--- Log opened Fri Feb 01 00:00:05 2013 2013-02-01T00:01:01 < emeb> don't really see how one gets any gains from loading 4 samples at a time. 2013-02-01T00:01:27 < emeb> since that requires 2 cycles of RAM access for 1 instruction it doesn't save any time. 2013-02-01T00:01:37 < Laurenceb_> hmm 2013-02-01T00:01:45 < Laurenceb_> for n regs it take N+1 cycles 2013-02-01T00:01:51 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Write error: Broken pipe] 2013-02-01T00:01:57 < Laurenceb_> so for 1 reg it takes 2 cycles 2013-02-01T00:02:02 < emeb> ok 2013-02-01T00:02:04 < Laurenceb_> for N == large 2013-02-01T00:02:11 < Laurenceb_> it tends to half the cycles/reg 2013-02-01T00:04:31 < emeb> right. current 5 instr iteration actually uses 8 cycles. 2013-02-01T00:06:06 < emeb> loading 4 samples would bring that down to 4 instr, 6 cycles, 2 samples. 2013-02-01T00:06:09 < emeb> That would be good. 2013-02-01T00:06:40 < emeb> err cross off that initial 4 samples :P 2013-02-01T00:07:03 < emeb> lets see how that would work... 2013-02-01T00:09:54 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-01T00:25:51 -!- fiendie [fiendie@ipv6.leela.fiendie.net] has quit [Ping timeout: 245 seconds] 2013-02-01T00:28:18 < emeb> not really seeing any noticeable improvement in execution time with that change. 2013-02-01T00:28:41 -!- fiendie [fiendie@ipv6.leela.fiendie.net] has joined ##stm32 2013-02-01T00:32:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-01T00:46:18 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-01T00:49:14 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-01T00:50:00 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-01T00:50:11 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-01T00:52:31 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 245 seconds] 2013-02-01T00:54:15 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-01T00:55:45 < dirty_d> usb protocol aint nothing to ^%#& with 2013-02-01T00:57:14 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-01T01:00:12 < emeb> bites hard? 2013-02-01T01:01:56 < dirty_d> mmhmm 2013-02-01T01:10:10 < ds2> USB is easy 2013-02-01T01:12:26 < Thorn> ok then what is the PID of data stage data packets (control transfer): 1) data0 2) data1 3) data0, data1, data0, ... 4) data1, data0, data1, ... 5) none of the above 2013-02-01T01:12:56 < emeb> don't bother us with details. :) 2013-02-01T01:13:02 < ds2> PID? that would depend on the device wouldn't it? :D 2013-02-01T01:13:22 < Thorn> FAIL 2013-02-01T01:13:52 < dirty_d> usb is rediculous 2013-02-01T01:15:38 < emeb> usb is delicious with a little sriracha. 2013-02-01T01:21:03 < dirty_d> not sure whats going on 2013-02-01T01:21:04 < dirty_d> i sent the dev desc, then get a ctr_tx 2013-02-01T01:21:04 < dirty_d> then send a zlp 2013-02-01T01:21:04 < dirty_d> then get a ctr_tx 2013-02-01T01:21:04 < dirty_d> then all hell breaks loose 2013-02-01T01:21:04 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 248 seconds] 2013-02-01T01:28:05 < dirty_d> im not sending a zlp after i get the address 2013-02-01T01:28:05 < dirty_d> you dont right? 2013-02-01T01:28:05 < dirty_d> also whats strange is that a decive descriptor is 18 bytes, but the GET_DESCRIPTOR packet asking for it has a length field of 64 2013-02-01T01:28:05 < dirty_d> and that matches up with the error from linux: [866627.229301] usb 6-1.1: device descriptor read/64, error -32 2013-02-01T01:28:05 < Laurenceb_> this is why i only ever used prewritten libraries for anything usb 2013-02-01T01:28:05 < Laurenceb_> i dont have time for this shit 2013-02-01T01:28:06 < Laurenceb_> some of the ST stuff is a bit poor but its easier to rewrite that than start from scratch 2013-02-01T01:28:06 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-01T01:28:24 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 264 seconds] 2013-02-01T01:38:50 < Laurenceb_> hmm back to debugging linear actuator fail :( 2013-02-01T01:38:51 < emeb> thought you'd solved that. 2013-02-01T01:40:29 -!- Thorn_ [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-01T01:41:28 < karlp> I thought he just discarded it as a bad idea 2013-02-01T01:43:59 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds] 2013-02-01T01:44:32 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Remote host closed the connection] 2013-02-01T01:45:22 -!- Gargantuasauce_ [~Gargantua@tangerine.silverorange.com] has joined ##stm32 2013-02-01T01:49:19 -!- Thorn [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-01T01:49:21 -!- Thorn_ [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Ping timeout: 248 seconds] 2013-02-01T01:50:48 < Laurenceb_> hehe 2013-02-01T01:50:48 < Laurenceb_> maybe i should 2013-02-01T01:50:48 < Laurenceb_> its oscillating still :( 2013-02-01T01:50:49 < emeb> dang. 2013-02-01T01:50:49 < Laurenceb_> quicksort helps - im taking 7ms windows of pressure readings at high sample rate now 2013-02-01T01:50:49 < Laurenceb_> then using quicksort to chuck everything outside the interquartile range 2013-02-01T01:50:49 < Laurenceb_> and average the rest 2013-02-01T01:50:49 < emeb> discard outliers 2013-02-01T01:50:49 < emeb> common statistical technique 2013-02-01T01:50:49 < Laurenceb_> turns out i was getting some biasing of the pressure sensor from emi 2013-02-01T01:50:49 < Laurenceb_> yes 2013-02-01T01:50:50 -!- Thorn [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Changing host] 2013-02-01T01:50:50 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-01T01:51:14 < karlp> so dongs/zyp: go tyour go pro / 2gig memory card yet? 2013-02-01T01:51:17 < karlp> or what came in your box? 2013-02-01T01:51:22 < Laurenceb_> but arms are all horrible and nonlinear 2013-02-01T01:51:30 < emeb> design new ones. :) 2013-02-01T01:53:17 < Laurenceb_> http://imagebin.org/index.php?mode=image&id=245001 2013-02-01T01:53:34 < Laurenceb_> wtf @ position 2013-02-01T01:53:54 < Laurenceb_> its doing a good job at maintaining pressure considering 2013-02-01T01:54:30 < Laurenceb_> if i can just kill the short period oscillation its now quite good 2013-02-01T01:54:32 < Laurenceb_> ...just 2013-02-01T01:54:49 < Laurenceb_> its got a triangular setpoint 2013-02-01T01:54:55 < Laurenceb_> 0 to 1 psi ramp 2013-02-01T01:56:14 < dongs> karlp: mine just landed in japan 2013-02-01T01:56:21 < dongs> ill have it in 2-3 days 2013-02-01T01:56:44 < Laurenceb_> ive also written a backlash tracker/remover 2013-02-01T01:56:48 < Laurenceb_> thats helping a lot 2013-02-01T01:57:05 < karlp> my friend said, "this gopro2 looks suspiciously like a 2gig sd card" 2013-02-01T01:59:08 < dongs> karlp: haha well shrug, ebay refunded my other item instantly when i disputed it 2013-02-01T01:59:21 < dongs> (the other godong that also got listing pulled0 2013-02-01T01:59:43 < dongs> so im sure if im getting a box of aids it will be no problem getting back shitpal and if not, its only 120bucks of fakemoney 2013-02-01T02:10:24 -!- Titan___ is now known as TitanMKD 2013-02-01T02:11:47 -!- Laurenceb_ [~Laurence@host86-178-5-12.range86-178.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-01T02:16:46 -!- DaKu is now known as daku 2013-02-01T02:43:34 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-01T03:03:32 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-01T03:36:25 < R2COM> hmm LPC4330 with M0 and M4 on one die, FPU, 204MHz... 2013-02-01T03:36:36 < emeb> ya. 2013-02-01T03:36:54 < dongs> R2COM: you mean like this?? https://youtube.com/watch?v=vBmp3k82Reo 2013-02-01T03:36:56 < R2COM> and...for a price of stm32f4 I guess... 2013-02-01T03:36:57 < emeb> zyp did an LPC435x board a few weeks ago 2013-02-01T03:37:33 < emeb> OK everyone can point and laugh now: https://github.com/emeb/f303_sdr 2013-02-01T03:37:58 < R2COM> 1 min video with gay music doesnt tell much 2013-02-01T03:38:00 < emeb> not done yet, but runs and seems to make sense. 2013-02-01T03:38:12 < dongs> cool copypasted asm 2013-02-01T03:38:13 < R2COM> reading datasheet anyhow 2013-02-01T03:38:29 < emeb> dongs: that was auto-generated from a perl script 2013-02-01T03:39:34 < emeb> perl code in the support dir 2013-02-01T03:53:30 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-01T03:54:36 < dongs> emeb: wat exactly are you doing with that anyway. 2013-02-01T03:57:41 < emeb> dongs: just trying out some ideas to see if the F303 is capable of doing minimal SDR. 2013-02-01T03:59:07 < emeb> Based on results so far it looks like it should be able to do simple AM broadcast reception. 2013-02-01T03:59:10 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-01T04:00:47 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-01T04:01:21 < dongs> funtimes 2013-02-01T04:01:48 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-01T04:06:38 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has quit [Ping timeout: 240 seconds] 2013-02-01T04:07:08 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-01T04:15:50 <+dekar> emeb_mac, just the internal adc? 2013-02-01T04:16:07 < emeb_mac> dekar: yep 2013-02-01T04:16:24 < emeb_mac> on the f303 the internal adc is good up to ~5MSPS 2013-02-01T04:16:29 <+dekar> one could probably use a faster one and hook it up to FSMC :) 2013-02-01T04:16:42 < emeb_mac> if there were an fsmc on an f303. 2013-02-01T04:16:53 < emeb_mac> but yeah - on an f4 that would work 2013-02-01T04:18:16 <+dekar> do you have an usb2 phy? 2013-02-01T04:18:33 <+dekar> you can probably stream the samples directly to a pc 2013-02-01T04:25:51 < emeb_mac> dekar: f303 has a usb fs device. 2013-02-01T04:26:36 < emeb_mac> so it could act as a source for low-rate data. 2013-02-01T04:28:33 <+dekar> use an f4 already :P 2013-02-01T04:29:54 < gxti> anyone done smartcard stuff before? 2013-02-01T04:31:16 < gxti> i'm looking to make some sort of handheld device that can be used for various cryptographic purposes, probably starting with a f4 and moving to a fpga later. gonna have a smartcard for storing secrets when locked/powered off 2013-02-01T04:34:58 <+dekar> emeb_mac, have you tried this? http://dekar.wc3edit.net/2012/11/07/rtl2832u-transformer-mod/ 2013-02-01T04:35:37 < emeb_mac> dekar: yep - saw that a while back. 2013-02-01T04:39:43 < ds2> hmmm OTP i2c EEPROMs don't seem to exists 2013-02-01T04:40:54 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-122.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-01T04:51:36 < dirty_d> ds2, mr usb expert, does get_descriptor happen before set_address? 2013-02-01T04:52:42 < ds2> no 2013-02-01T04:53:01 < dirty_d> hmm 2013-02-01T04:53:13 < ds2> reset - set_address - get descr 2013-02-01T04:53:40 < dirty_d> this is just weird 2013-02-01T04:55:06 < ds2> see - http://www.beyondlogic.org/usbnetshell/usb4.shtml#Control 2013-02-01T04:56:38 < dirty_d> im getting get_descriptor first somehow 2013-02-01T04:56:53 < ds2> sent to what address? 2013-02-01T04:56:57 < dirty_d> 0 2013-02-01T04:57:12 < ds2> hmmmm that would seem to be bad 2013-02-01T04:57:26 < ds2> what is the host? 2013-02-01T04:57:39 < dirty_d> my pc 2013-02-01T04:57:48 < dirty_d> usb things work fine 2013-02-01T04:58:02 < ds2> what OS 2013-02-01T04:58:09 < dirty_d> linux 2013-02-01T04:58:14 < ds2> hmmm 2013-02-01T04:58:19 < ds2> turn on USB snooping 2013-02-01T04:58:23 < dirty_d> i did 2013-02-01T04:58:25 < ds2> and see if it is sending a set address at all 2013-02-01T04:58:28 < dirty_d> i do see get_descriptor first 2013-02-01T04:58:30 < ds2> is it sending set_address? 2013-02-01T04:58:35 < dirty_d> yes afterward 2013-02-01T04:58:55 < ds2> is this a EHCI controller on the PC? 2013-02-01T04:59:00 < ds2> or older UHCI/OHCI? 2013-02-01T04:59:03 < dirty_d> ehci 2013-02-01T04:59:22 < ds2> and ur device is FS/LS? or HS? 2013-02-01T04:59:34 < dirty_d> FS 2013-02-01T04:59:37 < dirty_d> im usiont stm32f3 2013-02-01T04:59:42 < ds2> Hmmm 2013-02-01T04:59:47 < ds2> is there a hub in the middle? 2013-02-01T05:00:07 < dirty_d> not an external one 2013-02-01T05:00:22 < ds2> as in just root hub or your PC mobo has one? 2013-02-01T05:00:29 < dirty_d> but i have my st-link plugged into the same port and that works fine 2013-02-01T05:00:35 < dirty_d> im not sure 2013-02-01T05:01:07 < ds2> did you get a reset to start with? 2013-02-01T05:01:33 < dirty_d> yup 2013-02-01T05:01:54 < dirty_d> i thought i was getting set_address first then get_descriptor 2013-02-01T05:02:00 < dirty_d> but it seems it was the opposite order 2013-02-01T05:02:01 < ds2> something off 2013-02-01T05:02:04 < dirty_d> yea 2013-02-01T05:02:06 < ds2> time to plug in the beagle 2013-02-01T05:02:10 < dirty_d> i hope im not fooling myself 2013-02-01T05:02:13 < ds2> or some other analyzer 2013-02-01T05:02:25 < ds2> a catc would be nice though 2013-02-01T05:03:43 < dirty_d> wireshark actually sniffs usb 2013-02-01T05:04:03 < dirty_d> whe i do get the get_descr request 2013-02-01T05:04:08 < dirty_d> the length is 64 though 2013-02-01T05:04:09 < dirty_d> why is that? 2013-02-01T05:04:12 < ds2> i know but I want to look at it from a neutral observer 2013-02-01T05:04:18 < dirty_d> the info is only 18 bytes 2013-02-01T05:04:19 < ds2> 64 is the min. length, IIRC 2013-02-01T05:05:18 < dirty_d> hmm 2013-02-01T05:05:30 < ds2> it gets padded on the wire 2013-02-01T05:07:37 < ds2> bbl, going home 2013-02-01T05:09:01 < dirty_d> ok 2013-02-01T05:09:55 < R2COM> soo.. Keil can debug simulatenously cores M4/M0 of LPC, free IDE LPCXpresso cant 2013-02-01T05:10:38 -!- Gargantuasauce_ [~Gargantua@tangerine.silverorange.com] has quit [Ping timeout: 240 seconds] 2013-02-01T05:10:51 < dongs> keil is pretty awesome. 2013-02-01T05:11:12 < R2COM> looks like that 2013-02-01T05:11:17 < R2COM> lets see price 2013-02-01T05:11:40 < dongs> 5k 2013-02-01T05:12:01 < dongs> freetard version works up to 32k 2013-02-01T05:13:33 < R2COM> hmmm 5k... 2013-02-01T05:13:49 < R2COM> well maybe reasonable if one does frequently many different microcontroller work 2013-02-01T05:14:58 < R2COM> i think there should be limitation on debugging too for free version 2013-02-01T05:29:28 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-01T05:33:07 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-01T05:33:29 < dongs> there isnt any 2013-02-01T05:33:38 < dongs> only linker has limit 2013-02-01T05:33:44 < dongs> i think you can even build objects > 32k 2013-02-01T05:33:48 < dongs> but final link wont work 2013-02-01T06:03:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-01T06:09:34 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-01T06:09:41 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-01T06:11:53 -!- BJFreeman [~bjfree@159.sub-75-233-88.myvzw.com] has quit [Quit: had a good time] 2013-02-01T06:14:28 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-01T06:30:55 -!- fxd0h_ [~fx@186.18.172.240] has quit [Ping timeout: 256 seconds] 2013-02-01T06:33:42 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-01T06:38:03 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-01T06:38:49 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-01T06:48:03 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-01T06:49:17 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-01T07:18:29 -!- drgreenthumb [~fletch@drupal.org/user/72475/view] has quit [Read error: Connection reset by peer] 2013-02-01T07:22:49 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-01T07:24:46 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32 2013-02-01T07:47:27 -!- R2COM1 [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-01T07:47:27 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-01T08:07:28 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-01T08:10:46 -!- R2COM1 is now known as R2COM 2013-02-01T08:11:54 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Changing host] 2013-02-01T08:11:54 -!- dfletcher [~fletch@drupal.org/user/72475/view] has joined ##stm32 2013-02-01T08:11:57 -!- dfletcher is now known as drgreenthumb 2013-02-01T08:21:22 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-01T08:49:19 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 260 seconds] 2013-02-01T09:12:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-01T09:16:57 < emeb_mac> *crickets* 2013-02-01T09:28:53 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-01T09:37:45 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-01T09:46:28 -!- dekar_ [~dekar@212.255.42.215] has joined ##stm32 2013-02-01T09:46:28 -!- mode/##stm32 [+v dekar_] by ChanServ 2013-02-01T09:48:51 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-01T09:50:07 -!- dekar [~dekar@212.255.26.75] has quit [Ping timeout: 240 seconds] 2013-02-01T09:51:01 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Ping timeout: 246 seconds] 2013-02-01T09:54:02 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-01T09:56:39 -!- NonaSuomy [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-01T09:56:39 -!- NonaSuomy [~NonaSuomy@69.158.165.12] has quit [Changing host] 2013-02-01T09:56:39 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-01T09:57:08 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-01T09:57:17 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has quit [Ping timeout: 252 seconds] 2013-02-01T09:57:23 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-01T10:04:51 < ds2> hmmm got a strange desire to make a LED cube 2013-02-01T10:08:24 < dongs> http://www.tickld.com/images/content/39664.jpg 2013-02-01T10:40:14 < zyp> 03:51:36 < dirty_d> ds2, mr usb expert, does get_descriptor happen before set_address? 2013-02-01T10:40:17 < zyp> 03:52:42 < ds2> no 2013-02-01T10:40:18 < zyp> that's wrong 2013-02-01T10:40:21 < zyp> 03:53:01 < dirty_d> hmm 2013-02-01T10:40:23 < zyp> 03:53:13 < ds2> reset - set_address - get descr 2013-02-01T10:40:36 < zyp> normal enumeration is reset - get descr - set address 2013-02-01T10:42:16 < zyp> http://bin.jvnv.net/f/L5ecr.png <- MSC enumeration 2013-02-01T11:05:44 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds] 2013-02-01T11:15:19 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32 2013-02-01T11:36:55 -!- KennyMcCormic [~Kenny@94.28.162.28] has joined ##stm32 2013-02-01T11:40:41 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-01T11:58:41 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] 2013-02-01T12:10:18 -!- Netsplit *.net <-> *.split quits: R0b0t1_, ReadError 2013-02-01T12:10:29 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-01T12:12:05 -!- Netsplit over, joins: ReadError 2013-02-01T12:13:12 -!- R0b0t1_ [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32 2013-02-01T12:13:18 < dongs> http://i.imgur.com/hjggu12.png 2013-02-01T12:13:19 < dongs> how do i drc 2013-02-01T12:13:37 < dongs> made by: this awesome software http://www.numberone.com/prices.asp?currency=USD&product=easypc 2013-02-01T12:15:58 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-01T12:16:00 < zyp> ok, buying it 2013-02-01T12:16:59 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-01T12:17:56 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 245 seconds] 2013-02-01T12:19:47 -!- KennyMcCormic|2 [~Kenny@94.28.162.28] has joined ##stm32 2013-02-01T12:21:54 -!- KennyMcCormic [~Kenny@94.28.162.28] has quit [Ping timeout: 260 seconds] 2013-02-01T12:26:53 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 252 seconds] 2013-02-01T12:31:46 -!- KennyMcCormic|2 is now known as KennyMcCormic 2013-02-01T12:34:56 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-01T12:38:29 -!- |sark [~|sark@p54B627C3.dip.t-dialin.net] has joined ##stm32 2013-02-01T12:46:39 -!- Laurenceb_ [~Laurence@host86-178-5-12.range86-178.btcentralplus.com] has joined ##stm32 2013-02-01T13:03:42 -!- Laurenceb_ [~Laurence@host86-178-5-12.range86-178.btcentralplus.com] has quit [Ping timeout: 264 seconds] 2013-02-01T13:13:22 < trepidaciousMBR> Is it an unforgivably awful idea to use structs for data storage (just writing structs directly to file, network etc.)? 2013-02-01T13:13:46 < trepidaciousMBR> I understand that packing can change, breaking everythingm but can that be reliably avoided to get the benefit of much simpler code? 2013-02-01T13:15:14 -!- _BJFreeman [~bjfree@18.sub-75-233-167.myvzw.com] has joined ##stm32 2013-02-01T13:15:54 -!- _BJFreeman is now known as BJFreeman 2013-02-01T13:21:43 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-01T13:21:43 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host] 2013-02-01T13:21:43 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-01T13:24:08 < zyp> trepidaciousMBR, it's not 2013-02-01T13:24:37 < zyp> as long as you define the binary format of the file, and force the structs to always adhere to that format it will be fine 2013-02-01T13:24:51 < zyp> just keep in mind endianness 2013-02-01T13:26:18 < trepidaciousMBR> Yup, endianness is ok, I'll be using Java on the other end, Ive got code to make sure everything is correct 2013-02-01T13:26:33 < trepidaciousMBR> With the structs and GCC, what do I need to make sure I know exactly where each field is? 2013-02-01T13:26:44 < zyp> __attribute__ ((packed)) 2013-02-01T13:27:08 < zyp> then no padding will be inserted 2013-02-01T13:27:29 < trepidaciousMBR> Is that a problem on an STM32F4, I've seen some worrying stuff about non-aligned access etc.? 2013-02-01T13:27:40 < zyp> no, it's not a problem 2013-02-01T13:29:07 -!- |sark [~|sark@p54B627C3.dip.t-dialin.net] has quit [Ping timeout: 248 seconds] 2013-02-01T13:35:48 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 248 seconds] 2013-02-01T13:38:43 < jpa-> trepidaciousMBR: only problem with structs is that it will be difficult to change the format later 2013-02-01T13:38:55 < jpa-> i.e. if you add a new field, all your old files become invalid 2013-02-01T13:40:43 < trepidaciousMBR> Yup that's fine, I'm planning on having a new identified every time a struct changes, to avoid that 2013-02-01T13:41:32 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-01T13:42:57 < trepidaciousMBR> Well that all seems good - I was thinking I'd have to use something like protobuf, which looks good but a bit happy 2013-02-01T13:46:04 < trepidaciousMBR> I meant to type heavy, not happy. Wow ;) 2013-02-01T13:46:45 -!- enots [dimka@freelsd.net] has joined ##stm32 2013-02-01T13:50:16 < zyp> jpa-, just add it to the end and you'll be fine 2013-02-01T13:51:13 < karlp> trepidaciousMBR: jpa has nanobuff however, if you would like to use protobuffs. 2013-02-01T13:54:42 < trepidaciousMBR> I'll have a look at that - also adding to the end would be an option, actually I would like to use that for some stuff 2013-02-01T13:55:08 < trepidaciousMBR> Then I can initialise structs to defaults, then copy however much of the start of the struct I have in the file over the top, and hopefully get a sensible result. 2013-02-01T13:55:26 < jpa-> *nanopb 2013-02-01T13:55:28 < trepidaciousMBR> The attribute goes just after the struct, to make the whole thing packed, I guess? 2013-02-01T13:56:31 < jpa-> but yeah, if you have 2kB to spare and speed is not absolutely critical, i would recommend nanopb 2013-02-01T13:56:32 < trepidaciousMBR> I was looking at http://code.google.com/p/protobuf-embedded-c/ as well, which doesn't seem too bad, but still a little heavier 2013-02-01T13:56:56 < trepidaciousMBR> jpa-: Yup I've got a reasonable amount of space free, since it's a huge STM32F4 ;) 2013-02-01T14:00:40 < jpa-> yeah, i wouldn't worry about size then :) 2013-02-01T14:01:22 < jpa-> the "add to end" works only until you have an array of structs 2013-02-01T14:03:31 -!- KennyMcCormic [~Kenny@94.28.162.28] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 2013-02-01T14:06:01 < trepidaciousMBR> ah yeah 2013-02-01T14:27:42 < jpa-> gdb internal-error: could not find a target to follow mourn inferior it's so sad that the target died 2013-02-01T14:28:48 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-01T14:28:49 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-01T14:28:49 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-01T14:28:49 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-01T14:41:05 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-01T15:02:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has joined ##stm32 2013-02-01T15:03:43 < Laurenceb> grrr 2013-02-01T15:03:47 < Laurenceb> i hate control loops 2013-02-01T15:03:51 < Laurenceb> aka oscillators 2013-02-01T15:04:51 < Laurenceb> for it to oscillate there must be a phase lag somewhere right? 2013-02-01T15:04:57 < Laurenceb> yet i cant find the phase lag 2013-02-01T15:06:23 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 260 seconds] 2013-02-01T15:06:41 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-01T15:06:41 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host] 2013-02-01T15:06:41 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-01T15:08:04 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 248 seconds] 2013-02-01T15:13:51 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-01T15:20:59 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Read error: Operation timed out] 2013-02-01T15:21:16 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds] 2013-02-01T15:22:04 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-01T15:22:26 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-01T15:22:26 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host] 2013-02-01T15:22:26 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-01T15:25:27 < jpa-> hmm, any guesses about what is going wrong with nuttx spi sd card when dma is enabled? 2013-02-01T15:25:28 < jpa-> http://paste.dy.fi/F3H/plain 2013-02-01T15:26:02 < jpa-> i get that 0x01F824 sprayed all over the received data, but only during boot; a bit later, executed from shell, it works fine 2013-02-01T15:30:10 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection] 2013-02-01T15:30:20 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-01T15:32:41 < jpa-> hmm.. it has something to do with where the buffer is allocated.. but it's not in CCM so i would think that one ram is the same as any other ram.. 2013-02-01T15:35:22 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-01T15:38:21 < trepidaciousMBR> Nanopb looks great, thanks for that, I'll give it a go 2013-02-01T15:42:59 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-01T16:00:06 <+Steffanx> I'm so going to regret this. Wrote a little app that clears my clipboard when the text length <= 6 characters. To force me to remember such short strings. :D 2013-02-01T16:02:20 < dongs> i hope you wrote it in asm 2013-02-01T16:02:25 < dongs> and properly handled keyboard chaining 2013-02-01T16:02:32 < dongs> err 2013-02-01T16:02:58 <+Steffanx> Why in asm? 2013-02-01T16:02:59 < dongs> clipboard chaining 2013-02-01T16:03:23 < dongs> https://youtube.com/watch?v=2OL4FdIQrV4 2013-02-01T16:03:53 <+Steffanx> Whoaa, even the video is scary. 2013-02-01T16:04:01 <+Steffanx> That asks for a 3d video :P 2013-02-01T16:07:48 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-01T16:08:28 < dongs> 3D gopro 2013-02-01T16:09:55 <+Steffanx> Too bad 3D sucks ass 2013-02-01T16:15:49 -!- daku is now known as DaKu 2013-02-01T16:16:29 < Laurenceb> rageeee 2013-02-01T16:16:32 < Laurenceb> stupid oscillation 2013-02-01T16:17:12 -!- DaKu is now known as daku 2013-02-01T16:18:28 < Laurenceb> all i seem to be able to do is change the frequency 2013-02-01T16:18:59 < gxti> can you change it to 0? /s 2013-02-01T16:19:07 < Laurenceb> heh 2013-02-01T16:19:32 < Laurenceb> hmm maybe ill go back to trying extended kalman filters for _everything_ 2013-02-01T16:21:57 < Laurenceb> but that didnt work before :( 2013-02-01T16:22:24 < Laurenceb> maybe i should just give up and try pneumatics or hydralics 2013-02-01T16:23:12 -!- BJFreeman [~bjfree@18.sub-75-233-167.myvzw.com] has quit [Quit: had a good time] 2013-02-01T16:32:36 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Disconnected by services] 2013-02-01T16:32:47 -!- dirty_d [~adowning@209.213.71.70] has joined ##stm32 2013-02-01T16:34:20 < zyp> I don't think extended kalman filters for breakfast is a good idea 2013-02-01T16:35:02 <+Steffanx> So no breakfast for Laurenceb anymore 2013-02-01T16:37:27 < dirty_d> im having nightmares about usb protocol 2013-02-01T16:37:38 < karlp> Steffanx: nice app, shareable? 2013-02-01T16:38:33 <+Steffanx> No, not really. Was just a 'proof of concept'. Wrote it using Qt and it's VERY VERY VERY ugly 2013-02-01T16:38:51 < zyp> dirty_d, I saw you were asking some questions tonight 2013-02-01T16:39:09 < dirty_d> yea 2013-02-01T16:39:14 < dirty_d> im not sure whats going on 2013-02-01T16:39:19 <+Steffanx> It has a timer that checks the clipboard 24/7. It has to do that because the clipboard changed events don't work when the application isn't activated :( 2013-02-01T16:39:23 < dirty_d> i fired up wireshark 2013-02-01T16:39:23 < zyp> 03:51:36 < dirty_d> ds2, mr usb expert, does get_descriptor happen before set_address? 2013-02-01T16:39:27 < zyp> 03:52:42 < ds2> no 2013-02-01T16:39:28 < zyp> that's wrong 2013-02-01T16:39:34 < dirty_d> i thought so 2013-02-01T16:39:40 < zyp> get_descriptor does indeed happen before set_address 2013-02-01T16:40:00 < dirty_d> i send the descriptor, and wireshark sees a malformed response packet 2013-02-01T16:40:02 < dirty_d> 0 bytes 2013-02-01T16:40:09 < zyp> http://bin.jvnv.net/f/CcVQl.png <- like this 2013-02-01T16:40:31 < dirty_d> i just learned this morning that the host sends a zlp to tell you that it got the descriptor though 2013-02-01T16:40:34 < dirty_d> i dont handle that 2013-02-01T16:40:39 < zyp> bear in mind that the first get_descriptor only asks for first 8 bytes, not all 18 bytes 2013-02-01T16:40:50 < dirty_d> but i dont think it would cause wireshark to show what its showing 2013-02-01T16:40:57 <+Steffanx> karlp, i also think Qt is a little overkill, but it works :) 2013-02-01T16:41:00 < dirty_d> zyp, what? 2013-02-01T16:41:12 < zyp> see the screenshot I posted 2013-02-01T16:41:13 < dirty_d> zyp, the length field is 64 for me 2013-02-01T16:41:32 < zyp> no, it's not 2013-02-01T16:41:37 < zyp> then you are misinterpreting it 2013-02-01T16:41:59 < dirty_d> hmm 2013-02-01T16:42:17 < dirty_d> i keep seeing a different order of things everywhere i look, lol 2013-02-01T16:42:35 < zyp> the device descriptor is always 18 bits, so no sane host would ask for more than 18 bits when requesting the device descriptor 2013-02-01T16:42:41 < dirty_d> its possible that i missed that first packet 2013-02-01T16:42:59 < zyp> bytes* 2013-02-01T16:43:15 < dirty_d> zyp, look at #8 http://www.lvr.com/usbcenum.htm 2013-02-01T16:43:21 < zyp> and the first time, it always asks for the first 8 bytes, because it does not know the max packet size of EP0 2013-02-01T16:43:42 < dirty_d> that makes sense 2013-02-01T16:44:00 < dirty_d> zyp, i am seeing a read/64 -32 error 2013-02-01T16:44:04 < zyp> but it's always at least 8, and it's stored in the eight byte of the device descriptor 2013-02-01T16:44:59 < zyp> this is point 8 in the page you linked to 2013-02-01T16:45:31 < dirty_d> it says it always requests 64 bytes 2013-02-01T16:45:55 < zyp> it says windows always requests 64 bytes 2013-02-01T16:46:14 < zyp> might be true, I've not done much testing with windows, so I don't remember 2013-02-01T16:47:11 < zyp> either way, get_descriptor should never return more bytes than host asked for, and never more bytes than size of the descriptor 2013-02-01T16:47:22 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-01T16:47:25 < zyp> so the smaller value of those is the one to use 2013-02-01T16:48:06 < zyp> http://cgit.jvnv.net/laks/tree/usb/generic.h#n74 <- refer to line 89 for a practical implementation of that 2013-02-01T16:48:59 < dirty_d> zyp, see anything wrong with my write()? http://pastebin.com/n021GQvK 2013-02-01T16:49:03 < dirty_d> read() works fine 2013-02-01T16:49:33 < dirty_d> yea, i was looking how you did that 2013-02-01T16:50:00 < dirty_d> even if im sending too much data, wouldnt that show up as too much data? now a 0 sized response to get_desc 2013-02-01T16:50:06 < dirty_d> not* 2013-02-01T16:51:34 < zyp> you are not setting the endpoint to ready, as far as I can see 2013-02-01T16:51:53 < zyp> ah, you do it outside the write 2013-02-01T16:51:59 < dirty_d> yea 2013-02-01T16:52:01 < zyp> that's a bit non-intuitive 2013-02-01T16:52:19 < dirty_d> yea, i guess thta doesnt really need to be separate 2013-02-01T16:52:42 < dirty_d> i wasnt sure if there was any case where i want some data to be ready there, but not send it yet 2013-02-01T16:54:07 < zyp> nah, sounds unlikely 2013-02-01T16:54:24 < zyp> hmm, I'm not sure your set_tx_status is sane either 2013-02-01T16:54:49 < dirty_d> as in i dont need to use a temporary variable? 2013-02-01T16:54:58 < dirty_d> i dont think i do 2013-02-01T16:55:06 < zyp> no, the part where you are using |= 2013-02-01T16:55:22 < zyp> it will or back bits that you don't want to or back 2013-02-01T16:56:07 < dirty_d> which ones? 2013-02-01T16:56:30 < dirty_d> youre always supposed to set ctr_rx/tx 2013-02-01T16:56:54 < dirty_d> ahh i think i see what you mean 2013-02-01T16:57:03 < zyp> STAT_RX 2013-02-01T16:57:15 < dirty_d> yea that is messed up 2013-02-01T16:57:19 < zyp> you'll or them back so the ones that are set will be toggled and cleared 2013-02-01T16:58:02 < dirty_d> yea 2013-02-01T17:06:40 < dirty_d> zyp, this is correct right? http://pastebin.com/DuGHsj75 2013-02-01T17:06:47 < dirty_d> i feel like it could be shorter 2013-02-01T17:12:33 < zyp> of course it could 2013-02-01T17:13:40 < zyp> epr = (epr & (EP_TYPE | EP_KIND | CTR_RX | CTR_TX | EA | STAT_TX)) ^ (status << STAT_TX_bp); 2013-02-01T17:14:49 < dirty_d> cant use &= to make it sorter? 2013-02-01T17:15:49 < dirty_d> epr &= (epr & ~USB::EPR::STAT_TX_bm) | ((epr & USB::EPR::STAT_TX_bm) ^ (status << USB::EPR::STAT_TX_bp)) 2013-02-01T17:16:13 < zyp> no. 2013-02-01T17:16:15 < dirty_d> nope 2013-02-01T17:16:34 < zyp> well 2013-02-01T17:16:37 < dirty_d> epr &= (epr & ~USB::EPR::STAT_TX_bm & ~USB::EPR::STAT_RX_bm) | ((epr & USB::EPR::STAT_TX_bm) ^ (status << USB::EPR::STAT_TX_bp)) 2013-02-01T17:16:46 < zyp> no. 2013-02-01T17:16:50 < karlp> yeah, good lukc anyone ever reading that code again. 2013-02-01T17:17:00 < karlp> can't you let g++ deal with shrinking it? 2013-02-01T17:17:08 < dirty_d> hmm? 2013-02-01T17:17:11 < zyp> epr &= EP_TYPE | EP_KIND | CTR_RX | CTR_TX | EA | STAT_TX; epr ^= status << STAT_TX_bp; 2013-02-01T17:17:16 < zyp> that would work 2013-02-01T17:18:02 < dirty_d> is that the same thing? 2013-02-01T17:19:04 < zyp> sure, as long as epr is a temporary and not the hardware register 2013-02-01T17:19:14 < dirty_d> it is 2013-02-01T17:19:33 < zyp> but with the previous I wrote, you don't need the temporary 2013-02-01T17:20:04 < dirty_d> ok 2013-02-01T17:20:16 < dirty_d> now i just need to wait 6 hours to go home and try it, lol 2013-02-01T17:20:57 < zyp> at work? :p 2013-02-01T17:21:08 < zyp> or, school perhaps 2013-02-01T17:21:54 < dirty_d> work at school 2013-02-01T17:22:03 < dirty_d> i work at the school i went to school to 2013-02-01T17:22:09 < zyp> ah, I see 2013-02-01T17:23:19 < dirty_d> bit fiddling in the morning 2013-02-01T17:23:31 < zyp> I'm about to go home from work myself 2013-02-01T17:23:39 < zyp> 16:23 here now 2013-02-01T17:24:41 < dirty_d> where is that? 2013-02-01T17:24:46 < zyp> norway 2013-02-01T17:24:52 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-01T17:25:02 < dirty_d> cool 2013-02-01T17:25:31 < zyp> «6 hours to go home» sounds like US east coast 2013-02-01T17:25:43 < dirty_d> yup, massachusetts 2013-02-01T17:25:44 <+Steffanx> So you WORK at a school dirty_d ? 2013-02-01T17:25:51 < dirty_d> yup 2013-02-01T17:25:56 <+Steffanx> So no student? 2013-02-01T17:25:57 < dirty_d> programmer 2013-02-01T17:25:59 <+Steffanx> Ah, ok :) 2013-02-01T17:25:59 < dirty_d> not anymore 2013-02-01T17:26:08 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-01T17:26:09 < dirty_d> its not fun programming 2013-02-01T17:26:13 < dirty_d> mostly oracle sql 2013-02-01T17:26:34 <+Steffanx> Never tried it, but it sounds awful :) 2013-02-01T17:26:40 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-01T17:26:40 < dirty_d> it is 2013-02-01T17:26:50 < dirty_d> not hard, just not interesting 2013-02-01T17:27:39 <+Steffanx> At least you have a job.. not everyone can say that in the US 2013-02-01T17:27:45 < dirty_d> true 2013-02-01T17:27:55 < dirty_d> i dont think its as bad as most people think though 2013-02-01T17:41:57 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-01T17:42:13 < dirty_d> zyp, this is fine also right? usb.EPR[ep] = (usb.EPR[ep] & EP_TYPE_bm & EP_KIND_bm & EA_bm) | CTR_TX_bm | CTR_RX_bm | ((epr & STAT_TX_bm) ^ (status << STAT_TX_bp)) 2013-02-01T17:42:50 < dirty_d> epr on the end should be usb.EPR[ep] 2013-02-01T17:47:20 < zyp> you're reading it twice then, I don't like that 2013-02-01T17:47:25 < zyp> but you don't have to 2013-02-01T17:48:21 -!- emeb [~ericb@72.201.78.226] has joined ##stm32 2013-02-01T17:48:32 < zyp> hmm, it's not valid either 2013-02-01T17:49:16 < zyp> rewrite it like this: usb.EPR[ep] = ((usb.EPR[ep] & (EP_TYPE_bm | EP_KIND_bm | EA_bm | STAT_TX_bm) ^ (status << STAT_TX_bp)) | CTR_TX_bm | CTR_RX_bm; 2013-02-01T18:03:20 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-163.pptp.stw-bonn.de] has joined ##stm32 2013-02-01T18:06:39 < Laurenceb> http://www.veracode.com/blog/2013/01/brogrammer-killed-the-requirements-engineering-star/ 2013-02-01T18:06:43 < Laurenceb> lol borgramming 2013-02-01T18:06:47 < Laurenceb> *bro 2013-02-01T18:11:20 < dirty_d> zyp, its not valid? 2013-02-01T18:11:54 < zyp> EP_TYPE_bm & EP_KIND_bm would become 0 2013-02-01T18:12:15 < zyp> you need to or them together first, then and them with the register 2013-02-01T18:12:27 < zyp> but I need to catch a train, talk to you later 2013-02-01T18:12:31 < dirty_d> ahh yea 2013-02-01T18:12:34 < dirty_d> ok thanks, later 2013-02-01T18:12:38 -!- dekar_ [~dekar@212.255.42.215] has quit [Quit: Leaving] 2013-02-01T18:12:41 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-01T18:15:12 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 276 seconds] 2013-02-01T18:16:51 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-01T18:17:36 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has quit [Ping timeout: 264 seconds] 2013-02-01T18:18:09 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Quit: pokðt] 2013-02-01T18:18:15 < baird> Laurenceb: mentioning Lamport, but not that he wrote LaTeX while mentioning his actual programming credentials? :/ 2013-02-01T18:19:07 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 260 seconds] 2013-02-01T18:21:16 < baird> "Hey, this is one of the main guys behind the 'literate programming' idea.." 2013-02-01T18:22:35 < baird> "He wrote the book, and the software that wrote the book.." :) 2013-02-01T18:23:38 < Erlkoenig> my friend (studies art history) asked my how to do page numbers in libre office. i said he should try LaTeX... :-> 2013-02-01T18:27:15 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-01T18:28:19 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Ping timeout: 248 seconds] 2013-02-01T18:28:24 <+Steffanx> poor friend 2013-02-01T18:29:11 < karlp> yeah, he has friends who recommend latex.... ;) 2013-02-01T18:29:19 < karlp> (and not the fun kind) 2013-02-01T18:31:26 < emeb> and remember - when pronouncing LaTeX, be sure to have lots of saliva in the back of your throat to get the right gargling noise on the final "X". 2013-02-01T18:34:43 < Laurenceb> emeb: are you using LDMIA r8,{r0,r2,r9} 2013-02-01T18:34:51 < Laurenceb> or whatever registers 2013-02-01T18:34:53 < emeb> no 2013-02-01T18:34:55 < Laurenceb> to load 2013-02-01T18:35:05 < Laurenceb> that should be faster 2013-02-01T18:35:09 < emeb> why? 2013-02-01T18:35:30 < Laurenceb> n+1 cycles 2013-02-01T18:35:37 < Laurenceb> where n is number of registers 2013-02-01T18:36:28 < emeb> here's the current source: https://github.com/emeb/f303_sdr 2013-02-01T18:36:54 < emeb> using ldrd to load 2 regs, already n+1 2013-02-01T18:37:28 < emeb> strangely, didn't make any difference in execution time over just loading 1 reg. 2013-02-01T18:37:43 < Laurenceb> hmm 2013-02-01T18:38:35 < Laurenceb> wtf 2013-02-01T18:38:44 < Laurenceb> yeah thats as fast as what i worked out 2013-02-01T18:38:55 < Laurenceb> im confused... 2013-02-01T18:39:06 < emeb> by...? 2013-02-01T18:39:22 < Laurenceb> i worked out something at was 3cycles/sample 2013-02-01T18:39:38 < Laurenceb> your code is 3/sample 2013-02-01T18:40:11 < Laurenceb> oh wait.. you seem to have I and Q at once 2013-02-01T18:40:19 < Laurenceb> ldrd r4, [r1, #0] /* get LO I1,I2,Q1,Q2 */ <-huh? 2013-02-01T18:40:45 < emeb> yeah, what's wrong with that? 2013-02-01T18:41:06 < Laurenceb> whats the format of the LO data? 2013-02-01T18:41:13 < Laurenceb> int16_t ? 2013-02-01T18:41:18 -!- fxd0h [~fx@186.18.172.240] has quit [Ping timeout: 252 seconds] 2013-02-01T18:41:27 < emeb> it's interleaved 16-bit 2013-02-01T18:41:30 < dirty_d> you guys are crazy 2013-02-01T18:41:36 < Laurenceb> :P 2013-02-01T18:41:47 < dirty_d> old fangled assembly code 2013-02-01T18:41:55 < emeb> dirty_d: funny - that's what I was thinking when looking at your USB stuff. 2013-02-01T18:41:59 < dirty_d> lol 2013-02-01T18:42:01 < Laurenceb> _FAST_ asm cdoe 2013-02-01T18:43:09 < emeb> Laurenceb: look in amrx.c - that's where I build the LO array. 2013-02-01T18:43:16 < Laurenceb> oh i didnt know about ldrd 2013-02-01T18:43:25 < Laurenceb> thats a fast way to do it 2013-02-01T18:44:19 < Laurenceb> so you load R4 and r5 ? 2013-02-01T18:44:22 < emeb> ya 2013-02-01T18:44:26 < Laurenceb> nice 2013-02-01T18:44:41 < Laurenceb> ok, your code is just as fast as anything i'd thought up :P 2013-02-01T18:44:52 < emeb> whew! :) 2013-02-01T18:45:14 < Laurenceb> i was thinking of ldmia with three registers 2013-02-01T18:45:21 < Laurenceb> if you had that many spare... 2013-02-01T18:45:28 < Laurenceb> but its no faster than ldrd 2013-02-01T18:46:23 < emeb> only way to make it faster would be to interleave the LO and RF data 2013-02-01T18:46:45 < emeb> then load it all w/ one ldmia 2013-02-01T18:46:55 < emeb> but of course that's impossible with ADC DMA. 2013-02-01T18:49:48 < Laurenceb> yeah 2013-02-01T18:50:25 < emeb> did some PC-based tests on the DSP last night 2013-02-01T18:51:27 < emeb> looks like this approach gives ~50dB out-of-band rejection. Not great, but passable for first try. 2013-02-01T18:51:33 < Laurenceb> you could envelope your loopup 2013-02-01T18:51:45 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-01T18:51:59 < emeb> huh? 2013-02-01T18:52:03 < Laurenceb> http://upload.wikimedia.org/wikipedia/commons/0/07/Window_function_%28blackman-harris%29.png 2013-02-01T18:52:15 < Laurenceb> stick your lookup through that window 2013-02-01T18:52:22 < Laurenceb> but youd really need a longer lookup 2013-02-01T18:52:23 < emeb> Oh yeah - added a window in the PC testing last night. 2013-02-01T18:52:44 -!- daku is now known as DaKu 2013-02-01T18:52:45 < emeb> code here -> https://github.com/emeb/f303_sdr/blob/master/support/test_amrx.c 2013-02-01T18:52:50 < Laurenceb> this would be a nice project for ham bands 2013-02-01T18:53:05 < Laurenceb> nice 2013-02-01T18:53:11 < emeb> with the current sample rate it might work on 160m 2013-02-01T18:53:13 < Laurenceb> id have tried to cheat with matlab 2013-02-01T18:53:32 < Laurenceb> psk and stuff 2013-02-01T18:53:32 < jpa-> nice assembler 2013-02-01T18:53:35 < emeb> ordinarily I would too, but I wanted to test the actual C code. 2013-02-01T18:54:18 < emeb> but yeah - going to longer integrator would probably help a lot. 2013-02-01T18:54:55 < emeb> 64 samples now, 256 wouldn't be too much tougher to do and might get a few dB more selectivity. 2013-02-01T18:55:17 < Laurenceb> and tighten up the band edges 2013-02-01T18:55:23 < emeb> ya 2013-02-01T18:55:34 < emeb> 1k is probably not doable with the size of the CCM 2013-02-01T18:55:52 < Laurenceb> itd be very tight 2013-02-01T18:55:53 < emeb> the asm would grow > 8kB 2013-02-01T18:55:57 < Laurenceb> 512 might work 2013-02-01T18:56:23 < jpa-> you don't need to have it all unrolled, though 2013-02-01T18:56:33 < emeb> true 2013-02-01T18:56:35 < Laurenceb> oh and theres the offset twiddle code 2013-02-01T18:56:37 < Laurenceb> yeah 2013-02-01T18:56:38 < jpa-> a loop around 64 iterations will not hurt it much 2013-02-01T18:56:46 < Laurenceb> to get fractional frequencies 2013-02-01T18:57:05 < emeb> Laurenceb: fine tune is already in there. 2013-02-01T18:57:10 < Laurenceb> nice 2013-02-01T18:57:17 < Laurenceb> how is it implimented? 2013-02-01T18:57:30 < emeb> just a complex mult using a Sine LUT & NCO 2013-02-01T18:57:34 < Laurenceb> ah 2013-02-01T18:57:51 < Laurenceb> i was thinking of dithering the start address of the lookup 2013-02-01T18:57:55 < emeb> that runs at the decimated rate - makes almost no difference in execution time. 2013-02-01T18:57:58 < Laurenceb> that way there much less cpu time 2013-02-01T18:58:00 < Laurenceb> ok 2013-02-01T18:58:33 < emeb> the 6-th order IIR takes longer 2013-02-01T19:02:27 < emeb> probably should look if loading IIR code into CCM would help 2013-02-01T19:02:43 < emeb> it's ARM/CMSIS code & I didn't add the attributes for that. 2013-02-01T19:11:43 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Ping timeout: 246 seconds] 2013-02-01T19:11:44 -!- Steffann [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-01T19:11:44 -!- Steffann [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-01T19:11:44 -!- Steffann [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-01T19:11:46 -!- mode/##stm32 [+v Steffann] by ChanServ 2013-02-01T19:13:41 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-01T19:21:29 < emeb> heh - putting the IIR in CCM saved about 2us. 2013-02-01T19:22:27 < emeb> err.. no. Only about 1/2us. wrong scale on scope. Still, that's something... 2013-02-01T19:22:57 < Laurenceb> http://www.exitcertified.com/training-class/ipad-training-apple.html <-rofl 2013-02-01T19:22:59 <+Steffann> 2us .. per what? 2013-02-01T19:25:18 < Laurenceb> 15us ? 2013-02-01T19:25:23 < emeb> Steffann: every pass where the IIR is called, so 1/2us every 53us total 2013-02-01T19:25:31 <+Steffann> Ah, ok :) 2013-02-01T19:25:56 < Laurenceb> oh its not called every pass? 2013-02-01T19:26:05 < Laurenceb> i x 2013-02-01T19:26:30 < emeb> Laurenceb: the IIR is pipelined 4 samples, so running it on fewer than 4 samples wastes cycles. 2013-02-01T19:26:37 < Laurenceb> oh thats neat 2013-02-01T19:26:39 < emeb> I buffer up 4 samples and call it every 4th. 2013-02-01T19:26:43 < Laurenceb> yeah 2013-02-01T19:39:59 -!- tkoskine [tkoskine@kapsi.fi] has quit [Ping timeout: 255 seconds] 2013-02-01T19:40:00 < Laurenceb> aha 2013-02-01T19:40:08 < Laurenceb> found a issue with my actuator 2013-02-01T19:40:18 < Laurenceb> annoyingly this is hard to fix 2013-02-01T19:40:32 < emeb> not just a loose wire. 2013-02-01T19:40:38 < Laurenceb> im using a timer in toggle mode to drive a stepper controller 2013-02-01T19:40:54 < Laurenceb> then auto reload value controls speed 2013-02-01T19:41:09 < Laurenceb> but i have to use the preload stuff 2013-02-01T19:41:19 < Laurenceb> so that the timer can be updated at any time 2013-02-01T19:41:40 < Laurenceb> but this leads to lag when the motor is rotating slowly 2013-02-01T19:42:10 < Laurenceb> im struggling to think of a way to fix this... 2013-02-01T19:43:41 < emeb> ah - I saw the same thing when using a timer to generate a programmable clock freq. 2013-02-01T19:44:19 < Laurenceb> actually i have an idea 2013-02-01T19:44:20 < emeb> can't update the timeout value until the current one finishes. 2013-02-01T19:44:25 < Laurenceb> 1) read the timer 2013-02-01T19:44:33 < Laurenceb> 2) turn off reload buffering 2013-02-01T19:44:45 < Laurenceb> 3) load with timer value +2 or something 2013-02-01T19:44:52 < Laurenceb> 4) turn on buffering 2013-02-01T19:44:59 < Laurenceb> 5) load with the new period 2013-02-01T19:45:19 < Laurenceb> that should work.. providing another thread doesnt get called in the middle 2013-02-01T19:45:29 < emeb> there's the rub. 2013-02-01T19:45:58 < Laurenceb> i think i can do it from inside an ISR 2013-02-01T19:46:33 -!- barthess [~barthess@77.67.196.38] has joined ##stm32 2013-02-01T19:46:38 -!- tkoskine [tkoskine@kapsi.fi] has joined ##stm32 2013-02-01T19:47:55 < Laurenceb> it gives an extra edge, but its better than waiting for ~200ms for the timer to overflow 2013-02-01T19:48:02 < Laurenceb> if the speed is very low 2013-02-01T20:08:03 < Laurenceb> http://www.spiegel.de/static/360grad/kamtschatka/ 2013-02-01T20:13:03 < Erlkoenig> spiegel, the former news magazine 2013-02-01T20:15:57 < emeb> oh quit complaining. that's cool. 2013-02-01T20:15:59 < emeb> :) 2013-02-01T20:17:03 < Erlkoenig> well they got that title from hacktivists 2013-02-01T20:18:27 < emeb> I remember we used to have old issues of Spiegel in our high-school german class. It was interesting, but sometimes a bit fluffy. 2013-02-01T20:19:45 < Erlkoenig> yeay people learn german \o/ 2013-02-01T20:20:08 <+Steffann> Erlkoenig yeah, now it's your turn to learn dutch 2013-02-01T20:20:31 < dirty_d> whats an easy way to see your code size? 2013-02-01T20:20:41 < emeb> Plenty of folks take german here. Americanos aren't all blinkered idiots. :) 2013-02-01T20:21:13 < dirty_d> arm-none-eabi-size main.elf seems to do the trick 2013-02-01T20:21:25 <+Steffann> I just typed that :P 2013-02-01T20:21:52 <+Steffann> Dont answer your own questions :) 2013-02-01T20:22:16 < emeb> or rather, answer your own questions before asking them out loud. :) 2013-02-01T20:22:18 < Erlkoenig> hum i am very bad at languages... english and french should suffice .D 2013-02-01T20:22:23 <+Steffann> non 2013-02-01T20:22:30 < Laurenceb> https://raw.github.com/Laurenceb/Chibi-Spectro/master/tests/stepper/log36-crop.png 2013-02-01T20:22:32 <+Steffann> French is the worst language 2013-02-01T20:22:34 < Erlkoenig> although i forget everything french :D 2013-02-01T20:22:36 < Laurenceb> this is looking better 2013-02-01T20:22:46 < emeb> ooo - steppy! 2013-02-01T20:22:49 < zyp> I forgot the german I learned in school :p 2013-02-01T20:22:53 < Laurenceb> yeah 2013-02-01T20:23:03 < Laurenceb> at least its vaguely controlled 2013-02-01T20:23:09 <+Steffann> I know how it order patatoes in french :P 2013-02-01T20:23:12 < Laurenceb> with it knew what the steps were 2013-02-01T20:23:15 <+Steffann> and apples 2013-02-01T20:23:17 < Laurenceb> *I 2013-02-01T20:23:30 <+Steffann> and croissants of course 2013-02-01T20:23:32 < emeb> my problem is that I learned german when I was a toddler, and that's about as much as I remember now. 2013-02-01T20:23:49 < zyp> :) 2013-02-01T20:23:56 <+Steffann> uh emeb ? 2013-02-01T20:24:10 < emeb> Steffann: my mother is german - lived there for about a year when I was a kid. 2013-02-01T20:24:16 <+Steffann> aaah 2013-02-01T20:25:27 < Erlkoenig> hmm maybe i'll learn a tiny bit of chinese next term 2013-02-01T20:25:42 < emeb> I hear that's the language of the future! :P 2013-02-01T20:26:16 < Erlkoenig> that's why ;) 2013-02-01T20:26:45 < emeb> I for one welcome our new asian overlords... 2013-02-01T20:27:55 < Erlkoenig> a doc about the CCC on the radio... 2013-02-01T20:28:24 < dirty_d> what would be the fastest way for two stm32s to communicate? 2013-02-01T20:29:35 < Laurenceb> smoke 2013-02-01T20:30:02 < Erlkoenig> fsmc? :D 2013-02-01T20:30:17 < emeb> depends on which STM32 parts you're talking about. 2013-02-01T20:30:30 < emeb> some have faster interfaces than others. 2013-02-01T20:31:20 < emeb> but the fastest and most generic would probably be SPI. 2013-02-01T20:32:59 < jpa-> if you run them from the same clock, you can just DMA data directly into and out of GPIO port 2013-02-01T20:33:45 < jpa-> that will get you 16 bits per clock e.g. 144MB/s at 72 MHz, probably the fastest possible data rate 2013-02-01T20:34:06 < jpa-> (ok, maybe not, GPIO has max frequency below that) 2013-02-01T20:34:18 < jpa-> let's say every other clock :P 2013-02-01T20:34:31 < emeb> seems like synchronizing would be a nuisance. 2013-02-01T20:35:27 < jpa-> depends.. you could just transmit e.g. 0x0 0x0 0x0 0x0 0xFFFF to start and then strip that off in the receiving end 2013-02-01T20:35:48 < dirty_d> hmm 2013-02-01T20:35:51 < dirty_d> thta would be neat 2013-02-01T20:38:49 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-01T20:39:15 -!- _BJFreeman [~bjfree@234.sub-75-233-218.myvzw.com] has joined ##stm32 2013-02-01T20:39:50 -!- _BJFreeman is now known as BJFreeman 2013-02-01T20:39:57 < jpa-> another cool way would be to connect a SRAM to FSMC of both controllers, and then hook up the N_WAIT lines so that when the other accesses the ram the other sees it as busy 2013-02-01T20:40:10 < jpa-> that would be very easy to use in code, because it would be essentially a shared memory region 2013-02-01T20:40:39 < Laurenceb> https://raw.github.com/Laurenceb/Chibi-Spectro/master/tests/stepper/log36-crop.png 2013-02-01T20:40:46 < jpa-> it depends on your application if "a stream of data" or a "region of memory" is the easier way to communicate 2013-02-01T20:40:48 < Laurenceb> added backlash tracking 2013-02-01T20:40:54 < Laurenceb> dunno how much it helps 2013-02-01T20:41:41 < Laurenceb> i suspect maybe its overcorrecting 2013-02-01T20:42:10 < emeb> interesting how it's got more overshoot on the rising edges at the high end. 2013-02-01T20:44:11 < Laurenceb> thats viscoelasticity i think 2013-02-01T20:44:39 < Laurenceb> ill try adding a fudge factor to "underestimate" backlash 2013-02-01T20:44:43 < emeb> sesquipedalian prolixity! 2013-02-01T20:45:00 < Laurenceb> hopeful;y half way between the 2 behaviours will be better 2013-02-01T20:47:01 < Laurenceb> it still sucks 2013-02-01T20:47:13 < Laurenceb> should be possible to hit setpoints with a fraction of the error 2013-02-01T20:47:33 < Laurenceb> at least its no longer completely suxor 2013-02-01T20:50:44 < dirty_d> Laurenceb, what are you making? 2013-02-01T20:51:12 < Laurenceb> its a device to measure properties of tissue samples under applied pressure 2013-02-01T20:51:30 < Laurenceb> has to squash squashy stuff under a small arm 2013-02-01T20:52:17 < dirty_d> oh wow 2013-02-01T20:53:09 < Laurenceb> bbl 2013-02-01T21:01:46 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-01T21:11:43 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-01T21:25:36 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 240 seconds] 2013-02-01T21:34:29 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-01T21:37:54 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 264 seconds] 2013-02-01T21:42:51 < jpa-> hmph.. looks like seeed decided that my usb analyzer is not worth a prize :F 2013-02-01T21:43:14 < jpa-> i thought it would be a sure bet because there were only 2 entries, but i did know their rules allow them not to give out all the prizes 2013-02-01T21:44:03 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 248 seconds] 2013-02-01T21:45:20 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32 2013-02-01T21:46:29 < emeb> I feel your pain. 2013-02-01T21:46:45 < emeb> Entered a design in the TI Beaglebone Cape contest & ended up a loooooser. 2013-02-01T21:47:19 < jpa-> good thing that i didn't spend too much time on it.. though maybe if i had i might have won something :P 2013-02-01T21:47:29 < jpa-> did learn a lot of vhdl though 2013-02-01T21:47:36 < emeb> upside 2013-02-01T21:48:45 < jpa-> downside is that i really want to improve my logic analyzer also but now i don't feel like it :D 2013-02-01T21:49:23 < emeb> is your LA for sale somewhere? 2013-02-01T21:49:34 < jpa-> nah it's a program for the dso quad 2013-02-01T21:49:38 < emeb> right 2013-02-01T21:50:08 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds] 2013-02-01T21:50:12 < jpa-> should write some fpga code for it also, now it samples in software and it's limited to 500kSps 2013-02-01T21:50:22 < jpa-> should be easy enough to get up to 72 MSps on fpga 2013-02-01T21:50:46 < emeb> how many bits per sample? 2013-02-01T21:50:53 < ds2> think i need to be schooled by zyp a little bit 2013-02-01T21:51:14 < ds2> jpa-: is the dso quad ur project? 2013-02-01T21:51:45 < jpa-> ds2: no, i just write software for it because 1) i bought it and 2) the original software sucks 2013-02-01T21:52:13 < emeb> man - the schematic is a maze. 2013-02-01T21:52:20 < jpa-> emeb: it's 4 channels and for logic analyzer obviously 1 bit per channel (the ADC is 8 bits though) 2013-02-01T21:52:29 < ds2> jpa-: ah.. so you are one of the many improved firmwares out there? are you one of those that is selling the improved firmware? 2013-02-01T21:52:52 < jpa-> ds2: nope 2013-02-01T21:52:53 < ds2> the dso quad was very tempting but I got convinced into getting a bigger tek clone 2013-02-01T21:53:49 < emeb> I'm thinking about doing something with similar structure, but use a better FPGA & F4. That would be a good SDR platform. 2013-02-01T21:54:05 < ds2> emeb: low cost USRP? 2013-02-01T21:54:29 < emeb> ds2: sorta - wouldn't be as fancy of course. 2013-02-01T21:54:45 < ds2> emeb: what frequency range? 2013-02-01T21:54:55 < emeb> looked at HackRF but they use just a cheapie CPLD, not a real FPGA. 2013-02-01T21:55:41 < emeb> ds2: ADC sampling in the 50-100MHz range. Need to think about front-end though. 2013-02-01T21:56:13 < ds2> emeb: is it going to spit out audio or require a host to do that? 2013-02-01T21:56:30 < emeb> I'd probably put a codec on too so it could do audio. 2013-02-01T21:56:46 < ds2> now that'd somewhat interesting 2013-02-01T21:57:05 < emeb> yeah. Hook the F4 & FPGA together via FSMC. 2013-02-01T21:57:12 < ds2> the reason i got interested in the rtlsdr is the hope that I can add a dongle to a tablet and get little SW radio or a scanner on it 2013-02-01T21:57:15 < emeb> Use an F407 in the 100-pin pkg. 2013-02-01T21:57:35 < emeb> ds2: that would be cool. 2013-02-01T21:57:37 < ds2> is thefsmc big enough on the 100 pin pkg? 2013-02-01T21:57:52 < ds2> emeb: but I donno enough about the algs to hook it up to audio 2013-02-01T21:57:54 < emeb> ds2: on the 100pin parts the FSMC is 16-bit muxed Addr/Data. 2013-02-01T21:58:09 < ds2> aka intel style stuff? 2013-02-01T21:58:36 < emeb> ds2: kinda. some types of PSRAM support that format, but not most. 2013-02-01T21:58:43 < jpa-> for communicating with FPGA you often just want a stream and don't need mapped memory 2013-02-01T21:58:44 < emeb> It'd be easy to do w/ FPGA tho. 2013-02-01T21:58:49 < jpa-> therefore, you don't need addresses 2013-02-01T21:58:56 < emeb> jpa-: right. 2013-02-01T21:59:01 < ds2> finding a large PSRAM that supports it is a pain but if it is intel style, a 245 can fix that 2013-02-01T21:59:08 < ds2> or is t a 244? 2013-02-01T21:59:15 < emeb> someone else here was using the DCMI for SDR data input. 2013-02-01T21:59:40 < emeb> 245 is an 8-bit register. You'd need a couple for a full address latch. 2013-02-01T21:59:56 < ds2> they have 32 bit versions 2013-02-01T22:00:09 < emeb> not surprising. 2013-02-01T22:00:09 < jpa-> dcmi is nice in that it can buffer on the F4 side.. i guess the overhead may also be slightly smaller than with fsmc 2013-02-01T22:00:21 < ds2> i used them on the LCD interface as voltage xlators 2013-02-01T22:00:39 < emeb> jpa-: yeah. I'd probably hook both DCMI and FSMC up and decide down the road depending on application. 2013-02-01T22:00:57 < ds2> apparently the efm32's only offer the intel style bus 2013-02-01T22:01:04 < ds2> so the 245 is mandatory 2013-02-01T22:01:09 < emeb> blech. 2013-02-01T22:01:18 < ds2> unless you can find a distro that has intel style bus PSRAMs 2013-02-01T22:01:20 < jpa-> emeb: lots of routing there :) 2013-02-01T22:01:20 < fxd0h> hey guys , which oled voltage driver would you recommend? 2013-02-01T22:01:30 < ds2> micron lists them but no one seems to have them in stock 2013-02-01T22:01:34 < emeb> the FSMC in the 144pin pkg supports demuxed Addr/Data 2013-02-01T22:01:53 < ds2> yes, I am using that 2013-02-01T22:01:57 < emeb> jpa-: true, but a lot of DCMI and FSMC pins are shared. 2013-02-01T22:03:37 < emeb> hmmm - that ADC that the dso quad uses looks pretty decent. 2013-02-01T22:03:56 -!- koliqi [c0008071@gateway/web/freenode/ip.192.0.128.113] has joined ##stm32 2013-02-01T22:04:37 < emeb> only 8-bit, but that might be enough. 2013-02-01T22:04:50 < jpa-> it's not bad.. though they overclock it from 36 MSps to 72 MSps :P 2013-02-01T22:04:59 < jpa-> (in interleaved mode it can do 144 MSps) 2013-02-01T22:05:09 < emeb> I was wondering about that. 2013-02-01T22:05:16 < jpa-> (there is also a real 72 MSps version IIRC) 2013-02-01T22:05:32 < emeb> ya - 80MSPS version costs a few $$ more. 2013-02-01T22:05:37 < emeb> non-stock @ DK tho. 2013-02-01T22:07:00 < emeb> oh nice - fast T/H on the input w/ BW > 400MHz. 2013-02-01T22:07:14 < emeb> can do some decent IF sampling. 2013-02-01T22:12:15 < emeb> ds2: what tablet were you thinking for your RTL SDR experiment? 2013-02-01T22:14:26 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-01T22:16:42 -!- fiendie [fiendie@ipv6.leela.fiendie.net] has left ##stm32 ["WeeChat 0.3.2"] 2013-02-01T22:34:14 < Tectu> http://www.youtube.com/watch?v=5Y7eg0REXZM. 2013-02-01T22:49:07 <+Steffann> w00t random link 2013-02-01T22:49:40 -!- Laurenceb_ [~Laurence@86.143.186.156] has joined ##stm32 2013-02-01T22:49:43 < Laurenceb_> hi all 2013-02-01T22:49:50 <+Steffann> All just left 2013-02-01T22:49:58 < Laurenceb_> :( 2013-02-01T22:52:14 < jpa-> i came back 2013-02-01T22:52:19 < jpa-> just for you, laurie 2013-02-01T22:52:20 < Laurenceb_> http://www.bbc.co.uk/news/uk-21302925 2013-02-01T22:52:29 < Laurenceb_> literal rofl 2013-02-01T22:54:08 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer] 2013-02-01T22:54:31 -!- Thorn [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-01T22:54:32 -!- Thorn [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Changing host] 2013-02-01T22:54:32 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-01T22:57:57 <+Steffann> like i care Laurenceb_ 2013-02-01T22:58:00 < Laurenceb_> hmm back to stepper actuator fail 2013-02-01T22:58:28 -!- dirty_d [~adowning@209.213.71.70] has quit [Ping timeout: 245 seconds] 2013-02-01T22:59:12 -!- dirty_d [~andrew@76.118.112.158] has joined ##stm32 2013-02-01T23:00:04 -!- adowning_ [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-01T23:00:34 -!- adowning_ [~adowning@209-213-71-70.meganet.net] has quit [Client Quit] 2013-02-01T23:08:07 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32 2013-02-01T23:11:57 < emeb> success in failure 2013-02-01T23:14:10 < ds2> emeb: an OMAP4 or Tegra3 based on 2013-02-01T23:14:25 < ds2> emeb: be nice if it can happen on an OMAP3 based one 2013-02-01T23:17:09 < emeb> ds2: Andorrhoid or some small linux distro? 2013-02-01T23:17:21 < ds2> either 2013-02-01T23:17:36 < ds2> I have a small linux distro co-existing with that aboration 2013-02-01T23:18:00 < emeb> hmmm... Would have to be a newer version of droid - older ones didn't support OTG or host. 2013-02-01T23:18:30 < ds2> off hand - an Archos 80G9 seems ideal. it has a full size A connector 2013-02-01T23:18:42 < emeb> ideal. 2013-02-01T23:18:47 < ds2> but the Nexus 7 might work too..faster proc 2013-02-01T23:18:51 < emeb> does it run HS or FS? 2013-02-01T23:19:01 < ds2> HS, IIRC 2013-02-01T23:19:05 < emeb> dunno if RTL works FS... 2013-02-01T23:19:15 < ds2> the archos is a OMAP4 2013-02-01T23:19:22 < ds2> and I believe that port sits on the EHCI 2013-02-01T23:19:32 < ds2> well... all of them are HS capable...just how crappy is the HS 2013-02-01T23:19:43 -!- dirty_d [~andrew@76.118.112.158] has quit [Ping timeout: 245 seconds] 2013-02-01T23:20:14 < emeb> I've got an RTL working on beaglebone. Wouldn't be too hard to write code to do AM/FM/SSB/CW demod. 2013-02-01T23:20:46 < ds2> getting the RTL itself working on the USB level is easy (for me) 2013-02-01T23:20:53 < ds2> it is the demods that seems complex 2013-02-01T23:21:04 < emeb> ya. that's the easy part for me. :) 2013-02-01T23:21:22 < ds2> I'd be happy with a AM/NFM/WFM setup... SSB and CW would be a plus 2013-02-01T23:21:47 -!- DaKu is now known as daku 2013-02-01T23:21:49 < ds2> I have avoided gnu radio due to all the interconnects it needs 2013-02-01T23:21:59 < emeb> Yeah - too much for a tiny machine. 2013-02-01T23:22:13 -!- daku is now known as DaKu 2013-02-01T23:22:13 < emeb> I'd rather just do it all in a single cmd-line executable. 2013-02-01T23:22:20 < ds2> exactly! 2013-02-01T23:22:29 < emeb> base it off the current RTL test app. 2013-02-01T23:22:30 < ds2> and those chips on the tablets have NEON 2013-02-01T23:22:41 < ds2> u know alsa? 2013-02-01T23:22:48 < emeb> I've used it. 2013-02-01T23:22:59 < emeb> not too tough. 2013-02-01T23:23:29 < ds2> wonder how much harder is it to extract the MPEG stream from say ATSC 2013-02-01T23:23:41 < emeb> that's probably not doable. 2013-02-01T23:24:00 < emeb> I used to design DVB demods so I know a bit about what goes on there. 2013-02-01T23:24:23 < emeb> Can't squeeze that through the BW available in direct tuning mode on the RTL. 2013-02-01T23:24:27 < ds2> the extraction? 2013-02-01T23:24:31 < ds2> oh not enough info 2013-02-01T23:24:34 < emeb> right 2013-02-01T23:24:44 < emeb> only reliably runs at 1-2MSPS 2013-02-01T23:25:32 < ds2> so any signal smaller then about 500KHz of BW should work? 2013-02-01T23:25:40 < emeb> easily 2013-02-01T23:25:54 < emeb> WBFM is fine for example. 2013-02-01T23:26:37 < ds2> wifi or BW I think is within that range, isn't it? 2013-02-01T23:27:03 < emeb> probably not. 2013-02-01T23:27:15 < emeb> 802.11g is like 20MHz. 2013-02-01T23:27:35 < ds2> oh 2013-02-01T23:27:57 < ds2> a smart scanner app would still be nice 2013-02-01T23:29:32 < BrainDamage> wifi freq is also past tuner range 2013-02-01T23:29:39 < emeb> that too 2013-02-01T23:30:09 < BrainDamage> also, you could try to combine multiple tuners to acquire diff portions of the spectrum 2013-02-01T23:30:10 < BrainDamage> BUT 2013-02-01T23:30:23 < BrainDamage> you'll have to timestamp packets 2013-02-01T23:30:31 < ds2> or put a mixer in front 2013-02-01T23:30:36 < BrainDamage> usb2 has non deterministic latency 2013-02-01T23:30:46 < ds2> a mixer is easier 2013-02-01T23:31:00 < BrainDamage> a mixer doesn't magically increase adc bandwith 2013-02-01T23:31:00 < ds2> DBM + a ~2GHz LO 2013-02-01T23:31:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has quit [Remote host closed the connection] 2013-02-01T23:31:11 < emeb> transverter 2013-02-01T23:31:19 < ds2> have that bypassable with a FTDI chip 2013-02-01T23:31:54 < emeb> or, wire in an I2C port expander to the I2C bus on the RTL 2013-02-01T23:31:56 -!- BrainDamage [~BrainDama@78.12.233.1] has joined ##stm32 2013-02-01T23:32:09 < BrainDamage> sorry, I was saying 2013-02-01T23:32:10 < ds2> is the i2c bus accessible from the host? 2013-02-01T23:32:19 < BrainDamage> a mixer only rigidly shifts the spectrum 2013-02-01T23:32:24 < emeb> ds2: pretty sure it is - that's how they control the tuner chip 2013-02-01T23:32:39 < BrainDamage> your problem is 2-fold 2013-02-01T23:32:54 < BrainDamage> being outside tuner range means you have downconvert using a mixer 2013-02-01T23:33:01 < ds2> pick a LO so the image is outside of the pass band 2013-02-01T23:33:10 < BrainDamage> and being larger than adc bw means you'll have to sync multiple tuners 2013-02-01T23:33:15 < BrainDamage> the 2nd is not easy 2013-02-01T23:33:22 < emeb> ya 2013-02-01T23:33:28 < ds2> why do you need multiple tuners? 2013-02-01T23:33:43 < BrainDamage> because the adc bandwith is not enough to sample the whole channel 2013-02-01T23:33:48 < emeb> because wifi/bt is broader than the bw of the rtl 2013-02-01T23:33:49 < BrainDamage> you'll get only a tiny fraction 2013-02-01T23:33:53 < ds2> ohhhhh 2013-02-01T23:33:59 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has joined ##stm32 2013-02-01T23:34:03 < ds2> i wasn't even looking at the second problem 2013-02-01T23:34:16 < ds2> thought u were saying it is needed for the first problem 2013-02-01T23:34:29 < emeb> besides, if you want wifi/bt, better off just getting a cheap wifi dongle. 2013-02-01T23:34:37 < emeb> better performance/power 2013-02-01T23:34:43 < ds2> for WiFi maybe 2013-02-01T23:34:46 < ds2> for BT, that isn't an option 2013-02-01T23:34:59 < ds2> tehre are no cheap/free firmware that will give you a promiscious mode on BT 2013-02-01T23:35:08 < ds2> hence the ubertooth guys 2013-02-01T23:35:47 < ds2> but that's all wishful thinking 2013-02-01T23:35:48 < Laurenceb_> if you get one of the dongles with proper drivers theres tons you can do 2013-02-01T23:36:02 < Laurenceb_> you can get PHY level sniffing 2013-02-01T23:36:07 < ds2> Laurenceb_: the firmware for that is not free if you want to do itlegally 2013-02-01T23:36:19 < Laurenceb_> im not so sure 2013-02-01T23:36:39 < ds2> hell, right now, I'll even settle for hearing broadcast AM signals with a dongle on a tablet! 2013-02-01T23:36:52 < Laurenceb_> i investigated it for personnel tracking in dangerous environments 2013-02-01T23:36:56 < ds2> Laurenceb_: are you talking about the leaked CSR firmware from the sniffer company? 2013-02-01T23:36:56 < Laurenceb_> seems feasible 2013-02-01T23:37:00 < Laurenceb_> no 2013-02-01T23:37:06 < Laurenceb_> just look in the linux repos 2013-02-01T23:37:16 < ds2> Laurenceb_: BT or WiFi? 2013-02-01T23:37:23 < Laurenceb_> WiFi 2013-02-01T23:37:29 < ds2> Oh... wifi is not an issue 2013-02-01T23:37:31 < ds2> it is BT that is 2013-02-01T23:37:36 < Laurenceb_> ah 2013-02-01T23:37:51 < Laurenceb_> if you went with zigbee.... 2013-02-01T23:37:52 < ds2> wifi chips seems to let you do all that 2013-02-01T23:38:03 < ds2> if customers would pay for zigbee... 2013-02-01T23:38:06 < Laurenceb_> stm32w108 drivers can do PHY sniff 2013-02-01T23:43:10 -!- BrainDamage [~BrainDama@78.12.233.1] has quit [Remote host closed the connection] 2013-02-01T23:43:49 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has joined ##stm32 2013-02-01T23:48:47 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has quit [Remote host closed the connection] 2013-02-01T23:48:58 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] --- Day changed Sat Feb 02 2013 2013-02-02T00:03:34 -!- Steffann [~Steffanx@unaffiliated/steffanx] has quit [Quit: ….] 2013-02-02T00:03:57 -!- barthess [~barthess@77.67.196.38] has quit [Quit: Leaving.] 2013-02-02T00:04:07 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-02T00:04:10 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-02T00:14:40 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-02T00:15:21 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-02T00:19:19 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-02T00:22:20 -!- barthess [~barthess@5.100.200.218] has joined ##stm32 2013-02-02T01:00:49 -!- barthess [~barthess@5.100.200.218] has quit [Quit: Leaving.] 2013-02-02T01:01:37 -!- Ranewen2 [~Ranewen4@87.252.128.149] has joined ##stm32 2013-02-02T01:02:35 -!- Ranewen2 [~Ranewen4@87.252.128.149] has quit [Client Quit] 2013-02-02T01:02:57 -!- Ranewen2 [~Ranewen4@87.252.128.149] has joined ##stm32 2013-02-02T01:04:52 -!- Ranewen2 [~Ranewen4@87.252.128.149] has quit [Client Quit] 2013-02-02T01:05:51 -!- Lionhearted [~Ranewen4@87.252.128.149] has joined ##stm32 2013-02-02T01:29:12 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-02T01:31:22 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-02T01:57:09 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Stepper.c#L36 2013-02-02T01:57:36 < Laurenceb_> think i fixed the stepper stuff 2013-02-02T01:57:39 < emeb> works? 2013-02-02T01:58:38 < Laurenceb_> dunno, dont have hardwar ehere 2013-02-02T01:59:11 < emeb> heh - all software works until you actually test it. :) 2013-02-02T02:00:02 < Laurenceb_> im trying to adjust the step rate without introducing extra steps 2013-02-02T02:01:31 <+Steffanx> I see some spaces you forgot to remove there Laurenceb_ :P 2013-02-02T02:02:50 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-02T02:03:21 < emeb> seagull trolling - fly in, shit on things, fly away. 2013-02-02T02:06:42 < ds2> when did IEEE reversed its policy on being to sublease blocks of MAC addresses? 2013-02-02T02:12:29 * Tectu throws cookies into the room 2013-02-02T02:22:45 < emeb> OMG! crumbs! 2013-02-02T02:25:23 < gnomad> well, at least he didn't toss them... 2013-02-02T02:29:49 < Laurenceb_> emeb: found an antenni for you 2013-02-02T02:29:50 < Laurenceb_> http://www.radioarcala.com/nbspStation/TowersandAntennas/Tower7/tabid/358/Default.aspx 2013-02-02T02:30:07 < emeb> quite compact 2013-02-02T02:30:17 < emeb> I'm sure the neighborhood assn wouldn't mind that. 2013-02-02T02:31:10 * Laurenceb_ returns to watching lesbian vampires 2 2013-02-02T02:33:00 < Tectu> Laurenceb, link? 2013-02-02T02:34:24 < Laurenceb_> http://www.imdb.com/title/tt1020885/ 2013-02-02T02:35:25 < baird> whoa @ antenna 2013-02-02T02:38:54 -!- DaKu is now known as daku 2013-02-02T02:40:29 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-02T02:42:04 < dongs> goclones allocated to delivery. coming in <2 hours 2013-02-02T02:42:14 < Laurenceb_> goclones? 2013-02-02T02:42:48 < dongs> the $50 ebay gopro2s 2013-02-02T02:44:52 < Laurenceb_> oh, nice 2013-02-02T02:48:07 * Laurenceb_ checks his email for daily fail 2013-02-02T02:49:10 < Laurenceb_> someone wants me to do head impact assesment.. on a cardboard box in the back of a van 2013-02-02T02:53:40 < Laurenceb_> its kind of designed... for heads 2013-02-02T02:54:08 * Laurenceb_ zzz 2013-02-02T02:54:22 -!- koliqi [c0008071@gateway/web/freenode/ip.192.0.128.113] has quit [Quit: Page closed] 2013-02-02T02:57:33 -!- Lionhearted [~Ranewen4@87.252.128.149] has quit [Quit: Leaving] 2013-02-02T02:58:28 -!- Laurenceb_ [~Laurence@86.143.186.156] has quit [Ping timeout: 245 seconds] 2013-02-02T03:14:43 < BJFreeman> that is heady 2013-02-02T03:17:52 < dirty_d> ahhhhg 2013-02-02T03:22:29 < baird> "One of these things is not like the other..." http://www.imb.uq.edu.au/2009-imb-science-ambassadors 2013-02-02T03:29:03 < dirty_d> i need some way to debug this 2013-02-02T03:29:09 < dirty_d> im never gonna figure out whats going on 2013-02-02T03:31:13 < zyp> still usb? 2013-02-02T03:31:44 < ds2> zyp: question - which descriptor requests are done before set address? 2013-02-02T03:31:57 < zyp> the 8-byte get device descriptor request 2013-02-02T03:32:17 < ds2> that is the only one? 2013-02-02T03:32:19 < zyp> because it's the one that tells the host about EP0 MPS 2013-02-02T03:32:30 < zyp> well, it's the only one in OS X, at leasn 2013-02-02T03:32:33 < zyp> least* 2013-02-02T03:32:49 < zyp> different hosts are allowed to do different stuff 2013-02-02T03:33:15 < ds2> ok 2013-02-02T03:33:19 < zyp> you as a device should just comply with whatever the host asks you to :p 2013-02-02T03:33:49 < ds2> ok 2013-02-02T03:33:55 < zyp> http://bin.jvnv.net/f/CcVQl.png <- but this is how a normal enumeration looks with an OS X host 2013-02-02T03:35:05 < zyp> the four-byte get configuration is to get the length of the configuration descriptor before fetching the entire 2013-02-02T03:35:21 < ds2> interesting 2013-02-02T03:35:38 < zyp> wTotalLength is third and fourth byte of it ;) 2013-02-02T03:35:52 < zyp> it also does something similar with all the string descriptors 2013-02-02T03:36:04 < zyp> similar as in fetching length before fetching entire descriptor 2013-02-02T03:36:37 < dirty_d> zyp, yea still struggling 2013-02-02T03:36:55 < zyp> I assume it's because it's allocating buffers of the precise size, and then reading into them 2013-02-02T03:36:57 < dirty_d> i nitices your master laks_demo branch has the same errors as my code 2013-02-02T03:37:01 < dirty_d> but the dcd_acm works 2013-02-02T03:37:05 < dirty_d> but i cant see a damn difference 2013-02-02T03:37:12 < zyp> same errors? 2013-02-02T03:37:26 < dirty_d> malformed GET_DESCRIPTOR packet 2013-02-02T03:37:32 < dirty_d> im using wireshark 2013-02-02T03:37:45 < zyp> what do you mean by master? 2013-02-02T03:37:54 < dirty_d> git clone ... 2013-02-02T03:38:06 < zyp> umm, there is no master branch in repo 2013-02-02T03:38:10 < dirty_d> with no git checkout usb_cdc_acm 2013-02-02T03:38:25 < dirty_d> what do i get if i just git clone? 2013-02-02T03:38:36 < zyp> dunno, check for yourself 2013-02-02T03:38:42 < dirty_d> whatever i got from that didnt work 2013-02-02T03:38:54 < zyp> well, that's no surprise 2013-02-02T03:39:05 < zyp> aren't you working on stm32f3discovery? 2013-02-02T03:39:10 < dirty_d> yes 2013-02-02T03:39:24 < zyp> usb_cdc_acm is the only branch that is ported to F3 :p 2013-02-02T03:39:27 < dirty_d> oops 2013-02-02T03:39:57 < dirty_d> if i stick while(getTxStatus(ep) == USB::EPR::STAT_TX::VALID); after my write 2013-02-02T03:40:10 < dirty_d> the descriptor gets sent, and i see it in wireshark 2013-02-02T03:40:17 < zyp> usb_audio is for F4 discovery, and the other branches are for waveshare F4 board 2013-02-02T03:40:43 < dirty_d> but it still says Broken pipe (-EPIPE) (-32) 2013-02-02T03:40:51 < zyp> anyway, the malformed stuff sounds like a bug in wireshark 2013-02-02T03:41:07 < zyp> i.e. wireshark is telling you «I don't understand this» 2013-02-02T03:41:29 < zyp> so don't put too much faith in what wireshark tells you 2013-02-02T03:41:47 < dirty_d> yea but i can see tha the host keeps trying to get the descriptor when it says its malformed 2013-02-02T03:41:52 < zyp> I suggest you get some logging system in place on the device side, if you don't have it already 2013-02-02T03:41:55 < dirty_d> but when it doesnt, it only tries once 2013-02-02T03:42:13 < dirty_d> zyp, yea i was gonna ask you about that, how the hell do you get at your logs? 2013-02-02T03:42:33 < zyp> I just log into a buffer, then dump the buffer through gdb 2013-02-02T03:42:49 < dirty_d> that would make this a hell of a lot easier 2013-02-02T03:42:50 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out] 2013-02-02T03:43:17 < zyp> the logging is very low overhead, so I can leave it in constantly, and then just go and retrieve the logs when something bad happens 2013-02-02T03:43:34 < zyp> when the buffer is full, the start gets overwritten, so it always contains the latest activity 2013-02-02T03:43:45 < dirty_d> yea i saw that, its like a circular buffer 2013-02-02T03:43:52 < zyp> exactly 2013-02-02T03:44:01 < zyp> ring buffer log -> RBLog :p 2013-02-02T03:44:13 < dirty_d> so thats what it stands for 2013-02-02T03:44:22 < dirty_d> how do you do that with gdb? 2013-02-02T03:44:39 < zyp> http://cgit.jvnv.net/laks/tree/gdb_plugins/rblog.py <- with this 2013-02-02T03:45:22 < zyp> you could just print the buffer normally to get the contents, but my script also does printf style string formatting 2013-02-02T03:45:40 < dirty_d> yea 2013-02-02T03:46:24 < dirty_d> im gonna quit messing with this till i get t his working so i can actually see what the hell is happening 2013-02-02T03:46:36 < dirty_d> will this "just work" with openocd? 2013-02-02T03:46:52 < dirty_d> i havent done any kind of setup excet for using the config file that came with it 2013-02-02T03:47:23 < zyp> well, if you get gdb attached to openocd and has a gdb compiled with python support, it should work fine 2013-02-02T03:49:23 < zyp> oh, and you obviously have to have your code compiled with debugging symbols 2013-02-02T03:49:27 < dirty_d> yea 2013-02-02T03:49:32 < dirty_d> how do you attach it though? 2013-02-02T03:49:36 < zyp> I noticed it didn't have those when you sent me an elf 2013-02-02T03:49:38 < dirty_d> something remote #### 2013-02-02T03:49:54 < zyp> IIRC openocd defaults to port 3333 for gdb 2013-02-02T03:50:04 < zyp> so: target extended-remote :3333 2013-02-02T03:51:28 < dirty_d> i guess that worked 2013-02-02T03:51:41 < dirty_d> Remote debugging using :3333 0x00000000 in ?? () 2013-02-02T03:52:25 < zyp> great 2013-02-02T03:52:56 < zyp> remember debugging symbols 2013-02-02T03:53:05 < zyp> you need -ggdb in compiler flags 2013-02-02T03:53:27 < zyp> so gdb knows where the objects are stored in memory 2013-02-02T03:53:33 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-02T03:53:46 < dirty_d> i just added that 2013-02-02T03:54:12 < zyp> good 2013-02-02T03:54:31 < dirty_d> how do you get it to use that python extension? 2013-02-02T03:54:37 < zyp> source rblog.py 2013-02-02T03:54:57 < zyp> but I have a strong suspicion that your gdb is compiled without python support 2013-02-02T03:55:44 < dirty_d> i just ran that and it didnt do anything, which makes sense because I havent logged anything 2013-02-02T03:55:54 < zyp> oh, no error? 2013-02-02T03:55:57 < dirty_d> it would give an error if ther was no support right? 2013-02-02T03:55:59 < dirty_d> no error 2013-02-02T03:56:01 < zyp> good 2013-02-02T03:56:07 < zyp> then you've loaded the plugin 2013-02-02T03:56:22 < zyp> it creates a new command in gdb, called rblog 2013-02-02T03:57:03 < zyp> my logging object for usb is named usb_rblog in my code, so I dump it with «rblog usb_rblog» in gdb 2013-02-02T03:57:27 < zyp> it has to be instanced in your code before you can log to it, and then you use that name to dump it 2013-02-02T03:58:01 < zyp> http://cgit.jvnv.net/laks/tree/usb/generic.h <- instancing happens here 2013-02-02T03:58:23 < zyp> it's instanced with 256 entries and max 2 arguments 2013-02-02T03:59:00 < emeb> hola mundo! 2013-02-02T03:59:14 < dirty_d> hmm 2013-02-02T03:59:28 < zyp> also, I'm curious about where you got your gdb from 2013-02-02T03:59:35 < dirty_d> zyp, im opn archlinux 2013-02-02T03:59:41 < dirty_d> everything is bleeding edge and to the maxx 2013-02-02T04:00:00 < zyp> ah, you built it yourself? maybe a pkgbuild? 2013-02-02T04:00:18 < dirty_d> hmm, i called mine "log", but i get Python Exception No symbol table is loaded. Use the "file" command.: 2013-02-02T04:00:30 < zyp> yes 2013-02-02T04:00:35 < zyp> you have to load your elf in gdb 2013-02-02T04:00:37 < zyp> file foo.elf 2013-02-02T04:00:38 < dirty_d> nope, its built with python support in the default package 2013-02-02T04:00:50 -!- BJFreeman [~bjfree@234.sub-75-233-218.myvzw.com] has quit [Quit: had a good time] 2013-02-02T04:00:54 < zyp> umm 2013-02-02T04:00:59 < dirty_d> zyp, nice, it works 2013-02-02T04:01:01 < zyp> are you using normal x86 gdb? 2013-02-02T04:01:20 < zyp> or arm-none-eabi-gdb? 2013-02-02T04:01:40 < dirty_d> x86, i didnt think about that, but its working 2013-02-02T04:01:51 < zyp> ok, then it's probably ok 2013-02-02T04:02:07 < dirty_d> this is awesome 2013-02-02T04:02:14 < zyp> sure is 2013-02-02T04:02:24 < dongs> gdb, awesome? haha. 2013-02-02T04:02:36 < dirty_d> this is much better than moving an if statement around and looking to see if an LED is on or not 2013-02-02T04:02:41 < zyp> that logging class is possible the single most useful part of my library :p 2013-02-02T04:03:15 < baird> Needs to pulsate to a beat: http://www.coolproductsusa.com/shop/led-shower-head-with-temperature-sensor/ .. untz untz untz untz 2013-02-02T04:03:39 < dongs> haha 2013-02-02T04:03:42 < dirty_d> i feel bad for the guys that pioneered computer software 2013-02-02T04:04:25 < dirty_d> alright, just gotta log all these functions now 2013-02-02T04:06:37 < zyp> I'm considering a future extension to the rblog stuff by having a print server that can run in it's own thread and monitor the objects and send updates to some host, possibly via swo 2013-02-02T04:07:10 < zyp> so logs can be extracted without halting and dumping it 2013-02-02T04:07:39 < zyp> and since transfer to host will be async, it will still be fast enough to log in critical code 2013-02-02T04:08:10 < dongs> ya well, newsflash, normal commercial debugers can already do thsi 2013-02-02T04:08:39 < zyp> ok 2013-02-02T04:09:02 < zyp> what do you think takes me less time? working to earn enough money to buy commercial debugger or implementing it myself? :p 2013-02-02T04:09:05 < dongs> i havent really needed to do realtime log that couldn't be handled by just normal watch vars, for time-critical shit gpio+la works fine, for just checking values of shit at runtime without halting, normal realtime wachwindow is fine 2013-02-02T04:11:48 < zyp> also, this would not be limited to swo, but also normal uart or seperate ACM endpoint or whatever 2013-02-02T04:12:05 < zyp> so it could be used for normal logging outside a debug setting 2013-02-02T04:12:12 < qyx_> you are too enterprise oriented dongs 2013-02-02T04:12:34 < zyp> or even user reported shit 2013-02-02T04:15:47 < zyp> oh well, time for sleep 2013-02-02T04:15:55 < zyp> dirty_d, good luck 2013-02-02T04:21:32 < dirty_d> thanks 2013-02-02T04:22:53 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-02T04:26:36 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 264 seconds] 2013-02-02T04:34:23 -!- cjbaird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-02T04:34:24 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Read error: Connection reset by peer] 2013-02-02T04:42:59 < dongs> http://i.imgur.com/3thQamf.jpg < goclone2 2013-02-02T04:48:41 < dirty_d> whats with the sock 2013-02-02T04:49:35 < dongs> thats what i got 2013-02-02T04:51:59 < emeb> time to hit up ebay's complaint dept. 2013-02-02T04:53:15 < dongs> yas 2013-02-02T04:53:45 < dirty_d> dongle, no its not 2013-02-02T04:53:51 < Erlkoenig> what did you order? :D 2013-02-02T04:53:58 < cjbaird> $50 socks 2013-02-02T04:54:03 < emeb> pretty ballsy I guess. "Hi I'm selling a camera." "Oh, here's your socks!" 2013-02-02T04:54:35 -!- cjbaird is now known as baird 2013-02-02T04:55:07 < dirty_d> that actually happened? 2013-02-02T04:58:12 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-163.pptp.stw-bonn.de] has quit [Read error: Connection reset by peer] 2013-02-02T05:00:30 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-02T05:02:11 -!- emeb [~ericb@72.201.78.226] has left ##stm32 [] 2013-02-02T05:03:36 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-02T05:31:56 < baird> Commie Red: http://www.raspberrypi.org/archives/3195 (..worthing about the pricing?) 2013-02-02T05:34:08 < baird> *wondering 2013-02-02T05:50:53 < dirty_d> why is it not legal to import to usa? 2013-02-02T05:54:18 < baird> Emissions non-compliance (FCC, etc.) .. If it's anything like Australia, it is actually legal to buy them from overseas and mail them in personally. You just can't resell them. 2013-02-02T05:54:55 < baird> They're carrying-on about cost reduction, yet those worthless camera and lcd connectors are still there... 2013-02-02T05:58:01 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-02T05:59:02 < dirty_d> i thought the camera thing worked? 2013-02-02T05:59:25 < dirty_d> its cheap enough anyway 2013-02-02T05:59:43 < dirty_d> it does a lot for $35 2013-02-02T06:00:22 < dirty_d> mines a media center and wireless gateway 2013-02-02T06:00:27 < dirty_d> worth it 2013-02-02T06:04:43 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-02T06:06:16 < baird> I've got one of mine running NetBSD (and earlier, FreeBSD). Those are stable and useful enough now, at least headless. 2013-02-02T06:08:02 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-02T06:08:07 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-02T06:29:52 < gxti> hand-assembled QFN FTW 2013-02-02T06:47:54 < dirty_d> zyp's rblog is most useful 2013-02-02T07:08:53 < emeb_mac> gxti: what qfn did you put down? 2013-02-02T07:10:03 < gxti> 100mbit phy, 24 pin 4x4mm 0.5mm pitch 2013-02-02T07:11:20 < emeb_mac> teensy! 2013-02-02T07:11:49 < emeb_mac> paste + reflow? 2013-02-02T07:12:23 < gxti> no, tinned pcb and pads, drown in flux paste, run a bead of solder with the iron 2013-02-02T07:12:53 < gxti> i ended up putting it through the toaster oven anyway because it makes everything pretty 2013-02-02T07:13:08 < gxti> but the qfn looked pretty good before that even 2013-02-02T07:13:34 < emeb_mac> cool. I've wondered if that approach would work OK. 2013-02-02T07:14:15 < emeb_mac> I've soldered almost-qfn pkgs for xtals/ oscillators with solder/flux/iron with good results. 2013-02-02T07:16:33 < emeb_mac> main concern is that so many qfn pkgs have center/bottom exposed pad & that pretty much requires reflow. 2013-02-02T07:18:57 < gxti> yeah this is hardly dissipating anything so the pad doesn't matter 2013-02-02T07:31:31 < gxti> https://secure.flickr.com/photos/suigetsu/8437505440/sizes/o/in/photostream/ 2013-02-02T07:39:39 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-02T07:45:08 < emeb_mac> nice board 2013-02-02T07:45:15 < emeb_mac> is that your timeserver? 2013-02-02T07:46:51 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-02T07:47:16 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-02T07:48:37 < gxti> emeb_mac: yep 2013-02-02T07:48:50 < emeb_mac> so tiny! 2013-02-02T07:50:47 < emeb_mac> cute little power board too - the upside-down electrolytic made me do a double-take. 2013-02-02T07:58:32 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-02T08:18:39 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-02T09:10:30 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-02T09:45:40 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-02T09:45:58 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Client Quit] 2013-02-02T10:04:48 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-02T10:05:15 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-02T10:14:56 -!- dekar [~dekar@212.255.31.181] has joined ##stm32 2013-02-02T10:15:00 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-02T10:46:25 -!- claude is now known as Claude 2013-02-02T11:01:44 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-02T11:04:18 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] 2013-02-02T11:15:42 -!- barthess [~barthess@178.154.9.71] has joined ##stm32 2013-02-02T11:52:38 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-02T12:03:00 -!- daku is now known as DaKu 2013-02-02T12:32:55 -!- DaKu is now known as daku 2013-02-02T12:43:11 -!- Laurenceb_ [~Laurence@86.143.186.156] has joined ##stm32 2013-02-02T12:44:58 < zyp> dongs, awesome camera 2013-02-02T12:45:14 < dongs> yaman 2013-02-02T12:45:56 < zyp> they even saw you were living in japan, so they sent you one that you could put on your foot to take upskirt vids! 2013-02-02T12:48:16 < Laurenceb_> fapan 2013-02-02T12:49:55 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has joined ##stm32 2013-02-02T13:01:50 < Thorn> found a waifu for jpa- http://windytan.blogspot.fi/2012/11/the-sound-of-dialup-pictured.html 2013-02-02T13:09:12 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-02T13:12:53 -!- daku is now known as DaKu 2013-02-02T13:20:40 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-02T13:20:43 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-02T13:28:24 < dongs> wat 2013-02-02T13:30:15 <+Steffanx> pizza 2013-02-02T13:56:44 < jpa-> Thorn: whats a waifu? 2013-02-02T14:04:12 <+Steffanx> you 2013-02-02T14:08:32 < Viper168_> http://www.facebook.com/thorthethundercat some caturday goodness of my little goofball, like and spread if you like, I'd like to see him get thr exposure he deserved so for being so ridiculous and awesome 2013-02-02T14:09:25 <+Steffanx> Get your facebook crap out of here! 2013-02-02T14:10:04 < Viper168_> lol 2013-02-02T14:10:17 < Viper168_> he's the only justification for facebook 2013-02-02T14:10:34 < Viper168_> it'd a handy way to post him and comments etc... all together 2013-02-02T14:10:53 < Viper168_> instead oftumblr etc.. which doesn't present images well 2013-02-02T14:11:28 < Viper168_> if you see him and don't hate cats, you can't help but understand 2013-02-02T14:11:57 <+Steffanx> only when you have an account, so get out of here with that crap :P 2013-02-02T14:12:35 < Viper168_> no account required to check it out 2013-02-02T14:12:56 < Viper168_> facebooks is dumb i agree 2013-02-02T14:12:59 < Viper168_> -s 2013-02-02T14:13:12 < Viper168_> but for something silly like this it's fine 2013-02-02T15:21:51 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-02T15:36:28 -!- DaKu is now known as daku 2013-02-02T15:42:05 <+Steffanx> heh, it works like a 'charm' it gets a little annoying though. http://share.naffets.nl/a-20130202-144144.png :D 2013-02-02T15:44:56 < Tectu> zyp, when you use c++ on your STM32, do you have the STL and everything there? Isn't that a bit big? 2013-02-02T15:45:23 < jpa-> stl is somewhat big, but if you choose the parts wisely it's not horrible 2013-02-02T15:45:58 < jpa-> libstdc++ vectors etc. basic stuff are somewhere 10-20kB.. but iostreams would bring in locale support and go over 50kB 2013-02-02T15:46:10 < jpa-> uclibc is much smaller if you are fine with the license 2013-02-02T15:46:26 < jpa-> ustl is small also but the author doesn't care for embedded stuff 2013-02-02T15:46:33 < Tectu> what about newlib? or is that c only? 2013-02-02T15:46:41 < jpa-> newlib is c only 2013-02-02T15:46:50 <+Steffanx> isn't that the man who thinks embedded stuff will all get replaceed my big arse computers jpa- ? 2013-02-02T15:47:03 < Tectu> Steffanx, lolwhat? 2013-02-02T15:47:04 < jpa-> Steffanx: something like that 2013-02-02T15:47:09 <+Steffanx> :D 2013-02-02T15:47:13 < Tectu> wtf= 2013-02-02T15:47:27 < Tectu> might someone explain? ^^ 2013-02-02T15:47:38 < Tectu> jpa-, so libstdc++ is fine to go with? 2013-02-02T15:47:43 <+Steffanx> perhaps jpa- still has the link to it 2013-02-02T15:48:16 < jpa-> Tectu: somewhat yeah.. i am using it at work on STM32F4 with 1MB flash and it's nothing 2013-02-02T15:48:32 < jpa-> it's somewhat large on the dso quad with 32 kB flash, but ok even there 2013-02-02T15:48:39 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has quit [Read error: Connection reset by peer] 2013-02-02T15:49:29 < jpa-> Steffanx: nope, apparently the ustl bug tracker has been removed 2013-02-02T15:49:46 <+Steffanx> aw 2013-02-02T15:49:58 < jpa-> ah no, just changed location 2013-02-02T15:50:22 <+Steffanx> Heh, when you google it your github is second result :D 2013-02-02T15:50:38 <+Steffanx> the magic of google 2013-02-02T15:51:00 < jpa-> i should probably remove that github repo because i'm not going to fix it 2013-02-02T15:51:30 < Tectu> so that guy... he thinks that embedded devices are getting replaced by arse computers? 2013-02-02T15:51:41 <+Steffanx> big arse even 2013-02-02T15:51:44 <+Steffanx> linux and crap :) 2013-02-02T15:51:56 < Tectu> okay please... explain 2013-02-02T15:52:17 <+Steffanx> Can't explain. Not my word 2013-02-02T15:53:22 < jpa-> found the reply in my mailbox: http://paste.dy.fi/VU8/plain 2013-02-02T15:54:06 <+Steffanx> anyway, it's still there 2013-02-02T15:54:09 <+Steffanx> http://sourceforge.net/p/ustl/feature-requests/7/ 2013-02-02T15:54:41 <+Steffanx> zlog remembers the link you posted.. and that url works 2013-02-02T15:54:41 < zlog> Steffanx: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2013-02-02.html 2013-02-02T15:54:47 -!- daku is now known as DaKu 2013-02-02T15:54:55 < jpa-> ah, i just fail at using a web browser 2013-02-02T15:55:05 < jpa-> web browsers are a dying breed anyway 2013-02-02T15:55:15 <+Steffanx> :P 2013-02-02T15:57:50 < Tectu> jpa-, that mail you got after a pull request? 2013-02-02T15:58:04 < jpa-> a feature request, but anyway 2013-02-02T15:58:11 < jpa-> but i do agree on his points 2013-02-02T15:58:19 < Tectu> that RTOS will die? 2013-02-02T15:58:23 < jpa-> not those 2013-02-02T15:58:28 < jpa-> but the ones on his code 2013-02-02T15:59:11 < jpa-> i was just totally surprised that someone would write a 20kB STL implementation and target it at desktop systems with > 4 GB memories 2013-02-02T15:59:20 <+Steffanx> Haha, yeah :D 2013-02-02T15:59:49 < Tectu> jpa-, I do the same 2013-02-02T15:59:57 < Tectu> I guess it's not really faster / better than STL? 2013-02-02T16:00:14 < jpa-> nope, it is slower but smaller 2013-02-02T16:00:23 < jpa-> and not entirely compatible either 2013-02-02T16:00:51 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-02T16:01:36 < Tectu> why does he do it then? 2013-02-02T16:02:01 < jpa-> he likes to write it 2013-02-02T16:02:52 < jpa-> and if his code crosses a 64kB boundary, the next night he will see a nightmare of 286 segments 2013-02-02T16:04:02 < jpa-> the sad part is that neither ustl or uclibc++ support c++11 2013-02-02T16:04:19 <+Steffanx> poor us 2013-02-02T16:05:10 < Tectu> jpa-, why don't you or zyp write one? I guess it's easy for guys like you? xD 2013-02-02T16:05:21 < jpa-> i don't care enough 2013-02-02T16:05:28 < jpa-> libstdc++ works fine enough for me 2013-02-02T16:05:52 < jpa-> (and it really would mean just extending and improving ustl, it is a really good base) 2013-02-02T16:07:15 < Tectu> okay 2013-02-02T16:07:23 < Tectu> I am seriously considering moving to c++ 2013-02-02T16:07:31 < jpa-> go ahead 2013-02-02T16:11:25 < Tectu> jpa-, you might help me out here... 2013-02-02T16:11:32 < Tectu> why to choose one over the other? Foo:Foo() : _bar(0) { } and Foo::Foo() { _bar = 0; } 2013-02-02T16:12:07 < jpa-> the first one may theoretically allow the compiler to make better code in some situations 2013-02-02T16:12:23 < jpa-> it also gives nice warning messages if you type the fields in different order than they are declared 2013-02-02T16:12:40 < Tectu> isn't the second one more a disadventage? 2013-02-02T16:12:46 < jpa-> then you can have fun time sorting a list of values 2013-02-02T16:12:58 < jpa-> no no, it makes programming more fun 2013-02-02T16:13:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-02T16:14:18 <+Steffanx> MORE? 2013-02-02T16:14:38 < jpa-> more fun than snuggling with Steffanx 2013-02-02T16:15:01 <+Steffanx> You were drunk last time, so how you remember? 2013-02-02T16:15:05 <+Steffanx> *the first and last time 2013-02-02T16:15:06 < Tectu> jpa-, just to be sure here _bar is in both cases an int 2013-02-02T16:15:09 <+Steffanx> *the only time 2013-02-02T16:15:35 < jpa-> Tectu: assumed that 2013-02-02T16:15:43 < jpa-> Tectu: for objects it does make more sense 2013-02-02T16:15:47 < Tectu> jpa-, okay 2013-02-02T16:16:05 < jpa-> Steffanx: i've never been drunk 2013-02-02T16:16:41 <+Steffanx> That's what she said 2013-02-02T16:17:17 < jpa-> in our snuggleship, you are the 'she' 2013-02-02T16:17:45 < Tectu> can I ask another c++ question here? I guess we all know that ##c and ##c++ aren't that cool ;-) 2013-02-02T16:17:50 <+Steffanx> Man, you remember all things wrong. You WERE drunk 2013-02-02T16:17:57 <+Steffanx> ##c is very cool 2013-02-02T16:18:04 < jpa-> we don't promise answers 2013-02-02T16:18:07 <+Steffanx> You can get banned within 5 seconds 2013-02-02T16:19:05 < Tectu> so here you go: http://codepad.org/n8qRwRWz 2013-02-02T16:19:27 < Tectu> isn't it possible to use a getter methode of a class within the class? 2013-02-02T16:19:43 < Tectu> the error is thrown in line 25 of the past 2013-02-02T16:19:44 < Tectu> paste* 2013-02-02T16:20:31 < jpa-> uh, what's the friend for? 2013-02-02T16:20:45 < Tectu> jpa-, that's what I am trying to understand next ;-) 2013-02-02T16:20:48 < jpa-> i would go for the simple bool operator<(const Genome &rhs) 2013-02-02T16:20:59 < jpa-> ah 2013-02-02T16:21:21 < jpa-> but you can also have a *static* friend method outside the class if you really want 2013-02-02T16:21:28 < Tectu> I'm reading some code of a very simple neural network 2013-02-02T16:21:33 < Tectu> and that lines are from there 2013-02-02T16:21:34 < jpa-> the error is just that the method has to be const 2013-02-02T16:21:55 < Tectu> the method has to be const? why? 2013-02-02T16:22:00 < jpa-> because you are trying to compare const instances apparently 2013-02-02T16:22:22 < Tectu> well, the getFitness() method or the operator overloading? 2013-02-02T16:22:22 < jpa-> if you don't use const, the compiler doesn't care either (but being const-correct is better for the environment) 2013-02-02T16:22:25 < jpa-> both 2013-02-02T16:22:29 < Tectu> oh 2013-02-02T16:22:35 < Tectu> does it even make sense to use const here? 2013-02-02T16:22:39 < jpa-> it does 2013-02-02T16:22:40 -!- FransWillem [~fw@53533629.cm-6-4a.dynamic.ziggo.nl] has quit [] 2013-02-02T16:22:41 < Tectu> I mean we don't overwrite anything? 2013-02-02T16:22:46 < Tectu> I mostly don't see the point of const 2013-02-02T16:22:47 < jpa-> i assume you don't 2013-02-02T16:23:02 < jpa-> some people don't see the point and are too lazy to use const 2013-02-02T16:23:13 < jpa-> other people see a lot of points and want every language to have const 2013-02-02T16:23:35 < Tectu> the I don't see the point was a request for a small explanation ^^ 2013-02-02T16:23:58 < jpa-> well you don't overwrite something accidentally so easily if you use const where you mean it 2013-02-02T16:24:43 < Tectu> okay 2013-02-02T16:24:48 < Tectu> but I can simply not overwrite it? 2013-02-02T16:24:50 < jpa-> (also it lets the compiler optimize a bit better) 2013-02-02T16:24:58 < Tectu> how comes that? 2013-02-02T16:25:17 < Laurenceb_> can i connect timer input signals to a gpio pin internally? 2013-02-02T16:25:22 < Laurenceb_> on F4 2013-02-02T16:25:33 < Laurenceb_> or will the gpio multiplex stuff not allow that? 2013-02-02T16:25:41 < jpa-> Tectu: http://www.parashift.com/c++-faq/const-correctness.html 2013-02-02T16:26:18 < jpa-> Laurenceb_: you mean connect timer input to an output GPIO? 2013-02-02T16:26:29 < jpa-> i think it should work 2013-02-02T16:27:06 < Tectu> jpa-, I'll read on that, thanks! 2013-02-02T16:29:04 < Laurenceb_> so gpio multiplexer is just to select the peripheral? 2013-02-02T16:29:11 < Laurenceb_> i see 2013-02-02T16:29:29 < jpa-> gpio multiplexer mostly affects outputs i think.. but not 100% sure 2013-02-02T16:31:20 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-02T16:32:02 < Laurenceb_> Figure 17 2013-02-02T16:32:12 < Laurenceb_> looks like the input block would need to be powered 2013-02-02T16:37:10 < zyp> Tectu, currently I'm not compiling in any standard libraries 2013-02-02T16:37:21 < Tectu> zyp, you have your own stuff? 2013-02-02T16:37:37 < Laurenceb_> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex_mx_stm32%2FAlternate%20Function%20Input%20on%20STM32F4&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B¤tviews=471 2013-02-02T16:37:38 < zyp> I'm only using stdint.h to get integer types, otherwise I'm not using anything 2013-02-02T16:37:39 < Laurenceb_> eww 2013-02-02T16:37:45 < Laurenceb_> but true? 2013-02-02T16:38:09 < Laurenceb_> i should read chibios source i guess 2013-02-02T16:38:43 < jpa-> Laurenceb_: oh, interesting 2013-02-02T16:40:31 < Laurenceb_> to be fair thats in the ref manual 2013-02-02T16:40:36 < Laurenceb_> just not put very clearly 2013-02-02T16:40:43 < Laurenceb_> looks like i can do what i want 2013-02-02T16:40:55 < Laurenceb_> i want to use encoder mode to track my stepper motor 2013-02-02T16:41:03 < Laurenceb_> without any goofy loopback wires 2013-02-02T16:41:38 < Laurenceb_> so i use pwm +gpio to drive the stepper controller, and have another timer in encoder mode using the states of those pins 2013-02-02T16:42:01 < jpa-> isn't there some internal connections between timers that you could use? 2013-02-02T16:42:21 < jpa-> hmm.. probably not for encoder more 2013-02-02T16:42:24 < jpa-> *mode 2013-02-02T16:42:54 < Tectu> let's say I have a class called Utils where I have handsom stuff like random number generators in there... one of the function is float Utils::randomClamped(); can't I know use that in another class like float foo = Utils::randomClamped() without generating an object of the Utils class? 2013-02-02T16:43:13 < jpa-> Tectu: if the method is static, yeah 2013-02-02T16:43:24 < jpa-> Tectu: but why bother with a class if you make it static? 2013-02-02T16:43:29 < jpa-> you can make just a normal function then 2013-02-02T16:43:37 < Tectu> jpa-, so in static float Utils::randomClamped() ? 2013-02-02T16:43:46 < Tectu> ah, don't do it as a class then? 2013-02-02T16:43:59 < jpa-> yeah, just write a normal function if you have no use for an object 2013-02-02T16:44:05 < Tectu> just utils.cpp -> float randomClamped() { }; ? 2013-02-02T16:44:11 < jpa-> yes 2013-02-02T16:44:24 < Tectu> okay 2013-02-02T16:44:27 < Tectu> but why the static thing? o0 2013-02-02T16:44:39 < jpa-> if you really wanted it to be in the class 2013-02-02T16:44:56 < Tectu> ah 2013-02-02T16:44:59 < jpa-> (for example, to be able to touch the private parts of objects of the class) 2013-02-02T16:45:00 < Tectu> okay, seems to work now. Thanks! 2013-02-02T16:45:09 < Laurenceb_> jpa-: yeah the slave mode stuff can tbe applied to encoder mode 2013-02-02T16:45:18 < Laurenceb_> well it can, but only for one of the signals 2013-02-02T16:45:23 < Laurenceb_> - encoder mode 3 2013-02-02T16:45:41 < Laurenceb_> and then its the direction signal, which comes from gpio in my case, so useless 2013-02-02T16:46:58 < jpa-> you could have a timer count the repetitions and get the lowest 2 bits of the position from the current state of the GPIOs.. but it's difficult to synchronize 2013-02-02T16:47:29 < jpa-> actually no need to check the GPIOs, just the current count of the other timer 2013-02-02T16:48:02 < jpa-> a normal "read high order; read low order; read high order again. overflow? repeat." could work there 2013-02-02T16:48:24 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-02T16:48:51 < Laurenceb_> yeah 2013-02-02T16:49:03 < Laurenceb_> atm im doing this in software 2013-02-02T16:49:17 < Laurenceb_> but its a little tricky to catch everything without errors 2013-02-02T16:49:34 < Laurenceb_> im thinking a hardware motor tracker will be easier and more accurate 2013-02-02T16:49:38 < jpa-> yet another way would be to use DMA for toggling the GPIOs.. then you would have one timer overflow per step => easy to count 2013-02-02T16:49:43 < Laurenceb_> and i can relax timing everywhere 2013-02-02T16:50:15 < jpa-> true, being able to use the encoder input would make it quite easy 2013-02-02T16:52:13 < Laurenceb_> "When the I/O port is programmed as output: The Schmitt trigger input is activated A read access to the input data register gets the I/O state" 2013-02-02T16:52:28 < Laurenceb_> so a peripheral can read it too? 2013-02-02T16:52:53 < jpa-> not sure if the AFIO stuff multiplexes the inputs or not 2013-02-02T16:53:11 < zyp> I suspect it does 2013-02-02T16:53:15 < jpa-> maybe it does, because otherwise the same line would be connected to input from multiple gpio 2013-02-02T16:53:19 < Laurenceb_> figure 17 2013-02-02T16:53:47 < Laurenceb_> you mean the GPIOx_AFRL[31:0]/whatever register? 2013-02-02T16:53:55 < Laurenceb_> i can set the correctly, not an issue? 2013-02-02T16:53:59 < Laurenceb_> *that 2013-02-02T16:54:39 < jpa-> but if the pin is not configured for AFIO it won't help 2013-02-02T16:54:52 < Laurenceb_> figure17 2013-02-02T16:54:58 < Laurenceb_> its not gated, unlike output 2013-02-02T16:55:23 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-02T16:55:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-02T16:55:32 < jpa-> consider: Foo peripheral pin goes to PA1 and PA5, selectable; if it would be connected to PA1 when not in AFIO mode, then what happens when you actually want to connect it to PA5? 2013-02-02T16:56:15 < Laurenceb_> i think you are confused with remap? 2013-02-02T16:56:27 < jpa-> remap = afio on stm32f4, right? 2013-02-02T16:57:54 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-02T16:58:36 < Laurenceb_> really.... 2013-02-02T17:00:15 < Laurenceb_> oh i seee 2013-02-02T17:00:42 < Laurenceb_> but i dont see how that would stop you blowing something up 2013-02-02T17:01:19 < Laurenceb_> the AFIO registers will let you do it whatever the case 2013-02-02T17:01:26 < jpa-> hmm yeah 2013-02-02T17:01:32 < jpa-> there has to be some mechanism against that 2013-02-02T17:01:40 < jpa-> maybe a priority encoder in the multiplexer 2013-02-02T17:02:10 < jpa-> .. in which case it just might not be gated and instead the AFRL is all that matters 2013-02-02T17:02:17 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-02T17:02:47 < Laurenceb_> someone needs to try this... 2013-02-02T17:02:48 -!- barthess [~barthess@178.154.9.71] has quit [Read error: Connection reset by peer] 2013-02-02T17:03:29 < Laurenceb_> "I originally had corresponding GPIOs configured with MODER[1:0]='00' (Input mode) and the appropriate AFX bits set in the corresponding AFRL/AFRH to connect the timer peripheral to the correct GPIO pins. This does not work. I had to select MODER[1:0]='10' for it to work." 2013-02-02T17:03:37 < Laurenceb_> so doesnt work :( 2013-02-02T17:04:30 < Tectu> jpa-, am I supposed to do operator overloading in the public: ? 2013-02-02T17:04:48 < Laurenceb_> loopback wire time then i guess :-/ 2013-02-02T17:05:42 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] 2013-02-02T17:06:27 < jpa-> on the other hand, on STM32L1 (which has the same AFRL stuff), for example the USB peripheral requires you to configure pins as "INPUT" instead of "ALTERNATE" 2013-02-02T17:06:42 < jpa-> Tectu: yes 2013-02-02T17:10:49 < zyp> jpa-, really? on F3 I'm setting them to AF14 2013-02-02T17:11:39 < jpa-> https://svn.kapsi.fi/jpa/sphere2/src_tx/board.h 2013-02-02T17:11:52 < jpa-> i set the AFRH but it didn't work with PIN_MODE_ALTERNATE 2013-02-02T17:11:55 < zyp> on F1 they are apparently forced to USB regardless of what they are configured to when USB is enabled 2013-02-02T17:12:18 < zyp> but setting them to AF doesn't hurt 2013-02-02T17:13:03 < zyp> well, I haven't really tested F3 without setting them to AF 2013-02-02T17:14:10 < jpa-> hmm.. but this being my first usb thingy there were so many problems that i can't be 100% sure if AF would have worked otherwise.. 2013-02-02T17:14:23 < jpa-> i don't remember the details 2013-02-02T17:14:39 < Tectu> jpa-, I'm trying to overload the < operator for the sort() function... It's the one I showed you before which does take two objects and now the compiler complains that this function is only allowed to take exactly one argument? 2013-02-02T17:14:48 < Laurenceb_> actually i can do this simply 2013-02-02T17:14:54 < zyp> Tectu, member function? 2013-02-02T17:15:02 < jpa-> Tectu: yeah, it should take only one argument 2013-02-02T17:15:02 -!- Viper168 is now known as Viper168_ 2013-02-02T17:15:05 < Laurenceb_> the gpio is set by the same thread that will read the position 2013-02-02T17:15:09 < jpa-> Tectu: compare 'this' to the argument 2013-02-02T17:15:12 < zyp> yeå 2013-02-02T17:15:14 < zyp> yep 2013-02-02T17:15:15 < Laurenceb_> so i can just slave a timer off the pwm 2013-02-02T17:15:17 < Tectu> ah 2013-02-02T17:15:20 < Tectu> hmm 2013-02-02T17:15:27 < zyp> if you make a normal function, it takes two arguments 2013-02-02T17:15:30 < Tectu> hwo the hell did the other guy make his code running then 2013-02-02T17:15:37 < Laurenceb_> then read the value and multiply by + - 1 2013-02-02T17:15:42 < zyp> just put it outside the class 2013-02-02T17:15:48 < Tectu> it does take two objects, makes return (A.getFoo() > B.getFoo()) 2013-02-02T17:16:15 < jpa-> it was a friend function.. not sure if that automatically makes it static 2013-02-02T17:16:15 < zyp> sure, because it's located outside the class. 2013-02-02T17:16:29 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-02T17:16:37 < jpa-> or rather, moves it out of the class even though it is written inside it 2013-02-02T17:16:55 < zyp> static would «move it out of the class» 2013-02-02T17:17:24 < Tectu> huh? move it out of the class? so in the header.h file just outside the class { }; block? 2013-02-02T17:17:27 < jpa-> yeah, but it would still be FooClass::function.. but i'm not sure if that matters 2013-02-02T17:17:57 < zyp> it's an operator, so it doesn't have a name anyway :p 2013-02-02T17:17:58 < jpa-> why does C++ even have out-of-the-class operator overloads? 2013-02-02T17:18:07 < zyp> why not? 2013-02-02T17:18:27 < jpa-> someone across the country subtly declares an operator for your class 2013-02-02T17:18:42 < zyp> sure 2013-02-02T17:18:49 < Tectu> so I make it outside the class and static and then it can take two arguments? o.O 2013-02-02T17:18:50 < Tectu> what? 2013-02-02T17:19:05 < jpa-> Tectu: two ways to make operators in c++ 2013-02-02T17:19:05 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-02T17:19:08 < jpa-> in class, out of class 2013-02-02T17:19:38 < zyp> jpa-, say you make some sort of number class, and want to compare it with regular ints 2013-02-02T17:20:00 < zyp> consider operator<(int, myint) vs operator<(myint, int) 2013-02-02T17:20:14 < jpa-> ah yeah, just found stackoverflow saying the same thing 2013-02-02T17:20:16 < jpa-> makes sense 2013-02-02T17:21:06 < zyp> whenever your operator takes several types, it's usually easier to define it as a global function and not a member function 2013-02-02T17:25:30 < Tectu> jpa-, I found out that the compiler dosen't complain about only one argument when I make the friend thing. Instead I only get error: no match for ‘operator<’ now 2013-02-02T17:25:40 < Tectu> just the paste: http://codepad.org/vJW6wVRV 2013-02-02T17:26:07 < jpa-> Tectu: the only problem with the original one was the missing consts 2013-02-02T17:27:25 < Tectu> jpa-, huh? 2013-02-02T17:27:29 < Tectu> jpa-, orginal one? 2013-02-02T17:28:33 < Tectu> jpa-, ah, wait 2013-02-02T17:29:54 < Tectu> jpa-, ah, the operator< is supposed to take two consts... and now I must make the getter() function const as well, righT? 2013-02-02T17:30:28 < jpa-> yeppity yep 2013-02-02T17:30:37 < Tectu> I see 2013-02-02T17:32:07 < Tectu> now it compiles :) 2013-02-02T17:32:13 < Tectu> at least that part.. 2013-02-02T17:32:20 < jpa-> therefore it works 2013-02-02T17:32:53 < Tectu> hmm, the c++ (g++) error and warnings seem to be quite more detailed and helpful than the gcc ones 2013-02-02T17:33:05 < Tectu> I'm just not understanding them always ^^ 2013-02-02T17:33:18 < jpa-> wait until you get the template warnings 2013-02-02T17:33:22 < jpa-> they are not helpful 2013-02-02T17:33:51 < Tectu> thats what I am currently dealing with... 2013-02-02T17:33:54 < Tectu> vector stuff.. 2013-02-02T17:34:05 < Tectu> algorithm.cpp:24:70: error: passing ‘const std::vector’ as ‘this’ argument of ‘void std::vector<_Tp, _Alloc>::push_back(const value_type&) [with _Tp = float; _Alloc = std::allocator; std::vector<_Tp, _Alloc>::value_type = float]’ discards qualifiers [-fpermissive] 2013-02-02T17:34:06 < Tectu> what? 2013-02-02T17:34:49 <+Steffanx> Tectu is having a hard time 2013-02-02T17:35:12 < jpa-> you can't push back in const vector, obviously 2013-02-02T17:35:47 < Tectu> that I agree, it's just that there's no const...aaaaaakahrdlhsdf!!!! 2013-02-02T17:35:58 < Tectu> why the fuck did I make that method const :D 2013-02-02T17:36:21 < Tectu> thanks for all your help, jpa :) 2013-02-02T17:38:28 < dirty_d> zyp, i got this mostly working with your logger 2013-02-02T17:38:34 < zyp> great 2013-02-02T17:39:12 < dirty_d> i was doing usb.EPR[ep] &= ~USB::EPR::CTR_TX_bm; to unset the flags 2013-02-02T17:39:20 < zyp> ah 2013-02-02T17:39:21 < dirty_d> obviously screwing up the rest of the register 2013-02-02T17:39:25 < zyp> yeah 2013-02-02T17:41:51 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-02T17:41:56 < dirty_d> are you just ignoring that det device desriptor qualifier request? 2013-02-02T17:50:12 < zyp> yes 2013-02-02T17:50:25 < zyp> it's only for high speed devices 2013-02-02T17:50:33 < zyp> full speed only devices should stall it 2013-02-02T17:50:46 < zyp> (just like they should stall any control command they don't know how to handle) 2013-02-02T17:52:33 < dirty_d> hmm, im not using stall at all 2013-02-02T17:52:55 < dirty_d> does stall just mean dont send anything, or it it an actual response? 2013-02-02T17:53:25 < jpa-> i have been confused why the ST's usb example stuff never stalls anything even when it is required by the standard 2013-02-02T17:53:49 < jpa-> stall is a separate usb packet IIRC 2013-02-02T17:54:10 -!- barthess [~barthess@77.67.204.38] has joined ##stm32 2013-02-02T17:54:55 < dirty_d> hmm 2013-02-02T17:57:34 < zyp> stall is an actual response token 2013-02-02T17:58:49 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-02T17:59:37 < dirty_d> i thought i read that you cant reposnd with a stall or nak to a setup request 2013-02-02T18:01:22 < zyp> well, that's sort of correct 2013-02-02T18:01:50 < zyp> you can't respond to a setup token with a stall token, you always have to accept the setup packet 2013-02-02T18:02:23 < zyp> but you are allowed to stall the data or status stages of a control request 2013-02-02T18:02:39 < dirty_d> ahh, ok 2013-02-02T18:04:37 < dirty_d> its gonna be wierd to use this with interrupts 2013-02-02T18:04:44 < zyp> a new setup token clears any present stall condition on EP0, this is done by hardware 2013-02-02T18:04:56 < dirty_d> ahh, thats helpful 2013-02-02T18:05:19 < zyp> so when you stall it, the stall condition only lasts until you receive a new control request 2013-02-02T18:06:19 < dirty_d> yea 2013-02-02T18:06:20 < dirty_d> thats nice 2013-02-02T18:06:25 < dirty_d> i was worried about it sticking around 2013-02-02T18:06:36 < zyp> if you picture a state machine, a new setup token would be interpreted as a reset to the «new request» state 2013-02-02T18:06:56 < dirty_d> yea 2013-02-02T18:07:07 < dirty_d> so it goes back to VALID? 2013-02-02T18:07:14 < zyp> yes 2013-02-02T18:11:57 < dirty_d> should it literally stall after it sends that request? 2013-02-02T18:12:12 < dirty_d> it sits there for a good second or two, then gets a malformed qualifier response 2013-02-02T18:12:25 < dirty_d> im thinking it shouldnt 2013-02-02T18:12:32 < dirty_d> since its a real response 2013-02-02T18:13:39 < zyp> don't trust wireshark 2013-02-02T18:14:13 < jpa-> yeah, wireshark can be quite misleading for debugging usb peripherals 2013-02-02T18:14:23 < jpa-> i think it works better if you are writing the driver on the computer side 2013-02-02T18:20:01 < dirty_d> yea 2013-02-02T18:31:51 < dirty_d> zyp, heres a log from your logger, am i not using stall right? http://codepad.org/ZtGJdjky 2013-02-02T18:34:07 < dirty_d> everything does literally stall for a few seconds after i set stall 2013-02-02T18:42:51 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has joined ##stm32 2013-02-02T18:43:28 < zyp> ah 2013-02-02T18:43:31 < zyp> set STAT_TX to stall 2013-02-02T18:44:25 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-148.pptp.stw-bonn.de] has joined ##stm32 2013-02-02T18:44:28 < dirty_d> hmm 2013-02-02T18:44:34 < zyp> the STALL is sent as a response to either an IN (TX) or OUT (RX) token from the host, and in the get_descriptor case, the host will send an IN token to receive data from you 2013-02-02T18:44:37 < dirty_d> and i dont even have to write anything? 2013-02-02T18:45:11 < dirty_d> ok that makes sense 2013-02-02T18:45:17 < zyp> no, when you set STAT_TX to stall, host gets a STALL token instead of a DATA token 2013-02-02T18:45:58 < zyp> the best option is probably setting both TX and RX to stall when you have to stall EP0 2013-02-02T18:46:16 < zyp> that way it will stall regardless of what the host is expecting 2013-02-02T18:46:43 < zyp> I'm currently only doing TX, since host usually wants to read stuff from you 2013-02-02T18:48:59 < dirty_d> ok 2013-02-02T18:49:12 < dirty_d> it doesnt hang anymore 2013-02-02T18:49:20 < dirty_d> the host does try 3 qualifier requests though 2013-02-02T18:49:26 < dirty_d> i guess that doesnt matter 2013-02-02T18:49:37 < dirty_d> then the config request works fine after 2013-02-02T18:50:08 < dirty_d> this codes getting too complex 2013-02-02T18:50:26 < dirty_d> im using like a state machine to determine what to do at each tx and rx 2013-02-02T18:51:27 < dirty_d> im not using a loop to send multiple packets 2013-02-02T18:51:52 < dirty_d> because when i start using interrupts for this, i dont want to be looping in the isr 2013-02-02T18:58:23 < Tectu> dongs, how are you? 2013-02-02T18:58:47 < zyp> asleep, I suspect 2013-02-02T19:00:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-02T19:00:15 < gxti> slacker 2013-02-02T19:00:39 < emeb_mac> sleep is overrated 2013-02-02T19:01:02 < zyp> 2am in japan currently, nothing wrong with being asleep 2013-02-02T19:01:20 < gxti> definitely a slacker. 2013-02-02T19:05:37 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-02T19:08:40 -!- emeb [~ericb@72.201.78.226] has joined ##stm32 2013-02-02T19:24:18 < dirty_d> zyp, i think im officially enumerated, [149618.140100] cdc_acm 4-1.2:1.0: ttyACM0: USB ACM device 2013-02-02T19:24:49 < dirty_d> i used your configuration descriptor, but i think im going to make something more dynamic like you said 2013-02-02T19:35:05 < zyp> :) 2013-02-02T19:35:18 < zyp> I'm interested in seeing what you come up with then 2013-02-02T19:41:05 < dirty_d> i dont think there is anything wrong with using a static descriptor actually 2013-02-02T19:41:22 < dirty_d> as long as you can query how many interfaces, endpoints etc and the properties 2013-02-02T19:41:46 < dirty_d> and have the usb class set itself up by reading it 2013-02-02T20:12:51 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving] 2013-02-02T20:48:28 -!- barthess [~barthess@77.67.204.38] has quit [Ping timeout: 245 seconds] 2013-02-02T20:53:33 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-02T20:56:45 -!- barthess [~barthess@178.154.33.44] has joined ##stm32 2013-02-02T21:15:43 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-02T21:18:42 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has quit [Ping timeout: 264 seconds] 2013-02-02T21:26:48 -!- Claude is now known as claude 2013-02-02T21:35:41 -!- DaKu is now known as daku 2013-02-02T21:36:55 -!- daku is now known as DaKu 2013-02-02T22:09:24 -!- barthess [~barthess@178.154.33.44] has quit [Remote host closed the connection] 2013-02-02T22:11:07 -!- barthess [~barthess@77.67.249.10] has joined ##stm32 2013-02-02T22:23:01 -!- mrcan_ is now known as mrcan 2013-02-02T22:23:36 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-148.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-02T22:33:53 -!- barthess [~barthess@77.67.249.10] has quit [Quit: Leaving.] 2013-02-02T23:20:08 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] 2013-02-02T23:55:00 < BrainDamage> dongs & Laurenceb_ http://www.wikihow.com/Become-an-Ultimate-Anime-Fan 2013-02-02T23:55:56 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-212.pptp.stw-bonn.de] has joined ##stm32 --- Day changed Sun Feb 03 2013 2013-02-03T00:08:44 <+Steffanx> zyp is the animé fan here baird 2013-02-03T00:08:49 <+Steffanx> * BrainDamage 2013-02-03T00:09:16 < gxti> he doesn't need help though, he is already ultimate anime master 2013-02-03T00:09:17 < BrainDamage> well, they are the wannabees 2013-02-03T00:09:27 < BrainDamage> so they need the link 2013-02-03T00:09:28 <+Steffanx> True truee 2013-02-03T00:18:15 -!