 < kakimir64> it's not just somewhere 2019-10-19T22:40:53 < Jan-> okay found it 2019-10-19T22:40:55 < PaulFertser> Jan-: btw, does history involves studying multiple big partially contradictory sources and then trying to construct some theory that would be explaining most of it? 2019-10-19T22:41:17 < Jan-> basically found it by following the folder structure 2019-10-19T22:41:24 < Jan-> it's in mbed-os\targets\TARGET_STM\TARGET_STM32F4\TARGET_STM32F407xE\TARGET_ARCH_MAX\PinNames.h 2019-10-19T22:42:13 < Jan-> PaulFertser I think you may have outlined my concerns with all this quite well there! :D 2019-10-19T22:42:34 < Jan-> okay so it says LED1 = PB_3, which is I'm pretty sure not right for this board. 2019-10-19T22:44:02 < kakimir64> what the fuck Jan- 2019-10-19T22:44:23 < kakimir64> so convenient place to have board configuration 2019-10-19T22:44:34 < Jan-> okay so they're pa6 and pa7 on this board. 2019-10-19T22:44:52 < Jan-> can I just change the main.cpp or should I change the PinNames.h 2019-10-19T22:47:50 < PaulFertser> Jan-: so does knowing some basics about "group dynamics", politology etc help one to distinguish between "reasonable probable" and "highly unlikely" history theories? 2019-10-19T22:48:07 < Jan-> Paul have you ever seen MSDN. 2019-10-19T22:48:12 < Jan-> That's how you document software. 2019-10-19T22:48:21 < Jan-> Then you don't have to make everything into a detective story. 2019-10-19T22:48:43 < qyx> what 2019-10-19T22:48:45 < qyx> msdn? 2019-10-19T22:49:12 < kakimir64> microsoft software development network? 2019-10-19T22:49:23 < qyx> I know what does it mean 2019-10-19T22:50:19 < PaulFertser> Jan-: I've spend many hours reading MSDN to understand how ACLs for files/directories work in Windows, yes. And it was a detective story all right. No kidding. Plenty of information on many pages without many links between them. 2019-10-19T22:50:48 < Jan-> if it's any consolation st's uploader utility is being a pain right now 2019-10-19T22:51:04 < PaulFertser> And I didn't even need "programmer-level" info, I just wanted to fully understand how to use them. 2019-10-19T22:52:06 < Jan-> So anyway, I changed it to read "DigitalOut led1(PA_6);" and still no blinking. 2019-10-19T22:52:25 < Jan-> I don't really know if that's right though 2019-10-19T22:53:24 < kakimir64> is that even C language? 2019-10-19T22:53:44 < Jan-> I don't know. It's the mbed stuff. 2019-10-19T22:53:59 < Jan-> later in the code it just says "led1 = !led1" to toggle them 2019-10-19T22:54:04 < Jan-> and no I guess it's c++ really 2019-10-19T22:54:11 < kakimir64> what is DigitalOut 2019-10-19T22:54:17 < Jan-> I have absolutely no idea. 2019-10-19T22:54:33 < Jan-> https://os.mbed.com/handbook/DigitalOut 2019-10-19T22:54:50 < kakimir64> oh it's a type 2019-10-19T22:55:07 < Jan-> I was wondering at the syntax 2019-10-19T22:55:11 < Jan-> it's weird 2019-10-19T22:55:24 < kakimir64> unless it's a type 2019-10-19T22:55:31 < kakimir64> or something 2019-10-19T22:55:47 < Jan-> yeah but doesn't that then make led1 a... function? er. 2019-10-19T22:55:51 < Jan-> or a variable name 2019-10-19T22:55:54 < Jan-> with parentheses after it? 2019-10-19T22:57:19 < PaulFertser> Jan-: it's an old C++ syntax to call object constructors. 2019-10-19T22:57:27 < Jan-> I know a bit of C 2019-10-19T22:57:29 < Jan-> I know next to no C++ 2019-10-19T22:57:37 < kakimir64> why are you coding with mbed? 2019-10-19T22:57:44 < Jan-> it's a long story 2019-10-19T22:58:05 < PaulFertser> Jan-: probably it's time to return to cube ide or whatever it's called? 2019-10-19T22:58:08 < kakimir64> tell me short version 2019-10-19T22:58:23 < Jan-> does the syntax "DigitalOut myled(PA_6);" look reasonable 2019-10-19T22:58:30 < PaulFertser> Jan-: it does 2019-10-19T22:59:13 -!- boB_K7IQ [~boB_K7IQ@c-98-247-195-108.hsd1.wa.comcast.net] has joined ##stm32 2019-10-19T22:59:25 < Jan-> I wonder if there's some less weird way of doing it 2019-10-19T22:59:33 < Jan-> but still it should work at least 2019-10-19T23:00:13 < PaulFertser> Jan-: the less weird way in modern C++ would be myled{PA_6}; 2019-10-19T23:00:23 < PaulFertser> :D 2019-10-19T23:00:41 < Jan-> I was hoping for PORTA &= 1 << 6 or something even less weird. 2019-10-19T23:00:42 < PaulFertser> (but it's really true) 2019-10-19T23:01:19 < PaulFertser> Jan-: what makes you unhappy with cubeide? It's meant for C programmers, works fast, has nice GUI to configure peripherals etc? 2019-10-19T23:01:22 < Jan-> the biggest problem I have right now is that the st-link utility fails to find the board if I unplug the USB cable. 2019-10-19T23:01:34 < Jan-> (after plugging it back in natch) 2019-10-19T23:02:04 < Jan-> I'm doing this with a screenreader, the gui is not my friend. also the code the gui generates is hideous, and the HAL is really wordy 2019-10-19T23:02:32 < Jan-> or as the cube HAL would say HAL_WORDINESS_EVALUATION_VERBAL_ENGLISH_TEXT_WRITTEN("really") 2019-10-19T23:02:53 < PaulFertser> But it works... 2019-10-19T23:03:19 < Jan-> I'm not sure I can deal with programming stuff where every third line is some sort of /* MACHINE GENERATED COMMENT */ 2019-10-19T23:03:32 < srk> Jan-: look for 'baremetal stm32' 2019-10-19T23:03:44 < PaulFertser> It's only peripheral initialisation. You do it once, export the code, do not touch ever again. 2019-10-19T23:03:53 < Jan-> actually if I could make openocd work I'd be happier 2019-10-19T23:03:58 < Jan-> I mean mbed seems OK 2019-10-19T23:04:03 < Jan-> weird c++ isms 2019-10-19T23:04:04 < PaulFertser> You had openocd working already. 2019-10-19T23:04:13 < Jan-> I had openocd working under platformio 2019-10-19T23:04:14 < PaulFertser> It was flashing your board just fine. 2019-10-19T23:04:23 < PaulFertser> You can use the same, just give it new elf. 2019-10-19T23:04:39 < Jan-> but then I need an openocd configuration file and that's another pile of pain and being yelled at by you. 2019-10-19T23:05:01 < PaulFertser> Nope, you had it working with just configuration files already present in upstream distribution. 2019-10-19T23:05:19 < PaulFertser> Nothing special was used, you can reuse the command that already worked. 2019-10-19T23:05:31 < Jan-> it's almost like I have to leave the stlink disconnected for a while then plug it back in... 2019-10-19T23:06:32 < PaulFertser> btw, if stlink utility doesn't work then openocd won't work too, most probably. 2019-10-19T23:06:38 < Jan-> I thought that. 2019-10-19T23:07:26 < Jan-> there was some utility somewhere that would list all the connected ones 2019-10-19T23:08:33 < Jan-> erk, is it possible that all this screwing around has killed either the stlink or the board 2019-10-19T23:08:46 < kakimir64> software breakpoints are a bitch 2019-10-19T23:09:00 < Jan-> oh god is it sitting at a breakpoint 2019-10-19T23:09:23 < kakimir64> today I learned about jlink software breakpoints 2019-10-19T23:09:59 < Steffanx> dont you like em? 2019-10-19T23:10:14 < kakimir64> I don't 2019-10-19T23:10:36 < Steffanx> what changed 2019-10-19T23:10:37 < kakimir64> unless I explicitly allow them 2019-10-19T23:10:44 < Jan-> ok well it's finding the stlink because the led is flashing red and blue 2019-10-19T23:10:59 < kakimir64> Steffanx: those use stack did you know? 2019-10-19T23:11:30 < Jan-> ok we have a blinking light! 2019-10-19T23:11:36 * Jan- runs around in circles, waving her arms 2019-10-19T23:11:52 < kakimir64> I have minimal size stacks and mpu watching over 2019-10-19T23:12:04 < kakimir64> memfaulthandler 2019-10-19T23:12:09 < kakimir64> I needed to ask a pro 2019-10-19T23:12:11 < Steffanx> it does kakimir64? 2019-10-19T23:12:21 < srk> Jan-: congrats! 2019-10-19T23:12:28 < kakimir64> I talked with a pro Steffanx 2019-10-19T23:12:50 < Jan-> ok now to prove it is actually running our code 2019-10-19T23:12:55 < Steffanx> And how would it use stack? 2019-10-19T23:13:00 < Jan-> I'll change it from wait(0.2) to wait(1) 2019-10-19T23:13:14 -!- boB_K7IQ [~boB_K7IQ@c-98-247-195-108.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-19T23:14:54 < Jan-> OK, so when I change it, it doesn't seem to alter the blink rate. 2019-10-19T23:15:03 < Jan-> Do I have to tell the compiler that I changed something in main.cpp or will it figure that out 2019-10-19T23:15:59 < Steffanx> Im still not convinced kakimir64 :P 2019-10-19T23:17:42 < PaulFertser> kakimir64: jlink software breakpoints using stack? 2019-10-19T23:18:14 < kakimir64> pro said so 2019-10-19T23:18:22 < Jan-> hm ok seems to work as expected now 2019-10-19T23:18:24 < PaulFertser> Jan-: build system (like make) figures it out automatically when you remake. 2019-10-19T23:18:50 < Jan-> I tried it again with a much simpler one that didn't include their whole multi threaded OS thing. 2019-10-19T23:18:58 < Jan-> just uses the main mbed.h and not the threaded stuff. 2019-10-19T23:19:04 < Steffanx> and how did the pro come to this conclusion kakimir64 2019-10-19T23:19:27 < Jan-> will it ignore something called main.cpp.bak 2019-10-19T23:20:04 < Steffanx> you'll see soon enough Jan- ;) 2019-10-19T23:20:11 < PaulFertser> "A: A RAM Code, specially designed for this purpose, sets and clears Flash Breakpoints extremely fast; on micros with fast flash the difference between breakpoints in RAM and flash is hardly noticeable." 2019-10-19T23:20:36 < PaulFertser> Do you mean that RAM code is stored in stack? I doubt so. 2019-10-19T23:20:57 < kakimir64> I don't know 2019-10-19T23:21:00 < kakimir64> pro said 2019-10-19T23:21:00 < PaulFertser> Or is it using current stacks much? Like how tight were your stacks? 2019-10-19T23:21:07 < kakimir64> really 2019-10-19T23:21:09 < PaulFertser> Can you cite him/her? 2019-10-19T23:21:13 < kakimir64> really really tight 2019-10-19T23:21:22 < Steffanx> i think his pro talks in finnisher language 2019-10-19T23:21:22 < kakimir64> but I disabled breakpoints 2019-10-19T23:21:32 < kakimir64> and memfault still 2019-10-19T23:22:42 < Jan-> I would prefer to use openocd I guess. 2019-10-19T23:22:48 < Jan-> But what I have now does work. 2019-10-19T23:23:43 < Steffanx> does your pro use irc and can he tell us in ##stm32 kakimir64? 2019-10-19T23:27:29 < Jan-> Does this "DigitalOut myled(PA_6)" thing imply that the pin access is going to be slow, or at least comparatively slow? 2019-10-19T23:27:46 < kakimir64> oh he didn't say it uses stack 2019-10-19T23:28:13 < kakimir64> only implied that I have ran out of stack 2019-10-19T23:28:38 < PaulFertser> Jan-: one would need to see the implementation and probably disassembly listing to tell for sure. 2019-10-19T23:28:53 < Jan-> Yes I was looking for where it's actually written but I have no clue where that would be. 2019-10-19T23:30:14 < BrainDamage> that's where ides help 2019-10-19T23:30:19 -!- ekaologik [~quassel@p5DE86603.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-19T23:30:34 < Jan-> if I could do this in visual studio I would have 2019-10-19T23:30:59 < BrainDamage> nothing forbids you to? 2019-10-19T23:31:18 < BrainDamage> c/c++ is the same no matter what compiler you use 2019-10-19T23:31:33 < BrainDamage> as long as you set the paths correctly in visual studio, you can use it like an editor 2019-10-19T23:31:48 < Jan-> I think that's just going to be another layer of pain I don't need 2019-10-19T23:33:02 < Jan-> there are gpio_set functions in mbed but it seems very much c++ oriented 2019-10-19T23:33:09 < PaulFertser> Jan-: with a library properly written and LTO enabled that pin access via a C++ class can be as fast as inline assembly. 2019-10-19T23:33:29 < PaulFertser> But I have no idea if mbed is properly written. 2019-10-19T23:33:32 < BrainDamage> you can always observe the disassembler code 2019-10-19T23:33:39 < Jan-> neither do I, and I don't really have any way of finding out 2019-10-19T23:34:11 < Jan-> I think the c++ thing might be a bit of a deal breaker anyway 2019-10-19T23:34:26 < Jan-> I think to get any of the advanced stuff going you'd have to be more of a programmer than I am. 2019-10-19T23:42:53 < Jan-> OK, in the end it seems to come down to something like "obj->GPIOMEMBASE->R_STATE_W_SET = obj->gpioMask;" 2019-10-19T23:43:27 < specing> ugly 2019-10-19T23:43:47 < Jan-> I guess I sort of get what it does, even though I'm not really a c++ person. 2019-10-19T23:46:07 < Jan-> is that like...three lookups 2019-10-19T23:48:06 < kakimir64> my memfault is caused by creating one task with xTaskCreateRestricted and others with xTaskCreate 2019-10-19T23:48:25 < kakimir64> 3 user configurable MPU regions are not cleared between tasks 2019-10-19T23:49:13 < specing> Jan-: thats C 2019-10-19T23:49:16 < specing> Jan-: not C++ 2019-10-19T23:49:27 < Jan-> well OK, point remains 2019-10-19T23:49:39 < kakimir64> task created with normal xTaskCreate applies preceeding tasks configurable MPU regions 2019-10-19T23:49:44 < specing> Jan-: try Ada, its cool 2019-10-19T23:59:08 < kakimir64> PaulFertser: Steffanx: pro friend reasons why sw breakpoint would use stack is to store at least return address --- Day changed Sun Oct 20 2019 2019-10-20T00:01:39 < kakimir64> when I use TAD script with server things like fault monitor stop working in IDE 2019-10-20T00:02:04 < kakimir64> so I need to turn off server and then see memfault address in case 2019-10-20T00:02:25 < Jan-> specing: Why Ada 2019-10-20T00:02:27 < kakimir64> without the rtos TAD script with server 2019-10-20T00:03:30 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-20T00:03:53 < specing> Jan-: because the code looks much nicer than C/C++ and it catches a lot of errors at compile-time 2019-10-20T00:04:44 < kakimir64> soon Jan- will be developing his product with Haskell 2019-10-20T00:05:01 < kakimir64> just keep taking inputs from this channel Jan- 2019-10-20T00:05:37 < Jan-> I'm just wondering if a microcontroller is really the place for a heavily object oriented approach. 2019-10-20T00:05:49 < specing> Jan-: it could be 2019-10-20T00:06:01 < specing> 128+ k flash mcus are quite common 2019-10-20T00:06:11 < specing> so there is a lot of space for silliness 2019-10-20T00:06:54 < Jan-> it's more speed I think 2019-10-20T00:07:00 < Jan-> but then again this one does go at 168mhz 2019-10-20T00:07:08 < kakimir64> javascript it is 2019-10-20T00:07:10 < specing> depends on what you want to do 2019-10-20T00:07:30 < specing> you can do the time-critical stuff in assembly regardless of object orientation or not 2019-10-20T00:07:46 < specing> and if it is that time critical, it should probably be analog or something 2019-10-20T00:08:15 < Jan-> I'm not sure if I want my gpio set code to use a bunch of lookups and a conditional. 2019-10-20T00:08:43 < specing> huh? 2019-10-20T00:09:00 < Jan-> maybe some of that would get compiled out. 2019-10-20T00:09:21 < specing> where did you find this obj->GPIOMEMBASE->R_STATE_W_SET = obj->gpioMask;" ? 2019-10-20T00:09:38 < Jan-> some obscure sub sub sub directory in mbed 2019-10-20T00:09:46 < Jan-> it's how the gpio setting is actually implemented 2019-10-20T00:12:08 < Jan-> it's in a function called gpio_write 2019-10-20T00:13:09 < Jan-> hmm there's a whole discusison on it https://os.mbed.com/users/igorsk/notebook/fast-gpio-with-c-templates/ 2019-10-20T00:15:04 < specing> hahaha 2019-10-20T00:16:18 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-20T00:18:34 < Jan-> what's funny 2019-10-20T00:23:01 < BrainDamage> how much do you have to strobe your gpio btw? 2019-10-20T00:23:19 < BrainDamage> because with mcu clocking at 200MHz, you need to do plenty of things wrong to run out of time 2019-10-20T00:24:16 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-20T00:26:36 -!- RocketScientist [~User@31.192.14.195] has quit [Quit: Leaving.] 2019-10-20T00:31:04 < PaulFertser> Jan-: that discussion is relevant but it was in 2010, these days LTO might produce very different results, so that needs to be reevaluated. 2019-10-20T00:37:41 < Jan-> sorry LTO is a data tape format in my brain 2019-10-20T00:38:11 < PaulFertser> Link Time Optimisation. 2019-10-20T00:38:52 < Jan-> I guess I'm just used to AVRs where you can set a pin in one cycle (or is it two) 2019-10-20T00:38:57 < Jan-> just by hitting the register 2019-10-20T00:39:16 < Jan-> ok maybe it's a read modify write if you want to do it independently of other pins on that port. 2019-10-20T00:41:54 < specing> on stm32 it is one cycle as well 2019-10-20T00:42:05 < specing> and without r-m-w 2019-10-20T00:45:47 < Jan-> it has hardware to do that right 2019-10-20T00:50:36 < kakimir64> eclipse stopped on it's tracks 2019-10-20T00:50:41 < kakimir64> fuufu 2019-10-20T00:52:05 * Jan- hands kakimir64 the visual studio installer 2019-10-20T00:52:08 * Jan- hides :) 2019-10-20T00:52:27 < kakimir64> good idea 2019-10-20T00:52:40 < kakimir64> does it come in debian .deb package? 2019-10-20T00:52:51 < Jan-> yeah just like eclipse comes in .exe installer :D 2019-10-20T00:53:16 < kakimir64> holy shit 2019-10-20T00:53:43 * Jan- offers kakimir64 a cookie because that was mean 2019-10-20T00:53:48 < Jan-> sorry kakimir64 :/ 2019-10-20T00:53:58 < kakimir64> visual studio is available for debian 2019-10-20T00:54:11 * Jan- stops for a moment 2019-10-20T00:54:13 < Jan-> ...seriously? 2019-10-20T00:55:31 < kakimir64> https://code.visualstudio.com/ or is it some lite version? 2019-10-20T00:55:50 < kakimir64> doesn't look all right 2019-10-20T00:55:57 * Jan- headscratch 2019-10-20T00:55:58 < Jan-> I uhoh 2019-10-20T00:56:00 < Jan-> I'm a windows person 2019-10-20T00:56:17 < aandrew> kakimir64: actually yes the visual studio code *does* have a .deb 2019-10-20T00:56:17 < kakimir64> that is not visual studio 2019-10-20T00:57:33 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-20T00:57:35 < oz4ga> it's an add on to visual studio 2019-10-20T00:58:03 < oz4ga> sry. I was misreading that. pse to forget 2019-10-20T00:58:49 < catphish> arfgh the UK has requested another extension to brexit, when will this shit end? 2019-10-20T00:59:20 < Jan-> never :( 2019-10-20T00:59:23 * Jan- sob 2019-10-20T00:59:54 < catphish> i just hope to god this extension comes with a referrendum to decide the outcome for good 2019-10-20T01:00:02 < Jan-> I'm not sure that's a great idea 2019-10-20T01:00:12 < Jan-> if people vote leave again it solves nothing 2019-10-20T01:00:22 < Jan-> if people vote stay the shitstorm will be neverending 2019-10-20T01:00:25 < catphish> it is if the referendum provides real answers 2019-10-20T01:00:46 < catphish> like "should the UK leave with the deal negotiated by fuckface (boris)" 2019-10-20T01:01:08 < catphish> and "should the uk leave the eu with no ongoing agreement" 2019-10-20T01:01:26 < catphish> and "should the uk just bow down to our eu overlords after all" 2019-10-20T01:01:29 < Jan-> even then "no ongoing agreement" is a huge range of possibilities 2019-10-20T01:01:44 < Steffanx> Would be funny if the EU says "fuck you" the 31st of october 2019-10-20T01:01:56 < Steffanx> No deal for you. 2019-10-20T01:01:56 < BrainDamage> yeah, the eu has full authority to refuse delays 2019-10-20T01:01:57 < Jan-> it would in some ways 2019-10-20T01:01:59 < Jan-> but bear in mind 2019-10-20T01:02:11 < Jan-> there are 60 plus million people whose lives will be affected by this 2019-10-20T01:02:12 < BrainDamage> it's just not in its interest to 2019-10-20T01:02:13 < catphish> i like how our PM send the EU 2 letters, one saying "the uk requests an extension" 2019-10-20T01:02:28 < catphish> and another saying "i think this is a bad idea, signed, Boris" 2019-10-20T01:02:51 < Steffanx> Wasnt there this story where Hungary could use their veto to make it a no deal Brexit? 2019-10-20T01:02:57 < PaulFertser> So Boris sent a letter requesting an extension? I thought he promised to never do that? 2019-10-20T01:02:58 < catphish> but seriously, since we've already decided the people get to choose, just let the people choose 2019-10-20T01:03:25 < catphish> have another referrendum in 2 parts, one on the decision to leave, then on the leaving arrangement 2019-10-20T01:03:34 < BrainDamage> PaulFertser: it's ok, it's unsigned 2019-10-20T01:03:40 < PaulFertser> Haha 2019-10-20T01:03:50 < catphish> PaulFertser: he did, he didn't sign it 2019-10-20T01:04:03 < catphish> then send another signed letter personally saying they shouldn't grant it 2019-10-20T01:04:07 < catphish> *sent 2019-10-20T01:04:11 < Steffanx> I like him already 2019-10-20T01:04:18 < PaulFertser> catphish: what if people vote to leave in a way that makes the Irish unhappy? Then it's all the same stupid political game again. 2019-10-20T01:04:52 < catphish> PaulFertser: again, it's been made clear by *every* political force that the only thing that matters of the opinion of the majority 2019-10-20T01:05:01 < Jan-> my conclusion is that it is not reasonably possible for the UK to leave the EU at this point. 2019-10-20T01:05:07 < catphish> PaulFertser: so we might as well just settle what that is and get on with it 2019-10-20T01:05:08 < Jan-> Which does raise a lot of questions, like "how the hell did that happen." 2019-10-20T01:05:13 < BrainDamage> scottish people are also plenty unhappy and feel cheated from their referendum 2019-10-20T01:05:16 < Jan-> We sleepwalked into it. 2019-10-20T01:05:19 < catphish> i don't like this populist rule, but it's clearly what's going to happen 2019-10-20T01:05:33 < catphish> so we might as well get on with it 2019-10-20T01:05:43 < kakimir64> how does MPU like undefined address space? 2019-10-20T01:05:59 < Cracki> you're calling EU populist rule? the EU is the opposite of populist. they do everything that's UNpopular. 2019-10-20T01:06:09 < BrainDamage> https://pbs.twimg.com/media/EHRW92iW4AAdIiu?format=jpg&name=medium 2019-10-20T01:06:44 < PaulFertser> catphish: I have a feeling the UK politicians know that the majority will later regret their choice and so they're trying to avoid the blame by all means. 2019-10-20T01:07:00 < catphish> Cracki: me? no? i'm calling the decision of the UK parlaiment to implement a referrendum on EU membership and then vow to uphold its vague result "populist rule" 2019-10-20T01:07:16 < BrainDamage> the whole vote was just a political tool that didn't mean to pass from the own organizers 2019-10-20T01:07:26 < Cracki> populism is a synonym for democracy 2019-10-20T01:07:30 < BrainDamage> so when it did, they found themselves completely unprepared 2019-10-20T01:07:34 < catphish> i don't mind direct politics, but this affair has proved that it needs to me more specific 2019-10-20T01:08:10 < Cracki> brits won't regret brexit if brexit is done right 2019-10-20T01:08:20 < Cracki> they will regret voting polticians into office that fuck this all up 2019-10-20T01:08:25 < catphish> Cracki: there are many forms of democracy, direct democracy is great in theory, but this incident clearly show that if you're going to do direct democracy, you need a way to quickly clarify results 2019-10-20T01:08:37 < PaulFertser> BrainDamage: haha, but what if the file itself was PGP or S/MIME signed? How can we know? Or was it a material paper? 2019-10-20T01:08:51 < Cracki> the politicians right tooth and nail against brexit. the divide is just that. 2019-10-20T01:08:54 < Cracki> *fight 2019-10-20T01:09:13 < BrainDamage> done right is a bit ambigous 2019-10-20T01:09:19 < catphish> Cracki: brexit is interesting in the way it's shown clearly how UK politics works (and doesn't) 2019-10-20T01:09:23 < Cracki> this whole situation with this "indecisiveness" is plainly constructed by politicians 2019-10-20T01:09:48 < Cracki> if they did their jobs, this would have been over with by now 2019-10-20T01:09:56 < catphish> Cracki: i don't understand, surely a referendum that is split 48/52 clearly shows indecision 2019-10-20T01:09:59 < BrainDamage> so can you clarify what means done right? 2019-10-20T01:10:02 < Cracki> including negotiations to get back anything people fear they'd lose 2019-10-20T01:10:06 < Cracki> done right? 2019-10-20T01:10:16 < Cracki> well, what's wrong about how they did brexit so far? 2019-10-20T01:10:20 < Cracki> dragging their feet 2019-10-20T01:10:33 < BrainDamage> that still doesn't answer me 2019-10-20T01:10:33 < catphish> the problem is that most people in the UK don't wany brexit 2019-10-20T01:10:36 < Cracki> voting again because they didn't like the results 2019-10-20T01:10:45 < catphish> but we have this referendum that we should honour 2019-10-20T01:10:48 < Cracki> how does that not answer your question? 2019-10-20T01:10:51 < catphish> *want 2019-10-20T01:11:10 < Cracki> right, the people voted, and politicians won't honor that result 2019-10-20T01:11:10 < BrainDamage> you've showed an example of what not to, it'd be an answer only if there'd be a single alternative 2019-10-20T01:11:19 < BrainDamage> but there's thousands of options 2019-10-20T01:11:21 < Cracki> uh 2019-10-20T01:11:27 < BrainDamage> so no, it doesn't answer me 2019-10-20T01:11:37 < Cracki> I said what I think they should _not_ do. that's a suggestion for improvement. 2019-10-20T01:11:43 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-20T01:11:49 < catphish> Cracki: some people voted, but those most impacted were not given a vote 2019-10-20T01:11:54 < Cracki> you're gonna nail me because I said "done right" instead of "done better"? 2019-10-20T01:12:02 < Cracki> catphish, who was not given a vote? 2019-10-20T01:12:16 < catphish> non-british uk residents 2019-10-20T01:12:16 < BrainDamage> no, you're magically sweeping away their problem into those words 2019-10-20T01:12:23 < Cracki> whose problem? 2019-10-20T01:12:26 < catphish> ie EU immigrants 2019-10-20T01:12:31 < BrainDamage> they don't know themselves what to do 2019-10-20T01:12:32 < Cracki> they aren't UK citizens 2019-10-20T01:12:37 < catphish> Cracki: right 2019-10-20T01:12:41 < Cracki> uk citizens get a vote. 2019-10-20T01:12:44 < BrainDamage> hence they are hesitating, and yes, part of that is cowardice 2019-10-20T01:12:46 < catphish> Cracki: right 2019-10-20T01:13:00 < catphish> only citizens were allowed to vote, not those most affected 2019-10-20T01:13:03 < BrainDamage> the uk has an exception from schengen treaty 2019-10-20T01:13:16 < BrainDamage> the only country in eu 2019-10-20T01:13:26 < Cracki> so what's wrong with giving citizens a vote and not the rest of the eu? 2019-10-20T01:13:27 < BrainDamage> i find it hilarious that they were concerned with eu immigrants 2019-10-20T01:13:27 < catphish> Cracki: but more importantly, the vote was too vague :( 2019-10-20T01:13:43 * Jan- is theoretically a non british UK resident. 2019-10-20T01:14:02 < catphish> Cracki: a small majority votes to leave, which seems reasonable, but most of the propaganda at the time spoke of "a good deal" 2019-10-20T01:14:03 < Cracki> they as a nation get to decide, not anyone else 2019-10-20T01:14:08 < Cracki> >small 2019-10-20T01:14:10 < Cracki> >majority 2019-10-20T01:14:18 < catphish> Cracki: and that's the problem, this deal has not been reachable :( 2019-10-20T01:14:30 < Steffanx> Did you vote mr catphish ? 2019-10-20T01:14:49 < Cracki> the lack of a good deal is the fault of those politicians, nobody else's 2019-10-20T01:15:18 < catphish> Steffanx: i did, i voted to remain, but i'm a pretty obvious demagraphic, a young UK national with international business interests and strong beliefs in EU policy 2019-10-20T01:15:19 < Cracki> and instead of saying "oh well we can't do anything about it" people should take their politicans to task about their failure to carry out the people's decision 2019-10-20T01:15:51 < BrainDamage> the EU has plain stated that they won't have access to the EU common market if they leave, which they assured they'd be able to in the referendum 2019-10-20T01:16:02 < Jan-> I don't know about "assured" 2019-10-20T01:16:03 < catphish> Steffanx: with that said, i believe in direct democracy, and if a majority of voters believe in a *specific* course of action, i think we are obliged to accept it 2019-10-20T01:16:19 < Cracki> the EU made threats 2019-10-20T01:16:20 < BrainDamage> Jan-: it was one of the points of the propaganda 2019-10-20T01:16:29 < Cracki> it's ok to call that bluff 2019-10-20T01:16:43 < Cracki> EU making threats is propaganda for staying in the EU? 2019-10-20T01:16:46 < catphish> Cracki: it's ok unless it cripples your economy :) 2019-10-20T01:16:49 < Cracki> it's one of the reasons people voted to leave 2019-10-20T01:17:01 < catphish> Cracki: yes it is 2019-10-20T01:17:03 < BrainDamage> no, it was one of the arguments of the pro leave people 2019-10-20T01:17:03 < Jan-> BrainDamage: yeah, but nobody took that seriously did th... oh. 2019-10-20T01:17:05 < Cracki> trade negotiations have been invented before the EU 2019-10-20T01:17:37 < catphish> Cracki: it remains to be seen how that will end, clearly no side will risk economic losses when it comes down to it, if possible 2019-10-20T01:17:42 < Cracki> there's no reason you can't negotiate trade with individual countries. there's no reason to assume that *trade* has to suffer just because "being in the EU" also means free travel 2019-10-20T01:18:17 < Cracki> blackmail by the EU is an excellent argument for getting the fuck out of that abusive relationship 2019-10-20T01:18:35 < catphish> i voted because i'm an EU citizen and i don't want my small part of the EU to try to break away and form a new state 2019-10-20T01:18:46 < BrainDamage> how is it blackmail? they wanted to leave and the answer is that they won't be back halfway 2019-10-20T01:18:55 < catphish> but i realise that some people feel differently 2019-10-20T01:19:00 < catphish> and that part id purely personal 2019-10-20T01:19:03 < Cracki> europe isn't gonna break apart, only the EU bureaucracy 2019-10-20T01:19:35 < Cracki> it's almost as if nobody remembers a time before UK was in the EU 2019-10-20T01:19:41 < catphish> a lot of this seems to come down to purely personal opinion about how widely you believe regulations should be made 2019-10-20T01:20:18 < catphish> Cracki: i was not alive during any such time, i was born en EU citizen 2019-10-20T01:20:27 < Cracki> I believe threats by the EU aren't as serious. the UK isn't the underdog, doesn't have to beg. the UK has a strong enough position to negotiate a good deal. 2019-10-20T01:20:41 < Cracki> you don't need to be. you only need to be aware that history exists. 2019-10-20T01:21:00 < Cracki> the world isn't gonna end. 2019-10-20T01:21:02 < catphish> you just said peope don't remember it 2019-10-20T01:21:13 < catphish> i was agreeing with you, i don't remember it, i wasn't alive 2019-10-20T01:21:16 < Cracki> you think the world is gonna end because you're flooded with doomsday propaganda 2019-10-20T01:21:47 < catphish> the main thing is that i see no benefit to leaving the EU 2019-10-20T01:21:55 < Cracki> 1975 is when the Uk entered the EU, so anyone that's 54 years or older was born into this. 2019-10-20T01:21:55 < catphish> i'd be giving up so much, but i don't see the gain 2019-10-20T01:22:01 < Cracki> so only "old white men" will remember. 2019-10-20T01:22:29 < Cracki> politics is poker 2019-10-20T01:22:42 < Cracki> it's eternal negotiation 2019-10-20T01:23:08 < catphish> politics seems to be a weird battle between feelings and rational analysis, where most people can be pursuaded to go with feelings 2019-10-20T01:23:18 < Cracki> maybe it's a better negotiating position to leave and start fresh 2019-10-20T01:23:19 < catphish> the trick is stirring the right feelings in people 2019-10-20T01:23:37 < Cracki> the problem is that loss causes fear, which disables rational thinking 2019-10-20T01:23:48 < catphish> Cracki: maybe it's a better negotiating position to remain part of a large block :) 2019-10-20T01:24:04 < catphish> yes, loss is bad 2019-10-20T01:24:43 < Cracki> again, this is a case of stockholm syndrome. you're suffering under an opaque power structure that "provides for you". you will lose comfort in the short (!) term but you will remove your whole country from this outside control 2019-10-20T01:25:14 < Cracki> you lose money when you buy food. do you feel bad about losing money then? 2019-10-20T01:25:20 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-20T01:25:25 < Cracki> here it's immaterial. 2019-10-20T01:25:25 < catphish> Cracki: it seems apparant that everyone has to live under some regime of control by someone else 2019-10-20T01:25:42 < catphish> Cracki: the question is what geographic area that control should cover 2019-10-20T01:25:58 < Cracki> I prefer control closer to the area it controls 2019-10-20T01:26:01 < Cracki> roman empire 2019-10-20T01:26:12 < Cracki> why did it end? many opinions, sure 2019-10-20T01:26:17 < Cracki> but its size was a factor 2019-10-20T01:26:27 < Cracki> places were ruled from far away 2019-10-20T01:26:48 < Cracki> and "far away" doesn't mean distance 2019-10-20T01:26:53 < catphish> closer control seems to have its advantages, the downside is beaurocracy and cost 2019-10-20T01:27:05 < Cracki> but rulers being very abstracted from the people they rule over 2019-10-20T01:27:16 < Cracki> bureaucracy is what you get in huge empires 2019-10-20T01:27:26 < Cracki> you want less of that? you want smaller government? 2019-10-20T01:27:33 < catphish> i'd love it if i could make laws for my own house, but i'd have a LOT of negotiating to do 2019-10-20T01:27:39 < Cracki> I'm not saying tighter control, but local decisions 2019-10-20T01:27:49 < Cracki> "closer" is ambiguous 2019-10-20T01:27:56 < catphish> like if i wanted to work in the nearby town, would they let me? who knows 2019-10-20T01:28:17 < Cracki> it's a slider. you don't wanna make rules for japan, do you? 2019-10-20T01:28:22 < catphish> they might want me to ban cocaine in my house if i was going to let my residents work in their town 2019-10-20T01:28:24 < Cracki> or india, or whatever 2019-10-20T01:28:33 < catphish> it's a slider indeed 2019-10-20T01:28:37 < Cracki> consistency has value. that's something the EU does right. 2019-10-20T01:28:48 < Cracki> what it sucks at is respecting local peculiarities. 2019-10-20T01:28:51 < catphish> local control is great, but it has costs when trying to integrate 2019-10-20T01:29:31 < catphish> global control means less hassle moving and communicating between places, but more compromise 2019-10-20T01:29:51 < Cracki> accidental complexity and essential complexity https://en.wikipedia.org/wiki/No_Silver_Bullet 2019-10-20T01:30:00 < mawk> there are nations already in place which budgets to integrate like you say 2019-10-20T01:30:00 < catphish> it's easy to find similar issues in the private sector 2019-10-20T01:30:04 < mawk> it's not like it's an impossible task 2019-10-20T01:30:05 < Cracki> accidental complexity is that bureaucracy the EU is building 2019-10-20T01:30:33 < catphish> look at facebook, central control, american rules, easy exchange of information 2019-10-20T01:30:38 < Cracki> easy abuse 2019-10-20T01:30:45 < Cracki> look at google 2019-10-20T01:30:49 < catphish> yep, abuse becomes rife at that scale 2019-10-20T01:30:56 < Cracki> have you EVER spoken to an actual human at google when you wanted soemthing from them? 2019-10-20T01:31:02 < Cracki> they are an abstract entity 2019-10-20T01:31:18 < Cracki> abuse is the keyword 2019-10-20T01:31:34 < catphish> owning an AS means it's actually quite easy to talk to a human at any major internet provider 2019-10-20T01:31:34 < Cracki> people want to leave the eu because they see the abuses they're subjected to by the EU 2019-10-20T01:31:40 < catphish> but i get the point 2019-10-20T01:32:40 < Cracki> you're afraid of losing your comfortable life. 2019-10-20T01:32:45 < Cracki> consider people who have no comfortable life. 2019-10-20T01:32:48 < Cracki> they have nothing to lose. 2019-10-20T01:32:52 < Cracki> they can't be pressured. 2019-10-20T01:32:58 < Cracki> they take what they want. 2019-10-20T01:33:06 < Cracki> you can't sanction them. 2019-10-20T01:33:09 < catphish> unfortunately, those people have a weakness 2019-10-20T01:33:22 < Cracki> send those to prison, they celebrate the free meal and housing 2019-10-20T01:33:26 < catphish> they can easily be made to vote against the status quo, without though 2019-10-20T01:33:29 < catphish> *hought 2019-10-20T01:33:31 < catphish> *thought 2019-10-20T01:33:34 < Cracki> fine them, they laugh because they have no money 2019-10-20T01:33:43 < catphish> they are willing to risk nothing for possible gain 2019-10-20T01:34:05 < Cracki> brexiters know what they have to lose 2019-10-20T01:34:15 < Cracki> and what they can't lose 2019-10-20T01:34:24 < Cracki> they can't lose all the commerce that's inside the UK already 2019-10-20T01:34:34 < catphish> i've avtually literaly had people laugh at me because they have nothing to lose and will vote for change because things could only get better for them 2019-10-20T01:34:42 < Cracki> they can't lose anything that's covered by other trade treaties 2019-10-20T01:34:47 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Ping timeout: 276 seconds] 2019-10-20T01:34:58 < Cracki> do you respect their voice? 2019-10-20T01:35:04 < Cracki> do you respect their suffering? 2019-10-20T01:35:29 < Cracki> or is their voice worth less because they have less to lose? 2019-10-20T01:35:53 < Cracki> is a person's worth measured by how much can be taken away from them? 2019-10-20T01:36:34 < Cracki> the more a person has that can be taken away from them, the more I pity that person for the constant fear of loss they must be living under 2019-10-20T01:37:21 < Cracki> isn't it egotistical to vote only for your own interests, rather than for the interests of your whole nation? 2019-10-20T01:38:35 < Cracki> those that vote brexit often worry about the future of their nation under EU rule 2019-10-20T01:38:57 < mawk> I have a dutch contract now catphish , I'm going to move there very soon 2019-10-20T01:39:04 < mawk> come all join me and Steffanx in dutchland 2019-10-20T01:39:15 < Cracki> I can totally see why UK citizens don't want to be ruled by something that's essentially a francogerman empire 2019-10-20T01:39:34 < mawk> in france we just say it's a german empire lol 2019-10-20T01:39:44 < mawk> everybody says the EU is tailor made for germany and brings everyone else down 2019-10-20T01:39:46 < Cracki> in germany we say it's a german empire too :P 2019-10-20T01:39:54 < mawk> lol at least our countries agree 2019-10-20T01:40:28 < Cracki> although in germany it's mostly perceived as crazy german politicians burning german money and dancing hand in hand, naked, around a satanic bonfire 2019-10-20T01:40:39 < mawk> lol 2019-10-20T01:40:46 < mawk> don't forget the luxembourgisch politicians 2019-10-20T01:40:54 < mawk> our holy man Juncker 2019-10-20T01:41:22 < Cracki> and then there's macron with his "EU army", presenting his own forces without french flag, but with eu flag, while gassing his own citizens every weekend for over forty times now 2019-10-20T01:41:32 < mawk> yeah 2019-10-20T01:41:37 < Cracki> murica should have dronestriked him forty times already 2019-10-20T01:41:45 < mawk> then he gives lessons to china in HK 2019-10-20T01:42:05 < Cracki> I'm sure his lessons are good. he's in the position he is because what he does seems to work 2019-10-20T01:42:07 < mawk> and goes on all tv news broadcasts "this whole thing is just a misunderstanding problem, my reforms are good" 2019-10-20T01:42:14 < Cracki> yeah right 2019-10-20T01:42:39 < Cracki> EU rule contributes to what pisses the NL farmers off and causes them to demonstrate 2019-10-20T01:42:50 < catphish> sorry, wandered off to bed 2019-10-20T01:42:50 < mawk> I was in Delft on the NL farmer demonstration day 2019-10-20T01:42:59 < mawk> they're more tame than in france lol 2019-10-20T01:43:01 < Cracki> they're awesome people 2019-10-20T01:43:18 < catphish> Cracki: to clarify, i do not respect their suffering, they're idiots and i hope they all die 2019-10-20T01:43:22 < mawk> in france they pour manure in front of governors office 2019-10-20T01:43:31 < mawk> suffering of who ? 2019-10-20T01:43:39 < mawk> the brexiters ? 2019-10-20T01:43:47 < Cracki> and they get lectures on protecting nature from urbanites who haven't done a day of manual labor in their whole life 2019-10-20T01:44:00 < mawk> yeah 2019-10-20T01:44:04 < Cracki> I saw that video where they sprayed manure, it was awesome 2019-10-20T01:44:08 < mawk> geert wilders was being cheered at on the tractors 2019-10-20T01:44:15 < catphish> mawk: yes mostly them :) 2019-10-20T01:44:16 < mawk> while the green party guy was booed on stage 2019-10-20T01:44:21 < Cracki> catphish, thx for the clarification, unfortunately 2019-10-20T01:44:36 < mawk> you will go to hell with such statements catphish 2019-10-20T01:44:40 < mawk> love your neighbors 2019-10-20T01:44:44 < Cracki> remember that attitude when the inevitable happens 2019-10-20T01:46:32 < Cracki> so I'm watching a video from 2014 where someone reverse engineers an apple system management controller. 2019-10-20T01:46:36 < Cracki> it's basically CM4 2019-10-20T01:46:43 < Cracki> and no real protection on it 2019-10-20T01:47:11 < Cracki> the magic keys are plain ascii strings. one of the magic spells is literally Specialis Revelio 2019-10-20T01:47:37 < Cracki> his perspective is to hide malware, and to find and purge it 2019-10-20T01:49:08 < Cracki> have some evening entertainment https://www.youtube.com/watch?v=nSqpinjjgmg 2019-10-20T01:51:25 < mawk> nice 2019-10-20T01:51:33 < mawk> ah it's ionescu 2019-10-20T01:51:37 < mawk> he did a conference at my school 2019-10-20T01:51:51 < Jan-> we need an international cookie delivery service 2019-10-20T01:51:55 < Jan-> then I could send everyone cookies :D 2019-10-20T01:51:58 < mawk> lol 2019-10-20T01:52:03 < mawk> yeah I'd rather have cookies than beer 2019-10-20T01:52:10 < mawk> I don't drink, it's bad for the brain 2019-10-20T01:52:18 < Jan-> like interflora 2019-10-20T01:52:21 < Jan-> only for cookies not flowers 2019-10-20T01:52:23 < catphish> mawk: in case it wasn't clear, i was very much joking, while i have my own feelings about my nationality, and have my own interests to protect, which both make me campaign for EU membership, i'm really very open to opposing opinions, and while i have always advocated direct democracy, this incident has made me value understanding and compromise above everything else 2019-10-20T01:52:46 < Cracki> beer's value is the socializing you do while shuddering from the bitterness 2019-10-20T01:52:49 < catphish> in a way, i guess losing is a great way to learn humility 2019-10-20T01:53:06 * Jan- does quite like beer 2019-10-20T01:53:14 * Jan- is tiny and gets really whizzed on small amounts though :/ 2019-10-20T01:53:16 < Cracki> why do you think of it as losing? 2019-10-20T01:53:21 < Cracki> you're simply a minority 2019-10-20T01:53:41 < catphish> because the result of the national poll goes against my own personal feelings of national identity 2019-10-20T01:53:46 < Cracki> democracy doesn't HAVE to mean that you get screwed over. it just means you get proportional representation 2019-10-20T01:53:57 < Cracki> and decisions factor you in according to your weight 2019-10-20T01:53:59 < mawk> for yes/no question it's not very proportionnal 2019-10-20T01:54:03 < mawk> you can't leave at 55% 2019-10-20T01:54:11 < catphish> that's the problem with yes/no polls though, not proportional 2019-10-20T01:54:13 < mawk> referendums are dangerous 2019-10-20T01:54:21 < catphish> we're not getting 52% out of the EU 2019-10-20T01:54:23 < mawk> you know what else is not proportionnal ? MP election in france ! 2019-10-20T01:54:29 < Cracki> remainers aren't alone in "losing". brexiters "lose" the same things you would. 2019-10-20T01:54:32 < mawk> they did that only to kick out national front members of parliaments 2019-10-20T01:54:39 < Cracki> so they also have a vested interest in keeping those things. 2019-10-20T01:54:40 < mawk> we have 30% of the polls, but 5% of the seats 2019-10-20T01:55:05 < mawk> because they now the front lose on each face-to-face battle for each seat, so they instituted a battle-like system instead of a proportionnal system 2019-10-20T01:55:08 < Cracki> you aren't at war with the other half of your nation. 2019-10-20T01:55:11 < mawk> know* 2019-10-20T01:55:12 < Cracki> you are just disagreeing. 2019-10-20T01:55:31 < Cracki> propaganda makes you think you're at war, and that they want to murder you 2019-10-20T01:55:43 < Cracki> (given, leftists actually do want to murder nonleftists...) 2019-10-20T01:55:59 < mawk> yeah 2019-10-20T01:56:05 < mawk> my buddy Daniel got assaulted by antifas today 2019-10-20T01:56:15 < mawk> they kicked him in the face and threw rocks at him, when he was supposed to give a talk 2019-10-20T01:56:31 < Cracki> don't tolerate their intolerance 2019-10-20T01:56:35 < Cracki> show them the door 2019-10-20T01:56:38 < catphish> isn't it illegal there to throw rocks? 2019-10-20T01:56:41 < Cracki> pull up that ghastly xkcd 2019-10-20T01:56:55 < mawk> yeah it's illegal, but antifas seems to get a free pass for violence against right wing 2019-10-20T01:56:56 < Cracki> they are without sin, they are legally obligated to throw stones, dontchaknow 2019-10-20T01:56:57 < catphish> or are rocks free expression? 2019-10-20T01:56:58 < mawk> or against anyone, really 2019-10-20T01:57:08 < Cracki> yes, rocks are free expression, and words are violence 2019-10-20T01:57:12 < catphish> lol 2019-10-20T01:57:19 < Cracki> they declare words to be violence so they can respond with free expression of rocks 2019-10-20T01:57:32 < Cracki> that's how they justify the doublethink in their minds 2019-10-20T01:57:48 < catphish> i do enjoy talking politics with people who disagree with me but tolerate my opinion, it's refreshing 2019-10-20T01:57:53 < Cracki> they're the heroes, they fight like their grandpas in normandy fought against the evil nahtsees 2019-10-20T01:57:59 < mawk> it happens very rarely to me catphish 2019-10-20T01:58:10 < mawk> I got banned from numerous channels whenever I discuss politics, on french networks 2019-10-20T01:58:22 < Cracki> I'll let you in on a secret: you are getting groomed like that english youtube cunt groomed his pug 2019-10-20T01:58:36 < catphish> unfortunately most forefront politics is now things with no aprarrent compromise 2019-10-20T01:58:37 < mawk> lol 2019-10-20T01:58:48 < Cracki> the pendulum swings 2019-10-20T01:58:58 < Cracki> only today the pendulum is split and swings both ways (*g*) 2019-10-20T01:59:04 < Cracki> and it will meet in the middle... violently 2019-10-20T01:59:15 < Cracki> if they keep pulling their half further up, the right will respond 2019-10-20T02:00:19 < catphish> i thought brexit had compromise, and i do find it odd that the government are trying to push what i see as an "extreme" agenda, despite having insufficient public and political support to do so 2019-10-20T02:00:39 < catphish> neither side are talking compromise :( 2019-10-20T02:00:55 < Cracki> they are pushing for what's left 2019-10-20T02:01:01 < Cracki> compromise was tried 2019-10-20T02:01:03 < Cracki> and blocked 2019-10-20T02:01:17 < catphish> actually, theresa may rules out compromise on day 1 2019-10-20T02:01:26 < Cracki> so what they do now is escalate, in the hopes that the roadblocks start to move 2019-10-20T02:01:27 < mawk> the mommy 2019-10-20T02:01:48 < catphish> compromise was leaving the EU with single market membership 2019-10-20T02:01:49 < mawk> from where I stand remainers MPs looks like they are sabotaging everything the PM does, catphish 2019-10-20T02:02:02 < catphish> it's literally not been out on the table :( 2019-10-20T02:02:36 < Cracki> may is a remainer 2019-10-20T02:02:41 < Cracki> she sabotaged this from the start 2019-10-20T02:02:44 < mawk> lol 2019-10-20T02:02:44 < catphish> mawk: they are, because everything the PM does is "hard brexit", an extreme position that remainers cannot agree on 2019-10-20T02:02:44 < mawk> yeah 2019-10-20T02:02:55 < mawk> may was doing hard brexit ? 2019-10-20T02:02:58 < Cracki> she didn't want brexit with single market, she wanted to scare everyone 2019-10-20T02:03:10 < catphish> yes, i'm very very confused as to why may was trying to go hard brexit despite campaigning to remain 2019-10-20T02:03:16 < catphish> that part confuses me 2019-10-20T02:03:19 < catphish> greatly 2019-10-20T02:03:24 < Cracki> may is remain. giving her the reins was the dumbest idea ever. 2019-10-20T02:04:00 < catphish> well i hoped may would fight for a soft brexit, one that pleased nobody but appeased everyone 2019-10-20T02:04:01 < Cracki> it's like making a non-coffee-drinker brew coffee. 2019-10-20T02:04:14 < Cracki> she fought for a no-brexit 2019-10-20T02:04:24 < catphish> instrad she fracured everything violently by going down an unachieveable path 2019-10-20T02:04:24 < Cracki> delaying tactics, shitting in the pot, everything 2019-10-20T02:05:02 < catphish> but boris wants brexit, and he's doing exactly the same thing 2019-10-20T02:05:07 < Cracki> her future is probably in some EU position 2019-10-20T02:05:19 < Cracki> so she's safe ONLY if the EU has a place for her 2019-10-20T02:05:31 < Cracki> and that's the case if she drives brexit against the wall 2019-10-20T02:05:43 < catphish> and even opposition parties are refusing to put forward a soft brexit 2019-10-20T02:05:55 < catphish> i don't understand why, it's a shitty compromise, but we need on 2019-10-20T02:05:57 < catphish> *one 2019-10-20T02:06:01 < Cracki> because that's the only thing they can agree on? 2019-10-20T02:06:22 < Cracki> obviously, they'll travel the same way that far, but after that they'll have to fight again 2019-10-20T02:06:51 < catphish> i see no movement towards a shared position 2019-10-20T02:07:07 < Cracki> ask every one of them. they all think the single market is overall a good thing, maybe with opportunity for improvement 2019-10-20T02:07:14 < catphish> half the country is still angry and wants to leave by any means necessary, half want to remain because they're scared 2019-10-20T02:07:29 < Cracki> your population should be angry at your politicians, not at each other 2019-10-20T02:07:33 < Cracki> they fucked it up 2019-10-20T02:07:50 < catphish> for the most part, we *are* angry a politicians 2019-10-20T02:07:52 < Cracki> they could have given you an option that doesn't get 50% plus/minus change of support 2019-10-20T02:08:02 < Cracki> why haven't you hanged any of them yet? 2019-10-20T02:08:06 < catphish> no party wants an election right now 2019-10-20T02:08:11 < Cracki> or at least sprayed them with manure 2019-10-20T02:08:16 < catphish> because they're all unpopular 2019-10-20T02:08:17 < Cracki> they want to stay in power, obviously 2019-10-20T02:08:28 < Cracki> drag them out of their office rooms 2019-10-20T02:09:08 < Cracki> look at the french. they made rioting a fine weekend tradition. 2019-10-20T02:09:30 < catphish> personally i favour a new referrendum in which people are given a choice of EU membership, single market membership, or the best alternative deal, and honouring whatever comes out of it 2019-10-20T02:09:32 < Cracki> you gotta admire their sports spirit 2019-10-20T02:09:45 < Cracki> the EU won't offer single market brefix 2019-10-20T02:09:48 < Cracki> brexit 2019-10-20T02:09:55 < Cracki> that will only happen during negotiations 2019-10-20T02:10:05 < catphish> really? i thought they'd be all over it 2019-10-20T02:10:18 < Cracki> the only way to offer that to the population is by saying "we'll try our hardest to negotiate that" 2019-10-20T02:10:25 < Cracki> the EU wants UK IN the EU 2019-10-20T02:10:32 < Cracki> and no single market is the lever they use 2019-10-20T02:10:41 < Cracki> it's an abusive relationshio 2019-10-20T02:10:51 * Jan- points at Cracki 2019-10-20T02:10:54 < Jan-> what he said. 2019-10-20T02:10:55 < catphish> but everyone knows that's BS 2019-10-20T02:11:00 < Cracki> sure it's BS 2019-10-20T02:11:06 < catphish> we all assume they want us in a single market 2019-10-20T02:11:08 < Cracki> but brit politicians need to call that bluff 2019-10-20T02:11:17 < Jan-> brit politicians don't care 2019-10-20T02:11:24 < Jan-> I have never had less confidence in the government :/ 2019-10-20T02:11:29 < Cracki> the ultimate enemy is those unelected EU politicians 2019-10-20T02:11:39 < catphish> but no uk politician has put this option of single market to the people 2019-10-20T02:11:47 < catphish> they're 100% about "go it alone" 2019-10-20T02:11:57 < catphish> "negotiate our own deals, make our own regulations" 2019-10-20T02:12:04 < catphish> etc 2019-10-20T02:12:17 < catphish> this is alienating people like me 2019-10-20T02:12:17 < Cracki> that may be because they miss the mark and think protectionism needs to extend that far 2019-10-20T02:12:27 < Cracki> they can have their protectionism WHILE having a single market 2019-10-20T02:12:37 < Cracki> EU rules are nothing else but that 2019-10-20T02:12:48 < Cracki> which sausage may carry what name, and so on 2019-10-20T02:13:01 < catphish> indeed 2019-10-20T02:13:05 < Jan-> I actually think more EU integration is a good idea 2019-10-20T02:13:07 < Jan-> but not at any cost 2019-10-20T02:13:16 < Cracki> not at any cost, exactly 2019-10-20T02:13:31 < catphish> the cost is a matter for negotiation 2019-10-20T02:13:58 < Jan-> I don't want to be part of any organisation that has done what the EU has done to Greece. 2019-10-20T02:14:00 < catphish> IMO the biggest problem is the perception that most people votes against free migration of workers 2019-10-20T02:14:02 < Jan-> And to a lesser extent Spain and Italy. 2019-10-20T02:14:15 < catphish> no migration = no single market 2019-10-20T02:14:27 < Cracki> distinguish 2019-10-20T02:14:38 < Cracki> migration of neets, migration of unskilled labor, migration of highly skilled labor 2019-10-20T02:14:58 < catphish> the eu doesn't distinguish 2019-10-20T02:15:00 < Cracki> #1 means drain on social system 2019-10-20T02:15:04 < Cracki> #2 means depressing wages 2019-10-20T02:15:04 < catphish> and that's the political problem 2019-10-20T02:15:11 < Cracki> #3 means braindrain for originating country 2019-10-20T02:15:21 < catphish> the eu won't allow membership without all 3 2019-10-20T02:15:51 < Cracki> there's a middle ground between free migration (of 1/2/3) and "we won't let you leave" 2019-10-20T02:16:10 < Jan-> the problem is immigration has been made into a boogeyman issue 2019-10-20T02:16:13 < catphish> even though i favour all 3, again because of my personal birth into EU citizenship 2019-10-20T02:16:16 < Jan-> if you're anti-immigration you're a shit 2019-10-20T02:16:18 < Cracki> one middle ground could be applying to leave, and then requirement that the destination country actually has a job for you they can't fill locally 2019-10-20T02:16:30 < Jan-> which completely overlooks the sad reality of what happens if you just snap integrate two totally different economies. 2019-10-20T02:16:37 < Jan-> It was always going to cause problems and it did. 2019-10-20T02:16:39 < Cracki> which is what murica and canada do, I think. they need to see that the company can't hire your skill locally, so then you're ok to enter. 2019-10-20T02:16:54 < catphish> i'd be annoyed if there was a vote not to let me live in parts of england beyond the part i live in right now, and i feel that same way about the EU 2019-10-20T02:16:55 < Jan-> honestly that's what america and anywhere do. 2019-10-20T02:16:55 < kakimir64> how would you access linux machine desktop from windows? 2019-10-20T02:16:57 < Jan-> and what most countries do. 2019-10-20T02:17:05 < Jan-> kakimir64: ssh? 2019-10-20T02:17:07 < catphish> kakimir64: VNC 2019-10-20T02:17:28 < Cracki> vnc ¯\_(ツ)_/¯ 2019-10-20T02:17:32 < Cracki> there are other options tho 2019-10-20T02:17:59 < Cracki> everything is based on vnc or rdp (microsoft), and various levels of graphics acceleration support, and various levels of audio transport 2019-10-20T02:18:01 < catphish> i wouldn't be using windows in the first place 2019-10-20T02:18:53 < kakimir64> I want the exact screen I have in front of me 2019-10-20T02:18:59 < kakimir64> not some separate session 2019-10-20T02:19:04 < Cracki> just the screen? vnc and upwards 2019-10-20T02:19:13 < Cracki> sure, no terminal server, but screen sharing 2019-10-20T02:19:27 < Cracki> you might even be able to run an rdp server on linux 2019-10-20T02:19:41 < Cracki> then windows needs zero third party programs to connect. rdp client is included. 2019-10-20T02:20:43 < Cracki> xrdp, freerdp, freenx, x2go, vnc variants are some I've heard and seen before 2019-10-20T02:22:16 < Cracki> this looks like the most fully featured server I can currently find http://www.freerdp.com/ 2019-10-20T02:23:41 < Ultrasauce> https://houseoftheleg.bandcamp.com/album/area music 2019-10-20T02:23:46 < Cracki> rhel has "screen sharing" configurable form gui https://opensource.com/article/18/6/linux-remote-desktop 2019-10-20T02:24:10 < Cracki> (screen sharing vs "terminal server" type deal) 2019-10-20T02:24:43 < catphish> brexit: https://www.youtube.com/watch?v=pvEdaoi_XLk 2019-10-20T02:25:17 < Cracki> yes, the world isn't gonna end 2019-10-20T02:25:33 < catphish> indeed 2019-10-20T02:26:06 < catphish> personally i'll be fine either way 2019-10-20T02:27:16 < catphish> aside from potential mild inconvenience of reduced lebensraum 2019-10-20T02:28:13 < kakimir64> let's reboot this shiet 2019-10-20T02:28:18 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-20T02:28:23 < catphish> lol bye 2019-10-20T02:31:18 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-20T02:31:28 < catphish> i'd like to draw attention to the fact that i ran a 5k in 21:30 today, that's quite fast 2019-10-20T02:31:36 < Jan-> this may be a stupid question but it seems like microcontrollers in general (avr, stm32 etc) all have really small amounts of RAM. 2019-10-20T02:31:42 < Jan-> You know, compared to modern technology. 2019-10-20T02:31:48 < Jan-> Is there some reason it's hard to add lots of RAM to them? 2019-10-20T02:31:53 < Jan-> in terms of the manufacturing? 2019-10-20T02:32:02 < catphish> Jan-: it's the difference between SRAM and DRAM 2019-10-20T02:32:13 < Jan-> oh is it all static ram 2019-10-20T02:32:17 < Cracki> yes. 2019-10-20T02:32:18 < catphish> yes 2019-10-20T02:32:18 < Cracki> also cost. 2019-10-20T02:32:23 < Jan-> so it's just a crapton of flip flops. 2019-10-20T02:32:31 < Cracki> AND it's not that complicated to add SDRAM do it. 2019-10-20T02:32:33 < Cracki> (offchip) 2019-10-20T02:32:35 < catphish> totally different technology, DRAM is somewhat more complicated to integrate, and a pain to use 2019-10-20T02:32:41 < kakimir32> I can only see green screen 2019-10-20T02:32:47 < Jan-> can you not make dram using the same process as the rest of it, or something? 2019-10-20T02:32:47 < Cracki> leet 2019-10-20T02:32:55 < Cracki> take the raspi. 2019-10-20T02:32:56 < kakimir32> or is it cyan 2019-10-20T02:33:01 < catphish> and even though DRAM is cheap, remember that a STM32 only costs like £2 2019-10-20T02:33:02 < Cracki> it has ram bonded on top of the core 2019-10-20T02:33:23 < Jan-> well I think an f407 is a bit more 2019-10-20T02:33:46 < catphish> DRAM is a very different process afaik 2019-10-20T02:33:51 < catphish> though i don't know the details 2019-10-20T02:34:09 < catphish> it has capacitive elements rather than just gates 2019-10-20T02:34:18 < Jan-> I haven't looked but is there any straightforward way to put DRAM on an stm32 2019-10-20T02:34:27 < Jan-> or is there some other chip you'd go for 2019-10-20T02:34:53 < catphish> it's quite easy to add serial or parallel DRAM 2019-10-20T02:34:54 < Cracki> dunno how straight forward this is: https://www.st.com/content/ccc/resource/training/technical/product_training/group0/51/a3/68/fd/47/6d/43/b8/STM32F7_Memory_FSMC/files/STM32F7_Memory_FSMC.pdf/jcr:content/translations/en.STM32F7_Memory_FSMC.pdf 2019-10-20T02:35:02 < Cracki> f7 Flexible memory controller 2019-10-20T02:35:12 < Jan-> not on the f407 then :) 2019-10-20T02:35:18 < Cracki> f4 might have that too 2019-10-20T02:35:40 < Jan-> I guess what threw me is that stm32 is basically the same price as avr 2019-10-20T02:35:44 < Jan-> and yet lots of times faster 2019-10-20T02:35:49 < Jan-> which makes avr look stupidly expensive 2019-10-20T02:36:01 < Cracki> xmegas can use external ram, I think some atmega too already 2019-10-20T02:36:24 < Cracki> avr's worth is because it's simple and people seem comfy sitting there 2019-10-20T02:36:27 < catphish> Jan-: check out https://www.microchip.com/wwwproducts/en/23A1024 2019-10-20T02:36:51 < Jan-> the nice thing about avr is that you can just fire up the arduino editor, type main() and go from there 2019-10-20T02:36:53 < catphish> and similar serial RAM 2019-10-20T02:37:15 < Jan-> whereas the tools for stm32 are... um... 2019-10-20T02:37:18 < Jan-> ...hideous. 2019-10-20T02:37:25 < Cracki> you can have the same with those stm32 bluepills, once they have the right bootloader on them, or you've connected with an stlink (arduino can now use stlink) 2019-10-20T02:37:49 < Cracki> ST's support for arduino is getting better. they have specific target boards, and generic F0-F4 support 2019-10-20T02:38:26 < Jan-> the problem with all this cli stuff is that it is not discoverable 2019-10-20T02:38:32 < Jan-> you can't drop down a menu and look for likely options 2019-10-20T02:38:38 < Jan-> you just have to know everything up front. 2019-10-20T02:38:40 < Cracki> stm32f412 says that it has a "external static memory controller" 2019-10-20T02:38:52 < Cracki> so it'd be sram not dram, but eh you can buy plenty 2019-10-20T02:38:59 < catphish> Jan-: did you play with stm32cubeide? 2019-10-20T02:39:02 < Cracki> with arduino you can 2019-10-20T02:39:06 < Cracki> let me make you a screenshot 2019-10-20T02:39:21 < catphish> it's not arduino, but it's an easy / powerful starting point 2019-10-20T02:39:37 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-20T02:39:41 < Jan-> I did but the code it generates is just so horrible. 2019-10-20T02:39:43 < catphish> i still like my low leven shizzle though 2019-10-20T02:39:49 < catphish> *level 2019-10-20T02:40:09 < Jan-> as far as I know I can just use the f407 register definitions and go from there 2019-10-20T02:40:12 < Jan-> just like you would with an AVR 2019-10-20T02:40:14 < catphish> yep 2019-10-20T02:40:19 < Jan-> I would totally do that 2019-10-20T02:40:20 < catphish> if you like that, do it 2019-10-20T02:40:26 < Jan-> sure but I have no idea how. 2019-10-20T02:40:38 < Jan-> every IDE I have found has its own set of libraries 2019-10-20T02:40:39 < Cracki> here's the stm32duino stuff https://imgur.com/np8BKPk 2019-10-20T02:40:40 < catphish> well, cubeide probably lets you do that 2019-10-20T02:40:50 < kakimir32> it's a fucking mess when you try to do things with linux that are trivial in windows 2019-10-20T02:40:53 < Cracki> cubeide then 2019-10-20T02:40:54 < catphish> you don't have to use the included librries 2019-10-20T02:41:07 < Jan-> I guess you don't have to in platformio or mbed either 2019-10-20T02:41:13 < Cracki> st's LL is basically poking registers 2019-10-20T02:41:23 < Jan-> but if I'm going down there I might as well just use make and gcc 2019-10-20T02:41:34 < Jan-> it's like my main.cpp and the f407 defines 2019-10-20T02:41:34 < Cracki> same as in arduino, you can drop down below any library you use 2019-10-20T02:42:02 < catphish> you can always do it my way with ST's header files and a simple bash script to invoke GCC 2019-10-20T02:42:17 < catphish> nothing more than that is needed if you're hardcore 2019-10-20T02:42:35 < Jan-> well 2019-10-20T02:42:41 < Jan-> I think you'd want the basic register definitions 2019-10-20T02:42:44 < Jan-> for the sake of your sanity 2019-10-20T02:42:56 < catphish> "ST's header files" 2019-10-20T02:43:07 < Jan-> oh sorry I heard "do away with" :) 2019-10-20T02:43:07 < catphish> they include pointers to every register 2019-10-20T02:43:40 < catphish> they're literally just .h files with constants though, nothing that bloats your code 2019-10-20T02:44:19 < Jan-> would that mean dereferencing them every time you used them 2019-10-20T02:44:30 < catphish> no 2019-10-20T02:44:31 < Jan-> obviously that's just syntax but you know 2019-10-20T02:44:44 < catphish> you include a .h file 2019-10-20T02:44:57 < catphish> and the compiler will bring in whatever constants you use 2019-10-20T02:45:09 < mawk> but add nothing to the final executable, so it's not bloat 2019-10-20T02:45:34 < Jan-> I'm not saying it's bloat, just thinking how it works 2019-10-20T02:45:55 < Jan-> wouldn't you be like, *ODR_A |= ( 1 << n); 2019-10-20T02:46:12 < catphish> yes 2019-10-20T02:46:16 < catphish> that's how you'd use it 2019-10-20T02:46:50 < catphish> and the compiler would use the include to replace ODR_A with its appropriate value from the .h file at compile time 2019-10-20T02:46:53 < Jan-> so in the header it's something like volatile uint32_t* ODR_A = some number; 2019-10-20T02:47:24 < catphish> more like #define ODR_A some number; 2019-10-20T02:47:49 < catphish> it doesn't generate any code at all, it just defines a constant that gets replace in our final code 2019-10-20T02:47:51 < Jan-> well sure but then your C code just turns into 12345678 = ( 1 << n ); 2019-10-20T02:47:52 < Jan-> which is BS 2019-10-20T02:48:05 < catphish> not quite 2019-10-20T02:48:18 < catphish> *(12345678) = (1< uhm 2019-10-20T02:48:32 < Cracki> *(uint32_t*)0x12345678 = somevalue; 2019-10-20T02:48:33 < Cracki> yeah 2019-10-20T02:48:41 < Cracki> it behaves like a variable 2019-10-20T02:48:42 < Jan-> so you will need the asterisk. 2019-10-20T02:48:44 < Cracki> no 2019-10-20T02:48:47 < Cracki> that's the whole macro 2019-10-20T02:48:57 < Cracki> #define FOO *(uint32_t*)(someaddr) 2019-10-20T02:48:59 < catphish> the definition includes it 2019-10-20T02:49:03 < Jan-> oh right. 2019-10-20T02:49:05 < Cracki> FOO = newvalue; 2019-10-20T02:49:07 < catphish> ^ what Cracki saud 2019-10-20T02:49:09 < catphish> *said 2019-10-20T02:49:17 < Jan-> the reason I'm asking is that the arduino things, like PORTB, are complicated macros that actually run code 2019-10-20T02:49:19 < Jan-> and I never understood why 2019-10-20T02:49:25 < Cracki> uh no 2019-10-20T02:49:32 < Cracki> those macros often are just that 2019-10-20T02:49:42 < Jan-> I looked some of them up, some of them do run code 2019-10-20T02:49:44 < Jan-> which is weird 2019-10-20T02:49:47 < Cracki> you might have seen SFR8 or whatever 2019-10-20T02:49:51 < Cracki> that's taking care of that cast 2019-10-20T02:49:55 < catphish> well the macros vary in complexity, but generally they're pointers to structs 2019-10-20T02:50:04 < catphish> where the structs are also defines in the .h 2019-10-20T02:50:06 < Cracki> some may include code, maybe to wait for a flag or something 2019-10-20T02:50:09 < catphish> *defined 2019-10-20T02:50:20 < catphish> i've not seen any that include code 2019-10-20T02:50:21 < Jan-> also presumably by doing this I am locking it to exactly one model of chip. 2019-10-20T02:50:26 < catphish> only structs and ponters 2019-10-20T02:50:34 < Cracki> yeah well lowlevel isn't all that portable 2019-10-20T02:50:43 < Jan-> apparently ODR_B is at 0x40020414 and it will only be at that location in an STM32F407VET6. 2019-10-20T02:50:58 < Jan-> And nothing else. 2019-10-20T02:51:13 < Cracki> you can expect a close enough sister chip to have all the same names, and maybe different locations 2019-10-20T02:51:21 < catphish> Jan-: each chip has a .h file, you can often switch from one to another by just selecting the new .h file, but no guarentee the new chip will have the same register the old one did or that it'll behave the same 2019-10-20T02:51:24 < Cracki> if it has different versions of a peripheral, ¯\_(ツ)_/¯ 2019-10-20T02:52:02 < Jan-> well sure it's like going from an attiny to an atmega or whatever 2019-10-20T02:52:08 < catphish> this is the point of working at register level, it's nice and it's fast, but it's not portable 2019-10-20T02:52:14 < Cracki> atmegas are the same. one may have a single uart, one may have two numbered ones. one may have pin chance interrupts, the other doesn't 2019-10-20T02:52:25 < Cracki> (PCINT is like exti) 2019-10-20T02:52:31 < Jan-> Cracki: don't confuse your TCCR0A with TCCRA :) 2019-10-20T02:52:39 < Cracki> exactly that 2019-10-20T02:52:47 < Jan-> yeah no I get that. 2019-10-20T02:53:01 < Jan-> If you write sane code then all that low level stuff should be stacked neatly on a shelf anyway so you can tweak as necessary. 2019-10-20T02:54:21 < Jan-> But I still don't really have a way of doing this. 2019-10-20T02:54:28 < Jan-> the barrier is makefiles. 2019-10-20T02:55:33 < catphish> this is how i do it https://github.com/catphish/amiga-floppy/blob/master/make.sh 2019-10-20T02:55:44 < Jan-> amiga floppy? :) 2019-10-20T02:55:55 * Jan- was an amiga person a very long time ago 2019-10-20T02:56:04 < catphish> my current project is a usb floppy controller 2019-10-20T02:56:31 < catphish> primarily to read/write amiga disks, but in theory capable of handing any floppy disk 2019-10-20T02:56:54 < Jan-> if you can figure out how to get data off syquest ez-135 carts that were formatted in an amiga, do tell :/ 2019-10-20T02:57:02 < Jan-> we have the drive and the scsi card and everything 2019-10-20T02:57:13 < Jan-> but of course the carts are amiga formatted. 2019-10-20T02:57:29 < catphish> i used to have one of those drives as a HD 2019-10-20T02:57:49 < catphish> (in a PC) 2019-10-20T02:58:40 < Jan-> we do have an amiga 4000 somewhere 2019-10-20T02:58:50 < Jan-> probably have to use that but it hasn't been booted in 20 years :) 2019-10-20T02:58:57 < Jan-> and even then no ethernet 2019-10-20T02:59:07 < Jan-> what're we gonna do, parallel port? :D 2019-10-20T03:00:37 * Jan- shamelessly steals catphish's makething 2019-10-20T03:00:53 < Jan-> not that it'll work for me 2019-10-20T03:01:21 < catphish> you'll need the right header files (easily downloaded fro ST) 2019-10-20T03:01:36 < catphish> and you'll need the arm compiler (i use gcc from arm) 2019-10-20T03:01:51 < Jan-> I have arm-none-eabi-gcc and all that 2019-10-20T03:02:36 < Jan-> on windows I'd just write this in javascript or something 2019-10-20T03:03:03 < Jan-> but I'm sure a lot would have to change. 2019-10-20T03:03:38 < Cracki> oh uh, no concrete reason but does anyone here know of a place that can do xrays or even tomography in better than 20 micrometer resolution XY, better than 10 in Z? 2019-10-20T03:04:12 < Cracki> I would have the skills to analyze the volume data for some automated PCB RE but it needs good data to be easily automatable 2019-10-20T03:05:15 < Cracki> it's really not different from segmenting arteries in a tomograph 2019-10-20T03:07:26 < catphish> Jan-: in theory you just need arm-none-eabi-gcc and the STM32Cube_FW_xxxxx (according to what kind of chip you use) 2019-10-20T03:08:11 < catphish> the latter is quite a large library, but you only need to include a single file from it 2019-10-20T03:08:33 < Jan-> why do I need bits of cube 2019-10-20T03:08:55 < catphish> these are the header files 2019-10-20T03:09:20 < kakimir32> I did it 2019-10-20T03:09:35 < kakimir32> it works ggood 2019-10-20T03:09:52 < catphish> Jan-: the download contains lots of stuff, CMSIS etc, but you don't need to reference any of it 2019-10-20T03:10:02 < Jan-> I forget what the license is on cmsis 2019-10-20T03:10:33 < catphish> well you best check 2019-10-20T03:10:34 < Jan-> or what it actually gives me 2019-10-20T03:10:57 < catphish> it's apache 2.0 2019-10-20T03:11:20 < Jan-> I'm not totally clear on the difference between "the ST headers" and "cmsis" 2019-10-20T03:11:45 < catphish> yeah this is a tad blurred 2019-10-20T03:12:11 < catphish> cmsis is an ARM library, ST have added header files for their devices inside it 2019-10-20T03:12:27 < catphish> my make script references those header files direcrly 2019-10-20T03:12:56 < Jan-> I guess I'd assumed it would just be stm32f407xx.h 2019-10-20T03:12:59 < catphish> they're part of CMSIS, but kinda separate 2019-10-20T03:13:04 < catphish> yes, it is 2019-10-20T03:13:13 < Jan-> that's "the st header" right? 2019-10-20T03:13:16 < catphish> yes 2019-10-20T03:13:21 < catphish> and it's found inside CMSIS 2019-10-20T03:13:24 < Jan-> ok but inside it it says "@brief CMSIS STM32F407xx Device Peripheral Access Layer Header File." 2019-10-20T03:13:41 < catphish> the header file is designed to be used by CMSIS 2019-10-20T03:13:48 < Jan-> ok 2019-10-20T03:13:51 < catphish> but you can use it yourself directly 2019-10-20T03:14:09 < Jan-> it's fricken huge it's like eight thousand lines 2019-10-20T03:14:10 < catphish> i assume i'm not the only person that does this 2019-10-20T03:14:31 < catphish> that file also includes other more generic ST inclue files 2019-10-20T03:14:33 < catphish> it's huge 2019-10-20T03:14:38 < catphish> but... 2019-10-20T03:14:52 < Cracki> stm32 has lots of registers ¯\_(ツ)_/¯ 2019-10-20T03:14:57 < Cracki> *peripheral 2019-10-20T03:15:06 < Jan-> bit of an issue in that you then also need core_cm4.h and system_stm32f4xx.h 2019-10-20T03:15:07 < catphish> including it doesn't really add any code to your project, or to your compiled binary 2019-10-20T03:15:10 < Cracki> atmega headers are almost as large 2019-10-20T03:15:10 < Jan-> and whaever they include 2019-10-20T03:15:13 < Jan-> and whatever THEY include... 2019-10-20T03:15:27 < catphish> (except the specific parts you use) 2019-10-20T03:15:34 < Cracki> cm4 is a finite amount of dependency 2019-10-20T03:15:44 < Jan-> I get that catphish I'm just thinking what I need to obtain/. 2019-10-20T03:15:50 < catphish> if you wanted to be anal about it, you could copy the specific lines out into your orn .h file 2019-10-20T03:15:51 < Jan-> And somehow build. 2019-10-20T03:16:01 < catphish> *own 2019-10-20T03:16:48 < Jan-> Is there some reason why if I just #include everything, a single gcc call won't actually just build everything. 2019-10-20T03:17:08 < Cracki> a single gcc call would be enough 2019-10-20T03:17:12 < catphish> Jan-: it will 2019-10-20T03:17:13 < Cracki> if everything is actually a header 2019-10-20T03:17:23 < Jan-> well you can include any file you like... 2019-10-20T03:17:38 < catphish> Jan-: rather, it will read the header and compile those parts that your code uses 2019-10-20T03:17:47 < catphish> Jan-: read my make.sh, it's all there 2019-10-20T03:17:55 < Jan-> I have, I only partly understand it 2019-10-20T03:18:04 < catphish> no build steps are needed that are not in that file 2019-10-20T03:18:20 < catphish> it just passes the header directory to gcc as an include directory 2019-10-20T03:18:28 < catphish> that's it i think 2019-10-20T03:18:34 < Jan-> why isn't there a space between -I and $REPROROOT 2019-10-20T03:19:09 < catphish> that's just how -I works 2019-10-20T03:19:17 < catphish> it doesn't have spaces, it's weird 2019-10-20T03:19:22 < Jan-> oh I love open source software. "Why X?" "Because."£ 2019-10-20T03:19:24 < catphish> like -Ilib 2019-10-20T03:19:31 < catphish> or -lmath 2019-10-20T03:19:36 < Jan-> and what's in "/home/charlie/Applications/STM32Cube_FW_L4_V1.14.0" 2019-10-20T03:19:57 < catphish> Jan-: what's in there is something you need to downlaod from ST 2019-10-20T03:20:12 < catphish> but not l4, the right one for your chip 2019-10-20T03:21:02 < catphish> for exmple https://www.st.com/en/embedded-software/stm32cubel4.html 2019-10-20T03:21:11 < Jan-> yeah I just googled that up 2019-10-20T03:21:19 < Jan-> but that seems to be the whole HAL and everything 2019-10-20T03:21:25 < catphish> yes 2019-10-20T03:21:29 < catphish> we've been over this 2019-10-20T03:21:44 < catphish> the download contains the whole HAL, lots of drivers, everything 2019-10-20T03:21:57 < catphish> but you will only point GCC at the one header file 2019-10-20T03:22:11 < catphish> you can delete the rest if it makes you happy 2019-10-20T03:22:22 < Jan-> what. the stm32f407xx.h 2019-10-20T03:22:42 < catphish> yes (and whatever that file itself includes) 2019-10-20T03:22:45 < Jan-> I can get that from loads of places, I've probably already got that 2019-10-20T03:22:49 < catphish> ok then 2019-10-20T03:23:00 < catphish> if you have it, you don't need to download it again 2019-10-20T03:23:13 < catphish> just set that -I to the directory it's in 2019-10-20T03:23:25 < Jan-> it's part of mbed 2019-10-20T03:23:30 < Jan-> I guess it'd have to be 2019-10-20T03:24:07 < catphish> it's very likely that all frameworks will use those same header files 2019-10-20T03:24:21 < catphish> they have all the register addresses, no need to redo that work 2019-10-20T03:24:58 < Jan-> pat just needs to have the gcc arm-none-eabi tools, I guess 2019-10-20T03:25:00 < Jan-> path 2019-10-20T03:25:26 < catphish> yep 2019-10-20T03:25:35 < Jan-> and st-flash in your case 2019-10-20T03:25:41 < Jan-> what about startup_stm32l433xx.s 2019-10-20T03:25:58 < catphish> you'll want to find a copy of that file and take it 2019-10-20T03:26:03 < Jan-> oh that's in a subfolder near the .h 2019-10-20T03:26:13 < Jan-> oh it's the vector table 2019-10-20T03:26:18 < catphish> yeah mostly that 2019-10-20T03:26:35 < Jan-> is that so all the irq handlers work and such 2019-10-20T03:26:42 < catphish> yep 2019-10-20T03:27:06 < Jan-> some reason system.c is before main.c? 2019-10-20T03:27:24 < catphish> this lets you just define functions to handle IRQs like void TIM2_IRQHandler() {} 2019-10-20T03:27:39 < Jan-> yes I get that. 2019-10-20T03:27:44 < catphish> that's totally optional 2019-10-20T03:27:59 < Jan-> so system.c isn't doing any weird stm32 setup stuff 2019-10-20T03:28:01 < Cracki> order is arbitrary 2019-10-20T03:28:05 < catphish> kinda like arduino's startup and main functions 2019-10-20T03:28:06 < Cracki> content matters 2019-10-20T03:28:15 < Jan-> setup and loop. 2019-10-20T03:28:18 < Jan-> but I don't even use those. 2019-10-20T03:28:21 < catphish> that's it 2019-10-20T03:28:24 < catphish> it's nice to have them in separate files 2019-10-20T03:28:26 < Jan-> you don't want to run their main 2019-10-20T03:28:30 < catphish> but totally up to you 2019-10-20T03:28:38 < Jan-> so what's in system 2019-10-20T03:28:45 < Jan-> all the clock setup and suchlike? 2019-10-20T03:29:07 < catphish> i have a void SystemInit() that sets up the system clock 2019-10-20T03:29:12 < catphish> https://github.com/catphish/amiga-floppy/blob/master/system.c 2019-10-20T03:29:14 < Jan-> oh yeah I need to work that out :/ 2019-10-20T03:29:32 < Jan-> that's gonna be something I steal from elsewhere I bet 2019-10-20T03:29:50 < catphish> i'm 90% sure my code there is copied form somewhere 2019-10-20T03:30:11 < catphish> it's just standard code that sets up the high speed systen clock using a PLL 2019-10-20T03:30:13 < Jan-> well there must be files I have that do that, as I've been flashing code to this thing and it's been working. 2019-10-20T03:30:30 < catphish> you don't *have* to change the defaults 2019-10-20T03:30:48 < catphish> in fact if you flash nothing but main() it will work 2019-10-20T03:30:49 < Jan-> I was told you had to set all that up super carefully or it wouldn't even come out of reset 2019-10-20T03:31:00 < catphish> nah 2019-10-20T03:31:04 < catphish> defaults will work 2019-10-20T03:31:12 < catphish> but it will be slow 2019-10-20T03:31:25 < catphish> like atmega, it'll run at like 1MHz by default 2019-10-20T03:31:43 < catphish> until you set up a faster clock and switch to it 2019-10-20T03:32:20 < Jan-> where does that even get called 2019-10-20T03:32:28 < Jan-> or is "SystemInit" a special name that has some meaning 2019-10-20T03:32:32 < catphish> from startup.s 2019-10-20T03:32:45 < catphish> https://github.com/catphish/amiga-floppy/blob/master/startup_stm32l433xx.s#L93 2019-10-20T03:32:55 < Jan-> yeah I got it 2019-10-20T03:33:07 < Jan-> github is hateful but I am winning slowly 2019-10-20T03:33:21 < Jan-> so that has to be called SystemInit 2019-10-20T03:33:34 < catphish> pretty much everything there is optional, you can run your own code right from the startup vector if you want 2019-10-20T03:33:46 < catphish> you can rename it 2019-10-20T03:33:52 < Jan-> and er finally. -T STM32L433RCTx_FLASH.ld? 2019-10-20T03:33:59 < catphish> or just delete that line from startup.s and put the code in your main() 2019-10-20T03:34:13 < catphish> that file defines the memory layout 2019-10-20T03:34:28 < Jan-> I thought the .s did that... 2019-10-20T03:34:32 < catphish> again, copy/paste it 2019-10-20T03:34:34 < Jan-> again I just get that from st? 2019-10-20T03:34:46 < catphish> no, the .s is code and vectors 2019-10-20T03:35:11 < catphish> .ld defines the memory addresses for RAM and flash 2019-10-20T03:35:33 < catphish> and where stack goes, things i don't really understand 2019-10-20T03:35:44 < catphish> i'm not sure where that file came from 2019-10-20T03:35:47 < catphish> but i've not touched it 2019-10-20T03:35:52 < Jan-> why st-flash not openocd 2019-10-20T03:35:57 < Jan-> I've had trouble with openocd 2019-10-20T03:36:04 < catphish> ^ 2019-10-20T03:36:15 < catphish> st-flash "just works" for me 2019-10-20T03:36:21 < Jan-> again it's like every open source tool has to have its own private scripting language that is awful. 2019-10-20T03:36:26 < Jan-> make does, openocd does 2019-10-20T03:36:44 < catphish> i like st-flash because it's so simple 2019-10-20T03:36:56 < Jan-> what's the 0x8000000 2019-10-20T03:36:59 < catphish> no scripting needed, no options, just a simple command 2019-10-20T03:37:15 < Jan-> Yeah. it's like, jesus pole-vaulting christ, can openocd be more abstruse 2019-10-20T03:37:20 < catphish> that's the location to write the binary to, flash memory is at 0x8000000 2019-10-20T03:37:46 < Jan-> is it on the 407 I wonder 2019-10-20T03:37:47 < Jan-> I can check 2019-10-20T03:38:06 < catphish> it probably is, but do check 2019-10-20T03:38:12 < Jan-> yeah 2019-10-20T03:38:15 < Jan-> because that would be Bad. 2019-10-20T03:38:26 < catphish> you can't break it :) 2019-10-20T03:38:44 < catphish> the built in bootloader is read-only 2019-10-20T03:38:47 < Jan-> yeah but I can flash it full of partial gibberish that doesn't work. 2019-10-20T03:38:53 < catphish> well yes 2019-10-20T03:39:07 < Jan-> holy shit is that the time, I have to sleep 2019-10-20T03:39:14 < catphish> yes 2019-10-20T03:39:14 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-20T03:39:29 < catphish> and i need to do something productive that isn't helping you :) 2019-10-20T03:39:31 < Jan-> I am teaching a blind person how to not walk into objects tomorrow and I need to be there early to make her feel bad about not being awake. 2019-10-20T03:39:39 < Jan-> *cough* 2019-10-20T03:39:42 < catphish> lol 2019-10-20T03:39:51 < catphish> i say blind people are allowed to sleep in 2019-10-20T03:39:57 < Jan-> I'm not :/ 2019-10-20T03:40:31 < Jan-> but you can keep talking :D 2019-10-20T03:40:40 < Jan-> anyway cocoa and bed 2019-10-20T03:40:42 < catphish> i think i'll let you sleep :) 2019-10-20T03:40:46 * Jan- hands catphish a large mug of cocoa 2019-10-20T03:40:50 < catphish> yay 2019-10-20T03:41:01 < Jan-> and a white chocolate chip and macadamia nut cookie. 2019-10-20T03:41:29 < catphish> sounds pleasant :) 2019-10-20T03:41:57 < Jan-> thanks 2019-10-20T04:05:43 < kakimir32> https://github.com/devolo/dlan-greenphy-sdk/blob/master/Libraries/FreeRTOSv9.0.0/include/freertos_tasks_c_additions.h why I cannot find symbol FreeRTOSDebugConfig from map file? 2019-10-20T04:05:48 < Cracki> what are you guys doing, dog stuff? https://youtu.be/Av1XuYbL06Y?t=5570 2019-10-20T04:05:57 < kakimir32> this file gets included in task.c 2019-10-20T04:06:04 < Cracki> stripped maybe 2019-10-20T04:06:22 < Cracki> wait, it says extern there, do you build the C file too? 2019-10-20T04:06:32 < kakimir32> there is no C file 2019-10-20T04:06:33 < Cracki> nvm, it defines that right after 2019-10-20T04:06:41 < Cracki> maybe you have none of the defines 2019-10-20T04:06:50 < kakimir32> I have 2019-10-20T04:06:53 < Cracki> my money's on stripped 2019-10-20T04:06:56 < kakimir32> __GCC__ 2019-10-20T04:07:06 < Cracki> is that __GNUC__ 2019-10-20T04:07:10 < kakimir32> yes 2019-10-20T04:07:36 < kakimir32> I meant that 2019-10-20T04:09:40 < Cracki> is anything actually using that FreeRTOSDebugConfig? 2019-10-20T04:10:03 < Cracki> sounds like that might only be used in a debug context... so I'd ask if that's set or not 2019-10-20T04:11:22 < kakimir32> it's for de debugger 2019-10-20T04:11:31 < kakimir32> and freertos tad 2019-10-20T04:14:02 < kakimir32> https://mcuoneclipse.com/2017/07/27/troubleshooting-tips-for-freertos-thread-aware-debugging-in-eclipse/ 2019-10-20T04:14:47 < Cracki> I may need that... haven't checked how well ozone handles freertos 2019-10-20T04:20:42 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 252 seconds] 2019-10-20T04:21:10 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-20T04:23:05 < kakimir32> I have freertos TAD in mcueclipse 2019-10-20T04:23:13 < kakimir32> almost useless monitors 2019-10-20T04:23:25 < kakimir32> stack usages are okay 2019-10-20T04:23:38 < kakimir32> to see 2019-10-20T04:33:44 < kakimir32> I got the array visible to debugger 2019-10-20T04:33:59 < kakimir32> but I don't see any change in way data is visualized 2019-10-20T04:34:05 < kakimir32> no improvements 2019-10-20T04:35:29 < antto> unprovements 2019-10-20T04:38:56 < Cracki> maybe open the right panel or something 2019-10-20T04:39:46 < Cracki> does it let you block/unblock tasks or support debugging a single task while letting the rest run or anything else useful? 2019-10-20T04:40:30 < Cracki> one screenshot implies it's possible to debug a single task... 2019-10-20T04:40:43 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-20T04:41:26 < Cracki> note this BS https://www.freertos.org/FreeRTOS_Support_Forum_Archive/August_2018/freertos_v10.1.0_breaks_thread_aware_debugging_with_JLink_d406aae6j.html 2019-10-20T04:41:42 < Cracki> I thought that was when they went from 9 to 10 only, not in a point release 2019-10-20T04:52:43 -!- jadew [~rcc@82.76.177.192] has joined ##stm32 2019-10-20T04:52:43 -!- jadew [~rcc@82.76.177.192] has quit [Changing host] 2019-10-20T04:52:43 -!- jadew [~rcc@unaffiliated/jadew] has joined ##stm32 2019-10-20T04:58:05 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-20T05:08:32 < jadew> morning 2019-10-20T05:17:33 < Cracki> you're early 2019-10-20T05:18:08 < Cracki> sun won't rise for another two hours 2019-10-20T05:18:28 < jadew> and I went to sleep at 1 am 2019-10-20T05:18:31 < jadew> I'm getting old 2019-10-20T05:18:55 < Cracki> brain's plasticity wanes, needs less sleep to defragment 2019-10-20T05:19:05 < Cracki> can you hear your ear bones creak yet? 2019-10-20T05:19:16 < jadew> well, I slept 15 hours yesterday and the day before 2019-10-20T05:19:22 < jadew> so that might have something to do with it 2019-10-20T05:19:32 < jadew> I think it's some sort of personal record 2019-10-20T05:19:34 < Cracki> my distal thumb joints, of all things, crack sometimes 2019-10-20T05:19:57 < jadew> can't say I observed any new cracking noises 2019-10-20T05:20:12 < Cracki> pretty https://www.timeanddate.com/sun/romania/bucharest 2019-10-20T05:21:14 < jadew> that's nice 2019-10-20T05:24:08 < Cracki> only 4 months of days this length or shorter 2019-10-20T05:25:28 < jadew> I need money 2019-10-20T05:25:49 < Cracki> same 2019-10-20T05:26:00 < jadew> I'm short a couple of hundreds in order to place an order for some enclosures 2019-10-20T05:26:21 < jadew> Cracki, don't you have a cozy job? 2019-10-20T05:26:40 < Cracki> :> 2019-10-20T05:27:02 < Cracki> I was about to check my balance but they have maintenance at night 2019-10-20T05:27:19 < Cracki> it should be in the grayish green tho 2019-10-20T05:27:32 < Cracki> why, you want me to find the usurer in me? 2019-10-20T05:27:45 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T05:28:21 < Cracki> jobs cozy but that's more to do with the little time I need to spend on it and the general laissez faire 2019-10-20T05:28:23 < jadew> heh, no, thanks :) 2019-10-20T05:28:28 -!- con3 [~kvirc@154.119.40.228] has quit [Client Quit] 2019-10-20T05:28:34 < Cracki> good, I'm bad at usury 2019-10-20T05:28:46 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T05:28:51 < jadew> when I say "I" I mean the company 2019-10-20T05:29:05 < jadew> I could lend it some money myself, but I'm trying to avoid that 2019-10-20T05:29:05 < Cracki> my grandpa was also very very bad at usury. he has a certificate 2019-10-20T05:29:26 -!- con3 [~kvirc@154.119.40.228] has quit [Client Quit] 2019-10-20T05:29:29 < jadew> what kind of certificate is that? 2019-10-20T05:29:33 < Cracki> red tape when you lend your company money? 2019-10-20T05:29:36 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T05:29:57 < Cracki> the kind they gave to 18 year olds to make them cannon fodder 2019-10-20T05:30:04 < jadew> Cracki, well, I already did it a couple of times and I didn't get it all back yet 2019-10-20T05:30:18 < Cracki> uh 2019-10-20T05:30:29 < Cracki> is the business at least breaking even? 2019-10-20T05:30:33 < jadew> no lol 2019-10-20T05:30:37 < Cracki> >_> 2019-10-20T05:30:45 < Cracki> you need to bill your customers more 2019-10-20T05:30:57 < jadew> I need more customers 2019-10-20T05:31:03 < jadew> and that too 2019-10-20T05:31:11 < jadew> the consulting I'm doing, I'm doing for a very low rate 2019-10-20T05:31:18 < Cracki> price hike 2019-10-20T05:31:40 < Cracki> add a few percent, christmas anticipation tax 2019-10-20T05:32:00 < Cracki> you need to afford new socks for your thirteen poor children 2019-10-20T05:32:11 < jadew> I'll lead with that 2019-10-20T05:32:28 < Cracki> I'm gonna have to adjust my prices to match inflation too soon 2019-10-20T05:32:50 < Cracki> they always engage me two times each semester, start and end of lectures, for a little babysitting and video editing 2019-10-20T05:33:10 -!- con3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-20T05:33:11 < jadew> nice 2019-10-20T05:33:14 < Cracki> and they'll likely just go along with any price adjustments rather than bother finding someone else 2019-10-20T05:33:31 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T05:33:41 < jadew> I've lost contact with the world, so there aren't many normal people out there who know I can do stuff 2019-10-20T05:34:03 < jadew> and those that find out, I shut them off, because they want a website or some crap 2019-10-20T05:34:56 < jadew> my only option is to build a wildly successful business and become extremely rich 2019-10-20T05:35:39 < jadew> maybe I should start releasing some of the products I'm sitting on 2019-10-20T05:35:40 < Cracki> you could do what louis rossmann does, component level board repair 2019-10-20T05:35:46 < Cracki> fuck websites 2019-10-20T05:36:07 < jadew> Cracki, people here wouldn't pay what I'd want to be paid 2019-10-20T05:36:21 < Cracki> probably 2019-10-20T05:36:32 < jadew> I mean... I can fix broken stuff that I buy myself and earn a lot more than what I could earn by fixing other people's stuff 2019-10-20T05:36:49 < Cracki> socialize moar. find job fairs for the tech sector, go there to drop business cards 2019-10-20T05:37:14 < Cracki> some of those places may be better off contracting you rather than hiring 2019-10-20T05:38:06 < jadew> I was hoping you'd have different suggestions 2019-10-20T05:38:14 < jadew> something other than working 2019-10-20T05:38:21 < jadew> :D 2019-10-20T05:38:38 < jadew> I'm kinda fed up with working for other people tbh 2019-10-20T05:38:55 < jadew> that's the real reason why I think building a business is my the only option I have 2019-10-20T05:40:18 < jadew> anyway, I can work today for a client that needs some work done and I can afford those boxes tomorrow 2019-10-20T05:40:25 < jadew> but I don't enjoy doing it 2019-10-20T05:43:33 < jadew> https://www.youtube.com/watch?v=IAQClL3bZzg 2019-10-20T05:45:48 -!- splud [~noneya.bi@unaffiliated/splud] has quit [Ping timeout: 265 seconds] 2019-10-20T05:46:03 < Cracki> if not contract work, then you sell products, to consumers or businesses 2019-10-20T05:46:23 < Cracki> so you need to figure out who needs what 2019-10-20T05:46:30 < Cracki> still means shaking hands and networking 2019-10-20T05:47:14 < Cracki> only diff is you don't make one-offs but form viable products from what people wish they could buy 2019-10-20T05:47:31 < jadew> that's the plan, yeah 2019-10-20T05:47:43 < Cracki> imagine who are users/customers of RF applications 2019-10-20T05:48:01 < jadew> thing is, I already started working on a couple of things that I needed at one point 2019-10-20T05:48:13 < jadew> I didn't make any market research 2019-10-20T05:48:24 < Cracki> low budget ham operators, industrial too, maybe local internet providers that need point to point links 2019-10-20T05:48:28 < jadew> realized I should have done that at some point, but it was a bit too late 2019-10-20T05:49:06 < Cracki> emergency services (firefighters, ambulance) have their own stuff that's certified and all that, I would think 2019-10-20T05:49:18 < jadew> for sure 2019-10-20T05:49:28 < jadew> internet is all fiber here 2019-10-20T05:49:33 < jadew> even in the rural areas 2019-10-20T05:49:56 < jadew> and I'm sure there are providers for everything they need, that manufacture in extremely large volumes 2019-10-20T05:50:00 < jadew> so... low prices too 2019-10-20T05:51:25 < jadew> I'm not complaining, just moaning a bit 2019-10-20T05:52:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-20T05:52:43 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-20T05:52:44 < jadew> until I get at least a couple more products out, I won't have a clear picture of what it's like and if it's a trajectory that's worth pursuing 2019-10-20T05:53:32 < jadew> so far, my conclusion is that actual products take a lot of effort (and money) to produce 2019-10-20T05:54:18 < jadew> there's a big difference between building for yourself and building for others 2019-10-20T06:01:11 < Cracki> so... replacement parts eh 2019-10-20T06:01:21 < Cracki> HP frequency counters are low volume 2019-10-20T06:01:36 < jadew> yeah, the next one up is even lower volume hehe 2019-10-20T06:01:49 < Cracki> you might wanna look into replacement parts for cars 2019-10-20T06:02:05 < jadew> don't the chinese already have that covered? 2019-10-20T06:02:07 < Cracki> or reselling obd dongles, or selling "tuneups" and all that 2019-10-20T06:02:15 < Cracki> hm right, probably 2019-10-20T06:02:52 < Cracki> so what can they not compete on? 2019-10-20T06:03:18 < jadew> there's nothing they can't compete on 2019-10-20T06:03:36 < Cracki> turnaround time (time and distance) 2019-10-20T06:03:43 < jadew> especially since I see a trend where they do the absolute minimum, even if it's not right 2019-10-20T06:03:43 < Cracki> that not true 2019-10-20T06:03:45 < jadew> and people buy it 2019-10-20T06:03:49 < jadew> because it's cheaper 2019-10-20T06:03:51 < Cracki> they don't have those HP frequency counter parts 2019-10-20T06:04:11 < Cracki> people buy that, find out it sucks, and then look for something decent 2019-10-20T06:04:22 < jadew> I don't think that's how it goes 2019-10-20T06:04:32 < Cracki> keep your eyes open for how taiwan markets itself 2019-10-20T06:04:43 < jadew> I think people buy that, and they clench their thirst for the product even if it sucks 2019-10-20T06:04:43 < Cracki> they piss on china in terms of quality 2019-10-20T06:04:52 < jadew> so they stop using it, but they don't buy the good one either 2019-10-20T06:05:18 < Cracki> more networking. good ideas don't come to you, you go to them. 2019-10-20T06:05:55 < Cracki> troll monthly radio enthusiast meetings 2019-10-20T06:06:05 < jadew> I have lots of ideas myself tho 2019-10-20T06:06:14 < jadew> like I said, I'm even sitting on some products 2019-10-20T06:06:21 < jadew> I'm actually sitting on one for about 2 years 2019-10-20T06:06:25 < Cracki> most of them know enough to make their rods vibrate but they don't know much of the EE behind it 2019-10-20T06:06:33 < jadew> it's 90% ready - basically usable 2019-10-20T06:06:43 < Cracki> so make an announcement :> 2019-10-20T06:07:07 < Cracki> preorder or something :P 2019-10-20T06:07:10 < jadew> I actually built a website too 2019-10-20T06:07:15 < jadew> no preorder necessary 2019-10-20T06:07:34 < jadew> the idea came to me here, because it helps answers a particular question 2019-10-20T06:08:23 < jadew> I basically made a really comprehensive search engine for MCUs and other components 2019-10-20T06:08:38 < Cracki> octopart 2019-10-20T06:08:45 < Cracki> but? :> 2019-10-20T06:08:54 < jadew> like that, but much much higher resolution 2019-10-20T06:09:27 < jadew> and nicer to use I think 2019-10-20T06:09:45 < jadew> let me see when I bought the domain name 2019-10-20T06:10:05 < jadew> 2016 2019-10-20T06:10:07 < jadew> jeez... 2019-10-20T06:10:19 < jadew> it's been ready for almost 3 years 2019-10-20T06:11:20 < Cracki> i know, you need to make tools for perfectionists! 2019-10-20T06:11:52 < jadew> sounds like something I'd enjoy :) 2019-10-20T06:12:31 < Cracki> does romania have hipsters? they're easy to get money out of, if the thing looks shiny enough 2019-10-20T06:13:06 < Cracki> did you know, they sold a "fidget cube", basically an executive desk toy. it's a few types of switches in a cube, you can press them and fiddle with them 2019-10-20T06:13:16 < Cracki> does nothing else except feel clicky and snappy 2019-10-20T06:13:34 < Cracki> handheld of course 2019-10-20T06:13:44 < jadew> heh 2019-10-20T06:14:19 < Cracki> ! i can switch the scissor part under my F10 and scroll key. the F10 is broken and tilts... I hope it's the scissor and not the backpanel 2019-10-20T06:15:00 < Cracki> crap 2019-10-20T06:15:26 < jadew> what are you talking about? 2019-10-20T06:18:11 < dongs> whats this lunix pattern of doing (x + 500) / 1000 in kernel code is that something like ceil() or someshit? 2019-10-20T06:18:30 < jadew> that's round 2019-10-20T06:18:39 < jadew> round() 2019-10-20T06:19:04 < jadew> it's actually round(x / 1000.0) 2019-10-20T06:19:41 < dongs> well no the input is usually * 1000 before going there 2019-10-20T06:19:58 < dongs> so its basically like ((foo * 1000) + 500) / 1000 2019-10-20T06:20:21 < Cracki> (int)(x/1000 + 0.5) == round(x/1000.0) 2019-10-20T06:20:35 < Cracki> (int)(x/1000.0 + 0.5) == round(x/1000.0) 2019-10-20T06:20:41 < Cracki> yeah 2019-10-20T06:20:52 < Cracki> but why would kernel code need to do that 2019-10-20T06:20:59 < dongs> to avoid floats i guess 2019-10-20T06:21:01 < Cracki> they don't need the cycles, they can use floats can't they 2019-10-20T06:21:06 < dongs> cant 2019-10-20T06:21:12 < jadew> dongs, I don't understand why they're scaling it up and then back again 2019-10-20T06:21:14 < jadew> makes no sense 2019-10-20T06:21:16 < Cracki> oh right that shit needs to run on floatless systems 2019-10-20T06:21:20 < dongs> jade, to avoid flaots i geuss 2019-10-20T06:21:28 < Cracki> it's probably not scaled up before 2019-10-20T06:21:31 < Cracki> just the division 2019-10-20T06:21:36 < jadew> dongs, sure, but your example there yields the same result 2019-10-20T06:21:40 < jadew> ((foo * 1000) + 500) / 1000 == foo 2019-10-20T06:21:50 < Cracki> if it's scaled, it's probably a kind of fixed point number 2019-10-20T06:21:53 < Cracki> lol wat 2019-10-20T06:22:08 < jadew> the (foo + 500) / 1000 trick only works if foo is not scaled up by 1000 2019-10-20T06:22:15 < jadew> if you do scale it up, it will always yield foo 2019-10-20T06:23:16 < jadew> the way rounding works is that by adding 500, if foo is say... 1500 or more, then it will become 2000 or more 2019-10-20T06:23:22 < Ultrasauce> its not just to let it run on floatless systems, the fpu state would need to be preserved across context shifts and that's a pretty significant overhead 2019-10-20T06:23:23 < jadew> then you divide it by 1000 and you get 2 2019-10-20T06:23:31 < jadew> if foo is always a multiple of 1000 2019-10-20T06:23:39 < jadew> if you add 500, you never get to the next digit 2019-10-20T06:24:49 -!- fc5dc9d4 [~quassel@p5B3A8777.dip0.t-ipconnect.de] has joined ##stm32 2019-10-20T06:25:37 < jadew> what's the type of foo? 2019-10-20T06:25:49 < dongs> uint32 2019-10-20T06:26:00 < jadew> then in that case that code makes no sense 2019-10-20T06:26:04 < jadew> you can try it 2019-10-20T06:26:42 < dongs> i did, i mean i think result is always same 2019-10-20T06:26:50 < Cracki> it must be a fixed point value, nothing else makes sense 2019-10-20T06:27:15 < Cracki> three decimal digits, means 1000 2019-10-20T06:27:29 < jadew> http://cpp.sh/8zlcs 2019-10-20T06:28:01 < Cracki> I doubt anything does (foo*1000 + x) / 1000, that's just pointless 2019-10-20T06:28:09 < dongs> no it really does 2019-10-20T06:28:13 < Cracki> uh 2019-10-20T06:28:26 -!- fc5dc9d4_ [~quassel@p57A3271A.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-20T06:28:32 < dongs> https://github.com/torvalds/linux/blob/master/drivers/media/dvb-frontends/stv6111.c#L529 and https://github.com/torvalds/linux/blob/master/drivers/media/dvb-frontends/stv6111.c#L452 2019-10-20T06:28:34 < Cracki> do I grep for literally 1000 as a constant? 2019-10-20T06:28:35 < jadew> it's possible it's old code that did something 2019-10-20T06:28:38 < jadew> then someone fixed it 2019-10-20T06:28:43 < jadew> and they didn't know why it's there 2019-10-20T06:28:47 < Cracki> aha! 2019-10-20T06:29:07 < Cracki> shit that's badly commented 2019-10-20T06:29:10 < Cracki> needs si units 2019-10-20T06:29:16 < jadew> but local_frequency is not scaled 2019-10-20T06:29:21 < jadew> is it? 2019-10-20T06:29:22 < dongs> it is, on 1st line 2019-10-20T06:29:25 < dongs> on 529 2019-10-20T06:29:27 < Cracki> doesn't look like it 2019-10-20T06:29:42 < Cracki> it's doing a round to nearest. 2019-10-20T06:29:47 < Cracki> index does a ceiling 2019-10-20T06:29:50 < dongs> p->frequency is kHz. 2019-10-20T06:30:01 < dongs> * 1000 before passing to set_lo 2019-10-20T06:30:05 < dongs> which +500 /1000 it 2019-10-20T06:30:10 < Cracki> ah lol 529 2019-10-20T06:30:13 < jadew> dongs, it's a different function tho 2019-10-20T06:30:13 < dongs> yes. 2019-10-20T06:30:19 < dongs> doesnt matter 2019-10-20T06:30:30 < jadew> so the rounding part is expecting anything 2019-10-20T06:30:33 < Cracki> yes, smells like accident to me too 2019-10-20T06:30:51 < jadew> it just happens that it's being fed the correct data 2019-10-20T06:30:59 < dongs> jadew, its a static func 2019-10-20T06:31:00 < dongs> called once 2019-10-20T06:31:29 < jadew> the idea there is to make sure that it's getting kHz rounded 2019-10-20T06:31:44 < jadew> since the calling function is already feeding them rounded, yes, it's not necessary 2019-10-20T06:32:08 < jadew> but the function still behaves properly and it makes sense 2019-10-20T06:32:47 < dongs> it would only be behaving properly if it was commented that input to funciton is in Hz 2019-10-20T06:32:54 < jadew> if you had that in the same function/block, then yeah, that would have been a mistake 2019-10-20T06:32:55 < dongs> until it is, its just magical lunix bullshit 2019-10-20T06:33:25 < jadew> well if input is in Hz and they need kHz in that function, shouldn't they round it? 2019-10-20T06:33:52 < dongs> input is in kHz, in dvb_params struct. 2019-10-20T06:34:09 < dongs> they scale it up for no reason 2019-10-20T06:34:11 < jadew> I'm taking about set_lof 2019-10-20T06:34:21 < jadew> which takes local_frequency in Hz, no? 2019-10-20T06:34:33 < dongs> it doesnt need to but yes 2019-10-20T06:34:41 < jadew> and then converts that to kHz by dividing it by 1000 (and also rounding it in the process) 2019-10-20T06:35:11 < jadew> I agree with Cracki tho, that code needs more comments 2019-10-20T06:35:49 < dongs> http://bcas.tv/paste/results/Z3QGuK16.html anyway i refactored it into this shit because fucking cancer 2019-10-20T06:38:05 < dongs> also their code is broken if xtal is anything other than 16mhz 2019-10-20T06:38:21 < dongs> the *standard* 30mhz that comes with that part the shit would break 2019-10-20T06:39:21 < jadew> so what are you doing now? fixing linux? 2019-10-20T06:39:26 < dongs> haha no 2019-10-20T06:39:32 < dongs> i needed a driver for that chip 2019-10-20T06:39:37 < dongs> found that lunix code was disgusting 2019-10-20T06:39:51 < dongs> was gonna copy it first but it was too disgusting to deal with 2019-10-20T06:40:22 < dongs> so i did a kinda half-merge between ST reference code and lunix shit 2019-10-20T07:00:17 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 264 seconds] 2019-10-20T07:07:25 -!- splud [~noneya.bi@unaffiliated/splud] has joined ##stm32 2019-10-20T07:15:13 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-rrttdwbsdtzhtuzc] has quit [Ping timeout: 265 seconds] 2019-10-20T07:25:56 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-kdzwmpjnkykyrult] has joined ##stm32 2019-10-20T07:34:42 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-20T07:37:56 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 265 seconds] 2019-10-20T07:37:56 -!- day__ is now known as day 2019-10-20T07:38:25 -!- RocketScientist [~User@31.192.14.195] has joined ##stm32 2019-10-20T08:10:46 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-20T08:25:37 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-20T08:55:17 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T08:56:19 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-20T09:14:41 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 264 seconds] 2019-10-20T09:23:29 < dongs> huh i just discovered multiple carets thing in vs2017 2019-10-20T09:23:30 < dongs> thats pretty handy 2019-10-20T09:28:08 -!- qyx [~qyx@gw2.krtko.org] has quit [Read error: Connection reset by peer] 2019-10-20T09:31:55 -!- qyx [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-20T09:55:26 < jadew> all editors have that these days 2019-10-20T09:55:45 < dongs> well, i dont use hipster shit 2019-10-20T09:55:50 < dongs> so i havent noticed 2019-10-20T09:56:03 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c8a3:c3d6:b8:626c] has joined ##stm32 2019-10-20T09:56:04 < jadew> IIRC VS 2015 had it too 2019-10-20T09:56:50 < doomba> dongs writes code in notepad like a real g. 2019-10-20T09:58:52 < jadew> he uses keil, which uses scintilla as the editor 2019-10-20T09:59:11 < jadew> and that one had multiple cursors for a decade too 2019-10-20T09:59:25 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-20T10:01:02 < dongs> keil doesnt have that tho lol 2019-10-20T10:01:21 < jadew> are you sure? 2019-10-20T10:01:24 < dongs> yeah 2019-10-20T10:01:28 < jadew> hold alt + shift 2019-10-20T10:01:32 < jadew> and then down key 2019-10-20T10:01:46 < jadew> (make sure you have some lines of code down) 2019-10-20T10:02:16 < dongs> lul 2019-10-20T10:02:23 < jadew> works, eh? 2019-10-20T10:02:40 < dongs> well kinda but no 2019-10-20T10:02:46 < jadew> it's not the same as the one in VS 2019-10-20T10:02:52 < dongs> multi caret stuff in vs doesnt need to be on next line 2019-10-20T10:02:52 < jadew> but it's something 2019-10-20T10:03:12 < dongs> basicallyt its useful for refactoring, this scintilla shit sounds like its just vertical block select 2019-10-20T10:03:33 < jadew> yeah, but you can use for editing stuff too, if they're aligned 2019-10-20T10:03:56 < dongs> eifhr 2019-10-20T10:03:58 < dongs> right 2019-10-20T10:19:41 < antto> Code::Blocks, scintilla 2019-10-20T10:19:47 < antto> geany too 2019-10-20T10:19:58 < antto> iz noice 2019-10-20T10:20:11 < antto> such code, very text 2019-10-20T10:20:21 < jadew> aren't all of those scintilla based? 2019-10-20T10:20:37 < jadew> pretty sure geany is, don't know about Code:Blocks 2019-10-20T10:20:51 < antto> well, uh, i just.. said it above 2019-10-20T10:21:02 < antto> wxscintilla but still scintilla 2019-10-20T10:21:40 < antto> in dumb M$ IDE you can't even select a bitmap font 2019-10-20T10:21:48 < antto> wot dee fug 2019-10-20T10:21:59 < jadew> huh? 2019-10-20T10:22:16 < jadew> first of all, why would you use a bitmap font? 2019-10-20T10:22:25 < antto> coz it's f*cking noice 2019-10-20T10:22:51 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-kdzwmpjnkykyrult] has quit [Ping timeout: 240 seconds] 2019-10-20T10:22:59 < antto> do i sense discrimination against some kinds of fonts, boi? 2019-10-20T10:23:24 < jadew> I like my fonts crisp 2019-10-20T10:23:34 < antto> so? 2019-10-20T10:23:40 < qyx> are we in 1990? 2019-10-20T10:23:43 < antto> u got something against bitmap fonts? 2019-10-20T10:23:44 < qyx> bitmap fonts? 2019-10-20T10:23:52 < qyx> u no hinting? 2019-10-20T10:23:55 < antto> qyx u too? racist 2019-10-20T10:24:15 < antto> muh font is eggcelent 2019-10-20T10:31:58 < antto> https://i.imgur.com/19RU7ie.png 2019-10-20T10:33:41 < jadew> I like consolas 2019-10-20T10:34:05 < antto> i haven't found a vector font that doesn't look dumb at that size 2019-10-20T10:34:49 < antto> also i don't think i like consolas specifically 2019-10-20T10:35:18 < antto> liberation Mono is sorta okay 2019-10-20T10:53:06 < dongs> disgusting kikecad screenshot 2019-10-20T10:55:29 < Steffanx> Lol triggered 2019-10-20T10:55:30 < doomba> so 90's 2019-10-20T10:56:47 < doomba> h4x0riz3d xfce with themes from gnome-look 2019-10-20T11:01:21 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-fzkcxxijextsyxon] has joined ##stm32 2019-10-20T11:05:34 -!- renn0xtk9 [~max@2a02:810d:1540:2448:c8a3:c3d6:b8:626c] has quit [Quit: Konversation terminated!] 2019-10-20T11:22:46 < antto> dongs you're mostly seeing qpdfview there, not kicad 2019-10-20T11:22:52 < antto> 'o-o' 2019-10-20T11:26:29 < srk> hack font .. http://48.io/~rmarko/random/2019-10-20-102348_1920x1080_scrot.png 2019-10-20T11:27:22 < Steffanx> The background makes it complete 2019-10-20T11:28:09 < Steffanx> Doesnt the green hurt your eyes? 2019-10-20T11:28:33 < srk> no, I'm using redshift as well 2019-10-20T11:28:56 < srk> background http://48.io/~rmarko/random/luna.jpg 2019-10-20T11:29:26 < srk> they change randomly every day :) 2019-10-20T11:31:35 < antto> such much dreamy wallpaper 2019-10-20T11:32:05 < antto> Steffanx muh IRC is all green and red 2019-10-20T11:32:53 < Steffanx> The perfect eye test 2019-10-20T11:35:40 < antto> https://i.imgur.com/Ov0zQk1.png 2019-10-20T11:35:48 < antto> but now i use FixedSys 2019-10-20T11:36:17 < antto> one of the few gud things in crapdows, but i wonder if it was even made by M$ 2019-10-20T11:37:27 < antto> doomba it's LXDE akchually 2019-10-20T12:01:38 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-20T12:02:01 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-20T12:12:18 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T12:18:15 < RocketScientist> Anyone explain please, Mahony filter using gyroscope/accelerometer, has drift over time, right? Madgwick has not, because it also uses magnetometer so North pole stays still and gravity also, is that correct? 2019-10-20T12:21:51 < RocketScientist> Do Mahony filter also have drift for roll and pitch or only for yaw? 2019-10-20T12:30:46 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 252 seconds] 2019-10-20T12:39:34 -!- ekaologik [~quassel@p4FF1605D.dip0.t-ipconnect.de] has joined ##stm32 2019-10-20T13:26:47 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 268 seconds] 2019-10-20T13:26:58 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-20T13:33:44 -!- ekaologik [~quassel@p4FF1605D.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-20T13:33:55 -!- ekaologik [~quassel@p4FF1605D.dip0.t-ipconnect.de] has joined ##stm32 2019-10-20T13:45:54 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-20T14:36:13 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection] 2019-10-20T14:55:20 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-20T14:57:50 < dongs> RocketScientist: think about it, why would it have drift for roll/pitch when you have accelerometer vector to always keep it same orientation 2019-10-20T14:58:05 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-20T15:05:31 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T15:06:25 < RocketScientist> dongs: so it will not have any drift even if airplane for example or quadcopter is constantly moving, and we started calculating it with Mahony while moving? without knowing start position. 2019-10-20T15:38:03 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 250 seconds] 2019-10-20T15:40:30 < zyp> RocketScientist, have you looked at how the filters are working? 2019-10-20T15:43:19 < zyp> since the gyro measures angular velocity that is integrated over time, any offset in the measurement will also get accumulated and cause drift 2019-10-20T15:43:36 < RocketScientist> zyp: yeah, there was some math, just like alien written saint manuscript. understood nothing. 2019-10-20T15:43:56 < zyp> the filters combat this by estimating the offset and subtracting it 2019-10-20T15:44:07 < zyp> gtg, can explain more later 2019-10-20T15:44:40 < RocketScientist> zyp: so, it will not drift for only pitch/yaw ever? for roll it will drift I think. 2019-10-20T15:45:59 -!- ekaologik [~quassel@p4FF1605D.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-20T15:47:12 < RocketScientist> my Aerial Vehicle Bench Test Kit got AHRS display working based on Madgwick, no idea it will drift or not, I want to publish it as sonn as possible :) 2019-10-20T15:56:36 < catphish> after getting my usb IN throughput working nicely, i've realised that OUT throughput now sucks, might need to go back to double buffering for that 2019-10-20T15:58:06 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-20T16:18:16 < srk> RocketScientist: any IMU will drift due to integration errors 2019-10-20T16:18:50 < srk> need to add gps to the mix :) 2019-10-20T16:28:51 < zyp> RocketScientist, the gyro signals are integrated to give you an orientation reference, either as a rotation matrix or a quaternion 2019-10-20T16:29:07 < con3> eh i messed up and now altium is highlighting the current layer im on... anyone know a shortcut or way to stop this? 2019-10-20T16:29:58 < zyp> RocketScientist, then you take this reference and generate a vector pointing to what you believe is up 2019-10-20T16:30:25 < zyp> RocketScientist, accelerometer measures gravitational acceleration which is also a vector pointing up 2019-10-20T16:30:50 < zyp> both of these should be unit vectors, i.e. with magnitude 1, so the only difference would be in direction 2019-10-20T16:31:22 < zyp> and the difference in direction is the error between estimated up and measured up 2019-10-20T16:31:49 < zyp> so there's a feedback controller on that error feeding back to estimated gyro offset 2019-10-20T16:33:31 < zyp> when you add a magnetometer you'll get a north reference in addition to the up reference 2019-10-20T17:17:13 < deltab> con3: * + or - on the numeric keypad? 2019-10-20T17:17:28 < deltab> based on https://www.altium.com/documentation/altium-designer/pcb-cmd-setcurrentlayersetcurrentlayer-ad?version=18.0 2019-10-20T17:19:49 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-20T17:27:39 < con3> deltab: This seems to only change between layers 2019-10-20T17:27:43 < con3> will post a pic quick 2019-10-20T17:29:43 < con3> https://imgur.com/a/gimNQWy 2019-10-20T17:30:31 < deltab> can you switch it to anothe rmode that doesn't have a current layer? 2019-10-20T17:31:33 < deltab> I'm just guessing based in what I know of other software; never used altium 2019-10-20T17:48:17 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-20T18:04:10 -!- sterna [~Adium@2a02:aa1:1619:8e7f:c430:42e9:863e:ff1d] has joined ##stm32 2019-10-20T18:10:38 < zyp> con3, shift-S? 2019-10-20T18:25:16 < jadew> is there something better than chromecast? 2019-10-20T18:25:30 < mawk> chromecast works pretty nice 2019-10-20T18:25:50 < mawk> you have a multicast protocol whose name I forgot which works nice for media sharing, if your tv/stb supports it 2019-10-20T18:25:53 < mawk> then no need for an external device 2019-10-20T18:26:00 < mawk> but chromecast will solve all the headache 2019-10-20T18:26:00 < jadew> miracast 2019-10-20T18:26:03 < mawk> yes maybe 2019-10-20T18:35:30 -!- renn0xtk9 [~max@2a02:810d:1540:2448:7cd4:f0b5:987f:133] has joined ##stm32 2019-10-20T18:38:24 < aandrew> yeah I really like chromecast, the only thing I wish it *did* do was bluetooth audio 2019-10-20T18:38:41 < jadew> for headphones? 2019-10-20T18:38:50 < aandrew> the CC knows what teh audio delay should be / can adjust it, where the tv with an external bt dongle can't 2019-10-20T18:38:53 < aandrew> exactly 2019-10-20T18:39:50 < con3> zyp: you are a god among men 2019-10-20T18:39:56 < con3> sorry had a phonecall 2019-10-20T18:40:04 < con3> shift-s for the win 2019-10-20T18:40:18 < aandrew> altium? yes, shift-s is life 2019-10-20T18:45:17 -!- con3 [~kvirc@154.119.40.228] has quit [Ping timeout: 264 seconds] 2019-10-20T18:45:40 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-20T19:01:22 < kakimir64> I need freertos pros 2019-10-20T19:01:27 < kakimir64> where are the propros 2019-10-20T19:01:35 < invzim> I need someone to teach France English 2019-10-20T19:08:20 < BrainDamage> miracast uses wifi p2p 2019-10-20T19:08:35 < BrainDamage> so your device directly connects to the tv to stream data 2019-10-20T19:08:38 < BrainDamage> it's pretty nifty 2019-10-20T19:09:07 < BrainDamage> ( most modern wifi chips have dual radios, you can connect to the wifi while maintaining connection to the AP ) 2019-10-20T19:09:22 < BrainDamage> ( and by modern, I mean more recent than 10 years ) 2019-10-20T19:18:14 < kakimir64> well fuck 2019-10-20T19:18:38 < kakimir64> I used wrong attribute symbols all along 2019-10-20T19:18:47 < kakimir64> no wonder mpu didn't play ball 2019-10-20T19:19:02 < kakimir64> I used tskMPU ones 2019-10-20T19:19:14 < kakimir64> when I should have used portMPU ones 2019-10-20T19:20:35 < kakimir64> and I copied them from mpu demo provided with latest freertos 2019-10-20T19:21:24 < kakimir64> I need to rage to them now 2019-10-20T19:22:02 < aandrew> kakimir64: what's wrong with your freertos 2019-10-20T19:22:18 < kakimir64> nothing 2019-10-20T19:22:37 < kakimir64> but there is 2 sets of attribute symbols for mpu region attributes 2019-10-20T19:22:46 < kakimir64> tskMPU and portMPU 2019-10-20T19:22:56 < kakimir64> why 2019-10-20T19:27:09 < aandrew> https://imgur.com/a/iphr8Rd ok that's pretty slick 2019-10-20T19:27:27 < aandrew> 4x STM32s spitting data over UDP to influxdb and grafana looking at it 2019-10-20T19:28:34 < zyp> what data rate? 2019-10-20T19:30:28 < kakimir64> now I can start writing application tasks 2019-10-20T19:31:29 < aandrew> zyp: 250ms into the DB, but graphing updates once a second 2019-10-20T19:31:36 < aandrew> it's not fast display for sure 2019-10-20T19:31:57 -!- renn0xtk9 [~max@2a02:810d:1540:2448:7cd4:f0b5:987f:133] has quit [Quit: Konversation terminated!] 2019-10-20T19:31:57 < zyp> 4Hz is not really all that fast 2019-10-20T19:32:36 < zyp> wonder if it could do 1000Hz, or if that is approaching Laurenceb__-level of dumbness 2019-10-20T19:35:40 < Steffanx> does it involve memes data over audio? 2019-10-20T19:36:37 < sync> zyp: with influx and grafana? 2019-10-20T19:36:42 < sync> no wai, most likely 2019-10-20T19:41:36 < zyp> sync, yeah 2019-10-20T19:45:16 < jadew> BrainDamage, yeah, our main TV has miracast, but the old one has something that doesn't seem to work with my android phone 2019-10-20T19:45:55 < BrainDamage> android is killing miracast in recent versions 2019-10-20T19:46:05 < BrainDamage> because it wants everyone to switch to chromecast 2019-10-20T19:46:15 < BrainDamage> some manifacturers add it back, but ymmv 2019-10-20T19:46:43 < jadew> ah ha, maybe it's one of those things then 2019-10-20T19:46:48 < Jan-> hihi stm people 2019-10-20T19:46:51 < jadew> I can cast from my PC to both devices 2019-10-20T19:46:51 * Jan- has toast 2019-10-20T19:50:40 < kakimir64> could I link ISRs to privileged_functions? I mean ISRs are run in privileged mode and as MPU region of privileged functions must be 2^^ and total size of freertos privileged functions is about 10k there is 6k of unused space 2019-10-20T19:51:38 <@englishman> influxdb should have nanosecond precision no/ 2019-10-20T19:51:39 <@englishman> ? 2019-10-20T19:51:43 <@englishman> 1kHz is prolly fine 2019-10-20T19:52:09 <@englishman> it should certainly be able to take in data at >>1kHz 2019-10-20T19:52:18 < aandrew> zyp: right, that's what I was saying, I didn't need superfast, just needed to be able teo see what was going on 2019-10-20T19:52:48 < aandrew> englishman: yes it does 2019-10-20T19:52:52 < aandrew> but I don't care for that 2019-10-20T19:53:10 <@englishman> right, i don't care about your project either 2019-10-20T19:55:09 < aandrew> you're just sore becuase you're in Quebec 2019-10-20T19:55:21 <@englishman> how so 2019-10-20T19:55:28 <@englishman> the government just paid me $13k to buy a car 2019-10-20T19:55:36 < aandrew> what car did you get 2019-10-20T19:55:40 <@englishman> a tesla 2019-10-20T19:55:51 < aandrew> nice. in Ontario they killed that 2019-10-20T19:55:55 <@englishman> yep 2019-10-20T19:55:57 < aandrew> but hey, buck a beer 2019-10-20T19:56:11 <@englishman> alcoholics > enviroment 2019-10-20T19:56:18 < aandrew> tru 2019-10-20T19:56:27 < aandrew> only an alkie would buy the shitty beer that is $1 2019-10-20T19:57:05 <@englishman> il be installing influxdb @ work to chart and alert when shit fucks up in production 2019-10-20T19:57:50 <@englishman> mostly a fun excuse to buy an 80" tv tho 2019-10-20T19:57:51 < BrainDamage> how the hell can it have nanosecond precision when the master clock on computers has drifts in the order of tens of ppm 2019-10-20T19:58:30 <@englishman> precision != accuracy 2019-10-20T19:59:17 < BrainDamage> yes, I said _drift_ 2019-10-20T19:59:33 < BrainDamage> it can have nanosecond resolution 2019-10-20T19:59:57 < BrainDamage> but a random walk reduces precision 2019-10-20T20:09:35 < kakimir64> my bluetooth problems might be about usb-autosuspend 2019-10-20T20:09:40 < kakimir64> or usb powersaving 2019-10-20T20:20:32 < steve> stm32 SAI requires MCLK (master clock) to be 256x the frame sync freq. MCLK is used to clock an external codec. It's not used internally. So why the hard requirement of 256? 2019-10-20T20:28:19 < aandrew> BrainDamage: precision != accuracy 2019-10-20T20:28:24 < aandrew> oh yeah what englishman said lol 2019-10-20T20:28:42 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-20T20:40:41 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-20T20:40:45 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-20T20:44:29 < Cracki> https://www.vice.com/en_uk/article/ywx9y5/i-tried-living-on-bugs-to-see-if-ill-enjoy-the-future-of-food 2019-10-20T20:46:32 < aandrew> Cracki: I was reading https://www.vice.com/en_us/article/nzqayz/this-teen-hacked-150000-printers-to-show-how-the-internet-of-things-is-shit 2019-10-20T20:46:43 -!- scrts [~scrts@unaffiliated/scrts] has quit [Remote host closed the connection] 2019-10-20T20:46:44 -!- scrts2 [~scrts@d27-96-211-8.nap.wideopenwest.com] has quit [Remote host closed the connection] 2019-10-20T20:49:02 < Cracki> cute 2019-10-20T20:49:52 < Cracki> printers are now at the level windows xp was 2019-10-20T20:50:12 < Cracki> progressing from hoovering up all the viruses to having a firewall 2019-10-20T20:50:52 < Cracki> autoconf is a good thing for lans but it can and should know when something isn't coming from the lan 2019-10-20T20:53:15 -!- sterna [~Adium@2a02:aa1:1619:8e7f:c430:42e9:863e:ff1d] has quit [Quit: Leaving.] 2019-10-20T20:56:11 < Cracki> >You don't have computer science programs in the UK? No, yeah, we do, but they're terrible. 2019-10-20T20:56:23 < Cracki> I looked at some stuff harvard teaches on operating systems 2019-10-20T20:56:38 < Cracki> they actually touch upon real world stuff like the busses that are in your computer 2019-10-20T20:56:50 < Cracki> such as how to use the pci bus and so on 2019-10-20T20:57:00 -!- scrts [~scrts@d27-96-211-8.nap.wideopenwest.com] has joined ##stm32 2019-10-20T20:57:17 < Cracki> we stayed abstract. it was basically a math class with some exercises in bash 2019-10-20T20:57:34 -!- scrts2 [~scrts@d27-96-211-8.nap.wideopenwest.com] has joined ##stm32 2019-10-20T21:23:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-20T21:25:07 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-20T21:31:59 < kakimir64> freertos 101 2019-10-20T21:32:38 < kakimir64> I want task to measure temperature when requred and then signal back to process that required it that value is valid 2019-10-20T21:33:30 < PaulFertser> kakimir64: message queues? 2019-10-20T21:34:13 < kakimir64> is there like a thing where process unlock when there is semaphore to be taken 2019-10-20T21:34:23 < kakimir64> but without taking the semaphore 2019-10-20T21:34:59 < kakimir64> and somehow see from process that requires temperature measurement that semaphore has been taken 2019-10-20T21:36:19 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-20T21:36:29 < PaulFertser> I am not sure I follow, why can't you send the temp task a message that you ask for temp and then get a message from it with the temp value kakimir64 ? 2019-10-20T21:37:18 < kakimir64> I need to look into it 2019-10-20T21:40:26 < zyp> kakimir64, sounds like you want a future, but I don't think freertos has that concept 2019-10-20T21:43:05 < zyp> apparently somebody has made an attempt at one though: https://gist.github.com/ciniml/e85c08664ee307645321ec3ce3973cc6 2019-10-20T21:44:49 < kakimir64> I need to change those ISRs to tasknotify instead of semaphore 2019-10-20T21:49:28 < Cracki> signal back to process... does it keep running in the meantime? 2019-10-20T21:49:43 < Cracki> or does it block for the result 2019-10-20T21:49:53 < Cracki> I agree with the future/promise 2019-10-20T21:54:01 < PaulFertser> Is there a nice OS like FreeRTOS but with modern C++ in mind so that futures would work as usual? 2019-10-20T21:56:43 < kakimir64> block 2019-10-20T21:57:00 < kakimir64> basically one process needs another process to do a thing 2019-10-20T21:57:15 < kakimir64> and waits until thing has been done 2019-10-20T21:57:20 < kakimir64> blocks* 2019-10-20T21:57:44 < PaulFertser> In C I'd just send a message and then would wait for a message... 2019-10-20T22:00:35 < kakimir64> how about 2 semaphores? 2019-10-20T22:01:12 < kakimir64> one going one way and another going another way 2019-10-20T22:01:41 < kakimir64> It sounds and looks nasty but is it? 2019-10-20T22:02:00 < kakimir64> it's definitelly easy to comprehend at least 2019-10-20T22:02:07 < PaulFertser> But you also need to pass data 2019-10-20T22:02:43 < PaulFertser> Please tell me folks why using freertos message queue is a bad idea in this case? 2019-10-20T22:03:01 < kakimir64> I don't 2019-10-20T22:03:23 < kakimir64> I don't need to pass data I have struct of shared variables 2019-10-20T22:03:32 < PaulFertser> That's nasty 2019-10-20T22:03:39 < kakimir64> yes 2019-10-20T22:04:18 < Cracki> you can block on a queue, wait for the result to arrive 2019-10-20T22:05:01 < Cracki> message passing is the most convenient thing you can do 2019-10-20T22:05:20 < Cracki> if you have that data elsewhere already, yes semaphore 2019-10-20T22:05:34 < Cracki> producer increments semaphore, consumer decrements/blocks on semaphore 2019-10-20T22:05:38 < Cracki> (basically diy queue) 2019-10-20T22:06:18 < Cracki> so yes, two semaphores, if you have some kind of "bidirectional" communication 2019-10-20T22:06:40 < kakimir64> how about if I have multiple tasks that want to access temperature info 2019-10-20T22:06:50 < kakimir64> and have fresh or rather fresh info 2019-10-20T22:06:56 < Cracki> then that's a read on a variable 2019-10-20T22:07:14 < Cracki> and that read is either atomic (word size) or you need to use a lock to guard access 2019-10-20T22:07:27 < Cracki> if it's word size, don't bother fucking around, just read it 2019-10-20T22:07:40 < kakimir64> okay 2019-10-20T22:07:56 < Cracki> if you need to wait for actually new data, rather than polling, then there are other primitives called monitor/event/condition 2019-10-20T22:08:15 < Cracki> threads can block on that, and something pokes it, and that wakes one or all waiting threads up 2019-10-20T22:08:30 < Cracki> I'm assuming you have a single core 2019-10-20T22:08:33 < kakimir64> let's say only one task needs the temperature but temperature would be measured at it's own pace 2019-10-20T22:09:20 < kakimir64> but if temperature information is not fresh the system should just crash instead of continuing 2019-10-20T22:09:31 < Cracki> wat crash? 2019-10-20T22:09:41 < Cracki> sounds like you want to use a watchdog timer 2019-10-20T22:09:45 < kakimir64> yes 2019-10-20T22:09:49 < Cracki> then do that :P 2019-10-20T22:10:25 < kakimir64> but how do I detect the temperature is fresh? 2019-10-20T22:10:37 < kakimir64> combine it with xTickCount? 2019-10-20T22:10:52 < Cracki> temperature reader resets watchdog 2019-10-20T22:11:00 < kakimir64> no 2019-10-20T22:11:02 < Cracki> watchdog resets whole chip when it runs out 2019-10-20T22:11:12 < Cracki> wat no 2019-10-20T22:11:12 < kakimir64> watchdog reset needs to be much more complex than that 2019-10-20T22:11:18 < Cracki> huh 2019-10-20T22:11:23 < kakimir64> it needs to monitor multiple tasks 2019-10-20T22:11:28 < Cracki> oh well, sure, store a timestamp for that sensor value 2019-10-20T22:12:03 < kakimir64> ie. that thermostat task is much more important related to watchdog reset 2019-10-20T22:24:34 < kakimir64> should after_vectors be after vectors? 2019-10-20T22:24:47 < kakimir64> right after 2019-10-20T22:25:14 < kakimir64> I mean this mcuxpresso template puts them after data and bss 2019-10-20T22:26:40 < kakimir64> sauna> 2019-10-20T22:28:36 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-20T22:36:03 -!- con3 [~kvirc@154.119.40.228] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-20T22:47:56 < BrainDamage> dongs: since you struggle with git at times: https://girliemac.com/blog/2017/12/26/git-purr/ 2019-10-20T22:56:56 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-20T23:02:16 < Steffanx> But girls and mac. No way he will like it, BrainDamage 2019-10-20T23:27:12 < kakimir64> BrainDamage: I don't understand at all 2019-10-20T23:36:31 -!- RocketScientist [~User@31.192.14.195] has quit [Quit: Leaving.] 2019-10-20T23:36:51 -!- RocketScientist [~User@31.192.14.195] has joined ##stm32 2019-10-20T23:39:01 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-20T23:41:42 -!- RocketScientist [~User@31.192.14.195] has quit [Ping timeout: 265 seconds] --- Day changed Mon Oct 21 2019 2019-10-21T00:02:43 < antto> old macgurlie had a mac 2019-10-21T00:03:32 < antto> senpai didn't give a f*ck 2019-10-21T00:04:15 < Steffanx> E-I-E-I-O 2019-10-21T00:04:23 < antto> yeah, that too 2019-10-21T00:04:27 < Steffanx> You forgot that. 2019-10-21T00:04:35 < antto> i didn't know how to spell it >:/ 2019-10-21T00:09:17 < antto> old macgurlie went to church 2019-10-21T00:09:36 < antto> this gon b hard to rhyme 2019-10-21T00:10:17 < antto> senpai ran a google surch 2019-10-21T00:10:19 < antto> >:) 2019-10-21T00:12:20 < deltab> lurch, perch, merch, birch 2019-10-21T00:12:38 -!- gnom [~aleksande@178.150.7.153] has quit [Ping timeout: 276 seconds] 2019-10-21T00:13:39 < kakimir64> it's seems like marginal space saving to move .after_vectors inside privileged_functions 2019-10-21T00:14:05 < antto> kakimir64 is sooooo serious, i can't recognize him 2019-10-21T00:14:56 < kakimir64> most of routines are couple of bytes or what is takes to perform while(1) 2019-10-21T00:15:02 < Steffanx> The original song doesnt rhyme does it antto ? 2019-10-21T00:15:19 < antto> i don't fully know teh original song anyway ;P~ 2019-10-21T00:15:47 < antto> rhymes make things bett0r ya know 2019-10-21T00:18:11 < kakimir64> https://paste.ee/p/xtNHT what causes that massive fill of 0x4000 ? 2019-10-21T00:18:18 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-21T00:18:37 < kakimir64> line 39 2019-10-21T00:18:53 -!- gnom [~aleksande@178.150.7.153] has joined ##stm32 2019-10-21T00:19:07 < kakimir64> it should start all the shit after address 0x4000 but it gets it totally wrong 2019-10-21T00:20:23 < kakimir64> section_table_start should be 0x4000 2019-10-21T00:22:46 < kakimir64> linker doesn't understand me! 2019-10-21T00:24:14 < doomba> did you try switching back to kakimir32? :P 2019-10-21T00:24:37 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-21T00:26:09 < specing> forget kakimir32 2019-10-21T00:26:24 < specing> go back to kakimir8, no endianess madness 2019-10-21T00:26:43 < BrainDamage> little indian kakmir 2019-10-21T00:29:09 < kakimir64> trick of moving after_vectors to privileged_functions worked though 2019-10-21T00:29:30 < kakimir64> those start right after freertos privileged api 2019-10-21T00:29:31 < ub|k> if i want to sample an ADC continuously, does it suffice to set DDS, or do I also have to set CONT? 2019-10-21T00:29:59 < ub|k> sorry, forgot to mention, using DMA. 2019-10-21T00:50:23 < Steffanx> You still need the CONT. DDS is just a setting to keep the DMA requests going. 2019-10-21T00:50:31 < Steffanx> After a conversion 2019-10-21T01:10:06 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-21T01:17:58 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-21T01:20:19 < kakimir64> https://www.youtube.com/watch?v=LpOt2U8JwNo musics 2019-10-21T01:20:57 < Steffanx> Time to sleep, mr kakimir32 2019-10-21T01:21:03 < Steffanx> I mean kakimir64 2019-10-21T01:21:17 < kakimir64> I just got up? 2019-10-21T01:21:29 < Steffanx> Lolwut? 2019-10-21T01:21:39 < kakimir64> at least I feel like that 2019-10-21T01:21:43 < Steffanx> You moved to Australia? 2019-10-21T01:21:46 < Steffanx> Japan? 2019-10-21T01:21:57 < kakimir64> my internal clock moved somewhere 2019-10-21T01:22:04 < kakimir64> and it keeps moving 2019-10-21T01:22:10 < kakimir64> from place to place 2019-10-21T01:22:22 < Steffanx> Stop the drugs 2019-10-21T01:23:28 < kakimir64> it's not drugs 2019-10-21T01:23:30 < kakimir64> it's me 2019-10-21T01:24:06 < Steffanx> I see 2019-10-21T01:38:35 < kakimir64> I have hard time finding 2 used 2Tb drives steff 2019-10-21T01:38:55 < kakimir64> as new 2Tb is 70euros 2019-10-21T01:38:59 < kakimir64> but 4Tb is 110 2019-10-21T01:39:30 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-21T01:45:37 < kakimir64> I wonder if it makes sense to even buy 2TB 2019-10-21T01:45:38 < kakimir64> as new 2019-10-21T01:53:18 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-21T02:16:49 -!- onio [~onio@2a00:23c5:7a01:8600:bcb3:91dd:6f5a:79df] has quit [Quit: Leaving] 2019-10-21T02:22:33 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-21T02:23:16 < dongs> kikecad gerber output still not readable by cam350 because of retarded commetns at top 2019-10-21T02:23:21 < dongs> have to fucking open each one and delete 2019-10-21T02:27:25 < dongs> so how do people use fatfs async? just stick that shit in a separate thread?