--- Log opened Sat Sep 01 00:00:10 2012
2012-09-01T00:07:24 < Thorn> http://iichan.hk/tr/src/1340538856690.jpg http://iichan.hk/tr/src/1340538919422.jpg
2012-09-01T00:13:12 < cjbaird> lol idlechan is still going.. So 2004.
2012-09-01T00:14:16 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-01T00:14:29 < Thorn> this isn't idlechan
2012-09-01T00:14:47 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-01T00:15:50 < cjbaird> Invaded by teh Commies.
2012-09-01T00:18:49 < Laurenceb_> wut
2012-09-01T00:18:54 < Laurenceb_> why commies?
2012-09-01T00:20:07 < cjbaird> I got permbanned from idlechan for calling WaHA out on claming that putting 8 rand() syscalls inside the inner loop of his tripcode cracker didn't make it faster.. What a fucking gormless crybaby.
2012-09-01T00:20:31 < Laurenceb_> wtf
2012-09-01T00:20:41 < Laurenceb_>  tripcode cracker
2012-09-01T00:20:45 < Laurenceb_> what
2012-09-01T00:21:10 < Laurenceb_> some sort of weird virginity preservation society?
2012-09-01T00:21:26 < cjbaird> yep
2012-09-01T00:29:21 < Thorn> http://iichan.hk/tr/src/1340539623367.jpg
2012-09-01T00:38:03 -!- izua [~izua@188.26.165.54] has joined ##stm32
2012-09-01T00:38:03 -!- izua [~izua@188.26.165.54] has quit [Changing host]
2012-09-01T00:38:03 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T00:38:04 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T00:44:49 < Laurenceb_> so old
2012-09-01T00:44:55 < Laurenceb_> my dad has those on his wall
2012-09-01T00:45:31 < Laurenceb_> along with Oz "school kids" issue
2012-09-01T00:49:50 -!- Huitaryan1 [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-01T00:50:10 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-01T00:50:17 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-01T00:50:56 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-01T00:51:32 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Quit: leaving]
2012-09-01T00:51:52 < Laurenceb_> http://en.wikipedia.org/wiki/Schoolkids_OZ
2012-09-01T00:55:15 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-01T01:03:00 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 248 seconds]
2012-09-01T01:08:51 -!- barthess [~barthess@178.154.9.180] has quit [Read error: Connection reset by peer]
2012-09-01T01:27:20 -!- barthess [~barthess@164.177.210.15] has joined ##stm32
2012-09-01T01:42:01 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-01T01:43:03 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-01T01:43:03 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-01T02:19:34 -!- DaKu is now known as daku
2012-09-01T02:29:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-01T02:35:40 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 246 seconds]
2012-09-01T02:50:13 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-01T02:51:21 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-01T03:05:36 -!- Laurenceb_ [~Laurence@host81-159-189-200.range81-159.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-01T03:21:20 -!- barthess [~barthess@164.177.210.15] has quit [Quit: Leaving.]
2012-09-01T03:28:40 < dongs> sup trolls
2012-09-01T03:28:49 < dongs> flyback buying up drugs again i see
2012-09-01T03:31:31 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-09-01T03:39:06 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-01T03:59:03 < dongs> https://www.couchsurfing.org/activity/view/BFVYD0/mount-fuji-climb zyp is back in japan
2012-09-01T03:59:19 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-01T05:15:24 -!- DLPeterson_ [~AndChat16@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-09-01T05:35:17 < bgamari> Hmm, I'm not getting SysTick
2012-09-01T05:35:27 < bgamari> the interrupt vector appears to be set properly
2012-09-01T05:35:41 < bgamari> The tick is enabled in the Cortex
2012-09-01T05:36:02 < bgamari> I can verify that it's ticking down over JTAG
2012-09-01T05:36:06 < bgamari> but I never get the interrupt
2012-09-01T05:39:16 < dongs> you have breakpoint / led blinking in SysTick_Handler?
2012-09-01T05:42:54 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-01T05:44:08 < bgamari> dongs, I have a breakpoint
2012-09-01T05:44:16 < dongs> you setup nvic for systick right
2012-09-01T05:45:11 < bgamari>  *(SysTick_Type *) (0xe000e000+0x10) = {CTRL = 65543, LOAD = 15999,
2012-09-01T05:45:11 < bgamari>   VAL = 14181, CALIB = 1073760574}
2012-09-01T05:45:17 < dongs> what the christ
2012-09-01T05:45:31 < dongs> are you another one of those dudes who writes magic numbers into shit
2012-09-01T05:45:38 < bgamari> No
2012-09-01T05:45:41 < bgamari> that's output from gdb
2012-09-01T05:45:56 < dongs> good thing I never use gdb, i'd probaly kill myself if thats how i debugged shit
2012-09-01T05:46:12 < bgamari> 65543 is 10007
2012-09-01T05:46:17 < bgamari> Yeah, the formatting isn't ideal
2012-09-01T05:46:22 < bgamari> 0x10007 rather
2012-09-01T05:46:24 < dongs> anyway SysTick_Config() shiould do it
2012-09-01T05:46:29 < dongs> youve run that right?
2012-09-01T05:46:32 < bgamari> Yep
2012-09-01T05:46:56 < dongs> and youre using stdperiphlib?
2012-09-01T05:46:59 < dongs> or that libopencm3 shit
2012-09-01T05:47:12 < bgamari> The bottom three bits of STK_CTRL are set, which should mean it's counting and producing interrupts
2012-09-01T05:47:17 < bgamari> I'm porting libmanyuc
2012-09-01T05:48:11 -!- tech2077 [~tech2077@75.53.132.42] has joined ##stm32
2012-09-01T05:48:16 < dongs> right, does your vector name match with the shit in startup/assembly file?
2012-09-01T05:49:17 < bgamari> Yes
2012-09-01T05:49:37 < bgamari> x/4x *0x3c
2012-09-01T05:49:37 < bgamari> 0x10cd <SysTick_Handler>:	0xc10390f6	0x1a0300f2	0x01f10268	
2012-09-01T05:49:55 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-01T05:49:58 < dongs> are "interrupts" enabled :)
2012-09-01T05:49:59 < dongs> haha.
2012-09-01T05:50:13 < R2COM> if its about NVIC, just lookup stm32f4xxx.h file
2012-09-01T05:50:18 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-01T05:50:57 < R2COM> and then do something like:  NVIC->ISER[0] |= 1 << TIM2_IRQn;
2012-09-01T05:51:20 < R2COM> thats it
2012-09-01T05:51:32 < R2COM> its just lookup appropriate definition in the stm32f4 header file
2012-09-01T05:51:48 < R2COM> do not use that shitass stdperiph
2012-09-01T05:52:14 < bgamari> R2COM, I'm trying to port libmanyuc
2012-09-01T05:53:03 < R2COM> well, ok, never used that stuff
2012-09-01T05:56:36 < bgamari> It's definitely a low-level configuration issue
2012-09-01T05:57:00 < bgamari> R2COM, Wait, TIM2 != SysTick
2012-09-01T05:57:20 < R2COM> do something like: <------
2012-09-01T06:01:37 < dongs> bgamari: are interrupts enabled
2012-09-01T06:01:37 < bgamari> Like enabling the TIM2 interrupt?
2012-09-01T06:02:18 < dongs> cpsid or whatever
2012-09-01T06:02:28 < bgamari> I use __enable_irq()
2012-09-01T06:02:31 < dongs> right
2012-09-01T06:02:34 < dongs> that ought to work
2012-09-01T06:03:09 < bgamari> hmm
2012-09-01T06:05:42 < bgamari> Hah!
2012-09-01T06:05:44 < bgamari> success
2012-09-01T06:06:10 < bgamari> I guess I didn't succeed in getting the image with __enable_irq() to flash
2012-09-01T06:14:41 < dongs> thats what you get for using gdb
2012-09-01T06:18:37 < bgamari> dongs, Naw, that was just poor use of openocd
2012-09-01T06:18:52 < bgamari> I don't know if it's even possible to flash the device from gdb
2012-09-01T06:27:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-01T06:27:34 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-01T06:28:24 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-01T06:28:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-01T06:36:38 < LeelooMinai> Hmm, suddenly my uC endups in WWDG interrupt entry...
2012-09-01T06:37:12 < LeelooMinai> http://i.imgur.com/13ISI.png <- Not sure how it's possible, as WWDG is not even enabled
2012-09-01T06:39:31 < LeelooMinai> btw, I can flash from gdb through openocd, but I first have to unlock the flash, and openocd methds do not work - so I do that by using st-link utility once and later the openocd methods work... until next day, tath is when I have to do it again:)
2012-09-01T06:44:59 < dongs> LeelooMinai are you related to hackkitten aka maya posch
2012-09-01T06:45:31 < LeelooMinai> No, I just know her from ##electronics
2012-09-01T06:45:39 < dongs> "her".
2012-09-01T06:45:40 < dongs> bzzzzzzzzzt.
2012-09-01T06:46:25 < dongs> its a dude.
2012-09-01T06:46:27 < LeelooMinai> Well, she is from Netherlands I think and I live in Canda
2012-09-01T06:46:45 < LeelooMinai> No, she is not
2012-09-01T06:47:45 < dongs> penis between her legs says otherwise
2012-09-01T06:48:54 < LeelooMinai> I know her story... And just because your view on gender is severly limited, doesn't mean she is a "dude":)
2012-09-01T06:50:08 < dongs> right, right, dude.
2012-09-01T06:50:13 < dongs> so i guess you are related. thanks.
2012-09-01T06:52:01 < LeelooMinai> I find your train of thought disturbingly illogical...
2012-09-01T06:52:34 < LeelooMinai> Are you on some medication?
2012-09-01T06:52:38 < dongs> i find dudes on irc pretending to be girls in tech-related channels pretty illogical too
2012-09-01T06:53:08 < LeelooMinai> She is not pretending anything - she even has a website with her story
2012-09-01T06:54:23 < dongs> yea i can blog a shitstorm about how im a girl but i still got a dick
2012-09-01T06:55:13 < LeelooMinai> Well, you can always do that... what's your point? :)
2012-09-01T06:57:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-01T07:06:53 -!- AndTabSauce [~AndTabSau@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-01T07:13:33 < cjbaird> Dongs still needs a female blogger for his websight.
2012-09-01T07:15:40 < dongs> all irc "girls" have the same basic MO. and you can tell which of them have a dick after the first sentence they type up
2012-09-01T07:16:21 -!- DLPeterson [~AndChat16@107.197.105.35] has joined ##stm32
2012-09-01T07:18:17 < LeelooMinai> I think you seriously overestimate your clairvoyance skills
2012-09-01T07:20:51 < dongs> my years of irc trolling experience disagrees.
2012-09-01T07:22:34 < LeelooMinai> What do you care about here anyways?... It's her business...
2012-09-01T07:24:50 -!- DLPeterson [~AndChat16@107.197.105.35] has quit [Ping timeout: 268 seconds]
2012-09-01T07:31:24 < cjbaird> The Koppel-based gender determiners are my usual trick for outing alt'ing Furries... http://i.imgur.com/VpRKo.png
2012-09-01T07:34:48 < LeelooMinai> You guys have definitely too much time on your hands, obviously...
2012-09-01T07:38:45 < cjbaird> Aww, it's /fun/ trolling guys pretending to have a IRL fiancee in attempting to scam free artz for their virtual girlfriend.. (Protip: real girls don't have breast expansion fetish.. Your 'girlfriend' on FurAffinity keeps commissioning her top-heavy fursona has a present for you? Haaaaaaaaaaaaaaaa.)
2012-09-01T07:40:20 < cjbaird> Rule #41 of the Internet: Furries ruin everything.
2012-09-01T07:43:29 < LeelooMinai> Ok, that was pretty...  uncomprehensible
2012-09-01T07:44:00 < dongs> hahaha.
2012-09-01T07:50:30 < dongs> !zlog
2012-09-01T07:50:40 < dongs> cjbaird: wheres the log of this chan
2012-09-01T07:50:45 < dongs> i wanna run some other furries through the analyzer
2012-09-01T07:52:14 < cjbaird> I just grepped my ~/logs directory. Didn't have much from hackkitten, before you ask.
2012-09-01T07:52:22 < dongs> haha
2012-09-01T07:52:28 < dongs> hackkitten logs are all over internets
2012-09-01T07:52:55 < dongs> http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23robotics/2012-07-21.html
2012-09-01T07:52:57 < dongs> jesus fucking christ
2012-09-01T07:53:00 < dongs> look at all that yiffing
2012-09-01T07:53:07 < dongs> [11:39:03] * furrywolf bets hackkitten is one of those kittens that crawls onto people's heads while they're sleeping
2012-09-01T07:53:11 < dongs> [11:39:39] * hackkitten looks all innocent~
2012-09-01T07:53:50 < dongs> thats 1st hit for "hackkitten arduino" on google.
2012-09-01T07:54:22 < cjbaird> Gah. Reads like Furrymuck.
2012-09-01T07:54:57 < dongs> <GuShH> is another furry/dude
2012-09-01T07:55:37 < zippe> cjbaird: hmm, don't be so sure about the "real girls" anything
2012-09-01T07:55:57 < zippe> cjbaird: Unless you want to claim that girls in SF aren't real, at least.
2012-09-01T07:59:11 < cjbaird> Like the page goes at pains to explain, it's used as an estimate. I've used it a fair bit, and the nearest to a false positive I've seen from it was a 50/50 result.
2012-09-01T07:59:39 < AndTabSauce> lets talk about microcontrollers instead of furries and rp
2012-09-01T07:59:58 < AndTabSauce> does anyone have an opinion on libopencm3?
2012-09-01T08:00:10 < dongs> I do.
2012-09-01T08:00:26 < dongs> it unnecessary renames fucking everything, and is only buildable by gcc.
2012-09-01T08:00:30 < AndTabSauce> is it a real opinion or a dongs-opinion
2012-09-01T08:00:35 < AndTabSauce> ah
2012-09-01T08:00:38 < dongs> its somewhat legit.
2012-09-01T08:01:38 < dongs> i think renaming vectors is pretty asstastic.
2012-09-01T08:01:47 < dongs> the fact that it only works wiht gcc limits it to mostly freetard IDEs.
2012-09-01T08:01:56 < dongs> though it might work out OK in crossworks
2012-09-01T08:02:12 < AndTabSauce> ides are for chumps anyway
2012-09-01T08:02:15 -!- carp3 [~chatzilla@31.59.11.139] has quit [Quit: ChatZilla 0.9.88.2 [Firefox 15.0/20120824154833]]
2012-09-01T08:02:33 < dongs> yes of course
2012-09-01T08:03:59 < dongs> got a bad of megacheap 8mhz xtals from china lets see if they work
2012-09-01T08:05:41 < LeelooMinai> You guys seem to be very opinionated...
2012-09-01T08:06:12 < AndTabSauce> yeah well, that's just like
2012-09-01T08:06:23 < AndTabSauce> your opinion, man
2012-09-01T08:07:17 < AndTabSauce> i am glad to hear the library is ok, was worried i'd have to use the stm one
2012-09-01T08:07:29 < LeelooMinai> There's a thin line though between being opinionated and being close minded:)
2012-09-01T08:07:35 < cjbaird> libopencm3 can work, and you get it 'for free' when building summon-arm-toolchain...
2012-09-01T08:07:56 < cjbaird> Lots on the library doesn't, however. Like USB.
2012-09-01T08:08:17 < cjbaird> It's 'in alpha', basically.
2012-09-01T08:08:20 < AndTabSauce> was using chibios until i realized it's a piece of crap
2012-09-01T08:08:30 < AndTabSauce> i mean er does not suit my needs
2012-09-01T08:09:07 < dongs> AndTabSauce: try coos
2012-09-01T08:09:31 < dongs> just does rtos for you and stays out of the way for everything else
2012-09-01T08:09:35 < dongs> without forcing hal/shit on you.
2012-09-01T08:09:53 < dongs> theres nothing wrong wiht stm library.
2012-09-01T08:16:52 < dongs> i cant find any public logged channels with search that have hackkitten in them :(
2012-09-01T08:17:23 < dongs> 06:57:50 * hackkitten is developing an OS for her own CPU architecture written in VHDL and implemented on an FPGA
2012-09-01T08:17:30 < dongs> 06:58:06 * hackkitten is also crazy~
2012-09-01T08:17:52 < dongs> 07:48:41 * hackkitten hides thePowersGang's interrupts~
2012-09-01T08:18:37 < AndTabSauce> why are you so interested in this individual
2012-09-01T08:18:45 < dongs> im not
2012-09-01T08:18:54 < LeelooMinai> Right...
2012-09-01T08:21:58 < AndTabSauce> i am p sure ##electronics fits that bill
2012-09-01T08:22:30 < dongs> where are their searchable logs?
2012-09-01T08:22:35 < dongs> i only see shit from tom-itx
2012-09-01T08:22:37 < R2COM> -_-
2012-09-01T08:23:22 < AndTabSauce> use google for the searching
2012-09-01T08:23:45 < dongs> comeon, im talkin about typing a nick and getting all their blogs for the last x months
2012-09-01T08:23:52 < AndTabSauce> after explaining the reason for your infatuation to me
2012-09-01T08:25:34 < dongs> haha
2012-09-01T08:25:54 < dongs> i pasted "about me" shit off hackkitten blog into cjbaird's tool
2012-09-01T08:26:01 < dongs> Female Score: 1255
2012-09-01T08:26:02 < dongs> Male Score: 1669
2012-09-01T08:26:02 < dongs> The Gender Genie thinks the author of this passage is: male!
2012-09-01T08:26:09 -!- blocky [blocky@S0106c0c1c07746fb.gv.shawcable.net] has joined ##stm32
2012-09-01T08:26:27 < zippe> dongs:  "theres nothing wrong wiht stm library" … O_o
2012-09-01T08:26:37 < dongs> zippe: comeon, political/bullshit aside.
2012-09-01T08:27:07 < zippe> dongs: we've had this conversation before
2012-09-01T08:27:15 < zippe> dongs: The ST library is mostly crap
2012-09-01T08:27:25 < dongs> but it works
2012-09-01T08:27:27 < zippe> dongs: it doesn't actually *do* anything
2012-09-01T08:27:31 < dongs> right
2012-09-01T08:27:36 < dongs> and it doesnt need to
2012-09-01T08:27:46 < dongs> if you want hal/wahtever, use ti/stellaris shit
2012-09-01T08:27:47 < zippe> dongs: other than bloat your application and slow down your compiles
2012-09-01T08:28:14 < dongs> hardly any bloat.
2012-09-01T08:28:17 < zippe> So by those definitions alone, there are things wrong with it
2012-09-01T08:28:30 < zippe> Then we come to their execrable USB library
2012-09-01T08:28:39 < AndTabSauce> and clearly a simple statistical natural language analysis technique can accurately pigeonhole someone with both sets of hardware into the binary gender norm?
2012-09-01T08:28:40 < dongs> yea.. THATs a problem
2012-09-01T08:28:46 < zippe> Or their hopeless examples
2012-09-01T08:28:57 < R2COM> stmperiph is used by afghanistan electronics industry
2012-09-01T08:29:14 < dongs> USB library is my only complaint wiht stdperiph.
2012-09-01T08:29:22 < zippe> Which seem to be 99% IDE config files for moronic button-clickers, smeared with a thin layer of really bad application code
2012-09-01T08:29:37 < dongs> i never used their ide config files
2012-09-01T08:29:59 < zippe> I have met some really awful vendor code in my time, and ST are by no means the worst, but it's shocking what they expect you to put up with
2012-09-01T08:30:22 < zippe> The wasted man-years of work that their library represents is a sad thing
2012-09-01T08:30:27 < dongs> if you think stdperiph is bad.. have you seen st's code for their dtv/stb SoCs?
2012-09-01T08:30:30 < zippe> Like filling potholes on a closed road
2012-09-01T08:30:38 < dongs> but shit works and people use it
2012-09-01T08:30:39 < R2COM> any company making hardware, most of the time its application code is demonstration only
2012-09-01T08:30:40 < zippe> dongs: not that you heard it from me, but yes
2012-09-01T08:30:47 < R2COM> its funny to adapt that code for serious projects
2012-09-01T08:31:01 < R2COM> not any, but, lets say most
2012-09-01T08:31:24 < zippe> dongs: in my line of work, I've seen most of everything
2012-09-01T08:31:37 < zippe> dongs: perhaps not mil custom stuff
2012-09-01T08:32:02 < zippe> dongs: but if a vendor sells into the consumer space, chances are I or one of my co-workers has their cards in a box
2012-09-01T08:33:00 < R2COM> the only thing is, ST deals only with people buying tons of their chips
2012-09-01T08:33:02 < R2COM> thats it
2012-09-01T08:33:06 < zippe> R2COM: you'd think that, but LMI were a great example of a company that bootstrapped themselves by delivering a HAL that got their customers to market faster.
2012-09-01T08:33:10 < R2COM> and... its European...
2012-09-01T08:33:15 < R2COM> so, ... you know.
2012-09-01T08:33:41 < zippe> R2COM: Not necessarily.  The PX4 team has had quite a bit of support from ST
2012-09-01T08:34:02 < zippe> R2COM: it just depends on what you mean by "deals with"
2012-09-01T08:34:18 < zippe> R2COM: they're not going to make custom silicon for PX4
2012-09-01T08:34:36 < R2COM> I'm talking about support
2012-09-01T08:34:37 < zippe> R2COM: But early samples, FAE service, etc. is not beyond possible
2012-09-01T08:35:07 < dongs> haha, so how many millions chips did you buy?
2012-09-01T08:35:10 < R2COM> its support is not like, lets say support from Xilinx
2012-09-01T08:35:21 -!- spacekitteh [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T08:35:30 < R2COM> thats the thing, I did not buy millions of their chips
2012-09-01T08:35:34 < dongs> dudes I work with were considering using ST's sensors for some consumer proj with maybe 10k units/year prospect and they just told them to fuck off.
2012-09-01T08:35:43 < dongs> and thats a legit company with legit product line etc
2012-09-01T08:35:56 < dongs> not some opensauce hackspace
2012-09-01T08:36:03 < zippe> dongs: it's all in who you know and how you play it
2012-09-01T08:36:11 < R2COM> ok, change the company and stop bitching
2012-09-01T08:36:18 < dongs> R2COM: thats what they did
2012-09-01T08:36:22 < R2COM> good
2012-09-01T08:36:32 < zippe> R2COM: sadly, the sensor alternatives kinda suck
2012-09-01T08:36:51 < zippe> R2COM: ST have prettymuch nailed the sweet spot between "cheap" and "works"
2012-09-01T08:36:54 < R2COM> if someone is *really* serious about sensors, it can be done in a custom way
2012-09-01T08:37:17 < dongs> in consumer products, only "cheap" matters
2012-09-01T08:37:22 < dongs> doesnt need to be "serious"
2012-09-01T08:37:28 < R2COM> consumer..non military?
2012-09-01T08:37:31 < dongs> yeah
2012-09-01T08:37:34 < R2COM> well
2012-09-01T08:37:38 < zippe> dongs: "cheap" and "works well enough"
2012-09-01T08:37:40 < R2COM> then get what you can get and thats it.
2012-09-01T08:38:23 < dongs> ya not my project
2012-09-01T08:38:35 < dongs> just funny how ST places people on ignore unless they guarantee millions of units
2012-09-01T08:38:53 < zippe> Most vendors are like that
2012-09-01T08:39:14 < zippe> Unless you find someone that has a soft sport for hackers, or someone that sees you as a long-term opportunity
2012-09-01T08:39:24 < dongs> actually you can se how much ST cares about t heir clients/users by their website
2012-09-01T08:39:25 < zippe> I didn't say all … see above inre: LMI
2012-09-01T08:39:37 < zippe> I was sad when LMI got eaten by TI
2012-09-01T08:40:34 < zippe> dongs: yes, compare/contrast with, say, NXP's
2012-09-01T08:41:36 < zippe> … where you still can't resize the retarded flash widget they use for parametric search
2012-09-01T08:41:39 < dongs> nxp's err conexant err..whoever owns them now DTV product line was untouchable too
2012-09-01T08:41:48 < R2COM> by thw way, I see that NXp has now 200MHz chip
2012-09-01T08:41:49 < R2COM> armm4
2012-09-01T08:42:00 < R2COM> vs stm's 168MHz
2012-09-01T08:42:12 < zippe> R2COM: The lpc43xx is a weird thing
2012-09-01T08:42:13 < dongs> but does NXP have a DMA controller with more than one IRQ?
2012-09-01T08:42:19 < dongs> o snap.
2012-09-01T08:42:28 < R2COM> not sure, not looked thoroughly
2012-09-01T08:42:43 < zippe> dongs: who cares?  software interrupt demux is hardly brain surgery
2012-09-01T08:42:50 < dongs> zippe: yes yes, i know.
2012-09-01T08:43:03 < dongs> zippe b ut imagine all the overhead.
2012-09-01T08:43:03 < zippe> Biggest issue with the 43xx is that it doesn't come in a big-flash, big-ram config
2012-09-01T08:43:14 < zippe> dongs: it's about nine instructions.
2012-09-01T08:43:22 < dongs> heh
2012-09-01T08:43:32 < zippe> Not what I'd call difficult.
2012-09-01T08:44:02 < zippe> dongs: day job is still mostly v7a, where there is only one interrupt vector… I have no sympathy
2012-09-01T08:45:01 < dongs> zippe, how does stack work in that scenario
2012-09-01T08:45:10 < zippe> Ah hey. TI's parametric search widget sucks less.
2012-09-01T08:45:16 < dongs> you have a single vector that does a swtich() or something and you do what, call another function to handle?
2012-09-01T08:45:35 < zippe> dongs: interrupts are taken on a separate stack
2012-09-01T08:45:49 < dongs> each interrupt has their own?
2012-09-01T08:46:02 < zippe> No, in general you avoid doing any major work in interrupt context
2012-09-01T08:46:08 < dongs> well, obviousl
2012-09-01T08:46:29 < zippe> If you can't get it done in a microsecond or so, you delegate to an async handler in thread context
2012-09-01T08:47:07 < zippe> Then returning from interrupt context will usually switch to the thread context you just unblocked
2012-09-01T08:47:18 < zippe> unless something else more important is happening
2012-09-01T08:47:33 < zippe> If you've done Windows driver development, think DPC
2012-09-01T08:48:03 < dongs> ya, i understand the concept and about not spending time in interrupts
2012-09-01T08:48:15 < dongs> im just curious how a single handler deals with (potentially nested) interrupts
2012-09-01T08:48:21 < dongs> does the stack for eahc one just keep piling up or what
2012-09-01T08:48:21 < zippe> You don't nest
2012-09-01T08:48:25 < dongs> sure you do?
2012-09-01T08:48:31 < dongs> isnt that waht N in NVIC stands for?
2012-09-01T08:48:31 < zippe> v7a doesn't support nesting interrupts
2012-09-01T08:48:36 < dongs> oh.
2012-09-01T08:48:53 < zippe> The NVIC is an invitation to disaster
2012-09-01T08:49:28 < zippe> If you use more than 2 or 3 nesting levels, you end up with an unmanageable mess.
2012-09-01T08:49:43 < zippe> I had to dig out a project recently that had decided to give every interrupt its own priority
2012-09-01T08:49:59 < zippe> They couldn't understand why every thread needed 16k of stack
2012-09-01T08:50:46 < zippe> I made them draw a chart of all their interrupts and handlers, then told them to reset all their priorities to the default.
2012-09-01T08:51:02 < zippe> No complaints so far...
2012-09-01T08:51:02 < dongs> will it still nest if priority is same?
2012-09-01T08:51:06 < zippe> No
2012-09-01T08:51:09 < zippe> Only higher.
2012-09-01T08:51:13 < dongs> hmm i read somewehre it still will
2012-09-01T08:51:24 < zippe> Are you confusing nesting and chaining?
2012-09-01T08:51:40 < dongs> nesting = it'll fire an interrupt while another one is executing
2012-09-01T08:51:46 < zippe> Right, so no
2012-09-01T08:52:05 < zippe> Unless you mess with the current priority level
2012-09-01T08:52:16 < zippe> That's a not-uncommon trick
2012-09-01T08:52:28 < zippe> Do the first part of your work at level, then depress the level so that others at the same level can run
2012-09-01T08:52:43 < dongs> ill have to revisit my shit.
2012-09-01T08:53:10 < dongs> im getting some glitches which are i think due to interrupts nesting,, will ahve to check priority levels
2012-09-01T08:53:30 < zippe> nesting, or just plain interrupt latency?
2012-09-01T08:54:03 < zippe> In general, unless you have something that is really hard-realtime critical, nesting is just the wrong thing to do
2012-09-01T08:54:55 < dongs> i think nesting. in my osd shit, i have timer cc interrupt that catches hsync+starts spi dma, but if I have another timercc for a sensor that outputs data by pwm pulses, i get some noise/glitches in SPI output
2012-09-01T08:55:59 < zippe> You have them at the same priority?
2012-09-01T08:56:06 < dongs> i have not been able to capture the glitch so i dont know what exactly it affects (i.e. shifting pixels or dropping lines or what), but its definitely there if i dont conect the sensor
2012-09-01T08:56:17 < dongs> er -> definitely NOT there
2012-09-01T08:56:55 < dongs> hsync is 0,0 priority
2012-09-01T08:57:07 < dongs> sensor (less carea bout it) is like 3 priority
2012-09-01T08:57:32 < dongs> so uh.. it shouldnt nest then right?
2012-09-01T08:57:34 < zippe> Hmm, not sure what you mean by 0,0
2012-09-01T08:57:49 < dongs> sorry nvic has like 2 levels
2012-09-01T08:57:55 < dongs> channelprioremption and subpriority
2012-09-01T08:57:56 < dongs> or someshit
2012-09-01T08:58:05 < zippe> sec, I need to check the M4 TRM
2012-09-01T08:58:19 < zippe> Only a few of the MSBs in the priority are implemented
2012-09-01T08:58:26 < zippe> Most of the LSBs are ignored
2012-09-01T08:59:22 < dongs> right, but im using stdlib
2012-09-01T08:59:37 < dongs> i think it stuffs the preemption/subpriority shit into wahteve bits of nvic
2012-09-01T08:59:58 < dongs> anyway, point is, if my hsync is highest priority, why would it get glitched by timer capture with lower priority
2012-09-01T09:00:47 < dongs> maybe its actual bus traffic? i.e. spi DMA is moving bits, but sensor capture interrupt is doing some shit
2012-09-01T09:01:10 < dongs> so theres some latency between memory reads
2012-09-01T09:01:39 < zippe> Could certainly be that the DMA has lower priority on the fabric
2012-09-01T09:01:54 < zippe> and, hm.  I think my confusion about interrupt priority levels might be related to the M0
2012-09-01T09:02:13 < dongs> gonna check what trm says about dma priority levels
2012-09-01T09:02:18 < dongs> now my spi dma is at high
2012-09-01T09:02:22 < dongs> apparently trhere's'very high'
2012-09-01T09:02:25 < zippe> because the SCS register I'm thinking of is not in the M4 TRM
2012-09-01T09:02:48 < zippe> Any reason you're not using the capture to trigger the DMA directly?
2012-09-01T09:03:52 < dongs> i need to adjust buffer start and shit
2012-09-01T09:05:04 < zippe> Ah, right, B1.5.4
2012-09-01T09:05:36 < dongs> hm priority shit is just between other DMA requests.
2012-09-01T09:05:42 < dongs> so that doesnt matter
2012-09-01T09:06:17 < zippe> Ah, and 3.8 in the M4 TRM says "You configure the number of interrupts, and bits of interrupt priority, during implementation. Software can choose only to enable a subset of the configured number of interrupts, and can choose how many bits of the configured priorities to use."
2012-09-01T09:06:21 -!- dekar__ [~dekar@212.255.35.138] has joined ##stm32
2012-09-01T09:06:22 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-09-01T09:06:26 < dongs> yeah
2012-09-01T09:06:31 < zippe> Now, where is that SCS field
2012-09-01T09:06:31 < dongs> there's a function to set this
2012-09-01T09:06:41 < dongs> how many bits of priority shit to use or something.
2012-09-01T09:06:45 < dongs> i think mine is at default
2012-09-01T09:07:47 < dongs> anyway, i guess my glitches are just due to some traffic on internal buses
2012-09-01T09:08:51 < zippe> Oh, that worthless priority grouping stuff
2012-09-01T09:08:55 < dongs> cause i do hsync -> adjust buffer -> fire spi dma
2012-09-01T09:08:58 < zippe> I so totally put that out of my mind
2012-09-01T09:09:14 < zippe> Who cares which interrupt you choose to service first in a group?
2012-09-01T09:09:21 < dongs> heh
2012-09-01T09:09:27 < zippe> Unless they fire at exactly the same time, you're still going to block it
2012-09-01T09:10:01 < zippe> Sorry, where I was going earlier was that 0 and 3 are quite possibly the same thing
2012-09-01T09:10:01 -!- dekar_ [~dekar@212.255.35.25] has quit [Ping timeout: 245 seconds]
2012-09-01T09:10:12 < zippe> Try making your low-priority interrupt 255
2012-09-01T09:10:25 < zippe> That should guarantee that it's in a different group
2012-09-01T09:11:10 -!- AndTabSauce [~AndTabSau@blk-252-42-151.eastlink.ca] has quit [Quit: Bye]
2012-09-01T09:11:29 -!- AndTabSauce [~AndTabSau@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-01T09:11:32 < zippe> Ah, right
2012-09-01T09:11:40 < zippe> Which STm32 are you using?
2012-09-01T09:11:46 -!- AndTabSauce [~AndTabSau@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-01T09:11:47 < zippe> F103, IIRC
2012-09-01T09:12:00 < dongs> yea
2012-09-01T09:12:05 < zippe> The F4, at least, only implements the top 4 bits of the interrupt priority registers
2012-09-01T09:12:06 < dongs> pretty sure priority shit is same between f1/f4
2012-09-01T09:12:10 < dongs> yes, same
2012-09-01T09:12:31 < zippe> Ok, so I don't know what the library does with "3"
2012-09-01T09:12:58 < zippe> if that is really 0x30, then you should be OK
2012-09-01T09:13:08 < zippe> But if that's 0x03 then, no.
2012-09-01T09:13:19 < dongs> nah it shifts it around
2012-09-01T09:13:31 < dongs> http://bcas.tv/paste/results/AupQEC99.html
2012-09-01T09:14:20 < zippe> oh my fucking eyes
2012-09-01T09:15:40 < zippe> I honestly can't tell what that actually does
2012-09-01T09:15:41 < dongs> comeon, any worthwhile compiler will just optimize all that shit
2012-09-01T09:16:00 < zippe> I'm not talking about what the generated code looks like
2012-09-01T09:16:15 < zippe> I'm talking about the fact that no human should be forced to look at that
2012-09-01T09:16:51 < dongs> im pretty sure the bottom line is it puts priority, subpriority bits into relevant places in IP
2012-09-01T09:17:13 < zippe> You'd hope so
2012-09-01T09:17:14 < dongs> so you just specify [0..x] values per each bit set and it sticks them in correct places.
2012-09-01T09:17:26 < zippe> But it really couldn't go about it in a more obtuse fashion
2012-09-01T09:17:32 < dongs> haha
2012-09-01T09:17:33 < dongs> right
2012-09-01T09:17:59 < dongs> i think it pulls the number of used subpriroity bits (the amount of shift by) from aircr
2012-09-01T09:18:03 < zippe>  tmppriority = (0x700 - ((SCB->AIRCR) & (uint32_t)0x700))>> 0x08;
2012-09-01T09:18:23 < zippe> Yes, that's a pretty good capsule description of what's wrong with the ST library right there
2012-09-01T09:18:31 < dongs> heh
2012-09-01T09:18:38 < dongs> actuall i think thats CMSIS
2012-09-01T09:18:48 < dongs> ah no, thats misc,c
2012-09-01T09:18:50 < dongs> by ST
2012-09-01T09:18:52 < zippe> Yes
2012-09-01T09:19:13 < zippe> That code was written by someone that has no idea how the language works
2012-09-01T09:19:23 < zippe> And it is fundamentally the norm throughout the entire library.
2012-09-01T09:20:04 < zippe> What part of ((SCB->AIRCR >> 8) & 0x7) do they not understand?
2012-09-01T09:20:36 < dongs> but thats what it ends up being with compiler anyway
2012-09-01T09:21:03 < zippe> You don't seem to understand that source code is primarily written for the programmer to read, not the compiler.
2012-09-01T09:21:30 < zippe> Your code is what you're leaving for the next guy who comes after you to pick up
2012-09-01T09:22:07 < zippe> And this is the sort of code that pooper scooper laws were written about
2012-09-01T09:23:20 -!- claude is now known as Claude
2012-09-01T09:48:39 < zippe> Ah, I love chinese velcro
2012-09-01T09:48:49 < zippe> Where the adhesive doesn't actually stick to the back of the strip
2012-09-01T09:50:16 < dongs> :(
2012-09-01T09:51:14 -!- izua [~izua@188.26.165.54] has joined ##stm32
2012-09-01T09:51:15 -!- izua [~izua@188.26.165.54] has quit [Changing host]
2012-09-01T09:51:15 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T09:51:18 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T09:53:30 -!- izua_ [~izua@188.26.164.75] has joined ##stm32
2012-09-01T09:53:31 -!- izua_ [~izua@188.26.164.75] has quit [Changing host]
2012-09-01T09:53:31 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T09:53:32 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-01T09:56:19 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-01T10:10:22 < dongs> im having a fucked up problem of solder not sticking to ENIG'd pads between different batches of boards
2012-09-01T10:10:30 < dongs> (when pasting)
2012-09-01T10:10:40 < dongs> is it stencil or pcb problem?
2012-09-01T10:11:40 < dongs> gonna ask boardhouse on monday but figured if someone knows.
2012-09-01T10:37:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-01T10:57:57 -!- tavish3 [~tavish@125.63.106.158] has joined ##stm32
2012-09-01T11:01:00 < tavish3> hi, i have a probably silly question about USB OTG on stm32f4; can you use it as a usb host and usb device simultaneously without an external USB PHY chip?
2012-09-01T11:03:00 -!- tavish3 is now known as tavish
2012-09-01T11:03:11 -!- tavish [~tavish@125.63.106.158] has quit [Changing host]
2012-09-01T11:03:11 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-01T11:03:59 < dongs> tavish: thre's some dualcore samples in the usb-lib
2012-09-01T11:05:49 < dongs> < dongs> http://www.st.com/internet/com/SOFTWARE_RESOURCES/SW_COMPONENT/FIRMWARE/stm32_f105-07_f2_f4_usb-host-device_lib.zip
2012-09-01T11:05:52 < dongs> that one.
2012-09-01T11:06:48 < tavish> dongs: i see a dual role device example, and its readme says you have to choose the host/device role using a joystick on the eval board the example is for
2012-09-01T11:07:38 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-01T11:08:17 < dongs> thats about my involvement with it, so I dont know :)
2012-09-01T11:08:52 < tavish> :( ok, thanks anyways
2012-09-01T11:10:14 < dongs> http://cdn.vicorpower.com/documents/datasheets/Picor/ds_pi33xx.pdf thats a cool little reg
2012-09-01T11:10:23 < tavish> plus this USB-device library doesnt seem to compile. pdev->dev.test_mode doesnt exist, so do a few more members
2012-09-01T11:11:06 < dongs> sureley ST wouldnt ship shit that doesnt work.
2012-09-01T11:11:11 < dongs> i'd guess its ytour compiler or build settings.
2012-09-01T11:11:56 < tavish> it probably is, i have used it before
2012-09-01T11:12:04 < tavish> just can't find what
2012-09-01T11:25:27 -!- barthess [~barthess@178.154.68.55] has joined ##stm32
2012-09-01T11:26:37 -!- izua_ [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-01T11:30:43 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T11:32:07 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 268 seconds]
2012-09-01T11:39:05 -!- soafee-chan is now known as spacekitteh
2012-09-01T11:43:35 < cjbaird> Hmm. Yesterday's comments about the overkill unit testing for a APRS/GPS interface.. I recall they're trying to test it in ranges that exceed the US-military-mandated limits of consumer GPS modules.. (10kms altitude? 200m/s?)
2012-09-01T11:47:08 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T11:48:44 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 240 seconds]
2012-09-01T11:50:53 < cjbaird> COCOM limits: 60k feet altitute, /and-or/ 999 knot speed limit. Garmin eTrex only have the speed limit.
2012-09-01T11:55:56 -!- spacekitteh [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T11:56:31 -!- soafee-chan [~spaec@unaffiliated/traumapony] has quit [Disconnected by services]
2012-09-01T12:03:03 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 244 seconds]
2012-09-01T12:12:31 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-01T12:34:03 -!- barthess [~barthess@178.154.68.55] has quit [Quit: Leaving.]
2012-09-01T12:35:09 -!- spacekitteh [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T12:40:45 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-01T12:58:58 -!- dekar [~dekar@212.255.34.110] has joined ##stm32
2012-09-01T12:58:59 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-01T13:01:01 -!- dekar__ [~dekar@212.255.35.138] has quit [Ping timeout: 260 seconds]
2012-09-01T13:05:04 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-01T13:15:20 < dongs> http://www.digikey.com/Web%20Export/Supplier%20Content/ADI_505/MKT/NPE/2012/npe_head.jpg
2012-09-01T13:15:24 < dongs> way to go, internet
2012-09-01T13:38:45 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-01T13:44:29 < zyp> tavish, yes, it is possible
2012-09-01T13:45:16 < zyp> the point of the ID line is to detect whether a host or a device is plugged in and select the according mode
2012-09-01T13:45:45 < dongs> i think he meant at the same time.
2012-09-01T13:46:11 < dongs> keyword being "simultaneously"
2012-09-01T13:46:35 < zyp> well
2012-09-01T13:46:43 < dongs> and i was dumb for answering maybe
2012-09-01T13:46:51 < dongs> since thereis only one set of usb dp./dm lines
2012-09-01T13:47:02 < dongs> the other one would have to come from phy
2012-09-01T13:47:04 < zyp> each controller can only be in either host or device mode
2012-09-01T13:47:11 < tavish> dongs: there's 2 lines actually
2012-09-01T13:47:15 < dongs> tavish: ?
2012-09-01T13:47:17 < dongs> where
2012-09-01T13:47:23 < zyp> but there are two controllers, so you can have one in each mode simultanously
2012-09-01T13:47:33 < zyp> OTG_HS have an internal FS phy
2012-09-01T13:47:40 < zyp> in addition to OTG_FS
2012-09-01T13:48:44 < zyp> and I can guarantee that the internal PHY works, because I'm using that on my board
2012-09-01T13:48:44 < tavish> zyp: that's what i had been trying to find ot, since i have a olimex board woth both usb host and device ports
2012-09-01T13:49:15 < dongs> of course internal phy works
2012-09-01T13:49:20 < dongs> its on f4discovery board
2012-09-01T13:49:31 < zyp> dongs, no, that's OTG_FS
2012-09-01T13:49:51 < zyp> but OTG_HS also have one, on some other pins
2012-09-01T13:50:08 < tavish> yeah
2012-09-01T13:50:20 < zyp> OTG_FS is using PA9-12, OTG_HS is using PB12-15
2012-09-01T13:50:54 < zyp> and there is no reason both can't be used at the same time
2012-09-01T13:51:36 < dongs> ohh
2012-09-01T13:52:06 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 272 seconds]
2012-09-01T13:53:23 < dongs> lookin at datasheet
2012-09-01T13:54:17 < dongs> shit, urrite
2012-09-01T13:54:35 < dongs> now for the real mindfuck: can you take a Usb cable and loop the host into device ports
2012-09-01T13:55:09 < zyp> sure, it would be as useful as hooking a uart to another or a spi slave to a spi master
2012-09-01T13:56:18 < dongs> that is pretty cool though.
2012-09-01T13:56:29 < dongs> shit's even available on 64pin version..
2012-09-01T13:56:46 < zyp> sure, I've got a 64-pin version on my board
2012-09-01T13:56:56 < zyp> and I'm using OTG_HS with internal FS phy
2012-09-01T13:57:19 < zyp> because OTG_FS was supposed to be unavailable on F405
2012-09-01T13:57:27 < dongs> but it was?
2012-09-01T13:57:37 < zyp> they later revided the datasheet
2012-09-01T13:57:50 < dongs> ha
2012-09-01T13:57:50 < zyp> and the register block is present in my chips
2012-09-01T14:03:20 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has joined ##stm32
2012-09-01T14:03:21 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has quit [Changing host]
2012-09-01T14:03:21 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32
2012-09-01T14:44:04 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-01T14:44:05 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-01T14:44:05 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T14:44:06 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T14:45:54 -!- Laurenceb_ [~Laurence@host86-177-215-42.range86-177.btcentralplus.com] has joined ##stm32
2012-09-01T14:48:21 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-01T14:52:11 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-01T14:54:14 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-01T14:54:15 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-01T14:54:15 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T14:54:18 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T15:05:09 < cjbaird> lol KitaEri
2012-09-01T15:14:10 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-01T15:16:57 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-01T15:19:13 < dongs> sup?
2012-09-01T15:19:39 < dongs> sounds like cjbaird is fapping to some anniemay again
2012-09-01T15:20:33 <+izua> dongs: what is it with those japanese cartoons bro
2012-09-01T15:20:34 <+izua> it's like
2012-09-01T15:20:41 < dongs> why are you asking ME
2012-09-01T15:20:44 < dongs> cjbaird + Laurenceb_ are the experts
2012-09-01T15:20:52 < dongs> and zyp
2012-09-01T15:26:15 < cjbaird> tl;dr - anniemay producers did a rather savage prank on a newbie voice actor, possibly ruining his potential career. The real VAs laugh at the poor guy on TV. 2ch mad. 4chan mad. 4channers do a dickraid on the perp's girlfriend's twitter. A complete top-down Ubershitstorm; with the cartoon involved looking like it'll be cancelled mid-season.
2012-09-01T15:29:39 -!- barthess [~barthess@178.154.74.110] has joined ##stm32
2012-09-01T15:30:00 < zyp> which one?
2012-09-01T15:31:11 < cjbaird> Kokoro Connect. http://archive.foolz.us/a/thread/70914943/
2012-09-01T15:33:40 -!- Claude is now known as claude
2012-09-01T15:36:23 < dongs> looks like standard pedo shit.
2012-09-01T15:37:52 < cjbaird> Where is this week's ep of Space Bros. :/
2012-09-01T15:38:12 < zyp> it didn't air last weekend
2012-09-01T15:38:13 < Laurenceb_> PEADOS?! ON MY IRC? ITS MORE LIKELY THAN YOU THINK
2012-09-01T15:39:08 < Laurenceb_> lol EMF are launching balloonz
2012-09-01T15:39:10 < Laurenceb_> http://spacenear.us/tracker/
2012-09-01T15:39:37 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-09-01T15:39:40 < Laurenceb_> bet those nerds are enjoying their motorway noise
2012-09-01T15:39:48 < dongs> bad news for cjbaird Bandai Entertainment announced today the discontinuation of Blu-ray, DVD, comic and novel sales. The final shipment of comic and novels to retailers will be made at the end of October 2012
2012-09-01T15:40:02 < Laurenceb_> oh noes
2012-09-01T15:40:22 < BrainDamage> Laurenceb_:  why aren't you helping them?
2012-09-01T15:40:28 < BrainDamage> you belong to them anyway :p
2012-09-01T15:43:22 < cjbaird> _pay_ for anime and its merchandising?
2012-09-01T15:53:00 < Laurenceb_> in not _that_ much of a nerd
2012-09-01T15:56:19 -!- tech2077 [~tech2077@75.53.132.42] has quit [Read error: Connection reset by peer]
2012-09-01T15:57:31 < dongs> lawl, twitter account terminated
2012-09-01T15:57:35 < dongs> were there really dicks
2012-09-01T16:02:10 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-09-01T16:03:30 < Laurenceb_> erm i hope that thing doesnt burst over london
2012-09-01T16:03:50 < Laurenceb_> going to get missiled from the special olympics
2012-09-01T16:07:04 -!- Netsplit over, joins: @ChanServ
2012-09-01T16:07:14 < dongs> haha
2012-09-01T16:07:23 < dongs> youll be introuble
2012-09-01T16:07:29 < dongs> i think im going to sleep early
2012-09-01T16:09:33 -!- tech2077 [~tech2077@adsl-75-53-130-200.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-01T16:10:31 -!- daku is now known as DaKu
2012-09-01T16:10:32 < cjbaird> Predicted to kill a few Wombles.
2012-09-01T16:24:44 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 272 seconds]
2012-09-01T16:27:59 < Laurenceb_> http://shop.shopatcuretime.co.uk/epages/es122234.sf/en_GB/?ObjectPath=/Shops/es122234/Products/ALUMINIUMFIK01
2012-09-01T16:28:06 < Laurenceb_> this looks relevant to my interests
2012-09-01T16:28:36 < Laurenceb_> i should get back into diy explosives :P
2012-09-01T16:28:45 < cjbaird> 'joey' has started ascening rapidly..
2012-09-01T16:29:38 < Tzacks> Laurenceb_: r u into rocker propellents? :)
2012-09-01T16:29:43 < Tzacks> rocket*
2012-09-01T16:29:47 < Laurenceb_> i used to make pipe bombs
2012-09-01T16:30:12 < Laurenceb_> but ive got a large shrapnel mark down the side of my neck from that
2012-09-01T16:30:29 < Tzacks> i wanted to write thats quite dangerous :)
2012-09-01T16:30:35 < Laurenceb_> hehe
2012-09-01T16:30:39 < Laurenceb_> fuse went wrong
2012-09-01T16:30:57 < Tzacks> :)
2012-09-01T16:30:58 < cjbaird> I could mention something, but... publically-logged channel.. -_-
2012-09-01T16:31:06 < Laurenceb_> lol
2012-09-01T16:31:14 < Tzacks> mention something then :)
2012-09-01T16:31:18 < Laurenceb_> that could involve works of fiction
2012-09-01T16:31:34 < jpa-> "so, last night i had this dream that.."
2012-09-01T16:33:01 < cjbaird> In one of stories, a significant toilet-resembling landmark in Newcastle was destroyed by such a improvised munition..
2012-09-01T16:34:06 < cjbaird> I didn't write this story, either.. a housemate from when I lived in Armidale did. :)
2012-09-01T16:34:19 < Tzacks> :)
2012-09-01T16:36:01 < Laurenceb_> i was actually looking for carbon black
2012-09-01T16:36:06 < Tzacks> cjbaird: hope you didn't get shit all over the newcastle that day :)
2012-09-01T16:36:06 < Laurenceb_> anyone know where to buy it?
2012-09-01T16:37:05 < Tzacks> you mean like fibre?
2012-09-01T16:37:11 < Laurenceb_> no
2012-09-01T16:37:15 < Laurenceb_> powder
2012-09-01T16:38:46 < Tzacks> no idea but some pencils and a sandpaper would help :)
2012-09-01T16:45:34 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-01T16:45:46 < Thorn> http://www.supremepower.co.uk/Most%20Useless%20Machine.gif/view
2012-09-01T16:46:05 < jpa-> Laurenceb_: maybe http://www.ebay.com/itm/Graphite-Powder-100g-Bag-Pure-Natural-Graphite-/130447167247?pt=UK_Crafts_DrawingSupplies_EH&hash=item1e5f41cf0f  ?
2012-09-01T16:46:21 < karlp> damn you guys have been chatty the last couple of days
2012-09-01T16:46:31 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-01T16:54:34 < Laurenceb_> jpa-: im not sure its the same stuff
2012-09-01T16:54:42 < Laurenceb_> carbon black is formed from soot
2012-09-01T16:54:59 < Laurenceb_> and has micron sized particles with chaotic structure
2012-09-01T16:55:16 < Laurenceb_> graphite powder would suggest powdered graphite, which wont be as good
2012-09-01T16:56:06 < Laurenceb_> i need IR absorption in silicone rubber
2012-09-01T17:01:06 < cjbaird> Indian ink?
2012-09-01T17:13:32 < cjbaird> The balloon has frozen, or ran out of battery..
2012-09-01T17:20:42 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T17:22:18 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 264 seconds]
2012-09-01T17:25:27 -!- soafee-chan [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 244 seconds]
2012-09-01T17:25:34 < dongs> cjbaird: haha, i pasted hackkitten in another channel and one of hte dudes got back with this
2012-09-01T17:25:38 < dongs> http://www.youtube.com/watch?v=-Xpg7VVwmcY
2012-09-01T17:26:56 < BrainDamage> Laurenceb_: wrt aluminium powder, you can diy a ball mill using a plastic bottle, some ball bearings and a small dc motor with a reducer
2012-09-01T17:27:40 < BrainDamage> don't use iron if you plan to use the mill on gunpowder or similar
2012-09-01T17:27:49 < BrainDamage> ( for the ball bearings )
2012-09-01T17:30:38 < cjbaird> dongs: shit, an actual furry herm.
2012-09-01T17:31:16 < dongs> sounds pretty manly
2012-09-01T17:32:41 < cjbaird> Other content like: https://www.youtube.com/watch?v=NrabwiQnyOo
2012-09-01T17:33:09 < BrainDamage> one of my classmates is an actual furry
2012-09-01T17:33:44 < BrainDamage> if you pass behind him, he pourposedly puts on furry bs to bother you
2012-09-01T17:33:49 < cjbaird> Was one. Gave up 15 years ago, when it turned freaky and stopped being about a type of selfpub comics.
2012-09-01T17:34:07 < BrainDamage> he has a fursuit and everything
2012-09-01T17:34:13 < BrainDamage> and puts on pics of himself ...
2012-09-01T17:34:24 -!- spacekitteh [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-01T17:36:49 < dongs> found zyp's car http://a4.sphotos.ak.fbcdn.net/hphotos-ak-prn1/64113_281745218564765_1510110067_n.jpg
2012-09-01T17:38:59 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Client Quit]
2012-09-01T17:39:32 < cjbaird> found my scooter: https://secure.flickr.com/photos/brucetsao/7284533080/
2012-09-01T17:41:20 < cjbaird> haha https://secure.flickr.com/photos/brucetsao/7284298020/
2012-09-01T17:49:16 < cjbaird> dongs: Re: Maya .. Not too surprised. The most technically-literate girl I know IRL has a trisomy 'triple x chromosome'. She's still 100% female.. just often mistaken for a tranny. :/
2012-09-01T17:50:25 < dongs> cjbaird: uh yeah, well ,except this one doesn't need any dudeproofin'
2012-09-01T17:54:05 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-01T18:18:06 -!- tech2077 [~tech2077@adsl-75-53-130-200.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 264 seconds]
2012-09-01T18:29:59 -!- tech2077 [~tech2077@adsl-75-53-141-210.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-01T19:12:26 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-01T19:17:49 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-01T19:23:11 < Laurenceb_> doh
2012-09-01T19:23:33 < Laurenceb_> EMF crew crashed a payload into central london at high speed
2012-09-01T19:23:43 < Laurenceb_> their going to get locked up for that
2012-09-01T19:24:02 < Laurenceb_> luckly it seems no children were underneath
2012-09-01T19:33:29 < karlp> all of the aerial stuff's going to the shit regulated out of it in the next few years
2012-09-01T19:49:44 < qyx_> Laurenceb_: do you have some link?
2012-09-01T19:51:19 -!- tech2077 [~tech2077@adsl-75-53-141-210.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-01T19:51:50 < BrainDamage> karlp: it will eventually self regulate
2012-09-01T19:52:10 < BrainDamage> due to super retarded  american administration policies, they are selling helium severly underpriced
2012-09-01T19:52:21 < BrainDamage> they had the vastest helium deposite, and want to sell it off
2012-09-01T19:52:31 < karlp> BrainDamage: I didn't mean just balloons
2012-09-01T19:52:43 < BrainDamage> drones etc?
2012-09-01T19:52:45 < karlp> yeah
2012-09-01T19:52:57 < BrainDamage> there's already very restrictive measures in many countries
2012-09-01T19:53:03 < BrainDamage> like you're los limited
2012-09-01T19:53:09 < karlp> but yeah, I'd heard there was some crazy weird stuff wrt helium supplies
2012-09-01T19:53:51 < karlp> there's technically very restrictive measures in lots of places, but a whole lot of enforcement that's never actually had to deal wit the reality of there being hundreds or thousands of law breakers at a given time, not just one.
2012-09-01T19:58:44 < BrainDamage> I guess because the technology is neither cheap nor commonplace atm, the "dumb masses" didn't spoil the field yet
2012-09-01T19:58:54 < BrainDamage> I guess you're right with drone prices crashing
2012-09-01T20:07:07 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-01T20:10:29 -!- enots_ [dimka@freelsd.net] has joined ##stm32
2012-09-01T20:11:32 -!- barthess1 [~barthess@178.154.74.110] has joined ##stm32
2012-09-01T20:17:39 -!- barthess [~barthess@178.154.74.110] has quit [Ping timeout: 246 seconds]
2012-09-01T20:17:39 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 246 seconds]
2012-09-01T20:17:40 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-09-01T20:18:27 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 276 seconds]
2012-09-01T20:20:50 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-01T20:42:30 -!- enots_ is now known as enots
2012-09-01T21:51:20 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-01T22:00:11 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T22:00:11 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T22:00:47 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-01T22:07:54 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-01T22:07:55 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-01T22:07:55 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-01T22:07:58 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-01T22:09:39 -!- barthess1 [~barthess@178.154.74.110] has quit [Quit: Leaving.]
2012-09-01T22:49:19 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-01T22:49:19 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-01T23:16:59 < Laurenceb_> https://twitter.com/b3noxley/status/241895144847183872/photo/1/large
2012-09-01T23:17:03 < Laurenceb_> sausagecamp
2012-09-01T23:46:04 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Read error: Connection timed out]
2012-09-01T23:46:52 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
--- Day changed Sun Sep 02 2012
2012-09-02T00:00:45 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 276 seconds]
2012-09-02T00:09:35 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-09-02T00:13:39 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-02T00:24:31 < Laurenceb_> http://www.circlon-theory.com/HTML/EmcFallacies.html
2012-09-02T00:24:33 < Laurenceb_> wut
2012-09-02T00:30:12 -!- metaxa [metaxa@us.members.bombshellz.net] has quit [Quit: leaving]
2012-09-02T00:32:57 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-09-02T00:33:11 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-09-02T00:56:57 -!- tech2077 [~tech2077@adsl-75-53-141-210.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-02T01:07:52 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 272 seconds]
2012-09-02T01:11:56 < Laurenceb_> http://uk.farnell.com/bak/lp-523450p-is-3/battery-lithium-pol-3-7v-980-mah/dp/2077888
2012-09-02T01:12:06 < Laurenceb_> nice - farnell finally stocking lipo cells
2012-09-02T01:16:55 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-02T01:29:00 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-02T02:02:34 -!- barthess [~barthess@164.177.204.111] has joined ##stm32
2012-09-02T02:14:25 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-02T02:18:32 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-02T02:32:46 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-02T02:35:00 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-02T02:57:37 -!- Laurenceb_ [~Laurence@host86-177-215-42.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-02T02:58:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-02T03:03:11 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-02T04:21:48 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-02T04:23:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-02T04:46:18 -!- barthess [~barthess@164.177.204.111] has quit [Quit: Leaving.]
2012-09-02T04:52:31 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-02T04:59:27 < emeb_mac> Hello  hello   ello  llo  o  o  .   .
2012-09-02T05:01:24 < dongs> dead chan is dead
2012-09-02T05:17:57 < dongs> silly question
2012-09-02T05:19:17 < dongs> if I run a memory>periph dma transfer, memory_inc is on, once it runs to CNDTR, if I just re-enable it again what happens? will it continue at the end of memory block where it stopped?
2012-09-02T05:19:29 < dongs> or do  i need to reset cmar+cndtr each time
2012-09-02T05:21:39 < dongs> lets try it out
2012-09-02T05:21:59 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-02T05:22:00 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-02T05:22:00 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-02T05:22:03 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-02T05:32:41 < dongs> hmm.
2012-09-02T05:32:45 < dongs> not doing waht i expect.
2012-09-02T05:33:39 < emeb_mac> what's it do?
2012-09-02T05:38:01 < dongs> not waht i expect :\
2012-09-02T05:47:01 < emeb_mac> reading the RM - looks like it should pick up where it was when it stopped. If it ran to completion you need to reload the DMA stream settings.
2012-09-02T05:47:38 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-02T05:48:43 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 240 seconds]
2012-09-02T05:48:48 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-02T05:49:05 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-02T05:53:29 < dongs> define "completion"
2012-09-02T05:53:41 < dongs> my ram block for transfer is huge
2012-09-02T05:53:46 < dongs> but I only transfer x bytes at a time
2012-09-02T05:53:52 < dongs> but not at once
2012-09-02T05:53:58 < dongs> so i set cndtr to say 100 bytes, transfer
2012-09-02T05:54:10 < dongs> is that "completion", and I need to reload memory address too?
2012-09-02T05:54:15 < dongs> or just re-set cndtr for another 100 bytes
2012-09-02T05:54:20 < dongs> anyway, idid ^ and that dint work
2012-09-02T06:09:24 < emeb_mac> here's what the rm says: The stream suspends the transfer before the number of remaining data items to be transferred in the DMA_SxNDTR register reaches 0. The aim is to restart the transfer later by re-enabling the stream. In order to restart from the point where the transfer was stopped, the software has to read the DMA_SxNDTR register after disabling the stream by writing the EN bit in DMA_SxCR register (and then checking that it is at ‘
2012-09-02T06:09:24 < emeb_mac> to know the number of data items already collected. Then:
2012-09-02T06:09:24 < emeb_mac> –	The peripheral and/or memory addresses have to be updated in order to adjust the address pointers
2012-09-02T06:09:24 < emeb_mac> –	The SxNDTR register has to be updated with the remaining number of data items to be transferred (the value read when the stream was disabled)
2012-09-02T06:09:25 < emeb_mac> –	The stream may then be re-enabled to restart the transfer from the point it was stopped
2012-09-02T06:10:28 < dongs> hm
2012-09-02T06:10:31 < dongs> thjat does not apply to me
2012-09-02T06:10:40 < dongs> as far as the dma is concerned the transfer is "complete"
2012-09-02T06:10:58 < dongs> i am not stopping it halfway through.
2012-09-02T06:11:15 < dongs> anyway, its not that hard to advance buffer pointer.
2012-09-02T06:11:24 < dongs> i just figured i save an instruction or two by letting it do that automatically
2012-09-02T06:11:44 < emeb_mac> yeah - sounds like you need to handle that "manually"
2012-09-02T06:14:19 -!- tech2077 [~tech2077@adsl-75-53-141-210.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-02T06:25:25 -!- spacekitteh [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-02T06:26:11 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-02T06:26:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-02T06:26:22 -!- tech2077 [~tech2077@adsl-75-53-132-100.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-02T06:30:20 < dongs> hmm, rtos is fucking me over
2012-09-02T06:37:12 < dongs> as soon as it switches context all global vars are gone
2012-09-02T06:37:12 < dongs> hmm
2012-09-02T06:39:58 < zippe> "all global vars are gone"
2012-09-02T06:40:01 < zippe> Are you compiling pic?
2012-09-02T06:40:42 < dongs> no
2012-09-02T06:41:02 < dongs> once it switches context, my pointers to data become pointers  to... some crap in flash., usually random
2012-09-02T06:42:09 < dongs> static global vars shouldnt be affected by context switchign right.
2012-09-02T06:46:22 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Operation timed out]
2012-09-02T07:05:37 < zippe> It depends
2012-09-02T07:05:56 < zippe> If your toolchain is building PIC and the base register is not being respected, then yes, you'll be screwed
2012-09-02T07:06:32 < zippe> That's the most likely option
2012-09-02T07:06:36 < dongs> nah, its not ropi/rwpi.
2012-09-02T07:06:38 < dongs> hmm
2012-09-02T07:06:49 < zippe> Alternately, are you running at 0 or at 0x08?
2012-09-02T07:07:11 < dongs> you mean flash start?
2012-09-02T07:07:20 < zippe> yes
2012-09-02T07:07:28 < dongs> yeah, 0 i guess? defaults.
2012-09-02T07:07:38 < zippe> 0x08 should be the default
2012-09-02T07:07:41 < dongs> after switch, shit ends up being in 0x08xxxxxxx
2012-09-02T07:07:48 < dongs> stuff that was in ram before
2012-09-02T07:07:49 < zippe> Yes, that's expected
2012-09-02T07:08:10 < zippe> Hmm, is this just in the function that was pre-empted?
2012-09-02T07:08:16 < dongs> yeah
2012-09-02T07:08:20 < dongs> like, if im stepping it
2012-09-02T07:08:21 < zippe> Ah, ok.
2012-09-02T07:08:27 < zippe> Well, don't trust stepping
2012-09-02T07:08:39 < zippe> But if you are seeing runtime crashes then it sounds like save/restore issues
2012-09-02T07:08:41 < zippe> Which RTOS?
2012-09-02T07:08:44 < dongs> i do blah = foo(); // return ptr to some static buffer
2012-09-02T07:08:54 < dongs> on step, it points to ram, correct place
2012-09-02T07:09:14 < dongs> then i have while (1) { tickdelayblah(100); //this will yield etc
2012-09-02T07:09:23 < dongs> after tickdelay returns, blah = randomized
2012-09-02T07:09:31 < zippe> what register is it in?
2012-09-02T07:09:59 < zippe> and is tickdelay() a function or inlined?
2012-09-02T07:10:13 < dongs> its not in register, its in static memory block that belongs to that task
2012-09-02T07:10:28 < zippe> In that function, 'blah' will be in a register
2012-09-02T07:10:35 < zippe> when foo() returns it will be in r0
2012-09-02T07:10:42 < zippe> It will be copied somewhere else.
2012-09-02T07:10:45 < dongs> right, but i mean, i assign the return to a struct member
2012-09-02T07:10:50 < zippe> doesn't matter
2012-09-02T07:10:55 < dongs> localstruct.blah = foo();
2012-09-02T07:11:01 < zippe> yeah, irrelevant
2012-09-02T07:11:05 < dongs> then after tickdelay, localstruct is all trashed
2012-09-02T07:11:06 < dongs> including the blah.
2012-09-02T07:11:19 < zippe> localstruct is on the stack?
2012-09-02T07:11:24 < zippe> Or in bss?
2012-09-02T07:11:24 < dongs> no, its global static
2012-09-02T07:11:26 < zippe> or data segment?
2012-09-02T07:11:41 < zippe> Where is it relative to your stacks?
2012-09-02T07:11:52 < dongs> its near begining of sram so
2012-09-02T07:12:03 < dongs> like 0x20000600-ish
2012-09-02T07:12:04 < zippe> so … ?
2012-09-02T07:12:12 < zippe> What's msp, psp?
2012-09-02T07:12:12 < dongs> stacks should be further up?
2012-09-02T07:12:19 < zippe> stacks can be wherever you like
2012-09-02T07:12:37 < zippe> it's popular to put the interrupt stack at the bottom of RAM so that you hardfault when you overflow it
2012-09-02T07:12:42 < zippe> dumb if you ask me, but still
2012-09-02T07:12:52 < zippe> since you will go to lockup mode, but w/e
2012-09-02T07:12:54 < dongs> msp = 0x2000039c8; psp = 0x20000638
2012-09-02T07:13:10 < zippe> well, that would seem to be a problem then, wouldn't it?
2012-09-02T07:13:12 < dongs> hmm.. waht the fuck
2012-09-02T07:13:19 < dongs> thats like 4 bytes after my data struct
2012-09-02T07:13:45 < dongs> sturct is at 0x20000634
2012-09-02T07:14:02 < dongs> what did i fuck up? :p
2012-09-02T07:14:19 < zippe> Probably clicked the wrong button somewhere
2012-09-02T07:14:25 < zippe> Moved the 'stack size' slider to the wrong place
2012-09-02T07:14:37 < zippe> Maybe a mouse gesture in the compile window activated "super wizard mode"
2012-09-02T07:14:43 < dongs> ha ha.
2012-09-02T07:14:44 < dongs> super funny.
2012-09-02T07:15:15 < dongs> psp = task stack pointer?
2012-09-02T07:15:15 < dongs> or wat
2012-09-02T07:15:30 < zippe> I'm surprised you didn't get a pop-up telling you that for the small one-time charge of $799.95 you could talk to someone in a call-center in outer Mongolia where they woudl debug your code for just $11.95 per line...
2012-09-02T07:15:37 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-02T07:15:41 < zippe> psp is the "program" stack pointer
2012-09-02T07:15:54 < zippe> You can use it for non-interrupt code
2012-09-02T07:16:02 < zippe> i.e. thread mode
2012-09-02T07:17:05 < zippe> I don't know which RTOS, so I can't comment, but in general if I had to guess you want to go look real hard at its assumptions about where program stacks come from.
2012-09-02T07:17:20 < zippe> where 'program' includes the idle task
2012-09-02T07:17:58 < zippe> meanwhile, I will go debug someone else's linked list code.  Me, with a brain the size of a planet...
2012-09-02T07:20:41 < dongs> oops i think i found the problem
2012-09-02T07:21:14 < dongs> yep
2012-09-02T07:21:18 < dongs> loldongs.
2012-09-02T07:21:44 < dongs> i need to point to stack-1 when initing task
2012-09-02T07:21:57 < dongs> copypaste failed and i was pointing to end of stack buffer.
2012-09-02T07:40:14 < dongs> huh
2012-09-02T07:40:53 < dongs> i did NVIC_PriorityGroupConfig(NVIC_PriorityGroup_2 and all my glitches during spi transfers are gone
2012-09-02T07:41:43 < dongs> group2 is 2 bits for pre-emption priority and 2 bbits for sub
2012-09-02T07:42:57 < dongs> shit, even with usb running no glitches. ok, i win
2012-09-02T07:45:53 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-02T07:46:31 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-02T08:03:57 < cjbaird> I went to buy a new MP3 (+FM+mic) player just then. The Sony options had "YOU MUST NOT USE THIS IN GLORIOUS NIPPON" warnings on them..
2012-09-02T08:04:44 < dongs> wat
2012-09-02T08:04:59 < dongs> o rite
2012-09-02T08:05:04 < dongs> jap FM is like fucked band
2012-09-02T08:07:25 < cjbaird> You idiots took up NTSC for video too. Got an iRiver e300 instead.
2012-09-02T08:09:55 < cjbaird> I was thinking it was more because of Sony's love for region locked markets..
2012-09-02T08:33:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-02T08:33:21 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 245 seconds]
2012-09-02T08:43:08 < dongs> NTSC-J, bro
2012-09-02T08:43:15 < dongs> just enough to be slightly incompatible
2012-09-02T08:43:39 < dongs> but we got you wiht digital tv.
2012-09-02T08:43:52 < dongs> the most expensive standard, and some dick sucking made sure half of south america chose it too.
2012-09-02T08:44:17 < dongs> last thing poor shit countries needed was dtv standard that was so fucking expensive noone could afford even a converter box
2012-09-02T08:53:39 < cjbaird> DTV is called bittorrent here
2012-09-02T08:54:15 < cjbaird> (woohoo-- new Dr Who Dalek episode fresh from the cybernets)
2012-09-02T08:56:17 < dongs> australia has internet fast enough for buttorrent?
2012-09-02T08:57:39 < cjbaird> It's still faster than the television networks showing stuff.
2012-09-02T09:00:50 -!- dekar_ [~dekar@212.255.33.108] has joined ##stm32
2012-09-02T09:00:53 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-02T09:04:11 -!- dekar [~dekar@212.255.34.110] has quit [Ping timeout: 245 seconds]
2012-09-02T09:04:27 < dongs> You can't call CoSetFlag() to set the flags in the ISR, otherwise, the system would be in chaos.
2012-09-02T09:04:31 < dongs> nice docs.
2012-09-02T09:04:52 < zippe> CoOS?  You poor sucker.
2012-09-02T09:05:03 < dongs> nah it works fine
2012-09-02T09:18:17 -!- claude is now known as Claude
2012-09-02T09:19:25 < ABLomas> "<     dongs> the most expensive standard, and some dick sucking made sure half of south america chose it too." - haha, already in quote book =)
2012-09-02T09:21:54 < dongs> true story
2012-09-02T09:22:17 < dongs> it had to involve dick between brazil+jp because no sane country would choose ISDB (as noone has except japan)
2012-09-02T09:23:39 < cjbaird> JP tried a lot to get their DTV standard in AU as well.. Too bad for them, it took another 15 years for the Government to release it needed one.
2012-09-02T09:24:20 < cjbaird> *realise
2012-09-02T09:25:55 < dongs> ya they werent dumb enough to fall for it though
2012-09-02T09:27:34 < cjbaird> And then, all the Commercial television networks were going GIVE MONIES PLZ for the conversion-- and the 'conservative' government of the time did (and naturally, didn't want to give money to the public broadcasters).
2012-09-02T09:53:36 < dongs> meanwhile in soviet russia http://www.youtube.com/watch?v=sp5M9K3wZcY
2012-09-02T09:57:40 < cjbaird> https://www.youtube.com/watch?v=OmOjE9XKjdI
2012-09-02T09:59:21 -!- dekar__ [~dekar@212.255.44.209] has joined ##stm32
2012-09-02T09:59:24 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-09-02T10:02:43 -!- dekar_ [~dekar@212.255.33.108] has quit [Ping timeout: 240 seconds]
2012-09-02T10:08:23 < dongs> hmm what the fuck. I can't get PA1 to work as GPIO
2012-09-02T10:10:07 < dongs> or no, PA0 is the fail one
2012-09-02T10:10:31 < dongs> i probably need to do something retarded wiht it
2012-09-02T10:12:55 < dongs> nope.
2012-09-02T10:18:26 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-02T10:34:59 < dongs> oops, was a shitty solder.
2012-09-02T10:35:10 < dongs> I forgot i used some reject chips for this proto
2012-09-02T10:35:16 < dongs> the tray was cracked and some legs were bent
2012-09-02T10:35:22 < dongs> i bent some back but i guess not enough to make contact.
2012-09-02T10:49:09 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Ping timeout: 252 seconds]
2012-09-02T11:04:11 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-02T11:11:49 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-02T11:13:27 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Disconnected by services]
2012-09-02T11:13:29 -!- soafee-chan is now known as spacekitteh
2012-09-02T11:19:54 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Remote host closed the connection]
2012-09-02T11:22:00 < dongs> hmm..
2012-09-02T12:06:26 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-02T12:14:01 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-02T12:14:01 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-02T12:14:01 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-02T12:14:04 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-02T12:25:51 < dongs> http://www.foxnews.com/us/2012/09/01/new-jersey-teen-sticks-head-out-party-bus-dies/
2012-09-02T12:25:54 < dongs> ahhaah
2012-09-02T12:46:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-02T13:12:01 < cjbaird> Ha, this iRiver I got can do the 78-108 'japanese' bandplan. I'll see if I can tune an FM bug that sits in the 'undetectible78-88MHz band..
2012-09-02T13:12:56 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-02T13:13:21 < dongs> arent all FM chips full band and the limit is jus software
2012-09-02T13:15:37 < cjbaird>  It would seem so. This is the first device I've had (outside of a scanner) that's made <88MHz available.
2012-09-02T13:17:12 < cjbaird> ex-gf Thea once made a kit FM bug that I bought her, and our sexually-active housemates were rather paranoid for a while. :3
2012-09-02T13:17:55 < dongs> what did you do, rebroadcast thehir porn over FM to ???
2012-09-02T13:18:41 < Vutral> eh
2012-09-02T13:18:44 < Vutral> fmchips ?
2012-09-02T13:19:12 < dongs> Vutral: since cjbaird is like 90, probably some trash from heathkit
2012-09-02T13:19:12 < Vutral> which fm chips
2012-09-02T13:19:14 < Vutral> better do PSK spread spectrum
2012-09-02T13:19:15 < Vutral> ^^
2012-09-02T13:19:29 < Vutral> ic
2012-09-02T13:19:39 < Vutral> heathkit
2012-09-02T13:19:39 < Vutral> i got a battery eleminator from them
2012-09-02T13:19:44 < cjbaird> The kit was just one of those beginner PCBs kind of deal. /I taught a girlfriend how to solder/ :D
2012-09-02T13:20:00 < Vutral> and i got some rayovac mercury switches
2012-09-02T13:20:01 < Vutral> lol
2012-09-02T13:20:03 < Vutral> from old fridges
2012-09-02T13:20:30 < Vutral> cjbaird: build your own radio no problem
2012-09-02T13:20:32 < Vutral> isnt difficult
2012-09-02T13:21:03 < cjbaird> She would later go on to do upgrades for the sister's laptop, and such.. (but a while later she turned into 'an idiot hippy' that thought Technology was Teh Evil.. :/)
2012-09-02T13:22:30 < Vutral> lololol
2012-09-02T13:22:32 < Vutral> yeah
2012-09-02T13:22:36 < Vutral> i am a cool hippy
2012-09-02T13:22:40 < Vutral> because i love technology
2012-09-02T13:22:53 < Vutral> but i guess you call these kind of hipsters goa people ^^
2012-09-02T13:23:12  * Vutral throws a fleetwood mac after cjbaird 
2012-09-02T13:25:05 < cjbaird> As opposed to a neophillic 'smart hippy', of course. :)
2012-09-02T13:25:52 < cjbaird> If you start to talk about Burning Man, you shall be k-lined.
2012-09-02T13:26:30 < dongs> isnt that some hippie shit
2012-09-02T13:27:50 < cjbaird> More 'punk', in the heavily-commercialized-and-hyped-to-all-fuck sense.
2012-09-02T13:28:34 < cjbaird> Steampunk faggots fit right in.
2012-09-02T13:32:47 < cjbaird> I once knew some alternative-energy-generation 'Hippies' from the Nimbin and Elands areas, and showed them a bit of how a DIY computer network was put together from the APANA work.
2012-09-02T13:35:42 < cjbaird> The Elands guys were off-grid practically, and used a 386 laptop (this was in 1994) to do dial-up UUCP when the computer's batteries were charged. :)
2012-09-02T13:42:43 -!- lyra2 [n@78-56-227-237.static.zebra.lt] has joined ##stm32
2012-09-02T13:45:04 < lyra2> can some one confirm that it is possible to set ADC Vref to 1V instead of 3.3V?
2012-09-02T13:56:14 < dongs> how?
2012-09-02T13:56:26 < dongs> maybe if you hav a package that has avref pin.
2012-09-02T14:00:04 < lyra2> i don't know, i am just asking, i know that pic & atmel mcu offers such an option
2012-09-02T14:01:08 < lyra2> they have different internal reference voltages you can choose: like 3.3V, 1V
2012-09-02T14:03:35 < dongs> The analog power supply is used as the reference voltage for conversion.
2012-09-02T14:04:32 < dongs> You can only select reference voltage (external) on packages > 100pin
2012-09-02T14:04:44 < dongs> this is not tarduino.
2012-09-02T14:04:57 < dongs> on < 100pin, analog vcc is taken as vref, with no other  choice.
2012-09-02T14:05:19 < dongs> considering STM32F1 ADC can run at something liek 7msps, I think its quite OK that you can't  choose a internal vref.
2012-09-02T14:09:32 -!- Laurenceb_ [~Laurence@host86-177-215-42.range86-177.btcentralplus.com] has joined ##stm32
2012-09-02T14:11:04 < lyra2> i will go with 100, 144 or even bigger packages because i need a lot of ADC channels
2012-09-02T14:11:40 < dongs> in that case it will work as expected.
2012-09-02T14:11:45 < dongs> have a read through http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/APPLICATION_NOTE/CD00211314.pdf
2012-09-02T14:12:20 < Laurenceb_> see youve been trolling raspberry pi dongs
2012-09-02T14:12:25 < dongs> oh?
2012-09-02T14:13:44 < Laurenceb_> http://www.raspberrypi.org/archives/1901
2012-09-02T14:14:59 < lyra2> just to clarify, by 7msps you mean that if there are 20 ADC channels, each channel will be read 7*10^6/20 = 350k times per second, right?
2012-09-02T14:15:25 < dongs> there are 2 or 3 ADCs. you can scan channels.
2012-09-02T14:15:30 < dongs> read the datasheet to see what conversion time is.
2012-09-02T14:15:53 < dongs> you cant read all 20 channels "at once"
2012-09-02T14:16:13 < dongs> if you want that, you probably want an external ADC that will return shit back by spi/parallel/whatever.
2012-09-02T14:17:55 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 268 seconds]
2012-09-02T14:19:40 < dongs> Laurenceb_: L O L
2012-09-02T14:19:54 < dongs> Steve: “Eben Upton refused to have deal with Jews”
2012-09-02T14:19:55 < dongs> I think I will add some and write a great coverage about the customer
2012-09-02T14:22:53 < Laurenceb_> dongs: your OSD code - on rcgroups youre saying it handles hsync delay in hardware
2012-09-02T14:22:55 < Laurenceb_> how?
2012-09-02T14:23:10 < dongs> Laurenceb_: masterslave shit on timer
2012-09-02T14:23:17 < dongs> autoreset + output compare
2012-09-02T14:23:21 < Laurenceb_> ok
2012-09-02T14:23:30 < Laurenceb_> but how do you run spi dma off that?
2012-09-02T14:23:42 < dongs> i just fire it up in cc handler
2012-09-02T14:23:51 < Laurenceb_> oh
2012-09-02T14:24:02 -!- lyra2 [n@78-56-227-237.static.zebra.lt] has quit [Quit: Leaving]
2012-09-02T14:24:03 < Laurenceb_> then its not hardware only :P
2012-09-02T14:24:19 < dongs> yeah but you cant do it fully in hardware this way.
2012-09-02T14:24:30 < dongs> i was just blogging about this earlier
2012-09-02T14:24:42 < dongs> apparently i have to reset memory address + cndtr on each execute
2012-09-02T14:24:59 < dongs> I was hoping I could save a couple instructions by siply restarting  dma and it would continue running off next memory adress
2012-09-02T14:25:02 < dongs> and only reset on vsync
2012-09-02T14:25:09 < Laurenceb_> oh
2012-09-02T14:25:14 < Laurenceb_> cant you clk gate it?
2012-09-02T14:25:24 < dongs> surely im not changing hardware for this shit especially since it works
2012-09-02T14:25:36 < Laurenceb_> clk gate the spi peripheral?
2012-09-02T14:25:41 < dongs> what do you mean?
2012-09-02T14:26:02 < Laurenceb_> to enable/disable the output
2012-09-02T14:26:14 < Laurenceb_> during blanking interval
2012-09-02T14:26:22 < dongs> um. and how do you suppose I would do that?
2012-09-02T14:26:28 < dongs> its running as master
2012-09-02T14:26:33 < Laurenceb_> using RCC
2012-09-02T14:26:43 < Laurenceb_> but yeah still needs software control
2012-09-02T14:27:03 < dongs> haha.. im pretty sure starting up SPI 400 times per screen is gonna be mroe annoying than just incrementing memory pointer + starting dma
2012-09-02T14:28:17 < dongs> i dont have any complaints about my current code
2012-09-02T14:28:33 < dongs> i was testing earlier today, I could do  telemetry + gps  + usb at same time and it wasnt glitchign the screen
2012-09-02T14:28:37 < dongs> after i fixed the priorities shit in nvic
2012-09-02T14:28:49 < Laurenceb_> yeah
2012-09-02T14:29:14 < dongs> but now i have something else i wanna try
2012-09-02T14:29:20 < dongs> im doing screen erase by DMA
2012-09-02T14:29:27 < dongs> because it was faster than memset(10kbytes)
2012-09-02T14:29:32 < dongs> but im gonna try doing it line by line
2012-09-02T14:29:37 < dongs> in teh hsync handler
2012-09-02T14:29:52 < dongs> this way when i'm done with screne update im ready to draw without delay
2012-09-02T14:30:50 -!- barthess [~barthess@178.154.11.222] has joined ##stm32
2012-09-02T14:33:32 < Laurenceb_> http://www.b3tards.com/u/67f98b0e3909eff8f955/catcopterstarrynight.jpg
2012-09-02T14:37:17 < Laurenceb_> with enough ram you could try double buffering
2012-09-02T14:47:00 < dongs> not on f1.
2012-09-02T14:47:12 < dongs> if im gonna drop rez down i'd rather use it for black+white pixel buffers
2012-09-02T14:47:15 < dongs> than doublebuffer
2012-09-02T14:47:27 < dongs> hardware is setup for it but i'd need to go to like...
2012-09-02T14:47:29 < dongs> 256x200 or so
2012-09-02T14:47:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-02T14:56:48 < jpa-> render on the fly?
2012-09-02T14:57:06 < jpa-> if the data you need is mostly text
2012-09-02T14:58:09 < dongs> nah, its graphical. there's no problem with time, i have plenty of that to render.
2012-09-02T14:58:43 < dongs> if all else fails, i can make a 'render list' kinda thing of the primitives/etc i want to do while spi is clocking stuff out (long time) and then quickly draw them during vsync.
2012-09-02T14:59:02 < jpa-> i mean don't store the full buffer, just render one scanline at a time
2012-09-02T14:59:05 < dongs> right
2012-09-02T14:59:08 < dongs> i know.
2012-09-02T14:59:11 < jpa-> ok
2012-09-02T14:59:21 < dongs> if it was text that would work fine.
2012-09-02T15:00:39 < jpa-> well for graphs also
2012-09-02T15:00:48 < dongs> i'm not that hardcore
2012-09-02T15:24:26 < Laurenceb_> id just grab an F4 and double buffer
2012-09-02T15:27:01 < Laurenceb_> or a larger F1 maybe
2012-09-02T15:27:41 < Laurenceb_> https://groups.google.com/forum/?fromgroups=#!topic/ukhas/CJitzddevWQ
2012-09-02T15:27:43 < Laurenceb_> lol drama
2012-09-02T15:28:33 < dongs> Laurenceb_: F1 is $2. F4 is $12
2012-09-02T15:29:20 < dongs> I am really very, very unhappy that a flight was sent apparently knowingly over London and what would have been the arrival flight path for Heathrow.
2012-09-02T15:29:23 < dongs> haha
2012-09-02T15:35:28 < dongs> also, i ahve no space for a bigger f1. http://i.imgur.com/xKWDU.jpg
2012-09-02T16:08:01 < dongs> http://i.imgur.com/MjnDM.png surprisingly, timing is quite accurate. im just using CoTickDelay() to schedule these uart writes
2012-09-02T17:11:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-02T17:23:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-02T17:24:01 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-02T17:26:11 < cjbaird> I'm waiting on a visit from the Beer Fairy...
2012-09-02T17:32:35 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-02T17:33:53 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-02T17:33:53 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-02T17:33:54 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-02T17:33:54 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-02T17:35:41 < dongs> sleeptime
2012-09-02T17:48:02 < Tzacks> better to say morning time :O
2012-09-02T17:49:12 < Laurenceb_> sexy tiem
2012-09-02T17:49:34 < Laurenceb_> or bubba tiem in the case of flyback
2012-09-02T17:53:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-02T18:00:41 < Laurenceb_> http://www.bbc.co.uk/news/technology-19441861
2012-09-02T18:04:49 -!- Laurenceb_ [~Laurence@host86-177-215-42.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-02T18:17:46 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-02T18:19:46 -!- Claude is now known as claude
2012-09-02T18:25:34 < Laurenceb_> http://pipe2.darklomax.org/uk/cloud6/DSC_6242.jpg
2012-09-02T18:27:24 < Tzacks> hey Laurenceb_ i envy you the "festival" in uk, you have there everything :)
2012-09-02T18:27:47 < Tzacks> there could be funny and interesting things to see
2012-09-02T18:29:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-02T19:02:20 -!- claude is now known as Claude
2012-09-02T19:07:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-02T19:12:44 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-02T19:13:00 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-02T19:13:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-02T19:33:28 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-02T19:33:43 < TitanMKD> hello
2012-09-02T19:34:02 < zyp> hi
2012-09-02T19:34:45 < emeb> lo
2012-09-02T19:49:39 <+izua> Z
2012-09-02T19:54:37 < emeb> X
2012-09-02T20:09:41 -!- barthess [~barthess@178.154.11.222] has quit [Quit: Leaving.]
2012-09-02T20:23:43 <+izua> U
2012-09-02T20:27:02 < emeb> w00t - home-made mini JTAG -> SWD cable works.
2012-09-02T20:27:08 < emeb> http://imagebin.org/226892
2012-09-02T20:28:32 < emeb> probably be a good idea to put some heatshrink on for strain relief though...
2012-09-02T20:29:47 < emeb> need to make some BMPB + GDB scripts for startup, flash programming, etc
2012-09-02T20:29:59 < emeb> s/BMPB/BMPM/
2012-09-02T20:39:16 < zyp> http://cgit.jvnv.net/suzumebachi/tree/.gdbinit
2012-09-02T20:40:00 < emeb> heh - hoping you'd see that. Was googling the logs for the last time you posted that link
2012-09-02T20:50:54 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-09-02T20:53:19 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-02T20:53:34 < emeb> hmm - works, but had to hit it a few times and restart gdb.
2012-09-02T20:57:42 < emeb> http://pastebin.com/0nwqDCXs
2012-09-02T21:03:57 < zyp> looks like a bad connection somewhere
2012-09-02T21:04:27 < emeb> saying that it smells like hardware?
2012-09-02T21:05:36 < zyp> yes
2012-09-02T21:06:10 < emeb> oh wait - does BMP use hardware reset?
2012-09-02T21:06:21 < emeb> Bet that's it.
2012-09-02T21:06:40 < emeb> THis is the old 4-pin style SWD w/o reset.
2012-09-02T21:07:26 < emeb> yep - pushing reset button on the target after the first error gets it going.
2012-09-02T21:07:48 < zyp> I've used BMP without hardware reset connected, IIRC it works fine
2012-09-02T21:12:24 < emeb> seems to be required in this instance.
2012-09-02T21:13:40 < emeb> Could be that DMA running thing - I can hear the I2S spewing mangled audio while the flash process tries to run.
2012-09-02T21:18:54 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-02T21:35:06 < zippe> emeb: check which reset BMP is using; it might be the core-only reset bit
2012-09-02T21:35:30 < zippe> it's illegal to set both the core and external reset bits at the same time...
2012-09-02T21:38:12 < emeb> zippe: thx - will check.
2012-09-02T21:38:42 < emeb> Doesn't look like hooking up the HW reset line has any effect.
2012-09-02T21:44:32 -!- barthess [~barthess@178.154.6.118] has joined ##stm32
2012-09-02T21:45:05 < zyp> zippe, it's using SYSRESETREQ
2012-09-02T21:47:50 < emeb> hmm - wonder if it's related to the firmware version?
2012-09-02T21:47:56 < emeb> (in the BMPM)
2012-09-02T21:48:19 < emeb> Mine's marked 018 - no idea how that correlates with the git repo though.
2012-09-02T21:48:26 < zippe> zyp: it's pretty terrible if that doesn't stop DMA
2012-09-02T21:49:24 < zyp> emeb, were you the one with previous dma running bugs?
2012-09-02T21:49:52 < zyp> flyback, yes it does, that's what FSMC is for
2012-09-02T21:49:54 < emeb> zyp: yeah - first found that in the stlink
2012-09-02T21:50:09 < zyp> emeb, which chip are you using?
2012-09-02T21:50:14 < emeb> F405
2012-09-02T21:50:41 < emeb> flyback: why not read mods out of SD?
2012-09-02T21:50:52 < zippe> zyp: it probably needs to hit the RCC resets as well
2012-09-02T21:51:39 < emeb> zippe: and SYSRESETREQ doesn't hit the RCC?
2012-09-02T21:52:40 < zyp> SYSRESETREQ should do that.
2012-09-02T22:00:08 < zippe> zyp: based on the evidence above, no.
2012-09-02T22:07:09 < Laurenceb_> http://cd.textfiles.com/group42/ANARCHY/COOKBOOK/DISKBOMB.HTM
2012-09-02T22:07:12 < Laurenceb_> hehe genius
2012-09-02T22:07:33 < Laurenceb_> i should have tried that at school
2012-09-02T22:07:49 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-02T22:07:51 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-02T22:08:57 < BrainDamage> Laurenceb_ you should be able to achieve a similar result using acetone peroxide on a dvd drive
2012-09-02T22:09:15 < Laurenceb_> awesome
2012-09-02T22:09:22 < Laurenceb_> just need a victim :P
2012-09-02T22:10:48 <+Steffanx> I guess i'm at the wrong place at the wrong time
2012-09-02T22:18:21 < zippe> Steffanx: to get yourself fucked up by a bunch of stupid recipes that don't work?
2012-09-02T22:18:31 -!- barthess [~barthess@178.154.6.118] has quit [Quit: Leaving.]
2012-09-02T22:18:35 < zippe> flyback: are you sure about that?
2012-09-02T22:18:48 < zippe> I was looking for a tracker for F4 a little while back
2012-09-02T22:19:06 < Laurenceb_> zippe: lol
2012-09-02T22:19:20 < zippe> Got a suggestion from someone here actually, jpa- maybe
2012-09-02T22:19:24 < zippe> Let me see if I can find it again
2012-09-02T22:19:57 < BrainDamage> http://www.ebay.com/itm/100W-200W-300W-400W-808nm-QCW-High-Power-Laser-Diode-Bar-Array-DPSS-/150882951728?pt=LH_DefaultDomain_0&var=&hash=item2321532230#ht_3969wt_1377
2012-09-02T22:20:36 < zippe> http://www.serveurperso.com/temp/sds3m32.zip
2012-09-02T22:21:03 < zippe> You may, as you noted, have issues with sample memory
2012-09-02T22:21:56  * Laurenceb_ is trying to devise a way to track his balloon launched rocket
2012-09-02T22:22:04 < Laurenceb_> was thinking a time delay smoke bomb
2012-09-02T22:22:14 < zippe> There's a mod player in the same directory
2012-09-02T22:22:16 < Laurenceb_> and launch just after sunset
2012-09-02T22:24:59 -!- mrcan_ is now known as mrcan
2012-09-02T22:25:44 < Laurenceb_> BrainDamage: I've only got 1 gram mass budget for the "tracker" on the rocket
2012-09-02T22:26:16 < zippe> Laurenceb: how good does your tracking have to be?
2012-09-02T22:26:18 < Laurenceb_> and it needs a time delay of ~100 seconds to give a puff of smoke at apogee
2012-09-02T22:26:26 < Laurenceb_> +-10Km is fine
2012-09-02T22:26:34 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-02T22:26:38 < zippe> and what's your budget?  Radar tracking isn't a big deal if you can afford $100k, for example
2012-09-02T22:26:50 < Laurenceb_> the rocket is ~8cm long
2012-09-02T22:27:00 < Laurenceb_> ~impossible to pick up
2012-09-02T22:27:02 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds]
2012-09-02T22:27:15 < zippe> We used to track the ionisation trails left by micrometeorites (< grain of sand) at 70km
2012-09-02T22:27:26 < zippe> System was (in '98) about AUD$70k
2012-09-02T22:27:31 < Laurenceb_> nice
2012-09-02T22:27:47 < zippe> http://www.gsoft.com.au
2012-09-02T22:27:52 < zippe> They're still in business
2012-09-02T22:27:56 < Laurenceb_> hmm in my simulations to the rocket gets to ~80 to 90km
2012-09-02T22:28:28 < Laurenceb_> - little <100gram thing launched from 40km latex balloon
2012-09-02T22:28:44 < Laurenceb_> but i doubt it would leave a plasma trail
2012-09-02T22:28:58 < Laurenceb_> going at about 500m/s
2012-09-02T22:29:17 < zippe> So, much easier to track
2012-09-02T22:29:33 < zippe> Most of the gsoft work was watching air with radar
2012-09-02T22:29:40 < zippe> meteorites were almost too easy
2012-09-02T22:29:50 < zippe> rain, even more so
2012-09-02T22:30:31 < zippe> Like I said, all depends on $
2012-09-02T22:31:00 < Laurenceb_> i guess an 868mhz tracker would fit in the front and weigh ~1gram
2012-09-02T22:31:08 < Laurenceb_> with ceramic antenni
2012-09-02T22:31:48 < Laurenceb_> but that makes it more work
2012-09-02T22:32:54 < BrainDamage> phased array receiver?
2012-09-02T22:33:06 < Laurenceb_> nah
2012-09-02T22:33:20 < Laurenceb_> you can in fact do it with just a single ground station
2012-09-02T22:33:31 < Laurenceb_> if you can bounce packets off it and get doppler
2012-09-02T22:33:46 < Laurenceb_> the silabs transceivers support that
2012-09-02T22:34:06 < Laurenceb_> then fit doppler vrs time to a model
2012-09-02T22:34:13 < BrainDamage> wouldn't doppler shift be almost undetectable by oscillator drift?
2012-09-02T22:34:29 < BrainDamage> you'll get ridicolous temp diffences during the flight
2012-09-02T22:34:32 < Laurenceb_> if you bounce packets off it cancles out
2012-09-02T22:35:41 < Laurenceb_> silabs transceivers have AFC built into the modem and a delta F correction register
2012-09-02T22:35:59 < zyp> the rocket tracking stuff at the norwegian rocket range use doppler shift to track distance
2012-09-02T22:36:04 < Laurenceb_> you upload a packet then the AFC runs and chucks a packet back down the the current AFC setting
2012-09-02T22:36:46 < Laurenceb_> i have it working well in a matlab sim with realistic noise, update rates and onboard oscillator drift
2012-09-02T22:37:17 < zyp> the sender on the rocket have a good TCXO, and a frequency synth at the receiver station is tuned to exactly match it before launch
2012-09-02T22:38:13 < Laurenceb_> oh interesting
2012-09-02T22:38:19 < Laurenceb_> so not even transceiver
2012-09-02T22:38:33 < zyp> no, it's downlink only
2012-09-02T22:38:47 < Laurenceb_> well its doable then
2012-09-02T22:38:58 < Laurenceb_> but i like the idea of smoke cloud + 2 cameras on ground
2012-09-02T22:39:18 < Laurenceb_> its "visual proof" it works rather than something chucked out of a matlab script
2012-09-02T22:39:27 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-02T22:40:01 < zyp> smoke cloud where? at what distance are you going to observe it?
2012-09-02T22:45:14 < Laurenceb_> zyp: ~100km range
2012-09-02T22:45:37 < Laurenceb_> using a regular slr with ~10 second exposure it should show up fine
2012-09-02T22:45:46 < Laurenceb_> the background starfeild gives alignment
2012-09-02T22:46:06 < zippe> zyp: which range, of curiosity?
2012-09-02T22:46:07 < Laurenceb_> rocket is launched into the sunlight
2012-09-02T22:46:19 < zippe> zyp: I was at ESRANGE in '97
2012-09-02T22:46:47 < zippe> zyp: went briefly over to Andenes...
2012-09-02T22:47:13 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-02T22:47:45 < Laurenceb_> but a 100s delay igniter in <1gram mass seems hard
2012-09-02T22:48:54 < Laurenceb_> wheras oddly a full 868mhz transmitter isnt so hard.. crazy
2012-09-02T22:51:10 < BrainDamage> does the igniter have to incude the wire?
2012-09-02T22:51:19 < BrainDamage> I mean the weight
2012-09-02T22:51:23 < Laurenceb_> well...
2012-09-02T22:51:38 < Laurenceb_> in my sim code i allowed 1 gram of "payload"
2012-09-02T22:51:45 < BrainDamage> ok, so total
2012-09-02T22:51:56 < Laurenceb_> the larger the smoke generator the easier it is to track
2012-09-02T22:52:21 < Laurenceb_> so really the entire 100s delay has to be less than 500milligrams or so
2012-09-02T22:52:52 < BrainDamage> I think the main issue going electronic would be the battery
2012-09-02T22:53:08 < Laurenceb_> theres 200mg lipo cells
2012-09-02T22:53:15 < Laurenceb_> thats not an issue
2012-09-02T22:53:21 < Laurenceb_> or hearing aid batteries
2012-09-02T22:53:46 < Laurenceb_> tx only needs about 60ma
2012-09-02T22:53:48 < BrainDamage> then, tiny smd logic level mos, and some 8 pin smd uC
2012-09-02T22:54:15 < Laurenceb_> oh you mean for a pyro?
2012-09-02T22:54:22 < BrainDamage> yup
2012-09-02T22:54:27 < Laurenceb_> yeah electronic pyro has the issue of needing large current
2012-09-02T22:54:41 < Laurenceb_> so a 200mg lipo that would run at transceiver wont work
2012-09-02T22:54:56 < Laurenceb_> something like an si4432 running on 868mhz only uses 60ma
2012-09-02T22:55:12 < Laurenceb_> but pyros seem to need ~1A or so
2012-09-02T22:56:04 < BrainDamage> you could probably drop the current requirement by using a very heat sensitive inition material
2012-09-02T22:56:14 < BrainDamage> but you risk accidental triggers then
2012-09-02T22:56:33 < Laurenceb_> like from matches, and an 01005 smd resistor
2012-09-02T22:59:33 < Laurenceb_> i guess both approaches are similar in complexity
2012-09-02T22:59:58 < Laurenceb_> maybe i should try and get a spun launcher working first :P
2012-09-02T23:02:44 < BrainDamage> btw, your approach seems rather similar to those noobs @ wikisat
2012-09-02T23:02:55 < Laurenceb_> hehe
2012-09-02T23:03:05 < Laurenceb_> thats "unfortunate"
2012-09-02T23:03:08 < Laurenceb_> http://www.powerstream.com/p/GMB300910.pdf
2012-09-02T23:03:21 < Laurenceb_> thats 330mg when "trimmed down"
2012-09-02T23:04:02 < zyp> zippe, that's the one :)
2012-09-02T23:04:10 < zyp> I grew up right next to it
2012-09-02T23:04:17 < Laurenceb_> i was considering a pyro delay fuse, but aiui it needs to have a minimum diameter or it will "burn out"
2012-09-02T23:04:26 < Laurenceb_> zyp: thats _quite_ far north
2012-09-02T23:04:44 < Laurenceb_> so a 100 second delay gets very heavy
2012-09-02T23:05:14 < zyp> zippe, I mean Andenes, in case that was unclear
2012-09-02T23:05:44 < BrainDamage> perhaps something chemical
2012-09-02T23:06:06 < BrainDamage> like glycerine and potassium permangante self ignite
2012-09-02T23:06:15 < BrainDamage> permanganate*
2012-09-02T23:06:41 < BrainDamage> provided you have your parameters set, like granularity, mixing conditions, temp, etc
2012-09-02T23:06:47 < BrainDamage> you should get a reliable delay
2012-09-02T23:06:55 < Laurenceb_> yeah, and you then need to hit apogee with it
2012-09-02T23:06:59 < Laurenceb_> probably too hard
2012-09-02T23:07:18 < Laurenceb_> - smoke cloud at apogee
2012-09-02T23:10:05 < Laurenceb_> hmf radio tracker comes out looking more sane
2012-09-02T23:10:10 < Laurenceb_> but boring
2012-09-02T23:10:39 < Laurenceb_> BrainDamage: but wikisat plan to get to orbit, i was thinking 100km with "hobby" rockets
2012-09-02T23:12:13 < Laurenceb_> also lol @ wikisat attempts to build coke can rockets
2012-09-02T23:12:20 < Laurenceb_> each one fails epically
2012-09-02T23:14:09 < BrainDamage> mmm, I think sputnik was ~200km
2012-09-02T23:14:28 < Laurenceb_> erm... heigh is hardly important
2012-09-02T23:14:36 < Laurenceb_> you need 7km/s
2012-09-02T23:14:38 < BrainDamage> I know, you need to get it to speed
2012-09-02T23:15:03 < BrainDamage> and height is important, you don't want the atmosphere braking it :p
2012-09-02T23:16:12 < Laurenceb_> yes - 100km is "official boundary of space"
2012-09-02T23:16:48 < Laurenceb_> so i was thinking that reaching it with a "hobby" Estes black powder rocket would be a fun project
2012-09-02T23:20:39 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-02T23:26:54 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-02T23:30:14 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-02T23:30:15 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-02T23:30:15 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-02T23:30:18 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-02T23:53:56 -!- Claude is now known as claude
--- Day changed Mon Sep 03 2012
2012-09-03T00:01:36 < zippe> zyp: I was happy it was summer
2012-09-03T00:01:58 < Laurenceb_> bet its not nice in winter
2012-09-03T00:02:04 < Laurenceb_> its bad enough in the UK
2012-09-03T00:02:34 < Laurenceb_> -23C in my garden last year
2012-09-03T00:04:07 < zyp> Andenes never gets that cold, the gulf stream pass right next to it
2012-09-03T00:05:05 < Laurenceb_> north or arctic circle to
2012-09-03T00:05:08 < Laurenceb_> *of
2012-09-03T00:05:30 < zippe> I swam in the ocean there
2012-09-03T00:05:35 < zippe> it was *not* warm
2012-09-03T00:05:42 < zyp> :)
2012-09-03T00:08:09 < zyp> regarding the temperature, even tough it doesn't get very cold, it's usually always very windy, causing a lot of wind chill
2012-09-03T00:10:02 < zyp> I underestimated that when I went climbing the mountain on last new years eve :p
2012-09-03T00:13:55 < zippe> Heh, the stories about that mountain were amusing
2012-09-03T00:14:08 < zippe> We were told not to take photographs of it, or the MPs would come take us away
2012-09-03T00:14:30 < zyp> huh?
2012-09-03T00:14:48 < zippe> http://www.panoramio.com/photo_explorer#view=photo&position=310&with_photo_id=2074986&order=date_desc&user=112615
2012-09-03T00:14:58 < zyp> I know the military have some no-photo zones, but not on the mountains
2012-09-03T00:15:18 < zippe> This pic, the bay center-left is where we were; told not to take photos of the hillside behind
2012-09-03T00:15:44 < zippe> (I think you can see the LIDAR on the hilltop far left…)
2012-09-03T00:16:07 < zyp> yeah, ALOMAR is visible there
2012-09-03T00:16:42 < Laurenceb_> i didnt realise it was military
2012-09-03T00:16:57 < zyp> what is?
2012-09-03T00:17:10 < zyp> the rocket range is not military
2012-09-03T00:17:21 < zippe> There is/was a NATO base there, that's all
2012-09-03T00:17:27 < zyp> yes
2012-09-03T00:17:43 < zippe> big airfield, restaurant menus in 4 languages, etc.
2012-09-03T00:18:14 < zyp> http://bin.jvnv.net/f/yv1Ps.JPG <- here's the pic I went up to the mountain to snap
2012-09-03T00:18:14 < zippe> Boss and his girlfriend had a big argument about whether eating whale was OK or not
2012-09-03T00:18:17 < zippe> 8)
2012-09-03T00:18:23 < zyp> it's the opposite angle to the one you linked
2012-09-03T00:18:43 < zippe> Shot around noon or so, by the looks of it
2012-09-03T00:18:45 < zippe> 8)
2012-09-03T00:18:52 < zyp> :p
2012-09-03T00:21:34 < zyp> except that new years fireworks is launched at midnight :p
2012-09-03T00:23:27 < BrainDamage> so while everyone was down to party you were on the hills taking pics? :p
2012-09-03T00:24:21 < zyp> something like that
2012-09-03T00:24:38 < BrainDamage> also, bah, I hoped you were closer to lofoten, I wanted to see some pics of the maelstrom
2012-09-03T00:24:59 < zippe> New years? I'm surprised no trolls
2012-09-03T00:25:33 < zyp> most of my old friends from Andenes did like me and moved to other places, so I weren't aware of any parties I would go to anyway
2012-09-03T00:27:21 -!- zippe1 [~Adium@173.11.99.161] has joined ##stm32
2012-09-03T00:27:21 -!- zippe [~Adium@173.11.99.161] has quit [Read error: Connection reset by peer]
2012-09-03T00:50:50 < emeb> Starting to get the feeling that I'm beating my head against the wall.
2012-09-03T00:51:06 <+Steffanx> Means it's time for some sleep
2012-09-03T00:51:24 < emeb> that'd be bad - it's only 14:51 here.
2012-09-03T00:56:23 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-03T00:57:26 < zippe1> *sigh*  And this is why you don't throw the packaging away before you've tested it, folks
2012-09-03T00:58:44 < Laurenceb_> emeb: whats the problem?
2012-09-03T01:02:04 < emeb> Laurenceb_: Getting frustrated w/ the Linux-based devtools
2012-09-03T01:03:00 < emeb> all sorts of problems flashing my code via stlink & bmpm under Linux. When I try with the ST flash utility under Win7 it works fine...
2012-09-03T01:08:12 < Laurenceb_> hmm
2012-09-03T01:08:21 < Laurenceb_> what lines do you have connected?
2012-09-03T01:10:55 < zippe1> And how are you flashing?
2012-09-03T01:11:41 < Laurenceb_> i had issues with my F4 boards
2012-09-03T01:11:48 < Laurenceb_> until i disconnected reset
2012-09-03T01:13:33 < zippe1> I usually flash with the gdb 'load' command, if that helps any
2012-09-03T01:40:03 < emeb> I've got vref, swck, swdio, gnd hooked up
2012-09-03T01:40:58 < emeb> using the gdb load command with bmpm, using stlink-flash command with stlink.
2012-09-03T01:41:14 < emeb> (but I've also used gdb load w/ similar results)
2012-09-03T01:41:21 < zippe1> What results are those?
2012-09-03T01:41:23 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-03T01:41:27 -!- zippe1 is now known as zippe
2012-09-03T01:41:55 < emeb> zippe: depending on what code is running on the target at the moment I start flashing, I get different results.
2012-09-03T01:42:22 < emeb> but the general theme is that I need to flash multiple times, hit the reset button on the target, sometimes power cycle.
2012-09-03T01:42:55 < zippe> If the target is in a weird state when you stop it with gdb, you can certainly have flashing issues.
2012-09-03T01:43:17 < emeb> Yep.
2012-09-03T01:44:01 < emeb> Trouble is "weird state" seems to imply "anything that isn't a completely erased & stopped target"
2012-09-03T01:44:06 < zippe> caveat; usually I am only ever flashing a bootloader, however I usually use a breakpoint on the first instruction, then 'run' to reset the target
2012-09-03T01:44:23 < zippe> Well, you absolutely cannot expect to flash on a target that is generating interrupts of any sort
2012-09-03T01:44:47 < zippe> Or that has any DMA activity in progress
2012-09-03T01:44:54 < emeb> Yep - seen that.
2012-09-03T01:45:17 < zippe> Lockup mode also causes some major grief
2012-09-03T01:45:32 < emeb> So, it seems that the FOSS tools assume you're only ever going to flash a part that's fresh from the factory.
2012-09-03T01:45:36 < zippe> No
2012-09-03T01:45:43 < zippe> The FOSS tools assume that you know what you're doing
2012-09-03T01:45:55 < emeb> Thanks zippe - that's really helpful.
2012-09-03T01:46:04 < zippe> And it's fair to say that there are plenty of corner cases they don't cover
2012-09-03T01:46:09 < zippe> *shrug*  It's the truth
2012-09-03T01:46:10 < emeb> True.
2012-09-03T01:46:22 < zippe> I've tried to explain what I do that works for me.
2012-09-03T01:46:34 < emeb> Yeah - that's a nice data point.
2012-09-03T01:46:36 < zippe> And that in general, you're better off using a bootloader when you can.
2012-09-03T01:46:36 < zyp> I expect my tools to stop everything when flashing, and from my experience they usually do
2012-09-03T01:46:49 < emeb> zyp: that's what I would hope too.
2012-09-03T01:47:32 < emeb> but it seems that neither of the FOSS tools I've tried today *really* get the device into a known state before trying to flash.
2012-09-03T01:48:38 < emeb> In fact several times I've seen both bmpm and stlink complain they couldn't find the device at all. But plugging into Win7/ST-LINK found it and was able to erase.
2012-09-03T01:49:08 < emeb> Ah well - just venting here. I appreciate all the suggestions.
2012-09-03T01:49:11 < zyp> I'm observing that when I flash with the BMP, the USB gets reset when I flash
2012-09-03T01:49:20 < zippe> emeb: yup; that's fundamentally the difference between something hacked together by a bunch of well-meaning folks, and something assembled by the vendor
2012-09-03T01:49:29 < zyp> so it does reset _some_ peripherals
2012-09-03T01:49:37 < emeb> zippe: :) Yeah - I understand that.
2012-09-03T01:49:39 < zippe> Sadly, the vendor doesn't build tools that work for me, so I'm SoL
2012-09-03T01:49:40 < zyp> so I don't see why it wouldn't reset _all_ peripherals
2012-09-03T01:49:50 < emeb> Be glad that it works sometimes/at all
2012-09-03T01:50:07 < zippe> zyp: If I had to guess, it's because the ARM systemreset signal doesn't go everywhere it should
2012-09-03T01:50:16 < zyp> yeah, that's what I also suspect
2012-09-03T01:50:24 < zippe> zyp: I'd ask ST, but I don't currently have a line to the right team there
2012-09-03T01:50:28 < zyp> I checked the errata, but I couldn't find anything
2012-09-03T01:50:41 < zippe> They don't usually document debug errata
2012-09-03T01:51:10 < emeb> would be amusing to see a USB dump of what the vendor flash tool on Win does...
2012-09-03T01:51:24 < zippe> emeb: it could be arranged; I have a beagle here
2012-09-03T01:51:29 < zippe> Just not the time
2012-09-03T01:51:33 < emeb> probably goes through and mashes all the interal regs.
2012-09-03T01:51:43 < zyp> so do I, just not a usable windows computer
2012-09-03T01:51:46 < emeb> zippe: I could see where that would be a time consuming project.
2012-09-03T01:51:53 < zyp> but doesn't the texane repo contain some traces?
2012-09-03T01:52:11 < zippe> Some, but does it contain a trace that includes the probe recovering a device in lockup mode?
2012-09-03T01:52:28 < zippe> Their coverage is probably all "happy" workloads because those are the usual
2012-09-03T01:52:42 < zyp> emeb, anyway, are you sure you are not seeing different problems?
2012-09-03T01:53:11 < emeb> zyp: reasonably so - the problems remain consistent with different tools
2012-09-03T01:53:31 < zyp> the ones you pasted earlier tonight looked very similar to what I've had on a board with some bad soldering
2012-09-03T01:53:52 < zyp> or bad cable
2012-09-03T01:54:04 < emeb> zyp: Interesting.
2012-09-03T01:54:25 < zyp> that's why I suggested that you might have a bad connection
2012-09-03T01:54:45 < emeb> Well, I suppose it's worth checking again.
2012-09-03T01:54:50 < zyp> of course, that does not explain why the windows tools work
2012-09-03T01:54:59 < emeb> Yeah - there's that.
2012-09-03T01:55:03 < zyp> but you are not using the same cable for that, right?
2012-09-03T01:55:31 < dongs> lots of chat is ee
2012-09-03T01:55:35 < emeb> I use a Discovery F0 for both Win ST-LINK and Linux stlink
2012-09-03T01:55:49 < emeb> with the same USB cable and the same SWD cable.
2012-09-03T01:55:59 < zyp> yes, but forget about linux stlink
2012-09-03T01:56:15 < emeb> dongs: your favorite subject - debugging FOSS tools. :P
2012-09-03T01:57:12 < emeb> zyp: the bmpm only works w/ Linux and only with the little JTAG/SWD cable I made.
2012-09-03T01:57:23 < emeb> So there's only one data point there...
2012-09-03T01:57:25 < zippe> Oh, yay.  Don't initialise the P term with the D term parameter.  That's bad.
2012-09-03T01:57:52 < zippe> emeb: I use the BMP almost exclusively for the F4 work.
2012-09-03T01:58:16 < zyp> emeb, so what I'm suggesting is that the cable you're using with the BMP is bad, and that's unrelated to texane/stlink being shitty
2012-09-03T01:58:47 < emeb> zippe: cool - I was hoping that the BMPM would eliminate the flashing issues I've seen. No luck w/ that yet.
2012-09-03T01:59:54 < emeb> zyp: I suppose it's possible the home-made cable is bad. Worth ohming it out I guess.
2012-09-03T02:00:13 < zyp> emeb, I would like to verify your problems by reproducing them here, do you have an executable that provokes the problem that I can run on some of my boards?
2012-09-03T02:00:30 < Laurenceb_> try kill first?
2012-09-03T02:00:34 < Laurenceb_> before load
2012-09-03T02:00:50 < zyp> i.e. not being dependant on any external stuff being connected
2012-09-03T02:01:04 < emeb> zyp: I'll bundle up an elf for you to try.
2012-09-03T02:01:24 < zyp> oh, and it's late, so I'm not going to do that now
2012-09-03T02:01:26 < zippe> emeb: I'd be happy to see if I can repro here as well if I have something it will run on
2012-09-03T02:01:48 < zippe> F4 and VL discovery boards both available, f.e.
2012-09-03T02:04:38 < emeb> try this: http://ebrombaugh.studionebula.com/synth/stm32f4_codec/emeb_stm32f405_flash_problem.zip
2012-09-03T02:04:52 < emeb> should unzip to a main.elf that you can try flashing.
2012-09-03T02:06:34 < dongs> ive had zero problems with $20 jlink/ulink clones.
2012-09-03T02:06:39 < dongs> i need to try thenew jlinkob stuff.
2012-09-03T02:06:42 < dongs> thats only $15
2012-09-03T02:06:50 < emeb> That's targeted at my custom board. No idea if that's viable on an F4 Discovery. Would be an interesting experiment though
2012-09-03T02:07:09 < emeb> dongs: but you're using a full commercial license of Keil right?
2012-09-03T02:07:12 < dongs> emeb: http://cgi.ebay.fr/180923403461
2012-09-03T02:07:21 < dongs> jliknk shit works wiht freetard stuff too
2012-09-03T02:07:31 < emeb> openocd?
2012-09-03T02:07:31 < dongs> i think openocd suports it
2012-09-03T02:07:33 < dongs> yes
2012-09-03T02:07:57 < emeb> I've had good luck with openocd on other mfgs ARM parts (Atmel / NXP)
2012-09-03T02:08:09 < dongs> so its justfailing with stm?
2012-09-03T02:08:36 < dongs> i remember  spending weeks trying to get shit properly working wiht SAM7
2012-09-03T02:08:45 < dongs> i was using one of those overpriced olimex ftdi>shit breakouts
2012-09-03T02:08:56 < emeb> heh - that's what I've got.
2012-09-03T02:09:01 < emeb> works fine w/ SAM7
2012-09-03T02:09:16 < emeb> the current problem is with STM32 + SWD (not JTAG)
2012-09-03T02:09:47 < emeb> Works fine with simple / small apps on the low-end processors.
2012-09-03T02:10:04 < emeb> but on the STM32F405 + complex DMA it hangs constantly.
2012-09-03T02:11:31 < emeb> (during flashing - once code is loaded things are fine)
2012-09-03T02:11:32 < dongs> dunno, seems even zippe doesnt trust freetard shit with "Stepping"
2012-09-03T02:11:47 < zyp> emeb, no problems here
2012-09-03T02:11:51 < dongs> yet i was stepping through functions/interrupts/etc all fine
2012-09-03T02:11:57 < emeb> zyp: on your 405 board?
2012-09-03T02:12:03 < zyp> http://paste.jvnv.net/view/zeKff
2012-09-03T02:12:15 < zyp> waveshare F407 board
2012-09-03T02:12:26 < dongs> are  you using SWD on that though
2012-09-03T02:12:30 < dongs> waveshare is setup for jtag.
2012-09-03T02:12:39 < zyp> swd is a subset of jtag pins
2012-09-03T02:12:43 < zyp> and I connected in swd mode
2012-09-03T02:12:48 < emeb> so you can use either.
2012-09-03T02:12:49 < dongs> right, ok.
2012-09-03T02:13:21 < dongs> the download speed seems to be as terrible as usual.
2012-09-03T02:13:42 < emeb> zyp: big thanks for testing that - I'll get to work looking at potential HW issues...
2012-09-03T02:13:56 < zyp> dongs, nah, it's just the file that is small
2012-09-03T02:14:07 < dongs> 40k?
2012-09-03T02:14:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-03T02:15:10 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-03T02:15:57 < zyp> dongs, that's pretty small considering the flash is 1M, no?
2012-09-03T02:20:35 < dongs> shrugs
2012-09-03T02:21:31 < zyp> emeb, though, are you sure your code is starting DMA transfers when it's run on a board without the other stuff on your board hooked up?
2012-09-03T02:21:42 < zyp> I see it's always breaking in some i2c code
2012-09-03T02:22:29 < dongs> that was emeb;s coide?
2012-09-03T02:23:23 < zyp> yes
2012-09-03T02:23:55 < zyp> hmm, my own code is far from 40k, when you mention it :p
2012-09-03T02:24:37 < emeb> Ah - the I2C code is in the initializer for the audio codec - it's probably hanging there, never getting to the DMA setup. :P
2012-09-03T02:25:05 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-03T02:25:09 < emeb> (because you don't have pullups on the I2C bus that I'm using)
2012-09-03T02:25:09 < zyp> doesn't make for a very good test then :p
2012-09-03T02:25:15 < emeb> nope.
2012-09-03T02:25:32 < emeb> I could comment that section out easily and resend..
2012-09-03T02:25:35 < zyp> sure
2012-09-03T02:27:24 < dongs> emeb, yo uarent afraid zyp 's gonna reverse engineer your proprietary dsp algorithms?
2012-09-03T02:27:29 < emeb> bah
2012-09-03T02:27:36 < emeb> nothing cool in there.
2012-09-03T02:27:51 < emeb> OK - new zipfile up in the same place.
2012-09-03T02:28:06 < dongs> also, if you steal shitty i2c code from Laurenceb, it properly handles i2c errors too
2012-09-03T02:28:21 < dongs> or from zyp for that matter
2012-09-03T02:28:46 < dongs> though zyps code is about as readable as arm asm.
2012-09-03T02:29:16 < zyp> still no problems
2012-09-03T02:29:37 < emeb> cool
2012-09-03T02:29:40 < zyp> dongs, are you claiming my code is less readable than Laurenceb's?
2012-09-03T02:30:14 < dongs> in terms of er... you know magical numbers.
2012-09-03T02:31:29 < zyp> my i2c code is probably the most commented piece I have in my lib, all the numbers are commented
2012-09-03T02:31:46 < dongs> since when?
2012-09-03T02:31:57 < dongs> did you add comments since last time i looked at it? :p
2012-09-03T02:32:12 < zyp> dunno, I haven't touched it for ages
2012-09-03T02:32:18 < emeb> reflowed all the STM pins - still behaves the same for me.
2012-09-03T02:32:22 < zyp> http://cgit.jvnv.net/laks/tree/i2c/i2c.cpp <- just looked at it, it looks like this now
2012-09-03T02:32:54 < dongs> still magic shit in TRISE/CR2/etc.
2012-09-03T02:32:59 < zyp> hmm, only the ISR though :p
2012-09-03T02:33:06 < dongs> but at least looks like you wrote it out at least
2012-09-03T02:33:18 < zippe> Heh
2012-09-03T02:33:30 < zippe> I like that "handle_error" is not implemented
2012-09-03T02:33:39 < TheSeven> is it just me or is this i2c core the most complicated bullshit that was ever designed?
2012-09-03T02:33:43 < zippe> I finally did the GPIO banging crap … has resolved most of my issues with the ms5611
2012-09-03T02:33:55 < zyp> TheSeven, it's just you
2012-09-03T02:33:58 < zippe> TheSeven: Not by a long shot.  Biggest issue is just that I2C stinks
2012-09-03T02:34:07 < zippe> I have yet to see an I2C controller that doesn't suck
2012-09-03T02:34:13 < dongs> i have no problem with  5611 and hardware i2c.
2012-09-03T02:34:31 < zippe> dongs: It seems to be possible to get it into a state where it won't talk to you
2012-09-03T02:34:32 < zyp> the i2c core is piece of cake compared to the usb cores in F4
2012-09-03T02:34:38 < TheSeven> the fact that this one tries to be aware of slave addressing (in master mode) seems to make it unnecessarily complicated
2012-09-03T02:34:44 < dongs> zippe, do tell. have yet to see it.
2012-09-03T02:35:03 < dongs> zippe: you are clocking out scl on i2c reset to unstuck any peripherals that might think they're being addressed right
2012-09-03T02:35:09 < zippe> dongs: for recovery, yes
2012-09-03T02:35:15 < zippe> long scl train + start + stop
2012-09-03T02:35:19 < dongs> yea.
2012-09-03T02:35:21 < TheSeven> zyp: tbh said USB core seems much more straight forward to me than this I2C messie
2012-09-03T02:35:34 < zippe> 5611 seems to need that + a good breather before it will talk again
2012-09-03T02:35:51 < zippe> With that said, I am seeing runt clocks coming out of the F4 I2C as well
2012-09-03T02:36:02 < zippe> or whole extra bytes beyond those the software asks it to transfer
2012-09-03T02:36:02 < dongs> well, do you have a specific example of getting 5611 stuck? I reused my current i2c code on my flyingdong and have yet to see any problems wiht 5611, running under debugger/sitting on my desk overnight/etc
2012-09-03T02:36:09 < zyp> TheSeven, have you used both?
2012-09-03T02:36:18 < zippe> so it's possible we're doing something wrong that pisses the 5611 off.
2012-09-03T02:36:26 < TheSeven> I've used the USB core, and currently trying to understand how the I2C one works
2012-09-03T02:36:29 < zippe> e.g. stopping halfway through a byte, etc.
2012-09-03T02:36:52 < TheSeven> and the manual of the I2C core is really weird
2012-09-03T02:36:56 < zippe> I have a bunch of beagle traces, but I need to correlate them with software … it's "on the list"
2012-09-03T02:36:59 < TheSeven> you kinda have to reverse engineer it
2012-09-03T02:37:34 < dongs> ) -> decltype(pack(Configuration_desc(), r...)) {
2012-09-03T02:37:39 < dongs> what the fuck is this? c++11 or someshit?
2012-09-03T02:37:42 < zyp> yes
2012-09-03T02:37:51 < dongs> good luck co mpiling that outside gcc
2012-09-03T02:38:08 < zyp> should compile properly in any C++11 compliant compiler
2012-09-03T02:38:10 < TheSeven> I've worked with another I2C core in the past that just has some commands for "generate start", "generate stop", "tx byte", "rx byte", and a flag for ack generation during next rx
2012-09-03T02:38:25 < TheSeven> that seems to be much easier to work with
2012-09-03T02:38:30 < zyp> TheSeven, so does this.
2012-09-03T02:38:55 < zyp> those are the bits START, STOP and ACK in CR1
2012-09-03T02:39:06 < TheSeven> so far so good
2012-09-03T02:39:10 < dongs> yea except you gotta special case sending 1, 2, or 3+ bytes
2012-09-03T02:39:10 < TheSeven> but what about RX?
2012-09-03T02:39:16 < dongs> and receiving
2012-09-03T02:39:23 < dongs> and work around sillycoon bugs
2012-09-03T02:39:55 < TheSeven> why does this core treat the address byte special at all?
2012-09-03T02:40:27 < zyp> because address byte is always written
2012-09-03T02:40:37 < TheSeven> this seems to add unnecessary complexity to the software to account for a special case that's completely pointless
2012-09-03T02:40:45 < dongs> so what
2012-09-03T02:40:55 < dongs> ^ zyp
2012-09-03T02:41:10 < TheSeven> why can't I write it just like any other regular byte?
2012-09-03T02:41:53 < zyp> because the core is getting read/write mode from the address byte
2012-09-03T02:42:08 < TheSeven> why is it doing that? and not just letting *me* tell it what it's supposed to do?
2012-09-03T02:42:14 < dongs> http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/drv_i2c_soft.c i copypasted most of this from some china sample and it worked on first try
2012-09-03T02:42:22 < zyp> TheSeven, because it's pipelined
2012-09-03T02:42:53 < TheSeven> pipelined doesn't mean that it needs to be aware of this
2012-09-03T02:43:11 < emeb> so, I dropped back to an old "blinky LED" app for my F405 board & tried programming w/ stlink & bmpm
2012-09-03T02:43:14 < TheSeven> and also if you don't even properly pipeline the SPI core, why does the much slower I2C need that?
2012-09-03T02:43:26 < emeb> this code uses no DMA, etc. Just init & blink.
2012-09-03T02:43:32 < emeb> works great w/ stlink
2012-09-03T02:43:46 < emeb> bmpm still has issues & needs hand reset.
2012-09-03T02:44:13 < TheSeven> dongs: true, bitbanging seems easier than talking to this core
2012-09-03T02:44:29 < zyp> emeb, then it's either your hardware or some buggy software on the BMP
2012-09-03T02:44:37 < emeb> Yep.
2012-09-03T02:44:42 < TheSeven> zyp: what's your problem with that synopsys OTG core btw?
2012-09-03T02:44:48 < emeb> did a continuity test on the new cable - checks out OK.
2012-09-03T02:45:01 < emeb> will look w/ ohm-meter in low range for marginal connection...
2012-09-03T02:45:04 < TheSeven> (except for the HS one wrongly reporting its fifo size in the HWCFG registers)
2012-09-03T02:45:11 < TheSeven> ah, no, it was the FS one
2012-09-03T02:45:34 < dongs> so who did  ST buy the i2c core from?
2012-09-03T02:45:38 < dongs> its same on STM8 parts as well
2012-09-03T02:46:16 < zyp> TheSeven, you mean except for all the quirks?
2012-09-03T02:46:22 < TheSeven> which quirks?
2012-09-03T02:46:30 < TheSeven> I didn't need any so far
2012-09-03T02:47:26 < zyp> well, it's usually working fine, but when it locks up and start spewing garbage, even powercycling the board doesn't help
2012-09-03T02:47:36 < TheSeven> wtf? never happened to me
2012-09-03T02:47:52 < TheSeven> only way I could lock it up so far was wrong fifo config
2012-09-03T02:47:57 < zyp> the locking up is usually caused by me fiddling around with the debugger, so that's semi-ok
2012-09-03T02:48:02 < zyp> staying locked up is not ok.
2012-09-03T02:48:16 < zyp> hmm
2012-09-03T02:48:17 < TheSeven> a reset always fixed it for me
2012-09-03T02:48:23 < zyp> wrong fifo config how?
2012-09-03T02:48:35 < TheSeven> fifos beyond end of fifo memory
2012-09-03T02:48:57 < zyp> I don't have that
2012-09-03T02:49:32 < zyp> but I'm not configuring the fifos I'm not using, my usb guru suggested that might be a problem
2012-09-03T02:49:55 < TheSeven> shouldn't be as long as you don't enable them
2012-09-03T02:50:11 < zyp> that's what I'm also assuming
2012-09-03T02:50:22 < TheSeven> because you only have separate ones per endpoint on the TX side
2012-09-03T02:50:36 < TheSeven> and as long as the EP is disabled those won't do anything
2012-09-03T02:51:02 < TheSeven> do note however that the HS OTG's fifo seems to be only 1012, not 1024 bytes, contrary to what the datasheet says
2012-09-03T02:51:20 < TheSeven> (HWCFG correctly reports 1012 bytes)
2012-09-03T02:51:30 < zyp> http://bin.jvnv.net/f/qdrrJ.png <- but that doesn't explain why it's doing this and keeps doing this no matter how much I reset the board
2012-09-03T02:51:33 < TheSeven> on the FS one it was the other way round, datasheet was right, HWCFG was wrong apparently
2012-09-03T02:51:58 < TheSeven> what's that? bus level trace?
2012-09-03T02:52:02 < zyp> you mean words, not bytes?
2012-09-03T02:52:08 < zyp> yes
2012-09-03T02:52:09 < TheSeven> yes
2012-09-03T02:52:35 < TheSeven> well I haven't looked at it on the bus level yet
2012-09-03T02:52:51 < TheSeven> but from the software side it behaves just fine for me
2012-09-03T02:53:28 < zyp> oh, and you can't complain about the i2c core being complex if you don't mind the usb core special casing SetAddress
2012-09-03T02:53:36 < zyp> :p
2012-09-03T02:53:55 < TheSeven> hahaha yeah that one is interesting
2012-09-03T02:54:11 < TheSeven> but it's very clearly documented
2012-09-03T02:54:17 < TheSeven> and it's actually convenient
2012-09-03T02:54:29 < TheSeven> offloads handling of a special case from software to hardware
2012-09-03T02:54:54 < zyp> sure, I never had a problem with it
2012-09-03T02:55:42 < zyp> it's a bit silly when you have to handle the difference between that and other cores in an abstraction layer though
2012-09-03T02:56:12 < TheSeven> yeah, I ran into that well when first implementing a driver for that core (on a completely different SoC)
2012-09-03T02:57:03 < TheSeven> anyway, if you're having trouble with this core, I have a (so far) working and fairly efficient driver for it :)
2012-09-03T02:57:08 < TheSeven> btw, is there a reliable way to pipeline (especially RX on) the SPI core on f2/f4? I can't seem to find a race condition-free way to do it
2012-09-03T02:57:24 < zyp> but really, if you can handle the usb core, the i2c core should be piece of cake
2012-09-03T02:57:49 < TheSeven> the USB one has excellent documentation compared to the I2C one
2012-09-03T02:58:22 < TheSeven> which is why I've read through that section like 10 times now and still don't understand how it's supposed to work
2012-09-03T02:58:40 < TheSeven> many things are just left totally unclear/unspecified in these docs
2012-09-03T02:58:51 < zyp> oh yeah, then where is it defined how to use the internal phy of the OTG_HS? :p
2012-09-03T02:58:53 < TheSeven> like, for example, if TXE will go to 0 while transmitting an address byte
2012-09-03T02:59:33 < TheSeven> zyp: I said "compared to the I2C on"
2012-09-03T03:00:17 < TheSeven> sure, the docs of that synopsys core are't perfect either (but samsung's and nxp's are way worse, the stm32 one is the best I found so far, I wish I would have had that when I first dealt with that core)
2012-09-03T03:00:33 < TheSeven> but the I2C docs are just rubbish
2012-09-03T03:02:41 < TheSeven> and you said you were running into trouble while trying to plug the usb driver into an existing usb framework, I'm trying to do the same for I2C right now and it's much worse
2012-09-03T03:03:04 < TheSeven>     enum i2c_result (*start)(struct i2c_driver_instance* instance);
2012-09-03T03:03:04 < TheSeven>     enum i2c_result (*stop)(struct i2c_driver_instance* instance);
2012-09-03T03:03:04 < TheSeven>     enum i2c_result (*tx)(struct i2c_driver_instance* instance, const void* buf, int len);
2012-09-03T03:03:04 < TheSeven>     enum i2c_result (*rx)(struct i2c_driver_instance* instance, void* buf, int len, bool naklast);
2012-09-03T03:03:14 < zyp> no, I didn't say that, I'm writing my framework from ground up
2012-09-03T03:03:26 < TheSeven> how do I map that interface on top of this I2C core!?
2012-09-03T03:03:59 < zyp> ouch :)
2012-09-03T03:04:09 < TheSeven> addresses are taken care of by the framework already
2012-09-03T03:04:15 < TheSeven> and that interface seems *trivial*
2012-09-03T03:04:19 < TheSeven> well, for bitbanging at least
2012-09-03T03:04:33 < TheSeven> not for overly clever controllers that distinguish stuff that they shouldn't
2012-09-03T03:04:56 < zyp> abstract at a higher layer
2012-09-03T03:05:11 < TheSeven> and duplicate the whole addressing logic into all other i2c core drivers?
2012-09-03T03:05:14 < TheSeven> seems pointless
2012-09-03T03:05:21 < TheSeven> sure, you could have multiple stacked abstractions
2012-09-03T03:05:33 < zyp> yes, that's what I'm meanin
2012-09-03T03:05:34 < zyp> g
2012-09-03T03:05:38 < TheSeven> but why doesn't this stupid core allow me to just tell it what to do?
2012-09-03T03:06:34 < TheSeven> and the datasheet is totally confusing regarding how RX works
2012-09-03T03:06:54 < TheSeven> what if I just want to receive a single byte? (which seems to be a rather common case)
2012-09-03T03:07:15 < TheSeven> do I really have to take care of that in the code that's dealing with *addressing*!?
2012-09-03T03:07:38 < TheSeven> I'll probably just ignore it and read and discard some more bytes...
2012-09-03T03:09:14 < TheSeven> works at least for register based interfaces
2012-09-03T03:09:28 < zyp> TheSeven, you probably want to have a look at the POS bit in CR1
2012-09-03T03:09:40 < zyp> hmm, no
2012-09-03T03:09:44 < zyp> not relevant
2012-09-03T03:10:12 < TheSeven> "In case a single byte has to be received, the Acknowledge disable is made during EV6 (before ADDR flag is cleared) and the STOP condition generation is made after EV6."
2012-09-03T03:10:20 < TheSeven> which isn't very clear either
2012-09-03T03:10:34 < TheSeven> but seems really stupid
2012-09-03T03:10:50 < TheSeven> why can't I trigger RX on a byte per byte basis or just give it an RX byte count in advance?
2012-09-03T03:11:06 < TheSeven> (like on all other I2C cores I've dealt with so far)
2012-09-03T03:12:44 < zyp> EV6 is «address sent», it's the point where you either write outgoing data or decide whether to ACK or NAK incoming data
2012-09-03T03:13:22 < zyp> traditionally you would trigger a read there, but that's already implied
2012-09-03T03:14:24 < zyp> http://cgit.jvnv.net/laks/tree/i2c/i2c.cpp <- I'm not sure how much sense this would make to you, but at least it works :p
2012-09-03T03:14:55 < TheSeven> it can't make much less sense than this manual...
2012-09-03T03:16:40 < zyp> my i2c driver is making a write and/or read job, triggers a START and waits for the ISR to set busy = 0 to signal that it is done
2012-09-03T03:17:01 < TheSeven> hm, doesn't look too bad for such an event-driven approach
2012-09-03T03:17:28 < TheSeven> but for a synchronous approach this interface is just ugly...
2012-09-03T03:17:34 < zyp> no, it's kind of like the SetAddress thing, you move the state machine into the hardware :p
2012-09-03T03:18:00 < TheSeven> well what you did is kinda implementing a software state machine that's communicating with a hardware state machine through events
2012-09-03T03:18:17 < zyp> yes
2012-09-03T03:18:49 < TheSeven> but this approach doesn't work well with a synchronous procedure call interface from the outside world that acts at a much lower level
2012-09-03T03:19:02 < TheSeven> also how does your code behave if a device NAKs one of your bytes?
2012-09-03T03:19:11 < TheSeven> it looks a bit like that would lock up the whole driver
2012-09-03T03:19:26 < zyp> it will trigger irq_er()
2012-09-03T03:20:05 < zyp> which currently sends a STOP and sets busy = 0, ignoring the error
2012-09-03T03:21:33 < zyp> so it does not lock up, but errors are silently ignored
2012-09-03T03:21:50 < zyp> but that's an easy fix, I just haven't bothered yet
2012-09-03T03:26:10  * TheSeven wonders how well zyp's interface would work with non-register access protocols
2012-09-03T03:27:44 < zyp> :)
2012-09-03T03:30:57 < zyp> the driver itself should be general enough unless you need other compound transactions than write -> read
2012-09-03T03:32:43 < dongs> yeah, you were right about i2c driver being most commented
2012-09-03T03:32:47 < dongs> spi.reg.CR1 = 0x37f;
2012-09-03T03:32:50 < dongs> awesome shit.
2012-09-03T03:33:07 < zyp> the SPI driver is far from being done
2012-09-03T03:33:25 < dongs> thats ok, thats not an isolated example :)
2012-09-03T03:33:32 < dongs> http://bcas.tv/paste/results/r7Hzme24.html etc
2012-09-03T03:33:36 < zyp> everything is far from being done :D
2012-09-03T03:35:12 < dongs> fucking bored
2012-09-03T03:36:22 < TheSeven> 01
2012-09-03T03:36:44 < zyp> anyway, I think having a understandable interface is more important than the implementation being understandable without digging out the reference manual
2012-09-03T03:36:45 < dongs> zyp: what the hell does laks mean.
2012-09-03T03:36:52 < zyp> salmon.
2012-09-03T03:36:57 < dongs> o h .
2012-09-03T03:37:38 < dongs> totally understandable.
2012-09-03T03:38:35 < zyp> it's a name, it doesn't need to have any meaning
2012-09-03T03:38:54 < TheSeven> it looks a bit like an acronym
2012-09-03T03:40:04 < zyp> and I didn't come up with something good, so I just picked something dumb because it's short
2012-09-03T03:40:37 < dongs> almost as creative as libmanyuc
2012-09-03T03:41:49 < zyp> I read that as lib man yuck
2012-09-03T03:41:53 < dongs> exactly
2012-09-03T03:42:33 < TheSeven> but libmanyµc wouldn't be that much better, eh?
2012-09-03T03:42:51 < zyp> :p
2012-09-03T03:46:31 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-03T03:46:49 < dongs> looks like thats also along the lines of zyp-stuff
2012-09-03T03:46:52 < dongs> all obscure C++
2012-09-03T03:49:00 < zyp> more C than C++ though
2012-09-03T03:51:09 < zyp> looks like the authors had completely different stuff than me in mind when writing that though :)
2012-09-03T04:09:53 < dongs> zyp: thsi is pretty neat
2012-09-03T04:09:55 < dongs> http://bcas.tv/paste/results/fHDJjO68.html
2012-09-03T04:10:36 < cjbaird> "Communications will be handled between the server and the clients via XML over UDP. This has many advantages [...]
2012-09-03T04:10:41 < dongs> same shit im doing by placing byte in end of ram, but this one survives gcc's retarded startup that clears ram on reset
2012-09-03T04:12:22 < TheSeven> dongs: first thing I usually do is tell gcc to not generate any startup code :P
2012-09-03T04:12:33 < dongs> first thing I usually do is not use gcc :p
2012-09-03T04:12:48 < TheSeven> well, is there something better?
2012-09-03T04:13:11 < dongs> nothing freetard, no
2012-09-03T04:13:37 < TheSeven> I don't think even the proprietary ones can match it on performance and code size
2012-09-03T04:13:45 < dongs> eh
2012-09-03T04:13:53 < TheSeven> at least if you configure it right
2012-09-03T04:13:53 < dongs> my project is consistently  smaller wiht armcc than gcc
2012-09-03T04:14:01 < dongs> like 10k+ difference
2012-09-03T04:14:22 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-03T04:14:30 < dongs> we've been through this before, a bunch of dudes tried to compile my shit with freetard stuff and it was always bigger
2012-09-03T04:14:42 < cjbaird> Try asking dongs what he's doing here on freenode.. :)
2012-09-03T04:14:42 < zyp> I like the trick BMP is using; it's setting the button pin to output and pulling it low
2012-09-03T04:14:48 < TheSeven> 10k across how much total?
2012-09-03T04:14:51 < TheSeven> linking any bloated libgcc?
2012-09-03T04:14:56 < TheSeven> using -flto -Os?
2012-09-03T04:15:20 < dongs> like 38k with armcc, ~50ishk with gcc.
2012-09-03T04:15:25 < dongs> yeah i dunno, they tried a bunch of shit
2012-09-03T04:17:01 < zyp> doesn't tell anything about performance though
2012-09-03T04:27:53 < TheSeven> enum i2c_result (*txn)(struct i2c_driver_instance* instance, int address, const void* txbuf, int txlen, void* rxbuf, int rxlen);
2012-09-03T04:28:01 < TheSeven> think that covers every valid I2C use case?
2012-09-03T04:28:50 < zippe> TheSeven: no
2012-09-03T04:29:17 < zippe> dongs: Yes, but you used to claim it was 60k bigger
2012-09-03T04:29:22 < TheSeven> ok, which use cases do you think of?
2012-09-03T04:29:33 < zippe> TheSeven: multiple repeated starts
2012-09-03T04:29:41 < zippe> TheSeven: timeouts
2012-09-03T04:29:56 < zippe> TheSeven: dealing with devices like EEPROMs that block until they are not busy
2012-09-03T04:30:07 < zippe> TheSeven: retry attempts
2012-09-03T04:30:32 < TheSeven> well error handling could be wrapped around that (yes, it would require addition of a timeout)
2012-09-03T04:30:45 < TheSeven> what about multiple repeated starts? do such cases actually exist?
2012-09-03T04:31:14 < zippe> yes
2012-09-03T04:32:10 < zyp> isn't blocking usually achieved by clock stretching?
2012-09-03T04:32:24 < dongs> dudes (ab)use multiple repeat starts when they're bored
2012-09-03T04:32:33 < zippe> zyp: depends
2012-09-03T04:32:35 < dongs> how about a legit use case zippe
2012-09-03T04:32:44 < zippe> zyp: some devices simply don't respond while they're busy
2012-09-03T04:32:56 < dongs> that is true
2012-09-03T04:33:00 < zippe> dongs: for repeated starts?  couple of sensors I can't name only give you one byte at a time
2012-09-03T04:33:07 < dongs> Laurenceb (ab)used that wiht some ADC chip
2012-09-03T04:33:12 < zippe> running a separate transaction for each byte gives crap performance
2012-09-03T04:33:15 < TheSeven> "not responding" would be a nak while addressing, right?
2012-09-03T04:33:22 < dongs> zippe: right
2012-09-03T04:33:26 < zippe> TheSeven: it looks like the device just isn't there
2012-09-03T04:33:47 < TheSeven> which should effectively be the same
2012-09-03T04:33:58 < zyp> zippe, the difference in overhead is not that large
2012-09-03T04:34:08 < zippe> zyp: beg to differ
2012-09-03T04:34:17 < zippe> zyp: especially where audio is involved
2012-09-03T04:34:56 < zyp> repeat start or not, you still have to resend the address byte
2012-09-03T04:35:20 < zippe> zyp: but it's nice if you don't have to schedule a thread to commission a new transaction just to deal with it
2012-09-03T04:35:21 < dongs> repeatblog
2012-09-03T04:35:46 < zippe> If the interrupt-level state machine knows about it, you can keep the overhead manageable
2012-09-03T04:35:57 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-03T04:36:07 < TheSeven> enum i2c_result (*tx)(struct i2c_driver_instance* instance, int address, const void* buf, int len, bool stop);
2012-09-03T04:36:07 < TheSeven> enum i2c_result (*rx)(struct i2c_driver_instance* instance, int address, void* buf, int len, bool stop);
2012-09-03T04:36:07 < TheSeven> what about that? does it cover every possibility except for timeouts?
2012-09-03T04:37:03 < zippe> So what just about everyone does is supply an array of transaction structs
2012-09-03T04:37:20 < zippe> Each struct is marked 'read' or 'write' and includes a buffer pointer and length
2012-09-03T04:37:33 < dongs> lunix does that i think.
2012-09-03T04:37:42 < zippe> If you want to be sneaky, you can recycle the buffer pointer as a buffer for <= (sizeof void *) bytes.
2012-09-03T04:38:11 < zippe> The API is then i2c_transfer(txn * txns, int num_txns)
2012-09-03T04:38:28 < TheSeven> well that should be wrappable around my proposed interface, right?
2012-09-03T04:38:30 < zippe> And the result can be anything from pass/fail to "stopped in this txn at this offset for this reason)
2012-09-03T04:38:49 < zippe> TheSeven: No, because again, the interrupt-level state machine wants to see the txn array
2012-09-03T04:38:57 < zippe> So that it can e.g. issue repeated starts between txns
2012-09-03T04:39:00 < dongs> its too bad you gotta bang i2c irq for each fucking byte
2012-09-03T04:39:09 < zippe> dongs: don't fucking get me started
2012-09-03T04:39:10 < dongs> otherwise it would be nice to just configure hardware+buffers and forget it.
2012-09-03T04:39:24 < zippe> dongs: PA Semi had a kickass I2C + DMA engine
2012-09-03T04:39:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-03T04:39:37 < zippe> but nobody with any balls will commit it to silicon
2012-09-03T04:39:45 < TheSeven> zippe: so we really need to know about future transactions in the IRQ handler for this shitty core!?
2012-09-03T04:40:27 < zippe> TheSeven: almost all i2c cores suck; the ST one is about the middle of the road
2012-09-03T04:40:51 < zippe> TheSeven: but in short, if you want your performance to be any good, yes
2012-09-03T04:41:18 < TheSeven> I don't care about a couple lost microseconds somewhere
2012-09-03T04:41:40 < zippe> TheSeven: You will when you are trying to do a megabit and you only have eight microseconds between bytes
2012-09-03T04:41:49 < zippe> You can easily piss 50% of your performance away right there
2012-09-03T04:41:58 < TheSeven> so does this core need more information than "a repeated start is going to follow at some point" vs. "please send a stop"?
2012-09-03T04:42:18 < zippe> The core doesn't need that information
2012-09-03T04:42:22 < zippe> The interrupt handler needs it
2012-09-03T04:42:25 < zippe> the core is *stupid(
2012-09-03T04:42:49 < zippe> before you receive or send the last byte for a given txn, you need to know what you're going to do after that byte
2012-09-03T04:42:50 < TheSeven> if the core wouldn't need it (or something derived from it) the IRQ handler wouldn't need it either
2012-09-03T04:43:17 < zippe> Why do I feel like I'm teaching I2C 101 here?
2012-09-03T04:43:32 < zippe> The IRQ handler needs to know what to tell the core to do ahead of tiem.
2012-09-03T04:44:03 < TheSeven> well for all I2C cores I've dealt with so far that information wasn't necessary at all
2012-09-03T04:44:24 < zippe> Either it can always say "end the txn" and then force a trip back to the client to get a new txn to continue, or it can know what's coming next and just get on with it
2012-09-03T04:45:04 < zippe> The former is the sort of dumb-ass I2C driver design that makes me curse you and your get to the seventh generation, the latter can at least sometimes be worked with.
2012-09-03T04:45:45 < TheSeven> i2c usually isn't used for terribly performance critical stuff
2012-09-03T04:46:10 < zippe> You know, I was just writing something to the effect that people that think that performance and latency don't matter for I2C are idiots
2012-09-03T04:46:49 < TheSeven> people who are using I2C for stuff where that matters might be some
2012-09-03T04:46:58 < zippe> This is the same stupid that leads to things like the miserable frame rate on Android phones
2012-09-03T04:47:06 < zippe> Or their sucktastic user-level audio latency
2012-09-03T04:47:17 < dongs> > user-level audio latency
2012-09-03T04:47:20 < dongs> sounds liek youre talkin about lunix
2012-09-03T04:47:30 < zippe> because "nobody cares about low-latency audio", and "let's write the graphics stack in Java"
2012-09-03T04:48:05 < TheSeven> I doubt it's a few microseconds that counts there...
2012-09-03T04:48:09 < dongs> yep, lunix indeed
2012-09-03T04:48:12 < zippe> Have you looked at the latest Exynos SoCs?
2012-09-03T04:48:14 < dongs> TheSeven: you'd be surprised
2012-09-03T04:48:21 < zippe> Samsung put *eight* I2Cs on the thing
2012-09-03T04:48:28 < dongs> haha
2012-09-03T04:48:29 < dongs> why?
2012-09-03T04:48:37 < zippe> because nobody in Android-land can make multiple devices on a single I2C work well
2012-09-03T04:48:38 < dongs> so you could i2c while you i2c, yo dawg?
2012-09-03T04:48:42 < dongs> haha.
2012-09-03T04:48:45 < dongs> thats awesome.
2012-09-03T04:48:51 < zippe> So they have one interface per anticipated devices.
2012-09-03T04:49:04 < TheSeven> sounds like spi might be the better protocol for that
2012-09-03T04:49:17 < zippe> TheSeven: Sorry, I am coming over all aggro and jaded and that's probably not fair.
2012-09-03T04:49:32 < zippe> Please believe me; performance and reliability are very important for I2C
2012-09-03T04:49:47 < TheSeven> well not on whatever I'm dealing with right now at least
2012-09-03T04:49:50 < zippe> The bus was designed for things like IR receiver chips and telephone keypads
2012-09-03T04:50:14 < zippe> I knew, when I was younger, a chip designer at Philips that had a good deal to do with it in the early days
2012-09-03T04:50:56 < zippe> This is the sort of guy that had his first chip schematic, in pencil on an A4 sheet, framed on his wall.
2012-09-03T04:51:17 < zippe> It was never meant to be abused in the ways it's abused today.
2012-09-03T04:53:32 < zyp> don't support i2c abuse! don't write efficient i2c drivers!
2012-09-03T04:53:39 < zyp> *chants*
2012-09-03T04:53:53 < dongs> haha
2012-09-03T04:54:26 < zyp> seriously, consider what the code is going to be used for before making design decisions
2012-09-03T04:55:27 < zyp> otherwise you'll end up with a massively overengineered solution like Laurenceb's transaction-queue stuff that he spent weeks trying to get working
2012-09-03T04:55:33 < dongs> hahaha
2012-09-03T04:56:05 < dongs> ya i dunno why he borehred with that
2012-09-03T04:56:09 < dongs> i ripped that part out.
2012-09-03T05:01:17 < TheSeven> zippe: I'm not likely to run into I2C transfers that are >65535 bytes at any point, right?
2012-09-03T05:06:18 < TheSeven> or a transaction (from start to stop) that consists of more than 255 transfers
2012-09-03T05:06:28 < TheSeven> i.e. more than 254 repeated starts in a roqw
2012-09-03T05:06:29 < TheSeven> row*
2012-09-03T05:14:35 < TheSeven> does anyone here happen to know how 10bit i2c works?
2012-09-03T05:15:07 < TheSeven> assuming there would ever be a repeated start in the TX direction, would that send the low address byte again?
2012-09-03T05:15:18 < TheSeven> or does that only happen after the first (non-repeated) start?
2012-09-03T05:16:18 < dongs> In a Combined Transfer, after the Repeated Start in some cases only the first byte of the 10 bit address is transmitted. Please consult the I2C specification for details.
2012-09-03T05:16:33 < dongs> sounds like awesome aids to specialcase.
2012-09-03T05:16:45 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-03T05:25:45 < TheSeven> indeed
2012-09-03T05:31:56 < TheSeven> dongs: seems like the second byte is generally transmitted for writes, not dependent on whether it's a repeated start
2012-09-03T05:32:20 < dongs> yea, dunno, never dicked with anythign 10bit.
2012-09-03T05:32:24 < dongs> 7bit was aids enough.
2012-09-03T05:32:39 < dongs> i2c sucks, i wish people would stop using it.
2012-09-03T05:32:41 < dongs> 1wire forever.
2012-09-03T05:46:52 < dongs> where are the chats at
2012-09-03T05:53:46 < zippe> HDLC forever
2012-09-03T05:55:24 < dongs> zippe, should I be doing floating point in iSR
2012-09-03T05:55:35 < dongs> or should i move it to a thread and wait on flag
2012-09-03T05:55:51 < zippe> dongs: depends on how much and which CPU
2012-09-03T05:56:02 < dongs> same m3 shit. not much, just some mul/div
2012-09-03T05:56:13 < zippe> Does it have to be floating? Can you do it fixed?
2012-09-03T05:56:43 < dongs> probably could, but not worth the effort
2012-09-03T05:57:06 < zippe> armcc should support q16
2012-09-03T05:57:16 < zippe> That's pretty speedy
2012-09-03T05:58:09 < zippe> In general depends on latency, eh.  If you don't mind a couple hundred cycles worst case for the op, do it
2012-09-03T05:59:39 < dongs> data i need to dick with only comes every 500ms
2012-09-03T06:02:06 < dongs> right now its just simple cause i process everything in timer compare isr, and once i collect the data needed I run the process() function that does some float stuff (also isnide isr)
2012-09-03T06:02:22 < dongs> i'd have  to makea a separate thread and flag/wait or something,, but its probably not really worth the trouble
2012-09-03T06:04:14 < zippe> 500ms?  Hah
2012-09-03T06:04:24 < zippe> Just lowpri the isr
2012-09-03T06:04:34 < zippe> and PREPARE TO BE NESTED 8)
2012-09-03T06:04:38 < dongs> teehee.
2012-09-03T06:31:25 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-03T06:31:32 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-03T06:58:55 < dongs> oh, rockbox.
2012-09-03T06:59:43 < dongs> ok, so dma memory base address does NOT increment each time DMA transfer is done.
2012-09-03T07:00:03 < dongs> my ragememset stuff, i dont evne touch it (only change cndtr) and it keeps clearing
2012-09-03T07:00:06 < dongs> so yeah.. hmm.
2012-09-03T07:32:01 < emeb_mac> Heh - 8 I2C buses.
2012-09-03T07:33:03 < emeb_mac> Anyone who thinks I2C is simple - take a look at the PMIC they used on the Beagleboard - One I2C bus that's multiplexed between about a dozen subfunctions in one chip.
2012-09-03T07:54:43 < dongs> not clicking
2012-09-03T08:20:01 < emeb_mac> looks fun
2012-09-03T08:38:24 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-03T08:40:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-03T08:46:15 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-03T08:54:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-03T08:54:29 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-09-03T09:04:33 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-03T09:37:01 -!- claude is now known as claude_work
2012-09-03T09:37:48 < dongs> hmm
2012-09-03T09:38:07 < dongs> i think i need to remap USART3 on F1
2012-09-03T09:38:10 < dongs> before i can use it
2012-09-03T09:38:43 < dongs> hm no, default is that.
2012-09-03T09:38:44 < dongs> wierd
2012-09-03T10:02:03 -!- dekar__ [~dekar@212.255.44.209] has quit [Ping timeout: 244 seconds]
2012-09-03T10:05:02 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-03T10:07:30 -!- dekar__ [~dekar@212.255.40.250] has joined ##stm32
2012-09-03T10:07:31 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-09-03T10:18:15 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Ping timeout: 252 seconds]
2012-09-03T10:25:30 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-03T10:44:01 < dongs> dafUQ usart3 irq is not firing.
2012-09-03T10:45:38 < zyp> do you want me to ask «did you enable it?»?
2012-09-03T10:47:29 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-03T10:48:48 < dongs> yeah, i just ufcking noticed i didnt
2012-09-03T10:49:00 < dongs> i enabled iti n nvic but not in uart regs
2012-09-03T10:49:25 < dongs> wokrs.
2012-09-03T10:49:43 < zyp> is something wrong with your keyboard today?
2012-09-03T10:49:54 < dongs> you mean as opposed to any other itme?
2012-09-03T10:50:13 < dongs> here's what i shitted out in another channel just few minutes ago
2012-09-03T10:50:13 < dongs>  < dongs> why is this hcanenl aw laySDLFJSDlfj have haliek epoepel thath are liek "<somethnimc>+_DRUnK>"
2012-09-03T10:50:26 < dongs> at least i'm making some effort ot make sense in here.
2012-09-03T10:51:58 < zyp> so, you have been drinking? I seem to remember you claiming that you don't drink
2012-09-03T10:52:06 < dongs> i dont.,
2012-09-03T10:52:11 < dongs> drink.
2012-09-03T10:52:11 < dongs> ever.
2012-09-03T10:52:30 < zyp> except you just did? :p
2012-09-03T10:52:41 < dongs> do you see a dongsDRUNK after my nick?
2012-09-03T10:52:41 < dongs> nope.
2012-09-03T10:53:21 < dongs> zyp, best setup for UART: TX+GND+RX on a2.54 header.
2012-09-03T10:53:31 < dongs> wired it wrong? flip the connector
2012-09-03T10:54:24 < zyp> bad part: you can't use gnd as a reference point for remembering which way it's supposed to go
2012-09-03T10:54:46 < dongs> yes but you can just flip it if it doesnt work
2012-09-03T11:27:08 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-03T11:38:28 < zippe> You can do that with power as well if you have a 6-pin connector
2012-09-03T12:12:03 -!- pelrun [~pelrun@203-206-187-234.perm.iinet.net.au] has joined ##stm32
2012-09-03T12:14:03 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-03T12:14:04 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-03T12:14:04 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-03T12:44:11 < Laurenceb> zyp: yeah but my i2c works
2012-09-03T12:44:28 < Laurenceb> well... unless you try and submit something in the middle of an i2c isr
2012-09-03T12:48:31 < zyp> and how many weeks did it take to get it working? :p
2012-09-03T12:48:42 < Laurenceb> lol
2012-09-03T12:48:53 < Laurenceb> to be fair i was using a board with no jtag
2012-09-03T12:49:03 < Laurenceb> which i now see was a bit stupid
2012-09-03T12:49:19 < Laurenceb> debugging with scope
2012-09-03T12:49:52 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-03T12:52:26 < Laurenceb> wtf pcbtrain
2012-09-03T12:52:36 < Laurenceb> i ordered 5 day delivery 2 weeks ago
2012-09-03T12:52:44 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-03T12:57:59 < Laurenceb> http://i.imgur.com/jW4FC.png
2012-09-03T12:58:03 < Laurenceb> beats XD
2012-09-03T13:02:37 < Laurenceb> getting very clean performance with gated pwm - seems the isr lag jitter was causing noise before
2012-09-03T13:11:34 < Laurenceb> http://stash.warp.net/media/img/Kidsmac.jpg
2012-09-03T13:12:45 < dongs> totally safe
2012-09-03T13:18:23 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-03T13:19:49 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 246 seconds]
2012-09-03T13:19:54 -!- soafee-chan is now known as spacekitteh
2012-09-03T13:22:04  * Laurenceb yawnz
2012-09-03T13:22:13 < Laurenceb> where the hell are my pcbs
2012-09-03T13:22:18 < dongs> i haz them
2012-09-03T13:22:18 < Laurenceb> nothing to do here :(
2012-09-03T13:22:31 < Laurenceb> this country sucks
2012-09-03T13:22:32 < dongs> do you still order shit from seeeeeeedstudio
2012-09-03T13:22:41 < dongs> and wait like a month
2012-09-03T13:22:43 < Laurenceb> not for work stuff
2012-09-03T13:22:45 < dongs> oh
2012-09-03T13:22:49 < Laurenceb> its not "approved"
2012-09-03T13:22:50 < dongs> where frommmm for work stouff
2012-09-03T13:22:52 < dongs> haha
2012-09-03T13:22:55 < Laurenceb> has to be in EU
2012-09-03T13:22:57 < dongs> its barely "approved" for hobby shit
2012-09-03T13:23:07 < dongs> fucks cant drill worth a shit
2012-09-03T13:23:11 < dongs> so where hten, olimexi?
2012-09-03T13:23:59 < Vutral> soooo
2012-09-03T13:25:13 < dongs> fuck
2012-09-03T13:25:19 < dongs> something on my desk is high pitch noising
2012-09-03T13:25:22 < dongs> and i cant find what the hel lit is
2012-09-03T13:25:37 < Laurenceb> pcbtrain is the only place we use
2012-09-03T13:25:44 < dongs> like ~14khz or so
2012-09-03T13:26:12 < Laurenceb> http://www.flickr.com/photos/daveake/7915901300/in/set-72157630074184822/lightbox/
2012-09-03T13:26:17 < Laurenceb> slightly excessive
2012-09-03T13:26:56 < dongs> haha
2012-09-03T13:28:30 < dongs> omfg raging
2012-09-03T13:28:44 < dongs> the whining is raging me out
2012-09-03T13:28:49 < dongs> and i cant find what it is111111111
2012-09-03T13:30:42 < Laurenceb> power supply?
2012-09-03T13:30:46 < dongs> no idea
2012-09-03T13:31:40 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-03T13:32:34 < dongs> found it
2012-09-03T13:32:37 < dongs> was a fucking scanner power supply
2012-09-03T13:32:44 < dongs> on the desk behind me
2012-09-03T13:33:30 < BrainDamage> next time, a simple mic and an audio intensity monitor can work for those tasks
2012-09-03T13:34:01 < BrainDamage> connecting it to a stm32 controlled rocket launcher, is an exercise left to the reader
2012-09-03T13:37:31 < Laurenceb> lol
2012-09-03T13:52:42 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-03T13:52:43 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-03T13:52:43 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-03T13:52:46 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-03T14:07:21 < dongs> di.fm hardcore is playing 'ecuador'
2012-09-03T14:07:25 < dongs> thinkin' of assange
2012-09-03T14:12:01 < Laurenceb> lol
2012-09-03T14:12:23 < Laurenceb> i havent listened to di.fm for years
2012-09-03T14:12:37 < Laurenceb> used to listed in 2004 when i was at school
2012-09-03T14:12:40 < dongs> I have like 4 of their shits on a list, and when they start playing vocal trash i switch to another one
2012-09-03T14:12:42 < Laurenceb> *listen
2012-09-03T14:12:49 < Laurenceb> lol
2012-09-03T14:13:07 < dongs> hardcore/trance/goa/i tihnk some other channel
2012-09-03T14:13:27 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-03T14:13:45  * Laurenceb is considering milling some enclosures
2012-09-03T14:13:46 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-03T14:13:48 < Laurenceb> cant be bothered
2012-09-03T14:15:24 < Laurenceb> boss is like "shove a soldering iron into the side and ram it up and down"
2012-09-03T14:16:36 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 248 seconds]
2012-09-03T14:16:37 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-09-03T14:16:38 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-03T14:16:51 < Laurenceb> i work with some grade 1 retards
2012-09-03T14:17:49 < Laurenceb> off 2 milling rig
2012-09-03T14:50:53 -!- DaKu is now known as daku
2012-09-03T15:13:48 < Laurenceb> http://cdn-static.zdnet.com/i/story/70/00/003402/original/bristol-raspberry-jam-honnibol-1-620x.jpg?hash=ZwSvAGHkZJ
2012-09-03T15:13:51 < Laurenceb> lolling
2012-09-03T15:21:03 < BrainDamage> a pi that displays feeds about the lack of pi?
2012-09-03T15:26:52 < Laurenceb> with bbc model b emulation
2012-09-03T15:27:07 < Laurenceb> on original model b monitor
2012-09-03T15:27:13 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-03T15:34:40 < karlp> how zippe: isn't hdlc just a protocol ontop of something like i2c/spi/uart/anythingyoulike?
2012-09-03T15:34:54 < dongs> he was obviosuly trolling
2012-09-03T15:35:24 < karlp> hdlc is quite nice, and far superior to everyone making their own protocols on top of serial, but with i2c, yeah, I was getting abit confused
2012-09-03T15:37:46 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-03T15:39:18 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-03T15:41:07 < Laurenceb> whit wtf
2012-09-03T15:41:10 < Laurenceb> *shit
2012-09-03T15:41:21 < Laurenceb> trying to work out whats going on with traffic
2012-09-03T15:41:41 < Laurenceb> i can see a bazillion fire trucks and ambulances about 1 mile away with binoculars
2012-09-03T15:41:51 < Laurenceb> people in biohazard suits
2012-09-03T15:42:00 < dongs> they'er coming for you
2012-09-03T15:42:05 < karlp> response to that "joey" payload
2012-09-03T15:42:12 < dongs> they found your wank rags
2012-09-03T15:42:28 < dongs> (the reason for biohazard suits)
2012-09-03T15:42:39 < Laurenceb> lol
2012-09-03T15:42:58 < Laurenceb> 8 fire trucks 10 ambulances and about 20 police vehicles
2012-09-03T15:44:47 < dongs> you're bigger than i thought
2012-09-03T15:45:08 < Laurenceb> ah i see - crashed petrol tanker
2012-09-03T16:04:22 < Laurenceb> did you hear about the jewish peado?
2012-09-03T16:04:42 < Laurenceb> he hides in bushes and sells sweets to children
2012-09-03T16:05:18 < [7]> dongs: re high pitch whining: annoy-a-tron?
2012-09-03T16:05:57 < dongs> yeah, noone would dare to troll my lair.
2012-09-03T16:06:26 < dongs> [7]: why is rockbox font rendering so fucking obtuse
2012-09-03T16:06:35 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-03T16:06:39 < dongs> does font engine return a bitmap to display driver
2012-09-03T16:06:39 < dongs> or waht
2012-09-03T16:06:43 < dongs> cjbeard has joined
2012-09-03T16:06:46 < [7]> no idea, I always stayed away from the UI stuff
2012-09-03T16:06:48 < Laurenceb> alternative version: he has a van with "cheap sweetz" on the side
2012-09-03T16:08:21 < dongs> argh
2012-09-03T16:08:45 < dongs> [7]: what do you do there hten
2012-09-03T16:09:07 < [7]> I ported it to two devices
2012-09-03T16:09:22 < [7]> so lowlevel hw backend stuff
2012-09-03T16:09:54 < dongs> i wish shit would get ported to cretin zen
2012-09-03T16:10:04 < dongs> hardware is fine but fuck the original firmware is worthless
2012-09-03T16:10:30 < dongs> <creative> lets leave backlight ON while playing MUSIC so the battery lasts longer
2012-09-03T16:11:28 < dongs> what did you make it work on
2012-09-03T16:11:37 < [7]> ipod nano 2g and ipod classic
2012-09-03T16:11:38 < cjbaird> The original Zen had audio quality problems?
2012-09-03T16:11:46 < dongs> cjbaird: it did?
2012-09-03T16:11:47 < [7]> which are a real PITA to work with
2012-09-03T16:11:52 < dongs> zen MX had some shittastic audio
2012-09-03T16:11:59 < cjbaird> There was something major..
2012-09-03T16:11:59 < dongs> zen is ok
2012-09-03T16:12:06 < dongs> mx = gimped zen
2012-09-03T16:12:08 < [7]> because there's *nothing* documented, you have to reverse engineer the whole SoC from disassemblies
2012-09-03T16:12:10 < dongs> some totally differnt chip
2012-09-03T16:12:10 < cjbaird> Wait, I'm thinking of Zune. :)
2012-09-03T16:12:21 < dongs> lawl
2012-09-03T16:12:26 < [7]> and bootloaders with RSA signature verification and whatnot
2012-09-03T16:12:31 < [7]> it's apple...
2012-09-03T16:12:47 < dongs> sounds about as useful as porting NetBSD to Acorn
2012-09-03T16:12:47 < [7]> we have to boot through a buffer overflow exploit in the bootloader!
2012-09-03T16:12:53 < dongs> i.e. dead sw on dead hw
2012-09-03T16:14:55 < cjbaird> The iRiver T-series/prism players were great; handled OGG (w/ better bass than MP3), and the internal mic recorded to MP3.
2012-09-03T16:15:04 < dongs> > ogg
2012-09-03T16:15:12 < dongs> there's the problem
2012-09-03T16:15:16 < dongs> rockbox records to mp3 on any supported hardware.
2012-09-03T16:15:52 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-03T16:15:55 < dongs> waht rockbox lacks is usability testing
2012-09-03T16:16:03 < dongs> the UI is fucking homorific
2012-09-03T16:17:28 < Laurenceb> heh acorn
2012-09-03T16:17:32 < Laurenceb> i have one somewhere
2012-09-03T16:17:38 < cjbaird> ogg >>>>>>> mp3 on my NIN Halo encodes. When I first got the T30, I tried both formats directly from my CDs, and the difference was fucking incredible.
2012-09-03T16:18:29 < cjbaird> This new E300 I got yesterday probably doesn't do ogg, given that its mic records to wma :/
2012-09-03T16:20:20 -!- pelrun [~pelrun@203-206-187-234.perm.iinet.net.au] has quit [Ping timeout: 248 seconds]
2012-09-03T16:22:19 < cjbaird> Sweet! It does do ogg! :D
2012-09-03T16:26:28 < cjbaird> Retailers prosecuted for selling dangerous childrens toys: http://www.monthlychronicle.com/gen_images/newspic010512.jpg
2012-09-03T16:27:21 < dongs> what kinda dangerous shit could you do with that?
2012-09-03T16:27:23 < dongs> stick it in y our ass?
2012-09-03T16:28:40 < cjbaird> joke -> *   dongs ------------------------------------> *
2012-09-03T16:30:11 < dongs> i dont get it
2012-09-03T16:30:14 < dongs> even after explasnation
2012-09-03T16:33:09 < Laurenceb> peadorattles?
2012-09-03T16:41:21 < Tzacks> hey guys, sry to disturb, but do you know any software that i can use to generate a color-code map from bmp?
2012-09-03T16:45:46 < dongs> a what
2012-09-03T16:45:49 < cjbaird> As in a palette?
2012-09-03T16:46:11 < dongs> to answer this question you must  first answer the question.
2012-09-03T16:53:36 < cjbaird> The NetPBM (Unix..) suite has the ppmhist program, which does stuff like: http://dpaste.org/U38QZ/
2012-09-03T16:53:55 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-03T16:54:05 < dongs> haha pbm
2012-09-03T16:58:11 < cjbaird> http://netpbm.sourceforge.net/doc/ppmtomap.html
2012-09-03T16:59:00 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-03T16:59:01 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-03T16:59:01 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-03T16:59:04 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-03T17:02:19 < [7]> zyp: are you sure your driver is correct?
2012-09-03T17:02:36 < [7]> http://cgit.jvnv.net/laks/tree/i2c/i2c.cpp#n25
2012-09-03T17:03:06 < [7]> shouldn't that CR1 write be before the DR write? otherwise you might get a race condition, right?
2012-09-03T17:03:42 < [7]> er, I mean this one: http://cgit.jvnv.net/laks/tree/i2c/i2c.cpp#n39
2012-09-03T17:03:48 < [7]> before the DR read
2012-09-03T17:06:43 < Tzacks> ppm what is it?
2012-09-03T17:07:10 < dongs> some obsolete lunix graphics format
2012-09-03T17:07:21 < Tzacks> aha
2012-09-03T17:10:10 < Laurenceb> [7]: i seem to remeber that issue
2012-09-03T17:10:10 < Tzacks> well, nwm, maybe i'll pixel it out or... well.. i'm lazy... :)
2012-09-03T17:11:11 < Laurenceb> [7]: check my "Dactyl" project on github
2012-09-03T17:11:35 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-03T17:11:37 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-03T17:15:01 < dongs> Tzacks: im sure tyhere's something in "the gimp" tht will do it fo ryou
2012-09-03T17:15:26 <+Steffanx> nu
2012-09-03T17:15:46 < Laurenceb> gimpiest name ever
2012-09-03T17:21:24 < karlp> gimp will even save it as a .h file for you
2012-09-03T17:22:09 -!- daku is now known as DaKu
2012-09-03T17:23:59 -!- claude_work is now known as claude
2012-09-03T17:24:23 < BrainDamage> save to ppm format for some silly ascii art .h file https://en.wikipedia.org/wiki/X_PixMap
2012-09-03T17:24:51 -!- Piele [~peter@5ED0EA16.cm-7-1d.dynamic.ziggo.nl] has joined ##stm32
2012-09-03T17:25:14 < Tzacks> BrainDamage: that looks quite good
2012-09-03T17:25:53 < Tzacks> thx maybe i'll use it this way
2012-09-03T17:26:11 < Tzacks> dongs: how can you use gimp grrr :)
2012-09-03T17:26:21 < BrainDamage> he doesn't
2012-09-03T17:26:26 < Tzacks> :)
2012-09-03T17:26:39 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-03T17:26:58 < BrainDamage> the probability of dongs' using open source stuff is as much as Laurenceb getting laid
2012-09-03T17:27:08 < Laurenceb> BrainDamage: thanx for that
2012-09-03T17:27:18 < Laurenceb> that would be P=1 then
2012-09-03T17:27:23 < dongs> i gues Laurenceb is getting laid today
2012-09-03T17:27:30 < dongs> cuase i've got a bunch of opensores shit on my desktop :(
2012-09-03T17:27:33 < Laurenceb> unfortunately not
2012-09-03T17:27:45 < Laurenceb> i got laid a few months back tho
2012-09-03T17:28:34 < Tzacks> :)))))
2012-09-03T17:28:45 < Laurenceb> http://www.chickenalaska.com/store/media/images/t_laid.jpg
2012-09-03T17:31:47 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-03T17:41:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-03T17:41:08 < Laurenceb> oh shit
2012-09-03T17:41:23 < Laurenceb> in fact i once linked my ex gfs linkedin page on here
2012-09-03T17:41:36 < Laurenceb> shes gonna get so trolled
2012-09-03T17:42:00 <+Steffanx> You did?
2012-09-03T17:42:28 <+Steffanx> Where where? :P
2012-09-03T17:43:39 < Laurenceb> oh dear
2012-09-03T17:44:00 < Laurenceb> maybe it wasnt here actually
2012-09-03T17:44:21 <+Steffanx> You taught me to behave like this, Laurenceb
2012-09-03T17:47:34 < cjbaird> Here's my ex :P .. http://au.linkedin.com/pub/charlotte-brogden/31/a24/446
2012-09-03T17:47:48 <+Steffanx> We know that one cjbaird
2012-09-03T17:48:05 <+Steffanx> And we know zlog
2012-09-03T17:49:03 < cjbaird> Yes, I was just making sure that zlog knew..
2012-09-03T17:51:33 < Laurenceb> "cant fap to this"
2012-09-03T17:52:16 < Laurenceb> wow what a list of skillz
2012-09-03T17:53:50 <+Steffanx> "Annoy C.J. Baird" is missing :P
2012-09-03T17:54:08 < cjbaird> She omits the Cicada genital microphotography that she did after leaving Uni. :P :)
2012-09-03T17:57:15 < cjbaird> She and a few other friends were the frustrated Science Student type-- get the degree, get a science job, ....and discover that Dr Karl was a lying fuckface and science jobs are boring and unfulfilling and pay like crap. So dump that and get a job in the book industry.
2012-09-03T18:02:29 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-03T18:09:06 <+Steffanx> BrainDamage awake?
2012-09-03T18:09:50 < BrainDamage> yes
2012-09-03T18:09:57 < BrainDamage> I'm in EU too
2012-09-03T18:10:01 <+Steffanx> :P
2012-09-03T18:10:13 <+Steffanx> Iirc you have some link to a store on ebay with some nice camera modules
2012-09-03T18:10:17 <+Steffanx> Remeber which one that was?
2012-09-03T18:10:22 < BrainDamage> sure, sec
2012-09-03T18:11:39 < BrainDamage> http://stores.ebay.com/electronics-lee/LCD-CAM-Module-/_i.html?_fsub=1858785013&_sid=903337233&_trksid=p4634.c0.m322
2012-09-03T18:12:01 <+Steffanx> Looks like it, thanks
2012-09-03T18:12:18 < dongs> eh
2012-09-03T18:12:23 < dongs> check if waveshare cameras are cheaper
2012-09-03T18:12:29 < dongs> im prety sure tehy had some oVxxx breakouts
2012-09-03T18:12:43 < dongs> http://www.wvshare.com/product/OV7670-Camera-Board.htm
2012-09-03T18:12:59 <+Steffanx> dongs has waveshare stocks/shares ?
2012-09-03T18:13:03 < dongs> yes
2012-09-03T18:13:08 < dongs> i get paid 100% of each shit you order
2012-09-03T18:13:10 < dongs> from my irc links
2012-09-03T18:13:12 < dongs> !!!
2012-09-03T18:13:29 <+Steffanx> Inivisible referral link :D
2012-09-03T18:21:25 -!- DaKu is now known as daku
2012-09-03T18:48:47 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-03T18:53:51 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-03T18:55:07 < dongs> http://foo.is/~baldur/.cache/images.4chan.org/b/src/1346687284648.jpg attn cjbaird
2012-09-03T18:56:10 < karlp> last week's news,
2012-09-03T18:56:17 <+Steffanx> Who has the time to organize something silly like that?
2012-09-03T18:56:17 < karlp> they already won the comp didn't they?
2012-09-03T18:56:51 < dongs> http://www.boston.com/yourtown/news/allston_brighton/2012/08/allston_school_for_deaf_looks.html haha it really did
2012-09-03T18:57:40 <+Steffanx> Nice of the kids though
2012-09-03T18:58:04 < dongs> olh
2012-09-03T18:58:08 < dongs> they still have a month to go
2012-09-03T18:58:21 <+Steffanx> yeah
2012-09-03T18:58:26 <+Steffanx> "Voting in the Swift concert contest began on Aug. 21 and runs through midnight Sept. 23."
2012-09-03T19:00:27 < BrainDamage> I recall a similar thing where a guy got sent up in alaska at walmart, or similar
2012-09-03T19:00:32 -!- Rickta59 [463d49a6@gateway/web/freenode/ip.70.61.73.166] has joined ##stm32
2012-09-03T19:00:48 <+Steffanx> Yeah
2012-09-03T19:01:29 <+Steffanx> That guy, don't remember who, invited the "prankers" to join the concert
2012-09-03T19:02:19 <+Steffanx> Good luck inviting 4chan and crap to a concert
2012-09-03T19:02:32 < Laurenceb> need a gas mask for the BO
2012-09-03T19:03:08 <+Steffanx> BO ?
2012-09-03T19:03:21 < Laurenceb> body odour
2012-09-03T19:03:51 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-03T19:04:34 <+Steffanx> Oh, i was like "WTF is he up to this time" :P
2012-09-03T19:05:54 < Laurenceb> BrainDamage: I worked out how to track a rocket with 1gram tracker
2012-09-03T19:06:09 < BrainDamage> what solution did you pick?
2012-09-03T19:06:12 < Laurenceb> http://www.digikey.co.uk/product-detail/en/ANT-868-CHP-T/ANT-868-CHP-TTR-ND/1534434
2012-09-03T19:06:25 < Laurenceb> those with the msp430 chipcon transceivers
2012-09-03T19:06:34 < Laurenceb> then a 330mg lipo cell
2012-09-03T19:06:43 < Laurenceb> its doable in <1gram
2012-09-03T19:06:58 < BrainDamage> nice
2012-09-03T19:06:59 < Laurenceb> not sure if i can be bothered to build this thing...
2012-09-03T19:07:45 <+Steffanx> YES! YOU! CAN!
2012-09-03T19:07:50 < Laurenceb> was thinking an Estes D motor with nozzle extension and the outer casing taken off on a lathe
2012-09-03T19:07:50 < jpa-> how far can you receive suck tracker? or do you intend to track it with a directional antenna during flight?
2012-09-03T19:08:18 < Laurenceb> simulations suggest itd get to 80 to 90km launched from a 40km balloon
2012-09-03T19:08:38 < Laurenceb> jpa-: ukhas have managed ~900km with 10mw on 434mhz
2012-09-03T19:08:45 < Laurenceb> with a yagi on the ground
2012-09-03T19:09:20 < jpa-> nice
2012-09-03T19:09:25 < karlp> the only stm32's with eeprom are l1 right?
2012-09-03T19:09:34 < jpa-> i wonder if you could increase it with spread spectrum?
2012-09-03T19:09:44 < Laurenceb> why bother
2012-09-03T19:10:13 < Laurenceb> for a rockoon you only need ~100km
2012-09-03T19:10:18 < jpa-> probably no reason for that application
2012-09-03T19:10:34 < Laurenceb> annoyingly it seems 100km altitude is impossible with Estes motors
2012-09-03T19:10:36 < jpa-> i've been thinking about something else where i need both small receive & transmit antennas
2012-09-03T19:10:47 < Laurenceb> but thats hardly surprising :P
2012-09-03T19:10:55 < jpa-> i've been wondering if any small rx/tx system could easily accommodate spread spectrum
2012-09-03T19:11:14 < BrainDamage> DSSS is not really hard per se
2012-09-03T19:11:26 < Laurenceb> normally Estes motors get about 100m altitude XD
2012-09-03T19:11:49 < BrainDamage> the height boost comes from lower atmosphere density?
2012-09-03T19:11:54 < Laurenceb> yes
2012-09-03T19:12:00 < Laurenceb> hardly any drag
2012-09-03T19:12:25 < Laurenceb> then you get about twice the ISP with lower pressure air and a nozzle extension
2012-09-03T19:12:57 < Laurenceb> then you can get a much better mass fraction by stripping off most of the cardboard and sticking it in a carbon fibre tube
2012-09-03T19:12:58 < BrainDamage> ISP= ? is it the specific impuse?
2012-09-03T19:13:00 < Laurenceb> yes
2012-09-03T19:13:24 < Laurenceb> looks like about 120seconds at 40km with the nozzle
2012-09-03T19:13:42 < Laurenceb> and the rocket gets about 1.2Km/s delta v
2012-09-03T19:14:28 < Laurenceb> but Estes D motors contain only around 24grams of propellant, so the rest of the rocket has to be very light
2012-09-03T19:15:03 < emeb> why not make your own motors w/ more burn time?
2012-09-03T19:15:18 < Laurenceb> illegal in UK
2012-09-03T19:15:27 < emeb> bummer.
2012-09-03T19:15:27 < Laurenceb> but theres an easy and legal way to do it
2012-09-03T19:15:36 < Laurenceb> use APCP motors
2012-09-03T19:15:51 < Laurenceb> you can buy and store limited numbers of them without a license
2012-09-03T19:16:00 < emeb> A guy I used to work with made his own motors. ISTR that the ingredients were fairly easy to get.
2012-09-03T19:16:05 < Laurenceb> but its hard to buy them due to the limits on stockpiling
2012-09-03T19:16:20 < Laurenceb> but if i wanted 100km id do it that way
2012-09-03T19:16:57 < BrainDamage> the problem with the diy motor solution is that announcing about it would be a guilt confession
2012-09-03T19:16:59 < Laurenceb> i just like the idea of getting to >80km with hobby shop motors
2012-09-03T19:17:09 < BrainDamage> could you detach the rocket body from the tracker when it has exausted?
2012-09-03T19:17:14 <+Steffanx> I want 50 million km .. just to crash that mars rover :P
2012-09-03T19:17:31 < BrainDamage> you don't need that much, the gravitation well will help
2012-09-03T19:17:52 < Laurenceb> wikisat are pretty funny
2012-09-03T19:18:00 < emeb> getting to escape velocity is a bitch though
2012-09-03T19:18:01 < Laurenceb> they try coke can rockets over and over again
2012-09-03T19:18:02 < BrainDamage> you need to basically escape earth + sun + moon gravity wells
2012-09-03T19:18:08 < Laurenceb> they fail every time
2012-09-03T19:18:15 < emeb> definition of insanity
2012-09-03T19:18:38 < Laurenceb> with spin stabilised ammonium perchlorate motors you can do orbit with 2 stage
2012-09-03T19:18:47 < Laurenceb> launched from a balloon
2012-09-03T19:19:13 < Laurenceb> - off the shelf reloads in carbon fibre tube casings
2012-09-03T19:19:37 < BrainDamage> mmmm, building the spinner platform doesn't sound trivial tough
2012-09-03T19:19:39 < Laurenceb> i looked inot it for n-prize, then decided i couldnt be bothered
2012-09-03T19:19:55 < BrainDamage> you can't even counterbalance with fins since the atmopshere is ridicolous up there
2012-09-03T19:20:05 < Laurenceb> BrainDamage: its the one thing ""Monreo" has done successfully
2012-09-03T19:20:14 < Laurenceb> the Texas n-prize nutcase
2012-09-03T19:20:28 < Laurenceb> fins are heavy
2012-09-03T19:20:38 < Laurenceb> spin stabilised is the way to go
2012-09-03T19:20:52 < Laurenceb> problem with orbit is you need to change the spin axis for stage 2
2012-09-03T19:21:54 < Laurenceb> which is very tricky
2012-09-03T19:22:42 < Laurenceb> wtf
2012-09-03T19:23:05 < Laurenceb> im filling in a job application and about the 5th question is "What is your sexual orientation?"
2012-09-03T19:23:26 < emeb> is that legal in UK?
2012-09-03T19:23:26 < BrainDamage> North, definetly north
2012-09-03T19:23:37 < Laurenceb> "-"
2012-09-03T19:23:49 < emeb> "orthogonal to everything"
2012-09-03T19:23:54 -!- barthess [~barthess@178.154.70.211] has joined ##stm32
2012-09-03T19:23:55 < Laurenceb> emeb: apparently its illegal not to
2012-09-03T19:24:06 < emeb> bizarre.
2012-09-03T19:24:17 < Laurenceb> they have to keep records
2012-09-03T19:24:20 < emeb> "do ask, do tell"
2012-09-03T19:24:45 < Laurenceb> i now have negative orientation
2012-09-03T19:25:29 <+Steffanx> I think Laurenceb likes trolls
2012-09-03T19:25:51 < emeb> hot troll-on-troll action
2012-09-03T19:29:08 < Laurenceb> anyways
2012-09-03T19:29:30 < emeb> back to the rockets...
2012-09-03T19:30:03 < Laurenceb> if adc is running injected conversions triggered from an interrupt, can code outside the interrupt setup a normal convertion on one of the same channels used by the injected convertion group?
2012-09-03T19:30:38 < emeb> sounds unlikely
2012-09-03T19:30:57 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-03T19:30:58 -!- tavish_ [~tavish@120.59.131.163] has joined ##stm32
2012-09-03T19:31:22 < emeb> or rather, possible but difficult to choreograph.
2012-09-03T19:31:47 < Laurenceb> actually i seem to be doing it already
2012-09-03T19:32:22 < Laurenceb> yeah setting up a regular channel with software triggered conversion works
2012-09-03T19:35:00 < Laurenceb> i can't spell conversion
2012-09-03T19:35:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-03T19:39:47 <+Steffanx> ok
2012-09-03T19:41:18 < emeb> 3/5 success rate.
2012-09-03T19:41:18 < zyp> emeb, why would it be?
2012-09-03T19:42:06 < zyp> I don't see any reason why injected and normal conversions would conflict, only one is running at a time anyway
2012-09-03T19:42:24 < emeb> zyp: I guess it's not actually. Seems like making sure that the ADC is free when the injected IRQ comes along would be... fraught.
2012-09-03T19:43:04 < Laurenceb> it would
2012-09-03T19:43:14 < Laurenceb> but luckly its handled for you in hardware XD
2012-09-03T19:43:19 < Laurenceb> the ADC on stm32 is nice
2012-09-03T19:43:38  * Laurenceb -> home
2012-09-03T19:43:39 < Laurenceb> cya
2012-09-03T19:43:47 <+Steffanx> Have fun
2012-09-03T19:45:00 < emeb> The ADC on the STM32 is nice.
2012-09-03T19:45:02 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-03T19:45:46 < emeb> comparing the noisiness of the conversion data w/ dsPIC that I had been using in the past - it's a lot quieter.
2012-09-03T19:49:00 -!- barthess [~barthess@178.154.70.211] has quit [Quit: Leaving.]
2012-09-03T19:54:19 < karlp> heh, st app note with the "Nord" axis and the "east" axis
2012-09-03T19:54:59 < emeb> pick a language. stick with it.
2012-09-03T20:20:14 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-03T20:33:18  * emeb needs to do better QA on his soldering - reflowed one of my non-functional boards and it started working.
2012-09-03T20:38:56 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-03T20:39:49 < TitanMKD> hello
2012-09-03T20:44:22 < emeb> F05x (on the Discovery F0) has a 12-bit ADC.
2012-09-03T20:45:05 < emeb> looks pretty much the same as that on other STM32 parts.
2012-09-03T20:52:19 < zyp> emeb, solved the flashing problems?
2012-09-03T20:54:43 < emeb> zyp: no actually. But I confirmed that the BMPM behaves the same on both F405 boards that I've got.
2012-09-03T20:55:05 < emeb> after getting board #1 working...
2012-09-03T20:55:18 < zyp> I would suggest updating the firmware on it
2012-09-03T20:55:36 < emeb> zyp: yes - it's looking like that might be important.
2012-09-03T20:55:46 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Remote host closed the connection]
2012-09-03T20:56:00 < emeb> Need to figure that out - read the docs on the wiki & git repo.
2012-09-03T20:56:32 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-03T20:56:42 < zyp> it's very straight forward, provided you're able to compile it
2012-09-03T20:57:18 < emeb> good to know.
2012-09-03T20:57:37 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-03T20:58:05 < emeb> zyp: do you use github.com/gsmcmullin/blackmagic?
2012-09-03T20:58:11 < zyp> yes
2012-09-03T20:58:34 < emeb> I wonder if the firmware is different between BMP and BMPM?
2012-09-03T21:00:41 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-03T21:00:44 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-03T21:00:54 < zyp> not other than version, as far as I know
2012-09-03T21:02:31 < zyp> FWIW I'm currently running 4055706 with a patch on top to remove the USB IAD stuff, since the OS on my old workstation didn't support IAD
2012-09-03T21:03:05 -!- [7] is now known as [7|B0T]
2012-09-03T21:03:09 -!- [7|B0T] is now known as [7]
2012-09-03T21:03:09 < karlp> what's IAD?
2012-09-03T21:03:37 < zyp> Interface Association Descriptor
2012-09-03T21:03:54 -!- tech2077_ [~tech2077@adsl-75-53-132-100.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-03T21:04:48 < zyp> AFAIK it's mainly used for windows to be able to attach multiple drivers to a single device without making a custom driver
2012-09-03T21:05:01 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-09-03T21:05:19 < zyp> i.e. having two ACM interfaces on it
2012-09-03T21:06:30 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 246 seconds]
2012-09-03T21:06:30 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 246 seconds]
2012-09-03T21:06:30 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-03T21:06:31 -!- tech2077 [~tech2077@adsl-75-53-132-100.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-03T21:06:31 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-03T21:06:31 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-03T21:06:32 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-03T21:07:01 < zyp> https://github.com/gsmcmullin/blackmagic/commit/edaae39 <- this commit is the one I reverted
2012-09-03T21:07:53 < emeb> Hmmm - looks like flashing new firmware into the BMPM requires SWD through the serial port connector.
2012-09-03T21:08:04 < zyp> no, it has a bootloader
2012-09-03T21:08:14 < emeb> Ah - USB DFU?
2012-09-03T21:08:17 < zyp> yes
2012-09-03T21:08:21 < emeb> Better.
2012-09-03T21:08:33 < zyp> you just compile it and then run scripts/bootprog.py
2012-09-03T21:08:51 < emeb> sounds simple.
2012-09-03T21:09:49 < emeb> the HACKING file says "This requires an appropriate cable to connect the PC serial port to the probe.
2012-09-03T21:09:49 < emeb> See the schematic for more information." which sounds more involved than just USB DFU.
2012-09-03T21:10:36 < zyp> sorry, I remembered incorrectly, bootprog.py is for the hardware bootloader, I meant stm32_mem.py
2012-09-03T21:11:23 < emeb> Ah - that makes more sense.
2012-09-03T21:13:59 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-03T21:16:05  * emeb needs to better understand these USB bootloaders - could be useful.
2012-09-03T21:21:35 -!- barthess [~barthess@164.177.193.241] has joined ##stm32
2012-09-03T21:26:50 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-03T21:31:33 < zippe> emeb: another option would be https://github.com/PX4/Bootloader
2012-09-03T21:31:55 < zippe> That gives you the same bootloader protocol for serial and USB
2012-09-03T21:32:07 -!- claude is now known as Claude
2012-09-03T21:32:26 < emeb> zippe: ty - I'll check into that.
2012-09-03T21:32:32 < zippe> NP.  It works well for me.
2012-09-03T21:39:20 < karlp> zippe: but that means you ahve to use your programmer right? you can't just use defuse or whatever the generic dfu programmer is?
2012-09-03T21:39:32 < karlp> or is dfu not as useful as I believe?
2012-09-03T21:48:13 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-03T21:51:03 <+dekar__> karlp, I don't like DFU for the fact that you still have to install specific windows drivers matching the VID/PID
2012-09-03T21:52:08 < jpa-> blame windows for being windows :)
2012-09-03T21:52:36 -!- tech2077_ [~tech2077@adsl-75-53-132-100.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 248 seconds]
2012-09-03T21:53:39 -!- barthess [~barthess@164.177.193.241] has quit [Quit: Leaving.]
2012-09-03T21:58:39 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-03T22:03:12 < zippe> karlp: that's correct, you need to use the matching programmer
2012-09-03T22:03:25 < zippe> But that's less of an issue than you think, because the programmer is portable Python
2012-09-03T22:03:37 < zippe> So it runs just fine prettymuch anywhere
2012-09-03T22:03:47 < zippe> Wherease dfuse, etc. require specific platforms
2012-09-03T22:04:16 < zippe> (Windows can be a nuisance due to driver signing issues, but it's not insurmountable)
2012-09-03T22:04:17 -!- tech2077_ [~tech2077@76.247.251.117] has joined ##stm32
2012-09-03T22:05:10 < zippe> karlp: One other nice feature of the px4 bootloader is that the firmware transport format is JSON
2012-09-03T22:05:25 < zippe> karlp: so it's easy to add extra data to it if you need
2012-09-03T22:11:25 < emeb> buzzword compliant!
2012-09-03T22:13:35 < Laurenceb_> lol
2012-09-03T22:19:54 -!- daku is now known as DaKu
2012-09-03T22:24:49 < Titan___> does there's open source STM32 USB DFU (internal rom) for Windows/Linux ?
2012-09-03T22:25:02 -!- Titan___ is now known as TitanMKD
2012-09-03T22:25:22 < zyp> the point of DFU being standardized is that you don't need specialized tools
2012-09-03T22:25:28 < zyp> dfu-util should work fine
2012-09-03T22:26:46 < TitanMKD> ok i ask that because i never tested dfu I always use JTAG/SWD to program
2012-09-03T22:28:07 < zyp> the problem with DFU is that everybody seem to be either violating or extending the spec in some way or another
2012-09-03T22:28:25 < TitanMKD> maybe it is the case for STM32F4 bootrom
2012-09-03T22:29:22 < zyp> ST have some extensions they call dfuse
2012-09-03T22:29:35 < TitanMKD> anyway the best is maybe like LPC when plugged it simulate a USBKey MassStorage and you just need to copy the firmware on it
2012-09-03T22:29:43 -!- Claude is now known as claude
2012-09-03T22:29:49 -!- Rickta59 [463d49a6@gateway/web/freenode/ip.70.61.73.166] has quit [Quit: Page closed]
2012-09-03T22:30:03 < zyp> that sounds very overkill
2012-09-03T22:30:15 < TitanMKD> yes a friend tell me it really work like that ;)
2012-09-03T22:30:29 < TitanMKD> it is really plug & play with simple drag & drop
2012-09-03T22:30:30 < zyp> I assume he's talking about the mbed
2012-09-03T22:30:37 < emeb> primarily useful for n00b devs or mass-market devices.
2012-09-03T22:30:41 < zyp> they work like that; I own one
2012-09-03T22:30:44 < TitanMKD> yes or any LPC in fact it's in the ROM
2012-09-03T22:30:49 < zyp> no, it's not
2012-09-03T22:31:02 < zyp> the mbed contains a dedicated MCU to do that part
2012-09-03T22:31:13 < TitanMKD> but on LPC43xx it's the case in ROM
2012-09-03T22:31:27 < TitanMKD> with embedded mass storage
2012-09-03T22:31:37 < zyp> got any references on that?
2012-09-03T22:31:51 < TitanMKD> in datasheet but i never tested it
2012-09-03T22:33:13 < zyp> please tell me where I can find this datasheet and what page it says it on :)
2012-09-03T22:33:23 < TitanMKD> it's hidden feature ;)
2012-09-03T22:33:29 < TitanMKD> by default USB boot on DFU stuff
2012-09-03T22:33:33 < zyp> or like they say: [citation needed]
2012-09-03T22:34:07 < TitanMKD> it also requires rev B silicon because there's lot of bug on LPC43xx revA especially in ROM
2012-09-03T22:35:12 < TitanMKD> look UM10503 chapter 25.2 Introduction
2012-09-03T22:35:30 < TitanMKD> boot ROM contains USB driver with CDC, HID, MSC ;)
2012-09-03T22:35:49 < Laurenceb_> TitanMKD: I was writing stm32 firmware to do that
2012-09-03T22:35:56 < Laurenceb_> use mass storage and drag and drop
2012-09-03T22:36:00 < Laurenceb_> then i got bored :P
2012-09-03T22:36:12 < Laurenceb_> but its going to be big - using fatfs
2012-09-03T22:36:14 < TitanMKD> Laurenceb because RevA is too buggy ?
2012-09-03T22:36:22 < Laurenceb_> so noobs can use it
2012-09-03T22:36:23 < TitanMKD> ha yes
2012-09-03T22:36:34 < Laurenceb_> i.e. realistic end users
2012-09-03T22:36:46 < TitanMKD> yes it is a must for end user
2012-09-03T22:36:59 < Laurenceb_> i dont fancy sitting on phone all day guiding users through DFU
2012-09-03T22:37:00 < TitanMKD> especially we can program also the boot ROM it is OTP
2012-09-03T22:37:10 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-03T22:38:20 < Laurenceb_> ive taken the st example mass storage code and added some basic fatfs stuff
2012-09-03T22:38:32 < Laurenceb_> sorry no
2012-09-03T22:38:43 < Laurenceb_> mass storage is in the "application"
2012-09-03T22:38:59 < Laurenceb_> so it is possible to brick it, but you can remove the micro sd card if you do that
2012-09-03T22:39:20 < TitanMKD> or force OTP boot ;)
2012-09-03T22:39:23 < zyp> TitanMKD, I'm reading that it has a USB driver in ROM containing class drivers for ACM, HID and MSC
2012-09-03T22:39:24 < Laurenceb_> the bootloader is just a basic build of fatfs that searches for a firmware image and flashes
2012-09-03T22:39:44 < zyp> that's far from having a MSC-based drag&drop bootloader :p
2012-09-03T22:39:46 < Laurenceb_> then starts the main code - ive copied the libmaple stuff
2012-09-03T22:40:02 < Laurenceb_> zyp: yeah well its fairly small and runs on my hardware
2012-09-03T22:40:15 < TitanMKD> Laurenceb yes anyway basic fatfs can do the trick i have done that on less than 512bytes
2012-09-03T22:40:16 < zyp> Laurenceb_, I'm talking about the LPC4300
2012-09-03T22:40:24 < Laurenceb_> wut
2012-09-03T22:40:30 < Laurenceb_> 512bytes of flash?!
2012-09-03T22:40:42 < Laurenceb_> more like 5K surely?
2012-09-03T22:40:43 < TitanMKD> yes on very old mcu 8bits with EEPROM
2012-09-03T22:40:54 < TitanMKD> it was hard written fat ;)
2012-09-03T22:40:57 < Laurenceb_> oh
2012-09-03T22:41:08 < TitanMKD> without any fragment so you write it in straight
2012-09-03T22:41:09 < Laurenceb_> the smallest build of fatfs is something like 5K
2012-09-03T22:41:11 < zyp> fatfs is a library
2012-09-03T22:41:11 < Laurenceb_> i see
2012-09-03T22:41:18 < zyp> for interacting with FAT filesystems
2012-09-03T22:41:25 < Laurenceb_> erm... i know
2012-09-03T22:41:28 < TitanMKD> for a firmware updte fatfs is not really required ;)
2012-09-03T22:41:32 < Laurenceb_> oh sorry
2012-09-03T22:41:41  * Laurenceb_ fails to read scrollback
2012-09-03T22:41:58 < Laurenceb_> TitanMKD: if you want MSC its the easiest way to do it imo
2012-09-03T22:42:03 < TitanMKD> but yes for SDCARD management with FATFS there is no choice
2012-09-03T22:42:15 < Laurenceb_> then the actual MSC can be outside the bootloader
2012-09-03T22:42:27 < zyp> but I'm tired of pointless arguments now, have fun
2012-09-03T22:42:33 < Laurenceb_> and the bootloader just looks on the card
2012-09-03T22:42:57 < TitanMKD> Laurenceb ha ok you scan a sdcard for a firmware and if you find one you upgrade
2012-09-03T22:43:03 < TitanMKD> very good alternative too
2012-09-03T22:43:13 < Laurenceb_> so
2012-09-03T22:43:20 < Laurenceb_> bootloader is small that way
2012-09-03T22:43:38 < Laurenceb_> i need MSC in my app code anyway - its a datalogger
2012-09-03T22:44:04 < Laurenceb_> if its bricked the card can be removed and put in a reader
2012-09-03T22:44:05 < TitanMKD> yes you can also add safety / antibrick checks before to upgrade the FW to do not destroy the 1st sector where the bootloader is
2012-09-03T22:44:13 < Laurenceb_> yes
2012-09-03T22:44:31 < Laurenceb_> well its would make sure to skip the bootloader area when flashing
2012-09-03T22:44:37 < TitanMKD> yes
2012-09-03T22:44:48 < Laurenceb_> but you could add checksum for example
2012-09-03T22:44:58 < TitanMKD> yes a CRC32 will be good for that too
2012-09-03T22:45:41 < TitanMKD> to ensure integrity even if that do not protect against modified firmware by bad user ;)
2012-09-03T22:46:46 < Laurenceb_> so to flash, user drags image onto the device, then reboots it
2012-09-03T22:47:19 < TitanMKD> yes sounds very interesting
2012-09-03T22:47:28 < TitanMKD> to avoid such DFU pain to upgrade
2012-09-03T22:47:53 < BrainDamage> what would you use the data logger for?
2012-09-03T22:48:08 < Laurenceb_> i use it for medical logging
2012-09-03T22:48:14 < Laurenceb_> its on my github
2012-09-03T22:48:27 < Laurenceb_> runs my pneumatic spectrometer cuffs
2012-09-03T22:48:54 < BrainDamage> did you put binary files under revision control? :s
2012-09-03T22:49:04 -!- tavish_ [~tavish@120.59.131.163] has quit [Quit: z]
2012-09-03T22:49:04 < Laurenceb_> lol
2012-09-03T22:49:20 < BrainDamage> create a .gitignore file, tell it to ignore .o files
2012-09-03T22:49:24 < Laurenceb_> ok
2012-09-03T22:49:42 < BrainDamage> and .bin and .elf too ofc
2012-09-03T22:49:56 < TitanMKD> Laurenceb it use STM32 or LPC ?
2012-09-03T22:50:18 < BrainDamage> just writing *.o *.bin *.elf, one per line, should suffice
2012-09-03T22:50:38 < TitanMKD> Laurenceb because i'm building a really tiny STM32F4 tiny board 60mm*3.7mm with an aditional board i will add SDIO/MMC with 4 data lines
2012-09-03T22:50:40 < Laurenceb_> TitanMKD: STM32F103CBT6 to be precise
2012-09-03T22:50:46 < Laurenceb_> cool
2012-09-03T22:50:57 < TitanMKD> Laurenceb look my blog for the base board
2012-09-03T22:50:59 < Laurenceb_> 3.7mm?
2012-09-03T22:51:04 < TitanMKD> 3.7cm ;)
2012-09-03T22:51:07 < Laurenceb_> ah
2012-09-03T22:51:08 < BrainDamage> TitanMKD: aren't you the guy that's making the stm32f4 butt pirate?
2012-09-03T22:51:12 < TitanMKD> yes
2012-09-03T22:51:17 < Laurenceb_> lol butt pirate
2012-09-03T22:51:27 < TitanMKD> yes a hs bus pirate
2012-09-03T22:51:48 < Laurenceb_> i was making an stm32f4 SDR GPS dongle
2012-09-03T22:51:52 < Laurenceb_> but havent had time
2012-09-03T22:52:10 < Laurenceb_> to get further than running chibios on it
2012-09-03T22:52:50 < TitanMKD> anyway i plan to use chibios but maybe not really to use the scheduler/rtos behind
2012-09-03T22:53:06 < Laurenceb_> http://i.imgur.com/R1mre.png
2012-09-03T22:53:28 < TitanMKD> woo very little card
2012-09-03T22:53:40 < Laurenceb_> gps to usb :D
2012-09-03T22:53:42 < BrainDamage> I'd drop the usb connector and instead make the pcb itself have a small tab that acts as one
2012-09-03T22:53:53 < Laurenceb_> yeah
2012-09-03T22:54:04 < Laurenceb_> ive written some software gps in thumb2
2012-09-03T22:54:14 < TitanMKD> it seems it's just 1 layer
2012-09-03T22:54:15 < Laurenceb_> but i havent got time atm to develop it
2012-09-03T22:54:19 < Laurenceb_> 2
2012-09-03T22:54:21 < TitanMKD> ha ok
2012-09-03T22:54:26 < BrainDamage> you can see the vias
2012-09-03T22:54:27 < Laurenceb_> needs a good gnd plane for RF
2012-09-03T22:55:27 < TitanMKD> why do you use such a big STM32 instead of a LQFP64 ?
2012-09-03T22:55:31 < Laurenceb_> i made a few errors - usb is revered and the TCXO doesnt have enough swing to drive the stm
2012-09-03T22:55:42 < Laurenceb_> its is 64 pin
2012-09-03T22:55:45 < TitanMKD> oups it's already 64 ;)
2012-09-03T22:56:07 < Laurenceb_> i had to dead bug an sc70 not gate to get it to clk
2012-09-03T22:59:23 < Laurenceb_> http://i.imgur.com/MJjLi.png
2012-09-03T22:59:26 < Laurenceb_> F1 logger
2012-09-03T23:01:07 < TitanMKD> Laurenceb woo very dense board
2012-09-03T23:02:16 < Laurenceb_> not as dense as this
2012-09-03T23:02:32 < TitanMKD> it's a portable logger with LiPo ?
2012-09-03T23:02:41 < Laurenceb_> http://i.imgur.com/UubLM.jpg
2012-09-03T23:02:42 < Laurenceb_> yes
2012-09-03T23:03:07 < TitanMKD> woo the board on the bottom is "crazy"
2012-09-03T23:03:28 < Laurenceb_> F1 based autopilot
2012-09-03T23:03:32 < TitanMKD> it seems it's 6mil
2012-09-03T23:03:48 < TitanMKD> so any little PCB error is a major fail
2012-09-03T23:03:49 < Laurenceb_> yes
2012-09-03T23:03:59 < Laurenceb_> worked ok with seeedstudio
2012-09-03T23:04:06 < Laurenceb_> but i got 100% e test
2012-09-03T23:04:16 < TitanMKD> ha it is a good info
2012-09-03T23:04:30 < TitanMKD> i prefer itead it's cheaper and 100% etest is free
2012-09-03T23:04:37 < Laurenceb_> but seeedstudio vias only _just_  hit
2012-09-03T23:04:40 < TitanMKD> and IIRC it's same PCB company
2012-09-03T23:05:18 < Laurenceb_> yes
2012-09-03T23:05:26 < TitanMKD> Laurenceb and what does this board do ?
2012-09-03T23:05:28 < Laurenceb_> i used pcbtrain for work stuff
2012-09-03T23:05:34 < Laurenceb_> its an autopilot
2012-09-03T23:05:42 < Laurenceb_> imu+gps+radio
2012-09-03T23:05:54 < Laurenceb_> http://i.imgur.com/9cC04.jpg
2012-09-03T23:05:55 < TitanMKD> how many dof ?
2012-09-03T23:06:32 < zyp> over 9000
2012-09-03T23:06:42 < Laurenceb_> well i used 13 component state vector EKF from openpilot
2012-09-03T23:07:31 < TitanMKD> it's a 9 Degreees of Freedom ?
2012-09-03T23:07:37 < Laurenceb_> hmm....
2012-09-03T23:07:50 < Laurenceb_> i dont like defining things like that
2012-09-03T23:08:01 < Laurenceb_> it has accel+ magno+ gyro+ baro
2012-09-03T23:08:16 < zyp> Laurenceb_, for once I really agree with you
2012-09-03T23:08:32 < zyp> counting «dof» sounds so silly
2012-09-03T23:08:56 < TitanMKD> anyway with that you have an accurate X, Y, Z for autopilot and also maybe route ...
2012-09-03T23:09:15 < Laurenceb_> theres pitot and micro sd on the other side
2012-09-03T23:09:22 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-03T23:09:41 < Laurenceb_> but pitot is the only part that doesn twork :(
2012-09-03T23:09:50 < TitanMKD> Laurenceb and why not stm32f4 instead of the old stm32f1 ?
2012-09-03T23:09:58 < Laurenceb_> i designed this in 2009
2012-09-03T23:10:02 < TitanMKD> ha ok
2012-09-03T23:10:06 < Laurenceb_> http://i.imgur.com/4Tchw.jpg
2012-09-03T23:10:15 < Laurenceb_> was busy renovating my house
2012-09-03T23:10:19 < Laurenceb_> now im doing phd
2012-09-03T23:10:36 < Laurenceb_> pitot gets too much interference from the magno
2012-09-03T23:10:36 -!- Piele [~peter@5ED0EA16.cm-7-1d.dynamic.ziggo.nl] has quit [Quit: Leaving]
2012-09-03T23:10:55 < Laurenceb_> luckly honeywell have some new digital diff pressure sensors that are good enough to work :P
2012-09-03T23:12:07 < Laurenceb_> so yeah a new board with F4 and decent pitot is called for
2012-09-03T23:12:18 < Laurenceb_> but Dactyl 1.0 is good enough for some more dev work
2012-09-03T23:12:20 < TitanMKD> yes with F4 you have lot of power ;)
2012-09-03T23:12:24 < Laurenceb_> when i get time...
2012-09-03T23:12:40 < TitanMKD> Laurenceb in fact i think LPC43xx will be even better for that task with dual Cortex ;)
2012-09-03T23:13:01 < Laurenceb_> pita to program
2012-09-03T23:13:02 < TitanMKD> Laurenceb M0 can compute DOF stuff in background and M4 do other stuff more important or the reverse
2012-09-03T23:13:09 < BrainDamage> wait, how does the pitot pressure sensor gets interference _from_ the magnetometer?
2012-09-03T23:13:19 < BrainDamage> shouldn't it be the way around instead?
2012-09-03T23:13:34 < Laurenceb_> nope
2012-09-03T23:13:40 < Laurenceb_> honeywell magnos are crazy
2012-09-03T23:13:44 < TitanMKD> Laurenceb on all electronic for aeronautic we use external pitot probe
2012-09-03T23:13:49 < Laurenceb_> it draws about 3A peak or something
2012-09-03T23:13:54 < BrainDamage> o_O
2012-09-03T23:14:06 < Laurenceb_> through the 4.7uF external cap
2012-09-03T23:14:18 < Laurenceb_> that totally trashes a pitot reading each time it happens
2012-09-03T23:14:30 < Laurenceb_> due to inductance
2012-09-03T23:14:31 < BrainDamage> I can image with such ridicolous peak
2012-09-03T23:14:32 < TitanMKD> yes honeywell for pitot are just very bad
2012-09-03T23:14:45 < BrainDamage> perhaps consider separate psu for the magnetometer?
2012-09-03T23:14:50 < Laurenceb_> nope
2012-09-03T23:14:54 < Laurenceb_> that wont help
2012-09-03T23:15:02 < Laurenceb_> its inductance from the ocb traces
2012-09-03T23:15:05 < Laurenceb_> *pcb
2012-09-03T23:15:13 < Laurenceb_> honeywell already did the decoupling
2012-09-03T23:15:19 < BrainDamage> near field interference?
2012-09-03T23:15:25 < Laurenceb_> they did that well, but the inductive interference is the issue
2012-09-03T23:15:26 < Laurenceb_> yes
2012-09-03T23:15:41 < Laurenceb_> it only takes a few microvolts to wreck the pitot
2012-09-03T23:16:32 < Laurenceb_> 4 layer pcb might help a little
2012-09-03T23:16:48 < Laurenceb_> but honeywell now have digital sensors with everything on die
2012-09-03T23:16:59 < Laurenceb_> so ill use one of those on V2
2012-09-03T23:19:44 < Laurenceb_> pitot is tricky as the pressure is so low
2012-09-03T23:20:05 < Laurenceb_> what i was doing in 2008-10     http://i.imgur.com/gljQT.jpg
2012-09-03T23:21:00 < BrainDamage> looks nice for a bridge's underside
2012-09-03T23:21:01 < zyp> you were a woman? given that there is a kitchen in your picture?
2012-09-03T23:21:19 < BrainDamage> nah, bridge, trolls live under bridges
2012-09-03T23:21:39 < Laurenceb_> http://i.imgur.com/VgQRD.jpg
2012-09-03T23:21:45 < Laurenceb_> thats half way through work
2012-09-03T23:22:10 < BrainDamage> you're doing house carpentry all by yourself?
2012-09-03T23:22:25 < BrainDamage> I could see that being a huge time sink
2012-09-03T23:22:44 < Laurenceb_> yes
2012-09-03T23:22:58 < Laurenceb_> took my ~3 years full itme
2012-09-03T23:23:24 < Laurenceb_> it was 2 houses so i had to knock into one and take out rear wall
2012-09-03T23:23:35 < Laurenceb_> there is steel structure built in to take the load
2012-09-03T23:35:53 -!- DaKu is now known as daku
2012-09-03T23:36:42 -!- daku is now known as DaKu
2012-09-03T23:38:09 < TitanMKD> Laurenceb i'm redesigning a crazy board look http://imgur.com/yHHDZ
2012-09-03T23:38:22 < Laurenceb_> http://www.raspberrypi.org/phpBB3/viewforum.php?f=72&sid=e1cfee6afda147dcf6713e6e99c32da8
2012-09-03T23:38:23 < TitanMKD> especially the LBGA256 pins ;)
2012-09-03T23:38:25 < Laurenceb_> ^interesting
2012-09-03T23:38:45 < TitanMKD> on 2 layers ;) BGA escaping was a nightmare
2012-09-03T23:38:48 < Laurenceb_> nice
2012-09-03T23:38:55 < Laurenceb_> how are you going to solder it?
2012-09-03T23:39:08 < TitanMKD> i will try with my hot plate
2012-09-03T23:39:23 < TitanMKD> I have 3 LPC4350 ;)
2012-09-03T23:39:30 < TitanMKD> I hope i will have at least 1 working ;)
2012-09-03T23:39:49 < Laurenceb_> sounds fun
2012-09-03T23:39:53 < BrainDamage> careful with vias next to balls
2012-09-03T23:39:56 < TitanMKD> it's a challenge ;)
2012-09-03T23:39:57 < Laurenceb_> ive never tried
2012-09-03T23:39:59 < Laurenceb_> yeah
2012-09-03T23:40:05 < BrainDamage> you'll risk vias sucking the ball in
2012-09-03T23:40:08 < Laurenceb_> solder can cling to the edges
2012-09-03T23:40:19 < TitanMKD> BrainDamage yes there's some via near ball but i have no choice :(
2012-09-03T23:40:28 < BrainDamage> can you get tented vias?
2012-09-03T23:40:50 < Laurenceb_> zoom into my Dactyl board and you can see some issues with seeedstudio vias filling with solder paste
2012-09-03T23:40:55 < TitanMKD> but LBGA are 1mm spaced
2012-09-03T23:41:00 < Laurenceb_> s/paste/molten solder
2012-09-03T23:41:24 < TitanMKD> BrainDamage what's tented vias ?
2012-09-03T23:41:50 < BrainDamage> basically vias with a covering layer
2012-09-03T23:42:16 < Laurenceb_> you mean of solder stop?
2012-09-03T23:42:17 < BrainDamage> the solder mask covers the via hole, so you won't get the surface tension draw the ball into the pit
2012-09-03T23:42:26 < BrainDamage> yes
2012-09-03T23:42:26 < TitanMKD> I optimized the BGA using freeroute ;)
2012-09-03T23:42:35 < Laurenceb_> hmm
2012-09-03T23:42:41 < TitanMKD> he just removed about 10 via on the BGA ;)
2012-09-03T23:42:42 < BrainDamage> not all fabs can do it
2012-09-03T23:42:42 < Laurenceb_> i always do that
2012-09-03T23:42:53 < BrainDamage> and you have to ask explicitly for the via
2012-09-03T23:43:01 < Laurenceb_> oh
2012-09-03T23:43:05 < Laurenceb_> fills the via hole
2012-09-03T23:43:12 < TitanMKD> i will use Itead so i think they do not do that
2012-09-03T23:43:16 < Laurenceb_> yeah seeedstudio and pcbtrain dont do that
2012-09-03T23:43:36 < TitanMKD> anyway for 5cm*5cm and 9.9USD if that do not work it's not a huge problem ;)
2012-09-03T23:43:37 < BrainDamage> you could try clogging the via yourself
2012-09-03T23:43:38 < Laurenceb_> and their solder stop tends to leave a rim of exposed copper around the top of the via
2012-09-03T23:43:45 < Laurenceb_> that attracts solder
2012-09-03T23:44:18 < Laurenceb_> ive had issues with that causing shorts, even with a solder mask used
2012-09-03T23:44:26 < TitanMKD> maybe EING help to solder the LBGA
2012-09-03T23:44:40 < zyp> http://bin.jvnv.net/f/g17eb.JPG <- oshpark removes silk over via holes even though they are tented
2012-09-03T23:45:08 < Laurenceb_> i doubt they remove it
2012-09-03T23:45:14 < Laurenceb_> aiui its a liquid coating
2012-09-03T23:45:20 < zyp> I mean in software
2012-09-03T23:45:25 < Laurenceb_> oh
2012-09-03T23:45:33 < Laurenceb_> looks better than pcbtrain and seeed
2012-09-03T23:45:45 < zyp> look at the drawing, look how the lines are broken with rounded ends
2012-09-03T23:45:52 < Laurenceb_> you end up with thickes mask around the holes
2012-09-03T23:46:08 < TitanMKD> I was thinking to also add ULPI for USB1 ;)
2012-09-03T23:46:16 < TitanMKD> but USB0 is already 480mbps ;)
2012-09-03T23:46:18 < Laurenceb_> *thicker
2012-09-03T23:46:28 < Laurenceb_> seeed and pcbtrain is the opposite
2012-09-03T23:47:30 < TitanMKD> anyway for the price seeed/itead is really good
2012-09-03T23:47:32 < zyp> IMO oshpark is worth the price difference over seeed
2012-09-03T23:47:38 < TitanMKD> here same PCB cost just 10 times more
2012-09-03T23:48:14 < Laurenceb_> where are you?
2012-09-03T23:48:23 < zyp> the board in my pic is four layer, $45 for three shipped
2012-09-03T23:48:25 < TitanMKD> i live in France
2012-09-03T23:48:39 < TitanMKD> and PCB price are just crazy like 150euros for 3 boards 10cm*10cm ;)
2012-09-03T23:49:09 < Laurenceb_> sounds like pcbtrain
2012-09-03T23:49:42 < TitanMKD> else for cheap stencil i use http://www.smtstencil.co.uk/
2012-09-03T23:50:09 < TitanMKD> very good price for good quality
2012-09-03T23:50:21 < Laurenceb_> yes
2012-09-03T23:50:37 < Laurenceb_> shrdlu- or something on ##electronics
2012-09-03T23:50:41 < Laurenceb_> runs it
2012-09-03T23:50:51 < zyp> I've used http://www.ohararp.com/Stencils.html a couple of times
2012-09-03T23:51:48 < zyp> which may work out cheaper if you have several designs
2012-09-03T23:51:52 < Laurenceb_> bbl
2012-09-03T23:52:41 < TitanMKD> Laurenceb it is your GPS data logger http://www.ohararp.com/products.html ?
2012-09-03T23:54:23 < BrainDamage> I think all the discussions I had with shrdlu- would make me a giant hypocrite if I ordered him a stencil >_>
2012-09-03T23:55:52 < Laurenceb_> TitanMKD: why on earth would it be mine?!
2012-09-03T23:56:06 < Laurenceb_> BrainDamage: lol how come?
2012-09-03T23:56:10 < TitanMKD> Laurenceb because you design stuff like that ;)
2012-09-03T23:56:37 < Laurenceb_> heh
2012-09-03T23:56:48 < BrainDamage> Laurenceb_: shrdlu- has a very personal view on how physics should be and what is physical or not
2012-09-03T23:57:34 < BrainDamage> according to him "everything should be particles with spin", and with particles he means newtonian mechanics particles which are spinning
2012-09-03T23:59:13 < BrainDamage> oh, or that a charged particle, emits charge, not an E field
2012-09-03T23:59:37 < Laurenceb_> hehe
--- Day changed Tue Sep 04 2012
2012-09-04T00:00:11 < Laurenceb_> not as crazy as jean louis naudin
2012-09-04T00:02:06 < BrainDamage> holy shit that some serious crackpottery
2012-09-04T00:02:51 < BrainDamage> this pic is priceless http://jnaudin.free.fr/uav/hid/HIDflight.jpg
2012-09-04T00:04:56 < Laurenceb_> apparently he has reverse engineered steorn orbo
2012-09-04T00:05:59 < Laurenceb_> http://jnaudin.free.fr/steorn/indexen.htm
2012-09-04T00:06:40 < BrainDamage> I like the unnecessary umlauts on the i in toroidal
2012-09-04T00:07:21 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 244 seconds]
2012-09-04T00:07:51 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-04T00:10:53 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-04T00:15:35 < Laurenceb_> http://jnaudin.free.fr/steorn/html/steffecten.htm
2012-09-04T00:15:39 < Laurenceb_> ... and?!
2012-09-04T00:15:51 < Laurenceb_> how the hell does that give free energy
2012-09-04T00:20:45 < [7]> it clearly doesn't :)
2012-09-04T00:21:03 < [7]> I'd be more curious what would happen if that could would be connected for more than a second :P
2012-09-04T00:21:19 < Laurenceb_> its pretty basic stuff
2012-09-04T00:21:20 < [7]> should produce a good amount of smope *grin*
2012-09-04T00:21:28 < [7]> smoke*
2012-09-04T00:21:47 < Laurenceb_> the energy should all balance, you just need to think about whats happening to the dipoles
2012-09-04T00:21:50 < [7]> so you will have free smoke at least :P
2012-09-04T00:22:09 < Laurenceb_> heh
2012-09-04T00:22:44 < BrainDamage> Laurenceb_: obviously they neglect the energy conservation principle despite mountains of data points supporting it, otherwise they wouldn't even start
2012-09-04T00:23:08 < BrainDamage> I mean working on the subject
2012-09-04T00:24:00 < BrainDamage> and I guess thermodynamics is black magic too
2012-09-04T00:24:04 < [7]> only thing I could imagine to give "free energy" would be to find a way that converts heat (not temperature difference) to something else
2012-09-04T00:24:39 < BrainDamage> thermodynamic engines do convert heat to other types of work
2012-09-04T00:24:43 < [7]> basically picking up the IR radiation and converting it to electricity or whatever
2012-09-04T00:24:56 < zyp> say hello to entropy
2012-09-04T00:24:57 < BrainDamage> and it's still no free energy
2012-09-04T00:25:08 < [7]> BrainDamage: those convert a temperature difference to other types of work, not absolute heat
2012-09-04T00:25:14 < Laurenceb_> enthalpy vrs entropy is odd stuff
2012-09-04T00:25:26 < BrainDamage> absolute heat?
2012-09-04T00:25:41 < BrainDamage> you know that there's thermodynamic equilibrium even for EM heat?
2012-09-04T00:25:43 < [7]> as in "free" IR radiation, without requiring a cooled side
2012-09-04T00:25:48 < Laurenceb_> they are both measured in joules
2012-09-04T00:25:50 < Laurenceb_> hehe
2012-09-04T00:26:11 < Laurenceb_> i have a "free" energy idea that has not yet been disproved
2012-09-04T00:26:21 < Laurenceb_> it violates second law of thermodynamics
2012-09-04T00:26:21 < BrainDamage> hacking your power meter?
2012-09-04T00:26:24 < Laurenceb_> lol
2012-09-04T00:26:34 < Laurenceb_> thats bash.org worthy
2012-09-04T00:26:35 < [7]> Laurenceb_: exactly what I meant
2012-09-04T00:26:43 < Laurenceb_> ok anyway
2012-09-04T00:26:52 < Laurenceb_> so say you have a cantilever
2012-09-04T00:27:05 < Laurenceb_> if its small its going to oscillate with thermal energy
2012-09-04T00:27:11 < Laurenceb_> follow so far?
2012-09-04T00:27:35 < BrainDamage> you you're saying phonons have a prevalent direction
2012-09-04T00:27:39 < Laurenceb_> no
2012-09-04T00:28:00 < BrainDamage> how not, it's oscillating :p
2012-09-04T00:28:09 < Laurenceb_> due to its construction it will oscillate with greater amplitude along one axis
2012-09-04T00:28:32 < [7]> what I don't get is what exactly happens when mixing hot and cold water vs. pouring it over a peltier element (seebeck effect)
2012-09-04T00:28:54 < Laurenceb_> ok one thing at a time
2012-09-04T00:28:56 < Laurenceb_> ill finish
2012-09-04T00:28:57 < [7]> the final average temperature of the peltier/seebeck case just has to be lower than in the pure mixing case
2012-09-04T00:29:10 < Laurenceb_> hmm intriguing
2012-09-04T00:29:13 < [7]> but which one meets in the middle and which one doesn't?
2012-09-04T00:29:33 < Laurenceb_> so anyway
2012-09-04T00:29:42 < Laurenceb_> if the end of the cantilever is conductive metal
2012-09-04T00:29:53 < Laurenceb_> ans its placed in a vertical E field
2012-09-04T00:29:56 < Laurenceb_> *and
2012-09-04T00:30:11 < Laurenceb_> then if you have a small contact at top and bottom of travel
2012-09-04T00:30:20 < Laurenceb_> a current will flow between the two
2012-09-04T00:30:26 < Laurenceb_> even if the motion is random
2012-09-04T00:30:34 < Laurenceb_> -think about it :P
2012-09-04T00:31:17 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-04T00:31:56 < BrainDamage> http://www.twiddla.com/924180
2012-09-04T00:31:59 < BrainDamage> something like this?
2012-09-04T00:32:20 < Laurenceb_> yes
2012-09-04T00:32:38 < Laurenceb_> heh nice site
2012-09-04T00:32:55 < [7]> that looks like the oscillating of the cantilever would discharge the outer capacitor
2012-09-04T00:33:06 < Laurenceb_> yes
2012-09-04T00:33:14 < Laurenceb_> its a little more complex
2012-09-04T00:33:59 < Laurenceb_> but that has to not sheild the E field
2012-09-04T00:34:24 < BrainDamage> the instant the tip touches a plate and gets a charge, it'll get a force towards the other plate and transfer it, but I'm wondering the oscillations
2012-09-04T00:34:33 < BrainDamage> if the tip moves in the E field it gets eddy currents induced
2012-09-04T00:34:39 < Laurenceb_> sure
2012-09-04T00:34:43 < Laurenceb_> so it cools slightly
2012-09-04T00:34:48 < Laurenceb_> and pumps a little current
2012-09-04T00:34:54 < Laurenceb_> ... maybe
2012-09-04T00:35:06 < BrainDamage> minus the maybe, ramo's theorem
2012-09-04T00:35:18 < [7]> what would cool it?
2012-09-04T00:35:27 < Laurenceb_> its like maxwells demon
2012-09-04T00:35:33 < Laurenceb_> but no rectification stage
2012-09-04T00:36:35 < [7]> aaaanyway, I'm wrestling with this I2C core again. it just ignores any writes to its registers, except for the reset bit. smells clock gate-ish, but I can't seem to find the bug
2012-09-04T00:37:31 < [7]> RCC->APB1ENR |= RCC_APB1ENR_I2C1EN; should enable it, I'd think
2012-09-04T00:38:27 < zyp> except if you are using I2C2
2012-09-04T00:38:28 < Laurenceb_> what is ramos theorem?
2012-09-04T00:38:32 < BrainDamage> Laurenceb_ https://en.wikipedia.org/wiki/Shockley-Ramo_theorem
2012-09-04T00:38:32 < Laurenceb_> lol
2012-09-04T00:40:09 < Laurenceb_> [7]: i use periph lib the enable the i2c
2012-09-04T00:40:09 < BrainDamage> if we plug directly the value that <v>=0 due to the brownian motion nature, <I>=0
2012-09-04T00:40:29 < zyp> [7], how are you testing it? tried manipulating the registers in the debugger?
2012-09-04T00:40:31 < Laurenceb_> huh
2012-09-04T00:40:37 < BrainDamage> but ofc that doesn't say that it's not doing work,
2012-09-04T00:40:51 < Laurenceb_> the cantilever is moving
2012-09-04T00:40:59 < [7]> zyp: first tried to let my code init it, saw that it locked up at some point, and checked the register values, they weren't initialized
2012-09-04T00:40:59 < Laurenceb_> randomly between the two contacts
2012-09-04T00:41:09 < [7]> then tried accessing them directly through SWD
2012-09-04T00:41:09 < Laurenceb_> anyway im tired
2012-09-04T00:41:20 < Laurenceb_> i cant be bothered to argue about it
2012-09-04T00:41:25 < BrainDamage> k, np
2012-09-04T00:41:32 < [7]> if I do the latter, I can successfully flip the reset bit in CR1, but all other bits reset to zero
2012-09-04T00:41:34 < Laurenceb_> a scale model and multimeter would test the basic concept
2012-09-04T00:41:35 < BrainDamage> gimme a shout when/if you feel like it
2012-09-04T00:41:37 < [7]> and TRISE is stuck at 2
2012-09-04T00:41:39 < Laurenceb_> randomly wobble it
2012-09-04T00:41:39 < zyp> [7], are you inlining stuff? I found that I2C peripheral needs a few cycles after enable before it's ready to accept writes
2012-09-04T00:41:59 < [7]> well it doesn't even accept the enable bit in the first place
2012-09-04T00:42:16 < zyp> I mean enabling clock gate in RCC
2012-09-04T00:42:38 < zyp> also, did you clear the reset bit after setting it?
2012-09-04T00:42:41 < [7]> well that should be at least tens if not hundreds of clocks away
2012-09-04T00:42:50 < zyp> then it should be fine
2012-09-04T00:43:13 < [7]> I originally tried to clear reset and set enable in the same write
2012-09-04T00:43:25 < zyp> my code got inlined and reordered to the point where I2Cx_CR1 write happened directly after APB1ENR write :p
2012-09-04T00:43:29 < [7]> but it seems to generally not accept changes to any non-reset bits even in the debugger
2012-09-04T00:43:47 < [7]> I'm currently dealing with a -O0 debug build
2012-09-04T00:43:52 < zyp> ah
2012-09-04T00:43:54 < [7]> so it won't do that kind of stuff
2012-09-04T00:44:02 < zyp> I'm always working with -O2
2012-09-04T00:47:14 < [7]> what does that peripheral enable bit actually mean?
2012-09-04T00:47:30 < [7]> I would have thought it's supposed to be 0 while configuring clock dividers etc.
2012-09-04T00:47:42 < [7]> but apparently it completely blocks other register accesses?
2012-09-04T00:48:30 < [7]> the stdperiph lib seems to clear everything except for reset (and doesn't ever touch that bit), then configure everything, then enable the peripheral
2012-09-04T00:48:37 < [7]> that sequence doesn't seem to work for me at all
2012-09-04T00:49:36 < zyp> that's what my code also does, works for me
2012-09-04T00:49:56 < zyp> except that I'm doing a reset first
2012-09-04T00:50:03 < zyp> http://cgit.jvnv.net/laks/tree/i2c/i2c.cpp#n93
2012-09-04T00:50:36 < zyp> rcc enable, set reset, clear reset, configure clock dividers, set enable
2012-09-04T00:52:35 < [7]> hm, that seems to work now
2012-09-04T00:52:40 < [7]> but I'm not getting any IRQs :/
2012-09-04T00:53:19 < zyp> have you enabled the IRQs in the NVIC?
2012-09-04T00:54:51 -!- izua_ [~izua@unaffiliated/izua] has quit [Ping timeout: 276 seconds]
2012-09-04T00:55:03 < [7]> yes
2012-09-04T00:55:27 < zyp> have you enabled the IRQs in CR2?
2012-09-04T00:55:33 < [7]> yes
2012-09-04T00:55:38 < [7]> CR/SR values suggest that the start condition hasn't yet been sent, the bus is busy, and that it's still in slave mode
2012-09-04T00:55:49 < zyp> ah, right
2012-09-04T00:55:52  * [7] checks signal levels
2012-09-04T00:55:58 < zyp> you'll find that one is low
2012-09-04T00:58:01 < zyp> either that or you haven't configured the pins correctly so they just looks low to the core
2012-09-04T01:00:00 -!- DaKu is now known as daku
2012-09-04T01:00:05 -!- daku is now known as DaKu
2012-09-04T01:00:59 -!- DaKu is now known as daku
2012-09-04T01:01:04 -!- daku is now known as DaKu
2012-09-04T01:09:10 < Laurenceb_> dDafuk
2012-09-04T01:14:48  * [7] grabs his scope
2012-09-04T01:21:22 < [7]> a device is supposed to ack its address, eh?
2012-09-04T01:21:46 < zyp> yes
2012-09-04T01:22:46 < [7]> this bugger doesn't >.<
2012-09-04T01:23:08 < [7]> I'm trying to talk to the cs43l22 on the stm32f4discovery board
2012-09-04T01:23:39 < [7]> according to its datasheet the address should be 1001010
2012-09-04T01:23:55 < [7]> which is what I'm also seeing on my scope. but it doens't ack...
2012-09-04T01:26:48 < zyp> did you remember to shift it up one bit and add the rw bit?
2012-09-04T01:28:40 < zyp> bedtime, good luck
2012-09-04T01:29:26 < [7]> http://theseven.bounceme.net/~theseven/tmp/i2cnak.BMP
2012-09-04T01:31:36 < zyp> hmm, looks write
2012-09-04T01:31:55 < zyp> 1001010 W NAK
2012-09-04T01:34:16 < zyp> have you set PD4 high?
2012-09-04T01:34:21 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-04T01:34:44 < zyp> from CS43L22 datasheet: «The device will remain in the Power-Down state until the RESET pin is brought high. The control port is accessible once RESET is high»
2012-09-04T01:34:56 < zyp> RESET is connected to PD4
2012-09-04T01:37:32 < Laurenceb_> http://mayaposch.wordpress.com/2011/11/01/how-to-really-truly-use-qthreads-the-full-explanation/
2012-09-04T01:37:42 < Laurenceb_> a useful article from hackitten :P
2012-09-04T01:37:43 < [7]> zyp: oh, thanks for that hint :)
2012-09-04T01:38:06 < zyp> I just came across it when I went to double check the addr ;)
2012-09-04T01:38:27 < zyp> anyway, I was heading to bed, have a good night everyone
2012-09-04T01:38:31 < BrainDamage> gn
2012-09-04T01:42:36 < [7]> hm, do I have to set the i2c pins to open drain manually or does the i2c mode/af select already take care of that?
2012-09-04T01:42:52 < [7]> now the waveform looks like the cirrus chip doens't manage to pull it down
2012-09-04T01:42:59 < [7]> even though I've set it to opendrain now
2012-09-04T01:44:24 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-04T01:45:42 < [7]> hm, looks like there is a very strong pullup somewhere...
2012-09-04T01:51:09 < [7]> aha. bug in the gpio init code :/
2012-09-04T02:04:24 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-04T02:06:13 < dongs> sup trolls
2012-09-04T02:06:43 < dongs>  [7]> zyp: oh, thanks for that hint :)
2012-09-04T02:07:14 < [7]> hm?
2012-09-04T02:07:45 < dongs> that shoulda been obvious, no reply - its in reset :)
2012-09-04T02:07:58 < dongs> wait,a re you gonna get rockbox working on f4discovery
2012-09-04T02:08:01 < dongs> with something like SPI lcd?
2012-09-04T02:08:04 < dongs> that would be hilarious :)
2012-09-04T02:08:58 < [7]> nah
2012-09-04T02:09:33 < [7]> I'm just using this board to get the code somewhat working before I have a prototype of the final board
2012-09-04T02:09:51 < [7]> which is why I didn't bother to read the CS43L22 datasheet at all
2012-09-04T02:09:57 < [7]> except for looking up the i2c address
2012-09-04T02:10:16 < [7]> I want to talk to completely different devices later, just using it to test the driver
2012-09-04T02:10:38 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-04T02:11:22 < dongs> ah
2012-09-04T02:11:50 < [7]> this board doesn't have enough RAM for rockbox btw
2012-09-04T02:12:03 < [7]> 2MB is absolute minimum, 8MB for all features IIRC
2012-09-04T02:12:37 < [7]> and I don't think anyone has ported rockbox to a thumb MCU so far
2012-09-04T02:18:31 < emeb> Would be cool to hook an F40x up to a fat PSRAM via FSMC.
2012-09-04T02:25:19 < dongs> heh
2012-09-04T02:45:38 < [7]> yeah, RAM is the only thing this device is truly lacking
2012-09-04T02:59:48 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 248 seconds]
2012-09-04T03:02:07 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-04T03:02:59 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 244 seconds]
2012-09-04T03:03:35 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-04T03:06:46 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-04T03:27:04 < emeb> too bad none of the ST processors have an SDRAM controller on the FSMC.
2012-09-04T03:27:51 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 252 seconds]
2012-09-04T03:28:32 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-04T03:42:31 < zippe> emeb: I'm actually kinda glad
2012-09-04T03:42:37 < zippe> Since it would be a power monster
2012-09-04T03:44:56 < emeb> eh - you don't have to use it.
2012-09-04T03:48:40 < zippe> And an area monster
2012-09-04T03:48:49 < zippe> And it would complicate a huge number of pad drivers
2012-09-04T03:49:31 < emeb> hmmm
2012-09-04T03:49:44 < emeb> SDRAM - not DDR/DDR2/etc.
2012-09-04T03:52:34 < emeb> anyway - as is you need PSRAM if you want large external volatile memory. trouble with that is it's more $$ and most parts are only BGA.
2012-09-04T04:06:31 < dongs> how is that a problem
2012-09-04T04:09:12 < qyx_> not a problem but a complication
2012-09-04T04:09:47 < qyx_> i would rather grab one from these cheap freescale i.mx parts & ddr, both available in non-bga
2012-09-04T04:10:19 < emeb> build yourself a chumby!
2012-09-04T04:11:23 < qyx_> imx233 is in 128pin pqfp, 32/64MB ddr in tsop2 54
2012-09-04T04:11:28 < qyx_> or whatever it is
2012-09-04T04:12:24 < qyx_> yes, chumby is the same hw
2012-09-04T04:13:41 < qyx_> anyway, gn
2012-09-04T04:13:46 < emeb> l8r
2012-09-04T04:13:52 < dongs> imx233 is horribly slow
2012-09-04T04:19:07 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-04T04:39:41 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-04T04:42:54 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-04T05:03:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-04T05:05:34 < emeb_mac> Hmmm - STM32F3 Discovery shows up in the Digi-Key database. http://www.digikey.com/product-detail/en/STM32F3DISCOVERY/497-13192-ND/3522185
2012-09-04T05:05:49 < emeb_mac> $10.90, none in stock yet.
2012-09-04T05:05:50 < karlp> been in their shop for a few weeks now
2012-09-04T05:05:58 < karlp> but yeah, no stock yet :)
2012-09-04T05:06:04 -!- tech2077_ [~tech2077@76.247.251.117] has quit [Ping timeout: 246 seconds]
2012-09-04T05:06:07 < emeb_mac> nice price tho
2012-09-04T05:06:17 < karlp> just like all the discovery boards
2012-09-04T05:06:39 < emeb_mac> discovery boards are probably one of the smartest things ST does in terms of marketing.
2012-09-04T05:07:42 < emeb_mac> TI Launchpads come close to that in price/utility, if only the parts had more bang/buck.
2012-09-04T05:09:08 < emeb_mac> interesting that Mouser's price for the F3 discovery is $19.
2012-09-04T05:18:04 -!- tech2077_ [~tech2077@adsl-75-53-139-200.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-04T05:20:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-04T05:28:54 -!- tech2077_ [~tech2077@adsl-75-53-139-200.dsl.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-09-04T05:42:07 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-09-04T05:42:49 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 246 seconds]
2012-09-04T05:44:03 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-04T05:47:01 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-04T05:49:09 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-04T05:58:34 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-09-04T06:31:50 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-09-04T06:33:46 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-04T06:38:24 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Quit: leaving]
2012-09-04T06:40:37 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-04T07:08:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-04T07:15:48 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 248 seconds]
2012-09-04T07:56:16 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-04T07:56:27 -!- claude is now known as Claude
2012-09-04T08:19:51 < dongs> The Propeller chip is a full-custom, Parallax design in entirety, unencumbered from any third-party intellectual property obligations. Parallax endeavors to supply Propellers as long as we have demand (i.e., foreseeable future, a minimum 20 to 25 years). As an example of our commitment, Parallax still manufactures the BASIC Stamp I and it was designed nearly 20 years ago. For these reasons, customers should have no concern about the long-term supply
2012-09-04T08:25:48 < emeb_mac> yay
2012-09-04T08:25:55 < emeb_mac> old stuff available forever
2012-09-04T08:27:53 < emeb_mac> of course, you can still buy 1970s vintage TTL, CMOS, 80xx, 68xx, 65xx parts from Jameco.
2012-09-04T08:28:07 < dongs> old/shitty stuff more liek
2012-09-04T08:29:00 < dongs> who teh fuck wants / needs propeller chips
2012-09-04T08:32:44 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-09-04T08:32:45 < emeb_mac> damn straight. "Look dudes - we've got 8 processors and no peripherals except gpio"
2012-09-04T08:32:58 < dongs> lol
2012-09-04T08:33:15 < emeb_mac> "You want SPI or I2C - use a processor & bitbang away"
2012-09-04T08:33:15 < dongs> ya thats the excuse i usually use when trolling propfags
2012-09-04T08:33:23 < dongs> and dont forget their "spin" language
2012-09-04T08:34:20 < emeb_mac> yeah. I want to learn a whole new niche language.
2012-09-04T08:34:40 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-09-04T08:35:36 < emeb_mac> You could do an awesome demoscene with a STM32 if you wanted.
2012-09-04T08:35:59 < dongs> no you arent
2012-09-04T08:36:09 < dongs> you need to finish your retarded F0 eeprom reader shit first
2012-09-04T08:36:14 < dongs> which you havent even started on
2012-09-04T08:36:23 < emeb_mac> Just look at the crazy stuff that rossum does w/ AVRs.
2012-09-04T08:36:58 < emeb_mac> demoscene has nothing to do with the power of the MCU - it's all about imagination.
2012-09-04T08:37:41 < emeb_mac> word
2012-09-04T08:37:53 < dongs> i heard you like polling, so i made a chip with 8 cores so you can poll while you poll.
2012-09-04T08:38:08 < dongs> anyway prop is worthless
2012-09-04T08:38:11 < emeb_mac> lol
2012-09-04T08:38:32 < emeb_mac> xhibit - propeller edition
2012-09-04T08:39:00 < emeb_mac> gah! it's the end times.
2012-09-04T08:39:28 < dongs> no, you got it all wrong
2012-09-04T08:39:30 < dongs> more like spin.gif
2012-09-04T08:39:39 < dongs> pbx.mine.nu/artwork/spin.gif
2012-09-04T08:39:43 < dongs> flyback: would probably enjoy it
2012-09-04T08:40:16 < emeb_mac> it's best
2012-09-04T08:40:34 < emeb_mac> (not to)
2012-09-04T08:41:00 < emeb_mac> not exactly goatse, but it'll leave a mark.
2012-09-04T08:42:24 < emeb_mac> so, speaking of strange processors - anyone know if Chuck Moore is still working on that multiprocessing forth chip?
2012-09-04T08:42:57 < emeb_mac> doing the chip design entirely with his own CAD tools built with older forth chips?
2012-09-04T08:48:24 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-04T08:48:35 -!- soafee-chan [~spaec@unaffiliated/traumapony] has quit [Excess Flood]
2012-09-04T08:49:14 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-04T08:49:25 -!- soafee-chan [~spaec@unaffiliated/traumapony] has quit [Excess Flood]
2012-09-04T08:50:42 -!- spacekitteh [~spaec@unaffiliated/traumapony] has quit [Ping timeout: 244 seconds]
2012-09-04T08:51:02 -!- soafee-chan [~spaec@unaffiliated/traumapony] has joined ##stm32
2012-09-04T08:51:13 -!- soafee-chan [~spaec@unaffiliated/traumapony] has quit [Excess Flood]
2012-09-04T09:15:25 -!- pelrun [~pelrun@203-206-187-234.perm.iinet.net.au] has joined ##stm32
2012-09-04T09:16:30  * pelrun is currently having to work with keil
2012-09-04T09:16:42 < pelrun> god I can't wait to get back to a sane toolchain
2012-09-04T09:17:04 < pelrun> this acts like it hasn't been updated since 1985
2012-09-04T09:17:47 < dongs> ur so funny
2012-09-04T09:17:49 < dongs> more like
2012-09-04T09:17:53 < dongs> you got used to writing sloppy shit code
2012-09-04T09:17:56 < dongs> and now youre suffering from it
2012-09-04T09:18:18 < dongs> what does keil do that 'hasnt been updated since 1985'?
2012-09-04T09:18:24 < dongs> not allowing to declare vars mid-scope?
2012-09-04T09:18:51 < dongs> CHECK, NOT ANSI-C COMPLIANT BEHAVIOR
2012-09-04T09:18:59 < dongs> not allowing named struct initializers?
2012-09-04T09:19:01 < dongs> CHECK, SAME REASON
2012-09-04T09:19:03 < pelrun> uh
2012-09-04T09:19:11 < pelrun> EVERYTHING IS IN ALL CAPS
2012-09-04T09:19:15 < dongs> ?!