- RuslanPopov [~rad@ppp-188-65-14-36-bras2.istra.ru] has joined ##stm32 2013-02-03T00:18:18 < RuslanPopov> re all 2013-02-03T00:19:26 -!- Steffann [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-03T00:19:26 -!- Steffann [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-03T00:19:26 -!- Steffann [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T00:19:29 -!- mode/##stm32 [+v Steffann] by ChanServ 2013-02-03T00:19:39 < RuslanPopov> ppl, how to setup 36kHz PWM on TIM17 (F100)? i kill three days reading docs/specs/apnotes and sites... 2013-02-03T00:21:14 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Ping timeout: 255 seconds] 2013-02-03T00:21:18 < gxti> dunno lol 2013-02-03T00:21:45 <+Steffann> thats not funny 2013-02-03T00:21:57 < gxti> you're right :( 2013-02-03T00:23:18 < gxti> RuslanPopov: what have you tried? 2013-02-03T00:23:57 < RuslanPopov> https://github.com/RaD/armka_demos/blob/master/apps/ir-tester/main.c - I did this 2013-02-03T00:24:13 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer] 2013-02-03T00:24:16 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32 2013-02-03T00:24:20 < RuslanPopov> but frequency is very high 2013-02-03T00:24:33 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Quit: Leaving] 2013-02-03T00:27:58 < gxti> RuslanPopov: how fast is your sys_clk, and what frequency do you see from tim17? 2013-02-03T00:28:31 < RuslanPopov> I want to know it too :) 2013-02-03T00:29:10 < gxti> never mind, obviously mcuconf.h is there too 2013-02-03T00:29:29 < gxti> still, how do you know it's "very high" without knowing what it is? 2013-02-03T00:29:46 < RuslanPopov> I have 8MHz crystal on board and have such PWM: https://twitter.com/ruslan_popov/status/297632622916542466/photo/1 2013-02-03T00:30:21 < RuslanPopov> I did simple IR receiver on mic input and recorded a signal with audacity 2013-02-03T00:31:04 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 256 seconds] 2013-02-03T00:31:43 < gxti> ok, so pll is 3x and thus sys_clk is 24mhz 2013-02-03T00:32:27 < RuslanPopov> where did you see pll x3? 2013-02-03T00:32:32 < gxti> mcuconf.h 2013-02-03T00:32:48 < gxti> STM32_PLLMUL_VALUE 3 2013-02-03T00:33:00 < gxti> and all the dividers are DIV1 except for ADC which isn't relevant here 2013-02-03T00:33:03 < RuslanPopov> ok, I see 2013-02-03T00:33:17 < gxti> SW selects PLL, PLL is 3x its input which is HSE, hse is 8mhz 2013-02-03T00:34:08 < RuslanPopov> ok, that is clear 2013-02-03T00:36:58 < RuslanPopov> ok, I understand the part up to SYSCLK on clock tree diagram now 2013-02-03T00:37:49 < RuslanPopov> it seems STM32_PPRE2 is a APB2 bus prescaler 2013-02-03T00:38:04 < RuslanPopov> so, APB2 works on 24 MHz too 2013-02-03T00:38:15 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-03T00:38:48 < gxti> your recording looks like about 4700hz, not sure where that's coming from 2013-02-03T00:38:56 < gxti> but it's hard to calculate from a screenshot 2013-02-03T00:39:23 < RuslanPopov> just random values to check does pwm work at all 2013-02-03T00:40:38 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-03T00:43:00 < RuslanPopov> as I see, I should tune the tim17 psc: 24MHz / 36kHz = 666 (holy shit) 2013-02-03T00:43:01 < baird> Steffann: https://www.youtube.com/watch?v=RrSS058pjWw 2013-02-03T00:44:42 <+Steffann> anime in tha army.. yay 2013-02-03T00:49:32 < gxti> RuslanPopov: yes 2013-02-03T00:53:50 < RuslanPopov> TIM17_ARR determines PWM frequency. After timer's prescaler I have 36kHz on the TIM17. What value should I assign on TIM17_ARR in this case? ARR=1? 2013-02-03T00:55:29 < gxti> RuslanPopov: if the goal is for the pwm period to be always 36khz then you should leave the prescaler off (= 0) and set ARR=666 2013-02-03T00:55:45 < RuslanPopov> yes, this is my case 2013-02-03T00:55:48 < gxti> if the prescaled clock is 36khz then you can't control the pwm 2013-02-03T00:55:58 < RuslanPopov> understand 2013-02-03T01:06:09 < RuslanPopov> so, psc=0, arr=665, ccr1=332 (50 % duty cycle) 2013-02-03T01:08:58 < gxti> yes. well, arr=666 because it's actually 666.666 and you round to 667 then subtract 1. 2013-02-03T01:10:40 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-03T01:10:56 < RuslanPopov> with these values I have got 16kHz pwm, hmm 2013-02-03T01:31:23 -!- Laurenceb_ [~Laurence@86.143.186.156] has quit [Ping timeout: 245 seconds] 2013-02-03T01:54:56 -!- Steffann [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-03T02:14:07 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-03T02:29:10 < dirty_d> zyp, there is some kinda constexpr problem with your configuration descriptor 2013-02-03T02:29:24 < dirty_d> im trying to figure it out 2013-02-03T02:35:17 < zyp> how so? 2013-02-03T02:36:03 < Tectu> how big is newlib? o0 2013-02-03T02:41:33 < dirty_d> zyp, im not sure the error is very confusing 2013-02-03T02:41:41 < dirty_d> ill put some code up 2013-02-03T02:41:45 < zyp> pastebin the complete error 2013-02-03T02:41:59 < dirty_d> its a few files 2013-02-03T02:42:01 < zyp> and the code with line numbers corresponding to the error 2013-02-03T02:42:20 < zyp> but anyway, let me see the errors first 2013-02-03T02:44:33 < dirty_d> http://codepad.org/t2hcQZPD 2013-02-03T02:44:47 < dirty_d> also, clang++ and g++ seem to disagreeo n whats allowed in a constant expression 2013-02-03T02:45:02 < dirty_d> clang++ doesnt allow reinterpret cast or casting to void * 2013-02-03T02:45:41 < zyp> pastebin test.cpp 2013-02-03T02:47:20 < dirty_d> i have better 2013-02-03T02:47:35 < dirty_d> http://cisweb.bristolcc.edu/~adowning/test.tar.bz2 2013-02-03T02:47:49 < dirty_d> g++ -std=c++11 -Wall test.cpp -o test && ./test 2013-02-03T02:48:28 < dirty_d> i made some constexpr classes to read your descriptors 2013-02-03T02:48:51 < dirty_d> well, it will work with any descriptor, it just takes a void * to a descriptor 2013-02-03T02:49:48 < dirty_d> the error goes away if i comment out the interface descriptors inside your configuration descriptor 2013-02-03T02:54:04 < zyp> ok, hang on 2013-02-03T02:54:29 < zyp> uh, gcc on this computer is too old :p 2013-02-03T02:55:17 < zyp> but ok, I have arm-gcc 4.7 2013-02-03T02:55:25 < zyp> so that's fine for just compiling 2013-02-03T02:55:29 < zyp> I see the error 2013-02-03T02:55:39 < zyp> need to look into 2013-02-03T02:56:17 < dirty_d> zyp, it seems like youre not allowed to cast in a constant expression 2013-02-03T02:56:27 < dirty_d> and gcc is behind 2013-02-03T02:56:37 < dirty_d> that seems kinda silly though 2013-02-03T02:59:09 < zyp> heh 2013-02-03T03:06:47 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-03T03:07:26 < R2COM> interesting so turns out that on any timer TIMx, i am limited to measurement of maximum 2 signals for input capture pulse width and period 2013-02-03T03:08:06 < dirty_d> i thought it was only 1 2013-02-03T03:08:12 < R2COM> so, if I want for example to measure pulse width of 4 different signals (only pulse width not period) which are connected to CH1,CH2,CH3,CH4 to same timer TIM(2-5) it is not possible 2013-02-03T03:08:30 < gxti> haven't we been over this already 2013-02-03T03:08:34 < dirty_d> the timer gets reset 2013-02-03T03:08:40 < dirty_d> so it would mess up the other channel wouldnt it 2013-02-03T03:08:46 < R2COM> it getes reset depending how you set it for 2013-02-03T03:08:51 < gxti> well you don't *have* to use the reset 2013-02-03T03:09:20 < R2COM> you do to get your measurement correct 2013-02-03T03:09:54 < gxti> negatory 2013-02-03T03:10:19 < R2COM> it gets reset from TRGI which follows after TI1FP1, and TI2FP2 2013-02-03T03:14:54 < R2COM> also here is another thing, if I want to measure pulse width only, I am forced to record that into CCR1 register 2013-02-03T03:15:21 < R2COM> because from block diagram TI2FP1 goes to CCR1 and TI1FP1 also goes to CCR1 2013-02-03T03:15:56 < R2COM> and to measure pulse width, you need and "inverted" mode... when for example after rising edge later it goes down, and creates a pulse of specific width 2013-02-03T03:16:08 < zyp> that's what the 1 in TI2FP1 means 2013-02-03T03:16:16 < zyp> otherwise it would be TI2FP2 2013-02-03T03:16:22 < dirty_d> zyp, im not sure what to do about this 2013-02-03T03:16:48 < zyp> dirty_d, me neither, but it looks like a bad idea anyway 2013-02-03T03:17:03 < R2COM> what I am trying to say is, user is forced to use TixFP1 to measure the pulse width 2013-02-03T03:17:04 < R2COM> thats it 2013-02-03T03:17:28 < gxti> pretty sure you can crosswire IC 1-2 to pins 1-2 as much as you want 2013-02-03T03:17:28 < zyp> R2COM, you are wrong 2013-02-03T03:17:40 < R2COM> because from CC1P bit field description of CCER register, it says that Inverted/falling edge sensitivity is from TIxFP1 2013-02-03T03:17:51 < R2COM> at least what written in reference manual 2013-02-03T03:17:57 < R2COM> TIM2-5 general purspoe timers 2013-02-03T03:18:32 < zyp> so what? what about CC2P? 2013-02-03T03:18:50 < R2COM> in that case it is TI2FP1 2013-02-03T03:18:57 < R2COM> and still linked to CCR1 2013-02-03T03:19:12 < R2COM> otherwise they would say TIxFPx 2013-02-03T03:19:26 < R2COM> if you look at p.419 of reference manual 2013-02-03T03:19:38 < zyp> which reference manual and which revision? 2013-02-03T03:19:55 < R2COM> borth TI2FP1 and TI1FP1 are linked to IC1 2013-02-03T03:20:05 < zyp> sure 2013-02-03T03:20:10 < R2COM> stm32f4, Rev 3 2013-02-03T03:20:16 < R2COM> latest 2013-02-03T03:20:22 < zyp> and both TI1FP2 and TI2FP2 are linked to IC2 2013-02-03T03:20:36 < R2COM> yes 2013-02-03T03:20:56 < dirty_d> zyp, whats a bad idea? 2013-02-03T03:21:39 < R2COM> but again, Ti2FP1 and TI1FP1 linked to IC1, and CC1P bit description for CCER register says that IOnverted/falling edge sensitivity is only on TIxFP1... so to measure pulse width you are forced to use TI(1 or 2)FP1 2013-02-03T03:22:18 < R2COM> it means, that you cannot for example connect two different signals to CH1 and CH2, and then in CCR1 hold pulse width only of CH1 and in CCR2 hold pulse width of only CH2 2013-02-03T03:22:43 < R2COM> (again not intersted in period) 2013-02-03T03:24:11 < dirty_d> R2COM, maybe youre better off just using an external interrupt and timing the signals yourself 2013-02-03T03:24:21 < dirty_d> unless you cna afford using multiple timers 2013-02-03T03:24:41 < R2COM> idea was to measure pulse width of many independent pins.. 2013-02-03T03:24:55 < zyp> R2COM, sure you can, if you just use your head a bit rather than arguing about your misinterpretation of the reference manual 2013-02-03T03:24:56 < dirty_d> for many you probably need to do it yourself 2013-02-03T03:24:59 < R2COM> I can afford, but it would not be enough, if what I described isnt possible 2013-02-03T03:25:04 < zyp> anyway 2013-02-03T03:25:17 < zyp> what is the nature of the signals you are trying to measure? 2013-02-03T03:26:21 < R2COM> I am figuring things out, reading and trying to understand, meanwhile I am trying combinations in code to, and sure I do use head you asshole 2013-02-03T03:26:38 < R2COM> and I was not arguing, I was discussing things 2013-02-03T03:27:31 < zyp> sorry, I didn't mean to offend you, let's focus on solving your problem 2013-02-03T03:27:42 < zyp> so, what is the nature of the signals? 2013-02-03T03:28:13 < R2COM> thats ok, I will have to go from scratch and try every possible combination anyhow, and signals are simple PWM like slow speed (with period 15ms maximum) 2013-02-03T03:28:36 < R2COM> (thats why I am not interested in measuring of period, because I know its 15ms) 2013-02-03T03:28:39 < zyp> rc servo style signals? 2013-02-03T03:28:42 < R2COM> yes 2013-02-03T03:28:59 < zyp> trying to decode a signal from a transmitter? 2013-02-03T03:29:05 < zyp> receiver rather 2013-02-03T03:29:27 < R2COM> I can easily measure pulse width, from transmitter, it works for me, but for that, I am utilizing CH2 as input, and then CCR1... 2013-02-03T03:29:35 < R2COM> which is not quite what I want 2013-02-03T03:30:04 < zyp> yes, I understand that 2013-02-03T03:31:18 < zyp> is your purpose to solve this problem for any receiver implementation, or are you allowed to make assumptions about a specific kind of receiver? 2013-02-03T03:31:38 < R2COM> for examp[le like this: http://pastebin.com/0bNmeys9 2013-02-03T03:31:47 < zyp> because receivers usually have patterns in how the channels are timed in relation to each other 2013-02-03T03:32:26 < dirty_d> i dont think thats true for modern digital ones 2013-02-03T03:32:28 < R2COM> the code below does period too (I can exclude that one if I want) 2013-02-03T03:32:37 < R2COM> above* 2013-02-03T03:32:38 < zyp> dirty_d, I'm talking about modern digital ones 2013-02-03T03:33:04 < dirty_d> from what i hear, the DXM2 etc ones output all the signals at the same time 2013-02-03T03:33:06 < gxti> there's always *some* pattern, since it's being generated programmatically. the question is whether you can rely on it. 2013-02-03T03:33:19 < R2COM> yes better not rely on it. 2013-02-03T03:33:56 < dirty_d> how many channels do you need to decode? 2013-02-03T03:34:02 < zyp> http://bin.jvnv.net/f/LzwZB.png <- this is the first four channels from my turnigy 9x receiver, and the NOR of all channels combined 2013-02-03T03:34:07 < R2COM> as input... hmm 5 2013-02-03T03:34:15 < dirty_d> how many outputs? 2013-02-03T03:34:20 < R2COM> 4 2013-02-03T03:34:48 < dirty_d> do you have 5 timers you can use and a 6th for output? 2013-02-03T03:35:17 < R2COM> 4 outputs are easy, I hooked them to TIM3, and all I have to do is modify CCR1-4 to make any output 2013-02-03T03:35:26 < dirty_d> yea 2013-02-03T03:35:47 < dirty_d> R2COM, which stm32 are you using? 2013-02-03T03:36:49 < R2COM> for now, the discovery f4 board 2013-02-03T03:36:52 < zyp> R2COM, just consider this, each channel can only measure the time of a rising edge or a falling edge, and if you want to know the difference between them (pulse width), you need to know both, so if you can't make assumptions you'll have to use two channels for each signal 2013-02-03T03:37:07 < R2COM> but if idea is doable with chip, i can route custom board its not big deal 2013-02-03T03:37:55 < R2COM> zyp: well okay, so turns out I was right with my conclusions from that reference manual? 2013-02-03T03:37:57 < zyp> so each timer would only be able to decode two if you can't make assumptions 2013-02-03T03:38:10 < R2COM> (yes lets say i cant make assumptions) 2013-02-03T03:38:15 < zyp> ok 2013-02-03T03:38:43 < zyp> then you just hook one to CH1 and one to CH3, then double-route both, so they also use CH2 and CH4 respectively 2013-02-03T03:38:47 < gxti> although 1ms is probably slow enough to cheat and swap the channel between positive and negative :p 2013-02-03T03:39:06 < zyp> since you can't reset timer, you'll just let it run freely 2013-02-03T03:39:15 < dirty_d> wait what? i thought each timer can capture period and duty cycle? 2013-02-03T03:39:25 < dirty_d> duty cycle is pulse width 2013-02-03T03:39:31 < R2COM> exactly I wont be able to reset timer 2013-02-03T03:39:33 < zyp> timers know nothing about periods or duty cycles 2013-02-03T03:39:43 < zyp> they just capture edge times 2013-02-03T03:39:43 < dirty_d> is f4 different than f3? 2013-02-03T03:39:46 < zyp> no. 2013-02-03T03:40:01 < gxti> dirty_d: using the canonical "pwm input" pattern, you can input one signal per timer 2013-02-03T03:40:08 < dirty_d> gxti, yea 2013-02-03T03:40:13 < zyp> gxti, but we don't use that 2013-02-03T03:40:15 < gxti> dirty_d: you can measure two signals though if you do more work in software 2013-02-03T03:40:32 < zyp> the only «more work» is a simple subtraction 2013-02-03T03:40:35 < dirty_d> R2COM, how many spare 16 bit timers do you have that can do input capture? 2013-02-03T03:40:41 < gxti> zyp: well, you have to deal with wraparound 2013-02-03T03:40:48 < zyp> no 2013-02-03T03:40:55 < gxti> ok whatever 2013-02-03T03:41:00 < zyp> just subtract signed 2013-02-03T03:41:16 < zyp> as long as pulse time is much lower than half the timer perid, you'll be fine 2013-02-03T03:41:57 < zyp> so, yeah, set CH1 and CH3 to capture rising edges, CH2 and CH4 to capture falling edges 2013-02-03T03:42:05 < zyp> enable interrupts on CH2 and CH4 2013-02-03T03:42:23 < R2COM> well then one timer can do 2 inputs 2013-02-03T03:42:30 < R2COM> 2 different inputs 2013-02-03T03:42:37 < zyp> then when you get a CH2 interrupt, just take CH2 value and subtract CH1 value to get pulse width 2013-02-03T03:42:41 < zyp> same for CH4/3 2013-02-03T03:44:34 < zyp> hmm, forget about the signed part, you don't even have to do that :p 2013-02-03T03:44:55 < R2COM> yes, so since counter is running and I cant reset, that subtraction will work 2013-02-03T03:45:02 < zyp> yes 2013-02-03T03:45:05 < R2COM> this way one timer with 4 channels can serve two different outputs 2013-02-03T03:45:13 < zyp> exactly 2013-02-03T03:45:20 < dirty_d> R2COM, you have 10 general purpose 16-bit timers 2013-02-03T03:45:26 < dirty_d> why dont you just use a timer for each channel 2013-02-03T03:45:30 < zyp> and you can't do it any simpler than that, since you need to measure both rising edges and falling edges 2013-02-03T03:45:33 < dirty_d> if you have the resourcesm use them 2013-02-03T03:45:35 < R2COM> thats another option dirty_d 2013-02-03T03:45:37 < dirty_d> dont make life hard 2013-02-03T03:46:07 < zyp> oh well 2013-02-03T03:46:34 < dirty_d> if you come up with a brilliant way to do it all in software, youre going to have like 57 timers that you paid for sitting there doing nothing 2013-02-03T03:47:16 < zyp> IMO is modifying the receiver to give a combined signal a much neater solution, I only need to run a single signal wire from receiver to stm board :) 2013-02-03T03:47:21 < gxti> yeah, screw those extra timers. i used 4 timers to measure 1 signal. 2013-02-03T03:47:55 < zyp> http://bin.jvnv.net/f/SkMsg.JPG <- with a bunch of transistors to make a NOR gate 2013-02-03T03:48:06 < R2COM> oh well... 2013-02-03T03:48:22 < R2COM> those are more components then but I guess its ok 2013-02-03T03:48:28 < R2COM> by the way 2013-02-03T03:48:37 < zyp> IMO less wiring makes it worth it 2013-02-03T03:48:42 < gxti> of course it has to be cycle-accurate, has a period of 1 second, and deliberately hunts the point where things roll over 2013-02-03T03:48:51 < gxti> so that makes it more challenging than servos :p 2013-02-03T03:50:15 < dirty_d> the easiest way is also the best way in this case 2013-02-03T03:51:16 < R2COM> also, another thing, if someone wants to use for example input capture and under some circumstances pass those *same* receiver signals unmodified to output, then it maybe makes sense to pay for 1$ Multiplexer chip and thats it 2013-02-03T03:51:35 < R2COM> i.e. if I want to give out original signals untouched 2013-02-03T03:51:35 < dirty_d> huh 2013-02-03T03:51:49 < dirty_d> what do you mean, the signals are already there 2013-02-03T03:52:17 < dirty_d> they dont need to go through the mcu to go somewhere else 2013-02-03T03:52:21 < dirty_d> they just already exist 2013-02-03T03:52:36 < R2COM> no.. I mean I get for example 4 signals from receiver, and I dont want to do anything, just pass them to whatever as they are, then I dont need stm32 to measure their pulse width, then activate another output timer and pass as output, then I just can use output multiplexer chip 2013-02-03T03:53:00 < R2COM> yes... but in some cases you want them to be different, not the way how they arrived... 2013-02-03T03:53:11 < R2COM> but if not, then just pass them through the output multiplexer and thats it 2013-02-03T03:53:26 < zyp> my quadrotor board can do that :) 2013-02-03T03:53:54 < R2COM> if one combines them in single wire with logic and does things like that then yes 2013-02-03T03:53:56 < R2COM> if thats what you did 2013-02-03T03:54:11 < zyp> ah, that's not what I'm thinking of 2013-02-03T03:54:45 < dirty_d> why the hell are you not allowed to cast in a constexpr? 2013-02-03T03:54:49 < zyp> http://bin.jvnv.net/f/sTW8U.JPG <- this is the board I'm talking about 2013-02-03T03:54:59 < dirty_d> im not seeing how this wouldnt be well defined behavior 2013-02-03T03:55:04 < zyp> see the cpld sitting between the xbee sockets? :) 2013-02-03T03:55:19 < R2COM> oh well...cpld... sure 2013-02-03T03:55:29 < zyp> dirty_d, I bet it's because you are dealing with a pointer 2013-02-03T03:55:29 < R2COM> then all problems solved of course 2013-02-03T03:55:42 < zyp> well, you said $1 mux 2013-02-03T03:55:57 < R2COM> yea I did not go into specifics, yuea it could be cpld too 2013-02-03T03:56:04 < zyp> I think that cpld was $1, and my sole purpose for it was to use it as a flexible mux ;) 2013-02-03T03:56:10 < R2COM> yes 2013-02-03T03:56:14 < dirty_d> zyp, but if its a pointer to something that exists at compile time then its completely all well defined 2013-02-03T03:56:23 < dirty_d> its not even really a pointer 2013-02-03T03:56:29 < R2COM> but my idea was roughly same an external "mux" 2013-02-03T03:56:46 < zyp> dirty_d, except it's clearly casted to a void* and back. 2013-02-03T03:56:59 < dirty_d> zyp, i changed it to take a uint8_t * pointer 2013-02-03T03:57:14 < gxti> dirty_d: it may not be "completely well defined" until link time, and the linker may not be smart enough to understand 2013-02-03T03:57:21 < zyp> dirty_d, well, still 2013-02-03T03:57:25 < gxti> just because it makes sense doesn't mean it'll work :p 2013-02-03T03:57:25 < zyp> pointers are not buffers. 2013-02-03T03:57:48 < zyp> gxti, it errors out before link time 2013-02-03T03:57:54 < dirty_d> zyp, its not pointers that arenet allowed, its casting from one trype to the other 2013-02-03T03:58:17 < gxti> zyp: well yeah, but it could be because it can't express whatever he's doing in the object file 2013-02-03T03:58:28 < zyp> dirty_d, you mean casting pointers. 2013-02-03T03:58:30 < gxti> but i'll shut up now because i have no idea how it actually works :| 2013-02-03T03:58:41 < dirty_d> gxti, it can though because it works on g++ 2013-02-03T03:58:53 < dirty_d> not on clang++ though 2013-02-03T03:58:55 < zyp> except it doesn't? 2013-02-03T03:59:15 < dirty_d> no it really works with g++ 2013-02-03T03:59:26 < baird> I almost lost all my pretty skin-- I had a 12V SLA battery on quick-charge on the workbench about a metre from myself at the computer... and forgot about it. 4 hours later, I remember, and burn my fingers on the plastic while removing the power. If the container had ruptured... 2013-02-03T04:00:03 < gxti> or detonated :D 2013-02-03T04:00:43 < dirty_d> i dont know why im going through all this trouble just to have all this happen at compile time 2013-02-03T04:00:54 < zyp> dirty_d, anyway, it looks pretty dumb to first generate the descriptors and then go back to parse the resulting byte buffer :p 2013-02-03T04:01:16 < dirty_d> zyp, yea kinda, one class should do it all 2013-02-03T04:01:29 < baird> "He (would have) died doing what he loved doing--- fucking around with acids..." 2013-02-03T04:01:35 < dirty_d> its not really doing anything if its all constexpr though 2013-02-03T04:01:53 < zyp> huh? 2013-02-03T04:02:11 < dirty_d> its all just a bunch of constants 2013-02-03T04:02:19 < dirty_d> well it can be 2013-02-03T04:02:21 < zyp> I'm talking about the code looking dumb and not particularly fun to maintain, not that it's inefficient :p 2013-02-03T04:02:30 < dirty_d> oh yea, lol 2013-02-03T04:02:50 < dirty_d> well by reasoning was that it will work with a descriptor in any form 2013-02-03T04:03:11 < dirty_d> well not form, but type i guess youd say 2013-02-03T04:03:18 < dirty_d> it does have to have the same form in memory 2013-02-03T04:03:29 < zyp> why do you care? 2013-02-03T04:03:40 < dirty_d> OCD 2013-02-03T04:03:45 < zyp> why do you need to parse the descriptor? 2013-02-03T04:04:09 < dirty_d> zyp, so the usb class can initialize itself from a device descriptor and config descriptor 2013-02-03T04:04:28 < zyp> that sounds very backwards to me. 2013-02-03T04:04:37 < dirty_d> i dunno 2013-02-03T04:05:44 < dirty_d> i think defining a usb device with data in the form of a descriptor is simpler than defining it in a class 2013-02-03T04:06:00 < zyp> I disagree 2013-02-03T04:06:07 < dirty_d> one class could read any descriptor and just work 2013-02-03T04:06:41 < zyp> or you could just create a device with the required class drivers, and then it could generate all the required descriptors itself. 2013-02-03T04:06:44 < dirty_d> and it would just have to handle reads/writes on each endpoint 2013-02-03T04:07:17 < zyp> I mean 2013-02-03T04:07:42 < zyp> I want to be able to just instance the usb driver, then instance a cdc_acm driver attached to it 2013-02-03T04:08:05 < zyp> and then the cdc_acm driver would generate all the required descriptors, without me even having to see them 2013-02-03T04:08:26 < zyp> why should I have to care about all that if the driver can do it for me? 2013-02-03T04:08:39 < zyp> I just want a virtual serial port, I don't care about how. 2013-02-03T04:09:44 < dirty_d> you have to know the layout of the descriptor anyway though 2013-02-03T04:09:52 < zyp> why? 2013-02-03T04:09:52 < dirty_d> youre just generating it in a different way 2013-02-03T04:09:59 < zyp> what do you mean? 2013-02-03T04:10:10 < zyp> I'm thinking from a user perspective now 2013-02-03T04:10:15 < dirty_d> well i mean the class is going to generate the descriptor 2013-02-03T04:10:27 < zyp> yes? 2013-02-03T04:10:34 < dirty_d> but obviously you need to tell it how many interfaces and endpoints and the parameters of each etc 2013-02-03T04:10:50 < zyp> no, of course not 2013-02-03T04:11:02 < dirty_d> well how would it know how to generate it? 2013-02-03T04:11:23 < dirty_d> you have to write something that knows how to 2013-02-03T04:11:33 < zyp> the cdc_acm driver knows that acm requires two bulk endpoints, then asks the stack driver to allocate those 2013-02-03T04:11:52 < dirty_d> yea but you have to write the cdc_acm driver 2013-02-03T04:11:56 < zyp> so the driver would handle all that, and the user instancing the driver should not have to care. 2013-02-03T04:12:05 < dirty_d> is it easier to write a cdc_acm driver, or a cdc_acm descriptor? 2013-02-03T04:12:37 < zyp> it's certainly easier to instance the cdc_acm driver class than writing all the required descriptors for acm. 2013-02-03T04:12:48 < dirty_d> ye but each only has to be done once 2013-02-03T04:13:05 < zyp> and then copypasted to all your new projects? 2013-02-03T04:13:27 < dirty_d> pasted or included the same as the driver would have to be 2013-02-03T04:15:21 < zyp> so, by the way, how do you interface with the acm driver after it's instanced if it's instanced by the stack? 2013-02-03T04:17:10 < dirty_d> im not following 2013-02-03T04:17:28 < dirty_d> if it reads a descriptor to configure itself? 2013-02-03T04:19:02 < zyp> oh well, I don't have time to argue it now, it just seems like a horrible idea 2013-02-03T04:19:11 < dirty_d> im not trying to argue 2013-02-03T04:19:21 < zyp> but I am :p 2013-02-03T04:19:23 < dirty_d> lol 2013-02-03T04:19:35 < zyp> but it's 3:20 and I need to sleep :p 2013-02-03T04:19:41 < dirty_d> damn 2013-02-03T04:20:07 < zyp> anyway, my point is modularization and hiding away unimportant code 2013-02-03T04:21:24 < zyp> the descriptors used by a usb class could be considered implementation details of that class, and should therefore be hidden away inside the driver for that class, so that users of that driver would only deal with a clean interface presenting the features they want without having to deal with the implementation details 2013-02-03T04:23:59 < zyp> for cdc_acm that interface should include tx and rx buffers internally coupled to usb flow control so that the user don't have to worry about that 2013-02-03T04:24:35 < zyp> s/should/could/ 2013-02-03T04:24:39 < dirty_d> hmm i guess 2013-02-03T04:25:02 < dirty_d> it just seems like having a 15 line descriptor would really cut down on the size of the driver code 2013-02-03T04:25:19 < dirty_d> since you neecd all that information somewhere else anyway 2013-02-03T04:25:22 < zyp> I don't see how. 2013-02-03T04:25:36 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-233-1.clienti.tiscali.it] has quit [Ping timeout: 276 seconds] 2013-02-03T04:26:15 < zyp> and I'm not concerned about size of driver code 2013-02-03T04:26:21 < dirty_d> well i mean all of the fields in the device and configuration descriptor need to be expressed somwhere 2013-02-03T04:26:28 < zyp> I'm more concerned about the usefulness of it 2013-02-03T04:26:51 < dirty_d> expressing it as data is shorter than in code 2013-02-03T04:26:56 < zyp> heh 2013-02-03T04:27:40 < dirty_d> a descritor is already data, so it just seems like it should be defined as data in the program instead of writing code that generates it 2013-02-03T04:30:17 < zyp> generates how? 2013-02-03T04:30:59 < dirty_d> i dunno, however it would 2013-02-03T04:31:08 < dirty_d> thats what i mean 2013-02-03T04:31:41 < zyp> well, it wouldn't really be a huge difference 2013-02-03T04:32:25 < zyp> after all, the configuration descriptor is just a concatenated set of the descriptors used for each of the device's functions 2013-02-03T04:32:38 < dirty_d> a simple descriptor accurately defines a device in every way except for what it does when it receives/sends data 2013-02-03T04:33:02 < dirty_d> it just seems like the driver should only worry about what to do in those two cases 2013-02-03T04:33:36 < zyp> and that's why it's bad. 2013-02-03T04:33:58 < zyp> as a user, I don't want to define the device in every way 2013-02-03T04:34:33 < zyp> I just want to say which functions I want, and then have it define itself in whatever suitable manner 2013-02-03T04:34:43 < dirty_d> im taling from the perspective of us 2013-02-03T04:34:57 < dirty_d> to the user it doesnt matter 2013-02-03T04:35:06 < dirty_d> hes gonna include a couple files and use a class 2013-02-03T04:35:20 < dirty_d> im talking about how to implement these different types of device classes 2013-02-03T04:35:35 < zyp> so you are only going to support single function devices? 2013-02-03T04:35:58 < dirty_d> single configuration? 2013-02-03T04:36:06 < zyp> single function. 2013-02-03T04:36:16 < zyp> as in a device that only supports one usb class 2013-02-03T04:36:52 < zyp> have you considered composite devices at all? 2013-02-03T04:37:25 < dirty_d> im confused 2013-02-03T04:37:37 < zyp> do you know what a composite device is? 2013-02-03T04:37:41 < dirty_d> no 2013-02-03T04:38:08 < zyp> it's a device supporting multiple usb classes 2013-02-03T04:38:21 < zyp> so it would have several drivers attached at the same time 2013-02-03T04:39:15 < dirty_d> that would still work i think 2013-02-03T04:39:43 < zyp> and I want my users to be able to easily say «hey, I want to be a keyboard and also have two virtual serial ports», by instancing and attaching one hid driver object and two acm driver objects 2013-02-03T04:40:13 < dirty_d> how does that actually work though? 2013-02-03T04:40:55 < dirty_d> as in how are the descritors structured that it sends the host 2013-02-03T04:41:15 < zyp> you just concatenate the descriptors of each function and make sure the interface and endpoint numbers are unique 2013-02-03T04:41:40 < dirty_d> you cna only have one device descriptor though? 2013-02-03T04:41:56 < zyp> of course, I'm talking about the configuration 2013-02-03T04:42:00 < dirty_d> you mean concatenate configuration descriptors? 2013-02-03T04:42:07 < dirty_d> ok 2013-02-03T04:42:08 < zyp> no 2013-02-03T04:42:33 < zyp> I mean, you just add more interfaces 2013-02-03T04:42:49 < zyp> each function has one or more interfaces 2013-02-03T04:42:53 < dirty_d> yea 2013-02-03T04:43:23 < zyp> hid uses one interface, acm uses two, so the mentioned device would have a configuration with five interfaces in total 2013-02-03T04:43:42 < zyp> simple as that 2013-02-03T04:43:47 < dirty_d> i see 2013-02-03T04:45:37 < dirty_d> it would be harder for the user 2013-02-03T04:45:43 < dirty_d> to write the descritopr 2013-02-03T04:46:09 < zyp> so what I mean is that each of my drivers would just supply its own interface- and other descriptors, and they all get combined/concatenated to the configuration descriptor 2013-02-03T04:47:01 < zyp> since I would like to allocate free endpoint numbers automatically, they would be dynamic, but most other fields of the descriptor could still just be provided statically in the driver 2013-02-03T04:48:44 < dirty_d> how would you allocate the descriptor? 2013-02-03T04:49:33 < dirty_d> template/constexpr magic or dynamically? 2013-02-03T04:53:37 < zyp> haven't decided 2013-02-03T04:54:10 < zyp> I'm considering changing the API to allow partial writes to the transfer buffer 2013-02-03T04:54:58 < zyp> so each driver might be asked to just write it's part into the buffer in turn 2013-02-03T05:00:28 < dirty_d> hmm 2013-02-03T05:01:20 < dirty_d> so whats the deal with the cdc control endpoint? 2013-02-03T05:01:26 < dirty_d> i dont see it being used in your code 2013-02-03T05:01:35 < zyp> it's not 2013-02-03T05:02:14 < zyp> it's used for reporting state of inbound control signals 2013-02-03T05:02:36 < zyp> keep in mind that acm emulates a serial port 2013-02-03T05:03:37 < zyp> if you make a full serial converter complete with all the control signals, you would report changes in the inbound ones through that endpoint 2013-02-03T05:04:01 < zyp> outbound signals are controlled by a control request 2013-02-03T05:04:19 < zyp> I handle and just ignore that request as well 2013-02-03T05:05:15 < zyp> http://bin.jvnv.net/f/gffMk.png 2013-02-03T05:06:18 < dirty_d> ahh ok 2013-02-03T05:06:47 < dirty_d> what are you using to sniff usb traffic? 2013-02-03T05:07:28 < zyp> a beagle 2013-02-03T05:07:41 < zyp> http://www.totalphase.com/products/beagle_usb480/ 2013-02-03T05:08:17 < dirty_d> whoa, i dont know if that price is $1200, or $12.00 2013-02-03T05:08:59 < zyp> I'm just borrowing it :p 2013-02-03T05:09:23 < dirty_d> holy shit, its really $1200? 2013-02-03T05:09:36 < zyp> site says $1400 2013-02-03T05:09:44 < dirty_d> is it made of gold? 2013-02-03T05:09:50 < dongs> they had a nice sale few yerasr ago 2013-02-03T05:10:00 < dongs> was half off. shoulda got it that time. isntead i bought ellisys analyzer. 2013-02-03T05:10:07 < dongs> http://www.ellisys.com/products/usbex200/index.php 2013-02-03T05:10:15 < dirty_d> how is it that expensive? 2013-02-03T05:10:26 < dongs> because its test & measurement tools 2013-02-03T05:10:29 < dongs> why are scopes expensive? 2013-02-03T05:10:36 < zyp> professional test equipment usually are :p 2013-02-03T05:10:38 < dongs> or proper logic analyzers 2013-02-03T05:10:39 < dongs> etc 2013-02-03T05:11:20 < dirty_d> a scope is less 2013-02-03T05:11:24 < dongs> uh. no 2013-02-03T05:11:30 < dongs> a shit scope is less. 2013-02-03T05:11:36 < dongs> which is worthless the momentt it goes on your desk 2013-02-03T05:11:43 < zyp> it's not a high volume commodity, so there aren't as many users to distribute the development costs across 2013-02-03T05:12:39 < zyp> and $1400 is nothing to people making money from usb devices 2013-02-03T05:12:47 < dirty_d> true 2013-02-03T05:15:01 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-212.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-03T05:15:12 < zyp> still wondering how the openviszla will turn out 2013-02-03T05:15:37 < dongs> Heh 2013-02-03T05:16:08 < zyp> the hardware will probably be decent enough, but I wonder how long it will be before they have anything remotely non-shitty on the software side :p 2013-02-03T05:16:19 < dongs> lunix? never 2013-02-03T05:20:17 < zyp> oh well, time to sleep 2013-02-03T05:20:25 < zyp> for real this time :) 2013-02-03T05:21:15 < dirty_d> yea no shit its like 4:00 2013-02-03T05:22:12 < dirty_d> only 22:22 here though, still early enough to order some gresy dominos 2013-02-03T05:25:31 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-03T05:30:47 -!- emeb [~ericb@72.201.78.226] has quit [Quit: Leaving.] 2013-02-03T05:33:00 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-03T06:06:23 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-03T06:10:17 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-03T06:10:23 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-03T06:22:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-03T06:30:42 -!- drgreenthumb [~fletch@drupal.org/user/72475/view] has quit [Ping timeout: 264 seconds] 2013-02-03T06:41:02 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 255 seconds] 2013-02-03T06:49:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving] 2013-02-03T07:17:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-03T07:21:04 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has joined ##stm32 2013-02-03T07:36:25 < timemob> http://www.imgur.com/APTDNmw.jpeg 2013-02-03T07:38:13 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has quit [Quit: Bye] 2013-02-03T07:43:55 < gnomad> that's.... interesting. 2013-02-03T07:47:13 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-03T07:47:36 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-03T07:48:47 < emeb_mac> fuel gauge on an electric motorcycle? 2013-02-03T07:50:04 < gnomad> not a clue. 2013-02-03T07:54:33 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-03T08:06:51 -!- KennyMcCormic [~Kenny@95.139.182.5] has joined ##stm32 2013-02-03T08:10:33 < KennyMcCormic> morning 2013-02-03T08:12:01 < KennyMcCormic> guys, has anybody tested jitter on stm32l or f4 hsi jitter & stability? 2013-02-03T08:14:18 < KennyMcCormic> i want to create precise test signal and sample it afterwards, problem is that adc is always using HSI as clock , so either i need to create test signal using HSI or sync adc to HSE somehow relying on HSI short term stability 2013-02-03T08:15:16 < KennyMcCormic> what is better in this case? 2013-02-03T08:19:21 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has joined ##stm32 2013-02-03T08:20:09 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has quit [Client Quit] 2013-02-03T08:27:59 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32 2013-02-03T08:34:42 < dongs> KennyMcCormic: uh, ADC is NOT always using HSI as clock 2013-02-03T08:40:22 < baird> i r birthday boy today. My housemate gave me a bottle of Lagavulin wiskey.. The first-impression is having an aroma of asphalt... ~_~ 2013-02-03T08:41:02 < dongs> fucking animu smiley 2013-02-03T08:42:05 < KennyMcCormic> dongs: in stm32l it is always on HSI, i wonder if sample and hold circuit is also synchronized to it 2013-02-03T08:42:30 < dongs> oh, 32l 2013-02-03T08:42:54 < dongs> it cant do external xtal at all? or what 2013-02-03T08:43:09 < KennyMcCormic> as far as i can see - no 2013-02-03T08:43:44 < KennyMcCormic> 32l itself can work from xtal , but adc is connected to hsi 2013-02-03T08:43:52 < dongs> that is mega wierd 2013-02-03T08:43:56 < KennyMcCormic> yep 2013-02-03T08:44:05 < KennyMcCormic> thats why im asking 2013-02-03T08:44:11 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-03T08:45:45 < dongs> ADC clocked at 16 MHz < I dont see how thats possible off hsi 2013-02-03T08:46:03 < dongs> o wait 2013-02-03T08:46:06 < dongs> block diagram 2013-02-03T08:46:17 < dongs> looks like it has its *own* dedicated hsi 2013-02-03T08:46:26 < KennyMcCormic> no 2013-02-03T08:46:27 < dongs> probably beacuse you can run adc at full power while cpu is slow 2013-02-03T08:47:01 < dongs> HSIThe ADC clock which is always the HSI clock. A divider by 1, 2 or 4 allows to adapt the 2013-02-03T08:47:05 < dongs> clock frequency to the device operating conditions. For more details please refer to the 2013-02-03T08:47:08 < dongs> Operating Power Supply Range section in the PWR chapter. 2013-02-03T08:47:10 < dongs> yep 2013-02-03T08:47:13 < dongs> you're scrwed. 2013-02-03T08:47:44 < KennyMcCormic> so now it depends on how stable hsi is 2013-02-03T08:48:06 < KennyMcCormic> is it stable in stm32? 2013-02-03T08:48:25 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-03T08:50:04 < dongs> its... better than the one on AVR. 2013-02-03T08:50:16 < dongs> cant you just sync it to a TIMx thats running of external clock? 2013-02-03T08:50:24 < dongs> what do you need the adc clock t o be stable for> 2013-02-03T08:50:53 < KennyMcCormic> i have a sine generated off hse 2013-02-03T08:50:59 < KennyMcCormic> now i need to sample it 2013-02-03T08:51:29 < KennyMcCormic> if adc is not using the same clock i have +-1/16M jitter in measurements 2013-02-03T08:51:49 < KennyMcCormic> which is not good 2013-02-03T08:52:55 < KennyMcCormic> im averaging measurements so timing is very critical 2013-02-03T08:56:05 < R2COM> if its that critical then you better off with separate ADC on chip, clocked with clock synthesizer with subpicosecond jitter 2013-02-03T08:57:29 < KennyMcCormic> R2COM: not that critical, for me it is just enough to get 0 jitter between dac & adc clocks 2013-02-03T08:57:41 < R2COM> 0 jitter..? 2013-02-03T08:57:59 < KennyMcCormic> i mean they should use the same clocks 2013-02-03T08:58:22 < R2COM> there are logich synthesizers with several outputs, they are synchronous 2013-02-03T08:58:36 < KennyMcCormic> inside stm32l ? 2013-02-03T08:58:42 < R2COM> no.. 2013-02-03T08:59:19 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-03T08:59:46 < R2COM> for stm32, i guess they both clocked with APB1 related clock 2013-02-03T09:03:25 -!- DaKu is now known as daku 2013-02-03T09:04:10 -!- daku is now known as DaKu 2013-02-03T09:04:19 < dongs> R2COM: well, apparently in stm32L, they're not 2013-02-03T09:04:24 < dongs> ADC runs off its own HSI. 2013-02-03T09:04:33 < dongs> independent of apb/ahb/whatever 2013-02-03T09:04:42 < R2COM> well ok then 2013-02-03T09:05:32 < R2COM> its not expencive to check if it works out for application, if not, just buy separate components as described above, I'd recommend analog devices 2013-02-03T09:24:26 -!- DaKu is now known as daku 2013-02-03T09:24:44 -!- daku is now known as DaKu 2013-02-03T10:31:02 -!- claude is now known as Claude 2013-02-03T10:43:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-03T10:55:19 -!- barthess [~barthess@5.100.211.45] has joined ##stm32 2013-02-03T11:28:52 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds] 2013-02-03T11:30:03 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-03T11:49:19 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Remote host closed the connection] 2013-02-03T11:50:45 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-03T11:53:53 -!- barthess [~barthess@5.100.211.45] has quit [Quit: Leaving.] 2013-02-03T12:04:02 < dongs> spartan6 doesnt have ddr memory controller? huh weird i thought it did 2013-02-03T12:04:21 < jpa-> you can always program one on it :P 2013-02-03T12:04:30 < dongs> fuck that 2013-02-03T12:04:59 < dongs> fucking xilinx using opensores PDF tools 2013-02-03T12:05:02 < dongs> copypasting text out of them = comes back as unicode garbage 2013-02-03T12:05:25 < jpa-> which tool did they use? 2013-02-03T12:05:33 < dongs> lets see 2013-02-03T12:05:48 < dongs> haha acrocrap distiller 9.0 (Macintosh) 2013-02-03T12:05:55 < dongs> thats terrible 2013-02-03T12:06:05 < jpa-> yeah total crap open sores 2013-02-03T12:06:09 < dongs> from Indesign CS4 2013-02-03T12:07:13 < dongs> The key features and benefits of the Spartan-6 FPGA memory controller block are: 2013-02-03T12:07:14 < dongs> • DDR, DDR2, DDR3, and LPDDR (Mobile DDR) memory standards support 2013-02-03T12:14:01 < RuslanPopov> ppl, help with PWM please 2013-02-03T12:14:18 < RuslanPopov> f100, tim17, 36kHz PWM - https://github.com/RaD/armka_demos/blob/master/apps/ir-tester/main.c 2013-02-03T12:14:35 < RuslanPopov> I see 16kHz on the pin! 2013-02-03T12:15:39 < dongs> chibios?? 2013-02-03T12:15:48 < RuslanPopov> almost 2013-02-03T12:16:04 < dongs> does tim17 even need BDTR 2013-02-03T12:16:06 < RuslanPopov> but tim17 is unsupported in chibios now 2013-02-03T12:16:10 < dongs> i thought only TIM1/8 (advanced) timers needed it 2013-02-03T12:17:12 < RuslanPopov> tim17 has bdtr register 2013-02-03T12:18:43 < RuslanPopov> I can't understand this magic, huh 2013-02-03T12:20:18 < jpa-> verify your measurements, it seems unlikely to get 16kHz from that.. 18 kHz maybe 2013-02-03T12:20:49 < RuslanPopov> but even 18 differs from 36kHz 2013-02-03T12:26:26 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-03T12:30:16 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-03T12:34:59 -!- gnomad [~gnomad@71.203.29.67] has joined ##stm32 2013-02-03T12:38:44 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 248 seconds] 2013-02-03T13:02:25 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-03T13:26:28 -!- DaKu is now known as daku 2013-02-03T13:27:09 -!- daku is now known as DaKu 2013-02-03T13:30:59 -!- KennyMcCormic [~Kenny@95.139.182.5] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 2013-02-03T13:36:56 -!- DaKu is now known as daku 2013-02-03T13:37:07 -!- daku is now known as DaKu 2013-02-03T13:38:11 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-03T13:38:11 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-03T13:38:11 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T13:38:14 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-03T13:54:52 < Tectu> do you C++ guys always use prefixes such as C for classes, m_ f or members, i for integers, f for floats and so on? 2013-02-03T13:55:16 < zyp> no. 2013-02-03T14:08:44 -!- Claude is now known as claude 2013-02-03T14:17:00 < RuslanPopov> heh, it seems I just can't 36khz with mic input 2013-02-03T14:17:14 < RuslanPopov> lower freqs are visible\ 2013-02-03T14:18:41 < jpa-> Tectu: i use m_ for members, but nothing of the other stuff 2013-02-03T14:18:59 < jpa-> some people use C for classes and I for interfaces (fully abstract classes) 2013-02-03T14:19:07 < jpa-> only idiots use i for integers etc. 2013-02-03T14:20:03 < jpa-> RuslanPopov: ah, of course 2013-02-03T14:20:16 < RuslanPopov> I need an advice 2013-02-03T14:20:44 < RuslanPopov> I have to use AIC1863 - IR preamplifier to get the IR signal presence 2013-02-03T14:21:10 < RuslanPopov> I think that 36kHz PWM is good source signal for AIC1863 2013-02-03T14:21:44 < RuslanPopov> I need only detect signal presence, no data transmission happens 2013-02-03T14:22:17 < RuslanPopov> is there an error in my idea? 2013-02-03T14:22:39 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Read error: Connection reset by peer] 2013-02-03T14:25:39 < Tectu> jpa-, thanks! 2013-02-03T14:27:20 < Tectu> jpa-, reading some code here, every member type is with i, f, d, c etc... 2013-02-03T14:27:21 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-03T14:28:16 < jpa-> Tectu: probably a coder with windows COM background.. that stuff uses those prefixes 2013-02-03T14:28:46 < Tectu> jpa-, jup, it's a windows program which does also use the windows api to create windows etc 2013-02-03T14:28:52 < Tectu> now, that sounds funny... 2013-02-03T14:29:19 < jpa-> yeah, most winapi code looks like dog puke 2013-02-03T14:29:37 < Tectu> well, personally, I like the idea 2013-02-03T14:29:44 < Tectu> but I never saw it somewhere else 2013-02-03T14:29:50 < Tectu> and it looks like it makes the code a bit messy 2013-02-03T14:30:09 < Tectu> m_vectPop[i].m_iRate = 42; 2013-02-03T14:30:17 < RuslanPopov> I think it is crazy way 2013-02-03T14:30:28 < RuslanPopov> they must use Python :) 2013-02-03T14:31:38 <+Steffanx> Noo, now you know it's an integer.. how fantastic is taht Tectu :) 2013-02-03T14:31:55 < Tectu> stap it dude, stap it! 2013-02-03T14:32:28 <+Steffanx> Ok Boss 2013-02-03T15:10:40 < dongs> http://nicosound.anyap.info/sound/sm2428954 attn zyp/etc 2013-02-03T15:15:06 < zyp> what about that? 2013-02-03T15:16:23 <+Steffanx> anime! 2013-02-03T15:19:27 -!- DaKu is now known as daku 2013-02-03T15:25:03 -!- gnomad [~gnomad@71.203.29.67] has quit [Quit: This computer has gone to sleep] 2013-02-03T15:35:34 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 246 seconds] 2013-02-03T15:50:57 -!- Laurenceb_ [~Laurence@86.177.58.138] has joined ##stm32 2013-02-03T16:08:23 -!- claude is now known as Claude 2013-02-03T16:09:13 -!- daku is now known as DaKu 2013-02-03T16:29:58 -!- Claude is now known as claude 2013-02-03T16:30:26 -!- ReggieUK [ReggieUK@90.202.86.231] has joined ##stm32 2013-02-03T16:37:07 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32 2013-02-03T16:39:43 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-03T16:52:11 < mervaka> hmmmmm 2013-02-03T16:52:16 < mervaka> very interesting development 2013-02-03T16:52:30 < mervaka> just actually looked at my dma buffers 2013-02-03T16:52:40 < mervaka> and it seems it's only half filling :s 2013-02-03T16:56:08 < emeb_mac> odd 2013-02-03T16:56:25 < Laurenceb_> wrong size in dma controller? 2013-02-03T16:56:55 < emeb_mac> ht/tc flags? 2013-02-03T16:57:06 < mervaka> ah 2013-02-03T16:57:08 < emeb_mac> wordsize wrong? 2013-02-03T16:57:16 < mervaka> the data size is half word 2013-02-03T16:57:18 < Laurenceb_> oh nice - rfblade got some serious money :P 2013-02-03T16:57:24 < mervaka> i have a feeling that has a lot to do with it. 2013-02-03T16:57:24 < Laurenceb_> there you go then :P 2013-02-03T16:57:28 < jpa-> how to stop gcc from optimizing away the args array initializations? http://paste.dy.fi/uzZ/plain 2013-02-03T17:03:12 < jpa-> compilable version: http://paste.dy.fi/u8X/plain 2013-02-03T17:03:24 < jpa-> it totally drops the strings and everything 2013-02-03T17:07:36 < emeb_mac> bizarre. 2013-02-03T17:08:04 < emeb_mac> jpa-: what's that bkpt thing? some kind of call to system rom? 2013-02-03T17:08:53 < emeb_mac> Laurenceb_: yep - blade rf is about 1/2-way to the goal. 2013-02-03T17:09:03 * Laurenceb_ off 2013-02-03T17:09:10 < emeb_mac> $400 SDR for everyone. 2013-02-03T17:09:43 < jpa-> emeb_mac: it's for qemu-system-arm 2013-02-03T17:10:17 < emeb_mac> oh. never futzed with that. 2013-02-03T17:10:37 < jpa-> trying to benchmark some stuff 2013-02-03T17:10:45 < jpa-> (in simulator) 2013-02-03T17:13:12 < jpa-> making the array static fixes it for now 2013-02-03T17:13:54 < emeb_mac> I guess that the compiler doesn't know that the bkpt service uses it, so away it goes. 2013-02-03T17:17:14 < jpa-> but shouldn't it realize that unknown opcodes can do stuff to values in registers? 2013-02-03T17:17:41 < jpa-> i think the problem is more that it thinks the asm statement only cares about the pointer, and not about what the pointer points to 2013-02-03T17:20:16 < emeb_mac> makes sense 2013-02-03T17:21:24 < jpa-> but it would also make sense that there would be a way to tell that "hey, if i have a pointer i really want it to actually point to something!" 2013-02-03T17:27:01 <+Steffanx> lol @ how jpa- thinks that GCC thinks. :P 2013-02-03T17:29:10 < jpa-> no no, it's what i think 2013-02-03T17:29:38 < jpa-> gcc thinks "lololol this is not in C standard now i have an excuse to do whatever i want hahahaha" 2013-02-03T17:29:50 < emeb_mac> pwnd :P 2013-02-03T17:30:19 <+Steffanx> *to do whatever the programmer that wrote me wants me to do ( or whatever the one who uses me configured me to do ) 2013-02-03T17:30:33 < emeb_mac> do what I mean, not what I say 2013-02-03T17:33:33 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-03T17:55:12 < dongs> TIL: --c99 to armcc lets it compile zyp-like code 2013-02-03T17:55:35 < gxti> but supporting modern standards is counterproductive, dongs 2013-02-03T17:55:39 < dongs> yeah 2013-02-03T17:55:43 < dongs> i am not enabling that on any of my projects 2013-02-03T17:55:43 < gxti> better to use C the way god invented it 2013-02-03T17:55:45 < gxti> or whatever 2013-02-03T17:55:59 < dongs> i mean, it even allows variable declaration in the middle of scope 2013-02-03T17:56:02 < dongs> thats just retarded 2013-02-03T17:56:07 < dongs> makes for lazy coders 2013-02-03T17:57:06 < mervaka> ok, that's awesome. i've fixed my glitch entirely :D:D 2013-02-03T17:57:12 < mervaka> but only so far using memcpy 2013-02-03T18:10:19 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-03T18:16:06 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-03T18:16:12 < emeb_mac> mervaka: what did you do? 2013-02-03T18:16:29 < mervaka> multiplied the buffer size by two.. 2013-02-03T18:16:30 < mervaka> lol 2013-02-03T18:16:36 < emeb_mac> :P 2013-02-03T18:16:41 < mervaka> in the dma controller 2013-02-03T18:16:49 < mervaka> only running a memcpy at the moment 2013-02-03T18:17:13 < emeb_mac> circular mode or double buffer? 2013-02-03T18:17:27 < mervaka> double 2013-02-03T18:17:37 < mervaka> cba changing anything back 2013-02-03T18:17:48 < emeb_mac> huh. I've only ever used circular w/ ht/tc 2013-02-03T18:18:03 < emeb_mac> never had problems w/ that. 2013-02-03T18:18:20 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-03T18:20:24 < mervaka> actually, wait 2013-02-03T18:20:30 < mervaka> there's still a very minor glitch. 2013-02-03T18:20:34 < mervaka> fuck sake. 2013-02-03T18:20:35 < mervaka> lol 2013-02-03T18:22:30 < emeb_mac> interesting - the DMA controller on the F3 doesn't support double-buffered mode. Only circular. 2013-02-03T18:22:45 < mervaka> mm 2013-02-03T18:23:45 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Ping timeout: 260 seconds] 2013-02-03T18:24:11 < emeb_mac> mervaka: do you have your DMA buffer word aligned? 2013-02-03T18:24:25 < mervaka> i think so. 2013-02-03T18:24:34 < emeb_mac> wonder if that would cause any problems... 2013-02-03T18:25:57 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-03T18:27:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-03T18:28:00 < mervaka> well memcpy works ok, but i guess that doesn't matter 2013-02-03T18:28:16 < mervaka> i'll get the DSP back in and see how that goes. 2013-02-03T18:28:48 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-03T18:33:44 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-03T18:39:33 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-03T18:44:36 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Ping timeout: 276 seconds] 2013-02-03T18:52:50 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-03T19:04:07 < emeb_mac> Cardinal Fang - fetch... the DSP! 2013-02-03T19:13:13 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-03T19:13:42 < karlp> wat? searched for cpld on mouser, result number 2 is stm32f4 disco board. 2013-02-03T19:15:09 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-03T19:19:44 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Ping timeout: 255 seconds] 2013-02-03T19:20:14 -!- barthess [~barthess@5.100.194.142] has joined ##stm32 2013-02-03T19:20:19 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-03T19:35:02 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-03T19:38:18 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Ping timeout: 252 seconds] 2013-02-03T19:42:17 <+dekar> karlp, maybe there is a secret on chip PLD :D 2013-02-03T19:43:06 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has quit [Read error: Connection reset by peer] 2013-02-03T19:43:32 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-03T19:45:11 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-03T19:45:31 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Read error: Operation timed out] 2013-02-03T19:47:22 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-03T19:48:10 < mervaka> emeb: bizarre. the memcpy passthru only works well when the buffer size is 1024 words long 2013-02-03T19:48:41 < mervaka> 2048 glitches, 512/256/etc glitches 2013-02-03T19:49:11 < emeb> mervaka: that's very strange. 2013-02-03T19:49:21 < mervaka> yup :/ 2013-02-03T19:51:38 -!- dekar_ [~dekar@212.255.31.181] has joined ##stm32 2013-02-03T19:51:41 -!- mode/##stm32 [+v dekar_] by ChanServ 2013-02-03T19:51:55 < mervaka> i've also tried commenting out the bulk of the DSP code and reversing just one channel. both channels still glitch. 2013-02-03T19:52:03 < mervaka> still sounds like a timing issue. 2013-02-03T19:55:26 < emeb> be interesting to figure out where in the buffer w/ respect to start:end the glitch occurs. 2013-02-03T19:55:27 -!- dekar [~dekar@212.255.31.181] has quit [Ping timeout: 276 seconds] 2013-02-03T19:55:59 < emeb> I found it helpful to introduce known glitches by zeroing out specific regions of the buffer. 2013-02-03T19:56:22 < emeb> then compare your mystery glitch to the ones you expect. 2013-02-03T19:56:31 < mervaka> yeah definitely. i want to say that the halfword pair overlap buffers, but in my swap code, i return the data in exactly the same form. so confusing. 2013-02-03T20:01:39 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Read error: Connection reset by peer] 2013-02-03T20:03:12 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-03T20:05:46 < mervaka> well that's a sight. 2013-02-03T20:06:07 < mervaka> triggering off the ISR GPIO pin 2013-02-03T20:06:13 < mervaka> that i'm sending out 2013-02-03T20:06:28 < mervaka> and put the output sine on the other channel of the scope 2013-02-03T20:06:54 < mervaka> got a sinewave not syncing with the scope, but the glitch is sitting pretty :P 2013-02-03T20:19:00 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Ping timeout: 248 seconds] 2013-02-03T20:19:57 < gxti> i just got spam for some shitty indiegogo thing, sent to an email address i randomly generated for seeedstudio. thanks seeedstudio. 2013-02-03T20:20:56 < emeb> "if you're not paying for it, you are the product" 2013-02-03T20:21:12 < emeb> and even then... 2013-02-03T20:21:14 < gxti> i did pay for it 2013-02-03T20:21:35 * gxti files abuse report 2013-02-03T20:35:39 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-03T20:48:26 < jpa-> lol the slowest way ever to count number of instructions executed in a simulator.. 2013-02-03T20:48:39 < jpa-> infinite loop with 'si' in gdb, log to file, wc -l 2013-02-03T20:49:07 < jpa-> i get about 200 instructions per second :) 2013-02-03T20:50:59 < zyp> hah 2013-02-03T20:53:23 < jpa-> if i recompiled qemu there would be a dozen more efficient ways 2013-02-03T20:53:40 < jpa-> i just don't understand why they don't include a simple "instruction count" in the main qemu code 2013-02-03T20:53:50 < jpa-> it's a popular question on their mailing list 2013-02-03T21:12:33 < mervaka> hahaha 2013-02-03T21:12:43 < mervaka> well that's skirted round the issue. 2013-02-03T21:13:12 < mervaka> emeb: pushed the interrupt/glitch frequency above nyquist frequency 2013-02-03T21:13:25 < mervaka> buffer size of 4 :/ 2013-02-03T21:14:33 < emeb> interesting, but not exactly useful. 2013-02-03T21:15:12 < mervaka> nope :/ 2013-02-03T21:15:16 < mervaka> but it works. 2013-02-03T21:15:21 < mervaka> in a really hackey way. 2013-02-03T21:17:03 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-41.pptp.stw-bonn.de] has joined ##stm32 2013-02-03T21:18:03 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has joined ##stm32 2013-02-03T21:18:14 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has quit [Client Quit] 2013-02-03T21:26:25 < Tectu> knowing how to use GDB can somehow be quite helpful 2013-02-03T21:26:53 < Erlkoenig> possibly :D 2013-02-03T21:31:33 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds] 2013-02-03T21:33:26 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-03T21:35:15 < emeb> learning how to use GDB can be quite painful 2013-02-03T21:50:20 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-03T21:51:01 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-03T21:52:34 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-03T21:53:20 -!- alexn [~alexn@178.27.128.46] has joined ##stm32 2013-02-03T21:56:04 -!- alexn [~alexn@178.27.128.46] has quit [Remote host closed the connection] 2013-02-03T21:57:08 < karlp> jpa-: have you got qemu running elfs for stm32? what sort of things are you trying to do? 2013-02-03T21:57:37 -!- alexn [~alexn@178.27.128.46] has joined ##stm32 2013-02-03T21:58:09 < jpa-> karlp: just running generic cortex-m3 code 2013-02-03T21:58:27 < jpa-> i'm trying to run some benchmarks for algorithms automatically for each revision 2013-02-03T21:58:34 -!- rob_w [~rob_w@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-03T21:58:49 -!- alexn [~alexn@178.27.128.46] has quit [Remote host closed the connection] 2013-02-03T21:59:08 < Tectu> jpa-, does the std::vector::push_back of the stdlibc++ also allocate twice the space needed like the STL does? 2013-02-03T21:59:26 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-03T22:00:25 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has left ##stm32 [] 2013-02-03T22:00:31 < jpa-> Tectu: it uses x2 resizing, yes (it *is* STL) 2013-02-03T22:00:43 < rob_w> test fILE:/// 2013-02-03T22:00:47 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Remote host closed the connection] 2013-02-03T22:01:18 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-03T22:01:18 < jpa-> Tectu: you can use .reserve() if you know how much you need 2013-02-03T22:01:33 < Tectu> jpa-, I see 2013-02-03T22:01:54 -!- dirty_d [~andrew@76.118.112.158] has joined ##stm32 2013-02-03T22:02:28 -!- rob_w [~rob_w@unaffiliated/rob-w/x-1112029] has quit [Quit: Verlassend] 2013-02-03T22:04:31 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T22:04:34 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-03T22:04:52 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-03T22:07:31 < karlp> bleh, these 3.5mm head phone jacks have switches for when they're connected, 2013-02-03T22:07:35 < karlp> bu tit's super dodgy 2013-02-03T22:07:40 < karlp> maybe there were a little toooo cheap 2013-02-03T22:08:26 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Remote host closed the connection] 2013-02-03T22:09:37 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-03T22:09:37 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-03T22:09:37 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T22:09:39 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-03T22:15:16 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Remote host closed the connection] 2013-02-03T22:18:36 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-03T22:18:37 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-03T22:18:37 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T22:18:37 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-03T22:19:12 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Remote host closed the connection] 2013-02-03T22:20:09 -!- Steffanxx [524834a0@gateway/web/freenode/ip.82.72.52.160] has joined ##stm32 2013-02-03T22:20:19 < Steffanxx> This !@#$% rob_w guy left ? 2013-02-03T22:25:20 < Laurenceb_> !@#$% 2013-02-03T22:26:15 -!- dirty_d [~andrew@76.118.112.158] has quit [Quit: Leaving] 2013-02-03T22:29:47 < gxti> the nerve of that guy 2013-02-03T22:29:57 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-03T22:30:00 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-03T22:30:05 < gxti> fix ur internetz 2013-02-03T22:30:12 <+Steffanx> No, Apple fix your OS 2013-02-03T22:30:18 < gxti> also acceptable 2013-02-03T22:30:42 <+Steffanx> This rob_w guy that joined and quit(ed) a short while ago crashed my client 2013-02-03T22:30:49 <+Steffanx> thanks to fucked up Apple 2013-02-03T22:30:52 < gxti> k 2013-02-03T22:31:05 <+Steffanx> thanks to this: http://thenextweb.com/shareables/2013/02/02/typing-these-eight-characters-will-crash-almost-any-application-on-your-mac/?utm_source=dlvr.it&utm_medium=twitter 2013-02-03T22:31:42 -!- Steffanxx [524834a0@gateway/web/freenode/ip.82.72.52.160] has quit [Quit: Page closed] 2013-02-03T22:32:31 < gxti> quality software 2013-02-03T22:32:33 <+Steffanx> Had to 1@$% use zlog to find out WHAT was crahsing the client 2013-02-03T22:34:23 <+Steffanx> Tomorrow will be pay back time 2013-02-03T22:35:06 < karlp> hang on, so, let me get this right, just typing file:/// myself does the wrong thing to otherpeople? ;) 2013-02-03T22:35:14 <+Steffanx> Yes 2013-02-03T22:35:29 < karlp> or did it have be capitalised like fILE:/// 2013-02-03T22:35:56 <+Steffanx> Doesnt matter i think 2013-02-03T22:36:09 < karlp> I take it you're using a different client now then? 2013-02-03T22:36:12 <+Steffanx> and only when NSTextFields is used on lion 2013-02-03T22:36:26 <+Steffanx> *mountain lion 2013-02-03T22:36:36 <+Steffanx> No, i disabled the 'service' that causes the crash 2013-02-03T22:44:43 < Laurenceb_> test fILE:/// 2013-02-03T22:44:50 < Laurenceb_> did it work? 2013-02-03T22:44:54 < Laurenceb_> aww diabled :( 2013-02-03T22:46:28 <+Steffanx> Haha :P 2013-02-03T22:46:45 < Laurenceb_> pasting to all channels 2013-02-03T22:47:50 <+Steffanx> Dont forget tarduino lol :P 2013-02-03T22:48:09 < Laurenceb_> http://arduino.cc/forum/index.php?topic=110398.0 2013-02-03T22:48:12 < Laurenceb_> lulwut 2013-02-03T22:49:26 < gxti> look at that fabulously even paste application 2013-02-03T22:49:56 < gxti> haha no wonder https://secure.flickr.com/photos/9849051@N04/7390341786/in/photostream/ 2013-02-03T22:50:38 <+Steffanx> Who cares when it works? 2013-02-03T22:50:39 < Laurenceb_> Topic for ##apple set by Branes!~Branes@60-241-114-101.static.tpgi.com.au at Mon Jan 28 18:11:32 2013 2013-02-03T22:50:39 < Laurenceb_> test fILE:/// 2013-02-03T22:50:39 < Laurenceb_> Branes: need some help here ^ 2013-02-03T22:50:54 <+Steffanx> Which channels was that? 2013-02-03T22:51:11 < gxti> congratulations at continuing to be useless Laurenceb_ 2013-02-03T22:51:13 <+Steffanx> oh, apple :P 2013-02-03T22:51:19 -!- BrainDamage [~BrainDama@94.36.244.68] has joined ##stm32 2013-02-03T22:52:36 <+Steffanx> You know the feeling dont you gxti ? 2013-02-03T22:53:55 < gxti> no. instead of attempting to crash people's irc clients, i am going to go assemble 2 more ntp servers and probably write a bootloader. 2013-02-03T22:56:55 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-03T22:57:38 <+Steffanx> Have fun 2013-02-03T22:59:11 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2013-02-03T22:59:50 < Laurenceb_> http://pastebin.com/LQEJeNiL 2013-02-03T22:59:56 < Laurenceb_> conversation with cleverbot 2013-02-03T23:09:42 <+Steffanx> What gxti said Laurenceb_ :P 2013-02-03T23:22:25 < Erlkoenig> they should have set the topic to FILE:/// ... :D 2013-02-03T23:30:15 < emeb> videophool -> http://www.amazon.com/AudioQuest-Carbon-meters-White-Jacket/dp/B003CSU8EK/ 2013-02-03T23:30:21 < emeb> plus, read the reviews... 2013-02-03T23:41:14 -!- phantoneD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-03T23:43:53 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Ping timeout: 255 seconds] --- Day changed Mon Feb 04 2013 2013-02-04T00:25:50 -!- barthess [~barthess@5.100.194.142] has quit [Quit: Leaving.] 2013-02-04T00:26:12 -!- barthess [~barthess@77.67.235.2] has joined ##stm32 2013-02-04T00:54:53 < dongs> lul 2013-02-04T00:55:08 < dongs> low jitter 2013-02-04T01:04:48 <+Steffanx> good morning sir 2013-02-04T01:07:29 < dongs> totally legit projects keeping me busy :( 2013-02-04T01:07:30 < dongs> i hate that. 2013-02-04T01:08:25 -!- DaKu is now known as daku 2013-02-04T01:09:25 -!- barthess [~barthess@77.67.235.2] has quit [Quit: Leaving.] 2013-02-04T01:09:32 * emeb prefers illegit projects. 2013-02-04T01:10:30 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-04T01:13:14 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-04T01:23:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-04T01:26:48 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-04T01:27:59 -!- barthess [~barthess@77.67.235.2] has joined ##stm32 2013-02-04T01:28:48 -!- barthess [~barthess@77.67.235.2] has quit [Client Quit] 2013-02-04T01:32:33 < karlp> bleh, removing printfs broke things. 2013-02-04T01:32:42 < karlp> guess the timing wasn't as robust as I thought... 2013-02-04T01:33:11 < dongs> haha 2013-02-04T01:34:11 < karlp> I thought these prints were before and after the timing, guess not.... :) 2013-02-04T01:38:43 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-04T01:41:44 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-04T01:42:33 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 240 seconds] 2013-02-04T01:42:50 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-04T01:43:39 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 260 seconds] 2013-02-04T01:51:28 < BrainDamage> load bearing printf 2013-02-04T02:03:18 < Thorn> I had far weirder things happen with printf 2013-02-04T02:09:06 -!- uminded [~uminded@216.197.242.145] has joined ##stm32 2013-02-04T02:10:50 < uminded> Hello all 2013-02-04T02:11:10 < emeb> o/ 2013-02-04T02:12:40 < uminded> I am trying to toggle a pin on my STM32F4-Discovery via bit banding for maximum speed. I have the pin setup for 100MHz operation and the bitbanding toggle code is 8 lines of asm so at 168MHz it should be toggling the pin at 10.5MHz but its only going at 500kHz 2013-02-04T02:13:59 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Changing host] 2013-02-04T02:14:00 -!- dfletcher [~fletch@drupal.org/user/72475/view] has joined ##stm32 2013-02-04T02:14:01 -!- dfletcher is now known as drgreenthumb 2013-02-04T02:14:48 < emeb> got your clocks set up right? 2013-02-04T02:15:33 < emeb> ie - make sure that sysclk is actually running 168MHz and all the AHB/APB/GPIO clocks are as you expect. 2013-02-04T02:16:36 < uminded> my sistick is setup at SysTick_Config(SystemCoreClock / 1000) and outputting 1ms pulse (verified on logic analizer) 2013-02-04T02:17:24 < emeb> ummm - don't think that proves anything. the StdPeriph & CMSIS libs automatically scale for clock rate. 2013-02-04T02:18:08 < emeb> Try turning on MCO and directly measuring your clock w/ the scope. 2013-02-04T02:18:10 < uminded> AHB Prescaler is set to 1, and pll and HCLK are at 168mhz 2013-02-04T02:22:57 < uminded> I see the RCC_CFGR_MCO1 and RCC_CFGR_MCO2 structures but no documentation to set up... 2013-02-04T02:24:21 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-04T02:24:25 < Erlkoenig> such things are much easier to achieve via Timers/PWM ... 2013-02-04T02:24:59 < Thorn> +1 2013-02-04T02:26:10 < dongs> http://www.tradesparq.com/uploads/user/0000/0002/5345/product/102043/1287978765.jpg 2013-02-04T02:26:25 -!- BrainDamage [~BrainDama@94.36.244.68] has quit [Quit: Leaving.] 2013-02-04T02:26:41 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has joined ##stm32 2013-02-04T02:36:41 < uminded> http://pastebin.com/2B15UsbK this is my MOC output code but I do not get a clock on PA8 2013-02-04T02:37:44 < Thorn> you can't output 168MHz without dividing it down 2013-02-04T02:38:34 < Thorn> there's a limit for MCO, like 50MHz or whatever (see RM) 2013-02-04T02:39:38 < uminded> I have it div by 5 now and nothing. Do I setup the AF correctly? 2013-02-04T02:40:08 < Thorn> it's actually 100 MHz (5.2.10) 2013-02-04T02:40:40 < emeb> don't see you enabling clocks for GPIOA and AFIO - did you? 2013-02-04T02:41:49 < dongs> https://plus.google.com/u/0/112188920674787719023/posts/ATaPVxYmEPG 2013-02-04T02:41:50 < dongs> raspberry pi 2013-02-04T02:42:26 < emeb> rpi all the things 2013-02-04T02:42:38 < Thorn> >phone calls/Sms with only 3 relays 2013-02-04T02:43:12 < Thorn> does it also pulse dial? 2013-02-04T02:43:22 < dongs> probably 2013-02-04T02:44:03 < uminded> emeb: Got it outputting now. I changed to HSE/4 2013-02-04T02:44:19 < emeb> MCO works? 2013-02-04T02:44:23 < Thorn> and what's the freq 2013-02-04T02:44:28 < emeb> is it at the freq you expected? 2013-02-04T02:45:01 < uminded> emeb: very no... its 186khz 2013-02-04T02:45:08 < emeb> heh 2013-02-04T02:45:27 < dongs> waut 2013-02-04T02:45:27 < emeb> the clock setup code is rather "interesting" 2013-02-04T02:45:39 < dongs> just use the excel app ? 2013-02-04T02:45:42 < emeb> depends on whose library you're using, etc. 2013-02-04T02:45:58 -!- Laurenceb_ [~Laurence@86.177.58.138] has quit [Ping timeout: 245 seconds] 2013-02-04T02:46:10 < Thorn> you probably accidentally set the arduino compatibility bit 2013-02-04T02:46:10 < uminded> emeb: Yea their a few clocks to get right on these guys. The excep app fails horribly using LibreCalc, some VB error 2013-02-04T02:46:45 < emeb> heh. Libre fail. 2013-02-04T02:47:16 < emeb> well, I don't use the spreadsheet. Just read the code & set the bits the way you need. 2013-02-04T02:47:29 < dongs> lul, librecalc 2013-02-04T02:49:43 < uminded> LibreCalc has some room for improvement for for 99% of all work Libra is better than MS Office 2013-02-04T02:51:06 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 276 seconds] 2013-02-04T03:02:23 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-04T03:10:53 < R2COM> hmmm 2013-02-04T03:11:20 < R2COM> wondering of using one of the outputs of stm32 as a slow clock for some chip 2013-02-04T03:11:26 < R2COM> maybe not a good idea though 2013-02-04T03:15:05 < Erlkoenig> should work using a timer/pwm? 2013-02-04T03:15:13 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has quit [Quit: Leaving.] 2013-02-04T03:18:01 < R2COM> just concerned about how clean that clock would be 2013-02-04T03:19:23 < Erlkoenig> take an oscilloscope and look? 2013-02-04T03:20:57 < R2COM> yes. I'll just have to route a trace with appropriate connector to match it 2013-02-04T03:21:45 < R2COM> will make a breakout to test it. 2013-02-04T03:23:14 < emeb> how slow? 2013-02-04T03:24:19 < R2COM> 4MHz, but it might be more too, so well, I'll just test it out 2013-02-04T03:25:06 < emeb> If you're generating 4MHz just by dividing the main clock it should be very clean. 2013-02-04T03:29:56 -!- uminded [~uminded@216.197.242.145] has quit [Ping timeout: 248 seconds] 2013-02-04T03:44:38 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 252 seconds] 2013-02-04T03:49:50 < gxti> so i put together two more ntp servers, one i decided not to tin the qfn's pads first. did not go well. 2013-02-04T03:50:14 < gxti> i need a hot air station anyway... 2013-02-04T03:50:18 < emeb> huh. even w/ lots of flux? 2013-02-04T03:50:49 < gxti> yeah, just didn't wet very well. dicked with it a bit, put it through the toaster oven like i was going to anyway, dicked with it some more, no lights 2013-02-04T03:51:01 < gxti> might be something trapped underneath 2013-02-04T03:51:12 < gxti> the other one is perfect 2013-02-04T03:51:55 < emeb> I've got one of these: http://www.circuitspecialists.com/csihotgun-2.html 2013-02-04T03:51:58 < emeb> seems to work OK. 2013-02-04T03:53:43 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-04T03:54:14 < gxti> probably tinning both sides is necessary so that swabbing the iron across just melts the two together 2013-02-04T03:55:09 < emeb> I've heard that for hand soldering qfns it helps to modify the footprint so the pads extend well beyond the pkg. 2013-02-04T03:55:42 < gxti> i did extend them, although not ridiculously so 2013-02-04T03:55:58 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-04T03:56:22 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-04T03:58:46 < emeb> so what's this I hear - they're having a power outage during the SuperbOwl? 2013-02-04T04:07:21 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-04T04:09:38 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-04T04:10:17 -!- NonaSuomy_ is now known as TeknoJuce 2013-02-04T04:15:11 < gxti> poked it with the iron some more and now the lights all come up but it never hits dhcp... odd 2013-02-04T04:16:45 < emeb_mac> getting closer 2013-02-04T04:21:42 < Thorn> looks like strangest problems in circuits are caused by bad solder jobs 2013-02-04T04:23:24 < emeb_mac> bad solder has caused me a lot of wasted time. 2013-02-04T04:25:04 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-41.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-04T04:26:46 < R2COM> do guest picture uploads from imgur being deleted after some time? 2013-02-04T04:26:51 < R2COM> if person is not registered 2013-02-04T04:27:02 < R2COM> in patebin for example I can give expire date 2013-02-04T04:28:17 < R2COM> pastebin 2013-02-04T04:28:44 < Thorn> I don't think so 2013-02-04T04:28:59 < R2COM> too bad then its being loaded with trash and overloads 2013-02-04T04:35:25 < gxti> got it 2013-02-04T04:35:48 < gxti> that last pad was a bitch, kept going between dry and bridged 2013-02-04T04:38:29 < emeb_mac> happy now? 2013-02-04T04:38:42 < gxti> seems so 2013-02-04T04:38:46 < emeb_mac> sweeet 2013-02-04T04:39:38 < Thorn> I still don't know why my black magic clones work intermittently with lpc11c2x 2013-02-04T04:39:40 < emeb_mac> what's the phy you 2013-02-04T04:39:46 < emeb_mac> 're using? 2013-02-04T04:40:02 < gxti> ksz8031 2013-02-04T04:40:27 < Thorn> but both of them have the problem and I replaced the lpc11 after it got toasted and that didn't change things 2013-02-04T04:40:32 < gxti> i thought it was the cheapest available but apparently microchip has one that's a little bit cheaper 2013-02-04T04:41:37 < gxti> seems to fluctuate a bit. at one point i bought the last 3 KSZ8031 in stock at digikey and the next cheapest was 2x the cost 2013-02-04T04:42:06 < gxti> now there's the microchip thing, and also KSZ8081 2013-02-04T04:43:25 < gxti> same pinout, wonder what the downside is 2013-02-04T04:44:01 < emeb_mac> the mchp parts are actually smsc parts? 2013-02-04T04:44:10 < emeb_mac> lan87x0? 2013-02-04T04:45:04 < gxti> seems like it 2013-02-04T04:45:04 < emeb_mac> also qfn / csp 2013-02-04T04:45:09 < emeb_mac> only 2013-02-04T04:45:45 < gxti> most are, you don't see leaded packages until the $3 range 2013-02-04T04:46:40 < emeb_mac> yeesh - didn't check the prices on those. they're pretty cheap. 2013-02-04T04:47:00 < gxti> ah, mchp bought smsc in august 2013-02-04T04:47:04 < emeb_mac> yup 2013-02-04T04:47:27 < gxti> that part is bigger than the micrel ones though, since it does mii 2013-02-04T04:47:30 < emeb_mac> buddy who works there says that they're losing money fast because of it. 2013-02-04T04:47:47 < gxti> yeah well nobody really knows how microchip is still alive 2013-02-04T04:48:17 < emeb_mac> it's a strange business 2013-02-04T04:48:32 < gxti> hmm, the smsc part might be easier to hand-solder though since the pads wrap around the base 2013-02-04T04:48:43 < gxti> i already bought 10 of the micrels though :p 2013-02-04T04:48:46 < emeb_mac> aha - up the sides. 2013-02-04T04:49:13 < gxti> in any case any further ntp servers will be done with paste stencil 2013-02-04T04:54:54 < gxti> dang, the murata psu module kicks the shit out of mine 2013-02-04T04:55:40 < emeb_mac> the little custom one you did? 2013-02-04T04:57:04 < gxti> yeah, it's smaller, lighter, cheaper, muuuuch less noisy, and is a little less frightening when riding through an output short 2013-02-04T04:57:24 < gxti> although immediately after writing that i seem to have run into a problem 2013-02-04T04:59:34 < gxti> of course now i can't reproduce it. i had it oscillating or something. 2013-02-04T05:30:23 < dongs> time to do what i hate most, coding lunix shit 2013-02-04T05:30:45 < R2COM> come on bill gates uses linux 2013-02-04T05:30:58 < R2COM> he created win8 out of linux 2013-02-04T05:31:16 < dongs> i'd almost agree with that 2013-02-04T05:31:20 < dongs> with so much fail 2013-02-04T05:44:23 < gxti> win8 runs on arm. linux runs on arm. therefore, win8 = linux QFT 2013-02-04T05:50:44 < R2COM> long time ago I saw some article about winNT server I think being installed on some battle cruiser to control some systems as an experiment I guess, and it went shit 2013-02-04T05:51:13 < gxti> cool story 2013-02-04T05:57:44 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-04T06:05:55 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-04T06:06:05 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-04T06:52:23 < R2COM> actually here it is: http://en.wikipedia.org/wiki/USS_Yorktown_(CG-48) 2013-02-04T06:53:43 < R2COM> deputy technical director gave his comments about it 2013-02-04T06:55:57 -!- ReggieUK [ReggieUK@90.202.86.231] has quit [] 2013-02-04T07:10:37 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-04T07:33:13 -!- fxd0h [~fx@186.18.172.240] has quit [Quit: so long!] 2013-02-04T07:47:17 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-04T07:47:39 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-04T08:33:55 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-04T08:38:17 < dongs> http://i.imgur.com/gmaVQV6.jpg lol dongs 2013-02-04T08:40:46 < baird> lol if they'll actually work 2013-02-04T08:41:27 < dongs> I don't *usually* make fail 2013-02-04T08:41:30 < dongs> but when I do... 2013-02-04T08:44:18 < baird> I'm thinking I'll probably be up for two panels. Being functional will be something of a condition, though. 2013-02-04T08:49:24 < dongs> duh 2013-02-04T08:54:04 < emeb_mac> nice boards 2013-02-04T09:01:44 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-04T09:06:29 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-04T09:23:31 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-04T09:28:23 -!- cdh [~chatzilla@c-50-136-234-52.hsd1.ca.comcast.net] has quit [Ping timeout: 252 seconds] 2013-02-04T09:43:26 -!- cdh [~chatzilla@c-50-136-234-52.hsd1.ca.comcast.net] has joined ##stm32 2013-02-04T09:43:47 -!- dekar [~dekar@212.255.33.103] has joined ##stm32 2013-02-04T09:43:48 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-04T09:46:11 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-04T09:47:05 -!- dekar_ [~dekar@212.255.31.181] has quit [Ping timeout: 252 seconds] 2013-02-04T09:54:46 < dongs> [feenode] Xavierdarkness [~xavierdar@unaffiliated/xavierdarkness] requested unknown CTCP IMNEVERAL0NE from dongs: 2013-02-04T09:54:49 < dongs> wat 2013-02-04T10:25:32 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 240 seconds] 2013-02-04T10:55:46 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-04T10:56:25 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-04T11:00:36 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 248 seconds] 2013-02-04T11:04:32 < zyp> dongs, nice boards 2013-02-04T11:39:26 < R2COM> zyp: I decided to use cpld too, for that stuff we discussed the other day 2013-02-04T11:40:02 < R2COM> to switch between which group of PWM channels will be passed to output PWM channel 2013-02-04T11:40:25 < R2COM> and by the way I did the Input selector PWM measurement on cpld too 2013-02-04T11:40:36 < R2COM> so, dont need now to use Input compare from stm32 at all 2013-02-04T11:40:53 < karlp> gxti: 8081 looks like a pin compat replacement for 8031, but with lower power consumption 2013-02-04T11:41:01 < karlp> at leas,t that's the only differences I can find in the sheets 2013-02-04T11:41:07 < R2COM> so one control input (pwm like) is like a switch, it controls which signals will go to output 2013-02-04T11:41:50 < R2COM> CPLD costs 1.45$ 2013-02-04T11:42:04 < R2COM> already simulated design, it barely but fits the chip 2013-02-04T11:42:16 < R2COM> heres how I did it: 2013-02-04T11:42:17 < R2COM> http://pastebin.com/Vvda7zqX 2013-02-04T11:42:29 < R2COM> and its almost at the limit: http://i.imgur.com/pE75xOg.png 2013-02-04T11:42:35 < R2COM> http://i.imgur.com/aaqb8cD.png 2013-02-04T11:46:17 < dongs> vhdl is even less readable than veriblog 2013-02-04T11:47:24 < zyp> hmm, I don't get what it does, but ok 2013-02-04T11:47:36 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-04T11:47:37 < dongs> which cpld did you use? XC264A or something? 2013-02-04T11:48:17 < dongs> oh xc232a 2013-02-04T11:49:40 < R2COM> xc2c32a 2013-02-04T11:50:17 < R2COM> it simply does this: measure PWM pulse width of control signal, if its width is big enough pass port 1 PWM channels to output, if not pass port 2 PWM channels to output, thats it. 2013-02-04T11:50:53 < R2COM> xc232a is cheapest, and smallest 2013-02-04T11:55:04 < dongs> and pin compatible to 644 2013-02-04T11:55:12 < dongs> incase you fuck up and need it to do something more 2013-02-04T11:55:14 < R2COM> hmm not sure on that did not check 2013-02-04T11:55:18 < dongs> it is 2013-02-04T11:55:21 < dongs> ive used both in stuff here 2013-02-04T11:55:46 < dongs> if you buy more than one theres basically no price diff between 64 and 32 2013-02-04T11:55:58 < R2COM> yeah 2013-02-04T11:55:59 < dongs> and might acutally end up being cheaper 2013-02-04T11:56:01 < dongs> due to availability 2013-02-04T11:57:31 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-04T11:57:48 < R2COM> cpld's are great, this particularly one done on 0.18um process I guess 2013-02-04T11:58:10 < R2COM> pretty cheap for xilinx to fab it 2013-02-04T11:58:49 < R2COM> ok, will go take a nap 2013-02-04T11:59:32 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-04T11:59:32 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-04T12:00:21 < zyp> the only annoying part is the required toolchains to use them 2013-02-04T12:00:50 < dongs> indeed 2013-02-04T12:00:57 < dongs> xilinx suite is giant fucking bloatware 2013-02-04T12:19:07 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-04T12:21:41 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has quit [Ping timeout: 245 seconds] 2013-02-04T12:29:28 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-04T12:33:43 < dongs> does dickstarter allow separate billing/shipping adress 2013-02-04T12:33:49 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Remote host closed the connection] 2013-02-04T12:34:05 < dongs> i want to get me that bladerf trash but dickstarter doesnt do paypal 2013-02-04T12:34:10 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-04T12:38:38 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-04T12:46:13 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has joined ##stm32 2013-02-04T12:48:03 < jpa-> wtf, something keeps resetting my RTC on power on, even though Vbat remains nice and steady 3.3V all the time 2013-02-04T12:48:20 < jpa-> and it's not the software as reset by NRST doesn't do the same 2013-02-04T12:59:43 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-04T13:01:27 < Laurenceb_> hmm 2013-02-04T13:01:37 < Laurenceb_> i use rtc, but not like that 2013-02-04T13:01:54 < Laurenceb_> i leave vbat disconnected, and keep power applied all the time 2013-02-04T13:02:07 < Laurenceb_> then go to sleep mode 2013-02-04T13:02:25 < Laurenceb_> do you initialise the RTC at power on? 2013-02-04T13:04:33 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-04T13:04:35 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-04T13:05:34 < Laurenceb_> jpa-: isnt there some sort of power switch you need to set? 2013-02-04T13:06:44 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-04T13:17:55 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-04T13:25:33 -!- phantoneD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-04T13:26:02 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-04T13:35:41 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection] 2013-02-04T13:37:21 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-04T13:47:17 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-04T13:48:30 < baird> oh snap. http://gizmodo.com/5980842/there-is-blatant-racist-and-sexist-language-in-github-code 2013-02-04T13:51:27 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has joined ##stm32 2013-02-04T13:57:40 < karlp> anyone got any examples of triggering adc conversions off the trgo output of a timer? 2013-02-04T13:57:51 < karlp> I ge ta single conversion, but it doesn't keep triggering. 2013-02-04T13:58:24 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-04T14:08:55 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Quit: pokðt] 2013-02-04T14:12:29 -!- jef79m is now known as jef79m_lurking 2013-02-04T14:16:30 < dongs> baird: We reached out to GitHub to talk about what's going on here, and we'll update when we've heard back. 2013-02-04T14:16:34 < dongs> wuuuuuuuuuut 2013-02-04T14:17:03 < Thorn> many of the matches seem to be in stoplists and other censorship tools 2013-02-04T14:17:46 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-76.pptp.stw-bonn.de] has joined ##stm32 2013-02-04T14:18:10 < baird> "Git" itself is a curse-word in Commonwealth English... 2013-02-04T14:18:28 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-04T14:19:51 <+Steffanx> LOL WHAT A bullshit baird 2013-02-04T14:19:58 <+Steffanx> Some people care too much about certain things 2013-02-04T14:20:35 < baird> 'Professional Victims'... 2013-02-04T14:23:13 -!- claude is now known as Claude 2013-02-04T14:27:52 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Read error: Connection reset by peer] 2013-02-04T14:27:52 -!- scrts [~quassel@unaffiliated/scrts] has quit [Quit: No Ping reply in 180 seconds.] 2013-02-04T14:27:52 < Erlkoenig> what's bad about git? 2013-02-04T14:27:52 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-04T14:27:52 < baird> Australian Baby-boomers, at least, use it as you would 'bastard', or the like. 2013-02-04T14:27:52 -!- scrts_ [~quassel@2a01:348:70:46:17:57:19:0] has joined ##stm32 2013-02-04T14:27:52 < karlp> baird: I don't think he was calling bullshit on git, but on the article. 2013-02-04T14:27:52 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-04T14:28:27 < baird> "fuck off you short git", "ignorant git", etc. (karlp: was more replying to Erlkoenig) 2013-02-04T14:28:35 < Erlkoenig> oh yeay 2013-02-04T14:28:53 < Erlkoenig> at least it is googleable... unlike some other software o.O 2013-02-04T14:30:27 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-04T14:33:08 < karlp> so, this might be a problem: #define ADC_CR2_EXTSEL_TIM6_TRGO (10 << ADC_CR2_EXTEN_SHIFT) 2013-02-04T14:33:20 < karlp> use the right shift numbnums 2013-02-04T14:42:50 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has joined ##stm32 2013-02-04T14:47:12 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-04T14:51:40 -!- scrts_ is now known as scrts 2013-02-04T14:51:45 -!- scrts [~quassel@2a01:348:70:46:17:57:19:0] has quit [Changing host] 2013-02-04T14:51:45 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32 2013-02-04T14:55:46 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-04T14:56:03 < jpa-> karlp: set the CONT bit 2013-02-04T14:58:57 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-04T15:01:05 < Thorn> can anyone comment on my FSM technique? please no racial, misogynistic or otherwise discriminatory slurs http://pastebin.com/nJLWLJEi 2013-02-04T15:01:16 -!- daku is now known as DaKu 2013-02-04T15:13:41 < dongs> lol, gay freertos !! 2013-02-04T15:13:53 < dongs> wtf, variable declarattion mid-scope 2013-02-04T15:14:08 < Thorn> --std=gnu99 2013-02-04T15:14:13 < dongs> ugly, inconsistent formatting 2013-02-04T15:14:35 < dongs> ok thats all for my commetns :) 2013-02-04T15:14:35 < Thorn> where, braces? they're consistent 2013-02-04T15:14:43 < dongs> no they're not 2013-02-04T15:14:53 < dongs> if(++dp_blink_counter > 4) dp_blink_counter = 0; 2013-02-04T15:17:55 < Thorn> okay 2013-02-04T15:20:23 < Thorn> but my main question is how to organize the state machine better 2013-02-04T15:32:58 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-04T15:34:51 < Erlkoenig> basically a CPU is a FSM so one doesn't need to implement one... where the current state is represented by the program counter / the current line in the code AND the variables 2013-02-04T15:35:42 <+Steffanx> thanks for the info :P 2013-02-04T15:35:44 < karlp> jpa-: no, don't need cont, just needed to actually set the right trigger bits :) 2013-02-04T15:37:02 < Thorn> I agree it's weired that you need to implement a FSM on top of another FSM but is there an alternative. 2013-02-04T15:37:39 < zyp> silly argument 2013-02-04T15:38:54 < zyp> if you use that logic, you could argue that every latch in the cpu is a (very simple) fsm and it's dumb to use it to build a more complex fsm 2013-02-04T15:39:25 <+Steffanx> ha 2013-02-04T15:40:44 < Thorn> you can also say that your custom IP core is a single instruction CPU 2013-02-04T15:41:51 < Thorn> http://applejack.ponychan.net/chan/files/src/134854337706.jpg 2013-02-04T15:43:24 < Erlkoenig> also FSM are boring... implement a turing machine 2013-02-04T16:14:25 < karlp> Thorn: can you repaste that? pastebin's giving me invalid pasteid, did it expire? 2013-02-04T16:14:49 < Thorn> yes it had a 1 hour lifetime 2013-02-04T16:15:21 < Erlkoenig> it burned after reading 2013-02-04T16:15:42 < Thorn> http://pastebin.com/bYx3DWk2 2013-02-04T16:15:57 < jpa-> karlp: but i like the CONT bit 2013-02-04T16:16:41 < jpa-> but wtf is wrong with my rtc, it resets on shutdown 2013-02-04T16:17:50 < karlp> isn't there a mode for autoreset vs preserve? 2013-02-04T16:17:57 < karlp> one of the demos I looked at did that, one tick. 2013-02-04T16:18:01 < jpa-> haven't found info about any, STM32F4 2013-02-04T16:19:18 < jpa-> here is all i know: rev 1 prototype works, rtc keeps time. rev 2 prototype RTC resets about half of times at shutdown - i know this because the RTC oscillator stops also 2013-02-04T16:19:32 < jpa-> both running exactly same binary 2013-02-04T16:20:13 < jpa-> the fall time of Vdd is a bit slower (100ms vs. 20ms) on rev2 prototypedue to more capacitors 2013-02-04T16:20:30 < jpa-> but i don't understand why it should matter 2013-02-04T16:21:39 < jpa-> i have brown out detection on so NRST goes low quickly like it should on shutdown 2013-02-04T16:21:56 < karlp> In addition, the RTC keeps on running under system reset if the reset source is different 2013-02-04T16:21:59 < karlp> from the power-on reset one 2013-02-04T16:22:05 < karlp> differences in that? 2013-02-04T16:22:34 < jpa-> on both devices RTC runs fine if i just reset with NRST 2013-02-04T16:22:45 < jpa-> only when Vdd goes low there is a problem 2013-02-04T16:23:04 < karlp> same behaviour with BOD off? 2013-02-04T16:23:57 < jpa-> yes, then the NRST goes like --(3.3V)---__-(1.6V)-____ on shutdown but behaviour is the same 2013-02-04T16:26:54 < karlp> well, teh code I was looking at was just two different methods, one to "start and reset from zero" and one to "start" 2013-02-04T16:27:00 < karlp> so nothing that would be affecting you. 2013-02-04T16:27:02 < karlp> no idea sorry 2013-02-04T16:32:37 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-04T16:36:49 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [] 2013-02-04T16:46:21 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-04T16:48:13 < jpa-> does someone know what PDR_ON does on stm32f4? 2013-02-04T16:48:46 < zyp> didn't it get removed in later silicon revisions? 2013-02-04T16:49:28 < jpa-> wait.. who the heck has sold me an 'A' revision chip 2013-02-04T16:49:48 < emeb_mac> barrel scrapings 2013-02-04T16:49:56 < zyp> 2.1.6 in errata sheet 2013-02-04T16:50:12 < jpa-> i have LQFP144 2013-02-04T16:50:31 < jpa-> but should i connect it low or high? 2013-02-04T16:50:37 < jpa-> or leave unconnected 2013-02-04T16:51:17 < jpa-> maybe high 2013-02-04T16:51:26 < gxti> connect to 120VAC 2013-02-04T16:51:32 < jpa-> ok brb 2013-02-04T16:51:42 < zyp> 2.2.16 in datasheet 2013-02-04T16:51:51 < dongs> zyp, im a tool. i forgot the part# for SDIO angled header again 2013-02-04T16:51:57 < Laurenceb> baird: "casually riddled with things like bitch, slut, whore, and worse." 2013-02-04T16:52:07 < Laurenceb> wont someone think of the children??!!! 2013-02-04T16:52:30 < jpa-> zyp: ah, so high - thanks 2013-02-04T16:52:49 < zyp> dongs, S5484-ND 2013-02-04T16:53:05 < dongs> thankies 2013-02-04T16:53:46 < Laurenceb> i need therapy 2013-02-04T16:53:52 < Laurenceb> someone said the bitch word 2013-02-04T16:53:52 < jpa-> hmm i do have it connected high 2013-02-04T16:53:55 < jpa-> so it's not that 2013-02-04T16:54:09 < emeb_mac> Laurenceb: did a matlab/octave simulation of the SDR stuff yesterday. 2013-02-04T16:54:21 < emeb_mac> summary - won't work on F303 2013-02-04T16:55:03 < emeb_mac> front-end tuner/integrator needs to oversample by > 4x to get good out-of-band rejection. 2013-02-04T16:55:13 < emeb_mac> and F303 doesn't have the speed to do that. 2013-02-04T16:55:24 < Laurenceb> hmm 2013-02-04T16:55:30 < Laurenceb> cant you just do windowing 2013-02-04T16:55:36 < Laurenceb> with no oversampling 2013-02-04T16:55:47 < emeb_mac> windowing broadens the passband significantly 2013-02-04T16:56:06 < emeb_mac> w/o oversampling that means more aliasing 2013-02-04T16:57:16 < Laurenceb> yeah 2013-02-04T16:57:22 < Laurenceb> but it might be good enough 2013-02-04T16:57:24 < Laurenceb> i dunno 2013-02-04T16:57:36 < Laurenceb> cant hurt to try.. tho itd take a fair bit of work 2013-02-04T16:58:05 < emeb_mac> if you want to play with the processing the matlab/octave is here: https://github.com/emeb/f303_sdr/blob/master/support/tst_proc.m 2013-02-04T16:58:39 < Laurenceb> ive got lots to do already :P 2013-02-04T16:58:54 < emeb_mac> heh 2013-02-04T17:00:08 < jon1012> did you guys see the new boards for the stm32f4 discovery ? 2013-02-04T17:00:46 <+Steffanx> Laurenceb didn't post photos of his board yet afaik :P 2013-02-04T17:00:52 < Laurenceb> you mean with new caps and stuff? 2013-02-04T17:00:58 < Laurenceb> yeah ive got some like that 2013-02-04T17:01:15 < Laurenceb> http://ecx.images-amazon.com/images/I/41qXwIzI2oL._SL500_.jpg 2013-02-04T17:01:32 < Laurenceb> ^wtf - isnt expresscard too thin to fit usb? 2013-02-04T17:01:59 < dongs> too thin in waht sense? 2013-02-04T17:02:13 < Laurenceb> to fit usb connectors 2013-02-04T17:02:16 < dongs> its probably s ome breakout? 2013-02-04T17:02:23 < dongs> "micro" usb 3.0 is pretty thin 2013-02-04T17:02:25 < Laurenceb> yeah thats what im thinking 2013-02-04T17:02:27 < dongs> about same profile as microusb 2013-02-04T17:02:47 < Laurenceb> nothing about cable in description.... 2013-02-04T17:03:00 < Laurenceb> i need an adaptor for my bladerf 2013-02-04T17:03:02 <+Steffanx> Oh, Laurenceb has no fancy USB3.0 ports for his @#$% fantastic SDR? 2013-02-04T17:03:57 < Laurenceb> nope 2013-02-04T17:05:26 < jon1012> those new daugter boards, someone hard about it ? http://www.element14.com/community/docs/DOC-51084/l/element14-discover-more--an-expansion-board-for-stm32f4discovery 2013-02-04T17:05:38 < jon1012> for ethernet + lcd + cam 2013-02-04T17:05:45 < Laurenceb> old 2013-02-04T17:05:57 < jon1012> well rather this : http://www.st.com/internet/evalboard/product/255417.jsp 2013-02-04T17:05:59 < jon1012> oh, really ? 2013-02-04T17:06:13 < Laurenceb> oh wtf 2013-02-04T17:06:22 < Laurenceb> i was thinking farnell made them 2013-02-04T17:06:23 < Laurenceb> lo9l 2013-02-04T17:07:00 < Laurenceb> http://www.st.com/internet/com/MULTIMEDIA_RESOURCES/IMAGE/BOARD_PHOTO/stm32f4dis-ext.jpg 2013-02-04T17:07:11 < Laurenceb> that looks almost as bad as some arduino fail 2013-02-04T17:08:36 -!- NonaSuomy_ is now known as TeknoJuce 2013-02-04T17:08:39 < jon1012> bwah 2013-02-04T17:10:55 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has quit [Changing host] 2013-02-04T17:10:55 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-04T17:11:00 < dongs> ok got enough parts for 14 DP shits 2013-02-04T17:11:18 < zyp> nice 2013-02-04T17:11:23 < Laurenceb> dont use that word 2013-02-04T17:11:30 < Laurenceb> gizmodo will get angry 2013-02-04T17:11:31 < dongs> shut up nigger 2013-02-04T17:11:48 < dongs> that article is so fucking retarded 2013-02-04T17:11:57 < dongs> what exactly did they expect to hear back from github anyway? 2013-02-04T17:12:03 < Laurenceb> roflmao 2013-02-04T17:12:08 < dongs> "sorry, we're gonna rush out and delete all bad words asap"??? 2013-02-04T17:12:35 < Laurenceb> "your github project has been deleted due to: bad words" 2013-02-04T17:12:56 < Laurenceb> some time later : no projects left on github 2013-02-04T17:12:58 < dongs> "your fucking github has been deleted because you used the n-word" 2013-02-04T17:13:21 < Laurenceb> github has been deleted cuz it used the g word 2013-02-04T17:14:08 < emeb_mac> only a ginger can call another ginger ginger. 2013-02-04T17:14:12 -!- mode/##stm32 [+o Steffanx] by ChanServ 2013-02-04T17:14:17 -!- alexn was kicked from ##stm32 by Steffanx [alexn] 2013-02-04T17:14:34 < Laurenceb> lol what did alexn do? 2013-02-04T17:14:38 -!- mode/##stm32 [-o Steffanx] by ChanServ 2013-02-04T17:14:49 < emeb_mac> used the g word 2013-02-04T17:14:57 <+Steffanx> payback for what he and his collegue did yesterday :P 2013-02-04T17:15:01 < jon1012> woot ? 2013-02-04T17:15:06 < emeb_mac> file:///? 2013-02-04T17:15:06 <+Steffanx> test File:/// <= that :P 2013-02-04T17:15:11 < Laurenceb> hehehe 2013-02-04T17:15:58 < emeb_mac> mtn lion fail 2013-02-04T17:16:00 < Laurenceb> omgzorz rpi model A for sale 2013-02-04T17:16:09 < Laurenceb> *nerdfap* 2013-02-04T17:16:13 < emeb_mac> take my money! 2013-02-04T17:16:42 < zyp> now even cheaper and more useless! 2013-02-04T17:16:49 < Laurenceb> heh 2013-02-04T17:16:58 < emeb_mac> less memory, fewer I/O! 2013-02-04T17:17:07 < BrainDamage> 200% more hype 2013-02-04T17:17:13 < emeb_mac> not possible 2013-02-04T17:17:23 < BrainDamage> I'm gonna mark your words 2013-02-04T17:17:26 * jon1012 will sell some :) 2013-02-04T17:17:31 <+Steffanx> Wonder how many people actually want this model 2013-02-04T17:17:38 < dongs> modelA? 2013-02-04T17:17:43 < zyp> isn't the difference just that it lacks the usb-hub/ethernet chip? 2013-02-04T17:17:44 <+Steffanx> Yes 2013-02-04T17:17:50 < jon1012> I'll buy 4 and put them on ebay, let's try to see if they sell 2013-02-04T17:17:50 <+Steffanx> And less memory 2013-02-04T17:17:52 < emeb_mac> won't someone please think of the children? 2013-02-04T17:18:04 < Laurenceb> not you baird 2013-02-04T17:18:12 < dongs> .. 2013-02-04T17:18:15 < dongs> tehy didnt even relayout the shit 2013-02-04T17:18:20 < dongs> http://www.raspberrypi.org/wp-content/uploads/2013/02/Model-A.jpg 2013-02-04T17:18:24 <+Steffanx> No, why should they? 2013-02-04T17:18:26 < dongs> same insane connector placement 2013-02-04T17:19:11 < Laurenceb> lol they didnt swap to new stencils 2013-02-04T17:19:21 <+Steffanx> Who cares? :) 2013-02-04T17:19:34 < zyp> since it's not hardwired to a hub anymore, it could be used for OTG 2013-02-04T17:19:38 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-04T17:19:49 < zyp> if they had an OTG connector :p 2013-02-04T17:20:10 <+Steffanx> Time to test your (de)soldering skillz 2013-02-04T17:20:17 < jpa-> bah, i give up 2013-02-04T17:20:20 < dongs> http://i.imgur.com/FzeMyAn.png apparently this is the DAC on shitberrypi 2013-02-04T17:20:25 < jpa-> i can find no reason why my rtc resets :F 2013-02-04T17:20:41 < emeb_mac> PWM FTW! 2013-02-04T17:20:41 < jpa-> have to assemble a second prototype later and see if it does the same 2013-02-04T17:21:04 < BrainDamage> dongs: for audio? 2013-02-04T17:21:08 < emeb_mac> ya 2013-02-04T17:21:20 < zyp> that's just a bandpass filter 2013-02-04T17:21:29 < dongs> BrainDamage: yea 2013-02-04T17:21:35 < emeb_mac> if you want digital audio you have to use the hdmi connector 2013-02-04T17:21:49 < dongs> emeb_mac: bzzt. i bet the driver for that doesnt exist 2013-02-04T17:21:49 < BrainDamage> ok that sigma delta converters only need a single pole to work properly, but not even making it active is a bit pushing the cheap a lil far 2013-02-04T17:22:07 < emeb_mac> dongs: dunno, but wouldn't be surprised 2013-02-04T17:22:18 < BrainDamage> or you have to pay 2013-02-04T17:22:33 < emeb_mac> BrainDamage: but those pins aren't even hooked to S/D converters. 2013-02-04T17:22:41 < emeb_mac> just raw PWM. 2013-02-04T17:22:44 < Laurenceb> just to pwm 2013-02-04T17:22:46 < BrainDamage> oh god 2013-02-04T17:22:49 < Laurenceb> yeah 2013-02-04T17:23:08 <+Steffanx> "it works" 2013-02-04T17:23:14 < zyp> dongs, I think it does, judging by all the people buying them to hook to their TVs 2013-02-04T17:23:15 < dongs> "so does lunix" 2013-02-04T17:23:16 < Laurenceb> and limited current sink/source 2013-02-04T17:23:35 < zyp> if hdmi audio weren't working, there would be much rage about that 2013-02-04T17:23:38 < zyp> and I haven't heard any 2013-02-04T17:23:47 <+Steffanx> Audio works perfectly :) 2013-02-04T17:23:51 <+Steffanx> over hmdi 2013-02-04T17:25:07 < emeb_mac> so - everyone who's supporting bladerf have their rpis ready? 2013-02-04T17:25:35 <+Steffanx> You "need" USB3.0 too :) 2013-02-04T17:25:35 < dongs> still lookin for someone to paypal that for me 2013-02-04T17:26:01 < emeb_mac> yeah - sux that ks won't take pp 2013-02-04T17:26:19 < emeb_mac> fail 2013-02-04T17:27:17 < zyp> emeb_mac, I guess I need to ask for my rpi back in a few months time then 2013-02-04T17:27:22 <+Steffanx> Why support paypal when it is just as easy for people to get a CC? 2013-02-04T17:27:28 < zyp> depending on how delayed it becomes 2013-02-04T17:27:46 <+Steffanx> Just as delayed as that openviszla project 2013-02-04T17:28:30 < zyp> Steffanx, they have preprod boards now, so I suspect I might get it this year if I'm lucky! 2013-02-04T17:28:38 < dongs> haha 2013-02-04T17:28:50 < dongs> maybe your china lcd panel will arrive before openvizslaa 2013-02-04T17:28:58 <+Steffanx> They had board last year.. or was that back in 2011? 2013-02-04T17:29:02 <+Steffanx> *boards 2013-02-04T17:29:22 < zyp> I'm talking about last weeks update 2013-02-04T17:29:30 <+Steffanx> Oct. 21, 2011 :) 2013-02-04T17:29:41 <+Steffanx> Yes, those crappy backers only messages 2013-02-04T17:29:44 < zyp> preprod != prototype 2013-02-04T17:30:19 <+Steffanx> I know, but still. 2013-02-04T17:30:35 < zyp> dongs, I'd be surprised if not 2013-02-04T17:30:39 < dongs> just wait till you see awesome software 2013-02-04T17:30:48 < dongs> probably some hacked up trash on top of 'SUMP' 2013-02-04T17:31:13 < zyp> I'm just wondering if I'll get your adapter board or lcd first :p 2013-02-04T17:31:23 <+Steffanx> Ha, pwnd 2013-02-04T17:31:23 < dongs> dunno, im gona make some thursday 2013-02-04T17:31:38 < dongs> stencils + digikey shows up then 2013-02-04T17:31:48 < dongs> though maybe stencils a day earlier, and ive got parts for 4 here already anyway. 2013-02-04T17:32:21 <+Steffanx> I bet digikey sent you the wrong parts 2013-02-04T17:32:32 < dongs> no. i have enough right parts to know if shit works or not. 2013-02-04T17:33:36 <+Steffanx> We'll see 2013-02-04T17:36:05 -!- DaKu is now known as daku 2013-02-04T17:41:08 < Laurenceb> if i set timer ARR to n, is timer period n ticks or n+1 ticks? 2013-02-04T17:42:00 < dongs> pretty sure its +1 2013-02-04T17:42:05 < dongs> because zero 2013-02-04T17:42:53 < Laurenceb> i see 2013-02-04T17:43:11 < Laurenceb> so timer resets if it gets a clk and cnt==arr 2013-02-04T17:44:51 < dirty_d> zyp, where did you rad about what a cdc acm device driver actually has to do to communicate? 2013-02-04T17:44:53 < dirty_d> read* 2013-02-04T17:45:14 < dongs> "read"? dont you just open a bulk pipe and send/receive over it? 2013-02-04T17:45:20 < dongs> nothing special 2013-02-04T17:45:20 < dirty_d> i got it sort of half-working 2013-02-04T17:45:40 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-04T17:45:51 < dirty_d> if i do `echo "hello" > /dev/ttyACM0` it sits there for about 5 seconds, then my device reads it 2013-02-04T17:46:08 < dirty_d> if i try to write from my device, i never get the CTR_TX flag 2013-02-04T17:46:09 < dongs> you should check stuff in the USB interrupt 2013-02-04T17:46:12 < dongs> if theress om data, read it 2013-02-04T17:46:13 < dongs> or someshit 2013-02-04T17:46:15 < dirty_d> yup 2013-02-04T17:47:16 < dirty_d> ep 0 rx/tx works fine, ep 1 rx works, not sure about the huge delay though, and ep 1 tx seesm to not be working 2013-02-04T17:48:31 < karlp> make sure you don't have anything dumb like modem-manager trying to probe your device for at commands... 2013-02-04T17:48:51 < dirty_d> karlp, i havent gotten any thing on ep 2 2013-02-04T17:49:01 < dirty_d> which is the AT command ep 2013-02-04T17:49:25 < gxti> fuser /dev/ttyACM0 2013-02-04T17:49:59 < gxti> acm is kinda wonky sometimes too, i always have to run minicom after plugging something in then close it or pyserial just gives me garbage. need to figure out what the ioctl i'm missing is. 2013-02-04T17:50:15 < dirty_d> hmm 2013-02-04T17:50:28 < dirty_d> thats weird 2013-02-04T17:50:38 < Erlkoenig> i need to have pullups to +12V on my STM32F4's 5V-tolerant pins. Since these have a zero max injection current, i need to protect them from the overvoltage. Because the voltage regulation on the F4 Discovery can't sink any current, clamping diodes wouldn't work. Can i use zener diodes with 3.9V to GND instead? 2013-02-04T17:51:05 < zyp> should work fine 2013-02-04T17:51:31 < gxti> that would keep things from exploding but the line would be at 3.9V, so what's the point? 2013-02-04T17:52:09 < dirty_d> if you use a 3.9v zener isnt the line going to be at (12 - 3.9)V ? 2013-02-04T17:52:12 < Erlkoenig> then the F4 could pull it to GND / 3V when the pins are configured as output 2013-02-04T17:52:39 < gxti> Erlkoenig: the diode will always be there though, and it will always short any excess voltage past 3.9 2013-02-04T17:52:53 < gxti> Erlkoenig: if you want a 12v open drain line you need an external mosfet 2013-02-04T17:53:01 < Erlkoenig> yes but that won't be much current because the pullup has like 10k 2013-02-04T17:53:08 < Erlkoenig> wait a sec i think i should draw a small schematic 2013-02-04T17:53:15 < dirty_d> nevermind 2013-02-04T17:53:21 < gxti> current isn't the issue 2013-02-04T17:54:28 < gxti> either the internal protection diodes or the external one you supply will always clamp the voltage, regardless of the pin state 2013-02-04T17:54:56 < gxti> so plop down a cheap n-fet with appropriate drain voltage rating and you're golden 2013-02-04T17:55:35 < dirty_d> zyp, whats the deal with the ZLP on ep0 in your cdc_acm driver? 2013-02-04T17:55:40 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-04T17:55:56 < Erlkoenig> you mean pull the line to 0V via the fet and have the pullup pull it to 12V? 2013-02-04T17:56:04 < gxti> Erlkoenig: yep 2013-02-04T17:56:25 < gxti> npn works too but nfet is less parts. you're just making your own open-drain driver with a 60V rating (or whatever) instead of 5V 2013-02-04T17:56:30 < zyp> dirty_d, status stage 2013-02-04T17:56:50 < Erlkoenig> hm the pullup is probably to weak for fast switching, it's only there as a safety measure for when the F4 is not connected 2013-02-04T17:56:50 < karlp> One common pitfall is to neglect ground wirings and nonchalantly proceed with 2013-02-04T17:56:51 < karlp> connecting the meter to external hardware, such as a personal computer. The results 2013-02-04T17:56:52 < dirty_d> zyp, has nothing to do with cdc_acm right? 2013-02-04T17:56:54 < karlp> can be destructive. 2013-02-04T17:56:57 < dirty_d> just hte enumeration process? 2013-02-04T17:56:58 < zyp> dirty_d, it has 2013-02-04T17:56:58 < karlp> this old appnote is pretty casual 2013-02-04T17:57:08 < zyp> it has nothing to do with enumeration 2013-02-04T17:57:32 < dirty_d> i thought ep0 was only used for enumeration and such 2013-02-04T17:57:41 < zyp> haha, no 2013-02-04T17:58:13 < dirty_d> hmm 2013-02-04T17:58:19 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-04T17:58:21 < zyp> http://bin.jvnv.net/f/gffMk.png <- see this? 2013-02-04T17:58:36 < zyp> both set control line state and set line coding are control requests 2013-02-04T17:59:04 < zyp> they are respectively bRequest 0x22 and 0x20 2013-02-04T17:59:19 < zyp> see how I handle those in handle_setup? 2013-02-04T17:59:20 < dirty_d> ahh, i was getting that 2013-02-04T17:59:44 < zyp> I just ignore and ack both 2013-02-04T18:00:18 < dirty_d> RTS, and DTR RTS, arnt usb standard setup packets are they? 2013-02-04T18:00:22 < gxti> Erlkoenig: how fast? 2013-02-04T18:00:24 < dirty_d> i thought that was specific to the serial stuff 2013-02-04T18:00:28 < zyp> of course not, it's part of the acm spec 2013-02-04T18:00:31 < Erlkoenig> gxti: 2 kHz 2013-02-04T18:00:41 < zyp> anyway 2013-02-04T18:00:43 < dirty_d> zyp, then why isnt it send on the ep2 control ep? 2013-02-04T18:00:55 < Erlkoenig> gxti: there's a small fet 2013-02-04T18:01:02 < gxti> Erlkoenig: i think i ended up with a 1k pullup for 100khz switching in a 12v buck converter 2013-02-04T18:01:06 < zyp> because ep2 is an IN endpoint, host can't send on that. 2013-02-04T18:01:07 < gxti> maybe 100 ohms 2013-02-04T18:01:17 < gxti> you'll have to hook up a scope and see what works 2013-02-04T18:01:18 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] 2013-02-04T18:01:25 < dirty_d> zyp, this is strange 2013-02-04T18:01:26 < gxti> but 1khz is much easier 2013-02-04T18:01:29 < zyp> and there is no point wasting an endpoint on something that can be done by a control request 2013-02-04T18:01:46 < dirty_d> hmm 2013-02-04T18:02:03 < dirty_d> zyp, so lots of other device classes use custom setup packets on ep0? 2013-02-04T18:02:13 < zyp> anyway, the thing is that set line coding has a data stage, because the baudrate and stuff doesn't fit in the setup packet 2013-02-04T18:02:23 < dirty_d> ahh 2013-02-04T18:02:38 < zyp> so that's why handle_out() gets called, and after that I write a ZLP for the status stage 2013-02-04T18:02:50 < zyp> and yes, it's very common 2013-02-04T18:03:16 < zyp> lots of classes add their own control requests 2013-02-04T18:03:52 < zyp> http://cgit.jvnv.net/laks_demo/tree/main.cpp?h=usb_audio <- see the volume control handling here, for instance 2013-02-04T18:04:26 < zyp> http://bin.jvnv.net/f/sTrVv.png 2013-02-04T18:04:46 < dirty_d> i see 2013-02-04T18:05:28 < dirty_d> is there a bit set in the bmRequestType to tell that it is not a usb standard setup packet? 2013-02-04T18:05:56 < zyp> yes 2013-02-04T18:06:30 < dirty_d> ahh i see here, http://www.beyondlogic.org/usbnutshell/usb6.shtml 2013-02-04T18:06:33 < zyp> yep 2013-02-04T18:06:43 < dirty_d> standard, class, vendor 2013-02-04T18:06:47 < zyp> correct 2013-02-04T18:07:16 < dirty_d> so these fall under class commands right? 2013-02-04T18:07:24 < zyp> http://cgit.jvnv.net/suzumebachi/tree/main.cpp <- here I'm using my own vendor specific control requests 2013-02-04T18:07:43 < dirty_d> and if someone made some wierd custom cdc_acm driver they might add custom vendor setup packets? 2013-02-04T18:07:58 < dirty_d> ok 2013-02-04T18:08:01 < dirty_d> that all makes sense now 2013-02-04T18:09:24 < dirty_d> would ignoring those RTS, DTR RTS requests cause a write on the TX ep to not get an ACK? 2013-02-04T18:10:03 < zyp> impossible to answer without knowing the host driver 2013-02-04T18:10:19 < dirty_d> it must be why 2013-02-04T18:10:35 < dirty_d> so do you actually have to obey the baudrate? 2013-02-04T18:10:47 < zyp> of course not, it's only metadata 2013-02-04T18:10:50 < dirty_d> ok good 2013-02-04T18:11:21 < Erlkoenig> gxti: http://games.2g2s.de/zeug/f4-protection.png this is the setup, the dotted line is not always connected. in this case the pullup protects the IC or FET on the right of an undefined state. When the dotted line is present and the F4's pin is configured as an output, the diode should have no influence at all, and the output can drive the IC/FET input directly. when the F4's pin is not an output, the diode should protect the pin 2013-02-04T18:12:31 < gxti> Erlkoenig: right, and in that scenario the pin will indeed be protected 2013-02-04T18:12:59 < gxti> Erlkoenig: but the fact that you're pulling it to 12V suggests the IC or FET expects a 12V drive 2013-02-04T18:13:03 < dirty_d> Erlkoenig, why 3.9v though? 2013-02-04T18:13:11 < dirty_d> gxti, maybe its just output 2013-02-04T18:13:21 < gxti> Erlkoenig: as long as you're ok with that being at 0v or 3v whenever the F4 is connected, then go for it 2013-02-04T18:13:25 < Erlkoenig> luckily the IC/FET works with 3V... i use 12V because that's the only voltage available on the board on the right 2013-02-04T18:13:35 < gxti> Erlkoenig: ok then, carry on 2013-02-04T18:13:47 < Erlkoenig> dirty_d: because that's between 3V and 5V 2013-02-04T18:13:53 < Erlkoenig> gxti: okay thanks :) 2013-02-04T18:13:56 < dirty_d> oh 2013-02-04T18:14:39 < Erlkoenig> if i had 5V or 3V on the board on the right, i would have connected the pullup with that and wouldn't need to protect the F4's pin 2013-02-04T18:14:41 < gxti> just wanted to make sure it wasn't a 12v high-side switch 2013-02-04T18:14:50 < gxti> because that don't work too good 2013-02-04T18:15:21 < Erlkoenig> yeah of course ;) 2013-02-04T18:16:31 -!- Claude is now known as claude 2013-02-04T18:36:06 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer] 2013-02-04T18:41:24 < Laurenceb> http://pictures.mastermarf.com/blog/2010/100222-allah.jpg 2013-02-04T18:44:19 -!- barthess [~barthess@185.6.24.193] has joined ##stm32 2013-02-04T18:46:49 < dirty_d> arm-none-eabi-size main.elf is giving me 28584, thats a lot isnt it? 2013-02-04T18:46:54 < dirty_d> for .bss 2013-02-04T18:47:50 < emeb> depends on what your doing 2013-02-04T18:48:15 < dirty_d> that would be from global variables right? 2013-02-04T18:48:23 < dirty_d> or static 2013-02-04T18:49:17 < dirty_d> as far as i know all my global variables should add up to a couple hundred bytes 2013-02-04T18:49:32 < dirty_d> wait no 2013-02-04T18:49:35 < dirty_d> .data is 120 2013-02-04T18:49:38 < dirty_d> which seems right 2013-02-04T18:55:15 < dirty_d> ohhhhh, duh, it was zyp's logger, now its down to 500 bytes 2013-02-04T18:58:04 < dirty_d> wow, this really isnt bad so far, only 3.3k for all the sub stuff 2013-02-04T18:58:07 < dirty_d> USB 2013-02-04T18:59:56 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-04T19:06:37 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-04T19:09:37 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-04T19:12:32 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 240 seconds] 2013-02-04T19:13:56 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has quit [Ping timeout: 248 seconds] 2013-02-04T19:23:47 -!- claude is now known as Claude 2013-02-04T19:34:07 < Laurenceb> https://raw.github.com/Laurenceb/Chibi-Spectro/master/tests/stepper/log39-crop.png 2013-02-04T19:34:22 < Laurenceb> linear actuator is working relatively well now 2013-02-04T19:34:33 < Laurenceb> ~0.1PSI max error 2013-02-04T19:35:37 < emeb> what was the sol'n? 2013-02-04T19:38:19 <+Steffanx> is that thing powerful enough to crush an arm Laurenceb ? 2013-02-04T19:38:52 < Laurenceb> yes lolz 2013-02-04T19:39:05 < Laurenceb> emeb: main issue seems to be stepper control 2013-02-04T19:39:26 < emeb> the pulse width thing? 2013-02-04T19:39:32 < Laurenceb> yes 2013-02-04T19:39:39 < emeb> cool 2013-02-04T19:39:45 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Stepper.c#L38 2013-02-04T19:39:58 < Laurenceb> i added those two functions to clean up the rounding behaviour 2013-02-04T19:40:18 < Laurenceb> with 400hz control loop, off by one errors were significant 2013-02-04T19:40:28 < emeb> wow 2013-02-04T19:40:43 < Laurenceb> wut? 2013-02-04T19:41:11 < emeb> so sensitive to obi-wan errors 2013-02-04T19:41:17 < Laurenceb> yes 2013-02-04T19:41:22 < Laurenceb> 400steps/sec is a lot 2013-02-04T19:41:29 < Laurenceb> i think ive still screwed it up 2013-02-04T19:41:35 <+Steffanx> lol 2013-02-04T19:41:40 < Laurenceb> i think the position estimate is now off by one time period 2013-02-04T19:41:48 < Laurenceb> but ill fix that tomorrow... 2013-02-04T19:42:08 < emeb> so you serious - this thing is strong enough to cause injury? 2013-02-04T19:42:24 < Laurenceb> i had to take position estimation out of the isr to avoid it picking up the rounding 2013-02-04T19:42:34 < Laurenceb> yes, i havent measured max thrust 2013-02-04T19:42:51 < Laurenceb> when i had a firmware issue that maxed it out, it blew a bearing unit 2013-02-04T19:42:56 < emeb> so do you have to weave some fail-safe into it at some point? 2013-02-04T19:43:00 < Laurenceb> maybe 30Kgf 2013-02-04T19:43:26 < Laurenceb> its called a loose bolt on the hight adjustment 2013-02-04T19:43:30 < Laurenceb> so it slides up 2013-02-04T19:43:41 < Laurenceb> luckly no more bugs that bad so far 2013-02-04T19:43:41 < emeb> *shudder* 2013-02-04T19:44:04 < Laurenceb> anyways.. home time 2013-02-04T19:44:06 < Laurenceb> cya 2013-02-04T19:44:12 < emeb> l8r 2013-02-04T19:56:06 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Remote host closed the connection] 2013-02-04T19:58:59 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Read error: Connection reset by peer] 2013-02-04T20:12:02 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-76.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-04T20:12:15 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-04T20:28:26 < gxti> heh, so far seeedstudio disclaimed responsibility for selling/leaking my email address, and the company whose servers sent the spam said they "resolved the issue". still waiting on indiegogo. 2013-02-04T20:40:30 < emeb> surprised that you got that much 2013-02-04T20:41:04 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-76.pptp.stw-bonn.de] has joined ##stm32 2013-02-04T20:41:54 < gxti> i only bothered because it was so clear who was responsible, i don't even use the random email thing that often 2013-02-04T20:42:19 < gxti> if i sign up with a random email address and it gets spam it's pretty damn obvious who was responsible :p 2013-02-04T20:43:15 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 260 seconds] 2013-02-04T20:56:07 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-04T20:56:20 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has left ##stm32 [] 2013-02-04T20:57:29 -!- barthess [~barthess@185.6.24.193] has quit [Quit: Leaving.] 2013-02-04T21:13:16 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-04T21:21:33 < emeb> latest news on STM32F373 fail: 2013-02-04T21:21:47 < emeb> Our Division has confirmed the incorrect bootloader version issue with the STM32F373CCT6 parts that you have. This incorrect bootloader version does not support USB DFU. 2013-02-04T21:21:54 < emeb> The test programs were updated with correct bootloader version on 14th September 2012,  so all parts tested before this date  are impacted by this issue. 2013-02-04T21:22:13 < gxti> gg no re 2013-02-04T21:23:53 < emeb> They go on to say contact DK to return parts. Heh - not worth the bother for me. 2013-02-04T21:25:10 < ds2> hahahahahaha 2013-02-04T21:25:26 < ds2> did the guy who put the wrong one in get toasted? 2013-02-04T21:25:37 < emeb> we'll never know. 2013-02-04T21:26:08 < TitanMKD> emeb move to stm32f4 ;) 2013-02-04T21:26:12 < ds2> wonder how many were made this way 2013-02-04T21:27:53 < zyp> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/APPLICATION_NOTE/CD00167594.pdf <- oh, rev15 is now public 2013-02-04T21:29:11 < zyp> no mention of parts with wrong bootloader version though 2013-02-04T21:29:35 < zyp> but that's something for the errata sheet anyway 2013-02-04T21:30:38 < emeb> If that. 2013-02-04T21:30:58 * ds2 waits for grey market sources of cheap 373's 2013-02-04T21:31:06 < emeb> I doubt that there will be any general info on the bad lot/date codes. 