> 2019-10-21T02:29:14 < BrainDamage> dongs: if you have sed installed in windows, use sed /'^G04 #@!'/d -i * in the dir 2019-10-21T02:29:22 < BrainDamage> it'll delete all the top comments in the files 2019-10-21T02:29:59 < BrainDamage> or better, it'll delete all the lines that start with G04 #@! 2019-10-21T02:30:04 < BrainDamage> which are kicad top comments 2019-10-21T02:30:19 < dongs> its not G04, its the %FS before that 2019-10-21T02:30:21 < dongs> like 5-6 lines 2019-10-21T02:30:44 < BrainDamage> %FSLAX46Y46*% < this? 2019-10-21T02:30:53 < dongs> nope. sec i dleete them from all already lemme get the shit 2019-10-21T02:31:08 < dongs> %TF.GenerationSoftware,KiCad,Pcbnew,(5.1.4-0-10_14)*% 2019-10-21T02:31:13 < dongs> ah %TF sorry 2019-10-21T02:31:28 < dongs> i delete up to %FSLAX shit and it works fine 2019-10-21T02:31:33 < dongs> including G04 under that 2019-10-21T02:31:38 < dongs> G04 Gerber Fmt 4.6, Leading zero omitted, Abs format (unit mm)* 2019-10-21T02:31:38 < dongs> G04 Created by KiCad (PCBNEW (5.1.4-0-10_14)) date 2019-09-20 11:41:29* 2019-10-21T02:32:33 < BrainDamage> so all the lines up to, including that? 2019-10-21T02:33:06 < dongs> all %TF from top = works. so prolly sed /^%TF or whatever. 2019-10-21T02:33:57 < BrainDamage> yup, that should work, ^ means beginning of the line 2019-10-21T02:34:02 < BrainDamage> and /d delete 2019-10-21T02:34:20 < BrainDamage> you might have to escape the %, because sed by default uses regexes 2019-10-21T02:41:23 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-21T02:49:03 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-21T02:49:40 < deltab> % has no special meaning in regexes that I know of, other than vim's \% extensions 2019-10-21T02:57:13 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-21T03:00:06 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-21T03:00:12 < bitmask> Hi ppl 2019-10-21T03:09:55 < catphish> i've got my stm32 usb transfer rates up to 1.13MB/s now :) 2019-10-21T03:10:20 < catphish> but... unfortunately my host is shit and sees to throttle usb rates when the cpu is idle, not cool 2019-10-21T03:11:38 < Cracki> lol when idle 2019-10-21T03:11:52 < Cracki> sounds like it's messing with interrupts a little too much 2019-10-21T03:12:07 < Cracki> how'd you get the 1.13 MB/s? 2019-10-21T03:12:19 < Cracki> figured out the double buffering? 2019-10-21T03:13:23 < catphish> now got it up to 1.21MB/s :) 2019-10-21T03:13:41 < catphish> but yes, double buffering, and using large transfers on the host side 2019-10-21T03:17:37 < dongs> 1.21 niggawatts? 2019-10-21T03:24:52 < bitmask> I need some scary movie recommendations 2019-10-21T03:30:24 < Cracki> so... Scary Movie or Scream 2019-10-21T03:30:41 < Cracki> prefer a decade? 2019-10-21T03:53:13 < bitmask> newish 2019-10-21T03:53:19 < bitmask> something I havent seen 2019-10-21T03:53:37 < bitmask> I think I narrowed it down to 3 on netflix 2019-10-21T03:54:08 < bitmask> just browsing lists of scary movies on netflix now and these 3 seemed to be in every list 2019-10-21T03:55:37 < dongs> > netflix 2019-10-21T03:55:40 < dongs> found your problem 2019-10-21T03:56:08 < bitmask> well if you have any non netflix recommendations I'm all ears 2019-10-21T03:56:36 < BrainDamage> nuttflex 2019-10-21T04:16:29 < bitmask> hmm, what size resistor should I use between the base of Q8 and the collector of Q9 2019-10-21T04:16:30 < bitmask> https://i.imgur.com/pMfM3nz.png 2019-10-21T04:18:28 < BrainDamage> do you want the transistor fully on in saturation region, or do you need to regulate the current? and how much is vbat? 2019-10-21T04:22:54 < bitmask> vbat is 13V or less, really 12.6 or less but i'll say 13 to be safe 2019-10-21T04:23:25 < bitmask> Its just used to measure the battery voltage so I do want it fully on 2019-10-21T04:23:36 < BrainDamage> then you take the lower voltage, not the higher 2019-10-21T04:24:00 < bitmask> say 10V 2019-10-21T04:24:16 < BrainDamage> also bear in mind that when you measure the voltage, the transistor in saturation will have roughly a fixed drop 2019-10-21T04:24:22 < BrainDamage> which will offset your measurement 2019-10-21T04:24:32 < bitmask> yea I'll account for that 2019-10-21T04:25:10 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-21T04:26:06 < BrainDamage> so the collector sinks approx 100uA, assuming an hFE of 50, that means that ib has to be 1/50th of that 2019-10-21T04:26:30 < BrainDamage> better if much more 2019-10-21T04:27:36 < BrainDamage> slap another 100k and it'll be in saturation 2019-10-21T04:28:11 < BrainDamage> so you won't even need another resistor 2019-10-21T04:28:42 < BrainDamage> the requirement is R << (10V-0.7V-0.5V)/(100uA/50) which is ~4.4MOhm 2019-10-21T04:28:49 < BrainDamage> and 100k is much less than that 2019-10-21T04:28:55 < bitmask> I'm not used to regular transistors, i'll have to look at the datasheets :/ 2019-10-21T04:29:05 < bitmask> alright thanks 2019-10-21T04:53:40 < Ultrasauce> bitmask: play SOMA 2019-10-21T05:17:30 -!- con3 [~kvirc@154.119.40.228] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-21T05:45:48 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-21T05:49:12 < dongs> wait what the fuck are you even doing there? 2019-10-21T06:01:36 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 265 seconds] 2019-10-21T06:18:28 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-21T06:23:50 -!- fc5dc9d4_ [~quassel@p5B0816AA.dip0.t-ipconnect.de] has joined ##stm32 2019-10-21T06:27:14 -!- fc5dc9d4 [~quassel@p5B3A8777.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-21T06:42:09 < dongs> https://www.analog.com/en/products/adf4350.html#product-overview what would a nigga do with this 2019-10-21T06:43:17 < Cracki> evboard guide titled "Evaluation Board for Fractional-N/Integer-N PLL Frequency Synthesizer " 2019-10-21T06:43:36 < BrainDamage> it's just an oscillator 2019-10-21T06:44:01 < dongs> what do you do with oen tho 2019-10-21T06:44:20 < Cracki> generate something you can modulate onto? 2019-10-21T06:44:35 < BrainDamage> you drive a mixer, which in turn shift in frequency a signal 2019-10-21T06:44:53 < Cracki> datashit lists applications: wireless ifnrastructure, test equipment, ... clock generation 2019-10-21T06:45:00 < dongs> could this be used to shift say 3ghz shit into 1ghz band? 2019-10-21T06:45:01 < BrainDamage> modulation is done in baseband, those high freq oscillators are just for shifting 2019-10-21T06:45:03 < BrainDamage> yes 2019-10-21T06:45:09 < BrainDamage> but you'll also need a mixer 2019-10-21T06:45:18 < BrainDamage> and a filter 2019-10-21T06:45:32 < Cracki> I'm sure their ref designs use suitable mixers by them :P 2019-10-21T06:45:52 < Cracki> adl6375 in the first one 2019-10-21T06:46:06 < Cracki> *5375 2019-10-21T06:47:40 < BrainDamage> eg you output 4 GHz into the mixer, then (4-3)GHz gets folded into 1GHz but also (5-4), the filter is to exclude the 5 2019-10-21T06:58:36 < dongs> hmm i see 2019-10-21T06:58:51 < dongs> the board its on, i suspect they're doing this exct thing 2019-10-21T06:58:59 < dongs> shifting 3+ghz into lower band 2019-10-21T06:59:07 < dongs> but the shitty pics i have only show the ADF 2019-10-21T06:59:34 < dongs> https://i.imgur.com/bQRwYMx.jpg 2019-10-21T07:00:29 < dongs> https://i.imgur.com/RiUazVp.png 2019-10-21T07:33:22 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-21T07:36:17 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-21T07:36:17 -!- day__ is now known as day 2019-10-21T07:44:09 < aandrew> dongs: frequency synthesizer? could be front end for basic digital radio 2019-10-21T07:44:45 < aandrew> yeah it's basically an I/Q LO 2019-10-21T07:45:35 < aandrew> all those Rs and Ls that look like a castle wall look like a (really fucking long) filter network 2019-10-21T07:54:23 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-21T09:31:25 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-21T09:43:18 -!- jly [uid355225@gateway/web/irccloud.com/x-crrvclxymiqphiyi] has joined ##stm32 2019-10-21T09:44:39 < jly> ohayoo steffan sun 2019-10-21T09:48:00 < dongs> kys jly 2019-10-21T09:48:16 < jly> perfect! 2019-10-21T09:48:32 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-21T09:48:40 < jly> never a dull moment on irc 2019-10-21T09:49:50 < dongs> dont make me send you cock pics 2019-10-21T09:50:58 < doomba> we need to eat the babies. there's too much stm32! 2019-10-21T09:52:29 < jly> things have changed here.... ... 2019-10-21T09:56:50 < dongs> jly are you looking for a safespace 2019-10-21T09:57:00 < jly> n 2019-10-21T09:57:10 < jly> isn't that some sjw shit? 2019-10-21T09:57:18 < dongs> i think mr. hackkitten ran off to one of those 2019-10-21T09:57:27 < jly> hah you don't say 2019-10-21T09:57:32 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-21T09:58:17 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-21T10:05:18 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-21T10:07:38 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-21T10:14:34 < Steffanx> I dont speak that language, sorry jly 2019-10-21T10:32:07 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-21T10:35:22 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-21T10:53:14 < jly> who that 2019-10-21T10:56:10 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Quit: The Lounge - https://thelounge.github.io] 2019-10-21T10:58:26 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-21T10:59:08 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-21T11:10:30 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-21T12:44:42 < karlp> aandrew: ok, rechecked my grafana setup at work. I had a graphite-statsd docker container already, too do some statsd testing and see if graphite itself was enough. so then adding grafan was just "one more" on top of that. 2019-10-21T12:44:54 < karlp> but yeah, graphite is the datastore there 2019-10-21T12:46:32 < jly> ty kp 2019-10-21T12:47:41 < jly> are you a fan of 99 2019-10-21T12:50:12 < karlp> 99? 2019-10-21T12:56:29 < jly> https://en.wikipedia.org/wiki/Barbara_Feldon 2019-10-21T12:56:30 < jly> idk 2019-10-21T12:56:32 < jly> lol 2019-10-21T13:15:31 < Steffanx> Get Smart, Haohmaru 2019-10-21T13:16:38 < karlp> yes, I did love 99 when I was a kid 2019-10-21T13:25:08 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-21T13:26:22 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-21T13:26:28 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-21T13:34:09 < karlp> neat: https://www.eenewspower.com/news/formica-adds-wireless-charging-its-laminates?news_id=122294 2019-10-21T13:34:43 < dongs> SiC FET 2019-10-21T13:37:17 < dongs> uhh 2019-10-21T13:37:23 < dongs> Zilog got bought out by littelfuse> 2019-10-21T13:39:46 < dongs> https://www.businesswire.com/news/home/20151119005108/en/Zilog-Announces-New-ZNEO32%21-32-bit-Microcontroller-Family 2019-10-21T13:39:49 < dongs> has anyone used this 2019-10-21T13:41:34 < Steffanx> Looks like ZANO too much. 2019-10-21T13:41:55 < karlp> cheapish on digikey, btu yeah, never heard of it. https://www.digikey.com/product-detail/en/Z32F06423AKE/269-5040-ND/7035315 2019-10-21T13:42:06 < dongs> i cant even find the shit on zilog site 2019-10-21T13:42:11 < dongs> which is slow as fucking balls 2019-10-21T13:42:16 < dongs> there's 16bit version of Zneo on there 2019-10-21T13:42:18 < dongs> not even 32 one 2019-10-21T13:42:43 < dongs> huh that naming scheme looks suprrisingly liek ST's 2019-10-21T13:55:00 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-21T13:58:50 < con3> guys..quick question, if i want to add an internal cut-out or slot on the pcb. i just add a track designating the cutout area on the mechanical 1 layer , same layer as the board cutout? 2019-10-21T13:59:00 < dongs> uh 2019-10-21T13:59:09 < dongs> you might start by mentioning what pcb cad 2019-10-21T13:59:16 < con3> dongs: sorry 2019-10-21T13:59:20 < con3> altium 2019-10-21T13:59:32 < con3> one sec will screenshot 2019-10-21T13:59:40 < dongs> add a thing on keepout, and turn it into board cutout as well for visual 2019-10-21T14:00:24 < con3> Haohmaru: :p 2019-10-21T14:00:36 < dongs> kikecad prolly doesnt suppor board cutouts 2019-10-21T14:00:58 < con3> dongs: thank you ! will do that quick, thought i just add it on the same layer as the cutout :) 2019-10-21T14:01:08 < dongs> anyway draw the shit you wanna cutout on kee-out layer. select it all. T-V-E. goto properties and turn into board cutout instead of copper. done. 2019-10-21T14:04:18 < jpa-> con3: i've always just drawn cutouts on the outline layer, works fine atleast with jlcpcb 2019-10-21T14:04:38 < con3> jpa-: the one im using doesnt seem to recognise it like that 2019-10-21T14:04:54 < dongs> what 2019-10-21T14:05:11 < jpa-> so it detects outline, but not an internal slot that you've drawn exactly like you would draw outline? 2019-10-21T14:05:19 < con3> dongs: i dont see any option to change it from copper to cutout in the properties? 2019-10-21T14:05:20 < dongs> yea that makes no sense 2019-10-21T14:05:23 < dongs> unless you had copper and shit on it 2019-10-21T14:05:43 < jpa-> yeah, adding to keepout layer makes sense if your ground pour would otherwise extend to it 2019-10-21T14:05:47 < dongs> con3: did you convert the shit to region 2019-10-21T14:05:55 < dongs> wiht t-v-e 2019-10-21T14:05:56 < dongs> as i told you 2019-10-21T14:06:02 < con3> welp fuck 2019-10-21T14:06:08 < con3> one sec 2019-10-21T14:06:10 < con3> sorry 2019-10-21T14:06:22 < dongs> clicky one line, tab to select whole cutout, tve 2019-10-21T14:06:50 * con3 hugs dongs 2019-10-21T14:06:51 < dongs> fucking lazy latvians still havent shipped my wifi gear 2019-10-21T14:07:09 < dongs> o shit why is my order "pending" 2019-10-21T14:07:13 * con3 learnt something new 2019-10-21T14:07:20 * con3 writes down tve 2019-10-21T14:07:37 < dongs> there's a whole bunch of shit when you press t-v- that might be interesting 2019-10-21T14:10:43 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-21T14:11:03 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-21T14:11:54 < karlp> Haohmaru: why do you have such neat pcbs and and nice careful silk, and then just splat all this through hole stuff on it? 2019-10-21T14:16:21 < dongs> le autismo 2019-10-21T14:17:34 < karlp> love the socketed dip8, is that for "ermerhgerd, my burrbrown opamps!" audiophool shit I presume? 2019-10-21T14:18:01 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-21T14:18:06 < dongs> the one in that photo 2019-10-21T14:18:07 < karlp> on your board, 2019-10-21T14:18:12 < dongs> which is dip8 to sop8 adapter 2019-10-21T14:18:24 < dongs> the fuck 2019-10-21T14:18:27 * karlp wats 2019-10-21T14:18:40 < dongs> lmao 2019-10-21T14:18:51 < karlp> totally abiding by those switcher current loops... 2019-10-21T14:19:16 < karlp> it even comes in dip8? 2019-10-21T14:19:19 < karlp> oh, it will work. 2019-10-21T14:19:35 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-21T14:19:58 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-21T14:21:32 < dongs> livechatting this fucking latvian shit 2019-10-21T14:21:36 < dongs> i hope they're not trashing my order 2019-10-21T14:30:38 < dongs> oh for fucks sake 2019-10-21T14:30:43 < dongs> they're out of USA plug version of the shit 2019-10-21T14:30:52 < dongs> god damn fucking eurofags 2019-10-21T14:32:59 < dongs> allright order fixed 2019-10-21T14:43:21 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-21T14:50:40 < veverak> lol 2019-10-21T14:51:52 < ub|k> thanks Steffanx... I'll give it a try later. 2019-10-21T15:04:40 -!- tomeaton17_ [tomeaton17@unaffiliated/tomeaton17] has joined ##stm32 2019-10-21T15:05:05 -!- tomeaton17 [tomeaton17@unaffiliated/tomeaton17] has quit [Ping timeout: 276 seconds] 2019-10-21T15:06:41 < qyx> 33063 in 2019? 2019-10-21T15:12:53 -!- jly [uid355225@gateway/web/irccloud.com/x-crrvclxymiqphiyi] has quit [Quit: Connection closed for inactivity] 2019-10-21T15:56:52 < zyp> dongs, eurodk? 2019-10-21T15:57:08 < zyp> buying more mikrodiks? 2019-10-21T16:10:16 -!- con3 [~kvirc@154.119.40.228] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-21T16:14:49 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-21T16:20:48 < dongs> zyp: yah bought the audience 2019-10-21T16:20:57 < dongs> and a replacement for one of my SXT's 2019-10-21T16:22:00 <@englishman> just one? 2019-10-21T16:22:13 < dongs> yes 2019-10-21T16:22:29 < zyp> heh 2019-10-21T16:22:33 < dongs> one was kinda glitching, died suddenly and failed to boot a fwe times until i left it unpowered overnight 2019-10-21T16:22:42 < dongs> its working now but im gonna replace itr 2019-10-21T16:23:07 < dongs> and audience is replacemetn for my ancietn tpdink 2019-10-21T16:23:55 < zyp> the audience doesn't look all that useful to me 2019-10-21T16:24:14 < dongs> i have lots of shitty clients and a large space. so it kinda does to me 2019-10-21T16:29:10 < zyp> I prefer ceiling mounted APs 2019-10-21T16:29:21 < zyp> rather than having a box standing around 2019-10-21T16:30:26 < dongs> how do you get thernet and power to them? 2019-10-21T16:30:31 < dongs> well, besides poe but still ethernet 2019-10-21T16:31:08 < zyp> https://bin.jvnv.net/file/TWI5D.jpg <- like that 2019-10-21T16:32:11 < zyp> https://bin.jvnv.net/file/v3vXg.jpg 2019-10-21T16:35:12 < dongs> hackerman 2019-10-21T16:36:31 < zyp> I literally have at least one ethernet point in every single room of this house except the two bathrooms 2019-10-21T16:36:44 < zyp> including a ceiling point on each floor 2019-10-21T16:37:46 < zyp> originally expected to only need one of them, but ran to both floors just in case, turns out there's metal foil in the underfloor blocking much of the radio signals between the floors 2019-10-21T16:39:26 < doomba> same 2019-10-21T16:39:47 < doomba> but this huose is over 100 years old 2019-10-21T16:39:57 < doomba> i'm not credit floating for a new one 2019-10-21T16:40:28 < zyp> https://bin.jvnv.net/file/vkfgq.jpg <- I just need a new switch now, the one I've got is a bit crowded 2019-10-21T16:40:52 < doomba> +1 RouterOS 2019-10-21T16:41:15 < zyp> nah, fuck RouterOS 2019-10-21T16:41:29 < doomba> cmon TheDude. what's not to love about it? 2019-10-21T16:41:53 < zyp> the configuration shit? 2019-10-21T16:41:59 < dongs> winbox is awesome 2019-10-21T16:42:11 < karlp> "credit floating" ? 2019-10-21T16:42:48 < doomba> yea credit floating is how we muricans buy anything in this world. 2019-10-21T16:42:52 < zyp> but I mean, I don't mind RouterOS on the switch because it doesn't need much configuration 2019-10-21T16:43:19 < zyp> but I prefer ubnt for routing 2019-10-21T16:46:58 < doomba> https://mobile.twitter.com/hexdefined/status/1185864801261477891 2019-10-21T17:18:59 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-21T17:22:32 < Steffanx> Where did your fibre stuff go zyp? 2019-10-21T17:22:42 < dongs> lul www.real.com got malware'd 2019-10-21T17:22:51 < dongs> (the home of realplayer 2019-10-21T17:23:30 <@englishman> lol 2019-10-21T17:24:17 < dongs> main page is ded. if you goto /player then it looks ok 2019-10-21T17:24:27 -!- brdb [~basdb@2601:18c:8500:7f5b::9bb] has joined ##stm32 2019-10-21T17:34:47 < aandrew> karlp: I didn't know graphite did any of its own datastore 2019-10-21T17:35:31 < aandrew> zilog? heh z80 comes to 32-bit land 2019-10-21T17:36:12 < aandrew> I wonder what their business plan is... MCUs are not cheap to design and tapeout yet they created their own in the land of arm and riscv and mips 2019-10-21T17:36:29 < dongs> it is cortex m3 anyway 2019-10-21T17:36:33 < aandrew> oh the zilog chip is cortex m3 2019-10-21T17:36:36 < dongs> ya 2019-10-21T17:36:49 < dongs> the only peple crazy enough to do custom arch these days are renesas? 2019-10-21T17:36:50 < dongs> wiht their RX shit 2019-10-21T17:36:56 < dongs> of course nobody actually uses that shit. 2019-10-21T17:37:02 < dongs> (custom 32bit 2019-10-21T17:39:36 < aandrew> zyp> I literally have at least one ethernet point in every single room of this house except the two bathrooms 2019-10-21T17:39:46 < aandrew> wtf dude I have like 4 ethernet drops in my bathrooms, two at the shitter 2019-10-21T17:40:53 < aandrew> dongs: doesn't automotive have a big stiffy for RX? 2019-10-21T17:41:09 < dongs> does it? 2019-10-21T17:41:23 < dongs> ive seen it in jap products 2019-10-21T17:41:34 < dongs> i think these days even automotive shit is just using cortex-R 2019-10-21T17:41:41 < zyp> aandrew, if I built a house like 10-15 years ago, I probably would have considered that 2019-10-21T17:41:55 < karlp> renesas also just annoucned cortex-m parts... 2019-10-21T17:42:35 < aandrew> does anyone actually use 32-bit mode ARM insns anymore? 2019-10-21T17:42:39 < aandrew> seems everything is thumb 2019-10-21T17:43:23 < karlp> it's still good for testing that you've fucked up your linker and toolchain lib paths, when you try and execute it on cortex-m... 2019-10-21T17:43:34 < karlp> no idea on -a though, 2019-10-21T17:43:54 < zyp> my impression is that -a mostly builds in arm mode 2019-10-21T17:43:55 < karlp> I'm sure there's binary arm libs that are still being used on cortex-a from the old days 2019-10-21T17:44:00 < zyp> not sure what my -r workstuff do 2019-10-21T17:44:31 < aandrew> karlp: hah 2019-10-21T17:45:32 < karlp> aandrew: this is the one I ran, and it uses graphite's whisper piece. https://github.com/graphite-project/docker-graphite-statsd/blob/master/Dockerfile 2019-10-21T17:47:29 <@englishman> are modern bidets ethernet or wifi 2019-10-21T17:47:37 < doomba> bluetooth 2019-10-21T17:47:53 < aandrew> crapflooding for reals 2019-10-21T17:48:18 < dongs> dont ddos attack my commode 2019-10-21T17:49:51 < doomba> https://www.amazon.com/Elongated-Stainless-Functions-Nightlight-Oscillation/dp/B078XLBZZZ 2019-10-21T17:50:24 < doomba> because the buttons on the side are too far when you're 500lbs 2019-10-21T17:51:02 < dongs> remote control liek that is pretty standard 2019-10-21T17:53:30 <@englishman> only Americans remain behind in asshole cleaning technology 2019-10-21T17:53:59 <@englishman> scraping dead trees against their poor sphincters and getting hemherroids n shit 2019-10-21T17:54:25 -!- kakimir-pro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-21T17:54:40 < doomba> sheeit not me 2019-10-21T17:55:03 < doomba> i have a butt gun 2019-10-21T17:55:33 < doomba> once you try it, you never go back to TP 2019-10-21T17:55:37 < mawk> Advance your hygiene with Alpha Bidet. 2019-10-21T17:57:19 -!- maunix [~Mauricio@181.31.151.95] has joined ##stm32 2019-10-21T18:03:27 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-21T18:05:29 < Steffanx> Dutchland is scraping too, englishman 2019-10-21T18:05:36 < Steffanx> Better not go to dutchland mawk 2019-10-21T18:05:47 < Steffanx> CANCEL the contract 2019-10-21T18:06:19 <@englishman> I thought the dutch save all their turds for use in scat porn 2019-10-21T18:07:48 < Steffanx> Idk about that 2019-10-21T18:08:25 < Steffanx> But it's not most porn people watch is not made in their own country 2019-10-21T18:08:35 < Steffanx> With yankeeland being an exception 2019-10-21T18:09:06 < Steffanx> Better get antto out of his box 2019-10-21T18:11:40 < kakimir-pro> can I pass parameters to task function from task creation? 2019-10-21T18:12:26 < kakimir-pro> true 2019-10-21T18:13:11 < karlp> kakimir-pro: dude, the parameter after stack size and before priority is your user blob 2019-10-21T18:13:22 < kakimir-pro> pvParameters* 2019-10-21T18:13:25 < karlp> you can also use the thread local storage array if you like 2019-10-21T18:13:51 < kakimir-pro> static stack? 2019-10-21T18:14:02 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-21T18:14:27 < karlp> in my freertos world, I'm having bizarresness where main gets calleda gain. I stipped teh bootloader and iwdg out, to try trim the moving parts. just... "ok, we're in main again" 2019-10-21T18:14:30 < karlp> fucing weird 2019-10-21T18:14:46 < dongs> wtf 2019-10-21T18:14:53 < dongs> how does that work? 2019-10-21T18:14:58 < dongs> you think it resets or wat? 2019-10-21T18:15:08 < dongs> do you have a while(1) at end of main? 2019-10-21T18:15:22 < kakimir-pro> so does it reset? 2019-10-21T18:16:08 < karlp> yeah, just can't work out what's resetting it. 2019-10-21T18:17:46 < karlp> hrm, added a whiel loop at end of main, doesn't hit there. 2019-10-21T18:26:46 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-21T18:27:32 < karlp> ok. all my own bullshit 2019-10-21T18:27:34 < karlp> ffs. 2019-10-21T18:27:59 < karlp> this proto device is being detected as an old legacy device, and because it doesn't have any readings, it's triggering some "try and reboot it" recovery. 2019-10-21T18:28:12 * karlp smacks himself int eh face and gets more coffee 2019-10-21T18:29:50 < dongs> nice 2019-10-21T18:30:31 < karlp> well, it's doing what it was told to do. so.. that's good I guess .) 2019-10-21T18:30:53 < karlp> and was running things in the foreground to diagnose it so didn't see the logging that told me this :) 2019-10-21T18:34:11 < jpa-> now i'm too late to recommend you check the reset reason register :/ 2019-10-21T18:34:36 < karlp> well, I went through that :) 2019-10-21T18:34:48 < karlp> that's where I got it from, was "sft_reset" :) 2019-10-21T18:43:01 < karlp> gave up on trying to use freertos timers too. 2019-10-21T18:43:29 < karlp> between having a few blink patterns, and some priority levels for what states should be ont he led, trying to turn on / off the timers properly was code soup 2019-10-21T18:43:52 < karlp> ended up keeping a "busy" loop just with 10ms period instead. 2019-10-21T18:45:13 < jpa-> that's the pro way, having systick timer trigger interrupt that counts down on software timer that then proceeds to count down on a variable 2019-10-21T18:45:44 < karlp> it feels kidna gross, 2019-10-21T18:45:48 < jpa-> for some reason that is the simplest solution :) 2019-10-21T18:46:12 < karlp> it's j us that now i've got timer tick variables in my global state to ahndle the expiries and shit 2019-10-21T18:46:52 < karlp> still pissed that freertos ticks rollover 32bit with nowway out without patching, or inserting a complete hook on every tick. 2019-10-21T18:47:02 < karlp> (and hook on tick does wonders on tickless idle) 2019-10-21T18:48:08 < karlp> it seems like I _could_ use rtos primitives to manage it all, but doing it with just a low priority looping and timestamping task just _works_ and it's way less code. 2019-10-21T18:48:36 < jpa-> i wish more rtoses went with 64-bit clock cycle counters and true tickless, i.e. using hardware timer to schedule wakeup for even very short delays 2019-10-21T18:48:37 < karlp> so, that's the pro way then, use the rtos when it helps, ignore it when it's not :) 2019-10-21T18:49:48 < zyp> so far the only thing I use freertos timers for in workstuff is schedule a system reset to happen a few milliseconds later when switching between bootloader mode and runtime mode so response to the reset command have time to go out first :) 2019-10-21T18:49:57 < jpa-> i guess fault is partiall on ARM on not making systick counter 64-bit and with "interrupt when this count is exceeded" register, instead of the current half-ass way that doesn't work well with changing wakeup interval 2019-10-21T18:50:15 < zyp> jpa-, yea 2019-10-21T18:50:58 < karlp> zyp: yeah, I'm down from ~6 freertos timers to _maybe_ one. 2019-10-21T18:51:13 < karlp> and code is back to working and short. 2019-10-21T18:51:20 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-21T18:51:36 < karlp> 24bit systick always seemed like a wat moment. 2019-10-21T18:53:18 < zyp> for my workstuff, systick is not very relevant 2019-10-21T18:54:02 < zyp> all the important stuff is executing faster than systick 2019-10-21T18:54:07 < karlp> eh, I mean, do the subtractions properly, and it's not real either, just can't get "uptime" beyond 32bit rollover 2019-10-21T18:54:30 < karlp> there's an internal var in freertos that carries the rollovers, but it's not available 2019-10-21T18:56:04 < zyp> I've got a master task, a slave task and a network task, master runs 1 kHz, slave runs 5 kHz, network runs 1 kHz cycles with a bunch of back and forth per cycle 2019-10-21T18:56:21 < zyp> actually, I have too much latency at the moment for 1 kHz cycles, so I'm running 500Hz cycles 2019-10-21T18:58:08 < karlp> this is with your own timer and task stuff? 2019-10-21T18:58:20 < karlp> or is this freertos? 2019-10-21T18:58:25 < zyp> every device contains a switch that adds latency, so in default mode cycle time scales like O(n^2) 2019-10-21T18:58:55 < zyp> switches do have a repeater mode, but it was misbehaving last I tried, and I haven't had time to work out how to do that properly yet 2019-10-21T18:59:57 < zyp> this is freertos, but tasks are scheduled by timer interrupts running task notifies 2019-10-21T19:00:04 < zyp> also network interrupts running task notifies 2019-10-21T19:10:08 < karlp> ah, right. yeah, just blocking on xTaskNotifyWait()/Take() 2019-10-21T19:11:30 < karlp> how is that better/preferred than just vWaitUntil() ? 2019-10-21T19:12:26 < karlp> is it just because your wait until resolution is only in OS ticks, and you want them to be triggered more often? like 5khz for your slave? 2019-10-21T19:12:43 < bitmask> hmm, so I'm looking at the 3904 and 3906 transistors and its saying the emitter-base breakdown voltage is 6.0V, I didn't realize that was so low, isn't that a problem driving it from an avr pin? do I need a voltage divider? 2019-10-21T19:12:47 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-21T19:15:23 < jpa-> bitmask: that's in the opposite direction 2019-10-21T19:15:44 < bitmask> ohhhh 2019-10-21T19:16:00 < jpa-> bitmask: usually for BJT transistors, they'll require a series resistor and base-emitter voltage will not rise above 0.7V~1.0V 2019-10-21T19:17:24 < jpa-> (BJTs are current-controlled, not voltage-controlled) 2019-10-21T19:17:24 < bitmask> I dont know why I'm having such a hard time with BJTs, woulda thunk id have used em by now 2019-10-21T19:17:28 < bitmask> yea 2019-10-21T19:23:51 -!- friendofafriend [~ian@75.182.67.149] has joined ##stm32 2019-10-21T19:28:06 < BrainDamage> I don't know if it helps you, but for me it helps to view the diodes inside: https://en.wikipedia.org/wiki/Bipolar_junction_transistor#/media/File:Approximated_Ebers_Moll.svg 2019-10-21T19:28:26 < BrainDamage> base-emitter diode has to be on, and collector-emitter off 2019-10-21T19:30:28 < jpa-> *collector-base 2019-10-21T19:30:45 < BrainDamage> sorry, collector-base, yes 2019-10-21T19:31:26 < Thorn> lcsc webiste has got a new design 2019-10-21T19:31:34 < Thorn> still no STM32L476RCT6 in stock though :/ 2019-10-21T19:49:55 < kakimir-pro> fellow pros: https://paste.ee/p/xtNHT why does my script do 0x4000 fill in line 39? 2019-10-21T19:52:03 < jpa-> i would guess that maybe . inside a scope is relative to the start of the scope, and you are assigning 0x4000 to that relative address 2019-10-21T19:52:35 < kakimir-pro> oh 2019-10-21T19:52:58 < jpa-> why not just define FLASH_P and FLASH_NP for the different sections? 2019-10-21T19:53:28 < jpa-> keeping them in one memory area makes sense if you want to automatically assign the addresses, but if you don't, make them separate regions 2019-10-21T19:54:25 < kakimir-pro> p? np? 2019-10-21T19:55:03 < kakimir-pro> I just want datasection to stay clear from 16k boundary of privileged functions 2019-10-21T19:56:15 < kakimir-pro> hey jpa I understood what I need to do 2019-10-21T19:56:55 < kakimir-pro> move line 18 to end of privileged functions? 2019-10-21T19:57:26 < jpa-> dunno 2019-10-21T19:58:21 < jpa-> i'd just do different regions in MEMORY {}, and > FLASH_PRIVILEGED and > FLASH_NONPRIVILEGED for the different function types 2019-10-21T19:59:09 < jpa-> but if you don't really need to fix the size of the area for privileged functions, you could just ALIGN() to the size of the page you can assign in MPU 2019-10-21T19:59:46 < kakimir-pro> _Privileged_Functions_Region_Size = 0x4000 2019-10-21T20:01:33 < kakimir-pro> I just understood how address pointer thing works 2019-10-21T20:01:57 < kakimir-pro> indeed my binary has weight loss of over 10kilos now 2019-10-21T20:04:24 < kakimir-pro> every time there is section the address counter starts from 0 2019-10-21T20:04:28 < kakimir-pro> address pointer 2019-10-21T20:05:04 < kakimir-pro> so I made address pointer go all way to 0x4000 in privileged_functions 2019-10-21T20:05:10 < kakimir-pro> not after it 2019-10-21T20:07:16 < kakimir-pro> https://paste.ee/p/Nwbgb sweet 2019-10-21T20:09:41 < kakimir-pro> I was going at 57kilos and now I'm going at 44kilos 2019-10-21T20:12:18 < antto> such much overweight, ur LPC is gonna have a belly 2019-10-21T20:15:21 < kakimir-pro> I have 20kilos to implement the functionality 2019-10-21T20:15:48 < kakimir-pro> it's more than enough as I have all strings I need already included 2019-10-21T20:15:56 < antto> eLePhantController 2019-10-21T20:20:54 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-21T20:33:20 < aandrew> karlp: heh I had a similar bug where I"m writing MAC+IP info to an FPGA when ethernet comes up, but I have a bug in the HDL which my current workaround is to reset the FPGA every 5s until packets get through 2019-10-21T20:33:47 < aandrew> so I get an IP, update the FPGA, but don't see traffic so reset the FPGA.. then when the workaround kicks right and traffic flows, my src IP+MAC are zeroes. wtf.? heh 2019-10-21T20:36:24 < kakimir-pro> pro shared nice freertos trick 2019-10-21T20:36:37 < kakimir-pro> turn of led in idle hook before WFI 2019-10-21T20:36:41 < kakimir-pro> off* 2019-10-21T20:36:58 < kakimir-pro> turn it on in vTaskSwitchContextHook 2019-10-21T20:37:09 < kakimir-pro> it's an activity led 2019-10-21T20:38:33 < friendofafriend> Does anyone know a good source for a cheap ST-Link v2 that contains an STM32? I'd like to build one of those gnuk things, and the cheap knockoff programmer from eBay was really a TTL adapter. 2019-10-21T20:38:57 < specing> aren't all stlinks stm32? 2019-10-21T20:39:34 < kakimir-pro> aren't all ST-Links chaep as soap? 2019-10-21T20:40:38 < friendofafriend> There's a lot of eBay listings that show pictures of ST-Link v2s, and send you a USB TTL adapter instead. 2019-10-21T20:41:11 < kakimir-pro> buy a real one 2019-10-21T20:41:22 < kakimir-pro> it cannot be more than like 20bucks 2019-10-21T20:41:39 <@englishman> nucleo board with stlink on breakoff part 2019-10-21T20:41:42 <@englishman> like $10-15 2019-10-21T20:41:47 < kakimir-pro> even better 2019-10-21T20:41:52 < specing> friendofafriend: really? THey send you a TTL adapter? 2019-10-21T20:41:57 <@englishman> lol 2019-10-21T20:41:57 < specing> Can you even program stms that way? 2019-10-21T20:42:02 < friendofafriend> specing: Yep. https://www.ebay.com/itm/Programming-Unit-mini-STM8-STM32-Emulator-Downloader-M89-Pip-LU/273846371936 2019-10-21T20:42:02 < specing> dispute 2019-10-21T20:42:05 < specing> dispute 2019-10-21T20:42:07 < specing> dispute 2019-10-21T20:42:30 <@englishman> one of the pix says usb-ttl 2019-10-21T20:42:46 <@englishman> actually 3 do 2019-10-21T20:42:52 < specing> not one 2019-10-21T20:42:54 < specing> all of them 2019-10-21T20:42:56 < zyp> doesn't say st-link anywhere 2019-10-21T20:42:56 < specing> lol 2019-10-21T20:42:59 < specing> all but one 2019-10-21T20:43:04 < friendofafriend> Yes, it does. And it's always nice to have another USB TTL adapter. The listing picture does. 2019-10-21T20:43:15 <@englishman> supports st-link utility lol 2019-10-21T20:43:20 <@englishman> yeah that's a refundo 2019-10-21T20:43:34 < specing> $2 is quite expensive for an usb-ttl 2019-10-21T20:43:40 < friendofafriend> Are you mostly getting the programmer from Ali? 2019-10-21T20:43:59 <@englishman> no, all of mine have come from digikey. 2019-10-21T20:44:19 <@englishman> that's not true i got some nucleos from arrow. 2019-10-21T20:44:53 <@englishman> they are like $20 why would you waste all your money and most importantly your time on inferior chinese garbage. 2019-10-21T20:45:17 < antto> do then ship stuff by tieing a bag onto an arrow and shooting it with a bow it to yer door? 2019-10-21T20:45:21 < antto> * they 2019-10-21T20:45:21 < friendofafriend> Because I'd like to repurpose them, and $20 is about the same as a commercial offering. 2019-10-21T20:45:37 < antto> wow, such much typos 2019-10-21T20:45:40 * antto hides 2019-10-21T20:45:45 <@englishman> if you are working on such tight margins maybe it's time to make your own pcb 2019-10-21T20:46:02 < aandrew> yeah stlinkv2 on digikey is cheap 2019-10-21T20:46:14 < aandrew> or any nucleo board like englishman says 2019-10-21T20:46:41 < specing> lol englishman , what do you think we all do 2019-10-21T20:46:54 < friendofafriend> Thanks, I'll keep that in mind! :) I'm on a pocket-change-or-nothing-at-all budget. 2019-10-21T20:46:54 < specing> but waste time and money on inferior chinese garbage 2019-10-21T20:47:14 <@englishman> you? you just argue about why some obsolete programming language noone has used since 1979 still has relevance 2019-10-21T20:47:33 < antto> aww 2019-10-21T20:47:39 < specing> lol, you couldn't be more wrong if you tried 2019-10-21T20:54:06 < maunix> Hi all, I am considering purchasing the STM32MP157A kit. anyone with experience with it? Toughts? Are they ok? do they have some issue like overheating or some port that does not work? thanks in advance! 2019-10-21T20:54:33 < antto> a port wouldn't work?! o_O 2019-10-21T20:54:45 < antto> wut.. wot.. wuh.. eh? 2019-10-21T20:56:54 < maunix> antto: my question was general, if it has a bug it can be of many things. A port could be damaged if there's a problem with the PCB 2019-10-21T20:57:12 < BrainDamage> it's nearly impossible to get a microcontroller to overheat, however, I cannot say as much for the A7 2019-10-21T20:57:24 < Thorn> my issue with it is it's $100+ 2019-10-21T20:57:37 < kakimir-pro> when making a triac controller 2019-10-21T20:57:38 < antto> how would it help if someone told you that his stm32wutever is "fine" ? 2019-10-21T20:57:55 < kakimir-pro> is it sufficient to know only the last period of mains phase? 2019-10-21T20:58:09 < kakimir-pro> and start from zero crossing 2019-10-21T20:58:26 < antto> kakimir-pro r u multitasking on 73 different projects? ;P~ 2019-10-21T20:58:29 < jpa-> oh st is calling cortex-a's "STM32MP" now 2019-10-21T20:58:30 < BrainDamage> depends how inductive your load is, you need to know when current goes to 0, and when voltage goes to 0 2019-10-21T20:58:54 < kakimir-pro> or would you run like a virtual phase that is synced to actual phase 2019-10-21T20:58:55 < BrainDamage> the first can be omitted if the load has little inductance 2019-10-21T20:59:30 < kakimir-pro> interestings 2019-10-21T20:59:48 < BrainDamage> assuming you're doing phase chopping, your control is to delay the SCR firing so that the load receives only x% of the waveform 2019-10-21T20:59:55 < maunix> antto:i've read that the first raspberry pi 4 had overheating problems. i am not trying to be 100% precise in any statement.. just an input from someone that test them. 2019-10-21T21:00:00 < BrainDamage> delay from the start of the wave 2019-10-21T21:00:20 < BrainDamage> however, once fired, the SCR will remain on until the current goes to 0 2019-10-21T21:00:25 < antto> uh.. okay 2019-10-21T21:00:53 < BrainDamage> if there's no inductive component, voltage 0 means current 0, otherwise not necessarily and you'll need to measure both 2019-10-21T21:01:00 < steve> Hello, I'm confused about the DMA. If I say transfer size is N, does it transfer N bytes, or N * MSIZE bytes? 2019-10-21T21:01:03 < antto> but still, IMO STMicro is not of the same magnitude as whoever made the crapberry pi 2019-10-21T21:01:05 < kakimir-pro> heater elements have inductance 2019-10-21T21:01:09 < BrainDamage> yup 2019-10-21T21:01:37 < kakimir-pro> would virtual phase be anything useful? 2019-10-21T21:02:12 < BrainDamage> no idea what's virtual phase 2019-10-21T21:02:21 < jpa-> steve: N*MSIZE 2019-10-21T21:04:04 < steve> thanks jpa-, I figured since all documentation talks about "size" but no mention of bytes. Super confusing that the STM HAL_SAI_Transmit_DMA takes a uint8_t * pData, and a uint16_t size, but the size is not the number of uint8_t's, it's the number of whatever the DMA was configred at 2019-10-21T21:04:07 < jpa-> steve: to be more accurate, it transfers N*PSIZE on peripheral side and N*MSIZE on the memory side, throwing away or duplicating bits if MSIZE != PSIZE 2019-10-21T21:04:19 < kakimir-pro> BrainDamage: me neather 2019-10-21T21:07:21 < jpa-> hmm my previous line might be false, atleast for STM32F4; apparently it packs / unpacks the transfers if sizes differ 2019-10-21T21:12:14 < steve> packs? Like packs 2 MSIZE's into a size PSIZE, if say PSIZE is 2x MSIZE? 2019-10-21T21:12:47 < jpa-> yeah, if you enable the fifo 2019-10-21T21:13:49 < kakimir-pro> BrainDamage: do you have example code? 2019-10-21T21:14:03 < jpa-> STM32F1xx (with DMAv1 that has no fifo) just zero-pads or throws away bits when PSIZE != MSIZE 2019-10-21T21:15:39 < steve> that makes more sense to me... though I wouldn't want to rely on such hardware behavior anyway 2019-10-21T21:17:14 < BrainDamage> kakimir-pro: not here, no 2019-10-21T21:17:22 < BrainDamage> first do a calibration for mains period, so you can be frequency flexible, just measure T between two zero voltage crossings which is the semiperiod 2019-10-21T21:18:18 < BrainDamage> then fire the SCR for a whole period and measure when the current goes to 0, from the delay in degrees you can get the inductance 2019-10-21T21:18:44 < BrainDamage> so you can compensate 2019-10-21T21:18:55 < jpa-> (remember to use schmitt trigger style logic for zero crossing detection) 2019-10-21T21:19:15 < kakimir-pro> I guess 2019-10-21T21:19:28 < BrainDamage> remember to constantly adjust inductance because as the device heats up, it can change 2019-10-21T21:20:02 < aandrew> I have quite a bit of experience with phase control and being able to work in the presence of noise 2019-10-21T21:20:12 < aandrew> never ever ever tie your phase change input to an interrupt line 2019-10-21T21:20:32 < aandrew> timer capture with digital filter would work nicely and not cost your phase error 2019-10-21T21:20:46 < kakimir-pro> I plan to connect it to interrupts 2019-10-21T21:20:49 < kakimir-pro> gpio 2019-10-21T21:20:51 < aandrew> bad idea 2019-10-21T21:21:00 < aandrew> when you get hash on the line you get an interrupt storm 2019-10-21T21:21:19 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-21T21:21:30 < BrainDamage> use the ADC and slap a resonant IIR filter 2019-10-21T21:21:36 < jpa-> if you do that, make sure you have enough hysteresis somewhere - either schmitt-trigger GPIO pin with suitable resistor divider, or a comparator with hysteresis in front of the GPIO 2019-10-21T21:21:54 < Thorn> kakimir-pro: you might get away with it if you disable the interrupt for some time after it fires 2019-10-21T21:21:59 < jpa-> but indeed digital filters on timers would work well also 2019-10-21T21:22:03 < aandrew> or just use timer capture 2019-10-21T21:22:05 < Thorn> (make sure to also reset all pending flags) 2019-10-21T21:22:08 < aandrew> no need for interrupts 2019-10-21T21:22:25 < jpa-> you can get interrupt from the timer anyway if you want, you can just use it for the filtering 2019-10-21T21:22:48 < aandrew> yes, if you set it up that way 2019-10-21T21:22:51 < aandrew> no real need to 2019-10-21T21:22:54 < kakimir-pro> how does timer filter it? 2019-10-21T21:23:50 < jpa-> "wait for N clock cycles, if it toggles back during that time, don't act" 2019-10-21T21:23:52 < kakimir-pro> it only generates interrupt in time window or something? 2019-10-21T21:25:49 < kakimir-pro> well 2019-10-21T21:26:07 < aandrew> there are few ways to do it but yes 2019-10-21T21:26:20 < aandrew> you can also use a timer to poll the pin and use a voting routine 2019-10-21T21:26:31 < aandrew> last 3 samples were zero, next 3 are 1? that's a rising edge 2019-10-21T21:26:35 < aandrew> note time 2019-10-21T21:28:26 < aandrew> that's exactly how I did it in an industrial phase controller 2019-10-21T21:28:42 < aandrew> thing worked in VERY noisy environments without issues, and phase delay would have been bad 2019-10-21T21:30:11 < kakimir-pro> do you use constant time to fire SCR? 2019-10-21T21:30:38 < kakimir-pro> or stop firing at zerocrossing or something? 2019-10-21T21:32:34 < kakimir-pro> are SCRs still used in new industry machinery? 2019-10-21T21:32:46 < kakimir-pro> I mean power factor is pretty bad 2019-10-21T21:33:18 < jpa-> bad compared to what? 2019-10-21T21:33:49 < kakimir-pro> some switching regulator thing 2019-10-21T21:33:59 < kakimir-pro> with PFC 2019-10-21T21:34:19 < kakimir-pro> but maybe there are no such things for higher powers 2019-10-21T21:34:52 < BrainDamage> IGBT are starting to show here and there 2019-10-21T21:34:57 < BrainDamage> but those have their own problems 2019-10-21T21:35:14 < BrainDamage> IGBT's structure has its own parasitic SCR you don't want to turn on 2019-10-21T21:35:18 < jpa-> it's quite common to use SCRs so that they only switch at zero crossing, in which case they have no impact on power factor 2019-10-21T21:35:24 < BrainDamage> or it latches up and then you can't control it 2019-10-21T21:35:44 < kakimir-pro> yes on off switching jpa 2019-10-21T21:36:36 < kakimir-pro> act as substitute for a relay 2019-10-21T21:36:45 < kakimir-pro> relays arc and shit 2019-10-21T21:37:01 < kakimir-pro> and wear or even weld closed sometimes 2019-10-21T21:37:38 < BrainDamage> there's already prepackaged solid state relays 2019-10-21T21:37:56 < BrainDamage> ... which have internally SCRs 2019-10-21T21:38:39 < kakimir-pro> yes 2019-10-21T21:38:43 < kakimir-pro> I have them from china 2019-10-21T21:38:45 < day> isnt scr just a name for a concept? 2019-10-21T21:38:58 < kakimir-pro> hello day 2019-10-21T21:39:10 < BrainDamage> no, it's a physical device PNPN 2019-10-21T21:39:10 < kakimir-pro> evenings day 2019-10-21T21:39:11 < day> hi 2019-10-21T21:39:48 < kakimir-pro> when you log back in to your ircs you have 150 highlights to your nick? 2019-10-21T21:40:04 < day> isnt that a thyristor? 2019-10-21T21:40:22 < kakimir-pro> no but 2 oh them 2019-10-21T21:40:37 < kakimir-pro> in one package 2019-10-21T21:41:07 < day> hm nvm. i was thinking "active rectification" my bad 2019-10-21T21:42:12 < kakimir-pro> wait I meant triac is dual thyristor 2019-10-21T21:42:24 < kakimir-pro> https://en.wikipedia.org/wiki/TRIAC 2019-10-21T21:47:02 < aandrew> kakimir-pro: I was using block firing 2019-10-21T21:47:09 < kakimir-pro> ? 2019-10-21T21:47:23 < aandrew> kakimir-pro: basically at the phase-on point you keep firing and stop a ms or so before the zero cross to let it commutate 2019-10-21T21:47:48 < aandrew> phase control is still used because it's cheap compared to inverters with IGBTs 2019-10-21T21:47:58 < kakimir-pro> you had blocking delay there? 2019-10-21T21:48:04 < aandrew> "blocking delay" ? 2019-10-21T21:48:23 < kakimir-pro> while(almost_there){} 2019-10-21T21:48:50 < aandrew> no 2019-10-21T21:49:16 < aandrew> just interrupt at time to shut off 2019-10-21T21:49:26 < kakimir-pro> ok 2019-10-21T21:49:55 < aandrew> I was hitting the SCR gate with a 20kHz pulse train which went through an RC network which would give the start of each 20kHz pulse a very high/sharp current spike then backed down as the cap charged 2019-10-21T21:50:04 < aandrew> helped ensure the SCR turned on 2019-10-21T21:51:03 < aandrew> this was in a very noisy env so it was critical that you didn't just tap the SCR gate, you bitchslapped it continuously to make sure it stayed on 2019-10-21T21:53:49 < kakimir-pro> why not just drive current there? 2019-10-21T21:53:53 < kakimir-pro> constant current 2019-10-21T21:54:16 < kakimir-pro> rc network was in isolated mains side? 2019-10-21T21:55:09 < kakimir-pro> you have your SCR designs still laying around aandrew? 2019-10-21T21:55:21 < kakimir-pro> I look it my boards and think that I don't want to die yet 2019-10-21T21:55:28 < kakimir-pro> *at my boards 2019-10-21T21:56:01 < kakimir-pro> I think redesign is required 2019-10-21T21:57:21 < aandrew> kakimir-pro: well this was three phase, I was driving through pulse transformers 2019-10-21T21:57:32 < kakimir-pro> interesting 2019-10-21T21:57:37 < kakimir-pro> mine is 3 phase too 2019-10-21T21:57:59 < aandrew> pulse transformer give isolation and current boost (1:2) 2019-10-21T21:58:43 < aandrew> I would have to look, I'm not sure 2019-10-21T21:58:49 < aandrew> I can more or less remember it though 2019-10-21T21:59:24 < aandrew> output of transformer was rectified (just two diodes) and fed to gate/mt1 2019-10-21T21:59:36 < aandrew> er that's triac. gate/kathode 2019-10-21T22:03:49 -!- renn0xtk9 [~max@2a02:810d:1540:2448:98e8:c0fa:d5fc:5bbf] has joined ##stm32 2019-10-21T22:05:34 < kakimir-pro> that is how you kept sufficient current to gate 2019-10-21T22:11:15 < aandrew> well we were continously firing 2019-10-21T22:11:42 < aandrew> so you'd send a 20kHz pulse train with the shape I'd described into the primary of the pulse transformer, halfwave rectify the ouptut and feed to gate 2019-10-21T22:11:53 < aandrew> hm, memfault when I telnet to hte board 2019-10-21T22:11:55 < aandrew> interesting. 2019-10-21T22:12:05 < kakimir-pro> mpu enabled? 2019-10-21T22:12:13 < aandrew> yes, intentionally 2019-10-21T22:12:22 < kakimir-pro> stack overflow 2019-10-21T22:12:28 < aandrew> don't think so 2019-10-21T22:12:30 < aandrew> but decoding 2019-10-21T22:12:38 < kakimir-pro> https://paste.ee/p/uMJQN I'm sketching 2019-10-21T22:12:39 < aandrew> MemFault MMSR:0x00000082 MMAR:0x0000001c $pc:0x00254848 $lr:0x00254779 r[0x0 0xff 0x1e 0x2001d2dc] r12:0x200131bc 2019-10-21T22:12:55 < kakimir-pro> MMAR valid? 2019-10-21T22:13:18 < aandrew> SR=0x82: FAR valid, data access caused the fault 2019-10-21T22:13:31 < kakimir-pro> isn't that the address of violation? 2019-10-21T22:13:50 < aandrew> AR=0x1c puts that in a "you aren't allowed to access 0x0000001c, sonny boy" 2019-10-21T22:14:56 < aandrew> 0x254848 is in my telnet_process() routine 2019-10-21T22:15:01 < aandrew> so so far it is making sense 2019-10-21T22:15:27 < aandrew> looks like my telnet handle is invalid 2019-10-21T22:15:31 < aandrew> crash is looking at telnet->state 2019-10-21T22:16:18 < aandrew> which is odd, because it was reading telnet->state a few lines above 2019-10-21T22:16:20 -!- jly [uid355225@gateway/web/irccloud.com/x-lvagrzdlmvcwayhp] has joined ##stm32 2019-10-21T22:19:02 < aandrew> yeah this memfault is odd 2019-10-21T22:19:04 < kakimir-pro> aandrew: did your controller work at wider range of mains hz? 2019-10-21T22:19:06 < aandrew> mind you I noticed telnet acting up ealrier 2019-10-21T22:19:21 < kakimir-pro> like 45-55 and 55-65? 2019-10-21T22:19:26 < aandrew> kakimir-pro: it was aimed at 50-60 but IIRC worked down below 30 and above 120 2019-10-21T22:19:31 < aandrew> needed some mods to work at 400Hz 2019-10-21T22:19:48 < kakimir-pro> you made it works in 400hz? 2019-10-21T22:19:54 < aandrew> yes (aircraft) 2019-10-21T22:19:55 < kakimir-pro> why? and what were the mods? 2019-10-21T22:20:14 < aandrew> https://www.benshaw.com/ebooks/890001-15-07-2/html5forpc.html that is the manual for it if you're interested 2019-10-21T22:20:15 < kakimir-pro> gate driving? 2019-10-21T22:20:57 < aandrew> software needed to be modded to not fault out on mains freq fault, there was something else as well (new CTs I think?) 2019-10-21T22:21:17 < kakimir-pro> CT? 2019-10-21T22:21:25 < kakimir-pro> transformers 2019-10-21T22:21:25 < kakimir-pro> ? 2019-10-21T22:21:31 < aandrew> current transformer yes 2019-10-21T22:21:56 < kakimir-pro> but did one version work the whole range or was there seperate 400hz version? 2019-10-21T22:22:02 < aandrew> it was separate 2019-10-21T22:22:03 < aandrew> special mod 2019-10-21T22:22:12 < kakimir-pro> and how big motors they want to drive in airplanes with those? 2019-10-21T22:22:26 < aandrew> yeah the spec for that board as standard was 25-75hz 2019-10-21T22:22:45 < aandrew> kakimir-pro: I don't remember the exact sutiation that was put in to, I wasn't onsite for that 2019-10-21T22:23:05 < kakimir-pro> so it's not widelly adopted thing? 2019-10-21T22:23:12 < aandrew> nooooo it was a one off 2019-10-21T22:23:18 < kakimir-pro> cool 2019-10-21T22:23:29 < kakimir-pro> chemtrail pumps needed controller? 2019-10-21T22:23:33 < aandrew> https://cdn11.bigcommerce.com/s-46jufh/images/stencil/1280x1280/products/427742/332882/230958058912__61248.1543450571.jpg?c=2&imbypass=on is a closeup shitty pic of the board 2019-10-21T22:24:02 < kakimir-pro> yes 2019-10-21T22:24:04 < kakimir-pro> shitti 2019-10-21T22:24:20 < kakimir-pro> did it has rtos? 2019-10-21T22:24:22 < aandrew> https://i.ebayimg.com/images/g/YygAAOSwdGFY0Sxn/s-l1600.jpg better pic 2019-10-21T22:24:23 < aandrew> no 2019-10-21T22:24:31 < kakimir-pro> multithread? 2019-10-21T22:24:47 < aandrew> that started out as a PIC16C74 and ended up a PIC16F877 2019-10-21T22:24:49 < kakimir-pro> that looks legacy stuff 2019-10-21T22:24:49 < aandrew> no 2019-10-21T22:24:51 < aandrew> 100% assembly 2019-10-21T22:24:53 < aandrew> yes it's ancient 2019-10-21T22:24:56 < aandrew> almost 20y now 2019-10-21T22:25:03 < kakimir-pro> mlcc packages man 2019-10-21T22:25:07 < kakimir-pro> plcc* 2019-10-21T22:25:16 < kakimir-pro> whatever 2019-10-21T22:25:18 < aandrew> that design was ~25% of benshaw's total worldwide sales at one point 2019-10-21T22:25:21 < aandrew> PLCC yes 2019-10-21T22:25:47 < kakimir-pro> was that 90's? 2019-10-21T22:25:48 < aandrew> that was drawn up and laid out in oldschool OrCAD SDT/PCB386+ 2019-10-21T22:25:50 < aandrew> yep 2019-10-21T22:26:03 < aandrew> I started in '99 and that was my first project 2019-10-21T22:27:01 < aandrew> but you can see the pulseforming network there immediately to the left of the small pulse transformers on the right side of the board 2019-10-21T22:27:31 < aandrew> I do not remember any rectification on the primary of the transformer though, can't remember what those two diodes are doing there 2019-10-21T22:27:41 < aandrew> on the right side of the transformers though that's halfwave rectification 2019-10-21T22:28:36 < aandrew> that board definitely looks well-used though. the transformers are all yellowed and dark 2019-10-21T22:28:42 < aandrew> it was in a warm enclosure for a lot of hours 2019-10-21T22:30:08 < aandrew> https://www.ebay.com/itm/272809646982 is what a smaller unit (50HP) looked like when the americans made them 2019-10-21T22:30:15 < kakimir-pro> is sizeof array always sizeof type times array length? 2019-10-21T22:30:17 < aandrew> they didn't take pride in it at all, looks like shit 2019-10-21T22:30:28 < aandrew> kakimir-pro: packed or unpacked? 2019-10-21T22:30:35 < kakimir-pro> unknown 2019-10-21T22:31:03 < kakimir-pro> https://paste.ee/p/uMJQN line26 2019-10-21T22:31:43 < aandrew> well you have uint16s there, those are going to be blown out to 32 bit alignment 2019-10-21T22:31:49 < aandrew> but otherwise yes 2019-10-21T22:32:21 < aandrew> sizeof(xPhaseDetectorTaskpvParametersArray [who the fuck names that?!]) = 3 * sizeof(xPhaseTaskpvParameters_Struct 2019-10-21T22:32:27 < Steffanx> why nucleo-64 boards dont do usb? 2019-10-21T22:33:20 < kakimir-pro> yes I changed it to xPhaseTaskpvParameterArray aandrew 2019-10-21T22:33:51 < kakimir-pro> as I realized it makes no sense to have separate detector and seperate thing to drive SCR 2019-10-21T22:34:05 < kakimir-pro> :p 2019-10-21T22:35:53 < kakimir-pro> padding is never added to array itself? 2019-10-21T22:36:17 < kakimir-pro> let's say I packed that struct 2019-10-21T22:36:40 < kakimir-pro> nvm 2019-10-21T22:37:54 < Steffanx> non volatile memory? 2019-10-21T22:39:00 < kakimir-pro> yes 2019-10-21T22:39:10 < kakimir-pro> don't store what I said there 2019-10-21T22:41:26 < kakimir-pro> aandrew: only a period of time after firing SCR is critical to keep driving gate? but does SCR have problem staying on after you stop driving gate to detect zeropoint? 2019-10-21T22:42:39 < kakimir-pro> it has sufficient current flow by then and will shut off like it should? 2019-10-21T22:43:14 -!- renn0xtk9 [~max@2a02:810d:1540:2448:98e8:c0fa:d5fc:5bbf] has quit [Quit: Konversation terminated!] 2019-10-21T22:43:41 < kakimir-pro> at threshold current 2019-10-21T22:47:55 < kakimir-pro> aandrew: what do you think of current zero point measurement and automatically setting inductance value by that? 2019-10-21T22:48:30 < aandrew> kakimir-pro: I think you're overthinking it 2019-10-21T22:48:30 < kakimir-pro> inductance is then used to control timing of SCR firing 2019-10-21T22:49:19 < kakimir-pro> did your board have load inductance setting? 2019-10-21T22:49:30 < kakimir-pro> some dip switch thing to range it in? 2019-10-21T22:49:58 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-21T22:53:31 < kakimir-pro> your controller had current measurement aandrew 2019-10-21T22:54:57 < kakimir-pro> transformers smart 2019-10-21T23:17:22 -!- renn0xtk9 [~max@2a02:810d:1540:2448:5136:d5b4:2e79:2332] has joined ##stm32 2019-10-21T23:18:07 < Steffanx> hmm, why mouser still doesnt do paypal for non-yankees? 2019-10-21T23:18:31 < zyp> they don't? 2019-10-21T23:18:38 < zyp> fees, maybe? 2019-10-21T23:18:46 < Steffanx> i dont see the option 2019-10-21T23:20:28 < antto> yeah, no paypal 2019-10-21T23:21:08 < antto> neither paysecam 2019-10-21T23:21:25 < Steffanx> the what? 2019-10-21T23:21:30 < antto> PAL/SECAM 2019-10-21T23:21:31 < antto> bruh 2019-10-21T23:21:44 < antto> r u an NTSC guy? 2019-10-21T23:22:31 < Steffanx> i never gave a damn about that 2019-10-21T23:24:04 -!- sandeepkr [~sandeepkr@2a03:b0c0:2:d0::cac:7001] has joined ##stm32 2019-10-21T23:31:58 < kakimir-pro> they should take bitcoins 2019-10-21T23:35:03 < Steffanx> fuck no 2019-10-21T23:35:11 < antto> crapcoinz 2019-10-21T23:35:12 < Steffanx> is buttcoin dead yet? 2019-10-21T23:35:21 <@englishman> kakimir-verynonpro 2019-10-21T23:35:36 < antto> what's the opposite of pro? 2019-10-21T23:36:18 < kakimir-pro> noob 2019-10-21T23:36:25 -!- kakimir-pro is now known as kakinoob 2019-10-21T23:36:30 < antto> aww noes 2019-10-21T23:36:36 < kakinoob> lvl--; 2019-10-21T23:36:36 < antto> there must be a bett0r word 2019-10-21T23:36:50 < antto> dngrdmir 2019-10-21T23:36:52 < kakinoob> beginner 2019-10-21T23:36:56 < Steffanx> have some random song that is too much on the radio here lately: https://www.youtube.com/watch?v=q0hyYWKXF0Q 2019-10-21T23:36:58 < kakinoob> green 2019-10-21T23:37:01 < kakinoob> newbie 2019-10-21T23:37:08 < kakinoob> kaki? 2019-10-21T23:38:36 < kakinoob> extraordinary sound Steffanx 2019-10-21T23:38:40 < kakinoob> *voice 2019-10-21T23:38:49 < antto> which track is it? 2019-10-21T23:38:58 < kakinoob> dance monkey 2019-10-21T23:39:06 < antto> UGH 2019-10-21T23:39:16 < antto> GET OUTTA HERE 2019-10-21T23:39:40 < antto> lemme find u some bett0r track to listen to 2019-10-21T23:39:42 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-21T23:39:55 < kakinoob> now tell me how gay it's to control triacs with code and tickcounts instead of perfectly crafted timer system? 2019-10-21T23:40:10 * antto spits on el finger, and turns to teh pootube tab 2019-10-21T23:43:59 < Steffanx> tick .. tock.. tick.. tock.. antto 2019-10-21T23:44:02 < antto> https://www.youtube.com/watch?v=PauhYoJxMv4 2019-10-21T23:44:27 < Steffanx> nah ill pass :P 2019-10-21T23:44:43 < antto> why.. r u a.. CHICKEN? 2019-10-21T23:44:46 < antto> BOK BOK BOK 2019-10-21T23:45:16 < antto> open it, it's not rick astley 2019-10-21T23:45:17 < Steffanx> Are you croatian now? 2019-10-21T23:45:27 < Steffanx> bok = hi 2019-10-21T23:50:01 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-21T23:50:39 < antto> here's another one https://www.youtube.com/watch?v=3VQ5IjzbzY8 2019-10-21T23:50:46 < Steffanx> i opened it, and then moved on antto 2019-10-21T23:50:49 * Jan- attempts to write code for stm32 2019-10-21T23:50:54 < Jan-> :/ 2019-10-21T23:50:54 -!- maunix [~Mauricio@181.31.151.95] has quit [Quit: WeeChat 2.6] 2019-10-21T23:51:18 < antto> 2019-10-21T23:51:35 < antto> 2019-10-21T23:57:40 * kakinoob cheers Jan- 2019-10-21T23:57:52 < antto> https://www.youtube.com/watch?v=EWYr3QnT64M 2019-10-21T23:57:53 < Jan-> yeah I'm stuck at the "makefile" stage :( 2019-10-21T23:58:08 < antto> aww mekkfailz 2019-10-21T23:58:21 < kakinoob> get system where you can understand the build process IF you need to 2019-10-21T23:58:36 < antto> eye dee ee 2019-10-21T23:58:42 < kakinoob> antto: 5/5 2019-10-21T23:58:51 < kakinoob> very 90s 2019-10-21T23:59:02 < antto> what's very 90s? 2019-10-21T23:59:10 < kakinoob> musics 2019-10-21T23:59:16 < antto> which one? 2019-10-21T23:59:17 < kakinoob> or early 2000s 2019-10-21T23:59:19 < kakinoob> latest 2019-10-21T23:59:25 < antto> Hi Lo? 2019-10-21T23:59:26 < kakinoob> wait 2019-10-21T23:59:30 < antto> iz not old 2019-10-21T23:59:32 < kakinoob> jack the ripper 2019-10-21T23:59:38 < antto> that one is old 2019-10-21T23:59:48 < kakinoob> exact year? 2019-10-21T23:59:49 < antto> yet super phoochuristic --- Day changed Tue Oct 22 2019 2019-10-22T00:00:01 < antto> innovation 2019-10-22T00:00:06 < antto> i don't know 2019-10-22T00:00:08 < kakinoob> it sounds like innovating 2019-10-22T00:00:15 < antto> yeah 2019-10-22T00:00:46 < antto> them folks in the 90s when they heard tipper probably went "wat da fug?" 2019-10-22T00:01:03 < antto> "muh cocain is too weak for this sh*t" 2019-10-22T00:01:37 <@englishman> hmm 2019-10-22T00:01:46 <@englishman> fedex shipment of like $3k of shit says delivered 2019-10-22T00:01:50 <@englishman> name on signature i dont recognize 2019-10-22T00:02:01 <@englishman> called them up, they said oh yeah looks like it was delivered incorrectly 2019-10-22T00:02:05 <@englishman> ............ 2019-10-22T00:02:16 < antto> "it happens" 2019-10-22T00:02:29 <@englishman> if the phone dude can see that so easily, why the shit did the computer in the driver's hand not say that 2019-10-22T00:02:47 < antto> "comput0r bug" 2019-10-22T00:02:53 < antto> "they so pesky u know" 2019-10-22T00:03:48 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has joined ##stm32 2019-10-22T00:03:54 < Amun_Ra> they do not check an ID on delivery? 2019-10-22T00:04:00 < BrainDamage> almost never 2019-10-22T00:04:19 <@englishman> i was hoping they would just safe drop it 2019-10-22T00:04:31 <@englishman> as i don't live in an inner city ghetto 2019-10-22T00:04:36 * qyx running debian installer on a imx6 2019-10-22T00:04:50 < antto> nicebian 2019-10-22T00:04:55 < BrainDamage> I've even had costly material shoved in the mailbox slot ... deforming the hard envelope 2019-10-22T00:05:01 < BrainDamage> and my signature faked 2019-10-22T00:05:07 < antto> aww 2019-10-22T00:05:21 < BrainDamage> that took quite some yelling to get them to admit it 2019-10-22T00:05:34 <@englishman> is it coincidence that the guy who "signed" for it has the same name as the meme candidate for today's federal election 2019-10-22T00:05:50 < antto> spoodrman? 2019-10-22T00:05:58 < antto> dolan? 2019-10-22T00:06:06 < antto> no, wait, i don't know 2019-10-22T00:06:31 < antto> Fakey McTheiferson? 2019-10-22T00:06:36 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-22T00:11:56 < Jan-> So I need a system.c to suit my board with an stm32f407vet6 processor on it. 2019-10-22T00:12:04 < Jan-> I found one for the stm32f4discovery board 2019-10-22T00:12:21 < Jan-> but that has an stm32f407VG processor on it. 2019-10-22T00:12:24 < Jan-> so is that likely to be OK 2019-10-22T00:12:25 < antto> forget the board, get the chip started, then worry about the pinz 2019-10-22T00:12:35 < Jan-> that's the plan but it's not the same chip 2019-10-22T00:12:51 < antto> usually mcus keep uninitialized pins in a state that isn't too dangerous for the chip itself 2019-10-22T00:13:24 < antto> https://www.youtube.com/watch?v=L5c8rGWqAHY 2019-10-22T00:13:50 < antto> if dis don't make yo booty move.. 2019-10-22T00:22:22 < Steffanx> too much of the same, sorry 2019-10-22T00:22:48 < kakinoob> how does vTaskDelayUntil work when time is beyond one tickcount overflow? 2019-10-22T00:23:17 < kakinoob> I mean freertos kernel variable xTickCount or something 2019-10-22T00:25:52 -!- jly [uid355225@gateway/web/irccloud.com/x-lvagrzdlmvcwayhp] has quit [Quit: Connection closed for inactivity] 2019-10-22T00:26:39 < kakinoob> if the parameter value is current xTickCount + TickCount_Maximum then it's in future and if it's over that then it's in past? 2019-10-22T00:27:54 < kakinoob> I know vTaskDelay deals with it somehow so I need to check that implementation 2019-10-22T00:28:28 < Steffanx> i cannot imagine it does not deal with that 2019-10-22T00:30:01 < catphish> zyp: by the way, now i'm able to test properly, double buffering absolutely does make a difference to USB performance, good performance (750KB/s) is achieveable without it, but once i got double buffering working correctly, i can now achieve 1.2MB/s reliably in either direction 2019-10-22T00:31:38 < zyp> still on st_usbfs? 2019-10-22T00:31:45 < kakinoob> Steffanx: I see it puts it in seperate list 2019-10-22T00:31:49 < catphish> yes, l433 2019-10-22T00:32:14 < kakinoob> pxOverflowDelayedTaskList instead of pxDelayedTaskList 2019-10-22T00:32:51 < catphish> i'll publish my double buffering driver, because nobody else seems to have done so :) 2019-10-22T00:33:11 < antto> apt-get upgrade kakinoob 2019-10-22T00:33:45 < kakinoob> sudo apt-get install kakinoobinoob 2019-10-22T00:35:36 < catphish> https://github.com/catphish/stm32-doublebuffer 2019-10-22T00:36:14 < catphish> once i got off the drugs, the implementation was relatively straightforward 2019-10-22T00:38:23 < kakinoob> oh vTaskDelayUntil has too parameters 2019-10-22T00:38:28 < kakinoob> 2wo 2019-10-22T00:38:37 < kakinoob> 2wto 2019-10-22T00:39:30 < kakinoob> if I want it to unblock at absolute time I set previous time to time I want to unblock and increment I set to 0? 2019-10-22T00:45:36 < kakinoob> bullshit 2019-10-22T00:45:43 < kakinoob> I need to make my own delay function 2019-10-22T00:49:49 < kakinoob> unless 2019-10-22T00:49:51 < kakinoob> I trick it 2019-10-22T00:50:29 < kakinoob> I set previous time to time before current time 2019-10-22T00:50:36 < kakinoob> then add increment to make absolute time 2019-10-22T00:50:59 < kakinoob> sweet 2019-10-22T00:51:35 < kakinoob> let's see if current time is enough 2019-10-22T00:51:42 < kakinoob> for previous time parameter 2019-10-22T00:53:59 < kakinoob> indeed current time is ok 2019-10-22T01:05:20 < karlp> arm's rejected my documentation errata. 2019-10-22T01:05:42 < karlp> they're like, it might be inconsisten, but fuck you, we're not changing anything, and it's mentioned in afootnote elsewhere. 2019-10-22T01:12:11 < kakinoob> you tried to teach ARM? 2019-10-22T01:16:12 < karlp> I'm following up, asking them to be consistent about it, but not really hopefuly :) 2019-10-22T01:16:19 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-22T01:16:29 < karlp> nvic registers all say "usage constraints: Subject to standard PPB usage constraints, see General rules for PPB register accesses on 2019-10-22T01:16:31 < karlp> page B3-226. 2019-10-22T01:16:48 < karlp> other sections of the SCS (in the PPB) just says, "no restrictions" 2019-10-22T01:17:01 < kakinoob> I need feet heater 2019-10-22T01:17:02 < karlp> and you're meant to have gone specificallyh and read the PPB restrictions earlier. 2019-10-22T01:24:57 < qyx> so, I downloaded a debian installer iso 2019-10-22T01:25:03 < qyx> sha256 checksum ok 2019-10-22T01:25:28 < qyx> I format a microsd with ext4, copy the iso here, check sha256sum, differs 2019-10-22T01:25:37 < qyx> reformat with vfat, copy, check, checksum is ok 2019-10-22T01:26:52 < qyx> ok, so running the installer again with the iso on a vfat partition 2019-10-22T01:27:47 -!- renn0xtk9 [~max@2a02:810d:1540:2448:5136:d5b4:2e79:2332] has quit [Ping timeout: 276 seconds] 2019-10-22T01:28:08 < kakinoob> anyone have full version IDA? 2019-10-22T01:28:14 < kakinoob> IDA PRO 2019-10-22T01:30:16 < mawk> what for kakinoob ? 2019-10-22T01:30:20 < mawk> are you kakipirate now ? 2019-10-22T01:30:28 < kakinoob> idk. pro stuff 2019-10-22T01:31:50 < qyx> checksumming the same file on the imx6 yelds another incorrect sha256sum 2019-10-22T01:32:49 < kakinoob> movierecommd 2019-10-22T01:47:01 < kakinoob> need movie 2019-10-22T01:47:22 < mawk> joker 2019-10-22T01:47:39 < kakinoob> not available 2019-10-22T01:48:13 < mawk> what do you mean 2019-10-22T01:48:16 < mawk> download it 2019-10-22T01:48:46 < kakinoob> here it's playing in theaters? 2019-10-22T01:49:00 < kakinoob> have they released it online already? 2019-10-22T01:49:11 < mawk> I saw it in theater 2019-10-22T01:49:13 < mawk> but I'm sure it's online 2019-10-22T01:50:28 < mawk> not on my french torrent site tho 2019-10-22T01:59:18 < antto> kakinoob Titanic ;P~ 2019-10-22T01:59:20 * antto runs 2019-10-22T01:59:31 < kakinoob> gtfo 2019-10-22T01:59:47 < kakinoob> -.- 2019-10-22T01:59:58 < antto> EEEVRY NITE IN MAH DREAMS, EYE SEE U 2019-10-22T02:00:22 < antto> I F E E E E L U 2019-10-22T02:01:04 < antto> i watched "7 sisters" or some such, a few days ago.. it wasn't bad 2019-10-22T02:01:27 < antto> wut genre, kakinoob? 2019-10-22T02:01:47 < kakinoob> recommend genre 2019-10-22T02:01:58 < antto> drama 2019-10-22T02:02:00 < antto> Titanic! 2019-10-22T02:02:02 < kakinoob> no 2019-10-22T02:02:04 * antto runs fast0r 2019-10-22T02:02:20 < antto> don't watch Born twice 2019-10-22T02:02:22 < kakinoob> no drama it bores the shit out of me 2019-10-22T02:03:15 -!- canton7 [~canton7@about/csharp/regular/canton7] has quit [Quit: ZNC - http://znc.in] 2019-10-22T02:03:47 < antto> have u watched automata (i think) .. with antonio banderaz? 2019-10-22T02:04:17 < antto> iz iz shot in .bg ;P~ 2019-10-22T02:06:08 < antto> horrors and thrillers? 2019-10-22T02:06:51 < kakinoob> no 2019-10-22T02:07:04 < antto> i lieked the apparition 2019-10-22T02:09:14 < Steffanx> Lion king 2019-10-22T02:09:28 < Steffanx> Doom 2019-10-22T02:09:39 < Steffanx> Spiderman 2019-10-22T02:09:47 < antto> i was gonna suggest lion king but i thought that's too much after titanic 2019-10-22T02:09:59 < antto> spoodrman \o/ 2019-10-22T02:16:38 < antto> there was another one that was noice but i can't remember its name 2019-10-22T02:17:09 -!- canton7 [~canton7@about/csharp/regular/canton7] has joined ##stm32 2019-10-22T02:18:31 < dongs> ... my latvian order is pending again 2019-10-22T02:18:32 < dongs> what the fuck 2019-10-22T02:18:48 < antto> lat va f*ck 2019-10-22T02:19:18 < dongs> if my SXT was really dead i'd be pissed that it takes them a week to ship shit 2019-10-22T02:25:28 -!- con3 [~kvirc@154.119.40.228] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-22T02:28:58 < antto> is it me or does imdb search sux? 2019-10-22T02:29:08 < antto> can't find sh*t in it 2019-10-22T02:29:54 < antto> too bad i'm not sure who the main actress woz 2019-10-22T02:38:05 < zyp> dongs, norwegian webstores have started selling mikrotik stuff now, so I'll probably just buy from a domestic store next time I need something 2019-10-22T02:42:30 < zyp> I mean, the switch I'm planning to buy eventually is gonna cost me ~4k NOK to buy from eurodk with shipping and all, domestic it's 3.7 2019-10-22T02:43:09 < zyp> and buying domestic entitles me to five year warranty-ish as well 2019-10-22T02:45:42 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 265 seconds] 2019-10-22T02:46:02 < kakinoob> I actually needed to make 2 tasks per phase 2019-10-22T02:46:11 < kakinoob> one calculating shit and another firing triacs 2019-10-22T02:46:35 < kakinoob> because between firing task is otherwise blocked 2019-10-22T02:52:35 < aandrew> that sounds... bad 2019-10-22T02:53:15 < aandrew> ideally you would use timer outputs to drive the gate firing 2019-10-22T02:53:24 < aandrew> you load up the time to start firing and let the hardware handle it 2019-10-22T02:53:27 < karlp> that wouldn't be overthinking it enough... 2019-10-22T02:53:46 < aandrew> software phase control on an stm32 is a waste of the powerful timer output capabilities 2019-10-22T02:55:19 < kakinoob> I can replace firing tasks with timers 2019-10-22T02:56:42 < dongs> i think kikemir's oven shit is a fucking LPC 2019-10-22T02:57:12 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-22T02:57:16 < kakinoob> correct 2019-10-22T02:57:27 < dongs> zyp thats pretty expensive, their 24port/10gb sfp thing I bought was liek $<200 i thought 2019-10-22T02:57:27 < kakinoob> there is timers that can do shit 2019-10-22T02:57:45 < kakinoob> but I need to form this control in software first 2019-10-22T03:00:15 < kakinoob> I need like a thing 2019-10-22T03:00:28 < kakinoob> that let's other task to block and unblock other task 2019-10-22T03:00:46 < kakinoob> not just unblock but also block 2019-10-22T03:01:56 -!- Jan- [~IceChat9@87.114.135.182] has quit [Ping timeout: 268 seconds] 2019-10-22T03:02:35 < dongs> wait, it was less than $150 even 2019-10-22T03:03:02 < dongs> holy shit its only 110 on eurodk now 2019-10-22T03:03:18 < dongs> or $99 used 2019-10-22T03:03:21 < dongs> https://www.eurodk.com/en/products/crs/cloud-smart-switch-326-24g-2srm 2019-10-22T03:03:46 < kakinoob> you use eurodk too? 2019-10-22T03:04:56 < kakinoob> I want to see how much my mcu can do with core alone 2019-10-22T03:05:09 < dongs> did you solve your retarded shit with jlink 2019-10-22T03:05:14 < kakinoob> yes 2019-10-22T03:05:21 < dongs> how 2019-10-22T03:05:28 < kakinoob> which retarted shit? 2019-10-22T03:06:03 < dongs> idk 2019-10-22T03:06:07 < dongs> you had some shit that didnt work 2019-10-22T03:06:16 < kakinoob> it's been pretty nice with jlink 2019-10-22T03:06:28 < kakinoob> launch doesn't work 2019-10-22T03:06:33 < kakinoob> server launch from IDE 2019-10-22T03:06:43 < kakinoob> not an issue 2019-10-22T03:07:08 < kakinoob> with openocd I had problem with flash hole 2019-10-22T03:07:16 < kakinoob> it's gone with jlink server 2019-10-22T03:07:27 < aandrew> dongs: lpc? 2019-10-22T03:07:32 < dongs> yes 2019-10-22T03:07:43 < dongs> as in NXP LPC 2019-10-22T03:07:46 < aandrew> what is lpc other than a shitty processor 2019-10-22T03:07:47 < aandrew> ah 2019-10-22T03:07:52 < dongs> yes, exactly that 2019-10-22T03:07:54 < dongs> a shitty processor 2019-10-22T03:07:58 < aandrew> he's being sneaky getting help here 2019-10-22T03:08:06 < dongs> ikr 2019-10-22T03:08:16 < dongs> cuz #lpc is dead as expected 2019-10-22T03:08:16 -!- Jan- [~IceChat9@87.112.99.183] has joined ##stm32 2019-10-22T03:08:43 < kakinoob> I don't ask how to use peripherals 2019-10-22T03:09:11 < kakinoob> I don't ask much questions that cannot be derived from stm32 world 2019-10-22T03:09:34 < aandrew> I'm just kidding 2019-10-22T03:09:38 < aandrew> cortex is cortex 2019-10-22T03:09:46 < aandrew> not my problem if you choose to use LPC :-) 2019-10-22T03:10:00 < aandrew> but yeah this isn't really a good idea to do phase control in software 2019-10-22T03:10:01 < aandrew> you can do it though 2019-10-22T03:10:04 < aandrew> it'll work 2019-10-22T03:10:35 < aandrew> personally I think I'd have a single high priority task that controls phase and the other threads change its target phase angle as needed 2019-10-22T03:10:36 < kakinoob> there we go 2019-10-22T03:11:06 < kakinoob> aandrew: is there a thing to both unblock and block task from another task? 2019-10-22T03:11:37 < mawk> I guess you can always enter a critical section, do your two things, and exit it, kakinoob 2019-10-22T03:11:42 < mawk> then it's atomic 2019-10-22T03:11:47 < aandrew> kakinoob: I don't know, but I'd have to ask why you want to do that 2019-10-22T03:11:49 < mawk> critical section = disable interrupts 2019-10-22T03:11:51 < aandrew> task switching is shit 2019-10-22T03:12:35 < kakinoob> if I don't have valid timings for firing task 2019-10-22T03:13:04 < kakinoob> maybe I just keep giving semaphore 2019-10-22T03:13:04 < aandrew> all you can do is make the firing task high prio and at every opportunity compare and start firing if it's time 2019-10-22T03:13:29 < kakinoob> no it's driven with ticks 2019-10-22T03:13:30 < aandrew> is there a freertos task switching algo which will always bounce back to the highest prio after every task switch? 2019-10-22T03:13:38 < aandrew> I don't think it's needed if the priroities are done right 2019-10-22T03:14:16 < kakinoob> isn't highest priority tasks satisfied first? 2019-10-22T03:14:37 < kakinoob> until all in blocked state then move one step down in priority etc. 2019-10-22T03:19:49 -!- boB_K7IQ [~boB_K7IQ@c-98-247-192-179.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-22T03:22:19 < aandrew> yes I think it is 2019-10-22T03:22:55 < aandrew> so just make sure your highest prioity task is never blocking, and none of your other tasks will disable IRQs 2019-10-22T03:25:14 < kakinoob> then only highest priority task is ran? 2019-10-22T03:25:45 < kakinoob> it kinda defies the whole multitasking 2019-10-22T03:35:17 < aandrew> no 2019-10-22T03:35:22 < aandrew> highest priority gets its timeslice 2019-10-22T03:35:33 < mawk> Steffanx Steffanx Steffanx 2019-10-22T03:35:34 < aandrew> then it's put back on the list and the nexxt one up gets its turn 2019-10-22T03:35:46 < mawk> I need help Steffanx 2019-10-22T03:35:52 < kakinoob> steffpapa sleeps 2019-10-22T03:35:54 < aandrew> then after that task is done if the high task doesn't need its tick it goes on again to the next 2019-10-22T03:35:56 < mawk> ah :( 2019-10-22T03:36:13 < aandrew> so basically your main loop on the high task will be while (1) { do_shit(); yield(); } 2019-10-22T03:36:15 < mawk> 1175€ for 40m² in Delft, where am I going to live with that kind of rent price 2019-10-22T03:38:08 < Cracki> suburb? 2019-10-22T03:38:15 < Cracki> that is very pricey 2019-10-22T03:38:42 < mawk> I did 1h30 of subway to my work/school my whole life, I don't want to do that anymore 2019-10-22T03:38:55 < mawk> I'd rather eat dry bread and water everyday and be close to the job 2019-10-22T03:39:10 < Cracki> city where I live, you can have a 10 minute commute via train which is 5-10 km outside the city center 2019-10-22T03:39:21 < Cracki> that's better than almost anywhere in between because then you only get busses 2019-10-22T03:39:29 < Cracki> literally 5-10 minute commute 2019-10-22T03:39:33 < mawk> nice 2019-10-22T03:40:01 < Cracki> but requires you to live in a suburb, train comes every half hour or so 2019-10-22T03:40:23 < Cracki> it's a lot more plannable than busses 2019-10-22T03:40:36 < Cracki> still, 40 qm for 1175 is outrageous 2019-10-22T03:41:10 < Cracki> I pay 530 for 60 qm, admittedly it's an old building with shitty plumbing 2019-10-22T03:42:52 < Cracki> hm, random site for rents in delft says it doesn't go down much if you rent outside city centre 2019-10-22T03:43:02 < Cracki> maybe everyone earns just that much there 2019-10-22T03:43:40 < Cracki> or you found the most expensive offer 2019-10-22T03:43:51 < Cracki> 3 bedrooms 1080/month https://www.tudelft.nl/en/about-tu-delft/working-at-tu-delft/coming-to-the-netherlands-tu-delft/support-for-international-employees/cost-of-living/ 2019-10-22T03:45:54 < dongs> how many stroopwafels is that 2019-10-22T03:46:30 < mawk> yeah the job is in the most expensive neighbourhoud Cracki 2019-10-22T03:46:32 < mawk> Hof van Delft 2019-10-22T03:46:41 < Cracki> get a bicycle? 2019-10-22T03:46:43 < Cracki> electric? 2019-10-22T03:46:55 < Cracki> makes for an awesome commute 2019-10-22T03:47:07 < mawk> yeah why not 2019-10-22T03:47:27 < mawk> living in the centrum is nice tho, you have all the shops at walk distance 2019-10-22T03:47:28 < Cracki> bitch what 2019-10-22T03:47:36 < Cracki> it's ~5 km to the hague, ~10 to rotterdam? 2019-10-22T03:47:43 < mawk> yes NL is small 2019-10-22T03:47:50 < Cracki> that's like RIGHT AROUND THE CORNER with a bike 2019-10-22T03:47:53 < mawk> lol 2019-10-22T03:48:43 < Cracki> a half hour to stick your feet in ocean 2019-10-22T03:52:52 < mawk> on the dutch tv there is a show like "help us find the criminal" which shows security footage 2019-10-22T03:53:07 < mawk> and they make a physical description of the criminal, everytime they say "light tan" as an euphemism for morrocans 2019-10-22T03:53:23 < Cracki> kek 2019-10-22T03:54:35 < Cracki> the world needs an audio filter that replaces euphemisms with their true meaning 2019-10-22T03:55:29 < kakinoob> you can't say that word mawk 2019-10-22T03:55:49 < kakinoob> let's move to kakimir32> 2019-10-22T03:55:56 < mawk> lol 2019-10-22T03:58:51 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-22T03:59:29 < kakimir32> loitermir here 2019-10-22T04:00:17 < aandrew> huh, interesting 2019-10-22T04:00:30 < aandrew> four stm32f756 boards, very similar software 2019-10-22T04:00:35 < aandrew> two devboards and two custom 2019-10-22T04:01:07 < aandrew> I have almost twice the heap available on the 2 devboards and custom board running almost identical firmware 2019-10-22T04:01:20 < aandrew> the other custom board with similar but not THAT similar is much less RAM avail 2019-10-22T04:01:36 < aandrew> but I can't think of anything in the software that would eat up 6kB 2019-10-22T04:01:59 < aandrew> anyway 2019-10-22T04:02:00 < aandrew> 9pm 2019-10-22T04:02:03 < aandrew> time to find some food 2019-10-22T04:02:39 < kakimir32> aandrew is murican? 2019-10-22T04:02:46 < aandrew> canadian 2019-10-22T04:02:52 < kakimir32> fellow englishman 2019-10-22T04:05:08 < dongs> I should have never taken this fucking kikecad assembly. motherfucking garbage 2019-10-22T04:05:21 < dongs> oh well, i warned them if it doesnt work fuckem 2019-10-22T04:05:42 < kakimir32> what? 2019-10-22T04:06:03 < kakimir32> are you doing assembly now? 2019-10-22T04:06:27 < kakimir32> what is the problem? 2019-10-22T04:06:31 <@englishman> aandrew: someone named MBERNIER signed for my FedEx package 2019-10-22T04:06:47 < kakimir32> dongs: placement file is a mess? 2019-10-22T04:06:48 <@englishman> those fucking fascists are already taking shit out of my pocket 2019-10-22T04:07:01 < kakimir32> components are a mess? 2019-10-22T04:11:57 < kakimir32> otherwise I would do triac driving with timers 2019-10-22T04:12:37 < kakimir32> but how do I learn to play with scheduler if I don't put propper load on it 2019-10-22T04:12:52 < aandrew> englishman: nice, who knew he was into electronics 2019-10-22T04:13:18 < kakimir32> I didn't have a clue how it works before aandrew told me a minute ago in plain english 2019-10-22T04:13:26 < mawk> he sounds french englishman 2019-10-22T04:13:48 < aandrew> mawk: Maxime Bernier is the leader of "the people's party of canada" - someone trying to get elected tongiht 2019-10-22T04:13:55 <@englishman> mawk, Maxime Bernier is a meme politician here 2019-10-22T04:14:05 < mawk> ah lol 2019-10-22T04:19:10 < dongs> i think if i take any opensores shit in future im just gonna fucking charge "dealign wiht opensores" fee 2019-10-22T04:19:28 < dongs> kakimir32: yes, plus wrong footprints, plus just general bad decisions on top of bad decisions everywhere 2019-10-22T04:19:48 < dongs> plus they delayed the shit by 2 weeks because the fucking position file didnt include ALL THE PARTS 2019-10-22T04:19:56 < kakimir32> just charge by hour basis 2019-10-22T04:20:02 < kakimir32> to fix the shit 2019-10-22T04:20:02 < dongs> because apparently kikecad has some fucking flag for throughole/smt 2019-10-22T04:20:11 < dongs> and if they dont set that, it doesnt end up on dicnplace file 2019-10-22T04:20:30 < kakimir32> yes 2019-10-22T04:20:40 < kakimir32> THT or virtual part and there is no dickplace 2019-10-22T04:20:51 < kakimir32> it's an easy mistake if you don't check it 2019-10-22T04:21:00 < kakimir32> default for new part is THT 2019-10-22T04:21:31 < kakimir32> are they how new to all of this dongs? 2019-10-22T04:22:02 < dongs> no idea 2019-10-22T04:23:07 < dongs> i also cant tell for sure but it looks like they have a 1.8V plane pour on top layer which clearly has some vias going down to GND 2019-10-22T04:23:19 < dongs> like i said i dont care at this point fuckem 2019-10-22T04:23:45 < dongs> fucking shitty j-startup thinking they can make money with opensores 2019-10-22T04:33:25 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away] 2019-10-22T04:34:38 <@englishman> take their money with pride 2019-10-22T04:36:07 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-22T04:36:46 < dongs> they used some torex DC/DC integrated shit 2019-10-22T04:36:49 < dongs> that was like $3/ea 2019-10-22T04:37:03 < dongs> and datasheet had a completely fucing clear recommended layout 2019-10-22T04:37:10 < dongs> they went ahead and did it completely wrong 2019-10-22T04:37:20 < mawk> lol 2019-10-22T04:37:27 < mawk> some eng thought he was being smart 2019-10-22T04:37:33 < mawk> doing better than the manufacturer 2019-10-22T04:38:04 < dongs> https://www.torexsemi.com/file/xcl102/XCL102-103.pdf 2019-10-22T04:38:16 < dongs> page 13 2019-10-22T04:38:32 < dongs> https://i.imgur.com/re7X7Xi.png this is what the clowns did.. 2019-10-22T04:39:12 < mawk> lol 2019-10-22T04:39:20 < mawk> they put the caps the furthest away possible from the IC it seems 2019-10-22T04:39:24 < mawk> it's active sabotage 2019-10-22T04:39:47 < dongs> that and LX shit is just a tiny fucking trace 2019-10-22T04:39:51 < dongs> for both ends of the coil 2019-10-22T04:41:00 <@englishman> aandrew: I'm at a liberals campaign party 2019-10-22T04:41:07 <@englishman> don't ask me why or how 2019-10-22T04:41:11 < dongs> im sure its really hard to do pours in kikecad 2019-10-22T04:41:22 < mawk> lol englishman 2019-10-22T04:41:33 < mawk> did anyone bring you there by force ? 2019-10-22T04:41:42 <@englishman> no it's really happenstance 2019-10-22T04:43:32 <@englishman> she will lose so it'll be pretty fun to see 2019-10-22T04:49:56 -!- Tricky [~quassel@2409:8900:1b10:6167:15c3:e478:3fd9:5720] has joined ##stm32 2019-10-22T04:54:52 < aandrew> I was going to as why in fact 2019-10-22T04:55:12 < aandrew> I'm in detroit right now 2019-10-22T04:55:18 < aandrew> watching the live cbc converage and chatting 2019-10-22T04:55:37 < aandrew> drinking an old fashioned and plan on getting nicely drunk tonight 2019-10-22T04:55:43 < aandrew> not election related 2019-10-22T05:00:01 -!- Tricky [~quassel@2409:8900:1b10:6167:15c3:e478:3fd9:5720] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.] 2019-10-22T05:03:30 < kakimir32> why did you take kikecad startup job dongs? 2019-10-22T05:03:39 -!- zyp [zyp@zyp.no] has quit [Remote host closed the connection] 2019-10-22T05:06:42 < aandrew> he recognizes the superiority 2019-10-22T05:17:41 < aandrew> heh cbc predicts liberal gov, but can't say minority/majority 2019-10-22T05:17:47 < aandrew> I'm surprised NDP have done as well as they have 2019-10-22T05:28:47 < dongs> isnt this a convo more suitable for lwz, aandrew 2019-10-22T05:28:52 < dongs> nobody here gives a fuck about nigger politics 2019-10-22T05:32:39 < dongs> annd niggerrific assembly done 2019-10-22T05:32:43 < dongs> now i can tell them to fuck off 2019-10-22T05:33:16 < dongs> i dont think ive seen a worse USB connector before either 2019-10-22T05:34:19 < dongs> https://www.digikey.jp/product-detail/en/hirose-electric-co-ltd/ZX62R-B-5P-30/H125274CT-ND/5994760 2019-10-22T05:34:41 < dongs> its completely SMT, has NO alignment pegs of any kind, and the pins are under the cover 2019-10-22T05:40:10 < kakimir32> you assy protos for them? 2019-10-22T05:40:43 < aandrew> dongs: there's a fellow canadian here 2019-10-22T05:41:15 < aandrew> wow that looks like an aful connector 2019-10-22T05:41:16 < aandrew> awful 2019-10-22T05:41:39 < dongs> kikemir yea 2019-10-22T06:00:31 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-22T06:03:33 <@englishman> beep mawk 2019-10-22T06:03:55 <@englishman> https://i.redd.it/k7nvwyp18lt31.jpg 2019-10-22T06:22:50 -!- fc5dc9d4 [~quassel@p5B0810CD.dip0.t-ipconnect.de] has joined ##stm32 2019-10-22T06:26:27 -!- fc5dc9d4_ [~quassel@p5B0816AA.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-22T06:37:48 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-22T07:32:09 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-22T07:32:12 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-22T07:35:15 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 250 seconds] 2019-10-22T07:35:16 -!- day__ is now known as day 2019-10-22T07:43:48 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-22T07:44:03 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-fzkcxxijextsyxon] has quit [Ping timeout: 240 seconds] 2019-10-22T07:45:10 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-sehzcmqtjptozkgq] has joined ##stm32 2019-10-22T08:06:26 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-22T08:32:45 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-22T08:33:45 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Client Quit] 2019-10-22T08:46:00 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-22T09:19:56 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-22T09:25:09 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-22T09:27:58 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Read error: Connection reset by peer] 2019-10-22T09:30:42 < R2COM> dongs yo 2019-10-22T09:31:07 < R2COM> dongs which VR is better vive cosmos or vive index? 2019-10-22T09:32:49 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-22T09:38:25 -!- jly [uid355225@gateway/web/irccloud.com/x-wbvbiqnkvlumvjaa] has joined ##stm32 2019-10-22T09:39:32 < Haohmaru> welcome to teh real world, mr jly 2019-10-22T09:39:48 < jly> gooday sport 2019-10-22T09:40:12 < jly> sup Haohmaru 2019-10-22T09:40:22 < Haohmaru> not much 2019-10-22T09:53:51 < Haohmaru> soldering shizzle 2019-10-22T09:54:44 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-22T10:03:54 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-22T10:13:11 < Steffanx> mawk: not sure where you found that, but sounds pretty usual for a room in delft 2019-10-22T10:13:20 < dongs> R2COM: vive pro 2019-10-22T10:13:27 < Steffanx> Especially since the city is flooded by students 2019-10-22T10:26:04 < R2COM> hm index has same resol;ution but can do 120hz instead of 90hz vive pro 2019-10-22T10:32:19 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-22T10:43:21 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-22T10:55:18 -!- zyp [zyp@zyp.no] has joined ##stm32 2019-10-22T11:58:09 -!- jly [uid355225@gateway/web/irccloud.com/x-wbvbiqnkvlumvjaa] has quit [Quit: Connection closed for inactivity] 2019-10-22T12:03:30 -!- Jan- [~IceChat9@87.112.99.183] has quit [Quit: Light travels faster then sound, which is why some people appear bright, until you hear them speak] 2019-10-22T12:03:42 -!- Jan- [~IceChat9@87.112.99.183] has joined ##stm32 2019-10-22T12:04:04 -!- Jan- [~IceChat9@87.112.99.183] has quit [Client Quit] 2019-10-22T12:04:28 -!- Jan- [~IceChat9@87.112.99.183] has joined ##stm32 2019-10-22T12:11:01 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-22T13:02:11 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-22T13:19:10 -!- con3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-22T13:20:10 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-22T13:21:24 < karlp> man, st.com slow today 2019-10-22T13:22:58 < dongs> just today? 2019-10-22T13:23:26 < dongs> waht do yall do when you have a while(1) in some hardware shit with rtos do you just while(waiting_for_register) { yield(); } that shit? 2019-10-22T13:23:27 < karlp> nah, just worse 2019-10-22T13:23:30 < dongs> or is there more clever shit to do 2019-10-22T13:25:28 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-22T13:25:32 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-22T13:25:52 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-22T13:27:50 < zyp> dongs, the clever way is to suspend the task in some way and use an interrupt to trigger resuming it 2019-10-22T13:27:55 < jpa-> dongs: for short waits, i wouldn't bother to yield() even; for longer waits, wait on semaphore/event and use interrupt to set it 2019-10-22T13:28:13 < dongs> ah. ok zyp + jpa have the right answer 2019-10-22T13:28:14 < dongs> thanks 2019-10-22T13:28:23 < dongs> so something like DMA_TC setting the shit right 2019-10-22T13:29:06 < zyp> whatever you're waiting for 2019-10-22T13:29:09 < dongs> right 2019-10-22T13:30:16 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-22T13:30:35 < zyp> the net task in my workshit is woken either by rx complete, tx complete or a timer signalling it's time to start the next update cycle 2019-10-22T13:31:18 < zyp> eventually I'm gonna have to make it wake up at timeouts as well 2019-10-22T13:33:34 < karlp> do you bother using different task notification values for that? or just "wakup boi" 2019-10-22T13:35:01 < zyp> yes, I use the value as a bitfield with one bit per event type 2019-10-22T13:38:39 < zyp> https://paste.jvnv.net/view/Pjlgg 2019-10-22T13:40:49 < karlp> have you got ethertype as littleendian or something there? 2019-10-22T13:40:54 < karlp> or am I reading too much into that? 2019-10-22T13:40:57 < zyp> yes 2019-10-22T13:41:10 < karlp> that would drive me mad :) 2019-10-22T13:41:31 < zyp> yeah, newer code (ip/udp) bitswaps before switching 2019-10-22T13:41:54 < zyp> this code is pretty old and pending rewrite to improve the buffer handling 2019-10-22T13:42:08 < zyp> right now I'm only using two buffer descriptors; one for TX and one for RX 2019-10-22T13:42:17 < karlp> network order is BE right? and it's just ending up LE because you're accessing a 16bit value inside the rx bufffer? 2019-10-22T13:42:27 < zyp> yes 2019-10-22T13:44:16 < karlp> love undocumented registers: https://imgur.com/a/HJH4X5o 2019-10-22T13:46:28 < dongs> idonno about undocumented that sounds like flash unlock key type shit 2019-10-22T13:47:30 < karlp> no, this is related to magical temperature compensation 2019-10-22T13:47:53 < karlp> the 55aa sure, that's the same unlock code they use elsewhere, 2019-10-22T13:48:07 < karlp> but the registers 216 and 219 aren't documented anywhere. 2019-10-22T13:48:16 < dongs> ah 2019-10-22T13:49:50 < karlp> there's 16 registers I can write to to do temperature compensation in 8° ranges, based on teh internal sensor though, so that's cute. 2019-10-22T13:52:22 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-22T14:00:46 -!- kakinoob [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-22T14:06:03 < Haohmaru> karlp wut chip is that? 2019-10-22T14:09:44 < karlp> atm90e32 2019-10-22T14:10:05 < karlp> with it's shitty 1Mhz spi limit. 2019-10-22T14:10:16 < karlp> feels like i2c slow. 2019-10-22T14:10:36 < Haohmaru> ah, that thing 2019-10-22T14:10:53 < Haohmaru> what if u overclock it a lil bit ;P~ 2019-10-22T14:11:06 < karlp> hten you get garbage back. their specs are right, max spi is 1Mhz :) 2019-10-22T14:11:27 < Haohmaru> rub a very hot pepper on its thermal pad right before you solder it 2019-10-22T14:11:38 < karlp> ok, that was funny :) 2019-10-22T14:12:05 < Haohmaru> that's also how u overclock donkeys 2019-10-22T14:27:17 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 240 seconds] 2019-10-22T14:38:42 < dolanbatar> hi! is connecting VBAT on stm32l433 mandatory? even if there is not battery and i don't care about backup registers or rtc? 2019-10-22T14:38:46 < dolanbatar> please :) 2019-10-22T14:38:47 -!- con3|2 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-22T14:38:48 -!- con3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-22T14:39:12 < Haohmaru> doesn't the datasheet say it? 2019-10-22T14:39:36 < dolanbatar> nope, or i am blind 2019-10-22T14:45:33 < dolanbatar> okay, there is min and max for vbat in operating conditions 1.55 - 3.6V, so? connect it? i didn't do so on stm32g474 and all is fine 2019-10-22T14:47:05 < Haohmaru> put a little *shrug* on the silkscreen right next to the pin? 2019-10-22T14:48:54 < dolanbatar> :D 2019-10-22T14:48:56 < srk> :D 2019-10-22T14:49:13 < jadew> jeez.. I just had to take an electronics test to join a forum :/ 2019-10-22T14:49:22 < jadew> in russian 2019-10-22T14:49:27 < Haohmaru> wut kind of test? 2019-10-22T14:49:33 < jadew> was worried google translate will screw me over 2019-10-22T14:49:41 < Haohmaru> don't tell me it's resistor color codes 2019-10-22T14:49:42 < jadew> Haohmaru, general electronics repair test 2019-10-22T14:50:01 -!- con3|2 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-22T14:50:05 < Haohmaru> i avoid .ru websites 2019-10-22T14:50:07 -!- con3 [~kvirc@154.119.40.228] has joined ##stm32 2019-10-22T14:50:19 < jadew> well, I need some firmware from them 2019-10-22T14:50:20 < Haohmaru> wut'cha doin there 2019-10-22T14:50:26 < jadew> my TV is broken 2019-10-22T14:50:28 < Haohmaru> gud luck 2019-10-22T14:50:35 < Haohmaru> toss it out 2019-10-22T14:51:02 < Haohmaru> extract teh useful bits from it and toss the rest of it out ;P~ 2019-10-22T14:51:15 < jadew> I just bought a chromecast :/ 2019-10-22T14:51:29 < Haohmaru> buy some vodka with teh moniez, get drunk, and forget about teh TV 2019-10-22T14:51:31 < jadew> haven't used this TV in maybe a year 2019-10-22T14:51:33 < Haohmaru> TV? what TV? 2019-10-22T14:51:42 < jadew> worked for a bit and then it started rebooting 2019-10-22T14:51:54 < Haohmaru> is it one of those clev0r TVs? 2019-10-22T14:52:05 < Haohmaru> with crapdroid and webbrowz0rs? 2019-10-22T14:52:15 < jadew> no, it has some custom stuff 2019-10-22T14:52:31 < Haohmaru> okay, so it's not uber eggspensive then 2019-10-22T14:52:37 < dolanbatar> hmmmm, datasheet says the same thing about stm32g474, so maybe i should connect it there too :D 2019-10-22T14:53:28 < Haohmaru> i do remember on some chip (was it ATSAME?) where there was a BAT pin you had to at least connect a decoupling cap.. 100nF and maybe also 10uF or some such 2019-10-22T14:53:51 < Haohmaru> but i could be talking bullsh*t here 2019-10-22T14:53:58 < Haohmaru> memory iz rusty 2019-10-22T14:55:09 < dolanbatar> weirdest thing about this is that there is a switch on vbat drawn in schematic in datasheet 2019-10-22T14:55:50 < dolanbatar> so i though it would "get activated" only on power out 2019-10-22T15:01:54 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-22T15:03:49 < kakimir32> o/ 2019-10-22T15:03:54 < dolanbatar> \o 2019-10-22T15:19:58 < kakimir32> I had weird dream I bought ktm enduro bike 400cc 2019-10-22T15:22:03 < kakimir32> why 2019-10-22T15:25:55 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-22T15:29:14 < Haohmaru> u have dreams 2019-10-22T15:29:22 < Haohmaru> give me some 2019-10-22T15:29:25 < Haohmaru> no wait, don't 2019-10-22T15:29:32 < Haohmaru> i need muh own 2019-10-22T15:29:43 < kakimir64> trick is to sleep a lot 2019-10-22T15:29:47 < con3> i just have nightmares 2019-10-22T15:29:59 < Haohmaru> nightmares too 2019-10-22T15:30:03 < Haohmaru> anything 2019-10-22T15:30:30 < Haohmaru> Sir Sleep A Lot 2019-10-22T16:38:26 < doomba> Sir Sleep A Lot - Baby Got Nap 2019-10-22T16:38:56 < Haohmaru> i like big pillowz and i can not lie 2019-10-22T16:41:14 < doomba> and when the kitty walks in with a itty bitty face and puts razor sharp claws in your face you get SPRRRRUNG!!! it just cut your neck. now there's red stuff all over fluff. You try to breathe but you're choking on liquid. Now you're dead and you donno what hit ya. 2019-10-22T16:57:44 < Jan-> but kitties are cute :/ 2019-10-22T16:58:05 < Jan-> there is a cat around here somewhere and she would never stick her claws in without good reason 2019-10-22T16:58:21 < Jan-> although to be fair "good reason" might just mean "I was stretching." 2019-10-22T17:33:16 -!- con3 [~kvirc@154.119.40.228] has quit [Read error: Connection reset by peer] 2019-10-22T17:35:43 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-22T17:37:28 < jadew> any ideas on how to program a NAND flash without a programmer? 2019-10-22T17:37:55 < Haohmaru> with telekinesis 2019-10-22T17:38:21 < mawk> so, without the spi controller jadew ? 2019-10-22T17:38:23 < jadew> I have one of those TL866A but it doesn't work with this particular chip 2019-10-22T17:38:27 < mawk> or whatever kind of controller 2019-10-22T17:38:49 < jadew> I don't know... I think they can be programmed via JTAG, but I'm looking for confirmation 2019-10-22T17:39:07 < Haohmaru> he could attach 3 buttenz with pull resistors and punch in some SPI ;P~ 2019-10-22T17:39:27 < jadew> heh 2019-10-22T17:39:30 < mawk> if there is a spi controller you just use a random stm32 2019-10-22T17:39:50 < mawk> otherwise, well you implement the open nand flash interface or whatever, but I don't know the freq of that if you can bitbang it or not 2019-10-22T17:40:14 < jadew> I actually made a reader/writer a while ago... 2019-10-22T17:40:25 < jadew> but it was on a breadboard and I don't want to re-wire that shit up 2019-10-22T17:41:12 < jadew> it's why I bought this silly programmer 2019-10-22T17:42:12 < jadew> damn it... 2019-10-22T17:42:17 < mawk> the only difficulty I see is that NAND flash isn't always 3.3V logic 2019-10-22T17:42:23 < mawk> but short from that you can maybe bitbang it 2019-10-22T17:42:29 < jadew> I guess I'll have to wait for a week until the pre-programmed one arrives 2019-10-22T17:42:30 < mawk> but if there is a spi controller to your chip it's very very easy 2019-10-22T17:42:43 < mawk> just get a raspberry pi and a program like flashrom 2019-10-22T17:43:12 < Haohmaru> mm, crapberry 2019-10-22T17:46:30 < Steffanx> So found something outside delft yet, mawk? 2019-10-22T17:46:50 < mawk> I'm still looking inside delft Steffanx I want to be in the center 2019-10-22T17:46:56 < Steffanx> Hah, good luck 2019-10-22T17:47:04 < mawk> I don't mind if I pay a bit more 2019-10-22T17:47:08 < Steffanx> Competing with the shitload of students 2019-10-22T17:47:13 < mawk> yeah 2019-10-22T17:47:21 < mawk> but at least I have a real job and good guarantees 2019-10-22T17:47:25 < mawk> I'll beat all the students 2019-10-22T17:47:44 < Steffanx> Some have rich daddies 2019-10-22T17:47:49 < mawk> across the appartment of my parent there is the defltse studient society full of snobs in suit and tie that party every night 2019-10-22T17:48:13 < Steffanx> Hah the worst kind 2019-10-22T17:48:15 < mawk> I saw a very drunk one fall off the balcony, the ambulance came 2019-10-22T17:48:56 < Jan-> has anyone here ever set up a build environment from scratch on windows 2019-10-22T17:49:03 < Jan-> just to the point of having a working main() 2019-10-22T17:49:03 < Steffanx> You should go to the plattelaaand 2019-10-22T17:49:08 < mawk> yes Jan- 2019-10-22T17:49:21 < mawk> using openocd, gcc from arm 2019-10-22T17:49:25 < Steffanx> No students on the plattelaand. Only farmers 2019-10-22T17:49:29 < con3> i live in a student complex. i cant describe the horror of this place 2019-10-22T17:49:32 < Jan-> I talked to someone who had but not with the chip I have 2019-10-22T17:49:45 < mawk> but you already flashed it Jan- 2019-10-22T17:49:53 < mawk> but you keep changing your toolchain 2019-10-22T17:50:03 < Jan-> well I'm not particularly aiming to change the toolchain 2019-10-22T17:50:03 < Steffanx> It's not from scratch at all Jan- . 2019-10-22T17:50:05 < Haohmaru> i think the first eggzotic platform i smashed my head with was the dumb xtensa-lx106 or whatever it was 2019-10-22T17:50:15 < Haohmaru> i set it up to build with Code::Blocks 2019-10-22T17:50:18 < Jan-> but different libraries tend to come with different ones 2019-10-22T17:50:19 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-22T17:50:20 < Steffanx> Just get the pre compiled binaries 2019-10-22T17:50:34 < mawk> which library do you use now Jan- ? cube ? 2019-10-22T17:50:39 < Haohmaru> wasn't easy, i didn't know sh*t, all i knew was i ain't buying no makefiles 2019-10-22T17:50:54 < Jan-> honestly I'd just be happy to start with stm32f407xx.h and go from there 2019-10-22T17:50:56 < Steffanx> Doesnt the blocks use makefiles? 2019-10-22T17:51:14 < Jan-> I found someone in this channel who had given up on make and just wrote a shell script to do it 2019-10-22T17:51:19 < Jan-> but it isn't simple 2019-10-22T17:51:23 < mawk> yeah don't listen to that person 2019-10-22T17:51:26 < mawk> make is the simplest thing ever 2019-10-22T17:51:28 < Haohmaru> Steffanx it has a build system 2019-10-22T17:51:39 < Jan-> honestly it made more sense to me than make 2019-10-22T17:51:43 < Haohmaru> yes, you could tell it "nah i do have a makefile, use that" 2019-10-22T17:51:44 < mawk> why ?? 2019-10-22T17:51:52 < mawk> make is the reference specialized tool for building code 2019-10-22T17:52:06 < Jan-> yes, but the config is a nightmare 2019-10-22T17:52:09 < mawk> to build a single .c into an executable you don't even need a single line in your makefile ! 2019-10-22T17:52:10 < mawk> make file.c 2019-10-22T17:52:14 < zyp> make is crap 2019-10-22T17:52:21 < Haohmaru> Jan- i wouldn't use a shell script 2019-10-22T17:52:26 < Steffanx> You mean it generates a make file and runs make? 2019-10-22T17:52:40 < Jan-> neither would I, I'd write a little thing in javascript 2019-10-22T17:52:41 < Steffanx> Should we all use the scons, zyp? 2019-10-22T17:52:45 < Jan-> but only because I know javascript 2019-10-22T17:52:51 < zyp> Steffanx, maybe? 2019-10-22T17:52:56 < mawk> so Jan- , which lib ? cube ? 2019-10-22T17:52:56 < Haohmaru> Jan- nor that 2019-10-22T17:53:32 < Haohmaru> Jan- i don't talk java, but mIRC script, and i wouldn't write a script for compiling my C++ codez 2019-10-22T17:54:16 < zyp> Steffanx, if not scons, I'd look into other modern alternatives, perhaps meson with ninja as a backend 2019-10-22T17:55:01 < mawk> Jan- :( 2019-10-22T17:55:05 < mawk> which lib are you using 2019-10-22T17:56:27 < mawk> if you use cube you already wrote a main() and it was working 2019-10-22T17:56:42 < mawk> you were just complaining about the lenghty function names but that's not a real complaint, HAL works 2019-10-22T17:57:58 < Steffanx> Until it doesnt. 2019-10-22T17:58:09 < mawk> shhh 2019-10-22T17:58:16 < mawk> nobody needs to know 2019-10-22T17:58:53 < Jan-> If there's some way I can use mbed and just not have it include any of the libraries that would be top notch 2019-10-22T17:59:07 < mawk> why mbed ?? 2019-10-22T17:59:08 < mawk> :( 2019-10-22T17:59:12 < mawk> it's even worse 2019-10-22T17:59:19 < Jan-> not LGPL 2019-10-22T17:59:30 < Haohmaru> https://i.imgur.com/BVMXarc.png 2019-10-22T17:59:30 < Jan-> the API is really c++ ish 2019-10-22T18:00:26 < Jan-> which is not my favourite 2019-10-22T18:01:14 < mawk> HAL is BSD 2019-10-22T18:01:30 < mawk> there is no reason to use mbed 2019-10-22T18:01:33 < mawk> HAL is C ! 2019-10-22T18:01:39 < karlp> just don't use any of the libraries then... 2019-10-22T18:01:51 < mawk> this is ugly Haohmaru 2019-10-22T18:01:58 < srk> Jan-: you can also try https://modm.io/ 2019-10-22T18:01:58 < mawk> are you on windows 95 ? 2019-10-22T18:02:14 < Haohmaru> yes i'm on crapdows 2019-10-22T18:02:16 < srk> not sure about MPL thought 2019-10-22T18:02:30 < karlp> srk: what, no-one liked haskell*ivory so you're just generating more stuff now? 2019-10-22T18:02:54 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-22T18:03:01 < Jan-> srk thanks I'll check it out right now 2019-10-22T18:03:40 < mawk> HAL is BSD-3, Jan- 2019-10-22T18:04:10 < Jan-> HAL as in the one that goes "HAL_GPIO_WritePin(GPIOA, GPIO_PIN_6, GPIO_PIN_RESET);" 2019-10-22T18:04:47 < mawk> so what lol 2019-10-22T18:04:57 < mawk> that's a really bad reason to not use it 2019-10-22T18:05:11 < mawk> there are valid criticizing points, but this one is really not one of them 2019-10-22T18:05:25 < mawk> just do your one-letter-macros if you want to write 20 characters lines 2019-10-22T18:05:28 < Jan-> that was designed by someone who'd say "mellifluous felicitations upon the pre-meridian period of our physical coincidence" rather than "Good morning." 2019-10-22T18:05:34 < mawk> W(A, 6, R); 2019-10-22T18:06:23 < mawk> lol 2019-10-22T18:06:34 < Jan-> I guess I'm just used to PORTA &= (1<<6); 2019-10-22T18:06:37 < mawk> this makes self-documenting code and very easy to use library even in the absence of docs 2019-10-22T18:06:52 < mawk> just read the .h file, the method names are self-explanatory 2019-10-22T18:08:06 < srk> karlp: no, modm is not my project. I'm still working on device support for ivory tho. added automated AF and I2Cv2 timings calculators recently, about to tackle DMA :) 2019-10-22T18:08:44 < srk> ivory/tower is not for everyone, especially not for windows folks as it now requires nix and two different compilers :) 2019-10-22T18:09:06 < Jan-> Apparently you can sort of do that on STM32. 2019-10-22T18:09:15 < karlp> you can, you've been told how repeatedly... 2019-10-22T18:09:18 < srk> getting blink working is few commands away and probably easier than most toolchains :) 2019-10-22T18:09:18 < Jan-> uint32_t* ODR_B = (volatile uint32_t*) 0x40020414; *ODR_B |= (1<<13); 2019-10-22T18:09:32 < karlp> and the ODR_B type defintions are in the cmsis header files.. 2019-10-22T18:09:53 < karlp> so it's, "Jan- | I guess I'm just used to PORTA &= (1<<6);" just like you'd like. 2019-10-22T18:11:02 < Jan-> sure karl, it's just hard to create a setup where you can actually USE that. 2019-10-22T18:11:20 < zyp> Jan-, no, it's not hard, you're just bad at it 2019-10-22T18:11:25 < karlp> ^^ ^^ 2019-10-22T18:11:37 < Jan-> makefiles guys :( 2019-10-22T18:11:53 < srk> Jan-: https://vivonomicon.com/2018/04/22/bare-metal-stm32-programming-part-3-leds-and-buttons/ 2019-10-22T18:11:54 < zyp> makefiles has nothing to do with it 2019-10-22T18:11:56 < karlp> use catphishs's make.sh if you like. 2019-10-22T18:11:58 < karlp> or not make. 2019-10-22T18:12:04 < karlp> it's still "include this file, compile it" 2019-10-22T18:12:11 < mawk> why would you do that Jan- tho ? 2019-10-22T18:12:18 < Jan-> I like catphish's approach 2019-10-22T18:12:22 < mawk> no you don't 2019-10-22T18:12:22 < Jan-> it's at least understandable 2019-10-22T18:12:28 < mawk> you don't know how to do it, and he knows 2019-10-22T18:12:33 < mawk> just use a makefile until you learn 2019-10-22T18:12:36 < Jan-> well I don't. 2019-10-22T18:12:45 < Jan-> it needs a lot of changes for the board I have. 2019-10-22T18:12:52 < mawk> if you want to you can study cube's makefiles 2019-10-22T18:13:22 < karlp> srk's link is perhaps pretty good for how jan claims they want to work. 2019-10-22T18:13:26 < srk> mawk: proly ugly as fuck :)) 2019-10-22T18:13:50 < karlp> single makefile, much like the much copied winavr makefile 2019-10-22T18:14:09 < Jan-> people have said that to me, just use a preexisting one 2019-10-22T18:14:17 < Jan-> but it's kinda hard to find one that's exactly intended for what I'm doing 2019-10-22T18:14:37 < mawk> it has a lot of annotations srk but yeah it's a bit ugly (and was broken in previous cube version, I constantly had to delete it before regenerate code) 2019-10-22T18:15:01 < mawk> why hard ?? you just made it work for your board using cubeIDE ! 2019-10-22T18:15:08 < mawk> copy paste it, it worked 2019-10-22T18:15:13 < Jan-> copy paste WHAT 2019-10-22T18:15:19 < mawk> the makefile you made using cubeIDE 2019-10-22T18:15:29 < Jan-> cubeide makes makefiles? 2019-10-22T18:15:49 < karlp> did you ever read the windows you clicked on? 2019-10-22T18:15:54 < mawk> yes probably 2019-10-22T18:16:01 < karlp> it makes a project for you and you can choose what sort of project to make, one of them is makefiles 2019-10-22T18:16:06 < zyp> fuck the makefile, just make a shell script if you don't wanna learn make 2019-10-22T18:16:14 < karlp> that's what catphish gve them, make.sh 2019-10-22T18:16:18 < zyp> it's not like it makes a difference for a small project 2019-10-22T18:16:19 < karlp> it's just trolling at this point 2019-10-22T18:16:55 < zyp> the advantage of a makefile over a shell script is that it allows you to rebuild only the parts that changed, saving you time 2019-10-22T18:16:57 < Jan-> zyp I would totally do that. 2019-10-22T18:17:12 < zyp> but there's no point saving time on a small project that builds in less than a second anyway 2019-10-22T18:17:15 < Jan-> well I'd write a thing to take the md5 hash of the file contents and see if it changed. 2019-10-22T18:17:21 < Jan-> I've done that before it's fine 2019-10-22T18:17:24 < zyp> just rebuild everything 2019-10-22T18:17:27 < mawk> a makefile for a small project is easier to write than a shell script, in my opinion 2019-10-22T18:17:50 < mawk> for instance for my last android project https://paste.serveur.io/DOg0OdGB.mak 2019-10-22T18:17:58 < zyp> pff, for a small project you can just do gcc *.c -o foo.elf 2019-10-22T18:18:06 < zyp> no need to separate compilation and linking 2019-10-22T18:18:08 < mawk> the only real rule is $(OUT): $(OBJ) 2019-10-22T18:18:11 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-22T18:18:19 < mawk> the rest is just sugar 2019-10-22T18:19:06 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-22T18:19:39 < Jan-> ok so there is a file called "Makefile" in C:\Users\JanH\STM32CubeIDE\workspace_1.0.2\Test2\Debug 2019-10-22T18:20:05 < mawk> yes, and if you don't want to use cubeIDE just take the .mx project, open it in standalone stm32cubeMX and it will generate you a makefile 2019-10-22T18:20:12 < mawk> either case it works the same 2019-10-22T18:21:19 < Jan-> I notice that this project has a lot of the files mentioned it catphish's script 2019-10-22T18:21:48 < Jan-> the startup file (startup_stm32f407vetx.s) has a lot of assembly 2019-10-22T18:22:00 < mawk> yes, just use it vanilla 2019-10-22T18:22:03 < mawk> you will never have to edit it 2019-10-22T18:22:11 < zyp> it's probably full of unnecessary shit :) 2019-10-22T18:22:28 < mawk> it just has the interrupt vector 2019-10-22T18:22:32 < mawk> and a few lines for reset_handler 2019-10-22T18:23:11 < Jan-> https://pastebin.com/7VFExRFE 2019-10-22T18:23:12 < Jan-> there 2019-10-22T18:23:13 < Jan-> that 2019-10-22T18:23:42 < mawk> yes 2019-10-22T18:23:51 < zyp> all that stuff could be done in C, doesn't have to be assembly 2019-10-22T18:24:34 < mawk> it zeroes out the .bss section for your uninitialized global variables, it copies the .data section to SRAM, and calls your main function 2019-10-22T18:24:37 < mawk> it doesn't do much 2019-10-22T18:24:53 < Jan-> clocks? 2019-10-22T18:24:59 < mawk> also it defines the interrupt vector, and sets all the interrupt handlers as weak function pointing to a default handler 2019-10-22T18:25:02 < mawk> clock is later, it's in C 2019-10-22T18:25:34 < mawk> (if you open your project in standalone stm32cubeMX it will do an even smaller Makefile, in just one file) 2019-10-22T18:26:24 <@englishman> lol makefails 2019-10-22T18:26:27 < mawk> but you can use this one, it's fine 2019-10-22T18:26:31 <@englishman> all this opensores chat 2019-10-22T18:26:35 < Jan-> englishman: No kidding :/ 2019-10-22T18:26:39 * Jan- hands englishman a cookie. 2019-10-22T18:26:40 < mawk> make englishman.o 2019-10-22T18:26:47 <@englishman> instead of clicking new project in cube and shitting out a keilproj 2019-10-22T18:27:08 <@englishman> you could be in production by now 2019-10-22T18:27:24 < Jan-> there's quite a lot of coding to do in the meantime englishman :) 2019-10-22T18:27:45 <@englishman> when will that start 2019-10-22T18:28:14 < Jan-> when I have the prototype emitters probably 2019-10-22T18:28:31 < Jan-> although! I could probably try to get things like the serial I/O working for debug messages and suchlike. 2019-10-22T18:29:02 < Haohmaru> englishman by the time Jan- starts speeking in dongsish dialect, that's when 2019-10-22T18:29:15 < Haohmaru> so, still nope 2019-10-22T18:30:35 <@englishman> https://www.bell-labs.com/unix50/event/ 2019-10-22T18:32:47 < karlp> englishman: we already suggested keil, they didn't want that either :| 2019-10-22T18:33:13 <@englishman> lol 2019-10-22T18:33:25 <@englishman> make ##stm32 pro again 2019-10-22T18:33:57 < Jan-> I think keil basically ends up looking much like cube from where I'm sitting 2019-10-22T18:34:24 < Jan-> um, if I start adding more code files into a cube project, is it going to stuff them all full of these /* USER CODE BEGIN */ type comments 2019-10-22T18:34:27 < Jan-> or is that just main.c 2019-10-22T18:35:21 < Jan-> modm says that stm32f4 is "very well tested." 2019-10-22T18:35:23 * Jan- taps chin 2019-10-22T18:36:29 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-22T18:47:22 < Jan-> I quite like the look of modm 2019-10-22T18:47:30 < Jan-> they seem big on making sure everything is up to date and tested 2019-10-22T18:47:44 < mawk> hal too lol, it comes from st themselves 2019-10-22T18:47:54 < Jan-> yeah I'm sort of leaning back toward it 2019-10-22T18:53:11 < Jan-> I guess in the long run it's probably like c# 2019-10-22T18:53:19 < Jan-> if you use the microsoft tools it's probably best supported 2019-10-22T18:54:01 < srk> haven't tried it myself but it looks solid, especially if you want C++ interface. HAL is an abomination only good for use with cubemx .. 2019-10-22T18:54:18 < Jan-> well the hardware access layer in the cube ide is the thing I really didn't like 2019-10-22T18:54:20 < Jan-> it's super wordy 2019-10-22T18:54:24 < Jan-> why don't you like it 2019-10-22T18:56:59 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-22T18:57:49 < Steffanx> Is it bad that I remember your avr adventures started kinda in the same way, Jan- ? 2019-10-22T18:58:39 < Jan-> AVR wasn't this bad 2019-10-22T18:58:54 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-22T18:58:55 < Jan-> mainly because atmel's software was pretty ok 2019-10-22T18:58:57 < Jan-> nobody hated it 2019-10-22T18:59:03 < Jan-> but everyone shits on cube 2019-10-22T18:59:10 < mawk> not me ! 2019-10-22T18:59:15 < mawk> I use it 2019-10-22T18:59:29 < Jan-> everyone except mawk shits on cube. 2019-10-22T18:59:31 < mawk> :( 2019-10-22T18:59:45 < Jan-> and me I guess 2019-10-22T19:00:03 < Jan-> I just really hate the gui configurator and the code it generates is horriffic 2019-10-22T19:00:52 < Steffanx> I dont shit on cube either. The pin planner is nice 2019-10-22T19:01:17 < Steffanx> And LL stuff in the HAL isnt that bad. Worry sure, but.. I dont care 2019-10-22T19:01:27 < Steffanx> Wordy* 2019-10-22T19:01:31 < srk> Steffanx: yeah, pin planner is cool :) 2019-10-22T19:01:51 < srk> clock trees as well, the only two things I use it for 2019-10-22T19:04:06 < mawk> why horrific ? just a bit wordy sure, and it has tags like START USER CODE but it has to be that way, else you can regenerate code afterwards 2019-10-22T19:04:29 < Jan-> if I add more source files do they all end up like that 2019-10-22T19:04:54 < mawk> what do you mean by add ? 2019-10-22T19:05:00 < mawk> if you add your own source Cube won't touch them 2019-10-22T19:05:07 < Jan-> well I'm presumably not going to write everything in main.c 2019-10-22T19:05:08 < mawk> you can write them however you like 2019-10-22T19:06:40 < Jan-> I find that after a bit you end up not having much to do with the hardware nayway 2019-10-22T19:06:44 < Jan-> you're just writing your own stuff 2019-10-22T19:08:13 < Jan-> Oh. Also. This board has a USB socket on it. Is there some library that will make that USB port active and let us use it for code upload and debug like an arduino? 2019-10-22T19:08:19 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-22T19:08:21 < Jan-> Someone must have done that 2019-10-22T19:10:40 < Steffanx> Forget about tarduino. ;) 2019-10-22T19:10:46 < Ecco> Hi :) 2019-10-22T19:11:05 < Jan-> Steffanx: it's just a pain to have two usb ports, one for upload and one for seria. 2019-10-22T19:11:16 < Ecco> Imagine a low-end STM32-based device *without* any wireless connectivity 2019-10-22T19:11:27 < Steffanx> Can an tarduino even be debugged? I'm not talking about printf debugging 2019-10-22T19:11:30 < Ecco> question -> What would be the cheapest way to allow end-user to update the firmware? 2019-10-22T19:12:17 < Steffanx> Use the usb bootloader Jan- wants 2019-10-22T19:12:23 < Jan-> I was going to say Ecco 2019-10-22T19:12:27 < Jan-> for that you really do want USB 2019-10-22T19:12:32 < Ecco> well, usb is expensive 2019-10-22T19:12:37 < Ecco> -> Requires USB IP 2019-10-22T19:12:44 < Jan-> is it? 2019-10-22T19:12:46 < Ecco> -> Requires crystal (on most STM32s) 2019-10-22T19:12:51 < Ecco> -> Requires connector 2019-10-22T19:13:10 * karlp closes teh window again. 2019-10-22T19:13:18 < Steffanx> Hi karlp 2019-10-22T19:14:18 < Steffanx> I cannot think of another interface everyone and his mother has, ex 2019-10-22T19:14:27 < Steffanx> Ecco: * 2019-10-22T19:14:28 < Ecco> yeah 2019-10-22T19:14:30 < karlp> tada, "I agree, I will create an enhancement request for this feedback, thank you for bringing this to our attention." 2019-10-22T19:14:36 < Steffanx> Damn auto complete 2019-10-22T19:14:53 < karlp> convinced arm that the docs werne't good enough 2019-10-22T19:22:02 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-22T19:24:33 < Steffanx> Ecco: You can always go for the pcb style usb connector, but.. meh 2019-10-22T19:28:54 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-22T19:32:50 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Client Quit] 2019-10-22T19:37:17 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-22T19:43:19 < srk> Steffanx: there are some atmegas with JTAG 2019-10-22T19:43:39 < srk> but most ppl treat them as black boxes with print debugging 2019-10-22T19:44:52 < Steffanx> because the arduino ide doesnt support it. 2019-10-22T19:45:05 < Steffanx> they are working on something though, something official that is 2019-10-22T19:45:11 < srk> haha 2019-10-22T19:45:22 < srk> would be nice if they at least include simulator 2019-10-22T19:45:32 < srk> as simulation for atmegas is quite good due to precise intrusction timing 2019-10-22T19:45:58 < antto> something has unprecise instruction timing? 2019-10-22T19:46:06 < antto> pic? >:) 2019-10-22T19:46:10 < Steffanx> yeah, cant be that hard to add something like simavr 2019-10-22T19:46:26 < srk> anything with speculative executions, pipelines, flash wait cycles :) 2019-10-22T19:46:31 < srk> *execution 2019-10-22T19:46:42 < antto> well that's on teh thicc-er chips bruh 2019-10-22T19:46:53 < antto> are there any 8bit chips with any of that shizzle? 2019-10-22T19:47:07 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-22T19:47:21 < srk> dunno 2019-10-22T19:48:45 < antto> i remember writing a mIRC script to read the assembly dump of muh program and count the instruction cycles of the main loop in each possible branch x_x 2019-10-22T19:49:01 < antto> that sh*t wosn't fun 2019-10-22T19:49:01 < Steffanx> mirc script -_- 2019-10-22T19:49:37 < antto> that's not the problem ;] 2019-10-22T19:49:47 < srk> ;D 2019-10-22T19:49:55 * antto is a mIRC script mast0r 2019-10-22T20:10:39 < kakimir64> some guise still use mirc 2019-10-22T20:20:25 < kakimir64> and bad charsets 2019-10-22T20:21:52 < antto> pls 2019-10-22T20:22:06 < antto> i got FixedSys here 2019-10-22T20:27:13 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-22T20:29:42 < kakimir64> in finnish it's actual problem because special alphabet 2019-10-22T20:30:16 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-22T20:32:19 < antto> speshul 2019-10-22T20:32:51 < kakimir64> is there a macro to return maximum value of type? 2019-10-22T20:33:18 < antto> numeric_limits, oh wait 2019-10-22T20:33:24 < Jan-> is it evil if I rename HAL_GPIO_WritePin 2019-10-22T20:33:25 < antto> u wish 2019-10-22T20:33:26 * Jan- cackles 2019-10-22T20:33:29 * Jan- abuses refactoring 2019-10-22T20:33:46 < antto> oh bruh x_x 2019-10-22T20:33:50 < kakimir64> antto: speak to me in C 2019-10-22T20:34:12 < kakimir64> I don't understand this slang 2019-10-22T20:34:25 < antto> iz iz C with two pluses 2019-10-22T20:36:22 < kakimir64> no comprehendo 2019-10-22T20:36:54 < antto> kakimir64 well what are you akchually trying to do? 2019-10-22T20:37:11 < kakimir64> I have macro to select type for variable X 2019-10-22T20:37:13 < antto> don't tell me you're using sh*t like "int" 2019-10-22T20:37:14 < kakimir64> and 2019-10-22T20:37:37 < kakimir64> maximum value of type is used in logic of my functions 2019-10-22T20:37:45 < Jan-> OK so now I can't make cube work either :/ 2019-10-22T20:38:00 < kakimir64> work work Jan- 2019-10-22T20:38:09 < Jan-> "Error in STM32CubeProgrammer" followed by "Target is not responding, retrying..." forever. 2019-10-22T20:38:30 < kakimir64> your target is not responding 2019-10-22T20:38:45 < Jan-> Oh! Right. 2019-10-22T20:38:48 < kakimir64> did you touch the SWD pins? 2019-10-22T20:38:49 < Jan-> ... no wait 2019-10-22T20:38:52 < antto> well, it smells like you're trying to have flexiburu code, with macros 2019-10-22T20:38:59 < kakimir64> don't touch SWD pins 2019-10-22T20:39:01 < antto> kakimir64 just use templates bruh 2019-10-22T20:39:10 < antto> forget them macros 2019-10-22T20:39:57 < kakimir64> macros are perfect 2019-10-22T20:41:04 < antto> i wouldn't do anything semi complex in C 2019-10-22T20:41:12 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has quit [Ping timeout: 246 seconds] 2019-10-22T20:41:13 < kakimir64> Jan-: did you touch SWD pins? 2019-10-22T20:41:17 < kakimir64> with your code? 2019-10-22T20:41:21 < antto> cuz sooner or later imma have to use a macro x_x 2019-10-22T20:41:23 < Jan-> no 2019-10-22T20:41:27 < Jan-> it's a blink program 2019-10-22T20:41:32 < Jan-> well I mean 2019-10-22T20:41:42 < kakimir64> did it work earlier? 2019-10-22T20:41:43 < Jan-> it's using loads of ST's code I don't know what that does 2019-10-22T20:41:46 < Jan-> but anyway 2019-10-22T20:41:48 < Jan-> yes it did work earlier 2019-10-22T20:41:55 < Jan-> and yes the stlink dongle appears in device manager 2019-10-22T20:41:57 < kakimir64> what did you change? 2019-10-22T20:42:13 < Jan-> I've used the same dongle a lot with mbed and platformio in the meantime 2019-10-22T20:42:16 < kakimir64> error is different when probe is not available 2019-10-22T20:42:22 < Jan-> cube wanted to update the software so I let it 2019-10-22T20:42:44 < kakimir64> no change in code to: pins, clocks? 2019-10-22T20:43:07 < Jan-> I've been using lots of different libraries 2019-10-22T20:43:10 < Jan-> they could have changed stuff 2019-10-22T20:43:14 < Jan-> but not that I know 2019-10-22T20:43:31 < antto> it is maybe time, to take a smol break 2019-10-22T20:43:31 < kakimir64> cube updated what? 2019-10-22T20:43:53 < Jan-> firmware on the stlink 2019-10-22T20:43:57 < Jan-> it did that before and it worked OK afterward 2019-10-22T20:44:20 < kakimir64> you can always get the chip to ISP mode 2019-10-22T20:44:24 < kakimir64> maybe 2019-10-22T20:44:38 < kakimir64> and using serial recover the chip 2019-10-22T20:45:11 < kakimir64> you have some ttl level uart available? 2019-10-22T20:45:24 < Jan-> let's try putting the stlink into another usb port 2019-10-22T20:45:27 < Jan-> that's solved problems before 2019-10-22T20:45:44 < kakimir64> did it say target is not responding then? 2019-10-22T20:45:55 < Jan-> no I was in another IDE 2019-10-22T20:45:59 < Jan-> but it was a similar symptom 2019-10-22T20:47:06 < Jan-> ok that solved it 2019-10-22T20:47:24 < Jan-> only thing is I have to remember to hit f8 because it stops at a breakpoint before main 2019-10-22T20:50:13 < kakimir64> "have you tried turning it off and back on?" 2019-10-22T20:51:23 < mawk> in cube mx enable SWD for debugging 2019-10-22T20:51:32 < mawk> ah 2019-10-22T20:51:48 < mawk> her board has no BOOT0 jumper strangely 2019-10-22T20:51:52 < mawk> and it's a pretty big board 2019-10-22T20:52:03 < Jan-> it has a BT0 and a BT1 2019-10-22T20:52:54 < Jan-> ooh, I can pause the processor 2019-10-22T20:53:01 < Jan-> this is a bit more advanced than arduino 2019-10-22T20:53:24 < aandrew> heh weird 2019-10-22T20:53:34 < aandrew> so I have enabled stack checking and periodically dump that info to the console 2019-10-22T20:54:13 < aandrew> a few threads report 300-900 bytes on the stack all the time but if I reduce them the system explodes 2019-10-22T20:54:33 < aandrew> I think they must use a whack of it at some point but then the stack use drops significantl 2019-10-22T20:54:36 < aandrew> significantly 2019-10-22T20:55:27 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-22T20:57:31 < srk> curl 'https://www.st.com/bin/st/search-cx/searchsite?start=0&length=20000&search[value]=&search[regex]=false&q=*&category=resources&lang=en' | jq '.data[] | .title_man' 2019-10-22T20:57:35 < srk> all the docs 2019-10-22T21:02:06 < aandrew> srk: the [value] might be tricky 2019-10-22T21:02:32 < aandrew> or is that URL gonna return everything? 2019-10-22T21:02:50 < aandrew> oh DAYMN 2019-10-22T21:03:03 < aandrew> nice srk 2019-10-22T21:03:11 < aandrew> I fucking hate how they hide that shit 2019-10-22T21:03:31 < srk> value and regex is redundant as well, pasted older version 2019-10-22T21:03:41 < srk> curl 'https://www.st.com/bin/st/search-cx/searchsite?start=0&length=20000&q=*&category=resources&lang=en' | jq '.data[] | .title_man' 2019-10-22T21:03:44 < srk> this works as well 2019-10-22T21:03:54 < srk> original query is 10 times .. :D 2019-10-22T21:03:56 < srk> longer 2019-10-22T21:04:11 < Jan-> holy hell does the stm32f407 have like 6 serial ports 2019-10-22T21:04:29 < aandrew> Jan-: yeah it's got a lot 2019-10-22T21:05:36 < aandrew> srk: need to add -g to curl though 2019-10-22T21:05:38 < Jan-> is there anything bigger than a 407 2019-10-22T21:05:38 < srk> think I'll scrape some of the reference manuals, pdf2text them and store somewhere accessible like web dir listing.. 2019-10-22T21:05:47 < srk> Jan-: 427! 2019-10-22T21:05:48 < srk> Fx 2019-10-22T21:05:54 < srk> 7x 2019-10-22T21:05:56 < Jan-> what's that got that this hasn't 2019-10-22T21:05:56 < srk> F7x 2019-10-22T21:06:43 < srk> some stuff you probably won't need :D 2019-10-22T21:06:46 < srk> like 2nd DMA 2019-10-22T21:07:01 < srk> not sure tbh 2019-10-22T21:07:14 < Jan-> I saw a reference to the fact that you can do DMA with the SPI peripherals 2019-10-22T21:07:23 < Jan-> I assume that means I don't have to manually load every byte like an AVR? 2019-10-22T21:07:24 < srk> with any peripheral basically 2019-10-22T21:07:25 < aandrew> srk: might be worthwhile to just run that on a server to grab it all 2019-10-22T21:07:35 < srk> to reach high SPI speeds you need to use DMA 2019-10-22T21:08:00 < Jan-> argh I just stubbed my toe on a metal box 2019-10-22T21:08:02 < srk> aandrew: might be awful lot of data, some filtering needed :) 2019-10-22T21:08:46 * Jan- clutches her foot 2019-10-22T21:08:48 < Jan-> *hop* *wail* 2019-10-22T21:09:00 < srk> like starting with AN/RM/STM 2019-10-22T21:09:26 < srk> Jan-: https://vivonomicon.com/2019/07/05/bare-metal-stm32-programming-part-9-dma-megamix/ 2019-10-22T21:09:48 < aandrew> 16.5GB for all of it 2019-10-22T21:09:49 < srk> all of the DMA versions and their subtle differences 2019-10-22T21:09:50 < aandrew> that's not bad 2019-10-22T21:10:07 < srk> :)) 2019-10-22T21:10:26 < aandrew> hell I could run that query periodically to get the file list then use wget -m to just grab the newer ones 2019-10-22T21:10:32 < aandrew> srk: you're a fucking genius 2019-10-22T21:10:35 < Jan-> srk I don't think I can even get access to "bare metal programming" 2019-10-22T21:10:40 < Jan-> I can't make it work other than with some huge IDE 2019-10-22T21:10:42 < aandrew> Jan-: of course you will 2019-10-22T21:10:51 < aandrew> curiosity and patience is all it takes 2019-10-22T21:11:03 < srk> aandrew: yeah that's what I want to do as well - I have a folder with handful of these but no way I'm gonna update them by hand :) 2019-10-22T21:11:04 < Jan-> I think I'd rather just work with gcc and make 2019-10-22T21:11:11 < Jan-> but right now I think we're stuck in cube 2019-10-22T21:11:40 < kakimir64> you take from CubeMX what you need and then you go on and do the rest with tools that work for you 2019-10-22T21:12:05 < Jan-> in theory 2019-10-22T21:13:57 < kakimir64> show your project Jan-? 2019-10-22T21:14:08 < kakimir64> I assume it's bloated with libs 2019-10-22T21:15:25 < Jan-> I guess so 2019-10-22T21:15:28 < Jan-> I don't really know 2019-10-22T21:15:55 < kakimir64> you sound confused all the time 2019-10-22T21:16:06 < Jan-> the only code I put into it is two calls to HAL_GPIO_WritePin and two calls to HAL_Delay 2019-10-22T21:16:15 < Jan-> but I have no control over what else it is including 2019-10-22T21:18:06 < kakimir64> what is the problem currently? 2019-10-22T21:18:22 < Jan-> nothing as far as I'm aware 2019-10-22T21:18:39 < Jan-> we're trying to find the USB to serial converter dongles so we can maybe patch up the serial pins to another USB port and try sending some text. 2019-10-22T21:18:48 < aandrew> heh 2019-10-22T21:19:10 < kakimir64> throw arduino at it 2019-10-22T21:19:10 < Jan-> I have one but I'm not sure of its pin out or if it is 3.3v or 5v 2019-10-22T21:19:11 < aandrew> time nice cat links.txt | parallel --gnu "wget {}" 2019-10-22T21:19:22 < Jan-> and putting 5v into the stm32's serial receiver pins is probably a bad time. 2019-10-22T21:19:33 < kakimir64> altough once I have tried arduino it had weird uart 2019-10-22T21:19:37 < aandrew> this is definitely taxing my internet connection 2019-10-22T21:19:49 < aandrew> wonder if st even sees this 2019-10-22T21:20:01 < kakimir64> Jan-: some pins are 5v tolerant! 2019-10-22T21:20:04 < kakimir64> read datasheet 2019-10-22T21:20:10 < Jan-> yeah er 2019-10-22T21:20:12 < Jan-> let's not take the piss eh 2019-10-22T21:20:36 < aandrew> Jan-: I typically use cube to get the initial format then just use my own makefile afterward 2019-10-22T21:20:56 < kakimir64> Jan-: you can always do nasty resistor divider trick 2019-10-22T21:21:13 < Jan-> aandrew people keep telling me to do that, I have no idea how. 2019-10-22T21:21:23 < Jan-> kakimir32 that's what we did last time and it was fine. 2019-10-22T21:21:31 < Jan-> I was talking to an stm-based device with an avr 2019-10-22T21:21:33 < Jan-> just used a pot 2019-10-22T21:21:34 < Jan-> was fine 2019-10-22T21:23:00 < Jan-> ok so the dongle we have has a ch340g I don't know if they can do 3.3v 2019-10-22T21:23:56 < aandrew> srk: 2.7GB downloaded 2019-10-22T21:23:59 < aandrew> 1/8 of the way there 2019-10-22T21:24:03 < Jan-> crikey aandrew 2019-10-22T21:24:06 < Jan-> how long is it going to take 2019-10-22T21:24:55 < Jan-> ok so this device as 5V, VCC, 3.3v, TXD, RXD and GND. It is supplied with a jumper connecting TXD and RXD which makes no sense. 2019-10-22T21:25:04 < kakimir64> echo 2019-10-22T21:25:08 < Jan-> do you guys think the idea is that we connect the jumper between VCC and 5V or 3.3V accordingly? 