2012-09-04T09:19:16 < dongs> wuATU
2012-09-04T09:19:23 < pelrun> (this is 8051 btw)
2012-09-04T09:19:29 < dongs> oh alw d.
2012-09-04T09:19:33 < pelrun> ridiculous command line arguments
2012-09-04T09:19:43 < dongs> i dunno dude
2012-09-04T09:19:45 < dongs> im doing 8051 in keil
2012-09-04T09:19:47 < dongs> and none of hte shit is in ca ps.
2012-09-04T09:19:49 < pelrun> two different compilers and linkers, and the newer linker actually breaks code
2012-09-04T09:19:52 < dongs> and i dont need to do any command lines...
2012-09-04T09:20:04 < dongs> what shit are you using ? are we talkign about same toolchain?
2012-09-04T09:20:10 < dongs> i use c51
2012-09-04T09:20:11 < dongs> it works fine
2012-09-04T09:20:17 < pelrun> yeah, and bl51
2012-09-04T09:20:23 < dongs> yep
2012-09-04T09:20:28 < pelrun> fine unless you want it to actually remove unused functions
2012-09-04T09:20:34 < pelrun> then you need to use lx51
2012-09-04T09:20:46 < pelrun> and that has completely different semantics
2012-09-04T09:21:11 < dongs> *** WARNING L16: UNCALLED SEGMENT, IGNORED FOR OVERLAY PROCESS SEGMENT: ?PR?_TTY_BYTE?FW
2012-09-04T09:21:15 < dongs> works here.
2012-09-04T09:21:25 < dongs> you ARE using the IDE right?
2012-09-04T09:21:32 < dongs> instead of retarding around with mile-long command lines?
2012-09-04T09:21:35 < pelrun> can't use uVision for this project
2012-09-04T09:21:40 < dongs> why not?
2012-09-04T09:21:41 < pelrun> have to use the silabs ide
2012-09-04T09:21:49 < pelrun> pre-existing project that I'm unfucking
2012-09-04T09:21:54 < dongs> haha, well shit.
2012-09-04T09:21:56 < dongs> move it to uvision first
2012-09-04T09:22:00 < dongs> then deal wiht the fail
2012-09-04T09:22:06 < pelrun> and what were you saying about no all caps?
2012-09-04T09:22:12 < dongs> :p
2012-09-04T09:23:59 < pelrun> also can't define a located variable *and* initialise it
2012-09-04T09:24:42 < pelrun> have to drop back to asm to define the initial values
2012-09-04T09:24:53 < pelrun> and then it complains about overlapping memory
2012-09-04T09:24:57 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-09-04T09:25:14 < dongs> wut, that sounds like  your shit is broken dude.
2012-09-04T09:25:34  * pelrun doesn't like it when he's forced to accept/ignore warnings because there's no valid way to resolve them
2012-09-04T09:25:40 < pelrun> nope
2012-09-04T09:25:54 < dongs> int nig = 0; works here.
2012-09-04T09:26:14 < pelrun> uh, located variables, i.e. ones that have to be put in a specific memory location
2012-09-04T09:26:23 < pelrun> "at" keyword
2012-09-04T09:26:25 < dongs> oh
2012-09-04T09:26:34 < dongs> dunno only used those to refer to SFRs
2012-09-04T09:26:43 < dongs> i dont think any of my shit needed to be in specific places
2012-09-04T09:26:57 < pelrun> these are flash configuration variables
2012-09-04T09:27:02 < pelrun> for the application
2012-09-04T09:27:25 < dongs> oh well, stop dicking with command lines, get hte project building in uvision then you can start complainig :)
2012-09-04T09:27:39 < pelrun> pfft, the commandlines are shit but they are handled
2012-09-04T09:27:58 < pelrun> it's like a cp/m app
2012-09-04T09:28:05 < dongs> heh
2012-09-04T09:28:49 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-04T09:43:48 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 252 seconds]
2012-09-04T09:45:37 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-04T09:57:48 -!- dekar_ [~dekar@212.255.36.77] has joined ##stm32
2012-09-04T09:57:48 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-04T10:01:26 -!- dekar__ [~dekar@212.255.40.250] has quit [Ping timeout: 246 seconds]
2012-09-04T10:09:04 -!- Claude is now known as claude
2012-09-04T10:14:10 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 268 seconds]
2012-09-04T10:18:46 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-04T10:20:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-04T10:25:01 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-04T10:25:02 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-04T10:25:18 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-04T10:31:26 -!- claude is now known as claude_work
2012-09-04T10:36:55 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-04T11:00:05 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-04T11:00:05 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-04T11:00:05 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-04T11:33:47 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-04T12:07:41 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-04T12:08:41 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-09-04T12:10:00 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-04T12:28:27 -!- pelrun [~pelrun@203-206-187-234.perm.iinet.net.au] has quit [Quit: Leaving]
2012-09-04T12:56:59 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-04T12:57:00 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-04T12:57:00 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-04T12:57:03 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-04T13:42:48 < dongs> ST site is fucking useless.
2012-09-04T13:43:50 < dongs> The L3GD20 high-performance 3-axis digital-output gyroscope has a programmable full scale of ±250/±500/ ±2000 dps. With a resonant frequency above the audio bandwidth, it operates with greater accuracy and reliability, because it is completely immune to both mechanical vibrations and audio noise.
2012-09-04T13:44:03 < dongs> o'rly
2012-09-04T13:52:07 < zyp> whoa, that's awesome
2012-09-04T14:00:08 < Laurenceb> http://www.theregister.co.uk/2012/09/04/oracle_sparc_t5_processor/
2012-09-04T14:00:12 < Laurenceb> omfg 16 cores
2012-09-04T14:02:51 < dongs> http://www.bluelight-tech.com/BL-3GMod.htm also got email from tehse guys
2012-09-04T14:04:02 < karlp> emeb greenarrays stills seems to be in business, no idea how many customers though
2012-09-04T14:06:46 < Laurenceb> they took a decent gyro then put it on a retardo pcb
2012-09-04T14:19:55 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-04T14:20:27 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-04T14:21:35 < tavish> hi, anyone know if the CDC class has some kind of bandwidth limitation (if the endpoints are used for bulk transfers with libusb)
2012-09-04T14:23:15 < jpa-> just what usb 1.1 has
2012-09-04T14:23:54 < Laurenceb> lol @ raspberry pi bare metal
2012-09-04T14:24:09 < Laurenceb> thats some seriously overcomplex boot stuff
2012-09-04T14:25:45 < karlp> yeah, that sort of shit makes very little sensse to me.
2012-09-04T14:26:03 < karlp> plenty of other socs that would welcome more people fixing/creating the baremetal portions
2012-09-04T14:26:22 < tavish> jpa-: i am doing synchronous bulk tranfers with 64byte packets and i am getting _exactly_ 32000B/s. with 32byte packets, i get 16000B/s.
2012-09-04T14:26:47 < Laurenceb> https://www.olimex.com/dev/msp-ccrf.html
2012-09-04T14:26:51 < Laurenceb> ^nice
2012-09-04T14:27:12 < dongs> <Laurenceb> I just like it cuz it has 420 in the t itle
2012-09-04T14:28:43 < Laurenceb> http://hackaday.com/2012/09/03/an-stm32-processor-powers-this-pc/
2012-09-04T14:29:45 < tavish> jpa-: any idea where this limit could be from? i am not using the cdc-acm driver on linux
2012-09-04T14:31:42 < zyp> tavish, i.e. exactly 500 packets per second?
2012-09-04T14:32:03 < tavish> yeah
2012-09-04T14:33:57 < zyp> what about increasing packet size?
2012-09-04T14:34:45 < tavish> it full speed, so max size is 64
2012-09-04T14:41:17 < zyp> for bulk endpoints?
2012-09-04T14:44:06 < tavish> zyp: yeah, i think so. (though i'm not certain)
2012-09-04T14:44:17 < tavish> 512 bytes for high speed
2012-09-04T14:48:17 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-04T14:49:18 < Laurenceb> https://twitter.com/srgsrg/status/241907083858030594/photo/1/large
2012-09-04T14:49:41 < Laurenceb> lol thats the thing that nearly hit Heathrow
2012-09-04T14:53:06 < tavish> this is really weird. it seems to send those 500 packets all at once, wait for a second, send 500 more, wait for a sec :|
2012-09-04T14:58:18 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-04T15:01:07 < karlp> you have some weird interrupt blocking somewhere
2012-09-04T15:16:36 < dongs> 64bytes per endpoint for FS.
2012-09-04T15:17:41 < dongs> whats happening
2012-09-04T15:17:46 < dongs> any new trolling
2012-09-04T15:18:01 < dongs> Laurenceb: did you order your stellaris launchpad for $4.99
2012-09-04T15:18:29 < dongs> https://estore.ti.com/Stellaris-LaunchPad.aspx
2012-09-04T15:18:41 < dongs> max qty 2
2012-09-04T15:18:48 < dongs> i ordred 2, they take paypal
2012-09-04T15:19:10 < cjbaird> Ordered mine today via E14. Just the one.
2012-09-04T15:19:58 < cjbaird> Also preordered a Kinetis KL2 arduino wannabe
2012-09-04T15:20:11 < dongs> eh
2012-09-04T15:20:14 < dongs> url to kinetis?
2012-09-04T15:20:21 < dongs> i was all excited about K20/K40 shit
2012-09-04T15:20:34 < dongs> but it had no fpu and the chips were huge and the tools/libs were just terrible
2012-09-04T15:20:43 < dongs> codewarrior eclipse ripoff shit holy shit
2012-09-04T15:20:59 < cjbaird> http://au.element14.com/jsp/search/productdetail.jsp?SKU=2115294
2012-09-04T15:21:15 < dongs> haha da fuq
2012-09-04T15:21:16 < tavish> karlp: holy crap i think its the timer i am using for delays used by USB :( that's the only thing i can see
2012-09-04T15:21:57 < dongs> so its M0 core?
2012-09-04T15:22:03 < cjbaird> I hope you got 10 bux
2012-09-04T15:22:04 < dongs> wtf, yawh
2012-09-04T15:22:04 < dongs> hyawn
2012-09-04T15:22:09 < cjbaird> "M0+", whatever that is.
2012-09-04T15:22:13 < dongs> wehre can i order it for $10 w/paypal
2012-09-04T15:23:41 < cjbaird> http://www.element14.com/community/community/knode/dev_platforms_kits/element14_dev_kits/kinetis_kl2_freedom_board?ICID=hp_freedombanner .. more dox
2012-09-04T15:24:46 < cjbaird> Uses OpenSDA, whatever that is.
2012-09-04T15:25:01 < dongs> sophisticated usb debug interface
2012-09-04T15:25:12 < dongs> when is that new shit gonna get supported
2012-09-04T15:25:28 < dongs> CMSIS-DAP
2012-09-04T15:25:36 < dongs> that will kill all the overpriced vendors
2012-09-04T15:30:47 < cjbaird> KL2 product brief: http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=KL2&webpageId=133113337965471295E3EC&nodeId=01624698C9E3EC&fromPage=tax
2012-09-04T15:31:18 < dongs> looks like izua's job http://eluabrain.blogspot.ro/2011/11/brainz-brainz-braaaaaaaaainz.html
2012-09-04T15:31:37 < cjbaird> Does -40C to 105C.. Where's the balloon people? :)
2012-09-04T15:35:42 < dongs> argh i got more pcb orders and none of my projects to panel
2012-09-04T15:37:12 < dongs> what can i make tonight that will need a pcb
2012-09-04T15:38:30 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-04T15:38:34 < Laurenceb> cant be bothered to learn stellaris
2012-09-04T15:39:45 < karlp> does the same arm-gcc-launchpad work for stellaris and kinetis? or do you need other tools?
2012-09-04T15:40:02 < dongs> dunno about M0+
2012-09-04T15:40:11 < dongs> it should "compile" code, but good luck getting it on there
2012-09-04T15:40:39 < karlp> yeah, the getting it there was the question really :)
2012-09-04T15:40:47 < karlp> didn't word that properly.
2012-09-04T15:41:18 < cjbaird> Google Code was where I found the explanation for OpenSDA, but the bloody programmer is Spanish.
2012-09-04T15:42:15 < cjbaird> It's something preferred by the Android gadget scene?
2012-09-04T15:43:53 < Laurenceb> http://www.flickr.com/photos/daveake/7915918978/in/set-72157630074184822/
2012-09-04T15:43:57 < Laurenceb> thats slightly nuts
2012-09-04T15:44:44 < Laurenceb> lol its connects to gps too http://www.flickr.com/photos/daveake/7915809082/in/set-72157630074184822/
2012-09-04T15:44:54 < dongs> i have no fucking idea what opensda is
2012-09-04T15:45:08 < Laurenceb> opensad?
2012-09-04T15:45:28 < dongs> OpenSDA is an open-standard serial and debug adapter.
2012-09-04T15:46:14 < dongs> cjbaird: where did you even see googlecode shit.
2012-09-04T15:46:15 < Laurenceb> also known as an cp2102
2012-09-04T15:46:20 < dongs> Laurenceb: haha
2012-09-04T15:46:28 < cjbaird> Okay, the KL2 should be less of a brink than the NXP shit I've got, hopefully.
2012-09-04T15:46:32 <+izua> dongs: dekar_ is the lua guy bro
2012-09-04T15:47:12 < dongs> The board also has an on board “OpenSDA” Debugger and has an Arduino form factor, which will make it very interesting.
2012-09-04T15:47:50 < dongs> how does arduino form f actor make anything VERY INTERESTING
2012-09-04T15:47:50 < dongs> if anything, it makes it FUCKING RETARDED
2012-09-04T15:48:43 < cjbaird> https://code.google.com/p/opensdaframework/ .. dead project, but it's what I figured it out from.
2012-09-04T15:48:47 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-04T15:48:54 < dongs> cjbaird: that didnt look like anything debug related
2012-09-04T15:48:58 < dongs> they didnt even have any files for donwload
2012-09-04T15:49:08 <+izua> dongs: arduino form factor teaches you about cool layouts and connector placement
2012-09-04T15:49:16 < dongs> Adventista del Séptimo Día como alternativa del Sda Framework (NHiberbate+WCF+Silverlight) de la División Sudamericana.
2012-09-04T15:49:30 <+izua> dongs: what are you, still erasing your 10f pics with uv lamp? get on with the times, use an arduino!
2012-09-04T15:49:35 < dongs> whatever the fuck that says but it says "silverlight" and "WCF" in there so im prety sure its not debu real;ted
2012-09-04T15:49:39 <+izua> lamps/light
2012-09-04T15:50:07 <+izua> (in hindsight 10F PICs were pretty cool)
2012-09-04T15:52:48 < cjbaird> disable javascript to avoid needing a login, lol: http://www.element14.com/community/docs/DOC-49059/l/freescale-opensda-serial-debug-adapter-user-guide-rev-09
2012-09-04T15:53:01 < dongs> yeah i was wondering what hte fuck was up with htat
2012-09-04T15:53:13 < dongs> it showedf some shit under it that seemed legit then asked for e14 username
2012-09-04T15:53:22 < dongs> gonna see if chrome has a 'remove anything' extension to dick with dom
2012-09-04T15:53:59 < cjbaird> lynx works, ha
2012-09-04T15:54:15 < zyp> dongs, what, you aren't using IE?
2012-09-04T15:54:26 < zyp> I thought you only used enterprise quality software
2012-09-04T15:54:33 < dongs> yep
2012-09-04T15:54:37 < dongs> 4 clicks to remove elements and its gone
2012-09-04T15:55:22 < Laurenceb> e14 is runny
2012-09-04T15:56:26 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-04T15:56:29 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-04T15:56:48 < dongs> http://www.pemicro.com/opensda/pe_tools.cfm
2012-09-04T15:56:54 < dongs> oh
2012-09-04T15:56:57 < dongs> opensda is just a shit to upsell you to p&e micro crap
2012-09-04T15:59:21 < BrainDamage> giant wtf: http://wlanbook.com/free-public-wifi-ssid/
2012-09-04T16:02:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-04T16:05:52 <+Steffanx> lol BrainDamage
2012-09-04T16:07:38 < cjbaird> Wondering if the KL2duino will provide access to the SWD
2012-09-04T16:08:10 < karlp> in what way?
2012-09-04T16:08:20 < karlp> the opensda is connected to the swd pins.
2012-09-04T16:08:42 < karlp> or do you mean to be able to program your own kinetis devices, like with the discovery boards?
2012-09-04T16:09:15 < cjbaird> In case if the OpenSDA has issues for us members of the Free World, hacking onto the SWD might be the go.
2012-09-04T16:11:13 < dongs> why is shit like schematics etc not available
2012-09-04T16:11:29 < dongs> if its "open"
2012-09-04T16:12:03 < dongs> the only place mentioning it is that p&e micro page that says once youre ready for "real development" upgrade to other shit
2012-09-04T16:12:17 < karlp> yeah, doesnt really sound super promising
2012-09-04T16:12:22 < karlp> cmsis-dap sounds much better
2012-09-04T16:12:32 < karlp> or whatever arm is calling their interface
2012-09-04T16:12:37 < dongs> yes that
2012-09-04T16:33:47 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-04T16:53:10 < Laurenceb> not going to listen in case i catch the autism
2012-09-04T16:59:46 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-04T17:07:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-04T17:16:05 < dongs> what the fuck plays ogg in 2012
2012-09-04T17:16:08 < dongs> ogg is like lunix desktop
2012-09-04T17:16:13 < dongs> worthless since 1998
2012-09-04T17:16:19 <+dekar_> *click*
2012-09-04T17:16:22 <+dekar_> chrome plays it
2012-09-04T17:16:23 <+dekar_> :D
2012-09-04T17:16:36 < dongs> youre shitting me
2012-09-04T17:16:43 <+dekar_> what is that stuff anyway?
2012-09-04T17:16:49 < dongs> haha.. wow
2012-09-04T17:16:58 < dongs> had no idea
2012-09-04T17:17:08 <+dekar_> me neither :P
2012-09-04T17:17:34 <+izua> dongs: winamp 2 plays it pretty well
2012-09-04T17:17:38 <+izua> it has some sweet skins too!
2012-09-04T17:17:43 < karlp> ogg is actually one of the quite useful web formats for audio.
2012-09-04T17:25:20 <+dekar_> when do the mp3 patents end?
2012-09-04T17:25:25 <+dekar_> can't be too long
2012-09-04T17:26:26 <+izua> 3000 something
2012-09-04T17:27:14 <+izua> http://www.tunequest.org/a-big-list-of-mp3-patents/20070226/ here's what i have
2012-09-04T17:29:49 < jpa-> Granted: February 6, 2001  wtf, how could that be valid if it applies to mp3?
2012-09-04T17:33:03 < karlp> what, you think patent grants are meant to follow patent grant rules?
2012-09-04T17:33:03 < dongs> izua, with some extar drivers
2012-09-04T17:33:08 < karlp> what rock have you been living under?
2012-09-04T17:33:20 < Laurenceb> so is the music autistic?
2012-09-04T17:33:24  * Laurenceb has no speakers
2012-09-04T17:33:26 < dongs> my isntall of winamp 2.91 sure doesnt
2012-09-04T17:33:33 < Laurenceb> loooll winamp
2012-09-04T17:33:41 < karlp> don't know, I'm listening to this instead: http://africantapegroup.bandcamp.com/album/bon-sauvage
2012-09-04T17:33:55 < Laurenceb> bandcamp...
2012-09-04T17:42:54 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-09-04T17:44:01 < Laurenceb> http://i.imgur.com/RfOMM.jpg
2012-09-04T17:44:21  * Laurenceb is making lots of ^
2012-09-04T17:45:07 < BrainDamage> because everytime you test one, it's meant to be thrown away because it has ripped off a skin layer?
2012-09-04T17:45:21 < Laurenceb> lol
2012-09-04T17:45:36 < Laurenceb> its meant to be disposable
2012-09-04T17:45:47 < BrainDamage> the skin? :p
2012-09-04T17:46:09 < BrainDamage> btw, that connector looks rather odd
2012-09-04T17:46:22 < BrainDamage> reminds me of one of those cellphone connectors
2012-09-04T17:46:26 < Laurenceb> it is
2012-09-04T17:47:14 < Laurenceb> the most annoying part of making them is threading the cable down the hose
2012-09-04T17:47:38 < karlp> what's wrong with the cable itself?
2012-09-04T17:47:58 < karlp> might contain BPA ohnoes might kill baby?
2012-09-04T17:48:02 < Laurenceb> yes
2012-09-04T17:48:16 < Laurenceb> also cable down the side is messy
2012-09-04T17:48:37 < karlp> oh, so the hollow space is in the tube is important?
2012-09-04T17:48:48 < karlp> it's not just a "better" material?
2012-09-04T17:49:06 < Laurenceb> yes - its for the air supply
2012-09-04T17:49:15 < Laurenceb> thats what the black plug does
2012-09-04T17:49:25 < Laurenceb> connects to air adaptor
2012-09-04T17:50:34 < karlp> I thought you were just monitoring pulse or something?
2012-09-04T17:50:47 < Laurenceb> nope
2012-09-04T17:51:04 < dongs> he's monitoring erection
2012-09-04T17:51:10 < Laurenceb> lol
2012-09-04T17:51:12 < dongs> (his own)
2012-09-04T17:51:24 < Laurenceb> it measures blood flow
2012-09-04T17:51:38 < Laurenceb> so yes it may measure erection quite well
2012-09-04T17:51:54 < karlp> ok, but blood flow via your leds,
2012-09-04T17:51:59 < karlp> where does the air supply come in?
2012-09-04T17:52:34 < Laurenceb> you cant measure flow with LEDs  :P
2012-09-04T17:52:55 < Laurenceb> it squashes out the blood using applied pressure
2012-09-04T17:53:13 < Laurenceb> then releases it and looks at the blood flowing in
2012-09-04T17:53:20 < Laurenceb> to get volumetric flow info
2012-09-04T17:55:18 < Laurenceb> only IRL its clusterfucked
2012-09-04T17:55:27 < Laurenceb> the front keeps blowing off
2012-09-04T17:55:33 < karlp> oh, ok, I thought you were "applying science" to continually monitoring the colour/xxx of the skin to gauge flow that way
2012-09-04T17:55:48 < karlp> but you're just poking things and seeing how long it takes for the colour to come back.
2012-09-04T17:55:55 < karlp> but with moah robots!
2012-09-04T17:55:57 < Laurenceb> more or less
2012-09-04T17:56:17 < Laurenceb> im using the gradient
2012-09-04T17:56:32 < karlp> so, hack the system, and your device turns into an air cannon strapped onto baby's side.
2012-09-04T17:57:11 < Laurenceb> lol
2012-09-04T17:57:13 < Laurenceb> http://i.imgur.com/TPhEE.png
2012-09-04T17:57:58 < karlp> is vertical "colour" ?
2012-09-04T17:58:00 < Laurenceb> (gradient*constant for blood absorption* constant for path length) / (DC level)
2012-09-04T17:58:05 < Laurenceb> give the flow
2012-09-04T17:58:19 < Laurenceb> vertical is intensity
2012-09-04T17:58:48 < karlp> which is basically red(pink)vs white if I'm poking my skin?
2012-09-04T17:58:52 < Laurenceb> kind of
2012-09-04T17:59:00 < Laurenceb> but it uses 1 micron IR
2012-09-04T17:59:06 < Laurenceb> to get deep penetration
2012-09-04T18:01:50 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-04T18:04:59 < BrainDamage> blood absoption sounds like you're using a gas that is uncommon in air
2012-09-04T18:05:04 < BrainDamage> cyanide? :p
2012-09-04T18:05:28 < BrainDamage> altough it's probably something with small atomic number, so it penetrates more easily
2012-09-04T18:05:34 < BrainDamage> probably hydrogen or helium
2012-09-04T18:05:45 < Laurenceb> heh
2012-09-04T18:05:53 < BrainDamage> btw, are you aware this channel is publicly logged?
2012-09-04T18:05:54 < Laurenceb> yeah thats my secret plan
2012-09-04T18:05:59 < Laurenceb> doh
2012-09-04T18:06:27 < Laurenceb> yeah i want to try it using hydrgen gas absorption
2012-09-04T18:06:41 < Laurenceb> but this is all supposed to be secret
2012-09-04T18:07:10 < BrainDamage> I guess it's safe between this channel, and the google scraper bot that is indexing the logs
2012-09-04T18:07:42 < Laurenceb> technically im self employed
2012-09-04T18:08:33 < Laurenceb> but im planning to license the IP
2012-09-04T18:08:46 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-09-04T18:08:52 < Laurenceb> lol
2012-09-04T18:09:52 < Laurenceb> but that would involve removing it from github XD
2012-09-04T18:11:46 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-04T18:12:59  * dongs quickly mirrors
2012-09-04T18:13:30 < Laurenceb> http://cubieboard.org/
2012-09-04T18:13:32 < Laurenceb> omfg
2012-09-04T18:13:57 < Laurenceb> rpi is doa
2012-09-04T18:14:20 < zyp> *yawn*
2012-09-04T18:15:28 < jpa-> i wonder how easily it would be to do realtime stuff on those fast arm boards
2012-09-04T18:16:31 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-04T18:19:37 <+Steffanx> That project is still dead Laurenceb
2012-09-04T18:19:41 <+Steffanx> @ cubieboard
2012-09-04T18:19:59 <+izua> is dead?
2012-09-04T18:20:18 <+izua> why don't any of these cheap computers ever make it out in large numbers
2012-09-04T18:20:24 <+Steffanx> It's dead as long as it's not shipped :P
2012-09-04T18:20:52 <+Steffanx> And it's just another Allwinner crap board
2012-09-04T18:21:18 < zyp> izua, there is tons of allwinner-based crap everywhere
2012-09-04T18:21:52 <+izua> allwiener
2012-09-04T18:22:28 <+Steffanx> aufwiederschnitzel ?
2012-09-04T18:22:55 < zyp> izua, by the way, I mailed your board today, so it should be on it's way now
2012-09-04T18:23:38 <+izua> erdscheiße
2012-09-04T18:23:51 <+izua> zyp: awesome! how much for shipping?
2012-09-04T18:24:06 < zyp> ~$5 or so
2012-09-04T18:24:36 < zyp> cheaper than I had expected :p
2012-09-04T18:26:29 <+izua> Steffanx: actually i think that's himscheiße, the erd- being strawberries
2012-09-04T18:26:52 <+izua> zyp: well, let me know how much i owe you for shipping and where to paypal it
2012-09-04T18:27:11 <+Steffanx> 9.99$
2012-09-04T18:27:12 <+izua> and/or for the board/parts as well..
2012-09-04T18:28:16 <+Steffanx> zyp is rich, so he doesn't need more money izua
2012-09-04T18:28:35 < zyp> :p
2012-09-04T18:28:54 <+izua> well, he did spend his new year's eve taking pictures atop a mountain
2012-09-04T18:29:33 <+Steffanx> Oh, so this year you'll join him?
2012-09-04T18:29:47 <+izua> nah, i barely have money for my summer hiking gear
2012-09-04T18:30:22 < zyp> …taking pictures with camera equipment I probably spent $2000 on, Steffanx might have a point :p
2012-09-04T18:31:22 < Laurenceb> izua: what board did you buy?
2012-09-04T18:31:38 < zyp> I shipped him one of my old led matrix boards
2012-09-04T18:31:50 <+izua> i went all kawai on some tiny 16x16 led modules
2012-09-04T18:32:02 <+izua> and zyp decided to part with one
2012-09-04T18:32:06 < Laurenceb> ah
2012-09-04T18:32:26 <+izua> he still has 9 though, he could rig up a sweet 48x48.. game of life?
2012-09-04T18:32:30 < zyp> :p
2012-09-04T18:32:34 <+Steffanx> Now you both have 1/2 of the pong-board?
2012-09-04T18:32:46 < Laurenceb> wolfram would be impressed
2012-09-04T18:33:21 < zyp> I'm just looking forward to see somebody doing something cool with them
2012-09-04T18:33:36 < emeb> zyp: you get boards from oshpark?
2012-09-04T18:33:39 < zyp> they have just been collecting dust here for years :p
2012-09-04T18:33:46 < zyp> those boards are from seeed
2012-09-04T18:33:58 < karlp> oshpark is the new name for the dorkpot laen stuff right?
2012-09-04T18:34:03 < zyp> yes
2012-09-04T18:34:29 < emeb> I see that they use ENIG finish.
2012-09-04T18:34:38 < zyp> yeah, they are quite nice
2012-09-04T18:35:04 < emeb> you stencil/paste/reflow to assemble, or hand solder?
2012-09-04T18:35:05 < zyp> but the part I like the most is the cheap four layer boards
2012-09-04T18:35:11 < zyp> former
2012-09-04T18:35:33 < emeb> yeah - $10/sqin for 3 boards. Nice price.
2012-09-04T18:36:02 < zyp> http://bin.jvnv.net/f/FlKvz.JPG <- here is pre and post reflow of one of those led matrix boards
2012-09-04T18:36:17 < zyp> well, two of those…
2012-09-04T18:37:25 < emeb> hand place the components in the paste or do you use one of those jigs?
2012-09-04T18:37:31 < zyp> hand placed
2012-09-04T18:38:02 < emeb> haven't really figured out what those placement jigs bring to the party.
2012-09-04T18:38:37 < zyp> probably for people that are not steady enough with tweezers :p
2012-09-04T18:38:58 < emeb> yep. Hasn't been an issue for me (yet)
2012-09-04T18:39:14 <+Steffanx> emeb isn't old enough, yet
2012-09-04T18:39:32 < emeb> Steffanx: I doubt you know how old I am.
2012-09-04T18:39:56 < emeb> but yeah. that's true.
2012-09-04T18:40:04 < emeb> (shakes cane)
2012-09-04T18:40:14 < zyp> I just know Steffanx is slightly younger than me :p
2012-09-04T18:42:03 < emeb> thinking about getting some 4L boards from oshpark, but concerned that I have no experience soldering ENIG.
2012-09-04T18:42:20 < zyp> experience how?
2012-09-04T18:42:33 < zyp> it's not really different from anything else
2012-09-04T18:42:47 < emeb> always used HASL in the past - nice because I can solder fine pitch SMD just by reflowing the solder that's already there.
2012-09-04T18:43:43 < zyp> if you're working that way, just apply solder to the pads before placing the component
2012-09-04T18:44:01 < zyp> or apply paste
2012-09-04T18:44:27 < emeb> yep - that's a new area for me.
2012-09-04T18:45:04 < emeb> tried paste a few years ago but it didn't work well - wouldn't stick to the board.  Probably old paste.
2012-09-04T18:45:56 < zyp> I find that when I'm ordering a board, I might as well just order a stencil at the same time
2012-09-04T18:46:48 < emeb> what's a typical small stencil cost?
2012-09-04T18:46:56 < Laurenceb> free
2012-09-04T18:47:03 < emeb> :P
2012-09-04T18:47:09 < Laurenceb> if you're friends with shrudlu- on ##electronics
2012-09-04T18:47:11 < zyp> after all, PCB CAD output gerbers for stencils along with the other gerbers
2012-09-04T18:47:14 < Laurenceb> or wtf hes called
2012-09-04T18:47:42 < emeb> Yeah - pastemask layer is there.
2012-09-04T18:47:54 < zyp> http://www.ohararp.com/Stencils.html <- so far I've gotten mine from here
2012-09-04T18:48:08 < zyp> it's a bit expensive if you only have one design
2012-09-04T18:48:28 < zyp> but he lets you submit multiple as long as they all fit on one sheet
2012-09-04T18:48:40 < emeb> Makes sense
2012-09-04T18:49:15 < zyp> I got the stencil for that led matrix board for «free» since I just included it with another design
2012-09-04T18:50:56 < zyp> I have a couple of ideas for handy stuff I could make, so next time I'm ordering a PCB I'm going to do those other mini projects at the same time
2012-09-04T18:51:26 < emeb> JTAG -> SWD adapter boards. ;)
2012-09-04T18:51:45 < karlp> $43/sheet for overseas, that's far far beyond reasonable for a hobby project
2012-09-04T18:52:02 < Laurenceb> stm32f1 is so mainstream man
2012-09-04T18:52:28 < emeb> Laurenceb needs an obscure MCU family.
2012-09-04T18:52:37 < emeb> for hipster cred.
2012-09-04T18:52:54 < karlp> get cortex-r, for better piracy
2012-09-04T18:52:57 < dongs> Laurenceb: beep
2012-09-04T18:52:58 < emeb> How long before the F3 is available in a store near you?
2012-09-04T18:53:21 < dongs> GPS.print("$PUBX,41,1,0007,0003,4800,0*13\r\n");
2012-09-04T18:53:29 < dongs> what does the o ther shit mean in it
2012-09-04T18:53:31 < dongs> is there some doc
2012-09-04T18:53:41 < dongs> does some of the shit in pubx disable nmea output?
2012-09-04T18:54:44 < dongs> ya ok
2012-09-04T18:54:45 < dongs> argh
2012-09-04T18:55:05 < jpa-> those are nonstandard commands; google finds plenty of documentation
2012-09-04T18:55:12 < dongs> yeah found the original doc
2012-09-04T18:55:16 < Laurenceb> i only use ubx
2012-09-04T18:55:18 < Laurenceb> dunno
2012-09-04T19:03:02 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-04T19:03:27 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Remote host closed the connection]
2012-09-04T19:04:13 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-09-04T19:05:44 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-04T19:06:08 -!- Hoppo [~johnhopki@213.86.244.72] has joined ##stm32
2012-09-04T19:06:27 -!- claude_work is now known as claude
2012-09-04T19:07:07 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-04T19:07:11 < TitanMKD> hi
2012-09-04T19:07:34 -!- Hoppo [~johnhopki@213.86.244.72] has left ##stm32 []
2012-09-04T19:08:37 < Laurenceb> hi
2012-09-04T19:09:39 <+Steffanx> lo
2012-09-04T19:10:20 < zyp> z
2012-09-04T19:12:11 < emeb> x
2012-09-04T19:18:11 < dongs> y
2012-09-04T19:18:33 < Laurenceb> combo breaker
2012-09-04T19:18:39 < dongs> you faggot, it was going to be so kawaii
2012-09-04T19:21:29 < dongs> Program Size: Code=39608 RO-data=4068 RW-data=1220 ZI-data=6380
2012-09-04T19:21:35 < dongs> still beats gaycc
2012-09-04T19:23:08 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 252 seconds]
2012-09-04T19:23:18 <+Steffanx> No a man behind windows is nice dongs ? :P
2012-09-04T19:24:34 <+Steffanx> Note to self: Don
2012-09-04T19:24:37 <+Steffanx> t feed trolls
2012-09-04T19:25:38 < TitanMKD> for those interested in LPC43xx crazy board i just updated a design ;)
2012-09-04T19:25:40 < TitanMKD> http://imgur.com/Nlzy9
2012-09-04T19:26:03 < TitanMKD> it's LBGA256 LPC4350 on 2 layers and 5cm*5cm
2012-09-04T19:26:10 < emeb> Gah! BGA on 2 layers!
2012-09-04T19:26:41 < TitanMKD> yes but lot of GPIO are not routed ;)
2012-09-04T19:26:54  * Steffanx can't do BGA
2012-09-04T19:27:03 < TitanMKD> I have tried to output the maximum number of GPIO with just 2 layers
2012-09-04T19:27:05  * emeb can't either.
2012-09-04T19:27:10 < zyp> what's the point of using BGA if you're not using a lot of GPIO?
2012-09-04T19:27:12 < TitanMKD> anyway it's a challenge to solder that ;)
2012-09-04T19:27:24 < zyp> isn't it available in lower pin-count QFPs?
2012-09-04T19:27:29 < TitanMKD> zyp the point is i received thos LBGA chip freely ;)
2012-09-04T19:27:47 < emeb> worth what you paid.
2012-09-04T19:27:49 < TitanMKD> but for real world i will use LQFP ;)
2012-09-04T19:28:36 < TitanMKD> it's also to check if it is possible to solder such BGA using hot plate
2012-09-04T19:29:35 < emeb> TitanMKD: U2 is SPI Flash?
2012-09-04T19:29:43 < TitanMKD> it is QuadSPI ;)
2012-09-04T19:29:50 < TitanMKD> called also SPIFI ;)
2012-09-04T19:30:12 < emeb> isn't that just... special.
2012-09-04T19:30:20 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-04T19:30:26 < TitanMKD> it just run at up to 25MBytes/s ;)
2012-09-04T19:30:43  * emeb wonders why NXP made an MCU w/o flash.
2012-09-04T19:30:50 < TitanMKD> or even 40MBytes/s but i need to check it ;)
2012-09-04T19:31:17 < TitanMKD> anyway i use the spiflash just to boot and i load and execute all the code in SRAM
2012-09-04T19:31:27 < TitanMKD> for max speed @ 204MHz ;)
2012-09-04T19:31:31 < emeb> yep - pretty much have to do that w/ those parts.
2012-09-04T19:31:38 < dongs> wuat
2012-09-04T19:31:41 < dongs> how much sram is on that shit
2012-09-04T19:31:43 < dongs> liek 16k?
2012-09-04T19:31:49 < TitanMKD> 196KB IIRC
2012-09-04T19:31:58 < TitanMKD> in Thumb2 it is lot of code ;)
2012-09-04T19:32:04 < dongs> surely theye have some FSMC equivalent for nor/nand flahs?
2012-09-04T19:32:20 < TitanMKD> yes it support even DDR ;)
2012-09-04T19:32:21 < zyp> I've heard rumors that they support SDRAM
2012-09-04T19:32:44 -!- barthess [~barthess@178.154.93.221] has joined ##stm32
2012-09-04T19:32:46 < TitanMKD> yes SDRAM is supported by the controller
2012-09-04T19:33:12 < TitanMKD> especially on such big LBGA256 version but it requires 4 or even 6 layers to route it
2012-09-04T19:33:21  * Laurenceb looks at his 118KB project
2012-09-04T19:33:54 < TitanMKD> code can be also executed from SPIFI ;)
2012-09-04T19:34:24 < TitanMKD> there's an internal cache of 512bytes in fact so for little loop the max perf of 204mips are reached ;)
2012-09-04T19:34:44 < TitanMKD> I have done lot of test to find that because it is not documented
2012-09-04T19:35:33 < TitanMKD> oups not 512bytes cache but 256bytes ;)
2012-09-04T19:35:45 < TitanMKD> see details and source code here for those interested
2012-09-04T19:35:46 < TitanMKD> https://github.com/TitanMKD/hackrf/tree/master/firmware/startup_systick_perfo
2012-09-04T19:38:07 < emeb> executing code from spi flash. now there's a kludge.
2012-09-04T19:39:12 < TitanMKD> in fact it is the same on STM32F4 ;) it is not exaclty 0 wait state from onchip flash
2012-09-04T19:40:05 < TitanMKD> even if onchip flash is really faster than SPIFI
2012-09-04T19:41:29 < zyp> F4 got 1k of instruction cache
2012-09-04T19:41:50 < zyp> and 128B of data cache
2012-09-04T19:42:23 < TitanMKD> yes it is why for code > 1K there's some perf issues
2012-09-04T19:44:07 < zyp> the bus between flash and cache is 128 bits wide, so IIRC the bandwidth is high enough that it shouldn't be a problem in practice
2012-09-04T19:44:17 < dongs> 128bits, wow
2012-09-04T19:44:43 < zyp> yes, that's up to 8 instructions in one fetch
2012-09-04T19:44:56 < dongs> purty cool
2012-09-04T19:45:55 < TitanMKD> zyp but it depends also on flash latency
2012-09-04T19:46:24 < TitanMKD> I think internal flash are not > 60MB/s
2012-09-04T19:47:13 < TitanMKD> I could test it with same test as for LPC43xx
2012-09-04T19:47:24 < dongs> how do i flashed
2012-09-04T19:48:33 < emeb> meh. VHDL is such a PITA.
2012-09-04T19:48:40 < dongs> tell me about it.
2012-09-04T19:48:41 < BrainDamage> I'd have guessed that someone named dongs would know how to flash
2012-09-04T19:48:48 < dongs> its even more unreadable than verilog.
2012-09-04T19:49:07 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-04T19:49:10 < emeb> I hate the strong typing. Everything needs to be explicitly cast / converted.
2012-09-04T19:49:14 < jpa-> vhdl <3
2012-09-04T19:49:34 <+Steffanx> labview <3
2012-09-04T19:49:46 < jpa-> bat files <3
2012-09-04T19:49:50 < dongs> dongs <3
2012-09-04T19:50:14 <+Steffanx> dongs goes fap fap?
2012-09-04T19:51:50 < zyp> TitanMKD, if I'm reading datasheet correctly it supports a bus speed of up to 30 MHz, meaning that with a 16B wide bus, you can read it at 480 MB/s
2012-09-04T19:52:15 < TitanMKD> zyp hmm really not bad
2012-09-04T19:52:38 < zyp> so if the prefetcher/branch predictor is working correctly, it should pretty much be able to execute without waiting anything
2012-09-04T19:52:59 < TitanMKD> yes so maybe the 0 wait state is really true
2012-09-04T19:54:16 < zyp> s/correctly/perfectly/
2012-09-04T19:54:17 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Quit: leaving]
2012-09-04T19:55:01 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-09-04T20:13:34 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-04T20:26:51 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-04T21:08:15 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-04T21:49:27 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Quit: leaving]
2012-09-04T21:56:37 < TheSeven> hm, that laks i2c driver seems to be full of race conditions
2012-09-04T21:59:16 < TheSeven> and it seems like at least one of them actually can't be avoided
2012-09-04T21:59:53 < TheSeven> apparently if I'm in the ADDR state and set ACK, then wait for a while (single stepping), it will receive and acknowledge *two* bytes
2012-09-04T22:00:54 < TheSeven> i.e. I have to respond to the RXNE interrupt and clear ACK and set STOP before the pipelined next read proceeds to the acknowledge bit?
2012-09-04T22:01:29 < TheSeven> why does this stupid core clock stretch *after* that acknowledge, not before it, if it can't be sure whether the user actually wants that byte to be acked!?
2012-09-04T22:06:37 < TheSeven> oh, and another one
2012-09-04T22:07:38 <+Steffanx> laks .. by zyp, TheSeven ?
2012-09-04T22:10:36 < TheSeven> yes
2012-09-04T22:10:47 < emeb> wait - you're expecting to get accurate single-stepping on an interface with a real-time peripheral?
2012-09-04T22:11:18 < TheSeven> emeb: I2C should allow that, yes
2012-09-04T22:11:26 < emeb> Good luck with that.
2012-09-04T22:12:08 < emeb> for what it's worth, the I2C peripheral isn't running from the same clock that the CPU core is. Breakpoints won't stop its clock.
2012-09-04T22:12:12 < TheSeven> I don't really care about single stepping, but I do care about higher priority interrupts preemting the I2C IRQ handler and causing a race condition
2012-09-04T22:12:49 < TheSeven> IIUC there is just a fundamental flaw in the I2C core's interface
2012-09-04T22:12:54 < emeb> maybe wrap the I2C code with disable/enable IRQ?
2012-09-04T22:13:21 < TheSeven> well that kinda misses the point of giving it a lower priority than some more latency critical peripherals
2012-09-04T22:13:22 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-09-04T22:13:47 < TheSeven> IMO this is a bug. it shouldn't ACK the next byte before I've read the previous one
2012-09-04T22:14:07 < emeb> Send ST an email.
2012-09-04T22:14:34 < TheSeven> as if they were going to fix that
2012-09-04T22:14:59 < emeb> Yeah - not gonna happen, so work around it.
2012-09-04T22:15:25 < TheSeven> well it looks like there is no 100% stable workaround for this bug
2012-09-04T22:16:29 < emeb> why won't disabling IRQ work?
2012-09-04T22:17:36 < emeb> and, how granular can you get - do you need to disable for the entire I2C transaction, or just for this ACK section?
2012-09-04T22:23:23 < zyp> 20:56:37 < TheSeven> hm, that laks i2c driver seems to be full of race conditions
2012-09-04T22:23:31 < zyp> that is correct and known
2012-09-04T22:24:28 < zyp> you might want to read the Master Receiver chapter in the reference manual
2012-09-04T22:24:57 < zyp> see the notes on Method 1 and Method 2
2012-09-04T22:25:13 < zyp> «Method 1: This method is for the case when the I2C is used with interrupts that have the
2012-09-04T22:25:16 < zyp> highest priority in the application.»
2012-09-04T22:25:18 < zyp> that is what I'm doing
2012-09-04T22:25:41 < zyp> if you can't guarantee that the interrupt is serviced in a timely manner, you have to go with Method 2
2012-09-04T22:30:21 -!- DLPeterson [~hazelnuss@70.42.240.21] has joined ##stm32
2012-09-04T22:31:50 < emeb> which RM is that in? Not seeing it in the F4 RM.
2012-09-04T22:32:39 < zyp> oh, wait
2012-09-04T22:32:59 < zyp> I had both the F1 and F4 RMs open, I think I looked in the wrong one
2012-09-04T22:34:30 < emeb> yeah - in the F1 RM.
2012-09-04T22:35:33 < zyp> but yeah, I originally wrote the driver on F1, so it's written according to the F1 RM :p
2012-09-04T22:35:52 < emeb> So does that mean they fixed it in the F4?
2012-09-04T22:36:27 < zyp> I didn't change the driver when porting from F1 to F4
2012-09-04T22:37:22 < zyp> looks like the F4 RM only kept the «Method 2»
2012-09-04T22:37:53 < zyp> probably to avoid people having troubles with «Method 1»
2012-09-04T22:38:24 < emeb> So F4 doesn't need method 1 (hardware issue is resolved)
2012-09-04T22:38:43 < zyp> were there a hardware issue?
2012-09-04T22:39:58 < emeb> beats me - sounds like TheSeven thinks so.
2012-09-04T22:40:40 < emeb> and F1 Method 1 implies that there is some special effort required to handle I2C Rx when there's the possibility of IRQ.
2012-09-04T22:40:53 < emeb> whereas F4 seems to work w/o that special effort.
2012-09-04T22:41:44 < zyp> the difference is just that «Method 2» means a bit more special casing
2012-09-04T22:42:46 < zyp> ah, F1 errata have a chapter on that
2012-09-04T22:43:46 < emeb> oh - so f4 requires you to always use the special casing?
2012-09-04T22:43:58 < TheSeven> zyp:
2012-09-04T22:43:59 < TheSeven> <TheSeven> apparently if I'm in the ADDR state and set ACK, then wait for a while (single stepping), it will receive and acknowledge *two* bytes
2012-09-04T22:43:59 < TheSeven> <TheSeven> i.e. I have to respond to the RXNE interrupt and clear ACK and set STOP before the pipelined next read proceeds to the acknowledge bit?
2012-09-04T22:44:13 < zyp> if you rely on using Method 2 on F1, you have a race condition that can be worked around by temporary setting the lines as normal GPIOs and pulling them low
2012-09-04T22:45:20 < zyp> TheSeven, you are using F4, right?
2012-09-04T22:45:27 < TheSeven> yes
2012-09-04T22:45:45 < TheSeven> but that's just on the devboard, the final application will be f207
2012-09-04T22:45:57 < zyp> should be the same
2012-09-04T22:46:03 < TheSeven> I'd hope so
2012-09-04T22:46:16 < zyp> check page 584 in the reference manual
2012-09-04T22:46:19 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-04T22:46:24 < zyp> «For 2-byte reception»
2012-09-04T22:46:40 < zyp> that's a special case that should solve your race condition
2012-09-04T22:47:20 < Laurenceb_> let me guess....
2012-09-04T22:47:24 < Laurenceb_> i2c is breaking?
2012-09-04T22:47:29 < Laurenceb_> correct?
2012-09-04T22:47:34 < zyp> :)
2012-09-04T22:47:48 < Laurenceb_> hehe
2012-09-04T22:48:07 < zyp> Laurenceb_, looks like the F4 RM only documents the old «Method 2» now
2012-09-04T22:48:19 < Laurenceb_> reference manual + errata is actually quite helpful
2012-09-04T22:48:29 < Laurenceb_> but there are 3 undocumented issues iirc
2012-09-04T22:48:35 < TheSeven> zyp: that race condition seems to apply to all >1 byte transfers
2012-09-04T22:48:39 < Laurenceb_> yeah but method2 is nicer
2012-09-04T22:48:44 < Laurenceb_> once you get it running
2012-09-04T22:48:57 < Laurenceb_> i never fully debugged 1 byte transfer iirc
2012-09-04T22:49:06 < Laurenceb_> tested it once or twice but not fully
2012-09-04T22:49:11 < zyp> Laurenceb_, yeah, but I implemented «Method 1», and TheSeven is complaining about it not working right :p
2012-09-04T22:49:30 < Laurenceb_> ah
2012-09-04T22:49:31 < zyp> s/right/as he expected/
2012-09-04T22:49:35  * Laurenceb_ digs out links
2012-09-04T22:50:12 < Laurenceb_> https://github.com/Laurenceb/Dactyl/blob/master/i2c_int.c
2012-09-04T22:50:25 < Laurenceb_> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=https%3a%2f%2fmy.st.com%2fpublic%2fSTe2ecommunities%2fmcu%2fLists%2fcortex_mx_stm32%2fSTM32F103%20I2C%20interrupt%20driven%20method%202&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=399
2012-09-04T22:50:40 < zyp> according to F4 errata, there shouldn't be any timing related issues on F4 like F1 had
2012-09-04T22:50:41  * emeb is glad that his board only needs I2C TX
2012-09-04T22:51:53 < Laurenceb_> on F1 there is another issue where you get looping interrupts if you arent careful
2012-09-04T22:52:04 < Laurenceb_> i dont think i described it there
2012-09-04T22:52:30 < TheSeven> I'm getting two spurious TXE interrupts after sending a repeated START btw
2012-09-04T22:52:37 < TheSeven> not quite sure what's causing them
2012-09-04T22:52:38 < Laurenceb_> but start and stop have to be left to clear before _anything_ is done to the i2c peripheral
2012-09-04T22:52:49 < Laurenceb_> or it will lock up for good on F1
2012-09-04T22:53:13 < Laurenceb_> TheSeven: ah yes
2012-09-04T22:53:22 < zyp> Laurenceb_, I never had any problems of that sort
2012-09-04T22:53:28 < Laurenceb_> that was the other issue i couldnt remember
2012-09-04T22:53:41 < Laurenceb_> zyp: maybe its some sort of race condition and you by chance avoided it
2012-09-04T22:53:52 < Laurenceb_> maple had the same issue i had with start and stop aiui
2012-09-04T22:54:19 < Laurenceb_> so yeah 3 undocumented issues
2012-09-04T22:54:21 < zyp> might be, I'm never sending a start after a stop from the same ISR
2012-09-04T22:54:30 < zyp> I'm never sending a start from the ISR at all
2012-09-04T22:54:31 < Laurenceb_> start, stop and spurious interrupts
2012-09-04T22:55:08 < Laurenceb_> cant remember how to solve the spurious interrupts... its in my code somewhere
2012-09-04T22:57:07 < BrainDamage> bah, hams are full of anger and envy, they cannot reckognize the beauty of my "hamtroll 0.1" transmitter design: http://www.twiddla.com/925110
2012-09-04T22:57:28 < BrainDamage> it's beautiful in it's simplicity
2012-09-04T22:57:28 < TheSeven> Laurenceb_: so what do to about the spurious IRQs?
2012-09-04T22:57:41 < TheSeven> is there a known workaround to prevent that from happening?
2012-09-04T22:57:45 < TheSeven> or do I need to filter them out?
2012-09-04T22:58:00 < TheSeven> they seem to not happen when adding some delays at some points
2012-09-04T22:58:14 < Laurenceb_> hmm
2012-09-04T22:58:16 < TheSeven> oh, and I'm also getting an IRQ with SR1=0 SR2=3 which seems kinda pointless
2012-09-04T22:58:22 < Laurenceb_> ok ill read my code again
2012-09-04T22:58:49 < Laurenceb_> lmao BrainDamage
2012-09-04T22:59:16 < Laurenceb_> no i wont
2012-09-04T22:59:24 < Laurenceb_> my microwave killed my wifi again
2012-09-04T23:01:03 < Laurenceb_> TheSeven: iirc the spurious interrupts is due to how you read the status registers
2012-09-04T23:01:08 < Laurenceb_> hence this code https://github.com/Laurenceb/Dactyl/blob/master/i2c_int.c#L34
2012-09-04T23:01:36 < TheSeven> I'm doing that fairly similar
2012-09-04T23:01:52 < Laurenceb_> reads SR1 before trying anything else
2012-09-04T23:02:43 -!- DLPeterson [~hazelnuss@70.42.240.21] has quit [Ping timeout: 244 seconds]
2012-09-04T23:03:32 < Laurenceb_> all i can suggest is try my code
2012-09-04T23:11:17 < Laurenceb_> anyone here used ThreadX?
2012-09-04T23:15:58 < TheSeven> Laurenceb_: fixed by waiting for start bit to clear during repeated start before leaving the ISR
2012-09-04T23:16:08 < TheSeven> another undocumented quirk...
2012-09-04T23:16:26 < Laurenceb_> ah
2012-09-04T23:17:11 < Laurenceb_> yeah  https://github.com/Laurenceb/Dactyl/blob/master/i2c_int.c#L116
2012-09-04T23:17:51 < TheSeven> now I just have a stuck STOP command
2012-09-04T23:18:25 < TheSeven> which is set in TXE state
2012-09-04T23:18:41 < TheSeven> when doing repeated START in TXE and then STOP in RXNE it works
2012-09-04T23:19:07 < Laurenceb_> what causes the RXNE?
2012-09-04T23:19:22 < Laurenceb_> oh if you stop in RX mode
2012-09-04T23:19:27 < zyp> finished reception of a byte
2012-09-04T23:19:47 < Laurenceb_> yeah
2012-09-04T23:20:25 < Laurenceb_> TheSeven: wait for stop to clear before touching any registers
2012-09-04T23:21:07 < Laurenceb_> ive had the stop lockup too
2012-09-04T23:22:19 < TheSeven> well I'm already doing that
2012-09-04T23:22:33 < TheSeven> it works if I manually clear the stop bit btw
2012-09-04T23:22:39 < TheSeven> then I can start the next transaction
2012-09-04T23:22:51  * TheSeven grabs the scope to check whether that stop bit actually gets transmitted
2012-09-04T23:25:03 < TheSeven> bah
2012-09-04T23:25:10 < TheSeven> now it just stopped happening
2012-09-04T23:25:45 < Laurenceb_> in my experience it can jam up with stop set
2012-09-04T23:26:11 -!- barthess [~barthess@178.154.93.221] has quit [Quit: Leaving.]
2012-09-04T23:28:04 < TheSeven> bwahahaha
2012-09-04T23:28:07 < TheSeven> seems to finally be working
2012-09-04T23:28:12 < TheSeven> just 258 lines of code :P
2012-09-04T23:28:38 < TheSeven> (for comparison: my F4 USB device mode driver is 321 lines)
2012-09-04T23:28:45 < TheSeven> it almost caught up!
2012-09-04T23:28:52 < Laurenceb_> lol
2012-09-04T23:29:05 < Laurenceb_> is it online?
2012-09-04T23:29:39 < TheSeven> not in a public git or anything yet, but lingering somewhere on pastie
2012-09-04T23:30:43 < TheSeven> http://pastie.org/4664332
2012-09-04T23:31:03 < TheSeven> http://pastie.org/4664335
2012-09-04T23:31:30 < Laurenceb_> i meant the i2c lol
2012-09-04T23:31:53 < TheSeven> http://pastie.org/4664339
2012-09-04T23:32:03 < TheSeven> ok, lemme paste the i2c one :P
2012-09-04T23:34:05 < TheSeven> here it is:
2012-09-04T23:34:06 < TheSeven> http://pastie.org/4664350
2012-09-04T23:34:07 < TheSeven> http://pastie.org/4664349
2012-09-04T23:34:08 < TheSeven> http://pastie.org/4664347
2012-09-04T23:34:10 < TheSeven> http://pastie.org/4664346
2012-09-04T23:34:52 < TheSeven> (and yes, this is kinda C++-ish C code)
2012-09-04T23:35:01 < zyp> woah, you're actually using bitfields
2012-09-04T23:35:17 < karlp> living on the edge!
2012-09-04T23:35:31 < TheSeven> zyp: where did you spot one?
2012-09-04T23:35:52 < zyp> in the usb driver
2012-09-04T23:36:06 < TheSeven> ah, right
2012-09-04T23:36:17 < Laurenceb_> ok and i thought my code was hardcore
2012-09-04T23:36:38 < Laurenceb_> blerg
2012-09-04T23:36:45 < Laurenceb_> how did you fix stuck stop bit?
2012-09-04T23:36:46 < TheSeven> yes, after some USB nightmares I decided to put as much semantics as possible into this one
2012-09-04T23:37:13 < TheSeven> Laurenceb_: no idea, it went away by itself while trying to debug it
2012-09-04T23:37:22 < TheSeven> can't reproduce it anymore
2012-09-04T23:37:26 < TheSeven> we'll see if it comes back
2012-09-04T23:37:48 < TheSeven> <Laurenceb_> ok and i thought my code was hardcore << which particular paste is that referring to, and in what sense?
2012-09-04T23:37:57 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-04T23:38:26 < Laurenceb_> http://pastie.org/4664350
2012-09-04T23:39:04 < TheSeven> yeah, if-else monsters :P
2012-09-04T23:39:06 < Laurenceb_> i guess mine is about as long
2012-09-04T23:39:11 < TheSeven> that state machine is just horrible
2012-09-04T23:39:22 < Laurenceb_> yeah
2012-09-04T23:39:41 < Laurenceb_> looks pretty similar to my code
2012-09-04T23:39:53 < karlp> Laurenceb_: what's all that soc/stm32/ stuff?
2012-09-04T23:40:09 < zyp> TheSeven, out of curiosity, have you tested your usb driver on the OTG_HS peripheral with the internal PHY?
2012-09-04T23:40:09 < TheSeven> for reference, the same highlevel interface implemented on top of an abstraction of a generic dumb i2c core: http://pastie.org/4664385
2012-09-04T23:40:12 < TheSeven> looks MUCH simpler to me
2012-09-04T23:40:21 < karlp> and protocl/i2c?
2012-09-04T23:40:44 < TheSeven> zyp: yes
2012-09-04T23:40:53 < Laurenceb_> karlp: thats TheSeven code
2012-09-04T23:41:17 < Laurenceb_> TheSeven: so you have a transaction structure for i2c?
2012-09-04T23:41:25 < TheSeven> now I have one, yes
2012-09-04T23:41:36 < TheSeven> and that complicates matters even more
2012-09-04T23:41:47 < Laurenceb_> its nice tho
2012-09-04T23:41:50 < zyp> TheSeven, I did pretty much what you are doing, and I had to introduce a delay after phy selection before core reset, otherwise the usb core locked up during reset
2012-09-04T23:41:52 < Laurenceb_> ive done it like that
2012-09-04T23:41:56 < TheSeven> for the user, not for the implementer :)
2012-09-04T23:42:01 < Laurenceb_>    while (state->busy);   <- but you block until ready
2012-09-04T23:42:15 < Laurenceb_> i implimented a callback - very badly
2012-09-04T23:42:16 < TheSeven> Laurenceb_: yes, this is running in a singletasked environment
2012-09-04T23:42:26 < Laurenceb_> its just hacked into the code on mine
2012-09-04T23:42:37 < Laurenceb_> i could add function pointers for callbacks
2012-09-04T23:42:47 < TheSeven> zyp: you mean the AHB interface not going idle and similar crap?
2012-09-04T23:42:51 < TheSeven> I've seen that as well...
2012-09-04T23:42:57 < zyp> something like that
2012-09-04T23:43:16 < Laurenceb_> i also did a transaction queue
2012-09-04T23:43:51 < Laurenceb_> useful if you have a load of sensors to read on a uav
2012-09-04T23:44:08 < Laurenceb_> jst read them all then last callback schedules a task to process the data
2012-09-04T23:45:02 < Laurenceb_> or each callback is a bit of dsp filtering code to downsample the sensors
2012-09-04T23:45:21 < Laurenceb_> but mine ended up hackish
2012-09-04T23:45:21 < zyp> IMO it's simpler to just block the thread initiating the transaction
2012-09-04T23:45:30 < zyp> and just queue up transactions in that thread
2012-09-04T23:45:33 < Laurenceb_> yes
2012-09-04T23:45:55 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-04T23:47:00 < TheSeven> zyp: do you use the exact same ordering of the PCGCTL, GUSBCFG and GRSTCTL accesses?
2012-09-04T23:47:58 < TheSeven> also note that all the hw registers are declared volatile in my code base
2012-09-04T23:48:11 < zyp> I'm not touching PCGCTL, it defaults to everything on
2012-09-04T23:48:34 < zyp> and all registers are volatile in my code too
2012-09-04T23:49:58 < Laurenceb_> just reading this stuff is making me feel ill
2012-09-04T23:50:07 < Laurenceb_> waaaayyyy too complex
2012-09-04T23:50:08 < zyp> :p
2012-09-04T23:50:16 < zyp> says you :p
2012-09-04T23:50:21 < Laurenceb_> chibios usb drivers ftw
2012-09-04T23:51:46 < TheSeven> Laurenceb_: to be honest I think that one was easier to get working that this i2c bugger
2012-09-04T23:52:00 < Laurenceb_> hehe probably
2012-09-04T23:52:20 < TheSeven> but sure, it was a couple of days of debugging as well
2012-09-04T23:52:34 < TheSeven> but more of that being highlevel USB problems than HW-level crap like with I2C
2012-09-04T23:53:09 < karlp> all the i2c shit that goes on here makes me very likely to not really ever deliberately choose an i2c periph if there's a spi version available.
2012-09-04T23:53:57 < TheSeven> well that i2c fuckup is stm32-specific at least
2012-09-04T23:54:13 < TheSeven> i2c was one of the easiest things to get going on most other devices I worked with
2012-09-04T23:55:52 < zyp> FWIW I haven't had any I2C related problems since I was done writing my driver
2012-09-04T23:55:55 < karlp> to be honest, spi on the stm32 is rather silly too, the way it handles chipselct for masters is really stupid.
2012-09-04T23:56:14 < zyp> I don't seem to recall writing it was that major effort that everybody else makes it sound like either
2012-09-04T23:59:01 < karlp> just old war stories continually repeated?
2012-09-04T23:59:31 < Laurenceb_> lol
2012-09-04T23:59:36 < Laurenceb_> code warz
--- Day changed Wed Sep 05 2012
2012-09-05T00:00:34 < zyp> I'm aware of the potential pitfalls of my driver and what I should improve when I get around to it, but I'm not in a hurry since it currently works as is
2012-09-05T00:01:21 < Laurenceb_> ive just got a life
2012-09-05T00:01:34 < zyp> the only major issue is the silent ignoring of errors
2012-09-05T00:01:50 <+Steffanx> Laurenceb_ live.zip?
2012-09-05T00:01:55 <+Steffanx> *life
2012-09-05T00:01:56 < Laurenceb_> talking of which at some point im going to have to remove my github stuff :(
2012-09-05T00:01:58 < Laurenceb_> lol
2012-09-05T00:02:09 < zyp> Laurenceb_, by the way, how do you handle errors in the middle of your transaction queue?
2012-09-05T00:02:19 < Laurenceb_> skip and set a flag
2012-09-05T00:02:53 < Laurenceb_> there could be an error callback to retry some vital ones
2012-09-05T00:03:01 < Laurenceb_> but i never properly abstracted it
2012-09-05T00:05:29 < Laurenceb_> my logger project is going to be sold at some point...
2012-09-05T00:05:37 < Laurenceb_> for tons of loverly cash XD
2012-09-05T00:06:06 <+Steffanx> i see a kickstarter project
2012-09-05T00:06:30 < Laurenceb_> wtf no chance
2012-09-05T00:06:48 < zyp> of course not, nobody would want it :p
2012-09-05T00:06:54 < Laurenceb_> license it to GE
2012-09-05T00:07:07 <+Steffanx> GE ..
2012-09-05T00:07:14 < Laurenceb_> general electric
2012-09-05T00:08:35 < Laurenceb_> but i may have to port it to threadX so it looks more pro before something like that
2012-09-05T00:09:10 < Laurenceb_> https://www.youtube.com/watch?v=mQehX0rVYuY&feature=youtube_gdata_player
2012-09-05T00:09:40 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-05T00:12:01 < Laurenceb_> i spot honeywell sensors
2012-09-05T00:53:24 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-05T00:57:33 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Read error: Operation timed out]
2012-09-05T01:01:43 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-05T01:01:43 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-05T01:01:43 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-05T01:01:58 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Read error: Connection reset by peer]
2012-09-05T01:02:13 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-05T01:02:13 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-05T01:02:13 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-05T01:39:57 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-09-05T01:46:43 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-05T01:59:00 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 248 seconds]
2012-09-05T02:03:06 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-09-05T02:09:43 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-05T02:17:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-05T02:23:39 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-05T02:49:14 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-05T02:55:43 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-05T03:03:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-05T03:11:54 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-05T03:18:07 < dongs> why does that guy and eevblog guy look similar
2012-09-05T03:19:59 < cjbaird> Can't stand the eevblog guy. More irritatingly-trying-to-be-popular than Dr Karl.
2012-09-05T03:24:47 < cjbaird> There's a freelance documentary maker putting together a great-looking history of the electronics and microcomputer industry here in Australia (Dick Smith, Microbee, etc.) Lots of awesome interviews with the guys who're our equivalents of Steve Wozniak, etc. And then out of fucking nowhere, EEVBlog Guy trying to pretend being an authority on everthing.. :/
2012-09-05T03:29:00 < karlp> jealous much?
2012-09-05T03:30:51 < dongs> heh
2012-09-05T03:44:01 -!- DaKu is now known as daku
2012-09-05T03:51:40 < cjbaird> wut chu talkin bout willis
2012-09-05T04:10:57 < dongs> http://www.hobbyking.com/hobbyking/store/__27195__Atmel_Atmega_Socket_Firmware_Flashing_Tool.html
2012-09-05T04:17:41 < cjbaird> 20 bux
2012-09-05T04:18:15 < dongs> worth every penny
2012-09-05T04:18:28 < dongs> 5 nanoclips = $8*5 + effort to clip the m
2012-09-05T04:18:54 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-09-05T05:07:03 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-05T05:11:09 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-09-05T05:13:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T05:18:55 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-05T05:46:09 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Quit: Leaving.]
2012-09-05T05:48:44 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-05T06:24:23 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-05T06:32:19 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-05T06:32:25 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-05T06:32:32 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-05T06:55:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T06:58:46 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-09-05T07:01:02 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-05T07:04:36 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-05T08:21:48 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-05T08:30:27 -!- claude is now known as Claude
2012-09-05T08:31:15 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T08:52:15 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 252 seconds]
2012-09-05T09:25:32 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-09-05T09:30:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-05T09:30:56 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-05T09:35:58 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-05T09:43:31 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-05T09:44:38 -!- Thorn [~Thorn@95-29-30-243.broadband.corbina.ru] has joined ##stm32
2012-09-05T09:44:42 -!- Thorn [~Thorn@95-29-30-243.broadband.corbina.ru] has quit [Changing host]
2012-09-05T09:44:42 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-05T09:46:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T09:56:56 -!- dekar [~dekar@212.255.31.95] has joined ##stm32
2012-09-05T09:57:00 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-05T10:00:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-05T10:01:15 -!- dekar_ [~dekar@212.255.36.77] has quit [Ping timeout: 276 seconds]
2012-09-05T10:14:08 -!- Claude is now known as claude
2012-09-05T10:24:34 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-05T11:13:52 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-05T11:26:09 -!- claude is now known as claude_work
2012-09-05T11:39:10 < dongs> dongs
2012-09-05T11:39:45 < dongs> flyback: i see no flash on intel front page
2012-09-05T11:42:37 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-05T11:53:08 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-05T11:53:33 < claude_work> dongs , iirc intels NOR flash line is called numonyx now , and the NAND stuff is done by micron
2012-09-05T11:53:55 < dongs> yeah, not that sorta flash. flyback is trolling as usual
2012-09-05T11:54:12 < claude_work> ah :)
2012-09-05T11:55:00 < claude_work> so i got trolled by proxy-trolling then...
2012-09-05T11:55:18 < dongs> basically
2012-09-05T12:16:27 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-05T12:21:25 -!- jaeckel_ [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-05T12:22:23 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Ping timeout: 240 seconds]
2012-09-05T12:22:24 -!- jaeckel_ is now known as jaeckel
2012-09-05T12:22:24 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 240 seconds]
2012-09-05T12:22:26 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-05T12:33:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 272 seconds]
2012-09-05T12:33:39 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-05T12:48:40 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-05T13:07:05 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has joined ##stm32
2012-09-05T13:07:11 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has left ##stm32 []
2012-09-05T13:07:31 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has joined ##stm32
2012-09-05T13:08:31 < High-Fidelity> is everyone asleep?
2012-09-05T13:08:43 < dongs> yes
2012-09-05T13:09:06 < High-Fidelity> that's the way
2012-09-05T13:10:41 < cjbaird> Bought a stack of gold leaf foil to play with. Currently requiring a naked woman.
2012-09-05T13:12:13 < cjbaird> (or a hot trap who can passibly do the naked thing)
2012-09-05T13:14:26 < High-Fidelity> I can go naked and raise my voice a few octaves?
2012-09-05T13:18:21 < cjbaird> Play Pretend? Mmaybe.
2012-09-05T13:18:49 < High-Fidelity> anyone doing stm32 (or cortex dev in general) development that doesn't use an IDE… vim perhaps?
2012-09-05T13:19:54 < cjbaird> HF: Glourious makefiles are the go for most of us using s-a-m and TNT..
2012-09-05T13:21:04 < High-Fidelity> s-a-m?
2012-09-05T13:21:14 < karlp> sat
2012-09-05T13:21:16 < karlp> s-a-t
2012-09-05T13:21:24 < cjbaird> summon-arm-toolchain, yep.
2012-09-05T13:21:33 < High-Fidelity> ahhh… right
2012-09-05T13:21:41 < karlp> for people using that. just get the arm-gcc from launchpad and call it a day.
2012-09-05T13:23:05 < BrainDamage> you'll have just to check the right complier flags
2012-09-05T13:23:13 < BrainDamage> and write a tiny ldd script with the memory offsets
2012-09-05T13:23:15 < BrainDamage> but that's it
2012-09-05T13:23:48 < cjbaird> karlp: Assuming you're on x86..
2012-09-05T13:23:59 < High-Fidelity> have got an eclipse setup working, but would much prefer to use vim (and associated plugins)
2012-09-05T13:24:20 < High-Fidelity> I'll have to brush off my makefile skills… it's been a while
2012-09-05T13:26:04 < dongs> if I have a struct like: struct { char blah; float foo[12]; int bar; }; and I initialize it like lamestruct_t crap = { 'a', { 1.0, 2.0, 3.0 }, 3 }; how much space in flash does that take? does it dummy up the remaining 11 indexes?
2012-09-05T13:26:05 < cjbaird> The usual trap with makefiles is getting the compiler and linker architecture flags wrong, and getting ARM instructions on a thumb-only device, or M3 code on an M4, and so on.
2012-09-05T13:28:28 < BrainDamage> yes it should reserve the whole array space even if you don't use it all
2012-09-05T13:28:31 < High-Fidelity> cjbaird: I've got a working eclipse + gnuarmeclipse/yagarto environment working, so I can pull the generated makefiles from that for reference… should be OK with any luck
2012-09-05T13:28:51 < High-Fidelity> dongs: yeah… it'll use the whole space
2012-09-05T13:33:39 < dongs> suggestions on making it not do so?
2012-09-05T13:33:54 < karlp> cjbaird: sure, true. s-a-t had been the "recommended way" of getting it working on x86 though, I'd like to suggest that it's no longer necessarily the best
2012-09-05T13:34:30 < High-Fidelity> karlp: any suggestions then?
2012-09-05T13:34:40 < karlp> dongs: don't initialise it at all when you declare it?
2012-09-05T13:34:55 < karlp> then it goes into bss, and you just manually init the fields you want in your own startup code?
2012-09-05T13:35:11 < karlp> High-Fidelity: like I said, arm-gcc from launchpad
2012-09-05T13:35:22 < dongs> karlp: i have ~50 entries to init lke this
2012-09-05T13:35:31 < dongs> but they're all variable number of shit inside [12].
2012-09-05T13:35:49 < High-Fidelity> won't the initialisation code just take up the space instead?
2012-09-05T13:35:53 < karlp> suck it up and let armcc work it's magic? ;)
2012-09-05T13:35:58 < dongs> lul
2012-09-05T13:36:39 < BrainDamage> will the array have to ever fit larger strings?
2012-09-05T13:36:46 < BrainDamage> otherwise you could use []
2012-09-05T13:37:05 < BrainDamage> then the compiler will determine the array size from the initialization, and leave it at that
2012-09-05T13:37:23 < karlp> but he has different instances of the struct, and wants them all to be the same...
2012-09-05T13:37:34 < karlp> with varying amounts of the [] filled in.
2012-09-05T13:37:44 < dongs> yeah, different lengths of data inside, but max 12.
2012-09-05T13:39:18 < BrainDamage> if you want all the structs to be aligned, I think you'll just have to live with the wasted space
2012-09-05T13:40:01 < BrainDamage> the alternative would be some malloc + pointer
2012-09-05T13:47:51 < dongs> fixed it by making pointer + separate init but i am not sure i like it
2012-09-05T13:47:55 < dongs> gonna see about a better solution
2012-09-05T13:48:50 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-05T13:54:14 < karlp> what's that gcc flag to show all the extra things that got defined implicitly?
2012-09-05T13:57:00 < zyp> dongs, move the array to the end of the struct
2012-09-05T13:57:39 < dongs> i know what youre gonna say after that
2012-09-05T13:57:43 < dongs> and no thats not ansiC compliant.
2012-09-05T13:57:48 < dongs> thanks for trying though.
2012-09-05T13:57:52 < zyp> I'm not saying that
2012-09-05T13:58:03 < dongs> you were gonna say make it struct foo[] right?
2012-09-05T13:58:05 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-05T13:58:50 < zyp> the thing is that a variable length array is not allowed in the middle of a struct because that would make the offset of the elements after the array unknown
2012-09-05T13:59:45 < dongs> right
2012-09-05T13:59:46 < dongs> but its ok at the end?
2012-09-05T14:02:42 < zyp> as ok as a variable length array
2012-09-05T14:03:36 -!- daku is now known as DaKu
2012-09-05T14:03:42 < zyp> but I guess C89 doesn't have them at all?
2012-09-05T14:04:01 < dongs> it does
2012-09-05T14:04:09 < dongs> seemes to work, at least compiles with declaration
2012-09-05T14:04:11 < dongs> testing definition
2012-09-05T14:06:23 < dongs> argh
2012-09-05T14:06:25 < karlp> cmsis-dap: http://img829.imageshack.us/img829/1411/img110cd.jpg
2012-09-05T14:06:26 < dongs> of course it wont work
2012-09-05T14:06:33 < karlp> seems the simplecortex people are working on it?
2012-09-05T14:06:45 < dongs> cool crystal
2012-09-05T14:06:52 < karlp> yeah, cool bolts too
2012-09-05T14:07:13 < Laurenceb> what is cmsis-dap?
2012-09-05T14:07:18 < dongs> the new elite debugger
2012-09-05T14:07:22 < Laurenceb> lol
2012-09-05T14:07:24 < Laurenceb> oh
2012-09-05T14:07:25 < karlp> one stlink to rule them all
2012-09-05T14:07:25 < dongs> supported by KEIL
2012-09-05T14:07:27 < Laurenceb> yeah ive seen it
2012-09-05T14:07:59 < BrainDamage> boo @ chip codes etched
2012-09-05T14:13:41 < Laurenceb> filing is messy :P
2012-09-05T14:23:11 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-05T14:23:12 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-05T14:23:12 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-05T14:23:15 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-05T14:44:09  * Laurenceb has been filing holes in hammond enclosures
2012-09-05T14:44:11 < Laurenceb> funtimes
2012-09-05T14:44:59 < dongs> buttholes
2012-09-05T14:46:40 < Laurenceb> http://i.imgur.com/eQcJc.jpg
2012-09-05T14:47:08 < dongs> get a camera that isnt shit
2012-09-05T14:47:09 < dongs> its 2012
2012-09-05T14:47:18 < Laurenceb> lol
2012-09-05T14:47:24 < dongs> even cjbeard's shittycam takes better pics
2012-09-05T14:47:33 < Laurenceb> my nokia is a bit smashed
2012-09-05T14:47:43 < Laurenceb> it fell out of a skip
2012-09-05T14:47:56 < Laurenceb> well thats where the main board is from
2012-09-05T14:48:10 < Laurenceb> so i can have an untraceable phone
2012-09-05T14:48:40 < Laurenceb> the outer case is from a phone i bought
2012-09-05T14:49:09 < Laurenceb> but the camera module is squashed from where it was run over by a car
2012-09-05T14:49:32 < BrainDamage> no cnc mill?
2012-09-05T14:49:48 < Laurenceb> well yes
2012-09-05T14:49:52 < Laurenceb> i cnc the slots
2012-09-05T14:49:59 < Laurenceb> then finish off the edges with file
2012-09-05T14:50:05 < Laurenceb> made 24 of those things so far
2012-09-05T14:51:48 < Laurenceb> but i have a shitton of these enclosures to use now http://i.imgur.com/RVTNz.jpg
2012-09-05T14:52:31 < Laurenceb> should look a bit more pro
2012-09-05T15:06:06 < BrainDamage> dildo controller?
2012-09-05T15:09:16 < Laurenceb> http://a3.ec-images.myspacecdn.com/images01/52/12c5e313d9e43762d695b8150c66dd71/m.gif
2012-09-05T15:17:19 < dongs> looks like german engineering
2012-09-05T15:28:40 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-05T15:30:01 < dongs> commence trolling
2012-09-05T15:34:32 < Laurenceb> <trolling>
2012-09-05T15:38:55 < Laurenceb> http://www.themagpi.com/
2012-09-05T15:39:00 < Laurenceb> ^its sooo mad
2012-09-05T15:43:09 < dongs> is comparing to 0.0f "okay" with literals, not rsult of computation
2012-09-05T15:43:23 < dongs> i.e. im using 0.0f's as terminator for a variable list
2012-09-05T15:46:12 < Laurenceb> http://olimex.wordpress.com/2012/09/04/a10-olinuxino-preliminary-schematic-uploaded-on-github-and-we-are-open-for-discussions/
2012-09-05T15:48:33 < dongs> great
2012-09-05T15:48:47 < dongs> i'd buy it, but they have to come up with a less fucktarded sounding name
2012-09-05T15:49:38 < karlp> what's different with this a10 and the copperwhatsit you pasted yesterday laurence?
2012-09-05T15:50:38 < dongs> wai no pcie?
2012-09-05T15:52:17 < dongs> oh fuck i just realized its made wiht eagle
2012-09-05T15:52:21 < dongs> i feel so fucking sorry for them.
2012-09-05T15:53:14 < Laurenceb> lol
2012-09-05T15:53:28 < Laurenceb> karlp: cubieboard is slightly cheaper
2012-09-05T15:53:36 < Laurenceb> and smaller
2012-09-05T15:53:49 < Laurenceb> but the olimex thing has way more features
2012-09-05T15:53:55 < Laurenceb> i like the power supply
2012-09-05T15:54:01 < Laurenceb> it has a proper smps
2012-09-05T15:54:06 < dongs> Laurenceb: are you on the torrent
2012-09-05T15:54:07 < Laurenceb> with i2c control
2012-09-05T15:54:15 < Laurenceb> what torrent?
2012-09-05T15:54:31 < dongs> < dongs> 27 gigs of taiwanese porn made by some richkid leaked when he took his gaming rig for repair
2012-09-05T15:54:54 < dongs> < dongs> http://184.154.128.246/htm_data/7/1209/793672.html
2012-09-05T15:54:55 < Laurenceb> ...
2012-09-05T15:55:28 < dongs> apparently lots and lots of drugs were involved.
2012-09-05T15:56:05 < Laurenceb> enjoy your fapping dude
2012-09-05T15:56:18 < dongs> cubieboard is already agvailable??
2012-09-05T15:56:45 < Laurenceb> supposedly someone in another channel has one
2012-09-05T15:57:08 < Laurenceb> A10 looks epic
2012-09-05T15:57:14 < zyp> A10 looks like crap
2012-09-05T15:57:16 < Laurenceb> it has _everything_  integrated
2012-09-05T15:57:32 < zyp> everything what?
2012-09-05T15:57:44 < Laurenceb> like audio and vga and stuff
2012-09-05T15:57:48 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-05T15:57:49 < dongs> allwhiner
2012-09-05T15:57:51 < Laurenceb> ethernet
2012-09-05T15:59:00 < Laurenceb> "we still didn’t start to route, just arrange the components on the locations and will do full schematic connections re-check, but I hope we will be able to route everything on 4 layers as we did with A13"
2012-09-05T15:59:02 < Laurenceb> hahaha
2012-09-05T15:59:09 < Laurenceb> so some time to that shipping
2012-09-05T15:59:12 < dongs> right
2012-09-05T15:59:38 < dongs> doing that in eagle is gonna be insane
2012-09-05T15:59:38 < dongs> does eagle even do DDR routing?
2012-09-05T15:59:38 < dongs> trace length matching etc.
2012-09-05T15:59:50 < dongs> wat\
2012-09-05T15:59:55 < dongs> \oh, right
2012-09-05T16:00:17 < Laurenceb> "November-December is our prognosis to have these in production, maybe we could manage to start selling before Christmas"
2012-09-05T16:04:57 < dongs> http://i.imgur.com/WfPby.jpg
2012-09-05T16:05:21 < Laurenceb> lol @ pervert forum
2012-09-05T16:05:27 < Laurenceb> why do they all post 1024
2012-09-05T16:05:35 < dongs> haha no idea
2012-09-05T16:05:40 < dongs> i didnt read past 1st part
2012-09-05T16:06:00 < Laurenceb> that your board?
2012-09-05T16:06:47 < Laurenceb> 1024  硬盘塞满了  拙计啊
2012-09-05T16:06:54 < Laurenceb> is all i have to say
2012-09-05T16:13:19 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-05T16:13:27 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-05T16:16:35 < Laurenceb> now we know where the dongs secret pron forumz are
2012-09-05T16:17:40 < dongs> i started up shittorrent for the first time in like 4 years
2012-09-05T16:17:47 < dongs> didnt know people still used that.
2012-09-05T16:20:25 < Laurenceb> and immediately looked for pron
2012-09-05T16:21:12 < Laurenceb> its illegal in japan :P
2012-09-05T16:22:55  * Laurenceb passes dongs some tissues to clean up
2012-09-05T16:32:38 < dongs> Program Size: Code=38840 RO-data=4852 RW-data=1220 ZI-data=6540
2012-09-05T16:36:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T16:45:37 < Laurenceb> http://upload.wikimedia.org/wikipedia/commons/d/df/Flugtag_Ramstein_14.jpg
2012-09-05T16:45:39 < Laurenceb> oh shit
2012-09-05T16:46:37 < jpa-> any good ideas how to visualize 3D orientation in text format?
2012-09-05T16:46:57 < jpa-> a raw quaternion is not very easy to read.. SORA or euler angles are slightly better, but not by much
2012-09-05T16:49:13 < dongs> use libcaca
2012-09-05T16:49:16 < dongs> draw it in ascii.
2012-09-05T16:49:58 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-05T16:50:04 < jpa-> could work
2012-09-05T16:56:53 < zyp> jpa-, for what purpose?
2012-09-05T16:59:28 < jpa-> zyp: debugging
2012-09-05T16:59:52 < jpa-> (debugging accelerometer + magnetometer system)
2012-09-05T17:00:11 < zyp> why does it need to be text format?
2012-09-05T17:01:15 < zyp> anyway, if the raw quaternion doesn't make enough sense, you'll want to decompose it in a way that makes sense for your application
2012-09-05T17:01:36 < zyp> euler angles is a decomposition that makes sense for planes
2012-09-05T17:02:36 < jpa-> aha! i can just plug "0.6928   -0.0205i    0.0169j   -0.7206k" into wolfram alpha
2012-09-05T17:04:02 < Laurenceb> haha
2012-09-05T17:04:42 < dongs> how?
2012-09-05T17:04:59 < zyp> I made an opengl widget that draws a cube with different colors on each face oriented by the input quaternion that I use when debugging live quaternion data
2012-09-05T17:05:00 < dongs> whoa.
2012-09-05T17:05:26 < jpa-> zyp: did you bother to publish it?
2012-09-05T17:05:39 < dongs> jpa-: catch is, it only works on python on macos.
2012-09-05T17:05:58 < Laurenceb> lol
2012-09-05T17:06:22 < zyp> dongs, actually, I wrote it in C++
2012-09-05T17:07:39 < zyp> jpa-, I don't think I have, but I could dig it up later
2012-09-05T17:08:46 < zyp> otherwise you could write your own, it's not much work
2012-09-05T17:12:04 < zyp> http://paste.jvnv.net/view/EN3aq <- found it
2012-09-05T17:12:22 < zyp> just plug that into your favorite opengl framework
2012-09-05T17:12:55 < zyp> m_w, m_x, m_y and m_z is the quaternion
2012-09-05T17:18:13 < jpa-> thanks
2012-09-05T17:18:30 < dongs> flyback: except it wasnt
2012-09-05T17:18:31 < dongs> its 2012
2012-09-05T17:18:34 < dongs> people use ajax and shit.
2012-09-05T17:22:54 < dongs> zyp, http://bcas.tv/paste/results/ot6VSc48.html y/n/mare
2012-09-05T17:26:08 < karlp> so, between the embedded stlink on discovery boards, and the BMP, what else is in that sort of price range?
2012-09-05T17:26:23 < karlp> packaged stlink is $25, what else is out there of any use/note?
2012-09-05T17:26:24 < dongs> karlp: CMSIS-DAP
2012-09-05T17:26:30 < karlp> not out yet?
2012-09-05T17:26:32 < dongs> Jlink-OB
2012-09-05T17:27:04 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-05T17:27:36 < karlp> was more looking at not OB,
2012-09-05T17:27:52 < karlp> ie, work currently has a couple of my own discovery boards for programming/dbeugging the stm32 board,
2012-09-05T17:28:00 < karlp> I reckon they should "get their own"
2012-09-05T17:28:01 < dongs> well, china makes dongles out of -OB
2012-09-05T17:28:15 < dongs> www.hotmcu.com/usbminijtag-jlink-ob-jtagswd-emulator-with-usb-interface-p-29.html?cPath=3_25 shit like this.
2012-09-05T17:28:42 < karlp> currently tempted to just get the boxed stlink2... but it seems.... wrong
2012-09-05T17:28:58 < dongs> are t hey developing on a real os?
2012-09-05T17:28:59 < dongs> or lunix
2012-09-05T17:29:36 < karlp> I'm "dveloping" on lunix, but there are others on windows, currently they're just programming it though, not dev
2012-09-05T17:29:58 < zyp> dongs, looks ok
2012-09-05T17:33:01 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-05T17:33:05 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-05T17:33:59 <+Steffanx> nu
2012-09-05T17:41:25 < dongs> zyp: thx
2012-09-05T17:43:02 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-05T17:43:03 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-05T17:45:43 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-05T18:07:16 < dongs> http://www.maximintegrated.com/
2012-09-05T18:07:17 < dongs> whoa
2012-09-05T18:32:06 -!- barthess [~barthess@178.154.42.3] has joined ##stm32
2012-09-05T18:34:40 <+Steffanx> Fancy website dongs
2012-09-05T18:47:48 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-05T18:52:54 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-05T18:53:09 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-05T19:00:28 -!- claude_work is now known as claude
2012-09-05T19:09:56 < Laurenceb> its all greenwashed
2012-09-05T19:10:57 < emeb> Green! Beautiful Green. Green will save us from climate change. Buy lots and lots of Green.
2012-09-05T19:16:36 <+Steffanx> Whoa, "Made in Canada" how's that possible?!
2012-09-05T19:17:01 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-05T19:17:02 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-05T19:17:02 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-05T19:17:05 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-05T19:20:14 < Laurenceb> lol
2012-09-05T19:20:28 < Laurenceb> anyone here use openoffice?
2012-09-05T19:20:41 <+Steffanx> used
2012-09-05T19:20:56 < Laurenceb> it keeps crashing when i try to open a particular docx
2012-09-05T19:21:16 <+Steffanx> move to libre office :P
2012-09-05T19:21:50 < Laurenceb> heh
2012-09-05T19:22:25 < karlp> meh, arm-gcc from launchpad 2012q2 produces a binary 16bytes bigger than 2012q1
2012-09-05T19:22:42 < karlp> that was a useful upgrade
2012-09-05T19:22:44 < emeb> libreoffice here...
2012-09-05T19:23:12 < emeb> maybe the newer version is *faster*
2012-09-05T19:23:17 <+Steffanx> karlp .. and how large was the original binary?
2012-09-05T19:23:25 < emeb> 8 bytes
2012-09-05T19:23:31 < karlp> 16484 vs 16500
2012-09-05T19:23:43 < zyp> so what?
2012-09-05T19:23:59 <+Steffanx> It's 16 bytes between life and death!
2012-09-05T19:24:01 < karlp> not much :)
2012-09-05T19:24:04 < emeb> 0.09% increase! My stars!
2012-09-05T19:24:12 < zyp> do a diff on the disassembly, check what changed
2012-09-05T19:24:27 < karlp> meh, too much work :)
2012-09-05T19:24:39 <+Steffanx> lazy :)
2012-09-05T19:24:48 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: leaving]
2012-09-05T19:24:55  * emeb just got the TI M4 Launchpad email.
2012-09-05T19:25:04 < emeb> meh.
2012-09-05T19:25:22 < zyp> oh, so did I
2012-09-05T19:25:34 < zyp> is it worth getting?
2012-09-05T19:25:51 < emeb> For $5, sure. Can't say that I'd use it though.
2012-09-05T19:25:59 < emeb> Shelfware...
2012-09-05T19:26:22 < zyp> it'll probably end up like the MSP430 FRAM kit I got once
2012-09-05T19:26:26 < zyp> still sealed :p
2012-09-05T19:26:58 < zyp> but ok, if it's free shipping I'm getting one
2012-09-05T19:27:09 <+Steffanx> It is
2012-09-05T19:27:52 < emeb> You never know. Hey - it's got an RGB LED on it. Laurenceb can use it for his photo measurements.
2012-09-05T19:28:55 <+Steffanx> You can use it to port laks to, zyp
2012-09-05T19:30:35 < zyp> of course I'll have to port it if I want to use it for anything :p
2012-09-05T19:32:20 < zyp> but yeah, it can be a useful experience to see what I would have to adapt to make it run
2012-09-05T19:34:15 < emeb> "Please enter your company's URL"
2012-09-05T19:34:45 < zyp> huh?
2012-09-05T19:34:53 <+Steffanx> Since when you have CANUCKS in Norway flyback ?
2012-09-05T19:34:58 < emeb> just ignore - he's having a "moment"
2012-09-05T19:35:39 < zyp> flyback, because I don't have any useful purpose for it, so I see no reason to take it out
2012-09-05T19:35:56 < emeb> Oh look - TI estore takes Paypal.
2012-09-05T19:36:04 <+Steffanx> Yes, isn't it nice emeb ?
2012-09-05T19:36:12 < emeb> Indeed.
2012-09-05T19:36:17 < zyp> I noticed after entering my CC number :p
2012-09-05T19:36:24 < emeb> It's not like real money :)
2012-09-05T19:37:25 < emeb> sounds like some sort of ice-skating move.
2012-09-05T19:38:15 < zyp> by the way, what are the specs of this launchpad thing?
2012-09-05T19:38:15 < emeb> slip rings are easy - you can buy them from Adafruit.
2012-09-05T19:38:24 < zyp> I just ordered two, and I haven't even looked at it yet
2012-09-05T19:38:26 < emeb> now he asks.
2012-09-05T19:39:04 < emeb> 80 MHz Cortex M4F + ~256k flash, 32kSRAM, etc.
2012-09-05T19:39:08 < zyp> woah, 6 I2C controllers
2012-09-05T19:39:31 < emeb> SPI, SSI. No I2S though, so no audio for me.
2012-09-05T19:39:36 <+Steffanx> Why needs 6 i2c controllers :S
2012-09-05T19:39:38 <+Steffanx> *Who
2012-09-05T19:40:13 <+Steffanx>  /kick flyback Yeah, what are you doing here ? :P
2012-09-05T19:40:46 < karlp> heh, user manual for the stellaris launchpad links to www.ti.com/stellaris-launchpad/project0 which doesn't exist yet.
2012-09-05T19:41:09 < karlp> the question is, can you use the launchpad to program your own stellaris device, or do you then have to go and buy more tools?
2012-09-05T19:41:34 < karlp> looks like there's debug breakout pads at least
2012-09-05T19:41:41 < zyp> I'm hoping the chips have normal SWD available so I can hook up my BMP
2012-09-05T19:42:10 < zyp> s/hoping/expecting/
2012-09-05T19:43:16 < ntfreak> the launchpad includes the newer ICDI - chatting with TI people to open the protocol up - they are also considering supporting CMSIS-DAP.
2012-09-05T19:44:15 <+dekar> I'd imagine you could do 6 i2c in software at barely any cost anyway
2012-09-05T19:46:05 < zyp> flyback, as far as I can see, all GPIOs are 5V tolerant, so that would be a yes
2012-09-05T19:47:50 < BrainDamage> https://s3.amazonaws.com/data.tumblr.com/tumblr_lxq32wjlAY1qil96wo1_500.jpg
2012-09-05T19:48:21 < emeb> Hah!
2012-09-05T19:48:33 < emeb> hey-seuss.
2012-09-05T19:50:01 < zippe> ntfreak: this is the new lm4s launchpad?
2012-09-05T19:58:50 <+Steffanx>  
2012-09-05T19:59:37 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-05T20:03:59 <+dekar> izua, I made lua far more memory efficient by enabling a ton of hacks that by default are only enabled on 32bit PPC and x86er systems
2012-09-05T20:04:19 <+izua> dude, you need to show me how to do this lua thing on a stm32
2012-09-05T20:04:29 <+izua> you're making me jealous :P
2012-09-05T20:04:30 <+Steffanx> dekar you should also blog about it :P
2012-09-05T20:04:37 <+izua> or that
2012-09-05T20:04:41 <+Steffanx> The things you enabled / disabled etc. :)
2012-09-05T20:04:46 <+izua> or you should post it on our amazing wiki <self promotion>
2012-09-05T20:04:51 < karlp> how big is the footprint now?
2012-09-05T20:05:05 < karlp> and are you doing lua as the OS, or just writing the app in lua instead of C?
2012-09-05T20:06:34 <+dekar> karlp, footprint on flash? around 100kb, containing the compiler and VM
2012-09-05T20:07:00 <+dekar> I did embed lua into my firmware so I don't use it as the OS, but for scripting
2012-09-05T20:08:03 <+dekar> you can transfer LUA scripts over usb, the firmware then compiles them into bytecode and shuts most of the firmware down/frees buffers
2012-09-05T20:08:51 <+dekar> it then starts executing the bytecode in a lua VM, which has around 60kb ram at its disposal
2012-09-05T20:08:57 < BrainDamage> mmm woudn't you gain some code space if you removed the bytecode compiler and instead upload already compiled one?
2012-09-05T20:09:18 <+dekar> I would only gain flash space which I have plenty of
2012-09-05T20:09:42 <+dekar> also I like getting the compiler errors from the actual target, I retargeted stdout through USB
2012-09-05T20:09:55 <+izua> so the compiler doesn't waste memory just by being linked against?
2012-09-05T20:10:36 -!- barthess [~barthess@178.154.42.3] has quit [Quit: Leaving.]
2012-09-05T20:11:13 <+dekar> I don't think so, I haven't tried removing it though. My firmwares memory footprint goes down to 4kb ram during lua execution, I feel like thats good enough anyway
2012-09-05T20:11:31 <+dekar> 800 bytes of those are used by a ringbuffer for stdout
2012-09-05T20:12:52 <+dekar> those lua hacks are just ifdefs: http://pastebin.com/Fvmh3mrP
2012-09-05T20:13:24 <+dekar> all I did was defining them, they work fine on 32bit little endian arm (well besides the big endian ones I guess :P)
2012-09-05T20:15:19 <+dekar> double precision is needed for one of the hacks, switching to float gains no memory compared to double with the hack enabled
2012-09-05T20:15:59 <+dekar> I guess performance sucks on my stm32f1, but that's not a problem in my case
2012-09-05T20:17:41 <+dekar> izua, I guess I will blog/wiki about embedding lua in microcontroller firmwares
2012-09-05T20:18:23 <+izua> what about switching to ints?
2012-09-05T20:18:38 <+dekar> breaks math libs, I felt like keeping those
2012-09-05T20:18:41 <+izua> iirc, lua runs by default with floats
2012-09-05T20:18:52 <+izua> really? aw
2012-09-05T20:18:52 <+dekar> it defaults to double
2012-09-05T20:19:04 <+izua> s/with floats/with !ints
2012-09-05T20:19:08 <+izua> (pick your implementation)
2012-09-05T20:19:46 <+dekar> izua, I implemented the lua stuff for our Java/C# guys, they asked me why the lua OS lib wasn't working (os.date in particular)
2012-09-05T20:20:04 <+Steffanx> yay
2012-09-05T20:20:08 <+izua> lol
2012-09-05T20:20:18 <+dekar> I asked them how they think our device should receive the actual date when running stand alone :)
2012-09-05T20:20:33 <+izua> i think a much better question is "why"
2012-09-05T20:21:22 <+dekar> izua, I felt like staying as compatible to normal lua as possible for my co-workers
2012-09-05T20:22:10 <+dekar> one also copy/pasted some recursive function which made lua run out of stack space really fast
2012-09-05T20:23:13 <+dekar> but overall we're just toggling pins and doing basic string manipulation, no need for performance
2012-09-05T20:23:40 <+dekar> it still toggles LEDs faster than my eyes can see
2012-09-05T20:23:59 < BrainDamage> os.date could arrive from the RTC provided it's enabled
2012-09-05T20:24:21 < BrainDamage> but would need special hooks that are stm32 specific
2012-09-05T20:24:50 <+dekar> BrainDamage, it's a USB device possibly not connected to a PC (stand alone operation), also it has no battery
2012-09-05T20:25:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-05T20:25:32 <+dekar> you can flash lua scripts and it then executes them after boot
2012-09-05T20:31:40 < zippe> The big space/perf win with lua is having your own allocator
2012-09-05T20:32:55 < zippe> You can't use NANTRICK if you are using float rather than double, which made me sad
2012-09-05T20:33:04 < zippe> Flipside, all your numbers are smaller
2012-09-05T20:34:02 -!- dekar_ [~dekar@212.255.24.174] has joined ##stm32
2012-09-05T20:34:06 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-05T20:35:05 <+dekar_> yeah it has to fit pointers into the doubles, NANTRICK doesn't even work on x64 anymore
2012-09-05T20:35:51 -!- dekar [~dekar@212.255.31.95] has quit [Ping timeout: 245 seconds]
2012-09-05T20:35:59 <+dekar_> zippe, I didn't get what you said about the allocator though, lua uses newlibs dlmalloc
2012-09-05T20:36:31 <+dekar_> the garbage collector is just overhead on top of that (and convenience)
2012-09-05T20:39:30 < zippe> Ah, you're suffering with dlmalloc
2012-09-05T20:39:39 < zippe> You can do much better with a custom slab allocator
2012-09-05T20:39:47 <+dekar_> memory fragmentation all the way :)
2012-09-05T20:40:44 < zippe> Since lua allocates lots of objects of the same size, less than you might think
2012-09-05T20:41:39 < zippe> Still, the GC makes me sad
2012-09-05T20:42:00 <+dekar_> I guess, but I feel like I am done. Using those lua hacks and the 60kb ram I can now create an empty lua table and stuff 4000 doubles into it before it finally shuts the VM down telling me it ran out of memory
2012-09-05T20:42:16 < zippe> If it runs your app, all well and good
2012-09-05T20:42:21 < karlp> what f1 are you on with all the ram?
2012-09-05T20:43:14 < zippe> 103VE, I'd guess
2012-09-05T20:43:32 <+dekar_> STM32F103VET6
2012-09-05T20:43:51 < zippe> Though there are variants with 96K
2012-09-05T20:44:02 < zippe> (F/G)
2012-09-05T20:44:19 < zippe> The 103ZG is a crazy thing
2012-09-05T20:45:35 <+dekar_> I probably would have gone for the f4 if I had to redo the project now
2012-09-05T20:45:54 <+dekar_> but it wasn't out back then and the f1 is plenty for the stuff we do
2012-09-05T20:46:55 <+dekar_> doing floating point arithmetic without fpu leaves a bad feeling
2012-09-05T20:47:33 -!- eifeljogger [~eifeljogg@p5DDD6EC5.dip.t-dialin.net] has joined ##stm32
2012-09-05T20:47:59 <+dekar_> though if we ever run into performance issues I'll implement it as a C function
2012-09-05T20:49:56 <+dekar_> zippe, slab allocation sounds nice, hadn't heard of it before. I fear though that it barely works for lua since you can't allocate tables of a set size, instead you have to grow them by inserting into them
2012-09-05T20:51:08 <+dekar_> so when you go from 3999 elements to 4000 elements it probably copies all of it over (I read it usually over-allocates 32 elements to prevent constant re-allocation)
2012-09-05T20:51:58 <+dekar_> so I was even thinking about having a C function returning a pre-allocated lua table of a given size, but I don't have the need for that anymore
2012-09-05T21:03:36 < eifeljogger> Hi! Does anybody use the stmf4discovery, especially the cs43l22? I simply want to route a DAC signal (sinus) through the analog inputs to the earphone output. The signal can be watched at ANI1x and I can write the registers through I2C (I read them back for saftyness) but I get no output. Even the internal beep tone of the cs43l22 doesn't appear. But the original audio example works perfectly. A small example with working analog routing w
2012-09-05T21:04:07 < emeb> Coolness: http://t.co/Lw6uFzNv
2012-09-05T21:05:39 < emeb> eifeljogger: you do understand that in order to get sound out of the CS43L22 you need to send it a stream of data via I2S?
2012-09-05T21:07:11 < eifeljogger> even if I only want to amplify an analogue signal? When I watch the block diagram the analog inputs are behind the DAC
2012-09-05T21:08:59 < emeb> So you're trying to generate a signal via the STM32 on-chip DAC, send it via analog to the CS43L22 and route it through the on-chip mixer to the CS43L22 outputs?
2012-09-05T21:09:48 < eifeljogger> yes, from PA4 and PA5 (DAC1 and DAC2) to AIN1A and B
2012-09-05T21:10:10 < eifeljogger> and then simply in the mixer and to the earphones output
2012-09-05T21:11:57 < emeb> Should work, but I've never messed with it.
2012-09-05T21:12:43 < emeb> You can read/write the I2C interface & confirmed that the registers are set as you desire?
2012-09-05T21:13:03 < zyp> eifeljogger, «Even the internal beep tone of the cs43l22 doesn't appear.» - looks right
2012-09-05T21:13:22 < eifeljogger> Yes, I reread all register contents after writing and showing them on a display - all is perfect
2012-09-05T21:13:40 < eifeljogger> so the writing works
2012-09-05T21:14:11 < zyp> have you set the input levels?
2012-09-05T21:15:14 < emeb> Sometimes these audio chips have mute functions that depend on the presence of input data. Checked for that in the data sheet?
2012-09-05T21:15:37 < zyp> analog signals have no concept of «presence of input data»
2012-09-05T21:16:28 < zyp> eifeljogger, did you set PASSAVOL/PASSBVOL?
2012-09-05T21:17:51 < emeb> zyp: I had some trouble with the CS4370 a while back where sending it streams of 0 data in I2S would activate the mute function and shut down the analog paths.
2012-09-05T21:18:00 < eifeljogger> Yes, I set the Passthrua and b and the volumes.
2012-09-05T21:18:15 < zyp> eifeljogger, and output volume?
2012-09-05T21:18:20 < eifeljogger> Hmm emeb, I didn't found something similar in the data sheet
2012-09-05T21:18:30 < eifeljogger> Yes, output volume is at maximum :-)
2012-09-05T21:18:40 < emeb> eifeljogger: just a thought - probably doesn't apply to this type of chip.
2012-09-05T21:19:00 < eifeljogger> I have activated all mixer inputs and all outputs - but nothing
2012-09-05T21:19:22 < eifeljogger> emeb, thanks for the hint - I will contact cirrus fro this
2012-09-05T21:20:04 < eifeljogger> hmmm, I should give it a try with the audio example and then mix the signals to the embedded audio stream
2012-09-05T21:20:07 < emeb> eifeljogger: try modding the Discovery F4 audio example code - leave the I2S channel running and see if you can activate the beep?
2012-09-05T21:20:18 < eifeljogger> hehe ;-)
2012-09-05T21:20:20 < emeb> yep
2012-09-05T21:20:31 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-05T21:20:40 < emeb> I stripped the audio example down to just a sine generator a while back.
2012-09-05T21:21:06 < emeb> that was... interesting.
2012-09-05T21:21:17 < eifeljogger> So you delivered sinus data as I2S-Stream?
2012-09-05T21:21:23 < emeb> Yes
2012-09-05T21:21:43 < emeb> I had some hiccups getting the CS43L22 going though. Was a bit tricky.
2012-09-05T21:22:26 < eifeljogger> I also use the Codec_Init-Function for configurating the chip - but something seems to be wrong
2012-09-05T21:23:30 < eifeljogger> hmmm, I am looking at the code - there is a define switch between DAC input through PA4 and I2S .... so it should be possible to amplify without a stream
2012-09-05T21:25:41 < zyp> did you set PASSASEL and PASSBSEL correctly?
2012-09-05T21:25:45 < emeb> I do remember there was some seriously strange stuff with the volume levels not matching expectations.
2012-09-05T21:26:49 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-05T21:26:58 < emeb> also had to set the bottom two bits of reg 6
2012-09-05T21:27:40 < eifeljogger> zyp: yes, I set all eight input channels (to be save ;-)
2012-09-05T21:28:35 < eifeljogger> emeb, to 16-Bit  audio data?
2012-09-05T21:30:36 < emeb> yeah
2012-09-05T21:30:46 < emeb> probably not an issue for you.
2012-09-05T21:30:51 < eifeljogger> very strange, huh ? ;-)
2012-09-05T21:31:20 < emeb> I can send you the project if you want to look at it.
2012-09-05T21:32:03 < eifeljogger> That would be a great help :-)
2012-09-05T21:32:48 < eifeljogger> simply send it to: info@electronic-thingks.de
2012-09-05T21:33:11 < eifeljogger> (it's my company)
2012-09-05T21:33:43 < emeb> sent.
2012-09-05T21:34:23 < emeb> Oh - not a sinewave. Just a simple squarewave, but you can load the buffer with anything to test.
2012-09-05T21:36:40 < eifeljogger> thanks :-)
2012-09-05T21:36:57 < emeb> np
2012-09-05T21:36:59 < eifeljogger> Oh, I found this in codec.c - seem sto be from the original example
2012-09-05T21:37:09 < eifeljogger> " If these configurations are removed,
2012-09-05T21:37:09 < eifeljogger>       then a long delay should be added between powering off the Codec and switching
2012-09-05T21:37:09 < eifeljogger>       off the I2S peripheral MCLK clock (which is the operating clock for Codec).
2012-09-05T21:37:09 < eifeljogger>       If this delay is not inserted, then the codec will not shut down properly and
2012-09-05T21:37:09 < eifeljogger>       it results in high noise after shut down. "
2012-09-05T21:37:40 < emeb> funny.
2012-09-05T21:37:43 < eifeljogger> the term "which is the operating clock for the codec"  ...
2012-09-05T21:38:07 < eifeljogger> I currently have no clock there
2012-09-05T21:38:33 < eifeljogger> may be it is essential
2012-09-05T21:38:35 < emeb> Hmmm.
2012-09-05T21:38:39 < emeb> Could be.
2012-09-05T21:39:21 < emeb> So much depends on chip design details that are not revealed in the datasheet.
2012-09-05T21:39:51 < emeb> But I wouldn't be surprised if it needs a clock to do anything at all.
2012-09-05T21:40:11 <+dekar_> OT, did you guys watch Iron Sky yet? http://www.youtube.com/watch?v=He_PWsJqsVY
2012-09-05T21:40:44 < emeb> Heard about it - haven't seen anything but trailers.
2012-09-05T21:41:15 <+dekar_> I feel like this movie is extremely underrated
2012-09-05T21:41:28 < emeb> eifeljogger: and for added excitement, the STM32F4 MCLK outputs shut down when you're not feeding data to the I2S interface.
2012-09-05T21:42:09 < eifeljogger> hmmm
2012-09-05T21:42:44 < eifeljogger> ok, but that should be no problem - delivering always the same value for "no sound" :-)
2012-09-05T21:42:58 < emeb> Yeah - send it zeros.
2012-09-05T21:43:14 < emeb> You can set up the DMA to do that unattended.
2012-09-05T21:43:49 < eifeljogger> And at the end I will do it surely via I2S for better quality - but now I am doing experiments ;-)
2012-09-05T21:43:51 < eifeljogger> yes
2012-09-05T21:44:40 -!- tavish [~tavish@unaffiliated/tavish] has quit [Read error: Connection reset by peer]
2012-09-05T21:45:47 < emeb> eifeljogger: It can be done. A buddy of mine built a 32 voice additive synth w/ touchscreen and voltage control using the Discovery F4.
2012-09-05T21:46:33 < emeb> Used the CS43L22, but found that it was too noisy for really high quality audio. Probably a board layout issue.
2012-09-05T21:48:25 < eifeljogger> The final destination is to record sounds through the internal 12-bit ADCs and give them back with a selectable delay of fine granulation (100ns!). I want to do this with two circular buffers which are written via DMA with the ADC values and a send to I2S (or now to DAC) via another DMA. The "distance" between the DMA pointers then is the delay (clocking will be given with a 100ns-DMA-Clock). Could that work as I think?
2012-09-05T21:50:32 < eifeljogger> Did he use the Speaker outputs? PWM is always noisy
2012-09-05T21:56:30 < emeb> eifeljogger: should work. You'll have some "fun" working out the differences between the sample rate of the on-chip ADCs vs the CS43L22 though.
2012-09-05T21:57:17 < emeb> I've got a board that uses a codec which keeps the sample rates of audio in/audio out identical for that reason: http://ebrombaugh.studionebula.com/synth/stm32f4_codec/index.html
2012-09-05T21:57:19 < zyp> just clock everything from the same source and it'll be fine
2012-09-05T21:58:28 < eifeljogger> Using the "analog way" should work without problems - at least I thought this ;-)
2012-09-05T21:58:57 < emeb> eifeljogger: probably.
2012-09-05T21:59:00 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 248 seconds]
2012-09-05T21:59:13 < emeb> ie on-chip ADC & on-chip DACs.
2012-09-05T21:59:43 < emeb> all you have to deal with then are the 12-bit quantization, which isn't too bad depending on the audio material you're using.
2012-09-05T22:00:15 < emeb> s/are/is/
2012-09-05T22:00:33 < eifeljogger> Oh, that is from small embedded electrets in my ears :-)
2012-09-05T22:00:46 < eifeljogger> so for a first try this should work
2012-09-05T22:01:11 < eifeljogger> later I will use a microphone amplifier chip with gain control
2012-09-05T22:15:57 < eifeljogger> ok, I will try your proposals tomorrow and get this f... chip working - cu tomorrow guys, and thanks for the help!
2012-09-05T22:18:21 -!- barthess [~barthess@178.154.42.3] has joined ##stm32
2012-09-05T22:19:16 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-09-05T22:20:35 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-09-05T22:25:41 < emeb> good luck
2012-09-05T22:26:36 -!- eifeljogger [~eifeljogg@p5DDD6EC5.dip.t-dialin.net] has left ##stm32 []
2012-09-05T22:34:35 < ntfreak> zippe: yes this one: http://www.ti.com/stellaris-launchpad
2012-09-05T23:02:21 -!- barthess [~barthess@178.154.42.3] has quit [Quit: Leaving.]
2012-09-05T23:03:51 -!- barthess [~barthess@178.154.21.100] has joined ##stm32
2012-09-05T23:08:31 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-05T23:09:19 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-05T23:09:19 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-05T23:09:19 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-05T23:14:39 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-05T23:15:13 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-05T23:40:41 < Laurenceb_> https://www.miniand.com/products/Hackberry%20A10%20Developer%20Board#buy
2012-09-05T23:40:47 < Laurenceb_> rpi got properly pwned
2012-09-05T23:41:29 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-05T23:41:42 < karlp> you say that every two days Laurenceb_
2012-09-05T23:42:29 < karlp> also, isn't a10 actually a weaker cheap than the old a9 on the bcm part?
2012-09-05T23:42:41 < karlp> also, double the price
2012-09-05T23:43:17 < Laurenceb_> its allwinner a10
2012-09-05T23:43:22 < Laurenceb_> nothing to do with arm10
2012-09-05T23:43:31 < Laurenceb_> there is no arm10 in fact
2012-09-05T23:43:37 < Laurenceb_> its a cortex a8
2012-09-05T23:43:50 < karlp> ruight, but isn't rpi a cortex a9?
2012-09-05T23:43:56 < karlp> or am I mixiing it up with arm9
2012-09-05T23:44:00 < karlp> which is armv6 or something?
2012-09-05T23:44:05 < Laurenceb_> arm11 on rpi
2012-09-05T23:44:13 < Laurenceb_> yes arm11 is armv6
2012-09-05T23:44:17 < Laurenceb_> its veyr confusing
2012-09-05T23:44:29 < Laurenceb_> armv6==architecture
2012-09-05T23:44:41 < Laurenceb_> arm11 == what arm call it
2012-09-05T23:44:52 < Laurenceb_> a10/whatever is what the manufacturer calls it
2012-09-05T23:46:14 -!- barthess [~barthess@178.154.21.100] has quit [Remote host closed the connection]
2012-09-05T23:47:50 < zyp> the A10 got a better cpu than the bcm, but I'm not sure about the rest
2012-09-05T23:48:37 < zyp> and by todays measure the cortex-a8 is still pretty old and weak
--- Day changed Thu Sep 06 2012
2012-09-06T00:15:00 < emeb> looks like a good set of media I/O, but it'd be nicer if there was some GPIO/etc for interfacing.
2012-09-06T00:15:30 < emeb> Unless you're just building a set-top or game box I don't see what you'd do with this.
2012-09-06T00:15:43 < zyp> some? aren't there a lot?
2012-09-06T00:16:21 < emeb> Don't see any on that board outside of audio/video/networking.
2012-09-06T00:16:40 < zyp> ah, I didn't see that board
2012-09-06T00:16:40 < emeb> OK - 4-pin serial.
2012-09-06T00:17:03 < zyp> that's the third new A10 board I've seen during the last two days
2012-09-06T00:17:19 < emeb> Allwinner's marketroids are busy.
2012-09-06T00:17:23 < zyp> http://cubieboard.org/ <- I thought you were talking about this
2012-09-06T00:18:03 < emeb> <waiting for cubieboard.org>
2012-09-06T00:18:26 < emeb> Ah - nicer.
2012-09-06T00:19:08 < zyp> http://olimex.wordpress.com/2012/09/04/a10-olinuxino-preliminary-schematic-uploaded-on-github-and-we-are-open-for-discussions/ <- and third one is this
2012-09-06T00:19:29 < zyp> «GPIO-1,2,3,4 with total 132 GPIOs available for the user»
2012-09-06T00:19:39 < zyp> so, surely the SoC itself has enough GPIO :p
2012-09-06T00:19:43 < emeb> Even better.
2012-09-06T00:20:46 < emeb> Olimex board looks like a rehash of the iMX boards they're doing.
2012-09-06T00:21:31 < emeb> Nice to see SATA on those boards.
2012-09-06T00:21:44 < Laurenceb_> i want  silent cheap pc
2012-09-06T00:21:54 < emeb> Laurenceb_: word
2012-09-06T00:22:22 < emeb> I've got a little Intel Atom for a music server that would be easily replaced by something like this.
2012-09-06T00:23:43 < zyp> I wonder how much the Calxeda stuff could be had for
2012-09-06T00:24:42 < zyp> but I assume it's a pretty different price range than the one these cheap outdated chinese SoCs are in :p
2012-09-06T00:26:04 < emeb> targeting severs - probably.
2012-09-06T00:26:18 < zyp> not probably, very obviously
2012-09-06T00:26:19 < zyp> :p
2012-09-06T00:26:26 < Laurenceb_> arm is slightly annoying, but if you can install ubuntu its not too painful as a pc replacement
2012-09-06T00:26:58 < Laurenceb_> no flashplayer tho
2012-09-06T00:27:24 < emeb> Laurenceb_: use ffmpeg / mplayer?
2012-09-06T00:27:32 < Laurenceb_> unless theres a way to haxor the android version?
2012-09-06T00:27:35 < Laurenceb_> oh yeah
2012-09-06T00:28:09 < emeb> vlc will usually play .flv files too.
2012-09-06T00:28:21 < zyp> I would love having an ARM based server, but I'm not going to replace my xeon box with some toy bullshit
2012-09-06T00:29:00 < emeb> *nod* - xeon would be hard to replace with an a8 arm.
2012-09-06T00:29:41 < zyp> it's mostly idle, so it's not the cpu power I'm concerned about
2012-09-06T00:30:02 < emeb> I/O BW?
2012-09-06T00:30:07 < zyp> yeah
2012-09-06T00:30:22 < emeb> lots of SATA on those Calxeda parts.
2012-09-06T00:30:24 <+dekar_> I would wait for EFI on ARM, the linux situation atm is horrible on ARM imo
2012-09-06T00:30:40 < emeb> dekar_: you don't like u-boot?
2012-09-06T00:30:43 < emeb> :P
2012-09-06T00:31:00 <+dekar_> I don't like boardfiles
2012-09-06T00:31:10 <+dekar_> flattened device tree isn't that much better imo
2012-09-06T00:31:18 < Laurenceb_> SATA on cubieboard
2012-09-06T00:31:22 < emeb> boardfiles are... so retro.
2012-09-06T00:31:42 < Laurenceb_> why?
2012-09-06T00:31:55 < Laurenceb_> whats the alternative?
2012-09-06T00:32:09 <+dekar_> emeb, I have worked with some freescale SoC lately, the kernel it shipped with had 1000 patch files
2012-09-06T00:32:11 < zyp> my xeon box is my storage server, it saturates the GbE connection easily
2012-09-06T00:32:12 < Laurenceb_> if you want non standard stuff
2012-09-06T00:32:31  * emeb comes from beagleboard - I know boardfiles.
2012-09-06T00:32:38 < zyp> I'm imagining that most ARM board doesn't even come close
2012-09-06T00:32:43 < zyp> boards*
2012-09-06T00:32:55 <+dekar_> emeb, beagle has nice upstream support, you can probably boot a vanilla kernel on that
2012-09-06T00:33:15 < emeb> dekar_: now. 3 yrs ago not so much.
2012-09-06T00:33:40 < emeb> and Angstrom is still a huge bundle of kernel patches.
2012-09-06T00:33:55 <+dekar_> thus -> wait for EFI
2012-09-06T00:33:57 < emeb> yep
2012-09-06T00:36:14 <+dekar_> also the windows CE situation isn't much better, those board support packages seem to be a mess as well
2012-09-06T00:36:21 < Laurenceb_> http://www.instructables.com/id/EMP-shopping-cart-locker/step11/Install-device-on-human/
2012-09-06T00:36:21 < zyp> http://omapzoom.org/?p=kernel/omap.git;a=shortlog;h=refs/heads/p-linux-omap-3.4 <- lots of recent changes in omap branch too
2012-09-06T00:36:36 <+dekar_> I guess that's one of the reasons microsoft pushes for EFI
2012-09-06T00:38:21 < Laurenceb_> how does EFI work to setup the board?
2012-09-06T00:38:43 < Laurenceb_> i eman once everything is set up, how are the hardware drivers "mapped"  to the board?
2012-09-06T00:39:34 < zyp> huh?
2012-09-06T00:39:35 <+dekar_> Laurenceb_, efi passes structs with stuff, you can just call efi function pointers to read files from the filesystem etc
2012-09-06T00:39:55 <+dekar_> print to the display, write to efi non volatile memory
2012-09-06T00:40:03 <+dekar_> even do tcp networking
2012-09-06T00:40:14 < Laurenceb_> oh i see
2012-09-06T00:40:37 < zyp> EFI provides services similar to BIOS
2012-09-06T00:40:50 < Laurenceb_> yeah i dont understand how bios works :P
2012-09-06T00:40:59 < Laurenceb_> so you can call the functions at runtime?
2012-09-06T00:41:01 <+dekar_> usually the operating system unloads the EFI during boot and switches to its own drivers
2012-09-06T00:41:18 <+dekar_> for 3d acceleration etc
2012-09-06T00:41:23 < Laurenceb_> but how are the drivers mapped to the specific board?
2012-09-06T00:41:24 < zyp> Laurenceb_, BIOS used software interrupts to call BIOS routines to do various stuff
2012-09-06T00:41:34 < Laurenceb_> right
2012-09-06T00:41:41 < zyp> Laurenceb_, you call a function to tell you which hardware are present
2012-09-06T00:41:48 < Laurenceb_> oh right
2012-09-06T00:41:53 <+dekar_> Laurenceb_, what do you mean? your board comes with the matching EFI
2012-09-06T00:41:58 < Laurenceb_> yes
2012-09-06T00:42:06 < Laurenceb_> so the EFI matches the board
2012-09-06T00:42:27 < Laurenceb_> but if the runtime is generic, how does it manage to run the hardware
2012-09-06T00:42:34 < Laurenceb_> was my question
2012-09-06T00:42:43 <+dekar_> the EFI comes with all the information about the board hardcoded
2012-09-06T00:42:48 < zyp> it provides a way to let the OS enumerate the hardware
2012-09-06T00:42:57 < Laurenceb_> right
2012-09-06T00:43:03 < Laurenceb_> i kind of understand
2012-09-06T00:43:37 < zyp> on x86, PCI already provides a system for enumerating devices
2012-09-06T00:44:16 < Laurenceb_> yeah, pc architecture seems designed with this sort of stuff in mind
2012-09-06T00:44:18 < emeb> Here's another - shockingly, not an Allwinner: http://www.wandboard.org/
2012-09-06T00:44:28 < zyp> is it a marvell?
2012-09-06T00:44:31 < Laurenceb_> lol daft names
2012-09-06T00:44:42 < zyp> oh, i.MX
2012-09-06T00:44:57 <+dekar_> we have an iMX6 quad core at work
2012-09-06T00:45:00 < BrainDamage> I read that as wangboard
2012-09-06T00:45:07 < emeb> dongboard
2012-09-06T00:45:09 <+dekar_> that's the one with 1k of patches
2012-09-06T00:45:14 <+dekar_> stay away
2012-09-06T00:45:18 < zyp> :p
2012-09-06T00:45:38 < zyp> it's more modern than the allwinner stuff though
2012-09-06T00:46:05 <+dekar_> we have a board file that barely matches our hardware, it desperately polls for stuff that isn't there
2012-09-06T00:47:08 <+dekar_> it's just a mess :/
2012-09-06T00:47:38 < emeb> "dazzling 1GHz processor" - heh
2012-09-06T00:48:09 < zyp> I'm glad I'm not maintaining stuff like that at work
2012-09-06T00:48:27 <+dekar_> I am also glad it isn't my project
2012-09-06T00:50:47 <+dekar_> it just isn't worth the time unless you're like samsung and plan on selling a ton of those
2012-09-06T00:51:59 < Laurenceb_> dont round the corners
2012-09-06T00:52:08 <+dekar_> well if you use the kernel it ships with you can probably _use_ it, but I would be uber annoyed by the fact that upgrading the kernel is virtually impossible
2012-09-06T00:52:26 <+dekar_> try applying those 1000 patches to the latest vanilla kernel ;)
2012-09-06T00:52:57 < emeb> OE has a pretty good patch recipe system for stuff like that.
2012-09-06T00:53:09 < Laurenceb_> ive looked at rpi
2012-09-06T00:53:20 < Laurenceb_> it seems a hacked together mess to get the hardware running
2012-09-06T00:53:29 < emeb> but maintaining the recipes as the underlying kernel changes... oy!
2012-09-06T00:54:08 <+dekar_> Laurenceb_, at least raspi and beagle/panda are mainstream enough to go upstream
2012-09-06T00:54:53 <+dekar_> you will soon be able to compile your device tree blob for the raspi and just pass it to a vanilla kernel via uboot (maybe it's already possible)
2012-09-06T00:57:54 < Laurenceb_> nice
2012-09-06T00:58:27 <+dekar_> well I dislike the "device tree blob" part of it
2012-09-06T00:58:56 <+dekar_> and I don't like uboot too much
2012-09-06T00:59:13 <+dekar_> but it's probably the most pleasant linux-arm experience you can have atm
2012-09-06T01:00:55 <+dekar_> my panda board has weird driver bugs btw, network has weird freezes under load
2012-09-06T01:01:11 <+dekar_> :/
2012-09-06T01:01:27 <+dekar_> but I really don't feel like looking into it
2012-09-06T01:01:39 < zyp> maybe I should try to boot vanilla linux on one of the devboards at work one day I'm bored
2012-09-06T01:02:40 <+dekar_> I tried that on a tegra and a freescale SoC, both times I didn't even get serial output
2012-09-06T01:03:58 <+dekar_> I can't even stand the way you have the package the kernel on ARM systems, I mean zImage and uImage
2012-09-06T01:04:19 < Laurenceb_> better than when i tried to boot blackmagic code on my F1 board
2012-09-06T01:04:21 <+dekar_> EFI on the other hand will just read your filesystem
2012-09-06T01:04:38 < Laurenceb_> instant hardfault  :P
2012-09-06T01:04:59 < zyp> you just did something wrong :p
2012-09-06T01:06:30 < zyp> I've flashed over the bootloader on my OMAP board with a non-working one a few times
2012-09-06T01:07:03 < zyp> that's always fun
2012-09-06T01:10:59 <+dekar_> I like how macbook EFIs offer to download the OS once you replaced the HDD
2012-09-06T01:11:16 <+dekar_> they should make a standard for that, it's incredibly handy
2012-09-06T01:19:48 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-06T01:19:58 < TitanMKD> hi
2012-09-06T01:20:03 <+Steffanx> lo
2012-09-06T01:20:08 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-06T01:20:17 < TitanMKD> i just finished the case for my STM32F4 Debug board ;)
2012-09-06T01:20:30 < TitanMKD> see http://titanmkd.blogspot.com
2012-09-06T01:20:40 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-06T01:20:44 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-06T01:21:05 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-06T01:21:30 < emeb> TitanMKD: nice - how you making it?
2012-09-06T01:21:41 < emeb> Makerbot / equiv?
2012-09-06T01:21:57 < TitanMKD> emeb i did it with sketchup
2012-09-06T01:22:23 < TitanMKD> and it will cost less than 5US$ for a full case ;)
2012-09-06T01:22:43 < TitanMKD> for more details see http://dangerousprototypes.com/docs/Customizing_Sick_of_Beige_cases_for_laser_cutting
2012-09-06T01:23:27 < emeb> ah - laser cut acrylic. top & bottom only.
2012-09-06T01:23:55 < TitanMKD> yes
2012-09-06T01:24:02 < TitanMKD> but will include screw ...
2012-09-06T01:31:47 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-09-06T01:32:18 <+dekar_> why would you want to make a new bus pirate?
2012-09-06T01:32:51 < TitanMKD> dekar_ it is not really a BusPirate it's universal debugger
2012-09-06T01:33:08 < TitanMKD> dekar_ it will be also an ARM JTAG/SWD debugger ;)
2012-09-06T01:33:51 < TitanMKD> just all in one for high speed signals and special features like fast SDIO or Fast LCD/DVI/HDMI with Shield board
2012-09-06T01:33:54 <+dekar_> doesn't the bus pirate do that already?
2012-09-06T01:34:02 < TitanMKD> of course not
2012-09-06T01:34:11 <+dekar_> I means jtag/swd
2012-09-06T01:34:18 < TitanMKD> BP is based on Microchip @ 48Mhz max
2012-09-06T01:34:20 <+dekar_> *meant
2012-09-06T01:34:27 < TitanMKD> here it is ARM M4 @ 168MHz ;)
2012-09-06T01:34:31 <+dekar_> I know, I own one
2012-09-06T01:34:34 < TitanMKD> with 1MB flash and tons of RAM
2012-09-06T01:34:54 <+dekar_> the great thing about the bus pirate isn't the hardware though, it's the software
2012-09-06T01:35:02 < TitanMKD> yes i know
2012-09-06T01:35:14 < TitanMKD> anyway the BusPirate is for an other usage
2012-09-06T01:35:16 <+dekar_> a friend used mine with some special purpose firmware to flash a xilinx cpld once
2012-09-06T01:35:33 < TitanMKD> I plan to use my HW for Portable hacking without the need of a PC
2012-09-06T01:35:41 < TitanMKD> just an Android phone ;)
2012-09-06T01:35:59 <+dekar_> how do you power it?
2012-09-06T01:36:06 < TitanMKD> with LiPO ;)
2012-09-06T01:36:39 < TitanMKD> up to 3A LiPO for at least 1 day of full usage
2012-09-06T01:37:05 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-06T01:38:03 < TitanMKD> dekar_ I also added the RTC quartz ;)
2012-09-06T01:38:04 <+dekar_> I still don't really get it, why not make an android bus pirate instead?
2012-09-06T01:38:34 <+dekar_> the stm32 is too slow for HDMI etc anyway
2012-09-06T01:38:52 < TitanMKD> not really at 168Mhz
2012-09-06T01:39:09 < TitanMKD> you can refresh a screen quickly
2012-09-06T01:39:15 < TitanMKD> especially with DMA ;)
2012-09-06T01:39:33 < TitanMKD> anyway HDMI is not my AIM but it will be possible
2012-09-06T01:39:50 <+dekar_> at low resolution/refresh rate?
2012-09-06T01:40:03 < qyx_> firstly you do not have enough ram to store even single frame
2012-09-06T01:40:56 <+dekar_> TitanMKD, my point is, if you want a shield to do those things anyway, why not make a bus pirate shield or a stand alone logic analyzer
2012-09-06T01:41:18 < TitanMKD> dekar_ how do you manage SPI at 20MHz with a BP ?
2012-09-06T01:41:19 < TitanMKD> tell me
2012-09-06T01:41:32 < TitanMKD> how do you read NandFlash 8/16bits with a BP
2012-09-06T01:41:35 <+dekar_> TitanMKD, you can just slow SPI down as you like
2012-09-06T01:41:52 < TitanMKD> how do you read/write SDIO with 4bits .. at up to 48MB/s
2012-09-06T01:42:06 < TitanMKD> it is why for my needs BP is not enough
2012-09-06T01:42:25 <+dekar_> where do you get 48MB/s of data?
2012-09-06T01:42:26 < TitanMKD> even a fast serial port at 5mbps is not possible to be analyzed with BP
2012-09-06T01:42:53 <+dekar_> TitanMKD, well I also have that open bench logic analyzer (and a bunch of others at my disposal)
2012-09-06T01:43:12 < TitanMKD> dekar_ just using MMC 8bits @ 48MHz ;)
2012-09-06T01:43:12 <+dekar_> if you want to do fast stuff you don't want a micro controller
2012-09-06T01:43:59 <+dekar_> TitanMKD, USB isn't even fast enough
2012-09-06T01:44:07 < TitanMKD> dekar_ i have also the OBLS and it is very good
2012-09-06T01:44:39 < TitanMKD> dekar_ you can use SDIO to exchange data ;)
2012-09-06T01:45:09 <+dekar_> so why do you need the speed to begin with? just slow the communication down and use the bus pirate
2012-09-06T01:45:35 < TitanMKD> i just need to decode in realtime some data using internal memory
2012-09-06T01:46:04 <+dekar_> use some logic analyzer then?
2012-09-06T01:46:10 < TitanMKD> or also to use SPI or SDIO for fast data transfer
2012-09-06T01:47:10 < TitanMKD> anyway i know it is not ultimate board but all the code will be open source and C ansi with HAL
2012-09-06T01:47:22 < TitanMKD> for future board really amazing ;)
2012-09-06T01:47:35 < TitanMKD> (not based on STM32F4 ;)
2012-09-06T01:47:56 <+dekar_> I just feel like in most cases for most people the bus pirate will do the job
2012-09-06T01:48:06 < TitanMKD> of course it does
2012-09-06T01:48:23 < TitanMKD> it is just it will give other stuff to analyze/debug/hack
2012-09-06T01:48:49 <+dekar_> and for those special cases where the bus pirate doesn't your board probably won't do it either
2012-09-06T01:49:04 < TitanMKD> dekar_ ???
2012-09-06T01:49:29 < TitanMKD> my board as 44IO so anyway it can be interfaced with tons of hardware BP cannot
2012-09-06T01:49:48 <+dekar_> well for example I work with multi gigabit optical transceivers
2012-09-06T01:50:06 <+dekar_> your board is far too slow for those
2012-09-06T01:50:17 < TitanMKD> hehe of course
2012-09-06T01:50:23 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-06T01:50:35 < TitanMKD> but for that task it is not the same price ;)
2012-09-06T01:50:42 <+dekar_> so yeah your board might be a magnitude better than the bus pirate, but that doesn't matter usually
2012-09-06T01:51:00 < TitanMKD> my board is < 50US$ and you speak me about stuff requiring hw with FPGA with cost of more than 500 or 1000US$
2012-09-06T01:51:03 <+dekar_> but well, if you enjoy working on it :)
2012-09-06T01:51:38 < TitanMKD> anyway i do it for fun it is not my job
2012-09-06T01:52:56 <+dekar_> your blog says it does I2C up to 400khz, I would imagine it could do it much faster
2012-09-06T01:53:06 < TitanMKD> yes it's in hw
2012-09-06T01:53:17 < TitanMKD> SW I2C can be done at 1mbps ;)
2012-09-06T01:53:43 < TitanMKD> with bitbanging
2012-09-06T01:54:08 < BrainDamage> you have a ratio of line to ;) of 0.5
2012-09-06T01:54:28 < Laurenceb_> haha
2012-09-06T01:54:38 < TitanMKD> anyway I2C is not the most exciting protocol ;)
2012-09-06T01:55:39 < qyx_> i was considering doing s/;)$// ;)
2012-09-06T01:56:22 <+dekar_> well I use I2C a lot and the bus pirate often is too slow
2012-09-06T01:56:35 <+dekar_> as a sniffer I mean
2012-09-06T01:56:50 < TitanMKD> yes for that task fast MCU help a lot ;)
2012-09-06T01:57:29 < TitanMKD> espcially ARM MCU instruction are all 1cycle (except div and some misc inst)
2012-09-06T01:57:42 < TitanMKD> for PIC it requires 2 or 4 cycles ;)
2012-09-06T01:57:57 < TitanMKD> and PIC is not 32bits
2012-09-06T01:58:20 < TitanMKD> anyway i just hate PIC i worked on that and it is too limited for what i want to do
2012-09-06T01:58:31 < BrainDamage> pic32 is mips32
2012-09-06T01:58:45 < TitanMKD> i speak about PIC24 ...
2012-09-06T01:58:56 < TitanMKD> PIC32 is totallyn different and based on MIPS stuff
2012-09-06T01:59:39 < TitanMKD> I speak of course about old PIC12/16/24
2012-09-06T02:01:25 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 244 seconds]
2012-09-06T02:02:15 < TitanMKD> bye
2012-09-06T02:02:24 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-06T02:08:21 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-09-06T02:11:07 <+dekar_> I hate that android accessories have to supply power to the phone
2012-09-06T02:11:32 <+dekar_> it would be so much more useful if the phone would supply the power
2012-09-06T02:13:18 < zyp> sure, phone just have to act as usb host
2012-09-06T02:13:44 < zyp> a lot of the latest ones support that
2012-09-06T02:20:08 < Laurenceb_> http://lemote.kd85.com/
2012-09-06T02:20:19 < Laurenceb_> actually looks decent
2012-09-06T02:21:55 < Laurenceb_> that sound is RMS furiously fapping
2012-09-06T02:27:39 < cjbaird> Still no updates about availability on http://loogson.taobao.com/ ..
2012-09-06T02:28:39 < dongs>   The chip has a 16 bit A/D converter (realistically, 13 bits are usable), for much better analog input resolution than any Arduino compatible board.
2012-09-06T02:28:42 < dongs> "hi we have 16bit adc" .. "but only 13bit is usable"
2012-09-06T02:28:45 < dongs> freescaleis awesome
2012-09-06T02:29:00 < dongs> did yall see this? some retard put a kinetis k20 into tarduino form factor
2012-09-06T02:29:01 < dongs> http://www.kickstarter.com/projects/paulstoffregen/teensy-30-32-bit-arm-cortex-m4-usable-in-arduino-a
2012-09-06T02:29:15 < zyp> yes
2012-09-06T02:29:20 < zyp> weirdest part is bootloader
2012-09-06T02:30:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-06T02:30:28 < zyp> «Instead, to avoid having the bootloader take away valuable space, an overly ambitious plan was started to use another very small chip to hold the bootloader code, and copy it to the Kinetis chip only when needed.»
2012-09-06T02:30:55 < zyp> so they put a nuvoton M0 for that purpose
2012-09-06T02:32:28 < dongs> haha what.
2012-09-06T02:32:52 < Laurenceb_> lmao
2012-09-06T02:33:15 < zyp> I have no idea how it works, I can only imagine that it's using SWD or something to put the bootloader into SRAM and start executing it from there
2012-09-06T02:33:37 < Laurenceb_> stm32f4 should fit
2012-09-06T02:33:58 < Laurenceb_> yeah that 16bit adc kind of sucks
2012-09-06T02:34:08 < qyx_> $22K for .. that?
2012-09-06T02:34:24 < Laurenceb_> from the f3 datasheet it seems the 16bit adc is somewhat better
2012-09-06T02:34:26 < zyp> kind of dumb to go as far as a seperate chip for bootloading, and then not put a full debugger firmware on it
2012-09-06T02:34:35 < Laurenceb_> but limited speed
2012-09-06T02:34:48 < Laurenceb_> aiui you can approach that performance with an f1 and some dsp
2012-09-06T02:34:57 < Laurenceb_> fast adc then downsample
2012-09-06T02:35:02 < dongs> zyp: arduino... who the fuck needs debugging?
2012-09-06T02:35:37 < Laurenceb_> lol
2012-09-06T02:35:39 < zyp> who the fuck would care about reserving a few k of the flash for the bootloader?
2012-09-06T02:35:45 < dongs> The conversion speed is up to 16.6 ksps (kilo-samples per second) for each SDADC when
2012-09-06T02:35:49 < dongs> converting multiple channels and up to 50 ksps per SDADC if only one channel conversion
2012-09-06T02:35:52 < dongs> is used.
2012-09-06T02:35:52 < Laurenceb_> yeah F1 adc at 867ksps or whatever
2012-09-06T02:35:54 < dongs> only problem is it sslow.
2012-09-06T02:35:57 < Laurenceb_> then downsample
2012-09-06T02:36:16 < Laurenceb_> ^on F3 or freescale?
2012-09-06T02:36:17 < dongs> zyp, my  guess is K20 is actually a real fucking trash chip :p
2012-09-06T02:36:20 < dongs> Laurenceb_: f3
2012-09-06T02:36:23 < zyp> the M0 chip only have 4k of flash itself, so the bootloader image can't be very large either way :p
2012-09-06T02:36:23 < Laurenceb_> yeah
2012-09-06T02:36:34 < Laurenceb_> dongs: F1 and oversample is about as good
2012-09-06T02:36:45 < Laurenceb_> the fast adc F3 looks very decent tho
2012-09-06T02:37:11 < dongs> ADC conversion time:
2012-09-06T02:37:11 < dongs> – 1 .s at 56 MHz (1.17 .s at 72 MHz)
2012-09-06T02:37:12 < dongs> yea that
2012-09-06T02:37:16 < dongs> .s = us
2012-09-06T02:37:20 < zyp> no, I misread, it's the 16k variant
2012-09-06T02:37:23 < Laurenceb_> for f1 yes
2012-09-06T02:37:30 < dongs> no thats F3
2012-09-06T02:37:37 < Laurenceb_> no
2012-09-06T02:37:47 < Laurenceb_> thats the 12bit adc on the one with 16biut adc
2012-09-06T02:37:52 < Laurenceb_> look at the other f3
2012-09-06T02:37:57 < Laurenceb_> it has crazy fast adc
2012-09-06T02:38:09 < Laurenceb_> seems it can do 36Msps with interleaved
2012-09-06T02:38:17 < dongs> o shiiiet
2012-09-06T02:39:19 < dongs> neat
2012-09-06T02:39:45 < Laurenceb_> dunno if the memory bus can handle that
2012-09-06T02:39:52 < Laurenceb_> need to read the ref manual
2012-09-06T02:39:59 < Laurenceb_> that doesnt exist yet :(
2012-09-06T02:40:33 < Laurenceb_> M4 has a nices data bus matrix thingy
2012-09-06T02:40:36 < Laurenceb_> *nicer
2012-09-06T02:41:05  * Laurenceb_ zzz
2012-09-06T02:46:09 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-06T03:06:54 -!- tech2077 [~tech2077@adsl-75-53-139-200.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-06T03:37:20 < dongs> put yo dick down dewd
2012-09-06T03:39:39  * emeb still has dreams about being unprepared for college exams
2012-09-06T03:40:32 < emeb> OK navigation / control geniuses - lets see you do this: http://www.youtube.com/watch?feature=player_embedded&v=chPanW0QWhA
2012-09-06T03:41:01 < dongs> why is it being held by an arm
2012-09-06T03:41:38 < emeb> heh - probably because they don't have the vertical stability algos working yet.
2012-09-06T03:44:20 < dongs> god damn i gotta sanitize your fucking url before retweeting it
2012-09-06T03:45:28 < emeb> teh youtubes have some seriously strange URLs
2012-09-06T03:47:08 < dongs> emeb: someone should tell those dudes about wheels.
2012-09-06T03:47:16 < emeb> heh
2012-09-06T03:47:41 < emeb> guess they're going for all terrain applications.
2012-09-06T03:48:04 < qyx_> that alpha dog proto is also cool
2012-09-06T03:48:27 < qyx_> http://www.youtube.com/watch?v=SSbZrQp-HOk
2012-09-06T03:48:29 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-06T03:48:43 < emeb> turn a bunch of those loose on a battlefield and let 'em hunt down the bad guys or something.
2012-09-06T03:49:07 < emeb> same guys I guess.
2012-09-06T03:55:44 < qyx_> hm, traffic info, two cars are going in opposite direction on a highway
2012-09-06T03:55:45 < qyx_> wtf
2012-09-06T04:30:14 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-06T05:46:32 < cjbaird> lol OpenBSD (and Kuro5hin) https://www.kuro5hin.org/story/2011/10/5/1469/54284
2012-09-06T05:48:12 < karlp> posters name starts with troll, and you expect me to care about it?
2012-09-06T05:49:31 < karlp> man, openbsd has it's share of problems, but really? that's the best you can do?
2012-09-06T05:50:11 < cjbaird> OMG KARLP THAT IS THE COLMUNATION OF FIVE YEARS OF RESEARCH INTO OPENBSDH8!
2012-09-06T05:50:30 < cjbaird> Cognitive Dissonance much?
2012-09-06T05:51:49 < karlp> people have hated on theo on openbsd as long as I've know about the existence of openbsd
2012-09-06T05:52:07 < karlp> what's changed to make a hate post from last year humourous today?
2012-09-06T05:52:34 < cjbaird> before then, even. His playing games on the NetBSD codebase was one of the first claims to fame.
2012-09-06T05:54:34 < cjbaird> what has 'last year' go to do with anything?
2012-09-06T05:56:24 < cjbaird> Next, you'll be doing a zippe, and pretending that only 'critical' articles that're vetted by the people being critized have any validity..
2012-09-06T05:56:34 < dongs> cjbaird: trollaxor post?
2012-09-06T05:56:35 < dongs> oh
2012-09-06T05:58:48 < dongs> isnt kuro5hin fucking dead
2012-09-06T05:59:00 < cjbaird> That was what I was thinking..
2012-09-06T05:59:16 < cjbaird> It's only getting like a post every other month or so now.
2012-09-06T05:59:32 < dongs> all of trollaxor shit is on http://www.trollaxor.com/
2012-09-06T06:04:02 < cjbaird> \o/ http://www.trollaxor.com/2011/10/netbsd-designed-to-fail.html  warm fizzies
2012-09-06T06:04:06 < cjbaird> *fuzzies
2012-09-06T06:05:11 < cjbaird> fozzies
2012-09-06T06:06:13 < dongs> and MIPS, POWER, and SPARC are all has-been serving platforms that have been dying for well over a decade
2012-09-06T06:06:16 < dongs> lolol
2012-09-06T06:06:39 < cjbaird> fazzies
2012-09-06T06:08:24 < cjbaird> NetBSD should've kept at the 'Academic BSD' idea that was a big part of its beginnings ("and FreeBSD is the sell-out-get-rich-from-the-dot-coms BSD..") Much of the IPv6 work used Net' as the reference platform.
2012-09-06T06:08:49 < cjbaird> Also the 'Internet Land Speed' efforts too.
2012-09-06T06:10:06 < cjbaird> "4.5BSD" was said more than one time..
2012-09-06T06:20:05 < dongs> so whos got a good embedded printf without floats
2012-09-06T06:20:29 < dongs> and i dont mean the gcc provided trash
2012-09-06T06:20:33 < dongs> that needs sbrk or some other wank
2012-09-06T06:21:00 < dongs> or else im gonna steal avr's rprintf
2012-09-06T06:23:22 -!- tech2077 [~tech2077@adsl-75-53-139-200.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-06T06:28:14 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-06T06:28:15 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-06T06:35:24 -!- tech2077 [~tech2077@75.53.131.18] has joined ##stm32
2012-09-06T06:46:34 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-06T06:48:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-06T07:34:07 < emeb_mac> $0.10 cheaper on ebay?
2012-09-06T07:35:04 < emeb_mac> nope - $0.03 cheaper qty 10
2012-09-06T07:45:45 < dongs> hmm
2012-09-06T07:45:53 < dongs> lets see how much WS2811 costs from my chink supplier
2012-09-06T07:45:58 < dongs> ive got a reel of RGB 5050 leds
2012-09-06T07:46:03 < dongs> maybe i could make something retarded wiht it
2012-09-06T07:46:19 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-06T07:47:33 < dongs> obviously i dont care for that
2012-09-06T07:48:46 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 245 seconds]
2012-09-06T07:48:53 < zippe> dongs: hit up github for xprintf
2012-09-06T07:49:02 < zippe> dongs: MHO, it works fairly good
2012-09-06T07:49:13 < zippe> floating point optional
2012-09-06T07:49:13 < dongs> elm-chan:?
2012-09-06T07:49:20 < dongs> http://elm-chan.org/fsw/strf/xprintf.html
2012-09-06T07:49:21 < dongs> this shit?
2012-09-06T07:49:39 < zippe> what part of "elm-chan" looks like "github"?
2012-09-06T07:49:48 < dongs> shrug, name is same?
2012-09-06T07:49:54 < dongs> i ended up using http://www.sparetimelabs.com/tinyprintf/index.html but I can switch if xprintf is bettar
2012-09-06T07:50:10 < zippe> https://github.com/robbesol/xprintf
2012-09-06T07:50:15 < zippe> lrn2srch
2012-09-06T07:50:35 < dongs> how hte fuck do I use github
2012-09-06T07:50:42 < dongs> found the search box.
2012-09-06T07:50:54 < dongs> https://github.com/robbesol/xprintf
2012-09-06T07:50:55 < dongs> this!?
2012-09-06T07:51:12 < zippe> Correct
2012-09-06T07:52:01 < dongs> i think thats g onna end up being bigger than the shit i found
2012-09-06T07:52:27 < dongs> flyback: that isnt news
2012-09-06T07:52:45 < zippe> You said "decent"
2012-09-06T07:53:02 < zippe> There are lots of craptastic printfs out there
2012-09-06T07:53:09 < dongs> zippe, if I wanted something politically correct I'd just link to printf() in keil microlib.
2012-09-06T07:53:12 < zippe> This one is a reasonable balance between standards compliance and size
2012-09-06T07:53:14 < dongs> but that fucking breaks opensores shit
2012-09-06T07:53:30 < zippe> Why?  Most opensource folks have the (better) printf from newlib
2012-09-06T07:53:37 < dongs> it doesnt link
2012-09-06T07:53:42 < dongs> needs _sbrk or osme other crap
2012-09-06T07:53:50 < zippe> Oh, no, you're just being lame
2012-09-06T07:53:54 < dongs> nope.
2012-09-06T07:54:12 < zippe> what's the actual chain?
2012-09-06T07:54:27 < zippe> Are you linking nosys?
2012-09-06T07:54:42 < zippe> Or better still, where can I check your crap out?
2012-09-06T07:54:54 < zippe> Rather than playing 20 questions
2012-09-06T07:54:57 < zippe> Let's just fix it
2012-09-06T07:55:28 < zippe> is this still baseflight?
2012-09-06T07:56:08 < dongs> ya
2012-09-06T07:56:19 < zippe> which piece(s)?
2012-09-06T07:56:34 < dongs> i didnt commit printf bits yet. it builds, I can do it now, and you can not-link to printf.c/h and use something from gcc.
2012-09-06T07:56:37 < dongs> one sec
2012-09-06T07:57:25 < zippe> You need to not use _X and _Y in gps.c
2012-09-06T07:57:34 < dongs> no kidding.
2012-09-06T07:57:40 < zippe> _ is reserved for the implementation namespace, and you're going to break ctype.h with that
2012-09-06T07:57:48 < dongs> that entire file is full of fail
2012-09-06T07:57:54 < dongs> if it was my decision it wouldnt even be there
2012-09-06T07:58:05 < dongs> GPS is worthless shit anyway
2012-09-06T07:58:12 < zippe> It's not actually fatal by the looks of it
2012-09-06T07:58:56 < dongs> ARGH
2012-09-06T07:58:59 < zippe> ?
2012-09-06T07:59:01 < dongs> googlechoad is broeken again
2012-09-06T07:59:04 < dongs> cant commit wtf
2012-09-06T07:59:20 < zippe> github > g!code
2012-09-06T07:59:31 < zippe> flyback: it's a Japanese thing
2012-09-06T07:59:42 < zippe> flyback: problem is, he's kohai not sempai
2012-09-06T07:59:49 < zippe> it just hasn't sunk in yet
2012-09-06T08:01:45 < dongs> hmm
2012-09-06T08:01:53 < dongs> how would I even use printf() with dump to serial anyway
2012-09-06T08:02:03 < dongs> redirect stdout?
2012-09-06T08:02:46 < zippe> dongs: "with dump to serial"?
2012-09-06T08:02:56 < zippe> dongs: typically you implement putchar or similar
2012-09-06T08:03:07 < dongs> zippe: i mean purpose of printf() in all my shit is to print to uart.
2012-09-06T08:03:09 < dongs> right, i know
2012-09-06T08:03:24 < zippe> Ok, so newlib normally has you implement a bunch of basic stuff like write()
2012-09-06T08:04:12 < zippe> This is a bit over the top, but it's a reasonable example:
2012-09-06T08:04:12 < zippe> https://sites.google.com/site/stm32discovery/open-source-development-with-the-stm32-discovery/getting-newlib-to-work-with-stm32-and-code-sourcery-lite-eabi
2012-09-06T08:04:43 < zippe> It ties directly to the ST libs, so you should be able to virtually cut and paste it
2012-09-06T08:05:30 -!- dekar_ [~dekar@212.255.24.174] has quit [Quit: This computer has gone to sleep]
2012-09-06T08:06:14 < dongs> man
2012-09-06T08:06:20 < dongs> _sbrk is just as aids as I thought it was
2012-09-06T08:06:32 < dongs> this _ebss shit is all gcc dependent
2012-09-06T08:06:44 < zippe> If you're not using malloc, you can just return an error
2012-09-06T08:06:52 < dongs> nope.
2012-09-06T08:06:58 < dongs> if you use sprintf() it actually needs that
2012-09-06T08:07:03 < dongs> because i guess it internally mallocs or something
2012-09-06T08:07:14 < zippe> Why would you sprintf?
2012-09-06T08:07:23 < dongs> just sayin.
2012-09-06T08:07:44 < zippe> just askin'
2012-09-06T08:08:13 < zippe> Also, _ebss is pretty K&R, not very gcc anything
2012-09-06T08:15:34 < dongs> flyback: WS2803 claims 20 meters between devices without buffer, lol. gonna get some to try. in my tests i could never get anything that claimed that kinda distance to properly work
2012-09-06T08:35:11 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-06T08:35:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-06T08:54:05 < DLPeterson> anybody here using a Cortex-M4 and GCC toolchain?
2012-09-06T08:54:27 < dongs> i tried to but it sucked
2012-09-06T08:54:34 < dongs> well, ok i lied, I use m4 with crossworks
2012-09-06T08:54:35 < dongs> which uses gcc
2012-09-06T08:55:50 < DLPeterson> I'm wondering if I need to recompile my toolchain so it generates FP instructions rather than using softfp, or if I can just use a different compilation flag for something that is going on a CM4
2012-09-06T08:57:10 -!- dekar [~dekar@dslb-178-004-030-113.pools.arcor-ip.net] has joined ##stm32
2012-09-06T08:57:13 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-06T08:59:07 < zippe> DLPeterson: yes
2012-09-06T08:59:18 < zippe> DLPeterson: what are you trying to do?
2012-09-06T08:59:31 < DLPeterson> zippe, yes I need to recompile my toolchain or yes I can specify a compilation flag?
2012-09-06T08:59:51 < zippe> it depends on what you want to achieve
2012-09-06T09:00:14 < zippe> e.g. if you want to generate hardware FP instructions, then you need a compiler & libraries built to suit
2012-09-06T09:00:28 < zippe> DLPeterson: which platform are you on?
2012-09-06T09:01:01 < zippe> For Linux or Windows: https://launchpad.net/gcc-arm-embedded
2012-09-06T09:01:03 < DLPeterson> zippe, I'm currently using a CM3.  I have a toolchain that I trust.  I am upgrading my board to an STM32-F407, and I want to ensure hard floating point instructions are generated instead of the software implementations.
2012-09-06T09:01:29 < zippe> The link I just pasted is the GCC branch maintained by ARM
2012-09-06T09:01:39 < DLPeterson> zippe, I'm aware of it.
2012-09-06T09:02:00 < zippe> The binaries on that page include support for all of their current "embedded" architectures
2012-09-06T09:02:25 < zippe> The shortest path to getting hardware FP support for your CM4 is going to be to use those tools.
2012-09-06T09:02:44 < DLPeterson> zippe, It seems like perhaps it all boils down to these two flags passed to the compiler: -mfloat-abi and -mfpu and -fsingle-precision-constant
2012-09-06T09:02:49 < zippe> No
2012-09-06T09:03:00 < dongs> no i dont think so
2012-09-06T09:03:02 < zippe> DLPeterson: it starts by building the toolchain with the right multilib configuration
2012-09-06T09:03:23 < zippe> DLPeterson: perhaps you could start with 'gcc -print-multi-lib'
2012-09-06T09:03:31 < zippe> (where 'gcc' is your compiler driver)
2012-09-06T09:03:56 < zippe> You need this line:
2012-09-06T09:03:56 < zippe> armv7e-m/fpu;@mthumb@mfloat-abi=hard@march=armv7e-m@mfpu=fpv4-sp-d16
2012-09-06T09:04:05 < zippe> or a literal equivalent in the output
2012-09-06T09:04:19 < DLPeterson> my current toolchain gives: .;
2012-09-06T09:04:19 < DLPeterson> thumb;@mthumb
2012-09-06T09:04:22 < DLPeterson> fpu;@mfloat-abi=hard
2012-09-06T09:04:27 < dongs> troll'd
2012-09-06T09:04:33 < dongs> you NEED fpv4-sp-d16 shit
2012-09-06T09:04:35 < zippe> Ok; that's not going to help.
2012-09-06T09:04:35 < dongs> for m4/fpu
2012-09-06T09:05:19 < zippe> So if you are certain that you want to use your specific compiler source base, the easiest route (and I hate to say this, because it's not) is going to be to pull down ARM's bits and extract their multilib config
2012-09-06T09:05:37 < zippe> If you aren't familiar with building GCC this will be quite the nuisance, I'm sorry.
2012-09-06T09:05:58 < DLPeterson> I am familiar with building gcc.  I'm currently using summon-arm-toolchain.
2012-09-06T09:06:02 < zippe> While you're at it, take a look at their build script to see how they stage the compiler build to get the right architecture support.
2012-09-06T09:06:08 < zippe> Ah, ok.
2012-09-06T09:06:09 < DLPeterson> zippe, ok.
2012-09-06T09:06:23 < zippe> Well, you can go yell at Esden.  I told him to jump ship to this fork a long time ago.
2012-09-06T09:06:42 < zippe> I contributed a bunch to SAT before giving up; it was just too fragile for me.
2012-09-06T09:07:03 < zippe> It's not impossible to pull the multilib bits across.
2012-09-06T09:07:18 < zippe> And there are lots of SAT users that will thank you if you contribute the changes back.
2012-09-06T09:07:40 < zippe> Anyway, once you get the compiler happy, the options you want should be obvious from the multilib definition:
2012-09-06T09:07:42 < DLPeterson> zippe, what is the toolchain you are referring to?
2012-09-06T09:07:56 < zippe> -mthumb -mfloat-abi=hard -march=armv7e-m -mfpu=fpv4-sp-d16
2012-09-06T09:08:11 < zippe> DLPeterson: I'm sorry?
2012-09-06T09:08:43 < zippe> Did you mean, which one did I use to get that multilib line?
2012-09-06T09:08:50 < DLPeterson> zippe, "told him to jump ship to this fork" ... which fork?
2012-09-06T09:09:00 < dongs> i'd suspect the gcc-arm-embedded
2012-09-06T09:09:00 < zippe> Oh, to the ARM-maintained gcc
2012-09-06T09:09:21 < dongs> zipe, do ythey have working windows bins
2012-09-06T09:09:31 < zippe> dongs: yes
2012-09-06T09:09:42 < zippe> dongs: that's what we ship to our poor Windows users
2012-09-06T09:09:58 < DLPeterson> zippe, ah, I thought you might have been referring to the one that is in the channel message
2012-09-06T09:10:23 < dongs> TNT?
2012-09-06T09:10:34 < zippe> DLPeterson: sorry, no; last time I looked TNT had no support for anything other than vanilla cm3
2012-09-06T09:11:21 < zippe> Ouch.  SAT is still on 4.5.2, or the 2011 Linaro base?
2012-09-06T09:12:27 < zippe> I'm sorry.  I suggested the Linaro option back when I thought they were going to be doing v7m work as well.
2012-09-06T09:13:04 < zippe> But shortly after that I realised they were only interested in changes for the A profile; it helps some in that the T2 codegen is shared, but much of the optimisation is for the wrong pipeline.
2012-09-06T09:13:12 < zippe> And the VFP pieces are totally different.
2012-09-06T09:14:19 < zippe> DLPeterson: if you are on a Mac, I can point you at a homebrew bottle for the ARM-gcc-embedded toolchain if that helps any.
2012-09-06T09:14:29 < DLPeterson> zippe, I've modified his code to keep up with the latest linaro releases of gcc and gdb and a custom patch of OOCD that I use.
2012-09-06T09:14:51 < DLPeterson> zippe, I'm not on a Mac, I'm on gentoo.
2012-09-06T09:14:56 < zippe> DLPeterson: Fair enough.  It shouldn't be a major drama then to get the multilib pieces folded in.
2012-09-06T09:15:08 < dongs> haha gentoo
2012-09-06T09:15:13 < DLPeterson> haha dongs
2012-09-06T09:15:16 < zippe> The ARM guys have only just started on their 4.7 branch.
2012-09-06T09:15:37 < DLPeterson> zippe, do you notice the code size being larger with the ARM maintained gcc?
2012-09-06T09:15:59 < DLPeterson> for one of my projects the difference was 5kb... 35kb vs. 30kb.
2012-09-06T09:16:28 < dongs> try it with armcc :)
2012-09-06T09:16:30 < dongs> it might be like 15k
2012-09-06T09:20:52 < DLPeterson> dongs, is that what comes with the keil tools?
2012-09-06T09:21:14 < dongs> ya
2012-09-06T09:21:18 < dongs> realview arm or so
2012-09-06T09:21:33 < DLPeterson> do you use it?
2012-09-06T09:21:42 < dongs> but watch out it doesnt support any of gccisms or inline assembly, so if youre hardcore gcc (ab)user you'll be rewriting a lot of code
2012-09-06T09:21:51 < dongs> yes
2012-09-06T09:26:01 < DLPeterson> nah, most of my code doesn't depend in any way on the compiler
2012-09-06T09:36:56 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
2012-09-06T09:37:56 < emeb_mac> DLPeterson: I'm building for STM32F405 with floating point instructions. I use Codesourcery gcc on Linux and -mcpu=cortex-m4 -mfloat-abi=hard on the command line
2012-09-06T09:38:32 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has joined ##stm32
2012-09-06T09:38:39 < emeb_mac> No need to rebuild gcc, no special libs. Disassembly of the .elf files confirms float instructions.
2012-09-06T09:39:30 < DLPeterson> emeb_mac, good to know, thanks.  I'm going to fiddle around with my toolchain for a bit but may end up on ARM GCC or codesourcery
2012-09-06T09:41:46 < emeb_mac> I think there's some setup required in the startup.s file, but that's usually already handled.
2012-09-06T09:42:29 -!- cjbaird [~cjb@ppp121-44-206-194.lns20.syd7.internode.on.net] has left ##stm32 ["Konversation terminated!"]
2012-09-06T09:53:20 < DLPeterson> zippe, you earlier wrote -march=armv7e-m, I'm guessing you meant armv7-m?
2012-09-06T10:02:10 < zyp> no, cortex-m4 is v7e-m
2012-09-06T10:02:14 < zippe> DLPeterson: No,
2012-09-06T10:02:39 < zyp> the e is the other instructions added in M4, apart from the FPU
2012-09-06T10:03:36 < zippe> The other setup required is to enable cp10/cp11 to get the VFP bits going
2012-09-06T10:03:42 < DLPeterson> where is this flag documented?
2012-09-06T10:03:50 < zippe> Which one?
2012-09-06T10:03:50 < DLPeterson> gcc arm page: http://gcc.gnu.org/onlinedocs/gcc/ARM-Options.html
2012-09-06T10:03:54 < DLPeterson> doesn't list it
2012-09-06T10:04:14 < zyp> https://launchpadlibrarian.net/109296007/readme.txt <- here is a good overview
2012-09-06T10:05:24 < zippe> DLPeterson: no idea where or if it's documented
2012-09-06T10:05:32 < zippe> DLPeterson: it's in the source, at least
2012-09-06T10:06:07 < DLPeterson> zippe, huh. I guess maybe it is an arm patch that isn't in the mainline gcc or something
2012-09-06T10:06:21 < DLPeterson> or the documentation in the link I sent isn't up to date
2012-09-06T10:06:35 < DLPeterson> zyp, thanks
2012-09-06T10:06:46 < zippe> It's in mainline; http://gcc.gnu.org/ml/gcc-patches/2009-11/msg00941.html
2012-09-06T10:06:50 < zippe> been around for a while now
2012-09-06T10:07:05 < zippe> I think it was just never added to the docs
2012-09-06T10:08:40 < DLPeterson> zippe, yeah, probably.  Ok, well, I'm now fully convinced that the extra 5kb of code is worth the comfort of knowing that there are a lot of other people who have thought about this harder than I would like too :)
2012-09-06T10:08:48 < DLPeterson> I'm going to switch back to ARM GCC
2012-09-06T10:09:29 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-09-06T10:13:44 < zippe> 8)
2012-09-06T10:15:37 < zippe> dongs: stupid question for you
2012-09-06T10:15:41 < dongs> sup.
2012-09-06T10:15:58 < zippe> There's a character I've seen written a bit like /3\ that doesn't seem to be kana
2012-09-06T10:16:07 < dongs> ah
2012-09-06T10:16:11 < dongs> its some old hiragana shit
2012-09-06T10:16:25 < zippe> Him
2012-09-06T10:16:50 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection]
2012-09-06T10:16:54 < zippe> I went digging through all the hiragana fonts I have as well; I guess I must just be blind or it's a weird form
2012-09-06T10:17:15 < dongs> pretty sure its the one
2012-09-06T10:17:18 < dongs> lemme find it
2012-09-06T10:17:36 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-06T10:17:46 < zippe> Are the / and \ emphasis marks, or just part of the glyph?
2012-09-06T10:17:52 < dongs> U+3091 right?
2012-09-06T10:18:10 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-06T10:18:13 < dongs> its like a ru on top of squiggly shit
2012-09-06T10:18:16 < zippe> All I have is a bitmap
2012-09-06T10:18:20 < dongs> url
2012-09-06T10:18:24 < zippe> sec
2012-09-06T10:19:23 < zyp> sounds more like the kanji for small, with some fancy brush motions for the middle stroke
2012-09-06T10:19:49 < dongs> yeah or that
2012-09-06T10:19:58 < zippe> http://purgatory.org/~msmith/files/hiragana.png
2012-09-06T10:20:04 < dongs> oh
2012-09-06T10:20:06 < dongs> thats just fu
2012-09-06T10:20:07 < dongs> or wahtever
2012-09-06T10:20:11 < dongs> just faggoty written
2012-09-06T10:20:36 < zyp> ah, yep
2012-09-06T10:23:14 < dongs> confirmed by zyp the pro fu reader
2012-09-06T10:24:27 < zyp> yea, I'm reading fu like a badass
2012-09-06T10:25:51 < zippe> Weird.  It doesn't show up in the font browser at all; it's like it's just not a letter at all
2012-09-06T10:26:38 < zyp> ふ
2012-09-06T10:27:02 < zippe> Ah, stupid
2012-09-06T10:27:13 < dongs> http://ja.wikipedia.org/wiki/%E3%81%B5
2012-09-06T10:27:14 < dongs> this shit.
2012-09-06T10:27:18 < zippe> Apparently the "correct" romanisation is "hu"
2012-09-06T10:27:29 < zippe> According to some committee
2012-09-06T10:27:35 < zippe> Thanks much
2012-09-06T10:27:40 < dongs> depends which of the faggot weaboo rules youre following yeah
2012-09-06T10:30:22 < zippe> Just trying to make some sense of this crap.  Hangul is easier, at least.
2012-09-06T10:30:42 < zippe> Kanji totally kills me; I can't even find a decent single-character OCR server
2012-09-06T10:32:22 < zyp> what are you doing?
2012-09-06T10:32:22 < dongs> one of my pals was working on some jap/chinese ocr shit in .NET
2012-09-06T10:32:41 < dongs> zyp: probably playing some eroge
2012-09-06T10:33:58 < zippe> zyp: stretching my brain
2012-09-06T10:35:01 < zyp> I'm mostly wondering what you are planning to do with OCR
2012-09-06T10:35:24 < zippe> turn pdf bitmaps into glyphs I can feed to Google
2012-09-06T10:37:04 < dongs> google?
2012-09-06T10:37:25 < dongs> to google image search or something?
2012-09-06T10:37:26 < zippe> translate
2012-09-06T10:37:28 < dongs> oh
2012-09-06T10:37:46 < dongs> was pdf made with opensores shit and you cant copypaste it because they fucked up the charmap?
2012-09-06T10:38:02 < zippe> No, they're datasheets from a couple of japanese vendors
2012-09-06T10:38:21 < dongs> right, so why cant you copypaste
2012-09-06T10:38:28 < dongs> you can send htem to me for reading, if you want.
2012-09-06T10:38:29 < dongs> ^_^
2012-09-06T10:38:30 < zippe> I figure that it's an interesting way to learn a bit of the language that isn't "where can I find hourly hotel take foreigner"
2012-09-06T10:38:47 < zippe> Nah, the idea is to try to nut some of it out myself.
2012-09-06T10:38:53 < dongs> okay.
2012-09-06T10:39:10 < jpa-> do i have to bring NRST to SWD header with STM32F4? STM32F1 seems to work quite ok with just SWDAT + SWCLK
2012-09-06T10:39:11 < zippe> Separately I'm trying to teach myself enough to not totally embarrass myself as a tourist
2012-09-06T10:39:23 < dongs> jpa-: not "really"
2012-09-06T10:40:49 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-06T10:43:01 < zyp> I'm always amused by my co-travellers embarrasing themselves :)
2012-09-06T10:46:03 < dongs> zyp gets complimented how jouzu he is with his sticks by every jap at the after work nomikai.
2012-09-06T10:46:13 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-06T10:46:24 < zyp> :p
2012-09-06T10:47:38 < zippe> Ha
2012-09-06T10:47:38 -!- claude is now known as claude_work
2012-09-06T10:48:02 < zyp> we were at a restaurant this summer where a friend thought the server were asking «do you understand japanese?» so he answered «yes, a little bit»
2012-09-06T10:48:04 < zippe> I'll settle for accidentally propositioning the waitress, rather than ordering a glass of elephant semen
2012-09-06T10:48:32 < zyp> how the conversation really went: «do you guys live in japan?» «yes, a little bit»
2012-09-06T10:49:13 < zippe> Well, that's perhaps maybe close enough 8)
2012-09-06T10:50:28 < dongs> fo shizzle
2012-09-06T10:55:17 < zyp> another of my friends apparently managed to order some onion or something which was only meant as extra seasoning, except he only ordered that
2012-09-06T10:55:38 < zyp> I'm not sure exactly how that went, I weren't there
2012-09-06T11:05:39 < zippe> Heh
2012-09-06T11:05:59 < zippe> One more for the wizards; there's a kanji like 'seven', but with two cross strokes rather than one
2012-09-06T11:06:04 < zippe> Is that "17" by any chance?
2012-09-06T11:06:32 < zippe> i.e. http://en.wiktionary.org/wiki/%E4%B8%83
2012-09-06T11:12:15 < zyp> you mean the katakana mo?
2012-09-06T11:12:19 < zyp> http://en.wikipedia.org/wiki/%E3%83%A2
2012-09-06T11:18:48 < zippe> Ah! hiragana, just less florid than the font I was using as a reference, thanks
2012-09-06T11:35:06 < ntfreak> jpa-: i would always recommend bringing out srst, if the device gets locked then the only recovery would be by connecting while srst is asserted.
2012-09-06T11:35:57 < dongs> haha..
2012-09-06T11:36:01 < dongs> flyback: that chip is 16c
2012-09-06T11:36:11 < dongs> buying 100 cuz ... i fucking could.
2012-09-06T11:36:55 < zyp> what chip?
2012-09-06T11:37:42 < dongs> WS2811
2012-09-06T11:37:48 < dongs> 3ch led driver
2012-09-06T11:39:03 < zyp> google just turns up arduino people having «headaches» trying to get it working
2012-09-06T11:39:08 < dongs> haha
2012-09-06T11:39:15 < dongs> http://www.world-semi.com/en/Driver/Lighting_LED_driver_chip/WS2811/
2012-09-06T11:40:10 < ABLomas> lightningz!
2012-09-06T11:52:01 -!- tech2077 [~tech2077@75.53.131.18] has quit [Ping timeout: 246 seconds]
2012-09-06T12:42:13 < mansfeld> hi folks! little question about bootloaders for an STM32F100 : from reading app note 2606 it seems like it's only possible to boot into the bootloader with a specific pin setting.. is there no way of doing that in software?
2012-09-06T12:43:21 < mansfeld> (i.e. like most microcontrollers I know do : boot into bootloader by default, if no code received within N ms, then boot from flash)
2012-09-06T12:45:56 < zippe> not with the ROM bootloader, no
2012-09-06T12:46:09 < zippe> easy to do wth your own, of course
2012-09-06T12:46:26 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-06T12:48:54 < mansfeld> right, thanks
2012-09-06T12:49:23 < mansfeld> know of any decent ready-made flash bootloaders then? I'm on a pretty tight schedule for this :)
2012-09-06T12:50:04 < zippe> What host interface?
2012-09-06T12:50:26 < zippe> usb? serial?
2012-09-06T12:50:44 < zippe> There's a fairly tolerable DFU-based version in the libopencm3 distribution
2012-09-06T12:51:00 < zippe> ST's example bootloader is OK; a bit bulky but it works
2012-09-06T12:51:49 < mansfeld> it's a bit of a last minute hack and it would be nice to be able to do updates in the field
2012-09-06T12:52:22 < mansfeld> serial interface
2012-09-06T12:52:26 < zippe> The black magic debug probe has a DFU bootloader (same guy that did the libopencm3 stuff)
2012-09-06T12:52:58 < zippe> Hmm.  You might be interested in https://github.com/PX4/Bootloader
2012-09-06T12:53:23 < mansfeld> nice! thanks a lot
2012-09-06T12:53:25 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 252 seconds]
2012-09-06T13:03:00 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-06T13:33:24 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-06T13:48:11 < dongs> mansfeld: yes you can jump to bootloader from your software
2012-09-06T13:48:34 < dongs> mansfeld: you can get your shit in there once by jtag/bootloader pins, then you can have some command over serial/whatever to jump to rom bootloader.
2012-09-06T13:49:09 < dongs> you have TTL serial on board to USB or whatever?
2012-09-06T14:09:07 -!- tech2077 [~tech2077@adsl-75-53-136-13.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-06T14:18:56 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-06T14:23:54 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-06T14:39:07 < mansfeld> dongs: yes, I'll be connected with a UART
2012-09-06T14:39:17 < dongs> right
2012-09-06T14:39:34 < dongs> then you're OK. jump to bootloader from your code and use the documented ST bootloader prototocl
2012-09-06T14:39:40 < dongs> or the ST flash util or whatever, if you dont wanna write your own flasher
2012-09-06T14:40:03 < dongs> have your board listen on uart for some command that forces a reset
2012-09-06T14:40:08 < dongs> and have it jump to BL from there.
2012-09-06T14:40:49 < dongs> easy way to reset and remember, is write some magic number into BKP sram, force reset, and in reset vector read that magic and compare, if its there, jump to BL.
2012-09-06T14:41:44 < dongs> hopefully you dont care about encrypting firmware etc at this point.
2012-09-06T14:42:11 < mansfeld> right, but this is if I use my own bootloader that I've put in flash memory, yes?
2012-09-06T14:42:16 < zyp> if you are calling the bootloader instead of jumping to it, could you then say «BL to BL»? :p
2012-09-06T14:42:17 < dongs> no
2012-09-06T14:42:21 < dongs> there's a uart bootloader in ROM
2012-09-06T14:42:33 < mansfeld> so I can jump to ROM via software?
2012-09-06T14:42:37 < dongs> of course
2012-09-06T14:42:39 < dongs> its just an addres
2012-09-06T14:43:35 < mansfeld> ahaa, yes that makes sense
2012-09-06T14:43:38 < dongs> http://bcas.tv/paste/results/OUPprU29.html
2012-09-06T14:43:39 < dongs> on F103
2012-09-06T14:43:42 < dongs> with GCC.
2012-09-06T14:43:48 < dongs> check datasheet for where bootloader is on F100.
2012-09-06T14:43:52 < dongs> I suspect in similar or maybe even same place.
2012-09-06T14:44:00 < dongs> but you  can't do this from your app
2012-09-06T14:44:05 < dongs> cause youre gonna have pins wrong and interrupts running etc.
2012-09-06T14:44:15 < dongs> you want to reset and do that jump from reset vector.
2012-09-06T14:44:51 < mansfeld> right
2012-09-06T14:45:58 -!- tech2077 [~tech2077@adsl-75-53-136-13.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-06T14:48:58 < mansfeld> thanks for the advice, I'll try that
2012-09-06T14:49:27 < dongs> that'll be $129.95
2012-09-06T14:49:59 < dongs> fucking gmail keeps screwing with UI
2012-09-06T14:50:06 < dongs> you cant empty spam now without selecting all messages..
2012-09-06T14:50:09 < dongs> how fuCKing useful
2012-09-06T14:51:22 -!- High-Fidelity [~benlyall@123-243-232-159.static.tpgi.com.au] has quit [Quit: High-Fidelity]
2012-09-06T14:51:45 < mansfeld> don't you have "Delete all spam messages now" link?
2012-09-06T14:52:02 < zyp> is there a point in emptying spam?
2012-09-06T14:52:45 < dongs> 20.32 x 40.64mm ABLomas < how is this size for your purposes
2012-09-06T14:52:49 < dongs> mansfeld: not anymore
2012-09-06T14:52:53 < dongs> they got rid of it.
2012-09-06T14:53:07 < dongs> at least in apps.
2012-09-06T14:53:11 < dongs> i see it in my regular gmail still
2012-09-06T14:53:16 < dongs> ugh
2012-09-06T14:53:17 < mansfeld> oh, I still have it, in GApps
2012-09-06T14:54:15 < zyp> so do I
2012-09-06T14:54:41 < zyp> but I've never seen any point in clicking it
2012-09-06T14:55:27 < dongs> ill take a screneshot i get spam next time
2012-09-06T14:55:30 < dongs> it was def gone for m e
2012-09-06T14:55:37 < dongs> so i had to click select all, then the shit appeared.
2012-09-06T14:55:44 < dongs> maybe it doeesnt show up until theres > 1 page of spam
2012-09-06T14:56:03 < dongs> ABLomas: beep
2012-09-06T14:58:08 -!- tech2077 [~tech2077@adsl-75-53-131-118.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-06T15:05:48 < karlp> so arm launchpad gcc's readme file talks about how to use semihosting.  But it only talks about the linker options.
2012-09-06T15:05:55 < karlp> what else do you need to use that?
2012-09-06T15:09:19 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-06T15:10:20 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T15:11:11 < ABLomas> ping
2012-09-06T15:11:39 < dongs> yeah
2012-09-06T15:11:40 < ABLomas> dongs: i think it's OK (if talking about gliders, etc)
2012-09-06T15:11:43 < dongs> 0.32 x 40.64mm ABLomas < how is this size for your purposes
2012-09-06T15:11:46 < ABLomas> wait a sec, i measure my DLG
2012-09-06T15:11:51 < ABLomas> yes, i c, sec
2012-09-06T15:11:52 < ABLomas> ;-)
2012-09-06T15:11:58 < dongs> er 20.32
2012-09-06T15:12:24 < dongs> getting tired of waiting for that promini faggot.
2012-09-06T15:13:37 < dongs> fuck what hteh fuck is wrong with feenode
2012-09-06T15:13:40 < dongs> worthless garbage
2012-09-06T15:17:18 < ABLomas> what's wrong with freenode?
2012-09-06T15:17:21 < ABLomas> ;-)
2012-09-06T15:17:22 < ABLomas> measured
2012-09-06T15:17:23 < ABLomas> http://lag.lt/screens/misc/lektuveliai/dlg_internal_dims.jpg
2012-09-06T15:17:29 < dongs> the fact that you just saw the message
2012-09-06T15:17:31 < ABLomas> so yes, 20mm would be more than enough
2012-09-06T15:17:32 < dongs> i had like 300 seconds lag.
2012-09-06T15:17:41 < ABLomas> no, i was doing different things
2012-09-06T15:17:48 < ABLomas> irssi just hangs in background
2012-09-06T15:17:50 < dongs> no
2012-09-06T15:17:55 < ABLomas> that's only you lagging
2012-09-06T15:17:57 < dongs> this shit was failing
2012-09-06T15:18:03 < ABLomas> ,--8<-
2012-09-06T15:18:03 < ABLomas> |(15:09:19)-!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-06T15:18:04 < dongs> nope
2012-09-06T15:18:06 < ABLomas> anyway
2012-09-06T15:18:07 < dongs> its fucking feenode
2012-09-06T15:18:09 < ABLomas> |(15:10:20)-!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T15:18:10 < dongs> i have no problems on any other network
2012-09-06T15:18:11 < ABLomas> `-->8-
2012-09-06T15:18:21 < dongs> 8==D
2012-09-06T15:18:54 < ABLomas> some years ago we had alias in bash shell called 'updike'
2012-09-06T15:19:08 < ABLomas> so guys who managed systems with 4y+ uptime had longest =)
2012-09-06T15:19:16 < ABLomas> that was... eee... 2001?
2012-09-06T15:19:20 < ABLomas> ;-)
2012-09-06T15:19:42 < dongs> 1094days here. whatever the fuck that is in years.
2012-09-06T15:19:54 < dongs> almost 3.
2012-09-06T15:20:21 < ABLomas> back to board question - so, if you're talking about board for light airframe (autopilot), 20mm should fit everyone (maybe except Elf and others micro.... Who cares about them anyway)
2012-09-06T15:20:32 < ABLomas> but if you're thinking about "full-function" autopilot
2012-09-06T15:20:47 < ABLomas> i suggest to make ~30x60mm PCB
2012-09-06T15:20:53 < ABLomas> due to connectors amount
2012-09-06T15:21:09 < Laurenceb> anyone here driven lcd panels?
2012-09-06T15:21:21 < Laurenceb> i.e. the basic dumb LCD
2012-09-06T15:21:22 < dongs> like real ones?
2012-09-06T15:21:25 < ABLomas> 30mm should fit in every airframe (except maybe light gliders like i did in photo)
2012-09-06T15:21:36 < dongs> whatas a basic dumb lcd
2012-09-06T15:21:45 < dongs> character?
2012-09-06T15:21:48 < Laurenceb> one that takes ac signals aiui
2012-09-06T15:21:49 < dongs> 4bit HD18wahtever shit?
2012-09-06T15:21:50 < Laurenceb> yes
2012-09-06T15:21:51 < Laurenceb> no
2012-09-06T15:21:57 < Laurenceb> dumb lcd with no drivers
2012-09-06T15:21:59 < dongs> huh wat
2012-09-06T15:22:08 < Laurenceb> like in watches etc
2012-09-06T15:22:13 < dongs> o duno then. youll be the f irst
2012-09-06T15:22:17 < dongs> like 7seg?
2012-09-06T15:22:18 < Laurenceb> ill take this as a no
2012-09-06T15:22:20 < Laurenceb> yes
2012-09-06T15:22:24 < dongs> doesnt L152 has lcd driver
2012-09-06T15:22:26 < dongs> use that!
2012-09-06T15:22:30 < Laurenceb> heh
2012-09-06T15:22:36 < dongs> its got like 24? segment driver.
2012-09-06T15:22:37 < dongs> GO.
2012-09-06T15:22:38 < Laurenceb> was wondering if its possible with f1
2012-09-06T15:22:48 < Laurenceb> using pwm
2012-09-06T15:23:10 < Laurenceb> i dont quite understand how it works, i think you just pwm them
2012-09-06T15:23:23 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-06T15:23:26 < Laurenceb> but presumably the L152 uses lower voltage for it
2012-09-06T15:23:33 < Laurenceb> as it only takes like 800mv or something
2012-09-06T15:23:49 < Laurenceb> i was thinking of haxoring an omron m2
2012-09-06T15:23:52 < dongs> http://www.atmel.com/Images/doc2569.pdf
2012-09-06T15:24:13 < Laurenceb> http://www.amazon.co.uk/dp/B003CYK6FA/?tag=hydra0b-21&hvadid=9550951629&ref=asc_df_B003CYK6FA
2012-09-06T15:24:20 < Laurenceb> it has a toshiba 8 bit mcu
2012-09-06T15:26:33 < dongs> looks like pwm yea
2012-09-06T15:26:54 < Laurenceb> surely the dc problem can be solved using some caps
2012-09-06T15:27:08 < Laurenceb> rather than excessively complex antiphase pwm
2012-09-06T15:27:50 < karlp> huh, those kinetis parts are neat, they have the eeprom emulation stuff built in, in some separate block of flash
2012-09-06T15:27:56 < Laurenceb> not sure if i can be bothered to make an f1 board for the omron thingy
2012-09-06T15:28:02 < karlp> and a block of OTP too,
2012-09-06T15:28:05 < karlp> pretty
2012-09-06T15:28:20 < karlp> don't need those eui48+eeprom chips offboard
2012-09-06T15:28:42 < Laurenceb> ABLomas: i made an autopilot in 35x60mm
2012-09-06T15:28:50 < dongs> plz
2012-09-06T15:28:54 < ABLomas> yy, it's all OK
2012-09-06T15:29:01 < ABLomas> i'm not saying it's no possible, etc
2012-09-06T15:29:23 < ABLomas> but it's simply no point to make micro PCB and then connect 32497979 to it
2012-09-06T15:29:34 < Laurenceb> http://www.ultralightnews.com/pilotslounge/photopg/images/dactyl.jpg
2012-09-06T15:29:46 < ABLomas> now using UDB3 clone (with dsPIC on it), probably 28x40
2012-09-06T15:29:59 < ABLomas> and at least 16 different wires incoming
2012-09-06T15:30:15 < ABLomas> guess if anyone could see board behind wires...
2012-09-06T15:30:16 < Laurenceb> https://github.com/Laurenceb/Dactyl/tree/master/Hardware/Assembly
2012-09-06T15:32:23 < ABLomas> making it "micro" is possible of course (good example could be paparazzi lisa/m), but... then you need separate boards for servo, peripherials, everything becomes so complicated...
2012-09-06T15:32:54 < Laurenceb> yeah i integrated everything
2012-09-06T15:33:10 < Laurenceb> just needs servos, power, and antenni
2012-09-06T15:33:11 < ABLomas> offtopic: anyone could suggest small stepper motors with good price (smaller than in common printers, dunno what's average price for them)
2012-09-06T15:33:23 < ABLomas> making second wooden camera gimbal =)
2012-09-06T15:33:57 < ABLomas> "servos" - if it could drive FPV airframe, that means you need ~8 servo connectors
2012-09-06T15:34:08 < ABLomas> also, GPS probably
2012-09-06T15:34:15 < ABLomas> that's a lot already
2012-09-06T15:34:27 < ABLomas> you got telemetry? Another buch of wires
2012-09-06T15:34:50 < ABLomas> remote programming (or waypoints) via radio? another 4 wires
2012-09-06T15:35:28 < ABLomas> dongs already did uber small board for multicopters, which supports octo+gimbal
2012-09-06T15:35:53 < ABLomas> you know, i ended by soldering another board to use with servos (smth "servo and ESC distribution board")
2012-09-06T15:35:58 < zyp> amount of connectors is only relevant if you need dedicated cables for each
2012-09-06T15:36:00 < ABLomas> because original board was full of wires
2012-09-06T15:36:24 < ABLomas> so yes - 8 wires for ESC, 2 wires for servos, not counting telemetry, i2c, etc
2012-09-06T15:37:05 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 260 seconds]
2012-09-06T15:37:11 < ABLomas> see?
2012-09-06T15:37:12 < zyp> my amitions are CAN for stuff like that so I can run a single bus to all that
2012-09-06T15:37:14 < ABLomas> too small :P
2012-09-06T15:37:21 < zyp> can be distributed along with power
2012-09-06T15:37:43 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T15:37:57  * ABLomas slaps dongs 
2012-09-06T15:38:01 < Laurenceb> or just put it all on one pcb
2012-09-06T15:38:01 < ABLomas> get better ISP!
2012-09-06T15:38:03 < Laurenceb> http://www.lolotov.com/wordpress/wp-content/uploads/2009/07/anime-map.jpg
2012-09-06T15:38:29 < zyp> Laurenceb, yes, let's just put all the motors directly on the board too
2012-09-06T15:38:39 < Laurenceb> lol
2012-09-06T15:38:44 < Laurenceb> i wouldnt go that far
2012-09-06T15:39:15 < ABLomas> no problem: http://www.goodluckbuy.com/cjmcu-mwc-flight-controller-mini-4-axis-module-cjmcu328-micro-quadcopter-.html
2012-09-06T15:39:45 < Laurenceb> https://github.com/Laurenceb/Dactyl/blob/master/Hardware/Assembly/photo2.JPG?raw=true
2012-09-06T15:39:45 < dongs> funny level: anniemay
2012-09-06T15:44:08 < dongs> Laurenceb: yes that asks to DONWLOAD
2012-09-06T15:48:18 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-06T15:48:33 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T16:06:33 -!- izua [~izua@188.26.164.75] has joined ##stm32
2012-09-06T16:06:33 -!- izua [~izua@188.26.164.75] has quit [Changing host]
2012-09-06T16:06:33 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-06T16:06:36 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-06T16:07:29  * Oldboy deducts 10 points from Laurenceb for shitty lighting and shadows all over
2012-09-06T16:13:34 < dongs> looks like shitberrypi added 2 mounting holes
2012-09-06T16:20:19 < Laurenceb> http://makeamillion.de/wp-content/uploads/IMG_1942.jpg
2012-09-06T16:20:21 < Laurenceb> omfg
2012-09-06T16:20:26 < Laurenceb> fapfapfapfap
2012-09-06T16:20:37 < Laurenceb>  /sarc
2012-09-06T16:26:14 < dongs> the fuck is that
2012-09-06T16:26:14 < dongs> where are the mounting holes
2012-09-06T16:26:14 < dongs> oh
2012-09-06T16:26:14 < dongs> jesus fuck
2012-09-06T16:26:14 < dongs> tehy're not even on teh CORNERS of pcb?
2012-09-06T16:26:36 < zyp> no, it's like an asshole; right in the middle
2012-09-06T16:39:02 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 252 seconds]
2012-09-06T16:39:52 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T16:44:19 < Laurenceb> http://www.cca.org/d/maze-done.jpg
2012-09-06T16:49:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-06T17:03:42 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 264 seconds]
2012-09-06T17:04:11 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T17:08:03 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Client Quit]
2012-09-06T17:08:14 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T17:09:22 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-143.pptp.stw-bonn.de] has joined ##stm32
2012-09-06T17:11:16 <+izua> Laurenceb: amazing
2012-09-06T17:13:02 -!- claude_work is now known as claude
2012-09-06T17:14:24 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Quit: what the nigger fuck]
2012-09-06T17:14:38 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T17:15:10 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Client Quit]
2012-09-06T17:15:52 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-06T17:19:11 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has quit [Ping timeout: 245 seconds]
2012-09-06T17:27:13 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-06T17:27:16 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-06T17:27:42 <+Steffanx> nu
2012-09-06T17:27:56 < jpa-> hi steffie
2012-09-06T17:28:32 < dongs> nice yiff
2012-09-06T17:29:17 <+Steffanx> dongsie is jealous
2012-09-06T17:29:53 < dongs> lewl
2012-09-06T17:31:13 <+Steffanx> yeye
2012-09-06T17:33:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-06T17:44:38 < Laurenceb> jewlous
2012-09-06T17:45:33 <+Steffanx> jewless maybe
2012-09-06T17:47:31  * phantoxe is now away: gone
2012-09-06T17:48:05  * phantoxe is now away: gone
2012-09-06T17:51:57 < Laurenceb> "* dongs has quit (Quit: what the nigger fuck)"
2012-09-06T17:51:59 < Laurenceb> lmao
2012-09-06T17:53:15 <+Steffanx> You share multiple channels with him?
2012-09-06T17:54:57 < dongs> Laurenceb: i got klined from rcgroups irc for that quit message
2012-09-06T17:55:03  * phantoxe is back from: gone (been away for 6m)
2012-09-06T17:55:13 < dongs> those dudes are fucked.
2012-09-06T17:56:01 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has joined ##stm32
2012-09-06T17:56:07  * Laurenceb is bored
2012-09-06T17:56:14 < Laurenceb> staff training day today
2012-09-06T17:56:25 < Laurenceb> no-one to fill out order forms
2012-09-06T17:56:40 < Laurenceb> so epic mosfet shortage
2012-09-06T17:57:01 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-143.pptp.stw-bonn.de] has quit [Ping timeout: 252 seconds]
2012-09-06T17:57:05 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-06T17:57:09 <+Steffanx> What about going home?
2012-09-06T17:57:22 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has joined ##stm32
2012-09-06T17:57:45 < jpa-> wait, you need someone to help you get some mosfets? :P
2012-09-06T17:58:30 < Laurenceb> <eroomde> i carry i bottle of wine too
2012-09-06T17:58:47 < Laurenceb> <eroomde> good for bottling them
2012-09-06T17:58:59 < Laurenceb> <eroomde> and partly to try and cancel out the effects of Laurenceb's un self-aware tactlessness
2012-09-06T18:01:14 <+Steffanx> blabla?
2012-09-06T18:01:27 < dongs> wtf us tgus bkiggerghyt
2012-09-06T18:01:32 < dongs> i mean
2012-09-06T18:01:34 < dongs> waht the fuck is this bloggery
2012-09-06T18:01:52 <+Steffanx> Not happy with the blog?
2012-09-06T18:02:54 < Laurenceb> apparently 8 years ago i accused someone of stealing my lcd monitor
2012-09-06T18:03:04 < Laurenceb>  (irl)
2012-09-06T18:04:14 <+Steffanx> No, you just removed the gpu driver
2012-09-06T18:16:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-06T18:19:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-06T18:43:58 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-06T18:56:14 < Laurenceb> rofl
2012-09-06T18:56:21 < Laurenceb> rpi is being made by sony
2012-09-06T18:58:20 < jpa-> explains something ;)
2012-09-06T18:58:45 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-06T19:03:40 < zyp> hmm, my back-emf-sensing code is pretty fragile
2012-09-06T19:03:53 < zyp> http://bin.jvnv.net/f/0wW9G.png <- makes fun graphs though
2012-09-06T19:09:31 < karlp> what does that sense? when the motor rotates manually or something?
2012-09-06T19:09:37 < karlp> what is that a graph of?
2012-09-06T19:10:40 <+Steffanx> Laurenceb .. explain
2012-09-06T19:10:51 < zyp> first part is fixed commutation time, then I enable automatic commutation from back-emf sensing
2012-09-06T19:11:19 < zyp> it screws up and locks up the motor
2012-09-06T19:11:41 <+Steffanx> Can be very handy
2012-09-06T19:12:44 < zyp> I've got some old graphs of it working too
2012-09-06T19:12:48 < zyp> http://bin.jvnv.net/f/UErRM.png
2012-09-06T19:13:04 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-06T19:13:40 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-06T19:43:13 < zyp> http://bin.jvnv.net/f/GiRzr.png <- going from a state of too early commutation to too late
2012-09-06T19:44:22 < Laurenceb> plot "data.dat" using 1:2 WITH LINES
2012-09-06T19:44:41 <+Steffanx> python ? something
2012-09-06T19:45:14 < zyp> lines would just make a mess out of the stray samples :p
2012-09-06T19:46:12 <+Steffanx> ok, make it POINTS :)
2012-09-06T19:46:28 < jpa-> it is points? :)
2012-09-06T19:46:54 <+Steffanx> Whatever :P
2012-09-06T19:46:59 <+Steffanx> Little circles
2012-09-06T19:47:22 <+Steffanx> "Little blue circles with a black outline"
2012-09-06T19:47:41 -!- barthess [~barthess@164.177.211.101] has joined ##stm32
2012-09-06T19:47:49 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-06T19:48:28 < zyp> Steffanx, well, you wouldn't see the point if it were a point
2012-09-06T19:48:51 < zyp> if you want to see it, it needs some area, and as soon as it has area it's no longer just a point :p
2012-09-06T19:51:38 < emeb> http://www.loopinsight.com/2012/09/06/shit-apple-fanatics-say/
2012-09-06T19:53:56 <+Steffanx> emeb is it worth to watch it?
2012-09-06T19:54:09 <+Steffanx> I mean .. it's standard blabla
2012-09-06T19:54:20 <+Steffanx> 0
2012-09-06T19:54:40 < emeb> Steffanx: it's about 2mins longer than it needs to be.
2012-09-06T19:57:37 <+Steffanx> 3:46 actually :P
2012-09-06T19:57:43 < emeb> heh
2012-09-06T20:00:41 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-06T20:01:53 -!- dekar [~dekar@dslb-178-004-030-113.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-06T20:49:02 -!- BrainDamage [~yaaic@217.200.200.244] has joined ##stm32
2012-09-06T20:58:44 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-06T21:02:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-06T21:02:48 < TitanMKD> hi
2012-09-06T21:07:36 <+Steffanx> lo
2012-09-06T21:07:58 < BrainDamage> Hate.
2012-09-06T21:10:36 < karlp> love sex intelligence
2012-09-06T21:11:15 < emeb> brain = *pop*
2012-09-06T21:16:00 -!- BrainDamage [~yaaic@217.200.200.244] has quit [Quit: Yaaic]
2012-09-06T21:18:28 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-06T21:21:08 -!- dekar [~dekar@212.255.27.206] has joined ##stm32
2012-09-06T21:21:11 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-06T21:32:50 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-06T21:36:34 -!- eifeljogger [~eifeljogg@p549968F1.dip.t-dialin.net] has joined ##stm32
2012-09-06T21:40:33 < eifeljogger> @emeb &zyp: I solved the CS43L22 prob. It indeed was the missing MCLK signal. It is needed for the charge pumps etc.
2012-09-06T21:40:59 < eifeljogger> so there is no working amplifier without a clk signal
2012-09-06T21:58:16 < emeb> cool
2012-09-06T22:01:22 < Laurenceb_> anyone got an F1 board setup with usb and SD card?
2012-09-06T22:01:38 < Laurenceb_> actually no that might not work - i have sd on spi2
2012-09-06T22:01:55  * Laurenceb_ is trying to debug his usb mass storage
2012-09-06T22:03:06 < Laurenceb_> ive fixed the card disconnecting bug
2012-09-06T22:03:16 < emeb> what was it?
2012-09-06T22:03:19 < Laurenceb_> by reinitialising it :-/
2012-09-06T22:03:23 < Laurenceb_> i havent a clue
2012-09-06T22:03:28 < Laurenceb_> its not power
2012-09-06T22:03:33 < Laurenceb_> power looks fine
2012-09-06T22:03:58 < Laurenceb_> but in the last 30 or so bytes of the last sector before the card fails i get occasional 0x02
2012-09-06T22:04:08 < Laurenceb_> which obviously screws up the data
2012-09-06T22:04:16 < emeb> odd
2012-09-06T22:04:23 < Laurenceb_> so i need to come up with a proper bug fix for the origional problem
2012-09-06T22:04:33 < Laurenceb_> it works if i dont interleave usb and spi dam
2012-09-06T22:04:35 < Laurenceb_> *dma
2012-09-06T22:04:38 < emeb> What's std practice for bypassing sd card sockets?
2012-09-06T22:04:38 < Laurenceb_> but thats slow
2012-09-06T22:04:50 < Laurenceb_> 4u7/10u and 100n aiui
2012-09-06T22:05:01 < Laurenceb_> thats what i have
2012-09-06T22:05:23 < emeb> Serious bypass.
2012-09-06T22:05:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-06T22:06:11 < jpa-> 100n is probably not much use for SD socket
2012-09-06T22:06:36 < jpa-> if you want hotplug, you need more than 10µF, otherwise it'll give a serious spike on your 3.3V rail
2012-09-06T22:06:57 < Laurenceb_> its micro sd
2012-09-06T22:07:03 < emeb> I remember looking at bunny's teardown of SD cards and was surprised to see bypass caps in the cards.
2012-09-06T22:07:22 < Laurenceb_> anyway, power is 100% ok
2012-09-06T22:07:27 < jpa-> Laurenceb_: i doubt that makes any difference
2012-09-06T22:07:30 < Laurenceb_> checked with 5Gsps scope
2012-09-06T22:07:52 < Laurenceb_> no way theres anything bad hiding in there
2012-09-06T22:08:11 < emeb> microwaves...
2012-09-06T22:08:18 < Laurenceb_> if i turn off the dma/usb interleaving its 100% fine
2012-09-06T22:08:57 < Laurenceb_> which is... odd
2012-09-06T22:09:13 < jpa-> does SPI clock rate make any difference to the problem?
2012-09-06T22:09:18 < emeb> suggests that internal HW or SW is generating crosstalk?
2012-09-06T22:09:20 < Laurenceb_> no, tried that
2012-09-06T22:09:25 < Laurenceb_> hmm
2012-09-06T22:09:43 < Laurenceb_> slow spi clk and problem is unchanged
2012-09-06T22:09:52 < emeb> build timing deadzone into usb/dma interleave?
2012-09-06T22:10:20 < jpa-> hmm, slow spi clock kind of rules out most crosstalk / waveform issues
2012-09-06T22:10:26 < Laurenceb_> yeah
2012-09-06T22:10:40 < jpa-> Laurenceb_: you know that it occurs in the SPI interface and not USB interface?
2012-09-06T22:10:41 < Laurenceb_> as long as spi comms overlaps with usb at any point its unreliable
2012-09-06T22:10:49 < Laurenceb_> yes
2012-09-06T22:11:00 < Laurenceb_> ive set breakpoins and dumped corrupted buffers
2012-09-06T22:11:32 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/memory.c
2012-09-06T22:11:32 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 272 seconds]
2012-09-06T22:11:39 < Laurenceb_> thats the only st code ive changed
2012-09-06T22:11:56 < Laurenceb_> i made it call fatfs card access code
2012-09-06T22:12:02 < Laurenceb_> which works fine
2012-09-06T22:12:24 < Laurenceb_> its when i change the functions slightly to allow timing overlap with usb that i get the issues
2012-09-06T22:12:26 < jpa-> dso quad uses SPI + USB mass storage, but it doesn't interleave them
2012-09-06T22:12:35 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/memory.c#L75
2012-09-06T22:12:53 < Laurenceb_> if i change that line to wait for completion i have no issue
2012-09-06T22:12:57 < Laurenceb_> but its slow :(
2012-09-06T22:13:42 < Laurenceb_> wondering if its something in my code, but i cant see it
2012-09-06T22:13:55 < Laurenceb_> usb doesnt use dma as it has dedicated sram right?
2012-09-06T22:14:01 < Laurenceb_> so its not dma conjection
2012-09-06T22:14:05 < Laurenceb_> *conjestion
2012-09-06T22:14:08 < jpa-> try dumping the buffer addresses to serial port so that you know you aren't overwriting it too soon?
2012-09-06T22:14:44 < Laurenceb_> buffer addresses?
2012-09-06T22:14:56 < Laurenceb_> oh - i allocate the buffer
2012-09-06T22:15:08 < Laurenceb_> hmm if there was something else overlapping the buffer
2012-09-06T22:18:59 < Laurenceb_> hmm yeah i made sure to malloc a lot more space for the buffer
2012-09-06T22:19:10 < Laurenceb_> so its not being overwritten
2012-09-06T22:21:15 < Laurenceb_> if i had a decent logic anyliser i could see whats actually on the bus
2012-09-06T22:24:51 < TitanMKD> Laurenceb buy the OLS
2012-09-06T22:25:09 < TitanMKD> anyway OLS is too short in memory for long run and fast signal > 10MHz
2012-09-06T22:26:03 < TitanMKD> Laurenceb your problem is copy from USB -> microSD is slow ?
2012-09-06T22:26:18 < Laurenceb_> yeah i need to shift large files over
2012-09-06T22:26:24 < TitanMKD> I ask that because microSD are very slow for little transfer
2012-09-06T22:26:25 < Laurenceb_> spi is 18mhz
2012-09-06T22:26:41 < Laurenceb_> the stm is still the limiting factor
2012-09-06T22:26:50 < TitanMKD> I have done tons of benchmark on lot of SD and the conclusion is big buffer win
2012-09-06T22:27:13 < TitanMKD> or even better use multiple read/write to reach top speed else it stuck to 1MB
2012-09-06T22:27:51 < TitanMKD> of course if you use USB Full Speed it will be limited by 1MB/12Mbits of the USB
2012-09-06T22:28:47 < TitanMKD> Laurenceb the best for logger is to use nandflash ;) faster than sd but requires to do ECC by sw
2012-09-06T22:29:11 < Laurenceb_> yeah i cant be bothered with that
2012-09-06T22:29:16 < Laurenceb_> micro sd is very cheap
2012-09-06T22:29:21 < Laurenceb_> im using 4GB cards
2012-09-06T22:30:18 -!- eifeljogger [~eifeljogg@p549968F1.dip.t-dialin.net] has left ##stm32 []
2012-09-06T22:30:25 < TitanMKD> but for temperature < -40°C only nandFlash work
2012-09-06T22:30:27 < emeb> seems like it's getting hard to find 2GB, 4GB cards anymore.
2012-09-06T22:30:39 < emeb> costs too low for most retailers to bother.
2012-09-06T22:30:57 < Laurenceb_> my local store only has 1 and 2 GB cards :(
2012-09-06T22:32:39 < TitanMKD> Laurenceb just for fun HighSpeed NandFlash reach 200MB/s ;)
2012-09-06T22:32:54 < TitanMKD> but anyway with software ECC it reach 10MB/s ;)
2012-09-06T22:32:59 < Laurenceb_> guess ill try and get my hands on a logic anylizer
2012-09-06T22:33:08 < Laurenceb_> yeah but usb maxes out at 12mbps
2012-09-06T22:33:45 < Laurenceb_> but i suspect its happening onboard the ctm32
2012-09-06T22:33:47 < Laurenceb_> *stm
2012-09-06T22:34:09 < Laurenceb_> im not sure how to setup the LA either
2012-09-06T22:34:26 < Laurenceb_> as sometimes you have to transfer many MB of data before it fails
2012-09-06T22:34:45 < TitanMKD> I plan to interface FTDI FT2232H to have better transfer ;)
2012-09-06T22:34:54 < emeb> heisenbug - those are fun.
2012-09-06T22:34:58 <+izua> dekar says those chips are awesome
2012-09-06T22:35:28 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-06T22:35:33 <+izua> he totally loves them
2012-09-06T22:35:36 < TitanMKD> Laurenceb in that case you need something like Saleae logic to have tons of data
2012-09-06T22:35:52 < Laurenceb_> hmm im not sure if could cope
2012-09-06T22:35:59 < Laurenceb_> 18mhz for several minutes
2012-09-06T22:36:04 < TitanMKD> but Saleae16 have limitations too
2012-09-06T22:36:21 < TitanMKD> but 18mhz on 1 wire ?
2012-09-06T22:36:29 < Laurenceb_> 18mhz spi
2012-09-06T22:36:31 < TitanMKD> just simple SPI SD Card or you use 4bits ?
2012-09-06T22:36:35 < Laurenceb_> spi only
2012-09-06T22:36:57 < TitanMKD> ok so for 18MHz it requires at least 72MSPS
2012-09-06T22:37:09 < TitanMKD> saleae16 can do the trick
2012-09-06T22:37:14 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-06T22:37:16 < TitanMKD> but it is a bit expensive
2012-09-06T22:37:27 < Laurenceb_> ok
2012-09-06T22:37:38 < TitanMKD> Sample 2 channels at 100MHz
2012-09-06T22:38:16 < TitanMKD> 100Mhz is in fact 100MSPS ;) so it is something like 20MHz signal especially when it is not a simple square signal
2012-09-06T22:40:13 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-09-06T22:40:26 < Laurenceb_> the easiest solution would be to find out wtf is wrong
2012-09-06T22:40:58 <+izua> uh
2012-09-06T22:41:13 <+izua> you could have a timer generate data at a known speed
2012-09-06T22:41:31 <+izua> at least that would rule out dma issues. that's what i'd do anyway
2012-09-06T22:41:42 <+izua> (since i don't really fancy buying equpment :P)
2012-09-06T22:41:50 < Laurenceb_> how?
2012-09-06T22:41:54 < Laurenceb_> i dont understand
2012-09-06T22:42:01 < Laurenceb_> it would fill the buffer?
2012-09-06T22:42:01 <+izua> well
2012-09-06T22:42:06 <+izua> you said it works at slower speeds
2012-09-06T22:42:21 <+izua> have a timer fill the current spi buffer at various speeds
2012-09-06T22:42:41 < Laurenceb_> ok i see
2012-09-06T22:42:43 <+izua> dunno, pseudorandom data or heck, even just incrementing ints
2012-09-06T22:42:49 < Laurenceb_> i could do that inside the usb code
2012-09-06T22:42:52 < TitanMKD> Laurenceb to have no issue like that do some test on each function with different speed
2012-09-06T22:43:19 < Laurenceb_> i have a feeling this will work fine
2012-09-06T22:43:33 < TitanMKD> Laurenceb and add some canaries on border of each buffer because you have probably a buffer overflow ...
2012-09-06T22:43:58 < TitanMKD> also check alignment of your buffer especially for DMA ...
2012-09-06T22:44:17 < TitanMKD> some DMA requires 16bytes alignment ...
2012-09-06T22:44:19 < Laurenceb_> hmm i have a thought
2012-09-06T22:44:29 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-06T22:44:45 < Laurenceb_> at the end of each sector the while loop is going to be polling the dma register like mad
2012-09-06T22:44:57 < Laurenceb_> as the usb completes before the spi
2012-09-06T22:45:10 < Laurenceb_> and i see the corruption at the end of the sector
2012-09-06T22:45:40 < Laurenceb_> maybe polling very fast whilst dma is running fast makes the hardware unreliable
2012-09-06T22:45:47 < TitanMKD> it is on stm32f1 ?
2012-09-06T22:45:54 < Laurenceb_> yes
2012-09-06T22:45:55 < emeb> add delay to polling loop?
2012-09-06T22:46:00 < Laurenceb_> i will try tomorrow
2012-09-06T22:46:04 < Laurenceb_> hardware is at work
2012-09-06T22:46:04 < TitanMKD> yes maybe it's hw issue when you poll
2012-09-06T22:46:26 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/memory.c#L75
2012-09-06T22:46:28 < Laurenceb_> ^there
2012-09-06T22:47:16 < Laurenceb_> usually you poll for completion bit
2012-09-06T22:47:23 < Laurenceb_> im not doing that
2012-09-06T22:47:45 < Laurenceb_> perhaps a slight hardware glitch
2012-09-06T22:49:06 < TitanMKD> Laurenceb else move to chibios ;)
2012-09-06T22:49:15 < TitanMKD> driver are really cleaner than your example ;)
2012-09-06T22:49:27 < TitanMKD> without tons of while()
2012-09-06T22:49:40 < Laurenceb_> aiui its not interleaved
2012-09-06T22:49:54 < Laurenceb_> so effectively it has _loads_  of while
2012-09-06T22:50:31 < TitanMKD> to avoid that i use interrupt ;)
2012-09-06T22:50:44 < TitanMKD> i prefer to have event than to poll with while(1)
2012-09-06T22:50:55 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-09-06T22:50:58 < TitanMKD> it is also more power friendly ;)
2012-09-06T22:51:12 < TitanMKD> anyway you should really jump to chibios ;)
2012-09-06T22:51:30 -!- barthess [~barthess@164.177.211.101] has quit [Quit: Leaving.]
2012-09-06T22:51:33 < TitanMKD> even if you do not need multitask the driver with DMA are just a must
2012-09-06T22:51:46 < Laurenceb_> im using dma at the moment
2012-09-06T22:51:59 < Laurenceb_> but yeah im using chibios on my latest project
2012-09-06T22:52:02 < Laurenceb_> on F4
2012-09-06T22:52:05 < TitanMKD> but they use DMA in a clever way with events
2012-09-06T22:52:53 -!- Vutral [ss@vutral.net] has joined ##stm32
2012-09-06T22:52:53 < TitanMKD> there's also tons of errata on STM32F1/F4 maybe you are in such case
2012-09-06T22:52:57 -!- Vutral [ss@vutral.net] has quit [Changing host]
2012-09-06T22:52:57 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-06T22:54:20 < TitanMKD> Laurenceb else the best is to add in your code some on/off on GPIO and connect an oscilloscope
2012-09-06T22:54:29 < TitanMKD> i checked perf like that on my design
2012-09-06T22:54:46 < TitanMKD> it also show bugs/strange wait
2012-09-06T22:58:49 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Read error: Connection reset by peer]
2012-09-06T23:06:28 < Laurenceb_> yeah on my non interleaved version i polled for dma completion
2012-09-06T23:06:39 < Laurenceb_> which is a different register
2012-09-06T23:06:41 < emeb> +1 for toggling GPIO to scope for performance analysis
2012-09-06T23:08:06 < TitanMKD> the best is to have a debugger with trace ;)
2012-09-06T23:08:25 < TitanMKD> but too expensive for personnal project
2012-09-06T23:08:30 < emeb> word
2012-09-06T23:08:41 < emeb> dongs probably has one.
2012-09-06T23:08:58 < TitanMKD> you just trace the data access to specific registers
2012-09-06T23:09:07 < TitanMKD> and stop the trace when the bug appear
2012-09-06T23:09:49 < TitanMKD> but IIRC STM32Fx have no advanced Trace feature like that
2012-09-06T23:09:56 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Ping timeout: 248 seconds]
2012-09-06T23:10:14 < Thorn> you only need an expensive debugger for instruction trace, data trace (swv) can be done with a normal swd debugger if it supports that (st-link does)
2012-09-06T23:11:01 < TitanMKD> Thorn but i doubt a normal debugger do that in realtime with MCU running @ more than 70MHz
2012-09-06T23:11:20 < TitanMKD> and especially to be not intrusive
2012-09-06T23:11:26 < Laurenceb_> isnt it handled on the mcu?
2012-09-06T23:11:35 < TitanMKD> at work we have such tool and they cost more than 10KUSD
2012-09-06T23:11:51 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-06T23:12:11 < Thorn> read about the coresight architecture, serial wire viewer, etc.
2012-09-06T23:12:33 < Thorn> the debugger only acts as an interface converter (swo -> usb)
2012-09-06T23:13:01 < TitanMKD> yes maybe serial wire viewer but only for few data and not really in realtime like a real hardware trace
2012-09-06T23:13:29 < Thorn> the $10K thing you have is most likely an emulator
2012-09-06T23:13:30 < TitanMKD> bye
2012-09-06T23:13:34 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-06T23:14:12 < Laurenceb_> lol
2012-09-06T23:14:15 < Thorn> it *is* hardware...
2012-09-06T23:14:20 < Laurenceb_> he doenst want to admit being wrong
2012-09-06T23:14:56 -!- jaeckel [~jaeckel@jaeckel.eu] has joined ##stm32
2012-09-06T23:14:56 -!- jaeckel [~jaeckel@jaeckel.eu] has quit [Excess Flood]
2012-09-06T23:15:01 < Laurenceb_> debugging usb is painful
2012-09-06T23:15:11 < Laurenceb_> as the host will disconnect if you set a breakpoint
2012-09-06T23:15:20 <+izua> what speeds are you getting Laurenceb_ ?
2012-09-06T23:15:42 < Laurenceb_> 467KB/s with 9mhz spi and non interleaved
2012-09-06T23:15:59 < Laurenceb_> 605 with interleaved
2012-09-06T23:16:14 < Laurenceb_> 710KB/s with 18mhz non interleaved
2012-09-06T23:16:37 < Laurenceb_> about 1.1MB/s with 18Mhz and interleaved
2012-09-06T23:17:41 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 260 seconds]
2012-09-06T23:18:17 < Laurenceb_> 18MHz is a bit flakey on my 0.8mm FR4 boards for some reason
2012-09-06T23:18:23 < Laurenceb_> works 100% on the 1.6mm ones
2012-09-06T23:18:48 < Laurenceb_> trace impedance issues
2012-09-06T23:19:54 < emeb> impedance of the spi traces @ 18MHz is making a difference?
2012-09-06T23:20:15 -!- jaeckel [~jaeckel@jaeckel.eu] has joined ##stm32
2012-09-06T23:20:15 -!- jaeckel [~jaeckel@jaeckel.eu] has quit [Excess Flood]
2012-09-06T23:21:15 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-06T23:21:25 < Laurenceb_> seems to
2012-09-06T23:22:02 < emeb> interesting. 2lyr or 4lyr board?
2012-09-06T23:22:05 < Laurenceb_> 2
2012-09-06T23:22:13 < emeb> gnd plane under the spi traces?
2012-09-06T23:22:20 < Laurenceb_> with a scope connected you can see stuff bouncing around
2012-09-06T23:22:24 < Laurenceb_> yes, gnd plane
2012-09-06T23:22:49 < emeb> so you can see a difference in overshoot / ringing on the edges between the two boards?
2012-09-06T23:23:08 < Laurenceb_> yes
2012-09-06T23:23:16 < emeb> how long are the traces?
2012-09-06T23:23:36 < Laurenceb_> about 8cm
2012-09-06T23:23:53 < Laurenceb_> i think its a case of trace inductance + io capacitance forming a pi filter
2012-09-06T23:25:06 < emeb> 8cm ~= 3" ~= 0.25'. Rule of thumb is ~ 1ns/1', so your traces are about 0.25ns effective delay.
2012-09-06T23:25:25 < emeb> You must have very fast edges.
2012-09-06T23:25:33 < Laurenceb_> yeah the resonance is an order of magnitude too slow for that
2012-09-06T23:25:49 < Laurenceb_> thats why i thing its a resonance not a propogation effect
2012-09-06T23:26:01 < emeb> try source termination?
2012-09-06T23:26:14 < Laurenceb_> i have
2012-09-06T23:26:22 < Laurenceb_> it seems to help
2012-09-06T23:26:38 < Laurenceb_> but i wasnt able to get 100% reliability at 18mhz
2012-09-06T23:26:59 < Laurenceb_> tried 47,100, and 220 ohm
2012-09-06T23:27:08 < emeb> try load termination?
2012-09-06T23:27:40 < Laurenceb_> heh that would mean pcb mods
2012-09-06T23:28:19 < emeb> probably. could scrape off soldermask and put smd resistors right on the traces, wire to nearby gnd.
2012-09-06T23:28:26 < Laurenceb_> but i'd be happy with 9mhz spi if interleaving worked
2012-09-06T23:28:53 < Laurenceb_> yeah but i have 8pcbs
2012-09-06T23:29:05 < emeb> tedium.
2012-09-06T23:29:12 < Laurenceb_> exactly
2012-09-06T23:29:49 < Laurenceb_> ill try some more attempts at interleaving tomorrow
2012-09-06T23:30:17 < emeb> ever try touching the traces with that conductive black foam that DIP chips come in?
2012-09-06T23:30:41 < emeb> it's usally about 100ohm - 10kohm depending on how compressed it is.
2012-09-06T23:30:56 < Laurenceb_> oh interesting
2012-09-06T23:31:05 < Laurenceb_> http://www.google.co.uk/search?client=ubuntu&channel=fs&q=hypeberry&ie=utf-8&oe=utf-8&gl=uk&redir_esc=&ei=2wdJUIKKFqek0QWLmIHYDg#hl=en&safe=off&client=ubuntu&hs=CJP&channel=fs&gl=uk&sclient=psy-ab&q=hypeberry+pi&oq=hypeberry+pi&gs_l=serp.3...2998.3302.0.3403.3.3.0.0.0.0.144.234.1j1.2.0...0.0...1c.1.iWbWrJIbs_M&pbx=1&bav=on.2,or.r_gc.r_pw.r_qf.&fp=971d9a6b50a9aebc&biw=1366&bih=600
2012-09-06T23:31:08 < Laurenceb_> ^teehee
2012-09-06T23:31:11 < emeb> I've clipped it to a nearby ground and just "swabbed" the pins to see how loading effects things.
2012-09-06T23:31:13 < Laurenceb_> eww google url
2012-09-06T23:32:29 <+izua> still ends up here
2012-09-06T23:33:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-06T23:33:50 < Laurenceb_> eww- there was a spider in my oven
2012-09-06T23:33:57 < Laurenceb_> its now fried on the top of my pie
2012-09-06T23:38:29 < emeb> extra protein
--- Day changed Fri Sep 07 2012
2012-09-07T00:04:37 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-07T00:11:29 < Laurenceb_> oh shit
2012-09-07T00:11:39 < Laurenceb_> i know that guy who got killed in france
2012-09-07T00:11:51 < Laurenceb_> bullet proof vest time
2012-09-07T00:17:32 < Laurenceb_> oh dear http://www.davidicke.com/forum/showthread.php?s=d77c077debfdcb74800f1a2521706935&t=221659
2012-09-07T00:17:47 < Laurenceb_> <crackpottery ahead>
2012-09-07T00:22:09 -!- tech2077 [~tech2077@adsl-75-53-131-118.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-07T00:30:28 < Laurenceb_> that forum does nothing to cure paranoia
2012-09-07T00:33:21 <+Steffanx> "know" as in .. know who it is or speak with him once in a while Laurenceb_  ?
2012-09-07T00:33:41 <+Steffanx> *spoke
2012-09-07T00:38:20 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has joined ##stm32
2012-09-07T00:41:05 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-07T00:41:27 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-07T00:41:31 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-07T00:44:42 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-07T00:49:17 < emeb> Laurenceb_: Condolences - I heard about that on the news this morning. Nasty business.
2012-09-07T01:00:11 < Laurenceb_> i used to work in the same office
2012-09-07T01:00:30 < Laurenceb_> but only spoke to him once or twice
2012-09-07T01:00:40 < Laurenceb_> sstl is a big place
2012-09-07T01:09:57 < Thorn> best teardown ever? https://www.youtube.com/watch?v=mQehX0rVYuY
2012-09-07T01:10:55 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-09-07T01:14:03 < Laurenceb_> seen it already
2012-09-07T01:14:51 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-07T01:17:29 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-07T01:18:27 < Laurenceb_> "old"   :P
2012-09-07T01:22:53 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 252 seconds]
2012-09-07T01:29:16 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-07T01:35:06 -!- DaKu is now known as daku
2012-09-07T01:57:23 < emeb> Hmmm - just go email from Mouser that ST is enabling the 2nd USB peripheral on rev Y of the STM32F405.
2012-09-07T02:02:23 < zyp> huh?
2012-09-07T02:02:32 < zyp> isn't that old news?
2012-09-07T02:03:26 < emeb> The PDF is dated Aug 15
2012-09-07T02:03:32 < emeb> so probably yeah.
2012-09-07T02:04:07 < emeb> lot of latency in the distribution pipeline.
2012-09-07T02:05:08 < zyp> no, older than that
2012-09-07T02:05:28 < zyp> it's included in datasheet rev. 3, dated 31. may
2012-09-07T02:05:35 < zyp> Starting from Silicon revision Z, USB OTG full-speed interface is now
2012-09-07T02:05:36 < zyp> available for all STM32F405xx devices.
2012-09-07T02:08:58 < zyp> *yawn*
2012-09-07T02:10:29 < emeb> here's what they sent: http://www.mouser.com/PCN/STMicroelectronics_7299.pdf
2012-09-07T02:10:47 < emeb> ah - Jul 20
2012-09-07T02:10:59 < zyp> ah, right
2012-09-07T02:11:04 < zyp> Y of F2 and Z of F4
2012-09-07T02:11:13 < zyp> so that matches the datasheet
2012-09-07T02:12:51 < emeb> don't know where the rev letter is on the pkg - mine is marked "STM32F405RGT6 GH22S VG CHN GH 136"
2012-09-07T02:13:16 < zyp> that's what I also just was thinking
2012-09-07T02:13:21 < zyp> mine says exactly the same
2012-09-07T02:16:17 < zyp> it can be read from idcode register
2012-09-07T02:16:26 < zyp> hmm
2012-09-07T02:16:32 < emeb> nice
2012-09-07T02:17:39 < emeb> we used to put ID codes in the chips I worked on - remembering to update them when spinning the design was... problematic.
2012-09-07T02:18:52 < zyp> ah
2012-09-07T02:19:07 < zyp> revision id is the small letter after the ARM logo
2012-09-07T02:19:17 < zyp> mine says A there
2012-09-07T02:19:34 < zyp> which matches contents in idcode register
2012-09-07T02:20:36 < emeb> A here too. Haven't checked the reg.
2012-09-07T02:20:45 < emeb> Subtle though.
2012-09-07T02:21:11 < zyp> actually, the register claims that it's a rev B F2 :p
2012-09-07T02:21:24 < zyp> but that's a known bug for rev A F4, listed in the errata sheet
2012-09-07T02:21:37 < emeb> If a chip I bought a month ago was still rev A and they're up to Z at the factory - how deep is the pipeline. :P
2012-09-07T02:21:56 < zyp> I don't think there are anything between A and Z :p
2012-09-07T02:22:12 < emeb> That's the marketing dept at work.
2012-09-07T02:22:38 < zyp> version numbering is insane stuff
2012-09-07T02:23:11 < emeb> "no maps for these territories"
2012-09-07T02:23:32 < zyp> I've still not figured out exactly how the version numbering schemes we use at work works
2012-09-07T02:26:27 < zyp> they usually vary from three to five different components, and the last is usually a number that increments for each sequential release
2012-09-07T02:26:54 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-07T02:27:20 < zyp> but I have no idea how the other parts work, they seem to change at random times in random ways, and then the last number gets reset back to 1
2012-09-07T02:31:38 -!- peabody128 [~peabody12@128.249.96.127] has joined ##stm32
2012-09-07T02:35:07 < dongs>  Thorn> you only need an expensive debugger for instruction trace, data trace (swv) can be done with a normal swd debugger if it supports that (st-link does)
2012-09-07T02:35:28 < dongs> yeah, svw/swo works regardless of mcu speed
2012-09-07T02:35:58 < dongs> < Laurenceb_> he doenst want to admit being wrong
2012-09-07T02:36:00 < dongs> agreed
2012-09-07T02:36:52 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-07T02:37:33 -!- peabody128 [~peabody12@128.249.96.127] has quit [Ping timeout: 246 seconds]
2012-09-07T03:11:54 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 276 seconds]
2012-09-07T03:31:57 -!- tech2077 [~tech2077@adsl-75-53-141-174.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-07T04:05:29 < dongs> http://i.imgur.com/3orib.png kickstarter time. like teensy3 but without the aids
2012-09-07T04:06:05 < emeb> cute
2012-09-07T04:06:08 < emeb> what's the csp?
2012-09-07T04:06:20 < dongs> usb of course. bootloaders are for faggots.
2012-09-07T04:06:47 < emeb> usb socket on the opposite side?
2012-09-07T04:06:50 < dongs> yes
2012-09-07T04:06:53 < dongs> and its DIP
2012-09-07T04:07:00 < dongs> find any current proj with DIP usb!
2012-09-07T04:07:18 < dongs> smd USB sockets are utterly worthless
2012-09-07T04:07:18 < emeb> never seen that. Why did you pick that form?
2012-09-07T04:07:29 < emeb> ah - stronger mechanical connection.
2012-09-07T04:07:33 < dongs> its same size as tarduino pro mini + 5mm
2012-09-07T04:07:34 < dongs> yes
2012-09-07T04:07:43 < emeb> I've always wondered about that.
2012-09-07T04:07:46 < dongs> the usb is standard for dip
2012-09-07T04:07:48 < Erlkoenig> (03:07:18) dongs: smd USB sockets are utterly worthless <- what? why?
2012-09-07T04:08:04 < dongs> Erlkoenig: lol?? buy any modern shit and yank usb cable a few times
2012-09-07T04:08:15 < dongs> on second try the socket will come out right wiht the cable
2012-09-07T04:08:16 < emeb> An old mentor of mine lived by this maxim: "Solder has all the mechanical qualities of peanut butter."
2012-09-07T04:08:18 < dongs> tearing half hte pcb off
2012-09-07T04:08:50 < emeb> Never rely on solder alone for something that will need to absorb force.
2012-09-07T04:08:51 < Erlkoenig> ah, mechanical stability... indeed, my mobile phone's smd-usb got loose...
2012-09-07T04:09:14 < R2COM> i dont think so, i used many smt usb jacks, no problems
2012-09-07T04:09:24 < R2COM> some of them actually had some small part going into pcb
2012-09-07T04:09:28 < dongs> http://www.digikey.com/product-search/en?x=0&y=0&KeyWords=WM17115-ND this is a very standard through hole USB pinout, many manufactures clone/reuse same so
2012-09-07T04:09:40 < emeb> R2COM: yep. Little plastic legs on the ones I use.
2012-09-07T04:09:46 < R2COM> a hole in the pcb, so if someone tries to pull horizontally it will help the mechanical integrity
2012-09-07T04:09:50 < dongs> R2COM: right, but its still annoying and not as good as every part through hole
2012-09-07T04:10:16 < emeb> good for horizontal force, but not vertical.
2012-09-07T04:10:38 < R2COM> vertical is usually done by making nice sealed finished box which holds it all together
2012-09-07T04:11:05 < emeb> dongs: what's the main chip on that?
2012-09-07T04:11:14 < dongs> emeb: f103/f30x
2012-09-07T04:11:25 < R2COM> plus, usb if im correct requires some license if one plans to sell devices
2012-09-07T04:11:31 < emeb> what about the on-chip USB?
2012-09-07T04:11:49 < R2COM> what about them?
2012-09-07T04:11:50 < dongs> emeb: not used, thats not the target audience
2012-09-07T04:12:01 < dongs> R2COM: you need a vid/pid
2012-09-07T04:12:08 < dongs> no other license afaik.
2012-09-07T04:12:19 < emeb> agree
2012-09-07T04:12:20 < dongs> and if youre using someone elses usb>uart bridge or similar, then nothing at all.
2012-09-07T04:12:22 < R2COM> well yes, and it costs few $k if im correct
2012-09-07T04:12:26 < dongs> right
2012-09-07T04:12:46 < Erlkoenig> (03:09:50) dongs: R2COM: right, but its still annoying and not as good as every part through hole <-- and other people say "THT is dead, all-hail to SMD" ...
2012-09-07T04:12:55 < R2COM> what about the USB jack? the fact that im using the usb functionality?
2012-09-07T04:13:00 < emeb> those people are idiots.
2012-09-07T04:13:04 < dongs> R2COM: nope
2012-09-07T04:13:24 < emeb> that said, TH is an extra step in assembly = more $$
2012-09-07T04:13:33 < dongs> Erlkoenig: sorry, I was talking strictly USB sockets and other user-destroyable external interfaces
2012-09-07T04:13:42 < Erlkoenig> ah okay
2012-09-07T04:13:42 < R2COM> ok, so lets say i buy FTDI chip and put on my board, and connect to usb jack, i dont have to pay anything at all?
2012-09-07T04:13:43 < dongs> dip components are dead, indeed
2012-09-07T04:13:53 < dongs> R2COM: certainly not
2012-09-07T04:13:55 < R2COM> TH is really some extra step in assembly
2012-09-07T04:13:58 < dongs> R2COM: ftdi pays their vid/pid
2012-09-07T04:14:02 < Erlkoenig> dip is somewhat useful for beginners and hobbyists :D
2012-09-07T04:14:26 < R2COM> hmm
2012-09-07T04:14:45 < Erlkoenig> R2COM: you can "borrow" other vendor's VID/PID... as long as you don't distribute your device or install the driver for the real device, it will work...
2012-09-07T04:15:11 < dongs> I believe there's some private use ranges as well
2012-09-07T04:15:17 < dongs> been a while since i looked at usb-if shit
2012-09-07T04:15:18 < R2COM> well if someone selling his board, then he does distribute his device...
2012-09-07T04:15:35 < dongs> R2COM: there's no question with usb>serial stuff from a manufacturer.
2012-09-07T04:15:45 < R2COM> ok
2012-09-07T04:15:45 < Erlkoenig> R2COM: then he needs to buy a VID, and thats expensive
2012-09-07T04:16:10 < R2COM> ok, or he should use usb>uart from FTDI
2012-09-07T04:16:11 < Erlkoenig> and yes FTDI bought the VID for you, just like you don't need your own VID for your USB mouse
2012-09-07T04:16:39 < dongs> when I did usb>serial the other day I just reused tarduino 2560 vid+pid.
2012-09-07T04:16:44 < dongs> why?
2012-09-07T04:16:51 < dongs> 1) those fucks are too cheap to sign their windows drivers
2012-09-07T04:16:53 < R2COM> so it would be a little wasteful, having USB functionality in stm32, but because of licensing issues bringing FTDI chip to transfer data from USB to the chip
2012-09-07T04:17:00 < dongs> 2) i can just tell people "go donwload tarduino to instal ldrivers"
2012-09-07T04:17:19 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-07T04:17:24 < dongs> R2COM: tons of projects use on chip usb.
2012-09-07T04:17:33 < Erlkoenig> mine does <3
2012-09-07T04:17:38 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-07T04:17:48 < R2COM> i know, i am talking about the ones which you plan to sell, and be clear
2012-09-07T04:18:17 < Erlkoenig> for selling your stuff, you also need to pay to be allowed to stick the USB logo onto your device
2012-09-07T04:18:24 < R2COM> i used usb a lot before with microchips, but all that stuff were things designed for someone to operate within the lab, not for retail
2012-09-07T04:18:33 < emeb> Erlkoenig: but the logo is not required.
2012-09-07T04:18:38 < Erlkoenig> thats correct ;)
2012-09-07T04:18:44 < Erlkoenig> but increases trust in the device ;)
2012-09-07T04:18:45 < R2COM> Erlkoenig: so thats what i was wondering... turns out that usb logo because i use their jack is required
2012-09-07T04:19:05 < emeb> Erlkoenig: depends on your audience. If your making development tools, likely they don't care.
2012-09-07T04:19:49 < dongs> pretty sure olimex doesnt put USB logo on their shit
2012-09-07T04:19:54 < emeb> IOW - do you think Adafruit, Sparkfun, Olimex, etc bother to get logo qualified?
2012-09-07T04:19:56 < dongs> and all their stuff has usb sockets :)
2012-09-07T04:19:58 < dongs> right
2012-09-07T04:20:09 < Erlkoenig> there once was somebody who bought one VID and then sold the PID's belonging to that for a low price to hobbyists... unfortunately they sued him
2012-09-07T04:20:20 < R2COM> so does it mean they can be sued if someone wants?
2012-09-07T04:21:04 < Erlkoenig> darn moneysuckers
2012-09-07T04:21:04 < dongs> no
2012-09-07T04:22:10 < R2COM> stm32 has one uart which works on 10.5Mbps by the way
2012-09-07T04:22:54 < R2COM> but one would need a better cable if wants using it on longer ranges
2012-09-07T04:23:30 < Erlkoenig> it's a very enlightning moment when you plug in your selfmade usb device and the OS correctly enumerizes it and displays everything like a regular usb device ;-) http://2g2s.de/sub/usb-enum.png
2012-09-07T04:23:36 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-07T04:23:36 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-07T04:23:36 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-07T04:23:49 < Erlkoenig> R2COM: USB is faster!11
2012-09-07T04:24:05 < R2COM> i know its faster
2012-09-07T04:26:37 < Erlkoenig> but what's weird... USB devices can supply their own strings to describe the vendor and device name, but both windows and linux completely ignore them, and say "Unknown device" or something unless you supply the correct driver, and then the strings from the driver are displayed... the original strings from the device can be displayed, but are normally hidden...
2012-09-07T04:27:18 < Erlkoenig> it would probably help identifying unknown devices if the OS would just display the device name string the device sends
2012-09-07T04:28:35 < R2COM> ...
2012-09-07T04:31:08 < emeb> the word of the day is HABOOB!
2012-09-07T04:31:40 < Erlkoenig> i wonder why noone thought of some plug-and-play standard for RS232 devices...
2012-09-07T04:32:32 < emeb> Err - RS-232 is like >50 years old. No such thing back then.
2012-09-07T04:34:20 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-07T04:35:15 < Erlkoenig> so in these 50 years some company could have made something like that...
2012-09-07T04:39:47 < dongs> flyback: did you hear, that led driver is like 16c. i bought 100.
2012-09-07T04:40:09 < dongs> china cloning china? ya right
2012-09-07T04:41:41 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-07T04:41:42 < upgrdman> what led driver?
2012-09-07T04:41:44 -!- R2COM1 [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-07T04:42:09 < dongs> WS2811
2012-09-07T04:43:12 -!- R2COM1 is now known as R2COM
2012-09-07T04:47:27 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 255 seconds]
2012-09-07T04:53:53 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-07T04:55:18 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 264 seconds]
2012-09-07T05:02:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-07T05:06:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-07T05:47:06 < dongs> loldongs
2012-09-07T05:47:11 < dongs> where are the chats
2012-09-07T05:55:32 < R2COM> here
2012-09-07T05:55:54 < R2COM> im looking now if theres a way to make interrupt on TIM occur only when CCRx has specific value
2012-09-07T05:56:06 < R2COM> for input capture mode lets say
2012-09-07T05:57:10 < R2COM> not like, react on the pulse depending on its edges, but something like: monitor the input capture, but dont bother me until you detected that pulse width is longer than N
2012-09-07T05:58:02 < dongs> yes
2012-09-07T05:58:31 < dongs> masterslave mode, trigger but tifp1 or someshit i believe
2012-09-07T05:58:35 < dongs> i have that in one of my projects
2012-09-07T06:11:59 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-221.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-07T06:25:19 < dongs> http://captchaaaaaaaaaaaaaaaaaaaaaaaaa.ytmnd.com/
2012-09-07T06:26:54 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-07T06:27:01 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-07T06:53:59 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-07T06:54:19 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-07T06:58:28 < zippe> R2COM: I think you can have one of the timer inputs be an enable
2012-09-07T07:00:41 < R2COM> well
2012-09-07T07:01:15 < R2COM> im reading datasheet now, will see
2012-09-07T07:51:58 -!- claude is now known as Claude
2012-09-07T08:54:50 -!- Claude is now known as claude
2012-09-07T08:55:10 -!- claude is now known as Claude
2012-09-07T09:29:28 -!- dekar [~dekar@212.255.27.206] has quit [Quit: This computer has gone to sleep]
2012-09-07T09:40:41 -!- Piele [~peter@5ED0EA16.cm-7-1d.dynamic.ziggo.nl] has joined ##stm32
2012-09-07T09:43:48 -!- NonaSuomy [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has joined ##stm32
2012-09-07T09:43:48 -!- NonaSuomy [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has quit [Changing host]
2012-09-07T09:43:48 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-07T09:45:42 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 252 seconds]
2012-09-07T09:46:01 -!- Claude is now known as claude
2012-09-07T09:51:22 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 272 seconds]
2012-09-07T09:56:35 -!- High-Fidelity [~High-Fide@123-243-232-159.static.tpgi.com.au] has joined ##stm32
2012-09-07T10:20:07 -!- dekar [~dekar@dslb-188-107-170-026.pools.arcor-ip.net] has joined ##stm32
2012-09-07T10:20:09 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-07T10:26:48 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-07T10:30:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-07T10:36:21 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-07T10:37:42 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-07T10:43:03 < tavish> i am trying to do bulk transfers on stm32f4 OTG_HS in full speed, i am able to reach about speed that is possible using bulk tranfers, and i started by modifying the CDC class example. Any pointers to what could be wrong?
2012-09-07T10:46:48 < dongs> your question does not show the problem.
2012-09-07T10:46:56 < dongs> if you read it , it seems everything is working as expected.
2012-09-07T10:49:17 < tavish> dongs: I can go upto 450-500KB/s in OUT and 600 in IN bulk transfers, It be able to do 1.1MB/s or so
2012-09-07T10:50:49 < tavish> dongs: oh, wait. loopback
2012-09-07T10:50:55 < tavish> i think that's it
2012-09-07T10:53:11 < zyp> well, duh, USB is not duplex
2012-09-07T10:58:25 < tavish> :( that was it
2012-09-07T10:59:25 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-07T11:01:09 < zyp> (and before anybody mentions it, yes SS USB is duplex)
2012-09-07T11:01:40 -!- claude is now known as claude_work
2012-09-07T11:02:37 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-07T11:06:05 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-07T11:06:05 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-07T11:06:05 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-07T11:19:17 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 268 seconds]
2012-09-07T11:19:39 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-07T11:19:40 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-07T11:19:40 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-07T11:31:31 -!- zippe [~Adium@173.11.99.161] has quit [Remote host closed the connection]
2012-09-07T11:32:19 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 246 seconds]
2012-09-07T11:35:36 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-09-07T12:08:16 -!- Piele [~peter@5ED0EA16.cm-7-1d.dynamic.ziggo.nl] has quit [Quit: Leaving]
2012-09-07T12:17:31 < dongs> Localization... All my software is localized. Localized for the one place that counts, the United States of America (fuck yeah). Who cares if Alvarez or whoever still doesn't know what the words File, Edit and Save mean, or that pressing a printer icon will print the shit on the screen. They need to learn some fucking Enlglish instead of expecting pop up dialog boxes to start with upside down question marks.
2012-09-07T12:17:38 < dongs> lol
2012-09-07T12:25:51 < zyp> ha
2012-09-07T12:39:45 < karlp> emeb/dongs: what's "csp" mean?
2012-09-07T12:40:01 < dongs> chip scale package? i guess
2012-09-07T12:40:20 < zyp> yeah
2012-09-07T12:40:48 < zyp> also known as flip-chip
2012-09-07T12:41:28 < zyp> which is basicly a die with solder points
2012-09-07T12:42:55 < karlp> what part on that pcb rendering that dongs pasted was csp?
2012-09-07T12:43:04 < karlp> there's a qfn looking part?
2012-09-07T12:43:10 < dongs> yeah, he thought that was
2012-09-07T12:43:12 < dongs> which it wasnt
2012-09-07T12:43:26 < karlp> so, what really makes this one better than teensy3?
2012-09-07T12:43:38 < karlp> we've been over the "it's not the hardware, it's the software" before right?
2012-09-07T12:44:08 < karlp> I'mn not a kickstarter fan, but paul stoffgren(sp?) has made the teensy pretty damn easy to use.
2012-09-07T12:45:27 < dongs> haha sorry. its just STM32 with uart bootloader and all the pins I want broken out.
2012-09-07T12:45:30 < dongs> no tarduino hand holding here
2012-09-07T12:45:49 < dongs> notice the "i want" part, i was joking about doing anything with kickstarter wiht it.
2012-09-07T12:50:02 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-07T13:04:03 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-07T13:07:35 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-07T13:07:36 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-07T13:07:36 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-07T13:07:39 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-07T13:11:31 < dongs> status
2012-09-07T13:13:25 < karlp> fixing svn checkouts on jenkins
2012-09-07T13:13:43 < zyp> sounds shitty
2012-09-07T13:13:45 < karlp> and verifying workarounds for gdbtui
2012-09-07T13:13:47 < karlp> it is shitty
2012-09-07T13:13:53 < dongs> more like jenkems
2012-09-07T13:13:55 < karlp> both are things I'd rather not be doing :)
2012-09-07T13:14:08 < karlp> I'm tempted to say, fuckit, today's the day we use git
2012-09-07T13:14:30 < karlp> we're already using git svn to work with git locally anyway.
2012-09-07T13:31:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-07T13:45:51 < dongs> https://upload.wikimedia.org/wikipedia/commons/8/8c/Gldt.svg
2012-09-07T13:45:53 < dongs> hly shit
2012-09-07T13:46:10 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has joined ##stm32
2012-09-07T13:49:26 < dongs> and this is, kids, what is wrong with lunix.
2012-09-07T13:53:45 < zyp> what exactly?
2012-09-07T13:57:45 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-07T14:02:36 -!- daku is now known as DaKu
2012-09-07T14:16:27 <+izua> oh look, there's gentoo and openwrt too
2012-09-07T14:16:28 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds]
2012-09-07T14:23:02 < dongs> Laurenceb: http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATA_BRIEF/DM00052740.pdf
2012-09-07T14:27:56 < karlp> so, that's only $30,
2012-09-07T14:28:11 < karlp> how's that compare to the boards like this: https://www.sparkfun.com/products/10724 that are $100 ?
2012-09-07T14:29:51 < zyp> [insert analogy between discovery board and other stm32 devboard]
2012-09-07T14:30:16 < Laurenceb> omfg
2012-09-07T14:30:20 < Laurenceb> fapfapfap
2012-09-07T14:30:33 < zyp> «but it's shit»
2012-09-07T14:31:19 < Laurenceb> nope
2012-09-07T14:31:28 < Laurenceb> the lsm303 is pretty good
2012-09-07T14:31:39 < Laurenceb> and that st baro is as good as measurement specialities
2012-09-07T14:31:54 < Laurenceb> lol @ linux timeline
2012-09-07T14:32:01 < Laurenceb> i used to run morphix
2012-09-07T14:33:24 < Laurenceb> http://www.morphix.org/
2012-09-07T14:33:26 < Laurenceb> lol
2012-09-07T14:33:53 < dongs> karlp: lolll.
2012-09-07T14:34:07 < dongs> karlp: itg3200 = $3, adxl345 = $1.5, HMC5883L = $2
2012-09-07T14:34:11 < dongs> sparkfun = fucking jews.
2012-09-07T14:34:23 < dongs> thats how it compares.
2012-09-07T14:34:28 < dongs> i think theire itg3200 breakout is liek $49
2012-09-07T14:34:33 < dongs> which is insane
2012-09-07T14:34:40 < Laurenceb> that ad accel is piss poor
2012-09-07T14:34:56 < karlp> I was always curious why those sparkfun imus were so pricey
2012-09-07T14:35:02 < Laurenceb> but i bet the itg3200 beats the new st gyro still
2012-09-07T14:35:12 < Laurenceb> all my itg3200 gyros have been insanely good
2012-09-07T14:35:14 < qyx_> l3g4200 you mena?
2012-09-07T14:35:16 < qyx_> mean
2012-09-07T14:35:22 < Laurenceb> way lower noise than datasheet
2012-09-07T14:35:23 < Laurenceb> no
2012-09-07T14:35:35 < Laurenceb> L3GD20
2012-09-07T14:38:09 < qyx_> hm
2012-09-07T14:38:27 < Laurenceb> someone needs to do allen variance on these things
2012-09-07T14:38:38  * Laurenceb looks at peabody124
2012-09-07T14:44:36 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-61.pptp.stw-bonn.de] has joined ##stm32
2012-09-07T14:49:42 < dongs> fuck.
2012-09-07T14:49:48 < dongs> i cant get that fucking evalboard from jp digikey
2012-09-07T14:54:19 < tavish> otg_hs in device mode stopped working as soon as i added code for otg_fs in host mode
2012-09-07T14:54:30 < tavish> these problems shouls sort themselves out
2012-09-07T14:55:06 < dongs> 497-13014-ND = not found on digikey.jp :(
2012-09-07T14:55:14 < dongs> fucking export restriction scam
2012-09-07T14:55:23 < dongs> aybe its time i place a mouser order
2012-09-07T14:57:22 -!- tech2077 [~tech2077@adsl-75-53-141-174.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-09-07T15:02:48 < Laurenceb> arrgg rageee
2012-09-07T15:03:00 < zyp> like usual
2012-09-07T15:03:12 < Laurenceb> just spent 25 minutes walking around trying to find someone to sign uni order forms
2012-09-07T15:03:22 < Laurenceb> 15 possible people, every one on vacation
2012-09-07T15:03:45 < Laurenceb> maybe i should give up
2012-09-07T15:04:07 < Laurenceb> what a load of slackers
2012-09-07T15:04:45 < Erlkoenig> hm, wouldn't it, theoretically, be better if linker scripts where shipped with the toolchains, and the linker automatically chose the correct one (based on some command line switch)? To avoid every project having a copy of the same linkerscript...
2012-09-07T15:05:24 < BrainDamage> you could simply symlink
2012-09-07T15:05:38 < BrainDamage> or you can write a master linker script and get it included in the others
2012-09-07T15:06:24 < zyp> Erlkoenig, linker script and startup code depends on each other
2012-09-07T15:06:36 < Erlkoenig> oh hm one would have to ship the startup code as well
2012-09-07T15:06:59 < Erlkoenig> like "the toolchain provides an environment for C programs" and you don't have to mess with "providing a C environment"
2012-09-07T15:07:22 < zyp> that's pretty much how avr-gcc works
2012-09-07T15:07:38 < Erlkoenig> yeah... the arm-gcc might do that as well...
2012-09-07T15:08:01 < zyp> though, then you have consider that avr-gcc mainly have to handle atmel's devices
2012-09-07T15:08:14 < zyp> there is a lot more differences in arm devices
2012-09-07T15:08:33 < BrainDamage> mainly? there's avr-compatible devices that aren't manifactured by atmel?