2013-02-04T21:32:01 < emeb> And since distributors rarely give you visibility of the lot/date status of their stock it'll be pot-luck anytime you buy these. 2013-02-04T21:32:01 < ds2> does this mean the non USB DFU version of the bootloader code is free now? since it is not protected and they were aquired without a eula prohibiting otherwise 2013-02-04T21:32:33 < zyp> huh? 2013-02-04T21:32:37 < emeb> ?? 2013-02-04T21:32:44 < ds2> didn't you read out the bootloader? 2013-02-04T21:32:48 < emeb> anyone can read the system memory 2013-02-04T21:32:51 < zyp> I did, so what? 2013-02-04T21:32:52 < emeb> of any part 2013-02-04T21:32:59 < ds2> oh...thought it was locked? 2013-02-04T21:33:03 < emeb> nah 2013-02-04T21:33:17 < ds2> oh 2013-02-04T21:33:19 < zyp> I read both F303 and F373 ones, so we could compare them 2013-02-04T21:33:50 -!- dekar [~dekar@212.255.33.103] has quit [Quit: This computer has gone to sleep] 2013-02-04T21:34:00 < zyp> so if you want to reverse engineer it, you can 2013-02-04T21:34:11 < ds2> hmmm 2013-02-04T21:34:19 < emeb> dongs ran it through a disassembler. 2013-02-04T21:34:29 < zyp> decompilator actually 2013-02-04T21:34:45 < zyp> it's a bit more thorough than a disassembler 2013-02-04T21:34:51 < ds2> is that true of all the stm32s? 2013-02-04T21:34:52 < emeb> ya - would have gotten further except his register map was wrong for the F3xx 2013-02-04T21:35:06 < emeb> ds2: don't see why not. 2013-02-04T21:35:11 < ds2> Hmmmm 2013-02-04T21:35:12 < zyp> ds2, hard to tell without testing with all. 2013-02-04T21:35:13 < emeb> haven't tried tho 2013-02-04T21:35:44 < ds2> before doing that...anyone got a short quick piece of code to send data in and out of the UART? 2013-02-04T21:36:02 < emeb> of which part? 2013-02-04T21:36:05 < ds2> I cannot figure out why all I can get out of it is a corrupt 9600baud. anything else is totally broken 2013-02-04T21:36:08 < ds2> stm32f103 2013-02-04T21:36:12 < zyp> http://cgit.jvnv.net/laks/tree/usart/usart.h 2013-02-04T21:36:40 < ds2> i mean something that compiles and runs... want to dump the registers the working code is using and the registers I am configuring 2013-02-04T21:37:18 < ds2> the setup appears to be same as the reference code and the fragments people are posting on line 2013-02-04T21:37:36 < emeb> maybe BMP serial port code? 2013-02-04T21:37:45 < zyp> uh 2013-02-04T21:37:54 < zyp> how do you manage to get it wrong? 2013-02-04T21:38:06 < ds2> zyp: that is why I am so puzzled. it is a uart 2013-02-04T21:38:16 < ds2> emeb: hmmmm does that output text or binary? 2013-02-04T21:38:33 -!- daku is now known as DaKu 2013-02-04T21:39:21 < emeb> ds2: the BMP has a 2nd USB ACM port that acts like a generic USB/Serial port - tx/rx anything you want. 2013-02-04T21:39:28 < ds2> emeb: what's the name to search for? bmp programmer stm32 isn't turning up link for it 2013-02-04T21:39:33 < zyp> looking at the code I posted myself, I don't get how you manage to fuck it up 2013-02-04T21:39:44 < ds2> emeb: no... I want to drive the UART, not the USB 2013-02-04T21:39:49 < emeb> ds2: https://github.com/gsmcmullin/blackmagic 2013-02-04T21:40:16 < zyp> emeb, probably better to look at libopencm3 examples then 2013-02-04T21:40:21 < emeb> right 2013-02-04T21:40:43 < emeb> I've got uart running on F373 w/o problems using StdPeriph. 2013-02-04T21:40:49 < zyp> ds2, are you fucking up the fractional baudrate stuff? 2013-02-04T21:41:21 < ds2> zyp: I don't think so. I get the same figure for it if I do the math by hand. it is odd that it works only with 9600. scaling it up or down doesn't work 2013-02-04T21:41:34 < zyp> can I have a look at your code? 2013-02-04T21:41:40 < ds2> sure 2013-02-04T21:42:13 < ds2> zyp: http://www.hy-research.com/tmp/blink.c 2013-02-04T21:42:24 < zyp> 404 2013-02-04T21:42:25 < ds2> SetUart() is where I configure it. PutChar is the send 2013-02-04T21:42:43 < ds2> my bad 2013-02-04T21:42:46 < ds2> zyp: http://www.hy-research.com/tmp/blinky.c 2013-02-04T21:43:25 < ds2> and I have tried it with HSE, HSI 2013-02-04T21:43:29 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has joined ##stm32 2013-02-04T21:44:34 < zyp> uh 2013-02-04T21:44:40 < zyp> what's up with the #else part? 2013-02-04T21:44:42 < ds2> also tried adding 1 and subtracting 1 with a debugger. 2013-02-04T21:44:56 < ds2> zyp: that is an experiment. it was from some other code. neither case works 2013-02-04T21:45:02 < ds2> and results in the same value in BRR 2013-02-04T21:45:15 < ds2> also tried it w/o the +1 2013-02-04T21:45:33 < zyp> BRR = busclock / baudrate; works fine for me 2013-02-04T21:45:48 < zyp> without any +1 2013-02-04T21:46:03 < ds2> tried both since one page mentioned somehting about round off 2013-02-04T21:46:38 < zyp> there is no rounding at 115200baud 2013-02-04T21:46:55 < ds2> that's what I tried at first 2013-02-04T21:46:57 < zyp> BRR = 625 2013-02-04T21:47:10 < ds2> this mess of code is a snapshot in the experiments 2013-02-04T21:47:56 < ds2> what sysclock for BRR=625 and 115200? 2013-02-04T21:48:51 < zyp> 72MHz, and no divider on ABP2, so APB2 clock is also 72MHz 2013-02-04T21:49:14 < zyp> you are using USART1 which runs from APB2 2013-02-04T21:49:22 < ds2> yes 2013-02-04T21:49:27 < zyp> are you sure your bus dividors are correct? 2013-02-04T21:49:46 < ds2> not completely 2013-02-04T21:49:59 < emeb> funny how often problems can be traced back to bad clock setup 2013-02-04T21:50:03 < zyp> do you have the ability to read out registers now? 2013-02-04T21:50:32 < ds2> yes but board is not in front of me 2013-02-04T21:50:34 < zyp> if so, can you read out RCC_CFGR for me? 2013-02-04T21:50:53 < zyp> or just find out what you set RCC_CFGR to 2013-02-04T21:50:53 < ds2> look in SetClock() 2013-02-04T21:51:01 < zyp> ah, thanks 2013-02-04T21:51:02 < ds2> that register is set there 2013-02-04T21:51:13 < ds2> this code came from the StdLib stuff 2013-02-04T21:51:48 < zyp> I see 2013-02-04T21:51:54 < zyp> what is the value of RCC_CFGR_PPRE2_DIV1 ? 2013-02-04T21:52:27 < ds2> https://github.com/gsmcmullin/blackmagic 2013-02-04T21:52:29 < ds2> blah 2013-02-04T21:52:37 < ds2> stm32f10x.h:#define RCC_CFGR_PPRE2_DIV1 ((uint32_t)0x00000000) 2013-02-04T21:53:10 < zyp> ah, right 2013-02-04T21:53:18 < zyp> what about RCC_CFGR_PLLMULL4? 2013-02-04T21:53:39 < ds2> stm32f10x.h:#define RCC_CFGR_PLLMULL4 ((uint32_t)0x00080000) 2013-02-04T21:54:06 < zyp> and you are running with a 12MHz crystal? 2013-02-04T21:54:26 < ds2> yes 2013-02-04T21:54:54 < zyp> so, are you aware that you are running with SYSCLK = 48MHz? 2013-02-04T21:55:18 < zyp> 12*4 2013-02-04T21:55:23 < ds2> apparently not 2013-02-04T21:55:30 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-04T21:55:45 < zyp> you should probably use PLLMULL6 or something there 2013-02-04T21:56:11 < ds2> ok 2013-02-04T21:56:15 < zyp> because the PLLMULL4 define is defined to the value 2 2013-02-04T21:56:28 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-04T21:56:33 < zyp> which means that PLL gives a 4x factor 2013-02-04T21:56:56 < ds2> oh I see why I put a 4 in there 2013-02-04T21:57:09 < dirty_d> its multiplying half the input clock right? 2013-02-04T21:57:11 < ds2> wasn't sure if I F'ed something so I purposely slowed it down. 2013-02-04T21:57:18 < zyp> dirty_d, no 2013-02-04T21:57:24 < dirty_d> zyp, on mine it is 2013-02-04T21:57:30 < dirty_d> well, if its the internal one 2013-02-04T21:57:39 < zyp> HSI is divided by two, HSE is not 2013-02-04T21:57:45 < dirty_d> yea 2013-02-04T21:58:03 < ds2> Hmmm ok. let me change back and try that tonight 2013-02-04T21:58:48 < dirty_d> i need to come up with a project for this mill 2013-02-04T21:59:00 < dirty_d> i finally "finished" it saturday 2013-02-04T21:59:06 < ds2> is it possible to setup the pinmux wrong and have it break on higher baudrates? 2013-02-04T21:59:11 < ds2> dirty_d: mill? 2013-02-04T21:59:18 < dirty_d> ds2, i made a cnc mill 2013-02-04T21:59:28 < ds2> dirty_d: nice. what kind of spindle? 2013-02-04T21:59:44 < dirty_d> ds2, i used an existing manal machine 2013-02-04T21:59:50 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-04T21:59:51 < dirty_d> its only 1/2HP and 2000rpm 2013-02-04T22:00:04 < dirty_d> 18x7x13" travel 2013-02-04T22:00:09 < zyp> ds2, sounds unlikely 2013-02-04T22:00:16 < dirty_d> backlash is 0.0012" on all axes 2013-02-04T22:00:19 < dirty_d> came out pretty good 2013-02-04T22:00:28 < ds2> zyp: ok. I will focus on the clock setup. 2013-02-04T22:00:29 < dirty_d> i loaded some oversized balls in the ballnuts 2013-02-04T22:00:58 < ds2> dirty_d: what did you start with? one of those Seig's? 2013-02-04T22:01:07 < dirty_d> no, a grizzly 2013-02-04T22:01:17 < zyp> maybe if you fuck up and set the output to open drain, and the signal happen to be weakly pulled up 2013-02-04T22:01:29 < dirty_d> ds2, this http://www.grizzly.com/products/Drill-Mill-with-Stand/G0704 2013-02-04T22:01:30 < zyp> then it might work at slow speeds 2013-02-04T22:02:05 < ds2> I see. 2013-02-04T22:02:29 < zyp> but that sounds unlikely 2013-02-04T22:02:40 < zyp> you'd be able to tell with a scope 2013-02-04T22:03:27 < ds2> still trying to figure out the way the pinmux is on this chip... if I want to use a pin as non GPIO, what do I set it up as? I have seen code that sets it up as explicit input, explicit output...it doesn't seem to be a pattern 2013-02-04T22:03:38 < ds2> reason I asked that is I saw something funny with the scope 2013-02-04T22:03:41 < zyp> oh, wait 2013-02-04T22:03:49 < zyp> this is exactly what you do 2013-02-04T22:03:54 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 276 seconds] 2013-02-04T22:04:16 < ds2> ? 2013-02-04T22:04:37 < dirty_d> http://news.sky.com/story/1047004/mini-drones-army-deploys-tiny-helicopters 2013-02-04T22:04:50 < zyp> CRL/CRH defaults to 0x44444444, and you write |= 0xb into the field for the TX pin 2013-02-04T22:05:02 < zyp> 0xb | 0x4 = 0xf, and 0xf is AF open drain 2013-02-04T22:05:27 < ds2> doh 2013-02-04T22:06:34 < ds2> thanks. 2013-02-04T22:06:57 < emeb> zypdebugger FTW 2013-02-04T22:07:24 < ds2> dirty_d: nice.. R8! think that is a seig 2013-02-04T22:07:43 < ds2> been hacking at the damn uart for a while 2013-02-04T22:08:26 < dirty_d> ds2, nah this is someting else, seig makes the x3, this is a BF20 clone 2013-02-04T22:08:39 < dirty_d> ds2, wait till you get to USB 2013-02-04T22:08:45 < ds2> dirty_d: hmmm 2013-02-04T22:09:06 < ds2> dirty_d: USB looks to be F'ed on my board anyways so... 2013-02-04T22:09:13 < dirty_d> which board? 2013-02-04T22:09:23 < ds2> but I am really just not too worried about USB 2013-02-04T22:09:33 < ds2> dirty_d: I made a custom F103 board 2013-02-04T22:09:40 < dirty_d> ahh 2013-02-04T22:09:47 < dirty_d> whats wrong with the usb? 2013-02-04T22:09:53 < ds2> dirty_d: it'll be running uCLinux so the stack is mostly there 2013-02-04T22:10:17 < ds2> dirty_d: it lacks ESD protection and vbus sensing 2013-02-04T22:10:19 < dirty_d> the only reason im bothering with usb is because its just more convinient to use than serial, and a lot of pcs dont even have a serial port 2013-02-04T22:10:37 < dirty_d> ahh 2013-02-04T22:11:25 < zyp> you don't really need vbus sensing 2013-02-04T22:12:15 < ds2> is the USB block always looking for a reset condition? 2013-02-04T22:12:29 < ds2> it seemed like it needs something to wake up 2013-02-04T22:12:32 < zyp> yes, the usb block is not even vbus aware 2013-02-04T22:12:43 < ds2> oh 2013-02-04T22:12:47 < ds2> maybe it is not F'ed 2013-02-04T22:12:58 < zyp> probably not 2013-02-04T22:13:47 -!- dekar [~dekar@drms-590cef39.pool.mediaWays.net] has joined ##stm32 2013-02-04T22:13:50 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-04T22:15:47 < dirty_d> im looking at the f3disco schematic, i dont see what pin vbus is connected to 2013-02-04T22:16:26 < zyp> none. 2013-02-04T22:16:45 < zyp> (IIRC) 2013-02-04T22:16:58 < zyp> F3 is using same usb core as F1 2013-02-04T22:16:59 < dirty_d> i thought i saw something in your code 2013-02-04T22:17:13 < ds2> hmmm ok 2013-02-04T22:17:35 < zyp> dirty_d, the usb core used in F4 has vbus sensing 2013-02-04T22:17:42 < dirty_d> oh, that was it 2013-02-04T22:18:35 < zyp> but even so, it's still possible to configure the controller to ignore it 2013-02-04T22:24:24 -!- jef79m_lurking is now known as jef79m 2013-02-04T22:34:52 -!- dekar_ [~dekar@drms-4d014de6.pool.mediaWays.net] has joined ##stm32 2013-02-04T22:34:54 -!- mode/##stm32 [+v dekar_] by ChanServ 2013-02-04T22:36:48 -!- dekar [~dekar@drms-590cef39.pool.mediaWays.net] has quit [Ping timeout: 264 seconds] 2013-02-04T22:54:24 -!- jef79m is now known as jef79m_lurking 2013-02-04T22:55:37 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-04T23:05:06 < Laurenceb_> http://t2.gstatic.com/images?q=tbn:ANd9GcR3PvBIaSI8MLTpuktFemAGp5kjXt8hRRpIZh72J5lpCM--58ZHSQ 2013-02-04T23:06:29 < emeb> wonder if it would be possible to get a personalized license plate that said GOATSE 2013-02-04T23:08:03 < Laurenceb_> obligatory 2013-02-04T23:08:04 < Laurenceb_> http://www.dailyfailcenter.com/sites/default/files/fail/d77a3c7bc23f.jpg 2013-02-04T23:08:25 < Erlkoenig> SuperBowl 2013-02-04T23:09:15 < emeb> ark ark 2013-02-04T23:10:36 < Laurenceb_> http://www.dailyfailcenter.com/sites/default/files/fail/90b4ea3b78ef.jpg 2013-02-04T23:10:51 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] 2013-02-04T23:11:15 < dirty_d> hmm, i should just use libusb instead of serial over usb 2013-02-04T23:12:14 < Erlkoenig> libusb doesn't support asynchronous operation on windows :/ 2013-02-04T23:12:52 < dirty_d> i figured the windows support sucked 2013-02-04T23:13:00 < Erlkoenig> jup 2013-02-04T23:13:23 < Erlkoenig> also the whole thing has no nice integration in GUI Toolkits' mainloops 2013-02-04T23:15:15 < dirty_d> why because its blocking? 2013-02-04T23:15:48 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-04T23:16:10 < Erlkoenig> no the asynchronous API can not generate callbacks 2013-02-04T23:16:23 < Erlkoenig> and blocking is always bad when having a GUI 2013-02-04T23:18:51 < Laurenceb_> so have another thread? 2013-02-04T23:19:22 < dirty_d> yea 2013-02-04T23:19:54 -!- jef79m_lurking is now known as jef79m 2013-02-04T23:20:16 < zyp> if you are doing vendor specific stuff and care enough about windows support you might want to use winusb directly 2013-02-04T23:20:37 < Laurenceb_> never use winjungo 2013-02-04T23:20:39 < Laurenceb_> ewwww 2013-02-04T23:22:06 < Laurenceb_> someone used that on a work project and i had to fix their fail :-/ 2013-02-04T23:22:19 < R2COM> so.. looks to me that those Psocs have PLDs embedded on same die with microcontroller 2013-02-04T23:22:27 < R2COM> interesting... 2013-02-04T23:22:43 < zyp> I used windows-libusb on a work project, worked nicely 2013-02-04T23:22:58 < R2COM> so in that case one probably could do his thing with one die, without bringing in cpld on board 2013-02-04T23:23:14 < zyp> I think the usb backend ran in a seperate thread with in and out queues 2013-02-04T23:23:15 < R2COM> but it needs check of course, hard to tell 2013-02-04T23:23:40 < dirty_d> zyp, have you tested the throughput on your serial driver? 2013-02-04T23:23:50 < zyp> or maybe the data processing also run in that thread 2013-02-04T23:24:26 -!- jef79m is now known as jef79m_lurking 2013-02-04T23:24:46 < zyp> dirty_d, no, and I suspect that such attempts may break it :p 2013-02-04T23:25:03 < dirty_d> zyp, what do you figure the f3 can do? 2013-02-04T23:25:18 < dirty_d> max theoretical is 1.5MB/s right? 2013-02-04T23:25:26 < dirty_d> so maybe 0.75MB? 2013-02-04T23:26:01 < zyp> hmm, let's see 2013-02-04T23:26:18 < Laurenceb_> id imagine maybe 800 to 900K 2013-02-04T23:26:56 < zyp> max theoretical is 1216000 bytes per second on HS 2013-02-04T23:26:57 < Laurenceb_> for e.g. I get 825KB/s with mass storage 2013-02-04T23:27:08 < Laurenceb_> and zyp claims over 900KB/s 2013-02-04T23:27:29 < Laurenceb_> but that wasnt sustained for a huge file :P 2013-02-04T23:27:55 < zyp> did I claim that? the sd card I was testing with is slow as fuck 2013-02-04T23:28:19 < Laurenceb_> interesting 2013-02-04T23:28:26 < Laurenceb_> i was using ancient cards from 2002 2013-02-04T23:28:40 < dirty_d> thats fast enough for anything id want to do 2013-02-04T23:28:43 < zyp> and my msc code is not particularly efficient either 2013-02-04T23:28:44 < Laurenceb_> that id pulled out of old kit and sanded the contacts clean on.. 2013-02-04T23:29:11 < Laurenceb_> i should benchmark on decent cards 2013-02-04T23:29:16 < zyp> yeah, card surely gets faster when you sand the contacts :D 2013-02-04T23:29:39 < Laurenceb_> well 2013-02-04T23:29:42 < Laurenceb_> its works 2013-02-04T23:29:57 < Laurenceb_> as opposed to unsanded where theres too much corrosion 2013-02-04T23:30:07 < zyp> I understood that, I was just joking 2013-02-04T23:30:12 < Laurenceb_> cards had been outside for 10 years 2013-02-04T23:30:14 < dirty_d> it would appear that it's time to gtfo 2013-02-04T23:30:18 < dirty_d> later 2013-02-04T23:30:19 < zyp> http://bin.jvnv.net/f/e16v1.png 2013-02-04T23:30:29 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-04T23:30:34 < zyp> that's write operations 2013-02-04T23:30:52 < zyp> the gaps in CK is when usb is doing it's stuff 2013-02-04T23:31:12 < zyp> and the activity on Dx is when block is being transferred to card 2013-02-04T23:31:32 < Laurenceb_> is usb and sd interleaved? 2013-02-04T23:31:34 < zyp> and D0 low is when card is busy 2013-02-04T23:32:00 < zyp> yes, the code is still dumb and simple, and only do one thing at a time 2013-02-04T23:32:15 < Laurenceb_> oh 2013-02-04T23:32:19 < Laurenceb_> so not interleaved? 2013-02-04T23:32:29 < zyp> huh? 2013-02-04T23:32:36 < Laurenceb_> sd and usb transfers dont overlap? 2013-02-04T23:32:55 < zyp> correct 2013-02-04T23:33:23 < Laurenceb_> i see 2013-02-04T23:33:42 < Laurenceb_> i modified the ST code to use fatfs to access the card and interleave the transfers 2013-02-04T23:33:59 < Laurenceb_> interestingly its much faster if theres no hub inbetween 2013-02-04T23:34:32 -!- jef79m_lurking is now known as jef79m 2013-02-04T23:34:34 < zyp> I have more important issues to solve first, I have no error handling :p 2013-02-04T23:37:04 < gxti> that's fine as long as you don't create any errors, which seems like a good idea 2013-02-04T23:37:30 < zyp> the sd card tends to misbehave from time to time, which messes up my assumptions 2013-02-04T23:37:48 < zyp> especially the sdhc card I bought later 2013-02-04T23:38:41 < zyp> Laurenceb_, that sounds more like a bad hub 2013-02-04T23:38:48 < gxti> are there any f4 dev boards with usb and SD? 2013-02-04T23:38:55 < zyp> sure 2013-02-04T23:38:59 < gxti> or f2 i guess 2013-02-04T23:39:14 < zyp> http://bin.jvnv.net/f/vV9Bt.JPG <- even has highspeed usb 2013-02-04T23:39:15 < karlp> the netduino plus? ;) 2013-02-04T23:39:32 < gxti> refusing to buy on grounds of having 'duino' in name 2013-02-04T23:39:33 < Laurenceb_> zyp: well the same hub causes BSOD on winxp running winjungo 2013-02-04T23:39:49 < TitanMKD> gxti my STM32F4 debug have 2 USB port ;) 2013-02-04T23:39:50 < zyp> sounds unrelated 2013-02-04T23:39:51 < karlp> sorry, plus 2, 2013-02-04T23:39:57 < TitanMKD> gxti but SD is optional 2013-02-04T23:40:10 < gxti> TitanMKD: impressive 2013-02-04T23:40:17 < zyp> gxti, I can recommend the waveshare boards 2013-02-04T23:40:17 < karlp> gxti: well, you asked :) http://www.amazon.com/gp/product/B009QOYK2U?m=A15S1YSPRYIXOQ 2013-02-04T23:40:30 < karlp> sd, usb, ethernet, stm32f4 2013-02-04T23:40:53 < Laurenceb_> crappy ethernet tho 2013-02-04T23:41:03 < karlp> but shield compatible! 2013-02-04T23:41:08 < karlp> and you can use dot net micro! 2013-02-04T23:41:13 < gxti> $60 isn't worth it, i'd rather just design a crappy prototype of the thing i'm making anyway 2013-02-04T23:41:30 < zyp> the nice part about the waveshare boards is that none of the signals are hardwired to anything, so you can prototype whatever 2013-02-04T23:42:53 < zyp> and with a nice selection of accessory boards, it's easy to hook up other stuff as required as well 2013-02-04T23:43:13 < zyp> so I'd say it's well worth the price 2013-02-04T23:43:40 < gxti> better than the netduino in any case 2013-02-04T23:43:44 < zyp> yes. 2013-02-04T23:43:52 -!- Claude is now known as claude 2013-02-04T23:44:38 < zyp> the discovery boards are also nice, but then you have to keep in mind that some pins are used for stuff on the board, and can therefore not be used freely 2013-02-04T23:45:07 < Erlkoenig> annoying as hell... 2013-02-04T23:46:14 < zyp> but then they are intended to be cheap boards to demonstrate the strengths of the chip, not really tools for prototyping everything :p 2013-02-04T23:46:43 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-04T23:46:46 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Read error: Connection reset by peer] 2013-02-04T23:47:00 < gxti> made a decent programmer until i got my bmp 2013-02-04T23:47:10 < gxti> of course i only worked out the kinks after i got the bmp parts 2013-02-04T23:49:24 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-244-68.clienti.tiscali.it] has quit [Ping timeout: 256 seconds] 2013-02-04T23:51:11 < gxti> heh, some of the waveshare boards are just breakouts for the f4disco. cute. 2013-02-04T23:52:41 < zyp> yeah, and thereby suffer from the same problem with used pins ;) 2013-02-04T23:53:32 < gxti> i'll think about the 407z though 2013-02-04T23:53:40 < gxti> pluggable ramz would be nice 2013-02-04T23:54:03 -!- DaKu is now known as daku 2013-02-04T23:54:41 < zyp> I just got the full 407z kit with all the accessories, thinking they might come in handy some day 2013-02-04T23:55:01 < zyp> so far I've used the sd, ulpi and ethernet boards 2013-02-04T23:56:47 < emeb> thinking my next design will use a 407v 2013-02-04T23:57:28 < emeb> cheaper than the 405... 2013-02-04T23:57:40 < zyp> v is 100pin? 2013-02-04T23:57:54 < zyp> ah, it's because you don't want the full 1M flash? 2013-02-04T23:58:19 < emeb> 100 pin will be good - want FSMC 2013-02-04T23:58:32 < emeb> don't really need 1MB flash 2013-02-04T23:58:39 < gxti> i'm not even sure what i'm making yet, which is why a dev board is probably a good idea 2013-02-04T23:58:47 < zyp> «don't need it, don't want to pay for it» 2013-02-04T23:59:06 < emeb> didn't think I'd use ethernet, but 407v has it and gxti pointed out how cheap the phys are... 2013-02-04T23:59:26 < zyp> gxti, yeah, same here, I'm just making lots of toy snippets --- Day changed Tue Feb 05 2013 2013-02-05T00:00:17 < emeb> ditto 2013-02-05T00:00:45 * emeb looks at the shelf full of snippets from the last few years... 2013-02-05T00:02:03 * zyp looks at laks_demo repo: http://bin.jvnv.net/f/Soqzs.png 2013-02-05T00:03:38 < zyp> that's 13 branch heads 2013-02-05T00:03:51 < zyp> some more relevant than others 2013-02-05T00:04:18 < emeb> lots of coding 2013-02-05T00:05:00 < zyp> most of them is of the type «this is a hack, but there's some nice stuff here that I can use for something later» 2013-02-05T00:05:43 < emeb> yup - a lot of the hardware/software projects that I never finish end up as foundations for stuff I actually get paid for. 2013-02-05T00:05:53 < gxti> f2 is sort of the new f4, right? 2013-02-05T00:05:58 < zyp> no 2013-02-05T00:06:01 < gxti> k 2013-02-05T00:06:04 < zyp> f4 is the new f2 2013-02-05T00:06:12 < emeb> f2 = old n busted 2013-02-05T00:06:13 < gxti> o rly 2013-02-05T00:06:16 < gxti> k, never mind 2013-02-05T00:06:47 < emeb> f2 is M3, f4 is M4F 2013-02-05T00:07:22 < gxti> and then f3 is a cut-down f4? 2013-02-05T00:07:24 < zyp> f2 came some months before f4 IIRC, but then when f4 came with M4F and a higher maximum clock, it got completely overshadowed 2013-02-05T00:07:27 < zyp> no 2013-02-05T00:07:31 < gxti> blrubhrlburglh 2013-02-05T00:07:37 < emeb> F3 is M4F too, but slower. More like F1 2013-02-05T00:07:51 < zyp> F3 is an upgraded F1 2013-02-05T00:07:55 < emeb> but with specially extended analog 2013-02-05T00:08:07 < gxti> ok, so i was right but too vague about 'cut down' 2013-02-05T00:08:09 < zyp> got M4F, analog stuff and some other nice improvements 2013-02-05T00:08:19 < zyp> well 2013-02-05T00:08:29 < emeb> F37x has 16-bit ADCs, F30x has fast ADCs 2013-02-05T00:08:37 < gxti> since f1 is not m4 2013-02-05T00:08:38 < emeb> plus op-amps 2013-02-05T00:08:48 < gxti> anyway, i got what i needed which is to stick with f4 2013-02-05T00:08:50 < zyp> F1 with additions is more fitting than F4 with reductions 2013-02-05T00:09:23 < zyp> but yeah, performance wise it sits in between F1 and F4 2013-02-05T00:09:23 < gxti> zyp: i overestimate the importance of using m4 instead of m3, then 2013-02-05T00:09:45 < zyp> I'm thinking in terms of peripherals, not performance 2013-02-05T00:09:45 < emeb> huh? 2013-02-05T00:09:49 < gxti> k 2013-02-05T00:10:14 < emeb> m4 DSP stuff doesn't really bring much to the party. 2013-02-05T00:10:14 < ds2> in practice, does the F4 burn more power then the F2? 2013-02-05T00:10:20 < emeb> but floating pt is nice to have. 2013-02-05T00:10:40 < zyp> F3 would outdo F2 for floating point stuff, but on other stuff F2 might perform better due to the higher clock 2013-02-05T00:11:03 < zyp> not sure how much difference the SIMD stuff would do 2013-02-05T00:11:27 < zyp> but either way, F2 and F3 would each be fastest on different applications 2013-02-05T00:11:29 < emeb> I think the SIMD stuff is more of a marketing bullet. 2013-02-05T00:11:47 < zyp> they would both beat F1, and F4 would then beat both again 2013-02-05T00:12:06 < emeb> *nod* 2013-02-05T00:12:17 < emeb> and LPC43xx beats everything. :) 2013-02-05T00:12:29 < gxti> but yeah the waveshare board is looking like a winner, since they have memory and touchscreens and stuff 2013-02-05T00:12:29 < zyp> not too sure about that 2013-02-05T00:12:37 < emeb> ?? 2013-02-05T00:12:50 < zyp> I suspect F4 would be faster on execution from flash 2013-02-05T00:13:20 < emeb> flash on your 435x is slower? no cache? 2013-02-05T00:13:36 < zyp> remember that the F2/F4 flash is 128-bits wide 2013-02-05T00:13:46 < emeb> true 2013-02-05T00:14:16 < zyp> lpc doesn't appear to have that, so it might get more wait states 2013-02-05T00:14:25 -!- daku is now known as DaKu 2013-02-05T00:14:55 -!- DaKu is now known as daku 2013-02-05T00:15:05 < zyp> hard to say without benchmarking though 2013-02-05T00:15:25 < zyp> and then it might be irrelevant when you put code in ram instead 2013-02-05T00:15:40 < emeb> CCM is nice to have. 2013-02-05T00:15:44 < zyp> lpc43xx got five seperate sram banks :p 2013-02-05T00:15:46 -!- dekar_ [~dekar@drms-4d014de6.pool.mediaWays.net] has quit [Quit: Leaving] 2013-02-05T00:24:29 -!- jef79m is now known as jef79m_lurking 2013-02-05T00:25:08 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-05T00:25:08 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Changing host] 2013-02-05T00:25:08 -!- Thorn__ [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-05T00:26:02 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer] 2013-02-05T00:31:30 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-05T00:36:39 < emeb> zyp: did you get the SDRAM working on your LPC43xx? 2013-02-05T00:37:03 < zyp> haven't tried yet 2013-02-05T00:37:40 < zyp> I read over the documentation on that, then I switched context and started doing something else instead 2013-02-05T00:37:53 < emeb> ooo - shiny! :) 2013-02-05T00:38:20 * emeb is often sidetracked 2013-02-05T00:39:19 < zyp> latest news here is that I pulled out the quadrotor board and started working on that again 2013-02-05T00:39:49 < emeb> what needs doing on that? 2013-02-05T00:40:05 < zyp> want to get something flying again this year when spring comes :p 2013-02-05T00:40:46 < zyp> only show stopper is that I need to reprogram cpld for pwm outputs, it's currently configured for can 2013-02-05T00:41:15 < zyp> and I'm not sure I'll bother digging up xilinx toolchain before I get tired and move on to something else again :p 2013-02-05T00:41:32 < emeb> is it that hard to set up? 2013-02-05T00:42:03 < zyp> no, just a bother :p 2013-02-05T00:42:38 < emeb> I spent most of Saturday designing an AM signal generator in Xilinx to test out the F303 SDR project. 2013-02-05T00:42:54 < emeb> the Sunday I simulated it in Matlab and concluded there was no way to make it work. 2013-02-05T00:42:55 < emeb> :) 2013-02-05T00:43:03 < zyp> :p 2013-02-05T00:43:54 < zyp> well, I just have to dig up the old laptop where I have it installed, edit a few lines of verilog, then figure out again how I got the impact stuff working with my usb stuff 2013-02-05T00:43:59 < emeb> that's what happens when you dive into the low-level stuff before doing the system engineering... 2013-02-05T00:44:16 <+Steffanx> lol 2013-02-05T00:44:39 < zyp> probably impact still remembers the settings I used last time, so that's also fine 2013-02-05T00:44:51 < emeb> I _hate_ impact. 2013-02-05T00:44:58 < qyx_> zyp: few days ago i solved it with xc3sprog 2013-02-05T00:44:59 < emeb> worst pos tool 2013-02-05T00:45:10 < zyp> yeah, it's buggy as well 2013-02-05T00:45:15 < qyx_> but don't know if it works also fir cplds 2013-02-05T00:45:41 < emeb> It's like it's geological layers of cruft dating all the way back to the original 2000-series parts. 2013-02-05T00:45:55 < zyp> some times it connects to my jtag stuff, but doesn't give me the program option 2013-02-05T00:46:43 < zyp> so I have to dick around and disconnect and search for cables, and then connect to my stuff again and if I'm lucky program will be available 2013-02-05T00:46:52 < emeb> :P 2013-02-05T00:46:54 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-05T00:47:01 < emeb> qyx_: what's xc3sprog? 2013-02-05T00:47:22 < emeb> nm http://xc3sprog.sourceforge.net/ 2013-02-05T00:47:30 < emeb> looks like it does cpld 2013-02-05T00:47:31 < qyx_> emeb: fpga programmer 2013-02-05T00:47:43 < qyx_> i used it with generic ft232 cable 2013-02-05T00:48:08 < ds2> zyp: how big of a quadrotor craft do you fly? 2013-02-05T00:48:08 < emeb> I wrote my own bitstream programmer for loading XC3Sxxx parts via a SPI port. 2013-02-05T00:48:54 < zyp> ds2, current build is 80cm in diameter 2013-02-05T00:49:00 < zyp> first one was a bit smaller 2013-02-05T00:49:44 < Laurenceb_> emeb: i still think it will work 2013-02-05T00:49:50 < zyp> of course, arms could be cut shorter if needed 2013-02-05T00:50:00 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-05T00:50:03 < zyp> Laurenceb_, you always do that 2013-02-05T00:50:05 < emeb> Laurenceb_: I'd be interested in seeing how 2013-02-05T00:50:30 < zyp> Laurenceb_, and that's probably part of the reason you are having problems getting stuff to work :p 2013-02-05T00:50:40 < emeb> from what I can tell via simulation, you need to do a fair amount of oversampling. 2013-02-05T00:50:42 < Laurenceb_> longer lookup 2013-02-05T00:50:57 < Laurenceb_> 512 or 1024 samples 2013-02-05T00:50:58 < emeb> so that limits the amount of data you can process in the time available 2013-02-05T00:51:05 < Laurenceb_> true 2013-02-05T00:51:06 < ds2> zyp: nice size. what's the radio? 2013-02-05T00:51:08 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-05T00:51:17 < zyp> currently turnigy 9x 2013-02-05T00:51:33 < emeb> Laurenceb_: the oversampling means that you need to process 4x the number of samples collected by the ADC in a given time. 2013-02-05T00:51:34 < zyp> used some cheaper ebay shit previously 2013-02-05T00:51:44 < Laurenceb_> yes 2013-02-05T00:51:52 < emeb> CPU can't do that for any useful bandwidths. 2013-02-05T00:52:00 < Laurenceb_> but im not sure the alising is that bad 2013-02-05T00:52:05 < Laurenceb_> i think you can trade off 2013-02-05T00:52:08 < ds2> a real radio! 2013-02-05T00:52:23 < Laurenceb_> in fact... 2013-02-05T00:52:33 < Laurenceb_> if you make the lookup a little shorter 2013-02-05T00:52:43 < Laurenceb_> the aliasing will be out of the audio range 2013-02-05T00:52:44 < emeb> From the spectra I simulated, with 1x or 2x oversampling the aliases will be only about 20dB down. 2013-02-05T00:53:01 < Laurenceb_> i think theres an optimal window function and lookup length 2013-02-05T00:53:43 < zyp> hmm, I don't have any cool pics of the current build 2013-02-05T00:54:03 < zyp> http://bin.jvnv.net/f/TlmaY.JPG <- got one of it disassembled though 2013-02-05T00:54:16 < zyp> with old radio and old controller board 2013-02-05T00:54:21 < Laurenceb_> http://upload.wikimedia.org/wikipedia/commons/2/2a/Planck-taper.svg 2013-02-05T00:54:36 < Laurenceb_> maybe that with a slightly shorter lookup 2013-02-05T00:55:07 < Laurenceb_> 256 sample lookup i think 2013-02-05T00:55:10 < emeb> yeah - wider windows might help narrow up 2013-02-05T00:55:20 < emeb> I was using blackman-harris or hann. 2013-02-05T00:55:41 < Laurenceb_> short lookup -> too wide 2013-02-05T00:55:57 < Laurenceb_> long lookup -> aliasing at low frequencies 2013-02-05T00:56:06 < Laurenceb_> theres an optimal length... 2013-02-05T00:56:15 < emeb> I'm simulating 64 samples/period w/ 256 sample integration. 4x OS 2013-02-05T00:56:22 < Laurenceb_> hmm 2013-02-05T00:56:50 < emeb> That gives > 50dB OOB rejection 2013-02-05T00:57:09 < emeb> but the sample rate has to be too low to be useful to fit it all into the available time. 2013-02-05T00:57:12 < Laurenceb_> i get 185KHz offset is 70dB down 2013-02-05T00:57:28 < Laurenceb_> with 512 sample lookup 2013-02-05T00:57:40 < Laurenceb_> and planck-taper 2013-02-05T00:58:04 < Laurenceb_> then the aliasing will be quite minimal 2013-02-05T00:58:11 < emeb> Yeah, but 512 sample LUT takes about 40us to process 2013-02-05T00:58:15 < Laurenceb_> and should be out of the audio range 2013-02-05T00:58:16 < Laurenceb_> so? 2013-02-05T00:58:27 < Laurenceb_> oh 2013-02-05T00:58:38 < Laurenceb_> yeah you lose high frequencies... 2013-02-05T00:58:50 < ds2> zyp:does that use an off the shelf ESC? 2013-02-05T00:59:01 < Laurenceb_> ok maybe 256... 2013-02-05T00:59:07 < zyp> ds2, yes 2013-02-05T00:59:29 < ds2> zyp: what does the processor do? just the stablization? 2013-02-05T00:59:34 < Laurenceb_> not as wide but might still work 2013-02-05T00:59:52 < Laurenceb_> remember most HAM stuff uses 8Ksps 2013-02-05T00:59:53 < zyp> ds2, hard to answer when the firmware is not done. :p 2013-02-05T01:00:00 < Laurenceb_> and audio is lower bandwidth 2013-02-05T01:00:04 < Laurenceb_> <10KHz 2013-02-05T01:00:14 < Laurenceb_> on that frequency range 2013-02-05T01:00:32 < ds2> zyp: what will it do? :D 2013-02-05T01:00:36 < Laurenceb_> i dont think the aliasing is an issue if its properly tuned 2013-02-05T01:00:52 < Laurenceb_> its only going to alias noise 2013-02-05T01:00:56 < zyp> ds2, well, first I need to get the stabilization working properly :p 2013-02-05T01:01:02 < ds2> oh 2013-02-05T01:01:04 < Laurenceb_> as theres no audio up at those frequencies 2013-02-05T01:01:10 < zyp> then I can worry about other cool stuff to do :p 2013-02-05T01:01:43 < emeb> Laurenceb_: aliasing causes OOB signals to appear in-band at the wrong frequencies. You get other stations interfering 2013-02-05T01:02:17 < emeb> inverted, frequency shifted, etc. 2013-02-05T01:02:22 < Laurenceb_> hmmm 2013-02-05T01:02:39 < Laurenceb_> but other stations are going to be quite spaced apart 2013-02-05T01:03:27 < Laurenceb_> so you need to consider the rejection 2013-02-05T01:03:40 < emeb> AM band channel allocations are 9kHz in the US IIRC. at night with ionosphere you can get far away stuff on nearby channels. 2013-02-05T01:03:48 < Laurenceb_> hmmm 2013-02-05T01:04:00 < Laurenceb_> thats some serious filtering 2013-02-05T01:04:06 < emeb> yep :) 2013-02-05T01:04:06 < Laurenceb_> uk its not like that 2013-02-05T01:04:18 < Laurenceb_> yeah theres no way youll manage that on an F3 2013-02-05T01:04:27 < Laurenceb_> i was thinking more like 40KHz or so 2013-02-05T01:05:11 < Laurenceb_> very close stuff will end up a horrible aliased mess unless you are careful 2013-02-05T01:06:03 < emeb> yeah - that's why I concluded OS is needed. 2013-02-05T01:06:07 < Laurenceb_> planck-taper 256 sample will be about 20dB down for channels that close 2013-02-05T01:06:13 < Laurenceb_> i see 2013-02-05T01:06:53 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-05T01:07:49 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-05T01:08:06 < Laurenceb_> http://upload.wikimedia.org/wikipedia/commons/f/f2/Window_functions_in_the_frequency_domain.png 2013-02-05T01:08:21 < Laurenceb_> yeah, none of the functions drop off fast enough 2013-02-05T01:08:46 < Laurenceb_> your other stations will come in around 1 or 2 on the x axis 2013-02-05T01:08:52 < emeb> here's a BH window, 256 samples, 1x OS: http://imagebin.org/245482 2013-02-05T01:09:20 < Laurenceb_> i see 2013-02-05T01:09:23 < emeb> here's the same window, 256 samples, 4x OS http://imagebin.org/245483 2013-02-05T01:09:30 < Laurenceb_> yeah this would work in the UK... 2013-02-05T01:09:53 < Laurenceb_> what do you mean by oversample? 2013-02-05T01:09:56 < emeb> that's keeping the tuner at a fixed frequency and sweeping the stimulus signal across it 2013-02-05T01:10:00 < Laurenceb_> i see 2013-02-05T01:10:19 < emeb> By OS I mean that I integrate 4x more samples per period than the ADC collected. 2013-02-05T01:10:28 < Laurenceb_> hmmm 2013-02-05T01:10:30 < emeb> so I collect 64 samples and integrate 256 2013-02-05T01:10:42 < Laurenceb_> where do the extra ones come from? 2013-02-05T01:10:51 < emeb> history - previous samples saved 2013-02-05T01:11:04 < Laurenceb_> wont that kill your audio bandwdith? 2013-02-05T01:11:16 < emeb> yes. thats why I don't see it working. 2013-02-05T01:11:28 < emeb> unless you reduce the ADC rate significantly 2013-02-05T01:11:29 < Laurenceb_> im confused 2013-02-05T01:11:37 < Laurenceb_> how are you _meant_ to do this 2013-02-05T01:11:43 < emeb> and then it won't have a useful bandwidth. 2013-02-05T01:11:43 < Laurenceb_> must be a solved problem 2013-02-05T01:11:53 < Laurenceb_> wait 2013-02-05T01:12:05 < Laurenceb_> if the channels are at constant offset... 2013-02-05T01:12:10 < Laurenceb_> its just OFDM :P 2013-02-05T01:12:16 < Laurenceb_> so rectangular 2013-02-05T01:12:21 < Laurenceb_> are they spaced that well? 2013-02-05T01:13:51 < emeb> well, I suppose you could use OFDM techniques to channelize, 2013-02-05T01:14:12 < emeb> but that would give you data from all the channels you don't care about too. Wasted cycles for unused data. 2013-02-05T01:19:42 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-05T01:20:13 < dirty_d> im trying out pyusb, its pretty neat 2013-02-05T01:20:18 < dongs> sup blogs 2013-02-05T01:20:27 < Laurenceb_> i didnt mean that 2013-02-05T01:20:37 < Laurenceb_> i meant maybe rectangular will remove the other channels 2013-02-05T01:20:39 < dirty_d> literally 8 lines of code to do a usb bulk write 2013-02-05T01:20:44 < Laurenceb_> if they are at constant offsets 2013-02-05T01:20:49 < Laurenceb_> python? 2013-02-05T01:20:53 < Laurenceb_> inb4 dongs 2013-02-05T01:20:53 < dirty_d> yea 2013-02-05T01:21:00 < dirty_d> someone here heates python 2013-02-05T01:21:03 < dirty_d> hates* 2013-02-05T01:21:59 < emeb> Laurenceb_: yeah - rectangular window has narrower response, but rejection falls off slower. 2013-02-05T01:22:42 < dirty_d> why the hell would i get broken pipe errors with a bulk write? 2013-02-05T01:22:49 < dirty_d> my stm32 does get the data though 2013-02-05T01:23:15 < dirty_d> its kinda wierd too, when i do a single bulk write, i get a zlp, the data, then another zlp 2013-02-05T01:23:16 < Laurenceb_> emeb: this is odd... 2013-02-05T01:23:19 < dirty_d> all o the same bulk endpoint 2013-02-05T01:23:25 < emeb> and the bandwidth of the stations means that there's more energy besides just at the nulls. 2013-02-05T01:23:29 < Laurenceb_> its clearly possible to have a high enough cut off 2013-02-05T01:23:40 < Laurenceb_> but i dont quite see how you do it with dsp 2013-02-05T01:23:45 < zyp> dirty_d, how large are your bulk writes? 2013-02-05T01:23:53 < Laurenceb_> without cutting off the audio bandwidth 2013-02-05T01:23:53 < emeb> Laurenceb_: what do you mean? 2013-02-05T01:23:54 < dirty_d> zyp, 'Hello!' bytes 2013-02-05T01:23:59 < Laurenceb_> ill have to think.... 2013-02-05T01:24:02 < zyp> oh 2013-02-05T01:24:21 < zyp> that's not normal, are you sure you are doing it correctly? 2013-02-05T01:24:26 < dirty_d> im creating my endpoint the same as the ep0 one that works 2013-02-05T01:24:31 < dirty_d> except using bulk 2013-02-05T01:24:39 < emeb> Laurenceb_: It can be done w/ enough processing cycles/time. Don't think that F303 has the horsepower for the input bandwidth that would be useful tho. 2013-02-05T01:24:47 < dirty_d> zyp, correctly on which end? 2013-02-05T01:24:51 < zyp> both 2013-02-05T01:25:04 < dirty_d> zyp, im obviously not doing it right someehre 2013-02-05T01:25:14 < dirty_d> i think the python end is opk though 2013-02-05T01:25:34 < dirty_d> i do recieve 'Hello!' on the stm32 2013-02-05T01:25:43 < dirty_d> but i get a zlp before and after it 2013-02-05T01:25:52 < zyp> bulk transfers are terminated by non-MSP sized packets, which means ZLP for mod MSP sized transfers 2013-02-05T01:25:55 < dirty_d> thats not supposed to happen is it? 2013-02-05T01:26:13 < zyp> but that shouldn't apply to a six byte transfer 2013-02-05T01:26:28 < zyp> pastebin your python? 2013-02-05T01:27:12 < dirty_d> http://codepad.org/8DyiaeQD 2013-02-05T01:27:42 < zyp> I haven't tried bulk write with pyusb myself, I've only used control transfers and bulk IN 2013-02-05T01:28:17 < dirty_d> zyp, also after i get the CTR_RX and 'Hello!' on the stm32, i write a zlp, but i never get a CTR_TX for it 2013-02-05T01:28:52 < zyp> of course not, you are not reading it on host 2013-02-05T01:28:52 < dirty_d> im not sure if the low level usb drivers should do thta, or pyusb 2013-02-05T01:29:09 < dirty_d> oh ok 2013-02-05T01:29:13 < zyp> host only sends IN token when asked to read from a device 2013-02-05T01:29:25 < dirty_d> so maybe if the program doesnt crash with broken pipe, and i have a read after i will get CTR_TX 2013-02-05T01:29:36 < zyp> yes 2013-02-05T01:30:04 < dirty_d> do i have to do anything special once i read the data on the stm32? 2013-02-05T01:30:11 < dirty_d> or is the ACK it sends supposed to be all? 2013-02-05T01:30:38 < zyp> you don't have to do anything, bulk transfers are much simpler than control 2013-02-05T01:30:48 < zyp> (and unidirectional in nature) 2013-02-05T01:30:50 < dirty_d> hmm, weird 2013-02-05T01:31:00 < dirty_d> especially since its technically working 2013-02-05T01:31:37 -!- jef79m_lurking is now known as jef79m 2013-02-05T01:37:20 < zyp> I'd test it here, but not now, too late 2013-02-05T01:38:58 < dongs> blame lunix 2013-02-05T01:40:33 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32 2013-02-05T01:40:34 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host] 2013-02-05T01:40:34 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-05T01:43:44 < dirty_d> zyp, i gotta step away from this and get some drinks anyhow 2013-02-05T01:43:52 < dirty_d> dongle, damn lunix 2013-02-05T01:43:56 < dirty_d> dongle 2013-02-05T01:43:58 < dirty_d> i hate that word 2013-02-05T01:44:38 < Erlkoenig> sounds like a lunar unix 2013-02-05T01:45:47 < gxti> lusers use lunix 2013-02-05T01:46:10 < gxti> i think instead of buying a $115 dev boatd i will try to make something very simple work on my f4disco instead :p 2013-02-05T01:46:22 < Erlkoenig> what OS do lwinners use? 2013-02-05T01:50:26 * Laurenceb_ zzz 2013-02-05T01:52:53 < GargantuaSauce> the only people who win avoid software entirely 2013-02-05T01:55:00 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-05T02:06:50 < dongs> what $115 devboad 2013-02-05T02:07:23 < zyp> waveshare I guess 2013-02-05T02:08:10 < gxti> yeah, open407z 2013-02-05T02:08:46 < gxti> package a with the lcd and memories 2013-02-05T02:15:15 * emeb is a luser 2013-02-05T02:15:52 < emeb> Erlkoenig: lwinners use lwindows 2013-02-05T02:16:31 < Erlkoenig> hmm why? because of it being primary target of proprietary software vendors? 2013-02-05T02:17:06 < emeb> nah - because more malware. 2013-02-05T02:17:30 < GargantuaSauce> malware is proprietary too 8) 2013-02-05T02:23:12 < dongs> o 2013-02-05T02:23:43 < dongs> its obvious why noone writes malware for lunix 2013-02-05T02:23:46 < dongs> cuz there are no lunix users 2013-02-05T02:23:54 < dongs> and even if there were 2013-02-05T02:23:58 < dongs> you'd have to target like 200 different distros 2013-02-05T02:24:40 -!- Thorn__ [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer] 2013-02-05T02:25:00 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-05T02:25:00 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Changing host] 2013-02-05T02:25:01 -!- Thorn__ [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-05T02:40:44 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-05T02:47:54 < R2COM> linux is like Mafia, invincible 2013-02-05T03:10:07 < dongs> I would think "aids" is a better analogy 2013-02-05T03:10:53 < dongs> its loud/annoying supporters are everywhere yet nobody wants it 2013-02-05T03:10:56 < Erlkoenig> that one would fit for windows too... 2013-02-05T03:11:34 < dongs> yeah, except for "noone wants it" part. 2013-02-05T03:11:57 < Erlkoenig> i think noone wants windows because it is so good but rather because all the fancy software requires it 2013-02-05T03:13:17 < R2COM> well truth is truth, windows is still more convenient OS to work with, especially for simulation of EM related stuff, and other Eda/modeling related things. 2013-02-05T03:14:00 < Erlkoenig> because of its well design or because of the software that runs on it? 2013-02-05T03:14:01 < R2COM> sometimes when the project is hot, you want to concentrate on that, rather than trying to fix some broken library etc... 2013-02-05T03:14:57 < R2COM> Erlkoenig: not quite sure, but it just works more smooth, less shit happening, maybe both, because people who write software for it already familiar with its interface, and interface is firmly supported by Ms 2013-02-05T03:15:42 < Erlkoenig> but then on windows you have to work around weird quirks, understand scurrile WinAPI functions, have a lot of work for deployment 2013-02-05T03:15:45 < R2COM> but again thats not always case, for example Cadence IC design suite runs on linux only 2013-02-05T03:15:58 -!- Thorn__ [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer] 2013-02-05T03:16:04 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has joined ##stm32 2013-02-05T03:16:04 -!- Thorn__ [~Thorn@128-73-112-97.broadband.corbina.ru] has quit [Changing host] 2013-02-05T03:16:04 -!- Thorn__ [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-05T03:16:06 < R2COM> Erlkoenig: dont remember anything like that. sorry. 2013-02-05T03:16:31 < Erlkoenig> look e.g. at the CreateProcess API vs. fork()/exec() 2013-02-05T03:16:46 < Erlkoenig> it's a huge mess and can't do more than fork()+exec()+dup2() 2013-02-05T03:16:51 < R2COM> I used lots of big eda tools on windows, like Sonnet, ADS, Genesys, OrCad... in moderately complex and some bigger projects, never had anything to do with api or other stuff 2013-02-05T03:17:27 < Erlkoenig> yeah than you were lucky and somewhat far away from the actual OS... doing system-near stuff is more painful 2013-02-05T03:17:42 < dongs> windows API is perfectly fine 2013-02-05T03:17:47 < dongs> and whats most important, its DOCUMENTED 2013-02-05T03:17:48 < R2COM> i think... i clearly said before that it is convenient for people working with eda tools. 2013-02-05T03:17:49 < dongs> unlike lunix trash 2013-02-05T03:17:57 < Erlkoenig> POSIX is somewhat documented, too 2013-02-05T03:18:11 < dongs> "somewhat" doesnt cut it in 2013 2013-02-05T03:18:20 < Erlkoenig> feel the understatement 2013-02-05T03:18:32 < Erlkoenig> also windows has badly documented parts, too 2013-02-05T03:18:34 < dongs> im writing a lunix media device driver and the entire /media dir has zero comments 2013-02-05T03:18:35 < GargantuaSauce> there are windows apis with poor documentation 2013-02-05T03:18:41 < GargantuaSauce> and windows apis with WRONG documentation 2013-02-05T03:18:44 < Erlkoenig> like e.g. WinSxS - the only docs i found were outdated tutorials 2013-02-05T03:18:48 < dongs> or docs, or good practices of how to do shit etc. 2013-02-05T03:18:57 < dongs> there are like 5 different drivers and they do same shit 5 dsifferent ways 2013-02-05T03:19:06 < dongs> its actually fucking incredible how lunix fags get anything done 2013-02-05T03:19:08 < dongs> wiht this kinda shit 2013-02-05T03:19:10 < R2COM> longer time ago I did write some Apps for win, to talk with usb hardware through dlls too, as far as I remember was relatively painless process 2013-02-05T03:19:23 < Erlkoenig> using WInUSB? 2013-02-05T03:19:32 < dongs> back then winusb wasnt around 2013-02-05T03:19:33 < R2COM> not remember, but I used some .dll files 2013-02-05T03:19:36 < dongs> i usd cyapi for years 2013-02-05T03:19:41 < dongs> works perfectly fine 2013-02-05T03:19:50 < Erlkoenig> dongs: ok i was talking about the userspace... don't know how messy the kernels insides are 2013-02-05T03:20:01 < dongs> Erlkoenig: it doesnt matter. 2013-02-05T03:20:04 < dongs> whole thing is a fucking mess. 2013-02-05T03:20:14 < karlp> bang! http://www.youtube.com/watch?v=I9zBGgpzl0I 2013-02-05T03:20:15 < Erlkoenig> so is windows, seriously 2013-02-05T03:20:28 < Erlkoenig> ever tried shipping an application with dependencies? 2013-02-05T03:20:55 < gxti> lol you're still talking about this 2013-02-05T03:21:06 < Erlkoenig> hmm did i talk here before about that? 2013-02-05T03:21:12 < Erlkoenig> and yes also linux userspace contains lots of shit 2013-02-05T03:21:17 < R2COM> also remember writing some database application with C#, without actually much detailed understanding, just spent several hours, going through its help files, examples...and bam! done... so its relatively convenient for those who need some quick things without digging much into all detailed shit (unless their full time work is Application programmer) 2013-02-05T03:21:21 < dongs> 10:20 < Erlkoenig> so is windows, seriously 2013-02-05T03:21:22 < dongs> 10:20 < Erlkoenig> ever tried shipping an application with dependencies? 2013-02-05T03:21:23 < dongs> yes 2013-02-05T03:21:39 < dongs> ever tried shipping a lunix application with dependencies, LOLLLLLLLLLLLLL 2013-02-05T03:21:59 < Erlkoenig> yeah, you just have to specify the dependent libraries in your packages description file 2013-02-05T03:22:05 < dongs> hahaha. 2013-02-05T03:22:09 < Erlkoenig> like: DEPENDS="gtk-3 libusb sdl" 2013-02-05T03:22:10 < dongs> oh man 2013-02-05T03:22:11 < dongs> sitll laughing 2013-02-05T03:22:27 < gxti> hey 2013-02-05T03:22:31 < Erlkoenig> i found this easier than building an installer that installs gtk-3, libusb, sdl and my application 2013-02-05T03:22:34 < gxti> maybe you should talk about something else, this is obviously going nowhere 2013-02-05T03:22:40 < GargantuaSauce> and then prey sdl means sdl 1.2 and not 1.3 2013-02-05T03:22:54 < Erlkoenig> then write "sdl = 1.2" or something 2013-02-05T03:23:02 < karlp> (you should totally be talking about supersonic ping pong balls!) 2013-02-05T03:23:06 < gxti> ^ 2013-02-05T03:23:11 < GargantuaSauce> i think we should be able to come to the compromise that software is terrible and we are all huge masochists for pursuing its use and development 2013-02-05T03:23:22 < R2COM> karlp: can it kill? 2013-02-05T03:23:30 < Erlkoenig> GargantuaSauce: yes that sums it up well 2013-02-05T03:23:33 < dongs> Erlkoenig: you seem t o be fucking confused 2013-02-05T03:23:40 < dongs> Erlkoenig: noone on windows uses SDL, GTK-3, or libusb. 2013-02-05T03:23:45 < Erlkoenig> and you seem to be a fucking troll 2013-02-05T03:23:46 < dongs> Erlkoenig: the problem is you're using shit software on a proper OS 2013-02-05T03:23:49 < dongs> thats why you're having trouble 2013-02-05T03:23:54 < gxti> whoa, stop the goddamn presses 2013-02-05T03:23:58 < gxti> dongs is a *troll*? 2013-02-05T03:24:02 < dongs> noone fucking wants your lunix shit on windows. 2013-02-05T03:24:06 < dongs> GTK apps on windows look so fucking out of place 2013-02-05T03:24:11 < gxti> seems HIGHLY IMPROBABLE 2013-02-05T03:24:11 < dongs> keep that shit on your retarded lunix desktop 2013-02-05T03:24:16 < dongs> same goes for SDL 2013-02-05T03:24:23 < dongs> noone wants that shit. directx rapes the fuck out of it. 2013-02-05T03:24:25 < Erlkoenig> so what would you recommend for portable applications? 2013-02-05T03:24:36 < dongs> and libusb, use cyapi or winusb, problem fucking solved 2013-02-05T03:24:39 < Erlkoenig> portable as in Windows&Linux, because the client wanted so 2013-02-05T03:24:39 < GargantuaSauce> directx is a nightmare and you should know that 2013-02-05T03:24:39 < dongs> none of the opensores aids 2013-02-05T03:24:48 < dongs> i'll never install anything that "depends" on libusb on windows because those fucks cant code worth a shit 2013-02-05T03:24:56 < dongs> libusb installs a filter driver on top of EVERY usb device on your system 2013-02-05T03:24:58 < Erlkoenig> so give me an actual alternative, please 2013-02-05T03:25:03 < Erlkoenig> yes i know 2013-02-05T03:25:07 < dongs> and it fucking crashes all the fucking time and causes usb resets on EVERYTHING 2013-02-05T03:25:20 < Erlkoenig> not for me but whatever 2013-02-05T03:25:35 < Erlkoenig> if my program were WIndows-only, okay, but what if the client wants windows&linux? 2013-02-05T03:25:52 < dongs> tell them to fuck off 2013-02-05T03:25:57 < Erlkoenig> sounds good 2013-02-05T03:26:10 < Erlkoenig> also, if i were to use cyapi, is it easy to write an installer that installs it? 2013-02-05T03:26:20 < dongs> you dont need to install anything 2013-02-05T03:26:21 < Erlkoenig> and DirectX 2013-02-05T03:26:24 < dongs> ditto 2013-02-05T03:26:30 < Erlkoenig> why not? that stuff is not always installed 2013-02-05T03:26:35 < dongs> it doesnt need to be 2013-02-05T03:26:48 < Erlkoenig> so how to use it if it's not present? 2013-02-05T03:26:49 < dongs> cyapi is one dll that goes next to your app. nothing to install 2013-02-05T03:26:58 < dongs> directx is bundled wiht every fucking version of windows since like 98 2013-02-05T03:27:01 < dongs> where the fuck have you been? 2013-02-05T03:27:15 < dongs> if you're supporting anything below XP in 2013, you're wasting time anyway 2013-02-05T03:27:16 < Erlkoenig> ah okay messing around with Gtk was different because its like 1000 DLL's and stuff 2013-02-05T03:27:25 < dongs> yes, excatly 2013-02-05T03:27:29 < dongs> 1000 worthless dlls 2013-02-05T03:27:30 < Erlkoenig> are like all proprietary apps just one DLL? 2013-02-05T03:27:38 < Erlkoenig> s/apps/libs 2013-02-05T03:27:41 < gxti> lol no 2013-02-05T03:27:52 < Erlkoenig> that i would want to use from my app 2013-02-05T03:28:39 < dongs> static linking of stuff on windows is fucking great 2013-02-05T03:28:43 < dongs> look at mpc-hc for example 2013-02-05T03:28:53 < Erlkoenig> static linking is so 80ies 2013-02-05T03:28:55 < gxti> hahahahah 2013-02-05T03:28:56 < dongs> 14meg .exe that has EVERYTHING in it, all video/audio/stream/etc d ecoders/renderers/etc 2013-02-05T03:29:01 < dongs> then look at vlc 2013-02-05T03:29:05 < dongs> 1000 tiny dlls 2013-02-05T03:29:08 < dongs> one for ecah shit 2013-02-05T03:29:11 < dongs> fucking useless. 2013-02-05T03:29:29 < dongs> vlc takes like 10 seconds just to startup 2013-02-05T03:29:35 < dongs> while it fucking loads every piece of itself 2013-02-05T03:29:47 < dongs> im sure your gtk app takes ages to load too 2013-02-05T03:29:53 < dongs> while it grinds through 1000 bullshit opensores libs 2013-02-05T03:30:06 < Erlkoenig> well this is how it is done on POSIX/UNIX systems, but does not translate too well on windows 2013-02-05T03:30:39 < Erlkoenig> the downside of the static linking approach is if you have one huge library statically linked into 10 applications, and if you start these 10 applications, your code will end up 10x in your RAM 2013-02-05T03:30:51 < dongs> noone gives a fuck abou that in 2013 2013-02-05T03:30:52 < Erlkoenig> and you the HDD 2013-02-05T03:30:54 < dongs> my desktop has 16gig ram, yours? 2013-02-05T03:30:59 < Erlkoenig> 6 2013-02-05T03:31:07 < Erlkoenig> however, i'd still consider it bad design 2013-02-05T03:31:07 < dongs> stop using AMD garbage then 2013-02-05T03:31:22 < dongs> its not 2013-02-05T03:31:22 < Erlkoenig> i am not using AMD on this computer 2013-02-05T03:31:29 < R2COM> I am using AMD 2013-02-05T03:31:36 < Erlkoenig> and it doesn't matter after all 2013-02-05T03:31:54 < Erlkoenig> solving problems with "more ram" & "more HDD" doesn't seem to be the best solution 2013-02-05T03:32:14 < dongs> Erlkoenig: also your '10 apps ' argument is COMPLETELY worthless. 2013-02-05T03:32:26 < dongs> gtk + vlc on windows im sure share more than 50% of garbage thats inside thier separate dirs 2013-02-05T03:32:36 < dongs> but of course they're separate, both on disk and in memory. 2013-02-05T03:32:42 < gxti> f1 has a way to use an alternate vector table right? i remember seeing it but i can't remember where 2013-02-05T03:32:50 < Erlkoenig> yes because it doesn't work on windows 2013-02-05T03:32:52 < dongs> gxti: sure, vectors in rap 2013-02-05T03:32:54 < dongs> err ram 2013-02-05T03:33:05 < dongs> NVIC_Set(something vector address or whatever 2013-02-05T03:33:10 < Erlkoenig> on linux, if you start 10 Gtk Applications, the Gtk code will exist exactly ONCE in the RAM and on the HDD 2013-02-05T03:33:21 < gxti> i guess i should go look at stm32 bootloaders that already exist, and see how they do stuff 2013-02-05T03:33:37 < gxti> i'd like to minimize the amount of custom crap that has to go into the app's link script 2013-02-05T03:33:51 < dongs> Erlkoenig: unless your app needs gtk-2.1.1 and anotehr one needs gtk-2.0.1 and another needs 2.3.7 2013-02-05T03:33:58 < Erlkoenig> on Cortex-M4 there is the VTOR register don't know about the F1 though. 2013-02-05T03:34:15 < Erlkoenig> dongs: this rarely happens because the distributors compile them all against the latest version 2013-02-05T03:34:21 < Erlkoenig> s/compile/link 2013-02-05T03:34:35 < dongs> haha distributors 2013-02-05T03:34:36 < ds2> ] 2013-02-05T03:34:46 < dongs> ok wahtever this conversation is pointless. lunixfags will remain lunixfags. 2013-02-05T03:34:52 < gxti> yes 2013-02-05T03:34:58 < GargantuaSauce> that we will 2013-02-05T03:34:59 < dongs> just stop using GTK+trash on windows, apps look so fucking worthless when made with that 2013-02-05T03:35:01 < gxti> and winfags will remain winfags and nothing of value was gained by anyone 2013-02-05T03:35:05 < gxti> so stfu already 2013-02-05T03:35:19 < gxti> and let's not speak of the macfags 2013-02-05T03:35:21 < GargantuaSauce> fwiw i think gtk is garbage on any platform 2013-02-05T03:35:33 < Erlkoenig> and would you guys generally please stop using "fag" as a swearword, as this is insulting for some people 2013-02-05T03:35:42 < R2COM> lol 2013-02-05T03:35:46 < gxti> i'm not using it as a swearword and i don't care about some people 2013-02-05T03:35:51 < GargantuaSauce> what are you some kinda HOMO 2013-02-05T03:36:20 < Erlkoenig> people who don't care about other people are antisocial 2013-02-05T03:36:26 < gxti> yes 2013-02-05T03:36:28 < gxti> and? 2013-02-05T03:36:34 < GargantuaSauce> also -fag as a suffix isnt really an insult 2013-02-05T03:36:35 < gxti> when you're paying me i'll be sensitive 2013-02-05T03:36:42 < gxti> yes, what he said 2013-02-05T03:37:12 < R2COM> be social, like fags 2013-02-05T03:37:31 < GargantuaSauce> it's the ultimate honourific of the internet 2013-02-05T03:37:53 < gxti> ANYWAY 2013-02-05T03:38:03 < gxti> what is a good f1 bootloader to look at for ideas? 2013-02-05T03:38:08 < Erlkoenig> NOP 2013-02-05T03:38:21 * ds2 screams u-boot 2013-02-05T03:38:22 < ds2> ;) 2013-02-05T03:38:24 * ds2 runs 2013-02-05T03:41:06 < R2COM> hmm also I just wonder why to use gtk+ to create application on windows? 2013-02-05T03:41:21 < R2COM> wouldnt one who wants some quick app use Microsoft Visual Express? 2013-02-05T03:41:27 < gxti> don't start again, we just finished 2013-02-05T03:41:39 < R2COM> it was just question 2013-02-05T03:41:51 < Erlkoenig> because of its portability to Linux 2013-02-05T03:42:32 < gxti> ds2: i'd use it if it were available 2013-02-05T03:42:38 < gxti> but it looks like no :p 2013-02-05T03:42:56 < ds2> gxti: I been told by folks here it should be trivial to port the one for the f2 2013-02-05T03:43:37 < gxti> although i kind of want to make a vaguely user-friendly loader, with a web ui instead of tftp shenanigans, so perhaps not 2013-02-05T03:47:38 -!- upgrdman_ [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32 2013-02-05T03:56:44 < GargantuaSauce> in this web twenty world you'd better have a goddamn REST api 2013-02-05T04:00:29 < gxti> hmm, maybe for rev 3 i should just slap a sd card on the bottom and boot from that 2013-02-05T04:01:10 < gxti> then i can use an existing bootloader and not worry about having a nice UI because any monkey can pop out a sd card and overwrite a file 2013-02-05T04:01:39 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-05T04:03:07 < R2COM> dongs: so you run win 8? 2013-02-05T04:03:11 < dongs> of course 2013-02-05T04:03:16 < R2COM> I am thinking of putting it on my main dekstop 2013-02-05T04:03:25 < dongs> thats whaT I did 2013-02-05T04:03:28 < dongs> best decision ever 2013-02-05T04:03:28 < R2COM> but one thing I am still unsure is 2013-02-05T04:03:34 < dongs> i tried it in VM a few months ago, it was amazing 2013-02-05T04:03:43 < dongs> then I just got parts for newpc and loaded win8 2013-02-05T04:03:44 < R2COM> what about those rumors telling that it does not like some custom made drivers for custom hardware? 2013-02-05T04:03:51 < dongs> such as? 2013-02-05T04:04:20 < R2COM> I dont know? just any? if you put it in google there are tons of links on forums like"help it doesnt want to run uncertified driver" 2013-02-05T04:04:37 < R2COM> and they have to use some sort of safe mode for that, or bypass it somehow? there I different workarounds I see 2013-02-05T04:05:07 < R2COM> so lets say I have some custom hardware, with custom made drivers by someone, would it be problem for me to just install it on win 8? 2013-02-05T04:05:23 < R2COM> and select something like "Its OK install this driver even though its unsigned!" 2013-02-05T04:05:23 < R2COM> ? 2013-02-05T04:05:35 < dongs> no unsigned drivers in 64bit 2013-02-05T04:05:41 < dongs> which is fine with me as I can self-sign 2013-02-05T04:05:52 < dongs> and any reputable vendor should be providing proper signed drivers 2013-02-05T04:05:52 < R2COM> what one has to do to self-sign them? 2013-02-05T04:06:19 < dongs> code-signing cert 2013-02-05T04:06:30 < R2COM> it costs something i assume? 2013-02-05T04:06:35 < dongs> yes, obviously 2013-02-05T04:06:40 < dongs> and needs identity check and shit. 2013-02-05T04:06:44 < dongs> so you dont go signing virii 2013-02-05T04:06:54 < gxti> this virus approved by dongs 2013-02-05T04:06:56 < ds2> is there a simple and free disassembler that knows the F1 registers? 2013-02-05T04:07:09 < dongs> ds2: objdump? *ducks* 2013-02-05T04:07:09 < R2COM> hmm.. but ok, actually for those things I could use on older xp laptops too... 2013-02-05T04:07:21 < ds2> dongs: that doesn't know the F1 registers 2013-02-05T04:07:24 < R2COM> so things like Xilinx programmers/debuggers etc, will have no trouble installing on Win 8? 2013-02-05T04:07:31 < dongs> ds2: what do you mean 'registers' 2013-02-05T04:07:33 < ds2> it is simple though ;) 2013-02-05T04:07:36 < dongs> you mean ioregs for peripehrals etc? 2013-02-05T04:07:38 < ds2> dongs: RCC, etc 2013-02-05T04:07:40 < ds2> yeah 2013-02-05T04:07:45 < dongs> uh.. nothing will "know" 2013-02-05T04:07:47 < dongs> those 2013-02-05T04:07:49 < dongs> i wrote some scripts for ida 2013-02-05T04:07:54 < dongs> for F1 peripehrals 2013-02-05T04:07:56 < ds2> ida isn't officially free 2013-02-05T04:08:09 < dongs> well, i actually officially own it and have a support contract. but yeah. 2013-02-05T04:08:17 < dongs> doesnt freetard version work with arm? 2013-02-05T04:08:21 < ds2> but you paid it 2013-02-05T04:08:26 < dongs> right 2013-02-05T04:08:29 < ds2> the freetard version doesn't do .hex files 2013-02-05T04:08:39 < dongs> what does it do then? 2013-02-05T04:08:40 < dongs> .elf? 2013-02-05T04:08:54 < ds2> .elf 2013-02-05T04:08:59 < dongs> shittty. 2013-02-05T04:09:41 < R2COM> apparently i see now some links complaining with xilinx shit not working ok on win 8 2013-02-05T04:10:59 < gxti> people complain about xilinx shit not working on every os 2013-02-05T04:11:00 < R2COM> and there are some workarounds too though.. 2013-02-05T04:11:32 < R2COM> gxti: its not too bad actually (but its bad), I just use for now ISE 13.4 versions XX.4 usually are most patched and complete 2013-02-05T04:11:58 < R2COM> but xilinx tools are messy... yes.. 2013-02-05T04:12:16 < R2COM> and they even want money for it.. to develop for big chips. 2013-02-05T04:17:08 < dongs> i heard windows software is good bceause it costs money 2013-02-05T04:18:02 < R2COM> unless someone takes money spends it on hookers and then still writes bad code 2013-02-05T04:18:38 < R2COM> win 7 eos is 2015, so its ok to stick with it for a while 2013-02-05T04:18:55 < dongs> win7 has same restrictions on signed drivers 2013-02-05T04:18:59 < dongs> so I dont see what the problem is 2013-02-05T04:19:16 < R2COM> yes but in win 7 you can select "install anyway" or something like that 2013-02-05T04:19:20 < R2COM> and driver is still installed 2013-02-05T04:19:51 < R2COM> plus main tools, and xilinx tools are so far so good relatively stable on win 7 2013-02-05T04:27:57 < R2COM> ok...continuing smoking the datasheet of new buck converters 2013-02-05T04:32:31 < Erlkoenig> would an implementation of a parser for context-free grammars with optimization for low memory usage (->mikrocontrollers) be of use for anyone? 2013-02-05T04:35:14 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Quit: Leaving.] 2013-02-05T04:51:23 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-76.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-05T04:54:50 -!- Thorn__ is now known as Thorn 2013-02-05T05:03:55 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-05T05:17:53 -!- Viper168 [~Viper@node61.18.251.72.1dial.com] has joined ##stm32 2013-02-05T05:17:54 -!- Viper168 [~Viper@node61.18.251.72.1dial.com] has quit [Changing host] 2013-02-05T05:17:54 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-05T05:25:13 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-05T05:25:38 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-05T05:37:17 -!- upgrdman_ [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Quit: Ex-Chat] 2013-02-05T05:44:08 < dongs> haha, and here's perfect example of opensores app vs windows one 2013-02-05T05:44:16 < dongs> was lookin for something to do packet sniffing on win8 2013-02-05T05:44:25 < dongs> choice either wireshark or microsoft network monitor 2013-02-05T05:44:37 < dongs> wireshark: 50megs installed, 342394274 dlls, ugly chunky UI 2013-02-05T05:44:53 < dongs> microsoft netwrok monitor: 6 megs, runs great, has all sorts of neat stuff for filtering,highlighting,scripting engine etc 2013-02-05T05:45:03 < dongs> UI doesnt look like it came out of opensauce asshole 2013-02-05T05:45:39 < gxti> poot poot 2013-02-05T05:47:43 < gxti> k so i'm going to be lazy and just make this a tftp server, that's pretty easy for tards to use 2013-02-05T05:50:04 < dongs> pretty easy.. tards.. tftp.. 2013-02-05T05:50:05 < dongs> RIGHT. 2013-02-05T05:50:21 < dongs> i think you mean pretty easy for lunixfgts to use 2013-02-05T05:50:28 < gxti> even you could figure it out dongs 2013-02-05T05:50:56 < gxti> you might have to press some of the buttons on that big square thing next to the mouse, but not many 2013-02-05T05:52:01 -!- DaKu_ [DaKu@dakus.dk] has joined ##stm32 2013-02-05T05:53:26 -!- Netsplit *.net <-> *.split quits: daku 2013-02-05T05:53:26 -!- DaKu_ is now known as DaKu 2013-02-05T05:53:56 < emeb_mac> dongs: did you say you've used some wideband RF front-end tuner chips before? 2013-02-05T05:54:15 < dongs> duno about wideband, just regular VHF/UHF stuff 2013-02-05T05:54:33 < dongs> i designed some stuff with FC0013 (for satellite recevier 2013-02-05T05:54:49 < emeb_mac> OK - same as used in some of the RTLSDR dongles 2013-02-05T05:55:12 < emeb_mac> I guess data sheets for those are NDA only... 2013-02-05T05:55:27 < dongs> wait no, the satellite proj was M88TS2022 tuner 2013-02-05T05:55:33 < dongs> cause the demod was also from same chink place 2013-02-05T05:55:42 < dongs> well 2013-02-05T05:55:43 < dongs> i ahev all teh docs 2013-02-05T05:55:46 < dongs> for fc0013 also 2013-02-05T05:55:48 < dongs> that was some other proj here 2013-02-05T05:55:58 < dongs> i also have vendor code for both 2013-02-05T05:56:05 < dongs> not reverse engineer faggot shit that opensores dudes use 2013-02-05T05:56:20 < dongs> why? i thought you gave up on sdr on f3 2013-02-05T05:56:21 < emeb_mac> cool. 2013-02-05T05:56:31 < emeb_mac> don't want to do it w/ F3 2013-02-05T05:56:37 < emeb_mac> thinking of a "real" frontend now 2013-02-05T05:56:52 < dongs> geez just pay for bladerf 2013-02-05T05:56:58 < emeb_mac> blech 2013-02-05T05:57:03 < R2COM> what are the requirements? 2013-02-05T05:57:05 < R2COM> rf for what? 2013-02-05T05:57:05 < emeb_mac> 300MHz min 2013-02-05T05:57:09 < dongs> emeb_mac: pfft, 2013-02-05T05:57:11 < R2COM> 300Mhz what? 2013-02-05T05:57:14 < dongs> make your own analog frontend 2013-02-05T05:57:20 < dongs> and hook up zeroIF or wahtever to it 2013-02-05T05:57:26 < emeb_mac> bladerf min input is 300MHz 2013-02-05T05:57:37 < dongs> right, so just make your own downconverter 2013-02-05T05:57:39 < dongs> from wahtever frequency 2013-02-05T05:57:40 < emeb_mac> need a transverter or something 2013-02-05T05:57:53 < dongs> it has input for external shit 2013-02-05T05:57:55 < R2COM> whats the rf bandwidth? 2013-02-05T05:57:55 < emeb_mac> and... bladerf is still vaporware 2013-02-05T05:58:23 < emeb_mac> so pony up $400, wait 6mo and pffft. 2013-02-05T05:58:23 < dongs> emeb_mac: it seems more ready than most dickstarter projcs 2013-02-05T05:58:33 < emeb_mac> right 2013-02-05T05:58:48 < emeb_mac> I'll get interested when I can order one for next week delivery 2013-02-05T05:58:57 < dongs> so i suspect software notwithstanding, tehy should have hardware ready to ship by the time dickstarter is over 2013-02-05T05:59:01 < R2COM> what is the fucking bandwidth for rf needed? 2013-02-05T05:59:23 < emeb_mac> R2COM: what do you mean? I'm dreaming bluesky here. 2013-02-05T05:59:33 < emeb_mac> I'm thinking 20-40MHz or so 2013-02-05T05:59:41 < R2COM> not familiar with those, I thought you said you want custom front end 2013-02-05T05:59:44 < dongs> haha, not gonna happen 2013-02-05T05:59:55 < dongs> consumer/cheap tuners will have around 8mhz max 2013-02-05T06:00:25 < R2COM> you can do 20-40Mhz easily with Rf filter, lna, mixer, and if filter 2013-02-05T06:00:52 < R2COM> rf filter has to be designed though if you want some custom center frequency, non standard one 2013-02-05T06:02:03 < emeb_mac> dongs: yep. cheap tuners are for video. 2013-02-05T06:02:08 < R2COM> but components for such things available for example from hittitie.com for low cost 2013-02-05T06:02:09 < emeb_mac> that would be fine too. 2013-02-05T06:02:19 < emeb_mac> ya 2013-02-05T06:02:32 < dongs> hi tittie? 2013-02-05T06:02:42 < R2COM> yeah! 2013-02-05T06:02:49 < emeb_mac> Looking at ADI stuff, but that's $$$ 2013-02-05T06:03:31 < R2COM> if you looking for lower cost, look at minicircuits.com, but usually they want you buy stuff in qty 20, if one is serious about his project it shouldnt be big deal 2013-02-05T06:03:32 < emeb_mac> Maxim has some cheapo terrestrial TV tuners for 50-850 that might work 2013-02-05T06:03:41 < dongs> you cant get those 2013-02-05T06:04:06 < dongs> i tried to get some ebfore, shit needs NDA and noone actualyl stocks them 2013-02-05T06:04:06 < emeb_mac> um... I see some @ mouser? 2013-02-05T06:04:09 < dongs> really? 2013-02-05T06:04:17 < dongs> maybe shit changed, last i tried maxim rf shit was early 2000s 2013-02-05T06:04:30 < dongs> is there an actual datasheet + etc available for those? 2013-02-05T06:04:57 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-05T06:05:06 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-05T06:05:15 < dongs> http://www.hittite.com/products/view.html/view/HMCAD5831LP9BE hyeers what you need 2013-02-05T06:05:43 < emeb_mac> here @ mouser - but not low range... http://www.mouser.com/ProductDetail/Maxim-Integrated/MAX2120CTI+/?qs=sGAEpiMZZMujDqo0UVf7D2b0hsc2xdEI 2013-02-05T06:06:03 < dongs> i fucking hate mouser 2013-02-05T06:06:11 < dongs> once you click a country-specific link on it one time 2013-02-05T06:06:14 < dongs> it keeps forwarding there 2013-02-05T06:06:16 < emeb_mac> ooo - 3 bits. 2013-02-05T06:06:26 < dongs> i clicke some de.mouser.com trash once and now shit keeps coming up in german 2013-02-05T06:06:29 < emeb_mac> yeah - I've had that prob with the website too. 2013-02-05T06:06:49 < dongs> and if you change language, it goes back to main page 2013-02-05T06:06:50 < dongs> FUCK 2013-02-05T06:07:27 < emeb_mac> kill all the web monkeys 2013-02-05T06:07:39 < dongs> oh that maxim shit is for dvb-s 2013-02-05T06:07:44 < emeb_mac> ya 2013-02-05T06:07:51 < dongs> so 900m~2ghz range 2013-02-05T06:07:55 < dongs> but yeah, 40mhz bandwidth 2013-02-05T06:08:04 < dongs> cause typical S/S2 channel is anywhere from 30+mhz 2013-02-05T06:08:05 < emeb_mac> the low-band stuff is still not available through distys 2013-02-05T06:08:18 < dongs> it never was, cause noone actually uses maxim shit 2013-02-05T06:08:24 < emeb_mac> heh 2013-02-05T06:08:42 < emeb_mac> I got some good 16-bit fast ADCs for one project a few years back 2013-02-05T06:09:01 < emeb_mac> err DACs 2013-02-05T06:09:56 < emeb_mac> these tuners: http://www.maximintegrated.com/datasheet/index.mvp/id/7260/t/al 2013-02-05T06:10:38 < emeb_mac> available but no pricing. :P 2013-02-05T06:10:58 < ds2> surely, you can get samples.... ;) 2013-02-05T06:11:13 < emeb_mac> hah! 2013-02-05T06:12:18 < emeb_mac> this looks nice: http://www.analog.com/en/rfif-components/modulatorsdemodulators/adrf6806/products/product.html 2013-02-05T06:12:25 < emeb_mac> $20 @ DK 2013-02-05T06:15:11 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Read error: Connection reset by peer] 2013-02-05T06:15:33 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32 2013-02-05T06:15:34 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host] 2013-02-05T06:15:34 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-05T06:15:59 < R2COM> maxim sure does great stuff, and hell yea, their stuff is used a lot, by defense contractors as well 2013-02-05T06:16:40 < emeb_mac> I've used a fair amount of it in R&D projects. They've got a fairly nice web store. 2013-02-05T06:17:21 < ds2> besides the max232? 2013-02-05T06:17:50 < emeb_mac> heh 2013-02-05T06:18:08 < emeb_mac> lots of RF stuff 2013-02-05T06:18:13 < emeb_mac> converters 2013-02-05T06:18:16 < emeb_mac> etc 2013-02-05T06:18:33 -!- jef79m is now known as jef79m_lurking 2013-02-05T06:19:23 < R2COM> http://www.maximintegrated.com/datasheet/index.mvp/id/5746 2013-02-05T06:20:13 < R2COM> applications: "LO signal synthesis"... very economical solution for LO synthesis 2013-02-05T06:20:49 < R2COM> well, actually in some systems one wants to switch LO *fast* then yea... 2013-02-05T06:23:02 < emeb_mac> that's fast 2013-02-05T06:23:22 < dongs> 'request full dataasheet" 2013-02-05T06:23:39 < emeb_mac> don't forget to say "please" 2013-02-05T06:24:35 < R2COM> this stuff is for mostly projects sponsored by government not for hamtards 2013-02-05T06:25:05 < dongs> hamtards would just use rapeberrypi 2013-02-05T06:25:38 < gxti> no, hamtards would spend 6 months building something from discrete transistors on a copper board and it would still suck 2013-02-05T06:27:19 < ds2> nah... they'll salvage a tv tuner from the 80's 2013-02-05T06:27:29 < emeb_mac> lol 2013-02-05T06:27:54 < ds2> there was a spec an project using those in a rag 2013-02-05T06:28:18 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 276 seconds] 2013-02-05T06:53:30 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-05T07:13:10 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-05T07:13:32 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-05T07:29:31 < upgrdman> why wouldn't the make the full datasheet available? 2013-02-05T07:34:54 < dongs> because hardcore 2013-02-05T07:38:06 < emeb_mac> because afraid of competitors stealing sekrit sauce 2013-02-05T07:39:01 -!- BusError_ [~michel@host81-152-248-48.range81-152.btcentralplus.com] has quit [Ping timeout: 240 seconds] 2013-02-05T07:39:25 -!- BusError_ [~michel@host81-152-248-48.range81-152.btcentralplus.com] has joined ##stm32 2013-02-05T07:41:00 < gxti> i have an input i want to pull down and i do GPIOB->ODR = 0 and it's still being pulled up. this is suboptimal. 2013-02-05T07:41:17 < dongs> wut 2013-02-05T07:47:25 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-05T07:47:47 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-05T07:48:15 < upgrdman> gxti, pin set an output, open-drain? 2013-02-05T07:48:34 -!- jef79m_lurking is now known as jef79m 2013-02-05T07:48:55 < upgrdman> as* 2013-02-05T07:49:07 < gxti> no, input pullup/pulldown 2013-02-05T07:49:15 < upgrdman> you cant pull an input! 2013-02-05T07:49:17 < upgrdman> :) 2013-02-05T07:49:25 < gxti> . . . 2013-02-05T07:49:33 < upgrdman> make it an output, open-drain if all you need is pull-down-ability 2013-02-05T07:49:56 < upgrdman> "ODR" is the output data register.... OUTPUT :) 2013-02-05T07:50:05 < gxti> i know you're trying to be helpful 2013-02-05T07:50:08 < gxti> but you are failing 2013-02-05T07:50:25 < gxti> GPIOx->CNFy value 10 = input with pull-up/pull-down 2013-02-05T07:50:39 < gxti> ODR determines which way it pulls 2013-02-05T07:50:50 < upgrdman> what stm32? 2013-02-05T07:50:57 < gxti> it's working for every pin other than PB4, which is the one i need it to... it's coincidental with NTRST so maybe that one is special 2013-02-05T07:51:19 < gxti> this is f1 2013-02-05T07:52:00 < upgrdman> don't know the f1 :( but the f4 and f0 do not use ODR to dick with pull ups/down 2013-02-05T07:53:00 < upgrdman> maybe check the datasheet for limits on specific pins 2013-02-05T07:54:26 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [] 2013-02-05T07:55:05 < gxti> that was it, i had to use remap to disable jtag 2013-02-05T07:55:48 < gxti> i wonder why that wasn't causing more problems... 2013-02-05T07:58:16 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-05T08:03:06 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-05T08:05:06 -!- jef79m is now known as jef79m_lurking 2013-02-05T08:05:31 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 260 seconds] 2013-02-05T08:18:39 < emeb_mac> ya - jtag remap is fun. 2013-02-05T08:28:57 < dongs> typography in f303 datasheet is really nasty 2013-02-05T08:29:20 < dongs> for some dumb reason they swapped the package order from f103 datasheet 2013-02-05T08:29:24 < dongs> so 100pin shit is first 2013-02-05T08:29:32 < dongs> and its harder to read for altfunctions and shit 2013-02-05T08:32:07 -!- Ranewen [~Ranewen4@82.132.1.100] has joined ##stm32 2013-02-05T08:32:36 < emeb_mac> I'd noticed that too 2013-02-05T08:36:01 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds] 2013-02-05T08:39:04 -!- claude is now known as Claude 2013-02-05T08:58:43 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-05T09:23:27 -!- Ranewen [~Ranewen4@82.132.1.100] has quit [Quit: Leaving] 2013-02-05T09:38:32 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-05T09:40:47 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-05T09:44:28 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-05T09:47:07 < dongs> damn 2013-02-05T09:47:30 < dongs> PA0/PA5 on F303 are fucking useless!! 2013-02-05T09:47:42 < dongs> both TIM2_CH1 anda nothing else usable for pwm 2013-02-05T09:47:44 < dongs> ugh 2013-02-05T09:48:55 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T09:49:58 < dongs> rageeeeee 2013-02-05T09:57:40 -!- Claude is now known as claude 2013-02-05T09:59:44 < dongs> what the hell!! 2013-02-05T10:00:03 < dongs> PB3 is useless too 2013-02-05T10:00:21 < dongs> also TIM2_CH2 thats mirrored elsewehre, but no other alt pwm mapping! 2013-02-05T10:02:24 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T10:03:46 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T10:03:51 -!- barthess [~barthess@86.57.155.106] has quit [Client Quit] 2013-02-05T10:20:51 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T10:21:23 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-05T10:33:57 -!- jef79m_lurking is now known as jef79m 2013-02-05T10:42:31 -!- barthess [~barthess@86.57.155.106] has quit [Ping timeout: 245 seconds] 2013-02-05T10:44:34 -!- jef79m is now known as jef79m_lurking 2013-02-05T10:48:50 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T10:55:12 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-05T11:05:41 -!- jef79m_lurking is now known as jef79m 2013-02-05T11:06:03 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer] 2013-02-05T11:06:27 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-05T11:17:27 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-05T11:19:40 -!- jef79m is now known as jef79m_lurking 2013-02-05T11:22:06 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] 2013-02-05T11:25:17 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-05T11:25:43 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-05T11:34:55 -!- jef79m_lurking is now known as jef79m 2013-02-05T11:39:45 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T11:49:44 -!- jef79m is now known as jef79m_lurking 2013-02-05T11:52:29 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has joined ##stm32 2013-02-05T11:53:02 -!- jv-- [jv@pilsedu.cz] has quit [Read error: Operation timed out] 2013-02-05T11:53:09 -!- jv-- [jv@pilsedu.cz] has joined ##stm32 2013-02-05T12:02:42 < dongs> 5x5 ball array 2013-02-05T12:06:39 < Laurenceb_> now emeb has got me obsessed with how to do sdr properly 2013-02-05T12:07:35 < dongs> hey Laurenceb_ i got a pro/EE question 2013-02-05T12:07:45 < dongs> sayt i have voltage going into ADC through 10k/1k divider 2013-02-05T12:07:52 < dongs> wat happens if I use same pin as dac 2013-02-05T12:08:00 < dongs> what comes out of the voltage part 2013-02-05T12:08:06 < dongs> divider works backwards/ or wat 2013-02-05T12:08:19 < jpa-> divider is not a multiplier backwards :) 2013-02-05T12:08:20 < Laurenceb_> depends on the current you draw 2013-02-05T12:08:28 < dongs> jpa-: yes i gathered that much lol 2013-02-05T12:08:31 < Laurenceb_> with 0 current, just the DAC voltage 2013-02-05T12:08:44 < jpa-> usually DAC voltage or if it is too loaded then something lower 2013-02-05T12:08:50 < dongs> mhm 2013-02-05T12:09:12 < dongs> but it will ahve 1k pulldown 2013-02-05T12:09:16 < zyp> depends on DAC output impedance and sink impedance 2013-02-05T12:09:19 < dongs> 10K to out, 1K to gnd 2013-02-05T12:09:26 < dongs> this is on F3 2013-02-05T12:09:35 < zyp> I suspect DAC drives too strong for 1k to GND to matter much 2013-02-05T12:09:37 < dongs> PA5, ADC2_IN2 or DAC1_OUT2 2013-02-05T12:10:10 < zyp> so you can probably just consider that 1k nonexistant and only consider the 10k in series 2013-02-05T12:10:57 < dongs> if dac is going into something like a/v transmitter 2013-02-05T12:11:22 < dongs> that should have practically no current draw right 2013-02-05T12:12:05 < zyp> hard to tell, try it 2013-02-05T12:12:09 < dongs> right 2013-02-05T12:12:19 < dongs> its neat that PA5 is a useless pin but has that dualfunction on it. 2013-02-05T12:12:36 < dongs> im revising pinout for the F3 flyingthing, thats why im asking 2013-02-05T12:12:53 < Laurenceb_> aha 2013-02-05T12:13:00 < Laurenceb_> i think i worked it out 2013-02-05T12:13:02 < dongs> since it was pin incompatible enough to just tear it down and redo 2013-02-05T12:13:05 < zyp> OSD? 2013-02-05T12:13:07 < dongs> no 2013-02-05T12:13:11 < Laurenceb_> emeb needs IIR on I and Q 2013-02-05T12:13:15 < Laurenceb_> from his SDR 2013-02-05T12:13:16 < zyp> why a/v transmitter? 2013-02-05T12:13:21 < dongs> audio output 2013-02-05T12:13:27 < dongs> beeps or something. vario maybe 2013-02-05T12:13:34 < zyp> ah 2013-02-05T12:14:42 < zyp> do some fancy modulation and use it for telemetry data :p 2013-02-05T12:15:01 < dongs> emeb knows all about fancy modulations 2013-02-05T12:15:06 < zyp> then you can decode it on receiver side and render osd there :p 2013-02-05T12:15:08 < dongs> and de-modulations 2013-02-05T12:15:08 < dongs> yeah 2013-02-05T12:16:49 < dongs> 20 pwm outputs, wtf 2013-02-05T12:16:59 < dongs> i guess F3 is a bit better than F1 with peripheral assignment 2013-02-05T12:18:16 < zyp> F3 is way better, since you don't have to deal with remap groups and collisions and bullshit 2013-02-05T12:18:24 < dongs> http://i.imgur.com/9LMDzEC.png 2013-02-05T12:18:26 < dongs> got this so far 2013-02-05T12:18:28 -!- Laurenceb_ [~Laurence@host86-177-58-138.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-05T12:18:39 < dongs> stuff in column C is required 2013-02-05T12:19:16 < zyp> I suggest just reading the AF table in the datasheet 2013-02-05T12:19:28 < dongs> i did 2013-02-05T12:19:42 < dongs> i presume it can map anything to anything 2013-02-05T12:20:14 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-05T12:20:14 < zyp> when I did my F4 board I printed that, marked every function I wanted on every pin it's available on, then just started by picking the pins with only one interesting function on 2013-02-05T12:20:14 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-05T12:20:14 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-05T12:20:14 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-05T12:20:46 < dongs> is this in RM? 2013-02-05T12:20:50 < zyp> no, datasheet 2013-02-05T12:20:59 < zyp> end of pin mapping chapter 2013-02-05T12:21:09 < dongs> oh, right right after pin mappings 2013-02-05T12:21:13 < zyp> it's the one with cols for each AF number and rows for each pin 2013-02-05T12:21:21 < dongs> right 2013-02-05T12:21:26 < dongs> well, i did same thing using pin mappings 2013-02-05T12:21:37 < zyp> AF table is much more readable :p 2013-02-05T12:22:29 < dongs> so I can do PB3/AF7 and PA15/AF7 and it'll hookup to USART2? 2013-02-05T12:22:35 < dongs> no other faggotry needed? 2013-02-05T12:22:41 < zyp> correct 2013-02-05T12:22:43 < dongs> if so, then my plan will work 2013-02-05T12:23:03 < zyp> it's much more sane than the F1 bullshit with remapping entire groups of pins 2013-02-05T12:23:07 < dongs> right 2013-02-05T12:24:32 < dongs> I also like that TIM3/TIM16+17 are possible to remap to same pin 2013-02-05T12:24:47 < dongs> then I can finally have different pwm period across those pins 2013-02-05T12:25:02 < dongs> where on F1 its only TIM3, so they all have to be 400hz or whatever. 2013-02-05T12:31:12 -!- jef79m_lurking is now known as jef79m 2013-02-05T12:31:21 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-05T12:38:41 -!- jef79m is now known as jef79m_lurking 2013-02-05T12:43:16 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-05T12:49:25 < dongs> oh, cool. the pins I want are all FT :D 2013-02-05T12:49:47 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T12:49:54 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Read error: Connection reset by peer] 2013-02-05T12:50:34 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-05T13:04:58 < dongs> what the hel lis a "FM+ capable" 2013-02-05T13:06:52 < dongs> oh.. i2c crap? 2013-02-05T13:06:53 < dongs> fail 2013-02-05T13:07:29 < dongs> dOnGs 2013-02-05T13:14:19 < Thorn> received stellaris launchpad and meanwell PSUs which have a 400V cap on the high side. isn't it a little too small for ~220V? 2013-02-05T13:17:06 < Thorn> which is 380Vp-p 2013-02-05T13:21:12 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-05T13:27:52 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Quit: trepidaciousMBR] 2013-02-05T13:35:16 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Ping timeout: 248 seconds] 2013-02-05T13:35:42 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-05T13:37:33 < zyp> dongs, 1MHz or something 2013-02-05T13:38:06 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-05T13:46:04 < jpa-> Thorn: it's only 220 * sqrt(2) = 311 V 2013-02-05T13:46:25 < jpa-> 380 V is between phases 2013-02-05T13:48:23 < Thorn> ah right, I mixed up sqrt(2) and sqrt(3) 2013-02-05T13:48:27 <+Steffanx> heh 2013-02-05T13:48:31 < Thorn> well that's better 2013-02-05T13:48:47 < Thorn> there's some room to spare then 2013-02-05T13:53:24 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-05T13:58:45 < dongs> zyp: http://i.imgur.com/BsBEqF0.png what you think 2013-02-05T13:59:26 < jpa-> looks like excel 2013-02-05T14:00:28 < jpa-> dongs: does that device have the limitation on PC13/14/15 that "only one can be output"? 2013-02-05T14:00:29 < zyp> spi flash for config? 2013-02-05T14:00:29 < dongs> it is, because excel is a great tool 2013-02-05T14:00:42 < dongs> jpa, not that I know of? 2013-02-05T14:00:46 < dongs> PC13-15 have "low drive current" limitation 2013-02-05T14:00:50 < dongs> other than that i dont htink so? 2013-02-05T14:01:09 < dongs> zyp: or whatever, to avoid wasting device flash 2013-02-05T14:01:37 < dongs> jpa, which device has this limit? 2013-02-05T14:01:56 < zyp> dongs, looks decent enough 2013-02-05T14:02:10 < jpa-> dongs: F4.. hmm let me see if i remember totally wrong 2013-02-05T14:02:16 < dongs> i dont think i forgot anything.. 2013-02-05T14:02:37 < dongs> jpa-: hm, i thnk you're wrong.. i know for sure this is not a problem on F1... or F2 2013-02-05T14:02:52 < jpa-> "GPIOs PC13 to PC15 are restricted: only one I/O at a time can be used as an output" 2013-02-05T14:02:57 < dongs> huh... 2013-02-05T14:03:05 < dongs> lets see 2013-02-05T14:03:14 < jpa-> F4 reference manual in "PWR" "Power supplies" "Battery backup domain" 2013-02-05T14:03:19 < zyp> heh, news to me 2013-02-05T14:03:28 < dongs> PC13, PC14 and PC15 are supplied through the power switch. Since the switch sinks only a limited amount of current 2013-02-05T14:03:32 < dongs> (3 mA), the use of GPIO PC13 to PC15 in output mode is limited: 2013-02-05T14:03:33 < dongs> not on F3 2013-02-05T14:03:35 < dongs> - The speed should not exceed 2 MHz with a maximum load of 30 pF 2013-02-05T14:03:42 < dongs> this is same note as on F1 2013-02-05T14:03:42 < dongs> - These GPIOs must not be used as current sources (e.g. to drive an LED). 2013-02-05T14:04:08 < jpa-> F4 has those restrictions and in addition the "Only one output" 2013-02-05T14:04:12 < dongs> mhm 2013-02-05T14:04:29 < zyp> dongs, does PA2/3 also have TIM2 channels? 2013-02-05T14:06:26 < jpa-> dongs: is the 2 MHz speed enough for your SPI needs? so that you won't have to insert any dummy delays between CS set and start of SPI command 2013-02-05T14:08:16 < dongs> its just CS 2013-02-05T14:08:22 < dongs> as in chip enable 2013-02-05T14:08:25 < dongs> its gonna be going at like... 2013-02-05T14:08:26 < jpa-> dongs: in some cases you may want USB VBUS 2013-02-05T14:08:27 < dongs> ~500Hz 2013-02-05T14:08:52 < jpa-> dongs: yeah, but if you are running SPI at e.g. 25 MHz, the CS won't switch on fast enough without a delay 2013-02-05T14:08:53 < dongs> ah, i suppose i could waste PB5 for VBUS 2013-02-05T14:09:07 < dongs> i think sensors are 10mhz max 2013-02-05T14:09:10 < dongs> prob less 2013-02-05T14:09:16 < jpa-> ok 2013-02-05T14:09:27 < jpa-> yeah, probably it's not a problem and if it is you can add the delay 2013-02-05T14:09:53 < dongs> its GPIO though, so why does it matter 2013-02-05T14:09:58 < dongs> GPIO_set(); SPI_Transmit() 2013-02-05T14:10:13 < dongs> the time between those 2 things is enough even for 2mhz drive to switch im sure 2013-02-05T14:10:28 < jpa-> well depends on how slow your code is :) 2013-02-05T14:10:41 < dongs> zyp: PA2/3 have TIM3 i think 2013-02-05T14:10:41 < jpa-> 36 instructions is a lot 2013-02-05T14:10:47 < zyp> jpa-, it's using stdperiphlib, so it's pretty slow :) 2013-02-05T14:10:52 < jpa-> ah, then :D 2013-02-05T14:11:01 < dongs> TIM2 yea 2013-02-05T14:11:11 < jpa-> if it was zyp inlined++ it might be too fast :) 2013-02-05T14:11:20 < dongs> zyp: but its better to have separate timer domains for PWM out 2013-02-05T14:11:25 < dongs> i.e. servo + motors 2013-02-05T14:11:44 < dongs> so i have 3 separate domains there that can run at different period 2013-02-05T14:11:48 < dongs> actually 4 separate 2013-02-05T14:12:32 < zyp> yeah, I got that 2013-02-05T14:12:56 < zyp> but it's a bit pointless if you are gonna run four ESCs anyway 2013-02-05T14:13:00 < zyp> like I am 2013-02-05T14:13:08 < dongs> well, you can remap shit 2013-02-05T14:13:18 < zyp> yes, that's what I was asking about 2013-02-05T14:13:23 < zyp> anyway, looks nice 2013-02-05T14:13:30 < zyp> I'll be getting a couple 2013-02-05T14:13:34 < dongs> not like youre gonna use tim15 for other shit anyway 2013-02-05T14:13:56 < dongs> so many timers on F3 its like a party! 2013-02-05T14:14:04 < dongs> what about vbus shit? 2013-02-05T14:14:10 < dongs> should I just stick it on PB5? 2013-02-05T14:14:24 < zyp> sure, nice to have 2013-02-05T14:14:26 < dongs> I really dont like native usb :( 2013-02-05T14:15:04 < zyp> may be useful to know vbus state, but not strictly required 2013-02-05T14:15:35 < dongs> well its unused so yea 2013-02-05T14:15:59 < dongs> and its FT so i can directl stick usb 5V there 2013-02-05T14:16:17 < dongs> wait, how does that work if i wanna power shit by usb only? 2013-02-05T14:16:37 < dongs> i guess dual diode stuff like usual? 2013-02-05T14:16:50 < zyp> I was assuming you had that 2013-02-05T14:16:51 < dongs> yea ok. 2013-02-05T14:16:52 < jpa-> hmm.. are FT pins 5V tolerant when power is off? 2013-02-05T14:17:00 < zyp> no 2013-02-05T14:17:07 < jpa-> though if you power by usb then it is not a problem 2013-02-05T14:17:08 < zyp> I think they are Vcc+4V max 2013-02-05T14:17:15 < jpa-> yeah i remember the same 2013-02-05T14:17:29 < jpa-> somewhat annoying 2013-02-05T14:17:40 < dongs> it is power by usb 2013-02-05T14:17:47 < dongs> well, either or. 2013-02-05T14:17:56 < zyp> yeah shouldn't be a concern then 2013-02-05T14:18:01 < dongs> in my current shit i just have a BAT54C between ext5v and vbus 2013-02-05T14:19:47 < Laurenceb> what are you trying to do? 2013-02-05T14:19:53 * Laurenceb too lazy to scrollback 2013-02-05T14:19:58 < dongs> hack the gibson 2013-02-05T14:20:10 < Laurenceb> the fuck 2013-02-05T14:20:31 < dongs> im sure im forgetting something else 2013-02-05T14:20:31 -!- claude is now known as Claude 2013-02-05T14:20:41 < dongs> wonder if I should put unused pin as another lED 2013-02-05T14:20:43 < dongs> for PWM goodness 2013-02-05T14:21:11 < dongs> hm if it runs at same period as PWM1/2 then thats gonna be shitty 2013-02-05T14:24:08 < zyp> F4 datasheet says the same as F3 datasheet wrt. PC13-15 2013-02-05T14:24:28 < zyp> i.e. no notice about only one output 2013-02-05T14:25:16 < dongs> yea it makes no sense... 2013-02-05T14:25:22 < dongs> which ds are you reading jpa- 2013-02-05T14:25:29 < zyp> no, RM 2013-02-05T14:25:29 < dongs> i dont think it would have such retarded limitation 2013-02-05T14:25:32 < zyp> RM has that notice 2013-02-05T14:25:38 < dongs> oh 2013-02-05T14:25:46 < zyp> I'll check F3 RM 2013-02-05T14:26:11 < dongs> When the backup domain is supplied by VDD (analog switch connected to VDD), the 2013-02-05T14:26:12 < dongs> following functions are available: 2013-02-05T14:26:12 < dongs> . PC13, PC14 and PC15 can be used as GPIO pins 2013-02-05T14:26:14 < dongs> l;ooks ok 2013-02-05T14:26:25 < dongs> only speed 2013-02-05T14:26:28 < dongs> not one at a time 2013-02-05T14:26:50 < dongs> that would majorly fuck with my plans 2013-02-05T14:26:54 < dongs> if only one of those was out 2013-02-05T14:26:58 < dongs> because those pins are WORTHLESS otherwise. 2013-02-05T14:27:04 < zyp> F3 RM does not say that 2013-02-05T14:27:11 < dongs> aND cs is perfect use for them 2013-02-05T14:28:01 < zyp> well, those pins are intended for RTC shit, so not really worthless 2013-02-05T14:28:40 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-05T14:31:12 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-05T14:32:45 < dongs> worthless for my application 2013-02-05T14:36:25 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-05T14:42:07 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-05T14:45:51 < dongs> zyp, what happens if I put dac output through a transistor 2013-02-05T14:46:26 < dongs> like gate or whatever 2013-02-05T14:50:34 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T14:51:05 < dongs> hm nvm im gonna try something else 2013-02-05T14:52:47 -!- barthess [~barthess@86.57.155.106] has quit [Client Quit] 2013-02-05T14:53:43 < dongs> http://i.imgur.com/ix6bYOV.jpg nice pack of C4 2013-02-05T14:56:01 <+Steffanx> Time to remove the +- text and write a new +- on it :P 2013-02-05T14:56:15 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T14:57:19 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T14:57:25 < jon1012> hi 2013-02-05T14:57:33 < dongle> sup 2013-02-05T14:57:39 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T14:57:51 < jon1012> what is the best way to communicate fast and with low cpu between stm32fx (4 for example) and computer using usb ? 2013-02-05T14:58:07 < jon1012> I tried using ft232 chips, but seems unelegant and slow :) 2013-02-05T14:58:25 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T14:58:27 < zyp> umm, native usb? 2013-02-05T14:58:40 < jon1012> yup, but what protocol / driver / ? 2013-02-05T14:58:45 < dongs> anything you want? 2013-02-05T14:58:54 < dongs> just make custom bulk in/out shit 2013-02-05T14:58:56 < jon1012> which one uses less cpu ? 2013-02-05T14:59:00 < jon1012> oh 2013-02-05T14:59:11 < jon1012> but I want to avoid writing kernel driver on computer 2013-02-05T14:59:21 < zyp> you don't have to 2013-02-05T14:59:22 < jon1012> I can usb libusb or serial or whatever 2013-02-05T14:59:28 < zyp> just use libusb or winusb or whatever 2013-02-05T14:59:31 < jon1012> ok 2013-02-05T14:59:34 < dongs> and use cyapi/winusb on the pc side 2013-02-05T14:59:34 < dongs> (assuming windows, or you can go fuck yourself on lunix) 2013-02-05T14:59:34 < dongs> you dont need to write a driver 2013-02-05T14:59:54 < jon1012> ok 2013-02-05T15:00:25 < dongle> what kind of data is it? 2013-02-05T15:00:59 < jon1012> X*Y*3 matrixes 2013-02-05T15:01:09 < jon1012> of one byte 2013-02-05T15:01:14 < jon1012> (rgb pictures) 2013-02-05T15:01:28 < dongle> .. from fsmc or something? 2013-02-05T15:01:45 < jon1012> nope, from computer to mcu 2013-02-05T15:01:49 < dongle> ah 2013-02-05T15:01:54 < dongle> well, then yeah, whatever works 2013-02-05T15:02:47 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T15:02:54 < jon1012> ok 2013-02-05T15:02:57 < jon1012> thanks for the info 2013-02-05T15:03:00 < jon1012> I'll try 2013-02-05T15:03:09 < jon1012> never did raw libusb yet 2013-02-05T15:03:18 < dongle> well, if youre on windows I would recommend against it 2013-02-05T15:03:28 < dongle> since it doesnt actually properly work there. 2013-02-05T15:03:38 < jon1012> I develop on linux but target os may be windows 2013-02-05T15:03:43 < jon1012> depending on cases 2013-02-05T15:03:44 < dongle> too bad. 2013-02-05T15:03:54 < jon1012> right now I use pyserial 2013-02-05T15:04:00 < jon1012> with an ft232 2013-02-05T15:04:01 < zyp> dongle, worked for me when I used it 2013-02-05T15:04:10 < dongle> zyp, it "works" in the same sense that lunix "works" 2013-02-05T15:04:16 < zyp> windows libusb is just a wrapper for winusb anyway 2013-02-05T15:04:25 < dongle> zyp, its not 2013-02-05T15:04:34 < dongle> it installs filter driver over every usb device. 2013-02-05T15:04:37 < dongle> its not a wrapper 2013-02-05T15:04:40 < zyp> huh? 2013-02-05T15:04:43 < dongle> because you can "open" any usb device 2013-02-05T15:04:48 < dongle> with it. 2013-02-05T15:04:57 < dongle> plus, i dont think those fucks got their shit signed yet 2013-02-05T15:05:02 < dongle> which means its fucking useless on any modern pc 2013-02-05T15:05:12 < zyp> oh, by the way, there are several windows libusb variants 2013-02-05T15:05:19 < dongle> i'm sure they're all shit 2013-02-05T15:05:20 < zyp> the one I used was just wrapping winusb 2013-02-05T15:05:25 < dongle> why would I want a opensores wrapper over winusb 2013-02-05T15:05:28 < dongle> when i can just use winusb:? :) 2013-02-05T15:05:44 < zyp> well, I wanted it because I wrote some cross platform shit 2013-02-05T15:06:18 < zyp> but sure, if you only care about windows, no reason to not just use winusb 2013-02-05T15:07:01 < jon1012> ok 2013-02-05T15:07:12 * jon1012 wants cross platform :) 2013-02-05T15:07:25 < dongle> cross platform usually means it doesnt properly work on any platform. 2013-02-05T15:07:29 < dongle> plz prove me wrong 2013-02-05T15:09:09 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T15:09:12 < zyp> this shit worked nicely on both windows and linux without change, but whatever 2013-02-05T15:09:13 <+Steffanx> It seems there is some sort of signed libusb drver stuff dongle .. 2013-02-05T15:09:19 < zyp> arguing with you about that is useless anyway 2013-02-05T15:09:31 <+Steffanx> "The installer needs to run with administrative privileges, so you are strongly encouraged to use a version that has been digitally signed, such as any official version from v2.0.0.151." Version 2.xxx of some driver installation tool 2013-02-05T15:09:34 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T15:09:43 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T15:09:50 < jon1012> btw, what speed can I hope to achieve with libusb direct to an stm32 device ? 2013-02-05T15:09:51 < zyp> Steffanx, what stuff? 2013-02-05T15:10:01 < dongle> jon1012: usb_fs 2013-02-05T15:10:01 <+Steffanx> libusb 2013-02-05T15:10:02 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T15:10:06 < dongle> so <= 12mbit. 2013-02-05T15:10:17 < dongle> well, unless youre using the shit with highspeed phy or wahtever. 2013-02-05T15:10:27 -!- barthess [~barthess@86.57.155.106] has quit [Remote host closed the connection] 2013-02-05T15:11:45 < zyp> FS has 9.7mbit of actual bulk payload bandwidth, under ideal conditions 2013-02-05T15:12:27 < zyp> theoretical max is 19 packets per frame, which makes out to 1216000 bytes of payload per second 2013-02-05T15:12:57 < jpa-> jon1012: usually people have had like 900kB/s for usb mass storage on stm32 2013-02-05T15:13:07 < jon1012> ugh ok 2013-02-05T15:13:12 < dongle> 'ugh'? 2013-02-05T15:13:15 < dongle> did you expect better? 2013-02-05T15:13:18 < jon1012> that's low :( 2013-02-05T15:13:18 < jpa-> so i expect somewhere between 0.9 and 1.2 MB/s 2013-02-05T15:13:19 < jon1012> yeah 2013-02-05T15:13:20 < dongle> do you really need a F4 in there? 2013-02-05T15:13:37 < dongle> use ez-usb fx2 if you want to pump data to/from PC at nearly 400mbit 2013-02-05T15:13:40 < jon1012> other people use fpga, but I prefer a F4 :) 2013-02-05T15:13:42 < zyp> jon1012, what's your application? 2013-02-05T15:14:07 < dongle> seems hes sending video frames or osmething 2013-02-05T15:14:08 < zyp> if you want higher speeds than a megabyte per second, you need highspeed 2013-02-05T15:14:09 < dongle> or picture frames 2013-02-05T15:14:19 < jon1012> yup video frames 2013-02-05T15:14:25 < zyp> F4 with external phy would do it 2013-02-05T15:14:37 < jpa-> what does the F4 do with the video frames? 2013-02-05T15:14:39 < dongle> yeah but it wont have nough cpu/memory bandwidth to deal with it 2013-02-05T15:14:44 < jon1012> jpa-, display them 2013-02-05T15:14:49 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T15:14:52 < jon1012> do the scanning, pwm and all that 2013-02-05T15:15:17 < jpa-> so how much bandwidth do you need? (how large are your frames?) 2013-02-05T15:15:17 <+Steffanx> The two cameras i've worked with bot had a fx2 inside :) 2013-02-05T15:15:21 <+Steffanx> *both 2013-02-05T15:15:43 < jon1012> jpa-, potentially 512x512x3 byte, 30 to 50 times per second 2013-02-05T15:16:07 < jpa-> ok, so you definitiely need usb high speed 2013-02-05T15:16:20 < dongs> Steffanx: thats because thats how its done usuaslly 2013-02-05T15:16:28 <+Steffanx> Probably 2013-02-05T15:16:44 < jpa-> F4 + external phy or FX2 + perhaps some other stuff 2013-02-05T15:17:04 < jpa-> jon1012: how are you going to scan 512*512*3 display at 50 Hz? 2013-02-05T15:17:19 < jon1012> the scanning is faster and works 2013-02-05T15:17:25 < jon1012> right now 2013-02-05T15:17:26 <+Steffanx> FX2 + crystal => image chip 2013-02-05T15:17:56 < jpa-> jon1012: dma to gpio? 2013-02-05T15:18:08 < jon1012> just raw bitbanging 2013-02-05T15:18:12 < jon1012> jpa-, shooting the data at external shift regs, then go to next line and so one 2013-02-05T15:18:14 < dongle> i think he was doing the other way around.... 2013-02-05T15:18:17 -!- barthess [~barthess@86.57.155.106] has quit [Client Quit] 2013-02-05T15:18:26 < dongle> images from PC to display w/f4 2013-02-05T15:18:28 < jon1012> so on* 2013-02-05T15:19:14 <+Steffanx> You didn't even use the built-in stuff to read out the data jon1012 ? 2013-02-05T15:19:20 < jpa-> jon1012: and this is running at 512x512? 2013-02-05T15:19:34 < jon1012> Steffanx, no 2013-02-05T15:19:36 <+Steffanx> dmci? dcmi whatever it's called 2013-02-05T15:19:48 < jon1012> jpa-, 128x128 only now, but I'll try to go 512 2013-02-05T15:19:49 < zyp> Steffanx, that's input only 2013-02-05T15:20:06 < jon1012> Steffanx, what do you mean by built in stuff ? 2013-02-05T15:20:15 <+Steffanx> Nevermind 2013-02-05T15:20:26 < zyp> jon1012, camera interface, but it's input only 2013-02-05T15:20:33 < jon1012> oh ok 2013-02-05T15:20:40 < jpa-> jon1012: i bet that bitbanging is not going to work fast enough for 512*512, but i guess you could manage with some clever DMA stuff 2013-02-05T15:20:58 < jon1012> really ? that would be of interest to me :) 2013-02-05T15:21:08 < jon1012> I'll try to see how dma works, never used it 2013-02-05T15:21:16 < dongle> jon1012: youre sending images FROM F4 to PC? or PC to F4 2013-02-05T15:21:22 < jon1012> pc to f4 2013-02-05T15:21:31 < dongle> right, so DCMI is useless in this case 2013-02-05T15:21:35 < dongle> as its input only 2013-02-05T15:21:54 < jpa-> Steffanx has been smoking too much pot 2013-02-05T15:21:59 <+Steffanx> blablabla 2013-02-05T15:22:02 < dongle> i suggest you follow what "others" do in this situation, that is fx2+fpga 2013-02-05T15:22:04 < zyp> I think this would better be done with FX2 and FPGA 2013-02-05T15:22:08 < dongle> it would be a LOT easier to setup. 2013-02-05T15:22:16 < jpa-> i vote FX2 and FPGA also 2013-02-05T15:22:27 < jon1012> mmh, but f4 seemed easier :) 2013-02-05T15:22:34 < dongle> by far not easier 2013-02-05T15:22:36 < jon1012> and I don't have fx2 nor fpga dev boards 2013-02-05T15:22:44 < zyp> it's easier as long as it's fast enough 2013-02-05T15:22:48 <+Steffanx> Hobby project jon1012 ? 2013-02-05T15:22:55 < jon1012> nope, not a hobby project :) 2013-02-05T15:23:01 < jpa-> jon1012: your F4 dev board doesn't have USB high-speed either 2013-02-05T15:23:08 < jon1012> but a tight budget small company :) 2013-02-05T15:23:13 < jon1012> yeah true 2013-02-05T15:23:35 < dongle> i'd imagine driving rgb lcd or whaetver with fpga is trivial 2013-02-05T15:23:40 < dongle> and on F X2 side its just a 8/16bit fifo 2013-02-05T15:23:47 <+Steffanx> ( jpa- i don't smoke and NEVER will. Just like your said you dont drink ) 2013-02-05T15:24:03 < jpa-> Steffanx: nuh-uh, everyone in .nl smokes pot 2013-02-05T15:24:06 < zyp> I had some similar ideas when I was playing with my led matrix 2013-02-05T15:24:09 < dongle> lol stoners 2013-02-05T15:24:20 <+Steffanx> ( jpa- has never been in dutchland ) 2013-02-05T15:24:21 < jon1012> by the way, I was thinking of ft232h cpu-style FIFO for higher speed, do you think it would work with stm32f4 ? 2013-02-05T15:24:30 < jpa-> (i actually have, there were cows) 2013-02-05T15:24:40 <+Steffanx> There are cows in finland too 2013-02-05T15:24:42 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Read error: Connection reset by peer] 2013-02-05T15:24:46 < jpa-> wow 2013-02-05T15:24:56 < dongle> jon1012: you've been suggested the proper solution. now you can either nigger it up or follow the advice. 2013-02-05T15:25:02 < zyp> http://bin.jvnv.net/f/ldMgR.JPG <- but this is way smaller than 512x512, and monochrome :) 2013-02-05T15:25:09 < jon1012> dongle, ok I'll try that :) 2013-02-05T15:25:12 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-05T15:25:16 < jpa-> jon1012: if you don't compress the data, nothing USB FS will ever give you the bandwidth required 2013-02-05T15:25:23 < jon1012> ok 2013-02-05T15:25:32 < jon1012> ft232h is high speed 2013-02-05T15:25:37 <+Steffanx> JPEG encoding/decoding it is :) 2013-02-05T15:25:46 < jpa-> jon1012: ok, then there is some chance 2013-02-05T15:26:05 < jpa-> jon1012: you might even be able to plug it directly to your shift registers maybe? 2013-02-05T15:26:05 < dongle> USB to synchronous 245 parallel FIFO mode for transfers up to 40 Mbytes/Sec 2013-02-05T15:26:05 < zyp> I'd rather go with ulpi phy than ft323h 2013-02-05T15:26:06 < dongle> neat 2013-02-05T15:26:23 < dongle> so use that instead of FX2 and fpga. 2013-02-05T15:26:31 < dongle> but still F4 doesnt belong in that picture at all 2013-02-05T15:26:43 < zyp> ft232h* 2013-02-05T15:26:49 < jon1012> f4 is fast enough for the bit bang to my led display :) 2013-02-05T15:26:53 < jon1012> I tested it and it works 2013-02-05T15:27:10 < jpa-> doesn't help if you don't get the data to F4 2013-02-05T15:27:12 < jon1012> that's why I used it, that and the fact I never worked with fpga 2013-02-05T15:27:15 < jon1012> yup 2013-02-05T15:27:38 < Laurenceb> fpga is generally expensive and power hungry 2013-02-05T15:27:52 < Laurenceb> but F4 isnt really a multimedia provessor 2013-02-05T15:29:02 <+Steffanx> It doesn't really have to process media. 2013-02-05T15:29:09 <+Steffanx> *multimedia 2013-02-05T15:29:14 < dongle> why cant you use FSMC to drive lcd? 2013-02-05T15:29:17 < zyp> power hungry is a silly statement, I bet it would use less power than a software solution in this case 2013-02-05T15:29:55 <+Steffanx> It just has to process some huge stream of 'random' bytes Laurenceb :) 2013-02-05T15:30:06 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-05T15:30:16 < Laurenceb> pron bytes 2013-02-05T15:32:02 < dongle> http://www.raspberrypi.org/phpBB3/viewtopic.php?f=31&t=30882 2013-02-05T15:32:05 < dongle> attn all 2013-02-05T15:32:26 < jon1012> gargl i2c php 2013-02-05T15:32:28 <+Steffanx> ih what 2013-02-05T15:32:40 <+Steffanx> php... so what? 2013-02-05T15:34:07 <+Steffanx> or isn't that your point dongle? 2013-02-05T15:34:16 <+Steffanx> You would've used C# on IIS for sure 2013-02-05T15:34:33 <+Steffanx> man, why i feed him 2013-02-05T15:36:39 < dongle> OK so last unused pin 2013-02-05T15:36:44 < dongle> i guess ill do a nfet to buzzer or something 2013-02-05T15:42:15 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-05T15:44:41 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Disconnected by services] 2013-02-05T15:47:15 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-05T15:49:12 < Laurenceb> http://imgur.com/gallery/wb2Vh 2013-02-05T15:51:32 < Laurenceb> http://static.quickmeme.com/media/social/qm.gif 2013-02-05T15:51:34 < Laurenceb> lolz 2013-02-05T15:51:39 < dongle> nice fail 2013-02-05T15:51:48 < dongle> 1x1 pixel GIF 2013-02-05T15:51:58 < jon1012> (example of one of my display working : https://www.youtube.com/watch?v=e6mKPLnY854 ) 2013-02-05T15:52:01 < Laurenceb> c&p fail 2013-02-05T15:52:13 < jon1012> (the reason why I need fast usb) 2013-02-05T15:52:18 < dongle> oh its LED 2013-02-05T15:52:23 < Laurenceb> how do you drive that? 2013-02-05T15:52:31 < jon1012> shift registers 2013-02-05T15:52:31 < karlp> what's wrong with a tv? 2013-02-05T15:52:38 < Laurenceb> lolowned 2013-02-05T15:52:59 < jon1012> karlp, you can't burn eyes with a tv :p 2013-02-05T15:53:12 < dongle> jon1012: ive got something fo r you 2013-02-05T15:53:13 < karlp> that display didn't like any eyes were being burnt... 2013-02-05T15:53:47 < dongle> jon1012: http://i.imgur.com/epltwLl.jpg 2013-02-05T15:54:01 < jon1012> wow nice 2013-02-05T15:54:11 < dongle> getting vid, 1 sec 2013-02-05T15:54:32 < karlp> dongs, that must be fail though, that wasn't fx2 + fpgs wasn it? ;) 2013-02-05T15:54:38 < dongle> karlp: actually it was 2013-02-05T15:54:41 < dongle> its exactly waht itis :) 2013-02-05T15:56:08 < dongle> jon1012: http://www.youtube.com/watch?v=Ihdpn6MdAeM 2013-02-05T15:56:39 < dongle> and this can EASILY do huge matrixes 2013-02-05T15:56:41 < dongle> at 30fps+ 2013-02-05T15:57:03 < dongle> my ledpal is doin that shit 2013-02-05T15:57:10 < jon1012> nice ! 2013-02-05T15:57:22 < jon1012> so that's using fx2 and an fpga ? 2013-02-05T15:57:26 < dongle> yep 2013-02-05T15:57:55 < Laurenceb> how do the modules talk? 2013-02-05T15:58:09 < zyp> they shout. 2013-02-05T15:58:15 < dongle> i think up to 4 chained per channel, and any number of channels 2013-02-05T15:58:25 < Laurenceb> parrellel interface? 2013-02-05T15:58:28 < dongle> serial 2013-02-05T15:58:43 < Laurenceb> nice 2013-02-05T15:59:52 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-05T16:00:13 < jon1012> serial one per color ? 2013-02-05T16:00:21 < dongle> its 24bit rgb 2013-02-05T16:00:23 < dongle> per pixel 2013-02-05T16:00:34 < dongle> shit's just playing useless demo so it doesnt really show what it can do 2013-02-05T16:00:37 < jon1012> oh so your display handle the scanning themselves ? 2013-02-05T16:00:57 < karlp> dongle: oh, I was mixing that led grid up with one of your other led projects, sorry 2013-02-05T16:01:09 < dongle> karlp: yea, that one was f103 2013-02-05T16:01:25 < jon1012> mine are just dumb shift register with dumb chan p mosfets 2013-02-05T16:01:34 < dongle> jon1012: http://i.imgur.com/qls4B.jpg display 2013-02-05T16:01:34 < jon1012> and all the intelligence is in the controller 2013-02-05T16:01:45 < jon1012> ok 2013-02-05T16:02:03 < jon1012> really really nice 2013-02-05T16:03:33 < dongle> anyway, with fpga it doesnt really mater waht your target dispaly is 2013-02-05T16:03:51 < dongle> the chinese high pitch led panels are also just shift registers 2013-02-05T16:03:53 -!- DaKu is now known as daku 2013-02-05T16:03:54 < dongle> and like 1/8 scanning 2013-02-05T16:04:01 < dongle> and i think you can chain 8x modules per line in those 2013-02-05T16:04:21 < dongle> and you ahve 16 pixels (2 scans) per line 2013-02-05T16:04:28 < dongle> and clock out pixel data or wahtever 2013-02-05T16:05:32 < jon1012> yeah 2013-02-05T16:06:02 < jon1012> dongle, funny, I'm making 16x16 flexible pcb displays for a client right now 2013-02-05T16:06:10 < dongle> :) 2013-02-05T16:06:22 < jon1012> dongle, but using the same tech than for my rigid ones 2013-02-05T16:06:27 < dongle> i bet they're not ~60/ea or wahtever these run for me 2013-02-05T16:06:27 < jon1012> just dumb shift regs 2013-02-05T16:06:32 < jon1012> yeah 2013-02-05T16:06:53 < jon1012> and I'm using 3528 leds not 5050 2013-02-05T16:06:57 < dongle> right 2013-02-05T16:07:06 < dongle> high pitch stuff is fin 2013-02-05T16:07:07 < dongle> fun 2013-02-05T16:07:28 < jon1012> I tend to stick at 6mm pitch 2013-02-05T16:07:35 < jon1012> (some clients ask for 3 some for 25) 2013-02-05T16:08:03 < dongle> how wmuch does your finepitch flexpcb run per d isplay? 2013-02-05T16:08:22 < jon1012> let me check 2013-02-05T16:08:26 < dongle> per panel that is 2013-02-05T16:08:31 < dongle> you can /msg if NDA 2013-02-05T16:09:04 < dongle> cause we're looking for some highpitch shit, though lameway driving is a bit of a problme :) 2013-02-05T16:14:29 < dongle> bloggery 2013-02-05T16:16:52 < gxti> bliggty blag 2013-02-05T16:23:05 < Laurenceb> http://laughingsquid.com/wp-content/uploads/first_goatse_apple_2.jpg 2013-02-05T16:25:07 < dongs> nice 2013-02-05T16:27:14 -!- daku is now known as DaKu 2013-02-05T16:27:36 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has quit [Quit: Ex-Chat] 2013-02-05T16:29:16 <+Steffanx> You like photoshop don't you dongs? 2013-02-05T16:37:28 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-05T16:37:54 < dongs> yaman 2013-02-05T16:38:55 <+Steffanx> k 2013-02-05T16:41:06 < dongs> hmm, http://bcas.tv/paste/results/OuUbOb38.html how can i do something like this but with only one input capture? 2013-02-05T16:41:10 < dongs> capture on both edges? 2013-02-05T16:41:44 < zyp> this is pwm capture? 2013-02-05T16:41:49 < dongs> hm wait 2013-02-05T16:41:52 < dongs> yeah 2013-02-05T16:41:53 < Laurenceb> hi emeb_mac 2013-02-05T16:41:59 < emeb_mac> Laurenceb: heya 2013-02-05T16:42:01 < Laurenceb> i was thinking about SDR 2013-02-05T16:42:02 < zyp> each channel can only capture one of the edges 2013-02-05T16:42:20 < zyp> so you need to use two channels per signal if you need to capture both edges 2013-02-05T16:42:20 < Laurenceb> i think you need to think more about the IIR filter 2013-02-05T16:42:27 < dongs> maybe it doesnt matter, since TIM2_CH4 is unused anyway 2013-02-05T16:42:28 < emeb_mac> Oh? 2013-02-05T16:42:33 < Laurenceb> i think the key to making it work is IIR on the complex data 2013-02-05T16:42:37 < dongs> i was thinking of using spektrum/sbus pad to do frsky telemetry 2013-02-05T16:42:46 < Laurenceb> http://www.mathworks.co.uk/help/dsp/examples/multistage-design-of-decimators-interpolators.html 2013-02-05T16:42:47 < Laurenceb> this 2013-02-05T16:43:04 < gxti> dongs: it would have to be ch2, they're paired 1/2 and 3/4 2013-02-05T16:43:13 < emeb_mac> Laurenceb: If you look at the matlab sim I did there's a complex IIR after the decimator & before the AM detector 2013-02-05T16:43:15 < dongs> from lipo/current sensor 2013-02-05T16:43:15 < dongs> it sends shit as pulse-coded pwm 2013-02-05T16:43:15 < dongs> like pulse-width 2013-02-05T16:43:22 < gxti> dongs: internally though, it's not like the pin has to be connected 2013-02-05T16:43:27 < dongs> does 'slavemode_reset' apply to entire timer? 2013-02-05T16:43:27 < Laurenceb> i see 2013-02-05T16:43:31 < zyp> yes 2013-02-05T16:43:34 < emeb_mac> it is importatnt 2013-02-05T16:43:36 < dongs> i guess it does hm, not just channel? 2013-02-05T16:43:38 < zyp> reset resets the entire counter 2013-02-05T16:43:38 < Laurenceb> yes 2013-02-05T16:44:35 < dongle> hmm 2013-02-05T16:44:39 < emeb_mac> the windowed integration before decimation has BW larger than one "bin" 2013-02-05T16:44:42 < dongle> < gxti> dongs: it would have to be ch2, they're paired 1/2 and 3/4 2013-02-05T16:44:50 < dongle> tell me more 2013-02-05T16:44:57 < dongle> i have TIM2_CH4 available 2013-02-05T16:44:59 < emeb_mac> so oversampling is needed to avoid aliasing within the bin 2013-02-05T16:45:01 < Laurenceb> emeb_mac: yeah 2013-02-05T16:45:04 < dongle> CH3 not connected 2013-02-05T16:45:10 < dongle> but if reset resets whole thing thats not gonna work 2013-02-05T16:45:14 < zyp> dongle, ch4 input can be routed to ch3 and ch4 internally 2013-02-05T16:45:21 < Laurenceb> im sure theres a clever way to do this well 2013-02-05T16:45:23 < dongle> seriously? 2013-02-05T16:45:26 < emeb_mac> and complex IIR is needed to at the OS rate to narrow up the response. 2013-02-05T16:45:26 < gxti> yes 2013-02-05T16:45:29 < Laurenceb> but im not very experience with dsp... 2013-02-05T16:45:37 < zyp> you can use one to capture rising and one to capture falling, and then just subtract to find difference 2013-02-05T16:45:39 < gxti> ti3 and ti4 can be fed from ch3 or ch4 2013-02-05T16:45:46 < dongle> hmm 2013-02-05T16:45:46 < gxti> ti1 and ti2 can be fed from ch1 or ch2 2013-02-05T16:45:57 < dongle> so off same pin you mean? 2013-02-05T16:45:59 < gxti> so you can have ti3 capture ch4 rising, and ti4 capture ch4 falling 2013-02-05T16:46:02 < dongle> problem is i cant autoreset 2013-02-05T16:46:08 < dongle> that timer 2013-02-05T16:46:13 < gxti> right, you can just not do the reset part 2013-02-05T16:46:16 < dongle> okay 2013-02-05T16:46:21 < zyp> if you just capture both and calculate difference, you'll be fine 2013-02-05T16:46:25 < dongle> ok 2013-02-05T16:46:30 < emeb_mac> Laurenceb: ideally one could use a CIC / Hogenauer filter at the high rate. 2013-02-05T16:46:33 < zyp> a bit more work than resetting, but still possible 2013-02-05T16:46:43 < emeb_mac> but again, not enough cycles :( 2013-02-05T16:46:49 < dongle> would pullin that line up with 10K to vcc 2013-02-05T16:46:51 < dongle> affect anything? 2013-02-05T16:46:56 < dongle> when using it as uart 2013-02-05T16:47:06 < dongle> cuz frsky outputs are opendrain or someshit 2013-02-05T16:47:08 < zyp> no 2013-02-05T16:47:24 < gxti> pullup on uart is normal since they idle high 2013-02-05T16:47:27 < zyp> but why not just use internal pullup? 2013-02-05T16:47:34 < dongle> hmm.. 2013-02-05T16:47:37 < dongle> duno if strong enough 2013-02-05T16:47:57 < zyp> probably fine with external anyway 2013-02-05T16:48:42 < dongle> rigfht, i put 10k pullups on both uart lines before and used htem as i2c 2013-02-05T16:48:52 < dongle> i mean, switching either uart or i2c 2013-02-05T16:48:53 < dongle> and that worked 2013-02-05T16:49:26 < dongle> neat, so the useless spektrum/shitbus pad can also be moderately useful w/frsky input 2013-02-05T16:51:47 < dongle> http://i.imgur.com/mVSxTi4.png like this. 2013-02-05T16:54:26 < gxti> i wish there were a timer on uart4_rx, i want to do baud rate detection and now i have to go hardmode :< 2013-02-05T16:55:05 < dongs> doesnt stm32 bootloader autobaud 2013-02-05T16:56:36 < gxti> maybe. i can't use implementations that require a 0x55 sync or whatever though. 