2019-10-22T21:25:34 < kakimir64> sometimes there is copper bridge in 5v option 2019-10-22T21:25:46 < kakimir64> you cut that and place jumper to 3v3 option 2019-10-22T21:25:57 < jpa-> Jan-: i think the idea may be that the jumper is just for testing 2019-10-22T21:26:01 < jpa-> and you remove it in normal use 2019-10-22T21:26:05 < Jan-> okay 2019-10-22T21:26:20 < jpa-> but it may also be for connecting to VCC, who knows 2019-10-22T21:26:25 * Jan- headscratch 2019-10-22T21:26:26 < Jan-> hmm 2019-10-22T21:26:31 < Jan-> it's one of these: https://www.banggood.com/3_3V-5V-USB-to-TTL-Converter-CH340G-UART-Serial-Adapter-Module-STC-p-1232728.html?cur_warehouse=CN 2019-10-22T21:26:34 < Jan-> they are super-cheap 2019-10-22T21:26:40 < jpa-> you can measure voltage on TX pin to see what it outputs 2019-10-22T21:26:53 < Jan-> we have to drag the oscilloscope out for that :) 2019-10-22T21:27:05 < Jan-> and find some way of making it continuously send data, hmm 2019-10-22T21:27:14 < jpa-> multimeter is enough, it is high when idle 2019-10-22T21:27:30 < Jan-> ohhh 2019-10-22T21:27:34 * Jan- offers jpa- a cookie 2019-10-22T21:27:45 * jpa- hopes it has chocolate in it 2019-10-22T21:27:45 < aandrew> 25s per 100MB, their site is not keeping up with my internet connection 2019-10-22T21:27:57 < aandrew> too much server side script shit to retrieve a file 2019-10-22T21:28:05 < aandrew> probably running java 2019-10-22T21:28:19 < Jan-> jpa-: white chocolate chip and macadamia nut, like the ones subway have. 2019-10-22T21:28:28 < Jan-> aandrew: ewww. 2019-10-22T21:28:42 < Jan-> well hang on, we probably only get 4 megabytes per second here 2019-10-22T21:28:46 < Jan-> maybe 6 2019-10-22T21:29:45 < aandrew> https://imgur.com/gallery/z2WkWhb 2019-10-22T21:32:15 < Jan-> the vcc pin is at 3.59V 2019-10-22T21:32:27 < Jan-> the 5v pin is at 5v 2019-10-22T21:32:46 < Jan-> the 3.3v pin is at 3.09v 2019-10-22T21:33:00 < Jan-> data lines are at 3.59v 2019-10-22T21:33:05 < Jan-> what does that suggest? 2019-10-22T21:35:15 < kakimir64> that you get what you pay for 2019-10-22T21:35:27 < Jan-> they were super cheap :) 2019-10-22T21:35:38 < kakimir64> did you check both data lines? 2019-10-22T21:35:46 < kakimir64> or your multimeter is very cheap also 2019-10-22T21:35:57 < kakimir64> do an echo test now? 2019-10-22T21:36:05 < Jan-> the data lines are shorted by this jumper 2019-10-22T21:36:06 < kakimir64> jumper between rx and tx 2019-10-22T21:36:16 < Jan-> let's see if we can do that 2019-10-22T21:36:51 < kakimir64> realterm should get that done just before it crashes 2019-10-22T21:37:10 < Jan-> termite thanks 2019-10-22T21:37:29 < Jan-> yes that works fine 2019-10-22T21:37:53 < Jan-> I send foobar, I get foobar back 2019-10-22T21:38:29 < aandrew> kakimir64: how is your phase controlling going today? have you reached thread count nirvana? 2019-10-22T21:39:05 < aandrew> halfway to 16GB now 2019-10-22T21:39:08 < kakimir64> I'm still figuring out synchronisation between PhaseTask and TriacTask 2019-10-22T21:39:17 < aandrew> why do you need to synchronize? 2019-10-22T21:39:38 < kakimir64> to replace timings for halve of sine that is not currently occuring 2019-10-22T21:40:13 < aandrew> have PhaseTask set newPhase=x, newPhaseValid=true and have TriacTask cli(); phase=newPhase; newPhaseValid=false; sti(); 2019-10-22T21:40:33 < aandrew> that'll end up being like four insns which is a nothingburger 2019-10-22T21:40:49 < kakimir64> I don't like nothingburgers 2019-10-22T21:41:03 < aandrew> triacTask will only change phase when it is appropriate to do so and phaseTask doesn't need to know when that is 2019-10-22T21:41:04 < kakimir64> I think I'm going to ride my moped now> 2019-10-22T21:41:15 < aandrew> are you phase controlling the motor on the moped? 2019-10-22T21:41:32 < kakimir64> no I throttle it manually 2019-10-22T21:44:38 < kakimir64> aandrew: https://paste.ee/p/3evjG current situation 2019-10-22T21:45:22 < aandrew> vTriacTask 2019-10-22T21:45:26 < aandrew> xPhaseTaskpvParameters_Struct 2019-10-22T21:45:31 < aandrew> xBinarySemaphoreFromPhaseTask 2019-10-22T21:45:35 < aandrew> xTempTickCount 2019-10-22T21:45:40 < kakimir64> those are short names 2019-10-22T21:45:41 < aandrew> who hurt you, kakimir64 2019-10-22T21:45:48 < kakimir64> I did 2019-10-22T21:46:13 < kakimir64> I would write longer names if there was any meaningful words to add to them 2019-10-22T21:46:47 < aandrew> usSineHalve 2019-10-22T21:46:48 < aandrew> good god 2019-10-22T21:47:02 < aandrew> the Hungarians are attacking! 2019-10-22T21:47:14 < kakimir64> you are not focusings 2019-10-22T21:47:24 < kakimir64> you don't see pass the mere symbols 2019-10-22T21:47:27 < aandrew> what are you prefixing your functions and variables iwth x and v garbage for 2019-10-22T21:47:39 < aandrew> no, I can't because it's stabbing me in the eyes 2019-10-22T21:47:42 < kakimir64> freertos naming convention 2019-10-22T21:47:54 < aandrew> you're not writing freertos code, you're writing code that runs in freertos 2019-10-22T21:48:00 < kakimir64> oh 2019-10-22T21:48:11 < kakimir64> I figured but autism got really bad 2019-10-22T21:48:20 < aandrew> and shouldn't you be using the CMSIS names? (I admit I wander over to CMSIS and FreeRTOS functions which is bad) 2019-10-22T21:48:24 < kakimir64> I will refactor as I get back from mopeding 2019-10-22T21:48:36 < aandrew> don't refactor on my account. this makes sense to you so do it 2019-10-22T21:48:41 < aandrew> it's just really fucking with my comprehension 2019-10-22T21:48:54 < aandrew> especially that hungarian notation trash 2019-10-22T21:49:20 < kakimir64> I concider my code encrypted 2019-10-22T21:49:48 < kakimir64> very good only I can read it 2019-10-22T21:49:55 < aandrew> makes it hard for us to help :-) 2019-10-22T21:50:08 < aandrew> although I am thankful that your indentation scheme is sane 2019-10-22T21:50:11 < kakimir64> I don't need help at the moment 2019-10-22T21:50:29 < kakimir64> until I need help 2019-10-22T21:51:56 -!- renn0xtk9 [~max@2a02:810d:1540:2448:5136:d5b4:2e79:2332] has joined ##stm32 2019-10-22T21:51:56 < kakimir64> where did you see hungarian? 2019-10-22T21:52:06 < doomba> i'm hungarian rn https://postimg.cc/SnPxhk2k 2019-10-22T21:52:23 < aandrew> usSineHalve, ucOutputPowerSettingCTRL, usPhaseNumber 2019-10-22T21:52:31 < aandrew> kakimir64: why is your triac task waiting on ANYTHING 2019-10-22T21:52:39 < aandrew> that is a realtime task 2019-10-22T21:52:54 < aandrew> you do your checks and yield(), you never ever wait 2019-10-22T21:53:03 < Jan-> I have a pin listing for the stm32f4 discovery 2019-10-22T21:53:08 < kakimir64> oh 2019-10-22T21:53:14 < aandrew> if you really want to be fancy I guess you can wait on an edge interrupt or someshit 2019-10-22T21:53:24 < kakimir64> I can yield() 2019-10-22T21:53:37 < kakimir64> thanks for an idea 2019-10-22T21:53:41 < Jan-> this listing includes which pins are 5v tolerant 2019-10-22T21:53:42 < kakimir64> you said it before 2019-10-22T21:53:48 < kakimir64> but now i understand it 2019-10-22T21:53:58 < Jan-> and it seems like basically all of them are except A5 and A4 2019-10-22T21:54:10 < Jan-> is it likely that will also apply to my stm32f4 board? 2019-10-22T21:54:37 < kakimir64> I just don't want to lose synchronisation as Triac task can in certain circumstances run without any blocking delays 2019-10-22T21:54:58 < aandrew> kakimir64: so basically phasetask calculates the new angles to set in triactask whenever you get a zero cross? 2019-10-22T21:55:13 < kakimir64> absolute times 2019-10-22T21:55:31 < kakimir64> and it should fill in sine_halve number 1 2019-10-22T21:55:32 < aandrew> ok, and triactask doesn't run unteil phasetask has new values? 2019-10-22T21:55:49 < kakimir64> then wait synchronization 2019-10-22T21:55:57 < kakimir64> and then fill in sine_halve 0 2019-10-22T21:56:00 < aandrew> dunno, this sounds like a really weird arch 2019-10-22T21:56:08 < kakimir64> yes 2019-10-22T21:56:13 < kakimir64> kakiflavour 2019-10-22T21:56:16 < kakimir64> moped> 2019-10-22T21:56:51 < aandrew> just have phasetask wait for zero cross, do the calc and update the period/firing angle. occurs every 8.3ms for 60Hz system. 2019-10-22T21:57:47 < kakimir64> when ISR arrives 2019-10-22T21:57:47 < aandrew> I'm thinking just have a timer synchronized to the zero cross with the time value set to the desired phase angle, and have the output of that fire the SCR 2019-10-22T21:58:04 < kakimir64> firing task is firing sine_halve 0 2019-10-22T21:58:10 < aandrew> alternatively when the timer interrupt expires, it unblocks triactask which fires the SCR but more jittery 2019-10-22T21:58:13 < kakimir64> so those values cannot be changed 2019-10-22T21:58:30 < kakimir64> I'll figure it out 2019-10-22T21:58:34 < aandrew> I'm very sure LPC has PWM-style timer outputs though 2019-10-22T21:59:00 -!- catphish_ [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-22T21:59:04 < aandrew> just set your PWM period to the half-phase time minus 500us or so and set the PWM value to the angle you want to fire at 2019-10-22T21:59:07 < kakimir64> it has multiple of them 2019-10-22T21:59:09 < aandrew> let it work in hardware 2019-10-22T21:59:18 < kakimir64> this is motor controller MCU 2019-10-22T21:59:32 < aandrew> I don't really understand your architecture here it seems overly complicated and I haven't heard a valid reason why yet 2019-10-22T21:59:50 < kakimir64> me neather 2019-10-22T22:00:12 < kakimir64> for academic excercise 2019-10-22T22:00:29 < kakimir64> *"academic excercise" 2019-10-22T22:02:55 < kakimir64> how do I learn task synchronization and scheduler if not like this? 2019-10-22T22:03:15 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-22T22:03:20 < kakimir64> this will load it propperly in addition to other tasks 2019-10-22T22:21:15 < Jan-> is "USART 1" likely to be the same thing the pinout calls "Serial 1" 2019-10-22T22:23:38 < aandrew> kakimir64: personally if you're trying ot do that I'd just have tasks with printf() and play with semaphores 2019-10-22T22:51:52 -!- onio [~onio@2a00:23c5:7a01:8600:3584:8be1:a96f:a113] has joined ##stm32 2019-10-22T23:03:38 < Jan-> So apparently in order to use the serial port I have to make changes in the GUI thing 2019-10-22T23:04:27 < Jan-> I assume to get to that I hit the button "device configuration tool" 2019-10-22T23:04:54 < Steffanx> you dont HAVE to 2019-10-22T23:05:03 < Jan-> but when I do that it pops up a requester that says "An error has occurred. See error log for more details. java.lang.NullPointerException" 2019-10-22T23:05:07 < Steffanx> haha 2019-10-22T23:05:13 < Steffanx> Bitten by java 2019-10-22T23:05:48 < Jan-> what did I do :( 2019-10-22T23:06:54 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-22T23:09:49 < Jan-> can I fix this somehow? 2019-10-22T23:10:31 < Cracki> uhhhh 2019-10-22T23:12:45 < Jan-> ah crap we had it working now the fricken ide breaks :/ 2019-10-22T23:15:06 < Steffanx> we 2019-10-22T23:15:16 < Steffanx> Jan- is not alone in this :o 2019-10-22T23:15:34 < Jan-> it's a long story 2019-10-22T23:16:01 < Steffanx> it makes sense. kakimir64 writes his comments in the "we" form too 2019-10-22T23:16:23 < Jan-> I do have help here. 2019-10-22T23:16:26 < Jan-> Why is a long story. 2019-10-22T23:17:36 < Steffanx> i see 2019-10-22T23:18:34 < Steffanx> i have time :P 2019-10-22T23:20:27 < Jan-> well I restarted cube and now it doesn't produce the error anymore 2019-10-22T23:20:38 < Jan-> but all it does is resize some of the code and console windows :/ 2019-10-22T23:21:17 < Jan-> shouldn't this pop up th GUI configuration thing 2019-10-22T23:21:34 < Steffanx> to be honest. idk. i never use it to generate code et all 2019-10-22T23:22:05 < Jan-> the problem is I could presumably add the code myself 2019-10-22T23:22:21 < Jan-> but I would need to find the code implemented for this exact microcontroller, or one very near it 2019-10-22T23:25:13 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-22T23:25:30 < kakimir64> back 2019-10-22T23:25:40 < kakimir64> good rides 2019-10-22T23:26:30 < Steffanx> Hows the moped moving? 2019-10-22T23:27:22 -!- renn0xtk9 [~max@2a02:810d:1540:2448:5136:d5b4:2e79:2332] has quit [Quit: Konversation terminated!] 2019-10-22T23:28:59 < kakimir64> very little mileage left to the engine 2019-10-22T23:31:57 < kakimir64> Steffanx: freertos project recommnds? 2019-10-22T23:32:25 < aandrew> heh I think st.com blocked me after 16G 2019-10-22T23:32:28 < aandrew> I can't get https://www.st.com/content/ccc/resource/training/technical/product_training/group0/43/c2/9f/fc/b2/d8/46/cd/STM32WB-Security-Advanced-Encryption-Standard-HW-Accelerator-AES/files/STM32WB-Security-Advanced-Encryption-Standard-HW-Accelerator-AES.pdf/jcr:content/translations/en.STM32WB-Security-Advanced-Encryption-Standard-HW-Accelerator-AES.pdf 2019-10-22T23:33:15 < kakimir64> you ddos them? 2019-10-22T23:33:21 < Steffanx> I bought some h745 nucleo kakimir64 . Going to play with real dual core magic 2019-10-22T23:33:22 < aandrew> for the last hour, yes :-) 2019-10-22T23:33:35 < doomba> like network blocked or just browser 2019-10-22T23:33:43 < aandrew> looks like some high level blocking 2019-10-22T23:33:44 < kakimir64> but what you are going to do with it Steffanx? 2019-10-22T23:33:49 < Steffanx> Idk 2019-10-22T23:33:54 < aandrew> I'm downloading it from another IP without issue 2019-10-22T23:33:55 -!- kakimir64 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-22T23:34:18 < Jan-> is there some reason that cube mx might remove code I'd put in main() when it "generates code" 2019-10-22T23:34:29 < aandrew> yep 2019-10-22T23:34:32 < aandrew> because it's evil 2019-10-22T23:34:43 -!- kakimir-pro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-22T23:34:48 < Jan-> Serious question. 2019-10-22T23:34:52 < aandrew> if you don't put your code in the specific areas it says are safe, it'll overwrite it 2019-10-22T23:34:55 < aandrew> I am being serious 2019-10-22T23:35:01 < aandrew> code generators like this are fucking evil 2019-10-22T23:35:51 < zyp> yeah, TI has similar shit 2019-10-22T23:36:12 < Jan-> you're fucking kidding. 2019-10-22T23:36:16 < kakimir-pro> Steffanx: that's just another board then 2019-10-22T23:36:23 < zyp> I just avoid mixing own and generated code, if I need to change something I replace entire file 2019-10-22T23:36:34 < PaulFertser> kakimir-pro: what's with your engine, why low mileage? 2019-10-22T23:36:36 < Jan-> well I have to add some code to main somehow... 2019-10-22T23:36:51 < kakimir-pro> I hear from sound it's getting worse 2019-10-22T23:37:09 < deltab> does the generated main call another function? 2019-10-22T23:37:13 < kakimir-pro> -> little engine life left 2019-10-22T23:37:19 < deltab> one that you can change? 2019-10-22T23:37:48 < kakimir-pro> oh Jan- have one of those things where you don't touch main() 2019-10-22T23:38:15 < kakimir-pro> bbl> 2019-10-22T23:39:15 < Steffanx> Not really. I will play around with it, mr kakimir-pro 2019-10-22T23:39:29 < kakimir-pro> not that much 2019-10-22T23:39:37 < Jan-> well I have to write some sort of main otherwise how do I... 2019-10-22T23:39:39 < Jan-> you know 2019-10-22T23:39:41 < Jan-> write an application 2019-10-22T23:40:03 < kakimir-pro> demotivational-kaki 2019-10-22T23:40:43 < PaulFertser> kakimir-pro: eh, sound indicating what specific kind of failure? 2019-10-22T23:40:44 < antto> aww 2019-10-22T23:41:16 < jadew> so... I just realized something 2019-10-22T23:41:34 < jadew> that nanoVNA shit is ruining the market for my next product 2019-10-22T23:41:59 < aandrew> jadew: you look for the USER CODE 1 markers (something like that) 2019-10-22T23:42:03 < BrainDamage> what is your next product? 2019-10-22T23:42:05 < aandrew> and make sure your shit is inbetween those 2019-10-22T23:42:15 < aandrew> or just don't regenerate code after doign it once and fuck all that noise 2019-10-22T23:42:18 < deltab> Jan-: oh, are there sections /* BEGIN USER CODE */ ... /* END USER CODE */? 2019-10-22T23:42:34 < jadew> BrainDamage, I've been working on a tracking generator for older HP SAs 2019-10-22T23:42:45 < deltab> er, USER CODE BEGIN 2019-10-22T23:42:47 < Jan-> deltab: dozens. :( 2019-10-22T23:42:50 < jadew> been at it for quite some time and just recently got a prototype I'm happy with 2019-10-22T23:42:52 < jadew> http://82.76.177.192/stuff/linearity.png 2019-10-22T23:42:57 < deltab> those are places you can add code 2019-10-22T23:43:07 < Jan-> right now I need to figure out what type is returned by HAL_UART_Transmit. 2019-10-22T23:43:08 < jadew> I've been trying to get that ^ basically 2019-10-22T23:43:14 < jadew> which is a non-normalized traced 2019-10-22T23:43:39 < BrainDamage> is x axis linear? 2019-10-22T23:43:41 < jadew> the spectral purity is on par with the HP TGs 2019-10-22T23:43:42 < antto> Jan- doesn't you IDE/notepad.exe tell you 2019-10-22T23:43:44 < aandrew> jadew: I bought one of these: https://www.aliexpress.com/item/32968659654.html 2019-10-22T23:43:45 < jadew> BrainDamage, yeah 2019-10-22T23:43:55 < antto> like.. when you mouse over it 2019-10-22T23:44:17 < aandrew> jadew: that's flat. what sa are you using for that pic 2019-10-22T23:44:26 < Jan-> a HAL_StatusTypeDef, apparently 2019-10-22T23:44:31 < jadew> aandrew, an HP 8562E 2019-10-22T23:44:33 < antto> noice 2019-10-22T23:44:39 < antto> sounds soooo much like C x_x 2019-10-22T23:45:10 < aandrew> Jan-: I just grep the source 2019-10-22T23:45:17 < antto> "just" 2019-10-22T23:45:25 < jadew> aandrew, yeah, I was aware of that one too, but I wasn't too concerned about it 2019-10-22T23:45:43 < Jan-> aandrew I used the "open declaration" functino 2019-10-22T23:45:46 < jadew> thing is... the enclosure alone for my TG costs $50... 2019-10-22T23:45:53 < aandrew> HAL_StatusTypeDef HAL_UART_Transmit(UART_HandleTypeDef *huart, uint8_t *pData, uint16_t Size, uint32_t Timeout); 2019-10-22T23:45:58 < Jan-> which opened up a file called stm32f4xx_hal_uart.c on line 1019. 2019-10-22T23:46:06 < Jan-> which is that one that aandrew just posted 2019-10-22T23:46:07 < jadew> then there's custom machined shielding inside 2019-10-22T23:46:09 < aandrew> yep, and you can see it returns HAL_OK if there was no issue 2019-10-22T23:46:53 < antto> deltab my IDE has integration for wxWidgets, and the generated code sorta looks like grayed out so it's obvious that you shouldn't touch those parts, while you can write anywhere else 2019-10-22T23:46:55 < Jan-> ok that seems to work 2019-10-22T23:47:10 < jadew> actually the enclosure is not 50, but close, point being that I think the nanoVNA kinda killed my product before it was even out 2019-10-22T23:47:36 < jadew> because the usual nanoVNA user won't care about the specs anyway 2019-10-22T23:48:06 < Jan-> what I don't get is what "huart1" is and why I pass a pointer to it into HAL_UART_Transmit. 2019-10-22T23:48:10 < Jan-> Sounds like win32 internals code 2019-10-22T23:48:14 < Jan-> "handle to uart" 2019-10-22T23:48:40 < antto> Jan- doesn't your chip have multiple UARTs? 2019-10-22T23:48:48 < Jan-> six I believe 2019-10-22T23:48:52 < aandrew> Jan-: really easy 2019-10-22T23:48:57 < Jan-> it's an enum 2019-10-22T23:49:04 < antto> well, how would UART_Transmit know which one you want? 2019-10-22T23:49:14 < aandrew> Jan-: all the HAL peripherals have some data blob associated with them 2019-10-22T23:49:20 < aandrew> the "handle" is ust a pointer to that blob 2019-10-22T23:49:33 < Jan-> cripes I guess it's good this thing has more memory than an atmega328 2019-10-22T23:49:38 < jadew> BrainDamage, what do you think? is it worth going on with my project? 2019-10-22T23:49:48 < aandrew> you allocate the blob, fill it out, call HAL_xxx_Init(&handle) and the HAL library uses that to configure and track state of that peripherla 2019-10-22T23:49:59 < aandrew> all the HAL_xxx_() functions want that handle to work 2019-10-22T23:50:21 < Jan-> well I didn't fill it out 2019-10-22T23:50:29 < aandrew> also: HAL_xxx_Init() will call HAL_xxx_MspInit() which is a function YOU write to initialize the hardware (enable clocks, enable I/O, etc.) 2019-10-22T23:51:04 < antto> deltab, https://i.imgur.com/MlPHqlq.png the blue-ish stuff is the autogenerated stuffz 2019-10-22T23:51:11 < aandrew> there are basic basic HAL_xxx_MspInit() functions written and weakly linked so if you don't write your own the default is used and it's usually doing what you want 2019-10-22T23:51:28 < Jan-> actually it auto generated MX_USART1_UART_Init(void) 2019-10-22T23:51:32 < aandrew> Jan-: you need to take some time to learn how HAL expects to be used 2019-10-22T23:51:37 < aandrew> that will help you a LOT 2019-10-22T23:51:46 < BrainDamage> I'm not sufficiently familiar with your target market, the average EE might get their needs solved with an uncalibrated source, but it seems to me you're targeting high quality test instrumentation, so things like univs or RF companies 2019-10-22T23:51:48 < Jan-> I get it, I'm just not a big fan of pointers 2019-10-22T23:51:50 < aandrew> Jan-: I can't expect that that kind of response was easy to come to 2019-10-22T23:51:57 < Jan-> haven't you SEEN that poor plasticine binky 2019-10-22T23:52:01 < aandrew> Jan-: you are writing code in C; you best get used to liking pointers 2019-10-22T23:52:15 < Jan-> I've written a bit of C, I tend to avoid them :) 2019-10-22T23:52:28 < antto> aww 2019-10-22T23:52:31 < aandrew> ok but again, you're writing in C now... get used to it or perhaps think about a different platform 2019-10-22T23:52:36 < aandrew> I hear Rust is popular with the cool kids 2019-10-22T23:52:49 < jadew> BrainDamage, yeah.. that's a very small market tho 2019-10-22T23:52:49 < Jan-> I'm neither cool nor a kid. 2019-10-22T23:53:34 < antto> Jan- why, you should worship them point0rz 2019-10-22T23:53:39 < jadew> BrainDamage, thanks for the input 2019-10-22T23:53:47 < Jan-> pointers are great 2019-10-22T23:53:49 < antto> ur switch to C++ and abstract them away 2019-10-22T23:53:52 < jadew> I was actually betting on the average EE tbh 2019-10-22T23:53:53 < antto> * or 2019-10-22T23:53:54 < Jan-> the problem is that the C syntax for using them is absolutely mad 2019-10-22T23:54:04 < BrainDamage> jadew: mmm, how about a middle way? 2019-10-22T23:54:06 < antto> which syntax? 2019-10-22T23:54:09 < aandrew> Jan-: https://pastebin.com/FNn71yTk is how I init my UARTs 2019-10-22T23:54:20 < BrainDamage> jadew: you could turn your tool into a calibration source 2019-10-22T23:54:24 < aandrew> I'm not a fan of MSPInit() personally so I break the API a little by doing it here 2019-10-22T23:54:41 < aandrew> what's wrong with C's pointers? 2019-10-22T23:54:45 < aandrew> or their pointer syntax 2019-10-22T23:54:46 < BrainDamage> that would let you tap into the average EE market, with slightly more strict needs 2019-10-22T23:55:06 < aandrew> int foo; int *pfoo = &foo... very easy :-) 2019-10-22T23:55:07 < Jan-> too many asterisks and ampersands 2019-10-22T23:55:17 < jadew> BrainDamage, hah, that's actually not bad, but what's stopping china from releasing a measured ADF43xx IC for $5? 2019-10-22T23:55:18 < aandrew> I remember when I was learning C this was the trickiest bit to get 2019-10-22T23:55:22 < antto> Jan- smol hands? 2019-10-22T23:55:34 < aandrew> "Hmm that didn't work, add a *." "no? ok, two stars?" "try an &" 2019-10-22T23:55:36 < aandrew> eventually I got it 2019-10-22T23:55:39 < antto> there are reduced scale keyboards 2019-10-22T23:55:42 < BrainDamage> jadew: nothing, but your work is already done, it's a sunken cost 2019-10-22T23:56:01 < BrainDamage> if you can find a way to utilize it, then it's extra cash 2019-10-22T23:56:16 < aandrew> jadew: how was what I linked not what you're concerned about? 2019-10-22T23:56:17 < jadew> true 2019-10-22T23:56:22 < Jan-> When you get some asshat who's been super clever and it's all like **&foo* then it gets a bit much 2019-10-22T23:56:24 < BrainDamage> and until china delivers, you'll still have gained some 2019-10-22T23:56:37 < jadew> aandrew, it's less appealing than the nanoVNA 2019-10-22T23:56:44 < aandrew> Jan-: I've noticed something with your chats here in the last week or so 2019-10-22T23:56:58 < aandrew> you have a very ... negative outlook on all this work 2019-10-22T23:57:08 < Jan-> yeah well 2019-10-22T23:57:16 < Jan-> I've been bouncing off a lot of shitty open source software 2019-10-22T23:57:18 < antto> aandrew iz iz called realism 2019-10-22T23:57:20 < aandrew> I mean you're making your way through it which is great, but there's always something negative you've gotta comment about whatever it is that's suggested 2019-10-22T23:57:22 * Jan- does not like open source software 2019-10-22T23:57:30 * Jan- ends up in a foul mood 2019-10-22T23:57:36 < jadew> BrainDamage, I'll consider it, it's really not a bad idea 2019-10-22T23:57:49 < jadew> but it needs a bit of rework 2019-10-22T23:57:53 < aandrew> jadew: really? I find nanovna kind of shite 2019-10-22T23:58:01 < jadew> this thing is an actual tracking generator, in the classical sense 2019-10-22T23:58:05 < jadew> it requires an LO 2019-10-22T23:58:06 < antto> Jan- well, how about close all teh browser tabs of opensource "sh*t", and use something else? 2019-10-22T23:58:30 < Jan-> I have :) 2019-10-22T23:58:38 < antto> like, either get a closed-source chunk of software that is eggcelent, and innovate sumfin' 2019-10-22T23:58:39 < aandrew> what I really want to build is hforsten's VNA: http://hforsten.com/improved-homemade-vna.html 2019-10-22T23:58:40 < jadew> aandrew, me too, but that's not stopping people from buying it 2019-10-22T23:58:47 < aandrew> or at least an ethernet version of it 2019-10-22T23:59:00 < jadew> and for everyone buying that crap, there's one less potential customer for me 2019-10-22T23:59:36 < aandrew> jadew: ok, but let me ask you this: why would you spend so much time and effort and money trying ot make what looks like a very good TG for a $50 VNA? --- Day changed Wed Oct 23 2019 2019-10-23T00:00:00 < aandrew> i.e. if they are using a shitty $50 VNA why would they *care* to buy a super flat TG to go with it? 2019-10-23T00:00:01 < jadew> aandrew, what do you mean? 2019-10-23T00:00:16 < aandrew> the nanovna is a piece of shit 2019-10-23T00:00:19 < jadew> the TG goes with the old HP SAs 2019-10-23T00:00:23 < aandrew> oh 2019-10-23T00:00:27 < aandrew> see there's the difference 2019-10-23T00:00:29 < aandrew> my mistake 2019-10-23T00:00:57 < jadew> but people who own those, might just go for the nanoVNA and suddenly not need a TG anymore 2019-10-23T00:00:58 < aandrew> but now the problem becomes if someone is happy with a shittier TG in the nanoVNA it's unlikely they'll care about the better one 2019-10-23T00:01:10 < jadew> exactly 2019-10-23T00:02:42 < jadew> for example, I doubt you can properly characterise an active device with the nanoVNA 2019-10-23T00:02:46 < jadew> or the one you got for that matter 2019-10-23T00:02:55 -!- kakimir-pro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-23T00:03:51 < karlp> so nanovna is shit, and minivna is the ok one? 2019-10-23T00:04:08 < jadew> karlp, I've read that it's better than miniVNA 2019-10-23T00:04:17 < jadew> at least that's what some say, could be BS 2019-10-23T00:04:20 < aandrew> I was looking at that too; apparently if you are good with its limitations it's quite good 2019-10-23T00:04:32 < aandrew> basically you put it in an insulated box and let it warm up, do your cal, do your measurements 2019-10-23T00:04:51 < aandrew> doing that and sticking to that gives you repeatability. that's what I've heard but have not used one myself 2019-10-23T00:05:06 < karlp> I know zyp was quite content with one of them at least, 2019-10-23T00:05:35 < antto> interesting lil fact, in bulgarian, not many words end on "vna" .. one is "govna" which means "shits" (as in a collection of multiple chunks of sh*t, not the verb sh*t) 2019-10-23T00:06:15 < qyx> we have hovna 2019-10-23T00:06:31 < antto> which language is that? 2019-10-23T00:06:35 < qyx> with the same meaning 2019-10-23T00:06:36 < qyx> .sk 2019-10-23T00:06:40 < jadew> aandrew, got a proper SA or power meter? 2019-10-23T00:06:42 < antto> oh 2019-10-23T00:06:53 < jadew> would be interesting to see the actual output of the thing you bought 2019-10-23T00:06:53 < antto> and "govno" is for a single piece.. 2019-10-23T00:06:58 < aandrew> jadew: I just borrow it when I need one right now 2019-10-23T00:06:58 < qyx> yeah 2019-10-23T00:07:09 < aandrew> hence why I want to build hforsten's VNA 2019-10-23T00:07:16 < antto> >:) 2019-10-23T00:07:32 < aandrew> antto: lol 2019-10-23T00:08:03 < antto> but it's not often used, since we have better alternative words for teh brown substance 2019-10-23T00:08:22 < antto> and for teh act of producing them 2019-10-23T00:08:40 < qyx> if I may ask.. 2019-10-23T00:09:05 < antto> you may, what's the question? 2019-10-23T00:09:37 < qyx> which one? 2019-10-23T00:09:52 < qyx> aandrew: cannot it be bough? 2019-10-23T00:09:54 < antto> laino/laina is the simplest one 2019-10-23T00:10:00 < qyx> I was looking at it too 2019-10-23T00:10:13 < aandrew> qyx: not that I'm aware of 2019-10-23T00:10:17 < aandrew> but I want to make it ethernet anyway 2019-10-23T00:10:20 < aandrew> USB is shit 2019-10-23T00:10:40 < qyx> antto: heh 2019-10-23T00:11:31 < jadew> I was planning to make a VNA too (after finishing the TG), but that market is not there anymore 2019-10-23T00:11:55 < BrainDamage> aandrew: one thing that the pluto sdr does which is interesting is that it's ethernet based, but if you use the usb port it creates a usb network interface 2019-10-23T00:12:10 < aandrew> BrainDamage: haha that's kind of funny 2019-10-23T00:12:11 < BrainDamage> which gives you flexibility in connection 2019-10-23T00:12:31 < aandrew> the whole idea of using ethernet is that you don't have to use the USB stack 2019-10-23T00:12:36 < aandrew> not layer ethernet on top of it 2019-10-23T00:12:49 < aandrew> oh wait pluto has both eth and usb? 2019-10-23T00:12:50 < BrainDamage> yes, and you don't have to deal with the timing insanity of usb 2019-10-23T00:12:51 < jadew> I like ethernet too for instruments 2019-10-23T00:12:56 < BrainDamage> it has wifi too 2019-10-23T00:13:56 < jadew> GPIB is the worst 2019-10-23T00:14:11 < jadew> it's bulky 2019-10-23T00:14:33 < jadew> and you can only run one experiment on the bus 2019-10-23T00:14:36 < BrainDamage> oh god, I have plenty of bad experiences with GPIB 2019-10-23T00:15:16 < jadew> more than half of my instruments are probably GPIB connected :/ 2019-10-23T00:15:38 < BrainDamage> are you using a ethernet-gpib bridge from agilent with a ton of quirks as well? 2019-10-23T00:15:58 < jadew> no, I built this one several years ago 2019-10-23T00:16:17 < BrainDamage> because that gives me flasback of my bachelor thesis 2019-10-23T00:16:33 < BrainDamage> the thing would stop responding halfway a measurement 2019-10-23T00:16:42 < BrainDamage> ... which lasted 8h 2019-10-23T00:16:49 < jadew> ouch 2019-10-23T00:16:57 < jadew> connection problem maybe? 2019-10-23T00:17:28 < jadew> I have a weird issue with mine, where the bus gets stuck and I have to turn on the SA, take a screenshoot and then it works :D 2019-10-23T00:17:43 < BrainDamage> I never managed to debug it, the worst part is that the advisor was adamant to use labview to interface with the instrumentation because others would have to use it as well after me 2019-10-23T00:17:50 < jadew> I have no idea what's going on, but I don't have the desire to look into it either 2019-10-23T00:18:09 < BrainDamage> and I had to jump through hoops to make labview handle gracefully all the errors 2019-10-23T00:18:19 < BrainDamage> and write immediately all the measurement points 2019-10-23T00:18:35 < BrainDamage> so that if it failed, i wouldn't have to start from t=0 2019-10-23T00:19:00 < jadew> heh, at least the EOL stuff was handled automatically by the driver? 2019-10-23T00:19:12 < BrainDamage> yes 2019-10-23T00:19:13 < karlp> fuck labview looks like a repeatability disaster 2019-10-23T00:19:33 < karlp> I watched a guy build this giant programming rig in it, completely unversionable. zero way of recreating any of it. 2019-10-23T00:19:39 < karlp> "but no progrtamming!" 2019-10-23T00:20:21 < jadew> yeah... but most experiments are programmed, no? 2019-10-23T00:20:43 < BrainDamage> labview propagand itself to not require knowledge of programming 2019-10-23T00:20:53 < mawk> I became a android-superpro 2019-10-23T00:20:59 < BrainDamage> except unless you do, you'll still write a shitty labview code 2019-10-23T00:21:00 < mawk> I can make my phone vibrate from native C code 2019-10-23T00:21:50 < jadew> personally, I wrote a VXI client library and I have this thing that works with my GPIB controller, so I write all experiments in c++ 2019-10-23T00:21:59 < jadew> no VISA or any of that crap 2019-10-23T00:22:18 < Steffanx> So how well does the aliexpress analyser work, aandrew ? 2019-10-23T00:23:48 < jadew> Steffanx, it's the best thing since toast bread 2019-10-23T00:24:19 < Steffanx> Not sure how serious you are jadew 2019-10-23T00:24:29 < jadew> me neither :) 2019-10-23T00:25:06 < antto> Seriousness [#_________] 2019-10-23T00:25:20 < jadew> but I can think of single components in my TG, that are more expensive than that entire board 2019-10-23T00:25:37 < Steffanx> Cant expect too much for 50 euro.. 2019-10-23T00:25:46 < mawk> I will become a huge android celebrity I hope 2019-10-23T00:25:54 < mawk> if I publish what I found before any other, on xda-developers 2019-10-23T00:25:54 < Steffanx> Will you now? 2019-10-23T00:26:12 < Steffanx> You'll be forgotten in 2 seconds 2019-10-23T00:26:13 < jadew> Steffanx, what's bothering me is that it works and it sells :P 2019-10-23T00:26:14 < mawk> :( 2019-10-23T00:26:22 < antto> mawk u gonna b teh kardashian of teh crapdroid world? 2019-10-23T00:26:31 < mawk> yes, I found the way to change DNS servers on android 10; the previous method has been removed by android in a blink of a commit with reason "we refactor stuff, maybe add this back later" 2019-10-23T00:27:00 < jadew> mawk, neat, so blockada is back? 2019-10-23T00:27:08 < jadew> which btw, never worked on my phone... 2019-10-23T00:27:11 < jadew> I still get ads 2019-10-23T00:27:14 < mawk> ah, yeah I could patch them I guess 2019-10-23T00:27:26 < mawk> I did this for the wireguard vpn, which didn't work on android Q before 2019-10-23T00:27:38 < antto> oh, ur smartphone vomits google ads at you? 2019-10-23T00:27:42 < antto> hu hu hu 2019-10-23T00:27:49 < mawk> google phone vomiting google ads 2019-10-23T00:27:56 < jadew> all the time 2019-10-23T00:27:59 < BrainDamage> considering they also are trying to scrap adblocking, I wonder how accidental it was 2019-10-23T00:28:15 < jadew> obviously not accidental 2019-10-23T00:28:15 < mawk> BrainDamage: it's because they added DNS over TLS so had a big refactor of all the dns stuff in android 10 2019-10-23T00:28:27 < mawk> so the previous command line for changing them became too simple, so they just removed it instead of reworking t 2019-10-23T00:28:29 < antto> their ads on pootube are absolutely horrific lately 2019-10-23T00:28:40 < antto> and this is not even their final form 2019-10-23T00:29:27 < PaulFertser> Why are not all android users using "vanced.app" for youtube? Are there any gotchas? 2019-10-23T00:29:38 -!- friendofafriend [~ian@75.182.67.149] has quit [Quit: leaving] 2019-10-23T00:29:45 < jadew> what's vanced? 2019-10-23T00:29:49 < jadew> youtube no ads? 2019-10-23T00:29:54 < PaulFertser> Properly patched official youtube app. 2019-10-23T00:30:03 < mawk> yes, I used that PaulFertser 2019-10-23T00:30:13 < jadew> does it run in the background too? 2019-10-23T00:30:14 < BrainDamage> I use firefox 2019-10-23T00:30:16 < mawk> yes jadew 2019-10-23T00:30:18 < mawk> that's the trick 2019-10-23T00:30:20 < BrainDamage> and ublock 2019-10-23T00:30:26 < mawk> no need to pay 15€/month for background music 2019-10-23T00:30:26 < jadew> ok, I have to give it a try 2019-10-23T00:30:30 < jadew> BrainDamage, on android? 2019-10-23T00:30:31 < BrainDamage> no ads and run in the background too 2019-10-23T00:30:33 < BrainDamage> yes 2019-10-23T00:30:35 < mawk> if you're rooted install it as a magisk module 2019-10-23T00:30:43 < jadew> I'll have to try that too 2019-10-23T00:30:45 < mawk> it will replace the system youtube and crash way less than if you install it as .apk 2019-10-23T00:31:00 < Steffanx> PaulFertser: sounds like something that requires a rooted phone 2019-10-23T00:31:03 < jadew> I've been living with stupid ads in my music since forever 2019-10-23T00:31:10 < PaulFertser> Steffanx: and in fact it doesn't 2019-10-23T00:31:20 < Steffanx> But no playstore app 2019-10-23T00:31:20 < PaulFertser> Steffanx: so that's not answering my question 2019-10-23T00:31:27 < mawk> yeah not playstore, but easy to install 2019-10-23T00:31:34 < mawk> android isn't iOS, you're free to not use the playstore 2019-10-23T00:31:39 < mawk> go on https://vanced.app 2019-10-23T00:31:44 < mawk> install microG and vanced 2019-10-23T00:31:55 < Steffanx> Never heard of it, ever. 2019-10-23T00:31:56 < mawk> if you're not rooted you install side by side with google's youtube, it's not very reliable for me but better than nothing 2019-10-23T00:32:40 < jadew> which one has the bitcoin miner? 2019-10-23T00:32:45 < mawk> Steffanx: depending on your android version the setting to install "unknown" .apk is either global in the security settings, or app by app in the application setting 2019-10-23T00:32:56 < mawk> so in your case the app is Chrome, since you will download it using chrome 2019-10-23T00:33:06 < Steffanx> Or firefox 2019-10-23T00:33:11 < mawk> vanced.app is the official site jadew , there is a second site which is very sketchy 2019-10-23T00:33:28 < antto> i got a crapdroid phone, technically.. i rubbed it a little >:/ 2019-10-23T00:33:32 < jadew> the original app sounds sketchy enough, tbh 2019-10-23T00:33:40 < antto> installed firefox on it, and an IRC client 2019-10-23T00:33:47 < antto> and a keyboard thing 2019-10-23T00:33:48 < jadew> a *patched* version of copyrighted software 2019-10-23T00:34:16 < antto> with the default keyboard i can't f*cking type a dash >:( 2019-10-23T00:34:27 < jadew> no way it could contain malware, isn't it? 2019-10-23T00:34:46 < mawk> why ? 2019-10-23T00:34:48 < mawk> I don't think it's anything malicious jadew 2019-10-23T00:34:52 < mawk> yeah, it's just people who don't like to pay for bare minimum service 2019-10-23T00:34:54 < mawk> google siphons our data way enough to grant us the right to listen to music freely 2019-10-23T00:35:21 < Steffanx> It will make your phone vibrate with mawk's new hacks :P 2019-10-23T00:35:35 < antto> mawk but how will they be evil if they were fair? 2019-10-23T00:35:38 < antto> >:) 2019-10-23T00:36:10 < jadew> mawk, I get why it's done, it's just that in the past, robin hood used to pack other gifts too in crack.exe 2019-10-23T00:36:18 < mawk> jadew: it has been posted on xda-dev, I think if it contained anything sketchy people would've noticed 2019-10-23T00:36:27 < mawk> it's a community full of android hackers much more knowledgeable than me 2019-10-23T00:36:35 < mawk> yeah I get it 2019-10-23T00:37:06 < jadew> that's a good hacker name, btwm 2019-10-23T00:37:06 < mawk> Steffanx: https://paste.serveur.io/N8Sh43Mv.c here is my binder-based vibro 2019-10-23T00:37:08 < jadew> *btw 2019-10-23T00:37:11 < jadew> robin hood 2019-10-23T00:37:29 < antto> "a new code change, i gotta review it.. bruh, i feel lazy.. we are 20 people, surely the other will check the patch, imma lay low" 2019-10-23T00:37:32 < jadew> probably taken tho, it'll have to be robinhood2019 2019-10-23T00:38:23 < antto> rohinbood 2019-10-23T00:40:15 < jadew> I'm so bummed about my realization 2019-10-23T00:40:19 < mawk> binder is a kernel-based IPC mechanism Google bought when they bought Android, it's very undocumented 2019-10-23T00:40:36 < mawk> I asked on the blokada chat jadew and they already patched it for android 10 it seems 2019-10-23T00:41:01 < mawk> [ Peter ], [22.10.19 23:38] 2019-10-23T00:41:01 < mawk> [En réponse à mawk] 2019-10-23T00:41:01 < mawk> hi, 2019-10-23T00:41:01 < mawk> it does, you only need to disable Private DNS and Data Saver in system settings 2019-10-23T00:41:03 < jadew> I think the only way to still make a living building electronics is to move to China and build reference design boards for $2 and sell them for $2.1 2019-10-23T00:41:13 < mawk> for $2.01 you mean 2019-10-23T00:41:46 < jadew> no, $2.1, that's the only way to make your 2c profit 2019-10-23T00:42:07 < jadew> mawk, interesting 2019-10-23T00:42:13 < jadew> I'll have to upgrade 2019-10-23T00:46:39 < Steffanx> You cant have been the first one to have done this, can you mawk 2019-10-23T00:48:45 < mawk> why not ? 2019-10-23T00:48:54 < mawk> I did it in the first days after android Q came out 2019-10-23T00:48:58 < mawk> to fix my favorite project 2019-10-23T00:49:23 < mawk> before android Q there was no way to do it from a user application, so nobody could've done it before 2019-10-23T00:49:26 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-23T00:50:11 < Steffanx> I mean the vibrating from native C codes with the bindings et all 2019-10-23T00:50:31 < mawk> ah, well binder was not possibly called from C code like that before, it came out in Q too 2019-10-23T00:50:41 < mawk> but I guess maybe a very patient hacker that read all the AOSP source code did it too 2019-10-23T00:51:02 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-23T00:51:02 < mawk> but you'd maybe give up before finishing and just do it in Java which is the regular way 2019-10-23T00:51:08 < Steffanx> You were impatient and.. still did it? 2019-10-23T00:51:15 < mawk> yes, thanks to android Q 2019-10-23T00:51:23 < mawk> that has the /system/lib64/libbinder_ndk.so library 2019-10-23T00:51:59 < Steffanx> I'm not into or following android at all, so I've no real clue what it's all about:P 2019-10-23T00:52:51 < mawk> this is the library to allow user programs to call in all the good android services like the vibrator; before that you needed to use Java 2019-10-23T00:52:59 < qyx> antto: are you able to actually use the firefox? 2019-10-23T00:53:30 < qyx> I tried multiple times, I failed multiple times 2019-10-23T00:53:47 < qyx> it is simply 10x slower than any other chromium based browser 2019-10-23T00:53:52 < qyx> I am using kiwi browser now 2019-10-23T01:23:49 < antto> qyx i don't know 2019-10-23T01:24:14 < antto> i mean, i am absolutely disgusted by smartphones 2019-10-23T01:24:20 < antto> and google 2019-10-23T01:25:17 < antto> so, i think i only used the default browser (chrome?) when i figured it's unavoidable, in order to get firefox 2019-10-23T01:25:42 < antto> since then i used firefox, opened some wikipedia pages and some other sh*t 2019-10-23T01:26:05 < antto> it "works" .. as much as that's possible ot a device like this 2019-10-23T01:26:12 < antto> * on 2019-10-23T01:26:45 < antto> also, duckduckgo 2019-10-23T01:26:53 < antto> no google 2019-10-23T01:27:42 < antto> i tried to uninstall teh chrome/google sh*t but of course, they've thought about that 2019-10-23T01:30:19 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-23T01:31:03 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-23T01:31:47 < karlp> yuk. 2x16bit registers, final value is (a<<8|b>>8)/25600. 2019-10-23T01:32:11 < karlp> why mix up the /256 _and_ a scaling to "avoid" flaots. 2019-10-23T01:32:13 < karlp> weirdos 2019-10-23T01:32:13 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Client Quit] 2019-10-23T01:32:14 < antto> ADC? 2019-10-23T01:32:39 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-23T01:32:47 < antto> oh wait, is it the weird AT90 thing? 2019-10-23T01:35:39 < karlp> yeah 2019-10-23T01:38:50 < antto> is it supported by gcc? 2019-10-23T01:39:13 < antto> akchually i don't even know if "avr-gcc" is wut you use for them AVR32 chips 2019-10-23T01:43:29 < karlp> no, not at90, atm90, it' sjust and adc front end with a spi interface. 2019-10-23T01:44:20 < antto> oh, it's not an MCU? 2019-10-23T01:45:12 < karlp> no 2019-10-23T01:45:16 < antto> okay, muh cat sez "enough keyboard for today" 2019-10-23T01:46:43 < antto> mouse too >:( 2019-10-23T01:46:46 < antto> grrrrrr 2019-10-23T01:47:03 < antto> imma have to learn telekinesis 2019-10-23T01:49:43 < zyp> karlp, power metering? 2019-10-23T01:50:26 < karlp> zyp: yeah, newstuff. 2019-10-23T01:50:42 < zyp> same atmel chip that I used or another? 2019-10-23T01:50:47 < karlp> cheap,er meant to be easier to calibrate than some others 2019-10-23T01:50:58 < karlp> one ofthem, atm90e32 and e36 are ~similar enough for most stuff. 2019-10-23T01:51:22 < zyp> ah, yeah, I used e32 2019-10-23T01:52:10 < karlp> yeah, me too, wsa just having a weird moment with one of their registers. 2019-10-23T01:54:16 < karlp> better than fucking linux rootfs shit 2019-10-23T01:54:23 < zyp> https://paste.jvnv.net/view/EicDs <- here's the shit I used to test mine 2019-10-23T01:55:08 < karlp> yeah, you're just using the "top16bits" registers 2019-10-23T01:55:14 < karlp> line 34, 2019-10-23T01:55:33 < karlp> there's a second register with more decimal places, it was just a bit fiddly to join it properly. 2019-10-23T01:55:50 < zyp> right, there were some stuff like that, yeah 2019-10-23T01:56:08 < zyp> what confused me was those Ugain and Igain values 2019-10-23T01:56:34 < karlp> does chr+chr+chr do concat in python? 2019-10-23T01:56:37 < karlp> that's non obvious 2019-10-23T01:56:43 < karlp> yeah, the gain regs are dumb. 2019-10-23T01:56:48 < karlp> very metering world, but still lame. 2019-10-23T01:57:15 < BrainDamage> yes, sum of chars is string concatenation 2019-10-23T01:57:19 < zyp> datasheet didn't say how the value would be used or anything, just measure against a reference and correct 2019-10-23T01:58:14 < zyp> technically chr(x) makes a 1-character string from ascii (or maybe even unicode?) value x 2019-10-23T01:58:21 < zyp> and adding strings concatenates 2019-10-23T01:58:30 < zyp> this is old shitty python2 code 2019-10-23T01:58:40 < mawk> rm it 2019-10-23T01:58:56 < karlp> it's dumb in my mind having both gain/offset regs and defining some of the fields to be in native units, like volts/amps/watts, but then having an second config for the "pulse configs" that defines what goes in the "not kwh, it's pulses" registers 2019-10-23T01:58:59 < zyp> python3 has bytestrings which are nicer to convert between integers and back 2019-10-23T01:59:01 < karlp> but whatever, 2019-10-23T01:59:16 < mawk> in python2 or python3 I'd use the struct module I think 2019-10-23T01:59:29 < zyp> yeah, just too lazy to 2019-10-23T01:59:34 < mawk> too bad it doesn't speak the exact width integer but only the well known names (char, int, long, long long) 2019-10-23T02:00:34 < karlp> it's kinda "nice" how it doesn't ever talk about, "this is how you use your system variables to work out nominal gains" just "change this number until you get the right readings" 2019-10-23T02:00:53 < zyp> exactly, that's what annoyed me 2019-10-23T02:01:39 < zyp> line 63/64 is me reverse engineering what goes into the gains 2019-10-23T02:02:12 < karlp> 2.4 your burden? 2019-10-23T02:02:18 < zyp> IIRC 64 and 1.5 are some random factors that sounded good, 0.72 is the reference voltage and the rest are the voltage divider stuff 2019-10-23T02:02:25 < zyp> I think so 2019-10-23T02:02:49 < zyp> https://bin.jvnv.net/file/16UQA.pdf 2019-10-23T02:02:57 < karlp> just got to finish moving the test gear down to the new office 2019-10-23T02:04:10 < karlp> oh heh, you actually used seven series resistors like they said :) 2019-10-23T02:05:06 < karlp> now, if only I could get a fucking console on this alpine rootfs. 2019-10-23T02:07:16 < mawk> docker ? 2019-10-23T02:07:50 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-23T02:10:02 < karlp> no, testing some allwinner shit that I need a new kernel for, and need a suitable rfs for it. 2019-10-23T02:10:50 < qyx> are you doing allwinner shit? 2019-10-23T02:12:17 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-23T02:12:28 < BrainDamage> is the chip mainlined or still using the crappy kernel? 2019-10-23T02:13:38 < mawk> I saw allwinner support in recent 5.* kernels 2019-10-23T02:14:34 < karlp> BrainDamage: oh, it works fine for ~most things in 4.19, with a nice openwrt rootfs 2019-10-23T02:14:49 < karlp> I'm just trying to get the bluetooth portions working so I can send teh dts upstream properly, 2019-10-23T02:14:56 < karlp> and htat needs stuff that only landed in 5.2 2019-10-23T02:15:51 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 250 seconds] 2019-10-23T02:15:57 < karlp> and my own 5.3 kernel with openwrt's 4.19 rootfs plus their stuff makes the wireless stop working, and still doesn't give me bluettoh, and trying to make my own rfs is just giving me pain and heartache and 10000 "recipes" of cargo culting based on 15 years of how things have been at various stages. 2019-10-23T02:18:36 < mawk> their rootfs but still your initramfs ? 2019-10-23T02:18:50 < mawk> but I assume you built without external modules tho 2019-10-23T02:19:02 < karlp> well, their rootfs _as_ my initramfs (I'm just doing it all on initramfs) 2019-10-23T02:19:08 < karlp> yes, no modules, all built in 2019-10-23T02:19:10 < mawk> ah good 2019-10-23T02:21:43 < kakipro> Linux HP-Compaq-Pro-6300-MT 5.3.4-050304-lowlatency #201910051526 SMP PREEMPT Sat Oct 5 15:31:01 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux 2019-10-23T02:26:56 < karlp> 2019: "You should never use make -jN with Buildroot: top-level parallel make is currently not supported" 2019-10-23T02:29:18 < mawk> yeah 2019-10-23T02:29:22 < mawk> I already saw that 2019-10-23T02:29:39 < mawk> one of my teachers was one of the authors of buildroot, he stuffed it down our throats 2019-10-23T02:30:07 < mawk> two of my other teachers were among the authors of autotools, I think this school produced the most detested software of all time 2019-10-23T02:35:41 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away] 2019-10-23T02:37:22 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-23T03:01:01 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-23T03:02:11 < kakipro> what school? 2019-10-23T03:02:36 < mawk> EPITA 2019-10-23T03:03:57 < kakipro> kaki32 now> 2019-10-23T03:04:24 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-23T03:06:12 -!- kaki32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-23T03:21:48 * karlp watches autotools get to busily wasting time... speaking of :) 2019-10-23T03:21:57 < dongs> lol lunix 2019-10-23T03:22:43 < karlp> the worst is the people who "help" projects by turning up and going, "oh, you have your own clean simple makefile? let me fix that for you by converting to autohell so that it will build on sunos4.99" 2019-10-23T03:22:47 < karlp> lol indeed :) 2019-10-23T03:25:09 < dongs> right?? 2019-10-23T03:26:41 < karlp> rebuilding host python3. ffs. 2019-10-23T03:27:50 < karlp> docs says it just needs python2.7 or later. apparently that's just so it was able to build a 3.7. 2019-10-23T03:32:21 < dongs> https://techcrunch.com/2019/04/06/gps-rollover-is-today-heres-why-devices-might-get-wacky/ lol wut 2019-10-23T03:32:24 < dongs> how did i miss this 2019-10-23T03:34:15 < aandrew> clearly your GPS synchronized calendar rolled over 2019-10-23T03:34:50 < aandrew> wtf 2019-10-23T03:34:50 < aandrew> https://en.wikichip.org/wiki/hisilicon/kunpeng/920-6426 2019-10-23T03:36:33 < dongs> 8 octets, 16 nigglets 2019-10-23T03:42:27 < dongs> geeeez 2019-10-23T03:43:01 < dongs> i downloaded 3D model for a fucking headphoen jack from a jap company that i needed to make enclosure and they fuckign spammed me from a local branch "plz buy" 2019-10-23T03:43:04 < dongs> wtf 2019-10-23T03:44:56 -!- con3 [~kvirc@154.119.40.237] has quit [Remote host closed the connection] 2019-10-23T03:49:25 < jadew> I asked on the HP group, so it's official 2019-10-23T03:49:30 < jadew> nobody needs my product anymore 2019-10-23T03:50:18 < jadew> if I had it out like... 4 years ago, maybe 2019-10-23T03:53:46 < jadew> there's probably some wisdom in this story 2019-10-23T03:55:39 < jadew> "execute ideas when they come to you"? - probably this 2019-10-23T03:57:24 < jadew> to put salt on the injury, when I started working on this, I needed it 2019-10-23T03:57:27 < jadew> now I don't... 2019-10-23T04:04:30 < jadew> oh well, I have other projects almost ready, but this one was the one I enjoyed the most 2019-10-23T04:05:03 < jadew> turns out making a tracking generator in such a small space is extremely difficult 2019-10-23T04:20:11 < dongs> im trying to repair this dj headphone thingy and theres 3 wires at the connector 2019-10-23T04:20:24 < dongs> red blue and gold 2019-10-23T04:20:28 < dongs> is that some standard coloring 2019-10-23T04:24:59 < emeb_mac> must have gold - djs love the bling 2019-10-23T04:25:08 < dongs> heh 2019-10-23T04:26:36 < jadew> what does it matter? just put the meter on them and listen 2019-10-23T04:27:23 < dongs> the what meter? 2019-10-23T04:27:29 < dongs> tell m more 2019-10-23T04:27:55 < jadew> DMM in diode mode maybe? you should hear a pop in the correct ear piece 2019-10-23T04:30:48 < dongs> i gotta tin the ends first 2019-10-23T04:30:54 < dongs> its one of those weird unconductive shits 2019-10-23T04:40:49 < doomba> yea those little headphone wires are a bitch 2019-10-23T04:41:04 < doomba> i used to use a mig welder as a soldering iron to repair them in prison 2019-10-23T04:42:52 < BrainDamage> they are generally lacquered 2019-10-23T04:43:07 < BrainDamage> with some you can burn the lacquer, with others you have to physically scrape it 2019-10-23T04:43:13 < BrainDamage> once you sanded it, you can solder 2019-10-23T04:43:33 < jadew> is it not enamel? 2019-10-23T04:44:33 < BrainDamage> bleh, sorry, you're correct 2019-10-23T04:44:44 < BrainDamage> lacquer is a subtype of enamel 2019-10-23T04:44:48 < BrainDamage> generic term is enamel 2019-10-23T04:45:19 < jadew> if your soldering iron is hot enough, you can tin them sometimes 2019-10-23T04:45:32 < BrainDamage> that's the subtype that can be burned 2019-10-23T04:45:38 < BrainDamage> not all of them can 2019-10-23T04:45:39 < dongs> yeah it tinned right away with a normal temperature iron 2019-10-23T04:46:39 < jadew> some are impossible to get off with just an iron 2019-10-23T04:47:28 < jadew> I remember when I was little, there were schematics for enamel burners in electronics books 2019-10-23T04:47:38 < dongs> https://i.imgur.com/ajmKzSj.jpg 2019-10-23T04:47:45 < jadew> basically a coil that would heat up, and you would put the wire in 2019-10-23T04:48:07 < dongs> anyway just touching the end with iron worked great 2019-10-23T04:48:22 < dongs> i donno if the outer coating is some kinda cloth or something shit but it burned right off. 2019-10-23T04:48:30 < dongs> and yeah diode mode i could tap the stuff and hear clicks 2019-10-23T04:48:32 < dongs> great tip 2019-10-23T04:48:38 < jadew> np 2019-10-23T04:49:38 < dongs> https://www.slashgear.com/lenovo-yoga-book-c930-hands-on-specifications-price-availability-ifa-2018-30543732/ i fucked with this at costco a week ago 2019-10-23T04:49:44 < dongs> fucking terrible lol 2019-10-23T04:49:53 < dongs> e-paper keyboard with a bit of haptic feedback 2019-10-23T04:50:40 < jadew> so on the front it's a regular tablet, on the back an e-reader and if you fold it like that a laptop? 2019-10-23T04:50:44 < jadew> that's kinda neat 2019-10-23T04:51:03 < dongs> yes but its useless as laptop with that kinda keyboard 2019-10-23T04:51:28 < doomba> they keyboard buttons are epaper? 2019-10-23T04:51:31 < dongs> yes 2019-10-23T04:52:09 < jadew> lenovo is the kind of company that cashes in on the business hipsters 2019-10-23T04:52:26 < dongs> absolutely, just like they cashed in on the retarded lunxi faggots 2019-10-23T04:52:26 * jadew has their first tablet :P 2019-10-23T04:52:34 < doomba> oh god 2019-10-23T04:52:34 < dongs> who thought thinkpads were useful or something 2019-10-23T04:52:36 < dongs> ugh disgusting shit 2019-10-23T04:53:04 < jadew> the original thinkpads were great, but they gradually destroyed the brand 2019-10-23T04:53:25 * doomba pets his thinkpad running arch linux while whispering m'lady to an imagined princess in need of white-knighting. 2019-10-23T04:54:33 < jadew> who the hell needs a titanium lid on their laptop? 2019-10-23T04:54:56 < doomba> business hipsters 2019-10-23T04:55:17 < jadew> I thought it was cool back then, but now I realize just what a gimmick that was 2019-10-23T04:55:34 < dongs> how the fuck do you display arabic on LCD 2019-10-23T04:55:41 < dongs> isnt it one of those languages that neesd to be composed 2019-10-23T04:56:03 < dongs> like current letter modifies what previous ones look like 2019-10-23T04:56:11 < jadew> like japanese? 2019-10-23T04:56:14 < dongs> no 2019-10-23T04:56:27 < dongs> japanese is 1:1 theres no weirdness there 2019-10-23T04:56:33 < jadew> I saw an asian chick once typing several letters into a single one 2019-10-23T04:56:42 < jadew> I thought she's japanese 2019-10-23T04:56:51 < dongs> thats normal, you don't need to display the intermediate stuff 2019-10-23T04:57:28 < dongs> https://www.instructables.com/id/Write-in-Arabic-in-Arduino-Uno/ 2019-10-23T04:57:30 < dongs> lmao 2019-10-23T04:57:31 < dongs> allah.ino 2019-10-23T04:57:39 < jadew> lol 2019-10-23T04:58:20 < jadew> are there no arabic keyboards? 2019-10-23T04:58:44 < doomba> allah.ino IED8266? 2019-10-23T04:58:50 < jadew> btw, arabic is a good language to learn 2019-10-23T04:58:55 < dongs> https://github.com/idreamsi/arduino-persian-reshaper 2019-10-23T04:58:59 < jadew> lots of people speak it 2019-10-23T04:59:00 < dongs> yeah this is what im talking about 2019-10-23T04:59:05 < dongs> The characters change shape according to their surrounding characters. 2019-10-23T05:05:03 < doomba> e-ink touchpad keyboard. still can't get over this. 2019-10-23T05:05:06 < doomba> people will LOVE it too 2019-10-23T05:05:18 < dongs> well, its been on sale for a while. shit's even at costco 2019-10-23T05:05:22 < dongs> didnt look at price 2019-10-23T05:05:25 < dongs> it felt really tacky 2019-10-23T05:05:35 < dongs> like you prss key and epaper updates to make it look like you pressed it 2019-10-23T05:05:43 < dongs> and there's a little buzz from haptic feedback 2019-10-23T05:05:46 < doomba> and that update is probably slow. 2019-10-23T05:05:50 < dongs> of course 2019-10-23T05:06:11 < doomba> because who needs to type more than 10wpm? you should be swiping and tapping on ads, consumer. 2019-10-23T05:06:50 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Disconnected by services] 2019-10-23T05:06:51 < jadew> they can now pop ads right under your fingers 2019-10-23T05:06:52 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-23T05:06:56 < dongs> lewl 2019-10-23T05:07:11 < dongs> i guess the only real positive thing about it is you can change keyboard layouts 2019-10-23T05:07:16 < doomba> yep. make you click on products you don't even know you're about to buy! 2019-10-23T05:07:25 < dongs> oh and there's no touchpad 2019-10-23T05:07:28 < dongs> since entier thing is touchpad 2019-10-23T05:07:32 < doomba> but i can change keyboard layouts with setxkbmap 2019-10-23T05:07:36 < dongs> so you press on spacebar and it opens up a little touchpad area 2019-10-23T05:07:57 < dongs> dumbo, nomrmal people dont use lunix 2019-10-23T05:08:01 < dongs> or know what the fuck set xkbmap is 2019-10-23T05:08:10 < dongs> also, japanese keyboard layout is physically different 2019-10-23T05:08:13 < doomba> i know. 2019-10-23T05:08:25 < dongs> enter key, position of [], \ backspace, etc 2019-10-23T05:08:26 < doomba> all the normies use Windows 10 Telemetry Edition 2019-10-23T05:08:34 < dongs> yeah i love my win10 1903 2019-10-23T05:08:40 < dongs> i dont even care what telemetry it keeps on me 2019-10-23T05:08:44 < dongs> because it just works 2019-10-23T05:09:17 < doomba> remember those laser projection keyboards? 2019-10-23T05:10:17 < doomba> this is like that 2019-10-23T05:10:29 < BrainDamage> I don't see how's better or worse than a touchscreen keyboard 2019-10-23T05:11:27 < dongs> wow 2019-10-23T05:11:37 < dongs> so i sold some shit to brazil for around 400$ last year 2019-10-23T05:11:41 < dongs> they finally paid me for it 2019-10-23T05:11:44 < jadew> lol 2019-10-23T05:11:52 < dongs> i was like "who teh fuck is this transfer coming from" 2019-10-23T05:13:21 < dongs> the package also spent liek 2 months in customs 2019-10-23T05:13:27 < dongs> brazil must be a really shitty place to live in 2019-10-23T05:14:09 < BrainDamage> iirc brazil has very tight custom policies for electronic stuff "to make the internal market to survive" 2019-10-23T05:14:24 < dongs> yeah 2019-10-23T05:14:25 < BrainDamage> which means huge delays at customs and they are behind more than a decade 2019-10-23T05:14:38 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-23T05:15:50 < dongs> https://github.com/idreamsi/arduino-persian-reshaper/blob/master/Persian_Letters_Arduino.ino gawd that code 2019-10-23T05:16:04 < dongs> this is truly how fucking retarduiners code 2019-10-23T05:17:38 -!- catphish_ [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-23T05:21:15 < kaki32> oh wow 2019-10-23T05:22:36 < kaki32> whar are you using it for? 2019-10-23T05:23:01 < dongs> gotta make user-friendly IEDs 2019-10-23T05:23:22 < kaki32> ikr 2019-10-23T05:25:19 < kaki32> but seriously? 2019-10-23T05:25:26 < dongs> yeah, seriously 2019-10-23T05:25:30 < dongs> need to have some menus in arabic 2019-10-23T05:25:32 < dongs> on a lcd 2019-10-23T05:25:41 < dongs> already have them in jap/gook/chink 2019-10-23T05:29:00 < kaki32> remeber text needs to be right to left 2019-10-23T05:29:48 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-23T05:31:14 < kaki32> sounds better than your startup assembly gigs 2019-10-23T05:32:51 < dongs> laugh 2019-10-23T05:39:35 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-23T06:10:29 -!- qyx [~qyx@gw2.krtko.org] has quit [Ping timeout: 265 seconds] 2019-10-23T06:10:59 -!- qyx [~qyx@84.245.120.125] has joined ##stm32 2019-10-23T06:14:37 < jadew> BrainDamage, I heard that about Argentina 2019-10-23T06:15:24 < jadew> in the context of people complaining about how they can't get good equipment 2019-10-23T06:15:48 -!- qyx [~qyx@84.245.120.125] has quit [Ping timeout: 265 seconds] 2019-10-23T06:15:51 -!- qyx_ [~qyx@gw2.krtko.org] has joined ##stm32 2019-10-23T06:21:51 -!- fc5dc9d4_ [~quassel@p5B08107B.dip0.t-ipconnect.de] has joined ##stm32 2019-10-23T06:25:17 -!- fc5dc9d4 [~quassel@p5B0810CD.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-23T07:31:00 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-23T07:33:57 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-23T07:33:57 -!- day__ is now known as day 2019-10-23T08:07:53 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-23T08:08:26 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-23T08:11:37 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-23T08:24:38 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-23T08:43:17 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-23T08:45:48 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-23T08:54:21 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-23T09:02:52 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-23T09:11:22 -!- jly [uid355225@gateway/web/irccloud.com/x-qnakskxnjfecktnc] has joined ##stm32 2019-10-23T09:13:41 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-23T09:30:03 < jly> good morning Haohmaru 2019-10-23T09:38:37 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-23T09:55:13 < Steffanx> Is it a good morning, do you wish him one, do you want one. What's it jly? 2019-10-23T09:55:36 < jly> j,l,y 2019-10-23T10:02:34 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-23T10:03:33 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2019-10-23T10:04:26 < upgrdman> in altidongs, when making a footprint, in 2d view, is there any way to see the 3d model (with ortho perspective of course) ? trying to line up silkscreen lines with part of a 3D shape 2019-10-23T10:05:40 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-23T10:14:45 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-23T10:25:21 < Haohmaru> upgrdman u can.. in kicad 2019-10-23T10:25:22 * Haohmaru runs 2019-10-23T10:26:09 < Haohmaru> hai jey el why 2019-10-23T10:29:53 < jly> very good 2019-10-23T10:44:25 < jpa-> stupid trick of the day: you know how annoying it is to write stuff like array[ARRAY_LEN - 1 - index] when you want to index from the end? now you can save many characters by writing array[ARRAY_LEN + ~index] 2019-10-23T10:48:03 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 268 seconds] 2019-10-23T10:51:20 < Steffanx> Ty jpa- 2019-10-23T10:51:35 < Steffanx> Do you also use the !! Operator much? 2019-10-23T10:52:41 < jpa-> no, i don't trust bools anymore 2019-10-23T10:53:38 < Steffanx> Are your bools bipolar? 2019-10-23T10:55:16 < jpa-> no, and that's the problem! 2019-10-23T10:55:42 < jpa-> they're FILENOTFOUND and my compiler goes all "muahahahaa, now i can do whatever i want" 2019-10-23T10:56:33 < Steffanx> jly: can give you some files 2019-10-23T10:56:55 < jly> jly sleeps in the cat box 2019-10-23T10:57:08 < jly> or plays dead, steffan can show you 2019-10-23T10:59:06 < Steffanx> You mean: http://imgur.com/a/JbJgRCt ? 2019-10-23T10:59:36 < jly> yeah all its missing is ants crawling all over :\ 2019-10-23T11:07:58 < Haohmaru> jpa- i don't usually do it that way 2019-10-23T11:08:20 < Haohmaru> moar like index = len-1; array[index] 2019-10-23T11:09:02 < Steffanx> You use anything like this? Dont you have c++ magic for this? 2019-10-23T11:09:40 < Haohmaru> if you mean the fancy iterator stuff - i don't use that 2019-10-23T11:09:54 < Haohmaru> and that comes from std:: i don't have that 2019-10-23T11:10:50 < Haohmaru> how much magical can array access be? 2019-10-23T11:10:54 < Haohmaru> i got magic elsewhere ;P~ 2019-10-23T11:18:05 < Haohmaru> http://paste.debian.net/hidden/bbec996e/ 2019-10-23T11:19:29 < jpa-> Haohmaru: sure, if you are only going in the array in one direction 2019-10-23T11:19:48 < Haohmaru> i use multiple indeggzes then 2019-10-23T11:24:27 < Haohmaru> i forgot teh akchual template... here it is: http://paste.debian.net/hidden/c51917d4/ 2019-10-23T11:24:58 < Haohmaru> compile time tim0r calculat0r 2019-10-23T11:25:27 < Haohmaru> for le eggzmega 2019-10-23T11:28:03 < Haohmaru> Steffanx r u not impressed? ;P~ 2019-10-23T11:28:28 < Haohmaru> i had forgotten i wrote this, and now i'm looking at it and i'm kinda impressed ;P~ 2019-10-23T11:28:35 < Steffanx> I'm still recovering 2019-10-23T11:28:40 < Haohmaru> aww 2019-10-23T11:30:08 < Haohmaru> Steffanx basically it brute-forces teh possible tim0r prescalers to find a setting with least error.. all at compile time, and then the final code is just the SFR assignments (with compile time constants) in the last function 2019-10-23T11:31:18 < Haohmaru> ehm the prescaler and the period 2019-10-23T11:33:56 < Haohmaru> 2019-10-23T11:36:44 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-23T11:38:26 < jly> hmm 2019-10-23T11:38:48 < jly> i think steffan should help stm8 2019-10-23T11:39:21 < jly> we're using st visual disaster 2019-10-23T11:47:15 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Ping timeout: 268 seconds] 2019-10-23T11:49:16 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-23T11:49:16 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-23T11:49:16 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-23T11:58:55 < zyp> karlp, figured that was easier than picking some resistor long enough that it's okay to put 230V across it 2019-10-23T11:59:09 < zyp> considering creepage and whatever 2019-10-23T12:00:46 < Steffanx> Yeah with the cosmic compiler, jly? 2019-10-23T12:01:12 < jly> ye 2019-10-23T12:01:30 < jly> there's another option i've never had to pick 2019-10-23T12:01:41 < jly> like it compiles sea code right? 2019-10-23T12:02:24 < jly> all i wanted stm8 to do is supervise some stuff and spew out i2c 2019-10-23T12:11:55 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-23T12:20:48 < Steffanx> We used it to replace some i2c-7segment driver which was obsolete. The end result with the stm8 was even cheaper than the original product 2019-10-23T12:25:14 < jly> yeah i find things like that happen more often than expected 2019-10-23T12:25:28 < jly> replace xyz with something that should be more expensive... 2019-10-23T12:38:17 < Haohmaru> replace craptium with kicad 2019-10-23T12:38:19 * Haohmaru runs 2019-10-23T12:40:45 < doomba> altrightium divider 2019-10-23T12:44:49 < Haohmaru> Artherial Designer 2019-10-23T12:45:03 < Haohmaru> much blud, very flow 2019-10-23T12:45:24 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has joined ##stm32 2019-10-23T12:45:53 < ronox> hi, what is the GIPO number of PA0? i cant find it in the datasheet for the stm32f407vet6 2019-10-23T12:49:28 < jpa-> uh.. GPIOA 0? what other GPIO number is there? 2019-10-23T12:50:33 < ronox> i tried GPIOA0, it errord 2019-10-23T12:51:08 < ronox> im using the standard peripheral lubrary the code is GPIO_InitStructure.GPIO_Pin = GPIOA0; 2019-10-23T12:52:10 < jly> https://www.youtube.com/watch?v=UWoQMuzeA-g 2019-10-23T12:52:15 < jly> they're on their way 2019-10-23T12:52:29 < jly> you have 20 minutes to learn this before the nukes miss you and you surive 2019-10-23T12:52:34 < jly> *survive 2019-10-23T12:52:55 < jly> .................. seriously 2019-10-23T12:53:25 < jly> bad news for yanks 2019-10-23T13:01:25 < Steffanx> ronox: GPIO_PIN_0 I think. Its just pin 0 of GPIO A. 2019-10-23T13:01:39 < Steffanx> It's better have a look at the HAL user manual 2019-10-23T13:01:59 < Steffanx> -It's 2019-10-23T13:02:00 < karlp> zyp: meh, what installation category were you hoping for? I mean, yeah, you can probably just use whaever resistors you were already using if you use seven of them, but fuck that :) 2019-10-23T13:03:26 < ronox> why would i look at the HAL user manual? 2019-10-23T13:03:29 < ronox> im not even using it 2019-10-23T13:03:50 < ronox> ah sorry i never specified my library 2019-10-23T13:04:03 < ronox> ty, gpio-pin-0 works 2019-10-23T13:04:32 < ronox> im currently using the stdperiphlib, does it still reffer to the on board LED as "LEDBUILTIN"? 2019-10-23T13:13:39 < ronox> could someone help me figure out how to switch on the built in LEDs for this boaard? the stm32f407vet6? 2019-10-23T13:20:07 < zyp> karlp, idk, I'm just ghettoing up a «good enough» prototype 2019-10-23T13:24:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-23T13:25:05 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-23T13:27:23 < karlp> ronox: that's not a board, that's a mcu. 2019-10-23T13:29:39 < Haohmaru> em see you 2019-10-23T13:29:53 < ronox> stm32f407vet6 F4VE 2019-10-23T13:30:08 < ronox> this one https://os.mbed.com/users/hudakz/code/STM32F407VET6_Hello/shortlog/ 2019-10-23T13:30:19 < Haohmaru> mmmm em bed 2019-10-23T13:30:48 < karlp> ok, doesn't it say what the led is there? you'r elike hte third person ina week trying to use this board all of a sudden, what happened?! 2019-10-23T13:31:09 < Haohmaru> karlp must be organized 2019-10-23T13:31:20 < Haohmaru> mbed brainwashes unsuspecting makers 2019-10-23T13:31:21 < karlp> https://os.mbed.com/media/uploads/hudakz/stm32f407vet6_left02.png 2019-10-23T13:31:27 < karlp> LED1 LED2? 2019-10-23T13:31:58 < karlp> also here https://github.com/libopencm3/libopencm3-miniblink/blob/master/boards.stm32.mk#L121 2019-10-23T13:32:22 < Haohmaru> maybe they promise them "take our boardz and it's all easy, u won't have to read datasheets*" "* if you've already read them" 2019-10-23T13:48:21 < ronox> so, when that pin is high/low the LED will also be high/low? 2019-10-23T13:50:36 < Haohmaru> ronox generally you should check the schematic of your board, i guess this one here: http://wiki.stm32duino.com/images/5/5c/STM32_F4VE_SCHEMATIC.PDF 2019-10-23T13:50:43 < Haohmaru> and see how teh LED is wired up 2019-10-23T13:50:50 < ronox> what schematic 2019-10-23T13:50:59 < ronox> that links broken 2019-10-23T13:51:50 < Haohmaru> generally i wouldn't buy a demo/dev board that has no schematic 2019-10-23T13:52:00 < Haohmaru> ;P~ 2019-10-23T13:52:05 < Haohmaru> find the schematic 2019-10-23T13:52:18 < Haohmaru> eyeball it, see teh truth 2019-10-23T13:52:31 < Haohmaru> 2019-10-23T14:09:57 < ronox> i didnt know that when i bought it 2019-10-23T14:10:08 < ronox> that none existed 2019-10-23T14:10:23 < ronox> anyway i figured it out, seems the pin is just in parallel with the LED 2019-10-23T14:13:23 < ronox> could someone tell me whats wrong with this code and why it wont work? https://pastebin.com/qaft5f0i 2019-10-23T14:13:42 < ronox> namely in the adc.c part with how i configure the ADC 2019-10-23T14:14:16 < ronox> this is from colorchord 2019-10-23T14:17:40 < ronox> ive done everything logically, but i cant figure out where the disconnect is 2019-10-23T14:18:07 < ronox> if its reading audio, or if its reading audio correct, i do know that the RBG driver part is working fine 2019-10-23T14:18:17 < ronox> especially since i never altered it 2019-10-23T14:29:00 < karlp> ronox: whether the led is high or low or low or high depends on how it was wired, but is easy for you to test locally. what do youy mean by "the pin is paralle with the led" ?! 2019-10-23T14:29:31 < karlp> what exactly are you asking people to look at in that jumbo paste? 2019-10-23T14:29:38 < karlp> what does/doesn't work? 2019-10-23T14:30:15 < ronox> should the ADC code function? 2019-10-23T14:30:28 < ronox> its acting like its not 2019-10-23T14:30:54 < karlp> "function" ? 2019-10-23T14:31:05 < ronox> read 2019-10-23T14:31:18 < karlp> "read" ? 2019-10-23T14:31:20 < ronox> should the ADC be reading voltages 2019-10-23T14:31:30 < karlp> what is your "read" call, where is it, what does it do actually? 2019-10-23T14:31:51 < ronox> what, no! 2019-10-23T14:31:53 < karlp> is all of taht code actually used? 2019-10-23T14:32:02 < karlp> get rid of the code that isn't used so people have a chance of reading it. 2019-10-23T14:32:14 < ronox> i dont know what code is or isnt used 2019-10-23T14:33:16 < ronox> i mean, the adc.c section, the way the adc pin is configured/set up, is there any issue with it? with that setup, should it be able to sample audio? 2019-10-23T14:34:12 < ronox> the colorchord F3 project uses this same method, reading audio over the ADC, the F4 uses an on board microphone, however, my F4 has none such microphone 2019-10-23T14:34:54 < ronox> so i replicated an equivalent ADC section using the equivalent initializer functions and the sampling code 2019-10-23T14:34:54 < karlp> what are you trying to trigger the adc with? you've got leftovers splatted everywhere. 2019-10-23T14:35:11 < karlp> are you trying to use external trigger ADC_ExternalTrigConv_T1_CC1 or are you trying to use a timer irq to manualyl read it one by one? 2019-10-23T14:36:02 < ronox> i dont know, i dont even really understand most of it, im just trying to make it work on my board 2019-10-23T14:36:13 < ronox> i can show you the original F3 code however 2019-10-23T14:36:15 < ronox> 1 sec 2019-10-23T14:36:40 < ronox> https://github.com/cnlohr/colorchord/blob/master/embeddedstm32f303/adc.c 2019-10-23T14:37:00 < ronox> https://github.com/cnlohr/colorchord for refference here is the original project 2019-10-23T14:37:58 < ronox> also ill clean up the paste 2019-10-23T14:38:26 < karlp> you just changed adc4 to adc1 and the adc pin number right? that's all you should have done at least. 2019-10-23T14:39:51 < Haohmaru> make -f Makefile -pls -makeitwork -pls -pls -j4 2019-10-23T14:40:13 < Haohmaru> C:\> pray.exe 2019-10-23T14:41:17 < ronox> https://pastebin.com/cC7Xxj1y 2019-10-23T14:41:51 < ronox> i also used an entirely different library and ported the init over 2019-10-23T14:42:01 < ronox> the F3 code is not compatible with the F4 2019-10-23T14:42:13 < ronox> half the functions are dofferent/omitted 2019-10-23T14:42:43 < ronox> the F4 colorchord doesnt even have an adc.c because it reads off a microphone 2019-10-23T14:43:45 < ronox> anyway, so what your saying is then, i did it correctly right? that adc code i converted should set up the ADC in the same configuration as is set up in the F3 version 2019-10-23T15:01:07 < kakipro> https://www.youtube.com/watch?v=Bt2KDtDt940 2019-10-23T15:10:37 < ronox> i think i give up 2019-10-23T15:11:05 < ronox> ill never figure out how this damn program works in like 5 years of strict study 2019-10-23T15:11:17 < ronox> you have to know how to use it to use it and that makes no sense 2019-10-23T15:11:30 < ronox> means learning is impossible 2019-10-23T15:13:12 < karlp> you're trying to bite too much at once. 2019-10-23T15:13:24 < karlp> stop trying to do a single "massive paste and replace, omg it doesn't work" 2019-10-23T15:13:34 < karlp> stop trying to go "colorchord port doesn't work" 2019-10-23T15:13:46 < karlp> go "how to get adc working _at all_ for my port." 2019-10-23T15:14:04 < karlp> perhaps stop trying to copy f3 code if you'r ehaving problems, copy _other_ f4 adc code. 2019-10-23T15:14:13 < karlp> look at f4 adc examples. 2019-10-23T15:18:26 -!- jly [uid355225@gateway/web/irccloud.com/x-qnakskxnjfecktnc] has quit [Quit: Connection closed for inactivity] 2019-10-23T15:24:04 < Steffanx> Did you ever try arduino, ronox? 2019-10-23T15:27:07 < ronox> yes 2019-10-23T15:27:12 < ronox> i can do quite a bit in it 2019-10-23T15:27:55 < ronox> plus i did a beginners class in C programming, which is why i felt confident enough to even attempt this to begin with 2019-10-23T15:35:40 < kakipro> aandrew: what type of zeropoint detection you expect me to have? I should have checked my hardware before going crazy with code 2019-10-23T15:35:53 < kakipro> input to opto is rectified 2019-10-23T15:36:30 < kakipro> there is rising and falling edge for every phase halve 2019-10-23T15:38:55 -!- tomeaton17_ is now known as tomeaton17 2019-10-23T15:39:32 < kakipro> so it's trivial then 2019-10-23T15:51:55 < karlp> until you get harmonics and it crosses more than once. 2019-10-23T15:59:56 < BrainDamage> mains frequency is generally pretty stable, you can lock on it and ignore any trigger that isn't close to what you espect 2019-10-23T16:01:49 < aandrew> kakipro: input opto isn't bad, maybe throw a small RC on that to try to knock off any high freq without inducing too much phase shift 2019-10-23T16:01:55 < aandrew> kakipro: but take that to a timer input 2019-10-23T16:02:10 < aandrew> something you can capture the timer value when it changes 2019-10-23T16:02:10 < kakipro> k 2019-10-23T16:02:33 < Haohmaru> aww fiahfohs crashz0r3d 2019-10-23T16:02:43 < aandrew> that gives you the most flexibility. if for whatever reason you don't want or like that you can always use it as a regular GPIO input and poll the shit/filter in software 2019-10-23T16:02:48 < kakipro> aandrew: phase shift can be corrected 2019-10-23T16:02:52 < kakipro> correct? 2019-10-23T16:02:55 < aandrew> but I think timer caputre with some smarts would be best 2019-10-23T16:03:02 < aandrew> kakipro: if you know the frequency yes 2019-10-23T16:03:05 < BrainDamage> phase shift can be corrected for a fixed freq 2019-10-23T16:03:11 < BrainDamage> by a spike doesn't have one 2019-10-23T16:03:19 * Haohmaru runs kakipro thru an allpass filter 2019-10-23T16:03:28 * Haohmaru modulates teh cutoff freq 2019-10-23T16:03:35 < kakipro> firing triacs is pro level jubb 2019-10-23T16:03:57 < BrainDamage> so keep a table of say, last 10 semiperiods, and take an average, and use that as a reference to detect spurious zero crossings 2019-10-23T16:04:20 < BrainDamage> and don't insert a trigger if it's more than say, 10% away from that value 2019-10-23T16:05:09 < BrainDamage> don't rely on a fixed value because mains can drift over time 2019-10-23T16:05:36 < aandrew> BrainDamage knows how to filter 2019-10-23T16:05:38 < kakipro> da 2019-10-23T16:07:00 < Haohmaru> da ama ne 2019-10-23T16:10:06 < kakipro> bbl> 2019-10-23T16:11:23 -!- talsit [foobar@gromit.mixdown.ca] has joined ##stm32 2019-10-23T16:12:01 < Haohmaru> wait a minute, where ya going 2019-10-23T16:12:27 < aandrew> mopedding 2019-10-23T16:12:58 < Haohmaru> hm, if you ride a chopper, you're probably a rocker 2019-10-23T16:13:08 < Haohmaru> if moped - smol rocker? 2019-10-23T16:13:49 < aandrew> moped = kpopfan 2019-10-23T16:13:57 < Haohmaru> aww noes 2019-10-23T16:15:12 < PaulFertser> https://en.wikipedia.org/wiki/Mod_(subculture) 2019-10-23T16:25:23 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has quit [Quit: Leaving] 2019-10-23T16:26:04 < Haohmaru> https://upload.wikimedia.org/wikipedia/commons/7/75/MODs_in_Box_Hill.jpg 2019-10-23T16:26:17 < Haohmaru> so, do they still take amphetaminez 2019-10-23T16:29:41 < PaulFertser> That scooter with whiskers (on the left) certainly looks like drug-inspired. 2019-10-23T16:30:35 < PaulFertser> The one on the right with four rear-view mirrors installed on crash bars (on the right) is no better. 2019-10-23T16:31:12 < PaulFertser> So amphetamines or not, but I'd bet they get high every now and then. 2019-10-23T16:31:50 < Haohmaru> it's sad if grown ups take drugs and turn into living (sorta) sh*t 2019-10-23T16:32:10 < PaulFertser> Probably they live a happy life, why not then? 2019-10-23T16:32:28 < PaulFertser> Not all drugs were created equal 2019-10-23T16:32:41 < Haohmaru> you say some drugs are "okay" ? 2019-10-23T16:33:19 < PaulFertser> Well, I personally do alcohol occassionally. 2019-10-23T16:34:02 < Haohmaru> alcohol is pretty old and known afaik 2019-10-23T16:34:34 < Haohmaru> but some of this new sh*t is very fishy 2019-10-23T16:34:40 < Haohmaru> bath salts? 2019-10-23T16:34:48 < Haohmaru> krokodil? 2019-10-23T16:35:04 < Haohmaru> sniffing glue? 2019-10-23T16:36:07 < Haohmaru> i know some makers like to make things with a lot of glue 2019-10-23T16:36:16 < Haohmaru> like, even pointless things 2019-10-23T16:36:27 < Haohmaru> i wonder if they just like teh smell of teh glue >:) 2019-10-23T16:38:36 < PaulFertser> These days https://en.wikipedia.org/wiki/Toluene is less common 2019-10-23T16:40:41 < BrainDamage> alcohol is just impossible to regulate since it's so easy to make 2019-10-23T16:41:02 < karlp> doesn't stop people trying :) 2019-10-23T16:42:28 < PaulFertser> Haohmaru: despite being well-researched alcohol is still among the most dangerous drugs, in russia more people die of alcohol abuse than all other drugs combined afaik. 2019-10-23T16:43:21 < Haohmaru> alcohol is pretty nasty, but i don't think it can be stopped 2019-10-23T16:43:56 < Haohmaru> at least it's usually visible and sniffable when someone is drunk 2019-10-23T16:46:47 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-23T16:46:57 < BrainDamage> PaulFertser: its danger is more of a consequence of availability 2019-10-23T16:47:17 < BrainDamage> https://upload.wikimedia.org/wikipedia/commons/7/7e/Rational_harm_assessment_of_drugs_radar_plot.svg 2019-10-23T16:47:59 < BrainDamage> with the exception of opioids, most drugs are "about there" 2019-10-23T16:48:25 < Haohmaru> noice plot 2019-10-23T16:50:24 * karlp learns abotu new drugs. 2019-10-23T16:50:50 < karlp> why did they have to go and put opiods on different arbitrary places around taht graph 2019-10-23T16:51:36 < BrainDamage> I think the graph is sorted by interpolated mean 2019-10-23T16:51:42 < BrainDamage> from smallest to largest 2019-10-23T16:51:46 < BrainDamage> clockwise 2019-10-23T16:52:15 < karlp> right, ok. 2019-10-23T16:52:34 < PaulFertser> BrainDamage: I have a feeling if weed was available instead of alcohol the results would be less devastating. 2019-10-23T16:53:00 < karlp> I was under the impression that (clean) heroin wasn't all that high on the physical harm scale, 2019-10-23T16:53:48 * PaulFertser too 2019-10-23T16:54:54 < mawk> yes 2019-10-23T16:55:04 < mawk> heroin is not neurotoxic nor any kind of toxic 2019-10-23T16:55:26 < BrainDamage> i think the graph is for the effective 'street value' quality of the drug 2019-10-23T16:55:28 < mawk> as are most opiates with the exception of synthetic shits like MT-45 2019-10-23T16:56:01 < BrainDamage> ( because there's no other way to assess it otherwise ) 2019-10-23T16:56:09 < mawk> the danger of street heroin is contaminants when injected, we call that "dust" 2019-10-23T16:56:28 < BrainDamage> you cannot exactly ask people to become drug addicts to evaluate eg social harm 2019-10-23T16:58:00 < mawk> but all the harm can be mitigated if you change the route of administration 2019-10-23T16:58:11 < mawk> smoked heroin is as dangerous as IV'd heroin but you can't inject yourself stupid things 2019-10-23T16:58:22 < mawk> you'll maybe ruin your lungs overtime but that's better than nothing 2019-10-23T17:00:23 < mawk> your graph is a bit odd although BrainDamage 2019-10-23T17:00:26 < mawk> I'd rate alcohol much higher 2019-10-23T17:00:35 < mawk> morbid alcoholism is one of the worst conditions you can witness 2019-10-23T17:00:45 < mawk> on a physical level 2019-10-23T17:21:40 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-23T17:25:13 -!- Datz [~datz@unaffiliated/datz] has quit [Ping timeout: 265 seconds] 2019-10-23T17:25:45 -!- Datz [~datz@unaffiliated/datz] has joined ##stm32 2019-10-23T17:48:23 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 276 seconds] 2019-10-23T17:51:35 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-23T17:51:47 < bitmask> man I don't know anything 2019-10-23T17:52:11 < zyp> start learning then 2019-10-23T17:52:29 < bitmask> how far away from stuff do you need to keep a switching supply and does it need to be away from anything or just like adc stuff? 2019-10-23T17:53:21 < Haohmaru> put it in a lead box 2019-10-23T17:53:27 < bitmask> :) 2019-10-23T17:53:37 < Haohmaru> wrap teh box in tin foil 2019-10-23T17:53:47 < Haohmaru> for that extra 0.1% 2019-10-23T17:53:47 < bitmask> and then another lead box 2019-10-23T17:54:22 < Haohmaru> that would be innovative 2019-10-23T17:55:46 < bitmask> like this I kinda like, its on the bottom of a mosfet and some resistors but it fits pretty nicely, everything is on the bottom except the electrolytic and the feedback trace has a couple vias to jump over the gnd https://i.imgur.com/eFDMope.png 2019-10-23T17:55:59 < bitmask> you do need to to physically connect the ground pin with the tab right? 2019-10-23T17:56:17 < bitmask> I know its internally connected but for low impedance or whatever 2019-10-23T17:57:05 < bitmask> the polygon pours arent right yet, the connection is too thin, I just havent made a rule yet for them 2019-10-23T17:58:09 < BrainDamage> jadew: https://cdn.discordapp.com/attachments/438360283514404875/636576225896431636/Darkpainfulant_8416ed_7360629.png 2019-10-23T17:58:37 < bitmask> that made me smile 2019-10-23T17:58:50 < Steffanx> Me too, me too :) 2019-10-23T17:59:23 < Steffanx> I might even have made a little laughing sound 2019-10-23T17:59:29 < bitmask> :) 2019-10-23T17:59:32 < Steffanx> But I'm alone in the office. 2019-10-23T17:59:38 < BrainDamage> bitmask: not just low impedance, but larger thermal transfer too 2019-10-23T17:59:57 < BrainDamage> most of the heat will flow through the tab 2019-10-23T18:00:15 < BrainDamage> and if you have a large copper pour attached to it, it'll act as heatsink 2019-10-23T18:00:25 < bitmask> so any thoughts on my pic? its an lm2595 to263 2019-10-23T18:01:04 < bitmask> yea I was gonna do some vias for the gnd to the top for heat sinking 2019-10-23T18:01:15 < bitmask> I think gnd plane on top of board is easiest so thats what im doing 2019-10-23T18:02:07 < bitmask> plus I wont be stressing the IC, maybe 1/4-1/3rd rated power 2019-10-23T18:03:17 < karlp> that discord link, I saw that joke earlier today, and I still don't get it. what's it trying to say? 2019-10-23T18:03:47 < bitmask> https://youtu.be/A306Y7a4NRo 2019-10-23T18:03:57 < zyp> karlp, dragostea din tei 2019-10-23T18:04:44 < karlp> right. thanks. so totally glad I now know..... 2019-10-23T18:04:49 < zyp> you asked 2019-10-23T18:08:25 < bitmask> in altium, I have a bunch of vias to connect a pad to a polygon on the other side. I have everything on the same net but the polygon doesnt fill because the vias are there. any ideas? 2019-10-23T18:08:39 < karlp> I did, said I was glad didn't I? :) 2019-10-23T18:10:27 < aandrew> #drugchats 2019-10-23T18:10:40 < bitmask> drugs are bad mmmkay 2019-10-23T18:11:15 < aandrew> bitmask: I have rules for most of my power/gnd pours that there is no thermal reliefs at all 2019-10-23T18:11:40 < aandrew> haven't had any real issues with tombstoning when doing a proper preheat/soak of the board before entering the reflow profile 2019-10-23T18:12:48 < aandrew> bitmask: if you can, drop a shitload of vias there; they'll fill with solder and act as more of a heatsink 2019-10-23T18:13:31 < bitmask> yea I'll have to figure out the vias soon but I gotta figure out how to get these vias to cooperate with the polygon pour 2019-10-23T18:13:38 < aandrew> bitmask: you add a rule 2019-10-23T18:14:08 < bitmask> I dont think its a thermal relief issue 2019-10-23T18:14:42 < bitmask> https://i.imgur.com/1bbtbgi.png 2019-10-23T18:14:46 < bitmask> like see the blue outline 2019-10-23T18:14:51 < bitmask> that wont fill at all 2019-10-23T18:14:55 < bitmask> because those vias 2019-10-23T18:15:48 < bitmask> if I delete the vias it fills up 2019-10-23T18:16:20 < bitmask> i'll try no thermal relief but not sure what im looking for 2019-10-23T18:17:09 < aandrew> bitmask: https://imgur.com/a/pH2hziM 2019-10-23T18:18:06 < bitmask> hehe ahh thanks 2019-10-23T18:18:28 < aandrew> if I dicked around with net classes I could probably just say InNetClass() but whatever 2019-10-23T18:22:52 < bitmask> one more question, I have a trace and a pour the same net but I can't connect the trace to the polygon, pads to polygons works though 2019-10-23T18:23:31 < zyp> there's a parameter on the polygon pour, set it to pour over all same net objects 2019-10-23T18:23:41 < bitmask> thanks 2019-10-23T18:24:05 < bitmask> sweet 2019-10-23T18:24:09 < bitmask> ok now im getting somewhere 2019-10-23T18:24:12 < bitmask> thanks guys 2019-10-23T18:25:45 < karlp> (why would that not be the defualt?!) 2019-10-23T18:34:14 < bitmask> what size vias for heatsinking and for filling up pads to connect to other side, 0.3mm? 2019-10-23T18:35:56 < zyp> I prefer only using minimum size vias and just adding more of them if I want more conductivity 2019-10-23T18:36:35 < bitmask> ok so more like 0.2, I think thats about minimum 2019-10-23T18:40:31 < bitmask> think its ok the route the feedback pin of a smps around the IC? its either that or keep it shorter but use two vias to jump a trace 2019-10-23T18:40:52 < bitmask> routing it around should keep it further away from noise 2019-10-23T18:45:07 < bitmask> oh 0.3 is minimum 2019-10-23T18:50:33 < mawk> help help help Steffanx , I'm trying to recover my dutch citizenship and I sense they're asking trick questions "do you have another nationality other than the french nationality ?" if I say yes will they reject my application ? 2019-10-23T18:51:50 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-23T18:55:31 < Steffanx> mawk: is this something I could know? I'm Dutch-only;) 2019-10-23T18:55:37 < mawk> :( 2019-10-23T18:55:51 < Steffanx> Where is this? 2019-10-23T18:56:01 < mawk> https://www.paysbasmondial.nl/pays/france/resider-et-travailler/demander-un-passeport-ou-une-carte-didentite-si-vous-residez-a-letranger 2019-10-23T18:56:09 < mawk> on the NL website for recovering citizenship 2019-10-23T18:56:26 < mawk> here is it in dutch: https://www.nederlandwereldwijd.nl/wonen-werken/paspoort-en-id-kaart/paspoort-of-id-kaart-aanvragen-als-u-in-het-buitenland-woont 2019-10-23T18:57:00 < mawk> I have FR but also LU 2019-10-23T18:59:41 < Steffanx> Wouldn't it be easier to visit the dutch embassy or something? 2019-10-23T18:59:51 < Steffanx> There is one in paris 2019-10-23T19:00:03 < mawk> I called them, it took 30 minutes of waiting, they said "just go on the website" 2019-10-23T19:00:11 < mawk> you can't go there if you didn't fill in the form you have on the website anyway 2019-10-23T19:00:42 -!- Laurenceb [~laurence@14.200.208.46.dyn.plus.net] has joined ##stm32 2019-10-23T19:02:27 -!- learningc [~pi@121.122.105.195] has quit [Ping timeout: 240 seconds] 2019-10-23T19:02:32 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-23T19:02:45 < mawk> I want a ton of documents to get stamped by city hall and make officially translate to dutch 2019-10-23T19:02:48 < Steffanx> They cannot deny it when you have it, can they? 2019-10-23T19:02:53 < mawk> yeah I guess 2019-10-23T19:03:19 < mawk> the rules are funny, if your parents weren't married and your father didn't acknowledge you before you're 7 you must do a DNA test and have your father take care of you for 3 years 2019-10-23T19:04:02 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-23T19:04:36 < Steffanx> Lol 2019-10-23T19:05:13 < Steffanx> There are also situations where they can ask/force you to give up the other nationalities, but no clue how that works 2019-10-23T19:05:27 < mawk> I don't know if I'd agree to that 2019-10-23T19:05:43 < mawk> yeah before a recent law that's what they did, but recently they allowed 2 nationalities 2019-10-23T19:06:11 < mawk> but I don't know about 3 nationalities 2019-10-23T19:06:53 < mawk> ohhhh omg the procedure only works if your father was dutch 2019-10-23T19:06:58 < mawk> if it was a dutch mother you don't become dutch 2019-10-23T19:06:58 < mawk> lol 2019-10-23T19:07:03 < mawk> sexist dutchland 2019-10-23T19:08:36 < mawk> "You were married before 1 January 1985 to a non-Dutch husband and have therefore lost your Dutch nationality." ultra sexist dutchland ! 2019-10-23T19:10:08 < mawk> and you have to divorce to recover the nationality 2019-10-23T19:10:34 < Steffanx> Haha, old rules I guess 2019-10-23T19:10:52 < mawk> yeah 2019-10-23T19:11:04 < mawk> but still, if you're a dutch mother and a non-dutch father, your child aren't dutch, and that law still exists today 2019-10-23T19:11:23 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-23T19:11:24 < mawk> I hope twitter doesn't find out 2019-10-23T19:11:40 < mawk> #BoycottTheNetherlands #MeToo 2019-10-23T19:11:44 < Steffanx> Heh. 2019-10-23T19:11:52 < Steffanx> *dutchland 2019-10-23T19:12:13 < mawk> I saw the government started a communication campaign so that people quit saying "holland" 2019-10-23T19:12:27 < mawk> everyone says "hollande" here, they don't know "pays-bas" 2019-10-23T19:23:53 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-23T19:24:14 -!- renn0xtk9 [~max@2a02:810d:1540:2448:4b4:1f00:dc21:e5bd] has joined ##stm32 2019-10-23T19:24:32 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-23T19:25:41 < Steffanx> Really? I had some French and it's the first thing we learn 2019-10-23T19:25:59 < Steffanx> Les pays-bas 2019-10-23T19:26:08 < mawk> yeah 2019-10-23T19:26:21 < mawk> I force people to say "pays-bas" when they say "hollande" 2019-10-23T19:26:33 < mawk> but otherwise they're not aware that holland is just a region 2019-10-23T19:29:37 < Steffanx> Hollande was your president 2019-10-23T19:30:07 < mawk> yeah 2019-10-23T19:30:12 < mawk> François Pays-Bas 2019-10-23T19:31:02 < Steffanx> Heh 2019-10-23T19:31:20 < karlp> if you have a french passport, why do you need a dutch one? don't the four freedoms cover you? 2019-10-23T19:31:53 < mawk> to get an appartment it's more reassuring to landlords 2019-10-23T19:32:04 < mawk> it's difficult to rent in the city I want 2019-10-23T19:32:25 < karlp> report them to efta for violating your four freedoms? 2019-10-23T19:32:29 < mawk> also with it I can get speeding tickets in france and not pay them 2019-10-23T19:32:36 < mawk> if I had a car 2019-10-23T19:33:10 < Steffanx> Or a license? 2019-10-23T19:33:14 < mawk> yeah landlords take great liberties in accepting or rejecting applications, it's even harder if you're not white 2019-10-23T19:33:19 < Steffanx> *ans 2019-10-23T19:33:22 < Steffanx> And. Damnit 2019-10-23T19:34:59 <@englishman> four freedoms and a briefcase full of passports is better than four freedoms and nothing 2019-10-23T19:36:22 < Laurenceb> muh freedumz 2019-10-23T19:36:42 <@englishman> welcome back unfunnylaurenceb 2019-10-23T19:36:49 < Laurenceb> freedumz to have shitty handgun and get shot by p-olice 2019-10-23T19:36:54 < Laurenceb> hi b& hammer 2019-10-23T19:37:12 < Laurenceb> this is the only place I'm not b& 2019-10-23T19:37:27 < Laurenceb> actually no I'm unb& on julay.world and wizardchan 2019-10-23T19:37:33 <@englishman> will you be changing that 2019-10-23T19:37:36 < Laurenceb> kek 2019-10-23T19:37:49 < Laurenceb> I got b& from endchan last week 2019-10-23T19:38:10 < Laurenceb> and I'm permab& from 4chan 2019-10-23T19:40:38 < mawk> lol 2019-10-23T19:40:40 < mawk> how is that possible 2019-10-23T19:41:10 < Laurenceb> spamming a cat appreciation thread with video of kittens being dismembered with chainsaws iirc 2019-10-23T19:42:12 < oz4ga> To much Skip Morrow? 2019-10-23T19:45:43 -!- mode/##stm32 [+b *!*@14.200.208.46.dyn.plus.net] by englishman 2019-10-23T19:45:43 -!- Laurenceb was kicked from ##stm32 by englishman [Laurenceb] 2019-10-23T19:49:23 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-23T19:49:45 < bitmask> damn, only two deliveries for lunch? order food people 2019-10-23T19:50:15 < bitmask> still got $16.40 for 40 minutes of driving 2019-10-23T20:01:46 -!- kleinjl [~kleinjl@2001:1388:2:7992:598f:d04b:fb5:2709] has joined ##stm32 2019-10-23T20:08:05 < kleinjl> hi, please anyone known the max throughput (mhz) of gpio via dma in F1 and F4 ? 2019-10-23T20:10:39 < zyp> I guess DMA would compete for bus access with the CPU 2019-10-23T20:10:57 < zyp> so it'd probably depend a lot on what the CPU would be doing meanwhile 2019-10-23T20:11:06 < zyp> what's your goal? 2019-10-23T20:12:59 -!- renn0xtek9 [~max@ip5f5bee0f.dynamic.kabel-deutschland.de] has joined ##stm32 2019-10-23T20:13:00 -!- renn0xtk9 [~max@2a02:810d:1540:2448:4b4:1f00:dc21:e5bd] has quit [Ping timeout: 252 seconds] 2019-10-23T20:18:40 < karlp> "fastest gpio togglign for internet poitns" probably 2019-10-23T20:18:44 < karlp> only reason to ask such a question 2019-10-23T20:30:58 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-23T20:50:54 < PaulFertser> FX2 got really popular for being able to do fast GPIO for arbitrary protocols. So there's some non-bs niche for that kind of tasks. 2019-10-23T20:53:48 < kakipro> is fx2 fpga? 2019-10-23T20:56:06 < Steffanx> Will you keep the food hot. Then I'll order bitmask 2019-10-23T20:56:19 < bitmask> sure, I gots a thermal bag thingy :) 2019-10-23T20:57:33 < brdb> any recommendations on good setups for GSM and GPS (GNSS?) for the STM32 Cortex-M series? 2019-10-23T20:58:16 < brdb> I know there's the Teseo II/III, though I haven't found good sources for it 2019-10-23T21:02:24 < PaulFertser> kakipro: no, it's a microcontroller with a special peripheral which has its (apparently smart enough) state machine executing special kind of instructions. 2019-10-23T21:02:40 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-23T21:02:56 < kakipro> ASI 2019-10-23T21:02:58 < kakipro> C 2019-10-23T21:03:34 < PaulFertser> kakipro: yes, of course, fx2lp is how the non-EOL version called. 2019-10-23T21:05:02 < PaulFertser> brdb: are you asking about a module that integrates GSM, GNSS _and_ a Cortex-M microcontroller? Or are you just looking for something that would be easy to integrate with stm32? Why not the usual (among makers) SIM* stuff? 2019-10-23T21:05:47 < kakipro> baseband stm32? 2019-10-23T21:06:18 < PaulFertser> There's not such a thing :) 2019-10-23T21:07:04 < Cracki> TIL: photodiodes have appreciable capacitance, which is why the typical opamp circuit keeps it charged at zero 2019-10-23T21:08:02 < Cracki> kakipro, fx2 is also how cheap logic analyzers "work" 2019-10-23T21:08:31 < kakipro> i havev one 2019-10-23T21:08:50 < bitmask> hmm, if you choose 1 oz copper / 2 layer, does that mean each layer is really 1/2 oz? 2019-10-23T21:08:58 < bitmask> or are both layers 1 oz 2019-10-23T21:09:23 < bitmask> oh its 1 2019-10-23T21:09:32 < bitmask> reading helps 2019-10-23T21:11:36 < Cracki> nah 2019-10-23T21:11:42 < Cracki> outer layers will be 1oz each 2019-10-23T21:11:55 < Cracki> inner layers may sometimes be half ounce for some reason 2019-10-23T21:12:24 < Cracki> ounces of copper is always per layer, given for each layer 2019-10-23T21:12:56 < jadew> lol BrainDamage, took me a bit to get it :) 2019-10-23T21:13:04 < Cracki> btw: snow leopards have been called ounces too. etymology is interesting. 2019-10-23T21:13:29 < Cracki> haha 2019-10-23T21:14:53 < jadew> we had that song blasting in our car for the most part of last year 2019-10-23T21:16:21 < brdb> PaulFertser: additional components alongside the stm32, and not sure how available the SIM* devices are on a more production oriented design 2019-10-23T21:24:09 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-23T21:25:27 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-23T21:36:21 < srk> brdb: theres ublox m8n 2019-10-23T21:37:08 < srk> even m9n, mm 2019-10-23T21:43:54 -!- kow_ [~iccy@135.0.26.39] has quit [Read error: Connection reset by peer] 2019-10-23T21:45:55 < brdb> srk: was looking at that, thanks! 2019-10-23T22:02:14 < Cracki> I second the ublox stuff 2019-10-23T22:02:41 < Cracki> for positioning stuff anyway. haven't seen anything gsm from them. 2019-10-23T22:03:06 < Cracki> they do have that, but everyone knows them because of positioning 2019-10-23T22:04:22 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-23T22:04:44 < Cracki> til: swiss company, founders from ETH 2019-10-23T22:16:45 -!- kow_ [~iccy@135.0.26.39] has joined ##stm32 2019-10-23T22:21:19 < benishor> how can I measure the real configured frequency of a stm32f746? 2019-10-23T22:22:20 < antto> config usually means some SFR or fuse or such 2019-10-23T22:22:26 < antto> ya don't measure those things >:/ 2019-10-23T22:28:12 <@englishman> brdb: ublox is your one stop shop 2019-10-23T22:28:20 <@englishman> fantastic cellular modules 2019-10-23T22:28:44 < antto> how do you say it in dongsish? 2019-10-23T22:29:09 < antto> ubollocks? 2019-10-23T22:29:47 < Cracki> benis, can you configure cock tree to output it on a pin? 2019-10-23T22:30:34 < antto> Cracki NSFW that sh*t 2019-10-23T22:30:42 < brdb> cheers englishman 2019-10-23T22:30:58 <@englishman> antto: ucocks 2019-10-23T22:31:05 < antto> ah 2019-10-23T22:31:44 < Cracki> RM0385 talks about MCO 2019-10-23T22:31:50 < Cracki> so yes you can probably output that 2019-10-23T22:32:03 < Cracki> on particular pins tho 2019-10-23T22:32:09 < Cracki> https://www.st.com/content/ccc/resource/technical/document/reference_manual/c5/cf/ef/52/c0/f1/4b/fa/DM00124865.pdf/files/DM00124865.pdf/jcr:content/translations/en.DM00124865.pdf 2019-10-23T22:32:10 < brdb> now is that you-cocks or zed-cocks? 2019-10-23T22:32:35 < Cracki> you wanna conf that pin for high drive strength 2019-10-23T22:32:59 < antto> wait, isn't uCocks sorta the same thing as mikrodick? 2019-10-23T22:33:03 < antto> i see a conflict 2019-10-23T22:38:02 < qyx_> ublox makes totally different things than mikrotik 2019-10-23T22:39:49 < antto> i bet 2019-10-23T22:40:15 < antto> microdick is some rewter shizzle 2019-10-23T22:40:31 < antto> muh collegue loves them (and yet he pirates them) 2019-10-23T22:42:16 <@englishman> mikrodik is amazing 2019-10-23T22:42:33 <@englishman> but what does that have to do with ucocks 2019-10-23T22:43:09 < antto> nuffing 2019-10-23T22:43:15 < antto> u == micro 2019-10-23T22:43:19 < antto> i don't know 2019-10-23T22:43:36 < antto> there is just a conflict after translation to dongsish IMO 2019-10-23T22:43:56 < antto> i mean possibility for confusion 2019-10-23T22:52:12 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-23T23:03:38 -!- renn0xtek9 [~max@ip5f5bee0f.dynamic.kabel-deutschland.de] has quit [Ping timeout: 276 seconds] 2019-10-23T23:04:18 < karlp> benishor: there's app notes on using LSE to measure HSE/HSI via timers too, if you want to do it in situ 2019-10-23T23:13:48 -!- jly [uid355225@gateway/web/irccloud.com/x-ytdjtxgglblmzpig] has joined ##stm32 2019-10-23T23:26:59 -!- machinehum [~machinehu@d207-216-21-173.bchsia.telus.net] has quit [Quit: WeeChat 1.4] 2019-10-23T23:28:12 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-23T23:30:09 < Steffanx> Hello welcome, jly 2019-10-23T23:31:31 < jly> hi Steffanx 2019-10-23T23:31:39 < jly> i didn't know you used IRC 2019-10-23T23:31:57 < Steffanx> What is irc 2019-10-23T23:32:42 < jly> idk 2019-10-23T23:33:04 < jly> did you hear what laurence did 2019-10-23T23:42:38 -!- sterna [~Adium@c-3db9d954.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-23T23:48:52 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-23T23:49:12 < Steffanx> No what? 2019-10-23T23:49:35 -!- kaki32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-23T23:50:14 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-23T23:52:16 < catphish> quick sanity check - why would my timer run at the same rate regardless of the value of TIMx_PSC? 2019-10-23T23:53:09 < catphish> does the prescaler need to be enabled? 2019-10-23T23:55:41 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] --- Day changed Thu Oct 24 2019 2019-10-24T00:00:01 -!- renn0xtek9 [~max@2a02:810d:1540:2448:4b4:1f00:dc21:e5bd] has joined ##stm32 2019-10-24T00:00:50 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-24T00:02:27 < mawk> I don't think so no catphish 2019-10-24T00:02:31 < mawk> shouldn't run at the same rate 2019-10-24T00:02:37 < mawk> what's your ARR tho ? 2019-10-24T00:02:45 < mawk> also, how do you measure actual firing rate ? 2019-10-24T00:03:35 < catphish> oh, i think i need to generate an update event to load the value 2019-10-24T00:04:05 < catphish> i'm just letting it count up and regularly reading CNT 2019-10-24T00:04:26 < catphish> yep, triggering an update event fixed it 2019-10-24T00:04:28 < mawk> hm 2019-10-24T00:04:31 < mawk> ah 2019-10-24T00:04:33 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-24T00:04:36 < mawk> normally PSC is reloaded immediately 2019-10-24T00:04:44 < mawk> even in the middle of counting 2019-10-24T00:04:51 < catphish> yeah i thought so too 2019-10-24T00:05:36 < catphish> "The new prescaler ratio is taken into account at the next update event." 2019-10-24T00:06:01 < catphish> they count "on the fly" as being "without disabling the timer" i think 2019-10-24T00:07:13 < mawk> well on the fly is really in the middle of counting 2019-10-24T00:07:17 < mawk> but at the next update event yes 2019-10-24T00:07:25 < mawk> but I think you have a bit somewhere to autogenerate update events 2019-10-24T00:08:48 < qyx_> catphish: yes 2019-10-24T00:09:52 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 252 seconds] 2019-10-24T00:10:55 < qyx_> see timer preload 2019-10-24T00:10:59 < qyx_> The preload feature is available for: 2019-10-24T00:11:07 < qyx_> The timer prescaler register (TIMx_PSC) (cannot be turned off) 2019-10-24T00:11:24 -!- qyx_ is now known as qyx 2019-10-24T00:16:36 < catphish> all good now, thanks, just needed to trigger a reload, even though the timer had never been started 2019-10-24T00:21:59 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has quit [Read error: Connection reset by peer] 2019-10-24T00:21:59 -!- sandeepkr [~sandeepkr@2a03:b0c0:2:d0::cac:7001] has quit [Read error: Connection reset by peer] 2019-10-24T00:23:12 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-24T00:32:23 -!- renn0xtek9 [~max@2a02:810d:1540:2448:4b4:1f00:dc21:e5bd] has quit [Quit: Konversation terminated!] 2019-10-24T00:38:42 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T00:55:57 < bitmask> https://i.imgur.com/2FKIYZl.png 2019-10-24T00:55:59 < bitmask> starting to get somewhere 2019-10-24T00:57:24 < kakimir32> looking good 2019-10-24T00:57:32 < bitmask> thanks 2019-10-24T01:02:25 < kakimir32> what is it? 2019-10-24T01:04:04 < bitmask> heated jacket 2019-10-24T01:04:33 < kakimir32> oh wow 2019-10-24T01:04:36 < kakimir32> you went to town 2019-10-24T01:04:39 < bitmask> :) 2019-10-24T01:05:11 < bitmask> https://i.imgur.com/pirrW3n.png 2019-10-24T01:05:18 < bitmask> first time using a bare mcu, wanted to have some fun 2019-10-24T01:05:33 < bitmask> damn connectors take up too much space 2019-10-24T01:06:45 < kakimir32> xt30-series? 2019-10-24T01:07:40 < bitmask> xt30 for power, then found the mr30 which is the 3 pin version 2019-10-24T01:07:47 < bitmask> for the heating elements 2019-10-24T01:07:54 < kakimir32> 3-inline 2019-10-24T01:07:57 < bitmask> yea 2019-10-24T01:08:13 < kakimir32> there is 3 in triangle too iirc 2019-10-24T01:08:53 < bitmask> yea, not sure if they are pcb mount though 2019-10-24T01:09:59 < kakimir32> you have flat flex there? 2019-10-24T01:11:22 < bitmask> yea, thats going to a 'remote' control, oled + 4 buttons 2019-10-24T01:12:00 < bitmask> got some 8 pin 40cm cables in the mail a couple days ago 2019-10-24T01:12:22 < kakimir32> you find them easily in dimensions required? 2019-10-24T01:12:28 < kakimir32> slash inexpensive 2019-10-24T01:12:35 < kakimir32> lly 2019-10-24T01:13:15 < bitmask> https://www.aliexpress.com/item/32955888109.html 2019-10-24T01:13:29 < bitmask> $3.60 for 10 2019-10-24T01:14:09 < bitmask> seem decent 2019-10-24T01:28:39 < kakimir32> how about the heater elements? 2019-10-24T01:28:58 < Jan-> an-> So, I've got a project in cube ide that works okay, I have LEDs blinking and serial ports sending and all is well. 2019-10-24T01:29:06 < Jan-> I'm looking around for HAL documentation for the DAC. 2019-10-24T01:29:27 < Jan-> when I google I get lots of results for different IDEs and HALs, is there a central place for the official ST stuff? 2019-10-24T01:29:58 < kakimir32> did you hal come with examples? 2019-10-24T01:30:02 < Cracki> yes, they do doxygen docs and pack them in chm or pdf 2019-10-24T01:30:03 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-24T01:30:25 < Cracki> cubemx downloads the whole package for F4 or whatever and that's where it is 2019-10-24T01:30:34 < kakimir32> and examples Cracki? 2019-10-24T01:30:43 < Cracki> yes the package that contains the examples too 2019-10-24T01:30:48 < Cracki> those x*100 MB things 2019-10-24T01:30:48 < kakimir32> like it should 2019-10-24T01:31:07 < antto> eggzamples 2019-10-24T01:31:26 < kakimir32> eeggg! 2019-10-24T01:31:43 < antto> kewb em eggz 2019-10-24T01:32:04 < Jan-> egg samples! 2019-10-24T01:32:32 < antto> egg.c 2019-10-24T01:32:38 < antto> #include 2019-10-24T01:32:41 < Jan-> but I have no idea where there are any examples. 2019-10-24T01:32:48 < kakimir32> in the package 2019-10-24T01:32:51 < Jan-> I was sort of hoping for an intellisense thing in the IDE so I could type HAL_ and it would show me all the stuff 2019-10-24T01:32:53 < Jan-> but it doesn't do that 2019-10-24T01:33:05 < kakimir32> what it's based on? 2019-10-24T01:33:08 < kakimir32> eclipse? 2019-10-24T01:33:17 < Jan-> stm32cubeide? 2019-10-24T01:33:18 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-24T01:33:24 < kakimir32> ctrl + space 2019-10-24T01:33:31 < Cracki> cubeide has that... 2019-10-24T01:33:47 < kakimir32> ctrl + space is what you are looking for Jan- 2019-10-24T01:34:17 < bitmask> kakimir32 heating elements are carbon fiber yarn sandwiched between two pieces of silicone 2019-10-24T01:34:29 < Jan-> ok that's a few things :) 2019-10-24T01:35:33 < kakimir32> cudeide is just masked eclipse CDT 2019-10-24T01:35:47 < kakimir32> you can google "how to do X in eclipse" 2019-10-24T01:35:48 < Jan-> I was hoping it'd be based on visual studio 2019-10-24T01:35:53 < Jan-> because visual studio is great 2019-10-24T01:36:08 * antto spits his coffee 2019-10-24T01:36:14 < kakimir32> you know visual studio is propiertary tool 2019-10-24T01:36:26 < kakimir32> I have seen only atmel to work with ms 2019-10-24T01:36:31 < Cracki> you use it with a screen reader, eh? even that should give you a sense of whether cubeide is eclipse-based or VS-based 2019-10-24T01:36:39 < antto> Jan- then u'll "love" atmel studio 2019-10-24T01:36:46 < antto> i guess 2019-10-24T01:36:49 < Cracki> vs can be used by third parties 2019-10-24T01:36:53 < Jan-> I knew it wasn't visual studio 2019-10-24T01:36:54 < Cracki> atmel studio uses VS shell 2019-10-24T01:37:19 < Cracki> and in principle anyone else can use the vs shell 2019-10-24T01:37:24 < antto> can't use raster fonts in it 2019-10-24T01:37:28 < Cracki> ¯\_(ツ)_/¯ 2019-10-24T01:37:33 < Cracki> use vector 2019-10-24T01:37:37 < antto> no 2019-10-24T01:37:42 < antto> i want muh raster f0ntz 2019-10-24T01:37:46 < Cracki> turn raster into vector 2019-10-24T01:37:56 < antto> that's so.. wrong 2019-10-24T01:37:59 < Cracki> go back into your command line, you! 2019-10-24T01:38:43 < Cracki> I have nothing but fond memories of the good old times where pixels were big and fat and distinguishable 2019-10-24T01:38:51 < Cracki> and I want it to stay that way 2019-10-24T01:38:54 < antto> i do have a FixedSys clone on debian, which is vector 2019-10-24T01:39:28 < antto> i can see muh pixels fine 2019-10-24T01:40:28 < kakimir32> Jan-: when you have time to play with the tools there is visual studio for embedded 2019-10-24T01:40:49 < kakimir32> but please don't 2019-10-24T01:41:05 < kakimir32> play with the cube first a while 2019-10-24T01:41:06 < Cracki> I have 120 dpi at a barely OSHA-compliant 50 cm (20") distance. I remember screens where I could see the space between pixels 2019-10-24T01:41:47 < antto> Cracki i reject going below this certain dpi 2019-10-24T01:42:28 < antto> IMO if ya can't distinguish a pile of pixels then ur doing it wrong ;P~ 2019-10-24T01:42:40 < Cracki> windows 10 loupe needs a nearest neighbor interpolation mode. I can't recreate the nostalgia with this linear/cubic sampling or the speshul cel shading filter they offer 2019-10-24T01:43:23 -!- jly [uid355225@gateway/web/irccloud.com/x-ytdjtxgglblmzpig] has quit [Quit: Connection closed for inactivity] 2019-10-24T01:43:40 < antto> crapdows10 2019-10-24T01:44:17 < Cracki> I love 640x480 or less on 15" crt 2019-10-24T01:44:56 < antto> the monitor i recently bought is 1920x1200 24inch 2019-10-24T01:45:29 < antto> and i also still got this 1280x1024 17inch 2019-10-24T01:45:30 < Cracki> the extra you pay for those 120 rows you can put into 2560x1440 2019-10-24T01:45:49 <@englishman> I <3 80 char limit 2019-10-24T01:45:50 < antto> eh? 2019-10-24T01:45:57 < Cracki> 1200-1080 2019-10-24T01:46:14 < antto> then it'd have to be even moar than 24inch 2019-10-24T01:46:19 < Cracki> sure 2019-10-24T01:46:21 < antto> and 24 is already kinda huge 2019-10-24T01:46:34 < antto> i got a big desk but i got two comput0rz on it and a pile of sh*t 2019-10-24T01:46:45 < Cracki> next screen I'll buy is gonna be strictly larger than 24" 2019-10-24T01:46:58 < antto> why tho 2019-10-24T01:47:23 < antto> i bought this thing 2nd hand 2019-10-24T01:47:33 < antto> 200 euroz iirc 2019-10-24T01:47:39 < antto> iz iz IPS 2019-10-24T01:47:39 < Cracki> good q. probably because it's more area. I use this thing for watching movies too 2019-10-24T01:48:21 < kakimir32> any monitor aficionados on the channel? 2019-10-24T01:48:22 < Cracki> and right now, when I watch stuff that's very much moving around and engaging spatial perception, my eyes get fucked with because they wanna focus at the proper distance but need to focus at half a meter because there's the image 2019-10-24T01:48:26 < antto> so far i've never bought a brand new LCD 2019-10-24T01:48:32 < antto> they don't deserve it 2019-10-24T01:48:54 < antto> if imma buy a brand new monitor it gonna be something gud, like a CRT or OLED 2019-10-24T01:49:06 < Cracki> eh, all I know is that IPS is still good and anything else is still not as good 2019-10-24T01:49:18 < kakimir32> I have antto. once 2019-10-24T01:49:23 < kakimir32> it was 1kiloeur 2019-10-24T01:49:29 < kakimir32> I still use it 2019-10-24T01:49:31 < Cracki> crt isn't good in any comparison these days 2019-10-24T01:49:33 < kakimir32> eizo 2019-10-24T01:50:10 < antto> a thousand euros for a monitor? 2019-10-24T01:50:13 < antto> u mad? 2019-10-24T01:50:28 < antto> kakimad32 2019-10-24T01:51:25 < kakimir32> young money 2019-10-24T01:51:55 < kakimir32> I have certain rules though 2019-10-24T01:51:59 < antto> well i spent nearly a thousand on a photo camera 2019-10-24T01:52:17 < kakimir32> never buy under 500eur monitor, never buy under 500eur computer 2019-10-24T01:52:25 < kakimir32> (as new) 2019-10-24T01:53:14 < antto> imma make a comput0r webshop 2019-10-24T01:53:19 < kakimir32> same for a phone pretty much but I buy them almost 50% off when they sell older model stocks 2019-10-24T01:53:26 < antto> with speshul promos if ur name is kakimir 2019-10-24T01:54:32 < Jan-> all the google results I get from "stm32 dac cubeide" are for things like openstm32 2019-10-24T01:54:55 < antto> hot sale: muh old duron 650MHz with the cutting edge 3DNow!(TM) technology with two, i repeat TWO 128MB RAMz 2019-10-24T01:55:34 < antto> for the hot price of just 599.90euro 2019-10-24T01:55:56 < antto> GRAB IT NOW BEFORE U REGRET 2019-10-24T01:56:26 < kakimir32> Jan-: you are not looking for cube ide examples 2019-10-24T01:56:54 < Jan-> aren't I? 2019-10-24T01:57:02 < kakimir32> cubeide is ide 2019-10-24T01:57:07 < Jan-> yes? 2019-10-24T01:57:17 < kakimir32> the hal library is not "cubeide" 2019-10-24T01:57:56 < kakimir32> what is the name of stm32 hal? 2019-10-24T01:58:11 < Cracki> oh, someone is googling how to print with cout in netbeans 2019-10-24T01:58:27 < kakimir32> (I have never used stm32) 2019-10-24T01:58:36 < Cracki> stm32 hal is good enough term 2019-10-24T01:58:40 < Cracki> they just call it hal 2019-10-24T01:58:48 < Jan-> yes that's why it's hard to google 2019-10-24T01:58:53 < Jan-> is there not some msdn-style thing for it somewhere 2019-10-24T01:59:01 < Cracki> told ya, the shit isn't online 2019-10-24T01:59:13 < Cracki> it's only in the shit package that contains the hal files 2019-10-24T01:59:23 < Jan-> not the code, the documentation? 2019-10-24T01:59:26 < Cracki> yes 2019-10-24T01:59:47 < kakimir32> Cracki: well their naming sucks 2019-10-24T01:59:56 < antto> Jan- can't read properly, needs moar coffe 2019-10-24T02:00:01 < antto> or moar sleep 2019-10-24T02:00:08 < kakimir32> when I use lpc the package I download is LPCopen 2019-10-24T02:00:08 < antto> or both 2019-10-24T02:00:17 < Cracki> if you google "stm32f4 (or other) hal" you'll find some pdfs online by st, which are that 2019-10-24T02:00:26 < kakimir32> it includes liblpc hall and cmsis and example codes 2019-10-24T02:00:32 < kakimir32> *hal 2019-10-24T02:00:37 < Cracki> so it's "online" but really just a download 2019-10-24T02:00:41 < Cracki> (pdf) 2019-10-24T02:00:57 < Cracki> get the page for your chip on ST 2019-10-24T02:00:57 < Jan-> I found some PDFs but maybe I'm overlooking the API reference 2019-10-24T02:01:03 < Cracki> materials: documentation: user manual 2019-10-24T02:01:15 < Cracki> UM1725 is hal docs for f4 2019-10-24T02:01:21 < kakimir32> Cracki: is stm hal openstm32? 2019-10-24T02:01:23 < Cracki> wat chip do you have 2019-10-24T02:01:25 < Cracki> wat 2019-10-24T02:01:31 < Cracki> i don't know or gaf what openstm32 is 2019-10-24T02:01:35 < Cracki> afaik it's a random website 2019-10-24T02:01:42 < Jan-> oh maybe this 2019-10-24T02:01:42 < kakimir32> a 2019-10-24T02:01:42 < Jan-> https://www.st.com/content/ccc/resource/technical/document/user_manual/2f/71/ba/b8/75/54/47/cf/DM00105879.pdf/files/DM00105879.pdf/jcr:content/translations/en.DM00105879.pdf 2019-10-24T02:01:48 < Cracki> from a random company who slap "open" on their shit to market it better 2019-10-24T02:02:58 < Jan-> I'm not sure what chip that's for though 2019-10-24T02:03:31 < kakimir32> Jan-: it says it in header 2019-10-24T02:03:32 < Cracki> says in the title: STM32F4 2019-10-24T02:03:34 < kakimir32> F4 series 2019-10-24T02:03:43 < Cracki> did you miss that? 2019-10-24T02:03:53 < Cracki> I know it's a pdf and they don't have logical structure to them but... 2019-10-24T02:04:01 < kakimir32> he expected specific code 2019-10-24T02:04:19 < Cracki> on the internet, everyone knows you're a dog 2019-10-24T02:04:32 < kakimir32> :o 2019-10-24T02:04:33 < Cracki> or cat 2019-10-24T02:04:35 < Jan-> ooh so two DACs. 2019-10-24T02:04:41 < kakimir32> or kaki 2019-10-24T02:04:46 < Cracki> you learned that today, two dacs? 2019-10-24T02:04:51 < Cracki> what chip do you even have 2019-10-24T02:04:53 < Cracki> is that known? 2019-10-24T02:05:00 < Jan-> stm32f407vet6. 2019-10-24T02:05:08 < Jan-> I don't know how much of that matters. 2019-10-24T02:05:35 < Cracki> data sheet will state those things. reference manual will go into detail about how those peripherals work and how to configure them 2019-10-24T02:05:49 < Cracki> it matters because there are many F4 and they do differ in specifics 2019-10-24T02:06:03 < kakimir32> hal source code is the best reference for programming reference 2019-10-24T02:06:24 < Cracki> hal docs might tell you what to do for each variation, but they won't reiterate which chip has which variation 2019-10-24T02:06:34 < kakimir32> and examples using that hal 2019-10-24T02:06:36 < Cracki> that's what the RM is for 2019-10-24T02:07:05 < kakimir32> have you imported examples to your workspace Jan-? 2019-10-24T02:07:31 < kakimir32> if you find the package cubemx downloaded I assume cubeides import project button will accept that 2019-10-24T02:07:40 < kakimir32> and you can select what to import from the package 2019-10-24T02:08:04 < Jan-> I'm not going to futz with that 2019-10-24T02:08:22 < Jan-> I found the part of the PDF that lists the DAC functions 2019-10-24T02:08:24 < Jan-> there's a lot 2019-10-24T02:10:12 < Jan-> ok well it's too late now 2019-10-24T02:10:21 < Jan-> but I'll get back to this tomorrow 2019-10-24T02:10:26 < Jan-> I have the day off :) 2019-10-24T02:20:24 -!- kleinjl [~kleinjl@2001:1388:2:7992:598f:d04b:fb5:2709] has quit [Quit: kleinjl] 2019-10-24T02:21:49 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-24T02:41:16 < karlp> why is bitmask using an atmega644?! 2019-10-24T03:00:35 < mawk> lol 2019-10-24T03:01:58 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 245 seconds] 2019-10-24T03:02:29 < karlp> I love the shit aliexpress comes up with: https://www.aliexpress.com/item/33004674029.html 2019-10-24T03:08:12 < kakimir32> did you "invest"? 2019-10-24T03:09:47 < karlp> nah, shipping is too crap these days for toys on ali unless I can combine shipping 2019-10-24T03:10:06 < mawk> dongs is sad, he can't do https://fr.aliexpress.com/fuckjewnigger/33004674029.html anymore 2019-10-24T03:10:36 < karlp> adds about ~15-20 usd to every parcel on arrival thhese days. 2019-10-24T03:10:55 < kakimir32> I get 404 or so in french 2019-10-24T03:11:26 < karlp> that's what mawk was saying 2019-10-24T03:11:29 < mawk> yes kakimir32 I replaced "item" by "fuckjewnigger" 2019-10-24T03:11:44 < kakimir32> how nice of you 2019-10-24T03:11:59 < mawk> aliexpress finally fixed the super long links by putting "item" instead of the slug 2019-10-24T03:12:25 < kakimir32> does anyone know who dongs is? 2019-10-24T03:12:42 < kakimir32> and his etnicity 2019-10-24T03:14:10 < mawk> no 2019-10-24T03:14:28 -!- dobson` [~dobson@static.38.6.217.95.clients.your-server.de] has quit [Quit: Leaving] 2019-10-24T03:14:38 * kakimir32 sweats 2019-10-24T03:15:24 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Ping timeout: 252 seconds] 2019-10-24T03:15:33 < kakimir32> how should I calculate firing angle? 2019-10-24T03:15:38 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Ping timeout: 240 seconds] 2019-10-24T03:15:43 < kakimir32> if I know inductance 2019-10-24T03:16:19 < kakimir32> and I want linearize output 2019-10-24T03:16:29 < kakimir32> I don't know what I want to linearize 2019-10-24T03:16:33 < kakimir32> but I want to 2019-10-24T03:17:27 < kakimir32> should I concider integral of output voltage as power output? 2019-10-24T03:17:59 -!- dobson [~dobson@static.38.6.217.95.clients.your-server.de] has joined ##stm32 2019-10-24T03:18:06 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-24T03:20:43 < kakimir32> rather I should concider integral of output voltage^^2 as power output? 2019-10-24T03:23:32 < mawk> firing angle for doing what ? 2019-10-24T03:23:53 < kakimir32> driving thyristor/triac 2019-10-24T03:26:14 < kakimir32> oh wow 2019-10-24T03:27:31 < kakimir32> I suppose firing angle is linear to output power and what I need to do is convert angle to timing 2019-10-24T03:27:34 < Cracki> mawk, use instead https://fr.aliexpress.com/item/fuckjewnigger/33004674029.html 2019-10-24T03:27:48 < kakimir32> oh almost went for it second time 2019-10-24T03:28:59 < Cracki> how does voltage relate to any kind of firing angle, you building a rail gun or what 2019-10-24T03:29:21 < Cracki> thyristor/triac... smells like electrical angle 2019-10-24T03:29:49 < mawk> ah ok kakimir32 2019-10-24T03:29:59 < mawk> well 2019-10-24T03:30:03 < mawk> what is your output device ? 2019-10-24T03:30:07 < mawk> let's say pure resistive load ? 2019-10-24T03:30:26 < kakimir32> not but we may simplify it as resistive load 2019-10-24T03:30:31 < kakimir32> for now 2019-10-24T03:33:12 < mawk> so the angle is between 0 and 180° 2019-10-24T03:33:19 < kakimir32> no 2019-10-24T03:33:19 < mawk> 0° is firing immediately, 180° is never firing 2019-10-24T03:33:22 < mawk> what ? 2019-10-24T03:33:27 < mawk> 0 and pi if you want 2019-10-24T03:33:29 < kakimir32> we look at single halve of sine 2019-10-24T03:33:38 < kakimir32> of wait 2019-10-24T03:33:42 < kakimir32> 180 yes 2019-10-24T03:36:15 < mawk> let me compute then 2019-10-24T03:38:09 < kakimir32> ( x > 90)?(1+sin(x)):(sin(x)) ? 2019-10-24T03:38:32 < mawk> pretty far from linear 2019-10-24T03:38:39 < mawk> well 2019-10-24T03:38:49 < kakimir32> oh I forgot 2019-10-24T03:38:54 < kakimir32> ^^2 2019-10-24T03:39:37 < kakimir32> basic trigonometry is a challenge for me 2019-10-24T03:40:35 < kakimir32> ( x > 90)?(1+(sin(x)^^2)):(sin(x)^^2) ? 2019-10-24T03:40:41 < mawk> the power is proportional to x/(2π) + sin(4πfx)/(4π²f) 2019-10-24T03:40:47 < mawk> what's your ^^ ? 2019-10-24T03:40:54 < mawk> let me show you how I did it, it's quick 2019-10-24T03:41:06 < kakimir32> I'm calculating timing output 2019-10-24T03:41:19 < kakimir32> for linear power per x 2019-10-24T03:42:26 < kakimir32> x is power setting and return value 0 = don't fire, max = fire immidiatelly after zeroX 2019-10-24T03:44:37 < mawk> it's not linear 2019-10-24T03:44:39 < mawk> let me show 2019-10-24T03:48:04 < kakimir32> I see 2019-10-24T03:49:03 < kakimir32> mine is not correct 2019-10-24T03:50:35 < kakimir32> or is it 2019-10-24T03:53:48 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-24T03:56:39 < kakimir32> it isn't 2019-10-24T03:57:19 < mawk> kakimir32: https://maths.serveur.io/2NumQX4o 2019-10-24T03:57:22 < mawk> liek dis 2019-10-24T03:58:08 < kakimir32> oh my 2019-10-24T03:58:18 < mawk> just look at the last line 2019-10-24T03:58:20 < mawk> it's pretty simpl 2019-10-24T03:58:21 < mawk> e 2019-10-24T03:58:34 < mawk> ω is just 2πf 2019-10-24T03:58:37 < mawk> f signal freq 2019-10-24T03:58:40 < kakimir32> it's power? 2019-10-24T03:58:46 < mawk> yes 2019-10-24T03:58:50 < mawk> it's the multiplicative factor for the power 2019-10-24T03:58:53 < mawk> between the { } 2019-10-24T03:59:04 < mawk> and Veff² is the power with a pure sine wave 2019-10-24T03:59:08 < kakimir32> you have f there somewhere? 2019-10-24T03:59:17 < mawk> I forgot the /R everywhere do but don't worry 2019-10-24T03:59:20 < mawk> let's just say R = 1 2019-10-24T03:59:23 < mawk> yes 2019-10-24T03:59:27 < mawk> [02:58:34] .:mawk:. ω is just 2πf 2019-10-24T03:59:41 < mawk> hmmmmmm 2019-10-24T03:59:48 < mawk> I think I made a mistake maybe ? 2019-10-24T03:59:51 < mawk> or no 2019-10-24T04:00:17 < Cracki> sounds plausible at least 2019-10-24T04:00:19 < mawk> I divided by pi everywhere instead of the sine cosine period 2019-10-24T04:00:26 < mawk> small mistake, doesn't change the big shape 2019-10-24T04:00:26 < Cracki> omega in radians, would be 2pi f 2019-10-24T04:00:30 -!- kaki32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T04:00:45 < kaki32> idk man 2019-10-24T04:00:50 < mawk> the cos(ωt) has frequency f, period T = 1/f 2019-10-24T04:01:17 < Cracki> omega t, odd expression 2019-10-24T04:01:30 < mawk> yeah I think we can just kick out all ω everywhere 2019-10-24T04:01:44 < mawk> since it obviously doesn't change the power factor, does it 2019-10-24T04:01:46 < kaki32> I don't need absolute power 2019-10-24T04:01:48 < mawk> to have a higher freq or not 2019-10-24T04:01:52 < kaki32> I don't need voltages 2019-10-24T04:01:53 < kaki32> currents 2019-10-24T04:01:55 < mawk> ? 2019-10-24T04:01:56 < kaki32> frequencies 2019-10-24T04:02:07 < mawk> how do you define power without any of that 2019-10-24T04:02:13 < mawk> but don't worry I'm simplifying it 2019-10-24T04:02:14 < mawk> hold on 2019-10-24T04:02:19 < kaki32> I need conversion from angle to timing value 2019-10-24T04:03:55 -!- kakimir32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-24T04:04:18 < mawk> kaki32: https://maths.serveur.io/WKuV6eBe 2019-10-24T04:04:21 < mawk> what is timing value 2019-10-24T04:04:31 < mawk> you said something about linear power 2019-10-24T04:04:34 < kaki32> firing time 2019-10-24T04:04:38 < kaki32> of SCR 2019-10-24T04:04:44 < mawk> yeah 2019-10-24T04:04:48 < mawk> so you need the equation I did 2019-10-24T04:05:00 < mawk> just make a lookup table with my equation 2019-10-24T04:05:04 < mawk> with a 0.1% step for instance 2019-10-24T04:05:36 < mawk> or rather with a 1° step sorry, like you compute power % for 0°, for 1°, for 2°, ..., for 180° 2019-10-24T04:05:44 < mawk> then when you want 75% power you look up in the table 2019-10-24T04:05:50 < kaki32> is there any lib for C that I could just throw these maths at and let it generate code from it? 2019-10-24T04:06:12 < mawk> why do you need a lib ? you just want to generate a table 2019-10-24T04:06:19 < mawk> what precision you want for the power ? 2019-10-24T04:06:20 < mawk> 1% ? 2019-10-24T04:06:25 < kaki32> yeah 2019-10-24T04:06:38 < mawk> ok, yeah well just a for loop in python will do it I guess, let me try 2019-10-24T04:07:28 < kaki32> please don't 2019-10-24T04:07:42 < mawk> it's just like 5 lines 2019-10-24T04:08:45 < BrainDamage> build the table at boot 2019-10-24T04:08:51 < BrainDamage> takes no time 2019-10-24T04:09:29 < kaki32> I build one variable at boot 2019-10-24T04:09:39 < kaki32> it's size is like 2kilobytes 2019-10-24T04:09:44 < kaki32> of hash and strings 2019-10-24T04:10:02 < kaki32> it's my hash command array 2019-10-24T04:10:35 < kaki32> hash, id, pointer to name, pointer to description 2019-10-24T04:11:18 -!- Laurenceb [~laurence@110.184.147.147.dyn.plus.net] has joined ##stm32 2019-10-24T04:11:19 < mawk> let me find the right step for building the table 2019-10-24T04:11:27 < Laurenceb> muh table 2019-10-24T04:11:52 < kaki32> lurence 2019-10-24T04:15:07 < mawk> yeah my computation is correct I think 2019-10-24T04:15:07 < specing> laurence of kekabia 2019-10-24T04:15:07 < mawk> I got a pretty looking thing 2019-10-24T04:15:07 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has joined ##stm32 2019-10-24T04:15:08 < mawk> when I graph it it seems accurate 2019-10-24T04:15:56 < kaki32> so me the graphics! 2019-10-24T04:16:07 < mawk> 0.1° step seems to yield a 0.1% step for the power, pretty good 2019-10-24T04:17:29 < kaki32> mawk: are you a math gui? 2019-10-24T04:18:05 < mawk> kaki32: https://serveur.io/Thyristor.html 2019-10-24T04:18:08 < mawk> somewhat yes 2019-10-24T04:18:14 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 276 seconds] 2019-10-24T04:18:41 < kaki32> it looks like what I expected 2019-10-24T04:18:46 < mawk> good 2019-10-24T04:18:50 < mawk> then you have the formula now 2019-10-24T04:19:23 < mawk> the formula can't be analytically inverted (I think it requires the Lambert W function BrainDamage ?), that means you can't have angle as a function of power percentage, so that's why you build a table 2019-10-24T04:19:45 < mawk> or you could do it numerically like with a binary search or whatever 2019-10-24T04:19:45 < kaki32> lemme think a second 2019-10-24T04:20:05 < mawk> would take 7 or 8 iterations at the worst casre 2019-10-24T04:20:06 < mawk> -r 2019-10-24T04:20:08 < mawk> for every lookup 2019-10-24T04:20:18 < mawk> that's not so bad, and doesn't waste SRAM/flash 2019-10-24T04:20:18 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-24T04:20:20 < BrainDamage> mawk: just fyi, you shouldn't integrate from φ to π, rather from φ to α, where α = 2πt/T, t for which I=0 and T is the period 2019-10-24T04:20:50 < mawk> I took T = 2pi for simplification of the computation 2019-10-24T04:20:59 < mawk> since a bigger period doesn't change the power ratio I think 2019-10-24T04:21:08 < BrainDamage> but yes, it's a lambertian 2019-10-24T04:21:12 < mawk> and I integrated over a half period, also 2019-10-24T04:21:14 < mawk> to simplify too 2019-10-24T04:21:31 < mawk> since the thyristor firing angle is symmetric 2019-10-24T04:21:47 < dongs> mawk are you sure 2019-10-24T04:21:49 < mawk> but there's still the I=0 business 2019-10-24T04:22:03 < BrainDamage> yes, the wave is symmetric, but the on time has to be corrected for 2019-10-24T04:22:09 < mawk> ah, right 2019-10-24T04:22:11 < mawk> yes 2019-10-24T04:22:37 < dongs> oh wow they fxioed all links to be short autoamtically? 2019-10-24T04:22:41 < dongs> wouldnt that fuck with their SEO 2019-10-24T04:22:45 < BrainDamage> however, you table the whole function and just evaluate the two points 2019-10-24T04:22:50 < BrainDamage> then subtract them 2019-10-24T04:23:00 < BrainDamage> which is pretty trivial 2019-10-24T04:24:54 < kaki32> I don't understand 2019-10-24T04:25:23 < kaki32> I only understand the graph 2019-10-24T04:25:31 < kaki32> it makes sense 2019-10-24T04:28:05 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-24T04:29:15 < dongs> whats a name for a lamp that has like flexible neck 2019-10-24T04:29:20 < dongs> and a bright light source at end 2019-10-24T04:29:49 < dongs> oh hm flexible pipe 2019-10-24T04:30:08 < dongs> https://www.dhgate.com/product/gay/182005127.html this kinda shit i guess 2019-10-24T04:30:22 < dongs> how awful is chinese 3W 2019-10-24T04:30:25 < dongs> i need this to be really bright 2019-10-24T04:31:10 < mawk> graph is x = angle, y = ratio kaki32 2019-10-24T04:31:39 < mawk> the on-time you mean kaki32 should be synchronized with the output signal BrainDamage right ? 2019-10-24T04:31:46 < mawk> to know precisely when to trigger the SCR 2019-10-24T04:31:58 < mawk> then it will match the function I wrote 2019-10-24T04:32:09 < kaki32> that is the point 2019-10-24T04:33:09 < kaki32> thanks mawk 2019-10-24T04:33:57 < kaki32> did you enjoy it? 2019-10-24T04:34:10 < mawk> yeah 2019-10-24T04:34:12 < kaki32> maybe not right word 2019-10-24T04:34:16 < mawk> who doesn't enjoy writing maths ? 