2012-09-07T15:08:36 < Erlkoenig> then one could easily ship C-only code for some open source project, and everyone could compile it for any controller, just by specifying the desired controller on the command line
2012-09-07T15:08:48 < Laurenceb> theres avr clones
2012-09-07T15:08:51 < Laurenceb> soft cores
2012-09-07T15:09:23 < Erlkoenig> zyp: so what? one could ship a linker script and a startup code file for every different controller. maybe not in the "core toolchain package" but by "plugins" or something
2012-09-07T15:09:24 < zyp> BrainDamage, no, but you might still have relevant stuff like external memories and so on
2012-09-07T15:10:09 < Erlkoenig> that would all be only for the "default case"... if you intend to do "special stuff" you have to write your own script, but that would simplify the standard cases
2012-09-07T15:10:20 < BrainDamage> Erlkoenig: unfortunately, all the external periferials are not cross compatible between devices
2012-09-07T15:10:23 < zyp> true
2012-09-07T15:10:43 < BrainDamage> hence why projects like libopencm3 exist
2012-09-07T15:10:59 < BrainDamage> or OSs hardware abstraction layer in general
2012-09-07T15:11:28 < Laurenceb> on a similar note - does it seem feasible that a polling loop to check dma transfer counter could screw up the dma transfer?
2012-09-07T15:11:38 < Laurenceb> on the stm32f1
2012-09-07T15:11:44 < zyp> no
2012-09-07T15:12:19 < Laurenceb> heh
2012-09-07T15:12:40 < zyp> BrainDamage, and then the linker script issue is also solved
2012-09-07T15:12:46 < Laurenceb> well my spi2 sd card corruption seems to occur when there is polling going on
2012-09-07T15:13:19 < Laurenceb> looks like it reads occasional 0x02 and spits out junk
2012-09-07T15:13:32 < zyp> Laurenceb, what are you using the polled values for?
2012-09-07T15:13:51 < Laurenceb> to determine where i can transfer up to using usb
2012-09-07T15:14:04 < Laurenceb> so i can interleave sd read and usb write
2012-09-07T15:14:39 < zyp> oh
2012-09-07T15:14:45 < zyp> so you might have a race there?
2012-09-07T15:15:03 < Laurenceb> i cant see one
2012-09-07T15:15:30 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/memory.c#L75
2012-09-07T15:15:31 < zyp> is it guaranteed that the transfer counter is incremented only after the data has been committed to ram?
2012-09-07T15:15:33 < Laurenceb> modified st code
2012-09-07T15:15:42 < Laurenceb> yes i think so
2012-09-07T15:16:42 < Laurenceb> what does the spi transmit during an rx only dma transfer?
2012-09-07T15:17:19 < Laurenceb> like here https://github.com/Laurenceb/STM32-Logger/blob/master/Util/fat_fs/src/sd_spi_stm32.c#L439
2012-09-07T15:17:42 < zyp> no idea
2012-09-07T15:18:20 < Laurenceb> oh
2012-09-07T15:18:22 < Laurenceb> rw_workbyte[] = { 0xffff };
2012-09-07T15:18:26 < Laurenceb> its obvious
2012-09-07T15:19:07 < Laurenceb> if i swap the code to poll for dma completion there is no issue
2012-09-07T15:20:09 < zyp> try subtracting a word when polling
2012-09-07T15:20:38 < zyp> i.e. assume that you have one word less than the value you are reading
2012-09-07T15:22:26 < Laurenceb> theres actually already an offset of 2
2012-09-07T15:22:35 < Laurenceb> as its a 514byte transfer
2012-09-07T15:22:40 < Laurenceb> to fit the crc on the end
2012-09-07T15:23:26 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-07T15:31:26 < Tzacks> hey ppl, busy here?
2012-09-07T15:31:39 < zyp> always
2012-09-07T15:31:45 < Tzacks> zyp
2012-09-07T15:31:53 < Tzacks> help me, help me now :))))
2012-09-07T15:32:12 < Tzacks> im just kidding :)
2012-09-07T15:32:12 < zyp> what are you having problems with?
2012-09-07T15:32:15 < zyp> :p
2012-09-07T15:32:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-07T15:32:17 < Laurenceb> life
2012-09-07T15:32:22 < Tzacks> i am looking for a simple circuit
2012-09-07T15:32:42 < Tzacks> i got 5V solar panel, 4.2V lipo 5000mAh
2012-09-07T15:33:35 < Tzacks> and i would like to make some lamp that can charge battery durong the day and then use led at night
2012-09-07T15:33:48 < Tzacks> i need something simple and not too dangerous with that lipo :)
2012-09-07T15:34:18 < Tzacks> i mean something very simple :)))
2012-09-07T15:34:34 < Tzacks> i could find "garden light" but it was all for nimh :/
2012-09-07T15:35:55 < BrainDamage> http://dx.com/solar-powered-4-led-table-lamp-25504?item=1
2012-09-07T15:35:58 < Laurenceb> i use a bq24090 charger controller ic on my projects
2012-09-07T15:36:21 < BrainDamage> even less http://dx.com/solar-powered-self-recharge-4-led-table-lamp-with-flexible-neck-white-44005?item=22
2012-09-07T15:36:24 < Tzacks> i have student budget :/ cant afford to byu anything :/
2012-09-07T15:37:18 < BrainDamage> chances are it'll cost you more building it than buying it
2012-09-07T15:38:03 < Tzacks> i have some parts home etc, i wanted to build something
2012-09-07T15:39:37 <+izua> Tzacks: you can build an home made pyrotechnic/special fx device
2012-09-07T15:39:46 <+izua> all you need is your battery and some wire
2012-09-07T15:39:57 < Tzacks> i know :)
2012-09-07T15:40:47 < BrainDamage> you'll need a charge controller and a low voltage cutoff
2012-09-07T15:40:52 < Tzacks> but i'd prefer to have led light at night :)
2012-09-07T15:41:14 < BrainDamage> if you discharge the batt too much, it'll ruin it, and you'll risk fires
2012-09-07T15:41:18 < Tzacks> BrainDamage: i know :)
2012-09-07T15:41:50 < karlp> Erlkoenig: libopencm3 now includes and installs linker scripts that go with the startupcode for all the smt32f100 parts at least.  You need to specify it by name, but you don't have to include it every time.  not as magically easy as avr-gcc, but better than before
2012-09-07T15:42:15 < BrainDamage> try getting some samples from maxim
2012-09-07T15:42:37 < BrainDamage> they even ship by courier, but you'll have to use a professional sounding email
2012-09-07T15:42:43 < BrainDamage> your univ email will probably do
2012-09-07T15:43:03 < Erlkoenig> karlp: oh nice
2012-09-07T15:43:17 < karlp> (I got sick of having duplicate files all over the place)
2012-09-07T15:43:52 < Erlkoenig> a little bit of documentation on the libopencm3 web page would not hurt...
2012-09-07T15:44:52 < BrainDamage> you can generate it using doxygen
2012-09-07T15:45:18 < Erlkoenig> LGPL on a mikrocontroller library? Oo
2012-09-07T15:46:09 < BrainDamage> the more annoying thing is the _print overload function between libopencm3 and newlib
2012-09-07T15:46:25 < karlp> what's that?
2012-09-07T15:46:27 < BrainDamage> err, I mean _write
2012-09-07T15:46:38 < karlp> I had to use my own _write in my libopencm3 app?
2012-09-07T15:46:43 < BrainDamage> /usr/lib/gcc/arm-none-eabi/4.7.0/../../../../arm-none-eabi/lib/thumb/libc.a(lib_a-syscalls.o): In function `_write':
2012-09-07T15:46:44 < BrainDamage> /tmp/yaourt-tmp-cloud/aur-arm-eabi-newlib/src/build/arm-none-eabi/thumb/newlib/libc/sys/arm/../../../../../../../newlib-1.20.0/newlib/libc/sys/arm/syscalls.c:335: multiple definition of `_write'
2012-09-07T15:46:44 < BrainDamage> usart_irq_printf.o:/home/cloud/repositories/stm32/libopencm3/examples/stm32/f1/lisa-m-2/usart_irq_printf/usart_irq_printf.c:205: first defined here
2012-09-07T15:46:47 < Erlkoenig> GPL is evil ...
2012-09-07T15:46:53 < karlp> lgpl != gpl
2012-09-07T15:47:10 < Erlkoenig> yeah... but LGPL on a µC is still... weird?
2012-09-07T15:47:17 < Erlkoenig> because there is (often) no dynamic linking?
2012-09-07T15:47:18  * karlp shrugs.
2012-09-07T15:47:43 <+izua> cm3 went lgpl
2012-09-07T15:47:52 < karlp> the libopencm3 people imply that they mean you have to provide your .o, and the makefile,
2012-09-07T15:48:03 < karlp> even if you statically link.
2012-09-07T15:48:34 < Erlkoenig> that means you have to reveal all symbols (function names etc.)?
2012-09-07T15:48:51 < karlp> oh noes
2012-09-07T15:50:19 < karlp> see 4.d.0 vs 4.d.1
2012-09-07T15:50:27 < karlp> for shared linking vs "minimal corresponding source"
2012-09-07T15:51:20 < karlp> would a more open license have been better? probably.  is lgpl better than gpl (which libopencm3 was before)? definitely
2012-09-07T15:51:49 < Erlkoenig> yes, i agree
2012-09-07T15:51:56 <+izua> so what exactly can someone legally request from you if you link against cm3?
2012-09-07T15:52:29 <+izua> just object files and makefile?
2012-09-07T15:53:24 < Erlkoenig> don't object files always contain symbols? they don't request you to reveal your symbols, but how would you practically release your .o files without providing symbols?
2012-09-07T16:01:42 < karlp> .o and makefiles is my understanding, and only to people you've actually distributed to, not to the world.
2012-09-07T16:01:50 < karlp> and only on demand.
2012-09-07T16:03:20 < Erlkoenig> hm. this *might* scare potential "commercial" users of the lib, but i don't know...
2012-09-07T16:03:32 <+izua> like dekar?
2012-09-07T16:03:47 <+izua> but what's the point though?
2012-09-07T16:04:01 <+izua> i mean, with .o and makefiles, you can't make any changes, or see what the code does
2012-09-07T16:04:15 <+izua> you can barely compile and flash it (or make copies of it, which is pretty bad)
2012-09-07T16:07:15 < Erlkoenig> yes, but .o files neccessarily reveal all the symbols of your code, don't they?
2012-09-07T16:07:33 < zyp> no, not neccessarily
2012-09-07T16:07:39 < Laurenceb> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex_mx_stm32%2FReceiving%20SPI%20data%20through%20DMA%20-%20STM32%20internal%20hardware%20error&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=2678
2012-09-07T16:07:45 < Laurenceb> hmm could be my issue?
2012-09-07T16:07:53 < Erlkoenig> zyp: oh, how does linking work then without symbols?
2012-09-07T16:08:55 <+izua> Laurenceb: that gives me an idea
2012-09-07T16:09:01 <+izua> try using the sd card with the sdio peripheral
2012-09-07T16:09:11 < Laurenceb> its not on f103cbt6
2012-09-07T16:10:16 < zyp> Erlkoenig, consider static functions for instance
2012-09-07T16:10:52 < zyp> static functions are functions without visibility outside the translation unit
2012-09-07T16:11:59 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-07T16:12:41 < Erlkoenig> zyp: but sometimes programs contain more than one .o file, and with symbol interdependencies, so the symbols have to be visible?
2012-09-07T16:12:48 < zyp> in theory there should be possible linking several object files, outputting a new object file
2012-09-07T16:13:04 < zyp> symbols only used internally could then be resolved and hidden
2012-09-07T16:13:17 < Erlkoenig> yeah in theory... the problem was how to do it practically :D
2012-09-07T16:14:13 <+izua> there is -combine, but only for source files
2012-09-07T16:14:28 < zyp> but that still results in the same, no?
2012-09-07T16:15:07 <+izua> yeah
2012-09-07T16:15:12 <+izua> i think ar is the needed tool
2012-09-07T16:15:15 < karlp> Erlkoenig: we're using libopencm3 in a commercial device...
2012-09-07T16:15:27 <+izua> karlp: anyone asked for the code yet?
2012-09-07T16:15:32 < karlp> nope.
2012-09-07T16:15:36 <+izua> or you simply didn't advertise it as cm3 based?
2012-09-07T16:15:41 < karlp> and I don't expect them to either.
2012-09-07T16:15:47 < karlp> it's just a small part of a system.
2012-09-07T16:18:05 < karlp> I haven't looked at the literature recently, we did have a list of the opensoruce software in use, before the next batch gets made I'll review it again
2012-09-07T16:19:50 < dongs> um
2012-09-07T16:19:53 < dongs> isnt libopencm3 LGPL
2012-09-07T16:20:03 < dongs> so you can do wahtever the fuck you want with it.
2012-09-07T16:20:08 < Erlkoenig> no, not really
2012-09-07T16:20:24 < Erlkoenig> you have to make sure users of the software can replace the lib in their product
2012-09-07T16:20:32 < dongs> wut
2012-09-07T16:21:14 < dongs> replace with what
2012-09-07T16:21:28 < Erlkoenig> so users can supply their own libopencm3
2012-09-07T16:21:29 < karlp> with a bugfixed version of the library,
2012-09-07T16:21:37 < dongs> haha
2012-09-07T16:21:54 < Erlkoenig> so you either have to link your program to the lib dynamically (on a µC ... ) or provide your (potentially proprietary) .o files so users can link your program with libopencm3
2012-09-07T16:21:56 < BrainDamage> it's bsd that lets you do what you want with the code
2012-09-07T16:22:02 < Erlkoenig> and the latter might reveal your program's symbols
2012-09-07T16:22:34 < karlp> Erlkoenig: note however, that there's no real requirement that it's actually possible to _use_ the resulting relinked image.
2012-09-07T16:22:37 < karlp> just that it can be made.
2012-09-07T16:22:44 < karlp> if other things prevent reflashing....
2012-09-07T16:22:51 < Erlkoenig> haha nice :D
2012-09-07T16:22:53 < karlp> (bit of a dick move though)
2012-09-07T16:23:41 < Erlkoenig> well i just had the idea that companies *might* have a problem with revealing their symbols. like graphics card vendors not wanting to release their card's interfaces or something.
2012-09-07T16:23:42 < BrainDamage> doesn't the "v3" prevents that?
2012-09-07T16:24:12  * karlp shrugs
2012-09-07T16:24:14 < dongs> and this is why i use stdperiphlib.
2012-09-07T16:24:35 < karlp> isn't v3 just about the patent wars?
2012-09-07T16:25:06 < Erlkoenig> well uhm what again is the exact point of libopencm? why not just use CMSIS and StdPeriph?
2012-09-07T16:25:10 < BrainDamage> as far as I know, it was also made to prevent vendor lock-in
2012-09-07T16:25:35 < dongs> Erlkoenig: its free as in aids
2012-09-07T16:26:18 < Erlkoenig> no, you don't have to supply any .o files alongside your HI-Virus instances :P
2012-09-07T16:27:42 < BrainDamage> I think it'd be much wiser for them to switch to bsd
2012-09-07T16:27:55 < BrainDamage> this way even commercial companies could benefit from it
2012-09-07T16:28:47 < Erlkoenig> that's what i thought too...
2012-09-07T16:29:05 < BrainDamage> but some people bring it to a political stance
2012-09-07T16:29:23 < Erlkoenig> open source fanatics :D
2012-09-07T16:30:13 < BrainDamage> well, arguing with them puts you on the same level
2012-09-07T16:31:22 < Erlkoenig> instead of arguing i'd just use stdperiph ;)
2012-09-07T16:31:48 < dongs> < Erlkoenig> instead of arguing i'd just use stdperiph ;)
2012-09-07T16:31:55 < dongs> holy shit first dude in teh channel who makes sense
2012-09-07T16:32:17 < Erlkoenig> whaaaat
2012-09-07T16:32:40 < Erlkoenig> u wanna offend me?!?!
2012-09-07T16:33:03 < dongs> no, its a compliment you asshole
2012-09-07T16:33:05 < dongs> lrn2read
2012-09-07T16:33:17 < Erlkoenig> that... was the joke.
2012-09-07T16:41:46 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-07T16:45:05 < zyp> man, you guys are communicating as well as Laurenceb's sdcards
2012-09-07T16:46:01 < Erlkoenig> what do you think we are, girls? :D
2012-09-07T16:46:21 <+izua> my bet is on a delay 0x96 or whatever start byte they use
2012-09-07T16:46:30 <+izua> delayed*/late
2012-09-07T16:46:42 < Laurenceb> arg wtf
2012-09-07T16:46:52 < Laurenceb> 0k so ive replicated my sd card issue
2012-09-07T16:47:06 < Laurenceb> starting at 7753 byte offset into the file
2012-09-07T16:47:25 < Laurenceb> i get 0x01, 19bytes ok, 0x03, 19 ok, 0x00, 19ok
2012-09-07T16:47:43 < Laurenceb> then 0x01,03,00,02,01,02
2012-09-07T16:47:54 < Laurenceb> all at 20 byte spacings
2012-09-07T16:48:01 < Laurenceb> i.e. 19 ok bytes in between
2012-09-07T16:48:11 <+izua> dafuq
2012-09-07T16:48:16 <+izua> awesome
2012-09-07T16:49:01 < Laurenceb> i know whats happening and its repeatable
2012-09-07T16:49:05 < Laurenceb> but ... WTF
2012-09-07T16:49:11 <+izua> is it happening on a different card too?
2012-09-07T16:50:00 < Laurenceb> hmm good idea
2012-09-07T16:52:50 < Laurenceb> wtf
2012-09-07T16:52:56 < Laurenceb> usb wont mount now
2012-09-07T16:53:05 <+izua> neat
2012-09-07T16:53:11 <+izua> i broke your stuff without even touching it :D
2012-09-07T16:54:06 < Laurenceb> oh debug screw up
2012-09-07T16:54:14 < Laurenceb> helps to type run
2012-09-07T16:54:58 < Laurenceb> but why would it be every 20 bytes
2012-09-07T16:55:01 < Laurenceb> every odd
2012-09-07T16:55:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-07T16:56:17 < Laurenceb> what the fuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
2012-09-07T16:56:28 < Laurenceb> it needs jtag connected for usb to work
2012-09-07T16:57:23 < Laurenceb> laurence@desktop:/$ ls
2012-09-07T16:57:23 < Laurenceb> laurence@desktop:/$ cd media
2012-09-07T16:57:23 < Laurenceb> laurence@desktop:/media$ ls
2012-09-07T16:57:26 < Laurenceb> O K
2012-09-07T16:57:38 < Laurenceb> even my pc is fuxored
2012-09-07T16:57:51 < Erlkoenig> it's spreading :D
2012-09-07T16:58:18 < dongs> the aids of opensauce
2012-09-07T16:59:23 < Erlkoenig> http://xkcd.com/349/
2012-09-07T17:00:01 < Laurenceb> same issue on other card
2012-09-07T17:00:13 < zyp> Erlkoenig :D
2012-09-07T17:00:45 < dongs> i prefer this version goatkcd.com/349/
2012-09-07T17:06:35 < Laurenceb> yeah identical - 0x02/00/01/03 at 20 byte offsets
2012-09-07T17:08:23 <+izua> so you should be getting 02/00/01/03 bytes, one following the previous?
2012-09-07T17:09:32 < Laurenceb> no
2012-09-07T17:09:39 < Laurenceb> i have ascii data on there
2012-09-07T17:09:43 < Laurenceb> its a text file
2012-09-07T17:09:58 < Laurenceb> the characters are corrupted by 0x00/01/02/03
2012-09-07T17:10:11 < Laurenceb> at 20 byte intervals in the last half of a sector
2012-09-07T17:10:23 < Laurenceb> when this happens the card then becomes unresponsive
2012-09-07T17:10:34 < Laurenceb> and has to be restarted for the next sector
2012-09-07T17:10:45 < Laurenceb> i think theres data going out on mosi
2012-09-07T17:10:52 < Laurenceb> that si screwing with the card
2012-09-07T17:11:08 < Laurenceb> but im going from an experiment i did a few months ago
2012-09-07T17:11:14 < Laurenceb> im not going to set that up again
2012-09-07T17:11:25 < Laurenceb> been struggling with this bug for 6 months
2012-09-07T17:11:51 < Laurenceb> its so annying
2012-09-07T17:12:03 < Laurenceb> will crop up in the middle of many MB of data
2012-09-07T17:14:07 < Laurenceb> i guess i could set some gpio to disable mosi
2012-09-07T17:14:32 < dongs> you8 mean insteado  fixing it
2012-09-07T17:15:15 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-07T17:15:18 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-07T17:15:59 < Laurenceb> well im not going to be able to fix it
2012-09-07T17:16:06 < Laurenceb> as its an st hardware bug
2012-09-07T17:28:05 < Laurenceb> as far as i can see
2012-09-07T17:28:20 < Laurenceb> wtf  Util/fat_fs/src/sd_spi_stm32.c:338:29: error: 'GPIO_Pin_SD_MOSI' undeclared (first use in this function)
2012-09-07T17:28:32 < Laurenceb> its in discio.h which is included
2012-09-07T17:28:39 < Laurenceb> why wouldnt it see it?
2012-09-07T17:28:57 <+Steffanx> Typo?
2012-09-07T17:31:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-07T17:31:15 < Laurenceb> yes
2012-09-07T17:31:17  * Laurenceb fails
2012-09-07T17:32:05 <+Steffanx> You're welcome
2012-09-07T17:32:17 < Laurenceb> now work you stupid thing
2012-09-07T17:32:23 < Erlkoenig> nein
2012-09-07T17:32:29 <+Steffanx> jawohl
2012-09-07T17:40:31 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-07T17:44:22 < Laurenceb> hmm
2012-09-07T17:44:27 < Laurenceb> tried it 10 times
2012-09-07T17:44:33 < Laurenceb> with 10MB file
2012-09-07T17:44:40 < Laurenceb> no corrupted bytes found
2012-09-07T17:44:46 < Laurenceb> i cant be bothered to do this all day
2012-09-07T17:46:39 < Laurenceb> _seems_  to have fixed it
2012-09-07T17:46:48 < Laurenceb> it was failing about 70% of the time before
2012-09-07T17:47:01 < Laurenceb> just the weird jtag issue now
2012-09-07T17:47:18 < Laurenceb> with jtag unplugged it goes mad
2012-09-07T17:47:26 < Laurenceb> seems usb is power cycling
2012-09-07T17:57:43 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-07T17:57:44 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-07T17:57:44 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-07T17:57:47 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-07T17:59:01 <+izua> Laurenceb: this might be far fetched, but why not use a software spi?
2012-09-07T17:59:15 < Laurenceb> ewwwwwwwwwwwwww
2012-09-07T17:59:19 <+izua> yesbut
2012-09-07T17:59:26 < Laurenceb> what a horrifying idea
2012-09-07T17:59:31 <+izua> final user won't know what's under the hood
2012-09-07T17:59:37 <+izua> except if it's you
2012-09-07T17:59:38 < Laurenceb> lol
2012-09-07T17:59:39 <+izua> in which case
2012-09-07T17:59:59 < Laurenceb> hardware spi can use dma
2012-09-07T18:00:14 < Laurenceb> whilst the firmware faffs about with usb
2012-09-07T18:00:18 <+izua> well, it's better to have a working, but hard to maintain code, then awesome code and lots of fail :P
2012-09-07T18:00:46 <+izua> well then
2012-09-07T18:00:49 <+izua> you can do software sdio
2012-09-07T18:00:57 <+izua> at least you can fool yourself into hoping it will work faster :P
2012-09-07T18:01:45 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 276 seconds]
2012-09-07T18:03:38 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-07T18:04:12 <+izua> Laurenceb: i was planning to make a card reader with on-the-fly encryption, which would identify itself as a msc device
2012-09-07T18:04:17 < Laurenceb> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=https%3a%2f%2fmy.st.com%2fpublic%2fSTe2ecommunities%2fmcu%2fLists%2fcortex_mx_stm32%2fSTM32F103C8T6%20SPI%20problem&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=554
2012-09-07T18:04:19 < Laurenceb> blerg
2012-09-07T18:04:26 < Laurenceb> i cant be bothered to read tha
2012-09-07T18:04:31 <+Steffanx> The urls itself are blegh :P
2012-09-07T18:04:34 < Laurenceb> looks like i may have a known bug
2012-09-07T18:04:38 <+Steffanx> *is
2012-09-07T18:04:46 < Laurenceb> http://3.bp.blogspot.com/-HnkoHbTH7LQ/T2evo5BYZ1I/AAAAAAAAACM/UggCyzqXmME/s1600/meetJustin.jpg
2012-09-07T18:04:50 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-07T18:05:17 <+Steffanx> We know you are a bieber-fan Laurenceb
2012-09-07T18:06:41 -!- High-Fidelity [~High-Fide@123-243-232-159.static.tpgi.com.au] has quit [Ping timeout: 246 seconds]
2012-09-07T18:06:45 < dongs> ~haha stampede lunix is dead
2012-09-07T18:06:47 < dongs> i remember hearing about that
2012-09-07T18:07:08 < Laurenceb> linux on the desktop
2012-09-07T18:07:22 < Laurenceb> talking of which
2012-09-07T18:07:37 < Laurenceb> laurence@desktop:/$ ls
2012-09-07T18:07:52 < Laurenceb> laurence@desktop:/$ cd media
2012-09-07T18:07:53 <+Steffanx> Never heard of it dongs, so i guess it wasn't important
2012-09-07T18:07:57 < dongs> lol
2012-09-07T18:08:00 < Laurenceb> laurence@desktop:/$ ls
2012-09-07T18:08:05 < Laurenceb> laurence@desktop:/$ New Volume
2012-09-07T18:08:10 < Laurenceb> ^dafuk
2012-09-07T18:08:36 < dongs> disk label?
2012-09-07T18:08:46 < Laurenceb> it cant see /media
2012-09-07T18:08:53 < Laurenceb> or anything else
2012-09-07T18:11:31 -!- High-Fidelity [~High-Fide@123-243-232-159.static.tpgi.com.au] has joined ##stm32
2012-09-07T18:12:35 < Laurenceb> http://www.solid-run.com/store
2012-09-07T18:12:38 < Laurenceb> drool
2012-09-07T18:12:52 <+Steffanx> Nothing new ..
2012-09-07T18:14:53 < dongs> > 3- 2GByte microSD with Ubuntu pre-installed.
2012-09-07T18:14:55 < dongs> closed.
2012-09-07T18:15:29 < dongs> haha marvell soc
2012-09-07T18:15:30 < dongs> trash.
2012-09-07T18:16:43 <+Steffanx> Man.. you guys never seen that thing? It's probably as old as the rpi ( or at least ~ the time when the rpi became popular )
2012-09-07T18:17:24 < dongs> was that another kickstarted project
2012-09-07T18:17:33 < dongs> oshieeeet Solid-Run is a young startup company based in Israel which has been established end of 2010.
2012-09-07T18:17:51 <+Steffanx> Not a kickstarter project afaik
2012-09-07T18:18:22 < dongs> anyway, tehres a reason noone uses those marvel socs
2012-09-07T18:18:28 < dongs> and that reason is they suck ass
2012-09-07T18:18:34 <+Steffanx> if you say so
2012-09-07T18:18:50 < Laurenceb> anyways
2012-09-07T18:18:59 < Laurenceb> back to the importatn question
2012-09-07T18:19:02 < dongs> Steffanx: that cpu is from 2009
2012-09-07T18:19:09 < Laurenceb> why wont my usb connect
2012-09-07T18:19:43 < Laurenceb> if i have SWD connected it works fine
2012-09-07T18:20:01 < Laurenceb> without swd its not enumerating
2012-09-07T18:20:07 < Laurenceb> it tried but fails
2012-09-07T18:20:11 <+Steffanx> Cosmic intefrerence
2012-09-07T18:20:27 < Laurenceb> oh wait lol
2012-09-07T18:20:31 < Laurenceb> works on my other port
2012-09-07T18:20:43 < Laurenceb> my desktop is fuxored
2012-09-07T18:21:21 <+Steffanx> W00t msdnaa has vb6 D:
2012-09-07T18:21:43 < BrainDamage> vb6 rules
2012-09-07T18:21:50 < dongs> Steffanx: old news?
2012-09-07T18:22:04 < Laurenceb> visual basic?
2012-09-07T18:22:05 <+Steffanx> I never had access to it before dongs :)
2012-09-07T18:22:24 < dongs> pretty s ure VB6 has always been on msdn
2012-09-07T18:22:30 <+Steffanx> probably, yes
2012-09-07T18:22:33 < dongs> as well as visua sltudio 96 or wahtever the fuck it was bundled with
2012-09-07T18:22:36 <+izua> vb4 > vb6
2012-09-07T18:22:55 < Laurenceb> vbsqrt(-1)
2012-09-07T18:22:56 <+izua> Steffanx: it's a real thing bro
2012-09-07T18:23:10 <+Steffanx> "it" ?
2012-09-07T18:23:19 <+izua> people apparently study it too
2012-09-07T18:23:20 <+izua> https://en.wikipedia.org/wiki/Global_Consciousness_Project
2012-09-07T18:24:05 < Laurenceb> hahaha
2012-09-07T18:24:13 < Laurenceb> what the helll
2012-09-07T18:24:21 < dongs> more like global fucking trolling
2012-09-07T18:24:21 <+Steffanx> Laurenceb goes to princeton
2012-09-07T18:24:41 < Laurenceb> no oxford
2012-09-07T18:24:50 <+Steffanx> He'll show them his fancy LED  device
2012-09-07T18:25:14 < Laurenceb> blinkenlights
2012-09-07T18:25:23 < Erlkoenig> (17:23:20) izua: https://en.wikipedia.org/wiki/Global_Consciousness_Project <-- someone should make a wikipedia clone with a very similar URL, which only contains crazy hoaxes
2012-09-07T18:25:49 <+izua> WLKIPEDIA.COM?
2012-09-07T18:26:02 <+Steffanx> org
2012-09-07T18:26:06 <+izua> (your mileage might vary depending on your font)
2012-09-07T18:26:37 < Laurenceb> uncyclopedia
2012-09-07T18:26:46 <+Steffanx> WlKIPEDIA.ORG is btter in my case, izua
2012-09-07T18:26:51 <+Steffanx> *works better
2012-09-07T18:27:08 <+izua> reminds me of the recent xkcd
2012-09-07T18:27:16 <+izua> dongs: s/xkcd/goatkcd
2012-09-07T18:27:59 <+Steffanx> izua knows dongs too good
2012-09-07T18:28:44 < karlp> the marvel socs are used in seagate goflex devices,
2012-09-07T18:28:52 < karlp> also sheevaplug/guruplug
2012-09-07T18:29:01 < karlp> only problems they have is the power supply used with them.
2012-09-07T18:29:06 < karlp> they're otherwise just fine.
2012-09-07T18:29:29 < karlp> then it's just s difference in pricing.
2012-09-07T18:29:40 <+Steffanx> "No, they're crap" * more bloggery *
2012-09-07T18:29:55 < Laurenceb> aha
2012-09-07T18:30:02 < Laurenceb> my usb cable is fuxored
2012-09-07T18:30:12 < Laurenceb> it was free from XMOS
2012-09-07T18:30:18 <+izua> been there
2012-09-07T18:30:27 <+izua> first time i tested some usb example ._.
2012-09-07T18:30:44 < dongs> Domain Name:WLKIPEDIA.ORG
2012-09-07T18:30:44 < dongs> Created On:17-Dec-2005 05:06:36 UTC
2012-09-07T18:30:48 < dongs> too late.
2012-09-07T18:31:07 < Laurenceb> seems the spi was going wrong due to the data on mosi
2012-09-07T18:31:10 <+Steffanx> You know guys who can help us, dongs
2012-09-07T18:31:21 < Laurenceb> it loops through the shift register and out the other side
2012-09-07T18:31:27 < Laurenceb> fatfs code seems to be screwed
2012-09-07T18:31:40 < Laurenceb> dunno why it correleated with the buffers issue
2012-09-07T18:31:54 <+izua> isn't fatfs working at sector level?
2012-09-07T18:32:22 <+izua> maybe you can just buffer a whole request
2012-09-07T18:32:23 < Erlkoenig> that crazy "Coherent consciousness" stuff looks like the "Morphic fields" from torchwood...
2012-09-07T18:32:53 <+izua> and wait in the function that reads a whole sector to get it right from the card
2012-09-07T18:33:10 <+izua> i had some epic fails with nano(?)/tiny(?)/fatFS some months ago
2012-09-07T18:33:52 <+izua> Erlkoenig: well, at the very least it's an attempt to prove something like that exists (or doesn't)
2012-09-07T18:35:03 < Erlkoenig> one could argue something like that exists thanks to TV and the interwebz, but i doubt it influences such hardware RNG's ...
2012-09-07T18:35:47 < Laurenceb> izua: i read 512 byte sectors over dma
2012-09-07T18:36:03 < Laurenceb> then transfer 64 byte packets over usb up to the current transfer point
2012-09-07T18:36:57 <+izua> Laurenceb: yesbut
2012-09-07T18:37:02 < Laurenceb> sd spi spec says the card should get 0xff throughout the transfer
2012-09-07T18:37:03 < zyp> yesbutt
2012-09-07T18:37:04 -!- barthess [~barthess@178.154.97.136] has joined ##stm32
2012-09-07T18:37:10 < Laurenceb> pirate
2012-09-07T18:37:11 <+izua> cards have some pseudorandom delay before sending the start byte
2012-09-07T18:37:41 < Laurenceb> the fatfs code was echoing through the received data
2012-09-07T18:37:58 <+izua> it depends on how the 'read' is implemented
2012-09-07T18:37:59 < Laurenceb> for some reason it worked ok with the usb non interleaved
2012-09-07T18:38:21 < Laurenceb> probably as the read consisted of short bursts
2012-09-07T18:38:23 <+izua> since - at least with nanofatfs - you had to implemented the spi drivers yourself
2012-09-07T18:38:25 < Laurenceb> to memory
2012-09-07T18:38:45 < Laurenceb> yes - im using the stm32 drivers that are included with fatfs
2012-09-07T18:39:03 < Laurenceb> the st example spi code was mega lame
2012-09-07T18:39:14 < Laurenceb> single byte looped transfers
2012-09-07T18:39:21 < Laurenceb> that worked but it was super slow
2012-09-07T18:39:23 <+izua> so.. what happens if you echo 0xFF during a transfer
2012-09-07T18:39:42 < Laurenceb> i need a buffer filled with 0xff
2012-09-07T18:39:47 < Laurenceb> kind of a waste of ram
2012-09-07T18:40:51 <+izua> disable the pin as AFIO, set as pullup?
2012-09-07T18:40:56 < zyp> Laurenceb, why would you need that?
2012-09-07T18:41:05 <+izua> pullup + GPIO i mean, ofc
2012-09-07T18:41:14 < Laurenceb> yes
2012-09-07T18:41:19 < Laurenceb> thats what ive done
2012-09-07T18:41:21 <+izua> you'll get enough 0xFF to fill a few terabytes :P
2012-09-07T18:41:21 < zyp> just set the DMA read pointer to not increment, and point it to a single 0xff
2012-09-07T18:41:24 < Laurenceb> _seems_  to work
2012-09-07T18:41:34 < Laurenceb> zyp: neat
2012-09-07T18:41:38 < Laurenceb> ill try that
2012-09-07T18:41:48 < Laurenceb> better than gpio
2012-09-07T18:42:01 <+izua> you still waste one byte :P
2012-09-07T18:42:11 < zyp> two if writing 16-bit words :p
2012-09-07T18:42:14 < Laurenceb> yeah but gpio calls are fairly slow
2012-09-07T18:42:21 <+izua> not that a nonincrementing dma isn't cool, i had no idea it existed
2012-09-07T18:42:42 <+izua> Laurenceb: well, it's done only once per 512 bytes
2012-09-07T18:42:45 < Laurenceb> me neither
2012-09-07T18:42:52 < zyp> of course it exists, that's how you read from the SPI data register
2012-09-07T18:42:56 < karlp> how do you think dma from a periph is done izua? the address stays the same :)
2012-09-07T18:43:04 < Laurenceb> heh
2012-09-07T18:43:12 < Laurenceb> but both ends have to stay unchanged here
2012-09-07T18:43:18 <+izua> mhm
2012-09-07T18:43:25 < zyp> well, there is MINC and PINC bits for each end
2012-09-07T18:44:14 <+izua> it does make sense not to increment the address for a peripheral
2012-09-07T18:44:21 < Laurenceb> oh wait
2012-09-07T18:44:30 < Laurenceb> fatfs code already does this
2012-09-07T18:44:32 < Laurenceb> wtf
2012-09-07T18:44:41 < Laurenceb> oh crap
2012-09-07T18:44:45 < Laurenceb> i know whats wrong
2012-09-07T18:44:54 < Laurenceb> i overwrote the workbyte with stack
2012-09-07T18:45:01 <+izua> ...
2012-09-07T18:45:03 < Laurenceb> everything makes sense now
2012-09-07T18:45:28 <+izua> see, on avrs you have this awesome debug mechanism
2012-09-07T18:45:28 < Laurenceb> the dma is pointed to the stack
2012-09-07T18:45:33 <+izua> if you use too much ram
2012-09-07T18:45:35 <+izua> everything fucks up
2012-09-07T18:45:37 <+izua> instantly
2012-09-07T18:45:56 < zyp> izua, on ARM that's called a hardfault
2012-09-07T18:46:16 <+izua> yes but.. you need a debugger to see it
2012-09-07T18:46:20 < zyp> no
2012-09-07T18:46:25 <+izua> or at least write an isr for it
2012-09-07T18:46:29 < zyp> :)
2012-09-07T18:46:36 <+izua> s/isr/catch routine?
2012-09-07T18:46:47 <+izua> i know it's like an isr, but it has a special name. shame on me, i guess
2012-09-07T18:46:50 < zyp> I'd call it a hardfault handler
2012-09-07T18:47:28 <+izua> so hmm
2012-09-07T18:47:34 <+izua> i wonder if this solves Laurenceb's bug of 6 months
2012-09-07T18:47:57 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-07T18:48:04 < Laurenceb> i think so
2012-09-07T18:48:17 < Laurenceb> the dma was pointing to the stack
2012-09-07T18:50:47 < zyp> I'm planning to disallow DMA to stack in my stuff
2012-09-07T18:50:51 < Erlkoenig> (17:45:35) izua: everything fucks up <-- but you have to differ this fault from "you programmed something => everything fucks up" :D
2012-09-07T18:50:57 < zyp> because then I can place stacks in CCM
2012-09-07T18:51:11 < dongs> zyp, you cant DMA to CCM anyway.
2012-09-07T18:51:15 < zyp> exactly
2012-09-07T18:51:20 < Erlkoenig> CCM?
2012-09-07T18:51:31 < zyp> that's why it must not be allowed
2012-09-07T18:51:32 < dongs> 64k of core m emry shit
2012-09-07T18:51:44 < zyp> Erlkoenig, memory only accessible by cpu
2012-09-07T18:51:53 < Erlkoenig> ah
2012-09-07T18:51:59 < zyp> F4 got 64k of CCM and 128k of normal RAM
2012-09-07T18:52:32 < zyp> putting all thread's stacks in CCM seems like a nice way to get it utilized
2012-09-07T18:54:36 < dongs> yeah, it'llwork like that.
2012-09-07T18:54:44 < dongs> you just makea your own allocator-type shit
2012-09-07T18:55:15 < dongs> if you never free them you dont have to worry about fragmentation and shit
2012-09-07T18:55:39 < zyp> I was thinking static allocation
2012-09-07T18:56:06 < zyp> but yeah
2012-09-07T18:56:38 < dongs> just section(".ccm") or someshit on static var?
2012-09-07T18:56:45 < zyp> yep
2012-09-07T18:56:52 < dongs> sounds like gcc trash
2012-09-07T18:57:12 < karlp> #pragma section if you prefer :)
2012-09-07T18:57:14 < zyp> but really, allocating dynamically sounds like less code to write to create a thread, I like the sound of that
2012-09-07T18:58:02 < dongs> you can allocate a static block in ccm and just alloc chunks out of it keeping track.
2012-09-07T19:00:00 < zyp> I don't need to allocate a static block, I just let the linker script notify the allocator how much memory is available
2012-09-07T19:00:24 < dongs> well i guess.
2012-09-07T19:00:28 < dongs> if you dont plan to use it for anything else.
2012-09-07T19:00:38 < dongs> thread static data could go there too.
2012-09-07T19:01:06 -!- claude_work is now known as claude
2012-09-07T19:02:20 < zyp> I mean, the linker script just let the allocator know how much CCM is left over after putting global symbols into it, just like heap uses leftover ram
2012-09-07T19:02:45 < dongs> ah
2012-09-07T19:02:48 < dongs> yeah that'll work.
2012-09-07T19:03:24 < dongs> soon your shit will only work with gcc
2012-09-07T19:03:53 < zyp> soon? been that way all the time
2012-09-07T19:04:13 < dongs> yeah but you keepdoing more nad more gccisms
2012-09-07T19:05:19 < zyp> tell me when there is any other relevant compilers around that support C++11
2012-09-07T19:06:15 <+Steffanx> Don't most of them support it ( partially ) ?
2012-09-07T19:07:00 <+Steffanx> Oh, "relevant" nvm :P
2012-09-07T19:07:08 < zyp> clang might, but last time I checked, llvm support for cortex were not fully usable
2012-09-07T19:07:19 < zyp> cortex-m*
2012-09-07T19:08:34 < zyp> I don't know about dongs' beloved armcc, but I'm guessing that it will still be a long time before it will have C++11 support
2012-09-07T19:09:14 <+Steffanx> "Cross compilation in Clang is very broken at the moment. There was some effort to fix this last year, but I guess it lost momentum..."  LOLOL
2012-09-07T19:09:21 < zyp> and I'm not aware of any other compilers that are worth caring about
2012-09-07T19:09:55 <+Steffanx> A quote from a message of july 2012
2012-09-07T19:10:03 <+Steffanx> *on
2012-09-07T19:11:11 < zyp> oh, and of course clang support a lot of gccisms, so that might not be an issue either way
2012-09-07T19:12:21 <+Steffanx> You REALLY need C++11 ?
2012-09-07T19:12:30 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-07T19:13:23 < zyp> as I'm now using some syntax features that was added in C++11 here and there, it will not compile otherwise
2012-09-07T19:13:56 <+Steffanx> And it's hard to not use this new syntax?
2012-09-07T19:13:59 < TitanMKD> hi
2012-09-07T19:14:04 <+Steffanx> Lo
2012-09-07T19:14:45 < zyp> Steffanx, http://cgit.jvnv.net/laks/tree/usb/descriptor.h <- looked at this?
2012-09-07T19:14:59 <+Steffanx> Yeah, i've seen your fancy code
2012-09-07T19:16:03 < zyp> I couldn't do that like that without C++11
2012-09-07T19:16:47 <+Steffanx> Especially "constexpr auto pack(A a, R... r) -> X<A, decltype(pack(r...))> " looks nice :P
2012-09-07T19:16:49 <+Steffanx> *not*
2012-09-07T19:17:31 < zyp> don't mind how it's implemented, it's just a support function for the descriptor builders :p
2012-09-07T19:19:19 < zyp> the principle is simple, it takes a variable amount of arguments and packs them all together in a consecutive block of memory
2012-09-07T19:20:43 < dongs> blog of memory
2012-09-07T19:20:56 < zyp> it's done by recursively stuffing them into packed structs, but you don't have to care about that
2012-09-07T19:21:09 <+Steffanx> I do care :)
2012-09-07T19:21:29 < zyp> and it's way nicer than having to write a struct by hand specific to whatever descriptors you have
2012-09-07T19:23:09 < zyp> and then I can do neat tricks like line 96
2012-09-07T19:23:51 < zyp> it calculates the size of the descriptor with all subdescriptors and puts it into the wTotalLength field
2012-09-07T19:24:24 -!- dekar [~dekar@dslb-188-107-170-026.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-07T19:25:25 < zyp> I'm considering extending it to also do stuff like counting number of interfaces and putting it into bNumInterfaces
2012-09-07T19:26:02 < zyp> which all in all results in less stuff to care about when you want to just make a usb device
2012-09-07T19:26:41 <+Steffanx> And more 'cryptic' code for those who do :)
2012-09-07T19:26:46 <+Steffanx> *do care
2012-09-07T19:27:07 < zyp> it's not that cryptic
2012-09-07T19:27:20 < zyp> I mean, you do understand how it works?
2012-09-07T19:27:21 < Laurenceb> /dev/sdd:
2012-09-07T19:27:21 < Laurenceb>  Timing buffered disk reads:    4 MB in  5.59 seconds = 732.92 kB/sec
2012-09-07T19:27:22 < Laurenceb> nice
2012-09-07T19:27:46 <+Steffanx> Not exactly, but i'm not into all new things of zyp
2012-09-07T19:27:51 <+Steffanx> Have to read/play more with it
2012-09-07T19:28:06 <+Steffanx> *zyp = c++11 :P
2012-09-07T19:28:44 <+Steffanx> Is it windows and os x compatible Laurenceb ?
2012-09-07T19:28:46 < zyp> well, here's the short version: X is a templated struct, containing two variables of arbitrary types (A and B)
2012-09-07T19:28:53 < Laurenceb> Steffanx: yes
2012-09-07T19:29:07 <+Steffanx> And you know that because you tested it?
2012-09-07T19:29:12 < Laurenceb> yes :P
2012-09-07T19:29:16 <+Steffanx> k
2012-09-07T19:29:25 < Laurenceb> one of my friends actually runs windows
2012-09-07T19:29:27 < Laurenceb> fool
2012-09-07T19:29:35 < Laurenceb> and i found a macfag
2012-09-07T19:30:16 < zyp> there is two variants of the pack() function; one taking a single argument, and returns it right back again, and one taking an arbitrary number of arguments > 1
2012-09-07T19:30:30 < zyp> pack(1) will return 1
2012-09-07T19:30:52 < zyp> pack(1, 2) will do X(1, pack(2)) which becomes X(1, 2)
2012-09-07T19:31:36 < zyp> pack(1, 2, 3) will do X(1, pack(2, 3)) which becomes X(1, X(2, pack(3))), then X(1, X(2, 3))
2012-09-07T19:31:42 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-07T19:31:45 < zippe> Steffanx: "relevant" would be clang
2012-09-07T19:32:23 < zyp> since X is a packed struct, every elements end up sitting right next to the others
2012-09-07T19:32:26 <+Steffanx> Yes, but all the (for me) unknown c++11 stuff makes it 'cryptic' zyp :)
2012-09-07T19:33:00 < zippe> zyp: are you tagging the struct with the packed attribute, or its members?
2012-09-07T19:33:06 < zippe> The former is wrong, btw.
2012-09-07T19:33:18 < zyp> zippe, is it?
2012-09-07T19:33:34 < zyp> [citation needed]
2012-09-07T19:33:37 < zippe> Unless you plan on packing the struct inside another struct
2012-09-07T19:33:59 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-09-07T19:34:33 < zippe> zyp: hah, and they've gone and rewritten the documentation; please ignore.  Historically it did more than just what you wanted.
2012-09-07T19:34:45 < zyp> :)
2012-09-07T19:34:56 < zippe> zyp: however as currently documented it won't pessimise code that accesses well-aligned interior members
2012-09-07T19:35:14 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-229-156.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-07T19:35:49 < zyp> pessimise?
2012-09-07T19:36:01 < zippe> historically it would cause gcc to emit maximally paranoid code for member accesses, assuming that even members that were well-aligned vs. the base of the structure would be read using byte operations
2012-09-07T19:36:05 < zyp> as in the opposite of optimise? :p
2012-09-07T19:36:10 < zippe> on any platform with alignment constraints
2012-09-07T19:36:49 < zippe> zyp: correct
2012-09-07T19:37:00 < zyp> never heard anybody use that term before
2012-09-07T19:37:14 < zippe> #pragma pack(push, 1) / #pragma pack(pop) have long been the workaround
2012-09-07T19:37:44 < zippe> But if ((packed)) actually works less badly now, that's neat.  Thanks for making me go check. 8)
2012-09-07T19:39:42 < zyp> also, member accesses is not really relevant in this case, there is no code emitted
2012-09-07T19:40:26 < zyp> descriptors are generated compile time and emitted as a tree of packed structs, and then just read as a buffer
2012-09-07T19:40:45 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-07T19:46:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-07T19:52:18 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-07T19:52:37 < Laurenceb> http://pic.epicfail.com/wp-content/uploads/2010/02/keep-dogs-leashed-sign-fail1.jpg
2012-09-07T19:54:47 < emeb> I don't see the fail there.
2012-09-07T20:06:28 < Laurenceb> yeah more like epic win
2012-09-07T20:08:42 < zippe> zyp: you still have to put data into, and take data out of the struct at some point...
2012-09-07T20:08:59 < Laurenceb> how does c handle rounding floats when typecasting to int?
2012-09-07T20:09:18 < zyp> zippe, it's put into at compile time and read out as a continous buffer
2012-09-07T20:09:25 < BrainDamage> it should just truncate
2012-09-07T20:09:48 < BrainDamage> so implicit floor
2012-09-07T20:09:58 < Laurenceb> oh
2012-09-07T20:09:59 < Laurenceb> http://www.cs.tut.fi/~jkorpela/round.html
2012-09-07T20:10:05 < Laurenceb> yeah i see
2012-09-07T20:10:12 < zippe> zyp: damn, and here I was hoping you were talking about generating dynamic types.  Oh well. 8)
2012-09-07T20:11:25 < zippe> BrainDamage: floor is round down
2012-09-07T20:11:32 < zippe> BrainDamage: truncation is round towards zero
2012-09-07T20:12:32 < BrainDamage> not sure if I understand the difference, care to make a short example
2012-09-07T20:12:34 < BrainDamage> ?
2012-09-07T20:12:40 < zippe> floor (-1.5) is -2
2012-09-07T20:12:50 < zippe> trunc(-1.5) is -1
2012-09-07T20:13:06 < BrainDamage> ok, thanks
2012-09-07T20:13:42 < BrainDamage> I didn't consider negative numbers
2012-09-07T20:16:42 < zippe> 8)
2012-09-07T20:19:48 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-07T20:23:57 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 252 seconds]
2012-09-07T20:26:16 < Laurenceb> how does integer division differ?
2012-09-07T20:26:51 < emeb> interestingly, in 2s complement fixed point, truncation is same as floor.
2012-09-07T20:27:08 < Laurenceb> so if i have e.g. 6047/2?
2012-09-07T20:27:14 < Laurenceb> does it round up or down?
2012-09-07T20:27:28 < BrainDamage> integer division for power of 2 is implemented using bitshift
2012-09-07T20:27:37 < zyp> integer division always truncates
2012-09-07T20:27:56 < zyp> well
2012-09-07T20:28:04 < zyp> positive integer division always truncates.
2012-09-07T20:28:23 < zyp> I can't remember the rules for negative
2012-09-07T20:28:38 < BrainDamage> for non power of 2, you actually build a convergent series to invert the number
2012-09-07T21:28:21 -!- barthess [~barthess@178.154.97.136] has quit [Remote host closed the connection]
2012-09-07T21:59:10 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-07T22:13:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-07T22:16:30 -!- prog [prog@94.100.30.150] has joined ##stm32
2012-09-07T22:19:36 < karlp> and if you want to read all about interger behaviour in C, and defined and undefined, read all of this: http://blog.regehr.org/
2012-09-07T22:20:14 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-61.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-07T22:20:31 < karlp> well, not necessarily all of  it, this at least: http://blog.regehr.org/archives/category/software-correctness
2012-09-07T22:23:19 -!- barthess [~barthess@178.154.81.134] has joined ##stm32
2012-09-07T22:27:45 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-09-07T22:50:40 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-07T22:56:14 -!- barthess [~barthess@178.154.81.134] has quit [Quit: Leaving.]
2012-09-07T23:17:00 -!- barthess [~barthess@178.154.29.240] has joined ##stm32
--- Day changed Sat Sep 08 2012
2012-09-08T00:42:02 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-61.pptp.stw-bonn.de] has joined ##stm32
2012-09-08T00:49:10 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-08T01:01:28 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-08T01:02:23 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-08T01:06:01 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-09-08T01:08:22 -!- barthess [~barthess@178.154.29.240] has quit [Quit: Leaving.]
2012-09-08T01:09:39 -!- barthess [~barthess@164.177.216.122] has joined ##stm32
2012-09-08T01:17:11 < Laurenceb_> By getting more from technology to get more from life, ST stands for life.augmented.
2012-09-08T01:17:14 < Laurenceb_> obviously
2012-09-08T01:23:25 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-09-08T01:37:09 -!- NonaSuomy is now known as TeknoJuce
2012-09-08T02:13:10 -!- barthess [~barthess@164.177.216.122] has quit [Quit: Leaving.]
2012-09-08T02:49:09 < Laurenceb_> do chibios adc callback functions get a pointer direct to the dma buffer?
2012-09-08T03:03:05 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-09-08T03:12:57 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-08T03:48:36 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-08T03:48:41 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-08T03:48:41 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-08T03:48:44 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-08T04:14:22 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-08T04:22:56 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-08T04:37:04 -!- dekar [~dekar@212.255.26.121] has joined ##stm32
2012-09-08T04:37:07 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-08T04:39:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-08T04:49:33 < dongs> haha pjrc.com has a pic of my hardware in their 'counterfeit' section
2012-09-08T04:49:37 < dongs> http://www.pjrc.com/teensy/counterfeit_3.jpg
2012-09-08T04:54:15 -!- R2COM [~blackops@98.230.217.221] has joined ##stm32
2012-09-08T04:59:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-08T05:08:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-08T05:13:37 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-08T05:59:49 < emeb_mac> zzzzzzz.......
2012-09-08T06:02:55 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-09-08T06:06:33 < dongs> sup dongs
2012-09-08T06:08:07 < R2COM> lots of stuff
2012-09-08T06:15:15 < emeb_mac> where does dongs get prototype PCBs made?
2012-09-08T06:17:38 < emeb_mac> lol
2012-09-08T06:18:27 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-61.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-08T06:23:56 < dongs> emeb_mac: local place.
2012-09-08T06:24:01 < dongs> i usually panel shit with other customers orders.
2012-09-08T06:24:14 <+dekar> dongs, how's that even related to the teensy?
2012-09-08T06:25:09 < dongs> dekar: its back from ps3hax days
2012-09-08T06:25:20 < dongs> when they used some lufa? shit on teensy to do ps3 hax via usb
2012-09-08T06:25:29 < dongs> so every chink made aclone with just usb port and couple leds
2012-09-08T06:25:49 <+dekar> I used a teensy++ to read/write the nor flash on my ps3
2012-09-08T06:26:02 < dongs> in 8bit mode?
2012-09-08T06:26:20 <+dekar> dongs, https://dl.dropbox.com/u/28467113/ps3/IMG_20120411_143829.jpg
2012-09-08T06:26:21 < dongs> how did you get ~20 address lines?
2012-09-08T06:26:34 < dongs> rawl
2012-09-08T06:26:46 <+dekar> https://dl.dropbox.com/u/28467113/ps3/IMG_20120411_143856.jpg
2012-09-08T06:26:46 < dongs> i guess you did!
2012-09-08T06:27:59 <+dekar> https://dl.dropbox.com/u/28467113/ps3/IMG_20120411_171030.jpg
2012-09-08T06:28:18 < dongs> why a permanent install?
2012-09-08T06:28:23 < dongs> (i dont own a ps3 so i dont know)
2012-09-08T06:28:42 <+dekar> dunno, cause it easily fit inside
2012-09-08T06:28:51 < dongs> https://s3.amazonaws.com/digispark/newdesign.jpg
2012-09-08T06:29:09 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-08T06:29:12 <+dekar> I don't really use the ps3 anyway, someone gave it to me and I felt like hacking it :)
2012-09-08T06:29:13 < dongs> how do I 7805'd
2012-09-08T06:29:15 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-08T06:50:37 -!- tech2077 [~tech2077@adsl-75-53-141-174.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-08T06:59:24 < dongs> tech2077: any more ghetto crackberry pi hacks
2012-09-08T06:59:38 < tech2077> nope
2012-09-08T06:59:48 < tech2077> not really
2012-09-08T07:00:08 < tech2077> i got a free beaglebone from Ti and was basically told to work with it instead of the pi
2012-09-08T07:12:45 < dongs> i guess i should make a board for the ws2811
2012-09-08T07:12:51 < dongs> and some leds
2012-09-08T07:12:56 < dongs> or else how im gonna prototype it
2012-09-08T07:13:14 < dongs> finally flyback is gone
2012-09-08T07:13:24 < dongs> now i can throw away these dumb chinese chips
2012-09-08T07:23:20 < dongs> who
2012-09-08T07:24:25 < R2COM> ok bring on some interesting stuff
2012-09-08T07:24:42 < R2COM> dongs have you met some japanese girl already?
2012-09-08T07:25:19 < dongs> im only 13 so im only interested in dudes for now
2012-09-08T07:25:29 < R2COM> oh ok
2012-09-08T07:41:34 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-08T07:46:37 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-08T08:19:52 < dongs> argh pinout of that shit is designed so the led is on opposite side of hte PC
2012-09-08T09:12:33 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-08T09:14:54 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-08T09:41:11 < dongs> http://i.imgur.com/nzK8m.png
2012-09-08T09:44:48 < jpa-> aren't led strips already available everywhere?
2012-09-08T09:45:06 < dongs> im gonna test some chink led driver.
2012-09-08T09:45:30 < dongs> that claims to work over 20 meter distances between transceivers.
2012-09-08T09:45:38 < dongs> this is interesting for my purposes.
2012-09-08T09:47:29 < jpa-> ah yeah, at next orgy, led on every dong?
2012-09-08T09:54:37 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-09-08T09:54:52 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-08T09:58:04 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-08T09:58:25 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds]
2012-09-08T10:33:17 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-08T10:46:13 -!- claude is now known as Claude
2012-09-08T11:07:08 <+dekar> ws2811 was the SPI one right?
2012-09-08T11:07:32 <+dekar> I found some other one that controls 3 RGB LEDs per IC
2012-09-08T11:08:43 <+dekar> TM1809
2012-09-08T11:10:09 < dongs> dekar: not spi, onewire
2012-09-08T11:11:37 < dongs> haha o shit
2012-09-08T11:11:43 < dongs> it looks like 3 channel version of the WS2811
2012-09-08T11:12:03 < dongs> ah but separate voltage uspply
2012-09-08T11:12:05 < dongs> supply
2012-09-08T11:15:25 <+dekar> I know it isn't perfect, but the chinese told me it was cheap
2012-09-08T11:17:03 < dongs> yeah im sure it is.
2012-09-08T11:17:07 < dongs> since 2811 is liek 16cents
2012-09-08T11:17:11 < dongs> that one's probably liek 20
2012-09-08T11:17:12 < dongs> heh
2012-09-08T11:17:46 < dongs> only thing is...
2012-09-08T11:17:51 < dongs> that shit needs R per led??
2012-09-08T11:18:48 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-08T11:19:07 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-08T11:20:23 <+dekar> looks that way
2012-09-08T11:32:32 -!- barthess [~barthess@178.154.38.184] has joined ##stm32
2012-09-08T11:41:08 < dongs> http://coocox.org/Cox.html ooOooooo
2012-09-08T11:41:19 < dongs> finally a HAL with a name I can really enjoy
2012-09-08T11:47:13 < dongs> http://coocox.org/Cookie.html what the hell
2012-09-08T11:53:17 -!- R2COM [~blackops@98.230.217.221] has quit [Quit: Leaving.]
2012-09-08T11:59:06 <+dekar> what is the advantage of coocox? isn't that just GCC?
2012-09-08T12:00:26 <+dekar> they state the arduino only has a "simplified C language"
2012-09-08T12:00:33 <+dekar> what's that supposed to mean?
2012-09-08T12:52:42 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-08T13:05:11 < High-Fidelity> does anyone use a graphic interface to GDB when debugging using opened?
2012-09-08T13:06:04 < High-Fidelity> hmm… I mean openocd… got autocorrected
2012-09-08T13:09:21 <+dekar> I used eclipse
2012-09-08T13:11:23 < High-Fidelity> yeah… I'm trying to get away from eclipse… want to develop using vim… prefer it… I guess I can use the text interface
2012-09-08T13:11:27 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has joined ##stm32
2012-09-08T13:12:26 < BrainDamage> there should be a vim plugin for gdb
2012-09-08T13:13:31 < High-Fidelity> good point… forgot about that
2012-09-08T13:15:08 < High-Fidelity> Am getting back into development after a bit of a break… got everything running on windows using eclipse and yagarto.  My preference is to boot into OSX and work there though, windows is no fun for development
2012-09-08T13:15:40 -!- BjoernC_ [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has joined ##stm32
2012-09-08T13:15:51 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has quit [Ping timeout: 245 seconds]
2012-09-08T13:17:38 -!- BjoernC_ [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-08T13:21:05 -!- barthess [~barthess@178.154.38.184] has quit [Read error: Connection reset by peer]
2012-09-08T13:21:52 -!- barthess [~barthess@178.154.40.8] has joined ##stm32
2012-09-08T13:23:24 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-08T13:52:27 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-08T13:52:30 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-08T14:03:26 -!- Claude is now known as claude
2012-09-08T14:13:50 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-08T14:17:40 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has joined ##stm32
2012-09-08T14:23:37 < dongs> sup trolls
2012-09-08T14:24:07 <+Steffanx> Tha sky/ceiling of whatever is above you. Hi dongs
2012-09-08T14:27:19 -!- tavish [~tavish@120.56.238.134] has joined ##stm32
2012-09-08T14:27:20 -!- tavish [~tavish@120.56.238.134] has quit [Changing host]
2012-09-08T14:27:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-08T14:32:17 -!- High-Fidelity [~High-Fide@123-243-232-159.static.tpgi.com.au] has quit [Quit: High-Fidelity]
2012-09-08T14:38:04 < dongs> bathtime
2012-09-08T14:40:23 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-08T14:41:50 <+Steffanx> Have fun dongs
2012-09-08T15:05:47 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Read error: Connection reset by peer]
2012-09-08T15:07:08 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-08T15:07:11 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-08T15:10:54 < dongs> if I have 4 channels of timer on F1 pins
2012-09-08T15:10:59 < dongs> i mean TIMx_CH1..4
2012-09-08T15:11:11 < dongs> and I only use say CH1 and CH4 and dont initialize 2+3
2012-09-08T15:11:17 < dongs> and there's UART on alternate function on those pins
2012-09-08T15:11:22 < dongs> will uart work?
2012-09-08T15:11:29 < dongs> or does TIM take over all the pins when enabled
2012-09-08T15:11:38 < dongs> i mean i could find out but im lazy
2012-09-08T15:12:21 <+Steffanx> Yes, no
2012-09-08T15:13:14 < dongs> ok. time to wire my shit up that way and try then.
2012-09-08T15:13:26 < dongs> so if I  dont setup input capture or output compare on those 2 channels im ok?
2012-09-08T15:21:25 < Laurenceb_> theres an enable pwm bit
2012-09-08T15:21:45 < Laurenceb_> you can even have a pwm channel running and the io set to AF
2012-09-08T15:21:48 < Laurenceb_> and use the usart
2012-09-08T15:21:59 < Laurenceb_> if you dont enable the pwm output in the timer register
2012-09-08T15:22:19 < Laurenceb_> anyone here understand Sequential streams in chibios?
2012-09-08T15:24:07 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-8.pptp.stw-bonn.de] has joined ##stm32
2012-09-08T15:25:10 < dongs> Laurenceb_: thas what im hoping.
2012-09-08T15:25:14 < dongs> writing hacks to fix that now.
2012-09-08T15:29:47 < Laurenceb_> oh...
2012-09-08T15:30:07 < Laurenceb_> sequential stream is just a bunch of function pointers in chibios?
2012-09-08T15:30:13 < dongs> no idea
2012-09-08T15:30:16 < dongs> why are you fucking aroudn with chibios
2012-09-08T15:30:25 < dongs> did you see CoOS now has a proper hal
2012-09-08T15:30:27 < dongs> and it has a great name
2012-09-08T15:30:30 < dongs> COX
2012-09-08T15:30:30 < dongs> so..
2012-09-08T15:30:36 < dongs> fuck this animu shit, switch to the real communist stuff.
2012-09-08T15:30:53 < dongs> and its portable across liek 10 cortex MCUs
2012-09-08T15:31:04 < dongs> so when you ccant buy stm32 anymore you just recompile for some chinese nuvoton shit like nothing happened
2012-09-08T15:32:01 <+Steffanx> Whaat? And you don't use the CooCox IDE?!
2012-09-08T15:32:14 < dongs> nah. cuz its jsut a shitty rip of eclipse :p
2012-09-08T15:32:58 <+Steffanx> but you do use CooCox CoOS?
2012-09-08T15:33:28 < dongs> ya
2012-09-08T15:33:40 < dongs> my osdproj uses it
2012-09-08T15:36:38 < Laurenceb_> chibios works on my F4 and has usb support
2012-09-08T15:36:53 < Laurenceb_> but i cant work out if the usb read/write is blocking or not
2012-09-08T15:41:52 < dongs> so basically youre failing
2012-09-08T15:43:30 < Laurenceb_> https://github.com/mabl/ChibiOS/blob/master/os/various/shell.c#L245
2012-09-08T15:43:34 < Laurenceb_> so it blocks?
2012-09-08T15:43:35 <+Steffanx> Everyone does that.. some more than others :)
2012-09-08T15:43:50 < Laurenceb_>  if (chSequentialStreamRead(chp, (uint8_t *)&c, 1) == 0)
2012-09-08T15:43:50 < Laurenceb_>       return TRUE;
2012-09-08T15:44:07 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-08T15:44:15 < Laurenceb_> means read only returns zero if the connection is terminated
2012-09-08T15:44:24 < Laurenceb_> so otherwise it must block
2012-09-08T15:44:28 < Laurenceb_> yes?
2012-09-08T15:44:33 < jpa-> Laurenceb_: of course USB serial blocks if you use the blocking io
2012-09-08T15:44:43 < jpa-> there is separate function with timeout parameter
2012-09-08T15:44:50 < Laurenceb_> really?
2012-09-08T15:45:36 < jpa-> http://chibios.sourceforge.net/html/group__io__channels.html
2012-09-08T15:46:27 < Laurenceb_> hmm
2012-09-08T15:46:28 < Laurenceb_> #define 	chIOReadTimeout(ip, bp, n, time)   ((ip)->vmt->readt(ip, bp, n, time))
2012-09-08T15:46:38 < Laurenceb_> implied the usb code supports timeout?
2012-09-08T15:46:50 < Laurenceb_> - implies it has to support timeout for that to work
2012-09-08T15:47:22 < jpa-> of course it supports timeout
2012-09-08T15:47:41 < Laurenceb_> how do you know?
2012-09-08T15:47:45 < jpa-> i've used it
2012-09-08T15:47:50 < Laurenceb_> wheres the CDC source i cant find it
2012-09-08T15:48:03 < jpa-> and it would be one shitty USB serial implementation if it didn't support it
2012-09-08T15:48:08 < Laurenceb_> lol
2012-09-08T15:48:38 < jpa-> chibios/os/hal/src/serial_usb.c
2012-09-08T15:50:18 < Laurenceb_> ah i see
2012-09-08T15:50:37 < Laurenceb_> argg
2012-09-08T15:50:46 < Laurenceb_> but readt is a different function
2012-09-08T15:50:57 < Laurenceb_> where is the sequential sptream initialised?
2012-09-08T15:51:28 < jpa-> the virtual methods table is in that file, static const struct SerialUSBDriverVMT vmt = {
2012-09-08T15:51:33 < jpa-> and readt is there also
2012-09-08T15:52:07 < Laurenceb_> yeah
2012-09-08T15:52:21 < Laurenceb_> but how do i force the use of readt rather than read?
2012-09-08T15:52:29 < Laurenceb_> some kind of config somewhere....
2012-09-08T15:53:21 < jpa-> for what? sequentialstream?
2012-09-08T15:53:43 < jpa-> yeah, that is a weak point; the shell is written so that it always blocks
2012-09-08T15:53:56 < jpa-> basically you need to make a wrapper stream if you want to modify the blocking behaviour
2012-09-08T15:54:47 < Laurenceb_> whats that?
2012-09-08T15:55:43 < jpa-> a stream of your own that implements the sequentialstream (or even io channel) interface
2012-09-08T15:55:47 < Laurenceb_> arg wtf
2012-09-08T15:55:55 < jpa-> but why do you want the shell to timeout?
2012-09-08T15:55:57 < Laurenceb_> i cant even find the sheel example
2012-09-08T15:56:13 < Laurenceb_> so i dont have to spawn a dedicated thread for usb
2012-09-08T15:56:22 < jpa-> ok
2012-09-08T15:56:24 < Laurenceb_> and then impliment io queues to it or something
2012-09-08T15:56:29 < Laurenceb_> makes things simpler
2012-09-08T15:56:50 < jpa-> i have a wrapper stream here: https://svn.kapsi.fi/jpa/sphere2/src_common/powersave.c
2012-09-08T15:57:03 < Laurenceb_> thanks
2012-09-08T15:57:11 < Laurenceb_> it'd help if i didnt keep getting lost
2012-09-08T15:57:16 < jpa-> (it checks when the user types something and resets a powerdown timeout; for your case, just call readt from reads)
2012-09-08T15:57:23 < Laurenceb_> any idea where the F4 usb-cdc sheel example is?
2012-09-08T15:57:35 < jpa-> you mean: < Laurenceb> it'd help if i didnt keep failing
2012-09-08T15:57:49 < Laurenceb_> lol
2012-09-08T15:57:56 < Laurenceb_> i cant find it :(
2012-09-08T15:58:04 < Laurenceb_> had it 5 minutes ago
2012-09-08T15:58:41 < jpa-> i haven't checked out the F4 usb support yet
2012-09-08T15:58:56 < Laurenceb_> oh
2012-09-08T15:58:57 < Laurenceb_> https://github.com/mabl/ChibiOS/tree/master/testhal/STM32F4xx/USB_CDC
2012-09-08T15:59:21 < jpa-> demos/ARMCM3-STM32F103-FATFS/  is a shell example for F1, should have the same API
2012-09-08T15:59:44 < dongs> im gonna look into this CoX thing
2012-09-08T15:59:53 < dongs> its probably much better than this freetard shit
2012-09-08T16:00:08 < jpa-> dongs: isn't it equally free?
2012-09-08T16:00:20 < dongs> jpa-: its free but it not kawaii
2012-09-08T16:00:23 < jpa-> and it even has a google-imitated logo
2012-09-08T16:00:33 < Laurenceb_> https://github.com/mabl/ChibiOS/blob/master/testhal/STM32F4xx/USB_CDC/main.c#L470
2012-09-08T16:00:44 < dongs> jpa-: coocox you mean?
2012-09-08T16:00:49 < dongs> is that google font lol
2012-09-08T16:00:57 < Laurenceb_> kooklux
2012-09-08T16:01:00 < jpa-> dongs: yeah
2012-09-08T16:01:37 < Laurenceb_> so the SDU things is basically some function pointers?
2012-09-08T16:01:51 < Laurenceb_> and i want to change the read to the timeout read?
2012-09-08T16:02:27 < jpa-> Laurenceb_: nah, the virtual methods table is const; you want to create a new stream that then calls the SDU stuff
2012-09-08T16:02:35 < zyp> I'm thinking about writing an ACM driver and demo project today - just writing it here so I'll remember it :p
2012-09-08T16:02:39 < jpa-> and streams in chibios are just a bunch of function pointers
2012-09-08T16:02:47 < Laurenceb_> yeah
2012-09-08T16:02:52 < Laurenceb_> so cant i change the pointers?
2012-09-08T16:03:57 < Laurenceb_> so it points to readt rather than read?
2012-09-08T16:04:09 < Laurenceb_> where is it defined that it should used read?
2012-09-08T16:04:10 < jpa-> not in runtime because it is const table
2012-09-08T16:04:13 < Laurenceb_> sure
2012-09-08T16:04:24 < jpa-> you can go around modifying the chibios source, of course
2012-09-08T16:04:24 < Laurenceb_> but where is the file that defines read not readt?
2012-09-08T16:04:28 < Laurenceb_> yes
2012-09-08T16:04:44 < jpa-> well the shell calls sequential read, which is supposed to block
2012-09-08T16:05:00 < jpa-> and the serial_usb.c has a blocking function at that point of the vmt
2012-09-08T16:05:20 < Laurenceb_> hmm
2012-09-08T16:05:38 < Laurenceb_> doesnt the shell just call whatever is in SDU1?
2012-09-08T16:05:48 < Laurenceb_> which is configured for it
2012-09-08T16:05:56 < jpa-> whatever is in the stream you give it
2012-09-08T16:06:26 < Laurenceb_> yeah
2012-09-08T16:06:58 < Laurenceb_> hmm ok maybe its most sane to "create a new stream"
2012-09-08T16:07:04 < Laurenceb_> but what do you mean?
2012-09-08T16:07:13 < dongs> btw, uart+TIm shit works
2012-09-08T16:07:26 < dongs> with my super leet pwm in/out driver, I didnt even have to do anything
2012-09-08T16:07:27 < Laurenceb_> cool
2012-09-08T16:07:28 < jpa-> Laurenceb_: whatever i did in powersave.c
2012-09-08T16:07:28 < dongs> just skip those 2 channels
2012-09-08T16:07:36 < dongs> while loading the pwm map
2012-09-08T16:07:59 < jpa-> it's a bit of boilerplate the way i did it; there are probably easier ways
2012-09-08T16:08:16 < dongs> http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/drv_pwm.c#481
2012-09-08T16:09:07 < Laurenceb_> jpa-: so i define my own equivalent of the "BaseSequentialstream"  ?
2012-09-08T16:09:24 < jpa-> Laurenceb_: you make your own *implementation* of basesequentialstream
2012-09-08T16:09:39 < Laurenceb_> and that "stream" -  i.e. function pointer thingy calls readt
2012-09-08T16:10:12 < jpa-> yeah, or if you are lazy, just chIOReadTimeout
2012-09-08T16:10:21 < Laurenceb_> then i use chIOReadTimeout to read from it?
2012-09-08T16:10:23 < Laurenceb_> yeah
2012-09-08T16:10:37 < Laurenceb_> so it points to readt rather than read for reading
2012-09-08T16:10:40 < jpa-> i had the original stream as a pointer, but you can just hardcode SDU1
2012-09-08T16:11:07 < Laurenceb_> and i call chIOReadTimeout rather than the other ch read function
2012-09-08T16:11:26 < Laurenceb_> in the example code iirc SDU1 is defined in main.c
2012-09-08T16:11:36 < Laurenceb_> and initialised with the stock SDUinit code
2012-09-08T16:12:00 < Laurenceb_> then its passed to sheel, which has its own thread so its ok for it to block
2012-09-08T16:12:03 < Laurenceb_> *shell
2012-09-08T16:12:11 < jpa-> yep
2012-09-08T16:12:11 < dongs> why teh fuck do you keep writing sheel
2012-09-08T16:12:48 < Laurenceb_> i dunno lol
2012-09-08T16:13:25 < Laurenceb_> is there an example of how to init my custom SDU stream thingy?
2012-09-08T16:13:55 < jpa-> you can make your own examples :)
2012-09-08T16:15:34 < Laurenceb_> argggg
2012-09-08T16:15:35 < Laurenceb_> http://chibios.sourceforge.net/html/group__data__streams.html#ga5323b44ccf2eee6e0f3ef7b22fb55eeb
2012-09-08T16:15:39 < Laurenceb_> its not going to work
2012-09-08T16:15:48 < jpa-> how so?
2012-09-08T16:16:10 < jpa-> of course you don't put the readt in your read pointer directly (the types don't match), you create a wrapper function where you specify the timeout
2012-09-08T16:16:10 < Laurenceb_> oh i know how to do it
2012-09-08T16:16:22 < Laurenceb_> use a BaseSequentialStream to write
2012-09-08T16:16:29 < Laurenceb_> and whatever the other thing is the read
2012-09-08T16:16:40 < dongs> does openOCD do SWD over FTDI-ish type shit?
2012-09-08T16:16:44 < Laurenceb_> BaseChannel
2012-09-08T16:16:46 < dongs> i.e. usb wiggler type crap
2012-09-08T16:16:51 < Laurenceb_> use BaseChannel to read
2012-09-08T16:16:52 < dongs> or is taht jtag only
2012-09-08T16:18:08 < Laurenceb_> this looks sane now?
2012-09-08T16:18:16 < Laurenceb_> so existing stuff to write
2012-09-08T16:18:30 < Laurenceb_> as i dont mind a little block whilst it sends data
2012-09-08T16:18:50 < jpa-> well if you are going to modify the shell source code, it is no problem anyway
2012-09-08T16:18:53 < Laurenceb_> then i create a new BaseChannel that points to readt to read data with timeout
2012-09-08T16:19:01 < Laurenceb_> im not using shell
2012-09-08T16:19:07 < Laurenceb_> im just using it as an example
2012-09-08T16:19:17 < Laurenceb_> im creating my own shell to interface with some kit
2012-09-08T16:19:18 < jpa-> ah, well then just use the timeout function directly?
2012-09-08T16:20:08 < Laurenceb_> hmm i guess
2012-09-08T16:20:20 < Laurenceb_> the abstraction is potentially quite nice
2012-09-08T16:20:45 < Laurenceb_> i just wanted to avoid the necessary complication of a dedicated thread with io queues for the usb
2012-09-08T16:21:41 < Laurenceb_> theres a printf function as part of chibios isnt there?
2012-09-08T16:21:41 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has joined ##stm32
2012-09-08T16:22:21 < Laurenceb_> yeah
2012-09-08T16:22:22 < Laurenceb_> bbl
2012-09-08T16:24:48 < Laurenceb_> arg wtf
2012-09-08T16:24:56 < Laurenceb_> it talks to a basechannel
2012-09-08T16:25:00 < Laurenceb_> this is so confusing
2012-09-08T16:25:23 < Laurenceb_> maybe ill give up on chibios
2012-09-08T16:26:13 < Laurenceb_> https://github.com/mabl/ChibiOS/blob/master/testhal/STM32F4xx/USB_CDC/main.c#L333
2012-09-08T16:26:14 <+Steffanx> No you wont
2012-09-08T16:26:19 < Laurenceb_> ok now im really confused
2012-09-08T16:26:34 < Laurenceb_> chp is a BaseSequentialStream
2012-09-08T16:26:39 <+Steffanx> [15:22:23]  <Laurenceb_> bbl
2012-09-08T16:26:39 <+Steffanx> [15:24:48]  <Laurenceb_> arg wtf
2012-09-08T16:26:41 <+Steffanx> LOL :P
2012-09-08T16:26:43 < Laurenceb_> that talks to usb
2012-09-08T16:27:09 < Laurenceb_> chprintf talks to BaseChannels
2012-09-08T16:27:33 < Laurenceb_> which allow timeout
2012-09-08T16:28:39 < Laurenceb_> maybe its possible to typecast between the two?
2012-09-08T16:29:53 < Laurenceb_> chprintf uses chIOPut(ip, b)   ((ip)->vmt->put(ip, b, TIME_INFINITE))
2012-09-08T16:30:01 < Laurenceb_> so its taking a time argument
2012-09-08T16:30:07 < jpa-> of course you can cast between them
2012-09-08T16:30:16 < jpa-> basechannel is a subclass of basesequentialstream
2012-09-08T16:30:34 < Laurenceb_> but basesequentialstream doesnt allow timeout?
2012-09-08T16:30:41 < jpa-> yep
2012-09-08T16:30:57 < jpa-> because it's planned for files & such things
2012-09-08T16:31:02 < dongs> comeon freetards :(
2012-09-08T16:31:11 < dongs> does OpenoCD support SWD on ANY device or has to be specific device
2012-09-08T16:31:25 < Laurenceb_> so how the hell can chprintf call chIOput?
2012-09-08T16:31:33 < Laurenceb_> with a timeout argument
2012-09-08T16:31:55 < jpa-> Laurenceb_: it takes a basechannel, not a basesequentialstream
2012-09-08T16:32:11 < Laurenceb_> EXACTLY
2012-09-08T16:32:19 < Laurenceb_> https://github.com/mabl/ChibiOS/blob/master/testhal/STM32F4xx/USB_CDC/main.c#L333
2012-09-08T16:32:22 < Laurenceb_> ^WTF
2012-09-08T16:32:36 < jpa-> it has changed in the newest dev branch
2012-09-08T16:32:39 < Laurenceb_> (BaseSequentialStream *chp
2012-09-08T16:33:01 < Laurenceb_> that code is running on my F4
2012-09-08T16:33:19 < jpa-> then your os/various/chprintf.c should be using basesequentialstream also
2012-09-08T16:33:47 < Laurenceb_> oh
2012-09-08T16:33:50 < Laurenceb_> fair enough
2012-09-08T16:33:55 < Laurenceb_> anyway, printf works
2012-09-08T16:34:07 < Laurenceb_> with usb, so thats good so far
2012-09-08T16:34:22 < Laurenceb_> i can impliment a basic scanf later
2012-09-08T16:35:11 < Laurenceb_> thanx for the help
2012-09-08T16:35:12 < Laurenceb_> bbl
2012-09-08T16:36:27 < jpa-> scanf is way oldschool
2012-09-08T16:36:35 < dongs> ..scanf
2012-09-08T16:36:38 < dongs> who the fuck uses that
2012-09-08T16:36:42 < dongs> maybe same dudes who use chibios
2012-09-08T16:36:44 < jpa-> just use strtok_r and atoi, it works for any relevant case
2012-09-08T16:36:59 < jpa-> (or strtol)
2012-09-08T16:37:06 < dongs> strlol more like
2012-09-08T16:37:11 -!- BjoernC [~BjoernC@dyndsl-178-142-052-087.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-08T16:59:11 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-08T17:14:19 < dongs> http://tmrepository.com/fudtracker/five-things-desktop-linux-has-to-do-but-wont-to-be/ haha
2012-09-08T17:18:42 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-08T17:21:42 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 264 seconds]
2012-09-08T17:40:27 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-08T17:40:28 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-08T17:40:28 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-08T17:40:31 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-08T18:09:03 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-09-08T18:16:49 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-08T18:37:57 < zyp> fucking synopsys bullshit
2012-09-08T18:41:33 <+Steffanx> Yes yes
2012-09-08T18:42:28 < zyp> shit has locked up again, won't enumerate even though I'm loading known working code
2012-09-08T18:44:03 <+Steffanx> I think it's funny even a power cycle doesn't solve your issue :P
2012-09-08T18:44:18 < zyp> yeah
2012-09-08T18:46:14 <+Steffanx> and weird
2012-09-08T18:46:46 < zyp> well, this is not the same issue, it's not spewing garbage now
2012-09-08T18:47:13 < zyp> it's just failing to respond to the SetAddress command, the first GetDescriptor to addr 0 is fine
2012-09-08T19:01:48 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-08T19:42:37 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 240 seconds]
2012-09-08T19:44:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-08T19:54:57 <+dekar> dongs, the stable API stuff is BS imo
2012-09-08T19:55:34 <+dekar> the kernel is still compatible with old 2.4 binaries, probably even older ones
2012-09-08T19:55:51 <+dekar> X11 is a well defined protocol, ALSA has OSS support
2012-09-08T19:58:26 < qyx_> hm, is there any possibility to use dma for transferring gpio to mem?
2012-09-08T19:58:46 < qyx_> i would like to hook external high speed parallel adc to gpio or fsmc
2012-09-08T19:59:14 < qyx_> to grab short waveform at > 20msps
2012-09-08T19:59:40 -!- barthess [~barthess@178.154.40.8] has quit [Quit: Leaving.]
2012-09-08T20:00:15 <+dekar> using the fsmc it should work, probably using gpio as well
2012-09-08T20:00:25 < qyx_> there is no gpio trigger for dma, which seems ok to me..
2012-09-08T20:00:38 < qyx_> but
2012-09-08T20:01:09 < qyx_> i don't know how to set the frequency of triggering
2012-09-08T20:01:58 < qyx_> maybe if i could trigger at external int connected to adc's data ready output
2012-09-08T20:04:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-08T20:06:40 -!- tavish_ [~tavish@120.56.233.183] has joined ##stm32
2012-09-08T20:08:07 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-09-08T20:14:46 < qyx_> ah, i just found AN2548 with examples of doing that
2012-09-08T20:18:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-08T20:23:00 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-08T20:28:04 -!- tavish [~tavish@59.177.4.207] has joined ##stm32
2012-09-08T20:28:04 -!- tavish [~tavish@59.177.4.207] has quit [Changing host]
2012-09-08T20:28:04 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-08T20:30:38 -!- tavish_ [~tavish@120.56.233.183] has quit [Ping timeout: 255 seconds]
2012-09-08T20:46:33 -!- tavish_ [~tavish@120.56.137.193] has joined ##stm32
2012-09-08T20:47:23 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-09-08T20:53:34 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-08T20:54:05 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-08T20:54:09 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-08T21:17:57 < prog> qyx_: doing SDR ?
2012-09-08T21:20:16 < zyp> qyx_, you just trigger DMA of a timer and have it read from GPIO
2012-09-08T21:29:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-08T21:30:28 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-08T21:30:32 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-08T21:38:22 < qyx_> prog: no :)
2012-09-08T21:38:43 < qyx_> zyp: yep, thanks, they use it that way in the AN
2012-09-08T21:39:53 < zyp> dma request source and source/destination addrs are independent, so one can do a lot of weird stuff with dma :p
2012-09-08T21:46:01 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-09-08T21:46:53 -!- tech2077 [~tech2077@adsl-75-53-141-174.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-08T21:50:58 -!- barthess [~barthess@164.177.203.106] has joined ##stm32
2012-09-08T21:54:24 < Laurenceb_> you could trigger off a clk source that way
2012-09-08T21:54:29 < Laurenceb_> using timer input
2012-09-08T21:55:52 < jpa-> i wonder if Laurenceb_ could have DMA reconfiguring his timers :)
2012-09-08T21:56:20 < qyx_> stacked timers reconfiguring itselfs?
2012-09-08T21:56:20 < Laurenceb_> id need a dedicated timer i think
2012-09-08T21:56:33 < Laurenceb_> i used "stacked" timers gating one another
2012-09-08T21:56:52 < Laurenceb_> that way i can do as many pwm channels as i have timers
2012-09-08T21:57:44 < qyx_> ha, you can do pwm using dma!
2012-09-08T21:57:50 < qyx_> from a circular ram buffer as a source
2012-09-08T21:58:01 < Laurenceb_> rofl
2012-09-08T21:58:06 < Laurenceb_> i didnt think of that one
2012-09-08T21:58:52 < Laurenceb_> thatd take a ton of ram
2012-09-08T21:58:57 < qyx_> why
2012-09-08T21:59:03 < jpa-> not if you use a PWM timer to trigger it! ,)
2012-09-08T21:59:06 < qyx_> 256 words for 8bit pwm
2012-09-08T21:59:22 < qyx_> so 512B for 16 8bit pwm channels
2012-09-08T21:59:24 < jpa-> or if you can use the whole port, only 256 halfwords
2012-09-08T21:59:33 < jpa-> yeah
2012-09-08T22:00:11 < karlp> dongs, what hardware of yours is being mixed up with teensy?
2012-09-08T22:00:37 < Laurenceb_> i have 14bit pwm atm
2012-09-08T22:00:44 < karlp> never mind, I'll keep reading scrollback properly
2012-09-08T22:00:45 < Laurenceb_> and 4 channels
2012-09-08T22:00:55 < qyx_> 8K only
2012-09-08T22:01:00 < Laurenceb_> at 11khz in a sequence that repeats at 62hz
2012-09-08T22:01:06 -!- tech2077 [~tech2077@adsl-75-53-133-53.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-08T22:01:07 < Laurenceb_> thats a lot of data to do with dma
2012-09-08T22:01:13 < qyx_> ah
2012-09-08T22:01:43 < Laurenceb_> it works atm so *shrug*
2012-09-08T22:01:51 < jpa-> heck it works
2012-09-08T22:01:53 < Laurenceb_> also dma isnt completely gitter free
2012-09-08T22:01:54 < jpa-> i don't believe it
2012-09-08T22:02:01 < Laurenceb_> what?
2012-09-08T22:02:07 < Laurenceb_> my gated timers?
2012-09-08T22:02:10 < jpa-> yeah
2012-09-08T22:02:38 < Laurenceb_> got it working perfectly after a few days faffing with MSM bits
2012-09-08T22:02:47 < jpa-> "few days" :)
2012-09-08T22:02:59 < Laurenceb_> code is on github
2012-09-08T22:03:04 < jpa-> AFAIK you have been failing with timers for the past 8 months or so
2012-09-08T22:03:19 < jpa-> no way it's working now :)
2012-09-08T22:03:50 < Laurenceb_> ive tried a ton of techniques
2012-09-08T22:04:02 < Laurenceb_> this gated code is the best so far
2012-09-08T22:05:40 < Laurenceb_> its shot noise limited now
2012-09-08T22:06:23 < jpa-> still three leds or have you got more?
2012-09-08T22:07:14 < Laurenceb_> three
2012-09-08T22:07:21 < Laurenceb_> making a 5 led version atm
2012-09-08T22:07:34 < Laurenceb_> waiting for the LEDs to ship from the fab
2012-09-08T22:07:44 < Laurenceb_> getting custom designs fabbed in Austria
2012-09-08T22:08:39 < qyx_> what is it doing exactly?
2012-09-08T22:08:58 < Laurenceb_> tissue spectrometer
2012-09-08T22:09:02 < qyx_> besides they gate themselves to each other
2012-09-08T22:09:03 < qyx_> ah
2012-09-08T22:09:04 < qyx_> ok
2012-09-08T22:10:56 -!- tech2077 [~tech2077@adsl-75-53-133-53.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-08T22:19:28 < Laurenceb_> custom LEDs are surprisingly cheap
2012-09-08T22:20:57 < qyx_> what specs and price?
2012-09-08T22:21:18 -!- zippe [~Adium@173.11.99.161] has quit [Quit: Leaving.]
2012-09-08T22:23:26 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-09-08T22:24:03 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-08T22:24:35 < Laurenceb_> bare die infrared stuff
2012-09-08T22:24:38 -!- tech2077 [~tech2077@adsl-75-53-134-210.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-08T22:24:57 < Laurenceb_> ~40mw output, 1K @ around 2euros each
2012-09-08T22:25:31 < qyx_> hm, not much
2012-09-08T22:26:04 < Laurenceb_> GaAs substrates
2012-09-08T22:37:07 < Laurenceb_> jpa-: I can just use chnReadTimeout(ip, bp, n, time)
2012-09-08T22:37:24 < Laurenceb_> to do blocking read of usb with timeouyt
2012-09-08T22:46:40 -!- tech2077 [~tech2077@adsl-75-53-134-210.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-08T22:54:08 < Laurenceb_> does chibios mass storage actually work?
2012-09-08T22:54:18 < Laurenceb_> it looks like semi placeholder code to me
2012-09-08T22:56:26 < Laurenceb_> static uint8_t scsi_read_format_capacities(uint32_t *nblocks,
2012-09-08T22:56:26 < Laurenceb_>                                            uint32_t *secsize) {
2012-09-08T22:56:26 < Laurenceb_>   *nblocks = 1024;
2012-09-08T22:56:26 < Laurenceb_>   *secsize = 512;
2012-09-08T22:56:26 < Laurenceb_>   return 3; /* No Media.*/
2012-09-08T22:56:28 < Laurenceb_> ^lol
2012-09-08T22:58:49 -!- tech2077 [~tech2077@adsl-75-53-129-240.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-08T23:04:05 -!- barthess [~barthess@164.177.203.106] has quit [Quit: Leaving.]
2012-09-08T23:17:03 < jpa-> Laurenceb_: yes, that's what i said you to do :)
2012-09-08T23:25:41 < Laurenceb_> it isnt :P
2012-09-08T23:26:07 < Laurenceb_> good job i looked through the code in more detail myself
2012-09-08T23:29:07 -!- tavish_ [~tavish@120.56.137.193] has quit [Quit: leaving]
2012-09-08T23:35:11 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-08T23:35:12 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-08T23:35:12 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-08T23:35:15 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-08T23:45:01 < Laurenceb_> http://i.imgur.com/tpoOC.jpg
2012-09-08T23:46:03 < qyx_> baloon?
2012-09-08T23:49:18 < Laurenceb_> yes
2012-09-08T23:49:24 < Laurenceb_> from another balloon :P
2012-09-08T23:49:38 < BrainDamage> next time arm one of the payloads with an automated sentry gun
2012-09-08T23:49:48 < BrainDamage> and shoot down any competiting balloon
2012-09-08T23:50:02 < BrainDamage> so you can ensure you can maintain the altitude record
2012-09-08T23:55:10 -!- dekar_ [~dekar@212.255.31.74] has joined ##stm32
2012-09-08T23:55:14 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-08T23:57:42 -!- dekar [~dekar@212.255.26.121] has quit [Ping timeout: 268 seconds]
--- Day changed Sun Sep 09 2012
2012-09-09T00:30:35 -!- tech2077 [~tech2077@adsl-75-53-129-240.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-09T00:32:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-09T00:44:07 -!- tech2077 [~tech2077@76.247.249.32] has joined ##stm32
2012-09-09T01:50:05 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-09T01:57:25 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-09-09T01:59:05 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:03:45 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-09T02:08:30 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 264 seconds]
2012-09-09T02:09:40 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-09T02:10:54 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 268 seconds]
2012-09-09T02:12:32 < zyp> doh
2012-09-09T02:13:57 < zyp> I picked up the usb shit I was dicking around with earlier today, then realized it's failing because I moved the code from OTG_HS to OTG_FS, which has less FIFO RAM
2012-09-09T02:16:41 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:24:51 < zyp> no, it's still acting up, regardless of that.
2012-09-09T02:26:32 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 272 seconds]
2012-09-09T02:26:58 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:31:45 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-09T02:33:30 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-09-09T02:35:30 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 264 seconds]
2012-09-09T02:41:33 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:46:38 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-09-09T02:46:57 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:50:51 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 245 seconds]
2012-09-09T02:51:20 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-09-09T02:52:13 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T02:59:01 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-09-09T03:00:51 < dongs> karlp: yeah lol
2012-09-09T03:10:05 < dongs> ok so while im waiting for chinkdrivers to arrive
2012-09-09T03:10:11 < dongs> how do I go about controlling them
2012-09-09T03:10:22 < dongs> they do 1/0 by pwm duty cycle
2012-09-09T03:10:33 < dongs> wide/short = 1, short/wide = 0 or wahtever.
2012-09-09T03:10:41 < dongs> can I somehow use DMA+TIM to clock this shit out?
2012-09-09T03:11:01 < dongs> i.e. on each timer tick it would change duty based on some array of shit I precalculated
2012-09-09T03:30:38 < zyp> sounds like a waste of memory to precalc and use DMA, just use an ISR to set duty based on a single bit
2012-09-09T03:31:01 < dongs> zyp, really? memory is no problem, but wouldnt banging isr every xxus be worse?
2012-09-09T03:31:39 < dongs> also, teh timing i think is fairly strict
2012-09-09T03:31:43 < zyp> well, it obviously depends on your application
2012-09-09T03:31:58 < dongs> the low/high shit for 0 is 0.5us high, 2us low, with +-150ns
2012-09-09T03:32:11 < dongs> for 1, its 1.2us high, 1.3us low
2012-09-09T03:32:15 < zyp> the timing will still be perfect since you're using a timer with output compare to generate it
2012-09-09T03:33:07 < dongs> hmm period is 400khz
2012-09-09T03:33:09 < zyp> with preloading you just have to calculate the duty of the next period before the current period ends
2012-09-09T03:33:13 < dongs> right
2012-09-09T03:33:42 < zyp> and the isr would be pretty small, should be no problem
2012-09-09T03:34:42 < dongs> right, ok. i guess ill just do it that way then.
2012-09-09T03:34:50 < dongs> still, dma would not need cpu at all right?
2012-09-09T03:36:58 < dongs> anyhow, ill do it the lazy way first and then try the correct way.
2012-09-09T03:40:23 < zyp> dma would not need cpu while transferring, but would need memory for the whole precalced array
2012-09-09T03:42:49 < dongs> yeah but thats not a big deal?
2012-09-09T03:42:53 < dongs> i need to send in24 bit bursts
2012-09-09T03:42:59 < dongs> with 50us delay between
2012-09-09T03:43:24 < dongs> i could just precalc 24 bits worth of shit thatsonly gonna be like 48bytes
2012-09-09T03:43:56 < dongs> http://i.imgur.com/DepN1.png related to this channel
2012-09-09T03:48:20 < Erlkoenig> dongs: all that's impossible with the european standard plugs :-(
2012-09-09T03:49:16 < dongs> Erlkoenig: yeah, how terrible
2012-09-09T03:51:04 < Erlkoenig> also these plugs don't have a direction (about phase and "gnd") and are bulky... why do they even exist...
2012-09-09T03:51:29 < dongs> all the plugs except american ones suck
2012-09-09T03:51:40 < dongs> UK plugs are terrible
2012-09-09T03:51:59 < dongs> are you talking about 220V euro ones, that are just circular with 2 circular pins?
2012-09-09T03:52:27 < Erlkoenig> http://en.wikipedia.org/wiki/Schuko  yes, these... but they're 230V now ;)
2012-09-09T03:52:27 < dongs> http://upload.wikimedia.org/wikipedia/commons/8/8a/European_electrical_contour_plug_(not_earthed).JPG like this shit?
2012-09-09T03:52:42 < Erlkoenig> no thats the non-Earth version
2012-09-09T03:53:37 < BrainDamage> here plugs are https://en.wikipedia.org/wiki/File:Ita_plug_16A_10A.jpg
2012-09-09T03:53:49 < BrainDamage> apart that they are nonpolarized, they are not too bad
2012-09-09T03:53:59 < zyp> wtf
2012-09-09T03:54:08 < zyp> where do you have those?
2012-09-09T03:54:16 < dongs> haha 3 pin plugs
2012-09-09T03:54:22 < BrainDamage> in nearly every home?
2012-09-09T03:54:23 < dongs> well, pic says "ita"
2012-09-09T03:54:25 < dongs> im gonna guess italy?
2012-09-09T03:54:26 < BrainDamage> itay
2012-09-09T03:54:30 < Erlkoenig> http://de.wikipedia.org/w/index.php?title=Datei:AC_Power_Plug_Type_J.jpg&filetimestamp=20071111142936   these are the best.. earthed, small, and have a direction
2012-09-09T03:54:35 < zyp> you're not going to fit those into any outlet here :p
2012-09-09T03:54:35 < BrainDamage> italy, yes
2012-09-09T03:54:51 < BrainDamage> the europlug lacks a ground pin for a reason
2012-09-09T03:54:56 < qyx_> we have ireversible schuko here
2012-09-09T03:55:00 < qyx_> with earth pin
2012-09-09T03:55:10 < BrainDamage> because every european outlet put the ground pin where they wanted
2012-09-09T03:55:15 < qyx_> Erlkoenig: ^
2012-09-09T03:55:17 < zyp> also, why is having a direction an advantage?
2012-09-09T03:55:35 < qyx_> live conductor is always on the left here
2012-09-09T03:55:39 < dongs> it gets more AC-ey
2012-09-09T03:55:54 < zyp> seems to me like a huge disadvantage to always have to have chargers and shit in one direction
2012-09-09T03:56:06 < Erlkoenig> zyp: it might be useful in some cases, when connecting multiple AC-powered devices to eachother, like audio-stuff
2012-09-09T03:56:12 < zyp> «live conductor»
2012-09-09T03:56:14 < qyx_> no, chargers can be inserted also upside-down
2012-09-09T03:56:17 < qyx_> they are 2pin only
2012-09-09T03:56:22 < dongs> http://ecx.images-amazon.com/images/I/71szD24YuNL._SL1500_.jpg
2012-09-09T03:56:41 < Erlkoenig> because you might get less noise and levels-shift-problems when all devices hare the same GND
2012-09-09T03:57:14 < qyx_> zyp: what is bad with "live conductor"?
2012-09-09T03:57:38 < zyp> we don't have that concept here, both are «live»
2012-09-09T03:57:43 < BrainDamage> Erlkoenig: ground has nothing to do with polarity
2012-09-09T03:58:06 < BrainDamage> you'll get ground loops in both polarized and unpolarized plugs
2012-09-09T03:58:08 < Erlkoenig> erm, not ground as in earth, ground as in Neutral pin or however that's called
2012-09-09T03:58:25 < qyx_> the right one (neutral) is connected to protective earth in distribution box
2012-09-09T03:58:29 < BrainDamage> still applies, the ground loops are from the earthing pin
2012-09-09T03:58:49 < Erlkoenig> yes groundloops... but you might get a few less levelproblems or how they are called
2012-09-09T03:58:55 < zyp> qyx_, only in TN networks
2012-09-09T03:59:00 < BrainDamage> level problems?
2012-09-09T03:59:01 < zyp> most networks in norway are IT
2012-09-09T03:59:10 < zyp> see http://en.wikipedia.org/wiki/Earthing_system
2012-09-09T03:59:16 < qyx_> here are tn-s and tn-c
2012-09-09T04:00:12 < Erlkoenig> wah, uhm, like if you connect the GND's of the USB-plugs of 2 computers, you might get a circuit, because computer A's GND might be connected to the neutral pin while the other computer's is connected to the phase pin
2012-09-09T04:00:17 < dongs> japs dont have earthed or polarized plugs
2012-09-09T04:00:21 < dongs> even in new consturction
2012-09-09T04:00:23 < qyx_> hm, actually also tn-c-s
2012-09-09T04:00:25 < dongs> its absolutely fucking retarded.
2012-09-09T04:00:37 < qyx_> in new contructions tn-s is required
2012-09-09T04:00:46 < BrainDamage> Erlkoenig: no, pc power supply are isolated from live and neutral
2012-09-09T04:00:54 < Erlkoenig> BrainDamage: theoretically. :D
2012-09-09T04:00:59 < BrainDamage> even pratically
2012-09-09T04:01:14 < zyp> industry and some newer stuff here are using TN because of the increased power per conductor rating you get from running 400V
2012-09-09T04:01:23 < dongs> I was at a new construction just couple months ago and I needed sockets in some specific locations for teh stuff I needed to isntall there, and I told the guys "put grounded outlets here, and here and here blah' and they look at me blankly "why do you need grounded shit here"
2012-09-09T04:01:26 < zyp> but most of the country are using IT
2012-09-09T04:01:55 < Erlkoenig> dongs: wah, no Earth? isn't that very dangerous?
2012-09-09T04:01:55 < qyx_> here in slovakia is everything tn
2012-09-09T04:02:00 < BrainDamage> if they do, they cannot sell the device, you cannot have dc coupling
2012-09-09T04:02:13 < dongs> Erlkoenig: exactly
2012-09-09T04:03:23 < BrainDamage> here high power gets IT
2012-09-09T04:03:32 < zyp> also, if you just keep PE seperate from N in your equipment, you won't need to care whether it's connected one way or another
2012-09-09T04:03:45 < BrainDamage> I've always been wondering how the hell they can maintain well balanced load sides for the 3-phase
2012-09-09T04:04:04 < Erlkoenig> BrainDamage: one phase per room? :D
2012-09-09T04:04:05 < qyx_> hm, but you are right, 22kV, 110kV and 400kV are IT
2012-09-09T04:04:16 < dongs> Erlkoenig: like in america, all new installs use  3 wire conductors, one of htem being earth
2012-09-09T04:04:36 < Erlkoenig> yeah, as in germany
2012-09-09T04:04:37 < dongs> Erlkoenig: in jp, if you request earthed conenctor, they take a extra wire and tie it to some metal structure in teh building or someshit.
2012-09-09T04:04:42 < zyp> Erlkoenig, that's how it's done
2012-09-09T04:04:43 < dongs> lol.
2012-09-09T04:04:49 < Erlkoenig> dongs: wah...
2012-09-09T04:04:52 < zyp> Erlkoenig, as in one phase per circuit
2012-09-09T04:05:12 < dongs> when I built my house here I  wanted every plug earthed,
2012-09-09T04:05:12 < Erlkoenig> zyp: yes, i know
2012-09-09T04:05:22 < qyx_> funny such differences
2012-09-09T04:05:25 < dongs> but apaprently the contractors couldnt even BUY 3wiree conductors
2012-09-09T04:05:31 < qyx_> :D
2012-09-09T04:05:37 < zyp> distribution rails have connectors going 1 2 3 1 2 3 1 2 3, and then the fuses consume two and two points
2012-09-09T04:05:59 < Erlkoenig> dongs: so there are no "Residual-current devices" at all?
2012-09-09T04:06:02 < dongs> nope
2012-09-09T04:06:07 < dongs> they had no idea what that is at all
2012-09-09T04:06:13 < dongs> i wanted those in kitchen/bath etc
2012-09-09T04:06:16 < zyp> I used to work as an electrician some years ago
2012-09-09T04:06:22 < Erlkoenig> sounds like darwin-award-ly
2012-09-09T04:06:36 < zyp> also, japanese power installations looks fucking insane
2012-09-09T04:06:50 < Erlkoenig> my student's dorm has one of these devices per room...
2012-09-09T04:06:54 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-09T04:07:07 < zyp> some of the stuff I've seen in japan would be illegal as fuck in norway
2012-09-09T04:07:16 < qyx_> we have them mandatory
2012-09-09T04:07:26 < qyx_> in all new installations
2012-09-09T04:07:34 < zyp> all new installations in norway have one RCD per circuit
2012-09-09T04:07:34 < qyx_> and for lighting in bathrooms etc. also
2012-09-09T04:07:38 < Erlkoenig> zyp: having no earth is very much illegal in germany :D
2012-09-09T04:07:39 < qyx_> yep
2012-09-09T04:07:51 < BrainDamage> here rcd and earth are both mandatory
2012-09-09T04:08:21 < zyp> well, duh
2012-09-09T04:08:28 < zyp> no point in a RCD without earth :p
2012-09-09T04:08:29 < Erlkoenig> i'd just have them because the way they work is so cool. :D
2012-09-09T04:08:46 < dongs> america calls taht shit GFCI
2012-09-09T04:08:54 < dongs> i think i found one page in jap about that shit
2012-09-09T04:09:03 < dongs> electric contractors had no idea about it
2012-09-09T04:09:10 < qyx_> ground fault bla bla?
2012-09-09T04:09:10 < dongs> especially since in jap that shit sounded like some weird stuff
2012-09-09T04:09:12 < dongs> yeah
2012-09-09T04:09:39 < qyx_> i have seen that name only on PV installations
2012-09-09T04:09:43 < Erlkoenig> Fehlerstromschutzschalter, Fehlerstrom-Schutzeinrichtung, in german. bigger words ftw.
2012-09-09T04:09:47 < zyp> in norwegian, it's called a «ground fault switch»
2012-09-09T04:09:52 < BrainDamage> there's some countries that use earth wiring so thin with the excuse that there's rcd, that a case could go live to a deadly voltage without tripping the magnetothermic
2012-09-09T04:09:57 < zyp> «jordfeilbryter»
2012-09-09T04:10:04 < qyx_> :D
2012-09-09T04:10:11 < qyx_> "prúdový chránič"
2012-09-09T04:10:18 < zyp> hmm
2012-09-09T04:10:39 < BrainDamage> here's called "differential switch" "interruttore differenziale"
2012-09-09T04:10:49 < zyp> http://bin.jvnv.net/f/kWeIo.jpg <- found some nice pic of some shit I worked on :D
2012-09-09T04:11:00 < dongs> https://www.google.com/search?q=..........
2012-09-09T04:11:01 < dongs> err
2012-09-09T04:11:03 < dongs> lolw.
2012-09-09T04:11:16 < dongs> https://www.google.com/search?q=%E6%8E%A5%E5%9C%B0%E4%BA%8B%E6%95%85%E5%9B%9E%E8%B7%AF%E9%81%AE%E6%96%AD%E8%A3%85%E7%BD%AE
2012-09-09T04:11:16 < BrainDamage> zyp: that's the final result, right?
2012-09-09T04:11:20 < dongs> ^ jap version
2012-09-09T04:11:20 < zyp> of course
2012-09-09T04:11:41 < qyx_> do not have suitable font :S
2012-09-09T04:11:43 < Erlkoenig> 回 <- what a pretty letter
2012-09-09T04:11:56 < Erlkoenig> 接  i like that one too
2012-09-09T04:12:17 < qyx_> zyp: that is yuite ok
2012-09-09T04:12:24 < qyx_> i hate the look of american distribution boxes
2012-09-09T04:12:41 < dongs> < qyx_> do not have suitable font :S
2012-09-09T04:12:44 < dongs> are you on lunix or somethign
2012-09-09T04:12:50 < qyx_> :S
2012-09-09T04:12:54 < dongs> windows has CJK fonts in default install since 2001
2012-09-09T04:13:08 < Erlkoenig> dongs: my linux rendered it correctly...
2012-09-09T04:13:14 < dongs> just sayin'.
2012-09-09T04:13:15 < qyx_> debian base install
2012-09-09T04:13:26 < Erlkoenig> (03:12:24) qyx_: i hate the look of american distribution boxes <- these ones with cm-thick copper rails are nice :D
2012-09-09T04:13:30 < BrainDamage> qyx_: try déjà vu fonts
2012-09-09T04:13:45 < qyx_> i use them
2012-09-09T04:14:36 < BrainDamage> american house supply capability is ridicolous
2012-09-09T04:14:48 < zyp> I once worked in a distribution with cm-thick and 10cm wide rails
2012-09-09T04:14:52 < qyx_> dejavu sans do not have that glyphs
2012-09-09T04:14:53 < Erlkoenig> amerian way to build houses id ridiculous. :D
2012-09-09T04:14:55 < BrainDamage> I think houses are wired for excess of 100A
2012-09-09T04:15:18 < qyx_> that wires in wooden walls
2012-09-09T04:15:20 < zyp> BrainDamage, but stop and think about it, how many watts do they have available?
2012-09-09T04:15:22 < qyx_> without proper insulation
2012-09-09T04:15:25 < qyx_> and conduits
2012-09-09T04:15:46 < BrainDamage> zyp: 120V*100A= 12kW
2012-09-09T04:15:51 < qyx_> and these terribly looking wall switches
2012-09-09T04:15:57 < Erlkoenig> BrainDamage: that's not so much...
2012-09-09T04:16:07 < qyx_> and usually only 1 phase
2012-09-09T04:16:09 < dongs> lol
2012-09-09T04:16:10 < zyp> BrainDamage, then consider that a modern norwegian home got 25 kW available
2012-09-09T04:16:15 < dongs> I had to pay extra to get > 80A to my house
2012-09-09T04:16:22 < dongs> apparently thats like too much for jp
2012-09-09T04:16:23 < zyp> (63A*230V*√3)
2012-09-09T04:16:25 < BrainDamage> I said excess, I don't know the accurate figure
2012-09-09T04:16:27 < dongs> @ 100V
2012-09-09T04:16:33 < qyx_> same here
2012-09-09T04:16:37 < qyx_> 63A@230
2012-09-09T04:16:46 < dongs> american friends talk about having like 200A breakers
2012-09-09T04:16:50 < BrainDamage> I heard some people talk about 500A for new home, but that sounded too ridicolous
2012-09-09T04:17:03 < zyp> qyx_, and three phase, so *√3
2012-09-09T04:17:11 < qyx_> yes
2012-09-09T04:17:38 < qyx_> standard flat here has 32A
2012-09-09T04:17:40 < BrainDamage> you'd have a large bolt in place of a fuse
2012-09-09T04:17:46 < Erlkoenig> btw, wouldn't some sort of centralized low-voltage-DC-supply make sense? Supplying like 5V with one big PSU for one house/floor, for more energy and space efficiency?
2012-09-09T04:17:57 < BrainDamage> not with fixed voltage
2012-09-09T04:18:03 < zyp> I once lived in a house with 35A one-phase
2012-09-09T04:18:07 < BrainDamage> it'd be better if it could supply different ones
2012-09-09T04:18:08 < zyp> 8 kW total
2012-09-09T04:18:16 < BrainDamage> and each device negotiate
2012-09-09T04:18:21 < Erlkoenig> BrainDamage: one might add 12V or something
2012-09-09T04:18:34 < BrainDamage> I want negotiabile up to 25V
2012-09-09T04:18:35 < Erlkoenig> and use USB-compatible plugs, so one can charge the shitload of usb-charged devices
2012-09-09T04:18:36 < qyx_> 5V is too low considering wire resistance
2012-09-09T04:18:37 < BrainDamage> 25V 5A
2012-09-09T04:18:39 < zyp> it was also badly insulated, so when the winter arrived, we blew the main fuses a bunch of times
2012-09-09T04:18:48 < BrainDamage> so you can charge a laptop
2012-09-09T04:19:05 < Erlkoenig> BrainDamage: oh good idea
2012-09-09T04:19:21 < Erlkoenig> one could as well then use these plugs in public places, like trains or airports
2012-09-09T04:19:32 < qyx_> it would not be economical as you have to use higher current rated wires
2012-09-09T04:19:40 < zyp> Erlkoenig, consider that even the new high-power usb will use higher voltages than 5V
2012-09-09T04:20:08 < BrainDamage> the problem of usb connectors, is that they are rated iirc for max 2-3A, I forgot the exact number
2012-09-09T04:20:21 < zyp> and that's fine for usb devices
2012-09-09T04:20:26 < BrainDamage> the connection isn't that great either
2012-09-09T04:20:27 < Erlkoenig> BrainDamage: that should be enough for most devices
2012-09-09T04:20:40 < BrainDamage> they said the same for 0.5A 5V
2012-09-09T04:20:49 < dongs> < zyp> some of the stuff I've seen in japan would be illegal as fuck in norway
2012-09-09T04:20:54 < Erlkoenig> what uses more then 0.5A? only USB-HDD's, right?
2012-09-09T04:20:56 < dongs> zyp, i think illegal anywehre in teh world
2012-09-09T04:21:07 < zyp> anyway, what problem are you trying to solve by having low voltage outlets?
2012-09-09T04:21:11 < BrainDamage> nearly the totality of smarphones
2012-09-09T04:21:13 < dongs> like i said, i could understand MAYBE the shit in 50 years old houses
2012-09-09T04:21:17 < dongs> but NEW construction!?
2012-09-09T04:21:30 < Erlkoenig> BrainDamage: really? hm
2012-09-09T04:21:40 < BrainDamage> also all the tablets
2012-09-09T04:21:51 < qyx_> shit in 50y old houses is done using Al wires :)
2012-09-09T04:22:23 < BrainDamage> some of the tables use so much current that won't even charge if the screen is on, and the usb plug provides "only" 1A
2012-09-09T04:22:33 < BrainDamage> tablets*
2012-09-09T04:22:33 < zyp> al wires are used a lot everywhere here for thick cables
2012-09-09T04:22:46 < BrainDamage> that's because Al is cheaper
2012-09-09T04:22:50 < zyp> because al is about half the price of cu for a given rating
2012-09-09T04:22:58 < qyx_> for thick yes, but not standard home installation
2012-09-09T04:23:01 < qyx_> it is forbidden now
2012-09-09T04:23:05 < zyp> even though you need more of it :p
2012-09-09T04:23:40 < qyx_> we were buying 4x16mm^2 Cu, 12e/m
2012-09-09T04:23:44 < zyp> qyx_, yeah, but it's al into the distribution
2012-09-09T04:23:51 < qyx_> Al 4x25 was 2.5e/m
2012-09-09T04:23:57 < zyp> :)
2012-09-09T04:26:11 < BrainDamage> Erlkoenig: also, it's been a self regulating problem, due to the standard fixing the max current to 0.5A, some devices were simply never made for it
2012-09-09T04:26:26 < BrainDamage> no point in making a device usb powered if the usb plug cannot supply it
2012-09-09T04:28:13 < Erlkoenig> and while we are inventing at superduper stuff, there should be a single protocol for every connector - something a bit like USB, which can be used to connect hdds, sdds, sound cards, graphics cards, etc.etc. using multiple kinds of cables/connectors, for different speed requiremtents (slow cables for mouse, fast cables for graphics cards, etc.). but the software would not care about the speed and just see devices. This would probab
2012-09-09T04:28:51 < BrainDamage> mmm, I don't see that going out very well
2012-09-09T04:29:00 < BrainDamage> also your sentence got truncated
2012-09-09T04:29:46 < Erlkoenig> oh sorry
2012-09-09T04:29:52 < Erlkoenig> using multiple kinds of cables/connectors, for different speed requiremtents (slow cables for mouse, fast cables for graphics cards, etc.). but the software would not care about the speed and just see devices. This would probably increase interoperability, require less drivers etc.
2012-09-09T04:30:01 < BrainDamage> also, usb doesn't have a single protocol, it has multiple
2012-09-09T04:30:12 < BrainDamage> and on top of those, there's a whole stack of drivers
2012-09-09T04:30:40 < Erlkoenig> yes but these are only software problems
2012-09-09T04:30:45 < BrainDamage> oh and usb sucks from anything besides an user point of view
2012-09-09T04:31:03 < Erlkoenig> yes i didn't say use exactly usb, something a bit like USB ;)
2012-09-09T04:31:44 < BrainDamage> http://xkcd.com/927/
2012-09-09T04:32:02 < qyx_> also poe ethernet outlets could be used that way
2012-09-09T04:32:02 < Erlkoenig> i saw that coming.
2012-09-09T04:32:07 < qyx_> there is protocol defined already
2012-09-09T04:32:11 < qyx_> for pwr negotioation
2012-09-09T04:32:27 -!- Huitaryan [~Huitaryan@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-09T04:32:54 < BrainDamage> the problem with poe is that lan cables are not made for high current, and often crammed in conduits
2012-09-09T04:33:35 < qyx_> but the structure of the installation gives you minimum of 0.5A@48V
2012-09-09T04:33:39 < BrainDamage> they could get away with the excuse they didn't carry significant power, but with poe, I espect some fires on people with stupid electricians
2012-09-09T04:34:30 < qyx_> that 20/22AWG wire can withstand ad least 2A imho without causing fire
2012-09-09T04:35:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-09T04:35:52 < zyp> I think usb is pretty great, but it's a bit burdened by spec violations and dumb decisions
2012-09-09T04:36:20 < Erlkoenig> especially about the price of VID's >.<
2012-09-09T04:37:05 < qyx_> 16bit will be enough
2012-09-09T04:37:07 < zyp> 03:29:52 < Erlkoenig> using multiple kinds of cables/connectors, for different speed requiremtents (slow cables for mouse, fast cables for graphics cards, etc.).
2012-09-09T04:37:10 < zyp> it's called usb3
2012-09-09T04:37:44 < Erlkoenig> yeah
2012-09-09T04:37:52 < Erlkoenig> but my eSATA HDD won't connect to it o.O
2012-09-09T04:38:03 < Erlkoenig> so, why didn't the develop USB 3.0 instead of SATA?
2012-09-09T04:38:06 < Erlkoenig> *they
2012-09-09T04:38:28 < BrainDamage> because usb is an asyncronous half duplex protocol
2012-09-09T04:38:33 < zyp> no
2012-09-09T04:38:37 < Erlkoenig> i could have plugged in both my new superfast HDD into it, and my old ones
2012-09-09T04:38:39 < zyp> usb3 is full duplex
2012-09-09T04:38:46 < BrainDamage> oh about time
2012-09-09T04:39:03 < BrainDamage> the undeterministic latency was a giant pain
2012-09-09T04:39:49 < qyx_> there are combined esata/usb3 connectors :X
2012-09-09T04:40:08 < Erlkoenig> qyx_: yes, and they're expensive, require 2 chips, 2 drivers etc.
2012-09-09T04:40:53 < zyp> Erlkoenig, well, both SATA and USB mass storage are transports for SCSI commands
2012-09-09T04:41:06 < zyp> SCSI layer doesn't care about speed, it just sees the devices
2012-09-09T04:41:13 < zyp> isn't that kind of what you wanted?
2012-09-09T04:41:18 < Erlkoenig> that's the next question... why do we carry around that old SCISI legacy stuff?
2012-09-09T04:41:28 < Erlkoenig> -I
2012-09-09T04:41:32 < zyp> why not?
2012-09-09T04:41:42 < zyp> why invent a new command set when that's only software?
2012-09-09T04:41:42 < BrainDamage> protocol is not the transport layer
2012-09-09T04:41:58 < BrainDamage> it's like asking why do we still use tcp when there's wifi
2012-09-09T04:42:09 < zyp> :)
2012-09-09T04:42:15 < Erlkoenig> uhm
2012-09-09T04:42:51 < Erlkoenig> well i might have suspected that such an old protocol might have some disadvantages... whatever, disregard that
2012-09-09T04:43:31 < BrainDamage> there's other stuff that got routed trough tcp
2012-09-09T04:43:34 < BrainDamage> err usb
2012-09-09T04:43:37 < BrainDamage> like midi messages
2012-09-09T04:43:39 < zyp> I haven't checked the specific, but I'm under the impression that the SCSI command set is a good command set to operate on generic block devices regardless of the transport it's used on
2012-09-09T04:43:55 < BrainDamage> actually there's even midi over tcp now that I think about it
2012-09-09T04:45:04 < zyp> I believe fibre channel is also based on SCSI commands
2012-09-09T04:45:45 < Erlkoenig> hm, and why does an application have to use entirely different API's to communicate with other CPU's (i.e., computers: like socket API) than to communicate with periphal devices (like libusb)
2012-09-09T04:46:30 < BrainDamage> overhead, cost
2012-09-09T04:46:39 < qyx_> as there exists iscsi and fcoe, i thing fc is not iscsi
2012-09-09T04:46:43 < qyx_> but maybe i am wrong
2012-09-09T04:46:54 < qyx_> *fc is not scsi
2012-09-09T04:47:10 < zyp> qyx_, FC is a lower layer
2012-09-09T04:47:19 < zyp> but when used for disks, SCSI is used on top
2012-09-09T04:47:34 < zyp> FCoE is ethernet -> FC -> SCSI
2012-09-09T04:47:39 < qyx_> "that predominantly transports SCSI commands over Fibre Channel"
2012-09-09T04:47:41 < qyx_> so it is
2012-09-09T04:47:49 < zyp> iSCSI is ethernet -> IP -> TCP -> SCSI
2012-09-09T04:47:56 < qyx_> yep
2012-09-09T04:48:12 < BrainDamage> until 10 years ago, most of the devices didn't bundle a micro despite them costing 1-2$ and being a tiny fraction of the system's cost
2012-09-09T04:48:22 < BrainDamage> like washing machines, etc, used mechanical timers
2012-09-09T04:49:14 < BrainDamage> probably within 10 years most of the devices will talk trough tcp in a lan
2012-09-09T04:49:25 < zyp> it's also possible to run IP over FC, so you can probably run iSCSI on IP over FC :p
2012-09-09T04:49:39 < qyx_> might be useful
2012-09-09T04:49:48 < qyx_> in.. some circumstances
2012-09-09T04:49:49 < qyx_> probably
2012-09-09T04:49:57 < zyp> probably not :p
2012-09-09T04:50:40 < zyp> anyway, have you heard about HSIC?
2012-09-09T04:51:00 < zyp> it's a different physical layer for USB
2012-09-09T04:51:08 < zyp> suited for short (<10 cm) connections
2012-09-09T04:51:24 < zyp> for IC to IC links
2012-09-09T04:51:57 < qyx_> different than differential signalling?
2012-09-09T04:52:28 < BrainDamage> zyp: I hope it's full duplex
2012-09-09T04:52:37 < zyp> no, it's not
2012-09-09T04:52:47 < BrainDamage> :s
2012-09-09T04:52:54 < BrainDamage> like we need another i2c ...
2012-09-09T04:53:12 < zyp> the point of HSIC is to avoid the expensive PHY
2012-09-09T04:53:51 < Erlkoenig> it has the advantage, that the CPU just sees "oh there's an USB cardreader" and just needs to load the usb-cardreader driver, and not some motherboard-vendor-specific driver for an i2c cardreader
2012-09-09T04:53:52 < zyp> and leveraging the already existing higher layer USB framework
2012-09-09T04:54:04 < zyp> Erlkoenig, exactly
2012-09-09T04:54:19 < Erlkoenig> also, there are probably more USB cardreaders than I2C cardreaders
2012-09-09T04:56:55 < BrainDamage> i2c is not really for cpu
2012-09-09T04:57:01 < BrainDamage> it's for embedded
2012-09-09T04:57:13 < zyp> so is HSIC
2012-09-09T04:58:06 < Erlkoenig> yeah, so I2C is an entirely different thing
2012-09-09T04:58:19 < Erlkoenig> and HSIC seems to be a very good idea for e.g. notebooks
2012-09-09T05:01:14 < zyp> it's intended for stuff where bandwidth is needed
2012-09-09T05:02:35 < BrainDamage> and no deterministic latency basically
2012-09-09T05:02:46 < BrainDamage> and don't talk about isochronous mode :p
2012-09-09T05:03:04 < zyp> what do you need deterministic latency for?
2012-09-09T05:04:17 < BrainDamage> for nearly every data acquisition system, unless your samples are timestamped at source, you want deterministic latency
2012-09-09T05:04:19 -!- tech2077 [~tech2077@76.247.249.32] has quit [Ping timeout: 246 seconds]
2012-09-09T05:04:45 < BrainDamage> like say, an high speed adc
2012-09-09T05:05:23  * Erlkoenig is looking around is desk whether is computer is connected to a high speed adc
2012-09-09T05:05:30 < Erlkoenig> *his
2012-09-09T05:05:37 < BrainDamage> we were talking about embedded
2012-09-09T05:05:47 < BrainDamage> since hsic was meant for embedded
2012-09-09T05:06:52 < Erlkoenig> probably more for normal PC systems
2012-09-09T05:07:20 < zyp> no, it's used in a lot of today's smartphones
2012-09-09T05:07:34 < zyp> between cpu and radio chipset
2012-09-09T05:18:26 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-09T05:23:43 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-09-09T05:34:21 < dongs> http://img02.taobaocdn.com/imgextra/i2/50739708/T2JyWHXbFbXXXXXXXX_!!50739708.jpg
2012-09-09T05:34:34 < dongs> http://img04.taobaocdn.com/imgextra/i4/50739708/T2hUGKXX8XXXXXXXXX_!!50739708.jpg
2012-09-09T05:38:06 < dongs> eh, apparently LPC's external memory controller shiz can be configured to use SDRAM
2012-09-09T05:38:16 < zyp> old news
2012-09-09T05:38:20 < dongs> news2me
2012-09-09T05:38:26 < dongs> i dont use inferior hardware
2012-09-09T05:38:40 < Erlkoenig> hmm the music i hear indicates that i'm not the only one awake in my dorm at 04:38
2012-09-09T05:38:43 < Erlkoenig> (am)
2012-09-09T05:39:01 < zyp> everybody goes «boo hoo, stm32's FSMC can't use SDRAM, LPC is better because it can!»
2012-09-09T05:39:11 < dongs> heh
2012-09-09T05:39:16 < dongs> o'Rly
2012-09-09T05:39:19 < dongs> so did tehy rectify that iN F3
2012-09-09T05:39:22 < dongs> or still no sdram
2012-09-09T05:39:32 < zyp> probably not
2012-09-09T05:42:14 < dongs> so why are you awake
2012-09-09T05:42:40 < zyp> no idea, I should have been in bed hours ago
2012-09-09T05:42:47 < dongs> http://www.rcgroups.com/forums/showthread.php?t=1726998 oh god
2012-09-09T05:42:52 < dongs> new project using propeller
2012-09-09T05:42:58 < dongs> that guy must enjoy chopping his dick off
2012-09-09T05:43:09 < zyp> was dicking around with usb
2012-09-09T05:43:10 < dongs> I will be using the Parallax Propeller for the project, and initially I will be downloading and
2012-09-09T05:43:13 < dongs> modifying existing "Spin" & "PASM" code from the Parallax Object Exchange. Later I will
2012-09-09T05:43:16 < dongs> start writing my own code, and then convert it to C++.
2012-09-09T05:44:36 < zyp> http://bin.jvnv.net/f/PL6RR.png <- got some ACM shit going on
2012-09-09T05:45:09 < zyp> but my code for bulk out isn't fully working yet
2012-09-09T05:45:45 < dongs> funtimes.
2012-09-09T05:46:18 < dongs> http://propeller.wikispaces.com/Propeller+II
2012-09-09T06:05:34 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-09T06:17:44 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-8.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-09T06:24:47 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-09T06:24:53 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-09T07:12:52 < emeb_mac> heh - individual sigma/delta ADC & DAC on every pin? Go ahead - pull the other one.
2012-09-09T07:13:52 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-09T07:14:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-09T07:14:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-09T07:14:31 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-09T07:20:03 < mrnuke> hello
2012-09-09T07:28:52 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-09T07:29:07 < dongs> hi
2012-09-09T07:29:21 < emeb_mac> lo
2012-09-09T07:30:10 < mrnuke> I was wondering if you guys were able to builg arm-none-eabi-gcc from 4.7.1 sources
2012-09-09T07:30:26 < emeb_mac> why build from source?
2012-09-09T07:31:06 < mrnuke> I couldn't find it packaged for my distro
2012-09-09T07:31:18 < emeb_mac> err - shouldn't be distro dependent
2012-09-09T07:31:43 < dongs> https://launchpad.net/gcc-arm-embedded
2012-09-09T07:31:44 < dongs> you can thank me later.
2012-09-09T07:35:37 < mrnuke> dongs: and I was hoping I'd make my first impression as a geekish guy
2012-09-09T07:36:38 < dongs> wasting time is never making a good impression
2012-09-09T07:37:18 < mrnuke> when has autotools _not_ wasted time? :)
2012-09-09T07:37:29 < dongs> no idea. thats why I use windows.
2012-09-09T07:38:01 < mrnuke> ouch!
2012-09-09T07:40:01 < mrnuke> BTW, thanks for the link
2012-09-09T08:12:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-09T08:17:58 -!- GargantuaSauce [~quassel@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-09T08:23:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 272 seconds]
2012-09-09T08:27:37 -!- GargantuaSauce [~quassel@blk-252-42-151.eastlink.ca] has quit [Remote host closed the connection]
2012-09-09T08:37:08 -!- GargantuaSauce [~quassel@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-09T08:41:01 -!- tech2077 [~tech2077@adsl-76-247-249-32.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-09T09:16:12 < jpa-> mrnuke: https://github.com/EliasOenal/TNT  is quite fine if you like building stuff
2012-09-09T09:25:05 -!- claude is now known as Claude
2012-09-09T09:27:46 < zippe> jpa-: no M4 support
2012-09-09T09:27:50 < zippe> = fail
2012-09-09T09:32:25 -!- Claude is now known as claude
2012-09-09T10:10:40 -!- claude is now known as Claude
2012-09-09T10:33:05 -!- GargantuaSauce [~quassel@blk-252-42-151.eastlink.ca] has quit [Quit: sauce]
2012-09-09T10:33:30 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-09T11:55:36 -!- mrcan_ [~mrcan@78.173.220.3] has joined ##stm32
2012-09-09T11:55:37 -!- mrcan_ [~mrcan@78.173.220.3] has quit [Changing host]
2012-09-09T11:55:37 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-09T11:58:46 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 245 seconds]
2012-09-09T12:02:54 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-09T12:04:46 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 272 seconds]
2012-09-09T12:05:44 < Tectu> can anybody explain me the tamper stuff? is it like when you place a switch on a case and connect it to that tamper pin, and somebody opens the case, it looses any memory content for security reasons?
2012-09-09T12:26:41 <+dekar_> jpa-, TNT should support armv7-m with fpu by default, have you tried it and did it fail?
2012-09-09T12:27:38 <+dekar_> zippe, that was actually meant for you :/
2012-09-09T12:28:47 <+dekar_> Tectu, I would imagine it is just another interrupt
2012-09-09T12:46:21 -!- barthess [~barthess@164.177.212.102] has joined ##stm32
2012-09-09T13:00:10 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-09T13:23:30 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T13:29:58 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-09T13:30:56 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-09T14:04:34 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-09T14:06:18 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-09T14:07:11 < dongs> sup trolls
2012-09-09T14:26:18 < Laurenceb_> top - 12:25:28 up 1503 days
2012-09-09T14:26:21 < Laurenceb_> muha
2012-09-09T14:26:28 < Laurenceb_> beat that with windoze
2012-09-09T14:27:24 < dongs> \\dtvcap has been up for: 1024 day(s), 5 hour(s), 49 minute(s), 14 second(s)
2012-09-09T14:27:29 < dongs> ^ Windows XP
2012-09-09T14:34:03 < BrainDamage> universe has been up for: 4.339 ± 0.035 ×10^17 seconds
2012-09-09T14:36:21 < BrainDamage> also, I'd be more concerned about you both running a shit ancient kernel
2012-09-09T14:37:27 < dongs> dont worry
2012-09-09T14:37:32 < dongs> windows has no bugs
2012-09-09T14:57:12 < dongs> gonna watch MiB this evening.
2012-09-09T14:57:19 < dongs> acutally more like right now
2012-09-09T15:47:17 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-09T15:47:18 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-09T15:47:18 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-09T15:47:21 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-09T16:00:54 -!- theAdib [~adib@dslb-088-074-142-141.pools.arcor-ip.net] has joined ##stm32
2012-09-09T16:02:30 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-194.pptp.stw-bonn.de] has joined ##stm32
2012-09-09T16:26:41 -!- ntfreak_ is now known as ntfreak
2012-09-09T17:29:16 < dongs> nice movie
2012-09-09T17:53:39 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-74-115.pptp.stw-bonn.de] has joined ##stm32
2012-09-09T17:53:46 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-194.pptp.stw-bonn.de] has quit [Ping timeout: 245 seconds]
2012-09-09T17:55:50 < Oldboy> when was MiB released ? 1990 ?
2012-09-09T17:59:01 < dongs> mib3 of course.
2012-09-09T19:32:07 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-09T19:45:56 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-74-115.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-09T19:46:12 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-115.pptp.stw-bonn.de] has joined ##stm32
2012-09-09T19:50:27 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-09-09T19:51:07 < TitanMKD> hi
2012-09-09T19:51:29 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-09-09T19:51:30 -!- DaKu is now known as daku
2012-09-09T19:53:20 -!- tavish [~tavish@120.59.8.113] has joined ##stm32
2012-09-09T19:53:20 -!- tavish [~tavish@120.59.8.113] has quit [Changing host]
2012-09-09T19:53:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-09T19:54:49 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-09T19:54:53 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-09T20:34:08 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-09-09T20:35:08 -!- tavish [~tavish@120.59.8.113] has joined ##stm32
2012-09-09T20:35:08 -!- tavish [~tavish@120.59.8.113] has quit [Changing host]
2012-09-09T20:35:09 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-09T20:36:04 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-09T20:37:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-09T20:37:15 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-09T20:37:43 < Laurenceb_> http://pastie.org/4691225
2012-09-09T20:37:49 < Laurenceb_> anyone got any idea why?
2012-09-09T20:38:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-09T20:53:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-09T20:54:09 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-09T20:54:18 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-09T21:02:19 < Laurenceb_> works on my pc with old installation of chibios
2012-09-09T21:05:18 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-09T21:05:38 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-09T21:07:06 < TitanMKD> Laurenceb_ which version of Chibios ?
2012-09-09T21:08:52 < Laurenceb_> oh its the mcuconf format
2012-09-09T21:09:04 < Laurenceb_> its changed - this issue is covered on the forum
2012-09-09T21:12:49 <+Steffanx> Surprise surprise!
2012-09-09T21:19:51 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-09T21:22:46 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-09T21:41:32 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-09T21:41:33 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-09T21:41:33 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-09T21:52:22 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-09-09T21:52:39 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-09T21:52:39 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-09T21:52:39 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-09T21:52:42 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-09T22:24:40 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-09T22:24:47 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-09T22:25:21 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-09T22:59:54 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-09T23:00:12 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-09T23:00:13 -!- tech2077 [~tech2077@adsl-76-247-249-32.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-09T23:12:31 -!- tech2077 [~tech2077@adsl-75-53-131-76.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-09T23:19:52 -!- theAdib [~adib@dslb-088-074-142-141.pools.arcor-ip.net] has quit [Quit: Ex-Chat]
2012-09-09T23:20:05 < Laurenceb_> wohoo
2012-09-09T23:20:11 < Laurenceb_> more of my crap passed CE
2012-09-09T23:20:38 <+izua> you're becoming an exporting chinese person?
2012-09-09T23:21:36 < Laurenceb_> lol
2012-09-09T23:30:27 < Laurenceb_> wtf
2012-09-09T23:30:36 < Laurenceb_> undefined reference to `main'  in my chibios project
2012-09-09T23:34:07 <+izua> did you forget to add the WinMain@16
2012-09-09T23:34:30 < Laurenceb_> wut
2012-09-09T23:35:13 <+izua> you shouldn't have main bro, you should use winMain
2012-09-09T23:35:27 <+izua> anyway.
2012-09-09T23:35:35 <+izua> how did the bug with the sd card go?
2012-09-09T23:35:43 <+izua> was simply overwriting into the stack the main issue?
2012-09-09T23:36:00 < Laurenceb_> yes
2012-09-09T23:36:22 < Laurenceb_> chibios uses int main(void)   in the examples
2012-09-09T23:36:26 < Laurenceb_> ill try compiling one
2012-09-09T23:36:36 <+izua> so.. no more random errors?
2012-09-09T23:36:46 <+izua> what about speed?
2012-09-09T23:37:34 < Laurenceb_> ~800KB/s
2012-09-09T23:37:51 < Laurenceb_> works fine now
2012-09-09T23:38:10 < Laurenceb_> chibios examples compile and run... odd
2012-09-09T23:38:37 <+izua> makefile?
2012-09-09T23:38:59 < Laurenceb_> possibly
2012-09-09T23:39:03 <+izua> not sure how chibios organizes stuff, but i bet they have main() and you have to include some headers
2012-09-09T23:39:26 < Laurenceb_> the error is in the startup code
2012-09-09T23:39:32 < Laurenceb_> startup code cant see main
2012-09-09T23:41:33 < Laurenceb_> hmm no difference
2012-09-09T23:45:06 -!- tech2077 [~tech2077@adsl-75-53-131-76.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-09-09T23:47:34 -!- Claude is now known as claude
2012-09-09T23:47:44 < Laurenceb_> init code has extern void main(void);
2012-09-09T23:48:02 < Laurenceb_> but if i change my code to void main(void) theres no difference
2012-09-09T23:48:54 < TitanMKD> you just need standard int main(void) {
2012-09-09T23:49:25 < zyp> Laurenceb_, pastebin a full compile log please
2012-09-09T23:49:51 < TitanMKD> you should also use the makefile for chibios because it requires lot of libs ...
2012-09-09T23:49:52 < Laurenceb_> k
2012-09-09T23:50:13 < Laurenceb_> http://pastebin.com/cZKA9PvX
2012-09-09T23:50:56 < Laurenceb_> main.c contains main
2012-09-09T23:51:16 < Laurenceb_> chibios demo projects compile ok
2012-09-09T23:51:40 < zyp> that non-verbose output is fucking useless
2012-09-09T23:51:41 < Laurenceb_> i just copied and pasted the config and makefile over
2012-09-09T23:51:46 < Laurenceb_> lol
2012-09-09T23:52:06 < Laurenceb_> well crt0.c cant see main?
2012-09-09T23:52:09 < zyp> «Linking build/ch.elf» doesn't fucking tell which files it's linking
2012-09-09T23:52:22 < zyp> is it really linking main.o?
2012-09-09T23:52:29 < Laurenceb_> oh yes
2012-09-09T23:52:34 < Laurenceb_> the output is called ch
2012-09-09T23:52:48 < Laurenceb_> thats in the makefile somewhere - ch - chibi
2012-09-09T23:53:01 < zyp> I'm not talking about the output
2012-09-09T23:53:11 < zyp> I'm asking whether it's actually linking all the inputs
2012-09-09T23:54:24 < Laurenceb_> ena bling c++ doesnt help
2012-09-09T23:54:33 < Laurenceb_> its the stock chibios makefile
2012-09-09T23:54:50 < Laurenceb_>  /ChibiOS/demos/ARMCM4-STM32F407-DISCOVERY/Makefile
2012-09-09T23:55:02 < zyp> I never used it
2012-09-09T23:56:40 -!- tech2077 [~tech2077@adsl-75-53-137-130.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-09T23:56:47 < Laurenceb_> https://github.com/mabl/ChibiOS/blob/master/demos/ARMCM4-STM32F407-DISCOVERY/Makefile
2012-09-09T23:56:49 < zyp> I don't enjoy dicking around with outdated tools so I just listed the sources I needed from chibios and pulled them into my own build script
2012-09-09T23:56:55 < Laurenceb_> ill try moving that example
2012-09-09T23:57:05 < Laurenceb_> heh
2012-09-09T23:57:08 < zyp> anyway
2012-09-09T23:57:16 < zyp> try building verbose
2012-09-09T23:57:24 < zyp> USE_VERBOSE_COMPILE make
2012-09-09T23:57:37 < zyp> or maybe it's make USE_VERBOSE_COMPILE=yes
2012-09-09T23:59:22 < Laurenceb_> http://pastebin.com/TP9pzHv5
2012-09-09T23:59:49 < zyp> see
2012-09-09T23:59:53 < zyp> there's your problem
2012-09-09T23:59:57 < zyp> it's not linking main.o
--- Day changed Mon Sep 10 2012
2012-09-10T00:00:01 < zyp> just like I expected
2012-09-10T00:00:24 < Laurenceb_> hmm how odd
2012-09-10T00:01:36 < zyp> pastebin your makefile?
2012-09-10T00:01:59 < TitanMKD> you should add your main file in CSRC
2012-09-10T00:02:11 < TitanMKD> CSRC = main.c
2012-09-10T00:02:45 < TitanMKD> for good code becarefull to change -O0 to -O2 too
2012-09-10T00:02:58 < Laurenceb_> aha
2012-09-10T00:03:03 < Laurenceb_> typo in my makefile
2012-09-10T00:03:11 < Laurenceb_> compile now thanks
2012-09-10T00:03:34 < Laurenceb_> stupid gedit highlighting
2012-09-10T00:03:42 < Laurenceb_> meant i missed it
2012-09-10T00:04:14 < Laurenceb_> must have accidentally hit a key :P
2012-09-10T00:04:16 < TitanMKD> anyone tested OpenOCD 0.6.0 with STM32F4 in SWD & JTAG mode ?
2012-09-10T00:04:33 < TitanMKD> now it is officialy supported
2012-09-10T00:04:54 < TitanMKD> also flashing STm32F4 internal flash shall work
2012-09-10T00:05:11 < zyp> I tested some git version with stlink once
2012-09-10T00:05:14 < zyp> worked ok
2012-09-10T00:05:21 < TitanMKD> ha great
2012-09-10T00:05:29 < TitanMKD> anyone tested it on LPC43xx to ?
2012-09-10T00:05:31 < zyp> except that the gdbserver still sucked
2012-09-10T00:05:48 < TitanMKD> ha what happens with gdbserver it crash sometimes ?
2012-09-10T00:06:22 < zyp> no, it was incompatible with my gdb
2012-09-10T00:06:32 < BrainDamage> they wanted a fancy gdb version
2012-09-10T00:06:41 < zyp> as in old?
2012-09-10T00:06:48 < BrainDamage> either old or new
2012-09-10T00:06:53 < BrainDamage> by new I mean patched
2012-09-10T00:07:03 < BrainDamage> just not current :/
2012-09-10T00:07:15 < zyp> well, I didn't care to dick around with that, so I just tested with the built in flasher
2012-09-10T00:07:32 < zyp> I was mostly interested in flashing performance
2012-09-10T00:09:59 < TitanMKD> zyp i use this one GNU gdb (GNU Tools for ARM Embedded Processors) 7.3.1.20120316-cvs
2012-09-10T00:10:48 < TitanMKD> but never tested it in fact ;) as i use the GDB from STMSTudio
2012-09-10T00:10:53 < zyp> FWIW I'm currently using 7.4.1, but I can't remember what I was using when I tested openocd
2012-09-10T00:11:20 < zyp> I don't really care, as long as it works with the BMP
2012-09-10T00:12:11 -!- barthess [~barthess@164.177.212.102] has quit [Quit: Leaving.]
2012-09-10T00:12:21 < TitanMKD> the toolchain from https://launchpad.net/gcc-arm-embedded is very stable
2012-09-10T00:12:28 < TitanMKD> and they have longterm support
2012-09-10T00:12:53 < TitanMKD> even tested it with really bigger ARM ;)
2012-09-10T00:13:01 < TitanMKD> the Rx
2012-09-10T00:16:39 < Laurenceb_> how do you force all c files to be compiled in a makefile?
2012-09-10T00:17:11 < Laurenceb_> its not *.c    :P
2012-09-10T00:17:58 < TitanMKD> it's done with dependencies and rules
2012-09-10T00:18:15 < Laurenceb_> yeah
2012-09-10T00:18:29 < TitanMKD> just add each file in CSRC = toto.c titi.c ...
2012-09-10T00:18:29 < Laurenceb_> CSRC = $(PORTSRC) \
2012-09-10T00:18:29 < Laurenceb_>        $(KERNSRC) \
2012-09-10T00:18:29 < Laurenceb_>        $(TESTSRC) \
2012-09-10T00:18:30 < Laurenceb_>        $(HALSRC) \
2012-09-10T00:18:30 < Laurenceb_>        $(PLATFORMSRC) \
2012-09-10T00:18:30 < Laurenceb_>        $(BOARDSRC) \
2012-09-10T00:18:32 < Laurenceb_>        $(CHIBIOS)/os/various/shell.c \
2012-09-10T00:18:34 < Laurenceb_>        $(CHIBIOS)/os/various/chprintf.c \
2012-09-10T00:18:36 < Laurenceb_>        main.c
2012-09-10T00:18:45 < Laurenceb_> how do i force anything else in the working directory to be added?
2012-09-10T00:19:42 < TitanMKD> you can also create a .mk file
2012-09-10T00:19:45 < TitanMKD> and include it
2012-09-10T00:19:54 < TitanMKD> else add each .c in CSRC
2012-09-10T00:20:43 < TitanMKD> CSRC = $(PORTSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(KERNSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(TESTSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(HALSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(PLATFORMSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(BOARDSRC) \
2012-09-10T00:20:44 < TitanMKD>        $(CHIBIOS)/os/various/devices_lib/accel/lis302dl.c \
2012-09-10T00:20:45 < TitanMKD>        $(CHIBIOS)/os/various/chprintf.c \
2012-09-10T00:20:45 < TitanMKD>        main.c \
2012-09-10T00:20:46 < TitanMKD>        otherfile1.c file2.c
2012-09-10T00:20:57 < TitanMKD> the last line shall not contain \
2012-09-10T00:22:00 < TitanMKD> else to add a directory with tons of files add a file.mk file in this directory (root)
2012-09-10T00:22:17 < TitanMKD> and add the reference in CHIBIOS
2012-09-10T00:22:22 < TitanMKD> example
2012-09-10T00:22:22 < TitanMKD> CHIBIOS = ../..
2012-09-10T00:22:22 < TitanMKD> include $(CHIBIOS)/boards/ST_STM32F4_DISCOVERY/board.mk
2012-09-10T00:22:22 < TitanMKD> include $(CHIBIOS)/os/hal/platforms/STM32F4xx/platform.mk
2012-09-10T00:22:22 < TitanMKD> include $(CHIBIOS)/os/hal/hal.mk
2012-09-10T00:22:22 < TitanMKD> include $(CHIBIOS)/os/ports/GCC/ARMCMx/STM32F4xx/port.mk
2012-09-10T00:22:22 < TitanMKD> include $(CHIBIOS)/os/kernel/kernel.mk
2012-09-10T00:22:23 < TitanMKD> include $(CHIBIOS)/test/test.mk
2012-09-10T00:22:23 < TitanMKD> include $(CHIBIOS)/demos/ARMCM4-STM32F407-DISCOVERY_TRF7970A/trf7970a/trf7970a.mk
2012-09-10T00:22:39 < TitanMKD> trf7970a.mk <= Like a library but it is just some files
2012-09-10T00:23:45 < TitanMKD> and after add in CSRC = the name of the var declared in .mk
2012-09-10T00:23:53 < TitanMKD> for me TRFSRC
2012-09-10T00:24:05 < TitanMKD> $(TRFSRC) \
2012-09-10T00:24:28 < TitanMKD> i prefer to do that than tons of .c files ;)
2012-09-10T00:25:25 < TitanMKD> does it answer your question ?
2012-09-10T00:28:04 < TitanMKD> bye
2012-09-10T00:28:11 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-09-10T00:28:14 < qyx_> weird
2012-09-10T00:41:18 < zyp> tons of boilerplate
2012-09-10T00:43:37 < BrainDamage> tons of copypasta and no pastebin
2012-09-10T00:46:11 <+Steffanx> I bet you like that BrainDamage .. copypasta
2012-09-10T00:46:22 <+Steffanx> A 100% Italian product
2012-09-10T00:46:45 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-10T00:46:59 < BrainDamage> really? given the amount of historical inventions and innovators here I'd have tought it'd be a foreign product
2012-09-10T00:47:24 < BrainDamage> altough the area around naples even fakes chinese products, it's kinda ridicolous ...
2012-09-10T00:47:39 <+izua> Laurenceb_: i do it like
2012-09-10T00:47:47 <+izua> obj/%.o: src/%.c
2012-09-10T00:47:48 <+izua> 	@echo " CC		$<
2012-09-10T00:47:51 <+Steffanx> Oh true .. that type of pasta could be chinese too
2012-09-10T00:47:58 <+izua> 	@$(CC) $(CFLAGS) -c $< -o $@
2012-09-10T00:48:00 <+izua> that too
2012-09-10T00:48:03 < zyp> http://paste.jvnv.net/view/bgON1 <- this about the amount of boilerplate I'm okay with in my build system
2012-09-10T00:48:06 <+izua> last line is pretty important
2012-09-10T00:48:27 <+Steffanx> Yeah, but scons is "fucking slow", zyp
2012-09-10T00:48:34 <+Steffanx> ( not my words )
2012-09-10T00:48:35 <+izua> zyp: uhm, neat?
2012-09-10T00:48:46 < BrainDamage> mine perhaps?
2012-09-10T00:48:52 <+Steffanx> Yes BrainDamage
2012-09-10T00:48:54 < zyp> izua, of course it's neat, that's why I'm using it
2012-09-10T00:49:07 < BrainDamage> Steffanx: try fetching the game "spring"
2012-09-10T00:49:16 < BrainDamage> an old version, say, 2 years ago
2012-09-10T00:49:22 < BrainDamage> fire scons, fire cmake
2012-09-10T00:49:24 < BrainDamage> compare the 2
2012-09-10T00:49:33 <+izua> who the hell names these things
2012-09-10T00:49:37 <+izua> there's another game called "love"
2012-09-10T00:49:39 <+izua> try googling it
2012-09-10T00:49:53 < BrainDamage> love is a game engine
2012-09-10T00:49:56 < BrainDamage> made with lua
2012-09-10T00:50:06 < BrainDamage> so I already know the keywords :p
2012-09-10T00:50:08 <+izua> results for "spring the game": nothing related
2012-09-10T00:50:14 < BrainDamage> try spring rts
2012-09-10T00:50:15 < zyp> Steffanx, well, I dunno
2012-09-10T00:50:22 <+izua> try googling it without the proper keyboards :P
2012-09-10T00:50:27 < zyp> http://paste.jvnv.net/view/25kcq <- I'm pretty happy with my compile time
2012-09-10T00:50:37 <+izua> it says lua engine
2012-09-10T00:50:54 < BrainDamage> anyway, scons will take 2-3 minutes to process the spring sources, cmake 2-3 seconds
2012-09-10T00:51:31 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-09-10T00:51:50 <+izua> but». but.. it will only do that once, right?
2012-09-10T00:51:50 < BrainDamage> izua: the engine is c++ with some asm, the lua engine is for mod scripting
2012-09-10T00:51:58 < BrainDamage> no, every git fetch
2012-09-10T00:51:58 <+izua> when you change the spring script..
2012-09-10T00:52:05 <+izua> and then it will generate a nice and awesome makefile
2012-09-10T00:52:10 < BrainDamage> which can mean 3-5 times a day
2012-09-10T00:52:28 <+izua> so yeah, you can use the old makefiles if file structure hasn't change
2012-09-10T00:52:31 <+izua> d
2012-09-10T00:52:43 < zyp> izua, scons doesn't generate makefiles
2012-09-10T00:53:18 < zyp> also, 2-3 minutes sounds unreasonable
2012-09-10T00:53:26 <+izua> that sounds bad
2012-09-10T00:53:29 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-10T00:53:42 <+izua> so i have to waste 2-3 minutes every compile?
2012-09-10T00:53:51 <+izua> or.. meh.. you're making me check this out
2012-09-10T00:54:05 < zyp> I'm going to test it, I don't believe it takes minutes :p
2012-09-10T00:54:41 < BrainDamage> it took minutes with my pentium 4 1.8GHz which I had back then :p
2012-09-10T00:54:58 <+Steffanx> LOL
2012-09-10T00:55:03 < zyp> downloaded, I'll test
2012-09-10T00:55:18 < Erlkoenig> i'm currently making a new build system specially suitable for crosscompiling and multi-compiler-support... i'll tell you when it's ready :D
2012-09-10T00:55:19 < BrainDamage> get an old version
2012-09-10T00:55:24 < zyp> I just realized
2012-09-10T00:55:33 < zyp> there is only cmake files in this one
2012-09-10T00:55:40 < zyp> can you suggest one?
2012-09-10T00:55:43 < BrainDamage> yes because devs got tired to wait for scons
2012-09-10T00:55:51 < BrainDamage> mmm sec
2012-09-10T00:56:30 < BrainDamage> try one between 74 and 80
2012-09-10T00:56:57 < BrainDamage> the version string got changed, but that chunk of version got retained
2012-09-10T00:58:20 < BrainDamage> 0.81.2.1 should have both
2012-09-10T00:59:07 < zyp> bah, it's complaining about missing deps
2012-09-10T00:59:17 <+Steffanx> heh
2012-09-10T00:59:32 < zyp> tired of this
2012-09-10T01:00:02 < zyp> I'm not going to install a ton of deps just so I can build something to prove it's not as slow as you say :p
2012-09-10T01:00:26 < BrainDamage> hah, proof by unreachable source
2012-09-10T01:00:33 < zyp> :)
2012-09-10T01:01:01 < BrainDamage> btw if you're in a debian derived machine
2012-09-10T01:01:11 < BrainDamage> sudo apt-get build-deps spring
2012-09-10T01:01:16 < zyp> FWIW I have mongodb 2.0.6 sources lying around, that I used for some benchmarking once
2012-09-10T01:01:40 < zyp> scons uses about five seconds to do a full dep-scan of the mongodb tree
2012-09-10T01:01:41 < BrainDamage> mmm I've been meaning to try mongodb, how's it?
2012-09-10T01:01:56 < zyp> I was benchmarking compile times. :p
2012-09-10T01:02:05 < BrainDamage> bah
2012-09-10T01:02:17 < zyp> was testing different levels of parallel compiling, and I just needed something of decent size
2012-09-10T01:02:36 < BrainDamage> try touching the file timestamps too, because git does that
2012-09-10T01:02:44 < BrainDamage> apparently that irritates many buildsystems
2012-09-10T01:03:02 < zyp> scons maintains a list of hashes of all files
2012-09-10T01:03:19 < BrainDamage> I know
2012-09-10T01:03:19 < zyp> if timestamp changed, it can still know that contents are unchanged
2012-09-10T01:03:29 < BrainDamage> but I don't know if it rehashes on timestamp chance or not
2012-09-10T01:03:43 < BrainDamage> change*
2012-09-10T01:04:22 < zyp> it's kind of neat, because it means that if I change a source file but the output is still the same, it will realize that and not perform a new linking
2012-09-10T01:04:50 < zyp> I actually spotted a bug that way yesterday
2012-09-10T01:04:50 <+izua> sounds awesome for those 10k liners
2012-09-10T01:05:11 < zyp> I made some change, compiled, then learned that the change had no effect
2012-09-10T01:05:24 < zyp> turned out the entire block got optimized out due to a typo
2012-09-10T01:05:45 <+Steffanx> Oops
2012-09-10T01:05:47 < zyp> i.e. an if block that could never be true
2012-09-10T01:06:37 <+izua> you mean, the object output was still the same, and scons figure that out?
2012-09-10T01:06:58 < zyp> yes, because it hashes the object file
2012-09-10T01:07:36 < zyp> equal hashes means equal files ;)
2012-09-10T01:07:37 <+izua> that does sound pretty awesome
2012-09-10T01:08:08 < zyp> either way, scons might not be the fastest build system around, but IMO it's one of the nicest to use
2012-09-10T01:08:27 < zyp> and with the size of my projects, the speed is irrelevant
2012-09-10T01:08:51 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-10T01:09:09 < zyp> as demonstrated, a full build of one of my stm32 projects take half a second, with scons processing and the actual compiling combined
2012-09-10T01:12:09 < zyp> it's no wonder make would be faster, as make do no dependency scanning on it's own, and last time I checked cmake was useless for cross compiling
2012-09-10T01:12:27 < zyp> so I'm not sure what would be a better choice :)
2012-09-10T01:12:30 <+izua> do you need special files for building stm32 stuff?
2012-09-10T01:12:40 < zyp> what do you mean?
2012-09-10T01:12:48 <+izua> the 'environment' thing
2012-09-10T01:13:01 < zyp> are you referring to the SelectMCU stuff?
2012-09-10T01:13:02 <+izua> env.SelectMCU('stm32f407zg')
2012-09-10T01:13:06 <+izua> yeah
2012-09-10T01:13:08 < zyp> I made those rules myself
2012-09-10T01:13:14 <+izua> that's obviously not something common
2012-09-10T01:13:14 <+izua> oh
2012-09-10T01:13:29 < zyp> http://cgit.jvnv.net/laks/tree/build_rules <- it's part of my library
2012-09-10T01:14:00 < BrainDamage> zyp: actually spring uses cmake for cross compiling, atough not embedded
2012-09-10T01:14:03 < zyp> it's still in it's infancy, but it works rather well so far
2012-09-10T01:14:13 < BrainDamage> and I'd probably not want to use cmake either for embedded
2012-09-10T01:14:40 < BrainDamage> most of it's facilities are useless, like lib scanning, etc
2012-09-10T01:14:42 <+izua> i'm definitely going to try scons
2012-09-10T01:14:54 < BrainDamage> scons makes sense for embedded tough, I agree
2012-09-10T01:15:09 <+izua> and since it's pure python, it's not going to be so picky about the os
2012-09-10T01:15:52 < zyp> izua, http://cgit.jvnv.net/suzumebachi/tree/SConstruct?id=2fc77d2 <- here's the SConstruct I used before I wrote the SelectMCU rules
2012-09-10T01:17:10 < zyp> something like that should work without any magic stuff
2012-09-10T01:17:25 <+Steffanx> That IS magic stuff :P
2012-09-10T01:18:04 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-10T01:18:06 < zyp> http://cgit.jvnv.net/suzumebachi/tree/SConstruct?id=422a778 <- when I used chibios I just pulled out the extra sources I needed like this
2012-09-10T01:19:48 <+izua> i always thought i'm the only one doing that
2012-09-10T01:22:18 < zyp> I hate the idea of «copy this library and put your sources into this directory», and all makefiles seems to be written with that structure in mind
2012-09-10T01:22:39 < Erlkoenig> zyp: what language is that?
2012-09-10T01:22:52 < zyp> to me it makes much more sense to place the library in a subdirectory of my project, because then I can track it as a git submodule
2012-09-10T01:23:10 < zyp> so that's why I've designed my library to be used that way
2012-09-10T01:23:23 < zyp> Erlkoenig, it's python
2012-09-10T01:23:27 < Erlkoenig> ah thanks
2012-09-10T01:23:38 < zyp> see scons.org
2012-09-10T01:24:55 < Erlkoenig> looks so similar to ruby :D
2012-09-10T01:25:20 < zyp> well, ruby is kind of like a crossbreed between python and perl
2012-09-10T01:25:32 < zyp> syntax-wise :p
2012-09-10T01:25:46 < zyp> so that's no wonder
2012-09-10T01:25:48 < Erlkoenig> possible, i don't know perl...
2012-09-10T01:25:52 < BrainDamage> +1 to the git submodule zyp
2012-09-10T01:26:41 < zyp> it's the most sensible way to keep projects in sync with libraries
2012-09-10T01:27:32 < zyp> copying a library to your project would make it a maintenance hell to update it when new versions come out
2012-09-10T01:28:30 < zyp> and linking to it externally means that when the API changes, you can't just check out any old version of your project and compile it
2012-09-10T01:29:05 <+izua> it does, hmm
2012-09-10T01:29:13 <+izua> but what if the library isn't designed to be included that way
2012-09-10T01:29:28 <+izua> well, true, you can always mess with the lib paths in the makefile
2012-09-10T01:29:46 < Erlkoenig> when a library's API is modified, only by adding new function, the binary should be back-compatible with old programs linked against an old binary, right?
2012-09-10T01:30:24 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-10T01:30:32 <+izua> not if the new function replaces and old function you use, i guess
2012-09-10T01:30:35 < zyp> yes, but that disallows library authors to alter or remove features
2012-09-10T01:30:58 < zyp> and that's a good way to collect up a good bunch of legacy shit over time
2012-09-10T01:31:03 -!- tech2077 [~tech2077@adsl-75-53-137-130.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-09-10T01:31:22 <+izua> you can always write an interface class to your library, and send commands to it, and have the interface class parse them :P
2012-09-10T01:31:51 < zyp> we're doing embedded, we don't want to dick around with legacy overhead
2012-09-10T01:32:20 < Erlkoenig> i was wondering about PC systems... if you don't want to recompile every program that depends on a library, you probably should not modify/remove functions from that...
2012-09-10T01:32:38 < zyp> Erlkoenig, and that's how it works
2012-09-10T01:33:01 < Erlkoenig> "works" is relative... breaks all the time, on linux :D
2012-09-10T01:33:08 < zyp> that's usually the distinction between major and minor versions
2012-09-10T01:33:41 < zyp> minor keeps full compatibility, major may break
2012-09-10T01:35:24 < Erlkoenig> PC libraries dev's should try to never break compatibility, only if it's absolutely impossible otherways, so developing proprietary/closedsource software becomes easier because one version could run on many linux systems that satisfy the least required library versions...
2012-09-10T01:35:58 < Erlkoenig> and if support for closedsource software becomes better, there will be more nice big "killer apps" for linux
2012-09-10T01:36:12 < zyp> well, I'm not interested in that discussion now :)
2012-09-10T01:38:50 < BrainDamage> some libs don't even maintain abi compatibility with a simple recompile ...
2012-09-10T01:38:56 < BrainDamage> like c++ exported functions
2012-09-10T01:39:56 < Erlkoenig> also, an unified deployment systems for closedsource software that works on most distros would be nice - which makes sure all neccessary packages are installed (via the appropriate package manager) and then installs the prop. software in a way that's independant from the package manager. vendors would probably love that...
2012-09-10T01:40:20 < Erlkoenig> BrainDamage: sure? doesn't that apply for different compiler versions?
2012-09-10T01:40:26 < BrainDamage> it does
2012-09-10T01:40:46 < BrainDamage> I left the compiler change implicit
2012-09-10T01:40:55 < BrainDamage> still, distros change compiler rather often
2012-09-10T01:41:05 < BrainDamage> this means rebuilding the dep chain
2012-09-10T01:41:50 < Erlkoenig> compiler devs shouldn't do that :S
2012-09-10T01:42:08 < BrainDamage> also, even saying all libs have a large degree of retrocompatibility, the simple fact that multiple distros can have more or less updated libs
2012-09-10T01:42:15 < BrainDamage> some will have extra functions, some not
2012-09-10T01:42:18 -!- tech2077 [~tech2077@adsl-75-53-134-81.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-10T01:42:20 < zippe> This is one of the things OS X gets right
2012-09-10T01:42:29 < zippe> Because the "app" is really a directory
2012-09-10T01:42:43 < zippe> You can include the "right" versions of all the libraries/frameworks you need in the app
2012-09-10T01:42:47 < BrainDamage> this means lot of optional behaviour to code, so extra efforts
2012-09-10T01:43:00 < Erlkoenig> if i were to develop a library, could i make sure "old" C++ function names (from older versions of my library) don't change in my new binary?
2012-09-10T01:43:09 < BrainDamage> then you don't benefit from system lib updates
2012-09-10T01:43:20 < zippe> Erlkoenig: The issue is that you can't allow the vtable size to change
2012-09-10T01:43:27 < BrainDamage> say a lib has a vulnerability found, it gets patched and deployed
2012-09-10T01:43:33 < zippe> Erlkoenig: Or the offsets to any of the functions in the vtable
2012-09-10T01:43:43 < BrainDamage> since you shipped your own lib version, it means you'll have to rerelease it
2012-09-10T01:43:47 < zippe> Erlkoenig: If you prepare ahead of time, you can add padding slots to your object vtables
2012-09-10T01:43:50 < BrainDamage> it's not very different from static linking
2012-09-10T01:44:05 < zippe> Erlkoenig: it's possible, but it requires a lot of discipline
2012-09-10T01:44:16 < Erlkoenig> aah the vtable
2012-09-10T01:44:17 < BrainDamage> just slightly less problematic since most build systems assume shared linking
2012-09-10T01:46:17 < Erlkoenig> on java this all works :D
2012-09-10T01:46:27 < zippe> Erlkoenig: It's another reason why ObjectiveC is "better" than C++
2012-09-10T01:46:29 < BrainDamage> the main problem is having a sane policy on who should have the final word about what lib should be linked
2012-09-10T01:46:52 < BrainDamage> or even the default
2012-09-10T01:46:58 < zippe> "the main problem" is people who think there is "a main problem"
2012-09-10T01:47:14 < zippe> as opposed to many interlocking problems
2012-09-10T01:47:27 < BrainDamage> it was a rethorical form
2012-09-10T01:47:47 < BrainDamage> just like when people write literally when they mean figuratively
2012-09-10T01:48:33 < BrainDamage> is it right from a syntatical point of view? no, but language is defined by the people speaking it, get it wrong long enough and an exception will be created
2012-09-10T01:49:06 < BrainDamage> as long as you understand what I mean,  and criticize the form, you're just being unnecessarily anal about it
2012-09-10T01:50:05 < Erlkoenig> what's very funny, that the word "mean" seems to be suitable for more places than we where taught in school... :D
2012-09-10T01:51:00 < Erlkoenig> hm, the vtable issue might be fixable by some restructuring, e.g. not keeping the vtable with the object's memory or something... but that would require certain effort from compiler dev's, and can only be done by those nerds...
2012-09-10T01:51:22 < BrainDamage> the sane way is simply not exporting c++ symbols
2012-09-10T01:51:29 < BrainDamage> and using extern c
2012-09-10T01:51:45 < Erlkoenig> and throwing away all the nice OOP stuff. great...
2012-09-10T01:52:38 < BrainDamage> it's not ideal, but it works
2012-09-10T01:52:50 < BrainDamage> and it's what many use
2012-09-10T01:52:53 < Erlkoenig> why is there a vtable for every instance? couldn't be the vtable be a "global static" "object", with one instance per class, and every class has a pointer to it's appropriate "vtable object"?
2012-09-10T01:53:13 < Erlkoenig> well e.g. the GTKmm C++ API is very nice... would be a shame not to use it
2012-09-10T01:59:49 < zyp> huh?
2012-09-10T01:59:58 < zyp> «couldn't be the vtable be a "global static" "object", with one instance per class»
2012-09-10T02:00:03 < zyp> that's exactly how it is
2012-09-10T02:00:55 < Erlkoenig> really? uhm
2012-09-10T02:01:36 < Erlkoenig> well then what's the problem about it changing across library versions? the actual size could be determined during dynamic linking?
2012-09-10T02:02:18 < zyp> sure, there is nothing that prohibits appending more entries to the table
2012-09-10T02:02:30 < zyp> as long as the used part of the table doesn't change
2012-09-10T02:03:28 < Erlkoenig> well that should be no problem - the compilers could easily make sure that this doesn't happen
2012-09-10T02:03:57 < Erlkoenig> e.g. by explicitly specifying the order, or, during compiling, supplying the former binary library, so the compiler knows the desired order
2012-09-10T02:04:18 < zyp> in theory yes
2012-09-10T02:05:05 < zyp> also, vtable layout problems aren't really that different from class layout problems
2012-09-10T02:07:27 -!- tech2077 [~tech2077@adsl-75-53-134-81.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-09-10T02:08:38 < zyp> consider what happens when I add a field in a class in my library, and then you load that library into an application compiled with the old class definition
2012-09-10T02:09:03 < Erlkoenig> sizeof() will be invalid...
2012-09-10T02:09:08 < zyp> exactly
2012-09-10T02:09:15 < Erlkoenig> rather evil
2012-09-10T02:09:22 < Erlkoenig> couldn't sizeof be "dynamically linked" too...
2012-09-10T02:09:29 < Erlkoenig> e.g. known at program start time
2012-09-10T02:09:41 < zyp> so there wouldn't be allocated space for the new field
2012-09-10T02:10:19 < zyp> so it'll blow up as soon as you call an updated function trying to use the new field
2012-09-10T02:13:20 < zyp> of course, there are ways around problems like that
2012-09-10T02:14:23 < zyp> instead of having a shared API that might change, you can do a shared abstract API that doesn't change
2012-09-10T02:14:30 -!- Netsplit *.net <-> *.split quits: zippe
2012-09-10T02:14:48 < Erlkoenig> hm that might be against C++ performance ideas?
2012-09-10T02:15:00 < Erlkoenig> if you do it too abstract, you can use java as well
2012-09-10T02:15:05 < zyp> ha
2012-09-10T02:15:07 < zyp> :p
2012-09-10T02:15:34 -!- Netsplit over, joins: zippe
2012-09-10T02:15:38 < zyp> well
2012-09-10T02:16:14 < zyp> just avoid the problem altogether, don't change a shared API and expect it not to break.
2012-09-10T02:17:09 < Erlkoenig> well, not being able to add new functions/fields to C++ Classes is a rather evil constraint
2012-09-10T02:18:02 < zyp> non-virtual functions are no problem at all
2012-09-10T02:18:14 < Erlkoenig> at least.
2012-09-10T02:18:19 < zyp> they don't have any content in the class
2012-09-10T02:19:02 < Erlkoenig> but fields... new functions mean new fields :D
2012-09-10T02:19:05 -!- tech2077 [~tech2077@adsl-76-247-248-117.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-10T02:19:13 < zyp> and really, the problem is the same in C, if you change a shared struct
2012-09-10T02:19:39 < zyp> no, functions are not fields
2012-09-10T02:20:09 < zyp> non-virtual member functions are just normal functions that takes an instance pointer as an implicit first argument
2012-09-10T02:20:32 < zyp> they don't have anything to do with the layout of the class at all
2012-09-10T02:21:45 < Erlkoenig> yeah i got it... i meant, if you add functions, you most likely add new functionality, which most likely needs new fields
2012-09-10T02:34:12 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-09-10T02:35:24 -!- tech2077 [~tech2077@adsl-76-247-248-117.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-09-10T02:36:48 -!- daku [DaKu@dakus.dk] has quit [Quit: Bye]
2012-09-10T02:46:29 -!- tech2077 [~tech2077@adsl-76-247-248-149.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-10T03:02:48 -!- tech2077 [~tech2077@adsl-76-247-248-149.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-09-10T03:14:43 -!- tech2077 [~tech2077@adsl-76-247-248-230.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-10T03:30:12 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 248 seconds]
2012-09-10T03:54:53 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-10T04:02:51 <+dekar_> izua, http://pastebin.com/huyb8ukX
2012-09-10T04:04:14 < dongs> < Erlkoenig> well e.g. the GTKmm C++ API is very nice... would be a shame not to use it
2012-09-10T04:04:19 < dongs> > gtk
2012-09-10T04:04:22 < dongs> > c++
2012-09-10T04:04:25 < dongs> > nice
2012-09-10T04:04:27 < dongs> still laughing
2012-09-10T04:04:30 < Erlkoenig> wat?
2012-09-10T04:04:40 < Erlkoenig> show me a better GUI API
2012-09-10T04:05:09 <+dekar_> like Qt? :)
2012-09-10T04:05:22 < Erlkoenig> aren't Qt and GTKmm more or less equal...
2012-09-10T04:05:36 <+dekar_> tbh I've never even tried GTK
2012-09-10T04:05:40 < Erlkoenig> and even if not, my above statement applies for Qt as well
2012-09-10T04:06:01 <+dekar_> but I heard GTK wouldn't be object oriented
2012-09-10T04:06:15 < zippe> Erlkoenig: Appkit sucks less
2012-09-10T04:06:30 <+dekar_> I love the portability and the IDE of Qt
2012-09-10T04:06:36 < Erlkoenig> it's very (perhaps too) much object oriented. as far as possible in C. and GTKmm takes advantage of this in C++
2012-09-10T04:07:05 < zippe> Though I have to agree, gtk# beats the snot out of MFC
2012-09-10T04:07:10 < Erlkoenig> Gtk is portable as well, and has some sort of designer too... but you don't really need it
2012-09-10T04:07:25 < Erlkoenig> what about Gtk sucks?
2012-09-10T04:07:30 <+dekar_> well Qt is more than just C++, they have their own preprocessor to add new features to C++ that normally wouldn't be possible
2012-09-10T04:07:50 < Erlkoenig> AppKit is for Mac... so forget it
2012-09-10T04:08:03 < Erlkoenig> dekar_: yeah, and that's exactly the problem of Qt :D
2012-09-10T04:08:05 <+dekar_> Erlkoenig, GTK is far less portable though
2012-09-10T04:08:15 < Erlkoenig> hmm portable enough for me
2012-09-10T04:08:50 < Erlkoenig> dekar_: GTKmm does the same nice things by using the STL and libsigc++
2012-09-10T04:08:59 < Erlkoenig> and is thereby more c++-ish
2012-09-10T04:09:12 <+dekar_> I guess GTK runs on the major OS, but Qt is about everywhere - even on VxWorks and Symbian
2012-09-10T04:10:07 <+dekar_> Erlkoenig, hows that a problem? the way Qt implements threading and event driven programming wouldn't be possible with C++
2012-09-10T04:10:25 <+dekar_> I think it's awesome
2012-09-10T04:10:34 <+izua> dekar_: hey. neat
2012-09-10T04:10:38 < zippe> dekar: good luck building an app that works well with Qt on all those platforms. 8)
2012-09-10T04:10:46 <+izua> also +1 on Qt, but i don't really like the QtCreator
2012-09-10T04:10:57 < Erlkoenig> because a preprocessor adds more complexity... but whatever... what do you think exactly is not possible in C++ without a preprocessor?
2012-09-10T04:11:10 <+izua> zippe: it also has its own scripting language and a lot of helper libs
2012-09-10T04:11:14 < Erlkoenig> the libsigc++ based event system is also very nice, and type safe
2012-09-10T04:11:18 <+izua> so you don't really have to rely on os level libs
2012-09-10T04:11:43 <+dekar_> yeah Qt is more like the Java standard library, it's basically an operating system abstraction layer
2012-09-10T04:11:47 < zippe> izua: still, you're not going to build a single app with a the same UI on all those platforms.
2012-09-10T04:11:54 <+dekar_> makes stuff uber portable
2012-09-10T04:12:04 <+izua> no, actually it will use the native UI on each platform
2012-09-10T04:12:14 < zippe> I've seen enough non-portable Qt apps to find this fairly amusing
2012-09-10T04:12:15 <+izua> but you can force a given UI on every platform
2012-09-10T04:12:23 < zippe> izua: you aren't thinking straight
2012-09-10T04:12:32 <+izua> check the demos
2012-09-10T04:12:34 <+dekar_> zippe, depends on what you do, I use the same UI on OSX, Linux and Windows - for android, iPhone etc I would make a new one ofc
2012-09-10T04:12:35 < zippe> izua: the user interaction model is very different
2012-09-10T04:12:38 <+izua> you can force plastik on windows
2012-09-10T04:12:45 <+izua> xp looks on mac, and so on
2012-09-10T04:12:47 <+izua> ah
2012-09-10T04:12:50 <+izua> yes, definitely
2012-09-10T04:12:52 < zippe> dekar_: takes a lot of work to get the fonts to behave
2012-09-10T04:12:55 <+izua> but it just emulates the look
2012-09-10T04:13:05 <+izua> and it does it well
2012-09-10T04:13:20 < zippe> dekar_: far too often the UI will look fine on one system, but pick bad fonts and layout wrong on any other
2012-09-10T04:13:35 <+izua> then again, there's little reason to have the same UI on every platform
2012-09-10T04:13:41 <+dekar_> zippe, cause the other OS lacks the font, I had that before as well
2012-09-10T04:14:19 <+izua> dekar_: the idea is - as i've understood it - not to rely on system metrics at all, just put everything in the boxlayouts and give hints
2012-09-10T04:14:20 < zippe> It's not "lacks the font", it's that the layout tools embed knowledge of the original font into the layout, rather than dynamically basing their layout on the font
2012-09-10T04:14:24 <+dekar_> but it still is reasonably portable, I think it's comparable to java
2012-09-10T04:14:25 < zippe> Hell, even Tk gets that one right.
2012-09-10T04:14:51 -!- tech2077 [~tech2077@adsl-76-247-248-230.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-09-10T04:15:04 <+dekar_> izua, I had Qt switch on windows to some uber small font since the one I picked on OSX was missing
2012-09-10T04:15:19 < Erlkoenig> many windows layouts are based on the assumption "font = black, background = white" ... if you change that systemwide, everything breaks
2012-09-10T04:15:20 <+dekar_> but I guess I should just embed the font to begin with
2012-09-10T04:17:52 <+izua> i guess if you really want it to look the same, you can just embed the droid fonts
2012-09-10T04:18:12 <+izua> (not sure if there are any other better looking and free ui fonts)
2012-09-10T04:18:42 <+izua> and it would also help the 'feel' across the platforms, if that's what one's aiming for
2012-09-10T04:21:41 < qyx_> 03:13 <+izua> then again, there's little reason to have the same UI on every platform
2012-09-10T04:21:45 < qyx_> have to agree ^
2012-09-10T04:25:02 <+dekar_> http://pastebin.com/rcSU192u
2012-09-10T04:25:35 <+dekar_> multilib is a mess to get right
2012-09-10T04:25:45 <+dekar_> you basically have to blacklist everything that makes no sense
2012-09-10T04:26:21 -!- tech2077 [~tech2077@adsl-75-53-129-61.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-10T04:28:47 < Erlkoenig> what exactly do you need that for?
2012-09-10T04:29:17 <+dekar_> people were complaining about TNT not supporting weird architectures, so now I'm adding them all
2012-09-10T04:29:35 <+dekar_> I wish them fun compiling for weeks
2012-09-10T04:29:46 < Erlkoenig> what, just an ARM toolchain?
2012-09-10T04:30:11 <+dekar_> just multilib yeah
2012-09-10T04:30:35 < Erlkoenig> https://launchpad.net/gcc-arm-embedded won't this do the job?
2012-09-10T04:32:16 <+dekar_> https://answers.launchpad.net/gcc-arm-embedded/+question/196873
2012-09-10T04:33:01 <+dekar_> besides I'd have to compile it anyway
2012-09-10T04:34:42 < Erlkoenig> oh hm
2012-09-10T04:35:38 < dongs> basicallyt  if youre worried about code size or quality
2012-09-10T04:35:39 < dongs> just use armcc
2012-09-10T04:35:42 < dongs> problem fuckin' solved.
2012-09-10T04:35:53 <+dekar_> Erlkoenig, how do they configure newlib?
2012-09-10T04:35:56 < dongs> even latest gcc .47 doesnt come close to how good armcc code is generated
2012-09-10T04:36:19 < Erlkoenig> dekar_: i dunno? i only know it works :D
2012-09-10T04:36:38 < Erlkoenig> dongs: but armcc costs money doesnt it?
2012-09-10T04:37:12 < dongs> well, yes. thats why it works I suspect
2012-09-10T04:37:37 < Erlkoenig> so one either has to buy a bigger controller or a compiler... :D
2012-09-10T04:38:09 < Erlkoenig> hmm in a forum someone asks whether ARM's are as complicated to program as AVR's, or rather as easy as Arduino's.
2012-09-10T04:38:23 < dongs> Erlkoenig: kill it with fire.
2012-09-10T04:38:33 < Erlkoenig> harhar :D
2012-09-10T04:39:04 < Erlkoenig> the complicated part about AVR's being the many registers to be programmed bitwise
2012-09-10T04:39:41 < dongs> ??????????????????????????????????
2012-09-10T04:39:58 < dongs> as opposed to any register on stm32?
2012-09-10T04:39:59 < Erlkoenig> that's what i thought :D
2012-09-10T04:40:19 < dongs> stop reading tarduinoforumsforretards.com
2012-09-10T04:40:49 < Erlkoenig> it's just mikrocontroller.net there is also much useful stuff :D
2012-09-10T04:46:54 < dongs> only if you like dudes
2012-09-10T04:47:09 < dongs> and?
2012-09-10T04:47:22 < Erlkoenig> what does "dongs" mean?
2012-09-10T04:47:28 < Erlkoenig> oh
2012-09-10T04:47:30 < Erlkoenig> good
2012-09-10T04:47:58 < Erlkoenig> (03:46:55) flyback: it's all written in angry krout but neat site <-- in fact, many people are there are much more polite than in other forums
2012-09-10T04:48:16 < Erlkoenig> but it's the only forum where the admins may make fun of poor stubid newbs :D
2012-09-10T04:48:25 < Erlkoenig> ha
2012-09-10T04:48:29 < Erlkoenig> beware of the krauts \o/
2012-09-10T04:48:35 < Erlkoenig> what?!
2012-09-10T04:49:14 < Erlkoenig> well maybe *some* people are smart and develop neat things :D
2012-09-10T04:49:23 < dongs> disliking hitler is a crime
2012-09-10T04:49:44 < Erlkoenig> but yeah, that's really very funny, that basically every rocket tech is based on Werner van Braun
2012-09-10T04:50:07 < Erlkoenig> and it's first use was bombing london
2012-09-10T04:50:37 < Erlkoenig> hmm dunno much about him
2012-09-10T04:51:43 < Erlkoenig> i find it very amusin that there are nazi's in other countries too... do they hate 'their' own countries or what? :D
2012-09-10T05:14:14 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-115.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-10T05:17:41 < dongs> commence trolling
2012-09-10T05:30:15 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 276 seconds]
2012-09-10T05:35:47 < dongs> live stream please
2012-09-10T05:52:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-10T06:02:53 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-10T06:02:53 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-10T06:05:12 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-10T06:06:59 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-09-10T06:09:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-10T06:22:39 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-10T06:50:14 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-10T06:58:07 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-10T07:02:52 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 272 seconds]
2012-09-10T07:18:18 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-10T07:19:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has left ##stm32 []
2012-09-10T07:27:32 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Ping timeout: 248 seconds]
2012-09-10T07:29:57 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-09-10T07:30:49 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-10T07:53:22 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-09-10T08:21:45 -!- claude is now known as Claude
2012-09-10T08:22:13 < dongs> this is how i (t)roll http://i.imgur.com/dUJtM.jpeg
2012-09-10T08:30:00 < zippe> dongs: I should have taken a photo of the vice grips I used to use as a motor jig
2012-09-10T08:30:13 < zippe> Though I was pimped with some Wago terminals
2012-09-10T08:32:02 < dongs> heh
2012-09-10T08:32:06 < dongs> wago looks useless for my purpose
2012-09-10T08:32:16 < dongs> its much easier to push all 3 clips at once
2012-09-10T08:35:00 < GargantuaSauce> i have cuts on both hands from spinning up outrunners while holding onto them :V
2012-09-10T08:35:44 < GargantuaSauce> cuts from propellers that is
2012-09-10T08:39:29 < zippe> dongs: quick-releases on the Waygos… DIN rail mounting too
2012-09-10T08:41:44 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-10T08:46:05 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-10T09:22:34 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-10T09:25:40 -!- Claude is now known as claude
2012-09-10T09:39:29 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 240 seconds]
2012-09-10T09:39:56 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-10T10:09:05 -!- dekar_ [~dekar@212.255.31.74] has quit [Quit: This computer has gone to sleep]
2012-09-10T10:16:06 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-10T10:18:36 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-10T10:24:39 -!- carp3 [~chatzilla@31.59.27.40] has joined ##stm32
2012-09-10T10:24:57 < carp3> Hi, any have any sample  for using STM32F4 ADC in triple mode ?
2012-09-10T10:27:04 < dongs> Use STM32F2's Triple Regular Simultaneous Mode capable of ~ 6M samples per second
2012-09-10T10:27:12 < dongs> http://code.google.com/p/autoquad/source/browse/trunk/onboard/adc.c#382
2012-09-10T10:27:18 < dongs> F2 ADC/DMA is same.
2012-09-10T10:27:21 < dongs> as F4.
2012-09-10T10:31:34 -!- carp3 [~chatzilla@31.59.27.40] has quit [Remote host closed the connection]
2012-09-10T10:33:09 -!- carp3 [~chatzilla@31.59.27.40] has joined ##stm32
2012-09-10T10:34:23 < Laurenceb_> http://www.youtube.com/watch?v=6isXNVdguI8
2012-09-10T10:50:28 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-10T10:52:15 < Laurenceb_> seems applicable to flyback
2012-09-10T10:56:07 < Laurenceb_> http://robotgrrl.com/blog/2012/05/25/arduino-due-up-close-pics/
2012-09-10T10:56:09 < Laurenceb_> omg
2012-09-10T10:56:14 < Laurenceb_>  /sarc
2012-09-10T10:57:46 -!- dekar [~dekar@dslb-188-107-196-119.pools.arcor-ip.net] has joined ##stm32
2012-09-10T10:57:49 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-10T11:02:24 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-10T11:04:04 < dongs> Laurenceb jizzed at tarduino and pulled his network cable out
2012-09-10T11:15:27 -!- jon1012 [~jon@88.162.134.157] has joined ##stm32
2012-09-10T11:15:28 -!- jon1012 [~jon@88.162.134.157] has quit [Changing host]
2012-09-10T11:15:28 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-10T11:17:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-10T11:17:21 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-10T11:20:41 < dongs> whats proper invsqrt function
2012-09-10T11:20:56 < dongs> is it just 1/sqrt?
2012-09-10T11:20:57 < dongs> or wat
2012-09-10T11:24:22 -!- carp3 [~chatzilla@31.59.27.40] has quit [Remote host closed the connection]
2012-09-10T11:42:34 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-10T11:51:10 < jpa-> dongs: yeah, though accuracy may be better by directly calculating invsqrt
2012-09-10T11:51:33 < jpa-> how is USART in synchronous mode different from SPI? the waveforms are pretty similar
2012-09-10T11:51:45 < jpa-> (the register interface is of course different)
2012-09-10T11:54:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-10T11:54:27 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-10T11:55:46 < dongs> jpa, can you do usart in sync mode without start/stop bits?
2012-09-10T12:00:18 < jpa-> kind-of 
2012-09-10T12:00:19 < jpa-> Anturiportti 1:
2012-09-10T12:00:19 < jpa-> |_.Konfiguraatio	|_.Analog				|_.SPI1		|_.I2C1		|_.USART2	|_.TIMER 	|_.IRQ	| Rinnakkaisia nastoja 	|
2012-09-10T12:00:19 < jpa-> |_.Nasta 1		|ADC12_IN7 (PA7)			|MOSI (PA7) 	|SCL (PB6)	|		| 		|EXTI7	|2			|
2012-09-10T12:00:19 < jpa-> |_.Nasta 2		|ADC12_IN6 (PA6)			|MISO (PA6)	|		|RX (PA3)	|TIM2_CH4 (PA3)	|EXTI6	|2			|
2012-09-10T12:00:19 < jpa-> |_.Nasta 3		|ADC123_IN2 (PA2)			|SCK (PB3)	|		|TX (PA2)	|TIM2_CH3 (PA2)	|EXTI3	|2			|
2012-09-10T12:00:19 < jpa-> |_.Nasta 4		|ADC12_IN4 DAC1_OUT (PA4)		|		|SDA (PB9)	|		|TIM4_CH4 (PB9)	|EXTI4	|2			|
2012-09-10T12:00:19 < jpa-> Anturiportti 2:
2012-09-10T12:00:19 < jpa-> |_.Konfiguraatio	|_.Analog				|_.SPI2		|_.I2C2		|_.UART4	|_.TIMER 	|_.IRQ	| Rinnakkaisia nastoja 	|
2012-09-10T12:00:19 < jpa-> |_.Nasta 1		|ADC123_IN13 (PC3)			|MOSI (PC3)	|SCL (PB10)	|		|  		|EXTI10	|2			|
2012-09-10T12:00:19 < jpa-> |_.Nasta 2		|ADC123_IN12 (PC2)			|MISO (PC2)	|		|RX (PA1)	|TIM5_CH2 (PA1)	|EXTI2	|2			|
2012-09-10T12:00:19 < jpa-> |_.Nasta 3		|ADC123_IN0 (PA0)			|SCK (PB13)	|		|TX (PA0)	|TIM5_CH1 (PA0)	|EXTI13	|2			|
2012-09-10T12:00:26 < jpa-> gaaah
2012-09-10T12:00:27 < jpa-> sorry
2012-09-10T12:00:46 <+Steffanx> yay
2012-09-10T12:00:52 <+Steffanx> Slow paste ftw!
2012-09-10T12:00:59 < jpa-> "No clock pulses are sent to the SCLK pin during start bit and stop bit."  that
2012-09-10T12:02:49 <+Steffanx> At least we learnt some finnish
2012-09-10T12:02:58 <+Steffanx> nastoja = pins where nasta = pin :P
2012-09-10T12:03:21 <+Steffanx> Google translate is goood
2012-09-10T12:03:48 < jpa-> don't steal my secret pin assignments!
2012-09-10T12:06:23 < zyp> ei saa peitää
2012-09-10T12:07:10 <+Steffanx> zlog knows it all, jpa-
2012-09-10T12:07:11 < zlog> Steffanx: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-09-10.html
2012-09-10T12:07:29 < Oldboy> 'Anturiportti' sounds cute
2012-09-10T12:13:36 < dongs> < jpa-> "No clock pulses are sent to the SCLK pin during start bit and stop bit."  that
2012-09-10T12:13:39 < dongs> jpa-: thats useless.
2012-09-10T12:13:54 < dongs> if there's breaks between bytes, that would be fail for my purposes.
2012-09-10T12:13:59 < dongs> kinda like PIC with its SPI DMA
2012-09-10T12:14:11 < dongs> where it inserts a extra clock cycle between each byte transfer lo
2012-09-10T12:15:56 < jpa-> true, but works for me
2012-09-10T12:16:06 < dongs> so theres no way to have 0start/0stop bits?
2012-09-10T12:16:10 < dongs> and no parity bit
2012-09-10T12:16:11 < dongs> in usart
2012-09-10T12:26:24 < Laurenceb> <dongs> whats proper invsqrt function
2012-09-10T12:26:27 < Laurenceb> fail
2012-09-10T12:26:55 < Laurenceb> its called SQUARE root for a reason
2012-09-10T12:27:50 <+Steffanx> Explain Laurenceb
2012-09-10T12:28:10 < Laurenceb> the inverse of square root is square
2012-09-10T12:28:12 < Laurenceb> http://www.st.com/internet/mcu/subclass/1605.jsp
2012-09-10T12:28:18 < Laurenceb> FUCK YEAH
2012-09-10T12:28:23 < zyp> Laurenceb, invsqrt usually refers to the reciprocal
2012-09-10T12:28:30 < zyp> i.e. 1/sqrt
2012-09-10T12:28:41 < Laurenceb> then deivide?
2012-09-10T12:28:46 <+Steffanx> Still no reference manual Laurenceb
2012-09-10T12:28:51 < Laurenceb> erm
2012-09-10T12:29:06 < zyp> Laurenceb, but it's faster to calculate it directly than doing a sqrt() first and then inverting it
2012-09-10T12:29:17 < Laurenceb> ah, fairdoos
2012-09-10T12:29:17 <+Steffanx> Oh, there is one now
2012-09-10T12:29:33 <+Steffanx> dsf
2012-09-10T12:29:35 < zyp> see for instance http://en.wikipedia.org/wiki/Fast_inverse_square_root
2012-09-10T12:31:34 <+Steffanx> Since when the ref. man. for a F3 is online?
2012-09-10T12:31:44 < Laurenceb> 16 bit adc with 32x pga
2012-09-10T12:31:48 < Laurenceb> this morning
2012-09-10T12:31:58 < zyp> I did some even faster variant for quaternion normalization
2012-09-10T12:32:07 < Laurenceb> interesting
2012-09-10T12:32:31 < Laurenceb> you're writing a complimentary filter from scratch?
2012-09-10T12:32:54 < zyp> I did some shit for a work project last year
2012-09-10T12:33:02 < zyp> running on AVR with 8.24 fixedpoint
2012-09-10T12:33:02 < Laurenceb> ah
2012-09-10T12:33:06 < Laurenceb> lol
2012-09-10T12:33:18 < zyp> so float tricks doesn't apply
2012-09-10T12:33:28 < Laurenceb> grrr
2012-09-10T12:33:33 < Laurenceb> only one ref manual
2012-09-10T12:33:41 < Laurenceb> for the f3 with 16bit adc
2012-09-10T12:33:50 < Laurenceb> the fast 12 bit adc version doesnt have a manual
2012-09-10T12:33:55 < Laurenceb> .. yet
2012-09-10T12:34:09 <+Steffanx> You and your ADC-bits fetish :P
2012-09-10T12:34:22 < Laurenceb> i want to oversample it
2012-09-10T12:34:32 < Laurenceb> the 16bit adc is a bit poor
2012-09-10T12:34:39 < zyp> but anyway, since a quaternion is _almost_ normalized at any time, you just renormalize it to keep it from growing by errors adding up
2012-09-10T12:34:57 < zyp> so I just did a taylor expansion around 1
2012-09-10T12:35:06 < Laurenceb> yeah
2012-09-10T12:36:18 < Laurenceb> Infrared interface (IRTIM) looks new
2012-09-10T12:36:20 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-10T12:37:17 < Laurenceb> and a touch sensor interface too
2012-09-10T12:37:58 < Laurenceb> hdmi-cec hardware - isnt that one-wire compatible?
2012-09-10T12:39:07 < zyp> what's the difference between F30 and F37?
2012-09-10T12:39:21 < Laurenceb> 16bit adc rather than fast adc
2012-09-10T12:39:22 < zyp> and also F38?
2012-09-10T12:39:26 < Laurenceb> hmm
2012-09-10T12:39:36 < Laurenceb> theres only 2 datasheets
2012-09-10T12:39:44 < zyp> yeah, and no parts listed for F38
2012-09-10T12:39:54 < zyp> but reference manual only lists F37 and F38, not F30
2012-09-10T12:40:17 < Laurenceb> also the F30 supposidly has general purpose opamps for building analogue blocks
2012-09-10T12:40:26 < Laurenceb> *supposedly
2012-09-10T12:40:55 < Laurenceb> a bit like a Psoc
2012-09-10T12:41:37 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-10T12:43:38 < zyp> wasn't it supposed to have a CCM on the I-bus?
2012-09-10T12:43:47 < zyp> I see no mention of that in the RM
2012-09-10T12:46:39 < Laurenceb> hmm
2012-09-10T12:46:47 < Laurenceb> simple bus matrix that F4 too
2012-09-10T12:46:52 < Laurenceb> *than
2012-09-10T12:48:09 < zyp> yes, it looks similar to F1
2012-09-10T12:48:22 < zyp> there is no cache either (ART accelerator)
2012-09-10T12:49:10 < Laurenceb> looks like the fast adc version would saturate the memory bus pretty easily
2012-09-10T12:51:25 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-10T12:56:52 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 244 seconds]
2012-09-10T12:58:19 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-10T13:02:22 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-10T13:09:09 < zyp> hmm, there's going to be a NFC EEPROM on the F3discovery
2012-09-10T13:09:27 < zyp> http://www.st.com/internet/mcu/subclass/1412.jsp <- probably one of these
2012-09-10T13:12:46 < Laurenceb> odd ic
2012-09-10T13:13:47 < zyp> I'm just wondering why they put it on the F3 discovery
2012-09-10T13:13:57 < zyp> seems pretty unrelated
2012-09-10T13:23:06 < Laurenceb> odd- first conversion on adc on my F1 is screwed
2012-09-10T13:23:12 < Laurenceb> is that normal?
2012-09-10T13:27:15 < Laurenceb> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATA_BRIEF/DM00063389.pdf
2012-09-10T13:27:18 < Laurenceb> i dont see it
2012-09-10T13:27:38 < Laurenceb> also thats F30 - the fast adc kit XD
2012-09-10T13:27:53 < Laurenceb> Ten LEDs:
2012-09-10T13:27:57 < Laurenceb> omg blinky
2012-09-10T13:31:58 < karlp> hey, at least they're admitting that the output power is 3v, not 3v3 like they claim on the other boards.
2012-09-10T13:32:17 < karlp> (I have a sensor that only just works at 3.3v, and doesn't work at all with 3.0v)
2012-09-10T13:38:14 < Laurenceb> i dont understand the reason for the goofy diode to drop the voltage
2012-09-10T13:39:49 < Oldboy> diode ?
2012-09-10T13:41:44 < zyp> ah, no, I read wrong
2012-09-10T13:41:52 < zyp> it's only on the larger F3 demo boards
2012-09-10T13:42:46 < zyp> it said «both boards», I see now that it means «both F30 and F37 variants of the large board»
2012-09-10T14:14:31 < Laurenceb> wtf
2012-09-10T14:14:45 < Laurenceb> ive got some seriously odd adc behaviour
2012-09-10T14:15:21 < Laurenceb> first conversion gives me 1659
2012-09-10T14:15:35 < Laurenceb> then the next too
2012-09-10T14:15:44 < Laurenceb> but if i halt and wait i get 2583
2012-09-10T14:15:48 < Laurenceb> which is correct
2012-09-10T14:15:56 < Laurenceb> - approx actual voltage
2012-09-10T14:16:14 < Laurenceb> yet the voltage on the pin is stable
2012-09-10T14:20:32 < dongs> Laurenceb: are you calibrating ADC on power up
2012-09-10T14:20:41 < Laurenceb> yes
2012-09-10T14:20:46 < Laurenceb> this makes no sense
2012-09-10T14:20:54 < Laurenceb> its giving correct data
2012-09-10T14:21:02 < Laurenceb> but i have to jtag halt first
2012-09-10T14:21:31 < Laurenceb> ill try two conversions with a massive time delay
2012-09-10T14:22:39 < Laurenceb> hmm same issue
2012-09-10T14:22:40 < Laurenceb> wtf
2012-09-10T14:22:40 < dongs> uh...
2012-09-10T14:25:19 < Laurenceb> oh wait - i think watchdog keeps killing it
2012-09-10T14:25:26 < Laurenceb> and jtag is doing something weird
2012-09-10T14:31:44 < dongs> eh, touch sensing controller eh
2012-09-10T14:33:15 < Laurenceb> ah
2012-09-10T14:33:28 < Laurenceb> first adc conversion is corrupted
2012-09-10T14:33:45 < Laurenceb> and i need to leave a little gap between conversions due to high source impedance
2012-09-10T14:34:00 < dongs> shoulda put a blogamp in there
2012-09-10T14:35:23 < Laurenceb> im measuring battery voltage through some 30K resistors
2012-09-10T14:35:42 < Laurenceb> odd abotu the first conversion tho
2012-09-10T14:35:56 < Laurenceb> seems first software conversion on adc2 after powerup is bad
2012-09-10T14:36:08 < dongs> im certain youre doing it wrong
2012-09-10T14:36:24 < dongs> by "after power up" you mean "after i calibrated adC" right
2012-09-10T14:36:29 < Laurenceb> yes
2012-09-10T14:38:33 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/main.c#L155
2012-09-10T14:38:37 < Laurenceb> haxored it like that
2012-09-10T14:39:20 < dongs> christ.
2012-09-10T14:39:26 < Laurenceb> lolz
2012-09-10T14:39:28 < dongs> that code makes me wanna stab
2012-09-10T14:40:05 < Laurenceb> whats so bad?
2012-09-10T14:40:26 < dongs> nothing bad. just in general.
2012-09-10T14:40:37 < dongs> spaces missing EVERYWEHRE.
2012-09-10T14:41:47 < dongs> comparators on F3 look nutty
2012-09-10T14:48:24 < dongs> hmm
2012-09-10T14:48:29 < dongs> they fucked with FT io again
2012-09-10T14:49:14 < Laurenceb> i want to see the other ref manual
2012-09-10T14:49:19 < Laurenceb> F30x
2012-09-10T14:49:33 < Laurenceb> and see if its actually has configurable opamp blocks
2012-09-10T14:50:15 < dongs> The STM32F30xx embeds four operational amplifiers that can either be used as standalone
2012-09-10T14:50:18 < dongs> amplifiers or as follower / programmable gain amplifiers. The operational amplifier output is
2012-09-10T14:50:21 < dongs> internally connected to an ADC channel for measurement purposes.
2012-09-10T14:50:23 < dongs> it does.
2012-09-10T14:50:53 < Laurenceb> http://www.cypress.com/?id=2233
2012-09-10T14:50:57 < Laurenceb> or theres that
2012-09-10T14:51:02 < dongs> ew.
2012-09-10T14:51:03 < dongs> psoc
2012-09-10T14:51:05 < dongs> utteer fail
2012-09-10T14:51:21 < dongs> Laurenceb: http://i.imgur.com/KQYC6.png
2012-09-10T14:53:52 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-10T14:57:08 < Laurenceb> you have the ref manual?
2012-09-10T14:57:11 < Laurenceb> lol psoc is fail
2012-09-10T14:57:19 < dongs> haha, they put other functions on OSC_in/OSC_out
2012-09-10T14:57:20 < dongs> in F3
2012-09-10T14:57:26 < dongs> like timer/spi outputs
2012-09-10T14:57:27 < Laurenceb> i can see anything it does that stm cant
2012-09-10T14:57:40 < dongs> Laurenceb: you  get a gay GUI designer
2012-09-10T14:57:44 < dongs> with psoc
2012-09-10T14:58:20 < Laurenceb> lol
2012-09-10T14:58:28 < Laurenceb> wtf.. where are the opamp outputs?
2012-09-10T14:58:40 < dongs> /./ they're internally routed to adc, duh
2012-09-10T14:59:07 < Laurenceb> yeah but if they are opamps what will be unstable
2012-09-10T14:59:34 < dongs> also you can route shit to comparators
2012-09-10T14:59:35 < Laurenceb> you need a feedback path
2012-09-10T14:59:44 < Laurenceb> yes
2012-09-10T14:59:59 < zyp> yeah, no point in having opamps if there is no feedback
2012-09-10T15:00:26 < Laurenceb> the F37 has PGA
2012-09-10T15:00:32 < Laurenceb> i was interested in opamps
2012-09-10T15:00:33 < zyp> unless the programmable gain part means internal feedback
2012-09-10T15:00:38 < dongs> Standalone mode (external gain setting mode)
2012-09-10T15:00:39 < dongs> . Follower configuration mode
2012-09-10T15:00:39 < dongs> . PGA modes
2012-09-10T15:00:54 < dongs> the amplifier output pin is directly connected to the output pad to minimize
2012-09-10T15:00:55 < dongs> the output impedance. It cannot be used as a general purpose I/O, even if the amplifier is
2012-09-10T15:00:56 < Laurenceb> standalone sounds good
2012-09-10T15:00:58 < dongs> configured as a PGA and only connected to the ADC channel.
2012-09-10T15:01:01 < dongs> apparently there are opamp output pins
2012-09-10T15:01:04 < dongs> probably on 100qfp
2012-09-10T15:01:09 < dongs> did you check t he datasheet pinout?
2012-09-10T15:01:19 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 260 seconds]
2012-09-10T15:01:20 < Laurenceb> good plan
2012-09-10T15:02:59 < dongs> http://i.imgur.com/Z7Txe.png
2012-09-10T15:04:22 < zyp> nice
2012-09-10T15:04:37 < Laurenceb> AOP_OUT
2012-09-10T15:04:38 < Laurenceb> i see
2012-09-10T15:04:59 < Laurenceb> awesome
2012-09-10T15:05:12 < Laurenceb> my entire logger couple be on a single lqfp-48
2012-09-10T15:05:24 < Laurenceb> is there spec on voltage noise?
2012-09-10T15:08:26 < Laurenceb> for the opamps
2012-09-10T15:09:30 < dongs> i dunno, what would it be called?
2012-09-10T15:10:00 < dongs> i'd suspect that info is in datasheet?
2012-09-10T15:11:35 < dongs> fuck..
2012-09-10T15:11:40 < dongs> BOOT1 is in different place...
2012-09-10T15:12:09 < dongs> on lqfp48 :|
2012-09-10T15:12:11 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-10T15:12:22 < dongs> .. unless they fucked up in teh docs :p
2012-09-10T15:13:09 < dongs> most of IO on F3 is not 5V tolerant :(
2012-09-10T15:13:10 < dongs> what the fuck
2012-09-10T15:13:30 < dongs> infact its worse than F1
2012-09-10T15:13:40 < dongs> some of the shit is "3.3V that will fucking burn at anything higher"
2012-09-10T15:15:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T15:15:14 < Laurenceb> ah yeah its in datasheet
2012-09-10T15:15:17 < Laurenceb> but no noise spec
2012-09-10T15:17:17 < Laurenceb> opamps dont look too bad
2012-09-10T15:18:48 < Laurenceb> needs a 3.3v ldo built in
2012-09-10T15:37:52 < Laurenceb> http://i.imgur.com/WJwLO.jpg
2012-09-10T15:39:30 < Laurenceb> Lakenheath USAF base
2012-09-10T15:42:47 < dongs> cool avr bro
2012-09-10T15:43:33 -!- claude is now known as claude_work
2012-09-10T15:44:35 < Laurenceb> http://media.digikey.com/Photos/STMicro%20Photos/MFG_STM32%20F3%20Discovery.jpg
2012-09-10T15:44:46 < Laurenceb> looks good for developing imu code
2012-09-10T15:44:50 < dongs> cute
2012-09-10T15:44:56 < dongs> how so? isnt that just accel
2012-09-10T15:45:01 < dongs> or did they stick a gyro+acc there
2012-09-10T15:45:03 < Laurenceb> nope
2012-09-10T15:45:12 < Laurenceb> msl303 and the new gyro
2012-09-10T15:45:15 < Laurenceb> *lsm
2012-09-10T15:45:39 < dongs> ha
2012-09-10T15:45:49 < dongs> hueg f3 chip though
2012-09-10T15:45:49 < dongs> argh
2012-09-10T15:46:13 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-10T15:46:15 < BrainDamage> I wish ST would stop mounting pin headers at all on the boards and simply leave the holes and ship the pin headers with them separately
2012-09-10T15:46:33 -!- DaKu_ [DaKu@dakus.dk] has joined ##stm32
2012-09-10T15:50:31 < dongs> S0204: STM32F302xx, STM32F303xx and STM32F313xx Rev Z and Y device limitations
2012-09-10T15:50:32 < dongs> lol
2012-09-10T15:50:36 < dongs> already errata for f3
2012-09-10T15:50:58 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 244 seconds]
2012-09-10T15:51:19 < Laurenceb> fail
2012-09-10T15:51:47 < Laurenceb> presumably someone tried to program the f3discovery
2012-09-10T15:52:43 < dongs> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/ERRATA_SHEET/DM00063985.pdf
2012-09-10T15:52:46 < dongs> some epic fails
2012-09-10T15:52:53 < dongs> If the CPU writes to an address A in the SRAM memory and immediately (the cycle after)
2012-09-10T15:52:56 < dongs> reads an address B in the SRAM memory, while B = A+0x4000, the read operation will
2012-09-10T15:52:59 < dongs> return the content at address A instead of the content of address B.
2012-09-10T15:53:33 < Laurenceb> loool
2012-09-10T15:54:16 < dongs> and like 10 pages of i2C errata already
2012-09-10T15:54:20 < dongs> why dont they just fucking FIX it
2012-09-10T15:54:38 < dongs> i mean this is like the 4th chip revision of STM32, and nth if you count same i2c peripheral in STM8s
2012-09-10T15:54:42 < Laurenceb> lol
2012-09-10T15:55:10 < Laurenceb> "As of today, no compiler generates these particular instructions"
2012-09-10T15:55:14 < Laurenceb> so thats ok then
2012-09-10T15:55:16 < dongs> In Master mode, or in Slave mode when possible: configure the I2C with
2012-09-10T15:55:18 < dongs> NOSTRETCH=1 (clock stretching disable).
2012-09-10T15:55:18 < Laurenceb>  /sarc
2012-09-10T15:55:23 < dongs> fuck, same shit was in STM8 too
2012-09-10T15:55:53 < Laurenceb> "The likelihood of such condition to occur is rare"
2012-09-10T15:55:55 < Laurenceb> lolling
2012-09-10T15:56:09 < dongs> ya.
2012-09-10T15:56:15 < Laurenceb> this should be fine in out autopilot
2012-09-10T15:56:18 < Laurenceb> *our
2012-09-10T15:56:25 < dongs> the workaround for 0x4000 read bug is hilarious
2012-09-10T15:56:29 < dongs> "don't use more than 16k of ram"
2012-09-10T15:56:35 < dongs> and y ou'll be ok
2012-09-10T15:58:07 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T15:58:31 < dongs> hmm boot1 is not mentioned in datasheet
2012-09-10T15:59:07 < dongs> oh wtf
2012-09-10T15:59:10 < dongs> they moved that to a register
2012-09-10T15:59:14 < dongs> in option bytes
2012-09-10T15:59:18 < dongs> the pin.
2012-09-10T15:59:35 < dongs> After reset these bits take the value selected by the BOOT pin (BOOT0) and by the option
2012-09-10T15:59:38 < dongs> bite (nBOOT1).
2012-09-10T15:59:49 < dongs> someone was hungry.
2012-09-10T16:03:03 < dongs> ok well at least i dont have to worry about boot1 pin.
2012-09-10T16:03:51 < dongs> When using the small packages (48 and 64 pin packages), the GPIO pins which are not present on these packages, must
2012-09-10T16:04:10 < dongs> not be configured in analog mode.
2012-09-10T16:04:11 < dongs> uh...
2012-09-10T16:04:46 < dongs> no such limitation on f1
2012-09-10T16:15:09 < dongs> bloggery overload http://i.imgur.com/V46Hu.png
2012-09-10T16:23:26 < Laurenceb> /dev/sdc:
2012-09-10T16:23:26 < Laurenceb>  Timing buffered disk reads:    6 MB in  3.13 seconds =   1.92 MB/sec
2012-09-10T16:23:29 < Laurenceb> lulwut
2012-09-10T16:23:41 < Laurenceb> 12Mbps usb on F1
2012-09-10T16:23:49 < Laurenceb> 12/8==1.5MB/s
2012-09-10T16:24:45 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-10T16:24:46 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-10T16:24:46 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-10T16:24:49 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-10T16:25:50 < BrainDamage> 12Mbps is usb1.1
2012-09-10T16:27:50 < zyp> or 2.0
2012-09-10T16:28:35 < BrainDamage> I mean, it seems so spot on that maybe you got the mode wrong?
2012-09-10T16:28:44 < zyp> usb 2.0 added the highspeed spec, but also adopted LS and FS
2012-09-10T16:28:47 < zyp> heh
2012-09-10T16:28:52 < zyp> F1 only supports FS
2012-09-10T16:28:56 < zyp> as in 12 Mb/s
2012-09-10T16:29:36 < Laurenceb> looks like buffering
2012-09-10T16:29:51 < Laurenceb> /dev/sdc:
2012-09-10T16:29:51 < Laurenceb>  Timing buffered disk reads:    4 MB in  5.60 seconds = 731.51 kB/sec
2012-09-10T16:29:56 < Laurenceb> is about average
2012-09-10T16:30:06 < Laurenceb> fast enough for me
2012-09-10T16:30:33 < Laurenceb> theres lots of overhead - i worked out the fastest you can get is something like 820K
2012-09-10T16:30:49 < Laurenceb> 480Mbps adds longer packets - that helps a lot
2012-09-10T16:31:00 < zyp> not to mention higher signalling rate :p
2012-09-10T16:32:36 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 276 seconds]
2012-09-10T16:32:57 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-10T16:38:09 < Laurenceb> yeah but im getting like half the theoretical throughput
2012-09-10T16:40:45 < Laurenceb> 2012-9-10T13:41:32
2012-09-10T16:40:45 < Laurenceb> Battery: 4.174V
2012-09-10T16:40:45 < Laurenceb> Time, PPG(1), PPG(2)
2012-09-10T16:40:45 < Laurenceb> 0.000,313910,309577
2012-09-10T16:40:45 < Laurenceb> 0.016,313904,309570
2012-09-10T16:40:51 < Laurenceb> ^nice, everything works
2012-09-10T16:41:30 < Laurenceb> literally no more features to add :D
2012-09-10T16:41:44 < zyp> congratulations
2012-09-10T16:41:48 < Laurenceb> thanx
2012-09-10T16:41:59 < Laurenceb> well... apart from mass storage based bootloader
2012-09-10T16:45:09 < Laurenceb> grep seems to support multicore
2012-09-10T16:45:21 < Laurenceb> which is nice and fast.. but leaves my machine unusable
2012-09-10T16:49:32 < Laurenceb> http://www.flickr.com/photos/daveake/7970618830/in/set-72157631498431574/
2012-09-10T16:55:59 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-207.pptp.stw-bonn.de] has joined ##stm32
2012-09-10T16:58:54 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 264 seconds]
2012-09-10T17:02:50 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T17:05:59 < Laurenceb> one small issue
2012-09-10T17:06:11 < Laurenceb> logfile timestamps are in GMT
2012-09-10T17:06:15 < Laurenceb> not BST
2012-09-10T17:06:34 < Laurenceb> seems linux timestamps my files with gmt
2012-09-10T17:06:37 < zyp> UTC is the only thing that makes sense
2012-09-10T17:06:42 < Laurenceb> and the RTC grabs that
2012-09-10T17:07:13 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-09-10T17:07:15 < Laurenceb> but how come my local files on my pc are timestamped bst?
2012-09-10T17:07:27 < Laurenceb> nautilus corrects the timestamp to local?
2012-09-10T17:09:14 < zyp> something like that
2012-09-10T17:09:27 < zyp> it's the only sane way to use timezones
2012-09-10T17:09:50 < zyp> store all timestamps in UTC and add/subtract the necessary amount of hours when displaying it to a user
2012-09-10T17:10:54 < zyp> if you do it in any other way, you're in for a lot of bullshit when users move between timezones or in and out of DST
2012-09-10T17:11:17 < Laurenceb> hmm
2012-09-10T17:11:38 < Laurenceb> makes sense
2012-09-10T17:11:52 < zyp> I mean, you can of course store the timestamp in any timezone as long as you also store which
2012-09-10T17:12:38 < zyp> because then you know how to convert it to UTC, so you know how to convert it to anything else
2012-09-10T17:12:47 < Laurenceb> oh - if i browse the logger with nautilus its showing correct times
2012-09-10T17:12:56 < Laurenceb> i was grabbing RTC over jtag
2012-09-10T17:12:57 < zyp> but it's kind of pointless if you just store it in UTC in the first place
2012-09-10T17:12:57 < Laurenceb> cool
2012-09-10T17:13:36 < Laurenceb> RTC sets itself during connection using metadata from the files
2012-09-10T17:14:07 < zyp> is that precise?
2012-09-10T17:15:29 < zyp> I mean, how precise is that?
2012-09-10T17:17:45 < BrainDamage> I'd even go to the efforts to set the rtc to utc
2012-09-10T17:17:59 < BrainDamage> and instead calculate the offset only if you need to display
2012-09-10T17:18:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 276 seconds]
2012-09-10T17:18:10 < BrainDamage> this way all internal state is consistent
2012-09-10T17:18:10 < zyp> BrainDamage is a sane man
2012-09-10T17:18:44 < BrainDamage> I just don't like pointless headaches
2012-09-10T17:19:15 < Laurenceb> shrug
2012-09-10T17:19:18 < Laurenceb> its works atm
2012-09-10T17:19:28 < Laurenceb> only problem is it uses DST
2012-09-10T17:19:39 < Laurenceb> as code is stolen from mikrocontroller.net
2012-09-10T17:20:13 < Erlkoenig> yeay our nice german site rulez the µC world
2012-09-10T17:20:17 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T17:20:54 < BrainDamage> bah, predictive control theory is ridicolously boring
2012-09-10T17:21:18 < BrainDamage> not the concepts per se, but so many formalisms :s
2012-09-10T17:23:01 < Erlkoenig> need that for your studies?
2012-09-10T17:23:20 < BrainDamage> I have an exam about it Thursday
2012-09-10T17:24:11 < Erlkoenig> sounds like fun. what's your study subject?
2012-09-10T17:24:54 < BrainDamage> EE
2012-09-10T17:25:36 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-10T17:25:39 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-10T17:25:43 < BrainDamage> I should've picked physics instead in hindsight, but it's easy to be savy in retrspective ...
2012-09-10T17:29:29 < Laurenceb> heh
2012-09-10T17:29:35 < Laurenceb>   <- physics grad
2012-09-10T17:30:05 < Laurenceb> whats the size of F1 flash pages?
2012-09-10T17:40:07 <+izua> is there some easily parsable file for the registers and bits in F1 families, or at least f103? I guess something easier than the ton of includes in cmsis or cm3
2012-09-10T17:41:56 -!- DaKu_ is now known as daku
2012-09-10T17:42:05 -!- daku is now known as DaKu
2012-09-10T17:42:52 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-10T17:48:09 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 260 seconds]
2012-09-10T17:49:21 < Erlkoenig> izua: at least these defines/includes could be parsed by some standard C parser...
2012-09-10T17:52:35 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T17:54:58 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Remote host closed the connection]
2012-09-10T17:55:36 -!- claude_work is now known as claude
2012-09-10T17:57:48 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-10T17:58:40 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-10T18:01:06 < Laurenceb> is there a flash pagesize defined somewhere in the device  header  ?
2012-09-10T18:13:34 < Laurenceb> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/DispForm.aspx?ID=18579&RootFolder=https%3a%2f%2fmy%2est%2ecom%2fpublic%2fSTe2ecommunities%2fmcu%2fLists%2fcortex%5fmx%5fstm32%2fSD%20Card%20Bootloader%2c%20FAT%20FS%20Filesystem%2c%20USB%20Mass%20Storage&Source=https%3A%2F%2Fmy%2Est%2Ecom%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex%5Fmx%5Fstm32%2FFlat%2Easpx%3FRootFolder%3Dhttps%253a%252f%252fmy%252est%252ecom%252fp
2012-09-10T18:13:34 < Laurenceb> ublic%252fSTe2ecommunities%252fmcu%252fLists%252fcortex%255fmx%255fstm32%252fSD%2520Card%2520Bootloader%252c%2520FAT%2520FS%2520Filesystem%252c%2520USB%2520Mass%2520Storage%26FolderCTID%3D0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B%26currentviews%3D1119
2012-09-10T18:13:37 < Laurenceb> eeek
2012-09-10T18:14:25 <+Steffanx> Yeah, for sure
2012-09-10T18:16:01 < Erlkoenig> once upon a time, links looked like http://example.com/Pages/About.html
2012-09-10T18:16:44 <+Steffanx> The good old times
2012-09-10T18:16:46 < karlp> in many places, they still do.
2012-09-10T18:16:48 < Laurenceb> http://tinyurl.com/d2v9gc6
2012-09-10T18:16:58 < karlp> fucking tiny url,
2012-09-10T18:17:08 < karlp> just experiment with leaving out parameters until you get to the minium
2012-09-10T18:17:21 < karlp> the first line that yougot through was enough to get to the post by itself
2012-09-10T18:17:28 < Laurenceb> anyways, that looks perfect for my logger
2012-09-10T18:17:38 < karlp> ie, https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/DispForm.aspx?ID=18579
2012-09-10T18:17:41 < karlp> works just fine.
2012-09-10T18:19:07 < Erlkoenig> "asp" nooo
2012-09-10T18:19:20 <+Steffanx> + x
2012-09-10T18:20:34 < Erlkoenig> well the software is called "ASP" ;)
2012-09-10T18:20:46 < karlp> no, it's called ASP.Net :)
2012-09-10T18:20:48 < Laurenceb> aspergers spastic page
2012-09-10T18:20:52 <+Steffanx> And maybe it's fake Erlkoenig :P
2012-09-10T18:20:58 < karlp> ASP is the older one, that was just .asp, not .aspx
2012-09-10T18:21:06 < Erlkoenig> Steffanx: that would be funny :D
2012-09-10T18:21:10 < karlp> Steffanx: hehe
2012-09-10T18:21:37 <+Steffanx> Everything is better than something like page.dll?morecrap
2012-09-10T18:22:36 < Erlkoenig> even foo.php? :D
2012-09-10T18:23:16 <+izua> Steffanx: even about.exe ?
2012-09-10T18:23:40 <+Steffanx> Yeah.. people and their php phobia Erlkoenig :)
2012-09-10T18:23:51 <+Steffanx>  df
2012-09-10T18:23:57 < Erlkoenig> php phobia has a reason :D
2012-09-10T18:24:02 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-10T18:24:19 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-10T18:24:22 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-10T18:29:09 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-10T18:29:41 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-10T18:29:42 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-10T18:31:53 < Laurenceb> grr it wont compile
2012-09-10T18:32:05 < Laurenceb> Libraries/FATFS/ff.c:94:21: fatal error: fattime.h: No such file or directory
2012-09-10T18:32:06 < Laurenceb> wut
2012-09-10T18:32:26 < Erlkoenig> -Ipath/containing/fattime/
2012-09-10T18:35:43 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-10T18:38:32 < Laurenceb> seems to be some write IAR project
2012-09-10T18:38:36 < Laurenceb> *weird
2012-09-10T18:39:35 < jpa-> i wonder what crazy version of fatfs you have
2012-09-10T18:39:43 < jpa-> usually get_fattime is just a function in the usual glue
2012-09-10T18:39:59 < jpa-> (and it returns the current timestamp, for the purpose of setting it in modified files)
2012-09-10T18:41:13 < Laurenceb> ok wtf
2012-09-10T18:41:16 < Laurenceb> im confused now
2012-09-10T18:41:28 -!- dekar [~dekar@dslb-188-107-196-119.pools.arcor-ip.net] has quit [Ping timeout: 244 seconds]
2012-09-10T18:41:29 < Laurenceb> src/flash.c:36:95: error: 'FLASH_FLAG_BANK1_PGERR' undeclared (first use in this function)
2012-09-10T18:41:46 < Laurenceb> flash.h includes #include "stm32f10x_flash.h"
2012-09-10T18:41:50 < Laurenceb> where that is defined
2012-09-10T18:43:08 < Erlkoenig> probably a not satisfied #ifdef around the "#define FLASH_FLAG..."
2012-09-10T18:43:13 < Erlkoenig> like #define YOUR_CONTROLLER_NAME :D
2012-09-10T18:43:49 < jpa-> should be satisfied, there is #else
2012-09-10T18:44:46 < jpa-> Laurenceb: do you include flash.h in flash.c? :)
2012-09-10T18:44:50 < Laurenceb> yes
2012-09-10T18:45:29 < Laurenceb> jpa-: its the project i linked
2012-09-10T18:45:33 < Laurenceb> trying to compile it
2012-09-10T18:50:41 < Laurenceb> oh
2012-09-10T18:50:45 < Laurenceb> different periph lib
2012-09-10T18:50:46 < Laurenceb> wtf
2012-09-10T19:11:21 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-10T19:18:40 -!- dekar [~dekar@212.255.31.74] has joined ##stm32
2012-09-10T19:18:42 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-10T19:21:11 < Laurenceb> making slow progress....
2012-09-10T19:21:19 < Laurenceb> where the hell is __IO defined?
2012-09-10T19:23:47 < karlp> in the cmsis stuff isn't it?
2012-09-10T19:24:01 < karlp> can't you just press ctrl-click to get to it?
2012-09-10T19:25:28 < karlp> grep says core_cmX.h
2012-09-10T19:26:12 < karlp> it only resolves to volatile anyway.
2012-09-10T19:32:07 <+dekar> I need someone to test my multilib on armv7e-m with fpu: https://github.com/EliasOenal/TNT
2012-09-10T19:35:23 < Laurenceb> wow what a mess
2012-09-10T19:36:02 <+Steffanx> What is Laurenceb referring too?
2012-09-10T19:36:12 <+dekar> multilib I assume
2012-09-10T19:36:13 < Laurenceb> this IAr project
2012-09-10T19:36:20 <+dekar> oh well :P
2012-09-10T19:36:57 <+Steffanx> Too many spaces Laurenceb ? :P
2012-09-10T19:37:07 < zyp> hmm
2012-09-10T19:37:22 < Laurenceb> got it to compile
2012-09-10T19:37:27 < Laurenceb> just wont link now
2012-09-10T19:37:39 < Laurenceb> main.c:(.text+0x252): undefined reference to `__set_MSP'
2012-09-10T19:38:11 <+dekar> Laurenceb, just link it
2012-09-10T19:38:16 <+dekar> that's cmsis stuff
2012-09-10T19:39:12 < Laurenceb> hmm
2012-09-10T19:39:13 < Laurenceb> -I Libraries/CMSIS/CM3/CoreSupport/
2012-09-10T19:39:37 <+dekar> Laurenceb, that gets you the headers
2012-09-10T19:40:01 < Laurenceb> right
2012-09-10T19:40:07 < Laurenceb> what am i missing?
2012-09-10T19:40:57 < Laurenceb> oh i see
2012-09-10T19:40:59 < Laurenceb> ok..
2012-09-10T19:41:04 < Laurenceb> just need assert now
2012-09-10T19:41:31 <+dekar> core_cm3.c?
2012-09-10T19:41:40 < Laurenceb> yes
2012-09-10T19:42:15 <+dekar> I just links everything, the linker removes the stuff I don't need anyway :)
2012-09-10T19:42:37 <+dekar> *I just link
2012-09-10T19:43:36 < Laurenceb> ill need to impliment an assert
2012-09-10T19:43:38 < Laurenceb> .. tomorrow
2012-09-10T19:43:46 < karlp> -DNDEBUG?
2012-09-10T19:51:34 < zyp> http://paste.jvnv.net/view/VPo3I <- I wrote a class for fast logging to a ringbuffer
2012-09-10T19:51:43 < zyp> http://paste.jvnv.net/view/PtEBE <- and a gdb function to pretty-print it
2012-09-10T19:54:59 < zyp> I needed a way to log shit fast without introducing a delay to print each line
2012-09-10T20:01:03 < jpa-> zyp: thansk; i have wanted something like this for a while but never bothered to write it
2012-09-10T20:01:30 < zyp> :)
2012-09-10T20:02:11 < jpa-> mainly for the ability to check log through gdb
2012-09-10T20:02:26 < zyp> yeah
2012-09-10T20:02:41 < jpa-> for the speed issue, i've just used queues as I'm running with an RTOS anyway
2012-09-10T20:03:06 < zyp> I just started on the gdb stuff, it looks like this so far: http://paste.jvnv.net/view/82OVv
2012-09-10T20:03:12 < zyp> I'm not handling wraparound yet
2012-09-10T20:05:10 < zyp> queues how?
2012-09-10T20:06:19 < zyp> I don't want to have to deal with any formatting or copying strings at all when logging, just put a string pointer into a buffer
2012-09-10T20:06:24 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 260 seconds]
2012-09-10T20:06:42 < zyp> ideally it should be fast enough to call from an ISR
2012-09-10T20:07:00 -!- barthess [~barthess@178.154.62.0] has joined ##stm32
2012-09-10T20:07:11 < zyp> I'm also thinking about making it reentrant by doing an atomic update of the index when logging
2012-09-10T20:07:22 < zyp> I mean, thread/ISR safe
2012-09-10T20:08:38 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-10T20:08:58 < jpa-> well queue is just a ring buffer anyway, the point was that when using RTOS one is already provided, but not readable from gdb
2012-09-10T20:09:18 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-10T20:17:04 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-207.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-10T20:20:06 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-10T20:58:37 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-10T21:05:50 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-10T21:12:10 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-10T21:12:30 < TitanMKD> hello
2012-09-10T21:13:13 < zyp> hi
2012-09-10T21:14:04 < TitanMKD> anyone know an ARM Thumb2 instruction to do long jump (32bits) ?
2012-09-10T21:14:12 < TitanMKD> IIRC it is not possible and max is 24bits
2012-09-10T21:14:50 < zyp> yep
2012-09-10T21:14:51 < zyp> see http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0285b/Cchhgfaj.html
2012-09-10T21:15:48 < zyp> anyway, the trick is to put the addr into a register and jump to that
2012-09-10T21:16:15 < TitanMKD> yes and my problem is to do not use any register ;)
2012-09-10T21:16:28 < zyp> not possible
2012-09-10T21:16:47 < TitanMKD> the idea is to build Exception Vector table in flash 0x00000000 by default and jump to user code in RAM 0x08000000
2012-09-10T21:17:04 < TitanMKD> but because it's exception it shall not modify any register
2012-09-10T21:17:36 < TitanMKD> so it seems just impossible :(
2012-09-10T21:17:55 < zyp> why not? half the registers are already saved to stack at the time the ISR is called?
2012-09-10T21:18:07 < TitanMKD> and the worst is there's no remap hint on the ARM CPU I use to remap addr 0 to 0x08000000
2012-09-10T21:18:25 < zyp> what do you mean?
2012-09-10T21:18:46 < TitanMKD> yes the hint we found on Mx MCU to remap the addr 0 to anywhere ;)
2012-09-10T21:19:06 < TitanMKD> very useful to load code in RAM and remap all to ram
2012-09-10T21:20:01 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-10T21:20:02 < zyp> are you talking about NVIC.VTOR?
2012-09-10T21:20:17 < zyp> or what?
2012-09-10T21:20:18 <+dekar> TitanMKD, the stm32 has a pin to map either flash or ram to 0 iirc
2012-09-10T21:20:32 < zyp> I'm not sure exactly what the problem you're trying to solve is
2012-09-10T21:20:37 < TitanMKD> dekar the problem is my target is not a stm32 and it is not a Mx ;)
2012-09-10T21:21:02 <+dekar> wrong channel :P j/k
2012-09-10T21:21:03 < TitanMKD> zyp i just try to let run test code in RAM including Exception Vector
2012-09-10T21:21:16 < TitanMKD> instead of flashing always the internal flash memory
2012-09-10T21:21:33 < zippe> TitanMKD: Are you trying to do programmable vectors?
2012-09-10T21:21:38 < TitanMKD> zippe yes
2012-09-10T21:22:02 < zippe> TitanMKD: so as zyp said, put your vector table in RAM and use the VTOR register in the NVIC
2012-09-10T21:22:09 < TitanMKD> to load it one time in flash and jump to RAM at fixed addr ;)
2012-09-10T21:22:28 < TitanMKD> zippe on my AM CPU there's no VTOR ;)
2012-09-10T21:22:29 <+dekar> nor is nearly indestructible though, I have written my stm32s flash about a million times without failure
2012-09-10T21:22:34 < zippe> TitanMKD: which CPU?
2012-09-10T21:22:40 < TitanMKD> R4/R5
2012-09-10T21:22:46 < TitanMKD> it's CPU
2012-09-10T21:22:48 < zippe> TitanMKD: Get a refund
2012-09-10T21:22:58 < TitanMKD> and they have only fixed addr for Exception Addr
2012-09-10T21:23:30 < TitanMKD> of courser peripherals addr specific to each CPU/MCU is not fixed to any addr by i speak about Exception Handler
2012-09-10T21:23:58 < TitanMKD> in worst case i will trash register saved by the exception ;)
2012-09-10T21:24:09 < zippe> You'll have to do it in code; you can for example use bx <handler> and decode the LR to determine the vector number
2012-09-10T21:24:31 < TitanMKD> yes bx rx
2012-09-10T21:24:35 < zippe> you should have lots of spare registers courtesy of the exception stacking the caller-saved registers
2012-09-10T21:24:35 < TitanMKD> it's 32bits jump
2012-09-10T21:24:54 < zyp> so, this is cortex-r?
2012-09-10T21:25:01 < TitanMKD> zyp yes
2012-09-10T21:25:10 < zippe> Well, it depends on whether you need to handle anything in the ROM, or if the ROM is just flashed to support run-in-RAM
2012-09-10T21:25:27 < zippe> if you are just flashing to support run-in-RAM, then you can basically do
2012-09-10T21:25:34 < zippe> mov r0, pc
2012-09-10T21:25:56 < zippe> add r0, r0, <rom-to-RAM-offset>
2012-09-10T21:26:00 < zippe> mov pc, r0
2012-09-10T21:26:11 < zippe> in each of your flash-based exception vectors
2012-09-10T21:26:42 < TitanMKD> yes but i was searching a way to do not use any reg ;)
2012-09-10T21:26:55 < zippe> Why? you have a bunch of free registers...
2012-09-10T21:27:20 < zyp> are the vectors on cortex-r not 32-bit pointers like on cortex-m?
2012-09-10T21:27:57 < TitanMKD> in fact it support also full ARM instruction
2012-09-10T21:28:01 < TitanMKD> and Thumb2
2012-09-10T21:28:35 < TitanMKD> in fact there's not so much bunch of banked register ;)
2012-09-10T21:29:54 < TitanMKD> for my use i will really prefer to disable those banked registers ;)
2012-09-10T21:30:43 < TitanMKD> especially between supervisor & user mode
2012-09-10T21:38:34 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 260 seconds]
2012-09-10T21:51:05 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-09-10T21:51:30 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has joined ##stm32
2012-09-10T21:51:31 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has quit [Changing host]
2012-09-10T21:51:31 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-10T21:52:52 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-10T22:09:17 -!- barthess [~barthess@178.154.62.0] has quit [Quit: Leaving.]
2012-09-10T22:11:49 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-10T22:15:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-10T22:15:43 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-10T22:22:47 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-10T22:50:55 < Laurenceb_> text	   data	    bss	    dec	    hex	filename
2012-09-10T22:50:55 < Laurenceb_>   35920	    368	   1360	  37648	   9310	main.elf
2012-09-10T22:50:56 < Laurenceb_> ewww
2012-09-10T22:51:08 < Laurenceb_> finally got iar project to compile
2012-09-10T22:51:25 < Laurenceb_> 67% larger
2012-09-10T22:59:59 < zyp> hmm, I'm wondering if I should start working on adding support for the F1 USB driver
2012-09-10T23:00:36 < zyp> I think I need a break from the USB core in the F4 :p
2012-09-10T23:02:15 < zyp> and having two underlying drivers would probably help making a good abstraction on top of them
2012-09-10T23:02:37 < Laurenceb_> does building the periph lib as an asm library and linking use more flash?
2012-09-10T23:02:54 < zyp> huh?
2012-09-10T23:03:54 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-10T23:04:09 < zyp> if I were to guess, you are getting a 67% larger binary because you're now not stripping unused functions
2012-09-10T23:04:15 < Laurenceb_> doh
2012-09-10T23:04:33 < Laurenceb_> im compiling periph lib with a separate makefile
2012-09-10T23:04:36 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-10T23:04:36 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-10T23:04:52 < Laurenceb_> then creating libstm32.a and linking to it
2012-09-10T23:05:03 < zyp> so what? makefiles are just automation
2012-09-10T23:05:33 < Laurenceb_> is this going to include non essential junk?
2012-09-10T23:06:03 < zyp> how would I know?
2012-09-10T23:06:11 < zyp> I'm just making a guess
2012-09-10T23:06:37 < Laurenceb_> ok
2012-09-10T23:07:52 < zyp> I'm just thinking that the most probable reason for a huge gain in code size is that you're not throwing away code for unused functions
2012-09-10T23:07:56 < Laurenceb_> yet libstm32.a is 73KB
2012-09-10T23:08:02 < Laurenceb_> yeah
2012-09-10T23:08:18 < zyp> that's irrelevant, it's an archive of elf files
2012-09-10T23:08:27 < TitanMKD> or maybe there's debug in the lib too
2012-09-10T23:08:44 < TitanMKD> so the final exe will be maybe not so huge
2012-09-10T23:08:53 < zyp> debug info doesn't affect code size, that's just metadata in the elf
2012-09-10T23:09:02 < zyp> and that's why the size of the .a is irrelevant
2012-09-10T23:09:11 < zyp> same goes for size of the final .elf
2012-09-10T23:09:14 < TitanMKD> it's why i say it will not be so huge at end ;)
2012-09-10T23:09:19 < zyp> it's the size of the sections that matter
2012-09-10T23:09:23 < Laurenceb_> ill try changing the makefile
2012-09-10T23:09:33 < zyp> and I'm assuming that's what is 63% larger
2012-09-10T23:09:39 < zyp> sorry, 67%
2012-09-10T23:10:01 < TitanMKD> also check the lib is built with -O2 and not -O0 ;)
2012-09-10T23:12:13 < Laurenceb_> oh yeah smaller now
2012-09-10T23:12:22 < Laurenceb_> i might be able to get it down to 32KB
2012-09-10T23:12:29 < Laurenceb_> it was 22 before
2012-09-10T23:12:34 < Laurenceb_> with IAR
2012-09-10T23:12:58 < zyp> is this a gcc based toolchain?
2012-09-10T23:13:04 < Laurenceb_> yes
2012-09-10T23:13:16 < zyp> which CFLAGS and LDFLAGS are you using?
2012-09-10T23:13:58 < zyp> you want -ffunction-sections in CFLAGS and -Wl,--gc-sections in LDFLAGS
2012-09-10T23:14:08 < zyp> those two are responsible for removing unused code
2012-09-10T23:14:16 < karlp> datasections too...
2012-09-10T23:14:17 < Laurenceb_> -mcpu=cortex-m3 -mthumb -std=gnu99
2012-09-10T23:15:40 < Laurenceb_> HOLY SHIT
2012-09-10T23:15:44 < Laurenceb_> 18KB
2012-09-10T23:16:04 < Laurenceb_> and ive still got the screwed usb library code
2012-09-10T23:16:06 < TitanMKD> also -fomit-frame-pointer
2012-09-10T23:16:07 < TitanMKD> ;)
2012-09-10T23:16:28 < karlp> (onlyif you don't want the frame pointer ;)
2012-09-10T23:16:33 < zyp> Laurenceb_, also, double check that it still works :p
2012-09-10T23:16:37 < Laurenceb_> IAR just got annihilated
2012-09-10T23:17:00 < Laurenceb_> dont have hardware here
2012-09-10T23:17:03 < karlp> (unless you "garbage collected" code you needed...)
2012-09-10T23:17:09 < Laurenceb_> ill check the dissasem when im done
2012-09-10T23:17:17 < Laurenceb_> *diss asm
2012-09-10T23:17:22 < zyp> karlp, I had some problems with that once
2012-09-10T23:17:46 < zyp> one of the functions in newlib's libm broke when I turned on garbage collection
2012-09-10T23:17:47  * karlp too
2012-09-10T23:17:58 < Laurenceb_> this is pretty simple code
2012-09-10T23:18:08 < TitanMKD> -ffunction-sections -fdata-sections -fno-common
2012-09-10T23:18:08 < Laurenceb_> just just fatfs with flash read and write
2012-09-10T23:18:09 < TitanMKD> ;)
2012-09-10T23:18:14 < zyp> I never tried to track it down, so I'm not sure what caused it
2012-09-10T23:18:16 < Laurenceb_> and the USB mass storage code
2012-09-10T23:18:32 < zyp> Laurenceb_, then 18k should be pretty reasonable
2012-09-10T23:19:04 < zyp> it's over twice as large as any of my stuff as of yet
2012-09-10T23:19:17 < Laurenceb_> checks card for "firmware.bin"
2012-09-10T23:19:27 < Laurenceb_> if its finds it it flashes then jumps to app code
2012-09-10T23:19:43 < Laurenceb_> if not it checks the app code space for something sane looking
2012-09-10T23:19:56 < Laurenceb_> if it doesnt find it it stays as a usb mass storage thingy
2012-09-10T23:20:23 < Laurenceb_> and it deleted firmware.bin after flashing
2012-09-10T23:21:45 < TitanMKD> Laurenceb seems great features
2012-09-10T23:21:51 < zyp> hmm, gc-sections is only making a 7% difference on my quadrotor project
2012-09-10T23:22:01 < TitanMKD> Laurenceb it will be even better if it could be a library ;)
2012-09-10T23:22:06 < zyp> it goes from 7.9 kB to 7.3
2012-09-10T23:22:36 < TitanMKD> zyp it can be dangerous for vector table ;)
2012-09-10T23:22:55 < TitanMKD> zyp sometimes it trash it because it does not see any code linked to it
2012-09-10T23:23:07 < zyp> it's referenced explicitly in my linker script, so that's fine
2012-09-10T23:23:13 < TitanMKD> yes ;)
2012-09-10T23:24:42 < Laurenceb_> might be able to get 16KB now
2012-09-10T23:25:58 < zyp> after all, the vector table is the root of the dependency tree of references, since it's the one that references the reset vector, which references main, which references everything else ;)
2012-09-10T23:28:06 < TitanMKD> also have you looked for the code
2012-09-10T23:28:07 < TitanMKD>     while (dp < &_edata)
2012-09-10T23:28:07 < TitanMKD>       *dp++ = *tp++;
2012-09-10T23:28:20 < TitanMKD> such code is not really beautifull in asm :(
2012-09-10T23:28:53 < TitanMKD> i was a bit disappointed by generated code with -O2 but i have not tested -O3
2012-09-10T23:29:03 < TitanMKD> in Thumb2 of course
2012-09-10T23:29:28 < TitanMKD> with gcc-arm-none-eabi-4_6-2012q1-20120316.exe
2012-09-10T23:29:30 < TitanMKD> from Linaro
2012-09-10T23:30:34 < Laurenceb_> 15.5KB
2012-09-10T23:30:53 < TitanMKD> Laurenceb have you a sum of all options used ?
2012-09-10T23:31:24 < Laurenceb_> http://pastebin.com/wDLXbh1k
2012-09-10T23:32:08 < TitanMKD> ha ok with -Os
2012-09-10T23:32:17 < TitanMKD> so it's smallest possible code
2012-09-10T23:33:02 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-10T23:33:26 < TitanMKD> anyway Thumb2 is really a must especially on CortexM3/M4 the code execution take 1cycle for pratically all instructions !!
2012-09-10T23:33:37 < TitanMKD> and for DSP instruction also ;)
2012-09-10T23:33:48 < TitanMKD> (except div stuff like always)
2012-09-10T23:34:17 < zyp> TitanMKD, here's how mine turns out: http://paste.jvnv.net/view/idcZV
2012-09-10T23:34:46 < zyp> the loop itself is the five last instructions, looks pretty efficient to me
2012-09-10T23:34:47 < karlp> bleh, finally fixed my 802.15.4 code.
2012-09-10T23:34:57 < karlp> completely missed that the interrupt polarity was wrong.
2012-09-10T23:35:16 < TitanMKD> zyp ha not bad ::
2012-09-10T23:35:28 < TitanMKD> zyp it was maybe my options which was bad
2012-09-10T23:35:37 < zyp> this is with -O2
2012-09-10T23:35:41 < TitanMKD> zyp you used which compiler & which options ?
2012-09-10T23:35:50 < zyp> gcc 4.7
2012-09-10T23:35:59 < TitanMKD> like mine linaro stuff
2012-09-10T23:36:04 < TitanMKD> maybe you have some patch ?
2012-09-10T23:36:14 < zyp> no idea, I'm using the yagarto build
2012-09-10T23:36:23 < karlp> 4.7 is the "make code smaller" branch...
2012-09-10T23:36:26 < TitanMKD> mine was locked by using only r0 and r1
2012-09-10T23:36:34 < karlp> see the questions/answers on gcc arm embedded,
2012-09-10T23:36:42 < karlp> and the reason that dekar loves his TNT so much
2012-09-10T23:36:43 < TitanMKD> you imagine the mess it was always reloading r0 & r1 at each loop
2012-09-10T23:36:46 < zyp> karlp, I'm more interested in the «add C++11-support» part of it
2012-09-10T23:36:57 < Laurenceb_> 08000000 <__cs3_interrupt_vector>:
2012-09-10T23:36:57 < Laurenceb_>  8000000:	00 50 00 20 21 37 00 08 31 37 00 08 31 37 00 08     .P. !7..17..17..
2012-09-10T23:36:58 < Laurenceb_>  8000010:	31 37 00 08 31 37 00 08 31 37 00 08 31 37 00 08     17..17..17..17..
2012-09-10T23:37:01 < Laurenceb_> ^looks sane?
2012-09-10T23:37:06 < zyp> yes
2012-09-10T23:37:17 < karlp> I was more in the "which one will actually work without me sacrificing 20 virgins" camp
2012-09-10T23:37:27 < Laurenceb_> why 0x02000500?
2012-09-10T23:37:32 < zyp> stack addr
2012-09-10T23:37:37 < Laurenceb_> oh as its stack of course
2012-09-10T23:37:40 < Laurenceb_> end of sram
2012-09-10T23:37:43 < zyp> yep
2012-09-10T23:37:57 < TitanMKD> zyp anyway it could be better code using IT instead of cmp ;)
2012-09-10T23:38:30 < TitanMKD> but i have never see usage of IT by compiler
2012-09-10T23:38:35 < zyp> could it really? even with a branch?
2012-09-10T23:38:44 < TitanMKD> ITNE
2012-09-10T23:39:02 < zyp> I'm pretty sure I've seen gcc emitting IT
2012-09-10T23:39:10 < zyp> let me search a bit
2012-09-10T23:39:16 < TitanMKD> anyway it is still pratically perfect ;)
2012-09-10T23:39:30 < TitanMKD> my loop was using like 10 instructions with same code ;)
2012-09-10T23:39:38 < zyp> yep
2012-09-10T23:39:54 < zyp> http://paste.jvnv.net/view/Mu9Dd
2012-09-10T23:40:15 < TitanMKD> hehe ;)
2012-09-10T23:40:23 < TitanMKD> i love the IT stuff
2012-09-10T23:40:30 < TitanMKD> it just cost 0 cycle ;)
2012-09-10T23:41:12 < TitanMKD> IIRC you can combine IT+an instruction for 1cycle
2012-09-10T23:41:15 < TitanMKD> in special case
2012-09-10T23:42:37 < zyp> hmm
2012-09-10T23:42:41 < zyp> http://paste.jvnv.net/view/LgeHo <- this looks fun
2012-09-10T23:43:02 < zyp> compare followed by conditional compare
2012-09-10T23:43:19 < TitanMKD> anyway maybe i was using Thumb ;)
2012-09-10T23:43:43 < TitanMKD> i will check it is using Thumb2 and not the old ugly/slow Thumb(1)
2012-09-10T23:44:16 < TitanMKD> zyp yes strange to understand ;)
2012-09-10T23:44:32 < zyp> nah, it's pretty easy
2012-09-10T23:44:53 < karlp> heh, my avr code I was copying wasn't using the right edge either,
2012-09-10T23:44:58 < TitanMKD> yes it just do not do check in case eq
2012-09-10T23:45:00 < zyp> if the first compare is equal, the second will be performed
2012-09-10T23:45:02 < TitanMKD> or ne ;)
2012-09-10T23:45:05 < karlp> but the default was for low level triggered, so it mostly worked
2012-09-10T23:45:58 < zyp> after the first three instructions, you'll have the equal flag set only if both comparisons worked out equal
2012-09-10T23:46:41 < zyp> and that makes sense, because the code emitting those instructions look like this: if(bmRequestType == 0xc0 && bRequest == 0xf2)
2012-09-10T23:47:45 < TitanMKD> for those interested by a very good PDF about STM32F4 M4 core
2012-09-10T23:47:46 < TitanMKD> see
2012-09-10T23:47:47 < TitanMKD> http://armvn-repo.googlecode.com/files/STM32F4-Technical-Training.pdf
2012-09-10T23:48:27 < TitanMKD> bye
2012-09-10T23:48:36 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-10T23:52:52 < Laurenceb_> lol i had interrupt name issues
2012-09-10T23:52:57 < Laurenceb_> guess what the size is now
2012-09-10T23:53:04 < Laurenceb_> 21.78KB
2012-09-10T23:53:10 < Laurenceb_> IAR is 21.78KB
--- Day changed Tue Sep 11 2012
2012-09-11T00:03:28 < Laurenceb_> what are the chances
2012-09-11T00:04:09 < zyp> pretty high if you are really just using two instances of the same gcc shipped under different names
2012-09-11T00:05:18 < Laurenceb_> previously it was with IAR c compiler
2012-09-11T00:08:07 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-11T00:08:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-11T00:09:03 < Laurenceb_> oh yeah main if IAR is actually GCC  :P
2012-09-11T00:10:46 < Laurenceb_> i have a friend who works for IAR, maybe i should ask him XD
2012-09-11T00:14:45 -!- blocky [blocky@S0106c0c1c07746fb.gv.shawcable.net] has quit [Read error: Connection reset by peer]
2012-09-11T00:31:13 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-11T01:46:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-11T01:46:50 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-11T01:49:37 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-09-11T01:50:33 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-11T01:54:04 -!- grummund [~user@unaffiliated/grummund] has quit [Read error: Operation timed out]
2012-09-11T02:09:14 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-09-11T02:21:09 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-11T02:26:09 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-09-11T02:31:44 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-11T02:35:48 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 276 seconds]
2012-09-11T02:49:17 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-11T02:50:21 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-11T02:52:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-11T03:19:27 < dongs> happy 9/11 bitches
2012-09-11T03:26:57 -!- DaKu is now known as daku
2012-09-11T03:31:40 <+izua> dongs: no rc planes and model buildings?