2013-02-05T16:58:17 < gxti> since i also have to deal with gps that are mute after powerup anyway, i might end up not doing baud rate detection so much as cycling through rates every 5s and attempting to query it. 2013-02-05T16:58:30 < dongle> gxti: the latter is waht I use 2013-02-05T16:58:34 < dongle> but dont do that 2013-02-05T16:58:39 < dongle> are you working only wiht specific gps? 2013-02-05T16:58:42 < dongle> or any random ones 2013-02-05T16:59:00 < gxti> random ones, although from a smallish list 2013-02-05T16:59:01 < gxti> brb 2013-02-05T16:59:02 < dongle> what I do is cycle through common baudrates with a command that sets GPS to a specific rate I want 2013-02-05T16:59:10 < dongle> then switch to that baudrate :) 2013-02-05T16:59:28 < dongle> but that only works if your hardware is fixed. 2013-02-05T17:03:34 < zyp> doesn't work if your hardware is broken 2013-02-05T17:06:27 < gxti> i don't much care what baudrate it runs at, either way i have to discover it 2013-02-05T17:11:28 < dongle> i duno 2013-02-05T17:11:29 < dongle> if its gps 2013-02-05T17:11:34 < dongle> you can just cycle and wait for non-junk data. 2013-02-05T17:13:28 < gxti> some of the gps don't talk though 2013-02-05T17:13:36 < gxti> that's the annoying part 2013-02-05T17:13:41 < dongle> why? 2013-02-05T17:13:44 < dongle> until you send htem some shit? 2013-02-05T17:13:46 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-05T17:13:46 < dongle> stop supporting shitty stuff 2013-02-05T17:13:51 < gxti> i might 2013-02-05T17:13:55 < dongle> just support ublox and tell teh rest to fuck off 2013-02-05T17:14:17 < zyp> then drop NMEA support and only speak UBX? :p 2013-02-05T17:14:24 < gxti> zyp: irrelevant 2013-02-05T17:14:34 < gxti> i don't care *what* they talk, as long as they talk 2013-02-05T17:14:38 < zyp> I was commenting to dongs 2013-02-05T17:14:42 < gxti> k 2013-02-05T17:14:54 < dongle> yeah thats next wtep 2013-02-05T17:15:09 < gxti> dongle: where do i buy ublox in small quantities without overpaying? 2013-02-05T17:15:17 < gxti> not that i can do that with non-ublox either, but you might know :p 2013-02-05T17:15:37 < gxti> i'd really love to have a source for NEO-6T or LEA-6T but those are hard to find :( 2013-02-05T17:15:45 < dongle> gxti: neo6 modules are mega cheap, like $15 or so 2013-02-05T17:15:52 < dongle> ior you can jsut buy premade shit like http://www.rctimer.com/index.php?gOo=goods_details.dwt&goodsid=763 2013-02-05T17:15:53 < gxti> don't actually need the T for this particular project though 2013-02-05T17:15:55 < dongle> with antenna+etc 2013-02-05T17:16:04 < dongle> yeah T are expensive 2013-02-05T17:16:11 < gxti> as long as the non-T doesn't have shitty PPS 2013-02-05T17:16:13 < dongle> i think my T was 30-40ea 2013-02-05T17:16:17 < gxti> that jumps a microsecond whenever it feels like 2013-02-05T17:16:55 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-05T17:17:41 < gxti> hmm, not bad. what about just the module? 2013-02-05T17:17:54 < dongle> which one? neo6 you mean? 2013-02-05T17:17:59 < gxti> yeah 2013-02-05T17:19:16 < dongle> um im prety sure i remember seeing them for less than 20bux somewehre 2013-02-05T17:19:24 < dongle> or you can laways email china and ask to sell you some 2013-02-05T17:19:30 < dongle> they'll be glad to mark up $0.01 2013-02-05T17:19:35 < dongle> because that will feed their family for a week 2013-02-05T17:19:36 < gxti> i saw $18 on fleabay but it's probably got cancer 2013-02-05T17:20:24 < dongle> shittona of chinks selling NEO6 for <$15 on aliexpress/etc 2013-02-05T17:20:39 < gxti> definitely cancer. but probably worth a shot anyway. 2013-02-05T17:20:47 < gxti> as long as it's not socks 2013-02-05T17:20:51 < dongle> the onyl thing you'd get is neo6-0-000 2013-02-05T17:21:02 < dongle> just old firmware is all 2013-02-05T17:21:27 < gxti> upgradable? 2013-02-05T17:21:31 < dirty_d> zyp, seems libusb is getting a STALL when it writes 2013-02-05T17:21:43 < dirty_d> but i dont see how thats possible 2013-02-05T17:21:58 < dongle> i dunno what the deal with neo stuff, I think they're all mask rom basd 2013-02-05T17:21:59 < dongle> based 2013-02-05T17:22:13 < gxti> o rite, some are flash based but not the shitties from china 2013-02-05T17:22:17 < zyp> dongle, I heard the gopro you got were foot mountable, I guess they heard people in japan love upskirt vids 2013-02-05T17:22:28 < dongle> haha 2013-02-05T17:22:41 < dongle> gxti: its not the shitties/etc thing, neo stuff is value/checap trash 2013-02-05T17:22:45 < dongle> so tehres no reason to make it flash 2013-02-05T17:23:27 < dongle> http://www.ebay.com/itm/250866889973 you can always stock up on this shit :) 2013-02-05T17:23:48 < dongle> the pads on bottom are uart/USB/etc 2013-02-05T17:23:52 < gxti> haha, forgot about those. i might do it. 2013-02-05T17:24:20 < dongle> pretty slick 4layer board 2013-02-05T17:25:19 < gxti> i presume the pci-e is usb? 2013-02-05T17:25:26 < dongle> most likely 2013-02-05T17:25:37 < gxti> at that price it'd have to be, but i wonder what the chips are for then 2013-02-05T17:25:39 < zyp> it's standard 2013-02-05T17:25:48 < dongle> minipcie has usb lane in it 2013-02-05T17:25:51 < dongle> just like expresscard 2013-02-05T17:25:53 < zyp> minipcie form factor include s one pcie lane and usb 2013-02-05T17:25:55 < zyp> yep 2013-02-05T17:26:20 < zyp> and ublox 6 already have usb, so it's just a direct hookup 2013-02-05T17:26:22 < gxti> yes and it's annoying because 90% of embedded mobos with mini-pcie don't actually have the pci-e lane 2013-02-05T17:26:30 < dongle> bullshit 2013-02-05T17:26:34 < dongle> stop buying garbage 2013-02-05T17:26:41 < gxti> i haven't bought any actually 2013-02-05T17:26:45 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-05T17:26:49 < zyp> depends on what you're looking at 2013-02-05T17:26:59 < gxti> the cheap ones that don't actually have pci-e :p 2013-02-05T17:27:12 < gxti> this was a while ago though 2013-02-05T17:27:42 < zyp> minipcie 3G cards are commonly usb only, and the embedded boards with usb-only sockets I've seen have had the socket intended for 3G card 2013-02-05T17:28:05 < gxti> instead of doing that i ended up designing something from scratch with stm32f107 and here i am looking for gps modules to oem 2013-02-05T17:29:16 < dongle> 100% brand new and high quality 2013-02-05T17:29:28 < dongle> i like the chinkese "100% new and original" 2013-02-05T17:29:33 < dongle> saying, when yo ubuy ICs and shit from them 2013-02-05T17:29:55 < dongle> my chink supplier guy actually gives me a choice,, original or chinaclone 2013-02-05T17:29:58 < dongle> for whatever stuff. 2013-02-05T17:30:00 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] 2013-02-05T17:30:03 < dongle> so for shit i dont care about I usually go for chianclone 2013-02-05T17:30:10 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has joined ##stm32 2013-02-05T17:40:51 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-221.pptp.stw-bonn.de] has joined ##stm32 2013-02-05T17:41:42 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-05T17:43:11 < Laurenceb> ok this is weird 2013-02-05T17:43:17 < Laurenceb> ive killed the terminal on buntu 2013-02-05T17:44:16 < emeb> midas touch 2013-02-05T17:44:49 < Laurenceb> i was pissing about with usb on F4 2013-02-05T17:44:53 < Laurenceb> and killed it 2013-02-05T17:45:26 < Laurenceb> i dont see how this is possible 2013-02-05T17:45:39 < Laurenceb> all terminal sessions are separate threads right? 2013-02-05T17:45:53 < karlp> until the usb kernel driver explodes and blows up shit 2013-02-05T17:46:06 < karlp> I've had bad usb drivers take down the system before. 2013-02-05T17:47:31 < Laurenceb> hmf 2013-02-05T17:47:40 < Laurenceb> everything seems fine... 2013-02-05T17:48:04 < emeb> didn't even need a reboot? 2013-02-05T17:48:30 < Laurenceb> i have unsaved work 2013-02-05T17:50:56 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has quit [Read error: Connection reset by peer] 2013-02-05T17:51:12 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has joined ##stm32 2013-02-05T17:55:54 < dirty_d> man, this is crazy 2013-02-05T17:55:57 < dirty_d> makes 0 sense 2013-02-05T18:02:14 < zyp> hmm, now I'm confused 2013-02-05T18:02:30 < emeb> me too. but why? 2013-02-05T18:02:31 < dirty_d> why 2013-02-05T18:02:52 < gxti> baffling 2013-02-05T18:02:54 < zyp> found a package notice in my mail when I got home, was thinking «oh, nice, must be ipad lcd» 2013-02-05T18:03:11 < zyp> then I drive down to post office and pick up a small envelope 2013-02-05T18:03:21 < zyp> and I have no idea what it is 2013-02-05T18:03:31 < gxti> better blow it up, for science 2013-02-05T18:04:05 < emeb> beware of greeks bearing gifts 2013-02-05T18:04:10 < zyp> maybe it's supposed to be my gopro 2013-02-05T18:04:16 < emeb> socks! 2013-02-05T18:04:24 < zyp> but it's so small 2013-02-05T18:04:44 < emeb> socks are highly compressible 2013-02-05T18:05:12 < zyp> it's some small rigid shit inside 2013-02-05T18:05:19 < zyp> oh well, opening it 2013-02-05T18:05:31 < emeb> maybe the "gopro" folks send out random stuff. 2013-02-05T18:05:41 < zyp> wat 2013-02-05T18:05:43 < BrainDamage> gopro plushie 2013-02-05T18:05:51 < zyp> it's a microsd 2013-02-05T18:06:05 < zyp> 2gb microsd, with adapter to normal sd, in box 2013-02-05T18:06:13 < emeb> whatever you do DON'T PUT IT IN YOUR COMPUTER! 2013-02-05T18:06:26 < gxti> not a windows one at least 2013-02-05T18:06:45 < BrainDamage> emeb: what about putting it physically, but not electrically? 2013-02-05T18:06:50 < emeb> it'll give your computer the diabetes. 2013-02-05T18:06:57 < BrainDamage> like use the computer case as container 2013-02-05T18:07:08 < karlp> zyp: my friend's gopro was a microsd card, 2013-02-05T18:07:10 < karlp> not socks. 2013-02-05T18:07:14 < emeb> yeah - open the case & toss it in. 2013-02-05T18:07:19 < zyp> ah, then it's probably that 2013-02-05T18:07:28 < zyp> because I haven't ordered any sd 2013-02-05T18:07:31 < emeb> those clever ebay trolls 2013-02-05T18:07:46 < karlp> sd cards for norwegians and aussies, socks for japs 2013-02-05T18:08:02 < emeb> that's some kind of code 2013-02-05T18:08:58 < zyp> shows up as empty 2GB sd card 2013-02-05T18:09:05 < zyp> wonder if actual capacity is 2GB 2013-02-05T18:09:09 < emeb> OMG - You actually read it! 2013-02-05T18:09:37 < emeb> we're all doomed. 2013-02-05T18:09:43 < zyp> will be fun testing this with my SD code :p 2013-02-05T18:09:54 < dirty_d> this is pure madness 2013-02-05T18:10:01 < emeb> zyp makes lemonade... 2013-02-05T18:11:20 < dirty_d> zyp, i should not get CTR_RX and data and count if ACK was not sent to the host right? 2013-02-05T18:11:44 < dirty_d> my pc and the stm32 seem to be disagreeing on what the actual $*&% is going on 2013-02-05T18:12:33 < zyp> dirty_d, right 2013-02-05T18:13:02 < dirty_d> im getting "detected endpoint stall" from libusb 2013-02-05T18:13:11 < dirty_d> and ive tried libusb, and libusbx 2013-02-05T18:13:16 < dirty_d> so it has to be the stm32 i guess 2013-02-05T18:13:41 < dirty_d> LIBUSB_ERROR_PIPE (the endpoint halted) 2013-02-05T18:14:11 < dirty_d> i see "Hello!" with my eyes, but libusb says 0 bytes were transfered 2013-02-05T18:14:37 < dirty_d> is it possible that it sent ACK, but it wasnt recieved? 2013-02-05T18:14:52 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Read error: Connection reset by peer] 2013-02-05T18:14:58 < zyp> refresh my memory, are you using f3 discovery? 2013-02-05T18:15:55 < dirty_d> yea 2013-02-05T18:16:31 < zyp> ok, I can test it here with analyzer 2013-02-05T18:16:32 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32 2013-02-05T18:16:35 < Laurenceb> RAGE 2013-02-05T18:16:39 < zyp> no. 2013-02-05T18:16:40 < Laurenceb> whole system went down 2013-02-05T18:16:49 < zyp> NO RAGING 2013-02-05T18:16:49 < emeb> with unsaved work. 2013-02-05T18:16:59 < Laurenceb> not too much unsaved it seems 2013-02-05T18:17:18 < emeb> instead of RAGE, think of kittens and unicorns and rainbows. 2013-02-05T18:19:33 -!- alexn [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-05T18:21:09 < BrainDamage> won't work for him 2013-02-05T18:21:17 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer] 2013-02-05T18:21:27 < BrainDamage> he'd have to use lolis, animu, and kawaii 2013-02-05T18:21:41 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-05T18:21:41 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has quit [Changing host] 2013-02-05T18:21:41 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-05T18:21:54 -!- barthess [~barthess@178.154.1.114] has joined ##stm32 2013-02-05T18:21:55 < emeb> whatever floats one's boat I guess. As long as you keep it to yourself... 2013-02-05T18:22:44 < BrainDamage> you know he won't 2013-02-05T18:25:24 < dirty_d> zyp, http://cisweb.bristolcc.edu/~adowning/usb.tar.bz2 2013-02-05T18:25:49 < dirty_d> i have my main.elf for the stm32 in there, and the libusb test program is in usb_test 2013-02-05T18:26:32 < dirty_d> ps, you are the man 2013-02-05T18:26:33 < dirty_d> lol 2013-02-05T18:27:46 < Laurenceb> ok wtf 2013-02-05T18:27:53 < Laurenceb> bt 2013-02-05T18:27:54 < Laurenceb> #0 0xa0000000 in ?? () 2013-02-05T18:28:05 < Laurenceb> yet its enumerating and i see /dev/ttyACM0 2013-02-05T18:29:58 < dirty_d> hmm? 2013-02-05T18:34:21 < zyp> hmm, I broke your device 2013-02-05T18:34:52 < Laurenceb> and now its not even entering main 2013-02-05T18:35:03 < Laurenceb> oddly this si the same issue i had with TNT 2013-02-05T18:35:19 < zyp> dirty_d, what are you doing with control requests you don't understand? 2013-02-05T18:36:15 < dirty_d> zyp, i was sending a ZLP 2013-02-05T18:36:31 < dirty_d> but since im not using cdc_acm anymore im not getting any 2013-02-05T18:36:47 < dirty_d> i just have one bulk OUT ep 2013-02-05T18:36:53 < zyp> lsusb sent one that you are not handling 2013-02-05T18:36:56 < zyp> http://bin.jvnv.net/f/rKoSf.png 2013-02-05T18:37:43 < zyp> the second last one there is handled correctly (stalled), but the last confused your device and didn't get any answer, and now your device doesn't respond at all 2013-02-05T18:37:57 < dirty_d> hmmm 2013-02-05T18:38:11 < dirty_d> this is without running the libusb program? 2013-02-05T18:38:29 < zyp> yes, I was just running lsusb to check your descriptors 2013-02-05T18:38:48 < dirty_d> thats the get qualifier request right? 2013-02-05T18:38:56 < zyp> the stalled one is 2013-02-05T18:39:03 < zyp> the failing one is «get debug descriptor» 2013-02-05T18:39:19 < dirty_d> hmm 2013-02-05T18:39:40 < zyp> which should also stall 2013-02-05T18:40:06 < dirty_d> i dont get that on my pc 2013-02-05T18:40:24 < zyp> no, it's triggered by lsusb 2013-02-05T18:40:28 < zyp> but that doesn't matter 2013-02-05T18:40:42 < zyp> you still have to handle it correctly, by issuing a stall 2013-02-05T18:40:48 < dirty_d> ahh 2013-02-05T18:41:15 < zyp> as I said, you should stall every request you don't understand 2013-02-05T18:41:15 < dirty_d> but thats not the reason im having a problem is it? because when i run my libusb tset program i dont get it either 2013-02-05T18:41:27 < zyp> true, I haven't gotten to that yet 2013-02-05T18:41:29 < dirty_d> i dont get any other traffic on ep0 when i run it 2013-02-05T18:41:48 < zyp> I'm just complaining about errors in your control handling code ;) 2013-02-05T18:42:05 < dirty_d> ok, ill take care of that too 2013-02-05T18:42:37 < zyp> http://paste.jvnv.net/view/WBr7W <- log corresponding to screenshot, in case you're interested 2013-02-05T18:43:23 < dirty_d> ok 2013-02-05T18:43:54 < dirty_d> im initializing my ep1 after the last setup request, does that matter? 2013-02-05T18:44:26 < zyp> shouldn't 2013-02-05T18:44:29 < dirty_d> ok 2013-02-05T18:47:29 < zyp> uh 2013-02-05T18:48:32 < zyp> http://bin.jvnv.net/f/ZlDc5.png <- device is not acking 2013-02-05T18:49:07 < zyp> every out transaction should consist of an OUT token, a DATAx token and an ACK token 2013-02-05T18:49:17 < zyp> but the device is not replying with the acks 2013-02-05T18:49:19 < dirty_d> zyp, in the output from the rblog you see that i get CTR_RX right? 2013-02-05T18:49:39 < zyp> hmm, yes 2013-02-05T18:49:48 < dirty_d> so wtf is going on? 2013-02-05T18:49:48 < zyp> it shows len=0, len=6, len=0 2013-02-05T18:49:49 < dirty_d> lol 2013-02-05T18:49:53 < dirty_d> yea 2013-02-05T18:49:58 < dirty_d> that i dont get either 2013-02-05T18:50:05 < zyp> but on the bus, the entire packet is sent three times 2013-02-05T18:50:10 < dirty_d> yea 2013-02-05T18:50:25 < dirty_d> and you do see it recive Hello! in the log 2013-02-05T18:50:36 < dirty_d> which means that it should have sent ACK 2013-02-05T18:50:45 < dirty_d> according to the reference manual 2013-02-05T18:51:06 < zyp> yes, this is interesting 2013-02-05T18:51:13 < zyp> I've never seen this behavior before 2013-02-05T18:52:06 < Laurenceb> wtf 2013-02-05T18:52:14 < Laurenceb> is GAE using some weird sort of scanf? 2013-02-05T18:52:26 < Laurenceb> p scanbuff 2013-02-05T18:52:26 < Laurenceb> $4 = " 4 5\r", 2013-02-05T18:52:38 < Laurenceb> i entered "3 4 5\r" 2013-02-05T18:52:50 < Laurenceb> and tried to sscanf three integers 2013-02-05T18:53:03 < Laurenceb> its read the first one and whitespaced it in the string 2013-02-05T18:53:37 -!- Claude is now known as claude 2013-02-05T18:53:56 < dirty_d> zyp, any idea how tht could possibly happen? 2013-02-05T18:54:02 < zyp> not yet 2013-02-05T18:54:26 < zyp> Laurenceb, sounds like you are fucking up and overflowing something else, overwriting the start of the string 2013-02-05T18:57:00 < Laurenceb> hmm 2013-02-05T18:57:08 < Laurenceb> im passing a int16_t pointer 2013-02-05T18:57:19 < Laurenceb> thats probably pad for type int* 2013-02-05T19:00:05 < Laurenceb> same behaviour with int 2013-02-05T19:01:06 < dirty_d> zyp, whats the 19 mean in the 4th column? 2013-02-05T19:01:30 < dirty_d> oh, address 2013-02-05T19:01:32 < zyp> device addr 2013-02-05T19:01:39 < zyp> and next col is endpoint 2013-02-05T19:01:47 < zyp> anyway, found your problem 2013-02-05T19:02:01 < dirty_d> well thats exciting, lol 2013-02-05T19:02:03 < zyp> endpoint type is set to isochronous 2013-02-05T19:02:13 < dirty_d> oh my god 2013-02-05T19:02:25 < zyp> isochronous transfers are not acked, so that's why are receiving correctly, but host is not getting the ack 2013-02-05T19:02:31 < dirty_d> in the descriptor? 2013-02-05T19:02:37 < zyp> no, in EPR 2013-02-05T19:02:46 < zyp> (gdb) p/x usb.EPR[1] 2013-02-05T19:02:46 < zyp> $7 = 0x7421 2013-02-05T19:03:08 < dirty_d> i must not have translated between the number for bulk in the descriptor for the number the EPR register uses 2013-02-05T19:03:22 < dirty_d> i thought they were the same 2013-02-05T19:03:30 < zyp> :p 2013-02-05T19:03:40 < zyp> well, it explains the behavior 2013-02-05T19:03:53 < dirty_d> awesome, thanks 2013-02-05T19:04:05 < Laurenceb> oh wait 2013-02-05T19:04:16 < Laurenceb> unless sscanf has no float support? 2013-02-05T19:04:28 < zyp> Laurenceb, newlib-nano? 2013-02-05T19:04:32 < zyp> that's right 2013-02-05T19:04:35 < Laurenceb> GAE 2013-02-05T19:04:43 < Laurenceb> no float? 2013-02-05T19:04:46 < zyp> with --specs=nano.specs? 2013-02-05T19:04:52 < Laurenceb> i dunno 2013-02-05T19:05:05 < Laurenceb> its prebuilt 2013-02-05T19:05:15 < zyp> huh? 2013-02-05T19:05:25 < Laurenceb> GAE has prebuilt binaries 2013-02-05T19:05:36 < zyp> for what? 2013-02-05T19:05:40 < Laurenceb> oh you mean in my makefile? 2013-02-05T19:05:44 < zyp> of course 2013-02-05T19:05:50 < Laurenceb> im using default chibios thingy 2013-02-05T19:06:00 < zyp> well, I don't know what that means. 2013-02-05T19:06:23 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/Makefile 2013-02-05T19:06:26 < zyp> but newlib-nano defaults to no float support, to save space 2013-02-05T19:06:34 < Laurenceb> i see 2013-02-05T19:06:41 < Laurenceb> is there a way to get float support? 2013-02-05T19:06:49 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-05T19:06:54 < zyp> well, you don't appear to be using newlib-nano 2013-02-05T19:07:21 < Laurenceb> p scanbuff 2013-02-05T19:07:21 < Laurenceb> $15 = " 4.8 5.6\r" 2013-02-05T19:07:32 < Laurenceb> is what happens when i try to sscanf 3 4.8 5.6 2013-02-05T19:07:52 < zyp> try 13 instead of 3 2013-02-05T19:08:00 < zyp> (or some other dual-digit number) 2013-02-05T19:08:30 < zyp> or even some triple digit number 2013-02-05T19:08:40 < Laurenceb> " 3 4.5 6.7\r", '\000' 2013-02-05T19:08:45 < Laurenceb> looks like first digit 2013-02-05T19:08:47 < zyp> right 2013-02-05T19:09:02 < zyp> sounds like you're overflowing or something, breaking it 2013-02-05T19:09:26 < Laurenceb> it read 13 ok 2013-02-05T19:09:31 < Laurenceb> but nothing after that 2013-02-05T19:10:13 < Laurenceb> sscanf((const char*)scanbuff, "%d %f %f", (int*)&test, &per, &peak); 2013-02-05T19:10:47 < Laurenceb> looks sane right? 2013-02-05T19:10:53 < zyp> um 2013-02-05T19:10:58 < zyp> what's the type of test? 2013-02-05T19:11:11 < Laurenceb> int 2013-02-05T19:11:22 < zyp> so why are you casting to (int*)? 2013-02-05T19:11:36 < Laurenceb> as previously it was int16_t 2013-02-05T19:11:44 < zyp> is it still? 2013-02-05T19:12:02 < Laurenceb> nop, fixed that, but no difference to behaviour 2013-02-05T19:12:16 < zyp> because if so, you are going to overwrite whatever is stored in the two bytes after 2013-02-05T19:12:46 < zyp> why are you casting scanbuff to const char*? 2013-02-05T19:13:04 < Laurenceb> as thats what sscanf takes 2013-02-05T19:13:41 < zyp> and what is it originally? 2013-02-05T19:13:50 < Laurenceb> uint8_t 2013-02-05T19:14:08 < zyp> ok, fair enough 2013-02-05T19:15:40 < Laurenceb> yeah, reading three integers fails as well 2013-02-05T19:16:56 < zyp> did you check that the buffer looks sane before sscanf()? 2013-02-05T19:17:00 < Laurenceb> yes 2013-02-05T19:17:10 < zyp> because a const char is not mutable 2013-02-05T19:17:25 < zyp> sscanf() shouldn't be able to write it :p 2013-02-05T19:17:39 < Laurenceb> could the trailing \r screw it up? 2013-02-05T19:17:48 < zyp> no. 2013-02-05T19:18:31 < Laurenceb> how odd then 2013-02-05T19:20:53 -!- toyz [a135b3e2@gateway/web/freenode/ip.161.53.179.226] has joined ##stm32 2013-02-05T19:21:06 < toyz> gday... anyone into cadsoft eagle ? 2013-02-05T19:21:27 < dirty_d> zyp, excellent, working perfectly now 2013-02-05T19:21:29 < karlp> we lovz it 2013-02-05T19:21:30 < dirty_d> thanks again 2013-02-05T19:21:36 < zyp> no problem 2013-02-05T19:22:14 < zyp> I got to see a new interesting bug, so it was worth it for me ;) 2013-02-05T19:22:46 < dirty_d> heh, i guess, more human error 2013-02-05T19:24:02 < zyp> I like not only learning from my own mistakes, but also learning from other people's mistakes ;) 2013-02-05T19:24:07 < Laurenceb> oh i see 2013-02-05T19:24:12 < Laurenceb> its not using libc 2013-02-05T19:24:23 < Laurenceb> its using some random crap i accidentally linked it 2013-02-05T19:24:41 < Laurenceb> hmf 2013-02-05T19:24:42 < Laurenceb> sbrkr.c:(.text._sbrk_r+0xc): undefined reference to `_sbrk' 2013-02-05T19:24:52 < zyp> that's what you get from having all sorts of random crap everywhere 2013-02-05T19:24:56 < zyp> for* 2013-02-05T19:27:55 < Laurenceb> rage 2013-02-05T19:27:56 < Laurenceb> undefined reference to `_sstack' 2013-02-05T19:28:02 < Laurenceb> chibios is using its own names 2013-02-05T19:28:51 < gxti> spoilers: all names are arbitrary 2013-02-05T19:29:13 < Laurenceb> i need a chibios _sbrk(int incr) 2013-02-05T19:30:28 < gxti> os/various/syscalls.c 2013-02-05T19:30:29 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-05T19:32:31 < Laurenceb> aha 2013-02-05T19:32:32 < Laurenceb> thanks 2013-02-05T19:32:50 < Laurenceb> eeek 40KB of code 2013-02-05T19:33:12 < zyp> better remove all the spaces to make it smaller 2013-02-05T19:33:55 < gxti> use a precompiler like ioccc.org to get your code size down 2013-02-05T19:34:53 < Laurenceb> gxti: interesting 2013-02-05T19:35:07 < Laurenceb> p scanbuff$1 = "13 4.5 6.7\r", '\000' (gdb) p test$2 = 13gdb) p per$3 = 4.5(gdb) p peak$4 = 6.69999981 2013-02-05T19:35:13 < Laurenceb> seems to work nicely 2013-02-05T19:35:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-05T19:41:58 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving] 2013-02-05T19:42:48 < Laurenceb> bbl 2013-02-05T19:43:16 < Laurenceb> gxti: unsuccessful troll is unsuccessful 2013-02-05T19:43:24 < gxti> sloowwwwww 2013-02-05T19:46:47 < gxti> lol, indiegogo actually told the chinamen to stop spamming people. politely. 2013-02-05T19:49:12 < gxti> so of the three, 1) seeedstudio pretended they didn't leak/sell my email (yeah right) 2) inmotionhosting which relayed the email "dealt with the issue" presumably with guns or explosives and 3) indiegogo was all creamy cupcakey teddy bears and sunshine but did at least tell them to please stop. an interesting experiment. 2013-02-05T19:49:28 -!- DaKu is now known as daku 2013-02-05T19:54:06 <+Steffanx> pooor pooor gxti 2013-02-05T19:54:22 < gxti> i could have just pressed delete but i was bored 2013-02-05T19:54:24 <+Steffanx> someone sold your emailaddress? 2013-02-05T19:54:32 <+Steffanx> life is hard uh? 2013-02-05T19:54:53 < gxti> a disposible one yeah, which means i can be sure seeedstudio did it 2013-02-05T19:55:00 < gxti> not that they admit it 2013-02-05T19:55:05 < karlp> ffs, forgot the damn clocks _again_ 2013-02-05T20:02:10 < karlp> if I've got adc in continuous mode, with dma enabled, shouldn't it just keep doing the same dma transfer every time it finishes? 2013-02-05T20:02:19 < karlp> I seem to only get a single transfer 2013-02-05T20:02:48 < zyp> suer should, until dma runs out of room to store it 2013-02-05T20:02:50 < zyp> sure* 2013-02-05T20:03:02 < karlp> then I should get the ovfr bit. 2013-02-05T20:03:06 < zyp> which it won't if set to circular mode 2013-02-05T20:05:38 < gxti> look at the other flags, i guess 2013-02-05T20:05:55 < gxti> at least one of them must be set :p 2013-02-05T20:07:32 < karlp> I ugess. 2013-02-05T20:07:50 < karlp> will have to be tomorrow 2013-02-05T20:21:26 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-05T20:22:54 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-05T20:22:54 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has quit [Client Quit] 2013-02-05T20:23:07 -!- barthess [~barthess@178.154.1.114] has quit [Quit: Leaving.] 2013-02-05T20:25:00 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 240 seconds] 2013-02-05T20:25:27 -!- toyz [a135b3e2@gateway/web/freenode/ip.161.53.179.226] has left ##stm32 [] 2013-02-05T20:27:11 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-221.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-05T20:48:00 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-05T20:48:49 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c197:4727:f2c2:12e7] has joined ##stm32 2013-02-05T20:50:43 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 248 seconds] 2013-02-05T21:11:40 < dirty_d> zyp, copying a 10MB file through my stm32 to another file I got only 84KB/s 2013-02-05T21:11:50 < dirty_d> but at least the sha256sums match :p 2013-02-05T21:12:17 < zyp> how? 2013-02-05T21:12:29 < dirty_d> using pyusb 2013-02-05T21:12:37 < dirty_d> reading/writing 64 bytes at a time 2013-02-05T21:12:54 < zyp> you mean just echo? 2013-02-05T21:12:59 < dirty_d> yea 2013-02-05T21:13:05 < dirty_d> the stm32 writes what it reads 2013-02-05T21:13:44 < zyp> try doing larger transfer jobs, say 4096 bytes each 2013-02-05T21:14:02 < dirty_d> my max packet size is 64 bytes though? 2013-02-05T21:14:43 < zyp> sure, so larger transfers will be broken down into a series of 64B packets 2013-02-05T21:14:54 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has quit [Read error: Connection reset by peer] 2013-02-05T21:14:55 < dirty_d> oh, so you think its on the python end? 2013-02-05T21:15:28 < zyp> yes, it's likely host wasting too much time between each packet 2013-02-05T21:15:33 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has joined ##stm32 2013-02-05T21:15:58 < dirty_d> i commented out all your rblog stuff too before i did this 2013-02-05T21:16:06 -!- Laurenceb_ [~Laurence@host81-151-161-179.range81-151.btcentralplus.com] has joined ##stm32 2013-02-05T21:16:09 < zyp> that shouldn't matter, rblog is efficient 2013-02-05T21:16:32 < dirty_d> i need to set RX to NAK until i get CTR_TX right? 2013-02-05T21:16:34 < dirty_d> thats what im doing 2013-02-05T21:17:05 < zyp> not really 2013-02-05T21:17:06 < dirty_d> i dont let the host send me anything until im done writing the last thing sent 2013-02-05T21:17:30 < zyp> hmm, that can be a bit tricky to test then 2013-02-05T21:17:46 < zyp> you need to do async jobs so you can have a read job and a write job at the same time then 2013-02-05T21:18:03 < dirty_d> yea 2013-02-05T21:19:23 < dirty_d> it would only make it twice as fast though right 2013-02-05T21:19:51 -!- barthess [~barthess@178.154.1.114] has joined ##stm32 2013-02-05T21:19:53 < zyp> no, I mean, you need it for it to work at all if you do larger than 64B transfers 2013-02-05T21:20:15 < dirty_d> ahh 2013-02-05T21:20:52 < zyp> otherwise the host will not send IN tokens, so the first frame never get sent back, and therefore you never get ready to receive the second frame, and so on 2013-02-05T21:21:14 < dirty_d> ok 2013-02-05T21:21:34 < dirty_d> its fast enough for anything id use it for anyway 2013-02-05T21:21:34 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-221.pptp.stw-bonn.de] has joined ##stm32 2013-02-05T21:21:43 < zyp> or depending on how many buffers you can fill at a time, you can maybe receive three frames or so (one in tx buffer, one in sram and one in rx buffer) 2013-02-05T21:24:14 < dirty_d> hmm 2013-02-05T21:24:56 < Laurenceb_> emeb: i dont see whats eating all your sdr cycles 2013-02-05T21:25:07 < Laurenceb_> your code is 3 cycles/sample 2013-02-05T21:27:37 < emeb> I'd noticed that 2013-02-05T21:27:48 < emeb> but it still takes like 5us/64 samples 2013-02-05T21:28:58 < dirty_d> zyp, heh, its definitely not python slowing it down, it finsihes in less than a second if i skip sending it through the stm32 2013-02-05T21:29:07 < dirty_d> it takes 2 minutes through the stm32 2013-02-05T21:29:08 < dirty_d> for 10MB 2013-02-05T21:29:45 < dirty_d> i dont really care about speed though, it works! 2013-02-05T21:29:49 < dirty_d> lunchtime 2013-02-05T21:30:16 < emeb> Laurenceb_: the code is 7 cycles / 2 samples (3+2+1+1) = 3.5 cyc/samp 2013-02-05T21:30:27 < zyp> I'm not saying it's python, it's the latency of going back and forth between kernel stack and user space scheduling a new job for every 64 bytes 2013-02-05T21:30:54 < emeb> but my measurements on the scope say ~5us/64 samples = ~5.6cyc/samp 2013-02-05T21:31:13 < emeb> so maybe there's some bus stalling or something 2013-02-05T21:31:49 < zyp> you are wasting lots of time there that could be used for sending packets back to back if the host controller had a multi-packet job scheduled 2013-02-05T21:31:53 < zyp> hmm 2013-02-05T21:32:17 < zyp> tempted to make up some stuff to measure throughput now 2013-02-05T21:33:01 < emeb> Laurenceb_: it was interesting playing with the asm - the 3.5 cyc/samp version made no improvement over the earlier 4 cyc/samp code 2013-02-05T21:33:09 < Laurenceb_> emeb: including the IIR? 2013-02-05T21:33:23 < emeb> I'm leaving the IIR out 2013-02-05T21:33:32 < Laurenceb_> how odd 2013-02-05T21:33:39 < Laurenceb_> is the lookup in CCM? 2013-02-05T21:33:43 < emeb> yes 2013-02-05T21:33:59 < Laurenceb_> and processor at 72mhz? 2013-02-05T21:34:08 < emeb> AFAIK 2013-02-05T21:34:16 < Laurenceb_> very strange 2013-02-05T21:34:20 < emeb> I did measure MCO at one point. :) 2013-02-05T21:34:45 < Laurenceb_> its like 4 times too slow 2013-02-05T21:35:12 < emeb> well, it's somewhat slower than the cycle calcs would suggest. 2013-02-05T21:35:26 < emeb> but yes - about 4x slower than it needs to be to do the job 2013-02-05T21:36:19 < emeb> now, if one backed off the ADC sample rate from ~4.8MHz to ~1.2MHz it would work 2013-02-05T21:36:35 < emeb> but the input bandwidth wouldn't be enough to do much. 2013-02-05T21:37:26 < emeb> could overclock the CPU I suppose :P 2013-02-05T21:37:54 < Laurenceb_> something seems wrong 2013-02-05T21:37:57 < Laurenceb_> but i dunno what 2013-02-05T21:38:14 < Laurenceb_> like something is blocking one of the busses and its going into a wait state 2013-02-05T21:38:31 < emeb> that would make sense 2013-02-05T21:38:37 < emeb> question is what... 2013-02-05T21:39:24 < Laurenceb_> need better debug kit :P 2013-02-05T21:40:40 < Laurenceb_> unless... 2013-02-05T21:40:58 < Laurenceb_> you screwed up with NVIC config and there is an isr firing off all the time and nesting 2013-02-05T21:41:36 < emeb> only 1 ISR running. 2013-02-05T21:41:58 < emeb> just the DMA HC/TC 2013-02-05T21:45:56 < Laurenceb_> https://github.com/emeb/f303_sdr/blob/master/amrx.c#L125 2013-02-05T21:46:11 < Laurenceb_> you have all the IIR stuff inside the GPIO pin code 2013-02-05T21:46:14 < emeb> ccm ram attrib is in the .h file 2013-02-05T21:46:48 < emeb> yeah, but it only runs every 1/4 cycle, so I can measure the tuner/integrator timing by looking at the other three. 2013-02-05T21:47:15 < Laurenceb_> hmm 2013-02-05T21:47:21 < Laurenceb_> im unconvinced :P 2013-02-05T21:47:36 < emeb> :) I'm eager to be proven wrong. 2013-02-05T21:47:52 < emeb> because it would be cool if this could work. 2013-02-05T21:47:55 < Laurenceb_> maybe ill try this tomorrow or something 2013-02-05T21:48:04 < Laurenceb_> ive got a decent scope to use 2013-02-05T21:48:13 < emeb> yeah - give it a whirl. 2013-02-05T21:48:28 < Laurenceb_> runs of F3discoveyr right? 2013-02-05T21:48:32 < emeb> Look at the matlab/octave sim to to see if I missed anything in the architecture 2013-02-05T21:48:41 < emeb> it should run on the F3disco. 2013-02-05T21:48:44 < emeb> same processor. 2013-02-05T21:48:50 < emeb> just needs 8MHz clock. 2013-02-05T21:50:19 < Laurenceb_> k 2013-02-05T21:51:50 -!- Laurenceb_ [~Laurence@host81-151-161-179.range81-151.btcentralplus.com] has quit [Read error: Connection reset by peer] 2013-02-05T21:52:52 -!- Laurenceb_ [~Laurence@host86-174-23-86.range86-174.btcentralplus.com] has joined ##stm32 2013-02-05T21:57:47 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c197:4727:f2c2:12e7] has quit [Remote host closed the connection] 2013-02-05T21:58:24 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-05T22:01:50 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-05T22:03:24 < Laurenceb_> http://news.bbcimg.co.uk/media/images/54682000/jpg/_54682463_012681970-1.jpg 2013-02-05T22:03:25 < Laurenceb_> lol 2013-02-05T22:06:33 < Laurenceb_> i wonder how he cooks them 2013-02-05T22:14:37 -!- dekar [~dekar@212.255.44.126] has joined ##stm32 2013-02-05T22:14:40 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-05T22:15:03 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-05T22:15:39 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-05T22:21:14 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Quit: Leaving] 2013-02-05T22:21:27 < zyp> hmm, now I'm finding weird bugs in my own usb stack 2013-02-05T22:23:40 -!- Netsplit *.net <-> *.split quits: @ChanServ 2013-02-05T22:23:41 < emeb> finding them is good, right? 2013-02-05T22:23:51 < emeb> worse if they're there and you don't find them... 2013-02-05T22:24:01 < Thorn> >Businesses and individuals who wish to incorporate FreeRTOS+CLI into proprietary software for redistribution in any form must first obtain a a low cost (or free if being deployed on an NXP microcontroller) commercial license 2013-02-05T22:24:11 < Laurenceb_> OLD 2013-02-05T22:24:12 < zyp> emeb, true enough 2013-02-05T22:24:36 < Laurenceb_> theres still bugs in chibios usb :-/ 2013-02-05T22:24:48 < Laurenceb_> or at least if you break in the wrong place... 2013-02-05T22:25:30 < zyp> I was going to measure throughput here, so I made a simple bulk echo code 2013-02-05T22:26:38 < Laurenceb_> if i have a cdc-acm device that connect to my linux box, will it see any characters echoes back? 2013-02-05T22:26:40 < zyp> at some point, something goes wrong and even though packages are acked on bus, they don't seem to trigger interrupts 2013-02-05T22:27:10 < Laurenceb_> i want to detect nothing replied and use that to assume it should use default settings... 2013-02-05T22:27:26 < zyp> I wonder if it's a race condition of some sort 2013-02-05T22:27:28 < Laurenceb_> i.e. its not connecting to my python gui 2013-02-05T22:28:27 < Laurenceb_> is anything sent to a cdc-acm device when it first connects? 2013-02-05T22:29:55 < zyp> oh, I think I figured it out 2013-02-05T22:29:57 < Laurenceb_> actually i know the answer already 2013-02-05T22:30:32 -!- Netsplit over, joins: @ChanServ 2013-02-05T22:31:09 < zyp> ah, yep 2013-02-05T22:31:30 < zyp> well, this is silly. 2013-02-05T22:31:52 < zyp> not a bug in my stack, host is at fault 2013-02-05T22:33:02 < zyp> there are several kinds of data tokens, bulk endpoints interleave DATA0 and DATA1 tokens 2013-02-05T22:33:42 < zyp> so device received DATA0 correctly, then was expecting a DATA1 token 2013-02-05T22:34:10 < zyp> but every time I open device from host, it starts out with DATA0 for first packet 2013-02-05T22:34:40 < emeb> so you got 0,0,1,0... 2013-02-05T22:35:28 < zyp> well, rather 0, 0, 0, 0 2013-02-05T22:35:41 < emeb> init @ each comma 2013-02-05T22:36:14 < emeb> so can your stack detect open and reset which token it expects? 2013-02-05T22:36:27 -!- Netsplit *.net <-> *.split quits: @ChanServ 2013-02-05T22:36:44 < zyp> http://bin.jvnv.net/f/nlV5N.png 2013-02-05T22:37:13 < zyp> one succeding attempt, then three failing attepmts, then I poked the register in the debugger to reset it to expect DATA0, then a succeding attempt again 2013-02-05T22:37:53 < emeb> the "get device descriptor" is a new open 2013-02-05T22:37:58 -!- Netsplit over, joins: @ChanServ 2013-02-05T22:38:15 < jpa-> does the "get device descriptor" perhaps contain a DATA1 packet for the reply? 2013-02-05T22:38:22 < jpa-> or is it per endpoint? 2013-02-05T22:38:26 < zyp> per endpoint 2013-02-05T22:38:35 -!- Netsplit *.net <-> *.split quits: @ChanServ 2013-02-05T22:38:35 < zyp> libusbdoes get descriptor to search for devices 2013-02-05T22:39:15 < zyp> anyway, I think the correct solution is doing set_configuration every time libusb opens the device 2013-02-05T22:39:49 -!- Netsplit over, joins: @ChanServ 2013-02-05T22:40:02 < zyp> as far as I've understood, reapplying the same configuration is supposed to tear down the current config, flush all buffers, then reinitializing all endpoints again 2013-02-05T22:40:25 < zyp> but I don't support that in my stack yet :) 2013-02-05T22:41:37 < zyp> hmm, there might be some other control requests that would also do the trick for a specific endpoint 2013-02-05T22:45:32 -!- daku is now known as DaKu 2013-02-05T22:45:52 < zyp> ah, yes 2013-02-05T22:46:03 < zyp> «For endpoints using data toggle, regardless of whether an endpoint has the Halt feature set, a ClearFeature(ENDPOINT_HALT) request always results in the data toggle being reinitialized to DATA0.» 2013-02-05T22:50:46 < emeb> so you just weren't implementing the endpoint_halt yet. 2013-02-05T22:51:38 < zyp> well, host is not executing that command either 2013-02-05T22:52:00 < zyp> so host is at fault for not forcing the device to a known state 2013-02-05T22:53:46 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-221.pptp.stw-bonn.de] has quit [Ping timeout: 245 seconds] 2013-02-05T22:54:32 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-05T22:54:44 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-155.pptp.stw-bonn.de] has joined ##stm32 2013-02-05T22:57:06 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 264 seconds] 2013-02-05T22:57:06 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-05T22:58:25 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-155.pptp.stw-bonn.de] has quit [Client Quit] 2013-02-05T22:59:42 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-185.pptp.stw-bonn.de] has joined ##stm32 2013-02-05T23:05:50 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has left ##stm32 [] 2013-02-05T23:15:22 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-05T23:22:42 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-05T23:30:37 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-05T23:42:44 -!- Netsplit *.net <-> *.split quits: @ChanServ 2013-02-05T23:44:30 -!- DaKu is now known as daku 2013-02-05T23:45:22 < baird> http://i.imgur.com/VgItZ9O.jpg 2013-02-05T23:46:34 -!- Netsplit over, joins: @ChanServ --- Day changed Wed Feb 06 2013 2013-02-06T00:02:34 -!- barthess [~barthess@178.154.1.114] has quit [Quit: Leaving.] 2013-02-06T00:05:42 <+Steffanx> lol baird 2013-02-06T00:13:11 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-06T00:13:26 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-06T00:13:55 < dirty_d> so what exactly does a 3-axis megnetic sensor give you? 2013-02-06T00:14:12 < dirty_d> a vector pointing along the earth's magnetif flux lines towards the north pole or something? 2013-02-06T00:16:42 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-06T00:18:39 < Laurenceb_> kind of 2013-02-06T00:18:54 < Laurenceb_> vaguely towards the north pole 2013-02-06T00:19:08 < Laurenceb_> however, theres a world magnetic model package from USGS 2013-02-06T00:19:24 < Laurenceb_> that will give you predicted field at a given position and time 2013-02-06T00:19:37 < emeb> magnetic north != true north and the difference depends on your location. 2013-02-06T00:20:42 < emeb> plus magno sensors are very sensitive to other magnetic fields in your immediate area - car bodies, metal furniture, electronic devices, etc. 2013-02-06T00:21:19 < dirty_d> yea i dont care about where north is 2013-02-06T00:21:53 < dirty_d> it just seems like a good way to have an absolute direction vector 2013-02-06T00:22:04 < dirty_d> by combining the accelerometer and magnetic sensor 2013-02-06T00:22:45 < dirty_d> with just an accelerometer you can only get elevation and tilt 2013-02-06T00:23:02 < dirty_d> but with both you can get rotation around the gravity vector too 2013-02-06T00:24:02 < Laurenceb_> openpilot have a good build of the usgs code 2013-02-06T00:24:27 < dirty_d> cool 2013-02-06T00:24:40 < dirty_d> but is that was it is? a vector parallel with the flux lines that its inside? 2013-02-06T00:24:44 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-06T00:24:51 < Laurenceb_> erm 2013-02-06T00:24:56 < Laurenceb_> what is what? 2013-02-06T00:25:17 < dirty_d> if you take the x/y/z from the sensor and call it a 3d vector 2013-02-06T00:25:22 < Laurenceb_> yes... 2013-02-06T00:25:49 < dirty_d> is that vector pointing along the magnetic flux lines that the chip is currently inside? 2013-02-06T00:26:01 < dirty_d> or is it measuring something else? 2013-02-06T00:27:59 < Laurenceb_> yes 2013-02-06T00:28:05 < Laurenceb_> along the field lines 2013-02-06T00:28:17 < dirty_d> oh ok, cool 2013-02-06T00:44:00 -!- daku is now known as DaKu 2013-02-06T00:50:36 < dongs> sup, i see dirty_d is still bloggin 2013-02-06T00:51:00 < dongs> < emeb> plus magno sensors are very sensitive to other magnetic fields in your immediate area - car bodies 2013-02-06T00:51:06 < dongs> more like usb connectors near them 2013-02-06T00:51:11 < Laurenceb_> hehe 2013-02-06T00:51:16 < dirty_d> bloggin? 2013-02-06T00:51:21 < Laurenceb_> dont say i didnt warn you 2013-02-06T00:51:47 < Laurenceb_> maybe its just proximity to JEWS 2013-02-06T00:51:49 < Laurenceb_> /troll 2013-02-06T00:51:52 < dongs> fuck yea 2013-02-06T00:52:06 < Laurenceb_> jewfield of troncs win 2013-02-06T00:53:20 < ds2> arrrrggggggg... memory timing problems 2013-02-06T00:53:35 < dongs> http://lunduke.com/?page_id=2646 2013-02-06T00:53:42 < dirty_d> arrrrg maytee 2013-02-06T00:53:51 < ds2> is there a spreadsheet for computing FSMC configs? 2013-02-06T00:54:12 < dirty_d> dongs, wtf is this? lol 2013-02-06T00:56:55 < dirty_d> ds2, whats that 2013-02-06T00:57:22 < ds2> the external memory bus 2013-02-06T00:58:03 < dirty_d> oh 2013-02-06T00:58:48 < ds2> don't think I have the right values 2013-02-06T00:58:59 < ds2> got memory corruption but Linux sort of attempts to boot! 2013-02-06T01:00:10 < dongle> oh nice 2013-02-06T01:02:09 < Laurenceb_> hahaha wtf 2013-02-06T01:02:42 < dongle> ds2: so the board is not complete fail? i dont think you ever sent pics of that yet :( 2013-02-06T01:03:20 < ds2> dongle: not a complete fail... I had to write my own bootloader. Pictures will be done after my scheduling insanity dies down. I got 2 weeks to get that board and another board fully working for a demo 2013-02-06T01:03:57 < ds2> dongle: it seems all the code I am copying from the ST stuff has problems and I have to look it up to fix it. this is almost like a homework assignment :( 2013-02-06T01:04:33 < emeb> ds2: setting up FSMC was pretty easy for my app - just cut-n-paste from the StdPeriph example & tweak timings for my parts. 2013-02-06T01:05:09 < dirty_d> so with this FSMC thing you can have more ram? 2013-02-06T01:05:38 < ds2> emeb: yes. did all that... it is the tweaking part... from a glance, the part I have should be a superset of the ST evalboard 2013-02-06T01:05:54 < ds2> emeb: is the FSMC dependant on the sysclock? 2013-02-06T01:06:05 < ds2> things seems to work better at 24MHz then at 72MHz 2013-02-06T01:06:27 < emeb> ds2: pretty sure that FSMC is running from sysclk 2013-02-06T01:06:51 < emeb> dirty_d: yep - external memory & peripherals. 2013-02-06T01:07:01 < ds2> oh... as in oh crap :( 2013-02-06T01:07:15 < dirty_d> emeb, and it can be used like the internal ram at full speed? 2013-02-06T01:07:31 < ds2> wish ST had a spreadsheet to figure out the registers like TI has for their EMIF controller 2013-02-06T01:07:38 < emeb> dirty_d: it will insert waits as required to meet the timing you tell it. 2013-02-06T01:07:51 < emeb> probably can't run at full speed of internal mem 2013-02-06T01:07:53 < dirty_d> so its slower, but transparently slower? 2013-02-06T01:07:57 < emeb> yes 2013-02-06T01:08:01 < ds2> dirty_d: it is near painfully slow compared to internal RAM 2013-02-06T01:08:04 < dirty_d> oh, thats pretty damn neat 2013-02-06T01:08:25 < dirty_d> how big are the memory chips you can use? 2013-02-06T01:08:34 < emeb> ds2: yeah - I use it for audio delay & minimize reads/writes to the FSMC. 2013-02-06T01:08:39 < ds2> 8Mbyte seems sourceable 2013-02-06T01:08:56 < dirty_d> thats pretty cool 2013-02-06T01:08:59 < ds2> 16Mbyte and 32Mbyte should exists 2013-02-06T01:09:06 < ds2> and you can have like 4 of those chips 2013-02-06T01:09:08 < dirty_d> expensive? 2013-02-06T01:09:13 < ds2> not really 2013-02-06T01:09:23 < ds2> but you have to use the 100pin or more packages 2013-02-06T01:09:24 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-06T01:09:36 < ds2> this is just like the good old days when you did a microcontroller 2013-02-06T01:09:39 < emeb> 144 pin pkg is min for most stuff 2013-02-06T01:09:48 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T01:09:50 < emeb> since the 100 pin FSMC doesn't demux the addr/data 2013-02-06T01:10:05 < ds2> but a 16bit 245 can fix that 2013-02-06T01:10:26 < emeb> ya 2013-02-06T01:10:43 < emeb> so trade off larger STM32 pkg vs external addr latch 2013-02-06T01:10:44 < ds2> you ever did external memory on intel stuff, that should be nearly second nature 2013-02-06T01:11:05 < ds2> the way I see it is you trade one layout problem for another 2013-02-06T01:11:54 -!- DaKu is now known as daku 2013-02-06T01:14:24 -!- jef79m_lurking is now known as jef79m 2013-02-06T01:16:17 < dirty_d> holy crap, these xbee things can go 80km? 2013-02-06T01:19:40 < Laurenceb_> the high power ones can 2013-02-06T01:19:45 < Laurenceb_> pita to use in EU 2013-02-06T01:20:04 < Laurenceb_> people have done 60Km with an si4432 2013-02-06T01:20:18 < dirty_d> thats awesome 2013-02-06T01:21:00 < Laurenceb_> yagi on one end 2013-02-06T01:21:19 < Laurenceb_> other end on a high altitude balloon :P 2013-02-06T01:21:46 < dirty_d> the future is now 2013-02-06T01:22:02 < dirty_d> holy crap, they arent cheap 2013-02-06T01:26:01 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-06T01:26:14 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-06T01:28:13 < dongs> xbee is old news 2013-02-06T01:28:21 < dongs> noone cares about thier overpriced shit anymore 2013-02-06T01:28:44 < dongs> you can get 1W Si4432 breakouts for like 8 eur http://webshop.ideetron.nl/Webwinkel-Product-22676887/RFM23BP.html 2013-02-06T01:29:03 < dongs> inb4 Laurenceb_ bitching about how rfm sucks 2013-02-06T01:30:58 < dirty_d> im gonna have to google that 2013-02-06T01:31:16 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-06T01:32:43 < Laurenceb_> dongs: me and some guys from #highaltitude had a play around with si4432 breakouts 2013-02-06T01:32:53 < dongle> yea i know. and they sucked etc. tl;dr 2013-02-06T01:32:59 < Laurenceb_> turns out rfm sucks massively as it uses the capless thingy 2013-02-06T01:33:09 < dongs> but, shit works 2013-02-06T01:33:11 < Laurenceb_> using diodes on the ic as varicaps 2013-02-06T01:33:15 < dongs> have ytou looked at the 1W version though? 2013-02-06T01:33:26 < Laurenceb_> so we stuck on a TCVCXO 2013-02-06T01:33:30 < Laurenceb_> which was epic 2013-02-06T01:33:38 < Laurenceb_> thats rather naughty 2013-02-06T01:33:53 < Laurenceb_> also, using the ublox modules you can get 10MHz reference 2013-02-06T01:34:11 < Laurenceb_> we used a standalone PLL IC in sot-6 2013-02-06T01:34:25 < Laurenceb_> to get 30mhz reference off the ublox 2013-02-06T01:34:32 < Laurenceb_> but that used an additional 6ma 2013-02-06T01:34:50 < dongs> cool 2013-02-06T01:35:00 < GargantuaSauce> if I wanted, say, a megabit or two of throughput at that sort of transmission power 2013-02-06T01:35:04 < GargantuaSauce> what would you guys recommend 2013-02-06T01:35:05 < gxti> Laurenceb_: can you get 10mhz without 6t? 2013-02-06T01:35:12 < dongs> so youre saying ublox 10mhz+3x pll was more stable than the AWESOME crystal used on rfm? 2013-02-06T01:35:47 < dongs> GargantuaSauce: megabit? good luck. maybe wifi + directional antennas? :) 2013-02-06T01:36:07 < dongs> you'd need to be doing OFDM to get any kinda bandwidth through a narrow channel 2013-02-06T01:36:17 < dongs> all these shitty modules only do FSK type stuff 2013-02-06T01:37:23 < dirty_d> good ole wires 2013-02-06T01:37:55 < GargantuaSauce> yeah ok 2013-02-06T01:38:41 < GargantuaSauce> I am pissing around with nordic 2.4ghz ones now but probably won't be able to cram video through them at a reasonable framerate 2013-02-06T01:39:42 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-06T01:40:00 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-06T01:40:05 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T01:42:15 < Thorn> what applications use the 2.4GHz band (other than wifi, bluetooth and microwave ovens)? 2013-02-06T01:42:27 < GargantuaSauce> all sorts of comsumer shit 2013-02-06T01:42:54 < GargantuaSauce> wireless phones, remote controls for various things 2013-02-06T01:43:18 < gxti> it's unlicensed so pretty much anything can use it as long as it doesn't crap all over the spectrum 2013-02-06T01:43:33 < Thorn> nice 2013-02-06T01:44:15 < Thorn> looks like zigbee is there too 2013-02-06T01:46:25 < Thorn> "numerous cordless phones use a feature called Digital Spread Spectrum. This technology was designed to ward off eavesdroppers, but the phone will change channels at random, leaving no Wi-Fi channel safe from phone interference." 2013-02-06T01:46:59 < emeb> onoes! 2013-02-06T01:51:09 < Laurenceb_> yeah, maybe why my wifi is always screwed 2013-02-06T01:51:14 < Laurenceb_> channel 14 seems to help 2013-02-06T01:52:59 < Thorn> it would be nice to have a 2.4GHz SDR or something like that to analyze the situation 2013-02-06T01:53:21 < zyp> go buy bladerf then :p 2013-02-06T01:53:28 < dongle> haha 2013-02-06T01:53:38 < zyp> oh, they just passed their goal 2013-02-06T01:53:39 < gxti> hmm, dhgate has some more timing receivers than what i could find on ebay for about the same price 2013-02-06T01:54:22 < dongle> still waiting for someone to paypal it for me 2013-02-06T01:54:53 < dongs> https://s3.amazonaws.com/ksr/assets/000/323/650/3ba2e246e41135d8a83845647181a6c4_large.png?1357177552 is this eagle? 2013-02-06T01:56:26 < Thorn> 40MSPS sampling for 3.8GHz? they must do some kind of demodulation before the ADC 2013-02-06T01:56:40 < dongs> wutyt 2013-02-06T01:56:49 < zyp> dongs, somebody on irc channel said orcad 2013-02-06T01:56:54 < dongs> ok 2013-02-06T01:57:09 < dongs> yeah, just sayin pro shit isnt made wiht eagle. 2013-02-06T01:57:23 < dongs> though I kno this one french dtv demodulator company t hat used eagle for thier reference designs 2013-02-06T01:57:31 < dongs> they've sold themselves to Parrot now. 2013-02-06T01:59:10 < zyp> I think I'll ask for altium if I'm gonna do workstuff in the future 2013-02-06T01:59:12 < zyp> :p 2013-02-06T01:59:19 < dongs> yea. 2013-02-06T01:59:24 < gxti> good call 2013-02-06T01:59:42 < zyp> my eagle license isn't valid for commercial designs anyway 2013-02-06T01:59:56 < gxti> did you pay for it? 2013-02-06T02:00:00 < zyp> sure 2013-02-06T02:00:03 < gxti> for a non-com license? that's retarded 2013-02-06T02:00:18 < gxti> unless it's like $20 2013-02-06T02:00:22 < zyp> well, I paid for upgrade to eagle 6 2013-02-06T02:00:26 < dongs> dicktrace unlimited license is liek $600 2013-02-06T02:00:28 < dongs> for any use 2013-02-06T02:00:33 < dongs> which is more than reasonable 2013-02-06T02:00:41 < dongs> considering i made more than that with first board I routed in it 2013-02-06T02:01:04 < gxti> hmm, i should consider that as an alternative to "borrowing" altium 2013-02-06T02:01:14 < zyp> I got eagle 5 license for uni project, so I budgeted it along with parts and shit 2013-02-06T02:01:16 < gxti> i can't go back to kicad but if dicktrace is dongs-approved i'll check it out 2013-02-06T02:01:20 < zyp> so I didn't pay that myself :p 2013-02-06T02:01:29 < dongs> haha kicad 2013-02-06T02:01:42 < gxti> yeah 2013-02-06T02:01:46 < gxti> no. 2013-02-06T02:01:56 < dongs> sometimes dudes submit pcbs made with that shit 2013-02-06T02:02:05 < dongs> every single one made in kicad had DRC errors 2013-02-06T02:02:15 < dongs> pours bridging to shit, etc. 2013-02-06T02:02:17 < gxti> they probably couldn't be arsed to configure it correctly 2013-02-06T02:02:23 < dongs> shrug 2013-02-06T02:02:24 < gxti> it's pretty weird 2013-02-06T02:02:26 < dongs> yeah im sure 2013-02-06T02:02:35 < gxti> i didn't have any problems with the 3-4 boards i did before jumping ship 2013-02-06T02:02:50 < dongs> surpriose it took that long 2013-02-06T02:03:00 < gxti> it was still an upgrade from geda 2013-02-06T02:03:13 < gxti> first thing i tried, never fabbed anything from it 2013-02-06T02:03:37 < dongs> any 'EDA' tool that requires notpad to edit part footprings is fucking fail 2013-02-06T02:03:44 < gxti> indeed 2013-02-06T02:03:45 < dongs> altium footprint designer shit is fucking amazing 2013-02-06T02:03:46 < zyp> I haven't had problems with eagle either, but I wouldn't mind replacing it with something with more features 2013-02-06T02:04:54 -!- Laurenceb_ [~Laurence@host86-174-23-86.range86-174.btcentralplus.com] has quit [Ping timeout: 264 seconds] 2013-02-06T02:05:58 < dongs> about the only thing im really missing from dicktrace is teardropping/trace tapering and pinswap 2013-02-06T02:06:27 < dongs> apparently both are on todo 2013-02-06T02:06:34 < dongs> pinswap is ok manual but would be REALLY handy if was auto 2013-02-06T02:06:41 < zyp> is this when I'm gonna laugh and say that even eagle has pinswap? :p 2013-02-06T02:06:46 < dongs> yeah 2013-02-06T02:06:50 < dongs> well aware 2013-02-06T02:06:57 < dongs> but im sure it's as fucktarded to use as the rest of eagle 2013-02-06T02:07:07 < gxti> i don't think i've used pin swap, should figure that out 2013-02-06T02:07:21 < zyp> enable pinswap tool, click one pin, click pin to swap it with 2013-02-06T02:07:23 < dongs> pinswap is handy when you're lazy routing stuff 2013-02-06T02:07:25 < zyp> simple as that 2013-02-06T02:07:57 < zyp> only annoyance is that it's messing up the schematic at the same time, since it's just swapping the line ends there, no more intelligence 2013-02-06T02:08:31 < zyp> http://bin.jvnv.net/f/EF6TY.png <- so after swapping it looks like this 2013-02-06T02:08:57 < gxti> looks great 2013-02-06T02:09:23 < dongle> amazing. 2013-02-06T02:09:31 < dongle> personally if I was gonna pinswap anything I'd put them into a bus 2013-02-06T02:09:51 < zyp> I don't think it would be any more intelligent 2013-02-06T02:09:56 < zyp> but I haven't tried 2013-02-06T02:10:04 < dongle> cause (at least in dicktrace) the bus will keep pin names inside it 2013-02-06T02:10:11 < dongle> so i can kill one end and r econnect it 2013-02-06T02:10:17 < dongle> without having to re-create the net 2013-02-06T02:10:21 < dongle> so it makes things less aidsy 2013-02-06T02:10:37 < zyp> I think eagle buses work like that too 2013-02-06T02:10:56 < dongle> yeah but eagle buses get a huge name thats made up of everyt net in it 2013-02-06T02:11:09 < dongle> $fuck$you$eagle$lol$dongs[0..16] 2013-02-06T02:11:16 < zyp> probably true 2013-02-06T02:11:31 < zyp> I haven't used buses since I did the led matrix board 2013-02-06T02:12:03 < zyp> http://bin.jvnv.net/f/AXTMf.png 2013-02-06T02:12:59 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds] 2013-02-06T02:15:29 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Quit: Leaving] 2013-02-06T02:17:38 < emeb> lol buses 2013-02-06T02:17:48 < emeb> like the schematic for the dso quad 2013-02-06T02:17:56 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has joined ##stm32 2013-02-06T02:17:56 < emeb> put all nets into one big bus 2013-02-06T02:18:00 < dongle> heh 2013-02-06T02:18:23 < emeb> better off just leaving dangling named nets. 2013-02-06T02:19:08 < emeb> one of my buddies uses diptrace - very happy w/ it. apparently bugfixes & upgrades are well managed. 2013-02-06T02:20:05 < dongs> zyp: http://i.imgur.com/5WQMnDn.png fixed that for you 2013-02-06T02:20:32 < zyp> ha 2013-02-06T02:20:42 < dongs> first thing i thought of when I saw the original 2013-02-06T02:20:50 < dirty_d> zyp, got it up to 314KB/s 2013-02-06T02:21:05 < zyp> with larger transfers, or what? 2013-02-06T02:21:07 < dirty_d> yea 2013-02-06T02:21:28 < zyp> how are you doing in and out at the same time? 2013-02-06T02:21:43 < dirty_d> i just made two threads in python reading and writing at the same time 2013-02-06T02:22:06 < dirty_d> also, you cant have read() automatically set the ep back to VALID 2013-02-06T02:24:34 < dongs> .. threads.. python.. 2013-02-06T02:24:38 < dongs> maybe youre using macos too? 2013-02-06T02:24:43 < dirty_d> lol 2013-02-06T02:24:51 < dirty_d> dongs, why do you hate python? 2013-02-06T02:24:57 < dongs> its fucking retarded 2013-02-06T02:25:00 < dongs> it serves no purpose. 2013-02-06T02:25:00 < dirty_d> why? 2013-02-06T02:25:04 < dirty_d> what how? 2013-02-06T02:25:08 < dirty_d> i use it all the time 2013-02-06T02:25:18 < dongs> it doesnt do anything C doesnt do, requires huge runtime, the syntax is moronic, etc. 2013-02-06T02:25:22 < dirty_d> yea right 2013-02-06T02:25:29 < gxti> lol. 2013-02-06T02:25:33 < dirty_d> you can write a program in python in an hour that it would take a year to write in C 2013-02-06T02:25:41 < dongs> and it will be shit 2013-02-06T02:25:45 < zyp> dongs, which C compiler has interactive mode? 2013-02-06T02:25:45 < dongs> thats the fucking problem 2013-02-06T02:25:50 < zyp> :p 2013-02-06T02:25:51 < dongs> zyp, none, because you dont need it 2013-02-06T02:25:55 < gxti> hahahaha 2013-02-06T02:26:03 < dirty_d> dongs, it will be slower, but sometimes that doesnt matter 2013-02-06T02:26:06 < gxti> good stuff. 2013-02-06T02:26:47 < dongs> dirty_d: slower isnt the problem 2013-02-06T02:26:55 < dongs> the language teaches you to be a lazy cunt 2013-02-06T02:26:59 < dirty_d> dongs, a huge runtime isnt a problem, domputers have huge hard drives 2013-02-06T02:27:01 < dirty_d> its 2013 2013-02-06T02:27:17 < dongs> python "programmers" usually have no idea how to actually program 2013-02-06T02:27:20 < dongs> they just copypaste shit from forums 2013-02-06T02:27:24 < dongs> from other clueless retards 2013-02-06T02:27:28 < zyp> yea, I do that all the time 2013-02-06T02:27:35 < Erlkoenig> don't argue with dongs 2013-02-06T02:27:41 < dirty_d> its gets a job done 2013-02-06T02:27:50 < dirty_d> if you need to do something, there is a module for it 2013-02-06T02:27:52 < gxti> dirty_d: yesterday he argued that static linking was superior :p 2013-02-06T02:28:03 < zyp> isn't it? 2013-02-06T02:28:04 < Erlkoenig> /join #stm32-without-dongs 2013-02-06T02:28:12 < dirty_d> lol 2013-02-06T02:28:12 < dongs> gxti: it is, by far 2013-02-06T02:28:24 < dirty_d> i prefer most things without dongs 2013-02-06T02:28:38 < dongs> gxti: static linking to oddball shit NOTHIGN ELSE will use is superior to dynamic link that fills up your system dir with shit only one app uses. 2013-02-06T02:28:43 < zyp> I don't want dynamic libs on my stm32. 2013-02-06T02:29:04 < gxti> yeah totally different scenario zyp 2013-02-06T02:29:09 < dirty_d> there is a time and a place for C 2013-02-06T02:29:14 < dirty_d> and every program isnt the answer 2013-02-06T02:29:16 < gxti> dongs: so don't put it in the system dir, that's rude anyway 2013-02-06T02:29:24 < gxti> it's not 1996 2013-02-06T02:29:35 < dirty_d> diversify! 2013-02-06T02:30:16 < dongs> gxti: tell that to lunix 2013-02-06T02:30:29 < zyp> dynamic linking is nice for reusable libraries, but for obscure shit that you have to ship with your stuff anyway, there is not much point in not having it statically linked 2013-02-06T02:30:50 < dongs> gxti: i totally dont mind standalone apps which keep shit they use in theri own dir, thats why I have a d:\programs which survives across OS reinstlals, because all the apps in there can run without install/etc 2013-02-06T02:30:54 < zyp> esp. considering that the linker would only pull in the code for those functions that are actually used 2013-02-06T02:30:54 < gxti> you know, for a channel about microcontrollers we sure do have a lot of pointless discussions about operating systems 2013-02-06T02:31:02 < dongs> yeah, wat zyp said 2013-02-06T02:31:21 < dongs> gxti: feel free to start trolling about STM32 2013-02-06T02:31:38 < gxti> we could talk about pwm capture again that seems popular 2013-02-06T02:32:21 < zyp> though, the app-bundle stuff in OS X is a nice solution to packing shit along with your application 2013-02-06T02:33:15 < gxti> http://www.aliexpress.com/item/-/733044197.html shady y/n 2013-02-06T02:33:24 < zyp> a single .app can contain both the executable, dynamic libraries, resource files and other shit that your application might need 2013-02-06T02:34:32 < zyp> on the filesystem it's just a directory with a metadata file and a given structure, but in all the file explorers it's treated like a regular file and can be executed by doubleclicking it 2013-02-06T02:34:47 < dongs> gxti: looks legit 2013-02-06T02:34:55 < dongs> gxti: too bad china is on holiday until end of feb 2013-02-06T02:35:01 < dirty_d> neat 2013-02-06T02:35:13 < dongs> just make sure to ask for 001 :) 2013-02-06T02:35:17 < dongs> or they'll sendd you 000 2013-02-06T02:35:27 < gxti> ha 2013-02-06T02:35:29 < gxti> good call 2013-02-06T02:36:20 < gxti> yeah i'm probably not buying yet anyway, i mostly wanted to figure out how much it would cost to make sure it's a reasonable path to go down 2013-02-06T02:36:39 < zyp> best part is that I can just download any application, which usually comes packed in a .dmg, execute it directly from there, and then if I like it I can move it to /Applications together with other stuff that I might actually use again 2013-02-06T02:36:57 < zyp> IMO that's one of the things OS X got right 2013-02-06T02:37:19 < gxti> too bad there's no 6t floating around, best i could find is 4t and 5t soldered to stuff for like $30 2013-02-06T02:37:33 < dirty_d> zyp, im trying to figure out the overhead in calling this process() function, it takes 33 seconds to transfer 10MB in and out of the stm32 2013-02-06T02:37:45 < gxti> pretty much the going price for secondhand timing receivers it seems 2013-02-06T02:37:48 < dirty_d> and i think it takes 2 calls to process() for each 64 bytes 2013-02-06T02:38:44 < dirty_d> so its being called at about 20KHz? 2013-02-06T02:38:59 < dirty_d> wait no 2013-02-06T02:39:41 < zyp> why care? 40KHz should not be often enough to be concerned about 2013-02-06T02:39:42 < dirty_d> must be more 2013-02-06T02:40:08 < dirty_d> well my main concern was how much cpu time its taking 2013-02-06T02:40:25 < zyp> how much cpu time what is taking? 2013-02-06T02:40:30 < dirty_d> rocess() 2013-02-06T02:40:33 < dirty_d> process() 2013-02-06T02:40:42 < zyp> why does it matter? 2013-02-06T02:40:44 < dirty_d> it seems like not much 2013-02-06T02:40:55 < zyp> of course not 2013-02-06T02:40:55 < dirty_d> well i dunno, if i had a lot of other stuff to do 2013-02-06T02:41:18 < dirty_d> do you think its even worth making it interrupt driven? 2013-02-06T02:41:34 < dirty_d> that could possibly hurt other more time sensitive things right? 2013-02-06T02:41:46 < zyp> uh, depends 2013-02-06T02:42:25 < zyp> my plan is to have it interrupt driven, when I get a better scheduler 2013-02-06T02:42:32 < zyp> (or any scheduling at all, rather) 2013-02-06T02:43:00 < dirty_d> its pretty cool to have usb now though, for anything else i can stream data to my pc now 2013-02-06T02:43:03 < zyp> it would be required for any sort of sleep mode 2013-02-06T02:43:10 < dirty_d> i have too many nows in there 2013-02-06T02:43:17 < dirty_d> yea 2013-02-06T02:44:47 < zyp> my plan is to write a scheduler that keeps track of which threads are ready to run, have usb in it's own thread, then have the usb-interrupt set the usb-thread ready to run whenever something happens 2013-02-06T02:44:54 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has quit [Quit: Leaving.] 2013-02-06T02:45:00 < zyp> that way scheduler can put cpu to sleep whenever no threads are ready to run 2013-02-06T02:46:05 < dirty_d> i havent even tried to do anything like that 2013-02-06T02:47:01 < dirty_d> hows that even work? you have a timer interrupt switch threads? 2013-02-06T02:47:09 < Thorn> it's similar to how windows services interrupts 2013-02-06T02:47:26 < dirty_d> and store each threads registers when that happens? 2013-02-06T02:48:27 < zyp> currently I only have explicit yielding 2013-02-06T02:48:49 < dirty_d> ahh 2013-02-06T02:48:53 < zyp> every thread calls Thread::yield() when it doesn't have anything to do 2013-02-06T02:49:08 < zyp> (which triggers the service call interrupt) 2013-02-06T02:50:14 < zyp> then every register is pushed to the thread stack, stack pointer is saved to thread object, new stack pointer is loaded from new thread objects, and all registers are popped from new thread stack 2013-02-06T02:50:15 < Erlkoenig> in my system every thread is tightly coupled with an interrupt, and if that interrupt occurs, it switches to that thread. If a thread has nothing to do it calls Wait, and the next interrupt with less priority will occur and switch to its thread 2013-02-06T02:51:00 < dirty_d> hmm, interesting 2013-02-06T02:51:26 < zyp> http://cgit.jvnv.net/laks/tree/interrupt/fault.cpp <- the context switch itself looks like this 2013-02-06T02:51:33 < dirty_d> ive never used threads for any microcontroller stuff 2013-02-06T02:51:59 < Erlkoenig> and my context switch (push&pop) is copypasta'd from zyps :-> 2013-02-06T02:52:43 < Erlkoenig> i plan adding something which allows writing in a thread: WaitFor (USART1, SPI1, USB2, 0); or something 2013-02-06T02:52:45 < zyp> mine is still missing fpu stuff, so it would break once I get two threads that's both trying to use the fpu at the same time 2013-02-06T02:53:00 < dirty_d> i dont know any asm at all 2013-02-06T02:53:10 < Erlkoenig> zyp: doesn't the ARMv/ automatically push the FPU's registers on exception entry? 2013-02-06T02:53:15 < Erlkoenig> ARMv7 2013-02-06T02:53:26 < zyp> only half of them 2013-02-06T02:53:29 < Erlkoenig> ah. 2013-02-06T02:53:44 < zyp> similar to the other registers 2013-02-06T02:54:05 < Erlkoenig> okay good to know :-> 2013-02-06T02:54:39 < zyp> ARM has a nice app note about how to handle it, both wrt. normal interrupts and context switches 2013-02-06T02:55:07 < zyp> http://infocenter.arm.com/help/topic/com.arm.doc.dai0298a/index.html 2013-02-06T02:56:14 < Erlkoenig> hmm nice 2013-02-06T02:58:10 < Erlkoenig> hmm they disable the FPU, to save power? 2013-02-06T02:58:28 < Erlkoenig> ah no to get the exception 2013-02-06T02:58:39 < zyp> yeah 2013-02-06T02:59:00 < Erlkoenig> very clever ;) 2013-02-06T03:00:24 < dirty_d> does this thing have enough processing power to act as a digital guitar effect pedal? 2013-02-06T03:00:38 < zyp> probably 2013-02-06T03:00:41 < dirty_d> at 16-bit 48khz audio quality? 2013-02-06T03:00:46 < zyp> emeb is the man for that sort of questions 2013-02-06T03:00:54 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 272 seconds] 2013-02-06T03:01:00 < dirty_d> the adc seems fast enough 2013-02-06T03:01:04 < dirty_d> not sure about output though 2013-02-06T03:01:31 < zyp> you might want to use i2s and external codec chips for analog 2013-02-06T03:01:33 < emeb> dirty_d: I've done a lot of effects & synthesis with STM32 2013-02-06T03:01:49 < emeb> for high-quality though I don't use the on-chip ADC/DAC 2013-02-06T03:02:00 < emeb> instead use an off-chip codec via I2S 2013-02-06T03:02:16 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T03:02:32 < dirty_d> emeb, to a non-audiophile would it be indiscernible? 2013-02-06T03:02:49 < emeb> dirty_d: you could probably get away with it. 2013-02-06T03:02:57 < gxti> effects pedals have artistic leeway 2013-02-06T03:03:05 < dirty_d> yea 2013-02-06T03:03:16 < emeb> the ADC on these parts is pretty good too. not real noisy 2013-02-06T03:03:39 < dirty_d> its 12-bit though isnt it 2013-02-06T03:03:41 < dirty_d> the one i have i think 2013-02-06T03:03:42 < emeb> gxti: esp if the effects are distortion. 2013-02-06T03:03:43 < dirty_d> stm32f3 2013-02-06T03:03:48 < dirty_d> f303 2013-02-06T03:03:59 < emeb> dirty_d: yeah - 12-bit 2013-02-06T03:04:07 < emeb> or less if you want - it's settable. 2013-02-06T03:04:21 < emeb> the F373 has 16-bit ADCs too. 2013-02-06T03:04:33 < emeb> with 14-15 ENOB. 2013-02-06T03:04:40 < dirty_d> pshhh, im going for distortion anyway 2013-02-06T03:04:52 < emeb> then you could get by with GPIO. :) 2013-02-06T03:05:20 < dirty_d> i think you could simulate a tube amplifier pretty easily 2013-02-06T03:05:27 < emeb> HAH! 2013-02-06T03:05:31 < dirty_d> no? 2013-02-06T03:05:40 < emeb> you haven't looked into that much have you? :P 2013-02-06T03:05:45 < dirty_d> i kinda did 2013-02-06T03:05:54 < dirty_d> but i was looking into making an actual tube amp 2013-02-06T03:05:56 < emeb> accurately simulating tube amps is one of the more complex things you can do. 2013-02-06T03:06:02 < dirty_d> but i think the output is just a function of the input 2013-02-06T03:06:07 < dirty_d> its not? 2013-02-06T03:06:08 < emeb> there are nonlinearities ont top of nonlinearities. 2013-02-06T03:06:31 < dirty_d> yea but if you have a tube, cant you just make a lookup table of input vs output? 2013-02-06T03:06:33 < emeb> with some linear distortion thrown in for good measure and time-based effects too. 2013-02-06T03:06:42 < emeb> if only it were that simple. 2013-02-06T03:06:48 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-06T03:07:00 < dirty_d> what else is there? 2013-02-06T03:07:03 < emeb> for accuracy you also need to model the nonlinear response of the transformers. 2013-02-06T03:07:04 < dirty_d> besides capacitance etc 2013-02-06T03:07:12 < emeb> saturation, etc. 2013-02-06T03:07:19 < dirty_d> thats all doable too though right? 2013-02-06T03:07:29 < emeb> Maybe on a PC. 2013-02-06T03:07:31 < dirty_d> just map overall input voltage vs output voltage or something 2013-02-06T03:07:38 < emeb> not that simple 2013-02-06T03:07:57 < dirty_d> it never is 2013-02-06T03:08:06 < emeb> the nonlinearities are mixed with IIR time-based responses. 2013-02-06T03:08:17 < emeb> different orders of nonlinearities with different delays, etc. 2013-02-06T03:08:28 < dirty_d> pure madness 2013-02-06T03:08:48 < emeb> Don't get me wrong - you can do "good enough" models without too much trouble. 2013-02-06T03:09:00 < dirty_d> im a good enough kinda guy 2013-02-06T03:09:01 < dirty_d> lol 2013-02-06T03:09:03 < emeb> but to please the purists things need to be *just so* 2013-02-06T03:09:05 < Erlkoenig> do tubes "sound" soo good that you really want to simulate them perfectly? 2013-02-06T03:09:23 < dirty_d> i like the electric sound better anyway 2013-02-06T03:09:33 < dirty_d> transistor* 2013-02-06T03:09:37 < emeb> Erlkoenig: if you're selling a product that says "Emulates a Marshall" then you better know what you're doing. 2013-02-06T03:10:15 < Erlkoenig> yeah okay... but i meant why would somebody buy such a thing, does it really sound so good? 2013-02-06T03:10:17 < emeb> because there's a million garage-band guys out there who think they know what a Marshall sounds like and they'll be after you if you get it wrong. 2013-02-06T03:10:25 < Erlkoenig> ah :D 2013-02-06T03:11:07 * emeb got a job offer a few weeks back from a guy who wanted TOOBS. Turned him down... 2013-02-06T03:12:24 -!- phantoxeD [~destroy@a89-155-22-21.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds] 2013-02-06T03:13:12 < gxti> tube amps are quite far from 'transparent' but the distortion they introduce can be considered pleasant so for some inane reason people would rather have that than transparency 2013-02-06T03:13:44 < emeb> dirty_d: You might like this website - has lots of stuff on audio fx, including distortion: http://www.spinsemi.com/knowledge_base.html 2013-02-06T03:14:13 < emeb> it's geared to their specific processor, but you can extract algorithms from their discussion and use on other architectures. 2013-02-06T03:14:16 < Erlkoenig> gxti: maybe the recording label should apply the tube-like distortion to their CD's so you only need Transistor amps to get "that sound" 2013-02-06T03:14:49 < gxti> no, you're trying to use logic 2013-02-06T03:14:52 < gxti> logic does not apply here 2013-02-06T03:14:59 < Erlkoenig> bwahaha 2013-02-06T03:15:08 < emeb> audiophool... 2013-02-06T03:15:10 < gxti> only money and choobs 2013-02-06T03:15:16 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-06T03:16:26 < Erlkoenig> i tried to discuss with by bf (art history student) about logic, proofs, and quoting - turns out humanities sciences don't use logic either :D 2013-02-06T03:16:40 < gxti> no they do not 2013-02-06T03:17:31 < gxti> although they are probably more interesting to talk to than audiophools. 2013-02-06T03:17:55 < Erlkoenig> harhar, probably 2013-02-06T03:18:57 < Erlkoenig> at christmas eve i tried to explain "hilberts hotel" (thereby concepts of infinity) to my family.. that was fun :D 2013-02-06T03:22:29 < dirty_d> emeb, cool 2013-02-06T03:22:41 < dirty_d> emeb, i dont see anything about the speed of the DAC on the stm32 2013-02-06T03:30:11 < gxti> k bootloader time. got everything done except the actual flashing. 2013-02-06T03:37:44 < dongs> i thought stm32 dac could at least do audio 2013-02-06T03:37:49 < dongs> so ~48khz 2013-02-06T03:37:53 < dongs> its not fast by any means 2013-02-06T03:38:04 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32 2013-02-06T03:38:05 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host] 2013-02-06T03:38:05 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-06T03:39:19 -!- jef79m is now known as jef79m_lurking 2013-02-06T03:41:08 < emeb> dirty_d: the STM32 DACs can be driven at fairly high sample rates - controllable by a timer, or just whenever you happen to write to them. 2013-02-06T03:41:21 < emeb> They do have output buffers with limited slew rates though 2013-02-06T03:41:36 < emeb> Don't expect to get signals with > 100ksps or so out of them. 2013-02-06T03:41:44 < dirty_d> ok 2013-02-06T03:41:47 < dirty_d> thats fast enough 2013-02-06T03:41:55 < emeb> as dongs said they're plenty good for audio. 2013-02-06T03:43:02 < dongs> im wiring one up for audio telemetry 2013-02-06T03:43:04 < dongs> onflyingthing. 2013-02-06T03:43:08 < dongs> zyp volunteered to code. 2013-02-06T03:43:24 < emeb> sounds cool 2013-02-06T03:43:41 < emeb> you send the audio out via radio? 2013-02-06T03:43:58 -!- jef79m_lurking [~jef79m@124-168-171-28.dyn.iinet.net.au] has quit [Ping timeout: 272 seconds] 2013-02-06T03:44:03 < dongs> yea, a/v transmitter 2013-02-06T03:44:08 < dongs> audio channel is usally unused 2013-02-06T03:44:22 < dongs> or could send telemetry over L and audio over R 2013-02-06T03:44:22 < emeb> nifty. Just do FSK or something? 2013-02-06T03:44:23 < dongs> etc 2013-02-06T03:44:31 < gxti> cool. 2013-02-06T03:44:32 < dongs> right, some easy modulation 2013-02-06T03:44:46 < dongs> maybe some fec 2013-02-06T03:44:49 < dongs> so it doesnt get aidsed up 2013-02-06T03:44:57 < emeb> ya 2013-02-06T03:45:14 < emeb> simple parity or crc. 2013-02-06T03:45:20 < emeb> or RS if you want to get fancy. 2013-02-06T03:45:23 < dongs> well, could actually have something to recover 2013-02-06T03:45:46 -!- jef79m [~jef79m@124-169-138-54.dyn.iinet.net.au] has joined ##stm32 2013-02-06T03:47:05 < emeb> there's a good writeup on how to implement audio-rate MSK here: http://www.digitalsignallabs.com/msk.pdf 2013-02-06T03:47:26 < emeb> I used that to build an FPGA version a while back, but software would work too. 2013-02-06T03:48:31 < emeb> especially easy on the TX side... 2013-02-06T03:51:11 < baird> http://i.imgur.com/LZjJZ.jpg 2013-02-06T03:51:33 < emeb> heh - pretty cool 2013-02-06T03:51:57 < emeb> who knew modems say "please" and "thank you" 2013-02-06T03:53:24 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-06T03:56:01 < baird> You shouldn't anthropomorphise machines, they don't like it. 2013-02-06T04:33:22 -!- mrcan_ [~mrcan@78.173.217.83] has joined ##stm32 2013-02-06T04:33:23 -!- mrcan_ [~mrcan@78.173.217.83] has quit [Changing host] 2013-02-06T04:33:23 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T04:33:44 < dongs> http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=NP8P128A13T1760E-ND 2013-02-06T04:33:47 < dongs> wtf is this? 2013-02-06T04:33:48 < dongs> some variation on psram? 2013-02-06T04:36:32 < dongs> oh never mind some weird fucking flash shit 2013-02-06T04:37:10 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 272 seconds] 2013-02-06T04:41:35 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-06T05:06:49 < dongs> TOTEM POLE 2013-02-06T05:17:06 -!- jef79m [~jef79m@124-169-138-54.dyn.iinet.net.au] has quit [Ping timeout: 245 seconds] 2013-02-06T05:20:23 -!- jef79m [~jef79m@124-169-138-54.dyn.iinet.net.au] has joined ##stm32 2013-02-06T05:25:32 < baird> Needs more Totem Pole Dancing.. 2013-02-06T05:25:38 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-185.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-06T05:43:54 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-06T05:44:11 < R2COM> soo what pcb design software is hot these days.. 2013-02-06T05:44:26 < R2COM> except Casdence allegro of course. 2013-02-06T05:48:42 < dongs> EAGLE PROFESSIONAL 2013-02-06T06:02:02 -!- jef79m [~jef79m@124-169-138-54.dyn.iinet.net.au] has quit [Ping timeout: 272 seconds] 2013-02-06T06:04:00 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-06T06:04:09 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-06T06:04:17 -!- jef79m_lurking [~jef79m@124-169-138-54.dyn.iinet.net.au] has joined ##stm32 2013-02-06T06:04:25 -!- jef79m_lurking is now known as jef79m 2013-02-06T06:23:31 < emeb> oxymoron 2013-02-06T06:25:35 < R2COM> eagle professional, the only thing in eagle professional which exists which does not exist in Kicad is I guess diff-pair routing. thats it, what is the other reason to have eagle professional? 2013-02-06T06:25:53 < dongs> trolling, of course 2013-02-06T06:26:01 < R2COM> well, I used eagle 2013-02-06T06:26:05 < dongs> "i use eagle" looks great on resume 2013-02-06T06:26:38 < R2COM> "I use Cadence" looks way better, plus I dont care on resume 2013-02-06T06:27:09 < R2COM> so what else is in eagle professional, seems like just standard stuff 2013-02-06T06:27:22 < R2COM> hmm 2013-02-06T06:27:44 < dongs> i was joking, geez. 2013-02-06T06:27:44 < R2COM> well...autorouter not going to be worth a shit even for 1000$ anyhow.. 2013-02-06T06:27:58 < dongs> its shit, learn altium since its fairly standard everywehre 2013-02-06T06:28:06 < dongs> all dudes I work with either use altium or pads 2013-02-06T06:28:11 < R2COM> I was looking at Altium, but I heard already so much neagive shit about it 2013-02-06T06:28:44 < R2COM> I heard it integrates stuff in a shitty way, its spice engine is crap, and in general people again mostly use it as layout+schematic tool 2013-02-06T06:29:05 < R2COM> there is also Zuken Cadstar 2013-02-06T06:29:10 < dongs> haha, i think i trid that 2013-02-06T06:29:12 < R2COM> 2000$ starting 2013-02-06T06:29:16 < dongs> thats some obscure japanese thng 2013-02-06T06:29:17 < dongs> really weird 2013-02-06T06:29:24 < R2COM> not German? 2013-02-06T06:29:33 < R2COM> sounds German to me 2013-02-06T06:29:51 < R2COM> if its German, it gotta be like AMG 2013-02-06T06:29:54 < R2COM> xD 2013-02-06T06:30:23 < dongs> no, its not http://www.zuken.com/en/corporate 2013-02-06T06:30:28 < dongs> http://www.zuken.com/en/~/media/Images/Logos/Zuken%20Japanese%20Name.ashx?w=300&h=22&as=1 2013-02-06T06:30:31 < dongs> bla bla bla 2013-02-06T06:30:37 < R2COM> hm ok 2013-02-06T06:31:14 < R2COM> Zukei - Shori - Gijutsu - Kenkyujo 2013-02-06T06:31:16 < R2COM> heh 2013-02-06T06:34:45 < R2COM> and PADS I just toyed around a bit, it has some shit with bugs, app. closing, also if I remember correct it didnt allow similar pins on ICs or something like that, but maybe it got better now... 2013-02-06T06:39:16 < dongs> thats not really bad anyway 2013-02-06T06:39:22 < dongs> helps you organize better 2013-02-06T06:40:33 < R2COM> NC1 NC2 NC3... great organization... 2013-02-06T06:40:45 < R2COM> or LX pins for power converter 2013-02-06T06:42:04 < dongs> i prefer doing NC_1 NC_2 etc. 2013-02-06T06:42:37 < R2COM> yeah right, it helps confuse people who read schemtics later and cannot find XX_N pin because there is only XX pin in datasheet of component. 2013-02-06T06:42:47 < dongs> ;f 2013-02-06T06:44:59 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-06T06:47:15 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 260 seconds] 2013-02-06T06:51:09 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-06T06:51:27 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-06T06:56:57 < dongs> should I bother with 10k poulldown on fet gate for switch or will internal PD on STM32 be enough 2013-02-06T06:58:13 < R2COM> fet outside of stm32? 2013-02-06T06:58:24 < dongs> yea 2013-02-06T06:58:30 < dongs> gpio > fet gate 2013-02-06T06:58:34 < gxti> dongs: internal pulldown won't be there if the mcu is reset or explodes 2013-02-06T06:58:42 < dongs> oh hm good point 2013-02-06T06:58:43 < dongs> k 2013-02-06T06:59:00 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has quit [Ping timeout: 264 seconds] 2013-02-06T06:59:37 < dongs> i guess that'd be one way to tell if board is dead :) 2013-02-06T07:03:48 < dongs> ill just put a R down anyway incase i switch between npn or fet 2013-02-06T07:03:52 < dongs> depending wat i can get cheaper 2013-02-06T07:05:39 < gxti> except r for npn goes in series 2013-02-06T07:06:13 < dongs> both 2013-02-06T07:06:16 < dongs> im keeping both 2013-02-06T07:06:17 < gxti> k 2013-02-06T07:06:19 < dongs> R in series w/gate and pulldown 2013-02-06T07:11:47 < baird> My time machine works! .. http://i.imgur.com/qkKoePh.jpg 2013-02-06T07:12:51 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has joined ##stm32 2013-02-06T07:34:45 < emeb_mac> heh D10 cassettes. Short! 2013-02-06T07:37:31 < baird> I should get about 18kB on each side with these bad boyz. 2013-02-06T07:40:02 < gxti> but can you XIP a f4 on it 2013-02-06T07:44:12 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-06T07:47:19 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-06T07:47:52 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-06T08:01:28 < gxti> is there a way to readback flash using bmp? bootloader is almost working but the result is corrupt or something 2013-02-06T08:09:13 < gxti> never mind, got openocd to do it 2013-02-06T08:10:17 < baird> moar tiem mahcine http://i.imgur.com/YY6qruj.jpg 2013-02-06T08:14:44 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-06T08:14:51 -!- claude is now known as Claude 2013-02-06T08:17:11 < gxti> k, just forgot to call the finish function so the last page was blank 2013-02-06T08:17:15 < gxti> now it works \o/ 2013-02-06T08:58:00 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has joined ##stm32 2013-02-06T08:59:12 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-06T08:59:38 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 255 seconds] 2013-02-06T09:01:41 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-06T09:07:52 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-06T09:09:29 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has quit [Remote host closed the connection] 2013-02-06T09:10:05 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has joined ##stm32 2013-02-06T09:10:33 -!- jef79m is now known as jef79m_lurking 2013-02-06T09:22:32 -!- DLPeterson [~hazelnuss@dcn200-33.dcn.davis.ca.us] has joined ##stm32 2013-02-06T09:22:41 -!- Claude is now known as claude 2013-02-06T09:36:35 -!- DLPeterson [~hazelnuss@dcn200-33.dcn.davis.ca.us] has quit [Quit: leaving] 2013-02-06T09:36:52 -!- DLPeterson [~hazelnuss@168.150.200.33] has joined ##stm32 2013-02-06T09:38:14 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-06T09:42:26 -!- dekar_ [~dekar@212.255.41.49] has joined ##stm32 2013-02-06T09:42:29 -!- mode/##stm32 [+v dekar_] by ChanServ 2013-02-06T09:44:42 -!- dekar [~dekar@212.255.44.126] has quit [Ping timeout: 256 seconds] 2013-02-06T09:52:54 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has quit [Remote host closed the connection] 2013-02-06T10:04:40 -!- jef79m_lurking is now known as jef79m 2013-02-06T10:14:49 -!- jef79m is now known as jef79m_lurking 2013-02-06T10:18:35 -!- jef79m_lurking is now known as jef79m 2013-02-06T10:33:05 -!- jef79m is now known as jef79m_lurking 2013-02-06T10:41:41 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-06T10:42:51 -!- Laurenceb_ [~Laurence@host86-174-23-86.range86-174.btcentralplus.com] has joined ##stm32 2013-02-06T10:44:38 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-06T10:47:58 -!- jef79m_lurking is now known as jef79m 2013-02-06T10:48:39 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-06T10:53:19 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-06T10:57:05 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-06T10:58:57 -!- jef79m is now known as jef79m_lurking 2013-02-06T11:02:30 -!- Laurenceb_ [~Laurence@host86-174-23-86.range86-174.btcentralplus.com] has quit [Ping timeout: 264 seconds] 2013-02-06T11:08:08 -!- jef79m_lurking is now known as jef79m 2013-02-06T11:08:20 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-06T11:10:26 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 245 seconds] 2013-02-06T11:13:17 -!- jef79m is now known as jef79m_lurking 2013-02-06T11:28:44 -!- claude is now known as Claude 2013-02-06T11:32:01 -!- jef79m_lurking is now known as jef79m 2013-02-06T11:45:06 < zyp> gxti, you can just do compare-sections if all you want is to compare flash contents with elf contents 2013-02-06T11:46:31 < dongs> compare-sucktions 2013-02-06T11:46:56 < dongs> zyp, starting to layout this mess, apparently a 4L protorun is getting submitted tomorow so im gonna try to finish iht up before that 2013-02-06T11:47:22 < zyp> the F3 flight controller? 2013-02-06T11:47:37 < dongs> it *almost* looks doable on 2L if i try but I think ill do 4 anyway just for noise/etc 2013-02-06T11:47:40 < dongs> ya 2013-02-06T11:47:44 < zyp> nice 2013-02-06T11:59:20 < dongs> http://i.imgur.com/729Dzv0.png lots of free space ;d 2013-02-06T12:02:53 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has joined ##stm32 2013-02-06T12:03:08 < zyp> usb connector on bottom side? 2013-02-06T12:04:21 < dongs> im keeping same hardware layout as my current shit 2013-02-06T12:04:35 < dongs> so yes. 2013-02-06T12:11:02 -!- Roklobsta_ [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has joined ##stm32 2013-02-06T12:11:03 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has quit [Ping timeout: 256 seconds] 2013-02-06T12:15:35 -!- Roklobsta_ [~quassel@2001:44b8:41d3:6b00:7cf7:4012:5caf:a838] has quit [Ping timeout: 256 seconds] 2013-02-06T12:19:18 < dongs> unless you have real good reasons for kepeing it on top 2013-02-06T12:21:04 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c4f0:7b93:7a6c:ec5c] has joined ##stm32 2013-02-06T12:22:58 < zyp> not at all, I just noticed those resistors that would otherwise collide with it 2013-02-06T12:23:43 < zyp> not assuming they are in their final location, but yeah 2013-02-06T12:23:58 < dongs> they wouldtn actually 2013-02-06T12:24:06 < dongs> that socket ends exactly there 2013-02-06T12:24:07 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-06T12:24:10 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-06T12:24:12 < dongs> ah, but the ones on top yeah 2013-02-06T12:24:16 < dongs> cause you can see under it the keepout area 2013-02-06T12:24:53 < dongs> basically all connectors/blah im keeping in same positions, so I gotta work around that. 2013-02-06T12:32:04 < karlp> this is upgrading your existing f103 based flight controller to f3? 2013-02-06T12:32:12 < Laurenceb> sup 2013-02-06T12:32:12 < karlp> what more does the user get from that? 2013-02-06T12:32:23 < Laurenceb> dongs is making custom F3 flight board? 2013-02-06T12:32:45 < Laurenceb> what firmware will this run? 2013-02-06T12:33:18 < Laurenceb> hmm 2013-02-06T12:33:23 < Laurenceb> needs radio+ gps 2013-02-06T12:38:45 < zyp> I'm planning to get a few and run my firmware on it, to have cheaper stuff that I can crash without being too concerned 2013-02-06T12:47:12 < Laurenceb> im planning on watching pr0n 2013-02-06T12:47:17 < dongs> Laurenceb: has both. d/c about on-board radio or gps. 2013-02-06T12:50:05 < Laurenceb> http://bruxy.regnet.cz/web/linux/EN/housenka-bash-game/ 2013-02-06T12:54:27 < dongs> as retarded as it sounds. 2013-02-06T13:01:40 < dongs> hmm. i could use so-8 spi flash with so much space 2013-02-06T13:01:56 < dongs> im thinking of making a route cutout near the baro 2013-02-06T13:02:00 < dongs> so it can be wrapped easily 2013-02-06T13:04:20 < dongs> http://www.kkmulticopter.kr/index.html?modea=imgview&sn=flycam_black32&size=L like this guy 2013-02-06T13:04:23 < dongs> y/n/mare 2013-02-06T13:04:28 < baird> http://badassdigest.com/2013/02/05/eli-roth-showed-cannibal-holocaust-to-amazonian-tribe/ .. "we brought a television and a generator and we showed them Cannibal Holocaust. They thought it was the funniest thing that they had ever seen [..]" 2013-02-06T13:04:43 < dongs> haha i watched that 2013-02-06T13:04:47 < dongs> what a shit boring movie 2013-02-06T13:04:49 < dongs> and obviously fake as fuck 2013-02-06T13:04:57 < dongs> it had like maybe 2 hilarious moments in it 2013-02-06T13:05:00 < dongs> the rest was just shit 2013-02-06T13:07:56 < Laurenceb> i spit on your grave is better 2013-02-06T13:08:16 < karlp> is the wrapping of the baro so that the rushing wind of the raw power and speed of your rc vehicle doesn't interfere with the baro? 2013-02-06T13:08:27 < karlp> just like fluffy mic covers? 2013-02-06T13:10:11 < dongs> yeah 2013-02-06T13:10:22 < dongs> also shit is affected by light 2013-02-06T13:10:32 < dongs> so people usually put black soft foam and tape it up. 2013-02-06T13:10:58 < dongs> if its on the end of pcb, one could take shrinkwrap and shrink it. 2013-02-06T13:16:23 < dongs> http://i.imgur.com/4EYRKoC.png 2013-02-06T13:16:41 < dongs> pcb place is gonna love me for this i think 2013-02-06T13:17:15 < Laurenceb> apparently the metal encapsulated stuff is better 2013-02-06T13:17:29 < dongs> yeah, thats what i use 2013-02-06T13:17:35 < Laurenceb> cool 2013-02-06T13:17:42 < dongs> they still need to be covered from wind though. 2013-02-06T13:20:08 < Roklobsta> have any of you actually used the F4 extra DsP bells and whistles? 2013-02-06T13:20:19 < jpa-> you mean the instructions? 2013-02-06T13:20:25 < Roklobsta> yeah 2013-02-06T13:20:37 < jpa-> emeb has been using them lately to do SDR on the F3 2013-02-06T13:20:39 < Roklobsta> either using intrinsics or with autovector 2013-02-06T13:20:48 < Roklobsta> jpa oh cool 2013-02-06T13:20:58 < jpa-> but he has been doing it in assembler 2013-02-06T13:21:06 < Roklobsta> what fm or am or ssb 2013-02-06T13:21:23 < dongs> ttps://github.com/emeb/f303_sdr duno, take a look 2013-02-06T13:21:33 < Roklobsta> thanks 2013-02-06T13:23:16 < Roklobsta> oh of course m3 has dsp m4 had fpu 2013-02-06T13:23:30 < dongs> ? 2013-02-06T13:23:34 < dongs> F3 is M4 2013-02-06T13:23:35 < dongs> dsp+fpu 2013-02-06T13:23:42 < dongs> just lower mhz. 2013-02-06T13:23:43 < dongs> though I suspect it should overclock nicely 2013-02-06T13:23:55 < Roklobsta> oh ok 2013-02-06T13:23:57 < Roklobsta> duh 2013-02-06T13:24:25 < Roklobsta> is emb using audio input or something direct from an adc 2013-02-06T13:24:39 <+Steffanx> ADC iirc, but look at hte code :P 2013-02-06T13:24:43 < dongs> huh i dunno look at the code thats why i linked it 2013-02-06T13:24:50 <+Steffanx> adc.c is there :P 2013-02-06T13:25:18 < Roklobsta> oh i see adc built in 2013-02-06T13:26:18 < Laurenceb> hes doing something wrong.... 2013-02-06T13:26:31 < Laurenceb> im going to try that code on f3discovery.. later 2013-02-06T13:26:32 <+Steffanx> Too many spaces? 2013-02-06T13:26:36 -!- dongle [~no@bcas.tv] has quit [Ping timeout: 264 seconds] 2013-02-06T13:26:41 < Laurenceb> emeb thinks its 4 times too slow 2013-02-06T13:26:48 <+Steffanx> dongs what happened with your friend? 2013-02-06T13:26:56 < Laurenceb> i think hes calling a shit ton of floating point code all the time 2013-02-06T13:27:23 < Roklobsta> imperial or metric shit tonne? 2013-02-06T13:28:12 -!- dongle [~no@bcas.tv] has joined ##stm32 2013-02-06T13:31:53 < Roklobsta> what's the goal, a narrowbnd fm demod/ 2013-02-06T13:32:33 < Laurenceb> something like that 2013-02-06T13:42:22 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has joined ##stm32 2013-02-06T13:43:35 < Laurenceb> http://www.bitcraze.se/2013/02/pre-order-has-started/ 2013-02-06T13:43:52 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c4f0:7b93:7a6c:ec5c] has quit [Remote host closed the connection] 2013-02-06T13:54:12 < karlp> is it just me, or do those look absurdly fragile? 2013-02-06T13:54:28 <+Steffanx> it's not just you 2013-02-06T13:55:25 < zyp> shouldn't be that fragile, fr4 is pretty strong 2013-02-06T13:56:00 < karlp> tempted to go cut some out now, screw bolts into the corners and toss it across the room a few times. 2013-02-06T13:56:28 <+Steffanx> :P 2013-02-06T13:56:38 < karlp> hmm, I wonder if there's a scale in the office 2013-02-06T13:59:14 < dongs> Steffanx: wat friend 2013-02-06T13:59:24 < Laurenceb> dongs has no friends 2013-02-06T13:59:29 < dongs> ^ 2013-02-06T13:59:31 < Laurenceb> he trolled them all too much 2013-02-06T13:59:55 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-06T14:00:15 <+Steffanx> dongle, dongs 2013-02-06T14:00:45 < Laurenceb> http://s2.b3ta.com/host/creative/89718/1359979411/Mapman.gif 2013-02-06T14:01:25 <+Steffanx> uh? English humor Laurenceb ? 2013-02-06T14:01:36 < Laurenceb> yes, nvm 2013-02-06T14:01:42 <+Steffanx> *humour 2013-02-06T14:02:38 < dongle> sup? 2013-02-06T14:03:27 < dongle> hmm.. i could move leds inside the USBconnector like i did on the mini board 2013-02-06T14:03:35 < dongle> then i haev a shitton of space on lower right 2013-02-06T14:04:06 < Laurenceb> for silkscreen goatse 2013-02-06T14:04:10 < dongle> ya 2013-02-06T14:04:22 < dongle> that goes on the bottom 2013-02-06T14:05:52 < dongle> cuz so8 spi flash is way more available than 4x4 qfn 2013-02-06T14:13:54 < Laurenceb> http://img-europe.electrocomponents.com/euro/img/site/campaigns/ncjHome/heroBanner/heroBanner_npi_0502.jpg 2013-02-06T14:13:57 < Laurenceb> lol RS 2013-02-06T14:14:25 < Laurenceb> heroBanner more like fail banner 2013-02-06T14:14:48 < dongle> npi = niggerberrypi? 2013-02-06T14:15:00 < dongle> LOL 2013-02-06T14:15:05 < dongle> wasn't far off 2013-02-06T14:15:40 < karlp> what's that mbed board? 2013-02-06T14:15:50 < dongle> does that trash plug in on top of shitberry 2013-02-06T14:15:54 < dongle> into the useless gfpio connector? 2013-02-06T14:16:20 < Laurenceb> http://raspi.tv/ 2013-02-06T14:16:29 < Laurenceb> ^ASPI 2013-02-06T14:16:36 < Laurenceb> says it all 2013-02-06T14:19:52 < jpa-> huh, had a really difficult bug today 2013-02-06T14:20:04 < jpa-> adding one if condition caused touchscreen to become jittery 2013-02-06T14:20:36 < jpa-> reason: the change in code size caused the pages align in cache just so that timing changed slightly 2013-02-06T14:20:52 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-06T14:21:08 < jpa-> the max11802 controller signifies "data valid" status by the lowest 4 bits of the measurement result 2013-02-06T14:21:18 < jpa-> however, in SPI, there bits are clocked out last 2013-02-06T14:21:40 < jpa-> so it can happen so that the first byte is still invalid even though the second byte shows "data valid" 2013-02-06T14:22:21 < jpa-> .. so by a magical coincidence, that single if condition managed to change timing exactly so that this happened and the data was partially invalid 2013-02-06T14:22:57 < jpa-> and because of the exactness of this problem any breakpoint or debug line i added would change the timing and possibly cause the bug to disappear 2013-02-06T14:23:33 < zyp> uh, so you can clock out a message saying data valid, but start of message is still not valid? 2013-02-06T14:23:33 < dongle> nice 2013-02-06T14:23:41 < zyp> sounds dumb 2013-02-06T14:24:17 < zyp> you'd think it would latch at start of read so you can read the entire message atomic 2013-02-06T14:24:22 < jpa-> it does.. though a simple "read again after you have seen the "data valid" bit 2013-02-06T14:24:28 < jpa-> it seems to latch only bytes 2013-02-06T14:24:40 < jpa-> and the measurement result spans two bytes 2013-02-06T14:24:42 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has quit [Ping timeout: 264 seconds] 2013-02-06T14:25:44 < jpa-> http://paste.dy.fi/XLQ/plain there 0xff24 etc. are the corrupted values 2013-02-06T14:26:15 < jpa-> you can see one 0xff50 in also the "working" version, but it was so rare and random that it was not noticeable 2013-02-06T14:27:00 < jpa-> usually "not ready" is 0xffff 2013-02-06T14:35:02 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds] 2013-02-06T14:35:34 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-06T14:44:27 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-06T14:58:26 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-06T15:00:38 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 255 seconds] 2013-02-06T15:01:33 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-06T15:06:59 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 240 seconds] 2013-02-06T15:07:42 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-06T15:08:23 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-06T15:08:46 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-06T15:11:37 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: ….] 2013-02-06T15:16:01 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-06T15:16:01 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-06T15:16:01 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-06T15:16:01 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-06T15:22:08 -!- dongs_ [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-06T15:22:38 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds] 2013-02-06T15:22:51 -!- dongs_ is now known as dongs 2013-02-06T15:23:09 < dongle> fucking feenode 2013-02-06T15:23:55 < Laurenceb> the barn door is wide open on this SDR - so yes, if you had a signal at a harmonic, it will probably come down to your passband 2013-02-06T15:23:59 < Laurenceb> bladerf failz 2013-02-06T15:24:28 < jpa-> so will need a separate filter? 2013-02-06T15:24:35 < Laurenceb> yes 2013-02-06T15:24:40 < Laurenceb> epic fail 2013-02-06T15:24:57 < Laurenceb> that basestation ic is designed to have SAW on the front 2013-02-06T15:25:15 < Laurenceb> it has no inbuilt filtering, unlike the decent rtl-sdr stuff 2013-02-06T15:25:17 < dongle> oops 2013-02-06T15:25:21 < Laurenceb> like e4000 2013-02-06T15:25:25 < Laurenceb> dunno about the other stuff 2013-02-06T15:25:26 <+Steffanx> Laurenceb that is fail no. 1? :) 2013-02-06T15:25:42 < Laurenceb> they use single conversion mixier 2013-02-06T15:25:45 < dongle> is this in their irc shats 2013-02-06T15:25:52 < Laurenceb> any harmonics in LO and they are screwed 2013-02-06T15:25:53 < Laurenceb> yes 2013-02-06T15:25:58 < dongle> joining 2013-02-06T15:26:50 <+Steffanx> Always nice when you have to design your own > 1GHz filters Laurenceb. 2013-02-06T15:27:06 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds] 2013-02-06T15:27:27 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-06T15:28:17 < gxti> zyp: yep i knew about compare, but i needed to actually read back to see how flashing wasn't working 2013-02-06T15:35:58 < Thorn> stupid SDR question: if you have a 80MSPS ADC but want to receive a 40MHz band starting at 2.4GHz, how do you downconvert the signal? 2013-02-06T15:36:33 < Laurenceb> using a 2.4GHz oscillator 2013-02-06T15:37:05 < Thorn> so you need a programmable oscillator and filters? 2013-02-06T15:37:12 < Laurenceb> yes 2013-02-06T15:37:20 < Laurenceb> only that way is __hard 2013-02-06T15:37:29 < BrainDamage1> not necessarily programmable 2013-02-06T15:37:32 < Laurenceb> best to go for 2.42GHz 2013-02-06T15:37:40 < Laurenceb> and IQ mixing 2013-02-06T15:37:49 < BrainDamage1> if the band of your interest is fixed position, the LO can be fixed 2013-02-06T15:37:59 < Thorn> if 2.4GHz is a variable 2013-02-06T15:38:07 < BrainDamage1> then variable LO 2013-02-06T15:38:28 < Laurenceb> this is where bladerf failz 2013-02-06T15:38:39 < Laurenceb> they use a divided digital signal as LO 2013-02-06T15:38:45 < BrainDamage1> wat 2013-02-06T15:38:53 < BrainDamage1> and no filtering 2013-02-06T15:38:57 < BrainDamage1> are they crazy? 2013-02-06T15:39:03 < Laurenceb> yes 2013-02-06T15:39:18 <+Steffanx> You have to make the filter yourself BrainDamage1 :P 2013-02-06T15:39:21 < Laurenceb> that front end is designed to have SAW on the front 2013-02-06T15:39:26 <+Steffanx> Laurenceb wants his 400$ back? 2013-02-06T15:39:33 < Laurenceb> i never paid 2013-02-06T15:39:42 <+Steffanx> okok :) 2013-02-06T15:39:45 <+Steffanx> I thought you did 2013-02-06T15:39:57 < Laurenceb> needed to read datasheets 2013-02-06T15:40:01 < Laurenceb> to check for fail 2013-02-06T15:41:18 < Thorn> so they use a PLL in the FPGA as the LO or something? 2013-02-06T15:41:24 < Laurenceb> nope 2013-02-06T15:41:31 < Laurenceb> its all on the frontend ic 2013-02-06T15:42:38 < BrainDamage1> using digital dividers is common to do divide by 2^n, but requires filtering because it tends to add harmonics ( read: square it ) 2013-02-06T15:42:57 < BrainDamage1> same wrt pulse swallowing for instance 2013-02-06T15:42:59 < Laurenceb> yeah 2013-02-06T15:43:10 < Laurenceb> this is what bladerf missed 2013-02-06T15:43:20 < Laurenceb> this is also what was good about e4000 2013-02-06T15:43:32 < Laurenceb> but i think other front ends have rf filters on the front now too 2013-02-06T15:45:01 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 245 seconds] 2013-02-06T15:46:43 < Thorn> looks like there's a semi-official serial port module for qt 5 2013-02-06T15:47:35 <+Steffanx> Hmm, it's official now? 2013-02-06T15:48:38 < Thorn> qt5? it's been for a month or so, there's 5.0.1 already 2013-02-06T15:48:55 <+Steffanx> I mean the QSerialPort 2013-02-06T15:49:06 <+Steffanx> I've used that once and had all kinds of issues. Especially with events iirc. 2013-02-06T15:49:17 < Thorn> it's called qtserialport now 2013-02-06T15:49:46 <+Steffanx> Not sure if it's the same as i used though 2013-02-06T15:50:27 <+Steffanx> Oh, no. Looks totally different :) 2013-02-06T15:50:46 <+Steffanx> It was called QextSerialPort 2013-02-06T15:52:14 < dongs> i remember chink frsky shit uses qt 2013-02-06T15:52:21 < dongs> and some qt serial prot widget 2013-02-06T15:52:38 < dongs> it was shitty 2013-02-06T15:52:38 < dongs> qext is not official qt shit rihgt? 2013-02-06T15:52:38 < dongs> some opensores lib? 2013-02-06T15:53:43 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T15:54:00 < Thorn> trying to compile, getting some stupid error about an undeclared variable 2013-02-06T15:54:18 < karlp> who owns qt now anyway? 2013-02-06T15:54:33 < dongs> still nokia i think 2013-02-06T15:54:44 < Thorn> digia.no iirc 2013-02-06T15:54:44 < dongs> usd to be trawltech 2013-02-06T15:55:00 < dongs> wat? nokia sold out oo? 2013-02-06T15:55:06 <+Steffanx> Yes 2013-02-06T15:55:07 < Thorn> http://qt.digia.com/ 2013-02-06T15:55:39 < Thorn> they're .fi 2013-02-06T15:56:08 <+Steffanx> You started the old company, didn't you dongs? 2013-02-06T15:57:52 < Laurenceb> http://science.slashdot.org/comments.pl?sid=3437457&cid=42807253 2013-02-06T15:57:55 < Laurenceb> what the fuuuuuu 2013-02-06T15:58:01 < Laurenceb> not sure if troll 2013-02-06T15:59:49 <+Steffanx> " 25-35% of Americans are basically so disconnected from scientific and social reality" writer is one of those :P 2013-02-06T15:59:59 * Steffanx closes tab 2013-02-06T16:00:11 < Laurenceb> its funny if you read on 2013-02-06T16:00:38 < Laurenceb> "The least divisive, least disrupting course of action is for the government to internally and secretly set up an Executive Action team within one the intelligence agencies whose purpose is to discredit, attack and dismantle and neutralize the leaders of the denier terrorist movement. We all know who they are." 2013-02-06T16:02:48 <+Steffanx> Too much blabal 2013-02-06T16:05:30 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-06T16:13:50 < Thorn> do they ever try to compile their code? just fixed an extremely stupid bug in qtserialport 2013-02-06T16:14:20 <+Steffanx> open sores 2013-02-06T16:14:46 <+Steffanx> they can't do shit right 2013-02-06T16:15:13 <+Steffanx> 2013-02-06T16:15:39 < gxti> +3 insightful, gg no re 2013-02-06T16:29:41 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-116.pptp.stw-bonn.de] has joined ##stm32 2013-02-06T16:36:06 < BrainDamage1> http://www.kickstarter.com/projects/461687407/kickstarter-open-source-death-star 2013-02-06T16:40:15 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-06T16:40:28 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-06T16:41:16 <+Steffanx> 20 million yay 2013-02-06T16:41:20 <+Steffanx> dongs started a project iguess 2013-02-06T16:42:57 < Thorn> main laser turret will be goatse-shaped? 2013-02-06T16:45:08 < dirty_d> wait, 302 people donated over $100,000 in total to that bladerf project? 2013-02-06T16:45:25 < gxti> sad, innit 2013-02-06T16:45:28 < dirty_d> more like 302 companies i guess 2013-02-06T16:45:42 < dongs> dirty_d: no, more like 302 nerds 2013-02-06T16:45:48 < Laurenceb> no 2013-02-06T16:45:53 < Laurenceb> 302 people donated 2013-02-06T16:45:53 < dirty_d> nerds dont have that much money 2013-02-06T16:45:58 < Laurenceb> total is 100000 2013-02-06T16:46:01 < BrainDamage1> 331,12583$ 2013-02-06T16:46:08 < BrainDamage1> average, per person 2013-02-06T16:46:11 < dirty_d> thats a lot 2013-02-06T16:46:13 < BrainDamage1> doesn't seem absurd to me 2013-02-06T16:46:16 < dirty_d> but not that much 2013-02-06T16:46:23 < Laurenceb> very cheap 2013-02-06T16:46:26 < dirty_d> i figured people would donate like $20 2013-02-06T16:46:35 < BrainDamage1> they want to buy the device 2013-02-06T16:46:38 < gxti> but then you don't get anything 2013-02-06T16:46:38 < BrainDamage1> not really donate 2013-02-06T16:46:46 < gxti> and people don't really give a shit about the project, they just want stuff 2013-02-06T16:46:48 < dirty_d> truth 2013-02-06T16:47:31 < Laurenceb> thats not much money 2013-02-06T16:47:53 < dirty_d> how much will it cost? 2013-02-06T16:47:54 < Laurenceb> but it looks like the project is fail to me 2013-02-06T16:48:01 < Laurenceb> $400/board 2013-02-06T16:48:16 < emeb_mac> still cheaper than USRP 2013-02-06T16:48:19 < Laurenceb> im not getting one 2013-02-06T16:48:24 < Laurenceb> USRP works 2013-02-06T16:48:25 < dirty_d> daaaaamn 2013-02-06T16:48:55 < Laurenceb> emeb_mac: some permanoob stole my F3discovery :( 2013-02-06T16:48:55 < emeb_mac> I'm sure this works too. Specs are a bit off though 2013-02-06T16:49:06 < emeb_mac> bastard 2013-02-06T16:49:11 < Laurenceb> will try some SDR stuff laterz 2013-02-06T16:49:15 < emeb_mac> ya 2013-02-06T16:49:15 < dirty_d> tole? 2013-02-06T16:49:17 < dirty_d> stole 2013-02-06T16:49:21 < dongs> it was me 2013-02-06T16:49:22 < Laurenceb> if he doesnt wire it up to mains or something 2013-02-06T16:49:35 < Laurenceb> someone from the university 2013-02-06T16:49:46 < Laurenceb> for use in their n00b fail project 2013-02-06T16:49:51 < emeb_mac> downside of working in an open lab I guess 2013-02-06T16:50:07 < Laurenceb> they are going to use LEDs as light detectors.... 2013-02-06T16:50:13 < emeb_mac> old 2013-02-06T16:50:20 < Laurenceb> and fail 2013-02-06T16:50:26 < emeb_mac> ya 2013-02-06T16:50:43 < Laurenceb> ill take it back when they realise they cant port my projects to F3 2013-02-06T16:50:49 < Laurenceb> cuz they are a n00b 2013-02-06T16:51:09 < Laurenceb> or more likely cant build a toolchain 2013-02-06T16:51:14 <+Steffanx> I dont say im perfect, but it's funny when you say stuff like that :P 2013-02-06T16:51:29 < Laurenceb> hey i built one in the end 2013-02-06T16:51:37 < dirty_d> whats the cheapest bidirectional wireless solution i could use for a max distance of about 50 meters and at least 10KB/s of bandwidth? 2013-02-06T16:51:54 < Laurenceb> maybe stm32w108...? 2013-02-06T16:52:09 < emeb_mac> cups + string 2013-02-06T16:52:17 <+Steffanx> Cheap ass nrf24l01+ modules or perhaps even cheaper 433mhz stuff? 2013-02-06T16:52:23 < Laurenceb> emeb_mac: bladerf doesnt have a filter on the front 2013-02-06T16:52:33 < Laurenceb> stm32w has a free processor 2013-02-06T16:52:43 < Laurenceb> and drivers now 2013-02-06T16:52:48 < Laurenceb> so tis actually useful 2013-02-06T16:52:54 < emeb_mac> wat? doesn't that fancy femtobase chip do filters? 2013-02-06T16:52:58 < dirty_d> it would be neat to have a wireless link to a quadcopter board to see the input/output of the pid algorithm and be able to tune it 2013-02-06T16:53:01 <+Steffanx> you have nrf24l01+ modules for ~2-3$ on ebay, that is cheap enough? 2013-02-06T16:53:03 < dongs> haha... 2013-02-06T16:53:08 < Laurenceb> emeb_mac: no, so LO harmonics will screw it 2013-02-06T16:53:12 < dongs> i linked that deathstar kickstarter to a not so bright pal of mine on msn 2013-02-06T16:53:19 < dongs> that kickstarter shit is a scam 2013-02-06T16:53:20 < emeb_mac> seems like a problem 2013-02-06T16:53:28 < dongs> wtf people still pay? 2013-02-06T16:53:30 < Laurenceb> yeah 2013-02-06T16:53:39 < emeb_mac> esp with such a wideband frontend 2013-02-06T16:53:45 <+Steffanx> #bladerf ? 2013-02-06T16:53:51 < dongs> Steffanx: yea 2013-02-06T16:53:51 < Laurenceb> that sort of architecture needs a SAW on the front 2013-02-06T16:54:01 < Laurenceb> or it fails epically 2013-02-06T16:54:15 < emeb_mac> or ceramic or cavity or something 2013-02-06T16:54:23 < Laurenceb> they seem to have forgotten this small issue... 2013-02-06T16:54:25 <+Steffanx> dongs.. bright pal.. msn 2013-02-06T16:54:32 <+Steffanx> something is not right there 2013-02-06T16:54:36 < Laurenceb> hehe Steffanx 2013-02-06T16:55:30 < zyp> Steffanx, he said «not so bright» after all 2013-02-06T16:56:00 < dongs> Steffanx: yeah, i said NOT bright 2013-02-06T16:56:12 < gxti> durrr 2013-02-06T16:56:42 < emeb_mac> ah - IF filters but no tracking filter in front of the mixers. 2013-02-06T16:56:59 < Laurenceb> yes 2013-02-06T16:57:09 < zyp> dirty_d, that's sort of one of the reasons I started with usb 2013-02-06T16:57:16 < Laurenceb> http://www.limemicro.com/img/LMS6002D-block-diagram.png 2013-02-06T16:57:18 < emeb_mac> so could be subject to front-end overload too 2013-02-06T16:57:26 < Laurenceb> yes 2013-02-06T16:57:31 < dirty_d> zyp, what is? 2013-02-06T16:57:33 < zyp> dirty_d, I mean, I'm going with a wired usb link, because I want the speed 2013-02-06T16:57:38 < zyp> 15:52:58 < dirty_d> it would be neat to have a wireless link to a quadcopter board to see the input/output of the pid algorithm and be able to tune it 2013-02-06T16:57:47 < dirty_d> yea me too 2013-02-06T16:57:48 < emeb_mac> I've got a high-power municipal comms tower nearby 2013-02-06T16:57:54 < Laurenceb> hehe 2013-02-06T16:58:01 < emeb_mac> every time that one keys up my RTL dongle desenses. 2013-02-06T16:58:14 < zyp> wireless doesn't have the capacity to stream the data from every iteration 2013-02-06T16:58:24 < dirty_d> really? 2013-02-06T16:58:27 < dirty_d> its not that much data 2013-02-06T16:58:53 < zyp> says who? 2013-02-06T16:58:54 < zyp> :) 2013-02-06T16:59:03 < dirty_d> 16 bytes for the sensed orientation 2013-02-06T16:59:17 < dirty_d> actually 8 2013-02-06T16:59:17 < emeb_mac> maxim has single-chip TV tuners with tracking filters on-chip. 2013-02-06T16:59:27 < Laurenceb> emeb_mac: yeah, so did elonics 2013-02-06T16:59:28 < dirty_d> actually 2013-02-06T16:59:32 < Laurenceb> emeb_mac: http://www.limemicro.com/download/FAQ_v1.0r10.pdf 2013-02-06T16:59:39 < Laurenceb> see Graph 1 2013-02-06T16:59:44 < dirty_d> 6 bytes 2013-02-06T16:59:52 < emeb_mac> Laurenceb: my RTL dongle is elonics - doesn't help :( 2013-02-06T16:59:53 < dirty_d> at 400Hz 2013-02-06T17:00:02 < emeb_mac> besides, elonics is gone 2013-02-06T17:00:06 < dirty_d> and the pid output is another 6 bytes, right? 2013-02-06T17:00:06 < Laurenceb> emeb_mac: it does help, a lot 2013-02-06T17:00:08 < zyp> dirty_d, I want all the input sensor values, intermediate values and output values (motor speeds) 2013-02-06T17:00:22 < Laurenceb> emeb_mac: the problem is just so bad that even elonics suffers 2013-02-06T17:00:34 < dirty_d> zyp, whats that add up to? 2013-02-06T17:00:48 < emeb_mac> Laurenceb: *nod* 2013-02-06T17:01:30 < zyp> let's see, all sensors have 16-bit samples, so there's 20 bytes of raw values 2013-02-06T17:01:55 < zyp> the attitude quaternion alone will be 16 bytes 2013-02-06T17:02:20 < zyp> and the motor outputs are at least two bytes each, so that's another 8 bytes 2013-02-06T17:02:21 < Laurenceb> http://www.limemicro.com/download/LimeMicro-Whitepaper2.pdf 2013-02-06T17:02:23 < Laurenceb> hehe 2013-02-06T17:02:30 < Laurenceb> i see lots of SAW filters 2013-02-06T17:02:44 < zyp> double those figures for inputs and outputs if I pass those also as floats 2013-02-06T17:02:48 <+Steffanx> oops, sorry dongs 2013-02-06T17:02:52 < dirty_d> you could simplify that a lot 2013-02-06T17:03:08 < dirty_d> im just combining the accel and gyro into a vector representing the angle of the quadcopter 2013-02-06T17:03:18 < dirty_d> the gravity vector basically 2013-02-06T17:03:29 < zyp> then there are the accumulators of the I-part of any PID filters 2013-02-06T17:03:32 < dirty_d> so thats only 12 bytes 2013-02-06T17:03:45 < dirty_d> you dont really need that though 2013-02-06T17:03:54 < zyp> that sounds dumb, why not just pass the quaternion? 2013-02-06T17:04:00 < emeb_mac> Laurenceb: that's the problem with these wideband tuners 2013-02-06T17:04:02 < dirty_d> you just need input, you can calculate the intermediate stuff and output on your pc 2013-02-06T17:04:10 < dirty_d> since you know the algorithm 2013-02-06T17:04:13 < emeb_mac> good front-end filtering to the band of interest is needed 2013-02-06T17:04:33 < zyp> it's not about need, it's about wanting full insight in what the board is doing 2013-02-06T17:04:34 < emeb_mac> and for something like bladerf you don't know what the band of interest is. 2013-02-06T17:04:38 < Laurenceb> if someone you make a widely tunable LO itd be solved 2013-02-06T17:04:40 < Claude> emeb_mac, elonics is gone? does this mean no more dvb-t tuners with E4000 in the near future? 2013-02-06T17:04:41 < emeb_mac> different for every user 2013-02-06T17:04:46 < dirty_d> zyp, but you do know 2013-02-06T17:04:49 < dirty_d> you write it 2013-02-06T17:04:50 < emeb_mac> claude: elonics is up for sale 2013-02-06T17:04:50 < Laurenceb> without filling the LO full of harmonics 2013-02-06T17:04:55 < dirty_d> you just copy tour pid code to a pc program 2013-02-06T17:04:56 < Claude> :( 2013-02-06T17:05:05 < emeb_mac> but R820T is about as good 2013-02-06T17:05:12 < dirty_d> it will only differ if something is horribly wrong 2013-02-06T17:05:12 < emeb_mac> and they're still doing fine. 2013-02-06T17:05:15 < Laurenceb> emeb_mac: another way to do it is to mix up 2013-02-06T17:05:34 < Laurenceb> mix up rather than mixing down 2013-02-06T17:05:40 * Claude needs to stock up on E4000 dvb-t tuners then 2013-02-06T17:05:54 < Laurenceb> thats how some more old school stuff worked 2013-02-06T17:05:54 < emeb_mac> claude: probably too late for that 2013-02-06T17:06:05 < Claude> amazon.de has some for 14 euros 2013-02-06T17:06:08 < Claude> or 15 euros 2013-02-06T17:06:14 < emeb_mac> elonics-based RTL dongles are getting scarce 2013-02-06T17:06:27 < emeb_mac> as of about 6mo ago... 2013-02-06T17:06:52 < dirty_d> zyp, right? or am i missing something 2013-02-06T17:07:08 < dirty_d> if you did that, you only need 12KB/s for a 1000Hz pid loop 2013-02-06T17:07:09 < zyp> dirty_d, fair point, given that they start from an equal state 2013-02-06T17:07:13 < dirty_d> right 2013-02-06T17:07:14 < Laurenceb> R820T looks similar to elonics e4000 2013-02-06T17:07:21 < Laurenceb> probably about as good 2013-02-06T17:07:31 < emeb_mac> Laurenceb: high-side mixing has some advantages, but doesn't solve all problems 2013-02-06T17:07:50 < Laurenceb> like? 2013-02-06T17:07:59 < Laurenceb> you need very high frequency kit, true 2013-02-06T17:08:05 < dirty_d> zyp, are you using a chip that outputs a quaternion, or calculating it? 2013-02-06T17:08:12 < emeb_mac> and to be really helpful you need more superhet stages and very high freq IFs 2013-02-06T17:08:18 < zyp> dirty_d, anyway, 20kB/s, not 12, since there's 10 channels of 16-bit input data 2013-02-06T17:08:30 < zyp> calculating 2013-02-06T17:08:33 < Laurenceb> emeb_mac: yes 2013-02-06T17:08:50 < Laurenceb> to do wide tunability, as i see it you need to have crazy high IF 2013-02-06T17:08:58 < emeb_mac> right 2013-02-06T17:09:02 < Laurenceb> about twice the max covered frequency 2013-02-06T17:09:02 < dirty_d> zyp, but dont you reduce that data to a quaternion before the PID? 2013-02-06T17:09:14 < Laurenceb> then you dont care about LO harmonics 2013-02-06T17:09:35 < zyp> dirty_d, sure, but that part is also a tunable filter 2013-02-06T17:09:41 < dirty_d> oh ok 2013-02-06T17:09:45 < dirty_d> still isnt that pretty slow? 2013-02-06T17:09:50 < emeb_mac> and you can still have overload problems unless your 1st mixer has high IM3 2013-02-06T17:09:54 < zyp> slow? 2013-02-06T17:09:59 < dirty_d> 20KB/s? 2013-02-06T17:10:03 < emeb_mac> and integrated mixers usually don't have high IM3 2013-02-06T17:10:06 < Laurenceb> lna -> mix -> saw -> max down 2013-02-06T17:10:22 < Laurenceb> emeb_mac: yeah but LO harmonics are an even bigger issue 2013-02-06T17:10:40 < Laurenceb> s/max/mix 2013-02-06T17:10:48 < emeb_mac> ya 2013-02-06T17:10:55 < Laurenceb> advantage is you dont need complex mixers 2013-02-06T17:10:58 < zyp> dirty_d, it's a matter of perspective 2013-02-06T17:11:03 < Laurenceb> as in IQ 2013-02-06T17:11:09 < Laurenceb> but you need very high F 2013-02-06T17:11:11 < emeb_mac> there's a reason why good spectruma analyzers have YIG tracking filters. 2013-02-06T17:11:18 < Laurenceb> yes 2013-02-06T17:11:48 < Laurenceb> i looked into making an SDR dongle in ~2010 2013-02-06T17:11:48 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds] 2013-02-06T17:12:00 < Laurenceb> came to conclusion it wasnt feasible 2013-02-06T17:12:04 < emeb_mac> something like funcube? 2013-02-06T17:12:08 < Laurenceb> yes 2013-02-06T17:12:11 < zyp> dirty_d, 20kB/s is 200kbaud if you were sending it via an uart, and that pretty much rules out all serial-connected radio interfaces 2013-02-06T17:12:20 < Laurenceb> as it was too power hungry 2013-02-06T17:12:27 < Laurenceb> mixing to 6GHz or so 2013-02-06T17:12:31 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-06T17:12:41 < dirty_d> zyp, the xbees arent fast enough? 2013-02-06T17:12:51 < zyp> (I might be wrong, but I believe stuff like xbee usually tops out at 115200baud) 2013-02-06T17:12:55 < emeb_mac> that's hard to do on a 500ma budget 2013-02-06T17:13:01 < dirty_d> i see 350kb/s 2013-02-06T17:13:03 < Laurenceb> yeah 2013-02-06T17:13:03 < dirty_d> 250* 2013-02-06T17:13:20 < zyp> on uart or radio? 2013-02-06T17:13:24 < dirty_d> im not sure 2013-02-06T17:13:54 < zyp> I think they have an on-air-bitrate of 250kbit/s, but that's not indicative of true throughput 2013-02-06T17:14:02 < Laurenceb> tho linear have some nice newer parts 2013-02-06T17:14:04 < Laurenceb> http://www.linear.com/product/LT5560 2013-02-06T17:14:15 < zyp> so anyway, I'm going with usb :p 2013-02-06T17:14:27 < dirty_d> yea you just cant fly it at the same time 2013-02-06T17:14:32 < emeb_mac> nice 2013-02-06T17:14:43 < zyp> of course I can, as long as the cable is long enough 2013-02-06T17:15:01 < dirty_d> zyp, if your quad isnt tiny you could throw on a raspberry pi connected via usb and a usb wifi adapter 2013-02-06T17:15:29 < dirty_d> i might do just that 2013-02-06T17:15:33 < Laurenceb> emeb_mac: still need two of those and two frequency references... 2013-02-06T17:15:34 < zyp> sure, or a beaglebone and an ethernet cable 2013-02-06T17:15:34 < dirty_d> since the quad is 15 pounds, lol 2013-02-06T17:15:47 < zyp> 15 pounds? 