2019-10-24T04:34:25 < kaki32> but did you feel engaged 2019-10-24T04:34:33 < Ultrasauce> dongs: the style of neck is called a gooseneck 2019-10-24T04:34:39 < BrainDamage> dongs: use approximatively 100lm/w 2019-10-24T04:34:48 < BrainDamage> so that lamp would be approx 300lm 2019-10-24T04:34:49 < mawk> I just regret I don't type latex as fast as I write on paper 2019-10-24T04:34:51 < mawk> yes kaki32 2019-10-24T04:35:05 < kaki32> great 2019-10-24T04:36:05 < BrainDamage> dongs: to get a sense of scale, a headlight lamp in low intensity is about 700lm 2019-10-24T04:36:16 < dongs> headlight as in car? 2019-10-24T04:36:19 < BrainDamage> yes 2019-10-24T04:36:33 < dongs> gooseneck, check. 2019-10-24T04:36:36 < dongs> lemme see wat aliexpress has for that 2019-10-24T04:36:49 < dongs> ah yes. much more results now 2019-10-24T04:38:11 < dongs> https://www.aliexpress.com/item/32908380364.html this looks about waht im looking for mechanical-wise but lol 1W led 2019-10-24T04:38:15 < BrainDamage> mawk: yes 2019-10-24T04:39:20 < Ultrasauce> just get like 10 of them 2019-10-24T04:39:26 < Ultrasauce> make your desk look like a tentacle monster 2019-10-24T04:40:05 < Ultrasauce> maybe eliminate usb from the search results 2019-10-24T04:41:16 < BrainDamage> mmm, that actually might be a not too horrible suggestion, the only problem with uniform lighting is that without shadows your brain will feel plenty irked 2019-10-24T04:41:30 < BrainDamage> because it fucks up depth perception 2019-10-24T04:41:35 < dongs> i dont think alisexpress understands -USB for search 2019-10-24T04:41:56 < dongs> https://www.aliexpress.com/item/32956889443.html? mmm 2019-10-24T04:42:11 < dongs> > usb clip light 2019-10-24T04:42:12 < dongs> 8W 2019-10-24T04:42:16 < dongs> yea no 2019-10-24T04:42:25 < dongs> Dimmer Clip on Light: The Flexible clamp light has 3 colors mode & 10 levels light, Max 600 LM hmm 2019-10-24T04:42:41 < dongs> is that 600 chinese lumen 2019-10-24T04:42:56 < dongs> wait fucking what 2019-10-24T04:42:58 < dongs> its battery? 2019-10-24T04:43:00 < Ultrasauce> does aliexpress really have no advanced search 2019-10-24T04:44:01 < kaki32> ali has one of the worst search algos I have seen 2019-10-24T04:45:14 < kaki32> when you mistype something it just doesn't say yeah no we don't have it 2019-10-24T04:45:31 < kaki32> it throws all possible things at you 2019-10-24T04:45:47 < dongs> sounds better htan this jap auction site 2019-10-24T04:45:50 < dongs> when you search for something 2019-10-24T04:45:57 < dongs> it shows all the sold listsings 2019-10-24T04:46:04 < kaki32> perfect 2019-10-24T04:46:05 < dongs> https://www.mercari.com/jp/search/?keyword=LCT5 like this 2019-10-24T04:46:44 < dongs> the one that isn't sold is unsurprisingly, completely unrelated 2019-10-24T04:46:58 < kaki32> and another is that search words seem not to seek match from title of the item 2019-10-24T04:47:13 < kaki32> once I wanted to order a thing again 2019-10-24T04:47:31 < kaki32> i tried to write de title 2019-10-24T04:47:37 < kaki32> it gives crap 2019-10-24T04:47:48 < Ultrasauce> https://www.banggood.com/weed-p-1473897.html 2019-10-24T04:47:49 < kaki32> you need the magical keyword combination! 2019-10-24T04:48:00 < dongs> weed, NICE 2019-10-24T04:48:11 < dongs> Luminous flux1991lm 2019-10-24T04:48:43 < kaki32> perfect efficiency 2019-10-24T04:48:50 < kaki32> *efficancy 2019-10-24T04:49:14 < dongs> why are most of those grow lamps like purple or somethign 2019-10-24T04:49:52 < Ultrasauce> why are plants green 2019-10-24T04:50:09 < dongs> idk 2019-10-24T04:50:22 < Ultrasauce> cause they absorb light that isnt green 2019-10-24T04:51:15 < kaki32> once earth had sun with different colour? 2019-10-24T04:51:37 -!- Laurenceb [~laurence@110.184.147.147.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-24T04:53:50 < dongs> Ultrasauce: wait WHAT 2019-10-24T04:53:54 < dongs> scroll down that bangnigger page 2019-10-24T04:54:01 < BrainDamage> there's two types of clorophyl: a and b 2019-10-24T04:54:05 < dongs> last pic is eurofaggot AC adapter with USB plug 2019-10-24T04:54:18 < Ultrasauce> i am assuming that's just a mistakenly included image 2019-10-24T04:54:20 < BrainDamage> https://en.wikipedia.org/wiki/File:Chlorophyll_ab_spectra-en.svg 2019-10-24T04:54:29 < Ultrasauce> cause the other ones show an edison plug or whatever its called 2019-10-24T04:54:33 < BrainDamage> the overlap gives a hole in the mid where green is 2019-10-24T04:54:43 < BrainDamage> rest is absorbed and converted to nutrients 2019-10-24T04:58:18 < kaki32> maybe the hole is from early evolution of plant life 2019-10-24T04:58:22 < kaki32> in seas 2019-10-24T04:58:44 < kaki32> sea water reflects and absorbs maybe something 2019-10-24T05:00:20 < BrainDamage> water tens to absorb and scatter higher frequencies better than lower 2019-10-24T05:00:26 < BrainDamage> hence why you see the sea as blue 2019-10-24T05:00:32 < kaki32> yes 2019-10-24T05:00:56 < kaki32> so it's not the explanatiom 2019-10-24T05:00:57 < BrainDamage> it's probably just that those wavelengths were convenient for the materials at hand 2019-10-24T05:01:08 < kaki32> yes 2019-10-24T05:06:41 <@englishman> kaki32, zip has met dongs irl. as for ethnicity, dongs is a russian black bear that has learned to chat. which explains his anger and fascination with dicks 2019-10-24T05:07:31 < kaki32> zip? 2019-10-24T05:09:49 < doomba> i can only imagine meeting dongs irl 2019-10-24T05:10:19 < doomba> swings his swivel chair away from altium for a few seconds to say "FUCK YOU" and then goes back to doing whatever he's doing. 2019-10-24T05:10:51 < BrainDamage> I like to imagine him as linus torvalds, because it makes things more hilarious 2019-10-24T05:11:02 < BrainDamage> same angry attitude too 2019-10-24T05:12:41 < doomba> yes. a black russian linus torvalds with a desk covered in pcbs and high capacity magazines 2019-10-24T05:25:22 < Cracki> for some reason I had to think of FPSRussia 2019-10-24T05:25:32 < Cracki> got busted for 25 grams of hash oil 2019-10-24T05:30:43 -!- kaki32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-24T06:09:40 < Tordek> is a prescaler just to divide a counter before triggering it? or is there a benefit for using, say, a prescaler of 50k and autoreload of 5 vs the other way around 2019-10-24T06:09:56 < Tordek> if I only care about the interruption when the counter reaches 0 2019-10-24T06:19:31 < Cracki> you grasp the situation 2019-10-24T06:20:01 < Cracki> benefit is smaller prescaler is that basic timer ticks are finer 2019-10-24T06:20:15 < Cracki> that's what you might want for finer pwm control 2019-10-24T06:20:20 -!- fc5dc9d4 [~quassel@p5B08108E.dip0.t-ipconnect.de] has joined ##stm32 2019-10-24T06:24:02 < Tordek> kk thanks 2019-10-24T06:24:21 -!- fc5dc9d4_ [~quassel@p5B08107B.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-24T06:25:26 < Cracki> oh also to get it ticking at some nice round number, if that's important to you 2019-10-24T07:26:44 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has quit [Ping timeout: 276 seconds] 2019-10-24T07:27:11 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has joined ##stm32 2019-10-24T07:30:45 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-24T07:33:38 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 245 seconds] 2019-10-24T07:33:38 -!- day__ is now known as day 2019-10-24T07:40:01 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-24T07:40:45 < R2COM> dongs im buying Valve Index, get ready to be owed in VR games 2019-10-24T07:51:51 < dongs> thats the one that has no external tracking huh 2019-10-24T07:51:55 < dongs> enjoy your aids 2019-10-24T07:52:45 < R2COM> wat 2019-10-24T07:52:49 < R2COM> it does have external trackling 2019-10-24T07:52:58 < R2COM> its Cosmos which doesnt (uses camera) 2019-10-24T07:53:08 < R2COM> but valve index has proper high end external lighthouse trackers 2019-10-24T07:53:28 < R2COM> cosmos is cheaper, im not getting that one 2019-10-24T07:59:58 < dongs> ah ok 2019-10-24T08:03:14 < dongs> ah it comes with rev2 of lighthouse shits 2019-10-24T08:03:18 < dongs> but still can be tracked by v1 2019-10-24T08:03:53 < dongs> https://www.amazon.com/Valve-Index-VR-Full-Kit-PC/dp/B07VPRVBFF#customerReviews lmao 2019-10-24T08:07:56 < R2COM> its cuz higher price on amazon 2019-10-24T08:08:05 < R2COM> if you buy from steam its $400 less 2019-10-24T08:08:14 < dongs> i know, i mean, thats what all the reviews point out 2019-10-24T08:14:20 < antto> gam0rz 2019-10-24T08:14:27 < antto> vee arr 2019-10-24T08:21:13 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 2019-10-24T08:21:28 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has joined ##stm32 2019-10-24T08:21:57 -!- jly [uid355225@gateway/web/irccloud.com/x-vppwqykqfucjdeue] has joined ##stm32 2019-10-24T08:30:26 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 2019-10-24T08:45:14 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has joined ##stm32 2019-10-24T08:50:47 -!- sterna [~Adium@c-4aebe155.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-24T08:57:35 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has quit [Ping timeout: 268 seconds] 2019-10-24T09:08:37 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-24T09:35:28 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-24T09:35:52 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has joined ##stm32 2019-10-24T09:40:27 -!- boB_K7IQ [~boB_K7IQ@173-160-142-89-Washington.hfc.comcastbusiness.net] has quit [Ping timeout: 240 seconds] 2019-10-24T09:54:22 -!- sterna [~Adium@c-4aebe155.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-24T10:10:23 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has joined ##stm32 2019-10-24T10:16:20 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-24T10:19:54 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-24T10:20:26 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [Ping timeout: 240 seconds] 2019-10-24T10:20:38 < ronox> hi, how do i do the arduino equivalent of "digitalwrite(pin,state)" on stm32f407vet6 F4VE using std peripheral library? i cant find any simple example code how to do it, everything i find uses like timers and interrupts and stuff 2019-10-24T10:20:39 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-24T10:20:49 < ronox> so, just code to turn a pin on or off 2019-10-24T10:21:44 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-24T10:22:06 < ronox> i cant use the HAL library because im using the stm peripheral library, its to do some testng on a program that is written ith th SPL 2019-10-24T10:23:29 < ronox> im just looking to do a little learning so that i can do a little debugging before i finally drop attempting to make colorchord work 2019-10-24T10:23:47 < jpa-> ronox: GPIOA->BSRR = (1 << pin_number); to set, BRR for reset 2019-10-24T10:24:02 < jpa-> no need for any libraries 2019-10-24T10:24:09 < ronox> oh ok 2019-10-24T10:24:22 < ronox> is that all? dont have to like set the pin up or anything? 2019-10-24T10:27:37 < srk> of course you have to 2019-10-24T10:28:07 < ronox> how do i set it up? 2019-10-24T10:28:15 < srk> depends :) 2019-10-24T10:28:16 < srk> https://vivonomicon.com/2018/04/22/bare-metal-stm32-programming-part-3-leds-and-buttons/ 2019-10-24T10:29:46 < ronox> do i need to do the gpio init and mode? or anything else 2019-10-24T10:30:33 < Steffanx> I always expect Haskell when I open your links, srk :D 2019-10-24T10:31:28 < srk> :D 2019-10-24T10:33:18 < ronox> im not really sure what any of that code in that page you linked means, is that C? 2019-10-24T10:34:47 < Steffanx> Looks like it 2019-10-24T10:35:00 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-24T10:35:21 < ronox> i dont really understand what any of it means or how to implement it, ive never seen anything like that before 2019-10-24T10:35:57 < ronox> like this GPIOB->MODER |= (0x1 << (LED_PIN*2)); 2019-10-24T10:36:04 < ronox> i dont know what the 0x1 thing means 2019-10-24T10:36:15 < ronox> or why LED pin is multiplied by 2 2019-10-24T10:38:22 < jpa-> reference manual will tell how the registers work 2019-10-24T10:39:13 < ronox> registers? 2019-10-24T10:39:27 < ronox> this is sounding like a rabbit hole 2019-10-24T10:39:45 < ronox> i just quickly want to turn a pin on to prove the program is running at all 2019-10-24T10:41:05 -!- ronoxx [~oldmanbee@193-116-76-133.tpgi.com.au] has joined ##stm32 2019-10-24T10:41:11 < ronoxx> which manual do i need? 2019-10-24T10:41:23 < ronoxx> this chip has like 5 different kinds of manual 2019-10-24T10:43:23 < ronoxx> ok, i checked the main datasheet, there is no mention of any charts that tell me the registry value for each GPIO 2019-10-24T10:43:45 < srk> 'reference manual' 2019-10-24T10:44:52 -!- ronox [~oldmanbee@193-116-76-133.tpgi.com.au] has quit [Ping timeout: 265 seconds] 2019-10-24T10:45:17 < ronoxx> ok, got it, theres over 7000 mentions of the word registry in that 2019-10-24T10:45:23 < ronoxx> which one am i looking for 2019-10-24T10:46:41 < ronoxx> is it the register map for the GPIO banks? 2019-10-24T10:49:10 < srk> yep 2019-10-24T10:49:55 < ronoxx> and how do i use this information? like i have seriously no clue whats going on 2019-10-24T10:51:06 < ronoxx> i think i can see where "moder" comes from, but i still dont know why the led pin is multiplied by 2 2019-10-24T10:51:18 < Haohmaru> ronoxx have you coded anything on MCUs before this? 2019-10-24T10:51:26 < ronoxx> only in arduino 2019-10-24T10:51:32 < Haohmaru> crapduino doesn't count 2019-10-24T10:51:41 < ronoxx> and i ony have done a basics course in C 2019-10-24T10:52:01 < Haohmaru> okay, with crapduino, you didn't read any datasheets, did u? 2019-10-24T10:52:08 < srk> you just need to set bit(s) in correct registers :) 2019-10-24T10:52:19 < Haohmaru> dijital_rite(); // boohoo i'm much coder 2019-10-24T10:52:23 < ronoxx> occasionally just for finding out the pin numbers, but that was board specific 2019-10-24T10:52:32 < Haohmaru> right 2019-10-24T10:52:39 < Haohmaru> well, welcome to teh jungle then 2019-10-24T10:52:49 < Haohmaru> STM32 is harder than an atmega 2019-10-24T10:52:57 < ronoxx> never touched an atmega 2019-10-24T10:53:07 < Haohmaru> wut was the chip in ur crapduino? 2019-10-24T10:53:12 < ronoxx> ive always used stm32f1 or esp through arduino or platformio 2019-10-24T10:53:39 < Haohmaru> x_x 2019-10-24T10:54:13 < Haohmaru> get used to reading datasheets 2019-10-24T10:54:17 < ronoxx> my problem is just i cant find the beginner information, everything to do with GPIO use is at its lowest level super advanced 2019-10-24T10:54:33 < Haohmaru> true 2019-10-24T10:54:40 < Haohmaru> datasheets don't teach u things 2019-10-24T10:54:43 < ronoxx> its either super involved, or, is written in some stupidly convoluted way 2019-10-24T10:54:53 < ronoxx> like that link from before with GPIOB->MODER |= (0x1 << (LED_PIN*2)); 2019-10-24T10:55:44 < srk> << is just a bit shift 2019-10-24T10:55:45 < ronoxx> sure -> is meant to be the proper syntax for pointers, but then i dont know |=, what 0x1 means, i vaguely know what << means, and i still dont know why LED is multiplied 2019-10-24T10:56:07 < srk> it shift 0x1 to the left by LED_PIN*2 2019-10-24T10:56:18 < Haohmaru> aww 2019-10-24T10:56:20 < Haohmaru> boi 2019-10-24T10:56:39 < ronoxx> and above all else i have no clue how to rewrite this for pin A6 2019-10-24T10:56:42 < Haohmaru> get familiar with assignment operators and binary/bitwise shizzle 2019-10-24T10:56:53 < Haohmaru> and hexadecimal representation 2019-10-24T10:56:55 < srk> |= is an assignemnt operator which uses OR with the current value :) 2019-10-24T10:56:59 < srk> pretty much 2019-10-24T10:57:12 < Haohmaru> no datasheet would teach you this 2019-10-24T10:57:23 < srk> or use abstraction like arduiNO. :D 2019-10-24T10:57:34 < Haohmaru> nooo, toss that crap out 2019-10-24T10:57:38 < ronoxx> i will, but right now i just want to get this quickly over and done with, i just want to input a LED pin code into the adc read function for this damn thing 2019-10-24T10:57:58 < ronoxx> because apparently its too much to ask for to do serial printing or printf 2019-10-24T10:58:17 < Haohmaru> then you can only ask around and copy/paste chunks of alien codez 2019-10-24T10:58:22 < Haohmaru> until it works 2019-10-24T10:58:35 < ronoxx> serial monitoring takes a whole library and printf freezes the program altogether 2019-10-24T10:58:58 < srk> that's why we use debuggers instead of print debugging 2019-10-24T10:59:01 < ronoxx> i just want some sort of output so i can debug my code i wrote here 2019-10-24T10:59:19 < ronoxx> how do you debug stuff? i just want some indicator of what the adc variable is 2019-10-24T10:59:40 < srk> with SWD programmer and GDB for example 2019-10-24T11:00:22 < ronoxx> is GDB a device or does that stand for general debugging 2019-10-24T11:00:27 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-24T11:00:45 < ronoxx> because i have an stlink and i think the stlink software i use for uploading has the capacity for debugging 2019-10-24T11:01:18 < ronoxx> stm32-stlink-ultility" 2019-10-24T11:01:44 < srk> yes, you can setup stlink with openocd and connect to it with gdb 2019-10-24T11:03:32 < ronoxx> firefox is telling me openocd is a scam 2019-10-24T11:04:01 < ronoxx> https://gnutoolchains.com/arm-eabi/openocd/ is this the real site for it? 2019-10-24T11:08:05 < ronoxx> ok i got it, im a bit worried because i tried opening openocd and it just showed a command prompt for a moment then turns off 2019-10-24T11:09:37 < ronoxx> so, what now? 2019-10-24T11:09:51 < ronoxx> do i now have to write some batch file or something for it? 2019-10-24T11:10:51 < ronoxx> hello? 2019-10-24T11:11:18 < srk> yes you need to supply it with parameters or use an IDE that would do that for you 2019-10-24T11:12:18 < srk> I think you can use it with eclipse if command line is not your cup of tea 2019-10-24T11:12:27 < ronoxx> i dont know how to use ecliple 2019-10-24T11:12:32 < ronoxx> eclipse* 2019-10-24T11:13:23 < srk> http://ardupilot.org/dev/docs/debugging-with-gdb-on-stm32.html 2019-10-24T11:16:12 < ronoxx> how do i do that but in windows? 2019-10-24T11:16:35 < ronoxx> i feel like making my stlink be readable in vmware would be a whole extra rabbit hole 2019-10-24T11:16:59 < srk> not sure, with libvirt usb passthru is like two clicks 2019-10-24T11:17:12 < srk> try it, might be easier than installing openocd/gdb on windows 2019-10-24T11:18:31 < ronoxx> i still run into the same problem where i have no idea if or not any of it is working since i dont understand how to make it work 2019-10-24T11:18:57 < ronoxx> i think first ill give another try at writing code to make the on board LED blink 2019-10-24T11:19:36 < ronoxx> then if that fails ill try this master expert class stuff for reading the stm on vmware and while making the information readable by that debugger 2019-10-24T11:19:48 < ronoxx> on the plus side if it works i can just directly program from in ubuntu 2019-10-24T11:20:05 < ronoxx> currently ive been exporting the bin file rather than letting makefile do its thing 2019-10-24T11:20:18 < ronoxx> i tried connecting it once before but it said it couldnt see it 2019-10-24T11:23:49 < srk> you can check with dmesg 2019-10-24T11:25:06 < ronoxx> whats that? 2019-10-24T11:26:35 -!- [1]MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has joined ##stm32 2019-10-24T11:28:37 -!- MrMobius [~default@c-73-134-82-217.hsd1.va.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-24T11:28:37 -!- [1]MrMobius is now known as MrMobius 2019-10-24T11:42:41 -!- drozdziak1 [~drozdziak@83.ip-92-222-87.eu] has joined ##stm32 2019-10-24T11:51:07 -!- jly [uid355225@gateway/web/irccloud.com/x-vppwqykqfucjdeue] has quit [Quit: Connection closed for inactivity] 2019-10-24T11:54:15 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-24T12:26:53 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-24T12:45:09 -!- ekaologik [~quassel@p5DC6B28A.dip0.t-ipconnect.de] has joined ##stm32 2019-10-24T13:24:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds] 2019-10-24T13:25:10 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-24T13:56:52 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-24T14:21:38 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-24T14:24:43 -!- tprrt [~tprrt@217.114.204.178] has quit [Read error: Connection reset by peer] 2019-10-24T14:34:30 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-24T14:46:08 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-24T14:48:39 -!- Laurenceb [~laurence@110.184.147.147.dyn.plus.net] has joined ##stm32 2019-10-24T14:52:17 < doomba> Laurenceb caught on vid https://i.imgur.com/ooPrVVA.mp4 2019-10-24T14:52:25 < mawk> you're alive doomba 2019-10-24T15:14:18 < Steffanx> Is he now? 2019-10-24T15:18:13 -!- Laurenceb [~laurence@110.184.147.147.dyn.plus.net] has quit [Ping timeout: 245 seconds] 2019-10-24T15:31:30 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has joined ##stm32 2019-10-24T15:35:24 < Haohmaru> can't be him 2019-10-24T15:41:24 -!- mode/##stm32 [+b *!*@159.217.93.209.dyn.plus.net] by englishman 2019-10-24T15:41:24 -!- Laurenceb was kicked from ##stm32 by englishman [Laurenceb] 2019-10-24T15:43:45 < Mangy_Dog> :o 2019-10-24T15:44:04 < Haohmaru> o_O 2019-10-24T15:44:06 < Haohmaru> why? 2019-10-24T15:44:12 < Haohmaru> poor loirens 2019-10-24T15:52:27 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T15:53:27 < kakim32> ping 2019-10-24T15:53:35 < kakim32> good 2019-10-24T15:53:57 < doomba> Today we remember Laurenceb, who passed shortly after being shot in a duel with englishman. Laurence Blaxter was born at the Bearstead Jewish Maternity Hospital, Stoke Newington. His family lived in Ilford, Essex. He was educated at University College London (UCL). 2019-10-24T16:02:28 < Jan-> Stoke Newington 2019-10-24T16:02:32 < Jan-> Poor guy 2019-10-24T16:05:27 < Steffanx> Lol precautions, englishman ? 2019-10-24T16:05:53 < Steffanx> Nottingham, doomba 2019-10-24T16:06:04 < Steffanx> Uni. 2019-10-24T16:31:13 < Haohmaru> shot by a fellow englishman 2019-10-24T16:31:23 < Haohmaru> so unpatrioticf 2019-10-24T16:31:42 < Mangy_Dog> hay i lived near stoke newington 2019-10-24T16:32:23 < Mangy_Dog> newington green was my haunt 2019-10-24T16:35:24 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T16:35:48 < Steffanx> The finnisher arrived. 2019-10-24T16:35:59 < kakipro> morgon 2019-10-24T16:36:10 < kakipro> do I want this Steffanx https://www.youtube.com/watch?v=SOWi-S4UwkU 2019-10-24T16:36:51 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-24T16:37:47 < Steffanx> Kakinet failed the stress test 2019-10-24T16:37:49 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T16:38:00 < kakipro> do I want it steff? 2019-10-24T16:38:52 < Steffanx> Nah ty 2019-10-24T16:39:06 < Steffanx> I'll forget to charge it 2019-10-24T16:39:08 < mawk> sharkjack 2019-10-24T16:39:08 < mawk> lol 2019-10-24T16:40:12 < kakipro> you bet it won't run long 2019-10-24T16:41:59 < srk> /win 20 2019-10-24T16:42:13 < kakipro> 10minutes 2019-10-24T16:42:41 < kakipro> is this product like 1 day fresh? 2019-10-24T16:43:08 < doomba> kakivir32.exe and Steffanuxnet have taken over our critical systems 2019-10-24T16:47:47 < Jan-> IT'S RAINING. 2019-10-24T16:47:51 < Jan-> We were supposed to go for a walk. 2019-10-24T16:47:52 < Jan-> this sucks. 2019-10-24T16:48:06 < Haohmaru> go swimming instead? 2019-10-24T16:48:28 < Haohmaru> blend with teh circumstances 2019-10-24T16:48:33 < Haohmaru> adapt 2019-10-24T16:50:39 < Jan-> Well, the gym 2019-10-24T16:50:42 * Jan- does not like swimming 2019-10-24T16:50:46 < Jan-> but either of those we'd walk to 2019-10-24T17:12:27 -!- ronoxx [~oldmanbee@193-116-76-133.tpgi.com.au] has quit [Read error: Connection reset by peer] 2019-10-24T17:16:11 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Quit: The Lounge - https://thelounge.github.io] 2019-10-24T17:47:01 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-24T17:48:08 <@englishman> Steffanx: no precaution, he was already banned and was ban evading. 2019-10-24T17:48:42 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-24T17:51:38 < Haohmaru> Jan- i meant.. swim in teh rain instead of walking 2019-10-24T18:06:13 < Jan-> pah. 2019-10-24T18:06:16 < Jan-> dampness :( 2019-10-24T18:09:20 * Mangy_Dog slobbers jans face for the extra effect 2019-10-24T18:17:10 < Jan-> if Mangy_Dog is not good I will put him outside. 2019-10-24T18:17:12 < Jan-> IN THE RAIN. 2019-10-24T18:19:11 * Mangy_Dog cries 2019-10-24T18:19:18 < Mangy_Dog> but no one will see the tears in the rain 2019-10-24T18:44:59 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-24T19:00:10 -!- ekaologik [~quassel@p5DC6B28A.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-24T19:02:20 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-24T19:03:24 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-24T19:25:13 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-24T19:41:01 < con3> so i have a use case where i want to read data from an adc and take an action if the read value is in a specific interval but do nothing it the action is already taken. It feels janky to just set a variable when the action has already been taken? Would this be the best way to do it :/ 2019-10-24T19:44:38 < Steffanx> Use the window interrupt stuff and disable the interrupt afterwards? 2019-10-24T19:46:03 < Steffanx> But it might still involve variables depending on what you want 2019-10-24T19:54:38 < con3> didnt know there's a window interrupt, quickly checking 2019-10-24T19:54:41 < con3> thanks Steffanx 2019-10-24T19:57:02 < Steffanx> I never used it and not sure if its suitable or if you mcu even has it 2019-10-24T19:57:30 < con3> yeah just checking up on it, first time ive heard of it 2019-10-24T19:57:38 < con3> else ill do the little variable set thing 2019-10-24T20:07:09 < kakim32> https://www.youtube.com/watch?v=vwFO-seKnBM short film 2019-10-24T20:08:09 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-24T20:20:32 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-24T20:23:09 < bitmask> muh pcb is becoming a mess 2019-10-24T20:23:27 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-24T20:23:55 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-24T20:24:16 -!- onio [~onio@2a00:23c5:7a01:8600:3584:8be1:a96f:a113] has quit [Ping timeout: 252 seconds] 2019-10-24T20:24:43 < bitmask> https://i.imgur.com/EN3LbBk.png 2019-10-24T20:24:48 < bitmask> and I still have to route 5V and GND for the mcu 2019-10-24T20:29:48 -!- machinehum [~walker@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-24T20:30:33 < antto> wuh iz dis 2019-10-24T20:31:02 < bitmask> heated jacket 2019-10-24T20:31:20 < bitmask> avr, 7 heating panels with thermistors, and stuff 2019-10-24T20:31:32 < bitmask> first bare mcu project 2019-10-24T20:31:59 < antto> like, clothes with integrated radiator? 2019-10-24T20:32:03 < bitmask> yup 2019-10-24T20:32:07 < antto> wowz 2019-10-24T20:32:14 < antto> innovation? 2019-10-24T20:32:59 < bitmask> theres a bunch out there, but the batteries don't last long enough and this has pid temp control whereas the ones out there are usually just a pwm at a certain level or something 2019-10-24T20:33:06 < antto> in teh winter i mostly have problems with teh outter-most ends, hands and feet 2019-10-24T20:33:49 < machinehum> There's a lot of incels and nazi's on efnet 2019-10-24T20:33:55 < bitmask> oops gotta go, be back in half an hour 2019-10-24T20:35:07 -!- doomba [~npc@slipgate.logbook.pw] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-24T20:38:41 -!- roomba [~npc@slipgate.logbook.pw] has joined ##stm32 2019-10-24T20:39:17 -!- roomba is now known as doomba 2019-10-24T20:43:41 < Cracki> why are you advertising for efnet? 2019-10-24T20:44:06 < Cracki> got a channel you would highlight as a good example of those qualities? 2019-10-24T20:47:29 < Steffanx> #l0de 2019-10-24T20:48:11 < machinehum> lol 2019-10-24T20:54:58 < Steffanx> where is ReadError_? 2019-10-24T20:55:06 < Steffanx> did they find his body yet? 2019-10-24T21:07:01 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has joined ##stm32 2019-10-24T21:28:56 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-24T21:35:47 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-24T21:47:42 <@englishman> no 2019-10-24T21:50:47 < Steffanx> Still not sure how much of joke this is. or not at all. 2019-10-24T21:57:54 < antto> 2019-10-24T21:58:31 <@englishman> he's been gone like a year and a half 2019-10-24T21:58:46 <@englishman> just hangs out in youtube livestream chats now 2019-10-24T22:01:23 < kakipro> who? 2019-10-24T22:19:25 < doomba> oh hell yeah lode radio hour 2019-10-24T22:19:32 < doomba> the efnet talk show! 2019-10-24T22:20:16 < doomba> i love their channel. they have some pro-level ansi art botnets in there that illustrate the current state of world affairs quite well. 2019-10-24T22:27:00 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-24T22:27:38 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 240 seconds] 2019-10-24T22:27:39 -!- BrainDamage_ is now known as BrainDamage 2019-10-24T22:36:49 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 265 seconds] 2019-10-24T22:49:23 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-24T23:20:41 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 264 seconds] 2019-10-24T23:22:37 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-24T23:23:00 -!- con3 [~kvirc@154.119.40.237] has quit [Read error: Connection reset by peer] 2019-10-24T23:40:55 -!- mitrax [mitrax@lfbn-ncy-1-393-156.w83-196.abo.wanadoo.fr] has joined ##stm32 2019-10-24T23:50:40 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-24T23:54:51 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has quit [Ping timeout: 240 seconds] 2019-10-24T23:56:53 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 --- Day changed Fri Oct 25 2019 2019-10-25T00:08:07 < Jan-> I am not loving the cube ide experience :( 2019-10-25T00:08:15 < Jan-> it generates all this huge amont of code 2019-10-25T00:08:24 < Jan-> and you have no idea what it does or how to use it 2019-10-25T00:08:40 < Jan-> we turned on the DACs, and a huge function called MX_DAC_Init showed up 2019-10-25T00:08:57 < Jan-> there is no info on what it does 2019-10-25T00:09:07 < Jan-> or what to change if we might want to have it work differently. 2019-10-25T00:09:13 < Jan-> I am a bit depressed, we can't do this :( 2019-10-25T00:09:20 < mawk> just don't look at it 2019-10-25T00:09:24 < mawk> why can't you do this ?? 2019-10-25T00:09:29 < mawk> just don't look at the generated code 2019-10-25T00:09:33 < mawk> do your own stuff that's all 2019-10-25T00:09:39 < mawk> why do you have to read *generated* code ? 2019-10-25T00:12:43 < kakipro> Jan-: there are faster atmel 8bit chips than 328 2019-10-25T00:12:53 < kakipro> newer core and shit 2019-10-25T00:13:22 < kakipro> if your application doesn't necessarily require 32bit 2019-10-25T00:13:27 < kakipro> or benefit from it 2019-10-25T00:14:41 < kakipro> I bet MX_DAC_Init configures dac per setting you set to cubemx 2019-10-25T00:15:26 < kakipro> usually it just one big blob of code and then setting you set comment parts of it out 2019-10-25T00:16:01 < kakipro> can't you just ctrl+doubleclick it and see what there is? 2019-10-25T00:17:03 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Ping timeout: 250 seconds] 2019-10-25T00:17:03 < kakipro> it should use self-describing symbol naming 2019-10-25T00:17:11 -!- Steffann [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-25T00:17:51 < kakipro> congrats for new nick steff 2019-10-25T00:18:49 < Jan-> sConfig.DAC_Trigger = DAC_TRIGGER_NONE;? 2019-10-25T00:19:00 < kakipro> event source probs 2019-10-25T00:19:11 < machinehum> Jan-: libopencm3 2019-10-25T00:19:17 < kakipro> peripherals can generate events 2019-10-25T00:19:17 < Steffann> Ty kakipro 2019-10-25T00:19:25 < Jan-> machinehum: I would. But it's lgpl, and that doesn't work for us. 2019-10-25T00:19:27 < machinehum> Jan-: Probably becuase there's no trigger 2019-10-25T00:19:39 < kakipro> you want software trigger Jan-? 2019-10-25T00:19:55 < machinehum> Jan-: What's the cube licence? 2019-10-25T00:19:57 < Jan-> I assumed I could start it converting at any time. 2019-10-25T00:20:03 < machinehum> I thought it was more restrictive 2019-10-25T00:20:17 < kakipro> you need to look into reference manual Jan- 2019-10-25T00:20:38 < Jan-> machinehum: BSD 2019-10-25T00:20:41 < Jan-> the 3 clause one 2019-10-25T00:20:49 < Jan-> anything else would be crazy for firmware 2019-10-25T00:20:52 < kakipro> certainly you can set it to convert whenever you want 2019-10-25T00:20:56 < Jan-> lgpl is a bit silly for firmware to be honest 2019-10-25T00:21:16 < Jan-> kakipro I have googled around a lot for information, I was expecting some sort of API reference at ST's site. 2019-10-25T00:21:19 < Jan-> But there is none. 2019-10-25T00:21:55 < kakipro> have you opened the reference manual? 2019-10-25T00:22:28 < kakipro> after you see DAC section the hal-code is self-explanatory 2019-10-25T00:23:08 < specing> Jan-: I just released AGPLv3 firmware :) 2019-10-25T00:23:35 * specing hears the sound of torches and pitchforks in the distance 2019-10-25T00:23:47 < Jan-> if there was intellisense you could write HAL_DAC_ and see what popped up 2019-10-25T00:24:04 < kakipro> ctrl + space 2019-10-25T00:24:06 < mawk> don't put bad ideas in Jan-'s head kakipro 2019-10-25T00:24:12 < mawk> Jan- must carry on with STM32 2019-10-25T00:24:17 < mawk> it's the golden ticket to heaven 2019-10-25T00:24:24 < kakipro> and not become traitor like me 2019-10-25T00:24:29 < mawk> exactly 2019-10-25T00:24:30 < Jan-> I have to say I am not loving the stm32 experience so far 2019-10-25T00:24:37 < mawk> why are you saying this Jan- :( 2019-10-25T00:24:39 < Jan-> there is a lot of documentation 2019-10-25T00:24:41 < mawk> you have a custom board, you can't complain 2019-10-25T00:24:47 < Jan-> the problem is it's all for 300 different APIs. 2019-10-25T00:24:48 < machinehum> Jan-: What did you use before? 2019-10-25T00:24:54 < mawk> if you have a nucleo two clicks in any toolchain ever and you have all the blink blink in the world 2019-10-25T00:24:56 < Jan-> machinehum: atmega328 2019-10-25T00:25:04 < kakipro> you expect too much Jan- 2019-10-25T00:25:10 < machinehum> Jan-: Oh 2019-10-25T00:25:10 < mawk> use HAL and only HAL, that's 1 API 2019-10-25T00:25:11 < Jan-> mawk that's not really the problem I got it working fine in each IDE. 2019-10-25T00:25:14 < mawk> with 1 documentation and nothing else 2019-10-25T00:25:19 < mawk> and it works very fine 2019-10-25T00:25:19 < machinehum> Yeah I've made that switch beore 2019-10-25T00:25:22 < machinehum> before 2019-10-25T00:25:31 < mawk> why do you have to read MX generated code ? don't hurt your eyes, you don't have to modify it 2019-10-25T00:25:44 < Jan-> well presumably it sets options 2019-10-25T00:25:46 < machinehum> Just give in to the fact that you'll be calling functions you don't understand 2019-10-25T00:25:46 < mawk> yeah 2019-10-25T00:25:47 < Jan-> what if I need to modify them 2019-10-25T00:25:49 < mawk> so don't look at them 2019-10-25T00:25:54 < mawk> then you modify in cubeMX and generate again 2019-10-25T00:25:55 < kakipro> you cube them settings 2019-10-25T00:25:57 < mawk> don't look at the code 2019-10-25T00:26:07 < mawk> for the docs it's in the .c files Jan- interspersed with the code 2019-10-25T00:26:07 < karlp> (that's the whole _point_ of generated code) 2019-10-25T00:26:11 < Jan-> cube mx code is made of suck lose and fail 2019-10-25T00:26:19 < mawk> I believe if you run a quick doxygen command you have nice .html docs 2019-10-25T00:26:24 < mawk> it's all in doxy format 2019-10-25T00:26:29 < machinehum> mawk: yes 2019-10-25T00:26:45 < Jan-> or as cube mx would say, /* USER COMMENT BEGIN */ I hate cube MX /* USER COMMENT END */ 2019-10-25T00:26:49 < mawk> you're not constantly using cubeMX anyway Jan- , once you have your peripherals set up 2019-10-25T00:26:54 < kakipro> Jan-: what I have heard most of the time it does the exact thing you set it up to 2019-10-25T00:26:57 < mawk> you can make your own .c files 2019-10-25T00:27:15 < Jan-> er, /* RESPONSE TO MAWK */ Yeah I know but still /* RESPONSE TO MAWK END */ 2019-10-25T00:27:19 < machinehum> Jan-: Delete all that shit out. It's fucking gross 2019-10-25T00:27:21 * Jan- waves her arms 2019-10-25T00:27:26 < Jan-> it fucking is! 2019-10-25T00:27:39 < machinehum> Then just use commit before code generation 2019-10-25T00:27:52 < machinehum> and use a mergetool to merge in the new stuff 2019-10-25T00:27:55 < Jan-> oh sorry. /* EXPLETIVE LADEN RANT BEGIN */ Fucking fuck. /* EXPLETIVE LADEN RANT END */ 2019-10-25T00:28:10 * Jan- runs around in circles, waving her arms 2019-10-25T00:28:11 < mawk> that sounds annoying machinehum 2019-10-25T00:28:21 < mawk> versus just leaving generated code where it is and having your own code in your own files 2019-10-25T00:28:23 < mitrax> on F4 when using the independent watchdog (IWDG not WWDG), and a reset occurs is there a flag in some register that you can check to know if the reset was triggered by the watchdog? 2019-10-25T00:28:54 < machinehum> mawk: It's not, how often are you generating 2019-10-25T00:29:06 < Cracki> mitrax, I think there is such a thing. rcc registers maybe? 2019-10-25T00:29:09 < kakipro> oh I forgot it's not so much fun to read stm32 ref manual. every peripheral is like a swiss knife 2019-10-25T00:29:29 < Jan-> so what the hell is the HAL call to set a value on the DAC 2019-10-25T00:29:34 < Jan-> or at least to start the process of setting the value 2019-10-25T00:29:42 < Jan-> and more to the point where would I even find that out 2019-10-25T00:29:55 < machinehum> HAL_Set_Value 2019-10-25T00:29:57 < Cracki> hal docs, hal examples? 2019-10-25T00:30:03 < machinehum> HAL_DAC_Set_Value* 2019-10-25T00:30:06 < machinehum> Something? 2019-10-25T00:30:14 < machinehum> Read the top of each of the deps 2019-10-25T00:30:31 < machinehum> There's a bulleted list on how to use the periph 2019-10-25T00:30:39 < mitrax> nvm found it, there's RCC_FLAG_IWDGRST 2019-10-25T00:30:51 < mitrax> Cracki: yes indeed :) didn't see your message sorry 2019-10-25T00:31:28 < Cracki> 14.2.2 How to use this driver 2019-10-25T00:31:30 < Jan-> HAL_DAC_SetValue 2019-10-25T00:31:31 < Cracki> 14 DAC 2019-10-25T00:31:40 < Jan-> ctrl+space is your (well my) friend 2019-10-25T00:31:41 < machinehum> Jan-: Yeah sure that 2019-10-25T00:31:47 < Jan-> needs four arguments 2019-10-25T00:31:47 < karlp> APINAME_PERIPHERAL_VERB 2019-10-25T00:32:01 < Jan-> hdac sounds like win32 stuff again, like hwnd 2019-10-25T00:32:16 < machinehum> It's the dac handle 2019-10-25T00:32:29 < Jan-> ppft you can't have a fricken handle on a microcontroller 2019-10-25T00:32:31 < Jan-> that's windows stuff 2019-10-25T00:32:42 < Jan-> whatever next, hresults? 2019-10-25T00:32:44 < Jan-> :D 2019-10-25T00:34:05 < machinehum> PaulFertser: I say stlink V3 support was merged 2019-10-25T00:34:08 < machinehum> saw* 2019-10-25T00:34:12 < machinehum> That's exciting 2019-10-25T00:34:35 < Jan-> it seems to use hdac without ever defining it 2019-10-25T00:34:53 < Jan-> it says hdac.Instance = DAC; at one point 2019-10-25T00:35:01 < mawk> ? 2019-10-25T00:35:06 < mawk> it's a global, no ? 2019-10-25T00:35:18 < Jan-> actually yes it is... 2019-10-25T00:35:23 < Jan-> it's a DAC_HandleTypeDef 2019-10-25T00:35:31 < mawk> yes 2019-10-25T00:35:51 < mawk> which is a structure holding initial parameters and an instance (pointer to actual registers) 2019-10-25T00:36:11 < Jan-> god this is so much crapper than visual studio 2019-10-25T00:36:17 < Jan-> can you get information on what the arguments represent 2019-10-25T00:36:20 < machinehum> I wish hal was written in C++ 2019-10-25T00:36:32 < machinehum> That would make it so much better 2019-10-25T00:36:32 < mawk> just open the .c Jan- 2019-10-25T00:36:43 < mawk> hal_stm32fxxxxx_dac.h 2019-10-25T00:36:48 < kakipro> hover over function instance Jan- 2019-10-25T00:36:50 < mawk> and the .c has the docs 2019-10-25T00:37:21 < Jan-> when I google "hal_dac_setvalue" it should just pop up a page on ST's site 2019-10-25T00:37:23 < Jan-> but no 2019-10-25T00:37:27 < kakipro> sometimes it shows part of comment in hover view 2019-10-25T00:37:50 < Cracki> ST are a bunch of offliners 2019-10-25T00:38:09 < Jan-> no kidding 2019-10-25T00:38:11 < Jan-> this blows guys 2019-10-25T00:38:13 < Cracki> you'll have more chance of finding docs on GD or CKS or somesuch 2019-10-25T00:38:13 < kakipro> get the reference manual and go into woods 2019-10-25T00:38:39 < Cracki> hacking in the woods, aha 2019-10-25T00:38:45 < Jan-> OK so this argument is type uint32_t, okay fine, but the values are DAC_CHANNEL_1 or DAC_CHANNEL_2. Shouldn't that be an enum? 2019-10-25T00:39:03 < Cracki> you expect these people to know what an enum is? 2019-10-25T00:39:10 < Jan-> which people 2019-10-25T00:39:33 < Cracki> you know... THOSE people 2019-10-25T00:39:45 < mawk> no docs online Jan- 2019-10-25T00:39:47 < Jan-> OK so the actual definition isn't the numeric values 0 and 1, it's a bitfield so it's 1 and 2. 2019-10-25T00:39:48 < mawk> just generate them I said 2019-10-25T00:39:50 < mawk> or go look in the .c 2019-10-25T00:39:59 < mawk> it's very clearly explicated 2019-10-25T00:40:13 < mawk> why do you want to set values manually ? use the constants 2019-10-25T00:40:18 < mawk> your code will be self-documenting 2019-10-25T00:40:21 < Jan-> it wants an alignment, such as DAC_ALIGN_12B_R 2019-10-25T00:40:33 < mawk> so you can give it that 2019-10-25T00:40:35 < Jan-> that means it'll set the 12 least significant bits of a uint32_t as the DAC value? 2019-10-25T00:40:54 < kakipro> R = right most likelly 2019-10-25T00:41:00 < kakipro> or register 2019-10-25T00:41:03 < Jan-> R is right yes 2019-10-25T00:41:09 < Jan-> DAC_ALIGN_12B_R: 12bit right data alignment selected 2019-10-25T00:41:30 < Jan-> so it's basically between 0 and 4096 2019-10-25T00:41:38 < kakipro> it is how you said it is 2019-10-25T00:41:41 < mawk> yes 2019-10-25T00:41:57 < Jan-> okay. 2019-10-25T00:42:42 < Jan-> here goes nothing wish me luck 2019-10-25T00:42:43 * Jan- hits F11 2019-10-25T00:42:53 < mawk> build regularly 2019-10-25T00:42:58 < mawk> to avoid bad surprises 2019-10-25T00:43:03 < Jan-> I added 2 lines gimme a break 2019-10-25T00:43:10 < machinehum> fullscreen? 2019-10-25T00:43:11 < mawk> yeah 2 lines is good 2019-10-25T00:43:28 < mawk> build and run or something 2019-10-25T00:44:01 < Jan-> this is the same endian-ness as an avr right 2019-10-25T00:44:07 < mawk> little endian 2019-10-25T00:44:14 < mawk> ARM could be both, but stm32 chose little endian 2019-10-25T00:44:19 < mawk> so, not aligned like humans 2019-10-25T00:44:31 < Jan-> er 2019-10-25T00:44:32 < mawk> 0x123456 will be encoded as [0x56, 0x34, 0x12] 2019-10-25T00:44:43 < Jan-> I'm thinking of the alignment 2019-10-25T00:44:51 < mawk> yeah 2019-10-25T00:44:54 < Cracki> related 2019-10-25T00:44:57 < Jan-> if I set it to right aligned will it just take the lowest 12 bits of a uint32_t 2019-10-25T00:45:03 < mawk> well 0x00000012 will be encoded as [0x12, 0x00, 0x00, 0x00] 2019-10-25T00:45:08 < Cracki> correct, lowest 12 2019-10-25T00:45:19 < Cracki> alignment is relative to logical layout 2019-10-25T00:45:26 < Cracki> so independent of endianness 2019-10-25T00:45:42 < Cracki> typically you don't have to think about alignment at all. 2019-10-25T00:46:22 < Cracki> a word being 0xABCDEFGH, right alignment means 0xFGH is the value you get. 2019-10-25T00:46:41 < kakipro> unless you are about to communicate large values out 2019-10-25T00:47:01 < kakipro> in byte basis 2019-10-25T00:47:35 < zyp> AVR is 8-bit and as such doesn't have an inherent endianness, strictly speaking 2019-10-25T00:47:59 < zyp> but compilers need to choose one when laying out larger variables 2019-10-25T00:48:06 < karlp> well, it's memory is laid out in a certain way such that type punning matters.. 2019-10-25T00:48:15 < Jan-> so it says DAC_OUT1 is PA4 2019-10-25T00:48:21 < karlp> but hey, this is why type punning is bad :) 2019-10-25T00:48:21 < Jan-> we are measuring PA4 2019-10-25T00:48:32 < mawk> if you set PA4 as DAC input yes 2019-10-25T00:48:33 < mawk> in MX 2019-10-25T00:48:36 < zyp> karlp, AFAIK that's completely compiler side, not hardware side 2019-10-25T00:48:46 < Jan-> DAC is an output surely 2019-10-25T00:48:53 < mawk> yes output sorry 2019-10-25T00:48:54 < karlp> zyp: I agree it probably theoretically is :) 2019-10-25T00:49:02 < Jan-> you had me worried there :) 2019-10-25T00:49:05 < mawk> lol 2019-10-25T00:49:11 < Jan-> wtf, inp.. huh? 2019-10-25T00:49:28 < Jan-> and I'm like HAL_DAC_SetValue(&hdac, DAC_CHANNEL_1, DAC_ALIGN_12B_R, 1024); 2019-10-25T00:49:40 < Jan-> then wait a second, then HAL_DAC_SetValue(&hdac, DAC_CHANNEL_1, DAC_ALIGN_12B_R, 3072); 2019-10-25T00:49:49 < Jan-> we know the loop is running as it also flashes an LED 2019-10-25T00:49:52 < specing> JAN_IRC_SetComplains(true); 2019-10-25T00:50:03 * Jan- pushes specing into the scorpion pit 2019-10-25T00:50:10 < Jan-> *shove* 2019-10-25T00:50:50 < specing> Scorpion pits are unsupported by cubemx 2019-10-25T00:51:15 < kakipro> they say that you can do anything with cubemx 2019-10-25T00:51:16 < Jan-> oh right yes 2019-10-25T00:51:23 * Jan- pushes specing into the /* BEGIN PUNISHMENT ZONE */ scorpions(); /* END PUNISHMENT ZONE */ 2019-10-25T00:51:34 < Jan-> now THAT is supported by cube mx. 2019-10-25T00:51:54 < specing> have fun implementing scorpions() 2019-10-25T00:52:04 < Jan-> you can find anything on github 2019-10-25T00:52:12 < mawk> I hate other people's code 2019-10-25T00:52:17 < specing> Jan-: but its GPL 2019-10-25T00:52:17 < mawk> so I never search on github 2019-10-25T00:52:31 < mawk> I use HAL because it's master pro corporate code 2019-10-25T00:52:43 < Jan-> OK so PA4 and PA5 are both sitting at 0.45 volts 2019-10-25T00:52:56 < mawk> so it works ? 2019-10-25T00:53:13 < Jan-> well no it should be alternating between 1024 and 3072 2019-10-25T00:53:31 < mawk> you correctly enabled DAC in mx right ? 2019-10-25T00:53:39 < mawk> not just setting pin mode, but enabling DAC peripheral 2019-10-25T00:53:39 < Jan-> which should be like 0.8 volts and 2.47 volts 2019-10-25T00:53:50 < kakipro> which reference did you use? 2019-10-25T00:53:56 < mawk> if you're really nice I can get me stm32 down the closet and give you the code 2019-10-25T00:54:01 < kakipro> voltage reference* 2019-10-25T00:54:02 < mawk> why would you need a reference kakipro ? 2019-10-25T00:54:03 < mawk> ah 2019-10-25T00:54:10 < Jan-> the options are OUT1 configuration and OUT2 configuration which we checked 2019-10-25T00:54:22 < Jan-> then for each channel, output buffer is enabled and trigger is none 2019-10-25T00:54:31 < Jan-> I guess I need some other call to make it go 2019-10-25T00:54:38 < mawk> it should say in the RM 2019-10-25T00:54:41 < kakipro> did you configure the pin? 2019-10-25T00:54:44 < mawk> it should be your bible 2019-10-25T00:54:45 < mawk> yes kakipro 2019-10-25T00:55:01 < kakipro> 0.45 sounds like some floating voltage level 2019-10-25T00:55:09 < mawk> maybe bad probing 2019-10-25T00:55:10 < Jan-> oh do I need a HAL_DAC_Start or something 2019-10-25T00:55:10 < mawk> yeah 2019-10-25T00:55:18 < mawk> do you see it somewhere Jan- ? 2019-10-25T00:55:24 < mawk> otherwise yeah 2019-10-25T00:55:33 < mawk> there are always like blocking mode, irq mode, dma mode 2019-10-25T00:55:35 < mawk> for many peripherals 2019-10-25T00:55:44 < mawk> so when you have the choice you need to select that using _Start functions 2019-10-25T00:55:45 < Jan-> this is why I don't like this way of doing it 2019-10-25T00:55:46 < kakipro> use blocking mode 2019-10-25T00:55:56 < mawk> you don't like having choice ?? 2019-10-25T00:55:56 < Jan-> somwehere in mx_dac_init it might be 2019-10-25T00:56:05 < mawk> yes use blocking mode 2019-10-25T00:56:05 < Jan-> but I don't find a _start anywhere 2019-10-25T00:56:14 < mawk> yeah so add it yourself in *your* code 2019-10-25T00:56:17 < kakipro> blocking mode = thread waits while process is completed 2019-10-25T00:56:25 < Jan-> I'm OK with the thread continuing 2019-10-25T00:56:30 < Jan-> it's got 1 second 2019-10-25T00:56:35 < Jan-> it should be able to set a fricken dac value in 1 second 2019-10-25T00:57:03 < kakipro> yes 2019-10-25T00:58:25 < Jan-> yep that fixed it 2019-10-25T00:58:25 < kakipro> but let's say you drive audio to DAC 2019-10-25T00:58:26 < Jan-> we have dac 2019-10-25T00:58:35 < kakipro> blocking mode is no option then 2019-10-25T00:58:42 < Jan-> well no kidding 2019-10-25T00:58:54 < Jan-> in my use case I might be doing long slow fades of lights so it will be setting values super often 2019-10-25T00:59:00 < Jan-> so I might want something async then 2019-10-25T00:59:25 < Jan-> but for now this is fine 2019-10-25T01:00:03 < mawk> ah 2019-10-25T01:00:20 < mawk> what are you connecting the DAC to Jan- ? 2019-10-25T01:01:02 < Jan-> the main board of a super china brand B3603 power supply 2019-10-25T01:01:11 < Jan-> normally it has a manufacturer-supplied daughter board on top of it for setting the voltage 2019-10-25T01:01:25 < Jan-> but we are planning to control it externally. It takes analog voltages. 2019-10-25T01:01:27 < mawk> ah 2019-10-25T01:01:35 < mawk> and it doesn't source much current ? 2019-10-25T01:01:40 < mawk> you can't load too much the stm32 2019-10-25T01:01:57 < Jan-> I doubt it it's only got a tiny microcontroller on there by default 2019-10-25T01:01:58 < karlp> you can totally do more than 1Hz in blocking mode, what sort of maddness world are you in. 2019-10-25T01:03:39 < karlp> it's rated to 1Msps, you can do that in blocking mode too, if you don't want to do much else. 2019-10-25T01:03:39 < Jan-> but I hadn't really considered that mawk, good thought 2019-10-25T01:03:39 < Jan-> karlp: I may not need async then :D 2019-10-25T01:03:39 < Jan-> in the end we will need way more DACs than the stm32f4 has 2019-10-25T01:03:39 < karlp> no, you just need to stop moaning and hating and trolling. 2019-10-25T01:03:39 < karlp> that might be difficult though. 2019-10-25T01:03:49 < kakipro> hell.. 100hz even 2019-10-25T01:03:49 < kakipro> but depends on other load the mcu has 2019-10-25T01:04:54 < qyx> wheres grumpy cat, dit it died? 2019-10-25T01:06:48 < Jan-> god's sake karl who's trolling here 2019-10-25T01:06:57 < Jan-> I'm just sitting here going through some code, there's no need to be nasty 2019-10-25T01:07:12 < Jan-> anyway it's all working 2019-10-25T01:07:36 < Jan-> does it say somewhere here that this thing has ethernet? I guess that would require a whole tcp/ip stack to be useful. 2019-10-25T01:08:25 < kakipro> lwip 2019-10-25T01:08:35 < kakipro> and probs you should then have rtos too 2019-10-25T01:08:52 < Jan-> yes at that point, possibly mbed is the best way to go :) 2019-10-25T01:08:55 < mawk> your board has ethernet yes Jan- 2019-10-25T01:08:58 < mawk> why ???? 2019-10-25T01:09:01 < mawk> what are you saying now 2019-10-25T01:09:07 < Jan-> I was just surprised 2019-10-25T01:09:18 < mawk> HAL has ethernet support 2019-10-25T01:09:40 < Jan-> that would let you set a mac address and receive frames I guess 2019-10-25T01:09:46 < Jan-> but to be useful you would need the stack 2019-10-25T01:09:54 < mawk> yeah 2019-10-25T01:10:11 < Jan-> it's just that there are ethernet controlled lights out there 2019-10-25T01:10:22 < mitrax> Jan-: lwip is bundled with stm32cube 2019-10-25T01:10:44 < Jan-> crikey. 2019-10-25T01:10:46 < Jan-> so you can do it. 2019-10-25T01:11:07 < Jan-> I think I read that this board would need an external ethernet physical layer chip but you can get a little board? 2019-10-25T01:11:36 < mawk> I would've thought your board has its own PHY, but why not 2019-10-25T01:11:48 < Jan-> the board certainly doesn't if the chip doesn't 2019-10-25T01:11:56 < mawk> I mean the chip, yes 2019-10-25T01:12:18 < Jan-> the board has got the stm32f407, what I think is an SPI flash chip, and a battery. Oh and a usb socket and an sd card slot. 2019-10-25T01:12:31 < Jan-> that's it 2019-10-25T01:13:33 < Jan-> not sure what the spi flash chip is for... 2019-10-25T01:16:13 < kakipro> more flash 2019-10-25T01:16:28 < Jan-> can't put program code in it though? 2019-10-25T01:16:49 < Jan-> OK yes it would need an external PHY chip. 2019-10-25T01:16:52 < Jan-> It has a MAC chip. 2019-10-25T01:16:56 < Jan-> er, MAC peripheral? 2019-10-25T01:16:57 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-25T01:17:03 < kakipro> probs cannot execute it directly 2019-10-25T01:17:10 < kakipro> but from ram for sure 2019-10-25T01:17:28 < kakipro> more likelly chip would be used for data storage 2019-10-25T01:17:44 < kakipro> as you don't need to play with internal flash writing and stuff 2019-10-25T01:17:54 < Jan-> sure 2019-10-25T01:18:35 < Jan-> SD cards are basicall SPI devices anyway aren't they? 2019-10-25T01:21:43 < kakipro> some support SPI as compatibility mode 2019-10-25T01:21:53 < machinehum> wtf is #stm32 and why do I need an invite? 2019-10-25T01:21:54 < kakipro> not the big ones though 2019-10-25T01:22:21 < machinehum> I want to be part of the highly exclusive group of weiners 2019-10-25T01:22:32 < kakipro> iirc 2019-10-25T01:22:44 < mawk> it's a fake channel that redirects to ##stm32 machinehum 2019-10-25T01:22:56 < mawk> since you're already on ##stm32 it doesn't redirect and just fail 2019-10-25T01:23:07 < kakipro> really? 2019-10-25T01:23:27 < mawk> absolutely 2019-10-25T01:23:33 < machinehum> Sounds like a coverup 2019-10-25T01:23:35 -!- machinehum [~walker@d207-216-21-173.bchsia.telus.net] has left ##stm32 ["WeeChat 1.9.1"] 2019-10-25T01:23:38 -!- machinehum [~walker@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-25T01:23:40 < machinehum> Oh 2019-10-25T01:23:44 < machinehum> Nvermind 2019-10-25T01:28:32 < machinehum> Anyone wanna make a highly exclusive channel? 2019-10-25T01:28:35 < machinehum> Or what 2019-10-25T01:28:42 < Jan-> I have one :D 2019-10-25T01:28:55 < Jan-> Also, can I please be a neckbeard? Only, no beard. 2019-10-25T01:29:01 < Jan-> is it an equal opportunity thing? 2019-10-25T01:29:18 < machinehum> invite plx 2019-10-25T01:29:39 < Jan-> No. It's highly exclusive :D 2019-10-25T01:30:02 < Ultrasauce> i have no neck and i must beard 2019-10-25T01:30:23 < Jan-> I have no beard, and I must neck 2019-10-25T01:34:50 < machinehum> How do I get in? 2019-10-25T01:35:24 < karlp> WLCSP can withstand three times the previous recommended reflow profile to be 2019-10-25T01:35:26 < karlp> compatible with a double reflow when SMDs are mounted on both sides of the PCB plus 2019-10-25T01:35:28 < karlp> one additional repair. 2019-10-25T01:35:33 < karlp> gotta be careful with them little bitties 2019-10-25T01:36:48 < kakipro> I hase neckbeard 2019-10-25T01:38:38 < Jan-> Trim it. 2019-10-25T01:38:42 < Jan-> Then you won't look like a linux sysadmin. 2019-10-25T01:42:13 < karlp> heh, st notes explicitly warn against light exposure on the wlcsp devices :) 2019-10-25T01:56:54 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-25T01:58:02 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-25T02:04:12 < Jan-> what's a wlcsp 2019-10-25T02:06:13 < kakipro> sounds like bare die 2019-10-25T02:06:32 < kakipro> or die with ball grid under it 2019-10-25T02:07:37 < kakipro> https://c44f5d406df450f4a66b-1b94a87d576253d9446df0a9ca62e142.ssl.cf2.rackcdn.com/2018/01/WLCSP1.png 2019-10-25T02:12:42 -!- machinehum [~walker@d207-216-21-173.bchsia.telus.net] has quit [Quit: WeeChat 1.9.1] 2019-10-25T02:15:51 < kakipro> codings done for the day 2019-10-25T02:27:27 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-25T02:30:46 < Jan-> I only actually wrote about 3 lines of code 2019-10-25T02:30:48 < Jan-> but they were important! 2019-10-25T02:38:30 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-25T02:44:23 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 250 seconds] 2019-10-25T02:49:50 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-25T02:49:56 < Cracki> think of all the great things you could do with this! https://www.youtube.com/watch?v=imOhgdLNSOc 2019-10-25T02:50:35 < kakim32> how bad it would be to create and delete task 100times a second 2019-10-25T02:51:11 < kakim32> :o 2019-10-25T02:52:30 < Cracki> I think it's cheap 2019-10-25T02:52:36 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-25T02:52:40 < Cracki> allocate some memory, init, done 2019-10-25T02:52:57 < Cracki> I would wonder why you wouldn't just keep it around 2019-10-25T02:54:32 < Jan-> thanks for the help guys 2019-10-25T02:54:34 < Jan-> I'm going to sleep 2019-10-25T02:54:39 * Jan- leaves a box of cookies for everyone 2019-10-25T03:39:37 -!- zoobab [zoobab@vic.ffii.org] has quit [Ping timeout: 240 seconds] 2019-10-25T03:46:30 -!- zoobab [zoobab@vic.ffii.org] has joined ##stm32 2019-10-25T03:53:53 < dongs> https://timschabe.com/blog/2019/10/19/retro-gaming-console/ 2019-10-25T03:55:31 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has left ##stm32 [] 2019-10-25T03:56:06 < Cracki> 3d printed feet 2019-10-25T03:56:12 < Cracki> oh well 2019-10-25T04:04:37 -!- karlp [karlp@palmtree.beeroclock.net] has quit [Quit: WeeChat 2.5] 2019-10-25T04:13:14 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-25T04:29:25 -!- karlp [karlp@palmtree.beeroclock.net] has joined ##stm32 2019-10-25T04:46:31 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 268 seconds] 2019-10-25T06:19:20 -!- fc5dc9d4_ [~quassel@p5B0811B0.dip0.t-ipconnect.de] has joined ##stm32 2019-10-25T06:22:57 -!- fc5dc9d4 [~quassel@p5B08108E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-25T06:33:47 -!- scrts2 [~scrts@d27-96-211-8.nap.wideopenwest.com] has quit [Changing host] 2019-10-25T06:33:47 -!- scrts2 [~scrts@unaffiliated/scrts] has joined ##stm32 2019-10-25T06:42:51 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-25T06:50:28 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 268 seconds] 2019-10-25T06:53:21 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-25T06:56:57 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-25T07:29:28 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-25T07:33:01 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 268 seconds] 2019-10-25T07:33:02 -!- day__ is now known as day 2019-10-25T08:21:06 -!- machinehum [~machinehu@S01061cabc0ab4603.vc.shawcable.net] has joined ##stm32 2019-10-25T09:08:29 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-25T09:14:24 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-25T09:20:08 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-25T09:20:11 -!- X230t [x13@gateway/shell/suchznc/x-qwrfcjkwcwzmeema] has quit [Quit: So long fuckers] 2019-10-25T09:30:37 < R2COM> dongs cadence is shit 2019-10-25T09:30:45 < R2COM> at least its IC flow tools 2019-10-25T09:30:56 < R2COM> RIPoff $$$$$ 2019-10-25T09:31:05 < R2COM> and GUI worse than win 3.1 2019-10-25T09:31:28 < R2COM> working 8 hours a day in EDA through VNC-like client called EoD 2019-10-25T09:35:41 -!- onio [~onio@host109-150-120-196.range109-150.btcentralplus.com] has joined ##stm32 2019-10-25T09:39:41 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-25T09:50:25 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-25T09:52:25 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-25T10:07:01 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-25T10:23:56 < jpa-> https://github.com/NordicPlayground/nrf52-ble-app-uart-long-range/blob/master/ble_app_uart_c/main.c#L543 does someone know what the ram_start stuff does here? 2019-10-25T10:24:27 < jpa-> it seems it would fetch the current stack pointer at that stage of initialization, but why would the bluetooth stack want that? after all the application can use more stack later.. 2019-10-25T10:32:29 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-25T10:42:37 -!- sterna [~Adium@2a02:aa1:1011:658f:c189:fbe9:c8bf:41bb] has joined ##stm32 2019-10-25T10:54:27 < mitrax> i'm providing implementations for newlib functions such as _sbrk, abort() etc in a static library that i link to my executable, things work fine, but as soon as i remove one of those function from the static lib and move it to the executable (e.g only abort()) , the linker uses the default implementations for the newlib functions and doesn't link the code from the static lib anymore, i get 2019-10-25T10:54:27 < mitrax> that newlib functions are declared as weak but is it expected that if a compilation unit is not providing all the implementations they won't be linked? 2019-10-25T10:54:48 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-25T10:58:01 < jpa-> mitrax: it's probably an issue with library search order 2019-10-25T10:58:56 < jpa-> you should probably try to adjust linker parameters so that the default implementations are not included in the search at all 2019-10-25T10:59:45 < jpa-> are you linking against -lnosys? 2019-10-25T11:02:20 -!- X230t [x13@gateway/shell/suchznc/x-ltasebypeedewdzy] has joined ##stm32 2019-10-25T11:09:39 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-25T11:10:09 < mitrax> jpa-: i have --specs=nosys.specs which iirc is equivalent 2019-10-25T11:13:26 -!- sterna [~Adium@2a02:aa1:1011:658f:c189:fbe9:c8bf:41bb] has quit [Ping timeout: 252 seconds] 2019-10-25T11:17:15 < jpa-> mitrax: does it help if you link with -Wl,--whole-archive myfunnyfunctions.a -Wl,--no-whole-archive 2019-10-25T11:19:06 < jpa-> otherwise what probably happens is that main.c needs foo() from newlib, and foo() needs _sbrk(), but the linker order is main.c myfunctions.a newlib nosys so it looks into myfunctions.a, sees no foo(), determines it doesn't need stuff from there, moves on to newlib, notices it also needs _sbrk(), but it doesn't go back to look in myfunctions.a but continues forwards to nosys, which contains 2019-10-25T11:19:06 < jpa-> implementation 2019-10-25T11:19:39 < jpa-> but if you have abort() in myfuncs.a and main.c calls it, linker brings in whole file from myfuncs.a and that also includes _sbrk 2019-10-25T11:20:03 < jpa-> --whole-archive will tell it to always include everything from myfuncs.a 2019-10-25T11:22:33 < mitrax> i see, let me try 2019-10-25T11:23:56 < jpa-> ah, regarding that ram_start stuff i asked about, looks like the first function writes to ram_start and second function wants that value from there 2019-10-25T11:31:03 < mitrax> jpa-: with --whole-archive i get multiple definition errors from the linker from some other functions, but thanks a lot, i understand what's going on now 2019-10-25T11:31:18 -!- onio [~onio@host109-150-120-196.range109-150.btcentralplus.com] has quit [Quit: Leaving] 2019-10-25T11:31:51 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-25T11:32:28 < jpa-> mitrax: yeah, linker is quite simplistic in that it will either bring in a whole .o file or nothing - so if nosys contains multiple definitions in a single file, you can either take all or none of them 2019-10-25T11:35:23 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-25T11:36:31 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-25T11:42:17 < mitrax> jpa-: yeah i've ran into library search order issues before, but nothing involving weak symbols with that behavior. After removing some symbols that shouldn't have been there --whole-archive solves the issue indeed, thanks again 2019-10-25T12:43:18 < karlp> yeha, the not backtracking thing can be a pig 2019-10-25T12:58:54 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-25T13:21:37 < Jan-> has anyone ever written a basic interpreter for stm32 2019-10-25T13:21:40 < Jan-> I assume they have 2019-10-25T13:25:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 252 seconds] 2019-10-25T13:25:17 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-25T13:27:22 < srk> I would like to write forth interpreter when I have nothing better to do :D 2019-10-25T13:30:23 < Jan-> we had a request, can we make a retro pocket computer with a vacuum fluorescent graphic display and have it run BASIC :D 2019-10-25T13:30:37 < Jan-> 10 PRINT "JAN IS SKILL" 2019-10-25T13:30:40 < Jan-> 20 GOTO 10 2019-10-25T13:34:42 < karlp> srk: explain that for me, there's a few forth ports already, is it really that forth people _only- like doing the port, and then not using them? 2019-10-25T13:35:10 < karlp> I saw one person with a usb stack for f4 in forth, that was neat, but it was just a demo, I couldn't find any apps of their own even where they used it. 2019-10-25T13:35:34 < karlp> it always seems to be, "look, forth!" and then nothing beyond. 2019-10-25T13:35:37 < zyp> I think it's just something to waste time on 2019-10-25T13:40:14 < srk> mostly, I would write a compiler as well 2019-10-25T13:41:02 < srk> than it might be useful :D 2019-10-25T13:48:31 < zyp> useful != used 2019-10-25T13:50:38 < jadew> I know what's wrong with the internet today 2019-10-25T13:50:47 < Haohmaru> i know too 2019-10-25T13:50:54 < jadew> ok, you first 2019-10-25T13:51:03 < Haohmaru> www is what's wrong 2019-10-25T13:52:02 < jadew> I'll tell you, it's a combination of two things 2019-10-25T13:52:09 < jadew> search engines and advertising 2019-10-25T13:52:18 < jadew> which leads to the need for SEO 2019-10-25T13:52:19 < Haohmaru> basically, google 2019-10-25T13:52:29 < jadew> and SEO means garbage content 2019-10-25T13:52:36 < Haohmaru> of course 2019-10-25T13:52:52 < jadew> I just searched for an explanation on something both in english and romanian 2019-10-25T13:53:13 < jadew> all the english sites that deal with it have longer than needed explanations that beat around the bush 2019-10-25T13:53:23 < Haohmaru> which websearch did ya use? 2019-10-25T13:53:33 < jadew> the romanian result was from the help section of an accounting program so it was a straight delivery 2019-10-25T13:53:36 < jadew> extremely concise and clear 2019-10-25T13:53:42 < jadew> Haohmaru, google 2019-10-25T13:53:47 < Haohmaru> well that's ur first mistake 2019-10-25T13:53:50 < Haohmaru> bruh 2019-10-25T13:54:09 < jadew> well, google does seem to have the best algo 2019-10-25T13:54:22 < Haohmaru> google kinda sorta eyeballz who you are a little too much before it "decides" which "results" to show u 2019-10-25T13:54:49 < jadew> everyone does that 2019-10-25T13:54:58 < Haohmaru> but it's wrong 2019-10-25T13:55:13 < Haohmaru> are we searching or fooling around here? 2019-10-25T13:55:19 < jadew> so what's the solution? I shouldn't search stuff anymore? 2019-10-25T13:56:06 < Haohmaru> unfortunately, we're surrounded by this brownness 2019-10-25T13:56:16 < Haohmaru> aka sh*t situation 2019-10-25T13:56:53 < zyp> yeah, let's go back to webrings instead 2019-10-25T13:57:33 < Haohmaru> i no longer use google to search 2019-10-25T13:57:45 < Haohmaru> FU google 2019-10-25T13:58:00 < jadew> what else are you using? 2019-10-25T13:58:04 < jadew> duckduckgo.nsa? 2019-10-25T13:58:06 < Haohmaru> they know i hate them and they troll me 2019-10-25T13:58:16 < Haohmaru> yez 2019-10-25T14:01:22 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 252 seconds] 2019-10-25T14:02:53 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-25T14:10:08 < Haohmaru> i like allpcb's instant quote webpage 2019-10-25T14:10:50 < Haohmaru> on some fields that are required, there's a tooltip that says "pls" 2019-10-25T14:10:53 < Haohmaru> >:) 2019-10-25T14:11:12 < Haohmaru> customer, pls 2019-10-25T14:11:31 < kakim32> https://drive.google.com/file/d/1G54YNYsTl1tJeio4e1KvMzfrXB5GwCuL/view?usp=sharing 2019-10-25T14:15:02 < kakim32> I wonder if it has serial port 2019-10-25T14:15:14 < kakim32> that I could utilize for management purposes 2019-10-25T14:17:21 < kakim32> could I possibly have both management interface and bridge in same ethernet cable 2019-10-25T14:30:50 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-25T14:33:07 < Haohmaru> wut is this какимир? 2019-10-25T14:34:12 < kakim32> did you know: ARM = Acorn Risc Machine 2019-10-25T14:34:21 < Haohmaru> yez 2019-10-25T14:34:47 < Haohmaru> i've read teh weekee pedia page 2019-10-25T14:35:09 < Haohmaru> or is it wiikuhpeedia 2019-10-25T14:35:17 < Haohmaru> bluh 2019-10-25T14:35:50 < Haohmaru> if i ever make a cpu, imma name it LEG 2019-10-25T14:36:44 < Haohmaru> i akchually wonder, has ARM registered the trademark LEG so that no one can use it to troll them? ;P~ 2019-10-25T14:37:01 < Haohmaru> trademark or name or however it's called 2019-10-25T14:58:31 < kakim32> I wonder if there is routerboard schematics somewhere? 2019-10-25T15:13:41 < Ultrasauce> https://www.youtube.com/watch?v=MjTjlUzBU-Y musicspam 2019-10-25T15:16:59 < kakim32> I hear ultrasauce musics 2019-10-25T15:17:38 < Ultrasauce> sort of 2019-10-25T15:18:30 < Ultrasauce> http://omae.wa.mou.shindei.ru/music/doodles/26.ogg 2019-10-25T15:19:11 < Jan-> Wow that sounds awesome 2019-10-25T15:19:18 < Jan-> what does ultra sauce taste like 2019-10-25T15:19:50 < Ultrasauce> why dont you come find out ( ͡° ͜ʖ ͡°) 2019-10-25T15:20:08 * karlp grins 2019-10-25T15:20:34 < Jan-> someone should market something called ultra sauce. 2019-10-25T15:20:44 < Jan-> you could have ultra sauce savory and ultra sauce sweet 2019-10-25T15:20:48 < Jan-> for main course and dessert 2019-10-25T15:20:50 < Jan-> works on everything! 2019-10-25T15:20:58 <@englishman> mayonnaise 2019-10-25T15:21:24 < Jan-> no thanks 2019-10-25T15:21:28 * Jan- can't stand mayonnaise 2019-10-25T15:21:34 < mawk> me neither 2019-10-25T15:21:43 < Jan-> I went to amsterdam in september, the only food available in the netherlands is mayonnaise. 2019-10-25T15:21:46 <@englishman> tzatziki 2019-10-25T15:21:46 < Jan-> they put it on everything 2019-10-25T15:22:32 <@englishman> HP sauce 2019-10-25T15:23:30 < karlp> kakim32: of course it has a serial port: https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;h=faf64056ddd46992a75b1e277d94541c7251035c 2019-10-25T15:26:20 < kakim32> how did you find that? 2019-10-25T15:26:51 < Ultrasauce> https://i.imgur.com/cgctRT4.png theres the setup rn 2019-10-25T15:27:28 < BrainDamage> is that a tape deck on the bottom? 2019-10-25T15:27:32 < Ultrasauce> yes 2019-10-25T15:36:00 < karlp> kakim32: IdrinkandIknowthings.gif. 2019-10-25T15:36:58 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-25T15:47:08 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-25T15:54:35 < Haohmaru> englishman i think tzatziki is sorta like a thicc-er version of our tarator 2019-10-25T16:13:17 < srk> Ultrasauce: what a track <3 2019-10-25T16:20:20 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 252 seconds] 2019-10-25T16:53:55 < karlp> Ultrasauce: https://imgflip.com/i/3ebwyi 2019-10-25T16:56:01 < Ultrasauce> more jammy stoner rock or what 2019-10-25T16:56:08 < Ultrasauce> https://albinorhinodoom.bandcamp.com/track/upholder-uncut-part-2-part-3 2019-10-25T16:57:22 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has quit [Remote host closed the connection] 2019-10-25T16:59:14 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-25T16:59:42 < karlp> is that you too? .fi? 2019-10-25T17:00:20 < Ultrasauce> lol no the only stuff i have is in that doodles directory 2019-10-25T17:00:39 < Ultrasauce> most are pretty rough 2019-10-25T17:02:28 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 264 seconds] 2019-10-25T17:02:55 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-25T17:04:49 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-25T17:05:11 < kakim32> karlp: do you know if usb3 is routed to test points on board or to anywhere? 2019-10-25T17:05:15 < kakim32> SOC has usb3 2019-10-25T17:05:19 < kakim32> I want usb3 2019-10-25T17:05:29 < Haohmaru> Ultrasauce u make musicz? 2019-10-25T17:09:06 < Ultrasauce> mostly senseless noise 2019-10-25T17:09:19 < Haohmaru> kewl 2019-10-25T17:09:28 < Haohmaru> wut genre? 2019-10-25T17:09:50 < Ultrasauce> see above 2019-10-25T17:10:06 < Haohmaru> i'll hear it when i get home 2019-10-25T17:10:11 < Ultrasauce> i mean senseless noise lol 2019-10-25T17:10:18 < Haohmaru> now now 2019-10-25T17:10:23 < Haohmaru> roughly 2019-10-25T17:10:32 < Haohmaru> eggsperimental? 2019-10-25T17:12:21 < Haohmaru> oh u got rackz n sh*t 2019-10-25T17:13:09 < Haohmaru> modular? 2019-10-25T17:14:05 < Ultrasauce> yeah a eurorack moneysink 2019-10-25T17:16:30 < Haohmaru> :/ 2019-10-25T17:22:26 < Ultrasauce> which incidentally is about 30% stm32 powered 2019-10-25T17:22:30 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Ping timeout: 265 seconds] 2019-10-25T17:23:46 < Haohmaru> i wanna get in teh biddness too 2019-10-25T17:23:51 < srk> mutable instruments or custom? :) 2019-10-25T17:24:01 < Haohmaru> not modulars but instruments 2019-10-25T17:24:03 < Ultrasauce> lots of MI 2019-10-25T17:24:36 < srk> MI is quite an inspiration 2019-10-25T17:24:48 < Ultrasauce> i do plan on doing some custom stuff but it will probably be designs straight from the 70s 2019-10-25T17:30:10 < karlp> kakim32: no idea on the usb3, maybe they're using it for the internal sockets? 2019-10-25T17:30:25 < kakim32> there just is no data 2019-10-25T17:30:36 < kakim32> I would wish to connect those lines to my LTE modem 2019-10-25T17:30:36 < karlp> what? 2019-10-25T17:30:49 < karlp> surely routerboard documents what's on the slots? 2019-10-25T17:30:57 < kakim32> yee 2019-10-25T17:31:00 < kakim32> no usb3.0 there 2019-10-25T17:31:12 < kakim32> not in this one or in 33 model 2019-10-25T17:31:14 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-25T17:33:58 < Haohmaru> Ultrasauce aka analog(ue) ? or worse.. toobz? 2019-10-25T17:37:52 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-25T17:40:59 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Quit: The Lounge - https://thelounge.github.io] 2019-10-25T17:41:36 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-25T17:46:20 < Ultrasauce> no tubes just bjts and old opamps 2019-10-25T18:02:53 -!- ekaologik [~quassel@p5DC6B253.dip0.t-ipconnect.de] has joined ##stm32 2019-10-25T18:17:54 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-25T18:25:40 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-25T18:39:21 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-25T18:56:44 < Haohmaru> noice 2019-10-25T18:57:01 < Haohmaru> i do that kind of stuff too 2019-10-25T19:04:10 -!- X230t [x13@gateway/shell/suchznc/x-ltasebypeedewdzy] has quit [Remote host closed the connection] 2019-10-25T19:07:02 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-25T19:07:20 < bitmask> man, I want to restart this whole board but I don't think its worth it 2019-10-25T19:07:35 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-25T19:08:27 < bitmask> is it bad practice to have multiple copper pours on the same layer? 2019-10-25T19:08:35 < zyp> why? 2019-10-25T19:08:59 < zyp> https://bin.jvnv.net/file/44ZNT.png <- I have a bunch in this design 2019-10-25T19:09:02 < bitmask> I dont know, I don't know how people 'usually' set one layer to a full ground plane 2019-10-25T19:09:10 < bitmask> maybe I just think thats how people do it 2019-10-25T19:09:17 < zyp> huh? 2019-10-25T19:09:26 < bitmask> ok nevermind, just a misconception I had 2019-10-25T19:09:40 < zyp> if you want a full ground plane, that's only one pour, right? 2019-10-25T19:10:14 < bitmask> right, im saying I thought that was normal, but I couldn't make that work on my design, I have 5 pours on one layer and 2 on the other 2019-10-25T19:10:23 < zyp> and all are ground? 2019-10-25T19:10:52 < bitmask> no, combo of gnd, agnd, 12V and 5V 2019-10-25T19:11:07 < zyp> when you say two pours, do you mean two actual polygons or one cut in half by traces? 2019-10-25T19:11:17 < bitmask> two actual polygons 2019-10-25T19:11:44 < zyp> you're doing a 2L board, right? 2019-10-25T19:11:46 < bitmask> yea 2019-10-25T19:12:07 < zyp> I don't usually do other pours than ground on 2L 2019-10-25T19:12:48 < zyp> the pic you saw are 12V, 5V, 3.3V and -12V on layer 3 in a 4L design 2019-10-25T19:12:56 < bitmask> I see 2019-10-25T19:13:11 < zyp> I mean 2019-10-25T19:13:34 < zyp> one of the big arguments for power pours is that they ease routing 2019-10-25T19:13:49 < zyp> but that's not the case on 2L 2019-10-25T19:14:04 < bitmask> I did have 5v on traces but I changed it to a polygon because I thought it would make more sense and lower impedance or something, i dont know enough of about that stuff 2019-10-25T19:14:23 < zyp> on 2L I usually pour ground on both sides and stitch it together 2019-10-25T19:14:47 < bitmask> how many vias to stitch it together? 2019-10-25T19:14:51 < bitmask> just curious 2019-10-25T19:14:57 < zyp> I don't usually count 2019-10-25T19:15:16 < zyp> also depends on the design, more along edges that borders RF stuff 2019-10-25T19:15:17 < bitmask> like a few or a bunch 2019-10-25T19:15:41 < zyp> https://bin.jvnv.net/file/KmWnr.png <- go ahead and count 2019-10-25T19:15:53 < bitmask> ahh I see 2019-10-25T19:16:25 < bitmask> man that looks so clean, I gotta figure this out 2019-10-25T19:18:14 * aandrew spots idiotic tag-connect interface 2019-10-25T19:19:22 < aandrew> nice layout otherwise though 2019-10-25T19:20:14 < aandrew> bitmask: I almost never ever ever ever fuck with the ground plane, but power plane gets cut up fairly often 2019-10-25T19:20:35 < aandrew> and every once in a while I'll sneak a trace onto a plane layer because there just isnt' any other way to get the damn net connected 2019-10-25T19:20:54 < aandrew> but I dn't do that on ground plane unless I have exhausted every other possibility 2019-10-25T19:21:09 < bitmask> ok 2019-10-25T19:21:41 < aandrew> 99.9% of the time you want a single solid ground. 0.1% of the time you might have a very good reason for cutting it up or separating grounds, but it's very rare 2019-10-25T19:22:02 < aandrew> and usually because you could not figure out how to keep the noisier stuff closer to the power supply and away from the sensitive stuff 2019-10-25T19:24:49 < srk> aandrew: what's wrong with tag-connect? (expect that it's kinda expensive) 2019-10-25T19:25:07 < aandrew> heh we go through this every time I notice it on someone's board 2019-10-25T19:25:48 < aandrew> it's expensive, proprietary and useless for debugging unless you use the one with the plastic retention clips at which point you're better off using a way cheaper and smaller 0.050" 10 pin header like every other sane design 2019-10-25T19:26:13 < srk> yeah :) 2019-10-25T19:26:32 < srk> or pogoprog 2019-10-25T19:27:22 <@englishman> haha, nissan leaf is combating battery overheat problems with a little PC cooling fan 2019-10-25T19:27:58 < zyp> my $50 cable is amortized over a thousand and something saved connectors by now, it's a standardized footprint anybody can buy a premade cable for and the clip works perfectly for debugging 2019-10-25T19:29:42 < aandrew> that clip does not work perfetly for debugging, it wears out and pops off/makes bad connection driving you insane 2019-10-25T19:29:48 <@englishman> I bought a noob-connect for work and hated it 2019-10-25T19:29:54 <@englishman> it sits in a drawer 2019-10-25T19:30:07 < aandrew> (if you're talking about the little piece of FR4 with three metal sockets in it) 2019-10-25T19:30:38 < aandrew> the plastic clip on the humongous tag-connect variant works great but now yo've got four gigantic fucking holes in your board where a single 2x5 header is smaller and cheaper 2019-10-25T19:31:06 < aandrew> as far as premade cables go: 2x5 0.050" cables are also premade and anyone can buy 2019-10-25T19:31:11 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-25T19:34:14 < zyp> I've yet to wear out any clips 2019-10-25T19:34:19 < zyp> lost one or two though 2019-10-25T19:37:56 < zyp> over something like six years 2019-10-25T19:40:20 <@englishman> here we actually standardized on jst-sh as msp430 only has retarded ass connectors 2019-10-25T19:40:41 <@englishman> so got an intern to make a bunch of cables 2019-10-25T19:43:34 < bitmask> if you have a mosfet leading out to a connector (n channel) would you put the mosfet on the bottom ground layer or on the top with a via, it could be switching 2-3 A 2019-10-25T19:43:45 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-25T19:44:29 < BrainDamage> what voltage? 2019-10-25T19:44:36 < bitmask> 12 2019-10-25T19:45:02 < BrainDamage> if you have a gate driver, put it high side, if you don't, put it low side 2019-10-25T19:45:20 < zyp> bitmask, do you have other components on the bottom? 2019-10-25T19:45:30 < bitmask> low side meaning bottom? 2019-10-25T19:45:34 < bitmask> no gate driver 2019-10-25T19:45:39 < bitmask> zyp I'm gonna have to 2019-10-25T19:45:40 < BrainDamage> oh sorry, misread you 2019-10-25T19:45:44 < BrainDamage> you meant physical location 2019-10-25T19:45:47 < bitmask> yes 2019-10-25T19:46:14 < zyp> if you have to have components on both sides anyway, you might as well keep it on the side that shortens the power path 2019-10-25T19:46:46 < bitmask> well I guess I dont have to, I was going to because it would keep the board smaller, i'll have to think about that 2019-10-25T19:47:01 < zyp> but it's not a big issue, vias have good conductivity 2019-10-25T19:47:25 < zyp> if all other components were on the top, I'd put the transistor also on the top 2019-10-25T19:47:47 < bitmask> I was gonna put the two smps on the bottom and everything else on top 2019-10-25T19:48:09 < zyp> smps as in module or a whole collection of parts? 2019-10-25T19:48:12 < bitmask> maybe bypass caps on bottom but that might be a waste 2019-10-25T19:48:17 < bitmask> collection of parts 2019-10-25T19:48:30 < zyp> because of space or what? 2019-10-25T19:48:53 < bitmask> what, why im putting them on the bottom? 2019-10-25T19:48:55 < bitmask> yea 2019-10-25T19:49:06 < bitmask> or why im not using a module 2019-10-25T19:49:24 < zyp> in my experience, 2L boards don't really benefit that much from putting components on both sides, since you often need to route on the bottom under components on the top 2019-10-25T19:49:44 < bitmask> ok, maybe i'll suck it up and just increase the board size a bit 2019-10-25T19:50:41 < zyp> why have you decided on a particular board size before the design is done? 2019-10-25T19:51:51 < zyp> unless I'm targetting a particular enclosure or something, I generally start by laying out the design and when I see how big it is I decide on the board size 2019-10-25T19:52:39 < bitmask> well the smallest size is determined by the 8 xt30/mr30 connectors, 4 on top 4 on left, I was trying to fit it basically inside that 2019-10-25T19:52:49 < bitmask> but I can make it a bit wider 2019-10-25T19:53:04 < mitrax> gaaah i bought a genuine FTDI USB to 3V RS232C adapter cause the chinese clone i had wasn't working reliably and was causing BSOD... and the fucker is dead on arrival... won't even show up in device manager, great! 2019-10-25T19:54:06 < zyp> uh, if a USB device is causing BSOD, that sounds like a host problem, not a device problem 2019-10-25T19:54:23 < bitmask> yea thats usually a driver issue 2019-10-25T19:54:44 < mitrax> zyp: yeah but it probably cause the driver not to act properly because it's not a genuine chip, the crash is due to the driver of course 2019-10-25T19:55:26 < bitmask> didn't ftdi reverse the clone breaking update after backlash? I guess not 2019-10-25T19:55:31 < mitrax> zyp: the latest up-to-date driver that is WHQL certified recognize the chip as counterfeit so i ca'nt use it 2019-10-25T19:55:41 < zyp> fun 2019-10-25T19:56:15 < mitrax> bitmask: yeah but i was running an old driver with the clone because of that 2019-10-25T19:56:49 < kakim32> https://en.wikipedia.org/wiki/QUIC 2019-10-25T20:07:04 -!- Jybz [~jibz@91-169-131-97.subs.proxad.net] has joined ##stm32 2019-10-25T20:12:15 < bitmask> alright, deleted everything I did in the past 2-3 days :P time to start over 2019-10-25T20:16:08 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-25T20:19:48 < bitmask> via(s) for 3.5A , any help on size? 2019-10-25T20:20:30 < karlp> just imagine it as track. 2019-10-25T20:21:01 < karlp> (it can carry more than track for the same space, because it's got more surface area) 2019-10-25T20:21:50 < bitmask> is it better to use a big via or multiple 2019-10-25T20:25:11 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 276 seconds] 2019-10-25T20:25:16 < Cracki> depends on whether the wall of the via gets thicker too 2019-10-25T20:25:56 < Cracki> I'd use several normal sized ones like everyone else I see is doing it 2019-10-25T20:26:07 < bitmask> ok thanks 2019-10-25T20:27:03 < kakim32> bitmask: saturn pcb toolkit 2019-10-25T20:27:29 < bitmask> ty 2019-10-25T20:27:33 < zyp> I prefer multiple small ones over one big one 2019-10-25T20:27:58 < Cracki> http://www.saturnpcb.com/pcb_toolkit/ 2019-10-25T20:28:29 < Cracki> >consider filled vias too https://electronics.stackexchange.com/questions/345053/large-via-vs-several-small-vias-for-large-current 2019-10-25T20:28:52 < zyp> remember that surface area scales proportionally with hole size (2πr) so two 10-mil holes has as much surface area as one 20-mil hole despite the latter requiring four times the area 2019-10-25T20:29:52 < bitmask> so you would even stay at the minimum via size and just increase the count? 2019-10-25T20:30:01 < zyp> or put another way; a bigger hole wastes more area on air that doesn't carry current 2019-10-25T20:30:06 < zyp> yes 2019-10-25T20:30:49 < Cracki> wall thickness is important too. "smallest via" may go down in that parameter 2019-10-25T20:31:03 < Cracki> what he said about surface area 2019-10-25T20:31:39 < Cracki> when you take a crosssection through the via cylinder parallel to the board, you get a ring 2019-10-25T20:31:57 < bitmask> yea I know what you're talking about 2019-10-25T20:32:21 < Cracki> otoh once you have enough vias to match the area of your trace, you're about done 2019-10-25T20:32:35 < Cracki> cross sectional area* 2019-10-25T20:33:51 < Cracki> they can carry heat tho, so more is better if you have to put a component where it can't have a heatsink or immediate copper pour 2019-10-25T20:35:11 < bitmask> I think 4 x 0.4mm will be good 2019-10-25T20:38:08 < bitmask> hmm saturn pcb shows much different numbers 2019-10-25T20:38:13 < bitmask> than the other thing I was looking at 2019-10-25T20:39:35 < bitmask> one says 0.3 = 1A, saturn says 2A, another says 1.9, another says 1 2019-10-25T20:39:40 < bitmask> so either 1 or 2 heh 2019-10-25T20:40:10 < bitmask> I think the temp rise is part of it 2019-10-25T20:40:16 < bitmask> when it has an option I choose 10C 2019-10-25T20:41:20 < bitmask> i'll call it 1.4A per 0.3mm via 2019-10-25T20:41:36 < bitmask> and do 3 for 2-3.5A 2019-10-25T20:46:37 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-25T20:49:52 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has joined ##stm32 2019-10-25T20:54:58 < kakipro> what was that first line python script needs? 2019-10-25T20:55:06 < kakipro> something about python and charset 2019-10-25T20:55:16 < kakipro> I mean.. it's been 10years 2019-10-25T20:56:33 < srk> encoding is not needed if using py3 2019-10-25T20:56:45 < srk> #!/usr/bin/env python 2019-10-25T20:56:47 < srk> or python3 2019-10-25T20:57:49 < kakipro> what does that line mean 2019-10-25T21:00:08 < Cracki> kakipro, #!/usr/bin/env python3 2019-10-25T21:00:12 < Cracki> yeah 2019-10-25T21:00:21 < Cracki> python2.7 is better than just python 2019-10-25T21:00:36 < Cracki> just python implies you dgaf about which version is running your code, so it must be compatible to 2 AND 3 2019-10-25T21:01:15 < Cracki> nobody runs anything below 2.7 anymore and no 2.x comes after 2.7, so 2.7 is fine for old scripts. 2019-10-25T21:01:28 < bitmask> https://imgur.com/a/3xLQEjO 2019-10-25T21:01:34 < bitmask> which orientation would you prefer? 2019-10-25T21:01:48 < Cracki> that line tells whatever runs executables on linux to execute /usr/bin/env with that argument and then the script file's path 2019-10-25T21:02:07 < Cracki> env does a lookup for "python3", rather than you having to hardcode where python sitz 2019-10-25T21:02:28 < Cracki> you do have to hardcode /usr/bin/env. it doesn't do shell-like search in PATH 2019-10-25T21:02:36 < Cracki> bitmask, gay 2019-10-25T21:03:30 < bitmask> funny... 2019-10-25T21:03:34 < kakipro> when I used python 2.7 was the latest or so 2019-10-25T21:04:29 < Cracki> i see lots of blue and some red 2019-10-25T21:05:00 < Cracki> either way, position shouldn't matter so long as it's the only current path? 2019-10-25T21:07:58 < bitmask> I didnt know if was better or worse to have the current flowing in the path of all the vias in series or if it would be better to have em spread out 2019-10-25T21:21:09 < Cracki> found an app note from ti on thermal and high current fuckery 2019-10-25T21:21:17 < Cracki> (motor drivers) 2019-10-25T21:22:07 < Cracki> also talks about thermal relief pads and how they're inappropriate for high current high heat applications 2019-10-25T21:22:09 < Cracki> http://www.ti.com/lit/an/slva959a/slva959a.pdf 2019-10-25T21:22:11 < srk> ty! 2019-10-25T21:24:07 < srk> hehe, pics from this board http://www.ti.com/tool/BOOSTXL-DRV8323RS 2019-10-25T21:24:18 < srk> gonna hook this one up soon! 2019-10-25T21:26:07 < antto> Ultrasauce, modular noises indeed 2019-10-25T21:26:35 -!- ohama [ohama@cicolina.org] has quit [Ping timeout: 265 seconds] 2019-10-25T21:28:35 -!- ohama [ohama@cicolina.org] has joined ##stm32 2019-10-25T21:34:09 < kakipro> which way you would do deadband in case of input from pid-controller: 1) deadband scaled to input(from pid) value range lower values so output stays 0 until input value has reached deadband 2) deadband not scaled to input value. at input value 1 output value jumps to deadband minimum? 2019-10-25T21:34:43 < kakipro> this in case there is meaningful deadband for some reason 2019-10-25T21:37:42 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 265 seconds] 2019-10-25T21:52:01 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-25T21:53:29 < jpa-> kakipro: i don't understand the question 2019-10-25T21:53:47 < kakipro> you have device thats output has deadband 2019-10-25T21:54:01 < kakipro> do you want to show it or hide it from pid 2019-10-25T21:54:11 < kakipro> by scaling input 2019-10-25T21:54:29 < aandrew> https://aleksandrov.pfoo.org/files/СССР/56b6471ea998933a19904d04a45.jpg 2019-10-25T21:54:31 < aandrew> that's for dongs 2019-10-25T21:54:44 < kakipro> so that input value 1 will have non-zero output 2019-10-25T21:55:20 < jpa-> kakipro: why are you applying deadband on input side? 2019-10-25T21:55:40 < kakipro> input FROM pid 2019-10-25T21:56:20 < jpa-> so what's the purpose of the deadband? 2019-10-25T21:56:21 < kakipro> output of the pid goes to this "input" 2019-10-25T21:57:03 < aandrew> kakipro: I am a big fan of normalizing everything but it depends on the application 2019-10-25T21:57:51 < jpa-> https://www.researchgate.net/profile/Sankar_Padmanabhan/publication/272183391/figure/fig1/AS:294999502671876@1447344770923/a-Hard-Threshold-b-Soft-Threshold.png so are you asking about the difference between hard threshold and soft threshold functions? 2019-10-25T21:58:24 < kakipro> no 2019-10-25T21:58:31 < kakipro> wait a second 2019-10-25T22:03:37 -!- ekaologik [~quassel@p5DC6B253.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer] 2019-10-25T22:04:21 < kakipro> https://sketch.io/render/sk-41e180b5fecd0aecfaa7a37766c86fa2.jpeg 2019-10-25T22:04:41 < jpa-> isn't that the same as the graph i pasted? 2019-10-25T22:05:13 < kakipro> not exactly 2019-10-25T22:05:17 < kakipro> see the gradient 2019-10-25T22:05:28 < jpa-> ok, true 2019-10-25T22:05:39 < kakipro> it's like scaled 2019-10-25T22:06:05 < kakipro> but actually it's wrong because there is deadband it cannot start from zero 2019-10-25T22:06:06 < jpa-> but anyway, the purpose of deadbanding PID output is usually to reduce how often the actuator has to move; only the left graph makes sense for that 2019-10-25T22:06:19 < kakipro> i'm not touching the pid 2019-10-25T22:06:33 < kakipro> this comes after the pid 2019-10-25T22:06:38 < jpa-> you are deadbanding PID output, even though you want to call it "input from PID" 2019-10-25T22:06:49 < kakipro> function that interfaces between pid output and actuation 2019-10-25T22:06:54 < jpa-> it's the same thing, whether you do it as the last step in PID controller code or first step in actuator code 2019-10-25T22:07:16 < kakipro> there is minimum actuation 2019-10-25T22:07:24 < kakipro> that is the deadband 2019-10-25T22:07:44 < kakipro> I just wonder how I want convert pid output 2019-10-25T22:07:47 < kakipro> to actuation 2019-10-25T22:08:09 < jpa-> the graph on the right would start actuation at zero, so that wouldn't satisfy your minimum actuation 2019-10-25T22:08:24 < kakipro> yes 2019-10-25T22:08:33 < kakipro> it's wrong I make another one that is correct 2019-10-25T22:09:42 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-25T22:10:20 < kakipro> https://sketch.io/render/sk-2ae812b40910ed1f10746c5be540e0bd.jpeg 2019-10-25T22:10:21 < kakipro> now 2019-10-25T22:10:57 < jpa-> so the graph on the right would make the actuator endlessly switch between +-1 on tiny changes in PID controller output 2019-10-25T22:11:25 < kakipro> and graph on left would cause hunting? 2019-10-25T22:11:36 < jpa-> hunting? 2019-10-25T22:12:18 < kakipro> overshoot undershoot overshoot 2019-10-25T22:12:47 < jpa-> all PID controllers constantly overshoot undershoot overshoot 2019-10-25T22:13:06 < kakipro> so which one is easiest to tune for? 2019-10-25T22:13:12 < kakipro> left one? 2019-10-25T22:13:19 < jpa-> you can only change the amplitude of that oscillation and how fast it decays, but it will never decay to zero 2019-10-25T22:13:34 < kakipro> yes 2019-10-25T22:13:42 < jpa-> the left one is how it's usually done, at least - the right one looks like it would wear out the actuator fast 2019-10-25T22:14:56 < jpa-> if the system is very static and you want to hit the correct point exactly, you'll need to implement something like "if +- 1 movements for N time, take average of them to determine the best position, then move +2 and -X to get to the perfect in-between position" 2019-10-25T22:15:28 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 268 seconds] 2019-10-25T22:15:49 < kakipro> left one it is 2019-10-25T22:16:42 < kakipro> this deadband is minuscule 2019-10-25T22:18:05 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-25T22:18:48 < kakipro> maybe sauna 2019-10-25T22:19:09 < kakipro> sauna jpa-? 2019-10-25T22:19:22 < jpa-> no thanks 2019-10-25T22:19:32 < jpa-> not in docking mood today 2019-10-25T22:20:23 < kakipro> you sauna often? 2019-10-25T22:20:33 < jpa-> not really 2019-10-25T22:20:35 < kakipro> traditional once a week? 2019-10-25T22:20:44 < jpa-> maybe once a month 2019-10-25T22:21:25 < kakipro> when I lived in cube I didn't sauna for month 2019-10-25T22:22:31 < kakipro> yes private sauna now> 2019-10-25T22:23:14 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has joined ##stm32 2019-10-25T22:24:47 < Steffann> will you jump in some ice cold water afterwards kakipro? 2019-10-25T22:25:10 < Steffann> and see it shrink to miniscule proportions? 2019-10-25T22:31:32 -!- Steffann is now known as Steffanx 2019-10-25T22:52:02 < kakipro> no 2019-10-25T22:52:50 < kakipro> 1) because of practicality reasons 2) cold water is cold 2019-10-25T23:02:26 < kakipro> actually mainly 2 2019-10-25T23:08:10 < Steffanx> but thats the point. 2019-10-25T23:10:25 < kakipro> I can't make python print 2019-10-25T23:10:37 < kakipro> and I don't know how to debug python 2019-10-25T23:15:16 < Steffanx> cant make python print? 2019-10-25T23:15:50 < kakipro> a list 2019-10-25T23:16:49 < kakipro> okay now it did something 2019-10-25T23:16:58 < kakipro> there is like 10 different ways to print a list 2019-10-25T23:17:02 < Steffanx> hah yes 2019-10-25T23:17:51 < kakipro> and then there is old way and new way 2019-10-25T23:17:56 < kakipro> and python 3.6 way 2019-10-25T23:18:16 < kakipro> -.- 2019-10-25T23:32:44 < Steffanx> and the kakimir way 2019-10-25T23:34:39 < kakipro> "Python Decimal doesn't support being constructed from float; it expects that you have to convert float to a string first." 2019-10-25T23:36:52 < kakipro> aha 2019-10-25T23:36:55 < kakipro> decimal is not integer 2019-10-25T23:38:43 < jadew> float => string => int? 2019-10-25T23:38:45 < jadew> lol 2019-10-25T23:38:46 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-25T23:39:21 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-25T23:39:34 < Cracki> there may be a reason 2019-10-25T23:39:39 < Cracki> tho I don't quite see it 2019-10-25T23:39:50 < kakipro> decimal is fixed decimal point 2019-10-25T23:39:55 < kakipro> integer is integer 2019-10-25T23:39:57 < Cracki> binary floats have finite decimal representation 2019-10-25T23:41:12 < Cracki> python 3.7 can do float to decimal just fine 2019-10-25T23:41:22 < Cracki> i don't know where you got that information that it wouldn't 2019-10-25T23:41:27 < Cracki> import decimal 2019-10-25T23:41:35 < Cracki> decimal.Decimal(math.pi) 2019-10-25T23:42:01 < Cracki> ah, random shit overflow page 2019-10-25T23:42:17 < kakipro> now I get the integers out 2019-10-25T23:42:35 < kakipro> but I didn't ask for CR LF 2019-10-25T23:43:14 < deltab> >>> decimal.Decimal(0.1) 2019-10-25T23:43:14 < deltab> Decimal('0.1000000000000000055511151231257827021181583404541015625') 2019-10-25T23:43:44 < Cracki> because 0.1 in ieee754 float is exactly that 2019-10-25T23:43:49 < Cracki> are you surprised? 2019-10-25T23:43:51 < Cracki> do you want rounding? 2019-10-25T23:44:04 < deltab> I'm not, but others might be 2019-10-25T23:44:28 < Cracki> wtf print a list? 2019-10-25T23:44:31 < Cracki> print(yourlist) 2019-10-25T23:44:32 < Cracki> done 2019-10-25T23:44:46 < kakipro> as integer 2019-10-25T23:44:46 < Cracki> debugging? python -m pdb yourscript.py 2019-10-25T23:44:54 < Cracki> list "as integer"?? 2019-10-25T23:45:05 < kakipro> nvm 2019-10-25T23:45:12 < kakipro> it's done 2019-10-25T23:45:20 < karlp> (pycharm is free, and is super rad) 2019-10-25T23:48:10 < kakipro> is there a way to increase precision of maths in python 2019-10-25T23:48:53 < MrMobius> pycharm is kind of shit 2019-10-25T23:48:56 < MrMobius> just my 2 cents 2019-10-25T23:49:35 < kakipro> end result is off by 1.1% of expected 2019-10-25T23:49:41 < Steffanx> Your opinion does not matter MrMobius :P 2019-10-25T23:49:51 < kakipro> I need better math 2019-10-25T23:50:09 -!- Jybz [~jibz@91-169-131-97.subs.proxad.net] has quit [Quit: Konversation terminated!] 2019-10-25T23:50:21 < Steffanx> What are you matching? 2019-10-25T23:50:28 < Steffanx> Mathing* 2019-10-25T23:51:18 < MrMobius> Steffanx, in my opinion, you are correct 2019-10-25T23:52:03 < Steffanx> Ty 2019-10-25T23:52:20 < kakipro> https://serveur.io/Thyristor.html Steffanx 2019-10-25T23:52:24 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-25T23:53:02 < kakipro> phi = radians(180) should equal output of 0 2019-10-25T23:57:09 < deltab> kakipro: not for floats, but for decimals you can set the precision 2019-10-25T23:59:46 < kakipro> precision for floats in python is? 2019-10-25T23:59:54 < kakipro> single? 2019-10-25T23:59:56 < kakipro> double? --- Day changed Sat Oct 26 2019 2019-10-26T00:00:03 < deltab> double 2019-10-26T00:00:38 < Cracki> that little bit of math already gets too imprecise for you? 2019-10-26T00:00:53 < deltab> so I don't know how you're getting 1.1% error 2019-10-26T00:01:35 < Cracki> yeah that's closed form and the "worst" of it is the sin() 2019-10-26T00:02:03 < kakipro> https://paste.ee/p/viFJl 2019-10-26T00:02:25 < Cracki> and? you round() things 2019-10-26T00:02:52 < kakipro> last print should be "0, " 2019-10-26T00:02:58 < Cracki> why tho 2019-10-26T00:03:00 < kakipro> it's "3, " 2019-10-26T00:03:07 < Cracki> you don't go from 0 to 180, but from 0 to 179 2019-10-26T00:03:11 < Cracki> and your step is 1 2019-10-26T00:03:15 < kakipro> oh 2019-10-26T00:03:17 < Cracki> the last has i = 179 2019-10-26T00:03:28 < Cracki> range(10) is 0..9 2019-10-26T00:03:28 < kakipro> indeed 2019-10-26T00:03:41 < kakipro> it workkens 2019-10-26T00:05:34 < kakipro> thanks for that uhm 2019-10-26T00:23:32 <@englishman> blade runner takes place in november 2019 2019-10-26T00:27:50 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 265 seconds] 2019-10-26T00:28:29 < kakipro> I had a nap one day 2019-10-26T00:28:59 < kakipro> when I woke up time was moving really slowly 2019-10-26T00:29:40 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-26T00:33:30 < kakipro> mind was clear, thinking was a breeze 2019-10-26T00:33:48 < kakipro> why I don't wake up like that every time? 2019-10-26T00:38:16 < karlp> MrMobius: what are you using for python instead? (or, what specifics don't you like abotu pycharm?) 2019-10-26T00:42:00 < oz4ga> It's difficult not to like PyCharm 2019-10-26T00:47:48 < Cracki> it's slugish 2019-10-26T00:48:07 < Cracki> I think it tried to crawl/index my whole projects directory 2019-10-26T00:48:14 < Cracki> (in addition to being sluggish) 2019-10-26T00:54:17 < karlp> it may have been sluggish because it was trying to index your entire projects directory 2019-10-26T00:54:24 < karlp> why were you trying to get it to do that? 2019-10-26T01:00:13 < MrMobius> karlp, the big selling point is intalling packages which doesnt work more often than not 2019-10-26T01:00:33 < MrMobius> the console window wont work a lot of times 2019-10-26T01:00:50 < MrMobius> tries to use the directory path as a filename for the interpretter and wont let you change it 2019-10-26T01:00:53 < MrMobius> the list is long 2019-10-26T01:01:49 < Cracki> maybe they're drunk on the success of their company 2019-10-26T01:02:05 < Cracki> or maybe it's grown too big to handle 2019-10-26T01:02:11 < Cracki> (the software) 2019-10-26T01:03:17 < Cracki> it's trying to be smart, while being dumb in some fundamental ways 2019-10-26T01:03:38 < kakipro> is there a way to place variables so that those become one continuous chunk? 2019-10-26T01:03:44 < kakipro> without using struct 2019-10-26T01:03:54 < Cracki> still in python? 2019-10-26T01:03:57 < kakipro> C 2019-10-26T01:04:11 < Cracki> not struct? array then 2019-10-26T01:04:23 < Cracki> or do you wanna fuck around with the linker? 2019-10-26T01:04:27 < specing> kakipro: with linker script and __attribute__? 2019-10-26T01:04:35 < kakipro> specing: indeed 2019-10-26T01:04:48 < kakipro> is there a way to make variable to trail another variable 2019-10-26T01:05:12 < specing> kakipro: why not a struct, though? 2019-10-26T01:05:37 < kakipro> I already have struct 2019-10-26T01:05:46 < kakipro> I don't want struct on struct on struct 2019-10-26T01:06:44 < kakipro> especially when variables are only kinda the same area of things 2019-10-26T01:07:09 < specing> Why, because of notation or indirection? 2019-10-26T01:07:14 < Cracki> what benefit is there to placing them after each other? 2019-10-26T01:07:19 < specing> (I suppose the indirection to a static object gets optimised out) 2019-10-26T01:10:44 < mawk> why do you want a continuous chunk kakipro ? 2019-10-26T01:10:50 < kakipro> mpu 2019-10-26T01:10:59 < mawk> you can give them all 0 alignemnt 2019-10-26T01:11:01 < mawk> should work 2019-10-26T01:11:14 < mawk> __attribute__((__align__(0))) 2019-10-26T01:11:37 < kakipro> how does that work 2019-10-26T01:12:09 < specing> kakipro: mpu? you want to give certain code less privileges? 2019-10-26T01:12:28 < kakipro> it's about aligning and sizing regions 2019-10-26T01:14:31 < Cracki> linker, define smaller regions, use __attribute__ to assign data to regions, done? 2019-10-26T01:15:24 < Cracki> linker script should also be able to affect packing and alignment 2019-10-26T01:15:58 < kakipro> how do I define regions? 2019-10-26T01:18:28 < karlp> MrMobius: Installing what, python packages? oh, I've never ever wanted that sort of thing. I've always used it with a venv for that project, but ok, interesting to hear at least :) 2019-10-26T01:20:50 -!- sterna [~Adium@c-7de2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-26T01:26:50 < kakipro> Cracki: you mean sections? 2019-10-26T01:27:03 < kakipro> then I align them in linker script and shit? 2019-10-26T01:30:06 < karlp> seriously, w.t.f. do you want to do that for. 2019-10-26T01:30:16 < karlp> what on earth is wrong with just a struct? 2019-10-26T01:31:09 < kakipro> nothing I guess 2019-10-26T01:33:55 < karlp> saves a whole heap of custom attributes and linker fuckery 2019-10-26T01:34:42 < kakipro> I concur 2019-10-26T01:40:39 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-26T01:47:03 < aandrew> wtf 2019-10-26T01:47:27 < aandrew> kakipro: I think you should change your nick back first 2019-10-26T01:47:28 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-26T01:47:38 < aandrew> kakipro: then, formulate a plan on what exactly your goals are 2019-10-26T01:47:48 < aandrew> so you can move toward them with the minimum amount of fucking around 2019-10-26T01:47:50 < kakipro> I'm coding 2019-10-26T01:48:13 < aandrew> may I suggest kakicode 2019-10-26T01:49:34 < aandrew> what are you doing that requires you to worry about alignment or memory regions? 2019-10-26T01:49:42 < kakipro> mpu 2019-10-26T01:49:42 < aandrew> get it working, THEN get it good 2019-10-26T01:49:57 < aandrew> why are you buggering with the mpu 2019-10-26T01:50:10 < karlp> he got enticed by the fact that freertos offers it as a port for cm3 2019-10-26T01:50:12 < kakipro> I was just interested if there was an attribute to make variables trail each other 2019-10-26T01:50:24 < aandrew> that's just alignment 2019-10-26T01:50:28 < aandrew> and that's linker, not mpu 2019-10-26T01:50:28 < karlp> they "just do" ifyou actually look at your map files :) 2019-10-26T01:50:32 < kakipro> I don't have any issue with MPU anymore 2019-10-26T01:50:57 < kakipro> certain variables 2019-10-26T01:51:12 < aandrew> why though 2019-10-26T01:51:45 < aandrew> are you memory constrained? or are you trying to optimize way way way way way too early in the process? 2019-10-26T01:51:59 < kakipro> I said MPU 2019-10-26T01:51:59 < aandrew> (I'm not even sure that unaligned accesses have a cost with Thumb2) 2019-10-26T01:52:08 < kakipro> there is regions 2019-10-26T01:52:15 < aandrew> MPU has *nothing* to do with where variables are laid out in memory 2019-10-26T01:52:30 < kakipro> indeed 2019-10-26T01:52:40 < kakipro> but when I use regions it has 2019-10-26T01:52:52 < aandrew> it has what 2019-10-26T01:52:55 < Cracki> it's a case of https://twitter.com/organizedthings 2019-10-26T01:53:06 < kakipro> something to do with where variables are 2019-10-26T01:53:10 < aandrew> nope 2019-10-26T01:53:11 < aandrew> not at all 2019-10-26T01:53:15 < aandrew> never ever 2019-10-26T01:53:24 < kakipro> not directly 2019-10-26T01:53:25 < Cracki> pardon me, https://twitter.com/organizedthings?lang=en 2019-10-26T01:53:29 < Cracki> *https://twitter.com/organizedthings/media 2019-10-26T01:53:36 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-26T01:53:40 < aandrew> not even indireclty 2019-10-26T01:54:04 < aandrew> mpu might enforce aligned accesses to whatever is in a region, but it has nothing to do with the variable being placed in that region 2019-10-26T01:55:10 < kakipro> it doesn't 2019-10-26T01:55:32 < Cracki> memory map: https://pbs.twimg.com/media/CCJnjm_XIAMLKen.png 2019-10-26T01:59:53 < kakipro> aandrew: I just wanted to know if there was some tricks I have not heard of before 2019-10-26T02:00:02 < kakipro> I changed the architecture of triac control 2019-10-26T02:01:06 < aandrew> you've redesigned it to be even more complicated? 2019-10-26T02:01:58 < kakipro> nope 2019-10-26T02:02:22 < kakipro> most of it can be ported to use timers now 2019-10-26T02:10:11 < kakipro> everything is calculated on half-period basis and firing uses different interrupt and not synchronization event from calculation task 2019-10-26T02:11:03 < kakipro> calculation task sets the time of that interrupt though 2019-10-26T02:13:15 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-26T02:14:26 < kakipro> for now functions use tickcounts but you could just throw values based on any 32bit timer there 2019-10-26T02:20:06 < kakipro> plan is to make emulating task next that is the conductor of the whole mess and then see how it works before involving actual timers 2019-10-26T02:33:40 -!- kakipro83 [b237a3d9@178-55-163-217.bb.dnainternet.fi] has joined ##stm32 2019-10-26T02:33:57 < kakipro83> no way power shortage destroyed my project 2019-10-26T02:34:07 -!- kakim32 [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-26T02:34:35 -!- kakipro [b237aa36@178-55-170-54.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-26T02:35:12 < mawk> lol 2019-10-26T02:35:49 < kakipro83> main source file is totaled 2019-10-26T02:38:09 < mawk> karma 2019-10-26T02:39:54 < kakipro83> just when everything was going great 2019-10-26T02:43:07 < kakipro83> I was literally writing the last function 2019-10-26T02:44:23 < kakipro83> 22 2019-10-26T02:45:06 < kakipro83> file is just a pattern of mess 2019-10-26T02:46:06 < Cracki> say git init 2019-10-26T02:46:07 < kakipro83> well.. it's been a while since playing some CSGO 2019-10-26T02:46:21 < kakipro83> Cracki: too late for that 2019-10-26T02:46:34 < Cracki> you'll say that again next time 2019-10-26T02:46:40 < kakipro83> is there any magical recover process in eclipse? 2019-10-26T02:46:45 < Cracki> lol eclipse 2019-10-26T02:47:02 < Cracki> eclipse is the opposite of backup 2019-10-26T02:47:18 < kakipro83> it's not like even saying "whoops something went wrong lemme recover your files" 2019-10-26T02:47:31 < kakipro83> it doesn't seem to even notice 2019-10-26T02:47:41 < kakipro83> wait boi 2019-10-26T02:47:46 < kakipro83> elf 2019-10-26T02:47:50 < Cracki> at most it might notice some project/metadata being shot 2019-10-26T02:47:56 < Cracki> you wanna decompile your elf? 2019-10-26T02:47:56 < kakipro83> does it has my sourcecode? 2019-10-26T02:48:07 < Cracki> if you're lucky, it has some debug info 2019-10-26T02:48:08 < mawk> if you enabled debugging yes 2019-10-26T02:48:14 < Cracki> do $ strings fucky.elf 2019-10-26T02:48:28 < kakipro83> I mean I opened elf with ozone one day 2019-10-26T02:48:32 < mawk> if you enabled debugging there is your full source 2019-10-26T02:48:37 < mawk> it's just a bit painful to reconstruct 2019-10-26T02:48:41 < mawk> send me a box of snuff by mail and I do it 2019-10-26T02:48:59 < kakipro83> and it blew the whole source to the screen as it is in eclipse 2019-10-26T02:49:27 < kakipro83> my endless babbling comments too 2019-10-26T02:50:21 < kakipro83> too bad it doesn't have any of my triac controller code 2019-10-26T02:50:27 < kakipro83> as I didn't build 2019-10-26T02:50:44 < mawk> :( 2019-10-26T02:50:47 < mawk> no swap files ? 2019-10-26T02:50:50 < mawk> look around in a terminal 2019-10-26T02:50:56 < mawk> no ~ files or .swp files ? 2019-10-26T02:51:09 < mawk> file could be hidden, look with ls -la 2019-10-26T02:53:00 < kakipro83> yeah no 2019-10-26T02:53:18 < kakipro83> guess who didn't buy UPS yesterday? 2019-10-26T02:54:06 < kakipro83> it was on my list and then I was like naah.. everything will be fine even without it 2019-10-26T02:54:22 < kakipro83> bought only the missing drives to nas server 2019-10-26T02:55:07 < mawk> but you didn't ctrl-s ? 2019-10-26T02:55:08 < mawk> ???¿¿ 2019-10-26T02:55:20 < mawk> you just write thousands of line of code and never save ? 2019-10-26T02:55:59 < kakipro83> I did 2019-10-26T02:56:20 < kakipro83> every time I pause 2019-10-26T02:56:24 < kakipro83> ctrl-s 2019-10-26T02:56:47 < mawk> so why isn't it here 2019-10-26T02:56:48 < mawk> it's strange 2019-10-26T02:57:04 < kakipro83> maybe I just pressed ctrl-s 2019-10-26T02:57:17 < kakipro83> when lights went out 2019-10-26T02:58:32 < kakipro83> nothing on lost and found 2019-10-26T02:58:43 < kakipro83> nothing on tmp 2019-10-26T02:58:50 < Cracki> some programs save non-atomically 2019-10-26T02:59:09 < Cracki> still, file systems today use journals, so it's a halfway decent state 2019-10-26T02:59:38 < Cracki> that should prevent empty files just because you wanted to save something 2019-10-26T03:00:28 < Cracki> maybe try windows. it's pro 2019-10-26T03:00:38 < kakipro83> is there any tools to like do some digital forensics? 2019-10-26T03:00:45 < kakipro83> to ext4 2019-10-26T03:00:47 < mawk> photorec 2019-10-26T03:00:55 < mawk> but ext4 is journaled it can't lose your files 2019-10-26T03:00:59 < mawk> it's just eclipse that lost your files 2019-10-26T03:05:04 < kakipro83> ext4 doesn't snapshoot though? 2019-10-26T03:05:40 < mawk> no, but what for ? 2019-10-26T03:08:13 < kakipro83> idk 2019-10-26T03:09:39 < karlp> mawk: that's just it, journalling does nothing to prevent you from apps misbehaving 2019-10-26T03:09:50 < karlp> I've had a file trashed by power failure before too. 2019-10-26T03:11:10 < kakipro83> major? 2019-10-26T03:16:55 < karlp> eh, it wasn't much fun 2019-10-26T03:16:58 < karlp> but this is what git's for 2019-10-26T03:38:08 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-26T03:38:13 < bitmask> ahoy 2019-10-26T04:49:15 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 240 seconds] 2019-10-26T04:54:07 -!- kakipro83 [b237a3d9@178-55-163-217.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-26T06:18:15 -!- fc5dc9d4 [~quassel@p5B081B56.dip0.t-ipconnect.de] has joined ##stm32 2019-10-26T06:22:38 -!- fc5dc9d4_ [~quassel@p5B0811B0.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-26T06:55:32 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-26T07:04:16 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 264 seconds] 2019-10-26T07:04:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-26T07:28:07 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-26T07:30:57 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-26T07:30:57 -!- day__ is now known as day 2019-10-26T07:43:08 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-26T07:43:17 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-26T07:43:48 -!- brdb [~basdb@2601:18c:8500:7f5b::9bb] has quit [Ping timeout: 246 seconds] 2019-10-26T07:52:16 -!- fc5dc9d4_ [~quassel@p5B08158A.dip0.t-ipconnect.de] has joined ##stm32 2019-10-26T07:56:47 -!- fc5dc9d4 [~quassel@p5B081B56.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds] 2019-10-26T08:17:00 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-26T08:20:32 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 265 seconds] 2019-10-26T08:20:32 -!- day__ is now known as day 2019-10-26T08:39:16 < jpa-> eh, freecancer support 2019-10-26T08:41:21 < jpa-> they won't let me withdraw to business account directly => "You have to complete business verification" => "Where is the link, is it possible with personal account?" => ten emails of back-and-forth with screenshots => "Business verification is not possible with personal account".. 2019-10-26T08:41:32 < jpa-> well, paypal it is 2019-10-26T08:41:50 < jpa-> if i ever get a tax review, this is sure gonna look like money laundering :P 2019-10-26T08:42:43 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-26T08:51:41 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-26T09:01:04 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-26T09:11:59 -!- X230t [x13@gateway/shell/suchznc/x-pfxydcgfgwkgfcso] has joined ##stm32 2019-10-26T09:26:34 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-26T09:36:54 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-26T10:31:06 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds] 2019-10-26T10:31:14 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-26T11:09:06 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d167:7a13:23d7:efc0] has joined ##stm32 2019-10-26T11:15:07 -!- sterna [~Adium@c-b7e0e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-26T11:41:57 < antto> Ultrasauce, http://antonsavov.net/tmp/z_n0nf_ladder_wubwub-fart.mp3 2019-10-26T11:41:59 < antto> >:) 2019-10-26T11:44:57 < Steffanx> Ty for your morning farts antto 2019-10-26T12:00:05 < jadew> found this on huffingtonpost's website (NSFW): http://images.huffingtonpost.com/2014-07-19-care2.com.jpg 2019-10-26T12:00:07 < jadew> in an article 2019-10-26T12:01:25 < Steffanx> Ok. 2019-10-26T12:01:37 < jadew> is it not unusual? 2019-10-26T12:10:14 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-26T12:11:42 < Steffanx> Idk I dont read the Huffington post 2019-10-26T12:12:02 < antto> how bout the puffington host? 2019-10-26T12:12:41 < Steffanx> That's you this morning 2019-10-26T12:27:32 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-26T12:47:57 -!- sterna [~Adium@c-b7e0e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-26T12:57:08 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-26T14:08:27 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Ping timeout: 240 seconds] 2019-10-26T14:10:34 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-26T14:10:34 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-26T14:10:34 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-26T14:22:04 -!- sterna [~Adium@h-235-5.A357.priv.bahnhof.se] has joined ##stm32 2019-10-26T14:27:40 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 264 seconds] 2019-10-26T14:31:39 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-26T15:09:02 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d167:7a13:23d7:efc0] has quit [Ping timeout: 276 seconds] 2019-10-26T15:13:07 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-26T15:18:53 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 264 seconds] 2019-10-26T15:40:06 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-26T15:43:34 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-26T15:47:08 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-26T16:02:20 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 276 seconds] 2019-10-26T16:08:05 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 264 seconds] 2019-10-26T16:19:14 -!- doomba [~npc@slipgate.logbook.pw] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-26T16:30:19 -!- roomba [~npc@slipgate.logbook.pw] has joined ##stm32 2019-10-26T16:30:36 -!- roomba is now known as doomba 2019-10-26T16:37:07 -!- catphish [~catphish@unaffiliated/catphish] has quit [Read error: Connection reset by peer] 2019-10-26T16:37:24 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-26T16:58:25 -!- doomba [~npc@slipgate.logbook.pw] has quit [Quit: ZNC 1.7.4 - https://znc.in] 2019-10-26T17:11:04 -!- roomba [~npc@slipgate.logbook.pw] has joined ##stm32 2019-10-26T17:11:46 -!- roomba is now known as doomba 2019-10-26T17:17:15 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-26T17:30:47 -!- kakim32 [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-26T17:47:39 < kakim32> I have innovation 2019-10-26T17:51:26 < kakim32> modular UPS 2019-10-26T17:51:46 < Steffanx> i dont even understand how you mananged to lose all your shit by an simple power outage kakim32 2019-10-26T17:52:02 < kakim32> just the main file 2019-10-26T17:52:43 < Steffanx> even that 2019-10-26T17:52:58 < kakim32> it's fate 2019-10-26T17:53:00 < Steffanx> Typical kakievent it seems 2019-10-26T17:53:42 < kakim32> project resists at every turn 2019-10-26T17:54:13 < kakim32> I can't help to not take it as a signal eventually 2019-10-26T17:54:15 < Steffanx> will you do backups/version control now? 2019-10-26T17:54:31 < Steffanx> you even have a nas, instill gitlab on there. its even a freenas plugin 2019-10-26T17:54:36 < Steffanx> *install 2019-10-26T17:54:50 < kakim32> bought the missing drives but not yet installed 2019-10-26T17:55:09 < kakim32> but it's an idea though 2019-10-26T17:58:48 < aandrew> I use syncthing to share my projects folder with like 4 different systems, and it's also backed up with incrementals to s3 2019-10-26T17:59:09 < aandrew> not to mention gratuitous use of git 2019-10-26T17:59:27 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-26T18:22:32 < bitmask> https://cdn.canoodlesoup.com/wp-content/uploads/2019/06/hzUHeWQ.jpg 2019-10-26T18:22:46 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Remote host closed the connection] 2019-10-26T18:29:13 < aandrew> bitmask: looks like a fun friday night 2019-10-26T18:29:19 < bitmask> yup 2019-10-26T18:31:10 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-26T19:21:36 < Jan-> tch 2019-10-26T19:21:39 < Jan-> can't even go to the gym 2019-10-26T19:21:42 < Jan-> it's too rainy :( 2019-10-26T19:21:50 < aandrew> you gonna melt? 2019-10-26T19:21:57 < Jan-> me? 2019-10-26T19:22:01 < aandrew> yes 2019-10-26T19:22:05 < Jan-> no. 2019-10-26T19:22:09 < Jan-> Because I'm not going out in it :D 2019-10-26T19:22:13 < aandrew> lol 2019-10-26T19:22:14 * Jan- doesn't like this time of year :( 2019-10-26T19:22:26 < aandrew> autumn is my favourite time of year in Canada 2019-10-26T19:22:44 < aandrew> leaves changing colour, smell of people firing up wood stoves/fireplaces, general feeling of hunkering down and eating hearty foods 2019-10-26T19:22:46 < Jan-> last time I was in canada it was april and it was freezing 2019-10-26T19:22:57 < aandrew> depending on the year that can happen 2019-10-26T19:23:08 < Jan-> canada is great 2019-10-26T19:23:09 < aandrew> sometimes april is lovely, other times it's a wet, slushy, cold affair 2019-10-26T19:23:15 < Jan-> toronto is a nice town, everyone is nice 2019-10-26T19:23:16 < Jan-> but the weather 2019-10-26T19:23:21 < Jan-> cannot do the weather 2019-10-26T19:23:25 < aandrew> toronto is awful 2019-10-26T19:23:32 < aandrew> mind you I'm not a city person, never have been 2019-10-26T19:23:44 < Cracki> møøse 2019-10-26T19:24:58 < Cracki> does maple syrup go well with moose steak? 2019-10-26T19:25:25 < aandrew> all steak should have on it is some salt and pepper and maybe a little oil 2019-10-26T19:25:35 < aandrew> good steaks don't need marinade or special prep 2019-10-26T19:25:49 < Cracki> ketchup, like the king imperator eats it 2019-10-26T19:26:16 < Cracki> *god imperator of course 2019-10-26T19:26:48 < Cracki> ye meat has its own taste, a shame to overpower it with anything 2019-10-26T19:27:21 -!- irf21k [~irf21k@106.206.110.179] has joined ##stm32 2019-10-26T19:34:46 < kakim32> let's talk about my innovation 2019-10-26T19:34:55 < kakim32> network connected mini-UPS 2019-10-26T19:35:15 < kakim32> modular 2019-10-26T19:35:54 < Steffanx> "true" sine wave stuff? 2019-10-26T19:35:58 < kakim32> DC 2019-10-26T19:36:06 < Steffanx> oh meh 2019-10-26T19:36:14 < Steffanx> is it for your rpi? 2019-10-26T19:36:32 < kakim32> it could have 5v output module 2019-10-26T19:37:28 < kakim32> it could take power from POE or externally applied DC 2019-10-26T19:42:39 < Ultrasauce> what made this antto? 2019-10-26T19:43:02 < Ultrasauce> really like that filter when the resonance is up 2019-10-26T19:43:41 < antto> it's been one of my cascaded 4-pole filters 2019-10-26T19:44:02 < Jan-> more Ultrasauce! 2019-10-26T19:44:04 < antto> woz messing around with LFO when it started sounding funny so i've recorded it 2019-10-26T19:44:05 * Jan- loves that name 2019-10-26T19:44:34 < antto> ehm, it's a digital filter, of course 2019-10-26T19:52:57 < steve> antto that sounds cool, is that running on a stm32? 2019-10-26T19:53:04 < antto> no 2019-10-26T19:53:07 < antto> comput0r 2019-10-26T19:53:45 < antto> i wish i could run it on an arm corteggz 2019-10-26T19:53:47 < antto> one day 2019-10-26T19:54:36 -!- irf21k [~irf21k@106.206.110.179] has quit [Remote host closed the connection] 2019-10-26T19:56:44 < steve> i'm playing with the stm32 now. re-learned a bunch of DSP, ie recalculating filter coefficeints at runtime 2019-10-26T19:57:58 < steve> i heard you can port VSTs to run on embedded targets, that's on my list 2019-10-26T20:00:57 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-26T20:05:28 < antto> Ultrasauce, https://soundcloud.com/antto/digitalfilter 2019-10-26T20:07:02 < antto> iirc someone at kvr had posted a video of some modular autism guy playing with a filter with a similar sequence, and the guy from kvr was saying how you can't get such sound with comput0rz.. cuz they too digital.. so i made this ^ 2019-10-26T20:08:08 < Steffanx> thats some much better farts antto 2019-10-26T20:21:21 < Ultrasauce> lol more like well tuned fm on analog is impossible 2019-10-26T20:22:25 < antto> i don't think the question was about the FM, probably more about the resonance or envelopes or modulation.. 2019-10-26T20:46:31 < kakim32> are usb2.0 lines used in usb3.0? 2019-10-26T20:46:49 < kakim32> in that type A connector 2019-10-26T20:50:33 -!- boB_K7IQ [~boB_K7IQ@c-73-140-168-208.hsd1.wa.comcast.net] has joined ##stm32 2019-10-26T20:55:22 <@englishman> hmm esp32 has a can peripheral that isn't in the datasheet 2019-10-26T20:56:22 < fenugrec> englishman, my theory is too many embarassing errata 2019-10-26T20:57:55 <@englishman> fenugrec: https://i.redd.it/k7nvwyp18lt31.jpg 2019-10-26T20:58:24 <@englishman> the periph has code support and lots of documentation with examples but it just isn't in the datasheet/rm 2019-10-26T20:58:35 <@englishman> maybe it's exclusively in the errata sheet :) 2019-10-26T20:59:23 < fenugrec> HAH ! 2019-10-26T21:00:08 < fenugrec> englishman, yeah we were looking at the esp for some can-enabled BS and it got disqualified when we noticed the "missing chapter" in the RM 2019-10-26T21:01:37 <@englishman> nice 2019-10-26T21:01:43 <@englishman> what did you go with 2019-10-26T21:02:59 < kakim32> https://techcrunch.com/2019/10/21/nordvpn-confirms-it-was-hacked/ 2019-10-26T21:03:35 < kakim32> luckily I didn't or don't use finnish server 2019-10-26T21:04:04 < kakim32> I use steffland 2019-10-26T21:05:32 < fenugrec> englishman, stm32 of course 2019-10-26T21:06:37 <@englishman> are there any MCUs with multiple can peripherals 2019-10-26T21:07:13 < fenugrec> yea, G4 I think ^ 2019-10-26T21:07:34 <@englishman> nice 2019-10-26T21:07:35 < fenugrec> not sure 2019-10-26T21:07:44 <@englishman> are you going to the g4 seminar tuesdasy 2019-10-26T21:08:12 <@englishman> ooo there are soe with 3 2019-10-26T21:08:38 < fenugrec> is there a g4 semenar 2019-10-26T21:08:56 <@englishman> yes in mtl 2019-10-26T21:09:13 < fenugrec> lucky us, I'm not going, you ? 2019-10-26T21:09:24 <@englishman> yes and work paid 2019-10-26T21:09:27 -!- onio [~onio@2a00:23c5:7a01:8600:5ac:3ace:b590:c891] has joined ##stm32 2019-10-26T21:09:28 < fenugrec> damn ! 2019-10-26T21:11:48 <@englishman> subscribing to st spam pays 2019-10-26T21:14:43 <@englishman> oh this is weird 2019-10-26T21:14:46 <@englishman> there are zero g4 with can 2019-10-26T21:15:14 <@englishman> g0 neither 2019-10-26T21:17:53 <@englishman> 3 can is perfect, stm32f413 2019-10-26T21:20:11 < fenugrec> I got lost in their selection chart, I thought the F4/G4 were the kitchen sink editions. guess not 2019-10-26T21:20:34 < fenugrec> why do you need 3 can ? gateway ? 2019-10-26T21:21:05 < kakim32> xi endorsed blockchain 2019-10-26T21:21:31 < kakim32> buttcoin spiked 2019-10-26T21:39:10 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2019-10-26T21:53:16 -!- boB_K7IQ [~boB_K7IQ@c-73-140-168-208.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-26T21:53:33 -!- boB_K7IQ [~boB_K7IQ@c-73-140-168-208.hsd1.wa.comcast.net] has joined ##stm32 2019-10-26T22:00:37 < bitmask> how much space should you leave between a mcu pin and a passive?, 1mm? 2019-10-26T22:01:31 < bitmask> everyone says to put the bypass caps as close as possible but you need at least a little space 2019-10-26T22:02:31 < antto> in kicad, footprints have a courtyard outline 2019-10-26T22:03:33 < antto> i leave some space on TQFPs since i hand solder them 2019-10-26T22:03:56 < bitmask> what is some space 2019-10-26T22:04:17 < antto> that's enough so that you can go in there with teh iron 2019-10-26T22:04:31 < antto> i haven't measured it, i eyeball it in the 3D viewer 2019-10-26T22:04:32 < bitmask> which is how much 2019-10-26T22:04:35 < bitmask> ok 2019-10-26T22:04:48 < antto> lemme see if i got some pic here 2019-10-26T22:05:40 < antto> perhaps-ish, sorta like this: https://i.imgur.com/UYy40Y7.png 2019-10-26T22:06:14 < mawk> I got my contract Steffanx 2019-10-26T22:06:17 < mawk> I'm officially a dutch worker now 2019-10-26T22:06:28 < antto> i drag solder with a normal sharp iron tip 2019-10-26T22:06:36 < antto> so i put the tip sidewayz ;P~ 2019-10-26T22:06:39 < bitmask> is that 2.54 spacing? 2019-10-26T22:07:04 < antto> no, i work on a 0.5mm grid, and go down if i need to 2019-10-26T22:07:27 < bitmask> i meant for those lines of throughole pads 2019-10-26T22:07:36 < antto> yeah, that's DIP-40 2019-10-26T22:07:41 < bitmask> k thanks 2019-10-26T22:07:48 < antto> the passives are 0603 2019-10-26T22:07:53 < antto> i fink 2019-10-26T22:08:32 < machinehum> ttps://www.digikey.ca/product-detail/en/stmicroelectronics/STM32G031J6M6/497-19552-ND/10300265 2019-10-26T22:08:35 < machinehum> Cool part 2019-10-26T22:09:27 < bitmask> so its about 1.27mm away 2019-10-26T22:10:03 < bitmask> half the distance in pixels as the pin spacing 2019-10-26T22:10:25 < bitmask> alright i'll aim for that, I think as long as its less than 4-5mm its fine 2019-10-26T22:11:35 < Steffanx> ohno, mawk 2019-10-26T22:11:37 < Steffanx> :P 2019-10-26T22:11:39 < Steffanx> congratz 2019-10-26T22:12:31 < Steffanx> when will you start, mawk? 2019-10-26T22:12:37 < mawk> half of november I think 2019-10-26T22:13:20 < Steffanx> i thnk :P 2019-10-26T22:13:25 < Steffanx> isnt that in your contract? 2019-10-26T22:14:47 < mawk> no, it's the blank contract 2019-10-26T22:14:51 < mawk> they will fill up the date after 2019-10-26T22:14:54 < Steffanx> aha 2019-10-26T22:14:55 < mawk> when I know when I have a house 2019-10-26T22:16:09 < mawk> there are strange clauses in the contract 2019-10-26T22:16:14 < Steffanx> did you make your gf find a job too? 2019-10-26T22:16:27 < mawk> she will be a housewife 2019-10-26T22:16:29 < mawk> that's the job 2019-10-26T22:16:30 < mawk> lol 2019-10-26T22:16:43 < mawk> I try to get her interested in programming but it doesn't work 2019-10-26T22:16:50 < Steffanx> kids are on the way i hear? :P 2019-10-26T22:16:57 < mawk> I hope 2019-10-26T22:17:58 < Steffanx> cant be that hard for her to find a semi-random job. 2019-10-26T22:18:36 < jpa-> can always work as a waitress at huffie-duffie cafe 2019-10-26T22:19:13 < Steffanx> yeah 2019-10-26T22:19:57 < mawk> she doesn't want to be near human beings 2019-10-26T22:20:00 < mawk> except me of course 2019-10-26T22:21:42 < Steffanx> haha, great. 2019-10-26T22:21:56 < specing> really mawk 2019-10-26T22:21:59 < specing> where did you find her 2019-10-26T22:22:07 < mawk> at a techno party 2019-10-26T22:22:10 < specing> what is her education? 2019-10-26T22:22:23 < mawk> just highschool degree 2019-10-26T22:22:27 < specing> ... 2019-10-26T22:22:30 < mawk> and social educator too 2019-10-26T22:22:37 < mawk> but she didn't finish that diploma 2019-10-26T22:22:39 < jpa-> in reality, mawk just doesn't let her out so she doesn't run away with Steffanx 2019-10-26T22:22:42 < mawk> she worked at french red cross before 2019-10-26T22:22:44 < mawk> lol 2019-10-26T22:22:45 < mawk> :( 2019-10-26T22:23:04 < Steffanx> Nah shes safe. i dont go to the place where mawk will go. 2019-10-26T22:23:06 < Steffanx> probably not ever 2019-10-26T22:23:13 < mawk> :( 2019-10-26T22:23:16 < mawk> delft is the beautifulest city 2019-10-26T22:23:26 < Steffanx> the entire region is way to busy 2019-10-26T22:23:30 < Steffanx> traffic jams et all 2019-10-26T22:23:56 < mawk> there is the train ! 2019-10-26T22:24:19 -!- boB_K7IQ [~boB_K7IQ@c-73-140-168-208.hsd1.wa.comcast.net] has quit [Ping timeout: 250 seconds] 2019-10-26T22:24:39 < Steffanx> trains are busy too 2019-10-26T22:24:46 < Steffanx> if they go 2019-10-26T22:25:47 < BrainDamage> both are pretty weird profession choices for someone who doesn't want to socialize 2019-10-26T22:25:47 < Steffanx> better get a bicycle 2019-10-26T22:27:47 < Steffanx> dont get it stolen mawk 2019-10-26T22:27:52 < Steffanx> i had two bikes, both are gone 2019-10-26T22:28:13 < mawk> I will put a big chain 2019-10-26T22:30:41 < BrainDamage> get a u-lock and a regular chain 2019-10-26T22:30:43 < BrainDamage> https://cdn.shopify.com/s/files/1/0414/4777/products/ulock-holster-rear-panier-striped-pants-taking-out-lock_2048x.jpg?v=1527661540 2019-10-26T22:31:16 < BrainDamage> you need 2 different tools to break them, making it extremely unlikely for a common thief to have both 2019-10-26T22:31:40 < BrainDamage> generally they carry large bolt cutters 2019-10-26T22:31:54 < BrainDamage> which work on the chain, but not on the u-lock 2019-10-26T22:32:39 < BrainDamage> ( you need a jack to break a u-lock efficiently ) 2019-10-26T22:32:58 < Steffanx> on the other hand, you can buy a new one for like 10 euro from a random junky at the corner of every street :P 2019-10-26T22:33:15 < mawk> lol 2019-10-26T22:33:19 < BrainDamage> you can buy back your bike* 2019-10-26T22:33:34 < specing> I wish the police would do more to combat bike theft 2019-10-26T22:33:48 < specing> but these days the only thing they can do is issue fines 2019-10-26T22:33:55 < specing> and the junkies/homeless don't care about those 2019-10-26T22:34:11 < Steffanx> yours or any other BrainDamage 2019-10-26T22:35:16 < BrainDamage> I never got mine stolen 2019-10-26T22:35:29 < Steffanx> yet 2019-10-26T22:35:33 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 265 seconds] 2019-10-26T22:35:41 < BrainDamage> my sister had 2019-10-26T22:35:48 < BrainDamage> got it back 2019-10-26T22:36:54 < BrainDamage> just made a nice and friendly visit to the local delinquents 2019-10-26T22:37:17 < BrainDamage> they even made things easier to stand by it in the garage with the door open 2019-10-26T22:37:37 < Steffanx> haha 2019-10-26T22:42:04 <@englishman> nice mawk congrats on not being poor anymore 2019-10-26T22:42:45 <@englishman> lol a linux bike 2019-10-26T22:46:26 < Steffanx> mawk: dont forget i come from groningen, which is a like "friesland" . for dutchland standards pretty uncrowded 2019-10-26T22:52:09 <@englishman> https://www.sifive.com/ 2019-10-26T22:56:48 < qyx> Steffanx: trains meh 2019-10-26T22:57:29 < qyx> train fires are becoming quite common here 2019-10-26T22:57:40 <@englishman> 🚂🚃🚃🚃🚃 2019-10-26T22:57:51 < qyx> I mean locomotive fires 2019-10-26T22:57:55 < Steffanx> in dutchland we're way past the steam engine :P 2019-10-26T22:58:00 < qyx> if the are few decades old 2019-10-26T22:58:10 <@englishman> 🚈🚃🚃🚃🚃 2019-10-26T22:58:12 < qyx> but we have a bunch of new siemens vectrons! 