2012-09-11T04:04:37 < mrnuke> dongs: bithc please! It's still 9/10 in the US
2012-09-11T04:05:05 < dongs> mrnuke: its 9/11 anywhere taht matters, bro
2012-09-11T04:06:14 < mrnuke> dongs: so, the assholes with most nukes don;t matter?
2012-09-11T04:21:11 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 244 seconds]
2012-09-11T04:21:30 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-11T04:50:26 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-11T04:50:47 < dongs> http://web.ncsu.edu/abstract/science/wms-cockroach-steering/
2012-09-11T04:50:48 < dongs> lol
2012-09-11T05:00:45 <+izua> oook
2012-09-11T05:00:48 <+izua> that's pretty cool
2012-09-11T05:00:58 <+izua> you could make an army of those and have them invade your neighbour
2012-09-11T05:03:12 < dongs> http://www.youtube.com/watch?v=gmbEX7zDzog lol @ comments disabled
2012-09-11T05:03:17 < dongs> probably to avoid peta faggots
2012-09-11T05:03:18 < dongs> from whining
2012-09-11T05:04:15 <+izua> let's save the cockroaches
2012-09-11T05:04:32 <+izua> i wonder how much voltage it would take to trigger the poor guy
2012-09-11T05:04:39 <+izua> BIG LION TO YOUR LEFT HERP
2012-09-11T05:06:45 < dongs> next they just need to figure this shit out on birds, and all quadcopter faggots can pack up
2012-09-11T05:07:14 <+izua> it probably works on moths
2012-09-11T05:08:51 < dongs> I’m sure the Pentagon would also love these bionic critters for other functions. Oh, and they are doing the same with moths, so I’m sure there will be biological spies flying the skies of the world sooner than we imagine. Sometimes I wonder if these people have ever watched any horror sci-fi movie. You know, the ones in which bionic moths and cockroaches kill every human on Earth.
2012-09-11T05:08:57 < dongs> haha
2012-09-11T05:09:42 < dongs> ahaha and here we go I strongly disapprove with this kind of research and I ask you to reconsider the ethical meaning of what you are doing here. In my opinion, it is a manifestation of the human disrespectfulness, which is very harmful to this planet. You might say, well this is “only” a cockroach? It doesn’t matter. You are crossing a line, you are disrespectful towards Creation, to borrow a religous term here. Just because humans are superior in 
2012-09-11T05:12:08 <+izua> herp derp
2012-09-11T05:12:29 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-11T05:12:52 <+izua> dongs: here, more EE and ethics and paranormal and conspiracies: https://en.wikipedia.org/wiki/Global_Consciousness_Project
2012-09-11T05:13:07 < dongs> izua: saw that last time you linked it
2012-09-11T05:13:19 <+izua> oh lol
2012-09-11T05:13:34 <+izua> i wonder if they use a matrix font
2012-09-11T05:13:53 <+izua> and be like "i'm looking at the global consciousness"
2012-09-11T05:14:13 <+izua> "but all i see are blonds, redheads, no code at all"
2012-09-11T05:22:46 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-11T06:04:30 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-11T06:04:37 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-11T06:48:21 < dongs> why teh fuck there are no chats
2012-09-11T07:14:26 < mrnuke> dongs: SSSSSSSH! we're waiting on 9/11 to start
2012-09-11T07:16:16 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-11T07:24:08 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-11T07:24:11 < dongs> heh
2012-09-11T07:52:28 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 272 seconds]
2012-09-11T07:56:42 < dongs> http://www.koreatimes.co.kr/www/news/nation/2012/09/113_119626.html
2012-09-11T08:11:04 < dongs> http://i.imgur.com/FvNaE.jpg
2012-09-11T09:09:57 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-11T09:38:30 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-11T09:53:48 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 240 seconds]
2012-09-11T10:06:32 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-11T10:28:07 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-11T10:28:07 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-11T10:28:07 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-11T10:33:39 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-11T10:33:43 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T10:36:13 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-11T10:36:59 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-11T10:37:02 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T10:38:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-11T10:41:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 252 seconds]
2012-09-11T10:44:39 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-11T10:56:10 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-11T11:03:27 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-11T11:33:04 < Oldboy> dongs: where are you from before japan ?
2012-09-11T11:33:39 < dongs> israel, why do you ask
2012-09-11T11:34:11 < Oldboy> just curious. I thought you were an American living in Japan.
2012-09-11T11:54:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-11T11:56:49 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-11T12:03:39 < tavish> I am having a wierd problem with usb host for msc, when i write to a file, the first two and the last bytes are always \00. Any idea where to look for the problem?
2012-09-11T12:07:06 < dongs> im sure Laurenceb would know
2012-09-11T12:12:48 < zyp> tavish, host, as in you have a memory stick connected to the stm32?
2012-09-11T12:13:19 < zyp> and you're using fatfs to access it or what?
2012-09-11T12:17:34 < tavish> zyp: yeah, pen drive attached to otg_fs core and using fatfs
2012-09-11T12:19:02 < zyp> if you are getting corrupt data in files but an otherwise healthy file system I would expect that the problem is unrelated to usb
2012-09-11T12:19:57 < zyp> tried running on a ramdisk or some other kind of block device to see if the same problems are still present?
2012-09-11T12:22:20 < tavish> no, i just tried on a single pen drive
2012-09-11T12:22:43 < tavish> i will try block reads/writes then
2012-09-11T12:30:40 < Laurenceb> http://www.theregister.co.uk/2012/09/11/hab_ashes/
2012-09-11T12:30:46 < Laurenceb> lurn 2 haircut
2012-09-11T12:31:11 < Laurenceb> tavish: you have hosting a file system?
2012-09-11T12:31:50 -!- tech2077 [~tech2077@adsl-75-53-129-61.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection timed out]
2012-09-11T12:32:35 < tavish> Laurenceb: yes, i was testing a simple file write
2012-09-11T12:34:03 < Laurenceb> ah
2012-09-11T12:34:21 < Laurenceb> im using an F1 as a mass storage device
2012-09-11T12:34:27 < Laurenceb> im not hosting anything
2012-09-11T12:36:20 < tavish> i was doing that earlier with f4, with microsd. it didnt work reliably so i abandoned it
2012-09-11T12:37:02 < tavish> for now; may have to try IAR, they have their own sdio code.
2012-09-11T12:37:07 < Laurenceb> heh
2012-09-11T12:37:16 < Laurenceb> last night i ported IAR code to GCC
2012-09-11T12:37:25 < Laurenceb> code size was ~identical
2012-09-11T12:42:18 < Laurenceb> 21.7KB with about 40 bytes between them
2012-09-11T12:42:22 < Laurenceb> gcc slightyl larger
2012-09-11T12:44:21 < tavish> Laurenceb: were you having problems with the st library as well?
2012-09-11T12:44:50 < Laurenceb> no
2012-09-11T12:46:45 < Laurenceb> it was the fatfs spi mmc code i was using for lowlevel card interfacing
2012-09-11T12:56:21 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has joined ##stm32
2012-09-11T12:58:16 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-11T12:58:16 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-11T12:58:16 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-11T12:58:19 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-11T13:00:54 < Laurenceb> cant help you with host, sorry
2012-09-11T13:03:16  * Laurenceb worked out a fairly unbrickable way to do mass storage bootloader
2012-09-11T13:03:37 < Laurenceb> problem is that you need a way to mount the card and upload
2012-09-11T13:03:51 < Laurenceb> which is handled by the application code
2012-09-11T13:04:41 < Laurenceb> so im setting IWDG in the bootloader, before jumping, and then if the bootloader detects IWDG as the reset cause it will try to mount the device itself
2012-09-11T13:04:52 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-11T13:14:45 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-11T13:18:31 < Laurenceb> looks like it will fit into 24KB
2012-09-11T13:19:04 < zyp> that's a nice small bootloader
2012-09-11T13:22:26 < Laurenceb> heh
2012-09-11T13:22:42 < Laurenceb> it has to have all the USB drivers
2012-09-11T13:23:03 < Laurenceb> but ive got 128KB of flash so *shrug*
2012-09-11T13:23:18 < Laurenceb> about 10KB spare
2012-09-11T13:27:45 < tavish> zyp: block read/writes seem to be working
2012-09-11T13:37:41 < karlp> only 50% bigger than my application :)
2012-09-11T13:42:46 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-11T13:46:04 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-11T13:47:15 < zyp> tavish, as in directly with usb without fatfs?
2012-09-11T13:49:12 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 276 seconds]
2012-09-11T13:50:24 -!- tavish [~tavish@120.59.32.232] has joined ##stm32
2012-09-11T13:50:48 -!- tavish is now known as Guest69994
2012-09-11T13:51:35 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 255 seconds]
2012-09-11T13:54:50 -!- Guest69994 is now known as tavish3
2012-09-11T13:59:04 < Laurenceb> 23.86KB
2012-09-11T13:59:06 < Laurenceb> nice
2012-09-11T14:01:41 < Laurenceb> 104KB of code space to use
2012-09-11T14:01:53 < dongs> blogspace
2012-09-11T14:02:11 < Laurenceb> indeed
2012-09-11T14:03:12 < Laurenceb> http://pastebin.com/BYS2Xmqv
2012-09-11T14:04:13 < dongs> wat
2012-09-11T14:04:56 < zyp> dongs, you would love my new r-blog class
2012-09-11T14:05:12 < Laurenceb> ^bootloader
2012-09-11T14:05:17 < Laurenceb> its nice and simple
2012-09-11T14:05:35 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-11T14:05:59 < dongs> huh wheres the acutal bootloader code
2012-09-11T14:06:01 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-11T14:06:13 < Laurenceb> lol
2012-09-11T14:06:21 < dongs> oh, its part of your messy shit
2012-09-11T14:06:24 < dongs> you can tell teh parts you copypasted
2012-09-11T14:06:25 < Laurenceb> i stole it from the IAR project
2012-09-11T14:06:29 < Laurenceb> heh
2012-09-11T14:06:32 < BrainDamage> how can you tell it's not his code? there's spaces in it
2012-09-11T14:06:38 < dongs> yea
2012-09-11T14:07:41 < Laurenceb> http://pastebin.com/Ne2D4KFV
2012-09-11T14:08:06 < jpa-> sEE_FAIL   i see fail
2012-09-11T14:08:36 -!- Steffanx [~Steffanx@client-221-023.flexnet2.rug.nl] has joined ##stm32
2012-09-11T14:08:36 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T14:08:39 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Ping timeout: 244 seconds]
2012-09-11T14:08:42 < Laurenceb> G0AT_sEE
2012-09-11T14:10:18 < Laurenceb> http://www.flickr.com/photos/acudworth/6854390424/in/photostream
2012-09-11T14:10:29 < Laurenceb> otherwise someone might forget where the pins are
2012-09-11T14:11:05 < BrainDamage> for breadboarding, it's not an awful idea
2012-09-11T14:11:11 < BrainDamage> for a pcb ...
2012-09-11T14:13:26 < dongs> you can tell that trash was made in eagle
2012-09-11T14:14:17 < dongs> Laurenceb: you gotta stop using this avr trash
2012-09-11T14:14:52 < BrainDamage> that's not his photostream by the severe lack of naked kids pics
2012-09-11T14:16:27 -!- Steffanx [~Steffanx@client-221-023.flexnet2.rug.nl] has quit [Quit: Ik ga weg]
2012-09-11T14:23:09 < qyx_> lol that label with pinout
2012-09-11T14:23:51 < dongs> tardushit pinout
2012-09-11T14:23:54 < dongs> the most filthy one
2012-09-11T14:25:40 < Laurenceb> BrainDamage: you can talk
2012-09-11T14:26:00 < BrainDamage> ?
2012-09-11T14:26:45 < Laurenceb> " severe lack of naked kids pics"
2012-09-11T14:27:13 < Laurenceb> BrainDamage sounds dissapointed
2012-09-11T14:27:18 -!- jaeckel [~jaeckel@jaeckel.eu] has joined ##stm32
2012-09-11T14:27:18 -!- jaeckel [~jaeckel@jaeckel.eu] has quit [Changing host]
2012-09-11T14:27:18 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-11T14:27:48 < BrainDamage> not really
2012-09-11T14:27:57 < Laurenceb> http://www.flickr.com/photos/acudworth/6813808225/in/set-72157627177202669
2012-09-11T14:28:24 < Laurenceb> ^ can i ship that to customers?
2012-09-11T14:28:36 < tavish3> >:(
2012-09-11T14:29:05 < BrainDamage> looks too solid, you need to replace the perfboard with a breadboard, and mount at least 2 shields on top
2012-09-11T14:29:41 < tavish3> is it a arduino boat?
2012-09-11T14:29:45 < tavish3> *an
2012-09-11T14:30:09 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 276 seconds]
2012-09-11T14:30:57 < BrainDamage> arduino high altitude balloon gps tracker wth radio from the guys in #highaltitude
2012-09-11T14:31:13 < BrainDamage> tbh for what they do, the arduino is correctly sized
2012-09-11T14:31:27 < BrainDamage> if only they'd use the chip instead of the whole board ...
2012-09-11T14:31:50 < nopcode> i dont get the arduino craze
2012-09-11T14:31:55 < nopcode> avrs are pretty much self-sufficient
2012-09-11T14:32:02 < nopcode> they need close to zero external peripherals
2012-09-11T14:32:34 < tavish3> its the IDE and the examples/libraries
2012-09-11T14:32:40 < tavish3> prolly
2012-09-11T14:32:47 < BrainDamage> the IDE is not an ide
2012-09-11T14:32:48 < Laurenceb> yet avr studio6 pwns
2012-09-11T14:32:50 < BrainDamage> it's a text editor
2012-09-11T14:33:03 < BrainDamage> actually, it's even a bad text editor...
2012-09-11T14:33:35 < dongs> hes talking about arduino 'IDE'
2012-09-11T14:33:36 < tavish3> yeah it is
2012-09-11T14:34:20 < Laurenceb> IFE
2012-09-11T14:34:30 < dongs> Laurenceb: you will command all y our #highturd guys to switch to my stm32 shit immediately http://i.imgur.com/L29UB.png
2012-09-11T14:34:32 < Laurenceb> Integrated Fail Environment
2012-09-11T14:34:43 < tavish3> but its 5 seconds to get an example to work. that must be what people like. open box - push button - dance
2012-09-11T14:34:49 < dongs> its same width as t arduino pro mini
2012-09-11T14:34:50 < Laurenceb> hahaha
2012-09-11T14:35:31 < Laurenceb> thats brilliant
2012-09-11T14:35:38 < Laurenceb> perfect for trolling
2012-09-11T14:35:46 < dongs> yeah
2012-09-11T14:35:52 < dongs> I linked it to the girl from coocox
2012-09-11T14:35:55 < dongs> she was like "omg thats so cute"
2012-09-11T14:36:48 < Laurenceb> where can you buy?
2012-09-11T14:37:11 < Laurenceb> girls?! impossible
2012-09-11T14:37:16 < dongs> i kno rite
2012-09-11T14:37:30 < dongs> she's fat, for a chink though
2012-09-11T14:37:56 < Laurenceb> is there a coocox irc?
2012-09-11T14:38:00 < dongs> nah
2012-09-11T14:38:03 < Laurenceb> there no chibios irc :(
2012-09-11T14:38:18 < dongs> chibios irc is proabably #animu4fags
2012-09-11T14:38:43 < Laurenceb> is that board 4 sale?
2012-09-11T14:39:06 < dongs> yeah, probably once the test batch works this week I'll probably make some panels
2012-09-11T14:39:36 < dongs> all the IO is available, like 10 timers, 2 uarts, 2 i2c etc
2012-09-11T14:39:44 < dongs> just native usb is unused (but who needs that shit)
2012-09-11T14:39:59 < BrainDamage> what's the bga chip?
2012-09-11T14:40:08 < dongs> qfn, just usb>uart shit
2012-09-11T14:40:27 < dongs> for firmware/blah. cortex debug on teh back for proper worwk
2012-09-11T14:40:47 < karlp> wasn't tectu or titanmkd trying to get people to a chibios irc the other week?
2012-09-11T14:41:38 < zyp> there was some talk about that
2012-09-11T14:41:43 < tavish3> #chibios exists
2012-09-11T14:41:46 < dongs> they  can just talk about that shit here
2012-09-11T14:41:54 < dongs> its not like there's more than 2 users of chibios anyway
2012-09-11T14:42:48 < Laurenceb> omg 18 people
2012-09-11T14:44:34 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-11T14:44:40 < dongs> maybe tech2077 will be using chibios soon
2012-09-11T14:44:42 < dongs> he likes ghetto shit
2012-09-11T14:45:05 -!- daku is now known as DaKu
2012-09-11T14:45:26 < Tectu> karlp, here
2012-09-11T14:45:28 < Tectu> karlp, i did
2012-09-11T14:46:35 < Laurenceb> http://www.theregister.co.uk/2012/09/11/hab_ashes/
2012-09-11T14:47:16 < dongs> Laurenceb: you keep getting your shit in the news
2012-09-11T14:47:23 < dongs> next, you need to get something trollish in there
2012-09-11T14:48:15 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-11T14:49:05 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-11T14:51:20 < dongs> dongsssssssssssssssssssssssssssssssssssssssssssssssss
2012-09-11T14:51:28 < dongs> d o  n   g    s
2012-09-11T14:53:00 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-11T14:54:24 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Client Quit]
2012-09-11T14:55:38 < Laurenceb> http://www.st.com/internet/com/press_release/p3323.jsp
2012-09-11T14:56:11 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-11T14:56:26 < dongs> Geneva, September 11, 2012
2012-09-11T14:56:28 < dongs> laughing out loud.
2012-09-11T14:57:07 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-09-11T14:57:52 < karlp> pretty much says, "here's your missile development kit" doesn't it?
2012-09-11T14:58:02 < dongs> yep
2012-09-11T14:58:05 < dongs> i bet i cant order it from digikey.jp
2012-09-11T14:58:06 < dongs> checking
2012-09-11T14:58:27 < dongs> i can, too bad
2012-09-11T14:58:32 < dongs> i cant order any st gyros/accels
2012-09-11T14:58:32 < dongs> lol
2012-09-11T14:59:16 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-11T14:59:34 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 256 seconds]
2012-09-11T15:00:41 < dongs> so looks like chinaprice for F3 is gonna be aroudn 3 bucks
2012-09-11T15:00:44 < dongs> i can live wiht that
2012-09-11T15:01:31 < karlp> that press release says that the f303 contains an m4?
2012-09-11T15:01:39 < dongs> .. they all do?
2012-09-11T15:01:52 < dongs> m4 w/fpu at 72mhz
2012-09-11T15:02:07 < karlp> oh, I must have missed that, I thought f4 was still the only m4 part from st.
2012-09-11T15:02:08 < karlp> my bad
2012-09-11T15:02:12 < dongs> nope
2012-09-11T15:02:18 < dongs> thats whyt its so awesome.
2012-09-11T15:02:21 < dongs> AND
2012-09-11T15:02:25 < dongs> its pin compatible wiht F1 chips
2012-09-11T15:02:50 < Laurenceb> and it has groovy analogue stuff
2012-09-11T15:04:35 < dongs> STM32F3 Discovery kit firmware package, including 28 examples and preconfigured projects for 4 different IDEs
2012-09-11T15:04:38 < dongs> o cool
2012-09-11T15:04:41 < dongs> warez
2012-09-11T15:05:48 < dongs> http://www.st.com/internet/com/SALES_AND_MARKETING_RESOURCES/MARKETING_COMMUNICATION/MARKETING_BROCHURE/brstm32f3.pdf
2012-09-11T15:05:53 < dongs> gotta love ALLFUCKINGCAPS
2012-09-11T15:06:10 < dongs> lol @ front page of that
2012-09-11T15:06:13 < dongs> laurenceb material
2012-09-11T15:07:01 < Laurenceb> why am i the pedo
2012-09-11T15:07:06 < dongs> <Laurenceb> too old
2012-09-11T15:08:23 < Laurenceb> CCM-SRAM* (core coupled memory): 8 Kbytes of SRAM mapped
2012-09-11T15:08:23 < Laurenceb> to the instruction bus; critical routines loaded in the 8-Kbyte CCM
2012-09-11T15:08:23 < Laurenceb> at startup can be completed at full speed with zero wait states,
2012-09-11T15:08:23 < Laurenceb> achieving 94 Dhrystone MIPS and CoreMark score 155 at 72 MHz
2012-09-11T15:08:26 < Laurenceb> niceeee
2012-09-11T15:08:35 < Laurenceb> thats on the F30x
2012-09-11T15:09:03 < dongs> yeah, old news
2012-09-11T15:09:07 < dongs> we talkd about this liek a month ago
2012-09-11T15:09:18 < dongs> this will work good on esc32
2012-09-11T15:09:24 < dongs> can put the ADC interrupt there
2012-09-11T15:09:33 < dongs> that does averaging
2012-09-11T15:09:41 < dongs> right now thats what takes up most time
2012-09-11T15:09:52 < dongs> plus ADCs are faster to.
2012-09-11T15:10:05 < Laurenceb> •	 Up to 17 timers: 16 and 32 bits running up to 144 MHz*
2012-09-11T15:10:13 < Laurenceb> nice - they put them on a pll
2012-09-11T15:10:13 < dongs> wat
2012-09-11T15:10:18 < dongs> double the sysclk timere?
2012-09-11T15:10:21 < Laurenceb> yes
2012-09-11T15:10:28 < karlp> might have been nice if they'd fixed things from f1, rather than making them "api and pin compatible"
2012-09-11T15:10:33 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-11T15:10:42 < dongs> karlp: dont worry, i2c peripheral already has a mile of errata
2012-09-11T15:13:01 < karlp> what's with the 313 and 383? lower voltage, but higher minimum current?
2012-09-11T15:14:56 < Laurenceb> looks like ADC maxes out at 18msps due to memory bus
2012-09-11T15:15:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-11T15:15:15 <+dekar> dongs, stm32 has no problems bitbanging i2c anyway, so I don't see a problem with i2c errata
2012-09-11T15:15:21 < dongs> heh
2012-09-11T15:27:45 <+dekar> in my case I have to interface weird i2c slaves, so I need the flexibility to do it in software anyway
2012-09-11T15:28:44 < tavish3> how come block read/write is working and fat_fs isnt when it used to work earlier :|
2012-09-11T15:30:05 -!- tavish3 [~tavish@120.59.32.232] has quit [Quit: leaving]
2012-09-11T15:30:22 <+dekar> how long till those long filename patents run out?
2012-09-11T15:30:37 < Laurenceb> http://i.imgur.com/FvNaE.jpg  <- lmao
2012-09-11T15:31:58 < dongs> OLD
2012-09-11T15:32:03 < dongs> 14:10 < dongs> http://i.imgur.com/FvNaE.jpg
2012-09-11T15:32:03 < dongs> 21:30 < Laurenceb> http://i.imgur.com/FvNaE.jpg  <- lmao
2012-09-11T15:32:26 < karlp> I think he was trying to credit you
2012-09-11T15:32:30 < dongs> oh ok
2012-09-11T15:32:36 < karlp> and was too lazy to say, "that sand people image was funny"
2012-09-11T15:32:36 < Laurenceb> :P
2012-09-11T15:32:52 < karlp> because you know, typing vs pasting?
2012-09-11T15:33:02 < Laurenceb> the pedo
2012-09-11T15:33:06 < Laurenceb> oops
2012-09-11T15:35:18 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-11T15:36:13 < dongs> CMSIS-RTOS API: Generic RTOS interface for Cortex-M processor-based devices.
2012-09-11T15:38:34 < Oldboy> who pedo ?
2012-09-11T15:40:18 < dongs> nooooooooooo what the fuck
2012-09-11T15:40:27 < dongs> they kept the F1 DMA controller on F3!!!!!!!!!!!!!!!!
2012-09-11T15:40:32 < dongs> FUCKKKKKKKKKKKKK
2012-09-11T15:45:22 < karlp> api compatible man :)
2012-09-11T15:45:25 < karlp> aren't you happy?
2012-09-11T15:53:08 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-11T15:53:08 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-11T15:53:08 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-11T15:53:10 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-11T15:57:56 < dongs> so i guess then the HSI has same crap where you can only go to 64mhz from it?
2012-09-11T15:58:49 < dongs> directly as clock or /2 as pll input
2012-09-11T15:58:50 < dongs> yep
2012-09-11T15:58:51 < dongs> argh
2012-09-11T16:00:25 < Laurenceb> lol ST just visisted
2012-09-11T16:00:30 < Laurenceb> *visited
2012-09-11T16:00:34 < dongs> wehre's my free F3discovery
2012-09-11T16:00:37 < Laurenceb> apparently F3 doesnt exist
2012-09-11T16:00:49 < Laurenceb> the guy denied its existence
2012-09-11T16:00:58 < Laurenceb> tried to flog us F2
2012-09-11T16:01:00 < dongs> he probably denies holocaust too
2012-09-11T16:01:35 < Laurenceb> lol why did i know thats what youd say
2012-09-11T16:02:26 <+izua> you know dongs well
2012-09-11T16:02:30 <+izua> too well, one might say.
2012-09-11T16:02:50 < Laurenceb> eww
2012-09-11T16:03:14 < Laurenceb> "forthcoming products are covered by NDA"
2012-09-11T16:03:35 < dongs> ity> what is Ellisworth?
2012-09-11T16:03:41 < dongs> els> LadyAda with fewer piercings
2012-09-11T16:04:06 < Laurenceb> hahaha
2012-09-11T16:05:14 <+izua> https://en.wikipedia.org/wiki/Jeri_Ellsworth
2012-09-11T16:05:20 < dongs> yes, we know
2012-09-11T16:05:46 <+izua> yes but
2012-09-11T16:05:49 <+izua> pokemon argument
2012-09-11T16:06:46 < Laurenceb> nerd test: if you dont know who those people are you arent a nerd
2012-09-11T16:10:41 < BrainDamage> pokemon argument = https://en.wikipedia.org/wiki/Wikipedia:Pok%C3%A9mon_test this?
2012-09-11T16:11:16 < Laurenceb> my shrug code with sections is running XD
2012-09-11T16:11:37 < Laurenceb> made at Wikipedia:Articles for deletion, before specific Pokémon characters were merged into a list of Pokémon.
2012-09-11T16:11:42 < Laurenceb> ^as everyone knows
2012-09-11T16:14:12 < Laurenceb> -fno-exceptions
2012-09-11T16:14:21 < Laurenceb> NO-EXCEPTIONS O_o
2012-09-11T16:14:28 <+izua> NO EXCEPTIONS
2012-09-11T16:14:33 <+izua> -Wl,rul34
2012-09-11T16:15:18 <+izua> BrainDamage: yeah, that's sort of it
2012-09-11T16:17:07 < zyp> http://www.kickstarter.com/projects/kuy/galago-make-things-better <- did we bash this yet?
2012-09-11T16:18:32 < zyp> a lot of talk about a «debugger», I wonder if that's just a gdb stub on it
2012-09-11T16:22:16 < dongs> jesus fuyclkojgflfskjjh
2012-09-11T16:22:22 < dongs> ok thats fucking it im putting my stm32 trollboard up
2012-09-11T16:23:10 < zyp> oh, wait, I found the schematic
2012-09-11T16:23:24 < zyp> it's an attiny45 on the underside of the board
2012-09-11T16:23:31 < zyp> so it's some v-usb shit
2012-09-11T16:24:12 < zyp> which is used as a swd adapter
2012-09-11T16:24:13 < Laurenceb> haha wut
2012-09-11T16:26:34 < Laurenceb> nice - logger down to 76KB, bootloader is 23KB
2012-09-11T16:26:43 < Laurenceb> neatly fits into 128K devices
2012-09-11T16:28:59 < dongs> http://i.imgur.com/0QGH2.png putting this up on kickstarter, any homosexual product name ideas? i guess what I really need get china to make a matching debugger using jlinkOB or something
2012-09-11T16:29:46 < Laurenceb> cp2102?
2012-09-11T16:30:07 < dongs> yea
2012-09-11T16:31:28 < zyp> cute
2012-09-11T16:31:36 < zyp> what's going to be your selling point? price?
2012-09-11T16:31:51 <+izua> color
2012-09-11T16:31:54 <+izua> pink trolling device
2012-09-11T16:32:44 <+izua> plug it in to program it, it suddenly enumerates an extra pipe for a usb keyboard, randomly types crap
2012-09-11T16:33:05 < zyp> except cp2102 can't do that
2012-09-11T16:33:16 < dongs> zyp: yeah
2012-09-11T16:33:28 < dongs> izua: yeah, no native USB fuck that shit
2012-09-11T16:33:43 < BrainDamage> do you know you have to be usa citizen to open a kickstarter project?
2012-09-11T16:33:55 < BrainDamage> anyone from the world can donate but only americans can get the money ...
2012-09-11T16:33:59 <+izua> or you can have someone from us do it for you
2012-09-11T16:34:11 <+izua> but someone posted here an .eu clone of kickstarter
2012-09-11T16:34:31 < dongs> BrainDamage: bullshit, that UAE project that ripped dudes for a cool half million wasnt u.s.
2012-09-11T16:34:35 <+izua> sure, it had "help me buy a computer" or "i need a new camera", but at least it could take money and paper
2012-09-11T16:34:39 <+izua> pay*
2012-09-11T16:34:44 < dongs> http://www.kickstarter.com/projects/484889112/the-pocket-tv-makes-any-tv-a-smart-tv
2012-09-11T16:35:05 < dongs> Dubai, United Arab Emirates
2012-09-11T16:35:08 < dongs>  Ahmad Zahran (393 friends)
2012-09-11T16:35:14 < dongs> preeeeeeeeety sure that guy is NOT a u.s. citizen
2012-09-11T16:35:35 < BrainDamage> http://www.kickstarter.com/help/faq/creators#AmIEligToStarAKickProj
2012-09-11T16:36:00 -!- claude is now known as claude_work
2012-09-11T16:36:09 <+izua> you need someone to provide a us residence address i think
2012-09-11T16:36:27 < dongs> so how did that arab scam
2012-09-11T16:36:28 <+izua> doesn't have to be you, but they'll kick down his door if you make bombs, i guess
2012-09-11T16:36:42 <+izua> well, he didn't scam.. yet
2012-09-11T16:36:46 < dongs> he did
2012-09-11T16:36:51 < dongs> those are $40 at dealextreme
2012-09-11T16:36:54 < dongs> in exactly same cases too.
2012-09-11T16:36:58 < Laurenceb> lol @ screenshots
2012-09-11T16:36:59 < dongs> and they "lost all their software develoeprs"
2012-09-11T16:37:04 < dongs> according to lates update
2012-09-11T16:37:09 < Laurenceb> "using xbox kinect to detect autism"
2012-09-11T16:37:19 <+izua> wtf
2012-09-11T16:37:20 <+izua> lol
2012-09-11T16:37:25 < Laurenceb> anyone using it _is_ autistic
2012-09-11T16:37:29 <+izua> that's even worse than "help me buy a new computer"
2012-09-11T16:37:32 < Laurenceb> simple problem is simple
2012-09-11T16:38:40 < Laurenceb> kickstarter is for noobs
2012-09-11T16:38:53 < zyp> dongs, only problem with your board is «dick this pad to get it into bootloader mode», that seems tedious
2012-09-11T16:38:55 < Laurenceb> if you have a serious project its easy to get funding
2012-09-11T16:38:58 <+izua> angel investors is where it's at
2012-09-11T16:39:13 < Laurenceb> or just sell your project rights like a normal person
2012-09-11T16:39:37 < dongs> zyp, thats only for hardcore shit
2012-09-11T16:39:47 < dongs> obviosuly your app would have some bootloader jump
2012-09-11T16:39:49 < dongs> over serial.
2012-09-11T16:40:39 < dongs> some enterprising negro could just solder some momentary pushbutton there.
2012-09-11T16:40:42 < zyp> yea, unless the app shits itself and gets stuck
2012-09-11T16:40:46 < dongs> right
2012-09-11T16:40:55 < dongs> but real men develop with swd
2012-09-11T16:40:57 < dongs> where this is nonissue.
2012-09-11T16:41:09 < zyp> can probably launch bootloader from hardfault handler though
2012-09-11T16:41:22 < Laurenceb> im using iwdg
2012-09-11T16:41:34 < Laurenceb> bootloader detects iwdg flag
2012-09-11T16:46:01 < Laurenceb> http://www.b3tards.com/u/33c8c804092e39b83959/n5gdtlsu.jpg
2012-09-11T16:56:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-11T16:57:42 < Laurenceb> is there a way to kill IWDG is standby mode?
2012-09-11T16:57:51 < Laurenceb> so it doesnt keep making up the core
2012-09-11T16:57:55 < Laurenceb> *waking
2012-09-11T17:00:15 < dongs> <Laurenceb> *wanking
2012-09-11T17:01:11 < zyp> wouldn't that kind of defeat the point?
2012-09-11T17:01:34 < Laurenceb> http://cdn2.holytaco.com/wp-content/uploads/2008/07/bukkake3.jpg
2012-09-11T17:03:00 < dongs> retweeting everywhere
2012-09-11T17:18:21 < Laurenceb> grr this is complex
2012-09-11T17:18:27 < Laurenceb> my bootloader goes into a loop
2012-09-11T17:18:52 < Laurenceb> so... ill change it to read the vbus voltage and iwdg reset status bit
2012-09-11T17:19:14 < Laurenceb> no iwdg reset -> start watchdog, jump to app code
2012-09-11T17:19:22 < Laurenceb> iwdg -> check vbus
2012-09-11T17:19:47 < Laurenceb> vbus -> check for firmware and flash then jump if found any, else turn off
2012-09-11T17:19:59 < Laurenceb> no vbus -> turn off
2012-09-11T17:20:07 < Laurenceb> that should do it?
2012-09-11T17:23:10 < Laurenceb> s/else turn off/try attach as mass storage, turn off if failure
2012-09-11T17:23:56 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-11T17:39:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-11T17:55:39 < Laurenceb> https://www.olimex.com/dev/imx233-olinuxino-maxi.html
2012-09-11T17:55:49 < Laurenceb> i spot same smsc ic as raspberry
2012-09-11T18:10:06 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-11T18:10:06 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-231-162.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-11T18:13:49 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-11T18:13:50 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T18:16:33 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-11T18:25:20 <+izua> why.. does.. that.. even end in -ino
2012-09-11T18:26:58 <+Steffanx> why ? not ? ?
2012-09-11T18:27:08 <+Steffanx> * .. .
2012-09-11T18:32:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-11T18:38:22 <+dekar> github is down :/
2012-09-11T18:41:40 < emeb> probably on GoDaddy
2012-09-11T18:43:31 < BrainDamage1> ino is an italian suffix for "tiny and cute"
2012-09-11T18:44:05 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-11T18:44:17 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-11T18:45:04 < emeb> kinda like 'ito' in Spanish I guess.
2012-09-11T18:45:10 -!- claude_work is now known as claude
2012-09-11T18:45:12 < Laurenceb> wheres dong tiny cute pink board thne?
2012-09-11T18:45:26 < emeb> dongino?
2012-09-11T18:46:49 < Laurenceb> http://i.imgur.com/L29UB.png
2012-09-11T18:47:17 -!- barthess [~barthess@178.154.11.214] has joined ##stm32
2012-09-11T18:48:17 < emeb> most excellent.
2012-09-11T18:48:54 <+izua> rofl
2012-09-11T18:54:15 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-11T18:54:18 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T18:55:54 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 264 seconds]
2012-09-11T18:56:43 < emeb> Github down -> Cloud computing proponents get a reality check.
2012-09-11T18:57:29 < BrainDamage> github down is not a big issue, git was built to be distributed
2012-09-11T18:57:35 < BrainDamage> you got a local copy of the files
2012-09-11T18:57:38 < BrainDamage> just wait a little
2012-09-11T18:57:41 < emeb> Big if.
2012-09-11T18:57:57 < BrainDamage> I said no if
2012-09-11T18:58:10 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-11T18:58:14 < BrainDamage> it's when amazon service went down that was more funny
2012-09-11T18:58:45 < emeb> Wonder if it's possible to have a backup/secondary git repo to use if main github disappears?
2012-09-11T18:59:08 < BrainDamage> you can use any server, even just a file server
2012-09-11T18:59:17 < BrainDamage> you could push to dropbox if you want
2012-09-11T18:59:43 < emeb> Don't know enough about git
2012-09-11T19:00:01 < emeb> just seeing folks I know who use github for business complaining.
2012-09-11T19:00:55 < emeb> they use it for collaborative work & can't collaborate w/ github inaccessible.
2012-09-11T19:01:05 < BrainDamage> they are dumb
2012-09-11T19:01:21 < emeb> wondering if there are alternatives / backups for that sort of thing.
2012-09-11T19:01:31 < BrainDamage> basically, run git --bare init; in a folder, put that folder on a file server, git remote add path/to/server
2012-09-11T19:01:35 < BrainDamage> done
2012-09-11T19:01:46 < emeb> kewl
2012-09-11T19:02:12 < BrainDamage> sorry, git remote add aliasname /path/to/server
2012-09-11T19:02:34 < BrainDamage> then simply git push aliasname master, git pull aliasname master
2012-09-11T19:02:40 < BrainDamage> to send and receiver the master branch
2012-09-11T19:02:47 < BrainDamage> git keeps a local copy of the whole history
2012-09-11T19:02:51 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-11T19:02:58 < BrainDamage> so they'll have their whole repo copied in no time
2012-09-11T19:03:06 < Laurenceb> https://a248.e.akamai.net/assets.github.com/images/error/angry_unicorn.png?1334862345
2012-09-11T19:03:09 < Laurenceb> nice
2012-09-11T19:03:10 < emeb> so that handles the basic repo functions, but not all the other social/tracking/wiki junk that github does.
2012-09-11T19:03:44 < BrainDamage> right
2012-09-11T19:03:46 < emeb> hmm - wonder where he got that angry unicorn.
2012-09-11T19:03:48 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-11T19:04:12 < BrainDamage> but you can always use emails or other communication methods in the meantime
2012-09-11T19:04:24 < emeb> only old people use email.
2012-09-11T19:04:49 < BrainDamage> ok, they should use facebook chat then :p
2012-09-11T19:04:56 < emeb> do it all w/ twitter :P
2012-09-11T19:05:36  * emeb probably falls into the "old people" category.
2012-09-11T19:06:51 < emeb> sourceforge sends email advocating I "explore the black hole". *do not want*
2012-09-11T19:12:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-11T19:12:39 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-11T19:17:06 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-11T19:21:14 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-11T19:21:36 < Laurenceb> how do i sleep until a set time in chibios?
2012-09-11T19:21:46 < Laurenceb> i know theres an example of this somethere...
2012-09-11T19:21:51 < Laurenceb> *somewhere
2012-09-11T19:22:35 <+Steffanx> Find the example?
2012-09-11T19:22:46 < BrainDamage> speaking of vcs and issue tracking, this looks interesting: http://ditz.rubyforge.org/
2012-09-11T19:23:00 < Laurenceb> http://en.verysource.com/code/2932035_1/timing.dox.html
2012-09-11T19:23:04 < Laurenceb> ^found it
2012-09-11T19:24:19 <+Steffanx> Oh, it doesn't really sleep ?. ?
2012-09-11T19:27:06 < emeb> that thread/process sleeps, but does the whole system?
2012-09-11T19:27:21 < emeb> ie - not necessarily saving power, etc.
2012-09-11T19:27:49 <+Steffanx> That's what it looks like
2012-09-11T19:28:25 < Laurenceb> yes
2012-09-11T19:29:07 < emeb> of course if you're using an RTOS the intent is that it manages all that for you in the background.
2012-09-11T19:29:31 < emeb> presumably you'll have other threads w/ their own priority that get serviced as required.
2012-09-11T19:34:14 < Laurenceb> im rewriting all my sepectrometer stuff in chibios atm
2012-09-11T19:34:29 < Laurenceb> one thread at a time...  :P
2012-09-11T19:34:51 < Laurenceb> air reservoir pressure regulator thread atm
2012-09-11T19:34:56 < Laurenceb> keeps it at 6PSI
2012-09-11T19:40:28 < zyp> the F1 usb peripheral is so simple compared to the one in F4
2012-09-11T19:40:55 < zyp> five generic registers and one per endpoint
2012-09-11T19:41:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-11T19:41:42 < zyp> compared to ~30 generic registers and four per endpoint
2012-09-11T19:42:33 < Laurenceb> heh
2012-09-11T19:42:41 < Laurenceb> i wish stm32f4 adc was simpler
2012-09-11T19:43:04 < Laurenceb> looks like ill have to read ref manual to set up chibios properly
2012-09-11T19:43:14 < Laurenceb> but for now... im going home
2012-09-11T19:43:15 <+Steffanx> simpler and less advanced/features?
2012-09-11T19:43:29 < Laurenceb> chibios is basically just register access for f4 adc
2012-09-11T19:43:36 < Laurenceb> bbl
2012-09-11T19:45:27 < emeb> ADC isn't hard to use. Plenty of example code out there.
2012-09-11T19:46:41 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-11T19:47:05 < zyp> I just started on a driver for the F1 USB, wonder if I can get it to enumerate before I get tired of it and stops for today :p
2012-09-11T19:49:12 <+Steffanx> YES YOU CAN
2012-09-11T19:50:07 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 240 seconds]
2012-09-11T19:51:12 < zyp> ok
2012-09-11T19:53:08 < emeb> doit doit doit doit doit.
2012-09-11T19:53:38 < emeb> (count on IRC for validation and encouragement)
2012-09-11T19:53:54 < zyp> :)
2012-09-11T19:58:53 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-11T19:59:09 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-11T20:38:48 < BrainDamage> http://i.imgur.com/95Vc3.jpg
2012-09-11T20:41:15 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-11T20:41:32 < TitanMKD> hi
2012-09-11T20:45:10 <+izua> BrainDamage: i hear the sound of jimmies being rustled so hard by that picture
2012-09-11T20:46:09 < TitanMKD> zyp i just retested generated code and in fact i was building with default -O so -O0 it is why the asm code was ugly
2012-09-11T20:47:04  * emeb does not understand the significance of "rustling jimmies"
2012-09-11T20:54:00 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-11T20:56:16 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 245 seconds]
2012-09-11T20:58:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-11T20:58:37 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 268 seconds]
2012-09-11T20:59:35 < BrainDamage> http://knowyourmeme.com/memes/that-really-rustled-my-jimmies
2012-09-11T20:59:41 < BrainDamage> I had to google it too
2012-09-11T21:01:59 < emeb> memes. should have known - all is revealed to those who consult the googles.
2012-09-11T21:02:37 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-11T21:04:19 < jpa-> "It is one of the largest threads dedicated to unrustling jimmies to date.." i have to appreciate the effort some people put into bringing us these memes
2012-09-11T21:12:10 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: Lost terminal]
2012-09-11T21:55:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-11T21:57:35 -!- Netsplit *.net <-> *.split quits: |sark
2012-09-11T21:58:47 -!- Netsplit over, joins: |sark
2012-09-11T21:58:48 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-11T22:06:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-11T22:07:10 <+Steffanx> Any result(s) zyp ?
2012-09-11T22:22:56 < Laurenceb_> void result(char s) {
2012-09-11T22:23:32 <+Steffanx> return; }
2012-09-11T22:35:40 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-09-11T22:50:36 < Laurenceb_> how do i use find?
2012-09-11T22:50:54 < Laurenceb_> ls -R | grep "Pressure.c"
2012-09-11T22:50:56 < Laurenceb_> works
2012-09-11T22:51:11 < Laurenceb_> but i need to find where the file is
2012-09-11T22:52:22 < qyx_> find . | grep "Pressure\.c"
2012-09-11T22:52:33 < qyx_> or find . -name "Pressure.c"
2012-09-11T22:53:06 < qyx_> with \
2012-09-11T22:54:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-11T22:58:09 < Laurenceb_> thanx
2012-09-11T23:00:07 < Laurenceb_> so i wondered my chibios adc config is so complex
2012-09-11T23:00:21 < Laurenceb_> read F4 datasheet, its almost the same adc as f1...
2012-09-11T23:00:33 < Laurenceb_> turns out chibios _always_  uses DMA with ADC
2012-09-11T23:00:35 < Laurenceb_> so stupid
2012-09-11T23:01:38 < TitanMKD> and it is good ADC use DMA
2012-09-11T23:01:46 < TitanMKD> it does not consume CPU for nothing
2012-09-11T23:02:19 < Laurenceb_> yes but i want to so a single software triggered conversion
2012-09-11T23:03:03 < TitanMKD> you can also disable DMA in that case ;)
2012-09-11T23:03:15 < TitanMKD> just use llp api to disable DMA channel ...
2012-09-11T23:08:16 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-11T23:08:16 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-11T23:08:59 < Laurenceb_> sure - i want to use the HAL properly
2012-09-11T23:09:22 < Laurenceb_> in which case i have to let it use DMA and get callback for DMA ISR, not ADC ISR
2012-09-11T23:15:51 < Laurenceb_> oh well - its not like i cant afford clk cycles
2012-09-11T23:17:04 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-09-11T23:19:27 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-11T23:26:38 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-11T23:32:22 -!- barthess [~barthess@178.154.11.214] has quit [Quit: Leaving.]
2012-09-11T23:36:42 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-11T23:37:02 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-11T23:37:02 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-11T23:38:31 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
--- Day changed Wed Sep 12 2012
2012-09-12T00:11:52 < Laurenceb_> http://www.youtube.com/watch?v=b9S-NzsUD5w&feature=g-vrec
2012-09-12T00:14:47 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-12T00:20:54 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-12T00:37:21 -!- Erlkoenig [~erlkoenig@pptp-212-201-69-228.pptp.stw-bonn.de] has joined ##stm32
2012-09-12T00:39:56 < zyp> Steffanx, nah, I got invited to some friends, an old friend that lives abroad was in town tonight so I had to go say hi :p
2012-09-12T00:41:03 < zyp> but well, I got to the stage of detecting and logging bus reset events if that counts for anything: http://paste.jvnv.net/view/kujVL
2012-09-12T00:42:53 < Laurenceb_> zyp: ever used adc with chibios?
2012-09-12T00:43:07 < Laurenceb_> does adcConvert call the callback?
2012-09-12T00:44:39 < zyp> hmm
2012-09-12T00:44:50 < zyp> I actually have
2012-09-12T00:45:01 < zyp> but I didn't use any callback
2012-09-12T00:45:36 < Laurenceb_> hmm
2012-09-12T00:45:41 < zyp> http://cgit.jvnv.net/suzumebachi/tree/main.cpp?id=422a778#n196 <- I'm not sure if this code is even sane
2012-09-12T00:45:42 < Laurenceb_> using adcConvert?
2012-09-12T00:45:51 < Laurenceb_> does it run?
2012-09-12T00:45:53 < Laurenceb_> lol
2012-09-12T00:46:39 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 260 seconds]
2012-09-12T00:46:48 < Laurenceb_> adcstartconversion?
2012-09-12T00:46:55 < Laurenceb_> im pretty sure that does use callback
2012-09-12T00:47:08 < zyp> if you supply one, which I didn't
2012-09-12T00:49:19 < Laurenceb_> hmm
2012-09-12T00:49:27 < Laurenceb_> chibios is too clever to understand
2012-09-12T00:49:33 < Laurenceb_> - for me to
2012-09-12T00:49:34 < BrainDamage> fun stuff: https://isc.sans.edu/diary/End+of+Days+for+MS-CHAPv2/13807
2012-09-12T00:50:37 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 240 seconds]
2012-09-12T00:54:36 -!- blocky [blocky@S0106c0c1c07746fb.gv.shawcable.net] has joined ##stm32
2012-09-12T01:02:39 <+Steffanx> heh shit happens zyp :P
2012-09-12T01:03:04 < Laurenceb_> oh
2012-09-12T01:03:11 < Laurenceb_> you can pass a NULL pointer
2012-09-12T01:03:15 < Laurenceb_> and callback isnt used
2012-09-12T01:08:00 < zyp> Steffanx, it's okay, getting resets means that I just have to configure the control endpoint and wait for setup packets to dump in, shouldn't be too much work :p
2012-09-12T01:08:17 <+Steffanx> I was referring to what you said before that :)
2012-09-12T01:08:18 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-12T01:08:55 < zyp> well, I'm expecting that I'll get enumeration working tomorrow then :p
2012-09-12T01:09:42 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-12T01:09:55 < zyp> after all, the control command handling is already in place, I just have to write a few lines for endpoint configuration and handling incoming/outgoing packets :p
2012-09-12T01:12:38 -!- mrcan_ [~mrcan@78.173.75.48] has joined ##stm32
2012-09-12T01:12:39 -!- mrcan_ [~mrcan@78.173.75.48] has quit [Changing host]
2012-09-12T01:12:39 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-12T01:16:54 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 276 seconds]
2012-09-12T01:17:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-12T01:23:25 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 244 seconds]
2012-09-12T01:25:03 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-12T01:29:35 < Laurenceb_> zyp: isnt the an ST example?
2012-09-12T01:29:41 < Laurenceb_> theres a ton for the F1
2012-09-12T01:34:02 < zyp> what makes you think I would be interested in copypasting an example and hacking it to work?
2012-09-12T01:36:05 < zyp> http://mbed.org/blog/entry/mbed-Interface-v20-experiments-Adding-CM/
2012-09-12T01:36:13 < zyp> that's a bit interesting
2012-09-12T01:37:19 < Laurenceb_> well...
2012-09-12T01:37:31 < Laurenceb_> i was kind of thinking you cared about getting work done
2012-09-12T01:38:44 < zyp> I don't have any F1 based projects that's using USB, what is the work I want to get done? :p
2012-09-12T01:40:49 < zyp> I'm just adding the driver for F1 to get a break from the quirky OTG cores in the F4, and to get another perspective
2012-09-12T01:42:07 -!- dekar_ [~dekar@212.255.20.12] has joined ##stm32
2012-09-12T01:42:08 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-12T01:42:08 < zyp> it might also come in handy in the future, since both L1 and F3 are using the same core
2012-09-12T01:44:01 -!- dekar [~dekar@212.255.31.74] has quit [Ping timeout: 246 seconds]
2012-09-12T01:54:52 -!- prog [prog@94.100.30.150] has quit [Ping timeout: 246 seconds]
2012-09-12T02:01:05 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-12T02:02:45 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-12T02:04:31 -!- Guest54600 [~prog@bornbrilliant.org] has joined ##stm32
2012-09-12T02:07:55 < Laurenceb_> ewww
2012-09-12T02:08:09 < Laurenceb_> i _was_ planning to use an F3 on my F1 board
2012-09-12T02:09:20 < Laurenceb_> how does  CMSIS-DAP work without a driver?
2012-09-12T02:10:55 -!- Guest54600 is now known as prog
2012-09-12T02:21:02 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-12T02:21:27 -!- DaKu is now known as daku
2012-09-12T02:25:48 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 276 seconds]
2012-09-12T02:26:53 -!- Reggie__ [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-12T02:27:22 < zyp> huh?
2012-09-12T02:27:24 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Read error: Connection reset by peer]
2012-09-12T02:31:32 -!- Reggie__ [~ReggieUK@5aca5630.bb.sky.com] has quit [Ping timeout: 248 seconds]
2012-09-12T02:35:49 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-12T02:59:50 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-09-12T03:00:09 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-12T03:10:28 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 248 seconds]
2012-09-12T03:11:48 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-09-12T03:17:26 < dongs> probably HID, no?
2012-09-12T03:19:31 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-12T03:19:31 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-12T03:19:31 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-12T03:22:24 < emeb> Hope not - HID be very slow.
2012-09-12T03:23:17 < zyp> who said it works without a driver?
2012-09-12T03:25:05 < BrainDamage> "CMSIS-DAP driverless USB debug connection"
2012-09-12T03:25:25 < zyp> huh
2012-09-12T03:25:39 < zyp> I wouldn't trust that :p
2012-09-12T03:36:59 < qyx_> acm/cdc?
2012-09-12T03:37:30 < dongs> thats not driverless on any real OS
2012-09-12T03:56:07 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-12T03:57:57 < emeb> heh - "real" OS needs drivers for acm/cdc :P
2012-09-12T03:58:29 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-12T04:03:37 -!- Erlkoenig [~erlkoenig@pptp-212-201-69-228.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-12T04:03:50 < dongs> ur so funneh
2012-09-12T04:04:00 < dongs> if ur os was real it would have more than 1% installed base
2012-09-12T04:04:22 < dongs> as long as  windows is on 98% of desktops, none of the shit matter.s
2012-09-12T04:06:32 < qyx_> it should work on win without installing anything
2012-09-12T04:06:44 < qyx_> should, i have no working example
2012-09-12T04:09:57 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-12T04:10:04 < dongs> what is "it"
2012-09-12T04:10:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-12T04:10:14 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-12T04:11:03 < emeb> actually, I'd bet acm/cdc works w/o custom driver on Win, but you still need a .ini to set up the correct linkages.
2012-09-12T04:19:38 < dongs> correct
2012-09-12T04:19:41 < dongs> which you still need to sign
2012-09-12T04:19:48 < dongs> unless you wanna look like a pirate
2012-09-12T04:20:09 < emeb> arrr!
2012-09-12T04:25:53 < HTT-Bird> .inf, but yeah
2012-09-12T04:26:23 < emeb> derp
2012-09-12T04:26:30 < qyx_> steal some suitable ids
2012-09-12T04:27:06 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-12T04:32:03 < dongs> if tarduino faggots would get off their ass and whql their CDC/ACM infs peopel could just use those
2012-09-12T05:15:52 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-12T05:22:08 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-12T05:47:40 < Bird|lappy> as a rule of thumb, should I rely on an ARM uC vendor's supplied convenience libs, or use the "raw" CMSIS header definitions for their part directly, or is that completely vendor-dependent?
2012-09-12T06:00:18 < dongs> i have no problem using stdperiphlib :)
2012-09-12T06:03:38 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-12T06:03:44 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-12T06:04:12  * Bird|lappy nods
2012-09-12T06:05:41 < Bird|lappy> do you normally install the peripheral-libs in your system cross-tools setup, or keep them with your project?
2012-09-12T06:07:51 < dongs> keep
2012-09-12T06:07:59 < dongs> i use keil, and they fuck around even with cmsis shit
2012-09-12T06:08:03 < dongs> adding some of their own crap in there
2012-09-12T06:08:10 < dongs> so I have lib\ with original ST / ARM stuff in it
2012-09-12T06:08:45 < Bird|lappy> got it
2012-09-12T06:09:24 < dongs> anyway, better to knwo exactly what version stuff is built with
2012-09-12T06:09:38 < dongs> so even if they didnt fuck with it i'd probably still have required shit with each project
2012-09-12T06:17:17 -!- mrcan__ [~mrcan@78.173.75.48] has joined ##stm32
2012-09-12T06:17:18 -!- mrcan__ [~mrcan@78.173.75.48] has quit [Changing host]
2012-09-12T06:17:18 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-12T06:20:40 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-09-12T06:38:01 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Read error: Connection reset by peer]
2012-09-12T06:39:15 < emeb> trying to build the bmp firmware - looks like I need libnosys - anyone got a quick summary of what's neede?
2012-09-12T06:47:31 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-12T06:51:18 < zippe> emeb: libnosys is part of the standard newlib dist
2012-09-12T06:52:30 < emeb> zippe: yep found it on redhat
2012-09-12T06:52:46 < emeb> I guess I need to build that w/ the arm-none-eabi toolchain
2012-09-12T06:53:01 < emeb> sounds... "fun"
2012-09-12T06:54:20 < dongs> I hate it when people try to be smart and use some obscure crap lib in their app
2012-09-12T06:54:23 < dongs> and then its like the onyl app that needs it
2012-09-12T06:54:33 < dongs> and then its wasting space
2012-09-12T06:54:42 < dongs> at leat on windows people are smart enough theyt'll just statically link the shit
2012-09-12T07:00:58 < emeb> I kinda wish that bmp distributed firmware binaries rather than forcing me to figure out how to build it.
2012-09-12T07:11:03 < zippe> dongs: libnosys is part of the standard toolchain
2012-09-12T07:11:30 < emeb> unless you happen to use codesourcery, which doesn't appear to have it.
2012-09-12T07:13:43 < emeb> any suggestions on a pre-built toolchain that has newlib/libgloss/libnosys/etc already in place?
2012-09-12T07:14:18 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-12T07:17:21 < dongs> i geuss that arm-gcc-shit at launchpad  that the guys keep pimping
2012-09-12T07:17:34 < emeb> yeah - was just looking for the link...
2012-09-12T07:17:40 < dongs> launchpad.net/gcc-arm-embedded
2012-09-12T07:19:01 < emeb> ty
2012-09-12T07:19:49 < emeb> Oh Yay - http://phys.org/news/2012-09-raspberry-pi-supercomputer.html#jCp
2012-09-12T07:20:47 < dongs> python? what teh FUCK
2012-09-12T07:20:52 < dongs> since when garbage liek taht even makes news
2012-09-12T07:21:26 < emeb> since it involves _raspberry_pi_ which has teh kewl.
2012-09-12T07:21:34 < Bird|lappy> btw: do you put your -mcpu in with your CC= line or in your ALL_CFLAGS= line in your Makefiles?
2012-09-12T07:21:45 < Bird|lappy> or just directly in your compile command?
2012-09-12T07:21:58 < emeb> I put it in the cflags
2012-09-12T07:22:17 < emeb> really doesn't matter though - all ends up in the same place when you compile.
2012-09-12T07:25:00 <+dekar_> emeb, the proprietary codesourcery libs are basically nosys anyway
2012-09-12T07:25:25 <+dekar_> they distribute them with their toolchain
2012-09-12T07:31:13 < emeb> dekar_: hmmm
2012-09-12T07:32:26 < emeb> I see that there are some libs in arm-2012.03/arm-none-eabi/lib/
2012-09-12T07:32:37 < emeb> nothing explicitly called libnosys though...
2012-09-12T07:35:25 < emeb> but, there is a gcc-arm-none-eabi-4_6-2012q2/arm-none-eabi/lib/libnosys.a
2012-09-12T07:35:43 < emeb> so I'll give the launchpad version a try...
2012-09-12T07:41:41 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-12T07:44:13 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-12T07:46:12 < emeb> yay! it worked.
2012-09-12T07:46:56  * emeb now waits to find out what's wrong with the launchpad toolchain
2012-09-12T07:53:07 < emeb> well, bmp built, bootloader took the firmware, enumerated as acm.
2012-09-12T07:55:13 < emeb> Yay! bmp programs my balky F405 w/o errors on the first try.
2012-09-12T07:59:38 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 272 seconds]
2012-09-12T08:03:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-12T08:14:27 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-12T08:14:27 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-12T08:14:27 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-12T08:14:29 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-12T08:18:00 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-09-12T08:28:36 -!- tech2077 [~tech2077@adsl-75-53-131-5.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-09-12T08:34:54 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 264 seconds]
2012-09-12T08:42:13 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-12T08:49:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-12T09:32:57 < emeb_mac> good price?
2012-09-12T09:46:59 < emeb_mac> not bad - $10 for all the GPS you can stand.
2012-09-12T09:51:29 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 260 seconds]
2012-09-12T09:52:31 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has joined ##stm32
2012-09-12T09:52:32 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has quit [Changing host]
2012-09-12T09:52:32 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-12T10:05:07 < emeb_mac> here you go: http://www.analogindustries.com/blog/entry.php?blogid=1274763248269
2012-09-12T10:11:23 <+dekar_> emeb_mac, the codesourcery libs have cs3 in their name iirc
2012-09-12T10:11:48 < emeb_mac> dekar_: I'll check
2012-09-12T10:13:20 -!- dekar_ [~dekar@212.255.20.12] has quit [Quit: This computer has gone to sleep]
2012-09-12T10:18:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-12T10:21:08 -!- claude is now known as claude_work
2012-09-12T10:45:49 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-12T10:49:04 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-12T10:49:05 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-12T10:49:05 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-12T10:52:04 < dongs> trollb_
2012-09-12T10:53:58 < dongs> Laurenceb_: did you hear, F3 has the same shittastic DMA controller as F1.. :(
2012-09-12T10:54:02 < dongs> at least there's DMA2 on all parts...
2012-09-12T10:54:23 < Laurenceb_> yes
2012-09-12T10:54:45 < dongs> is USB same too? loll
2012-09-12T10:55:01 < dongs> i.e. 103-style vs 105/107/f2/f4 style
2012-09-12T10:55:14 < Laurenceb_> aiui it had f4 usb
2012-09-12T10:55:19 < Laurenceb_> worst of both worlds
2012-09-12T10:55:19 < dongs> rly?
2012-09-12T10:55:26 < Laurenceb_> according to zyp
2012-09-12T10:55:29 < dongs> hur
2012-09-12T10:55:36 < dongs> i looked at the sores
2012-09-12T10:55:39 < dongs> for f3discovery sampels
2012-09-12T10:55:43 < dongs> it looked like F1 usblib in there
2012-09-12T10:55:52 < Laurenceb_> havent read it
2012-09-12T10:57:15 < Laurenceb_> bbl
2012-09-12T10:58:03 < dongs> http://i.imgur.com/kFv39.jpg
2012-09-12T10:58:29 < zyp> 09:55:14 < Laurenceb_> aiui it had f4 usb
2012-09-12T10:58:35 < zyp> no, I didn't say that
2012-09-12T10:58:50 < zyp> I said that L1 and F3 has same usb core as F1
2012-09-12T10:59:01 < dongs> which f1, 103 or 105+?
2012-09-12T10:59:04 < zyp> 103
2012-09-12T10:59:11 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-12T10:59:12 < zyp> same as in non-otg
2012-09-12T10:59:16 < dongs> argh, great
2012-09-12T11:01:06 < dongs> https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-ash3/527012_10151091922546553_636796050_n.jpg lol, america
2012-09-12T11:01:22 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-12T11:02:56 -!- dekar [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has joined ##stm32
2012-09-12T11:02:59 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-12T11:05:19 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-12T11:05:31 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-12T11:06:03 -!- dekar_ [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has joined ##stm32
2012-09-12T11:06:07 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-12T11:07:27 -!- dekar [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has quit [Ping timeout: 246 seconds]
2012-09-12T11:13:33 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-12T11:30:32 <+dekar_> OSX sucks to hard: https://dl.dropbox.com/u/28467113/Screen%20Shot%202012-09-12%20at%2010.20.20%20AM.png
2012-09-12T11:31:51 < zyp> cute
2012-09-12T11:32:00 <+dekar_> *too
2012-09-12T11:32:25 < zyp> did you set backlog to unlimited and cat some GB's worth of files?
2012-09-12T11:32:29 <+dekar_> my system is incredibly laggy atm, that's just 21 days of uptime
2012-09-12T11:32:49 <+dekar_> I didn't change any settings and compile TNT a few times
2012-09-12T11:33:13 <+dekar_> if that's default, then it is pretty much broken
2012-09-12T11:33:31 <+dekar_> every time I open terminal up again it also claims that memory
2012-09-12T11:34:04 <+dekar_> besides that I think that kernel task shouldn't need that amount of memory either
2012-09-12T11:38:14 < ntfreak>  afraid currently CMSIS-DAP uses HID - hope this will change but unlikely
2012-09-12T11:39:37 <+Steffanx> The kernel task is 'ok' dekar_
2012-09-12T11:40:49 <+dekar_> Steffanx, my system is uber laggy, swap used is 5.5gb
2012-09-12T11:41:07 <+Steffanx> but still here its ~700MB too
2012-09-12T11:41:12 <+dekar_> I guess I either need 16gb ram or an SSD
2012-09-12T11:41:12 <+Steffanx> ( no laggyness)
2012-09-12T11:41:40 <+dekar_> what's your uptime?
2012-09-12T11:41:57 <+dekar_> my OSX always gets laggy after >2 weeks
2012-09-12T11:42:24 <+Steffanx> 2 days :)
2012-09-12T11:42:51 <+Steffanx> Never have issues with laggy 'behaviour' after a few weeks
2012-09-12T11:43:24 <+dekar_> Lion?
2012-09-12T11:43:38 <+Steffanx> mountain
2012-09-12T11:43:58 <+dekar_> I wonder why I have those problems, I had that back on snow leopard as well
2012-09-12T11:44:12 <+dekar_> and I formatted since then
2012-09-12T11:44:49 <+Steffanx> Here the backlog is set too "Limit to avaibable memory" .. that's probably the default setting
2012-09-12T11:45:09 <+Steffanx> So lets compile your toolchain a few times :)
2012-09-12T11:46:12 <+dekar_> Steffanx, https://dl.dropbox.com/u/28467113/mac_swap_fail.png
2012-09-12T11:46:45 <+Steffanx> Oh, you only have 4?
2012-09-12T11:46:47 <+Steffanx> G
2012-09-12T11:47:00 <+dekar_> I had back then
2012-09-12T11:47:08 <+dekar_> now I have 8gb and things got a bit better
2012-09-12T11:47:18 <+dekar_> but look at those 59gb of swap
2012-09-12T11:47:25 <+dekar_> I seriously have no idea what that was
2012-09-12T11:47:31 <+Steffanx> I had that with 4GB as well
2012-09-12T11:48:15 <+Steffanx> but that was probably because I used some virtual machines pretty often
2012-09-12T11:48:27 <+dekar_> it got much better with 8gb but it still starts swapping after about a week of uptime and slows down more and more from that point on
2012-09-12T11:48:38 <+dekar_> yeah I still use a ton of VMs
2012-09-12T11:49:01 <+dekar_> plugging in my display via display port took like 2min today
2012-09-12T11:49:12 <+dekar_> that's how laggy it is already
2012-09-12T11:49:14 <+Steffanx> heh
2012-09-12T11:49:28 <+dekar_> I'll probably reboot to fix it
2012-09-12T11:54:08  * Steffanx wonders when mr. D wakes up and says you should use windows :P
2012-09-12T11:55:01 < dongs> http://i.imgur.com/b5q3s.gif
2012-09-12T12:24:25 < zyp> dekar_, huh, how do you get 59 GB of swap from 32 GB of page outs?
2012-09-12T12:24:49 < Laurenceb> https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-ash3/527012_10151091922546553_636796050_n.jpg
2012-09-12T12:24:54 < Laurenceb> thats the UK dumbass
2012-09-12T12:25:18 < Laurenceb> us doesnt have a "parliament"
2012-09-12T12:25:22 < Laurenceb> or jeremy kyle
2012-09-12T12:25:45 < zyp> I thought page outs were a counter of total amount of data committed to swap, but I don't see how amount of swap used can be higher than amount of data written to swap
2012-09-12T12:26:17 < Laurenceb> vasectomy?
2012-09-12T12:26:19 < Laurenceb> lol
2012-09-12T12:36:16 < karlp> dekar_: your dropbox link is a 404?
2012-09-12T12:39:04 < prog> hello
2012-09-12T12:39:11 < prog> hi dekar_ glad to see you here
2012-09-12T12:40:11 < zyp> dekar_ got a friend?
2012-09-12T12:41:03 < prog> heh
2012-09-12T12:42:16 < prog> well.. any pointer to a dual ADC board to use with the STM32 Discovery? ~1MSPS is fine
2012-09-12T12:43:40 < karlp> what does "dual adc board" mean, andwith which discovery board?
2012-09-12T12:44:01 < karlp> most of the stm32 parts on discovery boards have dual adcs on board...
2012-09-12T12:44:30 < prog> the F4
2012-09-12T12:53:46 < karlp> that definitely has dual adcs on board.
2012-09-12T12:54:33 < dongs> even the shittiest F103 parts have triple adc..
2012-09-12T12:54:41 < dongs> or at least double, i forget
2012-09-12T12:54:44 < dongs> I know f4 has triple for sure
2012-09-12T12:54:52 < karlp> f100 only has one :)
2012-09-12T12:54:53 < dongs> cause im using it
2012-09-12T12:56:06 < Laurenceb> f103cbt6 has double
2012-09-12T12:56:52 < karlp> so prog, what are you really asking then?
2012-09-12T12:58:02 < karlp> cos even the single adc on the slowest cheapest f100 are 1mhz sampling rate,
2012-09-12T13:00:27 < dongs> maybe hes gonna make a leet-o-scope
2012-09-12T13:00:59 < dongs> https://dl.dropbox.com/u/25091878/Photo%20Sep%2009%2C%2010%2027%2040%20AM.jpg
2012-09-12T13:00:59 < dongs> https://dl.dropbox.com/u/25091878/Photo%20Sep%2009%2C%2010%2025%2042%20AM.jpg
2012-09-12T13:01:33 < dongs> $999
2012-09-12T13:01:39 < dongs> or $85 from china
2012-09-12T13:02:37 < dongs> $20 tops in parts, either way.
2012-09-12T13:02:52 < dongs> i was hoping usbeedx would at least have fpga on logic part
2012-09-12T13:02:59 < dongs> but its same garbage, directly going to FX2
2012-09-12T13:03:02 < dongs> fucking pathetic
2012-09-12T13:03:05 < karlp> what's that usbee dx likes?
2012-09-12T13:03:10 < karlp> oh, typed to slow...
2012-09-12T13:04:02 < dongs> http://www.usbee.com/dx.html these wankers want $999 foro it
2012-09-12T13:04:07 < dongs> chinaclone is 85 w /probes and all
2012-09-12T13:04:19 < dongs> at 85 its *almost* a ripoff, at 999... whew.
2012-09-12T13:05:08 < karlp> I've just got a rigol 1102, which is fine, but I sometimes would like the automagical decoding that you guys all seem to have.
2012-09-12T13:05:24 < dongs> http://www.usbee.com/products.htm
2012-09-12T13:05:36 < dongs> go ahead and get that dx clone. it'll do the job, as long as you dont need to sample any digital > 24mhz
2012-09-12T13:06:14 < karlp> that's fine, f100 only goes to 24mhz clock :)
2012-09-12T13:06:58 < karlp> how do you know which clone works and whcih clones don't?
2012-09-12T13:07:06 < karlp> do the clones use the software from usbee?
2012-09-12T13:07:19 < karlp> the software makes the hardware normally.
2012-09-12T13:08:02 < dongs> yes
2012-09-12T13:08:08 < dongs> infact they just link to usbee.com
2012-09-12T13:08:09 < dongs> for software
2012-09-12T13:08:09 < dongs> lol
2012-09-12T13:08:12 < karlp> heh
2012-09-12T13:08:38 < dongs> http://www.hotmcu.com/xzl-studio-dx-mixed-signal-oscilloscope-logic-analyzer-p-13.html?cPath=3_26
2012-09-12T13:08:42 < dongs> this is the shit me and my pal have
2012-09-12T13:08:52 < dongs> the case is actually nice milled aluminum
2012-09-12T13:08:57 < dongs> i expected it to be plastic
2012-09-12T13:10:20 < Laurenceb> http://www.hotmcu.com/cp2102-module-usb-to-33v-ttl-p-35.html?cPath=8
2012-09-12T13:10:21 < Laurenceb> nice
2012-09-12T13:11:50 < karlp> dongs so those pcbs you just linked to, are those from teh einsize of your xzl clone? or inside a real one?
2012-09-12T13:12:35 < Laurenceb> http://www.hotmcu.com/stm32f4discovery-evaluation-development-board-p-31.html?cPath=1_20
2012-09-12T13:12:37 < Laurenceb> lol
2012-09-12T13:12:42 < Laurenceb> the j-link is cheap
2012-09-12T13:16:03  * karlp giggles.
2012-09-12T13:16:24 < karlp> hotmcu reminds me of one of the "cheap" groceries here.
2012-09-12T13:16:29 < karlp> some things are much much cheapers
2012-09-12T13:16:34 < karlp> other things are way more expensive
2012-09-12T13:17:48 < zyp> clones doesn't need to recover software development costs, so it's obvious that they are cheap :p
2012-09-12T13:19:04 < dongs> karlp: clone
2012-09-12T13:20:31 < dongs> i also ordered http://www.hotmcu.com/usbminijtag-jlink-ob-jtagswd-emulator-with-usb-interface-p-29.html?cPath=3_25
2012-09-12T13:20:34 < dongs> its on the way
2012-09-12T13:20:38 < dongs> gonna try it out.
2012-09-12T13:20:50 < dongs> should be flashable with CMSIS-DAP once itsout.
2012-09-12T13:23:56 < Laurenceb> how does CMSIS-DAP use "no usb drivers"  ?
2012-09-12T13:24:16 < zyp> ntfreak said it's using HID
2012-09-12T13:24:50 < Laurenceb> ah
2012-09-12T13:24:59 < dongs> sounds a bit slow
2012-09-12T13:25:11 < zyp> sounds like a dumb windowsism
2012-09-12T13:25:20 < dongs> REALLY NOW.
2012-09-12T13:25:29 < zyp> :)
2012-09-12T13:26:16 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-12T13:26:34 < zyp> I mean, what other advantage does HID provide for stuff like that except for being able to access it from userspace in windows without any inf files or anything?
2012-09-12T13:27:07 < dongs> shrug
2012-09-12T13:27:53  * Laurenceb is using CDC-ACM on F4
2012-09-12T13:27:59 < dongs> no youre not
2012-09-12T13:28:02 < dongs> you copypasted that shti
2012-09-12T13:28:04 < dongs> from somewehre
2012-09-12T13:28:07 < Laurenceb> and
2012-09-12T13:28:14 < Laurenceb> what wrong with that
2012-09-12T13:28:35 < Laurenceb> i just want USB terminal
2012-09-12T13:28:44 < dongs> whose vid/pid did you steal
2012-09-12T13:28:47 < dongs> I didnt  feel like writing a inf
2012-09-12T13:28:55 < dongs> so i borrowed vid/pid from tarduino 2560
2012-09-12T13:29:02 < Laurenceb> the stuck stuff in chibios
2012-09-12T13:29:04 < dongs> oh
2012-09-12T13:29:06 < dongs> lal
2012-09-12T13:29:08 < Laurenceb> think its st
2012-09-12T13:29:10 < dongs> chibifail
2012-09-12T13:30:06 < dongs> http://www.chibios.org/dokuwiki/lib/exe/fetch.php?cache=&media=logo.png
2012-09-12T13:30:10 < dongs> thats not animu
2012-09-12T13:31:11 < zyp> maybe I should rename my stuff to «ponyos» so I can compete
2012-09-12T13:31:18 < zyp> would be a good match for dongs' ponyboard
2012-09-12T13:31:33 < dongs> http://images.bbgsite.com/news/2010/04/13/2000_1.jpg i expected somehign along these lines
2012-09-12T13:47:29 < ntfreak> zyp: very dumb to use hid
2012-09-12T13:48:05 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-12T13:48:58 < ntfreak> ther are no advantages - infact quite the opposite - http://www.linuxhacker.org/ezmlm-browse/index.cgi?list=gnupic&cmd=showthread&month=201208&page=1&threadid=fklgfnhpndilldgihcgd
2012-09-12T13:51:38 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-12T13:52:20 < dongs> nice wall of text
2012-09-12T13:52:48 < ntfreak> aswell as the slow down due to CMSIS-DAP over HID, also one of the reasons why i do not use virtual com
2012-09-12T13:53:09 < dongs> so is cmcsis dap opensores
2012-09-12T13:53:15 < dongs> do they have F1 version yet?
2012-09-12T13:53:21 < dongs> i wanna try running it on that jlinkob once it shows up
2012-09-12T13:55:23 < ntfreak> you can get the src now - https://silver.arm.com/browse/CMSISDAP
2012-09-12T13:55:40 < dongs> ya i dont have that shit, im not a hacker
2012-09-12T13:56:41 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 244 seconds]
2012-09-12T13:57:55 < Laurenceb> Service unavailable
2012-09-12T13:58:45 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-12T13:59:16 < zyp> «including Windows, starting with Windows 8»
2012-09-12T13:59:19 < zyp> nice
2012-09-12T13:59:38 < ntfreak> guess they are doing some work - try later
2012-09-12T13:59:49 < zyp> that's probably the first good thing I've heard about windows 8 :p
2012-09-12T14:00:26 < dongs> doing what work
2012-09-12T14:00:33 < dongs> that page works for me
2012-09-12T14:03:05 < prog> back
2012-09-12T14:03:16 < prog> I got disconnected..
2012-09-12T14:03:19 < prog> <karlp> so prog, what are you really asking then? <-
2012-09-12T14:03:35 < prog> I need "a clean" signal for SDR
2012-09-12T14:04:17 < prog> in fact, I need to sample a complex signal (analytic, real & imaginary) at 16bit @ 1+MSPS
2012-09-12T14:21:05 < Laurenceb> im not sure but i think you could have adc1 and 2 aligned
2012-09-12T14:25:37 < dongs> why not just use the right tools for hte job
2012-09-12T14:25:43 < dongs> and get a external 16bit+ ADC
2012-09-12T14:25:48 < dongs> and tarnsfer by SPI or someshit
2012-09-12T14:26:05 -!- tech2077 [~tech2077@75.53.137.201] has joined ##stm32
2012-09-12T14:26:07 < dongs> something like ADS1256T
2012-09-12T14:26:31 < dongs> eh i think that one is single channel/muxed though
2012-09-12T14:26:32 < dongs> but you get the idea
2012-09-12T14:26:45 < zyp> well, that was what he was asking about in the first place
2012-09-12T14:26:53 < dongs> was it?
2012-09-12T14:26:59 < zyp> 11:42:16 < prog> well.. any pointer to a dual ADC board to use with the STM32 Discovery? ~1MSPS is fine
2012-09-12T14:27:02 < dongs> oh
2012-09-12T14:27:14 < dongs> of course noone understood that
2012-09-12T14:29:07 < prog> that's it. I need an external ADC board
2012-09-12T14:31:37 -!- tech2077 [~tech2077@75.53.137.201] has quit [Ping timeout: 240 seconds]
2012-09-12T14:32:55 < dongs> prog: troll around analog.com or ti for evalboards of their ADCs.
2012-09-12T14:33:02 < dongs> for sure they will have a cheap one to do what you want
2012-09-12T14:33:13 -!- daku is now known as DaKu
2012-09-12T14:33:33 < prog> I'll look at them
2012-09-12T14:33:45 < prog> thank yo
2012-09-12T14:33:46 < prog> u
2012-09-12T14:39:34 < dongs> http://www.digikey.com/product-search/en?x=15&y=18&lang=en&site=us&KeyWords=EVAL-AD7356EDZ
2012-09-12T14:39:49 < dongs> only12bit but im not really sure how useful 16bit adc is gonan be even in SDR
2012-09-12T14:44:30 <+dekar_> karlp, https://dl.dropbox.com/u/28467113/osx_fail_swap2.png
2012-09-12T14:45:43 < prog> dynamic range
2012-09-12T14:46:07 <+dekar_> Laurenceb, also yeah my laptops name was vasectomy
2012-09-12T14:46:36 < dongs> prog: you arent gonna find 16bit and 2msps
2012-09-12T14:46:39 < BrainDamage> with SDR you often have the problem that you get an in-band strong signal that drastically reduces your dynamic range of your adc
2012-09-12T14:47:02 < BrainDamage> called blocker
2012-09-12T14:47:14 < Laurenceb> you mean 2x16bit?
2012-09-12T14:47:20 < prog> yep
2012-09-12T14:47:21 < Laurenceb> or 2x8bit == 16?
2012-09-12T14:47:23 < Laurenceb> oh
2012-09-12T14:47:31 < Laurenceb> you wont find that on stm32
2012-09-12T14:47:33 -!- tech2077 [~tech2077@75.53.130.216] has joined ##stm32
2012-09-12T14:47:34 < dongs> http://www.digikey.com/product-search/en?x=19&y=11&lang=en&site=us&KeyWords=AD9269-20EBZ
2012-09-12T14:47:38 < Laurenceb> you can use parallel
2012-09-12T14:47:39 < dongs> Laurenceb: he wants external board.
2012-09-12T14:47:40 < prog> 2 x 16bit x 2MSPS (or at least 1MSPS)
2012-09-12T14:47:41 < dongs> but it wont be cheap
2012-09-12T14:47:49 < Laurenceb> with gpio
2012-09-12T14:47:52 < Laurenceb> and dma
2012-09-12T14:48:22 < Laurenceb> but is 12bit really so bad?
2012-09-12T14:48:23 < BrainDamage> AD7641 has 2 MSPs at 18 bit res
2012-09-12T14:49:24 < Laurenceb> nice
2012-09-12T14:49:34 < Laurenceb> 18bit parallel is a lot of gpio tho
2012-09-12T14:49:53 < zyp> esp. if it's 2x18
2012-09-12T14:50:12 < dongs> BrainDamage: its not dual though
2012-09-12T14:50:22 < BrainDamage> he could use 2 and sync the conversion
2012-09-12T14:50:25 < dongs> sure
2012-09-12T14:50:43 < dongs> there's only 3 dual-channel 16bit ones
2012-09-12T14:50:45 < dongs> and they're all >= 20msps
2012-09-12T14:50:49 < dongs> and only output parallel
2012-09-12T14:51:02 < dongs> and they're all > $40 heh
2012-09-12T14:51:30 < dongs> http://www.digikey.com/product-search/en?x=0&y=0&KeyWords=EVAL-AD7357EDZ a bargain if you dont mind 14bi
2012-09-12T14:51:33 < dongs> bit
2012-09-12T14:51:54 < dongs> i guess its same shit i linked before but 14bit version
2012-09-12T14:52:33 < Laurenceb> wait for F3 and oversample the 12bit adc?
2012-09-12T14:53:18 < dongs> :)
2012-09-12T14:53:25 -!- tech2077 [~tech2077@75.53.130.216] has quit [Ping timeout: 246 seconds]
2012-09-12T14:53:47 < dongs> http://www.analog.com/en/mems-sensors/mems-gyroscopes/adxrs649/products/product.html attn Laurenceb
2012-09-12T14:56:37 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-12T14:56:38 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-12T14:56:38 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-12T14:58:02 < dongs> lol renesas is still trying to sell their warez
2012-09-12T14:58:04 < dongs> 32bit RX210
2012-09-12T14:58:07 < dongs> yawnn
2012-09-12T14:58:49 < dongs> http://www.digikey.com/Web%20Export/hp/en/tiles/adi-npe-september.jpg
2012-09-12T14:59:35 < BrainDamage> not sure if mouser is trolling: http://eu.mouser.com/framfordummies/
2012-09-12T15:00:29 < dongs> LOL
2012-09-12T15:01:04 < dongs> signed up
2012-09-12T15:01:12 < Bird|lappy> BrainDamage, no, they are not trolling actually
2012-09-12T15:01:23 < Bird|lappy> FRAM in this case = ferroelectric random access memory :)
2012-09-12T15:01:30 < BrainDamage> I know what a fram is
2012-09-12T15:01:33 < dongs> im sure he knows what htat is
2012-09-12T15:01:34 < Bird|lappy> ah. :)
2012-09-12T15:01:55 < dongs> BrainDamage: thanks for heads up. now i will finally receive some useful spam in my mailbox
2012-09-12T15:01:59 < dongs> instead of 10 hobbyking flyers
2012-09-12T15:02:18 < BrainDamage> heh
2012-09-12T15:02:29 < dongs> I have people send shit from HK to my office to forward to them,,, and apparnetly that adds my address (again) to the spam list for hteir mail magazine...
2012-09-12T15:02:39 < dongs> so today I received a fucking pile of their spam mags
2012-09-12T15:05:14 -!- tech2077 [~tech2077@adsl-75-53-134-145.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-12T15:20:01 -!- tech2077 [~tech2077@adsl-75-53-134-145.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-12T15:22:06 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-12T15:48:06 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 240 seconds]
2012-09-12T15:48:29 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 260 seconds]
2012-09-12T15:48:39 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has joined ##stm32
2012-09-12T15:48:40 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925075702.dsl.bell.ca] has quit [Changing host]
2012-09-12T15:48:40 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-12T15:51:08 < karlp> are they going to actually ship me a deadtree fram for dummies?
2012-09-12T15:51:08 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T15:51:12 < karlp> can't it just be an ebook?
2012-09-12T15:51:52 < karlp> they're probably just going to be hyping up the fram msp430s anyway right?
2012-09-12T15:52:59 < BrainDamage> deadtree
2012-09-12T15:53:28 < HTT-Bird> speaking of MSP430s...I wonder when TI will put a date on the MSP430FR58xx parts :)
2012-09-12T15:53:33 < BrainDamage> and yes, it's basically a marketing plan, I'd be curious if it has any decent infos on the internal structure tough
2012-09-12T15:54:03 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-5.pptp.stw-bonn.de] has joined ##stm32
2012-09-12T15:54:07  * HTT-Bird is pondering a FRAM-based counter using those FRAM state saver ICs from Ramtron
2012-09-12T16:05:52 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 246 seconds]
2012-09-12T16:08:48 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T16:24:01 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 252 seconds]
2012-09-12T16:26:43 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T16:45:17 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-12T16:45:18 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-12T16:45:18 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-12T16:45:30 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-12T16:49:54 < Laurenceb> http://www.theregister.co.uk/2012/09/12/iphone_five_reasons/
2012-09-12T16:52:27 < Laurenceb> An Apple user, Fry has cruelly been described as "a stupid person's idea of a clever person"
2012-09-12T16:52:33 < Laurenceb> lolling
2012-09-12T16:59:14 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 272 seconds]
2012-09-12T17:02:03 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T17:02:42 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 265 seconds]
2012-09-12T17:02:51 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-12T17:03:31 < Laurenceb> http://hackaday.com/
2012-09-12T17:03:37 < Laurenceb> pony version is called ofr
2012-09-12T17:06:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-12T17:06:28 < Erlkoenig> Laurenceb: ... ooold?
2012-09-12T17:07:17 < karlp> I think he might mean the arm board underneath, no thte white noise thing
2012-09-12T17:07:49 < Laurenceb> yes
2012-09-12T17:08:01 < Laurenceb> *for
2012-09-12T17:08:29 < Laurenceb> http://i.imgur.com/L29UB.png
2012-09-12T17:09:04 < Erlkoenig> pony? why ponies with arm?
2012-09-12T17:10:44 < karlp> my little pony, keep up Erlkoenig  :)
2012-09-12T17:10:55 < Erlkoenig> but why :D
2012-09-12T17:11:14 < Laurenceb> because 4chan
2012-09-12T17:12:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-12T17:12:34 < Erlkoenig> why did i even ask?
2012-09-12T17:13:00 < jon1012> if people are interested, I made a gpio.py for stm32 (tested on f4 but should work on others) gpios on python on a chip
2012-09-12T17:13:08 < jon1012> (speaking of ponies :p)
2012-09-12T17:16:24 < karlp> is that something that you put a custom firmware on the device, and then python knows how to communicate with it over serial?
2012-09-12T17:16:27 < karlp> or what?
2012-09-12T17:16:57 < jon1012> no, you code in python
2012-09-12T17:17:05 < jon1012> there is a vm in the micro controller
2012-09-12T17:17:16 <+izua> so, can that python thing interact with the peripherals?
2012-09-12T17:17:21 < jon1012> yeah
2012-09-12T17:17:30 < karlp> p14p?
2012-09-12T17:17:32 < jon1012> yeah
2012-09-12T17:17:40 <+izua> i had a very simple one that could do peek and poke
2012-09-12T17:17:52 <+izua> best experimenting rig i've found yet ><
2012-09-12T17:17:59 <+izua> oh
2012-09-12T17:18:05 <+izua> you write python on the chip
2012-09-12T17:18:08 <+izua> neat
2012-09-12T17:18:22 < karlp> have you thought about, you know, updating the p14p wiki, or writing it up?
2012-09-12T17:19:00 < jon1012> I'll make a sprint and workshop at pycon fr (starts tommorrow and ends on sunday)
2012-09-12T17:19:02 < jon1012> on that
2012-09-12T17:19:15 < jon1012> so I'll try to document and release all the code
2012-09-12T17:19:22 < Laurenceb> anyone come across a simple scanf implementation for embedded?
2012-09-12T17:19:31 < jon1012> (right now works perfectly on an stm32f4 discovery board I have on my desk)
2012-09-12T17:19:52 < jon1012> with code like :
2012-09-12T17:19:53 < jon1012> from gpio import pin_mode, digital_write, OUTPUT, INPUT, HIGH, LOW
2012-09-12T17:19:58 < jon1012> pin_mode("E", 7, OUTPUT)
2012-09-12T17:20:02 < jon1012> digital_write("E", 7, HIGH)
2012-09-12T17:20:03 < jpa-> oh, p14p is no longer GPL \o/
2012-09-12T17:20:25 < karlp> website still says it is.
2012-09-12T17:20:33 <+izua> do the peripherals work?
2012-09-12T17:20:46 <+izua> timer & pwm generators, spi, sdio, etc
2012-09-12T17:20:53 < jon1012> right now I mapped only the gpios :) I'll have to map the rest
2012-09-12T17:21:13 < jpa-> karlp: https://code.google.com/p/python-on-a-chip/source/browse/LICENSE
2012-09-12T17:21:32 < karlp> someone should update the p14p gogole code page itself then.
2012-09-12T17:21:46 < karlp> if the project details page says gpl, I don't go and look at the license file in the source to double check.
2012-09-12T17:26:29 < jpa-> indeed
2012-09-12T17:31:37 -!- DaKu is now known as daku
2012-09-12T17:42:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-12T17:56:48 -!- daku is now known as DaKu
2012-09-12T17:59:33 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-12T18:01:21 < Laurenceb> anyone know how adc clk is configured in chibios?
2012-09-12T18:02:32 < Laurenceb> http://forum.chibios.org/phpbb/viewtopic.php?f=3&t=526
2012-09-12T18:02:37 < Laurenceb> interesting
2012-09-12T18:13:16 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-12T18:20:08 < Rickta59> is there any advantage to using -mfloat-abi=softfp when you are using a m4 cortex chip?
2012-09-12T18:21:41 < jpa-> yes, if your code is too small and fast otherwise
2012-09-12T18:22:22 < Rickta59> i hate it when the code is too small
2012-09-12T18:22:27 < zippe> Rick: if you have to link to a library that you can't recompile
2012-09-12T18:22:50 < Rickta59> ic/ i was trying to figure out why if you had all the source you would use that options
2012-09-12T18:23:29 < Rickta59> thanks zippe
2012-09-12T18:23:50 < dongs> you cant mix soft and hard fp?
2012-09-12T18:23:58 < dongs> cant you just write a wrapper function?
2012-09-12T18:24:33 < dongs> hm, i guess thre'd hve eeto be some way to transfer fp registers to ram/normal regs
2012-09-12T18:24:47 < dongs> Laurenceb: chibios forum is fucking dead
2012-09-12T18:25:47  * Laurenceb is trying to make F4 adc timing match his F1
2012-09-12T18:26:05 < Laurenceb> they changed sample times and apb clking
2012-09-12T18:26:38 < zippe> dongs: you can certainly write thunks
2012-09-12T18:26:54 < zippe> But there is no way to mark e.g. a function prototype with a call flavor
2012-09-12T18:27:11 < zippe> e.g. no moral equivalent of 'extern "Pascal"'
2012-09-12T18:27:35 < dongs> attn Laurenceb http://bcas.tv/paste/results/G4Q5sF55.html
2012-09-12T18:28:19 < dongs> zippe: right. gcc doesnt have some hack for marking a function as softfp, so it loads any arguments into normal regs/stack?
2012-09-12T18:28:41 < dongs> seems with them having a hack for just a bout everything, they'd think of that.
2012-09-12T18:38:21 < zyp> can't find any function attribute for that with a quick google search
2012-09-12T18:39:41 < karlp> Laurenceb: when in doubt, add more xml!
2012-09-12T18:39:50 < karlp> (I can't say I recommend that approach)
2012-09-12T18:41:20 < Laurenceb> this is so annoying
2012-09-12T18:41:32 < Laurenceb> im going to have to rewrite my adc code completely
2012-09-12T18:42:27 < zyp> you mean find some other adc code to steal?
2012-09-12T18:42:49 < Laurenceb> nope
2012-09-12T18:42:59 < Laurenceb> its annoying to find prime factors
2012-09-12T18:43:12 < Laurenceb> ill have to have another search for something that works well
2012-09-12T18:45:39 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-12T18:48:08 -!- barthess [~barthess@178.154.68.91] has joined ##stm32
2012-09-12T18:49:13 < Laurenceb> unless i trigger adc off a timer
2012-09-12T18:49:22 < Laurenceb> that might simplify things
2012-09-12T18:54:55 < Laurenceb> as ive got lots of timers to use now im using F4
2012-09-12T18:55:21 < zippe> That's what 6 & 7 are for
2012-09-12T18:55:26 < Laurenceb> previously i had to find factors so that everything syncronised nicely with sysclk
2012-09-12T18:55:34 < Laurenceb> - on the F1
2012-09-12T19:02:02 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 244 seconds]
2012-09-12T19:07:05 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-12T19:07:06 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-12T19:09:39 -!- dekar_ [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-12T19:12:01 -!- dekar [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has joined ##stm32
2012-09-12T19:12:03 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-12T19:13:43 -!- dekar [~dekar@dslb-088-069-145-088.pools.arcor-ip.net] has quit [Client Quit]
2012-09-12T19:16:56 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-12T19:33:27 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-12T19:38:16 < jpa-> karlp: apparently he has some secret reason to keep it secret :)  https://groups.google.com/forum/?hl=en&fromgroups=#!topic/python-on-a-chip/xc29hCh0RYA
2012-09-12T19:39:19 < zyp> is v10 newer or older code?
2012-09-12T19:39:33 < jpa-> older, it seems
2012-09-12T19:41:43 < Laurenceb> zippe: 6&7 cant trigger adc  :(
2012-09-12T19:42:56 < zyp> no, they are for dac
2012-09-12T19:43:26 < zyp> adc can be triggered by 1-5 and 8
2012-09-12T19:56:03 -!- claude_work is now known as claude
2012-09-12T20:03:39 -!- qyx_ [~qyx@krtko.org] has quit [Read error: Connection reset by peer]
2012-09-12T20:15:53 -!- oPossum5150 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-12T20:22:57 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-12T20:27:52 < emeb> zyp: updating firmware in my bmpm got my F405 board flashing reliably now.
2012-09-12T20:28:03 < zyp> :)
2012-09-12T20:28:32  * emeb wonders when my bmpm was built
2012-09-12T20:29:43 < karlp> the dim dark past.
2012-09-12T20:32:54 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-12T20:32:59 < zippe> emeb: is it one of the minis,  or the old-school one with the big 20-pin ARM connector?
2012-09-12T20:33:04 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-12T20:33:29 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 244 seconds]
2012-09-12T20:35:50 < emeb> zippe: it's a mini
2012-09-12T20:36:29 < emeb> bought it here: http://thequadshot.com/products/black-magic-probe-mini
2012-09-12T20:39:48 < emeb> heh - all iPhone accessory makers are crapping their drawers about now.
2012-09-12T20:39:51 < zippe> Then it was built in the last month or two, tops
2012-09-12T20:39:59 < zippe> emeb: you have no idea
2012-09-12T20:40:02 < emeb> *back to the drawing boards*
2012-09-12T20:40:37 < dongs> emeb: is the trash announced yet
2012-09-12T20:40:47 < emeb> dongs: as we speak
2012-09-12T20:40:54 < emeb> http://www.engadget.com/2012/09/12/apple-iphone-5-liveblog/
2012-09-12T20:41:30 < dongs> waht hte fuck is this
2012-09-12T20:42:40 < dongs> > Got old docks? "We're creating a bunch of accessories to help you with that. loool
2012-09-12T20:42:50 < emeb> bring $$$
2012-09-12T20:43:16 < dongs> looks worthless
2012-09-12T20:43:22 < dongs> apple has been dead for over a decade
2012-09-12T20:43:37 < zyp> it's the first cortex-A15-based phone in the market
2012-09-12T20:43:49 < zyp> they went dual A15 rather than quad A9 like everybody else
2012-09-12T20:43:59 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T20:44:03 < emeb> should be interesting to see how it benchmarks.
2012-09-12T20:44:28 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 252 seconds]
2012-09-12T20:44:52 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-12T20:45:10 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-12T20:45:28 < karlp> no point in quad core, iOS is single tasking right? ;)
2012-09-12T20:47:23 < zyp> :p
2012-09-12T20:47:56 -!- barthess [~barthess@178.154.68.91] has quit [Quit: Leaving.]
2012-09-12T20:48:15 < zyp> well, if you put it like «current smartphone workloads doesn't parallelize well», you'd be right
2012-09-12T20:48:32 < dongs> my assdroid phone had some runaway lunix process yesterday
2012-09-12T20:48:41 < dongs> i wake up and its in suspend mode cuz battery was at 1%
2012-09-12T20:48:54 < dongs> was like 80% when i went to sleep
2012-09-12T20:49:10 < zippe> zyp: it's not an Q15
2012-09-12T20:49:12 < emeb> dongs: get a windows phone. I'm sure it will work much better.
2012-09-12T20:49:13 < zippe> er, A15
2012-09-12T20:49:36 < zyp> zippe, anandtech is reporting that it is, blame them :)
2012-09-12T20:49:51 < zippe> zyp: Anand is in for a nasty shock
2012-09-12T20:49:51 < dongs> analtech
2012-09-12T20:49:58 < dongs> zippe: what is it then
2012-09-12T20:50:10 < zippe> dongs: Something else.
2012-09-12T20:50:12 < zyp> so it's quad A9 then?
2012-09-12T20:50:12 < dongs> <deadsteve> we didnt bother upgrading the processor, bitches
2012-09-12T20:50:22 < zippe> Nope.
2012-09-12T20:50:34 < zyp> dongs, they claim it's twice as fast as previous
2012-09-12T20:50:42 < dongs> pfft.
2012-09-12T20:50:47 < dongs> twice as fast at doin nothing
2012-09-12T20:50:54 < dongs> maybe they upgraded to armcc from gcc-aids
2012-09-12T20:50:55 < zippe> If you ask "is it X", where X is anything you have previously heard about, the answer is "no"
2012-09-12T20:51:17 < emeb> home-made magic sauce
2012-09-12T20:51:38 < dongs> steve rised from teh grave to design apple's own arm soc
2012-09-12T20:51:50 < dongs> the point is, apple lost relevance years ago
2012-09-12T20:52:13 < dongs> (not that it ever had any)
2012-09-12T20:52:23 < dongs> even assdroid is beating them in sales, and assdroid is fucking garbage
2012-09-12T20:52:57 < dongs> "iPhone 5 is now the best phone we've ever made, and what would be a great phone without a killer video?"
2012-09-12T20:53:50 < zippe> dongs: who cares about sales-into-the-chanel numbers?  Look at where the profits are.
2012-09-12T20:54:25 < dongs> apple makes profit cuz their users are dumbfucks.
2012-09-12T20:54:31 < dongs> you're saying it like its something new
2012-09-12T20:54:50 < dongs> thats always been their way of things
2012-09-12T20:57:11 < dongs> so zippe, anything magical in new apple cpu?
2012-09-12T20:57:16 < zyp> zippe, so it's some arm-compatible design similar to the qualcomm stuff then?
2012-09-12T20:57:23 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-12T20:57:49 < TitanMKD> hi
2012-09-12T20:58:06 < BrainDamage> why so much fuss about a phone?
2012-09-12T20:58:11 < dongs> no idea
2012-09-12T20:58:17 < emeb> what else is going on?
2012-09-12T20:58:29 < dongs> Jony is back, talking about the manufacturing process, how they create the textured ring vs. the almost mirrored chamfered edge.
2012-09-12T20:58:38 < dongs> this is the kinda shit faggots with macbooks in starbucks talk about
2012-09-12T20:58:55 < dongs> "hey bro, i've got this new iphone5 with a CHAMFERED EDGE TEXTURED RING"
2012-09-12T20:58:56 < emeb> arteests.
2012-09-12T20:59:03 < dongs> "wank wank wank wank wank wankw ank"
2012-09-12T20:59:26 < emeb> The variances from product to product we now measure in microns." Cameras actually take images of each case and select the appropriate back plate to fit.
2012-09-12T20:59:36 < emeb> so much for interchangable parts
2012-09-12T21:00:00 < emeb> read as "repair parts will cost you more"
2012-09-12T21:00:14 < dongs> didnt thnk apple stuff is repairable
2012-09-12T21:00:47 < dongs> lol @ iphone 4/4s now free with contract
2012-09-12T21:00:50 < emeb> Oh they are. The glass backplates on the current iPhones shatter all the time. They don't charge too much to replace them.
2012-09-12T21:01:20 < dongs> emeb, youre supposed to buy an artfag case for it
2012-09-12T21:02:01 < HTT-Bird> *chuckles*
2012-09-12T21:02:05  * HTT-Bird doesn't have a smartphone at all
2012-09-12T21:02:14  * emeb doesn't either
2012-09-12T21:02:24  * emeb is a luddite
2012-09-12T21:02:35 < HTT-Bird> can't find a carrier that won't tell us to "go pay out our ears" when it comes to data plans
2012-09-12T21:03:53 < BrainDamage> here it costs me 6€/month for 1k sms and 1GB traffic
2012-09-12T21:04:06 < BrainDamage> which is reasonable, mostly for the sms part
2012-09-12T21:04:23 < BrainDamage> ofc it costs them almost nothing, so reasonable is relative
2012-09-12T21:05:19 < dongs> 11:06AM PDT - Integrated Facebook likes into every product page in the new store
2012-09-12T21:05:20 < dongs> that's it
2012-09-12T21:05:25 < dongs> preordering itrash5 asap
2012-09-12T21:05:57 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-09-12T21:08:28 < dongs> zippe: "A6 cpu"?
2012-09-12T21:08:53 < dongs> o right
2012-09-12T21:08:56 < dongs> the shitty apple marketing name
2012-09-12T21:09:09 < dongs> haha does anyone acutalyl seriosuly use itunes
2012-09-12T21:09:27 < TitanMKD> hehe yes a revolution A6 with just 100Mhz more than the previous one ;)
2012-09-12T21:09:27 < emeb> if you've got an iPhone or iPad you pretty much have to
2012-09-12T21:09:39 < dongs> good thing I dont
2012-09-12T21:10:18 < emeb> TitanMKD: a revolution that comes in tiny steps
2012-09-12T21:11:15 < TitanMKD> haha like ever their A6 is just an IP bought to ARM and customized just a bit and 1 year obsolete compared to samsung stuff ;)
2012-09-12T21:11:40 < dongs> zippe seems to claim it isnt
2012-09-12T21:14:44 -!- amitofu [~str@cannabis.dataforce.net] has joined ##stm32
2012-09-12T21:14:52 < dongs> ^ stoner
2012-09-12T21:16:23 < amitofu> lol
2012-09-12T21:19:10 < zippe> TitanMKD: obsolete, eh?  First mass-market A9 is "1 year obsolete"?
2012-09-12T21:21:00 < TitanMKD> zippe obsolete compared to samsung stuff already since lot of months ;)
2012-09-12T21:21:18 < TitanMKD> zippe anyway Apple have no fab house so they are always behind samsung ...
2012-09-12T21:21:21 < emeb> heh - "If you ever wonder what music people who don't really like music listen to, just check out an itunes presentation"
2012-09-12T21:22:29 < zyp> TitanMKD, you sound like you're just full of shit now
2012-09-12T21:23:06 < TitanMKD> ?
2012-09-12T21:23:25 < jpa-> yeah, just like nvidia is always behind ATI, they are fabless anyway
2012-09-12T21:23:43 < zippe> TitanMKD: Sorry, not getting what you're saying here.  What's "obsolete"?
2012-09-12T21:23:52 < jpa-> (well, AMD)
2012-09-12T21:23:57 < BrainDamage> jpa-: arm
2012-09-12T21:24:09 < BrainDamage> even better example imo .p
2012-09-12T21:24:16 < zippe> TitanMKD: and, who cares about the specs?  From a user perspective, what matters is "does it do interesting things".  Sales numbers and profits are your guide there.
2012-09-12T21:24:32 < TitanMKD> zyp and soon Apple will loose against samsung/htc because of 4G ;)
2012-09-12T21:24:34 < zippe> TitanMKD: really, you sound like a hipster
2012-09-12T21:24:38 < jpa-> BrainDamage: sure, but they don't directly publish any mass-market products
2012-09-12T21:24:46 < zyp> zippe, specs is all I care about, because I'm not buying one?
2012-09-12T21:24:47 < zippe> TitanMKD: and I mean that in a bad way
2012-09-12T21:24:52 < zyp> -?
2012-09-12T21:25:19 < TitanMKD> i'm not against apple product i'm against they lock people in a world very closed in fact
2012-09-12T21:25:34 < zyp> TitanMKD, I'm not sure what you are saying, but you are aware that Apple just announced an LTE-capable iPhone?
2012-09-12T21:26:03 < zippe> TitanMKD: against why? And "lock people in" to what?
2012-09-12T21:26:18 < TitanMKD> zyp yes but LTE patents are owned by samsung and HTC ;)
2012-09-12T21:26:34 < HTT-Bird> what I want to see is a Cortex-M0+ on the 130nm ULP process TI's working on
2012-09-12T21:26:54 < zippe> HTT-Bird: I'd like to see it on 28LP, personally
2012-09-12T21:27:01 < zyp> :)
2012-09-12T21:28:12 < HTT-Bird> also, integrated FeRAM? :)
2012-09-12T21:28:27 < zippe> TitanMKD: Basically, I'm hearing a lot of FUD but nothing substantial.  So I guess I'm not selling AAPL today.
2012-09-12T21:28:44 < HTT-Bird> but even what they're planning to do with it so far (the MSP430FR58xx "Wolverine" parts) is exciting
2012-09-12T21:33:45 -!- ntfreak_ is now known as ntfreak
2012-09-12T21:51:09 < BrainDamage> interesting: http://www.oklabs.net/category/skype-reverse/
2012-09-12T21:57:04 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-12T21:58:13 < Laurenceb_> fuuuuk
2012-09-12T21:58:22 < Laurenceb_> there goes my weekend :(
2012-09-12T21:58:53 < Laurenceb_> tenants just managed to screw up the shower at a place i rent out
2012-09-12T21:58:57 < Laurenceb_> this shit sucks
2012-09-12T21:59:09 < emeb> Fun being a landlord.
2012-09-12T21:59:21 <+Steffanx> Ha Laurenceb_
2012-09-12T21:59:38 < emeb> Wait until they rebuild an engine on the living room carpet.
2012-09-12T21:59:39 < Laurenceb_> well im renting atm
2012-09-12T21:59:47 < Laurenceb_> and renting out my house whilst im away
2012-09-12T22:00:20 < Laurenceb_> sounds like they aether cracked the shower tray or managed to disconnect the drain
2012-09-12T22:00:26 < Laurenceb_> *either
2012-09-12T22:00:39 < emeb> ouch.
2012-09-12T22:01:13 < emeb> drain seal to pan cracked on our 2nd floor shower. Big stain on ceiling of room below.
2012-09-12T22:01:22 < Laurenceb_> yeah the ceiling is screwed
2012-09-12T22:01:31 < Laurenceb_> this has happened twice before
2012-09-12T22:01:41 < emeb> yay! you know how to fix it!
2012-09-12T22:01:51 < Laurenceb_> i wish someone make cast cast stainless steel bathroom fixtures
2012-09-12T22:02:31 < Laurenceb_> im tempted to inject polyurethane foam under the whole thing
2012-09-12T22:02:36 < Laurenceb_> thats probably a bad idea
2012-09-12T22:02:43 < emeb> that'll just make it harder to fix next time.
2012-09-12T22:03:06 < emeb> should put a trap-door in the ceiling. easy access. :P
2012-09-12T22:03:14 < Laurenceb_> really i need to rip out the whole thing and make a plywood framed tray
2012-09-12T22:03:28 < Laurenceb_> but thats about a weeks work and they will have no bathroom in the meantime
2012-09-12T22:04:11 < Laurenceb_> this stupid fibre glass stuff is about 10 times weaker than it should be
2012-09-12T22:05:11 < emeb> love it when you can hear it shift/crinkle as you step on it.
2012-09-12T22:05:38 < Laurenceb_> exactly
2012-09-12T22:05:40 <+Steffanx> Don't rent your apartment to people with 'some' over weight
2012-09-12T22:05:58 < Laurenceb_> they are both massively overweight and shower like 4 times a day each
2012-09-12T22:06:00 < Laurenceb_> nutters
2012-09-12T22:06:01 <+Steffanx> *room/whatever
2012-09-12T22:06:27 < Laurenceb_> its a house :P
2012-09-12T22:06:31 < Laurenceb_> im not that poor
2012-09-12T22:06:38 < Laurenceb_> terraced tho :P
2012-09-12T22:07:48 < emeb> so - um, if you're close enough to work on it weekends, why did you rent it out whilst living elsewhere?
2012-09-12T22:08:06 < Laurenceb_> i make more cash that way
2012-09-12T22:08:25 < Laurenceb_> and dont have to commute as far
2012-09-12T22:08:35 < emeb> commuting - blech.
2012-09-12T22:08:58 < Laurenceb_> i travel 80 miles a day atm
2012-09-12T22:09:08 < emeb> double blech
2012-09-12T22:09:32  * emeb used to have 24mi each way commute. bad enough.
2012-09-12T22:09:33 < Laurenceb_> heh
2012-09-12T22:10:02 < emeb> now I work @ home. *bliss*
2012-09-12T22:10:22 < karlp> I drive 7km when I'm too lazy to ride a bike, and it's too long.
2012-09-12T22:10:25  * Laurenceb_ is wondering if he can fit his head underneath shower tray to fit wooden reinforcements
2012-09-12T22:10:25 < karlp> hooray for small cities!
2012-09-12T22:10:58 <+Steffanx> Hooray for small countries!
2012-09-12T22:11:10 < Laurenceb_> actually... if i used a usb webcam i would need to head to be in there XD
2012-09-12T22:11:31 < Laurenceb_> *wouldn't
2012-09-12T22:11:42 < emeb> can always tell a geek - brings a computer to a carpentry job.
2012-09-12T22:11:42 < Laurenceb_> thats a little nutty
2012-09-12T22:11:45 < Laurenceb_> hehe
2012-09-12T22:12:08 < BrainDamage> and leave the webcam pointed at the shower?
2012-09-12T22:12:21 < Laurenceb_> *leave*
2012-09-12T22:12:27 < Laurenceb_> for dongs to enjoy
2012-09-12T22:12:40 < Laurenceb_> <Laurenceb_> they are both massively overweight
2012-09-12T22:14:10 <+Steffanx> Some enjoy that ..
2012-09-12T22:14:36 < Laurenceb_> http://www.bbc.co.uk/news/technology-19572820
2012-09-12T22:15:03 -!- feurig_ is now known as feurig
2012-09-12T22:15:34 < Laurenceb_> The new screen offers a 16:9 ratio
2012-09-12T22:15:44 < Laurenceb_>  existing apps would be shown with black borders until developers updated their products.
2012-09-12T22:15:47 < Laurenceb_> so much fail
2012-09-12T22:16:16 < emeb> "It's the developers fault"
2012-09-12T22:16:23 < zyp> is it really fail?
2012-09-12T22:16:32 < zyp> isn't that the sensible way to handle the situation?
2012-09-12T22:17:09 < jpa-> hmph.. apparently my TIM3 was already in use :F
2012-09-12T22:17:38 < jpa-> as nice as AFIO on STM32L is, it's tricky when you don't realize you have already used up the peripheral
2012-09-12T22:17:48 < zyp> :p
2012-09-12T22:17:58 < emeb> ooops
2012-09-12T22:18:33 < jpa-> just have to implement the encoder interface by SW then.. not that bad
2012-09-12T22:18:34 <+Steffanx> zyp .. don't you know Laurenceb_ well enough to know he just has to find something to bash?
2012-09-12T22:18:39 < zyp> even if the F1 USB core is simple as hell, it has already started giving me troubles :p
2012-09-12T22:19:25 < zyp> for some reason the endpoint registers are disregarding any writes :p
2012-09-12T22:19:42 < zyp> not sure what I'm doing wrong, there is not much to do wrong :p
2012-09-12T22:23:25 -!- claude is now known as Claude
2012-09-12T22:24:35 < jpa-> zyp: you forgot the RCC clock enable ,)
2012-09-12T22:24:51 < zyp> no, I've checked that
2012-09-12T22:25:12 < zyp> there is only one enable for USB, and the generic registers are accepting writes and reporting reset events just fine
2012-09-12T22:25:31 < zyp> I've also checked that the divider is correctly set to 48 MHz
2012-09-12T22:25:51 < emeb> bad clock stalled me for a while.
2012-09-12T22:26:41 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-12T22:26:46 < zyp> it seems to be picking up sof frames too (frame number register is updating), so it looks like it can sync to the bus just fine
2012-09-12T22:26:59 < zyp> so I don't think there are any clock issues
2012-09-12T22:30:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-12T22:38:07 -!- qyx_ [~qyx@krtko.org] has joined ##stm32
2012-09-12T22:40:22 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-12T22:45:36 < Laurenceb_> got to be more fun than shower trays
2012-09-12T22:46:09 < Laurenceb_> and adc timing issues on f4 vrs f1
2012-09-12T22:46:12 < Laurenceb_> ragggeeeee
2012-09-12T22:46:17 < Laurenceb_> so much mess to fix
2012-09-12T22:48:01 -!- Claude is now known as claude
2012-09-12T23:08:35 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-12T23:08:35 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-12T23:08:35 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-12T23:19:36 -!- mrcan_ [~mrcan@78.173.75.48] has joined ##stm32
2012-09-12T23:19:36 -!- mrcan_ [~mrcan@78.173.75.48] has quit [Changing host]
2012-09-12T23:19:36 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-12T23:23:01 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-09-12T23:27:00 -!- barthess [~barthess@178.154.68.91] has joined ##stm32
2012-09-12T23:35:35 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-12T23:35:37 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-12T23:35:37 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-12T23:35:40 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-12T23:39:14 < Laurenceb_> wohooo
2012-09-12T23:39:25 < Laurenceb_> made F4 pwm/adc work
2012-09-12T23:39:45 < Laurenceb_> wrote a matlab script to scruch the numbers until it found something that worked
2012-09-12T23:40:24 < Laurenceb_> 5 pwm channels with ofdm sample outputs at 116.213323134blerrggg
2012-09-12T23:40:27 < Laurenceb_> Hz
2012-09-12T23:46:35 < Laurenceb_> http://www.indiegogo.com/SuperAwesomeNY2012
2012-09-12T23:46:37 < Laurenceb_> lolling
2012-09-12T23:47:45 < emeb> Laurenceb_: by ofdm what do you mean?
2012-09-12T23:48:05  * emeb comes from comm background and OFDM means a modulation format for high-speed data to him.
2012-09-12T23:48:29 < emeb> involving lots of FFTs, high-speed DACs, etc.
2012-09-12T23:50:37 < Laurenceb_> yes
2012-09-12T23:50:42 < Laurenceb_> that sort of ofdm
2012-09-12T23:50:50 < Laurenceb_> using only hardware pwm
2012-09-12T23:50:51 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-09-12T23:50:59 < emeb> low sample rate?
2012-09-12T23:51:14 < Laurenceb_> yes - its actually for a "spectrometer"
2012-09-12T23:51:22 < Laurenceb_> using leds and a photodiode
2012-09-12T23:51:59 < Laurenceb_> so it can separate the signal into bins, one for each led
2012-09-12T23:52:32 < Laurenceb_> its very tricky to do with hardware pwm and adc, and keep things phase aligned
2012-09-12T23:53:01 < emeb> so you're sending a known orthogonal sequence through each different LED color and then correlating that against the received photodiode signal to extract the different components?
2012-09-12T23:53:05 < Laurenceb_> if adc phase drifts wrt timers you get weird rippling interference
2012-09-12T23:53:07 < Laurenceb_> yes
2012-09-12T23:53:16 < Laurenceb_> its all NIR and MIR leds
2012-09-12T23:53:27 < Laurenceb_> 800nm to 1.4micron
2012-09-12T23:53:49 < emeb> Makes sense.
2012-09-12T23:54:06 < emeb> Nifty technique too.
2012-09-12T23:55:15 < emeb> Not exactly OFDM though. More orthogonal sequences than frequencies. More like DS spread spectrum.
2012-09-12T23:55:43 < Laurenceb_> yeah im not using constellations
2012-09-12T23:55:53 < Laurenceb_> just take sqrt(I^2+Q^2)
2012-09-12T23:56:26 < emeb> no phase - just mag
2012-09-12T23:56:30 < Laurenceb_> constellations _might_ be possible by adjusting the timers and using the right filtering
2012-09-12T23:56:32 < Laurenceb_> yes
2012-09-12T23:57:13 < Laurenceb_> the main reason i wanted hardware only was to avoid interrupt jitter
2012-09-12T23:57:22 < Laurenceb_> which gives noise on the signal
2012-09-12T23:57:26 < emeb> yup
2012-09-12T23:57:32 < Laurenceb_> getting about 17 ENOB/Channel now
2012-09-12T23:57:47 < emeb> Not bad
2012-09-12T23:59:07 -!- barthess [~barthess@178.154.68.91] has quit [Remote host closed the connection]
2012-09-12T23:59:20 < Laurenceb_> for a 12 bit adc XD
2012-09-12T23:59:24 < emeb> what's the sample rate on the ADC?
2012-09-12T23:59:42 < Laurenceb_> 1.4Msps
2012-09-12T23:59:54 < emeb> how long are the correlation sequences?
--- Day changed Thu Sep 13 2012
2012-09-13T00:00:16 < Laurenceb_> 1/116.blerg  seconds
2012-09-13T00:01:19 < emeb> so your extra ENOB comes from integrating over that period.
2012-09-13T00:01:32 < Laurenceb_> yes
2012-09-13T00:02:43 < HTT-Bird> in a discrete-time feedback system (think say a PID loop on a uC): what impact does decreasing the sampling frequency have?
2012-09-13T00:02:46 < emeb> about 12k integrations / sample
2012-09-13T00:03:00 < HTT-Bird> does that correspond to my "dominant pole" in the Bode plot?
2012-09-13T00:03:31 < Laurenceb_> bbl
2012-09-13T00:03:42 < Laurenceb_> btw, watch bbc4 if you are in uk
2012-09-13T00:03:51 < Laurenceb_> - documentary on alan bond
2012-09-13T00:04:26 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-13T00:24:39 < Laurenceb_> http://en.wikipedia.org/wiki/Alan_Bond_%28rocket_developer%29
2012-09-13T00:31:40 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-13T00:33:13 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-13T00:41:29 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 244 seconds]
2012-09-13T01:09:11 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-13T01:10:46 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-13T01:10:49 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-13T01:11:31 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds]
2012-09-13T01:34:37 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-13T01:35:07 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 240 seconds]
2012-09-13T02:05:41 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-13T02:07:30 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-5.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-13T02:10:05 -!- DaKu is now known as daku
2012-09-13T02:17:05 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-13T02:17:22 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-13T02:44:24 -!- tech2077 [~tech2077@adsl-75-53-132-237.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-13T02:45:07 < Bird|lappy> -fdelete-null-pointer-checks is correct for a Cortex-M3, rght?
2012-09-13T02:45:50 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-13T02:46:28 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-09-13T02:48:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-13T02:49:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Operation timed out]
2012-09-13T02:52:46 < zyp> it's implied by -O2
2012-09-13T02:55:50 < zyp> -fdelete-null-pointer-checks makes gcc assume that you can't access memory address 0
2012-09-13T02:56:15 < zyp> in stm32 you can, it will be aliased to either flash or ram
2012-09-13T02:56:55 < Bird|lappy> ah.  so I need -fno-delete-null-pointer-checks?
2012-09-13T02:57:05 < zyp> the question is rather whether your application rely on it or not
2012-09-13T02:57:28 < zyp> are you in any part of your code trying to read address 0 as a valid value?
2012-09-13T02:57:49 < zyp> if so, then you need -fno-delete-null-pointer-checks
2012-09-13T02:58:13 < Bird|lappy> ah.
2012-09-13T02:58:33 < zyp> if not, then you don't need to care
2012-09-13T02:59:09 < zyp> and really, if you need to read the first word of flash or ram, you can just access 0x08000000 or 0x20000000 respectively
2012-09-13T03:00:01 < Bird|lappy> now for the fun part: I have my system_<family>.c and startup_<family>.s files...those just compile into my application, right? (the CMSIS package for my uC ships without a Makefile or anything like that)
2012-09-13T03:00:39 < zyp> probably
2012-09-13T03:01:07 < Bird|lappy> also, what is the System_Init CMSIS hook for?
2012-09-13T03:06:54 -!- tech2077 [~tech2077@adsl-75-53-132-237.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-09-13T03:16:41 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-13T03:19:13 -!- blocky [blocky@S0106c0c1c07746fb.gv.shawcable.net] has quit []
2012-09-13T03:24:07 -!- tech2077 [~tech2077@76.247.250.82] has joined ##stm32
2012-09-13T03:35:26 < dongs> sup trolls
2012-09-13T03:38:01 < dongs> http://www.theonion.com/articles/apple-announces-new-iphone-with-nword-on-back-know,29531/
2012-09-13T03:39:58 < dongs> zippe: lots of places are saying its dualcore a15
2012-09-13T03:40:03 < dongs> zippe: what do you know that they dont
2012-09-13T03:41:53 < emeb> the onion - they so naughty
2012-09-13T03:43:25 -!- tech2077 [~tech2077@76.247.250.82] has quit [Ping timeout: 246 seconds]
2012-09-13T03:43:31 < emeb> note that Anandtech doesn't categorically say it's an A15, just that "it looks like Apple has integrated two ARM Cortex A15 cores".
2012-09-13T03:43:45 < dongs> yeah but some indian outfits are saying specifically its a15
2012-09-13T03:44:37 < emeb> wonder what their sources are. so far all references I've seen point back to Anandtech.
2012-09-13T03:44:58 < dongs> ARM describes Cortex A15 processor as "the highest-performance licensable processor the industry has ever seen". iPhone 5 is the first phone to feature this technology.
2012-09-13T03:46:36 < dongs> lol @ apple pricing though
2012-09-13T03:46:45 < dongs> +$100 for 32gb nand
2012-09-13T03:46:51 < dongs> 32gig microsd costs what... $20?
2012-09-13T03:46:59 < emeb> yep
2012-09-13T03:47:05 < emeb> they do it because they can.
2012-09-13T03:47:11 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds]
2012-09-13T03:47:37 < dongs> haha nonbrand 32gig microsd is like $15 actually
2012-09-13T03:47:39 < dongs> that is nuts
2012-09-13T03:47:43 < BrainDamage> there's a reason why no sd slot
2012-09-13T03:47:48 < emeb> exactly
2012-09-13T03:47:49 < dongs> BrainDamage: duh, right
2012-09-13T03:48:22 < dongs> even branded shit like sandisk ultra is still around 25bucks
2012-09-13T03:48:24 < emeb> and true believers will swallow that the built in nand is better quality than $20 SD
2012-09-13T03:48:51 < dongs> thre's no 64gig cards yet?
2012-09-13T03:49:04 < dongs> SDXC for 64gig?
2012-09-13T03:49:36 < dongs> oh wow there is
2012-09-13T03:49:57 < dongs> $60bucks for sandisk ultra
2012-09-13T03:50:09 < dongs> apple fucking rapaes.
2012-09-13T03:56:01 -!- tech2077 [~tech2077@adsl-75-53-140-154.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-13T04:01:15 -!- tech2077 [~tech2077@adsl-75-53-140-154.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-13T04:02:53 < Bird|lappy> do modern SDHC/SDXC cards still support SPI mode?
2012-09-13T04:03:09 < dongs> duno, but with STM32 having SDIO controller who cares?
2012-09-13T04:04:04 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-09-13T04:04:11 < Bird|lappy> hehe :)
2012-09-13T04:07:49 < Thorn> stm32f107 doesn't have sdio
2012-09-13T04:08:57 < Bird|lappy> should my _exit() stub sit and spin, or soft-reboot the CPU?
2012-09-13T04:10:11 < Thorn> depends on what you were smoking whrn you decided to call it
2012-09-13T04:10:28 < Thorn> you can halt a microcontroller, you can restart it, you can't exit()
2012-09-13T04:11:13 < Thorn> unless your rtos supports task destruction, which is a different matter
2012-09-13T04:13:05 < Bird|lappy> no RTOS here, just bare metal/cmsis/newlib
2012-09-13T04:13:29 < Thorn> well, exit() makes no sense in that environment
2012-09-13T04:13:47 < Bird|lappy> but I'd think halting would be a bit annoying, in addition to causing a big power drag.  and newlib insists I provide an _exit() stub
2012-09-13T04:14:23 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-13T04:15:26 < Thorn> if you never call it, does it matter what it does?
2012-09-13T04:15:43 < Bird|lappy> true :)
2012-09-13T04:16:21 < Thorn> maybe the default new handler or a generic exception handler calls it, no idea here (only relevant to c++)
2012-09-13T04:17:04 < Thorn> but those things will need to be overridden anyway
2012-09-13T04:17:15 < Thorn> for a microcontroller
2012-09-13T04:32:47 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-13T05:06:56 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-13T05:10:48 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-13T05:14:01 -!- Thorn__ [~Thorn@176.15.67.62] has joined ##stm32
2012-09-13T05:14:01 -!- Thorn__ [~Thorn@176.15.67.62] has quit [Changing host]
2012-09-13T05:14:01 -!- Thorn__ [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-13T05:14:09 -!- oPossum5151 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-13T05:14:22 -!- oPossum5150 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-13T05:14:24 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-13T05:15:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-13T05:15:49 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-09-13T05:27:45 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-13T05:57:21 -!- phantoneD is now known as phantoxeD
2012-09-13T06:06:05 < emeb_mac> quiet here... too quiet.
2012-09-13T06:06:48 < Bird|lappy> heh, quite
2012-09-13T06:07:10  * Bird|lappy has his toolchain working and spitting out a .elf file now though \o/
2012-09-13T06:09:38 < dongs> yeah no chatting wtf
2012-09-13T06:10:33 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has quit [Remote host closed the connection]
2012-09-13T06:14:26 < Bird|lappy> heh.  *is writing a couple-page paper for class*
2012-09-13T06:16:38 < Bird|lappy> perhaps you would like something to read, though?
2012-09-13T06:25:31 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 252 seconds]
2012-09-13T06:27:19 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-13T06:27:25 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-13T06:34:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-13T06:34:49 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-13T06:40:25 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 260 seconds]
2012-09-13T06:47:06 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-13T07:03:37 < zippe> dongs: what I know is that it's not the A15
2012-09-13T07:05:37 < dongs> so what is it then
2012-09-13T07:11:31 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-13T07:14:14 < zippe> Not the A15
2012-09-13T07:14:30 < emeb_mac> custom Apple jank
2012-09-13T07:15:51 < zippe> http://www.linkedin.com/pub/gerard-williams-iii/a/95a/278
2012-09-13T07:16:05 < zippe> Draw your own conclusions. He was the CortexA8 architect @ ARM.
2012-09-13T07:20:31 < dongs> so what, he designed someshit custom for crapple?
2012-09-13T07:21:38 < dongs> Technical lead of entire processor from concept, architecture, micro-architecture, verification to implementation.
2012-09-13T07:30:40 < dongs> lol @ coolest job in the world
2012-09-13T07:30:52 < dongs> more like "gonna be the fastest dying company of 21st century
2012-09-13T07:31:29 < zippe> http://www.google.com/finance?q=aapl
2012-09-13T07:31:32 < zippe> Looks pretty dead to me
2012-09-13T07:31:42 < dongs> agreed
2012-09-13T07:32:48 < dongs> its hte next facebook.
2012-09-13T07:37:19 < dongs> http://newyork.newsday.com/business/technology/iphone-5-release-pushes-facebook-stock-higher-1.3996006 ???
2012-09-13T07:37:34 < dongs> Facebook Inc. surged as much as 8.9 percent, on track for a record gain, after Apple Inc. said the iPhone 5 has a built-in application for the social network that enables photo sharing and voice-activated posts.
2012-09-13T07:37:38 < dongs> lol.
2012-09-13T07:48:45 < zippe> In totally random news, have you looked at the Silego GreenPak2 devices?
2012-09-13T07:49:07 < zippe> I'm kinda bummed about the crapulent drive strength, but apart from that, they're cute little toys
2012-09-13T07:52:48 < dongs> is that the fpga-with-no-io-pins thingies?
2012-09-13T07:53:20 < dongs> couldnt find any product I'd be manufacturing 10k+ of to use them in
2012-09-13T07:53:22 < dongs> but yeah neat otherwise
2012-09-13T07:54:41 < zippe> No, it's a small mixed-signal thing
2012-09-13T07:54:45 < zippe> It has a few I/Os
2012-09-13T07:54:58 < zippe> adc, pwm, etc.
2012-09-13T07:55:01 < zippe> Still trying to find a use
2012-09-13T07:56:28 < dongs> just looked at their site yeah ive looked at it before
2012-09-13T07:56:33 < dongs> didnt seem like they'd be interested in hobbyists
2012-09-13T07:56:36 < dongs> so 10k+ order minimum
2012-09-13T08:06:43 -!- claude is now known as Claude
2012-09-13T08:12:20 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-13T08:34:13 -!- Thorn__ is now known as Thorn
2012-09-13T09:08:19 < R2COM> i cant find a datasheet of that greenpak shit on their site
2012-09-13T09:08:58 < zippe> R2COM: You clicked on one of the products, and then the link marked "Datasheet"?
2012-09-13T09:09:12 < dongs> thats too hardr
2012-09-13T09:10:20 < R2COM> i dont see datasheet
2012-09-13T09:10:59 < zippe> http://www.silego.com/index.php?page=greenpak2
2012-09-13T09:11:10 < zippe> You might actually have to … wait for it … scroll down
2012-09-13T09:12:07 < R2COM> oh.. i was on buy section
2012-09-13T09:12:21 < R2COM> and for some reason when i click products from there, it didnt lead me to a page you showed
2012-09-13T09:12:25 < R2COM> i had to enter from main one
2012-09-13T09:15:41 < R2COM> dongs you need a rest
2012-09-13T09:15:48 < R2COM> like, a long rest
2012-09-13T09:15:54 < dongs> rly
2012-09-13T09:15:58 < dongs> from?
2012-09-13T09:18:12 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-09-13T09:21:47 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-13T09:23:11 -!- oPossum5151 is now known as oPossum5150
2012-09-13T09:33:58 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-13T09:53:06 -!- izua [~izua@86.121.79.148] has quit [Read error: Connection reset by peer]
2012-09-13T10:17:18 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-09-13T10:18:47 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-13T10:22:28 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-09-13T10:22:28 -!- ServerMode/##stm32 [+o ChanServ] by calvino.freenode.net
2012-09-13T10:31:21 -!- Claude is now known as claude
2012-09-13T10:51:30 -!- claude is now known as claude_work
2012-09-13T10:53:44 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-13T11:09:55 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-13T11:09:56 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-13T11:09:56 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-13T11:25:59 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-13T11:26:00 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-13T11:38:38 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-13T11:54:28 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-13T12:32:08 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-13T12:32:47 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-13T12:37:48 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-13T12:44:42 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-13T13:12:53 < dongs> The desktop Linux suckage we are seeing today - on basically all the major Linux distributions - are the final symptoms of mistakes made 10-20 years ago - the death cries of a platform.
2012-09-13T13:14:24 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-13T13:14:40 < dongs> shit, Linuxb_ just missed my troll
2012-09-13T13:14:46 < Laurenceb_> wut
2012-09-13T13:14:59  * Laurenceb_ is getting trolled by his email
2012-09-13T13:15:26 < Laurenceb_> about 5 different people trying to organise stuff and not copying each other in
2012-09-13T13:15:44 < Laurenceb_> first email - guy a : im interested in the job
2012-09-13T13:16:00 < Laurenceb_> guy b : heres clearer job description etc
2012-09-13T13:16:16 < Laurenceb_> guy a: sorry - not interested - doesnt copy everyone
2012-09-13T13:16:30 < Laurenceb_> everyone else about 5 times CCing everyone
2012-09-13T13:16:39 < Laurenceb_> awesome - welcome to the team etc
2012-09-13T13:16:42 < Laurenceb_> so much fail
2012-09-13T13:17:04 < Laurenceb_> about 100 emails bouncing around over nothing
2012-09-13T13:20:13 < dongs> I keep getting this spam from this Hale Dwoskin faggot
2012-09-13T13:20:26 -!- dekar [~dekar@212.255.20.12] has joined ##stm32
2012-09-13T13:20:30 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-13T13:22:27 < dongs> Laurenceb_ is into holistic releasing
2012-09-13T13:37:45 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-13T14:07:17 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has joined ##stm32
2012-09-13T14:37:22 -!- grummund_ [~user@unaffiliated/grummund] has joined ##stm32
2012-09-13T14:40:05 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-13T14:42:28 -!- Netsplit *.net <-> *.split quits: Thorn, grummund, R0b0t1
2012-09-13T14:44:17 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-13T14:44:17 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-13T14:44:17 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-13T14:45:37 -!- Netsplit over, joins: R0b0t1
2012-09-13T14:47:16 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 244 seconds]
2012-09-13T14:47:19 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-13T14:47:25 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-13T14:49:29 -!- dekar_ [~dekar@212.255.20.12] has joined ##stm32
2012-09-13T14:49:32 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-13T14:54:39 -!- dekar [~dekar@212.255.20.12] has quit [Ping timeout: 240 seconds]
2012-09-13T14:54:39 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-09-13T14:54:40 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-09-13T14:55:44 -!- tech2077 [~tech2077@adsl-75-53-141-224.dsl.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-09-13T14:59:59 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-09-13T15:04:57 -!- daku is now known as DaKu
2012-09-13T15:09:37 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-13T15:17:40 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 274 seconds]
2012-09-13T15:18:11 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 272 seconds]
--- Log closed Thu Sep 13 15:23:12 2012
--- Log opened Thu Sep 13 15:23:20 2012
2012-09-13T15:23:20 -!- jpa-_ [jpa@hilla.kapsi.fi] has joined ##stm32
2012-09-13T15:23:20 -!- Irssi: ##stm32: Total of 57 nicks [1 ops, 0 halfops, 1 voices, 55 normal]
2012-09-13T15:24:13 -!- Irssi: Join to ##stm32 was synced in 59 secs
2012-09-13T15:25:18 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 274 seconds]
2012-09-13T15:25:19 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 272 seconds]
2012-09-13T15:25:19 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 274 seconds]
2012-09-13T15:25:21 -!- jpa- [jpa@hilla.kapsi.fi] has quit [Ping timeout: 268 seconds]
2012-09-13T15:26:39 < claude_work> true
2012-09-13T15:27:37 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-13T15:28:35 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has joined ##stm32
2012-09-13T15:28:35 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has quit [Changing host]
2012-09-13T15:28:35 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32
2012-09-13T15:31:06 -!- Tectu [tectu@kunsmann.eu] has quit [Ping timeout: 246 seconds]
2012-09-13T15:31:44 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Ping timeout: 246 seconds]
2012-09-13T15:31:54 -!- Tectu [tectu@kunsmann.eu] has joined ##stm32
2012-09-13T15:33:06 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-13T15:33:20 -!- jaeckel [~jaeckel@jaeckel.eu] has joined ##stm32
2012-09-13T15:33:20 -!- jaeckel [~jaeckel@jaeckel.eu] has quit [Changing host]
2012-09-13T15:33:20 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:33:22 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:33:55 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:33:59 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:34:20 -!- jaeckel [~jaeckel@jaeckel.eu] has joined ##stm32
2012-09-13T15:34:21 -!- jaeckel [~jaeckel@jaeckel.eu] has quit [Changing host]
2012-09-13T15:34:21 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:34:55 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 260 seconds]
2012-09-13T15:34:55 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 260 seconds]
2012-09-13T15:36:23 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-13T15:36:49 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:37:35 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:37:57 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-13T15:40:45 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:42:55 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:42:55 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:43:35 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:44:26 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:44:59 -!- DaKu_ [DaKu@dakus.dk] has joined ##stm32
2012-09-13T15:45:16 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:45:47 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-13T15:48:13 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has quit [Excess Flood]
2012-09-13T15:48:46 -!- Ranzbak_ [~paul@wcrol.xs4all.nl] has joined ##stm32
2012-09-13T15:48:46 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-13T15:51:35 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-09-13T15:52:44 -!- Ranzbak [~paul@wcrol.xs4all.nl] has quit [Write error: Broken pipe]
2012-09-13T15:52:45 -!- DaKu [DaKu@dakus.dk] has quit [Write error: Broken pipe]
2012-09-13T15:53:22 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-13T15:53:45 -!- prog` [prog@bornbrilliant.org] has joined ##stm32
2012-09-13T15:54:57 -!- Steffanx [~Steffanx@client-208-073.flexnet2.rug.nl] has joined ##stm32
2012-09-13T15:54:58 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-13T15:57:02 -!- oPossum5151 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-13T15:57:25 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-13T16:02:25 -!- Netsplit *.net <-> *.split quits: oPossum5150, prog
2012-09-13T16:05:45 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-13T16:06:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-13T16:11:28 -!- claude_work_ [sbnc@h1682708.stratoserver.net] has joined ##stm32
2012-09-13T16:11:56 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-13T16:12:02 -!- claude_work [sbnc@h1682708.stratoserver.net] has quit [Ping timeout: 248 seconds]
2012-09-13T16:12:23 -!- dongs_ [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds]
2012-09-13T16:12:33 -!- gsmcmull1n [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-13T16:13:29 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 248 seconds]
2012-09-13T16:13:45 -!- CheBuzz [~CheBuzz@76.164.192.124] has quit [Ping timeout: 248 seconds]
2012-09-13T16:15:29 -!- ntfreak_ is now known as ntfreak
2012-09-13T16:16:36 -!- oPossum5152 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-13T16:16:46 -!- oPossum5152 is now known as oPossum5150
2012-09-13T16:17:17 -!- karlp [~karl@palmtree.beeroclock.net] has quit [Ping timeout: 248 seconds]
2012-09-13T16:17:22 -!- karlp_ [~karl@palmtree.beeroclock.net] has joined ##stm32
2012-09-13T16:17:36 -!- oPossum5151 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-13T16:17:58 -!- You're now known as jpa-
2012-09-13T16:30:03 -!- Steffanx [~Steffanx@client-208-073.flexnet2.rug.nl] has quit [Quit: Ik ga weg]
2012-09-13T16:33:15 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-13T16:36:16 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-13T16:55:41 < dongs> argh
2012-09-13T16:55:44 < dongs> spent half a day cleaning my office
2012-09-13T16:55:46 < dongs> and its still fucked
2012-09-13T16:59:02 < zyp> sounds like my living room
2012-09-13T17:05:41 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-13T17:11:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-13T17:12:06 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-13T17:15:59 -!- Laurenceb__ is now known as Laurenceb_
2012-09-13T17:44:35 -!- claude_work_ is now known as claude
2012-09-13T18:13:38 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-13T18:16:12 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-13T18:16:15 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-13T18:16:21 -!- barthess [~barthess@164.177.204.11] has joined ##stm32
2012-09-13T18:19:03 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-13T18:19:32 -!- tavish [~tavish@59.177.11.83] has joined ##stm32
2012-09-13T18:19:32 -!- tavish [~tavish@59.177.11.83] has quit [Changing host]
2012-09-13T18:19:32 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-13T18:23:01 -!- DaKu_ is now known as DaKu
2012-09-13T18:23:29 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-13T18:24:45 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-13T18:50:37 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 240 seconds]
2012-09-13T18:53:20 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-13T18:54:21 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-13T18:54:24 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-13T19:01:54 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 268 seconds]
2012-09-13T19:20:48 -!- enots [dimka@freelsd.net] has joined ##stm32
2012-09-13T19:44:11 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-13T19:44:29 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-13T19:54:29 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-13T19:54:49 < TitanMKD> hi
2012-09-13T19:54:53 <+Steffanx> lo
2012-09-13T19:57:08 < jpa-> having a mounting screw touching the V+ rail: [ ] Good idea  [X] Not a good idea
2012-09-13T19:57:50 <+Steffanx> Not fair.. you filled in the answer for use
2012-09-13T19:58:37 < jpa-> well, i kind of thought it might be too hard
2012-09-13T19:58:58 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-13T19:59:15 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-13T20:07:57 < zyp> jpa-, only to you, since you had to try it out first before deciding
2012-09-13T20:41:21 < zippe> Why is it bad? Mounting screws carrying power are a fine tradition.
2012-09-13T20:43:20 < zyp> when it's intentional :)
2012-09-13T20:44:50 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-13T20:45:33  * emeb wants to see mounting screws carrying HS USB.
2012-09-13T20:45:44 < emeb> not really
2012-09-13T20:56:48 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-13T20:56:49 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-13T20:56:49 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-13T20:56:52 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-13T21:10:36 < bgamari> for some reason my STM32F405 crashes as soon as I push SYSCLK above 32MHz
2012-09-13T21:10:48 < bgamari> which is well below the maximum clock of 168MHz
2012-09-13T21:11:08 < bgamari> Any ideas what might be going on here?
2012-09-13T21:11:09 < jpa-> have you set the voltage regulator range properly?
2012-09-13T21:11:40 < jpa-> also, the APB1 and APB2 domains have lower limits so you need to set a divider for them
2012-09-13T21:13:28 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-13T21:13:31 < zyp> bgamari, are you sure you are configuring the PLL correctly?
2012-09-13T21:13:45 < bgamari> jpa-, hmm, Where is the voltage regulator range set?
2012-09-13T21:13:58 < bgamari> jpa-, Yes, I am using a fairly conservative divider on the APB1 and 2 domains
2012-09-13T21:14:05 < bgamari> zyp, Fairly
2012-09-13T21:14:23 < zyp> bgamari, pastebin the working and non-working code please
2012-09-13T21:14:55 < zyp> remember that the PLL intermediates also have some limits
2012-09-13T21:17:09 < bgamari> zyp: http://hpaste.org/74699
2012-09-13T21:18:14 < jpa-> bgamari: ah, the voltage range setting is at maximum by default, so it should work (PWR->CR)
2012-09-13T21:19:43 < bgamari> zyp, The above works
2012-09-13T21:20:00 < bgamari> if I change N to 80 it also works as it turns out
2012-09-13T21:20:05 < bgamari> but above 85 it fails
2012-09-13T21:21:41 < bgamari> N can be found on line 176
2012-09-13T21:22:02 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-13T21:22:13 < zyp> yes
2012-09-13T21:23:56 < zyp> hmm, isn't the HSI 8 MHz?
2012-09-13T21:24:38 < zyp> you are using a division factor of 16 for M, which makes the PLL input 0.5 MHz
2012-09-13T21:24:54 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-13T21:24:57 < zyp> the valid range is 1-2 MHz with recommended value of 2 MHz
2012-09-13T21:26:28 < bgamari> Hmm, I was under the impression it was 16MHz
2012-09-13T21:26:29 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 260 seconds]
2012-09-13T21:26:52 < zyp> as far as I can see, your working code does HSI / 16 * 64 / 2
2012-09-13T21:26:58 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-13T21:27:08 < zyp> oh, you're right
2012-09-13T21:30:56 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 252 seconds]
2012-09-13T21:32:26 < bgamari> found one issue
2012-09-13T21:32:52 < bgamari> I miscompute PeripheralClock and SlowPeripheralClock
2012-09-13T21:33:05 < bgamari> which led me to violate my APB1 constraint
2012-09-13T21:34:38 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-13T21:38:00 < bgamari> damn
2012-09-13T21:38:04 < bgamari> something has gone horribly
2012-09-13T21:38:05 < bgamari> wrong
2012-09-13T21:38:18 < bgamari> I don't even get any response from JTAG now
2012-09-13T21:39:29 < bgamari> the chip just vanished
2012-09-13T21:41:29 < bgamari> zyp, Could screwing around with the PLL possibly destroy an MCU?
2012-09-13T21:46:01 < bgamari> wtf
2012-09-13T21:49:12 < jpa-> not usually
2012-09-13T21:49:24 < jpa-> but try connecting while holding reset
2012-09-13T21:49:37 < jpa-> because if it hangs the CPU, the debug unit may sometimes hang also
2012-09-13T21:52:49 < bgamari> jpa-, that did it
2012-09-13T21:52:53 < bgamari> jpa-, Thanks!
2012-09-13T21:53:00 < bgamari> now the question is why did that happen?
2012-09-13T21:56:07 < bgamari> Unfortunately now I can't halt the device and openocd won't flash it while running
2012-09-13T21:57:04 < zyp> it's halted if you keep reset asserted
2012-09-13T21:58:52 < bgamari> zyp, I can't convince openocd that's true unfortunately
2012-09-13T21:59:28 < jpa-> bgamari: what happens if you connect under reset but then let it go?
2012-09-13T21:59:48 < jpa-> that works for me using the stlink utility and swd
2012-09-13T22:00:01 < bgamari> ahh, there we go
2012-09-13T22:00:05 < bgamari> it worked the second time
2012-09-13T22:00:12 < bgamari> but not the first for whatever reason
2012-09-13T22:01:23 < bgamari> good, now I just have to figure out what's wrong with my PLL configuration
2012-09-13T22:08:16 < bgamari> hmm, yeah, things are just very unstable at high N
2012-09-13T22:17:55 <+dekar_> I was able to OC my stm32 to 136mhz without damaging it
2012-09-13T22:18:00 <+dekar_> *stm32f1
2012-09-13T22:18:36 < zyp> bgamari, try increasing the PLL input to 2 MHz as recommended in the RM
2012-09-13T22:18:48 < zyp> it is supposed to reduce jitter
2012-09-13T22:19:18 < bgamari> that is a good point
2012-09-13T22:20:21 < bgamari> I also started using HSE and things seem much better
2012-09-13T22:22:48 < bgamari> Before when I was running the chip on HSI near the edge of its capabilities occassionally I would see the SPI rate increase by a factor of several
2012-09-13T22:22:55 < bgamari> I'm driving a DAC
2012-09-13T22:23:07 < bgamari> and I'd be getting 30kHz updates
2012-09-13T22:23:28 < bgamari> whereas when under normal conditions I'd only get 10kHz updates
2012-09-13T22:23:31 < zyp> well, you're not allowed to complain about clock accuracy if you're only using HSI :)
2012-09-13T22:23:35 < bgamari> what could cause this?
2012-09-13T22:23:48 < bgamari> Could the clock jump in rate by that much?
2012-09-13T22:24:37 < zyp> how much?
2012-09-13T22:25:05 < zyp> anyway, the HSI is probably rather temperature dependant
2012-09-13T22:25:15 < zyp> and higher clock rates lead to more heat
2012-09-13T22:25:36 < zyp> so I wouldn't be surprised if there is a connection there ;)
2012-09-13T22:26:56 < bgamari> zyp, a factor of 3?
2012-09-13T22:27:52 < zyp> nah, that rather sounds like you messed up something
2012-09-13T22:28:34 -!- barthess [~barthess@164.177.204.11] has quit [Quit: Leaving.]
2012-09-13T22:28:35 < bgamari> hmm
2012-09-13T22:28:53 < bgamari> zyp, Probing one of the pins of a resonator should show its fundamental frequency, correct?
2012-09-13T22:29:03 < bgamari> I could have sworn this was a 16MHz resonator
2012-09-13T22:29:08 < bgamari> but the scope seems to show 8MHz
2012-09-13T22:52:36 < karlp_> HSI is 1%, that's more than sufficient for lots of use cases.
2012-09-13T22:52:49 < karlp_> far better than the 10% on the internal oscillators on avr's for instance
2012-09-13T22:56:19 < bgamari> damn, I still can't get the PLL to work at a reasonable frequency
2012-09-13T22:56:53 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-13T22:57:42 < bgamari> It's locked at what I believe should be 100MHz
2012-09-13T22:57:54 < bgamari> which is then divided down to 50MHz with P=0
2012-09-13T22:58:13 < bgamari> and then another factor of 2 for APB1
2012-09-13T22:58:28 < bgamari> yet the device dies the moment I switch to PLLCLK
2012-09-13T22:59:03 < Laurenceb_> M4 has 2 cycle division right?
2012-09-13T22:59:12 < Laurenceb_> that takes any register combination
2012-09-13T22:59:26 < zippe> Laurenceb: sounds right
2012-09-13T22:59:49 < Laurenceb_> so about as fast as bit shift
2012-09-13T23:00:10 < zippe> What are you trying to do?
2012-09-13T23:00:28 < zippe> If the divisor is a constant the compiler will pick shift vs. divide as appropriate
2012-09-13T23:00:41 < Laurenceb_> dsp code for doing DFT
2012-09-13T23:00:52 < zippe> If you know it's a power of 2, you should always use a shift because the compiler may be able to merge it with other operations
2012-09-13T23:01:13 < Laurenceb_> i can make the maths cleaner if i dont use powers of 2 everywhere
2012-09-13T23:01:20 < Laurenceb_> working out what the speed penalty is
2012-09-13T23:01:55 < karlp_> most recent regehr post is about performance of division :)
2012-09-13T23:02:13 < zippe> Laurenceb: measure it
2012-09-13T23:02:23 < zippe> Laurenceb: potential extra pressure from register operands...
2012-09-13T23:02:28 < Laurenceb_> yeah i will
2012-09-13T23:08:57 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 244 seconds]
2012-09-13T23:09:21 < Laurenceb_> should be fast enough tho - my core dsp code is MAC to decimate the sample rate
2012-09-13T23:11:25 < Laurenceb_> in fact this only runs at around 100KHz
2012-09-13T23:11:29 < Laurenceb_> so no issue at all
2012-09-13T23:20:11 < bgamari> That fact that I can only run SYSCLK up to 42 MHz before things go unstable seems very suspicious
2012-09-13T23:20:46 < bgamari> considering APB1 is supposedly being divided by 2 from that and has a maximum clock of 42MHz
2012-09-13T23:24:50  * bgamari is very confused
2012-09-13T23:25:34 < Laurenceb_> i suspect pll issues
2012-09-13T23:25:49 < Laurenceb_> f4 pll is hard to use correctly :P
2012-09-13T23:27:27 < bgamari> heh
2012-09-13T23:27:34 < bgamari> Laurenceb, Any hints?
2012-09-13T23:28:06 < Laurenceb_> other than read chibios code, no
2012-09-13T23:28:20 < bgamari> alright
2012-09-13T23:28:22 < bgamari> well that does help
2012-09-13T23:44:34 < bgamari> hmm
2012-09-13T23:44:49 < bgamari> well, if I pipe the PLL output to MCO2, I see that it's flat
2012-09-13T23:44:50 < bgamari> weird
2012-09-13T23:45:51 < bgamari> despite the fact that PLLRDY is set
2012-09-13T23:50:31 < zyp> I've never had any PLL issues on F4, but I'm using HSE
2012-09-13T23:51:46 < zippe> bgamari: are you trying to run the F4 off the HSI?
2012-09-13T23:53:47 < bgamari> zippe, nope, not any more
2012-09-13T23:53:59 < bgamari> I think HSI was giving me issues earlier
2012-09-13T23:54:03 < bgamari> now I'm using HSE
2012-09-13T23:54:10 < bgamari> yet experiencing many of the same problems
2012-09-13T23:54:29 < zippe> Have you set the Magic Bit?
2012-09-13T23:55:36 < bgamari> zippe, uh oh
2012-09-13T23:55:38 < bgamari> apparently not
2012-09-13T23:55:51 < bgamari> ahh, I guess MCO doesn't like outputting 50MHz
2012-09-13T23:56:16 < bgamari> if I set the MCO divider I no longer see the PLL output as being flat
2012-09-13T23:56:46 < zippe> In particular, check the VOS bit in PWR_CR
2012-09-13T23:56:52 < zippe> If it's zero, you will be very sad.
2012-09-13T23:58:06 < bgamari> oh no!
2012-09-13T23:58:07 < bgamari> it's 0
2012-09-13T23:58:58 < zippe> Yes. The manual says it defaults to 1, but I think the ROM messes with it.
2012-09-13T23:59:09 < bgamari> zippe, What does this do?
2012-09-13T23:59:34 < zippe> It controls the internal voltage regulator's scaling curve against frequency
2012-09-13T23:59:51 < zippe> basically, as you increase the clock frequency, the required core voltage goes up as well
--- Day changed Fri Sep 14 2012
2012-09-14T00:00:02 < zippe> This is why you fall over at about 40MHz
2012-09-14T00:00:33 < zippe> (roughly, it varies with temperature)
2012-09-14T00:00:41 < zippe> Bottom line, it needs to be 1 if you want to go fast.
2012-09-14T00:00:54 < zyp> hmm, I haven't touched that
2012-09-14T00:02:31 < zippe> The datasheet just says that AHB tops out at ~140MHz with it set to zero, but I have seen stability issues well below that with VOS=0
2012-09-14T00:02:36 < bgamari> zippe, Hmm
2012-09-14T00:02:49 < bgamari> zippe, VOSRDY never seems to go high after setting VOS
2012-09-14T00:03:12 < zippe> I've never checked it, TBH
2012-09-14T00:04:08 < zyp> I bet it just reads as 0 because PWREN in RCC_APB1ENR is not set :p
2012-09-14T00:04:28 < bgamari> yep
2012-09-14T00:04:31 < bgamari> that's the one
2012-09-14T00:04:36 < zippe> Hah, fun
2012-09-14T00:04:42 < zippe> So anyway, give that a shot, see if it helps
2012-09-14T00:05:17 < qyx_> one question related
2012-09-14T00:05:28 < zyp> bgamari, oh, by the way
2012-09-14T00:05:29 < qyx_> what is max dma req frequency?
2012-09-14T00:05:45 < zyp> bgamari, have you set the number of wait states on the flash?
2012-09-14T00:05:50 < zippe> qyx_: too vague, be more specific
2012-09-14T00:05:51 < bgamari> zyp, Nope
2012-09-14T00:05:54 < zyp> ah
2012-09-14T00:05:54 < bgamari> Uh oh
2012-09-14T00:05:58 < zyp> that's your problem
2012-09-14T00:06:13 < zippe> Yes, that will also screw you badly
2012-09-14T00:06:26 < bgamari> ahh
2012-09-14T00:06:29 < bgamari> register reference?
2012-09-14T00:06:34 < zyp> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/PROGRAMMING_MANUAL/DM00023388.pdf
2012-09-14T00:07:01 < zyp> right, at the default value you can run up to 30 MHz within spec
2012-09-14T00:07:14 < qyx_> zippe: dma triggered by timer copying gpio to sram buffer
2012-09-14T00:07:20 < bgamari> Thanks
2012-09-14T00:07:49 < zyp> I'm sorry I didn't think of it earlier
2012-09-14T00:08:09 < qyx_> i cannot find anything specific in the ref manual
2012-09-14T00:08:17 < qyx_> except that one appnote
2012-09-14T00:08:18 < zyp> I was looking at my own init code now, and just noticed that I call flash_init() before enabling the PLL :)
2012-09-14T00:09:17 < zippe> qyx_: expect each read to cost about 4 cycles on the bus
2012-09-14T00:09:29 -!- diydsp [41df9b1e@gateway/web/freenode/ip.65.223.155.30] has joined ##stm32
2012-09-14T00:09:32 < bgamari> yep
2012-09-14T00:09:35 < bgamari> that was it
2012-09-14T00:09:42 < bgamari> Thanks everyone!
2012-09-14T00:10:12 < zippe> best case; one beat to transfer address to GPIO block, one beat to return data to DMA block, one beat internal to DMA, one beat for posted write from DMA to memory
2012-09-14T00:10:46 < zippe> qyx_: You may want to measure it with the DMA free-running though
2012-09-14T00:11:08 < zippe> qyx_: There were some interesting quirks in dongs' memory-to-memory numbers I couldn't fully account for
2012-09-14T00:11:11 < qyx_> so 18MHz could be possible?
2012-09-14T00:11:19 < qyx_> hm
2012-09-14T00:11:38 < qyx_> on F2
2012-09-14T00:11:52 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-14T00:12:54 < zippe> qyx_: It's not beyond the realms of possible, no
2012-09-14T00:13:42 < qyx_> should try :S
2012-09-14T00:16:20  * bgamari has a tricky issue with his DAC
2012-09-14T00:16:41 < bgamari> it requires that CS stay high in between 32-bit frames
2012-09-14T00:16:46 < bgamari> for >80ns
2012-09-14T00:17:09 < bgamari> Is there any way to make that happen while using DMA?
2012-09-14T00:17:24 < zyp> zippe, you don't happen to have any experience with working directly with the usb core in the F1, do you?
2012-09-14T00:17:44 < bgamari> the STM32 SPI controller is a little inflexible it seems
2012-09-14T00:18:38 < zippe> zyp: sorry, no
2012-09-14T00:18:57 < zippe> bgamari: Yeah, not going to get that
2012-09-14T00:19:03 < bgamari> damn
2012-09-14T00:19:09 < bgamari> well that's a bummer
2012-09-14T00:19:11 < zippe> bgamari: blame your codec vendor
2012-09-14T00:19:16 < zippe> Any reason you're not using I2S?
2012-09-14T00:19:24 < bgamari> It's not an I2S device
2012-09-14T00:19:30 < bgamari> it's a general purpose DAC
2012-09-14T00:19:35 < zippe> Oh, ok
2012-09-14T00:20:01 < bgamari> I have read that the PCM Standard I2S protocol has this requirement
2012-09-14T00:20:11 < bgamari> but it seems that it uses 16 bit samples
2012-09-14T00:20:18 < bgamari> Does this sound familiar
2012-09-14T00:20:24 < zippe> Well, I2S is a different animal
2012-09-14T00:20:25 < zyp> I2S is not using chip select
2012-09-14T00:20:44 < zippe> The biggest issue there is usually L/R selection
2012-09-14T00:20:45 < zyp> I2S is using a flag that's toggling for every word to indicate channel
2012-09-14T00:21:33 < bgamari> hmm
2012-09-14T00:21:39 < bgamari> well this is a bummer
2012-09-14T00:22:13 < bgamari> I guess time to respin around a K20
2012-09-14T00:22:14 < bgamari> arg
2012-09-14T00:22:29 < zippe> Can you use the internal DAC?
2012-09-14T00:22:34 < zippe> Or is it not hifi enough?
2012-09-14T00:23:35 < bgamari> not really
2012-09-14T00:23:38 < bgamari> not enough channels
2012-09-14T00:23:44 < bgamari> and not enough resolution
2012-09-14T00:27:29 < bgamari> although hmm
2012-09-14T00:30:52 < bgamari> even at 82MHz I'm only pushing 10ksamples/s
2012-09-14T00:31:00 < bgamari> something seems off
2012-09-14T00:35:12 < bgamari> What exactly do the GPIO_OSPEED registers do?
2012-09-14T00:36:33 < bgamari> I guess this is the drive strength?
2012-09-14T00:37:38 < emeb> bgamari: can't use I2S DAC?
2012-09-14T00:38:29 < bgamari> emeb, Well, the hardware is already designed around the DAC8568
2012-09-14T00:38:40 < bgamari> so I'll have to redesign either way
2012-09-14T00:38:47 < bgamari> or flywire it temporarily I suppose
2012-09-14T00:43:25 < emeb> how many channels, what bit resolution needed on your DACs?
2012-09-14T00:44:25 < bgamari> emeb, 6 channels, 16 bit resolution would be nice
2012-09-14T00:44:32 < bgamari> 12 is probably acceptable
2012-09-14T00:45:00 < Laurenceb_> are you the guy whos trying to make a SDR?
2012-09-14T00:45:03 < emeb> ah - hard to get 6 channels out of I2S. Really only set up for 2
2012-09-14T00:45:32 < Laurenceb_> 6 simultaneous rickrolls
2012-09-14T00:45:34 < emeb> you can use both I2S peripherals and get 4 though
2012-09-14T00:45:54 < Thorn> bgamari: but you should be able to control CS in software (if that's your problem)
2012-09-14T00:45:56 < qyx_> why is everybody asking about sdr?
2012-09-14T00:46:03 < qyx_> Laurenceb_: ^
2012-09-14T00:46:04 < Laurenceb_> cuz its awesome
2012-09-14T00:46:07 < emeb> The issue w/ driving spi_cs during DMA bursts is a pretty common one.
2012-09-14T00:46:37 < bgamari> Thorn, Yeah, that's what I'm doing at the moment
2012-09-14T00:46:40 < Thorn> you should also have a dma complete interrupt
2012-09-14T00:46:56 < Thorn> unless I misunderstand your issue
2012-09-14T00:47:01 < bgamari> Thorn, This is true
2012-09-14T00:47:04 < bgamari> I've considered that
2012-09-14T00:47:31 < bgamari> Letting DMA do the 4 byte transfer, interrupt, fiddle with the CS, restart DMA
2012-09-14T00:47:49 < bgamari> but I'm not sure how much better than the current situation it will be
2012-09-14T00:47:51 < qyx_> Laurenceb_: i was trying to capture output of external high speed adc and get the same question
2012-09-14T00:48:03 < bgamari> how many cycles of overhead does an interrupt incur?
2012-09-14T00:48:08 < BrainDamage> because one user here is trying to
2012-09-14T00:48:11 < qyx_> but for the purpose of doing ethernet connected analog logger/oscope
2012-09-14T00:48:21  * emeb reads the history and is glad that he used system_stm32f4xx.c to setup his PLL, flash & power stuff. 
2012-09-14T00:48:22 < BrainDamage> I forgot his nick, he also lurks in ##rtlsdr
2012-09-14T00:48:32 < Thorn> bgamari: try it, look at cs with a scope and see if you like what you see or not
2012-09-14T00:48:46 < bgamari> emeb, Yep, you should feel very glad
2012-09-14T00:48:51 < bgamari> Thorn, Alright
2012-09-14T00:50:18 < emeb> IIRC the Atmel SAM7 spi peripheral could be set up to drive the CS line automatically. It'd be nice if everyone did that.
2012-09-14T00:51:15 < zyp> yeah, as long as you only have one slave
2012-09-14T00:51:31 < emeb> there's that.
2012-09-14T00:51:40 < bgamari> emeb, As can the Freescale SPI controller
2012-09-14T00:51:49 < bgamari> but it also handles the multi-slave case
2012-09-14T00:52:05 < Thorn> bgamari: The book (2nd ed) says as low as 12 cycles with zero wait states (page 152)
2012-09-14T00:52:30 < bgamari> yeah
2012-09-14T00:53:00 < Thorn> that assumes it's set to a high enough priority of course
2012-09-14T00:53:09 < Thorn> or it stops being deterministic
2012-09-14T00:53:24 < HTT-Bird> bgamari: whoa.  I know the EFM32's SPI handles auto-CS for single slaves
2012-09-14T00:53:28 < bgamari> So if I push 32 bits over SPI running at half my SYSCLK, then I'll only be able to do 60 instructions worth of work before being interrupted
2012-09-14T00:54:15 < bgamari> Not really sure if it's worth the effort to do DMA
2012-09-14T00:54:52 < zyp> bgamari, have you looked at the SSOE bit?
2012-09-14T00:55:47 < zyp> what if you use a timer to trigger DMA against the SPI peripheral and set it to have the appropriate delay between the words?
2012-09-14T00:56:10 < bgamari> hmm
2012-09-14T00:56:34 < bgamari> I guess that would be possible
2012-09-14T00:56:39 < HTT-Bird> yeah
2012-09-14T00:56:41 < Thorn> or how about a hardware delay using a counter
2012-09-14T00:58:20 < bgamari> alright
2012-09-14T00:58:24 < bgamari> well there are some ideas
2012-09-14T00:58:34 < bgamari> currently I can push 60ksamples/s
2012-09-14T00:58:40 < bgamari> which might be enough
2012-09-14T00:59:04 < bgamari> It is interesting to consider how this might be done though
2012-09-14T01:00:19 < zyp> timer-triggered dma transfers seem to be pretty versatile
2012-09-14T01:00:54 < bgamari> It's frustrating since CS only needs to go high for 8 SYSCLK cycles
2012-09-14T01:01:15 < emeb> so in that case does the SSOE bit allow the CS to assert while the transfer is active?
2012-09-14T01:02:19 < zyp> also, if you are triggering the DMA transfers from a timer, you could even run the CS from an OC channel
2012-09-14T01:03:33 < emeb> might be the more obvious approach
2012-09-14T01:10:27 < Laurenceb_> 4,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,10,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,5,5,4,3,2,2,1,0,-1,-2,-2,-3,-4,-5,-5,-6,-7,-7,-8,-9,-9,-10,-11,-11,-12,-12,-13,-13,-13,-14,-14,-14,-14,-15,-15,-15,-15,-15,-15,-15,-15,-15,-14,-14,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-
2012-09-14T01:10:28 < Laurenceb_> 7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,10,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,5,5,4,3,2,2,1,0,-1,-2,-2,-3,-4,-5,-5,-6,-7,-7,-8,-9,-9,-10,-11,-11,-12,-12,-13,-13,-13,-14,-14,-14,-14,-15,-15,-15,-15,-15,-15,-15,-15,-15,-14,-14,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,1
2012-09-14T01:10:28 < Laurenceb_> 0,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,5,5,4,3,2,2,1,0,-1,-2,-2,-3,-4,-5,-5,-6,-7,-7,-8,-9,-9,-10,-11,-11,-12,-12,-13,-13,-13,-14,-14,-14,-14,-15,-15,-15,-15,-15,-15,-15,-15,-15,-14,-14,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,10,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15
2012-09-14T01:10:33 < Laurenceb_> ,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,5,5,4,3,2,2,1,0,-1,-2,-2,-3,-4,-5,-5,-6,-7,-7,-8,-9,-9,-10,-11,-11,-12,-12,-13,-13,-13,-14,-14,-14,-14,-15,-15,-15,-15,-15,-15,-15,-15,-15,-14,-14,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,10,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,
2012-09-14T01:10:40 < Laurenceb_> 5,5,4,3,2,2,1,0,-1,-2,-2,-3,-4,-5,-5,-6,-7,-7,-8,-9,-9,-10,-11,-11,-12,-12,-13,-13,-13,-14,-14,-14,-14,-15,-15,-15,-15,-15,-15,-15,-15,-15,-14,-14,-14,-14,-13,-13,-13,-12,-12,-11,-11,-10,-9,-9,-8,-8,-7,-6,-5,-5,-4,-3,-2,-2,-1,0,1,2,2,3,4,5,5,6,7,7,8,9,9,10,11,11,12,12,13,13,13,14,14,14,14,15,15,15,15,15,15,15,15,15,14,14,14,14,13,13,13,12,12,11,11,10,9,9,8,8,7,6,5,5
2012-09-14T01:10:48 < qyx_> nice
2012-09-14T01:11:00 < Laurenceb_> ooh shit
2012-09-14T01:11:09 < Laurenceb_> thats not the terminal i want
2012-09-14T01:11:18 < gnomad> doncha hate when that happens?
2012-09-14T01:11:19 < Laurenceb_> sorry
2012-09-14T01:11:26 < qyx_> nearly *pseudo* random
2012-09-14T01:11:32 < gnomad> at least is wasn't a pr0n URL.
2012-09-14T01:15:02 < zippe> bgamari: assuming no FPU, interrupt is about 16 cycles in, 16 on the way out
2012-09-14T01:15:05 < zippe> IIRC
2012-09-14T01:15:15 < bgamari> zippe, Thanks
2012-09-14T01:17:07 < HTT-Bird> Laurenceb: ENOTTY!
2012-09-14T01:18:58 < qyx_> hm, another approach to capture adc could be to use external counters & sram
2012-09-14T01:19:15 < qyx_> and to generate counter clk and sram WE with stm32
2012-09-14T01:43:00 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-09-14T02:04:54 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-14T02:17:53 <+dekar_> that cortex a8 costs barely more than a stm32f4: http://www.digikey.com/product-detail/en/MCIMX503CVM8B/MCIMX503CVM8B-ND/2673213
2012-09-14T02:19:27 <+dekar_> is that still a µC? I mean it needs external RAM and ROM
2012-09-14T02:20:02 < BrainDamage> the package is a bit hard to use, 400-LFBGA
2012-09-14T02:20:03 < karlp_> massive price break at 100 parts
2012-09-14T02:22:44 <+dekar_> I guess I am not thinking too hobbyist, last production run we had at work was 250 stm32f1
2012-09-14T02:31:13 <+dekar_> it even seems to have internal 128kbyte ram and 96kbyte rom
2012-09-14T02:35:58 <+dekar_> I guess you could use it as a pretty fast µC, it has about anything, but ADCs/DACs
2012-09-14T02:39:25 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-14T02:39:30 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-14T02:39:30 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-14T02:43:49 < zippe> dekar_: the big issue with something like the i.MX503 is that it has complex external power requirements
2012-09-14T02:44:18 < zippe> You typically pair a device like that with a power management IC, as well as some DRAM and a mass storage device
2012-09-14T02:44:37 < zippe> Take a look at e.g. the Beaglebone for another example
2012-09-14T02:44:53 < zippe> The SoC is cheap enough, but by the time you add the rest of the required parts the system is quite large and complex
2012-09-14T02:58:39 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-14T03:16:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-14T03:16:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Excess Flood]
2012-09-14T03:16:43 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-14T03:20:32 -!- DaKu is now known as daku
2012-09-14T03:21:58 < dongs> http://foo.is/~baldur/.cache/images.4chan.org/b/src/1347554068714.png
2012-09-14T03:21:59 < dongs> haha
2012-09-14T03:25:10 < emeb> It's good strategic thinking, but as history shows didn't do them much good.
2012-09-14T03:25:27 < dongs> well, ACPI on lunix still doesnt work worht a fuck
2012-09-14T03:25:35 < dongs> even after a dozen of paid intel engineers wasted time for years on it
2012-09-14T03:25:52 < emeb> err whut?
2012-09-14T03:26:04 < emeb> power management on my linux laptop works fine.
2012-09-14T03:26:34 < dongs> bwahahah
2012-09-14T03:26:37 < dongs> funniest shit I heard this week.
2012-09-14T03:27:00 < dongs> windows 98 had better ACPI support than any iteration of lunix from that point until now.
2012-09-14T03:27:08 < Bird|lappy> well, power management on my laptop works fairly well, my desktop not so much
2012-09-14T03:27:24 < Bird|lappy> but the problem with my desktop is Intel's fault, and not a Linux-specific issue either :P
2012-09-14T03:28:04 < Bird|lappy> dongs, that's because BIOS writers write their tables to what the MS implementation accepts, not to the ACPI spec :P
2012-09-14T03:28:23 < dongs> typical; excuse lunix has been making for the last decade
2012-09-14T03:28:52 < BrainDamage> from an user POV tough, it doesn't matter who's fault
2012-09-14T03:29:16 < Bird|lappy> I think BIOS vendors are getting better about it though
2012-09-14T03:29:40 < Bird|lappy> the need for at least some ACPI functionality in server platforms I think is the driver
2012-09-14T03:30:27 < BrainDamage> "The ACPI specification contains numerous related components for hardware  and software programming, as well as a unified standard for  device/power interaction and for bus configuration. Linus Torvalds, creator of the Linux kernel, once described it as "a complete design disaster in every way" in relation to his view that "modern PCs are horrible".5 Some software developers have trouble implementing ACPI, and express c
2012-09-14T03:30:40 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-14T03:31:19 < dongs> uh huh
2012-09-14T03:32:01 < BrainDamage> well, at least it looks like they were successful at pissing him off
2012-09-14T03:33:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-14T03:33:19 < Bird|lappy> I will agree with his sentiment in terms of the sheer pile of kludges PCs have accumulated over the years
2012-09-14T03:33:37 < emeb> burn it down. start over.
2012-09-14T03:33:49 < Bird|lappy> heh, ARM has its own fair share of fun :>
2012-09-14T03:34:21 < Bird|lappy> not as bad as the PC environment, but still, there are some quirks floating around there
2012-09-14T03:34:29 < Bird|lappy> "what's 26 bit mode again?"
2012-09-14T03:35:17 < Bird|lappy> or your assembler complaining that "this instruction isn't valid in ARM mode"
2012-09-14T03:35:51 < Bird|lappy> speaking of that, why did nobody wire up -mcpu=cortex-m3 to imply -mthumb?
2012-09-14T03:36:49 < Bird|lappy> heck, -mcpu=cortex-m3 -marm should be a compiler error
2012-09-14T03:37:04 < Bird|lappy> error: invalid option: -mcpu=cortex-m3 does not support -marm
2012-09-14T03:45:14 <+dekar_> zippe, we had to supply 1V8 3V3 and 5V0 for the last product already so I don't think that would be a problem
2012-09-14T03:45:42 <+dekar_> using those awesome devices: http://www.murata-ps.com/data/power/okl-t3-w5.pdf
2012-09-14T03:48:48 <+dekar_> dongs, linux on my lenovo x220 in idle: https://dl.dropbox.com/u/28467113/x220_susi_wattage.png
2012-09-14T03:51:36 < dongs> dekar_: ok, press the power button rihgt now and suspend to disk
2012-09-14T03:51:46 < dongs> then come back after you power it back up.
2012-09-14T03:51:47 < dongs> kthx.
2012-09-14T03:52:48 < qyx_> why should somebody use suspend to disk?
2012-09-14T03:53:05 < qyx_> not an excuse, but whatever
2012-09-14T03:53:49 < BrainDamage> it generally makes faster booting to save to disk and resume
2012-09-14T03:54:11 < dongs> i havent powered off my windows laptop since I got it
2012-09-14T03:54:19 < dongs> i either rebot or suspend to disk
2012-09-14T03:54:28 < dongs> it takes like 10 seconds to power up this way.
2012-09-14T03:54:31 < qyx_> my 4y old laptop boots in 5s from ssd
2012-09-14T03:54:44 < qyx_> without any suspend to disk
2012-09-14T03:54:45 < BrainDamage> with all apps opened?
2012-09-14T03:55:10 <+dekar_> dongs, lol I only had that laptop for like 2 weeks until it broke :/
2012-09-14T03:55:15 < qyx_> no, if i want that i suspend to ram
2012-09-14T03:55:55 <+dekar_> I always suspend to ram as well
2012-09-14T03:56:01 < qyx_> theres no reason to suspend to disk
2012-09-14T03:56:03 <+dekar_> Bird|lappy, "error: target CPU does not support ARM mode"
2012-09-14T03:56:10 < qyx_> i do not need to have it suspended for weeks
2012-09-14T03:56:51 <+dekar_> Bird|lappy, arm-none-eabi-gcc -c -marm -march=armv7-m -mfloat-abi=soft -mcpu=cortex-m3
2012-09-14T03:57:24 <+dekar_> it _is_ a compiler error :)
2012-09-14T04:03:09 <+dekar_> dongs, the x220 ran really well though, it had a 80gb ssd as well as a hard disk
2012-09-14T04:03:30 <+dekar_> and I am pretty sure suspend to disk would have worked fine
2012-09-14T04:03:49 < dongs> under windows, yes.
2012-09-14T04:05:46 <+dekar_> dongs, I have an old (core2) samsung laptop here that runs debian squeeze and has fully working standby
2012-09-14T04:06:28 < dongs> dekar_: i dont need to care about old/new/whatever. i install windows 7, it works
2012-09-14T04:06:31 <+dekar_> the hardware is mostly failing (display backlight flickers, keyboard and touchpad buttons are partially unreliable), but the OS runs fine
2012-09-14T04:06:48 < dongs> thogh any new install will be of course windows 8
2012-09-14T04:06:52 <+dekar_> the x220 had an core i5 and also ran really well
2012-09-14T04:07:06 <+dekar_> I admit I pick my laptops for linux support
2012-09-14T04:07:14 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 260 seconds]
2012-09-14T04:07:14 <+dekar_> but I also pick my MacBooks for OSX support
2012-09-14T04:08:16 <+dekar_> also windows runs pretty badly on MacBooks, so it isn't a magic cure for battery life issues
2012-09-14T04:11:45 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-14T04:12:23 <+dekar_> dongs, if you prefer windows then that's fine, I have to support all major OS so I can't be picky about my choice of OS - Windows isn't that bad once you installed cygwin anyway :P
2012-09-14T04:44:14 < Bird|lappy> dongs, trust me, get an Intel BOXDP45SG and watch as power management falls over under Windows too :P
2012-09-14T04:44:49 < Bird|lappy> but, I digress
2012-09-14T04:48:10 < Bird|lappy> how concerned should I be about debouncing the manual reset pushbutton on my board?
2012-09-14T05:01:15 -!- diydsp [41df9b1e@gateway/web/freenode/ip.65.223.155.30] has quit [Ping timeout: 245 seconds]
2012-09-14T05:02:38 < dongs> why do you e ven have a reset button
2012-09-14T05:11:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-14T05:25:13 < bgamari> What could cause a TEIF error during a DMA transfer from and ADC to memory
2012-09-14T05:30:58 < Bird|lappy> dongs, 1) pulling the part out of "shutdown" mode in case it ever has to do that (due to running out of battery, although I suspect that there's a better way) 2) early-stage debugging (startup logic, basically)
2012-09-14T05:31:42 < dongs> both done wiht a proper debugger
2012-09-14T05:32:39 < Bird|lappy> dongs, good luck debugging a part that's had its debug interface turned off by the power management circuit for 1), and yes, I will have SWD debugging available to me
2012-09-14T05:33:57 < Bird|lappy> so 2) may or may not be an issue
2012-09-14T05:50:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Read error: Operation timed out]
2012-09-14T05:56:04 < Bird|lappy> do you folks prefer the 10-pin or the 20-pin version of the ARM debugging connector btw?
2012-09-14T06:06:41 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-14T06:10:36 < bgamari> What would cause a bus error during a peripheral to memory DMA transfer?
2012-09-14T06:28:32 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-09-14T06:29:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-14T06:39:54 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-14T06:53:04 < bgamari> man, FEIF sucks
2012-09-14T06:58:24 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-14T06:59:51 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-14T06:59:51 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-14T06:59:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-14T07:06:05 < zippe> bgamari: what memory address?
2012-09-14T07:06:26 < zippe> bgamari: you can't DMA to the CCSRAM, so if you have e.g. allocated your stack there, you're screwed.
2012-09-14T07:06:45 < zippe> (assuming you are DMA'ing to your stack, that is)
2012-09-14T07:19:36 < dongs> back of hte bus error.
2012-09-14T07:23:47 < zippe> dongs: btw, Windows ACPI interpreter has major bugs; they were introduced because the Windows folks refused to take the reference implementation from Intel, failed to understand the spec, and now every platform in existence has to cater to those bugs
2012-09-14T07:24:39 < zippe> dongs: This is why opensource ACPI implementations have so much trouble; vendors won't standardise on how they hack around the MS bugs, so you get e.g. AML that tests various parts of the reported OS name, or makes assumptions about the order in which various EC methods are invoked
2012-09-14T07:24:55 < dongs> right right
2012-09-14T07:25:13 < zippe> dongs: each subsequent Windows version has introduced new bugs, and platform vendors have introduced new hacks to compensate
2012-09-14T07:25:35 < zippe> (I worked on ACPI support for an opensource operating system back in the day)
2012-09-14T07:26:17 < zippe> You can lay the blame for the overall failure of ACPI pretty evenly on the platform vendors and Microsoft.  The Linux and IRL guys are largely blameless here
2012-09-14T07:26:27 < dongs> zippe, you forgot the most important part
2012-09-14T07:26:28 < zippe> Though, the Intel reference interpreter was a bit of a monster.
2012-09-14T07:26:29 < dongs> end-user
2012-09-14T07:26:31 < dongs> who doesnt give a flying fuck.
2012-09-14T07:26:42 < dongs> which seems to be the boat most lunix-like developers are missing
2012-09-14T07:26:49 < zippe> dongs: That's why end users seem to buy Macs.
2012-09-14T07:26:53 < dongs> if it doesnt work for end-user it doent fucking matter how eliter your system is
2012-09-14T07:27:16 < zippe> dongs: Since the end-user experience for WIndows users is shit-awful
2012-09-14T07:27:25 < dongs> zippe, 98% installed userbase disagrees
2012-09-14T07:27:35 < zippe> dongs: flies eat shit too
2012-09-14T07:27:39 < zippe> dongs: doesn't make it tasty
2012-09-14T07:27:53 < dongs> now youre just being psycho
2012-09-14T07:27:58 < dongs> macos UI is fucking unusable shit.
2012-09-14T07:28:17 < dongs> you cant fucking do anything in that shit without 1000 mouse clicks
2012-09-14T07:28:26 < zippe> dongs: You're advocating a platform where the #1 technical support response to any problem is "format and reinstall"
2012-09-14T07:28:30 < dongs> and im talking about users using it not some hakk0r clicking once to assign a hotkey to open terminal and maximize it
2012-09-14T07:28:52 < zippe> dongs: You can drive the entire OS X UI without touching the mouse
2012-09-14T07:28:57 < dongs> zippe: what do you think macfgaggots do when apple craps out a new decimal point release of their single-threaded crap OS
2012-09-14T07:29:06 < zippe> dongs: your problem is simply that you're scared to try new things
2012-09-14T07:29:26 < zippe> dongs: and as a consequence, you have no *experience* from which to speak
2012-09-14T07:30:02 < dongs> zippe, driving all of mactrash UI from keyboard?
2012-09-14T07:30:08 < dongs>  do tell, did something change in teh last 5 years?
2012-09-14T07:30:23 < zippe> dongs: No, it's been that way since at least 10.2, which makes it nearly 10.
2012-09-14T07:30:23 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-14T07:30:23 < dongs> all dialogs now have sense-making taborder that hits all controls, really?
2012-09-14T07:30:34 < dongs> and you can open menus from keyboard and shit?
2012-09-14T07:30:43 < zippe> It's a miracle, I tell you.
2012-09-14T07:30:53 < dongs> with proper access keys, instead of spending hours pressing arrow keys to get to a menu item?
2012-09-14T07:31:05 < dongs> zippe, you must be shitting me, please do file->open in your nearest mac app and tell me how long it took.
2012-09-14T07:31:14 < dongs> (how many keypresses)
2012-09-14T07:31:26 < zippe> Two, if you count the modifier separately.
2012-09-14T07:31:49 < dongs> im not talking about invoking open file hotkey.
2012-09-14T07:31:54 < dongs> im talking about menu items that dont have them
2012-09-14T07:31:57 < dongs> that you need to get to manually.
2012-09-14T07:32:14 < dongs> there's a difference between hotkey and accesskey, you ought to know it.
2012-09-14T07:32:22 < zippe> Well, file->open has a hotkey, and you asked.
2012-09-14T07:32:53 < zippe> Beyond that, what do you expect?  One for the live menu, one for each x/y axis traversal for anything that isn't otherwise bound.
2012-09-14T07:33:11 < dongs> eaxctly. fail.
2012-09-14T07:33:22 < zippe> Oh, I see, you want CUA
2012-09-14T07:33:23 < dongs> max 3 keypresses to get to any item on windows.
2012-09-14T07:33:34 < dongs> assuming 1 level deep.
2012-09-14T07:33:35 < zippe>  You are forgetting nested menus
2012-09-14T07:33:39 < dongs> im not.
2012-09-14T07:33:50 < zippe> Also, you are forgetting keystrokes stolen by the app camping the event stream.
2012-09-14T07:34:26 < dongs> whatever dude.
2012-09-14T07:34:30 < zippe> But, regardless.  More keystrokes != can't be done.
2012-09-14T07:34:35 < dongs> there's no accesskeys on mactrash, and it s been 10 years as you said.
2012-09-14T07:34:41 < zippe> "accesskeys" ?
2012-09-14T07:35:02 < zippe> Are you, again, referring to the CUA "requirement" that every entry have an activation key?
2012-09-14T07:35:12 < zippe> You do realise that's total fail in any non-7-bit language?
2012-09-14T07:35:14 < dongs> why do you "" requirement?
2012-09-14T07:35:34 < dongs> zippe, which is why they go after text in ex. japanese menus.
2012-09-14T07:35:50 < dongs> i do not agree with localizing it in shit like russian/etc (though it is done)
2012-09-14T07:35:55 < zippe> Because CUA is lipservice material; it's an excuse to avoid UI that doesn't require menu traversal in the first place.
2012-09-14T07:36:08 < zippe> Look at the ribbon, for example/
2012-09-14T07:36:35 < dongs> who do you think they got ribbon idea from
2012-09-14T07:36:39 < dongs> fucking either lunix or mac faggots.
2012-09-14T07:36:48 < zippe> Or the forest of micro-toolbar icons that your average SDI app features.
2012-09-14T07:36:56 < dongs> i never click toolbars
2012-09-14T07:37:25 < zippe> Sure.  So again, you're angry about things that are different, and you appear to have a mild case of Internet Tourette's.
2012-09-14T07:37:35 < zippe> I should be convinced because?
2012-09-14T07:37:51 < dongs> things that are DIFFERENT that make the system UNUSABLE sure
2012-09-14T07:37:57 < zippe> Meanwhile, people are buying things that make them happy, and the trend there is not towards your favorite platform.  Oh, wah.
2012-09-14T07:38:03 < dongs> i spend 95% of my time on windows when doing work wiht keyboard only.
2012-09-14T07:38:12 < dongs> you need to spend 95% of the time on teh fucking single-buttoned mouse
2012-09-14T07:38:13 < dongs> in macos.
2012-09-14T07:38:19 < dongs> to do ANYTHING.
2012-09-14T07:38:31 < zippe> Which "you" are you referring to?  Certainly not me, or anyone I know.
2012-09-14T07:39:15 < zippe> When I do use the pointer, it's because it's a better option than the keyboard.  *shrug*  Use the right tool for the job.
2012-09-14T07:39:20 < zippe> Not everything is a nail, neh?
2012-09-14T07:39:49 < dongs> shrug
2012-09-14T07:40:30 < zippe> (and as for pointing devices; I use a 5-button trackball.  blame Descent)
2012-09-14T07:47:17 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-14T08:07:13 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-14T08:07:13 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-14T08:07:13 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-14T08:48:18 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-14T08:49:06 < zippe> zyp: you here just now?
2012-09-14T09:14:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 245 seconds]
2012-09-14T09:15:45 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-14T09:30:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-14T09:40:34 < zyp> zippe, I'm here now
2012-09-14T09:47:13 < zippe> zyp: Just wondering if you'd ever seen an issue with the F4 I2C if you failed to set the "don't ack" bit for too long after reading the second-last byte
2012-09-14T09:47:31 < zippe> I think I have an interrupt race that actually causes it to kill the transfer late, in the middle of a byte no less
2012-09-14T09:51:50 < zyp> I haven't seen it myself, but it sounds plausible
2012-09-14T09:57:19 < zyp> or, it doesn't really sound plausible that it should kill the transfer in the middle of a byte
2012-09-14T09:58:33 < zyp> since the ACK bit is stored in CR1, could it be that you mess up some other part of CR1 while modifying it?
2012-09-14T10:14:41 < zippe> The code is pretty careful
2012-09-14T10:21:40 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-14T10:21:44 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-14T10:37:44 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-14T10:38:25 < Laurenceb_> http://www.bbc.co.uk/news/uk-19595221
2012-09-14T10:38:30 < Laurenceb_> cant fap to this
2012-09-14T10:39:13 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-14T10:39:21 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-14T10:44:17 -!- LeelooMinai_ [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-14T10:45:14 -!- LeelooMinai_ [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Client Quit]
2012-09-14T10:56:24 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Remote host closed the connection]
2012-09-14T10:59:32 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-14T11:02:34 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-09-14T12:02:57 -!- claude is now known as claude_work
2012-09-14T12:17:36 < jpa-> is there any information on how the IO pins behave when the mcu is not powered?
2012-09-14T12:18:45 < zyp> tried the datasheet?
2012-09-14T12:19:19 < jpa-> yeah didn't find; i would expect that the 5-volt toleran IOs would not use the usual diode to vdd
2012-09-14T12:19:23 < zyp> normally a strong high on any pin would power the chip through the TVS diodes
2012-09-14T12:19:26 < zyp> but yea
2012-09-14T12:22:01 < dongs> it does
2012-09-14T12:22:04 < dongs> and not evn so strong.
2012-09-14T12:22:05 < zyp> ah, datasheet has some info
2012-09-14T12:22:24 < dongs> i was lighting up a blue led when uart is connected
2012-09-14T12:22:25 < zyp> «Positive injection is not possible on these I/Os.
2012-09-14T12:22:28 < dongs> with main board unpowered
2012-09-14T12:23:22 < zyp> dongs, but that's not 5V tolerant IO
2012-09-14T12:23:37 < dongs> both uarts arent?
2012-09-14T12:23:52 < zyp> according to page 71 in F4 datasheet, you can't inject positive current into 5V tolerant IOs
2012-09-14T12:23:57 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-14T12:24:20 < zyp> jpa-, does that answer your question?
2012-09-14T12:24:31 < dongs> oh, f4.
2012-09-14T12:24:32 < jpa-> kind-of
2012-09-14T12:24:33 < dongs> this is f1.
2012-09-14T12:24:37 < dongs> isnt every pin on f4 FT.
2012-09-14T12:24:38 < jpa-> what is "not possible"
2012-09-14T12:24:41 < jpa-> what stops me :)
2012-09-14T12:24:42 < zyp> dongs, yes
2012-09-14T12:24:46 < jpa-> dongs: almost
2012-09-14T12:25:04 < zyp> jpa-, the lack of the internal diode conducting the current? :p
2012-09-14T12:25:46 < jpa-> but if i put a high level there, does it break? :)
2012-09-14T12:26:33 < zyp> it'll speced to Vcc+4V
2012-09-14T12:28:47 < jpa-> hmm, where does it say that?
2012-09-14T12:28:57 < zyp> page 70
2012-09-14T12:31:04 < jpa-> ah
2012-09-14T12:31:16 < jpa-> it's all in there and i just keep looking in the IO ports section :D
2012-09-14T12:31:34 < Laurenceb> i thought it had a diode to "VDD_5v"
2012-09-14T12:32:44 < zyp> and what is VDD_5v?
2012-09-14T12:33:05 < Laurenceb> zener to vdd?
2012-09-14T12:33:10 < Laurenceb> it doesnt say
2012-09-14T12:33:17 < Laurenceb> "Silicon Labs’ new Parametric Search iPad app "
2012-09-14T12:33:24 < Laurenceb> DO NOT WANT
2012-09-14T12:41:22  * Laurenceb is reading email
2012-09-14T12:41:26 < Laurenceb> "* * Want to feel great?  Do a "random act of kindness" for someone today! * *"
2012-09-14T12:41:33 < Laurenceb> should say trolling
2012-09-14T12:43:17 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-14T12:43:58 < Laurenceb> "<dongs> and as soon as I get a new 22nm desktop im installing win8 <dongs> yes, it is that awesome."
2012-09-14T12:44:06 < Laurenceb> ooh a proper troll
2012-09-14T12:46:05 < zyp> I also have a new 22nm desktop, maybe I should be as cool as dongs and install win8 on it
2012-09-14T12:46:48 < Laurenceb> i cant see your desktop
2012-09-14T12:52:36 < claude_work> of course you can't see his destop, you have to step back a bit : 22NM = 40744m
2012-09-14T12:52:59 < claude_work> win8 is huge!
2012-09-14T12:53:36 < karlp_> nm != N.M.
2012-09-14T12:53:46 < karlp_> computers at sea!
2012-09-14T12:59:42 < tavish> computers that run on sea
2012-09-14T13:03:23 < claude_work> 40744m == cloud computing
2012-09-14T13:05:27 < dongs> http://i.imgur.com/b1OWs.jpg loldongs
2012-09-14T13:06:06 < dongs> claude_work: loll
2012-09-14T13:07:53 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-14T13:08:28 < dongs> also got led bars for WS2811 testing
2012-09-14T13:08:32 -!- metaxa [metaxa@us.members.bombshellz.net] has quit [Read error: Operation timed out]
2012-09-14T13:09:08 -!- Ranzbak_ [~paul@wcrol.xs4all.nl] has quit [Remote host closed the connection]
2012-09-14T13:12:03 < Laurenceb> http://abusemark.com/store/images/naze32c_1.jpg
2012-09-14T13:12:08 < Laurenceb> nice silkscreen
2012-09-14T13:12:57 < Laurenceb> is that a second accel on there?
2012-09-14T13:16:38 < Laurenceb> im guessing thats not seeedstudio XD
2012-09-14T13:19:05 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-14T13:22:03 < zyp> dongs, blows.jpg
2012-09-14T13:22:31 -!- metaxa [~metaxa@us.members.bombshellz.net] has joined ##stm32
2012-09-14T13:24:00 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-14T13:28:22 < dongs> haha
2012-09-14T13:28:53 < dongs> Laurenceb: obviously not, seedstudio is useless even for most crap protos
2012-09-14T13:29:08 < Laurenceb> http://www.digitaltrends.com/computing/encyclopedia-dramatica-evolving/
2012-09-14T13:29:09 < Laurenceb> wut
2012-09-14T13:29:11 < Laurenceb> noooooo
2012-09-14T13:29:17 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 252 seconds]
2012-09-14T13:29:54 < dongs> yeah 2 accel.
2012-09-14T13:30:08 < Laurenceb> thats a freescale one?
2012-09-14T13:30:12 < dongs> yes
2012-09-14T13:30:16 < Laurenceb> cool
2012-09-14T13:30:17 < dongs> 8452/8451
2012-09-14T13:30:23 < Laurenceb> magno is too close to the usb tho
2012-09-14T13:30:25 < ABLomas> works ok, i tested :P
2012-09-14T13:30:27 < dongs> got a bag of both
2012-09-14T13:30:31 < Laurenceb> usb is steel usually
2012-09-14T13:30:40 < dongs> Laurenceb: hard iron offset takes that out
2012-09-14T13:30:46 < Laurenceb> hmm...
2012-09-14T13:30:50 < dongs> affects it, but not enough to matter.
2012-09-14T13:30:54 < Laurenceb> i see
2012-09-14T13:31:00 < dongs> i saw the sphere plots of hte shit or wahtever, looked ok
2012-09-14T13:31:10 < Laurenceb> ok
2012-09-14T13:32:06 < Laurenceb> lol @ http://cdn.digitaltrends.com/wp-content/uploads/2011/04/ed-wallpaper.jpg
2012-09-14T13:32:30 < dongs> is that wtc
2012-09-14T13:32:37 < dongs> yeah it is
2012-09-14T13:33:17 < Laurenceb> just spotted ass burgerzz   :D
2012-09-14T13:34:56 < dongs> haha, jewstars on planes
2012-09-14T13:36:06 < dongs> is the guy in center cho
2012-09-14T13:36:20 < Laurenceb> yes
2012-09-14T13:36:25 < dongs> lol
2012-09-14T13:37:12 < Laurenceb> fox5 news is the fake news article where they claim 4chan blew up their van
2012-09-14T13:37:39 < dongs> is ipod guy the dude who suicided over stolen ipod
2012-09-14T13:37:49 < Laurenceb> dunno lol
2012-09-14T13:37:54 < dongs> an hero
2012-09-14T13:38:06 < dongs> mitchel henderson
2012-09-14T13:38:13 < dongs> http://images.encyclopediadramatica.se/5/51/Anwinner.jpg
2012-09-14T13:38:27 < dongs> http://knowyourmeme.com/memes/an-hero
2012-09-14T13:38:31 < Laurenceb> hahaha
2012-09-14T13:40:53 < karlp_> eek, you got me reading the dramatica again
2012-09-14T13:41:02 < karlp_> that's a terribl way to start a day at work
2012-09-14T13:41:04 < Laurenceb> teehee
2012-09-14T13:41:14 < dongs> youll be there al lday
2012-09-14T13:41:34 < karlp_> nope, used my mad skillz to close it.
2012-09-14T13:42:02 < Laurenceb> lol javascript pr0n popupd
2012-09-14T13:43:53 < dongs> Just because 10% of the population of Canada is French, doesn't mean that I'm going to learn the language. Ten percent of the population is gay also, but I'm not going to learn how to suck a cock.
2012-09-14T13:58:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-14T14:09:53 < jpa-> huh.. http://www.mikroe.com/downloads/get/1834/mikromedia_for_stm32_schematics.pdf  does USB without HSE
2012-09-14T14:10:07 < jpa-> i wonder if really works so stable that one can use it in a product
2012-09-14T14:11:03 < jpa-> i guess calibrating it with the 32kHz RTC crystal helps
2012-09-14T14:12:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Read error: No route to host]
2012-09-14T14:13:56 < Laurenceb> i think its possible
2012-09-14T14:14:09 < Laurenceb> the temperature range will be restricted that all
2012-09-14T14:14:16 < Laurenceb> - for 12Mbps usb
2012-09-14T14:16:18 < jpa-> yeah it does work, though the datasheet says it is not supported
2012-09-14T14:17:53 -!- metaxa [~metaxa@us.members.bombshellz.net] has quit [Ping timeout: 260 seconds]
2012-09-14T14:18:46 < Laurenceb> i was going from datasheet tolerances
2012-09-14T14:20:49 < Laurenceb> http://www.mikroe.com/mikromedia/stm32-m4/
2012-09-14T14:20:50 < Laurenceb> nice
2012-09-14T14:22:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has joined ##stm32
2012-09-14T14:24:28 -!- metaxa [~metaxa@us.members.bombshellz.net] has joined ##stm32
2012-09-14T14:26:18 < Laurenceb> WTF
2012-09-14T14:26:22 < Laurenceb> CE failure
2012-09-14T14:26:27 < Laurenceb> "  No reference to risk of monitor & sensor becoming projectiles in event of a RTA. Ref BS EN 1789"
2012-09-14T14:27:07 < Laurenceb> they want me to risk asses risk of my kit being in a car accident and embedding in someones head
2012-09-14T14:27:14 < Laurenceb> surely trolling
2012-09-14T14:27:49 < tavish> embedded system
2012-09-14T14:27:56 < Laurenceb> loool
2012-09-14T14:28:10 < Laurenceb> i cant believe theres a standard for that
2012-09-14T14:29:19 < zyp> hah
2012-09-14T14:39:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-14T14:42:14 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-14T14:43:07 < dongs> zippe: http://www.youtube.com/watch?v=rdIWKytq_q4
2012-09-14T14:43:18 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-14T14:43:18 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-14T14:53:39 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-14T14:53:39 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-14T14:53:39 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-14T15:06:22 -!- R0b0t1` [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-09-14T15:28:30 -!- daku is now known as DaKu
2012-09-14T15:30:20 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-14T15:44:25 -!- dekar [~dekar@212.255.27.138] has joined ##stm32
2012-09-14T15:44:28 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-14T15:46:39 -!- dekar_ [~dekar@212.255.20.12] has quit [Ping timeout: 260 seconds]
2012-09-14T15:49:47 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-14T15:49:48 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-14T15:49:48 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-14T15:49:50 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-14T16:06:45 < Laurenceb> arg wtf
2012-09-14T16:06:53 < Laurenceb> chibios isnt using c99
2012-09-14T16:06:57 < Laurenceb> no wonder my code fails
2012-09-14T16:07:15 < dongs> http://i.imgur.com/D7HFz.jpg nasty handsoldered shit
2012-09-14T16:07:51 < Laurenceb> looks good to me
2012-09-14T16:08:04 < dongs> your code fails cuz you didn't pass --fno-more-spaces to the compiler
2012-09-14T16:08:52 < Laurenceb> whats the transistor for?
2012-09-14T16:09:06 < Laurenceb> also where are naked dutchess photos?!
2012-09-14T16:09:06 < dongs> diode?
2012-09-14T16:09:09 < Laurenceb> oh
2012-09-14T16:09:11 < dongs> external vcc vs usb
2012-09-14T16:10:20 < Laurenceb> cp2102 has proper rail isolation
2012-09-14T16:10:23 < Laurenceb> unlike ftdi
2012-09-14T16:10:51 < dongs> i wouldnt know, i'd never use ftdi shit in production
2012-09-14T16:10:52 < dongs> overpriced trash
2012-09-14T16:11:47 < Laurenceb> also ftdi 3v3 regulator is crap
2012-09-14T16:11:59 < dongs> i dont use the one on cp anyway.
2012-09-14T16:12:01 < Laurenceb> so many ardutards trying to run stuff of it and failing
2012-09-14T16:12:02 < dongs> i mean except internally.
2012-09-14T16:12:06 < dongs> haha
2012-09-14T16:12:07 < dongs> right
2012-09-14T16:12:18 < dongs> can it even do 100mA?
2012-09-14T16:12:18 < Laurenceb> due to noisy poorly regulated mess
2012-09-14T16:12:19 < dongs> i doubt it, to stay under usb bus poweredspec
2012-09-14T16:12:23 < Laurenceb> no - 50ma
2012-09-14T16:12:32 < Laurenceb> but even at 20ma it fails aiui
2012-09-14T16:13:12 <+izua> dongs: why is it not pink?:(
2012-09-14T16:13:44 < Laurenceb> # List all user C define here, like -D_DEBUG=1
2012-09-14T16:13:44 < Laurenceb> UDEFS = -std=gnu99
2012-09-14T16:13:47 < Laurenceb> fixed
2012-09-14T16:13:48 < dongs> izua: cuz i got the boards in 3 days instead of 10
2012-09-14T16:13:59 < Laurenceb> ...
2012-09-14T16:14:07 < Laurenceb> so you have pink ones shipping?
2012-09-14T16:14:10 <+izua> but they'll be pink in production, right?
2012-09-14T16:14:17 < dongs> yea when i do the panel
2012-09-14T16:14:19 < Laurenceb> omfg
2012-09-14T16:14:23 <+izua> awesome
2012-09-14T16:14:32 <+izua> you should ship them in a box with Laurenceb's ponies
2012-09-14T16:14:36 < Laurenceb> lol
2012-09-14T16:17:34 < Laurenceb> code going to githubz shortly
2012-09-14T16:18:10 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-14T16:18:44 < Laurenceb> http://www.b3tards.com/u/38d3fc0972e7b44e8466/kate-middleton-tits-prince-william-and-bubbles--herman.gif
2012-09-14T16:19:35 < BrainDamage> what's with the etched ic markings dongs?
2012-09-14T16:19:36 <+izua> ...
2012-09-14T16:20:16 < dongs> wat
2012-09-14T16:20:21 < dongs> i didnt etch shit
2012-09-14T16:20:27 < dongs> blame shit lights + camera
2012-09-14T16:20:51 < BrainDamage> ah k
2012-09-14T16:21:46 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-09-14T16:21:57 < dongs> its just f103cb
2012-09-14T16:22:00 < dongs> no sekrets
2012-09-14T16:22:21 <+izua> dongs:
2012-09-14T16:22:52 <+izua> why did you choose an usb adapter chip, instead of going with the internal usb hardware?
2012-09-14T16:23:07 -!- zippe [~Adium@173.11.99.161] has quit [Ping timeout: 240 seconds]
2012-09-14T16:23:20 < dongs> because: thats how I want to use it, and i cant be arsed writing a USB bootloader to reflash
2012-09-14T16:23:22 <+izua> except the comfort of not having to write drivers and use a usb library on the chip and full speed
2012-09-14T16:23:23 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-14T16:23:28 < dongs> yes
2012-09-14T16:23:37 <+izua> so only that?
2012-09-14T16:23:40 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-14T16:23:43 < dongs> correct.
2012-09-14T16:23:45 < dongs> i'm fucking lazy.
2012-09-14T16:23:54 <+izua> i always tend to think usb/serial chips are a compromise solution
2012-09-14T16:24:04 <+izua> (although i'm a fan of ft232 myself)
2012-09-14T16:24:20 < dongs> i have a reel of them.
2012-09-14T16:24:25 < zyp> he's on windows, so custom usb stuff is just pain with custom drivers
2012-09-14T16:24:26 < dongs> i use them in other shit.
2012-09-14T16:24:30 < dongs> yeah
2012-09-14T16:24:34 < zyp> :p
2012-09-14T16:24:54 <+izua> you do get driverless instalation, ends up as a com port so it's easy to interface with any software, plus, free cpu time and ram on the controller
2012-09-14T16:26:17  * Laurenceb wrote f103 bootloader in 1 day
2012-09-14T16:26:20 < zyp> a simple ACM stack doesn't really consume much runtime resources
2012-09-14T16:26:26 < Laurenceb> using copypasta iar project XD
2012-09-14T16:26:31 < dongs> shrug
2012-09-14T16:26:44 < dongs> this is how i roll
2012-09-14T16:27:14 < zyp> dongs, and it's fine really
2012-09-14T16:27:18 < Laurenceb> is that a murata resonator?
2012-09-14T16:27:38 < dongs> it might be
2012-09-14T16:27:41 < zyp> nothing stopping anybody for hooking up usb if they really want to
2012-09-14T16:27:42 < dongs> its hte hceapest one in that size on dkey
2012-09-14T16:27:44 < dongs> i know that much
2012-09-14T16:27:50 < zyp> from*
2012-09-14T16:27:51 < Laurenceb> isnt the tolerance piss poor?
2012-09-14T16:28:16 < dongs> its a lso one of recommended ones on the stm32 clocking appnote
2012-09-14T16:28:37 -!- qyx_ [~qyx@krtko.org] has quit [Ping timeout: 240 seconds]
2012-09-14T16:28:45 -!- qyx_ [~qyx@krtko.org] has joined ##stm32
2012-09-14T16:28:48 < dongs> http://www.digikey.com/product-detail/en/CCR8.0MXC8T/445-5232-1-ND/2179638
2012-09-14T16:28:49 < dongs> sorry its
2012-09-14T16:28:51 < dongs> tdk
2012-09-14T16:28:55 < dongs> and now im fucked cuz its nonstock.
2012-09-14T16:28:58 < dongs> so ill have to find osmething else.
2012-09-14T16:29:07 < dongs> luckily its a common type
2012-09-14T16:29:19 < dongs> ill prob get something from china.
2012-09-14T16:31:33 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-09-14T16:35:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-14T16:36:11 < Laurenceb> ±0.5%
2012-09-14T16:36:16 < Laurenceb> wow thats shit
2012-09-14T16:37:32 < karlp_> twice as good as HSI.
2012-09-14T16:37:39 < karlp_> what sort of accuracy are you demanding?
2012-09-14T16:37:41 < dongs> its a resonator, you fuckwit
2012-09-14T16:37:45 < dongs> WITH built in caps
2012-09-14T16:37:49 < karlp_> it's a breakout board anyway
2012-09-14T16:37:49 < dongs> in 3.2x1.2mm
2012-09-14T16:37:53 < dongs> find something better, asshole
2012-09-14T16:38:20 < dongs> lets see what PWM out looks like
2012-09-14T16:42:50 < dongs> http://i.imgur.com/65XUd.png this is pwm jitter. I think I'll manage.
2012-09-14T16:43:16 < dongs> on shitmega, that stuff walks around +-400ns
2012-09-14T16:43:20 < dongs> wiht intrc
2012-09-14T16:48:28 < Laurenceb> addified
2012-09-14T16:48:34 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro
2012-09-14T16:49:04 < karlp_> this is the one we're meant to clone before you have to pull it right?
2012-09-14T16:49:08 < Laurenceb> thats like a few ns
2012-09-14T16:49:13 < Laurenceb> lols
2012-09-14T16:49:18 < Laurenceb> no thats my logger
2012-09-14T16:49:33 < jpa-> it's quite nice that you are allowed to release it
2012-09-14T16:49:38 < Laurenceb> this is my fully threaded version.. it very boring
2012-09-14T16:49:45 < Laurenceb> im not, shhh
2012-09-14T16:50:02 < Laurenceb> but this is really only there for me to track
2012-09-14T16:50:09 < Laurenceb> as i said, its very very boring
2012-09-14T16:50:25 < BrainDamage> I'm impressed you didn't commit binary blobs this time
2012-09-14T16:50:32 < Laurenceb> lol
2012-09-14T16:51:40 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/PID_Control.c
2012-09-14T16:51:48 < Laurenceb> worlds most abstracted pid controller
2012-09-14T16:52:34 < dongs> how so
2012-09-14T16:52:36 < dongs> looks normal
2012-09-14T16:52:39 < dongs> ive seen similar shit elewhere
2012-09-14T16:52:41 < dongs> else
2012-09-14T16:52:46 < dongs> infact you probly copied it fromsomwehre
2012-09-14T16:53:16 < Laurenceb> hehe
2012-09-14T16:53:22 < Laurenceb> how did you work that out
2012-09-14T16:55:27 < Laurenceb> just need to write timer code
2012-09-14T16:55:42 < Laurenceb> chibios includes the stmf10x.h header from st right?
2012-09-14T16:56:01 < jpa-> it includes it, but it doesn't #include it
2012-09-14T16:56:01 < Laurenceb> i can just copy over register definitions
2012-09-14T16:56:07 < Laurenceb> obviously
2012-09-14T16:56:14 < Laurenceb> but its there... somewhere
2012-09-14T16:56:20 < dongs> also got my WS2811 breakout boards http://i.imgur.com/9wadN.jpg
2012-09-14T16:56:26 < jpa-> yeah, just type #include <stm32f10x.h>  and you'll have it
2012-09-14T16:56:32 < Laurenceb> cool
2012-09-14T16:56:58 < Laurenceb> looks like pcbtrain packaging
2012-09-14T16:58:12 < dongs> its obviously not
2012-09-14T16:58:22 < dongs> cuz silk isnt shittastic fail
2012-09-14T16:59:07 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-14T16:59:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 260 seconds]
2012-09-14T16:59:39 -!- karlp_ is now known as karlp
2012-09-14T16:59:59 < dongs> st i2c peripheral crashes hard if thres no pullups lol
2012-09-14T17:00:12 < Laurenceb> thats normal
2012-09-14T17:03:36 -!- metaxa [~metaxa@us.members.bombshellz.net] has quit [Ping timeout: 255 seconds]
2012-09-14T17:06:35 < Laurenceb> http://news.bbcimg.co.uk/media/images/62900000/jpg/_62900244_62900243.jpg
2012-09-14T17:06:38 < Laurenceb> NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
2012-09-14T17:07:15 < dongs> someone wtc'd KFC?
2012-09-14T17:07:27 < Laurenceb> i think it was the jewz
2012-09-14T17:07:44 < Laurenceb> KFC: inside job
2012-09-14T17:07:52 < Laurenceb> dont taze me bro
2012-09-14T17:09:48 < karlp> Laurenceb: so, more princess boobs, less animated stupdi heads on boobs
2012-09-14T17:10:10 < Laurenceb> yeah but the internet isnt providing them :(
2012-09-14T17:10:21 < dongs> btw, the druggie taiwankid was justin lee
2012-09-14T17:10:29 < karlp> but... it's in french _print_ media!
2012-09-14T17:10:36 < karlp> hang on, let me get some frenchies on teh case
2012-09-14T17:10:50 < Laurenceb> i dont actually want to see her tits
2012-09-14T17:11:03 < Laurenceb> or ill need to order some viagara
2012-09-14T17:11:07 < karlp> hehe
2012-09-14T17:11:30 < bgamari> zippe, Nope, it should be in .data
2012-09-14T17:12:45 < bgamari> zippe, Address is 0x10000a64
2012-09-14T17:13:01 < bgamari> re: What could cause a TEIF error during a DMA transfer from and ADC to memory?
2012-09-14T17:13:43 < jpa-> 0x1000xxxx is core-coupled memory, DMA cannot access it
2012-09-14T17:13:57 < jpa-> only 0x2000xxxxx and 0x0800xxxx is accessible to DMA
2012-09-14T17:14:33 < bgamari> Yeah, I guess so
2012-09-14T17:14:39 < bgamari> Bah
2012-09-14T17:17:59 < zyp> jpa-, and FSMC
2012-09-14T17:21:56 < jpa-> zyp: ah, yeah
2012-09-14T17:22:03 < jpa-> and peripheral areas anyway
2012-09-14T17:22:04 < Laurenceb> nice - my chibios project is only 18KB
2012-09-14T17:22:27 < Laurenceb> as opposed to 121 for my other version...
2012-09-14T17:22:42 < bgamari> Is there a conventional section name for things that can't go in CCM?
2012-09-14T17:23:19 < karlp> wouldn't you use a special section name for things that _can_ got into CCM?
2012-09-14T17:23:37 < jpa-> i would :)
2012-09-14T17:23:39 < bgamari> yes
2012-09-14T17:23:46 < bgamari> I suppose so
2012-09-14T17:23:52 < jpa-> but, i guess it's the same which way around you do it
2012-09-14T17:24:05 < jpa-> one way might be to put stack in CCM and bss in normal RAM
2012-09-14T17:24:09 < bgamari> My current linker script puts all of .bss into .CCM
2012-09-14T17:24:23 < bgamari> scratch the dot
2012-09-14T17:25:23 < bgamari> Are normal CCM accesses lower latency than normal RAM?
2012-09-14T17:26:07 < jpa-> yes
2012-09-14T17:26:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-14T17:26:22 < jpa-> 2-4 waitstates for normal ram IIRC (haven't checked)
2012-09-14T17:28:46 < claude_work> dongs , sorry for my ignorance . but who is justin lee , a clebrity or so? google didn't help but at these is a nice ad next to the article on yahoo :) http://i.imgur.com/36SxA.png
2012-09-14T17:29:08 < dongs> ya that guy
2012-09-14T17:29:18 < dongs> some dude in taiwan who was drugging up celebs and fucking them while they're passed out
2012-09-14T17:29:29 < dongs> then he took his g aming rig to a computer shop to get it fixed
2012-09-14T17:29:35 < dongs> and they found like 100s of gigs of pr0n recordeings
2012-09-14T17:29:39 < dongs> and promptly torrented them
2012-09-14T17:29:59 < claude_work> yeah i know that , but just some random dude with 100gigs of selfmade porn ?
2012-09-14T17:30:09 < dongs> yea, who's going to jail now i guess.
2012-09-14T17:30:14 < dongs> i dont think its legal to drug + rape..
2012-09-14T17:30:19 < claude_work> ah ok
2012-09-14T17:31:43 -!- metaxa [~metaxa@us.members.bombshellz.net] has joined ##stm32
2012-09-14T17:31:54 < claude_work> sounded like justin lee is a important person , thats why i was asking :)
2012-09-14T17:33:23 < dongs> nah dont think he was anyone important
2012-09-14T17:38:32 < claude_work> so leason learnd from that : keep two rigs , one for erhmm"business stuff" and the other for gaming
2012-09-14T17:44:56 < dongs> yeah.
2012-09-14T17:56:09 -!- tavish [~tavish@120.59.132.51] has joined ##stm32
2012-09-14T17:56:09 -!- tavish [~tavish@120.59.132.51] has quit [Changing host]
2012-09-14T17:56:09 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-14T18:16:33 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: Lost terminal]
2012-09-14T18:16:37 -!- diydsp [41df9b1e@gateway/web/freenode/ip.65.223.155.30] has joined ##stm32
2012-09-14T18:18:23 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-14T18:18:28 < Laurenceb> http://hackaday.com/2012/09/14/testing-30-brands-of-batteries/
2012-09-14T18:18:40 < Laurenceb> hmm who could this "Thomas"  be
2012-09-14T18:19:15 < diydsp> hey all!  I'm building this guy's board:  http://ebrombaugh.studionebula.com/synth/stm32f4_codec/index.html
2012-09-14T18:19:38 < diydsp> I'm filling out his BOM.xls for 100% Mouser if possible...
2012-09-14T18:19:58 < diydsp> having trouble with this component though from digikey:  http://www.digikey.com/scripts/DKSearch/dksus.dll?Detail&itemSeq=119605019&uq=634831427566303588
2012-09-14T18:20:05 < BrainDamage> talk to emeb, that's his board, he joins this channel from time to time
2012-09-14T18:20:25 < diydsp> it appears to be a coil, but not used in the way I'm used to... :)  inductance isn't specified.
2012-09-14T18:20:42 < diydsp> and yes emeb is awesome!  i've been following him for years on a synth diy mailing list
2012-09-14T18:20:43 < BrainDamage> also, the digikey link doesn't work
2012-09-14T18:21:33 < diydsp> http://www.digikey.com/product-detail/en/BLM18BD252SN1D/490-1028-2-ND/584239   hmm other one prolly had some session info
2012-09-14T18:23:03 < BrainDamage> it's an high frequency choke
2012-09-14T18:24:03 < BrainDamage> there's the impedance specified at 100MHz, that gives you ESR + 2*pi*100MHz*L
2012-09-14T18:24:40 < BrainDamage> given that ESR is 1.5Ohm, and neglectable, you can say that impedance is all given by L
2012-09-14T18:25:22 < BrainDamage> so L=2.5kOHm/(2⋅pi⋅100MHz) ~= 4uH
2012-09-14T18:25:37 < BrainDamage> this assumes ofc that self resonance is far off it, so it behaves purely inductively
2012-09-14T18:28:42 < Laurenceb> ok wtf
2012-09-14T18:28:51 < Laurenceb> table60 is missing from f4 ref manual
2012-09-14T18:29:44 < diydsp> wow BrainDamage, that almost makes sense :)  srsly thank you.
2012-09-14T18:29:49 < diydsp> I
2012-09-14T18:29:57 < diydsp> I'll share my BOM when it's done.
2012-09-14T18:36:25 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Read error: Operation timed out]
2012-09-14T18:37:42 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-14T18:38:22 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-235-219.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-14T18:41:39 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 260 seconds]
2012-09-14T18:49:19 < bgamari> Hmm, for some reason my DMA transfer from ADC to memory only runs once, despite CONT being set
2012-09-14T18:49:27 < bgamari> The ADC appears to still be sampling
2012-09-14T18:49:42 < bgamari> the stream is setup with CIRC
2012-09-14T18:50:01 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-14T18:50:05 < bgamari> NDTR returned to its shadow value on finishing the first run
2012-09-14T18:50:10 < bgamari> but isn't ticking down
2012-09-14T18:50:22 -!- claude_work is now known as claude
2012-09-14T18:56:03 < bgamari> None of the DMA status flags are set
2012-09-14T18:56:31 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-09-14T18:56:59 -!- BrainDamage [~BrainDama@62.10.7.25] has joined ##stm32
2012-09-14T18:58:09 < bgamari> ADC1_SR_STRT is set, so clearly the ADC is running
2012-09-14T18:59:40 < dongs> youre in circuilar DMA mode?
2012-09-14T18:59:44 < bgamari> I'm not sure whether it's the DMA or the ADC that's failing
2012-09-14T18:59:46 < bgamari> yep
2012-09-14T18:59:58 < dongs> scanconv is enabled, continuous conv is enabled?
2012-09-14T19:00:22 < dongs> http://bcas.tv/paste/results/2SXrei77.html this works for me
2012-09-14T19:00:25 < bgamari> dongs, Yep
2012-09-14T19:02:05 < bgamari> hmm
2012-09-14T19:08:53 < bgamari> It just doesn't make any sense
2012-09-14T19:09:10 < bgamari> I can stop the DMA transaction and restart it; doesn't make any difference
2012-09-14T19:10:23 < bgamari> stop the ADC and restart it; again, no differencem
2012-09-14T19:10:28 < bgamari> scratch that
2012-09-14T19:10:37 < bgamari> hmm, perhaps the ADC is hung?
2012-09-14T19:10:51 < bgamari> I can clear CONT and SR doesn't seem to change
2012-09-14T19:12:09 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-14T19:12:31 < Laurenceb> http://www.st.com/stonline/stappl/resourceSelector/app?page=resourceSelector&doctype=REFERENCE_MANUAL&SubClassID=1605
2012-09-14T19:12:34 < Laurenceb> niceeeeeeeee
2012-09-14T19:13:03 < jpa-> what's cool this time?
2012-09-14T19:13:20 < karlp> it's the "other" f3s
2012-09-14T19:13:38 < karlp> 37 and 38, not just 30 and 31
2012-09-14T19:14:09 < karlp> 200mA is "low draw" ?!
2012-09-14T19:14:26 < jpa-> less than rpi :)
2012-09-14T19:16:35 < Laurenceb> opamps
2012-09-14T19:17:08 < TitanMKD> anyone worked on a design to accurately (at least +/-1microAmp resolution) measure power consumption ?
2012-09-14T19:19:01 < jpa-> TitanMKD: precision is not a trouble if you don't need much range
2012-09-14T19:19:04 < Laurenceb> karlp: where is current spec?
2012-09-14T19:19:07 < jpa-> what is the maximum current you need?
2012-09-14T19:19:15 < karlp> on the battery tester page you linked
2012-09-14T19:19:46 < TitanMKD> jpa- yes it is more hard on full range from 1microA to 1000mA ;)
2012-09-14T19:20:08 < TitanMKD> but anyway it is better to ask for accurate current measure in %
2012-09-14T19:20:48 < TitanMKD> jpa- i ask that in order to build a board to help to develop more clever/power friendly software
2012-09-14T19:21:00 < Laurenceb> oh
2012-09-14T19:21:13 < Laurenceb> i thought it was current draw from f3  :P
2012-09-14T19:21:26 < Laurenceb> battery tester guy is the soldering fail guy from rcgroups
2012-09-14T19:21:41 < jpa-> TitanMKD: so, do you need to measure 1uA to 1A with 1% accuracy of the reading, or do you need to measure 1A currents with 1uA accuracy?
2012-09-14T19:22:31 < TitanMKD> jpa- the 1st is maybe not bad ;) but 1uA accuracy from 1uA to 1A will be a must
2012-09-14T19:22:45 < jpa-> ok
2012-09-14T19:22:52 < jpa-> so 20 bits; not impossible, but difficult
2012-09-14T19:23:02 < jpa-> how much bandwidth do you need?
2012-09-14T19:23:06 < TitanMKD> jpa- the problem is on low current with burden voltage ...
2012-09-14T19:23:42 < jpa-> instrumentation amplifiers can amplify very small voltages quite accurately
2012-09-14T19:23:46 < TitanMKD> jpa- something like 100Hz will be enough in fact
2012-09-14T19:23:49 < jpa-> nanovolts is not impossible
2012-09-14T19:24:39 < BrainDamage> you could use hall current sensors if burden voltage is an issue, you'd have to check their internal noise tough
2012-09-14T19:25:01 < TitanMKD> and the most difficult is to check multiple voltage line ;)
2012-09-14T19:25:06 < TitanMKD> 3.3V and 5V
2012-09-14T19:25:16 < jpa-> and getting 1ppm accuracy with hall sensors may be quite difficult
2012-09-14T19:25:25 < jpa-> it is a heck of a challenge using resistors also
2012-09-14T19:25:51 < jpa-> TitanMKD: do you need 1uA absolute accuracy or just resolution
2012-09-14T19:25:53 < TitanMKD> jpa- but the idea behind is to measure accurately realtime power consumption and average on 1s ...
2012-09-14T19:26:23 < TitanMKD> jpa- the best you can in fact ;)
2012-09-14T19:26:34 < jpa-> TitanMKD: then why do you need 1uA accuracy at 1A currents?
2012-09-14T19:26:35 < TitanMKD> jpa- for a price not exceeding 20us$ ;)
2012-09-14T19:26:45 < jpa-> wouldn't 1% accuracy overall be good enough?
2012-09-14T19:27:15 < TitanMKD> jpa- yes for 1A, 1uA resolution is a bit stupid so maybe just a percent accuracy is realist
2012-09-14T19:28:04 < jpa-> ok, then you can simply implement multiple measurement ranges (two could be enough) and automatically switch between them
2012-09-14T19:28:05 < TitanMKD> jpa- in fact with such tool we could also analyze which instruction is executed depending on voltage & current if it is really accurate
2012-09-14T19:28:30 < TitanMKD> but that will require very fast measure like 100msps ;)
2012-09-14T19:28:37 < TitanMKD> but it's not the purpose now
2012-09-14T19:29:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-14T19:29:38 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-14T19:31:01 < TitanMKD> jpa- maybe there's some magic components for that from TI
2012-09-14T19:31:49 < TitanMKD> jpa- like for the smart meter stuff
2012-09-14T19:31:58 < jpa-> smart meters would be AC
2012-09-14T19:32:22 < TitanMKD> yes AC is not interesting only DC
2012-09-14T19:32:54 -!- claude is now known as Claude
2012-09-14T19:33:11 < jpa-> but hall current sensors are a good way if e.g. 1% accuracy is enough
2012-09-14T19:33:38 < jpa-> you could put a 1A and a 100mA one in series to get better accuracy at lower currents
2012-09-14T19:33:59 < TitanMKD> yes like multimeter does
2012-09-14T19:34:45 < TitanMKD> maybe i should look the EEVBLOG http://www.eevblog.com/projects/ucurrent/
2012-09-14T19:34:53 < jpa-> usually multimeters have ranges that need to be switched manually
2012-09-14T19:34:57 < jpa-> like ucurrent has also
2012-09-14T19:34:57 < TitanMKD> it is exactly what i want but in an automatic range way ;)
2012-09-14T19:35:09 < TitanMKD> and done by a MCU
2012-09-14T19:35:20 < jpa-> that is a problem if your processor wakes up e.g. every 10 ms and consumes 100x more current then
2012-09-14T19:35:33 < TitanMKD> to output data in realtime through USB to a PC ...
2012-09-14T19:35:56 < TitanMKD> jpa- yes it is why it could be required to be fast like 10MSPS ;)
2012-09-14T19:36:11 < TitanMKD> and to average data later ...
2012-09-14T19:36:36 < BrainDamage> you won't necessarily see those peaks, there's averaging circuits on power rails, aka LC filters
2012-09-14T19:36:39 < TitanMKD> or 7.5MSPS for STM32F4 ;)
2012-09-14T19:36:50 < BrainDamage> as in, they'll be highly attenuated
2012-09-14T19:37:12 < TitanMKD> BrainDamage yes but sometimes it could be great to know current "glitch" ;)
2012-09-14T19:37:26 < TitanMKD> that can also show a potential problem
2012-09-14T19:37:47 < jpa-> TitanMKD: you can average analogically
2012-09-14T19:38:14 < TitanMKD> jpa- yes it could be an option depending on what we want to see ;)
2012-09-14T19:38:23 < TitanMKD> average current or instantaneous current
2012-09-14T19:38:47 < TitanMKD> anyway the Dave Jones ucurrent seems a must for the base design
2012-09-14T19:39:45 < Laurenceb> https://www.youtube.com/watch?&v=eky-q9CE_co#
2012-09-14T19:39:47 < Laurenceb> lolling
2012-09-14T19:39:49 < jpa-> it's not that special
2012-09-14T19:46:23 -!- prog` is now known as prog
2012-09-14T19:57:28 < TitanMKD> jpa- it seems hard to do it with just a 12bits ADC (which have in fact only 10bits)
2012-09-14T19:58:11 < jpa-> TitanMKD: you can put a 1000x preamplifier on one channel
2012-09-14T19:58:18 < jpa-> and then sample both of them
2012-09-14T19:59:36 < TitanMKD> jpa- yes
2012-09-14T19:59:39 < BrainDamage> you can also oversample and extract extra bits from fltering
2012-09-14T20:03:44 < TitanMKD> in fact with autorange and ucurrent design 10bits is just enough ;)
2012-09-14T20:04:23 < TitanMKD> it is 1mV/nA (from 0 to 1000nA) or 1mV/uA (from 0 to 1000uA) ...
2012-09-14T20:04:34 < TitanMKD> so only 1000points ;)
2012-09-14T20:04:40 < BrainDamage> the problem with autorange is fast transitions between each ranges, to prevent saturation ah wait, I forgot, ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ah wait, I forgot, ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) ;) 
2012-09-14T20:04:57 < HTT-Bird> BrainDamage: what was that? :p
2012-09-14T20:05:11 < BrainDamage> just a reference to his line-to-;) ratio
2012-09-14T20:05:42 < BrainDamage> + some copypaste fail  since I include 3 times "ah, wait I forgot"
2012-09-14T20:06:01 < TitanMKD> hehe
2012-09-14T20:06:12 < jpa-> TitanMKD: but implementing fast autorange in ucurrent is not easy
2012-09-14T20:06:44 < karlp> just put an extra digit on and don't autorange.
2012-09-14T20:07:03 < karlp> the autoranging meters I have are the slowest and most tedious to use.
2012-09-14T20:07:17 < TitanMKD> jpa- yes it is the challenge ;)
2012-09-14T20:07:39 < HTT-Bird> karlp: cheapies? or ones with a decent name on them?
2012-09-14T20:08:06 < BrainDamage> connect together 2 autoranging mm, one set to current, the other voltage and you get nice patterns from all the autoranging
2012-09-14T20:10:41 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-14T20:12:12 < TitanMKD> anyone know on uCurren what happen if the wrong range is used like 0-1000nA mode when there's 200mA as input ?
2012-09-14T20:12:22 < TitanMKD> I hope to does not kill any component
2012-09-14T20:12:27 < TitanMKD> and just saturate the value
2012-09-14T20:14:55 < TitanMKD> in that case autorange is very easy jsut check is value is saturate ;) and change mode +1 ;)
2012-09-14T20:15:08 < TitanMKD> could be done very quickly
2012-09-14T20:21:51 -!- barthess [~barthess@178.154.77.159] has joined ##stm32
2012-09-14T20:24:30 < jpa-> TitanMKD: it might fry the shunt resistor
2012-09-14T20:24:45 <+Steffanx> nu ;)
2012-09-14T20:24:54 < jpa-> and also, would not be done quickly enough, the voltage drop would increase and device might crash
2012-09-14T20:25:56 < TitanMKD> jpa- first 1 will try to build the original uCurrent ;)
2012-09-14T20:26:13 < TitanMKD> jpa- if some are interested in my spare time i will do a schematic+ocb with Kicad ;)
2012-09-14T20:26:20 < TitanMKD> ocb->pcb
2012-09-14T20:27:28 < jpa-> is your every line a joke?
2012-09-14T20:28:51 < TitanMKD> yes because it is funny project
2012-09-14T20:32:27 <+Steffanx> ucurrent .. i just received mine last week :)
2012-09-14T20:32:49 < TitanMKD> and it work like expected ?
2012-09-14T20:32:56 <+Steffanx> It looks ok
2012-09-14T20:33:14 <+Steffanx> ( not from eevblog though, but almost the same design for less money )
2012-09-14T20:33:17 < TitanMKD> it is amazing but i cannot buy one to Dave Jones as i do not live in Australia
2012-09-14T20:33:29 < TitanMKD> Steffanx where did you bought it ?
2012-09-14T20:33:42 <+Steffanx> A dutch "buy things together" website
2012-09-14T20:34:15 < TitanMKD> ok
2012-09-14T20:34:37 -!- DaKu is now known as daku
2012-09-14T20:35:50 <+Steffanx> I wonder if dutchland is the only country with a website like that
2012-09-14T20:38:41 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-14T20:46:57 < TitanMKD> anyone tested the STM32F4 ADC ?
2012-09-14T20:47:22 < TitanMKD> i'm curious to know if accuracy is really 12bits or maybe just 10bits (with 2bits noise)
2012-09-14T20:48:00 < TitanMKD> maybe it also depends on STM32F4 package & rev (+ of course the board pcb/design)
2012-09-14T20:49:25 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 260 seconds]
2012-09-14T20:51:14 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-14T20:59:51 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-14T21:05:02 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-14T21:05:42 < BrainDamage> Steffanx: groupon
2012-09-14T21:06:04 <+Steffanx> Nah, that's 10000% commercial
2012-09-14T21:06:48 < BrainDamage> I know lot of people here do "buy groups", but it's generally a local reality, not organized trough webpages, etc
2012-09-14T21:07:12 <+Steffanx> Yeah, it's more similar too that
2012-09-14T21:13:22 < bgamari> What is a good rule of thumb for JTAG interface speed?
2012-09-14T21:13:48 < bgamari> Should it be set as a fraction of the MCU clockrate?
2012-09-14T21:18:37 -!- tavish_ [~tavish@120.56.137.76] has joined ##stm32
2012-09-14T21:20:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-09-14T21:24:32 < Thorn> bgamari: jtag clock is completely separate from other clocks
2012-09-14T21:27:26 < bgamari> For some reason openocd has started giving me "corrupted fifo read pointer 0x43c0488" on the second "flash write_image" of every session
2012-09-14T21:28:03 < bgamari> Occasionally even the first write_image
2012-09-14T21:28:35 < Thorn> try lowering the frequency if it's high (like >1MHz)
2012-09-14T21:32:10 < HTT-Bird> yeah, what is a good rule of thumb for JTAG or SWD clock
2012-09-14T21:32:11 < HTT-Bird> ?
2012-09-14T21:32:13 < tavish_> jtag/st-link dont always work for me too. for openocd, i close openocd, start again; connect gdb without load or reset; close gdb; connect gdb with load and halt
2012-09-14T21:32:22 < HTT-Bird> I'm only running my part at 1MHz in the first place...
2012-09-14T21:32:24 < tavish_> works most of the times, but not always
2012-09-14T21:33:02 < bgamari> Thorn, Seems to be getting worse
2012-09-14T21:33:05 < bgamari> even happens with 500kHz
2012-09-14T21:33:12 < bgamari> Can't seem to program the part at all now
2012-09-14T21:33:14 < HTT-Bird> I mean, my CPU core clock is 1MHz
2012-09-14T21:33:52 < bgamari> I wonder if the FLASH is beginning on fail
2012-09-14T21:34:08 < bgamari> I've been reprogramming the part constantly for the last few days
2012-09-14T21:34:50 < TitanMKD> bgamari sometimes to program the flash it requires higher CPU freq else flash write fail
2012-09-14T21:35:25 < bgamari> TitanMKD, Higher MCU freq?
2012-09-14T21:38:06 < TitanMKD> yes i read that for STM32F4
2012-09-14T21:38:21 < TitanMKD> like the MCU shall run at more than 32MHz or something like that
2012-09-14T21:40:22 < qyx_> bgamari: reset the part to system loader and then try to program using jtag
2012-09-14T21:40:47 < bgamari> qyx_, Good idea
2012-09-14T21:53:47 < qyx_> does it work?
2012-09-14T22:21:54 < bgamari> qyx_, Seems to
2012-09-14T22:24:29 -!- Claude is now known as claude
2012-09-14T22:42:32 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-09-14T22:55:49 -!- tavish_ [~tavish@120.56.137.76] has quit [Quit: Z]
2012-09-14T23:06:10 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-14T23:06:10 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-14T23:06:10 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-14T23:06:13 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-14T23:16:15 <+Steffanx> \
2012-09-14T23:17:24 < zyp> /
2012-09-14T23:17:46 < prog> |
2012-09-14T23:18:35 < karlp> -
2012-09-14T23:20:17 <+Steffanx> o
2012-09-14T23:20:26 < prog> no
2012-09-14T23:20:32 < prog> /
2012-09-14T23:20:33 <+Steffanx> Or .
2012-09-14T23:20:40 <+Steffanx> ( lines smashed on the ground )
2012-09-14T23:21:02 < zyp> <
2012-09-14T23:21:05 < zyp> I broke it :(
2012-09-14T23:21:22 <+Steffanx> =
2012-09-14T23:26:25 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection]
2012-09-14T23:28:58 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-14T23:35:32 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-14T23:44:29 < bgamari> even despite the SPI inadequacies it seems I can still push samples to my DAC at 65 kHz
2012-09-14T23:45:52 -!- barthess [~barthess@178.154.77.159] has quit [Quit: Leaving.]
2012-09-14T23:57:01 -!- barthess [~barthess@178.154.77.159] has joined ##stm32
--- Day changed Sat Sep 15 2012
2012-09-15T00:30:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-15T00:50:04 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-15T01:01:57 < Laurenceb_> http://cdn03.cdn.egotastic.com/wp-content/uploads/2012/09/14/kate-middleton-topless-closer-06-580x435.jpg
2012-09-15T01:02:22 < Laurenceb_> i thought she was a bit flat chested then i realised i was looking at the one on the left
2012-09-15T01:04:57 -!- barthess [~barthess@178.154.77.159] has quit [Quit: Leaving.]
2012-09-15T01:05:35 -!- barthess [~barthess@178.154.9.47] has joined ##stm32
2012-09-15T01:22:13 < zyp> we already know you prefer looking at men
2012-09-15T01:37:14 -!- barthess [~barthess@178.154.9.47] has quit [Remote host closed the connection]
2012-09-15T01:39:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T01:46:25 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T02:04:15 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-15T02:05:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-15T02:19:52 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has left ##stm32 []
2012-09-15T02:58:10 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-15T03:03:32 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 248 seconds]
2012-09-15T03:24:01 -!- dekar_ [~dekar@212.255.20.51] has joined ##stm32
2012-09-15T03:24:04 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-15T03:25:44 -!- dekar [~dekar@212.255.27.138] has quit [Ping timeout: 255 seconds]
2012-09-15T03:42:29 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds]
2012-09-15T04:06:00 < dongs> hello trolls
2012-09-15T04:46:07 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-15T04:51:29 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T04:51:30 -!- BrainDamage [~BrainDama@62.10.7.25] has quit [Disconnected by services]
2012-09-15T04:51:46 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Client Quit]
2012-09-15T04:51:55 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T04:53:07 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 240 seconds]
2012-09-15T05:04:19 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Ping timeout: 255 seconds]
2012-09-15T05:05:04 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-15T05:10:11 < dongs> http://www.pjrc.com/teensy/beta/pcb_compare.jpg haha
2012-09-15T05:10:14 < dongs> iteadstudio strikes again
2012-09-15T05:10:42 < HTT-Bird> registration fail much?
2012-09-15T05:10:56 < dongs> ya looks like they havenet improved much
2012-09-15T05:11:06 < dongs> I had a similar pic where holes in a row were all drilled like a drunk chinese kid hand-drilled them
2012-09-15T05:11:13 < dongs> from like a year+ ago
2012-09-15T05:11:21 < HTT-Bird> looks like oshpark is the way to roll then, that is, if I don't roll the AC 66each offer that is
2012-09-15T05:11:30 < HTT-Bird> *shakes head* bah at grammar
2012-09-15T05:11:35 < HTT-Bird> that came out mangled, no?
2012-09-15T05:11:47 < dongs> yeah i have no idea what the fuck you said.
2012-09-15T05:12:16 < HTT-Bird> I mean, if I don't go with the Advanced Circuits 33each or 66each offers, I'll be going with oshpark
2012-09-15T05:12:17 < HTT-Bird> :)
2012-09-15T05:13:23 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Read error: No route to host]
2012-09-15T05:14:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T05:14:48 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-15T05:25:37 -!- dekar_ [~dekar@212.255.20.51] has quit [Ping timeout: 240 seconds]
2012-09-15T05:26:18 -!- dekar_ [~dekar@212.255.17.115] has joined ##stm32
2012-09-15T05:26:21 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-15T05:42:00 < HTT-Bird> If you want your microcontroller to monitor a slow-rising auxiliary supply voltage that feeds a few peripherals, is the pulse-stretching effect of a garden-variety reset-supervisor IC a good or bad thing?
2012-09-15T06:10:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-15T06:10:46 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-15T06:13:12 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-15T06:14:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T06:17:08 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Ping timeout: 248 seconds]
2012-09-15T06:29:43 -!- bgamari_ [~bgamari@pool-108-8-238-44.spfdma.east.verizon.net] has joined ##stm32
2012-09-15T06:30:01 -!- bgamari [~bgamari@pool-108-8-229-143.spfdma.east.verizon.net] has quit [Ping timeout: 268 seconds]
2012-09-15T06:48:21 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-15T06:53:01 < dongs> http://www.kickstarter.com/projects/digistump/digispark-the-tiny-arduino-enabled-usb-dev-board haha
2012-09-15T06:53:07 < dongs> those clowns got 300k and 25k devices to make
2012-09-15T06:56:22  * HTT-Bird can't figure out if a PoR/LVD IC will work for his application or if he needs something that's LVD-only *sigh*
2012-09-15T06:58:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T07:01:42 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-15T07:02:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T07:02:18 < dongs> HTT-Bird: what are you triyng to do
2012-09-15T07:02:50 < HTT-Bird> dongs: provide my uC with information regarding whether a sometimes-rather-slow-rising auxiliary supply is usable or not
2012-09-15T07:03:25 < dongs> i know that might be a silly suggestion but why not just have a voltage divider that puts shit on logic high when the supply is at usable levels
2012-09-15T07:03:28 < dongs> and just detect that.
2012-09-15T07:03:46 < dongs> well, assuming its above 3V
2012-09-15T07:03:49 < HTT-Bird> dongs: might be able to do that, but that might be a Delta-Icc problem in addition to voltage divider Iq
2012-09-15T07:04:42 < dongs> what does 'rather slow rising" have to do with it? if you use a power on reset thingy with some delay (like most usually are), it should work.
2012-09-15T07:05:06 < dongs> like you dont have to hook it up to reset/ just abuse its output as gpio/exti/whatever
2012-09-15T07:06:26 < HTT-Bird> dongs: yeah.  it'll be hooked up to a GPIO line.  and I was wondering if the "rather slow rising" factored in to whether the delay was needed/harmless/harmful
2012-09-15T07:07:02 < dongs> HTT-Bird: most have some sorta deadband / debounce type shit. some even have it configurable.
2012-09-15T07:07:20 < dongs> but as long as voltage reaches its threshold and stays there, it will be activated
2012-09-15T07:07:23 < dongs> after programmed delay.
2012-09-15T07:07:33 < dongs> well, or after whatever delay teh chip is configured for
2012-09-15T07:07:48 < dongs> some y ou jsut buy with like 'reset will go after 100ms' type shit.
2012-09-15T07:07:54 < dongs> others you can have a external resistor to set up the delay
2012-09-15T07:07:59 < HTT-Bird> yeah, either way the part'd have built-in hysteresis.  but yeah.  the delay from this rail coming up to PGOOD being allowed high should be OK here I think
2012-09-15T07:09:36 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-09-15T07:15:08 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T07:19:27 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-15T07:20:35 < dongs> http://i.imgur.com/P53VK.jpg saw this the other day
2012-09-15T07:21:21 < HTT-Bird> lol@engrish
2012-09-15T07:38:32 -!- R0b0t1` [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-15T07:39:34 -!- R0b0t1` is now known as iR0b0t1
2012-09-15T07:42:49 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-15T07:50:20 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-15T07:54:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Ping timeout: 240 seconds]
2012-09-15T08:02:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T08:52:35 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-15T09:18:46 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-15T09:28:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 268 seconds]
2012-09-15T09:45:13 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-15T09:52:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-15T09:58:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T10:34:00 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-15T11:07:00 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-09-15T11:07:00 -!- ServerMode/##stm32 [+o ChanServ] by sendak.freenode.net
2012-09-15T11:07:55 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-15T11:07:59 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-15T11:18:25 < dongs> sup trolls
2012-09-15T11:33:26 -!- ChanServ [ChanServ@services.] has quit [shutting down]
2012-09-15T11:34:01 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-09-15T11:34:01 -!- ServerMode/##stm32 [+o ChanServ] by sendak.freenode.net
2012-09-15T11:50:17 -!- dekar_ [~dekar@212.255.17.115] has quit [Quit: This computer has gone to sleep]
2012-09-15T11:50:54 -!- dekar [~dekar@212.255.17.115] has joined ##stm32
2012-09-15T11:50:56 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-15T12:14:40 -!- claude is now known as Claude
2012-09-15T12:17:23 < ABLomas> ...
2012-09-15T12:28:37 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Quit: Quitte]
2012-09-15T12:31:17 -!- barthess [~barthess@178.154.41.86] has joined ##stm32
2012-09-15T12:32:46 < dongs> no chats, sure sign that arduino is taking over
2012-09-15T12:37:54 -!- grummund_ [~user@unaffiliated/grummund] has quit [Ping timeout: 264 seconds]
2012-09-15T13:24:35 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T13:55:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-15T14:03:20 < dongs> haha fuckers
2012-09-15T14:03:28 < dongs> typical/max current consumption table of F3
2012-09-15T14:03:29 < dongs> is like
2012-09-15T14:03:32 < dongs> 1000 TBD's
2012-09-15T14:03:38 < dongs> any idea how much ti draws?
2012-09-15T14:03:42 < dongs> someone said 200mA, is that for real
2012-09-15T14:09:19 -!- dekar [~dekar@212.255.17.115] has quit [Quit: This computer has gone to sleep]
2012-09-15T14:09:45 -!- dekar [~dekar@212.255.17.115] has joined ##stm32
2012-09-15T14:09:46 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-15T14:10:12 < jpa-> dongs: nah, the 200mA was for something totally else :)
2012-09-15T14:10:35 < dongs> haha ok.
2012-09-15T14:10:54 < dongs> is at leats some approximate shit documented somewehre?
2012-09-15T14:10:56 < dongs> datasheet seems worhtless.
2012-09-15T14:11:40 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-15T14:24:01 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-15T15:05:06 < dongs> https://github.com/videlalvaro/gifsockets
2012-09-15T15:05:08 < dongs> oooooooookay.
2012-09-15T15:05:13 < dongs> Laurenceb_: were you involved in this
2012-09-15T15:07:45 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds]
2012-09-15T15:21:55 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-15T15:27:15 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 252 seconds]
2012-09-15T15:49:18 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-15T16:16:17 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-15T16:33:04 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-09-15T16:34:50 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T16:47:59 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-15T16:48:38 < Claude> gifsockets, pah shamelessly copied from the chinese : http://www.gg-dl.com/%E6%92%92%E5%A8%87-ebay.gif
2012-09-15T17:00:20 -!- daku is now known as DaKu
2012-09-15T17:02:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T17:28:52 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-15T17:28:52 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-15T17:28:52 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-15T17:30:10 < dongs> "People always ask me why my keyboard is so loud. Don't they know that C++ is strongly typed? "
2012-09-15T17:34:16 < zyp> :)
2012-09-15T17:43:07 < Bird|lappy> lol xD
2012-09-15T18:04:26 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-15T18:04:59 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T18:05:08 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-15T18:05:31 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has joined ##stm32
2012-09-15T18:49:34 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-15T18:49:34 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-15T18:49:34 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-15T18:49:34 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-15T19:06:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T19:11:48 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T19:32:27 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T19:48:25 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-15T19:51:30 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-15T19:58:34 -!- tavish [~tavish@120.56.143.88] has joined ##stm32
2012-09-15T19:58:42 -!- tavish [~tavish@120.56.143.88] has quit [Changing host]
2012-09-15T19:58:42 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-15T20:01:44 -!- DaKu is now known as daku
2012-09-15T20:06:17 -!- daku is now known as DaKu
2012-09-15T20:11:33 -!- barthess [~barthess@178.154.41.86] has quit [Quit: Leaving.]
2012-09-15T20:11:59 -!- zippe [~Adium@173.11.99.161] has left ##stm32 []
2012-09-15T20:13:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-15T20:29:11 < emeb> *crickets*
2012-09-15T20:33:41 < Bird|lappy> emeb, do you prefer the 10pin or the 20pin version of the ARM JTAG/SWD connector?
2012-09-15T20:34:09 < jpa-> :F
2012-09-15T20:34:11 < jpa-> 3 pins :)
2012-09-15T20:34:22 < Bird|lappy> heh, SWDIO, SWDCLK, GND?
2012-09-15T20:40:06 < emeb> Bird|lappy: I use SWD, either 4 or 6 pin version.
2012-09-15T20:40:32 < Bird|lappy> link me the pinouts for those?
2012-09-15T20:40:51 < emeb> errr.... look for the schematics of the Discovery boards on the ST site.
2012-09-15T20:41:23 < emeb> Seems like most folks here tend to use the 10-pin mini JTAG though.
2012-09-15T20:42:06 < emeb> Here's my schematic w/ 4-pin SWD: http://ebrombaugh.studionebula.com/synth/stm32f4_codec/stm32f4_codec_pg1.pdf
2012-09-15T20:43:32 < Bird|lappy> emeb, the Discovery board has the 4 pin connector on it
2012-09-15T20:43:36 < Bird|lappy> as to the 6 pin?