2013-02-06T17:15:51 < dirty_d> haha, yea 2013-02-06T17:15:52 < zyp> isn't that like 7-8kg? 2013-02-06T17:15:56 < dirty_d> yea 2013-02-06T17:15:59 <+Steffanx> dirty_d, go with this fancy new WIFI chip from TI 2013-02-06T17:16:05 < dirty_d> it has 2Kw motors 2013-02-06T17:16:12 < zyp> mine is <1.5kg, and I consider it rather large 2013-02-06T17:16:22 < emeb_mac> Laurenceb: was looking at this the other day: http://www.analog.com/en/rfif-components/modulatorsdemodulators/adrf6806/products/product.html 2013-02-06T17:16:41 < dirty_d> it flys at about half throttle 2013-02-06T17:16:43 < zyp> it's so large that I'm considering to get another smaller one that's more suitable for indoor use 2013-02-06T17:16:54 < emeb_mac> needs filters tho 2013-02-06T17:17:08 < Laurenceb> emeb_mac: yeah, LO divide wrecks it 2013-02-06T17:17:11 < dirty_d> and if im not mistaken, half-throttle is a lot less than half power 2013-02-06T17:17:31 < dirty_d> hmm, but it might be half thrust 2013-02-06T17:17:33 < emeb_mac> these might work too: http://www.maximintegrated.com/datasheet/index.mvp/id/7260 2013-02-06T17:18:49 < dirty_d> i could have just used a helicopter instead 2013-02-06T17:19:01 < dirty_d> for less money 2013-02-06T17:19:03 < Laurenceb> oh nice 2013-02-06T17:19:11 < Laurenceb> emeb_mac: yeah, that looks really good 2013-02-06T17:19:23 < dirty_d> a helicopter is a superior flying machine than a quadcopter 2013-02-06T17:19:27 < emeb_mac> don't know how hard they are to get tho 2013-02-06T17:19:45 < Laurenceb> fairly low F range 2013-02-06T17:19:58 < Laurenceb> but this stuff can often be driven outside the datasheet range 2013-02-06T17:20:19 < zyp> dirty_d, quadrotors are superior crashing machines, because they don't have the same mechanical complexity, so they are cheap to crash and get flying again :p 2013-02-06T17:20:24 < emeb_mac> right - they're spec'ed for certain market. 2013-02-06T17:20:35 < dirty_d> zyp, yea thats true 2013-02-06T17:20:47 < zyp> and I don't want toys that I have to be too careful with :p 2013-02-06T17:21:05 < dirty_d> i have a little blade 120sr heli that ive crashed so many times 2013-02-06T17:21:07 < gxti> stupid question... why quadcopter and not tricopter? 2013-02-06T17:21:09 < dirty_d> never broke 2013-02-06T17:21:15 < dirty_d> but my bird destroyed it 2013-02-06T17:21:22 < dirty_d> chewed threw like every wire on it 2013-02-06T17:21:24 < zyp> gxti, quads are less complex 2013-02-06T17:21:37 < zyp> dirty_d, http://bin.jvnv.net/f/YFQuy.jpg <- I think this was 800g :p 2013-02-06T17:21:52 < gxti> zyp: howso 2013-02-06T17:22:02 < Laurenceb> zyp spotted in the wild 2013-02-06T17:22:10 < dongle> http://i.imgur.com/bC6tUsT.png getting there.. 2013-02-06T17:22:14 < dirty_d> gxti, no moving parts besides the rotors 2013-02-06T17:22:19 < zyp> gxti, with four motors you spin two in each direction, giving a zero sum of torque 2013-02-06T17:22:23 < Laurenceb> dongle: hawt 2013-02-06T17:22:27 < gxti> zyp: oh, ok 2013-02-06T17:22:41 < dirty_d> looks neat 2013-02-06T17:22:42 < zyp> gxti, with three motors you have to have a servo to tilt one motor to control yaw 2013-02-06T17:22:42 < gxti> stupid torque. 2013-02-06T17:23:00 < zyp> on a quad you just control yaw with torque sum 2013-02-06T17:23:06 <+Steffanx> Yes, it's all your fault gxti 2013-02-06T17:23:37 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-06T17:25:41 < emeb_mac> Laurenceb: ever see these? http://www.ti.com/product/afedri8201 2013-02-06T17:26:02 < emeb_mac> nrnd tho 2013-02-06T17:26:42 < karlp> and no replacement? 2013-02-06T17:26:56 < karlp> radio's dead I take it :) 2013-02-06T17:27:18 < gxti> regrettably not 2013-02-06T17:27:25 < Laurenceb> nice 2013-02-06T17:27:43 < Laurenceb> aha 2013-02-06T17:27:48 < Laurenceb> ok i worked it out i think 2013-02-06T17:28:03 < dongs> lol hd radio 2013-02-06T17:28:26 < Laurenceb> lna -> linear tech mixer running off VCO thingy -> SAW at high frequency -> cheap TV front end 2013-02-06T17:28:44 < Laurenceb> so you mix up to a high IF 2013-02-06T17:29:00 < Laurenceb> then go to baseband with cheap shitty whatever front end 2013-02-06T17:29:23 < Laurenceb> that way you should get tunability without piss poor performance 2013-02-06T17:31:09 < emeb_mac> ya 2013-02-06T17:31:27 < emeb_mac> there are a lot of sat-based tuners out there with high input ranges 2013-02-06T17:32:12 < emeb_mac> dongs: what about hd radio? 2013-02-06T17:32:20 < emeb_mac> seems to be more of it these days. 2013-02-06T17:32:52 < emeb_mac> was using RTL dongle + SDR# and could see lots of FM stations w/ IBOC 2013-02-06T17:33:04 < Laurenceb> http://www.linear.com/product/LTC6946 2013-02-06T17:33:09 < Laurenceb> use one of those maybe 2013-02-06T17:33:21 < Laurenceb> and the mixer to mix up to something high... 2013-02-06T17:33:49 < emeb_mac> ya 2013-02-06T17:34:30 <+Steffanx> RF still is black magic 2013-02-06T17:34:39 < gxti> always 2013-02-06T17:35:00 < emeb_mac> lotsa variables to keep track of 2013-02-06T17:35:23 < BrainDamage1> I should start designing comb filters using satanic crosses 2013-02-06T17:35:30 < dongs> 8 nets to go :( 2013-02-06T17:35:40 <+Steffanx> no space left? 2013-02-06T17:35:44 < dongs> there is 2013-02-06T17:36:24 < dongs> plus i can always cheat 2013-02-06T17:36:30 < dongs> ive got layer2 entirely unused still 2013-02-06T17:36:40 <+Steffanx> layer 2 from 4? 2013-02-06T17:36:44 < dongs> yeah 2013-02-06T17:36:46 < dongs> the gnd plane 2013-02-06T17:45:29 < dongs> 4 2013-02-06T17:54:05 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-116.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-06T17:55:01 < dongs> mm, last one 2013-02-06T17:56:10 -!- alexn [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-06T17:56:42 <+Steffanx> gn 2013-02-06T17:56:55 < dongs> you arent sleeping 2013-02-06T17:57:30 <+Steffanx> no, but you will be ..... soon 2013-02-06T17:57:32 < dongs> mwahah, all done, pour time 2013-02-06T17:58:44 < dongs> no errors! 2013-02-06T17:58:50 < emeb_mac> pour a beer? pour some coffee? pour me? 2013-02-06T17:59:32 <+Steffanx> sleep 2013-02-06T18:00:27 < emeb_mac> needs gold plated audio output jack 2013-02-06T18:00:53 <+Steffanx> This channel is for raging so: !@#%ˆ&*&ˆ% crappy windows vista 2013-02-06T18:01:02 < gxti> lol wut 2013-02-06T18:01:02 <+Steffanx> Hangs during boot 2013-02-06T18:01:06 < emeb_mac> why you still worried about vista? 2013-02-06T18:01:15 < emeb_mac> get thee a newer version dude. 2013-02-06T18:01:17 <+Steffanx> too lazy to install something else 2013-02-06T18:01:17 < BrainDamage1> why are you using the most bugged windows version after me and 95? 2013-02-06T18:01:27 < dongs> winMe was great 2013-02-06T18:01:28 <+Steffanx> And it isn't THAT bugged 2013-02-06T18:01:31 < dongs> you're a shitty admin 2013-02-06T18:01:40 < gxti> too obvious dongs 2013-02-06T18:01:52 < gxti> try again 2013-02-06T18:01:59 <+Steffanx> Who installs something new on a laptop he never uses? 2013-02-06T18:02:06 <+Steffanx> like 1 time/year 2013-02-06T18:02:17 < emeb_mac> so stop using it? 2013-02-06T18:02:22 <+Steffanx> No cant 2013-02-06T18:02:23 < BrainDamage1> why keeping a laptop you only use it once a year' 2013-02-06T18:02:26 < BrainDamage1> ? 2013-02-06T18:02:36 <+Steffanx> it's the only windows OS here that works with this crappy pl2303 2013-02-06T18:02:47 < gxti> wot 2013-02-06T18:02:56 < emeb_mac> wag the dog 2013-02-06T18:03:01 <+Steffanx> I use the windows installation once/year BrainDamage1 :P 2013-02-06T18:03:15 <+Steffanx> That's not what i said, i know 2013-02-06T18:03:57 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-06T18:04:52 <+Steffanx> It took me hours to find a proper version of the pl2303 driver BrainDamage1 :) 2013-02-06T18:07:24 < gxti> insert linux troll here 2013-02-06T18:07:27 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-06T18:07:59 <+Steffanx> [enter best i cant use linux argument here] 2013-02-06T18:08:06 <+Steffanx> There is no alternative 2013-02-06T18:10:34 < emeb> there can be only one! 2013-02-06T18:14:17 < karlp> my adc sequence -> dma buffer addressing is off by one :| 2013-02-06T18:14:50 < karlp> first sequence entry goes to dmabuffer[1], last sequence entry goes to dmabuffer[0] 2013-02-06T18:17:19 < emeb> huh. wonder if that's similar to the problem that mervaka is having. 2013-02-06T18:18:14 < karlp> if I use 4 channels (only three I can easily poke) I see my three channels at buff[1,2,3] 2013-02-06T18:18:26 < karlp> no idea what's on buff[0] but it's getting it from somewhere 2013-02-06T18:18:56 < dongs> http://i.imgur.com/hovfJON.png done 2013-02-06T18:19:42 < emeb> purty. f303? 2013-02-06T18:19:46 < dongs> yea 2013-02-06T18:20:09 < emeb> look out world - floating point flyingthing. 2013-02-06T18:20:41 < emeb> why the notches around the device on the top? 2013-02-06T18:20:44 < karlp> but with 3 channels I get 3,1,2 in buf0,1,2. 2013-02-06T18:21:07 < dongs> so i can shrinktube it 2013-02-06T18:21:23 < emeb> what is that device? 2013-02-06T18:21:28 < dongs> presure sensor 2013-02-06T18:21:43 < emeb> what does shrinktube do for it? 2013-02-06T18:21:55 < dongs> prevent air from getting into it :) 2013-02-06T18:22:08 < emeb> uhh.... counterintuitive much? 2013-02-06T18:22:52 < dongs> no. 2013-02-06T18:23:16 < emeb> lowpass filter on the air pressure? 2013-02-06T18:23:23 < dongs> just prevents air from blowing into it 2013-02-06T18:23:28 < dongs> some foam around + shrinktube. 2013-02-06T18:23:38 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-06T18:23:39 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 244 seconds] 2013-02-06T18:23:42 < emeb> right - noise reduction. 2013-02-06T18:30:53 < emeb> apparently C pointers are so hard to understand that it takes 501 pages to describe them: http://www.amazon.com/gp/product/8176563587?ie=UTF8&camp=213733&creative=393177&creativeASIN=8176563587&linkCode=shr&tag=codemonkeyorg-20 2013-02-06T18:32:09 < Laurenceb> When I received the book, some pages started to fall off. Also, some examples given in the book are incorrect. 2013-02-06T18:32:20 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-06T18:32:25 < gxti> that's what happens when you start using c 2013-02-06T18:32:33 < gxti> bits start falling off 2013-02-06T18:32:33 < Thorn> some guru programmer said that ability to understand pointers is inborn 2013-02-06T18:32:48 < emeb> either you can or you can't? 2013-02-06T18:32:53 < Laurenceb> dongs: what about servo power? 2013-02-06T18:33:42 < Laurenceb> dongs: dont need the resistors on usb 2013-02-06T18:34:08 < emeb> extra sense of security... 2013-02-06T18:34:22 < Laurenceb> whats the ic next to the so8 flash? 2013-02-06T18:34:41 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer] 2013-02-06T18:34:51 < Thorn> his job interview is: a very simple question (need to answer without thinking, > 20 seconds = fail), a pointer question and a very hard question (opportunity to correct one's mistakes) 2013-02-06T18:35:09 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-06T18:35:09 -!- TeknoJuce [~NonaSuomy@69.158.165.12] has quit [Changing host] 2013-02-06T18:35:09 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-06T18:36:57 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 252 seconds] 2013-02-06T18:38:03 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-06T18:41:27 -!- barthess [~barthess@178.154.65.67] has joined ##stm32 2013-02-06T18:47:30 < dirty_d> i dont get whats so complicated about pointers 2013-02-06T18:48:00 < dirty_d> i think people have a hard time just ebcause theyve heard that theyre complicated and they go into it not in the right mindset, lol 2013-02-06T18:49:47 < dirty_d> so whats the most populat open source stm32 library? 2013-02-06T18:50:02 < Laurenceb> ask dongs 2013-02-06T18:50:09 < dirty_d> with a reasonable liscense 2013-02-06T18:51:07 <+Steffanx> You have that book Laurenceb ? :D 2013-02-06T18:51:36 < dirty_d> whats dongs gonna tell me? 2013-02-06T18:51:47 < Laurenceb> ask him 2013-02-06T18:52:43 < dirty_d> dongs, so whats the most popular open source stm32 library? 2013-02-06T18:52:51 <+Steffanx> dongs is in dreamland 2013-02-06T18:53:04 < zyp> dirty_d, it probably depends on your background, people coming from java or other languages without pointers might have a hard time wrapping their head around them 2013-02-06T18:53:18 < dirty_d> yea i guess 2013-02-06T18:54:11 <+Steffanx> At least GNOME is going JavaSCript 2013-02-06T18:54:48 < jpa-> dirty_d: libopencm3 is probably the only one; but chibios is also open source if you want more than just a "library" 2013-02-06T18:57:10 < jpa-> also i don't know what your "reasonable" is 2013-02-06T18:57:39 < dirty_d> ill have a look 2013-02-06T18:58:04 < emeb> what's the big deal with the ST StdPeriph license? 2013-02-06T18:58:23 < emeb> I see a lot of folks complaining about it but it doesn't seem so bad to me. 2013-02-06T18:58:37 < jpa-> emeb: it is gpl-incompatible 2013-02-06T18:58:42 < jpa-> for one 2013-02-06T18:58:48 < dirty_d> you cnat sell stuff 2013-02-06T18:59:02 < zyp> I'd care more about the API than the license :) 2013-02-06T18:59:10 < jpa-> yeah it has all sorts of crazy conditions that are not really enforced much 2013-02-06T18:59:23 < dirty_d> even if you did sell soemthing, how would they know what library you used? 2013-02-06T18:59:30 < emeb> wait - ST won't let you use StdPeriph in products that you sell? 2013-02-06T18:59:39 < jpa-> it does 2013-02-06T18:59:42 < jpa-> with some conditions 2013-02-06T18:59:50 < dirty_d> only if you sell it to them? 2013-02-06T19:00:35 < jpa-> dirty_d: it explicitly says "ST grants you .. a license to .. make, have made, use, sell, offer to sell ..." 2013-02-06T19:00:48 < dirty_d> jpa-, but at the end it says you cant sell 2013-02-06T19:00:50 < dirty_d> i dont get it 2013-02-06T19:00:58 < jpa-> dirty_d: at what point? 2013-02-06T19:01:07 < dirty_d> the last number or bullet etc 2013-02-06T19:01:13 < dirty_d> i dont have it infront of me 2013-02-06T19:01:22 < jpa-> http://www.st.com/internet/com/LEGAL_RESOURCES/LEGAL_AGREEMENT/LICENSE_AGREEMENT/software_license_agreement_liberty_v2.pdf 2013-02-06T19:01:24 < jpa-> there you go 2013-02-06T19:02:09 < jpa-> "Unless otherwise explicitly stated in this Agreement, You may not sell," 2013-02-06T19:02:13 < jpa-> but it was explicitly stated 2013-02-06T19:02:21 < emeb> right 2013-02-06T19:02:53 < dirty_d> damn lawers! 2013-02-06T19:02:55 < emeb> basically "you can do a, b,c but anything else is prohibited" 2013-02-06T19:03:09 < jpa-> but the problem is "any use, adaptation, ... for use with processors manufactured by .. an entity other than ST .. is a .. breach of this agreement" 2013-02-06T19:03:37 < zyp> why is that a problem? 2013-02-06T19:03:46 < emeb> right - so don't use StdPeriph with other mfgs processors. Big deal. 2013-02-06T19:03:55 < jpa-> so it conflicts with GPL "You may not impose any further restrictions on the recipients' exercise of the rights granted herein." 2013-02-06T19:04:04 < jpa-> so you cannot use stdperiph and gpl code in the same project 2013-02-06T19:04:28 < emeb> No problem there. 2013-02-06T19:04:50 < jpa-> how so? unless you don't care to use gpl anyway 2013-02-06T19:05:35 < emeb> right - I don't use randomly harvested code in stuff that I sell. 2013-02-06T19:05:55 < zyp> jpa-, what stops you from giving out your code under gpl while keeping stdperiphlib under its own license? 2013-02-06T19:06:10 < jpa-> zyp: if you want to distribute binaries 2013-02-06T19:06:17 < jpa-> (or devices with binaries) 2013-02-06T19:06:36 < zyp> then what? 2013-02-06T19:06:47 < jpa-> then the whole bunch has to be under gpl 2013-02-06T19:06:53 < jpa-> and stdperiph cannot be under gpl 2013-02-06T19:07:31 < zyp> are you sure about that? 2013-02-06T19:07:42 < jpa-> as far as i understand, i'm not a lawyer 2013-02-06T19:07:52 < emeb> depends on which version of gpl - v2, v2, LGPL, etc. 2013-02-06T19:08:00 < emeb> V3, etc 2013-02-06T19:08:04 < jpa-> all microcontroller code links to a single binary so it makes a derivative work quite surely 2013-02-06T19:08:10 < jpa-> LGPL is probably ok 2013-02-06T19:08:24 -!- Claude is now known as claude 2013-02-06T19:08:34 < zyp> also, if you are the copyright owner, nobody is forcing you to follow gpl yourself even though you make the source available under gpl 2013-02-06T19:08:41 < jpa-> sure 2013-02-06T19:08:45 < emeb> although with the way FSF is going these days it seems like they want to make GPL in general more restrictive. 2013-02-06T19:08:49 < jpa-> but if you want to use e.g. chibios or uclinux it is a problem 2013-02-06T19:09:40 < emeb> right - so those oses/libs provide their own interfaces to ST hardware that doesn't require StdPeriph. 2013-02-06T19:09:49 < jpa-> (ok, chibios has exceptions if you don't modify it, and in reality no-one seems to care) 2013-02-06T19:10:17 < jpa-> actually chibios uses stdperiph anyway 2013-02-06T19:10:22 < jpa-> or the headers, that is 2013-02-06T19:10:25 < emeb> :D 2013-02-06T19:10:36 < jpa-> but he is the copyright owner so it is so-so ok 2013-02-06T19:11:02 < emeb> well, sounds like in general it's a manageable thing. 2013-02-06T19:11:08 < jpa-> another slight issue with the ST license is that if they find some excuse to blame you for a breach, they can forbid you from using the license at all, for anything 2013-02-06T19:11:10 < emeb> not worth all the sound & fury I've seen. 2013-02-06T19:11:36 < zyp> let's see 2013-02-06T19:11:37 < zyp> «Both versions of the GPL have an exception to their copyleft, commonly called the system library exception. If the GPL-incompatible libraries you want to use meet the criteria for a system library, then you don't have to do anything special to use them; the requirement to distribute source code for the whole program does not include those libraries, even if you distribute a linked executable containing them.» 2013-02-06T19:11:55 < zyp> that sounds like something stdperiphlib might fall under 2013-02-06T19:12:26 -!- DLPeterson [~hazelnuss@168.150.200.33] has quit [Ping timeout: 272 seconds] 2013-02-06T19:12:32 < dirty_d> no chibios for f3? 2013-02-06T19:12:45 < jpa-> zyp: uh? really? didn't know about such exception 2013-02-06T19:13:11 < jpa-> why can't i find it in the license 2013-02-06T19:13:35 < jpa-> written in lawyer-talk :F 2013-02-06T19:13:46 < zyp> for gplv3 «system libraries» is defined in section 1 2013-02-06T19:13:49 < emeb> http://www.gnu.org/licenses/gpl-faq.html#SystemLibraryException 2013-02-06T19:14:11 < zyp> «The “System Libraries” of an executable work include anything, other than the work as a whole, that (a) is included in the normal form of packaging a Major Component, but which is not part of that Major Component, and (b) serves only to enable use of the work with that Major Component, or to implement a Standard Interface for which an implementation is available to the public in source code form.» 2013-02-06T19:14:24 < zyp> «standard interface» 2013-02-06T19:14:34 < zyp> sounds like we're on the right track 2013-02-06T19:14:47 < jpa-> GPLv2 is not that clear on it "However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable. 2013-02-06T19:16:25 < zyp> I'd argue that stdperiphlib is distributed along with the «major components» since it comes with the SDK or whatever from ST 2013-02-06T19:18:50 < Thorn> you could ask on #gnu 2013-02-06T19:19:18 <+Steffanx> I bet you can't 2013-02-06T19:19:37 <+Steffanx> Those guys always say: screw you.. fuck everything else that is not GPL 2013-02-06T19:19:41 < zyp> in layman's terms I'm reading this as «you are allowed to make use of standard libraries for interfacing with the platform without those libraries falling under the terms of the GPL» 2013-02-06T19:23:30 < Thorn> there's no Major Component here. also see GPLv2 text: "...unless that component itself accompanies the executable" - the executable is its own OS in this case 2013-02-06T19:23:55 <+Steffanx> Didn't dekar_ do all kind of 'research' and knows a lot about it? 2013-02-06T19:24:44 < Thorn> unless you count hardware itself as a Major Component 2013-02-06T19:25:08 < zyp> Thorn, the gplv3 doesn't phrase it like that 2013-02-06T19:25:14 < zyp> isn't it? 2013-02-06T19:26:42 < Thorn> okay, how does (a) from v3 text apply then? "is included in the normal form of packaging a Major Component" 2013-02-06T19:27:01 < zippe> Having spoken with these people over the years 2013-02-06T19:27:08 < zippe> The answer is going to depend on who you ask 2013-02-06T19:27:09 < Thorn> what is a Major Component, what is its form of packaging 2013-02-06T19:27:29 <+Steffanx> ha zippe :D 2013-02-06T19:27:34 < zippe> The best way to approach this is to find Eben Moglen somewhere and buy him dinner. 2013-02-06T19:27:45 < zippe> Or lunch, or a beer, or whatever. 2013-02-06T19:27:59 < zippe> Then secretly record him (in a state where single-party consent is law) 2013-02-06T19:28:03 < zyp> «A “Major Component”, in this context, means a major essential component (kernel, window system, and so on) of the specific operating system (if any) on which the executable work runs, or a compiler used to produce the work, or an object code interpreter used to run it.» 2013-02-06T19:28:07 <+Steffanx> can we find him somewhere neer rms zippe ? 2013-02-06T19:28:12 <+Steffanx> *near 2013-02-06T19:28:13 < zyp> I heard the cpu is an object code interpreter 2013-02-06T19:28:26 < zippe> Steffanx: generally better to stay as far away from RMS as possible 2013-02-06T19:28:43 <+Steffanx> Tell Eben Moglen that 2013-02-06T19:29:10 < zippe> I'm not "telling" Eben anything 2013-02-06T19:29:16 <+Steffanx> :( 2013-02-06T19:29:30 < zippe> We disagree at a pretty fundamental level, and that's more or less that. 2013-02-06T19:29:53 <+Steffanx> you've talked with/to him once? 2013-02-06T19:30:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-06T19:30:44 < zyp> Steffanx, it's rather the fact that he's one of the guys behind the GPL 2013-02-06T19:30:45 <+Steffanx> or read his manifest 2013-02-06T19:32:49 < zippe> We've talked briefly, and I've heard him speak more than once 2013-02-06T19:33:09 < karlp> hrmm, I wonder if it's doing another adc read first, and that's why it's rotated by one in the output buffer. 2013-02-06T19:33:13 < dirty_d> where on earch is the st usb library? 2013-02-06T19:33:16 < dirty_d> earth 2013-02-06T19:33:27 < Thorn> there're at least two 2013-02-06T19:33:49 < dirty_d> whats the difference? 2013-02-06T19:33:52 < Thorn> for the F102/F103 core and for synopsys 2013-02-06T19:34:11 < Thorn> not sure about F3 but it has same core as F102/103 2013-02-06T19:34:42 < dirty_d> hmm 2013-02-06T19:34:49 < dirty_d> i cant find the librry, only pdfs 2013-02-06T19:35:08 < Thorn> synopsys cores are device/host/otg, they're in F107, F2 and F4 2013-02-06T19:35:35 -!- toyz [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has joined ##stm32 2013-02-06T19:35:43 < toyz> gday.. 2013-02-06T19:36:08 < toyz> can some1 point me to some hydraulic oriented people ????? 2013-02-06T19:36:15 < toyz> and sorry for offtopic :) 2013-02-06T19:37:06 < jpa-> people hydraylics.. mmm 2013-02-06T19:38:26 < emeb> soylent hydraulics is peeeepol.. 2013-02-06T19:38:50 < dirty_d> lol 2013-02-06T19:39:16 < dirty_d> this is pretty neat, im programming my stm32 at home from work 2013-02-06T19:39:30 < dirty_d> usbreset was the missing link 2013-02-06T19:39:58 < Thorn> wrong gpio config -> short circuit -> overheating -> fire -> nobody at home 2013-02-06T19:39:58 < toyz> any other place i could find any hydraulics ? 2013-02-06T19:40:11 < dirty_d> Thorn, nothing is hooked up, lol 2013-02-06T19:40:14 < Thorn> toyz: why ask here of all things 2013-02-06T19:40:36 < toyz> Thorn: I LOVE YoU 2013-02-06T19:40:43 < toyz> i love the channel 2013-02-06T19:40:49 < Thorn> but I'm not into hydraulics 2013-02-06T19:40:56 < zyp> ooh, Thorn got a fan! 2013-02-06T19:41:04 < jpa-> hydraulics are into you 2013-02-06T19:41:05 -!- toyz is now known as Ranewen 2013-02-06T19:41:27 -!- Ranewen is now known as Lionhearted 2013-02-06T19:42:19 < Lionhearted> i got mine registerd nick,, now will try to post on some more places 2013-02-06T19:43:45 < emeb> jpa-: twss 2013-02-06T19:52:28 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Quit: Leaving.] 2013-02-06T19:52:40 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-46.pptp.stw-bonn.de] has joined ##stm32 2013-02-06T19:52:45 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-06T19:54:21 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T19:54:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-06T19:57:36 -!- dekar_ [~dekar@212.255.41.49] has quit [Quit: Leaving] 2013-02-06T20:05:11 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving] 2013-02-06T20:05:27 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:07:28 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-06T20:08:07 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:08:24 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:10:09 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:10:22 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:10:51 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 245 seconds] 2013-02-06T20:11:19 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:11:34 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:12:21 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:12:34 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:12:50 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:13:06 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:13:42 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:13:56 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:15:19 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:15:33 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:15:52 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-06T20:16:07 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:23:31 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving] 2013-02-06T20:23:47 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-06T20:44:00 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 252 seconds] 2013-02-06T20:54:13 -!- Lionhearted [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has quit [Ping timeout: 245 seconds] 2013-02-06T20:55:10 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/] 2013-02-06T20:55:33 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has joined ##stm32 2013-02-06T21:23:35 -!- Laurenceb_ [~Laurence@host86-177-214-70.range86-177.btcentralplus.com] has joined ##stm32 2013-02-06T21:33:37 < Laurenceb_> lol 2013-02-06T21:34:00 < Laurenceb_> owned of my pregnant cat got in touch 2013-02-06T21:34:05 < Laurenceb_> *owner 2013-02-06T21:34:19 < Laurenceb_> autocorrect knows 1337?! 2013-02-06T21:34:34 < Laurenceb_> it was an a shipping container of parts from germany 2013-02-06T21:34:48 < Erlkoenig> german parts are best :P 2013-02-06T21:35:09 < emeb> wait - the cat came from germany? 2013-02-06T21:35:18 < Thorn> the cat was in the container? 2013-02-06T21:35:24 < Thorn> or the owner? 2013-02-06T21:35:27 < zyp> how did you find the owner? cat was tagged? 2013-02-06T21:35:29 < Laurenceb_> emeb: yes 2013-02-06T21:35:40 < Laurenceb_> cat was in the container 2013-02-06T21:35:52 < emeb> wow! and they figured out it was you. 2013-02-06T21:35:57 < Laurenceb_> someone at JCB factory rang me, and we rang the factory in Germany 2013-02-06T21:36:15 < emeb> So can you send it back? 2013-02-06T21:36:18 < Laurenceb_> someone in Germany had contacted the german factory to ask if they had seen their cat 2013-02-06T21:36:19 < Laurenceb_> yes 2013-02-06T21:36:32 < emeb> Poor confused kitty. 2013-02-06T21:36:59 < emeb> What are the quarantine regs for UK/DE animal transport? 2013-02-06T21:37:32 < Laurenceb_> dunno, ive given it to RSPCA 2013-02-06T21:37:47 < Erlkoenig> ship it back "bad parts" 2013-02-06T21:37:52 < Laurenceb_> lol 2013-02-06T21:38:12 < Laurenceb_> looks like it jumped inside a container of gearboxes 2013-02-06T21:38:58 < zyp> «Ordered gearboxes, received cat. F-- would not buy again.» 2013-02-06T21:40:29 < emeb> worse than "ordered gopro, received socks/sd card..." 2013-02-06T21:41:30 < zyp> :) 2013-02-06T21:42:01 < zyp> I don't mind when the sd card was free :p 2013-02-06T21:42:11 < emeb> getting a refund? 2013-02-06T21:42:26 < zyp> got it weeks ago 2013-02-06T21:42:46 < zyp> that's why I was so surprised when something showed up anyway 2013-02-06T21:42:51 < emeb> got refund before you got the pkg? how's that? 2013-02-06T21:43:01 < Laurenceb_> i worked out how to make bladerf not suck 2013-02-06T21:43:15 < Laurenceb_> rf out -> linear tech mixer -> SAW filter -> rf input 2013-02-06T21:43:33 < emeb> right - build your own frontend. 2013-02-06T21:43:46 < Laurenceb_> lol yup 2013-02-06T21:43:52 < emeb> which you'd probably want to do anyway, since it can't go < 300MHz 2013-02-06T21:43:55 < zyp> emeb, I didn't care about waiting and just opened a case as soon as they advised me that the auction was pulled 2013-02-06T21:44:05 < zyp> got refunded almost immediately 2013-02-06T21:44:12 < zyp> no questions asked 2013-02-06T21:44:13 < emeb> zyp: makes sense. 2013-02-06T21:45:42 < zyp> ah, comment from ebay was «We didn't receive valid tracking information from the seller. We're sorry you had a problem with your purchase, and we're issuing you a refund for this case.» 2013-02-06T21:49:33 -!- toyz [c1c6a872@gateway/web/freenode/ip.193.198.168.114] has joined ##stm32 2013-02-06T21:49:41 < toyz> gday 2013-02-06T21:49:47 < toyz> can you forward me to a pc/laptop channel ? 2013-02-06T21:50:00 < zyp> try #pc and #laptop 2013-02-06T21:50:08 < upgrdman> ## 2013-02-06T21:50:31 < toyz> ty 2013-02-06T21:50:44 < toyz> wont work 2013-02-06T21:50:54 < zyp> strange 2013-02-06T21:59:11 <+Steffanx> Those channels dont exist. Can one 'create new' channgels without a registered nick? 2013-02-06T21:59:38 <+Steffanx> Some channels don't accept users without a registered nick anyway 2013-02-06T22:00:03 -!- toyz [c1c6a872@gateway/web/freenode/ip.193.198.168.114] has quit [Ping timeout: 245 seconds] 2013-02-06T22:00:07 <+Steffanx> hmpf 2013-02-06T22:00:58 -!- mrcan_ [~mrcan@78.173.255.25] has joined ##stm32 2013-02-06T22:00:58 -!- mrcan_ [~mrcan@78.173.255.25] has quit [Changing host] 2013-02-06T22:00:58 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-06T22:01:10 < emeb> we're so mean. 2013-02-06T22:01:48 < Laurenceb_> he was trolling 2013-02-06T22:01:56 < Thorn> you can create channels with /join 2013-02-06T22:03:53 < gxti> i don't feel bad 2013-02-06T22:03:57 <+Steffanx> LOL i wont 2013-02-06T22:04:00 < gxti> i've never seen him before and he's come in here twice today asking about random shit 2013-02-06T22:04:04 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 272 seconds] 2013-02-06T22:04:16 < zyp> I think he used to hang around here before 2013-02-06T22:04:21 -!- blablablabla [524834a0@gateway/web/freenode/ip.82.72.52.160] has joined ##stm32 2013-02-06T22:04:39 <+Steffanx> Oh, yes. It works :) 2013-02-06T22:05:21 <+Steffanx> but not really 2013-02-06T22:05:22 <+Steffanx> #pc 2013-02-06T22:05:30 < Laurenceb_> politically correct 2013-02-06T22:05:40 < Laurenceb_> where they inspect github code 2013-02-06T22:05:54 < Laurenceb_> and PUNISH YOU if your code is incorrect 2013-02-06T22:06:12 < gxti> fwiw the answer is probably ##hardware 2013-02-06T22:06:17 <+Steffanx> Only when they also punish for unreadable code/too few spaces 2013-02-06T22:06:28 < gxti> no Steffanx that is encouraged, it is 'agile' 2013-02-06T22:06:39 < gxti> there's no time for mantainability 2013-02-06T22:06:45 < Laurenceb_> agile development 2013-02-06T22:06:49 * blablablabla 2013-02-06T22:07:07 < zyp> is it politically correct to use norwegian names of fish as names for my github projects, or would I get in trouble for that? 2013-02-06T22:07:13 < Laurenceb_> wtf 2013-02-06T22:07:19 < Laurenceb_> * blablablabla 2013-02-06T22:07:19 < gxti> norwegian? you sick bastard 2013-02-06T22:07:22 < Laurenceb_> how you do that 2013-02-06T22:07:24 * gxti 2013-02-06T22:07:30 < blablablabla> I love you too Laurenceb_ 2013-02-06T22:07:35 * zyp 2013-02-06T22:07:38 <+Steffanx> 2013-02-06T22:07:41 < gxti> fail 2013-02-06T22:07:44 < Laurenceb_> lolz 2013-02-06T22:07:56 < Laurenceb_> oh i didnt spot you blablablabla 2013-02-06T22:08:10 <+Steffanx> my client is smart enough to see there is nothing behind /me 2013-02-06T22:08:23 < Laurenceb_> analrapist 2013-02-06T22:08:24 -!- neuro_sys [~neuro_sys@unaffiliated/neurosys/x-283974] has joined ##stm32 2013-02-06T22:08:32 < neuro_sys> love it love it love it 2013-02-06T22:08:40 < blablablabla> Love what? 2013-02-06T22:09:25 -!- blablablabla [524834a0@gateway/web/freenode/ip.82.72.52.160] has quit [] 2013-02-06T22:09:26 < Laurenceb_> psychotherapist 2013-02-06T22:16:14 < Thorn> Psycho the rapist? 2013-02-06T22:19:27 <+Steffanx> No, psychotherapist 2013-02-06T22:20:28 -!- neuro_sys is now known as neuro_sys|off 2013-02-06T22:25:11 < Thorn> I just discovered that there's PDF documentation for the SPL 2013-02-06T22:25:20 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-06T22:26:26 < Thorn> Thorn, ianal, but i'd say yes it fits the "system libraries" definition. <-- from #gpl 2013-02-06T22:26:37 < Thorn> er, #gnu 2013-02-06T22:30:17 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-06T22:38:07 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-06T22:41:15 < zyp> yay, somebody agreed with me 2013-02-06T22:42:01 < Thorn> the discussion is ongoing and it doesn't look very pretty 2013-02-06T22:42:24 < emeb> gnu discussions are never pretty 2013-02-06T22:43:46 < jpa-> is anyone eating his foot yet? 2013-02-06T22:45:40 < gxti> if there's one thing we can be certain of, it is that avoiding GNU solves the problem entirely 2013-02-06T22:46:01 < gxti> unfortunately i've acclimated to chibios so i'm not free of it :| 2013-02-06T22:46:03 < emeb> sadly, this is true. 2013-02-06T22:46:51 < jpa-> on the other hand the ST libraries are not magic and aren't that awesome either 2013-02-06T22:47:05 < jpa-> the best part of them is that they are "standard" 2013-02-06T22:47:07 < Erlkoenig> BSD License ftw... 2013-02-06T22:47:21 < jpa-> i wish libopencm3 wasn't lgpl shit 2013-02-06T22:48:26 < jpa-> but anyway, laks is the way to go :P 2013-02-06T22:49:10 < emeb> next library will be "lutefisk" 2013-02-06T22:49:22 < gxti> gross 2013-02-06T22:50:04 < emeb> he said norwegian fish... 2013-02-06T22:50:13 < Thorn> ST libraries work, which for things like USB is an achievement 2013-02-06T22:50:31 < jpa-> that is true 2013-02-06T22:50:57 < gxti> i need to rip out the code i copypastad to get my bootloader working 2013-02-06T22:50:57 < jpa-> though i'm still utterly confused by how their USB stuff manages to work 2013-02-06T22:51:02 < Laurenceb_> lol 2013-02-06T22:51:07 < Laurenceb_> it does.... 2013-02-06T22:51:11 < gxti> i already rewrote the ihex parser so i could feed data to it instead of having it pull 2013-02-06T22:51:12 < Laurenceb_> leave it at that 2013-02-06T22:51:22 < gxti> the rest should be easy 2013-02-06T22:51:32 < Thorn> jpa-: India has a huge population 2013-02-06T22:52:03 < jpa-> because when i try to compare it to other non-working implementations the st one seems to be furthest from how it is supposed to be :) 2013-02-06T22:52:33 < Thorn> #gnu discussion so far http://pastebin.com/KVRnttA9 2013-02-06T22:52:39 < Thorn> zyp: ^^ 2013-02-06T22:52:57 < Laurenceb_> im not going to read 2013-02-06T22:53:02 < Laurenceb_> my head will assplode 2013-02-06T22:54:32 < jpa-> it's just bing 2013-02-06T22:54:33 < jpa-> he is ok 2013-02-06T22:58:35 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Read error: Connection reset by peer] 2013-02-06T23:01:50 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-06T23:02:09 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Read error: Connection reset by peer] 2013-02-06T23:04:15 < Erlkoenig> Thorn: looks like someone doesn't have a good understanding of mikrocontrollers... 2013-02-06T23:15:39 < Thorn> embeddeve development 101 in progress 2013-02-06T23:15:45 < Thorn> *embedded 2013-02-06T23:16:28 -!- daku is now known as DaKu 2013-02-06T23:20:56 < dirty_d> why the hell couldnt st follow a sensible naming scheme for all these register bitmasks and such 2013-02-06T23:22:50 -!- barthess [~barthess@178.154.65.67] has quit [Quit: Leaving.] 2013-02-06T23:22:53 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-06T23:25:29 < dirty_d> GPIO_LCKR_LCK0 and then GPIO_BSRR_BR_0 2013-02-06T23:25:48 < dirty_d> some men just want to watch the world burn 2013-02-06T23:26:34 < Erlkoenig> this at least means that it's not autogenerated 2013-02-06T23:27:48 < dirty_d> it should be 2013-02-06T23:28:49 < gxti> the multi-bit ones are crap anyway, nobody wants to do FOOBAR_0 | FOOBAR_2 oh yeah obviously that means "hurrdurr mode" 2013-02-06T23:30:12 < dirty_d> i know 2013-02-06T23:30:22 < dirty_d> im trying to filter this all into a more sensible format 2013-02-06T23:30:30 < Erlkoenig> what really 2013-02-06T23:30:38 < Thorn> anyone happens to know who Bradley Kuhn is? 2013-02-06T23:30:39 < Erlkoenig> that would be neat 2013-02-06T23:33:06 < dirty_d> yea 2013-02-06T23:33:12 < dirty_d> time to gtfo 2013-02-06T23:33:14 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-06T23:55:22 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 --- Day changed Thu Feb 07 2013 2013-02-07T00:04:41 -!- Ranewen [~Ranewen4@87.252.128.182] has joined ##stm32 2013-02-07T00:15:02 < dongs> dont need resistors on usb? rly? 2013-02-07T00:16:48 < Laurenceb_> yup 2013-02-07T00:16:49 < Laurenceb_> wtf 2013-02-07T00:17:11 < Laurenceb_> why am i being invited to beyonce concerts 2013-02-07T00:17:36 < emeb> someone's got your number 2013-02-07T00:17:53 < Laurenceb_> yup 2013-02-07T00:17:59 < emeb> secret shame - beyonce fan. 2013-02-07T00:18:06 < Laurenceb_> ive found my way only spam central 2013-02-07T00:18:10 < Laurenceb_> *onto 2013-02-07T00:19:11 -!- scrts [~quassel@unaffiliated/scrts] has quit [Quit: No Ping reply in 180 seconds.] 2013-02-07T00:20:03 * Laurenceb_ is making custom adhesives 2013-02-07T00:20:13 < Laurenceb_> http://www.sigmaaldrich.com/catalog/product/sial/319902?lang=en®ion=GB 2013-02-07T00:20:21 < Laurenceb_> i like how you can order 200L drums 2013-02-07T00:21:23 < qyx_> dont adhese yourself to something 2013-02-07T00:21:41 < Laurenceb_> or firebomb anywhere :P 2013-02-07T00:22:28 < emeb> here you go: http://www.amazon.com/Passion-Natural-Water-Based-Lubricant-Gallon/dp/B005MR3IVO/ref=sr_1_11?ie=UTF8&qid=1360189331&sr=8-11&keywords=bulk+personal+lube 2013-02-07T00:22:30 < Laurenceb_> silicone resin bars dissolved in that stuff, then spray with air srayer thingy 2013-02-07T00:22:35 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32 2013-02-07T00:22:37 < dongs> emeb is blaxter trolling 2013-02-07T00:22:56 < dongs> did you have inline resistors on f3 breakout? 2013-02-07T00:22:57 < Laurenceb_> lolzors 2013-02-07T00:23:17 < dongs> on usb 2013-02-07T00:23:49 < zyp> dongs, I think datasheet says they are not required 2013-02-07T00:24:20 < dongs> mhm. is that f3 specific cause f1 def needed them 2013-02-07T00:25:03 < Laurenceb_> yes 2013-02-07T00:25:25 < dongs> also not like it matters but id rather not load 22r resistors if i dont have to 2013-02-07T00:25:37 < dongs> ok will read ds/rm 2013-02-07T00:26:34 < dongs> less parts to assemble the better 2013-02-07T00:27:24 < zyp> lpc43xx is fun in that aspect, USB0 doesn't require external resistors, USB1 does 2013-02-07T00:27:27 < zyp> :p 2013-02-07T00:27:51 < dongs> oh and ic next to so8 is just xtal. 2013-02-07T00:28:17 < dongs> and which usb did you connect on breakout? 2013-02-07T00:28:50 < zyp> USB0 of course, it's the HS one 2013-02-07T00:28:58 < dongs> orite 2013-02-07T00:29:25 < zyp> USB1 is also HS with ULPI, but only FS with internal phy 2013-02-07T00:30:12 < zyp> so difference in resistors is caused by different internal phys 2013-02-07T00:30:19 < dongs> ah. 2013-02-07T00:30:54 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-07T00:37:16 < Thorn> btw iirc lpc17xx needs 33R rather than 22R for USB like F1 and most(?) other controllers, I wonder why the difference 2013-02-07T00:38:07 < dongs> i used 33R on usb before,, forget with what though 2013-02-07T00:47:02 < dongs> i suppose i could just look at f3disco schematics. 2013-02-07T00:48:27 < dongs> um.. 2013-02-07T00:48:30 < dongs> tehy have 22R there. 2013-02-07T00:51:32 < zyp> page 100 in f303 datasheet 2013-02-07T00:51:38 < zyp> «No external termination series resistors are required on USBDP (D+) and USBDM (D-), the matching impedance is already included in the embedded driver.» 2013-02-07T00:51:50 < dongs> i was close, i was on 103 2013-02-07T00:52:03 < dongs> huh 2013-02-07T00:52:06 < dongs> which datasheet are you looking at 2013-02-07T00:52:11 < dongs> i'm on F303CCxx 2013-02-07T00:52:24 < dongs> but I have rev5 document 2013-02-07T00:52:29 < zyp> «Doc ID 023353 Rev 3» 2013-02-07T00:52:37 < zyp> i.e. the one I already had on disk 2013-02-07T00:52:38 < dongs> there it is. 2013-02-07T00:52:39 < dongs> its on 104 2013-02-07T00:52:53 < dongs> i was almost there, i was reading usb electrical specifications 2013-02-07T00:53:02 < dongs> hmm 2013-02-07T00:53:07 < dongs> well im gonna trash them then 2013-02-07T00:53:14 < dongs> less shit to swap in dicknplace 2013-02-07T00:54:30 < zyp> ok, sleeptime 2013-02-07T00:56:53 <+Steffanx> noo 2013-02-07T00:57:23 -!- DaKu is now known as daku 2013-02-07T00:59:37 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-07T01:00:08 < Laurenceb_> only trolls to talk to now 2013-02-07T01:00:45 < dongs> yep 2013-02-07T01:00:47 < dongs> resistors removed. 2013-02-07T01:05:00 < emeb> trollio iglesias 2013-02-07T01:05:29 < Laurenceb_> http://olimex.files.wordpress.com/2013/01/a10s-2.jpg 2013-02-07T01:05:36 < Laurenceb_> raspberry pi got pwned 2013-02-07T01:06:01 < emeb> not unless Oli is charging "$35" for that. 2013-02-07T01:06:25 < emeb> scare quotes because no one actually bought an rpi for $35. 2013-02-07T01:06:34 < dongs> right lol 2013-02-07T01:06:37 < emeb> but that's the price everyone remembers 2013-02-07T01:06:52 < dongs> they bought the aidsfruit RASPBERRYPI INVENTOR'S KIT 2013-02-07T01:06:54 < dongs> for liek $299 2013-02-07T01:07:04 < dongs> and got a fREE raspberry pi with that 2013-02-07T01:07:13 < emeb> lulzors 2013-02-07T01:08:52 < emeb> element14 quotes $35, but never has stock and charges like $25 S/H. 2013-02-07T01:10:48 < qyx_> interesting placement of decoupling caps 2013-02-07T01:10:50 -!- Ranewen [~Ranewen4@87.252.128.182] has quit [Quit: Leaving] 2013-02-07T01:10:52 < qyx_> on that olimex board 2013-02-07T01:11:09 < emeb> hah - hadn't noticed that. 2013-02-07T01:11:24 < emeb> totally legit! lump em all together like that. 2013-02-07T01:13:03 < emeb> bet I know why - spec prolly says 'one cap per vdd', but it's a BGA and there's no room close to the pkg. 2013-02-07T01:13:19 < emeb> plus they didn't want to put any parts on the back: http://olimex.files.wordpress.com/2013/01/a10s-3.jpg?w=580 2013-02-07T01:13:26 < Laurenceb_> looks a bit dubious 2013-02-07T01:13:44 < dirty_d> well i sort of sanitized st's bitmask defines 2013-02-07T01:13:46 < emeb> understatement ^^^ 2013-02-07T01:13:55 < dirty_d> a program can only do so much though 2013-02-07T01:14:01 < dirty_d> pesky humans 2013-02-07T01:16:04 < dongs> k sending that pcb off 2013-02-07T01:16:20 < Laurenceb_> magno too close to usb 2013-02-07T01:16:33 < Laurenceb_> /pedantic troll 2013-02-07T01:16:36 < dongs> pfft 2013-02-07T01:16:44 < dongs> it was even closer in last revision 2013-02-07T01:16:46 < dongs> and worekd fine 2013-02-07T01:17:07 < Laurenceb_> guess just 6 point cal will work relatively well to take out the effect 2013-02-07T01:17:20 < emeb> you spin me 'round. 6x 2013-02-07T01:17:44 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Quit: Leaving] 2013-02-07T01:19:39 < dongs> aw i forgot the layer mark display 2013-02-07T01:19:46 < dongs> lets see if i have somehere to put it 2013-02-07T01:25:27 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32 2013-02-07T01:32:07 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Remote host closed the connection] 2013-02-07T01:33:59 < dirty_d> tis done! http://codepad.org/jPcMwj5K 2013-02-07T01:34:20 < dirty_d> ran it through my program 2013-02-07T01:34:47 < Erlkoenig> dirty_d: you read in the register & bit def's? so you could provide a more OOP-like interface with bitfieds and structs and stuff 2013-02-07T01:36:10 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32 2013-02-07T01:36:45 < dirty_d> hmm, it got cut off before the end 2013-02-07T01:36:49 < dirty_d> its like 6000 lines long 2013-02-07T01:36:59 < dirty_d> Erlkoenig, i could 2013-02-07T01:37:23 < dirty_d> but bitfields arent good for some things 2013-02-07T01:37:33 < dirty_d> sometimes you need to use bitwise operations 2013-02-07T01:37:38 < dirty_d> i.e. the usb registers 2013-02-07T01:37:38 < Erlkoenig> one could provide both ;) 2013-02-07T01:37:49 < dirty_d> some bits are very wacky 2013-02-07T01:38:03 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Remote host closed the connection] 2013-02-07T01:38:03 < dirty_d> like writing 1 doesnt change the value, or it toggles the bit etc 2013-02-07T01:38:41 < dirty_d> Erlkoenig, the thing thats really missing are the bit group comfiguration values 2013-02-07T01:38:58 < dirty_d> they should have defined things like GPIO_OSPEEDR_OSPEED_2MHZ 2013-02-07T01:38:59 < dirty_d> etc 2013-02-07T01:39:15 < dirty_d> instead of having to look up and remember what bits need to be set for each speed 2013-02-07T01:39:27 < Erlkoenig> yeah indeed that would be good to have 2013-02-07T01:41:00 < Erlkoenig> what i also find problematic about the stdperiph... if i want my code to pass a GPIO pin as a parameter to some function, i need to specify register adress (like GPIOD), pin number (0-15), and hat RCC constant so the called code can actually do something with that pin. this is very annoying, it would be neat to have a data structure that encapsulates that all, so i can write MyClassInstance.useThisPin(GPIOD_14) or something... 2013-02-07T01:41:17 < Erlkoenig> similar things probably also apply to other periphals 2013-02-07T01:41:24 < dirty_d> yea definitely 2013-02-07T01:42:14 < Erlkoenig> or .useThisTimer(TIM4); ... ;-) 2013-02-07T01:43:20 < dirty_d> yea 2013-02-07T01:44:18 < dirty_d> its hard to make stuff like that convinient 2013-02-07T01:44:43 < dirty_d> i havent ever tried to make anything portable on a uc 2013-02-07T01:44:51 < Erlkoenig> no, but to make it convenient AND efficient ;-) 2013-02-07T01:44:57 < dirty_d> right 2013-02-07T01:45:30 < Erlkoenig> one could make excessive usage of const structures that get optimized away, or something 2013-02-07T01:45:50 < dirty_d> im sure you could if you use c++11 2013-02-07T01:46:17 < dirty_d> it would be a hell of a lot of work though 2013-02-07T01:46:30 < dirty_d> probably more than just hardcoding it to work on every uc it has to 2013-02-07T01:46:51 < dirty_d> ive never cared about my code working on any other chip than the one im using 2013-02-07T01:46:59 < Erlkoenig> well this wouldn't mainly be for portability but rather for prettier code 2013-02-07T01:47:32 < dirty_d> the peripherals are pretty simple 2013-02-07T01:47:40 < dirty_d> the complexity is mostly in the algorithms etc 2013-02-07T01:47:43 < dirty_d> and that is portable 2013-02-07T01:47:58 < dirty_d> so i just dont worry about the peripheral crap 2013-02-07T01:48:32 < dirty_d> atmel did really good with their xmegas 2013-02-07T01:49:14 < dirty_d> code written for one will pretty much just work on another if it has all the required peripherals 2013-02-07T01:50:23 < Erlkoenig> hmm yes that would be very good 2013-02-07T01:50:32 < Erlkoenig> lets unite all fools here and write such a super-F4-library \o/ 2013-02-07T01:51:40 < upgrdman> lol 2013-02-07T01:52:50 < upgrdman> when i got my first f4 it was my first uc after the arduinos... i started to write a lib that mimicked the arduino library (digitalWrite() etc...) but said fuck it after I realized how limited the arduino stuff is 2013-02-07T01:54:18 < upgrdman> i looked into opencm3, and it had promise but was far from complete :( 2013-02-07T01:56:05 < dirty_d> time to get some drinks 2013-02-07T01:56:07 < dirty_d> i dunno 2013-02-07T01:56:20 < dirty_d> the only real way to make stuff portable is a lot of manual labor 2013-02-07T01:56:34 < dirty_d> later 2013-02-07T01:56:47 < Erlkoenig> doesn't the ChibiOS provide something like that? 2013-02-07T01:56:49 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Quit: Leaving] 2013-02-07T01:57:42 < upgrdman> when i need uc code to be portable i wrap the implementation-specific code in functions or #defines... easy. 2013-02-07T02:01:04 < Erlkoenig> yuck #defines 2013-02-07T02:01:48 < upgrdman> embrace them! 2013-02-07T02:01:53 < upgrdman> :) 2013-02-07T02:02:41 < upgrdman> preprocessor directives can be abused, which is bad, but i've never understood why some people flat out hate them 2013-02-07T02:03:24 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32 2013-02-07T02:03:32 -!- Laurenceb_ [~Laurence@host86-177-214-70.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-07T02:04:16 < Erlkoenig> yeah of course like with many things, you can use them for good and for evil... and i must admit i have used them for evil some times... 2013-02-07T02:04:48 < upgrdman> "with great power comes great responsibility" blah blad ;) 2013-02-07T02:06:20 < upgrdman> #ifdef's are probably my favorite directive... 2013-02-07T02:20:05 < Erlkoenig> http://www.youtube.com/watch?v=qJe3cdM7f1c 2013-02-07T02:25:28 < upgrdman> wtf 2013-02-07T02:26:11 < Erlkoenig> pretty strong :D 2013-02-07T02:27:16 < upgrdman> i don't understand german, but its seems like a weird song 2013-02-07T02:27:38 < Erlkoenig> yeah, pretty weird :D its about a traditional meal... 2013-02-07T02:28:46 <+Steffanx> Time to get some sleep Erlkoenig 2013-02-07T02:28:57 < Erlkoenig> but ima not tired :o 2013-02-07T02:30:16 < upgrdman> you guys sleep together ?! :o 2013-02-07T02:30:44 < Erlkoenig> sure he likes me to hold him in my arms 2013-02-07T02:30:55 < upgrdman> mmmm spooning 2013-02-07T02:31:09 < upgrdman> Steffanx likes to be the small spoon? lol 2013-02-07T02:31:28 <+Steffanx> Nah, some here really believed he was a she 2013-02-07T02:31:34 <+Steffanx> but he told us he wasn't a she 2013-02-07T02:31:46 <+Steffanx> So now i don't like to be in her/his arms anymore 2013-02-07T02:31:49 < upgrdman> a she named erl? 2013-02-07T02:31:52 < Erlkoenig> what? my camouflage failed 2013-02-07T02:31:54 <+Steffanx> Yes 2013-02-07T02:32:10 < Erlkoenig> does "erl" mean something in english? :o 2013-02-07T02:32:16 <+Steffanx> It means Erlkoenig 2013-02-07T02:32:21 <+Steffanx> without the tab 2013-02-07T02:32:24 < Erlkoenig> ah :D 2013-02-07T02:32:44 <+Steffanx> Anyway, will see you in dreamland? 2013-02-07T02:32:45 < upgrdman> Erlkoenig, well Earl is a name ... i think with of euro decent? 2013-02-07T02:32:58 < upgrdman> is erlkoenig one name/word? 2013-02-07T02:33:16 < Erlkoenig> yes, the Erlkoenig is a ballad by Goethe and the name of a figure representing death in that ballad 2013-02-07T02:33:24 < upgrdman> oh 2013-02-07T02:33:27 < upgrdman> ok 2013-02-07T02:33:34 <+Steffanx> Erlkönig :P 2013-02-07T02:33:44 <+Steffanx> Fix your grammer 2013-02-07T02:33:44 < dongs> it has 'nig' in it so its funny. 2013-02-07T02:33:47 < Erlkoenig> hm yes i am uncreative 2013-02-07T02:33:59 <+Steffanx> donigs 2013-02-07T02:34:01 < upgrdman> so back to this spooning... did one of you "toot it and boot it"? http://www.youtube.com/watch?v=uWlMhiEL9Mc 2013-02-07T02:34:05 < Erlkoenig> IRC doesn't like 'ö' in nicknames... 2013-02-07T02:34:14 < upgrdman> UTF FTW! 2013-02-07T02:34:25 <+Steffanx> dongs, rename.. now 2013-02-07T02:34:26 < Erlkoenig> (01:34:01) upgrdman: so back to this spooning... did one of you "toot it and boot it"? http://www.youtube.com/watch?v=uWlMhiEL9Mc <-- "Unfortunately, this video is not available in Germany because it may contain music for which GEMA has not granted the respective music rights." >.< 2013-02-07T02:34:29 <+Steffanx> => donigs 2013-02-07T02:34:33 < dongs> haha 2013-02-07T02:34:34 < upgrdman> damn it 2013-02-07T02:34:41 -!- dongle is now known as donigs 2013-02-07T02:34:51 <+Steffanx> Now freenode will ban you 2013-02-07T02:34:58 < upgrdman> Erlkoenig, try googleing (or whatever you use) "yg - toot it and boot it" ... it a song by some black people 2013-02-07T02:35:00 < donigs> unlikely 2013-02-07T02:35:34 < Erlkoenig> http://www.dailymotion.com/video/xe4gsz_yg-toot-it-boot-it_music this? 2013-02-07T02:35:48 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-07T02:36:01 < upgrdman> yes 2013-02-07T02:36:17 < upgrdman> its a stupid/funny song 2013-02-07T02:36:21 < dongs> fucking dpboard stencil is late comeon piece of shit waws supposed to be here today 2013-02-07T02:36:35 < dongs> resisting urge to hand-solder this shit 2013-02-07T02:36:44 < Erlkoenig> hmm rap 2013-02-07T02:36:46 < dongs> wouldnt end up too well ;d 2013-02-07T02:37:16 < dongs> < Erlkoenig> hmm crap 2013-02-07T02:37:18 < dongs> ^ i saw this 2013-02-07T02:37:44 < Erlkoenig> that guy is too skinny 2013-02-07T02:37:46 -!- jef79m is now known as jef79m_lurking 2013-02-07T02:37:47 < upgrdman> well its not classy or anything... 2013-02-07T02:37:49 < upgrdman> ya 2013-02-07T02:38:17 < gxti> so many cuts in that video i'm getting nauseous 2013-02-07T02:38:21 < upgrdman> lol 2013-02-07T02:39:39 < Erlkoenig> now my neighbours will think i listen to rap... need to deliver them metal day&night 2013-02-07T02:41:07 < upgrdman> ever seen Supertroopers? there some supposedly german techno in it during a funny scene 2013-02-07T02:41:34 < Erlkoenig> hmm no never heard of 2013-02-07T02:42:17 < dongs> https://youtube.com/watch?v=uJrJ5FohmdI its this 2013-02-07T02:43:37 < upgrdman> https://www.youtube.com/watch?v=pWJQu12C2Ms 2013-02-07T02:44:39 < Erlkoenig> w.t.f. 2013-02-07T02:44:44 < upgrdman> :) 2013-02-07T02:45:18 < Erlkoenig> pretty good :D 2013-02-07T02:45:50 < upgrdman> the movie is pretty funny, you might torrent it if you don't mind english films 2013-02-07T02:46:01 < donigs> its onyl worth watching in hd 2013-02-07T02:46:15 < donigs> its out on bd, go rent it or somethign 2013-02-07T02:46:44 < Erlkoenig> hmm torrenting is not a good idea because this is a student dorm with supervised internet connection 2013-02-07T02:46:57 < upgrdman> :( 2013-02-07T02:47:36 < upgrdman> what kind of bandwidth (KB/MB per sec) do you get in a school dorm? 2013-02-07T02:47:43 < emeb> shouldn't that be dönigs? 2013-02-07T02:47:48 < Erlkoenig> 1 GBit 2013-02-07T02:47:52 < upgrdman> fuck 2013-02-07T02:47:58 < upgrdman> i hate my university 2013-02-07T02:48:04 < Erlkoenig> oh there is even a german version of that movie 2013-02-07T02:48:04 < gxti> emeb: irc is racist 2013-02-07T02:48:05 < Erlkoenig> harhar 2013-02-07T02:48:58 < upgrdman> i live in my own home, but at my university the bandwidth is like 50KBps, maybe 100KBps. i was like WTF?!?!?!?!?! even at home 15 - 30Mbps is common 2013-02-07T02:49:23 < donigs> upgrdman: thats because of al the cheap fucks like you who live on campus torrenting 2013-02-07T02:49:43 < upgrdman> donigs, read what i wrote you cock sucker. i dont live on campus 2013-02-07T02:49:50 < upgrdman> ;) 2013-02-07T02:50:01 < donigs> the living on campus wasnt addressed to you 2013-02-07T02:50:03 < donigs> the cheap fuck part was. 2013-02-07T02:50:16 < upgrdman> :) 2013-02-07T02:50:30 < Erlkoenig> yeah on campus our connection is not faster than yours, probably because the WLAN AP's are overloaded with 4chan and youtube traffic... but the dorms seem to have a better connection 2013-02-07T02:50:45 < upgrdman> ~$3000 per semester, and the fuckers at my uni charger $0.10 to print one page. 2013-02-07T02:50:52 < Erlkoenig> im germay dorms are not on the campus ;) 2013-02-07T02:51:13 < upgrdman> o 2013-02-07T02:51:40 < Erlkoenig> and the universitie's buildings are spread across the town... 2013-02-07T02:52:39 < upgrdman> public transit in the usa sucks :( 2013-02-07T02:52:56 < Erlkoenig> it sucks probably everywhere except Japan 2013-02-07T02:52:58 < upgrdman> if my uni buildings were strewn about it would be a pain in the ass 2013-02-07T02:53:40 < upgrdman> as it is, the parking sucks. walking at last 1/2 mile to most classes. and its all hills. 2013-02-07T02:54:08 < Erlkoenig> well as long as you study only one subject you won't have to go far... 2013-02-07T02:54:20 < upgrdman> and the douche bag campus designers put the library all the way in the corner, far from any parking. 1 mile walk... 2013-02-07T02:54:45 < Erlkoenig> genius... we have multiple libraries, every faculty has its own at their place 2013-02-07T02:54:51 < donigs> considering how average a merican is a fat fuck, I don'[t see how this is a problem 2013-02-07T02:54:54 < upgrdman> Erlkoenig, problem is that even classes related can be several buildings apart 2013-02-07T02:55:09 < donigs> 1 mile of walking is good for you 2013-02-07T02:55:13 < Erlkoenig> yeah... use a bike? :D 2013-02-07T02:55:15 < donigs> (wahtever the fuck 1 mile is, you ignorant american fuck) 2013-02-07T02:55:16 < upgrdman> donigs, true. i qualify as a fat fuck 2013-02-07T02:55:29 < upgrdman> donigs, maybe 1.4km 2013-02-07T02:55:40 < donigs> looks liek 1.6 2013-02-07T02:55:46 < upgrdman> k 2013-02-07T02:56:26 < upgrdman> some people bike or skateboard, but theres too many idiots for me to want to navigate through them 2013-02-07T02:56:39 < donigs> first world problems. 2013-02-07T02:56:46 < upgrdman> yes, very much so 2013-02-07T02:57:21 < upgrdman> i have another first world problem: i want to order pizza delivery, but im too lazy to walk over to my wallet so i can give them my card number 2013-02-07T02:57:58 < donigs> i dont doubt it 2013-02-07T02:59:00 < upgrdman> donigs, where are you? germany? 2013-02-07T02:59:04 < gxti> god help you when the pizza actually shows up 2013-02-07T02:59:12 < upgrdman> lol 2013-02-07T03:00:17 < upgrdman> one of these days, if i can overcome shame, i should answer the door in just boxers and a t-shirt 2013-02-07T03:00:37 < Erlkoenig> do you pay pizza with credit cards? duh 2013-02-07T03:00:54 < upgrdman> yes. i dont keep cash on hand 2013-02-07T03:01:02 < Erlkoenig> americans :D 2013-02-07T03:01:15 < upgrdman> and it a debit card, but close enough 2013-02-07T03:01:26 < upgrdman> Erlkoenig, im lazier and fatter than most. 2013-02-07T03:02:59 < upgrdman> its nice have an electronic record of all my purchased... and coins suck ass. 2013-02-07T03:03:30 < Erlkoenig> yeah but paying with card takes so much longer than paying cash... 2013-02-07T03:03:35 < upgrdman> no 2013-02-07T03:03:40 < upgrdman> quicker often 2013-02-07T03:03:42 < Erlkoenig> here it does, and it's annoying 2013-02-07T03:03:49 < upgrdman> how long does it take to auth a card? 2013-02-07T03:03:52 < Erlkoenig> the cashiers are much faster with their coins 2013-02-07T03:03:57 < Erlkoenig> dunno, a minute? 2013-02-07T03:03:58 < upgrdman> here its maybe 2 seconds at most 2013-02-07T03:04:01 < upgrdman> DAMN 2013-02-07T03:04:08 < Erlkoenig> including PIN entry and stuff 2013-02-07T03:04:27 < upgrdman> no need for pin. sign on the lcd and done. ~5 seconds total 2013-02-07T03:05:08 < Erlkoenig> ah with credit cards... nobody uses them here, only EC 2013-02-07T03:05:41 < Erlkoenig> the terminals here use analog modems and they tage ages for verification... 2013-02-07T03:05:46 < upgrdman> well all debit cards (except the ones for kids) can be used a credit cards... except theres no credit. it just pulls from your account 2013-02-07T03:05:54 < upgrdman> that sucks 2013-02-07T03:06:31 < Erlkoenig> well yeah 2013-02-07T03:06:53 < Erlkoenig> i dont even know if supermarkets here accept credit cards... only EC i think 2013-02-07T03:07:02 < upgrdman> ec == requires pin? 2013-02-07T03:07:38 < dongs> my cc "requires" pin 2013-02-07T03:07:41 < dongs> but i made it blank 2013-02-07T03:07:42 < dongs> fuck that shit. 2013-02-07T03:07:59 < Erlkoenig> yes... theoretically also signature, but nobody accepts that 2013-02-07T03:08:11 < upgrdman> pretty much everywhere here takes credit or debit cards. only exceptions i see are one-man-businesses, like a hot dog stand 2013-02-07T03:08:15 < Erlkoenig> EC directly instantly accesses your bank account 2013-02-07T03:08:21 < upgrdman> k 2013-02-07T03:08:32 < Erlkoenig> so you instantly get rid of your money 2013-02-07T03:08:49 < upgrdman> sure, but i dont run out. i know how to budget 2013-02-07T03:09:17 < upgrdman> which is rare among americans. i only two 3 people other than myself that don't live paycheck to paycheck 2013-02-07T03:09:28 < upgrdman> 3, not 2 2013-02-07T03:09:48 < Erlkoenig> duh... wtf 2013-02-07T03:09:58 < Erlkoenig> and people ask why there is/was a financial crysis 2013-02-07T03:10:09 < upgrdman> ya it pissed me off. 2013-02-07T03:10:51 < upgrdman> friends say they're out of cash... yet they have cable tv, a smartphone and other shit. how can any adult live paycheck to paycheck and waste money on shit like that 2013-02-07T03:11:22 < Erlkoenig> people... can't be understood :D 2013-02-07T03:11:33 < upgrdman> ya 2013-02-07T03:12:13 < upgrdman> one the plus side, its nice to freak people by dropping hundreds or a few thousand on a new toy without saving up... im always saving. its like a foreign concept to many people 2013-02-07T03:12:26 < Erlkoenig> i don't know about my friends financial issues but i never heard anyone say they had were out of money 2013-02-07T03:12:35 < Erlkoenig> ha great :D 2013-02-07T03:12:41 < Erlkoenig> yeah exactly 2013-02-07T03:12:54 < upgrdman> i bought a new tablet on a whim... stuff like that 2013-02-07T03:13:22 < dongs> useless. 2013-02-07T03:13:28 < dongs> why the fuck is this conversaion happening anyway 2013-02-07T03:13:28 < upgrdman> im planning to buy a new laptop after i sell my netbook that i havn't used in a while 2013-02-07T03:13:41 < upgrdman> dongs, why not. 2013-02-07T03:13:44 < dongs> no idea 2013-02-07T03:13:50 < upgrdman> feel free to bring up back on topic 2013-02-07T03:13:53 < dongs> it sounds weird, like Erlkoenig is a dickgirl or something. 2013-02-07T03:13:58 < gxti> better than talking about windows, gnu, or PWM capture 2013-02-07T03:14:09 < upgrdman> dickgirl? 2013-02-07T03:14:28 < dongs> my trap sense is tingling 2013-02-07T03:14:34 < Erlkoenig> i can't buy as much hardware as i can use... i still haven't taken full advantage of my F4 Disco 2013-02-07T03:14:59 < Erlkoenig> wait, the other way round 2013-02-07T03:15:07 < Erlkoenig> can't use as much hardware as i could buy 2013-02-07T03:15:28 < upgrdman> :) i made a servo tester with my f4disco board... its just an lcd + some buttons + one channel of pwm :) way overkills but who cares 2013-02-07T03:16:19 < gxti> one of these days i will knuckle down and start writing some crypto stuff for mine, but so far i ahve only used it as a SWD programmer. but for the price, i'm not complaining. 2013-02-07T03:16:45 < Erlkoenig> yeah nice, wanted to build such a thing too... but still have some ATmega laying around could use one of them for that... also have a supercheap alphanum LCD, but the cable is probably difficult to solder 2013-02-07T03:16:46 < upgrdman> ya. the f4disco is cheaper than the chinese arduino clones 2013-02-07T03:17:22 < upgrdman> Erlkoenig, i repair servos and other RC electronics, so i made a super-overkill servo tester. 2013-02-07T03:17:39 < Erlkoenig> oh neat 2013-02-07T03:17:42 < upgrdman> http://www.farrellf.com/projects/hardware/2012-11-23_Custom-Built_Servo_Tester/ 2013-02-07T03:18:38 < Erlkoenig> cool... well thats much more i was planning to build^^ 2013-02-07T03:20:46 < Erlkoenig> +than 2013-02-07T03:21:39 < upgrdman> i really only needed sweep mode, 2position mode, cycle limit and adjustable framerate, but i did the rest because i figure it will be the last time i need to make this tool. 2013-02-07T03:22:23 < dongs> shoulda used a touchscreen 2013-02-07T03:22:28 < dongs> not a shitty 2color lcvd 2013-02-07T03:22:48 < upgrdman> dongs, i have a 3.2" qvga touch lcd... but meh 2013-02-07T03:23:00 < upgrdman> b/w lcd was fine but it 2013-02-07T03:23:01 < upgrdman> for it 2013-02-07T03:23:05 < Erlkoenig> perfect would be an added potentiometer or other thing to measure the servos current angle... so one could measure the pulse width <=> angle function 2013-02-07T03:23:10 < upgrdman> and it was free, so the price was right 2013-02-07T03:23:20 < upgrdman> ya 2013-02-07T03:25:46 < dongs> upgrdman: can i run that shit headless if i dont have a lcd 2013-02-07T03:25:53 < dongs> o wait, what did you make the project in? 2013-02-07T03:25:55 < dongs> gcc or something? 2013-02-07T03:26:01 < upgrdman> yes 2013-02-07T03:26:20 < dongs> oh ok nvm prob too much work to make it work then 2013-02-07T03:26:33 < upgrdman> shouldn't be. i dont use any libs 2013-02-07T03:26:42 < upgrdman> only #include i didnt write is stm32f4xx.h 2013-02-07T03:27:06 < dongs> i just need shit to sweep full/\low 2013-02-07T03:27:09 < dongs> every 5 seconds 2013-02-07T03:27:49 < dongs> haha, no shit. nto even stdlib 2013-02-07T03:27:57 < dongs> importing into keil 2013-02-07T03:29:50 < upgrdman> i dont like stm's lib... so i dont use it 2013-02-07T03:30:03 < upgrdman> it makes my life harder sometimes, but this is just a hobby. 2013-02-07T03:30:06 < dongs> simple_gpio.h(18): warning: #1295-D: Deprecated declaration SysTick_Handler - give arg types 2013-02-07T03:30:09 < dongs> haha.. 2013-02-07T03:30:16 < upgrdman> ? :) 2013-02-07T03:30:19 < dongs> its not 1989 brto 2013-02-07T03:30:20 < dongs> bro 2013-02-07T03:30:35 < dongs> void shit_handler(); < stopped being relevant in 1985 2013-02-07T03:30:41 < gxti> you're the one using an asinine compiler dongs :p 2013-02-07T03:30:41 < dongs> void shit_handler(void); is the right way 2013-02-07T03:30:45 < gxti> maybe you're the one with the problem 2013-02-07T03:31:02 < dongs> 0b001???? 2013-02-07T03:31:05 < dongs> what the fuccccccccccck 2013-02-07T03:31:06 < upgrdman> wtf should i list void as the args... thats not required in c 2013-02-07T03:31:14 < dongs> upgrdman: newsflash, it is. 2013-02-07T03:31:26 < dongs> what garbage compiler allows 0bxxxx ??? 2013-02-07T03:31:30 < upgrdman> k&r says otherwise 2013-02-07T03:31:45 < upgrdman> dongs, thats standard syntax for radix2 2013-02-07T03:32:02 < dongs> upgrdman: not in any compiler except gcc 2013-02-07T03:32:46 < upgrdman> ok 2013-02-07T03:32:50 < upgrdman> well yay for gcc 2013-02-07T03:32:56 < upgrdman> wtf cant other support binary literals 2013-02-07T03:32:57 < gxti> \o/ 2013-02-07T03:33:15 < dongs> oh good fucking god 2013-02-07T03:33:20 < dongs> entire uc1608 shit is made with that 2013-02-07T03:33:24 < dongs> dude, wqha hte fuck 2013-02-07T03:33:24 < upgrdman> lol 2013-02-07T03:33:26 < upgrdman> yes 2013-02-07T03:33:30 < upgrdman> :) :) 2013-02-07T03:33:34 < gxti> the moral of the story is: don't give dongs your source code 2013-02-07T03:33:43 < gxti> because his ancient compiler can't handle it, and he wasn't going to say nice things anyway 2013-02-07T03:34:07 < upgrdman> i dont mind the criticism 2013-02-07T03:34:18 < upgrdman> im inexperienced and tough skinned 2013-02-07T03:34:32 < upgrdman> and i know he doesnt like linux or oss. too bad. 2013-02-07T03:34:43 < gxti> honestly i didn't even know about the (void) thing until i managed to accidentally turn on the warnings for it in gcc 2013-02-07T03:35:15 < gxti> except it was some baffling thing about seeing a definition when it was not previously declared, or something like that 2013-02-07T03:35:30 < dongs> gxti, () is C++ shit. 2013-02-07T03:35:35 < dongs> it must be (void) in C. 2013-02-07T03:35:39 < dongs> there's no ifs or buts about i t. 2013-02-07T03:35:45 < gxti> ok, well the only compiler i use doesn't give a shit 2013-02-07T03:35:53 < gxti> and it doesn't materially affect anything at all 2013-02-07T03:35:57 < dongs> thats grea,t good luck getting hired to places that use real compilers 2013-02-07T03:35:59 < dongs> it does. 2013-02-07T03:36:01 < dongs> the code looks like shit 2013-02-07T03:36:04 < dongs> (void) = no args 2013-02-07T03:36:09 < dongs> () = some faggot forgot to put args in 2013-02-07T03:36:12 < gxti> i don't do embedded for a living 2013-02-07T03:36:15 -!- jef79m_lurking is now known as jef79m 2013-02-07T03:36:17 < gxti> so i am continuing to not care 2013-02-07T03:36:37 < gxti> and code style is different everywhere anyway 2013-02-07T03:37:01 < upgrdman> lol 2013-02-07T03:37:04 < gxti> odds are some crotchety assclown will complain about indentation if nothing else 2013-02-07T03:37:21 < gxti> oh wait 2013-02-07T03:37:23 < dongs> servo.axf: Error: L6200E: Symbol micros multiply defined (by screens.o and main.o). 2013-02-07T03:37:24 < gxti> your asinine compiler does that too 2013-02-07T03:37:26 < dongs> o man 2013-02-07T03:37:32 < dongs> someone doesnt know how to declare prototypes/functions either? 2013-02-07T03:37:41 < upgrdman> i wrote that 16px font by hand before i found out about xbm... :o) 2013-02-07T03:37:52 < upgrdman> dongs, ? 2013-02-07T03:38:18 < dongs> volatile uint32_t micros; 2013-02-07T03:38:19 < dongs> ... 2013-02-07T03:38:22 < dongs> in HEADER file? 2013-02-07T03:38:24 < dongs> dude, wha the fuck? 2013-02-07T03:38:30 < gxti> haha, no excuses there 2013-02-07T03:38:39 < upgrdman> ok ya. that was fucked 2013-02-07T03:38:49 < dongs> surprised GAYCC didnt complain 2013-02-07T03:38:54 < gxti> it usually does. 2013-02-07T03:39:14 < upgrdman> should have put that in the .c 2013-02-07T03:39:27 < dongs> and there's d7/etc 2013-02-07T03:39:37 < upgrdman> d7? 2013-02-07T03:39:56 < dongs> enum GPIO_PIN d7, d6, d5, d4, d3, d2, d1, d0, cd, wr1; 2013-02-07T03:39:57 < dongs> ... 2013-02-07T03:39:58 < dongs> this 2013-02-07T03:39:59 < dongs> in header. 2013-02-07T03:40:04 < dongs> thas multiply included too 2013-02-07T03:40:12 < dongs> typedef that shiz. 2013-02-07T03:40:32 < gxti> honestly i'm surprised that you don't hate typedefs 2013-02-07T03:40:38 < gxti> it just seems more congruent with your overall attitude. 2013-02-07T03:41:51 < dongs> ok it builds 2013-02-07T03:41:57 < dongs> moved enum GPIO_PIN d7, d6, d5, d4, d3, d2, d1, d0, cd, wr1; 2013-02-07T03:41:59 < dongs> to the .c file 2013-02-07T03:42:49 < dongs> upgrdman: does it just run when pwoered on 2013-02-07T03:42:55 < upgrdman> yes 2013-02-07T03:42:55 < dongs> in two mode 2013-02-07T03:42:58 < dongs> as the sores says 2013-02-07T03:42:58 < dongs> kk 2013-02-07T03:42:59 < upgrdman> yes 2013-02-07T03:43:21 < dongs> A1 2013-02-07T03:43:25 < dongs> what pin is that?? 2013-02-07T03:43:40 < dongs> GPIOA_Pin1 or osmehting? 2013-02-07T03:43:50 < dongs> why the fuck not just write PA1 2013-02-07T03:43:53 < dongs> like normal people :d 2013-02-07T03:43:58 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-07T03:44:18 < upgrdman> p==pin... redundant. fuck that 2013-02-07T03:44:20 < R2COM> xc2c32a was available in quantity 410 yesterday in digikey, now its not, some asshole bought it probably already 2013-02-07T03:44:27 < dongs> haha 2013-02-07T03:44:29 < dongs> < 2013-02-07T03:44:57 < R2COM> seriously? you bought it? 2013-02-07T03:45:15 < dongs> just liek some fucker bought some other shit i needed after i blogged it in the channel :p 2013-02-07T03:46:04 < R2COM> ok well, I was about to blog today about cool SMPS power supply i found today for cheap price, and already analysed it, needs very little support elements too 2013-02-07T03:46:07 < R2COM> now not going to 2013-02-07T03:46:50 < dongs> :D 2013-02-07T03:47:10 < upgrdman> lineagetech? 2013-02-07T03:47:12 < dongs> i use RT8009 for general dc/dc stuff 2013-02-07T03:47:37 < dongs> its compatible with a few different big maker chips 2013-02-07T03:47:39 < dongs> pin/function 2013-02-07T03:47:45 < R2COM> oh, thats a shit banana part 2013-02-07T03:47:52 < gxti> "shit banana"? 2013-02-07T03:47:58 < gxti> never mind 2013-02-07T03:48:02 < gxti> don't want to know. 2013-02-07T03:48:49 < R2COM> I used semtech, aos, did not like them much 2013-02-07T03:49:00 < gxti> yes i am done with sc189, the foldback sucks 2013-02-07T03:49:18 < dongs> upgrdman: its not doing shit. 2013-02-07T03:49:22 < gxti> any brief overcurrent and it's stuck unless you have a supervisor to disconnect the load 2013-02-07T03:49:31 < upgrdman> dongs, sucks to be you :) 2013-02-07T03:49:39 < R2COM> gxti: what do you mean? 2013-02-07T03:49:41 < gxti> open sores strikes again 2013-02-07T03:49:43 < dongs> its in hardfault handler, of course. 2013-02-07T03:49:43 < R2COM> most of them have OCP 2013-02-07T03:49:44 < dongs> yep 2013-02-07T03:49:46 < upgrdman> dongs, want a .bin or .elf? 2013-02-07T03:49:52 < dongs> upgrdman: no of course not 2013-02-07T03:50:01 < gxti> R2COM: yes and once it triggers it will not exit foldback until load drops below a tiny value 2013-02-07T03:50:23 < gxti> practically speaking unless it's just a small mcu it will be stuck at brownout forever 2013-02-07T03:50:25 < R2COM> gxti: some parts I used could recover even after short 2013-02-07T03:50:29 < gxti> most can 2013-02-07T03:50:31 < gxti> but not this one 2013-02-07T03:50:44 < gxti> it's cheap and tiny and just needs an inductor and caps but this one thing blows chunks 2013-02-07T03:50:47 < R2COM> it shuts off power, and then poewrs again, what are you talking about why would it stuck?? 2013-02-07T03:50:57 < R2COM> it detects overcurrent, and shuts down power 2013-02-07T03:51:00 < gxti> sc189 does not shut off power, it lowers the voltage 2013-02-07T03:51:02 < gxti> rtfds 2013-02-07T03:51:19 < R2COM> you need to rtfds because you are talking bs 2013-02-07T03:51:25 < R2COM> I was not talking about sc189 2013-02-07T03:51:54 < gxti> and i was, which i clearly stated when i started ranting 2013-02-07T03:52:09 < dongs> trash is in hardfault as soo nas it jumps to main 2013-02-07T03:52:11 < upgrdman> dongs, well the logic to generate the servo pwm is pretty simple. theres a block in main.c that sets up the timer, and TIM5_IRQHandler() in interrupts.c does the logic 2013-02-07T03:52:26 < dongs> upgrdman: i know. 2013-02-07T03:52:29 < dongs> the problem is elsewhere. 2013-02-07T03:52:53 < R2COM> anyhow, going to try new ones soon.. hopefully that would be great one, I'm sick already of my PSU breakouts for testing 2013-02-07T03:53:29 < upgrdman> dongs, did you convert all the 0b's in main.c? 2013-02-07T03:53:39 < dongs> it wouldnbt build otherwise, so yes 2013-02-07T03:55:28 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-07T03:56:44 < dongs> ok its sweeping 2013-02-07T03:56:50 < dongs> FPU stuff wasnt handled 2013-02-07T03:56:53 < dongs> is why it was crashing 2013-02-07T04:00:44 < upgrdman> :) 2013-02-07T04:02:55 < upgrdman> if you're going headless you can just set the globals as needed and skip buttons altogether 2013-02-07T04:03:06 < upgrdman> volatile enum Mode currentMode = TWO_POS; 2013-02-07T04:03:16 < upgrdman> etc 2013-02-07T04:03:28 < dongs> shit's fucked, probabl wrong ranges 2013-02-07T04:03:33 < dongs> esc just beeps at it 2013-02-07T04:03:38 < dongs> gotta turn damn scope on again 2013-02-07T04:03:51 < upgrdman> u in 2pos mode? 2013-02-07T04:03:53 < dongs> yeah 2013-02-07T04:04:08 < upgrdman> is it a cheap esc? 2013-02-07T04:04:15 < upgrdman> it might want a slower framerate 2013-02-07T04:04:30 < upgrdman> try volatile uint16_t twoPos_rate = 15; 2013-02-07T04:05:49 < dongs> um no its 32hz now, ought to work 2013-02-07T04:05:54 < dongs> technically should be 50 2013-02-07T04:05:59 < upgrdman> also, some esc's will not arm unless they see a low-throttle signal for >1 second... so 2pos mode might be changing positions too quick for it to arm 2013-02-07T04:06:21 < upgrdman> dongs, theres no technically. the rc hobby has no standard 2013-02-07T04:06:32 < upgrdman> thats why i made the framerate adjustable 2013-02-07T04:07:36 < upgrdman> wait, 32hz? should be 100hz if you left rate at 10 (10ms)... 2013-02-07T04:07:37 < dongs> what hte fuck is this 2013-02-07T04:09:03 < upgrdman> if its 32hz im guessing theres a clock issue. is your f4 running at full speed? 2013-02-07T04:09:08 < upgrdman> whats wrong? 2013-02-07T04:09:17 < dongs> it is when I use normal code 2013-02-07T04:09:25 < dongs> did you fuck with anything in system.c? 2013-02-07T04:09:59 < upgrdman> i dont use a system.c 2013-02-07T04:10:07 < dongs> you do. 2013-02-07T04:10:08 < dongs> shit's in stm 2013-02-07T04:10:09 < dongs> ok works 2013-02-07T04:10:09 < upgrdman> -_- 2013-02-07T04:11:55 -!- bsdfox [~Bob@c-71-197-82-145.hsd1.ca.comcast.net] has joined ##stm32 2013-02-07T04:11:55 -!- bsdfox [~Bob@c-71-197-82-145.hsd1.ca.comcast.net] has quit [Changing host] 2013-02-07T04:11:55 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32 2013-02-07T04:14:05 < emeb> kumbayaa... 2013-02-07T04:14:06 < dongs> ok shiyt works good and makes things faster and i dont have to keep my rc on for an hour while doing these things 2013-02-07T04:14:34 < dongs> thanks dude 2013-02-07T04:34:07 < dongs> the time it holds at high point could be halved though. 2013-02-07T04:38:36 < dongs> ok fixed that 2013-02-07T04:45:08 < upgrdman> glad it worked out in the end 2013-02-07T04:45:25 < upgrdman> it was my best effort at a first-real-mcu-project. :) 2013-02-07T04:45:32 < upgrdman> i've got a ways to go 2013-02-07T04:45:54 < upgrdman> that reminds me, do people use javadoc comments in other languages? i kind of like the syntax 2013-02-07T04:46:52 < GargantuaSauce> doxygen is more common but javadoc is a thing people use too 2013-02-07T04:48:24 < GargantuaSauce> speaking of servos, check out this in all its busywaiting glory http://www.youtube.com/watch?v=1HPVbuaONO4 2013-02-07T04:52:59 < dirty_d> that a lot of servos 2013-02-07T04:53:07 < Erlkoenig> all signals generated by timers? 2013-02-07T04:53:28 < GargantuaSauce> just one timer is used, the four ICs on that protoboard are shift registers 2013-02-07T04:54:29 < dirty_d> why not just use more timers? 2013-02-07T04:54:40 < GargantuaSauce> there are not enough pwm channels :( 2013-02-07T04:54:43 < GargantuaSauce> believe me i wanted to 2013-02-07T04:54:49 < dirty_d> there arent? 2013-02-07T04:54:57 < Erlkoenig> aren'T there like 32 or so on an STM32F407? 2013-02-07T04:55:09 < dirty_d> i think i counted 16 servos 2013-02-07T04:55:20 < dirty_d> there are at least 4 4 channel timers no? 2013-02-07T04:55:25 < GargantuaSauce> well i didn't want to dedicate EVERY timer to servos, it has to do other stuff too 2013-02-07T04:55:32 < Erlkoenig> dirty_d: exactly... 2013-02-07T04:55:39 < dongs> GargantuaSauce: haha are you making a hexapod shite 2013-02-07T04:55:41 < GargantuaSauce> this is for a hexapod so i need 18 + more for grippers and sensors 2013-02-07T04:55:46 < GargantuaSauce> that i am 2013-02-07T04:55:53 < Erlkoenig> wow... 2013-02-07T04:56:00 < dirty_d> grippers? 2013-02-07T04:56:11 < GargantuaSauce> claws, arms, whatever you want to call em 2013-02-07T04:56:16 < dirty_d> thats scary 2013-02-07T04:57:12 < dirty_d> http://www.youtube.com/watch?v=0jyBiECoS3Q 2013-02-07T04:57:14 < dirty_d> no sleep for me 2013-02-07T04:57:16 < GargantuaSauce> and yes dongs i included the shot of the openocd console just 4 u 2013-02-07T04:57:34 < GargantuaSauce> ya zenta is my main inspiration for this shit 2013-02-07T05:03:21 < GargantuaSauce> i guess the reason for using shift registers was more for general gpio allocation than the pwm channels themselves come to think of it 2013-02-07T05:04:59 < GargantuaSauce> since i want to use the DCMI and foot switches and some sonar rangers and juggling all those ports was going to be a pain 2013-02-07T05:05:31 < dirty_d> how does it work with shift registers? 2013-02-07T05:05:36 < dirty_d> ive never used one 2013-02-07T05:06:52 < GargantuaSauce> they consist of a shift register and a storage register, serial data in fills the shift register and then you can strobe the storage register's clock to update it to the shift register's values 2013-02-07T05:06:56 < upgrdman> they're like a serial -> parallel converter 2013-02-07T05:07:28 < GargantuaSauce> so i just have the four shift registers chained....i push in 32 bits with SPI1 and then strobe the storage register when it comes time to turn off each servo in turn 2013-02-07T05:08:20 < dirty_d> hmm 2013-02-07T05:08:38 < dirty_d> so what kind of resolution are you getting each servo signal? 2013-02-07T05:08:50 < GargantuaSauce> a couple microseconds 2013-02-07T05:09:06 < GargantuaSauce> the spi runs happily at 42 MHz i think it was? 2013-02-07T05:09:17 < dirty_d> wow i didnt know spi could go that fast 2013-02-07T05:09:26 < upgrdman> ya, thats fast. damn 2013-02-07T05:09:31 < GargantuaSauce> SPI1 on an f4 can, it's hooked up to the faster ahb bus 2013-02-07T05:11:11 < GargantuaSauce> er apb 2013-02-07T05:13:27 < GargantuaSauce> ooh, apparently that's at least twice the rated clock speed of the 74hc595 at 3v 2013-02-07T05:13:32 < GargantuaSauce> i am now surprised it works at all 2013-02-07T05:15:39 -!- dekar [~dekar@212.255.41.49] has joined ##stm32 2013-02-07T05:15:43 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-07T05:25:40 < Erlkoenig> http://games.2g2s.de/git/?p=stm32;a=blob;f=src/hwcontrol/steer.cc;h=fc822df39a20485da0f4070281e3705aa60f740f;hb=refs/heads/master i made this TImer-Number,GPIO-Number-Agnostic Servo control class, which can then be instantiated like here http://games.2g2s.de/git/?p=stm32;a=blob;f=src/hwcontrol/sub_conf.cc;h=14af9c97fd8d5dc0aa63103205abe48eb43fc9e0;hb=refs/heads/master line 26 ... Actual values are set in line 68 2013-02-07T05:36:43 < dongs> 500 - Internal Server Error 2013-02-07T05:36:43 < dongs> syntax error at /etc/gitweb.conf line 14, near "$git_temp " 2013-02-07T05:36:44 < dongs> lol 2013-02-07T05:36:47 < dongs> lunix failurez 2013-02-07T05:47:21 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-07T05:55:23 -!- dekar [~dekar@212.255.41.49] has quit [Quit: Leaving] 2013-02-07T05:58:24 < R2COM> mmm TI or LTspice... 2013-02-07T05:58:43 < R2COM> 2x price difference... 2013-02-07T06:01:05 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 255 seconds] 2013-02-07T06:03:48 < dirty_d> ltspice is nice 2013-02-07T06:03:53 < dirty_d> free 2013-02-07T06:04:04 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out] 2013-02-07T06:05:50 < R2COM> I mean linear... err.. no ltspice.. 2013-02-07T06:06:22 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-07T06:07:02 < dirty_d> ok 2013-02-07T06:07:15 < donigs> https://www.youtube.com/watch?v=IPM8OR6W6WE 2013-02-07T06:08:28 -!- jef79m is now known as jef79m_lurking 2013-02-07T06:08:32 < dirty_d> cool 2013-02-07T06:11:53 < dirty_d> hmm, already have a hvlp gun 2013-02-07T06:11:59 < dirty_d> must be expensive 2013-02-07T06:14:00 < gxti> the good news is i figured out how to get leap second data out of this gps receiver 2013-02-07T06:14:15 < gxti> the bad news is that it's the raw almanac data and now i have to unscrew my head and attempt to read the GPS spec 2013-02-07T06:14:15 < emeb> we're saved! 2013-02-07T06:14:34 < emeb> boo! reading specs... 2013-02-07T06:16:01 < gxti> apparently it requires 8 different variables 2013-02-07T06:16:21 < dirty_d> leap seocnd data? 2013-02-07T06:16:29 < emeb> to compute leap second from almanac? 2013-02-07T06:16:35 < gxti> calculating a future leap second from almanac, yeah 2013-02-07T06:16:38 < Erlkoenig> hmm the question is how long that hydrophopic coating lasts before it wears off... and how toxic it is when it gets into the environment 2013-02-07T06:17:09 < dirty_d> gxti, i thought gps gave you utc time or something 2013-02-07T06:17:32 < gxti> yes, if i do nothing the time coming from the gps receiver will leap 2013-02-07T06:17:45 < dirty_d> ahh 2013-02-07T06:17:47 < gxti> but this is a ntp server and i have to give a day's warning to clients 2013-02-07T06:18:40 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-07T06:18:44 < dirty_d> i thoguth there was a simple formula for that 2013-02-07T06:18:49 < dirty_d> leap years at least 2013-02-07T06:18:59 < gxti> -_- 2013-02-07T06:19:11 < emeb> what do the do with a day's warning? slowly advance until it occurs? 2013-02-07T06:19:27 < gxti> dirty_d: leap seconds and leap years are totally unrelated 2013-02-07T06:19:34 < dirty_d> yea i figured 2013-02-07T06:19:38 < gxti> emeb: everyone has to do it at the same time 2013-02-07T06:19:45 < gxti> emeb: but ntp only polls every 10m or so 2013-02-07T06:19:49 < dongs> Erlkoenig: the dude is wearing a respirator and mask whlie spraying it 2013-02-07T06:19:54 < dongs> it must be toxic as fuck 2013-02-07T06:20:20 < emeb> gxti: ah - so they schedule it in their ntp clients to all happen at once? 2013-02-07T06:20:25 < gxti> dongs: doesn't have to be toxic to not want it in your lungs while it's aerosolized 2013-02-07T06:20:29 < gxti> emeb: yep 2013-02-07T06:20:37 < emeb> makes sense. 2013-02-07T06:24:12 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 276 seconds] 2013-02-07T06:24:24 -!- enots [dimka@freelsd.net] has joined ##stm32 2013-02-07T06:32:45 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-07T06:46:56 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-07T06:50:24 < gxti> so i guess i can ignore some of this shit, apparently there's provision for a linear correction over time as well. more oddly, it's not zero... 2013-02-07T06:51:17 -!- jef79m_lurking is now known as jef79m 2013-02-07T06:51:29 < gxti> like they tried to approximate zero using floats and didn't quite hit it 2013-02-07T06:55:07 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-46.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-07T06:55:25 -!- jef79m is now known as jef79m_lurking 2013-02-07T06:59:37 < dongs> hmm digikey is here 2013-02-07T07:19:54 < emeb> anything good? 2013-02-07T07:23:49 < dongs> got some loose hdmi headers in bubble wrap 2013-02-07T07:23:53 < dongs> err dp headers 2013-02-07T07:24:10 < dongs> mostly parts for my cubieboard breakout 2013-02-07T07:24:15 < dongs> which hasnt arrived yet fuck 2013-02-07T07:25:08 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Ping timeout: 255 seconds] 2013-02-07T07:27:45 < dongs> Feb 7 2013-02-07T07:27:46 < dongs> 13:37 2013-02-07T07:27:46 < dongs> Arrival at inward office of exchange 2013-02-07T07:27:48 < dongs> nice timestamp 2013-02-07T07:30:47 < dongs> haha, here's a failure i wont be reporting to digikey: 2013-02-07T07:32:39 < dongs> well, if imgur ever uploads shit 2013-02-07T07:33:20 < dongs> http://i.imgur.com/0ktsKfZ.jpg 2013-02-07T07:33:36 < emeb> a few extra 2013-02-07T07:34:31 < gxti> i once had a double-fill, two completely separate barcoded packets each with the 10 items i ordered 2013-02-07T07:34:47 < gxti> no idea how that happened 2013-02-07T07:34:57 < dongs> all i ever had from them was fail 2013-02-07T07:34:59 < dongs> this is the firstnonfail 2013-02-07T07:35:53 < gxti> or this: http://partiallystapled.com/~gxti/trash/2011/02/23-IMG_0082.jpg 2013-02-07T07:36:18 < dongs> wut 2013-02-07T07:36:41 < dongs> you mean you cant plug in AC cord into those? 2013-02-07T07:36:57 < gxti> sure you can, not very snug fit though 2013-02-07T07:41:25 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-07T07:45:06 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds] 2013-02-07T07:47:33 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-07T07:47:45 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-07T07:48:48 -!- jef79m_lurking is now known as jef79m 2013-02-07T07:58:59 -!- jef79m is now known as jef79m_lurking 2013-02-07T08:02:53 -!- jef79m_lurking is now known as jef79m 2013-02-07T08:02:59 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer] 2013-02-07T08:03:25 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32 2013-02-07T08:03:32 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has quit [Ping timeout: 248 seconds] 2013-02-07T08:16:04 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has joined ##stm32 2013-02-07T08:27:40 -!- DLPeterson [~hazelnuss@168.150.200.33] has joined ##stm32 2013-02-07T08:28:02 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-07T08:28:56 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-07T08:31:59 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-07T08:34:09 < dongs> holy shit 2013-02-07T08:34:13 < dongs> apple shipped back my jewpod 2013-02-07T08:34:27 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-07T08:34:32 < dongs> hmm, i got a new serial number 2013-02-07T08:34:39 < dongs> so they must have just swapped it. nice! 2013-02-07T08:36:05 < emeb_mac> got it replace? 2013-02-07T08:36:12 < dongs> seems so 2013-02-07T08:36:20 < dongs> http://mashable.com/2011/11/12/ipd-nano-recall/ < from this 2013-02-07T08:37:33 < emeb_mac> they're swapping out old nanos due to battery problems? 2013-02-07T08:37:49 < dongs> gues sso 2013-02-07T08:37:52 < dongs> and i had one from like 2005. 2013-02-07T08:37:53 < dongs> haha. 2013-02-07T08:37:57 < dongs> (which ig ot for free) 2013-02-07T08:38:02 < emeb_mac> nice 2013-02-07T08:38:31 < dongs> LOL! 2013-02-07T08:38:37 < dongs> just checked the serial 2013-02-07T08:38:42 < dongs> they replaced it with ipod nano 6th gen 2013-02-07T08:38:45 < dongs> teh square one 2013-02-07T08:39:12 < emeb_mac> wow - those have a nice screen. 2013-02-07T08:39:24 < dongs> http://support.apple.com/kb/SP593 this 2013-02-07T08:39:36 < dongs> 240x240 1.5" tft 2013-02-07T08:40:04 < emeb_mac> with touch, no? 2013-02-07T08:40:19 < emeb_mac> ISTR playing with one @ retailer. Had nice UI 2013-02-07T08:40:54 < dongs> i dont really see how you can touch on a 1.5" screen 2013-02-07T08:42:24 < dongs> i think this is the one people used to make watches out of 2013-02-07T08:43:07 < dongs> haha it is touch 2013-02-07T08:43:13 < dongs> retarded shit. 2013-02-07T08:43:14 < emeb_mac> ye[ 2013-02-07T08:43:34 < emeb_mac> yeah - but seemed to work well when I tried it 2013-02-07T08:43:41 < GargantuaSauce> it's only on quadrants though, no actual touchy interface 2013-02-07T08:43:53 < emeb_mac> aha 2013-02-07T08:43:56 < dongs> i wouldnt know, this would be my second apple device I own since 2005 2013-02-07T08:44:15 < emeb_mac> responded well to swipes, etc. 2013-02-07T08:45:00 < dongs> haha photos 2013-02-07T08:45:09 < dongs> im watching some retarded unboxing/whtever shit on youtube 2013-02-07T08:45:14 < dongs> so much useless shit 2013-02-07T08:47:19 < emeb_mac> lols 2013-02-07T08:53:03 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 276 seconds] 2013-02-07T09:10:26 -!- jef79m is now known as jef79m_lurking 2013-02-07T09:28:40 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-07T09:40:35 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-07T09:49:33 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T09:57:51 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-07T09:58:45 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-07T10:02:56 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-07T10:13:29 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-07T10:16:12 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-07T10:42:24 -!- claude is now known as Claude 2013-02-07T10:48:16 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-07T10:58:46 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T11:21:57 -!- DLPeterson [~hazelnuss@168.150.200.33] has quit [Read error: Operation timed out] 2013-02-07T11:27:13 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32 2013-02-07T11:30:31 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T11:34:50 -!- alexn_ [~alexn@93.104.205.194] has joined ##stm32 2013-02-07T11:36:53 -!- jef79m_lurking [~jef79m@124-169-138-54.dyn.iinet.net.au] has quit [Read error: Connection reset by peer] 2013-02-07T11:48:01 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-07T11:49:32 -!- jef79m [~jef79m@124-168-171-28.dyn.iinet.net.au] has joined ##stm32 2013-02-07T11:50:15 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-07T11:50:18 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-07T11:58:35 -!- BusError_ is now known as BusError 2013-02-07T12:43:00 -!- Laurenceb_ [~Laurence@host86-177-214-70.range86-177.btcentralplus.com] has joined ##stm32 2013-02-07T12:43:56 < dongs> no chats at all 2013-02-07T12:44:00 < dongs> so sad 2013-02-07T12:58:29 -!- daku is now known as DaKu 2013-02-07T13:00:18 -!- Laurenceb_ [~Laurence@host86-177-214-70.range86-177.btcentralplus.com] has quit [Read error: Operation timed out] 2013-02-07T13:05:12 < zyp> :( 2013-02-07T13:05:28 < Viper168> dongs, http://sphotos-b.xx.fbcdn.net/hphotos-ash3/531575_546686568685692_1321546185_n.jpg 2013-02-07T13:06:08 -!- mx [~christoph@ppp-93-104-17-11.dynamic.mnet-online.de] has joined ##stm32 2013-02-07T13:09:16 < Thorn> they finally fixed it! only 24 hours after it was introduced http://qt.gitorious.org/qtplayground/qtserialport/commit/00d68d29810184789c306aa5f74273be32d1dd6a/diffs/c07b1c4b615a5152dfbdb1c3e7cbcea9e3d82efa 2013-02-07T13:10:33 -!- DaKu is now known as daku 2013-02-07T13:24:07 < dongs> Viper168: mega old 2013-02-07T13:24:16 < dongs> you can tell how old that pic is by the jpeg artifacts 2013-02-07T13:24:22 < dongs> from each reupload on fecesbook 2013-02-07T13:28:10 < Viper168> of course, I've seen it before 2013-02-07T13:28:16 < Viper168> but I know how much you love dicks 2013-02-07T13:29:40 < zyp> dongs, any news? 2013-02-07T13:29:59 < dongs> zyp, stencil still not here. 2013-02-07T13:30:04 < dongs> was supposed to be today, failing. 2013-02-07T13:30:20 < dongs> (for dp shit). f3 shit submitted. 2013-02-07T13:31:03 < zyp> yeah, I saw you mentioned that earlier 2013-02-07T13:32:20 < dongs> just gonna go to sleep, nothign else to do. 2013-02-07T13:32:21 < dongs> bbl 2013-02-07T13:33:37 <+Steffanx> gn 2013-02-07T13:33:55 <+Steffanx> Whaa, 2013-02-07T13:34:04 <+Steffanx> 20:30 <= and dongs goes to sleep?! 2013-02-07T13:34:10 <+Steffanx> *bed 2013-02-07T14:14:56 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T14:21:39 < Laurenceb> he has waifu to sleep with 2013-02-07T14:27:20 <+Steffanx> waifus go to sleep around 20:30?! 2013-02-07T14:29:37 < Viper168> they do after you wake them up 2013-02-07T14:30:11 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T14:31:38 < Laurenceb> or put in the compressed air 2013-02-07T14:34:32 -!- jef79m is now known as jef79m_lurking 2013-02-07T14:42:37 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds] 2013-02-07T14:43:01 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-07T14:51:12 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-07T14:59:39 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 276 seconds] 2013-02-07T15:00:10 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-07T15:10:49 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T15:16:30 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-82.pptp.stw-bonn.de] has joined ##stm32 2013-02-07T15:17:34 < Laurenceb> man tail 2013-02-07T15:17:57 <+Steffanx> .... 2013-02-07T15:23:54 < jpa-> tail: wag your animatronic furry tail 2013-02-07T15:24:37 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T15:26:16 <+Steffanx> jpa- your mother didn't taught you to speak like that 2013-02-07T15:26:45 <+Steffanx> *talk 2013-02-07T15:31:54 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T15:31:56 < Laurenceb> http://cache.gawker.com/assets/images/4/2009/10/500x_MoodTail.jpg 2013-02-07T15:35:31 <+Steffanx> I know my blabla isn't interesting most of the time, but man.. you are even worse Laurenceb 2013-02-07T15:36:19 < Erlkoenig> some people attach those to buttplugs 2013-02-07T15:36:56 <+Steffanx> Some people don't know when to stop :P 2013-02-07T15:49:07 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 244 seconds] 2013-02-07T15:49:56 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-07T15:49:58 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Disconnected by services] 2013-02-07T15:50:26 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-07T15:52:49 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T15:53:10 -!- dirty_d is now known as Guest17091 2013-02-07T15:53:57 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 276 seconds] 2013-02-07T15:54:20 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-07T15:55:47 < dirty_d> so this seems pretty neat https://www.sparkfun.com/products/691 2013-02-07T15:56:01 < dirty_d> $20 for 100m range at 250kb/s 2013-02-07T15:56:18 < dirty_d> spi up to at least 4MHz 2013-02-07T15:56:28 < dirty_d> says it can do 2Mb/s too on the air 2013-02-07T15:56:48 < Erlkoenig> on ebay you can buy 20 of them for 20$ or something 2013-02-07T15:57:00 < dirty_d> really? 2013-02-07T15:57:21 < Erlkoenig> yeah from china 2013-02-07T15:58:41 <+Steffanx> Yeah, I told you dirty_d 2013-02-07T15:58:49 <+Steffanx> ~2-3$ per module 2013-02-07T15:59:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-07T15:59:23 <+Steffanx> *make it 1.4$ 2013-02-07T15:59:40 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 246 seconds] 2013-02-07T16:00:03 < Laurenceb> or just stm32w108 2013-02-07T16:00:17 <+Steffanx> You cant get those for 2$ 2013-02-07T16:00:31 < Laurenceb> but you get an F1 core too 2013-02-07T16:00:43 <+Steffanx> Who says he needs that? 2013-02-07T16:01:03 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-07T16:04:10 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-07T16:05:16 -!- ntfreak_ is now known as ntfreak 2013-02-07T16:06:12 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 264 seconds] 2013-02-07T16:06:30 < dirty_d> Steffanx, just the chip for $2-$3 or the whole board? 2013-02-07T16:06:39 <+Steffanx> http://www.ebay.com/itm/1X-NRF24L01-2-4GHz-Antenna-Wireless-Transceiver-Module-For-Microcontroller-EP98-/121062689188?pt=LH_DefaultDomain_0&hash=item1c2fe609a4 2013-02-07T16:06:40 < Laurenceb> http://uk.mouser.com/ProductDetail/STMicroelectronics/STM32W-RFCKIT/?qs=sGAEpiMZZMuMLSej4gvK4jSgQUk3ijMF 2013-02-07T16:07:06 <+Steffanx> Dont know how well they perform, but 1.4$ is worth a try 2013-02-07T16:07:15 <+Steffanx> I have a few of them, but never tried them D: 2013-02-07T16:07:28 < dirty_d> http://www.ebay.com/itm/10Pcs-NRF24L01-2-4GHz-Antenna-Wireless-Transceiver-For-Microcontroller-EP98-New-/121042000247?pt=LH_DefaultDomain_0&hash=item1c2eaa5977 2013-02-07T16:07:38 < dirty_d> $10 for 12 2013-02-07T16:07:49 < dirty_d> i mean $12 for 10 2013-02-07T16:08:56 < Laurenceb> bladerf update 2013-02-07T16:08:58 < Laurenceb> Laurenceb: 3x was ~ -10 dB 5x ~ -22 dB 7x ~ -35 dB 2013-02-07T16:08:58 < Laurenceb> Laurenceb: test setup was : tune LMS to 400 MHz. Then tx a narrow band fm signal at 400, then 800, 1200, 1600, ... and looks at relative levels. 2013-02-07T16:09:25 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-07T16:09:41 <+Steffanx> And Laurenceb is happy with that? 2013-02-07T16:09:58 < Laurenceb> no 2013-02-07T16:10:36 < dirty_d> i dont see the range mentioned anywhere 2013-02-07T16:12:15 < dirty_d> "Improved ShockBurst and UART are easy to connect with all kinds of low cost" 2013-02-07T16:12:18 < dirty_d> what the hell does that mean 2013-02-07T16:12:25 < dirty_d> i thought these were SPI? 2013-02-07T16:15:47 < zyp> Laurenceb, saw that last night 2013-02-07T16:15:59 < Laurenceb> its not very encouraging :-/ 2013-02-07T16:16:04 <+Steffanx> Just read the datasheet of the nrf24l01+ dirty_d :) 2013-02-07T16:16:20 < dirty_d> gonna do that 2013-02-07T16:16:34 <+Steffanx> anyway, for 1.4$ not much can go wrong :P 2013-02-07T16:16:36 < zyp> Laurenceb, doesn't really matter to me 2013-02-07T16:16:41 <+Steffanx> *2.8, you need to of course 2013-02-07T16:16:42 < zyp> why does it matter to you? 2013-02-07T16:17:18 < Laurenceb> for datalinks to high altitude balloons 2013-02-07T16:18:27 < zyp> to or from? 2013-02-07T16:18:35 < Laurenceb> id have to add a board with SAW and lna 2013-02-07T16:18:38 < Laurenceb> maybe both 2013-02-07T16:19:03 < Laurenceb> 10dB is nothing like enough attenuation 2013-02-07T16:21:01 < zyp> remember that it comes on top of the other attenuation, from your antenna and everything 2013-02-07T16:25:55 < dirty_d> zyp have you looked at nRF24L01+? 2013-02-07T16:26:29 < Laurenceb> zyp: but if you are trying to rx a balloon and there is a phone mast nearby... 2013-02-07T16:26:34 < zyp> dirty_d, not in depth 2013-02-07T16:26:46 < dirty_d> zyp, seems pretty neat for short range 2013-02-07T16:26:51 < dirty_d> 2Mb/s air rate 2013-02-07T16:26:58 < dirty_d> and SPI 2013-02-07T16:27:03 < zyp> but air rate is irrelevant 2013-02-07T16:27:11 < Laurenceb> personally id use stm32w108 on custom board 2013-02-07T16:27:23 < Laurenceb> now it has firmware 2013-02-07T16:27:27 < zyp> though, I've used nRF24AP2 2013-02-07T16:27:32 < dirty_d> zyp, it says SPI up to 10Mb/s 2013-02-07T16:27:51 < zyp> dirty_d, that's still very irrelevant 2013-02-07T16:28:01 < dirty_d> zyp, what else is there? 2013-02-07T16:28:11 < Laurenceb> throughput 2013-02-07T16:28:19 < dirty_d> hmm 2013-02-07T16:28:29 < dirty_d> how do i figure that out? 2013-02-07T16:28:41 <+Steffanx> trial and error 2013-02-07T16:28:46 <+Steffanx> No, i bet someone on the web tested that 2013-02-07T16:28:53 < zyp> the true bottleneck is the product of the size of the frames you can send, the rate you can send them at, and the percentage that will reliably get through 2013-02-07T16:29:10 < dirty_d> hmm 2013-02-07T16:29:48 <+Steffanx> https://forum.sparkfun.com/viewtopic.php?t=16453 .. 2013-02-07T16:30:05 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has joined ##stm32 2013-02-07T16:30:06 <+Steffanx> Not sure which module thay use 2013-02-07T16:30:11 < zyp> or rather, combine the latter two into «the rate of frames that will reliably get through after the overhead of handling re-transmitting lost frames» 2013-02-07T16:30:55 < zyp> which of course depends on packet loss, which is a variable depending on your link quality 2013-02-07T16:31:18 <+Steffanx> Only when you care about that :) 2013-02-07T16:31:47 < zyp> of course I do, I need to trust my data. 2013-02-07T16:31:51 <+Steffanx> but i guess dirty_d wants a reliable link? 2013-02-07T16:31:57 < dirty_d> yea 2013-02-07T16:32:35 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 248 seconds] 2013-02-07T16:33:29 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-07T16:33:50 < dirty_d> so about 46KB/s with 32 byte packets with no loss 2013-02-07T16:34:40 < zyp> which is about 1/30 of FS USB throughput 2013-02-07T16:34:50 <+Steffanx> zyp and his usb fetish 2013-02-07T16:35:14 < dirty_d> zyp, but it works with magic, not wires 2013-02-07T16:35:49 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-07T16:35:54 <+Steffanx> morning Sir 2013-02-07T16:36:29 < zyp> Steffanx, well, the easy way to avoid throughput problems is having a capacity an order of magnitude larger than the requirements :p 2013-02-07T16:39:20 <+Steffanx> This channel isn't good for me. When you say that i wonder if mr goatse was thinking the same zyp ... 2013-02-07T16:42:44 < mervaka> hmm 2013-02-07T16:42:57 < mervaka> what's arm's version of M_PI? 2013-02-07T16:43:12 -!- fxd0h_ [~fx@186.18.172.240] has joined ##stm32 2013-02-07T16:43:13 < mervaka> doesn't seem to be in math.h 2013-02-07T16:43:33 < mervaka> i could always declare my own i know.. 2013-02-07T16:43:51 < emeb_mac> Steffanx: who you calling Sir? 2013-02-07T16:43:59 < zyp> mervaka, it's not? 2013-02-07T16:44:19 < mervaka> zyp: unless i'm making a schoolboy error, seems now. 2013-02-07T16:44:19 < emeb_mac> mmmm... pie... 2013-02-07T16:44:21 < mervaka> not* 2013-02-07T16:44:40 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-07T16:44:59 < zyp> zyp@rin include $ grep M_PI math.h 2013-02-07T16:45:00 < zyp> #define M_PI 3.14159265358979323846 2013-02-07T16:45:01 -!- fxd0h [~fx@186.18.172.240] has quit [Ping timeout: 245 seconds] 2013-02-07T16:45:02 -!- fxd0h_ is now known as fxd0h 2013-02-07T16:45:10 < zyp> that's from my arm toolchain 2013-02-07T16:45:32 < zyp> i.e. latest gcc-arm-embedded 2013-02-07T16:46:38 < mervaka> ah 2013-02-07T16:46:51 < mervaka> i guess this is an MDK fail then. 2013-02-07T16:47:49 < mervaka> no mention of pi :( 2013-02-07T16:47:49 < emeb_mac> murder-death-kill? 2013-02-07T16:50:05 < karlp> mervaka: were you doing multi channel adc sequences with dma? 2013-02-07T16:50:11 < karlp> emeb suggested I talk to you, 2013-02-07T16:50:25 < karlp> I'm seeing the adc sequence channels rotated by one in the dma output buffer. 2013-02-07T16:50:46 < karlp> so dmadest buf[0,1,2,3] contains adc channels 3,0,1,2 2013-02-07T16:51:44 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-07T16:51:58 < mervaka> aha 2013-02-07T16:52:09 < mervaka> um, are you around tonight? i'm about to head off to uni 2013-02-07T16:52:22 < mervaka> in a sentence though, it's a ballache. 2013-02-07T16:53:13 < karlp> yeah, I'll be here 2013-02-07T16:53:17 < mervaka> cool 2013-02-07T16:53:26 < mervaka> should be back around 8pm GMT 2013-02-07T16:53:30 < karlp> works for me. 2013-02-07T16:55:26 < emeb_mac> karlp: which uP you using? 2013-02-07T16:55:27 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-221-155.clienti.tiscali.it] has quit [Ping timeout: 260 seconds] 2013-02-07T16:58:15 <+Steffanx> You emeb_mac, who else? 2013-02-07T16:58:42 < emeb_mac> well then, good morning to you too Sir. 2013-02-07T16:59:36 < karlp> emeb_mac: L1. 2013-02-07T17:00:08 < emeb_mac> karlp: ah. I've got working code for sampling multiple adc channels to dma on an F4. 2013-02-07T17:00:09 < zyp> is the ADC equivalent to F1? 2013-02-07T17:00:36 < zyp> I don't think I've done ADC other than on F1 2013-02-07T17:01:12 < emeb_mac> I've also done it on a F100VL, but only to the injected channels there. 2013-02-07T17:01:19 < emeb_mac> (w/o dma) 2013-02-07T17:01:47 < zyp> I've done both 2013-02-07T17:02:18 < karlp> emeb_mac: can I see it? 2013-02-07T17:02:26 < emeb_mac> injected channels are nice in that the storage is in the ADC if you use 4 or less so no dma neede. 2013-02-07T17:02:28 < karlp> the dma isn't exactly the same, l1 has the same dma as f1/f3, 2013-02-07T17:02:41 < emeb_mac> right 2013-02-07T17:02:53 < karlp> yeah, I know I can use the injected for less than 4, but I'll be doing 12 eventually, 2013-02-07T17:03:09 < emeb_mac> I did find that using injected chls on the F100 you can't do continuous mode. 2013-02-07T17:03:22 < karlp> I'm just using regular 2013-02-07T17:03:30 < emeb_mac> right. 2013-02-07T17:03:59 < karlp> regular, continuous, circular, much like some of the stdperiphlib examples and what my intpretation of the ref man suggests. 2013-02-07T17:04:15 < karlp> the values are all ok, just rotated in the dest buff. 2013-02-07T17:04:24 < zyp> http://cgit.jvnv.net/suzumebachi/tree/telemetry.cpp?id=c265553 <- I'm using a normal sequence with dma here, as far as I remember it works correctly 2013-02-07T17:04:24 < emeb_mac> weird 2013-02-07T17:04:25 < karlp> my theory is it's doing an extra conversion on something else first 2013-02-07T17:04:39 < karlp> possibly me turning on dma in the wrong order or something 2013-02-07T17:08:54 < karlp> zyp: that's f1 code?doing the double write to the aden bit to start conversion? 2013-02-07T17:09:30 < zyp> yes, that's F1 code 2013-02-07T17:10:01 < zyp> pretty old, since it got removed when I ported my quadrotor stuff to F4 2013-02-07T17:13:44 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-07T17:14:25 < zyp> http://bin.jvnv.net/f/Da1fb.png <- it's the code serving data for this :p 2013-02-07T17:18:16 < dirty_d> why did you choose f4 over f3? 2013-02-07T17:18:32 < dirty_d> thats cool 2013-02-07T17:18:41 < karlp> f4 was out well before f3 2013-02-07T17:18:45 < dirty_d> ahh 2013-02-07T17:18:58 < karlp> f3's still pretty fresh 2013-02-07T17:19:05 < karlp> witness the bootloader fuckups 2013-02-07T17:19:23 < dirty_d> i havent noticed anything 2013-02-07T17:24:25 < karlp> lucky you :) 2013-02-07T17:25:05 -!- fxd0h [~fx@186.18.172.240] has quit [Read error: Connection reset by peer] 2013-02-07T17:26:05 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-07T17:27:19 < dirty_d> karlp, whats the actual issue? 2013-02-07T17:31:03 < karlp> the f373 was meant to have a usb capable bootloader, but it doesn't. (or, lots of devices were shipped with the wrong bootloader) 2013-02-07T17:32:12 < karlp> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex_mx_stm32%2FSTM32F373%20Bootloader&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B¤tviews=234 2013-02-07T17:32:32 -!- mx [~christoph@ppp-93-104-17-11.dynamic.mnet-online.de] has left ##stm32 [] 2013-02-07T17:32:37 < dirty_d> oh, i have the 303 2013-02-07T17:33:03 < emeb_mac> seems that most F373 shipped before Sept '12 had the bootloader intended for the F383 w/o USB. 2013-02-07T17:33:09 < zyp> yep 2013-02-07T17:33:31 < zyp> which probably means a lot of distributor stock 2013-02-07T17:33:38 < emeb_mac> right 2013-02-07T17:33:45 < zyp> including the ones I bought in december 2013-02-07T17:33:56 < emeb_mac> o/ - me too 2013-02-07T17:36:00 < emeb_mac> hey lookie here: http://www.digikey.com/product-detail/en/BB-BBLK-000/BB-BBLK-000-ND/3884456 2013-02-07T17:36:12 < emeb_mac> $45 - rpi killer? 2013-02-07T17:37:51 < gxti> i am required by guild law to scoff anytime someone uses says something is a "X killer" 2013-02-07T17:38:02 < emeb_mac> of course. 2013-02-07T17:38:12 * emeb_mac uses it ironically 2013-02-07T17:38:18 < zyp> :p 2013-02-07T17:38:38 < zyp> hmm, it got a HDMI connector 2013-02-07T17:38:47 < emeb_mac> yah. and emmc 2013-02-07T17:38:49 < zyp> would probably capture the tv crowd 2013-02-07T17:39:13 * emeb_mac admits he has a pre-release of this. 2013-02-07T17:39:46 < zyp> but IIRC the SoC doesn't have hardware video decoding 2013-02-07T17:39:54 < zyp> so that might be irrelevant 2013-02-07T17:40:17 < emeb_mac> the am3359 has it, but the am3352 doesn't 2013-02-07T17:40:28 < emeb_mac> production boards will have the am3359. 2013-02-07T17:41:03 < zyp> ah ok 2013-02-07T17:41:09 < zyp> still cortex-a8? 2013-02-07T17:41:18 < emeb_mac> finkso 2013-02-07T17:41:27 < zyp> yeah 2013-02-07T17:41:40 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 248 seconds] 2013-02-07T17:41:42 < zyp> not significantly faster then 2013-02-07T17:41:48 < zyp> (as opposed to the odroid boards) 2013-02-07T17:41:51 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Ping timeout: 276 seconds] 2013-02-07T17:41:55 < emeb_mac> right 2013-02-07T17:42:04 < emeb_mac> but cheap & lots of i/o 2013-02-07T17:43:00 < zyp> odroid-u is not much more expensive 2013-02-07T17:43:16 < zyp> and I haven't seen the i/o of the old beaglebone killed rpi yet 2013-02-07T17:44:04 < gxti> the a10 olinuxino should be nice although the name continues to make me retch 2013-02-07T17:44:18 < gxti> not out yet though 2013-02-07T17:44:34 < zyp> a10 is also just singlecore cortex-a8 2013-02-07T17:45:02 < gxti> sosad 2013-02-07T17:45:19 < gxti> my dreams of a olimex build cluster, shattered :p 2013-02-07T17:45:24 < zyp> :p 2013-02-07T17:45:46 < zyp> I might be interested when they do an exynos 5 based odroid 2013-02-07T17:46:21 < zyp> then we're starting to talk enough cpu power to actually do something useful 2013-02-07T17:49:31 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-07T17:56:27 < dirty_d> can my discover program external chips with only SWD, or jtag too? 2013-02-07T17:56:33 < dirty_d> discovery* 2013-02-07T17:56:40 < dirty_d> and do i care? 2013-02-07T17:57:03 < gxti> pretty sure stlink doesn't do jtag 2013-02-07T17:57:10 < dirty_d> ok 2013-02-07T17:57:30 < dirty_d> i guess i dont care then 2013-02-07T17:57:43 < dirty_d> what advantage does jtag have? 2013-02-07T17:58:02 < gxti> you can have multiple chips on one chain 2013-02-07T17:58:25 < gxti> and things that are not mcus 2013-02-07T17:58:47 < dirty_d> ok, SWD is good enough for me 2013-02-07T17:58:50 < gxti> indeed 2013-02-07T17:59:47 < karlp> swd can do multiple too, or at least, its meant to be able to 2013-02-07T18:00:01 < Erlkoenig> jtag is said to be faster 2013-02-07T18:00:09 < karlp> [citation needed] 2013-02-07T18:00:41 < Erlkoenig> quote from blurp@#mikrocontroller.net@irc.euirc.net 2013-02-07T18:00:42 < Erlkoenig> :P 2013-02-07T18:00:44 < karlp> you can get fast and slow swd dongles, and fast and slow jtag dongles. 2013-02-07T18:00:56 < gxti> karlp: really all i know is someone was talking about lpc4343 or whatever that dual-core processor was, and it was easier to use jtag to address both 2013-02-07T18:01:01 < gxti> but i haven't done it so i don't know 2013-02-07T18:01:39 < zyp> yes 2013-02-07T18:01:53 < zyp> in swd mode, only the M4 in lpc43xx is available 2013-02-07T18:02:01 < zyp> so jtag has to be used to debug the M0 2013-02-07T18:03:04 < zyp> but this is a limitation of the chip, not the protocol 2013-02-07T18:03:26 < zyp> swd supports having several devices on the same bus, but I haven't seen any devices supporting that yet 2013-02-07T18:03:52 < zyp> and as for speed 2013-02-07T18:03:54 < gxti> any special wiring, or just i2c-style? 2013-02-07T18:04:28 < zyp> bmp does both swd and jtag, and swd mode is slightly faster 2013-02-07T18:04:41 < zyp> but not enough to really matter 2013-02-07T18:04:56 < zyp> just all in parallel 2013-02-07T18:05:20 < dirty_d> zyp, so i can program stm32f4's etc with my f3disco right? 2013-02-07T18:05:26 < karlp> yes. 2013-02-07T18:05:29 < zyp> keep in mind that swd is muxed on TMS and TCK, and in a traditional jtag setup they are connected in parallel on all chips 2013-02-07T18:05:30 < dirty_d> ok 2013-02-07T18:05:57 < karlp> (but leave the jumper on Idd!) 2013-02-07T18:06:04 < zyp> :) 2013-02-07T18:06:31 < dirty_d> isnt that one for measuring current? 2013-02-07T18:07:17 < dirty_d> anyone have a schematic for the most basic SWD setup? just being able to program the chip and run the code 2013-02-07T18:07:32 < zyp> do you even need one? 2013-02-07T18:07:37 < gxti> it's wires to a header 2013-02-07T18:07:51 < dirty_d> well when i make my own pcb i will 2013-02-07T18:07:52 < gxti> swdio, swclk, gnd, reset, and optionally: vcc and swo 2013-02-07T18:08:07 < zyp> reset is also pretty optional 2013-02-07T18:08:07 < karlp> reset is not required. 2013-02-07T18:08:09 < dirty_d> what about boot0 and boot1? 2013-02-07T18:08:15 < gxti> reset is pseudo-optional 2013-02-07T18:08:35 < karlp> there's a really nice app note from stm called "gettting started with stm32 hardwrae development" that covers this :) 2013-02-07T18:08:41 < gxti> i only solder header pins in for swdio, swclk, and gnd, but reset is there in case shit hits the fan :p 2013-02-07T18:08:53 < zyp> reset is not required unless you are using sleep modes or remaps the swd pins 2013-02-07T18:09:16 < gxti> yeah, and if i accidentally manage to do any of those things i ain't greenwiring it 2013-02-07T18:09:18 < dirty_d> ok 2013-02-07T18:09:38 < zyp> anyway, I suggest just using the standard connector 2013-02-07T18:09:43 < zyp> it makes life so much easier :p 2013-02-07T18:10:25 < zyp> http://bin.jvnv.net/f/Exo6T.JPG <- like the one on these boards 2013-02-07T18:10:40 < zyp> emeb would have the schematic 2013-02-07T18:10:44 < zyp> huh, where did he go? 2013-02-07T18:10:51 < dirty_d> zyp, the other end would plug into the header on the disco? 2013-02-07T18:10:57 < dirty_d> why dont they use the same connector there? 2013-02-07T18:10:59 < karlp> (or just use a 5 or 6 pin header like on the discovery baords :) 2013-02-07T18:11:18 < zyp> or just get a decent debugger :p 2013-02-07T18:11:29 < karlp> it's a perfectly decent debugger :) 2013-02-07T18:11:31 < dirty_d> this is good enough for me 2013-02-07T18:11:35 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-07T18:11:38 < dirty_d> especially with that logger you made 2013-02-07T18:12:08 < gxti> the nice thing about the disco header is you can just solder down the 3 pins for swdio, gnd, swclk, but the others are there in case you need them 2013-02-07T18:12:54 < zyp> the bad part about it is pins prone to getting bent 2013-02-07T18:13:03 < zyp> especially when they are the only pins on the board 2013-02-07T18:13:33 < zyp> I prefer a small shrouded header over that 2013-02-07T18:13:57 < zyp> and since the pinout is standardized, you might as well just keep to the standard for interopability 2013-02-07T18:14:02 < emeb> the mini-10 is nice 2013-02-07T18:14:22 < emeb> looks good on my little breakouts 2013-02-07T18:14:34 < zyp> I just posted a pic of those 2013-02-07T18:14:37 < dirty_d> the boot pins should just be set to boot from user flash right? 2013-02-07T18:14:49 < zyp> unless you want to use the bootloader 2013-02-07T18:15:04 < dirty_d> i dont think i do 2013-02-07T18:15:09 < dirty_d> im not sure what you even use that for 2013-02-07T18:15:17 < zyp> bootloading. :p 2013-02-07T18:15:29 < dirty_d> but bootloading what? 2013-02-07T18:15:29 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: ….] 2013-02-07T18:15:32 < zyp> i.e. flashing without a jtag/swd adapter 2013-02-07T18:15:39 < dirty_d> ahh 2013-02-07T18:15:55 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-07T18:15:56 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-07T18:15:56 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-07T18:15:56 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-07T18:15:56 < zyp> not very useful for development, but may be nice for user upgrades 2013-02-07T18:16:11 < dirty_d> i see 2013-02-07T18:16:49 < zyp> as for debuggers, I love these: http://bin.jvnv.net/f/n71NA.JPG 2013-02-07T18:16:52 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T18:19:13 < emeb> really glad I brought the aux serial on those out to 0.1" header 2013-02-07T18:19:19 < emeb> it's been very handy 2013-02-07T18:19:45 < zyp> :) 2013-02-07T18:19:58 < zyp> I haven't put headers on any of those yet 2013-02-07T18:20:17 < zyp> but the stlink headers are nice 2013-02-07T18:20:39 < emeb> yeah - good idea you had to label the pins 2013-02-07T18:20:41 -!- alexn_ [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-07T18:20:52 < emeb> makes them more useful for nonstd setups 2013-02-07T18:20:57 < gxti> what alternative headers are there? the 10-pin cortex is shorter but wider 2013-02-07T18:21:03 < gxti> and more esspensive 2013-02-07T18:21:21 < dirty_d> "Optional components: 0 ohm resistor" lol 2013-02-07T18:21:59 < zyp> gxti, too cheap to matter for boards I make for myself anyway 2013-02-07T18:22:11 < zyp> and can just be left out on shipping boards 2013-02-07T18:22:22 < zyp> as in unpopulated 2013-02-07T18:22:39 < emeb> gxti: what kind of qty do you do? I gather your ntp server will be build in volume... 2013-02-07T18:22:41 < karlp> zyp: how do you resolve "both modified" for submodules? 2013-02-07T18:22:50 < gxti> emeb: far too little to justify my scrooging 2013-02-07T18:22:55 < zyp> karlp, huh? 2013-02-07T18:23:08 < emeb> gxti: embrace your inner scrooge! 2013-02-07T18:23:15 < gxti> but i think about this stuff all the time for no explicable reason 2013-02-07T18:23:15 < karlp> I've done a git rebase master, and it's given a merge conflict in the submodule 2013-02-07T18:23:26 < emeb> *nod* 2013-02-07T18:23:36 < zyp> karlp, you are merging two branches that refer to two different submodule revisions? 2013-02-07T18:23:44 < karlp> apparently, wasn't deliberate 2013-02-07T18:23:48 < gxti> i think i'd take a nopop 4-pin header over a nopop cortex debug 2013-02-07T18:24:27 < zyp> just pick one that will work and mark the conflict as resolved 2013-02-07T18:24:31 < gxti> easier to make a programming jig for and easier for people who barely know which end of the iron gets hot to solder one down 2013-02-07T18:24:46 < karlp> "pick one" by going into that directory and git checkout [working code] ? 2013-02-07T18:25:03 < zyp> yes, I imagine that's what you'll have to do 2013-02-07T18:25:12 < zyp> I haven't come across that myself 2013-02-07T18:26:17 < karlp> ended up just doing git rebase --skip, somehow they both had the same update 2013-02-07T18:26:54 < zyp> gxti, otherwise I might get a tag-connect cable and use their 6-pad swd footprint 2013-02-07T18:27:54 < zyp> http://www.tag-connect.com/node/52 2013-02-07T18:28:06 < zyp> there you hav your programming jigs 2013-02-07T18:28:40 < zyp> but normal ribbon cables are still nicer for development 2013-02-07T18:32:05 < gxti> "about the same PCB space as an 0805 resistor" my ass 2013-02-07T18:32:18 < gxti> worth considering i suppose 2013-02-07T18:33:32 < emeb> huh. anyone else having trouble finding oshpark.com? 2013-02-07T18:33:43 < zyp> works for me 2013-02-07T18:33:48 < emeb> good. 2013-02-07T18:34:13 -!- BrainDamage1 [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Quit: Leaving.] 2013-02-07T18:34:27 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-07T18:35:04 < emeb> ok - working now. wonder what was up? 2013-02-07T18:35:43 < karlp> aliens 2013-02-07T18:35:55 < emeb> as good an explanation as any 2013-02-07T18:36:41 < emeb> wonder if there's a way to get paypal to stop offering me a credit card EVERY @#$%ING TIME I LOG IN. 2013-02-07T18:38:14 < zyp> move out of US, that's US only. 2013-02-07T18:38:22 < gxti> emeb: lol nope 2013-02-07T18:38:30 < gxti> well 2013-02-07T18:38:34 < gxti> you could get the credit card. 2013-02-07T18:39:08 < emeb> meh 2013-02-07T18:40:03 < karlp> stop using paypal? 2013-02-07T18:40:50 < emeb> shyeah - right. 2013-02-07T18:41:20 < karlp> what is it doing for you? 2013-02-07T18:41:29 < karlp> why not just pay with the card directly? 2013-02-07T18:41:51 < zyp> it's probably useful if you have an income on paypal 2013-02-07T18:42:26 < emeb> Some of my international clients pay via paypal. I leave the $$ in the account to pay for new development. 2013-02-07T18:43:35 < karlp> oh, you have income on paypal 2013-02-07T18:43:40 < emeb> ya 2013-02-07T18:44:08 < zyp> I was once planning to get rid of a credit card… 2013-02-07T18:44:24 -!- Claude is now known as claude 2013-02-07T18:44:56 < zyp> but I weren't allowed to cancel it from the web interface, and I didn't bother calling them up, so I still have it 2013-02-07T18:45:47 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T18:46:12 < emeb> emergency CC 2013-02-07T18:46:57 < zyp> well, if I still knew the pin :p 2013-02-07T18:48:14 < zyp> not that I really need it, considering I have like four other CCs 2013-02-07T18:48:22 < dirty_d> what the hell? there are stm32s that run java? 2013-02-07T18:48:26 < dirty_d> whyyyyyy!? 2013-02-07T18:48:27 -!- fxd0h [~fx@186.18.172.240] has quit [Ping timeout: 276 seconds] 2013-02-07T18:48:34 < Laurenceb> emergency c compiler? 2013-02-07T18:48:39 < Laurenceb> that would be useful 2013-02-07T18:48:39 < zyp> well, the hava vm is just software 2013-02-07T18:48:44 < zyp> java* 2013-02-07T18:48:45 < karlp> hrm, the order of these dma channels is correct after boot, but after the first reset and rerun it switches to this off by one. 2013-02-07T18:48:52 < karlp> getting somewhere... 