2019-10-26T22:58:26 < Steffanx> in dutchland they get "square" wheel due to leafs on the track 2019-10-26T22:58:29 < qyx> they are stopping randomly 2019-10-26T22:58:40 < Steffanx> and they stop going when the first snow is coming down 2019-10-26T22:58:50 < Steffanx> *wheels 2019-10-26T22:58:51 <@englishman> bullet train and high-speed train are two different emojis both introduced in unicode 6.0 2019-10-26T22:59:19 < qyx> usually show is not a problem here 2019-10-26T22:59:24 < qyx> except for trolley buses 2019-10-26T22:59:33 <@englishman> 🏂🏿 2019-10-26T22:59:43 < Steffanx> it shouldnt be a problem here too, but yet it is 2019-10-26T22:59:56 < Steffanx> full panic mode 2019-10-26T23:00:18 < qyx> heh, in the capital we have trolley buses going to the hills mostly 2019-10-26T23:00:24 < qyx> because less smoke, better aceleration, etc 2019-10-26T23:00:42 < qyx> and togeter with all-season tires 2019-10-26T23:00:51 < qyx> and freezing trolley cables 2019-10-26T23:01:03 < qyx> they are out of service when the winter comes 2019-10-26T23:01:27 <@englishman> they tried electric busses here about 10y ago 2019-10-26T23:01:35 <@englishman> when it got cold they didnt have enough power to get up the hills 2019-10-26T23:01:49 < qyx> we have some hybrid buses 2019-10-26T23:01:50 <@englishman> so people had to get off and walk up next to the bus 2019-10-26T23:01:59 < qyx> 100kW diesel generator with electric engine 2019-10-26T23:02:06 < qyx> they are ok 2019-10-26T23:02:09 < qyx> until they are full 2019-10-26T23:02:20 <@englishman> theres some all-electric school busses here with diesel heaters 2019-10-26T23:03:05 < qyx> so the 240kW electric could climb those 20% slopes easily 2019-10-26T23:03:11 < qyx> if it had enough power.. 2019-10-26T23:07:31 -!- aafsass [3e412fc0@192-47-65-62.dyn.estpak.ee] has joined ##stm32 2019-10-26T23:15:59 -!- aafsass [3e412fc0@192-47-65-62.dyn.estpak.ee] has quit [Remote host closed the connection] 2019-10-26T23:35:33 < specing> englishman: where is here? 2019-10-26T23:43:56 < Steffanx> england ofcourse. 2019-10-26T23:46:31 < specing> Steffanx: round here they are selling methane busses as the most eco around 2019-10-26T23:46:55 < specing> while Shenzhen has a fleet of 16000 BEV busses in their fleet... --- Day changed Sun Oct 27 2019 2019-10-27T00:27:50 -!- sterna [~Adium@h-235-5.A357.priv.bahnhof.se] has quit [Quit: Leaving.] 2019-10-27T00:30:53 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 264 seconds] 2019-10-27T00:55:24 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-27T00:56:08 -!- con3 [~kvirc@154.119.40.237] has quit [Client Quit] 2019-10-27T00:56:56 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-27T01:00:20 < sync> specing: gas is actually not too bad 2019-10-27T01:02:04 < specing> sync: the point is that they could have gone all the way and gotten full BEV ones 2019-10-27T01:13:27 < Steffanx> Gas is baaaad sync. They pumped everything out of the ground here. And now the ground is going down. Into the gaps. 2019-10-27T01:13:40 < specing> lol 2019-10-27T01:14:00 < jadew> Steffanx, sounds to me like more space is created 2019-10-27T01:14:01 < Steffanx> "Massive" earthquakes 2019-10-27T01:14:12 < Steffanx> Houses getting huge cracks 2019-10-27T01:14:26 < Steffanx> Government goes: ty, and fuck you 2019-10-27T01:14:47 < jadew> were those people paid for the gas underneath their house? 2019-10-27T01:14:56 < Steffanx> (Its actually a pretty serious issue in the region I live in) 2019-10-27T01:15:01 < jadew> or is that the property of "the state"? 2019-10-27T01:15:04 < Steffanx> Not directly, no. Jadew. 2019-10-27T01:15:23 < Steffanx> I guess they see it like that. 2019-10-27T01:15:39 < Steffanx> Muh gasss 2019-10-27T01:15:49 < Ultrasauce> https://www.youtube.com/watch?v=FsPUSHBQYx the good kind of gas 2019-10-27T01:16:01 < Steffanx> Cant see the vid. 2019-10-27T01:16:11 < Steffanx> Its unavailable 2019-10-27T01:16:25 < jadew> works here, maybe your country sucks? 2019-10-27T01:16:27 < Steffanx> Hmm it does a 400 2019-10-27T01:16:41 < Ultrasauce> https://gas0095.bandcamp.com/album/gas-0095 2019-10-27T01:16:44 < jadew> I'm trolling... it doesn't work here either 2019-10-27T01:17:13 < jadew> it's a song 2019-10-27T01:17:15 < Steffanx> The better 2019-10-27T01:17:42 < jadew> I need to learn to have debates 2019-10-27T01:17:48 < jadew> anyone up for a debate club? 2019-10-27T01:18:54 < jadew> I just had a heated argument with my in-laws 2019-10-27T01:18:55 < Steffanx> Best gas lately laughing gas. Very popular "drug" here lately. 2019-10-27T01:19:03 < Steffanx> Nitrous or whatever the English name is 2019-10-27T01:19:14 < jadew> and I want to improve my delivery, I understood their point of view, but couldn't get them to consider mine 2019-10-27T01:19:34 < jadew> so... debate club anyone? 2019-10-27T01:19:35 < Steffanx> I think c racki would like it. 2019-10-27T01:19:42 < mawk> I love it Steffanx 2019-10-27T01:19:45 < mawk> I did it when I was 12 2019-10-27T01:19:59 < mawk> I bought whipped cream cans at the supermarket and inflate rubber balloons with it 2019-10-27T01:20:02 < mawk> then breath in them 2019-10-27T01:20:10 < Steffanx> -_- 2019-10-27T01:20:14 < jadew> and you start laughing? 2019-10-27T01:20:30 < jadew> what are the side effects? 2019-10-27T01:20:32 < mawk> not really, it's more like anæsthetic 2019-10-27T01:20:39 < mawk> my head hurts after 2019-10-27T01:20:59 < mawk> it lasts only like 30 seconds, it's supposed to be mixed 50/50 with oxygen and breathed continuously, it's for anæsthesia of children normally 2019-10-27T01:21:21 < Steffanx> mawk: just breathed into the balloon for so long it was mainly the lack of oxygen that made you "trip"? 2019-10-27T01:21:26 < mawk> lol 2019-10-27T01:21:32 < mawk> no I did 1 breath of N2O and 1 breath of air 2019-10-27T01:21:37 < mawk> because I was 12 but not stupid 2019-10-27T01:22:00 < mawk> at that time I was reading on organic chemistry to know how to get the most out of drugs 2019-10-27T01:22:29 < Steffanx> Lol. I never cared. And still dont 2019-10-27T01:24:20 < mawk> I was very very bored 2019-10-27T01:24:55 < Steffanx> I'm hardly ever bored. I can perfectly do "nothing" 2019-10-27T01:25:01 < jadew> mawk, I think a lot of people are breathing in paint fumes to feel good 2019-10-27T01:25:08 < mawk> they're stupid jadew 2019-10-27T01:25:18 < mawk> that's the perfect route to sudden cardiac arrest 2019-10-27T01:25:21 < jadew> it's probably cheaper than a can of whipped cream 2019-10-27T01:25:42 < mawk> paint thinner and other benzene-related chemicals are very cardiotoxic when breathed in directly 2019-10-27T01:26:02 < mawk> while nitrous oxide is a real drug used in the medical field as a very safe anæsthetic 2019-10-27T01:26:32 < jadew> so... not that's not for having a good time either? 2019-10-27T01:27:32 < mawk> you can have a good time with nitrous if you don't forget to breath 2019-10-27T01:27:48 < jadew> I think most junkies here are using heroin 2019-10-27T01:27:50 < mawk> but if you want to huff paint thinner you may as well drink some bleach to have your ultimate trip before meeting your maker 2019-10-27T01:28:13 < jadew> I haven't seen anyone looking like the meth heads in online pictures 2019-10-27T01:29:04 < jadew> definitely injectable stuff 2019-10-27T01:30:19 < jadew> I don't exactly know what the street people were using before 2019-10-27T01:30:30 < jadew> (I actually saw one today, with one of those bags) 2019-10-27T01:30:34 < jadew> haven't seen them in a while 2019-10-27T01:30:58 < Steffanx> Alcohol 2019-10-27T01:31:03 < jadew> https://static4.libertatea.ro/wp-content/uploads/2011/11/aurolac4.jpg 2019-10-27T01:31:32 < mawk> lol 2019-10-27T01:31:38 < mawk> two roms huffing paint 2019-10-27T01:33:08 -!- sterna [~Adium@c-b7e0e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-27T01:33:21 < Steffanx> So found yourself a sleeping spot in the meantime mawk? 2019-10-27T01:33:38 < Steffanx> Still aiming for somewhere in the centre? 2019-10-27T01:34:18 < mawk> yes I rather be in the centre and be poor than in the shitty Tanthof suburbs in a 10 stories building 2019-10-27T01:34:25 < mawk> I will sleep at my father's house in the meantime 2019-10-27T01:34:34 < mawk> just next to the train station in the hof van delft neighborhood 2019-10-27T01:37:03 < jadew> last time we visited Paris, I saw an ad for storage closet (3 to 5 sqm), that was going for more than 30k eur 2019-10-27T01:37:10 < mawk> lol 2019-10-27T01:37:12 < mawk> yes 2019-10-27T01:37:17 < mawk> it's above 10k€/m² here 2019-10-27T01:37:24 < mawk> so 30k for 5m² is cheap 2019-10-27T01:37:31 < jadew> crazy 2019-10-27T01:42:59 < sync> specing: you have to have the infrastructure for that tho 2019-10-27T01:43:00 < Steffanx> Whokay mawk 2019-10-27T01:43:38 < Steffanx> When you're used to paris I can imagine you want to live I the centre. Imagine it being quiet at night ;) 2019-10-27T01:43:48 < Steffanx> In* 2019-10-27T01:44:37 < jadew> you know how invoices carry an obligation to pay? 2019-10-27T01:44:45 < jadew> does that work corss-border in the EU? 2019-10-27T01:44:50 < jadew> *cross 2019-10-27T01:45:23 < mawk> yeah Steffanx I don't like suburbs where you have to bike 20 minutes to get to civilization 2019-10-27T01:45:28 < jadew> I'm wondering if it's possible to deliver goods before being paid and still benefit from some sort of protection 2019-10-27T01:45:31 < mawk> I want to be able to make quick purchases 5 minutes before the store close 2019-10-27T01:45:54 < mawk> yes signed invoice carry an obligation to pay under a fine jadew 2019-10-27T01:46:00 < mawk> in france at least, but I think it's EU-wide 2019-10-27T01:46:09 < mawk> it's 45€ fine I think, for a payment delay 2019-10-27T01:46:10 < jadew> mawk, got it, that's good to know 2019-10-27T01:46:26 < mawk> but the invoice needs to be standardized, signed, and you need to have a preliminary offer signed before it 2019-10-27T01:46:38 < mawk> like invoice offer that the other party accepted and signed 2019-10-27T01:46:44 < mawk> I don't know how it is in english 2019-10-27T01:46:48 < mawk> "devis" in french 2019-10-27T01:46:57 < jadew> I see, in here invoices travel without stamp or signature 2019-10-27T01:47:02 < jadew> makes e-trade a lot easier 2019-10-27T01:47:09 < Steffanx> I live in a relatively small village and i can do that easily, mawk. 2019-10-27T01:47:09 < mawk> I guess e-signature works too 2019-10-27T01:47:18 < Steffanx> Small as in 3.9K people 2019-10-27T01:47:40 < mawk> I will come haunt your house when I go see the few family members still alive in drachten 2019-10-27T01:47:50 < jadew> Steffanx, jeez... how do you find a girl who's not your cousin in there? 2019-10-27T01:48:24 < specing> sync: its not hard to do 2019-10-27T01:49:08 < Steffanx> jadew: I don't know how it works in romania, but we are used to move all over the country 2019-10-27T01:49:43 < jadew> Steffanx, we move too, but rarely into small towns 2019-10-27T01:49:45 < mawk> of course, NL has the size of a french city 2019-10-27T01:50:11 < jadew> so the people there must all have quite a bit of common DNA 2019-10-27T01:50:20 < Steffanx> Not really. 2019-10-27T01:50:35 < mawk> fries are very different from dutch 2019-10-27T01:50:44 < mawk> with the weird language 2019-10-27T01:50:53 < mawk> they just got colonized by hollanders 2019-10-27T01:51:27 < Steffanx> All of the northerners are different. 2019-10-27T01:51:55 < sync> specing: meh, our local bus service tried it 2019-10-27T01:52:02 < sync> and they are now back to diesel hybrid 2019-10-27T01:52:10 < Steffanx> One part has a weird language, but their behaviour is similar 2019-10-27T01:54:15 < Steffanx> Sure there is some common DNA, we're human after all, jadew :P 2019-10-27T01:54:22 < jadew> heh 2019-10-27T01:54:37 < Steffanx> Imagine dutchland having 17M people and Romania just 2 million more. 2019-10-27T01:54:49 < jadew> went to a wedding recently and I could tell who my relatives were from their traits alone 2019-10-27T01:54:52 < mawk> dutchland is twice the population of paris 2019-10-27T01:55:28 < specing> sync: they bought a chinese BEV bus? 2019-10-27T01:56:08 < Steffanx> There isnt a single close family member in this village. 2019-10-27T01:56:18 < jadew> Steffanx, that's good 2019-10-27T01:56:26 < Steffanx> One has the same last name, but is not related at all. 2019-10-27T01:56:54 < Steffanx> Is it common in romania to stay in the same village/city forever? 2019-10-27T01:57:07 < jadew> Steffanx, until recently, yeah 2019-10-27T01:57:17 < jadew> nowdays the youth are fleeing to bigger cities tho 2019-10-27T01:57:28 < jadew> like I did 2019-10-27T01:57:29 < catphish> hello sirs 2019-10-27T01:57:42 < jadew> well, more like my parents did 2019-10-27T01:57:43 < Steffanx> Hello mr catphish 2019-10-27T01:57:45 < catphish> i have a question for you most esteemed ladies and gentlemen 2019-10-27T01:57:46 < jadew> they just dragged me along 2019-10-27T01:57:52 < mawk> I have an enormous amount of work to do for my android hacking idea, and I hesitate between two options: forking the AOSP component to do it in my own way, or working it from the ground up, but nicolas@desktop:~/Android/AOSP/frameworks/native/libs/binder$ find . -not -path './tests/*' -not -path './ndk/test/*' \( -name '*.c' -o -name '*.h' -o -name '*.cpp' \) -exec wc -l {} + # 21644 total 2019-10-27T01:58:34 < catphish> for USB firmware updated, should i use the STM32 hardwired bootloader, or should i flash something else? 2019-10-27T01:58:42 < catphish> *updates 2019-10-27T01:58:56 < mawk> does the stm32 rom support usb ? it's not always the case 2019-10-27T01:58:59 < zyp> catphish, for what thing? 2019-10-27T01:59:01 < mawk> if the answer is not you have your answer by default 2019-10-27T01:59:02 < Steffanx> Depends on your needs obviously 2019-10-27T01:59:25 < catphish> i have an stm32l433, it has a good built in bootloader with USB DFU 2019-10-27T01:59:40 < zyp> IMO the rom bootloader is not very end-user-friendly 2019-10-27T01:59:44 < Steffanx> The question is, do you need all of those 21k lines of code, mawk 2019-10-27T01:59:46 < catphish> the only real downside is that it needs a hardware pushbutton to access it 2019-10-27T01:59:52 < Steffanx> Or can you do it in 1k 2019-10-27T02:00:01 < Steffanx> Or 20. 2019-10-27T02:00:10 < catphish> and of course you need the right DFU software 2019-10-27T02:00:22 < zyp> and probably driver binding on windows 2019-10-27T02:00:33 < zyp> or does DFU automatically get bound to winusb nowadays? 2019-10-27T02:00:35 < mawk> my hope was that I could hack something in 10 lines but after reading a bunch of the files I fear it won't work like that Steffanx , but I guess I can at least steal the relevant code from AOSP, it's open sourz 2019-10-27T02:00:39 < catphish> so, are there better options? 2019-10-27T02:00:46 < zyp> write your own? 2019-10-27T02:00:49 < mawk> it doesn't need a hardware pushbutton catphish 2019-10-27T02:00:52 < mawk> you can trigger BOOT0 from software 2019-10-27T02:00:57 < zyp> bootloaders are dead simple 2019-10-27T02:01:50 < Steffanx> So what are you hacking mawk. Still the dns stuff? 2019-10-27T02:02:00 < catphish> mawk: getting into the ROM bootloader from software seems possible, but non trivial, and of course there's the downside of being able to brick the thing if you fail to flash it correctly 2019-10-27T02:02:14 < mawk> the dns stuff is over and pushed Steffanx , it's in the official app on the play store right now and got 150€ in donations for it 2019-10-27T02:02:26 < mawk> I'm just trying to make a android lib for hackers 2019-10-27T02:02:31 < mawk> to share the knowledge I built when doing the dns stuff 2019-10-27T02:02:50 < mawk> yes that's what I meant with banking catphish 2019-10-27T02:03:09 < zyp> catphish, I tend to use a simple design; it consists of a protocol part and a flashloader part 2019-10-27T02:03:12 < mawk> or did I say it ? I didn't send the message, what I wanted to say is that with your own bootloader you can have security mechanism such as banking to avoid bricking 2019-10-27T02:03:30 < mawk> or a security mechanism as a readonly bootloader, which is exactly what the ST rom is, you can't brick if you have ST rom 2019-10-27T02:03:39 < mawk> if you put an emergency pushbutton on the board for instance 2019-10-27T02:03:41 < catphish> mawk: i don't think you sent it :) 2019-10-27T02:03:46 < zyp> the flashloader part contains anything related to writing the flash, and has three interface functions; prepare/write/finalize 2019-10-27T02:04:14 < zyp> prepare unlocks the flash, write is called to fill in the next block of data and finalize relocks the flash 2019-10-27T02:04:29 < zyp> then slap whatever protocol you need on top of that 2019-10-27T02:04:38 < zyp> personal stuff uses usb, workstuff uses ethernet 2019-10-27T02:04:48 < catphish> mawk: yeah my first thought was a pushbutton, but then i thought maybe i could avoid that by having something that checksums a rom at boot and drops to a bootloader if it fails, plus the option to drop to the bootloader from the main program 2019-10-27T02:04:52 < zyp> for USB I've written both HID and DFU frontends 2019-10-27T02:05:01 -!- sterna [~Adium@c-b7e0e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-27T02:05:06 < mawk> catphish: the ST rom already does it ! you're in luck 2019-10-27T02:05:13 < mawk> if there is no valid code in flash it will default to bootloader mode 2019-10-27T02:05:37 < catphish> mawk: i didn't know it could check validity 2019-10-27T02:05:41 < catphish> that might be ideal 2019-10-27T02:05:42 < mawk> also yeah you can trigger BOOT0 from your own code, and you can arrange for that checksum-checking code to never be overwritten (unless the user wanted to brick his board I guess) 2019-10-27T02:05:52 < zyp> catphish, it can't 2019-10-27T02:06:01 < mawk> it checks if there is something in the flash, it's not an extended check 2019-10-27T02:06:15 < mawk> there could be a thousand meaning of "validity" 2019-10-27T02:06:17 < zyp> and «validity» is not brick-proof 2019-10-27T02:06:26 < catphish> yeah i saw it checks that the first adress is empty, but that's not a huge help 2019-10-27T02:06:42 < catphish> the safest option is of course a ROM and a pushbutton 2019-10-27T02:06:54 < catphish> but i wondered if most people had something else 2019-10-27T02:06:56 < mawk> but as zyp says your own bootloader is easy (or relatively easy when USB gets involved) and you can make your own anti-bricking arrangements 2019-10-27T02:07:01 < mawk> two banks of code, readonly stub, etc 2019-10-27T02:07:24 < zyp> if you rely on firmware to get into bootloader, you'll brick it as soon as you flash a misbehaving (but otherwise valid) firmware 2019-10-27T02:07:39 < mawk> yes 2019-10-27T02:07:47 < mawk> that's what I meant by "user wanting to brick his board" 2019-10-27T02:08:15 < zyp> I design my bootloaders so that I can always force-enter the bootloader before the firmware is started 2019-10-27T02:08:24 < zyp> by holding some buttons or something during startup 2019-10-27T02:08:52 < catphish> right now my board has no buttons 2019-10-27T02:09:06 < catphish> but actually, it has plenty of other pins that users could short in an emergency 2019-10-27T02:09:10 < zyp> https://cgit.jvnv.net/arcin/tree/bootloader.cpp <- here's a complete bootloaer in a little over 300 lines 2019-10-27T02:09:23 < zyp> flashloader at line 71, protocol at line 143 2019-10-27T02:09:35 < zyp> bootloader entry conditions at line 255 2019-10-27T02:10:33 < zyp> https://cgit.jvnv.net/arcin/tree/bootloader.cpp?id=9a09653 <- here's an older version of the same bootloader, with DFU protocol instead of custom HID-based, protocol at line 114 2019-10-27T02:12:12 < catphish> so i might consider having a physical button, but not on boot0 2019-10-27T02:12:27 < zyp> that's what I did on my nfc reader 2019-10-27T02:12:36 < zyp> I added a button just to force bootloader entry 2019-10-27T02:12:52 < zyp> arcin doesn't need that, since it hooks up to buttons 2019-10-27T02:13:34 < zyp> arcin force-enters bootloader if buttons 1 and 2 are held down when being powered on 2019-10-27T02:16:45 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-27T02:16:50 < catphish> on a related note, if jlpcb don't have the stm32l433 for assembly, what's the next cheapest option for small scale production? 2019-10-27T02:17:13 < zyp> in what direction? 2019-10-27T02:17:21 < mawk> doing it yourself ? 2019-10-27T02:17:32 < catphish> lets assume i'm an incompetent moron 2019-10-27T02:17:35 < mawk> assemble everything but the chip, if it's qfp or whatever you can do it yourself 2019-10-27T02:17:35 < mawk> ah 2019-10-27T02:17:45 < zyp> oh, I thought you meant alternatives to stm32l433 2019-10-27T02:18:15 < catphish> zyp: sorry, i meant alternative assemblers who will solder that chip 2019-10-27T02:18:31 < Steffanx> Then I'd advice you some positive self esteem group therapy 2019-10-27T02:18:39 < catphish> tbh i could probably redesign based on a chip they stock 2019-10-27T02:18:50 < catphish> but that's really annoying to have to do 2019-10-27T02:18:58 < mawk> why wouldn't you solder it yourself ? 2019-10-27T02:19:05 < mawk> it's not like BGA 2019-10-27T02:19:30 < zyp> BGA is not that hard 2019-10-27T02:19:56 < catphish> i don't have the skills to reliably solder batches of qfp 2019-10-27T02:20:13 < mawk> ah it's qfp, even easier 2019-10-27T02:20:25 < Steffanx> Why the l433 btw? It doesnt look that cheap to me at all 2019-10-27T02:20:30 < catphish> i mean, i can solder my own, and they work, but i don't think i'd want to do it in any quantity 2019-10-27T02:20:42 < catphish> Steffanx: really? seemed very cheap to me 2019-10-27T02:20:46 < mawk> maybe you can send your chips to jlcpcb assembly lab ? 2019-10-27T02:20:51 < zyp> catphish, how much? 2019-10-27T02:20:54 < mawk> cheap for you when you buy 10, but not when you buy 10000 2019-10-27T02:20:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-27T02:21:53 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-27T02:22:24 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has quit [Ping timeout: 246 seconds] 2019-10-27T02:22:34 < catphish> well the first problem is that there are 3 different ways to compare prices, and they rarely yield the same result 2019-10-27T02:22:45 < Steffanx> It does depends on your needs ofcourse, catphish . But their newish G4 family seems pretty nice price wise. 2019-10-27T02:22:47 < zyp> catphish, how much are you paying for the chip? 2019-10-27T02:23:00 < catphish> the stm32l433cb is $1.84 in official st pricing 2019-10-27T02:23:07 < mawk> that's alright I guess 2019-10-27T02:23:15 < catphish> it's £3 from UK suppliers in smallish batches 2019-10-27T02:23:29 < Steffanx> Oh I'd check real prices, I never trust the st prices 2019-10-27T02:23:39 < mawk> st price is batch price I think 2019-10-27T02:23:44 < catphish> so this is the problem 2019-10-27T02:23:46 < zyp> st prices are fairly irrelevant as long as you're not buying from st 2019-10-27T02:23:58 < catphish> there 1) mouser prices 2) st prices 3) china prices 2019-10-27T02:24:07 < Steffanx> Yeah 2019-10-27T02:24:16 < catphish> https://www.mouser.co.uk/ProductDetail/STMicroelectronics/STM32L433CBT6?qs=sGAEpiMZZMuoKKEcg8mMKF376ZS%2FEJoEDGvqY1vp%252BbNmCLxOWKh3%252Bw%3D%3D 2019-10-27T02:24:22 < catphish> that's mouser-pricing 2019-10-27T02:24:25 < zyp> the price that matters is the price you'd pay where you buy it 2019-10-27T02:24:53 < zyp> speaking of prices, I still have a ton of l052 to get rid off 2019-10-27T02:25:02 < catphish> do they have usb? 2019-10-27T02:25:05 < zyp> yes 2019-10-27T02:25:16 < catphish> can they run at 40MHz? 2019-10-27T02:25:19 < zyp> no 2019-10-27T02:25:25 < catphish> gay 2019-10-27T02:25:26 < zyp> I mean, maybe, but they are not rated to 2019-10-27T02:25:41 < catphish> those are basically my only 2 requirements 2019-10-27T02:26:15 < catphish> i went with the l433 because it has 1) crystal-less USB and 2) USB DFU ROM 2019-10-27T02:26:34 < catphish> but neither of those things are particularly critical 2019-10-27T02:27:02 < catphish> if i can find a chip that's more widely available, has usb, and runs at 40MHz, i can move 2019-10-27T02:27:32 < catphish> actually i'd rather it ran at 80MHz like the l433 (i divide the timer by 2) 2019-10-27T02:27:40 < Steffanx> Or 160? 2019-10-27T02:27:45 < catphish> (or 120/160) 2019-10-27T02:27:48 < catphish> sure 2019-10-27T02:27:56 < Steffanx> Pin count? 2019-10-27T02:29:07 < catphish> not many https://nutty.tk/floppy.pdf - currently using s 48 pin chip 2019-10-27T02:29:37 < catphish> but less would be fine 2019-10-27T02:30:01 < catphish> it needs USB and a 32 bit timer with 2 timer-attached pins 2019-10-27T02:30:08 < catphish> that's basically it 2019-10-27T02:30:35 < catphish> (plus 11 more GPIOs) 2019-10-27T02:31:30 < catphish> the problem i have is that the selection is overwhelming and ofen what looks good in stmcube turns out to be more expensive / less available 2019-10-27T02:31:59 < catphish> i could just start looking for the most highly available / common / cheap chips on ali and check if they're a fit 2019-10-27T02:32:55 < Cracki> then you'll get f103c 2019-10-27T02:33:22 < Cracki> tho not sure if they have any 32 bit timers already 2019-10-27T02:33:50 < catphish> the f103 is pretty interesting, but 72mhz is an incovenient clock speed 2019-10-27T02:34:10 < Cracki> works for usb, may be inconvenient for your sampling? 2019-10-27T02:34:10 < aandrew> yeah for me selection starts at digikey 2019-10-27T02:34:14 < aandrew> sort by price 2019-10-27T02:34:19 < aandrew> filter on features 2019-10-27T02:34:19 < mawk> f103 is mega old and you're a modern man catphish 2019-10-27T02:34:27 < catphish> actually, thinking about it, i could probably get away with a 16 bit timer now 2019-10-27T02:35:02 < catphish> for a 40mhz timer is non negotiable 2019-10-27T02:36:14 < catphish> not sure how easy it is to get a 40mhz timer on a 72mhz chip using 2 PLLs? 2019-10-27T02:36:59 < Cracki> aliseeks gives me this for the cheapest stm32: https://www.aliexpress.com/item/32694446110.html 2019-10-27T02:37:15 < Cracki> what needs a 40 mhz timer? 2019-10-27T02:37:40 < Cracki> 72 = 8*9, 40 = 8*5... 2019-10-27T02:37:49 < catphish> i need to time things with 25ns resolution 2019-10-27T02:38:03 < Cracki> if it's got an 8 mhz hse, maybe it's possible 2019-10-27T02:38:20 < catphish> your link no work for me 2019-10-27T02:38:23 < Steffanx> Lcsc seems to have the f401. Which is 84mhz and seems cheapish. 2019-10-27T02:38:35 < Cracki> same ;) I always thought they only block my country for some listings 2019-10-27T02:38:46 < catphish> ah 2019-10-27T02:38:49 < mawk> aliexpress racism 2019-10-27T02:39:25 < Steffanx> Will you join the racism train in dutchland? If so, please stay in France:P 2019-10-27T02:39:40 < mawk> :( 2019-10-27T02:39:49 < Cracki> it's natural 2019-10-27T02:40:00 < mawk> geert wilders looks like a monkey Steffanx 2019-10-27T02:40:02 < mawk> can I come now ? 2019-10-27T02:40:05 < Cracki> we should stop denying ourselves what everyone else on earth enjoys 2019-10-27T02:40:29 < catphish> oh, jlpcb actually do now list the l433 2019-10-27T02:40:53 < Steffanx> I mean the train with people that claims everyone and his mother is a racist, mawk :P 2019-10-27T02:41:01 < catphish> $3.4711 @ 100 from them 2019-10-27T02:41:08 < Steffanx> Not the train with racists 2019-10-27T02:41:49 < Steffanx> F401 beats that. Unless you dont like qfn 2019-10-27T02:43:35 < catphish> i might try mouser, sort by price 2019-10-27T02:44:36 < mawk> ah lol Steffanx , not a chance no 2019-10-27T02:44:38 < Steffanx> Have you seen jpa- price thingy based on digikey prices catphish ? 2019-10-27T02:44:48 < mawk> people already call me racist, so I can't join their crowd 2019-10-27T02:44:53 < Steffanx> Jpa -'s* 2019-10-27T02:45:02 < mawk> my neighbor told every other neighbor I'm racist to try to drive me away from the apartment 2019-10-27T02:45:15 < catphish> Steffanx: no 2019-10-27T02:45:20 < Steffanx> If not, see the topic. 2019-10-27T02:45:41 < Steffanx> Hmm is it broken? 2019-10-27T02:45:50 < catphish> it does look roken :( 2019-10-27T02:46:15 < Steffanx> Ohno, it was so perfect 2019-10-27T02:47:14 < catphish> well it seems clear that if i want a cheap chip, it's going to come down to whether a 72MHz chip can run a 40MHz timer 2019-10-27T02:49:17 < Cracki> you can run it at 40 or 80 mhz 2019-10-27T02:49:24 < Cracki> jiggle the clock tree? 2019-10-27T02:49:39 -!- Mangy_Dog [~Mangy_Dog@94.2.234.208] has quit [Ping timeout: 240 seconds] 2019-10-27T02:50:02 < Cracki> do you need the chip to be that cheap? 2019-10-27T02:50:11 < catphish> well technically no 2019-10-27T02:50:36 < catphish> but it costs more than the rest of my BOM combined 2019-10-27T02:50:42 < catphish> so seems sane to make savings 2019-10-27T02:50:53 < Cracki> you'll sell it for 10x markup I hope? 2019-10-27T02:51:08 < catphish> hopefully 2019-10-27T02:52:31 < catphish> Cracki: run what at 40 or 80MHz? the timer, or just everything? 2019-10-27T02:52:44 < Cracki> everything, when in doubt 2019-10-27T02:53:08 < Cracki> you work with something that runs 80 mhz, right? that's derived from an 8 mhz crystal? 2019-10-27T02:53:52 < Cracki> if you need a 32 bit timer, not sure how much choice you have 2019-10-27T02:54:14 < Cracki> boss decided on f303, which has at least one 32 bit timer, and it's cheaper than 3 bucks 2019-10-27T02:54:22 < catphish> right now i run everything at 80MHz from a 16MHz HSI 2019-10-27T02:54:24 < Cracki> dunno, 1.xx or 2.xx bucks was it 2019-10-27T02:54:50 < catphish> i may be fine with 16 bit timer if that's a big deal 2019-10-27T02:54:59 < Cracki> so even if you get an f103, run it at 80 and I think the timer's clock can be /2 to 40 mhz 2019-10-27T02:55:19 < catphish> will 72mhz chips run fine at 80? 2019-10-27T02:55:26 < catphish> seems, odd 2019-10-27T02:55:26 < Cracki> out of spec but far away from anything people reported as possible 2019-10-27T02:55:48 < catphish> i guess everything is slightly overclockable 2019-10-27T02:55:59 < Cracki> people experimented with 128 mhz and 108 2019-10-27T02:56:03 < Cracki> (blue pill crowd) 2019-10-27T02:56:32 < Cracki> depends on st/gd/cks/... 2019-10-27T02:57:13 < Cracki> not sure what suffers first. they may just have checked basic operation (core, execution from flash) 2019-10-27T02:57:24 < catphish> well now that i see jlpcb have the chip i already designed for, it doesn't seem a bad choice at all 2019-10-27T02:57:27 < Ultrasauce> probably wont work across the rated temperature and voltage range 2019-10-27T02:57:41 < catphish> but a $3.50 chip is a silly choice if a $1 chip will work 2019-10-27T02:58:39 < Cracki> it's also just a 2.50 difference, times number of boards, which you have to compare to time/cost of porting 2019-10-27T02:58:40 < catphish> so, everyone loves the f103, but everyone says it's old, is there not a new chip that is similarly ubiquitous? 2019-10-27T02:58:51 < catphish> right 2019-10-27T03:00:05 < Cracki> someone long ago picked the f103 for something, and that got popular, so now that subseries is easily available 2019-10-27T03:00:18 < Cracki> pick whatever. I doubt you'll have trouble soucing a few hundred at a time 2019-10-27T03:00:54 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Remote host closed the connection] 2019-10-27T03:01:24 < catphish> well i like my l433, the board works, the code works, i guess i'll stop worrying 2019-10-27T03:01:43 < Cracki> maybe design it as a shield/hat/cape/whatever on top of some existing board that's popular and capable enough 2019-10-27T03:01:49 < Cracki> yeh 2019-10-27T03:01:54 < Cracki> it's a nucleo, isn't it? 2019-10-27T03:02:04 < catphish> no, DIY 2019-10-27T03:02:06 < Cracki> hm 2019-10-27T03:02:35 < Cracki> but there is an L433RC nucleo, so you can expect people to take the firmware and run it on a nucleo with perfboard 2019-10-27T03:02:43 < catphish> Cracki: https://i.imgur.com/TEDFmct.jpg 2019-10-27T03:03:01 < catphish> i didn't know there was a l433 nucleo actually 2019-10-27T03:03:10 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-27T03:03:10 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-27T03:03:10 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-27T03:03:14 < catphish> that makes my firmware even more useful 2019-10-27T03:03:32 < Cracki> https://www.st.com/en/evaluation-tools/nucleo-l433rc-p.html 2019-10-27T03:03:51 < catphish> handy 2019-10-27T03:03:59 < Cracki> not sure those double row headers can be reused to be floppy. their pin mapping on nucleos can appear arbitrary 2019-10-27T03:04:25 < catphish> almost certainly not directly 2019-10-27T03:04:35 < catphish> they randomly put power pins all over the place 2019-10-27T03:04:58 < Cracki> and it has micro usb, not that nice full size usb a(?) 2019-10-27T03:05:10 < catphish> i like my USB B 2019-10-27T03:05:29 < catphish> feels more sturdy than micro 2019-10-27T03:05:50 < Cracki> quite 2019-10-27T03:06:09 < catphish> the main question that remains is whether to add a pushbutton, and if so, whater to make it boot0 or not 2019-10-27T03:08:22 < catphish> also the matter of whether to bother to add USB ESD protection 2019-10-27T03:09:07 < aandrew> fuck a duck 2019-10-27T03:09:17 < aandrew> I just lost 6mos+ of weather data from my little datalogger 2019-10-27T03:09:34 < catphish> :( 2019-10-27T03:10:35 < Cracki> until exploded 2019-10-27T03:11:19 < aandrew> rtl_433 -F csv | tee thelog instead of tee -A thelog 2019-10-27T03:11:25 < aandrew> sigh 2019-10-27T03:12:15 < Cracki> shouldn't have used command line 2019-10-27T03:12:24 < Cracki> make moar snapshots 2019-10-27T03:13:09 < Cracki> command line is fun because it's dangerous 2019-10-27T03:13:29 < aandrew> lol 2019-10-27T03:13:32 < aandrew> nothing dangerous about it 2019-10-27T03:13:33 < Cracki> also do daily log rotation 2019-10-27T03:13:44 < Cracki> never ever touch a file again that's not from today 2019-10-27T03:13:45 < aandrew> should've had a backup 2019-10-27T03:13:47 < aandrew> that's all 2019-10-27T03:13:54 < aandrew> and it was minor, just kind of a shitty thing 2019-10-27T03:14:05 < BrainDamage> a gui program would've had a checkbox for append/overwrite, and you'd have the same possible oversight 2019-10-27T03:14:06 < Cracki> things can be designed to be safe (like thorium reactors) or not 2019-10-27T03:14:23 < Cracki> a gui program could have refused to overwrite any file 2019-10-27T03:14:29 < Cracki> and made you delete it outside the program 2019-10-27T03:14:36 < BrainDamage> that sounds dumb 2019-10-27T03:14:38 < Cracki> (or in the file picker) 2019-10-27T03:14:42 < Cracki> that sounds safe 2019-10-27T03:15:17 < BrainDamage> also, a cli tool could have made the same choice, what you said is not inherent property to guis 2019-10-27T03:15:21 < Cracki> there's a reason people alias their rm to rm -i in interactive shells 2019-10-27T03:15:34 < Cracki> so it ASKS whether you're aware of what you're doing 2019-10-27T03:15:38 < BrainDamage> I have rm aliased to trash-cli 2019-10-27T03:15:46 < Cracki> see, that's safe 2019-10-27T03:16:14 < Cracki> humans aren't made for dealing with immediate and irreversible consequences 2019-10-27T03:16:49 < Cracki> the press of a button is too quick for regret to kick in and prevent the consequence 2019-10-27T03:17:01 < BrainDamage> what is really hell in cli is space character handling 2019-10-27T03:17:11 < Cracki> that's why a bunch of people teach their email clients to delay sending an email by minutes 2019-10-27T03:17:37 < Cracki> so you *can* run to the mailbox, bash it open, and get back that nastygram you didn't really mean to send 2019-10-27T03:21:25 < Cracki> I need to follow my own advice more. one particular program I have uses memory mapped files for a modest million-record time series, mostly so it can crash (it's research code) without losing data... but it should instead mmap that to some temp file, and recover, and ask to save or discard instead 2019-10-27T03:22:40 < Cracki> rsync knows what a "dry run" is. they implemented that "not because it is easy, but because it is hard" 2019-10-27T03:24:39 < mawk> every serious program should implement dry run 2019-10-27T03:25:05 -!- Jak_o_Shadows [~Jak@unaffiliated/jak-o-shadows/x-0479135] has joined ##stm32 2019-10-27T03:25:08 < Cracki> people make their embedded microcontrollers boot from banked flash so firmware update is allowed to fail (for any number of reasons) 2019-10-27T03:25:38 < mawk> yeah I suggered that to catphish 2019-10-27T03:26:07 < Cracki> apple's system management controller has two levels of bootloader so you can fuck the bootloader and still drop underneath it to fix that 2019-10-27T03:26:52 < catphish> mawk: sorry, i only saw that in passing, how does that work? write to a separate area, verify, them make live? 2019-10-27T03:26:57 < Cracki> I mostly have PTSD from Neo wannabes who dare to call themselves administrator while at the same time furiously hacking into a green on black terminal and saying "oops" more than once a year 2019-10-27T03:27:13 < Cracki> you can make that as complicated as you like 2019-10-27T03:27:52 < Cracki> such as checksums, or setting up watchdog before jumping into app code, or hardware selectable switch 2019-10-27T03:28:26 < Cracki> it should allow you to boot "the old" firmware too, with nothing but a switch or jumper 2019-10-27T03:29:32 < Cracki> such a scheme lets you make a bare bones bootloader, and permits the application to write to the other region (pulling firmware from wifi/...) 2019-10-27T03:30:44 < Cracki> basically you ensure that at all times there's at least one consistent set of code on it, so update can fail from power outage or whatever 2019-10-27T03:30:58 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-27T03:31:23 < aandrew> I usually just make my bootloader unwritable 2019-10-27T03:31:27 < Cracki> new application code could, once it determines that it has successfully started, flip something so it gets priority over the old version 2019-10-27T03:31:35 < mawk> yes catphish 2019-10-27T03:31:42 < mawk> have two areas for your code, A and B 2019-10-27T03:31:54 < mawk> and you arrange so that at any time at least one of them is in a valid state 2019-10-27T03:31:56 < mawk> so that you can't brick 2019-10-27T03:32:36 < catphish> does this require 2 compiles of the code? or can arm code easily be relocateable 2019-10-27T03:32:43 < catphish> i know very little about this 2019-10-27T03:32:56 < Cracki> you can ask for position independent code 2019-10-27T03:33:29 < Cracki> interrupt vector table prolly needs some attention 2019-10-27T03:33:55 < aandrew> it's not bad 2019-10-27T03:34:19 < aandrew> worst I had to do was to make a CM0 IVR able to bounce interrupts to the application IVR siince CM0 does not have relocatible vectors 2019-10-27T03:34:21 < Cracki> I think some chips help you with that and map the one or the other area into a fixed address range 2019-10-27T03:35:47 < Cracki> On-the-fly firmware update for dual bank STM32 microcontrollers https://www.st.com/content/ccc/resource/technical/document/application_note/group0/ab/6a/0f/b7/1a/84/40/c3/DM00230416/files/DM00230416.pdf/jcr:content/translations/en.DM00230416.pdf 2019-10-27T03:36:20 < catphish> handy 2019-10-27T03:36:28 < Cracki> > even demanding real time tasks can be executed during the update. 2019-10-27T03:36:33 < Cracki> that's probably the killer feature 2019-10-27T03:37:14 < Cracki> you could, if you require, even write code that "translates" the runtime state to the new firmware 2019-10-27T03:38:36 < aandrew> Cracki: https://pastebin.com/njTedi6a is how I bounced the ISR 2019-10-27T03:38:44 < Cracki> *takes notes* 2019-10-27T03:40:10 < catphish> now i must sleep 2019-10-27T03:40:18 < con3> night catphish 2019-10-27T03:40:52 < con3> anyone here ever touched node-red? 2019-10-27T03:41:28 < aandrew> not me 2019-10-27T03:41:56 < Cracki> looks funky 2019-10-27T03:42:03 < Cracki> >originally by IBM 2019-10-27T03:42:15 < con3> trying to figure out wtf people use it 2019-10-27T03:42:45 < Cracki> at the very least, sounds like graphical configuration 2019-10-27T03:42:59 < Cracki> >flows created in Node-RED are stored using JSON 2019-10-27T03:43:26 < con3> yeah, just not sure why the guy before me used node-red while putting together a production server 2019-10-27T03:43:27 < Cracki> https://flows.nodered.org/node/node-red-contrib-openplc 2019-10-27T03:43:46 < Cracki> looks like they want to drag electrical engineers out of the ladder diagram and structured text cave they've been hiding in 2019-10-27T03:44:02 < con3> it looks stupid 2019-10-27T03:44:08 < Cracki> looks scratchy 2019-10-27T03:44:08 < con3> but im not sure if im missing something 2019-10-27T03:44:24 < con3> like i dont like arduino, but this is ...eh... 2019-10-27T03:44:53 < con3> this is next level code gen 2019-10-27T03:45:12 < Cracki> ye looks like hyped webshit for scamming venture capital 2019-10-27T03:45:31 < Cracki> code gen is all the rage, and it's good when it works, and people should use it more, but it's also hard to do well 2019-10-27T03:45:47 < con3> I just need to figure out wtf he started using this in the server. Need to dig a bit more 2019-10-27T03:46:02 < Cracki> >version 1.0 released 30 Sep 2019 2019-10-27T03:46:05 < con3> shit that keeps me up at night 2019-10-27T03:46:15 < Cracki> uhhhhh maybe replace it all 2019-10-27T03:46:39 < con3> Cracki: i really want to, but the guy over me will want me to give good reasons why 2019-10-27T03:46:57 < Cracki> maybe the old guy you get to take over from needs to explain this stuff 2019-10-27T03:46:59 < con3> Anyone will dig a bit 2019-10-27T03:47:03 < Cracki> you aren't an archeologist 2019-10-27T03:47:31 < con3> He fucked off two years ago and now im staring at shit that boggles my mind 2019-10-27T03:47:56 < con3> he has rpi's everywhere that are just communicating with 2 external adc's and logging data 2019-10-27T03:47:56 < Cracki> 55 seconds "essentials" https://www.youtube.com/watch?list=PLyNBB9VCLmo1hyO-4fIZ08gqFcXBkHy-6&v=ksGeUD26Mw0 2019-10-27T03:48:00 < con3> my mind hurts 2019-10-27T03:48:15 < Cracki> looks a little like simulink 2019-10-27T03:48:19 < Cracki> and i can get behind simulink 2019-10-27T03:50:41 < Cracki> the biggest advantage to code generation is that magic adaptability is easier. if you wrote code, the apis would have to match. like this, you can wire it up, and some program can resolve mismatches and adapt 2019-10-27T03:50:57 < Cracki> *apis and data types and all that 2019-10-27T03:51:42 < Cracki> stuff traditionally expressed as code now becomes data 2019-10-27T03:53:03 < Cracki> say, do you speak some afrikaans? 2019-10-27T03:53:15 < con3> I do... 2019-10-27T03:54:02 < con3> eh streets are filled with drunk students 2019-10-27T03:54:03 < Cracki> how do you say "yeah can I get uhhh... none pizza with left beef?" 2019-10-27T03:54:07 < Cracki> kek 2019-10-27T03:54:11 < Cracki> why are you not one of them 2019-10-27T03:54:48 < con3> what do you mean by "none pizza with left beef"? 2019-10-27T03:54:55 < Cracki> oh boy one sec 2019-10-27T03:55:13 < Cracki> https://knowyourmeme.com/memes/special-delivery-instructions 2019-10-27T03:55:15 < con3> got to get stuff done, no time for drinky rn 2019-10-27T03:55:27 < Cracki> https://gizmodo.com/reflections-on-the-10th-anniversary-of-none-pizza-with-1819692097 2019-10-27T03:55:40 < con3> feel like i missed a hell of a meme 2019-10-27T03:56:20 < con3> omg 2019-10-27T03:57:13 < con3> "cut into isosceles triangles" 2019-10-27T03:57:38 < Cracki> >send your cutest delivery boy 2019-10-27T03:57:53 < con3> "cut into pentagram" 2019-10-27T03:57:59 < con3> this is amazing 2019-10-27T03:58:01 < Cracki> they do that to hotel room service too. "print this, put it on the pillow" 2019-10-27T03:58:41 < con3> On October 19th, 2007 -> ah this is way back 2019-10-27T03:01:12 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-27T03:18:55 < Thorn> is there a standard sdio protocol for wi-fi modules? is there a spec available? 2019-10-27T03:30:19 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Bye!] 2019-10-27T03:30:36 < Cracki> this looks like it may be device-specific and only based on SDIO but also it's broadcom so maybe it's just their extensions that warrant a special driver https://wiki.freebsd.org/SDIO 2019-10-27T03:31:19 < Cracki> haven't found any indication that wifi is a particular specification on top of SDIO 2019-10-27T03:32:24 < Thorn> murata wifi modules (using broadcom chips) and even esp have sdio interfaces. in case of murata it seems to be the only way to access wifi from a host cpu 2019-10-27T03:32:41 < Cracki> spec at least has "commands" https://www.ercankoclar.com/wp-content/uploads/2017/11/SD-InputOutput-SDIO-Card-Specification.pdf 2019-10-27T03:32:52 < Cracki> looks like generic bus 2019-10-27T03:33:00 < Thorn> so if sdio is a standard (rather than e.g. spi or xSPI) maybe there is a standard protocol / command set too 2019-10-27T03:33:06 < Thorn> like sd card 2019-10-27T03:33:39 < Cracki> this as a "wireless lan" spec https://www.sdcard.org/downloads/pls/ 2019-10-27T03:34:04 < Cracki> >dl.php 2019-10-27T03:34:06 < Cracki> bitch what 2019-10-27T03:34:29 < Thorn> https://esp32.com/viewtopic.php?t=8899 2019-10-27T03:35:05 < Cracki> oh, apparently that's for "memory cards" that do squirreling away of data over wifi 2019-10-27T03:36:45 < Cracki> AT commands... people need to be shot for using that today for anything other than controlling a modem 2019-10-27T03:43:10 < fenugrec> AT commands just won't die 2019-10-27T03:47:48 < con3> im about to start using AT commands :p 2019-10-27T03:48:49 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 268 seconds] 2019-10-27T03:52:52 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-27T03:59:46 -!- rue_shop1 [~rue_mohr@d50-92-152-244.bchsia.telus.net] has joined ##stm32 2019-10-27T03:59:53 < rue_shop1> are all the smart people asleep just now? 2019-10-27T04:00:28 < rue_shop1> I need assistance with an openocd glitch 2019-10-27T04:01:16 < aandrew> nah we're up 2019-10-27T04:02:00 < aandrew> wait 2019-10-27T04:02:06 < aandrew> you're rue_mohr 2019-10-27T04:02:41 < rue_shop1> openocd -f $(OPENOCD_SCRIPTS_DIR)/interface/stlink-v2.cfg -f $(OPENOCD_SCRIPTS_DIR)/target/stm32f1x.cfg -c init -c targets -c "halt" -c "flash write_image erase ./$(BINARY).elf" -c "verify_image ./$(BINARY).elf" -c "reset run" -c shutdown 2019-10-27T04:02:54 < rue_shop1> I am rue_mohr 2019-10-27T04:02:54 < aandrew> you're up near gibsons 2019-10-27T04:03:00 < rue_shop1> heh 2019-10-27T04:03:05 < rue_shop1> yes, in canada 2019-10-27T04:03:12 < aandrew> I follow you on twitter 2019-10-27T04:03:22 < rue_shop1> :) I hope I'm entertaining 2019-10-27T04:03:44 < rue_shop1> so, every second time I run that openocd sequence the stm32 hangs 2019-10-27T04:04:04 < rue_shop1> aandrew, gee, I only have about 550 followers too ;) 2019-10-27T04:04:05 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-27T04:04:15 < rue_shop1> (which is just insane) 2019-10-27T04:04:15 < aandrew> can you run those instructions one at a time to see *which* is causing the hang 2019-10-27T04:05:07 < rue_shop1> and of course it works fine now, hmm 2019-10-27T04:05:17 < rue_shop1> I'll keep working, its going to happen again... 2019-10-27T04:05:48 < rue_shop1> aandrew, do you use an IDE for flashing chips? 2019-10-27T04:05:50 < aandrew> I am (kind of) in vancouver 2019-10-27T04:05:58 < rue_shop1> ah good show! 2019-10-27T04:06:06 < rue_shop1> didn't expect to see anyone from canada 2019-10-27T04:06:10 < aandrew> I don't use an IDE at all 2019-10-27T04:06:16 < rue_shop1> k, makefile? 2019-10-27T04:07:03 < aandrew> yep 2019-10-27T04:07:15 < rue_shop1> do you have a `make install` ? 2019-10-27T04:07:20 < aandrew> make flash yes 2019-10-27T04:07:22 < aandrew> but I use jlik 2019-10-27T04:07:24 < aandrew> jlink 2019-10-27T04:07:27 < rue_shop1> ok 2019-10-27T04:07:34 < rue_shop1> jlink... 2019-10-27T04:07:42 < aandrew> lots and lots of people have no issues with stlink though so don't think that's the issue 2019-10-27T04:07:54 < aandrew> also PaulFertser is here, he's *the* person behind openocd 2019-10-27T04:08:06 < rue_shop1> what timezone? 2019-10-27T04:08:19 < aandrew> he's always got a few hints too 2019-10-27T04:08:24 < aandrew> IIRC he's in Moscoq 2019-10-27T04:08:26 < aandrew> Moscow 2019-10-27T04:08:44 < rue_shop1> 5am 2019-10-27T04:08:50 < rue_shop1> its ok, my night just started 2019-10-27T04:08:51 < aandrew> there's #openocd there as well 2019-10-27T04:08:58 < rue_shop1> ah 2019-10-27T04:09:01 < aandrew> yeah I'm presently in Ontario 2019-10-27T04:09:06 < aandrew> 10pm here 2019-10-27T04:09:20 < rue_shop1> oh, you were in vancouver a few weeks ago? 2019-10-27T04:09:22 < rue_shop1> moving? 2019-10-27T04:09:39 < aandrew> moving to vancouver, I'm an ontarioite 2019-10-27T04:09:41 < aandrew> ontarian 2019-10-27T04:09:48 < rue_shop1> 7pm, so, hopefull another 8 hours till bed 2019-10-27T04:10:02 < aandrew> wow you're workign late tonight 2019-10-27T04:10:04 < rue_shop1> ah, I think I recall who ya are 2019-10-27T04:10:06 < rue_shop1> work? 2019-10-27T04:10:09 < rue_shop1> hell no, play 2019-10-27T04:10:20 < aandrew> yeah I DM'd you about moving to vancouver 2019-10-27T04:10:33 < aandrew> well my work is my play too so I often conflate them 2019-10-27T04:10:33 < rue_shop1> I'm working on porting my filter profiler to an stm32 cause I need multiple capture channels 2019-10-27T04:10:49 < rue_shop1> your suit guy :) 2019-10-27T04:10:51 < aandrew> which STM32 are you running on? 2019-10-27T04:10:54 < aandrew> yes 2019-10-27T04:10:57 < rue_shop1> blue pill 2019-10-27T04:11:45 < aandrew> ok so F103 2019-10-27T04:11:52 < rue_shop1> yea 2019-10-27T04:13:22 < aandrew> the pressure sensitive bargraph was pretty cool 2019-10-27T04:13:33 < rue_shop1> it'll get finished 2019-10-27T04:13:46 < rue_shop1> its a distraction project tho 2019-10-27T04:14:10 < rue_shop1> I'm wondering if I should change the sensor to an LED thing so that its a more accessible project for people 2019-10-27T04:14:21 < rue_shop1> I dont know if most people can get conductive foam 2019-10-27T04:15:01 < aandrew> right 2019-10-27T04:15:29 < aandrew> a piece of standard foam on top of a pressure sensor would work too 2019-10-27T04:15:46 < rue_shop1> :) those are too pricey for me to bother with 2019-10-27T04:15:52 < rue_shop1> their like $5 ea? 2019-10-27T04:17:00 < aandrew> yeah they are pricier than I thought 2019-10-27T04:17:25 < rue_shop1> they look like they should be cheap too 2019-10-27T04:17:30 < aandrew> yes very 2019-10-27T04:17:52 < rue_shop1> ok, they are cheap, just not as much as I like 2019-10-27T04:18:09 < rue_shop1> if I was being paid to make a project with one I would't worry about it 2019-10-27T04:18:33 < aandrew> I did some low level code for this sensor: https://www.sparkfun.com/products/13162 2019-10-27T04:18:45 < aandrew> it's actually *really* interesting how he does that 2019-10-27T04:19:00 < aandrew> super cheap and works quite well 2019-10-27T04:19:40 < rue_shop1> neat 2019-10-27T04:19:53 < rue_shop1> I have gesture detection on my bucket list 2019-10-27T04:20:01 < aandrew> you could do something very similar for your sensor 2019-10-27T04:20:16 < rue_shop1> I'd like to keep it simpler 2019-10-27T04:20:37 < rue_shop1> something with an analog voltage is awesome cause I can just throw that tiny13 at it and be done 2019-10-27T04:21:15 < aandrew> well it could be adapted for that pretty easily 2019-10-27T04:21:20 < aandrew> it's designed for I2C output right now 2019-10-27T04:24:53 < rue_shop1> only have 1 line tho 2019-10-27T04:25:17 < rue_shop1> 8 pins, -2 power -1 reset, -4 charlieplexed leds = 1 :) 2019-10-27T04:26:21 < aandrew> right; just saying you could get a PWM output that would become analog 2019-10-27T04:26:30 < aandrew> that sensor is overkill though 2019-10-27T04:26:36 < aandrew> you only need one LED and one IR receiver 2019-10-27T04:27:24 < rue_shop1> ah!, cool 2019-10-27T04:27:33 < rue_shop1> just two leds ;) 2019-10-27T04:27:50 < aandrew> if you wanted to do it the harder way, yes 2019-10-27T04:28:09 < rue_shop1> I have to play with it, I'm pretty sure the ~1.4V generated by an led is enough signal for the ADC 2019-10-27T04:28:36 < aandrew> that particular sensor uses a typical 38kHz receiver and the LED is modulated the same way to help ignore ambient light 2019-10-27T04:29:47 < rue_shop1> yup 2019-10-27T04:30:01 < rue_shop1> me and a fellow once used TV ir gear as rangers for a robot 2019-10-27T04:30:17 < rue_shop1> you can draw back the range by *slightly* being off key with the 38Khz 2019-10-27T04:30:28 < rue_shop1> so a sweep can tell you how far away an object is 2019-10-27T04:30:52 < rue_shop1> *but* my buddy was wearing kinda fuzzy blue shoes, which the ir just couldn't see 2019-10-27T04:31:15 < rue_shop1> we decided is was not a realiable means of distance detection and also put wiskers on the robots 2019-10-27T04:32:03 < rue_shop1> http://ruemohr.org/%7Eircjunk/robots/multibots/smokingprogress/dscn6164.jpg 2019-10-27T04:32:29 < rue_shop1> that avr board was made *just* before the arduino came out 2019-10-27T04:32:38 < rue_shop1> by about 18 months 2019-10-27T04:32:51 < aandrew> was your friend Elvis? 2019-10-27T04:32:57 < rue_shop1> the idea of generic controller boards was about 2 years before that 2019-10-27T04:33:11 < rue_shop1> no.. not suade shoes 2019-10-27T04:33:43 < Cracki> wait, IR for ranging? 2019-10-27T04:33:51 < Cracki> going by rx signal strength? 2019-10-27T04:33:59 < Cracki> or even phase? 2019-10-27T04:34:05 < rue_shop1> the recievers have a threshold 2019-10-27T04:34:12 < Cracki> ic 2019-10-27T04:34:18 < aandrew> rue_shop1: nice tank 2019-10-27T04:34:21 < aandrew> my son would love that 2019-10-27T04:34:24 < rue_shop1> you can play with the filter by being offkey from the 38khz 2019-10-27T04:34:40 < rue_shop1> and thus drop the detection range 2019-10-27T04:35:12 < rue_shop1> but a red table leg and a blue fuzzy shoe will still have drastically different levels at the same distance 2019-10-27T04:36:01 < Cracki> ye red is closer to IR than blue is 2019-10-27T04:36:19 < Cracki> and fuzzy vs mirror surface 2019-10-27T04:37:22 < aandrew> I've been working on this for him: https://imgur.com/VCPXHHu 2019-10-27T04:37:26 < aandrew> wtf I thought i Had a better picture 2019-10-27T04:38:00 < Cracki> carrot slices 2019-10-27T04:38:01 < aandrew> Cracki: the XYZ basically just PWM'd the LED at 38kHz and then PWM'd the intensity of the 38kHz LED 2019-10-27T04:38:13 < aandrew> depending on where the hand was, the receiver would pick it up at different times 2019-10-27T04:38:19 < Cracki> ah, sweep 2019-10-27T04:38:34 < aandrew> with two LEDs and a sensor he could readily detect XZ 2019-10-27T04:38:40 < aandrew> with three he had basic XYZ support 2019-10-27T04:40:39 < rue_shop1> aandrew, yup, I have no idea what those are 2019-10-27T04:41:11 < aandrew> thos are the inner races for the tank treads 2019-10-27T04:41:19 < rue_shop1> aaah 2019-10-27T04:41:31 < aandrew> I have the body and tracks printed, but have to do some tweaking and probably reprint a few things 2019-10-27T04:42:21 < rue_shop1> https://www.thingiverse.com/thing:1739978 2019-10-27T04:42:26 < rue_shop1> I got a lot of that printed 2019-10-27T04:42:30 < rue_shop1> no wheels tho 2019-10-27T04:43:03 -!- mitrax [mitrax@lfbn-ncy-1-393-156.w83-196.abo.wanadoo.fr] has quit [] 2019-10-27T04:43:15 < jadew> I got my son starcraft, he can build all the tanks he likes 2019-10-27T04:43:46 < jadew> was actually thinking of starting a project with him 2019-10-27T04:43:50 < jadew> some sort of robot 2019-10-27T04:45:24 < rue_shop1> #robotics is a great way to combine electronics, mechanics, and programming ;) 2019-10-27T04:45:53 < mawk> mechatronics 2019-10-27T04:45:57 < jadew> I've been in there for years, but I haven't built a robot since more than 15 years ago 2019-10-27T04:46:18 < rue_shop1> I need to get that 4th arm going so I can do a video of some kind 2019-10-27T04:46:28 < rue_shop1> tho, having 3 tracking me like that was pretty creepy 2019-10-27T04:46:36 < rue_shop1> maybe I'm not cut out to be a leader 2019-10-27T04:47:27 < rue_shop1> hmm, my output isn't outputting 2019-10-27T04:47:37 < mawk> is it inputting ? 2019-10-27T04:48:56 < rue_shop1> just going over the code again 2019-10-27T04:49:02 < rue_shop1> I'm doing it differently than usual 2019-10-27T04:49:37 < rue_shop1> gpio_set_mode(GPIOB, GPIO_MODE_OUTPUT_2_MHZ, GPIO_CNF_OUTPUT_PUSHPULL, (GPIO10 | GPIO11 | GPIO1) ); 2019-10-27T04:49:39 < rue_shop1> is that valid? 2019-10-27T04:49:52 < rue_shop1> I looked at the library I think it should be 2019-10-27T04:50:21 < aandrew> rue_shop1: I'm not familiar with that API 2019-10-27T04:50:41 < rue_shop1> opencm3 2019-10-27T04:50:43 < Cracki> it may not like the oring and instead expect those defines to be bit indices rather than bit masks 2019-10-27T04:50:49 < rue_shop1> which I dont think is winning the race 2019-10-27T04:51:53 < rue_shop1> if I set an io to 2Mhz, the cpu does wait states when writing, right? 2019-10-27T04:53:13 < aandrew> hm? 2019-10-27T04:53:15 < aandrew> why would it do that? 2019-10-27T04:53:21 < Cracki> nah, that's drive strength 2019-10-27T04:53:24 < Cracki> slew rate 2019-10-27T04:55:14 < rue_shop1> hmm 2019-10-27T04:55:39 < rue_shop1> so, if I toggle the pin as fast as the processor can, the output will just take a dazed state of confusion 2019-10-27T04:57:39 < Cracki> scope it :) 2019-10-27T04:57:49 < rue_shop1> oh I have a scope on it 2019-10-27T04:58:10 < Cracki> you may see a good signal, if load is small enough 2019-10-27T04:58:30 < Cracki> capacitive load that is 2019-10-27T04:59:12 < rue_shop1> ok I can bit-bang a square wave 2019-10-27T05:00:14 < jadew> bad connection? 2019-10-27T05:01:15 < rue_shop1> I'll check the pin set/clear works and go back to the code thats supposed to be pushing serial commands 2019-10-27T05:03:01 < rue_shop1> one NOP between setting and clearing a pin may not be enough :) 2019-10-27T05:09:05 < rue_shop1> hahaha 2019-10-27T05:09:16 < rue_shop1> its wwaaaaaay too fast for the stm32 DSO 2019-10-27T05:15:25 < Thorn> the QSPI appnote is only 95 pages long 2019-10-27T05:24:36 -!- jly [uid355225@gateway/web/irccloud.com/x-mtdojqwomzzxjekl] has joined ##stm32 2019-10-27T05:24:49 < Cracki> stm32 dso has 1 mhz sampling or so? 2019-10-27T05:25:07 < Cracki> f103 adc are on that order of speed iirc 2019-10-27T05:27:43 < aandrew> rue_shop1: there are also bit-banding aliases for all the GPIO regs 2019-10-27T05:27:59 < aandrew> so you can literally write 1 to the bit_on and write another 1 to the bit_off 2019-10-27T05:28:12 < aandrew> so your toggle rate can by your APB bus rate/2 2019-10-27T05:35:15 < rue_shop1> #define A9833SPort GPIOB_BSRR 2019-10-27T05:35:15 < rue_shop1> #define A9833CPort GPIOB_BRR 2019-10-27T05:35:19 < rue_shop1> set port and clear port 2019-10-27T05:35:26 < rue_shop1> #define A9833StrobeH() A9833SPort = (1< #define A9833StrobeL() A9833CPort = (1< all the set and clear macros are working 2019-10-27T05:36:03 < rue_shop1> but the scope I'm using cant see activity on the clock and strobe lines 2019-10-27T05:36:35 < rue_shop1> https://paste.debian.net/1110936/ 2019-10-27T05:40:00 < Thorn> rue_shop1: is gpiob clock enabled 2019-10-27T05:40:16 < Thorn> also comment says PORTC while the code references GPIOB 2019-10-27T05:40:39 < rue_shop1> rcc_periph_clock_enable(RCC_GPIOB); 2019-10-27T05:40:56 < rue_shop1> yes, there are some comments from the project I ported it from 2019-10-27T05:42:31 < Thorn> are the pins set to output mode? 2019-10-27T05:42:44 < rue_shop1> gpio_set_mode(GPIOB, GPIO_MODE_OUTPUT_10_MHZ, GPIO_CNF_OUTPUT_PUSHPULL, (GPIO10 | GPIO11 | GPIO1) ); 2019-10-27T05:42:49 < Thorn> they're in input by default 2019-10-27T05:42:50 < rue_shop1> manually toggling them is working 2019-10-27T06:00:15 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-27T06:00:41 < rue_shop1> k the logic analizer says my dso is out to lunch 2019-10-27T06:00:50 < rue_shop1> its an answer I'll accept 2019-10-27T06:10:47 < rue_shop1> ok, whats up with floating point math on this thing 2019-10-27T06:11:26 < rue_shop1> as soon as I added code that did a fp multiply it all just stopped 2019-10-27T06:11:45 < rue_shop1> its like its going for a fpu that isn't there 2019-10-27T06:12:11 < Thorn> >floating point on a mcu 2019-10-27T06:12:17 < Thorn> mcdonalds is hiring 2019-10-27T06:12:33 < rue_shop1> this is not unreasonable, its slow, but can be done 2019-10-27T06:12:52 < rue_shop1> even an apple IIe with a 6502 can do floating point 2019-10-27T06:13:31 < Cracki> f103 has no fpu, but f4 and f3 (CM4) have fpu 2019-10-27T06:13:39 < rue_shop1> yea 2019-10-27T06:13:41 < Cracki> even without fpu it's still quite fast about that stuff 2019-10-27T06:13:42 < rue_shop1> so software it 2019-10-27T06:13:48 < rue_shop1> but is there a gcc flag I need to set? 2019-10-27T06:13:57 < Cracki> if it crashes, you must have told it to use an fpu 2019-10-27T06:14:05 < rue_shop1> -mthumb ? 2019-10-27T06:14:09 < Cracki> not enough 2019-10-27T06:14:23 < Cracki> it shouldn't assume an fpu unless you tell it 2019-10-27T06:14:30 < Cracki> so maybe something else is going on 2019-10-27T06:14:45 < rue_shop1> this is not my first encounter with math issues on the stm32 2019-10-27T06:15:22 < rue_shop1> https://paste.debian.net/1110937/ 2019-10-27T06:15:28 < rue_shop1> see anything that stands out? 2019-10-27T06:17:02 < rue_shop1> -mfloat-abi=softfp 2019-10-27T06:17:02 < rue_shop1> aha 2019-10-27T06:17:03 < rue_shop1> ? 2019-10-27T06:17:28 < Cracki> good idea but I would be surprised if it didn't do that already 2019-10-27T06:17:40 < Cracki> perhaps run the target under a debugger and see what it faults on 2019-10-27T06:19:21 < rue_shop1> I only have gdb, and I'm sure you dont have time to help me extract this out of that 2019-10-27T06:19:24 < Thorn> iirc if you tell gcc to use cortex-m3 as opposed to m4 it should assume soft fp by default because m3 can't have hard fp 2019-10-27T06:20:24 < Cracki> care to try segger ozone? 2019-10-27T06:20:30 < Cracki> you have a jlink so it's compatible 2019-10-27T06:21:11 < rue_shop1> oh maybe 2019-10-27T06:24:08 < rue_shop1> stlink 2019-10-27T06:24:59 < rue_shop1> I didn't tell it to use 4, 2019-10-27T06:25:07 < rue_shop1> https://paste.debian.net/1110937/ 2019-10-27T06:25:11 < rue_shop1> is the makefile 2019-10-27T06:27:10 < rue_shop1> anyone here do fp on stm32? 2019-10-27T06:27:19 < rue_shop1> er, blue pill... 2019-10-27T06:27:45 < Thorn> this is a red pill channel 2019-10-27T06:27:46 < Cracki> I did. cubemx set up the project just right (soft fp) 2019-10-27T06:27:57 < Cracki> it also set up hard floats for an f303 correctly 2019-10-27T06:28:07 < rue_shop1> so why is fp making this bail 2019-10-27T06:28:09 < Cracki> f303 BLACK PILL board 2019-10-27T06:28:22 < Cracki> what evidence do you have to say that fp makes it fail? 2019-10-27T06:28:25 < Cracki> and HOW does it fail? 2019-10-27T06:29:03 < rue_shop1> as soon as I add code that does a fp operation the whole thing just halts 2019-10-27T06:29:04 < Cracki> it may need to be told to link some library you aren't linking at the moment 2019-10-27T06:29:30 < rue_shop1> ok, linker thing... 2019-10-27T06:30:09 < Thorn> why no --specs in your makefile btw 2019-10-27T06:30:17 < Cracki> may this help you https://embeddedartistry.com/blog/2017/10/9/r1q7pksku2q3gww9rpqef0dnskphtc 2019-10-27T06:30:25 < Thorn> iirc newlib readme tells to use --specs 2019-10-27T06:31:41 < rue_shop1> a gcc flag or a linker flag? 2019-10-27T06:31:49 < Thorn> both iirc 2019-10-27T06:42:16 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-27T06:42:23 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-27T06:49:46 < rue_shop1> is a stm32F103 a cortex-m3 with thumb? 2019-10-27T06:51:05 -!- fc5dc9d4 [~quassel@p5B3A8038.dip0.t-ipconnect.de] has joined ##stm32 2019-10-27T06:52:14 < rue_shop1> I'm close to tearing apart the a******o plugin to see what it does 2019-10-27T06:54:55 < Cracki> they're all thumb 2019-10-27T06:55:03 < Cracki> yes it's m3 2019-10-27T06:55:40 -!- fc5dc9d4_ [~quassel@p5B08158A.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-27T07:01:19 < rue_shop1> ok, so basically I cant make a program work that does foo= 6.8; for(int i = 0; i < 8.4*foo; i++) NOP(); 2019-10-27T07:04:01 < Ultrasauce> rue why does your code always make me sad 2019-10-27T07:04:04 < Ultrasauce> stop doing that 2019-10-27T07:04:13 < rue_shop1> I need to do floating point 2019-10-27T07:04:30 < rue_shop1> it should just do software fp, but its not 2019-10-27T07:06:23 < Thorn> I've never needed any floating point in my embedded code. fixed point is far more natural 2019-10-27T07:06:47 < rue_shop1> I'm making a filter analizer 2019-10-27T07:07:15 < rue_shop1> I was doing it on a 16Mhz avr 2019-10-27T07:07:21 < rue_shop1> but I needed more capture timers 2019-10-27T07:07:30 < rue_shop1> the avr did great floating point 2019-10-27T07:07:38 < rue_shop1> atmega88 2019-10-27T07:08:37 < rue_shop1> I just need access to a makefile with the proper flags to make it do it in software 2019-10-27T07:09:14 < rue_shop1> interestingly enough, last week I was having issues with dividion 2019-10-27T07:09:26 < rue_shop1> integer division 2019-10-27T07:09:49 < rue_shop1> if I didn't explicitly use uint16_t it wouldn't work 2019-10-27T07:11:10 < Thorn> sounds like it's time to learn some C 2019-10-27T07:11:23 < rue_shop1> srrsly? 2019-10-27T07:11:54 < rue_shop1> your saying uint16_t/unsigned int should result in the wrong answer for the calc? 2019-10-27T07:12:42 < rue_shop1> gcc is simply doing the wrong things with the math cause of something I have/havn't said in the makefile 2019-10-27T07:15:40 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-27T07:19:13 < Thorn> I dunno, you can try pasting the code (both versions) to godbolt.org, it does support arm iirc 2019-10-27T07:19:17 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-27T07:19:18 -!- day__ is now known as day 2019-10-27T07:20:15 < Thorn> cortex-m3 supports integer division in hardware so no software functions to link 2019-10-27T07:20:19 < rue_shop1> its about the compile and link flags tho... 2019-10-27T07:20:33 < rue_shop1> yea, well gcc screwed up the integer thing on it 2019-10-27T07:20:52 < rue_shop1> in a really funkey way that I'm not gonna go back and recreate now 2019-10-27T07:20:54 < Thorn> cortex is upwards compatible, even v6m and v7m 2019-10-27T07:20:55 < Cracki> maybe it's just integer math in general 2019-10-27T07:21:08 < rue_shop1> cause I want to get a simple floating multiply to work without the processor locking up 2019-10-27T07:22:59 < Cracki> so I'm watching a husky rip into cardboard boxes and plastic shipping bags 2019-10-27T07:23:13 < rue_shop1> -msoft-float -mfloat-abi=soft 2019-10-27T07:23:32 < rue_shop1> to my understanding that should be pretty explicit about doing floating point in software 2019-10-27T07:23:58 < Cracki> I still think whatever's going on isn't to be blamed on how floating point is done 2019-10-27T07:24:21 < Cracki> lots of things can be described as "locking up" 2019-10-27T07:25:16 < Cracki> btw, plain floating point literals are implicitly double. that's why people always suffix the 'f' to literals: 12.345f 2019-10-27T07:25:20 < Cracki> to make them float/single 2019-10-27T07:25:25 < rue_shop1> https://paste.debian.net/1110955/ 2019-10-27T07:25:28 < rue_shop1> see main? 2019-10-27T07:25:48 < rue_shop1> see that basic little fp multiply I put in there? 2019-10-27T07:25:53 < Cracki> I would wonder if it ever enters the loop at all 2019-10-27T07:25:58 < rue_shop1> THAT stops it from looping 2019-10-27T07:26:07 < Cracki> i.e. whether line 67 does anything 2019-10-27T07:26:08 < rue_shop1> if I just make that i < 8, it works fine 2019-10-27T07:26:11 < Cracki> define "stop" 2019-10-27T07:26:19 < rue_shop1> the loop stops looping 2019-10-27T07:26:29 < rue_shop1> it ceases to run 2019-10-27T07:26:45 < Cracki> debugger 2019-10-27T07:26:59 < rue_shop1> which is crazy, cause gcc should notice its static and just compile it in 2019-10-27T07:27:00 < Cracki> I would wanna know whether it's staying in the for-loop or not, and why 2019-10-27T07:27:15 < rue_shop1> ok, so, lets debug it 2019-10-27T07:27:22 < rue_shop1> what command, openocd? 2019-10-27T07:27:35 < Ultrasauce> you are comparing an unsigned int and a double 2019-10-27T07:27:36 < Ultrasauce> dont do that 2019-10-27T07:27:44 < rue_shop1> it should do that just fine 2019-10-27T07:27:57 < rue_shop1> what debug command do you want me to run *now* 2019-10-27T07:28:05 < rue_shop1> it'll be halted right now 2019-10-27T07:28:13 < rue_shop1> its ready 2019-10-27T07:28:14 < Ultrasauce> would you like to argue with me about this or fix it 2019-10-27T07:28:31 < rue_shop1> no I'd like you to tell me what to run to find out what faulted it to halt 2019-10-27T07:28:45 < Cracki> set a break point on line 75, then run to there, then single step and watch variable values 2019-10-27T07:28:57 < rue_shop1> I'd also like a listing file to know if its compiling with floating point instructions or not 2019-10-27T07:29:05 < rue_shop1> Cracki, ok, give me a command 2019-10-27T07:29:25 < rue_shop1> cause I have no ability to use gdb 2019-10-27T07:29:32 < rue_shop1> or openocd 2019-10-27T07:29:36 < Cracki> I usually do that from an IDE so i don't have to deal with gdb and openocd directly :P 2019-10-27T07:29:42 < rue_shop1> mhm 2019-10-27T07:29:47 < Cracki> ozone counts too 2019-10-27T07:30:07 < rue_shop1> apt-cache search ozone 2019-10-27T07:30:07 < rue_shop1> root@freebee6:/files/programming/c/stm32/DDSctrl# 2019-10-27T07:30:19 < rue_shop1> aka, its not a package available 2019-10-27T07:30:31 < Cracki> you get it from segger's website 2019-10-27T07:30:45 < Cracki> and I'm not sure it's available on linux, but I can be mistaken 2019-10-27T07:30:57 < Cracki> oh it is 2019-10-27T07:31:01 < Cracki> excellent 2019-10-27T07:31:07 < rue_shop1> I need a base that can already do this 2019-10-27T07:31:15 < rue_shop1> arduino must be able to 2019-10-27T07:31:38 < rue_shop1> I'll put away my project and try it 2019-10-27T07:31:44 < Cracki> arduino can't debug. the best they have is instrumentation and device-side command line control of that. 2019-10-27T07:31:58 < rue_shop1> no but it can do software fp 2019-10-27T07:32:14 < Cracki> I'll go back to watching the husky 2019-10-27T07:32:17 < Cracki> he's fluffy 2019-10-27T07:33:58 < rue_shop1> you know, if I work this out, it means I know more than you guys do about stm32 2019-10-27T07:34:08 < rue_shop1> :P 2019-10-27T07:34:43 < Ultrasauce> if you work this out you'll find that i am not full of shit 2019-10-27T07:35:56 < rue_shop1> I'm really frustrated that you guys seem to be saying I shouldn't be trying to multiply 4.5*8.2 on a stm32 2019-10-27T07:36:08 < Cracki> sure you can do that 2019-10-27T07:36:23 < Cracki> C has a few rules about numbers and how they're promoted though 2019-10-27T07:36:32 < Cracki> I think in this case both sides of the comparison get promoted to float 2019-10-27T07:36:51 < Cracki> when both sides are int types and only differ in signedness, they get promoted to unsigned 2019-10-27T07:37:00 < Cracki> but float beats that... I think 2019-10-27T07:37:18 < Cracki> so anyway, everything's nonnegative and well within any value ranges 2019-10-27T07:37:48 < Cracki> the loop itself will take a lot more cycles than the single nop, so don't be surprised about that 2019-10-27T07:38:18 < rue_shop1> I'm measuring characteristics to freqs down to 1Hz 2019-10-27T07:38:26 < Cracki> let's say 100 cycles per iteration, that's 30k cycles, about 0.4 ms 2019-10-27T07:38:27 < rue_shop1> it takes 2 seconds to do the measurements 2019-10-27T07:38:39 < rue_shop1> if it takes 300ms for it to do the calcs, thats still ok 2019-10-27T07:39:24 < Cracki> anyway, debugger or figure out how to get assembly listing from your compiler. it might already dump that in some file. 2019-10-27T07:39:32 < rue_shop1> at this point even 4 hours isn't long enough for the stm32F1 to so 4.5*8.2 2019-10-27T07:39:45 < Cracki> if you have the elf, (gcc's?) objdump and similar tools can disassemble and show you 2019-10-27T07:40:02 < rue_shop1> gonna set up arduino for the stm32 and have it print me some floating point stuff 2019-10-27T07:44:20 < Ultrasauce> promoting an unsigned int to a double for the comparison is what is breaking here 2019-10-27T07:44:25 -!- jly [uid355225@gateway/web/irccloud.com/x-mtdojqwomzzxjekl] has quit [Quit: Connection closed for inactivity] 2019-10-27T07:44:30 < rue_shop1> it shouldn't break 2019-10-27T07:44:36 < rue_shop1> gcc should do conversions and be happy 2019-10-27T07:49:58 < Cracki> try turning that 5.5 into a 5.5f 2019-10-27T07:50:30 < Cracki> I don't think that'll do much. your code, on ordinary x64, runs just fine 2019-10-27T07:50:45 < rue_shop1> it runs fine on an 8 bit avr too 2019-10-27T07:51:00 < Cracki> but that's assuming there aren't any weird differences in handling of doubles by arm gcc 2019-10-27T07:51:16 < rue_shop1> its handing floating point wrong for this chip 2019-10-27T07:51:28 < Cracki> it's possible that it can't handle doubles 2019-10-27T07:51:36 < rue_shop1> I'm just working with arduino to see if its handling floats ok 2019-10-27T07:51:40 < Cracki> so give it a try, rewrite that constant to 5.5f 2019-10-27T07:52:21 < rue_shop1> void Ad9833SetFreq ( float f, uint8_t channel ) { 2019-10-27T07:52:21 < rue_shop1> / unsigned long div = f * 10.73741824; // make freq register from frequency (25Mhz) (for 20Mhz use 13.4217728); 2019-10-27T07:52:21 < rue_shop1> unsigned long div = f * 5497.558; // *512 for SIN lookup 2019-10-27T07:52:37 < rue_shop1> those are the actual lines of code causing the fault I'm tracing 2019-10-27T07:57:17 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-27T07:57:25 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-27T07:58:58 < rue_shop1> yup, ok, good with printing a static float value to the serial port 2019-10-27T07:59:07 < rue_shop1> lets try a loop with sin() values 2019-10-27T08:02:55 < rue_shop1> huh, that might have been pushing it too far, back off for a multiply 2019-10-27T08:04:30 < rue_shop1> multiply is ok 2019-10-27T08:06:32 < rue_shop1> oh no just a glitch with my programmer 2019-10-27T08:06:37 < rue_shop1> sin() is working just fine 2019-10-27T08:07:12 < rue_shop1> so what can arduino do that we all cant 2019-10-27T08:07:19 < rue_shop1> (floating point obviously) 2019-10-27T08:13:36 < rue_shop1> iiinteresting 2019-10-27T08:13:46 < rue_shop1> its not playing with the floating point settings at all 2019-10-27T08:15:14 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-27T08:20:03 < antto> o hai therr rue 2019-10-27T08:27:00 < rue_shop1> hi 2019-10-27T08:27:51 < rue_shop1> can you write a simple program that loops thru angles in steps of 0.005 radians and prints the sin() values for them to a serial port? 2019-10-27T08:28:24 < Cracki> it is possible. 2019-10-27T08:28:50 < Ultrasauce> pastebin your .map rue 2019-10-27T08:29:08 < rue_shop1> ok, 1 min. 2019-10-27T08:31:01 < rue_shop1> http://ruemohr.org/%7Eircjunk/tempimage/program.map 2019-10-27T08:31:44 < antto> btw, how much instruction cycles does expf() take on a cortex M4? 2019-10-27T08:32:06 < Ultrasauce> why are you using such an absurdly old compiler 2019-10-27T08:32:23 < rue_shop1> heh, is gcc out of date???? 2019-10-27T08:32:52 < rue_shop1> well, it IS debian, but let me see 2019-10-27T08:33:19 < antto> 5.4.1 is old 2019-10-27T08:33:38 < rue_shop1> gcc-arm-none-eabi is already the newest version (15:5.4.1+svn241155-1). 2019-10-27T08:33:46 < rue_shop1> according to debian 2019-10-27T08:34:12 < rue_shop1> lets aside that, I'm not gonna start random complex upgrades 2019-10-27T08:34:14 < antto> rue_shop1 on debian i just compile muh own compiler 2019-10-27T08:34:32 < rue_shop1> is it putting fpu insturctions in there? 2019-10-27T08:34:32 < antto> like, i got avr-gcc 7.3.0 or something 2019-10-27T08:34:50 < antto> ..since long ago 2019-10-27T08:34:52 < rue_shop1> I'm just walking the flags that arduino uses 2019-10-27T08:35:15 < rue_shop1> if you would like to encourage debian to get with it, I'll totally back you 2019-10-27T08:35:32 < antto> not gonna happen 2019-10-27T08:35:39 < antto> debian stable is about stability 2019-10-27T08:36:09 < rue_shop1> too bad stable and operational aren't mutual 2019-10-27T08:36:15 < antto> but for teh compil0r i make a compromise cuz i want teh newer one cuz it has noice C++ thangs, so i build my own 2019-10-27T08:36:37 < rue_shop1> arduino packages its own 2019-10-27T08:36:44 < rue_shop1> lets see what version it is 2019-10-27T08:36:52 < antto> i found some .sh script and modified it 2019-10-27T08:37:10 < antto> i think for arm gcc it would be even simpler 2019-10-27T08:37:47 < rue_shop1> arduino is using gcc 4.8.3 2019-10-27T08:38:01 < rue_shop1> and its floating point works 2019-10-27T08:38:22 < Ultrasauce> anyway i am pretty sure at least part of the issue is an incorrectly configured linker 2019-10-27T08:38:23 < antto> yer floatz don't work? 2019-10-27T08:38:35 < Ultrasauce> missing the architecture flags that tell it to use the correct libc 2019-10-27T08:38:48 < rue_shop1> oh? 2019-10-27T08:38:55 < Ultrasauce> (but idk if they changed the way that works in later saner versions) 2019-10-27T08:40:14 < rue_shop1> the older one is working... 2019-10-27T08:40:22 < rue_shop1> (packaged for arduino) 2019-10-27T08:40:31 < Ultrasauce> the older one is using a build system that was not hamfistedly modified by you 2019-10-27T08:40:32 < rue_shop1> which, might have a special config 2019-10-27T08:41:09 < rue_shop1> https://paste.debian.net/1110971/ 2019-10-27T08:41:15 < rue_shop1> this is the current makefile 2019-10-27T08:41:41 < rue_shop1> I tried to add the floating point flags, but the arduino compile sequence doesn't use them that I can find 2019-10-27T08:42:12 < antto> makefile :~( 2019-10-27T08:42:20 < rue_shop1> yup 2019-10-27T08:42:30 < rue_shop1> should I be using systemD? 2019-10-27T08:42:36 < antto> >:) 2019-10-27T08:42:49 < antto> wait, systemD or systemd? 2019-10-27T08:43:07 < rue_shop1> ... 2019-10-27T08:43:31 < rue_shop1> "-lm -lgcc -mthumb -Wl,--cref -Wl,--check-sections -Wl,--gc-sections -Wl," 2019-10-27T08:43:40 < rue_shop1> the options on the arudino compile are ... 2019-10-27T08:44:09 < rue_shop1> long... 2019-10-27T08:44:36 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-27T08:44:54 < rue_shop1> I dont know if c++ compilers build in floating point differently 2019-10-27T08:45:11 < antto> oh, u use C, right 2019-10-27T08:45:15 < antto> duuh 2019-10-27T08:45:50 < rue_shop1> honestly extra overhead isn't gonna matter on a 72Mhz controller I'm performing a 4Mhz task with 2019-10-27T08:46:15 < antto> overhead? 2019-10-27T08:47:18 < rue_shop1> back in 1995 someone said that there were no bennifits to coding in assembler 2019-10-27T08:47:37 < rue_shop1> I pointed out I could, in assembler, write hello world in 20 bytes 2019-10-27T08:47:42 < rue_shop1> and he didn't belive me 2019-10-27T08:47:56 < rue_shop1> he didn't admit defeat when I showed it to him either 2019-10-27T08:48:44 < antto> ehm, it's 2019, use C++ 2019-10-27T08:48:57 < antto> who needs C 2019-10-27T08:49:13 < antto> aint nobody got time for asm 2019-10-27T08:49:38 < rue_shop1> I'm not being paid 2019-10-27T08:49:40 < rue_shop1> I'm fine 2019-10-27T08:50:00 < antto> yeah, well if ya speek asm - sure, i don't 2019-10-27T08:52:06 < rue_shop1> ever used sprintf on an stm32f1? 2019-10-27T08:53:18 < antto> i've never used an stm32 2019-10-27T09:00:49 < rue_shop1> ... 2019-10-27T09:06:43 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 265 seconds] 2019-10-27T09:07:18 < jpa-> Steffanx, catphish: prices fixed now; looks like digikey changed their url format a bit; to see the graphs may need to press shift-F5 / ctrl-F5 to fully reload 2019-10-27T09:11:47 < jpa-> Steffanx: though the graphs start to be less useful now that there are almost 2000 points on each - hard to make anything out 2019-10-27T09:21:13 -!- rue_mohr [~rue_mohr@d50-92-152-244.bchsia.telus.net] has joined ##stm32 2019-10-27T09:21:24 < rue_mohr> arm-none-eabi-gcc -march=native -Q --help=target 2019-10-27T09:21:33 < rue_mohr> -mfloat-abi= soft 2019-10-27T09:22:02 < rue_mohr> -msoft-float 2019-10-27T09:23:01 < rue_mohr> if, by default, its using software fp, wtf is going wrong? 2019-10-27T09:26:12 < rue_mohr> cortex-m3 2019-10-27T09:28:20 < rue_mohr> ARMv7-M. A Thumb®-2 Instruction Set Architecture (ISA) subset, consisting of all base Thumb-2 instructions 2019-10-27T09:29:04 < rue_mohr> armv7-m 2019-10-27T09:29:18 < rue_mohr> -march=armv7-m 2019-10-27T09:29:37 < rue_mohr> -mcpu=cortex-m3 2019-10-27T09:30:17 < jpa-> rue_mohr: going wrong how? cortex-m3 doesn't have hardware floats anyway 2019-10-27T09:30:32 < rue_mohr> software float isnt' working 2019-10-27T09:30:43 < jpa-> isn't working how? 2019-10-27T09:30:52 < rue_mohr> I'm sure its cause of something I'm doing wrong in the compile/link flags, but I cant find it 2019-10-27T09:30:59 < rue_mohr> the processor locks if I do a fp operation 2019-10-27T09:31:18 < jpa-> what does the hardfault cause register say? (SCB->CFSR) 2019-10-27T09:31:36 < rue_mohr> I dont know how to get to it 2019-10-27T09:31:45 < jpa-> print in debugger 2019-10-27T09:32:02 < rue_mohr> can I do with withopenocd? 2019-10-27T09:32:06 < jpa-> also check the instruction at the fault point, perhaps you are linking against some hard float library - it's best to use "gcc" for linking instead of using "ld" directly 2019-10-27T09:32:19 < jpa-> rue_mohr: sure, just "p /x SCB->CFSR" in gdb 2019-10-27T09:32:20 < rue_mohr> I dont have an ide with a bunch of integrated stuff 2019-10-27T09:32:34 < rue_mohr> ok, can you help me connect gdb? 2019-10-27T09:33:02 < jpa-> start openocd, then in another terminal run "arm-none-eabi-gdb myfirmware.elf" and then type "tar ext :3333" 2019-10-27T09:34:21 < rue_shop1> ok, few mins, I have to reverse out some makefile options 2019-10-27T09:36:26 < rue_shop1> heh, and install gdb 2019-10-27T09:36:45 < jpa-> how did you install arm-none-eabi-gcc if it didn't come with matching gdb? 2019-10-27T09:36:57 < rue_shop1> apt-get install 2019-10-27T09:37:03 < rue_shop1> matching? 2019-10-27T09:37:17 < jpa-> ok 2019-10-27T09:37:24 < rue_shop1> oh, ... 2019-10-27T09:37:35 < rue_shop1> OH 2019-10-27T09:37:39 < rue_shop1> oops... 2019-10-27T09:37:54 < jpa-> well matching in that way it's usually best to install them both from same place, even though they are mostly version compatible, there are some problems with e.g. gdb 8 vs. gcc 4.7 2019-10-27T09:38:10 < rue_shop1> yea, I ... just did a silly thing and have to wait for it to finish 2019-10-27T09:43:13 < antto> apt-get install codeblocks <- u got an IDE now 2019-10-27T09:43:18 < antto> ;P~ 2019-10-27T09:43:42 < rue_shop1> ok, openocd is working 2019-10-27T09:45:22 < rue_shop1> lets try installing gdb-arm-none-eabi then... 2019-10-27T09:45:57 < rue_shop1> ok I got a prompt 2019-10-27T09:46:18 < rue_shop1> Remote debugging using :3333 2019-10-27T09:46:18 < rue_shop1> 0x00000000 in ?? () 2019-10-27T09:46:32 < Ultrasauce> >:3333 2019-10-27T09:46:41 < rue_shop1> undefined debug reason 7 - target needs reset on the openocd 2019-10-27T09:47:01 < jpa-> try "mon reset halt" and then "continue" 2019-10-27T09:47:44 < rue_shop1> continuing 2019-10-27T09:47:58 < jpa-> then ctrl-c and "bt" to see where it is 2019-10-27T09:48:17 < rue_shop1> void blocking_handler(void) │ 2019-10-27T09:48:18 < rue_shop1> >│103 { │ 2019-10-27T09:48:18 < rue_shop1> │104 while (1); │ 2019-10-27T09:48:18 < rue_shop1> │105 } 2019-10-27T09:48:21 < rue_shop1> wtf 2019-10-27T09:48:22 < antto> rue pls 2019-10-27T09:48:23 < jpa-> yeah, good 2019-10-27T09:48:29 < jpa-> that's the default hardfault handler 2019-10-27T09:48:29 < rue_shop1> sorry bad copy 2019-10-27T09:48:42 < rue_shop1> ok, so 2019-10-27T09:48:44 < jpa-> try if p /x SCB->CFSR works now 2019-10-27T09:49:21 < rue_shop1> I got ya wrong, "no SCB in current context" 2019-10-27T09:49:29 < jpa-> yeah, wait a moment 2019-10-27T09:49:33 < rue_shop1> thanks 2019-10-27T09:49:56 < jpa-> download this https://raw.githubusercontent.com/karlp/zypsnips/master/armv7m-vecstate-zippe.gdb 2019-10-27T09:50:08 < jpa-> and then "source ....../armv7m-vecstate-zippe.gdb" and then "vecstate" 2019-10-27T09:50:43 < jpa-> i usually use just p /x SCB->CFSR directly, but that only works if the headers have been included; that macro by zippe will work in any case 2019-10-27T09:51:31 < antto> aww, gdb doesn't know about the SFRs 2019-10-27T09:53:09 < jpa-> antto: gdb has no inherent knowledge about anything, only what is visible from debug symbols 2019-10-27T09:53:24 < antto> yeah.. 2019-10-27T09:53:25 < rue_shop1> np, having an argument with my fileserver 2019-10-27T09:53:51 < rue_shop1> no symbol SCB in current context 2019-10-27T09:54:10 < rue_shop1> oh sorry 2019-10-27T09:54:10 < antto> so does anyone happen to know roughly how many instruction cycles an expf() takes on a corteggz M4? 2019-10-27T09:54:20 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-27T09:54:21 < rue_shop1> ok, hold on 2019-10-27T09:54:26 < jpa-> antto: M4 or M4F? 2019-10-27T09:54:36 < antto> M4 with fpu of course 2019-10-27T09:54:41 < antto> are there M4 without fpu? 2019-10-27T09:54:47 < jpa-> sure 2019-10-27T09:54:48 < rue_shop1> UsageFault due to access to disabled/absent coprocessor 2019-10-27T09:54:54 < antto> o_O sh*t 2019-10-27T09:55:01 < rue_shop1> gcc must be trying to use a fpu 2019-10-27T09:55:02 < antto> then i gotta eyeball more carefully 2019-10-27T09:55:02 < rue_shop1> arg 2019-10-27T09:55:05 < rue_shop1> yes? 2019-10-27T09:55:13 < jpa-> rue_shop1: yeah, something is trying to use it 2019-10-27T09:55:23 < rue_shop1> the linker cant do that 2019-10-27T09:55:25 < jpa-> rue_shop1: see if there is anything in "bt" that could point to where 2019-10-27T09:55:27 < rue_shop1> has to be gcc 2019-10-27T09:55:29 < Ultrasauce> can i reiterate that the linker is loading the wrong libc now 2019-10-27T09:55:48 < jpa-> you can also do e.g. "frame 1" and "display /i $pc" to see what instruction it was trying to run 2019-10-27T09:56:06 < jpa-> rue_shop1: are you linking with "ld" directly, or with "gcc"? 2019-10-27T09:56:23 < rue_shop1> #1 0x080001c6 in main () at main.c:75 2019-10-27T09:56:23 < rue_shop1> Backtrace stopped: previous frame inner to this frame (corrupt stack?) 2019-10-27T09:56:28 < rue_shop1> erm 2019-10-27T09:56:37 < rue_shop1> https://paste.debian.net/1110971/ 2019-10-27T09:56:51 < rue_shop1> brb, I need another cup of tea... 2019-10-27T09:57:15 < antto> eye-oh-tea? 2019-10-27T09:57:27 < jpa-> rue_shop1: yeah, you are linking with LD 2019-10-27T09:57:33 < jpa-> no 2019-10-27T09:57:37 < jpa-> you're linking with gcc! 2019-10-27T09:57:47 < jpa-> so shouldn't be a multilib issue 2019-10-27T09:58:33 < rue_shop1> jpa-, if nothing else, I'm really glad you proved me right about the fpu thing 2019-10-27T09:59:05 < jpa-> yeah, try "frame 1" and "display /i $pc", because it seems the fpu usage comes from your main() instead of from some library 2019-10-27T09:59:25 < rue_shop1> should we reset it and clean the slate a bit? 2019-10-27T09:59:36 < jpa-> no 2019-10-27T09:59:41 < rue_shop1> (gdb) frame 1 2019-10-27T09:59:41 < rue_shop1> #1 2019-10-27T10:00:05 < jpa-> hum, i thought #1 was main() 2019-10-27T10:00:10 < rue_shop1> 1: x/i $pc 2019-10-27T10:00:10 < rue_shop1> => 0xfffffff9: movs r0, r0 2019-10-27T10:00:31 < rue_shop1> I put this in main to trip it: 2019-10-27T10:01:30 < rue_shop1> foo = 56.7; ... 2019-10-27T10:01:30 < rue_shop1> for (i = 0; i < 5.5 * foo; i++) { /* Wait a bit. */ 2019-10-27T10:01:58 < rue_shop1> it must have not optimized the calc, OR its trying to use a fp op in the compare 2019-10-27T10:02:42 < rue_shop1> knowing how to make a listing for it would be nice... 2019-10-27T10:03:35 < rue_shop1> oh objdump -D 2019-10-27T10:04:05 < jpa-> yeah, or just can also do "disasm main" in gdb 2019-10-27T10:04:15 < jpa-> (or was it "disas" 2019-10-27T10:04:52 < rue_shop1> disas 2019-10-27T10:04:59 < rue_shop1> sorry, I never got into gdb 2019-10-27T10:05:29 < rue_shop1> ok so, next level issue being that I dont know what arm fp instructions look like 2019-10-27T10:05:38 < jpa-> they start with 'v' i think 2019-10-27T10:05:51 < rue_shop1> 0x080001c2 <+42>: bl 0x8000904 <__floatunsidf> 2019-10-27T10:06:08 < jpa-> that's fine, it should call into the library softfloat implementation 2019-10-27T10:06:12 < rue_shop1> nothing starting with v 2019-10-27T10:06:22 < jpa-> pastebin the whole disassembly for main() 2019-10-27T10:06:27 < rue_shop1> iinteresting then 2019-10-27T10:07:07 < rue_shop1> https://paste.debian.net/1110987/ 2019-10-27T10:07:30 < rue_shop1> bl must be a branch... 2019-10-27T10:07:44 < jpa-> at some point you can also do this: "b main" to put breakpoint there, "mon reset halt" and "c" to start running, it should hit the breakpoint, then "layout asm" so you can see what it executes, then "stepi" to step single instruction, and keep pressing enter to see at which point exactly it crashes 2019-10-27T10:08:10 < jpa-> yeah, no floats there 2019-10-27T10:08:32 < jpa-> maybe it is in some library after all, though your linker line seems correct 2019-10-27T10:08:40 < rue_shop1> I wonder if its doing that thing I read about where it generates calls for the software library, but then uses the fpu 2019-10-27T10:08:54 < rue_shop1> lets fix the linker line 2019-10-27T10:09:09 < rue_shop1> I dont mind buckshot diagnostics 2019-10-27T10:09:54 < rue_shop1> oh correct 2019-10-27T10:10:10 < rue_shop1> great, I'm randomly inserting nots in what I read 2019-10-27T10:10:41 < rue_shop1> ok going for a breakpoint and reset 2019-10-27T10:11:56 < rue_shop1> will it tell me when it faults or will I just find myself stuck in the handler? 2019-10-27T10:13:05 < rue_shop1> :/ the opencm3 library is.... thick... 2019-10-27T10:18:03 < rue_shop1> geeesh, I really need to optimize my software serial thing 2019-10-27T10:19:47 < rue_shop1> jpa-, you still there? 2019-10-27T10:19:58 < rue_shop1> I'd like to change the source to get to the fault a bit quicker 2019-10-27T10:20:13 < rue_shop1> not sure how to get back to gdb here 2019-10-27T10:21:10 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d167:7a13:23d7:efc0] has joined ##stm32 2019-10-27T10:23:17 < rue_shop1> oh 2019-10-27T10:23:19 < jpa-> rue_shop1: you can type "finish" to get out of a function 2019-10-27T10:23:31 < jpa-> instead of stepping every line 2019-10-27T10:23:51 < rue_shop1> I just hit the float function 2019-10-27T10:24:06 < jpa-> is it using hard or soft floats? 2019-10-27T10:24:09 < rue_shop1> => 0x80011cc: stc2 10, cr4, [r7], {8} ; 2019-10-27T10:24:31 < jpa-> yeah, that's hard float stuff 2019-10-27T10:24:38 < rue_shop1> ok! 2019-10-27T10:24:57 < rue_shop1> so, gcc must have ignored my flag? 2019-10-27T10:25:16 < rue_shop1> LDFLAGS += -msoft-float -mfloat-abi=soft -march=armv7-m 2019-10-27T10:26:04 < jpa-> oh wait 2019-10-27T10:26:28 < jpa-> you're not specifying cpu for link step at all 2019-10-27T10:26:46 < jpa-> maybe just try LDFLAGS += $(CFLAGS) 2019-10-27T10:26:59 < jpa-> that's the easiest way to keep your ldflags and cflags in sync 2019-10-27T10:27:05 < rue_shop1> I'd added LDFLAGS += -msoft-float -mfloat-abi=soft -march=armv7-m 2019-10-27T10:27:08 < rue_shop1> at some point 2019-10-27T10:27:31 < rue_shop1> but I had a lot comming at me 2019-10-27T10:27:31 < jpa-> just -mcpu=cortex-m3 -mthumb would be enough 2019-10-27T10:27:42 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-27T10:28:20 < rue_shop1> ok let me try that out 2019-10-27T10:28:31 * rue_shop1 looks for the door.... 2019-10-27T10:28:34 < rue_shop1> finish 2019-10-27T10:29:32 < jpa-> arm-none-eabi-gcc -print-multi-lib can show what library options you have installed, with those linker flags it might be going for some non-thumb, hard-float with soft-abi library 2019-10-27T10:29:58 < rue_shop1> fpu;@mfloat-abi=hard 2019-10-27T10:30:01 < rue_shop1> :) 2019-10-27T10:30:21 < jpa-> that's the only option? 2019-10-27T10:30:58 < rue_shop1> no 2019-10-27T10:31:04 < rue_shop1> its the one that immediatly stands out to me 2019-10-27T10:31:16 < jpa-> it's not the one you want, though 2019-10-27T10:31:27 < jpa-> that command just lists everything your gcc supports 2019-10-27T10:31:48 < jpa-> and if your linker flags are correct, you should hope it picks up thumb/v7-m/nofp;@mthumb@march=armv7-m@mfloat-abi=soft or similar 2019-10-27T10:32:14 < jpa-> but because your LDFLAGS are missing -mcpu=cortex-m3 and -mthumb, it might not match to that 2019-10-27T10:32:18 < rue_shop1> oh now the programmer is playing up... 2019-10-27T10:32:39 < jpa-> https://github.com/PetteriAimonen/STM32_Trace_Example/blob/master/Makefile here's another makefile you can take a look at 2019-10-27T10:32:47 < rue_shop1> its ok, its only 1:30am here 2019-10-27T10:33:29 < rue_shop1> haha 2019-10-27T10:33:40 < rue_shop1> works better if I kill the *other* openocd thats running 2019-10-27T10:34:28 < rue_shop1> jpa 2019-10-27T10:34:30 < rue_shop1> you win 2019-10-27T10:34:36 < jpa-> you'll get the bugs out in no time with two debuggers 2019-10-27T10:34:44 < rue_shop1> can I offer you a prize? 2019-10-27T10:34:51 < rue_shop1> hahah 2019-10-27T10:34:56 < rue_shop1> two... 2019-10-27T10:35:24 < rue_shop1> ok, 2019-10-27T10:35:37 < rue_shop1> wow, that was painfull (till you got here) 2019-10-27T10:35:50 < jpa-> been there, done that 2019-10-27T10:36:03 < jpa-> it was very popular problem back in ~2011 2019-10-27T10:36:10 < rue_shop1> haha, I totally closed all the project files to debug that 2019-10-27T10:36:13 < jpa-> nowadays most people just get proper makefiles to start with ;) 2019-10-27T10:36:30 < rue_shop1> I couldn't find one! 2019-10-27T10:36:45 < jpa-> yeah, usually you find 100+ bad ones on google 2019-10-27T10:36:55 < rue_shop1> yea, I did that 2019-10-27T10:37:00 < jpa-> the ones from e.g. libopencm3 examples are fine though 2019-10-27T10:37:05 < jpa-> if you are going to use libopencm3, that is 2019-10-27T10:37:22 < rue_shop1> then took my avr makefile, stripped it down and rewrote it based on hand entered instructions from tutorials 2019-10-27T10:37:37 < rue_shop1> it looks like opencm3 isn't winning the race tho 2019-10-27T10:37:40 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 268 seconds] 2019-10-27T10:37:53 < rue_shop1> everyone seems to be using other things 2019-10-27T10:38:35 < jpa-> there are bazillion libraries, it's pure luck if any two people choose the same one and use it in the same way 2019-10-27T10:39:07 < rue_shop1> opencm3 is one where I find I have to reverse engineer it every time I go to do something 2019-10-27T10:39:17 < rue_shop1> which is working so far 2019-10-27T10:39:38 < rue_shop1> but stepping thru my software serial code was painfull 2019-10-27T10:39:43 < jpa-> looking into commented source code is not "reverse engineering" 2019-10-27T10:40:33 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-27T10:40:40 < rue_shop1> fair enough, but it would be nice to have basic application notes on use instead of opening up the library source each time your working out hwo to use a fn 2019-10-27T10:41:18 < antto> muh IDE makes that easy 2019-10-27T10:41:40 < jpa-> rue_shop1: like http://libopencm3.org/docs/latest/html/ ? 2019-10-27T10:42:16 < rue_shop1> the examples are.... ambigious to people who didn't write the library... 2019-10-27T10:42:38 < rue_shop1> there are a lot of docs for it and that is helping 2019-10-27T10:42:57 < rue_shop1> it would also help if I just cut myself off from using avrs for a while :) 2019-10-27T10:43:03 < jpa-> well, the examples are ambiguous to beginners 2019-10-27T10:43:13 < jpa-> but that's just the way most things are 2019-10-27T10:43:14 < rue_shop1> iirc when I was switching from BASIC to C it wasn't a clean transition 2019-10-27T10:43:36 < rue_shop1> omg, the jukebox played out 2019-10-27T10:43:43 * rue_shop1 hits random again 2019-10-27T10:44:55 < rue_shop1> oh, the tea machine auto turns off after 4 hours, better click that again too 2019-10-27T10:46:36 < rue_shop1> sweet, now I can plug in the DDS chip 2019-10-27T10:47:36 < Jan-> crikey 2019-10-27T10:47:42 < Jan-> another incarnation of rue :) 2019-10-27T10:48:40 < rue_shop1> we are many 2019-10-27T10:49:22 < rue_shop1> there aren't a lot of me, its just that the number of people in the universe is dwindling and the percentage of me is getting higher 2019-10-27T10:49:55 < rue_shop1> I'm outlasting everyone 2019-10-27T10:50:04 < rue_shop1> I use a mouse and a keyboard 2019-10-27T10:50:13 < rue_shop1> my mouse has a cord 2019-10-27T10:50:15 < rue_shop1> and a ball 2019-10-27T10:50:39 < rue_shop1> and my computer uses a copper wire to communicate with the internet 2019-10-27T10:50:44 < antto> i want one of those too 2019-10-27T10:50:46 < rue_shop1> 4 of them 2019-10-27T10:51:25 < Jan-> so does mine to be fair 2019-10-27T10:51:30 < rue_shop1> tho, my coffeemaker is http controlled via a radio link 2019-10-27T10:51:30 < Jan-> it's plugged in at the back 2019-10-27T10:51:43 < Jan-> I know that because the little latching tab on the ethernet cable has broken off and it is easy to accidentally pull it out. 2019-10-27T10:51:45 < antto> coffee maker huh 2019-10-27T10:51:54 < antto> i drink instant coffe 2019-10-27T10:51:56 < rue_shop1> it makes tea 2019-10-27T10:52:03 < rue_shop1> cause, I'm not that hardcore 2019-10-27T10:52:19 < Steffanx> rue_shop1: after you found out how to master stm32, can you teach Jan- ? 2019-10-27T10:52:30 < rue_shop1> sure 2019-10-27T10:52:37 < Steffanx> Awesome 2019-10-27T10:52:42 < rue_shop1> Jan-, have a bluepill? want to do hello world? 2019-10-27T10:52:43 < antto> where's muh popcr0nz 2019-10-27T10:52:57 < rue_shop1> and a usb-ttl serial adapter 2019-10-27T10:53:02 < rue_shop1> cause... yea 2019-10-27T10:53:38 < rue_shop1> can anyone tell me if that usb client CAN even emulate a serial port? 2019-10-27T10:54:01 < rue_shop1> cause I find it halarious to use a converter to talk to an stm32 2019-10-27T10:54:22 < rue_shop1> ok, I'm being distracted, where is my dds module? 2019-10-27T10:56:32 < PaulFertser> rue_shop1: yes, libopencm3 has an example implementing tty ACM on bluepill. 2019-10-27T10:57:19 < rue_shop1> hehe, I should have known that... 2019-10-27T10:57:28 < rue_shop1> :) 2019-10-27T11:00:38 < rue_shop1> so I'm building a filter profiler 2019-10-27T11:00:46 < rue_shop1> 1Hz to ~40khz 2019-10-27T11:01:42 < rue_shop1> something I can check the -3db of a mechanical servo with 2019-10-27T11:02:40 < Jan-> I need an expert in switching power supplies. 2019-10-27T11:02:47 < Jan-> ...who's cheap. 2019-10-27T11:03:12 < rue_shop1> try me 2019-10-27T11:03:22 < rue_shop1> I did them up for a while 2019-10-27T11:03:39 < rue_shop1> its all about filters 2019-10-27T11:04:12 < Jan-> it is indeed 2019-10-27T11:04:48 < Jan-> if you really don't mind, it's this https://imagebin.ca/v/4zpwW1ovwTTU 2019-10-27T11:05:13 < Jan-> the idea is that with pwm signals on inputs 5 (at the bottom left in the voltage section) you can change the output voltage. The problem is it reacts really slowly, it's laggy. 2019-10-27T11:05:17 < rue_shop1> hmm, pins are 5 0 k D C Y Y 2019-10-27T11:06:01 < rue_shop1> well look a dat filter! 2019-10-27T11:06:40 < rue_shop1> what size are the caps? 2019-10-27T11:06:58 < Jan-> hang on I need to get someone to look at the diagram for me 2019-10-27T11:07:09 < rue_shop1> I'm looking 2019-10-27T11:07:26 < Jan-> if it's not on the diagram we don't know 2019-10-27T11:07:28 < rue_shop1> but it does not have a value on the filter caps 2019-10-27T11:07:39 < rue_shop1> just says C 2019-10-27T11:07:58 < Jan-> we did try shorting out some of the resistors in the RC filters 2019-10-27T11:08:02 < Jan-> but it didn't make much difference 2019-10-27T11:08:09 < rue_shop1> not if the caps are huge, no 2019-10-27T11:08:27 < Jan-> they are tiny surface mount chip capacitors 2019-10-27T11:08:42 < rue_shop1> :) so maybe 10uF or less? 2019-10-27T11:09:28 < rue_shop1> another thing to check, is if the pwm is driving proprly 2019-10-27T11:09:40 < rue_shop1> make sure its driving to 3.3?V and ground properly 2019-10-27T11:09:45 < Jan-> basically we're unplugging the pwm 2019-10-27T11:09:48 < Jan-> so it goes high impedance 2019-10-27T11:09:53 < Jan-> to see how fast it will drop to zero 2019-10-27T11:10:04 < rue_shop1> at 1M, not fast 2019-10-27T11:10:17 < Jan-> is that R33 2019-10-27T11:10:24 < rue_shop1> yes 2019-10-27T11:10:39 < Jan-> we tried bridging that with 10k 2019-10-27T11:11:06 < rue_shop1> again, depending on the size of those caps, it could be slow 2019-10-27T11:11:31 < rue_shop1> the cap on the PI controller isn't marked either 2019-10-27T11:11:50 < rue_shop1> its like anthing with a filter value was left open-ended 2019-10-27T11:11:52 < Jan-> is that C10 or C14 2019-10-27T11:11:59 < Jan-> I can't remember one of them is in the voltage section and one is in the current 2019-10-27T11:12:01 < rue_shop1> 10 2019-10-27T11:12:12 < rue_shop1> but yea, same deal with 14 2019-10-27T11:12:15 < Jan-> again it's a little chip 2019-10-27T11:12:26 < rue_shop1> small chips can be up to 100uF 2019-10-27T11:13:04 < Jan-> it's annoying as this could work well for us but as an LED driver it would stop us doing things like strobe effects as it is too slow 2019-10-27T11:13:25 < rue_shop1> its slow so that its low ripple 2019-10-27T11:13:35 < rue_shop1> if you can take some ripple, you can lighten up on the caps 2019-10-27T11:13:58 < Jan-> we can increase the PWM frequency too. 2019-10-27T11:14:00 < rue_shop1> 3rd order filter is kinda ... heavy 2019-10-27T11:14:04 < rue_shop1> no 2019-10-27T11:14:16 < rue_shop1> it wont make a difference to the filter 2019-10-27T11:14:24 < Jan-> no it won't but it would reduce the ripple with a lighter filter 2019-10-27T11:14:33 < rue_shop1> the filters cutoff willbe way below the pwm 2019-10-27T11:14:47 < rue_shop1> yes 2019-10-27T11:14:54 < rue_shop1> see why I'm making a filter profiler? 2019-10-27T11:15:00 < Jan-> all we can do is remove the capacitors I guess 2019-10-27T11:15:13 < Jan-> we'd probably destroy them getting them off so it would be one way tricky 2019-10-27T11:15:26 < Jan-> er, trick too 2019-10-27T11:15:29 * Jan- cannot type today 2019-10-27T11:15:35 < rue_shop1> hot tweezers? 2019-10-27T11:15:46 < rue_shop1> are you jan- of #robotics? 2019-10-27T11:15:59 < Jan-> no, I'm Jan of a small town northeast of London. 2019-10-27T11:16:03 < rue_shop1> electronics or avr? 2019-10-27T11:16:12 < rue_shop1> its a familiar nick 2019-10-27T11:16:13 < Jan-> I hang out in electronics, we've met before :) 2019-10-27T11:16:22 < Jan-> you're rue_mohr, right 2019-10-27T11:16:23 < rue_shop1> oo long time ago 2019-10-27T11:16:34 < rue_shop1> that channel got stupid.... 2019-10-27T11:16:38 < rue_shop1> :) 2019-10-27T11:16:42 * Jan- hugs :) 2019-10-27T11:16:48 < rue_shop1> we started #robotics 2019-10-27T11:17:35 < antto> skynet 2019-10-27T11:17:52 < rue_shop1> yea, I dont know if your loop is going to have enough speed to keep the converter under control 2019-10-27T11:18:15 < Jan-> *sigh* 2019-10-27T11:18:26 < rue_shop1> I would suggest adjusting the natural feedback of it 2019-10-27T11:18:29 < Jan-> we've been looking around for a solution to this for literally years. 2019-10-27T11:18:41 < rue_shop1> ... years?.... 2019-10-27T11:18:45 < Jan-> on and off yes 2019-10-27T11:18:47 < Jan-> it keeps coming up 2019-10-27T11:18:49 < rue_shop1> what is the voltage/current of the leds? 2019-10-27T11:18:51 < Jan-> getting big banks of high power LEDs under microcontroller authority 2019-10-27T11:19:15 < Jan-> it's a series array of eight three-watt white emitting LEDs so they're about 3V each 2019-10-27T11:19:22 < rue_shop1> were not talking 3.3A at 94V 2019-10-27T11:19:31 < Jan-> and they draw about 0.3 amps at 24v ish 2019-10-27T11:19:36 < rue_shop1> yea 2019-10-27T11:19:39 < rue_shop1> k 2019-10-27T11:19:42 < Jan-> it's a test array 2019-10-27T11:19:46 < Jan-> in the end we would expect to use more 2019-10-27T11:20:19 < rue_shop1> so ~24V, 300mA 2019-10-27T11:20:41 < Jan-> In the final device we'd basically load it up until the converters ran out of oomph 2019-10-27T11:20:49 < Jan-> which is going to be a couple of amps at 36 volts 2019-10-27T11:21:10 < Jan-> the manufacturer says 3 or 4 amps but china quality so. 2019-10-27T11:21:21 < rue_shop1> thats a great circuit you have, but I dont think its your golden untitled goose game 2019-10-27T11:21:33 < rue_shop1> indeed 2019-10-27T11:21:42 < rue_shop1> is your supply 24V? 2019-10-27T11:21:52 < rue_shop1> 36? 2019-10-27T11:22:06 < Jan-> it needs to be a bit above the output 2019-10-27T11:22:08 < Jan-> it's a buck converter 2019-10-27T11:22:18 < rue_shop1> boost converters are really nice for led current control 2019-10-27T11:22:42 < Jan-> really we would have wanted that so we could go up to say 48v 2019-10-27T11:22:46 < Jan-> but this is what's on the market 2019-10-27T11:23:21 * rue_shop1 gets his book 2019-10-27T11:24:17 < rue_shop1> power supplies I think... 2019-10-27T11:24:58 < rue_shop1> hmm thats buck... 2019-10-27T11:25:49 < Jan-> if it helps 2019-10-27T11:25:50 < Jan-> https://imagebin.ca/v/4zq2ovQyLXeE 2019-10-27T11:26:09 < rue_shop1> I have a few supplies I did for things 2019-10-27T11:26:11 < rue_shop1> current control 2019-10-27T11:26:32 < Jan-> we would happily just have voltage control and put the current control in the software 2019-10-27T11:26:51 < rue_shop1> did you guys reverse engineer that and thats why none of the small caps have values? 2019-10-27T11:26:54 < rue_shop1> :) 2019-10-27T11:26:59 < Jan-> we didn't, someone did 2019-10-27T11:27:07 < Jan-> the problem is nobody here is a power supply expert and we do not want to get involved in basically anything that starts involvig coils and stuff. 2019-10-27T11:27:21 < Jan-> that board is off the shelf and works as far as it goes. 2019-10-27T11:27:27 < rue_shop1> I suppose you dont have a breadboard, an LM555 and a few discretes eh? 2019-10-27T11:27:38 < Jan-> we probably do but how would that help 2019-10-27T11:27:46 < rue_shop1> well i have this circuit here... 2019-10-27T11:27:54 < rue_shop1> heh 2019-10-27T11:28:20 < rue_shop1> hmm 2019-10-27T11:29:22 < rue_shop1> oh I see why it stays under control 2019-10-27T11:29:41 < rue_shop1> ooh ok 2019-10-27T11:29:52 < rue_shop1> your thing is better than it first appeared to be 2019-10-27T11:30:06 < rue_shop1> you know 2019-10-27T11:30:20 < Jan-> is it? 2019-10-27T11:30:21 < rue_shop1> you COULD just put a DAC on the PI filter input 2019-10-27T11:30:25 < rue_shop1> PI controller 2019-10-27T11:30:31 < Jan-> that was discussed 2019-10-27T11:30:38 < Jan-> we actually bought dac chips 2019-10-27T11:30:40 < rue_shop1> how many level do you want? 2019-10-27T11:30:44 < rue_shop1> what dac chips? 2019-10-27T11:30:51 < Jan-> can't remember but they're 16 bit 2019-10-27T11:31:00 < rue_shop1> k, might not be suitable 2019-10-27T11:31:02 < Jan-> I would describe that as "sufficient" :) 2019-10-27T11:31:11 < Jan-> jesus you want more? 2019-10-27T11:31:16 < rue_shop1> not if their what I think 2019-10-27T11:31:31 < Jan-> I do have an stm32f407 2019-10-27T11:31:33 < rue_shop1> I think their chips that center at 2.5V and only have a 2.5V range 2019-10-27T11:31:34 < Jan-> that has dac outputs 2019-10-27T11:31:41 < rue_shop1> thats probably better 2019-10-27T11:31:42 < Jan-> nono they're not the audio ones. 2019-10-27T11:31:47 < Jan-> didn't get the audio ones 2019-10-27T11:31:48 < rue_shop1> cause they will be full 0-3.3V range 2019-10-27T11:31:49 < Jan-> because that 2019-10-27T11:31:54 < rue_shop1> ok 2019-10-27T11:32:25 < rue_shop1> so then replace the lopass resisors with jumpers, and remove the caps 2019-10-27T11:32:46 < rue_shop1> tie to R35, you dont need to control the votlage, just the current 2019-10-27T11:33:02 < Jan-> actually we probably do need to control the voltage 2019-10-27T11:33:10 < rue_shop1> leds are current operated 2019-10-27T11:33:14 < rue_shop1> their voltage changes with temp 2019-10-27T11:33:16 < Jan-> in theory. 2019-10-27T11:33:25 < rue_shop1> I'v never seen otherwise 2019-10-27T11:33:33 < Jan-> in reality most of these things will not actually achieve zero output when you set zero current if voltage is set high 2019-10-27T11:33:42 < rue_shop1> you do want a limit if an led goes open circuit 2019-10-27T11:33:50 < Jan-> you have to track the voltage down to keep it a bit above what you think it will need 2019-10-27T11:33:56 < Jan-> I've written that algorithm before 2019-10-27T11:34:09 < rue_shop1> odd wonder if thats the driver 2019-10-27T11:34:14 < Jan-> it's because these power supplies are never ideal 2019-10-27T11:37:31 < Jan-> OK so the question is you're saying jump out R30-R33 and remove C15-C17 2019-10-27T11:37:44 < rue_shop1> and use the dac 2019-10-27T11:37:51 < rue_shop1> is a fine solution 2019-10-27T11:38:18 < Jan-> Or may be leave in one stage of the filter and use faster pwm 2019-10-27T11:38:22 < Jan-> the pwm right now is about 2khz 2019-10-27T11:38:46 < rue_shop1> it would help to know what the small caps are 2019-10-27T11:38:53 * Jan- ascii shrug smiley 2019-10-27T11:38:55 < Jan-> no idea sorry 2019-10-27T11:38:57 < Jan-> no way to tell 2019-10-27T11:39:05 < rue_shop1> cap meter, but yea 2019-10-27T11:39:18 < Jan-> we have one but it is basic and it gives no reading while they are in circuit 2019-10-27T11:39:26 < rue_shop1> I went thru about 200 smt caps stripped off boards to find two 0.02uF caps 2019-10-27T11:39:37 < rue_shop1> decided it was not ever worth it to do that agian 2019-10-27T11:40:07 < rue_shop1> then the analog filter didn't work anyhow 2019-10-27T11:40:44 < rue_shop1> and I decided analog sucks 2019-10-27T11:40:51 < Jan-> that's what I decided ages ago 2019-10-27T11:40:58 < Jan-> that's why we really really really don't want to futz with this damn thing 2019-10-27T11:41:13 < rue_shop1> you said it was off the shelf 2019-10-27T11:41:21 < rue_shop1> so take one and develop with it 2019-10-27T11:41:38 < Jan-> yeah and then if we want to make dozens of them we have to brew our own board. 2019-10-27T11:41:49 < Jan-> and it's a high speed switching power supply 2019-10-27T11:41:57 < Jan-> so that will be a ten-respin epic tale of pain and horror 2019-10-27T11:42:12 -!- jly [uid355225@gateway/web/irccloud.com/x-hsptkbfafjgowkyg] has joined ##stm32 2019-10-27T11:42:17 < rue_shop1> ok 2019-10-27T11:42:22 < rue_shop1> then use those chineese modules 2019-10-27T11:42:28 < rue_shop1> do you know the ones? 2019-10-27T11:42:38 < Jan-> this is a chinese module :) 2019-10-27T11:42:44 < rue_shop1> no 2019-10-27T11:43:19 < rue_shop1> https://www.aliexpress.com/item/32972024049.html 2019-10-27T11:43:23 < Jan-> we had some other ones which had serial control, or at least they could if you reflashed them with third party firmware 2019-10-27T11:43:23 < rue_shop1> see the two pots? 2019-10-27T11:43:40 < rue_shop1> one of them is for current, the other for votage (if I grabbed the right one) 2019-10-27T11:43:55 < Jan-> I can't see pics, but I know what you mean 2019-10-27T11:44:06 < rue_shop1> oooh yea jan 2019-10-27T11:44:17 < rue_shop1> gee, I got you and kat mixed up 2019-10-27T11:44:17 * Jan- doffs hat 2019-10-27T11:44:31 < rue_shop1> you can see a bit tho, right? 2019-10-27T11:44:55 < Jan-> not to any practical use. 2019-10-27T11:45:08 < rue_shop1> ok, their little dc-dc modules from china 2019-10-27T11:45:15 < Jan-> I know what you mean. 2019-10-27T11:45:16 < rue_shop1> pot adjustable votlage and current 2019-10-27T11:45:27 < Jan-> we did think about motorised pots 2019-10-27T11:45:30 < Jan-> seriously, we were that desperate 2019-10-27T11:45:33 < rue_shop1> if you want, switch 'em out for digital pots 2019-10-27T11:45:42 < Jan-> we tried this 2019-10-27T11:45:47 < rue_shop1> oh!? 2019-10-27T11:45:50 < Jan-> some of those little modules become really flickery and unstable 2019-10-27T11:45:57 < rue_shop1> hmm 2019-10-27T11:46:03 < Jan-> basically you're dividing down the feedback 2019-10-27T11:46:06 < rue_shop1> did you have them heatsinked? 2019-10-27T11:46:09 < Jan-> and that is a really twitchy thing 2019-10-27T11:46:29 < rue_shop1> yea, and leds are quite sensitive to wobble 2019-10-27T11:47:24 < rue_shop1> my dds isn't working on the stm32 2019-10-27T11:47:26 < rue_shop1> :( 2019-10-27T11:47:57 < Jan-> dds? 2019-10-27T11:48:30 < rue_shop1> direct digital synthesis 2019-10-27T11:48:34 < rue_shop1> its a freq gen 2019-10-27T11:48:59 < rue_shop1> I'm porting my filter profiler from avr to stm32 2019-10-27T11:49:22 < rue_shop1> cause none of the avrs I have, have the timing capture channels I need 2019-10-27T11:49:32 < rue_shop1> or want 2019-10-27T11:49:33 < Jan-> oh right 2019-10-27T11:49:42 < Jan-> that's basically why I got this stm32 board 2019-10-27T11:49:43 < rue_shop1> as I'm avoiding excessive design creativity 2019-10-27T11:49:44 < Jan-> more fasterness. 2019-10-27T11:50:13 < rue_shop1> I need to hardware capture 2 channels of zero crossings 2019-10-27T11:50:22 < rue_shop1> avr can only do 1 2019-10-27T11:50:29 < rue_shop1> stm32 can do 12 2019-10-27T11:50:59 < rue_shop1> its another project to battle my way into using stm32 instead of avr 2019-10-27T11:51:15 < rue_shop1> and I'm breaking all sorts of problems 2019-10-27T11:51:41 < rue_shop1> things like, "gee I didn't know I couldn't do floating point on these" 2019-10-27T11:51:49 < Jan-> I think the stm32 stuff itself is fine 2019-10-27T11:51:56 < Jan-> I am finding the documentation is beyond awful 2019-10-27T11:52:15 < rue_shop1> its like going from programming in BASIC to C 2019-10-27T11:52:23 < rue_shop1> everything is manual and there is a lot of it 2019-10-27T11:52:59 < rue_shop1> its 3am, I wonder If I'm gonna just crash suddenly 2019-10-27T11:53:48 < rue_shop1> tho, it looks like I stopped doing things that actaully require thinking 2019-10-27T11:54:01 * Jan- tucks rue_mohr up with blanket and teddy 2019-10-27T11:56:25 -!- fenugrec [~fenugrec@47.54.195.193] has joined ##stm32 2019-10-27T11:57:01 < rue_shop1> ok I think the two clocks are the right way around now 2019-10-27T11:57:59 < Jan-> I don't think stm32 is actually that much more complicated in theory 2019-10-27T11:58:04 < Jan-> there's more of it but it's not complicated 2019-10-27T11:58:11 < Jan-> it's just that there are 101 different libraries for it 2019-10-27T11:58:17 < Jan-> and none of them seem to be very well written up 2019-10-27T11:58:40 < rue_shop1> well, having to turn on perphial clocks is a new thing 2019-10-27T11:58:55 < antto> just wind it up 2019-10-27T11:58:56 < rue_shop1> I had a great time trying to get an io working that didn't have its clock source turned on 2019-10-27T12:00:04 < rue_shop1> do you remember going form BASIC to C 2019-10-27T12:00:17 < rue_shop1> and all that extra stuff you had to put in to make things work? 2019-10-27T12:02:10 < Steffanx> BASIC? 2019-10-27T12:02:24 < rue_shop1> remember, C64, Apple IIe, DOS? 2019-10-27T12:03:01 < rue_shop1> "the sinclare and the apple II, did what a computer was supposed to do" 2019-10-27T12:03:03 < Jan-> I don't mind how complicated things get 2019-10-27T12:03:09 < Jan-> so long as there's a decent explanation with examples 2019-10-27T12:03:14 < rue_shop1> "and you didn't need a fancy operating system" 2019-10-27T12:04:03 < Jan-> there is a realtime OS for stm32 I think 2019-10-27T12:04:10 < rue_shop1> of course there is 2019-10-27T12:04:19 < rue_shop1> its just that every OS sucks 2019-10-27T12:04:23 < rue_shop1> (and blows) 2019-10-27T12:04:59 < Jan-> I'm not sure that a quarter of a megabyte of ram is really enough to be honest 2019-10-27T12:05:02 < Jan-> and there's no memory protection 2019-10-27T12:05:08 < Jan-> so a rogue task can take out the whole thing 2019-10-27T12:05:26 < Jan-> (or is there? I've been assuming there isnt) 2019-10-27T12:05:28 < rue_shop1> 64k, it got them to the moon, it should be enough for you 2019-10-27T12:05:38 < Jan-> yeah but they didn't try to port fucking linux to it# 2019-10-27T12:05:43 * Jan- gripes 2019-10-27T12:05:53 < rue_shop1> :) 2019-10-27T12:06:32 < rue_shop1> to the person reading the logs years later, they aren't picking up the reference 2019-10-27T12:07:10 < rue_shop1> oooh 2019-10-27T12:07:24 < rue_shop1> what did I break 2019-10-27T12:08:00 < Jan-> anyway didn't it have like 2k of ram 2019-10-27T12:08:03 < Jan-> and the rest was rom 2019-10-27T12:08:08 < Jan-> an apollo guidance computer 2019-10-27T12:08:21 < rue_shop1> ferrite rope memory? 2019-10-27T12:08:34 < Jan-> ferrite core memory 2019-10-27T12:08:40 < Jan-> you had to weave the firmware like fabric 2019-10-27T12:08:48 < Jan-> then it was set it blocks of epoxy because vibration 2019-10-27T12:09:30 < rue_shop1> but unlike the usb drive I bought last year, the AGC still works 2019-10-27T12:09:31 < Jan-> ok, so actually it was more like 4K as it was 2048 words and each word was 16 bits. 2019-10-27T12:09:41 < Jan-> only really 15 bits plus parity 2019-10-27T12:09:42 < rue_shop1> bigger than 16 wasn't it? 2019-10-27T12:09:46 < rue_shop1> hmm 2019-10-27T12:10:17 < rue_shop1> oh maybe it was something about them putting multiple words on the same address and selecting whichone they wanted 2019-10-27T12:12:10 < Jan-> my understanding is that it worked mostly like we'd expect a modern computer to work 2019-10-27T12:13:03 < rue_shop1> have you ever investigated the 4004 datasheet? 2019-10-27T12:13:39 < Jan-> no. 2019-10-27T12:13:52 < rue_shop1> not a cpu like we know it today 2019-10-27T12:14:01 < Jan-> the AGC had registers and stuff 2019-10-27T12:14:05 < Jan-> I'm not sure if it had a stack 2019-10-27T12:14:30 < rue_shop1> I'm really starting to have challanges operating 2019-10-27T12:14:35 < rue_shop1> its Pi AM 2019-10-27T12:14:51 < Jan-> do we need to stretcher you through to your bedroom and tuck you up with an extra blanket for 10 straight hours of nap time 2019-10-27T12:15:21 < rue_shop1> but I want to get my project milestone 2019-10-27T12:15:31 < rue_shop1> I need the freq gen to work from stm32 2019-10-27T12:15:56 < Jan-> sleep on it 2019-10-27T12:16:04 < rue_shop1> I must be close tho 2019-10-27T12:16:10 < rue_shop1> I have serial data comming out the chip 2019-10-27T12:16:20 < rue_shop1> the thing just isn't doing the thing 2019-10-27T12:17:31 < Jan-> I'm really sorry I can't be more help 2019-10-27T12:17:33 < Jan-> I'm a bit stupid really 2019-10-27T12:18:14 < rue_shop1> your fine, a person would need to see a lot of this project for anything to make sense 2019-10-27T12:19:01 < Jan-> that's not even the issue 2019-10-27T12:20:15 < rue_shop1> wait 2019-10-27T12:20:17 < rue_shop1> was ... 2019-10-27T12:20:23 < rue_shop1> was I in the right project folder? 2019-10-27T12:20:50 < rue_shop1> wtf 2019-10-27T12:21:07 < rue_shop1> what have I been flashing to this controller for the past 20 minutes???? 2019-10-27T12:23:35 < rue_shop1> is... 2019-10-27T12:23:48 < rue_shop1> is stm32 opposite endian than avr? 2019-10-27T12:24:45 < rue_shop1> er woudl that even matter here 2019-10-27T12:26:19 < Jan-> I asked that 2019-10-27T12:26:23 < Jan-> never got a satisfatcory answer 2019-10-27T12:26:28 < Jan-> I'm really sorry I can't be more hel 2019-10-27T12:26:41 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-27T12:26:46 < rue_shop1> its ok, I think I'v hit me limit for tongiht 2019-10-27T12:26:57 < rue_shop1> no 1khz winesave 2019-10-27T12:27:03 < PaulFertser> rue_shop1: stm32 is little endian 2019-10-27T12:27:03 < rue_shop1> sinewave 2019-10-27T12:27:17 < rue_shop1> for( temp = 0x8000; temp != 0; temp >>= 1) 2019-10-27T12:27:26 < rue_shop1> I'm software shift ing bits out 2019-10-27T12:27:34 < rue_shop1> it shouldn't matter tho, right? 2019-10-27T12:27:57 < rue_shop1> only if I'm doing consecutive memory bytes 2019-10-27T12:28:10 < PaulFertser> Indeed 2019-10-27T12:28:25 < rue_shop1> the code is just like the avr 2019-10-27T12:28:28 < rue_shop1> and its not working 2019-10-27T12:28:43 < rue_shop1> the pulses are hapening 2019-10-27T12:28:49 < rue_shop1> the timing looks right 2019-10-27T12:29:04 < rue_shop1> no life from perphial 2019-10-27T12:29:41 < PaulFertser> rue_shop1: are you checking with an LA or other way to sniff data between the stm32 and peripheral? 2019-10-27T12:29:42 < Jan-> I'm pondering (again) whether we have any chance of making a board that does what we want here. 2019-10-27T12:29:58 < rue_shop1> Jan-, I recommand dac 2019-10-27T12:30:05 < Jan-> yes 2019-10-27T12:30:11 < rue_shop1> I did a LA capture, it looks good 2019-10-27T12:30:12 < rue_shop1> so 2019-10-27T12:30:23 < Jan-> but even if we can get that to work (by totally wrecking this existing board) we then might have to make our own 2019-10-27T12:30:25 < rue_shop1> unless I got my bits backwards 2019-10-27T12:30:30 < Jan-> and the LM series power supply controllers are twitchy as hell 2019-10-27T12:30:32 < rue_shop1> oh, I wonder about startup delay 2019-10-27T12:31:09 < rue_shop1> oh my bad leg is starting to twitch that means I AM done 2019-10-27T12:31:31 * Jan- makes rue_shop1 a hot cocoa 2019-10-27T12:31:33 < Jan-> bedtime rue 2019-10-27T12:31:51 < rue_shop1> one more thing 2019-10-27T12:31:54 < rue_shop1> a startup elay 2019-10-27T12:31:56 < rue_shop1> delay 2019-10-27T12:32:14 < rue_shop1> nope 2019-10-27T12:38:09 < rue_shop1> I'll have to capture a working packet tommorow and compare 2019-10-27T12:43:12 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-27T12:47:07 -!- fenugrec [~fenugrec@47.54.195.193] has quit [Ping timeout: 265 seconds] 2019-10-27T13:44:31 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-27T14:00:49 < karlp> cute, curvy traces https://www.aliexpress.com/item/4000103610226.html?spm=a2g0s.9042311.0.0.1fbc4c4dDiS2fQ 2019-10-27T14:00:59 < karlp> nice to see bluepill finally getting proper alternatives 2019-10-27T14:08:53 < Steffanx> alternatives for the bluepill you mean? 2019-10-27T14:09:21 < karlp> yeah 2019-10-27T14:09:28 < Jan-> hey karlp how's it going 2019-10-27T14:09:32 < specing> Are there any RF stm32? 2019-10-27T14:09:40 < Steffanx> yes 2019-10-27T14:09:41 < Steffanx> the wb 2019-10-27T14:09:47 < specing> the wb? 2019-10-27T14:10:05 < Steffanx> stm32wb 2019-10-27T14:10:18 < karlp> Jan-:eh, better than yesterday, can't complain :) 2019-10-27T14:10:26 < Jan-> why do I get ru.aliexpress.com 2019-10-27T14:10:29 < Jan-> it does this all the time 2019-10-27T14:10:43 < Steffanx> it knows you're a russian spu 2019-10-27T14:10:43 < Steffanx> y 2019-10-27T14:11:09 < Jan-> pizdets :/ 2019-10-27T14:11:33 < Jan-> is that like a blue pill with a 400 series stm32 on it 2019-10-27T14:11:57 < specing> China says 100 Mhz, STM says 84 MHz 2019-10-27T14:12:48 < Jan-> what's the difference between F401 and F407 2019-10-27T14:12:51 < Jan-> the board I have has F407 2019-10-27T14:16:57 < Steffanx> that's the f411, specing 2019-10-27T14:17:08 < Steffanx> f411 is the 2nd option which is 100 2019-10-27T14:18:27 < specing> f401 has boring peripherals 2019-10-27T14:18:33 < specing> Steffanx: ah ok 2019-10-27T14:19:29 < specing> F411 has boring peripherals, too 2019-10-27T14:20:50 < Steffanx> boring peripherals, lol 2019-10-27T14:21:29 < Steffanx> with Ada nothing is boring. 2019-10-27T14:21:56 -!- jly [uid355225@gateway/web/irccloud.com/x-hsptkbfafjgowkyg] has quit [Quit: Connection closed for inactivity] 2019-10-27T14:22:43 < specing> it doesen't have anything that my stm32f072 doesen't already have 2019-10-27T14:23:53 < Steffanx> except for a m4 core with fpu 2019-10-27T14:23:56 < Steffanx> and a few megahurtz more 2019-10-27T14:24:20 < specing> Now stm32wbx5 has non-boring peripherals 2019-10-27T14:24:36 < specing> Steffanx: yeah, and thats boring 2019-10-27T14:24:50 < Steffanx> oh i forgot ada is super effecient, sorry :) 2019-10-27T14:24:58 < specing> i'm not trying to mine bitcoin on stm32 2019-10-27T14:25:28 < PaulFertser> Hm, I do not remember pizdets word used here before. Thank you Jan- for spreading the knowledge. 2019-10-27T14:26:05 < karlp> specing:at least f4 hhas modern gpios.... 2019-10-27T14:26:15 < Steffanx> not that many russians here i think, PaulFertser 2019-10-27T14:26:34 < specing> karlp: is that like ... modern art? 2019-10-27T14:27:04 < specing> no? Isn't IRC full of slavs in general and russians in particular? 2019-10-27T14:27:15 < Jan-> da, tovarisch 2019-10-27T14:27:22 < PaulFertser> Steffanx: yep, but "suka blyat" seems to be an international curse now, so that wouldn't surprise me. But Jan- knows more than that. 2019-10-27T14:27:37 < Jan-> I have a bulgarian friend 2019-10-27T14:27:37 < specing> pozdravljena, tovarišica Jan- 2019-10-27T14:27:40 < Jan-> but she curses in russian 2019-10-27T14:28:27 < antto> we got pretty gud curses 2019-10-27T14:28:29 < Jan-> I can sing in Russian if you'd like! 2019-10-27T14:28:40 * Jan- sings "Katyusha" in very bad Russian 2019-10-27T14:29:08 < Steffanx> only because its all you hear the avarage russian say, PaulFertser :P 2019-10-27T14:29:49 < Jan-> antto: where are you from :) 2019-10-27T14:29:53 < antto> .bg 2019-10-27T14:30:00 < Jan-> ah right 2019-10-27T14:30:32 < Steffanx> .bg is still just a russian country outside russia right? 2019-10-27T14:30:35 < Jan-> don't you just add "ebasi" in front of everything 2019-10-27T14:30:41 < Steffanx> could call it a colony, but its different. 2019-10-27T14:30:41 < antto> Steffanx >:( 2019-10-27T14:30:46 < antto> pls 2019-10-27T14:30:58 < Jan-> ooh Steffanx 2019-10-27T14:30:59 < Jan-> never say that 2019-10-27T14:31:01 < Jan-> they get very upset 2019-10-27T14:31:02 < antto> Jan- not really 2019-10-27T14:31:21 < Jan-> when my friend calls her buddies back home and speaks bulgarian to them, I always tease her by saying it sounds like Russian 2019-10-27T14:31:25 < Jan-> she doesn't like that :D 2019-10-27T14:31:37 < antto> i don't understand russian 2019-10-27T14:31:42 < Steffanx> it looks like russian and sounds like it for an outsider. 2019-10-27T14:32:04 < Jan-> "da" is still "da" 2019-10-27T14:32:32 < Steffanx> njet is still njet? 2019-10-27T14:32:34 < antto> it sounds russian when a russian sez it 2019-10-27T14:32:46 < Jan-> anyway I have to go 2019-10-27T14:32:55 < Jan-> we're going out for lunch and I haven't even showered yet :D 2019-10-27T14:32:57 < antto> no such word here 2019-10-27T14:33:24 < Jan-> later all 2019-10-27T14:35:17 < Steffanx> how do you say "no" antto? 2019-10-27T14:35:23 < antto> не 2019-10-27T14:36:42 < Steffanx> aha 2019-10-27T14:45:15 < PaulFertser> antto: but do you folks in .bg nod your head meaning no? 2019-10-27T14:46:10 < antto> nod? 2019-10-27T14:46:46 < PaulFertser> Yes, I remember something like that, people get confused when you turn your heard left-right meaning yes and up-down meaning no. 2019-10-27T14:46:47 < antto> is that teh vertical movement? 2019-10-27T14:47:20 < antto> okay, yeah that's very popular about .bg but IMO it's wrong 2019-10-27T14:47:45 < antto> the axis of movement doesn't matter that much 2019-10-27T14:47:58 < antto> i think i nod vertically for both yes and no 2019-10-27T14:47:59 < Steffanx> so what is the proper russian movement when to say "no", PaulFertser? 2019-10-27T14:48:24 < antto> when it's a "no" it's just moving the head up 2019-10-27T14:48:46 < Steffanx> nono, you move your head from the left to the right and back :P 2019-10-27T14:48:48 < PaulFertser> Steffanx: left-right in arbitrary order arbitrary number of times 2019-10-27T14:48:48 < antto> for yes - downwards 2019-10-27T14:49:06 < antto> that could mean yes or no 2019-10-27T14:49:10 < antto> the thing is, it doesn't matter 2019-10-27T14:49:20 < antto> we use different things for the cue 2019-10-27T14:49:25 < Steffanx> In dutchland it matters much. 2019-10-27T14:49:25 < antto> not the akchual movement 2019-10-27T14:49:33 < Steffanx> when i dont want to talk i just shake or nod. 2019-10-27T14:49:43 < Steffanx> depending on no or yes. 2019-10-27T14:51:37 * Steffanx sends antto 2019-10-27T14:52:48 -!- X230t [x13@gateway/shell/suchznc/x-pfxydcgfgwkgfcso] has quit [Remote host closed the connection] 2019-10-27T14:57:22 < antto> Steffanx i mean... i watch american movies and shows and what not.. it have never noticed how they nod their head for yes/no nor have had my brain blink for "conflict" between the visuals and speech 2019-10-27T14:57:35 < antto> basically, the head nodding is low priority 2019-10-27T14:57:46 < Steffanx> im not sure if that's how it works in dutchland 2019-10-27T14:58:28 < antto> what i *do* notice is the head movements in these silly bollywood TV shows 2019-10-27T14:58:31 < antto> so annoying 2019-10-27T14:58:41 < antto> but oh well 2019-10-27T14:58:48 < Steffanx> illl try to see it often we dutchies do it 2019-10-27T14:58:55 < Steffanx> ill report later this week 2019-10-27T15:03:28 < Steffanx> unless you're week starts at a monday, then its later next week 2019-10-27T15:33:47 < antto> https://www.youtube.com/watch?v=IIvGwbW9Qk4 >:) 2019-10-27T15:43:35 -!- deskwizard [~quassel@unaffiliated/deskwizard] has joined ##stm32 2019-10-27T16:08:16 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-27T16:29:33 < PaulFertser> Why would a working week start with a day off? 2019-10-27T16:31:35 -!- kow__ [~iccy@135.0.26.39] has joined ##stm32 2019-10-27T16:33:05 -!- kow_ [~iccy@135.0.26.39] has quit [Ping timeout: 276 seconds] 2019-10-27T16:39:37 < Steffanx> i didnt say working week, PaulFertser 2019-10-27T16:39:55 < Steffanx> i wouldnt mind having a working week that stats with a day of 2019-10-27T16:40:06 < Steffanx> and perhaps the 2nd day, 3rd ... the 5th :P 2019-10-27T16:43:16 <@englishman> 4 day work weeks here are p common 2019-10-27T16:50:43 < bitmask> man I'm glad I decided to start over, this is looking much better 2019-10-27T16:51:52 < mawk> I will work from 8h30 to 18h Steffanx , is this common 2019-10-27T16:52:26 < Steffanx> 8.30-17h more more common. unless you have like a 1.5 hour break. 2019-10-27T16:53:04 < Steffanx> *is more 2019-10-27T16:54:13 < Steffanx> wouldnt it be more than 8 hours/day? 2019-10-27T16:55:07 < mawk> probably yeah 2019-10-27T16:55:11 < mawk> they said "full time" 2019-10-27T16:56:02 < Steffanx> unless friday is a very short day 2019-10-27T16:56:09 < Steffanx> otherwise it would be more than 40/week 2019-10-27T16:56:15 < Steffanx> which is pretty uncommon 2019-10-27T16:56:48 < mawk> « 5.1. Unless agreed otherwise or discussed to mutual agreement, the working hours run from 8:30 AM to 6:00 PM. The working week runs from Monday to Friday. » 2019-10-27T16:58:59 < Steffanx> but i assume there is also numbers of hours/week mentioned somewhere? 2019-10-27T17:04:17 < Steffanx> i think it cant be more than 40 hours/week. afaik there is some limit which limits it to 40 hours over a span of X weeks 2019-10-27T17:09:17 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Excess Flood] 2019-10-27T17:09:43 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-27T17:10:09 < Steffanx> didnt you check on this before you signed it, mawk? :P 2019-10-27T17:22:59 < mawk> it's not signed yet Steffanx 2019-10-27T17:23:38 < mawk> but here with my school I was used to working from 8am to 1am, so 8:30am to 6pm is vacations 2019-10-27T17:23:43 -!- deskwizard [~quassel@unaffiliated/deskwizard] has left ##stm32 ["http://quassel-irc.org - Chat comfortably. Anywhere."] 2019-10-27T17:24:22 < Steffanx> Nah I dont hate to work but I like my spare time more :D 2019-10-27T17:24:46 < Steffanx> Oh I thought you signed it, I'd them ask anyway 2019-10-27T17:33:58 -!- BrainDamage_ [~BrainDama@unaffiliated/braindamage] has joined ##stm32 2019-10-27T17:35:03 -!- BrainDamage [~BrainDama@unaffiliated/braindamage] has quit [Ping timeout: 246 seconds] 2019-10-27T17:35:05 -!- BrainDamage_ is now known as BrainDamage 2019-10-27T17:40:55 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-27T17:41:05 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-27T18:38:27 -!- dobson [~dobson@static.38.6.217.95.clients.your-server.de] has quit [Ping timeout: 240 seconds] 2019-10-27T18:38:48 -!- dobson [~dobson@static.38.6.217.95.clients.your-server.de] has joined ##stm32 2019-10-27T18:48:17 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-27T18:51:32 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-27T19:07:53 <@englishman> 8h30-1800 wut 2019-10-27T19:07:59 <@englishman> Do you get a 3 hour lunch 2019-10-27T19:11:43 -!- renn0xtk9 [~max@2a02:810d:1540:2448:d167:7a13:23d7:efc0] has quit [Quit: Konversation terminated!] 2019-10-27T19:14:15 < bitmask> damn, shoulda went with a higher ffc pin count, decided to add something and I'm one short 2019-10-27T19:15:11 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-27T19:24:42 -!- kow__ is now known as k\o\w 2019-10-27T19:24:48 -!- k\o\w [~iccy@135.0.26.39] has quit [Quit: Leaving] 2019-10-27T19:25:05 -!- k\o\w [~iccy@135.0.26.39] has joined ##stm32 2019-10-27T19:30:17 -!- k\o\w [~iccy@135.0.26.39] has quit [Quit: Leaving] 2019-10-27T19:30:31 -!- k\o\w [~iccy@135.0.26.39] has joined ##stm32 2019-10-27T19:31:06 -!- k\o\w [~iccy@135.0.26.39] has quit [Client Quit] 2019-10-27T19:31:20 -!- k\o\w [~afed@135.0.26.39] has joined ##stm32 2019-10-27T19:38:12 < bitmask> should you keep a smps ground separate from the rest of the gnd except for a connection at a single point or just use the ground plane for everything 2019-10-27T19:38:32 -!- renn0xtk9 [~max@2a02:810d:1540:2448:884b:1607:1e06:a0cb] has joined ##stm32 2019-10-27T19:38:35 < jpa-> bitmask: you should design it as if it only had a single point of contact, but then just connect it everywhere 2019-10-27T19:38:59 < bitmask> I dont follow 2019-10-27T19:39:15 < jpa-> that way you get the noise benefits of single-point contact, but if you mess up that design, you still don't get all the extra noise of messed up single point contact 2019-10-27T19:39:38 < jpa-> bitmask: the idea of single ground contact point is that also all signals / power traces pass at that point 2019-10-27T19:40:08 < jpa-> http://www.hottconsultants.com/techtips/split_plane.gif 2019-10-27T19:40:29 < jpa-> http://www.hottconsultants.com/techtips/split-gnd-plane.html is useful read 2019-10-27T19:40:35 < bitmask> I see, thanks 2019-10-27T19:42:05 < Jan-> and tonight I have to learn how to make this stm32 do pulse withd modulation 2019-10-27T19:42:21 < Jan-> and apparently I don't even know how to spell pulse width modulation. 2019-10-27T19:43:24 < Steffanx> thats why we call it PWM 2019-10-27T19:44:06 < Jan-> yeah. 2019-10-27T19:44:08 < Steffanx> jpa-: re the price graphs. yeah, time for an updated version jpa-. one that has some more filtering or something. 2019-10-27T19:44:48 < jpa-> Steffanx: yeah, i'll see about it in 2038 when i have time 2019-10-27T19:45:14 < Steffanx> but we are ##stm32. :P 2019-10-27T19:45:33 < Steffanx> And it seems it picks the wrong freq. for the H745 2019-10-27T19:45:35 < Steffanx> for example 2019-10-27T19:45:45 < Steffanx> whatever the definition of wrong freq. is. Since it has 2 2019-10-27T19:46:37 < Steffanx> can we hire you to do it bitmask? 