2012-09-15T20:44:07 < emeb> Bird|lappy: Discovery VL has 4-pin. F4 and F0 discovery (STLINK - V2) use 6-pin.
2012-09-15T20:44:55 < emeb> the extra 2 pins are optional - reset and SWO (trace / serial debug) aren't strictly required.
2012-09-15T20:45:07 < Bird|lappy> ah.
2012-09-15T20:46:15 < emeb> old-style 20-pin JTAG is so enormous. Better off using external adapter to SWD or mini JTAG.
2012-09-15T20:46:48 < Bird|lappy> well, I'm not using SWO, so I plan to bring a UART Tx/Rx out to the connector instead for a full test/debug console
2012-09-15T20:47:05 < emeb> that's what I do.
2012-09-15T20:47:33 < emeb> 4-pin 0.1" header is easy to use
2012-09-15T20:47:44 < Bird|lappy> I'll probably wind up using the 10pin mini-JTAG pinout for SWD with a 2.54mm pitch header instead of the 1.27mm pitch "official" version
2012-09-15T20:48:01 < emeb> whatevs
2012-09-15T20:48:12 < Bird|lappy> as the BusBlaster doesn't come with cables, so I plan to use some F-F jumpers I have laying around for hooking it up :)
2012-09-15T20:48:24 < emeb> the 1.27mm pitch aren't hard to find and are more standard.
2012-09-15T20:48:46 < emeb> direct connect to most newer debug pods
2012-09-15T20:48:52 -!- barthess [~barthess@178.154.41.86] has joined ##stm32
2012-09-15T20:49:13 < emeb> cables are cheap - ~$3 from uCPro
2012-09-15T20:49:20 < Bird|lappy> oh, linky?
2012-09-15T20:50:16 < emeb> momentado...
2012-09-15T20:50:52 < zyp> IMO the 1.27mm pitch connector is the most reasonable choice given that it's standardized by ARM
2012-09-15T20:51:57 < emeb> http://microcontrollershop.com/product_info.php?products_id=4517
2012-09-15T20:53:26 < emeb> I'll probably go that route on any future boards I do.
2012-09-15T20:57:06 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-15T20:57:06 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-247-126.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-15T20:57:12 < zyp> not using standardized pinouts means either making customized cables or connect single wires one by one
2012-09-15T20:58:13 < BrainDamage1> use wire wrap posts for jtag
2012-09-15T20:58:28 <+Steffanx> pogo pins!
2012-09-15T20:58:30 < zyp> and I can assure you that the latter gets annoying after connecting and disconnecting a few times
2012-09-15T20:59:06 < Bird|lappy> do you need to decouple the VTref line on a debug connector?
2012-09-15T20:59:34 < zyp> VTref? as in vcc out?
2012-09-15T20:59:58 < zyp> I've never done it
2012-09-15T21:01:43 < emeb> meither
2012-09-15T21:03:39 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-15T21:03:41 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-15T21:03:41 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-15T21:04:09 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-15T21:04:15 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-09-15T21:04:28 < TitanMKD> hi
2012-09-15T21:04:34 <+Steffanx> lo
2012-09-15T21:04:42 < jpa-> z
2012-09-15T21:05:08 < Thorn> where do you get those 1.27mm connectors anyway? last time I checked only digikey had them
2012-09-15T21:05:21 < zyp> I got mine from digikey
2012-09-15T21:05:22 <+Steffanx> So the answer is : digikey :P
2012-09-15T21:05:28 < Laurenceb_> so
2012-09-15T21:05:34 < Laurenceb_> sup
2012-09-15T21:05:35 < zyp> but I'm sure any decent distributor got 1.27mm headers
2012-09-15T21:05:44 <+Steffanx> Tha sky bro, Laurenceb_
2012-09-15T21:05:44 < jpa-> Laurenceb_: dongs
2012-09-15T21:05:59 < Thorn> farnell didn't
2012-09-15T21:08:56 < zyp> http://no.mouser.com/ProductDetail/Harwin/M50-3600542/?qs=sGAEpiMZZMtsLRyDR9nM114VzIC6WROoxRF84lzOi%2fQ%3d <- unshrouded at mouser here
2012-09-15T21:10:52 < emeb> http://www.digikey.no/product-detail/en/20021521-00010T1LF/609-4054-ND/2414951
2012-09-15T21:11:00 <+Steffanx> Whats wrong with this Thorn ? http://nl.farnell.com/fci/20021121-00006c4lf/header-vert-1-27mm-smt-6way/dp/1865277
2012-09-15T21:11:56  * Laurenceb_ has been fixing showers
2012-09-15T21:11:58 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-15T21:12:07 < emeb> did you use your computer?
2012-09-15T21:12:07 < zyp> Steffanx, it's 2x3, not 2x5 :p
2012-09-15T21:12:09 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-15T21:12:10 < Laurenceb_> knee deep in tampons
2012-09-15T21:12:11 -!- Claude is now known as claude
2012-09-15T21:12:11 < Laurenceb_> no
2012-09-15T21:12:24 < zyp> but they have 2x5 too
2012-09-15T21:12:27 < emeb> was it the pan, drain or seal?
2012-09-15T21:12:33 < Laurenceb_> tiles
2012-09-15T21:12:39 <+Steffanx> http://nl.farnell.com/fci/20021121-00010c4lf/header-vert-1-27mm-smt-10way/dp/1865279 yeah
2012-09-15T21:12:44 < emeb> oh yay - none of the above.
2012-09-15T21:12:50 < Laurenceb_> retiling it... slowly
2012-09-15T21:12:56 < emeb> urgh
2012-09-15T21:13:00 < Laurenceb_> have to wait for adhesive to dry
2012-09-15T21:13:03 < zyp> but I recommend shrouded ones
2012-09-15T21:13:35 < zyp> the pins are easy to bend so you don't want to risk that
2012-09-15T21:13:43 < Thorn> ˆ1,22 !!
2012-09-15T21:13:45 < emeb> and you won't hook 'em up backwards
2012-09-15T21:14:06 < zyp> emeb, yeah, that's another advantage
2012-09-15T21:14:07 <+Steffanx> Farnell prices, Thorn
2012-09-15T21:14:16 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-15T21:14:26 <+Steffanx> http://nl.farnell.com/fci/20021221-00010c4lf/header-vert-1-27mm-smt-10way/dp/1865302 like that zyp ?
2012-09-15T21:14:28 <+Steffanx> 1.66 :P
2012-09-15T21:14:33 <+Steffanx> *1.63
2012-09-15T21:14:39 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-15T21:14:39 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-15T21:15:04 <+Steffanx> Or with one gap on one side?
2012-09-15T21:15:16 <+Steffanx> Those are even more expensive
2012-09-15T21:16:19 < zyp> hmm, I think the one you linked is the exact ones I'm using, hang on
2012-09-15T21:17:10 < zyp> http://www.digikey.com/product-detail/en/20021521-00010T1LF/609-4054-ND/2414951 <- here's the one at digikey
2012-09-15T21:17:15 < zyp> FCI minitek 127 series
2012-09-15T21:19:07 < zyp> no, not same
2012-09-15T21:19:30 <+Steffanx> no, not really
2012-09-15T21:19:39 < zyp> http://no.mouser.com/Search/ProductDetail.aspx?qs=oJs1R%252bxK6kTFutD5Qb%252bmaQ== <- mouser have it though
2012-09-15T21:27:31 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-15T21:43:29 < Bird|lappy> on the Pixel Memory LCDs from Sharp, can one simply tie DISP high, or does it need to be actively managed by the processor?
2012-09-15T21:45:08 < zyp> if you are asking us to interpret the datasheet for you, you can at least give an url to it
2012-09-15T21:45:47 < Bird|lappy> http://www.sharpmemorylcd.com/resources/LS013B4DN02_Pre_Spec.pdf
2012-09-15T21:54:09 < Laurenceb_> can anyone help me with f4 timers?
2012-09-15T21:54:11 < zyp> it doesn't say, but I would guess that it's fine
2012-09-15T21:54:14 < Laurenceb_> im reading rm0090
2012-09-15T21:54:20 < Laurenceb_> page 437
2012-09-15T21:56:01 < zyp> Bird|lappy, the timing diagram on page 9 suggests that the purpose of the disp signal is to have the display show a white screen instead of garbage before it's initialized
2012-09-15T21:58:55 < zyp> you usually don't want to show your customers random garbage while you're booting your system to the point where it's ready to initialize the display, so it's better to keep the display blank until that time
2012-09-15T21:59:22 -!- DaKu is now known as daku
2012-09-15T21:59:39 < zyp> this is just guesswork on my side though, so you'll have to test it for yourself to see if it'll work
2012-09-15T22:00:43 < zyp> Laurenceb_, table 62?
2012-09-15T22:00:50 < Laurenceb_> yeah
2012-09-15T22:00:57 < Laurenceb_> so tim9 can slave off tim10
2012-09-15T22:01:00 < Laurenceb_> or tim11
2012-09-15T22:01:07 < Laurenceb_> yet look at p455
2012-09-15T22:01:13 < Laurenceb_> there no mms bits
2012-09-15T22:01:19 < Laurenceb_> *there are
2012-09-15T22:02:27 < Laurenceb_> they missed a register?
2012-09-15T22:02:51 < zyp> looks like there is no CR2
2012-09-15T22:03:16 < zyp> which likely implies that TRGO have a fixed function
2012-09-15T22:03:31 < Laurenceb_> or they missed it
2012-09-15T22:03:54 < Laurenceb_> they missed the slave mode description for the earlier timers
2012-09-15T22:03:59 < Laurenceb_> thats on the ST forum
2012-09-15T22:04:16 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 268 seconds]
2012-09-15T22:04:36 < Laurenceb_> guess ill have to try it and see
2012-09-15T22:04:39 < Laurenceb_> code compiles ok
2012-09-15T22:05:06 < Laurenceb_> but the f4xx header isnt that cleaver
2012-09-15T22:05:13 < Laurenceb_> *clever
2012-09-15T22:05:22 < zyp> of course not
2012-09-15T22:05:44 < Laurenceb_> just typedefs
2012-09-15T22:06:02 < zyp> the timers follow a common layout, so you just define the full featured one and reuse the definition for the reduced function ones
2012-09-15T22:06:19 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-15T22:09:46 < zyp> compare figures 153/154
2012-09-15T22:09:49 < zyp> page 415
2012-09-15T22:10:10 < zyp> TIM9/12 doesn't have TRGO, only TRGI
2012-09-15T22:10:33 < zyp> while the others doesn't have TRGI, only TRGO
2012-09-15T22:10:40 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-09-15T22:10:41 < zyp> hmm
2012-09-15T22:11:00 < zyp> the weird part is that CR2/MMS is described for TIM9/12 and not the others
2012-09-15T22:11:09 < zyp> and that only has anything to do with TRGO
2012-09-15T22:11:56 < zyp> so yeah, it might be a mistake in the documentation, they might have removed the CR2 description from the wrong register
2012-09-15T22:12:04 < zyp> the wrong timers*
2012-09-15T22:14:30 < zyp> http://paste.jvnv.net/view/2Xixg <- also, I got F1 to enumerate nicely today
2012-09-15T22:15:14 < Laurenceb_> cool
2012-09-15T22:15:27 < Laurenceb_> hmm but MSM bit needs to be used on masters
2012-09-15T22:15:40 < zyp> so now I have both F1 and F4 support in my stack
2012-09-15T22:15:47 < Laurenceb_> SMCR is "Not avaliable"
2012-09-15T22:16:09 < Laurenceb_> hopefully they got that wrong as they thought its NA due to no slave mode on the timer
2012-09-15T22:16:27 < Laurenceb_> according the the forum the timer documentation is very wrong in places
2012-09-15T22:16:53 < Laurenceb_> table60 is wrong for sure
2012-09-15T22:16:54 <+Steffanx> Who knows ..
2012-09-15T22:17:09 < Laurenceb_> going to have to try this
2012-09-15T22:20:10 <+Steffanx> Don't blow your leds
2012-09-15T22:20:57 < Laurenceb_> heh
2012-09-15T22:27:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 240 seconds]
2012-09-15T22:32:22 < Laurenceb_> ah look at figure 154
2012-09-15T22:32:40 < Laurenceb_> theres an internal output from CC1
2012-09-15T22:34:24 < Laurenceb_> wtf
2012-09-15T22:34:31 < Laurenceb_> why cant i upload to github
2012-09-15T22:34:35 < Laurenceb_> error 430
2012-09-15T22:35:28 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-15T22:35:33 <+Steffanx> They banned you
2012-09-15T22:38:41 < Laurenceb_> some kind of ssh thing....
2012-09-15T22:39:13 < Laurenceb_>  The requested URL returned error: 403
2012-09-15T22:40:50 -!- diydsp [41df9b1e@gateway/web/freenode/ip.65.223.155.30] has quit [Ping timeout: 245 seconds]
2012-09-15T22:55:06 < Laurenceb_> fixed using ssh
2012-09-15T22:58:04 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-15T23:12:36 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-09-15T23:14:43 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-09-15T23:22:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-15T23:23:07 < emeb> Hmmmm
2012-09-15T23:23:25 < emeb> trying to use the DSP FFT routines on my F4 project
2012-09-15T23:23:37 < emeb> if I call one FFT it works
2012-09-15T23:23:51 < emeb> if I call two then I get a hard fault
2012-09-15T23:24:12 -!- barthess [~barthess@178.154.41.86] has quit [Quit: Leaving.]
2012-09-15T23:25:40 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-15T23:28:12 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Timer.c
2012-09-15T23:28:16 < Laurenceb_> timer code ^
2012-09-15T23:29:19 < Laurenceb_> but im too tired to try this stuff now
2012-09-15T23:29:26 < Laurenceb_> same shit works on F1
2012-09-15T23:33:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-09-15T23:42:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-15T23:44:53 < Laurenceb_> http://forum.chibios.org/phpbb/viewtopic.php?f=8&t=157#p999
2012-09-15T23:45:00 < Laurenceb_> seems like _slight_ overkill
2012-09-15T23:54:58 <+Steffanx> Laurenceb_ and his hard to read code
--- Day changed Sun Sep 16 2012
2012-09-16T00:13:50 -!- barthess [~barthess@178.154.19.75] has joined ##stm32
2012-09-16T00:19:56 < Laurenceb_> http://content.animalnewyork.com/wp-content/uploads/2010/08/pig_blood_koran_02.jpg
2012-09-16T00:20:05 < Laurenceb_> is all i have to say
2012-09-16T00:29:37 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-16T00:29:40 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-16T00:30:03 -!- zippe1 [~Adium@173.11.99.161] has joined ##stm32
2012-09-16T00:33:19 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 272 seconds]
2012-09-16T00:38:26 < emeb> bizarre - FFT instance structure is corrupt on 2nd call.
2012-09-16T00:39:23 < Laurenceb_>  it converts in place?
2012-09-16T00:40:08 < emeb> kinda/sorta. FFT routine requires 3 inputs - an instance structure that defines the parameters, as well as input & output arrays.
2012-09-16T00:40:26 < emeb> (parameters = length, fwd/rev, etc)
2012-09-16T00:40:40 < zyp> so it changes the instance?
2012-09-16T00:41:05 < emeb> It shouldn't touch the structure, but for some reason the length is coming up 0 on the second call.
2012-09-16T00:41:15 < emeb> something's clobbering it.
2012-09-16T00:42:08  * emeb is happy with ddd + arm-none-eabi-gdb + BMPM
2012-09-16T00:44:01 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-16T00:45:33 < Laurenceb_> i tried to get nemiver running
2012-09-16T00:45:41 < Laurenceb_> it failed miserably
2012-09-16T00:45:45 < Laurenceb_> dd works for me
2012-09-16T00:45:47 < Laurenceb_> *ddd
2012-09-16T00:47:59 < Laurenceb_> http://farm2.static.flickr.com/1010/831709312_4cec2608d0_o.jpg
2012-09-16T00:48:01 < Laurenceb_> lolling
2012-09-16T00:48:10 < zyp> I like gdb's scriptability, the possibility to extract information directly from memory and format it in a nice way is handy
2012-09-16T00:48:38 < zyp> the logging stuff I wrote is proving to be pretty valuable
2012-09-16T00:49:41 < Laurenceb_> yeah nemiver will probably work
2012-09-16T00:49:47 < Laurenceb_> but gdb works for me :D
2012-09-16T00:50:50 < zyp> people can say what they want about how much more useful it is to have a debugger than printf-style-debugging, but you can't exactly step through routines handling realtime events from peripherals like usb :p
2012-09-16T00:51:00 < karlp> zyp, indeed.
2012-09-16T00:51:03 < emeb> word
2012-09-16T00:51:07 < karlp> was doing that yesterday :|
2012-09-16T00:51:13 < karlp> need to get swo going I guess :)
2012-09-16T00:51:26 < zyp> karlp, have you seen my logging stuff?
2012-09-16T00:51:41 < karlp> Laurenceb_: however, instead of ddd, have you tried gdbtui?
2012-09-16T00:51:45 < karlp> zyp: yeah, it looked neat :)
2012-09-16T00:52:17 < karlp> I have a serial port on the device though, I'm just goign to add a debug mode where instead of modbus, it just behaves as regualr serial port.
2012-09-16T00:52:26 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-101.pptp.stw-bonn.de] has joined ##stm32
2012-09-16T00:52:42 < gnomad> I generally gdb in emacs, but when that is broken, I use gdbtui.
2012-09-16T00:52:53 < Laurenceb_> nope
2012-09-16T00:53:01 < zyp> karlp, the problem with that is just that printing will be pretty slow
2012-09-16T00:53:10 < Laurenceb_> heh emacs
2012-09-16T00:53:25 < Laurenceb_> rms would be proud
2012-09-16T00:53:25 < gnomad> ddd and the other graphical interface to gdb are just ****waaaaayyyyyyy******* to slow.
2012-09-16T00:53:30 < karlp> yeah, but fast enough for what I'm doing.
2012-09-16T00:53:40 < zyp> then it's fine :)
2012-09-16T00:53:46 < gnomad> Emacs has been my IDE since 1988.
2012-09-16T00:54:03 < gnomad> you can pry it from my cold, dead, fingers.
2012-09-16T00:54:33 < zyp> come on, let's not turn this into a discussion about preferences
2012-09-16T00:54:38 < Laurenceb_> heh
2012-09-16T00:54:43 < Laurenceb_> in b4 flamewar
2012-09-16T00:54:50 < zyp> come on, let's not turn this into a flamewar
2012-09-16T00:56:47 <+Steffanx> zyp wants a flamewar?
2012-09-16T00:56:48 <+izua_> you can use the flamewar on his cold, dead, body though
2012-09-16T00:58:06 < Laurenceb_> f you make printf use dma it can work for realtime peripherals
2012-09-16T00:59:11 < gnomad> What I do for high-speed debugging is spit out a format string and the raw data.
2012-09-16T00:59:22 < gnomad> my PC does the data formatting.
2012-09-16T00:59:24 < zyp> gnomad, that's what I do
2012-09-16T00:59:40 < gnomad> format strings are indexed, too, for speed.
2012-09-16T00:59:53 < zyp> I throw a string pointer and some arguments into a ringbuffer, then extract the buffer with a gdb script and have it format the data
2012-09-16T01:00:22 -!- barthess [~barthess@178.154.19.75] has quit [Read error: Connection reset by peer]
2012-09-16T01:00:29 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-16T01:00:46 < zyp> http://cgit.jvnv.net/laks/tree/util/rblog.h <- logging class is here, gdb script is not in git yet
2012-09-16T01:02:41 < zyp> http://paste.jvnv.net/view/IRYgs <- looks like this, doesn't handle wraparound yet
2012-09-16T01:04:39 < Laurenceb_> so that talks to gdb?
2012-09-16T01:05:05 < Laurenceb_> oh theres a gdb module
2012-09-16T01:05:35 < zyp> yes, you load the script directly into gdb, it's a gdb plugin
2012-09-16T01:06:49 -!- barthess [~barthess@178.154.38.82] has joined ##stm32
2012-09-16T01:07:30  * emeb brain *pop*
2012-09-16T01:07:45 < emeb> yay - another scripting language to learn.
2012-09-16T01:08:04 < zyp> it's python
2012-09-16T01:08:45 < emeb> like I said.
2012-09-16T01:09:21 < zyp> it's your own fault you don't know it already, it's IMO the most versatile scripting language around
2012-09-16T01:09:24 < zyp> :p
2012-09-16T01:09:53 < emeb> no doubt. That's what was said about the last n-1 scripting languages I've come across. :)
2012-09-16T01:10:01 < zyp> that gdb is extendable by python is just a bonus to me
2012-09-16T01:10:15 < zyp> really? which ones?
2012-09-16T01:11:23 < emeb> Err. Lets see which ones I've used: Tcl/Tk, Scheme, Perl, Awk, just to name a few of the more common ones...
2012-09-16T01:12:18 < Laurenceb_> heh
2012-09-16T01:12:36 < karlp> awk}
2012-09-16T01:12:41 < karlp> you decided to write programs in awk?
2012-09-16T01:12:47 < karlp> you got what you deserved
2012-09-16T01:12:59  * emeb has been around a while.
2012-09-16T01:13:11 < karlp> sure, back when tcl seemed like an awesome idea
2012-09-16T01:13:20 < zyp> isn't awk only useful for relatively simple string processing? :p
2012-09-16T01:13:28 < emeb> You'd be surprised.
2012-09-16T01:13:28 < karlp> zyp: hoho
2012-09-16T01:13:33 < karlp> you can do alllll sorts of magic with perl
2012-09-16T01:13:35 < karlp> with awk
2012-09-16T01:13:38 < karlp> oops >(
2012-09-16T01:13:46 < emeb> it's a natural mistake.
2012-09-16T01:13:56 < Bird|lappy> lol
2012-09-16T01:13:59 < karlp> perl is awk with more keywords :)
2012-09-16T01:14:06 < emeb> Perl is kinda the bastard child of awk & C
2012-09-16T01:14:12 < Bird|lappy> Python's probably the cleanest of them save for Scheme if you're a Lispy type that is
2012-09-16T01:14:27 < zyp> kind of like how ruby is the bastard child of perl and python? :p
2012-09-16T01:14:38 < zyp> so awk would be ruby's grandfather?
2012-09-16T01:15:22 < emeb> Yeah - seems like Python is winning tho. Probably ought to figure it out.
2012-09-16T01:15:49 < karlp> python]s nice, but I'm starting to like lua a fair bit
2012-09-16T01:15:49 < Bird|lappy> Python also has the advantage that it's not "hog-tied" to an application domain
2012-09-16T01:15:51 < emeb> What's one more to stir up the pot.
2012-09-16T01:16:09 < karlp> wasn]t by choice, I had the same, +oh no, one more language to learn@
2012-09-16T01:16:12  * emeb used lua on his old Agenda PDA.
2012-09-16T01:16:13 < karlp> bleh, this keyboard is busted.
2012-09-16T01:16:35 < emeb> sure it's not yer fingers?
2012-09-16T01:16:47 < karlp> no, I]m in knoppix shuffling partitions,
2012-09-16T01:16:51 < karlp> and it has the wrong keymap
2012-09-16T01:17:15 < karlp> I'm trying to remember to touch type on a US keyboard again, and I've forgotten where a lot of the punctuation was.
2012-09-16T01:17:40  * emeb remembers being confused by DE kbds.
2012-09-16T01:17:51 < zyp> lua seems kind of nice, but I haven't yet had a reason to learn it
2012-09-16T01:18:43 < karlp> yeah, it]s the scripting language for the stock web admin console in openwrt
2012-09-16T01:18:54 < karlp> so it was write our own, or learn a bit of lua
2012-09-16T01:19:16 < karlp> now we've started writing some of the smaller things that had been small C apps in lua instead.
2012-09-16T01:19:34 < karlp> python is still too big unfortunately
2012-09-16T01:19:42 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-16T01:20:20 < zyp> it seems to fit most of the stuff I do
2012-09-16T01:21:06 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-16T01:21:06 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-16T01:34:20 < emeb> FFTs working now. Looks like I had the data arrays sized wrong.
2012-09-16T01:35:30 < Bird|lappy> is there a rule of thumb to estimate the parasitic trace capacitance on a FR-4 PCB?
2012-09-16T01:36:22 < zyp> yes
2012-09-16T01:36:37 < zyp> but it depends on the board thickness
2012-09-16T01:36:48 < Bird|lappy> .063" for me
2012-09-16T01:37:18 < zyp> my RF prof at uni had a sheet with some formulas to calculate stuff like that
2012-09-16T01:37:43  * Bird|lappy is trying to size some I2C pullups
2012-09-16T01:37:58 < zyp> he helped me do the calculations for making a 50 ohm transmission line on a board
2012-09-16T01:38:08 < Bird|lappy> and the maximum value for those depends on the Cload on the bus
2012-09-16T01:38:08 < zyp> but I can't remember the formula
2012-09-16T01:38:14 < emeb> http://www.technick.net/public/code/cp_dpage.php?aiocp_dp=util_pcb_imp_calculator
2012-09-16T01:38:44 < emeb> For I2C though it's probably not worth getting too complicated.
2012-09-16T01:38:52 < emeb> 4.7k should do it.
2012-09-16T01:39:16 < Bird|lappy> emeb, I am trying to stay away from resistors that small ;)
2012-09-16T01:39:23 < Bird|lappy> (power consumption)
2012-09-16T01:40:06 < emeb> but they only conduct during a bus transaction. not a lot of wasted power.
2012-09-16T01:40:11 < Bird|lappy> true.
2012-09-16T01:40:40 < zyp> wouldn't the capacitance of the devices on the bus be more significant than the traces itself?
2012-09-16T01:40:49 < zyp> unless your board is huge
2012-09-16T01:40:59 < Bird|lappy> zyp, the devices are a PCA9306 translator (Con = 12.5pF max)
2012-09-16T01:41:21 < Laurenceb_> something to note
2012-09-16T01:41:23 < Bird|lappy> and a TMP112 sensor (Cdevice not specified, but the device is capable of 3.4MHz operation so it has to be very small
2012-09-16T01:41:37 < Laurenceb_> according to stm32 datasheets you need 100r resistors in series with i2c
2012-09-16T01:41:46 < Laurenceb_> presumably to limit peak current
2012-09-16T01:41:51 < Laurenceb_> maybe to reduce emi too
2012-09-16T01:41:58 < Bird|lappy> ...and then a header going off board (currently unused, may hook up to a P82B96 in the future to accommodate remote sensors)
2012-09-16T01:42:08 < Bird|lappy> Laurenceb_, not using a STM32 here (some other ARM part :P)
2012-09-16T01:42:56 < zyp> Bird|lappy, also, with board thickness, I meant distance from ground plane
2012-09-16T01:43:08 < zyp> so if you have four-layer board, it's prepreg thickness :p
2012-09-16T01:43:18 < Bird|lappy> true :D
2012-09-16T01:43:36 < zyp> (assuming internal ground plane)
2012-09-16T01:43:49 < zyp> I used that with purpose on my board
2012-09-16T01:43:51 < Bird|lappy> probably will go double sided here though
2012-09-16T01:44:13 < Bird|lappy> as the fastest frequencies that go off-uC in my design are ~100kHz
2012-09-16T01:44:33 < zyp> the longer the distance is, the less capacitance you will have
2012-09-16T01:44:46 < zyp> so that's probably a good thing
2012-09-16T01:45:37 < zyp> in my case I wanted to reduce the width of the transmission line, so having a plane directly under it was beneficial
2012-09-16T01:45:38 -!- barthess [~barthess@178.154.38.82] has quit [Quit: Leaving.]
2012-09-16T01:56:25 < Laurenceb_> www.indiegogo.com/SuperAwesomeNY2012
2012-09-16T01:56:27 < Laurenceb_> lulwut
2012-09-16T01:56:58 <+izua_> that's super awesome
2012-09-16T02:00:35 < Bird|lappy> hrm, emeb, you think ~30k would be OK?
2012-09-16T02:00:42 < Bird|lappy> for my I2C pullups?
2012-09-16T02:00:48 < emeb> might work.
2012-09-16T02:00:58 < karlp> depends how fast you want to go.
2012-09-16T02:01:13 < emeb> http://26-26-54.hardwarebug.org/30
2012-09-16T02:01:54 < Laurenceb_> wut
2012-09-16T02:02:00 <+izua_> wat
2012-09-16T02:02:07 < emeb> old joke from the beagleboard community
2012-09-16T02:03:31 < zyp> base8
2012-09-16T02:04:16 < Laurenceb_> hmm
2012-09-16T02:04:25 < zyp> or what?
2012-09-16T02:04:53 < Laurenceb_> oh yeah
2012-09-16T02:04:53 < emeb> nah - just someone made a mistake and it "propagated"
2012-09-16T02:05:50 < zyp> well, then I don't got the joke :)
2012-09-16T02:06:00 < emeb> I guess you had to be there.
2012-09-16T02:06:20 < qyx_> where is the catch?
2012-09-16T02:06:25 < zyp> I'm just aware that there is a lot of jokes about shit that are true in other bases than base10
2012-09-16T02:06:25 < emeb> but the important think in this case was the line below - try it and find out.
2012-09-16T02:06:40 < emeb> like DEC25 = OCT31, etc
2012-09-16T02:08:58 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-16T02:09:56 < emeb> Looks like IFFT(FFT(x)) != x
2012-09-16T02:10:17 < Laurenceb_> rounding errors?
2012-09-16T02:10:23 < emeb> or, to be more precise, sounds like.
2012-09-16T02:10:47 < emeb> Laurenceb_: hard to say at this point. It's a floating pt. FFT, so that shouldn't be too bad.
2012-09-16T02:13:48 < emeb> Well, if I can resolve this, I may be able to do FFT-based audio processing in realtime.
2012-09-16T02:14:31 < emeb> Looks like 128point real FFT + IFFT + data move operations take < 120uS
2012-09-16T02:15:18 < emeb> of an F405 at full speed. My data frame period is > 600uS, so plenty of time.
2012-09-16T02:15:19 < Laurenceb_> oh F4?
2012-09-16T02:15:21 < Laurenceb_> ah
2012-09-16T02:16:03 < emeb> Need to use larger FFTs - 128 points/frame and need overlap.
2012-09-16T02:16:09 < Laurenceb_> fft -> ifft for audio processing in frequency space in the middle?
2012-09-16T02:16:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-16T02:16:15 < emeb> yep
2012-09-16T02:16:35 < Laurenceb_> 4096 should almost work
2012-09-16T02:16:44 < Laurenceb_> id try 2048
2012-09-16T02:17:22 < Laurenceb_> helium voice time :P
2012-09-16T02:17:28 < emeb> Thinking more like 512. 2K is largest supported by this lib.
2012-09-16T02:17:39 < emeb> Laurenceb_: shifters are one option.
2012-09-16T02:17:46 < emeb> manglers...
2012-09-16T02:17:49 < emeb> reverbs...
2012-09-16T02:17:54 < emeb> Phase vocoders...
2012-09-16T02:18:00 < Laurenceb_> yeah
2012-09-16T02:18:04 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-16T02:21:08 < emeb> OK, 512pts works time-wise. 450us
2012-09-16T02:21:42 < emeb> that'll give me 3/4 overlap which should be sufficient. Better freq resolution too.
2012-09-16T02:24:34 < emeb> Aha - fixed. Needed to enable the bit reversal.
2012-09-16T02:24:43 < emeb> Looks like this will work...
2012-09-16T02:38:42 < Laurenceb_> http://www.theregister.co.uk/2012/09/14/hp_whitman_smartphone_plans/
2012-09-16T02:38:58 < Laurenceb_> "There will be countries in the world where people will never own a tablet or a PC or a desktop. They will do everything on a smartphone."
2012-09-16T02:39:03 < Laurenceb_> whats she smoking
2012-09-16T02:40:15 < emeb> you disagree
2012-09-16T02:40:35 < Laurenceb_> how would you even type
2012-09-16T02:41:09 < emeb> virtual kbd on the touchscreen?
2012-09-16T02:41:17 < emeb> + thumbs
2012-09-16T02:42:32 < BrainDamage> because for stuff like a paper, you totally want to type it entirely on a smartphone
2012-09-16T02:42:38 < zyp> I've heard that a lot of people in japan claim they don't need any other computer than their phone
2012-09-16T02:42:53 < BrainDamage> yes, but not entire countries
2012-09-16T02:42:56 < emeb> but a lot of people will never need to write a paper.
2012-09-16T02:43:05 < zyp> emeb, exactly
2012-09-16T02:43:09 < emeb> they're not talking about professionals
2012-09-16T02:43:10 < BrainDamage> sure, I agree with that
2012-09-16T02:43:23 < BrainDamage> but not whole countries again, just some people
2012-09-16T02:43:23 < emeb> they're talking about the guy who drives the garbage truck.
2012-09-16T02:43:42 < BrainDamage> some people need a pc, some people can get away with a smartphone
2012-09-16T02:43:48 < BrainDamage> some won't even need the smartphone
2012-09-16T02:43:48 < emeb> right
2012-09-16T02:45:18 < zyp> I must admit that I kind of fail to see how people get by without owning a decent computer, but I recognize that a lot of people do :p
2012-09-16T02:45:40  * emeb remembers Stephenson's "Anathem" - all the tourists carrying "Jeejahs" - basically smartphones.
2012-09-16T02:47:01 < BrainDamage> mmm perhaps using bluetooth keyb + hdmi out on a tv, you might be able to do the task
2012-09-16T02:47:09 < BrainDamage> but I bet it'd be rather awkward
2012-09-16T02:47:15 < emeb> general purpose PC today provides a lot more power than most folks need/use in their daily lives.
2012-09-16T02:47:32 < emeb> Just realize we're our own kind of 1%. :)
2012-09-16T02:47:41 < BrainDamage> I totally agree, for instance, I replaced my parent's pc with an atom
2012-09-16T02:47:50 < BrainDamage> and they noticed very little difference
2012-09-16T02:48:34 < BrainDamage> other than now there's 0 noise, it uses a usb drive, and 0 fans
2012-09-16T02:48:34 < BrainDamage> no mobile parts
2012-09-16T02:48:34 < BrainDamage> other than the power switch
2012-09-16T02:48:59 < zyp> hey, I've always gotten by with rather economic computers, compared to gaming rigs :)
2012-09-16T02:49:10 < emeb> yep
2012-09-16T02:49:39 < zyp> the one I bought a couple of months ago is kind of unnecessary powerful :p
2012-09-16T02:49:47 < emeb> computers are tools. Not everyone needs a hammer or pipe wrench.
2012-09-16T02:50:56 < zyp> I don't know what you consider as a decent lifetime for a workstation, mine was four years old
2012-09-16T02:51:18 < emeb> Yeah - I get a new mobo about that often.
2012-09-16T02:51:38 < zyp> the one I had before that lasted five years
2012-09-16T02:51:44 < emeb> sometimes I need a new PS and/or case too, but that gets swapped out less often
2012-09-16T02:52:17 < zyp> so I'm actually only on my third desktop computer since I started buying my own
2012-09-16T02:52:34  * emeb has a garage full of old machines.
2012-09-16T02:52:51 < emeb> Need to start recycling them - they have no antique value.
2012-09-16T02:52:51 < zyp> so have I, but I didn't buy them new :p
2012-09-16T02:53:20 < BrainDamage> the psu has plenty nice parts, the ferrite chokes, the high voltage mosfets
2012-09-16T02:53:22 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-101.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-16T02:53:39 < emeb> meh - pulling parts is a last resort.
2012-09-16T02:54:13 < zyp> salvaging parts is something I only do when I need a part for prototyping some shit right now
2012-09-16T02:54:27 -!- zippe1 [~Adium@173.11.99.161] has left ##stm32 []
2012-09-16T02:54:30 < emeb> yup - can't afford to wait 3 days for mail order.
2012-09-16T02:54:33 < BrainDamage> remember one thing tough: if you plan to reuse say a pentium 4 as nas
2012-09-16T02:54:43 < emeb> lol
2012-09-16T02:54:55 < BrainDamage> the consumption it'll have will basically null any economic advantace
2012-09-16T02:55:02 < BrainDamage> advantage*
2012-09-16T02:55:09 < zyp> unless you're not paying for power
2012-09-16T02:55:19 < emeb> yep - be prepared for some noise too.
2012-09-16T02:55:28 < emeb> (or bring a huge heatsink)
2012-09-16T02:55:38 < zyp> neither is a problem here :)
2012-09-16T02:55:53 < BrainDamage> how do you get free power?
2012-09-16T02:55:58 < zyp> http://bin.jvnv.net/f/rxFpC.JPG <- I got a nice cool server room
2012-09-16T02:56:19 < BrainDamage> the rocks add a nice touch
2012-09-16T02:56:26 < BrainDamage> looks almost a landslide
2012-09-16T02:56:35 < emeb> OK, so you've got exposed bedrock in your basement?
2012-09-16T02:56:40 < zyp> yep
2012-09-16T02:56:53  * emeb wishes he had a basement.
2012-09-16T02:57:10 < BrainDamage> how do you get the free energy? and please don't tell me HHO :p
2012-09-16T02:57:31 < zyp> BrainDamage, simply by having somebody else pay
2012-09-16T02:57:48 < BrainDamage> ok, so you tap the neighbour's line, got it :p
2012-09-16T02:57:53 < zyp> I'm renting the basement, power for the basement is not metered separately.
2012-09-16T02:57:53  * emeb wishes he had someone else paying
2012-09-16T02:58:10 < BrainDamage> ah
2012-09-16T02:58:29 < BrainDamage> then I'd probably try all sort of dumb shit
2012-09-16T02:58:52 < BrainDamage> like bitcoin mining, and try to get some extra money from trading :p
2012-09-16T02:59:17 < emeb> sink a solid copper slug into that bedrock - heatsink!
2012-09-16T02:59:31 < zyp> I don't have hardware to make bitcoing mining worthwhile
2012-09-16T03:00:33 < zyp> I once considered picking up some cheap used GPUs for that purpose, but I scrapped the idea
2012-09-16T03:01:52 < zyp> on the other hand, I earned a bit on bitcoin trading last year
2012-09-16T03:02:08 < BrainDamage> the huge price variability?
2012-09-16T03:02:20 < zyp> bought some for fun when they was at $8 on the way up, and sold off at $30 just before they crashed
2012-09-16T03:02:40 < BrainDamage> it feels a bit like stock market speculation
2012-09-16T03:03:03 < zyp> it's sort of the same
2012-09-16T03:03:04 < BrainDamage> perhaps it'll never be a currency, but it's interesting as speculation tool
2012-09-16T03:03:57  * emeb reads the Bitcoin wikipedia article, sees the value vs time chart.
2012-09-16T03:03:58 < Laurenceb_> it seems flawed
2012-09-16T03:04:02 < Laurenceb_> as it cant scale?
2012-09-16T03:04:11 < Laurenceb_> theres a limited number of tokens
2012-09-16T03:04:13 < zyp> I also considered using it as an excuse to get a powerful FPGA board
2012-09-16T03:04:32 < Laurenceb_> reminds me of arduino and raspberry pi
2012-09-16T03:04:34 < karlp> laurence, no, you can @make@ more of them via mining.
2012-09-16T03:04:48 < Laurenceb_> in that they seemed to launch the pre 1.0 version
2012-09-16T03:04:51 < BrainDamage> karlp: there's a limit to that
2012-09-16T03:05:00 < zyp> Laurenceb_, so what? the point is to induce deflation
2012-09-16T03:05:03 < Laurenceb_> with a bit mor ework it could be implimented a bit better
2012-09-16T03:05:06 < Laurenceb_> hmm
2012-09-16T03:05:22  * Laurenceb_ zzz
2012-09-16T03:05:33 < zyp> as less gets generated, the ones that exist gets more valuable
2012-09-16T03:05:47 < zyp> or something like that
2012-09-16T03:05:48 < emeb> if demand goes up
2012-09-16T03:05:53 < BrainDamage> I guess the point is how much it can fractionate
2012-09-16T03:06:01 < zyp> yeah
2012-09-16T03:06:20 < BrainDamage> since the total of bitcoins ever generable is capped
2012-09-16T03:07:08 < emeb> time for some yardwork...
2012-09-16T03:07:27 < zyp> if demands goes up, the value increases, so the worth of all coins in existence goes up
2012-09-16T03:10:20 < zyp> the dangerous part about bitcoins is a fear-fueled crash of value
2012-09-16T03:11:42 < BrainDamage> or even intentional
2012-09-16T03:11:51 < zyp> if something triggers a drop in value causing everybody to want to get rid of them, which further drives the value down
2012-09-16T03:12:24 < BrainDamage> like several governments declare bitcoins outlawed because some idiotic reason, like used to send money to terrorists or whatever
2012-09-16T03:12:56 < BrainDamage> not having any official backer in the "real life" can be both an advantage and a burden
2012-09-16T03:13:49 < dongs> sup blogs
2012-09-16T03:23:59 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-09-16T03:28:15 -!- Netsplit over, joins: @ChanServ
2012-09-16T03:33:49 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds]
2012-09-16T03:53:27 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Ping timeout: 272 seconds]
2012-09-16T04:10:34 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 268 seconds]
2012-09-16T04:15:37 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-16T04:20:20 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-16T04:36:31 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-101.pptp.stw-bonn.de] has joined ##stm32
2012-09-16T04:50:04 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-16T04:56:54 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-16T05:02:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-16T05:25:17 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-16T05:25:18 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-16T05:34:07 -!- izua_ [~izua@unaffiliated/izua] has quit [Ping timeout: 240 seconds]
2012-09-16T05:44:04 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-101.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-16T05:50:27 < emeb_mac> this is probably old news: http://www.anandtech.com/show/6292/iphone-5-a6-not-a15-custom-core
2012-09-16T05:56:12 < karlp> well, anand was the only souce that said it was right}
2012-09-16T05:56:28 < karlp> and zippe pretty seriously said it definitely wasn't
2012-09-16T06:09:16 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-16T06:09:22 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-16T06:13:50 < Bird|lappy> so, what do you use for ESD/transient protection on external interfaces?
2012-09-16T06:14:14 < Bird|lappy> and also: do you need ESD protection devices on key switches (or keypads for that matter)?
2012-09-16T07:13:07 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-16T08:47:54 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 250 seconds]
2012-09-16T08:48:59 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-16T08:49:00 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-16T08:49:00 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-16T08:57:30 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-16T09:05:26 -!- claude is now known as Claude
2012-09-16T09:17:19 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-16T09:39:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-16T09:48:01 -!- Netsplit *.net <-> *.split quits: grummund, LeelooMinai, |sark, @ChanServ
2012-09-16T10:11:01 -!- Netsplit over, joins: LeelooMinai
2012-09-16T10:11:54 -!- Netsplit over, joins: grummund, |sark
2012-09-16T10:21:34 < Thorn> Intel promises a 5nm process in a few years. wtf
2012-09-16T10:24:51 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-16T10:25:08 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-16T10:30:30 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-16T10:30:46 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-16T10:33:19 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-16T10:33:27 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 252 seconds]
2012-09-16T10:34:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-16T10:37:15 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-16T10:37:52 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 265 seconds]
2012-09-16T10:39:54 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-16T10:40:11 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-16T10:49:57 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-16T10:50:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-16T10:58:31 -!- Claude is now known as claude
2012-09-16T11:07:19 -!- barthess [~barthess@178.154.67.199] has joined ##stm32
2012-09-16T11:18:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-16T11:22:31 < dongs> http://i3.kym-cdn.com/photos/images/newsfeed/000/246/770/000.jpg
2012-09-16T11:25:27 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-16T11:25:45 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-16T12:35:36 -!- claude is now known as Claude
2012-09-16T12:42:57 -!- Claude is now known as claude
2012-09-16T13:08:09 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-09-16T13:08:09 -!- ServerMode/##stm32 [+o ChanServ] by sendak.freenode.net
2012-09-16T13:09:00 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-16T13:48:00 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-16T14:04:13 < dongs> ok freertos is officially bloated garbage
2012-09-16T14:04:15 -!- barthess [~barthess@178.154.67.199] has quit [Quit: Leaving.]
2012-09-16T14:04:17 < dongs> not to mention unreadable code
2012-09-16T14:04:25 < dongs> with all teh vCrap xBullShit and 4-space tabs
2012-09-16T14:04:30 < dongs> shift-del'in it
2012-09-16T14:05:31 < zyp> :)
2012-09-16T14:05:51 < dongs> and why the fuck should I provide pvPortMalloc() and shit
2012-09-16T14:08:01 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-16T14:13:36 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Quit: leaving]
2012-09-16T14:14:08 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-16T14:14:39 < dongs> Laurenceb_: so far in order of gayness, ChibiOS >>> FreeRTOS >>>> all other shit >>> CoOS.
2012-09-16T14:14:51 < Laurenceb_> lol
2012-09-16T14:15:03 < Laurenceb_> found some trolling material for you
2012-09-16T14:15:20 < Laurenceb_> http://farm2.static.flickr.com/1010/831709312_4cec2608d0_o.jpg
2012-09-16T14:15:41 < dongs> haha
2012-09-16T14:21:20 -!- Tectu is now known as Tectu_
2012-09-16T14:22:14 -!- Tectu_ is now known as Tectu
2012-09-16T14:22:18 -!- Tectu is now known as Tectu_
2012-09-16T14:22:31 -!- Tectu_ is now known as Tectu
2012-09-16T14:26:13 -!- daku is now known as DaKu
2012-09-16T14:27:19 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-16T14:43:35 < dongs> took about 10 minutes of not reading docs to get freeRTOS to start a task blinking leds.
2012-09-16T14:44:49 < dongs> but yeah
2012-09-16T14:44:52 < dongs> whole vCrap isnt gonna fly
2012-09-16T14:50:37 < karlp> what's the motivation behind thee xperiment?
2012-09-16T14:51:27 < dongs> just to see how much it sucks
2012-09-16T14:51:34 < dongs> suckage confirmed, time to move along
2012-09-16T14:51:35 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-16T14:51:49 < dongs> coos is still my favorite for aids-free rtos that stays out of ht e way
2012-09-16T14:51:53 < dongs> and doesnt force idiot shit on end-user
2012-09-16T14:53:12 < dongs> and their spacing/indent style is ugh
2012-09-16T14:53:30 < dongs> ( ( ( ( prvFuckingGay ) ) ) ) ) / portSUCKS )
2012-09-16T14:57:06 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 264 seconds]
2012-09-16T15:01:09 < karlp> the complete antilaurenceb style
2012-09-16T15:01:59 < dongs> haha
2012-09-16T15:04:57 < zyp> :p
2012-09-16T15:05:12 -!- barthess [~barthess@178.154.67.199] has joined ##stm32
2012-09-16T15:13:25  * Laurenceb_ has been working on his timers
2012-09-16T15:13:26 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Timer.c
2012-09-16T15:13:33 < Laurenceb_> registerzzzz
2012-09-16T15:14:07 < Laurenceb_> does coos have a hal?
2012-09-16T15:16:18 < jpa-> yeah it was called penis or something
2012-09-16T15:16:27 < Laurenceb_> lol
2012-09-16T15:16:44 < Laurenceb_> http://www.coocox.org/CooCox_CoIDE.htm
2012-09-16T15:16:51 < Laurenceb_> System requirements:
2012-09-16T15:16:51 < Laurenceb_>     Windows XP SP3/Windows Vista/Windows 7
2012-09-16T15:16:53 < Laurenceb_> lolling
2012-09-16T15:21:22 < dongs> ?
2012-09-16T15:21:27 < dongs> < Laurenceb_> does coos have a hal?
2012-09-16T15:21:33 < dongs> yeah its called cox but obviosuly you dont have to use it
2012-09-16T15:21:37 < dongs> or even know it exists.
2012-09-16T15:23:42 < Laurenceb_> true
2012-09-16T15:27:24 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-16T15:30:37 -!- barthess [~barthess@178.154.67.199] has quit [Quit: Leaving.]
2012-09-16T15:55:50 < zyp> http://bin.jvnv.net/f/4HSHR.png <- I wonder what the fuck the host is doing, it seems to be pretty confused
2012-09-16T16:03:41 < Laurenceb_> whats that showing?
2012-09-16T16:04:40 < zyp> device enumerats fine, then host decides to reset the port, skip normal enumeration and do some weird shit instead
2012-09-16T16:05:55 < dongs> its tweeting at you
2012-09-16T16:05:59 < dongs> chirp chirp nyet
2012-09-16T16:09:18 < Laurenceb_> i meant what debugging software is that?
2012-09-16T16:09:48 < zyp> total phase data center
2012-09-16T16:30:02 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-114.pptp.stw-bonn.de] has joined ##stm32
2012-09-16T16:31:15 < Laurenceb_> wonder how long till f3 is in stock
2012-09-16T16:31:19  * Laurenceb_ wants
2012-09-16T16:32:56 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32
2012-09-16T16:43:46 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-16T17:26:59 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-16T17:45:50 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-16T17:55:31 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-16T18:32:58 -!- barthess [~barthess@178.154.67.199] has joined ##stm32
2012-09-16T18:37:42 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-16T18:48:42 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-16T18:53:21 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-16T18:53:27 < Laurenceb_> http://hipsterjew.com/2012/08/buy-all-your-indian-goods-at-hitler/  <-- attention odngs
2012-09-16T18:59:42 < zyp> old
2012-09-16T19:00:50 < Erlkoenig> dafuq is that site oO
2012-09-16T19:02:09 < Laurenceb_> not sure if its troll or 4 real
2012-09-16T19:02:32 < Erlkoenig> "Who Wants a HJ?" -- HJ = Hitlerjungend = Hitler's youth = The nazi's "youth organisation" to brainwash the children
2012-09-16T19:30:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-16T19:32:21 < Laurenceb_> anyone here used matlab?
2012-09-16T19:34:17 < Laurenceb_> can i init an array using a function that takes vertices as arguments without using a loop?
2012-09-16T19:36:38 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-16T19:45:46 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-16T19:45:46 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-16T19:45:46 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-16T19:45:49 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-16T19:53:59 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-16T20:15:42 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 264 seconds]
2012-09-16T20:17:10 -!- DaKu is now known as daku
2012-09-16T20:17:30 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-16T20:23:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-16T20:32:24 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-16T20:38:33 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 272 seconds]
2012-09-16T20:52:01 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has left ##stm32 []
2012-09-16T21:35:49 < Bird|lappy> does anyone know what -irlen value I should use when creating a SWD DAP in OpenOCD, or is that option ignored by 'swd newdap'?
2012-09-16T21:36:58 -!- olasd [~olasd@pdpc/supporter/active/olasd] has joined ##stm32
2012-09-16T21:38:30 < zyp> it's irrelevant for swd
2012-09-16T21:38:39 < Bird|lappy> ok, as I thought.
2012-09-16T21:49:51 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-16T21:50:49 -!- Oldboy_ [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-16T21:50:49 -!- Oldboy_ [myircshell@grid.sh3lls.net] has quit [Client Quit]
2012-09-16T22:02:44  * Laurenceb_ has finished his first chibios project
2012-09-16T22:02:50 < Laurenceb_> hopefully it will work...
2012-09-16T22:03:27 < Laurenceb_> final size is 30KB as opposed to 64KB for the previous "bare metal" code
2012-09-16T22:04:06 < zyp> I got my ACM echo shit to work on F1
2012-09-16T22:04:57 < zyp> heh, «bare metal»
2012-09-16T22:04:58 < Laurenceb_> i cant remember if it echoes in chibios...
2012-09-16T22:05:22 < zyp> no, I mean, I just made some shit that returns all received bytes
2012-09-16T22:05:38 < Laurenceb_> to be fair it had fatfs with long filenames and the st MSC code
2012-09-16T22:05:38 < zyp> http://paste.jvnv.net/view/aZmIa <- the very advanced ACM class driver looks like this
2012-09-16T22:05:47 < Laurenceb_> heh
2012-09-16T22:06:19 < Laurenceb_> wonder if you could use stripped down USB as mcu to mcu comms
2012-09-16T22:06:21 < Laurenceb_> using usb otg
2012-09-16T22:06:42 < zyp> of course you could
2012-09-16T22:06:59 < Laurenceb_> it seems fairly robust, and you have lots of hardware support for the low level stuff
2012-09-16T22:07:08 < zyp> ha
2012-09-16T22:07:17 < zyp> «fairly robust»
2012-09-16T22:07:34 < zyp> seems fairly fragile to me
2012-09-16T22:07:52 < Laurenceb_> ive done some electrically very bad stuff
2012-09-16T22:08:02 < Laurenceb_> like using the wrong resistors by mistake etc
2012-09-16T22:08:08 < Laurenceb_> and its still handled it ok
2012-09-16T22:08:19 < Laurenceb_> the connectors suck tho
2012-09-16T22:08:44 < zyp> if you are talking FS USB, SPI will probably beat it on every point
2012-09-16T22:08:56 < zyp> both speed, simplicity and robustness
2012-09-16T22:09:15 < Laurenceb_> needs more conductors
2012-09-16T22:09:24 < Laurenceb_> but yeah its simpler for sure
2012-09-16T22:09:31 < TitanMKD> zyp even RS422 beat USB FS ;)
2012-09-16T22:09:45 < TitanMKD> zyp and with very long cable
2012-09-16T22:09:55 < Laurenceb_> need external phy
2012-09-16T22:10:46 < zyp> on the other hand, when you are talking HS it starts getting more interesting
2012-09-16T22:10:58 < zyp> that's why HSIC exists
2012-09-16T22:11:45 < Laurenceb_> ah interesting
2012-09-16T22:15:25  * TitanMKD dream about a MCU 1GHz with USB 3.0 ;)
2012-09-16T22:17:39 < zyp> like fx3?
2012-09-16T22:18:09 < TitanMKD> yes but with a price for hobby ;)
2012-09-16T22:18:25 < TitanMKD> like 15/20usd per chip ;)
2012-09-16T22:19:00 < zyp> hmm, they claim fx3 starts at $25
2012-09-16T22:20:00 < zyp> http://www.digikey.no/product-detail/en/CYUSB3014-BZXI/428-3130-ND/2827561 <- cheapest at digikey is at $37
2012-09-16T22:20:03 < zyp> not that far off :p
2012-09-16T22:20:46 < TitanMKD> yes really true
2012-09-16T22:21:20 < TitanMKD> but to really use USB3.0 bandwidth will require at least 600MHz MCU with DMA on that peripherals ;)
2012-09-16T22:23:13 < zyp> oh, I don't know
2012-09-16T22:24:00 < zyp> I'm just glancing over the datasheet, but it claims that the GPIF supports up to 32 bits and up to 100 MHz operation
2012-09-16T22:24:30 < zyp> that's 3.2 Gb/s of data
2012-09-16T22:24:51 < TitanMKD> zyp yes ;) and the starter kit cost is just 2147USD ;)
2012-09-16T22:25:14 < TitanMKD> which one said USB3.0 will be cheap ;)
2012-09-16T22:25:25 < zyp> buy the chip and build your own board :p
2012-09-16T22:26:03 < TitanMKD> the chip cost is 235.76USD
2012-09-16T22:26:08 < TitanMKD> not 23.5
2012-09-16T22:26:23 < TitanMKD> and only BGA
2012-09-16T22:26:27 < zyp> huh?
2012-09-16T22:26:42 < zyp> http://www.digikey.no/product-detail/en/CYUSB3014-BZXI/428-3130-ND/2827561 <- $37 in single quantity here?
2012-09-16T22:26:44 < TitanMKD> check the link ttp://www.digikey.no/product-detail/en/CYUSB3014-BZXI/428-3130-ND/2827561
2012-09-16T22:27:01 < TitanMKD> here i see 1 for	235,76000
2012-09-16T22:27:12 < zyp> in what currency?
2012-09-16T22:27:20 < TitanMKD> oups ;)
2012-09-16T22:27:28 < TitanMKD> the link is NO
2012-09-16T22:27:33 < TitanMKD> i was thinking USD
2012-09-16T22:28:34 < TitanMKD> ok it is about 33euros
2012-09-16T22:29:29 < zyp> that should be pretty affordable if you actually have a use for usb3
2012-09-16T22:30:16 < TitanMKD> yes
2012-09-16T22:30:48 < TitanMKD> especially to transfer data in realtime for high end SDR for example ;)
2012-09-16T22:31:25 < zyp> I recently saw this project where it was used for that
2012-09-16T22:32:16 < TitanMKD> yes or for realtime logic analyzer 1GBPS ;)
2012-09-16T22:32:39 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-16T22:32:42 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-16T22:34:08 < zyp> izua, received the board yet?
2012-09-16T22:34:21 <+izua> hello
2012-09-16T22:34:31 <+izua> not yet
2012-09-16T22:46:21 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-16T22:48:59 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-09-16T22:56:09 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-09-16T22:56:42 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-16T22:59:06 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: z]
2012-09-16T22:59:45 <+izua> https://www.youtube.com/watch?feature=player_embedded&v=t3TAOYXT840
2012-09-16T23:01:56 -!- barthess [~barthess@178.154.67.199] has quit [Quit: Leaving.]
2012-09-16T23:12:22 < gnomad> methink an ARM with accelerated video would be more than sufficient for that...
2012-09-16T23:36:29 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-16T23:36:32 < Mazingaro> hi
2012-09-16T23:36:44 < Mazingaro> please I need help on systick + scheduler.
2012-09-16T23:37:01 < zyp> what do you need help with?
2012-09-16T23:37:05 < Mazingaro> I developed a small RTOS using SysTick as system tick
2012-09-16T23:37:25 < Mazingaro> I switch task context in SysTick_Handler
2012-09-16T23:37:40 < Mazingaro> but all I get is only one interrupt
2012-09-16T23:37:58 < Mazingaro> SysTick_Handler is executed only one time
2012-09-16T23:38:38 < Mazingaro> I'm using an olimex stm32-h107
2012-09-16T23:38:47 < Mazingaro> base on stm32f107
2012-09-16T23:39:45 < zyp> you're probably unintentionally disabling interrupts in your context switch
2012-09-16T23:39:52 < Mazingaro> ok
2012-09-16T23:40:28 < zyp> can you show me your code?
2012-09-16T23:40:32 < zyp> pastebin or something
2012-09-16T23:40:44 < Mazingaro> I read that cortex-m3 pushes from r0 to r3, r12, lr, pc and xpsr
2012-09-16T23:40:55 < zyp> yep
2012-09-16T23:41:00 < Mazingaro> before executing an exception
2012-09-16T23:41:26 < Mazingaro> shall I retrieve these values before exiting ISR?
2012-09-16T23:41:58 < zyp> no, they are popped automatically just like they are pushed
2012-09-16T23:42:20 < Mazingaro> but I did a naked ISR
2012-09-16T23:42:31 < Mazingaro> so my last instruction is bx lr
2012-09-16T23:42:50 < Mazingaro> where lr is loaded with the task address
2012-09-16T23:43:00 < zyp> http://cgit.jvnv.net/laks/tree/interrupt/fault.cpp <- you can have a look at my switch_context function here
2012-09-16T23:43:25 < zyp> I'm calling it from the svcall handler, but it would work just as well from the systick handler
2012-09-16T23:43:37 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 240 seconds]
2012-09-16T23:44:06 < Mazingaro> ok
2012-09-16T23:44:11 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- and the function for initializing new thread contexts are here
2012-09-16T23:44:33 < zyp> it might not be the best multithreading code around, but it works for me :)
2012-09-16T23:45:20 < Mazingaro> http://pastebin.com/pRggQ7iY
2012-09-16T23:45:26 < Mazingaro> this is the ISR
2012-09-16T23:46:13 < Mazingaro> http://pastebin.com/a2ipE4Ad
2012-09-16T23:46:23 < Mazingaro> these are SAVE and RESTORE
2012-09-16T23:47:04 < Mazingaro> http://pastebin.com/XehQz5F2
2012-09-16T23:47:18 < Mazingaro> this is the startScheduler executed only once
2012-09-16T23:48:35 < zyp> are you running separate privileged and unprivileged stacks?
2012-09-16T23:48:53 < Mazingaro> yes I will
2012-09-16T23:49:25 < Mazingaro> so the bx lr would change to unpriviledged
2012-09-16T23:49:35 < Mazingaro> or shall I copy lr on lr?
2012-09-16T23:50:29 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-16T23:50:30 < Mazingaro> arm code should be orr lr, lr, #4
2012-09-16T23:50:38 < Mazingaro> then bx lr
2012-09-16T23:51:15 < zyp> uh, I lost track of what you are talking abot
2012-09-16T23:51:56 < Mazingaro> ...
2012-09-16T23:52:00 < Mazingaro> did you see my code?
2012-09-16T23:52:47 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-16T23:52:47 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-16T23:53:12 < Mazingaro> it seems that after returning from SysTick ISR, the exception get disabled
2012-09-16T23:54:20 < Mazingaro> I saw that you disabled interrupt at start and reenabled at end, why? Are the irqs disabled during ISR?
2012-09-16T23:54:37 < TitanMKD> Mazingaro to avoid reentrancy
2012-09-16T23:54:50 < TitanMKD> Mazingaro but you can enabled it by sw if you want to manage that
2012-09-16T23:55:04 < Mazingaro> no I don't want reentrancy
2012-09-16T23:55:20 < TitanMKD> all cortex work like that
2012-09-16T23:55:35 < TitanMKD> anyway it is also configurable on other Cortex
2012-09-16T23:55:43 < TitanMKD> evn Mx IIRC
2012-09-16T23:57:15 < zyp> hmm, I can't remember why, being in the highest interrupt priority group should already do that
2012-09-16T23:57:45 < Mazingaro> ok
2012-09-16T23:57:59 < zyp> they were probably slapped on to test something, and then I never removed them
2012-09-16T23:58:19 < zyp> the lack of comments makes it plausible :p
2012-09-16T23:58:40 -!- daku is now known as DaKu
2012-09-16T23:59:54 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-16T23:59:57 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
--- Day changed Mon Sep 17 2012
2012-09-17T00:00:01 < Mazingaro> so?
2012-09-17T00:02:02 < Mazingaro> how can I reenable systick?
2012-09-17T00:02:59 < zyp> are you sure your code works at all after a context switch?
2012-09-17T00:03:15 < Mazingaro> yes it works
2012-09-17T00:03:23 < Mazingaro> it remains in task2 forever
2012-09-17T00:03:34 < Mazingaro> I tested with 2 task
2012-09-17T00:04:39 < Mazingaro> in fact there are no more call to SysTick_Handler
2012-09-17T00:07:04 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T00:07:06 < Mazingaro> ...
2012-09-17T00:07:59 < Mazingaro> gdb won't show all registers
2012-09-17T00:08:35 < zyp> then you're using an old version of gdb and/or a bad debugger
2012-09-17T00:08:37 < Mazingaro> zyp: are you using only the privileged stack?
2012-09-17T00:08:45 < karlp> get either a gdbserver that tells gdb about all registers and a gdb that reads and interptrets that, or don't get upset about it.
2012-09-17T00:08:54 < zyp> currently, yes
2012-09-17T00:08:59 < Mazingaro> arm-none-eabi-gdb from codesourcery
2012-09-17T00:09:57 < zyp> I'm planning to rewrite it to use thread stacks later when I have time to revisit that part of the code
2012-09-17T00:10:35 < zyp> but it works as it is now, it's just a bit more wasteful with the memory
2012-09-17T00:12:01 < zyp> having one handler stack for interrupts means that you don't have to allocate enough ram for any of the ISRs on every thread stack
2012-09-17T00:12:47 < zyp> memory is not scarce yet in my project, so it's not high on the list of stuff I'm planning to do :p
2012-09-17T00:14:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-17T00:15:04 < zyp> Mazingaro, oh, wait, I think I figured out what your problem is
2012-09-17T00:15:26 < Mazingaro> really?
2012-09-17T00:15:42 < zyp> where is the code that creates new threads?
2012-09-17T00:16:20 < Mazingaro> startScheduler
2012-09-17T00:16:44 < zyp> but where do you set up the initial contents of the new thread you are creating?
2012-09-17T00:16:52 < Mazingaro> ok
2012-09-17T00:17:22 < Mazingaro> http://pastebin.com/Wn5pC5Yj
2012-09-17T00:17:24 < Mazingaro> setTCB
2012-09-17T00:18:03 < zyp> ah, yes
2012-09-17T00:18:19 < Mazingaro> :)
2012-09-17T00:18:38 < zyp> you're never really returning from the systick handler, that's why it won't execute again
2012-09-17T00:18:48 < Mazingaro> zyp: yes I know
2012-09-17T00:18:55 < zyp> when you change to the new thread, it's still executing as the systick handler
2012-09-17T00:19:11 < Mazingaro> ah!
2012-09-17T00:19:12 < Mazingaro> yes
2012-09-17T00:19:14 < Mazingaro> then?
2012-09-17T00:19:17 < zyp> the problem is right there in the code you just pasted
2012-09-17T00:19:19 < zyp>   pTCB -> m_uiLR = (uint32_t) pTask -> m_pFunction;
2012-09-17T00:19:46 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- look at this again, I believe your m_uiLR corresponds to my lr_ex
2012-09-17T00:20:16 < Mazingaro> ok
2012-09-17T00:20:18 < Mazingaro> then?
2012-09-17T00:20:30 < Mazingaro> how can I return from ISR and jump in the task?
2012-09-17T00:20:55 < Mazingaro> should I pop the values on the stack?
2012-09-17T00:21:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T00:21:06 < zyp> on cortex, return from an ISR means loading PC with a special value (bx lr loads pc with the contents of lr)
2012-09-17T00:21:22 < zyp> i.e. 0xfffffff9 in my case
2012-09-17T00:21:45 < Mazingaro> so you load that value in PC before bx lr?
2012-09-17T00:22:17 < zyp> LR is filled with that value on interrupt entry, and when you return to LR, the cpu interprets it as a special return
2012-09-17T00:22:41 < Mazingaro> yes
2012-09-17T00:22:56 < zyp> so what you have to do is fill your m_uiLR field with that special value instead of the function of the new thread
2012-09-17T00:22:58 < Mazingaro> and if I pop these values?
2012-09-17T00:23:12 < Mazingaro> and use PC to jump?
2012-09-17T00:23:16 < zyp> the rest of your code probably works as it is
2012-09-17T00:23:35 < Laurenceb_> or you could use chibios
2012-09-17T00:23:38 < Laurenceb_>  /troll
2012-09-17T00:23:51  * Laurenceb_ is happy
2012-09-17T00:24:08 < Laurenceb_> guy who gratified my house just got shot in a drive by
2012-09-17T00:24:26 < zyp> also, in your case the value 0xfffffff9 might not be the correct one
2012-09-17T00:24:37 < Mazingaro> zyp: so I load LR with the stack one and pc with the task?
2012-09-17T00:24:51 < zyp> yes
2012-09-17T00:24:55 < Mazingaro> nah ok
2012-09-17T00:25:00 < Mazingaro> ah ok, tx
2012-09-17T00:25:18 < Mazingaro> I'm modifying it, I'll let you know
2012-09-17T00:25:40 < zyp> when you return to the special value you put into lr, the cpu will take care of loading the actual task
2012-09-17T00:26:10 < zyp> http://infocenter.arm.com/help/topic/com.arm.doc.ddi0337e/CIHJFFIG.html <- anyway, here is the documentation on that
2012-09-17T00:27:08 < zyp> you might want to use 0xfffffffd instead, if you are using process stacks
2012-09-17T00:35:05 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-17T00:39:56 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-114.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-17T00:40:54 < Mazingaro> zyp: so my code works only loading PC with 0xfffffffd and user bx lr?
2012-09-17T00:43:24 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-140.pptp.stw-bonn.de] has joined ##stm32
2012-09-17T00:43:35 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-140.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-17T00:46:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-17T00:51:01 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Quit: Ik ga weg]
2012-09-17T00:52:16 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T00:53:43 < zyp> Mazingaro, don't ask me, it's your code :p
2012-09-17T00:53:53 < Mazingaro> lol
2012-09-17T00:53:54 < Mazingaro> ok tx
2012-09-17T00:54:20 < zyp> did you read the documentation I linked to?
2012-09-17T00:54:33 < zyp> it explains what is going on
2012-09-17T01:08:44 < Laurenceb_> http://www.flickr.com/photos/swiss-strato/7982755829/in/photostream
2012-09-17T01:08:46 < Laurenceb_> insane
2012-09-17T01:10:08 < Mazingaro> zyp: yes I read, but loading 0xfffffffd in pc force processor to jump to HardFault_Handler :)
2012-09-17T01:10:16 < Thorn> stellaris m4 launchpad for $5 if somebody hasn't seen it yet http://www.ti.com/ww/en/launchpad_site/stellaris.html?DCMP=stellaris-launchpad&HQS=stellaris-launchpad-b
2012-09-17T01:10:50 < zyp> Mazingaro, that means that you haven't set up the rest of the stack correctly
2012-09-17T01:14:34 < Laurenceb_> http://www.flickr.com/photos/swiss-strato/7982677478/in/photostream/
2012-09-17T01:23:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-17T01:24:11 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds]
2012-09-17T01:33:12 < Mazingaro> zyp: lol, mrs r0, psp -> HardFault
2012-09-17T01:49:59 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-17T01:50:13 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 260 seconds]
2012-09-17T01:52:10 < Mazingaro> I didn't understand why ldr r0, =pointer -> HardFault, pointer is 8 byte aligned
2012-09-17T01:52:51 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-17T02:08:42 < Mazingaro> ;(
2012-09-17T02:30:24 < Mazingaro> zyp: I did not understand: the manual you show me tells that 0xfffffffx could be loaded in pc using bx too, how?
2012-09-17T02:44:57 < Thorn> Mazingaro: look up the bx instruction and what it does
2012-09-17T02:45:50 < Thorn> to simplify, bx reg is pc := reg
2012-09-17T02:47:18 < Mazingaro> bx lr
2012-09-17T02:47:28 < Mazingaro> returns from subroutine or ISR
2012-09-17T02:48:31 < Thorn> yes it does
2012-09-17T02:49:34 < Mazingaro> ok, but I did not understand how write into pc 0xfffffffd using bx and any register
2012-09-17T02:53:10 < Thorn> load your value into a register with ldr or something, then do bx <reg>. maybe you can load pc directly too, not sure
2012-09-17T02:53:29 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 272 seconds]
2012-09-17T02:54:22 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T02:54:26 < Mazingaro> ldr r0, 0xfffffffd
2012-09-17T02:54:27 < Mazingaro> ?
2012-09-17T02:54:41 < Thorn> something like that
2012-09-17T02:55:13 < Mazingaro> assembler won't assemble it
2012-09-17T02:57:36 < Mazingaro> :(
2012-09-17T02:57:46 < Thorn> if you're using gcc, you need ldr r0, =0xfffffffd afair
2012-09-17T02:58:32 < Thorn> and even if you're not
2012-09-17T02:59:19 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-09-17T02:59:37 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T02:59:54 < Mazingaro> ok, but how can I suppose to return to task code using bx r0 with r0 loaded with that value?
2012-09-17T03:02:03 < Mazingaro> in fact it jumps to HardFault directly after bx r0
2012-09-17T03:02:23 < Thorn> what are you doing, returning from an interrupt? the core loads lr with a correct value automatically on exception entry, all you need to do is bx lr (unless you destroyed lr)
2012-09-17T03:02:39 < Thorn> if it hardfaults, then you probably aren't in an exception
2012-09-17T03:03:58 < Mazingaro> I'm in naked SysTick Handler
2012-09-17T03:04:37 -!- grummund [~user@unaffiliated/grummund] has quit [Read error: Connection reset by peer]
2012-09-17T03:04:56 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:05:14 < Thorn> are you sure the core is in the handler mode?
2012-09-17T03:05:35 < Thorn> and you aren't messing with stacks?
2012-09-17T03:05:44 < Mazingaro> doesn't it enter at exception?
2012-09-17T03:06:22 < Thorn> because it takes the actual return address from the stack
2012-09-17T03:06:47 < Mazingaro> I did not touch sp
2012-09-17T03:07:12 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T03:07:39 < Thorn> and I assume you didn't overwrite stack memory either.
2012-09-17T03:07:50 < Mazingaro> no
2012-09-17T03:08:09 < Thorn> then it can't hardfault
2012-09-17T03:08:15 < Mazingaro> but it is
2012-09-17T03:08:30 < Mazingaro> ah
2012-09-17T03:08:37 < Mazingaro> you were right
2012-09-17T03:08:45 < Mazingaro> I'm not in an exception
2012-09-17T03:09:08 < Mazingaro> this happen in startScheduler which is not started in handler mode
2012-09-17T03:09:14 < Mazingaro> sorri in exception mode
2012-09-17T03:09:58 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-09-17T03:10:14 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:14:40 < Thorn> zomg these dies are HUGE http://www.anandtech.com/show/5875/dual-coregt2-ivy-bridge-die-measured-121mm2
2012-09-17T03:14:43 < Mazingaro> Thorn: problem was that I assume that startScheduler, which is called in main, should perform a context restoring to launch first task
2012-09-17T03:14:46 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-17T03:15:29 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:20:58 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-17T03:22:09 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:26:37 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-09-17T03:27:29 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:28:08 < upgrdman> any idea what HG1-3 would be? center of this photo: http://www.rcheliguy.com/temp/IMG_4265.JPG
2012-09-17T03:31:47 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-17T03:32:02 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-17T03:33:42 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:37:38 < Mazingaro> Thorn: it's a mess, nothing works... Damned arm it is worst than x86 :(
2012-09-17T03:37:39 < Mazingaro> lol
2012-09-17T03:38:02 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-09-17T03:39:09 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:41:49 < dongs> upgrdman: lol what is that
2012-09-17T03:43:40 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-17T03:43:58 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 268 seconds]
2012-09-17T03:44:07 < Thorn> what's black and white and should be read all over?
2012-09-17T03:44:38 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:45:05 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-17T03:46:35 < Thorn> http://i.imgur.com/2BCFQ.jpg
2012-09-17T03:48:39 < dongs> is this some new valve shit
2012-09-17T03:49:17 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-09-17T03:49:58 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:49:58 < Thorn> fan made shit
2012-09-17T03:50:05 < dongs> o
2012-09-17T03:50:07 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-17T03:50:07 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-17T03:50:07 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-17T03:50:10 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-17T03:54:11 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-09-17T03:55:17 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T03:59:40 < dongs> http://www.youtube.com/watch?v=4BLmPd4W-sY
2012-09-17T04:00:17 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-09-17T04:00:26 < Mazingaro> Thorn: I still didn't understand how it works...
2012-09-17T04:00:37 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-17T04:00:57 -!- DaKu is now known as daku
2012-09-17T04:01:06 < Mazingaro> Thorn: documentation is not clear
2012-09-17T04:01:40 < Thorn> what works? exception return?
2012-09-17T04:02:45 < Mazingaro> I did not understand how return from exception changing the return address to jump to other code
2012-09-17T04:05:25 < Thorn> do you have The Book?
2012-09-17T04:06:31 < Mazingaro> I've got a book on arm not cortex
2012-09-17T04:06:53 < Mazingaro> arm it's another stroy
2012-09-17T04:06:58 < Mazingaro> story
2012-09-17T04:07:01 < Thorn> The Definitive Guide to ARM Cortex-M3, 2nd ed.
2012-09-17T04:07:08 < Mazingaro> ah no
2012-09-17T04:07:26 < Thorn> written by the principal developer of the architecture
2012-09-17T04:07:48 < Mazingaro> ok
2012-09-17T04:07:52 < Thorn> it explains all of this in a remarkably clear language
2012-09-17T04:08:38 < Mazingaro> ok
2012-09-17T04:08:45 < Mazingaro> how much does it cost?
2012-09-17T04:09:20 < Thorn> http://www.amazon.com/Definitive-Guide-Cortex-M3-Second-Edition/dp/185617963X
2012-09-17T04:09:50 < Mazingaro> no money here!
2012-09-17T04:13:57 < Mazingaro> :(
2012-09-17T04:14:44 < Thorn> well you could try reading it on google books
2012-09-17T04:15:35 < Mazingaro> ?
2012-09-17T04:16:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T04:46:01 < dongs> feertos
2012-09-17T05:02:08 < Thorn> sort of electronics lab http://i.imgur.com/QJtwF.jpg
2012-09-17T05:06:31 < dongs> IM THAT NIGGA
2012-09-17T05:14:07 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-17T05:23:28 < upgrdman> dongs, didn't see your reply from hours ago. that photo is of a pcb from a 5.25" floppy disk drive
2012-09-17T05:23:31 < upgrdman> any idea what HG1-3 would be? center of this photo: http://www.rcheliguy.com/temp/IMG_4265.JPG
2012-09-17T05:25:30 < dongs> i would guess something hall related maybe? just a guess though.. if hte spinning disk on top has some sorta magnetic shits on them for position detection
2012-09-17T05:25:36 < dongs> i dunno why they're so close together though
2012-09-17T05:27:04 < upgrdman> theres a big ass rotor that covers those coils and chips... i don't think they do any sensing
2012-09-17T05:27:33 < upgrdman> i'll have to look up that big IC and see if the datasheet gives any clues
2012-09-17T05:35:49 < dongs> found some obsolete shit while cleaning http://i.imgur.com/5PDhC.jpg
2012-09-17T05:46:31 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-17T05:50:50 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-17T05:52:55 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-17T06:08:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-17T06:08:15 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-17T06:11:03 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-17T06:49:48 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-17T07:07:02 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 252 seconds]
2012-09-17T07:11:29 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-17T07:38:11 < dongs> I'm ordering some white PCBs...
2012-09-17T07:38:13 < dongs> http://i.imgur.com/ZAMp4.jpg
2012-09-17T07:39:15 < emeb_mac> demo boards for passives?
2012-09-17T07:39:21 < dongs> yeah.
2012-09-17T07:39:22 < dongs> lol.
2012-09-17T07:39:28 < dongs> gonna panel some 32minis with it.
2012-09-17T07:39:37 < emeb_mac> along for the ride
2012-09-17T07:41:31 < emeb_mac> "Frame proof"?
2012-09-17T07:47:38 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-17T08:18:15 < dongs> lmao
2012-09-17T08:18:18 < dongs> gonna get that fix'd
2012-09-17T08:23:22 -!- claude is now known as Claude
2012-09-17T08:28:42 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-09-17T08:29:07 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has joined ##stm32
2012-09-17T08:29:07 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has quit [Changing host]
2012-09-17T08:29:07 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-17T08:35:14 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-17T09:01:36 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T09:17:15 < dongs> http://www.ebay.ca/itm/290729233824 killer deal
2012-09-17T09:18:05 < emeb_mac> that's a lotta 555s
2012-09-17T09:19:00 <+izua> i think Thrashbarg in #electronics made a ntsc generator & pong game with ~8
2012-09-17T09:20:20 < emeb_mac> ingenious
2012-09-17T09:23:27 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-17T09:30:39 < gnomad> those timers do you long time!
2012-09-17T09:34:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-17T09:40:05 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 272 seconds]
2012-09-17T09:56:36 < dongs> http://www.kickstarter.com/projects/323660810/640-1280-2560tiny?ref=home_location
2012-09-17T09:56:40 < dongs> ...
2012-09-17T09:56:42 < dongs> more fail
2012-09-17T09:58:04 < dongs>  I have been hand soldering each one as of now, and the 100 pin TQFP package takes about 30 minutes without a re flow oven.
2012-09-17T09:58:07 < dongs> whaaaaaaaaaaaaaat
2012-09-17T09:58:18 < dongs> i hand-soldered that lqfp48 shit on mini32 in like half a minute
2012-09-17T09:58:24 < dongs> what hte dick
2012-09-17T10:13:40 -!- dekar [~dekar@212.255.17.115] has quit [Quit: This computer has gone to sleep]
2012-09-17T10:37:28 -!- Claude is now known as claude
2012-09-17T10:41:23 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-17T10:50:29 -!- claude is now known as claude_work
2012-09-17T10:53:04 -!- dekar [~dekar@dslb-088-069-146-196.pools.arcor-ip.net] has joined ##stm32
2012-09-17T10:53:07 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-17T10:54:39 < Mazingaro> hi
2012-09-17T10:57:10 < Mazingaro> I need help with a context switch routine I wrote... I could not get the processor return to task code after SysTick ISR.
2012-09-17T11:02:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-17T11:12:46 < Mazingaro> I could not get any of the thing said in the arm doc about returning from exception working
2012-09-17T11:17:24 < Mazingaro> especially
2012-09-17T11:17:43 < Mazingaro> I'm trying to load the EXC_RETURN in pc, but all I get is HardFault
2012-09-17T11:36:14 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-17T11:36:14 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-17T11:36:14 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-17T11:36:17 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-17T11:41:45 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-17T11:57:27 -!- grummund_ [~user@unaffiliated/grummund] has joined ##stm32
2012-09-17T11:58:54 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 264 seconds]
2012-09-17T12:04:20 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-17T12:05:57 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-17T12:08:45 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-17T12:08:48 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-17T12:09:24 < Mazingaro> ...
2012-09-17T12:09:39 < Mazingaro> please, is there anyone could help me? tx
2012-09-17T12:11:16 < zyp> pastebin your code
2012-09-17T12:15:05 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
2012-09-17T12:17:47 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds]
2012-09-17T12:22:42 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-17T12:23:55 <+Steffanx> You'll never see his code zyp
2012-09-17T12:28:28 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-17T12:55:57 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-17T13:20:16 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-17T13:20:19 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-17T13:23:20 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-17T13:27:27 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 272 seconds]
2012-09-17T13:51:14 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-17T14:03:29 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-09-17T14:05:42 -!- tavish [~tavish@120.56.169.81] has joined ##stm32
2012-09-17T14:06:06 -!- tavish is now known as Guest74938
2012-09-17T14:10:20 -!- Guest74938 is now known as tavish3
2012-09-17T14:31:23 < Laurenceb> http://www.anandtech.com/show/6292/iphone-5-a6-not-a15-custom-core
2012-09-17T14:33:14 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-115.pptp.stw-bonn.de] has joined ##stm32
2012-09-17T14:34:50 < zyp> old news
2012-09-17T14:48:15 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-17T14:57:06 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-17T14:57:06 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-17T14:57:13 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-17T14:58:47 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-09-17T14:59:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-17T15:00:20 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 248 seconds]
2012-09-17T15:00:32 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-09-17T15:01:08 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-17T15:12:03 -!- daku is now known as DaKu
2012-09-17T15:14:21 -!- SuicideFunky [~randy@vps-212-153-70-76.twenty-five.nl] has joined ##stm32
2012-09-17T15:24:37 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-17T15:33:20 -!- tavish3 [~tavish@120.56.169.81] has quit [Quit: leaving]
2012-09-17T15:46:28 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-17T16:14:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-17T16:19:45 < Laurenceb> arg
2012-09-17T16:19:51 < Laurenceb> stupid f4 ref manual
2012-09-17T16:19:59 < Laurenceb> its so full of errors
2012-09-17T16:31:21 < Erlkoenig> isn't there an errata sheet?
2012-09-17T16:32:58 < zyp> errata sheet are about hardware errors, not documentation errors :p
2012-09-17T16:35:28 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-09-17T16:39:25 < Thorn> how do you call an errata sheet for an errata sheet?
2012-09-17T16:40:09 < dongs> you're fired
2012-09-17T16:41:26 < Thorn> not gonna happen in ST
2012-09-17T16:42:46 < Thorn> also, why errata 'sheet' if it's up to 50 pages long
2012-09-17T16:44:57 < zyp> it's just a nice way of spelling «shit»
2012-09-17T16:47:43 < gnomad> In other news, it is often useful to check the size of eratta sheets before selecting a processor...
2012-09-17T16:49:38 < Erlkoenig> i like the way clocks are namen in the stm32f4 sheet - the names in the "clock tree" figure differ from the names elsewhere... :D
2012-09-17T16:49:48 < karlp> unfortunately, it's even more useful to check things like availability
2012-09-17T16:50:22 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-17T16:55:01 < BrainDamage> one of my ex univ acquaintances works at ST, he's in the layout dept, and worked on stm32, knowing him I espect he was part if the errata problem :p
2012-09-17T16:56:30 < zyp> haha
2012-09-17T16:58:08 < gnomad> I would imagine there would be software to detect layout errors...
2012-09-17T17:00:10 < Thorn> as I understand it, verification software is much more expensive (and engineers expericenced in verification are much more rare) than design software
2012-09-17T17:01:00 < BrainDamage> most of the layout process is automated
2012-09-17T17:01:18 < BrainDamage> but there's still need of manual intervenction when the software fails to meet design constraints
2012-09-17T17:03:44 < Erlkoenig> does "layout error" mean the circuit's schematic is correct, but during layout some fail happened?
2012-09-17T17:04:38 < BrainDamage> that the layout implementation does not match the schematic specs
2012-09-17T17:05:04 < BrainDamage> it simply means it's different, not that it doesn't work
2012-09-17T17:05:14 < Erlkoenig> when you have a layout and a schematic, is it really so difficult to compare & verify them, hmm
2012-09-17T17:05:35 < Thorn> I'd think logic verification is responsible for most bugs
2012-09-17T17:05:57 < BrainDamage> silicon implenetation is not like your discrete glueing of readily made parts
2012-09-17T17:06:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T17:06:08 < Erlkoenig> oh rly? :D
2012-09-17T17:06:27 < BrainDamage> or better, it can be done to certain extent
2012-09-17T17:06:45 < BrainDamage> but when you're splitting hairs, not really
2012-09-17T17:12:56 < Thorn> afaik signal integrity is mostly a non-issue in IC design unless you're intel or nvidia
2012-09-17T17:17:47 <+Steffanx> Why people always forget AMD? :P
2012-09-17T17:18:33 < Thorn> who's that
2012-09-17T17:19:01 <+Steffanx> Not your mother
2012-09-17T17:30:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-17T17:42:30 < Thorn> "Newhaven Display International, Inc. is excited to announce a breakthrough product line in the display industry. More details about the new product release are soon to come." <-- what an informative email.
2012-09-17T17:43:35 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-17T17:43:51 <+Steffanx> They made you curious ? mission accomplished
2012-09-17T17:46:05 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-17T17:47:08 < Thorn> they even used a picture of a box with a question mark in it. quality trolling
2012-09-17T17:58:25 < Laurenceb> coming soon... something
2012-09-17T18:02:56 <+Steffanx> You spectrometer too Laurenceb ?
2012-09-17T18:22:52 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 246 seconds]
2012-09-17T18:30:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T18:30:08 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-17T18:30:27 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-17T18:30:33 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-09-17T18:30:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-17T18:32:26 < Mazingaro> hi
2012-09-17T18:32:32 < Mazingaro> zyp: here's my code
2012-09-17T18:33:27 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-17T18:34:08 < Mazingaro> zyp: http://pastebin.com/4wtAn4CU
2012-09-17T18:36:58 -!- claude_work is now known as claude
2012-09-17T18:40:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T18:40:38 < karlp> yeah, what could possibly go wrong.
2012-09-17T18:48:30 -!- dekar [~dekar@dslb-088-069-146-196.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-17T18:59:09 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T19:00:08 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-17T19:10:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-17T19:30:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-17T19:30:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-17T19:33:25 -!- barthess [~barthess@178.154.85.217] has joined ##stm32
2012-09-17T19:40:21 < karlp> anyone know why "make" (with a Makefile in pwd) works, but "mv Makefile hate; make -f hate" fails?
2012-09-17T19:41:01 < karlp> it's not finding a rule somewhere, and fucks me what's wrong with it.
2012-09-17T19:41:11 < karlp> something to do with autogenerated prereqs I bet though
2012-09-17T19:41:23 < karlp> make -d's output is less that intelligible
2012-09-17T19:42:05 < Erlkoenig> is your rule in your "hate" file or somewhere else?
2012-09-17T19:42:31 < karlp> in another one, which hasn't changed names.
2012-09-17T19:42:32 < Erlkoenig> maybe there is a rule that depends on "Makefile", so it will be re-run when "Makefile" has changed
2012-09-17T19:42:38 < karlp> ahh, good call.
2012-09-17T19:42:41 < karlp> let me look for that
2012-09-17T19:42:58 < karlp> yep, there it is.
2012-09-17T19:43:01 < karlp> excellllent
2012-09-17T19:43:03 < karlp> thanks muchly.
2012-09-17T19:44:00 < Erlkoenig> you're welcome ;)
2012-09-17T19:44:31 < Erlkoenig> out of interest, which kind of while is regenerated this way?
2012-09-17T19:47:16 < karlp> Erlkoenig: http://pastebin.com/HpzqZ51H
2012-09-17T19:47:38 < karlp> guess-rev.sh came from openwrt's buildroot at some stage, can't remeber anymore.
2012-09-17T19:47:57 < karlp> it returns something like "-gdd1616c-svn-dirty"
2012-09-17T19:48:06 < Erlkoenig> wah... evil
2012-09-17T19:48:08 < karlp> (that's in a modified git-svn repo)
2012-09-17T19:48:19 < karlp> why evil?
2012-09-17T19:50:12 < Erlkoenig> hm probably not so evil :>
2012-09-17T19:50:17 < Erlkoenig> but a bit confusing maybe
2012-09-17T19:52:02 < karlp> what way do you get the version control information of a build into the binary?
2012-09-17T19:55:53 < Erlkoenig> maybe have the version.h depend on the version-control file that stores the current version... but that will cause problems when the internal implementation of the version control system changes
2012-09-17T19:56:15 < Erlkoenig> or just pass the version via the "-D" compiler switch
2012-09-17T19:56:33 < olasd> yeah, I'd use -DVERSION=bla
2012-09-17T19:56:37 < zyp> Mazingaro, your link doesn't work now
2012-09-17T19:57:13 < zyp> the problem with -DVERSION is that you don't get any dependancy tracking on that
2012-09-17T19:57:24 < zyp> so you'll have to recompile everything when the version changes
2012-09-17T19:58:06 < Erlkoenig> maybe the version control system should provide some file that contains the version in some format, and automatically modifies that file upon commit
2012-09-17T19:58:18 < karlp> no. that's a terrible idea.
2012-09-17T19:58:37 < karlp> I've always hated that behaviour in cvs
2012-09-17T19:59:22 < Erlkoenig> what problems would this generate, given that this feature would be optional?
2012-09-17T19:59:47 < zyp> IMO the best solution is to have a rule that always executes, which pulls the version information and then updates version.h
2012-09-17T20:00:04 < karlp> whcih is what mine is doing. (version.h is marked phony, so it always runs it)
2012-09-17T20:00:31 < karlp> Erlkoenig: you can't rely on people to check in the right file with manual chnges or auto changes
2012-09-17T20:00:35 < karlp> it needs to be automatic.
2012-09-17T20:00:51 < karlp> or you end up with multiple releases with the same version
2012-09-17T20:01:22 < Erlkoenig> well i'd say the version system would update this file automatically?
2012-09-17T20:01:34 < Erlkoenig> so you can't forget it
2012-09-17T20:02:15 < Erlkoenig> but if version.h is generated on  every "make" execution, every source file dependant on it will be regenerated every time?
2012-09-17T20:02:37 < olasd> which you want anyway
2012-09-17T20:02:51 < olasd> oh well, maybe
2012-09-17T20:04:48 < Erlkoenig> well you don't want it if version.h gets a new "last modified" date but actually the version numbers stayed the same?
2012-09-17T20:04:55 < Mazingaro> zyp: hi again
2012-09-17T20:05:20 < olasd> generate version.h.tmp, compare with version.h, only change version.h if relevant
2012-09-17T20:05:38 < olasd> (which starts getting a bit tricky)
2012-09-17T20:05:45 < Mazingaro> zyp: http://pastebin.com/3BCfa6AF
2012-09-17T20:05:59 < zyp> olasd, in make perhaps
2012-09-17T20:06:19 < zyp> scons is using checksum to see if a file has changed or not
2012-09-17T20:09:36 < zyp> Mazingaro, I'm not sure what you are actually trying to do there, can you put a comment on what the intention of each line is?
2012-09-17T20:12:36 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-17T20:12:36 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-17T20:12:36 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-17T20:12:39 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-17T20:21:37 < Thorn> http://www.ganssle.com/tem/tem228.html#article2 m4 vs m0 power consumption
2012-09-17T20:22:53 < Erlkoenig> i love the title.
2012-09-17T20:27:17 < zyp> results are as I expected
2012-09-17T20:29:06 < zyp> M4 is using less power to perform a given task, because it's done much faster, but also draws more current when the task is performed
2012-09-17T20:30:05 < Thorn> so if you're running off a battery use a fast chip and sleep as much as possible?
2012-09-17T20:30:24 < Thorn> (sleep current seems the same between them)
2012-09-17T20:30:27 < HTT-Bird> in many cases, basically.
2012-09-17T20:30:55 < zyp> fast doesn't imply efficient
2012-09-17T20:30:58 < HTT-Bird> the current app I'm working on can go with wakeups as infrequent as every 10-20s
2012-09-17T20:30:58 <+Steffanx> Too bad he didn't test a 'low power' ARM like the L1 family
2012-09-17T20:30:59 < zyp> but yeah
2012-09-17T20:31:14 < Thorn> I heard sleep modes make firmware development much more difficult
2012-09-17T20:31:28 < Thorn> *using sleep modes makes
2012-09-17T20:31:37 < HTT-Bird> the main problem there is that JTAG cuts out during sleep on some (but not all) ARM setups
2012-09-17T20:31:41 < zyp> Steffanx, L1 is just M3 with a low-power process
2012-09-17T20:31:49 <+Steffanx> Yeah, but still ..
2012-09-17T20:31:58 < Thorn> waiting for L0
2012-09-17T20:32:16 < Thorn> (doesn't nXP have something L0-like already?)
2012-09-17T20:32:27 < zyp> I expect that L0 vs L1 will have the same results as that test
2012-09-17T20:32:29 < HTT-Bird> Thorn: LPC1100L/XL
2012-09-17T20:32:38 < HTT-Bird> but the problem with those is that their BOD is a current hog
2012-09-17T20:32:47 <+Steffanx> 1100 .. where 00 means no internal flash?
2012-09-17T20:32:49 < HTT-Bird> no
2012-09-17T20:32:55 < HTT-Bird> series designation
2012-09-17T20:33:01 < zyp> L1 will be faster than L0 and thus able to sleep more, so it should be more power efficient
2012-09-17T20:33:11 < zyp> unless something keeps you from sleeping
2012-09-17T20:33:58 < zyp> well
2012-09-17T20:34:29 < Thorn> iirc on Curiosity they inplemented sleep modes for the RAD750 (which apparently doesn't have any) using FPGAs
2012-09-17T20:34:31 < zyp> a fair amount of the difference in that test can probably be attributed to FPU vs FPU-emulation, which doesn't apply to the M3
2012-09-17T20:35:08 < zyp> so I would expect the M3 to be somewhere in the middle when it comes to power efficiency
2012-09-17T20:36:07 <+Steffanx> Time for a better / more general test :)
2012-09-17T20:36:24 <+Steffanx> Not just 300 square root calculations
2012-09-17T20:38:16 < Erlkoenig> (19:31:37) HTT-Bird: the main problem there is that JTAG cuts out during sleep on some (but not all) ARM setups <-- not on STM32F4 ;-)
2012-09-17T20:38:24 < HTT-Bird> hehe
2012-09-17T20:38:37 < karlp> isn't it configurable whether it does or doesn't?
2012-09-17T20:38:39 <+Steffanx> Isn't that something you can enable/disable?
2012-09-17T20:38:51 < Erlkoenig> you can enable that? tell me!11 :D
2012-09-17T20:38:54 < Thorn> in fact I'd expect that CoreSignt takes care of that
2012-09-17T20:39:19 < Erlkoenig> using while(1) __WFI (); ... seems to lock the MCU so you can't reset/reflash it, at least when using SWD
2012-09-17T20:40:09 < zyp> Erlkoenig, see DBGMCU documentation in reference manual
2012-09-17T20:41:17 < Erlkoenig> ah, i tried    DBGMCU_CR |= 1;    but got an "DBGMCU_CR" not define, IIRC
2012-09-17T20:41:22 < Erlkoenig> defined
2012-09-17T20:41:49 < zyp> then define it?
2012-09-17T20:42:02 < Thorn> DBGMCU->CR?
2012-09-17T20:42:05 < Erlkoenig> yeah, to what? 0xF00BA8? :D
2012-09-17T20:42:25 < zyp> reference manual says 0xe0042004
2012-09-17T20:42:33 < Erlkoenig> hmmm i only remember that it "didn't work" somehow... but not why :D
2012-09-17T20:42:46 < Erlkoenig> that was half a year ago or something :S
2012-09-17T20:42:46 < zyp> it didn't work because you didn't do it right.
2012-09-17T20:42:51 < Erlkoenig> probably :D
2012-09-17T20:43:14 < karlp> also, there's very little / no support in texane/stlink for recovering from any of this.
2012-09-17T20:43:23 < karlp> it's an area that needs work.
2012-09-17T20:43:37 < Thorn> what about BMP?
2012-09-17T20:43:48 <+Steffanx> Never tried it
2012-09-17T20:43:54 < zyp> me neither
2012-09-17T20:44:08 < Erlkoenig> i think i recovered my F4 from that while(1) __WFI();   by holding reset, starting the debug tool on the PC and then releasing the reset
2012-09-17T20:45:02 < zyp> I'm not touching sleep until I'm writing a proper scheduler
2012-09-17T20:45:58 < zyp> the only time it's reasonable to go to sleep is when no threads are ready to run, and currently I don't have any notion of readyness, I just treat all threads as ready to run at any time
2012-09-17T20:46:07 < Erlkoenig> i made a super simple accelerometer->USB adapter on my F4 Discovery, and inserting the __WFI (); instead of while(1); reduced the power usage a lot
2012-09-17T20:46:40 < Erlkoenig> of course i did something wrong and i basically got some while(1) __WFI(); -style of thing
2012-09-17T20:46:45 < zyp> :p
2012-09-17T20:47:26 < Erlkoenig> i presented that toy to some friends, and they said "meh, my iphone can do that as well"
2012-09-17T20:47:29 < Thorn> it doesn't drop its usb connection?
2012-09-17T20:47:37 < zyp> Thorn, why would it?
2012-09-17T20:47:42 <+Steffanx> Haha Erlkoenig
2012-09-17T20:47:44 < Erlkoenig> Thorn: interrupts were still handled
2012-09-17T20:47:47 < zyp> it's just stopping the core until an interrupt comes in
2012-09-17T20:49:36 < Erlkoenig> Steffanx: he also told me "making nice clever Apps for the iFoo stuff is supercool and earns you money" ... whereas i thought that probably programs which actually do something (calculating numeric algorithms or something) is cooler and these can't be imitated that easily
2012-09-17T20:50:18 < Thorn> zyp: right
2012-09-17T20:51:19 < zyp> also, having an usb connection doesn't really require anything when it's idle
2012-09-17T20:51:47 < jpa-> you have to be ready to respond, though
2012-09-17T20:51:59 < zyp> only when there is something to respond to
2012-09-17T20:52:04 < Thorn> how does the host recognize when the device is disconnected then?
2012-09-17T20:52:14 < jpa-> L1 USB seems to take ~5mA as soon as it is connected
2012-09-17T20:52:55 < Erlkoenig> Thorn: probably because the device drives the data lines to some defined state, even in idle mode?
2012-09-17T20:53:04 < zyp> jpa-, is the code handling SOF interrupts?
2012-09-17T20:53:15 < zyp> Erlkoenig, exactly
2012-09-17T20:53:24 < jpa-> zyp: probably, it's the chibios code
2012-09-17T20:53:44 < Erlkoenig> ... also the Host polls the device regularly, and if it doesn't respond, it's probably considered as "disconnected". but between these polls, the device µC can as well got to sleep via __WFI()
2012-09-17T20:53:49 < zyp> presence of LS/FS devices is determined by a pullup on one of the data lines
2012-09-17T20:54:14 < zyp> removing the pullup makes the host see the device as disconnected
2012-09-17T20:54:40 < jpa-> Erlkoenig: actually, even if the processor crashes, linux seems to keep the device connected for quite long
2012-09-17T20:54:53 < Erlkoenig> really? uhm...
2012-09-17T20:54:55 < zyp> there is no polling of idle devices.
2012-09-17T20:55:14 < Erlkoenig> that happens if you let the StdPeriph do all the work... :D
2012-09-17T20:55:36 < zyp> I can have my device enumerate, then halt in the debugger and poke around, then resume and still have a working connection
2012-09-17T20:56:16 < Mazingaro> zyp: http://pastebin.com/E3qH0uhi
2012-09-17T20:56:29 < Thorn> in my case the device got disconnected as soon as I stopped it with a debugger, but the host was sending periodic requests
2012-09-17T20:56:39 < Mazingaro> zyp: http://pastebin.com/E3qH0uhi
2012-09-17T20:56:46 < Mazingaro> zyp this la
2012-09-17T20:56:49 < Mazingaro> zyp this last
2012-09-17T20:57:03 < zyp> in case of an incoming packet on an endpoint, the package will be buffered by the device controller and acked
2012-09-17T20:57:18 < Erlkoenig> hacked?
2012-09-17T20:57:23 < Erlkoenig> :D
2012-09-17T20:57:29 < zyp> since the buffer for that endpoint then is full, the next packet will be NAKed
2012-09-17T20:57:49 < zyp> i.e. normal usb flow control
2012-09-17T20:59:53 -!- dekar [~dekar@212.255.17.115] has joined ##stm32
2012-09-17T20:59:55 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-17T21:01:46 < zyp> Mazingaro, «set EXC_RETURN in lr» you don't need to do it like that
2012-09-17T21:02:38 < Mazingaro> zyp, lr is 0xfffffff9 I need 0xfffffffd
2012-09-17T21:02:44 < zyp> Mazingaro, consider this, in SAVE_CONTEXT, lr already contains EXC_RETURN, so you are saving it there
2012-09-17T21:03:26 < Mazingaro> ok but how can I switch to thread?
2012-09-17T21:03:32 < zyp> and then in RESTORE_CONTEXT you are reading back the saved lr, which still contains EXC_RETURN
2012-09-17T21:03:42 < zyp> well, it's easy
2012-09-17T21:04:32 < zyp> when you make new threads, you have to fill the lr field with EXC_RETURN so that's what RESTORE_CONTEXT reads the first time you are switching to that thread
2012-09-17T21:05:35 < Mazingaro> have the lr field be copied in lr during RESTORE_CONTEXT?
2012-09-17T21:05:57 < zyp> "stmdb r12!, { r0, lr }\n")  // save tcb pointer and lr
2012-09-17T21:06:13 < zyp> that lr field you are saving there contains the EXC_RETURN value
2012-09-17T21:06:31 < zyp> you have to put that back into lr in RESTORE_CONTEXT
2012-09-17T21:06:53 < Mazingaro> well SAVE_CONTEXT is completely wrong, my point for now is RESTORE_CONTEXT
2012-09-17T21:07:18 < Mazingaro> how can I jump back to thread?
2012-09-17T21:07:31 < Mazingaro> RESTORE_CONTEXT has to pop from PSP the PC too?
2012-09-17T21:07:40 < Erlkoenig> are you writing your own scheduler/OS ?
2012-09-17T21:07:52 < Mazingaro> Erlkoenig: yes
2012-09-17T21:08:03 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-17T21:08:04 < Mazingaro> Erlkoenig: I did that for atmega2560, but I need more power
2012-09-17T21:08:13 < Erlkoenig> hmm i guess every embedded developer has to do this once :D
2012-09-17T21:08:19 < Mazingaro> Erlkoenig: so I took a chance with stm32
2012-09-17T21:08:32 < Mazingaro> stm32 is amazing
2012-09-17T21:08:50 < Mazingaro> but I didn't know the difference from cortex-m3 to a real ARM
2012-09-17T21:09:23 < Erlkoenig> what's a "real ARM"? :D
2012-09-17T21:09:41 < zyp> something with not thumb-only
2012-09-17T21:09:41 < Erlkoenig> one that has the ARM-mode and not just Thumb?
2012-09-17T21:09:50 < zyp> I would guess :p
2012-09-17T21:10:04 < zyp> Mazingaro, first of all, do you really need a seperate handler stack?
2012-09-17T21:10:13 < Mazingaro> zyp: like STR912 based on arm926
2012-09-17T21:11:03 < Mazingaro> zyp: I prefer it so I would have separate stacks in thread mode non interfering with the machine stack
2012-09-17T21:11:42 < Mazingaro> zyp: anyway should I provide a value for PC too?
2012-09-17T21:11:58 < zyp> there aren't really any interference, interrupts would just stack on top of whatever the current thread stack is
2012-09-17T21:12:02 < zyp> but anyway
2012-09-17T21:12:44 < zyp> http://cgit.jvnv.net/laks/tree/interrupt/fault.cpp <- here is my code for no seperate handler stack, I'll try rewriting it to what it would have to be if I had a seperate handler stack
2012-09-17T21:13:04 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-17T21:16:38 <+Steffanx> You have plans to rewrite everthing, lol :P
2012-09-17T21:17:59 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-17T21:18:05 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-17T21:18:29 < Mazingaro> zyp: tx
2012-09-17T21:18:40 < zyp> http://paste.jvnv.net/view/IXRD7 <- something like this, I believe
2012-09-17T21:19:05 < Erlkoenig> I've thought of an OS/scheduler that does nothing (no context switches) as long as one thread is running happily, and only when the thread starts a blocking call (like waitForUART (); or something) checks whether there's an other thread that can be run and in that case switches to that one (else does __WFI()).
2012-09-17T21:19:06 < Erlkoenig> This switch would also be done if that other thread is considered of "higher priority", currently waiting e.g. on SPI-data and an SPI Interrupt occurs. This would cause minimum overhead (by only doing switches when absolutely neccessary) but allow for only one calculation-intensive thread. Does something like that exist?
2012-09-17T21:19:11 < zyp> obviously untested
2012-09-17T21:19:27 < Mazingaro> zyp: so I do not have to manipulate the PC in the machine SP
2012-09-17T21:19:46 < zyp> but the idea is that instead of operating directly on ps (which is now msp), I read psp to r12 and operate on that
2012-09-17T21:20:05 < zyp> except for that there is no change
2012-09-17T21:20:58 < zyp> Erlkoenig, I'm doing something similar to what you are describing
2012-09-17T21:21:03 < Laurenceb_> Erlkoenig: ChibiOS kind of works like that
2012-09-17T21:21:10 < zyp> i.e. cooperative multitasking
2012-09-17T21:21:24 < Laurenceb_> but also has task switching inside the systick hander
2012-09-17T21:21:38 < Laurenceb_> so it can switch inside an isr
2012-09-17T21:21:45 < Erlkoenig> wah, evil :D
2012-09-17T21:22:06 < zyp> Erlkoenig, I'm performing a context switch in the svcall interrupt
2012-09-17T21:22:33 < Erlkoenig> svcall was this kind of "software interrupt", right?
2012-09-17T21:22:38 < Laurenceb_> im currently using the adc dma interrupt to schedule a thread to process the data
2012-09-17T21:22:41 < zyp> which is invoked by the svc instruction, which I call from a function called Thread::yield()
2012-09-17T21:22:49 < Mazingaro> zyp: how can I get the thread address where SysTick exception occurred?
2012-09-17T21:23:00 < Mazingaro> zyp: I need it in SAVE_CONTEXT
2012-09-17T21:23:07 < zyp> Mazingaro, why?
2012-09-17T21:23:15 < Erlkoenig> zyp: ah, and you would call Thread::yield() when e.g. you're waiting for the UART to receive data?
2012-09-17T21:23:21 < zyp> Erlkoenig, exactly
2012-09-17T21:23:43 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 272 seconds]
2012-09-17T21:23:47 < zyp> http://cgit.jvnv.net/laks/tree/os/time.h <- see for instance how my sleep works
2012-09-17T21:23:55 < Erlkoenig> zyp: do you have some structure that only jumps back to this thread when the UART actually received something (i.e. the UART ISR occured)?
2012-09-17T21:24:21 < zyp> no, as I mentioned earlier, I treat all threads as ready
2012-09-17T21:24:26 < Laurenceb_> but how long does yield yield for?
2012-09-17T21:24:31 < zyp> it's really just a linked-list-loop
2012-09-17T21:24:42 < zyp> Laurenceb_, until you've tried all other threads once
2012-09-17T21:24:47 < zyp> i.e. one round of the loop
2012-09-17T21:24:51 < Erlkoenig> but this way of sleep() needs switching to this thread often, and that means you have a bit more overhead than what you get with "my" idea, right?
2012-09-17T21:24:52 < Laurenceb_> oh i see
2012-09-17T21:25:20 < Erlkoenig> zyp: do your threads run in some sort of unprivileged state, so that's why you use svcall ?
2012-09-17T21:25:24 < zyp> Erlkoenig, yes, but it's the simplest approach, and works for me until I have time to actually write a scheduler
2012-09-17T21:25:57 < zyp> Erlkoenig, context switch has to happen in an ISR, that's why I use svcall
2012-09-17T21:26:53 < Erlkoenig> is your ISR handled by the scheduler's code, or by some "user code"?
2012-09-17T21:27:02 < Mazingaro> zyp: I did not understand then, how can the cpu jump back to thread?
2012-09-17T21:27:07 < zyp> Erlkoenig, what do you mean?
2012-09-17T21:27:11 < HTT-Bird> ESD protection questions: 1) what interfaces should I protect? everything? or should I focus on things that are exposed on a regular basis/are connected to cables and such?
2012-09-17T21:27:24 < zyp> Mazingaro, it's already stored on the stack before the ISR executes
2012-09-17T21:27:42 < zyp> and when you return to EXC_RETURN the cpu knows that it must read it back from the stack
2012-09-17T21:27:44 < Mazingaro> zyp: but this is the current, I need the new
2012-09-17T21:28:30 < zyp> then you have to put it into the pc field of the hardware saved part of the state frame
2012-09-17T21:28:41 < Mazingaro> zyp: ah ok
2012-09-17T21:28:46 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- again, see how my code is doing it
2012-09-17T21:29:18 < zyp> the function of the new thread goes into the hardware saved pc, the EXC_RETURN value goes into the software saved lr
2012-09-17T21:29:22 < Erlkoenig> zyp: uhm... assuming i want to use your scheduler system, and want to write code that gets executed when an interrupt occurs - would i put that in the ISR (which is than an "user code ISR")? Or rather into my own thread that the sheduler would jump into, when it notices that the interrupt has occured?
2012-09-17T21:30:00 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-17T21:30:10 < zyp> depends
2012-09-17T21:30:18 < Mazingaro> zyp: str sp, [ r0, #24 ]
2012-09-17T21:30:34 < zyp> as I don't have a proper scheduler, I can't jump to a specific thread, I can just trigger a change to the next thread
2012-09-17T21:30:35 < Mazingaro> zyp: str r0, [ sp, #24 ]
2012-09-17T21:31:00 < Erlkoenig> hmm... one day i have to code something like that, just for fun... ;-D
2012-09-17T21:31:26 <+Steffanx> Tomorrow will be that "one day", Erlkoenig
2012-09-17T21:31:28 < zyp> for my USB code I'm currently not using interrupts at all
2012-09-17T21:31:46 < zyp> I'm just polling the interrupt status flags every time it executes
2012-09-17T21:31:51 <+Steffanx> :D
2012-09-17T21:31:56 < Erlkoenig> some sort of limited continuations implementation in C/C++ would be nice, on top of that one could implement the kind of scheduler i described very elegantly
2012-09-17T21:32:12 < Erlkoenig> Steffanx: what, for you? :D
2012-09-17T21:32:21 <+Steffanx> For you of course
2012-09-17T21:32:40 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/threads/PPG_Demod.c#L59
2012-09-17T21:32:44 < zyp> in the future I'll have a Thread::wait_for_interrupt(USB) kind of function that makes a thread sleep until woken up by a specific interrupt
2012-09-17T21:33:07 < Erlkoenig> Steffanx: tomorrow is ungood
2012-09-17T21:33:23 < Erlkoenig> and first i'll finish my own super-awesome build system
2012-09-17T21:33:30 < zyp> so the threads don't have to manage the waiting part themselves
2012-09-17T21:33:33 < Laurenceb_> ^ how i do it
2012-09-17T21:33:37 <+Steffanx> Your crystal ball tells you tomorrow is "ungood"?
2012-09-17T21:33:40 < HTT-Bird> how concerned should I be with RTOSy things on a device with 32k of flash?
2012-09-17T21:33:58 <+Steffanx> Not concerned
2012-09-17T21:34:04 < zyp> HTT-Bird, depends on what you want to do
2012-09-17T21:34:07 <+Steffanx> Unless you only blink a led
2012-09-17T21:34:12 < Erlkoenig> zyp: yep, and the thread should consume no time at all until the interrupt occured
2012-09-17T21:34:21 < Laurenceb_> my code i linked is 29KB
2012-09-17T21:34:35 < zyp> my code is still <10k, with four threads or so
2012-09-17T21:34:37 < Laurenceb_> including usb and printf/scanf terminal thingy
2012-09-17T21:34:38 < Erlkoenig> Steffanx: i have to do things in RL... unfortunately
2012-09-17T21:34:51 < HTT-Bird> zyp: project's a programmable thermostat using a low-power graphical display
2012-09-17T21:34:54 < Erlkoenig> like going out of my door... and see sunlight, yuck
2012-09-17T21:35:10 < zyp> HTT-Bird, will you benefit from having several threads?
2012-09-17T21:36:04 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 248 seconds]
2012-09-17T21:36:38 < HTT-Bird> hrm.  well, there's "read temp -> calculate how much heat we need (PID) -> turn heat/cool on/off -> update display if needed"
2012-09-17T21:37:12 < HTT-Bird> "RTC overflow, so lets go keep track of that"
2012-09-17T21:37:23 < zyp> the main advantage about using threads is that you don't have to make explicit state machines
2012-09-17T21:38:12 < HTT-Bird> "somebody pushed a button, so lets go do the appropriate thing" (which seems to be the most state-machine-ish)
2012-09-17T21:38:44 < HTT-Bird> ...and the downside is its easy to get code that doesn't work and you can only scratch your head as to why, no?
2012-09-17T21:39:37 < zyp> also depends on what you are doing :)
2012-09-17T21:39:41 < Erlkoenig> zyp: in fact by using threads you use the CPU's nature of being a state machine the most elegant way, because basically the instruction pointer determines the current state you are in
2012-09-17T21:39:47 < Erlkoenig> right? :)
2012-09-17T21:40:19 < zyp> the current position in program and contents of stack, yes
2012-09-17T21:41:43 < Laurenceb_> Erlkoenig: ChibiOS
2012-09-17T21:41:58 < Laurenceb_> if you want to get actual work done
2012-09-17T21:42:17 < Erlkoenig> work? nononono
2012-09-17T21:42:43 < Laurenceb_> see i do this for a living
2012-09-17T21:42:46 < Laurenceb_> it has to work
2012-09-17T21:43:00 < Erlkoenig> seems like to easy :D
2012-09-17T21:43:18 < Laurenceb_> thats why i do it XD
2012-09-17T21:43:38 < Erlkoenig> well i'd like to at least try to write that kind of "scheduler" :D
2012-09-17T21:48:02 < Mazingaro> zyp: not working :(
2012-09-17T21:51:17 < Mazingaro> zyp: is PC located 24 byte + SP?
2012-09-17T21:53:24 < zyp> depends on which value of sp you refer to
2012-09-17T21:54:13 < zyp> but yes, at interrupt entry, stored pc is located at sp+24
2012-09-17T21:54:23 < Erlkoenig> hmm i loved how our prof explained the Stack using labeled cardboard boxes he stacked onto each other :3
2012-09-17T21:55:04 < zyp> and at the time you return to EXC_RETURN, the cpu will get pc from sp+24
2012-09-17T21:56:58 < Mazingaro> zyp, it jumps to 0x00
2012-09-17T21:57:20 < Mazingaro> str r0, [ sp, #24 ]
2012-09-17T21:57:32 < Mazingaro> r0 = function address
2012-09-17T21:59:18 < Mazingaro> :(
2012-09-17T21:59:29 < Mazingaro> str r0, [ sp, #24 ]
2012-09-17T21:59:32 < Mazingaro> cpsie i
2012-09-17T21:59:35 < Mazingaro> bx lr
2012-09-17T21:59:50 < Mazingaro> after bx lr, lr = 1, pc = 0
2012-09-17T22:02:39 < zyp> http://paste.jvnv.net/view/HyKzT <- I was bored, so I made some dumb example of how threads are making your code simpler :p
2012-09-17T22:05:25 < Mazingaro> oh tx
2012-09-17T22:05:39 < zyp> since HTT-Bird asked
2012-09-17T22:05:39 < zyp> :p
2012-09-17T22:06:44 < Mazingaro> zyp: shall I use bx lr as last instruction?
2012-09-17T22:07:04 < zyp> well, that's the return
2012-09-17T22:07:27 < zyp> so nothing after it will be executed :p
2012-09-17T22:07:34 < Mazingaro> lr = 0xfffffffd
2012-09-17T22:10:18 < Mazingaro> didn't work at all :(
2012-09-17T22:21:07 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-17T22:21:41 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-17T22:31:51 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-17T22:31:51 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-17T22:31:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-17T22:34:22 < Laurenceb_> zyp:  http://www.google.co.uk/url?sa=t&rct=j&q=stm32f4%20tim10%20trgo&source=web&cd=1&ved=0CCAQFjAA&url=http%3A%2F%2Fwww.st.com%2Finternet%2Fcom%2FTECHNICAL_RESOURCES%2FTECHNICAL_LITERATURE%2FAPPLICATION_NOTE%2FDM00042534.pdf&ei=b3pXUJa3CIaq0QXB2YGIBQ&usg=AFQjCNEMv1iKqrQIOgTgLzDb9qXmrXfg1g&cad=rja
2012-09-17T22:34:24 < Laurenceb_> eeek
2012-09-17T22:34:38 < Laurenceb_> found a description of tim10 TRGO on F4
2012-09-17T22:34:45 < Laurenceb_> its fixed as OC1
2012-09-17T22:34:50 < Laurenceb_> so thats lucky for me
2012-09-17T22:38:58 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-17T22:40:42 < Laurenceb_> well as long as they enable MSM...
2012-09-17T22:41:34 < Erlkoenig> http://en.wikipedia.org/wiki/MSM  see the first link.
2012-09-17T22:42:10 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-17T22:42:19 < Laurenceb_> no
2012-09-17T22:42:25 < TitanMKD> hi
2012-09-17T22:42:26 < Laurenceb_> i dont want that to be enabled
2012-09-17T22:42:50 < Laurenceb_> the gay bit
2012-09-17T22:43:23 < Erlkoenig> Laurenceb_->SEXCTRL->CR |= (1 << SCB_GAY);
2012-09-17T22:43:48 < Laurenceb_> noooo
2012-09-17T22:44:23 < Erlkoenig> thou can not withstand!
2012-09-17T22:44:59 < Erlkoenig> this->body.attractivenes.attract (&Laurenceb_);
2012-09-17T22:50:07 <+Steffanx> Erlkoenig, you have to remove the spaces
2012-09-17T22:50:28 < Erlkoenig> never >:o
2012-09-17T22:50:37 <+Steffanx> Otherwise Laurenceb_ doesn't understand
2012-09-17T22:50:53 < Erlkoenig> Laurenceb_ is not C-compatible? :D
2012-09-17T22:54:57 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-17T22:55:56 < zyp> wtf.
2012-09-17T22:56:34 < zyp> I leave for half an hour, and then this happen?
2012-09-17T22:56:41 < zyp> can't you guys behave yourselves?
2012-09-17T22:58:26 < Mazingaro> zyp: :(
2012-09-17T22:58:31 < Mazingaro> zyp: the evil exceptions
2012-09-17T22:59:38 < Erlkoenig> zyp: as if you wouldn't have joined cheeringly
2012-09-17T23:01:01 < zyp> Mazingaro, I'm not sure what you are doing wrong, is it possible for me to compile and run your code to test it and see what is going on?
2012-09-17T23:01:39 < Mazingaro> mmmh
2012-09-17T23:04:29 < Mazingaro> zyp: ah I understand now! http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0552a/Babefdjc.html
2012-09-17T23:05:09 < Mazingaro> after loading 0xfffffffd in PC the cpu uses the thread stack, so there's no need to load PC
2012-09-17T23:05:13 -!- piele [~Unknown@bakzeil.creativeserver.net] has joined ##stm32
2012-09-17T23:05:25 -!- piele [~Unknown@bakzeil.creativeserver.net] has left ##stm32 []
2012-09-17T23:05:33 < zyp> I tried to say that.
2012-09-17T23:05:55 < Mazingaro> zyp: yes, but after reload r0-r3, r12, lr, pc and xpsr from sp?
2012-09-17T23:06:01 < Thorn> me too
2012-09-17T23:06:14 < Mazingaro> zyp: or before ?
2012-09-17T23:06:47 < Mazingaro> so the return stack will hold those 8 registers?
2012-09-17T23:06:47 < zyp> it's done at the point when you jump to EXC_RETURN
2012-09-17T23:07:00 < zyp> yes
2012-09-17T23:07:17 < Mazingaro> yes I did understand, the question is, where are the 8 registers?
2012-09-17T23:07:28 < Mazingaro> shall I pop SP too?
2012-09-17T23:07:43 < zyp> huh?
2012-09-17T23:07:53 < zyp> the registers are on the process stack
2012-09-17T23:08:26 < Mazingaro> ah they're copied as per the effect of 0xfffffffd in pc
2012-09-17T23:08:33 < Mazingaro> good
2012-09-17T23:08:34 < Mazingaro> ok
2012-09-17T23:08:35 < zyp> yes
2012-09-17T23:09:03 < zyp> it clearly says that in the description in the page you just linked to
2012-09-17T23:09:04 < zyp> «Exception return gets state from the process stack.
2012-09-17T23:09:28 < zyp> state refers to those 8 registers
2012-09-17T23:09:29 < Mazingaro> ok
2012-09-17T23:14:04 -!- barthess [~barthess@178.154.85.217] has quit [Quit: Leaving.]
2012-09-17T23:17:31 < Erlkoenig> http://www.youtube.com/watch?feature=player_embedded&v=sUtS52lqL5w
2012-09-17T23:18:52 < karlp> cool, dirt cheap cp2102 dongles arrived finally.
2012-09-17T23:18:54 < karlp> about 45days from china
2012-09-17T23:18:58 <+Steffanx> Not bad
2012-09-17T23:19:15 < karlp> kinda overlooked the fact that they come with a usb A plug, not on the end of a cable.
2012-09-17T23:19:24 < karlp> but I have a hub on my desktop, so that works ok.
2012-09-17T23:19:31 <+Steffanx> I never received my LCDs. at least some chinese are great :D
2012-09-17T23:19:35 <+Steffanx> Got a refund :)
2012-09-17T23:19:44 < karlp> also, yay me, customs let it slip through
2012-09-17T23:19:48 <+Steffanx> ( the chinese with his nice emails )
2012-09-17T23:20:10 < karlp> it's even labelled as electronics and has a $10 value marked on it, which is actually higher than I paid even.
2012-09-17T23:20:18 < karlp> thanks dongs for the recommendations :)
2012-09-17T23:20:38 < Mazingaro> zyp: the RESTORE_CONTEXT now works!
2012-09-17T23:20:45 < Mazingaro> zyp: I wrote
2012-09-17T23:20:58 < Mazingaro> zyp: str r1, [ r0, #24 ]
2012-09-17T23:21:02 < karlp> now I can stop using one of teensy's purely as a usb-serial adapter
2012-09-17T23:21:14 <+Steffanx> cp2102 is better than ftdi ( or that crap prolific ) ?
2012-09-17T23:21:14 < Mazingaro> zyp: where r1 is the thread pc and r0 is the psp
2012-09-17T23:21:33 <+Steffanx> *ft232rl
2012-09-17T23:22:19 < karlp> dongs says cp2102 is as good as ft232, but way cheaper, and to avoid the prolific one.
2012-09-17T23:22:39 <+dekar> for usb-serial the ftdi stuff works really well imo
2012-09-17T23:22:50 < karlp> yeah, but way pricier
2012-09-17T23:22:51 <+dekar> prolific is indeed horrible
2012-09-17T23:23:10 < karlp> these were 2.98 on boards, delivered, including some female-female cabling
2012-09-17T23:23:54 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-17T23:24:06 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has joined ##stm32
2012-09-17T23:27:45 < Erlkoenig> does anoyone use Linux with ftdi? my linux kernel freezes when disconnecting an ftdi adapter
2012-09-17T23:28:02 <+dekar> always worked fine for me
2012-09-17T23:28:18 <+dekar> though my linux system is on debian stable
2012-09-17T23:28:24 <+dekar> pretty dated
2012-09-17T23:28:26 <+Steffanx> I can make os x ( at least i could ) and linux crash Erlkoenig :D
2012-09-17T23:28:31 <+Steffanx> Erlkoenig is using minicom?
2012-09-17T23:28:34 <+Steffanx> *uses
2012-09-17T23:28:47 <+dekar> Steffanx, is there any proper alternative to minicom?
2012-09-17T23:28:59 < Erlkoenig> Steffanx: no
2012-09-17T23:29:07 < Thorn> I haven't seen any crashes (ubuntu 10.04)
2012-09-17T23:29:07 < Erlkoenig> crashing linux is easy :D
2012-09-17T23:29:26 <+Steffanx> With an ftdi chip, yes
2012-09-17T23:29:28 < Erlkoenig> windows is much more stable nowadays
2012-09-17T23:29:33 < Erlkoenig> even without ftdi
2012-09-17T23:29:57 <+dekar> I am pretty sure you can forkbomb them all :D
2012-09-17T23:30:18 <+Steffanx> windows and fork?
2012-09-17T23:30:25 <+Steffanx> ( cygwin doesn't count )
2012-09-17T23:30:33 < Erlkoenig> well CreateProcess-bomb then
2012-09-17T23:30:54 < Erlkoenig> thats even more easy on windows.. a while(1) and malloc(1024*1024*1024*8); will crash any windows system ;)
2012-09-17T23:31:16 < Erlkoenig> while on linux you might have a chance to type in the allmighty "killall"
2012-09-17T23:31:18 <+Steffanx> Crash or make it slow?
2012-09-17T23:31:45 < Erlkoenig> make it so slow so that it won't to anything anymore... essentially the same as a crash :D
2012-09-17T23:33:51 < Erlkoenig> crashing linux is as easy as starting and stopping X, trying to run linux on new hardware, trying to use FTDI, opening a (device) file and unplugging the respective device, and well, just using linux normally
2012-09-17T23:34:45 < zyp> are you also a windows fanboy?
2012-09-17T23:34:45 <+Steffanx> without root privileges...
2012-09-17T23:35:06 < Erlkoenig> Steffanx: the latter two work without root priv's
2012-09-17T23:35:24 < zyp> Mazingaro, good, so everything is working now? :)
2012-09-17T23:35:27 <+Steffanx> "just using linux normally" not really
2012-09-17T23:35:37 < Erlkoenig> Even switching to a virtual terminal, then back to X causes crashes sometimes, without even needing to login
2012-09-17T23:35:50 <+Steffanx> You hw is crap
2012-09-17T23:35:58 <+Steffanx> :P
2012-09-17T23:36:02 < Erlkoenig> (22:34:45) zyp: are you also a windows fanboy? <- not really, but that doesn't mean i like to see linux crashes :S
2012-09-17T23:36:14 < Erlkoenig> Steffanx: windows doesn't have any problems with my hardware :D
2012-09-17T23:36:22 < BrainDamage> what? I restart x plenty times without having crashes
2012-09-17T23:36:23 < TitanMKD> Steffanx on linux loop with fork also kill all ;) but maybe on latest kernel it is detected and killed
2012-09-17T23:36:24 <+Steffanx> windows doesn't use you hw :P
2012-09-17T23:36:25 < zyp> then why are you exaggregating it?
2012-09-17T23:36:58 < Erlkoenig> BrainDamage: consider yourself lucky :D
2012-09-17T23:37:12 < Erlkoenig> Steffanx: how does it run games without all the fancy HW? :D
2012-09-17T23:37:25 < Erlkoenig> zyp: am i? i'm just reporting :S
2012-09-17T23:37:25 <+Steffanx> How ? Just with 1 fps
2012-09-17T23:37:59 < Erlkoenig> no, i get more than 1 fps
2012-09-17T23:38:48 < Erlkoenig> at least TF2, SupCom, minecraft work perfectly
2012-09-17T23:39:02 < zyp> oh, minecraft
2012-09-17T23:39:19 < zyp> my old workstation had a nice bug that minecraft could trigger
2012-09-17T23:39:43 < zyp> resizing the window sometimes froze the entire graphical framework
2012-09-17T23:40:02 <+Steffanx> I have a game that crashes my poor os x machine :P
2012-09-17T23:40:32 < zyp> I could still ssh in to the machine, but it's not as easy to just restart like X is :p
2012-09-17T23:41:17 < zyp> likely a bug in the old graphics driver
2012-09-17T23:41:42 <+Steffanx> linux and graphics isn't a good combination anyway
2012-09-17T23:41:54 < zyp> I'm not talking about linux
2012-09-17T23:41:57 < Erlkoenig> seems like some graphics drivers are a mere heap of bugs held together by a few working LoC
2012-09-17T23:42:56 < zyp> isn't that why people are complaining about nvidia drivers being closed source?
2012-09-17T23:43:06 < zyp> not being able to touch it means not being able to fix it
2012-09-17T23:43:16 < Erlkoenig> that applies to AMD drivers just as well
2012-09-17T23:43:35 <+Steffanx> The nvidia drivers i use on my (older) notebook work prefectly though
2012-09-17T23:43:42 < Erlkoenig> i tried to set up some server using debian stable on new PC hardware, and it kept crashing every 1-3 days... no error messages, just "poweroff"
2012-09-17T23:43:44 <+Steffanx> The only game i play works better on linux than on windows :D
2012-09-17T23:43:46 < zyp> that said, I can't remember having any problems when I used linux on computers with nvidia hardware
2012-09-17T23:43:52 < zyp> but that's a few years ago
2012-09-17T23:43:59 <+Steffanx> And yes it use the gpu
2012-09-17T23:44:37 < BrainDamage> what game is it?
2012-09-17T23:44:42 < Erlkoenig> now i managed to start it via UEFI, and it didn't crash for 5 days now, i am curious how long it will work: http://minecraft.niklas.2g2s.de/
2012-09-17T23:44:55 < Laurenceb_> oh noes minecraft
2012-09-17T23:45:02 < Erlkoenig> minecraft <3
2012-09-17T23:45:02 < Laurenceb_> im not _that_ nerdy
2012-09-17T23:45:05 <+Steffanx> A wolfenstein ET mod, BrainDamage
2012-09-17T23:45:12 <+Steffanx> VERY OLD
2012-09-17T23:45:23 <+Steffanx> I'm one of the few who still play it :)
2012-09-17T23:45:56 <+Steffanx> At busy evenings ~30 plays active :P
2012-09-17T23:46:01 < Erlkoenig> oh i know somebody who plays ET too ;-)
2012-09-17T23:46:02 <+Steffanx> *players
2012-09-17T23:46:46 < zyp> IMO the problem with linux is not the graphics drivers, it's the graphic environments
2012-09-17T23:46:57 < zyp> they are all just different degrees of horrible :p
2012-09-17T23:47:33 < Laurenceb_> gnome is ok
2012-09-17T23:47:37 -!- metaxa [~metaxa@us.members.bombshellz.net] has quit [Ping timeout: 260 seconds]
2012-09-17T23:47:54 < Erlkoenig> gnome2 was nice, gnome3 is a p*le of sh*t
2012-09-17T23:47:55 < zyp> sure, but it's still horribe
2012-09-17T23:48:03 < BrainDamage> X11 itself is a complete mess
2012-09-17T23:48:21 < zyp> BrainDamage, but it works, so that part is ok
2012-09-17T23:48:46 < BrainDamage> yes, but you're doing everything on top of a stupidly designed, working layer
2012-09-17T23:48:58 < BrainDamage> I hope that wayland will bring some sanity in that regard
2012-09-17T23:49:17 < BrainDamage> the way decorations are done for instance ..
2012-09-17T23:49:29 <+Steffanx> "but it works" sure YOU typed that zyp ?
2012-09-17T23:49:50 < zyp> sure
2012-09-17T23:50:08 < zyp> I'm using gnome on the linux computer I have at work
2012-09-17T23:50:35 < Erlkoenig> wat? can you stand it? :D
2012-09-17T23:50:44 < Erlkoenig> you can't even move the f*cking panel
2012-09-17T23:50:49 < zyp> huh?
2012-09-17T23:50:55 < zyp> which panel?
2012-09-17T23:51:35 < Erlkoenig> the panel at the top
2012-09-17T23:51:51 < zyp> that might be, I haven't tried
2012-09-17T23:51:56 < Erlkoenig> if you have a low Y-Resolution but a high X-Res ("widescreen") it's comfortable to have a vertical panel at the left or right
2012-09-17T23:52:16 < Laurenceb_> luckly i have no suck panels
2012-09-17T23:52:18 < zyp> I disagree
2012-09-17T23:52:23 < Erlkoenig> also there's no easy way to define launchers
2012-09-17T23:52:23 < Laurenceb_> *such
2012-09-17T23:53:01 < zyp> I can't stand the unity shit because of the panel on the left side
2012-09-17T23:53:14 < zyp> it's horribly dumb to dictate that it must be there
2012-09-17T23:53:18 < Laurenceb_> lol it is very annoying
2012-09-17T23:53:49 < BrainDamage> last time I tried it was also a monolithical piece with the top bar
2012-09-17T23:53:53 < BrainDamage> so you got all or nothing
2012-09-17T23:54:03 < zyp> well, I'm not using unity
2012-09-17T23:54:22 < zyp> I'm using whatever ubuntu calls the «classic» environment
2012-09-17T23:54:31 < Erlkoenig> even in win2000 you can move your panel, gnome2 could do that, but gnome3... nope
2012-09-17T23:54:47 < zyp> isn't classic gnome2?
2012-09-17T23:55:03 < zyp> because I also have gnome3 installed
2012-09-17T23:55:14 < Erlkoenig> gnome3 fallback is a *bit* usable, but you still can't configure global shortcuts - i'd like to have Super+F9 to launch the calculator - nope, impossible
2012-09-17T23:55:16 < zyp> which I also didn't like for some reason, I forgot why
2012-09-17T23:56:02 < zyp> but see, it's just like I said, we agree that it's horrible :p
2012-09-17T23:56:51 <+Steffanx> I never ever ever tried gnome3 :D
2012-09-17T23:58:20 < Erlkoenig> in gnome2 you could supereasily define your own shortcuts for everything... gnome3 allows you a few standard-ones, but not even a terminal-shortcut
2012-09-17T23:58:23 < zyp> I used gnome3 on the work computer I used before I started as a contractor
2012-09-17T23:58:59 < zyp> but I didn't use ubuntu on that one
2012-09-17T23:59:10 < Mazingaro> zyp: context switching now works, but I've got a problem with psp, the first push in thread crashes the cpu, lol!
2012-09-17T23:59:35 < zyp> :)
2012-09-17T23:59:37 < Erlkoenig> SP not initialized at that point? :D
2012-09-17T23:59:50 < zyp> it should be, it's using sp to return to the thread
--- Day changed Tue Sep 18 2012
2012-09-18T00:00:37 < Mazingaro> yes, but PSP it is loaded with the address of the last element of an array of uint8_t - 4
2012-09-18T00:01:06 < zyp> huh?
2012-09-18T00:01:20 < zyp> that doesn't make sense?
2012-09-18T00:01:30 < Mazingaro> yes
2012-09-18T00:01:43 < Mazingaro> is stack decrementing in arm?
2012-09-18T00:01:48 < zyp> yes
2012-09-18T00:02:11 < Mazingaro> so I reserved an array of uint8_t per thread
2012-09-18T00:02:14 < zyp> so an empty stack would have the stack pointer pointing to the address just after the end
2012-09-18T00:02:35 < Mazingaro> should it aligned at 8 bytes?
2012-09-18T00:02:45 < zyp> which for uint8_t stack[1024] should be stack + 1024
2012-09-18T00:02:55 < Mazingaro> yes
2012-09-18T00:02:58 < Mazingaro> I did so
2012-09-18T00:03:11 < Mazingaro> to be more precise I used a stack structure
2012-09-18T00:03:13 < Mazingaro> where
2012-09-18T00:03:29 < Mazingaro> stack[1024] and TCB are two members
2012-09-18T00:03:37 < Mazingaro> tcb member follows stacks
2012-09-18T00:04:05 < Mazingaro> so PSP = (uint32_t) &Task.m_tTCB
2012-09-18T00:05:02 < zyp> yes
2012-09-18T00:05:16 < Mazingaro> well it hangs on the first thread push
2012-09-18T00:05:26 < zyp> but wait, where do you do this?
2012-09-18T00:05:46 < Mazingaro> in the setup function
2012-09-18T00:05:49 < zyp> because you shouldn't
2012-09-18T00:06:02 < zyp> I mean
2012-09-18T00:06:04 < Mazingaro> PSP is stored every context switch
2012-09-18T00:06:08 < zyp> yes
2012-09-18T00:06:20 < Mazingaro> so the stack could grow for every thread
2012-09-18T00:06:34 < Mazingaro> depending on the thread itself
2012-09-18T00:06:38 < zyp> before the thread is started, you fill the stack with those eight hardware saved registers, plus some software saved registers
2012-09-18T00:06:50 < zyp> so you have to subtract that from the initial sp value
2012-09-18T00:07:08 < Mazingaro> ?
2012-09-18T00:07:17 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- refer to my code again
2012-09-18T00:07:24 < Mazingaro> I modified only PC
2012-09-18T00:07:44 < zyp> sp = (int_frame_t*)((uint8_t*)stack + stack_size - sizeof(int_frame_t)); <- that is how I set the initial sp
2012-09-18T00:08:29 < Mazingaro> I forgot to put the apsr too :)
2012-09-18T00:19:00 -!- bgamari_ is now known as bgamari
2012-09-18T00:35:24 < Laurenceb_> zyp do you live on here?
2012-09-18T00:39:31 < zyp> on where?
2012-09-18T00:40:44 < Erlkoenig> " The researcher also found that the ARM Cortex-A9 easily beats the  NVIDIA/AMD GPUs and latest Intel/AMD workstation CPUs in  performance-per-Watt efficiency." hum
2012-09-18T00:40:51 < zyp> I usually keep an eye on irc when I'm sitting at a computer, if that's what you're asking about :p
2012-09-18T00:48:02 < Laurenceb_> really
2012-09-18T00:48:21 < Laurenceb_> i thought the new 22nm intel stuff with tri gate was better
2012-09-18T00:48:28 -!- ReggieUK is now known as notReg
2012-09-18T00:49:56 < Laurenceb_> the apple A6 looks interesting
2012-09-18T00:50:04 < Laurenceb_> custom armv7 apparently
2012-09-18T00:50:06 < Erlkoenig> Apple :o
2012-09-18T00:50:15 < Erlkoenig> geh wek
2012-09-18T00:51:17 < Laurenceb_> apparently its been benchmarked as fast as quad core A9
2012-09-18T00:59:43 < Erlkoenig> what's so sad with apple that they actually do technology stuff... if all that apple-fetishism would just be about jewellry or fashion or something, well, that would be it... But the thought that there are actually programmers and electric engineers are working there...
2012-09-18T01:05:49 < zyp> what's sad about that?
2012-09-18T01:06:13 < zyp> Laurenceb_, imagine an ARM core with intel's process then :p
2012-09-18T01:07:10 < Erlkoenig> zyp: well that they celebrate all their livestyle and design fetish stuff around sacred technology :D
2012-09-18T01:08:43 < zyp> you should probably stop hating :p
2012-09-18T01:09:14 < zyp> marketing bullshit and dumb customers are marketing bullshit and dumb customers either way :p
2012-09-18T01:10:05 < Erlkoenig> well their walled garden style also affects the technology as such
2012-09-18T01:10:31 < Laurenceb_> imagine an apple A6 made by intel
2012-09-18T01:11:22 < zyp> or a normal cortex-A15 for that matter
2012-09-18T01:11:23 < qyx_> there's no such thing
2012-09-18T01:11:30 < qyx_> imagination has also its constrains
2012-09-18T01:12:31 < Laurenceb_> supposedly A15 has dual issue and deep pipeline
2012-09-18T01:12:43 < Laurenceb_> apple A6 is treble issue or something
2012-09-18T01:12:51 < zyp> http://www.hardkernel.com/renewal_2011/products/prdt_info.php?g_code=G133999328931 <- by the way, I came across this today
2012-09-18T01:13:05 < Erlkoenig> why "apple a6"? did apple themselves design that one?
2012-09-18T01:13:24 < Laurenceb_> yes
2012-09-18T01:13:24 < zyp> Erlkoenig, the name is unrelated
2012-09-18T01:13:34 < Laurenceb_> its custom supposedly
2012-09-18T01:14:01 < zyp> Apple A6 is a SoC, containing a custom cpu which should be about on par with cortex-a15 in performance
2012-09-18T01:14:48 < Laurenceb_> http://www.geeky-gadgets.com/iphone-5-features-custom-apple-a6-processor-not-a15-or-a9-based-core-17-09-2012/
2012-09-18T01:14:58 < Erlkoenig> zyp: the question is wether it's als on par regarding price ;)
2012-09-18T01:15:11 < Erlkoenig> and why do all chinese webpages have a weird font for english? :D
2012-09-18T01:15:16 < Laurenceb_> buy phone, desolder iwht hot air
2012-09-18T01:15:20 < zyp> irrelevant, you're not going to get it without buying an iPhone :p
2012-09-18T01:15:36 < zyp> anyway, that board I linked
2012-09-18T01:15:49 < zyp> it's kind of fun to compare it to rpi
2012-09-18T01:16:17 < qyx_> price also
2012-09-18T01:16:21 < zyp> it has around 16 times the performance, for less than four times the price
2012-09-18T01:16:24 < Erlkoenig> at least the whole iphone is not on par with phones which have similar capabilities ;)
2012-09-18T01:17:16 < Erlkoenig> zyp: a PC has like 100 times the cost of an ATmega8 and like 10000x the performance? performance vs. price seems to be logarithmic or something
2012-09-18T01:17:21 < zyp> it says something about how much «bang for the buck» you really get when you buy an rpi
2012-09-18T01:18:21 < zyp> Erlkoenig, do you see people building clusters of avr chips? I believe I saw a cluster with 64 rpis the other day
2012-09-18T01:18:27 < HTT-Bird> zyp: is Intel still an ARM licensee?
2012-09-18T01:18:34 < Laurenceb_> http://en.wikipedia.org/wiki/ARM_Cortex-A15_MPCore
2012-09-18T01:18:39 < zyp> HTT-Bird, don't ask me
2012-09-18T01:18:46 < Laurenceb_> out-of-order speculative issue 3-way superscalar execution pipeline
2012-09-18T01:18:47 < Laurenceb_> wut
2012-09-18T01:18:48 < HTT-Bird> I know they were at one point
2012-09-18T01:18:52 < Laurenceb_> so whats the big deal
2012-09-18T01:18:54 < Erlkoenig> wah... i don't see why building clusters out of something that's below the performance of PC-hardware
2012-09-18T01:19:28 < HTT-Bird> Erlkoenig: some workloads might be better with many small machines than one big one?
2012-09-18T01:20:09 < Erlkoenig> what? if the big one has the combined power of all the small ones... no?!
2012-09-18T01:20:16 < Laurenceb_> i mean about apple
2012-09-18T01:20:19 < zyp> http://paritynews.com/hardware/item/302-engineers-build-supercomputer-using-raspberry-pi-lego
2012-09-18T01:20:23 < Erlkoenig> because you don't need all the crazy synchronization stuff
2012-09-18T01:20:27 < Laurenceb_> if A15 has triple issue
2012-09-18T01:21:03 < Erlkoenig> lego. wtf.
2012-09-18T01:24:01 <+Steffanx> Nothing wrong with LEGO!
2012-09-18T01:25:31 <+Steffanx> Erlkoenig never played with LEGO?
2012-09-18T01:25:49 <+Steffanx> *LEGO?
2012-09-18T01:25:59 < Erlkoenig> of course i did :D
2012-09-18T01:26:06 <+Steffanx> OK, good
2012-09-18T01:26:13 < Erlkoenig> but using lego for a supercomputer is still wtf-worthy :D
2012-09-18T01:26:41 < Erlkoenig> what's really really WTF (NSFW!!!): http://www.whokilledbambi.co.uk/public/2009/01/fetish_hightech1.jpg    http://www.whokilledbambi.co.uk/public/2009/01/fetish_hightech5.jpg
2012-09-18T01:27:27 < Erlkoenig> and http://www.whokilledbambi.co.uk/public/2008/12/lego-sm.jpg
2012-09-18T01:27:32 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-18T01:27:37 <+Steffanx> heh
2012-09-18T01:27:39 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-18T01:28:54 < HTT-Bird> if there's any process I'd want to see an ARM in
2012-09-18T01:29:08 < HTT-Bird> it'd be TI's upcoming 130-nm ultra-low-power process
2012-09-18T01:30:26 < HTT-Bird> Cortex-M0+ in that, with a good 32/64/128KB of on-chip FRAM? I'll take that any day of the week
2012-09-18T01:32:23 < Laurenceb_> Erlkoenig likes fetish stuff?!
2012-09-18T01:32:35 < Erlkoenig> only technology fetish. :D
2012-09-18T01:32:36 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-18T01:32:36 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-232-102.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-18T01:33:06 < Erlkoenig> HTT-Bird: wasn't the problem about FRAM that it has limited read AND write cycles?
2012-09-18T01:33:10 < zyp> HTT-Bird, or M4F, considering that it'll spend less power on the same task
2012-09-18T01:33:18 < HTT-Bird> zyp: hehe. :)
2012-09-18T01:33:23 < zyp> it would be perfect combined with FRAM
2012-09-18T01:33:33 < zyp> since it could wake up, do something and then fall asleep
2012-09-18T01:33:41 < HTT-Bird> Erlkoenig: you are correct in that you have the read-and-refresh issue
2012-09-18T01:33:48 < zyp> full sleep, since everything is kept in FRAM anyway
2012-09-18T01:33:55 < HTT-Bird> but with the sheer amount of write cycles available to you
2012-09-18T01:34:01 < HTT-Bird> I'm really not sure that that's a problem?
2012-09-18T01:34:10 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-18T01:34:31 < Erlkoenig> HTT-Bird: well if you're going to write to the (F)RAM like in every 2nd cycle, your individual cells will wear out in minutes
2012-09-18T01:34:38 < Erlkoenig> or even seconds
2012-09-18T01:34:48 < HTT-Bird> Erlkoenig: you seriously underestimate just how many cycles FRAMs can withstand
2012-09-18T01:35:06 < Erlkoenig> or read, which destroys the data so it has to be re-written
2012-09-18T01:35:20 < Erlkoenig> HTT-Bird: hm, i remembered something like the cycles of Flash... but i might be wrong
2012-09-18T01:35:30 < zyp> wikipedia claims > 10^16 cycles
2012-09-18T01:35:37 < Erlkoenig> oh hm
2012-09-18T01:35:53 < HTT-Bird> yeah, 10^12 and upwards are the expected numbers
2012-09-18T01:36:24 < HTT-Bird> makes me want to get a Flash Destroyer from DP and mod it to try to wreck a FeRAM
2012-09-18T01:38:33 < zyp> 10^16 means a bit over three years at 100 MHz continuously
2012-09-18T01:38:40 < zyp> and is that per bit?
2012-09-18T01:39:07 < HTT-Bird> yep
2012-09-18T01:39:50 < zyp> if that figure holds true, you're not going to wear it out over the lifetime of a normal application :p
2012-09-18T01:40:07 < Erlkoenig> as long as the registers are still normals flipflops... :D
2012-09-18T01:52:55 -!- DaKu is now known as daku
2012-09-18T01:53:31 < dongs> lol dongs
2012-09-18T01:53:59 -!- notReg is now known as ReggieUK
2012-09-18T02:04:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-18T02:17:59 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-09-18T02:23:30 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-18T02:36:48 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-18T03:12:07 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has quit [Ping timeout: 240 seconds]
2012-09-18T03:13:52 < Erlkoenig> "Contrary to other markets the mobile devices market is basically computer architecture agnostic." ... as if architectures were religions :D
2012-09-18T03:14:13 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-18T03:14:23 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-18T03:29:17 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-18T03:30:12 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-18T03:40:48 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 268 seconds]
2012-09-18T03:54:15 < emeb> they're no? :O
2012-09-18T03:59:54 < Erlkoenig> harhar...
2012-09-18T04:03:03 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-18T04:18:45 < dongs> sup t rolls
2012-09-18T05:14:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-18T05:29:02 < dongs> cool
2012-09-18T05:29:04 < dongs> WS2811 arrived
2012-09-18T05:30:13 < gnomad> someone else was talking about that
2012-09-18T05:30:18 < gnomad> aparently a clockless protocol
2012-09-18T05:31:21 < gnomad> you know, I am really torn between saying Apple is doing so well in the mobile marketplace because they are  the only ones with a well though-out platform....
2012-09-18T05:31:47 < gnomad> ... or because they really are just a logo and a status symbol without any other meaning.
2012-09-18T05:33:16 < dongs> the latter.
2012-09-18T05:33:19 < dongs> yeah, clockless
2012-09-18T05:33:35 < dongs> 400khz? period pwm w/zero/one as differnet pulse widths
2012-09-18T05:33:51 < dongs> gonna try to drive it in a bit, need to assemble the driver pcb
2012-09-18T05:35:16 < dongs> got a reel of 5050 samsung rbg leds
2012-09-18T05:35:24 < dongs> or wait not samsung.. itswell or someshit.
2012-09-18T05:36:27 < Erlkoenig> (04:31:21) gnomad: you know, I am really torn between saying Apple is doing so well in the mobile marketplace because they are  the only ones with a well though-out platform.... <-- well thought-out or just so closed you don't see the not-so-nice parts?
2012-09-18T05:36:48 < dongs> lol who cares about apple
2012-09-18T05:36:52 < dongs> yesterdays tech at inflated prices
2012-09-18T05:37:11 < dongs> http://theoatmeal.com/comics/apple sums it up perfectly
2012-09-18T05:37:16 < Erlkoenig> hmm, people who care about dongs, apparantly
2012-09-18T05:37:34 < gnomad> sometimes a black box isn't a bad thing...
2012-09-18T05:38:17 < gnomad> one of my interests is in design, particularly user interface design.
2012-09-18T05:38:23 < Erlkoenig> as long as the black box works :D ... and nobody can find the privacy problems caused by that :P
2012-09-18T05:38:28 < gnomad> Ihave been doing UI work for 20+ years now.
2012-09-18T05:38:32 < dongs> gnomad: apple failed in UI years ago.
2012-09-18T05:38:41 < gnomad> I disagree
2012-09-18T05:38:43 < dongs> infact, they never succeeded ther.
2012-09-18T05:38:50 < gnomad> apple has fabulous design at all levels.
2012-09-18T05:38:56 < dongs> haha... ookay.
2012-09-18T05:39:19 < Erlkoenig> (04:37:10) dongs: http://theoatmeal.com/comics/apple sums it up perfectly <-- no it misses tha part about lifestyle and coolness
2012-09-18T05:39:31 < dongs> i have no idea since i dont use their mobile shit, but their desktop UI is *horrible*
2012-09-18T05:39:34 < dongs> absolutely fucking horrible.
2012-09-18T05:39:39 < gnomad> my personal opinion -- and I must stress opinion -- is that apple has succeeded because they have the best user interface, whether people actively recognize it for that or not.
2012-09-18T05:39:53 < gnomad> I, on the other hand disagree again.
2012-09-18T05:40:00 < Erlkoenig> a gurl once said to me "apple products perfect for us, because their products are so pretty" ... my face was like wtf???
2012-09-18T05:40:26 < gnomad> I was using X/windows under Unix since before MS-windows came out.
2012-09-18T05:40:36 < Erlkoenig> and she absolutely didn't care about hardware performance or price or something - just the "design" - baffled me a bit
2012-09-18T05:41:05 < dongs>  < gnomad> I was using X/windows under Unix since before MS-windows came out.
2012-09-18T05:41:07 < dongs> thats your problem.
2012-09-18T05:41:12 < dongs> you have no idea what a proper UI is
2012-09-18T05:41:16 < gnomad> and although X/windows has plenty of flaws, I have always had to laugh at every version of Windows that has come out.
2012-09-18T05:41:22 < Erlkoenig> i prefer the Linux bash user interface about the windows areo interface, because it actually does what you want, rather than showing neat graphics effects
2012-09-18T05:41:22 < dongs> so your opinion is >fucking invalid<
2012-09-18T05:42:04 < gnomad> So really, I hated MS-Wndows and X/windows equally.  even though I used then equally.
2012-09-18T05:42:23 < Erlkoenig> ... at least Microsoft does't do censorship, they don't care about what perverted pornography you have on your PC
2012-09-18T05:42:36 < dongs> xwindows guys like motif and praise GTK
2012-09-18T05:42:41 < gnomad> Six year ago I was rolling in dough, and I needed a tax write-off, so I dropped $2500 on a 17" macbook pro without even thinking about it.
2012-09-18T05:42:43 < dongs> so whatever they think is irrelevant
2012-09-18T05:42:53 < gnomad> I never planned to use OSX when I bought it.
2012-09-18T05:43:15 < gnomad> I was going to dual boot windows and linux like the rest of my machines
2012-09-18T05:43:26 < Erlkoenig> gnomad: why didn't you just buy a 1200$ notebook with the same hardware specs? :D
2012-09-18T05:43:38 < gnomad> but I figured, WTF, I'll play with the thing before I wipe.
2012-09-18T05:44:07 < gnomad> and I just fell in love.
2012-09-18T05:44:11 < gnomad> srsly.
2012-09-18T05:44:29 < gnomad> I am still using that *same* machine today.  Six+ years later.
2012-09-18T05:44:51 < Erlkoenig> successfully became introduced to apples style and coolnes dictatorship? _D
2012-09-18T05:45:27 < gnomad> like I said, I am honestly not sure why Apple has become so opopular.
2012-09-18T05:45:49 < Erlkoenig> because of lifestyle and coolnes of having an ifone
2012-09-18T05:45:59 < gnomad> All I can say is that it is the only *sane* desktop for the only *sane* operating system.
2012-09-18T05:46:21 < Erlkoenig> the only sane OS?! dangerous statement
2012-09-18T05:46:41 < gnomad> And seriously, I wretch every time I am forced to work with Windows or Linux.
2012-09-18T05:46:57 < Erlkoenig> might just as well be your personal flaw :D
2012-09-18T05:47:29 < Erlkoenig> at least windows and linux isn't censored
2012-09-18T05:47:36 < Erlkoenig> *aren't
2012-09-18T05:49:55 < gnomad> well, I guess this is one area where I am staunchly capitalist.
2012-09-18T05:50:16 < gnomad> in that you choose with your pocketbook.
2012-09-18T05:50:54 < Erlkoenig> if you are righ enough you buy iStuff to show that you are?
2012-09-18T05:51:16 < gnomad> I suppose some people might do that
2012-09-18T05:51:51 < gnomad> I bought my mb pro because it tickled my design geek.
2012-09-18T05:51:57 < Erlkoenig> i suppose that's the main reason people by istuff
2012-09-18T05:52:13 < gnomad> Mac's were only beginning to become "hip" six year ago
2012-09-18T05:52:17 < Erlkoenig> to show others "hey i have an ifone, i am cool, i have style, and - i have money."
2012-09-18T05:52:45 < gnomad> when I got mine, that certainly was not an attitude.
2012-09-18T05:52:58 < Erlkoenig> that excuses you a bit
2012-09-18T05:53:15 < gnomad> back then it was like, "oh, you have an Apple, you must be an artist."
2012-09-18T05:53:31 < Erlkoenig> wah, i don't want to be told i'm an artist :o
2012-09-18T05:53:35 < gnomad> and then I try to explain....
2012-09-18T05:53:42 < gnomad> no. I am a Unix geek.
2012-09-18T05:53:49 < gnomad> I have been using Unix for 20 years.
2012-09-18T05:53:54 < Erlkoenig> nor do i want to be offended "creative"
2012-09-18T05:54:09 < Erlkoenig> so how much do you use the console stuff in Mac OS?
2012-09-18T05:54:10 < gnomad> Apple just happens to have the most *usable* Unix desktop on the market
2012-09-18T05:54:21 < gnomad> I spend most of my time in Emacs
2012-09-18T05:54:24 < Erlkoenig> i.e. how much does the Unix-part of Mac OS matter?
2012-09-18T05:54:40 < Erlkoenig> so you could as well use linux, which as well has Emacs
2012-09-18T05:54:41 < gnomad> I do all of my MCU development with the gnu toolchain
2012-09-18T05:54:45 < Erlkoenig> or windows, probably
2012-09-18T05:54:47 < gnomad> Yes, I could.
2012-09-18T05:54:55 < gnomad> but all of the linux desktops suck balls
2012-09-18T05:55:05 < Erlkoenig> did you try them all?
2012-09-18T05:55:08 < gnomad> yes
2012-09-18T05:55:25 < gnomad> I am one of those people who *likes* consistency.
2012-09-18T05:55:36 < Erlkoenig> about the UI? doesnt matter :D
2012-09-18T05:55:50 < gnomad> any desktop that can be "themed" is fundamentally flawed.
2012-09-18T05:55:58 < Erlkoenig> wat :D :D
2012-09-18T05:56:05 < gnomad> (ok, so there is some hyperbole in that statement...)
2012-09-18T05:56:12 < Erlkoenig> that's the best statement i've ever heard :D :D lololol
2012-09-18T05:56:46 < gnomad> I used to use the ATI graphics driver under windows as an example.
2012-09-18T05:56:50 < Erlkoenig> so you say that the style apple invented is the perfect one for everyone's taste?!
2012-09-18T05:57:16 < gnomad> the thing was this gaudy shaped window that was practically impossible to figure out
2012-09-18T05:57:34 < gnomad> it didn't function like *anything* else.
2012-09-18T05:57:49 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-09-18T05:58:02 < Erlkoenig> yep, but since you have to open that window like once per windows installation, it doesn't matter so much
2012-09-18T05:58:17 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-18T05:59:22 < gnomad> heh, no, I had to use this awful piece of crap all the time.
2012-09-18T05:59:28 < gnomad> and it crashed regularly
2012-09-18T05:59:49 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-18T05:59:51 < Erlkoenig> funny. but that's not microsoft's fault...
2012-09-18T06:00:22 < gnomad> Rickta59: you just stepped into another os war...
2012-09-18T06:00:43 < Erlkoenig> also, i'd say that catalyst is not so bad, i didn't have any problems using it... just perhaps looks a bit weird
2012-09-18T06:00:56 < gnomad> Rickta59 knows I tend to be rather opinionated.
2012-09-18T06:01:06 < gnomad> but that's just an example.
2012-09-18T06:01:26 < gnomad> one of the things that I like about OSX is that every application functions the same way.
2012-09-18T06:01:38 < gnomad> the same keys do the same things in every app.
2012-09-18T06:01:41 < Erlkoenig> the 5 applications that exist for OSX? :D
2012-09-18T06:01:44 < gnomad> ok... not every...
2012-09-18T06:02:02 < gnomad> Believe it or not, I rarely leave OSX
2012-09-18T06:02:03 < Erlkoenig> whatever, i has a brains, i can learns keys
2012-09-18T06:02:14 < Erlkoenig> also they can be configured
2012-09-18T06:02:35 < gnomad> I ran vmware a fair bit the first few years I had my mac.
2012-09-18T06:02:51 < Erlkoenig> is it really impossible to switch themes on OSX?
2012-09-18T06:02:54 < gnomad> mostly for obscure windows apps
2012-09-18T06:03:04 < Erlkoenig> like if i prefer dark themes with white font?
2012-09-18T06:03:20 < gnomad> yes, you can change colors... and a lot of things actually.
2012-09-18T06:03:25 < Erlkoenig> at least
2012-09-18T06:03:53 < gnomad> but in terms of apps...
2012-09-18T06:04:21 < gnomad> just about anything that can be compiled for Linux can be compiled for OSX.
2012-09-18T06:05:06 < Erlkoenig> argh, another annoying invention of apple... why do they call everything "app", while there are "applications", "daemons", "programs" etc.
2012-09-18T06:05:24 < Erlkoenig> ... and infect everyone to always say "app"
2012-09-18T06:05:34 < gnomad> uh, I've been using the term "app" for 20 years
2012-09-18T06:05:58 < Erlkoenig> that fits perfectly into the style-design-hipster-image
2012-09-18T06:06:09 < gnomad> if you say so
2012-09-18T06:06:16 < Erlkoenig> "application" is lame, while "app" is so cool?
2012-09-18T06:06:28 < gnomad> no, app is just an abbreviation for application
2012-09-18T06:06:44 < gnomad> and it has been in the unix world for over 20 years
2012-09-18T06:06:51 < Erlkoenig> not on linux, as it seems
2012-09-18T06:07:05 < gnomad> an application, or app for short, is any program that a user interacts with directly.
2012-09-18T06:07:11 < Erlkoenig> apples advertising seems to strongly use that term because its so cool
2012-09-18T06:07:26 < gnomad> program is a superset that includes applications and daemons
2012-09-18T06:07:54 < gnomad> scripts can be either applications or daemons
2012-09-18T06:08:02 < Erlkoenig> or just scripts ...
2012-09-18T06:08:02 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Read error: Operation timed out]
2012-09-18T06:08:06 < gnomad> all scripts are programs
2012-09-18T06:08:24 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-18T06:08:25 < Erlkoenig> oh rly
2012-09-18T06:08:30 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-18T06:09:09 < Erlkoenig> i'm going to bed now...
2012-09-18T06:09:11 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-115.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-18T06:09:23 < gnomad> Applogies for those here for STM32.... I didn't what windows I was in when this started.
2012-09-18T06:09:47 < gnomad> er, I didn't realie what window I was in when I responded.
2012-09-18T06:11:35 < dongs> its ok, now everyone knows youre a fag.
2012-09-18T06:12:32 < gnomad> btw, re: The Oatmeal Apple commentary.
2012-09-18T06:12:46 < gnomad> I've really only ever purchased two Apple products.
2012-09-18T06:12:54 < gnomad> 1.  An Apple 2 in 1979
2012-09-18T06:13:27 < gnomad> 2. A MB Pro in 1996.  The same one I am typing on at this very moment.
2012-09-18T06:15:58 < dongs> http://i.imgur.com/yIexr.jpg
2012-09-18T06:16:08 < dongs> http://i.imgur.com/yIexr.jpg time to figure out how to make ws2811 work
2012-09-18T06:26:03 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-18T06:30:51 < dongs> hm
2012-09-18T06:31:14 < dongs> pwm period of 2.5us,, with 0.1us resolution, what should I run thet timer at?
2012-09-18T06:31:59 < dongs> dont think 10mhz is a possible prescaler w/72mhz clock :(
2012-09-18T06:34:38 -!- grummund_ [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-18T06:41:38 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-18T06:43:25 < dongs> http://i.imgur.com/J5Sdq.jpg success.
2012-09-18T06:51:02 < gnomad> well that was quick!
2012-09-18T06:52:48 < dongs> not enough precision over pulse width though
2012-09-18T06:52:50 < dongs> gonna have to double timer speeed
2012-09-18T06:57:06 < dongs> that looks better.
2012-09-18T07:11:59 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-18T08:07:38 -!- dekar [~dekar@212.255.17.115] has quit [Ping timeout: 244 seconds]
2012-09-18T08:07:42 < dongs> http://i.imgur.com/2t42A.png
2012-09-18T08:07:42 < dongs> hmm
2012-09-18T08:07:44 < dongs> timing looks ok.
2012-09-18T08:09:22 -!- dekar [~dekar@212.255.22.171] has joined ##stm32
2012-09-18T08:09:23 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-18T08:10:07 < dongs> http://i.imgur.com/1ovnS.png
2012-09-18T08:10:12 < dongs> but, shit doesnt work.
2012-09-18T08:10:19 < dongs> just stays white :p
2012-09-18T08:16:28 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-18T08:16:31 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-18T08:29:51 -!- claude is now known as Claude
2012-09-18T08:33:18 < dongs> ok, wtf. http://i.imgur.com/2t42A.png + http://i.imgur.com/1ovnS.png + spec in http://www.world-semi.com/uploads/soft/120505/1-120505110346.rar if its really needed. timing looks ok, but shit's not working. discuss.
2012-09-18T08:34:35 <+izua> underpants gnomes
2012-09-18T08:34:48 <+izua> they figured out step 2
2012-09-18T08:35:14 < gnomad> ws2811 datasheet in RAR?
2012-09-18T08:43:09 < dongs> gnomad: i kno rite
2012-09-18T08:44:24 < gnomad> at least you found one...
2012-09-18T08:44:31 < dongs> .. its on manufacturer website..
2012-09-18T08:44:40 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-18T08:44:48 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-18T08:46:18 < emeb_mac> heh - just got an order status update from TI estore - the Stellaris Launchpad is due to ship... mid-November.
2012-09-18T08:46:25 < dongs> yeah..
2012-09-18T08:46:26 < dongs> same
2012-09-18T08:47:36 < emeb_mac> Well - for $5 can't complain too much.
2012-09-18T08:52:49 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 256 seconds]
2012-09-18T08:58:45 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-09-18T08:59:52 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-18T09:00:42 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 264 seconds]
2012-09-18T09:19:23 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-18T09:34:12 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-18T09:49:24 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-18T09:50:06 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has joined ##stm32
2012-09-18T09:50:06 -!- Tzacks [~opera@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-09-18T09:52:53 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 256 seconds]
2012-09-18T09:55:54 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 264 seconds]
2012-09-18T10:04:30 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-18T10:07:57 -!- Claude is now known as claude
2012-09-18T10:11:34 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-18T10:20:13 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-18T10:20:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-18T10:28:11 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-18T10:30:07 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-09-18T10:31:10 < dongs> http://www.youtube.com/watch?v=oIcwLXTcZ6w
2012-09-18T10:31:12 < dongs> success
2012-09-18T10:38:32 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-18T10:46:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-18T10:51:36 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-18T10:51:53 -!- claude is now known as claude_work
2012-09-18T10:53:07 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-18T11:16:57 < Mazingaro> hi
2012-09-18T11:17:09 < Mazingaro> is it possible to see psp register with gdb?
2012-09-18T11:23:56 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-09-18T11:24:39 < zyp> Mazingaro, sure, my gdb shows psp
2012-09-18T11:25:47 < dongs> <zyp> i had to recompile it from source and apply 10 years old patches from RMS to do that though
2012-09-18T11:26:11 < zyp> but it requires that the gdbserver (debugger interface) tells gdb about the register layout
2012-09-18T11:26:45 < zyp> dongs, nah, the problem is not gdb itself
2012-09-18T11:27:03 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-18T11:28:38 < zyp> Mazingaro, which debugger hardware are you using?
2012-09-18T11:29:16 < dongs> http://www.youtube.com/watch?v=T6MIvW0hglQ
2012-09-18T11:59:54 < Mazingaro> zyp: openocd + arm-usb-ocd-h by olimex
2012-09-18T12:00:15 < Mazingaro> zyp: openocd -d 4 -f interface/olimex-arm-usb-ocd-h.cfg -f board/olimex_stm32_h107.cfg
2012-09-18T12:05:51 < zyp> ah
2012-09-18T12:06:55 < zyp> openocd's gdb server is said to be shitty, so I'm not sure whether it's possible
2012-09-18T12:08:58 < zyp> I believe ntfreak is an openocd developer, so he might be able to give you some advice
2012-09-18T12:17:06 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-18T12:17:19 < ntfreak> Mazingaro: at the moment use the 'mon reg psp' or 'mon reg' to see all registers
2012-09-18T12:18:05 < ntfreak> zyp: shitty - thats a bit harsh
2012-09-18T12:19:23 < Mazingaro> tx
2012-09-18T12:22:32 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 260 seconds]
2012-09-18T12:26:53 < zyp> ntfreak, no offence intended
2012-09-18T12:47:55 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-18T12:56:01 < dongs> haha
2012-09-18T12:57:42 < ntfreak> :-)
2012-09-18T12:57:58 < dongs> here, I can repeat it for zyp
2012-09-18T12:58:02 < dongs> since my opinion doesnt matter.
2012-09-18T12:58:08 < dongs> but when I used openocd with sam7
2012-09-18T12:58:10 < dongs> it was the worst shit ever.
2012-09-18T12:58:38 < dongs> half broken shit that needed to start some commandline shit every time in background to use with eclipse etc.
2012-09-18T12:58:48 < dongs> im sure it's imrpoved somewehat in 4 or so years but i doubt it
2012-09-18T12:59:11 < Thorn> I'm currently using openocd + olimex ocd-h, it's much better than texane stlink
2012-09-18T12:59:37 < dongs> Thorn: its easy to be smart when surrounded by idiots
2012-09-18T13:00:05 < ntfreak> glad we don't all think like that, cannot use MDK for everything
2012-09-18T13:00:16 < dongs> i dont use mdk for everything.
2012-09-18T13:00:24 < dongs> i use crossstudio too.
2012-09-18T13:01:24 < Thorn> I have to go on with my BMP clone project though
2012-09-18T13:01:57 < ntfreak> crossstudio is a good choice - they are also good supporters of OpenOCD
2012-09-18T13:02:32 < Thorn> because that olimex thing doesn't support swd and openocd isn't exactly ideal
2012-09-18T13:03:03 < BrainDamage> wait what? cloning an open source project? what for?
2012-09-18T13:05:40 < Thorn> I want some hardware changes and several copies too, which unfortunately doesn't make it cost effective to buy
2012-09-18T13:06:18 < BrainDamage> so take the project and alter it, why starting from scratch?
2012-09-18T13:07:15 < zyp> who said anything about starting from scratch?
2012-09-18T13:08:29 < BrainDamage> isn't taking the existing stuff called forking?
2012-09-18T13:09:05 < zyp> I assumed he's talking about doing a new board to run the same firmware on
2012-09-18T13:09:27 < zyp> I've considered that myself
2012-09-18T13:09:40 < BrainDamage> oh, nvm then
2012-09-18T13:09:50 < Thorn> what zyp said
2012-09-18T13:10:03 < Thorn> no need to have to support a firmware fork
2012-09-18T13:10:25 < zyp> I'm kind of tempted to make a board with an edge connector for the 20-pin 0.1" pitch header, so I can plug it directly into boards using that connector, without any ribbon cable between
2012-09-18T13:10:48 < zyp> in addition to having a 0.05" header on it
2012-09-18T13:10:52 < Thorn> I just want several connectors onboard and maybe isolation, so that debugging externally powered stuff is safe
2012-09-18T13:11:12 < zyp> you mean isolated ground?
2012-09-18T13:11:59 < Thorn> the jtag/swd interface
2012-09-18T13:12:06 < Thorn> including ground
2012-09-18T13:12:49 < zyp> how are you going to isolate SWDIO, considering it's bidirectional?
2012-09-18T13:13:07 < dongs> isolol more like
2012-09-18T13:14:45 < Thorn> I think I've seen bidirectional isloated buffer ICs, but I'm not sure. need more research
2012-09-18T13:16:38 < Thorn> or something like http://uk.farnell.com/avago-technologies/acsl-6210-00re/optocoupler-smd-dual-bi-dir/dp/9130144?Ntt=ACSL-6210-00RE
2012-09-18T13:18:22 < BrainDamage> if the schematic here is any indicative of the internal structure, just looks like a lot of unidirectional optocoupler: http://www.farnell.com/datasheets/312935.pdf
2012-09-18T13:18:41 < zyp> yeah
2012-09-18T13:18:44 < Mazingaro> ntfreak: I'd like to show the cortex-m3 special registers in ddd
2012-09-18T13:19:18 < zyp> you'll probably need to use one for each direction and then pass through a signal to select direction
2012-09-18T13:23:15 < Thorn> here is the datasheet http://www.farnell.com/datasheets/1266120.pdf
2012-09-18T13:24:04 < dongs> why does farnell cache datasheets.
2012-09-18T13:24:06 < dongs> dumbest idea ever
2012-09-18T13:24:24 < dongs> do they spider manufacturer websites daily to check for outdated files?
2012-09-18T13:24:35 < Laurenceb> unlikely
2012-09-18T13:24:41 < karlp> no, it means they don't have to spider daily to keep the links working
2012-09-18T13:24:47 < dongs> uh,,,
2012-09-18T13:24:53 < karlp> some of the manufacturers (re)move their pdfs all the time.
2012-09-18T13:25:24 < karlp> but yeah, I still wouldn't cache them either if I was farnell.
2012-09-18T13:25:48 < karlp> in some places it could be viewed as a service to their local users, to save them paying overseas download charges?
2012-09-18T13:26:17 < zyp> «overseas download charges»
2012-09-18T13:26:18 < dongs> plz
2012-09-18T13:26:18 < zyp> wtf
2012-09-18T13:26:21 < dongs> that hasnt been relevant
2012-09-18T13:26:24 < dongs> for about a decade now
2012-09-18T13:26:38 < dongs> ausfags used to have some "out of country" bandwidth limits
2012-09-18T13:26:42 < karlp> so does iceland
2012-09-18T13:26:47 < dongs> like you had 10gb outside australia
2012-09-18T13:26:50 < karlp> I bet a lot of africa and south america does too
2012-09-18T13:26:51 < dongs> and 50 inside
2012-09-18T13:26:53 < dongs> or someshit
2012-09-18T13:27:00 < dongs> africans dont download electronics
2012-09-18T13:27:02 < Thorn> could become relevant again after the December ITU meeting
2012-09-18T13:27:05 < dongs> they just jack cars
2012-09-18T13:27:19 < dongs> neither mexicans
2012-09-18T13:27:21 < zyp> hey, you are chatting to me on the other side of the world, this is going to be soo expensive for you!
2012-09-18T13:27:29 < karlp> or just slow.
2012-09-18T13:27:49 < karlp> hey, I still think it's an odd thing for farnell to do, but at least I can see reasons why :)
2012-09-18T13:28:59 < Oldboy> and its a lifesaver for obsolete products wehre farnell still has stock, but the manufacturer has scrubbed all traces of that part from their website
2012-09-18T13:29:36 < dongs> why teh fuck would you design something with an obsolete product
2012-09-18T13:30:26 < Oldboy> and shitheads like honeywell that change the part number and sell the same thing with a new part numbr
2012-09-18T13:30:29 < Thorn> I heard SDR SDRAM production has been mostly stopped
2012-09-18T13:31:51 < dongs> Thorn: not liek anyone uses sdram in a new design
2012-09-18T13:31:56 < dongs> unless they're chumby or cheapfucks
2012-09-18T13:32:24 < Thorn> even in a microcontroller design?
2012-09-18T13:32:45 < dongs> Thorn: only liek 2 microcontorllers have sdram interface
2012-09-18T13:33:50 < dongs> Thorn: get some cheap fpga w/ddr controller and access it through fsmc or something
2012-09-18T13:33:51 < Thorn> afaik many cortex-m3/m4 based ones do (except stm32 that is)
2012-09-18T13:34:02 < Thorn> >cheap fpga
2012-09-18T13:34:06 < dongs> yea?
2012-09-18T13:34:09 < dongs> shit's cheap in qty.
2012-09-18T13:36:00 < dongs> i thought LPC has sdram,, but not ddr/ddr2  sdram
2012-09-18T13:36:12 < dongs> anyway, with all the faggot routing you need to do for highspeed ddr idoubt any hobbyists are gonna be doing it.
2012-09-18T13:36:40 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-09-18T13:36:50 < Thorn> exactly
2012-09-18T13:37:07 < Thorn> there're some cortex chips with ddr controllers too
2012-09-18T13:37:16 < dongs> name some
2012-09-18T13:38:19 < Thorn> freescale kinetis something
2012-09-18T13:38:30 < Thorn> was discussed here a few months ago
2012-09-18T13:38:44 < dongs> looks like K60/K70
2012-09-18T13:38:45 -!- tavish [~tavish@120.59.45.199] has joined ##stm32
2012-09-18T13:39:09 -!- tavish is now known as Guest4418
2012-09-18T13:39:12 < Laurenceb> what does stm32 fsmc interface with?
2012-09-18T13:39:18 < dongs> sram
2012-09-18T13:39:21 < dongs> nand/nor
2012-09-18T13:39:29 < dongs> i.e. it doesnt have the sdram refresh shiz
2012-09-18T13:41:36 < dongs> Supports connection of DDR, DDR2 and low-power DDR memories
2012-09-18T13:42:18 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-18T13:42:21 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-18T13:42:56 < dongs> i think its just kinetis...
2012-09-18T13:43:02 < dongs> I think some LPC shit has sdram only
2012-09-18T13:43:05 < dongs> no ddr
2012-09-18T13:43:09 < dongs> dunno of any others..
2012-09-18T13:45:00 <+izua> http://hackaday.com/2012/09/06/frequency-analyzer-built-from-the-new-stellaris-launchpad/
2012-09-18T13:45:05 <+izua> ^ this is not reddit guys
2012-09-18T13:45:36 <+izua> also, stellaris looks neat
2012-09-18T13:45:42 < Thorn> well, exactly, so sdr still has its uses in new designs
2012-09-18T13:46:06 < dongs> video after which break
2012-09-18T13:46:07 < dongs> i dont see shit
2012-09-18T13:47:27 < dongs> so its a fft?
2012-09-18T13:52:12 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Quit: leaving]
2012-09-18T13:52:22 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-18T13:53:20 < dongs> https://www.olimex.com/Products/MSP430/Booster/MOD-LED8x8/
2012-09-18T13:53:23 < dongs> how does this trash work?
2012-09-18T13:53:25 < dongs> just shift registers?
2012-09-18T13:54:21 < BrainDamage> looks like it
2012-09-18T13:54:23 < Thorn> data, latch, sck - looks like 595
2012-09-18T13:54:25 < BrainDamage> look at the board's back
2012-09-18T13:55:18 < Thorn> https://www.olimex.com/Products/MSP430/Booster/MOD-LED8x8/images/MOD-LCD8x8-03.jpg
2012-09-18T13:55:57 < Thorn> ^^ you can read chip marking
2012-09-18T13:56:07 < dongs> there's a schematic
2012-09-18T14:04:31 < zyp> looks similar to my 16x16 stuff
2012-09-18T14:05:26 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-18T14:05:28 <+izua> looks like it's only 8x8 tho and bigger
2012-09-18T14:05:38 < zyp> yep
2012-09-18T14:05:46 < zyp> but similar in how it's controlled
2012-09-18T14:06:01 < dongs> I like this onewire ws2811 shit
2012-09-18T14:06:05 < dongs> im g onna make a 20 meter cable
2012-09-18T14:06:10 < dongs> and see if it works as claimed over such distance
2012-09-18T14:06:14 <+izua> zyp - dude i've been looking for small led matrices since
2012-09-18T14:06:27 <+izua> i think you really have a small fortune there :P
2012-09-18T14:06:33 < dongs> olimexi is 5x5cm
2012-09-18T14:06:35 < dongs> haerdly small
2012-09-18T14:06:42 < dongs> wat
2012-09-18T14:06:45 < dongs> where's zyp's 16x16 stuff
2012-09-18T14:07:04 < zyp> dongs, the ones I have are 16x16 in 32x32mm
2012-09-18T14:07:18 <+izua> http://bin.jvnv.net/f/ldMgR.JPG ?
2012-09-18T14:07:28 < zyp> yep
2012-09-18T14:08:50 < dongs> zyp: oh.
2012-09-18T14:10:08 < zyp> found them in some store in akihabara when I was in japan in 2008
2012-09-18T14:10:30 <+izua> zyp: found an exploit
2012-09-18T14:10:31 <+izua> http://bin.jvnv.net/f/Q8RXd/dongsuino.png
2012-09-18T14:10:39 <+izua> "exploit"
2012-09-18T14:10:50 < zyp> exploit how?
2012-09-18T14:11:15 <+izua> exploit as in anonymous uploads enabled
2012-09-18T14:11:33 < zyp> it's a feature
2012-09-18T14:12:17 < zyp> it's like a pastebin, just for files
2012-09-18T14:13:53 <+izua> can i paste my collection of pi digits?
2012-09-18T14:15:32 < dongs> weow
2012-09-18T14:15:35 < dongs> thgis driver
2012-09-18T14:15:35 < dongs> still worsk
2012-09-18T14:15:39 < dongs> over liek 15 meter cat5
2012-09-18T14:15:40 < dongs> awesome
2012-09-18T14:15:45 <+izua> your keyboard driver?
2012-09-18T14:15:49 <+izua> seems to typo quite a bit though
2012-09-18T14:17:14 < zyp> izua, you can do whatever you want, unless you abuse it some way
2012-09-18T14:18:38 < zyp> of course, if you are wasting space for unnecessary bullshit, it will be deleted when the disk starts filling up :p
2012-09-18T14:20:51 < dongs> http://i.imgur.com/PsH6m.jpg
2012-09-18T14:21:05 < dongs> controller and led#1 is in my office
2012-09-18T14:21:13 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-18T14:21:16 < dongs> and thats about 15/20meteers of cat5
2012-09-18T14:21:47 < zyp> nice
2012-09-18T14:21:48 <+izua> details over protocol
2012-09-18T14:22:00 <+izua> is it like an ac thing, as with the model railroad dcc? or do you have extra data lines?
2012-09-18T14:22:00 < zyp> how does the addressing work?
2012-09-18T14:23:17 < dongs> this shit?
2012-09-18T14:23:35 < dongs> you clock out as many pixels as  you want and submit 50us+ sync low pulse.
2012-09-18T14:25:24 < karlp> oh, dekar/steffanx, re alternatives to minicom, I find picocom to be much nicer.
2012-09-18T14:25:42 < karlp> less of the built in windowing and menuing stuff, and none of the automagic AT command handling stuff
2012-09-18T14:26:15 < zyp> karlp, what benefits does it have over, say, screen?
2012-09-18T14:26:22 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Quit: No Ping reply in 180 seconds.]
2012-09-18T14:26:36 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-18T14:27:42 <+izua> linux users know nothing about bray's terminal
2012-09-18T14:28:29 < karlp> zyp: last I tried screen, it got very upset when the terminal came and went (usb plug/unplug stuff)
2012-09-18T14:28:46 < karlp> also, it was personally confusing having screen for detached remote terminals and also for the serial port
2012-09-18T14:29:06 < karlp> brays is that giant windows thing with buttons and checkboxes everywhere for every single signal in the universe right?
2012-09-18T14:34:53 < dongs> http://i.imgur.com/4HjyX.jpg lol
2012-09-18T14:46:40 <+izua> karlp: and with hex/ascii/bin simultaneous views, and graphing with parsing rules, and scripting
2012-09-18T14:46:53 <+izua> but yeah, you described its fails pretty well :P
2012-09-18T14:52:40 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-18T14:59:02 < Laurenceb> http://ec.europa.eu/consumers/dyna/rapex/rapex_archives_en.cfm
2012-09-18T14:59:09 < Laurenceb> we're gunna get rapexed
2012-09-18T14:59:11 < dongs> ooold
2012-09-18T14:59:36 < dongs> oh what
2012-09-18T14:59:38 < dongs> http://en.wikipedia.org/wiki/Anti-rape_device it used to be this
2012-09-18T14:59:42 < dongs> invented by some african
2012-09-18T14:59:48 < dongs> cuz everyone knows everyone in africa gets raped
2012-09-18T14:59:50 < Laurenceb> yeah i know
2012-09-18T15:00:25 < dongs> Initially called RapeX, the name was changed in 2006 upon discovering that RAPEX is also an EU warning system against dangerous goods on the market.
2012-09-18T15:00:28 < dongs> lulz
2012-09-18T15:02:59 < Laurenceb> someone at the eu is trolling
2012-09-18T15:03:42 < dongs> for sure
2012-09-18T15:15:14 -!- Guest4418 [~tavish@120.59.45.199] has quit [Quit: leaving]
2012-09-18T15:23:23 -!- daku is now known as DaKu
2012-09-18T15:28:12 < dongs> what the ufck
2012-09-18T15:28:28 < dongs> i touched 5V to GND pin of a lightbulb in teh house accidentally..
2012-09-18T15:28:33 < dongs> both panels tripped
2012-09-18T15:29:13 < dongs> ...??
2012-09-18T15:33:27 < zyp> gnd of a lightbulb?
2012-09-18T15:33:33 < zyp> your lightbulbs have gnd?
2012-09-18T15:35:25 < dongs> the outer rim
2012-09-18T15:35:32 < dongs> whatever the fuck signal that is
2012-09-18T15:35:36 < dongs> fucking trash jap nongrounded houses
2012-09-18T15:38:11 <+izua> uhm
2012-09-18T15:38:17 <+izua> and the other line wasn't really 5v
2012-09-18T15:42:15 < dongs> ?
2012-09-18T15:42:18 < dongs> waht other line
2012-09-18T15:42:29 < Laurenceb> http://en.wikipedia.org/wiki/April_20
2012-09-18T15:42:35 < Laurenceb> seems like a pretty cool day
2012-09-18T15:42:36 < dongs> i  was trying to stick my led shit behind a white diffuser
2012-09-18T15:42:40 < dongs> which was a light on side of a wall
2012-09-18T15:42:58 < dongs> i touched either gnd or 5V (which had around 3.3V) pin to the outer can of the lightbulb
2012-09-18T15:43:00 < dongs> both breakers tripped.
2012-09-18T15:43:09 < dongs> not just that circuit but entire house.
2012-09-18T15:43:51 < zyp> and you don't have any RCD?
2012-09-18T15:44:04 < dongs> right
2012-09-18T15:44:09 < dongs> cuz japs dont do that sorta trash
2012-09-18T15:44:46 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-18T15:45:25 < zyp> I mean, RCDs usually trip at 30mA, so they trigger at anything, but otherwise you must have had a pretty huge short
2012-09-18T15:45:55 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-18T15:46:20 < dongs> are you saying this reaction is something rcd would do?
2012-09-18T15:46:28 < zyp> yes
2012-09-18T15:47:07 < dongs> time to read the datasheet on the 2 breakers that tripped
2012-09-18T15:47:13 < dongs> they're the main house breakers.
2012-09-18T15:47:35 < dongs> whgen i asked the japs about installing rcd shit in bathroom/kitchen etc they had no fucking idea the shit evne existed
2012-09-18T15:48:02 < zyp> I once tripped an RCD at uni by touching the ground of my scope to a cooling fin that was not ground as I assumed but live
2012-09-18T15:48:25 < zyp> no sparks or anything, scope just went dark
2012-09-18T15:49:36 < zyp> a normal short is much more violent, since tripping the breakers is triggered by having several amps running trough the short
2012-09-18T15:49:53 < Laurenceb> http://forum.bodybuilding.com/showthread.php?t=352653&page=1
2012-09-18T15:49:57 < Laurenceb> a meme is born
2012-09-18T15:50:12 < dongs> right
2012-09-18T15:50:49 < zyp> also, the house I live in now have one single fucking 30mA RCD for the entire house
2012-09-18T15:51:30 < dongs> http://ntec.nito.co.jp/content/ppreview.html?code=C872-C1739-S3421
2012-09-18T15:51:32 < zyp> it almost trips just by the capacative load of all the circuits
2012-09-18T15:51:36 < dongs> zyp: thats what I got.
2012-09-18T15:51:37 < dongs> 30mA
2012-09-18T15:52:15 < zyp> well, then that's what happened
2012-09-18T15:52:29 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-18T15:52:32 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-18T15:52:59 < zyp> leakage current is the difference between the current out on one wire and back on the other, when leakage surpass 30mA it trips
2012-09-18T15:53:19 < zyp> so when you touched the other stuff to that, you had a leakage into the other circult
2012-09-18T15:53:27 < dongs> i see.
2012-09-18T15:53:39 < dongs> and my current limit on 3V supplywas > 30mA
2012-09-18T15:53:40 < dongs> lame.
2012-09-18T15:53:46 < dongs> fucking japs
2012-09-18T15:53:54 < zyp> usually the leakage is to protective earth, but it doesn't have to be
2012-09-18T15:54:42 < dongs> I'm sure theres a reason all of civilized world puts gfci shit only in places that need it
2012-09-18T15:54:42 < dongs> and not into e ntire house wiring
2012-09-18T15:54:42 < dongs> I guess they do this shit cuz they dont ground anything
2012-09-18T15:54:42 < dongs> is it?
2012-09-18T15:55:08 < zyp> huh?
2012-09-18T15:55:48 < zyp> in norway we have RCDs integrated into every normal circuit breaker now, for all circuits
2012-09-18T15:56:16 < zyp> old houses like the one I live in have a single RCD before the distribution out to the circuits
2012-09-18T15:56:28 < zyp> and that's fucking stupid, because it trips way too easy
2012-09-18T15:56:36 < dongs> thats what this shit is
2012-09-18T15:56:38 < zyp> like last night, when there was a thunderstorm
2012-09-18T15:56:44 < dongs> i have 2 panels
2012-09-18T15:56:48 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Remote host closed the connection]
2012-09-18T15:56:49 < dongs> office side and  house side for everythign else
2012-09-18T15:57:19 < zyp> I was about to head to bed, and then suddenly a power outage
2012-09-18T15:57:26 < dongs> both panels have that shit i linked in them
2012-09-18T15:57:26 < dongs> and individual 15A shits
2012-09-18T15:57:30 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-18T15:57:37 < zyp> except the neighbors still had their lights on
2012-09-18T15:58:31 < zyp> so I went upstairs and flipped it on again, and I didn't even manage to get downstairs again before the next lightning struck and it tripped again
2012-09-18T15:58:36 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-18T15:58:40 < zyp> so I just said fuck it and went to bed
2012-09-18T15:58:54 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-18T15:58:54 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-18T16:04:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 255 seconds]
2012-09-18T16:43:36 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-230.pptp.stw-bonn.de] has joined ##stm32
2012-09-18T16:44:36 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-230.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-18T16:49:40 < dongs> http://bcas.tv/paste/results/94jVVK15.html lets see tarduino do that
2012-09-18T16:51:13 < dongs> i was gona do DMA though
2012-09-18T16:51:19 < dongs> will need to try that now that this way works
2012-09-18T16:53:03 < zyp> dongs, given that avr also got timer interrupts and output compare and shit, it really can't be that hard doing the same on arduino
2012-09-18T16:53:26 < dongs> zyp, the only problem is at 16mhz, one clock cyucle is 65ns
2012-09-18T16:53:40 < dongs> you need minimum 10mhz clock to get close to needed precision.
2012-09-18T16:54:04 < dongs> 10mhz clock = for timer
2012-09-18T16:57:03 < jpa-> doesn't avr have that?
2012-09-18T17:00:03 <+Steffanx> it does
2012-09-18T17:00:27 -!- DaKu is now known as daku
2012-09-18T17:00:34 <+Steffanx> And when you talk about
2012-09-18T17:00:40 <+Steffanx> AVRs .. xmega also counts :P
2012-09-18T17:00:54 <+Steffanx> Which has also DMA and other fancy crap ( running at 32MHz )
2012-09-18T17:03:14 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-18T17:10:10 < dongs> jpa, surely you can have 10mhz clock (well, probably 16 since prescalers arepowers of 2 or whatever, but you dont have nough instructions to preload shit etc.
2012-09-18T17:11:25 < jpa-> so you can't do a nice led blinker with AVR? geez, i wonder how all the other people have managed that
2012-09-18T17:12:26 < dongs> jpa, not with this particular driver.
2012-09-18T17:12:40 < dongs> http://arduino.cc/forum/index.php/topic,80567.msg609684.html#msg609684
2012-09-18T17:16:16 <+Steffanx> "please help with WS2811 led IC" <= standard arduino way of how people ask for help :D :D
2012-09-18T17:16:40 < dongs> yes of course
2012-09-18T17:16:51 < jpa-> so pick a crappy ic and you have trouble :)
2012-09-18T17:21:57 < dongs> no
2012-09-18T17:22:05 < dongs> its a checapass IC which does the job perfectly ifne
2012-09-18T17:22:13 < dongs> and might actually be very useful as it seems to work fine over long distances
2012-09-18T17:22:20 < dongs> most other drivers with clock/data separtately fail
2012-09-18T17:24:45 <+izua> isn't that sort of what you're doing
2012-09-18T17:24:59 <+izua> led drivers with long lengths of wire
2012-09-18T17:26:03 < zyp> with seperate clock/data, you risk clock and data signals going out of sync
2012-09-18T17:26:55 < zyp> that chip is using a self-clocking signal, so it avoids the problem
2012-09-18T17:27:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-18T17:29:13 < Thorn> google doesn't even list any datasheets for WS2811 on the first page, they seem to be drowned in the arduino kids' pleas for HALP to get this chip to work
2012-09-18T17:31:54 < dongs> lol
2012-09-18T17:31:58 < dongs> yueah they're on manufacturers website
2012-09-18T17:32:00 < dongs> no worries there
2012-09-18T17:32:25 < SuicideFunky> hello
2012-09-18T17:32:34 <+Steffanx> Good afternoon
2012-09-18T17:33:13 < SuicideFunky> hi Steffanx
2012-09-18T17:33:43 < dongs> Thorn: http://www.world-semi.com/en/Driver/Lighting_LED_driver_chip/WS2811/
2012-09-18T17:33:56 < SuicideFunky> thought to join this channel again, been some time :)
2012-09-18T17:35:07 <+Steffanx> Too bad we didn't miss you :P
2012-09-18T17:35:18 < Thorn> 3 channels? I've been thinking about a M0 with CAN (some lpc11xx even have a builtin CAN driver) + MBI5026
2012-09-18T17:36:15 < Thorn> that MBI thnig is 5V so you'd need to do level translation for 3 or 4 signals
2012-09-18T17:36:40 < Thorn> but it has 16 constant current channels
2012-09-18T17:36:43 < SuicideFunky> haha :p
2012-09-18T17:38:09 < dongs> Thorn: that MBI isnt $0.15
2012-09-18T17:38:26 < dongs> and there's a 3 RGB channel version of 2811
2012-09-18T17:38:36 < dongs> which isnt much more expensive
2012-09-18T17:38:45 < Thorn> MBI should be pretty cheap
2012-09-18T17:39:36 < dongs> oyeah chink shit
2012-09-18T17:39:39 < dongs> i forgot why i didnt want to use htem
2012-09-18T17:39:47 < dongs> i do remember looking at them
2012-09-18T17:39:49 < Thorn> (unfortunately the 3.3V version was much more expensive last tiem I checked)
2012-09-18T17:40:03 < dongs> Thorn: yeah , this stuff also needs a buffer for 3V>5V
2012-09-18T17:40:05 < dongs> 2811
2012-09-18T17:40:12 < Thorn> I have ~20 of them, so they must have been cheap when I bought them
2012-09-18T17:40:15 < dongs> its VIO_H is VCC*0.7
2012-09-18T17:40:26 < dongs> yeah i asked my chinaguy to sample some 2811
2012-09-18T17:40:31 < dongs> he was like "shit's 15c/ea
2012-09-18T17:40:35 < dongs> im like yeah ok ill just buy 100
2012-09-18T17:40:36 < Thorn> I mean mbi5026
2012-09-18T17:42:53 -!- claude_work is now known as claude
2012-09-18T17:43:07 < dongs> http://arduino.cc/forum/index.php?topic=58027.0 haha first hit for MBI5026
2012-09-18T17:43:40 < Thorn> >because arduino cant see the mbi5026
2012-09-18T17:43:48 < dongs> lfao
2012-09-18T17:43:49 < Thorn> PnP didn't work right
2012-09-18T17:44:11 < Thorn> or VID&PID are unknown in this firmware version...
2012-09-18T17:44:48 < dongs> new hardware detection wizard in tarduino IDE has performed an illegal operation and will now close.
2012-09-18T17:46:08 < Thorn> and they haven't heard about pwm either.
2012-09-18T17:46:33 < dongs> so i dunno wat my next step should be with this shit. i could put a stm32 and buffer and USB or somethin on a small board.
2012-09-18T17:46:51 < dongs> i still need to test how this works with like 100 leds in chain
2012-09-18T17:47:03 < dongs> my ledpro dude says  shit wont work due to capacitance loading or some otehr crap
2012-09-18T17:47:13 < dongs> china claims it'll drive up to 512 leds per chain in this mode.
2012-09-18T17:48:12 < Thorn> that's why I considered CAN
2012-09-18T17:48:24 < dongs> you can only use what the driver provides.
2012-09-18T17:49:00 < Thorn> but you don't have to use a particular driver. :3
2012-09-18T17:49:11 < dongs> you do if cost/availability is an issue.
2012-09-18T17:51:21 < zyp> it's not if you only make stuff in your head.
2012-09-18T18:05:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-18T18:08:02 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-09-18T18:15:29 -!- daku is now known as DaKu
2012-09-18T18:24:30 -!- claude is now known as Claude
2012-09-18T18:24:59 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-18T18:31:47 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-18T18:52:45 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-18T18:52:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-18T19:00:56 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 240 seconds]
2012-09-18T19:16:57 -!- Claude is now known as claude
2012-09-18T19:27:07 -!- claude is now known as Claude
2012-09-18T19:37:54 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-18T19:39:19 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-18T19:39:22 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-18T19:41:24 -!- Claude is now known as claude
2012-09-18T19:43:54 -!- barthess [~barthess@178.154.89.210] has joined ##stm32
2012-09-18T19:44:31 <+izua> I use goto every time. I pity the fools who use languages that don't have them. It's a legit language construct; don't be a pussy, with power comes responsibility. If Djikstra's so smart, why is he dead then?
2012-09-18T19:44:47 < zyp> huh?
2012-09-18T19:44:56 <+Steffanx> Randon quote izua ?
2012-09-18T19:44:59 <+Steffanx> *random
2012-09-18T19:46:00 <+izua> just placing the bait
2012-09-18T19:47:07 <+Steffanx> k
2012-09-18T19:47:10 < dongs> gotos are fine
2012-09-18T19:47:12 < dongs> you cant troll me with that
2012-09-18T19:47:24 <+izua> i should probably drop that in #python
2012-09-18T19:47:43 <+izua> yeah, screw you dongs, it's not like i ruin your baits
2012-09-18T19:48:26 <+izua> http://entrian.com/goto/
2012-09-18T19:48:34 <+izua> meh, someone made goto for python as an april 1st
2012-09-18T19:51:50 < dongs> entire python is april 1st
2012-09-18T19:51:56 < dongs> unnecessary unfunny shit.
2012-09-18T19:52:20 < zyp> dongs, why the hate?
2012-09-18T19:52:28 <+Steffanx> troll .. feed .. something
2012-09-18T19:55:49 < dongs> zyp, pytyhon servese no purpose.
2012-09-18T19:56:11 <+Steffanx> <flamewar>
2012-09-18T19:56:15 <+izua> if you think really hard about it
2012-09-18T19:56:19 <+izua> neither do computers
2012-09-18T19:56:22 < zyp> dongs, so, what should I use instead of python?
2012-09-18T19:56:32 <+izua> zyp: ruby
2012-09-18T19:56:35 <+izua> or brainfsck
2012-09-18T19:56:43 <+Steffanx> pre-made tools
2012-09-18T19:56:48 <+Steffanx> with gui
2012-09-18T19:58:13 < Thorn> python is less worse than 99% of languages
2012-09-18T19:59:54 < emeb> heh - comefrom
2012-09-18T20:00:07 < jpa-> alter
2012-09-18T20:00:12 <+izua> that's pretty neat
2012-09-18T20:00:20 <+izua> useless, but still neat
2012-09-18T20:00:28 < emeb> great for obfuscation.
2012-09-18T20:00:58 <+Steffanx> btw dongs : "// applefag pulsing" <= how you know that?!
2012-09-18T20:01:05 < emeb> cause huge sections of code to be skipped and it's not obvious from the source until you see the comefrom statement later.
2012-09-18T20:01:14 < dongs> Steffanx: that  applefag is a sine?
2012-09-18T20:01:46 <+Steffanx> That THAT is "applefaq pulsing"
2012-09-18T20:01:58 -!- enots [dimka@freelsd.net] has joined ##stm32
2012-09-18T20:02:10 <+Steffanx> Only applefags know that
2012-09-18T20:02:19 <+izua> there was this great faq over becoming great at obfuscation
2012-09-18T20:02:26 <+izua> where you'd include the same header file three times
2012-09-18T20:02:45 <+izua> and each time it would #redefine stuff
2012-09-18T20:03:28 <+Steffanx> pre-processor only .. and that is gone :(
2012-09-18T20:04:04 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-18T20:05:07 <+izua> http://thc.org/root/phun/unmaintain.html
2012-09-18T20:05:13 <+izua> How To Write Unmaintainable Code  -  Ensure a job for life ;-)
2012-09-18T20:05:44 < emeb> a job you hate and that hates you
2012-09-18T20:05:48 < emeb> yay.
2012-09-18T20:05:56 <+izua> General Principles: Quidquid latine dictum sit, altum sonatur.
2012-09-18T20:05:57 < jpa-> http://stackoverflow.com/questions/6959360/goto-in-python   :)
2012-09-18T20:06:03 <+izua> (- Whatever is said in Latin sounds profound.)
2012-09-18T20:06:31 < emeb> Romanii ite domum.
2012-09-18T20:07:37 <+izua> romanes eunt domus
2012-09-18T20:08:24 <+Steffanx> zzzzz
2012-09-18T20:09:41 < emeb> caught the reference. and if it's not done by sunrise I'll cut your balls off.
2012-09-18T20:09:58 <+izua> By misspelling in some function and variable names, and spelling it correctly in others (such as SetPintleOpening SetPintalClosing) we effectively negate the use of grep or IDE search techniques.
2012-09-18T20:10:48 <+Steffanx> heh
2012-09-18T20:12:57 <+izua> Choose variable names that masquerade as mathematical operators, e.g.:
2012-09-18T20:12:57 <+izua>     openParen = (slash + asterix) / equals;
2012-09-18T20:14:39 <+Steffanx> it should be at least be openParen=(slash+asterix)/equals; :P
2012-09-18T20:14:50 <+Steffanx> Laurenceb-style
2012-09-18T20:16:37 -!- barthess [~barthess@178.154.89.210] has quit [Ping timeout: 252 seconds]
2012-09-18T20:36:09 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-18T20:37:32 -!- barthess [~barthess@178.154.10.132] has joined ##stm32
2012-09-18T20:48:31 -!- barthess [~barthess@178.154.10.132] has quit [Read error: Connection reset by peer]
2012-09-18T20:57:47 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-18T21:07:16 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-18T21:11:27 -!- barthess [~barthess@164.177.208.126] has joined ##stm32
2012-09-18T21:18:57 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-18T21:39:40 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-09-18T21:57:28 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-18T22:12:09 -!- DaKu is now known as daku
2012-09-18T22:15:32 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 248 seconds]
2012-09-18T22:23:51 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-18T22:24:30 < Laurenceb_> hi
2012-09-18T22:24:49 < emeb> lo
2012-09-18T22:24:49  * Laurenceb_ has been playing with blueman
2012-09-18T22:25:07 < Laurenceb_> and a class 1 bluetooth module + 2.4GHz yagi
2012-09-18T22:25:12 < Laurenceb_> perfect for trolling
2012-09-18T22:25:34 < emeb> or harvesting phone contacts
2012-09-18T22:25:36 < Laurenceb_> been trolling office across the road all day
2012-09-18T22:25:47 <+Steffanx> You also have a tv remove that can you use to troll your neighbours don't you?
2012-09-18T22:25:52 <+Steffanx> a universal one ..
2012-09-18T22:25:54 < Laurenceb_> set netwrk name to "peadobear"  etc
2012-09-18T22:31:59 < BrainDamage> if you have an android phone, they just made a raw driver for some broadcom chips
2012-09-18T22:32:14 < BrainDamage> this means you could break into weak wifi and start injecting junk
2012-09-18T22:32:23 < Laurenceb_> lol
2012-09-18T22:39:03 < zyp> sometimes it seems like all you guys do is injecting junk here
2012-09-18T22:41:05 <+Steffanx> Sometimes
2012-09-18T22:42:48 < emeb> watch the SNR drop.
2012-09-18T22:43:31 < BrainDamage> zyp: just sometimes? >_>
2012-09-18T22:55:07 < Laurenceb_> wtf
2012-09-18T22:55:21 < Laurenceb_> anyone here got the fatfs code?
2012-09-18T22:55:26 -!- daku is now known as DaKu
2012-09-18T22:55:32 < Laurenceb_> look in fattime.c
2012-09-18T22:55:53 < Laurenceb_> Martin Thomas stm32 version :  its timestamping with DST adjusted code
2012-09-18T22:56:49 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 244 seconds]
2012-09-18T22:58:09 < Laurenceb_> should stamp with UTC?
2012-09-18T23:00:57 < Laurenceb_> http://msdn.microsoft.com/en-us/library/windows/desktop/ms724290%28v=vs.85%29.aspx
2012-09-18T23:01:01 < Laurenceb_> WHAT
2012-09-18T23:01:09  * Laurenceb_ epic rage @ microshaft
2012-09-18T23:01:25 < Laurenceb_> " The FAT file system stores time values based on the local time of the computer"
2012-09-18T23:01:33 < Laurenceb_> idiots
2012-09-18T23:01:59 -!- barthess [~barthess@164.177.208.126] has quit [Quit: Leaving.]
2012-09-18T23:02:09 < olasd> well, yeah, they use local time to set the bios clock too
2012-09-18T23:02:18 < jkent> because fat is from DOS days.. its legacy compatability..
2012-09-18T23:02:37 < Laurenceb_> grrr
2012-09-18T23:02:47 < Laurenceb_> guess ill just swap DST to GMT then
2012-09-18T23:02:50 < Laurenceb_> *BST
2012-09-18T23:03:27 -!- qyx_ [~qyx@krtko.org] has quit [Quit: leaving]
2012-09-18T23:04:40 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-18T23:07:16 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-18T23:09:14 < Laurenceb_> hmm ok then - the fatfs code is all correct
2012-09-18T23:09:22 < Laurenceb_> yet my times are getting offset
2012-09-18T23:09:34 < Laurenceb_> i think ubuntu is using UTC for FAT timestamps
2012-09-18T23:09:45 < Laurenceb_> and nautilus corrects to BST/GMT
2012-09-18T23:09:59 < Laurenceb_> that would explain the behaviour im seeing
2012-09-18T23:11:40 < Laurenceb_> "we're linux developers, we'll fix microsofts problems for them"
2012-09-18T23:13:28 < Laurenceb_> http://www.linuxine.com/story/automounted-fat-fs-always-using-utc-time
2012-09-18T23:13:30 < Laurenceb_> lawl
2012-09-18T23:15:55 < zyp> how the fuck do you handle timestamps in local time in a sensible way when you consider DST?
2012-09-18T23:16:07 < zyp> not to mention moving between timezones
2012-09-18T23:17:42 < BrainDamage> it's simple /*giant clusterfuck magic code*/ time_t adjustTime( time_t rtctime ); { // nobody touched this since it's creator }
2012-09-18T23:24:24 < Laurenceb_> im just going to swap the fatfs dst code to do british summer time
2012-09-18T23:24:43 < Laurenceb_> seems its working correctly
2012-09-18T23:24:49 < Laurenceb_> its just linux tried to fix it
2012-09-18T23:25:04 < Laurenceb_> that forum describes the issue i see
2012-09-18T23:32:58 < emeb> is it possible that gcc optimization could reverse the index direction on an array copy operation?
2012-09-18T23:36:00 < gnomad> possible, yes.  likely, no.
2012-09-18T23:40:03 -!- qyx_ [~qyx@krtko.org] has joined ##stm32
2012-09-18T23:40:57 < zyp> emeb, as long as the result would be the same, it's allowed to
2012-09-18T23:41:32 < zyp> why? do you see it happening=
2012-09-18T23:45:52 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-18T23:45:52 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Disconnected by services]
--- Day changed Wed Sep 19 2012
2012-09-19T00:04:09 < Laurenceb_> http://en.wikipedia.org/wiki/British_Summer_Time
2012-09-19T00:04:17 < Laurenceb_>  The British Summer Time period begins on the last Sunday of March and ends on the last Sunday of October
2012-09-19T00:04:19 < Laurenceb_> wtf
2012-09-19T00:04:21 < Laurenceb_> RAGE
2012-09-19T00:04:29 < Laurenceb_> why cant they make it simple
2012-09-19T00:05:28 < olasd> if they wanted to make it simple, why would they have DST in the first place :)
2012-09-19T00:06:21 < Laurenceb_> i guess the DST code finds day of week
2012-09-19T00:06:36 < Laurenceb_> so i need day == sunday and less that 7 days in month
2012-09-19T00:08:36 < emeb> zyp: I'm moving data within an overlapping buffer.
2012-09-19T00:09:13 < emeb> 	for(i=0;i<3*MONO_BUFSZ;i++) save_in[i] = save_in[i+MONO_BUFSZ];
2012-09-19T00:09:50 < emeb> where save_in[] is 4*MONO_BUFSZ long.
2012-09-19T00:09:53 < Laurenceb_> http://en.wikipedia.org/wiki/Daylight_saving_time_in_Germany#Germany
2012-09-19T00:10:02 < Laurenceb_> " Daylight saving time was reintroduced by the Nazis in 1940"
2012-09-19T00:10:07 < Laurenceb_> so its bad right?
2012-09-19T00:10:18 < Laurenceb_> just in case there was any doubt that it was evil
2012-09-19T00:10:21 < emeb> AZ (where I live) doesn't do DST
2012-09-19T00:10:35 < emeb> we get enough daylight - no need to save it.
2012-09-19T00:11:46 < Laurenceb_>  In 1996, daylight saving time was harmonized throughout the European Union by Directive 2000/84/EC, which moved the end of DST to the last Sunday in October
2012-09-19T00:11:50 < Laurenceb_> lol EU directives
2012-09-19T00:11:58 < Laurenceb_> i think i dont actually need to do anything
2012-09-19T00:13:32 < Laurenceb_> the EU actually _saved_ me work ?! impossible
2012-09-19T00:13:35 < emeb> Yep - the buffer index was being reversed - caused the overlapping data to be trashed.
2012-09-19T00:14:32 < emeb> doing the move in 3 separate loops fixed it, but that's not very efficient. Wonder how to code so that doesn't happen?
2012-09-19T00:20:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-19T00:20:51 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-19T00:21:45 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-19T00:24:08 < karlp> emeb: part of AZ does daylight savings :)
2012-09-19T00:24:23 < karlp> unless the reservations stopped doing it in the last few years
2012-09-19T00:24:38 < emeb> karlp: It's  true.
2012-09-19T00:25:27 < karlp> anyone looked at keil's RTX (CMSIS-RTOS) thing yet?
2012-09-19T00:25:34 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-19T00:31:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-19T00:42:14 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/fat_fs/src/rtc.c#L26
2012-09-19T00:42:17 < Laurenceb_> brain melt
2012-09-19T00:42:34 < Laurenceb_> line 38 is confusing to say the least
2012-09-19T01:01:15 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 256 seconds]
2012-09-19T01:25:54 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-19T01:26:04 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-19T01:35:00 -!- mode/##stm32 [+o dekar] by ChanServ
2012-09-19T01:35:12 -!- dekar changed the topic of ##stm32 to: Prepare for the two minutes of /hate/love. This channel is publicly logged, ask zlog for more (also for free z80s). | Join us building the ##stm32 toolchain: https://github.com/EliasOenal/TNT (now with multilib) |  We have a wiki http://stm32.izua.ro | How engineers buy parts http://tim.cexx.org/?p=919
2012-09-19T01:35:26 -!- mode/##stm32 [-o dekar] by ChanServ
2012-09-19T01:41:49 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Remote host closed the connection]
2012-09-19T01:45:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-19T02:32:04 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Ping timeout: 248 seconds]
2012-09-19T02:36:33 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-19T02:44:33 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 252 seconds]
2012-09-19T03:00:58 -!- DaKu is now known as daku
2012-09-19T03:17:04 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds]
2012-09-19T03:54:09 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-19T03:57:45 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-19T04:39:43 < dongs> i might have to follow laurenceb style and use F103 with SD card on SPI :(
2012-09-19T04:39:49 < dongs> how terrible
2012-09-19T04:41:41 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Ping timeout: 256 seconds]
2012-09-19T05:00:11 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-19T05:43:24 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-19T05:43:24 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-19T05:43:24 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-19T05:43:27 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-19T06:24:47 < dongs> got jlinkOB in the mail.
2012-09-19T06:24:49 < dongs> lets see how it works
2012-09-19T06:25:02 < dongs> cute. all traces are curvy
2012-09-19T06:25:10 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-19T06:26:16 < dongs> Firmware: J-Link ARM-OB STM32 compiled Dec 15 2010 11:30:03
2012-09-19T06:26:16 < dongs> Hardware: V7.00
2012-09-19T06:33:24 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-19T06:33:29 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-19T06:54:05 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-19T07:00:24 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-19T07:12:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-19T07:17:07 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 245 seconds]
2012-09-19T07:46:42 < dongs> fuckback
2012-09-19T07:46:54 < dongs> flyback: http://i.imgur.com/QlNZh.jpg
2012-09-19T07:47:10 < dongs> flyback: http://www.youtube.com/watch?v=T6MIvW0hglQ
2012-09-19T07:47:35 < dongs> shit works.
2012-09-19T07:52:07 < dongs> yeah once you ahve individually controllable leds, the rest is just whatever
2012-09-19T08:14:59 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-19T08:26:02 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-19T08:43:13 < dongs> loldongs
2012-09-19T09:55:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-19T10:13:32 -!- dekar [~dekar@212.255.22.171] has quit [Quit: This computer has gone to sleep]
2012-09-19T10:20:57 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-19T10:26:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-19T10:36:41 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 245 seconds]
2012-09-19T10:38:14 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-19T10:49:35 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-19T10:53:13 -!- dekar [~dekar@dslb-088-069-128-239.pools.arcor-ip.net] has joined ##stm32
2012-09-19T10:53:16 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-19T11:07:37 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 252 seconds]
2012-09-19T11:30:48 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-19T11:33:44 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Read error: Connection reset by peer]
2012-09-19T11:35:54 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-19T11:50:47 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 244 seconds]
2012-09-19T11:52:23 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-19T12:11:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-19T12:41:35 < Laurenceb> sup
2012-09-19T12:42:11 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/fat_fs/src/rtc.c#L38
2012-09-19T12:42:17 < Laurenceb> ^thats tarded right?
2012-09-19T12:42:24 < Laurenceb> assumes 31days in a month
2012-09-19T12:44:07 < zyp> of course it does, it's only ever true for month == 10
2012-09-19T12:44:08 < dongs> sup trolls
2012-09-19T12:44:10 < dongs> Laurenceb: did you hear
2012-09-19T12:44:20 < dongs> Laurenceb: theyre making me make a f103 w/sd on SPI and USB
2012-09-19T12:44:30 < dongs> fucking rage.
2012-09-19T12:44:36 < zyp> or month == 3
2012-09-19T12:44:44 < Laurenceb> loldongs
2012-09-19T12:44:54 < Laurenceb> zyp: so
2012-09-19T12:44:56 < zyp> dongs, who are they?
2012-09-19T12:44:57 < Laurenceb> its still wrong
2012-09-19T12:45:02 < dongs> zyp, some fags.
2012-09-19T12:45:17 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/fat_fs/src/rtc.c#L15
2012-09-19T12:45:18 < zyp> Laurenceb, no, march and october is always 31 days.
2012-09-19T12:45:23 -!- claude is now known as claude_work
2012-09-19T12:45:24 < Laurenceb> hmm
2012-09-19T12:45:29 < Laurenceb> oh ok
2012-09-19T12:45:33 < Laurenceb> yeah
2012-09-19T12:45:37 < dongs> you gotta use the knucles rule
2012-09-19T12:45:52 < Laurenceb> which is?
2012-09-19T12:45:53 < dongs> start with january and left knuckle on right hand
2012-09-19T12:45:58 < dongs> the high ones are 31 days
2012-09-19T12:46:00 < dongs> low ones are 30 or less
2012-09-19T12:47:05 < Laurenceb> hmm
2012-09-19T12:47:16 < Laurenceb> ok, so why is it using 2am?
2012-09-19T12:47:36 < dongs> http://lifehacker.com/232828/macgyver-tip-use-your-knuckles-to-remember-each-months-days
2012-09-19T12:48:18 < dongs> i just go back on same hand in reverse though
2012-09-19T12:48:19 < dongs> same effect
2012-09-19T12:48:52 < zyp> Laurenceb, because that's when it changes?
2012-09-19T12:52:48 < Laurenceb> oh wait
2012-09-19T12:52:57 < Laurenceb> itd help if i understood how DST works
2012-09-19T12:53:07 < dongs> ...
2012-09-19T12:53:31 < dongs> Daylight Saving Time (United States) began Sunday, March 11, 2012, 2:00am, and ends Sunday, November 4, 2012, 2:00am.
2012-09-19T12:53:35 < dongs> Except Arizona and Hawaii. Move your clocks ahead 1 hour in spring and back 1 hour in fall ("Spring forward, fall back").
2012-09-19T12:54:10 < dongs> so it offsets whatever current time is vs utc by 1 hour
2012-09-19T12:55:08 < karlp> I can't imagine ever wanting to work on an embedded device that had to care about DST
2012-09-19T12:55:51 < karlp> firmware updates to handle changing start/stop dates and all that bullshit.
2012-09-19T12:56:04 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-19T12:56:06 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-19T12:56:16 < Laurenceb> no harm in implimenting it properly
2012-09-19T12:59:57 < karlp> huh, new f051 in lqfp32, with fat fingered friendly 0.8mm pitch, instead of 0.5mm pitch
2012-09-19T13:00:18 < Laurenceb> interesting
2012-09-19T13:00:28 < karlp> though, given that it's 7mmx7mm, just like the lqfp48, I hardly know how they can claim this in the press release, "is now housed in an even
2012-09-19T13:00:32 < karlp> smaller package."
2012-09-19T13:01:51 < karlp> oh, maybe they're talking about the qfn32,  that must be new too
2012-09-19T13:01:52 < Laurenceb> ah
2012-09-19T13:02:26 < Laurenceb> looks like the fatfs code wont handle logfile recorded between 1am and 2am quite right when the clocks are changing
2012-09-19T13:02:29 < karlp> oh, and they're claimning a 20pin package too, but that's not listed anywhere I can find.
2012-09-19T13:02:31 < Laurenceb> i can live with that
2012-09-19T13:04:52 <+izua> Laurenceb: you mean daylight saving
2012-09-19T13:04:58 <+izua> why.. would that even matter?
2012-09-19T13:05:16 < Laurenceb> if someone is up late at night taking data...
2012-09-19T13:05:19 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-19T13:05:44 <+izua> yes but, why does your device even care about dst?
2012-09-19T13:05:46 < Laurenceb> oh - DST matters as otherwise there will be odd 1 hour offsets
2012-09-19T13:06:01 <+izua> umm
2012-09-19T13:06:23 <+izua> store the local timezone somewhere?
2012-09-19T13:06:45 <+izua> or this is way over my head
2012-09-19T13:06:50 <+izua> inb4 device in orbit
2012-09-19T13:21:37 < zyp> izua, it matters because FAT saves timestamps in local time
2012-09-19T13:21:55 < zyp> so the problem is FAT being utterly dumb
2012-09-19T13:23:34 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-19T13:29:32 < Laurenceb> yes
2012-09-19T13:29:41 < dongs> hehe
2012-09-19T13:30:00 < Laurenceb> what happens if i chuck a load of data into a serial bluetooth module when it not connected?
2012-09-19T13:30:13 < Laurenceb> it'll buffer some and then start chucking it?
2012-09-19T13:30:59 < dongs> ur chucking it
2012-09-19T13:31:21 <+izua> how does windows deal with files created/modified during the dst changeover?
2012-09-19T13:31:36 < dongs> noone uses fat on windows
2012-09-19T13:31:43 < dongs> and NTFS stores shit in UTC i believe.
2012-09-19T13:32:18 <+izua> s/deal/dealt
2012-09-19T13:32:33 <+izua> and the fact that no one uses it, doesn't mean no one used it at some point :P
2012-09-19T13:38:59 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-19T13:41:00 < Laurenceb> its used on lots of stuff in my experience
2012-09-19T13:41:03 < Laurenceb> cameras etc
2012-09-19T13:43:36 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-19T13:44:46 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-19T13:44:49 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-19T13:50:39 -!- gsmcmull1n [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 260 seconds]
2012-09-19T13:50:50 < Laurenceb> http://www.theonion.com/articles/no-one-murdered-because-of-this-image,29553/
2012-09-19T13:50:54 < Laurenceb> very NSFW
2012-09-19T13:52:19 < karlp> OLD
2012-09-19T14:16:50 -!- daku is now known as DaKu
2012-09-19T14:19:11 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-19T14:29:34 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 240 seconds]
2012-09-19T14:38:12 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-19T14:41:26 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-19T14:51:34 -!- DaKu is now known as daku
2012-09-19T14:52:22 < zyp> hmm, somebody claimed the reason the L1discovery doesn't have an USB connector was that the pins are used for the LCD
2012-09-19T14:52:47 < zyp> according to datasheet that's not true, the USB pins are listed as free IO
2012-09-19T14:52:54 < zyp> (and has no LCD functionality)
2012-09-19T14:53:54 -!- daku is now known as DaKu
2012-09-19T14:58:07 < dongs> ST saved 0.01 by not mounting USB conector
2012-09-19T15:01:11 < prog> that will cost us a few tens of $
2012-09-19T15:01:52 < karlp> or, they didn't have an usb demos ready, so they just ditched it?
2012-09-19T15:03:45 < dongs> haha
2012-09-19T15:03:47 < dongs> they still dont
2012-09-19T15:04:33 < zyp> stuff should be directly portable from F1
2012-09-19T15:25:56 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-19T15:27:51 < Laurenceb> http://www.theregister.co.uk/2012/09/19/the_iphone_5_queue_london/
2012-09-19T15:28:00 < dongs> lol, idiots.
2012-09-19T15:28:07 < Laurenceb> "a very keen Apple enthusiast and his carer"
2012-09-19T15:28:09 < Laurenceb> lolling
2012-09-19T15:28:55 < dongs> so all this tarduino kickstarter garbage keeps succeeding
2012-09-19T15:29:01 < dongs> http://www.kickstarter.com/projects/kenburns/tinyduino-the-tiny-arduino-compatible-platform-w-s
2012-09-19T15:29:04 < dongs> wtf is this
2012-09-19T15:29:07 < dongs> !!!!
2012-09-19T15:29:13 <+izua> it's a tiny arduino compatbile platform
2012-09-19T15:29:28 < Laurenceb> whyyyy
2012-09-19T15:29:28 < dongs> http://www.kickstarter.com/projects/818225433/arduino-style-battery-shield?ref=category
2012-09-19T15:29:32 < dongs> ahahahah
2012-09-19T15:29:43 < dongs> arduino-style-battery-shield
2012-09-19T15:29:52 < dongs> The Battery Shield makes it easy to add 9 volts to any chipKIT, Arduino, CHIPINO or any other chipKIT / Arduino/ CHIPINO style module.
2012-09-19T15:30:18 <+izua> yes, because you can't get a battery holder and shovethe wires in the arduino
2012-09-19T15:30:21 <+izua> you need a shield
2012-09-19T15:30:29 < Laurenceb> "Apple queue veteran Zohaib Ali and his carer, who were there representing an autism charity."
2012-09-19T15:30:31 < dongs> and it uses AAAA batteries
2012-09-19T15:30:32 < Laurenceb> oh dear
2012-09-19T15:30:33 < dongs> i never seen those
2012-09-19T15:30:42 < Laurenceb> he actually does have a carer
2012-09-19T15:31:43 < dongs> http://www.kickstarter.com/projects/rowdyrobot/arduino-ble-shield-connecting-the-ios-and-the-ardu?ref=category
2012-09-19T15:31:54 < Laurenceb> The Morph suit looks like something for you dongs
2012-09-19T15:32:01 < Laurenceb> http://regmedia.co.uk/2012/09/19/iphone5_queue_4.jpg
2012-09-19T15:32:24 <+izua> dongs: if you open up a 9v battery
2012-09-19T15:32:27 <+izua> you might see 6
2012-09-19T15:32:28 < zyp> so, why 9V? doesn't the arduino use a linear reg? what's the dropout voltage of the reg? couldn't one get away with a battery or two fewer without any loss in capacity?
2012-09-19T15:32:50 < dongs> zyp: i think it has a huge dropout reg
2012-09-19T15:32:59 < dongs> like your good ol 7805 wiht liek 2V dropout
2012-09-19T15:33:04 < dongs> i thin ktarduino is spec'd 7V..35V
2012-09-19T15:33:37 < dongs> http://www.kickstarter.com/projects/1973311155/velocomputer-1-precision-bluetooth-smart-sensor-4?ref=category why does this guy want 100k
2012-09-19T15:33:57 <+izua> because money
2012-09-19T15:34:22 < dongs> http://www.kickstarter.com/projects/jaredbouck/sprout-board-the-ultimate-arduino-experimenters-bo?ref=category
2012-09-19T15:34:32 <+izua> At least 360 times more accurate than traditional cycling sensors
2012-09-19T15:34:35 < dongs> http://s3.amazonaws.com/ksr/assets/000/117/583/aa947b578974b3252468e038299f107f_large.JPG?1345157415
2012-09-19T15:34:39 < dongs> deliverables.
2012-09-19T15:34:58 <+izua> $2100 is reasonable
2012-09-19T15:35:21 < dongs> http://www.kickstarter.com/projects/985890579/mini-usb-host-microcontroller-board-arduino-compat?ref=category
2012-09-19T15:35:24 < dongs> haha.
2012-09-19T15:35:53 < Laurenceb> i need to do some work and stop trolling
2012-09-19T15:36:08 < dongs> Laurenceb: you need to throw some shit on kickstarter
2012-09-19T15:36:28 < dongs> http://www.kickstarter.com/projects/kuy/galago-make-things-better?ref=category
2012-09-19T15:36:30 < dongs> yaaa.
2012-09-19T15:36:36 < dongs> even my brony board is better than this shit
2012-09-19T15:36:47 < dongs> "...the Galago might just be the perfect ARM board for tinkerers weaning themselves off the Arduino."
2012-09-19T15:36:50 < dongs> 'weaning'
2012-09-19T15:37:11 <+izua> dongs: why do you browse a poni board
2012-09-19T15:37:21 < dongs> izua: wat
2012-09-19T15:37:50 <+izua> is that galago running on stm32?
2012-09-19T15:37:53 < dongs> no
2012-09-19T15:37:55 < dongs> some lpc shit
2012-09-19T15:37:58 < dongs> lpc2100 i think?
2012-09-19T15:37:59 < dongs> oro smehting
2012-09-19T15:37:59 <+izua> <dongs> even my brony board is better than this shit
2012-09-19T15:38:16 < dongs> lpc13xx
2012-09-19T15:38:23 < dongs> izua: i was talking about the pink stm32 shit.
2012-09-19T15:39:01 <+izua> i thought you meant "even my poni forum is better than this community of retards"
2012-09-19T15:39:06 <+izua> which.. kind of makes sense.
2012-09-19T15:39:21 <+izua> you should put it on kickstarter though
2012-09-19T15:39:35 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-19T15:39:57 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-19T15:40:42 <+izua> decent one: http://www.kickstarter.com/projects/jywarren/public-lab-diy-spectrometry-kit?ref=category
2012-09-19T15:41:44 <+izua> using only a dvd ^^
2012-09-19T15:41:53 <+izua> "spectrometer"
2012-09-19T15:44:49 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-19T15:46:11 -!- Netsplit *.net <-> *.split quits: |sark, ReggieUK
2012-09-19T15:47:31 < dongs> https://c9.io/
2012-09-19T15:47:32 < dongs> wat
2012-09-19T15:48:29 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Ping timeout: 244 seconds]
2012-09-19T15:48:31 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-19T15:50:19 < Laurenceb> wtf blueman
2012-09-19T15:50:20 < Laurenceb> Device added successfuly, but failed to connect
2012-09-19T15:50:26 < dongs> SWD on attiny45 w/ software usb
2012-09-19T15:50:33 < dongs> why
2012-09-19T15:51:25 < dongs> After the successor to Galago ships, Galago's firmware will be released so people can learn from how it works. I believe this is the most reasonable compromise between the needs of business and open source.
2012-09-19T15:51:32 < dongs> haha
2012-09-19T15:55:23 < dongs> how can i find more trollshit on kickstaerter
2012-09-19T15:55:43 < dongs> that /discover category only lists uselss shit
2012-09-19T15:55:45 < dongs> wheres more shit
2012-09-19T15:55:52 < dongs> dont teghy have a LIST
2012-09-19T15:56:32 < dongs> http://www.kickstarter.com/projects/1648703585/arduino-compitable-quad-dc-motor-control-board?ref=card
2012-09-19T15:58:46 < dongs> http://www.kickstarter.com/projects/1973759599/open-source-smd-parts-feeders?ref=card oh haha this guy got f unded
2012-09-19T15:59:43 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-19T16:00:02 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-19T16:01:20 < BrainDamage> the latter actually seems way more reasonable than many other bullshit projects
2012-09-19T16:01:34 < BrainDamage> at least that one allegedly exists and works
2012-09-19T16:01:38 < dongs> you mean pay me to motivate me to get docs out?
2012-09-19T16:01:57 < BrainDamage> to motive, more like pay me to write docs
2012-09-19T16:01:59 < BrainDamage> motivate*
2012-09-19T16:02:05 < dongs> well yeah sure
2012-09-19T16:03:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-19T16:16:40 < Laurenceb> wtf
2012-09-19T16:16:43 < Laurenceb> Error mounting: mount: /dev/sdc1: can't read superblock
2012-09-19T16:16:51 < Laurenceb> yet the filesystem is clean
2012-09-19T16:17:00 < Laurenceb> using fschk
2012-09-19T16:17:04 < Laurenceb> or whatever its called
2012-09-19T16:17:06 <+izua> dongs: http://i.imgur.com/7Af2s.jpg
2012-09-19T16:17:10 <+izua> they found you man
2012-09-19T16:18:02 < Laurenceb> oh shit
2012-09-19T16:18:12 < Laurenceb> the card seems to be screwed
2012-09-19T16:18:35 < dongs> supablog
2012-09-19T16:20:53 < Laurenceb> rageeeee
2012-09-19T16:21:04 < Laurenceb> when my stm part works the uSD card fails
2012-09-19T16:21:14 < Laurenceb> maybe it needs cleaning
2012-09-19T16:22:07 < Laurenceb> opps
2012-09-19T16:22:11 < Laurenceb> its screwed
2012-09-19T16:22:29 < Laurenceb> *epic micro sd card quest*
2012-09-19T16:23:03 < Laurenceb> i think it was dropped in 1molar nitric acid
2012-09-19T16:27:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-19T16:27:09 < dongs> a mole of moles.
2012-09-19T16:27:30 < dongs> http://what-if.xkcd.com/4/
2012-09-19T16:37:46 < Laurenceb> yeah yeah
2012-09-19T16:37:48 < Laurenceb> shit
2012-09-19T16:37:59 < Laurenceb> i seem to have made a micro Sd card bricking machine
2012-09-19T16:38:46 < Laurenceb> oh
2012-09-19T16:38:52 < Laurenceb> MBR was corrupted
2012-09-19T16:39:06 < Laurenceb> seems my haxor formatting code doesnt work on linux
2012-09-19T16:39:12 < Laurenceb> seemed to work on windows
2012-09-19T16:39:15 < dongs> What people don't appreciate, when they picture Terminator-style automatons striding triumphantly across a mountain of human skulls, is how hard it is to keep your footing on something as unstable as a mountain of human skulls. Most humans probably couldn't manage it, and they've had a lifetime of practice at walking without falling over.
2012-09-19T16:39:28 < Laurenceb> lol
2012-09-19T16:39:31 < dongs> haha
2012-09-19T16:40:03 < Laurenceb> what if i try to format volume in linux...
2012-09-19T16:42:06 < Laurenceb> i guess this is why ST just made it reply with error
2012-09-19T16:42:16 < dongs> heh
2012-09-19T16:43:08 < Laurenceb> yeah format volume works
2012-09-19T16:43:21 < Laurenceb> but then when i try to add a FAT32 partition it fails
2012-09-19T16:43:29 < Laurenceb> Error mounting: mount: /dev/sdc1: can't read superblock
2012-09-19T16:43:40 < Laurenceb> even tho filesystem is clean
2012-09-19T16:43:42 < Laurenceb> weird
2012-09-19T16:45:38 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/usb_scsi.c#L354
2012-09-19T16:45:42 < Laurenceb> wtf did i do wrong
2012-09-19T16:46:21 < Laurenceb> i should overwrite partition table with zero or something?
2012-09-19T16:46:53 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-19T16:47:57 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds]
2012-09-19T16:49:02 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-19T16:49:30 < Laurenceb> http://en.wikipedia.org/wiki/Master_boot_record#PT
2012-09-19T16:49:45 < Laurenceb> wont the partition entry just be overwritten with a new one?
2012-09-19T16:54:19 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-19T16:54:23 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-19T16:56:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-19T17:03:17 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-19T17:15:36 < Laurenceb> very odd - writing MBR record works
2012-09-19T17:15:45 < Laurenceb> and i can check it with card reader as ok
2012-09-19T17:15:52 < Laurenceb> creating a new partition fails
2012-09-19T17:20:26 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has quit [Ping timeout: 245 seconds]
2012-09-19T17:24:10 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-19T17:30:15 -!- Mazingaro [~Tetsuja@host57-4-dynamic.10-87-r.retail.telecomitalia.it] has quit [Ping timeout: 245 seconds]
2012-09-19T17:43:20 <+izua> dongs: http://gushh.net/tmp/arduinoshield.png
2012-09-19T17:45:36 <+Steffanx> LOLOL
2012-09-19T17:46:04 <+Steffanx> Too bad it's some crap connector
2012-09-19T17:55:54 < jpa-> hey, if you are lucky that only blows half of the IO pins
2012-09-19T18:04:57 < Thorn> is this anti-arduino or anti-electronics in general?
2012-09-19T18:05:21 < Thorn> (I mean the message)
2012-09-19T18:10:30 < jpa-> guess twice
2012-09-19T18:10:41 <+Steffanx> yes
2012-09-19T18:10:42 <+Steffanx> no
2012-09-19T18:11:14 < jpa-> yay! you were right!
2012-09-19T18:11:28 <+Steffanx> w00t
2012-09-19T18:26:18 -!- claude_work is now known as claude
2012-09-19T18:30:23 < Laurenceb> i was fixing a heater yesterday
2012-09-19T18:30:34 < Laurenceb> it didnt have a ground wire O_o
2012-09-19T18:33:08 <+Steffanx> ok
2012-09-19T18:36:21 <+Steffanx> I was on the toilet yesteday
2012-09-19T18:36:24 <+Steffanx> no toilet paper
2012-09-19T18:36:31 <+Steffanx> ( random fact )
2012-09-19T18:38:37 < zyp> ok
2012-09-19T18:39:18 <+Steffanx> :P
2012-09-19T18:56:11 -!- DaKu is now known as daku
2012-09-19T18:57:30 -!- dekar [~dekar@dslb-088-069-128-239.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-19T18:58:33 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-19T18:59:16 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-09-19T19:00:07 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-19T19:01:52 < Laurenceb> http://pastebin.com/ErwmkLQG
2012-09-19T19:01:56 < Laurenceb> dafuq
2012-09-19T19:15:01 < zyp> you're using F4?
2012-09-19T19:15:20 < zyp> I thought you only used F1 :p
2012-09-19T19:16:19 < Laurenceb> nope
2012-09-19T19:16:26 < Laurenceb> failing badly
2012-09-19T19:16:40 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-19T19:16:41 < Laurenceb> usb dies on my host unless F4 is help in reset
2012-09-19T19:16:56 < zyp> nice
2012-09-19T19:17:33 < Laurenceb> *held
2012-09-19T19:17:58 < Laurenceb> http://pastebin.com/ppxiqFr5
2012-09-19T19:22:20 < Laurenceb> i suspect sduConfigureHookI  isnt configured somehow
2012-09-19T19:25:25 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-19T19:26:14 < Laurenceb> RAGEEEE
2012-09-19T19:26:30 < Laurenceb> looks like chibios change how USB framework works between versions
2012-09-19T19:26:34 < Laurenceb> this is so stupid
2012-09-19T19:26:41 < Laurenceb> *changed
2012-09-19T19:32:55 < jpa-> you mean between stable and dev versions?
2012-09-19T19:35:02 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 245 seconds]
2012-09-19T19:48:07 < Laurenceb> i dont know
2012-09-19T19:48:11 < Laurenceb> i had some old code
2012-09-19T19:48:17 < Laurenceb> but its just as broken as ever now
2012-09-19T19:49:25 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/main.c#L75
2012-09-19T19:49:29 < Laurenceb> its hardfaulting
2012-09-19T19:49:52 < Laurenceb> wish stlink worked properly with f4discovery
2012-09-19T19:49:59 < Laurenceb> i could work out what was wrong then
2012-09-19T19:50:28 < Laurenceb> have to keep power cycling it to load new code
2012-09-19T19:50:39 < gnomad> the stlink has issues with the F4 discovery?
2012-09-19T19:51:09 < Laurenceb> tons
2012-09-19T19:51:18 < Laurenceb> its pretty unusable :(
2012-09-19T19:51:49 < gnomad> with STs software or OpenOCD or both?
2012-09-19T19:51:56 < Laurenceb> stlink
2012-09-19T19:52:04 < Laurenceb> or texane rather
2012-09-19T19:52:18 < Laurenceb> imn giving up for the day i think
2012-09-19T19:52:20 < Laurenceb> going home
2012-09-19T19:52:22 < gnomad> texane is the open source driver?
2012-09-19T19:52:26 < Laurenceb> yes
2012-09-19T19:52:27 < jkent> i've got a patch for texane's stlink.. i haven't used it enough to share it though
2012-09-19T19:52:33 < jkent> i
2012-09-19T19:52:37 < jkent> i'll send you a patch..
2012-09-19T19:52:44 < Laurenceb> going, sorry
2012-09-19T19:52:49 < Laurenceb> thanx for the offer
2012-09-19T19:52:53 < gnomad> so are the problems in texane, or are they in the h/w design of the stlink?
2012-09-19T19:52:59 < jkent> mmkay
2012-09-19T19:53:27 < jkent> gnomad: nRST never gets asserted, i think its texane
2012-09-19T19:53:32 < gnomad> which is why I asked... do ST's tools exhibit similar problems with the stlink?
2012-09-19T19:53:56 -!- claude is now known as Claude
2012-09-19T19:54:08 < jkent> i don't have enough experience though.. its just something that works for me (tm)
2012-09-19T19:54:21 < jkent> 'ey Claude
2012-09-19T19:54:25 < gnomad> BTW, i've had an F4 board gathering dust for a while now... been waiting for the gnu toolchain to stabilize for it...
2012-09-19T19:54:37 < gnomad> guess that still hasn't heppened yet...
2012-09-19T19:55:04 < Claude> hey jkent
2012-09-19T19:57:31 < jkent> http://pastebin.com/4tb2Du7h -- if anyone wants it, adds the monitor command jtag_reset
2012-09-19T20:08:54 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-19T20:09:01 < zyp> jkent, nice
2012-09-19T20:09:17 < zyp> what about doing a pull request for that?
2012-09-19T20:09:47 < jkent> zyp: its really not a proper fix, nRST probably should be asserted after code is loaded.. i didn't look that far into it
2012-09-19T20:10:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-19T20:11:34 < zyp> still better than nothing, shouldn't hurt?
2012-09-19T20:12:39 < jkent> as long as someone knows they need to manually issue a "mon jtag_reset" lol.. mmkay, i'll get to it here in a bit
2012-09-19T20:16:29 < HTT-Bird> actually, come to think of it, do you think it's better if I put my debug-console UART and my SWD interface on separate connectors?
2012-09-19T20:16:48 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-19T20:16:48 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Client Quit]
2012-09-19T20:17:10 < zyp> I prefer having a standardized debug connector, so yes
2012-09-19T20:22:20 -!- dekar [~dekar@212.255.22.171] has joined ##stm32
2012-09-19T20:22:23 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-19T20:25:00 < HTT-Bird> yeah, it'd make it easier to hook up two adapters at once, too
2012-09-19T20:26:58 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-19T20:27:13 < jpa-> i usually put them on the same one, because i am lazy
2012-09-19T20:27:41 < HTT-Bird> heh
2012-09-19T20:27:49 < jpa-> in fact, all the 3 STM32 projects i have done have all had different kind of debug connector :D
2012-09-19T20:27:53 < HTT-Bird> :P
2012-09-19T20:28:13  * HTT-Bird will have to flywire his BusBlaster anyway, which is why he wants to use a .1 header instead of the .05 that's the standard
2012-09-19T20:31:37 < zyp> flywires gets boring real quick
2012-09-19T20:31:52 < zyp> get/make an adapter to 0.05 and use that
2012-09-19T20:32:58 < HTT-Bird> zyp: well, if I went with the two connectors approach, I'd use a 5 pin header for SWD only and a 4 pin for the UART...
2012-09-19T20:34:15 < jkent> pull request sent
2012-09-19T20:34:35 < zyp> I currently have four different stm32 boards sitting on my desk, two is using the 0.05" connector and two is using the 20-pin 0.1" connector
2012-09-19T20:35:16 < zyp> I certainly like being able to move the debugger around between them by just unplugging and plugging a single connector
2012-09-19T20:35:29 < HTT-Bird> zyp: also: BusBlaster can't drive debug-UART and SWD/JTAG at the same time
2012-09-19T20:35:45 < HTT-Bird> (hardware limitation, the 2nd serial/MPSSE on the FT2232 is taken up by JTAGing the onboard CPLD
2012-09-19T20:36:18 < zyp> and you're sure you're going to keep on using a busblaster forever?
2012-09-19T20:37:48 < HTT-Bird> well, it'll be a while before I go hunting for a more sophisticated JTAG dongle :P
2012-09-19T20:39:27 < HTT-Bird> the problem with using EXTa/EXTb on a 10pin .05" header is that I'd need to split it out to two different things
2012-09-19T20:39:40 < HTT-Bird> a JTAG dongle and a separate USB-TTL serial adapter
2012-09-19T20:39:55 < HTT-Bird> well, SWD dongle, but w/e
2012-09-19T20:40:07 < jkent> that black magic probe caught my eye
2012-09-19T20:40:21 < zyp> jkent, I love mine
2012-09-19T20:40:44 < HTT-Bird> I could use the 10pin header, leave EXTa/EXTb unconnected
2012-09-19T20:41:11 < zyp> HTT-Bird, what's EXTa/EXTb?
2012-09-19T20:41:59 < HTT-Bird> zyp: an option on the ARM JTAG headers (check the ARM docs on the pinout for details)
2012-09-19T20:42:19 < HTT-Bird> but then, what would be a good choice for TXD/RXD/Vref/GND TTL serial?
2012-09-19T20:43:36 < zyp> ah, it's what would be TDI and TDI
2012-09-19T20:43:39 < zyp> TDO*
2012-09-19T20:45:26 < HTT-Bird> I'm thinking a 4 pin .1 header, but correct me if there's a better/more standard choice out there
2012-09-19T20:46:05 < zyp> not that I know of
2012-09-19T20:46:34 < zyp> the most standard in the hobbyist world would be the 6-pin pinout the ftdi cables use
2012-09-19T20:47:38 -!- daku is now known as DaKu
2012-09-19T21:00:21 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-09-19T21:12:55 -!- barthess [~barthess@164.177.210.72] has joined ##stm32
2012-09-19T21:22:10 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-19T21:24:05 <+dekar> apparently my mints have a "party in my pocket" O.o http://www.blinkmints.com/
2012-09-19T21:34:59 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-19T21:42:33 -!- DaKu is now known as daku
2012-09-19T21:46:05 -!- daku is now known as DaKu
2012-09-19T21:49:21 -!- Claude is now known as claude
2012-09-19T22:23:58 < TitanMKD> photo of my STM32F4 Debug v1.0
2012-09-19T22:24:00 < TitanMKD> http://postimage.org/image/8unt3sf2n/
2012-09-19T22:24:21 < TitanMKD> Itead have done a very good work (except the silkscreen which is not perfect but good)
2012-09-19T22:25:46 < jpa-> what does it do?
2012-09-19T22:26:24 < prog> looks nice
2012-09-19T22:26:28 < TitanMKD> all see here http://titanmkd.blogspot.fr/
2012-09-19T22:26:40 < zyp> I wonder how a bad silkscreen would look if you call that good :p
2012-09-19T22:26:58 < TitanMKD> it is intended to be used as an advanced Bus pirate with high speed feature and more peripherals supported
2012-09-19T22:27:03 < jpa-> yeah, silk screen alignment is crap
2012-09-19T22:27:26 < TitanMKD> zyp yes the silkscreen is weird ;) but the PCB trace are very good
2012-09-19T22:27:45 < zyp> by what measure?
2012-09-19T22:27:48 < TitanMKD> and look here all in one ;)
2012-09-19T22:27:48 < TitanMKD> http://postimage.org/image/oe0j9axh5/
2012-09-19T22:28:00 < TitanMKD> zyp the size is 60mm x 37mm
2012-09-19T22:28:03 < TitanMKD> very little
2012-09-19T22:28:29 < TitanMKD> pratically 2 time less than a credit card
2012-09-19T22:28:41 < TitanMKD> TRF7970A is just a bit bigger than a credit card
2012-09-19T22:29:04 < jpa-> but its within their spec, so the traces are just what should be expected
2012-09-19T22:29:09 < jpa-> are they 0.2mm or 0.15mm btw?
2012-09-19T22:29:26 < zyp> I mean, what are you comparing it to when you claim it is good?
2012-09-19T22:30:11 < TitanMKD> zyp i compare it to some photo of Itead PCB which was really ugly with hole totally too big not centered ...
2012-09-19T22:30:29 < zyp> like the one close to the C8 silk?
2012-09-19T22:30:55 < zyp> the other vias doesn't look centered either
2012-09-19T22:31:21 < jpa-> the hole alignment looks good for itead, but otherwise not particularly great
2012-09-19T22:31:24 < TitanMKD> yes but i have see some card very ugly ;)
2012-09-19T22:31:39 < jpa-> seeed seems to do better, even though they are supposed to use the same fab
2012-09-19T22:31:50 < TitanMKD> but we can clearly see the etest on each pad ;)
2012-09-19T22:32:07 < TitanMKD> it was not the case on a previous batch i bought
2012-09-19T22:33:33 < jpa-> true, it's nice that they atleast sometimes do it :P
2012-09-19T22:35:01 < TitanMKD> for a batch of 10 board it is ok ;)
2012-09-19T22:35:17 < TitanMKD> I just hope there's no short or problem on trace now
2012-09-19T22:39:35 < TitanMKD> strange the bottom layer silkscreen is really better
2012-09-19T22:40:05 < TitanMKD> and the photo is with an optical zoom >8 ;)
2012-09-19T23:06:03 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-19T23:06:33 < Laurenceb_> hi
2012-09-19T23:08:10 < Laurenceb_> has anyone got F4Discovery debug working correctly with texane?
2012-09-19T23:08:23 < Laurenceb_> or it that impossible due to lack of reset line?
2012-09-19T23:08:37 < Laurenceb_> or.... do i need a funky gdbinit script?
2012-09-19T23:13:28 < zyp> huh?
2012-09-19T23:13:40 < zyp> always more or less worked for me
2012-09-19T23:16:00 < Laurenceb_> i have to power cycle to run the code
2012-09-19T23:16:47 -!- barthess [~barthess@164.177.210.72] has quit [Quit: Leaving.]
2012-09-19T23:31:46 < Laurenceb_> do you have a special gdbinit trick?
2012-09-19T23:38:13 < zyp> no?
2012-09-19T23:39:01 < zyp> please pastebin your gdb log, I'll take a look and see if I can see something weird
2012-09-19T23:40:21 < Laurenceb_> i dont have hardware setup here
2012-09-19T23:40:28 < Laurenceb_> basically it will flash ok
2012-09-19T23:40:36 < Laurenceb_> cut kill and run dont do anything
2012-09-19T23:40:47 < Laurenceb_> the cpu seems to be locked after flashing
2012-09-19T23:40:55 < Laurenceb_> so i have to power cycle to run the code
2012-09-19T23:41:01 < Laurenceb_> reset doesnt solve it either
2012-09-19T23:41:24 < zyp> I just flash and run without any problems
2012-09-19T23:41:35 < Laurenceb_> hmm
2012-09-19T23:41:43 < Laurenceb_> what is in your gdbinit?