2013-02-07T18:48:53 < gnomad> yeah, so long as there is an ARM JVM..... 2013-02-07T18:49:20 < gnomad> (btw, is there a JVM for cortex?) 2013-02-07T18:50:26 < dirty_d> i need a project 2013-02-07T18:50:29 < dirty_d> im slacking 2013-02-07T18:50:53 < dirty_d> what can i make that involves cnc machining and a stm32? 2013-02-07T18:51:19 < BrainDamage> cnc machine a stm32 out of Si 2013-02-07T18:51:22 < karlp> yay, turning off the timer that stops the sampling after 100ms keeps it in proper ordering... 2013-02-07T18:52:36 -!- Guest17091 [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Ping timeout: 248 seconds] 2013-02-07T18:55:26 < dirty_d> hah 2013-02-07T18:56:32 < karlp> dirty_d: http://www.stm32java.com/portal/ 2013-02-07T18:58:08 < dirty_d> no java for me 2013-02-07T18:58:40 < Laurenceb> ewww 2013-02-07T18:58:51 < Laurenceb> thats disguising 2013-02-07T18:59:34 < emeb> we've found something that shocks Laurenceb. 2013-02-07T18:59:40 < Laurenceb> yes, java 2013-02-07T18:59:58 < Laurenceb> wtf.. they have J on the part number?! 2013-02-07T19:00:06 < Laurenceb> jazelle built in or something? 2013-02-07T19:01:23 < Laurenceb> lols $2600 2013-02-07T19:01:28 < Laurenceb> per year 2013-02-07T19:05:33 < karlp> built in licensing from the ics compnay that makes the micro jvm 2013-02-07T19:05:37 < karlp> not jazelle 2013-02-07T19:05:51 < zyp> i.e. just a hardware key? 2013-02-07T19:06:06 < karlp> possibly not even that. 2013-02-07T19:06:12 -!- Guest17091 [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-07T19:07:14 < dirty_d> crystal load caps should be C = C^2 / (C*2) + Cstray 2013-02-07T19:07:15 < dirty_d> right? 2013-02-07T19:07:27 < dirty_d> says 2-7pF typically 2013-02-07T19:07:36 < dirty_d> but the datasheet says io poins have 5pF typically 2013-02-07T19:07:41 < dirty_d> which gives 10pF for the load caps 2013-02-07T19:07:52 < zyp> huh? 2013-02-07T19:08:19 < zyp> look at crystal datasheet, take C_L value, double that, round down to closest common cap value 2013-02-07T19:09:07 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-07T19:09:21 < zyp> on my lpc board I used a crystal with 18pF C_L, put 33pF loading caps on that 2013-02-07T19:09:57 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-07T19:13:11 < dirty_d> Load capacitance CL has the following formula: CL = CL1 x CL2 / (CL1 + CL2) + Cstray where: Cstray is the pin capacitance and board or trace PCB-related capacitance. Typically, it is between 2 pF and 7 pF. Please refer to Section 5: Recommendations on page 20 to minimize its value. 2013-02-07T19:13:28 < dirty_d> the picture shows CL1 and CL2 as the two caps you add 2013-02-07T19:13:35 < dirty_d> CL is just the value of each right? 2013-02-07T19:19:16 -!- daku is now known as DaKu 2013-02-07T19:21:34 -!- barthess [~barthess@77.67.196.60] has joined ##stm32 2013-02-07T19:23:45 < dirty_d> oh nevermind, i get it 2013-02-07T19:23:57 < dirty_d> how critical is it? 2013-02-07T19:24:05 < dirty_d> like what if theyre off 5 or 10 pF? 2013-02-07T19:27:54 < zyp> not very 2013-02-07T19:28:32 < zyp> wrong load will affect the oscillation frequency a little bit, but usually not enough to matter 2013-02-07T19:28:49 < Laurenceb> unless its RF kit 2013-02-07T19:28:52 < zyp> yes 2013-02-07T19:29:18 < zyp> I messed up when I did the project using a nRF24AP2 2013-02-07T19:29:37 < zyp> I put 15pF when I should have put something like 30pF 2013-02-07T19:29:44 < dirty_d> ok 2013-02-07T19:29:57 < zyp> it made the oscillator run 0.015% too fast 2013-02-07T19:30:22 < zyp> which makes out to 400 kHz at 2.4GHz - a big deal when the RF channel width was 1 MHz :p 2013-02-07T19:32:06 < dirty_d> yea 2013-02-07T19:32:40 < dirty_d> and for Rext just do what it says and go 5-6 cryastal series resistances? 2013-02-07T19:36:34 < zyp> uh, resistor? 2013-02-07T19:36:37 < zyp> never did that 2013-02-07T19:37:15 < dirty_d> yea shows a 300 something oh resistor in series with the crystal 2013-02-07T19:39:17 < qyx_> i always use it 2013-02-07T19:39:24 < qyx_> around ~390ohm 2013-02-07T19:39:35 < qyx_> it is nearly in all stm32 reference schematics 2013-02-07T19:40:00 < dirty_d> sounds about right 2013-02-07T19:40:14 < dirty_d> the crystal i liied at said 80ohms 2013-02-07T19:40:18 < dirty_d> thats about 5 times that 2013-02-07T19:40:46 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-07T19:42:47 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-07T19:43:16 < dirty_d> this looks like it would be fun to make, http://technabob.com/blog/2007/06/23/spinning-led-clock-uses-image-persistence-to-tell-time/ 2013-02-07T19:46:07 < dirty_d> how does it know where it is? 2013-02-07T19:46:13 < dirty_d> a hall sensor and magnet or something? 2013-02-07T19:46:17 < gxti> gps lol 2013-02-07T19:46:26 < dirty_d> then interpolate based on rpms? 2013-02-07T19:46:33 < gxti> yeah probably 2013-02-07T19:46:43 < gxti> bonus points if you use a brushless motor 2013-02-07T19:46:57 < dirty_d> they sure look like ass when theyre not spinning 2013-02-07T19:47:04 < dirty_d> well of course 2013-02-07T19:47:29 < neuro_sys> there was an earth globe animation using image persistence on youtube 2013-02-07T19:47:34 < neuro_sys> that shit looked dope 2013-02-07T19:47:41 < gxti> pretty old hat at this point really 2013-02-07T19:50:20 < dirty_d> 1800roms is probably good for that right? 2013-02-07T19:50:29 < dirty_d> that gives "30fps" 2013-02-07T19:51:10 < gxti> no idea what the minimum is 2013-02-07T19:51:15 < dirty_d> better make it 50,000 2013-02-07T19:51:20 < gxti> k 2013-02-07T19:51:23 < dirty_d> haha 2013-02-07T19:51:32 < gxti> better add a cowling so it doesn't put shrapnel in your face 2013-02-07T19:51:43 < dirty_d> yea 2013-02-07T19:51:53 < dirty_d> if its too fast the image might smear 2013-02-07T19:52:37 < dirty_d> 3600rpm gicves you a 60fps HD clock 2013-02-07T19:53:25 < jpa-> leds are so fast that smear is not an issue, as long as you have something decent (like stm32 dma) to drive them 2013-02-07T19:53:35 < jpa-> the most annoying part is the noise 2013-02-07T19:53:42 < dirty_d> yea 2013-02-07T19:53:52 < dirty_d> what do you mean use dma? 2013-02-07T19:53:56 < dirty_d> not regular gpio? 2013-02-07T19:54:17 < jpa-> with timer triggered dma you can easily have very precisely controlled timing for the leds 2013-02-07T19:54:38 < jpa-> of course you can just use timer interrupts and write to gpio yourself, but that wastes cpu time :P 2013-02-07T19:54:44 < karlp> man, oversampling is cool! 2013-02-07T19:55:36 < emeb> huh? 2013-02-07T19:56:11 < dirty_d> jpa-, you mean you set up which leds need to be on in advance and when the timer goes off it turns them all on at once? 2013-02-07T19:56:43 < dirty_d> nevermind, i have no idea, lol 2013-02-07T19:56:51 < jpa-> dirty_d: yeah, and you can generate the data into a ring buffer also on the fly 2013-02-07T19:58:10 < jpa-> back when i did this https://www.youtube.com/watch?v=yPZptwe29yM i had only a lousy pic so i had to generate all the data in 1/10'th of the spin and then just waste rest of the time by copying data from ram to GPIO 2013-02-07T19:59:38 < dirty_d> thats cool 2013-02-07T19:59:51 < jpa-> that was the most boring "demo" ever made 2013-02-07T20:00:04 < dirty_d> it looks cool 2013-02-07T20:00:11 < jpa-> it was cool as a hardware 2013-02-07T20:00:21 < jpa-> i wasn't good enough on the creative side :) 2013-02-07T20:00:40 < jpa-> (128x16 pixels 3 colors is a bit limiting :) 2013-02-07T20:01:33 < dirty_d> yea 2013-02-07T20:02:16 -!- DaKu is now known as daku 2013-02-07T20:02:27 -!- daku is now known as DaKu 2013-02-07T20:02:47 < karlp> emeb: just took four samples and shifted right by 2, and it's more reliably detecting small changes now. 2013-02-07T20:02:55 < karlp> never actually done that myself before. 2013-02-07T20:03:16 < dirty_d> yea thats pretty neat 2013-02-07T20:03:19 < emeb> integrate & dump baby 2013-02-07T20:03:44 < emeb> or, get fancier and do a moving average window - higher output rate, same integration. 2013-02-07T20:03:46 < karlp> my head hurts when I try and understand most dsp code and theories though. 2013-02-07T20:03:52 < jpa-> karlp: at work, on first gen prototype i had a light sensor gain totally wrong.. so the ADC values were 0..2 range.. i oversampled 1000x and it actually gave quite sensible data 2013-02-07T20:04:35 -!- DaKu is now known as daku 2013-02-07T20:05:07 < dirty_d> a little noise is actually helpful when you oversample right 2013-02-07T20:05:11 < emeb> yep 2013-02-07T20:05:13 < dirty_d> enough to toggle the LSB 2013-02-07T20:05:13 < karlp> essential 2013-02-07T20:05:30 < jpa-> fortunately stm32 adc has +-2 noise usually :) 2013-02-07T20:05:35 < karlp> the st app notes even suggest how to use the dac to add white noise if your sample doesn't have enough 2013-02-07T20:05:45 < karlp> but it does not that "normally, the internal adc will have enough noise" 2013-02-07T20:05:50 < dirty_d> cool 2013-02-07T20:06:33 < emeb> the noise generator in the STM32 DAC is badly done. 2013-02-07T20:06:34 < Laurenceb> on some of my projects i use qsort 2013-02-07T20:06:50 < Laurenceb> then chuck the outliers and mean the interquartiles 2013-02-07T20:07:29 < jpa-> i have written a median buffer class that keeps a sorted linked list of past N samples.. it's somewhat faster than sorting every tie 2013-02-07T20:07:31 < jpa-> *time 2013-02-07T20:08:13 < emeb> nifty 2013-02-07T20:08:28 < Laurenceb> thats neat 2013-02-07T20:08:39 < Laurenceb> but im running every 10ms with ~1msps data 2013-02-07T20:08:43 < jpa-> not that much faster though.. a tree structure would probably be optimal 2013-02-07T20:09:08 < jpa-> or in reality, a histogram could work also 2013-02-07T20:10:59 < dirty_d> for weeding out bad samples? 2013-02-07T20:11:11 < jpa-> yeah for median/percentile filtering 2013-02-07T20:11:22 < jpa-> though it's not always necessary 2013-02-07T20:11:48 < jpa-> another one i use quite often is "median3(a,b,c)".. it's just a if construct to pick the middle one of three samples 2013-02-07T20:11:50 < dirty_d> cant you just compare it to the filtered current value? 2013-02-07T20:12:24 < jpa-> that will be problematic on step edges 2013-02-07T20:14:10 < dirty_d> yea 2013-02-07T20:19:01 -!- neuro_sys is now known as neuro_sys|off 2013-02-07T20:20:34 < BrainDamage> http://snarxiv.org/vs-arxiv/ 2013-02-07T20:21:17 < dirty_d> karlp, wait if you only take 4 samples and shift right by 2 youre not oversampling 2013-02-07T20:21:33 < jpa-> sure you are 2013-02-07T20:21:43 < dirty_d> you still only have 12 bits 2013-02-07T20:21:56 < dirty_d> because youre multiplying by 4, then dividing by 4 2013-02-07T20:21:56 < jpa-> but you have average of 4 values, which reduces noise 2013-02-07T20:22:06 < dirty_d> you take 16 samples and shift right by 2 2013-02-07T20:22:33 < dirty_d> whats that give you? 2013-02-07T20:22:35 < dirty_d> 2 more bits 2013-02-07T20:23:14 < emeb> BrainDamage: funny - was on a roll for a while then got a whole bunch of them wrong. Got cocky. 2013-02-07T20:24:00 < emeb> Laurenceb: working on ADC board for FPGA-based front-end: http://imagebin.org/245829 2013-02-07T20:24:28 < jpa-> dirty_d: sure, if you can afford 16 samples oversampling 2013-02-07T20:24:31 < Laurenceb> groovy 2013-02-07T20:24:45 < Laurenceb> emeb: but whats so good about ADc based SDR? 2013-02-07T20:24:59 < emeb> wat? 2013-02-07T20:25:06 < Laurenceb> why not try making an upconverting low frequency thing or something? 2013-02-07T20:25:13 < jpa-> however "take 4, don't shift right" is pretty much useless, as the noise would be double so you'd really gain at most 1 bit 2013-02-07T20:25:14 < Laurenceb> why fast ADC? 2013-02-07T20:25:17 < dirty_d> jpa-, for some things you have all the time in the world 2013-02-07T20:25:21 < emeb> want BANDWIDTH!!! 2013-02-07T20:25:25 < dirty_d> like measuring battery voltage 2013-02-07T20:25:28 < Laurenceb> lol 2013-02-07T20:25:42 < BrainDamage> you can sample the whole channel spectrum at the same time 2013-02-07T20:26:06 < Laurenceb> yeah but you need a decent processor to do anything with it 2013-02-07T20:26:10 < emeb> Laurenceb: but for sensitivity it's still better to do tuning/filtering in analog 2013-02-07T20:26:20 < Laurenceb> yes 2013-02-07T20:26:45 < emeb> ya - FPGA needed to do front-end tuning/filtering. Maybe even demod. 2013-02-07T20:26:50 < Laurenceb> i was thinking of making an RF thermometer 2013-02-07T20:27:04 < Laurenceb> annoyingly ultra wideband isnt feasible 2013-02-07T20:27:18 < Laurenceb> so youd need to make it narrow band at some clean frequency 2013-02-07T20:27:38 < emeb> what - using blackbody? 2013-02-07T20:27:57 < Laurenceb> yes 2013-02-07T20:28:03 < Laurenceb> then antenni array 2013-02-07T20:28:11 < emeb> hmm... finding a "clean frequency" is the trick. 2013-02-07T20:28:20 < Laurenceb> for a synthetic aperture building penetrating person finder 2013-02-07T20:28:24 < Laurenceb> terminator tech 2013-02-07T20:28:32 < emeb> ya 2013-02-07T20:28:49 < emeb> plus, delta-T of person & surroundings can be pretty small in some cases. 2013-02-07T20:28:55 < Laurenceb> the front ends rapidly get very hardcore 2013-02-07T20:29:12 < Laurenceb> it might work on a uav so you get some doppler 2013-02-07T20:29:23 < Laurenceb> i never worked out all the maths 2013-02-07T20:29:36 < emeb> sounds like a complex problem. 2013-02-07T20:30:09 < Laurenceb> you dont know the emitted signal 2013-02-07T20:30:41 < Laurenceb> i think this means you need a 2D array and scan using fourier slice theorem 2013-02-07T20:33:06 < emeb> radon transform! 2013-02-07T20:33:28 < dirty_d> are you from the future? 2013-02-07T20:36:11 -!- Lionhearted [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has joined ##stm32 2013-02-07T20:36:16 < Lionhearted> gday...... 2013-02-07T20:36:40 < Lionhearted> can you tell me where to find the largest community for arm uc's ? thanks. 2013-02-07T20:38:53 -!- Lionhearted [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has left ##stm32 [] 2013-02-07T20:39:45 < emeb> I guess we weren't "large" enough for him. 2013-02-07T20:46:23 -!- daku is now known as DaKu 2013-02-07T20:48:36 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-07T20:50:27 -!- enots [dimka@freelsd.net] has quit [Ping timeout: 240 seconds] 2013-02-07T20:57:07 < karlp> you take 4^n samples and shift right by 2^n, to get n more bits. 2013-02-07T20:57:19 < karlp> or so the st app note says 2013-02-07T20:58:14 < emeb> yes - you pick up 6dB for every 4x integration 2013-02-07T20:58:25 < emeb> 6dB ~= 1bit 2013-02-07T21:02:34 < jpa-> yep 2013-02-07T21:13:36 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 252 seconds] 2013-02-07T21:13:56 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-07T21:25:38 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-07T21:31:29 -!- Erlkoenig [~erlkoenig@pptp-212-201-74-82.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-07T21:31:57 -!- enots [dimka@freelsd.net] has joined ##stm32 2013-02-07T21:41:28 < CheBuzz> The "largest" would probably be #android, no? 2013-02-07T21:42:31 < gxti> but they probably don't much care for any hardware that hasn't been leased to them by a telco 2013-02-07T21:43:23 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-07T21:54:06 -!- Irssi: ##stm32: Total of 72 nicks [1 ops, 0 halfops, 1 voices, 70 normal] 2013-02-07T21:55:09 <+Steffanx> dirty_d never ever mention you are in #avr here :P 2013-02-07T21:56:17 < dirty_d> Steffanx, lol 2013-02-07T21:56:23 < dirty_d> you guys dont like avrs? 2013-02-07T21:56:55 <+Steffanx> Some don't 2013-02-07T21:58:06 < zyp> 8-bit microcontrollers are pretty irrelevant nowadays 2013-02-07T21:59:21 < gxti> unless you're selling 10k units yeah 2013-02-07T21:59:24 < gxti> forget about it 2013-02-07T22:00:46 < dirty_d> truth 2013-02-07T22:04:21 < zyp> unless you are on the cheap end, they aren't really competitive on price either 2013-02-07T22:05:16 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-07T22:05:23 < dirty_d> the last avr i used was pretty neat, an xmega for about $3 2013-02-07T22:05:36 < dirty_d> but i do like stm32 more now 2013-02-07T22:06:25 < zyp> I have a box of atmega32u4 leftovers from my last avr project 2013-02-07T22:06:29 < zyp> hmm 2013-02-07T22:06:37 < zyp> they are apparently $6 each on digikey 2013-02-07T22:07:17 < zyp> and easily beat on all points by a cheaper stm32 2013-02-07T22:07:25 < dirty_d> yea 2013-02-07T22:07:33 < dirty_d> and even by xmegas of the same brand 2013-02-07T22:07:35 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] 2013-02-07T22:07:45 < zyp> do you get xmegas with usb? 2013-02-07T22:07:51 -!- barthess [~barthess@77.67.196.60] has quit [Quit: Leaving.] 2013-02-07T22:07:52 < dirty_d> did you mean atxmegaa4u? 2013-02-07T22:08:03 < dirty_d> ive only used one, and it did have usb 2013-02-07T22:08:05 < dirty_d> that one 2013-02-07T22:08:10 < dirty_d> atxmega32a4u 2013-02-07T22:08:11 < zyp> what's the deal with xmega really? how do they differ architecturally from normal megas? 2013-02-07T22:08:17 -!- barthess [~barthess@77.67.196.60] has joined ##stm32 2013-02-07T22:08:33 < dirty_d> the layout of peripherals and registers and stuff just makes more sense 2013-02-07T22:08:46 < dirty_d> you have more options for which pins do what function etc 2013-02-07T22:09:29 < dirty_d> the peripherals are so much better than the regular mega's ones too 2013-02-07T22:09:51 < dirty_d> i think i had 16 x 16-bit pwm channels 2013-02-07T22:10:03 < dirty_d> much faster adc etc 2013-02-07T22:11:36 < zyp> but the core architecture is the same? 2013-02-07T22:12:05 < zyp> oh, dma and interrupt priorities 2013-02-07T22:12:15 <+Steffanx> it's more an AVR with some ARM-ish features 2013-02-07T22:12:27 <+Steffanx> Still the same instruction set etc. 2013-02-07T22:12:35 < zyp> yeah 2013-02-07T22:12:39 <+Steffanx> + some extra. AES and DES stuff 2013-02-07T22:12:42 < zyp> well, still 8bit 2013-02-07T22:13:01 <+Steffanx> They say 8/16-bit on their webstie :D:D 2013-02-07T22:13:38 < dirty_d> zyp, yea its the same 2013-02-07T22:14:08 < dirty_d> i think the only reason its /16 is that some regsiters are 16-bits 2013-02-07T22:14:13 <+Steffanx> The one fancy features is the event system 2013-02-07T22:14:18 < dirty_d> but its really two 8 bit regs technically i guess 2013-02-07T22:14:25 < dirty_d> Steffanx, yea that was nice 2013-02-07T22:14:28 < zyp> Steffanx, lpc43xx also has something like that 2013-02-07T22:14:39 <+Steffanx> Do stuff without wasting cpu cycles? 2013-02-07T22:14:41 < dirty_d> i used the event system for my RC reciever decoding 2013-02-07T22:14:47 < dirty_d> no CPU involved 2013-02-07T22:14:51 < dirty_d> same on stm32 though 2013-02-07T22:14:52 <+Steffanx> or without even a running cpu 2013-02-07T22:15:05 <+Steffanx> or whatever it is capable of doing 2013-02-07T22:15:29 < dirty_d> same deal as on stm32 for input capture etc 2013-02-07T22:15:42 < dirty_d> or when the comparator clears compare output 2013-02-07T22:16:04 < dirty_d> it might have been a littel more configurable though 2013-02-07T22:16:15 < dirty_d> i think you can you any event to trigger anything else 2013-02-07T22:16:21 < dirty_d> not just certain ones 2013-02-07T22:16:42 < zyp> efm32 also has something like that 2013-02-07T22:17:23 < zyp> http://www.energymicro.com/technology/peripheral-reflex-system <- this stuff 2013-02-07T22:17:23 <+Steffanx> efm32 isn't alive yet is it? 2013-02-07T22:17:29 < zyp> sure it is 2013-02-07T22:17:37 < zyp> or so I've heard 2013-02-07T22:17:42 <+Steffanx> Yeah, looks similar 2013-02-07T22:18:00 < jpa-> on STM32 you can DMA to peripheral control registers to trigger stuff :P 2013-02-07T22:18:03 < zyp> but I'm more interested in the upcoming efr4d-range 2013-02-07T22:18:09 < gxti> you can certainly buy it, the question is whether you can make good use of it without using wacky vendor tools 2013-02-07T22:18:16 <+Steffanx> oh, the rf stuff isn't avaiable yet indeed 2013-02-07T22:18:25 < gxti> k 2013-02-07T22:18:41 < zyp> Steffanx, well, I said efm32, not efr4d :p 2013-02-07T22:18:44 <+Steffanx> it's m3/m4 so .. looks ok to me gxti :) 2013-02-07T22:18:57 < dirty_d> jpa-, i need to learn about that 2013-02-07T22:18:58 < gxti> it's not the core you have to worry about 2013-02-07T22:19:00 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-07T22:19:08 <+Steffanx> Doesn't look like "using wacky vendor tools" to me 2013-02-07T22:19:19 <+Steffanx> I never worry about anything gxti 2013-02-07T22:19:28 < gxti> i noticed 2013-02-07T22:19:52 <+Steffanx> I only care about your guy caring to much about pointless matters 2013-02-07T22:19:57 <+Steffanx> *you 2013-02-07T22:20:14 <+Steffanx> *guys 2013-02-07T22:20:16 <+Steffanx> man, english 2013-02-07T22:20:21 < gxti> you just don't care 2013-02-07T22:20:23 < gxti> about the english 2013-02-07T22:20:46 < gxti> silly euros 2013-02-07T22:21:13 <+Steffanx> Yes boss 2013-02-07T22:24:45 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-07T22:32:43 -!- jef79m_lurking is now known as jef79m 2013-02-07T22:34:57 < dirty_d> jpa-, so basically what the DMA system lets you do is transfer data from one address to another both of your choosing when a specific event occurs right? 2013-02-07T22:35:09 < dirty_d> im reading about it now 2013-02-07T22:35:43 < BrainDamage> Laurenceb: http://www.ebay.it/itm/100-TRILLION-ZIMBABWE-DOLLARS-500-MILLION-BOSNIA-DINAR-/330521744238?pt=Paper_Money&hash=item4cf4a1936e&_uhb=1#ht_1834wt_1377 2013-02-07T22:38:48 -!- Laurenceb_ [~Laurence@host86-143-184-68.range86-143.btcentralplus.com] has joined ##stm32 2013-02-07T22:39:26 <+Steffanx> The paper is worth more than the actually value :D 2013-02-07T22:40:33 < gxti> dat mustache 2013-02-07T22:42:20 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-07T22:45:36 <+Steffanx> gxti is jealous 2013-02-07T22:45:42 < gxti> hell yeah i am 2013-02-07T22:46:14 < Laurenceb_> mustachio? 2013-02-07T22:46:19 < Laurenceb_> zlog 2013-02-07T22:46:19 < zlog> Laurenceb_: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2013-02-07.html 2013-02-07T22:52:32 < mervaka> karlp: you about? 2013-02-07T22:52:49 < mervaka> back on my chair again :P 2013-02-07T22:53:47 < Laurenceb_> hmm 2013-02-07T22:53:48 < Laurenceb_> http://www.raspberrypi.org/wp-content/uploads/2013/02/robjamesoffice.jpg 2013-02-07T22:54:01 < Laurenceb_> guy in the background is behind N-prize aiui 2013-02-07T22:54:51 <+Steffanx> ok, thanks for hte info 2013-02-07T22:55:24 < Laurenceb_> useful for trolling if you ever meet irl 2013-02-07T22:55:33 <+Steffanx> I don't think i'll remember his face 2013-02-07T22:58:09 < Laurenceb_> "One of the teams (Potent Voyager) were planning to recover hydrogen from the balloon for use as second-stage fuel" 2013-02-07T22:58:14 < Laurenceb_> http://www.halfbakery.com/idea/N-Prize 2013-02-07T22:58:16 < Laurenceb_> i lolled 2013-02-07T23:02:21 < Laurenceb_> http://www.b3tards.com/u/037e58b9e054b3c8dd7b/morphgoatse.gif 2013-02-07T23:03:28 -!- barthess [~barthess@77.67.196.60] has quit [Quit: Leaving.] 2013-02-07T23:04:45 -!- Ranewen [~Ranewen4@87.252.128.227] has joined ##stm32 2013-02-07T23:05:52 < karlp> mervaka: yeah man. 2013-02-07T23:05:58 < karlp> I actually "fixed" it. 2013-02-07T23:06:10 < karlp> I was using a timer interrupt to turn off the dma after 100ms. 2013-02-07T23:06:22 < karlp> I stopped doing that and let the adc->dma just run forever 2013-02-07T23:06:27 < karlp> and now it stays in the right channel order. 2013-02-07T23:06:39 < karlp> probably doing a single extran channel or something. 2013-02-07T23:06:42 * karlp shrugs 2013-02-07T23:08:19 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-07T23:10:52 < Laurenceb_> https://my.st.com/cas/login?service=https://my.st.com/esample/app%3Fpage%3Dredirecting%26pn%3DLPS331APTR 2013-02-07T23:11:12 < karlp> what is that? 2013-02-07T23:11:32 < karlp> don't make me login! 2013-02-07T23:11:41 < mervaka> huh 2013-02-07T23:12:00 < mervaka> you enabled circular buffer? 2013-02-07T23:12:10 < karlp> yeah 2013-02-07T23:12:12 < mervaka> cool 2013-02-07T23:12:30 < karlp> if I don't turn off the dma channel and restart it, it stays nicely ordered in the dma buffer. 2013-02-07T23:12:36 < Laurenceb_> karlp: seems like free pressure sensors 2013-02-07T23:12:37 < mervaka> how do you mean by channel order? as in the left/right interleaving? 2013-02-07T23:12:45 -!- jef79m is now known as jef79m_lurking 2013-02-07T23:12:49 < karlp> no, doing X channels in a sequence. 2013-02-07T23:12:58 < mervaka> huh 2013-02-07T23:13:00 < karlp> I was seeing them rotated in the buffer, 3,0,1,2 2013-02-07T23:13:08 < mervaka> you have TDM working? 2013-02-07T23:13:10 < karlp> but only after turning off the dma and restarting it later. 2013-02-07T23:13:27 < karlp> now it's just continuously dma-ing, and stays nicely in 0,1,2,3 order, 2013-02-07T23:13:37 < karlp> and in the 100ms timer irq I just copy some values out. 2013-02-07T23:13:42 < Laurenceb_> it is free pressure sensors 2013-02-07T23:13:44 < Laurenceb_> nice 2013-02-07T23:13:46 < mervaka> hang on 2013-02-07T23:13:48 < mervaka> you have TDM working? 2013-02-07T23:14:08 < karlp> I'm just writing a bunch of channels to the SQRx registers 2013-02-07T23:14:20 < karlp> and doing scan+continuous 2013-02-07T23:14:32 < mervaka> oh, hang on 2013-02-07T23:14:36 < mervaka> you're using the onboard ADC 2013-02-07T23:14:41 < karlp> yeah. 2013-02-07T23:14:42 < mervaka> i understand now 2013-02-07T23:14:58 < mervaka> i was thinking you were using the I2S standard, TDM 2013-02-07T23:15:06 < karlp> sorry, emebhad just said that I might have been seeing what you were seeing or something 2013-02-07T23:15:06 < mervaka> when someone says multichannel to me, i think of that. 2013-02-07T23:15:15 < karlp> but I don't remember what you were doing exactly 2013-02-07T23:15:19 < mervaka> i'm interfacing with the SPI peripherals 2013-02-07T23:15:24 < mervaka> audio project 2013-02-07T23:15:29 < karlp> well, I'm doing 3-12 channel, depending on what's connected. 2013-02-07T23:15:35 < karlp> with the onboard adc 2013-02-07T23:15:36 -!- neuro_sys|off is now known as neuro_sys 2013-02-07T23:15:39 < karlp> ok, never mind then :) 2013-02-07T23:15:51 < karlp> you were still getting glitches though right? 2013-02-07T23:16:04 < mervaka> i have problems with the interrupt sequence. nvic seems to totally fuck up. 2013-02-07T23:16:05 < karlp> Laurenceb_: got a better link? 2013-02-07T23:16:06 < mervaka> yeah 2013-02-07T23:16:18 < mervaka> interrupt related woes. 2013-02-07T23:16:18 < karlp> I'm logged in, but I'm not seeing anything free anywhere 2013-02-07T23:16:30 < ds2> Got a shell on the stm32f1! 2013-02-07T23:16:39 * karlp cheers 2013-02-07T23:16:40 < ds2> still some funny business with the memory though 2013-02-07T23:16:42 < Laurenceb_> karlp: go to samples 2013-02-07T23:16:43 < mervaka> ideally i'd be using a single clock, and a single interrupt. 2013-02-07T23:16:56 < Laurenceb_> ds2: linux? 2013-02-07T23:16:58 < mervaka> and perhaps even a single data bus 2013-02-07T23:17:01 < karlp> Sorry, you are encountering flash plugin problem / browser issue. Our system works better with IE & Firefox. 2013-02-07T23:17:04 < ds2> Laurenceb_: yep. uCLinux 2013-02-07T23:17:12 < mervaka> ds2: nice! 2013-02-07T23:17:16 < Laurenceb_> awesome 2013-02-07T23:17:21 < ds2> Laurenceb_: you know the innerards of uCLinux memory management? 2013-02-07T23:17:27 < Laurenceb_> nope 2013-02-07T23:17:30 < ds2> oh 2013-02-07T23:17:35 < mervaka> is that an RTOS? 2013-02-07T23:17:41 < ds2> guess I will learn about pfn_to_page() tonight :) 2013-02-07T23:17:49 < karlp> LPS331APTR Laurenceb_? 2013-02-07T23:17:50 < ds2> mervaka: depends on how RT you need :D 2013-02-07T23:17:59 < mervaka> as in, a single process. 2013-02-07T23:18:03 < gxti> on geological timescales all OS are RTOS 2013-02-07T23:18:06 < gxti> :P 2013-02-07T23:18:32 < Laurenceb_> karlp: yes 2013-02-07T23:20:48 < zyp> hmm 2013-02-07T23:21:07 < zyp> reminds me that I need to find time to test the sdram on the lpc board soon 2013-02-07T23:23:06 < Ranewen> how is the common smd "called" for arms ? 2013-02-07T23:23:15 < Ranewen> qtf? or how? 2013-02-07T23:23:17 < ds2> SMD? 2013-02-07T23:23:19 < BrainDamage> qfp? 2013-02-07T23:23:21 < gxti> wtf 2013-02-07T23:23:28 < gxti> omg 2013-02-07T23:23:29 < Ranewen> qtfp package or ? 2013-02-07T23:23:30 < zyp> huh? 2013-02-07T23:23:31 < BrainDamage> quad flat plackage 2013-02-07T23:23:34 < gxti> Ranewen: tqfp 2013-02-07T23:23:37 < ds2> what are you taking about? 2013-02-07T23:23:57 < BrainDamage> he's asking the common smd package for cortex chips 2013-02-07T23:24:06 < BrainDamage> so it'd be qfp 2013-02-07T23:24:10 < ds2> BGA 2013-02-07T23:24:17 < BrainDamage> and it's variants, l, t, etc 2013-02-07T23:24:19 < gxti> CDIP oh wait 2013-02-07T23:24:19 < zyp> there are several common ones 2013-02-07T23:24:32 < Ranewen> sorry ! i only worked with DIL :( and with crappy chips.. 2013-02-07T23:24:36 < BrainDamage> yes, I said qfp because it's the one that generally hands solders 2013-02-07T23:24:39 < zyp> the qfp group may arguably be the most common 2013-02-07T23:24:54 < ds2> aren't there SOIC packaged ARMs? 2013-02-07T23:24:58 < ds2> or maybe TSSOP 2013-02-07T23:25:04 < zyp> lpc have some 2013-02-07T23:25:23 < zyp> some low pin count ones 2013-02-07T23:25:31 < gxti> same one as they do in DIP probably 2013-02-07T23:25:34 < gxti> m0 stuff 2013-02-07T23:25:38 < ds2> no external memory 2013-02-07T23:25:42 < zyp> yep 2013-02-07T23:26:08 < zyp> well, the problem with pins on only two sides is the number of pins you can fit 2013-02-07T23:26:17 < Ranewen> i think i was somewhere an smd or tqfp to DIP converters 2013-02-07T23:26:27 < Ranewen> *i think i saw* 2013-02-07T23:26:35 < zyp> sure, there are lots of those 2013-02-07T23:26:47 < Ranewen> that would allow me to start studying on the cheap 2013-02-07T23:26:51 < BrainDamage> smd is too broad of a package family 2013-02-07T23:27:04 < BrainDamage> anyway yes 2013-02-07T23:27:10 < BrainDamage> just go on ebay, and search for like 2013-02-07T23:27:19 < BrainDamage> qfp-64 breakout 2013-02-07T23:27:25 < zyp> I have a bunch of different ones, that I bought because they could be nice to have, but every time I needed one I was always missing one with both correct pin count and correct pitch :p 2013-02-07T23:27:30 < BrainDamage> or anyway qfp-pincount breakout 2013-02-07T23:27:40 < ds2> blah 2013-02-07T23:27:43 < ds2> dip..yuck 2013-02-07T23:27:52 < ds2> waste of space 2013-02-07T23:28:07 < gxti> not everyone is as pro as you ds2 2013-02-07T23:28:13 < BrainDamage> that sensation when i have an exam tomorrow, I should still review some things and https://gs1.wac.edgecastcdn.net/8019B6/data.tumblr.com/tumblr_m6cmkh8Xv41rwy263o1_500.gif 2013-02-07T23:28:23 < ds2> gxti: SMDs are so much easier to proto then DIPs 2013-02-07T23:28:31 < gxti> k 2013-02-07T23:28:39 < ds2> DIPs are painful...got to drill and wire 2013-02-07T23:28:51 < ds2> some copper tape, blank board and you have SMD stuff going 2013-02-07T23:29:02 < zyp> I think the last time I was looking for a breakout was that time I was hooking up a cpld 2013-02-07T23:29:09 < zyp> ended up with this instead: http://bin.jvnv.net/f/qHoay.JPG 2013-02-07T23:29:15 < BrainDamage> dips are painful when you walk over one 2013-02-07T23:30:04 < Ranewen> zyp, very nice soldering 2013-02-07T23:31:45 < ds2> think the worse part of doing DIP is sticking the chips in one side, flip it over, chip falls out. if you bend a pin to keep it from falling out, it is hell on removing it later 2013-02-07T23:32:03 < zyp> heh 2013-02-07T23:32:10 < dirty_d> tqfp is nice 2013-02-07T23:32:19 < dirty_d> or lqfp 2013-02-07T23:32:23 < dirty_d> probably any qfp! 2013-02-07T23:32:27 < ds2> DIPs and TH parts was an unpleasnant period 2013-02-07T23:32:38 < dirty_d> qfn, not so much 2013-02-07T23:32:53 < BrainDamage> whenever I have some leftover dips 2013-02-07T23:32:54 < GargantuaSauce> i am still too much of a wuss to graduate from through hole 2013-02-07T23:32:57 < gxti> qfn is not scary but it does take too damn long to hand-solder 2013-02-07T23:32:57 < ds2> qfns aren't too bad. hot air seems to take over it 2013-02-07T23:33:00 < BrainDamage> I bend the legs 90° 2013-02-07T23:33:06 < BrainDamage> and solder them smd 2013-02-07T23:33:08 < gxti> i need hot air. 2013-02-07T23:33:18 < dirty_d> you can do qfn with hot air easily 2013-02-07T23:33:24 < BrainDamage> mostly because I diy pcb and cba to drill 2013-02-07T23:33:25 < dirty_d> i just dont like not seeing whats happening 2013-02-07T23:33:32 < dirty_d> i dont trust it 2013-02-07T23:33:33 < zyp> ds2, hold the board between the middle finger and ring finger on your non-dominant hand, use ring finger to keep chip in place, hold solder between thumb and index finger on same hand, then solder 2013-02-07T23:33:40 < dirty_d> id rather see whats going on 2013-02-07T23:33:46 < BrainDamage> if you ask why I diy pcb, because it takes me fast enough to get and cheap enough 2013-02-07T23:34:09 < zyp> you only need to use thumb and index finger to hold the solder, so use the other fingers to keep the components in place with soldering 2013-02-07T23:34:24 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-07T23:35:21 < zyp> while* 2013-02-07T23:35:33 < ds2> zyp: soldering acrobatics? 2013-02-07T23:35:40 < zyp> you don't do that? 2013-02-07T23:35:52 < ds2> I am getting old... not as nimble as I used to be 2013-02-07T23:36:04 < ds2> SMD doesn't require that sort of thing 2013-02-07T23:36:35 < zyp> still need to keep the component in place 2013-02-07T23:36:46 < dirty_d> i just tack 2 corners 2013-02-07T23:36:51 < dirty_d> while i hold it 2013-02-07T23:36:53 < dirty_d> then do the rest 2013-02-07T23:36:57 < emeb> OK - got a DMA question: buddy of mine is using an F405 to send I2S to two audio DACs. 2013-02-07T23:37:03 < ds2> if the board is flat, gravity works for me 2013-02-07T23:37:12 < zyp> oh well 2013-02-07T23:37:23 < emeb> but both I2S2 and I2S3 are on DMA controller #1 - different streams, but same channel. 2013-02-07T23:37:38 < emeb> That shouldn't cause any problems though should it? 2013-02-07T23:37:39 < dirty_d> rosin is pretty sticky anyway 2013-02-07T23:38:29 < emeb> for some reason when he tries to run both I2S ports it doesn't work. No idea why not. 2013-02-07T23:38:34 < dirty_d> emeb, from what i just red, either will trigger the DMA request 2013-02-07T23:38:36 -!- LeelooMinai [~leeloo@198.84.185.212] has joined ##stm32 2013-02-07T23:38:41 < dirty_d> if thats ok, then thats ok 2013-02-07T23:39:20 < emeb> Looking at the RM it seems that the streams are unique and channels between different streams are unrelated. 2013-02-07T23:39:21 -!- Ranewen [~Ranewen4@87.252.128.227] has quit [Quit: Leaving] 2013-02-07T23:39:28 < zyp> I remember in school once, a teacher was laughing at me because he saw me using my mouth as a third hand for applying the solder because I had the hands full :p 2013-02-07T23:39:31 -!- LeelooMinai [~leeloo@198.84.185.212] has quit [Remote host closed the connection] 2013-02-07T23:39:38 < dirty_d> haha 2013-02-07T23:39:42 < emeb> mmm... lead poisoning. 2013-02-07T23:39:51 < zyp> I haven't died yet :) 2013-02-07T23:39:58 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 256 seconds] 2013-02-07T23:40:03 < emeb> risk may be somewhat overstated. 2013-02-07T23:40:13 < ds2> I used to do that 2013-02-07T23:40:24 * emeb grew up around lots of molten lead / solder / etc. Hasn't gotten too twitchy. 2013-02-07T23:40:37 < zyp> yeah, so did I 2013-02-07T23:40:59 < emeb> used to chew on lead pipes. :P 2013-02-07T23:41:07 < emeb> when I was small & stupid. 2013-02-07T23:41:07 < zyp> that I didn't do :p 2013-02-07T23:41:46 < emeb> not small anymore. don't know about stupid tho... 2013-02-07T23:42:27 < emeb> if 2 dma streams on the same controller have the same priority, what happens if they both request at the same time? 2013-02-07T23:42:32 < dirty_d> i spilled molten lead all over my hand when i was little 2013-02-07T23:42:33 < dirty_d> lol 2013-02-07T23:42:44 < emeb> ow 2013-02-07T23:42:45 < dirty_d> it solidified on my skin 2013-02-07T23:42:50 < zyp> my dad's an electrician that dabbles some in electronics on the side 2013-02-07T23:42:51 < dirty_d> it wasnt that bad of a burn though 2013-02-07T23:42:52 < gxti> emeb: they have an innate priority iirc 2013-02-07T23:42:52 < emeb> get a scar? 2013-02-07T23:42:54 < dirty_d> nope 2013-02-07T23:43:00 < zyp> I was quite young when he learned me how to solder :p 2013-02-07T23:43:02 < dirty_d> 2nd degree maybe 2013-02-07T23:43:20 < emeb> gxti: yeah - seems like streams are round-robin priority or something. 2013-02-07T23:43:30 < dirty_d> i used to do dangerous stuff as a kid 2013-02-07T23:43:41 < zyp> doesn't all kids do? 2013-02-07T23:43:49 < ds2> dirty_d: were you the radioactive boyscout? :D 2013-02-07T23:43:55 < dirty_d> yea in different ways i guess, lol 2013-02-07T23:44:06 < emeb> all kids are immortal until proven otherwise. 2013-02-07T23:44:44 < dirty_d> well its gtfo time again 2013-02-07T23:44:50 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-07T23:45:24 < BrainDamage> wait, you guys actually stopped to do dangerous stuff? 2013-02-07T23:47:06 < zyp> from what I've heard, I was around two years old the first time I got an electric shock 2013-02-07T23:47:57 < ds2> 220V? 2013-02-07T23:48:04 < zyp> yeah 2013-02-07T23:48:57 < zyp> little me discovered that putting metal into a power outlet was a good idea 2013-02-07T23:49:21 < ds2> what kind of outlets did you guys have? 2013-02-07T23:49:32 < ds2> thought the europeon ones had safetly interlocks to prevent that? 2013-02-07T23:49:33 < zyp> the ones that accept two round pins 2013-02-07T23:49:42 < zyp> nowadays they do 2013-02-07T23:49:53 < zyp> in 1990 that was not so common :p 2013-02-07T23:50:15 < gxti> haha you're the same age as me. loser. 2013-02-07T23:50:32 < ds2> wow. didn't know you guys been changing outlets that often 2013-02-07T23:50:32 < mervaka> awesome, my BZT algorithm's working great. why is maths so much easier than peripheral implementation? :/ 2013-02-07T23:50:41 < ds2> don't think ours here really changed 2013-02-07T23:50:41 < zyp> they didn't change 2013-02-07T23:51:14 < gxti> the basic interoperation of the outlet hasn't changed, they just have to have extra safety features 2013-02-07T23:51:19 < zyp> form factor is the same, they just added the interlock feature 2013-02-07T23:51:25 < gxti> but probably there aren't any outlet police that go around fining people who haven't rewired their house 2013-02-07T23:51:42 < gxti> still lots of old shit in the US with 2-prong unpolarized outlets :p 2013-02-07T23:51:58 < Laurenceb_> problem? 2013-02-07T23:51:58 < zyp> actually they do 2013-02-07T23:51:59 < ds2> Ohhh 2013-02-07T23:52:05 < gxti> but only old people and hipsters have to deal with that 2013-02-07T23:52:09 < Laurenceb_> id rather not have outlet police 2013-02-07T23:52:27 < zyp> but having safety outlets is not really required or something 2013-02-07T23:52:38 < zyp> but they check earthing and stuff like that 2013-02-07T23:52:43 < BrainDamage> well, usually people care about their kids 2013-02-07T23:52:50 < zyp> nowadays all outlets are supposed to be earthed 2013-02-07T23:53:00 < Laurenceb_> here you only have to have it checked for rented houses 2013-02-07T23:53:33 < zyp> but unearthed outlets are allowed as long as they are not withing reasonable range from an earthed outlet 2013-02-07T23:53:36 < gxti> i did have to rewire my dryer for the older 3-prong outlet, but that's kind of a special case as they're the high-current 240V ones 2013-02-07T23:53:52 < gxti> i would have replaced the outlet myself even though i rent, except there was no earth wire behind the plate of course 2013-02-07T23:54:45 < zyp> I was working as an electrician for a year before I went to uni, handled a fair amount of those «we had an inspection and these violations were found»-cases 2013-02-07T23:54:59 < zyp> mostly due to stricter earthing regulations 2013-02-07T23:55:27 < gxti> around here i'm pretty sure as long as you leave stuff alone you don't get in trouble, but if you remodel or something then you might end up having to rewire the whole room 2013-02-07T23:55:37 < Thorn> new qtserialport seems to work well (knock on wood) 2013-02-07T23:55:40 < gxti> anything that needs inspecting 2013-02-07T23:55:52 < zyp> gxti, yeah, we also have some rules like that 2013-02-07T23:56:08 < gxti> but afaik you don't have to have an inspection for any reason other than making major changes to the house 2013-02-07T23:56:09 < zyp> some old kind of wiring, that's known to be fragile and horrible 2013-02-07T23:56:16 < BrainDamage> I'm curious if the inspection agencies offices would pass their own tests 2013-02-07T23:56:34 < zyp> if you leave it alone, it's ok, but if you have to move it, you have to replace it entirely --- Day changed Fri Feb 08 2013 2013-02-08T00:00:50 -!- jef79m_lurking is now known as jef79m 2013-02-08T00:01:00 -!- DaKu is now known as daku 2013-02-08T00:01:29 < emeb> my parents have a rental house built >100yrs ago. Has bare wires with ceramic insulators in the walls... 2013-02-08T00:01:58 < ds2> cool... sawzall lottery 2013-02-08T00:02:15 < emeb> big surprise awaits... 2013-02-08T00:08:58 < zyp> working as an electrician was fun, one time I swapped two wires by accident and shorted the power input to a house 2013-02-08T00:09:37 < zyp> silly bugs are not restricted to software :) 2013-02-08T00:11:08 -!- Guest17091 is now known as dirty_d 2013-02-08T00:11:22 < emeb> hardware invented silly bugs long before software was conceived. 2013-02-08T00:11:37 < zyp> was hooking up a power meter 2013-02-08T00:11:54 < zyp> they are usually wired like this: input 1, output 1, input 2, output 2 2013-02-08T00:11:58 < dirty_d> my parents house have cloth insulated wires 2013-02-08T00:12:06 < Thorn> disconnecting neutral is more fun, consumers will see e.g. 380V instead of 220 2013-02-08T00:12:08 < zyp> I wired it like input, input, output, output 2013-02-08T00:12:46 < zyp> Thorn, most installartions in Norway are IT configurations, not TN 2013-02-08T00:13:06 < Thorn> didn't know that, interesting 2013-02-08T00:13:08 < zyp> i.e. 230V delta instead of 400V star 2013-02-08T00:13:15 < Thorn> is it more safe? or less so? 2013-02-08T00:13:49 < zyp> dunno, it's less efficient wrt. power capacity per wire thickness 2013-02-08T00:14:09 < zyp> so some newer installations and lots of industry are using TN 2013-02-08T00:14:44 < zyp> so it's mostly legacy reasons, I guess 2013-02-08T00:16:13 < Thorn> but you do have PE? 2013-02-08T00:16:27 < zyp> yes 2013-02-08T00:16:34 -!- jef79m is now known as jef79m_lurking 2013-02-08T00:17:20 < zyp> even in TN installations, PE is run separately from N, even though they are usually tied together at some point upstream 2013-02-08T00:17:31 < Thorn> that's good because most old installations around here are TN-C or sometimes TN-C-S 2013-02-08T00:17:56 < zyp> I think TN-C-S is the term we use here 2013-02-08T00:18:09 < zyp> combined in to the distribution, separate from 2013-02-08T00:19:04 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-08T00:19:27 < Thorn> e.g. I still don't have PE in my apartment although building wiring has been replaced a couple years ago and PE wires were installed 2013-02-08T00:20:48 < zyp> you mean the outlets now? 2013-02-08T00:21:44 < zyp> except for the kitchen I only have non-earthed outlets in my current apartment 2013-02-08T00:22:19 < zyp> because the house is old, and it was normal to do it like this before 2013-02-08T00:22:25 < Thorn> I don't have any earthed ones currently 2013-02-08T00:23:05 < zyp> before, earthed outlets were only required in rooms with plumbing, since the pipes are earthed 2013-02-08T00:23:27 < gxti> loool 2013-02-08T00:24:10 < zyp> now everything should be earthed, but old unearthed installations are still allowed as long as you don't have earthed outlets within reasonable distance 2013-02-08T00:24:11 < emeb> meh - redundant. 2013-02-08T00:24:34 < emeb> shouldn't need more grounds :P 2013-02-08T00:25:00 < emeb> RF designers say "moar grounds!" 2013-02-08T00:25:11 < zyp> the concern is that if you have a device with a fault, in a non earthed outlet, so the enclosure becomes live, you should not be able to touch both that and something that is earthed 2013-02-08T00:25:33 < zyp> since that'll give you a shock 2013-02-08T00:25:40 < emeb> do you have requirements for GFCIs? 2013-02-08T00:25:54 < zyp> yes 2013-02-08T00:26:01 < zyp> I believe they are required on everything nowadays 2013-02-08T00:26:06 < emeb> funny how those have evolved. 2013-02-08T00:26:14 < zyp> at least we installed them on everything when I worked as an electrician 2013-02-08T00:26:40 < emeb> my house was built in '79 and only has 1 GFCI on a circuit that feeds all bathrooms and outdoor outlets. 2013-02-08T00:26:52 < zyp> same for my parents' house 2013-02-08T00:27:07 < zyp> except it was built in '92 2013-02-08T00:27:11 < emeb> so if some electric yard equipment trips the GFCI I have run to the master bathroom to reset it. 2013-02-08T00:27:18 < dongs> suup 2013-02-08T00:27:48 < zyp> the house I'm currently living in is older, but has a single GFCI in front of all the circuits 2013-02-08T00:27:55 < Thorn> around here most people install a combined breaker/gfci at the input of a panel and that's it 2013-02-08T00:27:57 < dongs> japs don't even know what gfci is 2013-02-08T00:28:00 < emeb> the sleeper awakes. 2013-02-08T00:28:05 < dongs> or grounded outlets 2013-02-08T00:28:26 < emeb> or indoor heating apparently. 2013-02-08T00:28:32 < dongs> yes 2013-02-08T00:28:32 < gxti> garoundu faruto kurrento interuputeroooo 2013-02-08T00:28:45 < dongs> lold 2013-02-08T00:28:48 < zyp> nowadays you get combined current breakers and gfcis as a single device 2013-02-08T00:28:56 < zyp> so all circuits are individually protected 2013-02-08T00:29:09 < dongs> that's great 2013-02-08T00:29:30 < zyp> so those are used for everything nowadays 2013-02-08T00:29:47 < dongs> except it powers down your entire house 2013-02-08T00:29:50 < zyp> of course, the main circuit breaker doesn't have one, that would be pointless 2013-02-08T00:30:08 < dongs> that's what japs do. 2013-02-08T00:30:20 < dongs> main breaker is like that 2013-02-08T00:30:38 < zyp> you've said so before 2013-02-08T00:30:51 < emeb> should have a giant GFCI at the neighborhood distribution point. 2013-02-08T00:30:51 < dongs> fuckers 2013-02-08T00:30:54 < zyp> but then japs are overall weird when it comes to power shit 2013-02-08T00:31:02 < zyp> or maybe just overall weird 2013-02-08T00:31:09 < gxti> no earth though 2013-02-08T00:31:13 < emeb> then if your power goes out you can all blame eachother. 2013-02-08T00:31:15 < gxti> so yeah 2013-02-08T00:31:23 < dongs> ok back to bed 2013-02-08T00:31:39 < emeb> yup - nothing interesting going on here. 2013-02-08T00:33:51 < Thorn> btw, why there're no safety systems that would detect current in PE with a current transformer and trip breakers? (where PE is actually correctly installed) 2013-02-08T00:34:31 < zyp> huh? 2013-02-08T00:34:37 < zyp> because that doesn't make sense 2013-02-08T00:36:01 < zyp> a gfci is detecting sum current in the circuit, and trips the breaker if it's significantly more than zero (30mA) and it doesn't matter where it's going 2013-02-08T00:36:34 < gxti> if the current is flowing from the hot wire, then GFCI would detect it 2013-02-08T00:36:41 < zyp> detecting on PE doesn't really make sense, because then you don't know where the current is leaking from, so you don't really know what to trip 2013-02-08T00:36:45 < gxti> if the current is flowing from somewhere else then it wouldn't, but tripping the breaker would not help 2013-02-08T00:36:54 < zyp> you can't just assume that it leaks from the same circuit 2013-02-08T00:37:42 < zyp> what do you mean by hot wire? non-N? 2013-02-08T00:38:07 < gxti> yes, as opposed to "the 11kv line fell on your face" 2013-02-08T00:38:09 < zyp> (we call them L here) 2013-02-08T00:38:31 < zyp> but anyway, it doesn't matter 2013-02-08T00:39:00 < gxti> i'm making the same point as you though, that measuring something other than the line being interrupted would be pointless 2013-02-08T00:39:22 < zyp> even if you short N to PE downstream, it could still trip the GFCI 2013-02-08T00:40:01 < zyp> because even though they are at the same potential, some current will flow in the PE wire instead 2013-02-08T00:40:11 < zyp> kirchhoff and all that 2013-02-08T00:40:25 < zyp> and therefore the sum current in L and N would still be impacted 2013-02-08T00:41:06 -!- jef79m_lurking is now known as jef79m 2013-02-08T00:42:50 < zyp> so, yeah, measuring current in PE is not done because it offers no advantages over measuring sum current in the circuit instead 2013-02-08T00:58:47 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Remote host closed the connection] 2013-02-08T01:01:15 < dongs> so what brought on this discussion anyway. 2013-02-08T01:01:21 < dongs> did someone get electrocuted 2013-02-08T01:06:20 < emeb> zyp talking about jamming metal objects into electric outlets as a child. 2013-02-08T01:07:21 < zyp> yeah, when I was like 2 2013-02-08T01:08:24 < zyp> still think I turned out okay, so A++ would recommend 2013-02-08T01:08:52 < emeb> yes - childhood electrocutions as a means of introduction to an engineering career. 2013-02-08T01:09:49 < zyp> I remember encountering an uncovered outlet at like 7 or so 2013-02-08T01:10:07 < zyp> «hey, wonder how it would feel to stick my hand in there» 2013-02-08T01:10:11 < zyp> and so I did 2013-02-08T01:10:39 < zyp> quite the scientist I'd say 2013-02-08T01:10:54 -!- jef79m is now known as jef79m_lurking 2013-02-08T01:11:57 -!- jef79m_lurking is now known as jef79m 2013-02-08T01:12:22 < qyx_> lol 2013-02-08T01:12:27 < Thorn> I've been shocked at the university during a lab. nothing stupid on my part, touched a metal switch and the apparatus apparently wasn't grounded 2013-02-08T01:13:03 < Thorn> (and I doubt it's ever been fixed) 2013-02-08T01:13:53 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T01:14:03 < dongs> thats how they keep the EE course student count low 2013-02-08T01:14:28 < emeb> weed 'em out. 2013-02-08T01:14:52 < BrainDamage> really? I tought the batshit crazy profs, elitism, and mysoginy driving womans out did 2013-02-08T01:15:24 < zyp> oh, reminds me of the teacher from the electrics course in high school 2013-02-08T01:16:07 < zyp> he went around with a megger and shocked everyone, claiming that electricians needed to know how it felt 2013-02-08T01:16:37 < dongs> i made a stungun out of a huge capacitor and brought that to school 2013-02-08T01:17:04 < dongs> the end just directly pluged in to AC outlet 2013-02-08T01:17:11 < zyp> ha 2013-02-08T01:17:28 < dongs> then i'd walk around zapping shit and making sparks. 2013-02-08T01:17:34 < dongs> donmt remember if I stuck it in anyone 2013-02-08T01:17:45 < zyp> we did similar bullshit in high school 2013-02-08T01:18:22 < zyp> I remember charging caps and just leaving them around waiting for somebody to randomly pick them up 2013-02-08T01:18:37 < BrainDamage> in high schools people had endurance tests with the bench instruments 2013-02-08T01:18:49 < BrainDamage> around 16kHz it was the most painful 2013-02-08T01:18:53 < BrainDamage> at varying voltages 2013-02-08T01:19:50 < BrainDamage> then there was the good old molten blob of solder on the back of a coin, to add heat capacity and warm it 2013-02-08T01:19:55 < zyp> I also hacked together a welder from a bank of caps and a current-limited supply feeding it 2013-02-08T01:19:59 < BrainDamage> then have it accidentally fall 2013-02-08T01:20:14 < zyp> used it to weld together two coins 2013-02-08T01:22:30 < Thorn> statistics from coursera astronomy course http://pastebin.com/cb7sM5xD 2013-02-08T01:23:25 < zyp> quite the dropout rate? 2013-02-08T01:24:22 < Thorn> nobody else seems to have published statistics so can't compare to other courses. but there're certainly some that are harder 2013-02-08T01:35:52 < dongs> http://i.imgur.com/t6wSOKI.jpg might be nsfw 2013-02-08T01:36:09 < mervaka> https://www.facebook.com/photo.php?v=10151481668943767 2013-02-08T01:36:10 < mervaka> \o/ 2013-02-08T01:36:19 < mervaka> filtering = done 2013-02-08T01:36:57 < mervaka> well, i might look into shelving etc, but i dunno if i can be arsed at this point. 2013-02-08T01:37:39 < dongs> is taht done with stm32f4 2013-02-08T01:37:48 < mervaka> yep 2013-02-08T01:37:56 < mervaka> 4x4 mixer 2013-02-08T01:37:58 < mervaka> :D 2013-02-08T01:38:10 < dongs> i dig hte knobs 2013-02-08T01:38:14 < mervaka> heh 2013-02-08T01:38:19 < dongs> but not the tune 2013-02-08T01:38:24 < mervaka> that's just a behringer BCF2000 2013-02-08T01:38:28 < dongs> or hte itunes 2013-02-08T01:38:42 < dongs> oh 2013-02-08T01:38:43 < dongs> its a midibox? 2013-02-08T01:38:44 < dongs> right. 2013-02-08T01:38:47 < mervaka> yeah 2013-02-08T01:39:03 < mervaka> but i've coded layers in, so the two are intimately linked. 2013-02-08T01:39:26 < mervaka> ie, a row of buttons reveal layers in the mix matrix 2013-02-08T01:41:54 < gxti> knoobs 2013-02-08T01:42:23 < mervaka> no, buttons :) 2013-02-08T01:43:05 < mervaka> ie, push a MIX button on an output channel, and the input faders change to show sends to that output 2013-02-08T01:43:36 < mervaka> push MIX on an input channel, and the output faders change to show sends from that input 2013-02-08T01:43:45 < gxti> knooooooooooobs 2013-02-08T01:43:47 < zyp> the faders are actuated so they can change to values? neat 2013-02-08T01:43:56 < mervaka> yer 2013-02-08T01:44:02 < mervaka> that's standard in digital desks though 2013-02-08T01:44:18 < mervaka> (apart from the Focusrite one, but they're getting no end of stick for that) 2013-02-08T01:44:19 < zyp> not in cheaper stuff :p 2013-02-08T01:44:30 < mervaka> i mean digital live desks 2013-02-08T01:44:31 < mervaka> :P 2013-02-08T01:44:40 < mervaka> which is my background 2013-02-08T01:47:34 -!- Gargantuasauce_ [~Gargantua@tangerine.silverorange.com] has joined ##stm32 2013-02-08T01:50:14 < emeb> cool 2013-02-08T01:50:29 < emeb> needs to show yer hardware a bit tho... 2013-02-08T01:50:41 < zyp> agreed 2013-02-08T01:51:19 < mervaka> yeah, it was a video to shamelessly brag to my mates my l33t dsp skillz 2013-02-08T01:51:21 < mervaka> :p 2013-02-08T01:51:38 < mervaka> i'll do a proper video sometime with a real video camera 2013-02-08T01:51:56 < zyp> can I claim dsp-skills for the distortion present in my usb sound card? :) 2013-02-08T01:52:20 < mervaka> huh? 2013-02-08T01:52:40 < zyp> my usb sound card distorts the sound periodically 2013-02-08T01:52:45 < mervaka> haha 2013-02-08T01:52:59 < mervaka> sounds very much like the problem i'm having with my buffering 2013-02-08T01:54:00 < zyp> it happens because the i2s output is clocked seperately from the usb input, so some times the read and write pointers in the ring buffer are operating in the same area 2013-02-08T01:54:22 < mervaka> yeah, that's a bitch. 2013-02-08T01:54:43 < mervaka> ideally it needs to use usb as a sync reference, i guess. 2013-02-08T01:54:51 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-08T01:55:12 < mervaka> or bang it through a FIFO 2013-02-08T01:55:17 < emeb> mervaka: what filter model did you use? 2013-02-08T01:55:24 < mervaka> what iir model? 2013-02-08T01:55:26 < zyp> I tried to enable adaptive sync on usb to make it follow the i2s rate, but I haven't gotten it to work yet 2013-02-08T01:55:28 < mervaka> direct IIt 2013-02-08T01:55:34 < mervaka> second order 2013-02-08T01:55:55 < emeb> so you're computing the coeffs realtime? 2013-02-08T01:56:09 < mervaka> upon input data, yes 2013-02-08T01:56:21 < zyp> and stm32f4 has no way to lock i2s rate to usb rate 2013-02-08T01:56:21 < emeb> interesting. 2013-02-08T01:56:24 < mervaka> it's actually really easy 2013-02-08T01:56:27 < mervaka> really easy. 2013-02-08T01:56:28 < zyp> without external components 2013-02-08T01:56:52 < mervaka> zyp: not without routing it to a pin, i don't think so :( 2013-02-08T01:57:19 < emeb> I just used one of the 4th-order Moog ladder models. 2013-02-08T01:57:29 < zyp> I can output sof and have it drive a pll to drive master clock, I just need the pll :p 2013-02-08T01:57:30 < mervaka> emeb: generating S coefficients then using BZT method to convert to Z coefficients 2013-02-08T01:57:47 < emeb> nice 2013-02-08T01:58:05 < emeb> the Moog model has direct inputs for corner freq and resonance. Very simple. 2013-02-08T01:58:13 < mervaka> fuck knows whether my S coefficients are right though. but they do the job :) 2013-02-08T01:59:07 < mervaka> BZT's a bit funny. you need to warp the frequency using a tan function 2013-02-08T01:59:17 < emeb> ya - pretty typical 2013-02-08T01:59:17 < zyp> hmm 2013-02-08T01:59:29 < zyp> emeb, do you happen to have an opinion on labview? 2013-02-08T01:59:46 < mervaka> is this a good point to mention i have a single board rio for sale? 2013-02-08T01:59:48 < emeb> zyp: not one that's based on experience. :) 2013-02-08T01:59:48 < mervaka> :p 2013-02-08T02:00:13 < dongs> any of you trolls ever replaced ccfl in a l aptop screen 2013-02-08T02:00:34 < emeb> zyp: but based on what I've heard from others it's great for setting up automated lab testing apparatus. Not so great for simulation. 2013-02-08T02:01:00 < zyp> I just remembered that we used labview in uni dsp course 2013-02-08T02:01:04 < zyp> I didn't like it :p 2013-02-08T02:01:05 < emeb> and since LV seems to be going after Matlab's customers that's kind of sad. 2013-02-08T02:01:23 < zyp> drawing code didn't sit right with me :p 2013-02-08T02:01:29 < emeb> heh. 2013-02-08T02:01:40 < emeb> It's designed with technicians in mind. 2013-02-08T02:01:45 < mervaka> my labview training was shear torture. i was jumping ballistic angry all the way through 2013-02-08T02:01:59 < emeb> Or PhDs who don't want to learn text-based coding... :) 2013-02-08T02:02:08 < mervaka> matlab on the other hand, is just pure win. 2013-02-08T02:02:41 < emeb> Gotta admit I'm biased towards Matlab. 2013-02-08T02:02:50 < emeb> but then I've been using it for... ages. 2013-02-08T02:02:54 < BrainDamage> I hate both methlab and failview 2013-02-08T02:03:14 < mervaka> what's wrong with matlab? 2013-02-08T02:03:15 < emeb> methlab... lol 2013-02-08T02:03:28 < mervaka> although i wanna get into octave now 2013-02-08T02:03:31 < BrainDamage> you have to memorize the builtin functions 2013-02-08T02:03:36 < mervaka> you do? 2013-02-08T02:03:38 < BrainDamage> and the awkward syntax 2013-02-08T02:03:42 < mervaka> i just wrote my own 2013-02-08T02:03:48 <+Steffanx> mervaka dont forget about his nicknae :D 2013-02-08T02:03:49 < BrainDamage> that's the wrong way to use it 2013-02-08T02:03:56 < BrainDamage> the interpreter is super slow 2013-02-08T02:04:04 < mervaka> only hinderance i have is that arrays start at 1 instead of 0 2013-02-08T02:04:20 < BrainDamage> you're supposed to use it ony to string together builtin functions 2013-02-08T02:04:20 < zyp> I've actually never used matlab 2013-02-08T02:04:27 < emeb> a lot of people complain about the 1-base indices. 2013-02-08T02:04:35 < BrainDamage> lua is the same, not a huge deal 2013-02-08T02:04:35 < emeb> just gotta get used to it. 2013-02-08T02:04:39 <+Steffanx> Are they arrays or matrices? 2013-02-08T02:04:43 <+Steffanx> or vectors or wahtever 2013-02-08T02:04:48 < mervaka> emeb: yep. not a massive hinderance 2013-02-08T02:04:58 < zyp> but I understand that the pylab-stuff is emulating matlab functionality in python 2013-02-08T02:04:58 < emeb> Steffanx: the base type is a matrix 2013-02-08T02:04:59 < mervaka> Steffanx: both, i guess 2013-02-08T02:05:08 < zyp> and I'm quite happy with that 2013-02-08T02:05:11 < mervaka> you can treat them as either depending on what operator you use 2013-02-08T02:05:17 < emeb> but matrices can have only one dimension if you want. 2013-02-08T02:05:22 < mervaka> prefix your operator with . and it does array functions 2013-02-08T02:05:43 < mervaka> so for example A*B is a matrix multiplication 2013-02-08T02:05:50 < mervaka> A.*B is an array multiplication 2013-02-08T02:05:55 <+Steffanx> isn't that the reason why it starts at 1? 2013-02-08T02:06:00 < mervaka> yep 2013-02-08T02:06:03 < emeb> mervaka: huh? 2013-02-08T02:06:10 <+Steffanx> So don't complain, it's a feature :P 2013-02-08T02:06:11 < mervaka> emeb: ? 2013-02-08T02:06:16 < emeb> you mean vector not array, right? 2013-02-08T02:06:28 < emeb> since arrays can have multiple dimensions... 2013-02-08T02:06:34 <+Steffanx> Ha 2013-02-08T02:06:44 < mervaka> vector is a 1xn or mx1 matrix 2013-02-08T02:06:51 < emeb> right 2013-02-08T02:07:00 < mervaka> arrays are treated as arrays. 2013-02-08T02:07:02 < dongs> the biggest question is.. how did you get STM32 to send/receive midi??????? i thought this can only be don wiht arduino 2013-02-08T02:07:06 < emeb> A.*B is vector mult. 2013-02-08T02:07:10 < dongs> with midi shield 2013-02-08T02:07:13 < mervaka> dongs: usart 2013-02-08T02:07:22 < mervaka> easy 2013-02-08T02:07:41 < BrainDamage> you pwm out the midi squarewave signal! 2013-02-08T02:07:45 < mervaka> nope 2013-02-08T02:08:01 < zyp> I thought that on stm32 you have to implement usb host mode and use a usb-midi interface 2013-02-08T02:08:34 < dongs> haha 2013-02-08T02:08:48 < mervaka> midi is just USART at 31250 baud and an optocoupled rx 2013-02-08T02:08:59 < mervaka> and OD input 2013-02-08T02:09:08 < mervaka> tx* 2013-02-08T02:09:27 < dongs> for some r eason tehy sent me 2 tubes 2013-02-08T02:09:36 < dongs> i wonder if its because replacing these is so fucking annoying 2013-02-08T02:09:49 < zyp> they knew you are going to break one trying to get it in place 2013-02-08T02:09:58 < zyp> well 2013-02-08T02:10:16 < zyp> I assume they degrade over time, so you replace both to ensure uniform lighting 2013-02-08T02:10:29 < dongs> theres only one 2013-02-08T02:10:34 < zyp> really? 2013-02-08T02:10:45 < dongs> ya 2013-02-08T02:10:50 < dongs> lg panel 2013-02-08T02:11:02 < zyp> I once had a laptop with a half-dark screen because one of the tubes was broken 2013-02-08T02:11:41 < emeb> half bright or half dark? psych test... 2013-02-08T02:12:07 < zyp> half dark, considering it was not supposed to be dark at all. 2013-02-08T02:12:51 < zyp> it would be half bright if it was supposed to turn off but one tube still remained on :p 2013-02-08T02:13:19 < dongs> k resoldered. 2013-02-08T02:13:19 < dongs> hmm 2013-02-08T02:13:25 < dongs> i hope the lcd stil works after i put it back together 2013-02-08T02:15:34 < mervaka> nn guys 2013-02-08T02:18:47 < dongs> oops :( 2013-02-08T02:18:51 < dongs> there's a ring around center of the tube 2013-02-08T02:18:58 < dongs> probably to stop it from banging around the case. 2013-02-08T02:19:04 < dongs> owelz 2013-02-08T02:24:34 -!- Laurenceb_ [~Laurence@host86-143-184-68.range86-143.btcentralplus.com] has quit [Read error: Operation timed out] 2013-02-08T02:29:05 < emeb> FPGA/RF input board: http://i.imgur.com/eOhJljY.png 2013-02-08T02:29:46 < zyp> looks like the old board you were showing off recently 2013-02-08T02:30:04 < emeb> Layout is very similar to the DAC board I did a while back. 2013-02-08T02:31:28 < emeb> has a little differential driver amp instead of transformer/filter arrangment that the DAC had. 2013-02-08T02:31:43 < emeb> plus optional on-board sampling freq osc. 2013-02-08T02:34:28 < gxti> om nom nom 2013-02-08T02:34:54 < gxti> going to put together a little carrier board for ublox, nothing so exciting though 2013-02-08T02:35:28 < gxti> either that or rework the whole ntp server board to include it directly, still can't make up my mind 2013-02-08T02:36:01 < emeb> a design is never done - eventually you just stop. 2013-02-08T02:36:51 < gxti> i suppose putting my mcu directly under the switching psu module is probably a bad idea 2013-02-08T02:37:13 < emeb> you never know - the em field might be good for it. 2013-02-08T02:37:41 < gxti> that and ditching the 5v rail (and the 5v -> 3.3v switcher) would free up room for the gps module without expanding much 2013-02-08T02:38:09 < emeb> so what - bring 3.3V in directly? 2013-02-08T02:38:28 < gxti> right now i have a murata module that does 7-30->5, then a sc189 that does 5->3.3 but it kinda sucks 2013-02-08T02:38:34 < gxti> the muratas also come in 3.3 2013-02-08T02:38:43 < emeb> ooo - simpler 2013-02-08T02:38:54 < gxti> and i only did 5 to support receivers that needed it, although the active antenna might work better with 5 than 3.3 2013-02-08T02:44:55 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-08T02:45:11 < gxti> bonus points for not having the sc189 anymore as it can't recover from overcurrent without power cycling, even a very brief one 2013-02-08T02:45:28 < gxti> like plugging it into a gps receiver after it was running would put it into overcurrent mode 2013-02-08T02:45:45 < gxti> super lame :| 2013-02-08T02:46:44 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Ping timeout: 248 seconds] 2013-02-08T02:52:56 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-08T02:53:32 < dongs> well, cocks 2013-02-08T02:53:35 < dongs> the screen is not any brighter 2013-02-08T02:53:40 < dongs> and i managed to ruin a column of drivers 2013-02-08T02:53:46 < dongs> so i ahve a shitty vertical line 2013-02-08T02:53:50 < dongs> time to just get a replacement screen. 2013-02-08T02:53:54 < emeb> I'm back - did you miss me? (blew a breaker) 2013-02-08T02:53:54 < dongs> its a uxga15" from 2002 2013-02-08T02:54:19 < emeb> that's an old one. 2013-02-08T02:56:05 < dongs> TOSHIBA V000021130 LAPTOP LCD SCREEN 15" UXGA MATTE CCFL $64.99 2013-02-08T02:56:05 < dongs> lol. 2013-02-08T02:56:05 < dongs> yeah, they dont make that kinda awesome shit anymore. 2013-02-08T02:56:05 < dongs> good luck finding a 15" matte screen. 2013-02-08T02:56:05 < dongs> wonder if my chinks have htem 2013-02-08T02:56:11 < dongs> Compatibility:101% OEM Compatible 2013-02-08T02:56:11 < dongs> oh my 2013-02-08T02:56:21 < emeb> they go the extra 1% 2013-02-08T02:56:25 < dongs> ya. 2013-02-08T02:57:46 < gxti> but not 110%, let's be realistic here 2013-02-08T03:03:00 < dongs> ordered. 2013-02-08T03:03:04 < dongs> lets see if yanks ship it today. 2013-02-08T03:04:15 < gxti> 5pm west coast, unless it's digibutt good luck 2013-02-08T03:04:44 < dongs> these fags are in NJ 2013-02-08T03:04:47 < dongs> yea prob not 2013-02-08T03:05:24 < gxti> bought some smartcards 1 week ago and haven't heard anything other than an order confirm, let's see what happens. US, too. 2013-02-08T03:09:09 <+Steffanx> poor gxti 2013-02-08T03:10:33 < dongs> scam'd 2013-02-08T03:10:37 < gxti> not concerned yet, just poor practice. worst case i chargeback and lose nothing. 2013-02-08T03:10:41 < dongs> "buying" smartcards is scam anyway 2013-02-08T03:10:44 <+Steffanx> Yes, you lost time 2013-02-08T03:10:46 < dongs> cause you shouldnt be buying htem 2013-02-08T03:10:54 < gxti> o rly 2013-02-08T03:11:01 < dongs> yarly 2013-02-08T03:11:05 < gxti> why is that 2013-02-08T03:11:09 <+Steffanx> because 2013-02-08T03:11:13 < gxti> ok 2013-02-08T03:11:17 < gxti> lesson learned then 2013-02-08T03:11:19 <+Steffanx> because dongs says so 2013-02-08T03:11:22 < gxti> don't, because 2013-02-08T03:14:25 <+Steffanx> ok 2013-02-08T03:20:06 < dongs> http://bcas.tv/paste/results/LvW4F179.html xilinx customer support is hilarious 2013-02-08T03:20:33 < gxti> ha ha ha you're fired 2013-02-08T03:23:46 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Ping timeout: 245 seconds] 2013-02-08T03:28:11 < upgrdman> lol 2013-02-08T03:28:31 < upgrdman> i work in customer service, sometimes i REALLY want to give an honest answer like that. 2013-02-08T03:29:51 < upgrdman> dongs, have the original forum link? 2013-02-08T03:31:51 < Thorn> isn't digilent an official maker of dev kits for xilinx 2013-02-08T03:33:15 < upgrdman> yes 2013-02-08T03:33:28 < upgrdman> i have a digilent nexys2 dev board 2013-02-08T03:33:50 < dongs> upgrdman: http://forums.xilinx.com/t5/3rd-Party-Other-Boards-and-Kits/Chain-initialization-error-with-my-Nexys-3-Spartan-6-Dev-kit/td-p/265428 2013-02-08T03:33:54 < dongs> just googled 1st sentence 2013-02-08T03:34:20 < dongs> i didnt post it, just something my pal linekd 2013-02-08T03:34:21 < dongs> linked 2013-02-08T03:35:20 < dongs> im pretty sure i have official xilinx 3E evalkit and it s made by digilent though. 2013-02-08T03:35:28 < dongs> so that guy is just a fucking asshole :) 2013-02-08T03:35:37 < upgrdman> glad to see the "customer" didnt get upset about the rep's comment 2013-02-08T03:36:30 < gxti> hahaha and it's in a 3rd party board too 2013-02-08T03:36:40 < gxti> phailure 2013-02-08T03:38:51 < dongs> i dunnowhy hes running adpet though 2013-02-08T03:38:57 < dongs> you can just use xilinx programmer directly 2013-02-08T03:39:07 < dongs> i mean you can flash it directly from the xilinx shitty IDE anyway 2013-02-08T03:39:13 < dongs> chain programming/select chip etc. 2013-02-08T03:39:25 < dongs> i n ever installed any digilent-specific junks 2013-02-08T03:39:26 < dongs> and it worked 2013-02-08T03:40:17 < ds2> is there a easy way to trap NULL ptr access on the stm32? 2013-02-08T03:41:37 < emeb> Austin has been around for ages and he's always acted like that. 2013-02-08T03:42:22 < upgrdman> dongs, the ref man for my nexys2 states that i must use Adept if i want to program through the usb port... or i can program with xilinx' tools if i use an external jtag programmer. are you saying the xilinx tools will work over the usb port on the dev board? 2013-02-08T03:42:38 < emeb> sometimes he'll come back a few days later and modulate the response somewhat. 2013-02-08T03:44:50 < dongs> oh hey, got refund for my 2nd batch of goclone socks 2013-02-08T03:45:12 < emeb> Free Socks! 2013-02-08T03:45:27 < dongs> yea, spawn already stole them too 2013-02-08T03:45:35 < dongs> i tossed the shit in a corner of my office and its gone now 2013-02-08T03:46:08 < emeb> and did you get a "Thanks Dad"? No, probably not. 2013-02-08T03:47:28 < dongs> upgrdman: oh right. using platform cable here 2013-02-08T03:47:31 < dongs> some clone probly 2013-02-08T03:47:45 < upgrdman> whats a platform cable? 2013-02-08T03:47:55 < emeb> Xilinx downloader 2013-02-08T03:47:58 < dongs> http://www.ebay.com/itm/300776609831 2013-02-08T03:47:58 < dongs> yeah 2013-02-08T03:48:52 < emeb> http://latimesblogs.latimes.com/lanow/2013/02/ex-cop-manhunt-newspaper-delivery-women-shot.html 2013-02-08T03:49:10 < emeb> Pro Tip - if you're driving a car like the LAPD is looking for, stay home. 2013-02-08T03:51:34 < dongs> "It was not immediately known what newspapers the women were delivering." 2013-02-08T03:52:58 < dongs> fuck 2013-02-08T03:53:03 < dongs> these morons used freertos 2013-02-08T03:53:09 < dongs> but only committed GCC portability files 2013-02-08T03:55:47 < Thorn> that's what I do too 2013-02-08T03:56:06 < Thorn> my code won't compile in anything but gcc anyway 2013-02-08T03:56:27 < baird> rofl: "Jemima Khan turns her back on Julian Assange, saying he risks becoming Australia's version of L Ron Hubbard." 2013-02-08T03:56:58 < emeb> Hail Xenu! 2013-02-08T04:00:08 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-08T04:04:23 < dongs> http://www.telegraph.co.uk/news/worldnews/europe/russia/9854619/Document-proclaiming-death-of-Soviet-Union-missing.html 2013-02-08T04:04:27 < dongs> haha 2013-02-08T04:04:32 < dongs> cue in soviet russia jokes 2013-02-08T04:05:22 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 244 seconds] 2013-02-08T04:05:38 < dongs> curious. 2013-02-08T04:05:48 < dongs> feeRTOS doesn't have a FPU port for CM3 2013-02-08T04:06:08 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-08T04:06:22 < Thorn> it has a port for cm4f 2013-02-08T04:06:50 < dongs> where? 2013-02-08T04:06:56 < dongs> i dont see it in rvds dir 2013-02-08T04:09:40 < Thorn> RVDS\ARM_CM4F? (7.3.0) 2013-02-08T04:09:45 < dongs> ah 2013-02-08T04:09:48 < dongs> im still using 7.0.2 2013-02-08T04:09:52 < dongs> or rather these trolls are 2013-02-08T04:09:55 < Thorn> old 2013-02-08T04:09:58 < dongs> ya 2013-02-08T04:09:59 < Thorn> really 2013-02-08T04:11:32 < gxti> bleh, freertos is modified gpl too? 2013-02-08T04:11:52 < Thorn> yes 2013-02-08T04:11:59 < dongs> freertos sucks anyway 2013-02-08T04:12:04 < gxti> k 2013-02-08T04:12:05 < dongs> im just using it because this project does 2013-02-08T04:12:10 < dongs> coos > freertos 2013-02-08T04:12:30 < ds2> uCLinux!!! 2013-02-08T04:12:31 < ds2> ;) 2013-02-08T04:12:36 < gxti> no uclinux 2013-02-08T04:12:38 < gxti> ass 2013-02-08T04:12:43 < Thorn> ucWindows 2013-02-08T04:14:59 < gxti> coos is bsd, i will have to check it out 2013-02-08T04:15:21 < gxti> it's not that i particularly like chibios it's just the first thing that looked not shit 2013-02-08T04:15:41 < gxti> but GNU's Not Useful 2013-02-08T04:15:47 < emeb_mac> lol 2013-02-08T04:15:57 < gxti> dumping that shit for the first viable alternative 2013-02-08T04:16:34 < gxti> you wanna write a compiler and GPL it whatever, i don't want to interact with it outside the command line anyway 2013-02-08T04:16:44 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Ping timeout: 244 seconds] 2013-02-08T04:18:00 < gxti> but libraries shouldn't taint your code especially ones that you're compiling directly into your app :| 2013-02-08T04:19:51 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-08T04:21:55 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:b98e:a8e7:7827:312e] has joined ##stm32 2013-02-08T04:23:00 < baird> Damn, probably best to go back to the proprietary libraries, then. Goddamn people setting licensing conditions on the code they've written. 2013-02-08T04:24:26 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-08T04:25:20 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-08T04:25:22 < gxti> they're welcome do it and i'm welcome to not use it 2013-02-08T04:26:11 < gxti> luckily there's a whole class of licenses that don't fit into your false dichotomy 2013-02-08T04:29:55 < baird> And they have a large contributing following behind them. 2013-02-08T04:30:08 < gxti> wtf, strtok pulls in malloc? 2013-02-08T04:36:17 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:b98e:a8e7:7827:312e] has quit [Remote host closed the connection] 2013-02-08T04:46:14 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Read error: Operation timed out] 2013-02-08T04:54:04 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-08T05:13:53 -!- Netsplit *.net <-> *.split quits: TheSeven, zyp, mervaka, claude, enots, R2COM, ReadError, RuslanPopov, CoolBear, DLPeterson, (+55 more, use /NETSPLIT to show all of them) 2013-02-08T05:21:23 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-08T05:22:35 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-08T06:27:01 -!- mansfeld [~andrew@robopoly/watson] has joined ##stm32 2013-02-08T06:27:01 -!- neuro_sys|off [~neuro_sys@unaffiliated/neurosys/x-283974] has joined ##stm32 2013-02-08T06:27:01 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-08T06:27:01 -!- ReadError [readerror@ec2-50-19-189-163.compute-1.amazonaws.com] has joined ##stm32 2013-02-08T06:27:01 -!- zyp [zyp@zyp.im] has joined ##stm32 2013-02-08T06:27:01 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has joined ##stm32 2013-02-08T06:27:01 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-08T06:27:01 -!- zippe [~Adium@173.11.99.161] has joined ##stm32 2013-02-08T06:27:01 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-08T06:27:01 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-08T06:27:01 -!- claude [sbnc@h1682708.stratoserver.net] has joined ##stm32 2013-02-08T06:27:01 -!- ohama [~ohama@92.240.248.100] has joined ##stm32 2013-02-08T06:27:01 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32 2013-02-08T06:27:01 -!- ABLomas [~abl@78-60-198-200.static.zebra.lt] has joined ##stm32 2013-02-08T06:27:01 -!- fergusnoble [fergusnobl@repl.esden.net] has joined ##stm32 2013-02-08T06:27:01 -!- gxti [~gxti@ada.partiallystapled.com] has joined ##stm32 2013-02-08T06:27:01 -!- badger [~tom@cxkg.net] has joined ##stm32 2013-02-08T06:27:01 -!- CheBuzz [~CheBuzz@76.164.192.124] has joined ##stm32 2013-02-08T06:27:01 -!- Simon-- [~sim@2606:6a00:0:28:5604:a6ff:fe02:702b] has joined ##stm32 2013-02-08T06:27:01 -!- evil_dan2wik [~evil_dan2@unaffiliated/evil-dan2wik/x-0106201] has joined ##stm32 2013-02-08T06:27:01 -!- R0b0t1_ [~dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32 2013-02-08T06:27:01 -!- tkoskine [tkoskine@kapsi.fi] has joined ##stm32 2013-02-08T06:27:01 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32 2013-02-08T06:27:01 -!- RuslanPopov [~rad@ppp-188-65-14-36-bras2.istra.ru] has joined ##stm32 2013-02-08T06:27:01 -!- BusError [~michel@host81-152-248-48.range81-152.btcentralplus.com] has joined ##stm32 2013-02-08T06:27:01 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32 2013-02-08T06:27:01 -!- ChanServ [ChanServ@services.] has joined ##stm32 2013-02-08T06:27:01 -!- donigs [~no@bcas.tv] has joined ##stm32 2013-02-08T06:27:01 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32 2013-02-08T06:27:01 -!- Posterdati [~antani@host107-26-dynamic.59-82-r.retail.telecomitalia.it] has joined ##stm32 2013-02-08T06:27:01 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-08T06:27:01 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-08T06:27:01 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-08T06:27:01 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-08T06:27:01 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-08T06:27:01 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has joined ##stm32 2013-02-08T06:27:01 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-08T06:27:01 -!- esden [esden@repl.esden.net] has joined ##stm32 2013-02-08T06:27:01 -!- jef79m [~jef79m@124-168-171-28.dyn.iinet.net.au] has joined ##stm32 2013-02-08T06:27:01 -!- enots [dimka@freelsd.net] has joined ##stm32 2013-02-08T06:27:01 -!- qyx_ [~qyx@krtko.org] has joined ##stm32 2013-02-08T06:27:01 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32 2013-02-08T06:27:01 -!- ServerMode/##stm32 [+o ChanServ] by holmes.freenode.net 2013-02-08T06:27:25 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-08T06:27:25 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32 2013-02-08T06:27:25 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-08T06:27:32 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-08T06:27:32 -!- Rious [~Rious@ec2-50-17-240-121.compute-1.amazonaws.com] has joined ##stm32 2013-02-08T06:27:32 -!- CoolBear [~hightower@ti0069a380-0800.bb.online.no] has joined ##stm32 2013-02-08T06:27:38 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-08T06:27:38 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-08T06:27:38 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-08T06:27:38 -!- drgreenthumb [~fletch@drupal.org/user/72475/view] has joined ##stm32 2013-02-08T06:27:38 -!- mervaka [~mervaka@mervaka.co.uk] has joined ##stm32 2013-02-08T06:27:38 -!- Nutter [~Nutter@199-195-151-246.dyn.novuscom.net] has joined ##stm32 2013-02-08T06:27:44 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32 2013-02-08T06:27:44 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-08T06:27:44 -!- karlp [~karl@palmtree.beeroclock.net] has joined ##stm32 2013-02-08T06:27:48 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-08T06:27:48 -!- cdh [~chatzilla@c-50-136-234-52.hsd1.ca.comcast.net] has joined ##stm32 2013-02-08T06:27:48 -!- olasd [~olasd@pdpc/supporter/active/olasd] has joined ##stm32 2013-02-08T06:27:52 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32 2013-02-08T06:27:52 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-08T06:27:52 -!- jv-- [jv@pilsedu.cz] has joined ##stm32 2013-02-08T06:27:52 -!- Tectu [tectu@irc.ipv6.kunsmann.eu] has joined ##stm32 2013-02-08T06:29:14 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 252 seconds] 2013-02-08T06:29:51 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving] 2013-02-08T06:31:34 -!- zlog [~zlog@ip68-102-192-159.ks.ok.cox.net] has quit [Ping timeout: 244 seconds] 2013-02-08T06:34:02 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving] 2013-02-08T06:37:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-08T06:40:49 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-08T06:42:14 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-08T06:42:25 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-08T06:46:39 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has joined ##stm32 2013-02-08T06:50:20 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-08T06:53:02 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-08T06:56:14 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-08T06:59:06 < emeb_mac> banana split 2013-02-08T06:59:11 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Client Quit] 2013-02-08T06:59:28 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-08T06:59:38 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-08T07:14:58 < dongs> again? 2013-02-08T07:15:52 < emeb_mac> looks like there were 2, about 1.25 hrs apart 2013-02-08T07:17:27 < dongs> feenode cant properly do shit even with all these fucking lilo donations 2013-02-08T07:17:34 < dongs> #ifndef __cplusplus 2013-02-08T07:17:34 < dongs> typedef enum {FALSE = 0, TRUE = !FALSE} bool; 2013-02-08T07:17:35 < dongs> #endif 2013-02-08T07:17:35 < dongs> .. 2013-02-08T07:17:37 < dongs> what hte fuck is this shit 2013-02-08T07:18:02 < dongs> i have a feeling opensoresfags messed with this 2013-02-08T07:21:36 < dongs> haha more 0b1123123 crap 2013-02-08T07:23:30 < dongs> http://bcas.tv/paste/results/OedEJg69.html is dis some gnu trash 2013-02-08T07:23:31 -!- jef79m is now known as jef79m_lurking 2013-02-08T07:23:37 < gxti> lol wut 2013-02-08T07:23:56 < emeb_mac> yoiks 2013-02-08T07:23:58 < ds2> wtf 2013-02-08T07:23:59 < emeb_mac> that's a mess 2013-02-08T07:26:41 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-08T07:26:55 < dongs> seems gnu because enabling gnu hacks for armcc let it build 2013-02-08T07:28:23 < emeb_mac> took me a while to even figure out what that was doing. 2013-02-08T07:30:44 < dongs> the fuckers who write shit like void foo(); should fucking die 2013-02-08T07:31:10 < ds2> what's wrong with that? 2013-02-08T07:31:14 < dongs> EVERYTHINGF 2013-02-08T07:31:22 < ds2> isn't that a standard function prototype? 2013-02-08T07:31:23 < dongs> ITS BEEN DEPRECATED SINCE LIKE 1985 2013-02-08T07:31:25 < dongs> ... 2013-02-08T07:31:28 < dongs> void foo(void); is 2013-02-08T07:31:50 < ds2> Ohhhhhh that detail 2013-02-08T07:32:11 < dongs> whats worse is these idiots arent even consistent 2013-02-08T07:32:11 < dongs> void systemStart(); 2013-02-08T07:32:11 < dongs> void systemWaitStart(void); 2013-02-08T07:32:13 < dongs> ???????????? 2013-02-08T07:32:15 < ds2> shouldn't it be void foo(int)? thought K&R assumes INT? 2013-02-08T07:32:18 < gxti> formidable nerd rage 2013-02-08T07:33:13 < dongs> lol @ return ENETDOWN 2013-02-08T07:33:17 < dongs> some bullshit lunix error code 2013-02-08T07:33:26 < dongs> no idea what header would taht trash even be in 2013-02-08T07:34:04 < ds2> ENETDOWN is not Linux 2013-02-08T07:34:07 < gxti> spoilers: errno.h 2013-02-08T07:34:10 < dongs> yes it fucking is 2013-02-08T07:34:23 < gxti> not that it really makes sense for rtos 2013-02-08T07:34:27 < gxti> because yeah 2013-02-08T07:40:53 < dongs> ugh 2013-02-08T07:41:09 < dongs> getting opensores lunix trash to compile with anything serious is always a fucking chore 2013-02-08T07:41:36 < gxti> it must be hard on you dongs 2013-02-08T07:42:47 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-08T07:48:00 < dongs> modules/src/info.c: crtpInitTaskQueue(crtpInfo); 2013-02-08T07:48:00 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-08T07:48:00 < dongs> modules/src/info.c: if (crtpReceivePacketWait(crtpInfo, &p, 1000) == pdTRUE) 2013-02-08T07:48:01 -!- R2COM1 [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-08T07:48:08 < dongs> the ONLY reference to crtpInfo in entire source tree. 2013-02-08T07:48:09 < dongs> awesome! 2013-02-08T07:48:23 < dongs> i bet they #defined it in the MAKEFILE or something 2013-02-08T07:49:02 < dongs> nope. 2013-02-08T07:49:03 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out] 2013-02-08T07:51:05 < ds2> maybe it is dynamically generated 2013-02-08T07:56:00 < dongs> by what 2013-02-08T07:57:16 < dongs> haha, more gcc fail 2013-02-08T07:57:37 < dongs> same named global vars in different .c files = t otally OK for gcc 2013-02-08T07:57:42 < dongs> let's just merge them in same mermoy space anyways! 2013-02-08T07:57:55 < dongs> (and opensores tools who write shit code liek that never notice it 2013-02-08T07:57:59 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving] 2013-02-08T07:58:07 < dongs> oh, they probably meant to use same var anyway. 2013-02-08T07:58:36 < gxti> main.c:68: multiple definition of `mac_addr' 2013-02-08T07:58:39 < gxti> ^ works for me 2013-02-08T07:59:25 < dongs> nope 2013-02-08T07:59:30 < gxti> yep 2013-02-08T07:59:50 < gxti> you can deliberately declare a symbol as mergeable but i forget how and i don't know how you could do it accidentally 2013-02-08T07:59:59 < ds2> uh... that is standard 2013-02-08T08:00:05 < ds2> even microsoft C did that 2013-02-08T08:00:09 < dongs> __IO uint32_t I2CDirection = I2C_DIRECTION_TX; in i2c.c and __IO uint32_t I2CDirection; < no error on gcc. 2013-02-08T08:00:18 < dongs> err, 2nd one in i2ccrap.c 2013-02-08T08:00:48 < ds2> seems fine 2013-02-08T08:00:53 < dongs> ? 2013-02-08T08:01:02 < ds2> if you put an initializer in the second one, it would be either undefined or error out 2013-02-08T08:01:27 < dongs> well this code builds in gcc 2013-02-08T08:04:07 < ds2> that stuff should work with MSC 2013-02-08T08:14:07 < dongs> ? 2013-02-08T08:14:14 < dongs> they're defined twice. 2013-02-08T08:14:17 < dongs> no ifs/buts about it. 2013-02-08T08:14:26 < dongs> and gcc doesnt complain becuase it just merges both in same memory 2013-02-08T08:14:31 < dongs> whether thats intentional or not 2013-02-08T08:14:43 < dongs> same name, same type? who cares, i'll put it in same place 2013-02-08T08:22:35 -!- zippe1 [~Adium@173.11.99.161] has joined ##stm32 2013-02-08T08:23:44 -!- zippe [~Adium@173.11.99.161] has quit [Ping timeout: 256 seconds] 2013-02-08T08:27:59 < ds2> and I am saying that is fine 2013-02-08T08:28:14 < ds2> unless you can quote C spec that says otherwise 2013-02-08T08:28:38 < dongs> .......... 2013-02-08T08:29:57 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-08T08:36:02 < dongs> http://bytes.com/topic/c/answers/944086-gcc-compiler-does-not-detect-multiple-definition-error 2013-02-08T08:36:35 < dongs> hm, no thats different but equally fucktarded. 2013-02-08T08:36:43 < dongs> cause/efect is same 2013-02-08T08:38:41 < gxti> my system toolchain seems to merge, while my embedded one errors 2013-02-08T08:38:52 < gxti> go figure 2013-02-08T08:39:17 < dongs> thats opensores for you. 2013-02-08T08:39:21 < dongs> unpredictable garbage 2013-02-08T08:40:08 < gxti> mmmhmm 2013-02-08T08:40:49 < dongs> more awesome would be if the code in both .c files actually expected the vars to be merged 2013-02-08T08:40:54 < dongs> and caused subtle bugs if they werent. 2013-02-08T08:40:58 < dongs> that'd be a fun one to track dowen 2013-02-08T08:41:00 < dongs> down 2013-02-08T08:41:13 < dongs> ididnt look at the code but I'd imagine i2cdirection is kinda important 2013-02-08T08:41:17 < gxti> if they're not merged then you get an error 2013-02-08T08:41:27 < gxti> it's not like you would end up with two separate vars 2013-02-08T08:41:50 < dongs> you think? 2013-02-08T08:41:52 < dongs> never know with opensauce 2013-02-08T08:41:58 < gxti> . . . 2013-02-08T08:42:01 < gxti> right. 2013-02-08T08:42:30 < dongs> temp++; //To avoid GCC warning! 2013-02-08T08:42:32 < dongs> yaaaaaa 2013-02-08T08:42:41 < dongs> (after return SUCCESS; 2013-02-08T08:42:44 < gxti> i feel smarter already, bedtime 2013-02-08T08:42:52 < dongs> which makes a whole NEW warning, statement is unreachabe 2013-02-08T08:43:06 < gxti> that's called "idiot programmer" 2013-02-08T08:43:51 < emeb_mac> that's 99% of the problem right there. 2013-02-08T08:44:56 < dongs> incidentally, the "free" availability of opensores usually leads to these retards ganging up together on projects. 2013-02-08T08:44:59 < dongs> and writing shit code. 2013-02-08T08:45:28 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-08T08:45:45 < ds2> that's not an opensource problem 2013-02-08T08:45:53 < ds2> that is a retard thinking all warnings should go away 2013-02-08T08:45:53 < emeb_mac> yeah. If everyone had to pay for tools to write programs there would be fewer amateur programmers hacking away out there. :P 2013-02-08T08:46:06 < ds2> a warning is a warning because it maybe fine. 2013-02-08T08:46:33 < dongs> https://www.google.com/finance/chart?q=NASDAQ:ATML&tlf=12&chst=vks&chs=327x96&chsc=1&p=5d 2013-02-08T08:47:21 < emeb_mac> what happened to atmel? 2013-02-08T08:47:26 < dongs> http://www.atmel.com/microsite/sama5d3/default.aspx BGA only? get the fuck out 2013-02-08T08:47:30 < dongs> where's the DIP version 2013-02-08T08:49:11 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving] 2013-02-08T08:49:50 < emeb_mac> never succeed w/o 5V I/O for arduino compatibility. 2013-02-08T08:52:57 < emeb_mac> 0.8mm ball pitch. 2013-02-08T08:56:02 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds] 2013-02-08T08:56:10 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-08T08:56:36 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 256 seconds] 2013-02-08T08:57:27 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-08T09:13:59 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-08T09:20:56 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Ping timeout: 240 seconds] 2013-02-08T09:25:39 -!- evil_dan2wik [~evil_dan2@unaffiliated/evil-dan2wik/x-0106201] has quit [Ping timeout: 252 seconds] 2013-02-08T09:27:28 -!- R2COM1 is now known as R2COM 2013-02-08T09:40:12 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-08T10:00:22 -!- jef79m_lurking is now known as jef79m 2013-02-08T10:04:26 < zippe1> dongs: apparently, you don't understand C and what 'common' symbols are 2013-02-08T10:04:35 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-08T10:07:25 < dongs> zippe1: enlighten me 2013-02-08T10:08:27 < dongs> looks like maybe some C99 'tentative definition' crap but ii doubt it 2013-02-08T10:08:28 < zippe1> aka 'tentative definitions' 2013-02-08T10:08:36 < dongs> to me just sounds like dumb coder. 2013-02-08T10:09:04 < zippe1> The link you pasted is to someone that doesn't get it 2013-02-08T10:09:14 < zippe1> But it's a legitimate (if stupid) feature of the language 2013-02-08T10:09:28 -!- neuro_sys|off is now known as neuro_sys 2013-02-08T10:09:33 < dongs> yea well. 2013-02-08T10:09:41 < dongs> gcc not warnign about it = fail 2013-02-08T10:09:51 < zippe1> Warning about what? 2013-02-08T10:09:57 < zippe1> There is nothing to warn about 2013-02-08T10:10:10 < zippe1> The language explicitly allows it 2013-02-08T10:10:35 < zippe1> There are optional warnings you can enable (-fwarn-commons, -fno-commons), but if you want to be c99, you are supposed to support them 2013-02-08T10:12:35 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-08T10:17:16 -!- jef79m is now known as jef79m_lurking 2013-02-08T10:18:55 < ds2> why do you even think that is wrong? 2013-02-08T10:19:05 < dongs> huh 2013-02-08T10:19:07 < ds2> it is global 2013-02-08T10:19:16 < dongs> yes, variables are defined in ONE PLACE 2013-02-08T10:19:18 < ds2> a variable is a name for an address 2013-02-08T10:19:27 < dongs> any other file, if you are actually intending to be using same variable, it must be extern. 2013-02-08T10:19:30 < dongs> no ifs/buts around it. 2013-02-08T10:19:38 < ds2> I'd argue the other way 2013-02-08T10:19:48 < ds2> if you really intend it to be another variable, it should say static 2013-02-08T10:19:49 < dongs> wuut 2013-02-08T10:19:58 < dongs> well no fucking shit 2013-02-08T10:20:01 < ds2> extern is redundant/obsolete 2013-02-08T10:20:05 < dongs> huh 2013-02-08T10:20:13 < neuro_sys> huh? 2013-02-08T10:20:54 < ds2> I have had this arguement with coworkers 2013-02-08T10:21:11 < ds2> I don't see why you would expect it to be a different variable 2013-02-08T10:21:25 < dongs> um 2013-02-08T10:21:27 < ds2> initializing it causes an error cuz it is a contradiction 2013-02-08T10:21:28 < dongs> its not about different/same 2013-02-08T10:21:33 < dongs> its about coding style 2013-02-08T10:21:46 < ds2> ok 2013-02-08T10:21:47 < dongs> if you have a variable youre accessing from other source files 2013-02-08T10:21:51 < dongs> you had better make it extern so people reading your shit know what he fuck is going on 2013-02-08T10:22:04 < dongs> just dumbly declaring same variable name in 10 source files is dumb 2013-02-08T10:22:15 < ds2> ok 2013-02-08T10:22:19 < ds2> that is your coding style 2013-02-08T10:22:20 < dongs> thats all im bitching about. 2013-02-08T10:22:36 < ds2> even MS Manuals described that so it isn't an opensource thing 2013-02-08T10:22:57 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T10:23:01 -!- jef79m_lurking is now known as jef79m 2013-02-08T10:23:03 < dongs> i'd be willing to bet 99% of its usages are due to opensauced dudes not understanding what hte fuck they're doing thoug. 2013-02-08T10:23:26 < dongs> and when it works, they go "yay magic" 2013-02-08T10:23:33 < dongs> i've seen retards sticking shit like int foo; in header files 2013-02-08T10:23:33 < ds2> that I can agree with 2013-02-08T10:23:41 < dongs> and then including said header into a million .c files 2013-02-08T10:23:42 < ds2> it is like the whole arguement against and for the ## operator 2013-02-08T10:23:49 < dongs> yeah, that shit is useless too 2013-02-08T10:23:56 < ds2> I find ##'s very useful 2013-02-08T10:23:57 < dongs> fuck preprocessor abuse in general 2013-02-08T10:24:15 < dongs> \ in preprocessor should be banned too 2013-02-08T10:24:24 < dongs> i've seen PAGES of CODE written inside preprocessor shit 2013-02-08T10:24:26 < dongs> with \'s 2013-02-08T10:24:29 < dongs> RAGE 2013-02-08T10:24:54 -!- neuro_sys|off is now known as neuro_sys 2013-02-08T10:25:14 < zippe1> dongs: it's nice that you have ideas about the language 2013-02-08T10:25:21 < zippe1> It's just sad that they're not always right 2013-02-08T10:25:55 < dongs> zippe, never seen this kinda crap in quality commercial code. 2013-02-08T10:26:12 < zippe1> I don't think you'd know "quality commercial code" if it bit you on the ass 2013-02-08T10:26:27 < dongs> next youre gonna tell me using linker scripts to allocated blocks of shit in ram is also cool? 2013-02-08T10:26:30 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-08T10:26:49 < dongs> and then assigning them types from inside .c 2013-02-08T10:26:55 < dongs> and just expecting compiler/linker to figure it out 2013-02-08T10:27:24 < zippe1> dongs: the language doesn't provide a mechanism for specifying locations for storage 2013-02-08T10:27:40 < zippe1> dongs: arguably a failing, but one shared by most other languages 2013-02-08T10:28:27 < zippe1> I'm not quite sure what you're actually talking about, since you didn't bother to form a parseable sentence 2013-02-08T10:28:57 -!- jef79m is now known as jef79m_lurking 2013-02-08T10:29:01 < zippe1> But if the choice is between writing a new linker, or scripting the one I already have, I know what I'd prefer. 2013-02-08T10:29:20 < zippe1> (btw, re: tentative definitions, see MISRA-C rule 23) 2013-02-08T10:30:20 < zippe1> And if you aren't using MISRA or the JSF standards, see above inre: toothmarks on your backside 2013-02-08T10:30:28 < dongs> zippe1: something like doing char *__crap; and then using linker sections to allocate a block of (sram|flash) for i t 2013-02-08T10:30:36 < zyp> hmm, these tentative definitions, do they still work between different translation units? are the emitted symbols marked so they will be merged by the linker? 2013-02-08T10:30:47 < zippe1> zyp: yes; they're emitted as commons 2013-02-08T10:30:49 < dongs> zyp: merged 2013-02-08T10:30:58 < dongs> char foo; int foo = 1; // globasl will be sizeof(int) 2013-02-08T10:31:07 < zippe1> dongs: fail 2013-02-08T10:31:12 < dongs> zippe1: but true 2013-02-08T10:31:17 < zippe1> dongs: false 2013-02-08T10:31:26 < dongs> compile it, bitch 2013-02-08T10:31:26 < zippe1> dongs: your second definition is definite, not tentative 2013-02-08T10:31:34 < dongs> thats beacuse one has to exist 2013-02-08T10:31:36 < zyp> and all globals that's not initialized in a translation unit is tentative? 2013-02-08T10:31:40 < zippe1> negative 2013-02-08T10:31:45 < zippe1> zyp: correct 2013-02-08T10:31:53 < zippe1> zyp: unless they have static scope 2013-02-08T10:31:57 < zyp> of course 2013-02-08T10:32:20 < zyp> and then they will be merged with at most one non-tentative symbol in another translation unit 2013-02-08T10:32:30 < zyp> two of those will create a collision, right? 2013-02-08T10:32:45 < dongs> sounds liek a shitty feature anyway. 2013-02-08T10:32:53 < dongs> that leads to shit coding 2013-02-08T10:32:53 < dongs> which was my original rant anyway. 2013-02-08T10:32:59 < dongs> now back to dicknplacing shit, this isnt going anywehre :) 2013-02-08T10:33:16 < zyp> C++ doesn't have this, as far as I know 2013-02-08T10:33:41 < zyp> and yeah, I would prefer to be explicit anyway 2013-02-08T10:33:53 -!- jef79m_lurking is now known as jef79m 2013-02-08T10:34:15 < zippe1> Good discussion here: http://stackoverflow.com/questions/1490693/tentative-definitions-in-c99-and-linking 2013-02-08T10:34:18 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-08T10:34:29 < zippe1> zyp: -fno-commons is good practice, IMO 2013-02-08T10:36:08 -!- jef79m is now known as jef79m_lurking 2013-02-08T10:36:47 < dongs> Your search - -fno-commons - did not match any documents 2013-02-08T10:37:54 -!- claude is now known as Claude 2013-02-08T10:39:25 < dongs> zyp, stencil tomorow 9am. 2013-02-08T10:40:16 -!- jef79m_lurking is now known as jef79m 2013-02-08T10:43:50 < zyp> http://paste.jvnv.net/view/3GThA 2013-02-08T10:44:01 < zyp> so yeah, g++ behaves as gcc with -fno-common 2013-02-08T10:48:05 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-08T10:48:10 -!- Laurenceb_ [~Laurence@host86-143-184-68.range86-143.btcentralplus.com] has joined ##stm32 2013-02-08T10:48:23 -!- TeknoJuce [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer] 2013-02-08T10:51:07 -!- jef79m is now known as jef79m_lurking 2013-02-08T10:53:36 -!- jef79m_lurking is now known as jef79m 2013-02-08T11:04:44 -!- evil_dan2wik [~evil_dan2@unaffiliated/evil-dan2wik/x-0106201] has joined ##stm32 2013-02-08T11:07:29 -!- jef79m is now known as jef79m_lurking 2013-02-08T11:23:25 -!- Laurenceb_ [~Laurence@host86-143-184-68.range86-143.btcentralplus.com] has quit [Ping timeout: 246 seconds] 2013-02-08T11:45:39 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-08T11:49:57 < baird> Hmm. 35 cents, really? .. http://au.element14.com/renesas/rx63nrdk/development-board-rx63n-mcu/dp/2253395 2013-02-08T11:51:11 < dongs> yea 2013-02-08T11:51:13 < dongs> thats the "free" one 2013-02-08T11:51:16 < dongs> that they were giving away 2013-02-08T11:51:37 < dongs> so e14 is preoablyt just doing 35c to cover their administrative shit. 2013-02-08T11:51:40 < dongs> and charge you shipping 2013-02-08T11:52:14 < baird> So... it's legit? 2013-02-08T11:52:16 < dongs> yes 2013-02-08T11:52:21 < dongs> if you like weird non-arm core 2013-02-08T11:52:24 < dongs> and renesas shit 2013-02-08T11:52:53 < baird> I can avoid the whole shipping costs deal (collection from the warehouse).. 2013-02-08T11:52:58 < dongs> cool 2013-02-08T11:53:01 < dongs> grab 2 2013-02-08T11:53:06 < dongs> i'll paypal you to ship me one 2013-02-08T11:53:47 < dongs> i hope i dont have to prepay you 35c 2013-02-08T11:54:25 < zyp> do you really think he can afford to take that risk? 2013-02-08T11:54:40 < dongs> thats why i asked 2013-02-08T11:55:22 < dongs> wtf 2013-02-08T11:55:26 < dongs> dicknplace managed to put a led on backwards 2013-02-08T11:55:29 < karlp> dongs: http://www.verifiedfacts.org/i/Go%20Pro%20from%20Ebay/Mjk0MzE1 2013-02-08T11:55:30 < dongs> i wodner if it was backwards on tape or what 2013-02-08T11:56:09 < baird> Order: done. 2013-02-08T11:57:51 < baird> I'll shout him the 35 cents for the Internet Hugs he's given me. Postage however is a different story. 2013-02-08T11:59:26 < dongs> or we can put them towards y our purchase of dp adapter board 2013-02-08T11:59:48 < dongs> which would have been done 2 days ago if the fucking stencil would get here. 2013-02-08T12:01:44 < baird> That's doable too. 2013-02-08T12:06:16 < dongs> but i have beat the tracking # out of the assholes and its here tomorwo 9am 2013-02-08T12:06:18 < dongs> so all is good 2013-02-08T12:06:30 < dongs> arrives together with my scammed ipod nano 6th gen. 2013-02-08T12:28:53 -!- Viper168 [~Viper@node154.18.251.72.1dial.com] has joined ##stm32 2013-02-08T12:28:54 -!- Viper168 [~Viper@node154.18.251.72.1dial.com] has quit [Changing host] 2013-02-08T12:28:54 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-08T12:29:47 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] 2013-02-08T12:30:29 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Client Quit] 2013-02-08T12:30:47 -!- Viper168 [~Viper@node154.18.251.72.1dial.com] has joined ##stm32 2013-02-08T12:30:48 -!- Viper168 [~Viper@node154.18.251.72.1dial.com] has quit [Changing host] 2013-02-08T12:30:48 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-08T12:32:58 -!- jef79m_lurking is now known as jef79m 2013-02-08T12:35:22 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-08T12:40:52 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-08T13:02:11 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-08T13:17:32 -!- jef79m is now known as jef79m_lurking 2013-02-08T13:19:13 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T13:21:13 -!- jef79m_lurking is now known as jef79m 2013-02-08T13:31:45 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-08T13:31:46 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-08T13:31:46 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-08T13:31:49 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-08T14:11:17 -!- neuro_sys|off is now known as neuro_sys 2013-02-08T14:15:55 < dongs> lol, another useless 3d printer on dickstarter 2013-02-08T14:15:57 < dongs> shocking 2013-02-08T14:15:59 < dongs> http://www.kickstarter.com/projects/deltamaker/deltamaker-an-elegant-3d-printer-0 2013-02-08T14:19:14 <+Steffanx> :P 2013-02-08T14:19:45 < Laurenceb> daily heil 2013-02-08T14:20:28 <+Steffanx> Go to a channel where they actually like your nazi refernces Laurenceb 2013-02-08T14:20:46 < Laurenceb> dongs is the nazi 2013-02-08T14:27:55 < baird> Needs more Wagner background music. 2013-02-08T14:30:04 < Viper168> wanna see a picture of dongs ? 2013-02-08T14:30:08 < Viper168> http://sphotos-b.xx.fbcdn.net/hphotos-ash3/531575_546686568685692_1321546185_n.jpg 2013-02-08T14:32:05 < baird> Funfact: the hardware in https://www.youtube.com/watch?v=kU8ZKVBi1Qg also had 'Ride of the Valkyries'. Intended it for 'battle mode' with the robot it went in, 2013-02-08T14:32:30 -!- jef79m is now known as jef79m_lurking 2013-02-08T14:38:50 < Laurenceb> im bored 2013-02-08T14:38:56 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-08T14:39:27 < Laurenceb> any non fail stuff to linx? 2013-02-08T14:43:59 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 260 seconds] 2013-02-08T14:45:44 < Laurenceb> aha 2013-02-08T14:45:46 < Laurenceb> http://pic.epicfail.com/wp-content/uploads/2013/02/appetizing-cake-fail-poop.jpg 2013-02-08T14:50:52 < dongs> retweeted 2013-02-08T14:52:37 <+Steffanx> Uhm, that image is old dongs . See the artifacts 2013-02-08T14:55:46 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-08T15:12:06 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-08T15:21:05 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection] 2013-02-08T15:21:18 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-08T15:36:34 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-08T15:38:22 < Laurenceb> funtimes 2013-02-08T15:38:31 < Laurenceb> writing EKF filters 2013-02-08T15:39:22 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has quit [Quit: Ex-Chat] 2013-02-08T16:01:29 -!- Ranewen [~Ranewen4@87.252.128.194] has joined ##stm32 2013-02-08T16:08:27 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-08T16:13:37 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-08T16:32:37 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-08T16:36:32 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-08T16:38:38 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 252 seconds] 2013-02-08T16:38:39 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-08T16:45:09 -!- mrcan_ [~mrcan@78.173.24.21] has joined ##stm32 2013-02-08T16:45:10 -!- mrcan_ [~mrcan@78.173.24.21] has quit [Changing host] 2013-02-08T16:45:10 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-08T16:47:48 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 248 seconds] 2013-02-08T16:49:21 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-08T17:06:33 -!- Ranewen [~Ranewen4@87.252.128.194] has quit [Read error: Connection reset by peer] 2013-02-08T17:07:06 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-08T17:12:12 < Thorn> http://www.misra.org.uk/Buyonline/tabid/58/Default.aspx £10 nice 2013-02-08T17:13:21 < Thorn> http://ftp.iar.se/WWWfiles/guides/MisraC.pdf 2013-02-08T17:13:29 < Laurenceb> lolz 2013-02-08T17:13:35 < Laurenceb> £10 or free 2013-02-08T17:13:39 < Laurenceb> hmm ill take free 2013-02-08T17:24:37 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-08T17:38:15 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-08T17:47:06 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 245 seconds] 2013-02-08T17:53:40 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-08T17:54:26 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-08T17:57:03 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T18:15:26 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-08T18:17:00 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 276 seconds] 2013-02-08T18:17:01 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-08T18:28:23 -!- neuro_sys|off is now known as neuro_sys 2013-02-08T18:32:55 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-08T18:38:18 -!- Claude is now known as claude 2013-02-08T18:40:17 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T18:44:49 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-08T18:58:45 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-32.pptp.stw-bonn.de] has joined ##stm32 2013-02-08T19:11:09 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal] 2013-02-08T19:12:54 -!- tavish [~tavish@120.59.6.182] has joined ##stm32 2013-02-08T19:12:54 -!- tavish [~tavish@120.59.6.182] has quit [Changing host] 2013-02-08T19:12:54 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-08T19:25:10 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-08T19:46:56 -!- ggh [ada09aa9@gateway/web/freenode/ip.173.160.154.169] has joined ##stm32 2013-02-08T19:49:29 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32 2013-02-08T19:52:54 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 276 seconds] 2013-02-08T19:55:25 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-32.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-08T19:56:36 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal] 2013-02-08T19:58:21 -!- tavish [~tavish@120.59.6.182] has joined ##stm32 2013-02-08T19:58:21 -!- tavish [~tavish@120.59.6.182] has quit [Changing host] 2013-02-08T19:58:21 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-08T20:00:55 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-08T20:04:09 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-08T20:06:48 -!- barthess [~barthess@77.67.204.123] has joined ##stm32 2013-02-08T20:10:10 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-08T20:25:16 < Thorn> how do you use queues that only store pointers and not objects themselves (like in coos)? don't you have to essentially create another queue with locking and stuff? 2013-02-08T20:28:30 < zyp> huh? 2013-02-08T20:30:17 < zyp> the pointer would point to some buffer 2013-02-08T20:30:51 < zyp> which would be allocated from either the heap or some specific pool 2013-02-08T20:30:54 < Thorn> objects will have to be either dynamically allocated or is in some static array 2013-02-08T20:30:56 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 255 seconds] 2013-02-08T20:30:56 < Thorn> right 2013-02-08T20:31:31 < zyp> and since you'll be allocating and freeing from different threads, the allocator for that would have to be threadsafe 2013-02-08T20:31:44 < qyx_> overload protection: momentary, protection type: broken 2013-02-08T20:31:45 < qyx_> wat 2013-02-08T20:31:49 < Thorn> in both cases you need to have locking or critical sections (dynamic allocation most likely has that already) 2013-02-08T20:32:22 < Thorn> which is something freertos queues avoid 2013-02-08T20:32:28 < zyp> newlib have some hooks for threadsafety mechanisms 2013-02-08T20:32:33 < zyp> but I haven't looked into that myself 2013-02-08T20:33:12 < zyp> avoid how? 2013-02-08T20:33:45 -!- neuro_sys|off is now known as neuro_sys 2013-02-08T20:34:28 < zyp> a queue that doesn't store pointers would be some sort of a ringbuffer that you write data directly into, right? 2013-02-08T20:34:52 < zyp> that data access still have to be managed in an atomic way 2013-02-08T20:35:10 < Thorn> freertos manages all that transparently 2013-02-08T20:35:20 < zyp> sure, but it's still done. 2013-02-08T20:35:48 < zyp> a threadsafe memory pool will also manage all that transparently 2013-02-08T20:36:07 < zyp> so I'm not sure I see your point? 2013-02-08T20:36:33 < Thorn> malloc() and friends are evil in embedded code, a threadsafe pool is something you will need to implement yourself 2013-02-08T20:36:48 < zyp> why is malloc evil? 2013-02-08T20:38:28 < TitanMKD> zyp because of mem fragmentation 2013-02-08T20:38:32 < Thorn> for one thing you can't set specific limits for it, there's no way to check if "the queue is full" 2013-02-08T20:38:40 < Thorn> also fragmentation 2013-02-08T20:38:48 < TitanMKD> zyp and also non predictive time to alloc blocks 2013-02-08T20:38:49 < Thorn> i.e. per-queue limits 2013-02-08T20:39:09 < Thorn> and that too 2013-02-08T20:39:29 < zyp> not entirely true, but ok 2013-02-08T20:39:54 < Laurenceb> i dodnt see why its unsafe if used sanely 2013-02-08T20:40:07 < Thorn> there was a great article called "malloc() considered harmful" or something like that, can't find now 2013-02-08T20:40:22 < zyp> well, that's besides the point anyway 2013-02-08T20:40:32 < gxti> if someone writes a blog post called "X considered harmful" then X is probably not the antichrist 2013-02-08T20:40:37 < Thorn> also afaik many guidelines for safety code only allow it when creating tasks, not inside task loops 2013-02-08T20:42:12 < Thorn> add to the list inefficient RAM utilization (related to fragmentation) and there's not much RAM in mcus 2013-02-08T20:42:47 -!- neuro_sys is now known as neuro_sys|off 2013-02-08T20:43:08 < Thorn> http://mil-embedded.com/articles/justifiably-apis-militaryaerospace-embedded-code/ 2013-02-08T20:43:30 < zyp> well, sure, then use a fixed size pool then. 2013-02-08T20:44:42 < zyp> full buffer queues also have disadvantages, considering that you have to copy the contents into them 2013-02-08T20:44:59 < zyp> which may be a big penalty for large messages 2013-02-08T20:45:51 < Thorn> well you can store pointers in a freertos-style queue if you want to 2013-02-08T20:46:19 < zyp> of course, the mechanics are the same anyway 2013-02-08T20:46:26 < Thorn> but you can *only* store pointers in a coos-style one 2013-02-08T20:46:59 < zyp> or four bytes of other data :p 2013-02-08T20:48:05 < Thorn> right, 3 pointers = 1 CAN message 2013-02-08T20:48:08 < zyp> so, the freertos-style of queue have a variable message size, while a coos-style queue have a fixed size? 2013-02-08T20:48:35 < Thorn> pretty much, with the fixed size being = platform pointer size (32 bits) 2013-02-08T20:49:08 < zyp> and then you have the middle ground which has a compile time configurable message size 2013-02-08T20:49:56 < Thorn> haven't seen these but seems crazy, I may need a pair of char queues for an UART and another queue for CAN mesasges 2013-02-08T20:50:32 < zyp> can messages would most efficiently be stored in a fixed size queue, and has close to none penalty for copying due to the small size 2013-02-08T20:51:01 < zyp> ethernet frames would be a completely different matter 2013-02-08T20:51:37 < Thorn> if you can define several queues of different sizes using macros at compile time it'd be fine with me 2013-02-08T20:52:07 < zyp> that's the thing C++ templates are well suited for 2013-02-08T20:52:13 < zippe1> jpa-: thanks for all the NuttX fixes, btw. 2013-02-08T20:52:25 < Thorn> right, that too 2013-02-08T20:53:14 < zyp> my logging class is pretty much a write-only queue with compile-time configurable fixed size messages 2013-02-08T20:53:17 < zippe1> Thorn: better to avoid queues for UART - spend the memory on FIFO space in the driver. 2013-02-08T20:54:00 < zyp> isn't a FIFO also covered by the queue term? 2013-02-08T20:54:04 < Thorn> it'd be great, but stm32 = no hardware fifo. so the queue is essentially the fifo 2013-02-08T20:54:08 < Thorn> right 2013-02-08T20:54:29 < zyp> a UART queue would be a ringbuffer of bytes 2013-02-08T20:55:25 < zyp> might even set up dma to do the writing part of it 2013-02-08T20:55:53 < Thorn> yes, one disadvantage of freertos queues is they aren't dma aware 2013-02-08T20:56:19 < zippe1> Thorn: that assumes you're not using DMA, but yes 2013-02-08T20:56:23 < zyp> are you sure freertos only have one type of queues? 2013-02-08T20:56:55 < Thorn> seems so http://www.freertos.org/a00018.html 2013-02-08T21:05:14 < emeb> for a moment I thought that said "freetos" 2013-02-08T21:05:38 < emeb> and then I thought "mmmm.... tasty snack" 2013-02-08T21:06:35 < gxti> GPL flavor, yuck /s 2013-02-08T21:08:59 < emeb> maybe it's BSD corn chips? 2013-02-08T21:09:03 < Thorn> coos has some support for memory pools http://www.coocox.org/CoOSguide/CoOS_Dynamic_Memory.htm 2013-02-08T21:10:01 < Thorn> no idea if they're concurrent safe 2013-02-08T21:10:07 < Thorn> doesn't appear so 2013-02-08T21:14:05 < zyp> dicking around with verilog now 2013-02-08T21:14:16 < emeb> what cha makin? 2013-02-08T21:14:25 < zyp> making a new config for the cpld on the quadrotor board 2013-02-08T21:14:59 < zyp> so I get motor outputs 2013-02-08T21:15:19 < zyp> maybe I'll get to crash something in my ceiling or something this weekend 2013-02-08T21:15:37 < jpa-> zippe1: did i fix something? 2013-02-08T21:16:07 < emeb> big fun 2013-02-08T21:16:23 < emeb> tape pillows to the ceiling 2013-02-08T21:16:47 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-08T21:17:01 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-08T21:17:24 < zyp> time to see if impact wants to be my friend today 2013-02-08T21:18:01 < emeb> quadrotor + impact -> doesn't bode well. 2013-02-08T21:18:46 < zyp> well, I'm referring to the bad kind of impact now 2013-02-08T21:19:19 < emeb> ya - Xilinx impact. Nasty thing. 2013-02-08T21:19:38 < emeb> Ordered RXADC parts from DK this morning, just shipped. That was fast. 2013-02-08T21:20:50 < zyp> :) 2013-02-08T21:20:51 < zyp> hmm 2013-02-08T21:21:05 < zyp> usual problem, impacts hides the «program» button in the menu 2013-02-08T21:21:30 < zyp> there used to be some trick to get it to show up, but I don't remember what 2013-02-08T21:22:35 < zyp> ah, yeah 2013-02-08T21:23:05 < zyp> disconnect after configuring scan chain, try autoconnect (which fails), then manually connect, and they magically shows up 2013-02-08T21:23:24 < zyp> just disconnecting and manually connecting again doesn't work 2013-02-08T21:24:12 < Thorn> you probably need to chant something in latin while doing that 2013-02-08T21:24:18 < zyp> :) 2013-02-08T21:25:00 < emeb> somebody wave the rubber chicken over it. 2013-02-08T21:25:16 < zyp> I should just write an xsvf player or something, so I don't have to deal with it 2013-02-08T21:26:40 < emeb> that app really needs to be burned to the ground and rebuilt. 2013-02-08T21:42:11 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has joined ##stm32 2013-02-08T21:46:36 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving] 2013-02-08T21:57:01 <+Steffanx> Hmm, chinese new year and the chinese post track and trace services is down? 2013-02-08T21:57:11 <+Steffanx> -s 2013-02-08T21:57:47 <+Steffanx> Not that the services is that useful 2013-02-08T21:57:52 <+Steffanx> -*service 2013-02-08T22:02:34 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-08T22:04:49 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-08T22:06:56 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-08T22:15:20 -!- gxti [~gxti@ada.partiallystapled.com] has quit [Ping timeout: 255 seconds] 2013-02-08T22:17:39 -!- gxti [~gxti@ada.partiallystapled.com] has joined ##stm32 2013-02-08T22:19:19 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Ping timeout: 246 seconds] 2013-02-08T22:21:50 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-08T22:25:24 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 264 seconds] 2013-02-08T22:25:24 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-08T22:50:45 -!- daku is now known as DaKu 2013-02-08T22:55:38 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7524:2eb:53bc:b2ec] has joined ##stm32 2013-02-08T22:56:35 -!- DaKu is now known as daku 2013-02-08T22:58:28 -!- daku is now known as DaKu 2013-02-08T23:07:59 -!- DaKu is now known as daku 2013-02-08T23:16:42 -!- daku is now known as DaKu 2013-02-08T23:34:45 -!- DaKu is now known as daku --- Day changed Sat Feb 09 2013 2013-02-09T00:01:23 -!- toyz [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has joined ##stm32 2013-02-09T00:02:30 < toyz> evening... id just want you to point this http://dangerousprototypes.com/2013/02/08/video-course-embedded-c-programming-with-arm-cortex-m/ as an starting point for all newbies like myself :) 2013-02-09T00:06:10 -!- toyz [a135b3e8@gateway/web/freenode/ip.161.53.179.232] has quit [Client Quit] 2013-02-09T00:26:23 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:7524:2eb:53bc:b2ec] has quit [Ping timeout: 256 seconds] 2013-02-09T00:29:53 -!- Laurenceb_ [~Laurence@host86-140-225-57.range86-140.btcentralplus.com] has joined ##stm32 2013-02-09T00:51:44 -!- neuro_sys|off is now known as neuro_sys 2013-02-09T01:07:36 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-09T01:12:00 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has joined ##stm32 2013-02-09T01:26:45 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-09T01:27:20 < dirty_d> its a freakin blizzard out there 2013-02-09T01:33:03 <+Steffanx> Yay 2013-02-09T01:33:12 < dirty_d> snow sucks 2013-02-09T01:33:20 <+Steffanx> Send it to me 2013-02-09T01:33:33 < dirty_d> you dont want no part of this shiiiit 2013-02-09T01:33:41 <+Steffanx> i do 2013-02-09T01:33:48 < dirty_d> where are you from? 2013-02-09T01:33:57 <+Steffanx> Dutchland remember 2013-02-09T01:34:07 -!- barthess [~barthess@77.67.204.123] has quit [Quit: Leaving.] 2013-02-09T01:34:20 < dirty_d> oh yea, dont you have a lot of snow there? 2013-02-09T01:34:28 <+Steffanx> No 2013-02-09T01:34:42 <+Steffanx> Maybe a few inches a few weeks/year 2013-02-09T01:35:01 < dirty_d> oh, but its definitely cold enough right? 2013-02-09T01:35:07 < baird> Cicadias have been at it since daybreak.. not a good sign. 2013-02-09T01:35:16 <+Steffanx> It's ~-2C now 2013-02-09T01:35:27 <+Steffanx> but it's 00.35 :P 2013-02-09T01:35:54 < baird> 10am, 30degC. 2013-02-09T01:36:07 < dirty_d> its 1C here now 2013-02-09T01:36:09 < baird> Probably get to near 40 2013-02-09T01:36:20 < dirty_d> are you from hell? 2013-02-09T01:37:00 <+Steffanx> Austalia is a close to hell one can get 2013-02-09T01:37:06 <+Steffanx> on earth taht is 2013-02-09T01:37:10 < baird> Our spiders are bigger. 2013-02-09T01:37:34 <+Steffanx> gn 2013-02-09T01:37:37 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-09T01:37:41 < Laurenceb_> http://www.google.co.uk/url?sa=t&rct=j&q=john%20crowe%20john%20radcliffe%20hospital%20thesis&source=web&cd=4&ved=0CEEQFjAD&url=http%3A%2F%2Fetheses.nottingham.ac.uk%2F2501%2F1%2FMedical_Device_Design_for_Adolescents_Thesis.pdf&ei=_YoVUcmkBciZ0QXj2YC4BQ&usg=AFQjCNFVBA4ZS6cuNbFDEWfiFMeknROOhg&bvm=bv.42080656,d.d2k&cad=rja 2013-02-09T01:38:35 < Laurenceb_> "Human Factors (HF) practices in the design of medical devices are increasingly being recognised as important contributors in the development of successful products. The recent introduction of ISO 62366:2008 Medical Devices: Application of usability engineering to medical devices (ISO 2008) raises the profile of HF in a healthcare context and formalises the need for developers to consider usability in their devices." 2013-02-09T01:38:47 < Laurenceb_> I. AM. SO. EXCITED!!! 2013-02-09T01:39:26 < baird> Radiation is Rad, dude. 2013-02-09T01:40:35 < baird> "Can you help me tie this knot?" "ArE U injecting Heroin???" "Nah, just insulin. But it makes me look badass." 2013-02-09T01:40:45 < Laurenceb_> thats like the most exciting thesis of all time 2013-02-09T01:41:00 < gxti> Laurenceb_: you are weird 2013-02-09T01:41:28 < Laurenceb_> erm... im trolling 2013-02-09T01:42:34 < Laurenceb_> protip: dont click that link or you _will_ die of boredom 2013-02-09T01:44:13 < dirty_d> lol 2013-02-09T01:46:38 < dongs> sup trolls 2013-02-09T01:49:38 < Laurenceb_> OMFG page 36 2013-02-09T01:49:43 < Laurenceb_> the pentogrammmmmm 2013-02-09T01:49:46 < baird> With kids under 10, you can get complice from them by using the Germ Threat. "You have to take this medicine every day, or otherwise you'll get GEEERRMMMSS under your SSKKIINNN..." 2013-02-09T01:59:59 -!- _BJFreeman [~bjfree@66.sub-75-196-116.myvzw.com] has joined ##stm32 2013-02-09T02:00:27 -!- _BJFreeman is now known as BJFreeman 2013-02-09T02:04:41 -!