2019-10-27T19:47:22 < Steffanx> is devembedded that busy nowadays? or is it the kid taking all of your attentionm, jpa-? 2019-10-27T19:47:24 < bitmask> lots of monies 2019-10-27T19:48:14 < Steffanx> what is lots of monies? 2019-10-27T19:48:28 < jpa-> Steffanx: yeah, digikey lists stm32h745 as 216 MHz 2019-10-27T19:48:34 < Steffanx> $100 is a lot? 2019-10-27T19:48:35 < bitmask> $10k 2019-10-27T19:48:38 < Steffanx> awh damn 2019-10-27T19:48:43 < jpa-> Steffanx: both, kid takes time and also have to try to bring some food to the table 2019-10-27T19:48:54 < Steffanx> is that what you ask you fellow ##stm32 mates, bitmask. 2019-10-27T19:49:04 < Steffanx> mr jpa- should charge you for his help. 2019-10-27T19:49:20 < Steffanx> *your fellow 2019-10-27T19:49:47 < bitmask> shhh 2019-10-27T19:50:30 < Steffanx> did you ever continue this draw bot thingy/web app? 2019-10-27T19:50:49 < Jan-> so how does pwm even work 2019-10-27T19:50:53 < Jan-> does it use one of the timers 2019-10-27T19:51:11 < jpa-> Steffanx: is this offer still valid? 2012-10-23T18:37:09 <+Steffanx> I'm here to babysit you and jpa-, zyp 2019-10-27T19:51:28 < qyx> :D :D 2019-10-27T19:51:30 < Steffanx> 2012, lol. 2019-10-27T19:51:50 < jpa-> we've been talking this same crap all over for a decade now 2019-10-27T19:52:07 < bitmask> no, the 3d printed design was too complicated and had too much play. wasn't worth it, i'll work on it when I build a better design 2019-10-27T19:52:10 < jpa-> "did you check the refman?" "what's a refman?" "reference manual" "i'm using stdperiph" "get out" 2019-10-27T19:52:46 < Jan-> how do I stop cube from always waiting at a breakpoint when I upload code 2019-10-27T19:52:52 < jpa-> Jan-: yeah, you take a timer, set one of the "capture/compare" channels to pwm and start it 2019-10-27T19:52:53 < Jan-> it's like press F11, wait, press F8 to make it actually run 2019-10-27T19:53:47 < Steffanx> 2011-08-05T21:05:05 <+izua> ain't a lot of activity here, usually, but sometimes there are on-topic discussions 2019-10-27T19:53:47 < Steffanx> 2011-08-05T21:05:38 < Steffanx> Mwah, more off-topic 2019-10-27T19:53:51 < Steffanx> nothing changed, indeed 2019-10-27T19:54:27 < Steffanx> the cube ide jpa-? 2019-10-27T19:54:29 < Steffanx> *Jan- 2019-10-27T19:54:40 < Jan-> yes. 2019-10-27T19:54:44 < Jan-> sorry for starting with J :/ 2019-10-27T19:54:53 < Steffanx> It must have a setting in the debug configuration/settings where you can tell it to not break it main 2019-10-27T19:54:57 < Steffanx> it is some checkbox in the settings 2019-10-27T19:55:07 < Steffanx> i dont have the cubeide, but its some semi-generic eclipse thing 2019-10-27T19:55:14 -!- renn0xtk9 [~max@2a02:810d:1540:2448:884b:1607:1e06:a0cb] has quit [Ping timeout: 276 seconds] 2019-10-27T19:55:32 < Steffanx> i actually like it, because i want to tell it when it can start 2019-10-27T19:55:45 < Steffanx> although sometimes that might be before it even hits main() 2019-10-27T19:56:07 < Jan-> I'm not sure where the breakpoint actually is 2019-10-27T19:56:23 < jpa-> Steffanx: so, i have bad memory, was zyp's baby born already? 2019-10-27T19:56:54 < Steffanx> uhm. yes 2019-10-27T19:57:22 < Steffanx> i think so at least 2019-10-27T19:57:33 < jpa-> ah yeah, in august 2019-10-27T19:58:19 < Steffanx> i keep forgetting sending stroopwafels to you in may. :P 2019-10-27T19:58:44 < Jan-> STROOP WAFFLES?! 2019-10-27T19:58:49 < Jan-> They're the only edible dutch food! 2019-10-27T19:58:52 < Jan-> They're awesome! 2019-10-27T19:58:56 < Steffanx> lies. 2019-10-27T19:58:57 < Jan-> You have some? 2019-10-27T19:59:06 < Steffanx> atm, no. 2019-10-27T19:59:07 * Jan- takes all of Steffanx's stroop waffles 2019-10-27T19:59:28 < jpa-> Jan-: the "wood shoes" are also edible, i think 2019-10-27T19:59:36 < BrainDamage> only used ones 2019-10-27T19:59:39 < Steffanx> but once i a while i ship waffles to fellow ##stm32 followers. 2019-10-27T19:59:54 < jpa-> and those two are all the dutch food there is, i think 2019-10-27T19:59:56 < BrainDamage> quit your job and start an internel selling business 2019-10-27T20:00:01 < Jan-> I think we can actually get them here 2019-10-27T20:00:03 < Jan-> but they are great 2019-10-27T20:00:07 < Jan-> they are really great 2019-10-27T20:01:01 < BrainDamage> it's sugary waffle, with caramel ( which is burned sugar ), with sugar granules on top 2019-10-27T20:01:15 < BrainDamage> kinda hard to get it wrong 2019-10-27T20:01:33 < Steffanx> sugar granules on top, wut? 2019-10-27T20:01:46 < BrainDamage> here there's variants with extra sugar 2019-10-27T20:01:57 < Steffanx> lol, EXTRA sugar. 2019-10-27T20:02:15 < Jan-> They're gorgeous 2019-10-27T20:02:20 < Jan-> chewy and crunchy at the same time 2019-10-27T20:02:24 < Jan-> they are great. 2019-10-27T20:02:44 < Steffanx> BrainDamage: stroop@waffle.express is real already, but i didnt quit my job. unlikely that will happen. 2019-10-27T20:04:22 < jpa-> i've quit every job i've ever had 2019-10-27T20:04:28 < Jan-> http://waffle.express? 2019-10-27T20:04:50 < Steffanx> Same here, except one jpa-. 2019-10-27T20:04:56 < jpa-> Jan-: it only works over gopher, of course 2019-10-27T20:05:06 < Steffanx> it only works over email. 2019-10-27T20:05:24 < Jan-> can I order waffles? :D 2019-10-27T20:05:35 < jpa-> (gopher means waffle in french) 2019-10-27T20:05:56 < qyx> interesting, we call them "gofri" 2019-10-27T20:06:04 < qyx> I didn$t know that 2019-10-27T20:06:45 < Steffanx> gaufre, but yes 2019-10-27T20:06:51 < qyx> read go-free 2019-10-27T20:07:06 < Jan-> I have to say 2019-10-27T20:07:07 < jpa-> qyx: waffles, or the rats? 2019-10-27T20:07:08 < Jan-> dutch food is not great 2019-10-27T20:07:14 < Jan-> but waffles make up for it 2019-10-27T20:07:24 < Steffanx> dutch food is awesome. 2019-10-27T20:07:52 < qyx> jpa-: wut rats? no, waffles 2019-10-27T20:08:26 < jpa-> dutch food is best made in dutch oven 2019-10-27T20:08:32 < Jan-> Dutch food is entirely made out of mayonnaise. 2019-10-27T20:08:34 < con3> stroop waffles? 2019-10-27T20:08:41 < Jan-> They put mayonnaise on pizza for god's sake 2019-10-27T20:08:42 < Steffanx> stroopwafels. 2019-10-27T20:08:45 < Steffanx> lol no. 2019-10-27T20:08:49 < Jan-> stroop waffles! 2019-10-27T20:08:56 < Steffanx> idk where you've been, but it wasnt dutchland 2019-10-27T20:09:02 < Jan-> amsterdam 2019-10-27T20:09:06 < Jan-> they put mayo on CHIPS 2019-10-27T20:09:08 < Jan-> it's gross 2019-10-27T20:09:11 < con3> ive hung around enough in ##electronics to know stroop waffles :p 2019-10-27T20:09:26 < jpa-> when we were in amsterdam we ate at mcdonalds - it was expensive and the burgers were smaller than here :D 2019-10-27T20:09:26 < Steffanx> we eat everything with "appelmoes" > " Apple sauce" or whatever it would translate to 2019-10-27T20:09:38 < con3> why would you put mayo on chips 2019-10-27T20:09:42 < Steffanx> no, we dont put it on chips, Jan- 2019-10-27T20:09:46 < Steffanx> we dip in in mayo. 2019-10-27T20:09:53 < Jan-> You make everything out of mayo 2019-10-27T20:09:57 < Jan-> mayo on pizza 2019-10-27T20:09:59 < Jan-> mayo on chips 2019-10-27T20:10:06 < con3> Steffanx: is mayo very common that side? 2019-10-27T20:10:07 < jpa-> mayo on waffles 2019-10-27T20:10:08 < Jan-> mayo on sandwiches, ok, but TOASTED sandwiches? 2019-10-27T20:10:12 < qyx> jpa-: how much is a burger in jan-land? 2019-10-27T20:10:13 < Jan-> all dutch food is mayo. 2019-10-27T20:10:43 < Steffanx> no mayo in pizza, for fuck sake :P 2019-10-27T20:10:49 < Steffanx> or waffles 2019-10-27T20:10:50 < Jan-> they do. 2019-10-27T20:10:53 < Steffanx> lies. 2019-10-27T20:10:55 < jpa-> qyx: i have no idea about jan-land, but in jpa-land it would be ~6 EUR for the meal and ~3 EUR for just a burger 2019-10-27T20:10:57 < Jan-> hang on, you're dutch? 2019-10-27T20:11:11 < jpa-> no, Steffanx is hollandaise 2019-10-27T20:11:31 < jpa-> from the nether regions 2019-10-27T20:11:36 < Jan-> I was gonna say 2019-10-27T20:11:37 < qyx> jpa-: such cheap 2019-10-27T20:11:57 < kakim32> I just fed my 2 friends 2019-10-27T20:11:58 < qyx> here it is ~8e for the meal 2019-10-27T20:12:05 < kakim32> 60eur platter 2019-10-27T20:12:12 < kakim32> a lot of chicken died for us 2019-10-27T20:12:16 < Steffanx> let me check the big mac index 2019-10-27T20:12:38 < con3> hmm... whats a really good dish/ kinda food you guys would recommend where you're from? 2019-10-27T20:13:04 < jpa-> bigmac meal is 7.15 EUR without discounts 2019-10-27T20:13:19 < Steffanx> you all know this? or did google tell you? 2019-10-27T20:13:26 < qyx> https://www.zomato.com/bratislava/elephant-diner-ra%C4%8Da-bratislava-iii 2019-10-27T20:13:29 < qyx> ^ thats here 2019-10-27T20:13:55 < jpa-> Steffanx: of course we know the prices of burgers, what else would we eat? 2019-10-27T20:14:18 < Steffanx> i prefer the burger king over mcdonalds 2019-10-27T20:14:28 < jpa-> i prefer hesburger 2019-10-27T20:14:32 < Steffanx> but the burger king on the most awful locations so i hardly go there 2019-10-27T20:14:34 < con3> treason! 2019-10-27T20:14:40 < Steffanx> *is on 2019-10-27T20:14:45 < BrainDamage> diy mayonnaise is not that terrible, at least it has lemon juice in place of the vinegar 2019-10-27T20:15:27 < kakim32> there is a new hamburger restaurant 2019-10-27T20:15:30 < BrainDamage> ( industrial nearly universally uses vinegar since it's cheaper, more consistent and lasts longer ) 2019-10-27T20:15:35 < kakim32> called burger king 2019-10-27T20:15:56 < kakim32> flame grilled makes a difference 2019-10-27T20:15:57 < BrainDamage> here we're getting invaded by tons of american fast food chains 2019-10-27T20:16:06 < BrainDamage> food ruckers, etc 2019-10-27T20:16:31 < con3> yeah we've got a few american chains here but man there's some i really want 2019-10-27T20:16:34 < con3> like taco bell 2019-10-27T20:16:38 < con3> i want to try a taco 2019-10-27T20:16:39 < Steffanx> So whats the latest happy meal toy you got 2019-10-27T20:16:42 < Steffanx> jpa-: 2019-10-27T20:16:45 < Steffanx> or did they stop doing that 2019-10-27T20:17:01 < BrainDamage> they even opened starbucks this year 2019-10-27T20:17:08 < jpa-> i can't even remember when i last went to mcdonalds, must have been many years ago 2019-10-27T20:18:00 < Steffanx> do they sell hotdog pizza yet, BrainDamage? 2019-10-27T20:18:06 < qyx> https://www.mouser.sk/ProductDetail/Octavo-Systems/OSD32MP157C-512M-BAA-ES?qs=sGAEpiMZZMu3sxpa5v1qrpNRHZ6GopPLTfWmsmUgsV8%3D 2019-10-27T20:18:13 < qyx> great, repackaged STM32MP1 2019-10-27T20:18:16 < qyx> with memory and stuff 2019-10-27T20:18:29 < jpa-> is qyx in slovakia now? 2019-10-27T20:18:37 < Steffanx> not just now. 2019-10-27T20:18:44 < BrainDamage> Steffanx: don't know, hardly ever get fast food 2019-10-27T20:18:50 < qyx> 302 Ball, 1mm Pitch 2019-10-27T20:18:55 < Steffanx> pizza is not fast food is, it BrainDamage? 2019-10-27T20:19:11 < BrainDamage> here's not considered fast food 2019-10-27T20:19:18 < Steffanx> jpa-: for some parts of europe cz and sk are still the same country. :P 2019-10-27T20:19:26 < qyx> jpa-: yeah 2019-10-27T20:20:02 < BrainDamage> also the amount of pizza and carbs you get here is nowhere near an american style 2019-10-27T20:20:10 < BrainDamage> unless you get the trancio version 2019-10-27T20:20:16 < BrainDamage> which is the fast food version 2019-10-27T20:20:40 < BrainDamage> https://vivimilano.corriere.it/wp-content/uploads/2017/03/www-da-pino-550x298.jpg 2019-10-27T20:20:52 < BrainDamage> but you typically have a single portion like that 2019-10-27T20:20:58 < BrainDamage> rather than a whole pizza 2019-10-27T20:21:10 < qyx> here it is called a pizza slice 2019-10-27T20:21:23 < qyx> usually considered fast food 2019-10-27T20:21:29 < Steffanx> hm 2019-10-27T20:21:32 < qyx> but you may get the whole proper pizza in a restaurant 2019-10-27T20:21:41 < BrainDamage> yes, that's the fast food 2019-10-27T20:21:41 < qyx> which is not fast food anymore 2019-10-27T20:21:48 < BrainDamage> yup, same here 2019-10-27T20:21:51 < qyx> oh I misread 2019-10-27T20:22:25 < BrainDamage> http://www.marcellofotia.it/wp-content/uploads/2014/06/pizza.jpg 2019-10-27T20:22:29 < BrainDamage> this is a reg pizza here 2019-10-27T20:22:38 < BrainDamage> and yes, it fits a normal flat plate 2019-10-27T20:22:50 < BrainDamage> and it's considered a whole meal on its own 2019-10-27T20:23:15 < Steffanx> to eat the outside crust or not to eat it, that's the question. 2019-10-27T20:23:55 < BrainDamage> cut it in a spiral and eat it without hands 2019-10-27T20:24:15 < qyx> iirc the nominal diameter of a pizza here is 33cm 2019-10-27T20:24:16 < Steffanx> lolwut? 2019-10-27T20:25:14 < qyx> ok I should do some work now. 2019-10-27T20:25:28 < aandrew> 33cm is a tiny pizza 2019-10-27T20:26:38 < Steffanx> too much yankee inside you, aandrew 2019-10-27T20:27:30 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-27T20:40:32 < aandrew> Steffanx: not at all 2019-10-27T20:40:36 < aandrew> if I want a personal pizza I'll get one 2019-10-27T20:40:44 < aandrew> but a foot wide pizza is tiny 2019-10-27T20:41:36 < steve> when using stm32 HAL, are you supposed to #include "stm32xyz_hal.h" , or 2019-10-27T20:51:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-27T20:52:08 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-27T20:55:55 -!- renn0xtk9 [~max@2a02:810d:1540:2448:884b:1607:1e06:a0cb] has joined ##stm32 2019-10-27T21:02:04 < Jan-> does ST have any online documentation for any of the cube stuff 2019-10-27T21:02:11 < Jan-> for instance we're trying to set up a timer for PWM 2019-10-27T21:02:17 < mawk> read the .c 2019-10-27T21:02:19 < Jan-> there's a bunch of drop down boxes that don't have a lot of explanation 2019-10-27T21:02:20 < mawk> the doc is in it 2019-10-27T21:02:23 < mawk> ah 2019-10-27T21:02:24 < Jan-> no that's not documentation, that's code 2019-10-27T21:02:28 < mawk> yes it's doc 2019-10-27T21:02:30 < mawk> the doc is in the code 2019-10-27T21:02:37 < Jan-> WHAT code? 2019-10-27T21:02:43 < mawk> the relevant .c 2019-10-27T21:02:49 < mawk> stm32xxx_pwm.h 2019-10-27T21:02:50 < Jan-> Is there no documentation on ST's site? 2019-10-27T21:02:57 < Jan-> maybe giving some explanation? 2019-10-27T21:03:20 < mawk> there are 4380943809 variants of HAL for each device, I guess that's why they don't do it 2019-10-27T21:03:26 < mawk> but you can generate it as html, easy 2019-10-27T21:03:30 < mawk> run doxygen on the .c 2019-10-27T21:03:51 < mawk> but it's not very needed, just read the relevant .c, the doc is there I promise 2019-10-27T21:03:51 < Jan-> surely they haven't shipped software with all this in it with no documentation 2019-10-27T21:04:00 < mawk> I just told you where the doc is 2019-10-27T21:04:03 < mawk> the doc is in the .c 2019-10-27T21:04:22 < Jan-> which one where 2019-10-27T21:05:01 < mawk> Drivers/STM32F1xx_HAL_Driver/Src/stm32xxxx_hal_tim.c 2019-10-27T21:05:13 < mawk> replace stm32xxxx by your chip of course 2019-10-27T21:05:29 < aandrew> Jan-: yes 2019-10-27T21:05:41 < aandrew> there is a CubeMX HAL overview doc 2019-10-27T21:05:46 < Jan-> I have a .d, an .o and a .su 2019-10-27T21:05:59 < aandrew> there are also tutorials and videos 2019-10-27T21:06:19 < mawk> in the Drivers directory Jan- 2019-10-27T21:06:41 < mawk> on my computer for instance it's in /home/nicolas/STM32CubeIDE/workspace_1.0.0/Blinky/Drivers/STM32L4xx_HAL_Driver/Src/stm32l4xx_hal_tim.c 2019-10-27T21:06:48 < Jan-> for instance what's the difference between "PWM Generation CH1" and "PWM Generation CH1N" 2019-10-27T21:06:58 < aandrew> Jan-: N is negative 2019-10-27T21:07:00 < aandrew> inverted 2019-10-27T21:07:04 < Jan-> yes I'm there mawk but there's no .c files 2019-10-27T21:07:09 < Jan-> there's .d files, .o files and .su files 2019-10-27T21:07:20 < aandrew> i.e it's not another output in the sense that it's independent, it's just the inverted output of CH1 2019-10-27T21:07:23 < mawk> in the Drivers directory ? 2019-10-27T21:07:45 < Jan-> C:\Users\JanH\STM32CubeIDE\workspace_1.0.2\Test2\Debug\Drivers\STM32F4xx_HAL_Driver\Src 2019-10-27T21:08:46 < mawk> why debug ? 2019-10-27T21:09:14 < Jan-> I uhoh, I didn't create it 2019-10-27T21:09:20 < mawk> anyway get the standalone code by another other mean if you don't find it in your project, you have the .c in them 2019-10-27T21:10:18 < Jan-> in cube mx, there is a little info button 2019-10-27T21:10:36 < Jan-> and it pops up a text area with information on the selected option 2019-10-27T21:10:45 < Jan-> the documentation for "Counter Mode" is "CounterMode." 2019-10-27T21:10:55 < Jan-> Oh. OK. Great. That makes it all clear. 2019-10-27T21:11:02 < mawk> everything is in the reference manual anyway 2019-10-27T21:11:12 < mawk> you don't expect a HAL documentation to explain how the chip works 2019-10-27T21:11:29 < Jan-> why not 2019-10-27T21:11:32 < Jan-> it's got a place for information 2019-10-27T21:11:41 < mawk> it explains what the functions do, etc 2019-10-27T21:11:45 < mawk> it doesn't explain how the chip works 2019-10-27T21:11:47 < Jan-> some of the options do have information, like "RepetitionCounter must be between 0 and 155" 2019-10-27T21:11:58 < Jan-> some of them just repeat the name of the option 2019-10-27T21:11:59 < Jan-> it's not great 2019-10-27T21:14:31 < aandrew> a lot of the time you end up going to the RM and looking for similar named text for the specific peripheral 2019-10-27T21:14:55 < mawk> I'm very nice Jan- I'm generating the html docs for you 2019-10-27T21:15:05 < mawk> but their doxygen is a bit broken sometimes iirc 2019-10-27T21:16:49 < Jan-> don't worry about it on my account 2019-10-27T21:17:02 < Jan-> is there some obviou way of finding out which pin channel 1 of timer 1 comes out on 2019-10-27T21:17:43 < mawk> lol 2019-10-27T21:17:47 < mawk> in RM maybe 2019-10-27T21:17:52 < mawk> otherwise for that I just look at very pin one by one 2019-10-27T21:17:56 < Jan-> oh yeah lol 2019-10-27T21:17:56 < mawk> but your chip has many pins 2019-10-27T21:17:59 < mawk> my chip is a 32 pins chip 2019-10-27T21:18:00 < Jan-> because that's completely hilarious 2019-10-27T21:18:01 < aandrew> Jan-: that is done in cube 2019-10-27T21:18:09 < Jan-> why would that be a fact anyone needs to know 2019-10-27T21:18:27 < mawk> aandrew: Jan- is trying to set up the pin as tim output in the MX pin view 2019-10-27T21:18:29 < aandrew> Jan-: what I do is look at the datasheet for the part (not the RM) - and look for TIM1_CH1 for example 2019-10-27T21:18:31 < mawk> I think 2019-10-27T21:18:38 < aandrew> you find pin mapping very quickly that way 2019-10-27T21:18:48 < Jan-> yeah I guess 2019-10-27T21:18:52 < Jan-> you'd think it'd be in cube mx though 2019-10-27T21:18:58 < aandrew> it is 2019-10-27T21:19:08 < Jan-> it doesn't say anywhere 2019-10-27T21:19:10 < aandrew> you go to tim1, you say CH1 has an output 2019-10-27T21:19:14 < aandrew> it selects one at random 2019-10-27T21:19:25 < aandrew> then I think it is control-click on that pin and any others that *can* have that same function light up 2019-10-27T21:19:30 < aandrew> it's not intuitive 2019-10-27T21:19:33 < Jan-> well we set "channel 1" to "PWM generation channel 1" 2019-10-27T21:19:42 < Jan-> PE9 is set to TIM1_CH1 2019-10-27T21:19:54 < Jan-> presumably we have to set up a trigger source or clock source as well? 2019-10-27T21:20:00 < aandrew> yep now control (or shift or alt) click on PE9 in the chip view 2019-10-27T21:20:40 < Jan-> clock source options are disable, internal clock, or etr2 2019-10-27T21:20:43 < Jan-> presumably we want internal clock 2019-10-27T21:20:53 < aandrew> yes, normally that is what you want for PWM 2019-10-27T21:21:05 < Jan-> trigger source can be disable because we don't want it to trigger on interrupts? 2019-10-27T21:21:17 < aandrew> correct 2019-10-27T21:21:25 < Jan-> slave mode? 2019-10-27T21:21:30 < aandrew> you would do very well to look at the tutorials on using the timers, they are very comprehensive and explain a lot 2019-10-27T21:22:14 < Jan-> I found this https://www.st.com/content/ccc/resource/technical/document/application_note/group0/91/01/84/3f/7c/67/41/3f/DM00236305/files/DM00236305.pdf/jcr:content/translations/en.DM00236305.pdf 2019-10-27T21:24:35 < aandrew> you'll have to give me the title 2019-10-27T21:27:21 < Jan-> Does anyone here understand the sentence, "The time-base unit is made by the timer counter in addition to a prescaler stage and a repetition counter" 2019-10-27T21:29:12 < Jan-> the phrase "trigger source" does not appear in that document 2019-10-27T21:31:14 < mawk> aandrew: General-purpose timer cookbook for STM32 microcontrollers 2019-10-27T21:31:28 < mawk> look at the reference manual of your chip Jan- , it contains everything 2019-10-27T21:31:33 < mawk> cookbooks and application notes are accessory 2019-10-27T21:32:11 < aandrew> yes but the timer app note is very, very good 2019-10-27T21:32:21 < aandrew> if my laptop wasn't so laggy atm I'd find it for you 2019-10-27T21:32:34 < mawk> she found it, it's the link she gave 2019-10-27T21:32:34 < aandrew> just google for "stm32 timer tutorial" I think 2019-10-27T21:32:50 < mawk> or is it ? "general-purpose timer cookbook" 2019-10-27T21:33:20 < Jan-> "trigger source" doesn't appear in the reference manual for the stm32f407 2019-10-27T21:33:32 < mawk> TRGO ? 2019-10-27T21:33:36 < mawk> TR something 2019-10-27T21:33:44 < Jan-> neither does "slave mode" 2019-10-27T21:33:46 < mawk> sometimes searching a pdf can take long, let the search run for a while 2019-10-27T21:33:51 < mawk> give it half a minute 2019-10-27T21:34:23 < aandrew> you do not need a trigger source for PWM unless you're doing something special, internal I think is the defalt 2019-10-27T21:34:33 < Jan-> fine aandrew 2019-10-27T21:34:39 < Jan-> but what worries me is basically this thing is undocumented 2019-10-27T21:34:47 < Jan-> there is no way of knowing other than me just asking you guys for every single thing 2019-10-27T21:34:55 < Jan-> is there no manual for cubemx? 2019-10-27T21:36:02 < mawk> not really, they expect you to know the terms since you're supposed to have read the reference manual 2019-10-27T21:36:08 < mawk> or at least the parts about the peripheral you're using 2019-10-27T21:36:14 < mawk> just open your RM to the general purpose timer page 2019-10-27T21:36:55 < Jan-> I'm supposed to have read and memorised every register definition in a 1749 page reference document. 2019-10-27T21:37:52 < PaulFertser> Not, you're supposed to understand the basic concepts used by particular peripheral block. 2019-10-27T21:45:26 < Jan-> I think they let the people who design the chip write the docs. 2019-10-27T21:45:30 < Jan-> Which is a really horrible idea. 2019-10-27T21:46:46 < kakim32> 1749 sounds like a lpc part id 2019-10-27T21:47:00 < Jan-> there is no way that a google search for "STM32F497 HAL PWM example" should not just link straight to an example on ST's site 2019-10-27T21:47:23 < kakim32> stop whining 2019-10-27T21:47:31 < Jan-> tell me I'm wrong 2019-10-27T21:47:38 < steve> there's examples in the Cube's 2019-10-27T21:47:49 < kakim32> indeed 2019-10-27T21:47:50 < steve> the Cube .zip's 2019-10-27T21:48:05 < kakim32> Jan- is not arsed to see that package 2019-10-27T21:48:17 < Jan-> how do I google for "the cube zips" 2019-10-27T21:48:27 < steve> lol 2019-10-27T21:48:31 < Jan-> Seriously. 2019-10-27T21:48:35 < Jan-> How am I supposed to know any of this exists. 2019-10-27T21:48:40 < BrainDamage> I keep reading the cube .zyps 2019-10-27T21:48:50 < kakim32> somebody tell where cube stores the packages so we can get over this 2019-10-27T21:49:06 < Jan-> the help menu option in stm32cubeide has no results for "slave mode" 2019-10-27T21:49:28 < Jan-> I mean seriously 2019-10-27T21:49:32 < Jan-> how shit is this 2019-10-27T21:50:58 < kakim32> you mean your IDEs help page doesn't tell you how to write an application? 2019-10-27T21:51:03 < steve> the vendors are conspiring against you to keep fortune 500 interests alive, you have to stay strong and fight back 2019-10-27T21:51:23 < Jan-> kakim32: no, my IDE's help page doesn't tell me about the purpose of a drop down in the IDE. 2019-10-27T21:51:24 < Jan-> Duh. 2019-10-27T21:51:48 < Jan-> And why the hell shouldn't it tell me how to write an application. It's the help function in a piece of software designed for writing applications. 2019-10-27T21:51:49 < Jan-> Jesus. 2019-10-27T21:52:39 < Jan-> OK so I found stm32f4xx_hal_tim.c 2019-10-27T21:52:54 < kakim32> find the package 2019-10-27T21:53:03 < kakim32> import example projects that are relevant 2019-10-27T21:53:11 < steve> *insert grandfather's "back in my day we had to key in programs with 1's and 0's" story here* 2019-10-27T21:53:24 < Jan-> it's got the usual doxygen style stuff, SetConfig()" /* Sets Config */ yes very helpful 2019-10-27T21:53:50 < kakim32> hals are not documented it's given 2019-10-27T21:54:01 < kakim32> you need to reverse them with RM and examples 2019-10-27T21:54:05 < Jan-> I was getting that impression. 2019-10-27T21:54:09 < Jan-> That's completely insane. 2019-10-27T21:54:33 < Jan-> the IDE isn't really documented either 2019-10-27T21:54:51 < kakim32> guise where is the path to cube downloads? 2019-10-27T21:55:09 < Jan-> I found that file in: 2019-10-27T21:55:16 < Jan-> C:\Users\JanH\STM32CubeIDE\workspace_1.0.2\Test2\Drivers\STM32F4xx_HAL_Driver\Src 2019-10-27T21:55:31 < kakim32> yes that is the workspace 2019-10-27T21:56:14 < kakim32> clearly it has all the drivers imported also cmsis 2019-10-27T21:56:21 < steve> you get the Cubes from the ST website. warning they're gigs in size 2019-10-27T21:56:32 < kakim32> gigs? 2019-10-27T21:56:44 < kakim32> LPCOpen for one chip is like 30megs 2019-10-27T21:56:53 < steve> around 1-2GB for the CubeF7 zip 2019-10-27T21:57:04 < kakim32> has hal, cmsis, one example for every peripheral 2019-10-27T21:57:22 < kakim32> what do they put inside those? 2019-10-27T21:57:54 < steve> theyre's a lot of non-source files like HTML, bitmaps, 100's of copies of thes same static libs, etc 2019-10-27T21:58:02 < mawk> the IDE is just eclipse 2019-10-27T21:58:47 < kakim32> steve: sounds awulf 2019-10-27T21:59:07 < kakim32> but when Jan- created his project 2019-10-27T21:59:16 < kakim32> did cube store that shit? 2019-10-27T21:59:16 < Jan-> According to the reference manual, I don't think I want any of the slave modes. 2019-10-27T21:59:22 < Jan-> kakim32: it's short for Janine. 2019-10-27T21:59:56 < steve> after you go thruogh the head-desk banging phase it becomes comedic relief 2019-10-27T22:00:38 < steve> i'd guess that the cube zip is bundled into teh IDE, I don't know I use vim and emacs and cmake 2019-10-27T22:00:47 < Jan-> I would rather have done that 2019-10-27T22:01:00 < Jan-> but that would require me to be able to write makefiles 2019-10-27T22:01:15 < Jan-> and there is a huge amount of very very complex stuff to deal with. 2019-10-27T22:01:21 < rue_mohr> what is the point of using a HAL for a project that not even going to last past the EOL of the system is was designed on? 2019-10-27T22:01:45 < Jan-> is that question directed at me 2019-10-27T22:01:47 < rue_mohr> seems like its just extra overhead 2019-10-27T22:01:53 < rue_mohr> Jan-, no 2019-10-27T22:01:58 < rue_mohr> I mean in general 2019-10-27T22:02:01 < steve> rue_mohr wat? 2019-10-27T22:02:23 < Jan-> so this prescaler thing 2019-10-27T22:02:24 < rue_mohr> HAL is for porting a project between hardware platforms 2019-10-27T22:02:29 < steve> is STM32 getting EOL'd? 2019-10-27T22:02:41 < Jan-> on timer 1 or 8 we have the option to divide the clock by any 16 bit integer? 2019-10-27T22:03:06 < BrainDamage> rue_mohr: HAL is the name of a lib which collects stm32 register locations 2019-10-27T22:03:11 < BrainDamage> not hal the concept 2019-10-27T22:03:28 < steve> every project uses a HAL regardless of portability concerns 2019-10-27T22:03:37 < rue_mohr> heh, we can out of acronyms 2019-10-27T22:04:08 < Jan-> when I am calculating this prescaler value should I assume that it is prescaling the 168mhz system clock 2019-10-27T22:04:14 < Jan-> I mean we set "clock source" to "internal clock" 2019-10-27T22:04:28 < Jan-> is that what "internal clock" means 2019-10-27T22:04:42 < kakim32> steve: mcuxpresso is under 1gig and it comes with lpcopen for all the LPC 2019-10-27T22:04:47 < mawk> yes Jan- 2019-10-27T22:04:55 < mawk> not especially 168MHz but the clock for the tim peripheral yes 2019-10-27T22:05:03 < mawk> the RM says which clock is for which 2019-10-27T22:05:05 < Jan-> well it's a 168mhz chip 2019-10-27T22:05:07 < mawk> which TIM ? 2019-10-27T22:05:10 < Jan-> 1 2019-10-27T22:05:20 < Jan-> hang on 2019-10-27T22:05:20 < mawk> yes, but that doesn't mean everything runs at 168MHz, it depends on how you configured the clock tree 2019-10-27T22:05:27 < mawk> you could run it at 1MHz I guess for instance 2019-10-27T22:06:00 < steve> kakim32 that's nice. Not that disk space is a real concern. Atmel studio is like... what, 5 gigs? Significantly more than 1 gig 2019-10-27T22:06:05 < Jan-> are we looking fpr "APB1 timer clocks" 2019-10-27T22:06:09 < mawk> APB1 timer clock is max 84MHz and APB2 timer clock is max 168MHz 2019-10-27T22:06:18 < mawk> maybe, does the RM says it so ? 2019-10-27T22:06:25 < Jan-> no idea 2019-10-27T22:06:33 < mawk> there's a diagram about clock/timer 2019-10-27T22:06:35 -!- X230t [x13@gateway/shell/suchznc/x-ciswnpaftfqcrcdf] has joined ##stm32 2019-10-27T22:06:37 < mawk> it's in the first pages 2019-10-27T22:06:39 < mawk> let me look 2019-10-27T22:07:20 < Jan-> what does AHB stand for 2019-10-27T22:07:25 < Jan-> or even APB 2019-10-27T22:07:36 < kakim32> steve: I don't think you need to install that full 5gigs 2019-10-27T22:07:36 < mawk> this is your reference manual https://www.st.com/content/ccc/resource/technical/document/reference_manual/3d/6d/5a/66/b4/99/40/d4/DM00031020.pdf/files/DM00031020.pdf/jcr:content/translations/en.DM00031020.pdf 2019-10-27T22:07:39 < mawk> dunno I forgot 2019-10-27T22:07:46 < mawk> APB and AHB are two system busses 2019-10-27T22:07:49 < mawk> or two system "domains" 2019-10-27T22:07:51 < kakim32> you can choose if to install 8 bit or 32bit stuff etc. 2019-10-27T22:08:02 < kakim32> iirc 2019-10-27T22:08:20 < mawk> advanced ... bridge 2019-10-27T22:08:31 < Jan-> AHB is defined in the reference manual 2019-10-27T22:08:40 < mawk> TIM1 is APB2 2019-10-27T22:08:48 < mawk> page 66 of the pdf 2019-10-27T22:09:06 < mawk> so, if you set everything at max freq it will be 168MHz 2019-10-27T22:09:17 < Jan-> for now it looks like 16 2019-10-27T22:09:18 < Jan-> ok 2019-10-27T22:10:42 < Jan-> if we set it to 8000 we should get 2000hz pwm, which we will probably increase later but that's OK 2019-10-27T22:11:05 < mawk> if you set what to 8000 ? 2019-10-27T22:11:13 < mawk> you have two values to set, prescaler and autoreload value 2019-10-27T22:11:22 < mawk> autoreload value is the period - 1, prescaler is the dividend for the base clock 2019-10-27T22:11:29 < mawk> for the internal clock, I mean 2019-10-27T22:11:33 < mawk> APB2 timer clock in your case 2019-10-27T22:12:12 < Jan-> if we set "prescaler (PSC - 16 bits value)" to 8000 we should get 2khz pwm if the timer clock is 16mhz. 2019-10-27T22:12:26 < mawk> to 7999 I think ? prescaler of 0 is division by 1 2019-10-27T22:12:29 < mawk> it's shifted 2019-10-27T22:12:50 < Jan-> near enough 2019-10-27T22:13:04 < mawk> :( 2019-10-27T22:13:16 < mawk> you don't want to make near enough stuff 2019-10-27T22:13:20 < mawk> you like *precision* 2019-10-27T22:14:10 < mawk> so 2019-10-27T22:15:08 < Jan-> I'll be happy once we have a pin toggling up and down 2019-10-27T22:15:11 < Jan-> which I bet we don't right now 2019-10-27T22:15:19 < mawk> if internal clock is 168MHz you want a prescaler of 19 for instance, that will divide by 20 2019-10-27T22:15:32 < mawk> you will have a timer base freq of 8400 Hz 2019-10-27T22:15:47 < mawk> or any other computation that works 2019-10-27T22:15:49 < mawk> you get the idea 2019-10-27T22:16:38 < rue_shop1> Jan-, which stm32? 2019-10-27T22:17:06 < mawk> in fine that will do 168000/(1 + prescaler)/(ARR + 1) hertz 2019-10-27T22:17:07 < mawk> I think 2019-10-27T22:18:39 < Jan-> right now for some reason we can't even upload code to it 2019-10-27T22:18:57 < Jan-> the debugger connects but it says "Encountered Error when opening C:\ST\STM32CubeIDE_1.0.2\STM32CubeIDE\plugins\com.st.stm32cube.ide.mcu.externaltools.cubeprogrammer.win32_1.0.0.201904021149\tools\bin\STM32_Programmer_CLI.exe" 2019-10-27T22:19:52 < rue_shop1> skip the ide? 2019-10-27T22:20:25 < rue_shop1> stlink, openocd, makefile ? 2019-10-27T22:20:26 < Jan-> I spent a week trying to avoid using this heap 2019-10-27T22:20:34 < Jan-> but the complexity of doing that is enormous 2019-10-27T22:20:54 < rue_shop1> well, if its not a bluepill, I cant really say for sure 2019-10-27T22:20:56 < mawk> openocd is a single command 2019-10-27T22:21:02 < Jan-> yeah right 2019-10-27T22:21:09 < mawk> it's not a bluepill it's a counterfeit black something 2019-10-27T22:21:10 < Jan-> openocd is a single command that requires a 500 line config file 2019-10-27T22:21:16 < mawk> stm32f407ve 2019-10-27T22:21:20 < mawk> no 2019-10-27T22:21:37 < rue_shop1> Jan-, hell no, it requires 2, with a command line not longer than 50 characters 2019-10-27T22:21:53 < mawk> you don't even need anything if your chip is already known 2019-10-27T22:21:54 < Jan-> that doensn't help if I don't know what any of those characters need to be. 2019-10-27T22:21:57 < rue_shop1> but I have something working I can try to walk you thru 2019-10-27T22:21:57 < mawk> which it probably is 2019-10-27T22:22:09 < rue_shop1> I have them in my makefile 2019-10-27T22:22:11 < mawk> openocd -f stuff -c 'flash jansfavoritebinary.elf' 2019-10-27T22:22:20 < Jan-> anyway just setting up the system clocks is 50 plus lines 2019-10-27T22:22:25 < rue_shop1> openocd -f $(OPENOCD_SCRIPTS_DIR)/interface/stlink-v2.cfg -f $(OPENOCD_SCRIPTS_DIR)/target/stm32f1x.cfg -c "program ./$(BINARY).elf verify reset exit" 2019-10-27T22:22:26 < Jan-> and that's reliant on lots of HAL calls 2019-10-27T22:22:32 < mawk> why do you care 2019-10-27T22:22:38 < mawk> you're not writing this code, MX is 2019-10-27T22:22:40 < Jan-> it'd take me weeks (literally) to work all that out from scratch. 2019-10-27T22:22:51 < mawk> you can keep your current code and just flash with openocd 2019-10-27T22:22:54 < rue_shop1> it took me about 5 months 2019-10-27T22:22:58 < mawk> you can use whatever you want 2019-10-27T22:23:38 < rue_shop1> Jan-, are you so tight to a deadline that your just freaking out now? 2019-10-27T22:23:46 < rue_shop1> or are you playing? 2019-10-27T22:24:05 < rue_shop1> casue I been at both 2019-10-27T22:24:11 < Jan-> no deadline particularly 2019-10-27T22:24:15 < Jan-> I mean I'd like to get it done 2019-10-27T22:24:25 < Jan-> anyway we solved the problem by swapping over USB ports 2019-10-27T22:24:27 < Jan-> it does that a lot 2019-10-27T22:24:30 < rue_shop1> would you like to set things aside and give my system a try? 2019-10-27T22:24:48 < rue_shop1> well, are you using stlink? 2019-10-27T22:24:53 < Jan-> in a world of infinite time and energy yes 2019-10-27T22:25:07 < Jan-> in the real world I looked at all this a week ago and decided cube was the best supported (ha!) 2019-10-27T22:25:10 < rue_shop1> I'm here if you ever want to try 2019-10-27T22:25:13 < rue_shop1> well, I'm not here 2019-10-27T22:25:17 < rue_shop1> I'm in #robotics 2019-10-27T22:25:36 < rue_shop1> well, I spend more time on twitter, but I'm always parked in #robotics 2019-10-27T22:25:43 < Jan-> I assume that just having the MX_TIM1_Init() called won't actually start the PWM 2019-10-27T22:25:45 < rue_shop1> cause irc kinda died 2019-10-27T22:26:26 < rue_shop1> sorry, I'm using a different library ( opencm3 ) 2019-10-27T22:26:49 < Steffanx> twatter is dying too sir. 2019-10-27T22:26:59 * rue_shop1 looks from the road he's standing on, to the road Jan- is standing on 2019-10-27T22:27:03 < rue_shop1> Steffanx, I'v noticed 2019-10-27T22:27:16 < Steffanx> should move to facebook or something 2019-10-27T22:27:21 < Jan-> we looked at libopencm3 2019-10-27T22:27:22 < rue_shop1> I'm starting to come up with proof that I'm one of the last 100 people left on the planet 2019-10-27T22:27:28 < Jan-> we would totally have used it 2019-10-27T22:27:31 < Jan-> the problem is, it is LGPL# 2019-10-27T22:27:49 < rue_shop1> ah ok, yea, the commercial thing is a dimension I dont have 2019-10-27T22:28:10 < Jan-> I don't even know if I really do 2019-10-27T22:28:16 < Jan-> but it would be silly to exclude the possibility 2019-10-27T22:28:44 < rue_shop1> what opencm3 does isn't really that complex, it could be substituted out 2019-10-27T22:28:48 < Jan-> the problem with cube IDE and the ST hal is that it is basically undocumented. 2019-10-27T22:28:59 < Jan-> literally the only way to find out how it works is to ask someone 2019-10-27T22:29:06 < steve> rue_shop1 what is your system you refer to? Using OpenOCD? 2019-10-27T22:29:07 < rue_shop1> I was just head-desking yesterday about what it does to set up a port 2019-10-27T22:29:27 < Jan-> right now I have set up this timer peripheral to do PWM, I think... 2019-10-27T22:29:28 < rue_shop1> steve, opencm3, openocd, and a makefile 2019-10-27T22:29:37 < Jan-> and I presumably need to tell it to start pwm-ing 2019-10-27T22:29:43 < Jan-> at some point I will need to tell it to stop PWM-ing 2019-10-27T22:29:47 < rue_shop1> Jan-, I can give you basic opencm3 source to reffer to 2019-10-27T22:29:55 < Jan-> I mean it's just stupid, how does anyone ever use this 2019-10-27T22:30:01 < rue_shop1> it might serve as a directional guide 2019-10-27T22:30:10 < rue_shop1> Jan-, I'v wondered the same thing so many times 2019-10-27T22:30:21 < steve> cool. openocd is the path to enlightenment 2019-10-27T22:30:25 < Jan-> rue what I need is a call that will be something like HAL_TIM_Enable(&htim...) etc 2019-10-27T22:30:34 < rue_shop1> stm32 should have completely overrun avr, and its still back in like ~2008 2019-10-27T22:30:43 < Jan-> but there is literally no API reference anywhere on ST's site, it's completely mad. 2019-10-27T22:30:52 < rue_shop1> Jan-, so yes? 2019-10-27T22:31:08 < Jan-> yes, stm32 should have overrun AVR, it is faster and cheaper and barely uses more power 2019-10-27T22:31:13 < mawk> I told you where the reference is 2019-10-27T22:31:17 < Jan-> but the documentation is just horrific 2019-10-27T22:31:19 < mawk> lol 2019-10-27T22:31:36 < Jan-> mawk, doxygen comments in source code is not really the sort of documentation I'm after. 2019-10-27T22:31:36 < mawk> you just don't put good will to it 2019-10-27T22:31:51 < rue_shop1> Jan-, at some point I'v love to know more about your you:computer interface 2019-10-27T22:32:01 < rue_shop1> one moment while I paste code 2019-10-27T22:32:15 < Jan-> mawk, this is how you write API docs: https://docs.microsoft.com/en-us/dotnet/api/?view=netstandard-2.1 2019-10-27T22:34:03 < rue_shop1> Jan-, http://ruemohr.org/%7Eircjunk/programming/c/stm32/stm32basicpwm.c 2019-10-27T22:34:07 < Jan-> Run a google search for HAL_GPIO_WritePin site:st.com and see what you get. Why is there not a web page with a discussion of it, and examples? Jesus. 2019-10-27T22:34:23 < rue_shop1> I think you will find the calls, tho not identicle, give away what needs to be done 2019-10-27T22:34:35 < Jan-> sure rue I get it from the reference manual 2019-10-27T22:34:37 < Jan-> at least in part 2019-10-27T22:34:51 < rue_shop1> that code works 2019-10-27T22:35:20 < Jan-> the problem is somewhere in this code is a call something like HAL_TIM1_Enable(...) or whatever 2019-10-27T22:35:24 < Jan-> I know that 2019-10-27T22:35:25 < Jan-> it must exist 2019-10-27T22:35:28 < Jan-> where is it? who knows. 2019-10-27T22:36:01 < Jan-> they have written a hardware access layer and as far as I can tell there is no comprehensive API reference for it. That is absolutely barking. 2019-10-27T22:36:36 < rue_shop1> Jan-, you gonna look over that code example of mine? 2019-10-27T22:36:42 < Jan-> we are 2019-10-27T22:36:50 < Jan-> what it basically says is "you have to enable the timer" 2019-10-27T22:36:52 < Jan-> well OK yes we do 2019-10-27T22:36:55 < Jan-> but how? 2019-10-27T22:37:21 < rue_shop1> ok, the opencm3 calls are pretty simple, as in what they do can usually be mapped back to simple register modifications 2019-10-27T22:37:59 < rue_shop1> so your just missing the call to enable the timer? 2019-10-27T22:38:04 < Jan-> I have no idea. 2019-10-27T22:38:07 < Jan-> Presumably something like that. 2019-10-27T22:38:14 < Jan-> I don't know what the auto code generation has done. 2019-10-27T22:38:40 < Jan-> There's a function called HAL_TIMEx_PWMN_Start() 2019-10-27T22:38:43 < mawk> you say "there is do docs" but when I point you to where the docs are you say "doxygen doesn't count" lol 2019-10-27T22:38:48 < mawk> N is for negative 2019-10-27T22:39:13 < Jan-> when I google for " HAL_TIMEx_PWMN_Start()" I get nothing at st.com 2019-10-27T22:39:27 < Jan-> well, community.st.com 2019-10-27T22:39:31 < qyx> wat 2019-10-27T22:39:54 < rue_shop1> I was annoyed that opencm3 does not have basic examples like "serial hello world" and "25% pwm" and "ADC to portB" 2019-10-27T22:39:58 < steve> that's probably a macro Jan- 2019-10-27T22:40:10 < qyx> rue_shop1: it has 2019-10-27T22:40:26 < Steffanx> so many annoyed people nowadays. do people not look well enough or ...? 2019-10-27T22:40:33 < rue_shop1> maybe I missed it, can you point me again? 2019-10-27T22:40:34 < mawk> why do you say that steve ? doesn't look like a macro 2019-10-27T22:41:00 < mawk> there is a "how to use this driver" section in stm32f4xx_hal_tim.c Jan- 2019-10-27T22:41:21 < qyx> Steffanx: in the past people were self-contained memory-wise, then google came and people don't ermember anything because they can search for the info, now it seems that even the google doesn't help 2019-10-27T22:41:32 < Steffanx> isnt the doxygen stuff in the cubef4 download? 2019-10-27T22:41:45 < steve> oh I read TIMEx as TIMx 2019-10-27T22:42:02 < qyx> people are like wth, someone is doing this and that differently, I don't know why and how, I refuse to use it 2019-10-27T22:42:07 < mawk> initialize using HAL_TIM_PWM_MspInit(), enable the clock using HAL_TIM_PWM_MspInit(), configure the timer using HAL_TIM_PWM_Init() and HAL_TIM_PWM_ConfigChannel() (presumably these three things are already done by CubeMX), then start the PWM using HAL_TIM_PWM_Start() 2019-10-27T22:42:09 < mawk> Jan- ^ 2019-10-27T22:42:29 < mawk> yes Steffanx 2019-10-27T22:42:46 < Jan-> yeah I know mawk 2019-10-27T22:42:47 < Jan-> I have that file 2019-10-27T22:42:50 < Jan-> MX has done some of that 2019-10-27T22:43:03 < mawk> then you know what to do 2019-10-27T22:43:06 < Jan-> oh do I 2019-10-27T22:43:13 < Jan-> the problem is that all of those functions take big structs full of options and none of those are explained 2019-10-27T22:47:37 < steve> the structs generally correspond to register values and the values you set them with are all enumerated in english 2019-10-27T22:47:44 < mawk> HAL_TIM_PWM_Start() just takes an int 2019-10-27T22:47:50 < mawk> and your handle 2019-10-27T22:49:40 < Jan-> yes 2019-10-27T22:50:00 < Jan-> so we set channel 1 to "pwm generation channel 1" 2019-10-27T22:50:14 < Jan-> trigger source and slave mode to "disable" 2019-10-27T22:50:34 < Jan-> prescaler to 8000 and everything else at defaults 2019-10-27T22:50:50 < Jan-> then called HAL_TIM_PWM_Start(&htim1, 1); 2019-10-27T22:50:56 < Jan-> the pin is high but it isn't pwm-ing 2019-10-27T22:51:08 < steve> the MSPInit() is the tricky part, did yuo define it? 2019-10-27T22:51:55 < Jan-> er, nope? 2019-10-27T22:52:44 < steve> there's a zero percent chance of any peripheral if you don't define MSPInit for that peripheral 2019-10-27T22:52:51 < Jan-> er 2019-10-27T22:52:54 < Jan-> right? 2019-10-27T22:52:58 < Jan-> never heard of it 2019-10-27T22:53:22 < steve> lol, me too. I believe it's ST specific crap-tecture 2019-10-27T22:53:45 < Jan-> Seriously. How is absolutely anyone at all supposed to know this when it is not documented anywhere. 2019-10-27T22:54:01 < Jan-> You just have to ask someone online for every single api call in your entire project? 2019-10-27T22:54:19 < steve> not a terrible idea to have a 2 layer architecture but they way they lay it out in teh HAL with the _weak symbols is just infuriating 2019-10-27T22:54:43 < Jan-> OK so in the c file there are a few references 2019-10-27T22:54:43 < Jan-> HAL_TIM_Base_MspInit() 2019-10-27T22:55:26 < Jan-> oh look fuck it this is just unusable 2019-10-27T22:55:56 < Jan-> what is there that has similar performance to an stm32, but is not an stm32 2019-10-27T22:56:22 < steve> lol 2019-10-27T22:56:39 < Jan-> I'm serious. 2019-10-27T22:56:43 * karlp laughs 2019-10-27T22:56:44 < Jan-> This is just ridiculous. 2019-10-27T22:56:46 < steve> not-stm32 is a good platform 2019-10-27T22:56:48 < BrainDamage> nxp lpc 2019-10-27T22:57:04 < karlp> Jan-: do you really expec it to be different? 2019-10-27T22:57:15 < Jan-> it could hardly be worse 2019-10-27T22:57:23 < karlp> you're willfully ignoring the packaged solutions, so you're going to get the same or worse on all theh others. 2019-10-27T22:57:30 < Jan-> what are you even on about 2019-10-27T22:57:33 < steve> i can tell you the Atmel ASF is worse 2019-10-27T22:57:36 < karlp> same shit, different day. 2019-10-27T22:57:48 < Jan-> you tell me, link me to some document, right now, where it explains that I have to write a function called MSPInit() containing... something. 2019-10-27T22:57:59 < Jan-> Where does it say I have to do that for PWM to work. 2019-10-27T22:58:03 < Jan-> Where is that written. 2019-10-27T22:58:20 < steve> actually it probably says that at the top of the file 2019-10-27T22:58:26 < Jan-> which damn file 2019-10-27T22:58:29 < karlp> so steve said some random "oh, did you call this function?!" and you took that as gospel, and that you'd have to add it manually? 2019-10-27T22:58:36 -!- jly [uid355225@gateway/web/irccloud.com/x-berwwqsqtbayytmz] has joined ##stm32 2019-10-27T22:58:41 < steve> atleast i remember it saying that for xyz_hal_uart.c 2019-10-27T22:58:48 < Jan-> no karl, I got it from the c source file 2019-10-27T22:59:14 < Jan-> where it says "Initialize the TIM low level resources by implementing the following functions depending on the selected feature: (++) Time Base : HAL_TIM_Base_MspInit()" 2019-10-27T22:59:21 < Jan-> Is that what you mean? 2019-10-27T22:59:30 < Jan-> Or is it one of the others? Is it HAL_TIM_IC_MspInit()? 2019-10-27T22:59:40 < Jan-> Or is it HAL_TIM_OC_MspInit()? 2019-10-27T22:59:42 < steve> there's nothing special about MSPInit. You have to write the code yourself, or get CubeMX to write it for you. 2019-10-27T22:59:52 < Jan-> Because steve mentioned a function just caleld MspInit(). 2019-10-27T23:00:03 < Jan-> Maybe it's none of those. And who knows what has to go in them? Where is any of this explained? 2019-10-27T23:01:00 < steve> actually I don't even use MSP init because I prefer to explicitly call the "low level" init routines before the HAL level routines 2019-10-27T23:01:19 < steve> not to be confused with the ll drivers 2019-10-27T23:07:13 < steve> Jan- you might be a good audience for the youtube videos I'm making. What do you think about this section startnig at 17:00 minutes in? https://www.youtube.com/watch?v=KRTDYod0b7I 2019-10-27T23:07:55 < Steffanx> More beer, more pizza. 2019-10-27T23:08:27 < rue_mohr> Jan-, so the only call your missing, compared to my code, is the timer enable? 2019-10-27T23:09:26 < steve> Steffanx et al. feedback appreciated :) 2019-10-27T23:09:48 -!- renn0xtk9 [~max@2a02:810d:1540:2448:884b:1607:1e06:a0cb] has quit [Quit: Konversation terminated!] 2019-10-27T23:10:34 < Jan-> rue_mohr: I have no idea. 2019-10-27T23:10:39 < Jan-> How can I have any idea 2019-10-27T23:10:43 < Jan-> it's not written down anywhere 2019-10-27T23:11:36 < Jan-> I think one of the issues is that there are a billion different libraries for stm32 and if you try to look up any concept you get 101 results for things other than the HAL and IDE you are using 2019-10-27T23:11:45 < Jan-> like this video that steve mentioned 2019-10-27T23:11:48 < Jan-> he's not using cube ide 2019-10-27T23:11:52 < Jan-> so it's no good to us at all 2019-10-27T23:11:53 < rue_mohr> well, if you compare my code to your pwm test code 2019-10-27T23:12:11 < Jan-> literally the only pwm test code I have written is "HAL_TIM_PWM_Start(&htim1, 1) 2019-10-27T23:12:20 < Jan-> the rest of it was auto generated but I don't know what it has done 2019-10-27T23:13:09 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-27T23:13:13 < Jan-> presumably at some point there will be a way to set the pwm level, presumably by setting the maximum count for timer 1 but again who knows 2019-10-27T23:14:13 < qyx> I highly recommend you to read the refman first 2019-10-27T23:14:21 < qyx> there will be no presumably then 2019-10-27T23:16:10 < steve> Jan- it's the same HAL layer, it doesn't matter what IDE you use 2019-10-27T23:22:45 < Jan-> steve, I don't think the line "osKernelStart();" will work in my environment 2019-10-27T23:23:35 < steve> Jan- do you want it to? 2019-10-27T23:23:44 < Jan-> I have no idea :( 2019-10-27T23:23:46 * Jan- sobs 2019-10-27T23:24:15 < Jan-> in all seriousness, I did look at mbed 2019-10-27T23:24:28 < steve> well, in that section starting at 17 minutes in, it's about the UART, so just take the stuff that applies to your purposes 2019-10-27T23:24:32 < Jan-> which has a full realtime os (or well sorta realtime I guess) 2019-10-27T23:24:43 < Jan-> but I felt it would be over the top for our purposes 2019-10-27T23:25:03 < jadew> wing chun (think ip man, bruce lee) vs what seems like kickboxing: https://youtu.be/_BJo7YA9tno?t=200 2019-10-27T23:25:10 < steve> MBed uses the same HAL, does it not? 2019-10-27T23:25:16 < Jan-> no idea 2019-10-27T23:25:26 < Jan-> it wasn't all like HAL_GPIO_WritePin etc 2019-10-27T23:27:35 < rue_shop1> Jan-, I think I know why I didn't go with the library your using 2019-10-27T23:28:39 < Jan-> I'm starting to realise why *I* am not going with the library I'm using 2019-10-27T23:29:11 < rue_shop1> if I had the time, I'd write you a 25% pwm demo, no idea of how I'd do it 2019-10-27T23:29:23 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-27T23:29:45 < Jan-> steve's video says to start with a working reference and modify accordingly 2019-10-27T23:29:47 < Jan-> which would be fine 2019-10-27T23:29:52 < Jan-> but I can't even do that 2019-10-27T23:31:01 < steve> what do you mean you can't? the Cube zips have working references which include Cube IDE project files (i believe) 2019-10-27T23:33:51 < Jan-> they do? news to me 2019-10-27T23:35:35 < kakim32> aren't drivers imported to workspace from that zip? 2019-10-27T23:35:43 < steve> they work in a frankenstein sense, but they do compile and run. I should probably stop scaring you. They work like kittens and ponies 2019-10-27T23:36:11 < Jan-> https://www.youtube.com/watch?v=Hp1SrxknB4A 2019-10-27T23:36:37 < kakim32> shitposting has begun 2019-10-27T23:36:45 < kakim32> you have been fully integrated to ##stm32 2019-10-27T23:36:48 < Jan-> IN SOME PLACES HORSE IS FOOD. 2019-10-27T23:36:55 * Jan- waves her arms 2019-10-27T23:37:04 < steve> kakim32 yes, the projects use either absolute paths to the HAL or linked resources 2019-10-27T23:38:26 < Steffanx> have you integrated kakim32? 2019-10-27T23:38:32 < kakim32> I never integrate 2019-10-27T23:38:36 < Steffanx> yeah. 2019-10-27T23:38:38 < Steffanx> fucking lpc user 2019-10-27T23:39:11 < Steffanx> should ask a channel op to ban you for life. 2019-10-27T23:39:11 < Jan-> I was trying to follow this: https://www.waveshare.com/wiki/STM32CubeMX_Tutorial_Series:_PWM 2019-10-27T23:39:17 < kakim32> time to play with EFM8 then 2019-10-27T23:39:25 < jadew> https://i.imgur.com/e6q1HIP.gifv 2019-10-27T23:39:26 < Jan-> but it's not like that one site has a million tutorials for everything 2019-10-27T23:39:52 < Steffanx> even worse mr kakim32 2019-10-27T23:39:55 < Steffanx> at least go for stm8 2019-10-27T23:40:16 < kakim32> I'll go full AVR 2019-10-27T23:40:28 < kakim32> that's it 2019-10-27T23:40:49 < Jan-> I so wish there was an AVR only... more stm32ish. 2019-10-27T23:41:05 < kakim32> did you look around? 2019-10-27T23:41:08 < Jan-> yes. 2019-10-27T23:41:14 < Jan-> people said xmega "sucks" 2019-10-27T23:41:30 < kakim32> they say it here too 2019-10-27T23:42:03 < kakim32> it's not optimum package of features / price eather 2019-10-27T23:42:06 < jadew> I like the SAM line 2019-10-27T23:42:22 < Jan-> I'm honestly not that price sensitive, within reason 2019-10-27T23:42:31 < kakim32> it's not too late to SAM it jadew 2019-10-27T23:42:37 < kakim32> Jan- 2019-10-27T23:43:19 < kakim32> sam has no irc channel though 2019-10-27T23:43:23 < Jan-> are the docs any better 2019-10-27T23:43:31 < jadew> atmel docs are the best 2019-10-27T23:43:41 < Jan-> I mean jesus I've seen open source projects better documented than stm32 2019-10-27T23:43:51 < Jan-> and that's saying something 2019-10-27T23:44:08 < kakim32> but not working as well as stm32 2019-10-27T23:44:21 < jadew> that's debatable 2019-10-27T23:44:50 < Steffanx> All you need is in the datasheet, forget about libs, forget about it all. go write into bare registers 2019-10-27T23:45:02 < Steffanx> *reference manual 2019-10-27T23:45:08 < Jan-> I would. 2019-10-27T23:45:12 < kakim32> then everything has documentation 2019-10-27T23:45:32 < kakim32> small thing called reference manual 2019-10-27T23:45:37 < steve> hmm wtf is HAL_TIM_MspPostInit ? Is that a CubeMX specific routine? 2019-10-27T23:45:42 < Steffanx> The only real difference is perhaps the clock setup (which you dont NEED) and you have to enable the peripherals you want to use 2019-10-27T23:45:45 < Jan-> steve: christ knows. 2019-10-27T23:46:02 < Jan-> apparently it takes six or seven API calls to even update a PWM value 2019-10-27T23:46:04 < Steffanx> if you do that, then its just very similar to the old avr wold 2019-10-27T23:46:05 < Steffanx> world. 2019-10-27T23:46:11 < Steffanx> so i dont see you problem jpa- 2019-10-27T23:46:13 < Steffanx> *Jan- 2019-10-27T23:46:13 < Steffanx> damnit 2019-10-27T23:46:32 < Jan-> seriously though? I have to go that hardcore? 2019-10-27T23:46:38 < Steffanx> no you dont, but you could 2019-10-27T23:46:45 < antto> who said xmega sux? >:( 2019-10-27T23:46:46 < Steffanx> we've been over this for like 15 times now i think :P 2019-10-27T23:46:47 < antto> dafuq 2019-10-27T23:46:51 < Steffanx> you antto 2019-10-27T23:46:52 < Steffanx> i heard it. 2019-10-27T23:46:58 < Jan-> steffan you say I don't but seriously 2019-10-27T23:47:05 < antto> ur ears r full of wax 2019-10-27T23:47:07 < Jan-> I am sitting here trying to make this thing do pwm 2019-10-27T23:47:09 < steve> PWMs are tricky because it's both a timer and a GPIO. If you get a GPIO working, and a timer working inedpenedntly, it'll be easier to get the PWM working 2019-10-27T23:47:14 < Jan-> I have literally no idea how to do that 2019-10-27T23:47:24 < Steffanx> antto: like this moment where you use the WEEXXX and then the errata says you cannot use the other pins on the same port. 2019-10-27T23:47:28 < Steffanx> Such awesome feature 2019-10-27T23:47:37 < BrainDamage> make just strobe the pin with delays 2019-10-27T23:47:42 < BrainDamage> and use the timer for another time 2019-10-27T23:47:53 < antto> Steffanx eh? 2019-10-27T23:47:53 < Jan-> seriously? that's the solution? 2019-10-27T23:48:05 < BrainDamage> you're trying to learn two things at once 2019-10-27T23:48:07 < kakim32> Jan-: did you find the package with everything in it? 2019-10-27T23:48:10 < BrainDamage> and you're struggling 2019-10-27T23:48:12 < Steffanx> it's a feature of one of the XMEGAs antto. The WEX and the awesome errata. 2019-10-27T23:48:14 < kakim32> including examples 2019-10-27T23:48:22 < Jan-> kakim I found this: https://www.waveshare.com/wiki/STM32CubeMX_Tutorial_Series:_PWM 2019-10-27T23:48:29 < antto> Steffanx dafuq is WEX? 2019-10-27T23:48:35 < Steffanx> learn to XMEGA antto 2019-10-27T23:48:37 < Jan-> but it doesn't compare totally with the code that cube mx is creating 2019-10-27T23:48:45 < antto> muh xmega don't have WEX 2019-10-27T23:48:58 < Steffanx> its some fancy waveform extension. 2019-10-27T23:49:00 < kakim32> Jan-: it's waveshare.. expect nothing 2019-10-27T23:49:12 < Jan-> in all seriousness 2019-10-27T23:49:15 < Jan-> what else am I supposed to use 2019-10-27T23:49:29 < Jan-> if I put site:st.com on the end of the google search, I get 101 people asking on the forums "how the fuck does any of this work." 2019-10-27T23:49:31 < kakim32> examples provided by st 2019-10-27T23:49:55 < kakim32> inside some massive f4 cube package 2019-10-27T23:50:19 < antto> jadew which SAM ya like? 2019-10-27T23:50:32 < Steffanx> A5. 2019-10-27T23:50:44 < antto> that's kinda.. thicc 2019-10-27T23:51:10 < jadew> antto, I generally like the atmel peripherals and have no preference for any particular chip 2019-10-27T23:51:21 < jadew> I just order by price 2019-10-27T23:51:25 < antto> yeah, i like teh periphs too 2019-10-27T23:51:40 < Steffanx> jadew: did you have a look at the SERCOM? 2019-10-27T23:51:47 < kakim32> it's okay to use SAM now.. but lpc.. bad 2019-10-27T23:52:00 < kakim32> 8] 2019-10-27T23:52:15 < Jan-> So anyway, I copied and pasted the user_pwm_setvalue function from that waveshare tutorial, and called it with a middling 16 bit value 2019-10-27T23:52:32 < Jan-> it compiles and runs but no pwm 2019-10-27T23:53:00 < steve> Jan- there are other youtube videos specifically using CubeMX and IDE 2019-10-27T23:53:12 < jadew> Steffanx, I haven't 2019-10-27T23:53:22 < antto> i have 2019-10-27T23:53:25 < Jan-> but steve I should not be reliant on youtube to tell me how to use st's chip 2019-10-27T23:53:28 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 264 seconds] 2019-10-27T23:53:28 < Jan-> where is the documentation 2019-10-27T23:53:43 < jadew> Steffanx, why do you ask? 2019-10-27T23:53:45 < jadew> is it bad? 2019-10-27T23:53:50 < Jan-> I have never ever felt like this with a bit of electronics before, the idea that there is basically no information and you just have to endlessly ask people who might randomly know. 2019-10-27T23:54:11 < antto> jadew it's bad when the datasheet is microsh*t styled 2019-10-27T23:54:12 < kakim32> https://www.st.com/en/embedded-software/stm32cubef4.html is it this? 2019-10-27T23:54:26 < Jan-> seriously, how am I supposed to know that part of setting a pwm value is "sConfigOC.OCPolarity = TIM_OCPOLARITY_HIGH;" 2019-10-27T23:54:34 < antto> coz you look thru 3 tablez to figure out which pins ya need 2019-10-27T23:54:36 < Jan-> it is literally documented nowhere 2019-10-27T23:54:42 < Jan-> you just have to magically know 2019-10-27T23:54:51 < Jan-> and even now it isn't actually working so what now. 2019-10-27T23:54:56 < steve> Jan- there is documentation, you're just refusing to read it. 2019-10-27T23:55:16 < kakim32> thing with code is that you can follow it down to chip header and look from ref man 2019-10-27T23:55:20 < Jan-> what, a youtube video 2019-10-27T23:55:59 < Jan-> look at this https://www.google.com/search?ei=mwa2XYiwEcKo8gKzprnIDg&q=%22sConfigOC.OCPolarity%22+site%3Ast.com 2019-10-27T23:56:06 < steve> no ther's a pile of pdfs, training slide shows, and I'll say it for the Nth time, the Cube zips 2019-10-27T23:56:26 < kakim32> steve: is that MCU package stored locally? 2019-10-27T23:56:28 < antto> now, does anyone know roughtly how much instruction cycles an expf() take on a cortex M4F? 2019-10-27T23:56:52 < Jan-> and bear in mind, whatever OCPolarity is, it has to be set fresh every time you change PWM value. 2019-10-27T23:57:02 < kakim32> steve: is it that link I provided? 2019-10-27T23:57:07 < kakim32> is it locally? 2019-10-27T23:57:20 < kakim32> do you need to download it or does cube contain it? 2019-10-27T23:58:03 < Jan-> I looked through the c file that was mentioned earlier, stm32f4xx_hal_tim.c and "OCPolarity" is mentioned a few times but there is no explanation of it 2019-10-27T23:58:26 < Steffanx> Is overly complex imho, jadew 2019-10-27T23:59:16 < jadew> i2c always seemed complex to me 2019-10-27T23:59:19 < kakim32> Jan-: open the link I provided 2019-10-27T23:59:36 < kakim32> i2c multimaster always seemed complex to me jadew 2019-10-27T23:59:37 < antto> jadew i've never succeeded with i2c 2019-10-27T23:59:41 < jadew> all that multi-master crap complicates things 2019-10-27T23:59:45 < antto> so i use SPI shizzle 2019-10-27T23:59:51 < jadew> kakim32, exactly --- Day changed Mon Oct 28 2019 2019-10-28T00:00:02 < kakim32> you don't need to implement that 2019-10-28T00:00:08 < kakim32> unless you need to 2019-10-28T00:00:24 < antto> wisem32 2019-10-28T00:00:29 < steve> kakim32 yes that is what I'm talking about. I'd venture a guess it is bundled inside cubeMX but i'm not sure 2019-10-28T00:00:29 < jadew> I know, but the fact that the hardware has to support it, results in a bunch of extra registers and extra stuff to get in your way 2019-10-28T00:01:13 < Jan-> is that file likely to be called something like "STM32F4xx_DSP_StdPeriph_Lib_V1.8.0" 2019-10-28T00:01:13 < jadew> I wonder if the multi-master thing is even used that often 2019-10-28T00:01:19 < jadew> probably close to never 2019-10-28T00:01:36 < qyx> in smbus dynamic addressing for example 2019-10-28T00:01:55 < kakim32> Jan-: direct to youself to first google search result of "stm32 f4 package" 2019-10-28T00:02:45 < Jan-> I think that's what this is 2019-10-28T00:02:48 < Jan-> I'm not sure 2019-10-28T00:03:31 < kakim32> download it 2019-10-28T00:03:43 < kakim32> let's not think now for a moment 2019-10-28T00:03:52 < steve> no it's not _DSP_ whatever, it'll say Cube in it 2019-10-28T00:03:55 < Jan-> it's a whole procedure to even download the file 2019-10-28T00:03:57 < Jan-> forget it 2019-10-28T00:04:03 < steve> lol 2019-10-28T00:04:07 < Jan-> I'm way beyond the "this is clearly never going to work" 2019-10-28T00:04:08 < qyx> wtf lol 2019-10-28T00:04:22 < steve> barrier to entry has clearly defeated Jan- 2019-10-28T00:04:35 < antto> me too 2019-10-28T00:04:45 < Jan-> I don't mind the barrier to entry, it's the fact that it is not going to change 2019-10-28T00:04:51 < Jan-> it's just a barrier to everything 2019-10-28T00:04:55 < qyx> ok, so 2019-10-28T00:04:58 < antto> i gave up and used atmel.start 2019-10-28T00:04:59 < qyx> me as a linux user 2019-10-28T00:05:04 < qyx> I'll give it a try 2019-10-28T00:05:07 < qyx> lets count minutes 2019-10-28T00:05:11 < steve> like the coding bootcamps. "Are you comforable sitting in a chair, and are you able to type on a keyboard?" Oh, if not, I'm afraid coding is not for you 2019-10-28T00:05:30 < specing> easy Jan- 2019-10-28T00:05:32 < specing> Jan-: https://www.st.com/content/st_com/en/about/careers/careers.html 2019-10-28T00:06:05 < Jan-> inside the zip file I have there is a folder called "stm32f4xx_stdperiph_examples" 2019-10-28T00:06:15 < Jan-> and in that there is a folder called TIM 2019-10-28T00:06:15 < kakim32> sweet 2019-10-28T00:06:23 < Jan-> and in that there is a folder called TIM_PWMOutput 2019-10-28T00:06:27 < kakim32> sweet sweet 2019-10-28T00:06:32 < steve> ok that soudns like it Jan- 2019-10-28T00:06:47 * qyx downloading 667MB over LTE 2019-10-28T00:07:06 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-28T00:07:10 < kakim32> so cube doesn't provide these resources locally? 2019-10-28T00:07:34 < kakim32> but you need to download a zip? ;..( 2019-10-28T00:08:05 < Jan-> OK so the problem here is that this is not for cube 2019-10-28T00:08:11 < Jan-> at least it isn't for the HAL library 2019-10-28T00:09:24 < kakim32> indeed stdperiph sounds like lower level 2019-10-28T00:09:43 < kakim32> is there HAL anywhere if you go back in hierarchy? 2019-10-28T00:09:55 < Jan-> Well, I don't have a "TIM_OC3PreloadConfig" to call for instance 2019-10-28T00:10:04 < Jan-> it literally isn't in the API 2019-10-28T00:10:07 < Jan-> so that doesn't work 2019-10-28T00:10:51 < qyx> actually Jan- wtf 2019-10-28T00:11:01 < qyx> there is a 1800 page long manual for F4 HAL 2019-10-28T00:11:10 < qyx> OCPolarity is clearly documented inside 2019-10-28T00:11:17 < kakim32> where is the doc? 2019-10-28T00:11:17 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-28T00:11:17 < qyx> I just downloaded it for the first time in my life 2019-10-28T00:11:22 < qyx> there are multiple sections 2019-10-28T00:11:26 < qyx> one documenting the whole timer thing 2019-10-28T00:11:32 < qyx> there are values described 2019-10-28T00:11:32 < Jan-> downloaded what 2019-10-28T00:11:41 < kakim32> same zip 2019-10-28T00:11:48 < qyx> open F4 2019-10-28T00:11:48 < qyx> pagehttps://www.st.com/content/st_com/en/products/embedded-software/mcu-mpu-embedded-software/stm32-embedded-software/stm32cube-mcu-mpu-packages/stm32cubef4.html?dl=8jx1mdfNcrLzFldwxb7JVA%3D%3D%2CO60oCUioXy8nw0N1vuShsp%2Fr1YDN58HpMY%2BWaQUwWeMr8C40WwzrPUdjBcsiFxEam2%2FUY32WkeJ%2FVeGCaszM9%2BoWBo0we4uQmBqqnTGrkC5KoEQzE88temnylNkdmPu83G2sUqx%2Bst2YY9OW5Lxc1e9raYeFz%2B6hMjFpkAFTco8bMn%2F2ProjvhYCXL4gHIbh1ulxHX2bj 2019-10-28T00:11:54 < qyx> Y5ZrYB5dmmyIRW6xQuCU8f4E6TqIjMl6xn2Hi3f1tpl4dFtseyUZ8FofaVm7ynChV7xsWD8r28Zrf6WR0%2BcAcSwg%2FwPoTUOqAsq%2Fi9KVFsrzCaFGN%2Bg0eR4MBnxBTgFE8VDayTEeZEgkw%3D%3D#resource 2019-10-28T00:11:57 < qyx> fuk the url 2019-10-28T00:12:05 < qyx> user manuals section 2019-10-28T00:12:14 < qyx> Description of STM32F4 HAL and LL drivers 2019-10-28T00:12:26 < qyx> no registration, direct download 2019-10-28T00:12:30 < antto> aww, they encoded yer DNA into teh URL 2019-10-28T00:12:32 < qyx> a single, concise pdf 2019-10-28T00:13:05 < qyx> there is everything on the same place 2019-10-28T00:13:18 < qyx> I found it in 2 minutes at most 2019-10-28T00:13:28 < qyx> including waiting for 40MB pdf 2019-10-28T00:14:02 < qyx> also cube examples for F4 2019-10-28T00:14:40 < kakim32> you need to know what to search for though 2019-10-28T00:14:58 < qyx> "stm32 F4 manual examples"? 2019-10-28T00:15:28 < qyx> and if not, there is a software section for F4 2019-10-28T00:15:31 < qyx> directly on st.com 2019-10-28T00:15:37 < kakim32> well that 2019-10-28T00:15:48 < qyx> you just need to want to find it 2019-10-28T00:16:24 < antto> hm, there's no exp instruction in teh FPU? 2019-10-28T00:16:33 < antto> so it must be a library function then? 2019-10-28T00:16:40 < kakim32> usually all the resources are in the vendors page 2019-10-28T00:16:50 < kakim32> under resources / software 2019-10-28T00:17:00 < Jan-> 640 megabytes 2019-10-28T00:17:05 < Jan-> this had better be an actual book 2019-10-28T00:17:06 < Jan-> hard cover 2019-10-28T00:17:09 < Jan-> bound in human skin 2019-10-28T00:17:47 < Mangy_Dog> :o 2019-10-28T00:17:51 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-28T00:17:59 < Jan-> ooh, dogskin! 2019-10-28T00:18:08 * Jan- strokes Mangy_Dog's fur... 2019-10-28T00:18:22 < Mangy_Dog> stm32 Necronomicon edition... Rituals and rites for the Sacrificial programming 2019-10-28T00:19:00 < qyx> the manual is 40MB 2019-10-28T00:19:10 < Jan-> good name for a menu in the restaurant of the damned 2019-10-28T00:19:17 < Jan-> Necronomnomnomicon 2019-10-28T00:19:38 < Mangy_Dog> sounds like a canabal restraunt 2019-10-28T00:19:40 < Mangy_Dog> :p 2019-10-28T00:20:10 < qyx> also, google for "stm32f4 reference manual", click first hit, download the refman as mentioned multiple times, 30MB 2019-10-28T00:20:25 < qyx> go to timer section, functional description, pwm mode 2019-10-28T00:20:31 < Jan-> oh I've got the "getting started" yes 2019-10-28T00:20:39 < Jan-> the problem is that talks about all of the demo projects that exist 2019-10-28T00:20:45 < Jan-> whereas none of them are actually designed for my board 2019-10-28T00:20:50 < qyx> sirprisingly, the 4th paragraph describes the OC polarity setting 2019-10-28T00:21:06 < qyx> no, I am talking about the reference manual 2019-10-28T00:21:06 -!- onio [~onio@2a00:23c5:7a01:8600:5ac:3ace:b590:c891] has quit [Quit: Leaving] 2019-10-28T00:21:11 < qyx> which is the same for all boards 2019-10-28T00:21:30 < Jan-> yes I have the reference manual 2019-10-28T00:21:31 < Jan-> that's fine 2019-10-28T00:21:45 < qyx> then I am curious wheres the problem 2019-10-28T00:22:01 < antto> Jan- is this the first time yer reading datasheets? 2019-10-28T00:22:12 < kakim32> reference manual is not a datasheet 2019-10-28T00:22:22 < antto> wutever 2019-10-28T00:22:30 < antto> "long technical PDFs" 2019-10-28T00:22:30 < kakim32> thanks 2019-10-28T00:22:33 < Jan-> the problem is that the reference manual does not tell you how to use the HAL as it exists in cube ide. 2019-10-28T00:22:48 < Jan-> you end up knowing okay, I need to set this, this and this 2019-10-28T00:22:51 < kakim32> it doesn't 2019-10-28T00:23:04 < Jan-> then you go into cube and it generates a pile of code and you think okay now what 2019-10-28T00:23:07 < Jan-> that's where I am now 2019-10-28T00:23:30 < kakim32> did you find that pdf of HAL? 2019-10-28T00:24:17 < antto> either find a documentation for the code generator thing or toss it and use the docs u got already and go barefoot ;P~ 2019-10-28T00:24:22 < antto> i mean baremetal 2019-10-28T00:24:27 < antto> heavymetal 2019-10-28T00:24:33 < Jan-> I looked in en.stm32cubef4.zip\STM32Cube_FW_F4_V1.23.0\Documentation 2019-10-28T00:24:39 < aandrew> lol Mangy_Dog 2019-10-28T00:24:48 < Jan-> there is one file called STM32CubeF4GettingStarted.pdf 2019-10-28T00:24:53 < Jan-> which does not contain the string "pwm" 2019-10-28T00:25:20 < antto> well, it contains the string "CubeF4" so 2019-10-28T00:25:21 < aandrew> the reference manual is for the part, not one possible library to use the part 2019-10-28T00:25:30 < kakim32> qyx: give the guy exat name of hal pdf 2019-10-28T00:25:52 < qyx> already closed it 2019-10-28T00:26:02 < qyx> isnt she a girl 2019-10-28T00:26:06 < kakim32> did it had HAL in name? 2019-10-28T00:26:23 < aandrew> Jan-: did you literally put "stm32cube getting started" into google (without the quotes) 2019-10-28T00:26:30 < Jan-> no. 2019-10-28T00:26:42 < Jan-> I downloaded en.stm32cubef4.zip as directed 2019-10-28T00:26:56 < aandrew> Jan in north america is typically a girl's name, but in europe it's a guy's name 2019-10-28T00:27:00 < Jan-> there is a folder in it called "STM32Cube_FW_F4_V1.23.0" 2019-10-28T00:27:04 < aandrew> not sure where Jan's from but it hardly matters 2019-10-28T00:27:09 < Jan-> inside that is a folder called "Documentation" 2019-10-28T00:27:22 < Jan-> It's short for Janine, I was born in California, I live near London now 2019-10-28T00:27:24 < aandrew> Jan-: funny, googling for 'getting started' is usually one of my first things when I'm trying to learn something new 2019-10-28T00:27:31 < Jan-> aandrew no kidding. 2019-10-28T00:27:39 < steve> CubeMX should initialize PWM entirely, ready to go, if you want maximum hand-holding and training wheels 2019-10-28T00:27:54 < Jan-> steve: I agree that it should but it does not give you any information on how to turn it on. 2019-10-28T00:27:57 < steve> and then generate the Cube IDE project 2019-10-28T00:28:12 < kakim32> qyx: repaste link then :o 2019-10-28T00:28:19 < antto> kewb eye-dee-ee 2019-10-28T00:28:26 < Jan-> as far as I know, the HAL API as provided in cube ide is has no documentation beyond the doxygen comments in the source. 2019-10-28T00:28:34 < Jan-> which is... er 2019-10-28T00:28:39 < Jan-> odd 2019-10-28T00:28:45 < Jan-> to put it mildly 2019-10-28T00:28:50 < kakim32> search for hal inside that zip 2019-10-28T00:28:58 < qyx> kakim32: https://www.st.com/en/embedded-software/stm32cubef4.html#resource 2019-10-28T00:29:00 < aandrew> Jan-: have you looked at any of these links: https://imgur.com/a/pCGY8h3 2019-10-28T00:29:08 < steve> well it's not going to autogenerate specific documentation for every possible permutation of using CubeMX to generate a project... 2019-10-28T00:29:11 < qyx> UM1725 2019-10-28T00:29:17 < qyx> Description of STM32F4 HAL and LL drivers 2019-10-28T00:29:32 < kakim32> aa user manuals opens a list of shit 2019-10-28T00:29:34 < aandrew> that sounds very familiar 2019-10-28T00:29:37 < kakim32> first one there 2019-10-28T00:30:13 < antto> maybe i'll stick to SAM chips 2019-10-28T00:30:14 < aandrew> Jan-: you've been at this at least all afternoon... 2019-10-28T00:30:27 < qyx> you mean the last two weeks 2019-10-28T00:30:31 < aandrew> antto: I've found mchip's docu WAY worse than stm 2019-10-28T00:30:43 < Jan-> aandrew I know it is completely absurd 2019-10-28T00:30:57 < aandrew> and while I have no first-hand experience I've been told that samd parts are buggy as all hell 2019-10-28T00:30:57 < Jan-> I don't think I'm particularly stupid, this should not be this hard 2019-10-28T00:31:08 < kakim32> UM1725 Jan- 2019-10-28T00:31:21 < aandrew> Jan-: I'm just confused as to why you haven't checked out "stm32cubemx getting started" on google and going from there 2019-10-28T00:31:26 < kakim32> DM00105879.pdf 2019-10-28T00:31:32 < antto> aandrew i have been eyeballing the SAME parts 2019-10-28T00:32:02 < Jan-> kakim32: Oh, right OK. 2019-10-28T00:32:13 < qyx> antto: did you get eye glaucome 2019-10-28T00:32:19 < Jan-> you just have to know to google for "ST UM1725" and it pops right up. 2019-10-28T00:32:28 < Jan-> obvious when you think about it 2019-10-28T00:32:39 < kakim32> or go to resources at F4 page on st.com 2019-10-28T00:32:41 < Jan-> makes you wonder what UM1724 is. 2019-10-28T00:32:47 < antto> qyx i cursed at microsh*t's gray tables 2019-10-28T00:33:05 < Jan-> oh I have seen this before yes 2019-10-28T00:33:17 < aandrew> Jan-: funny, first 3 hits on "stm32cubemx user manual" are relevant 2019-10-28T00:33:30 < kakim32> aandrew: but who would want to get started when you can pass that step 2019-10-28T00:33:59 < aandrew> I'm just countering the "you just have to google for UM1725 and it pops right up" snark 2019-10-28T00:34:12 < antto> yeah i don't wanna get started, i wanna b coding muh audio loop code already 2019-10-28T00:34:25 < kakim32> I mean I have started looking those gettings started stuffs and other resources just latelly 2019-10-28T00:34:43 < Jan-> the thing is that document is basically just a reiteration of the doxygen comments 2019-10-28T00:34:46 < Jan-> just formatted a bit 2019-10-28T00:34:52 < aandrew> I mean I'm not gonna say stm's docs are amaze but come on, some basic research skill is required 2019-10-28T00:35:07 < kakim32> cmon it has readable text in it Jan- 2019-10-28T00:35:09 < aandrew> Jan-: yes, and have you checked out any of the videos that show up when googling that? 2019-10-28T00:35:27 < Jan-> That's where I got "HAL_TIM_PWM_Start(&htim1, TIM_CHANNEL_1);" 2019-10-28T00:35:46 < Jan-> but I call that and it doesn't PWM, so presumably I got something in the cube mx part wrong 2019-10-28T00:36:05 < aandrew> hell "stm32cubemx first steps" gives you https://www.st.com/content/st_com/en/support/learning/stm32-education/stm32-step-by-step.html 2019-10-28T00:36:31 < qyx> step 5, build iot 2019-10-28T00:36:36 < qyx> step 6: profit 2019-10-28T00:36:37 < aandrew> yeah I lol'd at that 2019-10-28T00:37:16 < Cracki> uh, the topic is helplessness? 2019-10-28T00:38:02 < kakim32> 1) blink a led 2) iot 3) profit 2019-10-28T00:38:12 < aandrew> but step 2 would get you familiar with some of the basic HAL, and then googling for "stm32 pwm example" gives https://visualgdb.com/tutorials/arm/stm32/pwm/ as a first hit 2019-10-28T00:38:25 < aandrew> I gotta ask, Jan- what exactly are you trying to achieve here 2019-10-28T00:38:32 < Cracki> that meme must contain a "stealing undies" step 2019-10-28T00:38:32 < steve> "Internet of Stings" - Richard Stallman 2019-10-28T00:38:48 < Jan-> right now, my intention is to find something with slightly less awful documentation 2019-10-28T00:38:51 < aandrew> because it seems that you are intentionally acting like you're incapable of this 2019-10-28T00:38:55 < kakim32> aandrew: I believe that is not HAL 2019-10-28T00:38:56 < Jan-> I'm going to go out on a limb here 2019-10-28T00:38:57 < Jan-> and say 2019-10-28T00:39:00 < Cracki> trust your gut, aandrew 2019-10-28T00:39:07 < Jan-> this is absolutely the worst documented piece of tech I have ever come across in my entire life. 2019-10-28T00:39:14 < Jan-> it is utterly joke worthy. 2019-10-28T00:39:31 < qyx> then you have not encountered enough whit 2019-10-28T00:39:35 < steve> lol, you haven't even seen the tip of the iceberg 2019-10-28T00:39:39 < qyx> s* I mean 2019-10-28T00:39:40 < Cracki> if you wanna laugh at something, look at ffmpeg documentation 2019-10-28T00:39:43 < aandrew> none of the things I've googled for assume any prior knowledge of HAL or stm32, just the fact that I'm using an stm32 and I've got cubemx in front of me 2019-10-28T00:39:45 < Jan-> OK, I'll limit that to "other than open source" 2019-10-28T00:39:50 < Cracki> that's doxygen dump as well 2019-10-28T00:39:52 < Jan-> open source documentation is hilarious 2019-10-28T00:39:58 < Jan-> this is almost that bad 2019-10-28T00:40:09 < Cracki> basically, anything that uses doxygen has made a mistake 2019-10-28T00:40:16 < qyx> ok, you are simply too negativistic 2019-10-28T00:41:01 < aandrew> kakim32: I thought so too, it looks awful HALish, but adding HAL to the search term pops up https://electronics.stackexchange.com/questions/179546/getting-pwm-to-work-on-stm32f4-using-sts-hal-libraries (with me as the accepted answer) along with https://www.waveshare.com/wiki/STM32CubeMX_Tutorial_Series:_PWM in the top 3 results 2019-10-28T00:41:12 < kakim32> Jan-: sadly dongs live in jp timezone else you two could shit on open source big time 2019-10-28T00:41:27 < aandrew> it's frustrating trying to help someone who seems to genuinely appears to have no intrested in actually being helped 2019-10-28T00:41:43 < aandrew> interest* 2019-10-28T00:42:01 < Cracki> the goal is to trap you into wasting your time 2019-10-28T00:42:13 < Jan-> andrew what I get from all the documents you have posted is "there are some functions I might or might not have to call" 2019-10-28T00:42:27 < Jan-> there is no description of how to set up the peripheral in mx 2019-10-28T00:42:29 < aandrew> Jan-: so why not just follow an example, get something working, and experiment? 2019-10-28T00:42:31 < Jan-> That may be wrong 2019-10-28T00:42:35 < Cracki> there are things you don't need to respond to 2019-10-28T00:42:41 < qyx> true. 2019-10-28T00:42:45 < kakim32> page 873 on UM1725 Jan- 2019-10-28T00:42:56 < kakim32> or was it 837 idk idc 2019-10-28T00:43:01 < aandrew> are you one of those types of people who must have perfection and clear understanding before they even start? 2019-10-28T00:43:02 < kakim32> did you check all the steps 2019-10-28T00:43:03 < Jan-> I call HAL_TIM_PWM_Start(&htim1, TIM_CHANNEL_1); 2019-10-28T00:43:05 < Jan-> it compiles 2019-10-28T00:43:10 < Jan-> but no pulses 2019-10-28T00:43:20 < aandrew> Jan-: follow. an. example. 2019-10-28T00:43:25 < Cracki> there are things one should not respond to 2019-10-28T00:43:33 < steve> Jan- post your code 2019-10-28T00:43:33 < Cracki> <º))))>< 2019-10-28T00:43:37 < qyx> true. 2019-10-28T00:43:54 < kakim32> 2 weeks in and code review have been suggested 2019-10-28T00:44:01 < kakim32> this is new record 2019-10-28T00:44:27 < qyx> Jan-: again, kakim32 gave you a page with a list of things you have to do or my do if you want in order to get the timer working 2019-10-28T00:44:41 < Jan-> basically when I hit f8 to get past the reset breakpoint the pin we think should be pwm just goes high 2019-10-28T00:44:43 < Jan-> that's it 2019-10-28T00:44:47 < qyx> aandrew gave you an example which should be working containing all the steps you need 2019-10-28T00:44:53 < qyx> why you don't try them? 2019-10-28T00:44:59 < aandrew> hell my e.se answer (https://electronics.stackexchange.com/a/239783/17429) gives you step.by.step based on the code in the question but this must be too easy for you 2019-10-28T00:45:27 < Cracki> qyx, you make the mistake of assuming that this isn't about wasting people's time :) 2019-10-28T00:45:47 < kakim32> usually newbs just join and first thing dump their code and ask what is wrong.. got to give it to Jan- for not to 2019-10-28T00:45:51 < aandrew> Cracki: I'm giving Jan- the benefit of the doubt but I do feel like I'm being trolled 2019-10-28T00:46:06 < Cracki> duck typing. 2019-10-28T00:46:18 < Cracki> it doesn't have to be a troll. it just has to functionally be one. 2019-10-28T00:47:01 < Jan-> so hang on 2019-10-28T00:47:20 < Jan-> do I have to write HAL_TIM_PWM_MspInit myself 2019-10-28T00:47:26 < Jan-> or does that come from a library 2019-10-28T00:48:08 < aandrew> Jan-: this is where reading about how CubeMX and HAL is structured would give you that answer 2019-10-28T00:48:13 < Cracki> see 2019-10-28T00:48:21 < Cracki> that question is answered in aandrew's linked answer 2019-10-28T00:48:24 < kakim32> Jan-: it says you don't 2019-10-28T00:48:34 < Cracki> item 2 in his answer 2019-10-28T00:48:36 < steve> cubeMX should write it for you, but you can also just look in your project to see if it's defined 2019-10-28T00:48:47 < Jan-> so why does that guy on stackexchange say "Your MspInit() function is good, except you should also enable the clock of the GPIOC peripheral." 2019-10-28T00:48:49 < Cracki> she basically picks questions others have asked or answered, and asks them again 2019-10-28T00:48:56 < Jan-> in that case he has written the function himself presumably 2019-10-28T00:49:14 < Jan-> or has he 2019-10-28T00:49:25 < Cracki> if you had used cubemx, you would have had all this code generated for you already 2019-10-28T00:49:31 < aandrew> Jan-: HAL_x_MspInit() is a function you write which does the low level "connect a peripheral to a pin" kind of work. Enable the clock, configure the GPIO for an alternate function, etc. HAL provides "skeleton" code for this which CubeMX tries to guess exactly what you want and it can get it right a lot of the time 2019-10-28T00:49:44 < catphish> jlpcb don't have stm32l433 in stock, arse 2019-10-28T00:49:47 < kakim32> cube has done that function right? 2019-10-28T00:49:55 < aandrew> Jan-: I am that guy on stackexchange 2019-10-28T00:49:57 < Cracki> cube generates everything 2019-10-28T00:50:21 < Jan-> ok well let's go step by step 2019-10-28T00:50:31 < Jan-> do I need to "zero-out my stack allocated structs" 2019-10-28T00:50:46 < aandrew> if you initialize all the elements, no. I do it because I like the order it creates 2019-10-28T00:51:01 < Jan-> I haven't knowingly created any structs, stack allocated or otherwise 2019-10-28T00:51:07 < aandrew> sure you have 2019-10-28T00:51:14 < aandrew> TIM_MasterConfigTypeDef sMasterConfig; 2019-10-28T00:51:19 < aandrew> what do you think that does? 2019-10-28T00:51:26 < aandrew> TIM_OC_InitTypeDef sConfigOC; 2019-10-28T00:51:28 < aandrew> as well 2019-10-28T00:51:36 < Jan-> I haven't the faintest clue. 2019-10-28T00:51:38 < aandrew> those are stack-allocated structures which will contain garbage 2019-10-28T00:51:48 < qyx> you shall never use uninitialized automatic variable 2019-10-28T00:51:53 < aandrew> Jan-: may I ask, what is your programming experience elvel? 2019-10-28T00:51:55 < aandrew> level 2019-10-28T00:52:00 < Jan-> I didn't create those, mx did, those are in MX_TIM1_Init 2019-10-28T00:52:01 < aandrew> not in embedded, just in general 2019-10-28T00:52:16 < aandrew> Jan-: perhaps I missed the link, where is the link to your code? 2019-10-28T00:52:32 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Quit: veegee] 2019-10-28T00:52:33 < Cracki> cubemx generated code lacks HAL_TIM_PWM_Start() 2019-10-28T00:52:33 < Jan-> I've written quite a lot of c# and javascript, and c for avr 2019-10-28T00:52:47 < Cracki> otherwise it's complete in all the setup stuff 2019-10-28T00:53:00 < Jan-> The entirety of the code I have contributed to this is "HAL_TIM_PWM_Start(&htim1, TIM_CHANNEL_1);" 2019-10-28T00:53:03 < Jan-> which does nothing 2019-10-28T00:53:04 < aandrew> Jan-: ok, that isn't meant as an intimidating or posturing question, I'm trying to understand where you're coing from 2019-10-28T00:53:12 < Cracki> if you don't use cubemx, but do what the SO question does i.e. roll your own, then you have to pay more attention 2019-10-28T00:53:15 < Jan-> I have no idea if it's even useful to what I'm trying to do 2019-10-28T00:53:26 < aandrew> Jan-: that does not answer my question. you have a file with C code in it that you've added to 2019-10-28T00:53:47 < aandrew> paste a link to that code please. pastebin.com is handy if you don't have an online pastebin of choice. 2019-10-28T00:54:05 < aandrew> just copy and paste the etire file that MX_TIM1_Init and your code is in 2019-10-28T00:55:38 < steve> a mutation test framework would probably have figured out the correct initialization by this point :P 2019-10-28T00:55:47 < Jan-> https://pastebin.com/JXEP10G2 2019-10-28T00:55:58 < Jan-> steve you're not kidding, I was honestly considering a fuzz tester 2019-10-28T00:56:22 < steve> well if you can reliably test for the correct result, it's not a terrible idea 2019-10-28T00:56:22 < aandrew> https://pastebin.com/mjBNNT88 is my timer code which uses HAL to do interesting things with timer outputs 2019-10-28T00:56:37 < Jan-> and yes, the led on pin A6 blinks 2019-10-28T00:56:55 < Jan-> and yes, the DAC switches values, and the USART emits characters 2019-10-28T00:57:06 < aandrew> ok, so MX_TIM1_Init() goes about setting things up 2019-10-28T00:57:37 < aandrew> (notice how they initialize all their variables. e.g. TIM_MasterConfigTypeDef sMasterConfig = {0}; 2019-10-28T00:57:44 < Jan-> yeah, but that's an absolutely massive wad of code-we-don't-know-what-it-does. 2019-10-28T00:57:50 < kakim32> what are the intended pins of PWM? 2019-10-28T00:57:59 < aandrew> I dislike that way of doing it, but that's just preference 2019-10-28T00:58:11 < aandrew> Jan-: that's not hte point. let's continue. 2019-10-28T00:58:12 < Jan-> Also I have no way of knowing that they're initialising all of the struct members. 2019-10-28T00:58:26 < Jan-> If there were other struct members I couldn't tell. 2019-10-28T00:58:34 < aandrew> so they set up htim1 (handle for timer1) 2019-10-28T00:58:55 < aandrew> Jan-: this is why I asked your experience level. You know all elements are initialized because that's (one) way to do it in C 2019-10-28T00:59:16 < aandrew> memest(&sMasterConfig, 0, sizeof(sMasterConfig); is another 2019-10-28T00:59:24 < Jan-> honestly, I hadn't even looked at the contents of that 2019-10-28T00:59:28 < Jan-> none of it is going to have any meaning to me 2019-10-28T00:59:34 < aandrew> ok 2019-10-28T00:59:38 < aandrew> perhaps I am wasting my time here 2019-10-28T00:59:44 < Jan-> if I was going to go through it at that level I'd say screw the HAL and write it from scratch myself. 2019-10-28T00:59:46 < aandrew> you do not seem at all even remotely curious how the system works 2019-10-28T00:59:56 < aandrew> how do you normally approach a problem? 2019-10-28T01:00:27 < aandrew> do you wait for the documentation to be written and if there's a missing comma or a spelling mistake, send it back and declare "I'm sorry, but this is not right, I can't work with tis." 2019-10-28T01:00:42 < Jan-> I would have thought the entire purpose of having the automatic configuration tool was to avoid having to go through the entire thing in detail. 2019-10-28T01:01:01 < Jan-> the problem is they never give you the last step 2019-10-28T01:01:08 < Jan-> they set it up but they never tell you how to turn it on. 2019-10-28T01:01:10 < aandrew> the point of the HAL is to abstract away the register reads and writes. You still need to have a very good understanding of how it needs to be manipulated in order to achieve your goal 2019-10-28T01:01:20 < steve> Jan- your code is incomplete. Is taht the entirety of what CubeMX spat out? 2019-10-28T01:01:32 < aandrew> the point of the HAL is not to give you "main screen turn on" high level abstraction 2019-10-28T01:01:46 < aandrew> it's a way to abstract away the low level details across multiple processor families 2019-10-28T01:02:22 < aandrew> steve: there'll be a HAL_MSP file as well 2019-10-28T01:02:31 < Jan-> so why did they create the gui configuration tool 2019-10-28T01:02:34 < Jan-> I didn't even want to use it 2019-10-28T01:02:39 < kakim32> what is the part of this that links peripherals to pins? #newtostm32 2019-10-28T01:02:53 < aandrew> antto: perhaps the point is to learn 2019-10-28T01:02:57 < aandrew> er Jan- 2019-10-28T01:03:37 < aandrew> they created the gui config tool to help you write a whack of code so you just need to supply your "business logic" 2019-10-28T01:03:41 < aandrew> it works ... sort of 2019-10-28T01:03:59 < aandrew> it's a good starting point but again unless you're willing to learn the HAL you may as well just learn the raw registers 2019-10-28T01:04:16 < aandrew> one way or another, you'll need to learn the system you want to work within. HAL's a lot easier than the registers 2019-10-28T01:04:25 < steve> kakim32 that'd be in the xyz_hal_msp.c file, or the MspInit() functions you write yourself, or copy from refernce projects 2019-10-28T01:04:45 < Jan-> the thing is, if you went directly to the registers, you could at least use the reference manual and you'd have some idea what was going on. 2019-10-28T01:05:03 < kakim32> Jan-: show mspinit 2019-10-28T01:05:05 < catphish> this pcb assembly thing is hassle, i guess i'll just continue to do it by hand until i can commit to large volume 2019-10-28T01:05:34 < kakim32> catphish: are you getting any better at it? 2019-10-28T01:05:36 < catphish> jlpcb seems like a magic bullet, but being out of stock of the stm32 kills it dead :( 2019-10-28T01:05:38 < aandrew> steve: I was going ot get to that but honestly I'm a little put off because in the span of an afternoon of trying to get PWM ot work Jan hasn't even attempted to walk through the MX_TIM1_Init() code to see how it was calling HAL 2019-10-28T01:05:43 < Jan-> right now you put some hopefully correct settings into cube mx, it generates miles of stuff, and you have no clue whether it's right or wrong because it doesn't tell you anything about the settings, or anything about how to actually start the peripheral working. 2019-10-28T01:05:47 < catphish> kakim32: i don't do it enough to be getting better really 2019-10-28T01:06:04 < aandrew> I'm gonna bugger off for a while 2019-10-28T01:06:13 < BrainDamage> catphish: mike on youtube has good tips how to get more efficient 2019-10-28T01:06:15 < kakim32> do you paste it or hand solder catphish? 2019-10-28T01:06:19 < BrainDamage> like how to arrange tapes, etc 2019-10-28T01:06:20 < steve> aandrew ah my bad : \ 2019-10-28T01:06:20 < Jan-> andrew I can go through the code but all I can do is to read the source of say HAL_TIMEx_ConfigBreakDeadTime. 2019-10-28T01:06:21 < qyx> I'm gonna switch you all off 2019-10-28T01:06:39 < jadew> catphish, you mean with the pnp machine or by hand? 2019-10-28T01:06:44 < catphish> kakim32: depends, i prefer paste, but for a one off it's a waste of money to make a stencil 2019-10-28T01:06:54 < kakim32> it never is 2019-10-28T01:06:56 < jadew> catphish, get a paste dispenser 2019-10-28T01:07:03 < steve> Jan- it doesn't generate miles, it's prbaboy not even 500 lines 2019-10-28T01:07:09 < jadew> I use it a lot 2019-10-28T01:07:26 < kakim32> paste is a way to go catphish 2019-10-28T01:07:41 < kakim32> stencils cost like pff 2019-10-28T01:07:43 < kakim32> nothing 2019-10-28T01:07:48 < kakim32> 10usd 2019-10-28T01:08:02 < BrainDamage> your sanity is worth more than a stencil 2019-10-28T01:08:35 < catphish> maybe i'll try stenciling a batch :) 2019-10-28T01:08:44 < catphish> and get better at that 2019-10-28T01:08:56 < catphish> maybe even buy an oven 2019-10-28T01:08:59 < Jan-> steve for what it's worth this file has 378 lines and I've only written about 10 of them. 2019-10-28T01:09:02 < jadew> wait... you tried production without solder paste/stencil? 2019-10-28T01:09:25 < kakim32> I did baking on electric stove 2019-10-28T01:09:31 < kakim32> + reflow gun 2019-10-28T01:09:46 < kakim32> it was sweaty stuff 2019-10-28T01:09:56 < kakim32> radiated heat is suprisingly harsh 2019-10-28T01:10:20 < jadew> I keep my oven on the balcony 2019-10-28T01:10:31 < kakim32> Jan-: mspinit show it 2019-10-28T01:11:04 < Jan-> don't know what you mean kakim 2019-10-28T01:11:10 < steve> Jan- that's nothing just read it, look 1 level deeper, then you'll see the bottom 2019-10-28T01:11:13 < Jan-> there are lots in all the different the source files 2019-10-28T01:11:18 < Jan-> steve I am fully aware. 2019-10-28T01:11:35 < Cracki> don't do stenciling without an oven or at least a hotplate. it only makes sense if you don't have to go at it with an iron anyway 2019-10-28T01:11:42 < Cracki> *at all of it 2019-10-28T01:12:13 < kakim32> guise: where is mspinit stored? 2019-10-28T01:12:14 < Cracki> for prototypes a paste dispenser is just fine. you do wanna play with a hotplate or oven. 2019-10-28T01:12:45 < Jan-> if I actually do what aandrew suggested and go through everything to understand it, following all the #includes, I will end up in a file called stm32f4xx_hal_rcc_ex.h 2019-10-28T01:12:51 < Jan-> which is over seven thousand lines 2019-10-28T01:12:53 < Cracki> kakim32, https://knowyourmeme.com/memes/pee-is-stored-in-the-balls 2019-10-28T01:13:14 < Jan-> and yes, if you want to call me lazy, that is too much for me to go through to get a pwm output working. 2019-10-28T01:13:27 < kakim32> Jan-: you can write a call mspinit and autocomplete it then follow it where it resides 2019-10-28T01:13:29 < Cracki> you don't have to read those 7000 lines! 2019-10-28T01:13:32 < Cracki> you follow the function calls 2019-10-28T01:13:41 < Cracki> and if you don't care about RCC stuff, you step over it 2019-10-28T01:13:45 < Cracki> that's what a debugger does 2019-10-28T01:14:04 < Cracki> <º))))>< 2019-10-28T01:14:35 < Jan-> "no default proposals" 2019-10-28T01:15:19 < steve> of those 7000 lines, only a few functions are called, if not just 1 2019-10-28T01:16:35 < kakim32> HAL_TIM_PWM_Init did call mspinit? 2019-10-28T01:16:57 < Jan-> steve that's where RCC_OscInitTypeDef is 2019-10-28T01:17:18 < kakim32> HAL_TIM_PWM_MspInit() 2019-10-28T01:17:20 < Jan-> kakim32: it calls HAL_TIM_MspPostInit(&htim1); if that's it. 2019-10-28T01:18:08 < kakim32> no Jan- MX_TIM1_Init calls that 2019-10-28T01:19:13 < kakim32> HAL_TIM_PWM_Init is called in that too 2019-10-28T01:19:18 < Jan-> that's just the only reference to "msp" that exists in main.c 2019-10-28T01:19:26 < kakim32> yes 2019-10-28T01:20:24 < kakim32> aandrews stockexchange post described: Don't call HAL_TIM_PWM_MspInit() manually. It is automatically called when you call HAL_TIM_PWM_Init(), but you must make sure your handle variables are properly initialized (see #1 above) 2019-10-28T01:21:00 < Jan-> I am assuming that the auto code generator properly initialized the variables. 2019-10-28T01:22:13 < Jan-> bear in mind I have never seen this HAL_TIM_PWM_MspInit() 2019-10-28T01:22:35 < steve> lol, did you look for it? 2019-10-28T01:22:43 < kakim32> just paste it to your code - hilight it and ctrl+shift+g 2019-10-28T01:22:57 < kakim32> it finds all the files with reference to such symbol 2019-10-28T01:23:26 < kakim32> including the function definition 2019-10-28T01:23:49 < kakim32> nasty way I find stuff with eclipse 2019-10-28T01:24:01 < Jan-> it's declared in stm32f4xx_hal_tim.h 2019-10-28T01:24:09 < Jan-> but I don't know where the body of the function is 2019-10-28T01:24:13 < kakim32> is there .c ? 2019-10-28T01:24:21 < Jan-> no idea 2019-10-28T01:24:40 < kakim32> did ctrl-shift-g give you a list of occurrences? 2019-10-28T01:25:25 < Jan-> only one 2019-10-28T01:25:35 < Jan-> OK so it's in stm32f4xx_hal_tim.c 2019-10-28T01:25:58 < steve> no, it's in xyz_msp.c 2019-10-28T01:26:33 < Jan-> HAL_TIM_PWM_Init is in stm32f4xx_hal_tim.c 2019-10-28T01:26:48 < kakim32> certainly 2019-10-28T01:27:06 < Jan-> and it calls HAL_TIM_PWM_MspInit(htim); on line 1193 2019-10-28T01:27:13 < kakim32> good 2019-10-28T01:27:24 < kakim32> where is that function definition 2019-10-28T01:27:26 < steve> which is in blahblah_msp.c 2019-10-28T01:27:40 < Jan-> it's hard to tell as I'm just doing text searches on the files 2019-10-28T01:27:42 < Jan-> hang on 2019-10-28T01:27:53 < Cracki> do not do text searches 2019-10-28T01:27:58 < Cracki> use your IDE to find those things 2019-10-28T01:28:06 < Cracki> ctrl-click on the name, or f12, or whatever it has 2019-10-28T01:28:09 < Cracki> "go to definition" 2019-10-28T01:28:18 < kakim32> or ctrl+shift+g 2019-10-28T01:28:20 < Jan-> doing that just gets me the header 2019-10-28T01:28:28 < Cracki> then do that again 2019-10-28T01:28:38 < kakim32> yes you do it again in header 2019-10-28T01:28:38 < Jan-> no file matching *msp.c in "Drivers" 2019-10-28T01:28:44 < Cracki> your IDE should know all the places it's defined 2019-10-28T01:28:51 < steve> it's not in drivers, it's next to main.c 2019-10-28T01:28:53 < kakim32> Jan-: it's not a driver 2019-10-28T01:29:03 < kakim32> it's not in that hierarchy 2019-10-28T01:29:22 < kakim32> drivers are solid stuff that is not generated / user writen code 2019-10-28T01:29:53 < Jan-> I have a HAL_TIM_Base_MspInit in stm32f4xx_hal_msp.c 2019-10-28T01:30:08 < kakim32> those are kept clearly separate 2019-10-28T01:30:17 < Jan-> so why did we just end up going through thm 2019-10-28T01:31:09 < kakim32> base is not enough 2019-10-28T01:31:13 < kakim32> we need PWM 2019-10-28T01:31:31 < kakim32> and clearly yet another driver file 2019-10-28T01:31:43 < Jan-> no instance of "pwm" in stm32f4xx_hal_msp.c 2019-10-28T01:32:05 < kakim32> do you have teamviewer? 2019-10-28T01:32:11 < Jan-> no 2019-10-28T01:32:29 < Jan-> look this is a waste of time 2019-10-28T01:32:37 < Jan-> it's clearly not going to work with any reasonable amount of effort 2019-10-28T01:32:46 < kakim32> you are looking within one file 2019-10-28T01:32:48 < Jan-> even if we fix this, I'll hit the same problem again and again every time I need a new peripheral to work 2019-10-28T01:32:51 < kakim32> that is driver file 2019-10-28T01:33:01 < Jan-> I can't make you guys go through all this for every single thing I do 2019-10-28T01:33:32 < kakim32> HAL_TIM_PWM_MspInit 2019-10-28T01:33:39 < kakim32> paste it to main no matter where 2019-10-28T01:33:40 < steve> "teach a man to fish, and he can initialize ARM Cortex-M peripherals for a lifetime" 2019-10-28T01:33:58 < kakim32> highlight it and then ctrl+shift+g 2019-10-28T01:34:40 < Jan-> no instances of HAL_TIM_PWM_MspInit in main.c 2019-10-28T01:34:57 < kakim32> oh shit did I give you the wrong shortcut 2019-10-28T01:35:07 < steve> I could use some consulting work if you still need the job done :P 2019-10-28T01:35:11 < kakim32> is it alt-shift-g? 2019-10-28T01:35:41 < kakim32> it should search the whole shit not just one file 2019-10-28T01:36:14 < kakim32> muscle memory can't be repeated verbally 2019-10-28T01:36:26 < catphish> kakim32: i have a hot plate and a hot air station, hopefully i can find a good way to combine them, or just buy an oven 2019-10-28T01:36:51 < Cracki> hot plate plus ir thermometer (or practice) 2019-10-28T01:37:38 < catphish> i've had pretty much 100% success with hot plate (apart from QFNs that i always manage to float) 2019-10-28T01:37:51 < catphish> but it's kind of an unpleasant process 2019-10-28T01:38:57 < Cracki> float because of exposed center pad with a bit much paste? 2019-10-28T01:39:03 < kakim32> Jan-: ctrl + click that HAL_TIM_PWM_MspInit then? 2019-10-28T01:39:17 < Jan-> I don't have anything to select 2019-10-28T01:39:20 < Jan-> that term is not in this file 2019-10-28T01:39:38 < kakim32> can you teamviewer? 2019-10-28T01:40:28 < Jan-> not very practically 2019-10-28T01:40:50 < kakim32> you can write that function call to main 2019-10-28T01:40:56 < kakim32> then click it 2019-10-28T01:41:04 < kakim32> then remove it after clicking it 2019-10-28T01:41:28 < kakim32> nasty trick I do 2019-10-28T01:41:46 < kakim32> I don't know how to eclipse and I have done it a while 2019-10-28T01:41:47 < Jan-> OK fine 2019-10-28T01:41:51 < Jan-> just gets me the header again 2019-10-28T01:41:57 < kakim32> click the header 2019-10-28T01:42:08 < Jan-> click what 2019-10-28T01:42:21 < kakim32> fucntion prototype 2019-10-28T01:42:31 < Jan-> you can't go-to-declaration from a header 2019-10-28T01:42:34 < Jan-> it thinks that is the declaration 2019-10-28T01:43:05 < BrainDamage> catphish: do you have a thermocouple on your dmm? 2019-10-28T01:43:19 < catphish> i think so, if i can find it 2019-10-28T01:43:29 -!- jly [uid355225@gateway/web/irccloud.com/x-berwwqsqtbayytmz] has quit [Quit: Connection closed for inactivity] 2019-10-28T01:43:30 < Jan-> the actual declaration of it is in stm32f4xx_hal_tim.c on line 1250 2019-10-28T01:43:33 < Jan-> __weak void HAL_TIM_PWM_MspInit(TIM_HandleTypeDef *htim) 2019-10-28T01:43:37 < BrainDamage> just kapton tape it to a waste pcb and place it on the hot plate as well 2019-10-28T01:43:39 < catphish> but i also have an IR thermometer gun that i use for this 2019-10-28T01:43:49 < Jan-> it's not nothing in it, just UNUSED(htim); 2019-10-28T01:43:50 < BrainDamage> and you'll have accurate temp continously updated 2019-10-28T01:44:02 < kakim32> yes its weak Jan- 2019-10-28T01:44:02 < catphish> BrainDamage: that seems like a good idea 2019-10-28T01:44:16 < catphish> no idea if the gun is accurate or not 2019-10-28T01:44:18 < kakim32> because it's defined elsewhere it's just a placeholder to keep things happy 2019-10-28T01:44:31 < kakim32> *it should be defined elsewhere 2019-10-28T01:44:44 < Jan-> I forgot why we're even looking for it 2019-10-28T01:45:01 < kakim32> it defines the pins 2019-10-28T01:45:11 < kakim32> used for PWM 2019-10-28T01:45:15 < BrainDamage> catphish: hot plate can also be used as preheater btw 2019-10-28T01:45:20 < Jan-> not right now it doesn't 2019-10-28T01:45:32 < Jan-> doesn't the mx generated code do that 2019-10-28T01:45:47 < catphish> BrainDamage: yeah i was thinking hot plate + hot air might be better 2019-10-28T01:46:01 < kakim32> yes it should be in xyz_msp.c in your project not in drivers 2019-10-28T01:46:04 < BrainDamage> you can reflow fine with a hot plate provided you monitor temp 2019-10-28T01:46:13 < kakim32> replace xyz with something 2019-10-28T01:46:14 < BrainDamage> the preheater is for your convenience when reworking 2019-10-28T01:46:21 < catphish> BrainDamage: makes sense 2019-10-28T01:46:58 < catphish> anyway, i definitely can reflow with a hot plate, it works, it's just a bit unpleasant 2019-10-28T01:46:59 < kakim32> *_msp.c rather 2019-10-28T01:47:24 < catphish> also, only one board at a time really, not great for a panel 2019-10-28T01:47:31 < BrainDamage> does it have a thermostat at least/ 2019-10-28T01:47:36 < catphish> sure 2019-10-28T01:48:14 < Jan-> are you saying there should be a function called HAL_TIM_PWM_MspInit in my stm32f4xx_hal_msp.c 2019-10-28T01:48:20 < Jan-> because there isn't 2019-10-28T01:48:23 < kakim32> no 2019-10-28T01:48:25 < Jan-> there's a HAL_MspInit 2019-10-28T01:48:30 < catphish> really i suppose the main question is whether to embark on trying to make some boards with my hot plate, or buy a T962 2019-10-28T01:48:36 < Jan-> there's a HAL_DAC_MspInit. 2019-10-28T01:48:44 < kakim32> that sounds like a driver file 2019-10-28T01:48:47 < Jan-> a HAL_TIM_Base_MspInit 2019-10-28T01:49:07 < Jan-> there's a HAL_TIM_MspPostInit which does seem to talk about GPIO pins. 2019-10-28T01:49:25 < kakim32> guise is that hal file? 2019-10-28T01:49:38 < Jan-> it does talk about pin 9 2019-10-28T01:49:43 < Jan-> which is the one we asked it to use for the PWM 2019-10-28T01:50:03 < kakim32> well paste that file 2019-10-28T01:50:14 < Jan-> it does a HAL_GPIO_Init(GPIOE, &GPIO_InitStruct); after setting the .Pin in the struct to GPIO_PIN_9. 2019-10-28T01:50:19 < Jan-> so that has to be something to do with it 2019-10-28T01:50:39 < kakim32> what a weird file name for cube generated file 2019-10-28T01:50:54 < Jan-> it's in STM32CubeIDE\workspace_1.0.2\Test2\Src 2019-10-28T01:50:57 < Jan-> with my main.c 2019-10-28T01:51:10 < kakim32> then it's that 2019-10-28T01:51:20 < Jan-> so hang on 2019-10-28T01:51:22 < kakim32> paste dat .c 2019-10-28T01:51:24 < Jan-> are we saying this got so complicated 2019-10-28T01:51:31 < Jan-> that even you guys didn't know where to look 2019-10-28T01:51:49 < Jan-> https://pastebin.com/A1KYmvsu 2019-10-28T01:52:00 < kakim32> we knew but I have never done stm32 and that file name sounds like part of hal 2019-10-28T01:52:10 < catphish> looks like T962 ovens are garbage anyway 2019-10-28T01:52:17 < Jan-> so how on earth am I ever supposed to figure it out 2019-10-28T01:52:18 < kakim32> nobody else bothers at this point Jan- 2019-10-28T01:54:37 < kakim32> Jan-: you get a touch of things at time 2019-10-28T01:54:45 < kakim32> if you continue 2019-10-28T01:56:08 < steve> lol we know where to look, it's you that can't find the stuff. And it's not that complicated. You're just freaking out. Tomorrow it'll make a lot more sense 2019-10-28T01:56:49 < Jan-> so far setting up a PWM peripheral has involved six files containing a total of over ten thousand lines of code 2019-10-28T01:56:54 < Jan-> and I'm no nearer making it work 2019-10-28T01:57:09 < kakim32> hmm it doesn't have pwm mspinit but seems to do something in that other function 2019-10-28T01:57:45 < kakim32> I let others determine if that file has something off 2019-10-28T01:59:16 < steve> Does anyone have a preference: would you put a FreeRTOS task definition at the BSP level, if it was handling low level stuff, or do you keep all Tasks at the application level? 2019-10-28T02:00:26 < kakim32> application level 2019-10-28T02:02:05 < kakim32> Jan-: you could rename that function definition to be PWM.. I don't know what it would benefit but personally I would try that 2019-10-28T02:03:09 < kakim32> https://electronics.stackexchange.com/questions/179546/getting-pwm-to-work-on-stm32f4-using-sts-hal-libraries/239783#239783 I would also do the first step to make sure everything is zero 2019-10-28T02:05:18 < kakim32> then I would reCube 2019-10-28T02:05:34 < Jan-> look if you guys can't make it work, I have no chance 2019-10-28T02:05:45 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-28T02:06:08 < kakim32> others don't bother and I have 0 stm32 experience 2019-10-28T02:06:39 < kakim32> I'm literally looking at first time how to PWM F4 2019-10-28T02:06:41 < Cracki> this is like keyhole surgery. 2019-10-28T02:06:54 < Cracki> can't fault anyone for having to operate on your problem over a great distance. 2019-10-28T02:07:03 < Cracki> that makes some things impossible. 2019-10-28T02:07:26 < Cracki> your problem is that you are mentally resigned already 2019-10-28T02:07:27 < kakim32> it sure looks scary 2019-10-28T02:07:53 < Cracki> helplessness reinforces itself 2019-10-28T02:08:14 < steve> deep. 2019-10-28T02:08:33 < Cracki> another contributing factor is the relative inexperience with IDEs, which do a lot for you, if you let them 2019-10-28T02:08:43 < kakim32> but it probs hase amount of things to make pwm to work 2019-10-28T02:08:48 < Cracki> text search < look up definition 2019-10-28T02:08:51 < kakim32> at least in code quantity 2019-10-28T02:09:04 < Jan-> all we've done is set it up and let mx generate some code 2019-10-28T02:09:12 < Jan-> I am not claiming that code is right, I don't know if we set mx up properly 2019-10-28T02:09:23 < Jan-> and I don't know how to actually use what it has set up because none of it is documented anywhere. 2019-10-28T02:09:24 < kakim32> it didn't 2019-10-28T02:09:49 < Cracki> you should start assuming everything has worked ok. you can start questioning things when you have a reason for it. 2019-10-28T02:09:49 < kakim32> or wait 2019-10-28T02:10:04 < Cracki> and you can start putting in the work of reading the docs. 2019-10-28T02:10:23 < Cracki> and browsing the source, in small and justified parts 2019-10-28T02:10:55 < Jan-> well I'm questioning it because what we have doesn't work :) 2019-10-28T02:11:11 < Cracki> you could be a lot more methodical about this 2019-10-28T02:11:19 < kakim32> Cracki: simple question: should cube code output pwm with 2 calls in main()? 2019-10-28T02:11:25 < kakim32> init and start 2019-10-28T02:11:41 < Cracki> you question things so you can assure yourself that things either work or _how_ they fail 2019-10-28T02:11:52 < Cracki> "fuck it" isn't questioning 2019-10-28T02:12:08 < Cracki> cube doesn't generate the start() call 2019-10-28T02:12:28 < Cracki> it didn't do that for anything I did with it, it expects you do call start() 2019-10-28T02:12:51 < kakim32> yes 2019-10-28T02:12:58 < Cracki> it did (not) that for the pwm I did with it, and it did (not) that with the quadrature encoder input decoding 2019-10-28T02:13:13 < Cracki> +do 2019-10-28T02:13:18 < kakim32> but if those are called in main 2019-10-28T02:13:29 < kakim32> init 2019-10-28T02:13:32 < kakim32> and start 2019-10-28T02:13:39 < kakim32> is it all it requires in main? 2019-10-28T02:14:07 < kakim32> pwm configured for some output value 2019-10-28T02:14:07 < Cracki> I'm looking at a project where it does pwm and quadrature decoding successfully... 2019-10-28T02:14:12 < Cracki> and it generated the init calls in main 2019-10-28T02:14:22 < Cracki> but right after, _I_ had to insert the start calls 2019-10-28T02:14:29 < Cracki> user code begin 2, in my situation 2019-10-28T02:14:36 < Cracki> example: HAL_TIM_Encoder_Start(&htim1, TIM_CHANNEL_ALL); 2019-10-28T02:14:38 < Jan-> I understand that it would need me to put some code in to make it start 2019-10-28T02:14:40 < kakim32> in case of pwm it's one start? 2019-10-28T02:14:41 < Cracki> HAL_TIM_PWM_Start(&htim2, TIM_CHANNEL_3); 2019-10-28T02:14:46 < Cracki> one start per channel 2019-10-28T02:14:49 < Jan-> but it's never described exactly what that should be 2019-10-28T02:14:52 < kakim32> and then pwm comes out? 2019-10-28T02:14:59 < Cracki> I use four channels (two per motor) so I have four of those calls 2019-10-28T02:15:16 < Cracki> Jan-, there is docs and examples for that 2019-10-28T02:15:24 < Cracki> repeating the claim doesn't make it any more valid 2019-10-28T02:15:33 < Jan-> I've never seen them. 2019-10-28T02:15:45 < kakim32> Jan-: recube 2019-10-28T02:15:58 < kakim32> keep other things as those are 2019-10-28T02:15:58 < Cracki> we have a volk song for this "but what if the pot has a hole? then you fix the hole. what if you have nothing to fix the hole? then you get the material to fix the hole. ..." 2019-10-28T02:16:05 < Cracki> kakim32, yes and then pwm comes out 2019-10-28T02:16:15 < kakim32> play with that PWM section of things 2019-10-28T02:16:23 < Jan-> normally I'd google for info but that doesn't really work 2019-10-28T02:16:48 < machinehum> Jan-: ./STM32Cube/Repository/STM32Cube_FW_F3_V1.10.0/Projects/STM32F3-Discovery/Examples 2019-10-28T02:16:56 < kakim32> are you certain you are probing correct pin? 2019-10-28T02:16:59 < machinehum> Or whatever board you're using 2019-10-28T02:17:14 < kakim32> do you have oscope? 2019-10-28T02:17:17 < Cracki> the generated init calls consist of HAL_Init, SystemCLock_Config, MX_GPIO_Init, MX_TIM2_Init and so on.... and then my start calls, and then I can set TIM2->CCR3 to set duty cycle for channel 3 2019-10-28T02:17:38 < kakim32> is there initial duty cycle? 2019-10-28T02:17:42 < Jan-> I have a bunch of examples 2019-10-28T02:17:44 < Jan-> somewhere 2019-10-28T02:17:46 < Jan-> can't remember where 2019-10-28T02:17:50 < Cracki> >X doesn't work 2019-10-28T02:17:55 < machinehum> I just listed the dir 2019-10-28T02:17:56 < Cracki> well MAKE it work 2019-10-28T02:18:03 < Cracki> if something doesn't work, why do you give up? 2019-10-28T02:18:11 < Jan-> the thing is none of them are for the board I have 2019-10-28T02:18:16 < Cracki> why do you not ask WHY what you tried didn't work, and HOW you can make it work? 2019-10-28T02:18:20 < Cracki> so? 2019-10-28T02:18:27 < Cracki> you have docs for the HAL for your chip 2019-10-28T02:18:30 < kakim32> *laughters* 2019-10-28T02:18:34 < Cracki> the precise board is totally irrelevant 2019-10-28T02:18:55 < Cracki> you can't remember where you have examples? srsly? that's keeping you from finding the fucking examples? 2019-10-28T02:19:06 < Cracki> oh bother, it slipped my mind 2019-10-28T02:19:20 < Cracki> <º))))>< 2019-10-28T02:19:38 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-28T02:19:47 < Cracki> kakim32, initial duty cycle? uh... yes the structs get prepopulated from what I entered in cubemx 2019-10-28T02:19:51 < kakim32> Cracki: does config has initial duty cycle? 2019-10-28T02:19:54 < Cracki> that's in some file, I'll look 2019-10-28T02:19:59 < Cracki> yes, it does 2019-10-28T02:20:11 < Cracki> cubemx asks for prescaler, arr/period, duty cycle... 2019-10-28T02:20:15 < Cracki> all in the gui 2019-10-28T02:20:27 < Cracki> and it populates some init function I'll look up just now 2019-10-28T02:20:27 < kakim32> ok so start is when it should start happening 2019-10-28T02:20:44 < kakim32> if initial duty cycle is there and everything 2019-10-28T02:21:07 < Cracki> hm, i have a tim.c next to main.c and that contains all the structure init stuff for all the timers I use 2019-10-28T02:21:17 < Cracki> MX_TIM2_Init: htim2.Init.Period = 3600; 2019-10-28T02:21:25 < Cracki> sConfigOC.Pulse = 0; 2019-10-28T02:21:29 < Cracki> sConfigOC.OCMode = TIM_OCMODE_PWM1; 2019-10-28T02:21:33 < Cracki> and whatever else there is 2019-10-28T02:21:43 < kakim32> what if pulse is higher that period? 2019-10-28T02:21:59 < Cracki> then you fix the hole in the pot 2019-10-28T02:22:27 < Cracki> I would assume it counts until it wraps around and becomes equal to ARR again 2019-10-28T02:22:36 < Jan-> I wouldn't mind bit banging the PWM 2019-10-28T02:22:41 < Jan-> but even for that I would need a working timer 2019-10-28T02:22:44 < kakim32> what if period is 0? 2019-10-28T02:22:46 < Cracki> lol you don't 2019-10-28T02:22:49 < Cracki> you need HAL_Delay() 2019-10-28T02:22:50 < kakim32> like in Jan- cube code? 2019-10-28T02:22:59 < Cracki> kakim32, consider asking the reference manual for that 2019-10-28T02:23:04 < Cracki> or timer cock bock 2019-10-28T02:23:11 < kakim32> sounds bad without looking RM 2019-10-28T02:23:21 < Cracki> that's common behavior for timers 2019-10-28T02:23:34 < kakim32> Jan-: recube 2019-10-28T02:23:37 < Cracki> you can do funny things by jumping beyond ARR, such as single shot 2019-10-28T02:23:54 < kakim32> you need to set period Jan- 2019-10-28T02:24:13 < Cracki> so... what, failure to set interesting paremeters? 2019-10-28T02:24:31 < Cracki> cubemx is there so you CAN'T forget. 2019-10-28T02:24:41 < Jan-> from one example I saw there's about 6 api calls required to change the pwm period 2019-10-28T02:24:45 < Jan-> without the DMA mode 2019-10-28T02:24:48 < Cracki> just go through each parameter, figure out what you need to know to give that a proper value 2019-10-28T02:25:06 < kakim32> period should not be 0 2019-10-28T02:25:16 < kakim32> it's a bit I know 2019-10-28T02:25:22 < Cracki> 0 period is a bad corner case 2019-10-28T02:25:28 < Cracki> it may to something useful but usually not 2019-10-28T02:25:37 < kakim32> how does cube allow this? 2019-10-28T02:25:49 < Cracki> what params did she even set up? which did you set up, Jan- ? 2019-10-28T02:25:58 < Cracki> cube allows it because it may be useful 2019-10-28T02:26:11 < Jan-> That's the problem 2019-10-28T02:26:13 < Cracki> acceptable values are 0 .. 65535 for 16-bit timers 2019-10-28T02:26:27 < Jan-> I have no idea what the parameters are 2019-10-28T02:26:30 < Cracki> ... 2019-10-28T02:26:31 < Cracki> wh 2019-10-28T02:26:31 < Jan-> thta's the whole problem 2019-10-28T02:26:33 < Cracki> why 2019-10-28T02:26:44 -!- X230t [x13@gateway/shell/suchznc/x-ciswnpaftfqcrcdf] has quit [Remote host closed the connection] 2019-10-28T02:26:45 < kakim32> you need to know what cube asks you to fill in 2019-10-28T02:26:48 < Cracki> do you see the table of parameters in cubemx 2019-10-28T02:26:50 < Cracki> for that timer 2019-10-28T02:26:59 < Jan-> there's a bunch of stuff 2019-10-28T02:27:03 < Cracki> lol 2019-10-28T02:27:09 < Jan-> the documentation for most of it just repeats the name of the parameter 2019-10-28T02:27:28 < Cracki> I just need to know... do you use a screen reader, is that the reason you're tapping in the dark? 2019-10-28T02:27:42 < Cracki> because those things can be defeated by bad guis 2019-10-28T02:27:46 < Jan-> I do, when we did all that, I had help. 2019-10-28T02:27:55 < Jan-> because mx is basically unusable 2019-10-28T02:28:09 < Cracki> then what ou absolutely need to do is figure out how to get that screen reader to handle cubemx gui 2019-10-28T02:28:09 < Jan-> I did go through the auto generated stuff to see if any of it made any sense 2019-10-28T02:28:18 < kakim32> for lulz throw some value at period 2019-10-28T02:28:27 < kakim32> you don't need to recube even 2019-10-28T02:28:35 < kakim32> 20000 should do it 2019-10-28T02:29:03 < kakim32> period is like the most basic property of timer 2019-10-28T02:29:24 < kakim32> it's like how many counts it has 2019-10-28T02:29:30 < kakim32> 0 - count to zero 2019-10-28T02:29:58 < kakim32> and ofc. restore everything we might have messed up before this 2019-10-28T02:30:38 < kakim32> cube doesn't make mistakes 2019-10-28T02:30:48 < kakim32> afaik 2019-10-28T02:31:41 < kakim32> it's the perfect glorious machine cube magic codez create whole applications without typing a line of code thing 2019-10-28T02:32:07 < Jan-> so that ended up in htim1.Init.Period = 20000; 2019-10-28T02:32:12 < kakim32> yes 2019-10-28T02:32:20 < kakim32> and pulse should be smaller than that 2019-10-28T02:32:44 < Jan-> pulse is 10000 2019-10-28T02:32:51 < kakim32> it's the count where output state is changed 2019-10-28T02:33:37 < kakim32> from high to low most likelly with your setup 2019-10-28T02:34:29 < kakim32> it cannot change to low if period is 0 it's never reached 2019-10-28T02:35:14 < Jan-> I get that 2019-10-28T02:35:21 < Jan-> so pulse is 10000 and period is 20000 2019-10-28T02:35:32 < kakim32> anything? 2019-10-28T02:35:33 < Jan-> when I hit f8 to resume after the startup breakpoint the pin goes high and that is it 2019-10-28T02:36:13 < Cracki> HAL_TIM_Encoder_MspInit 2019-10-28T02:36:26 < kakim32> you restarted debugging to load new binary? 2019-10-28T02:36:35 < Cracki> must have the right gpios at the right port configured to output 2019-10-28T02:36:41 < Jan-> I hit F11 2019-10-28T02:37:12 < Cracki> nvm, wrong function name, but the rest is still valid 2019-10-28T02:37:28 < Cracki> HAL_TIM_Base_MspInit needs __HAL_RCC_TIM2_CLK_ENABLE, that's it for my setup 2019-10-28T02:37:37 < kakim32> Cracki: we looked at that there is a paste 2019-10-28T02:37:42 < kakim32> up there somewhere 2019-10-28T02:37:45 < Cracki> yes 2019-10-28T02:37:54 < kakim32> was there something wrong? 2019-10-28T02:37:59 < Cracki> HAL_TIM_MspPostInit has the pwm pins set to output in my stuff 2019-10-28T02:38:00 < Cracki> dunno 2019-10-28T02:38:04 < Cracki> just stating things 2019-10-28T02:38:10 < Cracki> as AF_PP 2019-10-28T02:38:13 < kakim32> same thing in that one 2019-10-28T02:38:16 < Jan-> seriously if you guys can't fix it what chance do I have 2019-10-28T02:38:25 < kakim32> shut up 2019-10-28T02:38:26 < Cracki> seriously we're doing keyhole surgery 2019-10-28T02:38:38 < steve> is MspPostInit just a name that CubeMX made up? 2019-10-28T02:38:55 < kakim32> it's called in that MX_init thing 2019-10-28T02:38:57 < kakim32> in main file 2019-10-28T02:39:00 < kakim32> last thing it calls 2019-10-28T02:39:26 < Cracki> for my f103 it also needed __HAL_AFIO_REMAP_TIM2_PARTIAL_2() to map the right pins to the timer 2019-10-28T02:39:42 < steve> right, which is super nasty because it confuses with the calls to MspInit buried in the HAL 2019-10-28T02:39:44 < kakim32> should cube take care of that? 2019-10-28T02:39:54 < Cracki> cube should take care of that 2019-10-28T02:40:11 < Cracki> cube should tell you if the pin's gpio config conflicts with the timer's claim to it 2019-10-28T02:40:28 < Jan-> hang on a second 2019-10-28T02:40:37 < Jan-> this thing is actually working 2019-10-28T02:40:38 < mawk> did you follow the sequence of functions I told Jan- ? 2019-10-28T02:40:39 < mawk> ah 2019-10-28T02:40:41 < mawk> good 2019-10-28T02:40:42 < Jan-> it is toggling the pin 2019-10-28T02:40:45 < Jan-> once every four seconds 2019-10-28T02:40:53 < Cracki> so... when did it start working 2019-10-28T02:40:53 < Jan-> I... 2019-10-28T02:40:56 < Jan-> I can't even. 2019-10-28T02:40:58 < mawk> you messed up ARR and prescaler 2019-10-28T02:41:00 < Jan-> It... the... but... 2019-10-28T02:41:04 < mawk> you put 4Hz instead of 4kHz 2019-10-28T02:41:08 < mawk> not even 4Hz 2019-10-28T02:41:11 < mawk> 0.25Hz 2019-10-28T02:41:14 < Cracki> maybe it's getting a horribly slow clock 2019-10-28T02:41:25 < mawk> yes 2019-10-28T02:41:31 < mawk> please show clock tree tab in MX Jan- 2019-10-28T02:41:33 < mawk> screenshot 2019-10-28T02:41:34 < Cracki> 20000 for a period... I assume 0 prescaler 2019-10-28T02:41:39 < Jan-> does ST think that 0.25 hertz is a reasonable rate for pwm 2019-10-28T02:41:54 < kakim32> it might be 2019-10-28T02:41:55 < Jan-> next to APB1 and APB2 it says 16 2019-10-28T02:41:57 < kakim32> you configured it 2019-10-28T02:42:00 < Cracki> or is the prescaler something other than 0 2019-10-28T02:42:06 < Jan-> both prescalers are divide by one 2019-10-28T02:42:15 < mawk> then it's not to 168MHz Jan- ... 2019-10-28T02:42:27 < kakim32> is it RTC? 2019-10-28T02:42:29 < mawk> I need to know "APB2 timer clock" Jan- 2019-10-28T02:42:30 < Cracki> "does ST think" 2019-10-28T02:42:31 < mawk> no kakim32 it's APB2 2019-10-28T02:42:37 < Cracki> it gives you the freedom 2019-10-28T02:42:39 < kakim32> or some other low speed clock? 2019-10-28T02:42:43 < mawk> you can screenshot and I will tell you Jan- 2019-10-28T02:42:44 < Cracki> and YES that can be a valid frequency for pwm 2019-10-28T02:42:52 < Jan-> X1 16 APB2 Timer clocks (MHz) 2019-10-28T02:43:09 < mawk> ok then it's 16MHz 2019-10-28T02:43:14 < mawk> which prescaler and ARR did you put ? 2019-10-28T02:43:26 < mawk> just put 168MHz in that box and hit entert 2019-10-28T02:43:30 < mawk> the auto clock solver will do the rest 2019-10-28T02:43:31 < Jan-> I mean it's not completely wrong as it is sending "test" down the serial bus every 0.5s which is what I would expect for delay(250) twice 2019-10-28T02:43:38 < Cracki> then prescaler isn't 0 or that statement about APB2 is wrong 2019-10-28T02:43:50 < Jan-> hang on I am getting help this is something I can't do 2019-10-28T02:44:18 < Jan-> er, can I do this by futzing with the SystemClock_Config() 2019-10-28T02:44:21 < mawk> yes Jan- that's why I proposed you screen and I tell you 2019-10-28T02:44:27 < mawk> well no 2019-10-28T02:44:33 < mawk> if you regenerate code afterwards your modifications will be lost 2019-10-28T02:44:43 < Jan-> oh for *&"£^*£&$ 2019-10-28T02:44:46 < Cracki> :D 2019-10-28T02:44:49 < mawk> lol 2019-10-28T02:44:53 < mawk> maybe if you change the .mx file 2019-10-28T02:44:54 < Cracki> those that are outside the "user code" comments 2019-10-28T02:44:59 < mawk> but you can't solve the constraints yourself 2019-10-28T02:45:02 < mawk> that's why there is a solver 2019-10-28T02:45:10 < Jan-> https://www.youtube.com/watch?v=gf_IH3rj0hY 2019-10-28T02:45:11 < Cracki> it maintains stuff you put between 'user code' comment markers 2019-10-28T02:45:42 < kakim32> does cubemx keep you settings if you want to regenerate? 2019-10-28T02:45:52 < kakim32> just change values you want to change? 2019-10-28T02:45:54 < Cracki> 0.25 Hz would be 16 MHz / 20000 (period) / (magic number) 3200 2019-10-28T02:46:10 < Jan-> the delay 250ms does seem to work 2019-10-28T02:46:11 < Cracki> so I'm questioning (!) a few statements 2019-10-28T02:46:17 < Cracki> sure, because that's systick 2019-10-28T02:46:24 < Jan-> assuming that HAL_Delay(250); should be 250ms 2019-10-28T02:46:29 < Cracki> delay uses systick and that's usually set up right because it's unrelated to timers 2019-10-28T02:46:29 < Jan-> not 250 microsecond or something 2019-10-28T02:46:40 < mawk> yes HAL_Delay is always ms 2019-10-28T02:46:40 < Jan-> by the way guys just so you know 2019-10-28T02:46:46 < Jan-> I am just screwing around now 2019-10-28T02:46:51 < Jan-> there is no way this system is in any way usable 2019-10-28T02:46:56 < mawk> why ? 2019-10-28T02:46:56 < Jan-> I'm never going to do anything serious with it 2019-10-28T02:47:05 < mawk> once it's set up you never touch cubeMX again 2019-10-28T02:47:10 < mawk> it's a code *generator* 2019-10-28T02:47:24 < mawk> you can even stop using it right now and just do everything yourself if you had the knowledge to 2019-10-28T02:47:25 < Jan-> guys, seriously, if it takes like four of us an a whole evening to set up one PWM output it's basically unusable 2019-10-28T02:47:29 < kakim32> if you want to play with some setting without affecting others mawk? 2019-10-28T02:47:40 < kakim32> without configuring the whole shiet? 2019-10-28T02:47:46 < kakim32> like clock system? 2019-10-28T02:47:48 < mawk> yes of course kakim32 2019-10-28T02:47:55 < mawk> you just open the mx project and change what you want 2019-10-28T02:47:58 < mawk> it keeps your old code 2019-10-28T02:48:01 < kakim32> Jan-: you have been saying that days 2019-10-28T02:48:26 < Jan-> so what's the question about clocks 2019-10-28T02:48:30 < kakim32> and we got it just doing something and it wasn't right the first time 2019-10-28T02:48:32 < Cracki> we're doing keyhole surgery and you are defeated by a screenreader-unfriendly gui 2019-10-28T02:48:39 < mawk> well that you need to set the clock frequency to something you want, Jan- 2019-10-28T02:48:42 < mawk> and know that value 2019-10-28T02:48:50 < Jan-> which clock there's ten 2019-10-28T02:48:55 < mawk> the issue here is not with HAL, it's that you had the wrong idea of the clock freq 2019-10-28T02:48:55 < kakim32> :D 2019-10-28T02:48:55 < Cracki> the normal way to do this is without screen reader and without half a world between with nothing but IRC 2019-10-28T02:48:59 < Jan-> there's one marked HCLK (MHz) which is set to 16 2019-10-28T02:49:04 < Jan-> we can also change all the APB clocks 2019-10-28T02:49:10 < Jan-> they're also 16 2019-10-28T02:49:18 < mawk> Jan-: set HCLK to 168 and hit enter 2019-10-28T02:49:23 < Cracki> x16, /16, ... 2019-10-28T02:49:25 < mawk> or better, do it like in my screen (hold on) 2019-10-28T02:49:34 < mawk> Jan-: https://pix.watch/XA2lfu/2FMMz0.png 2019-10-28T02:49:51 < Cracki> btw, greenshot has an option to send screenshot directly to imgur, and puts the link into the clipboard. 2019-10-28T02:50:02 < Jan-> by the way I really can't see pictures normally 2019-10-28T02:50:03 < Jan-> just be aware 2019-10-28T02:50:08 < Jan-> I have help right now 2019-10-28T02:50:10 < mawk> yes I know, but you said you had the guy that helps you near you 2019-10-28T02:50:12 < mawk> so I sent it 2019-10-28T02:50:15 < Cracki> do screen readers have OCR these days? 2019-10-28T02:50:22 < Jan-> so where yours says HCLK MHz 168 ours says 16 2019-10-28T02:50:30 < Cracki> bad value 2019-10-28T02:50:31 < mawk> yes you can edit the field and it enter Jan- 2019-10-28T02:50:33 < mawk> MX will auto solve 2019-10-28T02:50:40 < Jan-> Cracki: sort of, it says "image may contain text" then it will optionally try to ocr it 2019-10-28T02:50:43 < kakim32> autosolving :O 2019-10-28T02:50:44 < Jan-> my phone will anyway 2019-10-28T02:50:48 < Jan-> the computer doesn't 2019-10-28T02:50:50 < mawk> for max freq everywhere it should look like my screen 2019-10-28T02:50:56 < Cracki> autosolver may give you a suboptimal solution 2019-10-28T02:51:04 < Cracki> it's only for chasing out conflicts 2019-10-28T02:51:07 < mawk> yes 2019-10-28T02:51:13 < mawk> that's why I ask to set HCLK to maximum 2019-10-28T02:51:21 < mawk> when I tried the autosolver set maximum to everything 2019-10-28T02:51:30 < mawk> but if it doesn't, make it look like my screen by hand 2019-10-28T02:51:39 < Jan-> bear in mind that's only going to make it 10 times faster 2019-10-28T02:51:50 < mawk> that's no big deal 2019-10-28T02:51:54 < Cracki> mawk, that's some funky clock tree. /4 x168... 2019-10-28T02:51:56 < mawk> we have to review your prescaler and ARR values afterwards 2019-10-28T02:52:05 < mawk> yeah it's a bit odd Cracki 2019-10-28T02:52:07 < Cracki> I'd have done /1 x42 2019-10-28T02:52:14 < Cracki> but eh, it works i guess? 2019-10-28T02:52:29 < mawk> well to have HCLK at max, which is 168MHz, this looks like the only solution 2019-10-28T02:52:56 < Jan-> OK so we changed the number in the box 2019-10-28T02:53:01 < mawk> and hit enter ? 2019-10-28T02:53:01 < Jan-> but it didn't want to regenerate code or anything 2019-10-28T02:53:10 < mawk> change the number, hit the enter key 2019-10-28T02:53:15 < mawk> it must look like in my screen 2019-10-28T02:53:17 < Jan-> oh wowo ok 2019-10-28T02:53:26 < Jan-> now it has the apb2 timer clocks set to 168 2019-10-28T02:53:59 < Jan-> ok let's flash that and see wtf we get 2019-10-28T02:54:07 < mawk> that's not the right attitude lol 2019-10-28T02:54:11 < mawk> what did I say about reviewing your values 2019-10-28T02:54:29 < mawk> but alright APB2 timer clock has the max freq now, what is your prescaler ? what is your autoreload value ? 2019-10-28T02:54:35 < Cracki> ST RMs are shit. >The content of the preload register are transferred into the shadow register permanently or at each update event (UEV), depending on the auto-reload preload enable bit (ARPE) in TIMx_CR1 register. 2019-10-28T02:54:44 < Cracki> "permanently" is different from "continuously" 2019-10-28T02:54:49 < Jan-> Cracki I've found better than that 2019-10-28T02:54:52 < Jan-> I think they're translations 2019-10-28T02:54:59 < Jan-> stm32 docs are shit. 2019-10-28T02:55:01 < mawk> from what, chinese ? 2019-10-28T02:55:02 < Jan-> I have literally never had worse. 2019-10-28T02:55:06 < mawk> ST is french and italian 2019-10-28T02:55:10 < Jan-> it shows 2019-10-28T02:55:13 < mawk> aka best countries 2019-10-28T02:55:22 < mawk> but everyone speaks english in the tech domain in france, and I guess italia too 2019-10-28T02:55:24 < Cracki> well, they should employ better technical writers 2019-10-28T02:55:35 < Jan-> ../Startup/startup_stm32f407vetx.s:78: Error: bad instruction `on ldr sp,=_estack' make: *** [Startup/subdir.mk:15: Startup/startup_stm32f407vetx.o] Error 1 00:54:27 Build Failed. 3 errors, 0 warnings. (took 3s.364ms) 2019-10-28T02:55:37 < mawk> when I gave talks at my school it was in english 2019-10-28T02:55:46 < Cracki> ones whose command of english is on par with dutch people ideally 2019-10-28T02:55:54 < Cracki> they employ old farts 2019-10-28T02:55:55 < Jan-> mawk are you italian 2019-10-28T02:55:58 < Cracki> they don't know much english 2019-10-28T02:56:10 < mawk> your build is very fucked up Jan- , try to regenerate in mx ? 2019-10-28T02:56:13 < mawk> no Jan- french 2019-10-28T02:56:26 < kakim32> Jan-: are you saying that you are doing all this.. legally blind? 2019-10-28T02:56:50 < mawk> delete your startup_stm32f407xx.s file 2019-10-28T02:56:54 < mawk> and regenerate 2019-10-28T02:56:57 < Jan-> mawk: where's that 2019-10-28T02:57:05 < mawk> see Jan- , this is the first valid criticism of cubeMX we found ! 2019-10-28T02:57:10 < mawk> sometimes the generated code is all fucked up 2019-10-28T02:57:22 < mawk> in your project files, I don't know where it is on windows, you found them earlier 2019-10-28T02:57:33 < kakim32> startup directory Jan- 2019-10-28T02:57:41 < Jan-> it's windows 2019-10-28T02:57:43 < Jan-> it's not a directory 2019-10-28T02:57:45 < mawk> the .o is in startup directory, but the .s is a level above 2019-10-28T02:57:46 < Jan-> it's a folder 2019-10-28T02:57:48 < mawk> lol 2019-10-28T02:57:53 < Cracki> their description of the ARR suck balls. basically ARR is two things, the "preload" is where you get to write into, and the 'shadow" is what the timer actually respects during operation. 2019-10-28T02:58:05 < mawk> they should call it just "period" not ARR 2019-10-28T02:58:09 < Cracki> every time I have to touch timers, I bitch about it 2019-10-28T02:58:12 < Cracki> aye 2019-10-28T02:58:17 < mawk> but it's really period-1 2019-10-28T02:58:17 < Cracki> period. 2019-10-28T02:58:19 < mawk> so I don't no 2019-10-28T02:58:20 < mawk> know 2019-10-28T02:58:20 < Cracki> :) 2019-10-28T02:58:21 < kakim32> mawk: cubemx is not perfect? 2019-10-28T02:58:32 < mawk> yes kakim32 :( 2019-10-28T02:58:35 < mawk> but they improve over time 2019-10-28T02:58:39 < Cracki> atmegas have the same semantics. put 0 in it, period is 1. that's economical. 2019-10-28T02:58:40 < mawk> great new redesign recently 2019-10-28T02:58:46 < Cracki> put 65535 in it, period is 2**16 2019-10-28T02:58:51 < Jan-> OK I renamed it from .s to .snot 2019-10-28T02:59:02 < mawk> lol 2019-10-28T02:59:13 < mawk> regenerate now 2019-10-28T02:59:25 < mawk> if it doesn't want you to just shake it a bit to fake you changed something 2019-10-28T02:59:28 < mawk> set a pin, unset it 2019-10-28T02:59:30 < mawk> whatever 2019-10-28T02:59:46 * Jan- hits f11 2019-10-28T02:59:53 < mawk> is that reload ? 2019-10-28T02:59:59 < Cracki> that's "start debuggin" 2019-10-28T03:00:02 < mawk> a 2019-10-28T03:00:20 < Cracki> it may or may not work depending on whether you're already having an active debugging session 2019-10-28T03:00:30 < mawk> you still didn't answer Jan- , what is your prescaler ? what is your ARR/period ? 2019-10-28T03:00:30 < kakim32> I'm so pedestrian that I use them buttons 2019-10-28T03:00:40 < Jan-> Error! Failed to read target status Debugger connection lost. Shutting down... 2019-10-28T03:00:52 < Cracki> faulty wiring :P 2019-10-28T03:00:59 < mawk> yeah 2019-10-28T03:01:02 < mawk> target = your board 2019-10-28T03:01:32 < mawk> you lost all your HAL bitchin credit for today Jan-, you didn't want to tell me prescaler and ARR 2019-10-28T03:01:38 < mawk> you can't complain about it if it fails now 2019-10-28T03:01:46 < Jan-> huh? 2019-10-28T03:01:53 < Jan-> it says "download verified successfully" 2019-10-28T03:01:57 < Jan-> then "error failed to read target status" 2019-10-28T03:02:10 < mawk> ah 2019-10-28T03:02:18 < mawk> reboot the board 2019-10-28T03:02:26 < mawk> tell me, did you enable SWD in cubeMX ? 