2012-09-19T23:42:22 < zyp> http://cgit.jvnv.net/suzumebachi/tree/.gdbinit?id=2a3413
2012-09-19T23:43:04 < TitanMKD> anyone have seen the Teensy 3.0 ?
2012-09-19T23:43:26 < TitanMKD> why to choose an ARM4 Kinetis from freescale i really do not understand the choice of that MCU
2012-09-19T23:43:34 < zyp> everybody have seen it.
2012-09-19T23:43:43 < Laurenceb_> odd
2012-09-19T23:43:45 < TitanMKD> for me it is the worst choice ever
2012-09-19T23:43:57 < Laurenceb_> latest texane and F4 discovery?
2012-09-19T23:44:02 < TitanMKD> STM32F4 or LPCxx are really better and pratically for the same cost
2012-09-19T23:44:03 < Laurenceb_> with onboard stlink?
2012-09-19T23:44:13 < zyp> Laurenceb_, no, I haven't used it for a while
2012-09-19T23:44:18 < Laurenceb_> hmm
2012-09-19T23:44:23 < Laurenceb_> this is very odd
2012-09-19T23:44:53 < Laurenceb_> ill have t have a play tomorrow
2012-09-19T23:45:03 < zyp> the last version I was using was the one where I added the flashloader
2012-09-19T23:45:10 < zyp> I haven't tested any later versions
2012-09-19T23:45:21 < Laurenceb_> ok
2012-09-19T23:54:25 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
--- Day changed Thu Sep 20 2012
2012-09-20T00:05:13 <+Steffanx> http://www.thesun.co.uk/sol/homepage/news/4545386/Record-is-a-load-of-pants.html your dad Laurenceb_ ?
2012-09-20T00:23:26 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-20T00:59:55 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-20T01:11:05 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T01:21:54 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Ping timeout: 240 seconds]
2012-09-20T01:23:27 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-20T01:28:26 -!- DaKu is now known as daku
2012-09-20T01:40:05 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-20T01:43:53 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-20T01:49:12 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-20T02:16:44 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 244 seconds]
2012-09-20T02:44:40 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-20T02:44:40 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-20T02:44:40 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-20T03:01:21 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-20T03:14:24 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-20T03:18:25 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-20T03:19:03 < Thorn> I had the same reset after load problem with f4discovery & texane
2012-09-20T03:22:52 < dongs> blog after blog
2012-09-20T03:26:16 < dongs> http://xkcd.com/1110/ fuuuuuuuuuuuuuuuuuu
2012-09-20T03:26:49 < Thorn> http://www.electricstuff.co.uk/milboards.html
2012-09-20T03:28:35 < dongs> lol vax
2012-09-20T03:33:20 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 252 seconds]
2012-09-20T03:45:21 < Thorn> wth is "battle short"? http://www.electricstuff.co.uk/Image4.jpg
2012-09-20T03:56:56 < dongs> mayb its like
2012-09-20T03:57:05 < dongs> fight of the short circuits
2012-09-20T04:00:32 < Thorn> http://firecontrolman.tpub.com/14100/css/14100_23.htm
2012-09-20T04:02:03 < Thorn> and http://www.ruggedsystems.com/UPS_Products.htm
2012-09-20T04:02:38 < BrainDamage> dongs:  http://xkcd-map.rent-a-geek.de/
2012-09-20T04:03:57 < dongs> aww so its not all that big
2012-09-20T04:43:35 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 255 seconds]
2012-09-20T04:47:37 -!- Thorn__ [~Thorn@176.15.67.62] has joined ##stm32
2012-09-20T04:47:37 -!- Thorn__ [~Thorn@176.15.67.62] has quit [Changing host]
2012-09-20T04:47:37 -!- Thorn__ [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-20T04:48:10 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-20T04:50:10 -!- Thorn__ is now known as Thorn
2012-09-20T04:56:42 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-20T05:09:54 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-20T05:14:53 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-20T05:14:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T05:32:17 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-09-20T05:42:35 -!- Netsplit *.net <-> *.split quits: nashpa, mrcan, R0b0t1, emeb_mac, Oldboy, esden
2012-09-20T05:42:55 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T05:42:55 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-20T05:42:55 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-20T05:48:32 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-09-20T05:48:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T05:48:32 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-20T05:48:32 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-20T05:48:32 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-20T05:59:54 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T06:05:22 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-20T06:06:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T06:09:42 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-20T06:32:25 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-20T06:32:31 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-20T06:53:54 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-09-20T06:54:11 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-20T07:07:50 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 268 seconds]
2012-09-20T07:09:03 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T07:16:04 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-20T07:22:58 < emeb_mac> ssshhhhh.... be vewy quiet - I'm hunting wabbits.
2012-09-20T07:28:17 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 244 seconds]
2012-09-20T07:31:22 < dongs> more like hunting dongs
2012-09-20T07:34:30 < emeb_mac> speak of the devil
2012-09-20T07:43:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-20T08:03:23 -!- claude is now known as Claude
2012-09-20T08:04:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T08:06:40 < jpa-> is monitoring vbus the best way to detect if an usb host is connected?
2012-09-20T08:09:51 < HTT-Bird> jpa-: if you're the device? yeah, you definitely want to keep an eye on vbus
2012-09-20T08:10:08 < HTT-Bird> pushing current down the bus with Vbus=0 will get you tutted at all over the place
2012-09-20T08:10:22 < jpa-> what if i'm connected to dummy usb charger?
2012-09-20T08:10:39 < HTT-Bird> jpa-: as in something that is just providing Vbus/GND and not an actual host?
2012-09-20T08:10:50 < jpa-> yeah
2012-09-20T08:11:29 < jpa-> i wonder if enabling the internal weak pull-up on D+ and D-, and then checking if the host pull-downs show up
2012-09-20T08:11:50 < jpa-> (i.e. not the usb pull-up, but the GPIO one)
2012-09-20T08:12:04 < HTT-Bird> jpa-: you can't do that with Vbus=0
2012-09-20T08:12:18 < HTT-Bird> if you have Vbus though you might.  I don't know for sure.
2012-09-20T08:13:25 < jpa-> ah.. so you are thinking of a situation when there is a host connected but it is powered down?
2012-09-20T08:14:47 < HTT-Bird> jpa-: yep
2012-09-20T08:14:58 < HTT-Bird> you cannot force any current back down the USB bus in that circumstance
2012-09-20T08:15:21 < jpa-> ok
2012-09-20T08:15:32 < jpa-> i guess i'll have to make up some space for vbus trace then :D
2012-09-20T08:42:40 < dongs> just put a voltage divider on vbus
2012-09-20T08:42:49 < dongs> and stick it to some gpio
2012-09-20T08:53:21 < jpa-> yeah, did that
2012-09-20T08:53:42 < jpa-> 1/2 divider
2012-09-20T08:53:54 < jpa-> (with 10kohm - 10kohm resistors)
2012-09-20T08:54:21 < dongs> i think thats what I did on my usb shit.
2012-09-20T08:54:25 < dongs> (with 10k)
2012-09-20T09:01:08 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-20T09:16:28 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-09-20T09:23:40 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-20T09:27:13 -!- Claude is now known as claude
2012-09-20T10:14:41 < zyp> jpa-, detecting if an usb host is connected, for what purpose?
2012-09-20T10:17:36 < zyp> the host will pull both lines low, which will generate a usb reset interrupt from your device controller
2012-09-20T10:21:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-20T10:30:54 < Thorn> that assumes it's initialized
2012-09-20T10:31:27 < Thorn> which is not true for my CAN device because CAN conflicts with USB on f103
2012-09-20T10:31:45 < Thorn> so I have to detect vbus, disable can and enable usb
2012-09-20T10:32:11 < Thorn> but that's hardly a normal situation
2012-09-20T11:05:07 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-20T11:05:42 -!- claude is now known as claude_work
2012-09-20T12:09:54 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-20T12:24:53 < jpa-> zyp: mostly because when USB is connected, i want to disable normal user interface (to prevent file system corruption if host writes to it)
2012-09-20T12:25:28 < jpa-> so i'd like to get a signal early enough that i can sync filesystem, and only then enumerate
2012-09-20T12:26:06 < dongs> sync them dongs
2012-09-20T12:26:57 < Laurenceb> hi
2012-09-20T12:27:16  * Laurenceb returns to trying to get f4discovery working
2012-09-20T12:28:16 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T12:28:19 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-20T12:28:20 < dongs> izua: great link
2012-09-20T12:29:05 < Laurenceb> bt
2012-09-20T12:29:05 < Laurenceb> #0  _unhandled_exception ()
2012-09-20T12:29:05 < Laurenceb>     at ../ChibiOS/os/ports/GCC/ARMCMx/STM32F4xx/vectors.c:178
2012-09-20T12:29:05 < Laurenceb> #1  <signal handler called>
2012-09-20T12:29:05 < Laurenceb> #2  0xfffffffe in ?? ()
2012-09-20T12:29:06 < Laurenceb> #3  0xffffffff in ?? ()
2012-09-20T12:29:07 < Laurenceb> Backtrace stopped: frame did not save the PC
2012-09-20T12:29:12 < Laurenceb> ^dafuk
2012-09-20T12:29:18 < dongs> at ../ChibiOS/os/ports/
2012-09-20T12:29:19 < dongs> found the problem
2012-09-20T12:31:08 < Laurenceb> if i dissasemble,
2012-09-20T12:31:09 < Laurenceb> Disassembly of section .text:
2012-09-20T12:31:09 < Laurenceb> 08000190 <memset>:
2012-09-20T12:31:16 < Laurenceb> ^why does ti start there?!
2012-09-20T12:32:15 <+izua> .comefrom 08000000
2012-09-20T12:32:19 <+izua> 08000190 <memset>:
2012-09-20T12:37:57 < Laurenceb> izua: i dont follow
2012-09-20T12:40:13 < Laurenceb> *rage*
2012-09-20T12:40:26 < Laurenceb> i thought it might be the linker script
2012-09-20T12:40:35 < Laurenceb> but all chibios linker scripts give same result
2012-09-20T12:40:39 < Laurenceb> wtf is going on
2012-09-20T12:49:58 < dongs> Laurenceb: have you considered switching to COX
2012-09-20T12:50:44 < Laurenceb> i am now
2012-09-20T12:50:51 < Laurenceb> havent a clue whats wrong here
2012-09-20T12:51:05 < zyp> jpa-, that sounds pointless
2012-09-20T12:51:07 < Laurenceb> cant even work out where vector table went
2012-09-20T12:51:46 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/main.c#L75
2012-09-20T12:51:49 < Laurenceb> ^my code
2012-09-20T12:52:10 < zyp> jpa-, usb control commands are blocking, so you could sync filesystem during the enumeration, not before it
2012-09-20T12:52:27 < Laurenceb> weird thing is the example code works
2012-09-20T12:52:31 < Laurenceb> or did work
2012-09-20T12:52:35 < Laurenceb> with old chibios
2012-09-20T12:52:40 < zyp> either in the SET_CONFIGURATION command or in some MSC specific command
2012-09-20T12:53:17 < zyp> I believe you've got five seconds to perform a SET_CONFIGURATION before you have to return, should be plenty of time to sync?
2012-09-20T12:56:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-20T12:57:37 < Laurenceb> oh it works
2012-09-20T12:57:38 < Laurenceb> wtf
2012-09-20T12:57:55 < Laurenceb> some tard told me F4discovery had OTG1 connected
2012-09-20T12:57:58 < Laurenceb> its OTG2
2012-09-20T13:07:35 < dongs> I should bring the guy doing lordpil.com in here
2012-09-20T13:07:40 < dongs> and do laurenceb.com
2012-09-20T13:07:49 < dongs> call it diary of serial failure
2012-09-20T13:09:35 < dongs> http://www.lordpil.com/blog/archives/2012-09-13#947
2012-09-20T13:10:24 <+izua> dongs: that is the most retarded thing i've seen in quite a while
2012-09-20T13:10:34 <+izua> i get her and her band people beers
2012-09-20T13:10:34 <+izua> for an accordian girl pinup calandar
2012-09-20T13:10:34 <+izua> also i bought this art book 'a gasmask a day project'
2012-09-20T13:10:34 <+izua> is like maybe 50 pages of mostly ink and watercolor drawings of chicks in gasmasks
2012-09-20T13:10:34 <+izua> and one squirrel
2012-09-20T13:10:35 <+izua> with biohazard acorns
2012-09-20T13:12:02 < dongs> izua: wat
2012-09-20T13:12:13 <+izua> lordpil
2012-09-20T13:12:18 < dongs> oh haha
2012-09-20T13:12:24 < dongs> tehres some hilarious shit on there
2012-09-20T13:12:34 <+izua> it looks like the lame retarded brother of bash.org
2012-09-20T13:12:39 < dongs> http://lordpil.com/blog/archives/2012-09-06
2012-09-20T13:13:19 < dongs> lordpil on DIY lcd projectors http://lordpil.com/blog/archives/2012-09-04
2012-09-20T13:14:32 < dongs> http://www.lordpil.com/blog/archives/2012-08-06#857 haha
2012-09-20T13:14:42 <+izua> i know a guy who made a projector that way :P
2012-09-20T13:14:53 <+izua> but you know, light source, lcd, lens
2012-09-20T13:15:15 <+izua> bigass 1KW mercury bulb
2012-09-20T13:15:16 <+izua> but hey
2012-09-20T13:15:18 <+izua> it worked
2012-09-20T13:15:54 < Laurenceb> ok so wtf
2012-09-20T13:16:02 < Laurenceb> /dev/ttyACM0 is present
2012-09-20T13:16:20 < Laurenceb> but its stuck at  chThdCreateStatic(waThread1, sizeof(waThread1), NORMALPRIO, Thread1, NULL);
2012-09-20T13:16:30 < Laurenceb> to spawn a led blink thread
2012-09-20T13:17:50 < Laurenceb> dafuk
2012-09-20T13:18:02 < Laurenceb> works if i start thread before configuring usb
2012-09-20T13:18:30 <+izua> does the blinkage threard still work after configuring usb?
2012-09-20T13:19:13 <+izua> dongs: http://www.lordpil.com/blog/archives/2012-08-22#843
2012-09-20T13:19:34 <+izua> so who is this lordpil guy? he makes cncs at least
2012-09-20T13:21:35 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T13:27:47 < Laurenceb> no data out of usb tho
2012-09-20T13:34:01 < Laurenceb> i just put in a loop like chprintf(USBout, "Firmware compiled __DATE__, running ChibiOS\r\n");
2012-09-20T13:34:06 < Laurenceb> with a delay
2012-09-20T13:34:16 < Laurenceb> screen /dev/ttyACM0 gives nothing
2012-09-20T13:36:05 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-20T13:37:29 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-20T13:39:48 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 248 seconds]
2012-09-20T13:42:30 < Laurenceb> ok i think texane is failing
2012-09-20T13:43:19 < Laurenceb> think ill go back to F1
2012-09-20T13:43:26 < Laurenceb> it has the advantage of working
2012-09-20T13:47:43 < zyp> did you test the revision I was using?
2012-09-20T13:47:56 < Laurenceb> nope
2012-09-20T13:48:02 < Laurenceb> it does seem to be working
2012-09-20T13:48:08 < Laurenceb> its just unreliable
2012-09-20T13:48:19 < Laurenceb> but something is very wrong with usb
2012-09-20T13:48:26 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/main.c
2012-09-20T13:48:32 < Laurenceb> i just copied the example
2012-09-20T13:49:08 < Laurenceb> theres nothing output
2012-09-20T13:50:06 < Laurenceb> i can screen /dev/ttyACM0
2012-09-20T13:50:15 < Laurenceb> its enumerating etc
2012-09-20T13:50:30 < Laurenceb> but there is no data
2012-09-20T13:50:53 < zyp> anything fun in dmesg?
2012-09-20T13:52:17 < Laurenceb> cdc_acm 6-2:1.0: This device cannot do calls on its own. It is not a modem.
2012-09-20T13:52:23 < Laurenceb> the only interesting bit
2012-09-20T13:52:52 < Laurenceb> followed by
2012-09-20T13:52:53 < Laurenceb> cdc_acm 6-2:1.0: ttyACM0: USB ACM device
2012-09-20T13:53:44 < Laurenceb> maybe some config file is wrong...
2012-09-20T13:54:58 < Laurenceb> can you clone my project and try and an F4?
2012-09-20T13:55:01 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-20T13:55:02 < Laurenceb> *on an F4
2012-09-20T13:57:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-20T13:57:37 < dongs> gay4
2012-09-20T13:57:42 < dongs> do you have sd card on sdio?
2012-09-20T13:57:49 < dongs> i have f4discovery but sdcard is on 4bit sdio
2012-09-20T13:57:59 < dongs> and im not moving it off there.
2012-09-20T13:58:05 < Laurenceb> no
2012-09-20T13:58:23 < dongs> fuuuuuuuuuuuuuuuuuuuuuurget it then.
2012-09-20T13:58:24 < Laurenceb> its just usb atm
2012-09-20T13:58:31 < dongs> oh
2012-09-20T13:58:32 < TitanMKD> dongs have you checked performance of SDCard (a fast version) with SDIO 4bits on F4 ?
2012-09-20T13:58:41 < Laurenceb> theres nothing else going on
2012-09-20T13:58:46 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-20T13:58:46 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-20T13:58:46 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T13:58:48 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-20T13:58:52 < dongs> TitanMKD: yeah, its fucking fast
2012-09-20T13:58:59 < TitanMKD> dongs like 10MB/s ?
2012-09-20T13:59:03 < dongs> 20+meg/sec
2012-09-20T13:59:03 < dongs> thats megaBYTEs not bits
2012-09-20T13:59:13 < TitanMKD> 20MB/s !!!
2012-09-20T13:59:13 < dongs> more than that.
2012-09-20T13:59:16 < dongs> yea.
2012-09-20T13:59:20 < TitanMKD> woo very interesting
2012-09-20T13:59:26 < TitanMKD> faster than on my pandaboard ;)
2012-09-20T13:59:34 < dongs> ur pandaboard runs liunix
2012-09-20T13:59:40 < TitanMKD> I planned to test SDIO 8bits using MMC card
2012-09-20T13:59:48 < Laurenceb> pedoboard
2012-09-20T13:59:50 < TitanMKD> IIRC it is 48MB/s the max possible speed
2012-09-20T13:59:50 < dongs> of course its gonna be slow shit
2012-09-20T14:00:16 < TitanMKD> dongs but pandaboard run 2 CPU @ 1GHZ ;)
2012-09-20T14:00:25 < TitanMKD> anyway the driver on linux suxx ...
2012-09-20T14:00:38 < dongs> 8bits is kinda pointless.
2012-09-20T14:00:43 < TitanMKD> they are using polling mode ...
2012-09-20T14:00:48 < dongs> too much pins wasted
2012-09-20T14:00:48 < dongs> 4bit is fine
2012-09-20T14:00:48 < dongs> and performance is fine
2012-09-20T14:00:59 < TitanMKD> dongs yes anyway 4bits at 20MB/s is very good
2012-09-20T14:01:40 < TitanMKD> my idea is to use as test the 8bits mode and to link it to an FTDI FT2232H to reach USB HS ;)
2012-09-20T14:01:53 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T14:01:54 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-20T14:02:00 < Laurenceb> #chibios is dead :(
2012-09-20T14:02:05 < TitanMKD> Laurenceb why ?
2012-09-20T14:02:11 < TitanMKD> ha the chan ;)
2012-09-20T14:02:12 < Laurenceb> no relies
2012-09-20T14:02:16 -!- izua [~izua@unaffiliated/izua] has quit [Client Quit]
2012-09-20T14:02:17 < Laurenceb> *replies
2012-09-20T14:02:23 < Laurenceb> my usb is fuxored
2012-09-20T14:02:34 < Laurenceb> theres no data coming through
2012-09-20T14:02:40 < TitanMKD> Laurenceb ask on chibios forum
2012-09-20T14:02:47 < TitanMKD> Laurenceb it is very active
2012-09-20T14:02:52 < Laurenceb> yeah but that'll take all day still
2012-09-20T14:03:09 < TitanMKD> hehe
2012-09-20T14:03:31 < TitanMKD> else i have just received all components from mouser to build my STM32F4 Debug + LPC4350 LBGA256 ;)
2012-09-20T14:03:37 < dongs> TitanMKD: http://dl.dropbox.com/u/34019941/images/2011/20111111/20111111_006.png
2012-09-20T14:04:05 < TitanMKD> dongs wooo !!
2012-09-20T14:04:22 < TitanMKD> dongs what SD card have you used ?
2012-09-20T14:04:43 < Laurenceb> lol @ font
2012-09-20T14:04:52 <+izua_> holy shit
2012-09-20T14:05:13 < TitanMKD> dongs in addition it use FatFS !! so the speed on raw block can be even a bit faster
2012-09-20T14:05:42 < BrainDamage> dongs: needs more comic sans
2012-09-20T14:05:44 < TitanMKD> dongs and with a good sdcard you an also reach 21MB/s write speed
2012-09-20T14:06:02 < dongs> http://maagpw.bay.livefilestore.com/y1pykL_7PGqXWzUabrjApBIkIhISPm47zcVmZsf4jSEKppiOIAfPo8GTO4MziZQPzmAEBK4KhqQhuzw0EmI3OM2AkIGZqdqoNO9/20110617_001.png?psid=1
2012-09-20T14:06:02 < dongs> F2
2012-09-20T14:06:02 < dongs> ugh what a shit u rl
2012-09-20T14:06:02 < dongs> http://maagpw.bay.livefilestore.com/y1pzro6WIC6YsTHyl03bYt63FuTIFY4dAmq2DB0Dc70gFNV0qTFKTQsVQRiCA31oR6yUzaFX_I0Gq2JlU3P_Df4CbRTP587PLlc/20110530_006.png?psid=1
2012-09-20T14:06:03 < dongs> F1 SDIO
2012-09-20T14:06:11 < TitanMKD> dongs I have a Sandisk Extreme 30MB/s Class 10
2012-09-20T14:06:35 <+izua_> dongs: last pic is f1?
2012-09-20T14:06:49 < TitanMKD> dongs ha yes 24MHz twice less it is logic
2012-09-20T14:07:06 < TitanMKD> dongs anyway DMA really work fine i imagine using polling it will reach 3MB/s ;)
2012-09-20T14:08:07 < TitanMKD> dongs are you using chibios for that ?
2012-09-20T14:08:13 < TitanMKD> or your own driver ?
2012-09-20T14:08:31 < dongs> izua_: yes
2012-09-20T14:08:34 < dongs> TitanMKD: of course not
2012-09-20T14:08:57 < dongs> just fatfs + mostly unmodified ST SDIO code
2012-09-20T14:09:06 < dongs> fatfs thing has a terminal sample shiz
2012-09-20T14:09:09 < TitanMKD> ha ok
2012-09-20T14:09:15 < dongs> its all basically sample code, so it just shows what hardware is capable of
2012-09-20T14:09:23 < TitanMKD> does your example is available on github or other ?
2012-09-20T14:09:35 < TitanMKD> just to compare later with chibios SDIO driver
2012-09-20T14:10:02 < dongs> i gave it to Laurenceb before..
2012-09-20T14:10:39 < dongs> ohgawd
2012-09-20T14:11:50 < dongs> Laurenceb: haha the jap blogged about you
2012-09-20T14:11:58 < Laurenceb> where?
2012-09-20T14:12:00 < dongs> Laurenceb: you linked http://dl.dropbox.com/u/34019941/images/2011/20111016/20111016_001.jpg to high altitude
2012-09-20T14:12:26 < dongs> and he was like
2012-09-20T14:12:27 < dongs> that gayjin called my shit hentai
2012-09-20T14:12:36 < dongs> 'worlds most complex hentai viewer'
2012-09-20T14:12:37 < Laurenceb> lol
2012-09-20T14:12:39 < dongs> or wahever your comment was
2012-09-20T14:13:34 < Laurenceb> http://gayjin.info/los-angeles
2012-09-20T14:14:17 < dongs> haha
2012-09-20T14:14:33 < Laurenceb> http://en.wikipedia.org/wiki/Fritz_the_Cat_%28film%29
2012-09-20T14:14:36 < Laurenceb> o...k...
2012-09-20T14:15:01 < Laurenceb> - wikipedia example of early hentai
2012-09-20T14:18:21 < dongs> still laughing @ hentai viewer
2012-09-20T14:19:53 < dongs> http://dl.dropbox.com/u/34019941/images/2012/20120607/20120607_002.jpg nice evalboard
2012-09-20T14:28:47 < dongs> http://www.electricstuff.co.uk/laserc13.jpg "From the Ebay listing : "Adopt Japan imported advanced motherboard and microchip" - if that control board has been anywhere near Japan then  I'm a Chinaman."
2012-09-20T14:28:52 < dongs> lool
2012-09-20T14:36:07 < TitanMKD> bye
2012-09-20T14:36:13 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-20T14:39:54 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 265 seconds]
2012-09-20T14:40:03 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 255 seconds]
2012-09-20T14:40:53 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T14:45:48 < jpa-> heh.. looks like the RTC calendar on stm32f4 is not prepared for 2k1
2012-09-20T14:45:52 < jpa-> *year 2k1
2012-09-20T14:45:53 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-20T14:46:32 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T14:47:13 < Laurenceb> oh noes
2012-09-20T14:51:42 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 264 seconds]
2012-09-20T14:53:00 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T14:57:33 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-20T14:58:22 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T15:01:22 -!- daku is now known as DaKu
2012-09-20T15:01:56 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-09-20T15:04:36 < Laurenceb> aha
2012-09-20T15:04:41 < Laurenceb> chibios fixed
2012-09-20T15:04:51 < Laurenceb> i fail at rtos programming
2012-09-20T15:05:03 < Laurenceb> i used a while loop to check for connection
2012-09-20T15:05:22 < Laurenceb> due to the thread having high priority it would never yield
2012-09-20T15:05:40 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-20T15:05:49 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T15:06:57 <+izua_> dongs: +1 on the lordpil idea
2012-09-20T15:08:25 -!- DaKu is now known as daku
2012-09-20T15:10:53 < Laurenceb> maybe i should shut up
2012-09-20T15:11:16 -!- daku is now known as DaKu
2012-09-20T15:12:13 < dongs> haha
2012-09-20T15:13:25 < Laurenceb> We're no strangers to love
2012-09-20T15:13:34 < Laurenceb> You know the rules and so do I
2012-09-20T15:13:47 < Laurenceb> A full commitment's what I'm thinking of
2012-09-20T15:13:47 < Laurenceb> You wouldn't get this from any other guy
2012-09-20T15:13:47 < Laurenceb> I just wanna tell you how I'm feeling
2012-09-20T15:13:47 < Laurenceb> Gotta make you understand
2012-09-20T15:13:47 < Laurenceb> Never gonna give you up
2012-09-20T15:13:48 < Laurenceb> Never gonna let you down
2012-09-20T15:13:50 < Laurenceb> Never gonna run around and desert you
2012-09-20T15:13:52 < Laurenceb> Never gonna make you cry
2012-09-20T15:13:54 < Laurenceb> Never gonna say goodbye
2012-09-20T15:13:56 < Laurenceb> Never gonna tell a lie and hurt you
2012-09-20T15:13:58 < Laurenceb> We've known each other for so long
2012-09-20T15:14:00 < Laurenceb> Your heart's been aching but
2012-09-20T15:14:02 < Laurenceb> You're too shy to say it
2012-09-20T15:14:04 < Laurenceb> Inside we both know what's been going on
2012-09-20T15:14:06 < Laurenceb> We know the game and we're gonna play it
2012-09-20T15:14:08 < Laurenceb> And if you ask me how I'm feeling
2012-09-20T15:14:10 < Laurenceb> Don't tell me you're too blind to see
2012-09-20T15:14:12 < Laurenceb> Never gonna give you up
2012-09-20T15:14:16 < Laurenceb> Never gonna let you down
2012-09-20T15:14:18 < Laurenceb> Never gonna run around and desert you
2012-09-20T15:14:20 < Laurenceb> Never gonna make you cry
2012-09-20T15:14:22 < Laurenceb> Never gonna say goodbye
2012-09-20T15:14:24 < Laurenceb> Never gonna tell a lie and hurt you
2012-09-20T15:14:26 < Laurenceb> Never gonna give you up
2012-09-20T15:14:28 < zyp> oh shut up
2012-09-20T15:14:28 < Laurenceb> Never gonna let you down
2012-09-20T15:14:30 < Laurenceb> Never gonna run around and desert you
2012-09-20T15:14:32 < Laurenceb> Never gonna make you cry
2012-09-20T15:14:37 -!- mode/##stm32 [+q *!*@vlsi1.eee.nottingham.ac.uk] by ChanServ
2012-09-20T15:14:41 <+izua_> NO.jpg
2012-09-20T15:14:52 -!- mode/##stm32 [-q *!*@vlsi1.eee.nottingham.ac.uk] by ChanServ
2012-09-20T15:14:52 < Laurenceb> (Ooh)
2012-09-20T15:14:54 < Laurenceb> Never gonna give, never gonna give
2012-09-20T15:14:56 < Laurenceb> (Give you up)
2012-09-20T15:14:58 < Laurenceb> We've know each other for so long
2012-09-20T15:15:00 < Laurenceb> Your heart's been aching but
2012-09-20T15:15:02 < Laurenceb> You're too shy to say it
2012-09-20T15:15:07 -!- mode/##stm32 [+q *!*@vlsi1.eee.nottingham.ac.uk] by ChanServ
2012-09-20T15:15:40 -!- mode/##stm32 [-q *!*@vlsi1.eee.nottingham.ac.uk] by ChanServ
2012-09-20T15:15:40 < Laurenceb> Never gonna let you down
2012-09-20T15:15:42 < Laurenceb> Never gonna run around and desert you
2012-09-20T15:15:46 < Laurenceb> Never gonna make you cry
2012-09-20T15:15:48 < Laurenceb> Never gonna say goodbye
2012-09-20T15:15:50 < Laurenceb> Never gonna tell a lie and hurt you
2012-09-20T15:15:57 < dongs> http://www.digikey.com/product-detail/en/PRQC8.00CR5010X000/478-4584-1-ND/1645895
2012-09-20T15:16:00 < dongs> is this a Laurenceb approved resonator
2012-09-20T15:16:26 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-20T15:16:31 <+izua_> that's a smart spam script
2012-09-20T15:16:34 <+izua_> if it knows +q
2012-09-20T15:16:36 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T15:16:54 < zyp> isn't ±0.5% as bad as HSI?
2012-09-20T15:16:54 <+izua_> but i think you flooded dongs 57.6k modem off with too much rickrolling
2012-09-20T15:17:14 <+izua_> it's not as bad as those ceramic 3 pin ones
2012-09-20T15:17:28 <+izua_> oh wait. it is!
2012-09-20T15:17:54 < zyp> no point having external resonator if it's not better than HSI
2012-09-20T15:19:44 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-20T15:21:36 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-20T15:21:52 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T15:24:30 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-20T15:32:28 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 246 seconds]
2012-09-20T15:35:05 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-20T15:43:38 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T15:45:28 < dongs> feenode sucks so fucking bad
2012-09-20T15:45:46 < zyp> ok
2012-09-20T15:50:10 -!- dongle [~no@bcas.tv] has joined ##stm32
2012-09-20T15:50:18 < dongle> this is fucking ridiculous
2012-09-20T15:50:31 < dongle> fuck feenode and fuck that leven or whatever guy
2012-09-20T15:54:36 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-09-20T15:56:30 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 264 seconds]
2012-09-20T15:57:11 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:00:58 -!- izua_ [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-20T16:01:16 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-20T16:01:16 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-20T16:01:16 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T16:01:19 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-20T16:01:41 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-159.pptp.stw-bonn.de] has joined ##stm32
2012-09-20T16:04:59 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-20T16:05:37 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:09:05 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-09-20T16:09:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-20T16:11:37 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 260 seconds]
2012-09-20T16:16:37 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-20T16:18:41 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:21:43 < Laurenceb> muhaha
2012-09-20T16:22:01  * Laurenceb has mailboxes, three threads and adc running in chibios with usb terminal
2012-09-20T16:24:06 < zyp> cool story bro
2012-09-20T16:24:07 < Erlkoenig> what's a mailbox?
2012-09-20T16:24:24 < zyp> FIFO for interthread communication
2012-09-20T16:24:54 < Erlkoenig> ah... implemented as a ring buffer?
2012-09-20T16:25:20 < zyp> that's an implementation detail.
2012-09-20T16:26:29 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 246 seconds]
2012-09-20T16:28:01 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:29:49 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-09-20T16:31:45 < dongle> <Laurenceb> malebox
2012-09-20T16:35:53 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 260 seconds]
2012-09-20T16:36:46 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:42:41 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-20T16:43:41 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:45:12 < Thorn> forgot where channel logs are stored? google for stm32 dongs. it's not just the first result, it's the whole page
2012-09-20T16:46:02 < zyp> :)
2012-09-20T16:49:22 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 252 seconds]
2012-09-20T16:56:07 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T16:58:25 < zyp> nice, USB HID has a definition for «Magic Carpet Simulation Device»
2012-09-20T16:58:25 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T17:01:41 < zyp> http://jiri-svoboda.blogspot.se/2011/05/how-to-ride-magic-carpet.html <- I'm apparently not the only one amused by it
2012-09-20T17:01:48 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-20T17:02:03 -!- claude_work is now known as claude
2012-09-20T17:06:03 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-20T17:06:05 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-20T17:07:53 < Laurenceb> do i need to use AFRH on F4 for ADC?
2012-09-20T17:08:22 < zyp> no, you just set the pin in analog mode
2012-09-20T17:08:48 < zyp> AFRx only applies when pin is in AF mode
2012-09-20T17:11:36 < jpa-> huh, never knew all that stuff USB HID defines
2012-09-20T17:12:08 < zyp> it's insane
2012-09-20T17:12:20 < Laurenceb> ok thanx
2012-09-20T17:12:24 < Laurenceb> something is wrong
2012-09-20T17:12:25 < zyp> some other part of the specification lists pretty much all body parts except the penis
2012-09-20T17:12:32 < jpa-> i wonder how much of it is actually used or supported
2012-09-20T17:12:38 < Laurenceb> im trying to read the port C4 and 5
2012-09-20T17:12:40 < jpa-> it does list the "do not disturb" light
2012-09-20T17:13:08 < Laurenceb> im getting noise from the adc
2012-09-20T17:13:22 < Laurenceb> but its not responding to singal input
2012-09-20T17:13:46 < zyp> Laurenceb, does the internal channels work? (temp sensor and stuff)
2012-09-20T17:13:55 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T17:15:51 < Laurenceb> i dont know how to set this stuff up...
2012-09-20T17:16:17 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/threads/PID_Pressure.c#L57
2012-09-20T17:16:20 < Laurenceb> looks sane?
2012-09-20T17:16:30 < jpa-> fortunately the touch screen spec allows for penises: "Any human appendage used as a transducer,"
2012-09-20T17:16:42 < jpa-> it also defines some useful info like "barrel pressure" and "tip pressure"
2012-09-20T17:18:38 <+Steffanx> jpa- is happy now
2012-09-20T17:20:02 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection]
2012-09-20T17:20:35 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-20T17:29:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-20T17:34:36 < Laurenceb> so alt function register should be zeros for analogue?
2012-09-20T17:35:53 < jpa-> anything should work
2012-09-20T17:36:06 < jpa-> just make type register "analog"
2012-09-20T17:36:20 < Laurenceb> hmm
2012-09-20T17:36:28 < Laurenceb> temp sensor works
2012-09-20T17:36:47 < Laurenceb> im getting no change, just noise if i put 3.3v on port c , 4
2012-09-20T17:36:56 < Laurenceb> in14
2012-09-20T17:42:12 < zyp> did you put it in analog mode?
2012-09-20T17:42:33 < Laurenceb> yes
2012-09-20T17:43:11 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/board/board.h#L241
2012-09-20T17:56:43 < Erlkoenig> http://slashdot.org/poll/2443/how-i-treat-my-books  <-- what? do you americans use the word "Pfui"?
2012-09-20T17:58:38 < jpa-> why not?
2012-09-20T17:58:49 < Erlkoenig> it appears to be a german one? :D
2012-09-20T17:58:51 <+Steffanx> "you americans"  not that many (active) "americans" around here..
2012-09-20T18:00:17 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-20T18:00:30 < Erlkoenig> is that word generally used among english-speakers or just sometimes because it's funny to use other language's words?
2012-09-20T18:00:34 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-20T18:00:34 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-20T18:00:34 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T18:00:37 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-20T18:00:51 <+Steffanx> but at least we have jpa-
2012-09-20T18:01:37 < Erlkoenig> hmm this coil generates really nice sparks <3
2012-09-20T18:01:39  * jpa- haps Steffanx 
2012-09-20T18:07:23 -!- Thorn [~Thorn@unaffiliated/thorn] has quit []
2012-09-20T18:14:16 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-09-20T18:15:43 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-20T18:15:43 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-20T18:15:43 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-20T18:15:46 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-20T18:34:34 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-20T18:41:08 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-20T18:53:51 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 256 seconds]
2012-09-20T18:58:46 -!- DaKu is now known as daku
2012-09-20T19:00:52 -!- Steffanx_ [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-20T19:01:35 -!- daku is now known as DaKu
2012-09-20T19:02:57 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 245 seconds]
2012-09-20T19:05:19 -!- piele [~Unknown@bakzeil.creativeserver.net] has joined ##stm32
2012-09-20T19:08:11 -!- Steffanx_ is now known as Steffanx
2012-09-20T19:08:24 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-20T19:14:18 -!- barthess [~barthess@178.154.6.106] has joined ##stm32
2012-09-20T19:30:23 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-20T19:33:01 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-20T19:36:28 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-20T19:38:01 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-09-20T19:38:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-09-20T19:39:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-20T19:40:01 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-159.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-20T19:51:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-20T19:56:26 < karlp> whoever was talking about 0.5% resonators vs HSI, microchip's announced a new set of usb chips with 0.25% clocks onboard, so usb without crystals too
2012-09-20T20:00:19 < karlp> looks like it's tuning based on usb start of frame stuff
2012-09-20T20:01:03 < gnomad> which family?
2012-09-20T20:03:52 -!- nemuikuma [~hunyeti@188-143-110-2.pool.digikabel.hu] has joined ##stm32
2012-09-20T20:04:42 < karlp> I don't know, lots.
2012-09-20T20:04:48 < karlp> they're tlaking about it across all families
2012-09-20T20:05:27 < gnomad> aah
2012-09-20T20:05:53 < gnomad> laser trimming or something else?
2012-09-20T20:16:40 < gnomad> http://ww1.microchip.com/downloads/en/Market_Communication/Microchip%27s%208-bit%20USB%20MCUs%20Press%20Presentation_Final.pdf
2012-09-20T20:17:55 < gnomad> " 0.25% accuracy when connected to a USB Host .25% accuracy when connected to a USB Host"
2012-09-20T20:18:03 < gnomad> heh... I see what they are doing there!
2012-09-20T20:21:58 < jpa-> hehe :)
2012-09-20T20:22:32 < jpa-> i wonder if any manufacturer has yet integrated a crystal oscillator in the same package.. i don't see why not
2012-09-20T20:23:31 -!- claude is now known as Claude
2012-09-20T20:25:23 < BrainDamage> I have an rtc chip from maxim with integrated xtal
2012-09-20T20:25:32 < BrainDamage> it's worth like 6€ in 1k amounts
2012-09-20T20:25:54 < gnomad> 6 what?
2012-09-20T20:26:20 <+Steffanx> Arduinos
2012-09-20T20:28:00 < jpa-> true, rtc chips do that
2012-09-20T20:28:07 <+izua> do they have that special power socket shield?
2012-09-20T20:31:19 < BrainDamage> gnomad: euro, fix your encoding to utf-8
2012-09-20T20:37:33 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-20T20:37:49 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-20T20:37:50 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-20T20:39:55 < zyp> BrainDamage, I also have some of those
2012-09-20T20:40:24 < zyp> it's a bit amusing how much larger the package for the one with integrated crystal are compared to the one without :p
2012-09-20T20:40:59 <+Steffanx> standard 32KHz xtals are also large :)
2012-09-20T20:41:35 < qyx_> there are some 1206 available
2012-09-20T20:41:55 < qyx_> but with slightly higher ppm
2012-09-20T20:41:59 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-20T20:42:20 < zyp> hmm
2012-09-20T20:42:52 < zyp> in the process of updating my F4 usb code to match the changes I did when I wrote my F1 usb code
2012-09-20T20:43:02 < zyp> I miss the simplicity of the F1 peripheral :p
2012-09-20T20:44:24  * jpa- hates to live in a place with 70° magnetic inclination
2012-09-20T20:44:39 < jpa-> makes all magnetometer stuff 10x more difficult
2012-09-20T20:44:58 <+Steffanx> life is hard huh?
2012-09-20T20:45:00 < zyp> makes you do it right the first time around, you mean? :p
2012-09-20T20:46:07 < jpa-> when my battery moves, the metal shell of it apparently affects the magnetometer bias.. and it affects it so much that horizontal component gets lost
2012-09-20T20:46:18 < zyp> :)
2012-09-20T20:46:27 < BrainDamage> how much is north vs magnetic north distortion there?
2012-09-20T20:47:07 < jpa-> 10-20°
2012-09-20T20:47:11 <+Steffanx> huh
2012-09-20T20:47:12 < jpa-> i don't care about declination :)
2012-09-20T20:47:14 <+Steffanx> *heh
2012-09-20T20:47:31 < jpa-> i am trying to align two devices (transmitter and receiver) based on magnetometer
2012-09-20T20:48:01 < nemuikuma> hey! what do you guys think, which stm32 discovery board is the best to get started with 32bit micros? (best documentation, tutorials,etc, performance does not matter now)
2012-09-20T20:48:14 < zyp> hrm, fuck this shit
2012-09-20T20:48:30 < BrainDamage> what zyp said :p
2012-09-20T20:48:53 < jpa-> nemuikuma: not much difference in docs & tutorials, but avoid STM32F1 (value line discovery).. it is the oldest, all the others have better designed peripherals
2012-09-20T20:48:56 < zyp> the tx-fifo-registers in the usb core is not contiguous
2012-09-20T20:49:24 < BrainDamage> I'd say get an f3 or f4
2012-09-20T20:49:33 < zyp> there is a bunch of shit between DIEPTXF0 and DIEPTXF1, so I can't just make it a single array
2012-09-20T20:49:34 < BrainDamage> they added nice stuff on the board
2012-09-20T20:49:36 < jpa-> i like L1
2012-09-20T20:49:44 < zyp> I agree with jpa
2012-09-20T20:50:04 < zyp> both F4, L1, F0 and F3 should be nice boards
2012-09-20T20:50:19 < BrainDamage> pretty much anything except VL
2012-09-20T20:50:26 < emeb> F3 boards out yet?
2012-09-20T20:50:29 < zyp> I have no idea about documentation and tutorials, I just care about the chip itself
2012-09-20T20:50:33 < jpa-> F0 and F3 are the newest so perhaps not so much tutorials, but they are quite similar to the other ones
2012-09-20T20:50:36 < zyp> emeb, not shipping yet
2012-09-20T20:50:45 < emeb> zyp: thought so.
2012-09-20T20:51:09 < zyp> I might pick it up together with F0 just to complete the collection :p
2012-09-20T20:51:29 < emeb> "Be the first on your block to collect them all!"
2012-09-20T20:51:30 < zyp> I don't have VL, it seemed too old to be useful :p
2012-09-20T20:51:36 < nemuikuma> stm32f0 is discounted at rs components now so i might go with that one
2012-09-20T20:51:53 < emeb> and the stlink v1 sux.
2012-09-20T20:51:59 < zyp> emeb, exactly
2012-09-20T20:52:23 < zyp> though, I guess it's mostly a firmware issue
2012-09-20T20:52:41 < zyp> could probably replace it with BMP
2012-09-20T20:53:37 < BrainDamage> I have the VL, I bought it back when L1 wasn't even out
2012-09-20T20:54:15 < BrainDamage> I threw it in a box and forgot, mostly because I cba'ed to switch OS to dev on it
2012-09-20T20:54:26 < emeb> heh - loading BMP into the stlink MCU would rule.
2012-09-20T20:54:31 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-20T20:54:31 < BrainDamage> then gained interest back when I saw texane's stlink
2012-09-20T20:54:46 < emeb> you'd probably have to shift the I/O around tho.
2012-09-20T20:54:58 < zyp> emeb, there is already support for that in the repo
2012-09-20T20:55:08 < emeb> zyp: figures.
2012-09-20T20:56:30 < BrainDamage> there was even a tutorial
2012-09-20T20:56:34 < BrainDamage> sec that I'll dig it
2012-09-20T20:56:39 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-20T20:56:59  * emeb has been happy with BMP debugging lately. 
2012-09-20T20:57:21 < emeb> seems to be very reliable - stepping thru code to find actual bugs.
2012-09-20T20:57:41 < zyp> :)
2012-09-20T20:57:58 <+Steffanx> Just thank gsmcmullin :)
2012-09-20T20:58:04 < BrainDamage> http://juick.com/paulfertser@jabber.org/1995035
2012-09-20T20:58:52 < emeb> ty - bookmarked.
2012-09-20T21:00:26 < emeb> cleaned up a bunch of DMA setup issues in my F4 audio code yesterday. nice to be able to reliably display variables.
2012-09-20T21:05:13 < emeb> so - what's the done thing when one needs atan2()?
2012-09-20T21:05:17 < emeb> just use -lm?
2012-09-20T21:09:57 < karlp> I use the VL board all the time, it's really not that bad.
2012-09-20T21:10:46 < emeb> I guess it depends on what your target app is doing.
2012-09-20T21:11:15 < karlp> well, the only difference between v1 and v2 stlink is nrst and svo/swv
2012-09-20T21:11:21 < karlp> so, it more depends what your tools are doing :)
2012-09-20T21:13:18 < emeb> *nod* I guess it's not really a V1 vs V2 issue. My troubles were with stlink in general.
2012-09-20T21:14:06 < zyp> karlp, and the scsi transport stuff
2012-09-20T21:14:12 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 248 seconds]
2012-09-20T21:14:54 < karlp> but that's transparent to anyone using it.
2012-09-20T21:23:19 < emeb> F3 Discovery looks nice. Interesting that the HSE XTAL is DNI.
2012-09-20T21:23:50 < emeb> Is the HSI osc accurate enough for USB?
2012-09-20T21:24:51 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-20T21:32:05 < zyp> hmm, nice, my F4 code seems to be much more reliable now
2012-09-20T21:33:52 < zyp> at least the OUT part of it
2012-09-20T21:39:52 -!- bgamari [~bgamari@pool-108-8-238-44.spfdma.east.verizon.net] has quit [Ping timeout: 240 seconds]
2012-09-20T21:41:04 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-20T21:41:27 -!- nemuikuma [~hunyeti@188-143-110-2.pool.digikabel.hu] has quit []
2012-09-20T21:42:08 < ntfreak> v1 stlink has srst as long as you are using the v2 api
2012-09-20T21:42:27 < ntfreak> gets confusing i know
2012-09-20T21:44:41 < Tectu> hmm, I used OpenOCD the first time today...
2012-09-20T21:44:50 < Laurenceb_> hi
2012-09-20T21:45:12 < Laurenceb_> i think the F1 on my F4discovery get a little warm in use
2012-09-20T21:45:26 < Laurenceb_> its could be heat from the nearby reg, hard to say
2012-09-20T21:45:31 < Tectu> Laurenceb, got some liquid nitrogen?
2012-09-20T21:45:42 < Laurenceb_> lol
2012-09-20T21:45:50 < Laurenceb_> im talking ~15C above ambient
2012-09-20T21:45:57 < Laurenceb_> hope all the gpio are configured ok
2012-09-20T21:46:01 < Tectu> o0
2012-09-20T21:46:15 < Tectu> how to wrong configured GPIO heat up your F1 on your F4?
2012-09-20T21:46:37 < Laurenceb_> well if the st code was erranous
2012-09-20T21:46:46 < Laurenceb_> it could be shorting a pin
2012-09-20T21:46:58 < Laurenceb_> im guessing its heat from the 3v3 regulator
2012-09-20T21:47:26 < Tectu> should your output driver fix shortings?  and i mean... heat it up? lol?
2012-09-20T21:47:49 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-20T21:47:56 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-20T21:48:21 < Laurenceb_> im just wondering if st did something stupid for it to get warm like that
2012-09-20T21:49:15 < Tectu> i am pretty sure that the answer would be no
2012-09-20T21:55:39 < zyp> http://bin.jvnv.net/f/gffMk.png :)
2012-09-20T21:59:55 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: z]
2012-09-20T22:01:16 -!- tech2077_ [~tech2077@207.80.127.250] has joined ##stm32
2012-09-20T22:04:46 <+Steffanx> Hmm, that software shows data-length and time in the same column zyp ?
2012-09-20T22:04:53 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 260 seconds]
2012-09-20T22:06:17 -!- tech2077_ [~tech2077@207.80.127.250] has quit [Ping timeout: 260 seconds]
2012-09-20T22:06:36 < Tectu> Steffanx, all cool softwares do
2012-09-20T22:06:43 <+Steffanx> Yeah
2012-09-20T22:08:48 < zyp> Steffanx, yep, depending on what makes sense
2012-09-20T22:11:06 -!- ChanServ [ChanServ@services.] has quit [shutting down]
2012-09-20T22:13:26 < zyp> hmm, I need to do something fun with this code now, it kind of gets boring when everything is working :p
2012-09-20T22:14:25 < zyp> maybe I should do SDIO and then MSC
2012-09-20T22:16:17 < Laurenceb_> 1)  make a self replicating 3d printer that can replicate itself half size
2012-09-20T22:16:20 < Laurenceb_> 2) ???
2012-09-20T22:16:23 < Laurenceb_> 3) profit
2012-09-20T22:16:42 < Laurenceb_> zyp: good luck getting format to work
2012-09-20T22:17:43 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-09-20T22:17:43 -!- ServerMode/##stm32 [+o ChanServ] by sendak.freenode.net
2012-09-20T22:18:52 <+Steffanx> "working" as in .. hacked everything together but it works "working"?
2012-09-20T22:20:32 < Laurenceb_> yes
2012-09-20T22:21:00 < Laurenceb_> thats the only sort of working i understand
2012-09-20T22:23:24 <+Steffanx> Still wonder why you hate Arduino so much
2012-09-20T22:24:07 < Laurenceb_> it gives me meaning in my life
2012-09-20T22:24:45 <+izua> by meaning you mean "rage inducing headaches cause by noobs"
2012-09-20T22:24:56 <+izua> caused*
2012-09-20T22:25:04 <+izua> because you should add ubuntu to your meaningless life too
2012-09-20T22:26:33 < BrainDamage> well, as tool per se, it does what it claims to do
2012-09-20T22:26:38 < BrainDamage> makes simple stuff simple
2012-09-20T22:27:55 < BrainDamage> I don't like it attracted the noobs by lowering the entry barrier, but I have mixed feelings
2012-09-20T22:37:17 <+izua> i don't like the fact that - although it lowered the entry barrier, which is a good thing, more people in a field means faster improvement overall - it taught bad concepts from the beginning.
2012-09-20T22:37:49 <+izua> the pcb/shield layout is bad, libs were recently made to not block, and completely isolating the programmer from the lower level stuff
2012-09-20T22:38:40 < BrainDamage> the last is actually what made the entry barrier lower tough
2012-09-20T22:39:11 <+Steffanx> It's so easy to start a flamewar-ish talk by just mentioning "arduino" :P
2012-09-20T22:40:23 <+izua> it's like jesus
2012-09-20T22:40:32 <+izua> it's not jesus itself one hates, but its fanbois
2012-09-20T22:40:59 <+Steffanx> Wrong example, you should've used Apple :P
2012-09-20T22:41:26 <+izua> i don't like offending present people :P
2012-09-20T22:43:25 <+Steffanx> So your rpi runs at 1GHz yet izua ?
2012-09-20T22:43:45 <+Steffanx> ( without voiding the warranty )
2012-09-20T22:43:53 <+izua> it ran away from rpi foundation to my place
2012-09-20T22:44:00 <+izua> it was exhausted by the time it got here
2012-09-20T22:44:06 <+izua> so were stocks, when i decided to buy one :P
2012-09-20T22:44:14 <+Steffanx> Aw
2012-09-20T22:44:23 <+Steffanx> And now you decided you don't want one anymore?
2012-09-20T22:44:27 <+izua> well
2012-09-20T22:44:31 <+izua> hypeberry pi
2012-09-20T22:45:05 <+Steffanx> They made it almost 50% faster, don't you LOVE that?
2012-09-20T22:45:32 < feurig> +1 izua
2012-09-20T22:45:48 <+Steffanx> feurig doesn't know what he talks about
2012-09-20T22:46:12 < feurig> true I only work with hardware I can make myself :)
2012-09-20T22:46:14 < zyp> mine disappeared
2012-09-20T22:46:22 <+izua> you make microcontrollers? :P
2012-09-20T22:46:41 <+Steffanx> You didn't eat it did you zyp ?
2012-09-20T22:46:43 < feurig> no but i make microcontroller based devices
2012-09-20T22:48:07 < zyp> Steffanx, nah, I let my usb guy borrow it, he said he was going to fix the usb driver
2012-09-20T22:48:27 <+Steffanx> heh
2012-09-20T22:48:37 <+Steffanx> Someone fixed something in the USB driver
2012-09-20T22:48:41 < zyp> it came up once we were talking about something, and I mentioned that I had one that I don't use
2012-09-20T22:48:42 <+Steffanx> don't remember what and who
2012-09-20T22:49:00 < feurig> Steffanx: its true. I will admit to that.
2012-09-20T22:50:38 < BrainDamage> feurig: so you diy microcontrollers? what's you process resolution?
2012-09-20T22:50:42 <+Steffanx> feurig, in dutch i would say something like "Any fool his absence, blablabla", but it sounds terrible in English :P
2012-09-20T22:51:13 <+Steffanx> What didn't BrainDamage understand when he said "no but i make microcontroller based devices"
2012-09-20T22:51:21 <+izua> well
2012-09-20T22:51:26 <+izua> he could use rpi in a microcontroller based device.
2012-09-20T22:51:50 < BrainDamage> Steffanx: I missed that sentence because I didn't notice my scrollbar wasn't at bottom
2012-09-20T22:51:56 <+Steffanx> :)
2012-09-20T22:52:32 <+Steffanx> I even heat-sinked the ARM and USB/ethernet chip on my rpi :D
2012-09-20T22:53:11 < zyp> did I mention the rather affordable exynos board I came across the other day?
2012-09-20T22:53:14 <+Steffanx> ( the chinese crap seems to do it job ok-ish )
2012-09-20T22:53:15 <+Steffanx> Yes
2012-09-20T22:53:33 < zyp> it was on paper 16 times the performance for less than four times the price
2012-09-20T22:53:46 < zyp> I almost bought one
2012-09-20T22:53:49 <+Steffanx> Too bad they molded that base board onto the other one
2012-09-20T22:53:54 <+izua> Steffanx: wait
2012-09-20T22:53:57 <+Steffanx> or the otherway around
2012-09-20T22:53:57 < zyp> but then I realized I have as little use for it as I have for rpi :p
2012-09-20T22:53:57 <+izua> you're an rpi fanboi?
2012-09-20T22:54:05 <+Steffanx> No, izua
2012-09-20T22:54:17 <+izua> damn you should get under dongs' bridge
2012-09-20T22:54:33 < BrainDamage> rpi is ok as cheap router/nas
2012-09-20T22:54:42 <+Steffanx> It's actually on my desk ( and it's on ), but just because i wanted to try their super advanced improvements
2012-09-20T22:54:45 < zyp> Steffanx, I'm pretty sure it's detachable, they sell the processor board alone for $10 less or something
2012-09-20T22:54:45 < BrainDamage> it has better hw specs for similar price
2012-09-20T22:54:52 <+Steffanx> I don't have a good power supply for it anyway
2012-09-20T22:55:32 <+Steffanx> It now uses a DC/DC converter + my lab psu
2012-09-20T22:55:39 <+Steffanx> just because it's possible :)
2012-09-20T22:55:47 < zyp> BrainDamage, rpi would be a pretty shitty router and even more shitty nas
2012-09-20T22:55:55 <+Steffanx> Not really
2012-09-20T22:56:03 <+Steffanx> For 100mbit it's ok :P
2012-09-20T22:56:09 < zyp> what's real world ethernet performance on it?
2012-09-20T22:56:11 < BrainDamage> nah, as hw it's actually superior to many soho boxes
2012-09-20T22:56:18 <+Steffanx> 9-10MByte/s
2012-09-20T22:56:25 < Tectu> what's the  target/xxx.cfg    in openocd for a regular STM32F103VET?
2012-09-20T22:57:07 < zyp> I'm pretty sure the rpi would be a bottleneck with my connection
2012-09-20T22:57:11 < BrainDamage> if you look at openwrt project the list of hw in those soho boxes, they are incredibly appaling
2012-09-20T22:57:31 <+Steffanx> You and your gigabit fetish zyp :P
2012-09-20T22:57:33 < zyp> I've got an alix as a router here now, it's a bottleneck, and I'm pretty sure the rpi would do worse
2012-09-20T22:57:41 < BrainDamage> how much did your router cost tough?
2012-09-20T22:58:08 < zyp> no idea, I got it for free because it was supposed to be broken
2012-09-20T22:58:41 < zyp> reflashed it and got it working without problems
2012-09-20T22:59:00 < BrainDamage> http://hackaday.com/2012/09/20/programming-an-arm-with-basic/
2012-09-20T22:59:46 < zyp> but ok, it'll probably do fine as a router for the slow connections that a lot of people still have :p
2012-09-20T23:00:00 < zyp> but NAS?
2012-09-20T23:00:58 < zyp> what is the point of having a NAS if it's fucking slow?
2012-09-20T23:02:09 < zyp> I mean, at 100 Mb/s, transferring a terabyte will take a full day
2012-09-20T23:02:40 < BrainDamage> because I don't need to transfer a TB
2012-09-20T23:02:52 < zyp> what do you need to transfer?
2012-09-20T23:02:57 < zyp> what do you use the NAS for?
2012-09-20T23:03:09 < BrainDamage> I dump incremental backups from my laptop
2012-09-20T23:03:27 < BrainDamage> and use it as torrent client that's always on
2012-09-20T23:04:50 < zyp> right, so you get a ton of files sitting on it
2012-09-20T23:05:18 < BrainDamage> yup, but small activity
2012-09-20T23:05:32 < zyp> and then you're leaving for a trip, and bring your laptop for a bit of entertainment on the train or whatever, you just need to copy some movies or episodes from the NAS
2012-09-20T23:06:31 < zyp> ever did that?
2012-09-20T23:06:43 < BrainDamage> yes, in that case, the disks are in a lvm setup
2012-09-20T23:06:54 < BrainDamage> with music and videos in specific hds
2012-09-20T23:07:08 < zyp> and?
2012-09-20T23:07:17 < BrainDamage> I unmount them, and carry with me
2012-09-20T23:07:34 <+Steffanx> Ha
2012-09-20T23:07:55 < zyp> right, so because copying the contents is fucking slow, you just bring along the entire drive
2012-09-20T23:07:59 -!- barthess [~barthess@178.154.6.106] has quit [Read error: Connection reset by peer]
2012-09-20T23:08:32 < BrainDamage> sure, my system is designed to be cheap, and easily maintenable
2012-09-20T23:08:36 < BrainDamage> not for performance
2012-09-20T23:08:58 < BrainDamage> I never claimed high speeds
2012-09-20T23:09:36 < BrainDamage> also, I challenge your bandwith speed vs me carrying the hd physically :p
2012-09-20T23:09:48 < zyp> oh yeah?
2012-09-20T23:10:08 <+Steffanx> Ok, BrainDamage first goes to zyp
2012-09-20T23:10:18 <+Steffanx> and goes back while zyp sends the hdd data back to BrainDamage
2012-09-20T23:10:32 < zyp> http://bin.jvnv.net/f/ihnnt.JPG <- my drive is the box sitting in the bottom here, good luck carrying it fast :)
2012-09-20T23:10:57 < BrainDamage> as I said, diff design goals
2012-09-20T23:11:21 <+Steffanx> less money involved ..
2012-09-20T23:12:03 < zyp> it's used shit picked up from ebay, not really that expensive :p
2012-09-20T23:12:18 <+Steffanx> Why people buy shit on ebay?
2012-09-20T23:12:26 < zyp> why not?
2012-09-20T23:12:39 <+Steffanx> I go to the toilet when i need shit :P
2012-09-20T23:13:00 < zyp> while it might be outdated in enterprise environments, it still easily saturates the gigabit network I have at home :p
2012-09-20T23:13:17 < BrainDamage> that was really a ( •_•)        ( •_•)>⌐■-■         (⌐■_■) poo-rful poon
2012-09-20T23:13:40 < zyp> Steffanx, that sounds backwards, I go to the toilet when I need to get rid of shit.
2012-09-20T23:14:06 <+Steffanx> You have NO idea what i do :P
2012-09-20T23:14:23 < BrainDamage> and I think we are pretty happy with that
2012-09-20T23:14:34 <+Steffanx> but … you are right :)
2012-09-20T23:15:09 -!- Claude is now known as claude
2012-09-20T23:16:37 <+Steffanx> ( what's the reason behind that claude ?)
2012-09-20T23:16:45 <+Steffanx> C => c ?!
2012-09-20T23:17:41 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-20T23:31:30 -!- barthess [~barthess@178.154.48.165] has joined ##stm32
2012-09-20T23:50:31 < emeb> feeling low
2012-09-20T23:50:36 < emeb> (ercase)
--- Day changed Fri Sep 21 2012
2012-09-21T00:00:17 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-21T00:04:18 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 264 seconds]
2012-09-21T00:21:40 -!- barthess [~barthess@178.154.48.165] has quit [Quit: Leaving.]
2012-09-21T00:33:07  * Laurenceb_ rage at chibios
2012-09-21T00:33:18 < Laurenceb_> they keep changing the interface _all the time_
2012-09-21T00:37:49 < zyp> that's how interfaces get better
2012-09-21T00:42:36 < Laurenceb_> lol thats one big eval kit
2012-09-21T00:44:11 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-21T00:50:15 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-21T00:50:54 < zyp> huh?
2012-09-21T00:50:56 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-21T00:55:10 < ntfreak> Tectu: target/stm32f1x.cfg normally, if using an stlink it would be target/stm32f1x_stlink.cfg
2012-09-21T01:16:52 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 248 seconds]
2012-09-21T01:18:29 -!- bgamari [~bgamari@pool-96-240-253-151.spfdma.east.verizon.net] has joined ##stm32
2012-09-21T01:19:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-21T01:20:07 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-21T01:22:22 < Thorn> what usb controller does f3 use? same synopsys as most other stm32 families?
2012-09-21T01:28:18 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-21T01:28:27 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 252 seconds]
2012-09-21T01:30:35 -!- bgamari [~bgamari@pool-96-240-253-151.spfdma.east.verizon.net] has quit [Quit: ZNC - http://znc.in]
2012-09-21T01:36:04 < zyp> Thorn, same as F1/L1
2012-09-21T01:36:19 < Thorn> what
2012-09-21T01:36:33 < Thorn> I thought that one was dead
2012-09-21T01:36:42 < zyp> why?
2012-09-21T01:37:10 < Thorn> since they seemed to switch to synopsys for later designs
2012-09-21T01:37:30 < Thorn> so, no avoiding two sets of drivers
2012-09-21T01:38:15 < Thorn> (with incompatible APIs too)
2012-09-21T01:38:16 < zyp> it's probably much much smaller than the huge OTG block
2012-09-21T01:38:56 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has joined ##stm32
2012-09-21T01:40:44 < Thorn> ram and flash sizes are also similar to those of f1
2012-09-21T01:41:26 < Thorn> it better cost no more than f1 too
2012-09-21T01:41:53 < zyp> also, the driver is really just a small part of a usb stack
2012-09-21T01:42:33 < Thorn> that's if it is actually separated from the rest of the stack, which is not the case for f102/f103 drivers
2012-09-21T01:42:57 < zyp> that's not my problem, in mine it is :p
2012-09-21T01:43:18 < zyp> I just finished up a CDC demo that works on both F1 and F4 today :p
2012-09-21T01:43:45 < Thorn> that's the problem, you want usb in a project, prepare to spend 2/3 developent time doing your own usb stack
2012-09-21T01:44:14 < zyp> why?
2012-09-21T01:44:20 < gnomad> are there any decent open source usb stacks?
2012-09-21T01:44:36 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-21T01:44:38 < zyp> both chibios and libopencm3 support both F1 and F4 in their stacks
2012-09-21T01:45:24 < zyp> and mine of course, but I wouldn't call it decent yet :p
2012-09-21T01:45:52 < Thorn> you should publish it as a separate project on github
2012-09-21T01:45:58 < zyp> I have
2012-09-21T01:46:08 < zyp> https://github.com/zyp/laks
2012-09-21T01:47:00 < Thorn> nice
2012-09-21T01:47:58 < zyp> it's a mirror in case people want to make github forks, main repo is here: http://cgit.jvnv.net/laks/
2012-09-21T01:49:00 < Thorn> hopefully people will want to make pull requests
2012-09-21T01:49:04 < zyp> http://cgit.jvnv.net/laks/commit/?id=89baa2a <- and here is the required effort of adding F1 support to the existing F4 stack
2012-09-21T01:50:08 < Thorn> that's because it looks designed. the ST library is tens of Kb of code
2012-09-21T01:50:32 < Thorn> *well designed
2012-09-21T01:51:22 < zyp> the libopencm3 stack also seems pretty well designed, I've read through the sources to get some ideas
2012-09-21T01:51:40 < zyp> and that one is clearly more mature than mine :p
2012-09-21T02:00:37 -!- DaKu is now known as daku
2012-09-21T02:02:53 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 260 seconds]
2012-09-21T02:08:04 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 246 seconds]
2012-09-21T02:11:55 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 246 seconds]
2012-09-21T02:12:06 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 264 seconds]
2012-09-21T02:15:56 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-21T02:15:56 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-21T02:15:56 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-21T02:19:30 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-21T02:32:20 < upgrdman> Laurenceb, i fell your pain. rm0090 is such a pain sometimes!
2012-09-21T02:33:11 < upgrdman> and it didn't help that the f4 was my first "real" foray into mcu's. (arduinos don't count, etc.)
2012-09-21T03:42:32 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-65.pptp.stw-bonn.de] has joined ##stm32
2012-09-21T03:44:37 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-65.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-21T04:47:48 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-21T04:50:31 < dongs> sup dongs
2012-09-21T05:11:09 < karlp> ntfreak: re: < ntfreak> v1 stlink has srst as long as you are using the v2 api
2012-09-21T05:11:30 < karlp> what do you mean exaactly? there's only 4 pins on the "target" header
2012-09-21T05:15:28 < dongs> http://linux.slashdot.org/story/12/09/20/1245240/stubborn-intel-graphics-bug-haunts-ubuntu-1204 lol opensores
2012-09-21T05:18:17 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-21T05:24:35 < Bird|lappy> dongs, no, lol at Intel Extreme Buggy Graphics
2012-09-21T05:25:18 < dongs> sorry bro that logic fails
2012-09-21T05:25:21 < dongs> it works wonderfully on windows
2012-09-21T05:29:21 < Bird|lappy> Intel graphics chips have p. much always been redheaded stepchildren.  do keep in mind that I'm looking at this from a 3D viewpoint though
2012-09-21T05:30:27 < Bird|lappy> although Linux's tendency to play close to the spec has been known to expose whacked-out HW in the past.
2012-09-21T05:30:40 < dongs> sorry dude, wrong there again.
2012-09-21T05:30:52 < dongs> on both counts actually.
2012-09-21T05:30:58 < dongs> lol @ playing close to the spec
2012-09-21T05:31:14 < dongs> I remembver in early 2000s i had a P6DLS dual P2 board which dualbooted lunix and windows
2012-09-21T05:31:25 < dongs> there was some weird fucked up ram issues that lunix never hit because it never used most of hte ram
2012-09-21T05:31:33 < dongs> while windows would trip over that shit constantly
2012-09-21T05:33:07  * Bird|lappy chains dongs to a Tandem NonStop running Guardian
2012-09-21T05:33:16 < Bird|lappy> now you /wish/ you had lunix :P
2012-09-21T05:37:54 < dongs> link only for Laurenceb http://www.youtube.com/user/JemmaKuma
2012-09-21T05:44:24 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-21T05:46:18 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-21T06:01:59 -!- BrainDamage [~yaaic@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-21T06:03:22 -!- BrainDamage [~yaaic@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Client Quit]
2012-09-21T06:31:54 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 240 seconds]
2012-09-21T06:32:49 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-21T06:35:02 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Quit: Leaving]
2012-09-21T06:57:55 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-09-21T07:16:18 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-21T07:23:17 -!- Netsplit *.net <-> *.split quits: Nutter, bgamari
2012-09-21T07:28:19 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-21T07:28:26 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has joined ##stm32
2012-09-21T07:28:26 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-21T07:56:16 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-21T08:14:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-21T08:19:39 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-21T08:21:53 -!- claude is now known as Claude
2012-09-21T08:26:41 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 268 seconds]
2012-09-21T08:33:56 < emeb_mac> Hmmm - Mouser quoting 8wks until the F3 parts are in stock.
2012-09-21T08:37:47 < emeb_mac> hmm - don't know if I'd draw that conclusion. These are brand new parts. Never been in stock before.
2012-09-21T08:38:00 < emeb_mac> heh
2012-09-21T08:38:31 < emeb_mac> agree
2012-09-21T08:38:54 < emeb_mac> these are good parts, good prices, lots of features.
2012-09-21T08:40:11 < emeb_mac> I've got an audio product based on the f100 - cheap, easy to use.
2012-09-21T08:41:20 < emeb_mac> the 2-chl DACs + high-res ADC make a good combination for electronic music synthesis.
2012-09-21T08:42:13 < emeb_mac> yeah - not enough mem for typical MODs
2012-09-21T08:42:17 < dongs> you wanted to do a lot of shit
2012-09-21T08:42:21 < dongs> but you ekep failing at it
2012-09-21T08:42:30 < dongs> they do have psram or wahtever.
2012-09-21T08:42:58 < dongs> it does.
2012-09-21T08:43:04 < dongs> 128mbit of psram is only 12 bucks
2012-09-21T08:43:15 < dongs> no you moron
2012-09-21T08:43:16 < emeb_mac> lol
2012-09-21T08:43:16 < dongs> its 2012
2012-09-21T08:43:18 < dongs> not 1996
2012-09-21T08:43:30 < dongs> they have it in this: http://media.digikey.com/photos/Micron%20Photos/557-54-VFBGA.jpg
2012-09-21T08:44:40 < dongs> 8mbit is largest in tsop.
2012-09-21T08:44:48 < dongs> so 1meg.
2012-09-21T08:45:03 < dongs> could squeak by wiht som stuff.
2012-09-21T08:45:34 < emeb_mac> how bout this: http://www.mouser.com/ProductDetail/ISSI/IS66WV51216DBLL-55TLI/?qs=sGAEpiMZZMteY%252blXHTQ5JKR3c313XbVN
2012-09-21T08:45:42 < emeb_mac> cheap, not BGA.
2012-09-21T08:45:55 < emeb_mac> only 1MB though
2012-09-21T08:46:14 < dongs> < dongs> 8mbit is largest in tsop.
2012-09-21T08:46:42 < emeb_mac> yep. available at least, and easy to hand assemble
2012-09-21T08:48:21 < emeb_mac> downside is that to hook up the FSMC you have to go to the 144-pin STM32F4 package.
2012-09-21T08:48:46 < dongs> Right.
2012-09-21T08:49:26 < emeb_mac> there you go, or slap down 2x of those parts.
2012-09-21T08:49:57 < emeb_mac> they're only $2.67/ea
2012-09-21T08:50:37 < emeb_mac> size?
2012-09-21T08:52:38 < emeb_mac> *groan*
2012-09-21T08:53:19  * emeb_mac has a tray full of Micron 64Mb SDRAM - doesn't try to use them.
2012-09-21T08:54:02 < emeb_mac> although I have thought of making an SDRAM controller with an FPGA
2012-09-21T08:54:26 < emeb_mac> nope - no SDRAM controller on the FSMC
2012-09-21T08:55:26 < emeb_mac> its easy.
2012-09-21T08:55:49  * emeb_mac has been doing FPGAs for the day job for > 20 yrs.
2012-09-21T09:02:35 < R2COM> >20 years of fpga and sdram only controller now?
2012-09-21T09:02:36 < Claude> emeb_mac, cool trying somthing similar atm. altera de1 hooked up the f4 disco. fsmc stuff is working quite good allready (using the f4 mclk out to sync the fpga to the fsmc) . still need to get an sdram controller working within the fpga
2012-09-21T09:03:33 < emeb_mac> Claude: cool. does the de1 have an SDRAM onboard already?
2012-09-21T09:04:27 < Claude> yes 8MB , but upgraded mine to 32MB some time ago
2012-09-21T09:04:43 < Claude> also some sram (512k iirc)
2012-09-21T09:05:31 < Claude> well still need an sdram controller on the fpga :) there are ready made ip cores from altera and so , but they are not open or free
2012-09-21T09:05:48 < emeb_mac> Ah. that was the next question.
2012-09-21T09:06:31 < Claude> my plan is using this controller : http://hamsterworks.co.nz/mediawiki/index.php/SDRAM_Memory_Controller
2012-09-21T09:06:42 < emeb_mac> some guys working with the Xilinx/Digilent S3Esk board got an SDRAM controller working there.
2012-09-21T09:06:50 < emeb_mac> used for audio delays, etc.
2012-09-21T09:07:29 < Claude> i need texture and framebuffer memory ,  a lot of it
2012-09-21T09:07:58 < Claude> oh and some z-buffer
2012-09-21T09:08:30 < emeb_mac> ah - video.
2012-09-21T09:08:57 < Claude> yeah : http://www.youtube.com/watch?v=koG3Iu2gids
2012-09-21T09:08:59 < jpa-> 8 MB sram doesn't cost too much nowadays.. i've been thinking of using just SRAM instead of bothering with SDRAM
2012-09-21T09:09:02 < emeb_mac> R2COM: never needed SDRAM before - always doing high-speed DSP for communications.
2012-09-21T09:09:17 < Claude> running out of internal ram . so just one texture and no z-buffer here
2012-09-21T09:09:42 < emeb_mac> purty.
2012-09-21T09:10:07 < emeb_mac> is that over a SPI interface?
2012-09-21T09:10:15 < Claude> fsmc , 8 bit
2012-09-21T09:10:20 < emeb_mac> aha.
2012-09-21T09:10:29 < emeb_mac> spi would be a bottleneck
2012-09-21T09:10:43 < Claude> using the 128k sram as framebuffer , ccm for heap , stack and stuff
2012-09-21T09:10:57 < emeb_mac> right
2012-09-21T09:11:13 < emeb_mac> DMA?
2012-09-21T09:11:51 -!- dekar_ [~dekar@212.255.47.165] has joined ##stm32
2012-09-21T09:11:54 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-21T09:11:57 < Claude> not yet , just memcpy() the framebuffer to the tft
2012-09-21T09:12:11 < emeb_mac> not bad.
2012-09-21T09:12:27 < jpa-> Claude: using MCO to sync FPGA to FSMC is something i've been wondering about - how well does it seem to work? at what frequency are you running?
2012-09-21T09:12:28 < Claude> yeah i'm quite impressed how fast the f4 is
2012-09-21T09:13:05 < emeb_mac> it's def. zippy
2012-09-21T09:13:43 < Claude> jpa-, works quite good . downside is you can't output hclk on it because of the io speed . so i putput hclk/2 and do some waitstates on the fsmc
2012-09-21T09:14:18 < Claude> output even..
2012-09-21T09:14:40 < jpa-> so that's what.. 84 MHz?
2012-09-21T09:15:06 < Claude> iirc yes
2012-09-21T09:15:18 < jpa-> very fast still :)
2012-09-21T09:15:45 -!- dekar [~dekar@212.255.22.171] has quit [Ping timeout: 256 seconds]
2012-09-21T09:16:23 < Claude> before i got the texture loading stuff working i rendered a julia set in realtime as texture , even then the F4 was able to get a decent framerate :)
2012-09-21T09:18:40 < Claude> the multiplexed A/D bus on the 100pin F4 is also a bit of bottleneck , but around 50MB/s is possible
2012-09-21T09:19:34 < Claude> no 1080i then on the F4 ;) lol
2012-09-21T09:20:06 < jpa-> yeah.. though if you know what you are streaming you could skip A and just write data
2012-09-21T09:21:51 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-21T09:22:44 < Claude> right , fsmc burst mode it the way to go. also using sdram without burst is aweful slow compared to sram
2012-09-21T09:28:31 < jpa-> but they still have the spurious read bug in the FSMC burst
2012-09-21T09:28:44 < jpa-> so if you have a FIFO on the FPGA that you want to read from, you can't use burst mode
2012-09-21T09:31:03 < dongs> moving BOOT1 to option bytes on F3 was great idea
2012-09-21T09:31:29 < dongs> trying to route some shit and that signal gets in the fucking way
2012-09-21T09:31:30 < dongs> :(
2012-09-21T09:32:40 < jpa-> what do you need BOOT1 for anyway?
2012-09-21T09:32:59 < jpa-> or no space to hook it to GND?
2012-09-21T09:33:36 < Claude> jpa , you mean the 'dummy read' stuff ? iirc it's only when you try to read from the fsmc smaller than AHB size ?  e.g. reading a byte from a 16bit memory
2012-09-21T09:33:48 < jpa-> Claude: nah not that
2012-09-21T09:34:05 < jpa-> Claude: in the errata, when it does burst it makes one extra read
2012-09-21T09:34:41 < Claude> ehhh, that sucks !
2012-09-21T09:34:45 < jpa-> doesn't matter for memory, but sucks for fifos
2012-09-21T09:35:11 < Claude> maybe i can catch this case in the fpga , if it is somehow predictable
2012-09-21T09:36:22 < jpa-> yeah it's supposedly deterministic
2012-09-21T09:36:30 < jpa-> so could be worked around.. the errata has details
2012-09-21T09:37:25 < Claude> will have a look , thanks for the hint!
2012-09-21T09:49:25 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 248 seconds]
2012-09-21T09:51:07 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-21T09:59:14 -!- Claude is now known as claude
2012-09-21T10:12:15 -!- lawrenceseattle [~anonymous@216.243.49.130] has joined ##stm32
2012-09-21T10:14:47 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Quit: No Ping reply in 180 seconds.]
2012-09-21T10:15:00 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-21T10:22:57 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-21T10:23:54 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 244 seconds]
2012-09-21T10:24:28 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 268 seconds]
2012-09-21T10:31:10 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-21T10:31:27 -!- dekar_ [~dekar@212.255.47.165] has quit [Quit: This computer has gone to sleep]
2012-09-21T10:36:50 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-21T10:36:51 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-21T10:36:51 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-09-21T10:40:57 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-21T10:46:03 -!- lawrenceseattle [~anonymous@216.243.49.130] has quit [Quit: lawrenceseattle]
2012-09-21T10:57:10 -!- claude is now known as claude_work
2012-09-21T11:19:08 -!- BjoernC [~BjoernC@dyndsl-178-142-054-097.ewe-ip-backbone.de] has joined ##stm32
2012-09-21T11:25:56 -!- BjoernC [~BjoernC@dyndsl-178-142-054-097.ewe-ip-backbone.de] has quit [Read error: Connection reset by peer]
2012-09-21T11:26:13 -!- BjoernC [~BjoernC@dyndsl-178-142-054-097.ewe-ip-backbone.de] has joined ##stm32
2012-09-21T11:34:00 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-21T11:42:17 < tavish> i used to watch him on tv many years ago
2012-09-21T11:51:13 < jpa-> claude_work: for your clock-out, are you using MCO2 (SYSCLK) or MCO1 (PLL clock)?
2012-09-21T11:52:07 < claude_work> sysclk but with a /2 divider
2012-09-21T11:53:01 < jpa-> yeah
2012-09-21T11:54:18 < claude_work> in theory i could bring it back to sysclk with an pll inside the fpga ,but  don't know how much this will jitter
2012-09-21T11:54:49 < jpa-> and it could be 1 step off
2012-09-21T11:55:55 < claude_work> *if* i would do a board design with an stm32f4 + fpga i would put an clock buffer on the mco out and route the clk signal directly to the sdram as well . that way all should be synchronous and no need to compensate the sdram stuff
2012-09-21T11:56:15 < jpa-> hmph i don't have any pin left in my connector for NWAIT
2012-09-21T11:56:23 < ntfreak> karlp: i mean if you have a standalone stlink/v1 (firmware above J11) the hardware srst is supported.
2012-09-21T11:56:23 < ntfreak> it is not brought out to CN2 on the STM32VLDISCOVERY but is connected to the micro on the board.
2012-09-21T11:57:19 < claude_work> ehh nWAIT , can be difficult or slow to interface external stuff without nWAIT
2012-09-21T11:57:31 < jpa-> indeed
2012-09-21T11:57:40 < jpa-> i realized that i really need nWAIT
2012-09-21T11:59:38 < jpa-> i have to sacrifice one of my ground pins for that
2012-09-21T12:03:42 < claude_work> i would put a resistor for series termination next to the mco output , 84MHz single ended sucks :)
2012-09-21T12:04:06 < claude_work> 22 ohms or so
2012-09-21T12:05:40 < jpa-> i'll probably go with /4 anyway.. i wonder how far the internal GPIO slew rate limiting could go
2012-09-21T12:09:28 < claude_work> yeah the 84MHz isn't really a square wave anymore on the f4 disco , good enough for the fpga to sync but i wouldn't use that in an commercial product :)
2012-09-21T12:10:43 < jpa-> a nice trapezoid is better than square wave ,)
2012-09-21T12:13:15 < dongs> sup trolls
2012-09-21T12:15:55 < ABLomas> .
2012-09-21T12:16:49 < dongs> oh
2012-09-21T12:18:34 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has left ##stm32 []
2012-09-21T12:21:08 < dongs> claude_work: 84mhz, too fast for gpio?
2012-09-21T12:31:43 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-21T12:36:09 -!- Thorn [~Thorn@unaffiliated/thorn] has quit []
2012-09-21T13:09:36 < tavish> how would you design your application to run a USB MSC host, and to be able to go to sleep mode. Right now i have to continuosly run the USBH_process() function (from the MSC Host example) in a loop. I recieve a request packet(on USBD), and to handle it I change my application, and then I call USBH_process(), it is stuck in HOST_CTRL_XFER state forever. It seems to work otherwise if i just keep it inside a loop in main(). any ideas what the problem coul
2012-09-21T13:13:04 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-21T13:13:14 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-21T13:13:28 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-21T13:13:29 < dongs> http://www.digikey.com/product-detail/en/PRQC8.00CR5010X000/478-4584-1-ND/1645895 how shit is this resonator
2012-09-21T13:15:41 < zyp> tavish, your problem sounds very specific to whatever host stack you are using
2012-09-21T13:23:23 < tavish> zyp: Do you know offhand when this state would be entered? I am using exclusively IN/OUT bul endpoint transfers
2012-09-21T13:25:05 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-21T13:25:05 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-21T13:25:05 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-21T13:25:08 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-21T13:31:10 -!- tavish_ [~tavish@59.177.79.74] has joined ##stm32
2012-09-21T13:31:14 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-21T13:31:59 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-09-21T13:33:09 < dongs> Laurenceb_: < dongs> link only for Laurenceb http://www.youtube.com/user/JemmaKuma
2012-09-21T13:34:53 < Laurenceb_> lol cosplay
2012-09-21T13:50:17 < claude_work> dongs , yes iirc max is 100MHz for a stm32f4 io cell
2012-09-21T13:51:41 < claude_work> but it's not the frequency on the io pin , it's the signal integrity overall
2012-09-21T14:08:04 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-21T14:16:54 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-21T14:23:08 < dongs> makin a digikey order. im sure ill remember something the ne xt day and/or when it arrives.
2012-09-21T14:26:12 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 248 seconds]
2012-09-21T14:35:27 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-21T14:41:47 <+izua> dongs: leds
2012-09-21T14:41:54 <+izua> one never has enough leds
2012-09-21T14:46:14 < dongs> nah
2012-09-21T14:46:22 < dongs> i got reels of leds
2012-09-21T14:46:22 < dongs> at prices digikey cant touch
2012-09-21T14:46:38 < dongs> and they're fuckin bright
2012-09-21T14:53:00 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-21T14:53:03 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-21T14:54:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-09-21T14:55:28 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-21T14:55:32 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-21T15:04:03 -!- daku is now known as DaKu
2012-09-21T15:12:58 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-21T15:16:15 -!- nashpa_ [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-21T15:17:37 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 240 seconds]
2012-09-21T15:18:08 -!- nashpa_ is now known as nashpa
2012-09-21T15:21:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-21T15:25:54 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 240 seconds]
2012-09-21T15:26:57 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-21T15:31:28 -!- tavish_ [~tavish@59.177.79.74] has quit [Quit: leaving]
2012-09-21T15:41:33 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-21T15:48:36 -!- gnomad [~gnomad@71.203.29.67] has joined ##stm32
2012-09-21T15:57:39 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-21T16:00:15 < Laurenceb> omg i found a photo of the kitchen cupboard at dong's house
2012-09-21T16:00:18 < Laurenceb> http://www.flickr.com/photos/bigaldaviesishere/5909416552/lightbox/
2012-09-21T16:04:20 < Laurenceb> "   # Import Psyco if available    try:        import psyco        psyco.full()        print "Using Psyco...""
2012-09-21T16:04:23 < Laurenceb> wtf python
2012-09-21T16:04:56 <+Steffanx> First of April?
2012-09-21T16:05:49 < Laurenceb> maybe
2012-09-21T16:11:11 < BrainDamage> psyco is actually serious
2012-09-21T16:11:17 < BrainDamage> it's a jit compiler
2012-09-21T16:11:19 < dongs> i think its some python compiler shit
2012-09-21T16:11:24 < dongs> yeah
2012-09-21T16:11:33 < BrainDamage> but it's outdated now
2012-09-21T16:11:39 < BrainDamage> it's replaced by pypy
2012-09-21T16:13:44 < Laurenceb> is it possible to compile python to an exe that will run as a standalone on windoze?
2012-09-21T16:14:21 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-21T16:14:22 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-21T16:14:22 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-21T16:14:50 < dongs> http://www.digikey.com/product-detail/en/B3U-1000P/SW1020CT-ND/1534357
2012-09-21T16:14:53 < dongs> is this a disconnect switch
2012-09-21T16:14:54 < dongs> Laurenceb: yeah
2012-09-21T16:15:08 < dongs> Off-Mom
2012-09-21T16:15:13 < dongs> im not into moms so i dunno wat the fuck that means
2012-09-21T16:16:39 <+Steffanx> I would day: yes
2012-09-21T16:17:11 < zyp> what are you daying?
2012-09-21T16:17:29 < Laurenceb> lol i have a load of those switches
2012-09-21T16:17:33 <+Steffanx> Oh, no i would say no dong
2012-09-21T16:17:42 <+Steffanx> SPST-NO = Normally Open
2012-09-21T16:18:08 <+Steffanx> At least .. that's what i make of "NO" :)
2012-09-21T16:19:48 <+Steffanx> but off-moms probably says it's 'Momentary' off
2012-09-21T16:19:56 <+Steffanx> And google seems to agree with that
2012-09-21T16:20:41 < dongs> so what the hel ldoes it mean in english
2012-09-21T16:26:46 < BrainDamage> Laurenceb: yes it's possible
2012-09-21T16:27:01 < dongs> Laurenceb: you do?
2012-09-21T16:27:08 < dongs> Laurenceb: those exact mdoel?
2012-09-21T16:27:18 < BrainDamage> it kinda cheats it makes a binary blob with the interpreter and the compiled python bytecode
2012-09-21T16:27:34 < dongs> duh..
2012-09-21T16:28:02 < BrainDamage> Laurenceb: http://www.py2exe.org/
2012-09-21T16:34:21 < Thorn> until this day I thought ubuntu was a relatively decent desktop os
2012-09-21T16:34:45 < BrainDamage> then you tried something else?
2012-09-21T16:35:15 < dongs> <Thorn> then I installed windows 7
2012-09-21T16:35:19 < BrainDamage> s/something/anything
2012-09-21T16:35:33 < Thorn> then I opened a .zip
2012-09-21T16:35:39 < Thorn> ????? ?? ?????.jpg
2012-09-21T16:35:45 <+Steffanx> gnome?
2012-09-21T16:35:47 < Thorn> etc.
2012-09-21T16:35:49 < Thorn> yes
2012-09-21T16:35:59 < BrainDamage> http://www.menuetos.net/ the final troll
2012-09-21T16:36:35 < BrainDamage> gui os in asm :s
2012-09-21T16:37:09 < Thorn> turns out only one specific version of unzip doesn't have this problem, and you need to uninstall p7zip-full and a lot of other stuff, and disable auto-update for several packages
2012-09-21T16:37:31 < Thorn> thankfully some kind soul make a script
2012-09-21T16:38:32 < dongs> wut?
2012-09-21T16:38:36 < dongs> what exactly is the peroblem
2012-09-21T16:38:39 < dongs> problem
2012-09-21T16:38:49 < dongs> that lunix doesnt support unicode?
2012-09-21T16:39:18 < Thorn> actually zip archives don't
2012-09-21T16:39:22 < Laurenceb> dongs: yes, its very easy to break
2012-09-21T16:39:29 < Laurenceb> by overheating when soldering
2012-09-21T16:39:37 < dongs> Laurenceb: no doubt
2012-09-21T16:39:41 < dongs> but i reflow so its ok
2012-09-21T16:39:56 < Laurenceb> yeah ive broken 2 hand soldering
2012-09-21T16:42:53 < zyp> do your solderjobs look as horrible as your code? :p
2012-09-21T16:43:21 < BrainDamage> without spaces? :p
2012-09-21T16:43:55 <+Steffanx> picts or it didn't happen
2012-09-21T16:43:59 <+Steffanx> -t
2012-09-21T16:44:57 < Laurenceb> https://github.com/Laurenceb/Dactyl/tree/master/Hardware/Assembly
2012-09-21T16:45:43 < Laurenceb> some of my soldering
2012-09-21T16:46:26 <+Steffanx> Ugly brown spots on the white pcb :P
2012-09-21T16:47:15 <+izua> BrainDamage: menuet is cool
2012-09-21T16:47:28 <+izua> it enforces fixed width fonts everywhere
2012-09-21T16:47:35 < Thorn> use a lightbox when photographing pcbs
2012-09-21T16:47:44 < BrainDamage> I hate fixed width fonts
2012-09-21T16:48:16 < Laurenceb> http://www.theregister.co.uk/2012/09/21/iphone_5_queue_london_launch_day/
2012-09-21T16:48:18 < Laurenceb> lulwut
2012-09-21T16:48:29 <+Steffanx> What's new Laurenceb ?
2012-09-21T16:49:01 < Laurenceb> not a lot
2012-09-21T16:49:06 < BrainDamage> "China has a one child policy," he shouted to the crowd, "which is a  problem because we need as many Chinese people as possible, to assemble  iPhone 5s."
2012-09-21T16:49:12 < Laurenceb> struggling to find motivation to do any work
2012-09-21T16:49:16 < Laurenceb> lol
2012-09-21T16:49:28 < Laurenceb> trollocoaster
2012-09-21T16:50:34 <+Steffanx> People should complain less
2012-09-21T16:50:56 <+Steffanx> Ignore and move on
2012-09-21T16:52:13  * zyp ignores Steffanx and moves on
2012-09-21T16:52:28 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: leaving]
2012-09-21T16:52:47 <+Steffanx> It's all up to you zyp . Can't care less :)
2012-09-21T16:56:56 <+Steffanx> I bet Laurenceb hacked a plugin together for his browser that does: "If page contains Apple => post link in ##stm32" :P
2012-09-21T16:57:57 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-21T16:58:19 < Laurenceb>  && page_contains_trolling()
2012-09-21T16:58:33 <+Steffanx> * && pageIsSortOfTrollable
2012-09-21T16:58:41 <+Steffanx> - spaces
2012-09-21T17:03:24 <+Steffanx> Laurenceb you also trolled the girl here in a village close to me?
2012-09-21T17:04:07 <+Steffanx> She placed an invitation for a party of facebook, but forgot to check the checkbox that says it was a private party :)
2012-09-21T17:04:21 <+Steffanx> Now they have NO idea how many people to expect
2012-09-21T17:04:33 < zyp> I assume you are going?
2012-09-21T17:04:51 <+Steffanx> No, I didn't accept the invitation :P
2012-09-21T17:04:53 <+Steffanx> No facebook
2012-09-21T17:05:05 < zyp> aww, poor Steffanx have no friends
2012-09-21T17:05:24 <+Steffanx> :(
2012-09-21T17:06:53 <+Steffanx> i have jpa- and you zyp :D
2012-09-21T17:07:19 < zyp> :D
2012-09-21T17:07:28 < BrainDamage> and Laurenceb and dongs are sorta the abusive uncles?
2012-09-21T17:07:59 <+Steffanx> I don't cam with them, you?
2012-09-21T17:10:34 < BrainDamage> cam?
2012-09-21T17:11:30 <+Steffanx> Yeah, isn't using a webcam the only way to for them to play "the abusive uncles"
2012-09-21T17:11:41 <+Steffanx> Anyway, this is going nowhere
2012-09-21T17:11:46 <+Steffanx> </end>
2012-09-21T17:11:56 <+Steffanx> *</conversation>
2012-09-21T17:13:14 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-21T17:17:52 < dongs> http://sphotos-a.ak.fbcdn.net/hphotos-ak-snc6/s480x480/198761_509327679097447_774196665_n.jpg
2012-09-21T17:19:20 <+Steffanx> That picture is reposted at least 100 times?
2012-09-21T17:19:30 <+Steffanx> jpeg artifacts ftw
2012-09-21T17:19:33 < dongs> you can tell by jpeg compression artifacts
2012-09-21T17:19:35 < dongs> yeah.
2012-09-21T17:20:18 <+Steffanx> https://lh5.googleusercontent.com/-DM89XA7sXPc/UFuZh7OuR9I/AAAAAAAAFZ0/4UVUGOARcH4/s738/A3QARhSCIAA2R9U.jpg less artifacts for free
2012-09-21T17:22:22 < Laurenceb> ok wtf
2012-09-21T17:22:28 < Laurenceb> format still failing
2012-09-21T17:22:28 < dongs> thanks, boner.
2012-09-21T17:22:31 < dongs> k goin to sleep
2012-09-21T17:22:32 < Laurenceb> Error creating file system: helper exited with exit code 1: helper failed with: Total number of sectors (8019968) not a multiple of sectors per track (62)! Add mtools_skip_check=1 to your .mtoolsrc file to skip this test mkfs.vfat 3.0.7 (24 Dec 2009)
2012-09-21T17:22:42 < dongs> haha..
2012-09-21T17:22:45 < dongs> > 2009
2012-09-21T17:22:55 < dongs> maybe next time you should try using mkfs.ext2fs
2012-09-21T17:22:59 < dongs> from like 1985
2012-09-21T17:23:12 < zyp> mkfs.murderfs
2012-09-21T17:23:17 < dongs> haha
2012-09-21T17:23:27 < dongs> i still have murderfs.com
2012-09-21T17:24:13 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has quit [Quit: ZNC - http://znc.in]
2012-09-21T17:24:21 <+Steffanx> Great website dong
2012-09-21T17:24:38 < Laurenceb> why does it not happen with a card reader?
2012-09-21T17:24:59 <+izua> 'lol murderfs
2012-09-21T17:25:03 <+Steffanx> Your cardreader is connected to a real os Laurenceb ?
2012-09-21T17:25:05 <+izua> what is that, the offspring of reiserfs?
2012-09-21T17:25:14 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has joined ##stm32
2012-09-21T17:25:18 < dongs> izua: yeah
2012-09-21T17:28:29 < dongs> lol, news to me: http://usnews.nbcnews.com/_news/2012/07/17/12794688-software-guru-ordered-to-pay-60-million-to-his-kids-over-wifes-death
2012-09-21T17:28:39 < dongs> reiser is fucked, how is he gonna make 60mil while hes in jail
2012-09-21T17:29:06 < BrainDamage> kickstarter
2012-09-21T17:29:14 < dongs> ahaha.
2012-09-21T17:31:17 < Oldboy> this thing never ends
2012-09-21T17:31:24 < Oldboy> wtf dongs you have a lot of spare time.
2012-09-21T17:31:30 < dongs> wat
2012-09-21T17:38:02 -!- claude_work is now known as claude
2012-09-21T17:38:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-21T17:50:42 -!- bgamari_ [~ben@physicsnat56.physics.umass.edu] has joined ##stm32
2012-09-21T17:52:43 -!- bgamari_ [~ben@physicsnat56.physics.umass.edu] has quit [Client Quit]
2012-09-21T17:58:12 < jpa-> Steffanx: i'm your friend? yaayyyy
2012-09-21T17:58:22 <+Steffanx> You didn't know what?
2012-09-21T17:58:24 <+Steffanx> *that
2012-09-21T17:58:35 <+Steffanx> What kind of friend are you?
2012-09-21T18:24:20 < Laurenceb> ha fixed my format
2012-09-21T18:24:32 < Laurenceb> emtire file system was offset by one sector
2012-09-21T18:24:37 < Laurenceb> *entire
2012-09-21T18:25:12 < Laurenceb> crazy that it worked at all
2012-09-21T18:34:20 < Laurenceb> ukhas conference tomorrow
2012-09-21T18:34:27 < Laurenceb> apparently the register will be coming
2012-09-21T18:34:54 < Laurenceb> 5am start 4 me :(
2012-09-21T18:35:54 <+Steffanx> lol what a fail
2012-09-21T18:36:15 <+Steffanx> Tell your username and password on ustream.tv with 50k viewers :D
2012-09-21T18:50:20 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-21T19:02:15 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-21T19:04:07 < TitanMKD> hi
2012-09-21T19:04:07 -!- DaKu is now known as daku
2012-09-21T19:04:08 < TitanMKD> I just received and mounted my case for my STM32F4 Debug board see the real hw photo on my blog http://titanmkd.blogspot.com
2012-09-21T19:04:25 < TitanMKD> for info case+pcb cost is about 3USD ;)
2012-09-21T19:04:33 < TitanMKD> 4 USD sorry ;)
2012-09-21T19:07:01 < Laurenceb> dongs: where is your usart bootloader code ?
2012-09-21T19:07:15 < Laurenceb> for activating the bootloader without using boot0
2012-09-21T19:08:53 -!- BjoernC [~BjoernC@dyndsl-178-142-054-097.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-21T19:10:28 < emeb> TitanMKD: Purty - time to warm up that soldering iron and make the thing work!
2012-09-21T19:11:09 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-21T19:13:18 < TitanMKD> emeb yes i have even smt stencil for it ;)
2012-09-21T19:13:32 < TitanMKD> emeb it will look like professional board at end
2012-09-21T19:13:52 < emeb> Bah - just do it by hand. You'll have it built in an hour or two.
2012-09-21T19:14:02 < TitanMKD> no there's QFN ;)
2012-09-21T19:14:21 < TitanMKD> for the USB charging stuff
2012-09-21T19:14:34 < emeb> I've done QFN by hand. Takes a little more care.
2012-09-21T19:14:52 < emeb> But, hand-applied paste + hotplate works quick too.
2012-09-21T19:15:00 < TitanMKD> yes with hot air but i prefer to do it using smt stencil and my hot plate
2012-09-21T19:15:22 < TitanMKD> faster and better result
2012-09-21T19:15:28 < emeb> no doubt
2012-09-21T19:16:37  * emeb has seen BGA done by hand. dead-bug w/ fine gauge wire. eek!
2012-09-21T19:17:13 < TitanMKD> hehe i have also a LBGA 256 board to populate ;)
2012-09-21T19:36:46 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-21T19:38:27  * Laurenceb epic fail
2012-09-21T19:38:48 < Laurenceb> ive got a bluetooth module running off an smps thats driven by the stm32
2012-09-21T19:38:56 < Laurenceb> i want to flash the stm32 over bluetooth
2012-09-21T19:39:03 < Laurenceb> ... see where this is going?
2012-09-21T19:39:37 < Laurenceb> bluetooth gets killed by itself
2012-09-21T19:39:39 < zyp> towards more failing?
2012-09-21T19:39:47 < Laurenceb> when it pust stm32 into reset
2012-09-21T19:39:52 < Laurenceb> *puts
2012-09-21T19:40:02 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-21T19:40:06 <+izua> how simple is the external smps hardware?
2012-09-21T19:40:31 < Laurenceb> lt3467a
2012-09-21T19:40:43 < Laurenceb> i think the best way to make this work is like so
2012-09-21T19:40:51 < Laurenceb> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/STM32%20Internal%20Bootloader&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=2980
2012-09-21T19:41:27 < Laurenceb> only call a bootloader routine that sets the pin upo before jumping directly
2012-09-21T19:42:17 < BrainDamage> Laurenceb: stick a supercap to the  bluetooth module then
2012-09-21T19:42:34 < Laurenceb> heh
2012-09-21T19:42:38 <+Steffanx> huh
2012-09-21T19:42:40 <+Steffanx> hoh
2012-09-21T19:43:04 < BrainDamage> really, you can get like 0.5F, 5.5V caps for cheap nowdays
2012-09-21T19:43:39 < Laurenceb> bbl
2012-09-21T19:50:18 -!- gnomad [~gnomad@71.203.29.67] has quit [Quit: This computer has gone to sleep]
2012-09-21T20:01:17 <+izua> Report: iPhone 5 lines shorter than expected as early adopters try to use iOS 6 Maps to locate Apple Stores.
2012-09-21T20:02:57 < emeb> :P
2012-09-21T20:05:01 <+Steffanx> I actually 'know' someone who went to germany for one. People are weird
2012-09-21T20:05:36 <+Steffanx> Especially when they sleep in from of the store
2012-09-21T20:08:16 < emeb> misplaced priorities.
2012-09-21T20:10:12 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 248 seconds]
2012-09-21T20:11:34 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-21T20:16:29 -!- Avi0 [~Avi0@c122-107-128-37.eburwd5.vic.optusnet.com.au] has joined ##stm32
2012-09-21T20:16:51 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-21T20:16:53 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-21T20:20:00 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-21T20:23:30 -!- Avi0 [~Avi0@c122-107-128-37.eburwd5.vic.optusnet.com.au] has left ##stm32 ["Leaving..."]
2012-09-21T20:25:43 < ossifrage> anyone have some useful gdb aliases for debugging various system faults? like something that will decode some of the status registers?
2012-09-21T20:38:19 < jpa-> i've thought about it but haven't done it
2012-09-21T20:38:30 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 264 seconds]
2012-09-21T20:50:22 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-21T20:59:37 < ossifrage> jpa-, I wonder if ou can do it with the system description xml
2012-09-21T20:59:49 < ossifrage> add extra defs for common registers
2012-09-21T21:25:54 -!- barthess [~barthess@178.154.109.233] has joined ##stm32
2012-09-21T21:28:57 -!- tavish_ [~tavish@120.56.134.31] has joined ##stm32
2012-09-21T21:29:05 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Reconnecting]
2012-09-21T21:30:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-21T21:30:23 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-21T22:14:14 -!- dekar [~dekar@212.255.47.165] has joined ##stm32
2012-09-21T22:14:17 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-21T22:14:42 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-21T22:25:39 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-21T22:26:45 < Laurenceb_> hi
2012-09-21T22:27:23 <+Steffanx> Lo
2012-09-21T22:44:17 -!- tavish_ [~tavish@120.56.134.31] has quit [Quit: zz]
2012-09-21T23:22:22 < zyp> ossifrage, I've been thinking about the same
2012-09-21T23:22:48 < zyp> but I haven't generated any faults lately, so I haven't had to debug them :p
2012-09-21T23:49:23  * Laurenceb_ is getting confused by reset code
2012-09-21T23:49:25 < Laurenceb_> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex_mx_stm32%2FHelp%20with%20application%20jump%20to%20system%20memory%20boot&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=291
2012-09-21T23:49:43 < Laurenceb_> EXPORT  Reboot_Loader
2012-09-21T23:49:43 < Laurenceb_>                 LDR     R0, =0x1FFFF000 ; ROM Base
2012-09-21T23:49:43 < Laurenceb_>                 LDR     SP,[R0, #0]
2012-09-21T23:49:43 < Laurenceb_>                 LDR     R0,[R0, #4]
2012-09-21T23:49:43 < Laurenceb_>                 BX      R0
2012-09-21T23:50:12 < Laurenceb_> ^doesnt that set the stack pointer to 0x1ffff004
2012-09-21T23:50:42 < Laurenceb_> i mean 0x1ffff000
2012-09-21T23:58:11 < Thorn> it sets sp to contents of memory at that address, i.e. sp = *r0;
2012-09-21T23:59:39 < Laurenceb_> oh
2012-09-21T23:59:43  * Laurenceb_ fails at asm
--- Day changed Sat Sep 22 2012
2012-09-22T00:00:28 < Thorn> and then jumps to r0[1], which is logical if the vector table is at that address
2012-09-22T00:03:39 < Laurenceb_> wait
2012-09-22T00:04:03 < Laurenceb_> so address 0x1ffff000 is loaded with 0x1ffff000?
2012-09-22T00:04:23 < Thorn> wat no
2012-09-22T00:04:26 < Laurenceb_> LDR     R0,[R0, #4]
2012-09-22T00:04:33 < Thorn> r0 = r0[1]
2012-09-22T00:05:13 < Thorn> (if that were possible in c without casts)
2012-09-22T00:06:08 < Laurenceb_> hmm
2012-09-22T00:06:19 < Laurenceb_> it has to be 0x1ffff004
2012-09-22T00:06:32 < zyp> Laurenceb_, contents of it
2012-09-22T00:06:37 < zyp> [] means contents of
2012-09-22T00:06:46 < Laurenceb_> i see
2012-09-22T00:06:51 < zyp> so [r0, #4] means contents of r0+4
2012-09-22T00:07:35 <+izua> http://imgur.com/xSjIP
2012-09-22T00:07:36 <+izua> so
2012-09-22T00:07:37 < zyp> r0 = 0x1ffff000; r0 = *r0; r0 = *(r0 + 4); r0();
2012-09-22T00:07:40 <+izua> what is this and what does it do?
2012-09-22T00:07:48 < zyp> that's pretty much what your code does
2012-09-22T00:07:53 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-22T00:08:14 < Thorn> r0 = 0x1ffff000; sp = *r0; r0 = *(r0 + 4); r0();
2012-09-22T00:08:20 < zyp> it loads the second entry from a vector table located at 0x1ffff000 and calls it
2012-09-22T00:08:39 < zyp> and as we all know, second entry in a vector table is the reset vector
2012-09-22T00:08:42 < Laurenceb_> oh
2012-09-22T00:08:56 < zyp> i.e. the entry function of the rom bootloader
2012-09-22T00:09:22 < Laurenceb_> i see now
2012-09-22T00:10:16 < Laurenceb_> http://www.youtube.com/watch?feature=player_detailpage&v=cvKC-4tCRgw#t=168s
2012-09-22T00:10:21 < Laurenceb_> but how does that work?
2012-09-22T00:10:40 < zyp> how does what work?
2012-09-22T00:10:43 < Laurenceb_> oh
2012-09-22T00:10:48 < Laurenceb_> thats doing the same thing
2012-09-22T00:10:56 < zyp> yes.
2012-09-22T00:11:24 < Laurenceb_> i get it now
2012-09-22T00:25:15 < Laurenceb_> according to the video it works
2012-09-22T00:25:23 < Laurenceb_> but the st forum says it doesnt...
2012-09-22T00:25:35 < Laurenceb_> im wondering if its do do with sram initialisation
2012-09-22T00:25:54 < Laurenceb_> wait no it cant be or the reset asm trick with 0xDEADBEEF wouldnt work
2012-09-22T00:26:06 < Laurenceb_> that guy on the forum probably screwed up
2012-09-22T00:26:41 < Laurenceb_> i need to jump directly from application code so as to keep the bluetooth working
2012-09-22T00:29:48 < Laurenceb_> his RCC code doesnt look right to me for a start
2012-09-22T00:43:57 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-22T00:46:56 -!- barthess [~barthess@178.154.109.233] has quit [Quit: Leaving.]
2012-09-22T01:19:18 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-22T01:25:57 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 248 seconds]
2012-09-22T01:31:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-22T01:31:44 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-22T01:40:12 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-22T02:20:56 < zyp> hmm, now I got F4 enumerating as an audio device
2012-09-22T02:21:35 < zyp> maybe I should write up a demo for the discovery board then
2012-09-22T02:23:56 < emeb> zyp: cool. pls do.
2012-09-22T02:28:06 < zyp> it's a bit interesting how different operating systems handle insufficient bandwidth
2012-09-22T02:29:07 < emeb> mostly *choke* ?
2012-09-22T02:29:20 < zyp> OS X refused to output anything at all, while linux outputted at the maximum possible speed, which was 1/3 of required
2012-09-22T02:30:16 < zyp> I had max packet size set to 64B, which translates to max 64 kB/s, and the audio stream was 192 kB/s (stereo 16bit 48kHz)
2012-09-22T02:31:17 < emeb> is that just a value the device sends back to the host in the enumeration?
2012-09-22T02:31:49 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-22T02:31:59 < HTT-Bird> are FTDI's HID-based, CDC based, or a hybrid?
2012-09-22T02:32:41 < zyp> neither
2012-09-22T02:32:46 < zyp> they are vendor specific
2012-09-22T02:33:34 < HTT-Bird> ah.
2012-09-22T02:33:58 < HTT-Bird> on a somewhat different note: is it possible to run non-UART (say SPI, I2C, or JTAG) serial interfaces over CDC?
2012-09-22T02:35:02 < zyp> CDC contain a bunch of different subclasses, ACM is the serial port emulation
2012-09-22T02:35:27 < HTT-Bird> yeah.  all the other CDC subclasses are for Ethernet, ISDN, ATM, networking things
2012-09-22T02:35:34 < zyp> yep
2012-09-22T02:36:21 < HTT-Bird> the CDC spec does allow for custom subclasses it seems?!
2012-09-22T02:36:29 < zyp> unsure
2012-09-22T02:36:36 < HTT-Bird> makes 2 of us
2012-09-22T02:36:49 < zyp> but what's the point? if it's custom, why don't go full vendor specific?
2012-09-22T02:37:08 < HTT-Bird> zyp: hrm.  you have a point there.
2012-09-22T02:37:37 < zyp> I have custom control requests for doing both I2C and JTAG on my board
2012-09-22T02:37:47 < HTT-Bird> *nods*
2012-09-22T02:38:00 < HTT-Bird> I'd figure you'd want a bulk pipe or two for JTAG at decent speeds tho
2012-09-22T02:38:05 < zyp> http://cgit.jvnv.net/suzumebachi/tree/main.cpp#n48
2012-09-22T02:38:28 < zyp> I don't care so much for speed, I only use it for CPLD programming
2012-09-22T02:38:32 < HTT-Bird> *nods*
2012-09-22T02:50:30 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 264 seconds]
2012-09-22T03:25:13 < dongs> Laurenceb: mine is a mix of shit inside usermode and  in startup asm
2012-09-22T03:25:35 < dongs> http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/drv_system.c#159
2012-09-22T03:25:48 < dongs> the proper way would be to use backup memory to pass the boot flag
2012-09-22T03:26:01 < dongs> and then do bootloader jump in usermode after checking the said backup memory
2012-09-22T03:26:09 < dongs> BKP stuff.
2012-09-22T03:26:50 < dongs> http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/baseflight_startups/startup_stm32f10x_md.s#151
2012-09-22T03:27:36 < zyp> now he just needs to steal your failureMode(), and he'll be all set
2012-09-22T03:30:34 < dongs> lool
2012-09-22T04:15:45 -!- DLPeterson [~hazelnuss@70.42.240.22] has joined ##stm32
2012-09-22T04:17:53 < DLPeterson> when configuring a pin to be a AF input on the STM32F1, the port gets configured in input mode.  On the STM32F4, it seems like you have to configure it to be in AF PP mode.  Is anybody here familiar with Alternate function inputs on STM32F4?
2012-09-22T04:36:01 < upgrdman> i am, a little
2012-09-22T04:36:13 < upgrdman> i've used timers
2012-09-22T04:36:20 < DLPeterson> upgrdman, in input mode?
2012-09-22T04:36:31 < upgrdman> input and output
2012-09-22T04:36:36 < upgrdman> output pwm
2012-09-22T04:37:03 < upgrdman> input to start a counter based on a button press, with some extras
2012-09-22T04:37:38 < upgrdman> ermm ... by "extras" i mean the longer you hold the button the quicker a value increments or decrements... don't know what the name for that is
2012-09-22T04:37:51 < DLPeterson> have you used a TIM in input capture mode?
2012-09-22T04:38:43 < upgrdman> i've triggered on timXch1 ... i think that would be input capture mode?
2012-09-22T04:38:59 < upgrdman> o wait
2012-09-22T04:39:33 < upgrdman> i have also done externally clocked counters
2012-09-22T04:41:32 < upgrdman> here are my stm32f4 notes: http://www.farrellf.com/projects/hardware/ ... my notes aren't that good and perhaps even misleading because i oversimpify things. but maybe it will help you.
2012-09-22T04:41:57 < upgrdman> i have not posted my pwm notes though
2012-09-22T04:44:06 < DLPeterson> upgrdman, thanks
2012-09-22T04:46:29 < upgrdman> good luck
2012-09-22T04:48:34 < DLPeterson> upgrdman, thanks.  From what I can tell, you just have to leave it in output push pull mode, even though you are using it as an alternate function input.  RM0009 Table 14 makes it look like there is no such thing as an alternate function input, but all the alternate function input examples I have seen keep the pin in AF PP mode with no pull up or pull down enabled.
2012-09-22T04:48:51 < upgrdman> ya
2012-09-22T04:49:00 < upgrdman> for example, with pwm
2012-09-22T04:49:18 < upgrdman> you see a bit in timer register to make it an output
2012-09-22T04:49:22 < upgrdman> s/see/set
2012-09-22T04:50:01 < upgrdman> such as:
2012-09-22T04:50:03 < upgrdman> TIM5->CCMR1 |= TIM_CCMR1_OC2M_1 | TIM_CCMR1_OC2M_2;		// OC2M = 110 for PWM Mode 1 output on ch2
2012-09-22T04:50:16 < upgrdman> TIM5->CCER |= TIM_CCER_CC2E;							// Enable output for ch2
2012-09-22T04:50:34 < DLPeterson> yeah, that part makes sense, it is more the GPIO configuration that is a bit confusing
2012-09-22T04:50:48 < DLPeterson> check out Table 14 of RM0009
2012-09-22T04:51:30 < DLPeterson> anyway, I've got to go, I'll be back later....
2012-09-22T04:51:37 < upgrdman> ya
2012-09-22T04:51:41 < upgrdman> k
2012-09-22T04:51:45 -!- DLPeterson [~hazelnuss@70.42.240.22] has quit [Quit: leaving]
2012-09-22T05:30:52 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-09-22T05:36:29 < emeb> heh - frequency domain audio processing (FFT/IFFT) on an F405 working. Simple lowpass filter w/ variable cutoff & transition band.
2012-09-22T05:36:44 < emeb> sounds very... sterile.
2012-09-22T05:37:02 < akaWolf> sterile =)
2012-09-22T05:37:33 < emeb> filter doesn't have a lot of character. Probably need to try putting in some resonance.
2012-09-22T05:38:40 < emeb> Tried a phase vocoder. Must need better source material, but it basically cut off everything except some low-end drums.
2012-09-22T05:43:33 < akaWolf> so what do you want?
2012-09-22T05:43:58 < emeb> Nothing in particular. Just playing to see if it's worth pursuing further.
2012-09-22T05:44:10 < akaWolf> a...
2012-09-22T05:49:57 < akaWolf> DAC or PWM?
2012-09-22T05:51:21 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-22T05:52:11 < akaWolf> or only computation without sound out?
2012-09-22T05:52:33 < emeb> I2S codec
2012-09-22T06:09:33 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 272 seconds]
2012-09-22T06:16:14 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds]
2012-09-22T06:21:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-22T06:31:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-22T06:31:13 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-22T06:31:42 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-22T06:39:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-22T06:50:55 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-22T06:58:32 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-22T07:19:49 < dongs> sup dix
2012-09-22T07:22:32 < dongs> http://25.media.tumblr.com/tumblr_mapkrygesc1rhptwbo1_1280.jpg haha
2012-09-22T07:22:37 < dongs> San Francisco, after the earthquake. 9/20/2012. Never forget.
2012-09-22T07:30:07 < dongs> oooooooooooooooooooooold
2012-09-22T07:53:40 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 248 seconds]
2012-09-22T08:05:01 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-22T08:18:17 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 252 seconds]
2012-09-22T08:39:37 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-22T08:43:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-22T09:01:40 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-22T09:03:09 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-22T09:10:41 -!- dekar_ [~dekar@212.255.31.84] has joined ##stm32
2012-09-22T09:10:58 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-09-22T09:14:44 -!- dekar [~dekar@212.255.47.165] has quit [Ping timeout: 248 seconds]
2012-09-22T09:19:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-22T09:58:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-22T10:27:59 < dongs> dix
2012-09-22T11:36:37 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-22T11:36:40 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-22T11:53:06 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has joined ##stm32
2012-09-22T12:14:07 -!- barthess [~barthess@178.154.40.7] has joined ##stm32
2012-09-22T13:07:10 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-22T13:14:35 -!- barthess [~barthess@178.154.40.7] has quit [Ping timeout: 268 seconds]
2012-09-22T13:58:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-22T13:59:58 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-09-22T14:22:43 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-09-22T14:34:00 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-22T14:35:31 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-22T14:36:37 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 246 seconds]
2012-09-22T14:37:33 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-22T14:45:51 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 268 seconds]
2012-09-22T14:47:06 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Read error: Connection reset by peer]
2012-09-22T14:48:12 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-22T14:48:14 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-22T14:50:15 -!- claude is now known as Claude
2012-09-22T14:51:14 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-22T14:55:19 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-22T15:12:34 -!- zyp [zyp@zyp.im] has quit [Remote host closed the connection]
2012-09-22T15:13:43 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 244 seconds]
2012-09-22T15:15:34 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-22T15:22:21 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-22T15:22:21 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-22T15:22:21 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-22T15:22:24 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-22T15:46:15 -!- daku is now known as DaKu
2012-09-22T15:54:05 < dongs> attn ausfags http://www.youtube.com/watch?v=BzI04jXR-y0
2012-09-22T15:54:10 -!- zyp [zyp@zyp.im] has joined ##stm32
2012-09-22T15:54:29 -!- DaKu is now known as daku
2012-09-22T15:58:47 <+Steffanx> When this Gangnam crap is finally over?!
2012-09-22T15:59:29 -!- daku is now known as DaKu
2012-09-22T16:05:00 < dongs> loo
2012-09-22T16:06:43 <+Steffanx> I always have to look into the urbandictionary when dongs says something i don't understand
2012-09-22T16:06:49 <+Steffanx> too bad it doesnt help me this time
2012-09-22T16:06:53 < dongs> has cjbeard been here
2012-09-22T16:06:53 < dongs> lately
2012-09-22T16:06:53 < dongs> or did he /ragequit life over assange
2012-09-22T16:08:32 <+Steffanx> He only rageparted this channel
2012-09-22T16:08:55 < dongs> oh, he is on
2012-09-22T16:08:58 < dongs> hmm
2012-09-22T16:10:01 < dongs> wtf i cant /invite without ops
2012-09-22T16:10:05 < dongs> waht fucking jewish ircd is this
2012-09-22T16:10:15 <+Steffanx> wieso?
2012-09-22T16:11:44 <+Steffanx> You want him to be here dongs?
2012-09-22T16:12:01 -!- mode/##stm32 [+o Steffanx] by ChanServ
2012-09-22T16:12:01 <@Steffanx> :)
2012-09-22T16:12:33 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-22T16:12:51 -!- mode/##stm32 [-o Steffanx] by ChanServ
2012-09-22T16:12:58 <+Steffanx> Oh, he has auto join on invite :)
2012-09-22T16:13:02 < dongs> lol
2012-09-22T16:13:13 < dongs> cjbeard: you were missed
2012-09-22T16:13:21 < dongs> http://i.imgur.com/1tVdw.jpg awesome software name
2012-09-22T16:13:59 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-22T16:14:43 < zyp> oh, they have ported jubeat to android now? nice?
2012-09-22T16:14:53 < zyp> -?
2012-09-22T16:15:07 <+Steffanx> I wonder if the phone is really the thing the people look at when they see that picture :D
2012-09-22T16:15:22 -!- barthess [~barthess@178.154.15.230] has joined ##stm32
2012-09-22T16:15:32 < BrainDamage> there's a phone?
2012-09-22T16:15:35 < jpa-> i looked at the tits
2012-09-22T16:15:53 < dongs> http://i.imgur.com/yAANz.jpg she looked better.
2012-09-22T16:16:15 < dongs> "we're seeking fun"
2012-09-22T16:16:26 <+Steffanx>  You are biased dongs.. you ARE japanese
2012-09-22T16:17:04 <+Steffanx> We know 'better' :P
2012-09-22T16:18:15 < dongs> there was a lot that looked like men.
2012-09-22T16:18:19 < jpa-> wait.. dongs is japanese... wow! that explains everything
2012-09-22T16:18:35 <+Steffanx> </sarcasm>
2012-09-22T16:20:04 < dongs> http://i.imgur.com/OEgN0.jpg
2012-09-22T16:21:41 <+Steffanx> A Spaceship crashed into the water?
2012-09-22T16:22:28 <+Steffanx> Nevermind, ignore me. I should do my homework
2012-09-22T16:22:50 < dongs> homowork
2012-09-22T16:23:37 <+Steffanx> How large is your dictionary dongs?
2012-09-22T16:23:51 < dongs> not very large,
2012-09-22T16:24:12 <+Steffanx> I pee
2012-09-22T16:27:06 <+Steffanx> Hmm, why only the large L1's have built-in opamps?
2012-09-22T16:27:52 < jpa-> no idea
2012-09-22T16:27:57 < zyp> because the small ones lacks the pins?
2012-09-22T16:28:02 < zyp> or area?
2012-09-22T16:28:07 < BrainDamage> opamps compensation caps take a shitload of area
2012-09-22T16:28:23 < BrainDamage> at least compared to digital circuits
2012-09-22T16:28:24 <+Steffanx> lqfp 64 vs 48
2012-09-22T16:28:35 < BrainDamage> I don't know how much is the die area vs package in stm32
2012-09-22T16:28:57 <+Steffanx> Don't know
2012-09-22T16:29:52 <+Steffanx> zyp, actually lqfp-48 is large if you only need a few pins :P
2012-09-22T16:30:40 < jpa-> maybe the opamp versions don't really exist
2012-09-22T16:30:46 < jpa-> they are all "non-stock" on digikey
2012-09-22T16:30:57 <+Steffanx> Could be
2012-09-22T16:31:42 < jpa-> on farnell also
2012-09-22T16:40:02 < zyp> hmm, I've never seen the cpu crash with NMI before
2012-09-22T16:49:14 < zyp> oh well, now I have usb audio actually outputting received audio data over I2S
2012-09-22T16:53:04 < dongs> watcha hacking usb adio for
2012-09-22T16:53:12 < zyp> for fun
2012-09-22T16:53:31 < zyp> http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_audio
2012-09-22T16:54:47 < zyp> usb by itself is very useless without any class drivers, so I'm adding class drivers
2012-09-22T17:00:49 < cjbaird> What the hell? I deleted stm32 from my IRC autologin..
2012-09-22T17:01:00 < jpa-> cjbaird: you can't leave
2012-09-22T17:01:04 <+izua> someone missed you and invited you back
2012-09-22T17:01:24 < dongs> cjbaird: why would you do that :(
2012-09-22T17:01:54 < dongs> do you hate  me that much
2012-09-22T17:02:15 < cjbaird> Nah, I just took leave from the channel while I wasn't doing anything really stm32 related..
2012-09-22T17:02:20 -!- Irssi: ##stm32: Total of 62 nicks [1 ops, 0 halfops, 3 voices, 58 normal]
2012-09-22T17:02:43 < cjbaird> I caught myself randomizing the channel too often. :/
2012-09-22T17:02:58 < dongs> ha ha. its not like anyone actually does anything stm32 r elated in here
2012-09-22T17:05:31 < cjbaird> I spent the day test-riding tiny scooters (http://www.sydneyscooterexpo.com/) .. I'm looking for a replacement for when the Postie craps out something expensive to repair.. Sym Symba, Kymco Like125, Honda PCX125 and Dio110. The Dio (..Brando) won it hands-down.
2012-09-22T17:09:13 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 252 seconds]
2012-09-22T17:12:50 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-22T17:15:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-22T17:16:29 < cjbaird> I'll be arsed if I can find where the preference setting in konversation is for disabling that autojoin invite shit..
2012-09-22T17:27:32 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds]
2012-09-22T17:28:07 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-22T17:28:58 < dongle> fucking feenode
2012-09-22T17:46:21 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-22T18:02:09 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 244 seconds]
2012-09-22T18:41:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-22T18:50:17 -!- barthess [~barthess@178.154.15.230] has quit [Quit: Leaving.]
2012-09-22T18:52:13 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-22T18:52:14 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-22T18:52:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-22T18:53:23 -!- Simonious [~simonious@c-67-170-182-171.hsd1.or.comcast.net] has joined ##stm32
2012-09-22T18:59:10 -!- Simonious [~simonious@c-67-170-182-171.hsd1.or.comcast.net] has left ##stm32 []
2012-09-22T19:03:14 < zyp> ok, shit actually works
2012-09-22T19:03:34 < zyp> currently playing music through f4discovery
2012-09-22T19:04:31 < jpa-> usb is a strange beast.. it takes a lot of work to get stuff working, but seems quite simple at the end
2012-09-22T19:06:09 <+Steffanx> "<cjbaird> I caught myself randomizing the channel too often. :/"  I guess i have to leave as well
2012-09-22T19:07:15 < BrainDamage> I think I hold the record on OT talk here, at least Laurenceb provides us a blag about his failures
2012-09-22T19:07:26 < BrainDamage> I keep them for myself :p
2012-09-22T19:07:36 < BrainDamage> them = my failures
2012-09-22T19:08:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-22T19:12:09 < Oldboy> to not for
2012-09-22T19:13:16 <+Steffanx> What does the Audio Class send zyp ?
2012-09-22T19:13:30 <+Steffanx> Just some data you can forward to an DAC?
2012-09-22T19:13:34 <+Steffanx> *a
2012-09-22T19:13:49 < zyp> Steffanx, depends on the descriptors
2012-09-22T19:13:56 <+Steffanx> Of course :D
2012-09-22T19:14:33 < jpa-> i guess it's important to name them "asdf" and "abc" for proper function
2012-09-22T19:14:35 < zyp> but yes, mine describes that it expects stereo 16bit 48 kHz PCM
2012-09-22T19:15:46 < zyp> jpa-, ha, actually it's «audio control descriptor» and «audio streaming format descriptor» and «audio control endpoint descriptor»
2012-09-22T19:15:51 < zyp> :p
2012-09-22T19:15:56 < jpa-> what, really? :D
2012-09-22T19:16:10 < jpa-> and foo_t is field order object?
2012-09-22T19:16:14 < zyp> well, that's what I was thinking when I wrote them
2012-09-22T19:16:15 < zyp> :D
2012-09-22T19:16:50 < zyp> well, I got tired of writing descriptor generators after doing the as_general_desc() :p
2012-09-22T19:16:57 < zyp> so I just hardcoded the rest to test
2012-09-22T19:17:32 -!- barthess [~barthess@178.154.15.230] has joined ##stm32
2012-09-22T19:17:35 < zyp> I prefer the «get shit working first, make them pretty afterwards» method :p
2012-09-22T19:18:54 < jpa-> yeah.. you are the last person i could blame for not making it pretty eventually
2012-09-22T19:19:12 < jpa-> even with those hardcoded descriptors the code is very readable :)
2012-09-22T19:19:35 < zyp> :)
2012-09-22T19:19:44 <+Steffanx> When you know what they mean, yes
2012-09-22T19:20:11 < zyp> Steffanx, you don't have to know what they mean, you just need to see what you can ignore :p
2012-09-22T19:20:24 < jpa-> yeah well.. as far as it's just the descriptors i know that i can look it up in USB standards if i want to
2012-09-22T19:20:27 <+Steffanx> Of course I have to know
2012-09-22T19:20:48 < emeb> zyp: you actually getting audio out?
2012-09-22T19:21:01 < zyp> of course I am
2012-09-22T19:21:05 < zyp> 18:03:34 < zyp> currently playing music through f4discovery
2012-09-22T19:21:24 < emeb> cool
2012-09-22T19:21:56 < zyp> there is no sync between usb and i2s though, so it creates some distortion when usb is writing in the same area of the ringbuffer as i2s is reading
2012-09-22T19:22:01 < jpa-> whereas code like this makes me want to puke: http://paste.dy.fi/P2a
2012-09-22T19:22:08 < zyp> but most of the time they aren't
2012-09-22T19:22:36 <+Steffanx> Just because of the ifdef's, that make it unreadable jpa- ?
2012-09-22T19:23:30 < emeb> the old isochronous data issue
2012-09-22T19:23:55 < jpa-> Steffanx: partly, and because if you leave out the ifdefs it doesn't really do anything.. so you don't know at all which parts to skip and what is important
2012-09-22T19:24:01 < zyp> if any of you want to test on f4discovery but can't be assed to compile the code, you can flash this: http://bin.jvnv.net/f/TcZX3/demo.elf
2012-09-22T19:24:04 < zyp> :p
2012-09-22T19:24:06 < emeb> thought there was some sort of PLL feature to get around that.
2012-09-22T19:24:48 < zyp> there is a seperate I2S PLL, but I'm just running it with hardcoded values
2012-09-22T19:25:32 < emeb> where's the source again?
2012-09-22T19:25:53 < zyp> http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_audio
2012-09-22T19:25:54 < emeb> nm
2012-09-22T19:25:59 < emeb> yep just found it
2012-09-22T19:27:39 < emeb> mysterious descriptors
2012-09-22T19:28:37 < zyp> oh, and for you guys that love to compare sizes, I'm using 8600 bytes of flash :p
2012-09-22T19:29:28 <+Steffanx> With yagarto?
2012-09-22T19:29:37 < emeb> winner has the smallest. that's a change.
2012-09-22T19:30:45 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has joined ##stm32
2012-09-22T19:31:13 < jpa-> my chibios-based stuff bloats around 40kB and it doesn't even do anything special
2012-09-22T19:32:04 < jpa-> (but i have to admit it's my own code that has the bloat)
2012-09-22T19:33:36 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-09-22T19:33:49 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-22T19:35:39 < emeb> ok, why's objdump generating a bin file that 134MB?
2012-09-22T19:36:52 < jpa-> because you are dumping the whole address range, instead of your data only?
2012-09-22T19:37:04 < jpa-> (whole = significant part of)
2012-09-22T19:37:08 < emeb> that's what it looks like
2012-09-22T19:37:54 < emeb> hrm - what flags to tell it only flash?
2012-09-22T19:38:15 < zyp> -j .text -j .data
2012-09-22T19:38:19 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-22T19:39:11 < emeb> ty
2012-09-22T19:39:54 < emeb> that happened when I started using CCM to store an audio buffer.
2012-09-22T19:40:14 < jpa-> there is probably a bug in your linker script
2012-09-22T19:40:21 < emeb> no doubt
2012-09-22T19:40:54 < emeb> I was just tweaking sections. Probably told it something bogus.
2012-09-22T19:41:24 < jpa-> you probably need to specify something like  >ccm AT >rom 
2012-09-22T19:42:33 < jpa-> (and if you want initialized variables to work, also modify the startup code)
2012-09-22T19:49:38 < emeb> I'll check. Don't want to init that - just temp storage.
2012-09-22T19:51:45  * emeb doesn't know enough about .ld files
2012-09-22T20:11:31 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-22T20:24:41 < zippe> emeb: paste your linker script somewhere
2012-09-22T20:39:51 < jpa-> lol wtf
2012-09-22T20:39:59 < jpa-> this usb cable has 2.4 ohms end-to-end resistance
2012-09-22T20:41:04 < jpa-> on each pin
2012-09-22T20:44:40 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-22T21:00:38 < Bird|lappy> jpa-, heh
2012-09-22T21:01:47 < jpa-> kind of explains why my batteries haven't charged
2012-09-22T21:05:47 < Bird|lappy> here's a fun one: when writing up a BoM, how do you deal with a part whose manufacturer P/N may vary based on manufacturer/variant of chip and you don't know which mfg. or variant you will get when you order it?
2012-09-22T21:07:10 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-22T21:11:31 < zippe> Bird|lappy: Depends a lot on who is processing the BoM
2012-09-22T21:12:06 < Bird|lappy> zippe, this may vary.  I know I'm going to be ordering from at least 3 distributors here
2012-09-22T21:12:58 < Bird|lappy> and I have oh, 3 parts for which this holds
2012-09-22T21:22:22 < Bird|lappy> zippe, I do know that DK and Mouser will be doing most of the processing; however, I also know they will not have the parts in question
2012-09-22T21:25:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 240 seconds]
2012-09-22T21:50:52 < zippe> For both of those guys, you need to be 100% precise
2012-09-22T21:51:09 < zippe> At least for their CSV interface, they don't understand substitutions
2012-09-22T21:51:29 < zippe> You can permute your BoM, or submit the BoM with multiples and then hand-edit the order when you submit it
2012-09-22T21:51:46 < zippe> Or if you're a large enough customer, you can call them up and talk to them about it
2012-09-22T21:51:58 < zippe> What BoM generator are you using?
2012-09-22T21:53:41 < Bird|lappy> zippe, I'm doing this semi-manually here in a spreadsheet
2012-09-22T21:54:06 < Bird|lappy> sort of "stitching together" two temporary BoM files (one from a previous iteration, the other from gnetlist)
2012-09-22T21:54:46 < Bird|lappy> zippe, heh.  well, would it help you to know what parts I'm speaking of here?
2012-09-22T21:55:27 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 256 seconds]
2012-09-22T21:57:46 < zippe> Not really, at least per your original question
2012-09-22T21:58:03 < zippe> second-sourcing parts is pretty common, but I haven't met anyone that handles it well at the hobbyist level
2012-09-22T21:58:20 < zippe> Are you generating a BoM for an eagle-based design?
2012-09-22T21:59:03 < Bird|lappy> no, gEDA/gaf
2012-09-22T21:59:48 < Bird|lappy> and this is a case where I literally will not know what manufacturer/variation I have until after I receive the parts
2012-09-22T22:00:23 < zippe> Ugh
2012-09-22T22:00:28 < zippe> Can't help you much there
2012-09-22T22:23:38 < jpa-> Bird|lappy: what is the trouble with just having multiple columns, one per supplier?
2012-09-22T22:31:12 < Bird|lappy> jpa-, come to think of it, I know a simpler way.
2012-09-22T22:38:08 < karlp> Steffanx: only "large" l1's have op-amps, because they're more like l1.5's, they are only new.
2012-09-22T22:45:32 < upgrdman> i've seen BoM's that listed more than one possibility... for ex
2012-09-22T22:46:13 < upgrdman> " #xxxxxxxxx from y" next line: "or #aaaaaaaaaaaaaaaa from b" next line "or..."
2012-09-22T22:56:46 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 246 seconds]
2012-09-22T22:57:22 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-09-22T22:58:35 <+Steffanx> Hehe, L1.5's karlp :D
2012-09-22T22:58:51 <+Steffanx> I'll go with a low power external one for my project
2012-09-22T23:04:24 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-22T23:04:52 -!- Thorn [~Thorn@78-106-223-25.broadband.corbina.ru] has joined ##stm32
2012-09-22T23:04:52 -!- Thorn [~Thorn@78-106-223-25.broadband.corbina.ru] has quit [Changing host]
2012-09-22T23:04:52 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-22T23:46:37 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Ping timeout: 240 seconds]
2012-09-22T23:51:55 -!- zippe [~Adium@173.11.99.161] has quit [Quit: Leaving.]
2012-09-22T23:54:36 < Tectu> recommend movies
2012-09-22T23:55:28 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-22T23:55:29 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-22T23:55:29 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-22T23:55:32 -!- mode/##stm32 [+v izua] by ChanServ
--- Day changed Sun Sep 23 2012
2012-09-23T00:17:16 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T00:24:08 <+Steffanx> Try you own fantasy or dreamland Tectu
2012-09-23T00:24:59 < Tectu> Steffanx, lol
2012-09-23T00:36:03 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-23T00:37:36 <+dekar_> Tectu, iron sky :)
2012-09-23T00:38:07 < Tectu> dekar_, seen ;-)
2012-09-23T00:42:46 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has joined ##stm32
2012-09-23T00:50:10 -!- Claude is now known as claude
2012-09-23T00:52:07 < zyp> my usb guru friend invited me to dinner, so I brought the discovery board with the audio demo I wrote
2012-09-23T00:52:19 < zyp> now I have to debug some weird crash he managed to provoke :/
2012-09-23T00:58:09 <+izua> zyp: what demo the demo do?
2012-09-23T00:58:15 <+izua> what does*
2012-09-23T00:58:55 < zyp> it uses the f4discovery board as a usb soundcard
2012-09-23T00:59:15 <+izua> both in and out?
2012-09-23T00:59:16 < zyp> didn't you see it when I talked about it earlier today?
2012-09-23T00:59:20 < zyp> no, out only
2012-09-23T00:59:36 < zyp> 18:03:34 < zyp> currently playing music through f4discovery
2012-09-23T00:59:37 <+izua> uh, i recall something in the backlog :P
2012-09-23T00:59:45 <+izua> looked over some code
2012-09-23T01:00:02 < zyp> I even posted the elf if you would like to test it :p
2012-09-23T01:00:13 <+izua> i only have f1 boards :(
2012-09-23T01:01:27 <+izua> well, with the right descriptors i guess it might work
2012-09-23T01:02:25 < zyp> sure, the usb part of it is already supported on both F1 and F4
2012-09-23T01:02:41 < zyp> but you would have to adapt the audio output
2012-09-23T01:03:24 <+izua> you could just dump it on the DAC i guess
2012-09-23T01:03:39 <+izua> or on a pwm + filter
2012-09-23T01:04:10 -!- barthess [~barthess@178.154.15.230] has quit [Ping timeout: 246 seconds]
2012-09-23T01:05:17 < zyp> sure, if you want to write the code for it, you could just go ahead
2012-09-23T01:05:20 < emeb_mac> pwm - blech
2012-09-23T01:05:27 < zyp> :)
2012-09-23T01:05:46 <+izua> you could do some retarded trick like opening another audio in pipe
2012-09-23T01:05:57 < zyp> heh, like my ACM demo
2012-09-23T01:06:00 <+izua> and feeding the output stream back to the input one
2012-09-23T01:06:10 < zyp> http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_cdc_acm <- that's what I do here
2012-09-23T01:06:18 < emeb_mac> loopback!
2012-09-23T01:06:22 <+izua> granted, you could even experiment with dsp, but at that point, it's just easier to, you know, write a vst plugin..
2012-09-23T01:06:28 < zyp> :p
2012-09-23T01:06:41 < emeb_mac> :P :P
2012-09-23T01:06:42 -!- claude [sbnc@h1682708.stratoserver.net] has quit [Ping timeout: 264 seconds]
2012-09-23T01:06:52 < zyp> that ACM demo compiles fine for both F1 and F4
2012-09-23T01:06:58  * izua wishes he would understand the black magic behind awesome dsp guitar effects
2012-09-23T01:07:15 < emeb_mac> izua - which effects are awesome?
2012-09-23T01:07:17 <+izua> is that the stm32 peripheral libs?
2012-09-23T01:07:26 < zyp> no, not at all
2012-09-23T01:07:30  * emeb_mac does some effects
2012-09-23T01:07:31 < zyp> it's using my own stuff
2012-09-23T01:07:38 <+izua> it looks weird, that's why i asked.
2012-09-23T01:07:44 <+izua> i'm mostly playing with cm3
2012-09-23T01:07:57 <+izua> emeb_mac: distorsion/drive and cabsim-ming
2012-09-23T01:08:04 <+izua> i mean, the ones that sound cood.
2012-09-23T01:08:18 < emeb_mac> izua: those aren't too tough to do.
2012-09-23T01:08:28 < emeb_mac> combinations of filters, nonlinearities.
2012-09-23T01:08:37 < zyp> izua, laks is the name of my driver library, laks_demo is a repo of demo-projects using it :p
2012-09-23T01:08:39 <+izua> yes but how do you make them sound good?
2012-09-23T01:09:04 < emeb_mac> izua: the dsp is easy. listening/tuning is hard.
2012-09-23T01:09:06 < zyp> izua, http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=led_matrix <- it also contains the one I wrote for the led matrix
2012-09-23T01:09:11 <+izua> it's something rather abstract and can't be really 'explained' with a term (probably due to lack of a word)
2012-09-23T01:09:28 <+izua> i think "punchiness" and "balls" are the best terms that come to mind, regarding guitar fx
2012-09-23T01:09:36 <+izua> zyp: dude i still didn't get a summons :(
2012-09-23T01:09:42  * emeb_mac likes the look of zyp's laks stuff - very clean.
2012-09-23T01:09:54 < zyp> that's the goal
2012-09-23T01:09:59 < emeb_mac> wish I understood that c++ templates stuff better though.
2012-09-23T01:10:11 < zyp> learn it? :p
2012-09-23T01:10:26 < emeb_mac> yep -
2012-09-23T01:10:38 < emeb_mac> need to spend some time with it.
2012-09-23T01:11:30 -!- claude [sbnc@h1682708.stratoserver.net] has joined ##stm32
2012-09-23T01:11:41 <+izua> zyp: so if i flash the cdcacm demo it will enumerate as an audio I/O device
2012-09-23T01:11:50 < emeb_mac> izua: unfortunately, getting "punchiness" and "balls" is usually secret sauce - the guys doing that stuff hold it close.
2012-09-23T01:12:02 <+izua> and with feedback too?
2012-09-23T01:12:07 < zyp> izua, what? of course not, it will enumerate as an ACM device
2012-09-23T01:12:10  * emeb_mac is good friends w/ AudioDamage - and they don't tell me diddly about it.
2012-09-23T01:12:27 <+izua> well... :P
2012-09-23T01:12:35 <+izua> why was i lead to believe that? :P
2012-09-23T01:12:55 <+izua> emeb_mac: and the funny thing is
2012-09-23T01:13:08 < zyp> the audio demo enumerates as an audio device, the acm demo enumerates as an acm device :p
2012-09-23T01:13:21 < emeb_mac> funny that.
2012-09-23T01:13:23 <+izua> a lot of the good software, even from ancient versions where they sounded mediocre - or halfay decent - have antidebugging and antidisasm stuff
2012-09-23T01:13:34 <+izua> speaking of amplitube/ni guitar rig
2012-09-23T01:14:05 < emeb_mac> yup - they don't want anyone looking at their algos
2012-09-23T01:14:07 <+izua> but.. they need to learn from somewhere...
2012-09-23T01:14:35 <+izua> yeah
2012-09-23T01:14:53 < emeb_mac> yeah - lots of research - reading proceedings from AES, dspfx, IEEE signal processing, CMJ, etc.
2012-09-23T01:15:03 < zyp> emeb_mac, how is your audio stuff coming along?
2012-09-23T01:15:11 < emeb_mac> and then writing lots of code and throwing most of it away.
2012-09-23T01:15:26 < emeb_mac> zyp: the basic famework is fine.
2012-09-23T01:15:46 < emeb_mac> Tried a bunch of stuff with FFT/IFFT - basically works. Not too
2012-09-23T01:15:51 < emeb_mac> interesting though.
2012-09-23T01:16:07 < emeb_mac> will probably fall back and try some time-domain stuff next.
2012-09-23T01:16:26  * emeb_mac needs to get some experience w/ reverbs.
2012-09-23T01:16:38 < emeb_mac> talk about black magic...
2012-09-23T01:17:06 < emeb_mac> good thing is that F405 has plenty of horsepower for doing DSP at audio rates.
2012-09-23T01:17:22 < emeb_mac> much better than the dsPIC stuff I've used in the past.
2012-09-23T01:17:27 < zyp> :)
2012-09-23T01:18:03 < emeb_mac> 48kHz 16-bit stereo data has about 660us frame rate in my setup.
2012-09-23T01:18:13 < zyp> hmm, how would you handle multichannel audio with F4?
2012-09-23T01:18:19 < zyp> multi as in >2
2012-09-23T01:18:26 <+izua> my best dsp attempt was a parametric filter, sort of like a wah
2012-09-23T01:18:37 <+izua> after i messed a long time with filters.
2012-09-23T01:18:42 < emeb_mac> zyp: 2x I2S?
2012-09-23T01:18:48 < emeb_mac> for 4 chls
2012-09-23T01:19:08 < emeb_mac> izua: that's a start.
2012-09-23T01:19:11 <+izua> zyp: multichannel dsp?
2012-09-23T01:19:23 <+dekar_> izua, http://www.youtube.com/watch?feature=player_detailpage&v=eWC22NufDUo#t=509s
2012-09-23T01:19:27 <+izua> or just rendering/recording stuff?
2012-09-23T01:19:29 <+izua> dekar_: ohai
2012-09-23T01:19:33 <+dekar_> hey :)
2012-09-23T01:19:35 < zyp> emeb_mac, how many channels is the F4 capable of that way?
2012-09-23T01:20:05 < emeb_mac> zyp: F4 has 2 I2S ports max, 2chl in, 2chl out each.
2012-09-23T01:20:14 < emeb_mac> so 4 in max, 4 out max.
2012-09-23T01:20:20 < zyp> that's what I figured
2012-09-23T01:20:29 < zyp> what if I want more?
2012-09-23T01:20:52 < emeb_mac> probably have to do custom hardware.
2012-09-23T01:21:01 < emeb_mac> FPGA on the FSMC
2012-09-23T01:21:24 < zyp> I'm kind of tempted to build a device with multichannel balanced output
2012-09-23T01:21:24 <+izua> dekar_: xray machine at home. awesomeness
2012-09-23T01:21:26 <+dekar_> izua, isn't that what we all do when we find a potted device? XD
2012-09-23T01:21:47 < emeb_mac> w/ HD Audio or AC97 or AES/EBU serdes.
2012-09-23T01:22:21 <+izua> dekar_: well, if all of us had xray machines at home, i guess :P
2012-09-23T01:22:50 <+dekar_> I wish I had one, and the place to store it :D
2012-09-23T01:22:58 < HTT-Bird> heheh
2012-09-23T01:23:36 <+dekar_> I was so WTF when he took that device and went over to his frickin' X-ray
2012-09-23T01:24:16 <+izua> dekar_: i had a bookmark, can't remember way
2012-09-23T01:24:28 <+izua> a guy built a basic headphone amp, split power supply, the horrible kind
2012-09-23T01:24:40 <+izua> but encased it in some transparent resin, with some specific angles and leds
2012-09-23T01:24:48 <+izua> and it ended up as a solid block of awesome
2012-09-23T01:25:15 <+dekar_> without PCB?
2012-09-23T01:25:21 <+izua> yeah
2012-09-23T01:25:25 <+izua> dead bug style
2012-09-23T01:25:26 <+dekar_> that's cool :)
2012-09-23T01:25:44 <+izua> http://runawaybrainz.blogspot.co.uk/2012/04/audio-crystal-cmoy-freeform-headphone.html\
2012-09-23T01:25:45 -!- barthess [~barthess@164.177.203.244] has joined ##stm32
2012-09-23T01:25:48 <+izua> minus the \\
2012-09-23T01:25:53 <+dekar_> there is this other guy on youtube who built an electron microscope in his garage
2012-09-23T01:26:02  * izua gives the enter guy a look of disaproval
2012-09-23T01:26:10 <+izua> oh i know that one :D
2012-09-23T01:29:46 <+izua> i wonder how expensive is some of that clear resin
2012-09-23T01:33:17 <+dekar_> so pretty
2012-09-23T01:33:19 < emeb_mac> can't be too expensive - they use it by the ton for various novelties
2012-09-23T01:33:39 < emeb_mac> potted scorpions & tarantulas sold to tourists here in AZ.
2012-09-23T01:36:14 < zyp> hmm, I should have a seperate test system from the dev system here
2012-09-23T01:36:43 < emeb_mac> for usb?
2012-09-23T01:37:00 < zyp> I managed to trigger some error, but somehow it also killed the connection with the BMP
2012-09-23T01:37:08 < emeb_mac> oops
2012-09-23T01:37:48 < emeb_mac> wow - that resin gets hotter than I would have thought - 108C during cure
2012-09-23T01:44:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-23T01:45:07 < zyp> hmm, looks like the problem might be the hub
2012-09-23T01:55:59 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T02:00:18  * emeb tries zyp's demo
2012-09-23T02:00:52 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-23T02:01:06 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 268 seconds]
2012-09-23T02:01:14 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 240 seconds]
2012-09-23T02:01:30 < zyp> :)
2012-09-23T02:01:36 < zyp> let me know the results
2012-09-23T02:02:06 < emeb> I've never hooked anything up to the F4 micro USB before.
2012-09-23T02:03:01 -!- barthess [~barthess@164.177.203.244] has quit [Quit: Leaving.]
2012-09-23T02:03:24 < zyp> are you testing on discovery or your own board?
2012-09-23T02:03:43 < emeb> on an F4 Discovery
2012-09-23T02:04:02 < emeb> looks like it needs power from the programming USB port
2012-09-23T02:04:15 < zyp> yep, it's not designed to be powered from the micro port
2012-09-23T02:05:50 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-23T02:05:59 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-23T02:06:07 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-09-23T02:06:44 < emeb> OK, plug in micro USB to host. LD7 lights, but nothing in the syslog.
2012-09-23T02:07:55 < zyp> LD7 is powered directly from vbus on micro port
2012-09-23T02:08:07 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-23T02:08:16 < zyp> it doesn't enumerate at all?
2012-09-23T02:08:26 < emeb> no
2012-09-23T02:08:37 < zyp> hmm, strange
2012-09-23T02:08:41 < emeb> programmed & verified w/ st-flash utility.
2012-09-23T02:08:50 < zyp> do you have gdb available?
2012-09-23T02:09:03 < emeb> should be able to bring it up - just a sec...
2012-09-23T02:09:23 < zyp> if you can extract the internal usb log, it should tell why it didn't enumerate
2012-09-23T02:10:06 < zyp> http://paste.jvnv.net/view/Do5ao <- save as rblog.py, load in gdb with «source rblog.py»
2012-09-23T02:10:17 < zyp> then extract log with «rblog usb_rblog»
2012-09-23T02:10:24 < emeb> ok - got gdb working. will try
2012-09-23T02:11:45 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-23T02:12:20 < emeb> (gdb) source rblog.py
2012-09-23T02:12:21 < emeb> rblog.py:1: Error in sourced command file:
2012-09-23T02:12:21 < emeb> Undefined command: "class".  Try "help".
2012-09-23T02:12:29 < zyp> oh
2012-09-23T02:13:02 < zyp> sounds like your gdb doesn't have python support
2012-09-23T02:13:10 < emeb> launchpad version
2012-09-23T02:13:13 < zyp> try «python print 'foo'»
2012-09-23T02:13:30 < emeb> (gdb) python print 'foo'
2012-09-23T02:13:30 < emeb> Python scripting is not supported in this copy of GDB.
2012-09-23T02:13:34 < emeb> yup
2012-09-23T02:13:43 < zyp> yeah, so it's compiled without
2012-09-23T02:14:03 < emeb> could try the CS version - just a sec...
2012-09-23T02:14:04 < zyp> well, you could just print the usb_rblog object then
2012-09-23T02:14:37 < zyp> it wouldn't be as pretty, but should still be readable
2012-09-23T02:15:19 < emeb> not in CS either...
2012-09-23T02:16:04 < zyp> «set print pretty on» if you don't already have it on, and then just «p usb_rblog»
2012-09-23T02:19:02 -!- BjoernC [~BjoernC@dyndsl-178-142-056-170.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-23T02:19:28 < emeb> ok - that generates a lot of stuff. any way to redirect to a file?
2012-09-23T02:19:33 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-23T02:20:13 < zyp> uh
2012-09-23T02:20:20 < zyp> google says yes: http://www.slac.stanford.edu/comp/unix/package/rtems/doc/html/gdb/gdb.info.Logging_output.html
2012-09-23T02:22:21 < emeb> http://pastebin.com/KPmrR3DP
2012-09-23T02:22:30 < zyp> thanks
2012-09-23T02:22:39 < zyp> oh wtf
2012-09-23T02:23:09 < zyp> that doesn't make sense at all :p
2012-09-23T02:23:16 < emeb> not to me at least
2012-09-23T02:23:41 < zyp> strings should be either readable or zero-pointers
2012-09-23T02:24:03 < zyp> maybe flashing failed, try flashing the elf directly from gdb
2012-09-23T02:24:32 < zyp> that would also explain why it doesn't enumerate in the first place :p
2012-09-23T02:24:58 < emeb> OK.
2012-09-23T02:25:08 < emeb> just load <file> right?
2012-09-23T02:25:16 < zyp> file <file>
2012-09-23T02:25:17 < zyp> load
2012-09-23T02:25:22 < zyp> is what I usually do
2012-09-23T02:25:22 < emeb> OK
2012-09-23T02:27:01 < emeb> ok, load acted like it did something, verified jolly good, etc.
2012-09-23T02:27:06 < emeb> still no enumeration tho
2012-09-23T02:27:18 < zyp> did you do «run»?
2012-09-23T02:28:00 < emeb> p:)
2012-09-23T02:28:03 < emeb> yes
2012-09-23T02:28:19 < zyp> does the log look more sane now?
2012-09-23T02:28:36 < emeb> no
2012-09-23T02:28:44 < zyp> what does gdb say if you halt and do a backtrace?
2012-09-23T02:29:31 < emeb> (gdb) backtrace
2012-09-23T02:29:31 < emeb> #0  0x5460f44e in ?? ()
2012-09-23T02:29:31 < emeb> #1  0xfffffff8 in ?? ()
2012-09-23T02:29:31 < emeb> #2  0xfffffff8 in ?? ()
2012-09-23T02:29:39 < emeb> out in the weeds
2012-09-23T02:30:13 < zyp> hmm, isn't that the lockup addr, that texane/stlink doesn't know how to recover correctly from, or something like that?
2012-09-23T02:30:40 < zyp> try unplugging everything, then power up the board and plug in the micro connector
2012-09-23T02:30:58 < zyp> I suspect bad flash caused lockup, which caused further attempts to fail
2012-09-23T02:31:17 < zyp> if gdb flashed properly, it should work after powercycle
2012-09-23T02:31:28 < emeb> yay - enumerating
2012-09-23T02:31:41 < emeb> boo stlink. :P
2012-09-23T02:31:51 < zyp> yeah
2012-09-23T02:32:21 < zyp> I actually disconnected the onboard stlink and hooked up my BMP instead to avoid having to deal with it :p
2012-09-23T02:32:56 < zyp> I'm planning to flash BMP to the onboard controller too, but hooking up the BMP was faster :p
2012-09-23T02:33:19 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-232.pptp.stw-bonn.de] has joined ##stm32
2012-09-23T02:33:23 < emeb> just flywire to the jumpers?
2012-09-23T02:33:41 < zyp> to PA13 and PA14 pins down on the right side
2012-09-23T02:33:50 < emeb> of course
2012-09-23T02:34:01 < emeb> see if I can send some audio to the device now...
2012-09-23T02:35:15 < emeb> works!
2012-09-23T02:35:19 < zyp> :)
2012-09-23T02:35:29 < emeb> audacity ftw
2012-09-23T02:35:53 < emeb> doesn't stop cleanly though
2012-09-23T02:36:04 < emeb> seems to repeat the last buffer full when the stream stops.
2012-09-23T02:36:15 < zyp> yep, that's exactly what happens
2012-09-23T02:36:26 < emeb> stop DMA!
2012-09-23T02:36:53 < emeb> so I send some silence.
2012-09-23T02:36:53 < zyp> there is also some distortion every few minutes when the read and write pointers of the ringbuffer interfere because they are not running at the exact same speed
2012-09-23T02:37:05 < emeb> Yeah
2012-09-23T02:37:51 < zyp> but those flaws are expected, as handling them is not implemented yet ;)
2012-09-23T02:38:05 < emeb> Should be a way to sync the clock somehow IIRC
2012-09-23T02:38:17 < emeb> ALSA usb audio drive complains a lot during setup.
2012-09-23T02:38:24 < emeb> missing mixer function, etc.
2012-09-23T02:38:30 < zyp> yeah
2012-09-23T02:38:58 < zyp> it's requesting a bunch of min/max values for the mixers during enumeration
2012-09-23T02:39:04 < zyp> I just return 0 to everything
2012-09-23T02:39:17 < emeb> proof of concept...
2012-09-23T02:39:37 < zyp> it's a basis to continue working on
2012-09-23T02:40:11 < zyp> I prefer to get stuff working with a simple solution first, and then improve upon it while keeping it working
2012-09-23T02:40:16 < emeb> good enough for a rigged demo :)
2012-09-23T02:40:31 < emeb> yeah - that's the way to prototype.
2012-09-23T02:40:36 < zyp> if it breaks I just have to see what I recently changed
2012-09-23T02:40:48 < emeb> git ftw
2012-09-23T02:41:01 < zyp> much more workable than architecting a huge complex solution first and then struggling to get it working
2012-09-23T02:42:44 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 248 seconds]
2012-09-23T02:43:45 < emeb> kill bugs before the burrow too deep into the structure.
2012-09-23T02:47:18 < emeb> It's not clear to me how to resolve the audio sync problem. The I2S PLL isn't really locked to the host's USB rate AFAICT
2012-09-23T02:47:37 < zyp> exactly
2012-09-23T02:48:05 < emeb> It might be necessary to implement a buffer-driven interpolator.
2012-09-23T02:48:12 < emeb> blech
2012-09-23T02:48:39  * emeb did that for a communications system once. Worked OK.
2012-09-23T02:49:25 < zyp> I'm thinking about just continously adjust the PLL speed a bit based on ringbuffer level
2012-09-23T02:50:02 < emeb> yeah. Only problem is that it requires some fairly fine control of the PLL divide ratio
2012-09-23T02:50:11 < zyp> increase I2S speed a step when it's almost full, decrease when it's almost empty
2012-09-23T02:50:13 < emeb> and a realtime process to monitor the buffer level.
2012-09-23T02:51:01 < emeb> Yup - that's how  you do it.
2012-09-23T02:51:03 < zyp> DMA peripheral have some level interrupts
2012-09-23T02:51:21 < zyp> so that part can be monitored by hardware
2012-09-23T02:51:53 < emeb> wonder if the PLL regs are writeable while it's the active source?
2012-09-23T02:52:27 < zyp> hmm, no
2012-09-23T02:52:33 < emeb> afraid of that.
2012-09-23T02:53:18 < emeb> some PLLs react poorly to steps - bad transient response because the loop filters are sensitive.
2012-09-23T02:53:26 < zyp> yeah
2012-09-23T02:57:02 < emeb> trying to see if the ST USB Audio demo code did anything special for that <doubt it>
2012-09-23T03:00:07 < emeb> looks like they intentionally discard samples when the buffer drifts.
2012-09-23T03:00:38 < zyp> that sounds poor
2012-09-23T03:00:45 < emeb> no kidding.
2012-09-23T03:01:00 < zyp> oh
2012-09-23T03:01:00 < zyp> 00:20:05 < emeb_mac> zyp: F4 has 2 I2S ports max, 2chl in, 2chl out each.
2012-09-23T03:01:01 < zyp> 00:20:14 < emeb_mac> so 4 in max, 4 out max.
2012-09-23T03:01:05 < zyp> that's wrong
2012-09-23T03:01:13 < emeb> ?
2012-09-23T03:01:52 < zyp> each I2S peripheral can be used as a dual output instead of one in and one out
2012-09-23T03:02:08 < zyp> so you have 8 channels in total, where all can be used as outputs
2012-09-23T03:02:46 < emeb> I was wondering if the ext ports could go the same dir as the main.
2012-09-23T03:02:51 < zyp> unless I'm misreading something
2012-09-23T03:03:06 < zyp> «Both I2Sx and I2Sx_ext can be configured as transmitters or receivers.»
2012-09-23T03:03:20 < emeb> the ST lib only lets you set them in opposite directions, but hardware may allow them to be the same
2012-09-23T03:03:22 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-23T03:03:29 < zyp> unsure about that
2012-09-23T03:03:49 < zyp> «The extended I2Ss (I2Sx_ext) can be used only in full duplex mode.»
2012-09-23T03:04:11 < zyp> sounds a bit weird, but ok
2012-09-23T03:04:27 < emeb> looks like the I/O drivers are hardwired to be opposite
2012-09-23T03:04:37 < zyp> that might be
2012-09-23T03:04:39 < zyp> damn. :p
2012-09-23T03:04:58 < zyp> here I was hoping I could have a 8-output device :p
2012-09-23T03:05:02 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-23T03:05:43 < emeb> I could design you an FPGA with as many ports as you want. And a rate interpolator too. :)
2012-09-23T03:06:37 < zyp> sure, but so can I :p
2012-09-23T03:06:45 < zyp> hmm, maybe that would be a neat project
2012-09-23T03:07:08 < emeb> been thinking about an F405 board w/ FPGA on it...
2012-09-23T03:07:34 < emeb> probably want a 100-pin pkg to use at least the A/D FSMC
2012-09-23T03:08:24 < emeb> current board is kind of a "dip the toes in the water" lead up to that.
2012-09-23T03:10:09 < zyp> hmm, the cheapest spartan-3a chips start at $6 from digikey
2012-09-23T03:10:23 < emeb> the 50k?
2012-09-23T03:10:28 < zyp> yep
2012-09-23T03:10:32 < emeb> those are pretty tiny.
2012-09-23T03:10:39 < emeb> I use the 200k - about $12
2012-09-23T03:10:50 < zyp> probably, I haven't touched FPGAs for ages
2012-09-23T03:11:35 < zyp> I got one of those spartan-3e boards somewhere, that one has a 500k part if I'm not mistaken
2012-09-23T03:13:34 < emeb> The Xilinx / Digilent S3SK board (has LCD, encoder, buttons & switches)
2012-09-23T03:13:37 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-09-23T03:13:49 < emeb> Those have been around 5-6 yrs or so.
2012-09-23T03:14:20 < zyp> yes, it's over five years since I bought mine
2012-09-23T03:14:40 < emeb> the 3A parts are somewhat easier to use. Not as many supplies required.
2012-09-23T03:14:50 < zyp> ah, nice
2012-09-23T03:15:01 < emeb> (only 2 - 3.3V & 1.2V)
2012-09-23T03:15:28 < zyp> weren't the 3A series directed towards DSP or something like that? I seem to recall a higher amount of hard multipliers
2012-09-23T03:15:33 < emeb> they always overdesign the supplies on those demo boards too.
2012-09-23T03:15:48 < emeb> Yeah - the 3A does have a better ratio of multipliers.
2012-09-23T03:16:04 < emeb> but for really good DSP the S6 is better.
2012-09-23T03:16:16 < zyp> sure, but that's newer :p
2012-09-23T03:16:27 < emeb> they've got the DSP48 blocks with wider multipliers and accumulators/muxes added.
2012-09-23T03:16:40 < emeb> Surprisingly not a lot more $$ though.
2012-09-23T03:16:51 < emeb> But bigger packages :(
2012-09-23T03:16:58 < emeb> (if you don't want BGA)
2012-09-23T03:17:50 < zyp> I'm tempted to do some simple BGA design to try my hand at it
2012-09-23T03:17:50 < emeb> I did a little board w/ a XC3S200A + dsPIC : http://ebrombaugh.studionebula.com/synth/dsPIC_fpga/index.html
2012-09-23T03:18:15 < emeb> FPGA bitstream is loaded from SD card.
2012-09-23T03:18:42 < emeb> Being able to do BGA would open up a lot of options
2012-09-23T03:18:56 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-23T03:18:58 < emeb> and since you already do paste & reflow it's not a big jump for you.
2012-09-23T03:19:12 < zyp> yeah
2012-09-23T03:20:03 < zyp> after all, BGA is usually less dense than QFN, so I'm imagining it could even be easier to do :p
2012-09-23T03:20:22 < emeb> no doubt - the low-end BGA S6 parts on 1mm pitch.
2012-09-23T03:20:26 < emeb> huge
2012-09-23T03:21:30 < zyp> then the next challenge is routing
2012-09-23T03:22:11 < emeb> I think most of the low-end parts are set up so the IO is near the outer edge.
2012-09-23T03:22:34 < zyp> yeah
2012-09-23T03:22:53 < emeb> power & cfg near the inside - not a lot of density
2012-09-23T03:24:05 < zyp> on a four layer board, reserving one layer for ground, one could do outer two rings in top layer and then the next two in the other two layers
2012-09-23T03:25:33 < emeb> Yeah.
2012-09-23T03:27:18 < zyp> I've considered doing a rather simple breakout for lpc4300
2012-09-23T03:27:37 < emeb> oh - the NXP ones w/o flash
2012-09-23T03:27:41 < zyp> then I both get experience with bga soldering and the lpc4300
2012-09-23T03:27:59 < zyp> they are supposed to have flash
2012-09-23T03:28:13 < zyp> are still just the ones without flash shipping?
2012-09-23T03:28:15 < zyp> hmm
2012-09-23T03:28:20 < emeb> AFAIK
2012-09-23T03:28:23 < zyp> digikey just lists the ones without
2012-09-23T03:28:28 < zyp> so yeah
2012-09-23T03:28:49 < emeb> someone here was saying they can execute out of SPI flash. :P
2012-09-23T03:28:52 < zyp> but that's sort of fine for a board like that
2012-09-23T03:28:59 < zyp> yeah, I remember that
2012-09-23T03:29:02 < zyp> that's not dumb
2012-09-23T03:29:22 < zyp> just put a bootloader there, then load software to ram
2012-09-23T03:29:26 < emeb> yup
2012-09-23T03:29:36 < emeb> not dumb, but did surprise me.
2012-09-23T03:29:54 < emeb> was thinking more like a ROM bootloader to copy flash->SRAM
2012-09-23T03:30:07 < emeb> but direct execute is more flexible.
2012-09-23T03:30:51 < zyp> well, you can have a bootloader in SPI flash that loads the firmware itself from an SD card for example
2012-09-23T03:31:03 < emeb> even better
2012-09-23T03:31:17 < emeb> and huge SPI flash is cheap.
2012-09-23T03:31:35 < emeb> small SPI flash is mousenuts
2012-09-23T03:31:47 < zyp> of course, there is nothing stopping you from doing the same with stm32
2012-09-23T03:32:08 < emeb> doing same what?
2012-09-23T03:32:24 < zyp> loading to ram and executing from there
2012-09-23T03:32:27 < emeb> copy SD -> SRAM w/ code in STM32 flash.
2012-09-23T03:32:52 < emeb> Internal SRAM on the LPC4300 is how big?
2012-09-23T03:33:06 < zyp> 128k + 64k CCM
2012-09-23T03:33:21 < emeb> Ah - same as F4. IIRC they have an SDRAM controller.
2012-09-23T03:33:22 < zyp> sorry
2012-09-23T03:33:28 < zyp> I misread
2012-09-23T03:33:32 < zyp> I was thinking F4
2012-09-23T03:33:48 < zyp> digikey says in the range of 168k-264k
2012-09-23T03:33:54 < zyp> but I don't know what the layout is
2012-09-23T03:34:26 < zyp> SDRAM is nice but I imagine high latency
2012-09-23T03:34:41 < zyp> external SRAM is probably better but still not as fast as internal
2012-09-23T03:37:18 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-23T03:37:23 < emeb> which part/pkg were you thinking for a breakout board?
2012-09-23T03:37:25 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 260 seconds]
2012-09-23T03:37:39 < emeb> looks like they've got 256, 180, 100 BGA pkgs.
2012-09-23T03:38:45 < zyp> IIRC some are 1mm pitch and some are 0.8mm
2012-09-23T03:39:09 < emeb> yep - the 256 is 1.0, others 0.8
2012-09-23T03:39:39 < emeb> looks like the 256 are cheapest too.
2012-09-23T03:39:42 < zyp> then either 256 or 100, depending on how hard 0.8 is vs added cost and area of 256
2012-09-23T03:40:50 < zyp> unless 256 has some crazy amount of required pins on the inner rings :p
2012-09-23T03:41:42 < qyx_> hm, M4 and M0 in one package?
2012-09-23T03:41:47 < zyp> yes
2012-09-23T03:41:53 < emeb> crazy, no?
2012-09-23T03:42:05 < zyp> it's a bit interesting
2012-09-23T03:42:24 < emeb> reminds me of the OMAP SoCs - they've got ARM cores scattered all over.
2012-09-23T03:42:37 < zyp> I know
2012-09-23T03:42:43 < qyx_> it sounds like 1-and-half core
2012-09-23T03:42:52 < qyx_> why M0?
2012-09-23T03:42:53 < emeb> eh - M0 is pretty usable
2012-09-23T03:42:56 < zyp> OMAP4 is dual A9 and two additional M3s
2012-09-23T03:43:03 < zyp> qyx_, because it's small
2012-09-23T03:43:19 < zyp> small in area, i.e. cheap to produce
2012-09-23T03:44:50 < zyp> it would be a pretty neat development system if you could run a debugger interface on the M0 that could control the M4
2012-09-23T03:44:51 < emeb> zyp: the OMAP3 on the Beagleboard has an A8 + c64x + a couple hidden M3 cores they don't talk about.
2012-09-23T03:44:59 < qyx_> available at farnell and cheaper than stm32
2012-09-23T03:45:07 < zyp> emeb, interesting
2012-09-23T03:45:07 < qyx_> it seems to be worth of trying
2012-09-23T03:45:33 < emeb> one of the M3 cores is hanging off the c64x DSP for use primarily in color space conversion.
2012-09-23T03:45:51 < zyp> emeb, but the non-OMAP SoC they used on beaglebone have no M-cores, right?
2012-09-23T03:46:12 < emeb> zyp: right - it's got these odd PMU things.
2012-09-23T03:52:21 < emeb> got one of these the other day to try out on my F4: http://adafruit.com/products/358
2012-09-23T03:53:34 < zyp> :)
2012-09-23T03:54:01 < emeb> Tried it out on an Arduino - works, but is kinda slow loading even w/ hardware SPI.
2012-09-23T03:54:12 < emeb> Will see how fast it goes on a real MCU. :)
2012-09-23T03:59:35 -!- DaKu is now known as daku
2012-09-23T04:07:09 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-23T04:07:15 < cjbaird> WWHHYYY
2012-09-23T04:07:23 < emeb> WWHHYY NNOOTT??
2012-09-23T04:12:42 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 264 seconds]
2012-09-23T04:13:13 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-09-23T04:16:45 < emeb> laters
2012-09-23T04:16:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-23T04:19:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T04:19:53 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-09-23T05:22:33 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-232.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-23T05:29:30 -!- Oldboy [myircshell@grid.sh3lls.net] has joined ##stm32
2012-09-23T06:04:54 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Ping timeout: 264 seconds]
2012-09-23T06:11:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T06:29:23 < emeb_mac> for some reason the google chrome browser thinks the laks_demo git is in indonesian and offers to translate
2012-09-23T06:30:07 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-23T06:30:12 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-23T07:19:48 < HTT-Bird> emeb_mac: it's based on letter frequencies I reckon :p
2012-09-23T07:22:26 < emeb_mac> hilarious
2012-09-23T07:22:50 < emeb_mac> I should have let it translate and see what it came up with
2012-09-23T07:25:59 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-23T07:40:40 < emeb_mac> trying to get a recent version arm-none-eabi-gcc running on my MacBook. Looks like I'll need to compile from source.
2012-09-23T07:41:35 < emeb_mac> seems all binaries for recent versions are for Win/Linux, while all MacOS versions are ancient or not for this eabi
2012-09-23T07:52:13 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-09-23T07:53:41 < HTT-Bird> emeb_mac: yeah, I used summon-arm-toolchain on my Debian testing box
2012-09-23T08:07:03 < emeb_mac> Ah - Yagarto has a MacOS version of gcc 4.7.1
2012-09-23T08:12:33 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-23T08:36:36 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-09-23T08:56:41 < emeb_mac> this is the best we can do?
2012-09-23T09:01:17 < HTT-Bird> thing is, .0001% of people would ever vote for either of them if they weren't
2012-09-23T09:01:21 < HTT-Bird> *sigh at poly-ticks*
2012-09-23T09:04:17 < R2COM> vote for arnold schwarzenegger
2012-09-23T09:04:27 < R2COM> at least it would be fun
2012-09-23T09:08:54 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-23T09:11:13 < cjbaird> Can't. He wasn't born there. (There was an attempt to admend the constitution specifically to get Arnie in..)
2012-09-23T09:31:00 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-23T09:43:34 < dongs> http://www.youtube.com/watch?v=LO1s06Yb24k yesterday fireworks trollin
2012-09-23T09:49:10 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-23T09:52:29 -!- Netsplit *.net <-> *.split quits: amitofu, ABLomas, feurig
2012-09-23T09:56:14 < cjbaird> Got to wonder who much Google is losing to storage costs on tards uploading HD1080 videos of worthless crap..
2012-09-23T09:58:15 < Oldboy> rename your backups to .avi and upload them to youtube
2012-09-23T09:58:19 < Oldboy> free unlimited backups
2012-09-23T10:00:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-23T10:02:40 < dongs> cjbaird: i kno rite
2012-09-23T10:02:54 < dongs> cjbaird: 4.3gig of 1080p/24 AVCHD @ 24mbit
2012-09-23T10:03:04 < cjbaird> ..which you had to upload first..
2012-09-23T10:03:10 < dongs> well im not in australia
2012-09-23T10:03:20 < dongs> so its not a big deal
2012-09-23T10:03:21 < dongs> ^_^
2012-09-23T10:06:50 < cjbaird> Oooh, you've just made an enemy for life.
2012-09-23T10:10:12 -!- Netsplit over, joins: amitofu, feurig, ABLomas
2012-09-23T10:10:16 < dongs> heh
2012-09-23T10:37:55 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-23T10:44:48 -!- claude is now known as Claude
2012-09-23T10:45:05 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-23T11:01:51 -!- Claude is now known as claude
2012-09-23T11:07:26 < dongs> cjbaird: haha.
2012-09-23T11:54:26 -!- BjoernC [~BjoernC@dyndsl-178-142-057-122.ewe-ip-backbone.de] has joined ##stm32
2012-09-23T12:29:54 -!- barthess [~barthess@178.154.29.244] has joined ##stm32
2012-09-23T12:40:50 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-09-23T13:05:14 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-23T13:08:24 < TitanMKD> Anyone tested the TI BQ24230 LiPo/charging ? For me all work fine but the Enable/Disable(CE) charging just do nothing and it is always in charging mode
2012-09-23T13:18:12 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-23T13:22:37 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 256 seconds]
2012-09-23T13:33:05 < dongs> did you check datasheet to see waht it expects there
2012-09-23T13:33:20 < TitanMKD> yes
2012-09-23T13:34:03 < TitanMKD> Connect CE to a high logic level to place the battery charger in standby mode
2012-09-23T13:34:36 < TitanMKD> even i switch CE to 4.2V it does not enter standby mode and stop charging
2012-09-23T13:34:56 < dongs> you probably burned that pin.
2012-09-23T13:35:04 < TitanMKD> hmm
2012-09-23T13:35:10 < dongs> "high logic level" is probably > 2V or someshit.
2012-09-23T13:35:23 < dongs> did you see w hat max ratings for it was?
2012-09-23T13:35:32 < TitanMKD> it is connected to an internal pull down 285kOhm
2012-09-23T13:35:48 < TitanMKD> high logic is up to 6V IIRC
2012-09-23T13:35:51 < TitanMKD> so 4.2V is ok
2012-09-23T13:36:55 < zyp> how do you know it's still in charging mode?
2012-09-23T13:37:17 < TitanMKD> i check it with multimeter ;)
2012-09-23T13:37:26 < TitanMKD> and i see the battery is charging at 200mA
2012-09-23T13:37:40 < TitanMKD> same as when the CE is set to Low GND
2012-09-23T13:38:08 < TitanMKD> and also the CHG is ON and my LED is ON when charging
2012-09-23T13:39:12 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 245 seconds]
2012-09-23T13:41:01 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-23T13:54:25 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-23T14:00:07 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-23T14:00:07 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-23T14:00:07 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-23T14:00:10 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-23T14:12:34 < dongs> blog.
2012-09-23T14:18:44 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 252 seconds]
2012-09-23T14:41:12 < cjbaird> blog: trying to overlcock the raspberry pi to 1ghz and failing. :/
2012-09-23T14:42:32 < dongs> what does it normally rtun at
2012-09-23T14:43:24 < cjbaird> 700Mhz, but the binary-blog firmware now lets it go up to anything provided the core temp doesn't top 85C
2012-09-23T14:44:58 < cjbaird> update the booting firmware: check; update config.txt: check; compile a whole kernel on the cluster in less than an hour: nope
2012-09-23T15:05:48 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-23T15:20:52 -!- daku is now known as DaKu
2012-09-23T15:31:16 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-23T15:31:16 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-23T15:42:50 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-23T15:44:17 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-23T15:44:25 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has joined ##stm32
2012-09-23T15:55:45 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-255-140.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-23T15:55:46 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has joined ##stm32
2012-09-23T16:01:12 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 256 seconds]
2012-09-23T16:24:23 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-23T16:50:15 -!- barthess [~barthess@178.154.29.244] has quit [Read error: Connection reset by peer]
2012-09-23T17:10:48 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T17:19:37 -!- barthess [~barthess@178.154.102.20] has joined ##stm32
2012-09-23T17:26:44 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-23T17:26:45 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-23T17:26:45 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-23T17:34:59 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-23T17:36:29 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-23T17:36:35 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-23T17:36:53 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has joined ##stm32
2012-09-23T17:52:41 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-23T17:57:15 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-23T18:13:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-23T18:25:03 -!- Erlkoenig1 [~erlkoenig@pptp-194-94-198-57.pptp.stw-bonn.de] has joined ##stm32
2012-09-23T18:25:03 -!- Erlkoenig1 [~erlkoenig@pptp-194-94-198-57.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-23T18:28:17 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 260 seconds]
2012-09-23T18:40:46 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-57.pptp.stw-bonn.de] has joined ##stm32
2012-09-23T19:03:06 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-23T19:17:40 < emeb_mac> *yawn*
2012-09-23T19:19:36 < zyp> good afternoon
2012-09-23T19:26:12 < emeb_mac> hi
2012-09-23T19:26:45 < emeb_mac> 9:26am here...
2012-09-23T19:27:37 < emeb_mac> looking over my resampling code - see how much effort to rewrite it in floating point for F4.
2012-09-23T19:29:00 < emeb_mac> originally used here: http://www.analoguehaven.com/synthesistechnology/e580/
2012-09-23T19:29:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-23T19:31:21 < emeb_mac> zyp: managed to get laks_demo built on MacOS last night - needed to grab gcc 4.7.1 from yagarto & install scons, but otherwise went well.
2012-09-23T19:31:47 < zyp> I'm working on OS X myself :p
2012-09-23T19:31:55 < emeb_mac> heh
2012-09-23T19:32:34 < emeb_mac> I don't normally develop on MacOS, but it's the machine I surf with while watching TV in the evening.
2012-09-23T19:35:16 < zyp> :)
2012-09-23T19:39:18 < zyp> hmm, that reminds me of the time I bought my macbook
2012-09-23T19:39:28 < gnomad> what is "scons" ?
2012-09-23T19:39:37 < zyp> a make replacement
2012-09-23T19:40:52 < emeb_mac> seems pretty smart
2012-09-23T19:42:14 < emeb_mac> switching to big iron.
2012-09-23T19:42:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-23T19:42:47 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-23T19:42:51 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-23T19:42:59 < zyp> dependency tracking is better, multiple directories are handled in a more sane way and it's also more flexible
2012-09-23T19:45:40 <+Steffanx> nu
2012-09-23T19:45:51 < zyp> Steffanx, good morning
2012-09-23T19:46:11 <+Steffanx> :)
2012-09-23T19:47:37 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-23T20:03:27 < jpa-> there should be a standard USB device class for setting RTC time
2012-09-23T20:03:58 < zyp> :p
2012-09-23T20:04:14 < jpa-> it would benefit digital cameras, photo frames and all kinds of small widgets
2012-09-23T20:04:15 < emeb> ??
2012-09-23T20:04:39 < zyp> sounds like a nice idea
2012-09-23T20:04:40 < jpa-> every time you plugged in the operating system would just automatically send the current UTC time to the device
2012-09-23T20:04:51 < emeb> easier to just put GPS in everything. :D
2012-09-23T20:05:02 < karlp> not really.
2012-09-23T20:05:11 < karlp> not nearly as fast.
2012-09-23T20:05:15 < jpa-> yeah, gps is much easier than <100 lines of code :)
2012-09-23T20:05:21  * emeb left off his irony tags
2012-09-23T20:05:29 <+Steffanx> but it doesn't need USB :D
2012-09-23T20:05:31 < karlp> sorry about that :)
2012-09-23T20:05:36 < karlp> gps is just used for time so often,
2012-09-23T20:06:01 < jpa-> emeb: it's not unreasonable enough for sarcastic remarks :)
2012-09-23T20:06:07 < emeb> lol
2012-09-23T20:06:17 < emeb> too close to truth
2012-09-23T20:06:44 < jpa-> there are so small and cheap gps chips nowadays that it's almost a viable alternative... but i guess with the smallest antennas they cannot get even a time sync indoors
2012-09-23T20:07:06 <+Steffanx> USB won't help either, without a pc
2012-09-23T20:07:19 < emeb> my andorrhoid can't get GPS if there's a metal roof on the building...
2012-09-23T20:07:46 < jpa-> my hemarrhoid cannot get GPS lock either
2012-09-23T20:07:46 < emeb> as seems to be the case in most commercial establishments.
2012-09-23T20:07:54 < Laurenceb_> <jpa-> there should be a standard USB device class for setting RTC time
2012-09-23T20:07:56 < gnomad> I wonder if you can do NTP over USB...
2012-09-23T20:08:00 < Laurenceb_> ^yeah i had that issue
2012-09-23T20:08:17 < emeb> gnomad: sure - CDC Ethernet.
2012-09-23T20:08:33 < gnomad> well there you go!
2012-09-23T20:08:33 < emeb> only works if the host will route you tho.
2012-09-23T20:08:36 < Laurenceb_> that was my conclusion
2012-09-23T20:08:38 < gnomad> of course, that is a lot of oerhead
2012-09-23T20:08:42 < gnomad> overhead
2012-09-23T20:08:42 < Laurenceb_> seems to only way
2012-09-23T20:08:57 < Laurenceb_> in the end i used the timestamp from mass storage
2012-09-23T20:09:01 < jpa-> Laurenceb_: if you have USB mass storage, you can try .. yeah
2012-09-23T20:09:10 <+Steffanx> So, can't you send the USB people a draft of CDC RTC jpa- ?
2012-09-23T20:09:38 < emeb> and don't hold your breath.
2012-09-23T20:09:40 < jpa-> Steffanx: i'm not so sure i'm expert enough to write one :)
2012-09-23T20:09:56 < jpa-> (heck, i haven't implemented any USB driver myself)
2012-09-23T20:10:15 < emeb> tech experience counts against you when writing standards.
2012-09-23T20:10:17 <+Steffanx> It's not more than "If connected send UTC Time in some special format"
2012-09-23T20:10:22 < Laurenceb_> lol emeb
2012-09-23T20:10:36 < Laurenceb_> just sent 32bit timestamp or something
2012-09-23T20:10:37  * emeb has sat in on IEEE stds meetings.
2012-09-23T20:10:53 <+Steffanx> 64 bit Laurenceb_
2012-09-23T20:11:06 < Laurenceb_> or that
2012-09-23T20:11:21 < zyp> setting time via usb should just need a descriptor to advertise presence of the ability and a control request to set the time
2012-09-23T20:11:24 < emeb> what's the resolution in seconds if you try to express the age of the Universe in 64 bits?
2012-09-23T20:11:53 <+Steffanx> And some crap for the jitter perhaps?
2012-09-23T20:11:57 < zippe> emeb: that's not enough info; you need to express the end as well
2012-09-23T20:12:18 < emeb> zippe: assume now = 0.5
2012-09-23T20:12:54 < Laurenceb_> 24milliseconds
2012-09-23T20:14:38 < emeb> why'd I get 47?
2012-09-23T20:14:48 < emeb> Ah Laurenceb_ set now = 1.0
2012-09-23T20:16:04 < emeb> 2 * 4.339e17 / 2^64 = 0.047
2012-09-23T20:16:35 < zyp> http://www.wolframalpha.com/input/?i=age+of+the+universe+%2F+2**64+%2F+0.5
2012-09-23T20:17:19 < emeb> we live in wondrous times.
2012-09-23T20:18:40 < emeb> so if we set the universal systick to 100ms then we're roughly 1/4 of the way to the end of the universe.
2012-09-23T20:19:02 < Laurenceb_> lol
2012-09-23T20:19:04 < emeb> in the 64-bit calendar.
2012-09-23T20:22:54 < emeb> now express that in Plank time, or chronons - you pick. :P
2012-09-23T20:23:11 < emeb> s/Plank/Planck/
2012-09-23T20:23:17 <+Steffanx> I'll express in in the time you wasted on this conversation :P
2012-09-23T20:23:21 <+Steffanx> *-n
2012-09-23T20:23:23 <+Steffanx> *i
2012-09-23T20:23:28 < emeb> party pooper.
2012-09-23T20:23:49 <+Steffanx> Ha
2012-09-23T20:27:50 <+Steffanx> Don't blame me! Now you have time to get another party started emeb
2012-09-23T20:28:06 < emeb> heh
2012-09-23T20:40:05 < Laurenceb_> http://ukhas.org.uk/general:ukhasbadgeboard
2012-09-23T20:40:09 < Laurenceb_> lol issues already
2012-09-23T20:42:02 <+Steffanx> ....
2012-09-23T20:42:38 < Laurenceb_> ive got one to play with now
2012-09-23T20:43:35 <+Steffanx> went to the conference?
2012-09-23T20:44:18 < Laurenceb_> yes
2012-09-23T20:48:59 <+Steffanx> And interesting talks?
2012-09-23T20:49:50 < Laurenceb_> a few yes
2012-09-23T21:27:29 < upgrdman> any here use the f0? do you like it?
2012-09-23T21:28:54 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-57.pptp.stw-bonn.de] has quit [Ping timeout: 264 seconds]
2012-09-23T21:32:05 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-119.pptp.stw-bonn.de] has joined ##stm32
2012-09-23T21:36:06 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-119.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-23T21:54:17 -!- BjoernC_ [~BjoernC@dyndsl-178-142-057-122.ewe-ip-backbone.de] has joined ##stm32
2012-09-23T21:54:31 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-23T21:55:18 -!- BjoernC [~BjoernC@dyndsl-178-142-057-122.ewe-ip-backbone.de] has quit [Ping timeout: 264 seconds]
2012-09-23T22:00:56 < emeb> Not yet.
2012-09-23T22:01:19 < emeb> Seems OK, but I haven't quite figured out what it's got going for it that the F100/VL parts do.
2012-09-23T22:01:34 < emeb> (other than faster clock speed and newer peripherals)
2012-09-23T22:13:18 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 264 seconds]
2012-09-23T22:35:40 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 268 seconds]
2012-09-23T23:05:10 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-23T23:07:27 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Quit: No Ping reply in 180 seconds.]
2012-09-23T23:07:51 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-23T23:08:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 268 seconds]
2012-09-23T23:08:49 < upgrdman> anyone here use linux? im getting a linker error
2012-09-23T23:09:57 < upgrdman> compiling with "... -L../../stm32f0-discovery-basic-template/Libraries -llibstm32f0" but get an error from ld "cannot find -llibstm32f0". if i "cd ../../stm32f0-discovery-basic-template/Libraries" the libstm32f0.a file is right there.
2012-09-23T23:13:10 < upgrdman> nevermind. apparently i should have left lib out of the filename because ld automatically add it in
2012-09-23T23:17:02 < emeb> :)
2012-09-23T23:34:54 < Tectu> a random STM32F103VET does have an F4 with the same pinconfig which i can simply replace, right?
2012-09-23T23:34:56 < Tectu> or is that wrong?
2012-09-23T23:42:39 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-23T23:52:46 < zippe> Tectu: there are F2/F4 parts that mostly match the F1 devices
2012-09-23T23:53:06 < upgrdman> anyone here used the stm32f0disco board in linux? seems like stlink doesn't support it... but i don't know why since its the same stlinkv2 hardware as the f4 board
2012-09-23T23:53:08 < zippe> Tectu: There is one minor but critical difference that you have to take care of
2012-09-23T23:53:47 < zippe> Tectu: it's covered in the F2/4 datasheet though, so as long as you read the section on pin compatibility you should be fine
2012-09-23T23:56:30 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-23T23:58:12 < emeb> upgrdman: probably because stlink hasn't been patched to recognise the F0 parts.
2012-09-23T23:58:25 < upgrdman> ok
2012-09-23T23:58:27 < emeb> when I use the F0 discovery to talk to older processors it works fine.
2012-09-23T23:58:37 < upgrdman> does that mean openocd would be the only option?
2012-09-23T23:58:48 < emeb> got me.
2012-09-23T23:58:53 < emeb> or try BMP
2012-09-23T23:59:04 < upgrdman> bmp?
2012-09-23T23:59:13 < zyp> didn't somebody add F0 support?
2012-09-23T23:59:15 < emeb> http://www.blacksphere.co.nz/main/blackmagic
--- Day changed Mon Sep 24 2012
2012-09-24T00:00:01 < zyp> emeb, the point with F0 is price, it's supposed to be the cheapest one
2012-09-24T00:00:22 < emeb> zyp: that's what I would have thought, but the VL/F100 parts are cheaper.
2012-09-24T00:00:30 < emeb> (last time I looked)
2012-09-24T00:01:39 < zyp> hmm, actually digikey lists F050K4 as cheaper than F100C4
2012-09-24T00:04:43 < emeb> well then.
2012-09-24T00:06:42 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-24T00:07:46 < emeb> although the K4 is a QFN, while the C4 is a LQFP48
2012-09-24T00:08:29 < emeb> if you get the 050C4 then the price is higher than the F100C4
2012-09-24T00:10:21 < emeb> Anyone else notice that using floating pt constants often causes hard faults?
2012-09-24T00:11:04 < zyp> often how?
2012-09-24T00:11:47 < zyp> sporadic for a given executable, or always with a given executable?
2012-09-24T00:12:17 < emeb> always at a given line in the code
2012-09-24T00:12:50 < emeb> so if I have "a = 1.0/b;" it always segfaults there.
2012-09-24T00:12:51 < zyp> in other words, you are generating invalid instructions?
2012-09-24T00:12:56 < emeb> must be
2012-09-24T00:13:03 < emeb> question is why
2012-09-24T00:13:17 < emeb> esp - if I do "a = 1.0F/b;" then it works.
2012-09-24T00:13:33 < zyp> oh
2012-09-24T00:13:46 < emeb> where the "F" may or may not be required from one line to another.
2012-09-24T00:13:54 < zyp> you say constants, is b constant?
2012-09-24T00:13:58 < emeb> no
2012-09-24T00:14:06 < zyp> then there are no constants.
2012-09-24T00:14:18 < emeb> 1.0 isn't a constant?
2012-09-24T00:14:41 < zyp> it's a literal
2012-09-24T00:14:46 < emeb> err ok
2012-09-24T00:14:56 < emeb> improper terminology
2012-09-24T00:14:59 < zyp> 1.0/b generates a double-division
2012-09-24T00:15:12 < emeb> ok
2012-09-24T00:15:18 < zyp> I mean, it's not constant as in «expression evaluated in compile time»
2012-09-24T00:15:46 < zyp> the fpu in M4F is only single precision, so a double-division would have to be software emulated
2012-09-24T00:16:27 < emeb> ah ok - need to force the literals to be single w/ F
2012-09-24T00:16:38 < emeb> since all C defaults to double
2012-09-24T00:16:41 < zyp> so, maybe your soft-fpu-lib is not thumb?
2012-09-24T00:17:07 < emeb> trying to avoid soft
2012-09-24T00:17:07 < zyp> i.e. you get the dreaded blx and hardfault with INVSTATE
2012-09-24T00:21:55 < emeb> sorted - ty
2012-09-24T00:25:02 < emeb> the perils of embedded floating point. Things you rarely deal with when programming under Linux.
2012-09-24T00:26:39 < zyp> double literals are still not floats under linux :p
2012-09-24T00:27:27 < emeb> no, but if you use them wrong the pgm rarely crashes.
2012-09-24T00:28:00 < zyp> still shouldn't in your case
2012-09-24T00:28:21 < emeb> but I'm not linking in a soft fpu lib
2012-09-24T00:29:22 < zyp> you're lying :p
2012-09-24T00:29:31 < zyp> how are you then getting a double division? :p
2012-09-24T00:32:06 < emeb> obviously I don't understand everything that's happening here.
2012-09-24T00:33:08 < emeb> when I do a=1.0/b the literal defaults to a double and I get a segfault because... ?
2012-09-24T00:33:36 < emeb> but when I do a=1.0F/b then it works because everything is single and hard fpu works.
2012-09-24T00:34:08 < emeb> And BTW - my resampler is working! w00t.
2012-09-24T00:38:20 < Thorn> which controller do I use for a FS device in f4?
2012-09-24T00:38:28 < prog> can you help me porting my SDR to stm32f4 ?:D
2012-09-24T00:39:26 < emeb> what kind of SDR
2012-09-24T00:39:39 < prog> sdrsharp.com
2012-09-24T00:39:46 < Thorn> I take it the HS controller has more memory and no honorable mentions in the errata
2012-09-24T00:40:12 < emeb> ah - I've played with sdrsharp
2012-09-24T00:40:20 < prog> cool
2012-09-24T00:40:21 < zyp> Thorn, either
2012-09-24T00:40:22 < emeb> nice UI.
2012-09-24T00:40:34 < prog> I'm planning to port it someday to STM32
2012-09-24T00:40:39 < emeb> Doubt that the F4 has enough balls to run it though.
2012-09-24T00:40:42 < zyp> Thorn, both are working for me
2012-09-24T00:40:49 < BrainDamage> prog: isn't there a .net VM for micros?
2012-09-24T00:41:12 < prog> I reworked algos to be less cpu hungry
2012-09-24T00:41:16 < Thorn> zyp: did you test the HS one in FS mode with no external PHY?
2012-09-24T00:41:22 < prog> BrainDamage: it's very slow
2012-09-24T00:41:33 < prog> no JIT, only direct interpretation
2012-09-24T00:41:35 < zyp> Thorn, yes, that's exactly how I've been using it
2012-09-24T00:41:40 < BrainDamage> ouch
2012-09-24T00:41:42 < Thorn> ok nice thanks.
2012-09-24T00:41:47 < zyp> I haven't added support for external PHY yet
2012-09-24T00:42:02 < zyp> I might do it some time during next week
2012-09-24T00:42:23 < emeb> prog: you want to use the F4 as USB HS host for an RTLSDR?
2012-09-24T00:42:51 < prog> I want to do all the calculations in the F4
2012-09-24T00:42:52 < zippe> The netmf VM is not very fast
2012-09-24T00:43:04 < prog> and may be use the rtlsdr with it
2012-09-24T00:43:08 < zippe> Since it can't JIT - it's just a bytecode VM
2012-09-24T00:44:13 < emeb> prog: I really don't think the F4 can sustain the fpu throughput needed for the ~1MSPS data that comes from the RTL
2012-09-24T00:45:58 < prog> if it can stand 250msps, i can get it process 2MSPS
2012-09-24T00:46:12 < prog> 250ksps*
2012-09-24T00:46:29 < prog> we only need 250ksps at most at any time
2012-09-24T00:46:56 < prog> the current decimation scheme is very efficient
2012-09-24T00:49:32 < emeb> prog: I've got an integer resampler running right now. 48ksps audio data running through it. ~120 FIR taps, takes about 50% of CPU
2012-09-24T00:50:44 < prog> CIC is your friend
2012-09-24T00:50:57 < emeb> Not for audio it isn't.
2012-09-24T00:51:06 < prog> true
2012-09-24T00:51:15 < prog> for audio you need something flat
2012-09-24T00:51:27 < emeb> this is a JOS-type interpolated sinc resampler.
2012-09-24T00:51:42 < upgrdman> anyone here have a f0disco board? mine seems slow but im not sure if its normal or something is configured wrong.
2012-09-24T00:52:00 < emeb> ie - every FIR tap is interpolated on the fly from a table of sinc & deta-sinc values based on the current phase.
2012-09-24T00:53:06 < prog> if your samplerates are fixed you may consider Parks-McClellan optimized kernels
2012-09-24T00:53:28 < emeb> fixed resampling rates are easy. This is realtime variable. :)
2012-09-24T00:53:51 -!- BjoernC_ [~BjoernC@dyndsl-178-142-057-122.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-24T00:54:03 < prog> http://cpansearch.perl.org/src/MLEHMANN/PDL-Audio-1.2/remez.c ;)
2012-09-24T00:54:34 < emeb> yep - well aware of PM optimization  & Remez exchange.
2012-09-24T00:54:38 < prog> this is a bit extreme, but if you want to have fun
2012-09-24T00:55:06 < emeb> I'm just giving this as a benchmark for your consideration.
2012-09-24T00:55:47 < emeb> another point - earlier I had 128-pt real FFT & IFFT running - takes about 200us.
2012-09-24T00:56:19 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-24T00:56:24 < emeb> you can get a lot of math done on the F4. great for audio. Not so sure about IF rates though...
2012-09-24T00:56:52 < emeb> esp when you add on the USB HS overhead to get data from the RTL.
2012-09-24T00:57:59 < zyp> not that much overhead? most of it could be avoided with DMA
2012-09-24T00:58:00 < prog> I know some folks who have succeeded doing a complete SDR with the F4 at 800ksps input
2012-09-24T00:58:34 < prog> and still at only 40% of the core
2012-09-24T00:59:28 < upgrdman> sdr?
2012-09-24T00:59:32 < Laurenceb_> depends how you define sdr
2012-09-24T01:00:08 < emeb> prog: nice - what were they using for I/O, what modulations were they using?
2012-09-24T01:00:38 < Laurenceb_> if you wanted to do gps for example
2012-09-24T01:00:41 < Laurenceb_> a lot harder
2012-09-24T01:01:44 < prog> it's for HF
2012-09-24T01:01:56 < Laurenceb_> cool
2012-09-24T01:02:09 < Laurenceb_> whats the bandwidth of the DMA?
2012-09-24T01:02:11 < Laurenceb_> erm
2012-09-24T01:02:11 < emeb> prog: OK - so just OOK for code?
2012-09-24T01:02:15 < Laurenceb_> s/DMA/DAC
2012-09-24T01:02:26 < emeb> simple SSB demod.
2012-09-24T01:02:44 < prog> SSB, AM, SAM, NFM
2012-09-24T01:03:12 < emeb> probably doable.
2012-09-24T01:03:43 -!- barthess [~barthess@178.154.102.20] has quit [Ping timeout: 246 seconds]
2012-09-24T01:03:58 < emeb> just need to to a coarse tune & decimate up front to get the sample rate down to something reasonable.
2012-09-24T01:04:17 < prog> the IQ data (16 bit x 2) was streamed from SDR-IP via ethernet
2012-09-24T01:04:31 < prog> absolutely
2012-09-24T01:04:49 < emeb> aha - using the ethernet MAC in the F4x7?
2012-09-24T01:05:29 < prog> I have a video of it.
2012-09-24T01:05:36 < prog> hold on
2012-09-24T01:05:57 < Laurenceb_> i was wondering if you could use DAC for LO generation
2012-09-24T01:06:01 < Laurenceb_> probably too slow
2012-09-24T01:06:51 < emeb> Laurenceb_: the STM32 DACs have a fairly low slew rate.
2012-09-24T01:07:20 < emeb> can't get much more than 100kHz BW out of them, no matter how fast you feed them data.
2012-09-24T01:08:08 < Laurenceb_> yeah i see now
2012-09-24T01:08:10 < Laurenceb_> sucks
2012-09-24T01:08:23 < Laurenceb_> you could use DMA to a GPIO and a resistor ladder i guess
2012-09-24T01:08:58 < Laurenceb_> or to me more clever, use a couple of parallel interface DACs
2012-09-24T01:09:02 < emeb> Yeah. I'd go for a real parallel DAC though - just to keep the spectral purity.
2012-09-24T01:09:29 < emeb> resistor ladders from GPIO are really glitchy & have lots of PS feedthru.
2012-09-24T01:09:34 < Laurenceb_> yeah
2012-09-24T01:10:57 < emeb> Something like this might work for an HF LO: http://ebrombaugh.studionebula.com/radio/txdac/index.html
2012-09-24T01:11:10 < emeb> depends on how fast you can get DMA to run through a GPIO.
2012-09-24T01:11:12 < Laurenceb_> i was pondering a single conversion SDR device
2012-09-24T01:12:57 < Laurenceb_> wonder if you could do a "DDR" interface
2012-09-24T01:13:10 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-24T01:13:18 < Laurenceb_> clk in data to the I on rising edges, Q on falling
2012-09-24T01:14:57 < Laurenceb_> 42Mhz update might be possible
2012-09-24T01:15:03 < Laurenceb_> with 16 bit wide interface
2012-09-24T01:15:23  * Laurenceb_ zzz
2012-09-24T01:20:01 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-09-24T01:20:50 < Thorn> why not, it's memory
2012-09-24T01:24:28 -!- barthess [~barthess@164.177.208.109] has joined ##stm32
2012-09-24T01:25:34 < HTT-Bird> I wonder if you could DMA to/from bit band addresses
2012-09-24T01:26:02 < HTT-Bird> would allow you to bitbang serial at a rather insane rate xD
2012-09-24T01:26:48 < zyp> no
2012-09-24T01:27:06 -!- barthess [~barthess@164.177.208.109] has quit [Client Quit]
2012-09-24T01:27:10 < zyp> the bit banding is performed in the cpu
2012-09-24T01:27:32 < HTT-Bird> darn. :p
2012-09-24T01:27:33 < zyp> it's not visible to other bus masters, just like the CCM
2012-09-24T01:35:25 < zippe> flyback: depends on who is the flow controller
2012-09-24T01:35:46 < zippe> flyback: another way of thinking about it is that DMA-capable peripherals have flow control abilities, but memory doesn't
2012-09-24T01:59:09 < Thorn> as everybody knows, you send a STALL to indicate a control request isn't supported. in which stage of the control transfer is this STALL sent?
2012-09-24T02:05:09 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 260 seconds]
2012-09-24T02:11:29 < zyp> you discover the error during the setup stage, so you report it during the data stage
2012-09-24T02:11:54 < zyp> (or status stage if there is no data stage)
2012-09-24T02:14:17 < zyp> you set the device controller to perform a stall after you have interpreted the status packet, and then the device controller will return a stall handshake the next time the host attempts to do an in or out transaction
2012-09-24T02:14:44 < Thorn> makes sense
2012-09-24T02:14:53 < Thorn> I wonder why there's a status stage
2012-09-24T02:15:12 < Thorn> a zero length data stage could be used instead for no data requests
2012-09-24T02:15:23 < zyp> huh?
2012-09-24T02:15:35 < zyp> the status stage is a zero length data stage.
2012-09-24T02:15:43 < Thorn> yes
2012-09-24T02:16:24 < Thorn> I mean make the data stage required (with 0 length if needed) and delete the status stage
2012-09-24T02:16:44 < zyp> what's the difference?
2012-09-24T02:16:50 < Thorn> or is it designed to give the device time to parse the data?
2012-09-24T02:16:57 < zyp> yes.
2012-09-24T02:17:31 < Thorn> because you have to answer quickly while the transaction is in progress
2012-09-24T02:18:30 < zyp> the point of the status stage is that the device can answer «ok, I am done processing your request and ready for your next one»
2012-09-24T02:19:13 < zyp> or «all changes that you asked me to perform have now been performed»
2012-09-24T02:19:15 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-24T02:20:32 < zyp> take set_configuration for example
2012-09-24T02:21:05 < zyp> it's supposed to configure and enable all the endpoints in the selected configuration before it returns
2012-09-24T02:22:52 < zyp> consider that you might be changing from one configuration to another, the host would like to know when it can start using the endpoints as described by the new configuration and not risk sending new data to endpoints still using the old one
2012-09-24T02:23:46 < Thorn> makes sense.
2012-09-24T02:34:30 -!- DaKu is now known as daku
2012-09-24T02:38:45 < upgrdman> is this a good value for someone who wants to try out all of the add-ons: http://www.ebay.com/itm/STM32F4DISCOVERY-STM32F407-STM32-ARM-Cortex-M4-Development-Board-15-Modules-Kit-/261053431839?pt=LH_DefaultDomain_0&hash=item3cc7ff401f
2012-09-24T02:40:52 < Thorn> ULPI via pin headers?
2012-09-24T02:42:52 < zyp> I got the one with onboard F4 instead of discovery board
2012-09-24T02:43:01 < zyp> http://wvshare.com/product/Open407Z-Package-B.htm
2012-09-24T02:43:37 < zyp> it's a bit more flexible since you don't have to worry about stuff on the discovery board already occupying some peripherals
2012-09-24T02:44:48 < Thorn> and the stm32 has more pins
2012-09-24T02:45:13 < upgrdman> whats ulpi?
2012-09-24T02:45:29 < zyp> interface for external usb phy
2012-09-24T02:45:48 < zyp> one of the included addon boards have one
2012-09-24T02:45:49 < upgrdman> zyp, have you played with the add-ons? are the good (considered the price)
2012-09-24T02:46:21 < zyp> not yet, haven't had time
2012-09-24T02:46:46 < zyp> I bought it because it seemed useful to have around for when I need it
2012-09-24T02:47:43 < zyp> and the price seems pretty good to me
2012-09-24T02:50:51 < zyp> I'll use the usb phy board as soon as I have time to play with high speed, and I considered using the microsd board to make a mass storage demo for USB
2012-09-24T02:51:06 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-24T02:51:27 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-24T02:52:45 < upgrdman> is waveshare a brand, or oem, or what?
2012-09-24T02:53:15 < zyp> I believe it's some chinese company that makes these devboards
2012-09-24T03:03:49 < zippe> Yes, they are
2012-09-24T03:12:35 < dongs> http://i.imgur.com/hBq0a.gif
2012-09-24T03:17:49 < emeb> harr...
2012-09-24T03:20:18 -!- izua [~izua@86.121.79.148] has joined ##stm32
2012-09-24T03:20:19 -!- izua [~izua@86.121.79.148] has quit [Changing host]
2012-09-24T03:20:19 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-24T03:20:22 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-24T03:24:41 < dongs> they also have decent pricing on stm32 and cp2102 ):
2012-09-24T03:24:52 < dongs> speaking of which where the fuck is my tray of that shit from them
2012-09-24T03:31:28 < emeb> gcc -02 seems to do pretty well. Trying to do some obvious optimizations by hand and none seem to do better.
2012-09-24T03:40:01 < dongs> http://i.imgur.com/T1b1d.jpg more pcb trolling
2012-09-24T03:40:05 < dongs> im glad these guys are paying.
2012-09-24T03:40:20 < Thorn> ookay but what about SET_ADDRESS? when is the address actually set? before the status stage?
2012-09-24T03:41:17 < Thorn> i.e. is it changed in the middle of a transfer?
2012-09-24T03:42:16 < Thorn> I recall lthat some driver code saves the address for later when it's received
2012-09-24T03:43:37 < Thorn> device enumeration http://e2e.ti.com/cfs-file.ashx/__key/CommunityServer-Discussions-Components-Files/110/2161.C5515_5F00_CSL250_5F00_USB_5F00_example3_5F00_trace.JPG
2012-09-24T03:44:30 < dongs> cool scope bro
2012-09-24T03:45:02 < Thorn> I wish
2012-09-24T03:52:21 < dongs> ive got ellisys usb shiz
2012-09-24T03:54:10 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-24T03:55:18 -!- NonaSuomy [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has joined ##stm32
2012-09-24T03:55:18 -!- NonaSuomy [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has quit [Changing host]
2012-09-24T03:55:18 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-24T03:56:54 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 240 seconds]
2012-09-24T03:59:44 < dongs> supposedly revision C of my scope can do USB-FS by software + key (which I can hax) but mine is rev B :(
2012-09-24T04:00:32 < Thorn> This notification endpoint isn't implemented. According to CDC spec it's optional, but its absence causes a NULL pointer dereference in the Linux cdc_acm driver.
2012-09-24T04:00:39 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 268 seconds]
2012-09-24T04:00:39 < Thorn> lol
2012-09-24T04:01:02 < dongs> lunix cdc driver
2012-09-24T04:01:02 < dongs> great
2012-09-24T04:01:23 < Bird|lappy> Thorn, nobody's patched that? o.O
2012-09-24T04:01:31 < Thorn> http://libopencm3.git.sourceforge.net/git/gitweb.cgi?p=libopencm3/libopencm3;a=blob;f=examples/stm32/f4/stm32f4-discovery/usb_cdcacm/cdcacm.c;h=4af1039344c0168dd561c7cdcb7931cab9ebfe66;hb=HEAD
2012-09-24T04:01:52 < Thorn> you can test if you have a f4disco LOL
2012-09-24T04:02:00 < dongs> lol
2012-09-24T04:02:22 < Thorn> how do I crashed lunix with a usb device
2012-09-24T04:06:27 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T04:14:28 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 248 seconds]
2012-09-24T04:17:00 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-24T04:22:17 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-24T04:24:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 264 seconds]
2012-09-24T04:24:37 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Ping timeout: 246 seconds]
2012-09-24T04:45:26 < upgrdman> zyp, did your waveshare stuff come with any documentation for the modules?
2012-09-24T04:47:35 < dongs> 56like what?
2012-09-24T04:47:43 < dongs> it comes wiht a CD with schematics for all teh shit
2012-09-24T04:47:45 < dongs> what else do you need?
2012-09-24T04:47:57 < dongs> and design files too if I remember
2012-09-24T04:48:06 < upgrdman> ? 56?
2012-09-24T04:48:11 < dongs> ignore 56.
2012-09-24T04:48:18 < dongs> probably leftover from some shit i was trying to type.
2012-09-24T04:48:41 < upgrdman> are there datasheets for the various bits
2012-09-24T04:49:25 < dongs> ???
2012-09-24T04:49:28 < dongs> its all standard shit dude
2012-09-24T04:49:33 < dongs> go to manufacturer website and get it
2012-09-24T04:49:38 < dongs> but yea, all that shit is there too
2012-09-24T04:49:41 < upgrdman> ok
2012-09-24T04:49:58 < upgrdman> do you like your kit? which one did you get?
2012-09-24T04:50:50 < dongs> i got f207b
2012-09-24T04:50:54 < dongs> same shit as zyp but F2 proc.
2012-09-24T04:53:35 < upgrdman> like it?
2012-09-24T04:55:16 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-24T04:58:39 < upgrdman> anyone sell them i the usa?
2012-09-24T05:04:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-24T05:35:48 < emeb_mac> *crickets*
2012-09-24T05:44:16 < upgrdman> yes, lots of crickets
2012-09-24T05:51:24 < dongs> ;why not jut buy the shit from them?
2012-09-24T05:52:30 < dongs> thgey will send byt fedex for cheap, only takes liek 3-4 days to usa
2012-09-24T05:55:32 < dongs> just got my tray of f103 and cp2102 from them  this morning
2012-09-24T06:13:16 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-24T06:22:13 < Bird|lappy> hrm, speaking of usb->serial: I wonder if anyone here has used the Exar parts?
2012-09-24T06:28:50 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-24T06:28:57 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-24T06:29:41 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T06:31:04 < dongs> LIFX uses a mesh network. The standby power usage of each bulb is very small, in fact a AA battery would keep a LIFX smartbulb in standby mode for approximately 1-2 years.
2012-09-24T06:31:07 < dongs> waaaaaaaaaat
2012-09-24T06:33:08 < dongs> trollin the trolls
2012-09-24T06:45:34 < dongs> yep.
2012-09-24T06:45:42 < dongs> forgot to order a diode i needed for a lot of shit.
2012-09-24T06:45:42 < dongs> oops.
2012-09-24T06:46:29 < Bird|lappy> "Never has so much trouble been had by so many with so few pins."
2012-09-24T06:47:33 < dongs> dix :(
2012-09-24T06:53:57 < iR0b0t1> .
2012-09-24T06:54:03 < dongs> .
2012-09-24T06:54:09 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-24T06:54:09 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-24T06:54:15 -!- iR0b0t1 is now known as iR0b0t1_
2012-09-24T06:54:16 < iR0b0t1_> .
2012-09-24T06:54:19 < dongs> .
2012-09-24T06:54:58 < dongs> cool host bro
2012-09-24T06:57:09 -!- iR0b0t1_ [~dev@unaffiliated/r0b0t1] has quit [Quit:  HydraIRC -> http://www.hydrairc.com <- Go on, try it!]
2012-09-24T06:58:21 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T06:58:21 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-24T06:58:21 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-24T07:56:02 < emeb_mac> wub wub wub
2012-09-24T08:04:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-24T08:13:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-24T08:39:27 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-24T08:43:07 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-24T08:50:15 < dongs> http://www.everlight.com/index.php?option=com_k2&view=item&id=463:18-038&Itemid=323&lang=en
2012-09-24T08:50:18 < dongs> fuck, i want a reel of these
2012-09-24T08:59:03 < emeb_mac> those look very nice
2012-09-24T09:03:24 < dongs> ya man
2012-09-24T09:03:28 < dongs> 1mm^2 RGB
2012-09-24T09:04:04 < jpa-> zippe: hmm.. squirrel seems very cool in fact
2012-09-24T09:08:59 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-24T09:13:18 < dongs> http://www.everlight.com/index.php?option=com_k2&view=item&id=117:61-238&Itemid=185&lang=en maybe ill just get some of these
2012-09-24T09:13:23 < dongs> more normal sized.
2012-09-24T09:25:13 < zyp> Thorn, set_address is the exception, it has to return the status stage from address 0 before starting to accept packets on it's new address
2012-09-24T09:48:16 -!- claude is now known as Claude
2012-09-24T09:50:23 < Thorn> thought so. thanks much
2012-09-24T09:53:37 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-24T10:00:08 -!- Claude is now known as claude
2012-09-24T10:04:49 < Thorn> wtf is wrong with diptrace, no updates in a year
2012-09-24T10:08:34 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 252 seconds]
2012-09-24T10:09:00 -!- nashpa_ [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-24T10:09:16 -!- nashpa_ is now known as nashpa
2012-09-24T10:09:59 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-24T10:14:11 -!- dekar_ [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-09-24T10:39:29 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-24T10:41:17 < Laurenceb_> as we make you our bitch, ST stands for "Sexy Time"
2012-09-24T10:43:02 < zyp> that was the most useful thing Laurenceb_ is going to say today
2012-09-24T10:43:15 < cjbaird> pornblog: https://www.youtube.com/watch?v=PQuT-Xfyk3o
2012-09-24T10:43:47 < Laurenceb_> censored
2012-09-24T10:45:01 < cjbaird> an underage b& for you!
2012-09-24T10:45:46 < dongs>  < Thorn> wtf is wrong with diptrace, no updates in a year
2012-09-24T10:45:51 < dongs> Thorn: what? they have a beta up from like a month ago
2012-09-24T10:45:56 < dongs> with great new shit & all bugs I reported fixed
2012-09-24T10:48:02 < dongs> Thorn: what are you missing in it?
2012-09-24T11:00:14 < Thorn> dunno, I haven't seen the download link for it
2012-09-24T11:00:21 < Thorn> will test and see
2012-09-24T11:02:18 -!- dekar [~dekar@dslb-088-069-140-090.pools.arcor-ip.net] has joined ##stm32
2012-09-24T11:02:22 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-24T11:03:41 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-09-24T11:09:24 < dongs> http://www.ebay.com/itm/330798161444 damn, i could use this
2012-09-24T11:09:29 < dongs> i'd have a lifetime supply of jumpers.
2012-09-24T11:16:13 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-24T11:43:40 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-24T11:43:46 < Thorn> I have these, they're single use only
2012-09-24T11:44:37 < Thorn> I had some quality debugging time with them. 2 pins with a jumper on them, touch one pin with scope probe, signal, touch the other, nothing
2012-09-24T11:50:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-24T11:50:46 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-24T11:59:28 < dongs> Thorn: heh
2012-09-24T12:01:30 -!- claude is now known as claude_work
2012-09-24T12:29:50 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-24T12:57:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-24T13:05:06 < Laurenceb> http://www.ohloh.net/p/stm32-logger
2012-09-24T13:05:07 < Laurenceb> wut
2012-09-24T13:05:17 < Laurenceb> they have be analising my github
2012-09-24T13:06:10 < jpa-> 21 years of effort! :)
2012-09-24T13:06:13 < Laurenceb> http://www.ohloh.net/p/stm32-logger/estimated_cost
2012-09-24T13:06:14 < Laurenceb> hahaha
2012-09-24T13:14:45 < dongs> ahaha
2012-09-24T13:14:58 < dongs> ...
2012-09-24T13:14:58 < dongs> is mostly written in C
2012-09-24T13:14:59 < dongs> with a well commented source code
2012-09-24T13:15:00 < dongs> hahahaha
2012-09-24T13:16:47 < Laurenceb> https://github.com/Laurenceb/STM32-Logger/blob/master/bootload.c
2012-09-24T13:16:49 < dongs> ah
2012-09-24T13:16:52 < dongs> i found the real statistics
2012-09-24T13:16:53 < Laurenceb> hopefully this will work
2012-09-24T13:17:02 < dongs> Aggregate coding time: ~10m
2012-09-24T13:17:04 < dongs> for C
2012-09-24T13:17:15 < dongs> the rest of 21 man-years were copypaste
2012-09-24T13:19:15 < Laurenceb> where?
2012-09-24T13:19:43 < dongs> http://www.ohloh.net/p/stm32-logger/contributors/2576471424483786
2012-09-24T13:19:48 < dongs> bottom
2012-09-24T13:20:24 < karlp> texane/stlink is only $67k.
2012-09-24T13:20:29 < karlp> you're winning laurence :)
2012-09-24T13:26:12 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has joined ##stm32
2012-09-24T13:52:01 < Laurenceb> rage
2012-09-24T13:52:13 < Laurenceb> the watchdog kills bluetooth upload
2012-09-24T13:53:36 -!- Oldboy [myircshell@grid.sh3lls.net] has quit [Quit: leaving]
2012-09-24T13:54:13 < Laurenceb> theres only  one way to do this
2012-09-24T13:54:19 < Laurenceb> bluetooth gpio commands
2012-09-24T14:02:47 -!- BjoernC [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has joined ##stm32
2012-09-24T14:12:11 < Laurenceb> ive got a transflash card... its made it thailand
2012-09-24T14:13:09 < dongs> funny level: chick with dix
2012-09-24T14:13:58 -!- daku is now known as DaKu
2012-09-24T14:25:53 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-gcomyincbmqglsww] has joined ##stm32
2012-09-24T14:37:02 < dongs> zippe: RGMII is pretty standard right, i dont have to use some vendor specified PHYs if I dont want to?
2012-09-24T14:43:14 < zyp> isn't that the point of the MII standards?
2012-09-24T14:43:27 < dongs> who fucking knows
2012-09-24T14:44:22 < Thorn> you still need to use vendor extensions to configure the thing etc. if I'm not mistaken
2012-09-24T14:47:15 < Thorn> so I want to make a usb device with roughy speaking 5 serial ports
2012-09-24T14:47:46 < Thorn> more accurately, 2 CAN, 2 RS-485 interfaces and one radio (nordic module)
2012-09-24T14:48:04 < dongs> more like nordick
2012-09-24T14:48:13 < Thorn> I selected a STM32F405RGT6 and I'm going to use the HS controller with 5 EP pairs
2012-09-24T14:48:16 < dongs> all 5 at once?
2012-09-24T14:48:25 < Thorn> preferably yes
2012-09-24T14:48:31 < Thorn> how do I do that?
2012-09-24T14:48:45 < dongs> just make a composite device descriptor i guess
2012-09-24T14:48:48 < dongs> zyp would know.
2012-09-24T14:50:33 < zyp> there are a few different ways of doing that
2012-09-24T14:51:10 < Thorn> oh, and it needs to accept commands to toggle LEDs etc. (I guess control requests will do)
2012-09-24T14:51:36 < Thorn> my primary concern is how to let the host know there's data to receive
2012-09-24T14:51:45 < dongs> um
2012-09-24T14:51:48 < dongs> you dont 'let know' in usb
2012-09-24T14:51:55 < dongs> you get a interrupt every 1.25ms or someshit
2012-09-24T14:51:59 < dongs> if you got shit to send
2012-09-24T14:52:03 < dongs> you just send it at that time.
2012-09-24T14:52:04 < dongs> something like that.
2012-09-24T14:52:22 < dongs> device cant initiate any transfer to host
2012-09-24T14:52:53 < zyp> the host is supposed to poll each IN endpoint when it expects data, and gets a NAK if there is none
2012-09-24T14:53:39 < Laurenceb> arg wtf
2012-09-24T14:53:49 < Laurenceb> i cant get the CMD dialogue from my bluetooth
2012-09-24T14:53:55 < Thorn> well there're interrupt transfers where the host does the polling automagically
2012-09-24T14:54:14 < dongs> your uart shit is bulk transfers.
2012-09-24T14:54:14 < Thorn> but if I use bulk the driver or even the application will have to poll
2012-09-24T14:54:28 < zyp> nah
2012-09-24T14:54:33 < Thorn> how does FTDI do it with bulk EPs?
2012-09-24T14:54:39 < zyp> same way
2012-09-24T14:54:40 < Thorn> or CDC-ACM
2012-09-24T14:54:50 < zyp> same way
2012-09-24T14:55:11 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-24T14:55:11 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-24T14:55:11 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-24T14:55:18 < Thorn> because right now my FTDI CAN thing is polled by the application, which is kind of horrible
2012-09-24T14:55:49 < Thorn> but the reason for that is to see if the external controller has a received frame
2012-09-24T14:56:04 < zyp> the driver/application/whatever just says «do a bulk transfer», and then the host stack does all the polling, until it receives data which it returns
2012-09-24T14:56:06 < Thorn> if it were rs-485 I wouldn't need to do that manually
2012-09-24T14:56:23 < Thorn> oh that is nice
2012-09-24T14:56:58 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T14:56:58 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-24T14:56:58 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-24T14:57:00 < zyp> http://bin.jvnv.net/f/PL6RR.png <- see the IN-NAKs here? that's the host polling for new data on my ACM endpoint
2012-09-24T14:58:26 < Thorn> what type is this endpoint, bulk? I came to believe "periodic" only refers to interrupt INs
2012-09-24T14:59:31 < zyp> yes
2012-09-24T14:59:50 < zyp> I believe interrupt endpoints are mostly about priorities
2012-09-24T15:00:03 < zyp> they are supposed to be used for low-latency small-data stuff
2012-09-24T15:00:38 < zyp> while isochronous are fixed rate, and bulk are best-effort
2012-09-24T15:01:35 < Thorn> hmm. maybe I missed this in the stuff that I read, even though I spent all of last night reading usb complete
2012-09-24T15:02:19 < Thorn> what is the polling interval for bulk EPs then, once per frame? (and why is it configurable for interrupt?)
2012-09-24T15:02:19 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-09-24T15:02:36 < zyp> oh, and I'm by no means an expert on this stuff, I just have a bit of practical knowledge :p
2012-09-24T15:03:07 < zyp> I have no idea, I believe it's «as often as the host has free time»
2012-09-24T15:04:07 < dongs>  once per frame? < i think so
2012-09-24T15:05:43 < Thorn> I get it. zyp knows so much because he has an analyzer. UNFAIR!
2012-09-24T15:07:39 < zyp> :)
2012-09-24T15:08:28 < zyp> find some usb developers and befriend them :p
2012-09-24T15:08:41 < BrainDamage> or shell 500$ for one
2012-09-24T15:08:53 < Thorn> so I just use an endpoint pair per port (bulk) and custom control transfers for blinky stuff, right?
2012-09-24T15:09:17 < Thorn> although it would be nice to also report CAN error conditions for example
2012-09-24T15:09:56 < Thorn> maybe IN control requests
2012-09-24T15:16:49 < Thorn> but that will be a completely non-standard device
2012-09-24T15:17:03 < Thorn> how many EPs do you need for 2 CDC-ACM ports?
2012-09-24T15:22:56 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-09-24T15:23:21 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds]
2012-09-24T15:24:26 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T15:29:37 < dongs> well, you need one per rx/tx so.. 4?
2012-09-24T15:31:56 < dongle> fuck freenode
2012-09-24T15:35:07 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-24T15:35:33 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T15:37:24 < dongle> http://hackaday.com/2012/09/24/programming-a-microcontroller-one-bit-at-a-time/ the kinda shit that passes as "useful contents" on hackaday these days
2012-09-24T15:38:52 < Thorn> haha http://habrahabr.ru/post/152052/
2012-09-24T15:39:32 < Thorn> ah they did give a link
2012-09-24T15:42:07 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds]
2012-09-24T15:42:51 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T15:47:01 < cjbaird> lol, the only cryllic font on my system is from the VT220 DEC terminal ttf. 213374u http://images.4chan.org/g/src/1348490755541.gif (and imgur is still crapped)
2012-09-24T15:48:12 < dongle> rofl
2012-09-24T15:48:14 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-24T15:48:26 < dongle> the title of that image should be "lunix fonts in 2012"
2012-09-24T15:49:04 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T15:49:19 < Thorn> this is a matrix printer imitation?
2012-09-24T15:50:01 < dongle> no
2012-09-24T15:50:09 < dongle> this is some terminal from 1986 imitation
2012-09-24T15:50:31 < dongle> leave it to wikipedos to document useless shit
2012-09-24T15:50:32 < dongle> http://en.wikipedia.org/wiki/VT220
2012-09-24T15:52:00 < cjbaird> One of those as a Unix terminal set to 4800 baud can lay ruin to entire countries' Internets.
2012-09-24T15:52:31 < dongle> so can any modern laptop
2012-09-24T15:52:43 < Thorn> that terminal had such a large distance between scanlines or what?
2012-09-24T15:52:55 < Thorn> ok, time to lol @ google translate.
2012-09-24T15:52:56 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-24T15:53:00 < dongle> scatlines
2012-09-24T15:53:09 < cjbaird> And also what I got to use as an undergrad from 1989-1994. No-one would touch the VT100s when the '220s arrived..
2012-09-24T15:53:13 < Thorn> and we begin with the title: AVR are sewn by hand LOL
2012-09-24T15:54:05 < cjbaird> sensi.org/~svo/glasstty/
2012-09-24T15:54:16 < Laurenceb> arg wtf
2012-09-24T15:54:25 < Laurenceb> bluetooth flash still not working
2012-09-24T15:54:42 < Laurenceb> i get the stm into bootload mode using gpio on the bluetooth
2012-09-24T15:55:00 < Laurenceb> then Unknown response. 0x0: 0x4e
2012-09-24T15:56:21 < cjbaird> Ah, the article that details how the character rasterization was done, which I ment to post: http://vt100.net/dec/vt220/glyphs
2012-09-24T15:57:14 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 240 seconds]
2012-09-24T15:58:16 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T15:59:48 < Thorn> "Look, took a run" - the reader will say, - "I'm on a desert island. Where I'll get there triggers?
2012-09-24T16:01:22 < Laurenceb> oh
2012-09-24T16:01:24 < Laurenceb> fail
2012-09-24T16:01:37 < Laurenceb> you need to set parity on the module
2012-09-24T16:03:44 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-24T16:04:36 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T16:05:59 < Laurenceb> http://www.theregister.co.uk/2012/09/22/liquid_nitrogen_and_ping_pong_balls/
2012-09-24T16:06:00 < Laurenceb> wtf
2012-09-24T16:06:09 < Laurenceb> someone needs to get a life
2012-09-24T16:11:32 < dongle> Laurenceb: old. saw that like 2 days ago
2012-09-24T16:13:39 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-24T16:14:04 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-24T16:14:04 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-24T16:14:04 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-24T16:14:06 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-24T16:14:13 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T16:17:41 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-09-24T16:19:50 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-24T16:20:12 < Thorn> lulz. looks like yesterday's bug http://www.mail-archive.com/linux-usb@vger.kernel.org/msg04422.html
2012-09-24T16:21:50 < Thorn> ah no. "If a device specifies zero endpoints"
2012-09-24T16:22:15 < Bird|lappy> Thorn, yeah...wtf sort of USB device is /that/?
2012-09-24T16:22:41 <+izua> an usb lamp
2012-09-24T16:23:25 < Thorn> any usb device in development? someone's first attempt at ausb device? etc.
2012-09-24T16:23:26 < Bird|lappy> izua, do those even enumerate?
2012-09-24T16:23:38 < Bird|lappy> yeah. it's possible to trigger that while hacking on a device
2012-09-24T16:23:38 <+izua> not really
2012-09-24T16:23:39 < Bird|lappy> but o/
2012-09-24T16:23:48 <+izua> but as a consequence, they also have 0 endpoints
2012-09-24T16:24:07 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-24T16:24:42 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 264 seconds]
2012-09-24T16:25:28 < Thorn> usb lamps / fans / etc just draw power, they don't connect d+/d- to anything afaik
2012-09-24T16:25:30 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T16:25:36 < Thorn> (which is illegal)
2012-09-24T16:28:59 < zyp> the default control endpoint (0) is never specified
2012-09-24T16:29:12 < zyp> so that would apply for any control-only device
2012-09-24T16:30:02 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-24T16:30:14 < zyp> oh and yeah, that looks like the exact bug that was mentioned yesterday
2012-09-24T16:30:23 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T16:31:05 < zyp> «If a device specifies zero endpoints in its interface descriptor»
2012-09-24T16:31:40 < Thorn> that bug was about an acm device w/o an interrupt EP afair (with bulk ones present)
2012-09-24T16:31:41 < zyp> devices often have several interface descriptors, so I'm guessing this particular one is the CDC ACM control interface
2012-09-24T16:32:09 < zyp> (ACM devices have one control interface and one data interface, the interrupt endpoint is defined in the control interface)
2012-09-24T16:32:12 < Thorn> although I'm not yet familiar with the cdc class so I may be mistaken
2012-09-24T16:32:25 < zyp> while the two bulk endpoints are defined in the data interface
2012-09-24T16:32:54 < Thorn> ah so it results in a zero EP control interface
2012-09-24T16:33:35 < zyp> exactly
2012-09-24T16:33:52 < zyp> see http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_cdc_acm for example of ACM descriptors
2012-09-24T16:33:55 < Thorn> neat.
2012-09-24T16:36:35 < Thorn> so how many EPs do you need for 2 ACM ports? 2 pairs bulk + 2 IN interrupt?
2012-09-24T16:37:40 < zyp> I haven't checked the spec if the interrupt in is strictly required
2012-09-24T16:38:14 < zyp> but if you're not using it anyway, you could just write a descriptor and point it to a nonexistant endpoint number
2012-09-24T16:38:30 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 264 seconds]
2012-09-24T16:38:36 < Thorn> because if I try to make my rs-485 interfaces ACM that will leave me with only 1 free IN EP
2012-09-24T16:38:41 < zyp> the host would not get any activity from it, so it will just see it as NAKing and interpret it as no data
2012-09-24T16:40:07 < Laurenceb> are there different versions of the stm32 bootloader?
2012-09-24T16:40:13 < Thorn> synopsys, have you read the usb spec? 15 pairs of EPs are allowed! not 5! and definitely not 3
2012-09-24T16:40:14 < Laurenceb> im getting 0x69 not 0x79
2012-09-24T16:40:23 < Laurenceb> as init reply
2012-09-24T16:40:32 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-24T16:41:08 < Thorn> Laurenceb: yes, one version had an interesting bug iirc, it could short circuit some pin or something
2012-09-24T16:41:18 < Laurenceb> the hell
2012-09-24T16:41:21 < Laurenceb> hmm
2012-09-24T16:41:32 < Thorn> seen that in errata
2012-09-24T16:41:34 < Laurenceb> maybe ill change the code to accept 0x69 as a valid rely
2012-09-24T16:41:47 < Laurenceb> this is F1
2012-09-24T16:44:41 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-24T16:47:49 < Mazingaro> hi
2012-09-24T16:49:57 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 260 seconds]
2012-09-24T16:50:31 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T16:51:04 < Laurenceb> FUCK YEAH
2012-09-24T16:51:07 < Laurenceb> its working
2012-09-24T16:51:19 < Laurenceb> cool part: board is currently 500m away
2012-09-24T16:51:24 < Laurenceb> class 1 bluetooth+ yagi
2012-09-24T16:53:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-24T16:56:29 <+izua> 500 m away?
2012-09-24T16:57:08 < Thorn> yagi? that huge thing on a pole?
2012-09-24T16:58:26 < BrainDamage> http://dx.com/p/2-4ghz-16dbi-directional-high-gain-16-unit-antenna-for-wifi-wireless-network-rp-sma-33016?item=3
2012-09-24T16:58:38 < BrainDamage> that's a yagi
2012-09-24T17:01:16 < Laurenceb> flashed ok
2012-09-24T17:01:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-24T17:01:28 < Laurenceb> just its bricked it now
2012-09-24T17:01:50 < Laurenceb> i forgot to turn off the gpio line that prevents power off
2012-09-24T17:02:06 < Laurenceb> and the gpio config timeout has expired on the bluetooth
2012-09-24T17:02:35 <+izua> where have you hidden it?
2012-09-24T17:02:44 < Laurenceb> ill have to go over and power cycle it, then try again with a line to fix that at the start of the process
2012-09-24T17:02:54 <+izua> inb4 cast in concrete
2012-09-24T17:03:01 < Laurenceb> izua: in the suspended ceiling of a disused office
2012-09-24T17:03:04 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 246 seconds]
2012-09-24T17:03:04 < Laurenceb> bbl
2012-09-24T17:03:40 < Laurenceb> lol @ "fuckufreak" bluetooth name
2012-09-24T17:03:56 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T17:03:57 < Laurenceb> my "yagi base station" is called pedobear
2012-09-24T17:04:24 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Disconnected by services]
2012-09-24T17:04:25 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-24T17:06:23 < BrainDamage> create another with "Scotland Yard" or similar
2012-09-24T17:07:56 < Laurenceb> "Chris Hanson"
2012-09-24T17:09:23 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-24T17:09:45 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-24T17:11:23 < Thorn> ubuntu 12.10 will display ads lol.
2012-09-24T17:11:25 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 260 seconds]
2012-09-24T17:12:33 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T17:18:08 <+izua> lol, where?
2012-09-24T17:18:47 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds]
2012-09-24T17:19:41 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T17:21:05 < jpa-> in search
2012-09-24T17:21:23 < jpa-> can be disabled though
2012-09-24T17:23:07 < Thorn> I take it the vbus pin is mandatory for synopsys usb cores?
2012-09-24T17:25:28 < Thorn> because of builtin pullup resistors etc.
2012-09-24T17:27:50 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-24T17:28:20 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T17:31:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-24T17:31:41 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-24T17:40:21 < Laurenceb> does stm32 reboot automatically on exiting bootloader?
2012-09-24T17:41:25 < Laurenceb> oh theres a "go"  command
2012-09-24T17:41:28 < Laurenceb> i see now
2012-09-24T17:41:37 <+Steffanx> Yeah, incl. windows shutdown and boot sound
2012-09-24T17:41:57 <+Steffanx> You can choose the windows version
2012-09-24T17:41:59 < Laurenceb> lol
2012-09-24T17:42:27 < Laurenceb> anyone here used the bootload commands to access sram or peripherals?
2012-09-24T17:47:02 -!- potomac [~kris@static-87-105-185-38.ssp.dialog.net.pl] has joined ##stm32
2012-09-24T17:48:34 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 240 seconds]
2012-09-24T17:48:36 -!- claude_work is now known as claude
2012-09-24T17:54:32 -!- potomac [~kris@static-87-105-185-38.ssp.dialog.net.pl] has quit [Quit: Wychodzi]
2012-09-24T18:05:43 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-24T18:06:58 -!- BjoernC [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-24T18:09:01 -!- claude is now known as Claude
2012-09-24T18:11:58 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-24T18:13:34 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-24T18:15:19 < jpa-> dekar: have you used nuttx besides apparently getting it working on some dev board?
2012-09-24T18:17:52 -!- Claude is now known as claude
2012-09-24T18:29:08 -!- claude is now known as Claude
2012-09-24T18:32:59 < Tectu> zippe, can you tell me the conflict when replacing F1 with F4?
2012-09-24T18:38:11 < zyp> read the pdf on it?
2012-09-24T18:38:35 < zyp> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/APPLICATION_NOTE/DM00033267.pdf
2012-09-24T18:39:02 < zyp> F4 is pretty much F2 with M4F, so that document also applies for F4
2012-09-24T18:40:05 < jpa-> DMA, USB, all kinds of small differences
2012-09-24T18:40:24 < jpa-> but DMA, USB, (and AFIO, but that's easy) are the biggest AFAIK
2012-09-24T18:44:13 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 246 seconds]
2012-09-24T18:45:38 -!- Claude is now known as claude
2012-09-24T18:48:45 < Thorn> and RCC
2012-09-24T18:49:35 < zyp> no major changes there
2012-09-24T18:49:53 < emeb> hardware wise, F4 has internal Vcore that needs a separate caps on some pins that are VDD on F1
2012-09-24T18:50:38 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-24T18:56:01 -!- dekar [~dekar@dslb-088-069-140-090.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-24T18:56:08 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-24T19:04:53 < Laurenceb> muhahha
2012-09-24T19:05:00 < Laurenceb> all works now
2012-09-24T19:05:17 < emeb> remote bluetooth bootloader?
2012-09-24T19:05:29 < Laurenceb> managed to get a proper reboot at the end now
2012-09-24T19:05:32 < Laurenceb> emeb: yes
2012-09-24T19:05:42 < Laurenceb> _very_  remote
2012-09-24T19:06:27  * emeb makes note to pwnz0r Laurenceb_'s remotes.
2012-09-24T19:07:32 < emeb> what kind of range do you get on BT?
2012-09-24T19:08:31 < Laurenceb> ~500m
2012-09-24T19:08:36 < Laurenceb> class 1 with a yagi
2012-09-24T19:08:43 < emeb> that's a fair distance
2012-09-24T19:10:39 < Laurenceb> people have done several Km
2012-09-24T19:10:57 < Laurenceb> im on top of a 25 story building
2012-09-24T19:15:21 < Laurenceb> http://www.youtube.com/watch?feature=player_embedded&v=mbwxXzIPxdc
2012-09-24T19:15:24 < Laurenceb> wtf hair
2012-09-24T19:15:31 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-24T19:16:39 < Laurenceb> is that... windows7 ?!
2012-09-24T19:17:07 < BrainDamage> saves on nighttime illumination costs
2012-09-24T19:17:10 < Laurenceb> millions of nerds just had to order some viagara
2012-09-24T19:17:26 < Laurenceb> after discovering she runs windows
2012-09-24T19:17:44 < BrainDamage> wat?
2012-09-24T19:18:31 < Laurenceb> the pc is running wondoze
2012-09-24T19:18:48 < BrainDamage> ...so?
2012-09-24T19:19:01 < Laurenceb> she now has -1 geek points
2012-09-24T19:19:43 -!- barthess [~barthess@178.154.106.50] has joined ##stm32
2012-09-24T19:20:17 < BrainDamage> sometimes I wonder if you got to PhD by sexing your way trough univ
2012-09-24T19:33:34 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 246 seconds]
2012-09-24T19:34:40 <+Steffanx> You NEED a range of 500 meter Laurenceb ?
2012-09-24T19:35:16 <+Steffanx> How did you do it BrainDamage ?
2012-09-24T19:35:20 <+Steffanx> Hard work?
2012-09-24T19:35:41 < emeb> liquor is quicker
2012-09-24T19:37:03 < BrainDamage> Steffanx: didn't finish yet, I guess I am doing something very wrong
2012-09-24T19:37:43 < BrainDamage> in fact I have an exam tomorrow :s
2012-09-24T19:38:09 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 244 seconds]
2012-09-24T19:38:10  * emeb doesn't mind not having exams any more.
2012-09-24T19:38:27 <+Steffanx> Exam in what BrainDamage ?
2012-09-24T19:38:37 < emeb> prostate
2012-09-24T19:38:38 < BrainDamage> power electronics
2012-09-24T19:38:46 <+Steffanx> lol emeb
2012-09-24T19:39:19 < BrainDamage> smps, actuator control, that kind of crap
2012-09-24T19:39:22 < emeb> http://www.youtube.com/watch?v=EfzXpE0QB2U
2012-09-24T19:39:23 < BrainDamage> booooooooooooooring
2012-09-24T19:42:07 <+Steffanx> BrainDamage is a slow student or just a regular one?
2012-09-24T19:42:46 < BrainDamage> a dumb lazy one
2012-09-24T19:43:16 < emeb> if you want to be an engineer you can either be dumb or lazy, but not both.
2012-09-24T19:43:19 <+Steffanx> Hmm as in .. busy with your study for 6 years now where it should take 4?
2012-09-24T19:44:18 < BrainDamage> here takes 5, this'd be 7, and I didn't do much over the past year
2012-09-24T19:44:56 <+Steffanx> 5 = bachelor + master?
2012-09-24T19:45:00 < BrainDamage> yes
2012-09-24T19:45:06 < emeb> not taking classes, or blowing off the ones you do take?
2012-09-24T19:45:33 < BrainDamage> already finished taking classes, I just need to pass the exams
2012-09-24T19:45:49 < zyp> emeb, so you're saying that since I'm an engineer and lazy, I can't be dumb?
2012-09-24T19:46:02 < emeb> zyp: exactly
2012-09-24T19:46:04 < BrainDamage> but I cannot find the motivation to study anymore
2012-09-24T19:46:13 < emeb> smart + lazy still works
2012-09-24T19:46:16 <+Steffanx> In your case, yes .. zyp :P
2012-09-24T19:46:34 < emeb> often smart + lazy is better - more efficient solutions. :)
2012-09-24T19:47:05 < emeb> "how do I get this done with minimum amount of typing?"
2012-09-24T19:47:09 < BrainDamage> that's optimizing time to market, not necessarily the implementation is efficient
2012-09-24T19:47:22 < zyp> hey, that's why I try to make simple interfaces for my code, I'm lazy so I hate writing lots of boilerplate :p
2012-09-24T19:47:33 < emeb> zyp: my point.
2012-09-24T19:49:07 <+Steffanx> Smart boobs for Laurenceb : http://i.imgur.com/nxJoh.gif
2012-09-24T19:49:16 <+Steffanx> 'smart'
2012-09-24T19:49:30 < emeb> OMGWTFXYZBBQ!
2012-09-24T19:49:37 < zyp> I've heard dolphins are quite smart
2012-09-24T19:50:44 < BrainDamage> I prefer boobs that can't potentially bit my fingers
2012-09-24T19:51:36 < emeb> that's what I like about ##stm32 - always something unusual going on here.
2012-09-24T19:52:17 <+Steffanx> The thing cjbeard didn't like
2012-09-24T19:52:38 <+Steffanx> The channel is still growing so i guess most people don't care :D
2012-09-24T19:52:56 <+Steffanx> *channel population
2012-09-24T19:53:30 < emeb> lot of lurkers
2012-09-24T19:53:59 < BrainDamage> they just fap to Laurenceb's links when reading the backlog
2012-09-24T19:54:13 <+Steffanx> All they need is zlog ..
2012-09-24T19:55:25 < ABLomas> nah, just urllog in irssi
2012-09-24T19:55:33 < karlp> huh, there is a few here now.
2012-09-24T19:55:34 < BrainDamage> I'd still like to know where to get the free z80 :p
2012-09-24T19:56:00 <+Steffanx> A few what karlp ?
2012-09-24T19:59:26 < karlp> a few _more_ people now.
2012-09-24T20:02:22 -!- Irssi: ##stm32: Total of 60 nicks [1 ops, 0 halfops, 1 voices, 58 normal]
2012-09-24T20:02:38 < jpa-> yeah, we are way bigger than ##microcontrollers now
2012-09-24T20:02:54 < jpa-> and waaaay more random
2012-09-24T20:03:08 < karlp> s/random/awesome/g
2012-09-24T20:04:41 < jpa-> http://xob.kapsi.fi/~jpa/foo
2012-09-24T20:05:05 < Thorn> this is good kind of random, unlike ##electronics
2012-09-24T20:05:08 < Thorn> imo
2012-09-24T20:05:40 < jpa-> not all of it
2012-09-24T20:05:47 < jpa-> dongs can be quite annoying
2012-09-24T20:05:54 < jpa-> as can flyback
2012-09-24T20:05:55 <+Steffanx> Indeed, jpa-'s random is annoying
2012-09-24T20:06:20 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-202.pptp.stw-bonn.de] has joined ##stm32
2012-09-24T20:06:21 <+Steffanx> Can't deal with dongs jpa- ? He isn't THAT bad
2012-09-24T20:06:27 < emeb> as is the overuse of the word "awesome"
2012-09-24T20:06:34 < emeb> what is this, the 80's?
2012-09-24T20:06:38 < Thorn> dongs is the second best thing about this channel
2012-09-24T20:06:39 < karlp> s/awesome/awelots/g?
2012-09-24T20:06:47 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-202.pptp.stw-bonn.de] has quit [Client Quit]
2012-09-24T20:07:18 < emeb> shock and awelots?
2012-09-24T20:07:41 < jpa-> Steffanx: he isn't, but flyback i've had to ignore
2012-09-24T20:07:53 <+Steffanx> He didn't talk much lately
2012-09-24T20:08:01 < emeb> here's dongs next meal: http://kotaku.com/5945770/burger-king-is-launching-squid-ink-ketchup-and-a-black-hamburger
2012-09-24T20:09:21 <+Steffanx> emeb, randominator works very well
2012-09-24T20:09:31 <+Steffanx> *emeb's
2012-09-24T20:09:48 < jpa-> your autocomplete is also working awesomely
2012-09-24T20:09:54 < BrainDamage> I hate when flyback starts spamming and there's a legit convo, I try to avoid OT when there's channel relevant talk
2012-09-24T20:10:19 < jpa-> i hate how flyback drives interesting people away from the channel by being a dork
2012-09-24T20:10:30 <+Steffanx> That had nothing to do with my auto complete.. I failed mr jpa-
2012-09-24T20:10:40 <+Steffanx> And i'm VERY sorry for that
2012-09-24T20:10:50 < jpa-> yes you failed me
2012-09-24T20:11:10 < emeb> I've got a strange twitter feed - lots of odd stuff comes in that way.
2012-09-24T20:14:02 < zyp> BrainDamage, I have a z80 somewhere, if you come over I'll dig it out for you
2012-09-24T20:14:26 < Thorn> TI has 44 different CAN transceivers. wtf
2012-09-24T20:17:15 -!- mode/##stm32 [+o Steffanx] by ChanServ
2012-09-24T20:17:29 -!- Steffanx changed the topic of ##stm32 to: Prepare for the two minutes of /hate/love. This channel is publicly logged, ask zlog for more (no free z80s anymore). | Join us building the ##stm32 toolchain: https://github.com/EliasOenal/TNT (now with multilib) |  We have a wiki http://stm32.izua.ro | How engineers buy parts http://tim.cexx.org/?p=919
2012-09-24T20:17:45 -!- mode/##stm32 [-o Steffanx] by ChanServ
2012-09-24T20:19:39 < emeb> no more free z80s. sadness.