- ggh [ada09aa9@gateway/web/freenode/ip.173.160.154.169] has quit [Quit: Page closed] 2013-02-09T02:17:57 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-09T02:30:44 < dongs> dp stencil is here 2013-02-09T02:30:46 < dongs> time to see the fail 2013-02-09T02:32:15 < zyp> nice 2013-02-09T02:36:58 < dongs> istill gotta write backlight driver code i guess 2013-02-09T02:37:20 -!- Laurenceb_ [~Laurence@host86-140-225-57.range86-140.btcentralplus.com] has quit [Read error: Operation timed out] 2013-02-09T02:43:04 < dirty_d> any of you done 1-Wire on stm32? 2013-02-09T02:48:16 < dirty_d> my apartment feels like its gonna blow over 2013-02-09T02:48:44 < dongs> got my arduinoshield for cubieboard assembled 2013-02-09T02:51:05 < dongs> is imgur slow as shit fo reveryone else 2013-02-09T02:51:26 < GargantuaSauce> nope 2013-02-09T02:51:44 < dongs> http://i.imgur.com/q1GQHmU.jpeg 2013-02-09T02:52:31 < GargantuaSauce> composite+vga? 2013-02-09T02:53:01 < dongs> and mpeg2ts port on the bottom which was the main reason for making a pcb 2013-02-09T02:53:35 < dongs> but i figured if im wasting space might as well do it in style 2013-02-09T02:55:21 < GargantuaSauce> cool...is anything on that board driving them or does it just connect to the microcontroller's dacs? 2013-02-09T02:56:21 < GargantuaSauce> oh the cubie has hardware for that 2013-02-09T02:56:50 < dongs> A10 has vga out 2013-02-09T02:56:54 < dongs> just conencting pins 2013-02-09T02:57:53 < dirty_d> how would you do 1-wire on stm32? 2013-02-09T02:58:17 < dirty_d> set gpio to input, and switch between pull-up and pull-down for signaling? 2013-02-09T02:58:28 < dongs> just.. do it 2013-02-09T02:58:35 < dongs> i know someone who did onewire-ish stuff 2013-02-09T02:58:39 < dongs> i mean that kinda protocol. 2013-02-09T02:58:42 < dirty_d> looks slow and simple 2013-02-09T02:59:01 < dongs> http://code.google.com/p/esc32/source/browse/trunk/onboard/ow.c 2013-02-09T02:59:04 < qyx_> i was trying to do that but then i used i2c temp sernsor 2013-02-09T02:59:33 < dirty_d> i thnk you could communicate to 15 1-wire devices all at the same time with one port right? 2013-02-09T02:59:37 < qyx_> dirty_d: you need opendrain for that with relatively strong pullup 2013-02-09T02:59:55 < dirty_d> qyx_, what if you actually power the devices though 2013-02-09T02:59:57 < dirty_d> no reason not to 2013-02-09T03:00:22 < qyx_> yes, only if you use parasitic power 2013-02-09T03:00:34 < dirty_d> those temp sensors take 750ms to do a conversion 2013-02-09T03:00:54 < dirty_d> so i think talking to them all at the same time would be better 2013-02-09T03:01:37 < dirty_d> you could sample all 15 pins in one read 2013-02-09T03:01:46 < qyx_> you can start conversion on all devices on the bus 2013-02-09T03:01:53 < qyx_> and then just read the results 2013-02-09T03:02:07 < dirty_d> and shift the bits into a variable for each device or something 2013-02-09T03:02:46 < qyx_> 15 pins? 2013-02-09T03:02:50 < qyx_> why 15 pins? 2013-02-09T03:02:57 < dirty_d> im just sayin, if you had 15 temp sensors 2013-02-09T03:03:09 < qyx_> you just connect them in parallel 2013-02-09T03:03:10 < dirty_d> i think it would be doiable to have a pin dedicated to each and communicate in parallel 2013-02-09T03:03:20 < dirty_d> instead of communicating one at a time 2013-02-09T03:03:57 < qyx_> yes, but why? baudrate is somewhere around 15kbaud 2013-02-09T03:04:17 < dirty_d> meh, i dunno, i guess youre right 2013-02-09T03:04:28 < dirty_d> the 750ms is unavoidable 2013-02-09T03:04:42 < dirty_d> so i guess the time youd lose going 1 by 1 is negligible 2013-02-09T03:05:01 < qyx_> you dont have to wait 750ms for each sensor 2013-02-09T03:05:17 < qyx_> you start conversion on every sensor, then wait 750ms 2013-02-09T03:05:21 < dirty_d> right, yea 2013-02-09T03:05:22 < qyx_> and the read them all 2013-02-09T03:06:30 < qyx_> but there needs to be enough power for all sensors converting at the same time - dedicated vcc, strong pullup or transistor driven "pullup" during conversion 2013-02-09T03:06:50 < dirty_d> yea 2013-02-09T03:06:51 < dongs> k workshit done time to do dp board 2013-02-09T03:07:09 < GargantuaSauce> also is this cubieboard like the rpi in that it's "open" but peripherals like the onboard gpu require closed drivers? 2013-02-09T03:09:41 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-09T03:11:40 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-09T03:24:16 < dongs> at least its standard-ish drivers 2013-02-09T03:24:22 < dongs> not the totally closed videocore bullshit 2013-02-09T03:24:25 < dongs> its mali400 or someshit 2013-02-09T03:24:29 < dongs> there's even some opensauce attempts on making it work 2013-02-09T03:24:33 < dongs> since there are at least specs available 2013-02-09T03:24:45 < dongs> not that I give a fuck 2013-02-09T03:30:17 < GargantuaSauce> i was mostly wondering about the viability of the camera interface but it looks like it pipes it straight to ram as opposed to going through the video hardware like the rpi's SoC does 2013-02-09T03:30:23 < GargantuaSauce> so..sweet i am totally getting one of these 2013-02-09T03:30:57 < GargantuaSauce> cause i think doing CV on the f4 is probably pushing it a little 2013-02-09T03:30:58 < dongs> yeah CSI interface just dumps it to dma buffers in ram 2013-02-09T03:31:09 < dongs> for some fucked up reason they broke out CSI0 but not its clock lines 2013-02-09T03:31:11 < dongs> so only CSI1 is usable 2013-02-09T03:31:12 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-09T03:31:24 < dongs> CSI0 has some extra ISP stuff for image processing, CSI1 doesnt 2013-02-09T03:39:49 < GargantuaSauce> oh it has hardware video encoding too 2013-02-09T03:39:51 < GargantuaSauce> sploosh. 2013-02-09T03:45:42 < dirty_d> have to do anything special to enable the TIM1 update interrupt except define the function and set the UIE bit in the timer reg? 2013-02-09T03:45:50 < dirty_d> i have void TIM1_UP_TIM16_IRQHandler() defined 2013-02-09T03:49:52 < dongs> pretty sure tim1 has its own irq 2013-02-09T03:51:10 < dirty_d> yea 2013-02-09T03:51:16 < dirty_d> well thats the one 2013-02-09T03:51:34 < dongs> lol scampod nano arrived 2013-02-09T03:51:45 < dirty_d> seems i have to setup NVIC first 2013-02-09T03:52:04 < dirty_d> bvut i dont see anything in the at section in the manual except for external interrupt regs 2013-02-09T03:53:44 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-09T03:57:05 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-09T03:57:42 < dongs> its in teh oven 2013-02-09T03:57:56 < emeb_mac> what are we baking? 2013-02-09T03:58:04 < dongs> dp_adapter 2013-02-09T03:58:12 < emeb_mac> oooo... kewl 2013-02-09T03:58:41 < emeb_mac> vicarious hardware assembly friday (or saturday). 2013-02-09T04:04:14 < dongs> http://i.imgur.com/TPB91F5.jpg seems not bad 2013-02-09T04:04:29 < dongs> time to try testing in a bit 2013-02-09T04:09:31 < emeb_mac> looks good. 2013-02-09T04:10:03 < emeb_mac> code already written? 2013-02-09T04:10:40 < dongs> havent even started. 2013-02-09T04:10:54 < dongs> im pro like that 2013-02-09T04:11:55 < dirty_d> im super duper confused 2013-02-09T04:11:59 < emeb_mac> confidence 2013-02-09T04:12:08 < dirty_d> if i put TIM1->SR &= ~TIM_SR_UIF_gm; as the first thing in the ISR, it works 2013-02-09T04:12:20 < dirty_d> if i put it at the end, it doesnt even get called once 2013-02-09T04:12:22 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 256 seconds] 2013-02-09T04:12:24 < dirty_d> wtf? 2013-02-09T04:13:03 < GargantuaSauce> i noticed behaviour like that too...put the line back at the top and decided to choose another battle :V 2013-02-09T04:13:11 < dongs> wel, nothign smoked on usb pluigin 2013-02-09T04:13:13 < dongs> so i think im ok 2013-02-09T04:13:32 < dirty_d> GargantuaSauce, is it like stuck in an infine loop recalling the interrupt? 2013-02-09T04:13:36 < dirty_d> or some wachy madness 2013-02-09T04:13:50 < GargantuaSauce> perhaps, i dunno 2013-02-09T04:17:11 < dirty_d> im gonna lose freakin power 2013-02-09T04:17:14 < dirty_d> lights are flickering 2013-02-09T04:17:22 < dirty_d> apartment is shaking bad 2013-02-09T04:17:29 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has joined ##stm32 2013-02-09T04:17:44 < emeb_mac> erm... where you at? 2013-02-09T04:17:45 < GargantuaSauce> a victim of this dumb snowstorm? 2013-02-09T04:18:00 < GargantuaSauce> it's going to hit me in a few hours :( 2013-02-09T04:18:13 < dirty_d> emeb_mac, massachusetts 2013-02-09T04:18:17 < dirty_d> yea 2013-02-09T04:18:19 < emeb_mac> dirty_d: epicenter 2013-02-09T04:18:26 < dirty_d> yea buzzards bay 2013-02-09T04:18:30 < dirty_d> right in the worst part 2013-02-09T04:19:28 < GargantuaSauce> http://www.weatheroffice.gc.ca/data/satellite/goes_ecan_1070_100.jpg yeah that is pretty menacing 2013-02-09T04:19:28 < emeb_mac> good luck to you. 2013-02-09T04:19:46 < GargantuaSauce> and i am on that crescent-shaped hunk of sandstone just north of it 2013-02-09T04:19:48 < Erlkoenig> pussies, who cares about a bit of snow :P 2013-02-09T04:19:50 < emeb_mac> the armpit of cape cod? 2013-02-09T04:20:42 * emeb_mac lived in rockport/gloucester area as a kid 2013-02-09T04:24:48 < dirty_d> my apartmens is shaking worse than it did during hurricane sandy 2013-02-09T04:25:23 < Erlkoenig> what... american way to build things 2013-02-09T04:25:49 < dirty_d> very old 4-story 2013-02-09T04:28:19 < dirty_d> GargantuaSauce, seems that if i do TIM1->SR &= ~TIM_SR_UIF_gm; NVIC_ClearPendingIRQ(TIM1_UP_TIM16_IRQn); at the end of the ISR, everything is fine 2013-02-09T04:28:58 < dirty_d> The function clears the pending bit of an external interrupt. 2013-02-09T04:29:13 < dirty_d> external as in external to the M4 core, not the MCU right? 2013-02-09T04:30:23 < Erlkoenig> yeah thats's a nice trap 2013-02-09T04:30:47 < dirty_d> Erlkoenig, hmm? 2013-02-09T04:31:07 < Erlkoenig> not knowing that you have to clear the interrupt bit :) 2013-02-09T04:31:23 < Erlkoenig> else your ISR will occur always 2013-02-09T04:31:54 < emeb_mac> interesting. 2013-02-09T04:32:14 < emeb_mac> don't recall seeing that NVIC_ClearPendingIRQ() used in the example code 2013-02-09T04:32:30 < dirty_d> Erlkoenig, yea but, if i only clear the TIM1 flag bit at the beginning of the ISR, everything is fine 2013-02-09T04:32:33 < dirty_d> why is that? 2013-02-09T04:32:50 < dirty_d> emeb_mac, i think thats because its all done behind STs higher level functions 2013-02-09T04:32:52 < dirty_d> but im not sure 2013-02-09T04:33:18 < emeb_mac> hmmm... need to research this. 2013-02-09T04:33:20 < Erlkoenig> dirty_d: but not when at the end of the ISR? :o 2013-02-09T04:33:26 < dirty_d> Erlkoenig, correct 2013-02-09T04:33:30 < Erlkoenig> wait a sec... 2013-02-09T04:33:37 < dirty_d> does that make any sense? 2013-02-09T04:33:43 < dirty_d> all i do is toggle a GPIO pin in the ISR 2013-02-09T04:34:12 < dongs> i wonder if its safe to just power up backlight driver without first talkign to it :) 2013-02-09T04:35:01 < Erlkoenig> i use in my code: TIM_ClearITPendingBit(TIMx, TIM_IT_Update); 2013-02-09T04:35:07 < Erlkoenig> at the end of the ISR 2013-02-09T04:35:17 < dirty_d> maybe that function also does the other one 2013-02-09T04:35:20 < dirty_d> ill look 2013-02-09T04:35:47 < dirty_d> Erlkoenig, anyway, this is the code http://pastebin.com/NcuZ4nPi 2013-02-09T04:35:51 < dirty_d> maybe im doing something wacky 2013-02-09T04:36:01 < Erlkoenig> TIM_ClearITPendingBit does this: TIMx->SR = (uint16_t)~TIM_IT; 2013-02-09T04:36:12 < dirty_d> hmmm 2013-02-09T04:36:31 < Erlkoenig> clearing the NVIC pending flag is not so good i think 2013-02-09T04:36:44 < dirty_d> i dont have that bit 2013-02-09T04:36:48 < dirty_d> im on f3 2013-02-09T04:36:49 < Erlkoenig> because if the Timer's "pending" bit is still active, it will trigger a new interrupt instantly, i.e. activate the NVIC pending bit 2013-02-09T04:36:59 < Erlkoenig> NVIC_ClearPendingIRQ(TIM1_UP_TIM16_IRQn); <--- so what is this? :D 2013-02-09T04:37:14 < Erlkoenig> that's the NVIC's bit, not the Timer's bit 2013-02-09T04:37:27 < dirty_d> thats a lower level thing in the core i believe 2013-02-09T04:38:18 < Erlkoenig> wut... this clearly looks like a function for a NVIC bit :D 2013-02-09T04:38:44 < dongs> no smoke yet.. 2013-02-09T04:39:19 < dirty_d> Erlkoenig, yea, the interrupt controller of the cortex M4 2013-02-09T04:39:30 < Erlkoenig> for the M3: NVIC->ICPR[((uint32_t)(IRQn) >> 5)] = (1 << ((uint32_t)(IRQn) & 0x1F)); /* Clear pending interrupt */ 2013-02-09T04:39:48 < Erlkoenig> core_m3.h:986 2013-02-09T04:41:08 < Erlkoenig> but the NVIC's pending bit is triggered again by the timer, if you don't tell the timer to stop doing that, by clearing the TIMx_SR::TIM_IT bit 2013-02-09T04:41:53 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-09T04:42:12 < dirty_d> Erlkoenig, yea thats why i do both 2013-02-09T04:42:51 < dirty_d> but i dont understand why i have to only at the end of the ISR 2013-02-09T04:43:08 < Erlkoenig> i don't see a TIM_ClearITPendingBit in your code :D 2013-02-09T04:43:44 < Erlkoenig> you should only need a TIM_ClearITPendingBit anywhere in your ISR 2013-02-09T04:44:50 < dongs> 3.3 switcher OK 2013-02-09T04:45:14 < dongs> wau 2013-02-09T04:45:17 < dongs> 24V out of step-upo 2013-02-09T04:45:19 < dongs> also seems ok 2013-02-09T04:46:03 < dirty_d> Erlkoenig, doesnt that fucntion do the same thing im doing? 2013-02-09T04:46:10 < Erlkoenig> nooo 2013-02-09T04:46:14 < Erlkoenig> there are TWO bits 2013-02-09T04:46:21 < Erlkoenig> a bit in the NVIC, and a bit in the timer 2013-02-09T04:46:21 < dirty_d> oh crap 2013-02-09T04:46:39 < Erlkoenig> if the bit in the timer is '1', the bit in the NVIC will go to '1' again 2013-02-09T04:46:50 < dirty_d> yea but that function only clears the timers bit 2013-02-09T04:47:00 < Erlkoenig> yes and that should be enough 2013-02-09T04:47:18 < Erlkoenig> at least that works perfectly for me :o) 2013-02-09T04:47:20 < dirty_d> TIM1->SR &= ~TIM_SR_UIF_gm 2013-02-09T04:47:25 < dirty_d> am i doing that wrong? 2013-02-09T04:47:26 < Erlkoenig> that's not the TIM_IT bit 2013-02-09T04:47:28 < Erlkoenig> that's the UIF bit 2013-02-09T04:47:31 < dirty_d> there is no TIM_IT bit 2013-02-09T04:47:35 < dirty_d> TIM_IT is the parameter to the function 2013-02-09T04:47:55 < Erlkoenig> oh then it's different for the F4 2013-02-09T04:48:01 < dirty_d> void TIM_ClearITPendingBit(TIM_TypeDef* TIMx, uint16_t TIM_IT) 2013-02-09T04:48:19 < Erlkoenig> (3:36:01 AM) Erlkoenig: TIM_ClearITPendingBit does this: TIMx->SR = (uint16_t)~TIM_IT; 2013-02-09T04:48:26 < Erlkoenig> oh wait 2013-02-09T04:48:28 < Erlkoenig> fail 2013-02-09T04:48:29 < dirty_d> Erlkoenig, whats your functions definition? 2013-02-09T04:48:29 < dongs> dp part workds :) 2013-02-09T04:48:33 < dongs> i got new screen 2013-02-09T04:48:38 < dongs> ok, woot. 2013-02-09T04:48:42 < dongs> time to write code for proper backlight and shit 2013-02-09T04:48:45 < Erlkoenig> okay UIF is correct. sorry. :D 2013-02-09T04:48:56 < dirty_d> ok 2013-02-09T04:49:03 < dirty_d> also, that function is doing =, not &= 2013-02-09T04:49:06 < dirty_d> im not sure why 2013-02-09T04:49:32 < Erlkoenig> hmm probably because writing '1' to the status bits doesnt change anything 2013-02-09T04:50:01 < dirty_d> im writing 0 to that bit 2013-02-09T04:50:08 < dirty_d> with &= 2013-02-09T04:50:18 < Erlkoenig> hmm correct 2013-02-09T04:51:29 < Erlkoenig> okay so technically doing both should suffice :D 2013-02-09T04:55:25 < dirty_d> Erlkoenig, hmm, i tried doing the same exact thing as that function, and it doesnt work 2013-02-09T04:55:46 < Erlkoenig> does your ISR occur to often or what's your problem? 2013-02-09T04:56:18 < dirty_d> Erlkoenig, nope, once a second 2013-02-09T04:56:35 < dirty_d> it does work if i put it at the beginning of the isr though 2013-02-09T04:56:47 < dirty_d> but if i put it at the end, along with the NVIC function, it works 2013-02-09T04:56:51 < dirty_d> i dont get it 2013-02-09T04:57:13 < Erlkoenig> so TIM1->SR &= ~TIM_SR_UIF_gm; alone, at the beginning works, and both at the end, work? 2013-02-09T04:57:36 < dirty_d> yup 2013-02-09T04:58:16 < Erlkoenig> okay that makes sense 2013-02-09T04:58:23 < dirty_d> it does? 2013-02-09T04:58:52 < dirty_d> if i put TIM1->SR&=~TIM_SR_UIF_gm along at the end, the ISR doesnt even get called 2013-02-09T04:58:53 < dirty_d> lol 2013-02-09T04:59:43 < Erlkoenig> well the timer overflows or whatever, and sets the UIF bit, which results in the NVIC setting its bit. as soon as your ISR starts the NVIC's bit is cleared. But if you do some work and don't clear the timer's bit, the NVIC's bit gets set again... If you clear the UIF at the beginning, the NVIC's bit is not set again 2013-02-09T05:00:15 < dirty_d> whats that have to do with the ISR not even getting called once though? 2013-02-09T05:01:11 < Erlkoenig> dunno, thats weird :D 2013-02-09T05:01:14 < dirty_d> yea i know 2013-02-09T05:01:16 < Erlkoenig> try power-cycling :D 2013-02-09T05:01:25 < dirty_d> code that hasnt even been executed yet is altering the past 2013-02-09T05:01:29 < dirty_d> i think i have a time machine here 2013-02-09T05:01:43 < Erlkoenig> no, STM32's are just chaotic systems :> 2013-02-09T05:01:52 < Erlkoenig> or nondeterministic turing machines 2013-02-09T05:03:07 < dirty_d> well its probably not that 2013-02-09T05:03:07 < dirty_d> lol 2013-02-09T05:03:18 < dirty_d> i bet something is just happening so fast i dont see the LED turn on before it turns off 2013-02-09T05:03:44 < Erlkoenig> yes thats possible. use an oscilloscope :> 2013-02-09T05:04:00 < dirty_d> i dont have one :/ 2013-02-09T05:04:08 < dirty_d> i gotta read about the cortex M4 2013-02-09T05:04:23 < dirty_d> maybe you have to clear the pending bit x cycles after the ISR starts or something 2013-02-09T05:04:53 < Erlkoenig> not the pending bit but the UIF bit :) 2013-02-09T05:05:12 < Erlkoenig> an oscilloscope is very useful in debugging timers :> ... you could set a breakpoint in your ISR 2013-02-09T05:05:13 < dirty_d> yea that 2013-02-09T05:07:55 < dirty_d> i wonder if it has to do with using -O2 2013-02-09T05:08:38 < Erlkoenig> wah possible... that reorders functions 2013-02-09T05:08:48 < dirty_d> what the shit 2013-02-09T05:08:50 < dirty_d> that was it 2013-02-09T05:08:57 < Erlkoenig> use asm volatile (""); or something to prevent reordering :> 2013-02-09T05:09:23 < dirty_d> well im glad it was just the compiler, and not the hardware or somethting 2013-02-09T05:09:47 < dirty_d> i guess the function call re-ordered things the way they shouda been 2013-02-09T05:09:59 < Erlkoenig> yeah possible... check the disassembly ;) 2013-02-09T05:10:11 < dirty_d> i forgot how to do that 2013-02-09T05:10:14 < dirty_d> objdump? 2013-02-09T05:10:19 < Erlkoenig> objdump -d foo.elf 2013-02-09T05:10:28 < Erlkoenig> veeeeeeeeeery useful 2013-02-09T05:10:41 < Erlkoenig> i gotta go now... 04:00am, and lan-party today :> 2013-02-09T05:10:56 < dirty_d> ok, later 2013-02-09T05:10:59 < dirty_d> thanks 2013-02-09T05:11:06 < Erlkoenig> bye :) 2013-02-09T05:12:00 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-150.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-09T05:19:59 < dirty_d> whoa, what the hell? 2013-02-09T05:20:53 < dirty_d> what the hell is going on with this disassembly? http://pastebin.com/7JcbPmF9 2013-02-09T05:21:46 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-09T05:22:43 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 260 seconds] 2013-02-09T05:29:21 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-09T05:51:26 < dongs> i2c works 2013-02-09T05:53:06 < R2COM> thats quite an achievement 2013-02-09T05:53:35 < dirty_d> tadaaa 2013-02-09T05:53:43 < dongs> just copypasted my old driver that worked :) 2013-02-09T05:53:58 < R2COM> I did that too today, copypasted i2c code for lsm303 2013-02-09T05:54:03 < R2COM> work right out of the box 2013-02-09T06:05:19 < dongs> ooO led driver has a temperature sensor S:D 2013-02-09T06:18:34 -!- BJFreeman [~bjfree@66.sub-75-196-116.myvzw.com] has quit [Quit: had a good time] 2013-02-09T06:21:06 < emeb_mac> nifty! 2013-02-09T06:21:54 < dirty_d> wait a freeakin minute 2013-02-09T06:22:06 < dirty_d> where does the ISR epilog code come from etc? 2013-02-09T06:22:18 < dirty_d> do i need a compiler attribute 2013-02-09T06:22:25 < dirty_d> or do i have to code it in asm? 2013-02-09T06:22:25 < dongs> no 2013-02-09T06:22:28 < dongs> haha no 2013-02-09T06:22:30 < dongs> this isnt arm7 trash 2013-02-09T06:22:57 < dirty_d> dongs, so i dont have to do anything? 2013-02-09T06:23:01 < dirty_d> how does it know what to do 2013-02-09T06:24:54 < dongs> you ahve the startup .asm 2013-02-09T06:24:57 < dongs> with vector table 2013-02-09T06:25:09 < dirty_d> yup, see that 2013-02-09T06:26:28 < dirty_d> dongs, how is the ISR not trashing registers though? 2013-02-09T06:26:39 < dirty_d> its not saving any that its using 2013-02-09T06:27:03 < dirty_d> it just does its business and returns 2013-02-09T06:27:17 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-09T06:29:43 < dongs> i use this thing called a "compiler" 2013-02-09T06:29:49 < dongs> if theres any saving to be done, it takes care of it 2013-02-09T06:30:01 < dongs> whatever it does i dont really give a shit 2013-02-09T06:32:49 < dirty_d> lol 2013-02-09T06:32:51 < dirty_d> its not though 2013-02-09T06:33:04 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-09T06:34:54 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-09T06:39:02 < dongs> backlight driver replies by i2c. woot 2013-02-09T06:58:19 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-09T06:58:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-09T06:58:42 < dongs> sigh.. why doesnt shit properly specify us/ms/whatever 2013-02-09T06:58:46 < dongs> "Wait 200ms" 2013-02-09T06:58:49 < dongs> microseconds? 2013-02-09T06:58:51 < dongs> milliseconds? 2013-02-09T06:59:38 < gxti> ? 2013-02-09T06:59:52 < R2COM> millisecnds 2013-02-09T07:00:02 < R2COM> us for microsecond 2013-02-09T07:00:07 < dongs> ok 2013-02-09T07:00:08 < dongs> thx 2013-02-09T07:00:16 < gxti> srsly? 2013-02-09T07:00:45 < dongs> yeaman 2013-02-09T07:01:07 < dongs> im ESL, unless it says "milliseconds" i dunno wat the fuck ms means micro or milli 2013-02-09T07:01:17 < gxti> ms is always milli 2013-02-09T07:01:35 < dongs> TIL i guess. 2013-02-09T07:02:47 < dongs> now need to calculate the voltage for this shit 2013-02-09T07:02:48 < dongs> lets see 2013-02-09T07:06:59 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Quit: leaving] 2013-02-09T07:26:59 < emeb_mac> dongs: handy table for M/k/m/u etc - http://en.wikipedia.org/wiki/Order_of_magnitude 2013-02-09T07:29:11 < Simon--> table near bottom of http://en.wikipedia.org/wiki/Order_of_operations is always handy too :) 2013-02-09T07:29:45 < dongs> reliable information? in MY wikipedia? 2013-02-09T07:31:13 < emeb_mac> inorite? 2013-02-09T07:31:32 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-09T07:33:52 < dongs> bl driver so far replying with stuff as expected 2013-02-09T07:34:03 < dongs> gonna try messing with current/voltage params 2013-02-09T07:34:09 < dongs> since it defaults to 24V 2013-02-09T07:35:08 < emeb_mac> coo 2013-02-09T07:47:33 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-09T07:48:08 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-09T07:48:28 < dongs> 0x00 = 10V 2013-02-09T07:48:28 < dongs> 0x1f = 40V 2013-02-09T07:48:30 < dongs> is this linear? 2013-02-09T07:49:29 < dongs> o wait its just 1v=1bit 2013-02-09T08:36:27 < dongs> wtf. voltage settings are totalyl fucked 2013-02-09T08:36:32 < dongs> set for 12V: outputting 40V 2013-02-09T08:36:33 < dongs> k bbl 2013-02-09T09:36:36 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-09T09:44:46 -!- rigid [~daniel@178-26-71-67-dynip.superkabel.de] has joined ##stm32 2013-02-09T09:46:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-09T10:02:57 < dongs> ohh, backlight is ON. 2013-02-09T10:03:33 < dongs> brightness works too. 2013-02-09T10:04:48 < dongs> display reconnecting each tiem i upload shit is a bit rage-inducing though 2013-02-09T10:04:50 < dongs> THANKS ZYP 2013-02-09T10:14:56 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-09T10:21:19 < Thorn> dongs: how do you use coos queues? i.e. how do you store actual data objects (as queues only keep pointers)? 2013-02-09T10:22:07 < dongs> in some block and pass pointers to them 2013-02-09T10:22:32 < dongs> http://i.imgur.com/4RZ1QzJ.jpg 2013-02-09T10:22:38 < Thorn> is that block protected by mutexts/critical sections/whatever? 2013-02-09T10:23:27 < dongs> why would it be? 2013-02-09T10:24:29 < Thorn> it's not a ring buffer or something similar? 2013-02-09T10:25:21 < dongs> be specific, dude. maybe youre just doing it wrong. 2013-02-09T10:27:09 < Thorn> well. a queue supports 2 operations: insert and remove. say you have a CAN RX interrupt where you read a new message and insert it into a queue, and a task where you remove messages from it 2013-02-09T10:27:48 < Thorn> so how do you manage memory for these messages? 2013-02-09T10:30:48 < Thorn> given that the CAN interrupt can happen while you're removing a message from the queue 2013-02-09T10:31:19 < dongs> why dont youlook at sores and check 2013-02-09T10:31:21 < dongs> daym bro 2013-02-09T10:31:27 < dongs> this isnt some closed sores rtos 2013-02-09T10:31:59 < Thorn> but the queue only holds pointers 2013-02-09T10:32:10 < dongs> thats all you need to hold in it, nig'ga 2013-02-09T10:32:25 < dongs> ok time to wash dishes bbl 2013-02-09T10:39:27 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32 2013-02-09T10:39:53 < baird> Drink Wiskey, don't breathe it.. :/ 2013-02-09T10:41:56 -!- drgreenthumb [~fletch@drupal.org/user/72475/view] has quit [Ping timeout: 248 seconds] 2013-02-09T11:02:18 < dongs> http://cdn.memegenerator.net/instances/400x/34596197.jpg 2013-02-09T11:11:40 < Thorn> dongs: can you show am\n example of how you use queues (adding an object and removing one)? 2013-02-09T11:12:28 < dongs> Thorn: would love to, but Im not using coos queues. i thought i did in one place but I checked and i didnt. 2013-02-09T11:14:06 < Thorn> okay I see 2013-02-09T11:16:05 < dongs> but i dont see what the issue is with passing a pointer to allocated block of memory (large block, enough to contain xx mesages), and passing pointers to offsets inside that block. 2013-02-09T11:18:07 < Thorn> that block will be essentially a ring buffer, no? you'll need to keep information on what part of it is free, and that information needs to be protected from concurrent updates 2013-02-09T11:20:58 < baird> dongs: newbs don't know about my commodore iec serial port. 2013-02-09T11:21:19 < dongs> Thorn: right 2013-02-09T11:21:29 < Thorn> why do you need an os queue then 2013-02-09T11:21:45 < Thorn> you will have already implemented one 2013-02-09T11:22:23 < Thorn> this is what I don't understand about pointer queues. freertos queues store objects by value 2013-02-09T11:24:22 < Thorn> btw someone said you can malloc() objects. I don't think you can call malloc() from an ISR in many rtoses 2013-02-09T11:33:42 -!- barthess [~barthess@5.100.214.38] has joined ##stm32 2013-02-09T11:49:00 -!- barthess [~barthess@5.100.214.38] has quit [Quit: Leaving.] 2013-02-09T11:52:28 < zippe1> thorn: the idea with pointer queues is that you have type-stable storage somewhere else 2013-02-09T11:52:43 < zippe1> Then you create a queue big enough for a pointer to every one; that becomes your free queue 2013-02-09T12:04:53 -!- barthess [~barthess@178.154.7.203] has joined ##stm32 2013-02-09T12:27:32 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-09T12:27:35 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-09T12:34:50 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-09T12:35:51 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 276 seconds] 2013-02-09T13:00:35 -!- daku is now known as DaKu 2013-02-09T14:17:54 -!- neuro_sys is now known as neuro_sys|off 2013-02-09T14:27:49 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-09T14:29:06 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-09T14:34:21 -!- barthess [~barthess@178.154.7.203] has quit [Quit: Leaving.] 2013-02-09T14:58:06 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-09T15:07:38 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-129.pptp.stw-bonn.de] has joined ##stm32 2013-02-09T15:20:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-09T15:22:31 -!- Erlkoenig [~erlkoenig@pptp-212-201-73-129.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-09T15:29:11 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds] 2013-02-09T15:30:10 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-09T15:36:26 -!- theAdib [~adib@dslb-088-074-154-140.pools.arcor-ip.net] has joined ##stm32 2013-02-09T15:36:36 < donigs> oh look feenode failing agian 2013-02-09T15:44:45 <+Steffanx> i think it's you 2013-02-09T15:45:27 < zyp> donigs, dp board looks nice :) 2013-02-09T15:45:37 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has joined ##stm32 2013-02-09T15:48:31 <+Steffanx> zlog 2013-02-09T15:48:31 < zlog> Steffanx: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2013-02-09.html 2013-02-09T15:49:36 <+Steffanx> of course zyp, has a stm32 on it 2013-02-09T15:49:41 < zyp> :p 2013-02-09T15:50:30 < zyp> donigs, when do you have time to assemble one for me? 2013-02-09T15:54:45 < donigs> zyp: anytime 2013-02-09T15:54:53 < donigs> im gonna make a panel of 3 next time 2013-02-09T15:55:06 < donigs> since another dude and another dude want em 2013-02-09T15:55:28 < zyp> great 2013-02-09T15:56:26 < zyp> I have the lcd on the way already 2013-02-09T15:56:38 < zyp> by now I almost suspect lcd will arrive first anyway 2013-02-09T15:57:34 < zyp> since it's already over two weeks since I ordered it 2013-02-09T16:02:00 <+Steffanx> donigs received his DVB-T things yet? 2013-02-09T16:03:27 < zyp> http://i.imgur.com/q1GQHmU.jpeg <- you mean this? 2013-02-09T16:03:30 < zyp> or what was that? 2013-02-09T16:03:45 <+Steffanx> no, not those 2013-02-09T16:03:54 <+Steffanx> the one he ( and i ) bought on eba 2013-02-09T16:03:55 <+Steffanx> y 2013-02-09T16:04:11 < zyp> ah, rtlsdr dongles? 2013-02-09T16:04:14 <+Steffanx> Yes 2013-02-09T16:04:48 <+Steffanx> rtlsdr, that was the word i was looking for :) 2013-02-09T16:14:50 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-09T16:24:41 < donigs> Steffanx: haha no 2013-02-09T16:26:51 < dongs> zyp: led driver gets a bit warm 2013-02-09T16:27:04 < dongs> probably within reason, i have no idea what its normal temp is supposed to be 2013-02-09T16:27:57 < zyp> do you have it set to adaptive voltage mode? 2013-02-09T16:28:03 < dongs> thats the default yes 2013-02-09T16:28:22 < zyp> tried measuring voltage on cathode side? 2013-02-09T16:28:31 < dongs> is that the gnd side? 2013-02-09T16:28:43 < zyp> yes 2013-02-09T16:28:44 < dongs> when i measure at diode, i get liek ~24V 2013-02-09T16:28:52 < dongs> i dunno how to measure the stuff when its sinking 2013-02-09T16:29:13 < zyp> just measure the voltage on the low side 2013-02-09T16:29:19 < dongs> low to gnd? 2013-02-09T16:29:22 < zyp> yes 2013-02-09T16:29:25 < dongs> ok lets see 2013-02-09T16:29:48 < zyp> power converted to heat in sink circuitry is low side voltage * current 2013-02-09T16:30:09 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 276 seconds] 2013-02-09T16:30:35 < dongs> but I dont have a "a" low side 2013-02-09T16:30:37 < dongs> i have 6 channels 2013-02-09T16:30:42 < zyp> and since voltage drop across chain is constant, lowering high side voltage would lower low side voltage, lowering sink power wastage 2013-02-09T16:30:51 < zyp> measure them all, one by one 2013-02-09T16:31:28 < zyp> they should be pretty equal 2013-02-09T16:31:38 < dongs> 4.12V 2013-02-09T16:31:42 < dongs> on one 2013-02-09T16:31:48 < dongs> its pretty hard to measure cuz my fluke probes are hueg 2013-02-09T16:32:11 < zyp> it's the lowest one that would dictate how low you could go on high side 2013-02-09T16:32:18 < dongs> 4.12 2013-02-09T16:32:20 < dongs> 4.17 2013-02-09T16:32:21 < dongs> 4.19 2013-02-09T16:32:26 < dongs> on 3 pins 2013-02-09T16:32:37 < dongs> the other oens are on inner layer, cant see 2013-02-09T16:32:57 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-09T16:33:09 < zyp> sounds like you easily could lower high side by a few volts 2013-02-09T16:33:43 < dongs> i thoguht adaptive starts at preset and goes to needed voltage automatically 2013-02-09T16:34:01 < zyp> yes, that's the idea 2013-02-09T16:34:35 < dongs> but it doesnt work? or 2013-02-09T16:34:38 < dongs> what does 4.12 on low side mean 2013-02-09T16:34:43 < dongs> high side that much too much? 2013-02-09T16:35:14 < zyp> well, you still need a margin on low side to be able to do current regulation at all 2013-02-09T16:35:25 < zyp> but 4V sounds like unnecesssarily much 2013-02-09T16:35:42 < zyp> try setting high side to 21V if that's possible 2013-02-09T16:35:57 < dongs> should be 2013-02-09T16:36:01 < dongs> i think im running on default, lets see 2013-02-09T16:36:13 < zyp> should reduce wasted power in sink to one fourth 2013-02-09T16:36:40 < zyp> so I would expect it to give a measureable reduction in temperature 2013-02-09T16:38:17 < dongs> for some reason 24 = 0b01111 2013-02-09T16:38:20 < dongs> for voltage 2013-02-09T16:38:28 < dongs> (default) 2013-02-09T16:38:36 < dongs> well, actually i dunno if its exactly 24 2013-02-09T16:38:44 < dongs> measured with scope last time. it was more like 24.8 or something 2013-02-09T16:38:58 < dongs> if its 1lsb/volt and it starts with 10 2013-02-09T16:39:15 < dongs> 0b01111 should be 26, no? 2013-02-09T16:39:39 < zyp> binary value represents 15 2013-02-09T16:39:40 < dongs> page 39 of LP8545 2013-02-09T16:40:00 < dongs> right, but 0 = 10V 2013-02-09T16:40:12 < zyp> should be 25 then 2013-02-09T16:40:21 < dongs> ok, and it waas 24.8 or so, close enough 2013-02-09T16:40:51 < dongs> so i set to 12 2013-02-09T16:40:58 < dongs> 0b01100 2013-02-09T16:41:44 < dongs> panel still bright 2013-02-09T16:41:58 < dongs> driver still hot. 2013-02-09T16:42:10 < zyp> it shouldn't affect brightness 2013-02-09T16:42:21 < zyp> since current is constant 2013-02-09T16:42:47 < dongs> lowside voltage 3.98 now 2013-02-09T16:42:50 < dongs> go lower? 2013-02-09T16:43:06 < zyp> uh, that was not a huge difference 2013-02-09T16:43:16 < zyp> did it actually change high side? 2013-02-09T16:43:59 < dongs> let see 2013-02-09T16:44:12 < dongs> 22.6 on highside now 2013-02-09T16:44:14 < dongs> hmmm 2013-02-09T16:44:22 < dongs> this is with 0xa 2013-02-09T16:44:24 < dongs> for voltage. 2013-02-09T16:44:44 < zyp> hang on, I'm checking something in the datasheet 2013-02-09T16:45:16 < zyp> what do you have DRV_HEADR set to? 2013-02-09T16:45:31 < zyp> and are you sure EN_ADAPT are set to 1? 2013-02-09T16:45:47 < dongs> http://bcas.tv/paste/results/gvm2Qj26.html 2013-02-09T16:45:59 < dongs> they're at defaults 2013-02-09T16:46:00 < dongs> from eeeprom 2013-02-09T16:46:07 < dongs> there's a defaults list in code i just pasted 2013-02-09T16:46:09 < dongs> yes, adapt is on 2013-02-09T16:46:12 < zyp> DRV_HEADR only has a range of 0.3-1.2V 2013-02-09T16:46:13 < dongs> and headr is deafult, 10 i think 2013-02-09T16:46:24 < zyp> 010? 2013-02-09T16:46:44 < zyp> then you should see around 0.5V on the lowest chain 2013-02-09T16:47:11 < dongs> drv_headr is 000 2013-02-09T16:47:20 < zyp> maybe one of the chains that you can't measure has a significantly higher voltage drop 2013-02-09T16:47:24 < zyp> that's even higher 2013-02-09T16:47:30 < zyp> I mean, lower 2013-02-09T16:49:09 < dongs> acutally doesnt amtter what i set start voltage 2013-02-09T16:49:10 < dongs> to 2013-02-09T16:49:15 < dongs> it always stays around 22.x 2013-02-09T16:49:21 < dongs> becaues i guess thats what adapt shit means. 2013-02-09T16:49:28 < zyp> yes 2013-02-09T16:49:35 < zyp> hmm 2013-02-09T16:49:59 < zyp> wonder if that is related to you connecting strings in parallel 2013-02-09T16:51:41 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T16:51:56 < zyp> what happens if you disable adaptive mode and force it lower? 2013-02-09T16:52:05 < dongs> havent tried 2013-02-09T16:52:11 < zyp> does the brightness change? 2013-02-09T16:52:23 < dongs> i mean the temp doesnt bother me, its not hurt to hold finger on it for a while 2013-02-09T16:52:27 < dongs> and it doesnt solder itself off the board 2013-02-09T16:52:40 < dongs> its hard to tell if shit c hanges or not because each flash reupload I get display disconnect 2013-02-09T16:52:44 < dongs> :) 2013-02-09T16:52:47 < dongs> because dp_power is on a gpio 2013-02-09T16:52:56 < dongs> so i get a black screen for a bit 2013-02-09T16:52:59 < dongs> so i cant quickly do a/b 2013-02-09T16:53:00 < zyp> heh 2013-02-09T16:53:03 < dongs> unless I write some test code using buttons 2013-02-09T16:53:19 < dongs> or do some protocol over usb/hid/serial or wahtever to change shit 2013-02-09T16:53:22 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T16:53:23 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has quit [Changing host] 2013-02-09T16:53:23 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-09T16:55:09 < dongs> actually at higher brightness chip almost doesnt heatt at all 2013-02-09T16:55:13 < dongs> at LOWER brightness it does 2013-02-09T16:55:15 < dongs> makes sense?! 2013-02-09T16:55:35 < zyp> uh 2013-02-09T16:55:35 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has quit [Ping timeout: 252 seconds] 2013-02-09T16:55:49 < zyp> oh well 2013-02-09T16:56:08 < zyp> by the way, are you able to measure the current draw on vbus? 2013-02-09T16:56:17 < dongs> i could 2013-02-09T16:56:28 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Ping timeout: 256 seconds] 2013-02-09T16:56:29 < zyp> just curious whether we are staying under 500mA 2013-02-09T16:56:34 < dongs> pretty sure. 2013-02-09T16:56:42 < dongs> but lemme see if ican find my cut usb cable for current measuremetn 2013-02-09T16:57:12 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-09T16:59:39 < Thorn> wtf. extern int i3 = 3; is allowed in C 2013-02-09T17:00:26 < Thorn> static int i5; extern int i5; is also allowed 2013-02-09T17:01:17 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer] 2013-02-09T17:01:46 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T17:01:46 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has quit [Changing host] 2013-02-09T17:01:46 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-09T17:02:04 < dongs> jesus 2013-02-09T17:02:06 < dongs> 1 fucking amp 2013-02-09T17:02:07 < dongs> lol 2013-02-09T17:02:45 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T17:02:53 < dongs> i wonder if its jus tb ecause im driving the backlight wrong. 2013-02-09T17:03:43 < dirty_d> Thorn, so 2013-02-09T17:03:43 < dongs> ok 2013-02-09T17:03:46 < zyp> how much if you turn off lcd and only run backlight? 2013-02-09T17:03:48 < dongs> 300mA wihtout backlight 2013-02-09T17:03:53 < zyp> ah 2013-02-09T17:03:56 < dongs> for logic 2013-02-09T17:04:04 < dongs> hmm.. thats more than i thought 2013-02-09T17:04:21 < zyp> yeah, I thought you said 130mA for lcd 2013-02-09T17:04:32 < zyp> maybe stm draws a bunch? 2013-02-09T17:04:40 < dongs> pfft 2013-02-09T17:04:41 < dongs> why would it 2013-02-09T17:04:47 < dongs> iots not doing anyhtign 2013-02-09T17:05:01 < zyp> what happens to draw when you unplug dp? 2013-02-09T17:05:10 < dongs> lesee 2013-02-09T17:05:45 < dongs> drop to 280 2013-02-09T17:06:00 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has quit [Ping timeout: 244 seconds] 2013-02-09T17:06:01 < zyp> what about turning off lcd reg then? 2013-02-09T17:06:11 < dongs> right 2013-02-09T17:06:20 < zyp> (sounds like having that option is useful after all) 2013-02-09T17:06:40 < dongs> 30mA 2013-02-09T17:06:51 < dirty_d> zyp, have you ever had any problems with ISRs not working iff you compile with -O2? 2013-02-09T17:07:21 < dongs> wesll 2013-02-09T17:07:23 < zyp> right, so lcd consumes 270mA @5V, backlight consumes 700mA and rest consumes 30mA 2013-02-09T17:07:25 < dirty_d> I have the code and dissassembly of both -O0 and -O2 here, http://pastebin.com/tF3C3GqL http://pastebin.com/tnqUyGa5 2013-02-09T17:07:27 < emeb_mac> ~30ma just for STM32 sounds about right - that's what I've seen before. 2013-02-09T17:08:03 < dongs> http://i.imgur.com/oETpGLF.png 2013-02-09T17:08:48 < dongs> trryign to remember which ass did I pull 130ma from 2013-02-09T17:09:00 < emeb_mac> more than one! 2013-02-09T17:09:15 < emeb_mac> dongs is assularly gifted 2013-02-09T17:09:59 < zyp> oh well, sounds like we're just going to violate the usb power limits 2013-02-09T17:10:09 < zyp> like half of the other devices out there 2013-02-09T17:10:13 < zyp> :p 2013-02-09T17:10:27 < zyp> hmm 2013-02-09T17:10:51 < zyp> wonder if it's possible to run FS on an usb3 port and still negotiate for 900mA 2013-02-09T17:10:52 < dongs> i geuss if that png says 320mA avg with 280mA least case, thten its probably pretty close 2013-02-09T17:11:06 < dongs> why? usb3 allows higher curent? 2013-02-09T17:11:17 < zyp> yes, 900mA max 2013-02-09T17:11:18 < dongs> i thought power negotiaition was like 1 byte 2013-02-09T17:11:22 < dongs> and 0xff = 500mA 2013-02-09T17:11:29 < dongs> did they extend something? 2013-02-09T17:11:53 < dongs> or am i missing descriptor with actual negotiation 2013-02-09T17:11:57 < dongs> missing = mixing 2013-02-09T17:12:57 < zyp> it's a byte, in usb2 lsb is 2mA 2013-02-09T17:13:21 < zyp> I forget how they changed it in usb3, but they did 2013-02-09T17:13:26 < dongs> ok 2013-02-09T17:15:43 < donigs> oh well, stuff works hardwar-wise. 2013-02-09T17:15:46 < donigs> software can always be fixed 2013-02-09T17:15:51 < zyp> yeah 2013-02-09T17:15:58 < zyp> looks nice either way 2013-02-09T17:16:09 < donigs> even rgb led works. 2013-02-09T17:16:14 < donigs> if you wanna draw even more current. 2013-02-09T17:16:18 < zyp> :p 2013-02-09T17:16:24 < donigs> hmm 2013-02-09T17:16:26 < donigs> 890mA now. 2013-02-09T17:16:27 < emeb_mac> why can't I see the imgur pic @ the link in the log? 2013-02-09T17:16:29 < donigs> nothing changed. 2013-02-09T17:16:57 < donigs> haha 2013-02-09T17:16:59 < donigs> current draw changes 2013-02-09T17:17:02 < donigs> based on how much white tehre is 2013-02-09T17:17:05 < donigs> like 10-20mA+- 2013-02-09T17:17:10 < donigs> i can move windows around 2013-02-09T17:17:13 < zyp> ah, I checked usb3 spec 2013-02-09T17:17:16 < donigs> and current draw goes 2013-02-09T17:17:19 < zyp> 900mA is only in SS mode 2013-02-09T17:17:20 < zyp> «Expressed in 2-mA units when the device is operating in high-speed mode and in 8-mA units when operating in SuperSpeed mode. 2013-02-09T17:17:24 < zyp> (i.e., 50 = 100 mA in high-speed mode and 50 = 400 mA in SuperSpeed mode).» 2013-02-09T17:17:28 < donigs> ah 2013-02-09T17:18:00 < gxti> what will usb 4 be called, ludicrous speed? 2013-02-09T17:18:10 -!- Erlkoenig [~erlkoenig@minecraft.stw-bonn.de] has joined ##stm32 2013-02-09T17:18:10 < zyp> suspect that an usb3 device aren't even allowed to operate in FS 2013-02-09T17:18:15 < emeb_mac> obligatory spaceballs ref 2013-02-09T17:18:29 -!- Erlkoenig [~erlkoenig@minecraft.stw-bonn.de] has quit [Client Quit] 2013-02-09T17:19:06 < emeb_mac> so unless you're running superspeed, no high current for you! 2013-02-09T17:19:11 < donigs> emeb_mac: http://i.imgur.com/TPB91F5.jpg this ? 2013-02-09T17:19:32 < zyp> works for me 2013-02-09T17:19:41 < donigs> http://i.imgur.com/U9UGXhx.jpg or this 2013-02-09T17:19:49 < emeb_mac> nah - this: http://i.imgur.com/4RZ1QzJ.jpg 2013-02-09T17:19:55 -!- tavish [~tavish@120.56.137.217] has joined ##stm32 2013-02-09T17:19:55 -!- tavish [~tavish@120.56.137.217] has quit [Changing host] 2013-02-09T17:19:55 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-09T17:20:03 < zyp> oh, I missed the latter one 2013-02-09T17:20:08 < Thorn> what is that board? 2013-02-09T17:20:18 < Thorn> usb/lvds display interface? 2013-02-09T17:20:21 < donigs> Thorn: its the board we've been discussing for hte last 30 minutes 2013-02-09T17:20:51 < emeb_mac> 2nd one works. that's what I thot was in the log. 2013-02-09T17:21:45 < gxti> donigs: discuss yes, explain anything about whatsoever so that people who don't already know what it is might find out, no 2013-02-09T17:21:49 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has joined ##stm32 2013-02-09T17:22:10 < emeb_mac> it's all context. 2013-02-09T17:22:33 < gxti> so, you know, a one-sentence summary would be swell 2013-02-09T17:22:51 < dongs> its a displayport>eDP + backlight driver board. 2013-02-09T17:23:06 < donigs> and stm32 and accelerometer for hte lulz. 2013-02-09T17:23:15 < gxti> and RGB LED for the lulz too? 2013-02-09T17:23:19 < donigs> yep 2013-02-09T17:23:21 < gxti> kk 2013-02-09T17:23:28 < emeb_mac> for a cheap fruitco-style LCD panel 2013-02-09T17:23:33 < gxti> looks nice 2013-02-09T17:23:43 < donigs> it works, which is more important than nice 2013-02-09T17:24:04 < emeb_mac> even if it draws more current than the USB FS/HS spec allows... 2013-02-09T17:25:32 < emeb_mac> donigs: what are you doing for an enclosure? 2013-02-09T17:25:38 < zyp> hmm, I just noticed that white usb cable looks awfully similar to the one I have here 2013-02-09T17:25:41 < zyp> identical even 2013-02-09T17:25:50 < donigs> its a canon usb cable 2013-02-09T17:25:51 < emeb_mac> :P 2013-02-09T17:25:53 < donigs> from t heir cam 2013-02-09T17:25:58 < zyp> donigs, that's what I thought 2013-02-09T17:26:07 < zyp> because that's where I got mine from too :p 2013-02-09T17:26:15 -!- barthess [~barthess@178.154.25.106] has joined ##stm32 2013-02-09T17:27:14 < donigs> maybe a G7 or G9 or so 2013-02-09T17:27:17 < donigs> unsure which cam 2013-02-09T17:27:21 < donigs> i had it for quite a while 2013-02-09T17:27:48 < zyp> donigs, EE pal of mine is laughing of your silk labels for fiducials 2013-02-09T17:28:03 < donigs> too lazy to unmark them 2013-02-09T17:28:15 < donigs> usually i dont put refdes on silk at all. 2013-02-09T17:28:25 < zyp> mine came with the 30D I bought five years ago 2013-02-09T17:29:30 -!- neuro_sys|off is now known as neuro_sys 2013-02-09T17:29:41 < donigs> btw, whats so funny about teh fiducial other than the marking? 2013-02-09T17:29:46 < donigs> my dicknplace actually uses those 2013-02-09T17:30:58 < zyp> huh? 2013-02-09T17:31:14 < dongs> the fiducial mark 2013-02-09T17:31:17 < zyp> nothing funny with the fiducials themselves 2013-02-09T17:31:21 < zyp> just the label :p 2013-02-09T17:31:26 < dongs> pfft 2013-02-09T17:38:53 -!- rigid [~daniel@178-26-71-67-dynip.superkabel.de] has quit [Ping timeout: 240 seconds] 2013-02-09T17:51:47 < emeb_mac> zyp: just what you need - coding hardware designs in python - http://milkymist.org/3/migen-tutorial.pdf 2013-02-09T17:52:24 < Thorn> there was some haskell project too 2013-02-09T17:52:30 < zyp> I think I looked at something similar some years ago 2013-02-09T17:52:33 < zyp> possibly the same 2013-02-09T17:53:09 < dirty_d> wtf, http://pastebin.com/Br4URN8J 2013-02-09T17:53:25 < dirty_d> how can eor.w r1, r0, #8192 be the first instruction? 2013-02-09T17:53:30 < dirty_d> nothing has been loaded into r0 2013-02-09T17:54:34 < zyp> are you sure it is the first instruction? 2013-02-09T17:54:57 < dirty_d> yea, look at the disassembly 2013-02-09T17:55:04 < zyp> can I have a copy of the elf? 2013-02-09T17:55:13 < dirty_d> sure 2013-02-09T17:56:05 < Thorn> looks to me like there has to be a ldr before that which is missing 2013-02-09T17:56:32 < dirty_d> haha, damn my work must have lost power, cant scp a file over 2013-02-09T17:56:50 < zyp> yeah, I'm sure it's just a case of misinterpreting objdump -S output 2013-02-09T17:56:57 < dirty_d> maybe 2013-02-09T17:57:02 < dirty_d> where can i put a file? 2013-02-09T17:57:06 < zyp> dump.jvnv.net 2013-02-09T17:57:08 < zyp> no 2013-02-09T17:57:11 < zyp> bin.jvnv.net 2013-02-09T17:57:39 < dirty_d> http://bin.jvnv.net/f/9eHa2/main.elf 2013-02-09T17:58:32 < dirty_d> wait did i send the right one? 2013-02-09T17:58:39 < zyp> yes 2013-02-09T17:59:01 < zyp> http://paste.jvnv.net/view/kRBfb <- you missed the first six instructions 2013-02-09T17:59:15 < dirty_d> i did 2013-02-09T17:59:23 < dirty_d> i missed a bunch of code in the disassembly 2013-02-09T17:59:25 < dirty_d> oops 2013-02-09T17:59:37 < emeb_mac> http://www.impactlab.net/2013/02/08/the-right-of-the-american-people-to-keep-and-fly-drones/ 2013-02-09T18:00:00 < dirty_d> zyp, do you see any reason that the isr would not work when compiled like that? 2013-02-09T18:00:19 < zyp> not obviously so, why? 2013-02-09T18:00:48 < dirty_d> zyp, it doesnt work when compiled with -O2 2013-02-09T18:00:51 < dirty_d> -O0 is fine 2013-02-09T18:00:59 < zyp> this looks like -O2 2013-02-09T18:01:12 < dirty_d> this is 2013-02-09T18:01:26 < dirty_d> this is -O0 http://pastebin.com/zcNsQL7F 2013-02-09T18:02:47 < dirty_d> all the ISR does is toggle an LED every second 2013-02-09T18:03:10 < dirty_d> with -O2 it never turns on, if i change the toggle to just turn it on, it does turn on 2013-02-09T18:05:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-09T18:06:05 < zyp> I read through the O2 assembly, it looks right to me 2013-02-09T18:06:28 < dirty_d> hmm 2013-02-09T18:06:54 -!- DaKu is now known as daku 2013-02-09T18:07:53 < dirty_d> zyp, if i swap those two c statements around it works, ie clear the timer interrupt flag first 2013-02-09T18:08:29 < dirty_d> http://paste.jvnv.net/view/VTvwS 2013-02-09T18:08:42 < zyp> ah, I know what your issue is 2013-02-09T18:08:51 < zyp> the ISR is entered twixe 2013-02-09T18:08:59 < Thorn> why uxth r0, r1 ? 2013-02-09T18:09:06 < dirty_d> zyp, is it? 2013-02-09T18:09:25 < zyp> Thorn, because ODR is defined as a 16-bit field, so the upper 16 bits are cleared 2013-02-09T18:09:31 < zyp> dirty_d, yes 2013-02-09T18:09:38 < zyp> 80001a8: 6118 str r0, [r3, #16] 2013-02-09T18:09:39 < dirty_d> why is that happening? 2013-02-09T18:09:39 < zyp> 80001aa: 4770 bx lr 2013-02-09T18:09:43 < Thorn> no I meant why r0. looks correct 2013-02-09T18:10:00 < zyp> note how the store happens right before the interrupt return 2013-02-09T18:10:12 < dirty_d> yea 2013-02-09T18:10:23 < zyp> that store is the one writing back the cleared interrupt flag 2013-02-09T18:10:29 < dirty_d> ok 2013-02-09T18:10:46 < dirty_d> do i need dmb? 2013-02-09T18:11:11 < zyp> there is a latency from that write, until NVIC sees the interrupt state input getting cleared 2013-02-09T18:11:28 < zyp> so when you return, NVIC still sees interrupt as asserted, so ISR is entered again 2013-02-09T18:11:33 < dirty_d> ahh 2013-02-09T18:11:40 < zyp> so leds turn on, then immediately off again 2013-02-09T18:11:42 < dirty_d> this stuff can bite you in the ass huh? 2013-02-09T18:11:49 < zyp> :) 2013-02-09T18:11:58 < zyp> and yes, a dmb should do the trick 2013-02-09T18:12:22 < Thorn> I always clear interrupt flags at the very start of an isr 2013-02-09T18:12:37 < dirty_d> zyp, but im guessing it would be more efficent to clear the flag at the start of the isr> 2013-02-09T18:12:39 < Thorn> looks like this rule does have a meaning 2013-02-09T18:12:46 < dirty_d> since i dont have to add any more instructions that way? 2013-02-09T18:13:18 < Thorn> that's what I do 2013-02-09T18:13:58 < zyp> dirty_d, true 2013-02-09T18:14:28 < dirty_d> ok 2013-02-09T18:14:47 < Thorn> so what makes the difference are the 3 instructions between str and bx lr? nice 2013-02-09T18:14:52 < dirty_d> zyp, is there a definite number of cycles you need to wait until hat write actually happens? 2013-02-09T18:14:56 < Thorn> between -O2 and -O0 2013-02-09T18:15:33 < zyp> O0 has a mov and a pop in between 2013-02-09T18:15:50 < Thorn> yes 2 instructions 2013-02-09T18:16:05 < zyp> but not necessarily only two cycles ;) 2013-02-09T18:16:32 < Thorn> and one bus transaction (pop) which is what probably matters 2013-02-09T18:16:39 < dirty_d> hmm 2013-02-09T18:16:51 < dirty_d> i cnt believe people are actually trying to drive right now 2013-02-09T18:16:58 < dirty_d> the plow was just stuck in the road, lol 2013-02-09T18:17:13 < zyp> you guys not used to snow? 2013-02-09T18:17:29 < dirty_d> yea, just not this much all at once 2013-02-09T18:18:12 < dirty_d> looks like almost 3 feet in the street 2013-02-09T18:18:14 < zyp> :) 2013-02-09T18:18:42 < zyp> I grew up north of the arctic circle :) 2013-02-09T18:18:56 < dirty_d> how the hell do people get around when it snows like that? 2013-02-09T18:18:59 < dirty_d> more plows etc? 2013-02-09T18:19:09 <+Steffanx> dirty_d i still want you to give me some of that snow 2013-02-09T18:19:22 < dirty_d> take it all, i hate snow 2013-02-09T18:19:37 < zyp> well, we didn't really have problems with too much snow on the ground 2013-02-09T18:19:38 <+Steffanx> i dont :D 2013-02-09T18:19:45 <+Steffanx> We do, I don't 2013-02-09T18:19:51 < zyp> it was always too windy, so everything was blowing around instead 2013-02-09T18:19:54 <+Steffanx> *we = everyone but me :D 2013-02-09T18:19:56 < dirty_d> why skiing or something? 2013-02-09T18:20:00 -!- TeknoJucer [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T18:20:03 < dirty_d> zyp, ahh 2013-02-09T18:20:34 < dirty_d> the streets are especially bad because it was so window, there is no snow on any roofs 2013-02-09T18:20:39 < dirty_d> windy* 2013-02-09T18:20:43 <+Steffanx> Time to move dirty_d 2013-02-09T18:20:43 < dirty_d> it all collected in the streets 2013-02-09T18:20:50 <+Steffanx> Go florida or so :P 2013-02-09T18:20:52 <+Steffanx> *to 2013-02-09T18:20:58 < dirty_d> its very densely populated 2013-02-09T18:21:02 <+Steffanx> oh, no that gives you other crap 2013-02-09T18:21:11 < dirty_d> alligators, lol 2013-02-09T18:21:37 <+Steffanx> Nowhere in yankeeland one is safe 2013-02-09T18:21:45 < BrainDamage> when nerds are bored: traceroute -m 100 216.81.59.173 2013-02-09T18:21:54 < zyp> just did that 2013-02-09T18:21:56 < zyp> :D 2013-02-09T18:21:58 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-09T18:22:14 <+Steffanx> what is that BrainDamage ? 2013-02-09T18:22:22 < zyp> Steffanx, try it and you'll see 2013-02-09T18:22:42 <+Steffanx> not much seems to happen there on the other side 2013-02-09T18:22:56 < dirty_d> lol 2013-02-09T18:22:59 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Ping timeout: 244 seconds] 2013-02-09T18:23:34 < emeb_mac> heehee 2013-02-09T18:23:47 <+Steffanx> Hmm, nothing happens in my terminal 2013-02-09T18:23:59 <+Steffanx> but the network tools from os x does show it 2013-02-09T18:24:19 < BrainDamage> you need to get past ip 10.26.26.102 2013-02-09T18:24:47 < zyp> I don't have that one here 2013-02-09T18:24:58 <+Steffanx> http://share.naffets.nl/a-20130209-172447.png 2013-02-09T18:25:21 < zyp> heh 2013-02-09T18:25:27 < zyp> http://paste.jvnv.net/view/b8XnW <- it's supposed to look like this :p 2013-02-09T18:25:41 <+Steffanx> Yes, it does do something like that in the netweork tools 2013-02-09T18:26:26 <+Steffanx> someone was bored 2013-02-09T18:27:59 < Thorn> someone has way too many IP addresses 2013-02-09T18:28:12 < dongs> http://www.venom-group.com/Venom-Medion-Dual-10-Amp-Charger-with-Audio-Speakers.html wat 2013-02-09T18:28:37 < Thorn> CCIE.38168 - is that his certificate #? 2013-02-09T18:28:43 <+Steffanx> ok dongs :S 2013-02-09T18:28:52 < dirty_d> its gonna be bad when they plow, http://i.imgur.com/OnDmBb5l.jpg 2013-02-09T18:28:53 < Thorn> are there so many CCIEs out there? 2013-02-09T18:28:53 < zyp> Thorn, ip range is owned by Epik Networks, they have 4096 of them 2013-02-09T18:29:16 < emeb_mac> dongs: it's a dessert topping, no it's a floor wax, no it's both. 2013-02-09T18:30:54 < emeb_mac> http://www.cnx-software.com/2013/02/09/299-microsemi-smartfusion2-starter-kit-cortex-m3-fpga/ 2013-02-09T18:31:02 < dongs> i have that 2013-02-09T18:31:11 < dongs> well, not THAT 2013-02-09T18:31:15 < dongs> but the actel version of it 2013-02-09T18:31:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-09T18:31:53 < zyp> I think somebody said their toolchain are worse than xilinx' 2013-02-09T18:32:00 < dongs> they are 2013-02-09T18:32:01 < zyp> that's when I lost interest for it. 2013-02-09T18:32:01 < dongs> by far 2013-02-09T18:32:11 < dongs> xilinx is godly compared to actel tools 2013-02-09T18:32:13 < dongs> mostly because they didnt write them 2013-02-09T18:32:21 < dongs> but instead bvought/licensed/freetarded different versions 2013-02-09T18:32:28 < dongs> so none of the shit properly works together 2013-02-09T18:32:35 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T18:32:55 < dongs> i forget waht actel uses for fpga compiler.. some generic shit 2013-02-09T18:32:58 < gxti> i wish xilinx would do something about the size though 2013-02-09T18:33:12 < dongs> might evne be synoposys stuff 2013-02-09T18:33:36 < zyp> gxti, why? are you installing it from floppies? 2013-02-09T18:33:48 < Thorn> lattice bundles synplify pro and active-hdl with their diamond package 2013-02-09T18:33:49 < dongs> hes installing into the cloud 2013-02-09T18:33:56 < gxti> zyp: no, floppies are expensive. i use punch cards. 2013-02-09T18:34:17 < gxti> the download grows significantly every year but the set of features i use does not 2013-02-09T18:34:30 < dongs> they actually removed the only feature I used in it 2013-02-09T18:34:34 < dongs> since version 10 or 11 2013-02-09T18:34:34 < zyp> we're in 2013, 6GB is not much 2013-02-09T18:34:36 < emeb_mac> stop upgrading 2013-02-09T18:34:47 <+Steffanx> what zyp said :D 2013-02-09T18:34:47 < dongs> the testbench editor shit 2013-02-09T18:34:48 < gxti> i don't upgrade it. i just lose it when i move machines or something. 2013-02-09T18:34:53 < dongs> where you could draw graphs of testbecnehs 2013-02-09T18:34:55 < dongs> was trashed 2013-02-09T18:35:01 < dongs> and now you have to "write" them in HDL 2013-02-09T18:35:07 <+Steffanx> poor you 2013-02-09T18:35:24 < zyp> even on a SSD, 6GB is not considerably expensive nowadays 2013-02-09T18:35:37 <+Steffanx> 6$ or so? :) 2013-02-09T18:35:38 -!- TeknoJucer [~TeknkoJuc@69.158.165.12] has quit [Ping timeout: 256 seconds] 2013-02-09T18:35:48 < emeb_mac> I installed, removed stuff I didn't need and cut it down to 2GB 2013-02-09T18:36:18 < zyp> of course, xilinx toolchain includes a full arm toolchain and everything, for zynq 2013-02-09T18:36:27 -!- _BJFreeman [~bjfree@67.sub-75-244-137.myvzw.com] has joined ##stm32 2013-02-09T18:36:34 < zyp> probably one for microblaze aswell 2013-02-09T18:36:38 < emeb_mac> ya 2013-02-09T18:36:53 -!- _BJFreeman is now known as BJFreeman 2013-02-09T18:36:55 < dongs> i wonder if dudes at xilinx who made microblaze were stoned 2013-02-09T18:37:01 * emeb_mac is still running ISE 12.4 2013-02-09T18:37:06 < dongs> "still"? 2013-02-09T18:37:13 < dongs> i'm still running whichever one had testbench graphs. 2013-02-09T18:37:17 <+Steffanx> bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb 2013-02-09T18:37:18 <+Steffanx> oops 2013-02-09T18:37:21 < dongs> pretty sure that was either 9 or 10 2013-02-09T18:37:21 < emeb_mac> considering they're up to 14-xomething 2013-02-09T18:37:34 < zyp> I think I've got ISE 13 on my old windows laptop, that I dug up the other day 2013-02-09T18:37:39 < emeb_mac> cat-like typing detected 2013-02-09T18:37:53 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-09T18:37:54 <+Steffanx> No, was messing around with my gamerish new mouse 2013-02-09T18:38:33 -!- BrainDamage1 [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-09T18:39:06 < zyp> hmm, 13.4 2013-02-09T18:39:33 <+Steffanx> My penis is bigger. 2013-02-09T18:39:40 < emeb_mac> yay? 2013-02-09T18:39:41 <+Steffanx> Oh fuck that not what this discussion was about 2013-02-09T18:39:53 < zyp> Steffanx, I'm not talking about mm 2013-02-09T18:40:03 <+Steffanx> It was about who has the oldest xilinx stuff 2013-02-09T18:40:16 < emeb_mac> dongs wins 2013-02-09T18:40:19 < BrainDamage1> in that case, apparently, the smaller number is better 2013-02-09T18:40:34 < zyp> I remember I had to pirate xilinx stuff once 2013-02-09T18:40:46 < emeb_mac> you a bad man 2013-02-09T18:40:48 < zyp> because free version refused to run on 64-bit windows 2013-02-09T18:40:50 < dongs> my ledguy pirated 14.someshit recently for chipscope 2013-02-09T18:41:03 < emeb_mac> chipscope rulz 2013-02-09T18:41:13 < dirty_d> can you drive these directly from a GPIO pin? http://www.aliexpress.com/store/product/SSR-40-DA-Solid-State-Relay-40A-Output-24-380VAC-1PCS/602458_344627592.html 2013-02-09T18:41:15 < zyp> and I'm not setting up 32-bit windows just for bullshit like that 2013-02-09T18:41:21 < dirty_d> with a resistor inbetween maybe? 2013-02-09T18:41:28 < zyp> sure you can 2013-02-09T18:41:29 < dongs> no 2013-02-09T18:41:39 < dongs> o wait, ssr? yeah probbly 2013-02-09T18:41:39 < dirty_d> it says 3-32V 3-25mA drive current 2013-02-09T18:41:50 < dongs> i qA EWsinf ir bX KQWe 2013-02-09T18:41:52 < dongs> errrrrrrrrr 2013-02-09T18:41:55 < dongs> whoa, nice typing 2013-02-09T18:42:02 < emeb_mac> moonman lingo 2013-02-09T18:42:04 < dongs> i was reading it backwards and I thought it was one of those arduino 5V relays 2013-02-09T18:42:05 < dirty_d> i dont speak klingon 2013-02-09T18:42:07 < dirty_d> yet... 2013-02-09T18:42:24 < zyp> I'm using a SSR to turn on the speakers by my tv 2013-02-09T18:42:26 -!- BrainDamage [~BrainDama@ip-46-232-150-138.static.adsl.cheapnet.it] has quit [Ping timeout: 256 seconds] 2013-02-09T18:42:28 < gxti> it looks like you don't need any resistor 2013-02-09T18:42:49 < zyp> just hooked it to vbus in an usb cable and plugged it into the otherwise useless service port on the tv 2013-02-09T18:42:54 < zyp> so speakers comes on with the tv 2013-02-09T18:43:04 < dirty_d> is it basically saying, if you provide 3V it will sink 3mA? 2013-02-09T18:43:13 < dirty_d> and 32V 25mA? 2013-02-09T18:43:25 < gxti> probably but not necessarily 2013-02-09T18:43:37 < zyp> probably 2013-02-09T18:44:01 < zyp> it's probably driving the led in an optocoupler through a resistor 2013-02-09T18:44:06 < Laurenceb_> so did dongs get his display thingy to work? 2013-02-09T18:44:10 < dongs> yes 2013-02-09T18:44:13 < zyp> so higher voltage just gives a higher led current 2013-02-09T18:44:13 < Laurenceb_> nice 2013-02-09T18:44:24 < dongs> http://i.imgur.com/TPB91F5.jpg + http://i.imgur.com/U9UGXhx.jpg 2013-02-09T18:44:32 < gxti> there's a little diagram with a "control circuit" (probably a dorito chip) and resistors in series with LEDs 2013-02-09T18:44:44 < dirty_d> ohh yea 2013-02-09T18:44:58 < Laurenceb_> whats ic5 for? 2013-02-09T18:45:06 < dongs> ha ha 2013-02-09T18:45:14 < dongs> 14bit accelerometer. 2013-02-09T18:45:17 < zyp> Laurenceb_, for detecting when you drop it in the floor 2013-02-09T18:45:27 < zyp> to automatically ordering replacement lcd 2013-02-09T18:45:36 < dongs> zyp, one of my e-pals had an idea, bump it to turn on/off. 2013-02-09T18:45:41 < Laurenceb_> lolz wtf 2013-02-09T18:45:48 < Laurenceb_> randomzzz 2013-02-09T18:45:51 < emeb_mac> or, 90 tilt for page mode 2013-02-09T18:45:56 < Laurenceb_> i guess... 2013-02-09T18:45:59 < zyp> emeb_mac, that was the idea 2013-02-09T18:46:14 < dirty_d> i want an anodizing setup 2013-02-09T18:46:14 * emeb_mac doesn't think different 2013-02-09T18:46:17 < Laurenceb_> that flexi looks a bit fragile 2013-02-09T18:46:26 < Laurenceb_> expensive if you bend it too much.. 2013-02-09T18:46:32 < zyp> just $60 2013-02-09T18:46:37 < emeb_mac> luckly, you don't bend that part 2013-02-09T18:46:45 < emeb_mac> =I 2013-02-09T18:47:19 < gxti> unfortunately it contains neither an arduino nor a raspberry pi so it will not sell 2013-02-09T18:47:24 < dirty_d> how would you make a heater for an anodizing dye tank etc? 2013-02-09T18:47:29 < emeb_mac> Laurenceb: parts for RF ADC ordered... 2013-02-09T18:47:36 < Laurenceb_> cool 2013-02-09T18:47:36 < emeb_mac> PCB layout mostly done 2013-02-09T18:47:49 < dongs> was that the shit you linked the other day? 2013-02-09T18:47:55 < dongs> that was fpga + ??? but I idndt see the actual fpga 2013-02-09T18:48:01 < Laurenceb_> what architecture are you trying now? 2013-02-09T18:48:05 < Laurenceb_> yeah what dongs said 2013-02-09T18:48:08 < zyp> dongs, xilinx pmod connectors 2013-02-09T18:48:17 < dongs> some daughterboard shit? 2013-02-09T18:48:18 < emeb_mac> dongs: yeah - that hooks to std Digilent FPGA board 2013-02-09T18:48:19 < zyp> to hook up to generic xilinx boards 2013-02-09T18:48:21 < dongs> off a xilinxevalboard? 2013-02-09T18:48:21 < dongs> k 2013-02-09T18:48:26 < emeb_mac> but I have plans.... :P 2013-02-09T18:48:37 < dongs> beat dickstarterRF 2013-02-09T18:48:52 < emeb_mac> doubt that. 2013-02-09T18:49:16 < emeb_mac> they have HW already - didn't you see the stack of boards in the pix? 2013-02-09T18:49:31 < Laurenceb_> yeah but their hardware is a little flawed 2013-02-09T18:49:42 < zyp> so anyway, I had a silly idea yesterday, that nobody apparently have done yet 2013-02-09T18:49:45 < emeb_mac> for large values of "little" 2013-02-09T18:49:46 < Laurenceb_> like only 10dB harmonic rejection 2013-02-09T18:49:54 < dongs> quickly, tweet it on irc 2013-02-09T18:49:56 < zyp> a «stop shouting» usb dongle 2013-02-09T18:50:06 < dongs> awat 2013-02-09T18:50:12 < gxti> don't be a hater, everyone's product is unique and perfect 2013-02-09T18:50:19 < zyp> i.e. hid keyboard dongle, that turns off caps lock when on 2013-02-09T18:50:22 < Laurenceb_> a h8 dongle 2013-02-09T18:50:28 < zyp> (I said silly idea) 2013-02-09T18:50:29 < emeb_mac> what's that - kinda like "the clapper"? 2013-02-09T18:50:31 < dongs> anti-rage dongle? 2013-02-09T18:50:36 < dongs> what about shift pressed down 2013-02-09T18:50:40 < dongs> since thats how i usually type in c aps 2013-02-09T18:50:41 < BrainDamage1> zyp: every internet troll uses shift 2013-02-09T18:50:43 < dongs> just hold the fuckin shift down 2013-02-09T18:50:53 < BrainDamage1> it's more difficult, so just angers you more 2013-02-09T18:51:06 < zyp> sure, this is meant to troll folks that keeps caps lock on 2013-02-09T18:51:20 < Laurenceb_> when i was at school someone managed to get some ps2 adaptors that added trolling to your text 2013-02-09T18:52:04 < emeb_mac> kinda the opposite of a keylogger 2013-02-09T18:52:07 < zyp> wouldn't work on OS X though 2013-02-09T18:52:26 < zyp> OS X has per-keyboard caps lock 2013-02-09T18:52:27 < BrainDamage1> because osx doesn't work with non apple keybs? 2013-02-09T18:52:51 < zyp> if you hook up two keyboards, you can turn on the locks individually 2013-02-09T18:53:11 < zyp> so you could have a seperate keyboard for raging 2013-02-09T18:53:28 < Laurenceb_> dongs: how is the backlight powered? 2013-02-09T18:53:30 < dongs> does crapple have separate DPI for different monitors? 2013-02-09T18:53:36 < dongs> Laurenceb_: by usb 2013-02-09T18:53:36 < Laurenceb_> from usb? or is there 5v on DP? 2013-02-09T18:53:38 < Laurenceb_> ah 2013-02-09T18:53:40 < zyp> dongs, I think so 2013-02-09T18:54:05 < dongs> Laurenceb_: apparently it draws like 900mA 2013-02-09T18:54:12 < zyp> my pdf reader changes zoom level on content when I move it between monitors here 2013-02-09T18:54:16 < dongs> i mean entire setup 2013-02-09T18:54:25 < zyp> well, used to 2013-02-09T18:54:50 < zyp> doesn't now, for some reason 2013-02-09T18:55:10 -!- BrainDamage1 [~BrainDama@62.10.6.67] has quit [Quit: Leaving.] 2013-02-09T18:55:35 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-09T18:55:38 < dongs> hmpf. 2013-02-09T18:55:49 < dongs> ididnt notice i was running the panel off my main graphics card 2013-02-09T18:56:24 < dongs> ill try off intel next time 2013-02-09T18:56:25 < zyp> why did it matter? 2013-02-09T18:56:37 < dongs> no, it didnt, but I thought stuff was moving pretty quick between panel and my 2 other displays 2013-02-09T18:56:45 < zyp> ah 2013-02-09T18:56:54 < dongs> i was kinda impressed, i thought that would have been copied between 2 gpus 2013-02-09T18:57:58 < zyp> my motherboard apparently has this shit: http://www.lucidlogix.com/product-virtu-mvp.shtml 2013-02-09T18:58:01 < zyp> whatever that is 2013-02-09T18:58:05 -!- barthess [~barthess@178.154.25.106] has quit [Quit: Leaving.] 2013-02-09T18:58:10 < zyp> for running onboard and pcie gpu together 2013-02-09T18:58:41 < dongs> i dunno about together, but intel shit just shows up as a separate graphics card and I can connect more shit to it 2013-02-09T18:59:03 -!- barthess [~barthess@185.6.25.172] has joined ##stm32 2013-02-09T18:59:39 < zyp> I've never tested, it doesn't appear to work out of the box on OS X 2013-02-09T19:00:01 < zyp> and it doesn't really matter when I can hook four displays to nvidia card either 2013-02-09T19:00:20 < dongs> http://www.lucidlogix.com/technology-cloud.html 2013-02-09T19:00:23 < dongs> these guys also claim this shit 2013-02-09T19:00:30 < dongs> somethign about "cloud graphics virtualization" 2013-02-09T19:00:33 < zyp> hah 2013-02-09T19:03:44 < dongs> bedtime 2013-02-09T19:05:04 < dirty_d> i want one, http://www.youtube.com/watch?v=Y2Xa0PMHM1U 2013-02-09T19:05:18 < dirty_d> this stuff is expensive 2013-02-09T19:05:52 < zyp> dirty_d, by the way, are you a cnc guy? 2013-02-09T19:06:29 < dirty_d> zyp, i guess, lol but a new one 2013-02-09T19:06:37 < dirty_d> i havent made much of anything yet 2013-02-09T19:06:51 < zyp> cool, wanna fix me an enclosure for that ipad lcd and adapter board? :p 2013-02-09T19:08:31 < dirty_d> can i see it? 2013-02-09T19:08:57 < zyp> see pic dongs posted 2013-02-09T19:09:01 < zyp> http://i.imgur.com/U9UGXhx.jpg 2013-02-09T19:09:58 < dirty_d> how big is that? 2013-02-09T19:10:10 < zyp> 10", ipad panel 2013-02-09T19:10:34 < dirty_d> hmm 2013-02-09T19:10:43 < dirty_d> my mill only has 7" of y travel 2013-02-09T19:11:04 < dirty_d> i suppose the work could be repositioned 2013-02-09T19:12:07 -!- neuro_sys is now known as neuro_sys|off 2013-02-09T19:12:28 < dirty_d> a 12"x12"x0.5" aluminum plate is $61 2013-02-09T19:12:35 < dirty_d> how thick would it need to be? 2013-02-09T19:17:57 < zyp> dunno 2013-02-09T19:18:25 < emeb_mac> use 2 2013-02-09T19:18:32 < emeb_mac> clamshell... 2013-02-09T19:20:36 < dirty_d> yea 2013-02-09T19:25:36 < emeb_mac> lols - Xilinx backend content servers down. Can't get Spartan6 datasheets from anywhere since they all link back to Xilinx. 2013-02-09T19:26:43 < emeb_mac> good thing I've got archive copies... 2013-02-09T19:28:43 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-09T19:30:16 -!- neuro_sys|off is now known as neuro_sys 2013-02-09T19:31:48 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-09T19:36:35 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-09T19:42:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-09T19:44:58 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-09T19:45:56 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 248 seconds] 2013-02-09T19:54:20 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has joined ##stm32 2013-02-09T19:56:44 -!- Count_Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Ping timeout: 255 seconds] 2013-02-09T19:57:42 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving] 2013-02-09T20:10:23 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Changing host] 2013-02-09T20:10:23 -!- dfletcher_ [~fletch@drupal.org/user/72475/view] has joined ##stm32 2013-02-09T20:10:30 -!- dfletcher_ is now known as drgreenthumb 2013-02-09T20:12:04 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Ping timeout: 248 seconds] 2013-02-09T20:16:02 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-09T20:23:05 < Thorn> >Normally the Cortex-M3 wait until the previous write is completed before the interrupt return. 2013-02-09T20:25:20 < Thorn> http://www.keil.com/forum/18951/ <- Mr. Yiu replies 2013-02-09T20:28:14 < gxti> good to know i suppose 2013-02-09T20:34:07 < dirty_d> hmm 2013-02-09T20:34:51 < dirty_d> am i missing anything to turn on ADC1? http://pastebin.com/N9VbTiQC 2013-02-09T20:34:57 < dirty_d> get stuck in a forever while loop 2013-02-09T20:46:21 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-09T20:56:17 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Read error: Connection reset by peer] 2013-02-09T20:56:43 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T21:07:12 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Read error: Connection reset by peer] 2013-02-09T21:07:38 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T21:24:19 < karlp> "Current generations of DIY drones are controlled by Wi-Fi, so the distances they can travel are limited" wat? 2013-02-09T21:25:27 < Thorn> autonomous navigation using gps is impossible? 2013-02-09T21:27:27 < dirty_d> nope 2013-02-09T21:27:30 < gxti> or illegal :p 2013-02-09T21:27:38 < dirty_d> its not illegal is it? 2013-02-09T21:27:45 < gxti> depends on where you are i suppose 2013-02-09T21:27:49 < dirty_d> yea 2013-02-09T21:27:52 < gxti> pretty sure some places even require line of sight 2013-02-09T21:27:55 < dirty_d> that would be a fun project 2013-02-09T21:28:22 < dirty_d> it would be cool to use that for like shipping things short distances 2013-02-09T21:28:37 < dirty_d> like within a corporation etc 2013-02-09T21:28:43 < dirty_d> with buildings miles apart 2013-02-09T21:28:58 < dirty_d> have a heli or something drop it off 2013-02-09T21:29:40 < Thorn> you can track people with iphones etc. in real time right? 2013-02-09T21:30:09 < dirty_d> yea 2013-02-09T21:30:30 < dirty_d> use a phone as the radio communication channel 2013-02-09T21:30:35 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-09T21:30:40 < Thorn> so take an uav with e.g. 3G onboard and have it track people down and dump something nasty on them 2013-02-09T21:30:45 < dirty_d> lol 2013-02-09T21:31:13 < Thorn> (and by nasty I mean surprise birthday presents of course) 2013-02-09T21:32:25 < Thorn> might need radio homing using gsm signal and/or computer vision for better accuracy 2013-02-09T21:32:59 < karlp> dirty_d: with those SSRs that say 3v-30V or so, 2013-02-09T21:33:09 < karlp> tehy normally really mean 3.3 or more. 2013-02-09T21:33:29 < karlp> and on the ones I'v eused, the LED will turn on at 2.5V or something, but the SSR itself won't switch 2013-02-09T21:33:39 < karlp> which was confusing.... 2013-02-09T21:36:27 < karlp> 20,000 tonnes of dead herring.... http://www.tweak.net.au/pics2/2013/February/herringadventure/pichtml/web_2013_02_09-14_47_23--img_5350_jfr.html 2013-02-09T21:39:32 < dirty_d> karlp, so 3.3v is fine? 2013-02-09T21:40:35 < Thorn> in my current project I'm powering 3.3V SSRs from 5V to be sure 2013-02-09T21:42:22 <+Steffanx> i know what you eat tomorrow karlp 2013-02-09T21:43:21 < karlp> dirty_d: was on mine, but ymmv. 2013-02-09T21:43:31 < karlp> there's a heap of somewhat dodgy ssrs out there. 2013-02-09T21:43:39 < karlp> and mine needed 3.3 all the way, not a drop less. 2013-02-09T21:44:34 < karlp> nah, we're shipping that shit to denmark to feed mink. 2013-02-09T21:44:42 < karlp> the local mink farmers don't even want it 2013-02-09T21:45:48 < dirty_d> what are you making with the? 2013-02-09T21:45:50 < dirty_d> them 2013-02-09T21:46:02 < karlp> what? ssrs? or dead herring? 2013-02-09T21:47:20 < dirty_d> ssrs 2013-02-09T21:47:35 < karlp> beer fridge controller 2013-02-09T21:47:40 < dirty_d> oh nice 2013-02-09T21:47:53 < karlp> well, it would have been better if the original controller hand't died. 2013-02-09T21:48:04 < dirty_d> i made one of those, but i just tweaked hte mechincal thermostat somehow 2013-02-09T21:48:14 < karlp> but now I can change the temp remotely, woo, (which I've never done) 2013-02-09T21:48:16 < dirty_d> for homebrew 2013-02-09T21:48:21 < karlp> and monitor the temp remotely (which I don't bother doing) 2013-02-09T21:48:32 < dirty_d> yea 2013-02-09T21:49:51 < BrainDamage> karlp: is your target goal with the temp controller that giant plain of dead fishes? 2013-02-09T21:49:54 < zyp> Thorn, interesting 2013-02-09T21:50:03 < zyp> re. Yiu 2013-02-09T21:50:38 < zyp> «done» is a relative term 2013-02-09T21:51:09 < zyp> but the dummy read was a neat idea 2013-02-09T21:52:07 < zyp> I'll keep that in mind if I run into similar problems in future 2013-02-09T21:52:20 < Thorn> I wonder if DMA running in parallel could make things worse (if there's no bus accessese between resetting the IRQ and return) 2013-02-09T21:52:41 < Thorn> -e 2013-02-09T21:53:10 < zyp> how so? 2013-02-09T21:53:56 < Laurenceb_> Thorn: is the problem that isr is not reset and reenters? 2013-02-09T21:54:02 < zyp> yes 2013-02-09T21:54:06 < Thorn> it could delay the IRQ resetting transaction, so it would happen too late 2013-02-09T21:54:13 < Laurenceb_> ah ive seen this 2013-02-09T21:54:27 < Laurenceb_> i just moved isr reset to the start or isr :P 2013-02-09T21:54:33 < Laurenceb_> *of 2013-02-09T21:54:56 < Thorn> Laurenceb_: what chip was that? 2013-02-09T21:55:19 < Thorn> e.g. can F1 have this problem? 2013-02-09T21:55:21 < Laurenceb_> F103 2013-02-09T21:55:24 < Laurenceb_> yes 2013-02-09T21:55:40 < jpa-> it usually makes most sense to reset the ISR bit at start of interrupt anyway 2013-02-09T21:55:46 < zyp> Laurenceb_, we were debugging this issue for dirty_d earlier today, with code breaking on -O2 but working on -O0 2013-02-09T21:55:48 < Thorn> hm. it's not supposed to have the kind of cache that Joseph Yiu describes, is it? 2013-02-09T21:55:58 < dirty_d> Laurenceb_, you feel my pain? 2013-02-09T21:56:07 <+Steffanx> Why it makes sense jpa- ? 2013-02-09T21:56:14 < zyp> Thorn, it's probably the AHB-APB bridge 2013-02-09T21:56:24 < Laurenceb_> hehe 2013-02-09T21:56:30 < jpa-> Steffanx: that way if another interrupt occurs while you are processing the previous one, you won't miss it 2013-02-09T21:56:54 < zyp> Thorn, so I suspect you might encounter that with any APB peripherals 2013-02-09T21:56:56 < dirty_d> jpa-, you will miss it if you dont do that? 2013-02-09T21:57:13 < dirty_d> oh the same interrupt right? 2013-02-09T21:57:16 < Laurenceb_> yes 2013-02-09T21:57:18 < Laurenceb_> same isr 2013-02-09T21:57:21 < dirty_d> ahh 2013-02-09T21:57:28 < Thorn> ah, that bridge probably has a FIFO for clock domain crossing? 2013-02-09T21:57:28 < jpa-> yeah 2013-02-09T21:57:33 < Laurenceb_> needs to run again - it will tailchain 2013-02-09T21:58:09 < dirty_d> i guess it depends, you might not need it to run again in some situations i guess 2013-02-09T21:58:59 < jpa-> sure, and in many situations two interrupts will never occur so close together 2013-02-09T21:59:35 < zyp> I noticed a fun phenomenon when I was playing with stellaris 2013-02-09T22:00:17 < zyp> on LM4F, GPIO blocks are hooked to both AHB and APB, the latter for compatibility reasons, so they can be accessed by either 2013-02-09T22:00:46 < zyp> when the peripheral is disabled (unclocked), attempts at AHB accesses would fail 2013-02-09T22:01:05 < zyp> while APB reads would return 0 2013-02-09T22:02:46 < zyp> I don't know the technical details of how the buses work, but I interpret this as APB doesn't report back status of an access, while AHB does 2013-02-09T22:03:37 < zyp> IIRC AHB slaves can also insert wait states if they need to, I don't think APB slaves can do that 2013-02-09T22:04:23 < zyp> so this leads me to think that «write done» is reported by the AHB-APB bridge 2013-02-09T22:04:30 < jpa-> atleast FSMC and FLASH interfaces can, but those are both on AHB 2013-02-09T22:04:37 < zyp> exactly 2013-02-09T22:05:13 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Read error: Connection reset by peer] 2013-02-09T22:05:36 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-09T22:06:52 < Thorn> this might be wrong but I noticed that access to an uninitialized peripheral on a lpc11 will genereate a fault while in stm32 it will simply do nothing (in case of gpio etc.) 2013-02-09T22:07:00 < zyp> I've managed to hang the debugger port of lpc43xx by trying to access an enabled but not configured sdram bank, I believe it gets stuck waiting for the slave to report status of the AHB transaction 2013-02-09T22:07:26 < zyp> Thorn, AHB or APB peripheral? 2013-02-09T22:10:42 < Thorn> looks like APB actually 2013-02-09T22:11:35 < Thorn> UM10398 fig. 3 page 13 - all peripherals except GPIO are on the APB 2013-02-09T22:13:41 < zyp> hmm, and F3 returns 0 for disabled peripherals on AHB 2013-02-09T22:13:57 < zyp> oh well, that's one disproven theory :p 2013-02-09T22:20:00 < dirty_d> i cook the strangest things 2013-02-09T22:20:24 < dirty_d> mushrooms, garlic, butter, soy sauce and mozzarella cheese 2013-02-09T22:21:15 < BrainDamage> those are not strange 2013-02-09T22:21:26 < dirty_d> no? 2013-02-09T22:21:27 < BrainDamage> if you cooked iron bolts, now I'd be impressed 2013-02-09T22:21:31 < dirty_d> this is actually delicious 2013-02-09T22:21:41 < BrainDamage> each ingredient is not strange :p 2013-02-09T22:21:50 < dirty_d> lol, not apart 2013-02-09T23:33:57 < Bird|lappy> dirty_d, that actually sounds pretty tasty, I'd stick an escargot inside each mushroom cap though ;) 2013-02-09T23:34:11 < dirty_d> mmm 2013-02-09T23:35:15 < zyp> to me it mostly sounds like seasoning though 2013-02-09T23:35:19 < zyp> where's the food part? 2013-02-09T23:39:09 < Thorn> A differential-drive robot is equipped with a wheel encoder with 10 ''ticks" per revolution, wheel radius of 0.1m, and the two wheels are 0.2m apart. The robot starts at the origin (position and orientation is 0) and, during a short time interval of 0.5s a total of 5 ticks were recorded for the right wheel and 3 ticks for the left. Where is the robot approximately located after 0.5s? 2013-02-09T23:44:11 < dirty_d> the right wheel moved 0.314m and the left moved 0.1884m i think? 2013-02-09T23:44:18 < dirty_d> now fancy math 2013-02-09T23:44:50 < dirty_d> simultaneous equation i think 2013-02-09T23:49:52 -!- jef79m_lurking is now known as jef79m 2013-02-09T23:53:43 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-09T23:55:35 -!- barthess [~barthess@185.6.25.172] has quit [Quit: Leaving.] --- Day changed Sun Feb 10 2013 2013-02-10T00:01:08 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Quit: leaving] 2013-02-10T00:04:21 < dirty_d> d1 - d2 = (0.2 * 2), ((d1 * pi) / 0.314) = ((d2 * pi) / 0.1884) 2013-02-10T00:04:35 < dirty_d> d1 = 1m, d2 = 0.6m 2013-02-10T00:05:12 < dirty_d> so the right wheel moved in an arc along a 1m circle, and the left in a 0.6m circle 2013-02-10T00:05:16 < zyp> it's tracing a circle with radius relative to center of the robot = 0.347m 2013-02-10T00:05:27 < zyp> hmm 2013-02-10T00:05:36 < zyp> maybe I'm fucking up my math 2013-02-10T00:06:02 < dirty_d> i think the answer is a 2d coordinate 2013-02-10T00:06:09 < zyp> of course 2013-02-10T00:06:48 < dirty_d> so 0.314m around a 1m circle is what angle? 2013-02-10T00:07:23 < zyp> 1m circle? 2013-02-10T00:07:52 < dirty_d> i dunno thts what i came up with for the arc of the right wheel 2013-02-10T00:08:00 < dirty_d> 0.5m radius arc 2013-02-10T00:08:16 < dirty_d> 0.3m radisu arc for left 2013-02-10T00:08:26 < zyp> oh, wait, I had a typo here 2013-02-10T00:09:07 < zyp> ah, yes, now I get r = 0.4 for center of robot 2013-02-10T00:09:14 < zyp> that makes it easy 2013-02-10T00:09:35 < zyp> angle is 0.1 in radians 2013-02-10T00:10:23 < zyp> no, wait 2013-02-10T00:10:28 < zyp> I'm silly 2013-02-10T00:11:00 < BrainDamage> Laurenceb_: http://i.imgur.com/w4DCD0E.jpg 2013-02-10T00:11:58 < dirty_d> i gor 0.01 radians 2013-02-10T00:12:01 < zyp> it would be at approx. coordinate 0.6, 0.2 2013-02-10T00:12:55 < zyp> or maybe I'm still fucking up 2013-02-10T00:13:09 < zyp> 0.3, 0.05 might be right 2013-02-10T00:13:44 < Thorn> anyway the idea is to directly apply odometry equations given in the lecture: x' = x + D_c cos phi, y' = y + D_c sin phi, D_c = (D_l + D_r) / 2, phi' = phi + (D_r - D_l) / base, x = y = phi = 0, x' = 0.3, y' = 0, phi = 0.6 rad 2013-02-10T00:13:49 < zyp> fuck, I give up :p 2013-02-10T00:14:08 < Thorn> so I guess zyp is the winner :D 2013-02-10T00:14:39 < Thorn> * phi' = 0.6 rad 2013-02-10T00:15:51 < Thorn> (note that these are only valid for very short distances) 2013-02-10T00:15:57 < dirty_d> hmm 2013-02-10T00:17:58 < BrainDamage> I wonder if dongs helped them picking the name: http://www.potatosemi.com/potatosemiweb/aboutus.html 2013-02-10T00:17:58 < dirty_d> why is that? 2013-02-10T00:18:12 < BrainDamage> also, > 74-series GHz logical 2013-02-10T00:18:17 < BrainDamage> logic+ 2013-02-10T00:20:01 < Thorn> "make out chips as popular as potato chips" - this is what happens when the founders write advertising copy by themselves 2013-02-10T00:21:00 < gxti> if they're "locates" in San Jose maybe they could have gotten the americans to write their website 2013-02-10T00:29:37 < dirty_d> my girlfriend just spit out a mouthful of vodka, lmao 2013-02-10T00:29:45 < dirty_d> i made a glass with ice and she though it was icewater 2013-02-10T00:30:36 < zyp> ouch 2013-02-10T00:31:08 < dirty_d> she took it outa my hand, thought she knew what it was and was taking a sip 2013-02-10T00:36:20 < zyp> http://www.youtube.com/watch?v=a0KloWxffpc 2013-02-10T00:37:02 < dirty_d> what the hell? 2013-02-10T00:37:50 < dirty_d> they actually sell it like that? 2013-02-10T00:38:08 < Thorn> they used a mounting screw on the hdd to create an illusion of a speaker? that is clever 2013-02-10T00:38:08 < zyp> dunno, just retweeting 2013-02-10T00:39:09 < dirty_d> seems like that should be illegal 2013-02-10T00:39:59 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Remote host closed the connection] 2013-02-10T00:41:26 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-10T00:49:03 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 276 seconds] 2013-02-10T00:51:35 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-10T00:53:28 < gxti> so what's the story with futurelec? i know digikey isn't always cheap but futurelec has STM32F107RCT6 for just over half the cost of DK in singles. i don't even need the RC part, but when it costs 40% less than the RB at digikey why bother... 2013-02-10T00:55:00 <+Steffanx> uh lol zyp ? 2013-02-10T00:58:52 -!- jef79m is now known as jef79m_lurking 2013-02-10T01:01:09 < baird> They're a front for a Thai shop, that I know. Their supposed Global headquarters is run from a small villa apartment not far from where I used to live.. Never used them. 2013-02-10T01:02:08 < gxti> future electronics, not futurlec. sorry, didn't even realize those were separate things. 2013-02-10T01:08:59 -!- jef79m_lurking is now known as jef79m 2013-02-10T01:26:26 -!- neuro_sys is now known as neuro_sys|off 2013-02-10T01:32:03 -!- neuro_sys|off is now known as neuro_sys 2013-02-10T01:34:11 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-10T01:39:44 -!- neuro_sys is now known as neuro_sys|off 2013-02-10T01:59:52 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-10T02:00:59 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-10T02:24:25 < dongs> sup tweeters 2013-02-10T02:25:50 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-10T02:25:55 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-10T02:30:55 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-10T02:33:30 < Laurenceb_> THIS IS NOT TWITTER 2013-02-10T02:33:47 < Laurenceb_> ONLY TWATS TWEET 2013-02-10T02:41:04 < dongs> so far my delayextreme order is extremely disappointing 2013-02-10T02:41:15 < dongs> ordered "genuine" panasonic or something 16860 batts, got some unbranded shit 2013-02-10T02:41:18 < gxti> a dragon ate it 2013-02-10T02:41:21 < dongs> bt keyboard doesnt work 2013-02-10T02:41:41 < gxti> it's those chinese electrons, man 2013-02-10T02:41:45 < gxti> they just don't flow right 2013-02-10T02:44:10 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 246 seconds] 2013-02-10T02:51:58 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-10T02:56:20 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-10T03:00:28 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-10T03:01:07 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-10T03:01:57 < zyp> ha 2013-02-10T03:02:00 < zyp> dx and genuine 2013-02-10T03:02:02 < zyp> :p 2013-02-10T03:06:42 < dongs> hey it actually SAID genuine 2013-02-10T03:06:47 < dongs> and they were bit more expensive 2013-02-10T03:06:49 < dongs> than the usual garbage 2013-02-10T03:07:38 < dirty_d> at least you didnt get socks 2013-02-10T03:08:16 < dongs> true 2013-02-10T03:08:32 < zyp> is that going to be the usual response every time somebody receives an order with the wrong contents? 2013-02-10T03:08:35 < zyp> :p 2013-02-10T03:10:23 < dongs> did you see the other guy got some mafia ring type thing 2013-02-10T03:10:39 < dongs> http://i.imgur.com/EgehoCc.jpg 2013-02-10T03:11:12 < gxti> haha, customs form printed on the envelope 2013-02-10T03:11:12 < dirty_d> lol 2013-02-10T03:11:24 < zyp> ha , nice 2013-02-10T03:11:35 < dirty_d> the tatiff form 2013-02-10T03:12:00 < dongs> thats normal 2013-02-10T03:12:48 < gxti> i haven't bought that much shit from china to see it 2013-02-10T03:14:54 < dongs> http://www.analog.com/en/mems-sensors/mems-accelerometers/adxl362/products/product.html hmm! is this new? 2013-02-10T03:16:35 < dongs> mega funky package 2013-02-10T03:16:39 < dongs> 3x3.25mm 2013-02-10T03:18:12 < zyp> haven't heard about it 2013-02-10T03:18:25 < dongs> me neither 2013-02-10T03:18:49 < zyp> «micropower» sort of smells noisy 2013-02-10T03:19:15 < zyp> or in Laurenceb-terms «nice, but it's shit» 2013-02-10T03:19:58 < dongs> i wonder why only invensense uses separate ADCs 2013-02-10T03:20:02 < dongs> and everyone else muxes 2013-02-10T03:20:10 < dongs> seems like a conspiracy 2013-02-10T03:20:16 < dongs> wonder if they patented it 2013-02-10T03:20:19 < zyp> probably because muxes are cheaper :p 2013-02-10T03:20:23 < dongs> pft 2013-02-10T03:22:57 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-10T03:24:45 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-10T03:27:04 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-10T03:29:27 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has joined ##stm32 2013-02-10T03:52:47 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:8d9c:302a:af74:20c6] has joined ##stm32 2013-02-10T03:53:06 < emeb_mac> yo all 2013-02-10T03:54:24 < dongs> donging on dongs 2013-02-10T03:54:51 < emeb_mac> how's the dp board? 2013-02-10T03:54:56 < dongs> working 2013-02-10T03:55:00 < emeb_mac> sweet 2013-02-10T03:55:33 < emeb_mac> make sure the brightness "goes to 11" 2013-02-10T03:57:27 < dongs> thats for zyp to try with his iMac 2013-02-10T04:00:32 < emeb_mac> so what's with the cubieboard & transport stream output? 2013-02-10T04:01:03 < dongs> board done, need to start testing code 2013-02-10T04:01:15 < dongs> I wrote most of the driver framework,, need to make the shit get together 2013-02-10T04:01:23 < emeb_mac> coo 2013-02-10T04:04:27 < emeb_mac> the ts hardware is part of the allwinner SoC? 2013-02-10T04:06:53 < zyp> heh 2013-02-10T04:07:08 < zyp> my imac is way too old, doesn't have DP 2013-02-10T04:07:28 < dongs> whoa 2013-02-10T04:07:32 < dongs> some kid showed up with a 3ds 2013-02-10T04:07:36 < dongs> playing "ar games" by nintendo 2013-02-10T04:07:49 < dongs> http://www.nintendo.com/3ds/ar-cards this shit 2013-02-10T04:07:53 < dongs> pretty neat stuff 2013-02-10T04:07:58 < zyp> isn't that pretty old already? 2013-02-10T04:08:04 < dongs> new to me, i dont get out much 2013-02-10T04:08:33 < dongs> http://www.nintendo.com/3ds/built-in-software/#/4 playing that box killing game 2013-02-10T04:08:38 < zyp> I think a guy I know was showing it off like ages ago 2013-02-10T04:09:04 < dongs> mkay 2013-02-10T04:09:13 < dongs> might pickup a 3ds, thats kinda neat 2013-02-10T04:09:32 < dongs> some fucking kids stole my dslite + real acekard cart 2013-02-10T04:10:04 < zyp> my mess stole mine 2013-02-10T04:10:17 < zyp> I should still have it, but I don't know where the fuck it is 2013-02-10T04:10:27 < dongs> try pinging it. 2013-02-10T04:10:31 < zyp> :p 2013-02-10T04:11:19 < zyp> I remember bringing it on 2010-trip to japan, then I don't know where the fuck I put it after I got home 2013-02-10T04:13:45 < emeb_mac> time to tidy up 2013-02-10T04:13:45 < zyp> one day I should go through everything in my storage room, throwing out all the garbage I don't need and organizing the rest 2013-02-10T04:14:14 < emeb_mac> just toss it on the rocks in the basement with your server rack 2013-02-10T04:14:37 < zyp> I have like boxes of stuff that I packed when I moved three years ago 2013-02-10T04:15:07 < zyp> that was just sitting around, until I moved again one and a half year ago :p 2013-02-10T04:15:11 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Ping timeout: 260 seconds] 2013-02-10T04:15:51 < emeb_mac> boxes unopened for many moves. Who knows what's in them... 2013-02-10T04:16:18 < zyp> I was digging through some other boxes today 2013-02-10T04:16:29 < emeb_mac> find anything good? 2013-02-10T04:16:45 < baird> I have 6 newphews.. all of them spoiled pathologically rotten by their cashed-up bogan parents. They've probably got 3-4 3DS'es each. 2013-02-10T04:17:17 < dongs> will trade dp_adapter boards for 3ds 2013-02-10T04:17:30 < zyp> a friend I know needed a pcmcia network adapter for some reason, and I mentioned I had a couple «if I can find them», so he came over to pick it up 2013-02-10T04:17:38 < dongs> ha ha 2013-02-10T04:17:39 < zyp> :p 2013-02-10T04:17:55 < dongs> i got a 33.6k modem pcmcia somewhere. 2013-02-10T04:17:58 < emeb_mac> how'd that work out? 2013-02-10T04:17:58 < dongs> with XJACK!! 2013-02-10T04:18:16 < dongs> except I have a general idea of its location down to about 30cm radius 2013-02-10T04:18:48 < zyp> well, I think the reason was that the usb network adapter he intending to use turned out to be too slow, because the laptop only supported usb1 :p 2013-02-10T04:19:01 < dongs> sounds like luinix garbage 2013-02-10T04:19:20 < zyp> I think it was some old thinkpad 2013-02-10T04:19:27 < zyp> but whatever 2013-02-10T04:19:40 < zyp> it weren't in the box I thought it should be in 2013-02-10T04:19:46 < baird> Sounds like my 600E 2013-02-10T04:19:49 < zyp> so I went looking through random boxes 2013-02-10T04:20:06 < emeb_mac> sounds dangerous 2013-02-10T04:20:06 < zyp> found lots of shit that I even forgot I had 2013-02-10T04:20:23 < emeb_mac> it's like christmas in february! 2013-02-10T04:20:54 < gxti> only it's all regifted 2013-02-10T04:20:58 < zyp> http://www.linuxfordevices.com/c/a/Linux-For-Devices-Articles/Device-profile-Hitachi-VisionPlate-wireless-tablet/ <- including a stack of these 2013-02-10T04:21:39 < gxti> transmeta yiss 2013-02-10T04:21:40 < baird> Someone did get the idea right.. 2013-02-10T04:21:53 < zyp> for some reason, I have three of those. 2013-02-10T04:21:55 < emeb_mac> haha transmeta 2013-02-10T04:22:02 < dongs> haha transmeta! 2013-02-10T04:22:07 < dongs> blast from the dead past 2013-02-10T04:22:16 < dongs> officially deader than *BSD and lunix toreballs didnt even save them 2013-02-10T04:22:17 < baird> 800x600 kinda kills it as a wireless X terminal, though. 2013-02-10T04:22:39 < baird> lol Paul Allen. He was the money behind Transmeta? 2013-02-10T04:22:43 < dongs> at least its TFT 2013-02-10T04:22:56 < dongs> back in 2003 you coulda gotten STN stuff. 2013-02-10T04:23:10 < zyp> :p 2013-02-10T04:23:26 < dongs> haha i remember midori lunix. it was like some useless lunix w/one patch that did crusoe-specific shit 2013-02-10T04:24:46 < zyp> I picked those up on my first trip to japan, was thinking I could do something cool with them, never did :p 2013-02-10T04:24:59 < dongs> i suppose it could at least run windows xp 2013-02-10T04:25:00 < dongs> barely 2013-02-10T04:25:14 < dongs> what kinda memory? 2013-02-10T04:25:24 < dongs> standard dimm? 2013-02-10T04:25:28 < dongs> or some proprietaryt microdimm crap 2013-02-10T04:25:33 < zyp> sodimm 2013-02-10T04:25:45 < dongs> I have a sharp mebius laptop ~2001 or so it used some bullshit proprietar sharp-only modules 2013-02-10T04:25:51 < zyp> like PC100 or PC133 or something 2013-02-10T04:28:03 < emeb_mac> I've got a carton full of these things somewhere: http://www.elinux.org/JuiceBox 2013-02-10T04:28:42 < zyp> what's the story? 2013-02-10T04:28:44 < zyp> :) 2013-02-10T04:29:18 < emeb_mac> Saw them blowing out on Overstock.com or something. Got a half-dozen for like $20 2013-02-10T04:30:26 < dongs> sounds useless. 2013-02-10T04:30:35 < emeb_mac> pretty much 2013-02-10T04:30:45 < emeb_mac> processor is CoB 2013-02-10T04:38:26 < dongs> fucking android garbage 2013-02-10T04:38:35 < emeb_mac> huh? 2013-02-10T04:40:17 < dongs> fucking nexus4 doesnt connect to cell data after being off wifi 2013-02-10T04:40:19 < dongs> have to fucking reboot 2013-02-10T04:41:08 < dongs> wow, fucking great 2013-02-10T04:41:11 < emeb_mac> bizarre 2013-02-10T04:41:12 < dongs> now it doesnt connect to wifi either 2013-02-10T04:41:17 < gnomad> a regular reboot won't do it? 2013-02-10T04:41:47 -!- theAdib_ [~adib@dslb-088-074-158-233.pools.arcor-ip.net] has joined ##stm32 2013-02-10T04:41:55 < zyp> sounds like the shit I deal with at work :p 2013-02-10T04:42:31 < emeb_mac> uh, fun? 2013-02-10T04:42:32 < dongs> garbage. 2013-02-10T04:42:40 < dongs> oyu know waht else absolutely rages me about android 2013-02-10T04:42:48 < dongs> lack of camera profiles per app 2013-02-10T04:42:57 < dongs> i dont wanna upload 8mp shit with imgur 2013-02-10T04:43:08 < dongs> but i cant be arsed to change shit to 8mp when im taking pics of my dick 2013-02-10T04:43:16 < dongs> so like,, why cant i have imgur remember to take shit at 1mp 2013-02-10T04:43:28 < dongs> and regular camera app uses its own profile at 8mp or wahtever 2013-02-10T04:43:34 < dongs> fuck, you'd think by 2013 they would have that shit figured out 2013-02-10T04:43:40 < emeb_mac> so is that the OS's fault or the app? 2013-02-10T04:43:45 < dongs> OS of course 2013-02-10T04:43:52 < dongs> im sure camera api doens't have anything like this 2013-02-10T04:44:06 < dongs> wouldnt be surprised if its not evne programmatically configurable 2013-02-10T04:44:11 < dongs> i.e. app cant say "take pics at 1mp" 2013-02-10T04:44:35 < emeb_mac> andorrhoid apis do have some massive holes. 2013-02-10T04:44:47 < zyp> I'm too tired to go look in the api docs to prove you wrong, but whatever 2013-02-10T04:44:53 -!- theAdib [~adib@dslb-088-074-154-140.pools.arcor-ip.net] has quit [Ping timeout: 240 seconds] 2013-02-10T04:45:02 < dongs> pfft 2013-02-10T04:45:20 < emeb_mac> portable fast fourier transform 2013-02-10T04:45:25 < dongs> yep shit needed reboot to even get back on wifi 2013-02-10T04:45:30 < dongs> buggy trash, man... 2013-02-10T04:45:39 < emeb_mac> get an iPhone. 2013-02-10T04:45:50 < zyp> my old phone sometimes needed reboot to charge 2013-02-10T04:45:57 < emeb_mac> worship at the temple of Jobs 2013-02-10T04:45:57 < zyp> that was a fun one 2013-02-10T04:46:53 < emeb_mac> I was looking at the android apis for audio. lame 2013-02-10T04:47:11 < emeb_mac> basically you can record and play sounds and play mp3 files. 2013-02-10T04:47:20 < emeb_mac> no fine-grain control, etc. 2013-02-10T04:47:42 < dongs> zyp, but you do agree there are no app camera profiels 2013-02-10T04:47:44 < emeb_mac> and don't even think about low-level audio DSP unless you want to do it all from scratch. 2013-02-10T04:47:47 < dongs> handled by default by os. 2013-02-10T04:47:50 < dongs> or camera api or wahtever. 2013-02-10T04:47:57 < zyp> dunno 2013-02-10T04:48:18 < dongs> if ther is and fags arent using it, tehy deserve to die. 2013-02-10T04:48:31 < zyp> I found the camera in my phone to be pretty much useless 2013-02-10T04:48:42 < dongs> it is 2013-02-10T04:48:55 < emeb_mac> why? 2013-02-10T04:49:07 < emeb_mac> I like the camera in my Sammy. 2013-02-10T04:49:07 < dongs> but in the rare occasion I do use the camera app to take a pic of something I will usually want it at whatever max shitty MP it can do 2013-02-10T04:49:16 < dongs> but I dont want same multimegabyte images on imgur. 2013-02-10T04:49:30 < dongs> or random garbage im sending to people in whatsapp or whatever 2013-02-10T04:49:37 < dongs> since that shit rescales to 640x480 anyawy 2013-02-10T04:49:43 < dongs> but the fucking OS keeps a 8meg image in camera folder. 2013-02-10T04:49:51 < dongs> which I then have to remembert to delte or something. 2013-02-10T04:50:05 < zyp> emeb_mac, why? because I'm comparing it to this: https://fbcdn-sphotos-h-a.akamaihd.net/hphotos-ak-snc6/191165_4035976650036_320530184_o.jpg 2013-02-10T04:50:15 < emeb_mac> guess I'm glad that my Sammy only has a 3MP camera. :) 2013-02-10T04:50:37 < emeb_mac> zyp: then ur doin' it rong. 2013-02-10T04:50:44 < emeb_mac> apples/oranges 2013-02-10T04:50:46 < dongs> zyp: obviosuly, this is what you need: http://www.blogcdn.com/www.engadget.com/media/2012/08/padcastership8-22.jpg 2013-02-10T04:50:59 < zyp> :p 2013-02-10T04:51:28 < dongs> you know the world is sad 2013-02-10T04:51:35 < baird> That's a /furry/ t-shirt..? 2013-02-10T04:51:41 < dongs> when you think of something utterly retarded (like dslr lens connected ot ipad) 2013-02-10T04:51:46 < dongs> and there's an actual product in existence. 2013-02-10T04:51:55 < dirty_d> i have a galaxy nexus, the camera is pretty swet 2013-02-10T04:51:59 < dirty_d> sweet 2013-02-10T04:52:03 < emeb_mac> furry microphone too 2013-02-10T04:52:11 < dirty_d> i have that fast burst camera app, it can take pics at 40fps, lol 2013-02-10T04:52:30 < dirty_d> i can choose the photo resolution on it too 2013-02-10T04:52:49 < emeb_mac> per app? 2013-02-10T04:52:53 < zyp> oh yeah? my camera can take pics at 60fps, it's called video 2013-02-10T04:52:58 < emeb_mac> lol 2013-02-10T04:53:06 < dongs> nowai 2013-02-10T04:53:13 < emeb_mac> wai 2013-02-10T04:53:24 < dirty_d> zyp, if i just take video its lower quality 2013-02-10T04:53:29 < dirty_d> and darker 2013-02-10T04:53:35 < emeb_mac> and blurry 2013-02-10T04:53:36 < dirty_d> this is like just taking pictures 2013-02-10T04:53:42 < zyp> … right 2013-02-10T04:53:51 < dirty_d> strange 2013-02-10T04:54:01 < baird> I think the candybar nokia and canon 80is i've got has less otherall volume and weight, anyway.. 2013-02-10T04:54:43 < dongs> < dirty_d> this is like just taking pictures 2013-02-10T04:54:44 < dongs> asghashdahashdash 2013-02-10T04:55:25 < dirty_d> action shots 2013-02-10T04:55:50 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 255 seconds] 2013-02-10T04:57:01 < dongs> haha in scamcaster demovideo you can see the rounded lens outline in some corners 2013-02-10T04:57:16 < dongs> that shits gotta use only like tiny bit of the crapple sensor 2013-02-10T04:57:20 < dongs> and the rest is cropped off 2013-02-10T04:58:06 < emeb_mac> solution in search of a problem 2013-02-10T04:58:08 < zyp> but really, the exposure of a video frame is no different from a photo exposure (for electronic shutter) 2013-02-10T04:58:37 < emeb_mac> you sure? 2013-02-10T04:58:55 < zyp> are you insuinating I'm wrong? 2013-02-10T04:59:01 < emeb_mac> seems like longer exposure would be good for video since you only see it for a fraction of a second. 2013-02-10T04:59:17 < emeb_mac> and motion blur is OK in video - conveys motion 2013-02-10T04:59:19 < gxti> implementation may vary 2013-02-10T04:59:35 < zyp> emeb_mac, longer than what? 2013-02-10T04:59:48 < baird> Something I used when busting some taggers on a train last month. Just have the movie record mode going, and no need for the pricks to stand still and look at the camera, just get a swing in the right direction. 2013-02-10T05:00:02 < emeb_mac> zyp: well of course the exposure won't be longer than the frame rate 2013-02-10T05:00:21 < zyp> emeb_mac, and indoor photos are often in the 1/30-1/60-range 2013-02-10T05:00:31 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-10T05:00:36 < emeb_mac> true 2013-02-10T05:01:08 < emeb_mac> but that's not because it makes better photos. because the sensors aren't sensitive enough for short exposures. 2013-02-10T05:01:19 < zyp> sure they are 2013-02-10T05:01:34 < dongs> just blame lunix/apple 2013-02-10T05:01:46 < emeb_mac> sounds like a plan. :P 2013-02-10T05:01:46 < baird> Gee, rich white kids trying to be Street.. http://i.imgur.com/WeNeLBP.jpg 2013-02-10T05:02:38 < dongs> so what are we gonna do for hte case for the panel, zyp 2013-02-10T05:02:56 < dongs> i wonder if I can get my dude to mill something out of solid piece of aluminum at work 2013-02-10T05:04:26 < baird> Ha-- if you don't want to go the entire VESA mount idea, use a 1/4" camera mounting hole, and then we can use things like gorillapods.. 2013-02-10T05:04:29 < zyp> dunno, I've been talking to some cnc pals, but haven't gotten any good suggestions yet 2013-02-10T05:05:06 < zyp> baird, hm, that's actually not a bad idea 2013-02-10T05:06:19 < emeb_mac> those are cute. so grabby. 2013-02-10T05:07:05 < zyp> can even put it on a normal tripod then, might be neat for showing shit outdoors, like quadrotor telemetry or whatever 2013-02-10T05:07:24 < zyp> fits nicely with the portability of it being usb-powered 2013-02-10T05:07:44 < emeb_mac> win 2013-02-10T05:08:17 < zyp> emeb_mac, so anyway, I grabbed my camera and pointed it at my desk 2013-02-10T05:08:31 < zyp> it wants to use 1/30 at f/2.8 and ISO400 2013-02-10T05:08:39 < emeb_mac> ok 2013-02-10T05:08:45 < emeb_mac> because it's dark indoors 2013-02-10T05:09:03 < zyp> if I want more sensor sensitivity, I can dial it all the way up to 12800 2013-02-10T05:09:13 < zyp> ISO12800 2013-02-10T05:09:19 < emeb_mac> wow 2013-02-10T05:09:50 < zyp> of course, it'll probably look noisy as hell, but I can 2013-02-10T05:09:55 < emeb_mac> from a signal processing standpoint what's actually happening when trading off ISO and time in a sensor? 2013-02-10T05:10:22 < zyp> worse SNR 2013-02-10T05:10:40 < emeb_mac> yeah I get that, but what are they changing electronically? 2013-02-10T05:10:52 < emeb_mac> sensor bias voltages or just integration time? 2013-02-10T05:11:15 < zyp> gain, I guess 2013-02-10T05:11:27 < emeb_mac> aha. makes sesne 2013-02-10T05:11:34 < baird> q&d concept shoop: http://i.imgur.com/nZKGDb4.jpg 2013-02-10T05:12:45 < emeb_mac> so higher ISO is higher gain on the sensor pre ADC. 2013-02-10T05:12:59 < emeb_mac> which means more noise of course. 2013-02-10T05:13:15 < emeb_mac> as the exposure time drops... 2013-02-10T05:13:39 < zyp> http://bin.jvnv.net/f/g4OrE.JPG <- ISO12800 2013-02-10T05:13:50 < zyp> f/2.8, 1/640 2013-02-10T05:13:57 < dongs> < baird> q&d concept shoop: http://i.imgur.com/nZKGDb4.jpg 2013-02-10T05:13:58 < emeb_mac> that's really not bad 2013-02-10T05:14:05 < dongs> baird: whats thaT? jewpad panel? 2013-02-10T05:14:08 < zyp> probably better than phone :p 2013-02-10T05:14:13 < emeb_mac> but you've got a top-of-the-line camera 2013-02-10T05:14:19 < emeb_mac> no question better 2013-02-10T05:14:28 < emeb_mac> camera sensors suuuuuck 2013-02-10T05:14:42 < zyp> I mean, probably better than phone at a way lower ISO setting :p 2013-02-10T05:15:21 < emeb_mac> if the phone camera even gave you that degree of control... 2013-02-10T05:15:35 < baird> dongs: translucent jewmagicpanel 2013-02-10T05:16:16 < zyp> http://bin.jvnv.net/f/RD0mr.jpg <- phone, default settings 2013-02-10T05:16:21 < zyp> (whatever they are) 2013-02-10T05:16:29 < dongs> pure shit 2013-02-10T05:17:22 < dongs> great 2013-02-10T05:17:28 < dongs> i got this 10port hub from delayextreme 2013-02-10T05:17:31 < dongs> and something is rattling inside 2013-02-10T05:17:35 < zyp> 1/20, ISO1000 2013-02-10T05:17:38 < emeb_mac> blurrier on the phone and I bet the exposure is way longer 2013-02-10T05:17:57 < zyp> hmm, exif doesn't say aperture 2013-02-10T05:18:31 < emeb_mac> phone doesn't have variable aperture does it? 2013-02-10T05:18:38 < zyp> probably not 2013-02-10T05:18:47 < dongs> the caek is a lie 2013-02-10T05:18:57 < zyp> huge difference in DoF though 2013-02-10T05:19:10 < zyp> f/2.8 on DSLR gives a pretty shallow DoF 2013-02-10T05:19:25 < emeb_mac> ya 2013-02-10T05:19:35 < zyp> that's why both the monitor and the keyboard is out of focus 2013-02-10T05:19:43 < emeb_mac> spotted that 2013-02-10T05:20:06 < emeb_mac> so the phone aperture must be tighter than f2.8 2013-02-10T05:20:43 < emeb_mac> but they're basically pinholes, so that makes sense... 2013-02-10T05:20:45 < zyp> I believe smaller sensor also gives a wider DoF 2013-02-10T05:21:05 < zyp> deeper* 2013-02-10T05:21:28 < dongs> well at least it works 2013-02-10T05:21:30 < dongs> so thats not bad 2013-02-10T05:21:31 < emeb_mac> yeah - equivalent to tighter aperture - pinhole camera has everything in focus 2013-02-10T05:21:41 < dongs> and i have a toggle on/off switch 2013-02-10T05:21:43 < dongs> for entire hub. 2013-02-10T05:21:48 < dongs> perfect for resetting the shit 2013-02-10T05:22:52 < zyp> heh 2013-02-10T05:23:15 < dongs> now i have 4 free ports 2013-02-10T05:23:16 < emeb_mac> zyp: what's your dslr? 2013-02-10T05:23:20 < zyp> 60D 2013-02-10T05:23:48 < dongs> hub has a single TT 2013-02-10T05:23:49 < dongs> isnt that bad/ 2013-02-10T05:24:00 < zyp> sounds pretty normal 2013-02-10T05:24:07 < dongs> i forget what multi vs single tt does 2013-02-10T05:24:12 < dongs> somethign babout bandwidth sharing between fs/hs i think 2013-02-10T05:24:16 < zyp> yep 2013-02-10T05:24:38 < zyp> max 12megabit FS per TT 2013-02-10T05:24:55 -!- jef79m is now known as jef79m_lurking 2013-02-10T05:25:04 < zyp> so several FS devices on it would have to share 2013-02-10T05:25:07 < zyp> same goes for LS 2013-02-10T05:25:15 < dongs> oh. 2013-02-10T05:25:20 < dongs> so doesnt really matter 2013-02-10T05:25:24 < zyp> still sounds pretty normal 2013-02-10T05:25:26 < dongs> cuz fs shit is mice and other garbage 2013-02-10T05:25:42 < zyp> haven't seen any hubs with more than one TT yet 2013-02-10T05:25:47 < dongs> realy? 2013-02-10T05:25:49 < zyp> not that I've checked many hubs 2013-02-10T05:25:53 < dongs> i've used chips with multi tt 2013-02-10T05:26:09 < dongs> i think it was onl 4port though 2013-02-10T05:26:20 < zyp> hmm 2013-02-10T05:26:25 < dongs> i didnt think you could do 10 ports on single chip - usualyl i'e seen 7+4 2013-02-10T05:26:47 < zyp> there's no limitation 2013-02-10T05:26:55 < zyp> in protocol, I mean 2013-02-10T05:27:19 < dongs> right 2013-02-10T05:27:22 < zyp> was checking that myself once, because it's sorta weird that 7 is such a common number 2013-02-10T05:27:25 < dongs> but i never seen > 7port hub ics 2013-02-10T05:27:38 < dongs> well, its 1 upstream + 7 downstream 2013-02-10T05:27:39 < zyp> where did the number 7 come from? 2013-02-10T05:27:44 < dongs> so 8 2013-02-10T05:27:49 < dongs> probably that 2013-02-10T05:28:00 < zyp> ah, so 8 phys nicely laid out on die? 2013-02-10T05:28:02 < zyp> makes sense 2013-02-10T05:28:20 < zyp> why 4 port hub chips then? :p 2013-02-10T05:29:09 < dongs> prob 8 port with other shit not bonded out 2013-02-10T05:29:19 < dongs> but yeah dunno 2013-02-10T05:29:33 < dongs> was gonna get a usb3 hub but they're expensive evne at delayextreme 2013-02-10T05:29:37 < dongs> and none 10port. 2013-02-10T05:30:15 < emeb_mac> delayextreme. lol 2013-02-10T05:31:44 < dongs> http://i.imgur.com/Umq6TrU.jpg here's my amazing cameraphone pic. you can also see how fucking filthy my desk is 2013-02-10T05:32:07 < emeb_mac> wow - real wood 2013-02-10T05:33:14 < dongs> heh, its part of the building 2013-02-10T05:33:17 < dongs> the desk. 2013-02-10T05:35:04 < emeb_mac> built-in furniture? crazy. 2013-02-10T05:35:49 < emeb_mac> heh - Xilinx finally got their website back-end running again. can d/l datasheets now. 2013-02-10T05:38:45 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-10T05:41:18 < zyp> I apparently have two connected hubs supporting TT per port 2013-02-10T05:42:06 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:8d9c:302a:af74:20c6] has quit [Remote host closed the connection] 2013-02-10T05:42:42 < zyp> http://paste.jvnv.net/view/9HgiK 2013-02-10T05:45:57 < dongs> cheap apple fucks 2013-02-10T05:45:59 < dongs> with single tt 2013-02-10T05:46:04 < zyp> oh, one of those is the «5m usb cable with extender» I have lying across the room 2013-02-10T05:46:11 < zyp> the last one 2013-02-10T05:46:15 < dongs> what is that? juist an active hub? 2013-02-10T05:46:16 < zyp> only one of the ports are broken out 2013-02-10T05:46:27 < zyp> no, bus powered hub with 1 port 2013-02-10T05:46:45 < zyp> and 5m cable :p 2013-02-10T05:46:57 < zyp> because each segment can be max 5m 2013-02-10T05:48:12 < zyp> and the other is one of the hub chips in my dell monitor 2013-02-10T05:48:41 < zyp> both SMSC chips are in the monitor 2013-02-10T05:50:29 -!- jef79m_lurking is now known as jef79m 2013-02-10T05:50:31 < zyp> hmm 2013-02-10T05:52:27 < zyp> all the «Apple» ones, apart from the one in the keyboard are root hubs, which doesn't appear have a VID/PID of their own, so Apple driver just assigns its own 2013-02-10T05:53:30 < zyp> oh well 2013-02-10T05:54:06 < dongs> root hub doesnt have vid? 2013-02-10T05:54:08 < dongs> how does that work 2013-02-10T05:55:07 < zyp> I think root hub is just a virtual construct for managing the EHCI controller, or something like that 2013-02-10T05:56:38 < zyp> at least in the usb tree explorer in OS X, they are listed with PCI IDs instead :p 2013-02-10T05:56:51 < zyp> (which says intel) 2013-02-10T05:58:41 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-10T06:02:27 -!- BJFreeman [~bjfree@67.sub-75-244-137.myvzw.com] has quit [Quit: had a good time] 2013-02-10T06:03:16 < zyp> so anyway 2013-02-10T06:03:49 < zyp> I have two hubs with multiple TTs; one which only has one port and one which has HS devices connected to all ports 2013-02-10T06:04:01 < zyp> very useful. 2013-02-10T06:05:46 -!- jef79m is now known as jef79m_lurking 2013-02-10T06:07:45 -!- jef79m_lurking is now known as jef79m 2013-02-10T06:08:00 * emeb_mac had to use the googles to figure out what TT means. 2013-02-10T06:08:15 < emeb_mac> basically the bridge from FS<->HS 2013-02-10T06:08:47 < emeb_mac> which is only able to handle one FS port at a time, so multi TT is good for hubs with > 1 port. 2013-02-10T06:13:07 < zyp> yep 2013-02-10T06:13:34 < zyp> each TT acts as a single FS/LS host 2013-02-10T06:19:08 -!- jef79m is now known as jef79m_lurking 2013-02-10T06:29:14 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-10T06:32:50 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-10T06:35:53 -!- jef79m_lurking is now known as jef79m 2013-02-10T06:36:24 -!- jef79m is now known as jef79m_lurking 2013-02-10T07:00:33 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-10T07:00:43 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-10T07:03:43 * emeb_mac is surprised by the 1pcs price of 10/100 jacks w/ magnetics. 2013-02-10T07:08:53 < emeb_mac> this looks like about the lowest price part @ Mouser: http://www.mouser.com/ProductDetail/Pulse/J00-0045NL/?qs=sGAEpiMZZMvQhAhQbXdbBvRfPWqGf1OtfeO9PfCV0D4%3d 2013-02-10T07:09:03 < emeb_mac> DK is all about 2x that. 2013-02-10T07:10:59 < emeb_mac> nm - bad search fu. 2013-02-10T07:11:13 < emeb_mac> DK has 'em too for less. 2013-02-10T07:12:26 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-10T07:36:58 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-10T07:49:15 < gxti> emeb_mac: not much though 2013-02-10T07:49:25 < gxti> but yeah, that's the exact same part i'm using 2013-02-10T07:49:42 < gxti> second most expensive item on the BOM :[ 2013-02-10T07:49:43 < emeb_mac> oh good. 2013-02-10T07:49:58 < emeb_mac> zeroed right in on it. 2013-02-10T07:50:21 < gxti> although if i buy my stm32f107s from FE it might become the first most expensive 2013-02-10T07:51:00 < gxti> i don't understand how FE has them for $4.93, my jew alarm is going off 2013-02-10T07:51:13 < emeb_mac> and you're using the phy chip from micrel? 2013-02-10T07:51:18 < gxti> yep 2013-02-10T07:51:28 < gxti> works good, man 2013-02-10T07:51:35 < emeb_mac> any reason that the smsc/mchp parts wouldn't also work? 2013-02-10T07:51:42 < gxti> they weren't there when i was buying 2013-02-10T07:51:46 < emeb_mac> ah 2013-02-10T07:51:59 < gxti> the bottom of the PHY market is a volatile place apparenty 2013-02-10T07:52:16 < emeb_mac> looking at this one: http://www.digikey.com/product-detail/en/LAN8710AI-EZK/638-1083-ND/2166035 2013-02-10T07:52:26 < gxti> should be fine though, just don't buy 100 of them and solder it to something important 2013-02-10T07:52:32 < emeb_mac> the rmii-only one in the 24-pin pkg is not available 2013-02-10T07:52:52 < emeb_mac> heh. just proto now. 2013-02-10T07:53:10 < emeb_mac> so what's the deal with FE? 2013-02-10T07:53:24 < gxti> http://www.futureelectronics.com/en/technologies/semiconductors/microcontrollers/32-bit/Pages/9301003-STM32F107RCT6.aspx?IM=0 2013-02-10T07:53:30 < gxti> $4.93. WHY?! 2013-02-10T07:54:07 < gxti> it's probably full of anthrax or something 2013-02-10T07:54:12 < emeb_mac> wondering what their price is for '407 parts... 2013-02-10T07:54:41 < gxti> $7.10 2013-02-10T07:54:49 < gxti> less than f107rb at DK 2013-02-10T07:54:59 < emeb_mac> yep - $3 less than DK 2013-02-10T07:55:08 < emeb_mac> for same part... 2013-02-10T07:55:12 < emeb_mac> hmmm 2013-02-10T07:55:14 < gxti> but the f107rc at FE is particularly low 2013-02-10T07:55:19 < gxti> just really suspicious 2013-02-10T07:55:30 < emeb_mac> shipping from canada? 2013-02-10T07:55:39 < gxti> 1 moose leg 2013-02-10T07:56:22 < emeb_mac> got a client in toronto. I'll ask him what his experience with them is... 2013-02-10T07:57:01 < gxti> they seem pretty legit, although their search is not great. not everybody can be DK i guess. 2013-02-10T07:59:19 < emeb_mac> mail sent. He stays up late - should have an answer in the morning. 2013-02-10T08:00:22 < gxti> $9 for fedex ground 2013-02-10T08:01:23 < emeb_mac> about the same. 2013-02-10T08:01:38 < gxti> definitely worth it for a bulk order 2013-02-10T08:01:48 < emeb_mac> but I usually get DK shipments via USPS priority for < $6 2013-02-10T08:01:57 < emeb_mac> and faster than UPS/FedEx 2013-02-10T08:02:24 < emeb_mac> FE doesn't handle Xilinx/Altera... 2013-02-10T08:02:27 < dongs> still blogging? 2013-02-10T08:02:28 < dongs> geez 2013-02-10T08:02:35 < emeb_mac> ya 2013-02-10T08:02:47 < gxti> fpga margins are near zero 2013-02-10T08:02:54 < gxti> no price breaks, no markup 2013-02-10T08:03:04 < dongs> future electrollings 2013-02-10T08:03:07 < dongs> gxti: thats if you buy from digikey 2013-02-10T08:03:25 < gxti> digibuttocks 2013-02-10T08:03:39 < dongs> what about that futureelctronics stm32? 2013-02-10T08:03:45 < dongs> seems price is abouty right 2013-02-10T08:03:46 < emeb_mac> FPGA vendors will cut deals for volume. 2013-02-10T08:03:47 < dongs> a bit high 2013-02-10T08:03:55 < gxti> it's half what DK lists 2013-02-10T08:03:59 < dongs> thats fine 2013-02-10T08:04:05 < dongs> dk lists F103CB for $7 too 2013-02-10T08:04:08 < dongs> yet its price is ~$2 2013-02-10T08:04:17 < gxti> hah, gay 2013-02-10T08:04:43 < gxti> well, buying everything from one place is kinda important when prototyping but i'm hitting this shit up next time i do any quantity 2013-02-10T08:04:47 < emeb_mac> welcome to the monopoly 2013-02-10T08:05:44 < dongs> meh.. need to run a lot of dicknplace t oday 2013-02-10T08:05:51 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-10T08:05:55 < dongs> on fcuking sunday, rage 2013-02-10T08:06:08 < emeb_mac> bummin' 2013-02-10T08:06:22 < R2COM> thats Ok, I worked all day today too 2013-02-10T08:06:38 < emeb_mac> the economy thanks you 2013-02-10T08:13:39 < gxti> wonder if these are pin compatible: http://www.aliexpress.com/item/-/533015913.html 2013-02-10T08:13:56 < dongs> pin compatible to waht 2013-02-10T08:14:12 < dongs> lan jacks are prety fucked up, a few manufacturers make up their own shit :( 2013-02-10T08:14:18 < gxti> http://www.digikey.com/scripts/DkSearch/dksus.dll?lang=en&KeyWords=553-1619-5-ND&cur=USD 2013-02-10T08:14:43 < dongs> hmm 2013-02-10T08:15:36 < gxti> LED pins are different. how asinine. 2013-02-10T08:15:52 < dongs> http://www.aliexpress.com/store/product/PCB-RJ45-Modular-Network-jack-socket-LAN-plug-with-integrate-transformer-magnetics-Tab-up-with-LED/910432_534136567.html 2013-02-10T08:15:55 < dongs> looks legit 2013-02-10T08:17:25 < emeb_mac> mechanicals seem to match 2013-02-10T08:17:31 < dongs> ya 2013-02-10T08:18:16 < emeb_mac> gxti: what's different on the leds? seemed right to me. 2013-02-10T08:18:33 < gxti> on the star jack the diagonals face the same way 2013-02-10T08:18:52 < gxti> on the shit one they point towards each other 2013-02-10T08:19:17 < gxti> other than that all the dimensions match 2013-02-10T08:19:27 < emeb_mac> derp 2013-02-10T08:19:28 < emeb_mac> yep 2013-02-10T08:19:30 < gxti> you could make a footprint that matches both 2013-02-10T08:19:34 < emeb_mac> yeah 2013-02-10T08:19:37 < gxti> but that's some stupid shit 2013-02-10T08:19:53 < gxti> also no solder tabs on the case, kinda lame 2013-02-10T08:20:21 < emeb_mac> meh 2013-02-10T08:21:44 < gxti> the one dongs posted is decent 2013-02-10T08:21:49 < gxti> not compatible, but better quality 2013-02-10T08:22:38 < dongs> yea looks slightly more legit plus has actual pinout 2013-02-10T08:23:28 < emeb_mac> and internal ckt 2013-02-10T08:25:07 < dongs> http://img01.cp.aliimg.com/imgextra/i2/847414747/T2y2xUXdhdXXXXXXXX_!!847414747.jpg 2013-02-10T08:26:15 < emeb_mac> ugh. furries 2013-02-10T08:26:48 < dongs> attn baird i guess. 2013-02-10T08:58:00 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-10T09:02:55 < baird> Gives me 'Double the First' flashbacks. 2013-02-10T09:03:00 < baird> Gives me 'Double the Fist' flashbacks. 2013-02-10T09:09:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 244 seconds] 2013-02-10T09:47:43 -!- jef79m_lurking is now known as jef79m 2013-02-10T09:56:31 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-10T10:19:34 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c4b0:eb67:9147:dffa] has joined ##stm32 2013-02-10T11:14:29 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-10T11:17:01 -!- neuro_sys|off is now known as neuro_sys 2013-02-10T11:25:11 -!- jef79m is now known as jef79m_lurking 2013-02-10T11:29:03 -!- barthess [~barthess@77.67.204.238] has joined ##stm32 2013-02-10T11:34:45 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-10T11:42:34 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-10T11:43:52 -!- jef79m_lurking is now known as jef79m 2013-02-10T11:46:57 -!- theAdib_ [~adib@dslb-088-074-158-233.pools.arcor-ip.net] has quit [Remote host closed the connection] 2013-02-10T11:48:22 -!- jef79m is now known as jef79m_lurking 2013-02-10T11:58:58 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-10T12:01:06 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-10T12:02:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-10T12:10:05 < dongs> stonedddddddddddddddddddddddd 2013-02-10T12:11:00 < neuro_sys> COOL! 2013-02-10T12:27:30 -!