2019-10-28T03:02:42 < mawk> but you won't answer, I don't know why I'm asking 2019-10-28T03:02:50 < Jan-> the answer is I don't know 2019-10-28T03:04:07 < kakim32> why you need to enable swd? 2019-10-28T03:04:13 < mawk> to debug 2019-10-28T03:04:17 < kakim32> it disables it else? 2019-10-28T03:04:18 < mawk> I guess 2019-10-28T03:04:24 < mawk> well the other option is "disable" 2019-10-28T03:04:26 < Jan-> it also says "Break at address "0xb9337822" with no debug information available, or outside of program code." 2019-10-28T03:04:43 < mawk> ah, so it's not "connection lost" anymore ? what did you change ? 2019-10-28T03:04:56 < mawk> well it looks like a plain old programming error, or maybe clock error 2019-10-28T03:05:08 < Jan-> it still says that 2019-10-28T03:05:18 < Jan-> it pops up a window where the code editor would be and says "break at address..." 2019-10-28T03:05:25 < mawk> ah 2019-10-28T03:05:33 < Cracki> cubemx needs to be told to configure SWD, yes 2019-10-28T03:05:40 < mawk> your crystal on the board is 8MHz iirc Jan- right ? 2019-10-28T03:05:46 < Jan-> yes. 2019-10-28T03:05:54 < mawk> you configured the clock tree ***exactly*** as in my screen ? 2019-10-28T03:05:57 < Jan-> did we break the swd by changing the peripheral clocks 2019-10-28T03:06:01 < mawk> or you just changed that one value in the box and hit enter 2019-10-28T03:06:14 < Cracki> you can't break SWD that easily 2019-10-28T03:06:19 < mawk> https://pix.watch/OI3Imn/5dC8u1.png 2019-10-28T03:06:30 < mawk> if you have a boot0 button yeah it's easy to recover 2019-10-28T03:06:34 < mawk> but apparently Jan-'s board has none 2019-10-28T03:06:46 < Cracki> mawk, you got the same/similar f4? 2019-10-28T03:06:55 < mawk> no 2019-10-28T03:06:59 < mawk> olimex something something iirc 2019-10-28T03:07:02 < Cracki> ic 2019-10-28T03:07:09 < mawk> stm32f407vet6 2019-10-28T03:07:33 < Jan-> it won't let us exactly match what you have 2019-10-28T03:07:48 < Cracki> have your assistance post a screenshot of what you DO have 2019-10-28T03:08:01 < mawk> what do you mean it won't let us Jan- ? 2019-10-28T03:08:07 < mawk> if I can do it you can do it too 2019-10-28T03:08:09 < mawk> start by the left 2019-10-28T03:08:20 < Jan-> setting the dropdown next to "pll source mux" to "/4" means SYSCLK (MHz) becomes 336 which is then in red 2019-10-28T03:08:27 < mawk> set HSE to 8MHz, PLL source mux should be HSE, system clock mux should be PLLCLK 2019-10-28T03:08:32 < mawk> yes, no big deal 2019-10-28T03:08:34 < Cracki> sure, but THEN you change some more options 2019-10-28T03:08:49 < mawk> after you did the 3 things I said you can change HCLK to 168 and hit enter 2019-10-28T03:08:55 < mawk> then it will try to solve through the issues by itself 2019-10-28T03:09:06 < Cracki> can't send cars over a bridge you haven't *finished* building 2019-10-28T03:09:37 < Jan-> "debugger connected" 2019-10-28T03:09:42 < mawk> so Jan-: HSE must be 8MHz, PLL source mux must be HSE, system clock mux must be PLLCLK 2019-10-28T03:09:47 < mawk> then you change the values 2019-10-28T03:10:33 < Jan-> it is as you say 2019-10-28T03:10:37 < mawk> good ! very good 2019-10-28T03:10:49 < Jan-> have we basically bricked the board 2019-10-28T03:10:50 < mawk> see, if you don't tell that you couldn't set it exactly as I say I can't guess what's wrong 2019-10-28T03:10:55 < mawk> no, you can't brick it 2019-10-28T03:11:22 < mawk> and I'm sure it has boot0 anyway 2019-10-28T03:11:24 < mawk> it's a jumper or something 2019-10-28T03:11:35 < mawk> it's a counterfeit of this right https://www.olimex.com/Products/ARM/ST/STM32-E407/open-source-hardware 2019-10-28T03:11:40 < BrainDamage> the only way for you to brick it is to physically burn it 2019-10-28T03:11:49 < mawk> or maybe not exactly 2019-10-28T03:11:54 < BrainDamage> or set the read only fuse 2019-10-28T03:12:02 < Jan-> I think I might physically burn it on purpose 2019-10-28T03:12:08 < mawk> and even if you don't have BOOT0 jumper you can connect the NRST line on your stlink and it will work again 2019-10-28T03:12:12 * Jan- rubs her hands together 2019-10-28T03:12:13 * Jan- cackles 2019-10-28T03:12:15 < mawk> :( 2019-10-28T03:12:23 < mawk> I will call the STM32 protection fund 2019-10-28T03:13:01 < kakim32> why is NRST not connected? 2019-10-28T03:13:05 < Jan-> it says debugger connected at first 2019-10-28T03:13:17 < Jan-> then "error failed to read target status" and "debugger connection lost" 2019-10-28T03:13:20 < mawk> it's optional under some circonstances kakim32 2019-10-28T03:13:30 < kakim32> does the board has it? 2019-10-28T03:13:31 < Cracki> it's optional if swd pins remain swd pins 2019-10-28T03:13:47 < mawk> quick summary of how it should look in cubeMX: https://pix.watch/c1Inm3/Pfl6rD.png https://pix.watch/ojVeqj/uJHcgL.png 2019-10-28T03:13:49 < mawk> for SWD 2019-10-28T03:13:50 < Cracki> it needs to pull reset if swd pins aren't configured to be swd pins 2019-10-28T03:14:01 < Cracki> so cubemx must be told to enable swd 2019-10-28T03:14:18 < mawk> if it's a custom board like Jan-'s 2019-10-28T03:14:20 < Cracki> yeah, like in those screenshots 2019-10-28T03:14:27 < mawk> for a nucleo, MX already knows what is SWD and so on 2019-10-28T03:15:04 < Jan-> bear in mind 2019-10-28T03:15:06 < Jan-> this worked an hour ago 2019-10-28T03:15:10 < Jan-> before we started futzing with clocks 2019-10-28T03:15:46 < mawk> before we fixed the clocks you mean 2019-10-28T03:15:52 < mawk> can you screenshot your clock tree ? 2019-10-28T03:16:02 < Jan-> I have to sleep soon 2019-10-28T03:16:04 < mawk> and be sure that your cubemx is like my two screens 2019-10-28T03:16:04 < mawk> ah 2019-10-28T03:16:07 < Jan-> I have work tomorrow 2019-10-28T03:16:09 < Jan-> well LATER TODAY 2019-10-28T03:16:11 < mawk> https://pix.watch/c1Inm3/Pfl6rD.png https://pix.watch/ojVeqj/uJHcgL.png 2019-10-28T03:16:12 < mawk> lol 2019-10-28T03:16:17 < mawk> it's only 2am don't worry 2019-10-28T03:16:20 < mawk> sleep is facultative 2019-10-28T03:16:26 < mawk> take a good old rail of coke 2019-10-28T03:17:18 < Cracki> there must be high capacity braille pads today 2019-10-28T03:17:49 < Jan-> https://pasteboard.co/IDZrLaS.png 2019-10-28T03:17:55 < Jan-> for that you would need to be able to read braille 2019-10-28T03:18:15 < Cracki> screenshot looks good. 2019-10-28T03:18:31 < Cracki> do you read braille? 2019-10-28T03:18:37 < Jan-> no. 2019-10-28T03:18:40 < Cracki> hm 2019-10-28T03:18:56 < Cracki> so you have everything read to you? or is there some eyesight you can make use of? 2019-10-28T03:19:11 < Cracki> I know "legally blind" can mean "too bad to drive" 2019-10-28T03:19:17 < Jan-> freedomscientific.com/jaws 2019-10-28T03:19:27 < Cracki> muh freedom 2019-10-28T03:19:48 < Jan-> don't knock them they've been called that for decades 2019-10-28T03:19:55 < mawk> good Jan- , your clocks are good 2019-10-28T03:20:05 < Jan-> can't be that great they don't fucking work. 2019-10-28T03:20:35 < Cracki> worth checking if the crystal is actually 8 MHz? 2019-10-28T03:20:49 < mawk> the crystal on the board is 8MHz, we verified a week or so ago iirc 2019-10-28T03:20:56 < mawk> if I didn't get lied to 2019-10-28T03:21:02 < mawk> but yes it's worth a check 2019-10-28T03:21:24 < Jan-> if it wasn't, delay would be wrong 2019-10-28T03:22:21 < mawk> yes RCC.HSE_VALUE=8000000 2019-10-28T03:22:23 < mawk> on https://github.com/abelykh0/stm32f407-vga/blob/master/stm32f407-vga.ioc 2019-10-28T03:22:41 < mawk> same board "black STM32F407VET6" 2019-10-28T03:23:58 < Jan-> I've got nothing to try 2019-10-28T03:24:03 < mawk> and there is a boot0 jumper 2019-10-28T03:24:17 < mawk> try to set BOOT0 to 1 and flash 2019-10-28T03:24:25 < mawk> but make sure you enabled SWD 2019-10-28T03:24:31 < mawk> otherwise you have to do it everytime 2019-10-28T03:24:43 < Jan-> I have no idea how we would enable or disable it 2019-10-28T03:24:47 < Jan-> we have never done anything with it 2019-10-28T03:24:55 < mawk> so hmm with the board with the USB port up, on the right of the board on the array of pins there are two jumpers 2019-10-28T03:25:08 < mawk> a left one, and a right one 2019-10-28T03:25:14 < Jan-> BT0 and BT1? 2019-10-28T03:25:15 < mawk> yes 2019-10-28T03:25:28 < mawk> there is a jumper between BT0 and GND 2019-10-28T03:25:35 < Jan-> no there isn't 2019-10-28T03:25:41 < mawk> what 2019-10-28T03:25:46 < mawk> did someone remove them ? 2019-10-28T03:25:49 < mawk> https://raw.githubusercontent.com/mcauser/BLACK_F407VE/master/docs/STM32F407VET6.jpg 2019-10-28T03:25:51 < mawk> that's your board right 2019-10-28T03:25:51 < Jan-> there's no jumpers anywhere on the entire board 2019-10-28T03:25:55 < mawk> wtf 2019-10-28T03:26:02 < mawk> maybe someone thought they were useless and removed them ? 2019-10-28T03:26:04 < mawk> they are essential 2019-10-28T03:26:17 < mawk> so essentially, BOOT0 is floating on your board ? 2019-10-28T03:26:27 < mawk> and it is randomly getting into ST bootloader mode or not 2019-10-28T03:26:29 < Jan-> so both BT0 and BT1 should be grounded? 2019-10-28T03:26:35 < mawk> yes for regular operation 2019-10-28T03:26:41 < mawk> and for bootloaer mode, BT0 should be tied to 3.3V 2019-10-28T03:27:01 < Cracki> bootloader mode enables SWD. that's all you need it for. 2019-10-28T03:27:29 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-28T03:27:34 < Jan-> ok, grounded bt0 and bt1 2019-10-28T03:27:40 < mawk> good 2019-10-28T03:27:42 < mawk> so now it's in regular mode 2019-10-28T03:28:00 < Jan-> I don't think that's actually the problem anyway as it worked before 2019-10-28T03:28:06 < mawk> it was floating before 2019-10-28T03:28:07 < Jan-> and no it did the same thing again 2019-10-28T03:28:11 < Jan-> debugger connection lost shutting down 2019-10-28T03:28:12 < mawk> yes 2019-10-28T03:28:16 < mawk> that's expected 2019-10-28T03:28:21 < mawk> so now you need to tell cubeMX to enable SWD 2019-10-28T03:28:43 < mawk> in categories you have SYS under system core, and in the Debug box you tell Serial wire 2019-10-28T03:28:51 < mawk> you generate, compile, and so on 2019-10-28T03:29:00 < mawk> then instead of grounding BOOT0 you set it to 3.3V 2019-10-28T03:29:05 < mawk> it will force the boad in SWD 2019-10-28T03:29:37 < Jan-> debug is set to disable under sys 2019-10-28T03:29:40 < mawk> ah ! 2019-10-28T03:29:44 < mawk> so, set it to serial wire 2019-10-28T03:29:46 < mawk> otherwise you can't debug 2019-10-28T03:30:02 < Jan-> I don't really want to use the debugger I just want the chip to work 2019-10-28T03:30:08 < mawk> ......... 2019-10-28T03:30:13 < mawk> are you serious ? 2019-10-28T03:30:16 < Cracki> lol UMich works on a braille display. they claim a single line, today on the market, costs 3-5 k USD 2019-10-28T03:30:20 < mawk> you want the debugger to make the chip work, Jan- 2019-10-28T03:30:20 < Jan-> same result 2019-10-28T03:30:24 < Jan-> debugger connection lost shutting down 2019-10-28T03:30:26 < mawk> yes, it's not over 2019-10-28T03:30:32 < mawk> have a little bit of patience, please 2019-10-28T03:30:37 < mawk> you need to set BOOT0 to 3.3V 2019-10-28T03:30:48 < Cracki> jumper it to 3.3V 2019-10-28T03:30:54 < mawk> Cracki: is the cubeIDE really flashing ? 2019-10-28T03:31:00 < mawk> or it is just live loading the code or whatever 2019-10-28T03:31:06 < Cracki> uh? 2019-10-28T03:31:23 < Cracki> yes it's writing to flash and then running target 2019-10-28T03:31:31 < Cracki> not sure what live loading means 2019-10-28T03:31:42 < mawk> like loading the code only in RAM 2019-10-28T03:31:45 < Cracki> it verifies after programming. 2019-10-28T03:31:48 < Cracki> nah, no ram fuckery 2019-10-28T03:31:52 < mawk> and messing with the 0x0 base pointer to fake it being in flash 2019-10-28T03:31:52 < mawk> ah 2019-10-28T03:31:53 < mawk> good 2019-10-28T03:32:05 < mawk> since the option is called "run" or "debug" but not "flash" I was never sure 2019-10-28T03:32:21 < Cracki> blame it on them being french/italian 2019-10-28T03:32:24 < mawk> lol 2019-10-28T03:32:26 < Cracki> and on eclipse being eclipse 2019-10-28T03:32:32 < Jan-> now it stops at "download verified successfully" 2019-10-28T03:32:43 < mawk> good 2019-10-28T03:32:47 < Jan-> and pops up a text tab with "Break at address "0x1fff3da0" with no debug information available, or outside of program code." 2019-10-28T03:32:48 < Cracki> the "flash to target" is something you make a "tool" for that calls openocd 2019-10-28T03:33:09 < Cracki> the builtin "debug" function programs flash and then runs target under debugging control (again openocd) 2019-10-28T03:33:25 < Cracki> that's some weird behavior 2019-10-28T03:33:29 < mawk> now normally you could set back the BOOT0 jumper to GND, unplug the board, plug it back in, and it's back to normal 2019-10-28T03:33:35 < Cracki> it shouldn't jump to anywhere outside of your program code 2019-10-28T03:33:48 < mawk> but if your instruction pointer goes outside your code you have a bug somewhere 2019-10-28T03:33:56 < mawk> maybe it's flash wait state business ? I don't know very well about that 2019-10-28T03:34:01 < mawk> since now it's 168MHz 2019-10-28T03:34:12 < mawk> maybe someone here knows 2019-10-28T03:34:22 < Jan-> I think it's pretty clear we screwed it 2019-10-28T03:34:28 < mawk> https://pix.watch/sIS9Oh/h_WrIv.png these settings 2019-10-28T03:34:31 < Jan-> thanks guys that was really helpful 2019-10-28T03:34:35 < mawk> you had the clock to a nonsense value, Jan- 2019-10-28T03:34:37 < Jan-> woo change the clocks brick the board 2019-10-28T03:34:41 < mawk> omg 2019-10-28T03:34:52 < mawk> just go back to avr already with that attitude 2019-10-28T03:35:00 < mawk> you're very lucky we haven't walked away yet 2019-10-28T03:35:01 < Ultrasauce> tfw timecube destroys your hardware 2019-10-28T03:35:18 < kakim32> it's just a lockup 2019-10-28T03:35:35 < jadew> chinese PCB pricing is ridiculous... 2019-10-28T03:35:36 < mawk> we have no idea which code you wrote, you could have written the most horrible nonsense code and we don't know 2019-10-28T03:35:41 < mawk> and you blame it on us 2019-10-28T03:36:00 < Jan-> well I don't know what to tell you 2019-10-28T03:36:18 < mawk> you could show the code for instance 2019-10-28T03:36:20 < mawk> for once 2019-10-28T03:36:31 < mawk> just zip up the whole directory and host it somewhere 2019-10-28T03:36:33 < Jan-> https://pastebin.com/D1hHRrY0 2019-10-28T03:36:35 < mawk> so we could see your cubeMX settings and so on 2019-10-28T03:37:26 < Jan-> there is a check box in mx called "system wake up" 2019-10-28T03:37:35 < Jan-> that sounds like a good check box but I have no idea what it is really for 2019-10-28T03:37:52 < mawk> that will enable a pin you can use to wake up the chip when it's sleeping 2019-10-28T03:37:57 < Jan-> oh ok. 2019-10-28T03:38:29 < mawk> let me see what to do when the clock is at max frequency, if there is something to do with the flash 2019-10-28T03:38:35 < mawk> like disable instruction prefetch or whatever 2019-10-28T03:39:57 < Jan-> again 2019-10-28T03:39:59 < Jan-> I have to ask 2019-10-28T03:40:05 < Jan-> how on earth could I possibly have known any of this 2019-10-28T03:40:30 < kakim32> you couldn't 2019-10-28T03:40:43 < kakim32> it comes with experience 2019-10-28T03:40:46 < Jan-> if it's confusing you guys what hope do I have 2019-10-28T03:40:48 < kakim32> more than days 2019-10-28T03:40:58 < kakim32> more than one hello world 2019-10-28T03:41:10 < Ultrasauce> if only there were some sort of document, you could reference, 2019-10-28T03:41:20 < Ultrasauce> perhaps even a manual 2019-10-28T03:41:25 < Ultrasauce> it might help you understand the hardware 2019-10-28T03:41:28 < Jan-> that's the problem 2019-10-28T03:41:34 < Jan-> there really isn't 2019-10-28T03:41:37 < kakim32> but you don't just read it first Ultrasauce? 2019-10-28T03:41:39 < Jan-> there'a reference manual 2019-10-28T03:41:52 < Jan-> that tells you about the low level stuff 2019-10-28T03:42:10 < Jan-> but as far as I can tell there is no manual for any of this stuff in mx 2019-10-28T03:42:11 < kakim32> Jan-: now you know what happens when your chip locks you out 2019-10-28T03:42:20 < kakim32> you know the symptoms 2019-10-28T03:42:36 < Jan-> I do have another board if that will help. 2019-10-28T03:42:47 < Ultrasauce> isnt this issue just the flash under reset thing 2019-10-28T03:42:47 < mawk> by reading the manual, Jan- 2019-10-28T03:42:56 < mawk> that's why I told you like 10 times to read the manual 2019-10-28T03:43:03 < mawk> not read it all, but the peripherals you're using, at least the intro 2019-10-28T03:43:41 < mawk> 5 WS (6 CPU cycles) : 150 MHz < HCLK ≤ 168 MHz 2019-10-28T03:43:51 < mawk> that's already what's written in cubeMX, that sounds alright...? 2019-10-28T03:44:08 < Jan-> I have no idea what you're talking about. 2019-10-28T03:44:18 < Jan-> we have a tab called "pinout and configuration" 2019-10-28T03:44:23 < Jan-> one called "clock configuration" 2019-10-28T03:44:36 < Jan-> one called "project manager" and one called "tools" in the mx section of cube ide. 2019-10-28T03:47:17 < mawk> maybe the crystal on the board is shitty 2019-10-28T03:47:23 < mawk> since you have a counterfeit chinese board 2019-10-28T03:47:43 < Cracki> counterfeit, hah. they're all chinese designed and produced. 2019-10-28T03:48:09 < mawk> you select HSI in PLL source mux, and the /M value of the PLL source mux should become /8 2019-10-28T03:48:16 < mawk> then you're using the internal RC oscillator 2019-10-28T03:48:17 < Cracki> it may be worth trying a lower frequency, if components really are a little bad 2019-10-28T03:48:21 < mawk> instead of the shady chinese thing 2019-10-28T03:48:32 < mawk> yeah and maybe lower freq if that doesn't work 2019-10-28T03:49:44 < Jan-> I have to sleep anyway 2019-10-28T03:49:55 < Jan-> it doesn't really matter 2019-10-28T03:50:20 < Jan-> It's disappointing that we've basically gone backwards but in the end like I say 2019-10-28T03:50:32 < Jan-> there is no way I would put much more time into this thing 2019-10-28T03:50:44 < mawk> you're not very optimist 2019-10-28T03:50:53 < mawk> I had basically less knowledge than you when starting stm32 2019-10-28T03:51:04 < Jan-> I think I've pushed on a really long way with this thing 2019-10-28T03:51:09 < Jan-> I've been trying for hours 2019-10-28T03:51:12 < mawk> and I just read source code, reference manual, had regressions like you, but just tried again and again 2019-10-28T03:51:15 < Jan-> with all of your help 2019-10-28T03:51:30 < mawk> yeah, and maybe thousands are using STM32, it works, that's all 2019-10-28T03:51:37 < mawk> you can't deduce it doesn't work from your experience alone 2019-10-28T03:51:46 < mawk> successful commercial products are using it 2019-10-28T03:52:12 < Jan-> that doesn't really help me, though, does it. 2019-10-28T03:53:28 < mawk> this should help you quit this defeatist attitude 2019-10-28T03:54:08 < Jan-> well look at this point I am completely helpless, I don't even have a board I can program 2019-10-28T03:54:18 < Jan-> so I think we can conclude we're basically screwed at this point 2019-10-28T03:54:20 < mawk> the issue is with the code 2019-10-28T03:54:23 < mawk> not with the board 2019-10-28T03:54:34 < mawk> with the BOOT0 trick we showed you, you can program any board even when it doesn't want to 2019-10-28T03:54:49 < steve> FreeRTOS question: In my SysTick_Handler() I'll call xPortSysTickHandler() , which is defined in port.c , but is not declared in any header. In the demo project I have, it declares xPortSysTickHandler() in the ***_it.c file. Is that how you do it? 2019-10-28T03:55:36 < mawk> if the demo goes it like that I guess yeah 2019-10-28T03:56:04 < steve> lol I go by the opposite, the demo does everything wrong 2019-10-28T03:56:25 < Jan-> you keep saying this but I mean... here we are... 2019-10-28T03:56:25 < steve> but in this case I'm expecting the delcaration in a FreeRTOS header, and there isn't one. 2019-10-28T03:57:01 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has joined ##stm32 2019-10-28T03:57:33 < kakim32> steve is it just #define? 2019-10-28T03:57:37 < kakim32> not a call 2019-10-28T03:58:09 < steve> kakim32 my ***_it.c says extern void xPortSysTickHandler(void); , and then calls it in SysTick_Handler() 2019-10-28T03:59:12 < mawk> well can you zip up the whole project so I can see it at least Jan- ? and you can go to bed we will see it later 2019-10-28T03:59:21 < mawk> the whole stm32cubeIDE project 2019-10-28T03:59:25 < mawk> with the .ioc file and everything 2019-10-28T03:59:28 < steve> I also noticed this, which is not part of my project, but makes zero sense: FreeRTOSConfig_template.h:#define xPortSysTickHandler SysTick_Handler 2019-10-28T03:59:35 < Jan-> it's not worth it 2019-10-28T03:59:37 < Jan-> I'll find something else. 2019-10-28T03:59:41 < Jan-> I'm sorry to have wasted your time. 2019-10-28T03:59:45 < Jan-> I'm going to sleep now. 2019-10-28T03:59:52 < steve> : ( 2019-10-28T04:03:40 < mawk> lol 2019-10-28T04:03:50 < mawk> embedded development is not easy, Jan- 2019-10-28T04:03:57 < mawk> and stm32 is relatively easy 2019-10-28T04:04:24 < mawk> we wouldn't be relatively well paid if this job was doable by tetraplegic monkeys or AIs 2019-10-28T04:04:49 < steve> aw don't rub it in. That's like saying javascript is easy when someone can't figure out react in a day 2019-10-28T04:05:09 < mawk> lol 2019-10-28T04:05:23 < mawk> I meant stm32 is relatively easy compared to the rest 2019-10-28T04:05:34 < mawk> so no reason to abandon it 2019-10-28T04:07:23 < steve> has someone made an AI coder yet 2019-10-28T04:15:46 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-28T04:16:23 < kakim32> steve: it makes all the sense 2019-10-28T04:16:29 < kakim32> it's renaming 2019-10-28T04:16:49 < kakim32> you see xPortSysTickHandler 2019-10-28T04:17:19 < kakim32> but guess what you see in map file 2019-10-28T04:17:40 < kakim32> you see SysTick_Handler 2019-10-28T04:17:56 < kakim32> no xPortSysTickHandler 2019-10-28T04:18:29 < kakim32> or actually just after preprocessor it has changed it's name 2019-10-28T04:19:20 < mawk> macro maybe 2019-10-28T04:19:45 < kakim32> steve: don't worry about it 2019-10-28T04:24:11 < kakim32> they probably did it because else it would break freertos naming system 2019-10-28T04:24:57 -!- catphish [~catphish@unaffiliated/catphish] has quit [Quit: Leaving] 2019-10-28T04:25:06 < kakim32> xPortSysTickHandler should not be called anywhere 2019-10-28T04:25:09 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-28T04:26:01 -!- catphish [~catphish@unaffiliated/catphish] has quit [Client Quit] 2019-10-28T04:27:30 < kakim32> same goes for all freertos interrupt handlers 2019-10-28T04:27:45 < kakim32> all 3 of them or so 2019-10-28T04:28:15 < kakim32> those are just renamed by template according to system naming 2019-10-28T04:30:15 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 250 seconds] 2019-10-28T04:30:42 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-28T04:41:08 -!- rajkosto [~Rajko@cable-178-149-117-3.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-28T04:43:18 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-28T04:44:28 < Cracki> when catphish shows up, show him this: https://pastebin.com/Qq3xAuWj http://cowlark.com/fluxengine/index.html 2019-10-28T04:55:23 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 276 seconds] 2019-10-28T05:07:16 -!- rue_shop1 [~rue_mohr@d50-92-152-244.bchsia.telus.net] has left ##stm32 ["Leaving"] 2019-10-28T05:16:38 < Cracki> also https://hackaday.com/2019/02/19/flux-engine-reads-floppies/ https://hackaday.com/2019/01/08/preserving-floppy-disks-via-logic-analyser/ 2019-10-28T05:37:26 < dongs> lol @ half the comments tellhim him to use some other shitty MCU 2019-10-28T05:37:27 < dongs> including PIC32 2019-10-28T05:40:23 < aandrew> heh 2019-10-28T05:50:12 < Cracki> pic! srsly 2019-10-28T05:50:48 < Cracki> psoc5 is only a CM3 at 80 mhz but that's about the same catphish is using 2019-10-28T05:51:56 < Cracki> it's all digital so psoc particular stuff is wasted on that 2019-10-28T06:11:54 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-28T06:11:55 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-28T06:29:23 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-28T06:29:24 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-28T06:29:24 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-28T06:42:25 -!- rue_mohr [~rue_mohr@d50-92-152-244.bchsia.telus.net] has left ##stm32 ["Leaving"] 2019-10-28T06:45:26 < dongs> i didnt read too much into "how it works" but it sounds like hes just using a timer with input capture or someshit? 2019-10-28T06:50:07 -!- fc5dc9d4_ [~quassel@p5B081248.dip0.t-ipconnect.de] has joined ##stm32 2019-10-28T06:54:08 -!- fc5dc9d4 [~quassel@p5B3A8038.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-28T07:07:38 < Cracki> exactly 2019-10-28T07:07:48 < Cracki> everything else is hostside processing 2019-10-28T07:08:13 < Cracki> the other hackaday article shows someone do that with a logic analyzer and an usb uart for the most basic control signals to the floppy drive 2019-10-28T07:08:53 < Cracki> he dumps the pulse times (via python script) into an apparently quasi-standard dump file format, then uses third party software to decode that 2019-10-28T07:09:30 < Cracki> catphish's ambition is to do all that deviceside and possibly present a mass storage device to the host 2019-10-28T07:09:50 < Cracki> (but also the raw pulse time dump) 2019-10-28T07:14:21 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-28T07:17:17 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 240 seconds] 2019-10-28T07:17:17 -!- day__ is now known as day 2019-10-28T07:24:00 < dongs> jesus christ 2019-10-28T07:24:08 < dongs> i just chcked the size of STM32 cube folder in my VM 2019-10-28T07:24:25 < dongs> 18.7gb 2019-10-28T07:24:30 < dongs> what teh actual fuck 2019-10-28T07:24:46 < dongs> 18.7 GB (20,163,469,312 bytes) 266,373 Files, 72,742 Folders 2019-10-28T07:25:02 < dongs> i was wondering why my shit was almost out of space 2019-10-28T07:28:51 < antto> dongs tiem for DEATHME... i mean baremetal 2019-10-28T07:40:32 < dongs> found how to free ~6gb 2019-10-28T07:40:41 < dongs> apparently ST released F7 package version 1.13 2019-10-28T07:40:46 < dongs> and then removed it from thier versioning XMLs 2019-10-28T07:40:51 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-28T07:41:00 < dongs> so you have 1.12 1.14 1.15 and .13 is missing 2019-10-28T07:41:05 < dongs> but my shit downloaded and unpacked it 2019-10-28T07:56:51 < Cracki> heh 2019-10-28T07:57:18 < Cracki> they should split those packages into basic support and all those examples they don't trim down 2019-10-28T07:59:06 < Cracki> they could also go easy on the pdf downloads. 2019-10-28T07:59:20 < Cracki> I have an 84 MB RM for STM32MP157 2019-10-28T07:59:28 < Cracki> and I never asked for that 2019-10-28T08:06:45 < Thorn> is there HAL for stm32mp1 btw or is it all in lunix drivers? 2019-10-28T08:07:28 < Cracki> i see an ad for "openstlinux" for mp1... 2019-10-28T08:07:29 < Thorn> are cortex-m peripherals separate from what is accessible from cortex-a btw? 2019-10-28T08:07:32 < dongs> there's a download, im not touching it 2019-10-28T08:08:02 < Cracki> bus schematic? 2019-10-28T08:08:46 < Cracki> they must have a separate memory map for the cortex M and cortex A 2019-10-28T08:08:54 < Cracki> would be nice if they matched 2019-10-28T08:18:36 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-28T08:23:15 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-28T08:31:25 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-28T08:36:51 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-28T08:37:19 -!- jly [uid355225@gateway/web/irccloud.com/x-lngnuxacuyydzqcn] has joined ##stm32 2019-10-28T08:43:33 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-28T09:12:21 < qyx> Please note the long lead time reported on this product. 2019-10-28T09:12:35 < qyx> Factory Lead Time: 77 Weeks 2019-10-28T09:14:09 < Thorn> kickstarter? 2019-10-28T09:14:35 < qyx> no, TE axicom relays 2019-10-28T09:15:21 < jly> hmmm 2019-10-28T09:15:23 < Steffanx> Such relay. 2019-10-28T09:15:24 < jly> what are they 2019-10-28T09:15:29 < jly> are they generic? 2019-10-28T09:16:07 < jly> at a glance look like takamisawa clones 2019-10-28T09:16:20 < jly> as in a google hit 2019-10-28T09:35:00 -!- c10ud [~c10ud@emesene/dictator/c10ud] has joined ##stm32 2019-10-28T09:43:20 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-28T10:08:18 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-28T10:15:50 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-28T10:45:11 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-28T10:52:57 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-sehzcmqtjptozkgq] has quit [Ping timeout: 240 seconds] 2019-10-28T10:56:36 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-28T11:04:13 -!- Tordek [~tordek@gateway/shell/blinkenshell.org/x-ekeoandwqzhctheg] has joined ##stm32 2019-10-28T11:08:33 -!- Tordek [~tordek@gateway/shell/blinkenshell.org/x-ekeoandwqzhctheg] has quit [Ping timeout: 246 seconds] 2019-10-28T11:13:51 -!- hackkitten [~hackkitte@2a02:6d40:30e6:ef01:d9ec:62e3:b85f:c28] has joined ##stm32 2019-10-28T11:14:25 -!- Tordek [tordek@gateway/shell/blinkenshell.org/x-ymvbirouiwbupsct] has joined ##stm32 2019-10-28T11:37:24 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has quit [Ping timeout: 268 seconds] 2019-10-28T11:43:11 -!- fenugrec [~fenugrec@142.167.170.182] has joined ##stm32 2019-10-28T11:51:08 -!- fenugrec [~fenugrec@142.167.170.182] has quit [Ping timeout: 245 seconds] 2019-10-28T12:10:51 < karlp> steve: you don't need to handle systick irqs yourselef unless you comment out that define, and then implement it yourself. out of the box freertos has both systick setup and a systick irq handler. 2019-10-28T12:22:06 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-28T12:30:43 < Haohmaru> dee ache el eggspress 2019-10-28T12:33:49 < Haohmaru> Your parcel status: Package is being tossed around in china 2019-10-28T12:53:31 < jadew> https://www.ebay.co.uk/itm/1x-K9GAG08U0E-SAMSUNG-soft-NAND-UE32-UE37-UE40-UE46-D5500-D5700-Nice/113855789655 2019-10-28T12:53:46 < jadew> does that listing seem to imply it has firmware on it? 2019-10-28T12:54:21 < jadew> or do you think the seller is trying to sort of implying it, but without it being the case? 2019-10-28T13:13:34 < Haohmaru> ask sell0r 2019-10-28T13:13:50 < jadew> I asked, he didn't answer 2019-10-28T13:14:03 < Haohmaru> FU him then 2019-10-28T13:14:11 < jadew> well, I already bought that 2019-10-28T13:14:19 < jadew> and have it on my desk 2019-10-28T13:14:21 < Haohmaru> 2019-10-28T13:14:58 < jadew> so the question is: do I solder it back or buy from someone else, or buy a programmer? 2019-10-28T13:15:05 < jadew> or make one 2019-10-28T13:15:14 < Haohmaru> ah, decisions.. 2019-10-28T13:15:25 < Haohmaru> toss a multisided coin 2019-10-28T13:15:39 < Haohmaru> or a bitcoin 2019-10-28T13:16:07 < jadew> I'm annoyed that my mini-pro can't program it 2019-10-28T13:16:14 < jadew> I don't understand why 2019-10-28T13:16:48 < jadew> the thing is bitbanging everything else, why can't it bitbang this thing? 2019-10-28T13:17:26 < Haohmaru> maybe it got tired of bitbanging 2019-10-28T13:17:50 < Haohmaru> jadew try soaking it in RedBull 2019-10-28T13:18:34 * Haohmaru is full of solutions 2019-10-28T13:19:20 < Haohmaru> maybe i should start doing "lifehacks" videos 2019-10-28T13:19:40 < Haohmaru> and become rich as fugg 2019-10-28T13:21:11 < Haohmaru> jadew send me that samsung thing and the mini-pro thing that doesn't wanna program it, iz gonna b muh first video 2019-10-28T13:21:17 < Haohmaru> >:) 2019-10-28T13:21:23 < karlp> what'sa "mini-pro" ? 2019-10-28T13:21:37 < Haohmaru> some programmer, i guess 2019-10-28T13:21:37 < jadew> karlp, tl866ii plus programmer 2019-10-28T13:21:57 < jadew> http://www.autoelectric.cn/en/tl866_main.html 2019-10-28T13:24:00 < jadew> oh, I don't have the II+, I have the A version 2019-10-28T13:24:38 < karlp> well there you go then :) 2019-10-28T13:25:52 < jadew> wouldn't work with that one either 2019-10-28T13:26:26 < qyx> important tips on Piracy:（Click here See the pirated Picture） 2019-10-28T13:26:27 < qyx> heh 2019-10-28T13:26:32 < karlp> NAND FLASH: yes, no, no? 2019-10-28T13:26:35 < karlp> looks like it would? 2019-10-28T13:26:53 < karlp> if you could get a footprint adapter... 2019-10-28T13:27:00 <@englishman> maybe it doesn't work because you bought flash from a random chinese store selling "fuck you" lapel pins 2019-10-28T13:27:12 <@englishman> he was probably selling them as decorative items only 2019-10-28T13:27:50 < jadew> englishman, I didn't try it on that chip, but on the original one 2019-10-28T13:27:57 < jadew> (that came out of my TV) 2019-10-28T13:28:25 < jadew> karlp, I checked the support list and it's not in the list 2019-10-28T13:28:30 < jadew> I have the adapter... 2019-10-28T13:38:52 < Haohmaru> jadew that page has winzip icon 2019-10-28T13:50:03 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-28T13:53:32 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-28T14:15:03 -!- kakim32 [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-28T14:21:26 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has quit [Ping timeout: 268 seconds] 2019-10-28T14:27:05 -!- jly [uid355225@gateway/web/irccloud.com/x-lngnuxacuyydzqcn] has quit [Quit: Connection closed for inactivity] 2019-10-28T14:38:28 < karlp> well, this has been hilarious https://github.com/karlp/ignore-warnings-peril 2019-10-28T15:05:49 -!- tctw [~Tectu@82-197-160-105.init7.net] has joined ##stm32 2019-10-28T15:23:06 -!- tctw [~Tectu@82-197-160-105.init7.net] has quit [Quit: Leaving] 2019-10-28T15:37:38 < PaulFertser> karlp: do you mean you had tought time debugging an issue in someone else's code due to that? 2019-10-28T15:38:58 < karlp> no, my own. was moving some shit to a new project, hadn't cleaned up all the headers yet. 2019-10-28T15:39:16 < karlp> worst was that breaking into the debugger and doign "call get_float()" wsa giving the right answer. 2019-10-28T15:39:33 < karlp> (it doesn't in that project, because the ordering all matters) 2019-10-28T15:40:37 < karlp> with certain orderings, you'll actually get an error about conflicting definitions.... 2019-10-28T15:41:42 < karlp> but do things "right" and it just .... fixes it. In my case, the float gets returned by the function, the caller was built with a default proto that assumed an int for the return, so it casts it again, and you get garbage out. 2019-10-28T15:42:02 < karlp> cleaned up my warnings, code magically starts working properly :) 2019-10-28T15:43:05 < PaulFertser> :) 2019-10-28T15:43:56 < PaulFertser> I thought you're building with -Werror ;) Well, missing prototypes is not a warning someone should ignore, indeed. 2019-10-28T15:51:54 < karlp> I though it would only matter if I had multiple options to link with, or wanted to link later 2019-10-28T15:52:04 < karlp> I wsa under the beliefe that the linker would resolve it all. 2019-10-28T15:52:06 < karlp> that was wrong. 2019-10-28T16:00:07 -!- tprrt [~tprrt@217.114.204.178] has quit [Ping timeout: 250 seconds] 2019-10-28T16:19:06 -!- brdb [~basdb@2601:18c:8500:7f5b::9bb] has joined ##stm32 2019-10-28T16:35:49 -!- steve [~steve@ool-182f8dfd.dyn.optonline.net] has joined ##stm32 2019-10-28T16:46:54 < steve> thanks karlp 2019-10-28T16:47:42 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-28T16:55:12 -!- emeb1 [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-28T16:55:16 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Ping timeout: 264 seconds] 2019-10-28T16:56:58 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-28T17:02:57 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Ping timeout: 240 seconds] 2019-10-28T17:04:11 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-28T17:05:42 < bitmask> https://imgur.com/a/XecqThr 2019-10-28T17:05:45 < bitmask> mmm pretty, 2019-10-28T17:05:52 < bitmask> so much better than my first attempt 2019-10-28T17:05:58 < Haohmaru> y u no direct link to image 2019-10-28T17:06:08 < Haohmaru> oh, multiple 2019-10-28T17:06:20 < Haohmaru> y u no combine into one >:/ 2019-10-28T17:07:01 < bitmask> thats extra work 2019-10-28T17:07:43 < Haohmaru> as an MSPainter - ur fired! 2019-10-28T17:09:42 < karlp> Haohmaru: direct link deosn't allow comments and titles either 2019-10-28T17:10:01 < Haohmaru> who needs that 2019-10-28T17:10:14 < Haohmaru> that's only useful for cat picz 2019-10-28T17:14:03 < emeb1> haha - found bug in H7 HAL 2019-10-28T17:14:20 < emeb1> (another) 2019-10-28T17:14:24 < Haohmaru> u win a tinfoil medal 2019-10-28T17:14:49 < emeb1> was hoping for painted plastic 2019-10-28T17:15:08 < Haohmaru> that's if u fix it 2019-10-28T17:15:16 < emeb1> https://community.st.com/s/question/0D50X0000BWoPVxSQN/bug-stm32cubefwh7v150-haladcinit-doesnt-initialize-adc-ovss-bits 2019-10-28T17:18:31 -!- emeb1 [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-28T17:20:19 < karlp> if you report them to zephyr, you can get them fixed :) 2019-10-28T17:20:38 < karlp> st has staff there that are getting shit down, even if zephyr itself is not useful. 2019-10-28T17:22:28 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-28T17:24:25 < jadew> I removed one word to make it extra funny: "Williams-Sonoma had a $275 bread maker listed in their print catalog, and no one was buying it. When they introduced a similar bread maker for $429 and positioned it next to the $275 bread maker, sales of the $275 bread maker nearly doubled. 2019-10-28T17:24:27 < jadew> " 2019-10-28T17:28:36 < emeb> so 2 x "no one was buying it" = ??? 2019-10-28T17:28:54 < jadew> that's the funny part, the word I removed was "almost" 2019-10-28T17:29:00 < jadew> almost no one was buying it 2019-10-28T17:29:00 < emeb> aha 2019-10-28T17:29:31 < jadew> although, almost no one * 2 isn't much different 2019-10-28T17:30:01 < emeb> yeah. a subtle distinction that's only funny to the numerate. 2019-10-28T17:52:04 < steve> hmm, getting a HardFault in xTaskIncrementTick() , am I understanding that the scheduler has to be started before the first tick happens? 2019-10-28T17:53:27 < steve> ala the line in cmsis_os.c that says if (xTaskGetScheudlerState() != taskSCHEDULER_NOT_STARTED) xPortSysTickHandler(); 2019-10-28T17:53:48 < karlp> steve: just don't try and start it yourself. 2019-10-28T17:54:09 < karlp> don't do anything with it at all. 2019-10-28T17:54:23 < steve> what is "it" ? 2019-10-28T17:54:26 < karlp> systick 2019-10-28T17:54:49 < karlp> don't start it, don't turn on it's irqs, don't bother confifuring it 2019-10-28T17:54:56 < steve> but I need to call HAL_IncTick() and the freertos tick in the systick ISR 2019-10-28T17:55:29 < karlp> if you want to use hal, there's explicit readmes on how to make them playtogether. you're followed them right? 2019-10-28T17:56:27 < steve> nope. I thought I could figure it out. I'll take a look 2019-10-28T18:04:00 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 246 seconds] 2019-10-28T18:16:32 < steve> Can't find any docs, just forum posts. I think i'm right, scheduler has to be started before the first tick, otherwise the first tick will HardFault. I'm not using cmsis_os.c; calling FreeRTOS directly 2019-10-28T18:30:15 < steve> had 1 more issue where xQueueSendFromISR fired in an ISR before the queue was created, haha. now it works yay 2019-10-28T18:38:47 < qyx> thats right, schedular has to be started 2019-10-28T18:38:55 < qyx> s/a/e 2019-10-28T18:39:09 < steve> wat 2019-10-28T18:40:16 < qyx> I know the "issue" 2019-10-28T18:40:22 < qyx> 16:52 < steve> hmm, getting a HardFault in xTaskIncrementTick() , am I understanding that the scheduler has to be started before the first tick happens? 2019-10-28T18:40:40 < steve> i mean s/a/e 2019-10-28T18:42:07 < qyx> oh sed replace, schedular->scheduler 2019-10-28T18:42:30 < steve> derp 2019-10-28T18:42:41 < qyx> https://www.gnu.org/software/sed/manual/html_node/The-_0022s_0022-Command.html 2019-10-28T18:43:28 < qyx> to be pendantic, I was wrong because it would match all "a" characters in the sentence 2019-10-28T18:44:31 < qyx> ok, i need to get 5V/200mA from 230V AC 2019-10-28T18:44:35 < steve> yea there's one for "replace 1 occurrence per line". just starting to use sed recently 2019-10-28T18:44:41 < qyx> what would you use, irc pro EEs 2019-10-28T18:45:10 < qyx> considering https://www.tme.eu/sk/details/mpm-05-5/zdroje-pre-zabudovanie/mean-well/ 2019-10-28T18:45:22 < qyx> but huuuge 2019-10-28T18:46:27 < qyx> also pls less than 0.5W under no load.. 2019-10-28T18:46:41 < qyx> wtf there are 1W power supplies with 0.5W draw under no load conditions 2019-10-28T18:47:00 < qyx> this one should be less than 0.075W 2019-10-28T18:49:52 < aandrew> needs moar stm32: https://www.youtube.com/watch?v=qmYpLlBWHp4 2019-10-28T18:50:23 < qyx> irm01-5 is smaller 2019-10-28T18:57:53 < steve> qyx, this tick race issue suggests to me, to initialize BSP inside tasks, because there is not enoguh time in between enabling SYSTick and starting the scheduelr, to do all of the initialization, particularly for BSP inits requiring delays. would you agree? 2019-10-28T18:57:59 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-28T18:59:44 < steve> or I could do something where HAL ticks are enabled first, then FreeRTOS ticks are enabled just before the scheduler starts 2019-10-28T19:01:14 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-28T19:23:54 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-28T19:30:33 < bitmask> these mr30pw connectors came in, they are so cute 2019-10-28T19:33:56 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-28T19:48:07 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-28T20:23:56 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-28T20:32:25 < antto> Jan- is seeking shelter in pic32 x_x 2019-10-28T20:33:05 < antto> u unhelpful es-tea-em fanbois! shame on u! 2019-10-28T20:33:25 < steve> lol, why do I want to look... I don't know 2019-10-28T20:33:27 < specing> antto: it wont be long before she realises that the supposed shelter is actually a bear's den 2019-10-28T20:33:40 < antto> i told her ;P~ 2019-10-28T20:33:54 < antto> even microsh*t has a hint ;P~ 2019-10-28T20:34:10 < antto> "going from pic18 to pic24 to ATSAM" huhuhu 2019-10-28T20:36:02 < steve> are they going to phase out the PIC? 2019-10-28T20:36:46 < antto> maybe not but.. i wouldn't touch pic32 even with a spit 2019-10-28T20:37:04 < antto> i'd feel dirty 2019-10-28T20:37:07 < antto> filthy 2019-10-28T20:37:14 < Steffanx> antto: how long have you been in #avr? 2019-10-28T20:37:28 < antto> i'm not there 2019-10-28T20:37:36 < antto> i woz, long ago 2019-10-28T20:37:48 < Steffanx> I remember she joined #avr. It went EXACTLY like how she acts here 2019-10-28T20:38:38 < Steffanx> It's funny :P 2019-10-28T20:38:44 < Steffanx> Hi Jan- ;) 2019-10-28T20:39:35 < steve> Gartner intelligence gathering robot 2019-10-28T20:41:03 < steve> arduino due seems powerful enough 2019-10-28T20:41:15 < Steffanx> Yeah 2019-10-28T20:42:16 < antto> Steffanx i think muh first log is from Apr 01 22:09:30 2013 2019-10-28T20:42:48 < antto> before that i woz on #arduino because i thought that is related to atmegas and didn't know there is an #avr channel 2019-10-28T20:43:08 < antto> i didn't know what arduino is either 2019-10-28T20:43:14 < antto> ;P~ 2019-10-28T20:43:18 < antto> CRAPduino 2019-10-28T20:44:48 < steve> I personally thanked Massimo for creating Arduino 2019-10-28T20:46:00 < antto> i think i woz in ##electronics asking stuff about muh atmega2561 (i didn't know wut a datasheet is then), and they got annoyed and tossed me into #arduino so i sat there 2019-10-28T20:46:42 < antto> but those there didn't like me either cuz i woz not using the same sh*t they use, i was trying to glue together a bootloader 2019-10-28T20:47:20 < antto> or.. i was trying to program a bootloader onto the chip.. i don't remember 2019-10-28T20:48:19 < Steffanx> It was before that I think, antto 2019-10-28T20:48:42 < antto> u mean u've seen Jan- on #avr before 2013-ish? 2019-10-28T20:48:51 < aandrew> I was really trying to help but this person just did not want to take anything upon themselves and I have a very low tolerance for that 2019-10-28T20:49:57 < antto> aandrew i feel really similar like her, but i *know* i don't have better alternatives so i sit and try to accumulate courage for my next attempt at arm corteggz 2019-10-28T20:50:51 < aandrew> antto: listen... you have a problem. it's causing you frustration and anxiety and you just don't know what to do, so you ask some strangers to help. 2019-10-28T20:50:51 < antto> and in the meantime i deploy an ear here and there and try to learn some little details that could possibly help me later 2019-10-28T20:51:18 < aandrew> these strangers offer a few ideas, ask some questions, and you completely ignore them and continue to stare at the problem and refuse to try to fix it 2019-10-28T20:51:55 < aandrew> so you continue to lament and maybe poke a few things you've already poked a dozen times already, with the same results. you blog about it here and you get some additional questions/offers, which you also ignore 2019-10-28T20:52:06 < aandrew> if that is you, then I would not be able to help you either. 2019-10-28T20:52:51 < aandrew> there was absolutely NOTHING stopping this person from walking through the functions in the autogenerated code in the same goddamned file, googling the function names and stumbling across the "stm32cubemx manual" 2019-10-28T20:52:59 < antto> well i'm not exactly like that, when i jumped to stm32 i already sorta knew some things about working with code generators, building yer own flavor of gcc, and how to make a codeblocks project for some nasty shizzle that comes with a horrific makefile 2019-10-28T20:53:15 < aandrew> there was absolutely NOTHING stopping this person from googling "getting started with cubemx" or "pwm example cubemx" 2019-10-28T20:53:22 < antto> i also knew a lil bit of how muh code turns into .hex 2019-10-28T20:53:58 < aandrew> I don't know if this behaviour is learned or it's just how this person thinks they need to behave to get someone to hold their hand and do it for them, but I refuse to work in this way 2019-10-28T20:54:34 < antto> i absolutely understand your side.. 2019-10-28T20:55:41 < aandrew> If this person came in and said "I can't pwm on stm32 to work with cube, I see http://tutorial and http://docs and http://blogposts but my code just doesn't work. I'm new and my code is at http://paste" they would have received more than enough help and had been on their way in no time 2019-10-28T20:56:30 < aandrew> they did *zero* research on their own. They attempted *nothing* to try to fix it on their own, and expressed *no* curiosity nor made any attempt to see how things fit together 2019-10-28T20:56:42 < antto> i think it's her immune system rejecting any help because of bad impression from the whole situation or something like that 2019-10-28T20:57:34 < aandrew> when my kids say they want to learn something or want help with something and they are genuinely trying I don't care if I have to go over something a hundred times until they get it. I truly love helping people and seeing someone get that "eureka!" moment is one of my guilty pleasures. 2019-10-28T20:57:59 < aandrew> but when my kids stop thinking and just want me to do it for them I have to send them away because it really irritates the everloving shit out of me 2019-10-28T20:58:23 < antto> i know.. i would've helped too if i could 2019-10-28T20:58:52 < aandrew> that's a personal/character flaw I have and I recognize it and that is why I walked away yesterday and will not offer additional help to this individual until I see some effort on their part 2019-10-28T20:59:42 < aandrew> getting started is sometimes really difficult because you don't know what you don't know and don't know what to ask, but the fact that they didn't bother to google "stm32cube tutorial" or someshit and watch a few vids/read a few sites and come back with better questions is very, very telling. 2019-10-28T21:00:23 < antto> well, i don't wanna touch kewbz 2019-10-28T21:00:49 < aandrew> immune system rejecting... go take some educational benadryl and come back when your immune system isn't working against you. :-) 2019-10-28T21:00:54 < antto> so next time imma skip all of that crap and go barefoot 2019-10-28T21:01:02 < aandrew> you'll regret it 2019-10-28T21:01:08 < antto> why? 2019-10-28T21:01:12 < aandrew> there are libraries written for a reason 2019-10-28T21:01:43 < aandrew> bare metal is useful in very few situations. Unless this is for educational purposes or you have a serious need for it, you're just wasting your own productivity 2019-10-28T21:01:59 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-28T21:02:07 < antto> i might use such things just to help me lazily get the periph->pins i need, and perhaps as example for some init sh*t, but i don't generally like these code generators 2019-10-28T21:02:32 < aandrew> I'm saying this from the standpoint of doing this kind of work for almost 30 years... I'm very good at what I do and I hate seeing people fall into the same traps I fell in to 2019-10-28T21:02:41 < antto> i woz thinking libopencm3 counts as baremetal 2019-10-28T21:03:21 < aandrew> agreed, code generation sucks. I typically only use Cube's code generation as a template/for the broad strokes and then write it into my own stuff anyway (see that pastebin on the timers from yesterday) 2019-10-28T21:03:31 < aandrew> how is opencm3 any different from cubemx? 2019-10-28T21:03:50 < aandrew> it's still an abstraction library for the devices and its peripherals, no? 2019-10-28T21:04:19 < antto> uhm, like, i don't have to run some horrific java monster on a crapdows 2019-10-28T21:04:26 < aandrew> oh I see 2019-10-28T21:04:35 < steve> i tell you hwat, CubeMX code is better than MATLAB generated C code 2019-10-28T21:04:37 < antto> i'd rather RTFM 2019-10-28T21:04:38 < aandrew> meh, it's run once to get the clocks/pins set up and then not opened again. :-) 2019-10-28T21:04:44 < steve> CubeMX runs on linux 2019-10-28T21:04:51 < antto> aandrew right 2019-10-28T21:04:57 < aandrew> cubemx does have a very respectable pin configurator 2019-10-28T21:05:35 < antto> and tbh i like atmel.start slightly better, since at least i only poke it from the browser, while teh nasty work is done somewhere on their server or i don't care where 2019-10-28T21:06:01 < Steffanx> Atmel start. Good luck with that. 2019-10-28T21:06:01 < aandrew> I *hate* that idea with the brilliance of a thousand suns 2019-10-28T21:06:12 < Steffanx> When microchip takes the WEBSITE dow 2019-10-28T21:06:12 < aandrew> s/brilliance/intensity/ 2019-10-28T21:06:21 < antto> yeah it's bad in general 2019-10-28T21:06:32 < Steffanx> *poof* there goes your tool :P 2019-10-28T21:06:35 < BrainDamage> don't rely on a remote service for what can be done in local 2019-10-28T21:06:38 < aandrew> asf was cool, it was a git repo. I wish cube did that (and made the repo that their own devs used public) 2019-10-28T21:06:44 < antto> but it too is probably written in java or who knows what (probably equal or worse) 2019-10-28T21:06:47 < aandrew> in fact I told them that in their 2019 wishlist forum post 2019-10-28T21:06:57 < aandrew> BrainDamage: 100% agreed 2019-10-28T21:07:41 < Steffanx> There is no local in case of atmel start 2019-10-28T21:07:59 < antto> BrainDamage i won't rely on it, as i said, i'd only use it for parametrically picking a chip, and generating some basic init code and periph->pin map 2019-10-28T21:08:11 < antto> then i'd just use libopencm3 or whatever 2019-10-28T21:08:33 < BrainDamage> you can try to download the website, altough with javascript the success chance will be dim 2019-10-28T21:08:35 < antto> the generated codez are usually C, and i don't like that 2019-10-28T21:08:49 < antto> i won't.. 2019-10-28T21:09:06 < antto> BrainDamage i have a problem with crap that assumes you have several gigabytes of RAM 2019-10-28T21:09:07 < antto> ;P~ 2019-10-28T21:09:13 * antto is ghetto 2019-10-28T21:09:52 < Steffanx> Me gives antto some extra dedotated wam 2019-10-28T21:09:55 < antto> that pretty much automatically disqualifies java-based IDEs 2019-10-28T21:10:04 * Steffanx / 2019-10-28T21:10:23 < antto> and some other sh*t 2019-10-28T21:10:41 < antto> so i use codeblocks (luckily i've made a gud choice long ago without knowing it) 2019-10-28T21:12:27 < antto> anywayz.. i'd use atmel.start before i buy a chip, before i make a board, so if the webpage is down - i'd ask them "where is teh start thing pls?" and if they say "FU, iz iz gone" then imma say "okay, FU too, imma use some other corteggz thenz!" 2019-10-28T21:13:12 < steve> lol, with your mighty 1x volume 2019-10-28T21:13:38 < antto> hm? 2019-10-28T21:14:01 < antto> oh, sure.. they won't care 2019-10-28T21:14:06 < antto> i'm nobody 2019-10-28T21:14:22 < antto> i don't give a sh*t about that 2019-10-28T21:16:01 < antto> i already made a whole board with some atsame54 with atmel.start and a bunch of RTFM'ing and a few chats with microsh*t support 2019-10-28T21:16:22 < antto> it didn't get fabbed but i think it might have some chances to even work 2019-10-28T21:16:30 < antto> or not eggsplode too much ;P~ 2019-10-28T21:17:59 < antto> but for muh personal project, i need to accumulate moar courage 2019-10-28T21:18:20 < antto> i think imma use same54 cuz i already got a lil bit familiar with it 2019-10-28T21:18:39 < antto> i just don't know if it'd be able to run the sh*t i wanna run in realtime 2019-10-28T21:19:05 < antto> or if i should look for something chunkier/faster like same70 x_x 2019-10-28T21:19:34 < steve> they make soic package SAMs, easy to solder 2019-10-28T21:19:50 < antto> but i can do a few cheatz in analog(ue) so maybe it'll work fine with corteggz M4F even 2019-10-28T21:20:10 < antto> steve i solder TQFP 0.5mm pitch by hand 2019-10-28T21:20:21 < steve> I'm playing on a STM32F7 Discovery, I woulnd't even think about making my own hardware for that 2019-10-28T21:20:52 < antto> hm 2019-10-28T21:20:57 < antto> didn't think of that asspect 2019-10-28T21:21:03 < antto> oops, aspect 2019-10-28T21:21:28 < steve> it's blazing fast though, real time FFTs no problem 2019-10-28T21:21:32 < antto> but i was gonna copy the same70-xpld board as much as possble 2019-10-28T21:22:30 < antto> i want realtime audio ;P~ 2019-10-28T21:22:44 < antto> and FPU 2019-10-28T21:22:55 < antto> i hope 32bit floatz are gonna cut it 2019-10-28T21:23:08 < Steffanx> Go dualcore stm32 2019-10-28T21:23:11 < Steffanx> all you needs in there 2019-10-28T21:23:20 < antto> not sure how that'll help 2019-10-28T21:23:31 < Steffanx> much speed, many real time 2019-10-28T21:23:33 < antto> i'm not doing FFT 2019-10-28T21:23:58 < antto> audio processing usually needs moar MHz not moar parallel processing powah 2019-10-28T21:24:15 < Steffanx> 400Megahurtz + 200Megahurtz. moite. 2019-10-28T21:24:26 < Steffanx> need more hurtz? 2019-10-28T21:24:32 < antto> dayum, that's almost like muh Duron650MHz 2019-10-28T21:24:35 < Steffanx> damn, now i talk like the haohmary guy. 2019-10-28T21:24:41 < antto> >:) 2019-10-28T21:24:52 < BrainDamage> fft is easily parallelizable 2019-10-28T21:24:57 < antto> soon, u all gon talk like spoodrman+lolcatz 2019-10-28T21:25:18 < steve> nah, if you use stateless filters you can parallelize all day 2019-10-28T21:25:38 < antto> i need to make muh 303 into hardware 2019-10-28T21:25:51 < BrainDamage> the main reason why gpu aren't used to process audio is simply because there's a cost in pushing the data back and forth 2019-10-28T21:26:01 < antto> and if that works, imma reuse it to make other synths 2019-10-28T21:26:57 < antto> BrainDamage yes, and some sh*t just depends on the output of teh previous sh*t, which depends on the output of the moar previous sh*t.. etc.. 2019-10-28T21:27:22 < antto> somewhere in the beginning you feed a noice cake, and then it turns into 894329834 kinds of sh*t 2019-10-28T21:28:05 < bitmask> https://i.imgur.com/tgjQQVj.png 2019-10-28T21:28:16 < antto> bravo, bitmask 2019-10-28T21:28:36 < antto> for teh direct link 2019-10-28T21:28:43 < Steffanx> dont use such feature. 2019-10-28T21:28:43 < Steffanx> or does fart sounds need this? 2019-10-28T21:28:43 < Steffanx> -s 2019-10-28T21:28:55 < bitmask> I always use a direct link when its not an album 2019-10-28T21:29:00 < aandrew> purdy. I'd have taken those three traces up and over the pins of J5/J6 to save the vias/bottom trace jumpers 2019-10-28T21:29:29 < aandrew> or taken pin 1 of J5 and J6 and maybe J7 and routed them on the bottom (you've already got vias for them at the RC junctions) 2019-10-28T21:29:37 < antto> bitmask it sorta looks like a quake3 "capture teh flag" map ;P~ 2019-10-28T21:29:45 < aandrew> haha yes it does 2019-10-28T21:30:02 < aandrew> Q10 and Q11 are the only components on the bottom of the board; intentional? 2019-10-28T21:30:07 < Steffanx> Now im happy i never played the quake. 2019-10-28T21:30:32 < antto> Steffanx i meant top-view, u don't play in top-view obviously 2019-10-28T21:30:37 < bitmask> aandrew yea, I may move em to the top but it doesnt really make much sense to, they are low side battery connects so all the ground is going through them 2019-10-28T21:30:54 < bitmask> might as well keep em on the ground plane 2019-10-28T21:31:06 < bitmask> isntead of forcing it up through vias 2019-10-28T21:31:13 < aandrew> just makes assembly worse and also means the board won't lie flat anymore 2019-10-28T21:31:22 < bitmask> yea, still thinking about it 2019-10-28T21:31:50 < aandrew> also the thin trace that's under C15; why not feed that right in to pin 2 of that big inductor instead of going up and into the side of that poly 2019-10-28T21:32:49 < bitmask> its the feedback pin, it said to keep it away from the inductor so I wasnt sure if I should connect it to that pad 2019-10-28T21:32:56 < aandrew> something about that star point at the right side of C16 makes me unhappy but it's just me being fussy 2019-10-28T21:33:21 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-28T21:33:28 < antto> the colors make me unhappy 2019-10-28T21:33:32 < kakinull> didn't follow chats 2019-10-28T21:33:33 < aandrew> nice layout. put layer designators (TOP BOT) and some silk saying what it is and a date code 2019-10-28T21:33:35 < bitmask> yea I didnt like that either but wasn't sure a better way to transmit the 5v to all the pins without doing a pour 2019-10-28T21:33:45 < antto> ++kakinull; 2019-10-28T21:33:57 < kakinull> is she back or did she finally give up? 2019-10-28T21:34:12 < antto> kakinull she gave up and went to seek shelter in pic32 2019-10-28T21:34:15 < kakinull> we did a lot of works last night 2019-10-28T21:34:26 < antto> well, mission failed 2019-10-28T21:34:32 < bitmask> yea, still playing with pad connects (thermals and what not) as the rules arent correct right now, gotta add a usb connector and then i'll finalize the silk 2019-10-28T21:34:50 < kakinull> she misconfigured clocks and blamed on us for locking the chip 2019-10-28T21:34:55 < aandrew> bitmask: might also be time to turn on the paste layer and make sure none of your silkscreen designators are going to get blown away 2019-10-28T21:35:05 < kakinull> so grateful 2019-10-28T21:35:11 < bitmask> yea, im sure some will, havent gone over those yet 2019-10-28T21:35:58 < kakinull> Steffanx: root level zpool? what user and group? 2019-10-28T21:36:09 < kakinull> if I make pools inside the pool 2019-10-28T21:36:11 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-28T21:36:21 < kakinull> and never use root pool directly 2019-10-28T21:36:22 < Steffanx> im not sure what you are even referring to, kakinull 2019-10-28T21:36:30 < kakinull> me neather 2019-10-28T21:36:34 < antto> hahaah 2019-10-28T21:36:35 < Steffanx> me neither 2019-10-28T21:36:45 < antto> r u kaking with me? 2019-10-28T21:36:45 < kakinull> thanks for correction 2019-10-28T21:37:09 < kakinull> I have verbally requested to correct my english time to time 2019-10-28T21:37:32 < kakinull> keep it up steff 2019-10-28T21:37:34 < aandrew> bitmask: https://imgur.com/a/9ZR3dLF is how I usually do layer indicators for fab 2019-10-28T21:37:47 < antto> kakinull u mean correggshun? 2019-10-28T21:37:53 < antto> >:) 2019-10-28T21:38:06 < Steffanx> ive still no clue what you want, kakinull 2019-10-28T21:38:07 < kakinull> you have a thing with eggs 2019-10-28T21:38:14 < aandrew> https://imgur.com/53jPxYJ is how it looks on the board 2019-10-28T21:38:18 < kakinull> Steffanx: just a minute I figure out what I want 2019-10-28T21:38:27 < kakinull> or an hour more likelly 2019-10-28T21:39:03 < kakinull> is antto actually How to Basic? 2019-10-28T21:39:13 < aandrew> https://i.imgur.com/dv8OEJ7.jpg is a better pic, kind of 2019-10-28T21:39:15 < antto> le wot? 2019-10-28T21:39:16 < BrainDamage> I guess kekmir is talking about zfs pools 2019-10-28T21:39:32 < kakinull> antto: he has egg thing too 2019-10-28T21:39:40 < Steffanx> antto is more a bascom 2019-10-28T21:39:43 < bitmask> why do you indicate layers? isn't it separated in the gerbers by layer? 2019-10-28T21:39:48 < antto> dafuq is bascom 2019-10-28T21:39:55 < Steffanx> basic for avr 2019-10-28T21:40:08 < aandrew> bitmask: it's to help make sure the fab house doesn't fuck up the layer order 2019-10-28T21:40:08 < BrainDamage> is that a giant row of sma connectors? 2019-10-28T21:40:16 < aandrew> yes 2019-10-28T21:40:39 < Steffanx> wannabee kakimir? 2019-10-28T21:40:43 < BrainDamage> the meme lives on 2019-10-28T21:40:51 < antto> i've never had any f*ckups with kicad gerb0rz ;P~ 2019-10-28T21:41:09 < aandrew> it's not the EDA tool that does it 2019-10-28T21:41:16 < bitmask> I'm only doing two layers so its labeled top and bottom, dont think it will be an issue 2019-10-28T21:41:16 < aandrew> it's the fab house who swaps layers 2019-10-28T21:41:36 < Steffanx> you expect jlcpcb to even look at those numbers? 2019-10-28T21:41:53 < antto> kicad has this option to put "protel" file eggstensions to the gerb0rz, besides their akchual names being obvious 2019-10-28T21:41:55 < aandrew> "you put the layers in the wrong order" "no you supplied them in the wrong order" 2019-10-28T21:41:56 < antto> i use that 2019-10-28T21:47:02 < antto> btw, Jan- ur lucky loirens wosn't here the whole tiem injecting brown substances into teh wat0rz which are merky enough already ;P~ 2019-10-28T21:48:04 < antto> i fink teh "quality" *cough* of ##stm32 has improved recently 2019-10-28T21:48:36 < Steffanx> whats your definition of quality? 2019-10-28T21:48:50 < antto> foggy 2019-10-28T21:48:52 < antto> ;] 2019-10-28T22:38:27 < zyp> has it? 2019-10-28T22:38:47 < zyp> I haven't seen much interesting conversations lately, but I haven't been around that much either 2019-10-28T22:38:58 < Steffanx> See. it depends on your definition of quality. 2019-10-28T22:39:05 < qyx> https://www.mouser.sk/ProductDetail/Pulse-Electronics/PA0368050NLT?qs=sGAEpiMZZMuRbIF9I%252BxVGI96yZI6DpPV 2019-10-28T22:39:12 < qyx> any idea what footprint should this have? 2019-10-28T22:39:22 < Steffanx> but i did notice daddies get more grumpy over time, so.. perhaps you are suffering from that too, zyp :P 2019-10-28T22:39:38 < qyx> I assume it should be surface mounted 2019-10-28T22:39:45 < zyp> Steffanx, haha, I don't think so 2019-10-28T22:56:06 <@englishman> chat death 2019-10-28T22:59:40 < Steffanx> when will you engage chats @englishman? 2019-10-28T23:00:09 <@englishman> I'll chat about g4 tomorrow I guess 2019-10-28T23:00:28 <@englishman> anyone have any questions for the FAEs or anything? 2019-10-28T23:00:31 < Steffanx> i bought some of those G4 ESCs to play with. is that bad? 2019-10-28T23:00:52 <@englishman> it's so weird they they don't have CAN 2019-10-28T23:01:11 < Steffanx> they dont? im pretty sure this tiny board has 2019-10-28T23:02:26 < bitmask> what happens if you have a lot of current on a net and theres a short path that can't handle the full load but theres another longer path that can handle the rest easily? will the shorter path heat up at all or will it mostly go to the longer path as the heat increases resistance? 2019-10-28T23:03:41 <@englishman> not according to when I checked cube yesterday 2019-10-28T23:03:46 <@englishman> maybe cube is wrong 2019-10-28T23:03:53 < Steffanx> cube is wrong 2019-10-28T23:03:59 <@englishman> it said there was no can on g0 or g4 2019-10-28T23:04:02 <@englishman> ah lol 2019-10-28T23:04:06 < qyx> G4 == ST32G4? 2019-10-28T23:04:10 <@englishman> yes 2019-10-28T23:04:13 < qyx> ESC == motor controller? 2019-10-28T23:04:14 < Steffanx> STM32* 2019-10-28T23:04:17 < Steffanx> yes 2019-10-28T23:04:22 <@englishman> are there g4 weith 3x can? 2019-10-28T23:04:48 < Steffanx> uhmm 2019-10-28T23:05:26 < qyx> what for 2019-10-28T23:05:33 < Steffanx> it might be listed as FDCAN btw, mr englishman 2019-10-28T23:05:42 <@englishman> you got the 4x esc? did you get the drone? 2019-10-28T23:05:50 < Steffanx> So when you filter on CAN it might not show up. 2019-10-28T23:05:51 <@englishman> oh cool what's fdcan 2019-10-28T23:06:01 < qyx> a can for can fd? 2019-10-28T23:06:17 < Steffanx> ^ 2019-10-28T23:06:20 <@englishman> yes 3x fdcan nice 2019-10-28T23:06:23 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-28T23:06:29 < Steffanx> much flexible, much data-rate. 2019-10-28T23:06:39 < qyx> so bits 2019-10-28T23:06:44 < qyx> you can even do a can switch! 2019-10-28T23:06:48 <@englishman> oh thats cool 2019-10-28T23:07:09 < Steffanx> i only see 0-2 FDCAN 2019-10-28T23:07:16 < qyx> I should look into them 2019-10-28T23:07:36 < qyx> no fdcan support in locm3 probably though 2019-10-28T23:07:49 <@englishman> one could be shared like in f413 2019-10-28T23:08:02 < Steffanx> but libopencm3 is so awesome, you can add it yourself qyx 2019-10-28T23:08:35 <@englishman> lolz 2019-10-28T23:09:11 < qyx> hard is 2019-10-28T23:09:15 < qyx> the world of opensores 2019-10-28T23:09:22 < Steffanx> Yes, where to start 2019-10-28T23:09:33 <@englishman> suicide 2019-10-28T23:10:30 < qyx> I need no fdcan yet, so Im safe for now 2019-10-28T23:10:59 <@englishman> pins for 3xfdcan even on qfp48 2019-10-28T23:11:00 <@englishman> nice 2019-10-28T23:11:15 < Steffanx> my cube still only shows 2xfdcan 2019-10-28T23:11:17 <@englishman> st has really unfucked peripheral assignment 2019-10-28T23:11:22 < Steffanx> whats wrong with my cube 2019-10-28T23:11:23 <@englishman> g473 2019-10-28T23:11:30 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-28T23:11:35 <@englishman> shows as 3x 2019-10-28T23:11:40 <@englishman> on the site 2019-10-28T23:12:00 <@englishman> 473, 474, 483, 484 2019-10-28T23:12:03 < Steffanx> ah true 2019-10-28T23:13:23 <@englishman> not too expensive either 2019-10-28T23:15:01 <@englishman> so did you get the stm32 drone 2019-10-28T23:16:33 < kakinull> how do I boot into bios with system with option rom? 2019-10-28T23:16:41 < kakinull> multiple option roms 2019-10-28T23:16:49 < kakinull> bios promt not visible 2019-10-28T23:17:34 < qyx> wat 170MHz 2019-10-28T23:17:35 < kakinull> any other way than pulling the cards 2019-10-28T23:17:46 <@englishman> yeah qyx 2019-10-28T23:17:50 <@englishman> g4 is dope 2019-10-28T23:17:53 < qyx> cordic and fmac 2019-10-28T23:17:59 < qyx> is this for sdr decoding or what 2019-10-28T23:19:11 < qyx> I am curious why the L4+ family is the only one with octoSPI 2019-10-28T23:20:09 < qyx> 1617 timers: 2019-10-28T23:20:09 <@englishman> oh shit 2019-10-28T23:20:13 < qyx> lol 2019-10-28T23:20:15 <@englishman> that fmac 2019-10-28T23:20:33 < qyx> 15msps 4 channel DAC? 2019-10-28T23:20:41 < qyx> internal channels mhm 2019-10-28T23:20:44 <@englishman> for $4 2019-10-28T23:20:51 <@englishman> half the price of msp430 2019-10-28T23:23:45 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Remote host closed the connection] 2019-10-28T23:24:07 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-28T23:24:07 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-28T23:24:07 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-28T23:29:39 -!- machinehum [~machinehu@S01061cabc0ab4603.vc.shawcable.net] has quit [Ping timeout: 268 seconds] 2019-10-28T23:42:18 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-28T23:53:13 -!- machinehum [~machinehu@S01061cabc0ab4603.vc.shawcable.net] has joined ##stm32 2019-10-28T23:57:18 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-28T23:58:56 < kakinull> now I have pool --- Day changed Tue Oct 29 2019 2019-10-29T00:00:31 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-29T00:01:40 * antto goes in, teh wat0r turns yellow, goes out 2019-10-29T00:13:15 < Cracki> uranium dust 2019-10-29T00:14:29 < kakinull> that's bad 2019-10-29T00:14:34 < kakinull> don't breathe that shit 2019-10-29T00:16:43 < Cracki> it's in every coal power plant's ash 2019-10-29T00:16:57 < Steffanx> Will you turn into The Mask, kakinull ? 2019-10-29T00:17:09 < kakinull> no 2019-10-29T00:17:16 < Steffanx> It suits you 2019-10-29T00:20:10 < bitmask> oh damn I forgot mounting holes, 3.5mm for M3? 2019-10-29T00:21:24 < aandrew> englishman: which FAEs 2019-10-29T00:24:06 < Cracki> m3 OD is 3.0 mm, so you can go with 3.2 mm for the "nominal tight fit" 2019-10-29T00:24:15 < Cracki> or 3.1 or 3.0 if you have plastic 2019-10-29T00:24:59 < Cracki> I'd go for 3.0 if it's 3d printed plastic. makes for good alignment. 2019-10-29T00:25:15 < Cracki> 3.0 printed, not designed :> I know how these things are 2019-10-29T00:25:43 < bitmask> i was talking about my pcb 2019-10-29T00:25:47 < Cracki> ah! 2019-10-29T00:25:57 < Cracki> eh, 3.2 mm, I'm sure that's a standard drill they have 2019-10-29T00:26:56 < Cracki> random tables: https://blogs.mentor.com/tom-hausherr/blog/tag/pcb-mounting-holes/ 2019-10-29T00:27:05 < Cracki> 3.5 mm is "loose", "3.1" is tight 2019-10-29T00:27:31 < kakinull> is 30C the perfect temperature for HDD 2019-10-29T00:27:44 < kakinull> ? 2019-10-29T00:28:07 < bitmask> k 2019-10-29T00:28:13 < Cracki> https://upload.wikimedia.org/wikibooks/en/3/38/Average_temperatures_versus_failures_rates_for_HDDs.jpg 2019-10-29T00:28:31 < Cracki> that's from google or someone's big data collection 2019-10-29T00:28:39 < Cracki> they are designed for normal environments 2019-10-29T00:28:58 < jadew> kakinull, the perfect temp for hdd is stable 2019-10-29T00:29:06 < Cracki> :D https://www.backblaze.com/blog/hard-drive-temperature-does-it-matter/ 2019-10-29T00:29:20 < jadew> 30 °C sounds cold 2019-10-29T00:29:26 < jadew> which is good 2019-10-29T00:29:27 < kakinull> jadew: how about 0C? 2019-10-29T00:29:35 < kakinull> serious question 2019-10-29T00:29:36 < kakinull> 0 2019-10-29T00:29:39 < Cracki> joints may start to creak 2019-10-29T00:29:52 < BrainDamage> 0 is too cold 2019-10-29T00:29:57 < jadew> kakinull, check the datasheet 2019-10-29T00:30:00 < Cracki> hdd recovery peeps hot air the spindle for crudded up specimen 2019-10-29T00:30:10 < jadew> technically, cold = good for electronics, but you have mechanical stuff in there too 2019-10-29T00:30:18 < jadew> and stuff that needs to be aligned 2019-10-29T00:31:21 < Cracki> going by the google chart, I'd aim for 35-40 celsius 2019-10-29T00:31:22 < BrainDamage> you can insulate it and let the own consumed power to warm it up 2019-10-29T00:31:48 < kakinull> thing is that ambient will change 2019-10-29T00:32:02 < kakinull> I need controllable heatpipes 2019-10-29T00:32:10 < kakinull> space technology 2019-10-29T00:32:11 < jadew> personally, I'd use an SSD in those conditions 2019-10-29T00:32:12 < Cracki> maybe ssd 2019-10-29T00:32:20 < Cracki> or sd card 2019-10-29T00:32:20 < BrainDamage> insulate the whole pc and run idle tasks to warm it up depending on the temp >.> 2019-10-29T00:32:21 < kakinull> this is NAS 2019-10-29T00:32:36 < Cracki> nas with active ventilation 2019-10-29T00:32:38 < Cracki> ? 2019-10-29T00:32:44 < BrainDamage> I assume it's outside 2019-10-29T00:32:46 < kakinull> some ventilation 2019-10-29T00:32:48 < Cracki> then the fans will spin in reaction to temperature 2019-10-29T00:32:49 < jadew> kakinull, then put it in a box and listen to BrainDamage 2019-10-29T00:32:56 < jadew> it will heat itself up 2019-10-29T00:32:57 < Cracki> yeh box it. 2019-10-29T00:33:05 < kakinull> how about summer? 2019-10-29T00:33:11 < jadew> you take it out of the box 2019-10-29T00:33:13 < Cracki> you hose it down 2019-10-29T00:33:28 < kakinull> I know what to do 2019-10-29T00:33:34 < kakinull> install 1kw electric heater inside 2019-10-29T00:33:36 < BrainDamage> just have a panel you can open for the summer 2019-10-29T00:33:45 < jadew> or! 2019-10-29T00:33:45 < kakinull> with thermostat 2019-10-29T00:33:48 < BrainDamage> in fron the ventilation holes 2019-10-29T00:33:53 < jadew> do what everyone else does and keep it inside 2019-10-29T00:34:11 < BrainDamage> yeah, outside might not be great for other things like humidity 2019-10-29T00:34:17 < jadew> or thieves 2019-10-29T00:34:21 < kakinull> it's in cold storage 2019-10-29T00:34:38 < BrainDamage> underground? 2019-10-29T00:34:42 < kakinull> I wish 2019-10-29T00:34:44 < Cracki> outhouse 2019-10-29T00:35:40 < Cracki> what temp range will it face 2019-10-29T00:35:43 < kakinull> in playhouse 2019-10-29T00:36:05 < kakinull> one dude coded in playhouse 2019-10-29T00:36:22 < kakinull> he moved in playhouse and codez 2019-10-29T00:36:26 < bitmask> damn, I don't have room for m3 mounting holes, should I go with m2 or just design the case to clip it in place 2019-10-29T00:36:32 < Cracki> m2 is fine 2019-10-29T00:36:41 < Cracki> m3 is huge compared to some pcbs 2019-10-29T00:36:45 < kakinull> there is 2.5 too 2019-10-29T00:36:47 -!- kow_ [~afed@135.0.26.39] has joined ##stm32 2019-10-29T00:36:54 < bitmask> i dont know if its worth buying m2 screws, i probably have some somewhere but not sure what sizes 2019-10-29T00:36:55 < kakinull> let's play video games> 2019-10-29T00:37:11 < kakinull> I have piles of 2 2.5 3 4 5 2019-10-29T00:37:23 < Cracki> finland isn't gonna get all that hot in the summer, right? 2019-10-29T00:37:31 < Cracki> then all you _may_ need is active heating 2019-10-29T00:37:36 < ohsix> hello 2019-10-29T00:37:37 < kakinull> true 2019-10-29T00:37:39 < bitmask> i only KNOW that I have 3 and 4, I have a bag of loose screws that probably has 2 and 2.5 2019-10-29T00:37:48 < kakinull> find the balance Cracki 2019-10-29T00:37:54 < kakinull> maybe just mine buttcoin? 2019-10-29T00:37:58 < Cracki> thx for reminding me that my nearest hw store has nothing below m3 2019-10-29T00:38:01 < kakinull> use it as thermostat 2019-10-29T00:38:06 -!- k\o\w [~afed@135.0.26.39] has quit [Ping timeout: 268 seconds] 2019-10-29T00:38:15 < kakinull> go to bolt specialist 2019-10-29T00:38:17 < ohsix> i forgot who the openocd wizard was; need someone to say if you can do https://i.imgur.com/Zz5DZ6g.png like in rvds with openocd before i spend time on it 2019-10-29T00:38:31 < Cracki> I'll just order it off aliex. mailman doesn't see a dime but eh 2019-10-29T00:38:32 < BrainDamage> assuming your outside temp doesn't reach 30 degs, normal cooling will work fine 2019-10-29T00:38:48 < Cracki> ohsix, PaulFertser is said to be the one 2019-10-29T00:39:05 < ohsix> also, if that were wrong, would the auto probing of the scan chain fail, with zeros on tdo or some similar thing 2019-10-29T00:39:06 < BrainDamage> only make sure the device won't be exposed to direct sunlight 2019-10-29T00:39:12 < ohsix> Cracki: ahh yea that's the one; been a while 2019-10-29T00:39:19 < kakinull> http://tuukkavirtaperko.net/comic/1/ do this in bolt specialist store 2019-10-29T00:39:23 < steve> ohsix go to #openocd 2019-10-29T00:39:26 < BrainDamage> even just a cardboard panel in front can do wonders 2019-10-29T00:40:15 < ohsix> steve: that would be silly! 2019-10-29T00:40:46 < kakinull> is it good idea to run video survelance server on freenas? 2019-10-29T00:41:04 < kakinull> that would warm it up 2019-10-29T00:42:35 < BrainDamage> make sure to set properly the io and cpu scheduling nice 2019-10-29T00:42:47 < BrainDamage> so that it won't delay when you need to access it 2019-10-29T00:43:44 < Jan-> motherfu... 2019-10-29T00:43:47 < Jan-> sorry wrong window 2019-10-29T00:44:05 < Steffanx> Why isn't your nas in your room, kakinull ? 2019-10-29T00:44:10 < kakinull> it is 2019-10-29T00:44:10 < Steffanx> Mine is 2019-10-29T00:44:18 < Steffanx> Is your room that cold? 2019-10-29T00:44:26 < Steffanx> Isnt the house heated? 2019-10-29T00:44:48 < kakinull> it's +17 at floor level and +21 at ceiling 2019-10-29T00:44:57 < Steffanx> Perfect 2019-10-29T00:45:14 < kakinull> when I turn off heating it's around -5 - 0 2019-10-29T00:45:44 < Steffanx> You do that? 2019-10-29T00:46:05 < ohsix> so there's one or two people in here that change their name all the time; waht's their current name hehe 2019-10-29T00:46:31 < Steffanx> doomba: 2019-10-29T00:46:37 < Steffanx> kakinull: 2019-10-29T00:46:43 < ohsix> roger that 2019-10-29T00:46:47 < Steffanx> Much highlight 2019-10-29T00:46:52 < kakinull> jly? 2019-10-29T00:47:24 < kakinull> where is he anyways 2019-10-29T00:48:04 < kakinull> Steffanx: if I'm not around I turn off heating 2019-10-29T00:48:22 < ohsix> where does nxp hide their bsdl files D: 2019-10-29T00:48:34 < BrainDamage> kakinull: then the self heating solution will be perfect 2019-10-29T00:48:54 < kakinull> ohsix: lpc? 2019-10-29T00:48:55 < BrainDamage> there won't be task conflict either since you won't have to access the nas if you're away 2019-10-29T00:49:09 < kakinull> I might 2019-10-29T00:49:20 < kakinull> run kakicloud server 2019-10-29T00:49:43 <@englishman> aandrew: stm32 faes from the states not sure which ones yet 2019-10-29T00:49:45 < doomba> gotta keep the HR intelligence AI companies on their toes 2019-10-29T00:50:13 < kakinull> videogayms> 2019-10-29T00:51:09 < ohsix> kakinull: mx5 2019-10-29T00:51:23 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-29T00:51:45 < ohsix> it's got the weird scan chain junk 2019-10-29T00:53:15 < Steffanx> He's on telegram, kakinull 2019-10-29T00:55:21 <@englishman> this cheap hotel is fuckin nice 2019-10-29T00:56:41 < ohsix> we got one of the expensive jlinks for a project but i've just been using openocd and an olimex thing dongs sent me years ago 2019-10-29T00:56:56 < ohsix> purchaser and stuff is non technical, nobody understands the irony, it's great 2019-10-29T01:01:19 < Steffanx> Ozone is awesome ohsix 2019-10-29T01:01:31 < aandrew> yep I really like ozone 2019-10-29T01:01:41 < aandrew> I wonder if I can telnet to it to get its cli 2019-10-29T01:04:28 < ohsix> yea i'll probably use it if it's a new project but this is all poking at existing things 2019-10-29T01:05:08 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-29T01:07:32 < bitmask> hmm, 4mm or 5mm m2 screws 2019-10-29T01:08:25 < Steffanx> 6 2019-10-29T01:08:49 < bitmask> why 6 2019-10-29T01:10:13 < Steffanx> Why 4 or 5 2019-10-29T01:10:20 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-ooudrtnnczumfwhy] has quit [Quit: Connection closed for inactivity] 2019-10-29T01:10:39 < bitmask> touche 2019-10-29T01:33:02 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-29T01:55:10 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-29T02:13:40 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-29T02:39:18 < bitmask> https://i.imgur.com/9sPGhkF.png 2019-10-29T02:40:11 < bitmask> few more things to do 2019-10-29T02:41:26 < kakipro> multiple coils 2019-10-29T02:42:00 < bitmask> I'm using a separate smps as a phone charger 2019-10-29T02:42:52 < bitmask> brb 2019-10-29T02:52:48 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 265 seconds] 2019-10-29T02:56:36 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-29T03:00:37 < kakinull> that regulator looks nasty 2019-10-29T03:01:02 < kakinull> does the part id start with XL? 2019-10-29T03:06:42 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-29T03:09:24 < kakinull> which one is for phone charger? 2019-10-29T03:14:02 < bitmask> I'm using the cheap MP1593 for phone charger and LM2595 for mcu 2019-10-29T03:16:38 < bitmask> what looks nasty? the smallness of it? 2019-10-29T03:17:58 < kakinull> LM2595 2019-10-29T03:18:11 < kakinull> what is good about it? 2019-10-29T03:18:23 < bitmask> no idea, thought it was popular? 2019-10-29T03:18:35 < bitmask> I know the 2596 is 2019-10-29T03:19:05 < bitmask> what would you use for 1A and for 3A? 2019-10-29T03:19:50 < kakinull> well I mean it will work 2019-10-29T03:20:06 < bitmask> is it the frequency? package? 2019-10-29T03:20:20 < bitmask> external components? 2019-10-29T03:20:26 < kakinull> both 2019-10-29T03:20:31 < kakinull> that 2019-10-29T03:21:24 < kakinull> datasheet year of release 1999 2019-10-29T03:21:39 < bitmask> what would you use? I didn't know what else to go with 2019-10-29T03:21:53 < kakinull> some MP chip 2019-10-29T03:22:21 < bitmask> oh really? the MP stuff is good? hmm 2019-10-29T03:22:39 < kakinull> well more modern 2019-10-29T03:22:44 < bitmask> I just used it because i had a cheap china module for a phone charger 2019-10-29T03:22:52 < bitmask> so I figured id use the same ic 2019-10-29T03:25:48 < bitmask> think the 3A would be enough for both a phone charger and the rest of the circuit? it only draws like 300mA 2019-10-29T03:25:59 < bitmask> or is that a bad idea 2019-10-29T03:26:08 < kakinull> it's enough 2019-10-29T03:26:11 -!- turnip420 [~machinehu@ip-142-232-170-235.ptr.bcit.ca] has joined ##stm32 2019-10-29T03:26:26 < kakinull> if the circuit is tuned to work optimally 2019-10-29T03:28:02 < bitmask> I guess I can add a jumper and leave the lm2595 stuff unsoldered, if it doesnt work then I can add it 2019-10-29T03:28:23 < bitmask> or look at other 1A smps with smaller components 2019-10-29T03:28:23 < kakinull> nah man 2019-10-29T03:29:00 < bitmask> alright, i'll just leave it off 2019-10-29T03:29:21 < kakinull> I just teasing you for going with pre-millenium era switching regulator 2019-10-29T03:30:15 < bitmask> well id rather make it better, and if it doesnt need both supplies I'd rather not 2019-10-29T03:30:34 < kakinull> what is the voltage of heaters? 2019-10-29T03:30:48 < kakinull> vbat? 2019-10-29T03:30:48 < bitmask> the heaters are 12v 2019-10-29T03:30:50 < bitmask> yea 2019-10-29T03:33:50 < kakinull> do you have reference design of that MP part? 2019-10-29T03:33:55 < kakinull> *with that 2019-10-29T03:34:07 < kakinull> I assume chinaboard 2019-10-29T03:34:17 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-29T03:35:04 < kakinull> you can simply draw 3amps out of it and see what it does 2019-10-29T03:35:14 < kakinull> with the minimum battery voltage ofc 2019-10-29T03:35:46 < bitmask> yea 2019-10-29T03:35:48 < bitmask> true 2019-10-29T03:36:26 < kakinull> your battery voltage will be 9volts minimum? 2019-10-29T03:37:26 < kakinull> note: MP chip has exposed pad 2019-10-29T03:38:56 < bitmask> yea and yea 2019-10-29T03:42:30 < bitmask> mp2344 is 2A/600kHz in a TSOT23-6 2019-10-29T03:43:01 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-29T03:43:34 < kakinull> https://www.openhacks.com/uploadsproductos/datasheet_77.pdf 2019-10-29T03:43:40 < kakinull> this has more 2019-10-29T03:44:44 < kakinull> I pulled 2.4Amps from it at output voltage of 5V 2019-10-29T03:44:52 < kakinull> with battery voltage under 6volts 2019-10-29T03:45:47 < kakinull> on board that was hardly bigger than my 2 thumb nails combined 2019-10-29T03:46:44 < bitmask> heatsink? 2019-10-29T03:47:09 -!- learning1 [~pi@121.121.99.187] has joined ##stm32 2019-10-29T03:47:12 < kakinull> first it started thermal cycling at 1.9A but after applying optional tweaks mentioned in datasheet + the bestest 4mm coil or so I got it up to 2.4A 2019-10-29T03:47:19 < kakinull> no heatsink 2019-10-29T03:47:32 < kakinull> A-type usb socket though 2019-10-29T03:48:15 < kakinull> i don't remember if it was 4mm 5mm or 6mm coil but it was pretty small 2019-10-29T03:50:04 < kakinull> 2.4A is the current ipad pulls if it detects charger that can provide 2.4A 2019-10-29T03:50:17 < bitmask> how does it poll that? 2019-10-29T03:50:22 < bitmask> I was wondering 2019-10-29T03:51:02 < bitmask> does it just measure what it can draw? 2019-10-29T03:51:17 < kakinull> http://file2.dzsc.com/product/17/05/13/759615_155559576.pdf I threw this there and badabing badaboom 2019-10-29T03:51:21 < bitmask> or does it require usb pin nonsense 2019-10-29T03:51:31 < kakinull> everything charged at maximum power 2019-10-29T03:51:42 < kakinull> apple has it's own resistor divider values for data pins 2019-10-29T03:51:51 < kakinull> samsung has it's own 2019-10-29T03:51:56 < kakinull> sony 2019-10-29T03:51:59 < kakinull> etc. etc. 2019-10-29T03:52:15 < bitmask> interesting 2019-10-29T03:53:22 < kakinull> such emulator chip makes a guess based on what kind of polling sequence it sees on data lines 2019-10-29T03:54:21 < kakinull> you don't get that chip on your chinese usb charger module you order from ali 2019-10-29T03:54:46 < kakinull> it's just D+ D- short in all of them basically and it says something like "hey you can pull 1A" 2019-10-29T03:55:17 < kakinull> devices like iphone won't pull even that because it's not apple charger 2019-10-29T03:56:43 < bitmask> where do you even buy that IC? 2019-10-29T03:56:47 < bitmask> no one has it 2019-10-29T03:57:56 < kakinull> you don't find it from western sellers 2019-10-29T03:58:07 < bitmask> its not even on lcsc or aliexpress 2019-10-29T03:58:31 < kakinull> I think there is similar or even pin identical chip from some western vendor 2019-10-29T04:00:01 < kakinull> https://www.maximintegrated.com/en/products/interface/universal-serial-bus/MAX14630.html maxim makes variety of them 2019-10-29T04:01:32 < kakinull> just google usb charger emulator ic 2019-10-29T04:06:14 -!- con3|2 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-29T04:07:21 <@englishman> kakinull: go to bed 2019-10-29T04:07:38 <@englishman> and never buy anything from maxim 2019-10-29T04:07:56 < kakinull> hmm 2019-10-29T04:08:09 < kakinull> I might have some maxim chip 2019-10-29T04:08:46 -!- con3 [~kvirc@154.119.40.237] has quit [Ping timeout: 252 seconds] 2019-10-29T04:08:53 -!- con3|2 is now known as con3 2019-10-29T04:10:06 < kakinull> maybe some comparators 2019-10-29T04:10:13 < kakinull> nanopower 2019-10-29T04:10:41 <@englishman> stm32g4 is amazing 2019-10-29T04:10:42 < kakinull> https://datasheets.maximintegrated.com/en/ds/MAX921-MAX924.pdf yes 2019-10-29T04:11:15 < kakinull> englishman: yo dawg.. I'm already in bed 2019-10-29T04:11:49 < turnip420> englishman: +1 to not buying those fucker's chips 2019-10-29T04:12:02 < kakinull> why? 2019-10-29T04:12:05 < turnip420> What's your favorite part about the g4? 2019-10-29T04:12:23 < turnip420> kakinull: Because they're fucking impossible to get if you're not google 2019-10-29T04:12:33 -!- mirage335 [~mirage335@204.141.172.74] has quit [Ping timeout: 265 seconds] 2019-10-29T04:12:38 <@englishman> the single pair of power pins and non-shit pin assignment 2019-10-29T04:12:39 < turnip420> They have a shit ton of shit, good luck getting anything 2019-10-29T04:12:47 <@englishman> turnip420 is 100% correct 2019-10-29T04:12:54 < turnip420> aka machinehum 2019-10-29T04:13:00 <@englishman> for a stoned root vegetable you're pretty smart 2019-10-29T04:13:47 <@englishman> the amazing math acceleraters, tons of CCM and amazing low price do it too 2019-10-29T04:13:49 <@englishman> amaze 2019-10-29T04:14:00 < turnip420> CCM? 2019-10-29T04:14:05 < turnip420> tf is that? 2019-10-29T04:14:39 < turnip420> Also is it fast fpu's? or just shit to do a sin function 2019-10-29T04:14:53 < kakinull> bitmask: your board is so big and input voltage so high you have plenty of margin in your regulator design parameters.. if datasheet says 3Amps you are probably going to get 3Amps if you have good heat transfer from regulator and not completelly crap inductor 2019-10-29T04:14:57 <@englishman> yeah basically sse 2019-10-29T04:17:39 < bitmask> still not sure what i wanna do 2019-10-29T04:18:05 < bitmask> I shoulda done some more research on switchers before I ordered everything and designed this 2019-10-29T04:18:05 < kakinull> do something 2019-10-29T04:18:35 < kakinull> do you have physical reference design on that mp chip? 2019-10-29T04:19:43 < kakinull> I would load it Vin 9V and Cout 3A 2019-10-29T04:19:47 < bitmask> yea 2019-10-29T04:20:04 < turnip420> englishman: What's the deal with opamps inside the mcu? 2019-10-29T04:20:25 < turnip420> Like can you power their rails externally or? 2019-10-29T04:21:09 < kakinull> to mitigate design component count 2019-10-29T04:21:10 <@englishman> what does the datasheet say 2019-10-29T04:21:14 < turnip420> fine 2019-10-29T04:22:05 < kakinull> bitmask: do you have means to conduct such test? 2019-10-29T04:22:11 < bitmask> not really 2019-10-29T04:22:27 < bitmask> no power supply, no electronic load 2019-10-29T04:23:06 < kakinull> in that situation it's good idea to leave all designs to board 2019-10-29T04:23:39 < kakinull> maybe add them jumpers if it works with just 1 regulator 2019-10-29T04:24:01 < kakinull> ohms law is your poor mans electronic load bitmask 2019-10-29T04:24:03 < bitmask> yea i'm gonna think about it a bit before I do anything 2019-10-29T04:25:10 < kakinull> add charger emulator if you can find one 2019-10-29T04:25:18 < kakinull> it makes all the difference 2019-10-29T04:25:19 < bitmask> yea i'll go with the max 2019-10-29T04:27:45 < kakinull> if you do reference design provided in datasheet it just works 2019-10-29T04:29:40 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-29T04:30:34 < kakinull> you are scared now bitmask? 2019-10-29T04:31:57 < kakinull> englishman: my internal clock is perfect for quebec timezone 2019-10-29T04:34:09 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-dyohdytgyvzhlemr] has joined ##stm32 2019-10-29T04:36:57 < bitmask> yes im scared now 2019-10-29T04:40:57 < kakinull> good 2019-10-29T04:42:05 < kakinull> there wasn't problem before a started talking 2019-10-29T04:42:09 < kakinull> now there is 2019-10-29T04:43:53 < kakinull> my job is done 2019-10-29T04:44:06 < kakinull> I'm quite amused by this 2019-10-29T04:45:44 < kakinull> I wonder if there is jobs in EE world that are for special problem creators 2019-10-29T04:47:38 < bitmask> :) 2019-10-29T04:51:55 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-29T04:52:02 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-29T05:09:13 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has joined ##stm32 2019-10-29T05:09:53 < R2COM> yo 2019-10-29T05:31:24 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 246 seconds] 2019-10-29T05:41:52 -!- turnip420 [~machinehu@ip-142-232-170-235.ptr.bcit.ca] has quit [Quit: WeeChat 1.4] 2019-10-29T06:04:45 < dongs> sup dongs 2019-10-29T06:04:47 < dongs> pcbaing and raging 2019-10-29T06:05:04 < dongs> machine decided to radnomly stop picking up 0402 beads and 0402 18pfs 2019-10-29T06:05:15 < dongs> piece of jap trash 2019-10-29T06:08:04 < Cracki> have you tried turning it off and on again 2019-10-29T06:10:53 < dongs> yeah 2019-10-29T06:22:22 < aandrew> probably got a spec of natto stuck in the head 2019-10-29T06:33:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-29T06:33:14 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-29T06:41:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [] 2019-10-29T06:49:05 -!- fc5dc9d4 [~quassel@p57A32E83.dip0.t-ipconnect.de] has joined ##stm32 2019-10-29T06:52:26 -!- fc5dc9d4_ [~quassel@p5B081248.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-29T07:04:27 < jadew> local electronics shops suck 2019-10-29T07:04:49 < jadew> same screw I can get from TME is exactly 10 times more expensive 2019-10-29T07:13:04 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-29T07:16:04 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 246 seconds] 2019-10-29T07:16:05 -!- day__ is now known as day 2019-10-29T07:34:59 < bitmask> hmmm, MP2330C says it can handle 3A but its such a tiny package 2019-10-29T07:35:14 < bitmask> I think I'm gonna go with the older 2315 just because I dont trust it :P 2019-10-29T07:37:15 -!- Miyu [~hackkitte@2a02:6d40:30e6:ef01:55f9:a28b:188:fc21] has joined ##stm32 2019-10-29T07:40:54 -!- hackkitten [~hackkitte@2a02:6d40:30e6:ef01:d9ec:62e3:b85f:c28] has quit [Ping timeout: 246 seconds] 2019-10-29T08:23:35 -!- jly [uid355225@gateway/web/irccloud.com/x-guekpaxcohlvamdq] has joined ##stm32 2019-10-29T08:27:18 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-29T08:31:30 -!- Miyu is now known as hackkitten 2019-10-29T08:35:33 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-29T08:40:17 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-29T09:02:53 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-29T09:28:24 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-29T09:31:36 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-29T09:58:52 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-29T10:00:06 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-29T10:01:44 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-29T10:25:00 -!- R2COM [~R2COM@ip24-251-252-131.ph.ph.cox.net] has quit [] 2019-10-29T11:02:42 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-29T11:02:43 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Read error: Connection reset by peer] 2019-10-29T11:35:15 -!- tprrt [~tprrt@217.114.204.178] has quit [Ping timeout: 240 seconds] 2019-10-29T11:41:38 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 268 seconds] 2019-10-29T11:44:41 < karlp> qyx: if you find a nice solution to your power problems, do share :) 2019-10-29T11:46:37 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-29T11:51:58 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-29T11:53:33 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-29T11:55:18 < karlp> englishman: yah, g4 looks fab. 2019-10-29T11:56:24 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-29T11:56:39 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Read error: Connection reset by peer] 2019-10-29T12:11:25 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-29T12:15:24 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-29T12:24:38 -!- grummund [~grummund@unaffiliated/grummund] has quit [Remote host closed the connection] 2019-10-29T12:30:30 < Steffanx> Hello jly 2019-10-29T12:33:45 < jly> hi 2019-10-29T12:55:15 < kakipro> :o 2019-10-29T12:55:29 < Steffanx> Hello kakipro 2019-10-29T12:55:39 < Steffanx> Where is kakinull 2019-10-29T12:55:55 < kakipro> sleepin 2019-10-29T12:58:30 < kakipro> he is like that 2019-10-29T13:00:58 < Steffanx> The bastard 2019-10-29T13:11:35 < doomba> kaki learns Rust and changes name to kakiVec 2019-10-29T13:37:51 < Haohmaru> fancymir 2019-10-29T13:38:38 < Steffanx> Kakimaru. 2019-10-29T13:40:29 < Haohmaru> that's illegal, Steffanx 2019-10-29T13:40:33 * Haohmaru calls teh cops 2019-10-29T13:40:52 < Haohmaru> hm, no 2019-10-29T13:40:59 * Haohmaru calls teh Samurai Squad 2019-10-29T13:45:29 <@englishman> time for g4proing 2019-10-29T13:49:21 < Haohmaru> o hai therr breggzitman 2019-10-29T14:02:55 < jadew> speaking of countries with uncertain future, where's laurence? 2019-10-29T14:03:19 < Haohmaru> he got kicked out again 2019-10-29T14:04:01 < jadew> it's been a while 2019-10-29T14:14:27 <@englishman> no loss 2019-10-29T14:15:02 -!- mode/##stm32 [-b *!*@159.217.93.209.dyn.plus.net] by englishman 2019-10-29T14:15:50 -!- mode/##stm32 [-b *!*@14.200.208.46.dyn.plus.net] by englishman 2019-10-29T14:46:03 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [Ping timeout: 240 seconds] 2019-10-29T14:53:17 -!- jly [uid355225@gateway/web/irccloud.com/x-guekpaxcohlvamdq] has quit [Quit: Connection closed for inactivity] 2019-10-29T15:10:58 <@englishman> stm32g5 coming with cortex-m33 2019-10-29T15:15:42 < karlp> yeah, it's on their website dude. 2019-10-29T15:16:15 < karlp> oh, no, that was l5 that's m33 too. 2019-10-29T15:21:31 <@englishman> yeah 2019-10-29T15:22:34 < Steffanx> Fancy 2019-10-29T15:22:42 <@englishman> stm32g474 discovery board led is so bright it comes with a diffuser and a SAFETY WARNING not to remove it 2019-10-29T15:28:19 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-29T15:29:50 < jpa-> englishman: but is it blue? 2019-10-29T15:30:18 <@englishman> rgb 2019-10-29T15:31:03 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-29T15:32:10 < jpa-> wow 2019-10-29T15:32:12 < jpa-> such led 2019-10-29T15:40:34 < karlp> why would they do that? 2019-10-29T15:40:36 < qyx> karlp: this is for a play-pcb, so far I like murata ac/dc modules 2019-10-29T15:40:48 < dongs> why'd you get G474 disco instead of -nucleo 2019-10-29T15:40:51 < dongs> what does disco have thats exdtra? 2019-10-29T15:41:12 < qyx> the led? 2019-10-29T15:41:19 < dongs> Lol 2019-10-29T15:41:49 <@englishman> thats what they are giving out 2019-10-29T15:41:53 < dongs> oh 2019-10-29T15:41:57 < dongs> youre at that seminar thing 2019-10-29T15:42:01 <@englishman> y 2019-10-29T15:42:08 <@englishman> using all this great java soft 2019-10-29T15:42:15 < karlp> qyx: yeah, I've got a hilink module for test, but it's not suitabel for catIII and it's too big still, and _still_ wants an external X2 cap an pi filter for emc 2019-10-29T15:42:15 < dongs> yeah new cubemx is disgusting 2019-10-29T15:42:23 <@englishman> stm32cubeprogrammer too 2019-10-29T15:42:24 < dongs> new cube requires 64butt JVM for some retarded reason 2019-10-29T15:42:35 < dongs> im gonna have to find a 64bit version of win7 and make a VM with that 2019-10-29T15:42:41 < dongs> for all my cube needs 2019-10-29T15:42:50 <@englishman> the dude in front of me is on win7 2019-10-29T15:43:18 < steve> lol 2019-10-29T15:43:47 < dongs> ive been using win7 thinpc thing 2019-10-29T15:43:54 < dongs> thats like half the size installed of standard win 7 pro 2019-10-29T15:43:59 < dongs> and doesnt have any aids 2019-10-29T15:44:07 < dongs> but it looks like there's no such thing for 8 or 10 2019-10-29T15:44:15 < dongs> maybe some custom isntall of win10 IoT Enterprise 2019-10-29T15:45:04 <@englishman> at least the cubeprogrammer cli is nice 2019-10-29T15:45:11 < qyx> get rpi4 with win10 iot core 2019-10-29T15:45:19 < qyx> you can use kicad on it 2019-10-29T15:45:29 < dongs> disgusting 2019-10-29T15:45:35 <@englishman> /vote ban qyx 2019-10-29T15:46:28 < qyx> sry just laurencebing 2019-10-29T15:51:43 -!- mode/##stm32 [-o englishman] by ChanServ 2019-10-29T16:00:14 -!- learningc [~pi@121.122.92.70] has joined ##stm32 2019-10-29T16:02:17 -!- learning1 [~pi@121.121.99.187] has quit [Ping timeout: 240 seconds] 2019-10-29T16:35:57 < englishman> is j-dink supported on stlinkv3 yet 2019-10-29T16:36:01 < englishman> doesnt look like it 2019-10-29T16:38:08 < karlp> saw a segger forum thread where they said they're goign to work on it. 2019-10-29T16:42:12 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-29T16:45:09 < mawk> Steffanx you know Ibeo ? 2019-10-29T16:45:21 < mawk> the german autonomous vehicle company 2019-10-29T16:45:57 < mawk> I had a meeting with their Eindhoven branch, they want me, but I don't like the city 2019-10-29T16:45:59 < mawk> let's say if the salary is at least 50% higher than where I am now I go 2019-10-29T16:52:16 < Steffanx> Never heard of them 2019-10-29T16:52:43 < Steffanx> Eindhoven isnt that bad. I think. Unless you want delft because of the older city crap :P 2019-10-29T16:53:01 < mawk> it's a german company 2019-10-29T16:53:12 < mawk> and they made a startup offshoot in the netherlands 2019-10-29T16:53:52 < mawk> they're at the early stage so no code validation/security shit 2019-10-29T16:53:57 < mawk> so it's interesting to code 2019-10-29T16:53:58 < day> 350 ppl. 50% higher salary. kek 2019-10-29T16:54:05 < day> not.gonna.happen 2019-10-29T16:54:20 < day> except you arent making shit money right now 2019-10-29T16:54:29 < mawk> yes that's the case day 2019-10-29T16:54:34 < day> :( 2019-10-29T16:54:40 < mawk> lol 2019-10-29T16:54:41 < effractur> eindhoven is quite ok 2019-10-29T16:54:46 < effractur> lot of embedded companies 2019-10-29T16:55:09 < mawk> I like the little channels and old buildings of Delft 2019-10-29T16:55:21 < Steffanx> It reminds you of paris I guess? 2019-10-29T16:55:31 < day> pick something that has water and or hills 2019-10-29T16:55:36 < Steffanx> At least a little 2019-10-29T16:55:44 < Steffanx> Lol. Good luck with that in dutchland 2019-10-29T16:55:46 < day> eindhoven looks like wasteland 2019-10-29T16:55:58 < day> quite easy in NL :/ 2019-10-29T16:56:01 < mawk> highest hill in NL is 10m 2019-10-29T16:56:03 < day> theyve got water everywhere 2019-10-29T16:56:13 < Steffanx> Oh water OR hills :) 2019-10-29T16:56:17 < day> yeah 2019-10-29T16:56:38 < Steffanx> Fryslan it is then. Cant get more water 2019-10-29T16:56:43 < day> water and hills is the jackpot :P but good luck finding a job with those conditions 2019-10-29T16:57:15 < Steffanx> Time to move to Norway I guess 2019-10-29T16:57:50 < Steffanx> So what do you want me to say mawk? I cannot magically make you like eindhoven 2019-10-29T16:58:05 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-29T16:58:30 < day> also keep in mind the income tax in NL 2019-10-29T16:59:12 < mawk> way less than France 2019-10-29T16:59:21 < day> are you sure? 2019-10-29T16:59:34 < mawk> as independent my mother pays around 50% 2019-10-29T16:59:37 < Steffanx> It's hard to compare stuff like that. You also have to look into health insurance and stuff 2019-10-29T16:59:46 < day> NL taxes heavily from the first eur, theres no "no tax" bracket 2019-10-29T16:59:54 < Steffanx> And other crap you have to pay monthly. 2019-10-29T17:00:21 < karlp> no, the _only_ metric is the nominal income tax percentage. 2019-10-29T17:00:30 < karlp> anything else is communism, or worse, socialism. 2019-10-29T17:00:46 < karlp> you are not allowed to count other taxes, or attempt to count what benefits are/are not included. 2019-10-29T17:01:03 < day> depends 2019-10-29T17:01:10 < mawk> well just don't get sick Steffanx , why would you pay for health anyway 2019-10-29T17:01:21 < Steffanx> Its mandatory 2019-10-29T17:01:29 < mawk> ah :( 2019-10-29T17:01:36 < day> if they include the healthcare in the income tax. i have to compare it to the german income tax+ healthcare 2019-10-29T17:01:53 < Steffanx> 120+ euro/month (at least or something like that) 2019-10-29T17:03:11 < Steffanx> I meant there is more than just tax you have to pay in some/many countries. 2019-10-29T17:03:15 < Steffanx> @ karlp 2019-10-29T17:03:16 < mawk> without state health insurance and private health fund I'd pay 120€/month for my medication 2019-10-29T17:03:27 < day> 120eur/month? kek 2019-10-29T17:03:35 < day> im paying 290 :P 2019-10-29T17:03:42 < karlp> Steffanx: I wass't being serious, I thought my "anything else is communism" had covered that 2019-10-29T17:03:47 < Steffanx> With what they call "own risk" crap 2019-10-29T17:03:57 < Steffanx> Nah just clarifying 2019-10-29T17:03:58 < karlp> I can't _stand_ people comparing nominal income tax rates as if it's the only thing 2019-10-29T17:04:16 < Steffanx> Haha ok 2019-10-29T17:04:42 < mawk> with the insurance day ? 2019-10-29T17:04:52 < mawk> or you have non-covered meds you buy yourself 2019-10-29T17:05:20 < day> just the mandatory insurance. theres lots of stuff that isnt covered 2019-10-29T17:06:11 < mawk> I hope the meds are covered in NL too 2019-10-29T17:06:13 < emeb> gah - health insurance. 2019-10-29T17:06:31 < emeb> I pay $25k/year for me+wife. 2019-10-29T17:06:36 < mawk> my doctor made a letter in french because he can't type english and he expect me to show it to the NL doctor 2019-10-29T17:07:16 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Max SendQ exceeded] 2019-10-29T17:07:19 < day> emeb: o.o 2019-10-29T17:07:32 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-29T17:07:32 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-29T17:07:32 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-29T17:07:40 < emeb> and that's w/ $10k/ea deductible. 2019-10-29T17:07:42 < mawk> did you have cancer or something emeb 2019-10-29T17:07:57 < emeb> no - live in US where health costs are ridiculous 2019-10-29T17:08:00 < day> probably makes 7figures/pa 2019-10-29T17:08:11 < englishman> karlp: yeah from july 2019-10-29T17:08:27 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-29T17:08:38 < englishman> paying for health insurance? in socialist yurop? wot 2019-10-29T17:08:57 < englishman> lol emeb holy fuck 2019-10-29T17:08:59 < mawk> it's state health insurance 2019-10-29T17:09:02 < mawk> but you still pay 2019-10-29T17:09:20 < day> englishman: it s only social for the poor :< 2019-10-29T17:09:26 < mawk> for independent worker health and retirement is 24% 2019-10-29T17:09:31 < karlp> emeb: but, your income tax rates are lower! so you're winning! 2019-10-29T17:09:47 < emeb> karlp: Yay! 2019-10-29T17:10:56 < Steffanx> Mawk, the "own risk" is some amount of money (385 e) you have to pay. If you need more care/medicine/whatever they will pay. 2019-10-29T17:11:04 < Steffanx> They = the insurance company 2019-10-29T17:11:12 < mawk> that's not some amount that's a big amount 2019-10-29T17:11:29 < Steffanx> There are a few exceptions but.. meh 2019-10-29T17:11:29 < emeb> sounds like deductible 2019-10-29T17:11:45 < Steffanx> 385/year that is 2019-10-29T17:11:55 < mawk> ah 2019-10-29T17:11:59 < mawk> that's better 2019-10-29T17:12:21 < day> in germany you pay a ridiculous amount for basic healthcare, and then people drop even more money on additional healthcare (teeth, hospital upgrades - single/double bed rooms, etc.) 2019-10-29T17:12:23 < mawk> but it's not the state then 2019-10-29T17:12:32 < Steffanx> It's not the state. No 2019-10-29T17:12:56 < Steffanx> Its fucked up. Fucking commercial companies and its mandatory to have health insurance 2019-10-29T17:13:08 < mawk> lol 2019-10-29T17:13:11 < mawk> that's shady 2019-10-29T17:13:15 < Steffanx> Yes it is. 2019-10-29T17:13:28 < day> yeah fuck those commercial companies! D: 2019-10-29T17:13:28 < emeb> same here, but it's not mandatory 2019-10-29T17:13:31 < Steffanx> Those companies make a shitload of money 2019-10-29T17:13:32 < mawk> well if it's a delegation of public service with proper monitoring and accountability it's fine 2019-10-29T17:13:51 < emeb> so ppl get sick, go to hospital but can't pay $5k/day 2019-10-29T17:13:52 < Steffanx> "Proper" hah :D 2019-10-29T17:13:52 < day> im still pro mandatory healthcare. 2019-10-29T17:13:59 < Steffanx> Me too, day. 2019-10-29T17:14:03 < emeb> so hospital loads the cost onto those who can pay. 2019-10-29T17:14:07 < mawk> but often that's not how it goes, in france they privatised highways with the promise of frozen fees and upgrade, but the fee increase every year 2019-10-29T17:14:21 < Steffanx> Surprise!! 2019-10-29T17:14:42 < mawk> now they're privatising Paris airport "competition will lower prices" but the airport is the only one on a 300km radius 2019-10-29T17:14:48 < emeb> haha 2019-10-29T17:15:00 < mawk> so they're selling the monopoly to a private company, Vinci 2019-10-29T17:15:03 < emeb> just build more airports! 2019-10-29T17:15:14 < day> i thought most airports are private :/ 2019-10-29T17:15:14 < mawk> there are 5 airports but all the same entity 2019-10-29T17:15:51 < day> tbf. airports cant directly extort private people. they have to fight airlines, which will put up a proper fight 2019-10-29T17:16:21 < mawk> chinese bought Toulouse airport, now it's a disaster, they emptied the cashier 2019-10-29T17:16:21 < mawk> billions are gone to china 2019-10-29T17:16:21 < mawk> and the investor is reselling 2019-10-29T17:16:29 < emeb> wait 'till they privatize the streets in front of your house. then you have to pay a toll to open the door. 2019-10-29T17:16:51 < Steffanx> Schiphol is a private company, but still pretty much state owned. 2019-10-29T17:17:03 < mawk> public service delegation 2019-10-29T17:17:09 < mawk> but for paris it's a real privatisation 2019-10-29T17:17:25 < emeb> in US airports are usually owned & run by the city. 2019-10-29T17:17:36 < emeb> we're better socialists! 2019-10-29T17:17:38 < mawk> state sells all its shares to Vinci 2019-10-29T17:18:02 < Steffanx> I googled that. Seems like a term invented by the french. 2019-10-29T17:18:04 < mawk> and it's not like the airport has deficit, it's a cash cow 2019-10-29T17:18:16 < Steffanx> "Public service delegation" 2019-10-29T17:18:49 < day> theres quite a bit of competition here between airports tbf 2019-10-29T17:18:54 < mawk> so the opposition started a referendum on the airport, but it's an online referendum run by the government, so oddly it's full of bugs and there is absolutely no publicity of it in the media 2019-10-29T17:19:10 < day> we have a local one that is a wreck because another one ~2h away is doing much better 2019-10-29T17:19:13 < mawk> we need 300k signatures 2019-10-29T17:19:20 < Steffanx> Very cool emeb . Until you go into shutdown because the president refuses to sign some paper 2019-10-29T17:19:34 < mawk> people in Belgium drive to the paris airport day, that's how of a monopoly it is 2019-10-29T17:19:47 < englishman> last month i spent 4 days in hospital after some emergency surgery, had a view of the city from the 9th floor in a private room. cost = $6 for the prescription painkillers they gave me after i left 2019-10-29T17:19:57 < Steffanx> People in dutchland move to Germany often. So did I last time. 2019-10-29T17:20:02 < emeb> Steffanx: yes, but look at the upside - shutting down airports means less pollution. It's Green! 2019-10-29T17:20:03 < Steffanx> Drive* 2019-10-29T17:20:14 < day> sounds more like your belgium airports failed :P 2019-10-29T17:20:23 < Steffanx> emeb: airplanes aren't the problem 2019-10-29T17:20:34 < mawk> well there is a big hub in paris for whole europe 2019-10-29T17:20:35 < day> you cant really blame paris airport for that can you? 2019-10-29T17:20:57 < day> i wouldnt dream to drive to paris. i usually fly from düsseldorf 2019-10-29T17:21:05 < mawk> no, but this indicates it's a monopoly, so you can't count on concurrence to let the customers have the advantage 2019-10-29T17:21:21 * emeb remembers 9/11 when US air travel was halted for 1wk+ the skies were clear and quiet then... 2019-10-29T17:21:27 < mawk> even the airlines are against this, they know they will be screwed over the fees 2019-10-29T17:21:43 < Steffanx> emeb: I dutchland we say everything above 300ft is like "non existent". We have this NOx problem atm, but the airplanes only contribute 0.6% because of this 300ft rule. LOL 2019-10-29T17:21:46 < day> why dont they offer flights f rom a different airport? 2019-10-29T17:22:10 < mawk> all airports for the region belong to the same entity 2019-10-29T17:22:11 < emeb> Steffanx: problems eliminated by definition! 2019-10-29T17:22:20 < day> even the belgium once? o0 2019-10-29T17:22:22 < day> ones* 2019-10-29T17:22:32 < mawk> no, I guess Belgium sucks 2019-10-29T17:22:41 < mawk> I mean like 300-350km around Paris 2019-10-29T17:22:48 < day> i see 2019-10-29T17:23:12 < mawk> Belgium could drive to Amsterdam 2019-10-29T17:23:29 < mawk> Schipol airport is modern, big scanner machines 2019-10-29T17:24:03 < day> the security bullshit is probably also playing into their hands 2019-10-29T17:24:10 < day> gl. paying for that machinery if you are a small airport 2019-10-29T17:34:16 < karlp> so, two tasks that want to access spi. mutex? or a third task that handles spi and messages back and forth? 2019-10-29T17:36:30 < mawk> that depends on the peripheral I guess, are you okay with interleaving messages ? 2019-10-29T17:37:02 < qyx> I am using mutex 2019-10-29T17:37:45 < qyx> or better, I am doing it like: one SPI bus class/service, multiple SPI dev classes/services 2019-10-29T17:37:55 < qyx> or structs or whatever 2019-10-29T17:38:02 < qyx> spi bus has singhle mutex 2019-10-29T17:38:13 < qyx> spi devs have info about chipselects 2019-10-29T17:38:53 < qyx> so spi_xfer(&spidev, rx, rxn, tx, txn) handles everything 2019-10-29T17:39:33 < qyx> mawk: usually it doesn't matter what slave you use, as soon as you deassert the chipselect, the trnsaction is done for that particular slavbe 2019-10-29T17:39:39 < qyx> and you may free the mutex 2019-10-29T17:40:01 < mawk> I see 2019-10-29T17:40:49 < mawk> but then tasks are blocking waiting on the mutex, instead of queuing a message to the hypothetic third task of karlp and going on with the rest 2019-10-29T17:41:09 < mawk> that depends on the kind of work the tasks are doing 2019-10-29T17:42:07 < qyx> yes, blocking/async 2019-10-29T17:45:02 < steve> anyone use google test here? 2019-10-29T17:47:33 < mawk> once or twice 2019-10-29T17:52:49 < steve> googletest and cmake... I get a infinite loop where the CMAKE_C_COMPILER variable changes and reruns cmake. Maybe it has to do with cross-compiling in the top level CMakeLists, and host-compiling in the add_subdirectory() ? 2019-10-29T17:53:11 < mawk> sounds like a big mess 2019-10-29T17:53:20 < mawk> burn your computer and wash your hands 2019-10-29T17:54:55 < karlp> mawk: yeah, I think amutex is fine, the task doesn't really have anything else to be doing. 2019-10-29T17:57:34 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-29T17:57:51 < bitmask> man I cant decide what to do 2019-10-29T17:58:28 < bitmask> is a 3A smps enough for my project (300mA) + phone charger or should I have separate supplies 2019-10-29T18:10:23 < karlp> well, that was easy. mutex added, problem gone. 2019-10-29T18:10:28 < karlp> the magic of doing things properly. 2019-10-29T18:11:11 < karlp> steve: BuT CmAkE MaKeS EvErYtHiNg AwEsOmE..... 2019-10-29T18:12:08 < bitmask> is it 'bad' to run a smps at 10% its current rating 2019-10-29T18:12:57 < mawk> it may not have max efficiency 2019-10-29T18:13:20 < mawk> and you burn power for nothing 2019-10-29T18:13:24 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-29T18:16:10 < bitmask> fuck it, im designing it to use the same supply, if I need to use two i'll just order new pcbs 2019-10-29T18:23:12 < BrainDamage> bitmask: depends if the smps can handle dcm mode 2019-10-29T18:23:21 < bitmask> dcm? 2019-10-29T18:23:22 < BrainDamage> you may even burn the coil 2019-10-29T18:23:41 < Thorn> https://www.semtech.com/products/power-management/femtobuck-dc-dc-regulators-controllers/sc221 2019-10-29T18:23:48 < Thorn> >20MHz step-down regulator 2019-10-29T18:23:51 < bitmask> it has AAM which I think will help 2019-10-29T18:23:56 < BrainDamage> the current in a smps in the coil works like this \/\/ 2019-10-29T18:24:14 < BrainDamage> when the bottom of the wave touches 0, it's called dcm mode 2019-10-29T18:24:25 < BrainDamage> when you reach such condition, the top spikes increase a lot 2019-10-29T18:25:02 < bitmask> the datasheet im reading now says this: The AAM voltage is used to setting the transition 2019-10-29T18:25:02 < bitmask> point from AAM to CCM 2019-10-29T18:25:25 < bitmask> I should probably read up on the modes 2019-10-29T18:26:28 < BrainDamage> almost surely it won't operate in ccm if you're on 10% of nominal load, so make sure to check that 2019-10-29T18:27:02 < bitmask> would you just use separate supplies in this case or look for a smps that works in this situation 2019-10-29T18:27:16 < BrainDamage> just one that works 2019-10-29T18:27:31 < BrainDamage> you might have to also change inductor ratings a bit 2019-10-29T18:27:44 < BrainDamage> but that's it 2019-10-29T18:29:29 < bitmask> what kind of rating 2019-10-29T18:30:57 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-29T18:31:59 < BrainDamage> max current and inductance, just check the datashit 2019-10-29T18:37:41 < englishman> morning was pretty boring 2019-10-29T18:37:58 < englishman> adc have a 14 bit programmable output gain, oh wow 2019-10-29T18:38:07 < englishman> hey check out our shitty opamps acting as a pga, wow 2019-10-29T18:38:19 < englishman> corDICK and fmac are afternoon 2019-10-29T18:39:02 < englishman> the halibut with dill was amazing 2019-10-29T18:47:21 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 250 seconds] 2019-10-29T18:55:02 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-29T19:02:25 < bitmask> hmm, what usb type A female do you guys use, I was looking for the smd with pth supports for strength but those are hard to find, is it fine to use smd with 4 smd support pads? 2019-10-29T19:03:08 < aandrew> I don't think I've ever used USBA of any gender in a design 2019-10-29T19:03:09 -!- fenugrec_ [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-29T19:03:26 < aandrew> definitely agree on needing PTH for the shell, preferably at front and back 2019-10-29T19:04:11 < aandrew> bitmask: why would running an SMPS (vastly) underloaded be bad? you might get poorer regulation and shit efficiency but it won't hurt it 2019-10-29T19:04:39 < bitmask> ok just didnt know enough about it 2019-10-29T19:05:18 < bitmask> ive settled on the mp2315 as the sole supply, hopefully it can handle 2.7A occasionally 2019-10-29T19:05:49 < bitmask> actually, I doubt it will ever do 2.7, not sure what my phone charges at but I'm pretty damn sure its not 2.4A 2019-10-29T19:05:59 < bitmask> well I guess I won't have this phone forever 2019-10-29T19:14:36 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds] 2019-10-29T19:15:43 -!- fenugrec_ [~fenugrec@24.105.71.66] has quit [Ping timeout: 245 seconds] 2019-10-29T19:26:30 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-29T19:29:23 < karlp> bitmask: this is for you to plug your phone charger cable into right? 2019-10-29T19:29:29 < bitmask> yea 2019-10-29T19:29:31 < karlp> your "5v out" port? 2019-10-29T19:29:39 < karlp> yar, not much of that in our collection :) 2019-10-29T19:29:53 < karlp> we don't have a channel endorsed one at least :) 2019-10-29T19:29:58 < bitmask> heh ok 2019-10-29T19:30:01 < karlp> I'd just go ot lcsc and get one with legs :) 2019-10-29T19:31:40 < qyx> I used a molex recently 2019-10-29T19:31:40 < bitmask> yea, I guess I can just do it all pth if necessary, its not really a problem 2019-10-29T19:31:49 < qyx> pth 2019-10-29T19:32:32 < qyx> englishman: what do you think of fmac and cordick? 2019-10-29T19:32:44 < qyx> after taking the coiurse or whatever it was 2019-10-29T19:33:05 < englishman> cordick is handy it's about 6x faster than math libs. but requires fixed point 2019-10-29T19:33:31 < BrainDamage> cordic is best implemented with bitshits 2019-10-29T19:33:38 < BrainDamage> and look up tables 2019-10-29T19:33:46 < karlp> https://lcsc.com/product-detail/USB-Connectors_Jing-Extension-of-the-Electronic-Co-Jing-Extension-of-the-Electronic-Co-A-F90degree-Aparagraph-10PBTWhite-plastic6-3-Not-high-temperature_C42576.html has nice solid lugs for both front and back. hard to pull out 2019-10-29T19:33:50 < BrainDamage> it fits fpga really well 2019-10-29T19:33:53 < englishman> the fpu can do float->fixedpoint so its not so bad 2019-10-29T19:33:57 < englishman> the fmac seems really cool 2019-10-29T19:34:01 < qyx> BrainDamage: I mean hw cordic in G4 2019-10-29T19:35:43 < englishman> lots of internal memory in the fmac 2019-10-29T19:38:31 < bitmask> hmm, DHL is saying my package landed in germany and the US 2019-10-29T19:38:37 < Steffanx> Ohno 2019-10-29T19:38:41 < bitmask> both listed as destination countries 2019-10-29T19:39:01 < Steffanx> Will it be new U235 for Rob? 2019-10-29T19:39:21 < bitmask> :) 2019-10-29T19:54:26 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-29T20:00:43 -!- buZz [~buzz@unaffiliated/buzz] has joined ##stm32 2019-10-29T20:00:56 < buZz> hey, is there any DFU 101 howto or something? 2019-10-29T20:01:40 < buZz> i got a device thats capable of enabling DFU mode but i only seem to gain access to the bootloader partition of 16kb , but a replacement firmware that i want on it is over 150kb , so must be somewhere else 2019-10-29T20:01:53 < buZz> the device is a 'LG R100 360VR goggles' 2019-10-29T20:14:38 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-29T20:27:35 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-29T20:45:28 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-29T21:03:00 < aandrew> hm, libopencm3 looks nice but you'd have to layer on top of it a decent HAL 2019-10-29T21:03:12 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-29T21:03:44 < aandrew> I'm not going to spend 90% of my time |ing and &ing and <<>> shit into my code, *especially* when opencm3's entire point is to abstract things so you can choose different processors 2019-10-29T21:04:34 < englishman> just choose one processor 2019-10-29T21:05:16 < englishman> a good one 2019-10-29T21:05:54 < Steffanx> is that the point of libopencm3 aandrew? 2019-10-29T21:08:42 < englishman> qyx: yeah the fmac is cool 2019-10-29T21:08:52 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-29T21:09:24 < englishman> it'll take some learning to fully utilize I think but it makes real time multi-tap filters realistic 2019-10-29T21:11:00 < aandrew> Steffanx: yes it appears so 2019-10-29T21:11:50 < kakipro> https://www.youtube.com/watch?v=7NvhwvjZG00 2019-10-29T21:12:01 < aandrew> abstract away the "coretex M" along with some peripherals and leave you to bitmanip the rest 2019-10-29T21:12:32 < kakipro> I logged in to gmail on my ubuntu 2019-10-29T21:12:37 < kakipro> sorry mint 2019-10-29T21:12:48 < kakipro> using chromium browser 2019-10-29T21:13:34 < kakipro> it said in my google security center or whatever that "ubuntu" has full access to my google account 2019-10-29T21:14:02 < kakipro> not nice 2019-10-29T21:14:11 < bitmask> kakipro there you are 2019-10-29T21:14:36 < kakipro> did you get to it bitmask? 2019-10-29T21:14:45 < bitmask> mp2315, en pin should be 5V? the example uses Vin of 19V but that doesn't seem right 2019-10-29T21:15:15 < kakipro> have you read the datashiit 2019-10-29T21:15:22 < bitmask> yes 2019-10-29T21:15:44 < bitmask> doesnt say, it shows a voltage divider but for 19V in it says R2 is NS 2019-10-29T21:15:46 < kakipro> All Other Pins..................................-0.3V to +6V 2019-10-29T21:15:58 < Jan-> aandrew for what it is worth I would recommend doing a lot of careful tests before you pick a chip and a set of libraries. 2019-10-29T21:15:59 < bitmask> oh :) guess I didnt read it 2019-10-29T21:16:18 < kakipro> en rising threshold is parameter you want to check 2019-10-29T21:16:22 < kakipro> also en hysteresis 2019-10-29T21:16:23 < BrainDamage> kakipro: it doesn't do it automatically, you must've enabled it 2019-10-29T21:16:44 < kakipro> it did 2019-10-29T21:16:49 < kakipro> everything with default settings 2019-10-29T21:17:04 < BrainDamage> no, you need to explicitly click through in google to authorize a third party program to access your account 2019-10-29T21:17:13 < BrainDamage> you might've forgotten about it, but you did 2019-10-29T21:17:20 < kakipro> nevertheless I removed application "ubuntu" from full access to my google account 2019-10-29T21:17:40 < kakipro> now my desktop has logged out of google services 2019-10-29T21:21:04 < kakipro> I logged in to gmail now braindamage 2019-10-29T21:21:06 < kakipro> normal stuff 2019-10-29T21:21:18 < kakipro> give password no 2way auth this time 2019-10-29T21:21:27 < kakipro> I'm in gmail 2019-10-29T21:21:55 < kakipro> now ubuntu is back again in third party applications! 2019-10-29T21:22:50 < kakipro> can I trust this? 2019-10-29T21:22:53 < BrainDamage> o_o 2019-10-29T21:23:05 < BrainDamage> how are you logging in? your browser/ 2019-10-29T21:23:10 < kakipro> gmail.com 2019-10-29T21:23:20 < kakipro> it has my address there already 2019-10-29T21:23:27 < kakipro> I type my passwd 2019-10-29T21:23:31 < BrainDamage> i mean, you're using a browser 2019-10-29T21:23:38 < BrainDamage> not a desktop mail client, right? 2019-10-29T21:23:38 < kakipro> chromium 2019-10-29T21:23:41 < kakipro> yes 2019-10-29T21:23:50 < BrainDamage> check the browser extensions 2019-10-29T21:24:03 < BrainDamage> maybe there's one installed by the os that gives you 'integration' 2019-10-29T21:26:04 < kakipro> ublock origin 2019-10-29T21:27:19 < kakipro> https://askubuntu.com/questions/989359/why-has-the-ubuntu-app-full-access-to-my-google-account 2019-10-29T21:27:45 < Steffanx> kakipoop :) 2019-10-29T21:27:51 < kakipro> I don't use sync 2019-10-29T21:28:43 < kakipro> BrainDamage: nice chromium doesn't have a button to disable the thing 2019-10-29T21:28:44 < BrainDamage> ah, it's referring to the browser itself 2019-10-29T21:28:51 < kakipro> it's sign out 2019-10-29T21:29:03 < kakipro> and if I press that I expect that my gmail logs out too 2019-10-29T21:30:14 < BrainDamage> it's not that 2019-10-29T21:30:16 < BrainDamage> chrome://settings/privacy 2019-10-29T21:30:30 < BrainDamage> allow access to chromium 2019-10-29T21:30:33 < aandrew> Jan-: sure, support is as important as features 90% of the time 2019-10-29T21:31:19 < kakipro> I see 2019-10-29T21:31:31 < kakipro> I need to relaunch da browser 2019-10-29T21:31:38 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-29T21:32:08 < bitmask> kaki, sorry to keep bothering you, the rising threshold is 1.6V max, does that mean if it goes a little below that then its considered disabled? 2019-10-29T21:32:21 < bitmask> and above is enabled 2019-10-29T21:32:25 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-29T21:33:04 < kakipro> it's gone now 2019-10-29T21:33:18 < bitmask> kaki, sorry to keep bothering you, the rising threshold is 1.6V max, does that mean if it goes a little below that then its considered disabled? 2019-10-29T21:33:18 < kakipro> I can breathe once again 2019-10-29T21:33:19 < bitmask> and above is enabled 2019-10-29T21:33:59 < kakipro> your chip is enabled certainly at 1.6V 2019-10-29T21:34:04 < kakipro> that is max 2019-10-29T21:34:11 < Steffanx> déjà vu, bitmask. 2019-10-29T21:34:15 < bitmask> min is 1.2 2019-10-29T21:34:25 < kakipro> minimum is your chip may be enabled as soon as 1.2volts 2019-10-29T21:34:35 < kakipro> *as low as 2019-10-29T21:34:38 < kakipro> but not certain 2019-10-29T21:34:41 < bitmask> ok 2019-10-29T21:34:52 < bitmask> Steffanx why deja vu? 2019-10-29T21:35:08 < Steffanx> lol 2019-10-29T21:35:15 < kakipro> bitmask: did you read the section about enable pin too? 2019-10-29T21:35:17 < bitmask> glitch in the matrix? 2019-10-29T21:35:56 < kakipro> https://www.youtube.com/watch?v=XfEuxRDYiyc 2019-10-29T21:36:02 < bitmask> man, how did I miss that too 2019-10-29T21:36:06 < Steffanx> no, you said the same thing twice mr bitmask 2019-10-29T21:36:08 < bitmask> I need to take a break I think 2019-10-29T21:36:12 < bitmask> oh 2019-10-29T21:36:17 < bitmask> hah 2019-10-29T21:36:20 < bitmask> I didnt realize he left 2019-10-29T21:36:36 < mawk> are you ircing from the browser kakipro ??? 2019-10-29T21:36:43 < kakipro> like a pro 2019-10-29T21:36:44 < mawk> remove -pro from your nick immediately 2019-10-29T21:36:54 < specing> kakinoob 2019-10-29T21:37:10 < kakipro> I'm not restricted by any certain way of doing things 2019-10-29T21:37:58 < Steffanx> Received CTCP-VERSION answer from kakipro: Kiwi IRC ohno 2019-10-29T21:38:04 < Steffanx> banned 2019-10-29T21:38:14 < bitmask> oh so I dont even need the voltage divider since it has the internal zener 2019-10-29T21:38:32 < kakipro> for what exactly? 2019-10-29T21:38:39 < Steffanx> for being kaki 2019-10-29T21:39:44 < kakipro> bitmask: vbat to en? 2019-10-29T21:40:03 < bitmask> with a 100k pullup 2019-10-29T21:40:20 < bitmask> but i'll add the divider anyway, i'll make it shut off if the voltage gets too low 2019-10-29T21:40:21 < kakipro> sounds okay 2019-10-29T21:40:33 < kakipro> that sounds okay too 2019-10-29T21:40:41 < bitmask> although it doesnt sound very accurate 2019-10-29T21:40:55 < kakipro> you can get it pretty accurate 2019-10-29T21:41:07 < bitmask> yea with some testing 2019-10-29T21:41:31 < kakipro> like 30% tolerance 2019-10-29T21:41:31 < bitmask> i'll go for like 3.2V a cell or something 2019-10-29T21:42:07 < kakipro> another thing you can do: add boot button 2019-10-29T21:42:28 < kakipro> it boots up regulator and mcu 2019-10-29T21:42:34 < kakipro> mcu then drives EN pin 2019-10-29T21:43:01 < kakipro> and turns off regulator if undervoltage 2019-10-29T21:43:05 < bitmask> thats a better idea, I was thinking of something like that 2019-10-29T21:43:41 < kakipro> if mcu has comparator 2019-10-29T21:44:06 < kakipro> you can utilize it too 2019-10-29T21:44:23 < bitmask> pretty sure it does but I think the pins are used 2019-10-29T21:44:40 < kakipro> you need one analog input though 2019-10-29T21:44:47 < kakipro> from vbat divider 2019-10-29T21:45:00 < bitmask> I already have that 2019-10-29T21:45:22 < kakipro> you can also add force power off button 2019-10-29T21:45:37 < kakipro> for "development purposes" 2019-10-29T21:45:46 < kakipro> that grounds EN 2019-10-29T21:46:24 < kakipro> what resistor values you have in vbat divider? 2019-10-29T21:46:35 < bitmask> 110K and 10K 2019-10-29T21:48:01 < kakipro> what is voltage range to ADC pin 2019-10-29T21:48:15 < bitmask> I'm using the 1.1V reference 2019-10-29T21:48:41 < kakipro> well that divider wont boot the mcu then 2019-10-29T21:49:15 < kakipro> what is the discharge current of that divider? 2019-10-29T21:49:26 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-29T21:50:31 < bitmask> not sure, I'm using transistors to enable/disable it though 2019-10-29T21:50:35 -!- renn0xtk9 [~max@2a02:810d:1540:2448:14a8:6e2c:d010:9e4f] has joined ##stm32 2019-10-29T21:52:07 < kakipro> it's not absolutelly necessary 2019-10-29T21:52:16 < bitmask> I know 2019-10-29T21:52:16 < kakipro> even with as low values as 100k 2019-10-29T21:53:36 < bitmask> someone recommended it as a high side switch to prevent the mcu from getting powered by the divider or something like that 2019-10-29T21:54:19 < kakipro> if wont get powered if you set brownout level to 4.8volts 2019-10-29T21:54:29 < kakipro> and have divider voltage of 1.1v 2019-10-29T21:55:10 < kakipro> http://jahonen.kapsi.fi/Electronics/ResOptimizer/ 2019-10-29T21:56:00 < karlp> aandrew: it wasn't realllllly built as a hal at all, but "free" headers when st wasn't licensing them properly. 2019-10-29T21:56:21 < karlp> I've done some work to unify apis, but no, it's not a hal in the sense of what you're looking for. 2019-10-29T21:56:47 < karlp> there's some higher level helper routines that make it pretty portable, but only really within stm32. 2019-10-29T21:56:54 < kakipro> you should mind reset state of GPIO pins in case MCU is powered but staying in reset condition 2019-10-29T21:56:59 < kakipro> bitmask 2019-10-29T21:57:17 < karlp> I don't really have any bit |=and &= in my code at all though... 2019-10-29T21:57:54 < bitmask> ok 2019-10-29T21:58:39 < kakipro> do you know gpio pins reset state bitmask? 2019-10-29T21:59:07 < bitmask> no, i'll have to look it up 2019-10-29T21:59:35 < kakipro> it's 1) inactive or 2) weak pull-up 2019-10-29T22:01:04 < kakipro> I think some older AVR had weak pull-up but fresh model had inactive 2019-10-29T22:01:37 < karlp> in multi target code I have, it's mostly just a different main.c with clock setup and pin configs. 2019-10-29T22:04:48 < kakipro> with your 1.1v divider regulator is not booted even if there was 1.1V at all GPIO pins 2019-10-29T22:05:01 < kakipro> as regulator minimum rising threshold is 1.2V 2019-10-29T22:06:11 < bitmask> yea I know 2019-10-29T22:06:32 < kakipro> but other things like switches operating heaters may activate 2019-10-29T22:06:51 < kakipro> unless signals are conditioned with external components(resistors) 2019-10-29T22:07:01 < bitmask> they are 2019-10-29T22:08:16 < kakipro> you can also triple resistance of that +vbat divider in this purpose without any noticeable difference 2019-10-29T22:08:34 < kakipro> to measurement 2019-10-29T22:08:37 < bitmask> doesnt matter, its not even on all the time 2019-10-29T22:08:50 < kakipro> yes 2019-10-29T22:09:57 < kakipro> if you are some day going to commercialize jacket heating then it would matter to absolutely starve the design from components 2019-10-29T22:10:32 < kakipro> for a few of them, dozen of them, hundred of them, thousand of them it doesn't matter 2019-10-29T22:10:48 < bitmask> right 2019-10-29T22:26:12 < kakipro> bitmask: https://sketch.io/render/sk-1189390958843f1160955ffa18cf05d1.jpeg 2019-10-29T22:28:10 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-29T22:28:13 < kakipro> 1k resistor for MCU to knock it out quick 2019-10-29T22:28:23 < kakipro> not leave it running on capacitance 2019-10-29T22:28:43 < bitmask> whats the bottom right switch say? 2019-10-29T22:28:52 < kakipro> kakicad 0.1 2019-10-29T22:28:57 < bitmask> above htat 2019-10-29T22:29:04 < kakipro> force off 2019-10-29T22:29:06 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-29T22:29:09 < bitmask> ohh 2019-10-29T22:29:27 < kakipro> zener size of maybe.. around 4volts should do 2019-10-29T22:29:39 < kakipro> there is some current from MCU 2019-10-29T22:30:15 < kakipro> less the closer you get to 5v with that zener 2019-10-29T22:30:46 < kakipro> it's there in case there is no load in 5volt line and you keep boot pressed 2019-10-29T22:31:09 < kakipro> en will rise to 6.5v and 5v line will to 2019-10-29T22:31:10 < kakipro> too 2019-10-29T22:31:47 < kakipro> AVR can take 6.5volts but it's not recommended 2019-10-29T22:32:59 < kakipro> another option is to replace zener with pulldown resistor that results to EN voltage around 5volts at maximum vbat 2019-10-29T22:33:18 < kakipro> oh wait 2019-10-29T22:33:24 < kakipro> this doesn't work 2019-10-29T22:33:51 < kakipro> you press boot and MCU will draw all the current from EN line 2019-10-29T22:35:05 < bitmask> I think I"m just gonna use a stm1061 to control the gates of the battery disconnect mosfets 2019-10-29T22:36:24 < bitmask> disable everything 2019-10-29T22:36:35 < kakipro> it will work 2019-10-29T22:37:02 < kakipro> you get total power off then 2019-10-29T22:37:16 < bitmask> which will save the batteries which is the main goal 2019-10-29T22:38:39 < kakipro> relying on software on something like that is a bit nasty 2019-10-29T22:39:43 < bitmask> I'm still gonna use software to monitor the batteries and maybe limit or cut off the heaters at a certain point but leave the system running. 2019-10-29T22:40:11 < bitmask> but then if it gets too low just have it all shut off 2019-10-29T22:40:23 * kakipro nods 2019-10-29T22:41:05 < qyx> whats stm1061 2019-10-29T22:41:32 < bitmask> low power voltage detector 2019-10-29T22:41:32 < kakipro> some voltage monitor 2019-10-29T22:41:40 < bitmask> precision voltage reference with comparator in one 2019-10-29T22:41:52 < kakipro> all in sot23 2019-10-29T22:42:00 < qyx> why not using stm32 directly? 