2012-09-24T20:20:39 < jpa-> wiki still sucks
2012-09-24T20:20:51 <+Steffanx> Yes, you didn't update it
2012-09-24T20:21:06 < Thorn> I wanted to ask for a wiki login btw
2012-09-24T20:21:35 < emeb> wait until izua is around?
2012-09-24T20:21:36 < Thorn> I've got a private wiki with lots of info, can copy some
2012-09-24T20:21:39 < jpa-> yeah, wiki login would help
2012-09-24T20:24:52 < emeb> you need this -> http://www.amazon.com/gp/product/B008M56YHS?ie=UTF8&camp=213733&creative=393185&creativeASIN=B008M56YHS&linkCode=shr&tag=codemonkeyorg-20
2012-09-24T20:26:17 < zyp> it was at $149 a few hours ago
2012-09-24T20:27:15 < emeb> maybe if you wait another few hours it'll be $129
2012-09-24T20:27:19 < zyp> :)
2012-09-24T20:28:35 < zyp> I'm not really interested though, I already have one of the 520-series
2012-09-24T20:28:39 < karlp> codemonkeyorg-20?
2012-09-24T20:28:43 < karlp> who'se getting the money there?
2012-09-24T20:30:00 < zyp> I expect SSDs to have dropped considerably in price the next time I need one, so it's kind of pointless to buy one now to fill a demand which doesn't exist yet
2012-09-24T20:33:14 < Thorn> are they usable for swap?
2012-09-24T20:33:52 < Thorn> (or pagefile.sys if you insist)
2012-09-24T20:34:04 < zyp> who needs swap?
2012-09-24T20:35:02 < BrainDamage> setting up hibernation to swapfile is annoying
2012-09-24T20:35:05 < Thorn> I do. try opening 50 tabs in firefox and 50 more in chrome. hard mode: with firebug activated in at least one tab
2012-09-24T20:35:05 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-24T20:35:40 < Bird|lappy> zyp, I may be crazy, but I often allocate 2-4x main RAM as swapspace
2012-09-24T20:36:10 < zyp> Bird|lappy, so I should just buy a whole ssd and dedicate it to swap then?
2012-09-24T20:36:49 < Bird|lappy> well, with the size of SSDs these days, you probably could just use one SSD for root/boot/paging
2012-09-24T20:37:24 < zyp> well, you're suggesting that I allocate 64-128GB to swap
2012-09-24T20:38:27 < karlp> last time I was looking at swap recommendations it was, 1-3x ram, but capped at 4gig.
2012-09-24T20:38:39 < karlp> so it had options when it ran out of ram, but not absurd options
2012-09-24T20:39:30 < zyp> ram is cheap, just buy more instead of arguing about how much swap one should have :p
2012-09-24T20:39:52 < emeb> ^ this
2012-09-24T20:40:29 < emeb> I find that if my machine starts swapping, things are going to get bad. But I run huge simulations.
2012-09-24T20:40:31 < TitanMKD> anyone know which could be the root cause of STM32F4 very very hot ?
2012-09-24T20:40:43 < zyp> TitanMKD, power dissipation :D
2012-09-24T20:40:53 < emeb> PLL = 336MHz?
2012-09-24T20:41:09 < TitanMKD> zyp it is not flashed just comes from factory
2012-09-24T20:41:15 < emeb> shorted GPIO?
2012-09-24T20:41:18 < TitanMKD> so i suspect it run at very low freq
2012-09-24T20:41:23 < TitanMKD> yes i suspect that too ;)
2012-09-24T20:41:38 < TitanMKD> or maybe +VCC on GND pin ...
2012-09-24T20:42:15 < emeb> someone needs to make cheap IR cameras for finding hotspots.
2012-09-24T20:42:21 -!- FridaJ [~FridaJ@217.200.185.247] has joined ##stm32
2012-09-24T20:42:36 < TitanMKD> I was thinking the TPS63031 was badly soldered and does short and i removed it
2012-09-24T20:42:51 < TitanMKD> and replaced by the other option a standard LM1117 (very easy to solder)
2012-09-24T20:43:03 < TitanMKD> and the MCU heat is horrible after 10s
2012-09-24T20:43:20 < emeb> this on your little buttpirate board?
2012-09-24T20:43:24 < TitanMKD> and 3.3V is 2.9V ;) so i really suspect a short ...
2012-09-24T20:43:28 < TitanMKD> emeb yes ;)
2012-09-24T20:43:46 < zyp> then you have a short.
2012-09-24T20:43:57 < TitanMKD> I think the MCU is probably dead too ;)
2012-09-24T20:44:12 < emeb> continuity checks on all adjacent pins usually works for me.
2012-09-24T20:44:25 < TitanMKD> that also explain why the TPS63031 output was 0V it was in security mode ;)
2012-09-24T20:44:49 < emeb> but I usually do that *before* applying power. :P
2012-09-24T20:45:06 < TitanMKD> me too i checked basicly GND & +VCC
2012-09-24T20:45:27 < TitanMKD> and between GND and +VCC there was 5KOhm like on a STM32F4 Discovery
2012-09-24T20:45:36 < emeb> probably wouldn't hurt to have a current limit bench supply set to about 100ma
2012-09-24T20:45:46 < TitanMKD> yes that wil help me
2012-09-24T20:46:06 < TitanMKD> anyway i failed to solder it with my hot plate too ;)
2012-09-24T20:46:07 < emeb> measured the power on my F4 board last night - ~93ma @ 5V input.
2012-09-24T20:46:18 < TitanMKD> too much paste with my stencil ...
2012-09-24T20:46:29 < TitanMKD> 5V input ??
2012-09-24T20:46:32 < emeb> the bigger the blob the better the job.
2012-09-24T20:46:38 < TitanMKD> on datasheet it is 3V to 3.6V max
2012-09-24T20:46:52 < emeb> main supply to my board is 5V - first thing is an LM1117-3.3 LDO
2012-09-24T20:47:02 <+Steffanx> :)
2012-09-24T20:47:03 < TitanMKD> ha ok you measured the main voltage for total current
2012-09-24T20:47:08 < emeb> yup
2012-09-24T20:47:20 < emeb> that includes a few LEDs running too
2012-09-24T20:47:23 < TitanMKD> anyway those LM1117 just suxx as they do not protect against short ;)
2012-09-24T20:47:48 < emeb> the solution is AVOID SHORTS!
2012-09-24T20:48:04 < TitanMKD> next time I will solder all the power and check all before to solder the MCU to be sure all is good ;)
2012-09-24T20:48:14  * emeb does that
2012-09-24T20:48:34 < TitanMKD> I was thinking it will work immediatly ;)
2012-09-24T20:48:40 -!- FridaJ [~FridaJ@217.200.185.247] has quit [Quit: Communi 1.1.3 for Symbian]
2012-09-24T20:48:43 < emeb> success-oriented planning.
2012-09-24T20:48:52 < TitanMKD> anyway I also suspect Itead PCB as the quality is the the best
2012-09-24T20:48:56 < emeb> you learn not to do that eventually.
2012-09-24T20:49:00 < Thorn> I normally make sure power is goos, then solder the MCU and JTAG, check it, then everything else
2012-09-24T20:49:01 < TitanMKD> is not the best
2012-09-24T20:49:16 < Thorn> *good
2012-09-24T20:49:25 < TitanMKD> yes but i was testing my SMT Stencil and I soldered all in one ;)
2012-09-24T20:49:30 < TitanMKD> and on 2 boards ;)
2012-09-24T20:49:46 < emeb> $ka-ching$
2012-09-24T20:49:57 -!- FridaJ [~FridaJ@217.200.185.247] has joined ##stm32
2012-09-24T20:50:12 < TitanMKD> I also suspect too much solder under resistor/capacitor and that does some short ...
2012-09-24T20:50:27 < TitanMKD> already see that on a TRF7970A board and that produce crazy problems
2012-09-24T20:51:01 < emeb> OK - time to post pix of your disgrace. :D
2012-09-24T20:51:45 < TitanMKD> hehe anyway the board is really clean ;)
2012-09-24T20:52:07 < TitanMKD> but the good part is lipo charging work fine ;)
2012-09-24T20:52:32 < emeb> uh, yay?
2012-09-24T20:52:50 -!- FridaJ [~FridaJ@217.200.185.247] has quit [Client Quit]
2012-09-24T20:53:02 < zyp> what about lipo exploding? does it work fine?
2012-09-24T20:53:34 < TitanMKD> lipo charging is independant from the other part ;)
2012-09-24T20:53:37 < emeb> what kind of bus-pirate was it? The exploding kind!
2012-09-24T20:54:30 < TitanMKD> hehe prototypes does not always work the first time, it is the challenge ;)
2012-09-24T20:54:54 < TitanMKD> i'm sure it is related only to soldering problems
2012-09-24T20:55:18 < jpa-> lipo's don't explode, they just burn fast :)
2012-09-24T20:55:55 -!- FridaJ [~FridaJ@217.200.185.247] has joined ##stm32
2012-09-24T20:58:06 < emeb> the usual reason my boards don't work at first is soldering on fine-pitch pins. Solution is more flux & reflow.
2012-09-24T20:59:31 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-09-24T20:59:35 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-24T21:07:53 < TitanMKD> emeb yes it was the error for me ;)
2012-09-24T21:08:07 < Thorn> STM32F1 is the first and (hopefully) buggiest attempt of [http://www.st.com/ ST Microelectronics] at a [[Cortex-M]] microcontroller line. Despite the bugginess, it turned out to be one of the most popular Cortex-M families thanks to good availability from the launch and aggressive pricing and advertisement, including the [[STM32VL-Discovery]] board aimed at hobbyists.
2012-09-24T21:08:16 < Thorn> ^^ good enough for wiki?
2012-09-24T21:08:24 < TitanMKD> i fixed lot of pin on STM32 as LQFP is very easy to fix but on QFN it is really harder
2012-09-24T21:09:54 < emeb> Thorn: looks good.
2012-09-24T21:10:08 < emeb> although, what bugs :P
2012-09-24T21:10:21  * emeb forgot irony tags again.
2012-09-24T21:11:39 <+dekar> Thorn, our wiki or wikipedia?
2012-09-24T21:11:45 < Thorn> our
2012-09-24T21:12:06 <+dekar> they'd [citation needed] that on wikipedia :P
2012-09-24T21:12:15 < Thorn> I'm fighting the urge to switch to fat trolling mode
2012-09-24T21:12:40 < Thorn> my wiki has that comic in the title image
2012-09-24T21:13:10 <+dekar> fat like in the filesystem?
2012-09-24T21:13:33 < Thorn> fat 32-bit trolling
2012-09-24T21:13:52 < emeb> Oh - I thought he meant this: http://www.daniele-tirinnanzi.com/slideshow.php?idfoto=11&gallery=characters
2012-09-24T21:13:53 < TitanMKD> ok let's check the photo
2012-09-24T21:13:55 <+dekar> oh, why wouldn't you like 32bit?
2012-09-24T21:14:03 < TitanMKD> you will see it look not so bad ;)
2012-09-24T21:14:48 < Thorn> one of the wikis that I edit anyway. http://wiki.zadraw.ch/
2012-09-24T21:15:25 -!- FridaJ [~FridaJ@217.200.185.247] has quit [Ping timeout: 268 seconds]
2012-09-24T21:15:39 <+dekar> :D
2012-09-24T21:15:47 -!- FridaJ [~FridaJ@217.200.185.212] has joined ##stm32
2012-09-24T21:15:48 -!- FridaJ [~FridaJ@217.200.185.212] has quit [Client Quit]
2012-09-24T21:15:54 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T21:15:54 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-24T21:15:54 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-24T21:16:08 < emeb> google translate FTW.
2012-09-24T21:16:37 < emeb> a godsend for us lazy-ass Americans who can't be bothered to learn other languages.
2012-09-24T21:18:08 <+dekar> Oh well, I decided learning english would be enough. So I guess Germans aren't any better :P
2012-09-24T21:18:12  * emeb is constantly awed by furriners who speak(type) better English than most folks here.
2012-09-24T21:18:31 -!- FridaJ [~FridaJ@217.200.185.210] has joined ##stm32
2012-09-24T21:18:40 <+dekar> I had Latin in school, but I barely remember a word
2012-09-24T21:19:09 < zyp> I had german in school, but I don't remember shit
2012-09-24T21:19:18 < TitanMKD> see the board
2012-09-24T21:19:19 < TitanMKD> http://postimage.org/image/pfwan6j1f/
2012-09-24T21:19:37 < TitanMKD> the most funny stuff is there's 2 Point on STM32F4 LQFP64
2012-09-24T21:19:42 < zyp> better than tom
2012-09-24T21:19:45 < emeb> I had some German and some Spanish. German because that's where my mom is from, Spanish because, hey, Mexico.
2012-09-24T21:19:47 < TitanMKD> but the point for pin 1 is the little ;)
2012-09-24T21:19:47 < BrainDamage> I had french, but I can only remember how to say "je ne parle pàs le française"
2012-09-24T21:20:22 < emeb> TitanMKD: that's not so bad.
2012-09-24T21:21:17 <+dekar> I really think English is enough, why bother learning any other languages?
2012-09-24T21:21:44 < zyp> because it's fun?
2012-09-24T21:21:51 < TitanMKD> emeb yes but i'm sure there's short behind some capacitor/components to produce short ...
2012-09-24T21:21:57 <+dekar> For some it might, and they're free to do so
2012-09-24T21:22:37 <+dekar> I never enjoyed learning languages of any kind
2012-09-24T21:22:40 < zyp> TitanMKD, sounds unlikely
2012-09-24T21:23:13 < zyp> I'd learn a lot of languages if I just had time
2012-09-24T21:23:25 < Thorn> test caps with a DMM in continuity mode. but why would the chip heat up then
2012-09-24T21:23:57 < zyp> I enjoy travelling, and I like being able to understand what people are saying
2012-09-24T21:24:21 < TitanMKD> Thorn yes why the chip heat, maybe there's a trace on a layer crossing with a short
2012-09-24T21:24:47 < TitanMKD> Thorn the boards are 100% eTest ;)
2012-09-24T21:24:59 < karlp> !define etest
2012-09-24T21:25:16 < TitanMKD> in etest they do not check via in fact
2012-09-24T21:25:32 < TitanMKD> because via are covered with silkscreen with no contact to test
2012-09-24T21:25:33 < Thorn> etest, n: an item in your pcb manufacturing bill
2012-09-24T21:26:43 < zyp> no point in checking the via itself, you just need to check that all pads in one net is connected, and that there is no connection between any nets
2012-09-24T21:27:09 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-24T21:27:09 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-24T21:27:09 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-24T21:27:09 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-24T21:28:00 <+dekar> izua, I'd say that beats the transparent headphone amp: http://www.youtube.com/watch?v=TLPWqCMb7DE
2012-09-24T21:28:49 < zyp> oh, I remember we played with stuff like that when I went to space tech school
2012-09-24T21:29:14 <+Steffanx> You went to space tech school?
2012-09-24T21:29:35 < zyp> surely I've said that?
2012-09-24T21:29:47 <+Steffanx> Thought you where EE
2012-09-24T21:29:52 <+Steffanx> *were
2012-09-24T21:30:00 < zyp> yes, in uni
2012-09-24T21:30:23 <+Steffanx> space tech school isn't "uni"?
2012-09-24T21:30:50 < zyp> no, last year of high school
2012-09-24T21:31:08 <+Steffanx> Weird system
2012-09-24T21:33:56 < emeb> Too bad he didn't measure the thrust.
2012-09-24T21:34:10 <+Steffanx> Yeah, for sure
2012-09-24T21:35:28 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-09-24T21:36:54 -!- FridaJ [~FridaJ@217.200.185.210] has quit [Quit: Communi 1.1.3 for Symbian]
2012-09-24T21:37:41 < emeb> lol apple: http://gizmodo.com/5945889/some-third+party-adapters-might-not-work-with-your-new-iphone
2012-09-24T21:38:29 <+Steffanx> Good thing :P
2012-09-24T21:38:44 <+Steffanx> I actually own a ipod touch and cheap chinese crap adapter
2012-09-24T21:39:01 < emeb> touch isn't bad.
2012-09-24T21:39:14 <+Steffanx> The adapter is terrible .. the ipod goes mad, non-functional touch screen etc :P
2012-09-24T21:39:23 <+Steffanx> So a ~anti-clone is good
2012-09-24T21:39:32  * emeb likes his Sammy Galaxy Player 5
2012-09-24T21:39:55 <+Steffanx> Of course the chinese will clone the chip
2012-09-24T21:40:04 < emeb> 'natch
2012-09-24T21:40:24 < emeb> in 3... 2... 1...
2012-09-24T21:41:13 <+Steffanx> I'm too afraid to open up the adapter too see what's inside
2012-09-24T21:41:35 <+Steffanx> The adapter is "Made in Califomia" :D :D :D
2012-09-24T21:42:03 < Bird|lappy> lol, that IS bad
2012-09-24T21:44:01 < emeb> fomy!
2012-09-24T21:45:42 -!- Bird|lappy is now known as Bird|test
2012-09-24T21:45:46 -!- zyp_ [zyp@zyp.im] has joined ##stm32
2012-09-24T21:45:58 -!- Bird|test is now known as Bird|lappy
2012-09-24T21:47:23 < zyp> http://www.vilbli.no/?Lan=3&Kurs=V.ELROM3----&Side=1.2
2012-09-24T21:47:23 < zyp> it's a bit changed since I went there though
2012-09-24T21:47:26 -!- Netsplit *.net <-> *.split quits: karlp, zyp
2012-09-24T21:47:34 <+Steffanx> bb
2012-09-24T21:47:53 < zyp_> heh
2012-09-24T21:47:56 -!- zyp_ is now known as zyp
2012-09-24T21:47:58 <+Steffanx> Whoa, we have real data on the wiki :D
2012-09-24T21:48:31 -!- Netsplit over, joins: karlp
2012-09-24T21:48:44 < zyp> http://www.narom.no/bilder/AVS_satellitt_elever_1000.jpg <- I digged up this, I'm one of the guys in the pic :p
2012-09-24T21:49:08  * Steffanx looks for beardman
2012-09-24T21:49:28 < zyp> good choice :p
2012-09-24T21:49:50 <+Steffanx> Yes, mr X right from the rocket
2012-09-24T21:50:19 < zyp> what do you mean?
2012-09-24T21:50:24 <+Steffanx> That's you
2012-09-24T21:50:52 < zyp> well, yea
2012-09-24T21:50:59 < zyp> but what about the X?
2012-09-24T21:51:19 <+Steffanx> Mr. X because he was unknown, until now
2012-09-24T21:51:23 < zyp> :p
2012-09-24T21:51:24 <+Steffanx> The application form said "Girls not allowed" uh? :P
2012-09-24T21:51:58 <+izua> "girls not recommended into engineering"
2012-09-24T21:52:01 < zyp> not really, but somehow we didn't have any that year :p
2012-09-24T21:52:38 < zyp> just a bunch of weird guys
2012-09-24T21:52:41 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-24T21:53:32 <+Steffanx> So how far did the rocket get up into the sky?
2012-09-24T21:53:40 < zyp> approx 10km or so
2012-09-24T21:53:44 <+Steffanx> Or did it try to go the otherway?
2012-09-24T21:54:53 -!- gsmcmull1n [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-24T21:54:55 < zyp> nah, it went nicely
2012-09-24T21:55:07 -!- barthess1 [~barthess@178.154.106.50] has joined ##stm32
2012-09-24T21:55:11 -!- mrnuke_ [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-24T21:55:34 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-24T21:56:38 -!- barthess [~barthess@178.154.106.50] has quit [Ping timeout: 246 seconds]
2012-09-24T21:56:38 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 246 seconds]
2012-09-24T21:56:39 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 246 seconds]
2012-09-24T21:56:39 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 246 seconds]
2012-09-24T21:56:39 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds]
2012-09-24T22:08:30 -!- mrcan_ [~mrcan@78.173.75.48] has joined ##stm32
2012-09-24T22:08:30 -!- mrcan_ [~mrcan@78.173.75.48] has quit [Changing host]
2012-09-24T22:08:30 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-24T22:09:27 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has joined ##stm32
2012-09-24T22:13:14 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-24T22:13:14 -!- DaKu is now known as daku
2012-09-24T22:13:16 -!- dongs_ [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-24T22:13:27 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 274 seconds]
2012-09-24T22:13:27 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has quit [Ping timeout: 274 seconds]
2012-09-24T22:13:27 -!- zlog_ [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 274 seconds]
2012-09-24T22:13:27 -!- claude [sbnc@h1682708.stratoserver.net] has quit [Ping timeout: 274 seconds]
2012-09-24T22:13:27 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 274 seconds]
2012-09-24T22:13:54 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has joined ##stm32
2012-09-24T22:14:30 -!- claude [sbnc@h1682708.stratoserver.net] has joined ##stm32
2012-09-24T22:24:53 -!- barthess1 is now known as barthess
2012-09-24T22:30:11 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 244 seconds]
2012-09-24T22:30:23 -!- enots [dimka@freelsd.net] has joined ##stm32
2012-09-24T22:43:19 -!- nashpa_ [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-24T22:43:41 -!- mrcan__ [~mrcan@78.173.75.48] has joined ##stm32
2012-09-24T22:44:37 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 260 seconds]
2012-09-24T22:45:14 -!- nashpa_ is now known as nashpa
2012-09-24T22:46:13 -!- mrcan__ [~mrcan@78.173.75.48] has quit [Changing host]
2012-09-24T22:46:13 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-24T22:48:39 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 245 seconds]
2012-09-24T22:50:34 -!- daku is now known as DaKu
2012-09-24T23:01:08 <+Steffanx> izua
2012-09-24T23:01:20 <+izua> Steffanx:
2012-09-24T23:01:27 <+Steffanx> You NEVER guess what happend
2012-09-24T23:01:30 <+Steffanx> ned
2012-09-24T23:01:36 <+izua> ned.
2012-09-24T23:01:38 <+izua> winter is coming.
2012-09-24T23:01:48 <+Steffanx> No, not GoT
2012-09-24T23:02:06 <+Steffanx>  /topic and click on the wiki-link
2012-09-24T23:02:13 <+izua> but.. that could have happened
2012-09-24T23:02:27 <+izua> we got hijacked and spammed
2012-09-24T23:02:40 <+izua> ehmagerds
2012-09-24T23:03:04 <+izua> and there's even a working link
2012-09-24T23:03:14 <+Steffanx> ehmagerds .. google knows that wordt, but doesn't know what it means
2012-09-24T23:03:18 <+Steffanx> -t
2012-09-24T23:04:39 <+izua> it's an 'oh my god' sort of exclamation
2012-09-24T23:05:39 < emeb> http://knowyourmeme.com/memes/ermahgerd
2012-09-24T23:05:56 < emeb> spell it right, y'all.
2012-09-24T23:06:01 <+Steffanx> emeb he didn't write it right.. and i don't care about memes
2012-09-24T23:06:02 <+izua> whoa
2012-09-24T23:06:06 <+izua> there are several people that made edits
2012-09-24T23:06:28 <+Steffanx> dekar I demand you to write something about your toolchain.. NOW
2012-09-24T23:07:07 < emeb> Steffanx: you know enough about memes to realize he spelled it rong. :)
2012-09-24T23:07:57 <+Steffanx> Not really. I clicked on your link
2012-09-24T23:08:14 < emeb> oh - my bad.
2012-09-24T23:10:37 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-24T23:10:38 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-24T23:10:38 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-24T23:10:41 < emeb> lol reddit - using facial recognition to track down the ermahgerd girl.
2012-09-24T23:10:48 < Laurenceb_> my bluetooth flash code is on github if anyone interested :P
2012-09-24T23:11:03 < Laurenceb_> obvious virgins
2012-09-24T23:11:14 <+Steffanx> linky link?
2012-09-24T23:11:14 < emeb> HAH!
2012-09-24T23:11:18 < Laurenceb_> - it uses an rn-41
2012-09-24T23:11:29 < emeb> Steffanx: scroll down on the link I gave above.
2012-09-24T23:11:42 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/jtag/stm32loader.py
2012-09-24T23:12:02 <+Steffanx> Laurenceb_ dongs will never forgive you
2012-09-24T23:12:09 < Laurenceb_> lol
2012-09-24T23:12:14 <+Steffanx> time.sleep(0.1) :D :D
2012-09-24T23:12:34 < emeb> dongs doesn't like python?
2012-09-24T23:12:43 <+Steffanx> He says he doesn't
2012-09-24T23:12:45 < Laurenceb_> indeed
2012-09-24T23:12:46 <+Steffanx> like it
2012-09-24T23:12:52 < Laurenceb_> thats the only way to make it work
2012-09-24T23:12:58 < Laurenceb_> rn-41 is laggy
2012-09-24T23:13:33 < Laurenceb_> and has internal buffer issues
2012-09-24T23:13:49 <+Steffanx> It doesn't reply with a "command acknowledged" message?
2012-09-24T23:14:06 < Laurenceb_> yes it does
2012-09-24T23:14:22 < Laurenceb_> but im too lazy to do it like that
2012-09-24T23:14:25 < Laurenceb_> maybe tomorrow
2012-09-24T23:14:26 <+Steffanx> Aaah
2012-09-24T23:14:35 < prog> why Python
2012-09-24T23:14:42 < Laurenceb_> a "send command" function
2012-09-24T23:14:43 < prog> that's intellectual terrorism
2012-09-24T23:14:52 < Laurenceb_> that loops for ok reply
2012-09-24T23:14:54 < Laurenceb_> lol prog
2012-09-24T23:14:55 < zyp> prog, because that's what he had a source to cut and paste from
2012-09-24T23:15:01 <+Steffanx> Doesn't matter in his case prog :P
2012-09-24T23:15:06 < Laurenceb_> exactly
2012-09-24T23:15:08 < emeb> burrrn!
2012-09-24T23:15:40 <+Steffanx> The copyright message is still there, so .. it's ok :)
2012-09-24T23:16:40 < prog> a good advice: keep away from the GPL as much as possible
2012-09-24T23:16:51 <+Steffanx> Too late
2012-09-24T23:18:03 < emeb> anyone ever used those HC-05 BT modules?
2012-09-24T23:18:21 < emeb> got one from DX a while back - haven't done anything with it yet.
2012-09-24T23:18:25 < Laurenceb_> the rn-41 timing issues are actually useful
2012-09-24T23:18:40 < Laurenceb_> as if there is "$$$" in the firmware it wont enter command mode
2012-09-24T23:18:42 <+Steffanx> I have some modules similar to that one, not sure if they are "HC-05 BT" emeb
2012-09-24T23:18:48 <+Steffanx> Never used them though :P
2012-09-24T23:19:41 <+Steffanx> I still have to send my pcb design to seeed and finish my reflow oven with usb and bluetooth :D
2012-09-24T23:19:43 < emeb> All based on the CSR chip. Different firmware loads though, enable different I/O
2012-09-24T23:19:54 < Laurenceb_> the same trick should work
2012-09-24T23:20:05 < Laurenceb_> i origionally tried RTS/CTS
2012-09-24T23:20:14 < Laurenceb_> but the lines do odd things during connection
2012-09-24T23:20:35 < Laurenceb_> had to use gpio mode
2012-09-24T23:22:45 < Laurenceb_> can you set pyserial timeout after connection is made?
2012-09-24T23:22:59 <+Steffanx> Try it?
2012-09-24T23:23:15 < Laurenceb_> or bother to read the docs...
2012-09-24T23:23:57 < Laurenceb_> aha
2012-09-24T23:24:00 < Laurenceb_>  .timeout
2012-09-24T23:24:02 < Laurenceb_> simple
2012-09-24T23:24:31 < Laurenceb_> so ill set timeout to ~0.1 seconds, write, flush, and readline and loop until its "OK"
2012-09-24T23:24:43 < Laurenceb_> then set timeout back to 5seconds for flashing
2012-09-24T23:24:44 <+Steffanx> Thorn doesn't have more fancy "Selection Guides"? They're nice
2012-09-24T23:25:53 < Thorn> I'll probably do more for f4 and maybe others too
2012-09-24T23:55:02 -!- Netsplit *.net <-> *.split quits: Nutter`, mrnuke_, zyp, claude, enots, BrainDamage, oPossum5150, esden, zippe, grummund,  (+52 more, use /NETSPLIT to show all of them)
--- Day changed Tue Sep 25 2012
2012-09-25T00:04:27 -!- Tectu_ [tectu@kunsmann.eu] has joined ##stm32
2012-09-25T00:04:27 -!- oPossum430 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-25T00:04:28 -!- Netsplit over, joins: @ChanServ, [7], phantoxeD, mrcan__, nashpa, enots, claude, BrainDamage, dongs_, zlog (+48 more)
2012-09-25T00:08:33 -!- BjoernC [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has joined ##stm32
2012-09-25T00:08:35 -!- FridaJ [~FridaJ@151.58.22.212] has joined ##stm32
2012-09-25T00:09:09 < Titan___> hehe gret
2012-09-25T00:09:09 < Titan___> i tested the 2nd board 3.3V and i obtain 44mA
2012-09-25T00:09:09 < Titan___> STM32F4 defaut factory comsumption 44mA includin a power LED
2012-09-25T00:11:27 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-25T00:15:28 -!- prog` [prog@bornbrilliant.org] has joined ##stm32
2012-09-25T00:16:27 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 256 seconds]
2012-09-25T00:17:20 -!- prog [prog@bornbrilliant.org] has quit [Ping timeout: 240 seconds]
2012-09-25T00:17:20 -!- DaKu [DaKu@dakus.dk] has quit [Ping timeout: 240 seconds]
2012-09-25T00:17:48 -!- DaKu [DaKu@dakus.dk] has joined ##stm32
2012-09-25T00:24:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 248 seconds]
2012-09-25T00:24:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-25T00:24:36 -!- dongs_ [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds]
2012-09-25T00:24:45 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-25T00:25:08 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 248 seconds]
2012-09-25T00:25:36 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-25T00:33:14 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 240 seconds]
2012-09-25T00:33:38 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-25T00:33:38 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-25T00:33:38 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-09-25T00:33:55 < Laurenceb_> Titan___: sounds about right
2012-09-25T00:34:02 -!- lawrenceseattle [~anonymous@66.192.187.106] has joined ##stm32
2012-09-25T00:34:04 < Laurenceb_> 168Mhz?
2012-09-25T00:34:47 < Mazingaro> hi
2012-09-25T00:35:40 < Mazingaro> please I'm using ddd to debug an olimex stm32-h107, but I cannot show all cpu registers in the "DDD: Registers" window
2012-09-25T00:36:50 < emeb> hmph - never much cared about the Registers window.
2012-09-25T00:38:39 < emeb> what are you using to connect to your target?
2012-09-25T00:41:55 < Mazingaro> lol
2012-09-25T00:43:59 < emeb> Mazingaro: which registers are you missing?
2012-09-25T00:44:31 -!- lawrenceseattle_ [~anonymous@66.192.187.106] has joined ##stm32
2012-09-25T00:45:10 < Mazingaro> psp
2012-09-25T00:45:48 < Mazingaro> primask
2012-09-25T00:45:53 < emeb> I'm seeing psp
2012-09-25T00:45:57 < Mazingaro> I can access them with monitor reg
2012-09-25T00:46:11 < emeb> which gdb target are you using?
2012-09-25T00:46:24 < Mazingaro> arm-none-eabi-gdb
2012-09-25T00:46:34 < Mazingaro> ddd --debugger arm-none-eabi-gdb
2012-09-25T00:46:38 < emeb> yeah - but how do you connect to the hardware??
2012-09-25T00:46:43 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-09-25T00:46:54 < Mazingaro> openocd + olimex arm-usb-ocd-h
2012-09-25T00:46:59 < emeb> ah
2012-09-25T00:47:13 < emeb> no experience w/ openocd on Cortex-Mx
2012-09-25T00:47:32 < emeb> their gdb server must not handle all the regs.
2012-09-25T00:47:40 < emeb> I'm using BMP.
2012-09-25T00:49:11 -!- lawrenceseattle [~anonymous@66.192.187.106] has quit [Ping timeout: 240 seconds]
2012-09-25T00:49:11 -!- lawrenceseattle_ is now known as lawrenceseattle
2012-09-25T00:49:30 < Mazingaro> but I can see registers with the command "monitor reg"
2012-09-25T00:50:32 -!- piele [~Unknown@bakzeil.creativeserver.net] has quit [Ping timeout: 248 seconds]
2012-09-25T00:50:32 -!- gsmcmull1n [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 248 seconds]
2012-09-25T00:53:04 -!- piele [~Unknown@bakzeil.creativeserver.net] has joined ##stm32
2012-09-25T00:54:15 < emeb> what rev of DDD you using?
2012-09-25T00:54:20 < emeb> I've got 3.3.12
2012-09-25T00:56:36 < Mazingaro> same here x86_64-pc-linux-gnu
2012-09-25T00:58:53 < emeb> what rev of gdb? I've got 7.3.1.20120613-cvs
2012-09-25T00:59:57 < emeb> are you doing "target remote <port>" or "target extended-remote <port>"?
2012-09-25T01:00:15 < Mazingaro> the first
2012-09-25T01:02:32 < Mazingaro> but target extended-remote:3333 is the same
2012-09-25T01:05:02 -!- mrcan__ is now known as mrcan
2012-09-25T01:06:25 < Mazingaro> now the second
2012-09-25T01:06:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-25T01:07:01 < Mazingaro> GNU gdb (Sourcery CodeBench Lite 2012.03-56) 7.2.50.20100908-cvs
2012-09-25T01:07:14 -!- BjoernC_ [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has joined ##stm32
2012-09-25T01:07:23 < emeb> looks like your gdb is older than mine.
2012-09-25T01:07:35 < emeb> no idea if that's the problem tho.
2012-09-25T01:08:37 < emeb> I'm using the launchpad toolchain: https://launchpad.net/gcc-arm-embedded
2012-09-25T01:08:37 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-25T01:09:16 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-25T01:09:16 -!- BjoernC [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has quit [Ping timeout: 246 seconds]
2012-09-25T01:09:16 -!- Tectu_ [tectu@kunsmann.eu] has quit [Ping timeout: 246 seconds]
2012-09-25T01:09:54 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-25T01:10:00 < emeb> http://www.yagarto.de/
2012-09-25T01:10:06 < Mazingaro> is that better?
2012-09-25T01:10:15 < emeb> they've got even newer toolchains there (4.7.1)
2012-09-25T01:10:21 -!- BjoernC_ [~BjoernC@dyndsl-178-142-052-111.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-25T01:11:09 -!- Tectu [tectu@kunsmann.eu] has joined ##stm32
2012-09-25T01:11:32 < emeb> but it depends on your OS - Yagarto is Mac / Win only.
2012-09-25T01:11:53 < Mazingaro> I'm running wheezy
2012-09-25T01:11:58 < emeb> haven't seen binaries of 4.7.x for Linux. Think you have to build your own.
2012-09-25T01:12:25 -!- R0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-25T01:12:25 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-25T01:12:33 < Laurenceb_> https://github.com/Laurenceb/STM32-Logger/blob/master/jtag/stm32loader.py
2012-09-25T01:12:37 < Laurenceb_> i made it nice :P
2012-09-25T01:13:32 < emeb> dongs'll still hate on it. :)
2012-09-25T01:13:35 < Laurenceb_> heh
2012-09-25T01:14:10 < Laurenceb_> apparently it will work on the "firefly" bluetooth modules too
2012-09-25T01:17:12 < Mazingaro> ...
2012-09-25T01:18:33 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 245 seconds]
2012-09-25T01:18:55 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-09-25T01:19:20 < Mazingaro> I'm trying to load in r12 the address of top of a stack using ldr r12, =pOldTCB, but as the instruction it's executed it jams in HardFault_Handler... Should I align the pOldTCB value to 8 bytes? And how? __attribute__ ((aligned(8), __packed__ )) ? tx
2012-09-25T01:24:42 -!- Laurenceb_ [~Laurence@host86-179-248-143.range86-179.btcentralplus.com] has quit [Ping timeout: 264 seconds]
2012-09-25T01:30:41 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-09-25T01:31:02 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-25T01:32:53 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-09-25T01:33:45 < TitanMKD> re
2012-09-25T01:33:51 < TitanMKD> grr i have found my error !!!
2012-09-25T01:34:00 < TitanMKD> it is why i fried 2 STM32 grr
2012-09-25T01:34:22 < TitanMKD> I have put VCAP_1 and VCAP_2 to 3.3V !!
2012-09-25T01:34:36 < TitanMKD> instead of GND
2012-09-25T01:45:21 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has joined ##stm32
2012-09-25T01:45:22 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-237-37.clienti.tiscali.it] has quit [Disconnected by services]
2012-09-25T01:56:36 < gnomad> doh!
2012-09-25T02:07:25 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-09-25T02:15:31 < dongs> Tectu: lol
2012-09-25T02:15:33 < dongs> err
2012-09-25T02:15:36 < dongs> ugh he /quit
2012-09-25T02:18:26 -!- DaKu is now known as daku
2012-09-25T02:21:25 -!- R0b0t1_ [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-25T02:21:43 -!- R0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-09-25T02:21:43 -!- piele [~Unknown@bakzeil.creativeserver.net] has quit [Ping timeout: 248 seconds]
2012-09-25T02:21:44 -!- piele [~Unknown@bakzeil.creativeserver.net] has joined ##stm32
2012-09-25T02:30:40 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 272 seconds]
2012-09-25T02:31:24 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has joined ##stm32
2012-09-25T02:31:24 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has quit [Changing host]
2012-09-25T02:31:24 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32
2012-09-25T02:36:55 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 394 seconds]
2012-09-25T02:37:01 -!- enots_ [dimka@194.190.195.85] has joined ##stm32
2012-09-25T02:40:24 -!- FridaJ [~FridaJ@151.58.22.212] has left ##stm32 []
2012-09-25T03:15:57 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 246 seconds]
2012-09-25T03:18:23 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-25T03:18:32 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has joined ##stm32
2012-09-25T03:46:18 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 264 seconds]
2012-09-25T03:50:30 -!- oPossum430 is now known as oPossum5150
2012-09-25T04:35:25 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-25T04:46:16 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-25T04:56:34 -!- lawrenceseattle [~anonymous@66.192.187.106] has quit [Quit: lawrenceseattle]
2012-09-25T04:59:30 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-25T05:35:56 -!- Nutter` is now known as Nutter
2012-09-25T05:46:31 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-25T05:46:32 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-25T05:46:32 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-25T06:05:24 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-09-25T06:11:14 < dongs> total lack of blogging today
2012-09-25T06:19:33 < zippe> Why would we share our trade secrets with you?
2012-09-25T06:27:47 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-25T06:27:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-25T06:35:58 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Read error: Operation timed out]
2012-09-25T06:39:24 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-25T06:44:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-25T06:45:00 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has joined ##stm32
2012-09-25T06:45:11 -!- R0b0t1_ is now known as iR0b0t1
2012-09-25T06:45:14 -!- iR0b0t1 [~dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-25T06:45:14 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-25T06:46:31 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-25T06:48:08 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-25T06:53:55 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-25T06:54:26 < emeb_mac> hi
2012-09-25T06:56:51 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-09-25T06:59:55 -!- GargantuaSauce [~sauce@blk-252-42-151.eastlink.ca] has quit [Ping timeout: 246 seconds]
2012-09-25T07:21:30 < dongs> zippe: lol.
2012-09-25T07:29:57 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 256 seconds]
2012-09-25T07:37:40 < dongs> how come no site mentioned what magnetometer is in itrash5
2012-09-25T07:41:47 < cjbaird> Bodge up an ip5-controlled quadcopter, get on all the blogz!
2012-09-25T07:58:15 < cjbaird> Woo! The Living God of Manchildren is coming! :D .. http://tech.slashdot.org/story/12/09/25/0044255/woz-applying-for-australian-citizenship-because-of-the-nbn
2012-09-25T08:06:12 < Thorn> cjbaird: that guy, svo, whose font you're using. I remembered him
2012-09-25T08:06:45 < Thorn> about 8 years ago or so he used to have a webside with lots of nude pictures of his underage girlfriend
2012-09-25T08:08:34 < Thorn> his nickname was vaguely familiar but the word 'webam' made it certain
2012-09-25T08:30:20 -!- claude is now known as Claude
2012-09-25T08:36:42 -!- NonaSuomy is now known as TeknoJuce
2012-09-25T08:49:54 < dongs> laurenceb probably has whole thing mirrored
2012-09-25T10:08:26 -!- Claude is now known as claude
2012-09-25T10:08:34 -!- enots_ is now known as enots
2012-09-25T10:12:28 < cjbaird> torrent plz
2012-09-25T10:14:24 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-25T10:22:06 < dongs> float val =  A / (B / 1000); val = A * 1000 / B
2012-09-25T10:22:09 < dongs> is this equivalent?
2012-09-25T10:22:41 < zyp> yes
2012-09-25T10:22:44 < dongs> why?
2012-09-25T10:22:51 < dongs> im a bit dumb w/math
2012-09-25T10:23:48 < dongs> wolframalpha to teh rescue
2012-09-25T10:23:56 < zyp> because A / (B / 1000) == (A * 1000) / (1000 * B / 1000) == A * 1000 / B
2012-09-25T10:23:57 < Thorn> overflows/underflows could occur with large/small A/B depending on how it's compiled
2012-09-25T10:24:04 < zyp> yes
2012-09-25T10:24:32 < zyp> it might not be equivalent in terms of overflow and rounding
2012-09-25T10:25:02 < zyp> but in math the expressions are equivalent
2012-09-25T10:26:33 < dongs> k
2012-09-25T10:35:33 < cjbaird> 1000x = 8 * 125 * x = x << 3 + (x << 7) - x - x -x, too..
2012-09-25T10:35:58 < cjbaird> or something like that..
2012-09-25T10:36:56 < cjbaird> I did a trick like that years ago to avoid overflow on a 68hc11..
2012-09-25T10:46:47 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-25T10:47:52 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-25T10:49:13 -!- claude is now known as claude_work
2012-09-25T11:05:16 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-09-25T11:14:09 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has left ##stm32 ["Konversation terminated!"]
2012-09-25T11:14:44 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-25T11:36:22 < dongs> heh
2012-09-25T11:38:57 < dongs> http://www.youtube.com/watch?v=Zg6iMDfOl9E
2012-09-25T11:48:48 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-09-25T11:58:14 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-25T11:58:15 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-25T11:58:15 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-25T12:01:29 -!- Netsplit *.net <-> *.split quits: ossifrage, esden
2012-09-25T12:01:34 -!- Netsplit *.net <-> *.split quits: qyx_
2012-09-25T12:01:38 -!- qyx [~qyx@krtko.org] has joined ##stm32
2012-09-25T12:01:52 -!- Netsplit over, joins: ossifrage
2012-09-25T12:02:52 -!- Netsplit over, joins: esden
2012-09-25T12:08:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-25T12:08:47 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-25T12:17:10 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-25T12:24:25 -!- ntfreak_ is now known as ntfreak
2012-09-25T12:28:07 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-09-25T12:47:33 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-25T12:50:26 -!- cjbaird_ [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-25T12:52:45 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has joined ##stm32
2012-09-25T12:58:37 <+dekar> "When you have a fire in an aircraft, there's no place to go, exactly, there's no – and you can't find any oxygen from outside the aircraft to get in the aircraft, because the windows don't open. I don't know why they don't do that. It's a real problem. So it's very dangerous." -Mitt Romney
2012-09-25T12:59:09 < dongs> who
2012-09-25T12:59:24 <+Steffanx> what
2012-09-25T13:01:10 < cjbaird_> What's Dan Quayle doing these days?
2012-09-25T13:01:11 <+Steffanx> You should ignore this american election bull crap dekar
2012-09-25T13:01:25 <+Steffanx> The return of cjbaird_ ?
2012-09-25T13:05:49 <+Steffanx> Too bad your buddy flyback left us, cjbaird_
2012-09-25T13:05:50 <+dekar> Steffanx, I think it's quite entertaining tbh
2012-09-25T13:07:09 <+Steffanx> True, but only if it's not misinterpreted or totally out of context
2012-09-25T13:07:23 <+Steffanx> It's probably both, like always
2012-09-25T13:07:55 <+dekar> Steffanx, it's not, he is talking about how his wife was in a plane that had a fire in the passenger compartment
2012-09-25T13:08:10 <+Steffanx> You've seen the video?
2012-09-25T13:08:54 <+dekar> nope, but the longer quote on some newssites
2012-09-25T13:10:12 < cjbaird_> Got one! http://scootercommunity.com.au/forums/p/18125/140194.aspx#140200
2012-09-25T13:10:17 -!- cjbaird_ is now known as cjbaird
2012-09-25T13:11:28 <+Steffanx> cjbaird, one what?
2012-09-25T13:11:58 <+dekar> Steffanx, is there a video? link?
2012-09-25T13:12:07 < dongs> http://petegailey.smugmug.com/Motorcycles/Scooter-Expo-2012/i-86BJFTV/0/L/DSC0344-L.jpg
2012-09-25T13:12:10 < dongs> whatsa going on here
2012-09-25T13:12:20 <+Steffanx> I didn't see a video dekar
2012-09-25T13:12:21 < dongs> http://petegailey.smugmug.com/Motorcycles/Scooter-Expo-2012/i-j5nHTN5/0/X2/P1040870-X2.jpg this is laurenceb
2012-09-25T13:12:50 <+Steffanx> Impossible
2012-09-25T13:13:18 <+dekar> Steffanx, why did you ask me about a video then?
2012-09-25T13:14:01 <+Steffanx> Nevermind .. already cared way too much about this american bullcrap :P
2012-09-25T13:16:10 < cjbaird> DSC0344: stunt rider from glorious Nippon who showed up for the day. Strangely enough, the guy who took the pics had his licence suspended a month ago.. Still pondering how2troll him, or better, the event organizers, who let him do test rides.
2012-09-25T13:17:08 <+Steffanx> That's not trolling, that's just silly. Mr cjbaird
2012-09-25T13:21:43 <+Steffanx> It's ok to troll the organizer though :P
2012-09-25T13:21:50 <+Steffanx> *event organizers
2012-09-25T13:23:46  * Laurenceb has been doing something much worse
2012-09-25T13:23:53 < Laurenceb> ive been coding in python
2012-09-25T13:25:27 <+Steffanx> We already discussed that yesterday :)
2012-09-25T13:32:14 < karlp> what's wrong with python?
2012-09-25T13:32:26 < karlp> I'm coding python right now.
2012-09-25T13:32:35 < karlp> (well, actually deploying python things)
2012-09-25T13:33:57 < dongs> fuck python and git.
2012-09-25T13:37:11 < cjbaird> python >>>>>>> ruby, at least.
2012-09-25T13:37:50 < dongs> a flying turd is >>>> ruby
2012-09-25T13:37:50 < dongs> so yeah
2012-09-25T13:41:37 <+dekar> what's wrong with ruby?
2012-09-25T13:42:12 < karlp> syntax
2012-09-25T13:42:56 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-09-25T13:42:56 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-09-25T13:42:56 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-25T13:44:00 < cjbaird> The ridiculous thing about ruby to me is that the guy was trying to make a procedural LISP, thinking it would automatically inherit all the LISPAWESOME..
2012-09-25T13:44:30 < cjbaird> So it really belongs in the LOGO familay of languages..
2012-09-25T13:44:57 < Laurenceb> the google "people" thing is funny
2012-09-25T13:45:05 < Laurenceb> especially if you type in lisp
2012-09-25T13:45:19 < Laurenceb> column of bearded gurus
2012-09-25T13:45:50 < cjbaird> ..and money-scabbing faggots who're trying to suck-up to Paul Graham..
2012-09-25T13:46:05 <+dekar> I only use C/C++ these days and I find it really hard to motivate me to learn new languages, especially interpreted ones
2012-09-25T13:46:50 < zyp> dekar, then you're not lazy enough
2012-09-25T13:47:12 <+dekar> I do like Java, but it usually doesn't do the job for me
2012-09-25T13:47:25 < zyp> I use python because it lets me accomplish more with fewer lines of code
2012-09-25T13:47:59 <+dekar> Does it have easy access to native APIs?
2012-09-25T13:48:16 <+dekar> I find it really annoying to write C wrappers for use with JNI/Java
2012-09-25T13:48:22 < dongs> lol, native apis
2012-09-25T13:48:28 < zyp> dekar, yes
2012-09-25T13:49:55 < Laurenceb> native aids
2012-09-25T13:50:25 < cjbaird> I've blinded myself with a 25 LED board I was testing the current draw of.. :/
2012-09-25T13:50:41 < zyp> ctypes makes shared libraries almost no-effort to use: http://docs.python.org/library/ctypes.html
2012-09-25T13:50:41 < Laurenceb> blinkenlights
2012-09-25T13:50:50 < Laurenceb> you need an arduino for that
2012-09-25T13:52:09 < karlp> there's also pyrex.... http://www.cosc.canterbury.ac.nz/greg.ewing/python/Pyrex/
2012-09-25T13:52:25 < zyp> karlp, and a bunch of other libraries
2012-09-25T13:52:34 < cjbaird> I have an old project for a manchester-encoded optical link. Was wondering about how fast this LED pseudo-globe could go.
2012-09-25T13:52:35 < karlp> cython too.
2012-09-25T13:52:35 < zyp> I've used boost.python to wrap C++ libraries before
2012-09-25T13:54:39 < Laurenceb> http://arduino.cc/forum/index.php/topic,107198.msg805223.html#msg805223
2012-09-25T13:54:47 < Laurenceb> they did something useful?!
2012-09-25T13:54:49 < Laurenceb> never
2012-09-25T13:55:31 < karlp> is that gnomad from this channel?
2012-09-25T13:56:04 <+Steffanx> Maybe it's time to stop your random bashing Laurenceb ?
2012-09-25T13:56:20 < cjbaird> My scooter does the same thing with a $2 thermocouple.
2012-09-25T13:58:31 <+Steffanx> karlp, according to gnomads ip he is in the US ( or at least his pc/server is )
2012-09-25T13:59:38 < dongs> haha
2012-09-25T13:59:43 < dongs> they kept the fucktarded spacing in due
2012-09-25T13:59:50 < dongs> awesome.
2012-09-25T14:00:51 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-25T14:00:52 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-25T14:00:52 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-25T14:00:55 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-25T14:00:59 < dongs> how many pins are wasted
2012-09-25T14:01:29 <+izua> dongs: if you don't like pin wastage
2012-09-25T14:01:41 <+izua> you should get one of those PICs in 3 pin SOT23
2012-09-25T14:01:55 <+Steffanx> Atmel also released a new c-m4. it uses less µA/MHz than the STM32L1 …
2012-09-25T14:02:14 < dongs> but does it still use same retarded AVR 10bit ADCs?
2012-09-25T14:02:23 <+Steffanx> I've no idea
2012-09-25T14:03:15 < dongs> > ARM® CortexTM-M4 running at up to 48 MHz
2012-09-25T14:03:16 < dongs> lol.
2012-09-25T14:03:21 <+Steffanx> And the AVR ADCs work pretty well for the things i used them for :)
2012-09-25T14:03:41 < dongs> > High precision 4/8/12 MHz factory trimmed internal RC oscillator
2012-09-25T14:03:41 < dongs> lol
2012-09-25T14:04:17 < dongs> > One 16-channel ADC 300Ksps (ADC) with up to 12 Bits Resolution
2012-09-25T14:04:50 <+Steffanx> No, not very fast :P
2012-09-25T14:04:52 < cjbaird> That reminds me, the backorders for the KL25 and Launchpad Stellaris should be shipped soon..
2012-09-25T14:04:52 -!- izua [~izua@unaffiliated/izua] has quit [Client Quit]
2012-09-25T14:05:00 < dongs> wut
2012-09-25T14:05:06 < dongs> i got spam that stellaris wasnt shipping until end of october
2012-09-25T14:06:46 < cjbaird> I got a spam from E14 saying something was shipping.. but the order code was for a product that'd been deleted. :/
2012-09-25T14:06:50 < Laurenceb> wow that sucks
2012-09-25T14:07:01 <+Steffanx> Another gadget for the 'nice to have but won't use' box, dongs?
2012-09-25T14:07:11 < Laurenceb> http://i.imgur.com/slWoJ.jpg
2012-09-25T14:07:14 < Laurenceb> teehee
2012-09-25T14:07:18 < dongs> Est. Ship Date 11/09/2012
2012-09-25T14:07:19 < dongs> haha
2012-09-25T14:07:21 < dongs> and thats  M/D/Y
2012-09-25T14:07:31 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-25T14:07:31 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-25T14:07:31 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-25T14:07:34 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-25T14:07:57 < dongs> awesome RF bro
2012-09-25T14:08:12 <+Steffanx> Any background info/context Laurenceb ?
2012-09-25T14:08:25 < dongs> looks like 800mW? amp for his bluetooth trolling
2012-09-25T14:08:36 < Laurenceb> from #highaltitude
2012-09-25T14:08:53 < Laurenceb> hmm 800mw bluetooth troll
2012-09-25T14:08:56 < Laurenceb> sounds like a plan
2012-09-25T14:08:58 < dongs> high trollitude
2012-09-25T14:09:01 < dongs> more liek
2012-09-25T14:09:24 < dongs> is that a tarduino
2012-09-25T14:09:31 < dongs> the pin spacing gives it away
2012-09-25T14:09:41 <+Steffanx> Dongs is just jealous. He doesn't have the RF/analog skillz
2012-09-25T14:09:50 < dongs> Steffanx: i have no skillz whatsoever
2012-09-25T14:11:29 < cjbaird> http://i.imgur.com/7FSPO.png
2012-09-25T14:11:54 <+Steffanx> heh
2012-09-25T14:11:54 < dongs> what is taht
2012-09-25T14:11:56 < dongs> stellaris?
2012-09-25T14:12:01 <+izua> cjbaird: awesome
2012-09-25T14:13:15 < cjbaird> The best I can figure out, it was something I flagged back in July.
2012-09-25T14:16:26 < zyp> dongs, that's unfair, mine doesn't ship until 11/16/2012 :(
2012-09-25T14:16:48 < dongs> aww
2012-09-25T14:16:49 < Laurenceb> http://www.fat-pie.com/dogofman.htm
2012-09-25T14:17:39 < cjbaird> Aha, it's the old KL25 order code.
2012-09-25T14:17:45 < dongs> KL25 is what
2012-09-25T14:17:48 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-25T14:17:49 < Laurenceb> http://www.flickr.com/photos/62825935@N00/8022873998/in/photostream
2012-09-25T14:19:04 < cjbaird> The Freescale Cortex-M0 Arduino-wannabe.
2012-09-25T14:19:30 < dongs> oh i remember that somewhat
2012-09-25T14:27:46 < Laurenceb> http://www.digikey.co.uk/product-detail/en/STM32373C-EVAL/497-13191-ND/3522186
2012-09-25T14:27:51 < Laurenceb> get them while they last
2012-09-25T14:28:30 < dongs> haha
2012-09-25T14:28:46 < dongs> time to poke my chinacontacts for F3 samples
2012-09-25T14:28:57 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 268 seconds]
2012-09-25T14:31:24 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-09-25T14:43:58 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 265 seconds]
2012-09-25T14:52:05 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-09-25T14:52:06 -!- Bird|lappy [~Bird|ub3r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-09-25T14:52:06 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-25T15:05:11 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
2012-09-25T15:19:36 < Laurenceb> is there such a thing as usb mini male?
2012-09-25T15:19:45 < Laurenceb> thats square ?
2012-09-25T15:21:01 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-25T15:21:05 < Laurenceb> http://uk.rs-online.com/web/p/usb-cable-assemblies/0411188/
2012-09-25T15:22:18 < karlp> you mean mini-A?
2012-09-25T15:22:27 < karlp> there was, but it's deprecated
2012-09-25T15:22:33 < Laurenceb> http://uk.rs-online.com/web/p/usb-cable-assemblies/0411188/
2012-09-25T15:22:43 < karlp> that's the same linke dude.
2012-09-25T15:22:51 < Laurenceb> oops
2012-09-25T15:23:12 < Laurenceb> is that mini b ?
2012-09-25T15:23:58 < karlp> hard to see.
2012-09-25T15:24:07 < karlp> looks like it.
2012-09-25T15:24:18 < karlp> the sort that connects to a discovery board for instance.
2012-09-25T15:25:46 < Laurenceb> yeah
2012-09-25T15:25:53 < Laurenceb> ive seen some square ended cables
2012-09-25T15:26:17 < karlp> micro A?
2012-09-25T15:26:46 < karlp> would fit into the OTG port of the f4 discovery for instance?
2012-09-25T15:31:52 < zyp> no
2012-09-25T15:31:59 < zyp> that cable is some bullshit connector
2012-09-25T15:32:23 < Laurenceb> zyp: you think so?
2012-09-25T15:32:30 < Laurenceb> it looked like something odd to me
2012-09-25T15:32:52 < dongs> http://www.zsmls.com/en/productshow.asp?productid=2636&catalogid=18&type_id=122
2012-09-25T15:32:54 < dongs> do wnat
2012-09-25T15:32:56 < dongs> want rather
2012-09-25T15:33:54 < Laurenceb> rage
2012-09-25T15:34:03 < karlp> less rage, more cookies.
2012-09-25T15:34:07 < Laurenceb> where do i find retractible usb a to mini b  ?
2012-09-25T15:34:24 < karlp> dx?
2012-09-25T15:34:30 < karlp> ebay?
2012-09-25T15:34:35 < Laurenceb> has to be farnell/rs etc
2012-09-25T15:34:43 < zyp> dongs, your link doesn't work
2012-09-25T15:34:48 < dongs> bullshit
2012-09-25T15:34:55 < Laurenceb> works for me
2012-09-25T15:35:06 < Laurenceb> zyp had cp filtering
2012-09-25T15:35:14 < dongs> maybne doesnt worwk in lunix
2012-09-25T15:35:28 < Laurenceb> ubuntu here
2012-09-25T15:35:44 < Laurenceb> patched through to T1 line with NO filtering
2012-09-25T15:35:51  * Laurenceb has bypassed cleanfeed
2012-09-25T15:36:11 < zyp> I get two lines of chinese gibberish and some link to www.safedog.cn
2012-09-25T15:36:31 < Laurenceb> woof
2012-09-25T15:36:52 < zyp> oh
2012-09-25T15:36:54 < zyp> The server prohibit proxy access!
2012-09-25T15:36:55 < zyp> Internet security laboratory from a security dog - the website security dog friendly software prompted, learn more, please click
2012-09-25T15:37:01 < Laurenceb> lol
2012-09-25T15:37:05 < zyp> that's what it says
2012-09-25T15:37:22 < Laurenceb> the cp has been taken down then
2012-09-25T15:37:36 < zyp> so, what am I missing out on?
2012-09-25T15:37:38 < Laurenceb> luckily i cached it
2012-09-25T15:37:53 < dongs> zyp: do you get some safedog.cn shit?
2012-09-25T15:38:01 < Laurenceb> woof
2012-09-25T15:38:33  * Laurenceb heads off to piratebay
2012-09-25T15:38:36 < Laurenceb> muhahahaha
2012-09-25T15:39:50 < zyp> dongs, that's what I just said?
2012-09-25T15:40:08 < Laurenceb> http://uk.rs-online.com/web/p/usb-cable-assemblies/7589333/
2012-09-25T15:40:11 < Laurenceb> dafuk
2012-09-25T15:40:23 < dongs> oh
2012-09-25T15:40:24 < Laurenceb> arduino sell the cheapest cables
2012-09-25T15:40:30 < dongs> weied
2012-09-25T15:41:31 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-25T15:43:20 < dongs> wut
2012-09-25T15:43:30 < dongs> Laurenceb: i pay < $1 for my 1M usb cables
2012-09-25T15:43:55 < Laurenceb> a million cables?
2012-09-25T15:43:58 < Laurenceb> nice
2012-09-25T15:45:04 < dongs> 1 meter.
2012-09-25T15:45:17 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-25T15:47:46 < Laurenceb> i know
2012-09-25T15:52:25 < dongs> http://i.imgur.com/Rpu1Z.png
2012-09-25T15:52:27 < dongs> it might work.
2012-09-25T15:53:33 < Laurenceb> enjoy the soldering
2012-09-25T15:56:46 -!- daku is now known as DaKu
2012-09-25T16:06:57 < dongle> maybe i should use 3 channel  driver
2012-09-25T16:07:02 < dongle> then i only need one per 3 leds.
2012-09-25T16:09:31 -!- dongs_ [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-25T16:09:34 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Read error: Connection reset by peer]
2012-09-25T16:10:42 -!- BrainDamage [~yaaic@217.200.200.237] has joined ##stm32
2012-09-25T16:11:57 -!- dongs_ is now known as dongs
2012-09-25T16:12:02 -!- BrainDamage [~yaaic@217.200.200.237] has quit [Client Quit]
2012-09-25T16:20:25 -!- dekar [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-09-25T16:32:31 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds]
2012-09-25T16:33:12 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-25T16:45:59 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-25T16:49:07 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has joined ##stm32
2012-09-25T16:49:10 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-25T16:50:26 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has quit [Client Quit]
2012-09-25T16:51:13 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has joined ##stm32
2012-09-25T16:51:17 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-25T16:55:14 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has quit [Ping timeout: 240 seconds]
2012-09-25T16:59:29 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-25T16:59:29 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-25T16:59:29 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-25T16:59:32 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-25T17:12:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-25T17:39:00 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-25T17:54:11 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-25T17:55:16 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds]
2012-09-25T17:56:09 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-25T17:56:27 < Laurenceb> what the fuuuu
2012-09-25T17:56:33 < jpa-> ck
2012-09-25T17:56:34 < Laurenceb> gedit is using 8GB ram
2012-09-25T17:56:55 < Laurenceb> i have had it running for 3 years...
2012-09-25T17:57:00 < jpa-> it's because it always buffers up to the next whitespace character
2012-09-25T18:00:58 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has joined ##stm32
2012-09-25T18:01:20 <+Steffanx> heh jpa-
2012-09-25T18:06:10 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has joined ##stm32
2012-09-25T18:09:55 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-25T18:09:55 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-25T18:09:56 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-25T18:20:48 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-25T18:25:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-25T18:25:28 -!- DaKu is now known as daku
2012-09-25T18:28:21 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-25T18:28:58 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-25T18:30:31 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has joined ##stm32
2012-09-25T18:30:35 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-25T18:34:57 < Mazingaro> hi
2012-09-25T18:35:31 < Mazingaro> please help, I developed a tiny RTOS, but I've got problem with stack alignment
2012-09-25T18:36:17 < Mazingaro> it seems that when cpu pushes the eight registers on PSP their 24 byte misaligned...
2012-09-25T18:37:50 < zyp> misaligned?
2012-09-25T18:38:25 < Mazingaro> yes psp points to 6 bytes before the correct eight bytes
2012-09-25T18:41:48 < karlp> it's not just pushing more than you think?
2012-09-25T18:42:42 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 264 seconds]
2012-09-25T18:44:00 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 252 seconds]
2012-09-25T18:48:25 < Mazingaro> when cpu executes an ISR pushes 8 registers
2012-09-25T18:48:47 < Mazingaro> r0 thru r3, r12, lr, pc and xpsr
2012-09-25T18:49:28 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-25T18:49:36 < karlp> 8*4 = 32 bytes?
2012-09-25T18:52:20 < zippe> There's also an option that will push a padding word if the stack is not 8-aligned
2012-09-25T18:52:56 < zyp> zippe, before or after the registers?
2012-09-25T18:53:05 < zippe> zyp: above
2012-09-25T18:53:11 < zyp> so before
2012-09-25T18:53:16 < zippe> Mazingaro: how are you reading the psp?
2012-09-25T18:53:23 < zippe> zyp: it's written after, but above
2012-09-25T18:53:34 < zippe> or rather, it's not actually "written"
2012-09-25T18:53:59 < zippe> but ARM use a full-descending stack, so the stack pointer is decremented and then the context is written in incrementing order from the new stack pointer up
2012-09-25T18:54:02 < zyp> when you say above, you mean above in memory, as in earlier on the stack?
2012-09-25T18:54:09 < zippe> At a higher address
2012-09-25T18:54:12 < zyp> right
2012-09-25T18:54:28 < Mazingaro> mrs psp, r0
2012-09-25T18:54:40 < zyp> order doesn't really matter, it's all done before the ISR is entered
2012-09-25T18:55:40 < zippe> Mazingaro: you're doing that inside your ISR, yes?
2012-09-25T18:56:33 < Mazingaro> yes
2012-09-25T18:56:37 < zippe> Mazingaro: and I'b betting your ISR is written in C.  And you haven't marked the ISR function as naked, so the compiler is emitting a prolog that is stacking some local variables.
2012-09-25T18:56:51 < Mazingaro> ISR is nacked
2012-09-25T18:57:00 < zippe> The 6-byte alignment is a bit suspicious thoug
2012-09-25T18:57:27 < zippe> nothing the compiler or the machine does will give you a non-4-aligned stack pointer
2012-09-25T18:59:16 < BrainDamage> if anyone here has an ieee access, you'd better be resetting your pw: http://ieeelog.com/
2012-09-25T19:09:35 < Mazingaro> arm it definitely a messed processor :)
2012-09-25T19:10:24 < Thorn> except most others are messed up even more
2012-09-25T19:10:47 -!- dekar [~dekar@studpool-wlan-74-223.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-09-25T19:13:02 < Mazingaro> not consider the x86
2012-09-25T19:16:11 < Thorn> x86 is by far the worst
2012-09-25T19:17:09 < emeb> Processor design is like other kinds of engineering - a whole boatload of compromises.
2012-09-25T19:17:14 < zyp> arm looks pretty clean to me
2012-09-25T19:17:35 < emeb> Trouble is that in order to sell a processor you have to show everyone those compromises.
2012-09-25T19:18:18 < Thorn> I think most FUD about arm is generated by avr/pic people who don't expect so much complexity in a microcontroller
2012-09-25T19:19:50 < emeb> heh
2012-09-25T19:20:21 < emeb> people who think ARM is ugly haven't seen the PIC.
2012-09-25T19:20:48 <+Steffanx> I need ieee access BrainDamage :P
2012-09-25T19:20:49 < Thorn> or grew up on PIC and don't see the ugliness, like php programmers
2012-09-25T19:21:43  * emeb started out on the 6502. That's perspective.
2012-09-25T19:22:24 < emeb> At least we don't have any Propeller fans here. :P
2012-09-25T19:22:39 < emeb> propeller. fan. see what I did there?
2012-09-25T19:23:46 <+Steffanx> It's not funny to sort of explain a joke, emeb :P
2012-09-25T19:24:23 < BrainDamage> and it wan't phunny even without explanation
2012-09-25T19:24:23 < emeb> Steffanx: it wasn't that funny of a joke to start out. No great loss.
2012-09-25T19:24:39 <+Steffanx> Ok, good to hear
2012-09-25T19:25:31 < zippe> Mazingaro: Which ARM? There are so many to choose from
2012-09-25T19:25:34 < zippe> Most aren't really so bad
2012-09-25T19:25:57 < zippe> Mazingaro: it's obvious you haven't dealt with very many
2012-09-25T19:26:19 < zippe> go play with mips32 for a bit, or any of a dozen or so historical architectures
2012-09-25T19:26:30 < zippe> Alpha might be a good start.
2012-09-25T19:26:32 < Mazingaro> cortex-m3 for example
2012-09-25T19:26:40 -!- daku is now known as DaKu
2012-09-25T19:26:45 < zippe> You mean v7m?  It's delightful.
2012-09-25T19:26:51  * emeb read MIPS documents once. The horror.
2012-09-25T19:26:53 < zippe> So simple, so consistent
2012-09-25T19:27:12 < zippe> There are really only a couple of very minor gotchas with v7m
2012-09-25T19:27:25 < zippe> Apart from that, it's easier than falling off a log
2012-09-25T19:27:26 < Thorn> I've never actually used mips but it looks very nice (and they were planning to compete with cortex-m too, might be interesting)
2012-09-25T19:27:43 < zippe> Thorn: The new high-end PIC parts are mips32
2012-09-25T19:27:54 < zippe> A bit slow, but not terrible
2012-09-25T19:27:55 < emeb> Thorn: Microchip is using MIPS in their PIC32.
2012-09-25T19:28:01 < Thorn> yeah, thry were planning a cortex-m like line of cores
2012-09-25T19:28:05 < Thorn> *they
2012-09-25T19:28:27 < Thorn> to license to multiple silicon manufacturers
2012-09-25T19:28:28 < zippe> Mazingaro: Honestly, if you think that v7m is hard, you need a new job.
2012-09-25T19:28:35 < emeb> It's such a fine-grained instruction set - needs a lot of instructions to do simple things.
2012-09-25T19:29:58 < zippe> emeb: this is why we have compilers 8)
2012-09-25T19:30:02 < Mazingaro> zippe: did you design an RTOS with it?
2012-09-25T19:30:13 < karlp> if _I_ can write startup code for it, it's easy.
2012-09-25T19:30:22 < karlp> hell, no asm required for startup! whoo!
2012-09-25T19:30:24 < emeb> zippe: yep - compilers are needed for MIPS. And lots of program memory too.
2012-09-25T19:31:02 < Thorn> one mistake that I notice beginners do is limit themselves to a vendor reference manual (it mostly describes peripherals). you need ARM ARM and books by Joseph Yu
2012-09-25T19:31:16 < zippe> Mazingaro: I'm on my third
2012-09-25T19:32:00 < Mazingaro> why three? the other two didn't work?
2012-09-25T19:32:06 < zippe> Different applications
2012-09-25T19:32:24 < zippe> Mazingaro: your question makes it pretty clear you haven't sorted out how to handle context switching yet.
2012-09-25T19:32:44 < zippe> In particular, are you trying to context switch inside an interrupt handler?
2012-09-25T19:33:39 < Mazingaro> yes
2012-09-25T19:33:46 < zippe> Don't do that
2012-09-25T19:34:12 < zippe> You've failed to understand PendSV
2012-09-25T19:34:29 < Mazingaro> I'm using SysTick
2012-09-25T19:34:33 < zippe> Don't
2012-09-25T19:35:14 < Mazingaro> isn't supposed to be used for RTOS development?
2012-09-25T19:35:14 < zippe> You need a hand-rolled PendSV handler that stacks the rest of your full context frame and then calls a second-level handler with the frame address as an argument.
2012-09-25T19:35:21 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-25T19:35:22 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-25T19:35:22 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-25T19:35:23 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-25T19:35:28 < zippe> SysTick is just a timer.
2012-09-25T19:35:35 < zippe> You don't context switch using it
2012-09-25T19:35:53 < zippe> You schedule a PendSV and context switch in the PendSV handler.
2012-09-25T19:35:54 < Mazingaro> why PendSV should be better?
2012-09-25T19:36:12 < zippe> Because then you don't waste time stacking a full context for every interrupt
2012-09-25T19:36:25 < Mazingaro> ok
2012-09-25T19:36:29 < zippe> The compiler will take care of it for you in the prolog
2012-09-25T19:36:53 < zippe> Stacking a full context on v7e-m is murder
2012-09-25T19:37:11 < zippe> Since you have all the FP registers as well
2012-09-25T19:37:21 < Mazingaro> I need fp registers too
2012-09-25T19:37:40 < Laurenceb> argg
2012-09-25T19:37:43 < zyp> zippe, how are you handling that? just stacking everything for every context switch?
2012-09-25T19:37:46 < Laurenceb> can anyone help me with chibios?
2012-09-25T19:37:48 < zippe> You really, really haven't understood the v7(e-)m exception model
2012-09-25T19:37:55 < Laurenceb> its hardfaulting in halinit
2012-09-25T19:37:55 < zyp> or are you doing some tricks with the lazy stuff?
2012-09-25T19:37:58 < zippe> zyp: several different ways, depending on the system
2012-09-25T19:38:11 < zippe> zyp: the fastest way is to enable lazy save
2012-09-25T19:38:37 < zippe> zyp: then in the PendSV handler when you touch the callee-saved FPRs that triggers the lazy save
2012-09-25T19:39:24 < zippe> Your ISRs pay for the lazy save stack space but unless they touch the FPU not the stores and loads
2012-09-25T19:39:28 < Mazingaro> zippe: ok, but how do context switch, could you provide me a tutorial?
2012-09-25T19:39:36 < zippe> Mazingaro: no.
2012-09-25T19:39:49 < Mazingaro> zippe: good...
2012-09-25T19:39:52 < zippe> Mazingaro: unless we agree on a contract
2012-09-25T19:40:06 <+Steffanx> $$
2012-09-25T19:40:20 < zippe> I bill in ten-hour blocks; my charity rate is $250 an hour
2012-09-25T19:40:34 < Mazingaro> oh I take less
2012-09-25T19:40:35 < zyp> zippe, that's how I've understood it, but I haven't implemented it yet
2012-09-25T19:41:34 < zippe> Mazingaro: I've given you all the google keywords you need.  Just read.
2012-09-25T19:41:59 < zippe> zyp: TBH the current system I'm working on just stacks everything up front because of some legacy portability decisions
2012-09-25T19:42:13 < Mazingaro> ok tx
2012-09-25T19:42:15 < zyp> I haven't needed to use the FPU yet, so I haven't taken the time to write the code to enable and handle the FPU
2012-09-25T19:42:21 < zippe> zyp: It's ugly, but latency is still in the ~2µs range which is good enough for our current purposes
2012-09-25T19:42:33 < zippe> zyp: most of the FPU handling is trivial.
2012-09-25T19:42:51 < zippe> zyp: the thing that broke my brain with the autosave stuff is correctly switching the autosave context
2012-09-25T19:42:51 < zyp> yeah, it's mostly the context switch that I have to care about
2012-09-25T19:43:17 < zippe> zyp: if you use PendSV it's much easier, because you have one set of rules for interrupts and another for context switches
2012-09-25T19:43:40 < zippe> zyp: nuttx does ASTs on interrupt return, and it's a PITA
2012-09-25T19:43:44 < zyp> I just figure that in a normal application, only a few of the threads will actually use the FPU
2012-09-25T19:43:50 < zyp> so adding the overhead for all seems silly
2012-09-25T19:43:57 < zippe> zyp: Don't jump to that assumption
2012-09-25T19:44:08 < zyp> s/silly/unnecessary/
2012-09-25T19:44:09 < karlp> what's AST?
2012-09-25T19:44:20 < zippe> zyp: with 4.7 the compiler is more likely to autovectorise things
2012-09-25T19:44:31 < zippe> karlp: asynchronous system traps
2012-09-25T19:44:36 < zippe> karlp: e.g. signal delivery
2012-09-25T19:45:06 < zippe> karlp: also thread timers, etc. etc.
2012-09-25T19:45:35 < zippe> karlp: it will also context switch on the interrupt return side, which annoys me greatly
2012-09-25T19:45:53 < zippe> But it's the price you pay for a portable architecture
2012-09-25T19:46:00 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-25T19:49:20 < karlp> is that so that the same code works on archs that don't have PendSV for instance?
2012-09-25T19:49:29 < karlp> where doing the switch in the interrupt is the only option?
2012-09-25T19:50:13 -!- tavish [~tavish@120.59.135.93] has joined ##stm32
2012-09-25T19:50:13 -!- tavish [~tavish@120.59.135.93] has quit [Changing host]
2012-09-25T19:50:13 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-25T19:50:25 < zippe> karlp: nuttx works on AVR, armv4/v7, mips, z80, etc. etc.
2012-09-25T19:50:39 < zippe> karlp: most platforms don't have a dedicated context switch trap, no
2012-09-25T19:51:02 < zippe> Mazingaro: if you're looking for a simple example, I recommend scmRTOS
2012-09-25T19:54:47 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-09-25T19:55:13 < TitanMKD> hi
2012-09-25T19:59:37 <+Steffanx> lo
2012-09-25T20:00:25 < zyp> zippe, I weren't aware of the purpose of pendsv myself, I've been using svcall for context switching so far
2012-09-25T20:00:35 < zyp> zippe, does this pseudocode look sensible: http://paste.jvnv.net/view/nAYMa
2012-09-25T20:00:41 < zyp> ?
2012-09-25T20:05:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-09-25T20:06:04 -!- tavish_ [~tavish@120.59.135.93] has joined ##stm32
2012-09-25T20:11:11 < zyp> seems more sensible to perform the scheduling ahead of setting pendsv, and then just let pendsv change to whatever thread is currently pending
2012-09-25T20:14:40 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has joined ##stm32
2012-09-25T20:17:45 < Mazingaro> zippe: tx
2012-09-25T20:18:33 < Mazingaro> zippe: I'm reading the STM32F10x standard peripherals firmware library, just to see how setup the pendsv handler
2012-09-25T20:27:23 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 256 seconds]
2012-09-25T20:27:36 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-09-25T20:29:06 < zippe> Mazingaro: That's a terrible, terrible idea
2012-09-25T20:29:16 < zippe> Mazingaro: the ST stdperiph library is awful
2012-09-25T20:29:26 < zippe> Mazingaro: it will teach you bad code habits, and confuse you greatly.
2012-09-25T20:29:41 < Mazingaro> zippe: it's a bad doc
2012-09-25T20:29:47 -!- tavish_ [~tavish@120.59.135.93] has quit [Quit: Reconnecting]
2012-09-25T20:29:55 < zippe> zyp: that's a waste of work, since you can see several interrupts back to back that repeatedly change your idea of what's runnable
2012-09-25T20:30:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-25T20:31:52 < zyp> hmm, true
2012-09-25T20:32:20 -!- tavish [~tavish@unaffiliated/tavish] has quit [Client Quit]
2012-09-25T20:33:14 < zyp> http://paste.jvnv.net/view/AZl4o <- something like this then
2012-09-25T20:34:23 < zyp> not very different, depending on how the scheduler is implemented
2012-09-25T20:35:52 < zippe> zyp: that's one way of looking at it, yes
2012-09-25T20:36:24 < zippe> zyp: though in the systick handler, if you're just ticking quanta (a dumb idea; using systick for deadlines works better) then you can just schedule a pendsv without bothering to ask the scheduler
2012-09-25T20:36:56 < zippe> zyp: since when you call the scheduler from the pendsv handler it's going to look at who's the most-runnable & do quantum expiration there
2012-09-25T20:38:23 < zyp> I'm planning to go with deadlines when I actually bother writing a scheduler
2012-09-25T20:38:50 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-25T20:44:34 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-25T20:46:18 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-09-25T20:51:48 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 265 seconds]
2012-09-25T20:52:10 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-25T21:08:28 < Mazingaro> zippe: if I understood, the PendSV handler executes the scheduler without caring about context switching ISR stack manipulation
2012-09-25T21:09:08 < zyp> no
2012-09-25T21:09:20 < Mazingaro> and what?
2012-09-25T21:09:35 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-25T21:11:28 < jpa-> it probably depends solely on how zyp has implemented it :)
2012-09-25T21:12:07 < zyp> this has nothing to do with me
2012-09-25T21:12:19 < jpa-> ah crap
2012-09-25T21:12:23 < jpa-> zyp and zippe all confused
2012-09-25T21:12:57 -!- tech2077_ [~tech2077@207.80.127.250] has joined ##stm32
2012-09-25T21:13:06 < zyp> the point zippe is making is that pendsv handler should be used for context switching, and all other handlers that needs to trigger a context switch just needs to set pendsv
2012-09-25T21:13:18 < Thorn> pensdv is also an interrupt
2012-09-25T21:13:45 <+Steffanx> exception :)
2012-09-25T21:14:10 < Thorn> yes, that's the correct term
2012-09-25T21:14:21 < Thorn> exceptions = interrupts + faults (+reset)
2012-09-25T21:15:14 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 240 seconds]
2012-09-25T21:16:19 < Mazingaro> zyp: so I still need to manipulate the stack to provide the 8 registers to return to thread...
2012-09-25T21:17:24 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-09-25T21:17:28 -!- tech2077_ [~tech2077@207.80.127.250] has quit [Ping timeout: 246 seconds]
2012-09-25T21:18:30 -!- barthess [~barthess@178.154.99.208] has joined ##stm32
2012-09-25T21:41:07 < Tectu> zippe, ping
2012-09-25T21:43:03 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-25T21:43:07 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-25T21:46:01 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-09-25T21:49:14 -!- dekar [~dekar@drms-4d014d4d.pool.mediaWays.net] has joined ##stm32
2012-09-25T21:49:18 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-25T21:49:26 < Thorn> >When I sent you the file, I explicitly said: "Please note that all the shapes in the GBO file are actually non plated holes." and the title of my mail was "PCB file for order 57739 /!\ MULTIPLE NON PLATED HOLES IN .GBO"
2012-09-25T21:49:39 < Thorn> >You forgot to drill the square shapes in the GBO file and therefore I have now a plain PCB with the shapes drawn on the silkscreen.
2012-09-25T21:50:36 <+Steffanx> seeed? itead?
2012-09-25T21:50:39 < Thorn> seeed
2012-09-25T21:50:41 < TitanMKD>  yes fixed STM32F4 Debug board
2012-09-25T21:51:22 < TitanMKD> anyway it is interesting to know VCAP_1/2 really does not like direct connection to 3.3V as the MCU is just dead
2012-09-25T21:51:32 < Thorn> that's hardly the correct way to specify non-plated holes, is it
2012-09-25T21:52:21 < Thorn> TitanMKD: I was going to ask you why you have pairs of caps seemingly in parallel on your pcb but went to sleep
2012-09-25T21:53:29 < TitanMKD> hehe ;)
2012-09-25T21:53:40 <+Steffanx> "a plain PCB" Thorn ?!
2012-09-25T21:53:52 <+Steffanx> plain as in .. silkscreen only?
2012-09-25T21:53:53 < TitanMKD> next time i will avoid to do schematic too late to do such big mistake
2012-09-25T21:54:11 < Thorn> Steffanx: no pics unfortunately
2012-09-25T21:54:17 <+Steffanx> Take them :)
2012-09-25T21:55:05 < Thorn> no address to travel to either
2012-09-25T21:55:42 < Thorn> it's from a question on their website.
2012-09-25T21:57:22 <+Steffanx> Ah
2012-09-25T21:58:11 < Thorn> TitanMKD: are you sure you didn't short vcap pins to ground?
2012-09-25T21:58:20 <+Steffanx> but hey "sorry for the inconvenience, we have redo for you" that's nice Thorn :)
2012-09-25T21:58:38 < Laurenceb_> Thorn: seed cannot do unplated holes
2012-09-25T21:58:42 < Thorn> yeah, poor seeed
2012-09-25T21:58:42 < Laurenceb_> try pcbtrain
2012-09-25T21:58:55 < Thorn> afair those caps are for LDO stability
2012-09-25T22:01:14 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-09-25T22:01:15 <+Steffanx> Uh? They can't do unplated holes Laurenceb?!
2012-09-25T22:01:16 < TitanMKD> Thorn no in fact I have connected VCAP1/2 to +3.3V instead of 2.2uF capa
2012-09-25T22:01:17 <+Steffanx> + _
2012-09-25T22:01:50 < Thorn> ah, then you connected the internal 1.8V rail to 3.3V
2012-09-25T22:01:55 < TitanMKD> yes ;)
2012-09-25T22:02:25 < TitanMKD> result the chip was very hot and is dead (no access with SWD/JTAG)
2012-09-25T22:03:28 < TitanMKD> Thorn the actual fix just do not connect VCAP1/2 to anything
2012-09-25T22:03:36 < TitanMKD> so ADC shall be totally wrong ...
2012-09-25T22:04:37 < Thorn> ADC?
2012-09-25T22:04:49 < TitanMKD> ha no it is internal voltage regulator in fact
2012-09-25T22:05:04 < Thorn> that LDO powers the core afaik
2012-09-25T22:05:29 < TitanMKD> i will try to connect them to 2.2uF CAP on my fix until i buy new batch of working boards
2012-09-25T22:05:40 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 252 seconds]
2012-09-25T22:07:38 < Oldboy> this page http://www.diyertool.com/dso203-pocket-sized-digital-oscilloscope-arm-dso-nano.html refers to a  STM32VCT6 chip. google has no hits for that part number.
2012-09-25T22:07:42 < Oldboy> any idea what it might be ?
2012-09-25T22:08:05 < Thorn> stm32f1vct6
2012-09-25T22:08:16 < Thorn> *stm32f103vct6
2012-09-25T22:08:27 < zyp> probably
2012-09-25T22:09:04 < Oldboy> VCT6 is package specification ?
2012-09-25T22:09:14 < zyp> the C is the package
2012-09-25T22:09:22 < Thorn> see http://stm32.izua.ro/wiki/STM32F1
2012-09-25T22:09:46 < zyp> sorry, no, the V is package, C is memory
2012-09-25T22:10:42 < Oldboy> that page is nice. thank you
2012-09-25T22:12:22 < Laurenceb_> <Steffanx> Uh? They can't do unplated holes Laurenceb?!
2012-09-25T22:12:24 < Laurenceb_> ^correct
2012-09-25T22:12:36 < Laurenceb_> you need an extra drill rig on the end of the line to do that
2012-09-25T22:12:46 < Laurenceb_> they are cheapscates so they dont
2012-09-25T22:17:24 < jpa-> Oldboy: btw, DS203 is DSO Quad, not DSO Nano (just to avoid some confusion caused by bad product title)
2012-09-25T22:18:12 < Oldboy> jpa-: :) I'm watching mikeselectricstuff teardown.
2012-09-25T22:18:17 < jpa-> Oldboy: schematics: http://essentialscrap.com/dsoquad/DS203V2.png
2012-09-25T22:20:10 <+Steffanx> According to my sources they CAN do it.. when you use labels in the milling layer Laurenceb
2012-09-25T22:20:29 <+Steffanx> like http://www.naffets.nl/share//a-20120925-212021.jpg
2012-09-25T22:21:35 <+Steffanx> You asked them Laurenceb_?
2012-09-25T22:24:48 -!- dekar [~dekar@drms-4d014d4d.pool.mediaWays.net] has quit [Quit: This computer has gone to sleep]
2012-09-25T22:26:11 < Laurenceb_> no
2012-09-25T22:26:11 -!- barthess [~barthess@178.154.99.208] has quit [Read error: Connection reset by peer]
2012-09-25T22:26:21 <+Steffanx> heh
2012-09-25T22:26:29 < Laurenceb_> ive sent jobs like that
2012-09-25T22:26:35 < Laurenceb_> and got plated holes
2012-09-25T22:26:47 < jpa-> plated holes with no annular ring?
2012-09-25T22:27:19 < zyp> is it very important that the holes are not plated?
2012-09-25T22:27:56 < Laurenceb_> it can reduce risk of shorts
2012-09-25T22:28:25 -!- DaKu is now known as daku
2012-09-25T22:28:29 < zyp> shorts with what?
2012-09-25T22:28:33 <+Steffanx> As long as the diameter is ok .. for me not zyp
2012-09-25T22:28:56 < zyp> diameter is always finished size after plating anyway
2012-09-25T22:30:18 <+Steffanx> You can never trust them :P
2012-09-25T22:33:25 < TitanMKD> anyone know why they do not upgrade DSO Quad with an STM32F4 ? at least to have faster FFT and more memory for buffer ...
2012-09-25T22:34:23 <+Steffanx> jpa- the DSO Quad is your thing :)
2012-09-25T22:38:45 -!- daku is now known as DaKu
2012-09-25T22:42:31 < Laurenceb_> http://pipe2.darklomax.org/pics/2011-05-21_Foxhunting_Prep/
2012-09-25T22:42:36 < Laurenceb_> meanwhile in ham land
2012-09-25T22:42:46 < Mazingaro> zyp: I still did not understand how it works :( grrrr
2012-09-25T23:03:58 -!- enots [dimka@194.190.195.85] has quit [Ping timeout: 252 seconds]
2012-09-25T23:05:08 < Oldboy> TitanMKD:I think the STM32 on the DSO Quad doesn't do anything - maybe just the menus. everyting is in the FPGA.
2012-09-25T23:05:15 < Oldboy> just guessing though.
2012-09-25T23:05:42 < zyp> jpa- would know, he has been playing around with the firmware
2012-09-25T23:05:43 < TitanMKD> Oldboy i doubt the FPGA do FFT ;)
2012-09-25T23:05:54 -!- enots [dimka@freelsd.net] has joined ##stm32
2012-09-25T23:06:00 -!- amitofu [~str@cannabis.dataforce.net] has quit [Ping timeout: 276 seconds]
2012-09-25T23:13:44 < Thorn> if it has hardware multipliers, why not.
2012-09-25T23:15:16 < zyp> doesn't most, except maybe the smallest, FPGAs have multiplier blocks?
2012-09-25T23:15:31 < emeb> most I know of do.
2012-09-25T23:15:39 < emeb> except ollld stuff.
2012-09-25T23:16:14 < Thorn> they used something obscure there afair. but if it has internal ram, it should also have multipliers
2012-09-25T23:16:41 < emeb> multipliers & SRAM & clock cycles are all you need. :)
2012-09-25T23:17:05 < emeb> and Xilinx FFT cores are free.
2012-09-25T23:17:15 < Thorn> it's not xilinx
2012-09-25T23:17:39 < emeb> Altera?
2012-09-25T23:17:57 < Thorn> Lattice
2012-09-25T23:18:00 < Oldboy> altera cyclone
2012-09-25T23:18:22 < Thorn> this is the schematics? http://essentialscrap.com/dsoquad/DS203V2.png
2012-09-25T23:18:35 < Oldboy> as per jpa-
2012-09-25T23:18:52 < Thorn> ice65 old obsolete lattice something
2012-09-25T23:19:11 < Oldboy> yeah I don't think that's the correect schematic
2012-09-25T23:19:29 < Oldboy> the teardown video showed an Altera cyclone II
2012-09-25T23:20:03 < emeb> no multipliers in the ice65l04f
2012-09-25T23:20:11 < Thorn> fwiw, this doesn't show any multipliers http://www.latticesemi.com/documents/iCE65Datasheet_120330.pdf
2012-09-25T23:20:43 < emeb> looks like the main bullet point on that part is low power.
2012-09-25T23:21:25 < Mazingaro> zyp: I was able to switch to the task, but the first push in the task code jams the cpu in FaultHandler
2012-09-25T23:21:49 < Thorn> wrong sp value?
2012-09-25T23:22:17 -!- amitofu [~str@cannabis.dataforce.net] has joined ##stm32
2012-09-25T23:23:29 < Mazingaro> Thorn: I guess, but I use the same stack to retrieve the lr value for returning from ISP...
2012-09-25T23:24:14 < Mazingaro> and the same stack frame is used to return to task from ISR
2012-09-25T23:24:25 < Thorn> wrong load/store order (and lr is in the middle)? :)
2012-09-25T23:27:38 < Mazingaro> Thorn: it would be wrong PC too
2012-09-25T23:34:42 < Mazingaro> really don't know
2012-09-25T23:57:12 < TitanMKD> Mazingaro are you in user mode or privileged ?
2012-09-25T23:57:45 < Mazingaro> from ISR switching to unpriviledged mode
2012-09-25T23:58:28 < TitanMKD> it is on CortexMx or other ?
--- Day changed Wed Sep 26 2012
2012-09-26T00:00:23 < Mazingaro> M3
2012-09-26T00:02:25 < TitanMKD> anyway without code it is really hard to tell anything
2012-09-26T00:07:55 -!- oPossum5150 [~opossum51@adsl-108-68-179-94.dsl.klmzmi.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-09-26T00:10:37 -!- oPossum5150 [~opossum51@adsl-76-253-107-152.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-26T00:12:48 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has quit [Remote host closed the connection]
2012-09-26T00:13:09 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-09-26T00:13:21 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Quit: Leaving]
2012-09-26T00:13:41 -!- oPossum5151 [~opossum51@adsl-76-253-108-184.dsl.klmzmi.sbcglobal.net] has joined ##stm32
2012-09-26T00:13:43 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-26T00:15:11 -!- oPossum5150 [~opossum51@adsl-76-253-107-152.dsl.klmzmi.sbcglobal.net] has quit [Ping timeout: 256 seconds]
2012-09-26T00:20:52 < Mazingaro> TitanMKD: I'm taking a look to: http://www.coactionos.com/embedded-design/36-context-switching-on-the-cortex-m3.html
2012-09-26T00:20:59 < Mazingaro> but it is very confusing
2012-09-26T00:21:41 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-26T00:21:42 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-26T00:21:42 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-26T00:23:23 < TitanMKD> Mazingaro it is easy in fact
2012-09-26T00:24:03 < TitanMKD> Mazingaro just becareful to save the user/system context using ^ at end
2012-09-26T00:24:17 < TitanMKD> and not to save the irq/svc/... context
2012-09-26T00:24:20 < Mazingaro> TitanMKD: it is a cortex-m3
2012-09-26T00:24:24 < TitanMKD> especially for r13/r14 ...
2012-09-26T00:24:41 < Mazingaro> does ^ works?
2012-09-26T00:24:53 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-26T00:25:20 < TitanMKD> yes
2012-09-26T00:25:39 < TitanMKD> at end of pop/push it means save user registers instead of irq/fiq...
2012-09-26T00:26:27 < TitanMKD> but in my case i save full context
2012-09-26T00:28:35 < Mazingaro> ^ is not supported
2012-09-26T00:30:35 < TitanMKD> you can switch to system mode in that case
2012-09-26T00:31:03 < Mazingaro> assembler won't assemble anything with ^
2012-09-26T00:31:18 < Mazingaro> I think that thumb2 mode hasn't got ^
2012-09-26T00:31:43 < TitanMKD> i'm sure it is supported
2012-09-26T00:32:13 < TitanMKD> look the ref card
2012-09-26T00:32:14 < TitanMKD> http://infocenter.arm.com/help/topic/com.arm.doc.qrc0001l/QRC0001_UAL.pdf
2012-09-26T00:32:32 < TitanMKD> it is supported
2012-09-26T00:32:33 < TitanMKD> LDM{IA|IB|DA|DB} Rn{!}, <reglist+PC>^
2012-09-26T00:33:46 < TitanMKD> you can save the full context or the partial context but it is safer to save all registers
2012-09-26T00:37:00 < Mazingaro> I did that: Error: Thumb load/store multiple does not support {reglist}^ -- `stmdb r0!,{r4-r11}^'
2012-09-26T00:37:45 < TitanMKD> in fact it is easier on CortexM as there's only main stack & process stack
2012-09-26T00:37:46 < Thorn> stmdb.w?
2012-09-26T00:38:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-26T00:39:55 < TitanMKD> you can do that
2012-09-26T00:39:56 < TitanMKD> LDMFD	LR, {R0-R14}^
2012-09-26T00:40:08 < TitanMKD> for restore context
2012-09-26T00:40:20 < TitanMKD> and also restore return addr
2012-09-26T00:40:26 < TitanMKD> then
2012-09-26T00:40:26 < TitanMKD> SUBS	PC, LR, #4
2012-09-26T00:40:46 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-26T00:40:53 < TitanMKD> and to save the context do
2012-09-26T00:40:54 < TitanMKD> STMDB	LR,{R0-LR}^
2012-09-26T00:41:41 < Mazingaro> why use lr
2012-09-26T00:41:44 < Mazingaro> ?
2012-09-26T00:42:21 < TitanMKD> lr=r14
2012-09-26T00:42:25 < TitanMKD> you can also write
2012-09-26T00:42:40 < TitanMKD> stmdb r14, {r0-r14}^
2012-09-26T00:43:52 < TitanMKD> anyway write ContextSave & ContextRestore code and test it with svc/swi
2012-09-26T00:44:37 < TitanMKD> in the SVCHandler add ContextSave then ContextRestore and you shall return just after the svc instruction without trashing anything
2012-09-26T00:45:22 < Mazingaro> Error: Thumb load/store multiple does not support {reglist}^ -- `stmdb lr,{r0-r14}^'
2012-09-26T00:45:49 < TitanMKD> you should use GCC ARM
2012-09-26T00:45:51 < TitanMKD> it works
2012-09-26T00:45:57 < Mazingaro> I'm using it
2012-09-26T00:45:58 < TitanMKD> in thumb mode
2012-09-26T00:46:03 < Mazingaro> in thumb mode
2012-09-26T00:47:25 < TitanMKD> very strange
2012-09-26T00:47:32 < TitanMKD> maybe you gcc is too old
2012-09-26T00:48:02 < Mazingaro> gcc version 4.6.3 (Sourcery CodeBench Lite 2012.03-56)
2012-09-26T00:48:09 < TitanMKD> try this one
2012-09-26T00:48:09 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Read error: Connection reset by peer]
2012-09-26T00:48:09 < TitanMKD> https://launchpad.net/gcc-arm-embedded
2012-09-26T00:48:18 < TitanMKD> it is an official version and it works
2012-09-26T00:48:29 < TitanMKD> Sourcey is just c****
2012-09-26T00:48:57 < Mazingaro> I'm on linux box
2012-09-26T00:49:27 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-09-26T00:49:39 < TitanMKD> there's linux version too
2012-09-26T00:49:54 < TitanMKD> https://launchpad.net/gcc-arm-embedded/4.6/4.6-2012-q2-update/+download/gcc-arm-none-eabi-4_6-2012q2-20120614.tar.bz2
2012-09-26T00:52:15 < zyp> TitanMKD, I'm not sure I understand you correctly, what do you say the purpose of the ^ is?
2012-09-26T00:52:39 < TitanMKD> zyp in fact it seems it's not really for CortexM3
2012-09-26T00:52:54 < TitanMKD> as i use real ARM cpu i forgot Mx are more basic about that
2012-09-26T00:53:08 < zyp> right, that's what I thought
2012-09-26T00:53:21 < zyp> cortex-m doesn't have banked registers except for ps
2012-09-26T00:53:27 < Mazingaro> ^ could not be used on M3
2012-09-26T00:53:38 < TitanMKD> so this stuff can be used
2012-09-26T00:53:39 < TitanMKD> http://www.coactionos.com/embedded-design/36-context-switching-on-the-cortex-m3.html
2012-09-26T00:57:31 < TitanMKD> Mazingaro read that http://infocenter.arm.com/help/topic/com.arm.doc.ddi0337e/DDI0337E_cortex_m3_r1p1_trm.pdf
2012-09-26T00:57:40 < TitanMKD> §5.11 Setting up multiple stacks
2012-09-26T00:57:49 < TitanMKD> very interesting for what you want to do
2012-09-26T00:57:57 < Mazingaro> I read that ;(
2012-09-26T00:58:18 < Mazingaro> and I'm trying to implement that
2012-09-26T00:58:34 < TitanMKD> Mazingaro you can also look at FreeRtos for CortexM
2012-09-26T00:59:03 < Mazingaro> I know freertos, but I need to develop my RTOS I could not use third party software
2012-09-26T00:59:44 < TitanMKD> like here
2012-09-26T00:59:44 < TitanMKD> http://freertos.svn.sourceforge.net/viewvc/freertos/trunk/FreeRTOS/Source/portable/GCC/ARM_CM3/
2012-09-26T00:59:57 < Mazingaro> yes
2012-09-26T01:01:04 < TitanMKD> look this
2012-09-26T01:01:04 < TitanMKD> void xPortPendSVHandler( void )
2012-09-26T01:01:22 < TitanMKD> it save context (to TCB)
2012-09-26T01:01:52 < TitanMKD> then schedule/switch context for the task to run => vTaskSwitchContext
2012-09-26T01:01:56 < TitanMKD> then restore context
2012-09-26T01:02:43 < TitanMKD> sorry for confusion with ARM 32bits code as i'm more familiar with them
2012-09-26T01:04:57 -!- dekar [~dekar@drms-590c4ee4.pool.mediaWays.net] has joined ##stm32
2012-09-26T01:05:01 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-26T01:09:19 < Mazingaro> np
2012-09-26T01:09:20 < Mazingaro> tx
2012-09-26T01:11:56 < TitanMKD> all is very well explained here
2012-09-26T01:11:56 < TitanMKD> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0338g/ch01s02s01.html
2012-09-26T01:12:07 < TitanMKD> Programmers model chapter
2012-09-26T01:13:17 < TitanMKD> bye
2012-09-26T01:13:22 < Mazingaro> bye tx
2012-09-26T01:13:28 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-09-26T01:13:32 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has joined ##stm32
2012-09-26T01:22:15 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Read error: Operation timed out]
2012-09-26T01:39:09 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-09-26T01:40:00 -!- dekar [~dekar@drms-590c4ee4.pool.mediaWays.net] has quit [Quit: This computer has gone to sleep]
2012-09-26T01:41:07 -!- BjoernC [~BjoernC@dyndsl-178-142-051-202.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-26T01:46:34 -!- Netsplit over, joins: @ChanServ
2012-09-26T01:46:34 < Mazingaro> please, why mrs r0, psp in SysTick_Handler jams the cp?
2012-09-26T01:46:40 < Mazingaro> cpu=
2012-09-26T01:46:41 < Mazingaro> ?
2012-09-26T01:48:05 -!- qyx is now known as qyx_
2012-09-26T01:53:33 < Mazingaro> ok found that
2012-09-26T01:57:32 -!- claude_work is now known as claude
2012-09-26T01:59:10 -!- esden [esden@repl.esden.net] has quit [Ping timeout: 246 seconds]
2012-09-26T02:01:25 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-09-26T02:07:05 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-26T02:16:42 < dongs> https://www.youtube.com/watch?v=BMsLArefSOw attn @ all
2012-09-26T02:28:34 < emeb> the stupid. it burnssss.
2012-09-26T02:35:22 -!- DaKu is now known as daku
2012-09-26T02:36:03 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-26T02:49:49 < dongs> http://essentialscrap.com/dsoquad/pawn.html
2012-09-26T02:49:49 < dongs> lol
2012-09-26T03:05:02 -!- oPossum5151 is now known as oPossum5150
2012-09-26T03:34:54 < dongs> where is the blog
2012-09-26T03:37:19 < dongs> http://www.theregister.co.uk/2012/09/26/nypd_iphone_theft_statistics/
2012-09-26T03:39:35 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-26T03:39:35 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-26T03:39:35 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-26T03:39:36 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-26T03:40:45 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 268 seconds]
2012-09-26T03:41:22 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-26T04:03:06 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 264 seconds]
2012-09-26T04:30:43 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-26T05:04:32 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-26T05:42:55 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-26T05:47:59 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-26T05:48:15 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-26T05:54:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 264 seconds]
2012-09-26T06:00:55 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-26T06:26:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-26T06:26:41 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-26T06:33:20 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-26T06:33:21 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-26T06:33:21 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-26T06:36:13 < Bird|lappy> quickie: is there a specific part you folks use for the 10pin CM3 SWD connector?
2012-09-26T06:39:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-26T06:52:12 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-26T06:56:59 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-26T07:05:44 < dongs> yeah
2012-09-26T07:06:03 < dongs> S9012E-05-ND
2012-09-26T07:06:09 < dongs> cheapest unshrouded
2012-09-26T07:06:15 < dongs> zyp has a shrouded version somewehre.
2012-09-26T07:33:28 < emeb_mac> amazing how hard it can be to find stuff like that in their online catalog
2012-09-26T07:35:42 < Bird|lappy> yeah
2012-09-26T07:36:04 < Bird|lappy> all the shrouded parts I can come up with don't accommodate the keying "hump" on the side of the plug it seems?
2012-09-26T07:37:56 < emeb_mac> I've got a link to the right shrouded header but it's in the bookmarks on my other computer...
2012-09-26T07:38:27 < emeb_mac> FCI part IIRC
2012-09-26T07:40:45 < Bird|lappy> ok
2012-09-26T07:41:00 < Bird|lappy> is it this one? http://www.digikey.com/product-detail/en/20021521-00010T1LF/609-4054-ND/2414951
2012-09-26T07:41:45 < Bird|lappy> also: what are the benefits of a grounded case on a tact (pushbutton) switch?
2012-09-26T07:44:16 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-26T07:44:25 < emeb> here you go: http://www.digikey.com/product-search/en?x=17&y=13&lang=en&site=us&KeyWords=609-4054-ND%09
2012-09-26T07:45:21 < Bird|lappy> yepyep, same part
2012-09-26T07:45:25 < Bird|lappy> as what I just linked
2012-09-26T07:58:52 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-26T08:13:21 < jpa-> Oldboy: then it was a tear down of something totally else
2012-09-26T08:13:38 < jpa-> Oldboy: DS203 / DSO Quad definitely has ice65L04
2012-09-26T08:15:07 < jpa-> Oldboy: dso quad looks like this inside: http://koti.kapsi.fi/jpa/stuff/pix/DSO_Quad_guts.jpg   and like this outside   http://essentialscrap.com/dsoquad/rot_1_1.jpg
2012-09-26T08:25:53 < emeb_mac> looks pretty nice.
2012-09-26T08:26:16 < jpa-> it's not bad
2012-09-26T08:26:24 < jpa-> the software is
2012-09-26T08:27:42 < jpa-> there is open-source fork based on 2.51 in which basic scope stuff (triggering etc.) works nicely, and then there is new "P100" version that has much nicer user interface but triggering etc. doesn't work perfectly.. and they didn't release source code for P100
2012-09-26T08:30:40 < emeb_mac> meh
2012-09-26T08:30:55 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-26T09:00:34 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-26T09:02:42 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-26T09:08:32 < Oldboy> jpa-: this is the teardown I was watching: http://www.youtube.com/watch?v=CJv2iCieeeM
2012-09-26T09:10:49 < Oldboy> its not a DSO Quad.
2012-09-26T09:11:30 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 264 seconds]
2012-09-26T09:12:53 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-26T09:16:12 < jpa-> ahh haha, i watched that also yesterday
2012-09-26T09:16:17 < jpa-> it's total crap
2012-09-26T09:16:30 < jpa-> software even worse than dso quad and hardware 10x worse
2012-09-26T09:21:16 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 245 seconds]
2012-09-26T10:34:00 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-09-26T10:35:15 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-26T10:37:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-26T10:43:54 -!- dekar [~dekar@drms-590c4ee4.pool.mediaWays.net] has joined ##stm32
2012-09-26T10:43:57 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-26T11:03:27 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-26T11:20:02 -!- dekar [~dekar@drms-590c4ee4.pool.mediaWays.net] has quit [Quit: This computer has gone to sleep]
2012-09-26T11:33:21 < dongs> DIY.. er.. TOOL
2012-09-26T11:33:51 < dongs> is this another of those mp3 / mp4 players rebadged as scope
2012-09-26T11:42:20 < jpa-> are there more than just that one?
2012-09-26T11:42:29 < jpa-> but yeah, qdso is just that utter crap
2012-09-26T11:42:50 < jpa-> i wonder where the "Q" comes from as it is just single-channel
2012-09-26T12:08:31 -!- BjoernC [~BjoernC@dyndsl-178-142-057-145.ewe-ip-backbone.de] has joined ##stm32
2012-09-26T12:12:33 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-26T12:19:56 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-09-26T12:19:57 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-26T12:30:21 < dongs> maybe its Qute
2012-09-26T12:37:58 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Read error: Operation timed out]
2012-09-26T12:45:07 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-26T12:45:08 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-26T12:45:08 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-26T13:04:11 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-09-26T13:17:18 < karlp> so, what should I get from mouser today? work's buying...
2012-09-26T13:19:22 < Thorn> karlp: ask them for this http://electronics.stackexchange.com/questions/42079/processor-running-c-natively
2012-09-26T13:20:04 < Mazingaro> hi
2012-09-26T13:20:16 < karlp> Thorn: hehe :)
2012-09-26T13:20:30 < Mazingaro> zyp: my RTOS seems to work properly... I have to add the fp stack handling too
2012-09-26T13:20:31 < karlp> going to get some 802.15.4 radios, as mouser actually obeys the laws and ships them,
2012-09-26T13:20:37 < karlp> not sure what else to try and snapple up
2012-09-26T13:23:27 < dongs> > 4. what happens when C loses popularity
2012-09-26T13:23:28 < dongs> lool
2012-09-26T13:23:33 < dongs> he must be one of those python/ruby newfags
2012-09-26T13:24:38 < Thorn> if C loses popularity, there would be much rejoicing. it would mean a better replacement is available
2012-09-26T13:24:51 < Mazingaro> is anyone interested in implment a lisp machine on stm32??? lol
2012-09-26T13:25:26 < dongs> no.
2012-09-26T13:25:28 < Thorn> I mean something that can be compiled to machine code, such as ada
2012-09-26T13:25:39 < dongs> Thorn: lets make a new language, call it AIDS
2012-09-26T13:26:02 < Thorn> gaycc -c main.gay
2012-09-26T13:27:10 < Thorn> someone did want to add cortex-m support to gnat btw.
2012-09-26T13:27:32 < dongs> Ghana National Association of Teachers ?
2012-09-26T13:28:04 < Thorn> the only freely available ada compiler
2012-09-26T13:28:12 < dongs> you know why right
2012-09-26T13:28:28 < dongs> cuz nobody uses that shit
2012-09-26T13:30:05 < karlp> I havent used it since uni, and I do not miss it one bit
2012-09-26T13:30:08 < Thorn> popularity of a technology seems to be inversely dependent on its merits, cf. php, mysql, arduino, etc.
2012-09-26T13:33:17 < dongs> haha
2012-09-26T13:33:19 < dongs> so fucking true
2012-09-26T13:33:24 < dongs> arduino is utterly fucking worthless
2012-09-26T13:33:40 < karlp> not that far, it does actually work.
2012-09-26T13:33:58 < dongs> for a pretty narrow definition of "workR"
2012-09-26T13:34:04 < dongs> er,  "work".
2012-09-26T13:34:10 < Thorn> technologies that allow you to quickly hack some crap togerher are the winners
2012-09-26T13:34:47 < jpa-> ada is not bad
2012-09-26T13:35:01 < zyp> Thorn, even though you are winning special olympics, you are still retarded
2012-09-26T13:35:04 < jpa-> it's strict, and a bit clumsy, but it has its uses
2012-09-26T13:36:17 < karlp> zyp: except that quickly hacking together crap that from the outside is indistuinguishable from the others doesn't make it special olympics
2012-09-26T13:37:04 < karlp> bleh, power inductors with 4 ohms resistance
2012-09-26T13:37:05 < karlp> no thanks
2012-09-26T13:37:33 < zyp> are you saying that quickly hacked together stuff doesn't have flaws that properly hacked stuff would avoid?
2012-09-26T13:46:34 < cjbaird> needs moar liboli.a linkage
2012-09-26T13:48:59 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-26T13:49:08 -!- Thorn [~Thorn@78-106-223-25.broadband.corbina.ru] has joined ##stm32
2012-09-26T13:49:08 -!- Thorn [~Thorn@78-106-223-25.broadband.corbina.ru] has quit [Changing host]
2012-09-26T13:49:08 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-26T14:13:23 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has joined ##stm32
2012-09-26T14:21:54 < Laurenceb> can anyone help me with git?
2012-09-26T14:22:36 < Laurenceb> git pull origin master From github.com:Laurenceb/Chibi-Spectro
2012-09-26T14:22:36 < Laurenceb>  * branch            master     -> FETCH_HEAD
2012-09-26T14:22:36 < Laurenceb> Updating 56e0fd8..4a43dac
2012-09-26T14:22:36 < Laurenceb> error: Untracked working tree file '.dep/can_lld.o.d' would be overwritten by merge.  Aborting
2012-09-26T14:22:54 < zyp> rm .deb/can_lld.o.d
2012-09-26T14:23:10 < zyp> also, put shit like that into .gitignore
2012-09-26T14:23:25 < zyp> and fucking don't commit it ever.
2012-09-26T14:24:00 < Thorn> I've seen github repos full of .o files
2012-09-26T14:24:14 < zyp> that's probably what Laurenceb have
2012-09-26T14:27:59 < dongs> haha
2012-09-26T14:32:48 < Laurenceb> hmm
2012-09-26T14:32:55 < Laurenceb> chibios is faulting in halinit
2012-09-26T14:33:28 < dongs> more like failinit()
2012-09-26T14:33:38 < dongs> have you switched to CoOS and COX yet
2012-09-26T14:34:06 < Laurenceb> nope
2012-09-26T14:34:10 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-26T14:34:14 < dongs> you should start
2012-09-26T14:36:43 < zyp> dongs, will you be his support guy then?
2012-09-26T14:36:47 < zyp> :p
2012-09-26T14:38:09 < dongs> no
2012-09-26T14:38:09 < dongs> that shit doesnt need support
2012-09-26T14:38:09 < dongs> it fucking works
2012-09-26T14:40:30 < zyp> ha
2012-09-26T14:40:42 < zyp> remember that we're talking about Laurenceb
2012-09-26T14:41:35 < BrainDamage> "instructions not clear, got my dick stuck in hal api"
2012-09-26T14:41:46 < dongs> one of my led reels is trackable
2012-09-26T14:41:58 < dongs> i picked up 2 1k rgb led reels for $45/ea
2012-09-26T14:42:18 < zyp> then you'll have two kilopixels!
2012-09-26T14:43:18 < dongs> i really wanna do liek a 16x16 panel of htem but i have a feeling shit will get kinda insane
2012-09-26T14:43:34 < Laurenceb> hmf
2012-09-26T14:43:39 < Laurenceb> its totally screwed
2012-09-26T14:44:00 < dongs> Laurenceb: run your code through indent -kr -nut -l400
2012-09-26T14:44:18 < Laurenceb> ill try the demo code one more time
2012-09-26T14:44:20 < dongs> and it will work
2012-09-26T14:44:22 < Laurenceb> it worked last time
2012-09-26T14:44:31 < Laurenceb> diff doesnt show up anything relevent
2012-09-26T14:44:34 < dongs> I got led blinking in a task with freeRTOS without even looking at thier demos
2012-09-26T14:44:51 < dongs> in about 30 minutes
2012-09-26T14:44:57 < dongs> and without using freetard tools
2012-09-26T14:45:25 < dongs> but freertos sucks.
2012-09-26T14:45:27 < dongs> vFUCKYOU
2012-09-26T14:45:40 < Laurenceb> what the fuuuuuuuuuuu
2012-09-26T14:45:52 < Laurenceb> ../../../os/hal/platforms/STM32/OTGv1/usb_lld.h:182:2: error: #error "the USB OTG driver requires a 48MHz clock"
2012-09-26T14:45:52 < Laurenceb> make: *** [build/obj/test.o] Error 1
2012-09-26T14:46:01 < dongs> yeah..
2012-09-26T14:46:20 < BrainDamage> dongs: https://makezineblog.files.wordpress.com/2012/02/peggy2.jpg
2012-09-26T14:46:39 < dongs> BrainDamage: what the fuck is it.
2012-09-26T14:46:51 < dongs> someone was really fucking bored if the back of that is leds.
2012-09-26T14:47:13 < BrainDamage> some genious took a matrix led display, and replaced the leds with ribbon cables to extend it ...
2012-09-26T14:47:37 < BrainDamage> at least take the matrix coordinate wires, not the fucking led pairs
2012-09-26T14:48:09 < dongs> thats not a rgb matrix
2012-09-26T14:48:31 < BrainDamage> it isn't
2012-09-26T14:48:46 < BrainDamage> oh wait, it is
2012-09-26T14:48:55 < BrainDamage> http://www.evilmadscientist.com/2008/peggy-version-2-0/
2012-09-26T14:49:23 < Laurenceb> arg what
2012-09-26T14:49:32 < Laurenceb> they swapped to olimex boards
2012-09-26T14:49:36 < Laurenceb> so much fail
2012-09-26T14:49:44 < Laurenceb> this is like fail/0
2012-09-26T14:50:32 < dongs> http://hackaday.com/2009/05/06/peggy-2-super-pixels/ lol
2012-09-26T14:51:19 < dongs> BrainDamage: its not rgb
2012-09-26T14:51:48 < BrainDamage> well, just 1 rgb pixel is made by multiple leds :p
2012-09-26T14:51:58 < dongs> yeah
2012-09-26T14:52:06 < BrainDamage> awesome resolution ...
2012-09-26T14:53:39 < BrainDamage> in other news, for the second time in 2 years, one of the trains I take regularry had a spectacular crash
2012-09-26T14:53:51 < BrainDamage> and I didn't take the crashed one by pure luck
2012-09-26T14:56:43 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-09-26T14:56:47 < Laurenceb> can anyone help me with this?
2012-09-26T14:56:57 < Laurenceb> anyone got an f4discovery + chibios set up?
2012-09-26T14:57:07 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro
2012-09-26T14:58:18 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 264 seconds]
2012-09-26T14:58:20 < dongle> fucking cock
2012-09-26T14:58:23 < dongle> fuck feenode
2012-09-26T14:58:30 < dongle> http://i.imgur.com/Dy1em.jpg lol.
2012-09-26T14:58:41 < dongle> also: im redoing that peggy shit with rgb leds
2012-09-26T14:58:45 < dongle> and stm32., of course
2012-09-26T14:58:55 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-26T14:59:09 < karlp> what's wrong with those jumbo sample boards?
2012-09-26T14:59:31 < dongle> karlp: nothing, they're orders from some customer
2012-09-26T14:59:52 < dongle> easiest pcb artwork evar
2012-09-26T15:00:01 < karlp> hang on, smart isn't just handing them out?
2012-09-26T15:00:10 < karlp> or did smart ask you to make those boards for them?
2012-09-26T15:00:13 < dongle> the latter.
2012-09-26T15:00:17 < karlp> ahh, ok :)
2012-09-26T15:00:30 < BrainDamage> doesn't look like a bad idea
2012-09-26T15:00:56 < BrainDamage> looks good and it's a neat way to display them
2012-09-26T15:01:48 < dongle> http://arxiv.org/pdf/1203.1895v1.pdf
2012-09-26T15:05:40 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds]
2012-09-26T15:05:48 < cjbaird> http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0044111
2012-09-26T15:06:12 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-26T15:06:36 < karlp> huh, gnome-terminal exploded trying to open that link cjbaird
2012-09-26T15:06:41 < karlp> pastes just fine...
2012-09-26T15:06:42 < dongle> haha
2012-09-26T15:06:49 < dongle> thats lunix for you
2012-09-26T15:06:50 < dongle> in 2012.
2012-09-26T15:07:02 < karlp> meh, it's in the link detection and auto converting to links.
2012-09-26T15:09:01 < dongle> cjbaird: some high quality Laurenceb material there
2012-09-26T15:09:28 < cjbaird> Hmm, gnome-terminal instantly coredumps here..
2012-09-26T15:09:36 < dongle> cjbaird: on that link?
2012-09-26T15:09:42 < karlp> mine just pops an error saying it cna't open it.
2012-09-26T15:09:46 < karlp> had to manually paste it.
2012-09-26T15:09:57 < dongle> gonna join #gnome and paste it.
2012-09-26T15:10:01 < karlp> hehe
2012-09-26T15:11:41 < cjbaird> g-t gets a SIGABRT assertion during initialization.. Wouldn't be surprised if it were barfing on some old config file
2012-09-26T15:12:02 < karlp> oh, you mean just _starting_ g-t?
2012-09-26T15:12:24 < cjbaird> yup
2012-09-26T15:13:00 < cjbaird> prob this: https://bugzilla.redhat.com/show_bug.cgi?id=720119
2012-09-26T15:13:16 < dongs> 720k bugs in redhat?
2012-09-26T15:14:40 < cjbaird> How many are still Open, though.
2012-09-26T15:14:49 < dongle> how many are closed, wontfix
2012-09-26T15:14:56 < dongle> because the "develoeprs" are fucking jerks
2012-09-26T15:15:36 < karlp> clever, duped the old bug to a newer bug
2012-09-26T15:16:43 < karlp> cjbaird: https://bugzilla.redhat.com/show_bug.cgi?id=755992#c51 has the workaround
2012-09-26T15:17:02 < karlp> well, for some people
2012-09-26T15:17:03 < karlp> bleh
2012-09-26T15:17:31 < karlp> dongs, you'll love this: between two versions of linux, the order of the serial port and the gpio driver being initialized has switched.
2012-09-26T15:17:52 < karlp> so my admittedly ugly hack for stealing a gpio for rs485 no longer works as gpio hasn't been setup yet
2012-09-26T15:17:58 < dongle> karlp: thats the reason for crashage?
2012-09-26T15:17:59 < dongle> oh
2012-09-26T15:18:01 < karlp> I'm so thrilled
2012-09-26T15:18:04 < dongle> on crapberrypy?
2012-09-26T15:18:15 < karlp> no, work board
2012-09-26T15:18:18 < dongle> ic
2012-09-26T15:18:24 < dongle> awesome.
2012-09-26T15:18:28 < karlp> indeed.
2012-09-26T15:18:39 < dongle> I gotta port some lunix driver for shit I sell from 2.6 to 3.0
2012-09-26T15:18:46 < dongle> apparently so much shit changes the custoemr cant figure out waht hte fuck to do
2012-09-26T15:18:57 < karlp> yeah, this is somewhere between 2.6.30 and 3.3.x
2012-09-26T15:19:06 < dongle> christ, shit is on 3.3 now?
2012-09-26T15:19:12 < karlp> but hey, the wireless driver doesn't occasioanlly wander off into lala land.
2012-09-26T15:19:15 < karlp> 3.5 iirc
2012-09-26T15:19:28 < karlp> they decided to stop dicking around with 2.6.30.1012
2012-09-26T15:19:33 < karlp> and just use less numbers
2012-09-26T15:19:34 < cjbaird> Okay, g-t up and running (had to restart gconfd as well). I'm not seeing the fun with pasting the link, though. :)
2012-09-26T15:20:48 < karlp> ctrl-click the link doe sthis for me: http://palmtree.beeroclock.net/~karlp/screenshot-gt-exploding.png
2012-09-26T15:21:53 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-26T15:26:23 < dongle> oh
2012-09-26T15:26:27 < dongle> peggyshit is not dimmable
2012-09-26T15:27:55 < dongle> http://s3.amazonaws.com/evilmadscientist/source/peggy2_GrayTest.pde oh fuck is this some laurenceb code
2012-09-26T15:30:16 < qyx_> O_o
2012-09-26T15:30:17 < qyx_> 24x if
2012-09-26T15:30:57 < cjbaird> Video generation? Might explain it.
2012-09-26T15:52:51 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-26T15:53:56 < dongle> http://whitequark.org/blog/2012/09/25/why-raspberry-pi-is-unsuitable-for-education/
2012-09-26T15:55:58 < cjbaird> Unfortunately, it just rehases the same-old idea that "not 100% stallman-approved == not good for education!!!111" ... The Z80 and 6502 weren't open, either.
2012-09-26T15:56:27 < Bird|lappy> grrr, Xilinx...
2012-09-26T15:56:46 < Bird|lappy> making your MSL data hard to get is a bad idea, mmkay?
2012-09-26T15:57:21 < cjbaird> Someone instead really need to lay the boot into how the RPi Foundation are giving too much attention to the Manchild 'Maker' market. "Hey guise! We're flying off overseas for another manchild event!"
2012-09-26T15:58:26 < Bird|lappy> ...I wonder if a 100TQFP fits under a 40PDIP
2012-09-26T15:58:35 < dongle> no
2012-09-26T15:58:38 < Bird|lappy> darn :p
2012-09-26T15:58:46 < Bird|lappy> would have to go to BGA to pull that one off
2012-09-26T15:58:49 < dongle> i think 100 is 14x14
2012-09-26T15:58:55 < dongle> typical 0.5mm shit anyway
2012-09-26T16:01:47 < Bird|lappy> heh.  one of my class projects is an 8086 board, and all my glue logic got vacuumed up by an XC95144XL (well, save for a few inverters, an output latch, and a couple of flops, but the latch is there due to drive requirements and the flops are the reset-release synchronizer for the CPLD)
2012-09-26T16:04:18 < dongle> what the hell., ST has a USB IC that has built in 33R resistors and 1.5K pullup
2012-09-26T16:05:03 < Bird|lappy> dongle, not terribly surprising.  I wonder if it has on-board soft-enumeration support
2012-09-26T16:05:40 < cjbaird> I go to replace a blown bulb in the scooter's headlight assembly, and discover the entire thing is only being held on by half a busted mounting.. :/ http://i.imgur.com/ikfvK.jpg
2012-09-26T16:05:46 < Bird|lappy> :<
2012-09-26T16:07:15 < Bird|lappy> anyone here ordered from Avnet btw?
2012-09-26T16:07:51 < Bird|lappy> going to have to put a small order in with them soon, but don't know if they have any sort of minimum order amount or the likes
2012-09-26T16:07:57 < dongle> USBUF01P6
2012-09-26T16:08:00 < dongle> Bird|lappy: lol no
2012-09-26T16:10:00 < dongle> tiny little fucker
2012-09-26T16:10:36 < Bird|lappy> :) turns out no they don't
2012-09-26T16:10:40 < Bird|lappy> \o/
2012-09-26T16:11:49 < Bird|lappy> got to order 10 of a rather old NOR Flash that's going EOL in 2013, and 1 of a voltage translator chip that nobody else wants to stock in gull-wing packages
2012-09-26T16:13:58 < Laurenceb> 0x08003d46 in dmaInit ()
2012-09-26T16:13:59 < Laurenceb>     at ../ChibiOS/os/hal/platforms/STM32F4xx/stm32_dma.c:429
2012-09-26T16:13:59 < Laurenceb> 429	    _stm32_dma_streams[i].stream->CR = 0;
2012-09-26T16:13:59 < Laurenceb> (gdb) p i
2012-09-26T16:13:59 < Laurenceb> Cannot access memory at address 0x1fffef2c
2012-09-26T16:14:01 < Laurenceb> ^wtf
2012-09-26T16:16:24 < Laurenceb> its clusterfucked
2012-09-26T16:22:01 < karlp> snalu?
2012-09-26T16:23:19 -!- daku is now known as DaKu
2012-09-26T16:23:21 < karlp> huh, getting upset about jazelle extensions on rpi is pretty absurd.
2012-09-26T16:23:29 < dongle> haha
2012-09-26T16:23:36 < dongle> yeah i thought that was kinda odd too
2012-09-26T16:23:41 < dongle> its not like anyone acutalyl uses that shit for years
2012-09-26T16:23:57 < karlp> it's faster ot use a decent modern jit, rather than give the bytecode to jazelle
2012-09-26T16:24:06 < dongle> jizzelle
2012-09-26T16:24:44 < dongle> ive ordered from avnet
2012-09-26T16:24:47 < dongle> their shipping sucks
2012-09-26T16:24:50 < dongle> too expensive.
2012-09-26T16:25:06 < dongle> but they had some gennum ASI serializer shits I needed that noone else stocks.
2012-09-26T16:25:19 < dongle> and gennum fucks wouldnt sell to me direct.
2012-09-26T16:25:32 < dongle> "hi guys im trying to buy your shit"
2012-09-26T16:25:36 < dongle> "fuck off, we're not interested"
2012-09-26T16:25:39 < karlp> doncha love that :)
2012-09-26T16:25:44 < dongle> yeah. its hilarious.
2012-09-26T16:25:50 < karlp> see broadcom :)
2012-09-26T16:26:03 < dongle> broadcom is a whole different pile of stupid
2012-09-26T16:26:09 < dongle> tehy wont even talk to you unless youre buying 100k of somesthing
2012-09-26T16:26:24 < dongle> marvell too
2012-09-26T16:26:31 < dongle> i had some project wiht marvell ethernet phy
2012-09-26T16:26:37 < dongle> (why the fuck woudl that need NDA anyway)
2012-09-26T16:26:46 < dongle> (marvell shit was specified by customer)
2012-09-26T16:27:07 < dongle> so I was like.. public datasheet is enough but i might need the nDA'd one.... contacted them..
2012-09-26T16:27:30 < dongle> they're like.. sorry.. we cant give you the nda or the data.. basically fuck off.
2012-09-26T16:27:34 < dongle> wtf.
2012-09-26T16:28:22 < dongle> copied design off some opensores plugcomputer shite, it worked, moved on
2012-09-26T16:28:32 < dongle> but still fucking silly. will never use marvel shit for anything. obviously.
2012-09-26T16:28:38 < karlp> complaining that it's a 3v3 device?! yeah, fuck that article off
2012-09-26T16:30:33 < karlp> that guy wrote quite a bit of early txane/stlink code by the way
2012-09-26T16:30:37 < karlp> fwiw
2012-09-26T16:30:43 < dongle> I don't usually test cars, but when I do.. http://www.youtube.com/watch?v=Fu5HfHpf3mI
2012-09-26T16:30:47 < dongle> which guty?
2012-09-26T16:30:49 < dongle> guy
2012-09-26T16:30:52 < dongle> the one bitching about shitberry?
2012-09-26T16:30:55 < karlp> whitequark
2012-09-26T16:30:56 < karlp> yeah
2012-09-26T16:31:11 < dongle> weird.
2012-09-26T16:32:19 < dongle> http://whitequark.org/blog/2011/10/07/my-first-factory-made-pcbs/
2012-09-26T16:32:20 < dongle> wut
2012-09-26T16:32:53 < dongle> > 27 Jan 2012 update: the boards have arrived and I.ve assembled them, but they have some nasty crosstalk bugs and are unfortunately unusable. http://whitequark.org/images/tft-lvds/pcbs-small.jpeg
2012-09-26T16:33:02 < dongle> NO YOU THINK
2012-09-26T16:33:12 < zyp> so, how long until hobbyist troubles will be 3.3V vs 1.8V instead of 5V vs 3.3V?
2012-09-26T16:33:20 < Laurenceb> 27 Jan 2012 update: the boards have arrived and I’ve assembled them, but they have some nasty crosstalk bugs and are unfortunately unusable. DO NOT USE THIS DESIGN
2012-09-26T16:33:23 < Laurenceb> looolll
2012-09-26T16:33:31 < karlp> zyp, probably not for a few more years
2012-09-26T16:33:49 < karlp> because most of the 1.8v stuff I've seen is still 3v3 specced
2012-09-26T16:33:50 < dongle> Laurenceb: looking at the picture, im NOT FUCKING SURPRISED
2012-09-26T16:33:59 < zyp> :)
2012-09-26T16:34:02 < karlp> but there's plenty of stuff tha tis NO MORE THAN 3v3 EVER!
2012-09-26T16:36:30 < dongle> also: that usb shite from ST is called IPAD, lol.
2012-09-26T16:36:33 < dongle> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00003596.pdf
2012-09-26T16:36:50 < dongle> with a (TM) even.
2012-09-26T16:37:26 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-26T16:37:26 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-26T16:37:26 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-26T16:37:29 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-26T16:37:29 < Laurenceb> CAPITALS
2012-09-26T16:38:02 < Laurenceb> why is chibios failing?
2012-09-26T16:38:03 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-26T16:38:22 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-26T16:39:45 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-26T16:41:42 < Laurenceb> http://olimex.wordpress.com/2012/09/25/a13-olinuxino-running-debian-preview/
2012-09-26T16:41:46 < Laurenceb> fapfapfap
2012-09-26T16:43:40 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-26T16:49:40 < Laurenceb> lol
2012-09-26T16:49:44 < Laurenceb> gcc was failing
2012-09-26T16:51:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-26T16:51:52 < Laurenceb> http://www.google.com/trends/?q=+linux&ctab=0&geo=all&date=all&sort=0
2012-09-26T16:51:55 < Laurenceb> ^oh noes
2012-09-26T16:53:29 < dongs> of course its dying
2012-09-26T16:55:53 < dongs> wow, 2007 must  NOT have been a year of lunix desktop
2012-09-26T16:56:02 < dongs> the news volume totally  fucking tanked
2012-09-26T16:56:05 < dongs> for entire year
2012-09-26T16:56:06 < dongs> i wonder why hmm
2012-09-26T16:56:23 < Laurenceb> http://www.ifixit.com/Teardown/iPhone-5-Teardown/10525/2
2012-09-26T16:56:33 < dongs> emeb_mac: http://i.imgur.com/Dy1em.jpg
2012-09-26T16:56:38 < Laurenceb> do they really use a "3.8v"  lipo?
2012-09-26T16:56:55 < dongs> its liion
2012-09-26T16:56:58 < dongs> not lipo.
2012-09-26T16:57:08 < karlp> Laurenceb: there's a nicer teardown where they sent it to chipworks...
2012-09-26T16:57:16 < Laurenceb> cool
2012-09-26T16:57:19 < dongs> what compass IC are they using?
2012-09-26T16:57:26 < Laurenceb> "lilthium ion polymer"
2012-09-26T16:57:29 < Laurenceb> do hell
2012-09-26T16:57:30 < zyp> who cares what they write? the voltage varies from 4.2 to 3.5V during a use cycle
2012-09-26T16:57:34 < dongs> still AKM shit?
2012-09-26T16:57:41 < emeb_mac> look at that.
2012-09-26T16:57:43 < Laurenceb> i know
2012-09-26T16:58:02 < karlp> Laurenceb: http://www.ifixit.com/Teardown/Apple-A6-Teardown/10528/
2012-09-26T16:58:08 < emeb_mac> how thin are those cards dongs?
2012-09-26T16:58:26 < dongs> emeb_mac: 1mm
2012-09-26T16:59:06 < dongs> and 10x15cm in size.
2012-09-26T16:59:31 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-09-26T16:59:51 < Laurenceb> http://guide-images.ifixit.net/igi/IDA4xuyWFSXlmTwd.medium
2012-09-26T17:00:05 < Laurenceb> interesting comparison of component density between those two
2012-09-26T17:00:28 < karlp> what?
2012-09-26T17:01:45 < zyp> I like the component density
2012-09-26T17:02:22 < Laurenceb> The Apple A6—labeled APL0598 on the package marks and APL0589B01 on the inside—is fabricated by Samsung on their 32 nm CMOS process and measures 9.70 mm x 9.97 mm.
2012-09-26T17:02:27 < Laurenceb> thats one big die
2012-09-26T17:02:30 < karlp> dongs: L3G4200D is the gyro?
2012-09-26T17:02:53 < dongs> yea
2012-09-26T17:02:59 < dongs> which is funny..
2012-09-26T17:03:03 < dongs> cuz that gyro has been EOL for over a year
2012-09-26T17:03:22 < karlp> and LIS331DLH  for accel
2012-09-26T17:03:24 < zyp> maybe they're just using leftover stock
2012-09-26T17:03:34 < zyp> later L3Gs are pin compatible
2012-09-26T17:03:39 < Laurenceb> step11 is interesting
2012-09-26T17:03:40 < zyp> so direct swap
2012-09-26T17:05:29 < Laurenceb> interesting that they used a wifi module
2012-09-26T17:05:37 < Laurenceb> almost like ardutards :P
2012-09-26T17:05:56 < karlp> hehe: "While there hasn't been any official word on the pin-out for the Lightning connector, we can confidently say that two of them will be used for power and ground. "
2012-09-26T17:06:46 < Laurenceb> http://guide-images.ifixit.net/igi/6H5Dnqk1ZJkmwxXX.medium
2012-09-26T17:06:47 < emeb_mac> hmmm - with apple that might not be a safe bet
2012-09-26T17:06:51 < Laurenceb> thats some crazy shit
2012-09-26T17:07:06 < Laurenceb> on die LC stuff
2012-09-26T17:07:08 < dongs> what the fuck are those eshits
2012-09-26T17:07:15 < dongs> coils?
2012-09-26T17:07:20 < Laurenceb> yes
2012-09-26T17:07:34 < dongs> http://guide-images.ifixit.net/igi/6H5Dnqk1ZJkmwxXX.huge
2012-09-26T17:07:36 < emeb_mac> analog / rf chips always look odd.
2012-09-26T17:07:41 < dongs> also link to properly sized images, fgt
2012-09-26T17:11:10 < dongs> http://www.st.com/jp/com/MULTIMEDIA_RESOURCES/IMAGE/BOARD_PHOTO/image_steval-ill015v1.jpg i wonder which stm32 that is
2012-09-26T17:11:39 < Mazingaro> hi
2012-09-26T17:11:46 < Mazingaro> zippe: are you there?
2012-09-26T17:11:47 < zyp> dongs, I'm reading C6T6
2012-09-26T17:11:54 < dongle> me too, but does that even exist?
2012-09-26T17:12:16 < zyp> sure
2012-09-26T17:12:18 < dongle> heh their bom just says "microcontroller cortex, STM32F103Cx
2012-09-26T17:12:35 < dongle> oh hahah wow it does
2012-09-26T17:12:38 < dongle> super gheto flash size
2012-09-26T17:12:43 < dongle> and half the ram
2012-09-26T17:12:45 < dongle> never knew.
2012-09-26T17:13:07 < zyp> remember that C8/CB is medium-sized F1s
2012-09-26T17:13:22 < zyp> so of course you have to have something smaller then :p
2012-09-26T17:13:46 < dongle> wiht CBT6 being 2bucks/per i dont see the point of going lower
2012-09-26T17:14:17 < karlp> we're using C4 here, it was enough cheaper
2012-09-26T17:14:24 < zyp> not a point in paying for more space than you need in volume production
2012-09-26T17:14:24 < emeb_mac> interesting to see all those die with I/O structures interior.
2012-09-26T17:14:44 < karlp> C8 wsa cheaper than C6, but more than C4.
2012-09-26T17:14:49 < emeb_mac> when I was designing chips 8 years ago that was still fairly rare.
2012-09-26T17:14:53 < karlp> C6 seems to be the unwanted middle child
2012-09-26T17:15:29 < dongle> how do they break the shit with io inside the die, its upside down and wired to bga/wahtever pads?
2012-09-26T17:16:23 < emeb_mac> probably flip chip w/ balls directly on the die to a custom pkg substrate.
2012-09-26T17:16:47 < emeb_mac> but one of the x-rays you can see bondwires going to the center of the die too.
2012-09-26T17:18:29 < Laurenceb> heh i cant see a magno on the iphone5
2012-09-26T17:18:42 < dongle> Laurenceb: right
2012-09-26T17:18:45 < dongle> me neither.
2012-09-26T17:18:49 < dongle> and none of the sites mention waht it is.
2012-09-26T17:18:54 < Laurenceb> fail
2012-09-26T17:18:54 < dongle> there's gotta be one there though
2012-09-26T17:19:16 < Laurenceb> maybe its some custom st accel + magno
2012-09-26T17:19:18 < dongle> but apaprently all the previous ones were AKM
2012-09-26T17:19:20 < emeb_mac> where was I reading - someone mentioned a magno + gyro on one die.
2012-09-26T17:19:20 < dongle> no
2012-09-26T17:21:06 < Laurenceb> maybe they got the part wrong in teardown
2012-09-26T17:21:15 < Laurenceb> and its a accel+magno
2012-09-26T17:21:28 < dongle> yea possible
2012-09-26T17:21:38 < dongle> but the acc+mag shit is rectangular
2012-09-26T17:22:52 < Laurenceb> yeah
2012-09-26T17:22:56 < Laurenceb> maybe a custom ic
2012-09-26T17:23:08 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-26T17:23:08 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-26T17:23:15 < emeb_mac> I'm not convinced how useful magno is.
2012-09-26T17:23:26 < dongle> uh..
2012-09-26T17:23:28 < Mazingaro> zyp: the stack save and load finally works... I tested it with one task only :) Now I'd like to know more about PendSV, should I trigger the task switching using PendSV?
2012-09-26T17:23:30 < dongle> maps and shit uses it
2012-09-26T17:23:51 < emeb_mac> on my sammy it seems useless as compass function - always vectors off to the nearest hard iron.
2012-09-26T17:24:24 < Laurenceb> yeah theres no way around that
2012-09-26T17:25:19 < emeb_mac> and the calibration procedure is clumsy - rotate 1-2x around each axis? pleas.
2012-09-26T17:26:30 < dongle> lawl. at least they dont ask y ou to put it in the freezer and slowly thaw it out for each axis
2012-09-26T17:26:35 < dongle> to calibrate
2012-09-26T17:26:49 < Laurenceb> hehe
2012-09-26T17:29:08 < cjbaird> Charlotte has an Garmin 60CS with that calibration, which got called the 'geocaching dance'. She threatened to kick me if I recorded it. :P
2012-09-26T17:29:14 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-26T17:29:14 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-26T17:29:14 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-26T17:29:17 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-26T17:29:30 < emeb_mac> don't piss off the female.
2012-09-26T17:31:46 < emeb_mac> time for a bike ride...
2012-09-26T17:31:50 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-26T17:33:31 < cjbaird> I spent nearly $200 in new bike parts earlier this afternoon.. (tires, tubes, plus a helmet). Was considering either working on that or the scooter..
2012-09-26T17:35:46 < cjbaird> Was served by the obligatory hipster-with-tattoos-up-both-arms.. :/
2012-09-26T17:37:52 < Laurenceb> The captain had a daughter Who fell into the water; Delighted squeals Revealed that eels Had found her sexual quarter
2012-09-26T17:38:33 < dongs> is there a 24bit out equivalent of 74hc154
2012-09-26T17:38:43 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-26T17:39:04 < dongs> i guess there wouldnt be since 5 bits would be 32 out
2012-09-26T17:39:25 < dongs> 74HC138..
2012-09-26T17:39:57 < dongs> wait no
2012-09-26T18:07:55 < Oldboy> you can combine 3 decoders. 3-8 line.
2012-09-26T18:11:48 < dongs> yeah, but I just realized i dont even need 24bit decoder
2012-09-26T18:11:55 < dongs> since im doing rgb ill only need 8 rows
2012-09-26T18:12:07 < dongs> 24bit is for the columns.
2012-09-26T18:34:36 < dongs> http://i.imgur.com/sLxKX.gif
2012-09-26T18:35:15 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-26T18:35:16 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-26T18:35:16 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-26T18:35:17 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-26T18:35:17 < dongs> attn Laurenceb http://www.kickstarter.com/projects/346390315/boys-will-be-boys-0
2012-09-26T18:36:00 < Laurenceb> 4real?!
2012-09-26T18:36:17 < dongs> its odd it hasnt got 500% funding though
2012-09-26T18:36:33 < dongs> seeing how 99% of backers on kickstarters are gay
2012-09-26T18:37:21 <+izua> dongs: Feature Film about a Bisexual Down Low NCAA Basketball Player, who impregnates his secret Gay Friend, a Rare Reproducing Hermaphrodite.
2012-09-26T18:37:24 < dongs> and shit, i dont even need any decoder at all, since I can just connect 8 rows to a GPIO and drive that shit
2012-09-26T18:37:25 <+izua> dongs: wtf.
2012-09-26T18:37:40 < dongs> izua: shit, i shoulda did attn hackkitten on that one
2012-09-26T18:37:55 < dongs> infact, im doing it right now
2012-09-26T18:38:26 <+izua> hackkitten is maya?
2012-09-26T18:38:49 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-26T18:38:54 < dongs> izua: yea
2012-09-26T18:38:54 < dongs> lol
2012-09-26T18:39:04 < dongs> "maya"
2012-09-26T18:56:53 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-26T18:56:53 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-26T18:56:53 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-26T19:00:40 < emeb> How to annoy the wife: http://www.youtube.com/watch?v=2hJfBM2sRkQ&feature=youtu.be
2012-09-26T19:02:47 < karlp> wht language is that?
2012-09-26T19:02:54 < emeb> Portugese
2012-09-26T19:10:29 < karlp> Ið'm not sure that needed to be 10 minutes long
2012-09-26T19:10:46 < Laurenceb> wohoo
2012-09-26T19:10:54 < Laurenceb> chibios is full of bugs
2012-09-26T19:11:00 < emeb> shocked!
2012-09-26T19:11:05 < zyp> oh no
2012-09-26T19:11:10 < Laurenceb> stack is getting annihilated
2012-09-26T19:11:18 < Laurenceb> in halInit()
2012-09-26T19:11:24 < zyp> sounds like a good fit for your code then
2012-09-26T19:11:32 < emeb> zing!
2012-09-26T19:11:52 <+Steffanx> Isn't it just you doing something wrong Laurenceb ?
2012-09-26T19:12:10 < Laurenceb> well other people have replicated the issue
2012-09-26T19:12:14 < zyp> Steffanx, yes, but it always best to blame somebody else
2012-09-26T19:12:18 < zyp> I can't do that :(
2012-09-26T19:12:26 <+Steffanx> No collegues to blame?
2012-09-26T19:12:38 < zyp> I mean, I'm the only one working on laks :p
2012-09-26T19:13:37 < emeb> no one else to point the finger at...
2012-09-26T19:13:47 <+Steffanx> You can blame ST zyp
2012-09-26T19:13:48 < emeb> maybe the compiler?
2012-09-26T19:13:55 <+Steffanx> and the compiler
2012-09-26T19:14:02 <+Steffanx> And the designers of c++
2012-09-26T19:14:14 <+Steffanx> and your parents
2012-09-26T19:14:14 < emeb> that's it - it's all Bjarne's fault.
2012-09-26T19:15:56 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 265 seconds]
2012-09-26T19:16:14 < emeb> yay - pitch shifter works on the F405.
2012-09-26T19:16:45 < Laurenceb> anyone here ever used chibios?
2012-09-26T19:16:52 < Laurenceb> and able to help me?
2012-09-26T19:18:20 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-26T19:19:36 < emeb> googling the logs suggests that Thorn & Tectu are the main exponents of chibios here
2012-09-26T19:20:14 < Thorn> wat.
2012-09-26T19:20:29 <+Steffanx> Tectu, for sure
2012-09-26T19:20:53 <+Steffanx> Don't forget jpa- emeb
2012-09-26T19:21:04 < Laurenceb> well its failing badly
2012-09-26T19:21:05 < emeb> yep
2012-09-26T19:21:10 < Thorn> not me
2012-09-26T19:21:14 < Laurenceb> http://forum.chibios.org/phpbb/viewtopic.php?f=3&t=645&p=6505#p6505
2012-09-26T19:21:14 < emeb> sorry
2012-09-26T19:22:28 < Laurenceb> maybe i should use zyp code
2012-09-26T19:23:01 < jpa-> i'm the logarithm
2012-09-26T19:23:21 < jpa-> Laurenceb: maybe you should stop failing
2012-09-26T19:23:24 < emeb> you multiply by adding?
2012-09-26T19:23:57 < jpa-> yes, with Steffanx 
2012-09-26T19:24:13 < emeb> there's an image that will linger
2012-09-26T19:24:15 < Laurenceb> im not failing
2012-09-26T19:24:23 < Laurenceb> chibios is
2012-09-26T19:24:36 < Laurenceb> or how else would it fail in halinit
2012-09-26T19:25:28 < jpa-> maybe you compile/call/do something else wrong :)
2012-09-26T19:25:51 < jpa-> step by step bring a little piece of your code to the stm32f4 discovery example, and see when it fails
2012-09-26T19:26:39 < jpa-> i don't have F4 discovery so can't help :/
2012-09-26T19:26:48 < karlp> where's this code?
2012-09-26T19:26:59 < karlp> I have an f4 disco here and I'm waiting on some other compiles if you want.
2012-09-26T19:27:42 < Laurenceb> jpa-: ive tried that approach
2012-09-26T19:28:22 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/main.c#L134
2012-09-26T19:28:27 < Laurenceb> that line breaks it
2012-09-26T19:32:14 < jpa-> then check under what conditions it breaks
2012-09-26T19:32:23 < jpa-> if you remove everything else, does that line still break it? :P
2012-09-26T19:32:44 < Thorn> it makes halinit fail (which is called earlier)?
2012-09-26T19:32:50 < Thorn> or what
2012-09-26T19:33:29 < jpa-> stack corruptions are difficult.. try enabling all the checks (stack overrun etc.) in the chibios conf
2012-09-26T19:33:41 < Thorn> have you compiled with -O0 -g3?
2012-09-26T19:36:17 < Laurenceb> ok i may have fixed it
2012-09-26T19:36:23 < Laurenceb> 8KB process stack
2012-09-26T19:36:28 < Laurenceb> and it boots
2012-09-26T19:36:52 < karlp> how much did you ahve before?
2012-09-26T19:37:00 < Laurenceb> 1
2012-09-26T19:37:02 < Laurenceb> bbl
2012-09-26T19:37:36 < emeb> remember boys & girls - always allocate enough memory for your stack.
2012-09-26T19:37:47 < jpa-> always enable stack overflow check
2012-09-26T19:37:48 < jpa-> always.
2012-09-26T19:38:03 < emeb> compiler switch?
2012-09-26T19:38:09 < jpa-> usually rtos switch
2012-09-26T19:38:15 < dongs> heh
2012-09-26T19:38:17 < jpa-> a bit different for single-thread apps
2012-09-26T19:38:36  * emeb hasn't used an rtos yet.
2012-09-26T19:38:38 < dongs> i once called task create shit with stack_size instead of stack_size - 1
2012-09-26T19:38:54 < jpa-> for bonus points, implement a function that tells you how much of the stack is actually used
2012-09-26T19:39:25 < emeb> and don't forget to add in the stack used by that function...
2012-09-26T19:40:46 < jpa-> i have one here   https://svn.kapsi.fi/jpa/sphere2/src_common/sphere_shell.c   and it gives very useful info:   http://paste.dy.fi/iWT/plain
2012-09-26T19:40:56 < Thorn> come mcus have hardware stack overflow traps iirc
2012-09-26T19:41:00 < Thorn> *some
2012-09-26T19:41:12 < zyp> Thorn, you can do that with MPU too
2012-09-26T19:41:36 < zyp> just protect the area directly below the stack and it will be trapped when you grow into it
2012-09-26T19:41:44 < jpa-> or by putting your stack at the bottom of ram
2012-09-26T19:41:54 < Thorn> speaking of which, does anybody actually use cortex MPU?
2012-09-26T19:42:07 < jpa-> for some reason linker scripts often put it at the top
2012-09-26T19:42:26 < zyp> jpa-, because it then has the maximum area to grow in
2012-09-26T19:42:50 < Thorn> and corrupt
2012-09-26T19:42:51 < zyp> heap grow upwards and stack grow downwards and it's all fine until you run out of memory
2012-09-26T19:43:02 < jpa-> yeah, depending on how many variables & heap allocations you happen to use
2012-09-26T19:43:29 < jpa-> but yeah, it gives you maximum room and absolutely no clue when you run out of RAM
2012-09-26T19:43:56 -!- barthess [~barthess@164.177.203.166] has joined ##stm32
2012-09-26T19:44:07 < zyp> depends on how you run out of ram
2012-09-26T19:44:24 < jpa-> how so? result is always same, heap & stack overlap and corrupt eachother
2012-09-26T19:44:26 < zyp> usually sbrk checks for collision
2012-09-26T19:44:49 < jpa-> hmm, based on current stack pointer?
2012-09-26T19:44:52 < jpa-> that's not enough.
2012-09-26T19:44:52 < zyp> yes
2012-09-26T19:45:07 < zyp> well, you could set a minimum distance
2012-09-26T19:45:15 < zyp> that's up to you
2012-09-26T19:45:23 < zyp> either way the check is done before shit happens
2012-09-26T19:45:27 < jpa-> you could set the stack size in the first place
2012-09-26T19:45:37 < jpa-> and avoid all the trouble
2012-09-26T19:45:44 < zyp> how do you even know the stack size?
2012-09-26T19:45:53 < jpa-> checkstack.pl
2012-09-26T19:46:26 < jpa-> and by then checking at runtime how much is actually used (some filler value & go through the stack area in a utility func)
2012-09-26T19:46:58 < zyp> I was toying around with static analysis to find stack size last year
2012-09-26T19:47:17 < emeb> if (*ptr != 0xDEADBEEF) {...}
2012-09-26T19:47:18 < jpa-> gives a rough idea, but of course cannot handle recursion
2012-09-26T19:47:25 < dongs> http://farm8.staticflickr.com/7040/7033325281_685b74afbe_o.jpg ...
2012-09-26T19:47:43 < zyp> jpa-, biggest problem is function pointers
2012-09-26T19:47:52 < jpa-> that also
2012-09-26T19:48:09 < emeb> dongs: spinny thing?
2012-09-26T19:48:24 < dongs> guess so
2012-09-26T19:48:48 < jpa-> dongs: will that fly or just blink?
2012-09-26T19:49:05 < dongs> not mine
2012-09-26T19:49:20 < dongs> oh i found that crazyfuck who was working on gameboy cart in  2012
2012-09-26T19:49:33 < dongs> http://farm8.staticflickr.com/7101/7385161466_29e34d64cd_o.jpg
2012-09-26T19:53:35 < Mazingaro> zyp: I do not like the PendSV task switching method
2012-09-26T19:53:45 < zyp> why not?
2012-09-26T19:54:05 < Mazingaro> zyp: http://books.google.it/books?id=mb5d_xeINZEC&pg=PA129&lpg=PA129&dq=cortex-m3+pendsv&source=bl&ots=kNjIWFCOf5&sig=URJspINj-ELGg6eK5yOZrebC3Cg&hl=en&sa=X&ei=uBxjUMKfD-Hm4QTqxYDYDw&sqi=2&ved=0CEgQ6AEwBQ#v=onepage&q=cortex-m3%20pendsv&f=false
2012-09-26T19:54:22 < Mazingaro> zyp: pag 129
2012-09-26T19:55:11 < zyp> what about it?
2012-09-26T19:55:33 < Mazingaro> zyp: point 1 "Task A calls SVC for task switching..."
2012-09-26T19:55:56 < zyp> you are misunderstanding.
2012-09-26T19:56:04 < zyp> it's an example.
2012-09-26T19:56:06 < Mazingaro> zyp: isn't it cooperative multitasking?
2012-09-26T19:56:57 < Tectu> <-sBNC> [Wed September 26 2012 18:20:30]: Steffanx (~Steffanx@ip4da726ae.direct-adsl.nl) (##stm32): Tectu, for sure
2012-09-26T19:56:58 < zyp> see point 6, there is your preemptive multitasking
2012-09-26T19:57:00 < Tectu> Steffanx, ^   what was it?
2012-09-26T19:57:20 < Mazingaro> zyp: mmmh
2012-09-26T19:57:24 < Tectu> emeb, ping
2012-09-26T19:57:28 < zyp> Mazingaro, you don't need to have either or, you can have both.
2012-09-26T19:57:29 <+Steffanx> You (ab)using chibiOs Tectu
2012-09-26T19:57:40 < Tectu> Steffanx, right
2012-09-26T19:57:47 <+Steffanx> Not?
2012-09-26T19:57:50 < zyp> Mazingaro, if one thread doesn't have anything to do, there is no point in waiting until systick before switching to another thread
2012-09-26T19:57:51 < Tectu> Steffanx, emeb needs some help?
2012-09-26T19:57:54 < Tectu> Steffanx, right, as i said :D
2012-09-26T19:57:55 < Mazingaro> zyp: I did context switch in SysTick both load and save context
2012-09-26T19:58:16 < Mazingaro> zyp: so I suppose to implement task switching in PendSV handler
2012-09-26T19:59:01 < Mazingaro> zyp: so?
2012-09-26T19:59:07 < zyp> the advantage of using pendsv is that pendsv can be triggered by both systick and other stuff that needs to trigger a task switch.
2012-09-26T19:59:15 < Mazingaro> zyp: ok
2012-09-26T19:59:29 < Mazingaro> zyp: shall I load and save context in SysTick?
2012-09-26T19:59:52 < zyp> no, that's what pendsv is for.
2012-09-26T20:00:26 < Mazingaro> zyp: so PendSv -> save -> switching -> load
2012-09-26T20:00:36 < zyp> yes
2012-09-26T20:00:38 < Mazingaro> zyp: and SysTick = trigger for pendsv?
2012-09-26T20:00:44 < zyp> yes
2012-09-26T20:00:52 < Mazingaro> okù
2012-09-26T20:01:00 < Mazingaro> I'm trying it right now
2012-09-26T20:05:38 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has left ##stm32 []
2012-09-26T20:05:51 < emeb> Tectu: whut?
2012-09-26T20:10:31 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 246 seconds]
2012-09-26T20:16:37 < Tectu> <-sBNC> [Wed September 26 2012 18:19:36]: emeb (~ericb@ip72-201-79-123.ph.ph.cox.net) (##stm32): googling the logs suggests that Thorn & Tectu are the main exponents of chibios here
2012-09-26T20:16:45 < Tectu> emeb, need anything?   ^
2012-09-26T20:17:02 < emeb> Tectu: not me - Laurenceb was looking for chibios help/
2012-09-26T20:17:13 < emeb> looked like you had some background.
2012-09-26T20:18:53 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-09-26T20:19:18 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Quit: Leaving]
2012-09-26T20:19:28 < Tectu> emeb, indeed, there's  #chibios
2012-09-26T20:19:35 < Tectu> but Laurenceb already found it
2012-09-26T20:20:09 < emeb> seems he figured out his issue - stack corruption due to sizing stack wrong.
2012-09-26T20:22:11 < Tectu> aye
2012-09-26T20:23:47 <+Steffanx> heh
2012-09-26T20:25:12 < Tectu> Steffanx, you open for some doctor playing?
2012-09-26T20:25:21  * emeb runs away.
2012-09-26T20:25:30 <+Steffanx> uhm.. :S
2012-09-26T20:29:19 < Tectu> that wasn't a no :3
2012-09-26T20:29:55 <+Steffanx> Nah, it depends on who's the patient :P
2012-09-26T20:33:15 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-09-26T20:34:32 < Tectu> hmm, how would you like to have it?
2012-09-26T20:35:09 <+Steffanx> email
2012-09-26T20:35:27 <+Steffanx> it
2012-09-26T20:35:38 < Tectu> done
2012-09-26T20:37:09 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-26T20:52:34 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-227-97.clienti.tiscali.it] has quit [Ping timeout: 240 seconds]
2012-09-26T20:52:57 -!- BrainDamage [~BrainDama@62.10.4.130] has joined ##stm32
2012-09-26T21:10:25 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-26T21:10:36 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-09-26T21:10:40 < TitanMKD> hi
2012-09-26T21:11:08 < emeb> lo
2012-09-26T21:11:37 < TitanMKD> for those interested the final fixed board STM32F4 Debug v1.0 Rev1 ;)
2012-09-26T21:11:39 < TitanMKD> http://postimage.org/gallery/k7lwbq0/e5f2f14f/
2012-09-26T21:12:25 < emeb> got it working?
2012-09-26T21:15:11 < TitanMKD> yes it work fine now ;)
2012-09-26T21:15:28 < TitanMKD> I blink a led and program it with SWD
2012-09-26T21:15:32 < emeb> So it's all just software now. :P
2012-09-26T21:15:48 < zyp> nice cuts
2012-09-26T21:15:50 < TitanMKD> I still need to tests special features like USB Host ;)
2012-09-26T21:16:29 < zyp> but did you forget to hook up the vcap pins after cutting off the traces? I see no wires
2012-09-26T21:16:30 < TitanMKD> at least the basic version with only Converter 5V to 3.3V without battery work fine
2012-09-26T21:16:35 <+Steffanx> Who made the plexiglass case TitanMKD ?
2012-09-26T21:16:37 <+Steffanx> Seeed?
2012-09-26T21:16:46 <+Steffanx> or did you did it by hand?
2012-09-26T21:16:52 < TitanMKD> zyp yes it was for test i need to add some wire to connect VCAP1/2 to the 2.2uF capacitor
2012-09-26T21:17:28 < TitanMKD> Steffanx yes Seeed with my own file and using DP 60*37
2012-09-26T21:17:50 <+Steffanx> DP 60*37 ?
2012-09-26T21:19:04 < TitanMKD> yes i bought some board as they includes screw ... and i made 5 top case with laser cutting service
2012-09-26T21:19:32 < TitanMKD> look this http://dangerousprototypes.com/docs/Customizing_Sick_of_Beige_cases_for_laser_cutting
2012-09-26T21:19:41 < TitanMKD> the tutorial is very cool i used that
2012-09-26T21:20:01 < TitanMKD> for laser cutting service it is here http://www.seeedstudio.com/depot/laser-cutting-service-p-1009.html?cPath=185
2012-09-26T21:20:48 < TitanMKD> and i just bought 5 case like that http://www.seeedstudio.com/depot/sick-of-beige-basic-case-v1-dp6037-p-1223.html?cPath=178_184
2012-09-26T21:20:59 <+Steffanx> Actually my question was. What is "DP 60*37"? :)
2012-09-26T21:21:58 <+Steffanx> DangarousPrototypes [size] ?
2012-09-26T21:22:27 < TitanMKD> yes Dangerous Prototypes
2012-09-26T21:26:08 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-26T21:34:25 < Thorn> what crystal do I use with f4? are there limitations imposed by usb like in f1?
2012-09-26T21:34:32 < TitanMKD> yes
2012-09-26T21:34:42 < TitanMKD> 8MHz is the recommended one
2012-09-26T21:34:51 < TitanMKD> but 12MHz can work too
2012-09-26T21:35:06 < emeb> bootloader works with either, but it assumes 8 first.
2012-09-26T21:35:39 < TitanMKD> the bad stuff is they have not integrated USB1 in bootloader
2012-09-26T21:35:43 < TitanMKD> only USB0
2012-09-26T21:35:49 < Thorn> hm, for some reason I though discovery has 25MHz, it's actually 8
2012-09-26T21:36:05 < emeb> 25MHz on F1 IIRC
2012-09-26T21:36:13 < TitanMKD> anyway i use Discovery ref design as the board work fine ;)
2012-09-26T21:36:14 < Thorn> what is USB1? USB HS?
2012-09-26T21:36:17 < emeb> don't know why they changed.
2012-09-26T21:36:24 < TitanMKD> Thorn USB1 is LS only
2012-09-26T21:36:34 < TitanMKD> only USB0 can be HS IIRC
2012-09-26T21:36:35 < emeb> LS or FS?
2012-09-26T21:36:39 < TitanMKD> FS sorry
2012-09-26T21:37:25 < Thorn> I've never seen them called USB0/1. OTG_HS & OTG_FS are correct designations afaik
2012-09-26T21:37:37 < jpa-> indeed
2012-09-26T21:37:49 < Thorn> emeb: I used 8MHz for a f103 project with usb
2012-09-26T21:38:06 < Thorn> TitanMKD: thanks.
2012-09-26T21:38:35 < emeb> Thorn: yeah - 8MHz works, but drivers default to 25. You need to tweak to get 8 working.
2012-09-26T21:39:13 < Thorn> I may be mistaken but that's for f105/f107. SPL expects 8MHz for f103
2012-09-26T21:39:29 < TitanMKD> I speak about STM32F4 !!
2012-09-26T21:39:36 < TitanMKD> I do not know anything on other version
2012-09-26T21:39:56 < Thorn> that was for emeb
2012-09-26T21:40:03 < TitanMKD> and IIRC stm32f are not even FS
2012-09-26T21:40:06 < emeb> Thorn: I may be wrong - thought that the ref designs & ST's peripheral libs used 25.
2012-09-26T21:40:22 < emeb> (For F103/5)
2012-09-26T21:40:23 < Thorn> for f107 definitely
2012-09-26T21:40:35 < TitanMKD> emeb i have STM32F4 Discovery and i can see the Quartz is 8MHz
2012-09-26T21:40:55 < TitanMKD> even for the ST-Link MCU it is 8MHz
2012-09-26T21:41:09 < TitanMKD> and i have found very cheap 8MHz quartz at mouser
2012-09-26T21:41:28 < emeb> I did an F105 design & used 8MHz - ended up having to modify the startup code for 8 because it defaulted to 25.
2012-09-26T21:41:39 < Thorn> correct.
2012-09-26T21:41:39 < TitanMKD> and they are the recommended ones for STM32F4 (in datasheet)
2012-09-26T21:42:08 < Thorn> but not for 103. 103 wants 8MHz
2012-09-26T21:42:15 < emeb> Ah
2012-09-26T21:42:46 < Thorn> it's got a different usb core and differences in the clock tree
2012-09-26T21:43:49 < emeb> makes sense
2012-09-26T21:45:06 < zyp> 20:34:25 < Thorn> what crystal do I use with f4? are there limitations imposed by usb like in f1?
2012-09-26T21:45:10 < zyp> what limitations?
2012-09-26T21:45:44 < zyp> in F1, usbclock can be generated by either sysclock or sysclock/1.5
2012-09-26T21:45:59 < zyp> so as long as you can generate 48 or 72 MHz sysclock, it's fine
2012-09-26T21:47:13 < zyp> in F4 usbclock is generated from a seperate divisor from vcoclock, I'm using 336/7 there
2012-09-26T21:49:04 < zyp> I'm using 8MHz crystals in both cases, but other values would work nicely as long as they are divisable
2012-09-26T21:54:10 < TitanMKD> anyway on F4 for HS it requires ULPI chipset
2012-09-26T21:54:22 < TitanMKD> they should really have implemented that natively !!
2012-09-26T21:54:27 < TitanMKD> like on LPC43xx !!
2012-09-26T21:54:33 < zippe> The usual clocking story for F4 is any integer multiple of 1MHz
2012-09-26T21:54:37 < Thorn> I don't really recall exactly what limitations, but it's been descussed here about a year ago. and I think it had to do with 25MHz and connectivity line
2012-09-26T21:54:39 < TitanMKD> it is why i will switch to LPC43xx
2012-09-26T21:54:44 < zippe> Since you always prescale to 1-2MHz and then multiply up with the PLL
2012-09-26T21:54:45 < TitanMKD> and also for SGPIO ;:)
2012-09-26T21:55:02 < Thorn> *discussed
2012-09-26T21:55:26 < zyp> Thorn, isn't it ethernet that requires 25MHz to drive phy?
2012-09-26T21:55:33 < zippe> Basically, buy whatever is cheapest in the form factor that matters to you
2012-09-26T21:55:36 < Thorn> most likely
2012-09-26T21:55:55 < zippe> Yes, but you clock the PHY separately anyway
2012-09-26T21:56:45 < Thorn> no, it shouldn'thave to do with PHY, 25MHz is the standard crystal for connectivity line
2012-09-26T21:56:59 < TitanMKD> also ULPI requires an additional vreg for 1.8V !!
2012-09-26T21:57:33 < TitanMKD> it cost less to replace that by FT2232H ;)
2012-09-26T22:00:40 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: Connection reset by peer]
2012-09-26T22:01:34 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-26T22:01:39 < emeb> USB HS is one of the reasons that using the F4 to do SDR with the RTL dongle is going to be complicated. -> prog
2012-09-26T22:02:27 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-26T22:03:01 < zyp> hmm, if it weren't so late already, I'd start looking into getting HS going with my stack now
2012-09-26T22:03:11 < zyp> maybe tomorrow
2012-09-26T22:04:23 < emeb> you have an HS PHY handy?
2012-09-26T22:04:46 < zyp> yes
2012-09-26T22:04:53 < emeb> nice.
2012-09-26T22:05:04 < zyp> was included with the waveshare kit
2012-09-26T22:05:09 < zyp> one of the reasons I bought it
2012-09-26T22:05:16 < emeb> oh yeah - you mentioned that the other day.
2012-09-26T22:06:08 <+Steffanx> It's not late zyp
2012-09-26T22:06:23 < emeb> Just after noon in fact. :)
2012-09-26T22:06:25 <+Steffanx> 9PM ..
2012-09-26T22:06:59 < emeb> Steffanx gets more done after 9PM than most people do all day.
2012-09-26T22:07:19 < zyp> I'm planning to sleep tonight :p
2012-09-26T22:07:25 < emeb> overrated
2012-09-26T22:07:32 <+Steffanx> 3 hours remaining zyp
2012-09-26T22:08:03 < zyp> I've been tired all day, I'm aiming to not repeat it tomorrow
2012-09-26T22:08:22 <+Steffanx> Had a long night this night?
2012-09-26T22:08:25 < emeb> zyp: have you used SDIO on F4? I see it in laks, but wondered if it's tested.
2012-09-26T22:09:12 <+Steffanx> Register definitions only :D
2012-09-26T22:09:16 < zyp> not yet, I was planning to test it last week, so I wrote the register definitions, but then I ended up playing with usb audio instead
2012-09-26T22:09:57 < emeb> you have an SD socket as part of the waveshare kit?
2012-09-26T22:10:01 < zyp> yes
2012-09-26T22:10:33 <+Steffanx> zyp, in util we can find your logger. You have your gdb script/plugin somewhere too?
2012-09-26T22:11:28 < zyp> http://paste.jvnv.net/view/Z4QSG
2012-09-26T22:12:11 < TitanMKD> emeb it is fully supported SDIO 4bits on chibios ;)
2012-09-26T22:12:36 < emeb> TitanMKD: nice
2012-09-26T22:12:42 < TitanMKD> i plan to test it ;)
2012-09-26T22:12:45 <+Steffanx> but chibios is buggy
2012-09-26T22:12:51 < emeb> just ask Laurenceb
2012-09-26T22:12:53 < TitanMKD> of course it is not ;)
2012-09-26T22:12:57 < zyp> the waveshare kit is quite nice, I'm not sure there is any peripherals I can't test with it
2012-09-26T22:13:04 < TitanMKD> Steffanx where have you see any bug on chibios ?
2012-09-26T22:13:09 <+Steffanx> ask Laurenceb
2012-09-26T22:13:35 < TitanMKD> I used it with DMA and SPI and it really work fine
2012-09-26T22:13:41 < TitanMKD> never see any crash or problem
2012-09-26T22:13:54 < TitanMKD> i also used in parallel USART at 5MBPS with DMA
2012-09-26T22:16:23 -!- dekar [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-09-26T22:20:39 < TitanMKD> hehe just for fun
2012-09-26T22:20:48 < TitanMKD> do you know peak package temperature for STM32F4 ?
2012-09-26T22:21:00 < TitanMKD> Mouser tell me 260°C !!
2012-09-26T22:21:24 < emeb> mmm... Toasty!
2012-09-26T22:21:39 <+Steffanx> Is that was you do with your stm32's emeb ?
2012-09-26T22:21:41 <+Steffanx> -'
2012-09-26T22:21:54 < TitanMKD> but of course i never use more than 215°C especially for leaded solder
2012-09-26T22:22:00 < emeb> Steffanx: I try not to.
2012-09-26T22:23:36 < Laurenceb_> my chibios code is running now
2012-09-26T22:23:45 < Laurenceb_> for some reason it required 8kb of stack
2012-09-26T22:24:03 < TitanMKD> Laurenceb and what is crashing ?
2012-09-26T22:24:25 < Thorn> main() { int  tmp[2048]; } ?
2012-09-26T22:25:00 < TitanMKD> hehe ;)
2012-09-26T22:25:47 <+Steffanx> Who was that who commented on TitanMKD and his use of ;) ? :D
2012-09-26T22:27:01 < TitanMKD> if anyone are interested i will buy a batch of PCB for rev2 ;)
2012-09-26T22:27:15 < TitanMKD> the case cost is about 4US$ ;)
2012-09-26T22:27:48 < TitanMKD> and i will sell some PCB for something like 10US$ including shipping
2012-09-26T22:28:18 < Laurenceb_> http://static.fjcdn.com/pictures/Elephant+sexy+time.+Other+funny+stuff+by+me+http+adfoc.us+261191_45d627_3336804.jpg
2012-09-26T22:28:45 < jpa-> TitanMKD: ping me when you have usable software for it :)
2012-09-26T22:29:00 < TitanMKD> jpa- i ahve already a full TRF7970A stack ;)
2012-09-26T22:29:14 < TitanMKD> jpa- i will do a video to see the sniffing in action ;)
2012-09-26T22:29:28 < TitanMKD> and it run with chibios the latest version ;)
2012-09-26T22:29:31 < jpa-> ah, so this is about rfid sniffer not the bus pirate clone?
2012-09-26T22:29:41 < TitanMKD> bus pirate is coming
2012-09-26T22:30:07 < TitanMKD> it will be not done in 1 day and depends on feedback as primary stuff was just for me for my trf7970a board
2012-09-26T22:30:37 < jpa-> expected as much :)
2012-09-26T22:31:18 < TitanMKD> and for those which do not want Lipo charging there's option to just use a simple LDO LM1117 and to no solder other components
2012-09-26T22:31:44 < TitanMKD> and the cost is 8euros less and about 30components less to solder (total is 51 for the full Lipo+USB host version)
2012-09-26T22:32:29 < TitanMKD> also a shield could be done with ULPI ;) and USB0 HS and other cool stuff
2012-09-26T22:33:11 < TitanMKD> could even do USB HS sniffing if coupled with SDIO ;)
2012-09-26T22:33:34 < Thorn> >shield
2012-09-26T22:33:34 < Thorn> http://newspaper.li/static/93ee24b317cbe3092dbbf3ca852384d9.jpg
2012-09-26T22:34:05 < jpa-> TitanMKD: the hardware is simple and stm32f4 is a powerful platform, but it is all very much useless without software
2012-09-26T22:34:14 < jpa-> (heck, it's just another discovery board)
2012-09-26T22:34:29 < TitanMKD> jpa- of course but in fact all is here with chibios
2012-09-26T22:34:48 < TitanMKD> jpa- just for a shell and basic command for spi/i2c/usart can be done in few days
2012-09-26T22:34:57 < jpa-> chibios does not sniff usb, it doesn't even support usb HS
2012-09-26T22:35:11 < TitanMKD> to sniff usb it an other purpose ;)
2012-09-26T22:35:30 <+Steffanx> ;) ;)
2012-09-26T22:35:35 < TitanMKD> anyway it requires only lower layer to receive raw packet and save them with DMA on SDIO for example
2012-09-26T22:35:46 <+Steffanx> Any context Thorn ?
2012-09-26T22:36:06 < jpa-> TitanMKD: have you considered nuttx instead of chibios?
2012-09-26T22:36:11 < TitanMKD> as ULPI stuff is full DMA ;)
2012-09-26T22:36:22 < jpa-> it has more complete shell out-of the box (built-in filesystem commands)
2012-09-26T22:36:39 < TitanMKD> jpa- if you give me good arguments why not
2012-09-26T22:36:50 < TitanMKD> we can also mix them ;)
2012-09-26T22:37:15 < TitanMKD> using nuttx for shell and chibios for driver + multitask(depending on needs)
2012-09-26T22:38:14 < TitanMKD> anyway for BusPirate emulation it does not really requires multitask except to open multiple VCOM session and use different peripherals in parallel ;)
2012-09-26T22:38:15 < Laurenceb_> if you can make it all talk nicely
2012-09-26T22:38:54 < TitanMKD> but to be serious if i'm alone on this project i will just use chibios and some custom drivers
2012-09-26T22:39:17 < jpa-> TitanMKD: i kind of like nuttx's pseudo filesystem.. similar to linux and possible to do the same dd/cat/echo magic for random purposes  http://nuttx.sourceforge.net/NuttShell.html#cmdmount
2012-09-26T22:39:24 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-09-26T22:39:27 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-26T22:39:44 < Laurenceb_> i never quite understood it
2012-09-26T22:39:59 < TitanMKD> jpa- ha yes the shell seems very good !! with all commands like linux
2012-09-26T22:40:03 < Laurenceb_> can you mount actual storage on the filesystem?
2012-09-26T22:40:09 < Laurenceb_> or is it just for io?
2012-09-26T22:40:15 < TitanMKD> jpa- does it also support history and stuff we see on any linux box ?
2012-09-26T22:40:57 < TitanMKD> the only stuff i want to avoid is to have linux kernel behind !!
2012-09-26T22:41:11 < TitanMKD> as it will not be realtime OS after
2012-09-26T22:41:25 < TitanMKD> i do not want some context switch in 5ms
2012-09-26T22:41:58 < jpa-> TitanMKD: i haven't got my own stm32f4 board yet.. it has some readline() style interface but it doesn't seem to work properly in simulation
2012-09-26T22:42:07 < jpa-> but command history should be easy to add :P
2012-09-26T22:42:09 < TitanMKD> (when it take less than 400ns on chibios)
2012-09-26T22:42:24 < jpa-> hehe, yeah, linux is a bit different beast :)
2012-09-26T22:42:44 < jpa-> i don't know if the shell supports piping, but the OS does
2012-09-26T22:42:45 < TitanMKD> yes i just want the shell like linux but not the huge kernel and drivers ;)
2012-09-26T22:43:47 < TitanMKD> else i will port/rewrite the "shell" of bus pirate but it is very limited compared to this nuttx
2012-09-26T22:44:37 < TitanMKD> there's also busybox but a bit huge
2012-09-26T22:45:06 < TitanMKD> for me a shell shall be less than 200KB (Thumb2 mode) else it is not for embedded stuff
2012-09-26T22:46:50 < zyp> with or without commands?
2012-09-26T22:47:12 < emeb> Thorn: English motherfucker! Do you speak it?
2012-09-26T22:47:16 < TitanMKD> just minimum commands and script if possible
2012-09-26T22:47:27 < zyp> 200kB without commands is fucking huge
2012-09-26T22:47:43 < TitanMKD> yes it is why i say max 200Kb ;)
2012-09-26T22:48:25 < zyp> pfd, should be easily done in less than 20
2012-09-26T22:49:29 < TitanMKD> with python interpreter will be a must ;)
2012-09-26T22:49:33 <+Steffanx> Do you emeb ?
2012-09-26T22:51:07 < emeb> Steffanx: usually.
2012-09-26T22:51:29 <+Steffanx> k
2012-09-26T23:01:04 < TitanMKD> ha great NuttX
2012-09-26T23:01:05 < TitanMKD> On the other hand, typical, richly featured NuttX builds require more like 64K (and if all of the features are used, this can push 100K).
2012-09-26T23:01:38 < TitanMKD> it also support Realtime multitask deterministic
2012-09-26T23:31:53 -!- barthess [~barthess@164.177.203.166] has quit [Quit: Leaving.]
2012-09-26T23:35:10 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
--- Day changed Thu Sep 27 2012
2012-09-27T00:04:44 -!- tech2077_ [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-27T00:08:30 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 264 seconds]
2012-09-27T00:14:02 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-09-27T00:19:10 -!- BjoernC [~BjoernC@dyndsl-178-142-057-145.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-27T00:50:10 -!- mrnuke_ [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-27T00:58:09 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 252 seconds]
2012-09-27T01:04:31 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-09-27T01:04:51 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-27T01:05:40 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-27T01:08:18 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-27T01:08:19 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-27T01:08:19 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-27T01:09:29 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-27T01:27:14 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-27T01:32:14 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-27T01:47:30 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-27T01:48:53 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 265 seconds]
2012-09-27T01:55:09 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-09-27T01:57:42 -!- tech2077_ [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 264 seconds]
2012-09-27T02:09:31 -!- DaKu is now known as daku
2012-09-27T02:18:35 < dongs> http://www.sunstar.com.ph/cebu/local-news/2012/09/20/cybercrime-law-helps-fight-human-trafficking-243767
2012-09-27T02:19:01 < dongs> terrible news for Laurenceb
2012-09-27T02:19:36 < dongs> also: cpu autopsy
2012-09-27T02:22:21 < Laurenceb_> doesnt apply to me
2012-09-27T02:22:28 < Laurenceb_> irc is an antisocial network
2012-09-27T02:24:14 < Laurenceb_> also they'd have to kill my cpu to do an autopsy
2012-09-27T02:24:33 < Laurenceb_> and its been running for so long that seems unlikely
2012-09-27T02:38:33 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Read error: Operation timed out]
2012-09-27T02:45:54 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-27T02:54:11 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 245 seconds]
2012-09-27T03:24:25 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-09-27T03:25:48 < emeb> Laurenceb: do you nurture your uptime?
2012-09-27T03:42:12 < dongs> \\dtvcap has been up for: 1041 day(s), 19 hour(s), 3 minute(s), 50 second(s)
2012-09-27T03:42:13 < dongs> i nurture mine
2012-09-27T03:42:25 < dongs> ^ windows XP professional
2012-09-27T03:48:48 -!- BrainDamage [~BrainDama@62.10.4.130] has quit [Quit: Leaving.]
2012-09-27T03:58:46 < dongs> where is chat
2012-09-27T04:02:01 -!- WorkerBuzz [~Buzz@50.20.134.98] has joined ##stm32
2012-09-27T04:10:04 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-27T04:31:25 < dongs> fucking flyback
2012-09-27T04:31:30 < dongs> messages me on efnet
2012-09-27T04:31:37 < dongs> "you should talk to dongs on feenode sometimes"
2012-09-27T04:32:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-27T04:43:11 < dongs> http://www.seeedstudio.com/depot/60mm-square-88-led-matrix-super-bright-rgb-p-113.html?cPath=163_165 just ordered 3 of these (and now pasting link in channel so faggots cant buy them before I did)
2012-09-27T04:54:04 < dongs> im gonna use ST 24bit driver for columns and switch rows with GPIO/DMA
2012-09-27T05:01:15 < emeb_mac> aside from price, how's that different from these: https://www.sparkfun.com/products/683 ?
2012-09-27T05:03:34 < dongs> well, pinout is absolutely retarded on teh sparkfun version.
2012-09-27T05:03:43 < dongs> and you probably cant stack them next to each other for continous display.
2012-09-27T05:03:52 < dongs> cause they got those weird grooves on sides
2012-09-27T05:05:55 < emeb_mac> actually, those grooves mate like puzzle pieces.
2012-09-27T05:06:11 < emeb_mac> (I've got one of the sparkfun ones)
2012-09-27T05:06:38 < emeb_mac> but yeah - pinout is totally odd.
2012-09-27T05:07:06 < dongs> oh do they?
2012-09-27T05:07:13 < dongs> ah, right
2012-09-27T05:07:18 < dongs> they're not symmetrical
2012-09-27T05:07:26 < dongs> ok, maybe thats better then
2012-09-27T05:07:29 < dongs> but yeah fuck that pinout hard.
2012-09-27T05:07:49 < dongs> hows the brightness on it?
2012-09-27T05:08:00 < emeb_mac> Not bad.
2012-09-27T05:08:16 < dongs> what are you driving it with
2012-09-27T05:08:40 < emeb_mac> I've got the sparkfun driver board - it only gives on/off on each LED, so only 7 colors.
2012-09-27T05:09:14 < emeb_mac> want to try pulling it off and running with a custom PWM scheme though.
2012-09-27T05:10:01 < dongs> im gonna make something wiht STP24DP05 i think
2012-09-27T05:10:48 < dongs> link to sparkrip driver?
2012-09-27T05:11:04 < dongs> https://www.sparkfun.com/products/759 is it this hsit?
2012-09-27T05:11:04 < dongs> no
2012-09-27T05:11:30 < dongs> https://www.sparkfun.com/products/760 this shit?
2012-09-27T05:12:00 < emeb_mac> yep
2012-09-27T05:12:02 < dongs> sounds like it by desription
2012-09-27T05:12:15 < emeb_mac> just an AVR w/ SPI interface
2012-09-27T05:12:20 < dongs> yeah.
2012-09-27T05:12:40 < emeb_mac> Might be able to reprogram it to do PWM.
2012-09-27T05:12:54 < emeb_mac> but likely need a whole new circuit.
2012-09-27T05:18:59 -!- tech2077_ [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-27T05:23:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-27T05:27:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-27T05:45:25 < dongs> emeb_mac: also seed is common anode, spark is common cathode
2012-09-27T05:45:57 < emeb_mac> hmmm - which is easier to drive?
2012-09-27T05:46:05 < dongs> thats what im trying to figuret out right now lol
2012-09-27T05:47:49 < dongs> the drivers i got are for common anode.
2012-09-27T05:48:13 < dongs> ST driver as well.
2012-09-27T05:48:22 < dongs> i need something to switch rows on teh high side
2012-09-27T05:48:30 < dongs> my analog-fu isnt good enough to figure it out
2012-09-27T05:55:39 < dongs> dumb common sense says use some pfets but thats lame
2012-09-27T05:55:45 < dongs> maybe a 8x darlington driver?
2012-09-27T05:57:47 < emeb_mac> I was looking at this a while back...
2012-09-27T05:58:04 < emeb_mac> Someone out there has a 8x high-side driver chip.
2012-09-27T05:58:54 < dongs> no, the R/G/B columns I will drive with that ST part
2012-09-27T05:59:09 < dongs> but the 8x rows, i wanna switch with something else
2012-09-27T05:59:29 < dongs> china used some obsolete fucking part
2012-09-27T06:00:43 < dongs> lol just found the same fucking matrixes for liek 6 bucks on shitbay..
2012-09-27T06:00:48 < dongs> fuck seeed, gonna cancel
2012-09-27T06:01:02 < emeb_mac> yow!
2012-09-27T06:02:13 < emeb_mac> check these: http://www.allegromicro.com/en/Products/Motor-Driver-And-Interface-ICs/High-and-Low-Side-Drivers.aspx
2012-09-27T06:04:31 < dongs> hmmz
2012-09-27T06:04:41 < dongs> noninverting, too..
2012-09-27T06:04:42 < dongs> thats neat
2012-09-27T06:04:58 < dongs> A2982
2012-09-27T06:05:10 < dongs> a bit spendy buty
2012-09-27T06:05:12 < dongs> but
2012-09-27T06:05:17 < dongs> hmm
2012-09-27T06:06:28 < dongs> why hte fuck did they go with 20pin apckage with 2 pins NC...
2012-09-27T06:06:46 < dongs> uh..
2012-09-27T06:06:48 < emeb_mac> you want 19?
2012-09-27T06:06:50 < dongs> its turnon/off delay is horrible
2012-09-27T06:06:57 < dongs> like in the us range
2012-09-27T06:07:02 < dongs> darlington arrays are in ~100ns range
2012-09-27T06:08:07 < dongs> emeb_mac: no, 18
2012-09-27T06:08:20 < dongs> since all you really need is 8 inputs, 8 outputs, Vs and Gnd
2012-09-27T06:08:26 < emeb_mac> look at the schematic - it is a darlington. just has a pre-driver so you can interface it to lv logic easily.
2012-09-27T06:08:36 < Bird|lappy> dongs, maybe their packaging people don't provide 18-PDIP leadframes or the likes?
2012-09-27T06:08:44 < dongs> Bird|lappy: right, right.
2012-09-27T06:08:56 < dongs> (also this is not dip lol)
2012-09-27T06:09:07 < dongs> emeb_mac: well why does it take 10x longer to on/off
2012-09-27T06:10:01 < dongs> but I guess allegro dudes know better
2012-09-27T06:10:03 < dongs> since they do ledshit.
2012-09-27T06:10:32 < emeb_mac> dongs: it turns on fast - 300ns
2012-09-27T06:10:52 < emeb_mac> turns off slow because turnoff isn't active - has to self discharge.
2012-09-27T06:11:00 < dongs> right
2012-09-27T06:11:13 < Bird|lappy> yeah, those darn minority carriers :P
2012-09-27T06:25:37 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-27T06:25:43 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-27T06:27:46 < dongs> http://i.imgur.com/ug04f.jpg lol
2012-09-27T06:29:48 < emeb_mac> oooohhh - sekrit chips!
2012-09-27T06:30:20 < emeb_mac> dongs:  these ebay arrays: http://www.ebay.com/itm/RGB-Full-Color-Dot-Matrix-LED-8x8-Display-60x60mm-Arduino-/271056288881?pt=LH_DefaultDomain_0&hash=item3f1c36bc71 ?
2012-09-27T06:30:32 < dongs> yes thats the one i was looking at
2012-09-27T06:30:40 < dongs> 1/3 the price of scamstudio
2012-09-27T06:30:43 < emeb_mac> yep
2012-09-27T06:33:01 < dongs> http://www.ebay.com/itm/110886286616?
2012-09-27T06:33:02 < dongs> lawl
2012-09-27T06:34:00 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-27T06:34:34 < dongs> shipping not availale to jp though.. must be hatin' on the islands
2012-09-27T06:35:22 < emeb_mac> it's that islands thing...
2012-09-27T06:35:59 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-27T06:36:40 < emeb_mac> a lot of other ebay vendors selling arrays for similar prices.
2012-09-27T06:37:02 < emeb_mac> maybe some ship to JP
2012-09-27T06:39:01 < dongs> yeah m, ill just get that one from lee or whatever
2012-09-27T06:39:02 < dongs> the shit you linked
2012-09-27T06:40:53 < dongs> lol just bought 8 for the price of 3 on scamstudio
2012-09-27T06:43:59 < emeb_mac> heh - see the number sold just jumped up.
2012-09-27T06:44:20 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-27T06:49:21 < dongs> only one guy on google used that allegro driver
2012-09-27T06:49:27 < dongs> there must be a cheaper alternative
2012-09-27T06:53:57 < emeb_mac> would be cool if you could find one with a shift reg built in.
2012-09-27T06:54:57 < dongs> you mean?
2012-09-27T06:55:02 < dongs> oh
2012-09-27T06:55:04 < dongs> heh
2012-09-27T06:55:07 < dongs> why?
2012-09-27T06:55:19 < dongs> wouldnt it be easier to just waste 8 bits of GPIO
2012-09-27T06:55:37 < dongs> instead of shifting it or 3:8 decode
2012-09-27T06:55:55 < dongs> we're not limited by tarduino here
2012-09-27T06:56:51 < emeb_mac> w/ SR you only need 2 bits - clk & data. clock a '1' in, then shift it down the line.
2012-09-27T06:57:03 < dongs> yes but why complicate it?
2012-09-27T06:57:08 < dongs> could spend time on PWM isntead
2012-09-27T06:57:22 < emeb_mac> if you can get it in a single chip it's simpler
2012-09-27T06:57:38 < emeb_mac> the ST driver handles the brightness modulation.
2012-09-27T06:57:47 < dongs> no it doesnt
2012-09-27T06:57:57 < dongs> its on/off only
2012-09-27T06:58:01 < emeb_mac> No? thought it had current control.
2012-09-27T06:58:10 < dongs> not individual/pwm, no
2012-09-27T06:58:16 < emeb_mac> bummer
2012-09-27T06:58:29 < dongs> the st evalboard it comes with has some lib for software pwm.
2012-09-27T06:58:43 < dongs> i could also use one of ti 24ch drivers i guess...
2012-09-27T06:59:11 < dongs> then i get free pwm.
2012-09-27T06:59:26 < emeb_mac> yeah
2012-09-27T06:59:58 < emeb_mac> but 8x GPIO would work fine too
2012-09-27T07:02:34 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-27T07:05:43 < dongs> i think ill stick with ST driver and just use their software pwm lib
2012-09-27T07:06:00 < dongs> not to mention its filthy cheap
2012-09-27T07:07:21 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-27T07:11:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-27T07:14:44 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-27T07:15:42 < emeb_mac> sounds like a plan
2012-09-27T07:17:02 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Read error: Operation timed out]
2012-09-27T07:17:14 < dongs> if tarduino can do 8x8 with 32level pwm out of 3 pins and 74hc595's , im sure so can STM32
2012-09-27T07:17:32 < dongs> no it doesnt.
2012-09-27T07:17:38 < dongs> only fags use that shit.
2012-09-27T07:18:04 < dongs> shit's so oldschool its only available in DIP
2012-09-27T07:18:25 < emeb_mac> dipdipdipdip
2012-09-27T07:18:39 < dongs> flyback ordered digikey's entire stock of LPCs in DIP
2012-09-27T07:20:00 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-27T07:20:00 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-27T07:20:00 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-27T07:20:38 < dongs> obvious troll is obvious
2012-09-27T07:24:16 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-09-27T07:26:01 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-09-27T07:28:07 < dongs> http://www.digikey.com/product-detail/en/ULN2803ADWR/296-15777-1-ND/598088 this is so much cheaper though..
2012-09-27T07:28:13 < dongs> duno if its usable for source driver though
2012-09-27T07:30:33 < emeb_mac> nope - low side
2012-09-27T07:38:11 < dongs> fail.
2012-09-27T07:38:48 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-09-27T07:38:54 < dongs> http://www.instructables.com/files/deriv/FYP/GZUS/GWWZSVR4/FYPGZUSGWWZSVR4.LARGE.jpg
2012-09-27T07:38:57 < dongs> lol.
2012-09-27T07:40:52 < cjbaird> how much did that manchildtoy cost..
2012-09-27T07:41:48 < dongs> in man-hours?
2012-09-27T07:42:13 < emeb_mac> so much effort put into bending the wires.
2012-09-27T07:47:29 < dongs> haha advanced circuits shit was filed into 'junk'label in my gmail
2012-09-27T07:47:32 < dongs> but now it ended up spam too
2012-09-27T07:47:44 < dongs> ASSEMBLY FALL SPECIALS!!
2012-09-27T07:47:45 < dongs>      It's Fall...Leaves are Beginning
2012-09-27T07:47:45 < dongs>    to Drop...and so are our Prices!
2012-09-27T07:47:54 < dongs> they still want something liek $600 for 4 layer protos
2012-09-27T07:48:02 < dongs> like get the fuck out.
2012-09-27T07:49:01 < dongs> ohh new software for dicknplace
2012-09-27T07:49:10 < dongs> i wonder waht shit is fucked now
2012-09-27T07:49:17 < dongs> each new upograde is like "upgrade and hope it doesnt fuck shit up"
2012-09-27T07:55:21 < emeb_mac> digi-key search fail - looking for high-side drivers, keeps giving me low-side.
2012-09-27T07:57:11 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-09-27T07:59:11 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-27T08:10:57 < dongs> ya cuz you gota dick with it to place shit.
2012-09-27T08:13:52 < dongs> emeb_mac: free pcbs for you if you figure out a good 8channel shit for high side
2012-09-27T08:14:43 < dongs> though i suspect that allegro part will do fine
2012-09-27T08:16:59 < emeb_mac> dongs: still looking. All the ones I've found so far are more expensive than the 2982
2012-09-27T08:31:19 -!- claude is now known as Claude
2012-09-27T08:40:56 < emeb_mac> dongs: here's one a little cheaper, 18pin pkg: http://www.digikey.com/product-detail/en/MIC2981%2F82YWM/576-1159-ND/771628
2012-09-27T08:41:36 < dongs> hmm yea
2012-09-27T08:41:51 < dongs> and should have chinaclones available
2012-09-27T08:42:02 < dongs> schema looks mostly like the A part
2012-09-27T08:42:26 < dongs> turn on delay is 1us though :(
2012-09-27T08:42:42 < emeb_mac> yep - a bit slower
2012-09-27T08:42:58 < emeb_mac> how fast you want to plex it?
2012-09-27T08:43:24 < dongs> not sure actually, so i was going faster is better :)
2012-09-27T08:43:33 < dongs> but
2012-09-27T08:43:39 < dongs> mic is 5V input only?
2012-09-27T08:44:19 < dongs> Vce is what matters for input levels right
2012-09-27T08:44:26 < emeb_mac> min 2.4V for "on"
2012-09-27T08:44:36 < emeb_mac> should be OK for 3.3V MOS outputs.
2012-09-27T08:45:11 < dongs> what hte hell is the diff between 81 and 82>
2012-09-27T08:46:28 < emeb_mac> no diff I can see - they named it both for xref
2012-09-27T08:47:34 < emeb_mac> so figure turn-on & turn-off should be < 10% of total cycle.
2012-09-27T08:48:10 < emeb_mac> that means ~100us / row = 800us/ for whole array = 1200Hz refresh rate.
2012-09-27T08:48:20 < emeb_mac> That should be too fast to see.
2012-09-27T08:51:34 < dongs> mmkay
2012-09-27T08:51:46 < dongs> well, it seems 18pin would be better.. as there's higher chances of finding similar shit cheaper
2012-09-27T08:52:12 < emeb_mac> prolly
2012-09-27T08:52:54 -!- tech2077_ [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 264 seconds]
2012-09-27T08:55:31 < dongs> lets see what my chinaman says for the micrel parts cost
2012-09-27T08:57:20 < emeb_mac> checked Vce sat & current - should be fine for driving 24 LEDs @ 20mA ea
2012-09-27T08:57:42 < emeb_mac> 24 * 20 = 480 < 500mA rated current
2012-09-27T08:58:19 < dongs> right. most of these shits have like 20mA, <= 100mA peak
2012-09-27T08:59:27 < emeb_mac> might need to supply LEDs w/ 6V though -
2012-09-27T08:59:56 < emeb_mac> data sheet for array says Vf for green = 3.85V max @ 20ma.
2012-09-27T09:00:11 < jpa-> this looks like a quite neat way to handle gpio config: http://svn.code.sf.net/p/nuttx/code/trunk/nuttx/arch/arm/src/stm32/chip/stm32f40xxx_pinmap.h
2012-09-27T09:00:26 < emeb_mac> 6V - 1.9V (Vcesat) = 4.1V > 3.85V Vf
2012-09-27T09:01:55 < emeb_mac> jpa-: glad I didn't have to type that.
2012-09-27T09:02:41 < jpa-> type once, use many times
2012-09-27T09:02:56 < dongs> too bad its nuttx
2012-09-27T09:04:22 -!- tech2077_ [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-27T09:05:33 < jpa-> dongs: what's wrong with that?
2012-09-27T09:05:40 < jpa-> you like cox better than nuttx?
2012-09-27T09:08:15 < dongs> heh remap includes for f10x are nasty
2012-09-27T09:09:01 < jpa-> f10x remap is nasty :)
2012-09-27T09:20:59 < emeb_mac> here's one for the apple haters: https://www.youtube.com/watch?feature=player_embedded&v=ew6fv9UUlQ8
2012-09-27T09:25:25 < dongs> funny level: american talk show
2012-09-27T09:26:19 < emeb_mac> not as funny as a japanese gameshow.
2012-09-27T09:27:15 < dongs> thats never been funny
2012-09-27T09:27:50 < emeb_mac> there you have it.
2012-09-27T09:32:50 < dongs> is 6V led voltage a problem with allegro part as well?
2012-09-27T09:33:12 < emeb_mac> lets see...
2012-09-27T09:33:31 < emeb_mac> yep
2012-09-27T09:34:00 < emeb_mac> but that's only if you need to run them at 20ma
2012-09-27T09:34:26 < emeb_mac> they may be bright enough at 10ma
2012-09-27T09:35:28 < dongs> ill prob make the driver pcb with pinout for that chinq matrix, and make my own matrix using 5050 leds as well,  hmm those are only 20mA also. i guess maybe tis ok
2012-09-27T09:37:48 < emeb_mac> those square ones you linked the other day?
2012-09-27T09:38:23 < dongs> i have a reel i got from itswell or something (some korea place) and getting 2 more from china,, yeah, those are like standard.. 5x5mm, plcc6 case
2012-09-27T09:41:01 < dongs> 8 pfets is expensive no matter what compared to a 8ch driver heh
2012-09-27T09:42:02 < emeb_mac> yep, and a nuisance to solder.
2012-09-27T10:12:00 -!- dekar [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-09-27T10:22:41 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-27T10:25:09 -!- Claude is now known as claude
2012-09-27T10:39:01 -!- claude is now known as claude_work
2012-09-27T11:00:57 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-27T11:01:00 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Excess Flood]
2012-09-27T11:01:18 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-27T11:03:17 -!- dekar [~dekar@dslb-188-107-173-096.pools.arcor-ip.net] has joined ##stm32
2012-09-27T11:03:21 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-27T11:50:29 < dongs> huh, technically, i dont need 24bit driver, i can just use 8 of these WS2811s and do the row switcharoo,
2012-09-27T11:50:33 < dongs> then i dont need to care about PWM...
2012-09-27T11:51:09 < dongs> maybe ill ghetto a test up on protoboard and see how it works
2012-09-27T11:51:16 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-27T11:56:47 < dongs> DoNgS
2012-09-27T11:58:59 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-27T12:15:21 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-09-27T12:24:18 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-27T12:24:21 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-27T12:43:56 < dongs> ouch, mic part is expensive from china
2012-09-27T12:52:07 < dongs> need to order some reels again.
2012-09-27T12:54:03 < dongs> It has been thought about, some inputs are wrong, negative tim rather than positive for example so you lose a PWM channel, you lose the flexi port also so no external I2C, and it's just another platform to support. There is a lot of "code compatible" claims that just are not true, supporting another target is a fair amount of work.
2012-09-27T12:54:08 < dongs> wat
2012-09-27T12:54:12 < dongs> I thought F1/F3 is api compatible
2012-09-27T12:54:45 < dongs> to the point where I can just swap the chips and load F1 code and it will work wihtout FPU
2012-09-27T12:56:01 <+Steffanx> That is a quote from a random guy on the web?
2012-09-27T12:57:04 < dongs> david ankers, leader of THE openpilot project.
2012-09-27T12:57:34 < dongs> i looked and it looked same, but w hat hte fuck do i know
2012-09-27T12:57:50 <+Steffanx> It would be nice if people had real arguments/code examples when they claim things like that :D
2012-09-27T12:58:27 < dongs> i guess by negative TIM he 's referring to TIMx vs TIMx_N shit
2012-09-27T12:58:34 < dongs> but that would fuck with pin/api compatibility
2012-09-27T12:58:38 < dongs> so that sounds a bit farfetched
2012-09-27T12:58:58 < dongs> if anything, its even MORE timers even on _MD devices
2012-09-27T12:59:08 < dongs> so i can finally use some non-connected timers for like.. timing stuff.
2012-09-27T13:23:27 -!- BrainDamage [~BrainDama@62.10.4.130] has joined ##stm32
2012-09-27T13:31:34 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-27T13:31:35 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-27T13:31:35 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-27T13:40:54 -!- cjbaird [~cjb@ppp121-44-222-26.lns20.syd7.internode.on.net] has quit [Ping timeout: 264 seconds]
2012-09-27T13:42:26 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-27T13:45:36 < karlp> hmm, dongs' led array from ebay won't ship to iceland.
2012-09-27T13:45:41 < karlp> or monaco.
2012-09-27T13:45:58 < karlp> heh, or japan
2012-09-27T13:46:07 <+Steffanx> :S
2012-09-27T13:46:12 <+Steffanx> Any reason for that?
2012-09-27T13:46:18 < karlp> it even lists "africa" as a country they don't ship to.
2012-09-27T13:46:26 <+Steffanx> *good reason
2012-09-27T13:47:50 < dongs> huh?
2012-09-27T13:47:52 < dongs> the 4.99 one?
2012-09-27T13:47:56 < dongs> or 6.70sonmething one
2012-09-27T13:48:36 < dongs> http://www.ebay.com/itm/271056288881 this is the shit i bought
2012-09-27T13:48:40 < dongs> and i think emeb
2012-09-27T13:49:28 <+Steffanx> "Arduino" heh
2012-09-27T13:49:51 < dongs> :(
2012-09-27T13:50:03 < dongs> i will make the ultimate 32bit driver board of doom for them
2012-09-27T13:50:09 <+Steffanx> What you use them for? Big ass display?
2012-09-27T13:50:11 < dongs> WITHOUT FILTHY ARDUINO
2012-09-27T13:50:15 < dongs> maybe i dunno
2012-09-27T13:50:20 <+Steffanx> I knew it
2012-09-27T13:50:28 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-09-27T14:05:52 < dongs> wtf why are caps so fucking expensive
2012-09-27T14:05:56 < dongs> 20bucks for  a reel of 18PF
2012-09-27T14:05:59 < dongs> this is fucking bullshit
2012-09-27T14:07:45 < dongs> hm does it make sense to waste GPIO for 8 rows or  I should just use 3:8 decoder hmm...
2012-09-27T14:13:41 < karlp> are you planning on doing anything else?
2012-09-27T14:14:00 < dongs> no, probably not
2012-09-27T14:14:09 < karlp> if it's going to just have a couple of ways of telling it "write this shit there" or "make me lots of rainbows" then just use the gpios
2012-09-27T14:15:19 < dongs> thats what im thinking
2012-09-27T14:15:51 < dongs> the only optimization is i'd be able to do something like GPIOA = counter++; with a 3:8 otherwise i'd need to shift a bit around. probably doesnt matter
2012-09-27T14:16:55 < dongs> ah i do need some sorta buffer for 3V>5V shit cuz this chink ws2811 shit works off 5V logic
2012-09-27T14:20:19 < dongs> http://www.digikey.com/product-detail/en/SN74LVC1G125DBVR/296-11603-1-ND/385742
2012-09-27T14:20:21 < dongs> sounds liek a winner
2012-09-27T14:20:54 < dongs> agh no
2012-09-27T14:20:59 < dongs> 0.7VCC
2012-09-27T14:20:59 < dongs> wtf
2012-09-27T14:21:25 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-27T14:22:09 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-27T14:26:30 < dongs> retarded
2012-09-27T14:26:30 < dongs> 8bit buffer is cheaper than 1bit
2012-09-27T14:26:35 < dongs> all i fucking need is one bit.
2012-09-27T14:28:53 < dongs> http://www.digikey.com/product-detail/en/74HC1G125GV,125/568-7728-1-ND/2753872 ok thatll work
2012-09-27T14:31:07 < dongs> wtf Laurenceb must have been writing this datasheet
2012-09-27T14:31:13 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-27T14:31:23 < dongs> page5, VIH: Min: 4.2V, Typ: 3.2V Max: ???
2012-09-27T14:32:00 <+Steffanx> Too many spaces, so not Laurenceb's work
2012-09-27T14:32:15 < dongs> how can min be higher than typical
2012-09-27T14:33:10 < zyp> if they mean worst case and not minimal
2012-09-27T14:33:34 < zyp> for VIL, worst case is lowest, for VIH worst case is highest
2012-09-27T14:33:41 < dongs> huh ,maybe i need HCT version and not HC
2012-09-27T14:34:01 < dongs> VIH: 2.0 min, 1.6 typ
2012-09-27T14:34:23 < dongs> zyp: what does that mean in english
2012-09-27T14:34:32 < dongs> will it switch off 3V I/O?
2012-09-27T14:34:55 < zyp> it means that any value over 2.0 will be high in any case
2012-09-27T14:34:58 < zyp> or some shit like that
2012-09-27T14:35:07 < zyp> so 3V IO should be fine
2012-09-27T14:35:10 < Bird|lappy> yeah.
2012-09-27T14:35:34 < Bird|lappy> 3V logic should have no trouble driving anything with a 5V TTL-compatible input
2012-09-27T14:35:37 <+Steffanx> should .. so you're not sure :P
2012-09-27T14:35:52 < dongs> ill just get hct version
2012-09-27T14:36:26 < Bird|lappy> Steffanx, you could get into a situation where the TTL input is too much load for a weak CMOS output driver
2012-09-27T14:36:29 < Bird|lappy> but that'd be odd
2012-09-27T14:37:12 <+Steffanx> It was actually a reply to zyp, but you also said "should" :)
2012-09-27T14:37:56 < Bird|lappy> but that's really only for actual TTL, not TTL-threshold CMOS
2012-09-27T14:38:14 < Bird|lappy> the issue with TTL-threshold CMOS being driven from 3V logic is you'll get some Delta-Icc going on
2012-09-27T14:38:25 < zyp> Steffanx, I never trust weird documentation 100%
2012-09-27T14:40:53 <+Steffanx> it's not THAT weird
2012-09-27T14:47:37 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-27T14:52:04 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T14:54:26 -!- tech2077_ [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-09-27T15:02:03 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-27T15:04:31 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 246 seconds]
2012-09-27T15:05:05 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T15:05:52 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-27T15:07:37 -!- daku is now known as DaKu
2012-09-27T15:08:21 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T15:12:43 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 256 seconds]
2012-09-27T15:15:34 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-27T15:15:35 -!- BrainDamage [~BrainDama@62.10.4.130] has quit [Disconnected by services]
2012-09-27T15:18:54 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T15:26:02 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-09-27T15:29:40 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T15:33:37 < Laurenceb> chibios == epic fail
2012-09-27T15:33:43 < Laurenceb> i cant even make it boot up
2012-09-27T15:33:51 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-27T15:34:02 <+Steffanx> Laurenceb == fail
2012-09-27T15:35:45 < Laurenceb> theres got to be example st code somewhere
2012-09-27T15:35:55 < karlp> Steffanx: no, Laurencb==fail
2012-09-27T15:36:16 <+Steffanx> :P
2012-09-27T15:37:52 < karlp> or, in titanmkd speak, ;)
2012-09-27T15:38:30 < Laurenceb> wut 2 do
2012-09-27T15:38:42 < Laurenceb> this project is massively behind schedule
2012-09-27T15:38:54 < Laurenceb> should never have tried chibios :(
2012-09-27T15:39:19 < karlp> you were in love with a few weeks aog
2012-09-27T15:39:26 < Laurenceb> yeah
2012-09-27T15:39:33 < Laurenceb> but it doesnt actually work
2012-09-27T15:41:03 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds]
2012-09-27T15:41:18 < zyp> except that it works for everybody else
2012-09-27T15:41:40 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-27T15:42:04 < dongle> fuckin feenode failing
2012-09-27T15:44:31 <+Steffanx> You failed
2012-09-27T15:46:32 < dongle> nope, i dont fail
2012-09-27T15:47:19 <+Steffanx> Yeah, it's hard to admit
2012-09-27T16:13:02 < Laurenceb> im using about 6KB of ram
2012-09-27T16:13:10 < Laurenceb> yet its eating everything
2012-09-27T16:24:21 < zyp> you have a hungry memory leak
2012-09-27T16:25:05 < Laurenceb> yet it fails in halinit
2012-09-27T16:25:09 < Laurenceb> first line
2012-09-27T16:25:20 <+Steffanx> again?
2012-09-27T16:26:49 < Laurenceb> yes
2012-09-27T16:27:00 < Laurenceb> its intermittent
2012-09-27T16:27:05 < Laurenceb> nothing i can do seems to eliminate it
2012-09-27T16:27:21 < zyp> then do the impossible!
2012-09-27T16:27:25 < Thorn> >Occasionally, a runaway CAN controller unintentionally sends messages that completely tie up the bus (what is
2012-09-27T16:27:25 < Thorn> referred to in CAN jargon as a babbling idiot).
2012-09-27T16:27:44 < zyp> :)
2012-09-27T16:28:34 < Laurenceb> do the impossible?
2012-09-27T16:28:37 <+Steffanx> Thorn .. where you get all these 'random' messages from?
2012-09-27T16:28:51 < Thorn> random sources of course.
2012-09-27T16:42:28 < Thorn> I've got a long wiki page on USB, do we need that in the channel wiki?
2012-09-27T16:42:53 < zyp> does it say something interesting?
2012-09-27T16:44:36 < Thorn> enumeration process and standard requests aren't documneted yet
2012-09-27T16:44:55 < Thorn> but there's some info about the HID class
2012-09-27T16:45:00 < zyp> «page on USB» doesn't say much
2012-09-27T16:45:14 <+Steffanx> Just post it and we'll see .. Thorn ?
2012-09-27T16:45:47 < Thorn> yeah, except it needs to be translated from dokuwiki markup to mediawiki
2012-09-27T16:45:59 <+Steffanx> No automatic way to do that?
2012-09-27T16:48:04 < Thorn> found this http://pastebin.com/Vaeeek9r
2012-09-27T16:48:22 <+Steffanx> heh
2012-09-27T16:49:22 < dongs> hah my rgb matrix shit shipped
2012-09-27T16:49:56 < Laurenceb> did you get the pink pcbs yet?
2012-09-27T16:50:09 < dongs> had them a while back, waiting on stencil
2012-09-27T16:50:12 <+Steffanx> How long does it take chinese to get something to japan dongs?
2012-09-27T16:50:15 < dongs> did 2 by hand, fuck that
2012-09-27T16:50:20 < dongs> Steffanx: like a week or so
2012-09-27T16:50:21 <+Steffanx> *the chinese
2012-09-27T16:50:52 < Laurenceb> photos required
2012-09-27T16:51:27 < dongs> of?
2012-09-27T16:51:48 <+Steffanx> the pcbs ofcourse
2012-09-27T16:58:50 < dongs> oh
2012-09-27T17:00:37 < Thorn> I think I'll just use a standard IDC10 connector with the arm debug pinout for jtag/swd
2012-09-27T17:01:06 < zyp> why?
2012-09-27T17:01:46 < Thorn> 1.27mm ones are expensive and hard to find
2012-09-27T17:01:53 < Thorn> at least good shrouded ones
2012-09-27T17:02:15 < zyp> except that they are $1 and sold everywhere
2012-09-27T17:02:34 < Thorn> my Black Magic Clone should have IDC10 as one of the choices
2012-09-27T17:02:48 <+Steffanx> You cloned it?!
2012-09-27T17:03:15 < Thorn> will get to it after I finish this board I'm working on
2012-09-27T17:03:36 < Thorn> there's a holiday in China until the end of next week anyway
2012-09-27T17:04:09 < Thorn> I'll probably combine both in the same order
2012-09-27T17:04:18 < Thorn> (PCB from seeed I mean)
2012-09-27T17:08:01 < dongs> haha
2012-09-27T17:08:12 < dongs> ebay guy who shipped the rgb shits sent me a tracking + email
2012-09-27T17:08:18 < dongs> 'for any information + source etc, go here
2012-09-27T17:08:25 < dongs> http://www.bestelecs.com/wiki/index.php?title=Support_materials#STM32_Board
2012-09-27T17:09:05 < zyp> ha
2012-09-27T17:09:17 <+Steffanx> ho
2012-09-27T17:09:48 < Laurenceb> hmm
2012-09-27T17:09:58 < Laurenceb> ive got chibios booting about 50% of the time
2012-09-27T17:10:22 <+Steffanx> heh
2012-09-27T17:10:33 < Laurenceb> if i increase the working area for one of the threads by 256 byte the probability decreases by 20%
2012-09-27T17:10:35 < Laurenceb> wtf
2012-09-27T17:10:42 -!- cjbaird_ [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-27T17:10:48 < dongs> how do i stacked
2012-09-27T17:10:50 < Laurenceb> - probability of it working
2012-09-27T17:10:54 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Ping timeout: 240 seconds]
2012-09-27T17:11:12 < Laurenceb> yet it failing in halinit
2012-09-27T17:11:20 < Laurenceb> so crazy
2012-09-27T17:11:42 < Laurenceb> halinit just chews up all the memory
2012-09-27T17:11:53 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-27T17:14:30 < Thorn> wth is wrong with 0805 LEDs, nobody has them
2012-09-27T17:15:01 <+Steffanx> I do
2012-09-27T17:15:52 < Laurenceb> hmm guess ill have to live with fuxored device
2012-09-27T17:15:57 < Laurenceb> its only for testing
2012-09-27T17:16:07 < Laurenceb> at least everything runs nicely once it boots up
2012-09-27T17:17:34 < Laurenceb> shouldnt halinit be deterministic?
2012-09-27T17:17:53 < jpa-> Thorn: they are everywhere :)
2012-09-27T17:18:26 < zyp> Laurenceb, are you sure it crashes in halinit? :p
2012-09-27T17:18:34 < Laurenceb> 100%
2012-09-27T17:18:40 < jpa-> Laurenceb: are you booting by reset or is this your own bootloader?
2012-09-27T17:18:40 < Laurenceb> ive stepped through
2012-09-27T17:18:47 < Laurenceb> reset
2012-09-27T17:19:09 < Thorn> btw, is there any merit in adding a halt input to a debugger? so you can stop the cpu when e.g. a logic analyzer triggers
2012-09-27T17:19:10 < zyp> Laurenceb, then you would know where in halinit it crashes, wouldn't you?
2012-09-27T17:19:16 < Laurenceb> yes
2012-09-27T17:19:21 < jpa-> quite strange to crash there.. usb writes crash for me non-deterministically sometimes if the device is connected while it is booting
2012-09-27T17:19:56 < Laurenceb> http://forum.chibios.org/phpbb/viewtopic.php?f=3&t=645
2012-09-27T17:20:08 < zyp> jpa-, that's because usb delays are non-deterministic
2012-09-27T17:20:09 < Laurenceb> ooh
2012-09-27T17:20:16 < Laurenceb> jpa-: interesting
2012-09-27T17:20:53 < Laurenceb> aha got it
2012-09-27T17:20:59 < Laurenceb> yeah its the usb
2012-09-27T17:21:25 < Laurenceb> wtf
2012-09-27T17:21:36 < Laurenceb> so basically you cant make a chibios usb device
2012-09-27T17:22:10 < zyp> heh
2012-09-27T17:22:13 < zyp> «halInit() is called before the kernel is initialized»
2012-09-27T17:22:18 < zyp> and it does enable usb?
2012-09-27T17:22:31 < Laurenceb> yes
2012-09-27T17:22:37 < zyp> wtf
2012-09-27T17:22:37 < Laurenceb> no wait
2012-09-27T17:22:56 < Laurenceb> ah yes it seems so
2012-09-27T17:23:01 < Laurenceb> SDU is started later
2012-09-27T17:23:13 < Laurenceb> but the usb itself is initialised there it seems
2012-09-27T17:23:16 < jpa-> Laurenceb: sure you can.. you just need to fix the chibios usb driver :)
2012-09-27T17:23:24 < Laurenceb> this is working 100% now
2012-09-27T17:23:26 < jpa-> the STM32F4 usb driver is just beta anyway
2012-09-27T17:23:38 < Laurenceb> im applying power before plugging in
2012-09-27T17:24:25 < Laurenceb> and its connecting and running nicely
2012-09-27T17:25:43 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has joined ##stm32
2012-09-27T17:25:45 < zyp> but if you are resetting in debugger, it fails?
2012-09-27T17:25:58 < dongs> he uses fucking gdb
2012-09-27T17:26:09 < Laurenceb> theres no reset line connected
2012-09-27T17:26:10 < dongs> how can anything ever get done wiht that shit
2012-09-27T17:26:20 <+Steffanx> blablablablabla dong
2012-09-27T17:26:21 <+Steffanx> s
2012-09-27T17:26:27 < zyp> dongs, nothing wrong with gdb, it even supports python plugins!
2012-09-27T17:26:28 < zyp> :D
2012-09-27T17:27:32 <+Steffanx> I wonder how many food dongs need to get satisfied
2012-09-27T17:27:39 <+Steffanx> *much
2012-09-27T17:27:45 < Abhishek_> Hello All
2012-09-27T17:27:55 <+Steffanx> Lo
2012-09-27T17:28:00 < Abhishek_> New in this community
2012-09-27T17:28:13 < dongs> watc out
2012-09-27T17:28:24 < dongs> Laurenceb will turn your code into no-spaces
2012-09-27T17:28:41 < Abhishek_> I'm Using STM32F4 discovery
2012-09-27T17:28:51 < Abhishek_> and a contributor to ChibiOS/GFX
2012-09-27T17:29:48 <+Steffanx> Just like Tectu ?
2012-09-27T17:29:59 < dongs> isnt tectu just a troll
2012-09-27T17:30:04 < dongs> and Laurenceb is the only user of chibios
2012-09-27T17:30:33 < jpa-> Abhishek_: hello :)
2012-09-27T17:30:35 <+Steffanx> You are 'just a troll'.. Tectu is a troll and chibios user
2012-09-27T17:30:55 < Abhishek_> I ain't
2012-09-27T17:31:14 <+Steffanx> no, Mr. D is :)
2012-09-27T17:31:20 < Laurenceb>  Mata inte trollen
2012-09-27T17:32:03 <+Steffanx> AGreed Laurenceb
2012-09-27T17:32:36 < Abhishek_> Swedish?
2012-09-27T17:33:53 < dongs> fuck, i need to submit this order.. i guess tomorrow... amybe ill remember something else I need to get
2012-09-27T17:35:40 < Abhishek_> even I'm using chibios
2012-09-27T17:36:02 < Thorn> kawaii desu
2012-09-27T17:38:57 < dongs> lol'd
2012-09-27T17:39:10 < dongs> Thorn: will you call your next stm32 project pinku bento box
2012-09-27T17:39:33 < Thorn> I only use green solder mask.
2012-09-27T17:39:44 < dongs> terrible design choice
2012-09-27T17:41:40 < zyp> Laurenceb, your shit, will it run on F4discovery?
2012-09-27T17:41:50 < Laurenceb> yes
2012-09-27T17:41:54 < zyp> ok, going to test
2012-09-27T17:42:04 < Laurenceb> ill play about with  usbDisconnectBus(serusbcfg.usbp);
2012-09-27T17:42:35 < Laurenceb> going to restaurant for free food, bll
2012-09-27T17:42:48 < dongle> <Laurenceb> here's my shitty code guise, fix this for me im gonna go eat
2012-09-27T17:43:22 < zyp> just going to see what he is doing wrong, so I can avoid doing the same mistake later :p
2012-09-27T17:43:41 < jpa-> :D
2012-09-27T17:44:02 < zyp> learning from other peoples failures is so much better than having to learn from my own :p
2012-09-27T17:45:15 < zyp> well fuck
2012-09-27T17:45:30 < zyp> I have no ../ChibiOS
2012-09-27T17:46:26 < Abhishek_> peace
2012-09-27T17:46:32 < zyp> bullshit repo that doesn't contain neither the chibios files or any reference to what is needed
2012-09-27T17:46:40 < dongle> lawl
2012-09-27T17:47:07 < Abhishek_> which repo?
2012-09-27T17:47:09 < karlp> where is this repo?
2012-09-27T17:47:15 < karlp> I asked him yesterday but didn't get an answer
2012-09-27T17:47:16 < zyp> Laurenceb's shit
2012-09-27T17:47:20 < zyp> https://github.com/Laurenceb/Chibi-Spectro
2012-09-27T17:47:27 < karlp> there's an f4 disco here on my desk too, and I'm waiting on compilers again
2012-09-27T17:48:23 < Thorn> Chibi-Supekuturo (not sure if Japanese has the sound 'p')
2012-09-27T17:48:42 < Abhishek_> get it here: https://github.com/mabl/chibios
2012-09-27T17:49:00 < zyp> Abhishek_, but which revision?
2012-09-27T17:49:07 < Abhishek_> i compiled laurence's code
2012-09-27T17:49:34 -!- cjbaird_ is now known as cjbaird
2012-09-27T17:49:45 < zyp> stuff like that should be done with submodules
2012-09-27T17:49:47 < Abhishek_> the current one
2012-09-27T17:50:11 < Abhishek_> however you can also get it from sf.net/projects/chibios
2012-09-27T17:51:03 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-09-27T17:51:14 < zyp> Abhishek_, I've used chibios before, so I'm quite aware of that, it's just dumb to use external repos without any reference to what revision I'm supposed to use
2012-09-27T17:51:31 < karlp> 46meg?! what in the hell is in this repo?
2012-09-27T17:52:11 < karlp> oh, 93meg of list files in a build directory
2012-09-27T17:52:13 < karlp> thanks Laurenceb
2012-09-27T17:52:17 < zyp> :)
2012-09-27T17:52:34 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has joined ##stm32
2012-09-27T17:52:34 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has quit [Changing host]
2012-09-27T17:52:34 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-27T17:52:43 -!- claude_work is now known as claude
2012-09-27T17:53:12 < jpa-> karlp: :D
2012-09-27T17:53:25 < zyp> hmm, wtf
2012-09-27T17:53:26 < zyp> (gdb) restart
2012-09-27T17:53:26 < zyp> Program received signal SIGSEGV, Segmentation fault.
2012-09-27T17:53:26 < zyp> 0x55555554 in ?? ()
2012-09-27T17:53:54 < karlp> oh even better, not just once, but he repetedly committed the new .lst files.
2012-09-27T17:56:22 < zyp> something is apparently very broken here, I'm not getting to main :p
2012-09-27T17:56:40 < jpa-> 0x55555555   is the fill-value chibios uses for thread stacks
2012-09-27T17:57:15 < zyp> what's the name of chibios' entry function?
2012-09-27T17:57:24 < Abhishek_> chSysInit
2012-09-27T17:57:35 < Abhishek_> crt entry function is resethandler
2012-09-27T17:57:38 < Abhishek_> this calls
2012-09-27T17:57:41 < jpa-> zyp: ResetHandler
2012-09-27T17:57:42 < Abhishek_> early_init
2012-09-27T17:57:46 < Abhishek_> late_init
2012-09-27T17:58:01 < Abhishek_> and then after filling memory
2012-09-27T17:58:04 < Abhishek_> calls main
2012-09-27T17:58:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-27T17:58:23 < jpa-> os/ports/GCC/ARMCMx/crt0.c
2012-09-27T17:58:24 < Abhishek_> in main chSysInit starts up the OS
2012-09-27T17:58:29 < jpa-> perhaps his linker script is crap
2012-09-27T17:58:51 < Abhishek_> he modified the __process_stack_size__ = 0x8000
2012-09-27T17:58:55 < zyp> going to singlestep this and see
2012-09-27T17:59:05 < Abhishek_> All the Best
2012-09-27T17:59:07 < Abhishek_> ;)
2012-09-27T17:59:24 < Abhishek_> i did that yesterday
2012-09-27T17:59:30 < Abhishek_> thrice
2012-09-27T17:59:55 < jpa-> he modified but didn't put the modified ch.ld on his github?
2012-09-27T18:00:01 < Abhishek_> where does it trap?
2012-09-27T18:00:05 < jpa-> or i can not see that
2012-09-27T18:00:11 < Abhishek_> he's put it on the forum
2012-09-27T18:00:26 < jpa-> "great"
2012-09-27T18:00:38 < jpa-> "let's just make everyone collect my code all over the intertubes"
2012-09-27T18:00:45 < Abhishek_> look: http://forum.chibios.org/phpbb/viewtopic.php?f=3&t=645
2012-09-27T18:01:07 < Abhishek_> btw jpa- what does that mean?
2012-09-27T18:01:29 < jpa-> what?
2012-09-27T18:01:47 < Abhishek_> your last intertubes one?
2012-09-27T18:02:15 < jpa-> it's a quote of what Laurenceb was thinking
2012-09-27T18:02:21 < jpa-> i can read his mind
2012-09-27T18:02:45 < Abhishek_> hmm'
2012-09-27T18:02:49 < Abhishek_> mmmm
2012-09-27T18:04:20 < Laurenceb> back
2012-09-27T18:04:56 < Laurenceb> zyp: i use __main_stack_size__     = 0x0F00;
2012-09-27T18:04:56 < Laurenceb> __process_stack_size__  = 0x8000;
2012-09-27T18:05:03 < Laurenceb> at top of the linker script
2012-09-27T18:13:46 < Laurenceb> "<kokey> at least since the schools and universities opened many online forums had a significant drop in trolling"
2012-09-27T18:13:49 < Laurenceb> ^lol
2012-09-27T18:14:36 < karlp> Laurenceb: do this: git filter-branch --tree-filter 'rm -rf build .dep' HEAD
2012-09-27T18:14:40 < karlp> then git push -f
2012-09-27T18:14:50 < Laurenceb> yeah i know
2012-09-27T18:14:59 < Laurenceb> sudo rm -rf
2012-09-27T18:15:05 < zyp> Laurenceb, your stack is really messed up
2012-09-27T18:15:06 < Laurenceb> also fixes it
2012-09-27T18:15:11 < zyp> not sure why, but it is
2012-09-27T18:15:12 < Laurenceb> you dont say :P
2012-09-27T18:15:13 < karlp> no, sudo rm -rf doens't fix it.
2012-09-27T18:15:14 < Laurenceb> yeah
2012-09-27T18:15:19 < karlp> you need to kill it with fire in the git history
2012-09-27T18:15:28 < Laurenceb> im trolling
2012-09-27T18:15:36 < karlp> otherwise cloning your repo downloads all of it just in case I want to go back ever look at your .lst files.
2012-09-27T18:15:45 < jpa-> you're not in a position to troll
2012-09-27T18:15:45 < Laurenceb> heh
2012-09-27T18:15:58 < karlp> also, git submodule add blah.path.to.chibios might help too.
2012-09-27T18:16:02 < Laurenceb> yeah im not on holiday
2012-09-27T18:16:05 < Laurenceb> trolliday
2012-09-27T18:18:06 < Laurenceb> karlp: Cannot rewrite branch(es) with a dirty working directory.
2012-09-27T18:18:09 < karlp> heh, it won't let me submit a pull request with it all fixed for you.
2012-09-27T18:18:28 < karlp> git stash save; fix shit like I said; git stash pop
2012-09-27T18:18:59 < Abhishek_> what was the problem btw?
2012-09-27T18:19:57 < Laurenceb> dont know yet
2012-09-27T18:20:04 < Laurenceb> its not as simple as just the usb
2012-09-27T18:20:11 < Laurenceb> usb seems to be a symptom
2012-09-27T18:21:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-27T18:21:43 -!- dekar [~dekar@dslb-188-107-173-096.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-09-27T18:22:03 < Laurenceb> git updated
2012-09-27T18:24:10 < karlp> much nicer!
2012-09-27T18:24:14 < karlp> see how much faster it is?
2012-09-27T18:24:40 < karlp> now do this: https://github.com/karlp/Chibi-Spectro/commit/9bf199a933fff670583ee9b389a0296defc2fc2b
2012-09-27T18:25:54 < Abhishek_> @karlp, what did you fix?
2012-09-27T18:25:56 < Laurenceb> ok
2012-09-27T18:26:02 < Laurenceb> can i merge that?
2012-09-27T18:26:08 < Laurenceb>  <- git n00b
2012-09-27T18:26:44 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-27T18:28:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-27T18:29:11 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-09-27T18:32:22 < karlp> no, they're different lines.
2012-09-27T18:32:24 < karlp> just do this...
2012-09-27T18:32:31 < karlp> git submodule add
2012-09-27T18:32:44 < karlp> git submodule add git://github.com/mabl/ChibiOS.git
2012-09-27T18:32:54 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-27T18:33:40 < karlp> and then fix your makefile.
2012-09-27T18:34:52 < Laurenceb> ok
2012-09-27T18:35:55 < karlp> as for your actual chibi problems, no clue :)
2012-09-27T18:36:49 < Laurenceb> ok thanks for the help
2012-09-27T18:36:53 < Laurenceb> its a bit cleaner now
2012-09-27T18:37:21 < Laurenceb> yeah it seems maybe usb is a symptom not the cause
2012-09-27T18:37:47 < Abhishek_> the demo runs for an hour  now
2012-09-27T18:37:52 < Abhishek_> my usb-cdc demo
2012-09-27T18:37:57 < Abhishek_> on disco
2012-09-27T18:38:41 < jpa-> Abhishek_: have you tried pulling out the usb cable and putting it back in?
2012-09-27T18:38:48 < Abhishek_> yes
2012-09-27T18:38:52 < Abhishek_> it enumerates
2012-09-27T18:38:56 < jpa-> also try to wait 10 minutes in between while it is writing data to the port
2012-09-27T18:38:57 < Laurenceb> Abhishek_: thats not where it breaks
2012-09-27T18:38:57 -!- barthess [~barthess@178.154.56.70] has joined ##stm32
2012-09-27T18:39:07 < Laurenceb> it breaks when you connect the cable
2012-09-27T18:39:12 < Laurenceb> works reliably after that
2012-09-27T18:39:32 < Abhishek_> is there a short-circuit due to your pin configs?
2012-09-27T18:39:39 < zyp> Laurenceb, I don't have any cable connected here, and it's still broken
2012-09-27T18:39:44 < Abhishek_> check for pin conflicts
2012-09-27T18:39:54 < Laurenceb> zyp: oh
2012-09-27T18:40:00 < Laurenceb> i havent actually checke dthat
2012-09-27T18:40:47 < Abhishek_> i'm drawing using the touchscreen on the lcd
2012-09-27T18:41:02 < Laurenceb> zyp: oh yes i see
2012-09-27T18:41:17 < Abhishek_> and dumping the co-ordinates onto the terminal
2012-09-27T18:41:18 < Laurenceb> breaks for me too
2012-09-27T18:41:43 < Abhishek_> if i remove the cable the thread gets stuck
2012-09-27T18:41:43 < Laurenceb> seems to be the same if i swap to the chibios board file
2012-09-27T18:42:01 < Abhishek_> i.e. i am unable to draw points
2012-09-27T18:42:09 < Abhishek_> other threads seem to work
2012-09-27T18:48:38 < Abhishek_> I think the USB-CDC methods are blobking ones
2012-09-27T18:49:24 < Abhishek_> this means they don't return until and unless data is churned out of the usb IP core
2012-09-27T18:49:35 < jpa-> all write methods are blocking by default
2012-09-27T18:49:43 < jpa-> but they don't unhang even if you reconnect USB
2012-09-27T18:49:47 < Abhishek_> yeah
2012-09-27T18:49:56 < Abhishek_> When I reconnect USB, it works
2012-09-27T18:50:03 < jpa-> ah, for you it works
2012-09-27T18:50:05 < jpa-> for me it does not
2012-09-27T18:50:10 < jpa-> it stays stuck for ever
2012-09-27T18:50:20 < Abhishek_> it worked got stuck
2012-09-27T18:51:13 < Abhishek_> after a few writes
2012-09-27T18:51:22 < Abhishek_> then when  fired up a terminal
2012-09-27T18:51:26 < Abhishek_> on the PC
2012-09-27T18:51:31 < Abhishek_> it starts working
2012-09-27T18:52:00 < Abhishek_> perpetually
2012-09-27T18:52:10 < Abhishek_> I guess this is what is going on:
2012-09-27T18:52:30 < Abhishek_> when USB is unconnected, the data is pushed to the FIFO
2012-09-27T18:52:39 < Abhishek_> i.e. the methods return when called
2012-09-27T18:52:43 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-09-27T18:52:57 < Abhishek_> once the buffer becomes full, it blocks
2012-09-27T18:53:19 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-27T18:53:21 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-27T18:53:27 < Abhishek_> when I reconnect USB, it works as the data is now pushed out of the FIFO
2012-09-27T18:53:32 < Abhishek_> onto the PC
2012-09-27T18:54:42 < Tectu> Steffanx, me not troll
2012-09-27T18:54:45 < Tectu> me swiss guy
2012-09-27T18:54:51 < zyp> http://paste.jvnv.net/view/S2I3y <- wtf is going on here?
2012-09-27T18:55:36 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-27T18:56:02 < zyp> nvicEnableVector is called with lr=0xffffffff, so it returns there when it's done
2012-09-27T18:56:12 < zyp> i.e. crashes
2012-09-27T18:56:54 < Abhishek_> u debugging laurence's code?
2012-09-27T18:57:03 < zyp> yes
2012-09-27T18:57:54 < Abhishek_> so stack overflows
2012-09-27T18:58:09 < zyp> ah, yeah
2012-09-27T18:58:33 < zyp> it starts out with some silly low amount
2012-09-27T18:58:50 < Abhishek_> do you able to see the stack growth
2012-09-27T18:59:15 < Laurenceb> i tried __main_stack_size__     = 0x0F00;
2012-09-27T18:59:16 < Laurenceb> __process_stack_size__  = 0x8000;
2012-09-27T18:59:21 < Laurenceb> but it doesnt fix it
2012-09-27T18:59:38 < zyp> where?
2012-09-27T19:00:16 < Abhishek_> in the ld script
2012-09-27T19:00:19 < Abhishek_> check in
2012-09-27T19:00:24 < zyp> found it
2012-09-27T19:00:29 < Abhishek_> ok
2012-09-27T19:00:39 < zyp> ah, right, it starts out with 0x400 for each
2012-09-27T19:01:48 < zyp> right, now it looks sane
2012-09-27T19:01:49 < Laurenceb> yeah
2012-09-27T19:03:16 < zyp> doesn't enumerate though
2012-09-27T19:04:36 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-27T19:04:48 < Laurenceb> aha
2012-09-27T19:04:54 < Laurenceb> i added   SYSCFG->PMC = 0; at top of main
2012-09-27T19:05:04 < Laurenceb> seems to work reliably...
2012-09-27T19:05:09 < Laurenceb> 20 attaches ok
2012-09-27T19:07:41 < zyp> doesn't do shit here
2012-09-27T19:07:50 < Abhishek_> wait
2012-09-27T19:07:57 < Abhishek_> RM0090 stm32f4
2012-09-27T19:07:59 < zyp> and what has that to do with anything? it's only some ethernet flag
2012-09-27T19:08:05 < Abhishek_> lists it as reserved bit!
2012-09-27T19:08:30 < Abhishek_> they are zero anyway
2012-09-27T19:08:56 < Laurenceb> hmm
2012-09-27T19:09:11 < Laurenceb> that was someones suggestion from #chibios
2012-09-27T19:09:35 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-09-27T19:09:39 -!- mode/##stm32 [+v dekar] by ChanServ
2012-09-27T19:09:40 < Abhishek_> he's using a STM32L remember
2012-09-27T19:09:52 < Abhishek_> this a STM32F4
2012-09-27T19:10:09 < Abhishek_> they have different register organisation
2012-09-27T19:10:19 < zyp> and a very different usb peripheral :p
2012-09-27T19:10:39 < Abhishek_> yep
2012-09-27T19:10:42 < zyp> anyway, shit works but doesn't enumerate
2012-09-27T19:10:58 < Abhishek_> I faced the same yesterday
2012-09-27T19:11:03 < Abhishek_> using Win7
2012-09-27T19:11:06 < zyp> I increased main stack, that seemed to be the problem
2012-09-27T19:11:18 < Abhishek_> how much?
2012-09-27T19:11:23 < Abhishek_> maybe 0xFFFF ?
2012-09-27T19:11:27 < zyp> 0x2000
2012-09-27T19:11:48 < Abhishek_> 8K ?
2012-09-27T19:11:53 < zyp> yep
2012-09-27T19:11:55 < Abhishek_> yes
2012-09-27T19:12:18 < Abhishek_> if there is no wait and any infinite loop is there
2012-09-27T19:12:31 < Abhishek_> USB will NOT enumerate
2012-09-27T19:12:40 < Abhishek_> mind this
2012-09-27T19:12:58 < zyp> but I'm tired of this, I'm going back to my own code that works :p
2012-09-27T19:13:55 < Laurenceb> its f4 not l1
2012-09-27T19:13:56 < Abhishek_> i'm sitting on it
2012-09-27T19:14:04 < Abhishek_> yes i know
2012-09-27T19:14:07 < Abhishek_> ;)
2012-09-27T19:19:43 < Laurenceb> you have to clk peripherals before you can access the registers right?
2012-09-27T19:19:57 < zippe> Laurenceb: yes
2012-09-27T19:20:17 < zippe> zyp: -finstrument_functions
2012-09-27T19:20:28 < zippe> zyp: also -ffixed_r10
2012-09-27T19:20:55 < zippe> zyp: now you can write a handler for the function instrumenter that compares SP against R10 and panics on overflow
2012-09-27T19:21:17 < zippe> The instrumenter is called after the prolog does the stack reservation for the function
2012-09-27T19:21:34 < zyp> interesting
2012-09-27T19:22:00 < zippe> https://github.com/PX4/Firmware/blob/master/nuttx/arch/arm/src/armv7-m/up_stackcheck.c
2012-09-27T19:23:55 < Laurenceb> do i need to set RCC_APB1ENR_TIM5EN?
2012-09-27T19:23:59 < Laurenceb> zippe: nice
2012-09-27T19:24:00 < zippe> It's not lovely from a performance perspective, but it works
2012-09-27T19:25:29 < zippe> Laurenceb: only if you want to use TIM5
2012-09-27T19:25:42 < Laurenceb> yeah
2012-09-27T19:25:46 < Laurenceb> obviously :P
2012-09-27T19:32:18 < zyp> zippe, are you only enabling that code when you need to debug stack overflow problems?
2012-09-27T19:32:56 < zippe> zyp: we actually run with it all the time
2012-09-27T19:33:06 < zippe> If you don't tell your users, nobody notices.
2012-09-27T19:33:15 < zippe> Single-stepping through it is vaguely annoying
2012-09-27T19:33:57 < zippe> But the total overhead is ~20 cycles per function and I no_instrument_function a few of the super-critical hotpaths
2012-09-27T19:34:08 < zippe> The rest of the time it's just invisible.
2012-09-27T19:34:31 < zyp> what about the unavailability of r10?
2012-09-27T19:34:42 < zippe> What about it?
2012-09-27T19:34:52 < zippe> You have lots of GPRs
2012-09-27T19:35:03 < zippe> What's one less between friends?
2012-09-27T19:35:28 < zyp> so it has no noticeable impact?
2012-09-27T19:35:43 < zippe> Not at the macro level that I've been able to measure on our code
2012-09-27T19:35:54 < zippe> I'm sure you could construct any number of strawmen where the register pressure is a problem
2012-09-27T19:36:52 < zippe> But in general use, gcc rarely has more than half a dozen really hot registers
2012-09-27T19:36:54 < zyp> nah, I'm not well versed in code known to have high register pressure :p
2012-09-27T19:37:32 < zippe> Well, there you go
2012-09-27T19:37:59 < zippe> Anyway, it's worth at least what you paid for it, and if it helps you find a stack overflow or two, you'll be ahead 8)
2012-09-27T19:38:09 < zyp> I'm assuming that inlined functions doesn't get instrumented?
2012-09-27T19:39:03 < karlp> seeing as inline is only a hint, not a rule, I'd guess you'd want to be explicit about it...
2012-09-27T19:39:25 < zyp> I mean functions that end up being inlined
2012-09-27T19:39:40 < karlp> well, it's not a function then is it :)
2012-09-27T19:40:00 < jpa-> chibios has it's own check that is done on contex switches, but it doesn't help during bootup
2012-09-27T19:41:56 < zyp> but it can't trap overflows as they happen, only after the fact
2012-09-27T19:43:09 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-27T19:43:09 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-27T19:43:09 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-27T19:43:16 < zyp> I've been considering using the MPU to protect the area below the active stack
2012-09-27T19:44:31 < zyp> but that either requires that you waste some memory for an explicit guard band, or know that the active thread doesn't need to access that memory in any sane way
2012-09-27T19:46:23 < jpa-> you could set write watchpoints using the debug unit, if you don't need them for other purpose :)
2012-09-27T19:47:13 < zyp> that would not be very different from the MPU approach
2012-09-27T19:47:24 < zyp> in either case you would trap writes to that area
2012-09-27T19:47:24 < jpa-> except that you would only lose 1 word
2012-09-27T19:47:36 < jpa-> not a whole block of memory
2012-09-27T19:47:43 < Laurenceb> wohoo timers working
2012-09-27T19:47:54 < Laurenceb> but the gating doesnt quite seem right
2012-09-27T19:48:04 < zyp> jpa-, what's the granularity of the MPU then?
2012-09-27T19:48:05 < Laurenceb> either i screwed up or the datasheet is wrong
2012-09-27T19:48:48 < zyp> jpa-, either way, you can risk that the compiler allocates a huge frame but doesn't touch that specific word that you set a watchpoint on
2012-09-27T19:50:16 < jpa-> zyp: true
2012-09-27T19:50:44 < Laurenceb> this is odd- timers are half the speed they should be
2012-09-27T19:50:59 <+Steffanx> That's not odd
2012-09-27T19:51:03 <+Steffanx> That's even
2012-09-27T19:51:07 < jpa-> 32 bytes is the minimum region for MPU
2012-09-27T19:51:08 < jpa-> not too bad
2012-09-27T19:51:15 < Laurenceb> im getting APB frequency
2012-09-27T19:51:23 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Remote host closed the connection]
2012-09-27T19:53:01 < jpa-> i wonder why ARM doesn't just provide a block where you can set "stack pointer may not go below the value of this register"
2012-09-27T19:53:18 < zyp> jpa-, I can just put all stacks next to each other and write protect the other threads' stacks, disallowing a thread to write into another thread's stack (through a buffer pointer or something) is not that unreasonable
2012-09-27T19:54:05 < zyp> if I'm putting the stacks into ccm, I already have to disallow using stack buffers for dma
2012-09-27T19:54:33 < jpa-> true
2012-09-27T19:55:44 < Laurenceb> lol wtf
2012-09-27T19:55:59 < Laurenceb> looks like the datasheet master slave mode mapping is all wrong
2012-09-27T19:56:02 < Laurenceb> for f4 timers
2012-09-27T19:56:05 < Laurenceb> funtimes
2012-09-27T19:56:22 < jpa-> bug is more often in your code than in the datasheet
2012-09-27T19:57:08 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-09-27T19:58:28 < Tectu> lol
2012-09-27T20:02:13 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-09-27T20:03:15 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-27T20:03:34 -!- claude is now known as Claude
2012-09-27T20:06:16 -!- BjoernC [~BjoernC@dyndsl-178-142-054-065.ewe-ip-backbone.de] has joined ##stm32
2012-09-27T20:16:23 < Thorn> so if you drive both driver enable and /receiver enable of something like adm3485 to inactive, it will effectively enter a standby (low power) mode?
2012-09-27T20:17:52 < zyp> what about reading the «LOW POWER SHUTDOWN MODE» section in the datasheet?
2012-09-27T20:18:57 < Thorn> if you insist.
2012-09-27T20:22:24 < zippe> zyp: it
2012-09-27T20:22:33 < zippe> it's very hard to use the MPU to protect the stack in all cases
2012-09-27T20:22:57 < zippe> zyp: easier if you have static stack allocation of course
2012-09-27T20:23:15 < zippe> But the power-of-2 size, power-of-2-align behaviour of the MPU is exceptionally retarded.
2012-09-27T20:31:18 -!- Claude is now known as claude
2012-09-27T20:48:09 < zyp> I haven't read the MPU documentation closely yet, so I'm not aware of all the limitations
2012-09-27T20:57:53 -!- barthess [~barthess@178.154.56.70] has quit [Remote host closed the connection]
2012-09-27T21:00:38 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-09-27T21:03:08 -!- claude is now known as Claude
2012-09-27T21:03:43 -!- barthess [~barthess@178.154.53.213] has joined ##stm32
2012-09-27T21:13:44 -!- tavish [~tavish@59.177.6.188] has joined ##stm32
2012-09-27T21:13:44 -!- tavish [~tavish@59.177.6.188] has quit [Changing host]
2012-09-27T21:13:44 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-27T21:24:44 -!- Claude is now known as claude
2012-09-27T21:43:06 -!- DaKu is now known as daku
2012-09-27T22:02:15 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-27T22:14:32 < Thorn> what copper thickness do seeedstudio boards have?
2012-09-27T22:15:11 < Thorn> their website says "inner layer: 0.01...0.018, outer >0.03mm"
2012-09-27T22:15:18 < jpa-> so that is what it is
2012-09-27T22:15:28 < jpa-> 35µm probably
2012-09-27T22:15:42 < emeb> get out your micrometer and do some QA?
2012-09-27T22:16:40 < Thorn> I'm looking at another manufacturer and they have same tolerances as seeed for 18um, for 35um tolerances are larger
2012-09-27T22:17:18 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-27T22:17:26 < jpa-> who cares about copper thickness?
2012-09-27T22:17:56 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-27T22:20:21 < zyp> people who are going to push some current through it
2012-09-27T22:20:53 < emeb> people who are counting on certain parasitics for RF & analog design.
2012-09-27T22:20:55 < jpa-> anyone in their right mind would put some safety margin there anyway
2012-09-27T22:21:23 < jpa-> for accurate impedances, the board material would probably be more important; now it's just "what you happen to get"
2012-09-27T22:21:50 < Thorn> there's a question about that, they say 1.18mil
2012-09-27T22:26:26 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 244 seconds]
2012-09-27T22:37:37 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-27T22:37:46 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-27T22:38:20 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has quit [Ping timeout: 245 seconds]
2012-09-27T22:39:46 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-27T22:44:04 < Laurenceb_> sexy tiem
2012-09-27T22:44:42 < Laurenceb_> my datalogger CE emission results came back
2012-09-27T22:45:16 <+Steffanx> TEST FAILED
2012-09-27T22:45:22 <+Steffanx> *NOT PASSED
2012-09-27T22:46:27 < Laurenceb_> how did you know?
2012-09-27T22:46:35 < Laurenceb_> yeah it failed
2012-09-27T22:47:14 < TitanMKD> Laurenceb because it never work at 1st time ;)
2012-09-27T22:47:30 < Laurenceb_> well.. kind of
2012-09-27T22:47:44 < Laurenceb_> it only just failed at some frequencies some of the time
2012-09-27T22:47:46 < TitanMKD> at my job ther was always problems with CEM stuff
2012-09-27T22:47:56 < Laurenceb_> we are going to redo the test with some creative fudges
2012-09-27T22:48:08 < Laurenceb_> and it should pass
2012-09-27T22:48:11 < TitanMKD> ferrite works very fine sometimes ;)
2012-09-27T22:48:17 < Laurenceb_> nah
2012-09-27T22:48:25 < Laurenceb_> just arrangement in the chamber
2012-09-27T22:48:32 < Laurenceb_> and data averaging
2012-09-27T22:49:00 < TitanMKD> I remember a guy validating a bluetooth hw it was a real pain and it failed totally
2012-09-27T22:49:14 -!- barthess [~barthess@178.154.53.213] has quit [Ping timeout: 240 seconds]
2012-09-27T22:49:29 < TitanMKD> in fact the hw just burn with some test
2012-09-27T22:50:27 < Laurenceb_> interestingly there is stronger emission at 144mhz than 72mhz
2012-09-27T22:50:33 < Laurenceb_> from the 72mhz F1
2012-09-27T22:51:13 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-27T22:51:54 < TitanMKD> Laurenceb hmm such sensibility is on Power Supply ?
2012-09-27T22:52:03 < Laurenceb_> huh?