2019-10-29T22:42:12 < kakipro> *laughters* 2019-10-29T22:42:13 < aandrew> karlp: aha, thank you for clarifiying 2019-10-29T22:42:15 < bitmask> haha 2019-10-29T22:42:19 < qyx> if you plan t rely on software 2019-10-29T22:42:24 < aandrew> it certainly looks like it would be a great starting point for a unified HAL across all devices 2019-10-29T22:42:32 < bitmask> no one said they planned on relying on software 2019-10-29T22:42:53 < qyx> oh 2019-10-29T22:43:18 < kakipro> qyx: he just want to add hard safety nobody should have problem with that 2019-10-29T22:43:58 < qyx> then there are dedicated battery protectors 2019-10-29T22:44:05 < qyx> with gate drivers and all that stuff 2019-10-29T22:44:17 < kakipro> and individual cell monitor 2019-10-29T22:44:54 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-29T22:45:35 < antto> i can protect your moniez 2019-10-29T22:45:37 < kakipro> but I think if bitmask adds that project is not moving 2019-10-29T22:45:39 < antto> physically 2019-10-29T22:45:54 < kakipro> and I think he will have separate charger 2019-10-29T22:46:03 < bitmask> I'm using a cheap china BMS already, just don't know exactly what cutoff voltage it uses 2019-10-29T22:46:17 < kakipro> so you have triple safety? 2019-10-29T22:46:27 < kakipro> doubled hard safety and soft safety? 2019-10-29T22:46:33 < bitmask> oh yea I guess haha 2019-10-29T22:46:53 < bitmask> maybe I can find what voltage this uses 2019-10-29T22:47:12 < kakipro> overkill alert 2019-10-29T22:47:26 < kakipro> show the BMS? 2019-10-29T22:47:46 < kakipro> lipos are not so fragile you think 2019-10-29T22:47:49 < bitmask> Over discharge voltage range: 2.3~3.0V ± 0.05V 2019-10-29T22:48:13 < bitmask> I think the IC comes in different versions and it doesnt tell which it uses 2019-10-29T22:48:20 < bitmask> and those voltages are pretty low 2019-10-29T22:48:35 -!- friendofafriend [~ian@75.182.67.149] has joined ##stm32 2019-10-29T22:48:51 < kakipro> why to even have that BMS then? for other protection functionality? 2019-10-29T22:49:22 < bitmask> ehh it does say 3V in the q&a 2019-10-29T22:49:31 < bitmask> yea and balance charging 2019-10-29T22:49:52 < bitmask> ok i'll just use this bms 2019-10-29T22:49:54 < bitmask> should be fine 2019-10-29T22:50:07 < kakipro> yes 2019-10-29T22:50:23 < karlp> aandrew: it's enough work looking after it as it is, rescoping to try and do a serious hal seems like a somewhat excessive, and also probably unwanted chunk of development. 2019-10-29T22:50:38 < karlp> there's definitely room for a clean layer of it, but I sure don't have time to work on it :) 2019-10-29T22:51:19 < karlp> (but I do try and make all the peripheral code consistent and unified wherever possible) 2019-10-29T22:51:23 < kakipro> chinese protection module + soft safety is 100% safety 2019-10-29T22:51:45 < kakipro> 50% + 50% 2019-10-29T22:52:29 < kakipro> videogayms> 2019-10-29T22:53:55 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-dyohdytgyvzhlemr] has quit [Quit: Connection closed for inactivity] 2019-10-29T22:53:58 < aandrew> karlp: I wasn't trying to assign work to you, I haven't got the time nor patience to run an OSS library so I already have considerable respect for those who do 2019-10-29T22:54:54 < aandrew> lol @ resistor optimizer .exe 2019-10-29T22:55:04 < aandrew> I just have a spreadsheet from TI I think it is 2019-10-29T22:55:09 < aandrew> there are a few out there 2019-10-29T22:55:48 < BrainDamage> for 99% of the tasks, the value of the resistor is not critical 2019-10-29T22:56:04 < BrainDamage> and for the remaining 1%, you buy high precision resistors in the first place 2019-10-29T22:56:45 < aandrew> almost all my resistors these days are 1% as there's no cost difference 2019-10-29T22:56:52 < aandrew> I certainly don't need that accurate but why not 2019-10-29T22:57:10 < aandrew> I do have some 0.1% for one or two very special occasions but that's also rare 2019-10-29T22:57:43 < aandrew> even at 0.1% you're fooling yourself about accuracy if you're not looking at tempco and dift 2019-10-29T22:58:03 < BrainDamage> and you need to be careful about EMI as well 2019-10-29T23:01:03 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-qnfuwgwwhruqzped] has joined ##stm32 2019-10-29T23:20:13 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-29T23:24:24 -!- catphish [~catphish@unaffiliated/catphish] has quit [Remote host closed the connection] 2019-10-29T23:37:30 -!- renn0xtk9 [~max@2a02:810d:1540:2448:14a8:6e2c:d010:9e4f] has quit [Quit: Konversation terminated!] 2019-10-29T23:46:05 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-29T23:49:20 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 --- Day changed Wed Oct 30 2019 2019-10-30T00:10:42 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-30T00:33:05 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-30T00:36:18 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T00:58:38 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has quit [Remote host closed the connection] 2019-10-30T01:03:54 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-qnfuwgwwhruqzped] has quit [Quit: Connection closed for inactivity] 2019-10-30T01:05:08 -!- futarisIRCcloud [uid222239@gateway/web/irccloud.com/x-sfynlrhigkrdtysm] has joined ##stm32 2019-10-30T01:10:33 -!- con3 [~kvirc@154.119.40.237] has quit [Read error: Connection reset by peer] 2019-10-30T01:18:53 < kakipro> no source in elf file 2019-10-30T01:19:14 < kakipro> ozone was just so smart it finds the source file and shows it 2019-10-30T01:24:11 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T01:29:59 < kakipro> is there like "2 direction" switching regulators? 2019-10-30T01:30:22 < kakipro> let's say I have energy solution and I want to charge battery from DC bus 2019-10-30T01:30:46 < con3> you mean bidirectional dcdc converters? 2019-10-30T01:30:55 < kakipro> but then other source of DC bus power goes down 2019-10-30T01:31:04 < kakipro> and I want to drive DC bus from battery 2019-10-30T01:31:33 < con3> yep kakipro their called bidirectional dcdc converters if i understand what you mean 2019-10-30T01:32:42 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-30T01:36:48 < kakipro> not cheap 2019-10-30T01:41:02 -!- fenugrec_ [~fenugrec@104.221.51.116] has joined ##stm32 2019-10-30T01:41:45 < englishman> kakipro: yes plenty 2019-10-30T01:42:14 < englishman> bq25703a 2019-10-30T01:44:29 -!- mirage335 [~mirage335@2001:470:8ede:0:216:3eff:fe97:ac6d] has joined ##stm32 2019-10-30T01:49:42 < kakipro> I don't think so englishman 2019-10-30T01:49:51 < karlp> aandrew: nah, I know, just trying to say a bit more abotu what it is, and what it could be, but isn't likely to ever be :) 2019-10-30T01:49:59 < kakipro> that is buckboost charger 2019-10-30T01:50:19 < karlp> given the bsd license on the header files now, the whole project is kinda..... "so you don't like structs and capitals? and you like license complications? have we got the project for you!" 2019-10-30T01:50:56 < karlp> ("the" in this case referring to the st cmsis headers) 2019-10-30T01:51:55 < aandrew> buckboost just regulates higher and lower than input 2019-10-30T01:52:00 < kakipro> englishman: did you get part id right? 2019-10-30T01:52:05 < aandrew> that doesn't mean regenerative capabilities 2019-10-30T01:52:28 < aandrew> what is complicated about bsd license? 2019-10-30T01:52:37 < englishman> kakipro: idk 2019-10-30T01:53:28 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Quit: Leaving] 2019-10-30T01:57:40 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 264 seconds] 2019-10-30T01:59:41 < kakipro> is there any more technical term for bidirectional? 2019-10-30T01:59:50 < kakipro> that could be used as search parameter 2019-10-30T02:00:17 < antto> bisexual 2019-10-30T02:00:19 < antto> oh wait 2019-10-30T02:00:39 < kakipro> yes 2019-10-30T02:01:45 < mawk> duplex kakipro 2019-10-30T02:02:27 < antto> due direzioni 2019-10-30T02:02:30 < antto> ;P~ 2019-10-30T02:03:04 < antto> la pizza e calda! 2019-10-30T02:03:38 < con3> actually not sure, i just know from journals their referred to as bidirectional. 2019-10-30T02:27:02 < kakipro> it has 4switch topology 2019-10-30T02:27:35 < kakipro> like 4switch buck-boost 2019-10-30T02:28:03 < antto> quatro-switcheroni 2019-10-30T02:32:22 < specing> 2 bucks 1 choke 2019-10-30T02:32:38 < kakipro> maybe there is a way to trick buck-boost converter to do bidir 2019-10-30T02:32:51 < kakipro> there needs to be signal crossing though 2019-10-30T02:34:59 * antto adds two diodes that look at each other 2019-10-30T02:37:28 < kakipro> https://www.ti.com/lit/ug/tidt046/tidt046.pdf 2019-10-30T02:37:42 < kakipro> the controller is over 10eur 2019-10-30T02:38:28 < specing> why not just connect two bucks in reverse? 2019-10-30T02:38:38 < antto> well it's by TEGGSAS 2019-10-30T02:39:38 < kakipro> specing: where is the boost then? 2019-10-30T02:41:40 < kakipro> http://www.ti.com/lit/ds/symlink/ucd3138.pdf it's a beast though 2019-10-30T02:41:47 < antto> libboost <- here it is 2019-10-30T02:41:49 < kakipro> for 10eur 2019-10-30T02:42:22 < specing> kakipro: you need boost as well? 2019-10-30T02:42:28 < jadew> another customer asked for schematic... should I make a note on the product page (not opensource)? 2019-10-30T02:42:29 < antto> if that's a beast, then u must b teh beauty 2019-10-30T02:42:53 < jadew> is there a not-opensource logo? 2019-10-30T02:42:59 < specing> lol 2019-10-30T02:43:29 < antto> jadew put the opensource logo in a white circle with red outline, and a diagonal red line 2019-10-30T02:43:57 < jadew> that sounds off-putting 2019-10-30T02:44:05 < kakipro> designed by dongs 2019-10-30T02:44:25 < antto> no one said reality is all sunshine and roses 2019-10-30T02:44:36 < jadew> I'll use 5pt font and put it in the footer 2019-10-30T02:44:49 < antto> "dongsware" 2019-10-30T02:44:59 < jadew> it has to be visible 2019-10-30T02:45:05 < jadew> or clear in some way 2019-10-30T02:45:21 < antto> make a logo 2019-10-30T02:45:29 < antto> "this project is bare gerborz" 2019-10-30T02:45:39 < jadew> I wonder how many tried to open the shielding can and broke the PCB... 2019-10-30T02:45:54 < jadew> maybe that's why they didn't answer to my followup email several months later 2019-10-30T02:46:11 < antto> and wut did u hide inside? 2019-10-30T02:46:34 < jadew> circuits with components that had their markings removed 2019-10-30T02:46:45 < antto> u bastard 2019-10-30T02:47:14 < jadew> I don't like it either, but it's a difficult to design device that is super easy to clone 2019-10-30T02:47:31 < antto> pics or it didn't happen ;P~ 2019-10-30T02:47:40 < jadew> of what? 2019-10-30T02:47:41 < kakipro> https://www.youtube.com/watch?v=qbWfvaqnneU movie I have not seen yet!" 2019-10-30T02:48:13 < kakipro> maybe I start software project 2019-10-30T02:48:16 < antto> nuffin, i'm kidding 2019-10-30T02:48:18 < kakipro> kakirouter 2019-10-30T02:48:25 < jadew> video unavailable 2019-10-30T02:48:27 < kakipro> the smartest autorouter in world 2019-10-30T02:49:15 < kakipro> but is this men in black anymore 2019-10-30T02:49:28 < kakipro> this looks like those superhero movie series things 2019-10-30T02:52:39 < antto> mv kakipro kakiman 2019-10-30T02:52:55 < antto> ur liek a superhero now 2019-10-30T02:54:07 < kakipro> $su kakinull 2019-10-30T02:55:47 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-30T02:58:59 < kakinull> https://www.imdb.com/title/tt2283336/ rating is right in the sweet spot 2019-10-30T03:01:43 < aandrew> 5.6 is a sweet spot? 2019-10-30T03:01:54 < aandrew> in my experience any movie 7.0 or higher with more than 50k votes is worth the watch 2019-10-30T03:02:27 < kakinull> too posh 2019-10-30T03:03:57 < kakinull> potentially 2019-10-30T03:05:11 < kakinull> I think I might stop trying to pursue career in electronics 2019-10-30T03:05:25 < kakinull> move to software side 2019-10-30T03:05:57 < kakinull> so much logistics is involved in electronics 2019-10-30T03:06:30 < kakinull> software is just computer or computer and target 2019-10-30T03:07:10 < kakinull> *workstation(computer being too casual term) 2019-10-30T03:12:25 < kakinull> besides electronics require math and professional skills 2019-10-30T03:12:34 < jadew> I want to see a movie too 2019-10-30T03:12:40 < jadew> ah, I've seen that one 2019-10-30T03:12:47 < jadew> do not remember a thing 2019-10-30T03:12:50 < jadew> just the chick 2019-10-30T03:13:12 < kakinull> ? 2019-10-30T03:14:06 < bitmask> kakinull have you seen this shit? https://www.monolithicpower.com/en/design-tools/design-tools/dc-dc-designer-online.html 2019-10-30T03:14:08 < jadew> that's how memorable the movie is 2019-10-30T03:14:33 < jadew> oh, I know how to end this night 2019-10-30T03:14:40 < jadew> movie and icecream 2019-10-30T03:14:42 < kakinull> no bitmask 2019-10-30T03:14:44 < jadew> and beer! 2019-10-30T03:14:49 < kakinull> i havenot 2019-10-30T03:14:52 < bitmask> pretty cool 2019-10-30T03:15:24 < jadew> bitmask, most manufacturers have something like that 2019-10-30T03:15:33 < bitmask> fine, be that way :P 2019-10-30T03:15:38 < jadew> :) 2019-10-30T03:15:47 < jadew> ok, it's very cool :P 2019-10-30T03:16:18 < jadew> which it is, btw, I'm just saying it's not uncommon :) 2019-10-30T03:17:02 < bitmask> I wonder why the values differ from the datasheet though 2019-10-30T03:17:11 < bitmask> always complicating shit 2019-10-30T03:17:22 < jadew> someone made a typo 2019-10-30T03:17:44 < jadew> or you get ideal values somewhere, and standard values somewhere else 2019-10-30T03:18:27 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has quit [Ping timeout: 240 seconds] 2019-10-30T03:20:20 < jadew> anyone seen rattlesnake? 2019-10-30T03:21:22 < bitmask> kaki it even has uvlo 2019-10-30T03:22:56 < bitmask> oh maybe not what I wanted but doesnt matter 2019-10-30T03:22:56 < bitmask> oh maybe not what I wanted but doesnt matter 2019-10-30T03:24:00 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-30T03:24:37 < kakinull> bitmask: using en pin? 2019-10-30T03:24:56 < kakinull> bitmask: using en pin? 2019-10-30T03:25:02 < bitmask> yea 2019-10-30T03:25:10 < bitmask> it just does the divider for you 2019-10-30T03:26:16 < kakinull> input and output capacitors you must select package that has availibility with your capacitor parameters 2019-10-30T03:27:58 < kakinull> not just few 2019-10-30T03:28:12 < kakinull> those are some weird packages usually or mislabeled if you get only few 2019-10-30T03:28:21 < kakinull> you need to see a ton of parts in parametric search 2019-10-30T03:29:10 -!- fenugrec_ [~fenugrec@104.221.51.116] has quit [Remote host closed the connection] 2019-10-30T03:32:06 < aandrew> hm interesting 2019-10-30T03:32:14 < aandrew> I'm telnetted to an stm32 running libtelnet 2019-10-30T03:32:51 < aandrew> from osx when I telnet to it and I backspace, I see ^? on my local screen but the remote never sees the character (proved with tcpdump) so the ^? is happening on osx 2019-10-30T03:32:54 < bitmask> hmm, the datasheet says input of 22uF, output 44uF, the website says input 3.3uF, output 22uF, thats a pretty big difference 2019-10-30T03:33:07 < aandrew> from win10 the delte key does delete without ^? 2019-10-30T03:33:16 < bitmask> oh sorry 6.6 input 2019-10-30T03:33:18 < aandrew> I thought this was a problem on the stm32 side but it's an osx telnet issue 2019-10-30T03:33:56 < kakinull> bitmask: my design had nominal 10 and 40 2019-10-30T03:34:00 -!- Cracki [~cracki@unaffiliated/crackwitz] has quit [Quit: Leaving] 2019-10-30T03:34:12 < kakinull> measured values were higher 2019-10-30T03:34:37 < kakinull> maybe it was 10 and 40 i don't remember 2019-10-30T03:36:39 < bitmask> yea I think i'll go with 2x4.7 and 2x22 2019-10-30T03:37:15 < aandrew> hm what's the difference between ^? and ^H 2019-10-30T03:37:17 < kakinull> https://drive.google.com/file/d/1UtzO0Mfh0sXrRx83bBBy8g1HPXT2EAwf/view?usp=sharing 2 caps next to usb datapins are output capacitors 2019-10-30T03:37:26 < kakinull> 22 x 2 2019-10-30T03:38:54 < kakinull> those big caps are like 100n 2019-10-30T03:39:11 < kakinull> big package only for optimizing routing 2019-10-30T03:39:33 < kakinull> I think I threw that same 22 cap to input too 2019-10-30T03:39:35 < kakinull> but only 1 2019-10-30T03:39:50 < kakinull> wait no those caps at output have only 10volt rating 2019-10-30T03:40:05 < kakinull> input cap was one 10u maybe 2019-10-30T03:40:33 < bitmask> makes sense 2019-10-30T03:41:04 < bitmask> is it better to use two to get lower esr? 2019-10-30T03:41:27 < kakinull> esr is a factor 2019-10-30T03:41:35 < kakinull> maybe 2019-10-30T03:41:37 < jadew> 1000Z scope with two channels and 200 MHz bw: https://beyondmeasure.rigoltech.com/acton/attachment/1579/f-f55c8579-4293-4d7f-aec4-ef7ea08ad8dc/1/-/-/-/-/DS1000Z-E%20Datasheet-EN.pdf 2019-10-30T03:41:56 < kakinull> depends if those 2 have smaller ESR than one 2019-10-30T03:42:14 < bitmask> wouldnt it always? 2019-10-30T03:42:15 < kakinull> assuming using different cap 2019-10-30T03:42:18 < bitmask> oh 2019-10-30T03:42:28 < kakinull> if same cap then sure 2019-10-30T03:43:02 < kakinull> you see what I meant by size of my 2 thumb nails? 2019-10-30T03:43:16 < bitmask> I already ordered 10uF/25V and 22uF/10V so I may as well use the same as you 2019-10-30T03:43:23 < bitmask> yup 2019-10-30T03:43:38 < bitmask> well I ordered 1206, are yours 0805? 2019-10-30T03:43:53 < bitmask> or do they come in min 1206 2019-10-30T03:44:09 < bitmask> they looks small but its hard to tell 2019-10-30T03:44:28 < kakinull> my output caps look like 0603 2019-10-30T03:44:47 < kakinull> 22u/10V 2019-10-30T03:44:48 < bitmask> I was gonna say they even look smaller than 0805 but I didnt think they came that small 2019-10-30T03:44:49 < kakinull> maybe 2019-10-30T03:44:59 < kakinull> I didn't either 2019-10-30T03:46:31 < kakinull> I had hard time to believe it when I measured them 2019-10-30T03:47:12 < kakinull> input is 0805 2019-10-30T03:47:38 < kakinull> 10u / 25v iirc 2019-10-30T03:47:58 -!- Cracki [~cracki@unaffiliated/crackwitz] has joined ##stm32 2019-10-30T03:48:00 < kakinull> coil is 4.7uH 2019-10-30T03:51:08 < bitmask> Im going with 6.8 2019-10-30T03:51:59 < bitmask> does it make a difference? 2019-10-30T03:53:52 < bitmask> did you use a fancy compensation circuit? 2019-10-30T03:55:39 < kakinull> no 2019-10-30T03:55:52 < kakinull> I added bootstrap diode and capacitor 2019-10-30T03:55:57 < kakinull> and some other capacitor 2019-10-30T03:56:50 < kakinull> to make it more efficient 2019-10-30T03:57:07 < kakinull> you don't need to as you have all the copper in world to conduct the heat 2019-10-30T03:57:55 < kakinull> SRU8043-4R7Y coil I used 2019-10-30T03:58:38 < kakinull> 17milliohms 2019-10-30T03:59:13 < kakinull> within same package size if your inductance goes up your saturation current goes down 2019-10-30T04:02:51 < kakinull> I saw my china contacts going way bellow the reference design inductor inductance finding the smallest reasonable well working inductance > smaller inductor package + cheaper while keeping low resistance and high saturation current 2019-10-30T04:03:54 < kakinull> reference design is sure-fire thing 2019-10-30T04:04:43 < kakinull> you have all the space also so just throw some generic 10mm or bigger coil there 2019-10-30T04:06:46 < kakinull> you ofc could add NC footprints for bootstrap diode and cap in case you run into problems bitmask 2019-10-30T04:07:33 < kakinull> your thermal design not working or using lower than intended input voltages etcetc 2019-10-30T04:14:35 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T04:25:01 < englishman> wtf kakinull 2019-10-30T04:25:03 < englishman> dont watch that movie 2019-10-30T04:25:39 < kakinull> i didnt 2019-10-30T04:35:35 < qyx> kakinull: 22u/10V in 0603? 2019-10-30T04:35:45 < qyx> whats that, Y5V? 2019-10-30T04:35:54 < kakinull> probs 2019-10-30T04:36:03 < kakinull> not X7R certainly 2019-10-30T04:36:10 < qyx> are you expecting to get more than 1u on them when dc biased? 2019-10-30T04:40:21 < kakinull> jeez that derating is massive 2019-10-30T04:40:44 < kakinull> 70% or so 2019-10-30T04:40:55 < bitmask> is ESR lower on larger mlcc packages? 2019-10-30T04:41:16 < kakinull> bitmask: and always remember to use at least x5r 2019-10-30T04:41:25 < bitmask> yea, definitely 2019-10-30T04:41:25 < kakinull> or x7r 2019-10-30T04:41:38 < bitmask> c0g for everything :P 2019-10-30T04:42:12 < kakinull> qyx: well there is 2 more 0603 footprints free 2019-10-30T04:42:26 < kakinull> in case of problems 2019-10-30T05:09:51 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away] 2019-10-30T05:12:02 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32 2019-10-30T05:14:22 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-30T05:15:13 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T05:17:02 -!- con3 [~kvirc@154.119.40.237] has quit [Max SendQ exceeded] 2019-10-30T05:17:29 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T05:20:03 -!- con3 [~kvirc@154.119.40.237] has quit [Client Quit] 2019-10-30T05:38:33 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-30T06:00:14 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-30T06:16:05 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-30T06:16:14 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-30T06:38:09 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-30T06:45:12 < machinehum> After watching mr robot a couple of nights ago I went looking for #sex on freenode 2019-10-30T06:45:19 < machinehum> Doesn't exist yo 2019-10-30T06:48:06 -!- fc5dc9d4_ [~quassel@p57A3217C.dip0.t-ipconnect.de] has joined ##stm32 2019-10-30T06:51:37 -!- fc5dc9d4 [~quassel@p57A32E83.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds] 2019-10-30T06:54:35 < machinehum> no weiners huh 2019-10-30T06:54:58 < machinehum> turnip420: wbu 2019-10-30T07:12:40 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-30T07:15:43 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 245 seconds] 2019-10-30T07:15:43 -!- day__ is now known as day 2019-10-30T07:36:57 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-30T07:37:37 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-30T07:45:44 < BrainDamage> kakinull: https://i.imgur.com/Hd488Tf.png 2019-10-30T07:46:03 < BrainDamage> this should make a buckboost converter work in reverse 2019-10-30T07:46:24 < BrainDamage> you need explicitly one that has a complementary driver to replace the diode 2019-10-30T07:46:40 < BrainDamage> then you swap the sense voltage pin 2019-10-30T07:46:59 < BrainDamage> also need a cap at the input 2019-10-30T07:47:12 < BrainDamage> unless I fucked up the circuit analysis, the vo/vi eq in ccm mode is the same 2019-10-30T07:48:04 < BrainDamage> so the circuit will start regulating the input instead of the output 2019-10-30T07:49:46 < BrainDamage> I might actually try that insanity to regen brake my bike 2019-10-30T08:09:19 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-30T08:19:38 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-30T08:30:35 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-30T08:34:38 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-30T08:35:23 -!- dongs [~dongs@bcas.tv] has quit [Quit: upgrading] 2019-10-30T08:38:18 -!- dongs [~dongs@bcas.tv] has joined ##stm32 2019-10-30T08:47:05 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-30T08:47:39 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-30T08:48:22 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-30T08:51:32 -!- jly [uid355225@gateway/web/irccloud.com/x-hilatfbmfmkwjaur] has joined ##stm32 2019-10-30T09:05:26 < Steffanx> Hello jly 2019-10-30T09:11:28 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-30T09:13:22 < jly> is that a timer? 2019-10-30T09:13:28 < jly> Hi stffn 2019-10-30T09:32:42 < jly> didn't tell me how his cat is going 2019-10-30T09:33:52 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-30T09:54:06 -!- sterna [~Adium@h-158-174-145-238.NA.cust.bahnhof.se] has joined ##stm32 2019-10-30T10:04:54 -!- tprrt [~tprrt@217.114.204.178] has joined ##stm32 2019-10-30T10:14:03 -!- ThatDamnRanga [~ThatDamnR@unaffiliated/wiretap] has quit [Ping timeout: 245 seconds] 2019-10-30T10:21:16 -!- jon101249 [~jon1012@lns-bzn-40-82-251-162-219.adsl.proxad.net] has joined ##stm32 2019-10-30T10:44:53 -!- ThatDamnRanga [~ThatDamnR@unaffiliated/wiretap] has joined ##stm32 2019-10-30T10:58:36 < jpa-> why are all vendor libraries this ridiculous 2019-10-30T10:59:05 < jpa-> on nRF52, setting for interrupt sources is trivial, as they are all in their own registers: "if (NRF_RTC1->EVENTS_TICK)" 2019-10-30T10:59:58 < jpa-> but the official code goes: drv_rtc_tick_pending(rtc_instance) -> evt_pending(p_instance, NRF_RTC_EVENT_TICK) -> nrf_rtc_event_pending(p_instance->p_reg, event) -> finally read the actual register 2019-10-30T11:11:18 < effractur> [A[B[A[B 2019-10-30T11:33:24 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has quit [Quit: ZNC 1.6.1 - http://znc.in] 2019-10-30T11:33:42 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has joined ##stm32 2019-10-30T11:33:42 -!- MrFahrenheit [~zumbi@ihaveahuge.wang] has quit [Changing host] 2019-10-30T11:33:42 -!- MrFahrenheit [~zumbi@unaffiliated/rageofthou] has joined ##stm32 2019-10-30T11:33:53 < Steffanx> Ty effractur 2019-10-30T11:34:34 < Steffanx> You should join the complain about nrf crew in #embedded, jpa- 2019-10-30T11:34:45 < Steffanx> ##embedded * 2019-10-30T11:38:11 < jpa-> oh no, more channels 2019-10-30T11:41:26 < karlp> aandrew: if "what is complicated about bsd license" was for me, then nothing, but originally, the ST headers _weren't_ bsd licensed. 2019-10-30T11:43:03 < karlp> jadew: everything about "it's a difficult to design device that is super easy to clone" just screams, "this is not a viable product to be trying to sell" 2019-10-30T11:51:03 < jpa-> or perhaps, just sell the complement 2019-10-30T11:51:59 < jpa-> such as PC software to go with it, because weirdly software seems less prone to copying than hardware now 2019-10-30T11:57:05 < karlp> well it's his plugin module for obsolete hardware anyway. so... :) 2019-10-30T12:02:25 < karlp> BrainDamage: let me know if it works: https://github.com/karlp/zypsnips/blob/master/buckboost-reverse.md 2019-10-30T12:05:54 < jpa-> looks like altium schematic 2019-10-30T12:18:38 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-30T12:31:52 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-30T12:40:41 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-30T12:50:26 -!- jly [uid355225@gateway/web/irccloud.com/x-hilatfbmfmkwjaur] has quit [Quit: Connection closed for inactivity] 2019-10-30T13:06:51 < jadew> karlp, I know that now, but then, that's the case with most electronics 2019-10-30T13:09:44 < jpa-> jadew: has anything you've made actually ever been cloned? 2019-10-30T13:09:53 < jadew> jpa-, yes 2019-10-30T13:10:05 < jadew> not electronics tho 2019-10-30T13:10:24 < jadew> software that has been copied 2019-10-30T13:10:40 < jadew> but these devices are not without competition 2019-10-30T13:11:01 < jadew> they have competition from china, and those are direct HP clones 2019-10-30T13:11:49 < jadew> I just don't want to feed them any key information 2019-10-30T13:12:59 < jadew> I've seen the designs that come out of china, and they're struggling in various areas 2019-10-30T13:15:55 < jadew> karlp is right tho, not the best product to sell 2019-10-30T13:16:31 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Ping timeout: 260 seconds] 2019-10-30T13:20:17 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-30T13:21:37 -!- sterna [~Adium@h-158-174-145-238.NA.cust.bahnhof.se] has quit [Ping timeout: 240 seconds] 2019-10-30T13:22:03 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-30T13:24:18 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-30T14:00:50 -!- fsasm [~fsasm@62-178-93-7.cable.dynamic.surfer.at] has joined ##stm32 2019-10-30T14:22:39 -!- ac_slater [~weechat@144.121.38.133] has joined ##stm32 2019-10-30T14:23:02 < ac_slater> hey guys, I have a STM32F413xH and I cannot for the life of me find the register mappings to things like SPI 2019-10-30T14:23:10 < ac_slater> I see the datasheet, but that has no registers 2019-10-30T14:27:22 < effractur> ac_slater: table 13? 2019-10-30T14:28:06 < ac_slater> effractur: interesting 2019-10-30T14:28:08 < ac_slater> this is useful 2019-10-30T14:28:34 < ac_slater> but doesn't really show what each register does. I found "reference manuals" for other STM32 chips that have SPI/I2C state machines and such and outline how to configure them 2019-10-30T14:29:02 < ac_slater> typically this data is in the datasheet, but it seems external 2019-10-30T14:30:05 < effractur> the spi docs are 2019-10-30T14:30:06 < effractur> https://www.st.com/content/ccc/resource/technical/document/application_note/group0/b0/7e/46/a8/5e/c1/48/01/DM00227538/files/DM00227538.pdf/jcr:content/translations/en.DM00227538.pdf 2019-10-30T14:31:14 < ac_slater> oh "all" STM32 have the same SPI circuitry? 2019-10-30T14:33:05 < ac_slater> what about I2C and other register configurations? :-/ 2019-10-30T14:33:45 < effractur> easiest is i think just to use for example STM32CubeMX 2019-10-30T14:33:50 < effractur> to generate configureations 2019-10-30T14:33:56 < ac_slater> ah I found it https://www.st.com/content/ccc/resource/technical/document/reference_manual/group0/81/ea/88/1f/97/9e/4a/d0/DM00305666/files/DM00305666.pdf/jcr:content/translations/en.DM00305666.pdf 2019-10-30T14:34:37 < ac_slater> effractur: I get that. I haven't had CubeMX spit out LL code yet, just HAL code. I need to adjust SPI timing registers for my slaves and that's not available in cubeMX 2019-10-30T14:41:35 < ac_slater> thanks effractur 2019-10-30T14:42:57 < karlp> ac_slater: you're looking for the reference manual, not the datasheet. 2019-10-30T14:43:29 < ac_slater> right, I said above I found other reference manuals but couldnt find the STM32F413xH since it's not listed on the documentation page 2019-10-30T14:43:34 < karlp> try harder? 2019-10-30T14:43:53 < ac_slater> karlp: I wasn't sure if it was available at all, pretty new to STM32 2019-10-30T14:43:55 < ac_slater> sorry 2019-10-30T14:44:03 -!- maunix [~Mauricio@181.31.151.95] has joined ##stm32 2019-10-30T14:44:19 < karlp> https://www.st.com/en/microcontrollers-microprocessors/stm32f413-423.html#resource and choose reference manual 2019-10-30T14:45:51 < karlp> the datasheet has pin descriptions and electrical properites, and a summary overview of the peripherals, 2019-10-30T14:45:57 < karlp> the reference manual has all the programming information 2019-10-30T14:46:23 < karlp> the "programming manual" is about programming flash, and is normally just a subset of the reference manual, you can normally ignore it completely unless you're working on debug tooling 2019-10-30T14:47:20 < karlp> huh, except now the programming manual seems to be more about system programming issues, so possibly interesting, but not useful for the peripherals 2019-10-30T14:48:05 < ac_slater> karlp: I really appreciate it 2019-10-30T14:49:31 < ac_slater> karlp: I expected it here https://www.st.com/content/st_com/en/search.html#q=STM32F413xH-t=resources-page=1 2019-10-30T14:50:09 < ac_slater> if I remove the "xH", I get a lot more resources 2019-10-30T14:50:11 < ac_slater> thanks mate 2019-10-30T14:50:45 < qyx> yeah because xH is not a valid part number 2019-10-30T14:50:49 < ac_slater> exactly 2019-10-30T14:51:02 < ac_slater> I appreciate the help 2019-10-30T14:51:45 < day> why do poeple not install termination resistors D: 2019-10-30T14:51:59 < day> *i dont understand it, therefore i don't need it* 2019-10-30T14:52:08 < qyx> they are redundant 2019-10-30T14:52:13 < qyx> exactly like bypass caps 2019-10-30T14:52:15 < day> totally kek 2019-10-30T14:52:43 < qyx> also, why use 10x 100nF, I just use a single 1u instead 2019-10-30T14:53:06 < Haohmaru> datasheet sez so 2019-10-30T14:53:13 < Haohmaru> *shrug* 2019-10-30T14:53:14 < qyx> ok, 2019-10-30T14:53:19 < qyx> back to work 2019-10-30T15:01:31 < karlp> day: or, "I don't need it, so I don't need it" ? 2019-10-30T15:01:57 < karlp> most people aren't actually pushing the boundaries of their transmission lines :) 2019-10-30T15:03:27 < day> im talking about building installations with 100m+ worth of cables 2019-10-30T15:03:56 -!- kow_ [~afed@135.0.26.39] has quit [Ping timeout: 265 seconds] 2019-10-30T15:13:18 < karlp> probably still not pushing boundaries of the lines :) 2019-10-30T15:14:23 < karlp> besides, what's 1-3% bit error when youv'e got crc in the protocols, just read it again, what's a bit of bus efficiency loss between friends :) 2019-10-30T15:15:00 < karlp> what happened in your day today to bring this up? 2019-10-30T15:17:21 < Thorn> why do poeple not install termination resistors D: 2019-10-30T15:17:30 < Thorn> because they're using active OCT? 2019-10-30T15:18:04 < qyx> obsessive compulsive termination? 2019-10-30T15:23:19 < Haohmaru> resistorman 2019-10-30T15:29:58 < karlp> qyx: here you go, greenpak to solve all your problems: https://www.dialog-semiconductor.com/an-cm-290?utm_source=Newsletter&utm_campaign=f2c7d6f180-EMAIL_CAMPAIGN_2019_10_28_11_23&utm_medium=email&utm_term=0_c4ee402843-f2c7d6f180-580037879 2019-10-30T15:30:12 < karlp> ech, just this one https://www.dialog-semiconductor.com/an-cm-290 2019-10-30T15:30:51 < karlp> personally I'd never seen DF as a selection criteria for capactiors before 2019-10-30T15:31:33 < karlp> I guess it's something like this https://www.digikey.com/product-detail/en/tdk-corporation/CGA9P4X7T2W105K250KA/445-7952-2-ND/2781378 2019-10-30T15:33:04 < englishman> karlp: i always thought the programming manual was about the cortex part of the chip, nvic etc 2019-10-30T15:34:18 < karlp> englishman: yeah, it is, I mixed it up with the old PMxxxx thigns that were flash programming 2019-10-30T15:34:37 < karlp> the ones linked now under programming manual do all seem to be coretex instrunction shit 2019-10-30T15:35:33 < karlp> heh. threadX, now called Azure RTOS :) 2019-10-30T15:36:07 < karlp> man, that was back in april they aquired express logic 2019-10-30T15:36:10 < qyx> ažúr 2019-10-30T15:37:28 < qyx> I cannot see much isolation there 2019-10-30T15:37:40 < qyx> karlp: yeah for not isolated applications it looks good 2019-10-30T15:39:44 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-30T15:39:55 < karlp> didyou need isolated too? 2019-10-30T15:40:25 < qyx> I was searching for something usable for PELV 2019-10-30T15:40:37 < karlp> I sitll don't get how they can jsut use this tiny ceramic? cap they have in their demo. gotta have that x rated series cap. 2019-10-30T15:41:07 < karlp> heh, threadx propaganda pdf uses the same example code as "clean" and "unclean" 2019-10-30T15:41:17 < qyx> I have basically two options, either isolate the bus or the power supply 2019-10-30T15:41:52 < qyx> I want to use small signal relays with 2.2mm clearance between coil and contact, so not enuff 2019-10-30T15:42:07 < qyx> probably isolating the bus is the best option 2019-10-30T15:43:05 < qyx> and also, basically fukit as it is just a play-project for my window screening driver 2019-10-30T15:43:09 < qyx> idk how is it called properly 2019-10-30T15:44:15 < qyx> exterior roller blinds 2019-10-30T15:49:53 * Haohmaru intensifies 2019-10-30T16:03:55 < aandrew> nrf51/52 libraries are terrible 2019-10-30T16:03:59 < aandrew> I've a LOT of experience with that 2019-10-30T16:08:15 < zyp> luckily nrf has good documentation so you don't need the libs 2019-10-30T16:09:03 < aandrew> zyp: yes, their docs are decent despite having ot look in one place for the base address and antoher place for the actual register definitions but that's minor 2019-10-30T16:09:33 < aandrew> I wrote a frequency hopping TDMA radio library for nrf51/52 that sixense is using for some of their devices 2019-10-30T16:10:17 < aandrew> gazell is way too low throughput so I just sit on a channel until interference gets above a certain percentage then coordinate a move to another frequency 2019-10-30T16:10:32 < aandrew> we're working on fine tuning that to eke out more performance right now 2019-10-30T16:13:02 < zyp> I still need to find time some time to continue on my zigbee shit 2019-10-30T16:15:59 < qyx> so it is not a truly hopping then? just avoiding interference? 2019-10-30T16:31:45 < Haohmaru> just duct-tape yer receiver onto a grass-hopp0r 2019-10-30T16:54:36 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-30T17:11:21 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T17:13:02 -!- renn0xtk9 [~max@2a02:810d:1540:2448:ccde:39f1:ab98:594f] has joined ##stm32 2019-10-30T17:13:44 -!- con3 [~kvirc@154.119.40.237] has quit [Max SendQ exceeded] 2019-10-30T17:14:28 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T17:16:19 -!- con3 [~kvirc@154.119.40.237] has quit [Max SendQ exceeded] 2019-10-30T17:16:47 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T17:16:49 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-30T17:59:26 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has joined ##stm32 2019-10-30T18:04:37 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-30T18:27:25 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has quit [Ping timeout: 265 seconds] 2019-10-30T18:29:19 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has joined ##stm32 2019-10-30T18:36:16 < Haohmaru> o haitherr loirens 2019-10-30T18:46:26 < karlp> hrm, -Wmaybe-unitialized is warning me about a variable that is set in all three cases of a switch on an enumerated type, that handles all the enumerations. 2019-10-30T18:46:51 < karlp> is taht because C doesn't _realllly_ have enums, and it's just an int behind the scenes? 2019-10-30T18:46:58 < mawk> yes 2019-10-30T18:47:14 < mawk> you can put a default:assert(0 && "unexpected case"); or something to settle that 2019-10-30T18:47:32 < karlp> adding a default: tot he switch made the warning go away, but means I won' having the "unandled enumeration 2019-10-30T18:49:56 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 252 seconds] 2019-10-30T18:51:10 -!- renn0xtk9 [~max@2a02:810d:1540:2448:ccde:39f1:ab98:594f] has quit [Quit: Konversation terminated!] 2019-10-30T18:51:24 < karlp> I think I'd rather init to zero rather than add a default. 2019-10-30T18:51:27 < qyx> I am enforcing default in switch {} 2019-10-30T18:51:39 -!- tprrt [~tprrt@217.114.204.178] has quit [Quit: leaving] 2019-10-30T18:52:11 < qyx> writing switch without default is like writing one-statement if without {} 2019-10-30T18:52:54 < karlp> having default prevents you from getting -Wswitch warnings on unhandled cases though 2019-10-30T18:54:03 < mawk> what's wrong with beautiful if qyx ? 2019-10-30T18:54:31 < qyx> ocpd hurts 2019-10-30T18:56:15 < emeb> -Wall + fixing every error/warning has saved me a lot of time. 2019-10-30T18:56:45 < emeb> latest gcc gives pretty damned good warnings too - even suggests fixes sometimes. 2019-10-30T18:57:02 < mawk> does that hurt qyx ? https://paste.serveur.io/AqfmdEPj.c 2019-10-30T18:57:29 < jpa-> -Wunused -Werror is a bit annoying combination when commenting stuff out for debugging 2019-10-30T18:57:39 < emeb> pirate coding - arr 2019-10-30T18:58:16 < emeb> yeah, -Wunused can be a PITA. 2019-10-30T18:59:04 < emeb> I also use -Wno-strict-aliasing - some of the libs I use generate a lot of those. 2019-10-30T18:59:42 < mawk> you can disable warnings just for the libs 2019-10-30T18:59:48 < mawk> you mean they generate it in the .h right 2019-10-30T19:00:11 < mawk> #pragma GCC diagnostic "-Wno-strict-aliasing" just before the #include, then you can pop the flags stack 2019-10-30T19:00:26 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has quit [Ping timeout: 240 seconds] 2019-10-30T19:00:31 < emeb> It's been a while since I added that. I *think* those warnings were from HAL 2019-10-30T19:00:42 < karlp> mawk: fuck your {}less ifs... 2019-10-30T19:00:49 < jpa-> i usually add -Wno-unsigned-compare 2019-10-30T19:01:21 < jpa-> emeb: just -Wno-strict-aliasing or also -fno-strict-aliasing? 2019-10-30T19:01:46 < emeb> jpa-: just -W... 2019-10-30T19:01:55 < jpa-> sounds pretty daring 2019-10-30T19:02:03 < emeb> orly? 2019-10-30T19:02:27 < jpa-> yeah, in that the compiler will still use the assumptions for optimizations but just won't warn about it 2019-10-30T19:02:56 * Haohmaru always puts scope blockz 2019-10-30T19:03:14 < emeb> hmm... well I suppose the -f would be good too them. 2019-10-30T19:03:24 * emeb curses Stack Overflow. :P 2019-10-30T19:03:37 < Haohmaru> some call it toiletoverflow 2019-10-30T19:05:01 < emeb> hmm... removed the -W and rebuilt. No warnings so I guess this code is OK. 2019-10-30T19:05:09 < emeb> just cargo cult then 2019-10-30T19:07:01 < emeb> aha - there it is: https://pastebin.com/jx7pXZAt 2019-10-30T19:07:20 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/] 2019-10-30T19:07:26 < emeb> lots of those in hal_spi 2019-10-30T19:07:49 < emeb> jpa-: so what's the safest way to fix that? 2019-10-30T19:08:07 < jpa-> -fno-strict-aliasing should be safe 2019-10-30T19:08:49 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-30T19:08:50 < jpa-> it'll just result in slightly slower code 2019-10-30T19:09:10 < emeb> cool. thx. 2019-10-30T19:09:31 < mawk> if you pray that the HAL developers correctly did their job you can just silence the warnings no ? 2019-10-30T19:10:18 < emeb> The thing is that I don't even use the functions where that's happening. 2019-10-30T19:10:24 < jpa-> mawk: well even that line kind of looks like it will not be valid with respect to strict-aliasing, so there is nothing there to hope for 2019-10-30T19:10:41 < jpa-> but what you can hope for is that without -flto, the compiler won't find anything worth optimizing there 2019-10-30T19:12:00 < jpa-> it's kind of annoying that there isn't a nice syntax to just do that as hspi->pTxBuffPtr[0] | (hspi->pTxBuffPtr[1] << 8) 2019-10-30T19:12:34 < mawk> using an union maybe 2019-10-30T19:14:16 < Haohmaru> didn't C++ have some solution for this? 2019-10-30T19:14:23 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has joined ##stm32 2019-10-30T19:14:49 < Haohmaru> wot woz it.. aggregate struct, POD, or somefink like that 2019-10-30T19:15:53 < jpa-> such precise 2019-10-30T19:16:33 -!- ekaologik [~quassel@p5DE86D7F.dip0.t-ipconnect.de] has joined ##stm32 2019-10-30T19:16:46 < emeb> wonder if there's a way to disable strict aliasing just for that one damned file. 2019-10-30T19:17:11 < jpa-> give -fno-strict-aliasing when compiling just that file? 2019-10-30T19:18:07 < emeb> I suppose that means a special target in the makefile for that file 2019-10-30T19:18:11 < Haohmaru> -fpls-make-it-work-thankz 2019-10-30T19:18:38 < jpa-> gcc is pretty much the only compiler that really likes this strict-aliasing stuff 2019-10-30T19:19:06 < jpa-> and many big projects give -fno-strict-aliasing when building "just to be sure" 2019-10-30T19:20:47 < emeb> ok - unique target for hal_spi.o did the trick. 2019-10-30T19:25:03 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-30T19:25:21 < Haohmaru> one post on toiletoverflow sez that the proper way to do this sh*t is to use memcpy() 2019-10-30T19:25:33 < emeb> rofl 2019-10-30T19:25:37 < jpa-> it's a valid way 2019-10-30T19:25:47 < Haohmaru> or std::memcpy() 2019-10-30T19:25:54 < jpa-> but any access through char* is equally ok 2019-10-30T19:26:35 < Haohmaru> another post sez that the char* thing is via an eggception to the rule and only works in one direction or somefink 2019-10-30T19:26:47 < Haohmaru> any kind of char btw 2019-10-30T19:26:57 < jpa-> Haohmaru: why do you write like a lolcat? 2019-10-30T19:27:10 < Haohmaru> >:) 2019-10-30T19:27:12 < emeb> this kind of crap is why whitequark says C is a terrible language 2019-10-30T19:27:43 < Haohmaru> jpa- maybe i *think* like a lolcat 2019-10-30T19:27:56 < emeb> or is a lolcat 2019-10-30T19:27:57 < Haohmaru> this realisation suddenly starts to concern me 2019-10-30T19:29:52 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-30T19:30:13 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 268 seconds] 2019-10-30T19:32:09 < kakinull> Steffanx: do you use encrypted pool? 2019-10-30T19:32:43 < Steffanx> Nope 2019-10-30T19:33:08 < jpa-> encrypted pool means something like swimming in a grave? 2019-10-30T19:33:27 < kakinull> Steffanx: any encryption at any phase? 2019-10-30T19:33:31 < kakinull> client side? 2019-10-30T19:34:27 < kakinull> I want transparent client side encryption 2019-10-30T19:34:48 < kakinull> access it like a harddrive do whatever you want 2019-10-30T19:34:57 < Haohmaru> wut r u encrypting, kakiman? 2019-10-30T19:35:21 < kakinull> what are you not encrypting Haohmaru? 2019-10-30T19:35:32 < Haohmaru> my farts 2019-10-30T19:36:01 < Haohmaru> i transmit them in their natural pure aroma 2019-10-30T19:36:24 < kakinull> k 2019-10-30T19:36:38 < Haohmaru> but tell me, don't avoid muh question 2019-10-30T19:36:47 < Haohmaru> i must know ur sikritz 2019-10-30T19:37:11 < Haohmaru> well, some of them only 2019-10-30T19:37:39 < Haohmaru> don't b such a kryptomir 2019-10-30T19:37:55 < kakinull> no sikrit just privacy 2019-10-30T19:38:33 < Steffanx> It's all local, no encryption. Except for what samba maybe does 2019-10-30T19:38:36 < kakinull> I concider my nas machine as "non-trusted platform" 2019-10-30T19:38:54 < kakinull> but at same time give everything for google 2019-10-30T19:39:03 < Steffanx> Mine cannot be accessed from the outside world without vpn 2019-10-30T19:39:30 < Steffanx> If you're in my network I'm lost anyway 2019-10-30T19:40:00 < Haohmaru> i'm in ur comput0r, sorting ur filez 2019-10-30T19:40:05 -!- boB_K7IQ [boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-30T19:40:21 < kakinull> once my brother used "free" vpn 2019-10-30T19:41:01 < Steffanx> kakinull has a brother :o 2019-10-30T19:41:08 < Steffanx> Is he more or less kaki? 2019-10-30T19:41:18 < kakinull> totally not kaki 2019-10-30T19:41:44 < kakinull> so there was attacker inside the network via that free vpn 2019-10-30T19:42:22 < Steffanx> Hm 2019-10-30T19:43:26 < kakinull> national telecommunications agency sent a letter that your local network is spewing malicious communications to internets 2019-10-30T19:44:10 < kakinull> and if you don't do anything they must K&B 2019-10-30T19:44:12 < Haohmaru> oopz 2019-10-30T19:45:27 < kakinull> is there any accelerator card for my NAS? 2019-10-30T19:45:41 < kakinull> to compress faster etc. 2019-10-30T19:45:46 < kakinull> pci-e 1x 2019-10-30T19:46:07 < kakinull> compress faster and encrypt decrypt faster 2019-10-30T19:46:24 < kakinull> because such old processor I doubt it has encrypt decrypt engine 2019-10-30T19:46:32 < Haohmaru> put the thing in a pressure-cooker 2019-10-30T19:47:27 < kakinull> https://ark.intel.com/content/www/us/en/ark/products/65693/intel-core-i3-3220-processor-3m-cache-3-30-ghz.html 2019-10-30T19:47:34 < kakinull> AES instructions - NO 2019-10-30T19:48:17 -!- Haohmaru [~Haohmaru@195.24.53.110] has quit [] 2019-10-30T19:49:19 < kakinull> maybe network card that has accelerator? 2019-10-30T19:52:15 < kakinull> what kind of things are offloaded to network card? 2019-10-30T19:52:22 < kakinull> ip package framing? 2019-10-30T19:52:26 < kakinull> and that's all? 2019-10-30T19:56:10 < jpa-> that doesn't sound like a thing that would be offloaded 2019-10-30T19:57:17 < jpa-> https://www.kernel.org/doc/html/v4.17/networking/dpaa2/overview.html this is one kind of network card acceleration 2019-10-30T19:57:27 < jpa-> but not particularly common 2019-10-30T19:57:30 < kakinull> I meant ethernet frame 2019-10-30T19:58:13 < jpa-> why do you need encryption on your NAS? 2019-10-30T19:58:17 < kakinull> I think it's time to buy xeon 2019-10-30T19:58:40 < kakinull> benchmark score doesn't matter as much as special instructions 2019-10-30T20:02:23 < kakinull> jpa-: accelerate client-server communications? 2019-10-30T20:03:31 < jpa-> what does that mean? 2019-10-30T20:04:11 < mawk> checksumming kakinull 2019-10-30T20:04:16 < mawk> and framing yes 2019-10-30T20:04:17 < kakinull> I mean isn't communication encrypted between nas client and host? 2019-10-30T20:05:33 < jpa-> kakinull: depends totally on what client you use 2019-10-30T20:05:55 < jpa-> smb, nfs, sshfs all support non-encrypted traffic also 2019-10-30T20:06:25 < kakinull> I didn't see an option 2019-10-30T20:06:34 < kakinull> I assume encrypted comms 2019-10-30T20:07:23 < jpa-> and is your NAS cpu maxing out currently? 2019-10-30T20:07:49 < kakinull> no 2019-10-30T20:08:20 -!- sterna1 [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-30T20:12:08 -!- turnip420 [~machinehu@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-30T20:18:42 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-30T20:21:11 < kakinull> Steffanx: how is performance with your 9211 card? 2019-10-30T20:22:03 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-30T20:37:55 -!- fsasm [~fsasm@62-178-93-7.cable.dynamic.surfer.at] has quit [Ping timeout: 265 seconds] 2019-10-30T20:41:29 < Steffanx> Didnt really test it, but it's fine. 2019-10-30T20:41:31 < kakinull> I suspect mine works in 1x pci-e mode 2019-10-30T20:41:42 < kakinull> just crappy mobo 2019-10-30T20:42:18 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-30T20:42:19 < Steffanx> I have something not too old with a simple pentium g4560 2019-10-30T20:42:20 < kakinull> friend said that your basic homepc / gamepc mobo may use wide pci-e connector in 1x mode if you don't have GPU in there 2019-10-30T20:42:31 < jpa-> isn't 1x pci-e plenty fast for gigabit ethernet? 2019-10-30T20:42:44 < kakinull> but how about SAS card? 2019-10-30T20:42:59 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-30T20:43:05 < jpa-> is your data going somewhere faster than to the network? 2019-10-30T20:43:14 < kakinull> to drives 2019-10-30T20:43:19 < kakinull> softraid 2019-10-30T20:43:34 < kakinull> total throughbutt must be higher 2019-10-30T20:43:48 < Steffanx> Best thing about g4560 is that it even does ecc memory 2019-10-30T20:44:34 < jpa-> SAS with softraid sounds like a funny combo, but 500 MByte/s still gets you pretty far even with RAID1 2019-10-30T20:45:11 < Steffanx> Works fine with freenas 2019-10-30T20:45:15 < Steffanx> The comb9 2019-10-30T20:45:17 < Steffanx> O 2019-10-30T20:45:31 < jpa-> yeah but aren't SAS drives a bit costly? 2019-10-30T20:45:56 < Steffanx> SAS -> 4x sata 2019-10-30T20:46:14 < Steffanx> That's what kakinull forgot to mention 2019-10-30T20:46:27 < Steffanx> (And me too) 2019-10-30T20:46:28 < jpa-> ok so basically just using SAS card as a SATA card, or is there an active converter in between? 2019-10-30T20:46:42 < kakinull> no 2019-10-30T20:46:51 < kakinull> just cable harness 2019-10-30T20:47:00 < kakinull> or extender 2019-10-30T20:47:05 < jpa-> ok, might as well call it SATA card then :P 2019-10-30T20:47:40 < kakinull> with extenders the card may have total of 256 drives 2019-10-30T20:49:44 < Steffanx> No, comes with a SAS connector jpa- . So its SAS :P 2019-10-30T20:52:05 < kakinull> yes 2019-10-30T20:56:44 < qyx> karlp: https://octavosystems.com/ 2019-10-30T20:57:05 < qyx> re: small packaged linux 2019-10-30T20:57:15 < qyx> available at mouser 2019-10-30T20:58:45 < antto> dedmau5 2019-10-30T20:59:45 < kakinull> https://drive.google.com/file/d/1H3wIyh1AfobrCH3bvQtk_7Xx7TVfC2nd/view?usp=sharing card removed 2019-10-30T20:59:48 < kakinull> time to use mobo sata 2019-10-30T21:04:01 < Steffanx> awwwwh 2019-10-30T21:15:48 -!- X230t [x13@gateway/shell/suchznc/x-tpabxgugphajeghq] has joined ##stm32 2019-10-30T21:24:18 < englishman> qyx: have you seen these SiFive bros 2019-10-30T21:37:13 < kakinull> card didn't work like it should have in my mobo 2019-10-30T21:37:16 < kakinull> steff 2019-10-30T21:37:29 < kakinull> I get 75MBps now on copper 2019-10-30T21:37:39 < kakinull> samba 2019-10-30T21:38:01 < kakinull> it was before at 20-30 2019-10-30T21:38:41 < Steffanx> isnt 75 still a bit slow? 2019-10-30T21:38:53 < kakinull> ye 2019-10-30T21:38:56 < kakinull> but usable 2019-10-30T21:39:24 < kakinull> it may have been also if there was bad connection and io errors or something 2019-10-30T21:39:37 < kakinull> but idk. I'm not familiar with this stuff 2019-10-30T21:40:52 < kakinull> but I strongly believe it's mobo issue 2019-10-30T21:41:58 < kakinull> jpa-: you need 9211-8i? 2019-10-30T21:42:41 < kakinull> with 2x minisas-4sata adapter cables? 2019-10-30T21:48:40 < kakinull> cpu load peaked at 10% 2019-10-30T21:48:48 < kakinull> new record 2019-10-30T21:50:25 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-30T21:53:08 < Steffanx> finternet. 2019-10-30T21:59:56 -!- emeryth [emeryth@boston-packets.hackerspace.pl] has quit [Ping timeout: 276 seconds] 2019-10-30T22:02:15 -!- emeryth [emeryth@2a0d:eb00:2137:2::10] has joined ##stm32 2019-10-30T22:09:45 < antto> same53n20 for stereo audio in/out, does it sound like a plan? 2019-10-30T22:11:31 < antto> 120MHz 2019-10-30T22:18:27 < zyp> you're on ##stm32, why do you ask if using a sam part is a good idea? 2019-10-30T22:18:40 < Cracki> first ipod had a 90 mhz arm 7tdmi 2019-10-30T22:18:44 < Cracki> *two of those 2019-10-30T22:28:23 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2019-10-30T22:33:09 < Steffanx> I assume you want more than just in/out antto ? 2019-10-30T22:33:24 < Steffanx> So more than just some i2s-ing or whatever 2019-10-30T22:41:40 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-30T22:42:26 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-30T22:44:51 -!- ac_slater [~weechat@144.121.38.133] has quit [Ping timeout: 246 seconds] 2019-10-30T22:45:27 -!- ac_slater [~weechat@144.121.38.133] has joined ##stm32 2019-10-30T22:45:51 < antto> perhaps it's time i find a more generic arm corteggz irc channel 2019-10-30T22:47:44 < mawk> nooooooo stay with us antto 2019-10-30T22:48:14 -!- maunix [~Mauricio@181.31.151.95] has left ##stm32 ["WeeChat 2.6"] 2019-10-30T22:48:38 -!- emeryth [emeryth@2a0d:eb00:2137:2::10] has quit [Ping timeout: 252 seconds] 2019-10-30T22:48:58 -!- emeryth [emeryth@boston-packets.hackerspace.pl] has joined ##stm32 2019-10-30T22:52:29 < Steffanx> but why go for SAM when you have the fancy new G4 for example. Why antto? 2019-10-30T22:52:32 < Steffanx> ^ legit question 2019-10-30T22:54:00 < antto> i happen to be a little bit familiar with the same5x already.. very little, but it's still above zero 2019-10-30T22:55:03 < Cracki> so what does it need to do 2019-10-30T22:56:00 < mawk> my new company is using stm32 2019-10-30T22:56:14 < antto> basically i have a VST which (IMO) deserves to run on a chip in a hardware box 2019-10-30T22:56:21 < Steffanx> Your new company? 2019-10-30T22:56:22 < mawk> well the useless coffee-drinking HR guys in a suit said so, but they don't understand anything at all to tech stuff so I don't know 2019-10-30T22:56:23 < antto> it's a synth 2019-10-30T22:56:38 < mawk> yes Steffanx , in Delft 2019-10-30T22:56:46 < mawk> I'm not working there yet so it's still new 2019-10-30T22:56:59 < antto> it's also maybe a gud eggscuse to try a hit my head against the arm cortex wall 2019-10-30T22:57:11 < Cracki> so using stm32 was someone's executive or marketing decision, not based on engineering? 2019-10-30T22:57:37 < mawk> no Cracki it's just that I don't know if they know what stm32 even is, they said "oh yes we are using that" but maybe they just overheard a conversation between two devs 2019-10-30T22:57:44 < Cracki> heh 2019-10-30T22:57:47 < Steffanx> hah lol 2019-10-30T22:58:10 < Cracki> maybe they actually do use it and a suit has a clue for once 2019-10-30T22:58:29 < mawk> well after that I talked about my experience with the RTMS RTOS and they said "you think we should use that ?" 2019-10-30T22:58:34 < mawk> "uhhhhh... it depends ?" 2019-10-30T22:58:47 < Steffanx> not unlikely they use stm32 :) 2019-10-30T22:59:03 < mawk> RTEMS not RTMS 2019-10-30T22:59:08 < Steffanx> you didnt talk with engineers? 2019-10-30T22:59:11 < Steffanx> *the engineers 2019-10-30T22:59:29 < antto> reminds me of when muh boss suddenly decided we should make our new things with "arm" chips 2019-10-30T22:59:31 < antto> everything 2019-10-30T22:59:35 < mawk> lol 2019-10-30T22:59:36 < antto> cuz "arm" 2019-10-30T22:59:55 < Cracki> that's like saying you once owned an audi and these pedestrians ask if they should buy audis now 2019-10-30T23:00:12 < mawk> yeah 2019-10-30T23:00:20 < mawk> I had a meeting with an engineer but we didn't talk much about what they were doing Steffanx , they just asked about me 2019-10-30T23:00:29 < Cracki> antto, what did you guys use before the reeducation towards arm? 2019-10-30T23:00:33 < antto> yeah, shall i put a smol audi in muh kitchen? 2019-10-30T23:00:41 < antto> or is it too much 2019-10-30T23:01:05 < Cracki> if they had some clue, they'd get you to talk about RTOSes in general 2019-10-30T23:01:15 < antto> Cracki after i lost a pile of time, we didn't change anything 2019-10-30T23:01:15 < Cracki> very brand-aware, these suits 2019-10-30T23:01:36 < Cracki> using pic? 2019-10-30T23:01:41 < Cracki> z80? 2019-10-30T23:01:44 < antto> no, xmega 2019-10-30T23:01:50 -!- turnip420 [~machinehu@d207-216-21-173.bchsia.telus.net] has left ##stm32 ["WeeChat 1.4"] 2019-10-30T23:01:57 < Cracki> acceptable 2019-10-30T23:02:00 < Steffanx> *eggsmega 2019-10-30T23:02:08 < antto> we do use pics for some old crap we keep dragging around :~( 2019-10-30T23:02:14 < Steffanx> ohno 2019-10-30T23:02:24 < Cracki> you're like that shop I worked at who used perl for everything 2019-10-30T23:02:29 < antto> when i came here the dev guy was a pic fan who wrote ASM only 2019-10-30T23:02:33 < mawk> lol 2019-10-30T23:02:36 < Cracki> but nobody they employed knew perl anymore, they just keep the shit working 2019-10-30T23:02:57 < Cracki> and then the head honcho said "we'll use python now" and they employed noobs who worked through the python tutorial 2019-10-30T23:03:06 < antto> we still drag a pile of .hex files for pics 2019-10-30T23:03:07 -!- turnip420 [~machinehu@d207-216-21-173.bchsia.telus.net] has joined ##stm32 2019-10-30T23:03:21 < Cracki> and then I come along wanting to use list comprehensions and they're like "omg no that's too complicated we can't expect our monkeys to understand that" 2019-10-30T23:03:27 < mawk> lol 2019-10-30T23:03:28 < antto> copies of similar projects with minute variations cuz there's no "settings" in anything 2019-10-30T23:03:34 < mawk> that would've made me want to quit Cracki 2019-10-30T23:03:43 < Cracki> and that's what I eventually did 2019-10-30T23:03:57 < Cracki> but mostly because the work was boring 2019-10-30T23:04:02 < mawk> in my maths school they just invented the Python course so they took a teacher at random who just read the wikipedia page of python and started to teach us on that, and I was the only one in class with knowledge of it 2019-10-30T23:04:13 < mawk> so he came next to me during class to take notes on what I was doing 2019-10-30T23:04:19 < antto> i seriously don't know how u gon sit down and write something semi-big in ASM 2019-10-30T23:04:29 < Steffanx> why not antto. 2019-10-30T23:04:32 < Cracki> srsly such a teacher should just throw money at you and make you hold the class 2019-10-30T23:04:50 < Steffanx> the w register is booming the the pic-world. 2019-10-30T23:04:57 < mawk> lol 2019-10-30T23:04:58 < Cracki> like any good but incompetent high school "comp sci" teacher does. they pick the wiseasses to hold class 2019-10-30T23:04:59 < antto> the wat 2019-10-30T23:05:57 < Steffanx> The working register, antto 2019-10-30T23:06:15 < antto> le grande shrug 2019-10-30T23:07:01 < Steffanx> AVR assembly isnt tooo bad either 2019-10-30T23:07:38 -!- ac_slater [~weechat@144.121.38.133] has quit [Ping timeout: 240 seconds] 2019-10-30T23:07:40 < antto> i remember when one day teh boss came and said we gotta make somefin' with voice recognition 2019-10-30T23:08:07 < antto> in his mind, he was already seeing teh cash flow 2019-10-30T23:08:22 < antto> that's when i told him how this ain't gonna happen ;P~ 2019-10-30T23:08:28 < antto> "why?!" 2019-10-30T23:08:30 < Cracki> lol 2019-10-30T23:08:35 < Cracki> voice recog on embedded? 2019-10-30T23:08:37 < antto> "cuz it's f*cking hard" 2019-10-30T23:08:46 < antto> "but muh phone does it" 2019-10-30T23:08:55 < mawk> you never heard of eDgE computing Cracki ? 2019-10-30T23:09:03 < Cracki> yuh phone sends the sound to a server that does it 2019-10-30T23:09:03 < Steffanx> why didnt you pick an stm32 and did live encoding to mp3, antto? 2019-10-30T23:09:06 < mawk> embedded you connect to the internet to delegate the processing to big servers 2019-10-30T23:09:07 < antto> "no, ur phone probably just records teh audio and sends it to mothership" 2019-10-30T23:09:49 < Steffanx> and send that to the web indeed 2019-10-30T23:09:51 < Steffanx> such stm32 power 2019-10-30T23:09:56 < Steffanx> you should never tell your boss no 2019-10-30T23:09:57 < antto> he even suggested "we can get chips like this, i KNOW it probably won't work on a pic" <- no shizzle 2019-10-30T23:09:58 < Steffanx> just spend munny 2019-10-30T23:10:02 < Steffanx> and more of it 2019-10-30T23:10:05 < Steffanx> until he sees it was a bad iea 2019-10-30T23:10:27 < mawk> lol 2019-10-30T23:10:30 < mawk> then he will blame you 2019-10-30T23:10:33 < Cracki> kek so I do some video editing on the side along with a few others, we share two editing workstations, and some asshole comes along saying "just virtualize it". yeah sure let's have the system waste resources on compressing and transmitting the whole desktop and audio over network 2019-10-30T23:10:42 < antto> we don't have infinite moniez to fool around with absolutely ded projects 2019-10-30T23:10:49 < Cracki> voice recog chips? uh... maybe 2019-10-30T23:10:59 < Cracki> for a very limited set of words 2019-10-30T23:11:25 < Steffanx> so what does the company you work for make, antto? Audio related stuff or is that just hobbying for you? 2019-10-30T23:11:37 < antto> nothing audio 2019-10-30T23:11:50 < antto> this audio shizzle is for muhself 2019-10-30T23:11:51 < Cracki> customers expect modern recog performance and that means deep neural networks 2019-10-30T23:12:25 < antto> i don't believe in authentication via voice recognition 2019-10-30T23:12:54 < antto> "i am Obama, open teh door" 2019-10-30T23:12:56 < Steffanx> so what is it then? all your xmega work must do something in the end. 2019-10-30T23:13:10 < antto> access control mostly 2019-10-30T23:13:16 < antto> and related sh*t 2019-10-30T23:13:20 < antto> boring crap 2019-10-30T23:13:39 < antto> relayz, inputz, rfid 2019-10-30T23:13:57 < antto> blink this LED, beep that buzz, send this packet.. 2019-10-30T23:14:03 < Steffanx> i see 2019-10-30T23:14:31 -!- deskwizard [~quassel@unaffiliated/deskwizard] has joined ##stm32 2019-10-30T23:15:12 < antto> i'm interested in pretty things like audio DSP, graphics/video/pixels.. 2019-10-30T23:15:16 < mawk> I will do bycicle parking measurements Steffanx 2019-10-30T23:15:19 < mawk> pure dutch tech 2019-10-30T23:15:24 < antto> don't care bout opening dumb doors 2019-10-30T23:15:24 < Steffanx> lolwut 2019-10-30T23:15:25 < mawk> bicycle 2019-10-30T23:15:37 < mawk> like, detect if the velo parking spot if empty or not 2019-10-30T23:15:49 < mawk> well I guess that project is over now, but that's the kind of thing 2019-10-30T23:16:00 < Steffanx> sounds like one of those random projecst that will never be real. 2019-10-30T23:16:02 < Cracki> use the blockchain 2019-10-30T23:16:10 < mawk> it's already done Steffanx , it's used in several dutch cities 2019-10-30T23:16:12 < Cracki> or induction coil in the ground 2019-10-30T23:16:16 < mawk> they do it with CV 2019-10-30T23:16:22 < Steffanx> One of those EU subsidized projects 2019-10-30T23:16:25 < mawk> at first with laser or stuff but that was too difficult 2019-10-30T23:16:30 < Cracki> so I hang a photo of a bicycle in front of the camera 2019-10-30T23:16:31 < mawk> the velo could be parked in a nonsense way and not detected 2019-10-30T23:16:36 < mawk> lol 2019-10-30T23:17:03 < Cracki> car parking garages use sonar or radar on every space 2019-10-30T23:17:04 < antto> velo anarchists 2019-10-30T23:17:20 < mawk> they do the parking detection in the Amsterdam Zuid velo station for instance Steffanx 2019-10-30T23:17:22 < antto> wut if it's a monocycle? 2019-10-30T23:17:23 < deskwizard> Howdy. I was wondering if I could get some pointers for choosing the right STM32 board, I'm having a look at what's available on digikey, and it's a bit overwelming... long story short, USB CDC, USB HID, a couple of uart and some i²c would be pretty much what I'm looking for. I had a bluepill and I'm tired of fighting the damn thing 2019-10-30T23:17:28 < Cracki> I guess velos don't reflect as well as a car roof 2019-10-30T23:17:31 < antto> some clowns ride them 2019-10-30T23:17:36 < antto> f*cking clowns! 2019-10-30T23:17:46 < Cracki> deskwizard, get a nucleo or disco? 2019-10-30T23:17:51 < Cracki> pick whatever 2019-10-30T23:18:00 < Steffanx> sure sure its real, mawk. but it does sound like "one of those projects" 2019-10-30T23:18:02 < mawk> you have a selector stuff on st website deskwizard 2019-10-30T23:18:06 < mawk> way easier than looking at digikey 2019-10-30T23:18:14 < Steffanx> sure its cool, but .. it doesnt really work or isnt something we really need :P 2019-10-30T23:18:20 < deskwizard> mawk: Oh, thanks! I'll have a look 2019-10-30T23:18:22 < mawk> you select the flash size, the features you want, the category (low power, normal, high power, ...), and you get your references 2019-10-30T23:18:44 < mawk> yeah Steffanx I hope that one is over lol 2019-10-30T23:19:14 < mawk> and yeah get a nucleo board if you can, for ethernet you have nucleo-144 boards with ethernet port and usb user port 2019-10-30T23:19:26 -!- ekaologik [~quassel@p5DE86D7F.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-30T23:19:35 < buZz> antto: dont beep me! 2019-10-30T23:19:37 < buZz> :) 2019-10-30T23:19:41 < Steffanx> buZz: buZz 2019-10-30T23:19:43 < mawk> for instance Nucleo-H743ZI if you want the biggest board deskwizard 2019-10-30T23:19:46 < mawk> to feel like a man 2019-10-30T23:19:48 < buZz> ^_^ 2019-10-30T23:19:58 * antto goes into hi-Z state 2019-10-30T23:20:03 < Steffanx> H743 isnt dual core is it? 2019-10-30T23:20:28 < Cracki> get the smallest possible one if you want less headache in taming it 2019-10-30T23:20:29 < mawk> not that I know of Steffanx 2019-10-30T23:20:42 < Steffanx> so get the H745 instread :) 2019-10-30T23:20:50 < deskwizard> lmao for the moment I'd prefer something smaller, the more there is the more I can play with and the less I'll get done :P 2019-10-30T23:20:53 < antto> bruh, isn't that M7 2019-10-30T23:21:04 < Cracki> bluepill or similar would have been just fine. the software/firmware side sounds like your problem 2019-10-30T23:21:04 < Steffanx> +M4 2019-10-30T23:21:36 < deskwizard> Cracki: indeed 2019-10-30T23:21:38 < mawk> yeah for usb you want either nucleo-144 or bluepill, the other nucleo don't have usb port 2019-10-30T23:21:45 < Cracki> so, cubemx 2019-10-30T23:21:47 < mawk> you have to solder one yourself or whatever 2019-10-30T23:22:00 < Cracki> cubemx generates setup code 2019-10-30T23:22:22 < Cracki> you'll need docs for stm32F1 HAL to make it actually do stuff 2019-10-30T23:22:45 < englishman> one of the h7 is dual core I'm pretty sure 2019-10-30T23:22:50 < Steffanx> No, not bluepill, forget about f103. https://www.aliexpress.com/item/4000103610226.html it is. F401/F411 much. 2019-10-30T23:22:51 < englishman> m4 coprocessor 2019-10-30T23:23:43 < englishman> deskwizard: stm32g0 or G4 nucleo-64 probably 2019-10-30T23:23:57 < Steffanx> englishman is all into the G now. 2019-10-30T23:24:02 < englishman> yes 2019-10-30T23:24:08 < englishman> day fmac 2019-10-30T23:24:10 < englishman> dat 2019-10-30T23:24:17 < Steffanx> and cordicks 2019-10-30T23:24:19 < englishman> dat fmac 2019-10-30T23:24:19 < Steffanx> -s 2019-10-30T23:24:26 < englishman> and cordicks 2019-10-30T23:24:35 < englishman> and secure boot 2019-10-30T23:24:40 < englishman> and dual bank 2019-10-30T23:24:59 < englishman> and 170MHz 2019-10-30T23:25:10 < englishman> and still 1/4 price of msp430 2019-10-30T23:25:13 < mawk> nucleo-64 has no usb port it's ugly 2019-10-30T23:26:24 -!- smvoss [~smvoss@207.191.220.92] has quit [Ping timeout: 268 seconds] 2019-10-30T23:26:50 < Steffanx> do you watch dumpertreeten every wednesday evening yet, mawk? 2019-10-30T23:26:56 < Steffanx> must be part of you dutchifiying 2019-10-30T23:27:10 < Steffanx> You have to know at least a few dumpert quotes 2019-10-30T23:27:23 < Steffanx> "blikje in die water" "Hedde drugs op" etc. 2019-10-30T23:27:48 < mawk> lol 2019-10-30T23:28:19 < deskwizard> bah soldering an USB port isn't an issue really. 2019-10-30T23:28:40 < deskwizard> the more I look the more it feels like I'm out of my league hehehe I'll keep hunting. Thanks for the help ! 2019-10-30T23:29:15 < Steffanx> Cant go much wrong with a 3 dollar aliexpress board 2019-10-30T23:29:20 < Steffanx> except for it not working but .. :P 2019-10-30T23:29:39 < mawk> why out of your league deskwizard ? 2019-10-30T23:29:41 < mawk> don't worry 2019-10-30T23:29:45 < mawk> I started the same way 2019-10-30T23:29:56 < Steffanx> Jan- is here to help you. 2019-10-30T23:30:09 < mawk> lol 2019-10-30T23:30:26 < Steffanx> Sorry, im not helpful 2019-10-30T23:31:11 < deskwizard> Steffanx: a 3$ board from that part of the world is precisely why I'm into this mess :P 2019-10-30T23:31:12 < karlp> deskwizard: what part of "fighting" the bluepill are you hoping to avoid? or thinking a different board will help you with? 2019-10-30T23:31:29 < Steffanx> hah, excellent deskwizard 2019-10-30T23:32:22 < karlp> mawk: if you want user usb, there's plenty of disco boards with that, just only nucleo144 if you want nucleo stylee 2019-10-30T23:32:31 < karlp> deskwizard: so, what _was_ the problem with that board then? 2019-10-30T23:33:10 < deskwizard> karlp: getting usb working indeed 2019-10-30T23:33:32 < mawk> out of 5 bluepills 1 had a badly soldered usb connector, so that's still 80% success rate ! 2019-10-30T23:34:09 < deskwizard> mawk: yeah I got the one ... :P fixed that, and the resistor that's wrong (R10 iirc but don't quote me on that) 2019-10-30T23:35:49 < mawk> yeah but it works with 99% of computers anyway 2019-10-30T23:35:56 < mawk> but the resistor value is bad indeed 2019-10-30T23:41:09 < mawk> lol the translation of your link Steffanx « The factory burned the breathing lamp and USBCDC test procedures. » 2019-10-30T23:42:33 -!- CheBuzz [~CheBuzz@unaffiliated/chebuzz] has quit [Ping timeout: 265 seconds] 2019-10-30T23:45:54 -!- CheBuzz [~CheBuzz@unaffiliated/chebuzz] has joined ##stm32 2019-10-30T23:51:37 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a873:5df4:f382:a0d7] has joined ##stm32 2019-10-30T23:59:42 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 --- Day changed Thu Oct 31 2019 2019-10-31T00:07:52 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-31T00:40:32 < kakinull> sweet 2019-10-31T00:40:46 < Steffanx> Found you 2019-10-31T00:40:54 < Steffanx> r waffle stack? 2019-10-31T00:40:58 < kakinull> lte module has some smaller than typical connector 2019-10-31T00:41:03 < kakinull> not u.fl 2019-10-31T00:41:10 < kakinull> slightly smaller 2019-10-31T00:42:21 < Steffanx> Mhf4? 2019-10-31T00:44:54 < kakinull> ipex mhf4 2019-10-31T00:45:04 < kakinull> I wonder if they will make it even smaller after that 2019-10-31T00:45:33 -!- CygniX [~CygniX@opensuse/member/CygniX] has quit [Max SendQ exceeded] 2019-10-31T00:47:07 -!- kakipro [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-31T00:47:39 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-31T00:49:46 -!- CygniX [~CygniX@opensuse/member/CygniX] has joined ##stm32 2019-10-31T00:51:30 < kakinull> oh mhf3 2019-10-31T00:51:38 < kakinull> that's the smaller one 2019-10-31T00:52:55 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-31T00:54:07 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has joined ##stm32 2019-10-31T01:01:21 < kakinull> https://medium.com/@melissamcewen/who-killed-the-junior-developer-33e9da2dc58c 2019-10-31T01:07:51 -!- k\o\w [~afed@135.0.26.39] has joined ##stm32 2019-10-31T01:08:58 < englishman> kakinull: buy a fucking synology already 2019-10-31T01:09:19 < englishman> you've been fixing your nas for literally a year 2019-10-31T01:09:30 < kakinull> it has high performance now 2019-10-31T01:09:41 < kakinull> it has raid 2019-10-31T01:09:46 < kakinull> it's running 2019-10-31T01:09:57 < kakinull> now I need to just start using it 2019-10-31T01:11:11 < kakinull> it's sitting there with blue power light glowing 2019-10-31T01:16:53 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a873:5df4:f382:a0d7] has quit [Ping timeout: 276 seconds] 2019-10-31T01:20:51 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has joined ##stm32 2019-10-31T01:21:30 < kakinull> https://drive.google.com/file/d/1azbLMOjP90ZzH8OQPVM873463SvUmaST/view?usp=sharing there.. 2019-10-31T01:21:37 < kakinull> englishman 2019-10-31T01:34:17 < englishman> not clicking your spynet link 2019-10-31T01:40:29 < kakinull> fair enough 2019-10-31T01:42:01 < mawk> englishman: https://pix.watch/cHl_rh/9BDyIo.png 2019-10-31T01:42:13 < mawk> Ei turvallinen kakinull ! 2019-10-31T01:42:24 < mawk> looks like a degenerated latin 2019-10-31T01:42:55 < englishman> not clicking pedonet links either 2019-10-31T01:43:05 < mawk> :((( 2019-10-31T01:43:07 < mawk> I'm not pedonet 2019-10-31T01:43:15 < mawk> my site is very respectable 2019-10-31T01:45:21 < kakinull> to be honest I don't klick mawks links either 2019-10-31T01:45:48 < mawk> what ! 2019-10-31T01:46:01 < mawk> kakibetrayer 2019-10-31T01:46:29 < mawk> pix.watch is just 200 lines of python, it's not pedo at all 2019-10-31T01:46:41 < mawk> plus I was 16 when I did it, can't be for pedos without being an adult myself 2019-10-31T01:47:14 < kakinull> interesting perspective 2019-10-31T01:47:44 < mawk> plus I fought the pedos 2019-10-31T01:47:53 < mawk> even stopped some of the URLs and lose a whole bunch of users 2019-10-31T01:47:56 < mawk> to kick them away 2019-10-31T01:48:13 < mawk> I had 90% of my users through Tor but I stopped it so the pedos go away 2019-10-31T01:50:00 < mawk> my site was a big thing in the french "not so legal" community, I received like 400€ of donations to thank me for running it 2019-10-31T01:50:06 < mawk> and there were no pædos there 2019-10-31T01:51:40 < mawk> just 1TiB drives kakinull , 2019-10-31T01:51:44 < mawk> ? 2019-10-31T01:51:47 < mawk> why are they named ada ? 2019-10-31T01:51:47 < kakinull> 2TB 2019-10-31T01:51:51 < mawk> ah 2019-10-31T01:51:55 < mawk> with a RAID1 or something ? 2019-10-31T01:51:55 < kakinull> ask freebsd 2019-10-31T01:51:59 < mawk> a 2019-10-31T01:52:02 < kakinull> raid-z2 2019-10-31T01:52:10 < mawk> wassthat 2019-10-31T01:52:59 < kakinull> idk 2019-10-31T01:53:14 < kakinull> it's like mirror and stripe combined 2019-10-31T01:53:32 < kakinull> maybe 2019-10-31T01:53:49 < kakinull> it was the bestest option 2019-10-31T01:53:58 < kakinull> I would add 4 x 1tb pool too 2019-10-31T01:54:49 < kakinull> but as the mobo is commercial garbage from 2012 it will not perform with sas card 2019-10-31T01:54:57 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-31T01:55:17 < kakinull> there is brigdes and other bottlenecks and the thing is probs running with 1x line instead of 8x 2019-10-31T01:55:19 < mawk> when I installed a SAN for a company I used two servers with 16 3TiB drives each, then linked the two servers with 10GiB ethernet and put a redundancy between the block devices, then on top of that RAID5 2019-10-31T01:55:39 < mawk> then iSCSI for accessing data 2019-10-31T01:55:45 < mawk> + a vpn under that for authentication 2019-10-31T01:55:48 < mawk> and voilà 2019-10-31T01:56:02 < mawk> got paid around 1000€ 2019-10-31T01:56:09 < kakinull> sounds hard 2019-10-31T01:56:28 < mawk> I spent most of my time waiting for RAID array building 2019-10-31T01:56:33 < mawk> and researching iSCSI docs 2019-10-31T01:57:08 < kakinull> oh wow windows has iSCSI initiator 2019-10-31T01:57:15 < kakinull> I might try that tonight 2019-10-31T01:57:25 < mawk> yes 2019-10-31T01:58:43 < kakinull> what is radius? 2019-10-31T01:59:25 < mawk> auth server from MS 2019-10-31T01:59:43 < mawk> it's used to validate user:pass couples 2019-10-31T01:59:53 < mawk> for authenticated ethernet for instance, or wifi enterprise 2019-10-31T02:00:07 < mawk> it's like baby kerberos 2019-10-31T02:00:12 < buZz> radius isnt from MS 2019-10-31T02:00:17 < mawk> ah maybe 2019-10-31T02:00:32 < mawk> there's a radius server builtin in Windows server though 2019-10-31T02:00:56 < buZz> its by some ISP from 1991 2019-10-31T02:01:01 < mawk> I see 2019-10-31T02:01:09 < buZz> sure, its mandatory for many enterprise network setups 2019-10-31T02:01:19 < buZz> windos built a server for it aswell 2019-10-31T02:01:50 < kakinull> does iSCSI have any auth / encryption itself? 2019-10-31T02:01:52 < mawk> I used that for the wifi enterprise network of a company, it's pretty cool when integrated with windows logon 2019-10-31T02:02:04 < buZz> kakinull: i dont think so 2019-10-31T02:02:07 < mawk> probably something but we didn't use it 2019-10-31T02:02:13 < mawk> if it exists it's very basic 2019-10-31T02:02:18 < mawk> hence the VPN we used ourselves 2019-10-31T02:02:21 < buZz> wpa2 enterprise is nice yeah 2019-10-31T02:02:28 < buZz> especially with decentral auth 2019-10-31T02:02:49 < buZz> you could have miss@yourhome.nl auth to same network as you@yourwork.nl 2019-10-31T02:02:56 < buZz> sry for .nl 2019-10-31T02:03:12 < mawk> don't be sorry we're all dutchlanders here 2019-10-31T02:03:18 < buZz> :) 2019-10-31T02:03:21 < mawk> even kakinull is honorary dutchlander 2019-10-31T02:03:33 < buZz> underlander 2019-10-31T02:03:42 < mawk> I got my birth certificate to apply for a passport at the dutch consulate 2019-10-31T02:03:56 < mawk> they called me "miss" when they sent me the certificate 2019-10-31T02:04:02 < buZz> nice, i got my birth certificate from the gemeentehuis in amsterdam 2019-10-31T02:04:15 < buZz> :) 2019-10-31T02:04:23 < kakinull> can windows use bitlocker for iscsi drive? 2019-10-31T02:04:39 < mawk> I'm dutch but no dutch ID whatsoever, so I must find a bunch of documents to prove I'm dutch 2019-10-31T02:04:47 < buZz> maybe thats a question for ##windows , kakinull 2019-10-31T02:04:47 < mawk> I think kakinull , wouldn't surprise me 2019-10-31T02:05:00 < buZz> mawk: marry someone 2019-10-31T02:05:01 < buZz> ez pz 2019-10-31T02:05:18 < mawk> I'm already dutch, I just have to get all the docs 2019-10-31T02:05:36 < buZz> step into the treadmill of bureaucrazy 2019-10-31T02:05:39 < mawk> I found out if your mother is dutch you're not dutch, and she loses her Dutch nationality, even (before 1985) 2019-10-31T02:05:41 < buZz> maybe you have a spare year? 2019-10-31T02:05:45 < mawk> lol 2019-10-31T02:06:13 < buZz> i'm in it for ~3 months for ~something now, and havent gotten past introductionary stages yet 2019-10-31T02:06:28 < buZz> and met ~12 people 2019-10-31T02:06:39 < mawk> they must justify their salary 2019-10-31T02:06:48 < buZz> maybe 2 of em, i've had >1 talk with 2019-10-31T02:07:35 -!- jadew [~rcc@unaffiliated/jadew] has quit [Ping timeout: 276 seconds] 2019-10-31T02:13:07 < Jan-> free gear! 2019-10-31T02:13:12 < Jan-> two stm32f407 dev boards, one possibly broken, and two stlink programmers 2019-10-31T02:13:17 < Jan-> receiver pays postage from london 2019-10-31T02:14:16 < kakinull> veracrypt does iscsi 2019-10-31T02:14:21 < mawk> ??¿¿ 2019-10-31T02:14:23 < mawk> ah 2019-10-31T02:14:32 < mawk> yeah, iSCSI presents a raw block device kakinull 2019-10-31T02:14:45 < mawk> it's exactly like a regular hard drive for all intents and purposes 2019-10-31T02:14:45 < kakinull> Jan-: you back 2019-10-31T02:14:53 < mawk> so what would be surprising is if veracrypt wouldn't do iSCSI 2019-10-31T02:14:58 < mawk> omg Jan- you're surrendering 2019-10-31T02:15:08 < mawk> you're disappointing all of us 2019-10-31T02:15:47 < Jan-> my heart bleeds 2019-10-31T02:16:06 < mawk> I hope 2019-10-31T02:16:12 < mawk> go back while you still can 2019-10-31T02:16:19 < Jan-> I ordered ESP32 2019-10-31T02:16:27 < karlp> that will be heaps more fun... 2019-10-31T02:16:28 < Jan-> on the basis it actually has API documentation 2019-10-31T02:16:38 < mawk> ........... 2019-10-31T02:16:47 < karlp> on your bike troll 2019-10-31T02:16:52 < mawk> you order a counterfeit of stm32 because you expect it has better docs ??? 2019-10-31T02:16:58 < mawk> I hope it's a joke Jan- 2019-10-31T02:17:00 < Jan-> well it actually has better docs. 2019-10-31T02:17:04 < Jan-> I have the docs. 2019-10-31T02:17:13 < Jan-> anyway, legitimate offer, stm32 gear free to good home, take it or leave it 2019-10-31T02:17:19 -!- friendofafriend [~ian@75.182.67.149] has quit [Quit: Lost terminal] 2019-10-31T02:17:36 < mawk> shady chinese counterfeit will be better supported than a popular chip of a popular western european well established brand ?? 2019-10-31T02:17:37 < kakinull> ask catphish when he is around 2019-10-31T02:17:38 < Jan-> it's that or it's going in the waste electronics bin 2019-10-31T02:17:50 < kakinull> he is uk 2019-10-31T02:17:59 < Jan-> mawk I don't think it's a counterfeit of anything, is it? 2019-10-31T02:18:00 < mawk> you can mail it to me Jan- 2019-10-31T02:18:05 < mawk> how long does it take ? 2019-10-31T02:18:17 < mawk> well it's maybe a clean room copy if you want 2019-10-31T02:18:21 < mawk> but still, a copy of stm32 2019-10-31T02:18:33 < kakinull> seriously mawk? 2019-10-31T02:18:42 < mawk> what kakinull ? 2019-10-31T02:18:43 < Jan-> yeah seriously mawk it isn't 2019-10-31T02:18:48 < Jan-> an ESP32 has a lot less peripherals 2019-10-31T02:18:49 < kakinull> copy of smt? 2019-10-31T02:19:03 -!- Netsplit *.net <-> *.split quits: forrestv, futarisIRCcloud, ohama, Jan-, zoobab, turnip420, Laurenceb, mawk, Cracki, marble_visions_, (+129 more, use /NETSPLIT to show all of them) 2019-10-31T02:19:03 -!- varesa [~varesa@ec2-52-49-18-111.eu-west-1.compute.amazonaws.com] has quit [Max SendQ exceeded] 2019-10-31T02:26:13 -!- Netsplit over, joins: srk, [7], learningc, hackkitten, scrts2, mrec, specing, aandrew, forrestv, bvernoux (+129 more) 2019-10-31T02:26:13 -!- bvernoux [~Ben@chl26-1-88-183-106-15.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2019-10-31T02:26:14 < Cracki> when I let my front door fall shut without having the key, I write off my whole home. there's no way to get back in. 2019-10-31T02:26:24 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has quit [Ping timeout: 265 seconds] 2019-10-31T02:26:30 < Jan-> Fine, take it, fix it, whatever 2019-10-31T02:28:04 -!- varesa [~varesa@ec2-52-49-18-111.eu-west-1.compute.amazonaws.com] has joined ##stm32 2019-10-31T02:30:43 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-31T02:33:11 -!- jadew [~rcc@79.115.58.118] has joined ##stm32 2019-10-31T02:33:11 -!- jadew [~rcc@79.115.58.118] has quit [Changing host] 2019-10-31T02:33:11 -!- jadew [~rcc@unaffiliated/jadew] has joined ##stm32 2019-10-31T02:35:55 < kakinull> is iscsi share fixed size? 2019-10-31T02:36:13 < kakinull> or can client trim it? 2019-10-31T02:36:24 < doomba> kakiSome(T) 2019-10-31T02:36:56 < kakinull> I run out of space today if I start making iscsi shares 2019-10-31T02:39:54 < Cracki> iscsi... do you expect an scsi command that enlarges a disk? 2019-10-31T02:40:16 < Cracki> I would expect that to only be possible from the iscsi host 2019-10-31T02:40:48 < kakinull> no 2019-10-31T02:42:02 < Cracki> there's this. maybe they use their words wrong. maybe there is an iscsi command that does it... https://docs.microsoft.com/en-us/powershell/module/iscsitarget/resize-iscsivirtualdisk?view=win10-ps 2019-10-31T02:42:52 < Cracki> ah, target = host 2019-10-31T02:48:31 < jadew> look how hot these two chicks are: https://www3-cdn.sherdog.com/_images/pictures/20170413092708_A44I1133.JPG 2019-10-31T02:48:38 < jadew> (and both of them can beat you senseless) 2019-10-31T02:49:42 < kakinull> they look violent but not muscular 2019-10-31T02:49:52 < jadew> I know, right? 2019-10-31T02:50:18 < jadew> and the one the left is the welterweight champion of the world in muay thai (like kickboxing, but with more ways of hitting) 2019-10-31T02:50:18 < Cracki> I expected deep fried chicken 2019-10-31T02:50:39 < Cracki> show me buckets of KFC 2019-10-31T02:50:50 < jadew> I wouldn't mind some KFC right now 2019-10-31T02:52:21 < jadew> they're hotter than the ring girls 2019-10-31T02:52:40 < jadew> the champ is 2019-10-31T02:53:42 < Cracki> how frumpy are the ring girls 2019-10-31T02:54:06 < jadew> eh, I'm kidding 2019-10-31T02:54:10 < jadew> they're hot 2019-10-31T02:54:56 < Cracki> have you checked their pulse 2019-10-31T02:55:28 < jadew> they seem very relaxed 2019-10-31T02:57:43 < Cracki> I think you need a few more wives 2019-10-31T02:57:48 < jadew> made a mistake and looked up her record, thought to myself I won't look at the names, so I wouldn't know the outcome when I'm watching a fight 2019-10-31T02:57:51 < jadew> she won them all :/ 2019-10-31T02:58:16 < jadew> Cracki, I tried to pitch that to my wife, she's not ok with it 2019-10-31T02:58:27 -!- smvoss [~smvoss@207.191.220.92] has quit [Ping timeout: 240 seconds] 2019-10-31T02:58:46 < Cracki> make moar kids. you sound horny enough :> 2019-10-31T02:59:40 < jadew> I just appreciate beauty 2019-10-31T02:59:59 < jadew> but yeah, that too 2019-10-31T03:01:54 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2019-10-31T03:02:51 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has quit [Ping timeout: 268 seconds] 2019-10-31T03:05:57 < Thorn> https://caseyhandmer.wordpress.com/2019/10/29/the-spacex-starship-is-a-very-big-deal/ 2019-10-31T03:06:44 -!- Rajko [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-31T03:08:07 < Cracki> is that 5000 words of "wow"? 2019-10-31T03:08:46 < Cracki> ah, the typical "we can imagine flying cars, but not that someone would invent the laser" 2019-10-31T03:09:20 < jadew> lol Cracki 2019-10-31T03:09:47 < jadew> yeah, it it seems like a paid article 2019-10-31T03:10:23 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has quit [Ping timeout: 265 seconds] 2019-10-31T03:10:27 < Cracki> I'm not seeing a structure to that article yet 2019-10-31T03:10:37 < Cracki> >Starship is a devastatingly powerful space access and logistical transport mechanism that will instantly crush the relevance of every other rocket ever built. 2019-10-31T03:10:43 < Cracki> yeah ok, we knew that 2019-10-31T03:10:50 < Cracki> it's not a horseless carriage, it's more than that 2019-10-31T03:11:05 < Cracki> 640k, the world needs at most five computers, etc. 2019-10-31T03:11:46 < Cracki> his only mistake is picking _one particular_ horseless carriage and saying _that_ is revolutionary 2019-10-31T03:11:54 < jadew> maybe they'll do one day trips to space in our lifetimes 2019-10-31T03:11:59 < jadew> that would be pretty cool 2019-10-31T03:12:06 < Cracki> they'll do that within 10 years 2019-10-31T03:12:19 < Cracki> within 5 years you'll have people paying a million or so to fly up 2019-10-31T03:12:30 < jadew> I need it to be a bit cheaper than that 2019-10-31T03:12:39 < Cracki> all those companies with shit public relations already send people up to 100 km and futher 2019-10-31T03:13:02 < jadew> think we'll be able to visit the moon? 2019-10-31T03:13:04 < Cracki> fermi calculation: price will drop by order of magnitude every 10 years 2019-10-31T03:13:05 < jadew> than would be amazing 2019-10-31T03:13:09 < Laurenceb> wait wut 2019-10-31T03:13:13 < Cracki> you won't, you're not murican 2019-10-31T03:13:16 < Laurenceb> did Musk write that? 2019-10-31T03:13:25 < Cracki> are you taking drugs? 2019-10-31T03:13:29 < Laurenceb> oh I see 2019-10-31T03:13:38 < specing> starship wont crush anything 2019-10-31T03:13:47 < Laurenceb> I was going to say, didnt think he was that bad 2019-10-31T03:13:53 < specing> I don't think it'll be used to transport anything into space 2019-10-31T03:14:00 < specing> its just a can for interplanetary travel 2019-10-31T03:14:10 < Laurenceb> orbital cannabis farm 2019-10-31T03:14:11 -!- smvoss [~smvoss@207.191.220.92] has joined ##stm32 2019-10-31T03:14:14 * Laurenceb patents 2019-10-31T03:14:14 < specing> a rather large can 2019-10-31T03:14:20 < Cracki> the revolution happening right now is that so many companies take shots at their own launch vehicles 2019-10-31T03:14:22 -!- pennTeller [~pennTelle@unaffiliated/pennteller] has joined ##stm32 2019-10-31T03:14:37 < jadew> specing, they're probably building two. when this one explodes, they'll use the secret one to go and visit the aliens 2019-10-31T03:14:47 < Laurenceb> starship a is retarded name 2019-10-31T03:14:53 < Cracki> a million monkeys on a million typewriters. every company can try something different, and what works will stick around. 2019-10-31T03:15:06 < Laurenceb> interplanetary transport system was at least accurate 2019-10-31T03:15:23 < Cracki> before spacex nobody considered propulsive landing. they put the engineering into it and made it work. they'll make refueling in space work too. 2019-10-31T03:15:24 < Laurenceb> should be called Autistic Latina Thot Hareem Launcher 2019-10-31T03:15:33 < Laurenceb> ALTHL 2019-10-31T03:15:51 < Laurenceb> that would reflect what it will actually be used for 2019-10-31T03:15:57 < jadew> Cracki, apparently NASA already did that before spacex 2019-10-31T03:15:58 < Cracki> aim for the sun 2019-10-31T03:16:09 < Cracki> yes, i saw footage that looked 60s-era 2019-10-31T03:16:10 < jadew> not sure why they didn't take it further tho 2019-10-31T03:16:18 < Cracki> buuuut for some reason people scrapped it 2019-10-31T03:16:31 < Cracki> the propulsive landing of Crew Dragon got axed too because nasa forced them to 2019-10-31T03:16:37 < Cracki> nasa wants good old parachutes 2019-10-31T03:16:43 < jadew> there's probably more money in throwing things away and building them all over again 2019-10-31T03:17:08 < Cracki> they're pushing away from innovation, towards spacex being their domestic alternative to russian capsules 2019-10-31T03:17:28 < kakinull> *propulsive landing on earth 2019-10-31T03:17:32 < Cracki> spacex will continue to develop propulsive landing for free market customers 2019-10-31T03:17:48 < Cracki> there's more money in reusing things. that's what spacex keeps demonstrating. 2019-10-31T03:18:09 < BrainDamage> a lot of nasa drive for development was to allow tech to be reusable for military development 2019-10-31T03:18:31 < BrainDamage> eg the shuttle capability to recover sats was meant to snatch enemy sats 2019-10-31T03:18:48 < BrainDamage> in practice this was used once or twice, I forgot 2019-10-31T03:18:57 < jadew> really? that sounds great 2019-10-31T03:19:13 < jadew> I guess now they boobietrap them? 2019-10-31T03:19:27 < BrainDamage> and all parts were sub-contracted to companies which were also sourcing military 2019-10-31T03:19:36 < BrainDamage> as consequnce shuttle costs were insane 2019-10-31T03:19:48 < BrainDamage> no, now they just aim an anti sat missile 2019-10-31T03:20:04 < Laurenceb> X-37 2019-10-31T03:20:36 < jadew> yeah, but I mean if you have a satellite that does something fancy, you wouldn't want anyone to be able to just snatch it from orbit 2019-10-31T03:20:50 < jadew> so you make it blow up if it detects it's not on course anymore 2019-10-31T03:21:01 < jadew> and it destroys the ship in the process too 2019-10-31T03:21:36 < Laurenceb> god this guy is annoying 2019-10-31T03:21:40 < Laurenceb> >not a thing 2019-10-31T03:21:40 < BrainDamage> oh sure, they likely have self destruct procedures now 2019-10-31T03:21:46 < Laurenceb> fucking buzzfeed man 2019-10-31T03:22:01 < Laurenceb> kill, kill, kill 2019-10-31T03:23:57 < Cracki> you give buttfeed power by taking it seriously 2019-10-31T03:24:02 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-31T03:24:07 < Cracki> they're all hacks and anyone seriously reading it is a moron 2019-10-31T03:24:13 < BrainDamage> also, both nasa and russian space agencies were mostly funded during the cold war, funding that was withdrawed later on 2019-10-31T03:24:38 < Cracki> these failures get laid off and will be sitting on san francisco streets shooting up in no time 2019-10-31T03:24:41 < BrainDamage> so they're mostly operating with now's 30 year old tech 2019-10-31T03:25:02 < Cracki> someone must find helium 3 on the moon. then they'll race again 2019-10-31T03:25:23 < BrainDamage> granted, not everything is 30yo tech, but some of the things they do require massive investments 2019-10-31T03:25:37 -!- Rajko [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has quit [Ping timeout: 240 seconds] 2019-10-31T03:31:31 < BrainDamage> oh another thing, the remaining space agency (ESA), is a pile of weird bureacracy where they try to appease each partecipating member by subcontracting them work rather than just buy at market cost 2019-10-31T03:31:42 < BrainDamage> hence costs are stupidly inflated as well 2019-10-31T03:32:33 < jadew> well.. maybe by doing that costs are much lower 2019-10-31T03:32:56 < BrainDamage> except not, when you buy at market costs you allow competition 2019-10-31T03:32:58 < jadew> since some of the money they invest comes back into the country 2019-10-31T03:33:39 < BrainDamage> spacex is not massively innovative, rather, it's everyone else that's mostly shit for one reason or the other 2019-10-31T03:33:44 < jadew> yeah, but I mean if you invest 100m and you make 50 back, it's like investing only 50m and also creating some jobs 2019-10-31T03:33:57 < jadew> while you might not get the same parts for 50m otherwise 2019-10-31T03:34:08 < specing> BrainDamage: they but at market costs, but only member states can participate in said market? 2019-10-31T03:34:38 < BrainDamage> no, they pretty much subcontract stuff to give a fair share of partecipation to each 2019-10-31T03:34:51 < BrainDamage> neglecting costs 2019-10-31T03:35:00 < jadew> yeah... might not be that bad after you draw the line 2019-10-31T03:35:19 < Laurenceb> he has been smoking the musk coolaid 2019-10-31T03:35:22 < Laurenceb> https://caseyhandmer.wordpress.com/2019/09/20/no-really-space-based-solar-power-is-not-a-useful-idea-literature-review-edition/ 2019-10-31T03:35:39 < Cracki> fuck all that, use nuclear 2019-10-31T03:35:41 < BrainDamage> me? no, spacex is also shit for other reasons 2019-10-31T03:35:44 < Laurenceb> musk dislikes space solar power as it makes his other businesses worthless 2019-10-31T03:36:28 < BrainDamage> my argument is that the market is somewhat open to anyone willing to invest the money 2019-10-31T03:36:33 < jadew> Laurenceb, last time I heard about space solar power, which was before spacex, the problem was getting it back to earth 2019-10-31T03:36:43 < jadew> have they solved that? 2019-10-31T03:36:49 < Laurenceb> jadew: yeah microwaves sort of work 2019-10-31T03:36:50 < jadew> uBeam maybe? 2019-10-31T03:36:55 < Laurenceb> with a pilot beam 2019-10-31T03:36:58 < Laurenceb> lol uBeam 2019-10-31T03:37:02 < Cracki> microwave... sounds like simcity 3000 2019-10-31T03:37:08 < Laurenceb> I wonder what happened to uBeam 2019-10-31T03:37:22 < Laurenceb> anyway, stratosolar ftw 2019-10-31T03:37:26 < jadew> Laurenceb, they almost went belly up and started focusing on different markets I think 2019-10-31T03:37:28 < BrainDamage> I want a uBeam with high density link 2019-10-31T03:37:37 < Laurenceb> uTaser 2019-10-31T03:37:41 < BrainDamage> so you can use it as a satellite weapon 2019-10-31T03:37:45 < jadew> Laurenceb, I don't think microwaves will work 2019-10-31T03:38:00 < Laurenceb> jadew: sure,but previously launch was too expensive 2019-10-31T03:38:02 < jadew> maybe with a laser 2019-10-31T03:38:07 < Laurenceb> musk has a dilemma 2019-10-31T03:38:20 < BrainDamage> laser losses tend to be worse than microwaves 2019-10-31T03:38:21 < Cracki> megawatt per square meter, I think that'll fry humans quick enough 2019-10-31T03:38:30 < Laurenceb> now launch is cheap but he wants his battery/panel monopoly to hold up 2019-10-31T03:38:33 < jadew> BrainDamage, at all wavelengths? 2019-10-31T03:38:42 < Laurenceb> so he spreads fud about space solar power 2019-10-31T03:38:52 < BrainDamage> jadew: nearly so 2019-10-31T03:38:55 < jadew> BrainDamage, I'd think the problem with microwaves is that they'd scatter 2019-10-31T03:38:59 < Laurenceb> wut 2019-10-31T03:39:09 < BrainDamage> jadew: and laser no? 2019-10-31T03:39:10 < Laurenceb> the panel is kind of large 2019-10-31T03:39:20 < Laurenceb> so a phased array is very focussed 2019-10-31T03:39:20 < BrainDamage> jadew: raileight scattering 2019-10-31T03:39:20 < jadew> BrainDamage, it would scatter less 2019-10-31T03:39:51 < Laurenceb> ^full retard 2019-10-31T03:40:18 < BrainDamage> scattering is pretty much proportional to freq 2019-10-31T03:40:23 < Laurenceb> short wavelength light scatters much more than microwaves 2019-10-31T03:40:41 < mawk> if I can read all sectors of a mifare classic NFC tag that means I can 100% clone it ? 2019-10-31T03:40:54 < BrainDamage> ( it's also why air is blue ) 2019-10-31T03:40:59 < mawk> and be it undistinguishable by the reader 2019-10-31T03:41:10 < kakinull> what type of smart test I want to run on my server? 2019-10-31T03:41:14 -!- Rajko [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-31T03:41:19 < mawk> I looked at commercial docs of the manufacturer of that tag and they say it's impossible to clone but I don't get why 2019-10-31T03:41:29 < BrainDamage> kakinull: short 2019-10-31T03:41:40 < mawk> I tried 2 tags for the same building entrance and they use the exact same set of keys 2019-10-31T03:41:50 < kakinull> thx BrainDamage 2019-10-31T03:42:04 < mawk> the last sector even contains a "do not reproduce ! copyrighted !" text 2019-10-31T03:42:21 < Cracki> they are supposed to do crypto 2019-10-31T03:42:29 < kakinull> daily or weekly BrainDamage? 2019-10-31T03:42:32 < Cracki> but classics are borken 2019-10-31T03:42:33 < Laurenceb> >Additionally, more than 2/3 of current production growth is in renewables, whose installation costs are falling at up to 10%/year 2019-10-31T03:42:37 < BrainDamage> kakinull: weekly is fine 2019-10-31T03:42:39 < Cracki> desfires have DES at least 2019-10-31T03:42:40 < Laurenceb> Casey Handmer is a tard 2019-10-31T03:42:40 < mawk> but it's exact same data Cracki and I have all the keys 2019-10-31T03:42:44 < mawk> so I can clone right ? 2019-10-31T03:42:51 < Laurenceb> that doesnt account for capacity factor 2019-10-31T03:42:58 < Cracki> try it? no clue what functions they have beyond data storage 2019-10-31T03:43:14 < mawk> the blank tags are on their way, they arrive tomorrow by amazon prime 2019-10-31T03:43:17 < BrainDamage> kakinull: smart pretty much catches slow degradation, and even then, 1/3 of the drives will fail without any parameter in smart going off 2019-10-31T03:43:30 < Cracki> they are supposed to do some crypto, so figure out what they use for that 2019-10-31T03:43:34 < mawk> the app I found bruteforces the tag to find the keys, using a huge db of leaked tag encryption keys 2019-10-31T03:43:45 < mawk> "mifare classic tool" 2019-10-31T03:44:01 < Cracki> if their crypto isn't used, but they're basically wireless usb drives, ¯\_(ツ)_/¯ 2019-10-31T03:44:22 < mawk> I hope 2019-10-31T03:44:38 < mawk> they're just there to open a door, so it wouldn't surprise me 2019-10-31T03:44:42 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has quit [Ping timeout: 265 seconds] 2019-10-31T03:44:52 < Cracki> anyway, yes, their crypto is weaker than the desfire ones, so it's likely that you can bruteforce anything that's not immediately accessible 2019-10-31T03:45:27 < BrainDamage> kakinull: oh and don't get confused like everyone, the column "type" is not an on/off switch, it's just a label 2019-10-31T03:45:45 < BrainDamage> it's the "when failed" column that tells you that a flag triggered 2019-10-31T03:47:03 < BrainDamage> http://ix.io/20nM 2019-10-31T03:47:32 < BrainDamage> the value column counts backwards from 200/255 to 0 2019-10-31T03:47:44 < BrainDamage> sorry, 100/200/255 to 0 2019-10-31T03:47:55 < kakinull> resilver priority page? it sets to resilver between begin and end time instead of whenever? 2019-10-31T03:50:50 < Laurenceb> >One article that should be considered for this article is your own “The SpaceX Starship is a very big deal,” which says that SpaceX could soon be launching large amounts of cargo to LEO for $35/kg. 2019-10-31T03:50:56 < Laurenceb> keek destroyed in the comment 2019-10-31T03:52:01 < jadew> had to look it up, I'm dumb: https://en.wikipedia.org/wiki/Scattering#Electromagnetic_scattering 2019-10-31T03:52:28 < Laurenceb> haha oh wow 2019-10-31T03:52:38 < Laurenceb> Casey Handmer works for hyperloop 2019-10-31T03:53:02 < Laurenceb> no wonder he is an idiot musk fan 2019-10-31T03:53:38 < jadew> well, is it that bad that he promotes spacex? 2019-10-31T03:53:54 < Laurenceb> it is bad that he is a brainless fanboy 2019-10-31T03:54:00 < Cracki> you are arguing with a britbong 2019-10-31T03:54:04 < Laurenceb> not ban for Musks bank balance of course 2019-10-31T03:54:04 < jadew> sure, Musk will get filthy rich, but there are some cool stuff being developed in the process 2019-10-31T03:54:09 < Laurenceb> *bad 2019-10-31T03:54:40 < jadew> and there's nothing wrong with being filthy rich to begin with 2019-10-31T03:55:30 < Laurenceb> >quora top writer 2019-10-31T03:55:32 < Laurenceb> oh god 2019-10-31T03:55:39 -!- pennTeller [~pennTelle@unaffiliated/pennteller] has quit [Ping timeout: 240 seconds] 2019-10-31T03:56:53 < Laurenceb> oh wow keeeek https://www.bbc.co.uk/news/av/stories-50223895/could-electric-roads-spark-a-green-transport-revolution 2019-10-31T03:57:02 < Laurenceb> Elon Musk on suicide watch 2019-10-31T03:57:31 < BrainDamage> is there a chance for you to communicate like a human being? 2019-10-31T03:58:13 < Laurenceb> >Pantograph on a truck is the thing of Musks nighmares 2019-10-31T03:58:31 < Laurenceb> his bank balance: destroyed 2019-10-31T03:58:39 * jadew thinks Laurenceb is thunderf00t 2019-10-31T03:58:45 < Laurenceb> lol 2019-10-31T03:59:09 < Laurenceb> Thunderf00t versus Sargon: fight of the britbongs 2019-10-31T03:59:41 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-31T04:00:15 < englishman> didn't sim city 2000 have space solar 2019-10-31T04:00:36 < englishman> then the beam would misalign and go around destroying your city 2019-10-31T04:00:44 < Laurenceb> lmao 2019-10-31T04:01:01 < Laurenceb> pilot beam solves that problem 2019-10-31T04:01:25 < kakinull> Laurenceb: that will happen 2019-10-31T04:01:28 < kakinull> electric roads 2019-10-31T04:01:49 < kakinull> no bullshit just overhead lines 2019-10-31T04:01:51 < englishman> lol the trolleybus is your solution Laurent? 2019-10-31T04:01:58 < Laurenceb> Musk will probably hang himself off one of the poles 2019-10-31T04:02:18 < Laurenceb> englishman: no but if it triggers Musk epic style (it will), then good 2019-10-31T04:02:44 < kakinull> problem is that for sedan you would need 5meter pantogram 2019-10-31T04:02:46 < englishman> you seem to care deeply about the feelings of a person you have never met 2019-10-31T04:03:17 < Laurenceb> too much time on /cow/ 2019-10-31T04:13:05 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Remote host closed the connection] 2019-10-31T04:13:28 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-31T04:39:02 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has quit [Ping timeout: 276 seconds] 2019-10-31T04:40:19 < kakinull> https://www.youtube.com/watch?v=d-ruFmgTpqA 2019-10-31T05:11:04 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-31T05:23:55 -!- pennTeller [~pennTelle@unaffiliated/pennteller] has joined ##stm32 2019-10-31T05:24:25 < mawk> Cracki: apparently "mifare classic" relied in security by obscurity 2019-10-31T05:24:29 < mawk> but it has been reverse engineered 2019-10-31T05:24:43 < mawk> so since I have the encryption keys for my building I can clone the card 2019-10-31T05:25:04 < mawk> in France there is a nation-wide system called Vigik, designed by the post office to let firemen and postmen enter buildings without the key 2019-10-31T05:25:20 < mawk> if someone could clone the tag used by them it would be instant access everywhere 2019-10-31T05:25:50 < mawk> previously they were using physical keys, also to open mailboxes to deliver mail, and these were cloned and sold illegally since long ago 2019-10-31T05:27:16 < Cracki> :) 2019-10-31T05:31:12 < BrainDamage> if it's not pubic key auth, you could just get your hands on one of the readers 2019-10-31T05:31:35 < mawk> in one commercial doc they claim all the readers are connected to a central station or something 2019-10-31T05:31:43 < mawk> and they mention public key crypto iirc, let me find it back 2019-10-31T05:32:54 < mawk> yeah they talk about RSA 2019-10-31T05:35:17 < mawk> lol I found an unprotected directory listing with all their internal documentation and programming software http://notices.hexact.fr/ 2019-10-31T05:35:22 < mawk> this sounds promising 2019-10-31T05:35:25 < mawk> let's wget -r this 2019-10-31T05:41:03 < dongs> hmm my mikrodik audience isnt booting 2019-10-31T05:44:22 < Thorn> musics https://www.youtube.com/watch?v=eclMFa0mD1c 2019-10-31T05:54:09 < dongs> oh lol it only made a fucking wifi interface to conect to it and no lan 2019-10-31T05:54:10 < dongs> disgusting 2019-10-31T05:57:27 < BrainDamage> mawk: if it's two way com to a central station, a compromised key may get revoked 2019-10-31T05:57:54 < BrainDamage> and it'd be kind of easy to automate alarms based off key usage, since there'd be no emergency reported to the place 2019-10-31T05:58:14 < mawk> yes 2019-10-31T06:08:44 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has quit [Ping timeout: 265 seconds] 2019-10-31T06:14:50 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2019-10-31T06:14:57 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2019-10-31T06:15:35 < dongs> wow th3e new "quickmount" mikrodik shit is fucking assinine 2019-10-31T06:15:38 -!- dan2wik [dan2wik@hellomouse.net] has joined ##stm32 2019-10-31T06:15:38 -!- dan2wik [dan2wik@hellomouse.net] has quit [Changing host] 2019-10-31T06:15:38 -!- dan2wik [dan2wik@unaffiliated/dan2wik] has joined ##stm32 2019-10-31T06:46:42 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Leaving] 2019-10-31T06:47:00 -!- fc5dc9d4 [~quassel@p5B08147D.dip0.t-ipconnect.de] has joined ##stm32 2019-10-31T06:51:38 -!- fc5dc9d4_ [~quassel@p57A3217C.dip0.t-ipconnect.de] has quit [Ping timeout: 268 seconds] 2019-10-31T07:11:06 -!- day__ [~Unknown@unaffiliated/day] has joined ##stm32 2019-10-31T07:15:02 -!- day [~Unknown@unaffiliated/day] has quit [Ping timeout: 276 seconds] 2019-10-31T07:15:03 -!- day__ is now known as day 2019-10-31T07:46:56 -!- jly [uid355225@gateway/web/irccloud.com/x-owliqwovtsybcvuq] has joined ##stm32 2019-10-31T07:57:08 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has joined ##stm32 2019-10-31T08:00:32 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 276 seconds] 2019-10-31T08:01:16 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-31T08:11:35 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 276 seconds] 2019-10-31T08:21:35 < Thorn> I need a diode that can withstand a 18650 short circuit current until the fuse blows. how do I select one 2019-10-31T08:21:48 < Thorn> and how do I make sure the fuse blows before the diode 2019-10-31T08:22:04 < Thorn> because voltage drop on the diode will be much larger 2019-10-31T08:22:41 < jly> h ithr 2019-10-31T08:23:33 < qyx> to247 to the rescue then 2019-10-31T08:23:58 < jly> what is that massive diode that is bigger than someones hand 2019-10-31T08:24:03 < jly> put that in too 2019-10-31T08:24:27 -!- irf21k [~irf21k@106.206.97.118] has joined ##stm32 2019-10-31T08:34:59 < Thorn> maybe I should take a TVS and use it as a diode, they have guaranteed energy specs 2019-10-31T08:35:05 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has quit [Quit: Gone to sleep...] 2019-10-31T08:35:55 -!- Rajko [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has quit [Read error: Connection reset by peer] 2019-10-31T08:38:58 < qyx> reverse polarity protection? 2019-10-31T08:39:28 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-31T08:40:35 < qyx> what about thjis? https://circuitdigest.com/sites/default/files/inline_users/u1759/JD3LH.jpg 2019-10-31T08:41:16 < BrainDamage> was about to suggest a mosfet too 2019-10-31T08:41:28 < BrainDamage> way less drop and can withstand huge currents 2019-10-31T08:44:55 < dongs> reverse protection? yeah fet like that is waht i use 2019-10-31T08:46:12 < jly> i'll just stick a diode bridge in \o/ (dumb) 2019-10-31T08:48:27 < Thorn> there is a little problem 2019-10-31T08:48:31 < Thorn> battery needs charging 2019-10-31T08:49:07 < qyx> why is it a problem? 2019-10-31T08:49:22 < jly> yeah i was struggling to think of the problem tbh 2019-10-31T08:49:26 < Thorn> that mosfet circuit (and a diode bridge for that matter) don't protect from reverse voltage 2019-10-31T08:49:40 < jly> my diode bridge was a joke for a start (pumped) 2019-10-31T08:49:43 < Thorn> they protect from reverse flow of current 2019-10-31T08:50:01 < Thorn> which is what happens when you're charging the battery 2019-10-31T08:50:25 < qyx> the mosfet conducts both ways 2019-10-31T08:50:37 < qyx> but I am not a mosfet pro 2019-10-31T08:50:47 < BrainDamage> yes, the mosfet will conduct both ways 2019-10-31T08:52:59 < Thorn> why are people recommending these complicated circuits then http://blog.deconinck.info/post/2017/12/22/18650-Battery-charger-reverse-polarity-protection 2019-10-31T08:54:28 < jly> i'd have to build them and autism my own conclusion 2019-10-31T09:00:09 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-31T09:01:24 < Thorn> https://electronics.stackexchange.com/questions/213961/is-it-possible-to-use-a-mosfet-for-li-ion-reverse-polarity-protection-when-prote 2019-10-31T09:01:48 -!- irf21k [~irf21k@106.206.97.118] has quit [Remote host closed the connection] 2019-10-31T09:19:44 < jpa-> Thorn & others: the problem with the simple mosfet circuit in a charger is that the charger side will always be correct way around, thus it will activate the mosfet and if you then connect a battery wrong way around, it won't disconnect until the charger voltage drops to below 1V 2019-10-31T09:20:27 < jpa-> i guess if the charger has current limit and can handle that amount of voltage dropping and the battery doesn't mind either, it might work 2019-10-31T09:22:02 < Thorn> charger is always supposed to have a current limit, but it's not rated for any reverse voltage at the bat pin (maybe -0.3V) 2019-10-31T09:22:20 < Thorn> if it's not charging and the battery is reversed, then what 2019-10-31T09:23:27 < jpa-> if it is not charging, then mosfet G-S voltage is 0V, mosfet is off, and the negative voltage from battery won't get through it (blocked by the parasitic diode) 2019-10-31T09:25:16 < jpa-> (ah, further problem: even if the charger has current limit, it will cause the mosfet to stay in a half-on state, just so that S will remain at ~2V - heating the mosfet and draining the battery) 2019-10-31T09:25:26 < PaulFertser> Sidenote: I have a cheap chinese 18650 charger and it's somehow autodetecting the polarity and charges either way. 2019-10-31T09:25:54 < jpa-> time to open it up, steal their superior engineering 2019-10-31T09:26:08 < PaulFertser> I'd open it up right atm but it's at my parents'. 2019-10-31T09:26:30 < Thorn> you can do that if the MCU is powered before the battery h-bridge (I assume) is switched on 2019-10-31T09:26:53 < jpa-> yeah, many of those chargers also feature automatic voltage detection 2019-10-31T09:27:15 < Thorn> i need to power my device from the battery in the absence of external supply (that's what batteries are for, aren't they) 2019-10-31T09:42:25 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2019-10-31T09:49:34 < upgrdman> any opengl pros in here? 2019-10-31T09:49:41 < upgrdman> ##opengl fags are sleeping i guess 2019-10-31T09:49:57 < upgrdman> i think i recall a way to use "variants" (wrong name probably) of a shader? like instead of checking an if(someUniform) in the shader, there was a more efficient way to do it that didn't waste clock cycles on the if 2019-10-31T09:49:58 < upgrdman> right now im using two Programs, but the only difference is one line in my vertex shader, so it seems like a waste. 2019-10-31T09:51:17 < jpa-> i have no idea about shaders, but in opencl you can do #ifdef and then pass option when you compile the kernel 2019-10-31T09:51:36 < jpa-> https://software.intel.com/en-us/blogs/2012/03/26/using-ifdef-in-opengl-es-20-shaders 2019-10-31T09:51:38 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-31T09:58:38 < dongs> what what a completely useless exaple 2019-10-31T09:59:04 < dongs> its like #ifdef COCKS / #endif #ifdef DONGS / #endif but the 2 parts #ifdef'd are compeltely different anyway 2019-10-31T09:59:07 < dongs> waht does this even achieve? 2019-10-31T09:59:11 < dongs> youre not reusing any code or whatever 2019-10-31T09:59:15 < dongs> fucking dumb shit 2019-10-31T09:59:45 < upgrdman> dongs, maybe it's for pros that love to put all of their code into a single giant file. 2019-10-31T09:59:55 < dongs> sup chinaman 2019-10-31T10:00:12 < upgrdman> theres some visual basic software at more work... >20k lines of code, in one file. 2019-10-31T10:00:31 < upgrdman> life is ok 2019-10-31T10:00:36 < dongs> i believe it, cuz VB doesnt really promote multi-file usage 2019-10-31T10:00:43 < dongs> especially VB6 2019-10-31T10:00:44 < dongs> lulz 2019-10-31T10:00:51 < upgrdman> haha ok, TIL 2019-10-31T10:03:17 < jpa-> dongs: yeah, in that case it just achieves putting different types of shaders into same file, because apparently everything just has to be called "main()" 2019-10-31T10:03:40 < dongs> jpa, wouldn't you do like 2019-10-31T10:03:45 < dongs> shader cocks() { } 2019-10-31T10:03:48 < dongs> shader dongs() { } 2019-10-31T10:03:51 < dongs> main() { 2019-10-31T10:04:02 < dongs> #ifdef COCKS cocks(); #else dongs(); #endif 2019-10-31T10:04:31 < dongs> just looks like an awful example 2019-10-31T10:04:55 < jpa-> not sure if that is any clearer; i don't know opengl, but in opencl one can just give the functions names and then call them based on the name, instead of just using "main()" as default 2019-10-31T10:04:58 < upgrdman> slightly related, Intel GPA is pretty cool and easy to use. 2019-10-31T10:05:01 < PaulFertser> "I learned the git concepts once and now I can use this thing. There has never been a situation where I failed so badly that I could not recover from it. And this beautiful and simple concept makes up for a lot of inconsistent UI." http://lucumr.pocoo.org/2015/2/17/ui-and-hidden-consistency/ 2019-10-31T10:05:05 < jpa-> i agree, looks quite silly 2019-10-31T10:05:51 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has joined ##stm32 2019-10-31T10:06:53 < jpa-> some GLSL compilers are pretty stupid though so maybe that's a reason to keep the code stupid also 2019-10-31T10:10:12 < upgrdman> also it seems like they could make runtime optimizations better, at least intel could. 2019-10-31T10:11:54 < upgrdman> if i do something like "gl_position = vec4(someVarying + someUniform, 0, 0, 1);" ... if someUniform is 0, it still wastes time adding 0. like 15% drop in performance compared to a shader without the +someUniform code 2019-10-31T10:12:58 < upgrdman> so i have to glUseProgram(someUniform == 0 ? program1 : program2); on the CPU :( 2019-10-31T10:14:22 < dongs> upgrdman: when you entering a glsl shader demoscene compo 2019-10-31T10:15:19 < upgrdman> dongs, good idea 2019-10-31T10:15:52 < dongs> man what the FUCK IS THIS DETECTING RADAR NONSENSE 2019-10-31T10:16:14 < upgrdman> but for now, im just optimizing my Telemetry Viewer shit. finally switched over to shaders. managed to improve (reduce) CPU usages after lots of profiling and tweaking. 2019-10-31T10:16:24 < dongs> oh, hah nice 2019-10-31T10:16:29 < dongs> lol shaders in java? 2019-10-31T10:16:37 < upgrdman> opengl in java 2019-10-31T10:16:38 < upgrdman> ya 2019-10-31T10:16:39 < dongs> well, i guess it works. cuz minecrap and opengl or awhtever 2019-10-31T10:16:39 < dongs> yeah 2019-10-31T10:17:01 < upgrdman> BBL 2019-10-31T10:17:41 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Leaving] 2019-10-31T10:18:20 < PaulFertser> dongs: some 5GHz WiFi channels mandate the radar detection capabilities so that your equipment wouldn't interfere with radars. 2019-10-31T10:20:00 < dongs> well aware 2019-10-31T10:20:07 < dongs> but itt takes like 10 minutes for that shti 2019-10-31T10:20:09 < dongs> and its not optional 2019-10-31T10:20:11 < dongs> fuck off 2019-10-31T10:20:24 < dongs> also im in the middle of nowhere, i need the [x] leave me alone button 2019-10-31T10:20:48 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has joined ##stm32 2019-10-31T10:27:52 < jly> THE FUCKING DOGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGssssssssssssssssssssssssssssssssssss 2019-10-31T10:27:57 < jly> sen; 2019-10-31T10:32:27 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has joined ##stm32 2019-10-31T10:39:03 -!- Haohmaru [~Haohmaru@195.24.53.110] has joined ##stm32 2019-10-31T10:39:22 < jly> dunno how the fuck that landed here :\\\\ 2019-10-31T10:43:02 < Thorn> >No external resistor pull-up is required, but it may be desirable if multiple interrupt lines are connected. 2019-10-31T10:43:14 < Thorn> this is about an /irq pin 2019-10-31T10:43:30 < Thorn> what does it mean, is it open drain or push pull 2019-10-31T10:44:15 < jly> maybe the internal pull-up is weak 2019-10-31T10:44:16 < jly> idk 2019-10-31T11:14:45 -!- kuldeep [~kuldeep@unaffiliated/kuldeepdhaka] has joined ##stm32 2019-10-31T11:18:08 -!- codyps [~codyps@richard.einic.org] has joined ##stm32 2019-10-31T11:33:38 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has quit [Ping timeout: 245 seconds] 2019-10-31T11:50:40 < dongs> hmmm 2019-10-31T11:52:48 < qyx> dongs: your mikrotik is detecting radars? 2019-10-31T11:56:10 < dongs> yeah they all do since like 2016 2019-10-31T11:56:19 < dongs> if you pick 160mhz channel 2019-10-31T11:58:29 < qyx> and if you select dfs option no-radar-detect in wireless advanced mode? 2019-10-31T11:58:39 < dongs> there isnt one 2019-10-31T11:58:43 < qyx> wut 2019-10-31T11:58:47 < karlp> then it checks the wireelss reg db and sees if you're allowed to... 2019-10-31T11:58:57 < karlp> and ignores your request in jurisdictions that require it :) 2019-10-31T12:00:03 < qyx> interesting, in the past there was an anarchy mode 2019-10-31T12:00:10 < jly> sounds like the government or something 2019-10-31T12:00:23 < dongs> there's no advanced /dfs and i see mikrodik posts complaining about not beaing able to disable it for years noiw 2019-10-31T12:00:35 < dongs> worse, it takes like 10 mins for initail radar scan 2019-10-31T12:00:40 < dongs> 10 fucking minutes lmao 2019-10-31T12:00:55 < dongs> time to switch to nigbiquity 2019-10-31T12:01:58 < qyx> in .sk 802.11h is mandatory and yet I was always alowed to chamnge mode to dynamic freq, no-radar-detect 2019-10-31T12:02:12 < qyx> Ill check later 2019-10-31T12:02:37 < dongs> i mean my shit lets me select country but wikipedo dfs list is retarded 2019-10-31T12:03:42 < dongs> i don' 2019-10-31T12:03:46 < dongs> t even want this shit to hop anything. 2019-10-31T12:03:54 < dongs> im in middle of nowhere, entire 2.4 and 5ghz band is clear here 2019-10-31T12:12:01 < zyp> the ubnt bridge I ran before I got fiber picked a channel at startup, and depending on which channel it picked it could also spend 10 minutes scanning for radar before starting 2019-10-31T12:12:28 < zyp> but apparently that's not required for all channels 2019-10-31T12:17:50 -!- fsasm [~fsasm@62-178-93-7.cable.dynamic.surfer.at] has joined ##stm32 2019-10-31T12:18:21 < Thorn> if I put the mcu into deep sleep where it's not keeping gpio config does it mean I have to put pullups/downs on all output nets to avoid them floating? 2019-10-31T12:20:41 < jly> depends if you put it to sleep on ketamine or something harder 2019-10-31T12:25:07 < jly> once my pic16 entered a k-hole, took 7 weeks to wake that little shit up 2019-10-31T12:32:07 < karlp> this shows what happens: https://www.youtube.com/watch?v=zDdbiKQTE-w 2019-10-31T12:34:26 < qyx> fuk dfs 2019-10-31T12:34:32 < qyx> fuk regulations 2019-10-31T12:34:45 < jly> nwa said something similar 2019-10-31T12:35:01 < jly> https://www.youtube.com/watch?v=MUW_6YkMRF8 2019-10-31T12:35:32 < jly> you see? 2019-10-31T12:37:02 < Steffanx> I cant see 2019-10-31T12:41:10 < jly> ou're the problem! You're the fucking problem you fucking Dr White honkin' jam-rag fucking spunk-bubble! I'm telling you Aitch you keep looking at me I'll put you in the fucking ground, promise you! 2019-10-31T12:41:21 < englishman> 2:30 ad for duckduckgo 2019-10-31T12:41:24 < englishman> wtf 2019-10-31T12:41:27 < jly> hi sir 2019-10-31T12:42:12 < jly> Just been admiring sir ben kingsley's magic 2019-10-31T12:45:00 < buZz> jly: did you see that marvell short? 2019-10-31T12:45:18 < buZz> https://www.imdb.com/title/tt3438640/ 2019-10-31T12:45:25 < buZz> really a amazing performance by him 2019-10-31T12:45:39 < buZz> -l 2019-10-31T12:53:42 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has joined ##stm32 2019-10-31T13:02:20 -!- kakinull [3e71bf21@62-113-191-33.bb.dnainternet.fi] has quit [Remote host closed the connection] 2019-10-31T13:02:47 -!- Laurenceb [~laurence@159.217.93.209.dyn.plus.net] has quit [Ping timeout: 276 seconds] 2019-10-31T13:09:48 -!- Haohmaru [~Haohmaru@195.24.53.110] has left ##stm32 [] 2019-10-31T13:11:38 -!- Mangy_Dog [Mangy_Dog@94.2.234.208] has joined ##stm32 2019-10-31T13:12:09 < jly> ty 2019-10-31T13:25:26 -!- X230t [x13@gateway/shell/suchznc/x-tpabxgugphajeghq] has quit [Remote host closed the connection] 2019-10-31T13:33:23 < englishman> https://www.anandtech.com/show/15036/sifive-announces-first-riscv-ooo-cpu-core-the-u8series-processor-ip 2019-10-31T13:38:08 < jly> ty 2019-10-31T14:46:00 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has joined ##stm32 2019-10-31T15:11:15 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has quit [Ping timeout: 240 seconds] 2019-10-31T15:38:02 -!- Laurenceb__ [~laurence@128.243.2.33] has quit [Remote host closed the connection] 2019-10-31T15:42:50 -!- Laurenceb [~laurence@128.243.2.33] has joined ##stm32 2019-10-31T15:45:43 -!- jly [uid355225@gateway/web/irccloud.com/x-owliqwovtsybcvuq] has quit [Quit: Connection closed for inactivity] 2019-10-31T15:52:36 -!- pennTeller [~pennTelle@unaffiliated/pennteller] has quit [Ping timeout: 265 seconds] 2019-10-31T16:02:29 < aandrew> I have Howard Johnson's high speed signal propagation videos on my plex server now 2019-10-31T16:02:37 < aandrew> basically 36h of signal integrity talk 2019-10-31T16:03:14 < aandrew> lol PIC on ketamine 2019-10-31T16:11:56 < qyx> lol you are on influxdb 2019-10-31T16:12:06 < qyx> do you use it? antto ^ 2019-10-31T16:12:10 < qyx> aandrew: ^ 2019-10-31T16:12:38 < aandrew> qyx: I am a complete newb at it but yes I've got a server set up and sucking in data from UDP 2019-10-31T16:21:16 < englishman> incucksdb is gr8 2019-10-31T16:21:36 < englishman> going to use it to monitor the production side of the company 2019-10-31T16:22:23 < qyx> except it doesn't have pow() function 2019-10-31T16:22:50 < englishman> https://github.com/johanvandenbroek/InfluxDB-Client-LabVIEW 2019-10-31T16:23:17 < qyx> fuk labview 2019-10-31T16:23:22 < englishman> lul 2019-10-31T16:23:25 -!- bitmask [~bitmask@c-73-215-237-27.hsd1.nj.comcast.net] has joined ##stm32 2019-10-31T16:23:38 < englishman> python has mostly replaced it 2019-10-31T16:34:20 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has joined ##stm32 2019-10-31T16:38:50 < bitmask> does it matter which way an inductor is wired? 2019-10-31T16:42:15 < Cracki> they have no polarity 2019-10-31T16:42:50 < bitmask> didnt think so, not sure why coilcraft marks them 2019-10-31T16:43:10 < Cracki> it might do the same to resistors 2019-10-31T16:43:32 < Cracki> or someone marked the symbol's pads such 2019-10-31T16:43:36 < bitmask> maybe to have the label facing the right way? i dunno 2019-10-31T16:43:46 < karlp> dude, of course they're polarised 2019-10-31T16:43:53 < karlp> you need to know whether it's going clockwise or not. 2019-10-31T16:44:01 < karlp> right ahnd rule man, don't you guys rememeber anything? 2019-10-31T16:44:13 < bitmask> I didn't go to school for this 2019-10-31T16:44:26 < qyx> enlighten me a bit 2019-10-31T16:44:38 < qyx> if I rotate it 180°, does it change anything? 2019-10-31T16:45:03 < Cracki> magnetic field 2019-10-31T16:45:11 < karlp> (I'm just kidding) 2019-10-31T16:45:16 < bitmask> hah 2019-10-31T16:45:17 < Cracki> if you don't care abut its sign, only about the throttling effect of the inductor, ¯\_(ツ)_/¯ 2019-10-31T16:45:37 < qyx> fuk the sarcasm 2019-10-31T16:45:58 < Cracki> you could call the coils in a motor "inductors" but that wouldn't imply you using them as part of a motor 2019-10-31T16:46:16 < Cracki> you can use motor coils as inductors tho. it's great fun. 2019-10-31T16:47:01 < Cracki> they'll behave weirdly... I'm sure there's an audiophile market for that 2019-10-31T16:48:16 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-31T16:54:42 -!- fenugrec [~fenugrec@24.105.71.66] has joined ##stm32 2019-10-31T16:59:24 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has joined ##stm32 2019-10-31T17:07:49 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-31T17:14:39 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-31T17:15:21 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Client Quit] 2019-10-31T17:21:52 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 268 seconds] 2019-10-31T17:37:53 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has joined ##stm32 2019-10-31T17:42:38 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has joined ##stm32 2019-10-31T17:46:40 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds] 2019-10-31T17:48:15 < jadew> "Before I started working my current job, there was a guy in it that... commented everything just with a date and nothing else. And he had a diary where he kept what he wrote each day. 2019-10-31T17:48:17 < jadew> And of course, he took that diary when leaving the job. 2019-10-31T17:48:19 < jadew> " 2019-10-31T17:49:24 < Steffanx> Hah. 2019-10-31T17:50:14 < qyx> it resembles wiping the disk encryption key to invalidate all data 2019-10-31T18:04:27 -!- sterna [~Adium@h-90-120.A137.corp.bahnhof.se] has quit [Quit: Leaving.] 2019-10-31T18:05:39 -!- drzacek [~drzacek@b941c009.business.dg-w.de] has quit [Quit: Leaving] 2019-10-31T18:13:14 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has joined ##stm32 2019-10-31T18:21:56 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has quit [Ping timeout: 276 seconds] 2019-10-31T18:28:30 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has joined ##stm32 2019-10-31T18:53:17 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-31T18:53:44 < aandrew> it's unbelievable that someone who is no stranger to the lecture circuit does *not* repeat asked questiosn before answering them 2019-10-31T18:53:51 < aandrew> this is amateur hour bullshit 2019-10-31T18:55:15 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has joined ##stm32 2019-10-31T19:02:50 < jadew> what are you watching? 2019-10-31T19:06:04 < aandrew> the Howard Johnson videos I mentioned earlier 2019-10-31T19:06:14 < aandrew> sign up for plex I'll share the lib with you 2019-10-31T19:10:55 < jadew> how advanced is it? I think I have a pretty good understanding of signal integrity 2019-10-31T19:15:16 < jadew> found a ToC 2019-10-31T19:15:36 -!- renn0xtk9 [~max@2a02:810d:1540:2448:a873:5df4:f382:a0d7] has joined ##stm32 2019-10-31T19:21:34 < Cracki> >diary for code 2019-10-31T19:21:52 < Cracki> code is probably as shit as the comments 2019-10-31T19:22:27 < Cracki> like a mason signing every single fucking brick he lays 2019-10-31T19:24:01 < jadew> his idea was brilliant tho 2019-10-31T19:24:21 < jadew> he took his comments with him when he left - that's genius 2019-10-31T19:25:26 < Cracki> I can check in object files or compiler generated assembly. 2019-10-31T19:26:31 < jadew> check what? 2019-10-31T19:26:41 < Cracki> commit 2019-10-31T19:27:04 < jadew> are we talking about the same thing? 2019-10-31T19:27:08 < Cracki> yes, code 2019-10-31T19:27:17 < Cracki> he obfuscates his code by taking away comments 2019-10-31T19:27:27 < jadew> yeah, but the comments were never there to begin with 2019-10-31T19:27:31 < Cracki> the next step is checking in more obfuscated code 2019-10-31T19:27:37 < Cracki> right, he kept them private 2019-10-31T19:27:38 < jadew> ah 2019-10-31T19:27:43 < jadew> I see 2019-10-31T19:27:53 < jadew> I see what you mean, I thought you thought you found a solution :P 2019-10-31T19:28:47 < Cracki> there are many ways to ensure job security. committing shit/obfuscated code is rather a guarantee that one gets fired. 2019-10-31T19:29:33 < Cracki> btw I compile my more complicated python code with cython so I can distribute binaries that aren't as simple to decompile as java code. 2019-10-31T19:29:51 < jadew> when I was employed, part of my job was to write the most complex bits, so when I left, there was nobody left that was good enough to work on that code 2019-10-31T19:30:10 < Cracki> good practice when you gotta expect people to fuck you over for your free work :P 2019-10-31T19:30:25 < Cracki> yeah that's real job security 2019-10-31T19:30:33 < Cracki> they'll need to hire or contract you later 2019-10-31T19:30:44 -!- c10ud [~c10ud@emesene/dictator/c10ud] has quit [Read error: Connection reset by peer] 2019-10-31T19:31:11 < jadew> they didn't do that, instead, they hired a team of developers and shortly after the company went belly up 2019-10-31T19:31:20 < Cracki> heheeeee 2019-10-31T19:31:52 < Cracki> btw there are tons of videos on youtube of people burning stuff such as piles of wood and brush 2019-10-31T19:32:18 < Cracki> if you like watching something burn and you've slashed and burned everything around you already, that's a decent substitute 2019-10-31T19:38:57 -!- boB_K7IQ [~boB_K7IQ@c-73-193-6-103.hsd1.wa.comcast.net] has quit [Ping timeout: 240 seconds] 2019-10-31T19:41:28 -!- sterna [~Adium@c-3fe2e253.016-35-62726f1.bbcust.telenor.se] has quit [Quit: Leaving.] 2019-10-31T19:48:37 < aandrew> jadew: this is the guru of SI and EMC 2019-10-31T19:48:51 < aandrew> I think he's on par with Ott 2019-10-31T19:53:07 -!- ac_slater [~weechat@pool-173-48-114-24.bstnma.fios.verizon.net] has quit [Quit: WeeChat 2.6] 2019-10-31T19:55:39 -!- veegee [~veegee@ipagstaticip-3d3f7614-22f3-5b69-be13-7ab4b2c585d9.sdsl.bell.ca] has quit [Ping timeout: 240 seconds] 2019-10-31T19:56:08 -!- emeb [~ericb@ip68-2-121-171.ph.ph.cox.net] has left ##stm32 [] 2019-10-31T19:59:12 -!- veegee [veegee@gateway/vpn/privateinternetaccess/veegee] has joined ##stm32 2019-10-31T20:06:40 < bitmask> hmm, how far should a USB-A female connector hang off the pcb 2019-10-31T20:07:04 < aandrew> bitmask: depends entirely on your enclosure really. 2019-10-31T20:07:33 < aandrew> for my own stuff I tend to have the connector hang over only enough for the flare to be off the board 2019-10-31T20:07:42 < bitmask> I see 2019-10-31T20:07:59 < bitmask> I think I'll go with like 2mm since i'll be 3d printing the enclosure 2019-10-31T20:08:05 < bitmask> maybe 3 2019-10-31T20:08:30 < aandrew> no matter what you choose the first one will be wrong :-) 2019-10-31T20:08:35 < aandrew> may as well just pick one and get it out of the way 2019-10-31T20:08:36 < bitmask> heh of course 2019-10-31T20:17:02 < Cracki> I'd err inwards 2019-10-31T20:17:21 < Cracki> if it sticks out too much, it may stick out of the enclosure 2019-10-31T20:17:39 < Cracki> which may force you to make the enclosue larger but ok 2019-10-31T20:18:26 < Cracki> i'd let it hang off the pcb a bit so it can anchor the pcb into the cutout of the enclosure 2019-10-31T20:19:29 < Cracki> I'd also plan for some small gap between pcb and enclosure, in addition to enclosure wall thickness and however much you want the connector face to be recessed from enclosure surface 2019-10-31T20:32:48 < aandrew> just be careful becuase the USB A plug is only so long 2019-10-31T20:33:07 -!- con3 [~kvirc@154.119.40.237] has joined ##stm32 2019-10-31T20:36:45 -!- boB_K7IQ [~boB_K7IQ@152.44.129.202] has joined ##stm32 2019-10-31T20:50:03 -!- con3 [~kvirc@154.119.40.237] has quit [Quit: KVIrc 4.9.3 Aria http://www.kvirc.net/] 2019-10-31T21:09:38 -!- fsasm [~fsasm@62-178-93-7.cable.dynamic.surfer.at] has quit [Ping timeout: 276 seconds] 2019-10-31T21:20:24 -!- rajkosto [~Rajko@cable-178-149-123-201.dynamic.sbb.rs] has joined ##stm32 2019-10-31T21:27:25 -!- kakipro [b237d8f7@178-55-216-247.bb.dnainternet.fi] has joined ##stm32 2019-10-31T21:29:14 -!- catphish [~catphish@unaffiliated/catphish] has joined ##stm32 2019-10-31T21:32:25 -!- catphish [~catphish@unaffiliated/catphish] has quit [Client Quit] 2019-10-31T21:36:36 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Quit: Whop whop] 2019-10-31T21:39:01 < kakipro> bye 2019-10-31T21:46:30 -!- con3 [~kvirc@154.119.40.183] has joined ##stm32 2019-10-31T22:22:55 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-31T22:25:20 -!- Steffanx [~steffanx@unaffiliated/steffanx] has quit [Client Quit] 2019-10-31T22:31:25 -!- Steffanx [~steffanx@unaffiliated/steffanx] has joined ##stm32 2019-10-31T22:37:19 -!- ekaologik [~quassel@p4FF16502.dip0.t-ipconnect.de] has quit [Quit: https://quassel-irc.org - Komfortabler Chat. Überall.] 2019-10-31T23:02:10 -!- Jybz [~jibz@2a01:e0a:198:e110:4a51:b7ff:fe84:99e6] has quit [Quit: Konversation terminated!] 2019-10-31T23:18:55 -!- fenugrec [~fenugrec@24.105.71.66] has quit [Ping timeout: 268 seconds] 2019-10-31T23:38:29 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has joined ##stm32 2019-10-31T23:41:31 -!- PaulFertser [paul@paulfertser.info] has quit [Ping timeout: 250 seconds] 2019-10-31T23:46:26 -!- emeb_mac [~ericb@ip68-2-121-171.ph.ph.cox.net] has quit [Quit: Leaving.] 2019-10-31T23:49:40 < buZz> ok, my stm32 device now finally kicks into the right DFU mode 2019-10-31T23:49:41 < buZz> https://hastebin.com/raw/iyituzejex 2019-10-31T23:50:08 < buZz> i tried dfu-util -U alt1 -a 1 ; dfu-util -U alt2 -a 2 2019-10-31T23:50:14 < buZz> but its all outputting the same 16kb 2019-10-31T23:50:20 < buZz> there should be more data .. 2019-10-31T23:52:48 < buZz> weird, whatever 'expected size' i specify, its ; Limiting default upload to 16384 bytes 2019-10-31T23:53:24 < Steffanx> Mine seems to do fine with at least 350MByte/s kakipro. 2019-10-31T23:54:55 < buZz> anyone has a clue what > "@Internal Flash /0x08000000/04*016Kg,01*064Kg,03*128Kg" would mean? :D --- Log closed Fri Nov 01 00:00:19 2019