2012-09-27T22:52:21 < Laurenceb_> i think its coming from the decoupling on the VDD lines yes
2012-09-27T22:52:37 < TitanMKD> so maybe an easy fix is to add the right capacitor
2012-09-27T22:58:38 < Laurenceb_> the excessive emission is at slightly less than 72mhz
2012-09-27T22:58:45 < Laurenceb_> seems to be from spi or something
2012-09-27T23:15:59 < Laurenceb_> http://www.youtube.com/watch?v=Wi-yZsTbCYE
2012-09-27T23:19:00 < Laurenceb_> Pull is not possible because you have unmerged files.  <- how do i fix that from git?
2012-09-27T23:22:02 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-27T23:22:03 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-27T23:22:03 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-27T23:22:06 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-27T23:30:11 < Thorn> why does f405r (lqfp64) have 4 power pins and only 2 ground?
2012-09-27T23:30:15 < zyp> Laurenceb_, by not having unmerged files
2012-09-27T23:30:24 < Laurenceb_> grumble
2012-09-27T23:30:29 < zyp> git stash
2012-09-27T23:30:44 < zyp> then pull, then stash pop
2012-09-27T23:31:03 < BrainDamage> stash = puts your noncommitted changes away temporarily
2012-09-27T23:31:18 < Laurenceb_> git stash
2012-09-27T23:31:18 < Laurenceb_> No local changes to save
2012-09-27T23:31:26 < zyp> oh, right
2012-09-27T23:31:32 < zyp> what does status say?
2012-09-27T23:31:43 < Laurenceb_> build/ch.dmp: needs merge
2012-09-27T23:31:44 < Laurenceb_> build/ch.hex: needs merge
2012-09-27T23:31:45 < Laurenceb_> etc
2012-09-27T23:31:47 < Laurenceb_> binaries
2012-09-27T23:31:55 < BrainDamage> why did you commit binaries ...
2012-09-27T23:32:09 < zyp> do you have any uncommitted changes?
2012-09-27T23:32:12 < BrainDamage> git add the version you want to keep
2012-09-27T23:32:18 < zyp> if not, do: git reset --hard HEAD
2012-09-27T23:32:22 < BrainDamage> I men git add the binary files
2012-09-27T23:32:25 < BrainDamage> then git commit
2012-09-27T23:32:33 < BrainDamage> this is git's way to solve merge conflicts
2012-09-27T23:34:06 < BrainDamage> also, remove the binary files from the repo ...
2012-09-27T23:34:55 < Thorn> I now remember where I saw a repo full of .o files on github.
2012-09-27T23:35:08 < Laurenceb_> wtf
2012-09-27T23:35:16 < Laurenceb_> loads of junk in the files now
2012-09-27T23:35:17 < Laurenceb_> >>>>>>> 0279a01e953f0a20d644fdfe0bad7bb303b7cd5a
2012-09-27T23:35:27 < BrainDamage> that's git's way to show the conflicts
2012-09-27T23:35:40 < Laurenceb_> ewww
2012-09-27T23:35:53 < BrainDamage> it's actually rather standard
2012-09-27T23:35:55 < zyp> git and every other merge tool
2012-09-27T23:35:59 < Laurenceb_> hmm
2012-09-27T23:36:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-27T23:36:05 < Laurenceb_> ill wipe the directory thewn
2012-09-27T23:36:09 < Laurenceb_> and clone again
2012-09-27T23:36:11 < BrainDamage> it won't fix the problem
2012-09-27T23:36:22 < BrainDamage> the problem is that the changes are conflicting
2012-09-27T23:36:27 < BrainDamage> and git has no idea what to do
2012-09-27T23:36:28 < Laurenceb_> hmm
2012-09-27T23:36:31 < zyp> Laurenceb_, you don't have to do that, just use reset --hard
2012-09-27T23:36:35 < BrainDamage> so it's asking you
2012-09-27T23:36:38 < Laurenceb_> i did
2012-09-27T23:36:47 < BrainDamage> do you have a diff tool like meld?
2012-09-27T23:36:49 < BrainDamage> git mergetool
2012-09-27T23:36:56 < BrainDamage> and it'll open the files in your diff editor
2012-09-27T23:37:06 < BrainDamage> so it'll be way more understandable
2012-09-27T23:37:12 < BrainDamage> provided you know how to use a diff editor
2012-09-27T23:37:23 < zyp> BrainDamage, no point in wasting time merging shit you don't have to merge though
2012-09-27T23:37:28 < Laurenceb_> yeah
2012-09-27T23:37:36 < Laurenceb_> the stuff at this end is some random junk
2012-09-27T23:37:40 < Laurenceb_> i dont care about it
2012-09-27T23:37:47 < Laurenceb_> there werer a couple of changes to some files
2012-09-27T23:37:52 < BrainDamage> then git reset --hard headyouwanttouseasreference
2012-09-27T23:37:58 < Laurenceb_> hmm
2012-09-27T23:38:04 < zyp> yeah
2012-09-27T23:38:07 < BrainDamage> like say, git reset --hard otherrepo/master
2012-09-27T23:38:08 < Laurenceb_> so the reference can be ages in the past?
2012-09-27T23:38:12 < BrainDamage> anything
2012-09-27T23:38:13 < Laurenceb_> oh
2012-09-27T23:38:15 < BrainDamage> even another repo
2012-09-27T23:38:42 < zyp> a branch is really just a pointer to a commit, i.e. the tip of that branch
2012-09-27T23:38:55 < Laurenceb_> got it
2012-09-27T23:38:57 < zyp> git reset changes that pointer to whatever
2012-09-27T23:38:58 < Laurenceb_> i see
2012-09-27T23:39:06 < Laurenceb_> its sane now, thanks
2012-09-27T23:39:09 < zyp> and --hard updates the files along with it
2012-09-27T23:39:42 < BrainDamage> there's one downside of using reset tough
2012-09-27T23:39:48 < BrainDamage> if you change your history, and push
2012-09-27T23:39:54 < BrainDamage> the histories won't align
2012-09-27T23:40:06 < BrainDamage> and git will hate you, and ask to use force to rewrite history
2012-09-27T23:40:11 < zyp> ah, right
2012-09-27T23:40:25 < zyp> this shit happened because you rewrote the history and force-pushed it
2012-09-27T23:40:40 < zyp> and then tried to pull at another copy
2012-09-27T23:41:04 < zyp> you shouldn't have used pull at all, just fetch and reset
2012-09-27T23:43:16 < zyp> pull is just fetch and merge
2012-09-27T23:43:41 < zyp> so you fetched the rewritten history from upstream and then tried to merge that with the original history in the local repo
2012-09-27T23:48:27 < BrainDamage> Laurenceb_: http://dl.dropbox.com/u/40949413/git-pretty.png
2012-09-27T23:49:10 < zyp> oh
2012-09-27T23:49:11 <+izua> zyp: not familiar with git.
2012-09-27T23:49:20 < zyp> izua, hmm?
2012-09-27T23:49:29 <+izua> doesn't that esentially mean that you always code+libs that worked at some point
2012-09-27T23:49:32 < zyp> BrainDamage, nice, I like that
2012-09-27T23:49:46 <+izua> and if some guy that branched long time ago wishes to update, it's his job to solve conflicts and not the developers?
2012-09-27T23:50:06 < Laurenceb_> heh
2012-09-27T23:50:15 <+izua> hmph. that does seem smart, plus, no need for a central server i think.
2012-09-27T23:50:30 < zyp> izua, yes? but I'm not really sure what you are asking…
2012-09-27T23:50:46 <+izua> not asking anything
2012-09-27T23:50:48 < zyp> git is decentralized, it's a feature
2012-09-27T23:51:05 <+izua> just thinking out loud
2012-09-27T23:51:24 < zyp> you can use it in a centralized fashion, but every clone of it is a fully standalone repository
2012-09-27T23:51:47 < zyp> you had a question mark after your text
2012-09-27T23:51:59 < zyp> 22:49:46 <+izua> and if some guy that branched long time ago wishes to update, it's his job to solve conflicts and not the developers?
2012-09-27T23:52:08 < zyp> conflicts are solved by the merger
2012-09-27T23:52:38 < Laurenceb_> http://en.wikipedia.org/wiki/Islamic_Rage_Boy
2012-09-27T23:52:41 < Laurenceb_> ^lulwut
2012-09-27T23:52:42 < zyp> if you want to merge updates to your copy, you are responsible for any conflicts
2012-09-27T23:52:49 -!- daku is now known as DaKu
2012-09-27T23:53:32 < zyp> BrainDamage, I like the «I do not think it means what you think it means.» note on revert :p
2012-09-27T23:54:25 < BrainDamage> we should also determine whenever ##stm32 counts as lynch mob
2012-09-27T23:54:55 < zyp> are we downstream?
2012-09-27T23:55:59 < BrainDamage> mmm, when Laurenceb_ asks for help, it might not be a bad idea to fetch, so somehow downstream yes
2012-09-27T23:56:26 < Laurenceb_> wut
2012-09-27T23:56:41 < zyp> but we are not keeping working branches of his stuff, so there won't be any merge conflicts
2012-09-27T23:56:45 < zyp> so it's pretty irrelevant :p
2012-09-27T23:57:31 < BrainDamage> still annoyed from seeing history rewritten over & over :p
2012-09-27T23:58:13 < Thorn> why would anyone use git when there's mercurial?
2012-09-27T23:58:28 < zyp> why would anyone use mercurial when there's git?
2012-09-27T23:58:58 < Thorn> mercurial actually cares about your files and will never clobber them
2012-09-27T23:59:21 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-146.pptp.stw-bonn.de] has joined ##stm32
2012-09-27T23:59:26 < zyp> how is that supposed to be an argument?
2012-09-27T23:59:46 < BrainDamage> if you bother start with an argument, don't handwave stuff
--- Day changed Fri Sep 28 2012
2012-09-28T00:01:03 < zyp> anyway, I'm not intending to start a flamewar here, I've never had to use mercurial, but from what I understand it's about on par with git
2012-09-28T00:01:36 < BrainDamage> yes except for small stuff, distributed vcs are mainly the same nowdays
2012-09-28T00:01:42 < BrainDamage> and steal features from eachother
2012-09-28T00:01:44 < Thorn> it has less features, is much simpler to use, messages are actually understandable, etc.
2012-09-28T00:02:13 < zyp> I've been using git for my projects for a few years now, and I'm using git daily at work, so it's the tool I know and love :p
2012-09-28T00:02:15 < Thorn> and it will never do anything destructive unless you really ask it to
2012-09-28T00:02:23 < zyp> neither would git.
2012-09-28T00:04:01 <+izua> neither would dd.
2012-09-28T00:04:14 < Thorn> exactly
2012-09-28T00:04:46 < zyp> Thorn, what are you finding destructive about git?
2012-09-28T00:06:00 < Thorn> ability to edit history
2012-09-28T00:06:10 <+izua> zyp: you were using scons, right?
2012-09-28T00:06:15 < zyp> izua, yes
2012-09-28T00:06:27 <+izua> is your set of goodies for stm32 available anywhere?
2012-09-28T00:06:41 < zyp> izua, of course, check the laks repo
2012-09-28T00:06:55 < zyp> Thorn, you can't do that in mercurial?
2012-09-28T00:07:50 < Thorn> no (or maybe there is some extremely limited way, I only used hg revert to undo last commit)
2012-09-28T00:07:50 < zyp> according to a quick google search, you can do hg commit --amend, that's editing history.
2012-09-28T00:08:29 < zyp> hmm, there is also a hg histedit, which is comparable to interactive rebasing
2012-09-28T00:08:39 < zyp> so I'm calling bullshit on your argument
2012-09-28T00:09:05 < Thorn> you found that because you're a git user.  I'm pretty sure it's not in any tutorial or guide
2012-09-28T00:09:21 < BrainDamage> ...
2012-09-28T00:09:37 < zyp> either way, history editing is not destructive
2012-09-28T00:10:21 < zyp> disruptive would probably be a better word :p
2012-09-28T00:10:54 < Erlkoenig> just don't use it if you don't like it :D
2012-09-28T00:11:49 < BrainDamage> I think like language "features" the annoyance stems when someone elses you collaborate more or less directly with does use them, and this annoys you
2012-09-28T00:13:14 < BrainDamage> so basically enforce your ideas by tool choice
2012-09-28T00:22:54 -!- BjoernC [~BjoernC@dyndsl-178-142-054-065.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-28T00:40:57 <+Steffanx> Hey flyback is ba ck :D
2012-09-28T00:41:11 < zyp> flyback, url?
2012-09-28T00:43:19 <+Steffanx> freescale ..
2012-09-28T00:43:56 <+Steffanx> For me freescale = freescale
2012-09-28T00:44:32 <+Steffanx> And Freescale isn't part of motorola anymore
2012-09-28T00:45:16 <+Steffanx> So freescale = freescale :)
2012-09-28T00:46:03  * Steffanx throws a wooden shoe at flyback 
2012-09-28T00:46:20 < zyp> hmm, it's even in stock
2012-09-28T00:46:37 < Thorn> http://stm32.izua.ro/wiki/USB
2012-09-28T00:46:39 < zyp> I'd order one right now if it got free shipping
2012-09-28T00:46:50 <+Steffanx> in stock @ element14?
2012-09-28T00:46:50 < zyp> just because I like collecting crap
2012-09-28T00:46:54 < Thorn> hopefully converted more or less well
2012-09-28T00:46:54 <+Steffanx> *in stock
2012-09-28T00:47:51 < zyp> Thorn, that article seems to suffer from information overload
2012-09-28T00:49:06 < zyp> it tries to cover a lot, but doesn't really say much about each
2012-09-28T00:49:22 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-28T00:49:28 < zyp> and I'm not sure how comprehensible it is to people new to USB either
2012-09-28T00:50:04 < Thorn> fixed some links
2012-09-28T00:51:07 < Thorn> well. that's basically my notes on the subject
2012-09-28T00:51:27 < Thorn> maybe these can be made into something generally useful
2012-09-28T00:52:20 < zyp> yeah, I thought it mostly looked like personal notes :p
2012-09-28T00:52:32 < Thorn> it is
2012-09-28T00:54:56 < zyp> «On device, all traffic flows to/from endpoints (= memory buffers for data received or waiting to be sent).»
2012-09-28T00:55:34 < zyp> I prefer to view endpoints as ports
2012-09-28T00:56:39 < zyp> just like tcp ports separate stuff on one ip, endpoints separate stuff on one device
2012-09-28T00:57:23 < zyp> how the memory buffers behind them work is pretty hardware dependant
2012-09-28T00:57:39 < Thorn> the point is you have to copy data into the buffer beforehand
2012-09-28T00:58:07 < zyp> the otg controllers use a single fifo for all the receiving endpoints
2012-09-28T00:59:07 < Thorn> anyway, a wiki is meant to be edited.
2012-09-28T00:59:18 < zyp> that's an implementation detail
2012-09-28T01:00:11 < zyp> a usb controller with a master memory port and guaranteed low enough memory latency could just pull data directly out of main memory while transmitting
2012-09-28T01:00:24 < zyp> or write directly to main memory
2012-09-28T01:00:37 < zyp> :p
2012-09-28T01:00:56 < Thorn> $ cat>wiki plz.
2012-09-28T01:01:18 < Thorn> what is the state of data lines when no device is connected btw?
2012-09-28T01:01:58 < zyp> SE0
2012-09-28T01:02:00 < zyp> host has a weak pulldown on both
2012-09-28T01:02:21 < zyp> and then you connect a stronger pullup on one of the lines to signal presence
2012-09-28T01:03:48 < Thorn> why do they say everywhere that "host reset the device when it's detected"? it's only logical that in order to detect a pullup, you need a low level, and that is reset
2012-09-28T01:04:14 < zyp> no.
2012-09-28T01:04:53 < zyp> initially the bus is weakly pulled down, then you get a stronger pullup to signal presence
2012-09-28T01:05:22 < zyp> then the host asserts a reset by driving down both lines, i.e. overpowering the pullup
2012-09-28T01:05:36 < Laurenceb_> tldr
2012-09-28T01:05:39 < Laurenceb_> its complicated
2012-09-28T01:05:43 < Thorn> ah
2012-09-28T01:06:41 < zyp> the pulls only matter when the bus is idle, they are too weak to matter when one end is driving the bus
2012-09-28T01:09:27 < Laurenceb_> this is odd
2012-09-28T01:09:40 < Laurenceb_> my pwm frequency halves after a while
2012-09-28T01:09:52 <+izua> what?
2012-09-28T01:10:24 <+izua> your scope's knob could be randomly jumping to the next value, too
2012-09-28T01:10:31 < Laurenceb_> ive got  a chain of gated timers
2012-09-28T01:10:36 <+izua> random guess is random.
2012-09-28T01:11:40 < Laurenceb_> lol
2012-09-28T01:11:56 < Laurenceb_> im suspecting the f4 datasheet is wrong
2012-09-28T01:12:04 < Laurenceb_> but it doesnt explain this weirdness
2012-09-28T01:14:49 < Laurenceb_> grumble
2012-09-28T01:15:03 < Laurenceb_> tim10 output to tim9 doesnt seem to work either
2012-09-28T01:15:04 < Laurenceb_> sucks
2012-09-28T01:16:08 <+izua> are you changing the reset value at run time? like in pulse-frequency modulation?
2012-09-28T01:17:00 < Laurenceb_> nah
2012-09-28T01:17:06 < Laurenceb_> thats why im confused
2012-09-28T01:17:13 < Laurenceb_> oh wait...
2012-09-28T01:17:26 < Laurenceb_> unless they something really stupid with the usb in chibios
2012-09-28T01:17:51 < Laurenceb_> the st f1 code clks down the processor in suspend mode
2012-09-28T01:18:21 < Laurenceb_> and the problem occurs a few seconds after i hit a breakpoint
2012-09-28T01:18:41 < Laurenceb_> no wait
2012-09-28T01:18:55 < Laurenceb_> its not actually happening during the breakpoint
2012-09-28T01:19:11 < Laurenceb_> its like something is happening with clk speed
2012-09-28T01:19:23 < Laurenceb_> the usb breaks
2012-09-28T01:19:51 < Laurenceb_> i wonder if chibios is being too clever for its own good
2012-09-28T01:23:36 <+izua> it's power efficient and green
2012-09-28T01:54:42 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-28T01:54:43 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-28T01:54:43 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-28T02:14:04 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-09-28T02:35:08 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-09-28T03:23:04 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-28T04:13:52 < dongs> hello blogs
2012-09-28T04:14:03 < dongs> is Laurenceb still failing with kawaiiOS
2012-09-28T04:16:58 < Bird|lappy> dongs, I just got one of those "your computer is infected" phone calls
2012-09-28T04:17:17 < Bird|lappy> cracked me up right good, with the poor sap still on the other end of the line
2012-09-28T04:17:20 -!- DaKu is now known as daku
2012-09-28T04:17:37 < dongs> lolwat
2012-09-28T04:17:44 < dongs> is this an american thang
2012-09-28T04:18:11 < Bird|lappy> not American per se
2012-09-28T04:19:10 < Bird|lappy> it's an attack where the crooks use some low-rent call center somewhere (usually in India) to dial people up and tell them "your computer is infected, do this and that and this other thing to fix it"
2012-09-28T04:19:28 < Bird|lappy> but the steps they give you actually download malware onto the mark's computer
2012-09-28T04:19:39 < dongs> haha, apparently http://www.hobbyking.com/hobbyking/store/__26944__ was put at cheapass price against the ideafly wish, so all their customers are fucked now
2012-09-28T04:19:44 < dongs> cause its below dealer cost at hobbyking
2012-09-28T04:20:47 < dongs> ah, thats why its out of stock
2012-09-28T04:20:49 < dongs> they stopped selling to them
2012-09-28T04:20:50 < dongs> heh
2012-09-28T04:41:30 <+izua> Bird|lappy: that's awesome
2012-09-28T04:41:33 <+izua> you should do that
2012-09-28T04:41:39 <+izua> and download the malware in the vm
2012-09-28T04:41:46 <+izua> and pretend to be a real retard
2012-09-28T04:41:49 <+izua> like, how do i download
2012-09-28T04:41:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-28T04:41:53 <+izua> how do i mouse
2012-09-28T04:42:00 <+izua> then reset your vm
2012-09-28T04:42:26 <+izua> and then explain the scammer you restored an earlier clone of your machine
2012-09-28T04:42:57 < qyx_> ask them if you can call them back if something doesn't work with the downloaded software
2012-09-28T04:43:32 <+izua> "can i call you back if the virus will not virus my box right?"
2012-09-28T04:45:04 < dongs> haha
2012-09-28T04:49:10 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 252 seconds]
2012-09-28T05:03:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-28T05:14:13 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 256 seconds]
2012-09-28T05:14:44 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-28T05:17:18 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-28T05:36:58 < emeb_mac> not a creature was stirring, not even a mouse.
2012-09-28T05:42:41 < dongs> total lack of chats
2012-09-28T05:51:49 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-09-28T05:51:50 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-09-28T05:51:50 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-09-28T05:56:50 < HTT-Bird> anyone here worked with the LTC3108 by any chance?
2012-09-28T05:58:59 -!- Erlkoenig [~erlkoenig@pptp-212-201-78-146.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-09-28T06:07:16 < dongs> sounds expensive
2012-09-28T06:07:16 < dongs> every LTC part in china is a clone
2012-09-28T06:07:16 < dongs> and usually shitty one
2012-09-28T06:07:43 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-28T06:10:03 < GargantuaSauce> >mfw i waste 4 hours because i forgot to enable a gpio
2012-09-28T06:15:58 < HTT-Bird> GargantuaSauce: :P gotta love debugging
2012-09-28T06:23:21 < cjbaird> E14.au should be getting their Freescale KL25 boards on Oct 4th.. (just spent more time talking on the mobile to them, to the point where the call cost more than the board..)
2012-09-28T06:24:28 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-28T06:24:35 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-28T06:33:05 < emeb_mac> dongs: got yer 8x8 matrix designed?
2012-09-28T06:34:59 < dongs> emeb_mac: still thinking what to use parts wise. thinking maybe for drivers just use this WS2811 shits I have, this will give me 8bit pwm for free and just toggle rows w/driver+gpio
2012-09-28T06:35:31 < dongs> they're so8 though so.. id have to see if I can fit them in 6x6
2012-09-28T06:37:32 < dongs> 8 of them is half hte price of ST driver
2012-09-28T06:37:43 < emeb_mac> aren't ws2811 I2C?
2012-09-28T06:37:45 < dongs> no
2012-09-28T06:37:50 < dongs> onewire pwm
2012-09-28T06:37:58 < dongs> so its really easy to drive
2012-09-28T06:38:53 < dongs> oh but hmm,, thier refresh is onlt 400 or 800khz
2012-09-28T06:39:06 < emeb_mac> yeah - was wondering about that.
2012-09-28T06:41:12 < dongs> lets see.. in 400khz mode, 2.5us per  bit, 24bits for RGB + 50us pause
2012-09-28T06:41:29 < dongs> 110us to push in one line
2012-09-28T06:43:49 < emeb_mac> heh - and you though the MIC2981 was slow.
2012-09-28T06:43:53 < dongs> yeah
2012-09-28T06:44:14 < dongs> so never mind on 2811s, st shit can clock data in at like 25mhz
2012-09-28T06:45:48 < dongs> im a bit confused which side does pwm to do different brightness
2012-09-28T06:46:01 < dongs> when using STs shit
2012-09-28T06:46:15 < dongs> the gpios driving high side?
2012-09-28T06:46:33 < emeb_mac> since RGB are all common on the high-side driver the ST chip needs to do PWM
2012-09-28T06:47:01 < emeb_mac> Otherwise you just get grayscale, or you need to run RGB on separate cycles.
2012-09-28T06:48:52 < dongs> st has STEVAL-ILL015V1
2012-09-28T06:49:18 < dongs> and has some pwm lib to do it with but they're not scanning rows
2012-09-28T06:49:27 < dongs> they just directly drive 2*8 leds from 2 driver chips heh
2012-09-28T06:49:42 < dongs> the lib sores is a .exe so i didnt bother installing it yet
2012-09-28T06:51:35 < emeb_mac> cute board
2012-09-28T06:59:08 < dongs> so yeah, prob sticking with the ST driver. only wondering if PWM shit will work with multiplexing rows
2012-09-28T06:59:09 < dongs> ..
2012-09-28T06:59:09 < dongs> worthless feenode
2012-09-28T06:59:23 < dongs> lagged trash
2012-09-28T06:59:55 < dongs> emeb_mac: what do you think
2012-09-28T07:00:12 < emeb_mac> well - software PWM, so the software knows where the start/stop points are.
2012-09-28T07:00:24 < emeb_mac> should be fairly easy to insert scanning into that.
2012-09-28T07:00:51 < emeb_mac> Main question is if the PWM rate is high enough that you get a good refresh rate.
2012-09-28T07:01:10 < dongs> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/APPLICATION_NOTE/CD00213809.pdf
2012-09-28T07:01:15 < dongs> page 25
2012-09-28T07:02:08 < emeb_mac> looking...
2012-09-28T07:02:26 < dongs> dunno how they're getting 600+ leds, is that the numbers from chaining multiple drivers?
2012-09-28T07:06:42 < cjbaird> It's damned hot.. And I have to go out in motorcycle gear to pay the rent.. :/ Not even the whiney cat wants to go outside.
2012-09-28T07:06:52 < dongs> 26C outside
2012-09-28T07:07:00 < dongs> perfect weather
2012-09-28T07:07:01 < emeb_mac> cjbaird: where?
2012-09-28T07:07:08 < dongs> sounds like ausfag
2012-09-28T07:07:12 < dongs> isnt it winter time there
2012-09-28T07:07:20 < dongs> or summer i guess whatever.
2012-09-28T07:07:25 < dongs> those weird upside down living peopel
2012-09-28T07:08:27 < cjbaird> 30C here, up from 22 a week ago.
2012-09-28T07:09:49 < cjbaird> That's 30 inside. Was just testing the contactless thermo outside, and it scored from 33-37C
2012-09-28T07:09:51 < emeb_mac> getting cooler here in AZ - only 98F today. Supposed to be > 100F over the weekend though
2012-09-28T07:10:35 < cjbaird> The offshore ocean temp is still 'cold', 18C.. Charl and I used to wait until it was at least 23.
2012-09-28T07:10:37 < emeb_mac> dongs:  looks like they're driving all the leds in parallel.
2012-09-28T07:10:52 < dongs> emeb_mac: right
2012-09-28T07:11:37 < dongs> emeb_mac: but, tarduino can do 32level grayscale with 595s and 16mhz cpu, surely stm32 can do better.
2012-09-28T07:12:22 < emeb_mac> dongs: I'm sure it can.
2012-09-28T07:12:54 < emeb_mac> this isn't rocket science. Just need to come up with a good timing plan.
2012-09-28T07:14:51 < dongs> btw that mic part was more expensive from china than digikey heh
2012-09-28T07:14:59 < emeb_mac> bizarre
2012-09-28T07:15:37 < dongs> i need to check if A and mic are same width and ill prob just do 20pin soic pattern
2012-09-28T07:15:46 < dongs> this way they can be interchangeable
2012-09-28T07:16:25 < emeb_mac> what's the advantage of making it work with the A part?
2012-09-28T07:16:53 < dongs> none i guess, other than possibility of clone/cheper parts available in soic20
2012-09-28T07:17:04 < dongs> the pinout is same
2012-09-28T07:17:09 < dongs> just bottom 2 pins are unused
2012-09-28T07:17:22 < dongs> so for soic18 it'll jsut have 2 pads open
2012-09-28T07:18:24 < dongs> looks liek it ll work
2012-09-28T07:18:25 < dongs> same soicw
2012-09-28T07:19:38 < emeb_mac> might as well.
2012-09-28T07:21:06 < dongs> what to do about chaining morethan one block?
2012-09-28T07:22:18 < emeb_mac> add more MIC chips and scan more rows.
2012-09-28T07:22:22 < dongs> no
2012-09-28T07:22:22 < dongs> i mean
2012-09-28T07:22:29 < dongs> more than one 8x8 block
2012-09-28T07:22:45 < emeb_mac> yeah - do 8x16 with 2 MIC chips
2012-09-28T07:22:53 < emeb_mac> still on ST chip
2012-09-28T07:23:02 < emeb_mac> s/on/one/
2012-09-28T07:23:27 < emeb_mac> timing depends on how much PWM res you want.
2012-09-28T07:23:31 < dongs> right, but i mean if I wanna chain like 10 of htem together.
2012-09-28T07:23:37 < emeb_mac> Ah
2012-09-28T07:23:38 < dongs> should do some sorta serial in/out shit
2012-09-28T07:23:54 < emeb_mac> separate STM32 per 8x8 block?
2012-09-28T07:23:59 < dongs> yeah
2012-09-28T07:24:33 < emeb_mac> set them up as SPI slaves with clock-thru?
2012-09-28T07:24:47 < emeb_mac> don't know if STM32 supports that tho.
2012-09-28T07:25:24 < dongs> is tehre 3 SPIs on F1013Cx? i thought theres only 2. and no i dont think it does clock through
2012-09-28T07:26:36 < emeb_mac> hmmm
2012-09-28T07:27:49 < dongs> Up to 2 SPIs (18 Mbit/s)
2012-09-28T07:28:52 < dongs> one will be gone for hte driver chip
2012-09-28T07:30:23 < emeb_mac> yeah
2012-09-28T07:31:28 < emeb_mac> wonder if you can simulate clock-thru so mosi = miso + delay?
2012-09-28T07:32:18 < dongs> uarts will be open but thats probably too slow?
2012-09-28T07:32:45 < emeb_mac> by reading then writing same data.
2012-09-28T07:33:12 < emeb_mac> probably
2012-09-28T07:43:54 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-09-28T07:47:15 < dongs> http://seeedstudio.com/wiki/images/0/05/Rainbowduino_V3.0b.pdf
2012-09-28T07:47:19 < dongs> tarduino just passes uart across.
2012-09-28T07:47:43 < emeb_mac> I've got it! CAN!
2012-09-28T07:47:56 < dongs> haha
2012-09-28T07:48:00 < dongs> i KNEW you were gonna say that.
2012-09-28T07:48:08 < emeb_mac> heh :P
2012-09-28T07:50:13 < dongs> but how would you chain it with buffering? 2 drivers?
2012-09-28T07:50:24 < dongs> plus then each shit would need to have some sorta can ID preset
2012-09-28T07:50:31 < emeb_mac> CAN bus is a multi-drop network
2012-09-28T07:50:35 < dongs> yes
2012-09-28T07:50:42 < emeb_mac> all nodes hook to the same two lines.
2012-09-28T07:50:50 < emeb_mac> not chain - star
2012-09-28T07:51:19 < dongs> right, but thats not really useful in terms of plugging each module into another
2012-09-28T07:51:46 < emeb_mac> sure it is - just pass the data lines thru from module to module
2012-09-28T07:51:56 < emeb_mac> but yeah - need addressing setup somehow.
2012-09-28T07:51:59 < dongs> wont the signal turn to shit by the time youre in nth module?
2012-09-28T07:52:20 < dongs> since its passing across the pcb in each one
2012-09-28T07:52:40 < emeb_mac> it could if you don't design the parasitics & terminations right.
2012-09-28T07:53:12 < mrnuke> aaargh stm32 library sucks
2012-09-28T07:53:21 < dongs> no!! yuo sucks!! its great
2012-09-28T07:54:14 < mrnuke> oh come on, it gets people to write comments like "Initialize User Button"
2012-09-28T07:54:30 < mrnuke> the button is there, there's nothing to initialize
2012-09-28T07:54:41 < mrnuke> oh wait, you're using a library with too much indirection
2012-09-28T07:54:51 < mrnuke> sorry man, it's just in my sanezone :)
2012-09-28T07:55:02 < mrnuke> *just not in my sanezone
2012-09-28T07:56:32 < emeb_mac> the STM32 peripheral library is a bit of a disaster.
2012-09-28T07:56:50 < emeb_mac> I've been using it. it works, but it's top-heavy.
2012-09-28T07:56:59 < emeb_mac> lots of code bloat.
2012-09-28T07:57:09 < dongs> ohh!!
2012-09-28T07:57:15 < dongs> emeb_mac: http://www.my-semi.com/content/products/product_mean.aspx?id=4 the shit rainblowduiono uses
2012-09-28T07:57:22 < dongs> 12channels but PWM
2012-09-28T07:57:39 < dongs> up to 16bit grayscale!
2012-09-28T07:58:01 < emeb_mac> nice
2012-09-28T07:58:35 < mrnuke> dongs: dang man, the fact that it works don't make it good
2012-09-28T08:00:03 < dongs> emeb_mac: i got it
2012-09-28T08:00:04 < dongs> DMX512
2012-09-28T08:00:13 < dongs> (for chain)
2012-09-28T08:00:26 < dongs> though i guess its just as retarded as can
2012-09-28T08:00:29 < dongs> still need addressing
2012-09-28T08:06:16 < emeb_mac> don't know much about dmx
2012-09-28T08:06:21 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-28T08:06:21 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-28T08:06:21 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-28T08:06:35 < dongs> my pal does some led crap wiht dmx
2012-09-28T08:06:41 < dongs> its just rs485 w/some protocol
2012-09-28T08:08:34 < dongs> http://www.my-semi.com/content/products/product_mean.aspx?id=14 hmm this is neat... it has built in shit for scanning
2012-09-28T08:09:04 < dongs> but 16channel, fail
2012-09-28T08:11:32 < emeb_mac> http://www.digikey.com/product-detail/en/TLC5947RHBT/296-23820-1-ND/1958758
2012-09-28T08:11:38 < dongs> right
2012-09-28T08:11:40 < dongs> i made stuff with that before
2012-09-28T08:11:47 < emeb_mac> $$$
2012-09-28T08:12:02 < emeb_mac> costs more than the MCU
2012-09-28T08:12:10 < dongs> 3x more than mcu yeah.
2012-09-28T08:12:22 < dongs> though i think my ledguy is buying those in reels so it might be cheaper but still $
2012-09-28T08:13:09 < emeb_mac> a little cheaper: http://www.digikey.com/product-detail/en/TLC5951RHAR/296-25567-1-ND/2197046
2012-09-28T08:13:21 < emeb_mac> only 2x MCU
2012-09-28T08:13:41 < dongs> huh is that new
2012-09-28T08:13:53 < dongs> oh, non stock
2012-09-28T08:13:55 < dongs> must be somew eird shit
2012-09-28T08:14:16 < emeb_mac> http://www.ti.com/product/tlc5951 <- 2009
2012-09-28T08:15:00 < dongs> wtf it has separate clocks for rgb?
2012-09-28T08:15:10 < LeelooMinai> Hi, I just looked at CPAL layer or whatever they call it - des enyone here use it or you just use the std library or even direct register access to handle I2C. I ask because I don't have a god feeling about using it:)
2012-09-28T08:15:14 < dongs> looks really fucked to interface
2012-09-28T08:15:30 < dongs> LeelooMinai: fuck no, cpal is such horribly overengineered shit
2012-09-28T08:15:42 < dongs> i think i looked at it once
2012-09-28T08:15:48 < dongs> theres literally liek 5 layers of abstraction
2012-09-28T08:16:05 < emeb_mac> dongs: you could prolly chain the RGB channels together. Makes one device look 3x longer.
2012-09-28T08:16:20 < LeelooMinai> Well, it also seems that they kind of plan it for some protocols, but only I2C is used... so... maybe no point in bothering with that
2012-09-28T08:17:00 < LeelooMinai> How about the peripheral library?  Is it being used by you?
2012-09-28T08:17:24 < emeb_mac> I'm using it - it's bloated but easy & has lots of example code.
2012-09-28T08:17:51 < LeelooMinai> Hmm, I think I will jus go with it for I2C and skip that strange CPAL
2012-09-28T08:25:08 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-09-28T08:38:54 -!- claude is now known as Claude
2012-09-28T08:51:34 < cjbaird> I've got to wonder what application this http://www.ezwind.net/brazil/BrazilPDP.jpg is being used for, to keep it running 2012..
2012-09-28T08:52:23 < dongs> drugs
2012-09-28T08:53:18 < dongs> how many combinations are there aa..ZZ
2012-09-28T08:53:40 < emeb_mac> (26*2)^2
2012-09-28T08:53:55 < dongs> mkay
2012-09-28T09:03:52 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Disconnected by services]
2012-09-28T09:03:53 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-28T09:05:01 -!- Nutter` is now known as Nutter
2012-09-28T09:05:31 < dongs> hmm. i only get 676
2012-09-28T09:06:04 < dongs> o crap
2012-09-28T09:14:04 < emeb_mac> (26*2)^2 = 2704
2012-09-28T09:14:15 < emeb_mac> 26^2 = 676
2012-09-28T09:17:14 < dongs> yaeh.. my loop was wrong
2012-09-28T09:23:28 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-28T09:23:45 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-28T09:24:12 < cjbaird> If you owned a HP35s like yours truly, you would never have made such a trivial error.
2012-09-28T09:36:34 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-28T09:47:21 < emeb_mac> like the HP35?
2012-09-28T09:47:45  * emeb_mac just used octave
2012-09-28T09:55:43 < cjbaird> BSD's seq can only do numbers, not stuff like [0-9a-Z], foo.
2012-09-28T10:03:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-28T10:21:29 -!- Claude is now known as claude
2012-09-28T11:01:45 < dongs> hmmm
2012-09-28T11:01:50 < dongs> MY9221 is $.50
2012-09-28T11:12:30 -!- claude is now known as claude_work
2012-09-28T11:13:26 -!- Thorn [~Thorn@unaffiliated/thorn] has quit []
2012-09-28T11:14:15 < dongs> it certainly looks like a better choice than ST chip, since it can do pwm and shit without extra work.
2012-09-28T11:14:51 < zyp> what are you making?
2012-09-28T11:15:01 < dongs> zyp,  led driver
2012-09-28T11:15:08 < zyp> for what?
2012-09-28T11:15:16 < dongs> just fucking around
2012-09-28T11:15:17 < dongs> 8x8 matrix
2012-09-28T11:25:59 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-28T11:27:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-28T11:33:39 < cjbaird> That sounds like a unique and worthwhile product.
2012-09-28T11:41:15 -!- BrainDamage [~yaaic@217.200.200.237] has joined ##stm32
2012-09-28T11:44:50 < dongs> totally.
2012-09-28T11:45:02 < dongs> i will be an instant MAKE:R celebrity
2012-09-28T11:45:09 < dongs> might even take it to kickstarter
2012-09-28T11:49:11 < cjbaird> And your very own female blogger with tattoos.
2012-09-28T11:52:18 < cjbaird> Hey, I already have one of those. It's a Transformers Autobot tat, though. :/.
2012-09-28T11:55:36 < dongs> too bad youre neither female or blogger
2012-09-28T11:58:08 -!- BrainDamage [~yaaic@217.200.200.237] has quit [Quit: Yaaic]
2012-09-28T12:03:18 < cjbaird> http://ozphoenix13.deviantart.com/art/My-Autobot-tattoo-165682379 :/ .. She might not have gone through with that if she'd know she was going to score a boyfriend a few months later. :)
2012-09-28T12:05:20 < GargantuaSauce> great success http://workman-industries.net/images/f4_mb.jpg
2012-09-28T12:06:33 < cjbaird> soft-float or hardFP on the mandelbrot?
2012-09-28T12:06:42 < GargantuaSauce> hard
2012-09-28T12:06:53 < GargantuaSauce> it's the f4
2012-09-28T12:07:00 < cjbaird> yep, why I asked.
2012-09-28T12:07:35 < GargantuaSauce> that lcd controller is kind of a pain
2012-09-28T12:07:45 < GargantuaSauce> it's the ssd1289
2012-09-28T12:08:33 < GargantuaSauce> if i double the max iterations of the fractal it sometimes get bored and forgets i'm jamming pixels into it and fucks up
2012-09-28T12:09:24 < dongs> youre using way to many wires to talk to it
2012-09-28T12:09:30 < dongs> i have one of those chinq lcds and i did it by spi
2012-09-28T12:09:36 < GargantuaSauce> it's a 16 bit parallel interface
2012-09-28T12:09:50 < GargantuaSauce> i could use a bunch of shift registers but fuck that
2012-09-28T12:11:11 < GargantuaSauce> pretty convenient that it's a whole gpio though, i can just copy the word to the register and strobe the write line on another one
2012-09-28T12:12:08 < GargantuaSauce> sure wouldn't be getting a 2.5 mhz pixel clock with a serial interface
2012-09-28T12:12:40 < GargantuaSauce> (but i'm certainly not with the quadratic map being iterated for each pixel)
2012-09-28T12:15:19 < dongs> lol what the fuck
2012-09-28T12:15:23 < dongs> why arent you using fsmc
2012-09-28T12:16:27 < GargantuaSauce> because i am wallowing in noobery
2012-09-28T12:19:44 < dongs> must b e
2012-09-28T12:20:32 < GargantuaSauce> i didn't know that was a thing until you mentioned it
2012-09-28T12:34:21 < dongs> GargantuaSauce: its good for this kinda shit.
2012-09-28T12:34:40 < dongs> should make a shitload easier to interface to 8/16bit wide lcds like that.
2012-09-28T12:34:58 < zyp> IM JUST GOING TO HOOK THIS SHIT UP HERE AND BANG THIS PIN ON AND OFF
2012-09-28T12:40:11 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-28T12:47:08 < jpa-> and when you use FSMC, please hook R/S to an address line
2012-09-28T12:47:10 < jpa-> not to gpio
2012-09-28T12:58:28 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-09-28T13:02:55 -!- Tunilame [29e5fec2@gateway/web/freenode/ip.41.229.254.194] has joined ##stm32
2012-09-28T13:05:27 -!- Tunilame [29e5fec2@gateway/web/freenode/ip.41.229.254.194] has quit [Client Quit]
2012-09-28T13:08:17 < Mazingaro> hi
2012-09-28T13:08:56 < Mazingaro> is the PendSV a normal exception and so returning with lr = 0xfffffffx or not?
2012-09-28T13:10:46 < zyp> yes
2012-09-28T13:17:04 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-28T13:28:34 -!- BjoernC [~BjoernC@dyndsl-178-142-056-208.ewe-ip-backbone.de] has joined ##stm32
2012-09-28T13:57:41 -!- daku is now known as DaKu
2012-09-28T13:58:31 -!- BjoernC [~BjoernC@dyndsl-178-142-056-208.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-28T13:59:31 < karlp> dongs: re multidrop, don't the f1 usarts have support built in for multimaster, and one of the uarts at least is faster than the others,
2012-09-28T13:59:40 < karlp> should be just fine for just sending commands along?
2012-09-28T14:00:09 < dongs> yes, should be, what about signal integrity though.. buffer it up on output?
2012-09-28T14:01:02 < karlp> how long ar eyou planning on having between the matrixes?
2012-09-28T14:01:10 < dongs> just board to board
2012-09-28T14:01:20 < karlp> if you want to go erally fast, you can stick some rs485 transceivers on there.
2012-09-28T14:01:22 < dongs> so socket to next board
2012-09-28T14:01:33 < karlp> if you're doing single master multi slave you're automatically hafl duplex anyway
2012-09-28T14:01:43 < karlp> but that's another 40c per board
2012-09-28T14:02:02 < karlp> how much data do you need to be sending to each board anyway?
2012-09-28T14:02:16 < dongs> havent even figured that out yet
2012-09-28T14:03:35 < dongs> this is as cjbaird puts it a manchild project
2012-09-28T14:04:42 < dongs> if im gonna do rs485 i might as well control the s hit by DMX
2012-09-28T14:05:04 < dongs> only problem is ... with dmx im only gonna be able to string 8 of these thigns togther max
2012-09-28T14:05:35 < dongs> shit never mind, e ven less
2012-09-28T14:05:46 < dongs> only like... 2.5 lol.
2012-09-28T14:05:54 < dongs> (if using dmx protocol)
2012-09-28T14:05:55 < karlp> don't know enough abotu dmx sorry
2012-09-28T14:06:10 < dongs> its basically 512 8bit values sent with a header.
2012-09-28T14:06:10 < karlp> I know about as much as " dmx... lighting shit right?"
2012-09-28T14:06:28 < dongs> each device receives and counts incoming shit from header
2012-09-28T14:06:31 < dongs> until it reached the nth byte
2012-09-28T14:06:34 < dongs> which is thier address.
2012-09-28T14:06:42 < dongs> and uses those numbers to light shit up or whatever.
2012-09-28T14:06:59 < dongs> so 8x8 #1 would be address 0... it would use up 192 bytes...
2012-09-28T14:07:04 < dongs> next one would be addres 192.. etc
2012-09-28T14:07:09 < dongs> but theres only 512 of these things in a frame
2012-09-28T14:09:37 < karlp> so yeah, stage lighting shit,
2012-09-28T14:09:45 < karlp> not individual pixels on a big matrix
2012-09-28T14:09:54 < dongs> right
2012-09-28T14:10:07 < dongs> peopel do use it for ledbar shit but it gets horribly wasteful with matrix.
2012-09-28T14:10:20 < karlp> you could use dmx to tell the master to say, "give me matrix pattern A!"
2012-09-28T14:10:40 < karlp> but only for   what, 255 commands?
2012-09-28T14:10:56 < karlp> attack pattern delta!
2012-09-28T14:14:20 < dongs> well, i could use  custom shit but hten it wont be  dmx. could address 2 matrixes with one dmx512 packet, 1st byte choosing address and then following bytes data for nth and nth+1 matrix, but hte problem is due to the way dmxshit works, max 512packet refresh rate is 44Hz
2012-09-28T14:17:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-28T14:17:41 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-28T14:17:41 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-28T14:17:41 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-28T14:18:14 < karlp> dongsmx.
2012-09-28T14:18:18 < karlp> moah standards!
2012-09-28T14:20:37 < dongs> i kno rite
2012-09-28T14:24:06 -!- izua [~izua@86.121.85.17] has joined ##stm32
2012-09-28T14:24:06 -!- izua [~izua@86.121.85.17] has quit [Changing host]
2012-09-28T14:24:06 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-28T14:24:08 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-28T14:36:34 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-09-28T14:41:15 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-28T14:44:53 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-28T14:52:08 < Mazingaro> zyp: well it seems that 0xfffffffd in lr works in SysTick_Handler and not in PendSV_Handler
2012-09-28T14:54:35 < zyp> then you're doing something wrong.
2012-09-28T14:56:03 -!- BjoernC [~BjoernC@dyndsl-178-142-056-208.ewe-ip-backbone.de] has joined ##stm32
2012-09-28T14:56:17 < Mazingaro> I moved the context save and load code from SysTick_Handler to PendSV_Handler and place the PendSV trigger in SysTick_Handler
2012-09-28T14:59:41 < Mazingaro> zyp: HardFault occur returning from interrupt
2012-09-28T15:00:50 < Thorn> from which interrupt?
2012-09-28T15:01:13 < Thorn> if you attempt to return to thread mode with pending exceptions, a usage fault is triggered
2012-09-28T15:01:27 < Mazingaro> zyp: yes that is the case
2012-09-28T15:01:45 < Mazingaro> zyp: but how select it in PendSV?
2012-09-28T15:02:04 < Mazingaro> Thorn: I noticed that lr = 0xfffffff1 in PendSV_Handler
2012-09-28T15:03:00 < zyp> oh
2012-09-28T15:03:13 < zyp> you have your priorities messed up?
2012-09-28T15:03:36 < zyp> pendsv should be lowest priority
2012-09-28T15:04:08 < Mazingaro> zyp: ah, right! I forgot to assign correct priority in NVIC
2012-09-28T15:04:34 < Mazingaro> zyp: there's no docs in STM32 STD PERIPH library...
2012-09-28T15:05:02 < zyp> that's not my problem :p
2012-09-28T15:05:06 < Mazingaro> I read this -> http://stackoverflow.com/questions/2752151/which-cortex-m3-interrupts-can-i-use-for-general-purpose-work
2012-09-28T15:05:31 < dongs>  < Mazingaro> zyp: there's no docs in STM32 STD PERIPH library...
2012-09-28T15:05:32 < dongs> ???
2012-09-28T15:05:35 < dongs> what docs do you want
2012-09-28T15:05:53 < Thorn> ARM ARM
2012-09-28T15:06:04 < Mazingaro> dongs: how to change priority for PendSV
2012-09-28T15:08:08 < dongs> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0395b/CIHJHFJD.html
2012-09-28T15:08:12 < dongs> typed your question in google
2012-09-28T15:09:01 < BrainDamage> karlp: you're in iceland, right? you guys got your share of weird: http://imgur.com/gallery/yBvhp
2012-09-28T15:09:23 < karlp> oh, I thought that was tonight.
2012-09-28T15:09:25 < karlp> must have been last night
2012-09-28T15:09:30 < karlp> yeah, our mayor's pretty rocking
2012-09-28T15:09:57 < karlp> when you elect a punk rocker comedian, that sort of thing happens.
2012-09-28T15:10:06 < karlp> he's been a pretty good mayor too
2012-09-28T15:10:22 < Mazingaro> dongs: I'd like to use std peripherals library if possible
2012-09-28T15:10:35 < Mazingaro> dongle: but if this is the only way...
2012-09-28T15:12:31 < Mazingaro> dongs: tx, I'll try to implement that
2012-09-28T15:15:12 < Mazingaro> dongs: this is for M1, I've got M3...
2012-09-28T15:18:05 < zyp> irrelevant.
2012-09-28T15:18:09 < Mazingaro> ok
2012-09-28T15:19:12 < dongs> Mazingaro: just directly use register/macros shits from cmsis
2012-09-28T15:19:13 < dongs> its all in there
2012-09-28T15:19:36 < Mazingaro> dongs: ok tx
2012-09-28T15:23:45 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-09-28T15:25:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-28T15:26:48 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-28T15:37:46 < Laurenceb> http://www.theregister.co.uk/2012/09/28/icelandic_museum/
2012-09-28T15:38:24 < zyp> the penis museum?
2012-09-28T15:38:51 < zyp> oh
2012-09-28T15:39:11 < zyp> so they want a vulva museum to complement it
2012-09-28T15:45:09 < karlp> no, that was just a prank.
2012-09-28T15:50:39 < karlp> bleh, I hve to register to help them with their 25 year thing.
2012-09-28T15:50:48 < karlp> not sure I feel like going that far.
2012-09-28T15:51:24 < Mazingaro> is ICSR register == CONTROL register?
2012-09-28T15:56:39 -!- BjoernC [~BjoernC@dyndsl-178-142-056-208.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-28T16:17:37 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-28T16:34:45 < Laurenceb> http://www2.b3ta.com/host/creative/6776/1348490239/newswears.jpg
2012-09-28T16:39:14 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-28T16:41:12 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-28T16:41:14 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-28T16:46:44 -!- corecode [~2@0x2c.org] has joined ##stm32
2012-09-28T16:47:43 < corecode> in case somebody wants stm32 breakout boards: https://twitter.com/MCHCK/status/251667770041856000
2012-09-28T16:51:45 < karlp> how's the MC HCK project coming along?
2012-09-28T16:52:14 < corecode> doing kinetis k20 now
2012-09-28T16:52:34 < karlp> so what, nuvotron, L1, what was r3? now k20?
2012-09-28T16:52:35 < corecode> last thing i did was hacking openocd to support swd
2012-09-28T16:52:44 < corecode> r3 was nuc120 again :)
2012-09-28T16:53:17 < corecode> now i can do swd with my busblaster
2012-09-28T16:53:33 < corecode> but the upper swd layers still need work
2012-09-28T16:53:39 < karlp> so who do you know that got you k20 parts on any reasonly timeframe and price?
2012-09-28T16:53:53 < corecode> freescale samples k20
2012-09-28T16:54:05 < corecode> takes a week, then you have them
2012-09-28T16:54:32 < corecode> fabs in china/taiwan said they could get it with 1 month lead time
2012-09-28T16:54:47 < corecode> but that was always for 10k units
2012-09-28T16:54:58 < karlp> what are the traces on the back of the usb stick end?
2012-09-28T16:55:09 < corecode> of the r4?
2012-09-28T16:55:12 < zyp> mini-b-socket
2012-09-28T16:55:21 < corecode> yea the mini-b was a fail
2012-09-28T16:55:33 < corecode> didn't realize that you can't plug in the cable if it is recessed
2012-09-28T16:55:43 < corecode> now it is micro ab/b
2012-09-28T16:57:13 < karlp> if you get them for the price you're targetting I'd buy one or three stuffed, but no way I'm going to assemble my own.
2012-09-28T16:57:54 < karlp> you're still going to need a swd dongle to work with these right?
2012-09-28T16:57:57 < zyp> with that part count, it can't take many minutes hand-soldering it
2012-09-28T16:58:00 < karlp> or doe sthe k20 have a usb bootloader in rom?
2012-09-28T16:58:16 < karlp> zyp: sure, but I could be doing my own layout with the few extra parts I want.
2012-09-28T16:58:28 < karlp> it's the assembled price I'd prefer.
2012-09-28T16:58:46 < corecode> i started writing a usb lib and dfu bootloader
2012-09-28T16:58:52 < corecode> it would come with that
2012-09-28T16:59:12 < corecode> and the plan is that you can use one mchck as swd adapter for another
2012-09-28T16:59:14 < corecode> to debug it
2012-09-28T16:59:15 < karlp> so you have how big a pile of rev2 boards you're trying to lose now?
2012-09-28T16:59:19 < corecode> yea
2012-09-28T16:59:25 < corecode> what you can see
2012-09-28T16:59:31 < karlp> oh, that's not so bad then :)
2012-09-28T16:59:33 < zyp> what's «mchck»?
2012-09-28T16:59:41 < corecode> zyp: http://mchck.org
2012-09-28T17:01:44 < Thorn> so you're using chips not supported by libopencm3, chibios hal etc?
2012-09-28T17:01:55 < karlp> will be eventually I guess.
2012-09-28T17:02:01 < karlp> kinetis is just so much newer
2012-09-28T17:02:15 < karlp> cjbaird bought a kinetis board, he's going to submit code now right? ;)
2012-09-28T17:02:27 < karlp> teensy3 is kinetis too, that will bring in a few more.
2012-09-28T17:03:05 < zyp> adding support can't be that much work
2012-09-28T17:03:09 < karlp> nuttx has some very basic k40 and k60 support
2012-09-28T17:03:57 -!- izua_ [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-28T17:03:59 -!- izua__ [~izua@86.121.85.161] has joined ##stm32
2012-09-28T17:03:59 < zyp> when you already have a decent abstraction framework in place, adding support for new devices underneath it is not very much effort
2012-09-28T17:04:04 < corecode> i didn't have a thorough look at libopencm3 yet
2012-09-28T17:04:21 < corecode> i heard people being unhappy with it tho
2012-09-28T17:04:30 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds]
2012-09-28T17:04:41 < karlp> well, people were definitely unhappy with it when it was gpl3 only
2012-09-28T17:04:54 < corecode> ah
2012-09-28T17:04:56 < cjbaird> karlp: I bought it, it still hasn't arrived though. :/
2012-09-28T17:05:13 < corecode> cjbaird: let's keep in touch
2012-09-28T17:05:14 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-09-28T17:06:00 < cjbaird> E14 phoned me yesterday wanting to mention something about the KL25-- I called back today, and no-one had a clue what I was called for.
2012-09-28T17:06:15 < corecode> heh
2012-09-28T17:06:18 < corecode> freedom board
2012-09-28T17:06:38 < corecode> too bad no <=48pin versions of the KL2* exist yet
2012-09-28T17:07:05 < Thorn> are there free tools that can join gerbers?
2012-09-28T17:07:10 < cjbaird> <- also a Raspberry Pi owner, so none of this shit surprises me..
2012-09-28T17:07:38 < Mazingaro> zyp: please look: http://pastebin.com/MzyEQ8Wa
2012-09-28T17:07:46 < corecode> Thorn: kicad can import multiple gerbers i think
2012-09-28T17:09:35 < Thorn> will try thanks
2012-09-28T17:09:58 < zyp> Mazingaro, no time now, need to catch a bus
2012-09-28T17:10:06 < Mazingaro> lol
2012-09-28T17:10:07 < Mazingaro> ok
2012-09-28T17:10:39 < karlp> corecode: there's almost no < 48 pin parts in cortext at all.
2012-09-28T17:10:49 < karlp> there's a couple of qfn 32,
2012-09-28T17:10:57 < karlp> and that vapourware NXP in dip.
2012-09-28T17:11:37 < corecode> i mean <=
2012-09-28T17:11:41 < corecode> including 48
2012-09-28T17:12:13 < Laurenceb> theres 36 pin stm32
2012-09-28T17:12:16 < cjbaird> Btw: The KL25 Freedom board-- I was told that E14.au are getting their shipment of 2000 on the Oct 4th.
2012-09-28T17:12:19 < Laurenceb> in qfn
2012-09-28T17:12:31 < Laurenceb> feedom
2012-09-28T17:13:03 < Thorn> there're stm32f1 ICs in lqfp48
2012-09-28T17:14:19 < corecode> so is stm32l1
2012-09-28T17:14:32 < corecode> but the kinetis KL2 doesn't exist in that yet
2012-09-28T17:15:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-28T17:15:14 < Laurenceb> http://uk.farnell.com/stmicroelectronics/stm32f103t6u6/mcu-32bit-arm-cortex-m3-36vfqfpn/dp/1606328
2012-09-28T17:16:12 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-28T17:16:12 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-28T17:17:44 -!- izua__ [~izua@86.121.85.161] has quit [Read error: Connection reset by peer]
2012-09-28T17:18:00 -!- izua__ [~izua@86.121.85.161] has joined ##stm32
2012-09-28T17:23:03 -!- izua__ [~izua@86.121.85.161] has quit [Read error: Connection reset by peer]
2012-09-28T17:23:04 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-28T17:23:08 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-28T17:26:50 < dongs> noy this fucking garbage again
2012-09-28T17:28:33 <+Steffanx> Yes yes
2012-09-28T17:29:12 < Laurenceb> https://maps.google.co.uk/maps?q=garrard+street+reading&hl=en&ll=51.453539,-0.974629&spn=0.013091,0.02738&sll=53.800651,-4.064941&sspn=12.672499,28.037109&hnear=Garrard+St,+Reading+RG1,+United+Kingdom&t=m&z=16&layer=c&cbll=51.453585,-0.974457&panoid=RMo8PjXUeM0mUE-OkobKXQ&cbp=12,355.84,,0,-4.51
2012-09-28T17:31:08 <+izua_> Laurenceb: rofl
2012-09-28T17:31:23 <+Steffanx> picture of Laurenceb behind the window?
2012-09-28T17:31:34 <+izua_> close
2012-09-28T17:31:50 <+Steffanx> zlog
2012-09-28T17:31:51 < zlog> Steffanx: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-09-28.html
2012-09-28T17:33:20 <+Steffanx> me doesn't see what's funny :)
2012-09-28T17:33:32 <+Steffanx> So it's probably not 'important' :)
2012-09-28T17:34:13 < Laurenceb> http://en.wikipedia.org/wiki/Sweeney_Todd
2012-09-28T17:35:17 <+Steffanx> Ah, nothing special :P
2012-09-28T17:38:47 < Mazingaro> please help, I followed this example http://www.coocox.org/show_Exam.php?id=a1c28590-78b1-11e1-b133-001d0f27e54d&comid=f6e7a291-7883-11e1-8100-001d0f27e54d&comname=NVIC
2012-09-28T17:39:23 < Mazingaro> to  change priority to PendSV, but when I try to enable PendSV IRQ the cpu jams in HardFault_Handler
2012-09-28T17:39:41 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-28T17:39:41 < Laurenceb> http://pic.epicfail.com/wp-content/uploads/2012/09/epic-costume-win-drunk-guy-makes-out-with-hottest-chick-at-party.jpg
2012-09-28T17:41:16 < Laurenceb> omg its dongs
2012-09-28T17:41:17 < Laurenceb> http://pic.epicfail.com/wp-content/uploads/2012/09/boyfriend-fail-hinata-hover-hands.jpg
2012-09-28T17:46:19 < karlp> the other side of that sign is redtube...
2012-09-28T17:46:37 < Mazingaro> :(
2012-09-28T17:47:08 < karlp> Mazingaro: have you had much luck writing applicatons for stm32? or are you writing the OS first?
2012-09-28T17:52:27 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-28T17:58:56 < dongs> im sure im gonna remember something i forgot to order again as soon as i submit this digikey shit
2012-09-28T18:14:16 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-09-28T18:17:46 < dongs> http://i.imgur.com/zlT5u.jpg got a visitor
2012-09-28T18:21:40 < karlp> nice colours
2012-09-28T18:23:00 <+Steffanx> Whoa, I'm glad we don't have spiders like that here
2012-09-28T18:27:28 < zyp> fucking bus with fucking shitty internet
2012-09-28T18:28:02 <+Steffanx> A bus with internet?
2012-09-28T18:28:03 < dongs> http://i.imgur.com/iSt0J.jpg old pic
2012-09-28T18:28:25 <+Steffanx> nice nice
2012-09-28T18:28:29 < zyp> Steffanx, isn't that normal?
2012-09-28T18:28:51 <+Steffanx> 3G with your own phone maybe, but wifi in a bus?
2012-09-28T18:29:21 < zyp> pretty much all long distance buses have wifi here
2012-09-28T18:29:48 <+Steffanx> long distance as in .. 50km?
2012-09-28T18:30:02 < zyp> uh
2012-09-28T18:30:19 < zyp> haven't checked that
2012-09-28T18:30:23 < zyp> I'm going to Oslo
2012-09-28T18:30:43 <+Steffanx> Some 'long' distance trains have wifi, but that's also crap
2012-09-28T18:31:58 -!- tavish [~tavish@59.177.12.183] has joined ##stm32
2012-09-28T18:31:58 -!- tavish [~tavish@59.177.12.183] has quit [Changing host]
2012-09-28T18:31:58 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-28T18:32:40 < Mazingaro> karlp: I'm writing a little RTOS for STM32F (olimex stm32-h107)
2012-09-28T18:33:01 < Mazingaro> karlp: I do not want to use state machines and loops
2012-09-28T18:33:20 < Mazingaro> karlp: I need something smarter
2012-09-28T18:33:51 < dongs> Mazingaro: why not look at CoOS. its simple, no aids, no hal, no crap.
2012-09-28T18:33:58 < dongs> simple task management that stays out of the way.
2012-09-28T18:34:07 < dongs> reinventing the wheel is a bit silly.
2012-09-28T18:34:10 < Mazingaro> yes, but I cannot use third party software
2012-09-28T18:34:16 < dongs> wtf not?
2012-09-28T18:34:45 < Mazingaro> I can't.
2012-09-28T18:35:12 < dongs> yeah, reason?
2012-09-28T18:35:36 < Mazingaro> company politic
2012-09-28T18:36:38 < dongs> coos is 1) free 2) opensores 3) communist
2012-09-28T18:36:40 < dongs> waht more could you ask for
2012-09-28T18:37:04 <+Steffanx> 100k$ would be nice
2012-09-28T18:37:25 < Mazingaro> ok
2012-09-28T18:37:29 < Mazingaro> tx for the hints
2012-09-28T18:37:41 < dongs> if all fags on irc who owe me shit would pay up i would be 100k richer.
2012-09-28T18:38:37 < Mazingaro> anyway I'd like to understand why it hangs at NVIC_EnableIRQ :)
2012-09-28T18:38:48 <+Steffanx> I your case I don't doubt that…. you have a long IRC history
2012-09-28T18:38:53 < dongs> define hangs?
2012-09-28T18:39:04 < dongs> do you have a proper debugger?
2012-09-28T18:39:11 < Mazingaro> gdb
2012-09-28T18:39:15 < dongs> .. oh
2012-09-28T18:39:18 < dongs> no idea then
2012-09-28T18:39:28 < dongs> maybe zyp knows.
2012-09-28T18:39:31 < dongs> when he's off the bus.
2012-09-28T18:39:34 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 244 seconds]
2012-09-28T18:39:36 < Mazingaro> it executes HardFault_Handler
2012-09-28T18:39:40 < dongs> right
2012-09-28T18:39:56 < dongs> what irq are you enabling?
2012-09-28T18:39:59 < dongs> pensv?
2012-09-28T18:40:00 < Mazingaro> PendSV
2012-09-28T18:40:08 < dongs> and does it have any code in it
2012-09-28T18:40:12 <+Steffanx> You can find out what sort of caused the hard fault, but don't remember the details
2012-09-28T18:40:14 < Mazingaro> I changed the priority
2012-09-28T18:40:16 <+Steffanx> zyp does ):
2012-09-28T18:42:18 < Mazingaro> http://pastebin.com/zSXBqYmi
2012-09-28T18:42:33 < Mazingaro> dongs: http://pastebin.com/zSXBqYmi
2012-09-28T18:43:07 < Mazingaro> this is executed after SysInit in main
2012-09-28T18:43:26 < dongs> um
2012-09-28T18:44:30 < Mazingaro> if I remove NVIC_EnableIRQ
2012-09-28T18:44:58 < dongs> http://bcas.tv/paste/results/u2praK91.html
2012-09-28T18:45:04 < Mazingaro> it works, but when in SysTick_Handler I try to trigger the PendSV exception it jumps to HardFault
2012-09-28T18:46:12 < Mazingaro> ok no needs to enable IRQ?
2012-09-28T18:47:12 < Mazingaro> ok, tx
2012-09-28T18:49:19 -!- claude_work is now known as claude
2012-09-28T18:58:15 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-28T19:01:35 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 256 seconds]
2012-09-28T19:02:27 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-28T19:11:27 < emeb> ZZZzzzz....
2012-09-28T19:14:00 <+Steffanx> Good night
2012-09-28T19:14:25 < emeb> Hmm? Huh.. uhhh - oh. Awake now.
2012-09-28T19:18:50 <+Steffanx> It was nice in dreamland?
2012-09-28T19:19:48 < emeb> I dreamt I was persecuted by the government for having mutant eyes that could see through shoes.
2012-09-28T19:23:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-28T19:23:44 <+Steffanx> Yeah, if you have a foot fetish that's very useful
2012-09-28T19:24:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-09-28T19:24:52 < emeb> Or if you're a shoe salesman. I am neither though so it was something of a non-sequitur.
2012-09-28T19:27:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-28T19:29:09 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-28T19:40:07 < Laurenceb> blergggggg
2012-09-28T19:40:16 < Laurenceb> EM testing is boring
2012-09-28T19:40:53 < emeb> repeat the happy mantra: "I'm getting paid for this"
2012-09-28T19:41:30 < Laurenceb> lol
2012-09-28T19:51:14 -!- claude is now known as Claude
2012-09-28T19:58:51 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: leaving]
2012-09-28T20:02:20 -!- Claude is now known as claude
2012-09-28T20:07:00 <+Steffanx> Laurenceb you are a physisist not?
2012-09-28T20:07:10 < Laurenceb> i am yes
2012-09-28T20:07:37 <+Steffanx> physists are not meant for programming crap. Can't you get someone who does it for you? :)
2012-09-28T20:08:57 <+Steffanx> wouldn't that save you ages?
2012-09-28T20:10:42 < karlp> him maybe,
2012-09-28T20:13:07 < karlp> but not his boss?
2012-09-28T20:13:30 <+Steffanx> His boss should be smart enough too see he can't handle it on his own? At least not in a reasonable time
2012-09-28T20:13:46 < karlp> haha
2012-09-28T20:15:44 < Laurenceb> what?
2012-09-28T20:15:48 <+Steffanx> I dont mean i don't think Laurenceb can't do it, but in general physists aren't meant to write code :)
2012-09-28T20:15:52 < Laurenceb> why cant i handle it?
2012-09-28T20:16:18 <+Steffanx> You said it yourself Laurenceb
2012-09-28T20:16:33 <+Steffanx> Something like "The project takes way too long now"
2012-09-28T20:16:50 < emeb> "You want the code? You can't handle the code!"
2012-09-28T20:19:18 <+Steffanx> It's not a hobby bobby project, is it Laurenceb ?
2012-09-28T20:21:28 < Laurenceb> Steffanx: I mean the chibios crap
2012-09-28T20:21:34 < Laurenceb> Steffanx: no
2012-09-28T20:29:16 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-28T20:29:17 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-28T20:29:17 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-28T20:37:37 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-09-28T20:48:04 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-09-28T21:15:16 -!- izua_ [~izua@unaffiliated/izua] has quit [Ping timeout: 248 seconds]
2012-09-28T21:17:57 < Mazingaro> hi
2012-09-28T21:26:28 < emeb> lo
2012-09-28T21:27:32 <+Steffanx> hi
2012-09-28T21:28:29 < emeb> nu
2012-09-28T21:28:56 < emeb> (what does nu mean? Greek letter? Next in series is xi then)
2012-09-28T21:29:16 <+Steffanx> It's sort of "Hey/hi" in latvian
2012-09-28T21:29:28 < emeb> Cool.
2012-09-28T21:29:36  * emeb has been to Riga. Nice place.
2012-09-28T21:30:02 <+Steffanx> I'm not SURE if it means that. Someone from latvia who joined here a few times said that
2012-09-28T21:30:35 < emeb> good enough.
2012-09-28T21:31:20 < emeb> http://translate.google.com/#lv/en/nu
2012-09-28T21:31:44 <+Steffanx> Good translate is wrong
2012-09-28T21:32:40 <+Steffanx> *google
2012-09-28T21:48:14 -!- DaKu is now known as daku
2012-09-28T22:03:49 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-09-28T22:12:32 < gnomad> imagine that, computer language translation gets something wrong...
2012-09-28T22:15:02 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-09-28T22:16:38 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-28T22:17:35 < Laurenceb_> ok this is weird
2012-09-28T22:17:40 <+Steffanx> yes
2012-09-28T22:17:42 < Laurenceb_> my timers were misbehaving
2012-09-28T22:17:51 < Laurenceb_> so i look at tim2->cnt
2012-09-28T22:18:00 < Laurenceb_> the upper bits are full of junk
2012-09-28T22:19:07 < emeb> free random number generator
2012-09-28T22:19:13 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-09-28T22:21:39 -!- barthess [~barthess@5.100.209.223] has joined ##stm32
2012-09-28T22:28:13 < Mazingaro> zyp: hi
2012-09-28T22:37:49 < Laurenceb_> reset doesnt fix it
2012-09-28T22:38:41 < emeb> constant junk, or changing?
2012-09-28T22:40:49 < Laurenceb_> seems to be constant
2012-09-28T22:41:53 -!- izua [~izua@188.26.167.40] has joined ##stm32
2012-09-28T22:41:53 -!- izua [~izua@188.26.167.40] has quit [Changing host]
2012-09-28T22:41:53 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-28T22:41:56 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-28T22:42:26 < Laurenceb_> its not setup now - looking over logs
2012-09-28T22:42:41 < Laurenceb_> ill try it tomorrow - maybe it needed power cycling
2012-09-28T22:43:21 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Timer.c#L22
2012-09-28T22:43:24 < Laurenceb_> thats the code
2012-09-28T22:43:35 < Laurenceb_> maybe i need to assert APB reset?
2012-09-28T22:43:43 < Laurenceb_> as part of the initialisation
2012-09-28T22:44:15 < Mazingaro> hi
2012-09-28T22:45:04 < Mazingaro> why using NVIC_SetPendingIRQ(PendSV_IRQn) causes an hard fault? I did change PendSV priority at 0xff with NVIC_SetPriority()
2012-09-28T22:49:05 < Laurenceb_> http://i.imgur.com/9bnvR.png
2012-09-28T22:50:38 <+izua> neat.
2012-09-28T22:51:39 <+izua> found a k555пaз in a baggie on the street, thrown away with other oddities
2012-09-28T22:51:57 < Laurenceb_> oh you should
2012-09-28T22:52:02 <+izua> production date, 8506, that means it was made in the soviet union
2012-09-28T22:52:07 < Laurenceb_> wut
2012-09-28T22:52:12 <+izua> wonder if it still works
2012-09-28T22:52:24 < Laurenceb_> - you should reset using the apb command
2012-09-28T22:52:30 < Laurenceb_> probably why it was wrong
2012-09-28T22:52:38 < Laurenceb_> how the hell did you find that?!
2012-09-28T22:52:43 < Laurenceb_> what country do you live in?
2012-09-28T22:52:59 < emeb> izua: wtf is k555пaз
2012-09-28T22:53:02 <+izua> .ro
2012-09-28T22:53:06 < Laurenceb_> ah
2012-09-28T22:53:15 < Laurenceb_> people just chuck out bags of ics?
2012-09-28T22:53:15 <+izua> it's likely a 555 soviet clone
2012-09-28T22:53:29 <+izua> yes, some weekends are "random shit disposal day on the street"
2012-09-28T22:53:35 < Laurenceb_> lol
2012-09-28T22:53:53 < emeb> "freecycle"
2012-09-28T22:54:04 <+izua> i found several rack servers, a scope, several non working ones and some assorted tube ham radios
2012-09-28T22:54:17 <+izua> (overall)
2012-09-28T22:54:23 < Laurenceb_> wow
2012-09-28T22:54:47 <+izua> nothing biggie, it's a 1MHz single trace scope
2012-09-28T22:54:47 < emeb> all I ever see at the curb is old furniture.
2012-09-28T22:55:07 <+izua> but hey, after some percussive maintenance, it almost works
2012-09-28T22:55:10 < BrainDamage> at most a brokenwashing machine or crt tv
2012-09-28T22:55:49 <+izua> this idea is to harmonize with EU's electronic waste disposal stuff
2012-09-28T22:56:11 < Thorn> K555 is a discrete schottky-TTL family
2012-09-28T22:56:30 <+izua> but since we don't have any actual e-waste centers, they just paint over regular trucks and haul the stuff on the streets to the regular dumps
2012-09-28T22:56:45 <+izua> and don't forget the marketing! efficient! green! speciality disposal pit and processing center!
2012-09-28T22:56:52 < Laurenceb_> lol
2012-09-28T22:57:19 < Laurenceb_> i have to have WEEE compliant disposal instructions and a policy on my hardware
2012-09-28T22:57:24 < Laurenceb_> so stupid
2012-09-28T22:57:28 < zyp> izua, still no matrix?
2012-09-28T22:57:35 <+izua> nope :(
2012-09-28T22:57:40 < BrainDamage> how many certfications do you have to pass?
2012-09-28T22:57:41 < zyp> weird
2012-09-28T22:57:44 <+izua> maybe they flew it by plane
2012-09-28T22:57:49 <+izua> and over america and back
2012-09-28T22:58:00 <+izua> it's probably in customs
2012-09-28T22:58:11 <+izua> they just are really fast and efficient at processing packages
2012-09-28T22:58:12 < Laurenceb_> BrainDamage: a stupid number
2012-09-28T22:58:27 < BrainDamage> I guess it being a medical item doesn't help
2012-09-28T22:58:41 < Laurenceb_> has to include road traffic accident standards
2012-09-28T22:58:42 < Laurenceb_> yes
2012-09-28T22:59:00 < zyp> disposal instruction: throw it in a fire
2012-09-28T22:59:09 <+izua> inhale deeply
2012-09-28T22:59:24 < BrainDamage> perhap it'll be more efficient if you ust distribute it from the chinese
2012-09-28T22:59:31 < BrainDamage> like uk -> chinese -> uk
2012-09-28T22:59:51 <+izua> you mean, chinese export certification?
2012-09-28T23:00:06 < BrainDamage> yup
2012-09-28T23:00:22 <+izua> "designed and assembled in uk. shipped from china, because retarded laws"
2012-09-28T23:01:03 < Laurenceb_> heh
2012-09-28T23:01:09 <+izua> Laurenceb_: among one of the most amazing things i've seen thrown away is a 6-7 meter long horizontal lathe
2012-09-28T23:01:15 < Laurenceb_> wow
2012-09-28T23:01:24 < Laurenceb_> im actually not selling stuff
2012-09-28T23:01:32 < Laurenceb_> but it needs to be tested in a hospital
2012-09-28T23:01:42 < Laurenceb_> for that to happen it has to pass all the same stuff
2012-09-28T23:01:53 <+izua> it has two operators posts, and the rotating grip (chuck?) is about as big as a guy in the vetruvian position
2012-09-28T23:02:06 <+izua> needless to say, 5 years later, it's still in the exact spot.
2012-09-28T23:03:18 < emeb> rusting
2012-09-28T23:03:47 <+izua> i could put it on craigslist i guess
2012-09-28T23:03:48 < corecode> anybody interested in trying/improve swd support in openocd?
2012-09-28T23:03:50 < corecode> https://github.com/corecode/openocd-swd
2012-09-28T23:04:35 <+izua> "lathe seeking good and spacious home/workshop. if you don't have any use for it, at least you can take a ride in the chuck! will not provide transport. picture might not be to scale"
2012-09-28T23:04:57 < corecode> cnc?
2012-09-28T23:05:08 <+izua> and then one of those silly perspective deformation pictures, where i seem to hold it in my hand
2012-09-28T23:05:09 < emeb> surprised that metal thieves haven't cut it up and hauled it away.
2012-09-28T23:05:39 <+izua> it looks very heavy
2012-09-28T23:05:44 < emeb> (unguarded metal doesn't last long around here)
2012-09-28T23:05:59 < emeb> storm grates, manhole covers, you name it.
2012-09-28T23:06:01 < BrainDamage> corecode: I think most here use either st-link tool or black magic probe
2012-09-28T23:06:06 <+izua> emeb: awesome
2012-09-28T23:06:06 < corecode> BrainDamage: :/
2012-09-28T23:06:27 < BrainDamage> corecode: I'd seiously consider porting the bmp to your device tough
2012-09-28T23:06:40 < corecode> bmp?
2012-09-28T23:06:43 < zyp> corecode, have you talked with ntfreak? he's an openocd developer
2012-09-28T23:06:46 < BrainDamage> bmp = black magic probe
2012-09-28T23:06:47 <+izua> here, the gyppos cut wires (sometimes even live ones and major comm lines), burn the insulation off and sell the copper
2012-09-28T23:06:52 < corecode> zyp: yes, talking with him
2012-09-28T23:07:07 < BrainDamage> since you wanted a debugging firmware, that'd fit your requirement perfectly
2012-09-28T23:07:20 < BrainDamage> and unlike other debuggers, that one has it's own gdb server in the firmware
2012-09-28T23:07:29 < emeb> izua: same here. not the roma though - meth heads.
2012-09-28T23:07:39 < corecode> BrainDamage: awesome!
2012-09-28T23:08:02 < BrainDamage> gsmcmullin made it
2012-09-28T23:08:10 < emeb> +1 for BMP.
2012-09-28T23:08:37 < corecode> very cood
2012-09-28T23:08:39 < corecode> cool*
2012-09-28T23:08:47 < BrainDamage> corecode: https://github.com/gsmcmullin/blackmagic
2012-09-28T23:08:52 < corecode> sounds like versaloon, but integrated gdb
2012-09-28T23:09:09 < corecode> i wouldn't want to steal his business tho
2012-09-28T23:09:24 < emeb> it's all FOSS - HW and SW.
2012-09-28T23:09:28 < corecode> yea
2012-09-28T23:09:32 < corecode> exactly for that reason
2012-09-28T23:09:47 < corecode> if it was closed just a bit, i wouldn't have any concerns
2012-09-28T23:10:22 < emeb> you're trying to do mass production / commercial sales?
2012-09-28T23:10:32 < zyp> «I don't mind using other people's code, except when I'm perfectly allowed to»
2012-09-28T23:10:45 < emeb> :D
2012-09-28T23:11:00 < corecode> well, i wouldn't feel bad re-implementing the idea
2012-09-28T23:11:09 <+izua> zyp: it's the thrill of doing illegal stuff.
2012-09-28T23:11:19 < BrainDamage> talk to him, and see if he's ok with it
2012-09-28T23:11:25 < emeb> ^ this
2012-09-28T23:11:38 < corecode> well clearly the mchck does not offer the same set of features
2012-09-28T23:12:03 <+izua> zyp: speaking of the matrix - which i pray to $_GOD it wasn't lost - do you still have the schematic/example code around?
2012-09-28T23:12:19 < zyp> of course
2012-09-28T23:12:29 < zyp> example code is in a branch in laks_demo repo
2012-09-28T23:13:03 < zyp> http://bin.jvnv.net/f/AXTMf.png <- schematic is here
2012-09-28T23:13:30 < corecode> emeb: my hope is that the board will be commercially available at $7, $5 volume
2012-09-28T23:13:55 < corecode> emeb: and so far it seems like i will have to do it, because nobody does instead of me
2012-09-28T23:14:13 < emeb> corecode: nice price. BMP is for sale in the US for ~$70
2012-09-28T23:14:32 < emeb> (made in NZ IIRC)
2012-09-28T23:14:33 < corecode> yea, the BMP is a tool that works
2012-09-28T23:14:45 < corecode> the mc hck is ment as a development platform
2012-09-28T23:14:49 < corecode> like arduino, just better
2012-09-28T23:14:59 <+izua> lolduino
2012-09-28T23:15:02 <+izua> zyp: thanks
2012-09-28T23:15:10 < corecode> izua: exactly
2012-09-28T23:15:51 < emeb> zyp: what are the FETs?
2012-09-28T23:15:57 < corecode> arduino is easy to use, that's the big plus
2012-09-28T23:16:00 < emeb> (Part Num)
2012-09-28T23:18:20 <+izua> zyp: hmm, ilike how the the latch on the row/led driver advances to the next column. saves a pin, but forces your refresh direction
2012-09-28T23:19:03 <+izua> i hope they won't wreck it in the customs ._.
2012-09-28T23:20:21 < emeb> "hmmm - complicated electronics with no discernible purpose. Must be terrist stuff - trash it."
2012-09-28T23:20:22 < corecode> emeb: http://mchck.org
2012-09-28T23:20:36 < corecode> in case you need context
2012-09-28T23:20:45 < emeb> cute
2012-09-28T23:21:04 <+izua> they could tore the lcd matrix apart
2012-09-28T23:21:11 <+izua> thinking you've hid weed in it or something
2012-09-28T23:22:10 < zyp> emeb, don't remember, some generic one with enough current driving capability
2012-09-28T23:22:31 < emeb> k
2012-09-28T23:23:40 < zyp> izua, how would you else do it?
2012-09-28T23:24:13 < emeb> izua: if you reload/reclock it fast enough then it can appear to scan in reverse.
2012-09-28T23:24:23 < emeb> POV lets you get away with a lot.
2012-09-28T23:28:06 <+izua> zyp: i would separate the lines for v and h
2012-09-28T23:28:08 < zyp> you can in theory have a line clock >1MHz, so it's not like scan direction matters :p
2012-09-28T23:28:15 < zyp> izua, why?
2012-09-28T23:28:22 <+izua> so i could choose h or v scanning. but i guess it doesn't really mater
2012-09-28T23:28:43 < zyp> choose how?
2012-09-28T23:29:08 < zyp> it's not possible to drive more than one row at a time in a sane way
2012-09-28T23:29:21 <+izua> the choice of lighting up a row or a column
2012-09-28T23:29:29 < zyp> yes
2012-09-28T23:29:30 <+izua> yeah, it's irrelevant :P
2012-09-28T23:29:35 < zyp> it's not possible
2012-09-28T23:30:31 < zyp> keep in mind that current regulation happens per col
2012-09-28T23:31:07 < zyp> so if you turn on multiple rows, the current gets distributed across multiple rows, making the pixels less bright
2012-09-28T23:31:58 < zyp> also, the col/row terms are reversed in the schematic, so don't be confused by that :p
2012-09-28T23:33:03 < zyp> and the row driver shift registers doesn't have latches
2012-09-28T23:33:26 < emeb> rows, columns - just a ' apart. (in Matlab at least)
2012-09-28T23:33:36 < zyp> so you can't clock in a new set of values while displaying the previous
2012-09-28T23:33:52 < emeb> better move fast
2012-09-28T23:33:56 < zyp> :)
2012-09-28T23:34:09 < zyp> or just use it as intended :p
2012-09-28T23:34:16 <+izua> yeah, lol :P
2012-09-28T23:34:48 <+izua> you could turn the max off, latch it, advance to next scanned row/col, dump data in the max
2012-09-28T23:35:02 <+izua> and that would overcome the lack of a latch though
2012-09-28T23:35:10 < zyp> sure
2012-09-28T23:35:31 < zyp> but that doesn't solve the current limiting :p
2012-09-28T23:35:38 <+izua> true
2012-09-28T23:37:05 <+izua> so hm. the idea is to dump 16 bits in the max, and clock a shifting zero to advance the columns
2012-09-28T23:37:18 <+izua> heh, 10 minute thing
2012-09-28T23:38:27 < zyp> izua, the idea is that you shift in a 1 when you latch data for the first row, then pass that down as you latch in new rows until it shifts out at the end
2012-09-28T23:38:32 < zyp> and then you shift in a new 1
2012-09-28T23:38:42 <+izua> ah
2012-09-28T23:38:53 < zyp> ah, except that the FETs are active low, so it's a 0
2012-09-28T23:38:56 <+izua> right, reverse logic
2012-09-28T23:39:04 <+izua> so, 0 turns on the column?
2012-09-28T23:39:18 < zyp> well, I call it row .p
2012-09-28T23:39:28 <+izua> wait
2012-09-28T23:39:30 <+izua> lol
2012-09-28T23:39:47 <+izua> so the max6979 drivers the columns, but the lines on its bus are called "row x"? :P
2012-09-28T23:40:06 < zyp> 22:31:58 < zyp> also, the col/row terms are reversed in the schematic, so don't be confused by that :p
2012-09-28T23:40:27 < zyp> anyway, I suggest looking at http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=led_matrix
2012-09-28T23:41:40 < zyp> vsync is the input to the shift register, hsync is the latch signal,
2012-09-28T23:41:40 <+izua> wow
2012-09-28T23:41:49 <+izua> you're pretty good at protecting your work :P
2012-09-28T23:42:09 <+izua> i never thought that would be possible when code & schematics are provided
2012-09-28T23:42:59 < zyp> well, the thing is that I first drawed the schematic, and then I rotated the matrix 90 degrees when I routed the board :p
2012-09-28T23:45:11 < Thorn> apparently outer copper layers are thicker because of electroplating
2012-09-28T23:45:23 < Mazingaro> zyp: :)
2012-09-28T23:45:57 < Thorn> one guide says it gives +0.2mm of copper to outer layers
2012-09-28T23:48:52 < Mazingaro> zyp: I was able to change the pendsv priority, but now I've got Fault on triggering PendSV
2012-09-28T23:49:06 < zyp> what kind of fault?
2012-09-28T23:50:03 < Mazingaro> Hard Fault
2012-09-28T23:50:19 < zyp> read SCB_CFSR to check what is causing the hardfault
2012-09-28T23:50:34 < Mazingaro> with gdb?
2012-09-28T23:50:45 < zyp> with whatever
2012-09-28T23:50:50 < zyp> I'd use gdb
2012-09-28T23:52:10 < Mazingaro> 0xe000ed28
2012-09-28T23:52:48 < Mazingaro> 0x00
2012-09-28T23:53:11 < zyp> it's a 32-bit register, read all 32 bits
2012-09-28T23:53:53 < Mazingaro> ok
2012-09-28T23:54:31 < Mazingaro> 0x00000000
2012-09-28T23:55:10 < Mazingaro> gotcha!
2012-09-28T23:55:17 < zyp> well, you obviously have to read it when you trap the hardfault to see what caused the hardfault
2012-09-28T23:55:20 < Mazingaro> 0x00820000
2012-09-28T23:55:37 < zyp> INVPC, PRECISERR
2012-09-28T23:55:50 < Mazingaro> bit 17 and 23
2012-09-28T23:57:53 < zyp> oh, right
2012-09-28T23:57:59 < zyp> disregard what I said
2012-09-28T23:58:07 < Mazingaro> INVSTATE and RESERVED
2012-09-28T23:58:24 < Mazingaro> the pc is wrong?!?!?
2012-09-28T23:59:15 < Mazingaro> what does it mean?
2012-09-28T23:59:47 < Mazingaro> is that big endian error?
--- Day changed Sat Sep 29 2012
2012-09-29T00:00:03 < zyp> no, probably address error
2012-09-29T00:00:06 < Mazingaro> so it would be 0x00008200
2012-09-29T00:00:21 < zyp> huh?
2012-09-29T00:00:33 < zyp> how did you read that register?
2012-09-29T00:00:54 < Mazingaro> I read 0xe000ed28 : 0x00 0x82 0x00 0x00
2012-09-29T00:01:11 < Mazingaro> with GDB memory
2012-09-29T00:01:21 < zyp> x/w 0xe000ed28
2012-09-29T00:01:41 < zyp> but yeah, that's 0x00008200
2012-09-29T00:01:49 < zyp> BFARVALID, PRECISERR
2012-09-29T00:01:49 < Mazingaro> ok
2012-09-29T00:02:01 < zyp> so you are getting a busfault
2012-09-29T00:02:14 -!- theAdib [~adib@dslb-088-074-140-062.pools.arcor-ip.net] has joined ##stm32
2012-09-29T00:03:52 < Mazingaro> PC = 0x80002e7
2012-09-29T00:04:04 < zyp> what's that?
2012-09-29T00:04:22 < Mazingaro> in that location there is the loop of HardFault_Handler
2012-09-29T00:04:35 < zyp> ah, right
2012-09-29T00:04:39 < Mazingaro> PC must contain the address of the instruction that bus fault
2012-09-29T00:04:45 < zyp> you need to check the old value of PC
2012-09-29T00:04:50 < zyp> i.e. stacked value of PC
2012-09-29T00:05:54 < Mazingaro> I put in SysTick_Handler, NVIC_SetPendingIRQ(PendSV_IRQn) to trig the PendSV for context svitch
2012-09-29T00:06:22 < Mazingaro> SysTick_Handler is a normal void function
2012-09-29T00:09:36 < zyp> http://seclists.org/nmap-dev/2012/q3/1050
2012-09-29T00:10:07 < Mazingaro> I read that exceptions like HardFault, BusFault and so on are already enabled
2012-09-29T00:18:01 < Mazingaro> zyp: working!
2012-09-29T00:18:26 < Mazingaro> zyp: I'm using *((uint32_t volatile *) 0xe000ed04) = 0x10000000; to trig the PendSV
2012-09-29T00:20:19 < Mazingaro> :D
2012-09-29T00:20:22 < zyp> :)
2012-09-29T00:21:21 < Mazingaro> it seems to work, as SAVE_CONTEXT - LOAD_CONTEXT test I didn't switch context among the two and test is it jumps to the correct place between exceptions
2012-09-29T00:22:27 < Mazingaro> registers on stack and PC seem correct since task execution take place in the same point when IRQ occurred
2012-09-29T00:23:06 < Mazingaro> zyp: I understand now, what zippe said...
2012-09-29T00:24:52 < Mazingaro> zyp: as I understood PendSV is a way to let the context switch take place without care about other interrupts
2012-09-29T00:25:26 < Mazingaro> as PendSV will be always executed "after" them
2012-09-29T00:25:34 < Mazingaro> this is a neat cpu feature
2012-09-29T00:25:57 < Thorn> as long as it has lowest priority.
2012-09-29T00:26:11 < Mazingaro> yes
2012-09-29T00:27:18 < Mazingaro> I put SysTick to 0xee and PendSV to 0xff they are the unique IRQs for now, so PendSV will be executed on SysTick tail
2012-09-29T00:33:47 -!- barthess [~barthess@5.100.209.223] has quit [Quit: Leaving.]
2012-09-29T00:37:27 < Mazingaro> did other manufacturers build controllers based on real ARM (like TI sitara)
2012-09-29T00:39:29 < Mazingaro> I worked on str912, but ST support is not so good...
2012-09-29T00:40:25 < Thorn> wtf is real ARM
2012-09-29T00:41:27 < Mazingaro> not Thumb processor
2012-09-29T00:41:31 < zyp> a core which supports arm mode, I assume
2012-09-29T00:41:37 < Mazingaro> yes
2012-09-29T00:41:39 < zyp> i.e. cortex-a
2012-09-29T00:41:46 < Mazingaro> yes
2012-09-29T00:42:40 < Mazingaro> thumb processors are cheaper than arm mode processors
2012-09-29T00:42:44 < Thorn> any advantages over a core without ARM mode?
2012-09-29T00:43:02 < zyp> Mazingaro, of course, that's the point
2012-09-29T00:43:16 < zyp> simpler, smaller and cheaper
2012-09-29T00:45:08 < Mazingaro> Thorn: the same between cpu for laptop and cpu for desktop
2012-09-29T00:45:09 < Thorn> so cortex-m is now a "thumb processor". trolling +1
2012-09-29T00:45:27 < Mazingaro> thumb2
2012-09-29T00:46:25 < zyp> well, the difference between -a and -m is much larger than arm/thumb vs thumb only
2012-09-29T00:46:45 <+izua> it also has a jazelle mode
2012-09-29T00:46:48 <+izua> with a secret opcode
2012-09-29T00:47:01 < Mazingaro> arm mode has got 6 privileged modes
2012-09-29T00:47:19 < Mazingaro> banked registers
2012-09-29T00:47:55 < Mazingaro> MMU
2012-09-29T00:48:15 < zyp> Mazingaro, so, what were your question again?
2012-09-29T00:48:49 < Mazingaro> who sells ARM mode processor?
2012-09-29T00:49:01 < zyp> everybody.
2012-09-29T00:49:08 < Mazingaro> ST = cortex
2012-09-29T00:49:16 < Mazingaro> TI = cortex
2012-09-29T00:49:39 < Mazingaro> NXP = cortex
2012-09-29T00:49:44 <+izua> arm just designs the cpu
2012-09-29T00:49:48 < Thorn> btw, TI seems to have stopped developing the OMAP line
2012-09-29T00:49:50 <+izua> and other people implement it in actual chips
2012-09-29T00:50:02 < zyp> Thorn, I heard some talk about that the other day
2012-09-29T00:50:13 < Thorn> wth.
2012-09-29T00:50:28 < zyp> was some talk during lunch at work, but I haven't seen any source on it yet
2012-09-29T00:50:30 < Mazingaro> Thorn: omap are not so cheap, not like the cortex
2012-09-29T00:50:46 < zyp> Mazingaro, you're confusing terms
2012-09-29T00:51:02 < Thorn> core i7 isn't as cheap as AVR either
2012-09-29T00:51:05 < zyp> omap contains both cortex-a and cortex-m cores
2012-09-29T00:51:15 <+izua> Thorn: but there's not a single pin you can bitbang stuff on an i7
2012-09-29T00:51:17 < Mazingaro> ok
2012-09-29T00:51:31 < Mazingaro> I mean cortex-m
2012-09-29T00:51:33 <+izua> well. probably there is.
2012-09-29T00:51:37 < zyp> and cars aren't as cheap as bicycles either.
2012-09-29T00:51:44 <+izua> but i dobut anyone can solder it
2012-09-29T00:51:44 < Mazingaro> sure
2012-09-29T00:52:21 < Thorn> also Stratix V is ridiculously expensive compared to a 74HC00
2012-09-29T00:52:28 < zyp> :)
2012-09-29T00:52:43 <+izua> but the startix v can't run on a wide range of voltages!
2012-09-29T00:53:08 < zyp> doesn't it require a wide range of voltages all at once? :p
2012-09-29T00:53:20 <+izua> touche
2012-09-29T00:53:22 < Thorn> or be used in linear mode as an amplifier
2012-09-29T00:54:04  * Bird|lappy has a 74ACT74 doing reset-release sync for a CPLD in one of his designs
2012-09-29T00:54:44 < Mazingaro> Thorn: did you refer to Sitara cpu?
2012-09-29T00:55:44 < Thorn> when?
2012-09-29T00:56:50 < Mazingaro> you said that TI stopped developing OMAP
2012-09-29T00:57:11 < zyp> omap != sitara
2012-09-29T00:57:42 < Mazingaro> ok
2012-09-29T00:58:15 < Laurenceb_> whats the "low power" APB enable flag for?
2012-09-29T00:58:19 -!- WorkerBuzz [~Buzz@50.20.134.98] has quit [Ping timeout: 252 seconds]
2012-09-29T00:58:32 < Mazingaro> are they both based on cortex-a8?
2012-09-29T00:59:51 < zyp> OMAP3 is, OMAP4 is A9 and OMAP5 is A15
2012-09-29T01:06:26 < Mazingaro> ah ok
2012-09-29T01:06:45 < Mazingaro> did anyone of you worked on str912?
2012-09-29T01:07:59 < Mazingaro> ?
2012-09-29T01:08:47 < Laurenceb_> you i heard about that on the news
2012-09-29T01:08:54 < Laurenceb_> what did Ti actually announce?
2012-09-29T01:09:59 < Mazingaro> a clone of the str912
2012-09-29T01:10:03 < zyp> Mazingaro, sounds pretty outdated by now, it's ARM9 after all
2012-09-29T01:10:17 < Thorn> nothing here http://newscenter.ti.com/index.php?s=32851
2012-09-29T01:10:17 < Mazingaro> arm926
2012-09-29T01:10:22 < Mazingaro> it wasn't so bad
2012-09-29T01:10:52 < Mazingaro> was a cpu intended for electrical motor control
2012-09-29T01:12:04 < emeb> He's probably talking about this: http://appleinsider.com/articles/12/09/27/apple-actively-courting-ti-employees-as-a6-rival-omap-chip-put-on-ice
2012-09-29T01:13:18 < emeb> Note - it doesn't say they're getting out of the OMAP business, but that they are shifting from consumer embedded to industrial, automotive.
2012-09-29T01:14:37 < Thorn> this means cheaper and less powerful CPUs?
2012-09-29T01:15:20 < emeb> who knows. Probably just means TI is tired of competing with Sammy, nVidia, etc.
2012-09-29T01:16:12 < emeb> might not even mean a big shift in engineering - more marketing. who they try to sell to.
2012-09-29T01:16:41 < zyp> it's more that they won't see samsung and apple on the customer list now that they are designing their own SoCs
2012-09-29T01:17:02 < emeb> yep
2012-09-29T01:17:17 < zyp> it's obvious that phone vendors aren't as attractive a market when the largest ones make their own SoCs
2012-09-29T01:17:51 < emeb> and tablet vendors are rapidly approaching that point too.
2012-09-29T01:18:37 < Thorn> so now thanks to ARM, every big guy can have their own CPU
2012-09-29T01:18:49 < zyp> well, phone vendors = tablet vendors, pretty much
2012-09-29T01:18:52 < emeb> that's what licensed IP is all about.
2012-09-29T01:19:36 < emeb> except for dime-a-dozen asian "craplets" based on awinner or whatever's cheap this week.
2012-09-29T01:20:22 < zyp> I don't have those on my customer list, so I don't care about those :p
2012-09-29T01:21:05 < emeb> that market is too volatile to count on anyway.
2012-09-29T01:22:18 < Mazingaro> Thorn: I recall when the arm was used on acorn archimedes first
2012-09-29T01:22:18 < Thorn> http://www.eetimes.com/electronics-blogs/semi-conscious/4397281/Why-OMAP-cant-compete-in-smartphones
2012-09-29T01:22:37 < Mazingaro> english pc with a english cpu
2012-09-29T01:22:41 < Mazingaro> lol
2012-09-29T01:24:16 <+Steffanx> eetimes, people read that?
2012-09-29T01:24:51 < emeb> yep - I still get a paper version in the mail!
2012-09-29T01:24:56 < emeb> pretty thin though.
2012-09-29T01:24:57 < Laurenceb_> oi
2012-09-29T01:24:59 < Laurenceb_> im english
2012-09-29T01:25:08 < Laurenceb_> stop humiliating my country
2012-09-29T01:25:29 < Laurenceb_> also acorn archimedes was epic
2012-09-29T01:26:08 < Laurenceb_> oh dear
2012-09-29T01:26:15 < Laurenceb_>  !Draw was good
2012-09-29T01:27:38 < Thorn> "It is a sad day for ARM world in general though. TI provides some of the best documentation in business."
2012-09-29T01:28:17 < emeb> *choke*
2012-09-29T01:28:53 < emeb> if you rate best by how many MB the pdf files are I guess.
2012-09-29T01:29:51 < zyp> OMAP4 TRM is pretty good
2012-09-29T01:30:12 < Laurenceb_> yeah
2012-09-29T01:30:31  * emeb has spent too long looking at OMAP3530 trm.
2012-09-29T01:30:43 < Laurenceb_> so they are dropping omap?
2012-09-29T01:30:51 < zyp> no
2012-09-29T01:31:15 < zyp> just reducing the focus on the smartphone market
2012-09-29T01:31:57 < zyp> emeb, I spent some time looking in the OMAP4 TRM a few days ago
2012-09-29T01:32:09 < zyp> I was poking hardware registers directly while debugging some issue
2012-09-29T01:32:45 < emeb> Oh - do they actually give the addresses now?
2012-09-29T01:33:13 < emeb> Used to be you had to know the sekrit place in the TRM where all the indirections were resolved.
2012-09-29T01:33:34 < zyp> heh
2012-09-29T01:37:48 < Mazingaro> Laurenceb: I'm not humiliating you
2012-09-29T01:37:57 < Mazingaro> Laurenceb: archimedes wasn't so bad
2012-09-29T01:38:20 < Thorn> trolling +2
2012-09-29T01:38:26 < zyp> also, the size of the pdf files should be roughly correlating to the amount of information in them, no?
2012-09-29T01:38:49 < zyp> more information is usually better, so that's not a bad thing :p
2012-09-29T01:39:02 < emeb> *nod*
2012-09-29T01:39:14 < Mazingaro> Laurenceb: these were epic times... ah the time passed on Amiga :)
2012-09-29T01:39:18 < emeb> organization counts too.
2012-09-29T01:46:17 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-09-29T02:01:27 < Thorn> Texas Instruments engineering technician Ruben Voluntad hasn’t always been trim. Though active, the scale read 210 pounds when he first started at TI’s Santa Clara, Calif., campus.
2012-09-29T02:03:32 < emeb> wtf?
2012-09-29T02:04:02 < Thorn> wtf indeed. that was a quote from ti.com
2012-09-29T02:06:38 < Laurenceb_> haha
2012-09-29T02:06:44 < Laurenceb_> OMAP keeps you fit too
2012-09-29T02:06:53 < Laurenceb_> hes going to be putting on weight now
2012-09-29T02:07:16 < Thorn> they also have a bring your daughter to work day
2012-09-29T02:08:06 < Thorn> no mention of potato batteries though
2012-09-29T02:09:53 < Laurenceb_> pedobear approves
2012-09-29T02:11:08 < zyp> Laurenceb_ doesn't work there, so bringing daughter to work shouldn't do any harm
2012-09-29T02:11:19 < Laurenceb_> lol
2012-09-29T02:13:29 < emeb> probably not a good idea to train the cat to play with used SMT tape.
2012-09-29T02:13:43 < Thorn> "What will you take back to campus with you this fall? A great tan? A perfected golf swing? Tons of Instagrams of your hiking trip through Europe with nothing but 500 euro and a pair of socks?"
2012-09-29T02:14:12 < emeb> risque!
2012-09-29T02:40:51 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-29T02:43:07 < corecode> is there a opencm3 channel as well?
2012-09-29T02:43:20 < Thorn> #libopencm3
2012-09-29T02:49:04 < corecode> thanks
2012-09-29T02:51:49 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-29T02:57:38 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-09-29T02:59:53 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 255 seconds]
2012-09-29T03:24:21 < dongs> hello trolls
2012-09-29T03:43:19 -!- daku is now known as DaKu
2012-09-29T03:48:26 < upgrdman> hello
2012-09-29T03:52:39 < emeb> yo dongs
2012-09-29T04:00:04 -!- izua [~izua@unaffiliated/izua] has quit [Ping timeout: 248 seconds]
2012-09-29T04:05:06 < dongs> sup
2012-09-29T04:05:11 < dongs> emeb: the mysemi driver is cheap
2012-09-29T04:05:16 < dongs> and seems really nice w/pwm etc
2012-09-29T04:05:23 < dongs> like <50c/driver
2012-09-29T04:05:38 < dongs> so i think im going wiht that
2012-09-29T04:05:48 < dongs> http://www.my-semi.com/content/products/product_mean.aspx?id=4 this shit.
2012-09-29T04:06:07 < Thorn> dongs: what is your project?
2012-09-29T04:06:13 < dongs> Thorn: mostly trolling
2012-09-29T04:07:10 < emeb> dongs: what do they cost?
2012-09-29T04:07:15 < Thorn> I thought you were designing something commercial with those led drivers
2012-09-29T04:07:21 < dongs> nah
2012-09-29T04:07:23 < emeb> and how fast can you load new data to them?
2012-09-29T04:07:52 < dongs> emeb: <50c/ea, 10 or 20mbit
2012-09-29T04:08:01 < dongs> ya 20mbit max
2012-09-29T04:08:03 < emeb> yow!
2012-09-29T04:08:12 < emeb> 2x for the 8x8.
2012-09-29T04:08:15 < dongs> yes
2012-09-29T04:08:21 < dongs> way cheaper than ST driver and free pwm
2012-09-29T04:08:27 < emeb> no kidding.
2012-09-29T04:08:42 < emeb> multi-color is like falling off a log.
2012-09-29T04:09:06 < dongs> so im gona pickup a 100 of them or so once china holiday is over next week
2012-09-29T04:09:10 < emeb> where do you get them? Not Digi-Key or Mouser for sure.
2012-09-29T04:09:11 < dongs> and design teh crap  for it meanwhile
2012-09-29T04:09:26 < dongs> yeah, my usual supplier of crap from china has the m
2012-09-29T04:09:41 < dongs> i got hte quote right before they went on a holiday of doom
2012-09-29T04:09:55 < dongs> http://en.wikipedia.org/wiki/Golden_Week_(China)
2012-09-29T04:10:26 < emeb> they seem to have a lot of those.
2012-09-29T04:10:31 < dongs> haha no shit
2012-09-29T04:11:19 < dongs> one thing im not sure is what package type they had
2012-09-29T04:11:31 < dongs> i did say ssop24.. hm im gonna ask if one of their dudes is stil around
2012-09-29T04:12:01 < dongs> theres usualyl common package that everyone uses and other shit is much more expensive/ahrder to get
2012-09-29T04:12:05 < emeb> that would be easier.
2012-09-29T04:12:14 < emeb> QFN isn't a huge deal though.
2012-09-29T04:12:26 < dongs> qfn probably has better power dissipation though
2012-09-29T04:12:42 < emeb> yup - big thermal pad in the middle.
2012-09-29T04:13:11 -!- |sark [~|sark@109.235.51.202] has quit [Read error: Connection reset by peer]
2012-09-29T04:13:19 < Thorn> there's a ebay user named my9221
2012-09-29T04:13:26 < emeb> heh
2012-09-29T04:13:28 < dongs> haha
2012-09-29T04:13:32 < Thorn> but he has no my9221 in stock
2012-09-29T04:14:07 < dongs> looks liek ssop24 is what everyone has
2012-09-29T04:14:49 < emeb> datasheet at seeed, but they don't appear to stock them.
2012-09-29T04:15:04 < dongs> seeed used them for rainblowduino, or at least one revision of it
2012-09-29T04:22:58 < emeb> bbl
2012-09-29T04:23:02 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-29T04:35:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-29T04:42:04 -!- theAdib_ [~adib@dslb-088-074-159-205.pools.arcor-ip.net] has joined ##stm32
2012-09-29T04:45:36 -!- theAdib [~adib@dslb-088-074-140-062.pools.arcor-ip.net] has quit [Ping timeout: 260 seconds]
2012-09-29T04:46:24 < emeb_mac> and they're yummy!
2012-09-29T04:55:37 < dongs> http://www.youtube.com/watch?v=XkGphC5hxuM#t=92s
2012-09-29T05:00:35 < emeb_mac> dongs: possible alternative to MIC2981 high-side driver - http://www.mouser.com/ProductDetail/ON-Semiconductor/NVTR0202PLT1G/?qs=sGAEpiMZZMvsEea7gdidXsXmKFm06sQ%252bMVMOL7oTytc%3d
2012-09-29T05:01:07 < emeb_mac> see zyp's matrix driver w/ discrete PFETs for reference.
2012-09-29T05:01:30 < dongs> isnt rdson pretty terrible on that part?
2012-09-29T05:02:12 < dongs> 1ohm @ 5V
2012-09-29T05:03:40 < emeb_mac> 1 ohm @ 400mA = 0.4V - a lot less drop than the 1.9V Vcesat of the MIC
2012-09-29T05:04:50 < dongs> ah. how about a dual/quad package of those?
2012-09-29T05:05:20 < dongs> they're not that cheap though
2012-09-29T05:05:31 < emeb_mac> 8x is cheaper than 1 MIC
2012-09-29T05:05:36 < dongs> is it?
2012-09-29T05:05:52 < dongs> checking
2012-09-29T05:07:05 < emeb_mac> MIC in 25qty = $1.81
2012-09-29T05:07:50 < emeb_mac> PFET in 25qty = $0.18 X8 = $1.44
2012-09-29T05:10:31 < dongs> dual packages seems cheaper
2012-09-29T05:10:47 < dongs> i.e random shit like this  http://www.digikey.com/product-detail/en/DMP2240UDM-7/DMP2240UDMDICT-ND/1774243
2012-09-29T05:11:43 < emeb_mac> that's good - a few cents less per chl and higher current handling.
2012-09-29T05:12:01 < emeb_mac> the 400ma of the original was marginal - would like > 480ma.
2012-09-29T05:12:32 < dongs> but in case of individual, youre only driving 6 leds each isnt it
2012-09-29T05:12:39 < dongs> so thats onyl like 120ma
2012-09-29T05:12:51 < dongs> er, individual is 3 leds each
2012-09-29T05:13:19 < emeb_mac> each column of the matrix has 8 x 3 LEDs in it -  24 x 20ma = 480ma
2012-09-29T05:14:39 < dongs> oh, oops, yourrite
2012-09-29T05:15:13 < dongs> not sure which package is mor common though
2012-09-29T05:15:16 < dongs> sot23-6 or so8
2012-09-29T05:15:38 < dongs> looks like so8
2012-09-29T05:16:02 < dongs> hm but those are more expensive
2012-09-29T05:16:15 < emeb_mac> so23-6 is smaller too
2012-09-29T05:17:12 < dongs> hmm that diodes part is actually cheapest in sot23
2012-09-29T05:17:18 < dongs> i just picked it at random
2012-09-29T05:17:46 < emeb_mac> lower Rdson too - only 0.2ohm if you keep Vgs > 2V.5V
2012-09-29T05:17:53 < dongs> right
2012-09-29T05:18:27 < emeb_mac> so drop is 0.096V
2012-09-29T05:18:35 < emeb_mac> @480ma
2012-09-29T05:19:29 < emeb_mac> But - discrete PFETs have a slight disadvantage -
2012-09-29T05:19:59 < dongs> turn off time?
2012-09-29T05:20:26 < emeb_mac> if you're driving the LEDs from a voltage != 3.3V then you need 5V tolerant OC outputs on GPIO and a pullup to 5V.
2012-09-29T05:20:52 < emeb_mac> and the green LEDs in those arrays need > 3.6V for full bright.
2012-09-29T05:22:48 < dongs> hm, why the 5V requirement? i thought these are logic level gates anyway
2012-09-29T05:24:29 < emeb_mac> the Vgs is referenced to the PFET source, which is tied to +5V.
2012-09-29T05:24:41 < emeb_mac> to turn off, you pull the gate to +5V
2012-09-29T05:25:01 < emeb_mac> to turn on you pull the gate down
2012-09-29T05:25:54 < emeb_mac> the MCU running @ 3.3V can't turn the PFET off by itself, so you use a pullup to +5V and OC gpio to pull down.
2012-09-29T05:26:24 < dongs> right but if its logic level gate which this is, the vgs is way below 5V
2012-09-29T05:26:45 < emeb_mac> I don't see where it says this is a logic level gate.
2012-09-29T05:26:50 < Bird|lappy> actually, go for a higher Vgs(th) on the PMOS
2012-09-29T05:27:05 < Bird|lappy> such that 1.7V Vgs is not enough to turn the part on
2012-09-29T05:27:28 < emeb_mac> Bird|lappy: you have a particular part in mind?
2012-09-29T05:27:30 < dongs> emeb_mac: the digikey parametric search
2012-09-29T05:27:39 < dongs> which i immediately filtered on 'logic level gate' feature
2012-09-29T05:28:22 < dongs> http://www.fairchildsemi.com/ds/ND/NDS356AP.pdf ive used these wiht 3.3V logic and they surely worked
2012-09-29T05:28:27 < dongs> without any opencollector/pullup fuckery
2012-09-29T05:30:43 < emeb_mac> yep
2012-09-29T05:30:49 < emeb_mac> that looks just right.
2012-09-29T05:33:04 < emeb_mac> Vgs = -1.6V @250ua
2012-09-29T05:33:53 < emeb_mac> Vgs th can go as low as -0.6 over temp, so you might get more leakage w/ 3.3V outs, but shouldn't be enough to turn on the wrong columns.
2012-09-29T05:34:14 -!- WorkerBuzz [~Buzz@65-113-103-130.dia.static.qwest.net] has joined ##stm32
2012-09-29T05:35:07 < emeb_mac> and $0.14 / chl in qty 25, so cheaper than MIC too.
2012-09-29T05:35:21 < dongs> yeah but that  fet is EOL since like 2 years ago i think
2012-09-29T05:35:26 < dongs> i used that shit long time ago
2012-09-29T05:35:38 < emeb_mac> bummer
2012-09-29T05:35:48 < dongs> nrnd
2012-09-29T05:36:30 < emeb_mac> but dk has 12k of them :)
2012-09-29T05:37:17 < Thorn> what do you need, logic level mosfets? how about some IRLML*?
2012-09-29T05:37:44 < dongs> i guess cant use nfets because of high side drive?
2012-09-29T05:37:56 < emeb_mac> right
2012-09-29T05:39:09 < dongs> well i do ha ev a bunch of them heh
2012-09-29T05:39:11 < emeb_mac> Oh wait -
2012-09-29T05:39:19 < emeb_mac> I was looking at something different:
2012-09-29T05:39:24 < dongs> and you got pricing wrong
2012-09-29T05:39:28 < dongs> its 0.14 @ 1k not 25
2012-09-29T05:39:32 < dongs> i didnt think the  shit was that cheap
2012-09-29T05:40:28 < emeb_mac> yeah
2012-09-29T05:40:41 < emeb_mac> I was looking at some lame Toshiba parts
2012-09-29T05:40:51 < emeb_mac> high Rdson and low current.
2012-09-29T05:47:31 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-09-29T05:48:27 < emeb_mac> "picking discretes is haaaard."
2012-09-29T05:51:16 < Bird|lappy> emeb_mac, not off the top of my head
2012-09-29T05:51:23 < Bird|lappy> you may need to go with a non-LL-gate part
2012-09-29T05:51:39 < emeb_mac> there are a bunch that come close.
2012-09-29T05:56:19 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-29T05:56:29 < Thorn> nice.
2012-09-29T05:57:01 < Thorn> how to disable weendows by installing a driver.
2012-09-29T05:57:10 < dongs> youre doing it wrong
2012-09-29T05:57:42 < Thorn> I found a usb sniffer called busgof, ans installed it
2012-09-29T05:57:49 < Thorn> I have a 64-bit win 7
2012-09-29T05:57:57 < Thorn> guess what happens next
2012-09-29T05:58:27 < Thorn> all usb devices get disconnected and newly connected ones don't even enumerate
2012-09-29T05:58:36 < dongs> awesome. you picked wrong shit
2012-09-29T05:58:41 < dongs> the one that installs filter driver above all usbh
2012-09-29T05:58:44 < Thorn> or maybe get disconnected right after enumeration.
2012-09-29T05:58:45 < dongs> and i bet you had libusb installed
2012-09-29T05:58:49 < Thorn> no mouse, no keyboard
2012-09-29T05:58:50 < dongs> which also does same faggotry
2012-09-29T05:59:00 < dongs> i fucking hate opensores
2012-09-29T05:59:03 < dongs> DOING IT WRONG.
2012-09-29T05:59:27 < Thorn> all because of an unsigned driver LOL
2012-09-29T05:59:43 < dongs> anyone who cant afford to sign a driver in 2012 should just die
2012-09-29T05:59:56 < dongs> see: all tarduino dudes who still refuse to sign their custom cdc infs
2012-09-29T06:01:39 < Bird|lappy> at least Microsoft has figured out that providing a user-mode adapter layer for vendor-specific USB devices is a good idea these days.  *points at WinUSB.sys*
2012-09-29T06:02:00 < dongs> sure. they even use it for MTP shit.
2012-09-29T06:02:10 < dongs> i mean internally with windows stuff.
2012-09-29T06:02:27 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Read error: Connection reset by peer]
2012-09-29T06:02:35 < Thorn> no isochronous transfers with that crap
2012-09-29T06:02:37 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-29T06:02:37 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-09-29T06:02:38 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-09-29T06:02:46 < dongs> who the fuck needs those
2012-09-29T06:02:59 < Bird|lappy> Thorn, if you need them, you probably want a kernel-mode driver anyway, just saying...
2012-09-29T06:03:24 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-29T06:03:40 < Bird|lappy> don't want the scheduler ruining your perfectly timed isochronous transfer now, do we? :)
2012-09-29T06:05:40 < Bird|lappy> besides, look at the torrential tide of proprietary crapware coming out.  bad software is everybody's problem, so blaming any one subgroup of software developers for all of it isn't helping.
2012-09-29T06:06:01 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Read error: Connection reset by peer]
2012-09-29T06:11:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-29T06:12:06 < Thorn> wtf, why doesn't windows allow changing device configuration?
2012-09-29T06:22:01 < Thorn> http://code.google.com/p/busdog/issues/detail?id=9
2012-09-29T06:22:16 < Thorn> "Beware this program!!"
2012-09-29T06:22:23 < dongs> Thorn: haha, you installed OPENSORES USB sniffer???
2012-09-29T06:22:38 < Thorn> If you find busdog useful and would like to donate towards a code signing certificate:
2012-09-29T06:22:49 < Thorn> what else is there.
2012-09-29T06:25:41 < dongs> i used some payware stuff that did the job nicely
2012-09-29T06:25:46 < dongs> andf didnt install a fucking filter driver on al ldevices
2012-09-29T06:26:45 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-29T06:26:48 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-29T06:26:56 < dongs> hhd usb analyzer
2012-09-29T06:28:19 -!- WorkerBuzz [~Buzz@65-113-103-130.dia.static.qwest.net] has quit [Ping timeout: 252 seconds]
2012-09-29T06:34:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-29T06:37:29 < Thorn> this is nice but it doesn't actually capture data http://vusb-analyzer.sourceforge.net/
2012-09-29T07:09:51 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-29T07:17:52 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-29T07:28:31 < zippe> Thorn: If you have  VM, capturing USB traffic from it is ez
2012-09-29T07:29:34 < zippe> dongs: Now, if Windows had a working userland USB API, life would suck less
2012-09-29T07:31:45 < Thorn> I'm reading docs for the synopsys core and it apparently waits for SET_ADDRESS after a reset and refuses to do anything else. but I've seen GET_DESCRIPTOR requests with address==0 (afair windows always does this). wtf
2012-09-29T07:32:16 < Thorn> zippe: yes I've read about that too
2012-09-29T07:34:04 < Thorn> (and why does it handle control requests in hardware?)
2012-09-29T07:49:42  * mrnuke is bitching about stm32 peripheral library
2012-09-29T07:49:47 < mrnuke> just ignore me
2012-09-29T07:50:58 < dongs> haha
2012-09-29T07:56:01 < mrnuke> oh wait, I don't have to use the periph lib :D
2012-09-29T08:06:16 < Thorn> wait until you see usb drivers.
2012-09-29T08:07:38 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-09-29T08:23:52 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-29T08:52:20 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-29T08:52:20 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-29T08:52:20 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-29T09:02:43 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-09-29T09:06:41 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-29T09:11:36 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-09-29T10:26:40 < dongs> http://i.imgur.com/xAjae.gif
2012-09-29T10:27:31 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-29T10:43:00 < dongs> http://i.imgur.com/mAfRq.png any idea how to genrate this with hardware pwm?
2012-09-29T10:46:57 < jpa-> DMA transfer to timer ARR?
2012-09-29T10:47:20 < jpa-> and toggle-type capture output
2012-09-29T10:47:37 < dongs> im checking if implementation depends on slightly shorter initial pulse
2012-09-29T10:47:51 < dongs> if not, im thinking just run pwm at say 60us period, and 0 is 50,50 duty and 1 is 100,0
2012-09-29T10:47:52 < jpa-> then you can just feed a bunch of pulse lengths there
2012-09-29T10:47:55 < dongs> each bit taking up 2 periods
2012-09-29T10:48:34 < dongs> arr is period?
2012-09-29T10:48:38 < jpa-> yeah
2012-09-29T10:48:47 < dongs> hmm.
2012-09-29T10:49:32 < jpa-> i would think the different lengths of on and off pulses are just a matter of your digital channel threshold
2012-09-29T10:50:22 < jpa-> and therefore you could just write "010101" or "00110011011" on SPI channel configured for 29.4kHz
2012-09-29T10:50:42 < dongs> o'rly
2012-09-29T10:50:47 < dongs> HMMM
2012-09-29T10:51:13 < dongs> thing is
2012-09-29T10:51:16 < dongs> this is onewire protocol
2012-09-29T10:51:23 < dongs> i guess I could swap SPI modes
2012-09-29T10:51:39 < dongs> only problem is theres no clock
2012-09-29T10:51:41 < dongs> for receive
2012-09-29T10:51:52 < jpa-> you need to receive also?
2012-09-29T10:51:54 < dongs> yes
2012-09-29T10:52:02 < dongs> but yeah output by SPI sounds like a win
2012-09-29T10:52:11 < dongs> i totally didnt think of that
2012-09-29T10:52:28 < dongs> receive i guess I c an just run a timer and watch for exti on my pin
2012-09-29T10:52:34 < dongs> and count # of transitions
2012-09-29T10:52:45 < dongs> 1 transition per tick = 1, >1 = 0
2012-09-29T10:53:27 < jpa-> you can run input capture channel and collect the timestamps
2012-09-29T10:53:52 < jpa-> then if "pulse length = 34µs, ok it was zero and skip the 3 following pulses" or similar
2012-09-29T10:54:05 < dongs> nah, its self-clocking, the period could be different
2012-09-29T10:54:21 < dongs> the overall proportion stays
2012-09-29T10:54:42 < jpa-> well equally easy if you have the pulse lengths in a buffer
2012-09-29T10:54:45 < dongs> 1010 = 0, 1100 = 1
2012-09-29T10:54:55 < dongs> yeah
2012-09-29T10:55:06 < dongs> and since its half duplex...
2012-09-29T10:55:15 < dongs> hmm only problem is..
2012-09-29T10:55:21 < dongs> no i dont have spi available on teh board (easily)
2012-09-29T10:55:25 < dongs> lame.
2012-09-29T10:55:50 < dongs> looks like arr dicking it is
2012-09-29T10:56:07 < jpa-> do you have timer output channel available? :)
2012-09-29T10:56:15 < dongs> ya
2012-09-29T10:56:17 < dongs> that one is easy
2012-09-29T10:56:19 < jpa-> good :)
2012-09-29T10:56:25 < dongs> ill do it with timer onyl i guess.
2012-09-29T10:56:31 < jpa-> yeah, not too bad
2012-09-29T10:56:46 < jpa-> takes somewhat more buffer space but otherwise quite easy
2012-09-29T10:56:52 < dongs> thats not a problem
2012-09-29T10:57:02 < dongs> its a onetime setup, space/performance doent matter
2012-09-29T10:59:54 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-29T11:00:05 < dongs> cjbeard, i had some imgur link for you
2012-09-29T11:00:08 < dongs> oh no, not imgur
2012-09-29T11:00:14 < dongs> Please link deviantART or FurAffinity profiles as a "resume" of sorts, or specific pieces from them.
2012-09-29T11:02:21 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-09-29T11:05:58 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 246 seconds]
2012-09-29T11:18:06 < cjbaird> dongs: wtf are you on about?
2012-09-29T11:18:34 < dongs> haha nothing just that quote.
2012-09-29T11:38:33 <+Steffann> I bet you have a devianART account dongs
2012-09-29T11:38:47 < dongs> i bet i dont
2012-09-29T11:40:14 <+Steffann> That's what i would say as well, bad for your reputation
2012-09-29T11:42:03 < cjbaird> The most selective pedos like ourselves would exclusive pixiv members.
2012-09-29T11:44:27 < cjbaird> I need to get my dinner cooked and ate before the housemate's landwhale AD&D roleplay Most Entitled friends show up. :/
2012-09-29T11:45:12 <+Steffann> Time to find a new housemate
2012-09-29T11:45:24 < ABLomas> =)
2012-09-29T11:45:37 < cjbaird> There's a shortage of landlords who'll do $100/wk rent deals...
2012-09-29T11:45:53 <+Steffann> Oh, he is the owner of the house?
2012-09-29T11:47:15 < cjbaird> Back when Charl and I were still going out, I'd made her a Snickers/Mars Bar Cheesecake for her birthday party.. Queen Landwhale tries to claim the rest of the cake (about 3 kg of it) "..to take home for my boyfriend". That wasn't happening Ms 200 kilos.
2012-09-29T11:50:02 <+Steffann> Sometimes i start to think she was your imaginary gf.. or you are just not over her yet :P
2012-09-29T11:56:26 -!- barthess [~barthess@178.154.57.174] has joined ##stm32
2012-09-29T11:57:50 -!- theAdib_ [~adib@dslb-088-074-159-205.pools.arcor-ip.net] has quit [Quit: Ex-Chat]
2012-09-29T11:58:03 < cjbaird> Hah. 'Queen Landwhale' of the Sydney AD&D scene actually ended up marrying the (gormless..) boyfriend, and she also keeps dropping in conversation about how he works for Google--- he's 37 years old, and works at the /help desk/. She never mentions that part. :P
2012-09-29T12:27:30 <+Steffann> Man, why you know so many about people cjbaird?
2012-09-29T12:28:00 -!- tavish3 [~tavish@125.63.110.142] has joined ##stm32
2012-09-29T12:29:32 < tavish3> how do you find the reason for a hardfault?
2012-09-29T12:29:55 <+Steffann> zyp is the expert in that
2012-09-29T12:30:11 <+Steffann> There's a register which contains some info about the cause, but i don't remember which one
2012-09-29T12:30:35 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-29T12:30:54 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-29T12:31:01 <+Steffann> SCB.CFSR it is
2012-09-29T12:32:54 < tavish3> Steffann: thanks, i'll try to see what it has
2012-09-29T12:33:35 <+Steffann> http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-06-29.html there some talk about it around 09:59:26
2012-09-29T12:33:39 <+Steffann> *there is
2012-09-29T12:33:58 <+Steffann> Just google for "zyp hardfault" :D
2012-09-29T12:35:20 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-09-29T12:38:58 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-29T12:39:16 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-29T12:39:16 < jpa-> SCB->CFSR
2012-09-29T12:39:27 < jpa-> you can find details in the STM32 Cortex-M3 programming manual
2012-09-29T12:42:11 <+Steffann> Zyp's summary in this channel is better jpa-
2012-09-29T12:47:26 <+Steffann> Would be nice info for the wiki :)
2012-09-29T12:48:03 < jpa-> the probable causes list would be
2012-09-29T12:48:20 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-29T12:48:44 -!- iR0b0t1 [~dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-29T12:49:08 <+Steffann> Isn't that list endless? :)
2012-09-29T12:49:24 < jpa-> nope
2012-09-29T12:51:28 < jpa-> cfsr invalid state => address bottom bit is not 1 => you are not compiling for thumb2 or stack corruption of return address;  cfsr invalid instruction => you are not compiling for thumb2 or you are linking to non-thumb libs;   bus fault => reading from invalid pointer   etc.
2012-09-29T12:53:17 <+Steffann> I know what you mean. i was just being 'funny' :P
2012-09-29T12:53:24 <+Steffann> As you now sort of summarized the causes
2012-09-29T12:57:14 < jpa-> maybe you can now put it on the wiki
2012-09-29T12:57:26 < jpa-> and also take a nice screenshot of the relevant registry description in the pdf
2012-09-29T12:59:11 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-29T13:03:13 < tavish3> i'll do it, once I figure it out
2012-09-29T13:03:24 -!- tavish3 [~tavish@125.63.110.142] has quit [Quit: Lost terminal]
2012-09-29T13:03:59 < Thorn> there's a large table in The Book (several pages) that lists exceptions and possible causes
2012-09-29T13:04:56 <+Steffann> "the book" ?
2012-09-29T13:05:26 < Thorn> by Joseph Yu
2012-09-29T13:05:33 -!- barthess [~barthess@178.154.57.174] has quit [Quit: Leaving.]
2012-09-29T13:21:07 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-29T13:39:55 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-29T13:42:19 < Thorn> what does this mean exactly? they don't ship libraries that are compiled with hard floating point instructions enabled? https://sourcery.mentor.com/sgpp/lite/arm/portal/kbentry27
2012-09-29T13:45:22 < zyp> «… but are not included with CodeSourcery's Lite or Personal Edition toolchains.»
2012-09-29T13:45:29 < zyp> so yes
2012-09-29T13:46:11 < Thorn> I mean, is hard float ABI == hard float instructions?
2012-09-29T13:46:28 < zyp> no
2012-09-29T13:46:45 < Thorn> they recommend -mfpu=vfp3 -mfloat-abi=softfp for example
2012-09-29T13:47:06 < zyp> ABI defines the interface of a function
2012-09-29T13:47:08 < Thorn> will coe compiled with these flags have VFP instructions? it probably will
2012-09-29T13:47:16 < zyp> i.e. how arguments to a function is passed
2012-09-29T13:47:21 < Thorn> right
2012-09-29T13:47:47 < zyp> -mfloat-abi=hard will pass float-arguments in FPU registers, -mfloat-abi=softfp will pass float-argument in CPU registers
2012-09-29T13:48:55 < Thorn> I'm looking at http://www.freddiechopin.info/en/articles/34-news/82-dwa-nowe-przyklady-stm32f4-i-lpc1769 , it says "it is not possible to use CodeSourcery toolchain, because it lacks libraries supporting floating point coprocessor"
2012-09-29T13:49:00 < zyp> the former is only possible if you have an FPU, the latter is always used when you don't have an FPU
2012-09-29T13:51:03 < Thorn> but that FAQ entry only says only hard float ABI libraries aren't shipped, and nothing about libraries that use VFP
2012-09-29T13:51:14 < zyp> remember that you have three possibilities a function/library can be compiled for: softfpu functions and softfpu ABI, FPU instructions and softfpu ABI, FPU instructions and hardfpu ABI
2012-09-29T13:51:40 < zyp> the two former are compatible since they use the same ABI, so they can be linked into the same application
2012-09-29T13:52:01 < zyp> so you can have some functions using the FPU and some using only softfpu
2012-09-29T13:52:17 < Thorn> so does sourcery lite include #2 or not?
2012-09-29T13:52:59 < zyp> I don't know
2012-09-29T13:53:11 < Thorn> that's the question.
2012-09-29T13:53:12 < zyp> does it matter?
2012-09-29T13:54:02 < zyp> are you calling library functions that would benefit from using FPU instructions?
2012-09-29T13:55:31 < Thorn> I'm concerned about libm, if it's soft fp only it's not good for f4m targets
2012-09-29T13:56:08 < Thorn> I tried arm-none-eabi-gcc -print-multi-lib and it doesn't say anything about floating point
2012-09-29T13:56:29 < Thorn> so there's probably only one version of libm
2012-09-29T13:56:51 < jpa-> Thorn: libm that comes with codesourcery 2011.09 atleast seems to be softfp
2012-09-29T13:56:54 < zyp> sounds right
2012-09-29T13:56:59 < jpa-> not too much work to recompile it
2012-09-29T13:58:03 < zyp> looks like yagarto also only includes softfp
2012-09-29T13:58:10 < jpa-> it does however use the float multiply etc. routines from gcc, so i don't know if it might still benefit a little from the FPU
2012-09-29T13:58:29 < Thorn> see last 3 comments, my output for sourcery is almost the same as theirs for yagarto http://www.freddiechopin.info/en/articles/34-news/82-dwa-nowe-przyklady-stm32f4-i-lpc1769
2012-09-29T13:58:55 < zyp> ah, I get that same output for yagarto
2012-09-29T13:59:11 < Thorn> for things that don't compile to single instructions, definitely
2012-09-29T13:59:44 < jpa-> nothing float-related compiles to single instructions without FPU :)
2012-09-29T14:00:01 < zyp> I'll probably recompile newlib to add a hardfpu variant when I start using it then
2012-09-29T14:00:03 < jpa-> but you won't get full benefit of the FPU with codesourcery libm
2012-09-29T14:00:25 < jpa-> i'll recompile just libm part and ditch the rest of newlib :P
2012-09-29T14:00:26 < Thorn> I'm not talking about no-VFP targets
2012-09-29T14:00:56 < jpa-> Thorn: i know, but that's what the codesourcery binary of libm is compiled for
2012-09-29T14:01:01 < Thorn> looks like switching to the arm toolchain is even better
2012-09-29T14:01:20 < jpa-> why?
2012-09-29T14:03:17 <+Steffann> Dekar's toolchain supports it too doesn't it?
2012-09-29T14:03:37 <+Steffann> oh, yes "Multilib support for several thumb targets: armv6s-m (cortex-m0/cortex-m1) armv7-m (cortex-m3) armv7e-m (cortex-m4/cortex-m4f including FPU and DSP instruction support)"
2012-09-29T14:04:23 < jpa-> i wonder what is the easiest way to use the DSP instructions (packed 4x8-bit and 2x16-bit) in GCC
2012-09-29T14:04:30 < Thorn> gcc-arm also supports cortex-r in addition to that (has anyone used cortex-r btw?)
2012-09-29T14:06:20 < zyp> I know there is cortex-r-cores in some of the stuff I'm playing with at work
2012-09-29T14:06:49 < Thorn> clarified http://stm32.izua.ro/wiki/Sourcery_CodeBench_Lite a little.
2012-09-29T14:08:44 <+Steffann> Uhm, zyp … is there any uc that isn't one of the project you work with? :D
2012-09-29T14:08:51 <+Steffann> *projects
2012-09-29T14:09:32 < zyp> hmm?
2012-09-29T14:11:37 < zyp> I work in the smartphone industry, is it strange that I'm around arm cores of all kinds? :p
2012-09-29T14:11:49 <+Steffann> No, that was my point
2012-09-29T14:14:10 < jpa-> but arm cores are just a small part of µcs :)
2012-09-29T14:14:16 < jpa-> do you work with 8051? :D
2012-09-29T14:14:22 < dongs> zyp works for winphone division of nokia
2012-09-29T14:14:49 < zyp> jpa-, not that I know :p
2012-09-29T14:15:44 < claude> zyp , tensilica cores?
2012-09-29T14:16:05 < zyp> never heard about
2012-09-29T14:16:11 < claude> ah okay nm
2012-09-29T14:16:19 < Thorn> zyp: thanks for explanation.
2012-09-29T14:24:58 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has joined ##stm32
2012-09-29T14:27:30 <+dekar> jpa-, CMSIS-DSP maybe?
2012-09-29T14:29:00 <+dekar> also I would assume GCC has intrinsics for the DSP instructions
2012-09-29T14:29:23 < jpa-> the dsplib works for some cases, yeah
2012-09-29T14:29:33 < jpa-> i haven't found the intrinsics yet, but haven't looked very hard
2012-09-29T14:30:02 < jpa-> liborc in a statically compiled way might work
2012-09-29T14:31:08 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has joined ##stm32
2012-09-29T14:34:41 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-09-29T14:41:59 -!- tavish [~tavish@59.177.69.138] has joined ##stm32
2012-09-29T14:42:24 -!- tavish is now known as Guest24939
2012-09-29T14:44:20 -!- Guest24939 is now known as tavish83
2012-09-29T14:47:17 < dongs> zyp, i was told  you made a rgb array
2012-09-29T14:49:50 <+dekar> gcc arm embedded doesn't seem to be targeted at micros
2012-09-29T14:49:55 <+dekar> http://pastebin.com/RYEg4Kuu
2012-09-29T14:50:43 <+dekar> their newlib has huge buffers etc :/
2012-09-29T14:51:11 < dongs> http://www.tradekorea.com/prod-asiaDtl/P00040030/OMB.html
2012-09-29T14:51:23 < dongs> i didnt notice this jap bath i went to had this
2012-09-29T14:51:27 < dongs> and i was sitting in that shit for like 5 minutes
2012-09-29T14:51:40 < dongs> it felt sorta like mild electric shock or something
2012-09-29T14:51:45 < dongs> but now my hands are totally fucked
2012-09-29T14:51:49 < dongs> like almost no powerl
2012-09-29T14:51:58 < dongs> lol
2012-09-29T14:52:57 < dongs> hm 40 to 60hz
2012-09-29T14:55:05 < Laurenceb_> did they have "maids"  ?
2012-09-29T15:03:07 < dongs> http://i.imgur.com/xAcMW.jpg lool
2012-09-29T15:04:11 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-29T15:04:11 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-29T15:04:11 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-29T15:23:20 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-09-29T15:31:33 -!- tavish83 [~tavish@59.177.69.138] has quit [Quit: leaving]
2012-09-29T15:35:22 < dongs> zyp must be on a bus still
2012-09-29T15:35:43 -!- izua [~izua@188.26.165.44] has joined ##stm32
2012-09-29T15:35:44 -!- izua [~izua@188.26.165.44] has quit [Changing host]
2012-09-29T15:35:44 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-29T15:35:47 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-29T15:38:32 < zyp> dongs, nah, I'm visiting a friend now
2012-09-29T15:38:51 < dongs> o
2012-09-29T15:38:57 < dongs> emeb said you  made some rgb array
2012-09-29T15:39:08 < zyp> no, it's only red
2012-09-29T15:39:25 < dongs> oh is that that small matrix
2012-09-29T15:40:00 < zyp> that's the only led matrix stuff I've been playing with, so unless it's that I don't know what you are talking about
2012-09-29T15:42:13 < dongs> yeah
2012-09-29T15:42:17 < dongs> must be
2012-09-29T15:42:32 < dongs> i got those chinq rgb matrix shits on the way
2012-09-29T15:42:37 < dongs> for like 6.something/ea shipped
2012-09-29T15:42:53 < dongs> gonna make a driver for them, and for bigger panels ill later make with 5050leds
2012-09-29T15:43:16 < zyp> anyway, rgb matrices are just normal led matrices with three leds per pixel
2012-09-29T15:43:22 < dongs> right
2012-09-29T15:44:22 < zyp> driving theory is not fundamentally different, except for the fact that different colors may need different currents to achieve same brightness
2012-09-29T15:44:37 < dongs> well, im gonna be using a proper constant current driver
2012-09-29T15:44:48 < zyp> that's what I also did
2012-09-29T15:44:49 < dongs> w/pwm for brightness shit
2012-09-29T15:45:57 < dongs> now i guess we're stuck on finding cheap/good enough highside driver
2012-09-29T15:46:07 < dongs> so either pfet or some sorta onechip 8channel thing
2012-09-29T15:46:51 < zyp> I didn't find any usable one chip things, so I went with discrete pfets
2012-09-29T15:48:35 < dongs> how many are you scanning? 1/8 or 1/16
2012-09-29T15:51:47 < zyp> 1/16
2012-09-29T15:52:18 < dongs> so thats like a lot of fets
2012-09-29T15:52:19 < dongs> heh
2012-09-29T15:52:26 < zyp> ha, my friend needed to solder some stuff so he bought some cheap solder station today
2012-09-29T15:53:13 < zyp> it worked for a few minutes, and then it blew up
2012-09-29T15:53:57 < zyp> it has a led display to show temperature, the inside of the display window is now covered with capacitor innards
2012-09-29T15:54:17 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-29T15:54:43 < zyp> http://bin.jvnv.net/f/xZq63.JPG <- fets
2012-09-29T15:55:04 < dongs> zyp: lolll @ exploded caps
2012-09-29T15:57:18 -!- qyx_ [~qyx@krtko.org] has quit [Quit: Changing server]
2012-09-29T15:57:45 < Laurenceb_> http://i.imgur.com/q2czO.png
2012-09-29T15:58:00 < Laurenceb_> EMC test results
2012-09-29T15:58:07 < Laurenceb_> strong 3rd harmonic from the stm32f1
2012-09-29T15:58:41 < Laurenceb_> 6th harmonic creeping into the ISM band at 434MHz
2012-09-29T15:58:45 < Laurenceb_> that kind of sucks
2012-09-29T15:59:00 -!- qyx_ [~qyx@krtko.org] has joined ##stm32
2012-09-29T16:00:04 < Thorn> I read something about a separate ground polygon under a chip, connected to the rest of the ground in one place, for EMI prevention
2012-09-29T16:00:06 < zyp> I think we've alreadly laughed about you only doing a two layer design
2012-09-29T16:00:54 < zyp> no wonder your board is noisy when you look at how it's designed :p
2012-09-29T16:01:03 < dongs> its the code
2012-09-29T16:01:27 < dongs> less spaces = more EMC noise
2012-09-29T16:02:33 < Laurenceb_> lol
2012-09-29T16:02:45 < Laurenceb_> its actually borderline
2012-09-29T16:02:48 < Laurenceb_> might pass
2012-09-29T16:02:55 < Laurenceb_> need to recalibrate the chamber
2012-09-29T16:03:03 <+Steffann> lol
2012-09-29T16:06:05 < jpa-> Laurenceb_: isn't the ISM band the only place where it doesn't matter? :)
2012-09-29T16:07:36 < jpa-> i did a 2-layer design for proto and will add a ground & supply layer when it gets to production
2012-09-29T16:09:43 < zyp> I just do four layer protos
2012-09-29T16:10:56 <+Steffann> You are just lazy
2012-09-29T16:11:12 < zyp> why?
2012-09-29T16:11:25 <+Steffann> because i say so :P
2012-09-29T16:11:38 < zyp> if I were lazy, I'd just hit autoroute and call it a day :p
2012-09-29T16:11:41 <+Steffann> ARgh, only parents can use that argument
2012-09-29T16:12:49 < jpa-> they can't if their child is not a lazy bum
2012-09-29T16:13:13 < dongs> hmm
2012-09-29T16:13:14 < dongs>  http://www.digikey.com/product-detail/en/FDC6306P/FDC6306PCT-ND/965583
2012-09-29T16:13:31 < dongs> P-CHAN, KAWAII DESU
2012-09-29T16:13:54 <+Steffann> They can use that argument for everything jpa- :(
2012-09-29T16:14:28 < jpa-> Steffann: bored of living with parents?
2012-09-29T16:14:48 <+Steffann> I don't have the "because we say so" argument age anymore :P
2012-09-29T16:15:07 < dongs> wondering if that'll work for highside shit.
2012-09-29T16:15:23 < jpa-> dongs: sure
2012-09-29T16:15:26 < zyp> no, it's not kawaii enough
2012-09-29T16:15:39 < jpa-> i like the p-chan + n-chan ones, very easy way to make an H-bridge
2012-09-29T16:15:43 < dongs> jpa, emeb had some concerns about 3V io drive etc
2012-09-29T16:16:00 < dongs> that would have to opencollector gpio and pullup to 5V
2012-09-29T16:16:06 < dongs> to make it work properly
2012-09-29T16:16:07 < jpa-> yes
2012-09-29T16:16:11 < zyp> yeah, it says Vgs 4.5V
2012-09-29T16:16:32 < zyp> the ones I'm using are 2V or so
2012-09-29T16:16:38 < jpa-> zyp: it's for high side, so doesn't matter
2012-09-29T16:16:53 < dongs> where teh fuck do you see 4.5
2012-09-29T16:16:55 < zyp> of course it does matter
2012-09-29T16:16:55 < jpa-> you need to get the P-channel FET's gate up to the source voltage for it to properly switch off
2012-09-29T16:17:10 < jpa-> so yeah, pull-up + open collector
2012-09-29T16:17:13 < qyx_> 4.5?
2012-09-29T16:17:17 < jpa-> 74HC series buffer is a good alternative
2012-09-29T16:17:18 < qyx_> it is specified also at 2.5V
2012-09-29T16:17:30 < zyp> jpa-, doesn't help if input voltage is 3.3
2012-09-29T16:17:45 < jpa-> zyp: yes yes, but totally different issue
2012-09-29T16:18:07 < jpa-> Vgs test voltage doesn't mean shit anyways
2012-09-29T16:18:32 < jpa-> Vgs_th can give an idea, and the characteristic graph gives the real numbers
2012-09-29T16:18:54 < Thorn> dongs: https://ec.irf.com/v6/en/US/adirect/ir?cmd=catSearchFrame&domSendTo=byID&domProductQueryName=IRLTS2242  ?
2012-09-29T16:19:23 < Thorn> (first random hit, they have a lot of dirrefent ones)
2012-09-29T16:19:32 < dongs> thats gotta be fucking expensive
2012-09-29T16:19:33 < dongs> for 7A
2012-09-29T16:20:26 < jpa-> i did this for a h-bridge  http://koti.kapsi.fi/jpa/stuff/pix/hbridge.png
2012-09-29T16:20:26 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-29T16:20:28 < zyp> hmm, I'm using FDV302P on my board
2012-09-29T16:21:03 < Thorn> http://www.irf.com/product-info/datasheets/data/irf7210.pdf
2012-09-29T16:21:10 < dongs> RDSOn 13ohm?!
2012-09-29T16:21:38 < zyp> doesn't really matter much
2012-09-29T16:21:41 < dongs> Thorn: no all that shit is huge
2012-09-29T16:21:46 <+Steffann> Why this 'weird' 7uF jpa- ?
2012-09-29T16:21:48 < dongs> sot23 is about the samllest
2012-09-29T16:21:49 <+Steffann> Why 7?
2012-09-29T16:21:58 < jpa-> Steffann: because i had them for some crazy reason
2012-09-29T16:22:05 <+Steffann> :S
2012-09-29T16:22:16 < jpa-> (can't remember the exact value, 4.7µF would work just as well if it pleases you :)
2012-09-29T16:22:25 < zyp> more power dissipated in the high side driver equals less power dissipated in the constant current low side driver
2012-09-29T16:22:58 < dongs> well each led is 20mA, so 24 of htem per row
2012-09-29T16:23:06 < dongs> 20ma at max anyway
2012-09-29T16:23:26 < Thorn> dongs: https://ec.irf.com/v6/en/US/adirect/ir?cmd=catProductDetailFrame&productID=IRLML6302GTRPBF https://ec.irf.com/v6/en/US/adirect/ir?cmd=catProductDetailFrame&productID=IRLML6402GTRPBF
2012-09-29T16:24:36 < dongs> i dont wanna do the oc + pullup shit, so any of those usable directly from 3.3V io?
2012-09-29T16:25:19 < Thorn> I don't see why not, they're specified from 2.7 to 4.5V and max gate is 12V
2012-09-29T16:25:21 < zyp> are you using 3.3V as supply for leds as well?
2012-09-29T16:25:27 < dongs> zyp: no.
2012-09-29T16:25:28 < dongs> 5V.
2012-09-29T16:25:48 < zyp> then you need pullups, since pfets are active low
2012-09-29T16:25:58 < dongs> :[
2012-09-29T16:26:09 < dongs> but but
2012-09-29T16:26:50 < zyp> don't you get some with built-in resistor between gate and source?
2012-09-29T16:27:29 < dongs> well
2012-09-29T16:27:39 < dongs> the problem wiht that is then gpio needs to be OC and 5V tolerant
2012-09-29T16:27:45 < dongs> y/n
2012-09-29T16:28:06 < zyp> yes
2012-09-29T16:28:14 < Laurenceb_> isnt this simple
2012-09-29T16:28:17 < dongs> well thats a fail
2012-09-29T16:28:21 < Laurenceb_> pfet
2012-09-29T16:28:32 < Laurenceb_> with pullup, then output open collector
2012-09-29T16:28:55 < jpa-> dongs: how fast do you want to switch? pull-up can be a bit slow
2012-09-29T16:28:58 < zyp> Laurenceb_, that's what we are already discussing
2012-09-29T16:29:03 < jpa-> 74HC series buffer is cheap and fast
2012-09-29T16:29:31 < Laurenceb_> oh lol
2012-09-29T16:29:37 < dongs> i need to find some chinaplce that does highside  dirvers taht are cheap and nonfail
2012-09-29T16:29:41 < dongs> wtf, they have to exist
2012-09-29T16:30:05 < jpa-> dongs: at 5V, 74HC works fine as a highside driver
2012-09-29T16:30:18 < dongs> jpa, for ~500mA current?
2012-09-29T16:30:26 < jpa-> you need 500mA gate current?
2012-09-29T16:30:34 < dongs> oh you mean for fet driver?
2012-09-29T16:30:40 < jpa-> i thought you meant that :)
2012-09-29T16:30:57 < dongs> jpa-: i mean led multiplex specific shit
2012-09-29T16:31:02 < jpa-> aha
2012-09-29T16:31:12 <+izua> there are UDN chips as a complement ot the ULN ones
2012-09-29T16:32:04 < dongs> ive looked at uln8309 or wahtever
2012-09-29T16:32:12 < dongs> apparently those are a problem because they drop something like ~2V
2012-09-29T16:32:32 < dongs> so at 5Vcc, its gonna be ~3V output, and green leds in the matrix need 3.6V for max bright
2012-09-29T16:32:59 < dongs> shit like ULN2803A
2012-09-29T16:33:08 < dongs> what does the udn part do?
2012-09-29T16:33:52 < zyp> I guess PNP transistors is the way to go then
2012-09-29T16:33:58 < dongs> haha..
2012-09-29T16:34:06 < zyp> they don't need pullup
2012-09-29T16:34:17 < zyp> and thus have no 5V tolerance issues
2012-09-29T16:34:38 < dongs> but dont they have a pile of other issues
2012-09-29T16:34:57 <+izua> dongs: ulns are low side
2012-09-29T16:35:01 <+izua> udns are high side drivers
2012-09-29T16:35:13 < dongs> izua: uh no?
2012-09-29T16:35:16 < zyp> dongs, you have to have a resistor in series with base signal instead
2012-09-29T16:35:43 < dongs> or wait, i dunno, maybe im dumb
2012-09-29T16:36:24 <+izua> dumb people make bad trolls
2012-09-29T16:36:35 <+izua> you don't want to have to turn in your badge, now do you?
2012-09-29T16:36:40 < Thorn> there're resistor-equipped transistors
2012-09-29T16:36:51 < dongs> yeah prebiased shit
2012-09-29T16:36:54 < dongs> or wahtever?
2012-09-29T16:36:56 < zyp> ah, yeah
2012-09-29T16:36:57 < Thorn> pdta114e / pdtc114e
2012-09-29T16:37:06 < Thorn> I have 200 of these
2012-09-29T16:37:43 < Thorn> they have a series base resistor and another from emitter to base
2012-09-29T16:37:46 < dongs> ok yeah uln is lowside
2012-09-29T16:38:02 < dongs> i must have been smoking something wehn i thought i saw it used on high side
2012-09-29T16:38:55 < Thorn> it's a low side switch but it's darlington so 2x voltage drop
2012-09-29T16:39:19 < zyp> still only 1.2V
2012-09-29T16:40:33 < dongs> ok fuck this tomrorow a bunch of shit arrievs that i have to assemble . bed time
2012-09-29T16:40:40 < dongs> noted all the urls
2012-09-29T16:40:44 < dongs> will see what emeb blogs about it tomrow
2012-09-29T16:42:44 <+Steffann> Hmm, 3AA batteries gives an annoying voltage. 2 times 1.2 ( rechargeable batteries ) is just enough to power a 3.3V device.
2012-09-29T16:42:59 <+Steffann> but when they're getting 'empty' it's not enough :(
2012-09-29T16:43:15 <+Steffann> *3 times
2012-09-29T16:43:25 < zyp> that's why people use voltage regulators.
2012-09-29T16:43:43 <+izua> lipo should be declared empty at ~0.9-1V/cell i think
2012-09-29T16:43:50 <+izua> so that's 2.7-3?
2012-09-29T16:43:55 < jpa-> lipo?
2012-09-29T16:44:01 <+izua> ehmj\
2012-09-29T16:44:07 <+izua> i mean the 1.2 chemestry, nimh
2012-09-29T16:44:07 < zyp> lipo is empty at 3.5V or so
2012-09-29T16:44:07 < Laurenceb_> wut
2012-09-29T16:44:10 < Laurenceb_> yeah
2012-09-29T16:44:11 <+Steffann> zyp, buck/boost  perhaps
2012-09-29T16:44:13 < jpa-> yep
2012-09-29T16:44:15 < Laurenceb_> never go below 3v
2012-09-29T16:44:15 < zyp> yep
2012-09-29T16:44:24 < jpa-> 2.5V with LDO could work
2012-09-29T16:44:42 < jpa-> or just use a single cell with boost regulator
2012-09-29T16:44:54 < Laurenceb_> i use lipo with 3.3v and a very low dropout reg from Ti
2012-09-29T16:45:06 <+Steffann> I want to use a pre-made case, which has 3 AA batteries
2012-09-29T16:45:06 <+izua> joule thief?
2012-09-29T16:45:21 < Laurenceb_> then use an smps
2012-09-29T16:45:45 < jpa-> Steffann: so what is the minimum voltage for your chips?
2012-09-29T16:45:49 < zyp> http://bin.jvnv.net/f/AcHkc.png <- I measured the lipo discharge curve in a project once
2012-09-29T16:45:50 < Laurenceb_> i want to make an stm32 controlled lipo charger
2012-09-29T16:45:53 < zyp> x axis is minutes
2012-09-29T16:45:58 < Laurenceb_> but im not sure if its worth it
2012-09-29T16:46:09 <+Steffann> I need around 3V jpa-
2012-09-29T16:46:11 < zyp> it's pretty clear when the battery is almost empty :p
2012-09-29T16:46:18 < jpa-> Steffann: no, exact number
2012-09-29T16:46:32 <+Steffann> A L0 can go pretty low, but the pir needs 3V
2012-09-29T16:46:44 < jpa-> zyp: yeah, i usually consider lipo/li-ion empty at 3.6V
2012-09-29T16:46:50 < Laurenceb_> as you need a constant current source
2012-09-29T16:47:00 < jpa-> zyp: the voltage varies a bit with the load
2012-09-29T16:47:03 < Laurenceb_> so a lipo charger ic isnt going to take up much more board area
2012-09-29T16:47:12 <+Steffann> Most buck/boost regulators ahave a pretty high Quiescent current :(
2012-09-29T16:47:26 < jpa-> Laurenceb_: lipo charger ic's are small, sot-23-5
2012-09-29T16:47:27 < karlp> Steffann: this is why I use 2 batteries, not 3.
2012-09-29T16:47:37 < Laurenceb_> jpa-: exactly
2012-09-29T16:47:43 < karlp> chargers that operate in pairs ar emore common/cheaper too
2012-09-29T16:47:47 <+Steffann> True karlp, but the case is made for 3 :(
2012-09-29T16:47:49 < Laurenceb_> Steffanx: but how much current does circuit use?
2012-09-29T16:47:54 < karlp> get a new case :)
2012-09-29T16:48:00 < karlp> mcp1623/1624,
2012-09-29T16:48:04 < zyp> jpa-, this is rather low load
2012-09-29T16:48:10 < karlp> cheap, works great, will run off a single cell too.
2012-09-29T16:48:11 <+izua> zyp: is that self discharge?
2012-09-29T16:48:16 <+izua> or under a constant load
2012-09-29T16:48:20 < zyp> constant load
2012-09-29T16:48:27 < Laurenceb_> constant current?
2012-09-29T16:48:31 < jpa-> Steffann: you could use a boost up to 5V and an LDO from there.. then you turn off the boost regulator until the voltage in capacitor falls below 4 volts
2012-09-29T16:48:35 < karlp> a little tedious that it doesn't come in fixed output versions, but so be it.
2012-09-29T16:48:41 < zyp> it's some sensors, an avr and an ant radio
2012-09-29T16:48:53 < Laurenceb_> i use ltc3467a
2012-09-29T16:49:01 < Laurenceb_> as sepic convertor
2012-09-29T16:49:17 <+Steffann> LT has some that use 16uA … maybe I want too much :P
2012-09-29T16:49:22 <+Steffann> 16uA should be ok
2012-09-29T16:52:13 <+Steffann> jpa-, maybe that's something nice to play with
2012-09-29T16:58:29 < Thorn> damn olimex board with leds between buttons so you can only see them from one specific angle
2012-09-29T17:01:35 <+Steffann> I could get another case karlp, but .. that would suck :(
2012-09-29T17:09:05 <+Steffann> With this one i just have to replace their pcb with mine and i have my battery powered pir security thingy
2012-09-29T17:26:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-29T17:28:16 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-29T17:32:21 < karlp> I think I missed what you were working on :)
2012-09-29T17:34:09 <+Steffann> I work on 1000 and 1 things :)
2012-09-29T17:34:11 <+izua> pir security thingy
2012-09-29T17:34:29 <+Steffann> don't forget wireless izua
2012-09-29T17:34:39 <+izua> it is wireless
2012-09-29T17:34:49 <+izua> you aproach it and it detects you with having to punch you in the face
2012-09-29T17:38:04 <+Steffann> I think i nee dyou for the second part
2012-09-29T17:38:07 <+Steffann> *last
2012-09-29T17:42:47 < emeb_mac> "please walk over there - don't be surprised at what happens"
2012-09-29T18:08:59 < Laurenceb_> software reset on stm32 generates a pulse on nrst pin
2012-09-29T18:09:03 < Laurenceb_> thats useful
2012-09-29T18:09:40 < jpa-> it does? cool
2012-09-29T18:09:53 < jpa-> though my software reset still does not work :)
2012-09-29T18:09:59 < zyp> nice, so you can reset external hardware off it
2012-09-29T18:11:22 < Laurenceb_> yeah
2012-09-29T18:19:42 <+Steffann> and that is documentated behaviour?
2012-09-29T18:31:23 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
2012-09-29T18:32:43 < Laurenceb_> yes
2012-09-29T18:50:44 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-29T18:55:15 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-29T19:03:11 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-29T19:14:41 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-29T19:23:19 -!- WorkerBuzz [~Buzz@65-113-103-130.dia.static.qwest.net] has joined ##stm32
2012-09-29T19:34:04 <+Steffann> Hmpf, hate it when you find the perfect chip and find out it isn't available anyway
2012-09-29T19:35:29 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-09-29T19:38:34 < Laurenceb_> what are you trying to do?
2012-09-29T19:38:50 <+Steffann> magic :P
2012-09-29T19:38:55 <+Steffann> Nah, nothing special
2012-09-29T19:39:17 <+Steffann> Just a very low (waste of) power buck-boost regulator
2012-09-29T19:40:19 < Laurenceb_> why not SEPIC?
2012-09-29T19:41:40 < jpa-> implement the smps controller in software :)
2012-09-29T19:42:24 < Laurenceb_> heh
2012-09-29T19:42:25 < qyx_> and then you do halt on mcu and mosfets make pfuuuuu
2012-09-29T19:43:16 < jpa-> actually the stm32 timers have all kinds of fancy features that could possibly avoid that :)
2012-09-29T19:43:47 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-29T19:44:37 < Laurenceb_> yeah
2012-09-29T19:44:54 < Laurenceb_> but id just go with lt3467a
2012-09-29T19:45:29 <+Steffann> I dont
2012-09-29T19:46:22 <+Steffann> "mplement the smps controller in software" <= waste of power :P
2012-09-29T19:47:13 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-29T19:50:16 -!- Posterdati [~quassel@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-29T19:52:46 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-09-29T19:56:51 -!- barthess [~barthess@178.154.70.228] has joined ##stm32
2012-09-29T20:02:20 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-29T20:02:22 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Excess Flood]
2012-09-29T20:02:48 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-29T20:07:49 -!- WorkerBuzz [~Buzz@65-113-103-130.dia.static.qwest.net] has quit [Ping timeout: 252 seconds]
2012-09-29T20:11:39 -!- barthess [~barthess@178.154.70.228] has quit [Quit: Leaving.]
2012-09-29T20:12:52 <+Steffann> Laurenceb_, the Quiescent current of that one is more than the uc + sensor use :P
2012-09-29T20:13:06 <+Steffann> + op amp
2012-09-29T20:17:33 < Laurenceb_> cant you turn it off?
2012-09-29T20:17:49 < Laurenceb_> actually it may be unsuited
2012-09-29T20:17:54 < Laurenceb_> whats you max current?
2012-09-29T20:20:46 <+Steffann> 100mA max
2012-09-29T20:21:02 < Laurenceb_> ah
2012-09-29T20:21:07 < Laurenceb_> thats quite low
2012-09-29T20:21:12 < Laurenceb_> mine was 800ma
2012-09-29T20:21:33 <+Steffann> Oh, even less than 100mA
2012-09-29T20:21:42 < Laurenceb_> - 800 max
2012-09-29T20:21:53 < Laurenceb_> so you want to run off 3x AA  ?
2012-09-29T20:21:57 <+Steffann> Yes
2012-09-29T20:22:22 <+Steffann> Maybe i can find another pir sensor with uses 2.. or 4
2012-09-29T20:22:25 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-29T20:22:48 < jpa-> or put AA-sized li-ion cells in the slot :)
2012-09-29T20:22:56 < Laurenceb_> what is the minimum voltage the sensors will run at?
2012-09-29T20:23:10 <+Steffann> AA sized liions exist?
2012-09-29T20:23:32 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-09-29T20:23:39 < karlp> sure.
2012-09-29T20:23:43 < karlp> 16850s
2012-09-29T20:23:49 < karlp> or 16550s, or whatever tehy're called
2012-09-29T20:24:09 < jpa-> i have some
2012-09-29T20:24:13 < jpa-> so yeah, they exist
2012-09-29T20:24:27 < jpa-> 14500
2012-09-29T20:24:37 < jpa-> 1200mAh
2012-09-29T20:24:40 <+Steffann> could be a solution
2012-09-29T20:25:08 < emeb> http://www.kickstarter.com/projects/18182218/freesoc-and-freesoc-mini
2012-09-29T20:25:20 < emeb> "best microcontroller in the world" <- lol
2012-09-29T20:25:22 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-29T20:26:27 <+izua> arduino?
2012-09-29T20:26:37 <+izua> please let it not be arduino
2012-09-29T20:26:42 < TitanMKD> hi
2012-09-29T20:27:35 < emeb> PSoC
2012-09-29T20:27:53 < jpa-> huh.. what is that PSoC creator supposed to be? a schematic editor?
2012-09-29T20:28:20 < emeb> jpa-: PSoC creator is sort of like an FPGA editor app.
2012-09-29T20:28:54 < emeb> PSoC has a bunch of "soft" hardware resources that can be configured @ runtime. PSoC creator helps to define the configuration.
2012-09-29T20:29:02 < jpa-> ah; not bad
2012-09-29T20:29:40 < emeb> It's an interesting approach - I did some work with them about 6-7 years ago.
2012-09-29T20:30:09 < emeb> Would have used them more except their CPU was pretty underpowered back then.
2012-09-29T20:30:19 < jpa-> seems to be that still
2012-09-29T20:30:36 < jpa-> they have cortex-M3 models but no M4
2012-09-29T20:30:48 < emeb> yep
2012-09-29T20:31:02 <+izua> so hm
2012-09-29T20:31:03 < jpa-> and prices start at $14
2012-09-29T20:31:03 < emeb> when I used it the CPU was more like a 6502.
2012-09-29T20:31:09 <+izua> that cypress chip is part mcu part fpga?
2012-09-29T20:31:15 < emeb> sorta
2012-09-29T20:31:17 <+izua> like atmel used to make (probably still do)
2012-09-29T20:31:23 < jpa-> pic32 has similar stuff
2012-09-29T20:31:49 < Laurenceb_> i dont get the point
2012-09-29T20:31:50 < emeb> there are digital resources that can be configured as counters, shifters, PRNGs, etc.
2012-09-29T20:31:57 < jpa-> it would be quite convenient to be able to just connect the closest pin to each sensor & connector etc.
2012-09-29T20:32:04 < Laurenceb_> its doesnt seem to do anything stm or similar cant
2012-09-29T20:32:05 <+izua> mhm
2012-09-29T20:32:08 < jpa-> would easy up pcb layout dramatically
2012-09-29T20:32:09 < emeb> and there are analog resources that can be filters, amps, ADCs, DACs.
2012-09-29T20:32:26 <+izua> Laurenceb_: you can design your own peripherals i guess
2012-09-29T20:32:33 < emeb> yeah kinda.
2012-09-29T20:32:37 < Laurenceb_> didnt really look like it to me
2012-09-29T20:32:38 < emeb> there are limits
2012-09-29T20:32:50 < emeb> and the peripherals use up some of the CPU bandwidth.
2012-09-29T20:32:51 < Laurenceb_> looked like hype to me
2012-09-29T20:32:55 < Laurenceb_> yeah
2012-09-29T20:33:21 < emeb> It's great for low-rate data acquisition apps
2012-09-29T20:33:23 < karlp> dongs only tools too right?
2012-09-29T20:33:31 < emeb> but you can't push it too hard.
2012-09-29T20:33:44 < emeb> karlp: you min Windows-only? Yeah I think so.
2012-09-29T20:33:45 < jpa-> i think cypress is aiming too low
2012-09-29T20:33:47 < karlp> also, not having an m4 doesn't mean it's the dark ages, lots of people don't need gigaflops of processing.
2012-09-29T20:34:06 < karlp> and hey, having an m3 is so much nicer than lots of radio socs for instance, that still use embedded 8051s
2012-09-29T20:34:14 < emeb> yep
2012-09-29T20:34:15 < jpa-> it makes a M3 class processor too expensive, but if it was in the absolute high end of MCU market it could be worth it
2012-09-29T20:35:48 -!- Posterdati [~quassel@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-09-29T20:35:50 < Laurenceb_> well the analogue looks like hype
2012-09-29T20:35:56 < Laurenceb_> compared to say F3
2012-09-29T20:36:06 < emeb> Maybe
2012-09-29T20:36:06 < Laurenceb_> but the digital blocks might be useful i guess
2012-09-29T20:36:21 < Laurenceb_> im reading psoc5 datasheet
2012-09-29T20:36:27 < emeb> The analog is pretty flexible - it does switched cap
2012-09-29T20:36:38 < emeb> so sigma/delta converters are possible.
2012-09-29T20:36:47 < emeb> or audio-rate filtering
2012-09-29T20:37:12 < Laurenceb_> wonder if you could do GPS with the UDB
2012-09-29T20:37:28 < emeb> doubt it.
2012-09-29T20:38:51 < Laurenceb_> yeah doesnt look like enough logic
2012-09-29T20:39:41 < emeb> only 8 product terms x 2
2012-09-29T20:40:10 < Laurenceb_> figure 8-2
2012-09-29T20:40:24 < Laurenceb_> take away the hype and its very similar to what F3 can do
2012-09-29T20:40:44 < Laurenceb_>  in PSoC® 5: CY8C55 Family Datasheet
2012-09-29T20:41:00 < emeb> Oh - I'm looking at 52 ds.
2012-09-29T20:41:11 < emeb> Figure 8-2 is analog interconnect.
2012-09-29T20:41:22 < Laurenceb_> theres just no filters on F3
2012-09-29T20:41:27 < Laurenceb_> - analogue filters
2012-09-29T20:41:32 < emeb> yup
2012-09-29T20:41:49 < jpa-> i have a project at work that could benefit from that kind of reconfigurable analog block
2012-09-29T20:41:51 < emeb> but adcs/dacs/comparators/pgas are similar
2012-09-29T20:41:59 < emeb> probably better on F3
2012-09-29T20:42:11 < jpa-> it needs to be able to support all kinds of sensors, while making the sensor assembly very simple
2012-09-29T20:42:16 < emeb> jpa-: yeah the reconfiguration is fun
2012-09-29T20:42:22 < emeb> you can do it on the fly.
2012-09-29T20:42:32 < emeb> it's all controlled by registers accessible from the CPU
2012-09-29T20:42:36 < jpa-> STM32 seems to have the best ADC & DAC stuff among the similar microcontrollers
2012-09-29T20:42:41 < Laurenceb_> yes
2012-09-29T20:42:43 < jpa-> LPC43xx also loses to it
2012-09-29T20:42:50 < Laurenceb_> certainly now with the F3 stuff
2012-09-29T20:43:02 < emeb> The STM32 12-bit ADC is the quietest of any MCU I've used.
2012-09-29T20:43:08 < emeb> (least lsb noise)
2012-09-29T20:43:09 < Laurenceb_> with an M4 core you can do loads of DSP processing
2012-09-29T20:43:10 < Laurenceb_> yeah
2012-09-29T20:43:16 < jpa-> even the most basic-of-basic F1 stuff is very good, and it's only better in the newer families
2012-09-29T20:43:31 < Laurenceb_> as its so fast you can oversample loads
2012-09-29T20:43:34 < jpa-> yeah.. we are now substituting analog filters with digital DSP on F4
2012-09-29T20:43:34 < emeb> I really like the 12-bit DACs on the F100 VL parts.
2012-09-29T20:43:38 < Laurenceb_> im getting about 20 effective bits
2012-09-29T20:43:40 < emeb> great for audio.
2012-09-29T20:43:55 < emeb> and you can't beat the price.
2012-09-29T20:43:56 < jpa-> (the previous version of the product had like 300+ parts for analog filters, lots of opamps and resistors..)
2012-09-29T20:44:09 < emeb> jpa-: that's a nice bit of system optimization
2012-09-29T20:44:32 < emeb> I used to use ADI ADuC parts for their 12-bit ADC/DAC
2012-09-29T20:44:36 < jpa-> also a big win for reconfigurability
2012-09-29T20:44:38 < Laurenceb_> im going to make an F3 pulse oximeter that connects directly to usb
2012-09-29T20:44:43 < Laurenceb_> with just F3 + vreg
2012-09-29T20:44:46 < emeb> put their CPU is old ARM 7 and they're very expensive.
2012-09-29T20:44:55 < jpa-> the old units were sometimes modded with jumper wires for specific applications
2012-09-29T20:45:28 < Laurenceb_> F3 can directly connect to a photodiode
2012-09-29T20:45:29 < emeb> Really looking forward to seeing what can be done w/ the F3 when it shows up.
2012-09-29T20:46:03 < emeb> Laurenceb_: seems ideal for data aqc apps.
2012-09-29T20:46:11 < jpa-> Laurenceb_: so what kind of input hardware is there on F3? internal pull-ups & opamps for analog, or something more complex?
2012-09-29T20:46:41 < Laurenceb_> jpa-: theres "analogue groups"
2012-09-29T20:46:53 < Laurenceb_> they are used for switched cap touch sensors
2012-09-29T20:47:06 < jpa-> the opamp stuff on the not-available L1 high end parts seems cool, i assume it's something similar but more advanced
2012-09-29T20:47:11 < Laurenceb_> configurable analogue connections
2012-09-29T20:47:53 < Laurenceb_> between the DAC, ADC, opamps and comparators
2012-09-29T20:48:29 < Laurenceb_> the touch input functionality uses them for charge redistribution stuff
2012-09-29T20:52:03 < emeb> The F373 parts look pretty good - lots of ADC/DAC options.
2012-09-29T20:53:18 < jpa-> cheaper than psoc, too
2012-09-29T20:53:25 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has joined ##stm32
2012-09-29T20:54:23 < emeb> no kidding - looks like ~$7.00 @ mouser. 1/2 price of PSoC
2012-09-29T20:55:04 -!- Posterdati [~quassel@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-29T20:55:07 < emeb> 303 parts are pretty good too - ~$6 - $9 in small qty
2012-09-29T20:55:19 < jpa-> but psoc is a good compromise if you don't know what you want to do :P
2012-09-29T20:55:24 < emeb> heh
2012-09-29T20:55:34 < jpa-> a bit pricey when you add the arduino tax
2012-09-29T20:55:56 < jpa-> $60 for a mcu board.. with nothing but processor + programmer
2012-09-29T20:56:04 < jpa-> + crystal
2012-09-29T20:56:41 < emeb> If ST can keep the analog specs good then the F3 should find a lot of sockets.
2012-09-29T20:58:24 < jpa-> F3 looks to be aimed at high-precision low-speed analog stuff
2012-09-29T21:00:26 < emeb> because of the SD ADCs?
2012-09-29T21:00:53 < jpa-> yeah
2012-09-29T21:01:06 < jpa-> and slower fast adc than F4
2012-09-29T21:03:15 < Laurenceb_> not so for the F30x
2012-09-29T21:03:23 < Laurenceb_> thats got really fast adc
2012-09-29T21:03:26 < Laurenceb_> - 12bit
2012-09-29T21:03:42 < jpa-> how fast?
2012-09-29T21:04:13 < jpa-> 37x seems to have only 1MSps, compared with 2.4MSps on F4
2012-09-29T21:04:31 < Laurenceb_> fast enough to saturate the dma
2012-09-29T21:05:28 < jpa-> maybe
2012-09-29T21:05:31 < Laurenceb_> http://www.flickr.com/photos/comradethompski/8020027773/in/photostream/
2012-09-29T21:06:36 < Laurenceb_> looks like my kind of place
2012-09-29T21:06:48 <+dekar> parts of a dog?
2012-09-29T21:07:09 < Posterdati> or selvatic pig?
2012-09-29T21:07:11 <+Steffann> Time to stop pasting those random links Laurenceb_?
2012-09-29T21:07:16 <+Steffann> *posting
2012-09-29T21:10:56 -!- tech2077 [~tech2077@107-1-61-118-ip-static.hfc.comcastbusiness.net] has quit [Ping timeout: 246 seconds]
2012-09-29T21:13:00 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has joined ##stm32
2012-09-29T21:15:54 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-29T21:15:55 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-29T21:15:55 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-29T21:17:19 < Laurenceb_> http://www.flickr.com/photos/comradethompski/sets/72157629240164790/with/6854799934/
2012-09-29T21:19:19 < zyp> http://www.kickstarter.com/projects/18182218/freesoc-and-freesoc-mini
2012-09-29T21:19:24 < zyp> so, how shitty is psoc 5?
2012-09-29T21:19:33 < zyp> I see that psoc 5 is cortex-m3 based
2012-09-29T21:20:34 < emeb> zyp is late to the conversation. :)
2012-09-29T21:20:42 <+Steffann> Yet another kickstarter project. Is it work to click on it?
2012-09-29T21:20:45 <+Steffann> *worth
2012-09-29T21:20:57 < zyp> emeb, did you already have it?
2012-09-29T21:21:17 < emeb> we were talking about that about an hour ago
2012-09-29T21:21:19 < zyp> I'm sorry I missed it, I'm not home this weekend so I'm not sitting at a computer all the time :p
2012-09-29T21:21:36 < zyp> care to summarize?
2012-09-29T21:21:37 < Laurenceb_> like the loser you are
2012-09-29T21:21:54 < emeb> Looks useful. F3 is prolly better
2012-09-29T21:22:20 < emeb> zyp: have you used PSoC?
2012-09-29T21:22:28 < zyp> emeb, no, that's why I'm asking
2012-09-29T21:22:39 < emeb> The analog & digital resources are sort of FPGA-like
2012-09-29T21:22:47 < zyp> Laurenceb_, I'll rather be the loser I am than the loser you are
2012-09-29T21:22:48 < emeb> not as fine-grained.
2012-09-29T21:23:58 < emeb> and the configuration tools for defining the analog & digital resources are likely Win only.
2012-09-29T21:24:13 < Laurenceb_> loser i am
2012-09-29T21:24:22 < emeb> yoda talk?
2012-09-29T21:26:54 -!- Posterdati_ [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-29T21:27:13 -!- Mazingaro [~Tetsuja@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 246 seconds]
2012-09-29T21:27:30 -!- Posterdati [~quassel@host110-237-dynamic.20-87-r.retail.telecomitalia.it] has quit [Ping timeout: 252 seconds]
2012-09-29T21:32:34 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-09-29T21:34:34 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-29T21:34:49 <+Steffann> Doh, I'm a loser too
2012-09-29T21:36:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-29T21:40:11 < Laurenceb_> http://www.flickr.com/photos/comradethompski/5814794582/in/photostream/lightbox/
2012-09-29T21:40:12 -!- Mazingaro [~Tetsuja@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-29T21:40:18 < Laurenceb_> aha my perfect house
2012-09-29T21:40:57 <+Steffann> I don't need 3 AA at all.. i can just bypass one AA in the case ( of course )  :(
2012-09-29T21:41:02 <+Steffann> Why i didn't think of that before?!
2012-09-29T21:41:45 <+Steffann> As in: just solder a wire to another metal part of the case
2012-09-29T21:43:05 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 255 seconds]
2012-09-29T21:46:54 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has joined ##stm32
2012-09-29T21:46:55 -!- TeknoJuce [~TeknoJuce@bas1-london15-2925200962.dsl.bell.ca] has quit [Changing host]
2012-09-29T21:46:55 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-29T21:48:59 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-09-29T21:48:59 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-09-29T21:49:00 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-29T21:50:22 < emeb> Laurenceb_ isn't that some Soviet-style remnant high up in the mountains somewhere?
2012-09-29T21:50:32 < emeb> think I saw that on BoingBoing a while back.
2012-09-29T21:50:34 < Laurenceb_> yeah
2012-09-29T21:52:04 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 246 seconds]
2012-09-29T21:55:49 < emeb> http://humanplanet.com/timothyallen/2012/02/buzludzha-buzludja-bulgaria/
2012-09-29T22:03:22 <+izua> https://secure.flickr.com/photos/comradethompski/5814786826/in/photostream/
2012-09-29T22:03:41 <+izua> Laurenceb_: what is tis
2012-09-29T22:07:04 <+izua> nvm, found it https://en.wikipedia.org/wiki/Buzludzha
2012-09-29T22:16:09 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-29T22:31:45 -!- barthess [~barthess@178.154.23.252] has joined ##stm32
2012-09-29T22:45:30 -!- barthess [~barthess@178.154.23.252] has quit [Remote host closed the connection]
2012-09-29T22:45:48 -!- barthess [~barthess@178.154.23.252] has joined ##stm32
2012-09-29T22:47:26 -!- barthess [~barthess@178.154.23.252] has quit [Client Quit]
2012-09-29T23:40:34 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
--- Day changed Sun Sep 30 2012
2012-09-30T00:04:12 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has joined ##stm32
2012-09-30T00:40:58 -!- barthess [~barthess@178.154.23.252] has joined ##stm32
2012-09-30T00:53:09 -!- Posterdati_ [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-09-30T00:54:00 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-09-30T00:55:56 < dongs> lol @ dongs tools only
2012-09-30T00:56:36 <+Steffann> Isn't it time to sleep mr dongs?
2012-09-30T00:56:57 < dongs> sorta
2012-09-30T00:57:10 < karlp> hoped you'd like it ;)
2012-09-30T00:57:13 < dongs> im going back to snooze
2012-09-30T00:57:20 < emeb> Bwahahahah..... reverb works!
2012-09-30T00:57:43 < dongs> psoc is prettty fail, windows tools or not.
2012-09-30T00:58:42 < dongs> emeb, and after a few hours of trolling last night the suggestion was to go back to pnp transistor for high side lol
2012-09-30T00:59:13 < emeb> can't find any good pfets?
2012-09-30T01:00:07 < dongs> well, or pfet and use a 74hc as driver.
2012-09-30T01:00:43 < emeb> that's getting complicated
2012-09-30T01:00:43 < dongs> to avoid havinf to deal with oc and pull up and 5v tolerant gpio.
2012-09-30T01:00:58 < emeb> yeah - that's a pita
2012-09-30T01:00:59 < dongs> not rly?
2012-09-30T01:01:13 < dongs> 3:8 decoder
2012-09-30T01:01:26 < dongs> and done
2012-09-30T01:01:44 < emeb> or a 595 and clock a bit thru
2012-09-30T01:01:46 < dongs> power it off 5v
2012-09-30T01:02:12 < dongs> naah, THAT is complicated.
2012-09-30T01:02:28 < emeb> whut? 2 gpio bits?
2012-09-30T01:02:38 < emeb> clk & data. done.
2012-09-30T01:02:41 < dongs> you're not
2012-09-30T01:02:47 < dongs> err
2012-09-30T01:03:28 < dongs> you're only saving one io and over complicating stuff
2012-09-30T01:03:42 < dongs> by doing serial
2012-09-30T01:04:06 < dongs> but yea, anything wrong with that way?
2012-09-30T01:05:33 < emeb> nah - should work fine. an hc138 is smaller pkg too.
2012-09-30T01:06:46 < dongs> then I can just pick any random cheap dual pfet and call it done. or maybe even singles if there's space.
2012-09-30T01:07:17 < emeb> yeah - with 0V/5V output swing the PFET will turn off solid.
2012-09-30T01:07:48 -!- barthess [~barthess@178.154.23.252] has quit [Quit: Leaving.]
2012-09-30T01:07:59 < emeb> no messing around with whacky Vgs(th)
2012-09-30T01:08:29 < dongs> ok, 3:8 it is. Back to sleep for me.
2012-09-30T01:13:15 < zyp> dongs, what makes psoc fail?
2012-09-30T01:14:12 < zyp> I mean, newest gen is apparently cortex-m3
2012-09-30T01:15:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-09-30T01:24:02 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-09-30T01:24:10 <+dekar> I feel like bees are listening in on my phone calls O.o
2012-09-30T01:24:18 <+dekar> http://www.springerlink.com/content/bx23551862212177/fulltext.pdf?MUD=MP
2012-09-30T01:24:23 <+izua> dekar: it's your government
2012-09-30T01:25:06 <+dekar> so you mean bees were invented by the german government to spy for them? :D
2012-09-30T01:25:30 < Thorn> http://www.edri.org/files/cleanIT_sept2012.pdf
2012-09-30T01:26:46 <+izua> dekar: http://web.ncsu.edu/abstract/science/wms-cockroach-steering/
2012-09-30T01:28:14 <+dekar> izua, I totally want one of those :)
2012-09-30T01:28:28 <+izua> i don't think it's very hard to make one at home
2012-09-30T01:28:37 <+izua> all you need is a cockroach and an arduino :P
2012-09-30T01:31:15 <+dekar> I wonder whether one is enough to carry an arduino
2012-09-30T01:31:38 <+dekar> especially once I've added an NFC shield :D
2012-09-30T01:32:05 <+izua> cockroaches are pretty big though
2012-09-30T01:32:21 <+izua> you can grow your own and selectively breed the bigger ones :P
2012-09-30T01:32:30 <+dekar> I guess I could just add several roaches for more carrying capacity
2012-09-30T01:32:37 < BrainDamage> trailing cart
2012-09-30T01:32:45 < BrainDamage> or in this case, trolling cart
2012-09-30T01:33:38 <+dekar> I wonder whether you could also stick some electrodes into their abdomen and use their stomach acid as a battery
2012-09-30T01:35:34 <+izua> without spilling it?
2012-09-30T01:35:39 < BrainDamage> the surface area will be tiny, it'll had to be an highly efficient circuit
2012-09-30T01:35:51 < BrainDamage> because the current capability will be ridicolous
2012-09-30T01:36:03 < karlp> you'd probably do better with a piezo thing
2012-09-30T01:36:09 < karlp> let their movement power you
2012-09-30T01:36:24 < BrainDamage> or a dynamo on the trolling cart that holds the arduino
2012-09-30T01:37:46 <+izua> but then you'd need roach food too
2012-09-30T01:38:05 <+dekar> awesome, I'll make a regenerative-breaking-arduino-shield :P
2012-09-30T01:38:56 < Laurenceb_> http://i826.photobucket.com/albums/zz187/youngexplorers69/peters/StPeters2008.jpg
2012-09-30T01:39:07 < Laurenceb_> that grammar error looks fatal
2012-09-30T01:39:32 < karlp> I'm more interested in what's normally in that fridge if not bodies
2012-09-30T01:41:19 < Laurenceb_> undertakers sandwiches
2012-09-30T01:52:08 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-09-30T01:59:25 -!- BjoernC [~BjoernC@dyndsl-178-142-052-101.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-30T02:40:32 < dongs> zyp: havent played with new ones. last time i touched it back when it was something like 8051
2012-09-30T02:40:37 < dongs> and the GUI was truly buggy
2012-09-30T02:41:13 < dongs> so much for sleep
2012-09-30T02:41:31 < dongs> 8.10: "oh today youre supposed to go cut grass for free for some bullshit community effort to clean shit up"
2012-09-30T02:41:36 < dongs> "no, fuck you, im trying to sleep"
2012-09-30T02:45:12 < emeb> poor dongs
2012-09-30T03:00:37 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-09-30T03:06:40 < dongs> better switch to sam4l
2012-09-30T03:07:43 < qyx_> trolling or no?
2012-09-30T03:08:13 < dongs> supposed low power
2012-09-30T03:08:33 < qyx_> i used to hade sam7s
2012-09-30T03:08:34 < qyx_> uhm
2012-09-30T03:08:37 < qyx_> *hate
2012-09-30T03:09:07 < upgrdman> if i wait more than a few seconds after plugging in my usb cable, the flash utility fails. any ideas why? im using the stlinkv2 on the f4disco board. both texane/stlink and openocd fail after more than a few seconds
2012-09-30T03:09:10 < dongs> ya, horrible shit
2012-09-30T03:09:51 < HTT-Bird> heheh
2012-09-30T03:10:14 < HTT-Bird> not sure if you really need to do that
2012-09-30T03:10:34 < HTT-Bird> you can get ARM parts these days that are quite competitive with say a MSP430 in terms of standby power
2012-09-30T03:10:52 < dongs> maybe it's better now with atmel framework.
2012-09-30T03:23:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-30T03:27:38 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-30T03:43:12 < dongs> finally got my 100>125V stepup transformer
2012-09-30T03:43:19 < dongs> now i can be awesome again.
2012-09-30T03:43:39 < dongs> all these cool american irons that barely warm up off shit jap power can be useful
2012-09-30T03:46:47 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-30T03:48:53 -!- Laurenceb_ [~Laurence@host86-136-118-129.range86-136.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-09-30T04:09:03 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-09-30T04:13:51 < zyp> you can come here and plug them into norwegian power :p
2012-09-30T04:16:41 < dongs> haha
2012-09-30T04:16:44 < dongs> what does norgay use
2012-09-30T04:16:45 < dongs> 220V?
2012-09-30T04:17:07 < zyp> 230
2012-09-30T04:17:10 < dongs> nice
2012-09-30T04:18:08 < zyp> yeah, except when I smoked some shit I bought in japan in 2008
2012-09-30T04:18:27 < zyp> took it to norway and plugged it in, and it couldn't handle all the awesomeness
2012-09-30T04:30:07 < Bird|lappy> heh
2012-09-30T04:37:47 -!- Netsplit *.net <-> *.split quits: ReggieUK
2012-09-30T04:38:06 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: No buffer space available]
2012-09-30T04:38:54 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-09-30T04:38:55 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-09-30T04:38:55 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-09-30T04:40:10 < dongs> lul
2012-09-30T04:40:54 < dongs> hello kitty vibrator couldnt deal wiht it
2012-09-30T04:42:22 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-30T04:50:31 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-30T04:53:00 -!- Netsplit *.net <-> *.split quits: ReggieUK, emeb_mac
2012-09-30T04:58:07 -!- Netsplit over, joins: emeb_mac, ReggieUK
2012-09-30T04:58:26 < dongs> finally got stencils + time to assemble the fuckers http://i.imgur.com/VoYWM.jpg
2012-09-30T04:58:47 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-30T05:00:20 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-30T05:02:32 < karlp> fuck dude, it only just aired, spoilers much?!
2012-09-30T05:02:46 < karlp> it hasn't even finished downloading yet!
2012-09-30T05:03:55 <+dekar> I guess your source is too slow then :P
2012-09-30T05:04:11 < dongs> dr.unfunny
2012-09-30T05:04:35 < karlp> well, I only started it a few minutes ago, it's almost finished
2012-09-30T05:07:05 < Thorn> CP2102?
2012-09-30T05:07:08 < Thorn> wtf
2012-09-30T05:09:39 < qyx_> usb to uart?
2012-09-30T05:10:35 < Thorn> minidongduino?
2012-09-30T05:14:24 < cjbaird> hehe-- General Grievous... as a Japanese Schoolgirl... with braces :) .. http://www.pixiv.net/member_illust.php?mode=medium&illust_id=26038315
2012-09-30T05:16:01 < Thorn> how do I get call stack before a fault from within a handler? (gdb)
2012-09-30T05:16:23 < upgrdman> bt
2012-09-30T05:16:28 < upgrdman> backtrace
2012-09-30T05:16:30 < upgrdman> iirc
2012-09-30T05:17:09 < Thorn> not in my case, maybe because I'm using a rtos
2012-09-30T05:17:17 < dongs> wtf is pixiv
2012-09-30T05:17:40 < upgrdman> if i wait more than a few seconds after plugging in my usb cable, the flash utility fails. any ideas why? im using the stlinkv2 on the f4disco board. both texane/stlink and openocd fail after more than a few seconds
2012-09-30T05:19:37 < dongs> even if you clean out your firmware?
2012-09-30T05:20:27 < Thorn> after plugging the cable into st-link? or into your device?
2012-09-30T05:20:56 < Thorn> also what do they print? openocd has a debug switch (-d 2 / -d 3)
2012-09-30T05:25:53 < upgrdman> only happens with my current firmware
2012-09-30T05:26:03 < upgrdman> and shit, now i cant find my dev board
2012-09-30T05:26:29 < upgrdman> im using the stlinkv2 to program the f4 on the discovery board
2012-09-30T05:31:22 < dongs> does your firmware disable jtag after boot?
2012-09-30T05:33:55 < upgrdman> no
2012-09-30T05:34:02 < upgrdman> and it uses swd
2012-09-30T05:35:14 < upgrdman> found my dev boardf
2012-09-30T05:35:19 < upgrdman> heres the openocd error
2012-09-30T05:35:54 < upgrdman> at file "embedded:startup.tcl", line 58
2012-09-30T05:35:55 < upgrdman> in procedure 'init' called at file "stm32f4-openocd.cfg", line 5
2012-09-30T05:35:55 < upgrdman> make: *** [install] Error 1
2012-09-30T05:36:21 < upgrdman> will try with d 2
2012-09-30T05:37:17 < upgrdman> in procedure 'init' called at file "stm32f4-openocd.cfg", line 5
2012-09-30T05:37:17 < upgrdman> Debug: 184 34 stlink_interface.c:99 stlink_interface_quit(): stlink_interface_quit
2012-09-30T05:38:23 < dongs> quitter interface
2012-09-30T05:38:25 < dongs> !!
2012-09-30T05:38:33 < upgrdman> ?
2012-09-30T05:44:13 < dongs> emeb_mac: !! http://www.ti.com/product/tlc59213
2012-09-30T05:44:59 < emeb_mac> looking...
2012-09-30T05:46:21 < emeb_mac> blech
2012-09-30T05:46:28 < dongs> its similar to mic shit iguess.
2012-09-30T05:46:31 < emeb_mac> Vce sat @ 350mA > 2V
2012-09-30T05:46:43 < dongs> http://www.dahlsys.com/hardware/led_matrix/index.html according to this guy
2012-09-30T05:46:44 < emeb_mac> worse
2012-09-30T05:46:49 < dongs> I wanted to run the design off of a single 5V supply. The TLC59213 accepts input levels down to 2V, even when running off of a 5V supply. So it can source around 4.4V on each of its output lines based on 3.3V signals from the STM32, which leaves a comfortable margin for the voltage drop over the LEDs and over the low side driver.
2012-09-30T05:47:04 < dongs> or is he miscalculating somethign
2012-09-30T05:47:33 < emeb_mac> Yeah - he completely missed the Vce sat vs Iout
2012-09-30T05:47:47 < dongs> hehe mkay
2012-09-30T05:48:14 < emeb_mac> also, it's got a 10% duty cycle limit at the high current rating
2012-09-30T05:48:32 < dongs> was just closing some bro-wser windows and noticed i still had that open and not read since last time
2012-09-30T05:48:41 < emeb_mac> not good for a x8 muxed system (duty = 12.5%)
2012-09-30T05:49:16 < dongs> This means that there is little, if any, benefit to using MOSFET / DMOS when driving them with 3.3V, such as that provided by the GPIO pins on the STM32 – which is probably why I was unable to find good options for such drivers for this design in the first place.
2012-09-30T05:49:20 < emeb_mac> PFET + HC138 seems like best be so far.
2012-09-30T05:49:22 < dongs> at least he came to same conclusion for individual fets
2012-09-30T05:49:22 < dongs> yeah
2012-09-30T05:50:07 < emeb_mac> hmmm - that a quote from this guy's design notes?
2012-09-30T05:50:11 < dongs> ya
2012-09-30T05:50:30 < dongs> Not finding any good options for arrays, I researched the option of using 16 discrete MOSFETs. I found that even for “logic level” MOSFETs, 3.3V switching voltage is too low to bring them into their saturated regions (they don’t turn on fully).
2012-09-30T05:50:58 < dongs> so the same issue basically that gate needs to be at 5V
2012-09-30T05:51:07 < dongs> tarduino guys done care cuz they run at 5V.
2012-09-30T05:53:31 < emeb_mac> yup
2012-09-30T05:54:00 < emeb_mac> but using HC logic to convert 3.3V -> 5V works fine.
2012-09-30T05:55:46 < dongs> sounds good. so ill start drawing wiht that in mind and that MY9221 or wahtever for drivers
2012-09-30T05:56:06 < dongs> gonna grab 100 of them once holidays are over, i need to order some other shit anyway
2012-09-30T05:58:13 < emeb_mac> cool
2012-09-30T05:58:15 < emeb_mac> I've got some of those matrix displays coming from ebay.
2012-09-30T05:58:38 < dongs> about the only remaining thing im still unsure on waht to do is block to block communication
2012-09-30T05:59:32 < dongs> thinking maybe just breakout usart1/2 or maybe 2/3, so can remap them into i2c for ghetto stuff, or maybe.. somethin dunno.
2012-09-30T05:59:53 < dongs> or just string rx/tx across the board and tap it
2012-09-30T06:00:26 < emeb_mac> might be able to ghetto a spi clockthru w/ software.
2012-09-30T06:00:52 < emeb_mac> that could go pretty fast - if it works.
2012-09-30T06:01:28 < dongs> so you mean have like CLK+data_in on one side, going to sck+miso? or whatever and then buffer sck out and forward as bufferdsck+mosi?
2012-09-30T06:01:41 < dongs> should i bother buffering clk for short chains of shit?
2012-09-30T06:01:47 < dongs> wiht like some 74xx buffer thingy
2012-09-30T06:03:21 < emeb_mac> depends on how long you plan to chain.
2012-09-30T06:04:30 < emeb_mac> but one of those little single-gate buffers in an SOT-23-5 would probably work fine.
2012-09-30T06:04:39 < emeb_mac> if it's less than 5 or 6 modules per chain then you could likely get away w/o.
2012-09-30T06:05:23 < emeb_mac> not like those driver chips that have clock duty cycle correction - they're designed to work in really long chains (hundreds)
2012-09-30T06:05:53 < dongs> right
2012-09-30T06:06:10 < dongs> i do have some sot23-5 buffers because i need them to bump 3V io to 5V for WS2811 stuff im doing for other shit
2012-09-30T06:06:23 < emeb_mac> yeah - that would work fine.
2012-09-30T06:06:31 < dongs> i think they add something like 5ns delay to clk dunno if that matters
2012-09-30T06:06:58 < dongs> need a latch, too, right? NSS sorta thing
2012-09-30T06:07:26 < emeb_mac> don't think so - what for?
2012-09-30T06:07:48 < dongs> i dunno. each crap just displays data received right? i guess maybe dont need.
2012-09-30T06:07:59 < dongs> but like led drivers usually have clk/data/latch
2012-09-30T06:08:27 < emeb_mac> yeah - a signal broadcast to all modules in the chain telling them to latch the data they now have.
2012-09-30T06:08:48 < emeb_mac> so each module gets spi in -> FIFO -> spi out
2012-09-30T06:09:05 < emeb_mac> when the NSS signal pops they use the data in their FIFO to set the display.
2012-09-30T06:10:27 < emeb_mac> assume 8-bit/color x 3 colors X 64 pixels/matrix = 192 bytes in the FIFO
2012-09-30T06:10:44 < dongs> that driver can do 16bit for some 48bit color hax
2012-09-30T06:10:59 < emeb_mac> OK - so 384 bytes.
2012-09-30T06:11:39 < dongs> so youre saying keep the latch right
2012-09-30T06:12:41 < emeb_mac> yeah
2012-09-30T06:13:01 < emeb_mac> run it to all modules in parallel
2012-09-30T06:13:29 < emeb_mac> running spi clock @ 10MHz x 384 x 8 ~307us to load one module w/ all data
2012-09-30T06:13:54 < emeb_mac> assuming the sw can keep up.
2012-09-30T06:15:52 < dongs> thtas enough to refresh a fuckton at 30hz
2012-09-30T06:16:35 < emeb_mac> yeah - and remember this is just to update the data. Each STM stores locally & refreshes its own matrix.
2012-09-30T06:16:54 < dongs> right
2012-09-30T06:16:59 < dongs> you can do 100 at 30hz
2012-09-30T06:17:06 < emeb_mac> so you don't have to have the SPI running continuously - just when the display updates.
2012-09-30T06:17:13 < emeb_mac> yep
2012-09-30T06:17:32 < emeb_mac> 80x80
2012-09-30T06:17:40 < dongs> the clock will probably be 10 seconds delayed by the time it gets to the 100th one.
2012-09-30T06:17:47 < emeb_mac> heh
2012-09-30T06:17:53 < dongs> lemme see which buffer i have for it
2012-09-30T06:18:09 < dongs> 568-7728-1-ND
2012-09-30T06:18:22 < dongs> http://www.nxp.com/documents/data_sheet/74HC_HCT1G125.pdf
2012-09-30T06:18:41 < dongs> oh what the crap
2012-09-30T06:18:58 < dongs> fucktastic rise/fall time on that shit
2012-09-30T06:19:10 < emeb_mac> that should work
2012-09-30T06:19:20 < dongs> o wait im looking at hc part
2012-09-30T06:19:24 < dongs> i got hct
2012-09-30T06:19:30 < emeb_mac> prop delay about the same as clk-data on the STM SPI output.
2012-09-30T06:19:36 < dongs> page 8
2012-09-30T06:19:38 < dongs> that looks better.
2012-09-30T06:19:48 < dongs> er page 9
2012-09-30T06:19:58 < dongs> propagation delay 11ns
2012-09-30T06:21:09 < emeb_mac> need to dblchk the spi prop delay on the STM
2012-09-30T06:21:48 < dongs> hmm wait... if youre free-running the clock, how is it gonna tell when to clock stuff in over spi?
2012-09-30T06:22:17 < dongs> latch has to go through the chip
2012-09-30T06:22:26 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-09-30T06:22:34 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-09-30T06:24:29 < emeb_mac> haven't thought it all the way through but here's what I'm thinking:
2012-09-30T06:24:57 < emeb_mac> every STM in the chain has an internal queue 384 bytes long. SPI data goes in one end, out the other.
2012-09-30T06:25:29 < emeb_mac> when NSS drops and SPI clocks start coming each STM loads it's own SPI tx reg with data from queue output.
2012-09-30T06:26:03 < dongs> clock will come immediately to all of them (and so is nss/latch, if its directly forwarded)
2012-09-30T06:26:09 < emeb_mac> when 8 clocks have come then they write that data into queue.
2012-09-30T06:26:13 < dongs> at this point the last in chain chip is getting nothing at all
2012-09-30T06:26:25 < dongs> just clock and nulls.
2012-09-30T06:26:32 < emeb_mac> yes - all clocks and nss at once, last chip in chain is getting 0 data
2012-09-30T06:27:01 < emeb_mac> finally all chips have their queues full after 384 x chips in chain
2012-09-30T06:27:06 < dongs> then you write through all the data, once 1st block is full, it just forwards the data to next one, etc
2012-09-30T06:27:07 < emeb_mac> then raise NSS
2012-09-30T06:27:11 < emeb_mac> yep
2012-09-30T06:27:13 < dongs> ok, i see
2012-09-30T06:27:41 < dongs> i think thats backwards of what led drivers do
2012-09-30T06:28:52 < emeb_mac> Oh - do they load front first and then forward when each is full?
2012-09-30T06:29:03 < dongs> yeah
2012-09-30T06:29:09 < dongs> latch goes through teh chip
2012-09-30T06:29:11 < emeb_mac> dropping nss along the way
2012-09-30T06:29:12 < dongs> right
2012-09-30T06:29:20 < emeb_mac> well, you could do that too
2012-09-30T06:29:28 < dongs> not saying this is better or worse, i dont know
2012-09-30T06:29:33 < emeb_mac> just need to run the NSS output from GPIO
2012-09-30T06:29:33 < dongs> but for this case latch would have to be going through the chip
2012-09-30T06:30:20 < emeb_mac> also means that chips are forwarding data immediately when they're passing thru
2012-09-30T06:31:01 < dongs> thats the case in either situation
2012-09-30T06:31:43 < dongs> hmm i think ill fwd it through the chip
2012-09-30T06:31:57 < emeb_mac> should work
2012-09-30T06:32:01 < dongs> just in case, it can be easily enough handled either way
2012-09-30T06:32:13 < emeb_mac> jumpers
2012-09-30T06:32:21 < dongs> oh hm
2012-09-30T06:32:26 < dongs> eyah not a bad idea
2012-09-30T06:32:40 < emeb_mac> or do it in software.
2012-09-30T06:33:00 < dongs> well thats what i meant for "can be easily enough to do either way
2012-09-30T06:33:10 < emeb_mac> yup
2012-09-30T06:34:31 < emeb_mac> so, since each multiplexed array can draw up to 480ma a x100 chain could draw 48A. Get a big supply!
2012-09-30T06:34:36 < dongs> haha
2012-09-30T06:34:53 < dongs> ya led power is pretty nutty
2012-09-30T06:35:07 < dongs> my pal is doing some led bar shits, they're using those big industrial DC supplies
2012-09-30T06:35:12 < dongs> for liek 24V/XXA or someshit
2012-09-30T06:35:19 < emeb_mac> no doubt
2012-09-30T06:35:33 < emeb_mac> I wonder what those LED billboards & jumbotrons draw.
2012-09-30T06:36:14 < dongs> probably a lot
2012-09-30T06:36:28 < emeb_mac> http://www.adafruit.com/products/420
2012-09-30T06:36:48 < emeb_mac> 2A, 5V
2012-09-30T06:37:09 < dongs> that looks like the same shit ebay hs
2012-09-30T06:37:43 < dongs> http://www.ebay.com/itm/310381741527 i.e. shit like this
2012-09-30T06:37:57 < dongs> 5V/35W
2012-09-30T06:38:17 < dongs> that aution doesnt show backside
2012-09-30T06:38:37 < dongs> http://i.ebayimg.com/t/P4-PH4-64-64-Dot-Matrix-RGB-Full-Color-LED-Module-Board-For-Video-Wall-/00/s/ODAwWDgwMA==/$(KGrHqNHJCEE92!KG8IqBPjRSnLkBw~~60_3.JPG
2012-09-30T06:38:41 < dongs> this one d oes
2012-09-30T06:39:29 < emeb_mac> oh - that's what that was.
2012-09-30T06:39:34 < dongs> http://www.ebay.com/itm/190644060517 and heres the shit aidsfruit is reselling
2012-09-30T06:39:41 < dongs> with a nice markup
2012-09-30T06:39:49 < emeb_mac> save a few $$
2012-09-30T06:40:07 < emeb_mac> Oh - it's strawberries.
2012-09-30T06:40:30 < dongs> Keep in mind that these displays are designed to be driven by FPGAs or other high speed processors: they do not have built in PWM control of any kind.
2012-09-30T06:40:33 < dongs> lawl
2012-09-30T06:40:51 < dongs> nothanks even if density is high
2012-09-30T06:41:12 < dongs> once i get the china8x8 shit working ill make a led carrier board for my 5050leds
2012-09-30T06:41:19 < dongs> with same pinout or w ahtever
2012-09-30T06:41:56 < emeb_mac> oh - PCB carrier for the discretes w/ same pinout as the matrix module? Good idea.
2012-09-30T06:42:22 < emeb_mac> Get a CM to churn out a batch for you.
2012-09-30T06:42:44 < dongs> the ledpal can do 8x8 matrixes on white pcb or osmeshit
2012-09-30T06:42:58 < dongs> can put some smd pin headers on the back
2012-09-30T06:43:02 < dongs> for the connections.
2012-09-30T06:43:13 < dongs> and plug the driver board to it just like behind chinamodule
2012-09-30T06:43:51 < emeb_mac> yep
2012-09-30T06:44:47 < dongs> <aidsfruit> At this time we do not have wiring documentation for the MEGA, but we will update our tutorial in the next week to add MEGA
2012-09-30T06:44:50 < dongs> "next week"
2012-09-30T06:45:01 < dongs> i wonder when the product went on sale
2012-09-30T06:47:38 < emeb_mac> they're like "whatevs".
2012-09-30T06:48:04 < emeb_mac> I've bought stuff from them - they ship fast and prices aren't unreasonable. Fine w/ me.
2012-09-30T06:49:14 < dongs> just dont like the attitude
2012-09-30T06:49:31 < dongs> bitch thinks she's a pro make:r
2012-09-30T06:49:50 < dongs> right up there with jerry "lier" ellisworth etc.
2012-09-30T06:50:06 < dongs> </rant>
2012-09-30T06:50:22 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 246 seconds]
2012-09-30T06:50:56 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-09-30T06:52:44 < cjbaird> jeri "the common factor in all my failed businesses... is Men!" ellisworth..
2012-09-30T06:53:05 < dongs> lo
2012-09-30T06:53:22 < dongs> tell'em like it is cjbaird
2012-09-30T06:54:01 < dongs> hm
2012-09-30T06:54:16 < dongs> emeb_mac: how about using MCO to buffer clock
2012-09-30T06:54:45 < dongs> actually never mind maybe ill hookup mco to the clock input on the led drivers
2012-09-30T06:54:53 < dongs> since it seems to have some options to run off intrc or external clock
2012-09-30T06:57:40 < emeb_mac> yeah - mco wouldn't be any good for the spi clock. Probably good for the PWM driver tho.
2012-09-30T06:58:00 < dongs> driver doesnt "need" a clock, but some options change when its available
2012-09-30T06:58:09 < dongs> no harm done if its unused so ill connect that.
2012-09-30T06:59:28 < emeb_mac> options - u haz dem.
2012-09-30T06:59:43 < dongs> fuck... google changed something and undirect script no longer kills the bullshit google/url? redirection shit in search results
2012-09-30T06:59:55 < dongs> i wish i saved whatever the fuck I used last time since that worked very well
2012-09-30T07:01:03 < dongs> Since 2012-09-04, it seems that
2012-09-30T07:01:04 < dongs> ugh
2012-09-30T07:01:08 < dongs> no wonder it broke in september
2012-09-30T07:02:55 < dongs> found it.http://userscripts.org/scripts/show/121923
2012-09-30T07:02:56 < emeb_mac> meh - jwz's "youtubedown" had to be revved the other day too.
2012-09-30T07:03:07 < dongs> i use yousabletubefix
2012-09-30T07:03:28 < dongs> haha wtf perl script
2012-09-30T07:04:37 < cjbaird> youtube-dl python eliteness 4 lyfe (which needed a bump too..)
2012-09-30T07:37:13 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Remote host closed the connection]
2012-09-30T07:37:51 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-30T07:43:49 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-30T08:00:16 < dongs> http://i.imgur.com/GEIUA.jpg gotta love the silk resolution of  my pcb fab
2012-09-30T08:01:56 < Bird|lappy> dongs, nice
2012-09-30T08:02:00 < qyx_> whats that thing near bmp085?
2012-09-30T08:03:29 < dongs> gyroacc
2012-09-30T08:04:24 < dongs> mpu6050 or watev
2012-09-30T08:14:25 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-09-30T08:30:14 < dongs> emeb_mac: lol http://www.ti.com/product/tpic6595
2012-09-30T08:30:22 < dongs> retweet from #arduno
2012-09-30T08:31:05 < dongs> thats one hardcore 595
2012-09-30T08:31:33 < emeb_mac> relay driver or something.
2012-09-30T08:32:49 < dongs> yea
2012-09-30T08:33:45 < Bird|lappy> yeah.
2012-09-30T08:34:01 < Bird|lappy> it's a 595 and a ULN2803A rolled into one package
2012-09-30T09:02:28 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-09-30T09:02:35 < dongs>  < slackguru> I asked if anyone knew of a 64bit processor being integrated on an arduino board yet.
2012-09-30T09:04:04 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-30T09:20:07 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-09-30T09:39:27 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-30T09:39:28 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-30T09:39:28 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-30T09:58:14 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-30T10:01:25 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-30T10:31:28 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-09-30T10:59:33 < dongs> attn cjbaird
2012-09-30T10:59:34 < dongs> -!- AeriCELL [~a@CPE001346b84eab-CM001a6682e3cc.cpe.net.cable.rogers.com] has joined #electronics
2012-09-30T10:59:37 < dongs> <AeriCELL> On cell.
2012-09-30T10:59:57 < dongs> <AeriCELL> http://theflog.geekandsundry.com/episodes/felicia-at-dragoncon/
2012-09-30T10:59:57 < dongs> <AeriCELL> Oppss...
2012-09-30T10:59:57 < dongs> lool.
2012-09-30T11:02:03 < cjbaird> "Geek" branding.. Rage level: rising.
2012-09-30T11:03:31 < cjbaird> Felicia as in attentionwhore, not Felicia the streetfighter catgirl: rage++
2012-09-30T11:04:54 < cjbaird> You interrupted the latest episode of Space Bros for that? :/
2012-09-30T11:05:34 < dongs> :(
2012-09-30T11:07:33 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 272 seconds]
2012-09-30T11:08:42 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-09-30T11:14:19 < cjbaird> My college-era boasting for the day: http://boards.4chan.org/a/res/72284369#p72284655
2012-09-30T11:14:36 < qyx_> felicia \o/
2012-09-30T11:15:18 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has joined ##stm32
2012-09-30T11:15:18 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has quit [Changing host]
2012-09-30T11:15:18 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32
2012-09-30T11:16:01 < dongs> ughhhhhhhhhhhhh
2012-09-30T11:16:12 < dongs> seeedstudio datashet for rgb matrix: 15x2 pins
2012-09-30T11:16:19 < dongs> schematic: 32 pins
2012-09-30T11:25:21 -!- barthess [~barthess@178.154.41.18] has joined ##stm32
2012-09-30T11:32:38 < dongs> i mean obviosuly makes sense for it oth ave 32 pins
2012-09-30T11:32:42 < dongs> but the fucking drawing! is wrong
2012-09-30T11:32:52 < dongs> for pin count + location.
2012-09-30T11:33:12 < dongs> fucking garbage.
2012-09-30T12:04:43 -!- barthess [~barthess@178.154.41.18] has quit [Quit: Leaving.]
2012-09-30T12:18:11 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer]
2012-09-30T12:47:56 < dongs> does turning off JTAG also disable traceswo?
2012-09-30T12:48:01 < dongs> (leaving SWD part on)
2012-09-30T12:48:06 < dongs> i guess it does
2012-09-30T12:51:25 < dongs> since thats on jtdo
2012-09-30T13:36:21 < zyp> on F1? probably
2012-09-30T13:36:59 < zyp> on F4 you can just set jtdi to a normal GPIO and keep jtdo as swo
2012-09-30T13:45:38 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-30T13:46:42 < dongs> cjbaird: i dont get it
2012-09-30T13:47:08 < dongs> now i do
2012-09-30T13:47:12 < dongs> took a while to figure out hwo to use 4chan
2012-09-30T13:49:05 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-30T13:55:59 < cjbaird> Tonight's project: trying to resurrect the 2MB RAM Linux laptop..
2012-09-30T13:58:03 < dongs> resurrect it in the recycle bin
2012-09-30T14:04:26 < cjbaird> lol, the 486 NetBSD laptop still boots. The LCD screen is a bit fuxored, though.
2012-09-30T14:05:17 < BrainDamage> with the amount of time wasted on that, you could probably buy a new laptop if invested in money making activities
2012-09-30T14:07:31 < cjbaird> It's a laptop heavy enough to do a good amount of physical damage to the next smartarse Gen-Y
2012-09-30T14:07:41 < dongs> haha
2012-09-30T14:08:42 < cjbaird> Ha, 790 days since I last used it.
2012-09-30T14:08:47 < zyp> I tried running linux on a 486 box a few years ago, turned out that the 2.6 kernel requires more than 2MB RAM
2012-09-30T14:09:04 < dongs> surprised it doesnt require something > 386
2012-09-30T14:10:33 < cjbaird> The 386 laptop had a very heavily hacked 2.0.39 kernel. It was pretty-much rolled back to the SysV-era (no networking, IPC, ...) but it worked well as a console-only Unix..
2012-09-30T14:11:13 < BrainDamage> but what you'd use it for? run calculations?
2012-09-30T14:11:32 < BrainDamage> or console only text editing?
2012-09-30T14:12:28 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has joined ##stm32
2012-09-30T14:12:58 < cjbaird> Well, it was my first laptop-- I bought it for $100 soon after Win95 came out. It could run Minix no problems, but for /nice/ stuff like zgv, Emacs, gcc, I had a go at getting a custom Debian Linux install running-- and eventually succeeded..
2012-09-30T14:13:19 < dongs> run hurd on it.
2012-09-30T14:13:49 < cjbaird> I'm thinking it would be useful to make a backup of the system disk image, and use that for MCU emulators, like that ATtiny8 emulator did a few months ago.
2012-09-30T14:15:15 < cjbaird> The 386 made itself very useful when I was stuck out in Charleville for a month. I wrote a satellite-tracking program with it while I was there.
2012-09-30T14:17:13 < cjbaird> (..and 640x480 greyscale porn is better than nothing, when you're staying at a girlfriend's parent's place, and she's too Catholic to do anything funky like skinny-dipping in the huge dam on the property. :P)
2012-09-30T14:17:22 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-30T14:17:25 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-09-30T14:17:41 < BrainDamage> unrelated: wtf happened in 2006? https://www.google.com/trends/explore?hl=en#q=anal%20fisting
2012-09-30T14:19:03 < dongs> lol zurich
2012-09-30T14:20:42 < jpa-> i don't know, but it was noticed in 2011 :P   https://www.google.com/trends/explore?hl=en#q=2006%20anal%20fisting
2012-09-30T14:21:17 < dongs> lawl
2012-09-30T14:21:40 < prog`> dude you're a genius! you just discovered a precusor signal to generalized economic depressions
2012-09-30T14:30:48 < cjbaird> woo, the 386 laptop can at least still power-up and enter the BIOS
2012-09-30T14:33:07 < cjbaird> the hell... the onboard clock still has the correct time (+/- 10 minutes) after 10 years without power..
2012-09-30T14:35:05 < jpa-> 2ppm.. not too bad
2012-09-30T14:37:12 < cjbaird> For a part that must be 23-25 years old, too.
2012-09-30T14:42:14 < cjbaird> I did replace the CMOS battery the last time..
2012-09-30T14:48:41 < dongs>  * AeriCELL backie
2012-09-30T14:48:42 < dongs> bzzt
2012-09-30T14:48:45 < dongs> furry detected
2012-09-30T15:07:34 < cjbaird> Compaq harddrives make totally-authentic Space Invaders noises when they fail from a head-crash. :)
2012-09-30T15:08:04 < cjbaird> *doooop* *doooooop* *dooooooop* *SMAASSSSSSSSSSSSSSSHHHH*
2012-09-30T15:14:04 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-30T15:16:34 < corecode> i wonder whether there is anybody on this planet actually knows SWD
2012-09-30T15:19:19 < cjbaird> http://i.imgur.com/6tivP.jpg :<
2012-09-30T15:19:47 < karlp> corecode: what are you struggling with? swd in openocd?
2012-09-30T15:21:53 < corecode> karlp: i think it is related to the swd port on the kinetis, but not sure
2012-09-30T15:22:12 < corecode> karlp: i implemented swd for busblaster in the blackmagic software
2012-09-30T15:23:07 < corecode> but somehow the protocol loses sync when there is a memory access to an invalid address
2012-09-30T15:34:57 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-30T15:38:29 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has joined ##stm32
2012-09-30T15:40:06 < corecode> http://imgur.com/0NCfa
2012-09-30T15:40:07 < corecode> :/
2012-09-30T15:41:29 < karlp> sounds like you're working a lot lwoer with swd thatn most of us.
2012-09-30T15:41:34 < karlp> did you get a hold of gsmcmullin?
2012-09-30T15:41:43 < karlp> I haven't seen him active here as much recently,
2012-09-30T15:43:28 < corecode> didn't yet
2012-09-30T15:43:46 < corecode> the swd code looks also not as sophisticated
2012-09-30T15:49:51 < dongs> swd for busblateR? what does htat mean
2012-09-30T15:50:24 < corecode> busblaster is a jtag/swd adapter
2012-09-30T15:50:28 < corecode> ftdi-based
2012-09-30T15:51:11 < dongs> oh.
2012-09-30T15:51:12 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-09-30T15:51:13 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-09-30T15:51:13 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-30T15:52:57 < dongs> but blackmagic = for cortex?
2012-09-30T15:53:01 < dongs> er, stm32
2012-09-30T15:53:07 < dongs> so how does that have to do with ftdi
2012-09-30T15:53:51 < karlp> the black magic code can run on linux and target via a ftdi cable
2012-09-30T15:53:58 < karlp> it's slower, but the codebase runs there too.
2012-09-30T15:54:45 < karlp> presumably it was pretty helpful for gareth while it was in development
2012-09-30T15:55:11 < dongs> thats kinda what i figured
2012-09-30T15:56:28 < corecode> yea
2012-09-30T15:56:34 < corecode> i added mpsse support
2012-09-30T15:56:48 < corecode> bus turnaround takes some time
2012-09-30T15:56:54 < corecode> but still fairly quick
2012-09-30T16:00:48 < dongs> try using it on windows wiht D2xx library
2012-09-30T16:01:19 < corecode> what's D2xx
2012-09-30T16:01:56 < dongs> the nonvcp version of ftdi
2012-09-30T16:02:01 <+dekar> the proprietary FTDI lib, works much better usually
2012-09-30T16:02:02 < dongs> http://www.ftdichip.com/Support/Documents/ProgramGuides/D2XX_Programmer's_Guide(FT_000071).pdf
2012-09-30T16:02:06 < dongs> right
2012-09-30T16:02:12 < corecode> oh the ftdi is not a problem
2012-09-30T16:02:33 <+dekar> it should be faster
2012-09-30T16:02:43 < corecode> not a problem either
2012-09-30T16:03:14 <+dekar> well at least when I did I2C on windows and linux using an FTDI the D2xx windows lib had the better performance
2012-09-30T16:03:24 < corecode> fair enough
2012-09-30T16:03:29 < corecode> but the ftdi side is covered
2012-09-30T16:04:13 < corecode> my problem is that the kinetis swd port stops driving the SWDIO line
2012-09-30T16:06:14 < dongs> sounds liek kinetis problem
2012-09-30T16:07:38 < corecode> or i'm not handling the low level protocol correctly
2012-09-30T16:07:47 < Thorn> my f107 enumerates
2012-09-30T16:08:05 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-09-30T16:09:54 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-09-30T16:10:12 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-09-30T16:23:05 < Thorn> windows requests config descriptor 8 times, wth.
2012-09-30T16:23:27 < Thorn> no, 9 times
2012-09-30T16:25:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-09-30T16:26:58 < karlp> it gets lonely, feels the need to talk a lot.
2012-09-30T16:27:18 < karlp> my windows laptop at work re-requests an IP over dhcp every 2 minutes.
2012-09-30T16:28:04 < zyp> Thorn, does it transfer correctly then?
2012-09-30T16:28:09 < corecode> https://community.freescale.com/thread/300702
2012-09-30T16:28:44 < zyp> I've observed that host stack likes to ask again several times if they are not happy with the answer they get
2012-09-30T16:28:53 < zyp> so multiple requests is a sign of something being wrong
2012-09-30T16:29:04 < dongs> "i didnt hear the shit i wanted to hear the first time, so im gonna ask it again
2012-09-30T16:29:44 < Thorn> dunno yet, but it started identifying the device when it started sending this stuff.
2012-09-30T16:30:06 < Thorn> I need a windows tool to show descriptors
2012-09-30T16:30:18 < karlp> goodluck corecode!
2012-09-30T16:31:02 < corecode> thanks
2012-09-30T16:31:13 < corecode> but this is something odd
2012-09-30T16:31:25 < corecode> because i'd expect that swd port IP core to come from arm directly
2012-09-30T16:32:09 < zyp> and you've checked this exact scenario with other chip vendors?
2012-09-30T16:32:12 < karlp> so, have you tried the exact same thing with an other chip?
2012-09-30T16:32:15 < karlp> yeah, what he said :)
2012-09-30T16:32:35 < karlp> surely you've got enough stm32L1 and nuvotron parts lying around?
2012-09-30T16:32:40 < corecode> i do
2012-09-30T16:32:42 < corecode> have them
2012-09-30T16:33:08 < corecode> the nuc120 works
2012-09-30T16:33:11 < corecode> without a hitch
2012-09-30T16:33:27 < karlp> with the same code and hardware you're using now?
2012-09-30T16:34:14 < corecode> yep
2012-09-30T16:34:20 < corecode> just plugged it in
2012-09-30T16:34:43 < corecode> oh fun, the stm32l doesn't?
2012-09-30T16:35:18 < corecode> did before
2012-09-30T16:35:32 < corecode> doesn't talk swd to me anymore
2012-09-30T16:36:43 < corecode> maybe i killed the chip
2012-09-30T16:37:05 < corecode> anyways, it works perfectly with the nuc120
2012-09-30T16:40:39 < corecode> openocd still works
2012-09-30T16:40:45 < corecode> jesus, this stuff is brittle
2012-09-30T16:41:00 < karlp> yup
2012-09-30T16:42:11 < karlp> one wrong move and it's all over
2012-09-30T16:42:33 < corecode> if they would at least document their own hardware
2012-09-30T16:42:42 < dongs> all over? comeon
2012-09-30T16:42:46 < dongs> what can you ruin over SWD
2012-09-30T16:42:47 < dongs> nothing.
2012-09-30T16:44:40 < qyx_> you would be surprised
2012-09-30T16:45:07 < qyx_> my jtag didn't like messing around with 12V
2012-09-30T16:45:15 < dongle> lol.
2012-09-30T16:47:43 < corecode> reset sequence brittleness
2012-09-30T16:47:47 < dongle> fucking trash feenode
2012-09-30T16:51:14 < dongs> test
2012-09-30T16:51:27 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-09-30T16:51:51 < dongs> argh
2012-09-30T16:52:03 < dongs> new pcb order but I dont have led driver board ready yet to panel (:
2012-09-30T16:52:21 < qyx_> can you advise some nice temp sensor with digital interface, absolute precision <1C and resolution at least 0.25C?
2012-09-30T16:52:31 < qyx_> besides dallas 1-wire things
2012-09-30T16:54:02 < karlp> that's going to start getting spendy...
2012-09-30T16:55:47 < karlp> dongs: by all over, I meant, unplug replug, start again. not that you'd destroyed anything.
2012-09-30T16:56:43 < dongs> just use internal stm32 temperature sensor ^_-
2012-09-30T16:57:07 < qyx_> you mean that one which is uncalibrated? :X
2012-09-30T16:57:10 < dongs> most baros have a decent temperature sensor since they need to compensate...
2012-09-30T16:57:36 < qyx_> i need to sense temp outside of stm32
2012-09-30T16:57:54 < dongs> MS5611 has <0.01C resolution for temperature
2012-09-30T16:57:54 < dongs> and +-0.8C accuracy
2012-09-30T16:57:55 < qyx_> mcp9800 seems to be quite good, i2c, 12bit
2012-09-30T16:58:11 < dongs> 24bit for 5611.
2012-09-30T16:59:20 < qyx_> unavailable at farnell :S
2012-09-30T16:59:50 < corecode> interestingly enough the stm32l does something similar
2012-09-30T17:00:06 < corecode> it outputs a wait, then doesn't react anymore
2012-09-30T17:02:59 < dongs> yeah you dont buy it from farnell.
2012-09-30T17:03:05 < dongs> europlaces have them
2012-09-30T17:03:06 < dongs> is this a oneoff?
2012-09-30T17:03:07 < dongs> or wat
2012-09-30T17:03:20 < dongs> you can score some breakout boards for them form ebay for around 20-25bucks.
2012-09-30T17:03:24 < dongs> (m,s5611)
2012-09-30T17:04:18 < dongs> the pressure shit on it ois so fucking accurate, i was displaying a graph and walked into my office, closing the door, and it registered as a dip on teh graph
2012-09-30T17:04:29 < karlp> what are you watching that you want the 0.25C resolution?
2012-09-30T17:05:02 < karlp> yeah, I had fun with a capacitive humidity sensor that dipped when I opened the front door.
2012-09-30T17:05:24 < karlp> or when I left my desk and it was further from my beathing
2012-09-30T17:06:02 < qyx_> temp stability over time in server room
2012-09-30T17:06:12 < qyx_> rooms
2012-09-30T17:06:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-30T17:06:37 < qyx_> i found mcp9808
2012-09-30T17:06:50 < qyx_> 0.86e @ 100pcs
2012-09-30T17:06:56 < dongle> that'll work.
2012-09-30T17:07:21 < qyx_> res 0.0625C, accuracy +-0.25
2012-09-30T17:14:42 < dongs> ok dongs
2012-09-30T17:18:28 < dongs> started on ledtrash schema. mostly done except pinswap to the leds and then input/output stuff for block communication.
2012-09-30T17:19:30 < dongs> http://i.imgur.com/lQzzY.png drawin' in FRITZING
2012-09-30T17:20:19 < dongs> now do I add CAN or USB
2012-09-30T17:20:22 < dongs> TOUGH CHOICE.
2012-09-30T17:23:07 < dongs> bedtime cuz WEEKDAY
2012-09-30T17:23:43 < BrainDamage> dongs: https://www.youtube.com/watch?v=d3CWLCoQu7c
2012-09-30T17:25:51 < Thorn> ok wtf. if I make my descriptor local (AND return a pointer to it), it doesn't crash. as soon as I declare it static const, it crashes in some unrelated place
2012-09-30T17:26:19 < Thorn> is returning a pointer to a local variable from a function supposed to be the wrong thing?
2012-09-30T17:26:33 < Thorn> *isn't
2012-09-30T17:27:39 < Thorn> in fact if it's static but not const it still works.
2012-09-30T17:45:32 < ABLomas> lol@link
2012-09-30T17:47:59 < Thorn> wtf. VID 0x1010 turns out to be Fukuda Denshi Co., Ltd
2012-09-30T17:48:36 < Thorn> how many of them are already allocated anyway?
2012-09-30T17:49:38 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-09-30T17:56:57 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-09-30T18:02:49 < Thorn> windows seems to re-request config descriptor when I e.g. run listdevs from libusb or refresh device manager
2012-09-30T18:05:11 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-30T18:07:17 < zyp> that's normal
2012-09-30T18:07:28 < zyp> so does libusb on other oses
2012-09-30T18:09:45 < Thorn> winusb api is ridiculously overcomplicated.
2012-09-30T18:10:56 < zyp> I played with a gamepad yesterday that had a physical switch to choose between HID and some microsoft-made mode
2012-09-30T18:11:39 < zyp> ran it through the analyzer and looked at the difference between the modes
2012-09-30T18:13:04 < jkent> i've said this before somewhere else, i wish there was a special vid:pid pair allocated that allows you use a guid in a string descriptor guid to identify a device
2012-09-30T18:13:16 < jkent> .. blah, goofed that up
2012-09-30T18:13:16 < jkent> haha
2012-09-30T18:13:53 < Thorn> I'm pretty sure guid is rot13 for evil
2012-09-30T18:20:04 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-09-30T18:20:38 < jkent> nah, rot13 of evil is rivy ;)
2012-09-30T18:39:38 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-09-30T18:40:27 < cjbaird> gots the 2M RAM Linux laptop restored and working \(^_^)/
2012-09-30T18:40:51 < cjbaird> protip: preserve soft links when restoring a unix filesystem...
2012-09-30T18:41:53 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-09-30T18:42:35 < cjbaird> Might be an idea to delete-zero-write the 1980s-era (straight) furry porn from the filesystem before sharing the image, though..
2012-09-30T18:43:25 < TitanMKD> hi
2012-09-30T18:52:15 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-09-30T18:55:21 < Laurenceb_> lol cjbaird
2012-09-30T18:55:33 < Laurenceb_> its ok if its straight
2012-09-30T18:57:16 < cjbaird> "From a mystical time before Ken Sample started on that Macro shit.."
2012-09-30T18:58:19 < cjbaird> The 68hc11 gcc cross-compiler is on here too..
2012-09-30T19:03:02 < Laurenceb_> lol
2012-09-30T19:03:11 < Laurenceb_> avr cant swap clk source at runtime
2012-09-30T19:05:08 < Laurenceb_> avr sucks so bad
2012-09-30T19:06:39 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-09-30T19:36:46  * emeb_mac designed a glitchless live clock switch for an ARM SoC once.
2012-09-30T19:37:21 < emeb_mac> it's not rocket surgery - wonder why AVR doesn't support it.
2012-09-30T19:41:55 < cjbaird> patents?
2012-09-30T19:42:38 < emeb_mac> nah - this isn't patentable territory. Ancient stuff.
2012-09-30T19:43:50 < emeb_mac> what am I saying though - that never stops anyone these days.
2012-09-30T19:44:30 < emeb_mac> patent office will approve just about anything - even stuff from expired patents.
2012-09-30T19:44:38 < emeb_mac> then let 'em fight it out in court.
2012-09-30T19:44:52 < emeb_mac> the party with the most expensive lawyers wins.
2012-09-30T19:49:11 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-09-30T19:49:23 < cjbaird> Was surprised to read a while ago that Chuck Moore, the Forth guy, has gotten megarich-- but not from the language or the companies involved, but from a patent involved with CPU pipelining..
2012-09-30T19:52:19 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-09-30T19:52:30 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-09-30T19:53:56 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-30T19:56:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-30T19:56:48 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-30T19:57:05 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-30T19:58:42 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-30T19:59:03 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-09-30T20:02:00 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-09-30T20:03:54 < Thorn> this OTG driver is pure gold. USBD_SetConfiguration calls USBD_SetCfg. all the latter does is call 2 callbacks: class->Init and user->DeviceConfigured
2012-09-30T20:05:19 <+dekar> RF is black magic: http://www.youtube.com/watch?v=eyA2lWQrxwg#t=1012s
2012-09-30T20:05:20 < Thorn> the former is in usbd_req (where standard requests are handler), the latter is in usbd_core.c (where you'd expect some heavy magic but in reality it's a layer that calls callbacks)
2012-09-30T20:05:45 < Thorn> prepare your ana^H^H^Hhandler stack
2012-09-30T20:08:37 <+Steffann> dekar, yeah. They care about every mm of coper but do some ugly solder jobs on some places
2012-09-30T20:08:47 < zyp> are you writing your own stack?
2012-09-30T20:09:33 < BrainDamage> Steffann: most of the signal is between the copper trace on the upper & bottom layer
2012-09-30T20:09:44 < BrainDamage> ie within the pcb's material
2012-09-30T20:09:51 <+Steffann> yes, but still
2012-09-30T20:11:00 < Thorn> nope, I'm writing a (vendor specific) class for the ST OTG stack. I have an itch to do my own stack but the lack of a usb analyzer is stopping me.
2012-09-30T20:11:26 <+Steffann> The vias at 18:12 are nice :D
2012-09-30T20:11:48 <+dekar> BrainDamage, I think it actually is on the top, they had insane metal shielding screwed onto the pcb
2012-09-30T20:12:28 < Thorn> now why is my device unconfigured, is it supposed to be in that state until I manually configure it? (vendor-specific + winusb.sys)
2012-09-30T20:12:31 <+Steffann> no dekar . RF = black magic
2012-09-30T20:13:28 <+Steffann> Too bad the last time I did some RF is > 1.5 years ago
2012-09-30T20:13:36 <+Steffann> Forgot probably most of it
2012-09-30T20:15:01 < BrainDamage> there's a shitload of delay lines in that analyzer
2012-09-30T20:15:19 < BrainDamage> I see also some power divisors
2012-09-30T20:15:44 < BrainDamage> and directional couplers
2012-09-30T20:16:03 <+Steffann> And some fancy pcb only filters?
2012-09-30T20:16:32 < BrainDamage> yes, the filters are made with the directional couplers and delay lines
2012-09-30T20:17:01 < BrainDamage> the wedge shapes are to maintain constant impedance over broadband
2012-09-30T20:21:06 <+dekar> I like the wedge things, if I'll ever make a PCB I'll make sure to add them randomly :D
2012-09-30T20:29:06 < Thorn> #^@&#%^ windows, restart required
2012-09-30T20:33:03 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-09-30T20:33:06 -!- mode/##stm32 [+v izua] by ChanServ
2012-09-30T20:37:32 -!- Thorn [~Thorn@unaffiliated/thorn] has quit []
2012-09-30T20:46:48 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-09-30T20:49:08 < Thorn> okay after rebooting it seems configured
2012-09-30T20:49:58 < Thorn> ftdi drivers didn't require reboot. why does winusb?
2012-09-30T21:10:48 <+dekar> cause windows?
2012-09-30T21:11:05 < emeb> as good an excuse as any.
2012-09-30T21:11:51 <+dekar> they should move as many drivers as possible to the userland
2012-09-30T21:12:23 < emeb> give intel a good excuse to sell faster, more expensive processors?
2012-09-30T21:12:59 <+dekar> microkernels have <10% overhead
2012-09-30T21:13:43 <+dekar> also HID drivers are userland already, it just makes no sense atm
2012-09-30T21:13:51 < BrainDamage> the tendency is hibridization of kernels
2012-09-30T21:14:06 < BrainDamage> slow, but happening
2012-09-30T21:14:12 < BrainDamage> both win, linux, etc
2012-09-30T21:14:24 < emeb> HID has pretty low bandwidth - probably doesn't hurt them too much.
2012-09-30T21:14:25 <+dekar> OSX and Linux have userland usb support
2012-09-30T21:14:50 < emeb> libusb FTW
2012-09-30T21:15:00 <+dekar> emeb, userland doesn't generate much overhead anyway
2012-09-30T21:15:15 <+dekar> context switches are bad, but usb is high latency anyway
2012-09-30T21:15:36 < emeb> dekar: agree - low overhead, but context can cause latency issues.
2012-09-30T21:16:11 < BrainDamage> isn't usb3 going to be full duplex? that alone will make latency much more under control
2012-09-30T21:17:03 <+dekar> emeb, usb transmits up to 1000 packets per second, that's not too much
2012-09-30T21:17:56 < emeb> depends on the packet size.
2012-09-30T21:18:26 <+dekar> I don't think you can transmit over 1000 per second even for small sizes
2012-09-30T21:18:38 < Bird|lappy> dekar, for LS/FS, true
2012-09-30T21:18:42 < emeb> If you've got a SDR sampler running a few MSPS then getting blocked could be disastrous.
2012-09-30T21:18:47 < Bird|lappy> HS uses microframes though
2012-09-30T21:19:32 <+dekar> emeb, unless you're using low throughput interrupt endpoints you'll never get a guarantee concerning the available bandwidth
2012-09-30T21:19:44 < Bird|lappy> dekar, or isochronous
2012-09-30T21:20:02 <+dekar> Bird|lappy, well if you don't care about transmission errors...
2012-09-30T21:21:22 <+dekar> I have watched my FTDI bitbang i2c - totally failing once I copied data from my external HDD :/
2012-09-30T21:21:43 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-09-30T21:21:47 <+dekar> so just don't assume realtime guarantees and you don't get burned :)
2012-09-30T21:22:00 < emeb> wise advice
2012-09-30T21:22:31 <+dekar> also concerning overhead in microkernels, I don't think I care
2012-09-30T21:22:46 <+dekar> I would accept 20% overhead for a proper kernel
2012-09-30T21:22:49 < emeb> leave that to the OS weenies.
2012-09-30T21:24:34 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-30T21:24:38 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-30T21:24:58 -!- izua [~izua@unaffiliated/izua] has quit [Read error: Connection reset by peer]
2012-09-30T21:26:55 -!- barthess [~barthess@178.154.13.227] has joined ##stm32
2012-09-30T21:28:36 -!- barthess [~barthess@178.154.13.227] has quit [Remote host closed the connection]
2012-09-30T21:29:39 -!- barthess [~barthess@178.154.13.227] has joined ##stm32
2012-09-30T21:42:45 < Thorn> okay I can blink a led via control requests
2012-09-30T21:45:09 <+dekar> that's what I did O.o
2012-09-30T21:54:30 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-09-30T21:54:32 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-09-30T21:56:52 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 248 seconds]
2012-09-30T21:58:16 < Laurenceb_> i just realised F1 connectivity line and F4 have spare PLLs that can drive MCO pin(s)
2012-09-30T21:58:19 < Laurenceb_> very useful
2012-09-30T22:02:58 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-09-30T22:03:46 < Tectu> when i used a dedicated stlinkv2 with texane/stlink, I could do    arm-none-eabi-gdb foo.elf   then i had all the debugging symbols in my gdb which was really handy. Now I use some OpenOCD. When I give the .elf as an argument of gdb, i cannot connect to the target
2012-09-30T22:03:51 < Tectu> anyone know about that?
2012-09-30T22:04:01 < Tectu> I use  target remote localhost:3333 to connect to the OpenOCD server
2012-09-30T22:04:40 < Tectu> i get
2012-09-30T22:04:50 < Tectu> Remote debugging using :3333
2012-09-30T22:04:51 < Tectu> Remote 'g' packet reply is too long: 0000000019000000dfff000004000000f0bc000025000000210000003700000019000000000000000f0000002d0000004001000000040020ffffffff615000080000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000020000001
2012-09-30T22:05:19 < Laurenceb_> 0xFFFFFFFFFFFFFFFFUUUUUUUUCCCCKKKKKKk
2012-09-30T22:05:27 < Tectu> o.O
2012-09-30T22:05:33 < jpa-> i agree with gdb, the reply is too long
2012-09-30T22:05:33 < Laurenceb_> dont use openocd here
2012-09-30T22:05:41 < Tectu> jpa-, ;-)
2012-09-30T22:07:23 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-09-30T22:10:19 < Thorn> try extended-remote?
2012-09-30T22:10:50 < Thorn> data OUT works.
2012-09-30T22:11:11 < Tectu> Thorn, can you explain?
2012-09-30T22:11:27 < Thorn> target extended-remote localhost:3333
2012-09-30T22:15:06 < Tectu> Thorn, what's exactly the difference, Thorn ?
2012-09-30T22:15:30 < jpa-> different protocol
2012-09-30T22:15:33 < jpa-> (slightly)
2012-09-30T22:15:38 < Tectu> i am getting the same issue
2012-09-30T22:16:03 < jpa-> i think extended-remote affects more the commands gdb sents than the replies it expects
2012-09-30T22:21:52 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-09-30T22:22:36 < Thorn> KILL C++ WITH FIRE
2012-09-30T22:22:59 < Thorn> data IN works.
2012-09-30T22:22:59 <+Steffanx> Same with mr Thorn
2012-09-30T22:25:03 < Thorn> ?: has lower precedence than <<.
2012-09-30T22:29:56 < Tectu> is OpenOCD really better than texane/stlink? o0
2012-09-30T22:36:20 -!- izua_ [~izua@unaffiliated/izua] has quit [Ping timeout: 248 seconds]
2012-09-30T22:44:22 -!- izua_ [~izua@86.121.85.113] has joined ##stm32
2012-09-30T22:44:23 -!- izua_ [~izua@86.121.85.113] has quit [Changing host]
2012-09-30T22:44:23 -!- izua_ [~izua@unaffiliated/izua] has joined ##stm32
2012-09-30T22:44:24 -!- mode/##stm32 [+v izua_] by ChanServ
2012-09-30T22:49:08 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 248 seconds]
2012-09-30T22:57:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-09-30T22:57:32 < BrainDamage> do you really thing that's a sane question to make in a place with opinionated people?
2012-09-30T22:57:37 < BrainDamage> is software x really better than y?
2012-09-30T22:58:02 < BrainDamage> if all agree, ok, but if some disagree ...
2012-09-30T22:59:14 < zyp> Tectu, my answer: in some ways, but not all
2012-09-30T23:01:00 < BrainDamage> openocd requires manual configuration of target chip with tcl scripting, is more known and used, stlink is more specific for stm32 but supports automatic reckognition of target chip and programmer
2012-09-30T23:01:19 -!- mansfeld [~andrew@robopoly/watson] has quit [Ping timeout: 272 seconds]
2012-09-30T23:01:22 < BrainDamage> openocd also supposedly supports certain debug functions that stlink lacks
2012-09-30T23:02:21 < Tectu> zyp, :)
2012-09-30T23:04:18 < zyp> openocd has a bunch of deps and the tcl stuff, so it's more work to get going, and I haven't heard anything about it being able to send those xml descriptions to gdb yet
2012-09-30T23:04:41 < zyp> but on the other hand, the code seem to be of higher quality and less bug-ridden
2012-09-30T23:05:13 < Thorn> VBUS_SENSING_ENABLED is of course not documented anywhere.
2012-09-30T23:08:14 < zyp> in the driver for the OTG controller?
2012-09-30T23:08:25 < Thorn> yes
2012-09-30T23:08:35 < zyp> VBUSBSEN flag is pretty well documented in the reference manual :p
2012-09-30T23:08:52 < Thorn> it doesn't call connected/disconnected callbacks by default
2012-09-30T23:09:28 < Thorn> the functions that call them are only compiled in when that #define is #defined
2012-09-30T23:09:49 < Thorn> not sure if it even has anything to do with that flag.
2012-09-30T23:09:52 < zyp> oh, sounds like something different then
2012-09-30T23:09:54 < zyp> yeah
2012-09-30T23:10:08 < Thorn> also, suspended/resumed callbacks are somewhat erratic
2012-09-30T23:10:17 < zyp> well, I haven't studied the ST driver
2012-09-30T23:10:27 < zyp> how so?
2012-09-30T23:10:52 -!- barthess [~barthess@178.154.13.227] has quit [Quit: Leaving.]
2012-09-30T23:11:20 < Thorn> suspended is almost always called before disconnected, resumed is often called after connected, but sometimes suspended seems to be called before reset
2012-09-30T23:12:32 < Thorn> e.g. I see either on_connected, on_reset, or on_connected, on_resumed, on_reset in my log
2012-09-30T23:12:55 < zyp> well, a suspend is a lack of traffic, and a disconnect is a lack of vbus, so I guess what's detected first depends on stuff like vbus capacitance
2012-09-30T23:13:21 < zyp> IIRC a suspend is no reception of SOF for three consecutive frames
2012-09-30T23:13:32 < Thorn> or even on_connected, on_resumed, on_suspended, on_reset once
2012-09-30T23:14:13 < zyp> I wouldn't worry too much about the order of those events
2012-09-30T23:14:18 < Thorn> if you low-powered everything in on_suspended, that would be a nasty surprise
2012-09-30T23:14:43 < Thorn> it's not the order
2012-09-30T23:14:47 < zyp> the reset should always be the last before enumeration, right?
2012-09-30T23:15:12 < Thorn> windows resets at least twice
2012-09-30T23:15:19 <+dekar> izua_, why buy a rubidium frequency standard when you can just use the one in the nearest GSM tower? http://steve-m.de/pictures/rtlsdr_external_clock.jpg
2012-09-30T23:15:23 < zyp> yes, I see that with OS X here too
2012-09-30T23:15:57 <+izua_> dekar: does that work only with a receiver?
2012-09-30T23:16:01 <+izua_> or do you need to send it any data?
2012-09-30T23:16:08 < Thorn> IIRC the synopsys core has early suspend and late suspend events but this driver doesn't seem to support that
2012-09-30T23:17:00 <+dekar> izua_, I think it just works, the 0Hz at the bottoms shows the difference to the tower
2012-09-30T23:17:38 <+dekar> they always have to stay in sync for time multiplex etc
2012-09-30T23:18:46 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has joined ##stm32
2012-09-30T23:18:46 -!- mansfeld [~andrew@2001:1af8:4400:a013:3::12c] has quit [Changing host]
2012-09-30T23:18:46 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32
2012-09-30T23:20:51 < karlp> Tectu: that's an argument with openocd and gdb about how to describe target registers. you won't get it fixed short of getting "correct" gdb versions.
2012-09-30T23:20:53 < Thorn> so looks like you can have from 2 to 4 events when plugging in the cable
2012-09-30T23:21:26 < Thorn> on_connected, [on_resumed], [on_suspended], on_reset
2012-09-30T23:21:28 < zyp> Thorn, found a log from F4 here: http://paste.jvnv.net/view/fTjRl
2012-09-30T23:21:47 < zyp> I'm just logging RESET and ENUMDNE interrupts though
2012-09-30T23:21:52 < Tectu> karlp, so I cannot "import" my elf into gdb? o0
2012-09-30T23:22:02 < karlp> no, you can, but you can't make openocd talk to your gdb
2012-09-30T23:22:13 < karlp> paul fertzer has the fix in a gmane posting somewhere
2012-09-30T23:22:26 < karlp> and openocd wants you to use a newer/differrent gdb
2012-09-30T23:22:31 < karlp> what oocd version are you using?
2012-09-30T23:22:46 < Tectu> 0.60
2012-09-30T23:22:48 < karlp> you're using oocd with the stlink as the swd device right?
2012-09-30T23:22:48 < Tectu> 0.6.0
2012-09-30T23:22:57 < Tectu> karlp, nope
2012-09-30T23:23:08 < Tectu> karlp, i use openocd with interface/jtagkey2.cfg
2012-09-30T23:23:12 < Tectu> it's some ftdi chip
2012-09-30T23:23:27 < karlp> well, that g reply too long is still between the oocd and the gdb versions
2012-09-30T23:23:33 < karlp> sorry for bringing up the stlink part.
2012-09-30T23:23:42 < Tectu> no problem
2012-09-30T23:23:53 < Tectu> so i simply should go back to use my old stlink with texane? :D
2012-09-30T23:23:54 < karlp> there's #openocd here too...
2012-09-30T23:24:09 < karlp> whatever works for you... :)
2012-09-30T23:24:28 < karlp> openocd can target about a 10000 times more devices, but stlink actually works... (for me at least)
2012-09-30T23:24:52 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-09-30T23:25:15 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-09-30T23:25:17 < Tectu> karlp, i asked there, no response yet
2012-09-30T23:25:30 < zyp> Thorn, I can add logging of the other events in my code and do some traces if you are interested
2012-09-30T23:25:37 < karlp> (it's not nearly as busy a channel)
2012-09-30T23:25:40 < zyp> (later)
2012-09-30T23:26:23 < zyp> I'd probably benefit from knowing that behavior in detail later anyway :p
2012-09-30T23:26:55 < Tectu> karlp, but not sure: is there now a solution for my problem, or am I just doing it wrong?
2012-09-30T23:27:04 < Thorn> well, it may be interesting to see connect/disconect & suspend/resume events also
2012-09-30T23:27:12 < karlp> can't remember.
2012-09-30T23:27:22 < Thorn> especially if you're going to rely on them in any way
2012-09-30T23:27:23 < karlp> presumably you had some reason to be using a jtagkey and oocd?
2012-09-30T23:28:19 < Tectu> karlp, well, got one for free from my university
2012-09-30T23:28:26 < Tectu> it has JTAG + UART
2012-09-30T23:29:07 < zyp> I have a jtagkey somewhere, I haven't used it since I received the f4discovery board
2012-09-30T23:30:17 < zyp> I seem to recall that it weren't very reliable
2012-09-30T23:30:42 < zyp> but I don't remember the exact issues I had
2012-09-30T23:35:22 < karlp> Tectu: are you on f4?
2012-09-30T23:35:28 -!- barthess [~barthess@178.154.13.227] has joined ##stm32
2012-09-30T23:35:30 < karlp> you can try set arm abi stuff...
2012-09-30T23:40:30 -!- DaKu is now known as daku
2012-09-30T23:41:55 < Thorn> http://www.computerhistory.org/atchm//wp-content/uploads/2012/08/williams-thermometer-sculpture.jpg
2012-09-30T23:48:12 < karlp> nice
2012-09-30T23:49:45 < Thorn> http://www.computerhistory.org/atchm/an-analog-life-remembering-jim-williams/
--- Log closed Mon Oct 01 00:00:43 2012