- neuro_sys is now known as neuro_sys|off 2013-02-10T12:33:11 -!- daku is now known as DaKu 2013-02-10T12:36:54 -!- jef79m_lurking is now known as jef79m 2013-02-10T12:38:44 -!- claude is now known as Claude 2013-02-10T12:44:32 -!- jef79m is now known as jef79m_lurking 2013-02-10T12:45:51 -!- jef79m_lurking is now known as jef79m 2013-02-10T12:48:21 -!- Claude is now known as claude 2013-02-10T12:55:35 -!- DaKu is now known as daku 2013-02-10T12:55:49 -!- daku is now known as DaKu 2013-02-10T12:57:07 -!- DaKu is now known as daku 2013-02-10T12:57:59 -!- daku is now known as DaKu 2013-02-10T12:59:51 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Ping timeout: 256 seconds] 2013-02-10T13:01:00 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-10T13:18:38 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:c4b0:eb67:9147:dffa] has quit [Ping timeout: 256 seconds] 2013-02-10T13:32:08 -!- jef79m is now known as jef79m_lurking 2013-02-10T13:56:02 < dongs> This version of the firmware allows ULINK2 to be used as a CMSIS-DAP compliant debug unit in addition to the traditional ULINK2 operation mode. 2013-02-10T13:56:06 < dongs> nice! 2013-02-10T13:57:31 < dongs> watch it brick my ulink clone :) 2013-02-10T13:57:35 < dongs> i'll be pissed. 2013-02-10T13:57:40 < dongs> cause its the best debugger for keil. 2013-02-10T13:57:46 < dongs> but I'll prob just buy the real thing if it does. 2013-02-10T13:59:41 -!- mrcan_ [~mrcan@78.173.78.247] has joined ##stm32 2013-02-10T13:59:42 -!- mrcan_ [~mrcan@78.173.78.247] has quit [Changing host] 2013-02-10T13:59:42 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-10T14:00:17 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-10T14:00:18 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-10T14:00:18 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-10T14:00:21 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-10T14:03:21 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 276 seconds] 2013-02-10T14:17:44 < dongs> seems to work. 2013-02-10T14:23:43 < baird> Real Life Diving By Zero:- finding the Google Streetview car in Google Streetview... http://i.imgur.com/UozoPCa.jpg 2013-02-10T14:35:41 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-10T14:46:44 < RuslanPopov> ppl, can you share an example of pwm capture code? 2013-02-10T14:58:01 -!- TeknoJucer [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-10T15:00:57 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has quit [Ping timeout: 252 seconds] 2013-02-10T15:06:06 -!- DaKu is now known as daku 2013-02-10T15:11:11 < dongs> sure 2013-02-10T15:11:29 < dongs> RuslanPopov: http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/drv_pwm.c enjoy 2013-02-10T15:11:37 < dongs> capture and output. 2013-02-10T15:12:56 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-10T15:22:10 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has joined ##stm32 2013-02-10T15:25:38 -!- theAdib [~adib@dslb-088-074-158-233.pools.arcor-ip.net] has joined ##stm32 2013-02-10T15:33:36 < RuslanPopov> thanks 2013-02-10T15:45:48 < RuslanPopov> hmm, it is impossible to understand without synthetic drugs :) 2013-02-10T15:53:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds] 2013-02-10T15:55:10 < zyp> «At this point we are clearly behind schedule, but we will work hard to make up ground over the next few weeks as we finish up the board routing and start building the first batch of boards.» 2013-02-10T15:55:28 < zyp> parallella kickstarter, should I be surprised? :p 2013-02-10T15:55:28 -!- tavish [~tavish@120.59.13.175] has joined ##stm32 2013-02-10T15:55:29 -!- tavish [~tavish@120.59.13.175] has quit [Changing host] 2013-02-10T15:55:29 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-10T15:57:16 < BrainDamage> we need to link the retarded guy from the rc forums now :p 2013-02-10T16:00:24 < Laurenceb_> tom66 :P 2013-02-10T16:00:35 < Laurenceb_> what kickfail project is this? 2013-02-10T16:00:57 < zyp> parallella 2013-02-10T16:01:05 < zyp> http://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone 2013-02-10T16:02:12 < zyp> apart from the delay, there were only good news though 2013-02-10T16:02:31 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal] 2013-02-10T16:02:52 < zyp> they swapped the zynq-7010 with a 7020 (three times as much fpga logic) and added more IO 2013-02-10T16:03:49 -!- claude is now known as Claude 2013-02-10T16:06:43 < Laurenceb_> omg it has I2C 2013-02-10T16:06:47 < Laurenceb_> so it must be good 2013-02-10T16:07:05 < Laurenceb_> does the usart support arduino bootloader? 2013-02-10T16:07:45 < zyp> clearly. 2013-02-10T16:08:34 -!- Claude is now known as claude 2013-02-10T16:16:45 -!- theAdib_ [~adib@dslb-088-074-185-058.pools.arcor-ip.net] has joined ##stm32 2013-02-10T16:17:45 < zyp> «I'm wondering how the Parallella Board can be used to offload processing from a Raspberry Pi, especially for OpenCL usage since the Raspberry Pi does not have an OpenCL implementation.» 2013-02-10T16:18:12 -!- _BJFreeman [~bjfree@221.sub-75-233-174.myvzw.com] has joined ##stm32 2013-02-10T16:18:14 < zyp> because a dual core cortex-a9 is not cool enough… 2013-02-10T16:18:47 -!- _BJFreeman is now known as BJFreeman 2013-02-10T16:19:54 -!- theAdib [~adib@dslb-088-074-158-233.pools.arcor-ip.net] has quit [Ping timeout: 264 seconds] 2013-02-10T16:23:18 < TitanMKD> zyp there's also possibility to program the FPGA ;) 2013-02-10T16:24:31 < zyp> well, yeah, that's why I bought it 2013-02-10T16:24:48 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-10T16:24:49 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has quit [Changing host] 2013-02-10T16:24:49 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-10T16:25:38 < zyp> or backed it or whatever 2013-02-10T16:25:41 < TitanMKD> zyp we too even if 28K cells could be not enough for some stuff it's a must for 99.9% of stuff in addition to 16 cores and 2 A9 !! 2013-02-10T16:26:14 < zyp> did you miss the latest update? it now has 85k cells 2013-02-10T16:26:19 < TitanMKD> and for 100us$ it is just a top price for such highly versatile hw 2013-02-10T16:26:30 < TitanMKD> ha really even the 100US$ version ? 2013-02-10T16:26:33 < zyp> 15:02:52 < zyp> they swapped the zynq-7010 with a 7020 (three times as much fpga logic) and added more IO 2013-02-10T16:26:42 < TitanMKD> so they switched to XC7Z020 ? 2013-02-10T16:26:48 < TitanMKD> wooo !!! 2013-02-10T16:27:11 < TitanMKD> and also now 560K memory in FPGA yes !! 2013-02-10T16:27:32 -!- TeknoJucer [~TeknkoJuc@69.158.165.12] has quit [Ping timeout: 248 seconds] 2013-02-10T16:28:33 < TitanMKD> ha yes i'm backer and i have not even see this news 2013-02-10T16:28:37 < TitanMKD> amazing ! 2013-02-10T16:29:18 -!- neuro_sys|off is now known as neuro_sys 2013-02-10T16:29:59 < TitanMKD> ha the news if from today ;) 2013-02-10T16:30:02 < TitanMKD> is 2013-02-10T16:30:32 < zyp> yes, that's why we're talking about it today 2013-02-10T16:30:36 < zyp> old news is no fun :p 2013-02-10T16:30:36 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-10T16:33:52 < TitanMKD> i'm very impatient to know details of connectors 2013-02-10T16:36:48 < zyp> they will connect 2013-02-10T16:37:40 -!- theAdib_ [~adib@dslb-088-074-185-058.pools.arcor-ip.net] has quit [Remote host closed the connection] 2013-02-10T16:38:15 < TitanMKD> the only draw back is the 2nd Gigabit port not available directly 2013-02-10T16:38:32 < TitanMKD> could have been a must for network stuff ;) 2013-02-10T16:39:13 < zyp> and an unnecessary cost for everything else 2013-02-10T16:40:49 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-10T16:41:09 < TitanMKD> yes anyway can be added easily on additional board 2013-02-10T16:43:10 < qyx_> everyone want full featured board for everything :S 2013-02-10T16:43:58 < qyx_> it seems to me like finding the one and only ultimate truth of life and the universe and.. 2013-02-10T16:44:33 < TitanMKD> hehe yes we are never happy and we search always something better, faster ... ;) 2013-02-10T16:45:19 < qyx_> zyp: how is the lpc board? 2013-02-10T16:45:22 < TitanMKD> but i think in 5 or 10 years hw will be frozen and we will reuse more and more same MCU ... 2013-02-10T16:47:13 < zyp> qyx_, haven't had time to play much with it yet, currently I'm trying to get back on track with quadrotor stuff instead 2013-02-10T16:47:54 < qyx_> ah 2013-02-10T16:48:21 < zyp> I have too many toys ;) 2013-02-10T16:49:09 < qyx_> me too, most of them unfinished :/ 2013-02-10T16:54:22 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-10T17:07:09 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-10T17:21:59 < emeb_mac> lol parallela 2013-02-10T17:22:46 < emeb_mac> offload rpi. oh, these people. 2013-02-10T17:23:12 < TitanMKD> why offload rpi when it replace it easily ;) 2013-02-10T17:23:44 < TitanMKD> especially with 1GB ram when RPI have only 256MB ;) 2013-02-10T17:23:47 < emeb_mac> zactly 2013-02-10T17:24:21 < emeb_mac> the only tool they have is a hammer... 2013-02-10T17:25:08 < TitanMKD> of course it requies lot of developer and cool application hw/hacks but with all potential behind there will be tons of stuff 2013-02-10T17:31:41 < zyp> http://bin.jvnv.net/f/sUitf.JPG 2013-02-10T17:33:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-10T17:37:53 < emeb_mac> purty! 2013-02-10T17:38:14 < emeb_mac> try not to break your props 2013-02-10T17:46:16 -!- neuro_sys is now known as neuro_sys|off 2013-02-10T18:02:26 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 255 seconds] 2013-02-10T18:02:57 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-10T18:05:36 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 264 seconds] 2013-02-10T18:08:33 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-10T18:27:24 < Tectu> http://www.geocities.jp/ja3npl/60mhz_db_scope/6record06.png 2013-02-10T18:36:17 < Laurenceb_> hardcore 2013-02-10T18:39:28 -!- neuro_sys|off is now known as neuro_sys 2013-02-10T18:41:48 < emeb_mac> nice. what is it though? 2013-02-10T18:48:29 < emeb_mac> looks like it's the DDS section from this: http://www.geocities.jp/ja3npl/60mhz_db_scope/60mhz_sa.html 2013-02-10T18:55:48 < emeb_mac> http://f.cl.ly/items/1I2s1D2E0K3d1C14292G/image.jpg 2013-02-10T19:01:39 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32 2013-02-10T19:34:22 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-10T19:41:28 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-10T19:43:19 <+Steffanx> Whoa, image post day 2013-02-10T19:43:46 < Robint91> hi all 2013-02-10T19:43:49 < Robint91> Hi Steffanx 2013-02-10T19:43:55 <+Steffanx> Lo 2013-02-10T19:44:34 < Robint91> why is every free ide sucky 2013-02-10T19:44:37 < Robint91> damm 2013-02-10T19:44:42 < Robint91> coocox is really shit 2013-02-10T19:44:49 <+Steffanx> 'shit'? 2013-02-10T19:45:34 < Robint91> Steffanx, can't open two project, moving a project == forgetting every setting, no c++ support, bad debugger 2013-02-10T19:45:49 < emeb_mac> "works for me" 2013-02-10T19:45:57 < emeb_mac> but yeah, all those things. 2013-02-10T19:46:06 < Robint91> no C and H file templates 2013-02-10T19:46:08 < Robint91> etc 2013-02-10T19:46:27 < emeb_mac> Still, I use it because one of my clients does. 2013-02-10T19:47:15 < Robint91> and every eclipse tutorail you find is half baked 2013-02-10T19:47:29 < Robint91> and requires that you do bazillion steps to get a project compiling 2013-02-10T19:47:32 < emeb_mac> don't even get me started on eclipse 2013-02-10T19:47:56 < emeb_mac> coocox is stripped-down eclipse and just barely makes it tolerable. 2013-02-10T19:48:36 < Robint91> coocox has stripped al the nice parts 2013-02-10T19:49:33 < karlp> ==> netbeans.... 2013-02-10T19:50:05 < gxti> agreed that coide is pretty crappy, not that i really like "full" eclipse or netbeans any more 2013-02-10T19:50:41 < Robint91> also I will not use a normal editor and makefiles 2013-02-10T19:50:53 < Robint91> that is back going to the stone age 2013-02-10T19:51:22 < gxti> good thing i'm not using a normal editor then, i'm using vim :P 2013-02-10T19:52:21 < Robint91> vim = normal editor 2013-02-10T19:52:31 < emeb_mac> *cough* 2013-02-10T19:52:41 < emeb_mac> how about geany? :) 2013-02-10T19:53:10 < gxti> there are plenty of options for doing IDE-like crap in vim although i don't use them either 2013-02-10T19:53:20 < gxti> pretty much just syntax highlighting and smart indent 2013-02-10T19:54:29 * Steffanx waits for the standard vim is great everything else sucks discussion 2013-02-10T19:54:38 < Robint91> gxti, so you use makefiles? 2013-02-10T19:54:43 < Robint91> or something like that 2013-02-10T19:54:48 <+Steffanx> scons :P 2013-02-10T19:54:50 < gxti> Steffanx: nah, i'm not into that shit 2013-02-10T19:54:54 < gxti> people can use whatever 2013-02-10T19:55:10 < gxti> Robint91: yes, makefiles. i like having my projects not tied to any particular IDE. 2013-02-10T19:55:31 < gxti> although i guess eclipse can generate makefiles 2013-02-10T19:55:36 < gxti> never gave it much thought 2013-02-10T19:55:58 <+Steffanx> How long did it take you to get productive using vim gxti ? 2013-02-10T19:56:25 < gxti> dunno, that was a long time ago 2013-02-10T19:56:47 <+Steffanx> The good old times 2013-02-10T19:58:59 < Robint91> gxti, how do you debug? 2013-02-10T19:59:45 < emeb_mac> I debug with ddd sitting on top of gdb 2013-02-10T19:59:56 < emeb_mac> works pretty well 2013-02-10T20:00:05 < gxti> using bare gdb right now, not especially fond of it 2013-02-10T20:00:14 < emeb_mac> bare gdb is a PITA 2013-02-10T20:00:17 < gxti> mostly because gdb's cmdline is hard to discover 2013-02-10T20:00:20 < emeb_mac> ya 2013-02-10T20:00:25 < gxti> i know what i want to do but not how to tell it :| 2013-02-10T20:00:51 < emeb_mac> ddd is pretty good for that - menus, source, etc. 2013-02-10T20:01:05 < emeb_mac> not that it doesn't have its share of warts though. 2013-02-10T20:01:48 < Robint91> see this is what I don't like having to use 13 different tools to get the job done 2013-02-10T20:02:29 < emeb_mac> but if you try to use just one tool then you're locked into what that tool's authors think is "the right way" 2013-02-10T20:02:37 <+Steffanx> Here ddd always tends to crash/hang .. 2013-02-10T20:02:53 < emeb_mac> which, after you've been working for a while likely isn't your way. 2013-02-10T20:03:21 < emeb_mac> Steffanx: sadness. I wonder if that's an OS/distro thing. 2013-02-10T20:05:49 < Laurenceb_> ddd is _old_ 2013-02-10T20:06:01 < Laurenceb_> there seem to be no updates since like 1999? 2013-02-10T20:07:06 < Laurenceb_> (2009/02/11) DDD 3.3.12 2013-02-10T20:07:13 < Laurenceb_> oh, still old tho 2013-02-10T20:07:24 <+Steffanx> You are old 2013-02-10T20:07:26 < emeb_mac> ya. looks old. feels old 2013-02-10T20:07:33 < emeb_mac> but works. 2013-02-10T20:07:43 < emeb_mac> and I don't see any viable alternatives. 2013-02-10T20:07:57 < gxti> Laurenceb_ is older than facebook, he must be irrelevant 2013-02-10T20:08:00 < Laurenceb_> nemiver 2013-02-10T20:09:17 < emeb_mac> huh 2013-02-10T20:09:24 < TitanMKD> Laurenceb Eclipse is really better for debug ;) 2013-02-10T20:09:27 < emeb_mac> give it a go 2013-02-10T20:09:29 < TitanMKD> but heavier 2013-02-10T20:09:54 -!- daku is now known as DaKu 2013-02-10T20:10:00 < emeb_mac> perspectives... blargh. 2013-02-10T20:10:38 * Robint91 wants a visual studio based IDE 2013-02-10T20:10:42 < Robint91> like AVR studio 2013-02-10T20:11:50 < karlp> so? why not use avr studio? 2013-02-10T20:11:56 < karlp> they make arm chips. 2013-02-10T20:12:01 < karlp> what's the problem? 2013-02-10T20:12:20 < Laurenceb_> but i couldnt get nemiver to work with arm 2013-02-10T20:12:35 < Laurenceb_> cant get it to connect 2013-02-10T20:12:43 < emeb_mac> hmm. 2013-02-10T20:13:18 < emeb_mac> ddd --debugger arm-none-eabi-gdb 2013-02-10T20:13:27 < emeb_mac> works dandy 2013-02-10T20:13:32 < gxti> they make weird arm chips :p 2013-02-10T20:13:37 < karlp> gdbtui is pretty decent. 2013-02-10T20:14:01 < Laurenceb_> yeah nemiver has all the right config stuff 2013-02-10T20:14:02 < Robint91> karlp, using ST part with AVRstudio 2013-02-10T20:14:06 < Robint91> -_- 2013-02-10T20:14:09 < Laurenceb_> but i couldnt make it work :-/ 2013-02-10T20:18:10 < Laurenceb_> feel free to try :P 2013-02-10T20:18:23 < emeb_mac> what the heck... 2013-02-10T20:18:43 < karlp> nemiver really doesn't want to do anything useful wtih a remote target. 2013-02-10T20:19:20 < karlp> "sorry it's impossible to restart a remote inferior" 2013-02-10T20:20:03 < karlp> meh, enough of that. 2013-02-10T20:21:35 < Laurenceb_> hehe 2013-02-10T20:21:40 < Laurenceb_> thats what i concluded 2013-02-10T20:23:27 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-10T20:32:38 < emeb> hmmm... 2013-02-10T20:32:57 < emeb> got nemiver installed, found syntax for invoking arm-none-eabi-gdb 2013-02-10T20:33:16 < emeb> no gdb command line though, so can't issue "attach_swd" command... 2013-02-10T20:34:08 < mervaka> write one? :P 2013-02-10T20:34:17 -!- DaKu is now known as daku 2013-02-10T20:34:18 < mervaka> i assume nemiver uses gdb? 2013-02-10T20:34:23 < emeb> you so funny... 2013-02-10T20:34:31 < emeb> nemiver sits on gdb, yes. 2013-02-10T20:34:55 < mervaka> so open the nemiver source code, and stick in a command line :) 2013-02-10T20:35:07 < emeb> "It's just so simple!" 2013-02-10T20:35:11 < mervaka> yep :) 2013-02-10T20:35:12 < mervaka> lol 2013-02-10T20:35:29 < mervaka> it's possible, but obviously not as quick as i make it out to be. 2013-02-10T20:36:43 -!- barthess [~barthess@77.67.204.238] has quit [Remote host closed the connection] 2013-02-10T20:37:14 -!- barthess [~barthess@77.67.204.238] has joined ##stm32 2013-02-10T20:37:19 -!- barthess [~barthess@77.67.204.238] has quit [Client Quit] 2013-02-10T20:38:53 < emeb> Well, looks pretty, but too GUI-centric & lots of necessary features missing or dumbed-down to the point of uselessness. 2013-02-10T20:39:31 < emeb> probably would work fine for for native debugging, but not so useful for cross... 2013-02-10T20:40:31 < Laurenceb_> i failed to get it to attach 2013-02-10T20:40:37 < emeb> ya 2013-02-10T20:40:43 < Laurenceb_> i managed to get it using arm-none-eabi-gdb 2013-02-10T20:40:54 < Laurenceb_> and set it to localhost:4242 2013-02-10T20:41:06 < Laurenceb_> but it still failed - like what karlp saw 2013-02-10T20:41:44 < emeb> Aha! got it working. 2013-02-10T20:41:54 < emeb> put the attach commands into .gdbinit 2013-02-10T20:41:57 < karlp> fuck that 2013-02-10T20:42:01 < karlp> that's just too dumb 2013-02-10T20:42:07 < karlp> and gets in the way of using gdb normally 2013-02-10T20:42:46 < emeb> ya - inelegant, but works. 2013-02-10T20:42:47 < Laurenceb_> oh nice 2013-02-10T20:43:06 < Laurenceb_> emeb: so what exactly did you do? 2013-02-10T20:43:37 < zippe1> Looking at the bug database, nemiver seems moribund 2013-02-10T20:43:40 < emeb> Laurenceb_: I'm using my BPM2 to talk to my F303 breakout. 2013-02-10T20:43:45 < Laurenceb_> ok 2013-02-10T20:43:56 < Laurenceb_> that should be similar to stlink 2013-02-10T20:44:07 < emeb> I've got a .gdbinit that sets up the BMP but doesn't automatically attach, so I just added the attach cmd. 2013-02-10T20:44:16 < Laurenceb_> hmm 2013-02-10T20:44:23 < emeb> Now single stepping, running, inspecting, etc. 2013-02-10T20:44:25 < Laurenceb_> i dont think stlink has that 2013-02-10T20:45:24 < emeb> Here's my .gdbinit (mostly from zyp): http://pastebin.com/UEdrDcqF 2013-02-10T20:45:39 < emeb> shouldn't be hard to modify for stlink stuff 2013-02-10T20:46:27 < Laurenceb_> ok so im missing the attach 1 2013-02-10T20:46:37 < Laurenceb_> but does stlink even need that? 2013-02-10T20:46:44 < emeb> beats me 2013-02-10T20:46:46 < Laurenceb_> hmm i dont have a discovery board here to test 2013-02-10T20:47:00 < Laurenceb_> guess ill try tomorrow - at least it seems possible now 2013-02-10T20:47:04 < emeb> you need to bring work home. :P 2013-02-10T20:47:20 < gxti> that's for BMP 2013-02-10T20:47:21 < gxti> not stlink 2013-02-10T20:47:32 < Laurenceb_> gxti: erm yeah i know 2013-02-10T20:47:43 -!- Tekno01 [~TeknkoJuc@69.158.165.12] has joined ##stm32 2013-02-10T20:47:55 < gxti> for stlink it depends on what your gdbserver is 2013-02-10T20:48:10 < gxti> if you mean texane's stlink no idea, i never got it working 2013-02-10T20:50:17 <+Steffanx> That is what texane's stlink is know for :) 2013-02-10T20:50:43 < emeb> known for not working? 2013-02-10T20:51:16 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has quit [Ping timeout: 246 seconds] 2013-02-10T20:51:42 <+Steffanx> not with external tools at least :) 2013-02-10T20:51:42 < emeb> ugh - slow mouse hover 2013-02-10T20:52:18 <+Steffanx> uh? 2013-02-10T20:52:34 < karlp> Laurenceb_: stlink doesnn't need the attach. 2013-02-10T20:52:41 < Laurenceb_> yeah i thought so 2013-02-10T20:52:49 < Laurenceb_> so that doesnt explain why it fails for me 2013-02-10T20:52:50 -!- tavish is now known as bat_man 2013-02-10T20:52:57 < emeb> nemiver uses mouse hover to display variables. You have to let it sit for a while before the pop-up. 2013-02-10T20:53:53 < emeb> here's a *great* idea: http://pypi.python.org/pypi/pytesla/ 2013-02-10T20:55:04 < Laurenceb_> mycar.honk_horn() 2013-02-10T20:55:07 < Laurenceb_> epic lol 2013-02-10T21:16:23 < BrainDamage> Laurenceb_:https://gs1.wac.edgecastcdn.net/8019B6/data.tumblr.com/tumblr_lvzsy2exja1r08qs8o1_500.jpg 2013-02-10T21:24:16 -!- bat_man [~tavish@unaffiliated/tavish] has quit [Quit: leaving] 2013-02-10T21:32:31 < Thorn> there's a programming languages class on coursera at this time, using SML for first 4 weeks. The number of SML questions on stackoverflow has increased at least 5-fold during this month. /join #sml and someone is having homework from this class explained to them. honor code ftw 2013-02-10T21:33:12 < emeb> gxti: my guy in Toronto says Future Electronics is reliable. 2013-02-10T21:34:27 < emeb> he doesn't buy much from them lately mainly because DK/Mouser have wider variety and better delivery times. 2013-02-10T21:35:22 < gxti> yeah, not replacing DK anytime soon. but it looks like my latest product is going to be popular so even buying just 10 STM32F107 from there would be nice 2013-02-10T21:37:09 < emeb> save a few $$ 2013-02-10T21:43:04 <+Steffanx> popular .. 10 gxti ? :P 2013-02-10T21:43:51 < jpa-> 10 is popular for some products :) 2013-02-10T21:44:32 < BrainDamage> selling even 100 of a 300GHz spectrum analyzer would be a worldwide success 2013-02-10T21:44:38 < gxti> for that part 10 is about where i'd draw the line on whether it's worthwhile to buy somewhere other than DK 2013-02-10T21:44:49 < gxti> not saying i'd sell only 10 units :P 2013-02-10T21:46:13 <+Steffanx> what product btw? 2013-02-10T21:48:18 < gxti> http://partiallystapled.com/2013/01/laureline-gps-ntp-server/ 2013-02-10T21:49:46 < BrainDamage> cool idea 2013-02-10T21:50:02 < gxti> rev 3 will have the gps onboard 2013-02-10T21:58:28 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-10T21:58:54 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-10T21:59:00 < jpa-> i hope they have implemented the ntp protocol properly and fully 2013-02-10T21:59:17 < jpa-> unlike, say, windows time services does :) 2013-02-10T22:00:42 < gxti> it's definitely better than the client "time service". i've never seen the server one so i don't know if it really exists like the docs say. 2013-02-10T22:01:12 < gxti> that board is just a server though, so it's not really a "full implementation" -- just "translating" gps to ntp 2013-02-10T22:01:34 < jpa-> (w32time is too lazy to take part in the network latency calculations.. instead it gives some bogus values that seem to actually amplify the network errors in the time.. causing +-2 second errors over LAN) 2013-02-10T22:01:38 < gxti> i installed meinberg's ntp build on my windows box and it seems to perform really really poorly 2013-02-10T22:02:15 < jpa-> i have been in search of a working ntp server for windows.. haven't found 2013-02-10T22:02:16 < gxti> like wildly varying offsets up to 300ms 2013-02-10T22:02:26 < gxti> but people on time-nuts say it works from them so *shrug* 2013-02-10T22:02:57 < gxti> it's still better than w32time though 2013-02-10T22:03:03 < gxti> which would go 30 seconds before correcting sometimes ;p 2013-02-10T22:04:00 < jpa-> my current workaround is ntpd detached from the system clock, with ntpdate running in a 2-minute loop to syncronize the time.. for some reason the ntpd build i have doesn't manage to drift the clock fast enough to keep it in sync 2013-02-10T22:04:53 < jpa-> (or sometimes it drifts too fast; i once saw a computer with the seconds ticking so fast that the difference was obvious to the eye.. minute lasted 50 seconds or so :) 2013-02-10T22:06:38 <+Steffanx> yay 2013-02-10T22:07:33 < gxti> i ran into a machine once where the RTC was busted, you could pull up the bios and watch the clock tick like 10x fast 2013-02-10T22:07:38 < emeb> time's arrow is apparently not a constant. 2013-02-10T22:07:58 <+Steffanx> a wise man has spoken 2013-02-10T22:08:04 <+Steffanx> or a man has spoken wise words 2013-02-10T22:08:11 < jpa-> maybe it was just moving really slow 2013-02-10T22:08:45 < emeb> or maybe you were on speed. 2013-02-10T22:09:08 < BrainDamage> I c what you did there 2013-02-10T22:09:37 < BrainDamage> but relatively speaking, it's not a big deal 2013-02-10T22:11:08 < emeb> http://www.youtube.com/watch?v=xCc-RWIp7XU 2013-02-10T22:12:37 < BrainDamage> bah, one can't even make some shitty jokes to get shunned with a youtube link 2013-02-10T22:13:29 < emeb> what do you want? It's a reaction, and Lurch rules! 2013-02-10T22:20:16 < gxti> screwed again by sc189, tried to power up my rev1 board so i could hack on features without blowing up something i've already sold and it's putting out 1.2v after powerup. oh well, i'll just inject 3.3v into one of the headers. 2013-02-10T22:21:06 < emeb> that's on the module you designed? 2013-02-10T22:21:16 < emeb> vs the bought ones 2013-02-10T22:21:24 < gxti> no, that's the secondary step-down from 5 to 3.3 2013-02-10T22:21:35 < gxti> it's going away in rev 3 2013-02-10T22:21:39 < emeb> right 2013-02-10T22:22:02 < emeb> sounds like those are kinda touchy 2013-02-10T22:23:23 < gxti> hrm, that was weird. so when i first plugged it in everything powered up but i went to connect SWD and that's when it flaked. 2013-02-10T22:23:31 < gxti> and even after disconnecting everything it stayed flaked. 2013-02-10T22:23:59 < gxti> so i bypassed it and used the bench supply and it was current-limiting so something is sort-of shorted... i cranked up the amps because it obviously wasn't a dead short, and the problem went away. 2013-02-10T22:25:10 < gxti> like steady-state 0.3A @ 2.something volts, then i increase current to 0.5A and it starts up after a second and drops down to 0.15A which is normal. 2013-02-10T22:25:14 < emeb> so whatever it was, you burned it away. :) 2013-02-10T22:25:19 < gxti> heh yeah 2013-02-10T22:25:55 < qyx_> what smps controller are you using here? 2013-02-10T22:26:58 < qyx_> ah schematic 2013-02-10T22:45:20 < Robint91> has someone worked with the FSMC? 2013-02-10T22:45:27 < Robint91> on a F4? 2013-02-10T22:46:14 -!- jef79m_lurking is now known as jef79m 2013-02-10T22:47:07 < Robint91> also my current design, http://i.imgur.com/bd6QhWx.png , STM32F4 + XC6SLX9 + some DDR SDRAM, it is a object tracker, this pcb is connected to a cmos camera and the pc, I can/hope to get 60fps @ 720p, and a object tracking update rate of 60Hz and a interpolated resolution of 4k 2013-02-10T22:49:57 < gxti> mmm yeah 2013-02-10T22:51:43 < Thorn> that fpga has builtin ddr termination? 2013-02-10T22:52:10 < Robint91> Thorn, sortof, short traces + correct io drivers 2013-02-10T22:53:05 < Robint91> Thorn, if the SDRAM is further away you need to terminate more, but with traces under 30mm it doesn't matter that much 2013-02-10T22:53:34 < Robint91> and also VTT termination is only needed when you multiple chips 2013-02-10T22:54:52 < Robint91> why not DDR2/DDR3, those chips come only in 0.65mm pitch bga, which isn't that nice to solder 2013-02-10T22:55:00 < Robint91> 0.8mm pitch is much easier 2013-02-10T22:55:22 < Robint91> Thorn, I'm only going to run at 133MHz 2013-02-10T23:06:51 <+Steffanx> This is for your vmarker Robint91 ? :) 2013-02-10T23:07:13 < Robint91> Steffanx, Yes it is, going away from the custom NDA sensor 2013-02-10T23:07:30 < Robint91> and going to an other more normal CMOS camera that also requires an NDA 2013-02-10T23:07:40 <+Steffanx> Something wrong with the sensor? 2013-02-10T23:07:44 < Robint91> Steffanx, but the whole object detection is going on the fpga 2013-02-10T23:08:22 < Robint91> Steffanx, to low resolution, to small sensor array, can't find any lenses for it, the wide angle lens that we have at test now is rubbish 2013-02-10T23:08:25 < Robint91> etc etc 2013-02-10T23:14:17 -!- jef79m is now known as jef79m_lurking 2013-02-10T23:18:43 -!- neuro_sys is now known as neuro_sys|off 2013-02-10T23:20:33 -!- jpqyt [~chatzilla@unaffiliated/gkwhc] has joined ##stm32 2013-02-10T23:28:48 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-10T23:28:54 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2013-02-10T23:29:03 -!- TitanMKD [~Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-10T23:30:16 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-10T23:33:46 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 245 seconds] 2013-02-10T23:37:31 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 245 seconds] 2013-02-10T23:38:31 -!- TitanMKD [~Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-10T23:49:46 < dirty_d> hmm, is having interrupt PriorityGroup from 0-4, and a PreemptionPriorityGroup from 0-15 any different than if there was only a PreemptionPriorityGroup from 0-79? 2013-02-10T23:50:10 < dirty_d> i.e. is it only so its easier to organize priorities? 2013-02-10T23:50:27 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-10T23:51:14 < dirty_d> it seems like the "real" preemption priority is just the prioritygroup multiplied by the preemption priority 2013-02-10T23:51:31 < zyp> huh? 2013-02-10T23:52:20 < dirty_d> oh, actually seems there are some restrictions to the preemption priorities depending on the priority group 2013-02-10T23:52:42 < zyp> I'm not sure what you are asking about, but there are four bits in the priority field 2013-02-10T23:52:52 < zyp> so you have 16 different priorities 2013-02-10T23:52:56 < dirty_d> zyp, im trying to understand what all this priority stuff means 2013-02-10T23:53:17 < dirty_d> http://www.aimagin.com/learn/index.php/STM32_Interrupt_Service_Routine_Priority 2013-02-10T23:53:26 < dirty_d> theres also a subpriority though 2013-02-10T23:53:36 < dirty_d> so 8 bits right? 2013-02-10T23:53:43 < zyp> no 2013-02-10T23:54:00 < zyp> the four bits are split into priority and subpriority, depending on the priority group setting 2013-02-10T23:54:20 < dirty_d> ahh 2013-02-10T23:54:46 < zyp> so you can have one group with 16 priorities, 16 groups of one, 8 groups of 2, 4 groups of 4 2013-02-10T23:54:49 < zyp> and so on 2013-02-10T23:55:28 < zyp> the difference is that interrupts of different sub-priorities, but within the same group can't preempt 2013-02-10T23:56:23 < zyp> so if you are processing one interrupt within a group, and then another higher priority interrupt from the same group arrives, it will have to wait until processing the first one is done 2013-02-10T23:56:23 < dirty_d> __STATIC_INLINE void NVIC_SetPriorityGrouping(uint32_t PriorityGroup) 2013-02-10T23:56:31 < dirty_d> there isnt a parameter for which irq channel 2013-02-10T23:57:22 < zyp> but if an interrupt from a higher priority group arrives, it will interrupt your current ISR 2013-02-10T23:57:49 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-10T23:57:51 < zyp> are you following? 2013-02-10T23:58:01 < mervaka> so priority grounds have preemption, yeah? 2013-02-10T23:58:09 < zyp> yes 2013-02-10T23:58:09 < mervaka> groups* 2013-02-10T23:58:49 < dirty_d> zyp, in what im reading here it says if theyre in the same group, but one has a higher preemption priority, then it can be preempted 2013-02-10T23:59:00 < zyp> no. 2013-02-10T23:59:16 < zyp> that's the whole point of grouping 2013-02-10T23:59:26 < zyp> no preemption within groups. --- Day changed Mon Feb 11 2013 2013-02-11T00:00:13 < dirty_d> EXTI1 is NVIC_PriorityGroup4 and NVIC_IRQChannelPreemptionPriority15 TIM2 is NVIC_PriorityGroup4 and NVIC_IRQChannelPreemptionPriority0 ADC3 is NVIC_PriorityGroup1 2013-02-11T00:00:14 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has joined ##stm32 2013-02-11T00:00:17 < dirty_d> If EXTI1 interrupt is being serviced and TIM2 interrupt occurs, EXTI1 interrupt is paused and put in a queue and TIM2 is serviced. This is because EXTI1 and TIM2 has the same NVIC_PriorityGroup4 level, and TIM2 has a higher NVIC_IRQChannelPreemptionPriority level. 2013-02-11T00:00:20 < zyp> if you set priority grouping to have one big group of 16, you will not have any preemption at all 2013-02-11T00:00:22 < dirty_d> so this is oncorrect? 2013-02-11T00:00:24 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-11T00:00:37 < mervaka> hang on 2013-02-11T00:00:42 < mervaka> i'm confused now too 2013-02-11T00:00:44 < mervaka> lol 2013-02-11T00:01:15 < dirty_d> zyp, so the prioritygrouping is global not per IRQ channel? 2013-02-11T00:01:24 < zyp> yes 2013-02-11T00:01:26 < dirty_d> ahh 2013-02-11T00:01:29 < mervaka> there's priority groups, preemption priority and subpriority 2013-02-11T00:01:30 < zyp> hmm 2013-02-11T00:01:46 < mervaka> so 2013-02-11T00:01:54 < mervaka> priority groups can preempt? 2013-02-11T00:02:14 < mervaka> same group, different preemption priority can preempt? 2013-02-11T00:02:25 < zyp> hang on 2013-02-11T00:02:33 < dirty_d> wait iwat 2013-02-11T00:02:34 < dirty_d> EXTI1 is NVIC_PriorityGroup4 and NVIC_IRQChannelPreemptionPriority15 TIM2 is NVIC_PriorityGroup4 and NVIC_IRQChannelPreemptionPriority0 ADC3 is NVIC_PriorityGroup1 2013-02-11T00:02:47 < dirty_d> it cant be global, if some are in group4, and others in group1 2013-02-11T00:02:57 < zyp> it's dumb, don't read it 2013-02-11T00:03:04 < zyp> hang on, I'll find some real documentation 2013-02-11T00:03:40 < zyp> http://infocenter.arm.com/help/topic/com.arm.doc.dui0553a/BABHGEAJ.html 2013-02-11T00:04:37 < dirty_d> oh i see 2013-02-11T00:04:47 < dirty_d> so the preemptionpriority determines the prioritygroup that its in 2013-02-11T00:05:01 < dirty_d> then what the hell is __STATIC_INLINE void NVIC_SetPriorityGrouping(uint32_t PriorityGroup) for? 2013-02-11T00:05:13 < mervaka> huh 2013-02-11T00:05:25 < zyp> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0553a/Cihehdge.html <- it's for setting the PRIGROUP field here 2013-02-11T00:05:52 < zyp> see the «Binary point» section 2013-02-11T00:06:27 < zyp> also, keep in mind this: «Implementations having fewer than 8-bits of interrupt priority treat the least significant bits as zero» - stm32 only has 4 2013-02-11T00:06:59 < dirty_d> ok 2013-02-11T00:08:14 < mervaka> so 2013-02-11T00:08:25 < mervaka> different priority groups can preempt? 2013-02-11T00:08:29 < zyp> yes 2013-02-11T00:08:44 < mervaka> same priority group, different preemption priority can preempt? 2013-02-11T00:08:59 < jpqyt> Hi, might anyone worked with DSP/BIOS rtos? I was wondering where it does all the system configuration 2013-02-11T00:09:08 < mervaka> same priority group, same preemption priority, different subpriority cannot preempt? 2013-02-11T00:09:33 < zyp> mervaka, there is nothing called «preemption priority» 2013-02-11T00:09:33 < dirty_d> so if you set PRIGROUP to 3, then you get 4 priority groups, and 4 subpriorities? 2013-02-11T00:10:10 < mervaka> zyp: i'm going by what the stdperipheral drivers say 2013-02-11T00:10:12 < mervaka> misc.c 2013-02-11T00:10:30 < zyp> dirty_d, no, you set it to 5 to get 4 groups of 4 2013-02-11T00:10:58 < dirty_d> oh i see, since hte 4 LSBs are 0 2013-02-11T00:11:14 < zyp> mervaka, then I have no idea what that means 2013-02-11T00:11:22 < mervaka> heh 2013-02-11T00:11:24 < zyp> dirty_d, correct 2013-02-11T00:12:07 < mervaka> apparently in one register, bits are shared according to what group they're in 2013-02-11T00:12:22 < mervaka> so in PG0, all 4 bits control subpriority 2013-02-11T00:12:51 < mervaka> in PG2, 2 bits subpri, 2 bits preempt pri 2013-02-11T00:13:07 < mervaka> in PG4, all 4 bits control preemption priority 2013-02-11T00:13:24 < mervaka> or something like that 2013-02-11T00:13:25 < zyp> that sounds about right 2013-02-11T00:14:45 < mervaka> so going back to my earlier justifications, am i on the mark? :P 2013-02-11T00:18:09 < dirty_d> zyp, so on an stm32 use `priority = ((group << 2) | sub) << 4` where group and sub can be 0-3 with an even split? 2013-02-11T00:18:40 < zyp> if you access the registers directly 2013-02-11T00:19:02 < zyp> some of the libraries already do the << 4 for you 2013-02-11T00:19:14 < dirty_d> ahh i see, yea it does 2013-02-11T00:19:45 < zyp> which may make sense, but is not very nice for portability 2013-02-11T00:20:24 < zyp> it's a reason that it's the lower bits that's not implemented and not the upper ones 2013-02-11T00:20:55 < dirty_d> so i guess unless you are going to have any interrupts taht will happens at the same time to just use all bits for the group 2013-02-11T00:21:06 < dirty_d> or if its rare 2013-02-11T00:21:40 < dirty_d> unless you dont want preemption on some 2013-02-11T00:21:46 < zyp> it's all up to what you need/do 2013-02-11T00:22:13 < dirty_d> i wonder why they only have 4 bits, complicates the device a lot? 2013-02-11T00:22:23 < zyp> I myself haven't even bothered changing priorities on anything 2013-02-11T00:22:29 < zyp> because I don't need to 2013-02-11T00:22:44 < dirty_d> does everything default to 0? 2013-02-11T00:22:56 < zyp> I don't remember 2013-02-11T00:25:05 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-11T00:45:51 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-11T01:04:41 -!- LeelooMinai_ [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-11T01:14:56 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-11T01:19:30 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T01:23:43 < emeb> Robint91: I've used FSMC on an F4 - works fine. 2013-02-11T01:24:19 < Robint91> emeb, okay, did you use normal SRAM? 2013-02-11T01:24:26 < Robint91> or something more special 2013-02-11T01:24:29 < emeb> I used PSRAM 2013-02-11T01:24:40 < emeb> Looks like SRAM, but is actually DRAM inside. 2013-02-11T01:25:04 < Robint91> emeb, the one that only comes in VFBGA 2013-02-11T01:25:27 < emeb> I've used both the Micron parts in the BGA and also ISSI has parts in a TSSOP. 2013-02-11T01:25:37 < emeb> but they're smaller capacity (ISSI) 2013-02-11T01:26:12 < Robint91> emeb, it is kind a stupid that they don't have a SDRAM controller 2013-02-11T01:26:23 < emeb> Are you going to do the DRAM controller in the S6 and emulate SRAM on the FSMC? 2013-02-11T01:26:33 < Robint91> emeb, sort of 2013-02-11T01:26:50 < emeb> Robint91: yeah - a nuisance that ST doesn't have a real DRAM controller. 2013-02-11T01:27:05 < Robint91> emeb, the S6 has a hardware DRAM controller that does DDR DDR2 and DDR3 2013-02-11T01:27:17 < Robint91> and has multiple interfaces 2013-02-11T01:27:20 < emeb> right. 2013-02-11T01:27:36 < emeb> but only works w/ DDR, not SDRAM. 2013-02-11T01:27:52 < Robint91> emeb, the idea is to write the image directly into the memory wiht channel A and connect the ST on channel B 2013-02-11T01:28:08 < Robint91> emeb, yeahn, but SDRAM is easy to do in logic 2013-02-11T01:28:19 < emeb> so the DRAM is an image buffer 2013-02-11T01:28:32 < emeb> set up as a dual-port 2013-02-11T01:29:01 < Robint91> emeb, yep, fpga calculates the region of interrest in realtime, pass that to the ST 2013-02-11T01:29:08 < Robint91> yes dual port 2013-02-11T01:29:27 < Robint91> ST calcutates the midpoint of those RoI's 2013-02-11T01:29:32 < Robint91> and passes them to the PC 2013-02-11T01:31:00 < Robint91> emeb, also I'm going to try to have the F4 configure the fpga from a bitstream stored on a SPI flash 2013-02-11T01:32:02 < Laurenceb_> how does DRAM differ from an interface point of view? 2013-02-11T01:32:16 -!- LeelooMinai_ [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Quit: Ex-Chat] 2013-02-11T01:32:56 < Robint91> Laurenceb_, a lot, SDRAM for example needs special commands that are transferred via the adress bus to refresh the memory, for example 2013-02-11T01:32:59 -!- LeelooMinai_ [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-11T01:33:45 < Robint91> Laurenceb_, SRAM is just powerup and go, SDRAM needs to be initialised 2013-02-11T01:33:55 < Laurenceb_> i see 2013-02-11T01:34:01 < Laurenceb_> but it refreshes itself? 2013-02-11T01:34:34 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-11T01:34:34 -!- LeelooMinai_ [~leeloo@198-84-185-212.cpe.teksavvy.com] has quit [Remote host closed the connection] 2013-02-11T01:34:36 < Robint91> Laurenceb_, not exaclty 2013-02-11T01:34:47 < Robint91> Laurenceb_, the memory controller needs to do that 2013-02-11T01:35:03 < Laurenceb_> it has to request a refresh? 2013-02-11T01:35:12 < Robint91> Laurenceb_, yep 2013-02-11T01:35:20 < Laurenceb_> i see 2013-02-11T01:35:38 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-11T01:36:43 < Robint91> Laurenceb_, but SDRAM lets you pack much more memory on the same area as SRAM 2013-02-11T01:37:17 < Robint91> Laurenceb_, if the F4 for example had a good SDRAM controller it would make it a nice uclinux system 2013-02-11T01:37:19 < zyp> it's also read and written in bursts, so the controller would normally have read and write buffers 2013-02-11T01:37:30 < zyp> kinda like a cache 2013-02-11T01:38:04 < Robint91> zyp, yeah, 2 words for DDR, 4 for DDR2, ... 2013-02-11T01:38:11 < Robint91> it is nice for streaming data 2013-02-11T01:38:30 < Robint91> but less good for random accessess 2013-02-11T01:38:45 < zyp> therefore the buffers. 2013-02-11T01:39:09 < Robint91> yep 2013-02-11T01:39:50 -!- LeelooMinai [~leeloo@198-84-185-212.cpe.teksavvy.com] has joined ##stm32 2013-02-11T01:39:55 < zyp> http://bin.jvnv.net/f/Mv7rw.JPG <- I made this board, but I haven't had time to test the sdram yet 2013-02-11T01:40:18 < zyp> the top end lpc chips do have sdram controllers 2013-02-11T01:43:21 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 245 seconds] 2013-02-11T01:44:59 < Robint91> zyp, yeah 2013-02-11T01:45:14 < Robint91> zyp, seen mine board that I'm designing? 2013-02-11T01:46:34 < zyp> I saw the one you pasted earlier tonight 2013-02-11T01:46:41 < zyp> looks nice 2013-02-11T01:46:54 < zyp> how many layers will it be? 2013-02-11T01:47:08 < dongs> at least hes using altium 2013-02-11T01:47:11 < dongs> and not some "eagle" shit 2013-02-11T01:47:16 < Robint91> zyp, 4 2013-02-11T01:47:23 < Robint91> dongs, :p 2013-02-11T01:47:32 < dongs> where is DDR? cause i dont see ddr routing 2013-02-11T01:47:35 < dongs> i see fail 2013-02-11T01:47:49 < Robint91> dongs, terminations is for pussies 2013-02-11T01:48:06 < dongs> also, bga fpga but not bga sdram? comeon 2013-02-11T01:48:17 < Robint91> and for traces shorter than 30mm the internal termination of the FPGA is good 2013-02-11T01:48:33 < dongs> really 2013-02-11T01:48:45 < dongs> and what about length matching and shit 2013-02-11T01:49:06 < Robint91> dongs, IO delays in the IOBs 2013-02-11T01:49:12 < dongs> haha 2013-02-11T01:49:26 < dongs> are you using internal XC6 sdram controller? 2013-02-11T01:49:30 < dongs> or writing your own shit 2013-02-11T01:49:32 < Robint91> dongs, 2013-02-11T01:49:38 < Robint91> dongs, using MCB 2013-02-11T01:49:40 < dongs> ok. 2013-02-11T01:49:49 < Robint91> dongs, there is lenght matching but it is hidden 2013-02-11T01:50:07 < Robint91> dongs, without any wiggly traces 2013-02-11T01:50:20 < dongs> o'rly? 2013-02-11T01:52:39 < emeb> no wiggly traces? How can that be? 2013-02-11T01:52:57 < dongs> surely trollen 2013-02-11T01:53:03 < dongs> most important question is, wehre is he gonna fab it 2013-02-11T01:53:10 < dongs> 2013-02-11T01:53:14 < emeb> heh 2013-02-11T01:53:23 < Robint91> eurocircuits 2013-02-11T01:53:26 < Robint91> maybe pcbcart 2013-02-11T01:53:37 < dongs> sounds like you arent going for quality then 2013-02-11T01:55:47 < upgrdman> wiggly traces are to match trace lengths, right? 2013-02-11T01:55:48 < dongs> hmpf, how does eurocircuits quote works 2013-02-11T01:55:56 < emeb> upgrdman: ya 2013-02-11T01:55:59 < dongs> its not printing it out 2013-02-11T01:56:05 < dongs> Price calculation fail. 2013-02-11T01:56:06 < dongs> lol 2013-02-11T01:56:11 < gxti> faillll 2013-02-11T01:56:27 < dongs> wtf 2013-02-11T01:56:37 < upgrdman> and you only need to be careful about matching trace lengths for very high speed stuff, right? 2013-02-11T01:56:45 < emeb> generally yes 2013-02-11T01:56:48 < dongs> sdram counts as high speed 2013-02-11T01:56:48 < upgrdman> k 2013-02-11T01:56:53 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-11T01:56:55 < gxti> 1mhz sdram olol 2013-02-11T01:57:12 * upgrdman should find the SIMMS that were in his old 486sx tower 2013-02-11T01:57:54 < emeb> 1 millihertz? 2013-02-11T01:57:56 < dongs> Robint91: so t ell me more how youre doing length matching "hiddenly" 2013-02-11T01:58:13 < Robint91> dongs, swapping DQ lines 2013-02-11T01:58:24 < upgrdman> maybe he's using shit tons of vias to adjust trace lengths ;o) 2013-02-11T01:58:35 < Robint91> dongs, also that 2013-02-11T01:58:37 < emeb> a stitch in time 2013-02-11T01:59:07 < gxti> doesn't save a drill bit 2013-02-11T01:59:14 < dongs> dunno, you got altium i'd just configure nets correctly and hit autoroute 2013-02-11T01:59:35 < Robint91> dongs, do you think? 2013-02-11T01:59:38 < dongs> yeah 2013-02-11T01:59:42 < Robint91> -_- 2013-02-11T01:59:56 < Robint91> dongs, do you use altium? 2013-02-11T01:59:59 < dongs> curvy traces look cooler than a shitton of vias 2013-02-11T02:00:10 < Robint91> dongs, did you every use an autoroute 2013-02-11T02:00:18 < dongs> in altium? no. 2013-02-11T02:00:34 < dongs> but I know it has routing modes that help w/routing length matched/diffpair stuff. 2013-02-11T02:01:01 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has quit [Read error: Operation timed out] 2013-02-11T02:01:13 < emeb> is it even legal to swap DQ lines - doesn't that screw up the command formats? 2013-02-11T02:01:33 < Robint91> emeb, you can swap DQ with in the bytelane 2013-02-11T02:01:45 < Robint91> and you can swap byte lanes them selfs 2013-02-11T02:02:12 < emeb> I suppose you have to tell the MCB about how things are wired up? 2013-02-11T02:02:25 < emeb> in a UCF or something? 2013-02-11T02:02:51 < Robint91> emeb, for the MCB you need to use specific pins to work 2013-02-11T02:03:04 < Robint91> but you still need to configure it in a UCF file 2013-02-11T02:03:20 < dongs> yeah, built in controller works on a set of pins 2013-02-11T02:03:37 < emeb> uh huh - didn't think it was too flexible. 2013-02-11T02:03:46 < dongs> kinda like fsmc on f4 :) 2013-02-11T02:04:37 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-11T02:04:39 < upgrdman> speaking of ucf's... with an fpga, does it matter which pins you use to make, say, an array of pins? seems like you could define at pin to be something ... 2013-02-11T02:04:50 < emeb> pretty flexible 2013-02-11T02:04:51 < upgrdman> any pin* 2013-02-11T02:04:59 < dongs> flexible unless its clock related or you care about delays etc 2013-02-11T02:05:03 < upgrdman> k 2013-02-11T02:05:16 < dongs> also depends on io blokc 2013-02-11T02:05:17 < emeb> Or if it hooks to a hard macro like the MCB or PCI blocks 2013-02-11T02:05:19 < upgrdman> what kind of delays >1 clock cycle? 2013-02-11T02:05:19 < dongs> if youre using different voltages 2013-02-11T02:05:21 < dongs> etc 2013-02-11T02:05:34 < Thorn> cyclones have some restriction on the number of outputs per bank or something 2013-02-11T02:05:45 < upgrdman> hmmm ok 2013-02-11T02:05:54 < Thorn> someone in ##fpga ran into it some time ago 2013-02-11T02:06:03 < emeb> yay cyclone 2013-02-11T02:15:30 < Robint91> dongs, this http://www.sioi.com.au/shop/product_info.php/cPath/30_24/products_id/48 lenght matching, my ass 2013-02-11T02:16:55 < emeb> heh 2013-02-11T02:17:22 < emeb> I've often wondered how critical that is. 2013-02-11T02:17:36 < emeb> no personal experience w/ DRAM 2013-02-11T02:18:30 < emeb> I have done some ~250MHz FPGA to FPGA and FPGA to ADC/DAC interfaces and we did length match those. 2013-02-11T02:19:06 < emeb> Never had any big problems on other fast ADC boards that were _not_ matched at similar speeds though. 2013-02-11T02:19:33 < Robint91> emeb, it all depends on the speed 2013-02-11T02:19:49 * Robint91 goes to sleep 2013-02-11T02:19:51 < Robint91> bye all 2013-02-11T02:19:54 < emeb> nn 2013-02-11T02:21:03 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Read error: Connection reset by peer] 2013-02-11T02:21:14 -!- jef79m_lurking is now known as jef79m 2013-02-11T02:22:09 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:e408:4c7a:8313:ef63] has joined ##stm32 2013-02-11T02:25:03 < dongs> hmmmpf 2013-02-11T02:26:14 < dongs> rofl, that sioi board made wiht gEDA 2013-02-11T02:26:21 < dongs> I thought it looked like ass 2013-02-11T02:26:38 < dongs> and that explains lack of length matching :) 2013-02-11T02:26:54 < gxti> other than length matching it looks "okay" 2013-02-11T02:27:29 < dongs> wahts bandwidht of say pc100/pc133 sdram 2013-02-11T02:27:33 < dongs> as in megabytes/sec 2013-02-11T02:29:34 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has left ##stm32 ["Konversation terminated!"] 2013-02-11T02:29:38 < dongs> 100mhz*8bytes = 800meg/sec? 2013-02-11T02:29:59 < karlp> heh, new toys, sif bothering with uclinux on f4: http://3cmade.en.alibaba.com/product/705655886-214330059/openwrt.html 2013-02-11T02:30:27 < dongs> is that some asseros shit 2013-02-11T02:30:40 < karlp> ya mon 2013-02-11T02:31:00 < karlp> mips all the way down 2013-02-11T02:31:08 < karlp> who cares as long as gcc makes it go? 2013-02-11T02:31:15 < dongs> the gcc part 2013-02-11T02:31:29 < karlp> gcc in, gcc during, gcc after, not caring :) 2013-02-11T02:31:38 < karlp> I'm not even actually choosing the compiler, 2013-02-11T02:31:51 < karlp> I go "menu, pick board, press go, get image" 2013-02-11T02:31:58 < karlp> who cares what compiler it is? 2013-02-11T02:32:46 < karlp> speaking of gnu, how do I make "x" in gdb print bytes as unsigned rather than signed? 2013-02-11T02:33:21 < karlp> x /26 *(unsigned char*)&kp gives me negaive shits. 2013-02-11T02:36:51 < dongs> im sure some awesome mile-long python macro will helpy ou 2013-02-11T02:37:05 < Thorn> x/26x or x/26u maybe 2013-02-11T02:37:30 < karlp> thanks thorn, I though the casting to unsigned char was enoguh, 2013-02-11T02:37:48 < karlp> x /26xu (void*)&kp 2013-02-11T02:37:53 < karlp> is enough :) 2013-02-11T02:39:39 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-11T02:41:55 < baird> E14 have sent an order confirmation for those 30 cent Renasis boards. Looks like I'll be getting them.. (and no price changes on the docs) 2013-02-11T02:42:24 < karlp> good to have more places to collect dust 2013-02-11T02:42:36 < karlp> they can hang out and party with zyp's transmeta boards! 2013-02-11T02:42:40 < upgrdman> 30 cent Renasis boards? 2013-02-11T02:42:58 < gxti> renesux 2013-02-11T02:42:59 < baird> I don't have to take shit about hording from you people... 2013-02-11T02:43:03 < dongs> baird: LOL 2013-02-11T02:43:23 < baird> upgrdman: http://au.element14.com/jsp/search/productdetail.jsp?SKU=2253395 2013-02-11T02:43:41 < karlp> baird: yes. yes you do. 2013-02-11T02:43:46 < dongs> good stuff. 2013-02-11T02:43:47 < Thorn> I've got 2 TI launchpads, msp430 and lm4f. guess what I did with them 2013-02-11T02:43:58 < dongs> i got 2 lm4f launchnpads 2013-02-11T02:44:00 < upgrdman> lol 2013-02-11T02:44:04 < dongs> one is still in box 2013-02-11T02:44:07 < dongs> !! 2013-02-11T02:44:09 < karlp> baird: you're the one with the c64! 2013-02-11T02:44:42 < gxti> THERMITE 2013-02-11T02:44:46 < gxti> solves every problem 2013-02-11T02:45:12 < gxti> including excessive electronic crap, excessive stock of powdered aluminium oxide, and having a home 2013-02-11T02:45:14 < upgrdman> have an unwanted erection? THERMITE! 2013-02-11T02:45:45 < dongs> so iguess even PC100 sdram will be plenty for my shit 2013-02-11T02:46:03 < dongs> and wont need length matching. so will just hook that up to spartan6 and try dicking with it 2013-02-11T02:46:17 < gxti> dickdickdick 2013-02-11T02:46:35 < dongs> max input data rate to the stuff is liek ~40meg/sec and its only gonna be used as buffer/fifo type stuff. 2013-02-11T02:46:38 < dongs> sholud be no problem. 2013-02-11T02:47:25 < dongs> hmm wifi on nexus4 broken again 2013-02-11T02:47:26 < dongs> what the fuck 2013-02-11T02:47:36 < dongs> fucking lunix 2013-02-11T02:48:46 < dongs> how do you feel knowing that when ytou slide the wifi on/off button into on position, some script runs /bin/bash with a mile-long commnad line to do something wiht wpa_supplicant or some other obscure lunix utility? 2013-02-11T02:49:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-11T02:50:00 < gxti> nobody cares 2013-02-11T02:51:36 < dirty_d> dongs, what would you rather happen? 2013-02-11T02:51:43 < dirty_d> it would explode? 2013-02-11T02:51:51 < gxti> he would rather the windows login sound plays 2013-02-11T02:51:52 < dongs> dirty_d: shell scripts, in 2013? 2013-02-11T02:51:55 < gxti> and then he's in windows 2013-02-11T02:52:05 < gxti> being fucked in the ass by steve baller 2013-02-11T02:52:23 < gxti> ballmer*, not that baller is inapproprate 2013-02-11T02:52:35 < dongs> so much hate for shit that WORKS. 2013-02-11T02:52:39 < dongs> i've ne ver had problems with wifi on windows. 2013-02-11T02:52:54 < baird> Still preferable to it being controlled by Java... which is probably the actual case. 2013-02-11T02:53:16 < gxti> well, does the "shell script" work? 2013-02-11T02:53:26 < Thorn> dongs can find an excuse to rage at linux even when using an fpga 2013-02-11T02:53:44 < dongs> gxti: obviously not 2013-02-11T02:53:46 < Thorn> for the record I had many wifi problems on windows 2013-02-11T02:53:47 < dongs> as my phone currently has no wifi 2013-02-11T02:53:49 < dongs> until i reboot it 2013-02-11T02:53:49 < gxti> k 2013-02-11T02:53:52 < gxti> good luck with that 2013-02-11T02:54:13 < gxti> i don't use wifi :p 2013-02-11T02:54:38 < dongs> thats bvecause your phone internet is probably faster than 64kbit 2013-02-11T02:55:30 < gxti> my phone doesn't get the "internet" 2013-02-11T02:56:30 < karlp> what was that? you wanted some stupid shell scripts? https://github.com/cesarbs/shnake 2013-02-11T02:58:01 < dirty_d> dongs, there is systemd 2013-02-11T02:58:06 < dirty_d> im using it now 2013-02-11T03:00:27 < qyx_> dongs: do you want to see windows dying while trying to connect to simple 802.1x wifi? 2013-02-11T03:01:19 < qyx_> it is shame cos it works out of box even on lunex 2013-02-11T03:04:51 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:e408:4c7a:8313:ef63] has quit [Ping timeout: 256 seconds] 2013-02-11T03:04:57 -!- Roklobsta_ [~quassel@2001:44b8:41d3:6b00:e408:4c7a:8313:ef63] has joined ##stm32 2013-02-11T03:05:01 < baird> I don't think I've ever seen a Windows7 box have faultless wifi.. 2013-02-11T03:06:32 < gxti> that's because every vendor ships their own shitty thing 2013-02-11T03:07:02 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-11T03:07:36 < dirty_d> i gotta learn how to use gdb 2013-02-11T03:07:50 < qyx_> without shell scripts apparently 2013-02-11T03:13:24 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 264 seconds] 2013-02-11T03:24:40 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-11T03:35:33 < dirty_d> how do you get watchpoints to work? 2013-02-11T03:35:44 < dirty_d> isnt it supposed to break when the variable changes? 2013-02-11T03:35:47 < dirty_d> it doesnt 2013-02-11T03:36:01 < zippe1> dirty_d: you can only watchpoint memory, not variables 2013-02-11T03:36:02 < dirty_d> but i can do `print var` and it does show the value 2013-02-11T03:36:18 < zippe1> That's because gdb knows where the current value of the variable is 2013-02-11T03:36:26 < zippe1> It's probably in a register, not in memory 2013-02-11T03:36:32 < dirty_d> ahhh 2013-02-11T03:37:30 < dirty_d> so then there is no way to do it? 2013-02-11T03:37:47 < zyp> to do what? 2013-02-11T03:37:57 < dirty_d> zyp, if i have a variable i want to watch 2013-02-11T03:38:04 < dirty_d> and i dont know if its in a reg or mem 2013-02-11T03:38:12 < zyp> what do you want to watch for? 2013-02-11T03:38:27 < karlp> "break methodnaame if (argument_name == blah)" works for me. 2013-02-11T03:38:37 < dirty_d> lets say its a counter and i want to see when it gets to 10 2013-02-11T03:38:46 < karlp> break filename:linenumber if (variable == blah) works for me too 2013-02-11T03:38:51 < dirty_d> hmm, ok 2013-02-11T03:38:54 < zyp> like karlp said 2013-02-11T03:38:59 < zyp> conditional breakpoints 2013-02-11T03:38:59 < dirty_d> neat 2013-02-11T03:39:08 < dirty_d> i cant believe im just learning this all just now 2013-02-11T03:39:15 < dirty_d> this is incredibly useful 2013-02-11T03:39:16 < dirty_d> lol 2013-02-11T03:39:24 < karlp> dirty_d: don't sweat it, I had no idea either until recently. 2013-02-11T03:39:35 < karlp> we all learn, every day. 2013-02-11T03:39:50 < karlp> sometimes, very rarely, if we're lucky, we get to teach zyp something ;) 2013-02-11T03:40:12 < karlp> maybe one day, if the stars align well, I'll teach zippe something! ;) 2013-02-11T03:40:14 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Remote host closed the connection] 2013-02-11T03:40:15 < zyp> I can admit that I've never used conditional breakpoints myself 2013-02-11T03:40:35 < zyp> I know that they exists, but syntax doesn't sit in the fingers yet :p 2013-02-11T03:40:50 < dirty_d> that worked 2013-02-11T03:40:54 < dirty_d> and its awesome 2013-02-11T03:42:38 < zyp> karlp, you guys are teaching me new stuff every day by having me figuring out your problems ;) 2013-02-11T03:42:40 < dirty_d> break 24 if varname==something works too 2013-02-11T03:42:54 < dirty_d> zyp, lol true 2013-02-11T03:43:08 < dirty_d> where 24 is a line number 2013-02-11T03:43:23 < dirty_d> i dunno how it knows i mean main.cpp though 2013-02-11T03:43:27 < karlp> zyp: hehe, you are impressive with your remote debugging :) 2013-02-11T03:44:02 < karlp> I felt so impotent today though, reading the backlog and seeing people talk about clearing interrupt flags at the end of the routine vs the beginning 2013-02-11T03:44:45 < karlp> I had written about it back in http://false.ekta.is/2012/06/interrupt-service-routines-double-firing-on-stm32/ 2013-02-11T03:44:55 < karlp> and just wasn't in the right channel at the right time 2013-02-11T03:45:05 < karlp> (was too busy drinking beer and .....) 2013-02-11T03:45:19 < zyp> :) 2013-02-11T03:45:33 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-11T03:48:01 < dirty_d> karlp, you saw why that was happening right? i cant remember 2013-02-11T03:48:39 < upgrdman> hmmm. i always clear the flaq as the last instruction and don't have problems. but im not using a lib. just, for example, TIM1->SR &= ~TIM_SR_UIF; 2013-02-11T03:48:57 < dirty_d> upgrdman, thats what i was doing and it wasnt working 2013-02-11T03:49:06 < upgrdman> weird 2013-02-11T03:49:36 < dirty_d> upgrdman, it was because the actual write to memory wasnt finished before the ISR returns 2013-02-11T03:49:56 < dirty_d> so it immediately runs again 2013-02-11T03:49:57 < upgrdman> i believe you, but i wonder why i didnt have problems. 2013-02-11T03:50:01 < dirty_d> im not sure 2013-02-11T03:50:10 < zyp> compiler version, compiler settings, etc… 2013-02-11T03:50:12 < dirty_d> was it compiled with -O2 2013-02-11T03:50:19 < zyp> and hardware in queston 2013-02-11T03:50:24 < dirty_d> upgrdman, i bet thats not actually the last instruction in the ISR 2013-02-11T03:50:38 < zyp> not all peripherals have the same amount of latency 2013-02-11T03:50:41 < upgrdman> well i use gcc. but dongs mod'd my code so he could use the $$ ide he uses... and i dont think he had problems 2013-02-11T03:50:44 < dirty_d> like if you disassembled it 2013-02-11T03:51:00 < upgrdman> dirty_d, maybe. ok 2013-02-11T03:51:02 < dirty_d> in mine it was the very last instruction 2013-02-11T03:51:14 < dirty_d> when i compiled it with -O0, it wasnt and it worked fine 2013-02-11T03:54:55 < dirty_d> is tehre a way to do like a `print TIM1->ISR` without looking up the address? it doesnt work since thats a macro 2013-02-11T03:56:55 < zyp> um 2013-02-11T03:57:04 < zyp> try compiling with -gdwarf2 2013-02-11T03:57:17 < karlp> dirty_d: no, I just remember having had _exactly_ the same problem, and I'd worked ou tthe solution (with help from others on ##stm32 and a bit of experimentation) and felt bad when I read the (lengthy) backlog and couldn't jus tsave you all some time 2013-02-11T03:57:19 < zyp> or -gdwarf-2, or whatever that setting was 2013-02-11T03:57:32 < zyp> I think it's supposed to also include macros then 2013-02-11T03:58:05 < karlp> (use -ggdb3= 2013-02-11T03:58:05 < zyp> but I haven't tried that myself, I don't have those as macros in my own code 2013-02-11T03:58:08 < karlp> (use -ggdb3) 2013-02-11T03:58:22 < karlp> like -g, but with moah. 2013-02-11T03:58:23 < zyp> ah, maybe that was the one 2013-02-11T03:59:05 < dirty_d> ahh cool 2013-02-11T03:59:08 < dirty_d> thanks 2013-02-11T03:59:17 < upgrdman> http://gcc.gnu.org/ml/gcc-help/2009-02/msg00130.html 2013-02-11T03:59:31 < upgrdman> "-ggdb3 produces extra debugging information, for example: including macro 2013-02-11T03:59:31 < upgrdman> definitions." 2013-02-11T03:59:35 < gxti> hrm... so if i use lwip in single-thread, NO_SYS mode i go insane trying to implement anything in TCP 2013-02-11T03:59:35 < karlp> http://www.quickmeme.com/meme/3sxpb0/ 2013-02-11T03:59:48 < gxti> but if i use the chibios example for thread mode, latency sucks ass 2013-02-11T03:59:54 < upgrdman> lol 2013-02-11T03:59:57 < karlp> and now, bedthyme 2013-02-11T04:00:13 < gxti> need to merge the tcpip thread and the mac thread, then the critical stuff is all in one thread 2013-02-11T04:00:16 < dirty_d> later 2013-02-11T04:00:23 < gxti> but i can still do fisher price socket mode in a separate thread 2013-02-11T04:03:46 < gxti> hmm, if i turn on the EXPERIMENTAL OMG DEVELOPER MODE flags it goes fast. what could possibly go wrong? 2013-02-11T04:06:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T04:07:03 -!- BJFreeman [~bjfree@221.sub-75-233-174.myvzw.com] has quit [Quit: had a good time] 2013-02-11T04:09:50 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 255 seconds] 2013-02-11T04:14:10 < karlp> gxti: what exactly is "fisher price" socket mode? 2013-02-11T04:15:14 < gxti> easy mode compared to single-threaded raw api :p 2013-02-11T04:15:26 < gxti> probably netconn (the intermediate layer) rather than bsd sockets 2013-02-11T04:21:53 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-11T04:24:33 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-11T04:38:13 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-11T04:52:30 < dirty_d> when you use a conditional breakpoint theres an overhead right? 2013-02-11T04:52:44 < dirty_d> or is that done in hardware 2013-02-11T04:53:07 < dirty_d> i figured theres a real breakpoint every time and gdb just doesnt report it unless the condition is true 2013-02-11T05:00:54 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-11T05:09:10 -!- Roklobsta_ [~quassel@2001:44b8:41d3:6b00:e408:4c7a:8313:ef63] has quit [Remote host closed the connection] 2013-02-11T05:11:14 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has joined ##stm32 2013-02-11T05:11:41 < timemob> http://www.imgur.com/6asRdKb.jpeg thinking of baird 2013-02-11T05:12:43 < emeb_mac> a classic! 2013-02-11T05:13:13 < timemob> Picked up some new speakers today from the trash 2013-02-11T05:13:40 < emeb_mac> do they work? 2013-02-11T05:13:42 < timemob> If can't use can pawn them off for some bucks 2013-02-11T05:13:44 < timemob> Yes 2013-02-11T05:14:01 < timemob> 3 element 2013-02-11T05:14:19 < emeb_mac> turned down some old speakers that someone is getting rid of. Giant things, no use. 2013-02-11T05:14:29 < timemob> Right 2013-02-11T05:14:50 < timemob> Like big square boxes 2013-02-11T05:14:51 < timemob> With grille 2013-02-11T05:15:45 < emeb_mac> ya - these were big enough to use as tables. Nowhere to put 'em 2013-02-11T05:15:55 < timemob> Lo 2013-02-11T05:16:04 < timemob> Tunnel time 2013-02-11T05:16:07 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has quit [Client Quit] 2013-02-11T05:17:58 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has joined ##stm32 2013-02-11T05:18:33 < timemob> Dongs 2013-02-11T05:19:15 -!- jef79m is now known as jef79m_lurking 2013-02-11T05:19:59 < emeb_mac> back from the underground 2013-02-11T05:20:38 < timemob> Haha 2013-02-11T05:21:09 -!- timemob [~dongs@g1-27-253-251-108.bmobile.ne.jp] has quit [Client Quit] 2013-02-11T05:21:20 < emeb_mac> dongs emerges from a tunnel - how freudian. Fellini and his trains have nothing on that. 2013-02-11T05:26:11 < baird> The Performa520 I think was only good for MkLinux.. 2013-02-11T05:26:36 < gxti> hard faultz 2013-02-11T05:26:36 * emeb_mac has a 660AV 2013-02-11T05:26:45 < emeb_mac> with DSP! 2013-02-11T05:29:31 < baird> An LCIII around somewhere. Was a NetBSD system (68000 Unix at 25MHz! neato) until the SCSI HDD stickoned-up. I compiled Gimp-2.2 on it once, and it took a week. 2013-02-11T05:30:02 < baird> I could probably make it boot from a SCSI Zip drive... 2013-02-11T05:31:02 < emeb_mac> oh dear. 2013-02-11T05:31:31 < emeb_mac> those old systems - I'd forgotten how slow they could be. 2013-02-11T05:31:35 < dongs> haha 2013-02-11T05:31:41 < dongs> i remmeber trying to run some *BSD on some dead mac 2013-02-11T05:31:51 < dongs> it wouldnt work properly because tehre was some bug in openfirmware 2013-02-11T05:31:57 < dongs> like clock speed would be shit or something 2013-02-11T05:33:36 < baird> 8/16/25MHz Unix was still preferable and more useful than what else was around. "We have this thing called 'the Internet'..." 2013-02-11T05:34:55 < gxti> btw whoever mentioned x64 windows not loading unsigned drivers, there is of course a way to fix that 2013-02-11T05:35:33 < baird> "ARPANet? That's ssshhhhiiiiittttt. I can dialup a dozen local Fidonet boards running on Amigas and download PORNO .iff files after I give the SysOp some drrrruuugggsss." 2013-02-11T05:35:36 < dongs> thats not a "fix" 2013-02-11T05:38:10 * baird wonders what "FCU RAM" is on the Resenas chips. 2013-02-11T05:43:32 < ds2> wh000000p wh000p 2013-02-11T05:43:39 < ds2> uCLInux works 2013-02-11T05:43:50 < ds2> 2000000 nasty little hacks but it works 2013-02-11T05:48:32 < dongs> cool 2013-02-11T05:48:40 < dongs> figured out fsmc timing shit? 2013-02-11T05:49:03 < ds2> yeah 2013-02-11T05:49:09 < ds2> fsmc problem was easy 2013-02-11T05:49:19 < ds2> ST's appnote 2016 is the problem 2013-02-11T05:49:26 < dongs> haha, no doubt 2013-02-11T05:49:29 < ds2> it cannot work as written, had to fix their crap 2013-02-11T05:49:41 < ds2> Linux has a table to track free memory 2013-02-11T05:49:58 < ds2> it also has a config option for number of banks/nodes 2013-02-11T05:50:17 < ds2> st's notes define 2 nodes and have memory start at 2000 0000 2013-02-11T05:50:37 < ds2> but they configured linux to allow only 1 node so the table to track memory lived in 6800 0100 or so 2013-02-11T05:51:19 < dongs> does lunix use onboard sram at all? 2013-02-11T05:51:23 < ds2> this meant all valid memor locations were located at tablet entries with addresses like 0x6890 0000 which is invalid. I hacked it to have mem start at 0x6800 0000 2013-02-11T05:51:24 < dongs> or do you just map all lunix memory to fsmc 2013-02-11T05:51:39 < ds2> the appnote seems to try to use it but they didn't set it up right 2013-02-11T05:51:48 < ds2> for now, I just ignore it 2013-02-11T05:51:51 < dongs> right, bcause 2000xxx is sram, isnt it 2013-02-11T05:51:57 < ds2> yep 2013-02-11T05:52:16 < dongs> good times 2013-02-11T05:52:20 < dongs> so the board is not a complete loss :) 2013-02-11T05:52:28 < ds2> as provided by ST it cannot possibily work 2013-02-11T05:52:46 < ds2> yep. going to do write ups and pictures in a few weeks 2013-02-11T05:52:53 < ds2> now I got to go back to working on MWC stuff 2013-02-11T05:52:56 < ds2> u going to MWC? 2013-02-11T05:56:00 < upgrdman> whats mwc? 2013-02-11T05:56:30 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 264 seconds] 2013-02-11T05:56:45 < gxti> hmm 2013-02-11T05:56:54 < gxti> i should do a uclinux board too. 2013-02-11T05:57:16 < ds2> it is "trivial" ;) 2013-02-11T05:57:18 < gxti> not as fun as real loonix 2013-02-11T05:57:26 < gxti> but easier than something like am1707 :p 2013-02-11T05:57:36 < ds2> gxti: u going to ELC? 2013-02-11T05:57:43 < gxti> i don't know what that is so no 2013-02-11T05:57:57 < ds2> Embedded Linux Conference in SF 2013-02-11T05:58:07 < gxti> definitely not then 2013-02-11T05:58:25 < ds2> n/m 2013-02-11T05:58:40 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Read error: Connection reset by peer] 2013-02-11T05:58:50 < ds2> do a F4 board :D 2013-02-11T06:02:00 < baird> Time to do some more shopping at E14.. \o/ 2013-02-11T06:03:44 < dongs> haha 2013-02-11T06:03:52 < dongs> baird: when are you picking up those free renesas shits 2013-02-11T06:04:09 < dongs> it says no stock does it mean they send out for them and email you later wehn its available? 2013-02-11T06:05:43 < dongs> embedded lunix conference 2013-02-11T06:05:49 < dongs> sounds like manchild dream event 2013-02-11T06:05:55 < baird> I'm thinking it's a Pre-Order kind of deal. The page mentioned the UK warehouse as the supplier, so 6 days on top of day as well. 2013-02-11T06:06:02 < dongs> ok 2013-02-11T06:06:22 < dongs> sounds like same kinda trash RS pulls 2013-02-11T06:06:34 < dongs> they mark shit in stock, and write in fine print "in UK warehouse" 2013-02-11T06:06:38 < dongs> so it takes 2 fucking weeks to get it to jp 2013-02-11T06:06:46 < dongs> and you cant filter that shit out 2013-02-11T06:06:55 < dongs> as in "i only want shit from local fucking warehouse" 2013-02-11T06:07:28 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-11T06:08:12 < gxti> newark or whatever does that too, but they have a 'direct ship' box you check to filter it out 2013-02-11T06:08:28 < gxti> or some other non-obvious name 2013-02-11T06:08:45 < dongs> i just stopped ordering shit at RS because of that trash 2013-02-11T06:08:55 < dongs> that and hte fact the website is almost fucking unusable 2013-02-11T06:09:02 < dongs> filtering sucks, UI sucks, search sucks, etc 2013-02-11T06:09:12 < dongs> I search for "power supply" it starts off listing with fucking barrel connectors 2013-02-11T06:10:32 < baird> Ha, 10 bux expansion boards for the VLDisco now: http://au.element14.com/kentec-electronics/eb-stm32discovery-lcd/exp-board-lcd-for-stm32vldiscovery/dp/2008817 2013-02-11T06:11:16 < dongs> ha, lcd driver 2013-02-11T06:11:37 < dongs> made by kentec display, for thier lcds. 2013-02-11T06:11:38 < dongs> clever 2013-02-11T06:11:43 < baird> It's not for the F4Disco, like nearly everything else. 2013-02-11T06:11:48 < dongs> right 2013-02-11T06:11:53 < dongs> i think i have a vldisco32 2013-02-11T06:11:57 < dongs> is that the one wiht 7seg lcd on it? 2013-02-11T06:12:58 < baird> No, looks more like a little brother of the F3Disco 2013-02-11T06:16:12 < baird> I don't actually have VL; thought I did.. 2013-02-11T06:32:41 -!- zippe1 [~Adium@173.11.99.161] has quit [Quit: Leaving.] 2013-02-11T06:35:48 -!- zippe [~Adium@173.11.99.161] has joined ##stm32 2013-02-11T06:45:37 < upgrdman> mmmmm so $10 for a 0.1" header -> SMD flat flex connector? 2013-02-11T06:47:07 < baird> backlight driver and such. 2013-02-11T06:48:00 < dongs> im gonna have to charge a bit more for my adapter, I think :) 2013-02-11T06:48:06 < dongs> (dp > flex + backlight driver. 2013-02-11T06:55:45 < jpqyt> Hi, might anyone have experience with TI's BIOS/DSP rtos? i am new to the system and an example code would always hang at SIO_issue()..any suggestions on how to debug such issue? 2013-02-11T06:55:58 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-11T06:56:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-11T07:01:34 < zippe> jpqyt: License precludes use on stm32 8) 2013-02-11T07:01:55 < zippe> (if I had to guess, you've probably failed to turn on the SIO clock( 2013-02-11T07:23:09 < jpqyt> zippe: oh its also used on the stm32? 2013-02-11T07:23:23 < dongs> time to turn on happyhardcore and load dicknplace 2013-02-11T07:24:50 < emeb_mac> crank out the units 2013-02-11T07:25:40 < jpqyt> zippe: the example code should work since it was designed for the c5515 platform. im only porting it to the c5535 platform, which shouldnt change anything OS, should it? 2013-02-11T07:26:44 < dongs> is that evne stm32 related 2013-02-11T07:27:12 < emeb_mac> OT 2013-02-11T07:28:47 < jpqyt> my apologies, I couldn't find a TI DSP related channel, so I thought I'd try here 2013-02-11T07:31:16 < dongs> maybe zipep knows 2013-02-11T07:35:47 < emeb_mac> could also try on #beagle - there's a lot of TI users there and beagleboards have a C6x DSP in them that some folks use. 2013-02-11T07:50:03 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [Read error: Connection reset by peer] 2013-02-11T08:19:12 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-11T08:39:44 -!- claude is now known as Claude 2013-02-11T08:45:04 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-11T08:45:58 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds] 2013-02-11T08:48:20 -!- _BJFreeman [~bjfree@173.sub-75-233-12.myvzw.com] has joined ##stm32 2013-02-11T08:48:52 -!- _BJFreeman is now known as BJFreeman 2013-02-11T08:59:17 < dongs> http://i.imgur.com/c6uGmm9.jpg 2013-02-11T09:00:14 < emeb_mac> wooden quad? 2013-02-11T09:00:25 < emeb_mac> that should be entertaining. 2013-02-11T09:00:38 -!- neuro_sys|off is now known as neuro_sys 2013-02-11T09:00:44 < dongs> probably the best material for it, i was more pointing out hte shape 2013-02-11T09:00:53 -!- Claude is now known as claude 2013-02-11T09:01:13 < dongs> swastikopter 2013-02-11T09:01:27 < emeb_mac> right - how could I have missed that... 2013-02-11T09:01:54 < emeb_mac> but it's backwards. 2013-02-11T09:02:35 < emeb_mac> zzzzzzzzzzz 2013-02-11T09:02:40 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-11T09:06:03 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out] 2013-02-11T09:11:21 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 240 seconds] 2013-02-11T09:27:30 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-11T09:30:33 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds] 2013-02-11T09:39:00 -!- rigid [~daniel@178-26-71-67-dynip.superkabel.de] has joined ##stm32 2013-02-11T09:39:53 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-11T09:58:58 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-11T10:07:59 -!- jon1012 [~jon@AMontsouris-151-1-37-25.w82-123.abo.wanadoo.fr] has joined ##stm32 2013-02-11T10:08:00 -!- jon1012 [~jon@AMontsouris-151-1-37-25.w82-123.abo.wanadoo.fr] has quit [Changing host] 2013-02-11T10:08:00 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-11T10:10:21 < baird> A nice little find today-- a casio f91w knockoff watch in Daiso ($2.80) Small and thin, and appropriate for underneath cuffs of motorcycle jackets. 2013-02-11T10:16:57 < baird> Considering that the jewelers here in AU flog the F91W for $70, and even the online shops want $20 for them.. 2013-02-11T10:20:38 -!- neuro_sys is now known as neuro_sys|off 2013-02-11T10:21:58 -!- neuro_sys|off is now known as neuro_sys 2013-02-11T10:27:38 -!- neuro_sys is now known as neuro_sys|off 2013-02-11T10:38:28 -!- neuro_sys|off is now known as neuro_sys 2013-02-11T10:39:25 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-11T10:42:01 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 244 seconds] 2013-02-11T10:54:12 -!- neuro_sys is now known as neuro_sys|off 2013-02-11T11:17:37 -!- jpqyt [~chatzilla@unaffiliated/gkwhc] has left ##stm32 [] 2013-02-11T11:19:21 < dongs> gonna stick 3 dp adapters into dicknplace after dinner 2013-02-11T11:19:24 < dongs> or maybe before, lets see 2013-02-11T11:19:42 < zyp> sounds nice 2013-02-11T11:19:46 < zyp> one of them mine? 2013-02-11T11:21:47 < dongs> should be 2013-02-11T11:23:16 < R2COM> finally optimized caps for my new power supply and some new parts ... 2013-02-11T11:23:28 < R2COM> almost, 28$ I guess total 2013-02-11T11:23:34 < R2COM> for just poewr 2013-02-11T11:24:51 < R2COM> too much cap. testing and simulation today 2013-02-11T11:27:20 -!- BJFreeman [~bjfree@173.sub-75-233-12.myvzw.com] has quit [Quit: had a good time] 2013-02-11T11:40:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-11T11:50:34 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has joined ##stm32 2013-02-11T11:57:48 -!- neuro_sys|off is now known as neuro_sys 2013-02-11T12:16:27 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-11T12:17:22 -!- Laurenceb_ [~Laurence@host86-179-84-27.range86-179.btcentralplus.com] has quit [Ping timeout: 246 seconds] 2013-02-11T12:27:59 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:e813:7ce0:4c30:9e76] has joined ##stm32 2013-02-11T12:31:16 < dongs> http://www.idealindustries.com/prodDetail.do?prodId=spliceline&div=0&l1=push-in 2013-02-11T12:31:23 < dongs> * Crimp-free, push-in butt splice 2013-02-11T12:31:59 < zyp> «lol, butt» 2013-02-11T12:33:33 < dongs> ya, thought it'd be relevant to this channel 2013-02-11T12:38:50 < baird> push-in butt 2013-02-11T12:39:56 < dongs> baird perks up when these links appear 2013-02-11T12:41:49 < baird> You keep posting them to gain my attention. Whhhyyyyy. 2013-02-11T12:42:51 < dongs> http://i.imgur.com/Cq1c4Li.jpg this is my speaker loot from earlier today 2013-02-11T12:43:03 < dongs> kinda dark, its dark outside. 2013-02-11T12:43:41 < baird> A White Van with speakers inside. Looks legit. 2013-02-11T12:44:14 < dongs> totally 2013-02-11T12:57:58 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-11T12:58:01 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-11T13:07:35 -!- jon__ [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-11T13:09:36 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 245 seconds] 2013-02-11T13:09:48 < dongs> http://www.chinaeds.com/zl/laoli%5CToshiba%5C%E6%97%A5%E6%96%87%5CT%5CTC%5CTC90A73U.pdf 2013-02-11T13:09:52 < dongs> whoa. 2013-02-11T13:09:53 < dongs> talk about totally weierrd shit 2013-02-11T13:11:12 <+Steffanx> china is on a 56k connection again 2013-02-11T13:12:23 <+Steffanx> Hmm, it's japanese. How can that no be 'weierrd'? 2013-02-11T13:12:28 <+Steffanx> *not 2013-02-11T13:15:02 -!- claude is now known as Claude 2013-02-11T13:24:25 -!- Roklobsta [~quassel@2001:44b8:41d3:6b00:e813:7ce0:4c30:9e76] has quit [Remote host closed the connection] 2013-02-11T13:25:21 < zyp> I'm reading «data slice IC» 2013-02-11T13:26:36 -!- Claude is now known as claude 2013-02-11T13:40:12 -!- TitanMKD [~Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-11T13:51:11 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-11T13:51:50 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 240 seconds] 2013-02-11T13:53:12 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T13:54:20 < dongs> k got rid of family faggots, now time to dicknplace. 2013-02-11T13:55:24 < baird> The VIC20 programs I saved to tape /30/ years ago still load (and run). 2013-02-11T14:03:38 < Laurenceb> dongs has a family 2013-02-11T14:03:39 < Laurenceb> wtf 2013-02-11T14:07:08 < dongs> wow nice i got sot23-5 orientation right on the first try 2013-02-11T14:07:14 < dongs> usually i have to rotate them after 1st dicknplace :) 2013-02-11T14:13:43 < Laurenceb> http://24.media.tumblr.com/tumblr_md122d1kcx1qhvghto1_500.jpg 2013-02-11T14:14:51 < Laurenceb> something for dongs 2013-02-11T14:19:59 < baird> 30 year old cassette tapes was here: http://images.4chan.org/g/src/1360585143370.jpg 2013-02-11T14:21:28 < Laurenceb> http://farm3.static.flickr.com/2036/2211449229_536f11f146.jpg 2013-02-11T14:21:37 < Laurenceb> your waifu? 2013-02-11T14:29:22 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-11T14:55:23 < Laurenceb> http://www.youtube.com/watch?v=yL_-1d9OSdk 2013-02-11T15:00:19 < dongs> http://i.imgur.com/SAyNB22.jpg out of teh oven 2013-02-11T15:01:25 < Laurenceb> i was expecting jews 2013-02-11T15:02:25 -!- TitanMKD [~Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-11T15:10:17 < Laurenceb> the outside world is a strange place, and it is hard to say what is real and what is hidden 2013-02-11T15:10:17 < Laurenceb> one thing thats real: you're a virgin 2013-02-11T15:10:17 < Laurenceb> Laurenceb, that's not funny. 2013-02-11T15:10:42 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-11T15:11:13 < dongs> is that a retweet from #chibidongs or somethign 2013-02-11T15:11:21 < Laurenceb> yup 2013-02-11T15:14:47 <+Steffanx> Laurenceb just talked about his life. Not that i care though 2013-02-11T15:18:46 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-11T15:20:23 < Tectu> dongs, nice, whta does it looks like 2013-02-11T15:22:01 < dongs> hm? 2013-02-11T15:22:10 < Tectu> http://i.imgur.com/SAyNB22.jpg out of teh oven 2013-02-11T15:22:19 < dongs> thats how it looks like 2013-02-11T15:22:22 < dongs> what other view do you need 2013-02-11T15:22:50 < Tectu> lol just re-read my post, that makes no sense... 2013-02-11T15:22:56 < Tectu> it was menat to be "those look nice, what are they" 2013-02-11T15:23:37 < dongs> backlight drivers for edp panel 2013-02-11T15:23:50 < gxti> if there's one thing we can all agree on, it is that Laurenceb is not funny 2013-02-11T15:24:07 <+Steffanx> Whaaaa, you know you just said that out loud gxti ? 2013-02-11T15:24:16 <+Steffanx> ( can't agree more though :D ) 2013-02-11T15:24:30 < BrainDamage> wait, was it even open to discussion? 2013-02-11T15:24:46 <+Steffanx> No, we all agree 2013-02-11T15:25:05 <+Steffanx> but we do love him, of course 2013-02-11T15:25:20 < gxti> speak for yourself 2013-02-11T15:26:02 <+Steffanx> Yes boss 2013-02-11T16:06:08 < baird> http://olduse.net/blog/ .."First mention of AIDS on Usenet. (Although not the first time the word appeared there; there was earlier an unfortunately named host named AIDS-UNIX on Usenet.)" 2013-02-11T16:06:48 < dongs> how would you implement a long-press 2013-02-11T16:07:10 < dongs> say trigger exti on button down, then start some timer and if timer fires and button still pressed then do long-press code? 2013-02-11T16:08:08 < Laurenceb> long press is long 2013-02-11T16:08:54 < dongs> blogpress 2013-02-11T16:09:25 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T16:12:25 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-11T16:13:44 < BrainDamage> obligatory xkcd reference: https://sslimgs.xkcd.com/comics/workflow.png 2013-02-11T16:15:29 <+Steffanx> :P 2013-02-11T16:16:23 < jpa-> hmph, the USB DFU bootloader doesn't allow reading the unique device id 2013-02-11T16:16:37 < dongs> which one/ In f3? 2013-02-11T16:17:09 < jpa-> F4 2013-02-11T16:17:44 < jpa-> it checks the memory regions (smart, to avoid busfaults), and the list of allowed regions doesn't have the unique id area 2013-02-11T16:17:59 < dongs> ah. 2013-02-11T16:26:42 <+Steffanx> No secret command to execute instructions to read it jpa- ? 2013-02-11T16:26:55 -!- jon__ [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Ping timeout: 246 seconds] 2013-02-11T16:27:05 < jpa-> Steffanx: ah, of course 2013-02-11T16:27:06 <+Steffanx> Oh 'secret' 2013-02-11T16:27:15 < jpa-> st's bad coding to the rescue 2013-02-11T16:27:36 < jpa-> if you start a read in an allowed region, you can read until the pointer overflows the region to whatever other area you want 2013-02-11T16:34:22 < gxti> \o/ 2013-02-11T16:39:23 < zyp> hah 2013-02-11T16:39:48 < zyp> or you could just upload code to sram and execute it? 2013-02-11T16:41:32 -!- jon__ [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-11T16:44:50 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-11T16:45:19 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-11T16:47:31 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-11T16:51:09 < Laurenceb> http://i.imgur.com/vF44p5N.png 2013-02-11T16:51:25 < Laurenceb> kalman filter running on my F4discovery 2013-02-11T16:51:53 < emeb_mac> gnuplot ftw 2013-02-11T16:51:55 < Tectu> Laurenceb, now you get all the bitches 2013-02-11T16:51:58 < Laurenceb> fitting 2nd order polynomial to pressure vrs position data 2013-02-11T16:52:23 < Laurenceb> correct values should be green =2, blue=0.4 2013-02-11T16:52:33 < emeb_mac> lol Tectu 2013-02-11T16:53:00 < Laurenceb> pink = measured pressure, red= target pressure 2013-02-11T16:53:05 < emeb_mac> too bad there's no such thing as math/DSP groupies 2013-02-11T16:53:16 < Laurenceb> i added a load of noise to target to get some movement 2013-02-11T16:53:17 < Laurenceb> lol 2013-02-11T16:53:46 < BrainDamage> can't find the red 2013-02-11T16:53:48 < Tectu> emeb_mac, they do exist 2013-02-11T16:53:51 < Laurenceb> this is HITL, but i swapped out pressure sensor data 2013-02-11T16:54:04 < Tectu> wtf is HITLR? 2013-02-11T16:54:08 < Laurenceb> BrainDamage: its hidden under pink 2013-02-11T16:54:13 < Laurenceb> lol 2013-02-11T16:54:17 < zyp> Tectu, nazi algorithm 2013-02-11T16:54:20 < Laurenceb> hardware in the loop 2013-02-11T16:54:23 < Tectu> Laurenceb, it would have been wise to choose black instead of read and overlay it 2013-02-11T16:54:26 < emeb_mac> hardware in the loop rad 2013-02-11T16:54:39 <+Steffanx> "red= target pressure" <= red? 2013-02-11T16:54:45 < Laurenceb> yes 2013-02-11T16:54:55 < Laurenceb> its hidden cuz its working so well :P 2013-02-11T16:54:57 <+Steffanx> me no sees red 2013-02-11T16:55:03 < gxti> or because you added all that noise 2013-02-11T16:55:05 < emeb_mac> Steffanx: is not a bull 2013-02-11T16:55:06 < Laurenceb> pink overlays directly over the top 2013-02-11T16:55:16 < BrainDamage> that's due to his excellent choice of overlapping colours 2013-02-11T16:55:23 < Laurenceb> lol 2013-02-11T16:55:34 < zyp> hmm 2013-02-11T16:56:05 < Laurenceb> the triangular profile is there to tune it, then i chucked in a sinusoid 2013-02-11T16:56:43 < zyp> I just thought that emeb_mac was awfully early up today, then I realized that it's getting late and is almost time to home from work 2013-02-11T16:56:43 < BrainDamage> is the noise injection artificial, or actually measured? 2013-02-11T16:56:52 < Laurenceb> artificial 2013-02-11T16:56:52 < zyp> +go 2013-02-11T16:57:09 < BrainDamage> because, damn, that's noisy 2013-02-11T16:57:16 < emeb_mac> :) 2013-02-11T16:57:25 <+Steffanx> Have a good ride zyp 2013-02-11T16:57:26 < emeb_mac> the alarm cat is always on time 2013-02-11T17:01:36 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [] 2013-02-11T17:01:52 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32 2013-02-11T17:01:57 < zyp> Steffanx, I said almost, not leaving entirely yet :p 2013-02-11T17:02:23 <+Steffanx> That doesn't change the fact that you should have a good ride 2013-02-11T17:03:04 < Laurenceb> lmao 2013-02-11T17:04:13 < Laurenceb> http://englishrussia.com/2007/09/19/good-ride/ 2013-02-11T17:04:39 <+Steffanx> :P 2013-02-11T17:05:25 <+Steffanx> Yet another example of GTA:Russia 2013-02-11T17:06:49 -!- Bird|ghosted [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32 2013-02-11T17:07:07 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has quit [Ping timeout: 276 seconds] 2013-02-11T17:07:07 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 276 seconds] 2013-02-11T17:07:07 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 276 seconds] 2013-02-11T17:07:42 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has joined ##stm32 2013-02-11T17:09:17 < Thorn> freertos scheduler crashes, nice 2013-02-11T17:09:51 < Thorn> looks like it tries to remove a task from an empty event list 2013-02-11T17:10:33 < dongs> ur a task 2013-02-11T17:14:12 < emeb_mac> http://www.youtube.com/watch?v=gsYT8YHL-R0 2013-02-11T17:15:23 < emeb_mac> attn dongs: for your cubieboard - http://www.cnx-software.com/2013/02/08/headless-connected-oscilloscope-based-on-cubieboard-or-beaglebone/?utm_source=rss&utm_medium=rss&utm_campaign=headless-connected-oscilloscope-based-on-cubieboard-or-beaglebone 2013-02-11T17:15:27 < emeb_mac> long URL is long 2013-02-11T17:16:37 < jpa-> zyp: sure, but then the uploaded code would have to somehow either get back to bootloader or handle USB protocol by itself 2013-02-11T17:17:01 < zyp> of course 2013-02-11T17:17:25 < zyp> you could probably do a full second-stage bootloader that way 2013-02-11T17:18:48 < zyp> emeb_mac, why would anyone want to use their phone as a scope interface? 2013-02-11T17:19:38 < emeb_mac> zyp: that's not a particularly useful application of the technology 2013-02-11T17:19:57 < emeb_mac> but I could see it being handy in remote data acquisition. 2013-02-11T17:19:58 < jpa-> using a tablet computer as scope interface would make sense 2013-02-11T17:20:04 < zyp> emeb_mac, agreed 2013-02-11T17:20:41 < zyp> card itself looks nice, but I sort of wonder why they just didn't put down a SoC on it 2013-02-11T17:21:01 < emeb_mac> baby steps? 2013-02-11T17:21:23 < emeb_mac> more interested in the scope side of the design than in the SoC 2013-02-11T17:21:38 < jpa-> yeah, all the cruft for the SoC add a lot of development time 2013-02-11T17:22:00 < emeb_mac> not having to dick around with all the details of building a functional computer when they're already available for cheap 2013-02-11T17:22:39 < zyp> but a single board solution would be more manageable than a stack of differently shaped boards 2013-02-11T17:22:51 < emeb_mac> no doubt 2013-02-11T17:23:27 < donigs> http://www.youtube.com/watch?v=6EsoWpTO2qg#! haha 2013-02-11T17:25:48 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-35.pptp.stw-bonn.de] has joined ##stm32 2013-02-11T17:26:10 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-11T17:32:28 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-11T17:32:45 < baird> Top Gear /Korea/??? .. At least it can't fail as badily as TG Australia.. 2013-02-11T17:33:22 < gxti> will it have starcraft? 2013-02-11T17:36:47 < dongs> more like top gayer 2013-02-11T17:37:50 < gxti> more like dramatic voiceovers, hyper-vignetting, and music you've probably never heard of. with cars. 2013-02-11T17:38:04 < Laurenceb> http://www.sickipedia.org/joke/view/1430796 2013-02-11T17:38:30 < baird> Top Gear Japan -- racing K-Cars. 2013-02-11T17:39:30 < baird> Laurenceb: saving that. 2013-02-11T17:39:39 < Thorn> what are meaningful values for NVIC_IRQChannelPreemptionPriority? 0...15 or something else? 2013-02-11T17:39:55 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-11T17:45:18 < dongs> better read the docs 2013-02-11T17:45:24 < dongs> i remember preemeption shit was pretty hardcore stuff 2013-02-11T17:45:34 < dongs> there's some relation to groups also 2013-02-11T17:45:42 < dongs> and if you enable groups , meaning of priority changes 2013-02-11T17:46:08 < Thorn> good summary here (but no reference to stm32 spl) http://www.freertos.org/RTOS-Cortex-M3-M4.html 2013-02-11T17:46:10 < zyp> meaning is the same, just not all priority levels can preempt each other 2013-02-11T17:46:41 < zyp> Thorn, ah, that one looks nice 2013-02-11T17:47:09 < Laurenceb> aiui theres only 4 bits 2013-02-11T17:47:16 < Thorn> one of my interrupts had priority 0 but it doesn't call any freertos functions, so the crash is still unresolved 2013-02-11T17:47:16 < zyp> yep 2013-02-11T17:47:25 < Laurenceb> and you set a mode that divides them between group and pre-emption 2013-02-11T17:47:54 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-11T17:48:01 < Thorn> number of bits depends on an implementation (and $$$ that ARM asks for the license) iirc 2013-02-11T17:48:13 < dongs> lotta bits 2013-02-11T17:48:59 < zyp> you must be doing some pretty hardcore stuff if you even manage to use all four in a meaningful way 2013-02-11T17:50:14 < Thorn> so what's the significance of configMAX_SYSCALL_INTERRUPT_PRIORITY? that doc seems to say that freertos sets BASEPRI to that value to prevent current interrupt from being interrupted in a critical section 2013-02-11T17:50:25 < Laurenceb> http://www.complex.com/rides/2011/11/gallery-the-25-worst-cars-ever-made#26 2013-02-11T17:50:40 <+Steffanx> Laurenceb is bored? 2013-02-11T17:51:07 -!- Bird|ghosted is now known as HTT-Bird 2013-02-11T17:51:31 < dongs> Laurenceb: old 2013-02-11T17:52:56 < dongs> (both link and car) 2013-02-11T17:54:03 <+Steffanx> Meh, my new mouse has some !@#$%ˆ& coil whine-ish sound 2013-02-11T17:54:04 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-11T17:54:19 <+Steffanx> Even opened it up, but as far as i could see there is no coil in it 2013-02-11T17:54:40 <+Steffanx> There is no more annoying sound that that 2013-02-11T17:57:04 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-11T17:57:19 <+Steffanx> Couldnt find the component that makes the sound either. Perhaps I couldnt find it because logitech used this stacked pcbs soldered together with a header 2013-02-11T17:57:26 <+Steffanx> I HATE it when they do that 2013-02-11T17:57:52 < gxti> get with the 3d revolution Steffanx 2013-02-11T17:58:11 < gxti> parts in multiple dimensions 2013-02-11T17:58:25 < dongs> We're supporting Syria's struggle by sending camera equipment + civilian UAVs to allow. protesters to. film crack downs more safely. Anybody willing to DONATE EQUIPMENT or HELP TRANSFER EQUIPMENT to Syria please contact us. We'll be continuing our support in other countries, once ASS+ is out. Contact us+ forward the message to people you think could help. Cheers! 2013-02-11T17:58:49 < gxti> giev stuffs plz 2013-02-11T17:59:49 < Thorn> I'm surprised they haven't donated a few civilian Predators yet 2013-02-11T18:02:06 < Thorn> also if anyone hasn't seen yet http://blog.frankvh.com/2011/12/07/cortex-m3-m4-hard-fault-handler/ 2013-02-11T18:02:29 < dongs> thats if you dont have a real debugger 2013-02-11T18:02:35 < dongs> i.e. the lunix way of life. 2013-02-11T18:03:39 -!- neuro_sys is now known as neuro_sys|off 2013-02-11T18:04:17 < Thorn> or maybe if your devices are used in production 2013-02-11T18:05:05 < jpa-> i have been using code like that for a while 2013-02-11T18:05:24 < jpa-> i also dump the whole RAM to a file on the filesystem so that i can load the crash dump back to debugger 2013-02-11T18:06:18 < Thorn> and display "Segmentation fault - core dumped (yes really)" on the screen 2013-02-11T18:06:19 < jpa-> but yeah, the purpose is mostly to have atleast something to work on if the program crashes for some reason when there is no debugger connected 2013-02-11T18:06:56 < zyp> that's why every real debugger can load a core dump 2013-02-11T18:07:36 < Thorn> can gdb load a non-elf formatted core dump? 2013-02-11T18:07:40 < jpa-> too bad that some arm-none-eabi-gdb builds lack the simulator target.. so loading a crash dump is more difficult on those 2013-02-11T18:07:45 < jpa-> Thorn: yes 2013-02-11T18:08:16 < Thorn> interesting 2013-02-11T18:08:23 < jpa-> restore memory.dmp binary 0x20000000 2013-02-11T18:08:53 < jpa-> i store the registers in a static structure and then run lines like this after loading the dump set $pc=CRASHDATA.PC 2013-02-11T18:09:15 < jpa-> all in a nice little script of course 2013-02-11T18:09:25 < zyp> hmm, that's real neat 2013-02-11T18:10:06 < zyp> and of course, with the logging stuff I use, all the log objects would be included in the dump 2013-02-11T18:11:05 < Thorn> when I was trying to analyze my crash earlier today gdb was complainig "can't access memory at address 0x2c" for a simple local variable 2013-02-11T18:11:17 < Thorn> wtf 2013-02-11T18:11:18 < jpa-> the code at the end of this file uploads the crash dump to web using gprs modem :P https://svn.kapsi.fi/jpa/paatti-chibios/bootloader/mini_gprs.c 2013-02-11T18:11:40 < jpa-> Thorn: maybe corrupted stack pointer 2013-02-11T18:11:53 < Thorn> stack pointer points to ram 2013-02-11T18:12:13 < Thorn> is there a frame pointer register in eabi btw? 2013-02-11T18:12:22 < jpa-> depends on compiler 2013-02-11T18:12:34 < jpa-> gcc sometimes uses $r7, but it depends on optimizations 2013-02-11T18:12:59 < zyp> optimized code won't have it unless it's useful for that particular function 2013-02-11T18:13:32 < Thorn> well it hasn't crashed any more yet, can't check 2013-02-11T18:13:41 < Thorn> btw can you dump RAM from gdb? 2013-02-11T18:13:52 < zyp> of course 2013-02-11T18:13:53 < jpa-> dump memory etc 2013-02-11T18:13:57 < Thorn> so you can save a snaphot for future analysys 2013-02-11T18:14:15 < Thorn> why didn't I know that an hour ago x_x 2013-02-11T18:15:01 < emeb> "The more you know..." (TM) 2013-02-11T18:16:42 < gxti> need to start using that. although last time i tried it it seemed to not work. "hard fault" for me usually means "something's broke but the evidence has already been destroyed" :| 2013-02-11T18:17:29 < emeb> ditto 2013-02-11T18:17:31 < Thorn> why, stack frames are saved unless there's a stack overflow 2013-02-11T18:18:00 < Thorn> simple problems can be diagnosed relatively easily 2013-02-11T18:18:07 < gxti> it works 100% of the time, 40% of the time 2013-02-11T18:18:57 < jpa-> gxti: depending on if you are running on an rtos, you may have to do a few tricks to get into the process stack 2013-02-11T18:19:33 < jpa-> http://koti.kapsi.fi/jpa/stuff/other/stm32-hardfault-backtrace.html 2013-02-11T18:20:08 < dongs> ~When you reference an illegal pointer on STM32 (and possibly other Cortex-M3 processors), it jumps to the hardfault handler. All fine and good, but gdb doesn't quite understand this and gives a rather unhelpful stack trace: 2013-02-11T18:20:12 < dongs> found your problem 2013-02-11T18:20:28 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-11T18:20:51 < trepidaciousMBR> I really can't get gdb going with an STM32F4... 2013-02-11T18:20:54 < jpa-> dongs: do you know of a free debugger that understands the shadowed stack pointer of Cortex-M*? 2013-02-11T18:20:59 < gxti> the problem is that we listen to you at all, dongs 2013-02-11T18:22:22 < zippe> jpa-: "do you know of a free debugger" … you could stop there 2013-02-11T18:22:49 < jpa-> i wonder if he knows of a commercial debugger that does that either 2013-02-11T18:22:56 < jpa-> surely there has to be some 2013-02-11T18:24:13 < zyp> can't really be too hard to add support for when it already understands signal handlers? 2013-02-11T18:24:49 < zyp> just have to examine the EXC_RETURN value in lr to see which stack it was using previously 2013-02-11T18:25:08 < jpa-> yep 2013-02-11T18:25:22 < jpa-> though getting at the PSP without messing up some other register may take a few tricks 2013-02-11T18:25:51 < zyp> no, gdb can read psp 2013-02-11T18:25:56 < jpa-> hum 2013-02-11T18:26:02 < zyp> note how it shows up in «info registers» 2013-02-11T18:26:11 < zyp> so that should not be a problem 2013-02-11T18:26:15 < jpa-> but can it read psp when msp is selected? 2013-02-11T18:26:23 < jpa-> because it didn't with texane stlink back when i tried it 2013-02-11T18:26:27 < jpa-> but things may have changed 2013-02-11T18:26:31 < zyp> of course, «info registers» read both psp and msp 2013-02-11T18:26:41 < jpa-> good 2013-02-11T18:26:51 < zippe> it's not a question of whether "info reg" can read it, but whether the stub can 2013-02-11T18:26:55 < zippe> And the answer is "yes" 2013-02-11T18:27:32 < Thorn> recent gdb versions should show correct stack trace, hardfault or not 2013-02-11T18:27:35 < zyp> http://paste.jvnv.net/view/6kjwj <- grabbed the closest board on my desk and attached 2013-02-11T18:27:40 < dongs> http://www.nyan.cat/retro.php haha im sure old as fuck 2013-02-11T18:27:56 < zyp> psp is 0 since I'm only using msp 2013-02-11T18:28:41 < dongs> in keil i cna just riht click on hardfault handler and choose caller or callee or wahtever and it goes to what caused the shit 2013-02-11T18:28:44 < dongs> so shrug 2013-02-11T18:28:51 < dongs> no need for mile-long python command lines 2013-02-11T18:29:06 < jpa-> nothing is good without python 2013-02-11T18:29:46 < dongs> https://youtube.com/watch?v=5-Ur37XuRzc ledguy is trolling harder 2013-02-11T18:30:15 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 260 seconds] 2013-02-11T18:31:00 -!- RuslanPopov [~rad@ppp-188-65-14-36-bras2.istra.ru] has quit [Remote host closed the connection] 2013-02-11T18:31:36 < trepidaciousMBR> Ah ha, got gdb running, command line at least, no idea what to do with it but it looks plausible 2013-02-11T18:31:45 < Laurenceb> troll dongs with it 2013-02-11T18:33:58 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-11T18:36:05 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 255 seconds] 2013-02-11T18:36:52 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has quit [Quit: Ex-Chat] 2013-02-11T18:43:07 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-11T18:45:17 -!- alexn_ [~alexn@93.104.205.194] has joined ##stm32 2013-02-11T18:46:07 -!- alexn_ [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-11T18:46:22 -!- alexn_ [~alexn@93.104.205.194] has joined ##stm32 2013-02-11T19:00:31 < trepidaciousMBR> how do I get gdb to show me the memory pointed to by a global pointer? 2013-02-11T19:01:32 < trepidaciousMBR> Just trying to work out why chibios hangs, it is in the port_halt() function, so should have something set 2013-02-11T19:02:08 < jpa-> trepidaciousMBR: print *(uint32_t*)0x123456 2013-02-11T19:02:18 < jpa-> or if you have the pointer as symbol, just print *pointer 2013-02-11T19:02:22 < trepidaciousMBR> ah cool 2013-02-11T19:03:01 < trepidaciousMBR> Ah, no symbol in current context 2013-02-11T19:05:10 -!- alexn_ [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-11T19:05:39 < Laurenceb> http://yro.slashdot.org/comments.pl?sid=3451461&cid=42860419 2013-02-11T19:06:03 -!- barthess [~barthess@178.154.7.77] has joined ##stm32 2013-02-11T19:06:55 < emeb> all ur social networking are belong to us 2013-02-11T19:07:07 < Laurenceb> nothing changed there then 2013-02-11T19:08:11 < trepidaciousMBR> My fault - I don't think I have it enabled at compile 2013-02-11T19:08:22 < trepidaciousMBR> I'll recompile and try again, thanks 2013-02-11T19:09:03 < Thorn> wtf, now it's not crashing anymore 2013-02-11T19:14:01 < emeb> whoo-hoo. your done! 2013-02-11T19:14:35 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 260 seconds] 2013-02-11T19:16:22 < Thorn> somehow I doubt it 2013-02-11T19:31:54 < Laurenceb> http://www.zug.com/live/89639/The-Fifty-Shades-of-Grey-Experiment.html 2013-02-11T19:34:50 <+Steffanx> Laurenceb is bored? 2013-02-11T19:35:03 < Laurenceb> lol 2013-02-11T19:35:41 < emeb> "Oh baby, nothing gets my blood racing like poor dental hygiene." 2013-02-11T19:37:18 -!- daku is now known as DaKu 2013-02-11T19:37:54 < Laurenceb> http://www.zug.com/live/86128/Holy-War-Which-Religious-Book-Will-Stop-a-Bullet.html 2013-02-11T19:37:58 < Laurenceb> irl lolled 2013-02-11T19:47:42 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-11T19:48:04 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-11T19:50:59 -!- DaKu is now known as daku 2013-02-11T19:54:00 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-11T20:02:21 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-11T20:08:26 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 252 seconds] 2013-02-11T20:20:14 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving] 2013-02-11T20:22:05 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-35.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-11T20:23:52 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-11T20:39:37 -!- jon__ [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Ping timeout: 246 seconds] 2013-02-11T20:44:11 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 260 seconds] 2013-02-11T20:48:53 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T20:56:51 -!- _BJFreeman [~bjfree@228.sub-75-233-151.myvzw.com] has joined ##stm32 2013-02-11T20:57:31 -!- _BJFreeman is now known as BJFreeman 2013-02-11T21:08:38 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 255 seconds] 2013-02-11T21:12:55 < emeb> DK order arrived, footprints checked, RX ADC boards ordered. 2013-02-11T21:14:08 < emeb> this business of ordering DK on Friday and getting delivery on Monday rules. 2013-02-11T21:14:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-11T21:16:34 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-11T21:31:44 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32 2013-02-11T21:31:46 < Robint91> hi all 2013-02-11T21:33:03 < Robint91> I have a serial datastream, that I reciever with the uart @115200, it always starts with 0xFF 0x00 0xFF 0xFF a predefined length of data 2013-02-11T21:33:26 < Robint91> I'm thinking, use the DMA controller or a interrupt based routine 2013-02-11T21:37:56 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-11T21:42:38 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-11T21:49:20 -!- Laurenceb_ [~Laurence@host86-179-248-139.range86-179.btcentralplus.com] has joined ##stm32 2013-02-11T21:53:59 -!- daku is now known as DaKu 2013-02-11T21:56:33 < Thorn> it still hasn't crashed 2013-02-11T21:57:05 < emeb> it's waiting for you to lose interest. 2013-02-11T21:58:02 < BrainDamage> a watched program never crashes? 2013-02-11T21:58:35 < emeb> zactly 2013-02-11T22:02:11 -!- ntfreak_ is now known as ntfreak 2013-02-11T22:04:02 < Robint91> DMA or interrupt based, to recieve UART data 2013-02-11T22:04:06 < Robint91> ? 2013-02-11T22:05:04 < Thorn> yes 2013-02-11T22:05:04 -!- barthess [~barthess@178.154.7.77] has quit [Quit: Leaving.] 2013-02-11T22:05:28 < Robint91> Thorn, yes? 2013-02-11T22:05:50 < Thorn> both ways work 2013-02-11T22:09:08 < jpa-> Robint91: if you want to react to input immediately (e.g. "someone pressed enter"), use interrupts 2013-02-11T22:09:21 < jpa-> Robint91: if you want to transfer a lot of data and you know the size in advance, use DMA 2013-02-11T22:09:37 < Robint91> jpa-, it is more a device that outputs data is specific method 2013-02-11T22:10:01 < jpa-> for output DMA is easy so use it 2013-02-11T22:10:09 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-11T22:10:30 < Robint91> jpa-, only 0.01% of the time I'm going to send something back 2013-02-11T22:10:55 < jpa-> maybe i fail at parsing your sentences 2013-02-11T22:11:18 < Robint91> jpa-, the stm is only going to recieve 2013-02-11T22:11:34 < Robint91> that isn't almost any communication back 2013-02-11T22:11:57 < gxti> just use interrupts already, geez 2013-02-11T22:12:14 < Laurenceb_> gxti: calm down 2013-02-11T22:12:19 < gxti> NO 2013-02-11T22:27:46 < dirty_d> Robint91, use dma 2013-02-11T22:29:07 < dirty_d> in the transfer complete interrupt you can decide to send a reply 2013-02-11T22:29:12 < Laurenceb_> ive used dma in circular mode to talk to gps modules 2013-02-11T22:31:56 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Quit: Leaving.] 2013-02-11T22:32:29 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-11T22:33:10 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Read error: Connection reset by peer] 2013-02-11T22:34:57 < zyp> dma in circular mode would be a decent option when you don't know the transfer length and still don't want to be interrupted for every single byte 2013-02-11T22:35:47 < Laurenceb_> ive used it as a FIFO before 2013-02-11T22:35:59 < Laurenceb_> using the transfer index register 2013-02-11T22:36:13 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-11T22:42:57 < emeb> circular mode is handy 2013-02-11T22:52:06 -!- johntramp [~john@122-62-203-214.jetstream.xtra.co.nz] has joined ##stm32 2013-02-11T22:53:37 -!- |sark [~|sark@37.247.88.155] has quit [Ping timeout: 244 seconds] 2013-02-11T23:10:00 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-11T23:14:26 < Laurenceb_> http://www.axsem.com/www/micros/ultra-low-power-rf/ax8052f143 2013-02-11T23:14:40 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Excess Flood] 2013-02-11T23:16:43 -!- |sark [~|sark@apt-get.debian.org.ve] has joined ##stm32 2013-02-11T23:18:15 < Laurenceb_> better than silabs O_o 2013-02-11T23:19:38 < Laurenceb_> http://www.axsem.com/www/rfics/narrow-band/ax5043 2013-02-11T23:19:46 < Laurenceb_> they do transceiver only as well 2013-02-11T23:20:26 < zyp> huh, digikey just mailed me with an EOL notice 2013-02-11T23:20:29 < zyp> http://ordering.digikey.com/ObsoletePart.aspx?dkcid=3163&WT.z_email=obsolete&pn=568-9603-ND 2013-02-11T23:21:46 < emeb> that's a surprise! 2013-02-11T23:21:56 < emeb> those parts haven't been around long. 2013-02-11T23:22:10 < zyp> I wonder what the deal is 2013-02-11T23:26:56 < zyp> http://www.nxp.com/products/microcontrollers/cortex_m4/series/LPC4300.html#quickreference <- according to this it's not, but a number of other variants are 2013-02-11T23:28:27 < emeb> oh - just certain packages 2013-02-11T23:28:59 < zyp> doesn't seem like a consistent pattern 2013-02-11T23:29:09 < emeb> looks like no one likes the 208qft 2013-02-11T23:29:16 < emeb> tqfp 2013-02-11T23:29:31 < emeb> err lqpf 2013-02-11T23:29:34 < emeb> lol 2013-02-11T23:30:04 < emeb> spazzy fingers 2013-02-11T23:31:13 < dirty_d> OH MY GOD 2013-02-11T23:31:26 < dirty_d> for the hell of it i was looking up FPGAs 2013-02-11T23:31:34 < dirty_d> http://www.mouser.com/ProductDetail/Altera-Corporation/EP4SGX530HH35C3N/?qs=sGAEpiMZZMvoScKlWpK8TBFFO0FjFVkjATu6blQW04E%3d 2013-02-11T23:31:56 < emeb> heh 2013-02-11T23:32:11 * emeb used some Xilinx parts that were about that much a while ago. 2013-02-11T23:32:51 < dirty_d> what would something that expensive be used for? 2013-02-11T23:32:51 < emeb> actually built a board w/ 2 of 'em like that. Expensive board. 2013-02-11T23:33:10 < emeb> A lot of those are used for rapid prototyping. 2013-02-11T23:33:18 < dirty_d> ahh makes sense 2013-02-11T23:33:22 < dirty_d> time to gtfo 2013-02-11T23:33:25 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-11T23:33:38 < emeb> big boards with lots of those parts on them that you can load VHDL/Verilog designs into for test before building ASICs. 2013-02-11T23:34:15 < Thorn> what is 2 + j in polar form (amplitude only)? sqrt(2^2 + j^2) = sqrt(3), right? 2013-02-11T23:34:33 < emeb> ya 2013-02-11T23:34:46 < emeb> angle is atan(1/2) 2013-02-11T23:34:58 < Thorn> assignment solution says sqrt(5) wtf 2013-02-11T23:35:11 < emeb> right 2013-02-11T23:35:15 < emeb> 2^2 = 4 2013-02-11T23:35:21 < emeb> 1^=1 2013-02-11T23:35:23 < emeb> 5 2013-02-11T23:35:39 < emeb> you don't put the j in the mag calc. 2013-02-11T23:35:54 < emeb> just the component 2013-02-11T23:36:26 < emeb> (sorry - didn't notice the 3 you wrote ) 2013-02-11T23:36:33 < Thorn> d'oh thanks --- Day changed Tue Feb 12 2013 2013-02-12T00:10:43 -!- alexn_ [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-12T00:22:03 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-5.pptp.stw-bonn.de] has joined ##stm32 2013-02-12T00:30:58 -!- dirty_d [~andrew@76.118.112.158] has joined ##stm32 2013-02-12T00:33:52 < Laurenceb_> hmm i now have a new favourite transceiver 2013-02-12T00:34:15 < Laurenceb_> this axsemi stuff is 256byte fifo too 2013-02-12T00:40:00 -!- timemob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has joined ##stm32 2013-02-12T00:40:31 < timemob> Lol zyp did you get obsolete parts notify on that lpc 2013-02-12T00:41:59 < emeb> ewwww - 8052 2013-02-12T00:53:00 < Laurenceb_> yeah 2013-02-12T00:53:22 < Laurenceb_> but they do it as a discrete transceiver 2013-02-12T00:53:27 < zyp> 22:20:25 < zyp> huh, digikey just mailed me with an EOL notice 2013-02-12T00:54:28 < Laurenceb_> attn dongs 2013-02-12T00:54:29 < Laurenceb_> http://www.axsem.com/www/rfics/narrow-band/ax5043 2013-02-12T00:54:38 < Laurenceb_> silabs got pwned 2013-02-12T00:54:59 -!- DaKu [DaKu@dakus.dk] has quit [Ping timeout: 252 seconds] 2013-02-12T00:57:51 < dirty_d> Laurenceb, what kinda stuff would you use that for? 2013-02-12T00:58:10 < Laurenceb_> planes/quadcopters 2013-02-12T00:58:20 < emeb> ah - 5043 stand-alone. 2013-02-12T00:58:23 < emeb> bettar... 2013-02-12T00:58:44 -!- daku [DaKu@dakus.dk] has joined ##stm32 2013-02-12T00:59:24 < dirty_d> Laurenceb, so like long range low data rate? 2013-02-12T01:00:27 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-12T01:02:30 < Laurenceb_> yes 2013-02-12T01:02:37 -!- t1memob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has joined ##stm32 2013-02-12T01:02:47 < Laurenceb_> someone managed 60Km with a si4432 2013-02-12T01:02:56 < Laurenceb_> and that things a bit better... 2013-02-12T01:06:09 -!- timemob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has quit [Ping timeout: 248 seconds] 2013-02-12T01:12:46 < Laurenceb_> ooh the transceiver one has single ended ant 2013-02-12T01:13:41 < emeb> http://www.mouser.com/ProductDetail/AXSEM/AX5043-QFN28/?qs=sGAEpiMZZMuWfa4oKpdyCUrJIeVfcPcg 2013-02-12T01:13:52 < emeb> $8/ea qty1 2013-02-12T01:14:40 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-12T01:15:10 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-12T01:15:29 < t1memob> Huh, 70 to 1050mhz? 2013-02-12T01:16:11 <+Steffanx> ordered some cheap nrf24l01+ modules yet dirty_d ? 2013-02-12T01:16:52 < Laurenceb_> t1memob: id imagine itd want a filter on the front 2013-02-12T01:17:43 < t1memob> You mean to block out shit that's outside of the freq you're using? 2013-02-12T01:19:28 < emeb> looks like the recommended antenna circuit is tuned LC. 2013-02-12T01:19:39 < emeb> probably works pretty narrow band. 2013-02-12T01:20:47 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-12T01:21:51 < Laurenceb_> yeah 2013-02-12T01:22:08 < Laurenceb_> also it has a proper input switch, so no external switch to go for high power 2013-02-12T01:22:33 < Laurenceb_> looks like less board area that si4432 2013-02-12T01:22:52 < Laurenceb_> and much lower current draw 2013-02-12T01:25:27 < dirty_d> 60km 2013-02-12T01:25:36 < dirty_d> on like an air baloon? 2013-02-12T01:26:04 < dirty_d> how far do you figure you could get from inside a building to a car? 2013-02-12T01:26:14 < Laurenceb_> heh 2013-02-12T01:26:25 < Laurenceb_> i managed about a Km on the ground 2013-02-12T01:26:41 < Laurenceb_> with stubby antenni doing 4.8Kbps 2013-02-12T01:28:01 < Laurenceb_> http://maps.google.co.uk/maps?q=surrey+research+park&hl=en&ll=51.240896,-0.608625&spn=0.006676,0.021136&client=ubuntu&channel=fs&oe=utf-8&gl=uk&hq=research+park&hnear=Surrey,+United+Kingdom&t=h&z=16 2013-02-12T01:28:21 < Laurenceb_> woods on left to car park on right 2013-02-12T01:29:15 -!- t1memob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has quit [Quit: Bye] 2013-02-12T01:29:39 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Quit: Ik ga weg] 2013-02-12T01:29:59 -!- dirty_d [~andrew@76.118.112.158] has quit [Quit: Leaving] 2013-02-12T01:30:23 < Laurenceb_> ironically lime microsystems is somewhere in the middle there 2013-02-12T01:30:29 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-12T01:35:37 < dirty_d> Laurenceb_, not sure if i was still connected when i asked about the 80km range conditions 2013-02-12T01:37:16 < Laurenceb_> ah 2013-02-12T01:37:23 < Laurenceb_> that was to a balloon 2013-02-12T01:37:33 < Laurenceb_> rather poor wire on the balloon 2013-02-12T01:37:47 < Laurenceb_> it wasnt professionally made at all - some random ham guys 2013-02-12T01:37:56 < Laurenceb_> then 12dBi yagi on the ground 2013-02-12T01:38:16 < Laurenceb_> its was probably possible to get better range, the xtals were mismatched 2013-02-12T01:38:23 < Laurenceb_> they were using crappy rfm22b modules 2013-02-12T01:38:44 < emeb> heh - mismatched xtals. 2013-02-12T01:38:49 < emeb> how did they manage that? 2013-02-12T01:41:06 < Laurenceb_> well the tolerance is piss poor on that rfm crap 2013-02-12T01:41:20 < Laurenceb_> then it was at about -10C onboard a balloon 2013-02-12T01:41:29 < gxti> needs a gpsdo ;) 2013-02-12T01:41:41 < Laurenceb_> it was rather hard to much about with the tuning on the ground and keep pinging it 2013-02-12T01:41:58 < Laurenceb_> as the temperature was slewing the frequency everywhere 2013-02-12T01:42:17 < gxti> if you've already got a microcontroller you can replace the crystal with a cheap vcxo and get really good frequency stability actually 2013-02-12T01:42:28 < gxti> microcontroller and gps* 2013-02-12T01:45:17 < Laurenceb_> yeah sure 2013-02-12T01:45:20 < Laurenceb_> its not hard 2013-02-12T01:45:37 < Laurenceb_> unless you are an arduino using permanoob 2013-02-12T01:45:41 < Laurenceb_> anyway... 2013-02-12T01:45:44 * Laurenceb_ zzz 2013-02-12T01:45:47 < gxti> N00BZ 2013-02-12T01:45:52 < Laurenceb_> indeed 2013-02-12T01:46:39 < gxti> i should make a ghetto gpsdo just for that sort of thing, just a crappy vcxo and a crappy micro and maybe a PLL chip and you feed it a PPS and it makes your radios not suck 2013-02-12T01:47:06 < gxti> or i guess if i used a stm32 (not necessarily a given) the internal pll could work 2013-02-12T01:47:25 < gxti> depends on what the radio needs 2013-02-12T01:47:32 < gxti> whatever, not going to 2013-02-12T01:50:19 -!- Laurenceb_ [~Laurence@host86-179-248-139.range86-179.btcentralplus.com] has quit [Ping timeout: 244 seconds] 2013-02-12T02:06:33 -!- |sark [~|sark@apt-get.debian.org.ve] has quit [Ping timeout: 276 seconds] 2013-02-12T02:19:21 < dongs> raeg 2013-02-12T02:21:11 < emeb> what's buggin' ya? 2013-02-12T02:21:48 < dongs> nothing yet 2013-02-12T02:22:16 < emeb> sort of pre-emptive rage? 2013-02-12T02:22:27 < dongs> yea. 2013-02-12T02:24:11 < dongs> http://bcas.tv/paste/results/mxcxOF73.html 2013-02-12T02:24:17 < dongs> these are the kind of opensores changes that piss me off 2013-02-12T02:27:58 < emeb> little shit edits that don't do anything beneficial except change the format? 2013-02-12T02:28:28 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-12T02:28:41 < dongs> and actaully make shit less readable 2013-02-12T02:30:30 < karlp> so emeb, you made the front page of imgur: http://imgur.com/gallery/ovx0sG3 2013-02-12T02:31:05 < dongs> i dont get it 2013-02-12T02:41:54 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-5.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-12T02:50:57 < emeb> karlp: troof! 2013-02-12T02:52:04 < emeb> dongs: in Arizona it rarely rains, so 1) oily gunk builds up on roads and 2) people are inexperienced. When it does rain the roads get more slippery than in places where it rains regularly and inexperienced drivers get in a lot of minor accidents. 2013-02-12T02:52:07 -!- Nutter [~Nutter@199-195-151-246.dyn.novuscom.net] has quit [] 2013-02-12T02:52:28 * emeb has been rear-ended a few times because of this. 2013-02-12T02:52:52 < dongs> makes sense. 2013-02-12T02:53:08 < dongs> may be it can be blamed on lunix. 2013-02-12T02:53:15 < emeb> :P 2013-02-12T02:54:15 < karlp> it's actually the same in northern california, wher eit rains all winter, but everyone seems to forget every summer 2013-02-12T02:54:26 < karlp> first autumn rain's a good day to work from home. 2013-02-12T02:54:43 < karlp> same here with first snow days too, everyone still on their summer tires really. 2013-02-12T03:10:36 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-12T03:12:21 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 252 seconds] 2013-02-12T03:12:22 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-12T03:25:46 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-12T03:27:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-12T03:32:55 -!- Nutter [~Nutter@199.195.151.246] has joined ##stm32 2013-02-12T03:38:33 < dongs> Bit 15 SWAP: Swap TX/RX pins 2013-02-12T03:38:35 < dongs> awesome! 2013-02-12T03:40:32 < dirty_d> people change tires with seasons? 2013-02-12T03:41:31 < qyx_> it is obligatory here for example 2013-02-12T03:41:36 < dirty_d> oh, wow 2013-02-12T03:41:59 < BrainDamage> here's mandatory too to get either winter tires or tire chains onboard in the winter period 2013-02-12T03:42:08 < qyx_> yep 2013-02-12T03:42:10 < BrainDamage> because snow can arrive in copious amounts 2013-02-12T03:42:28 < dirty_d> where? 2013-02-12T03:42:36 < BrainDamage> I am in northern italy 2013-02-12T03:42:39 < dirty_d> ahh 2013-02-12T03:42:45 < qyx_> slovakia 2013-02-12T03:43:10 < qyx_> there was ~70cm of new snow last week in some parts 2013-02-12T03:43:34 < BrainDamage> 30cm snowed here just today 2013-02-12T03:43:38 < BrainDamage> more will arrive tomorrow 2013-02-12T03:43:50 < dirty_d> i got hit with that blizzard friday 2013-02-12T03:43:52 < dirty_d> i hate snow 2013-02-12T03:44:29 < dirty_d> i must have seen at least 15 cars get stuck outside my house over the weekend 2013-02-12T03:44:39 < dirty_d> they did a real shitty job of plowing 2013-02-12T03:44:42 < dongs> what instruction set is A8? ARMv7, so does it do both arm and thumb2? 2013-02-12T03:44:58 < dongs> right, looks like i t 2013-02-12T03:44:59 < dongs> hmm 2013-02-12T03:49:54 < dongs> holy shit 2013-02-12T03:49:55 < dongs> emeb: beep 2013-02-12T03:49:57 < dongs> AN2606 is released 2013-02-12T03:50:55 < dongs> The bootloader embedded in the STM32F37xxx devices supports three serial interfaces: 2013-02-12T03:50:59 < dongs> USART1, USART2 and DFU (USB). 2013-02-12T03:51:01 < dongs> haha. 2013-02-12T03:54:01 < qyx_> ok, another project, rfid door opener with ldap client :S 2013-02-12T04:08:55 < emeb> dongs: boop 2013-02-12T04:09:16 -!- Gargantuasauce_ [~Gargantua@blk-222-208-237.eastlink.ca] has quit [Remote host closed the connection] 2013-02-12T04:09:25 < dongs> status of laughing at officially released an2606 2013-02-12T04:09:26 < emeb> lol - wonder if they'll ever put out errata about the F373 fuckup 2013-02-12T04:09:42 < dongs> STill doing it wrong. 2013-02-12T04:11:02 < upgrdman> what fuckup? 2013-02-12T04:11:16 < dongs> the part where it doesn't do USB bootloader. 2013-02-12T04:11:30 < dongs> despite many claims of it doing so. 2013-02-12T04:11:37 < upgrdman> o 2013-02-12T04:11:40 < dongs> invarious datasheets, refrence manuals, and now boot methods appnote. 2013-02-12T04:12:26 < gxti> the monkey running the programming station was on strike that week so they had to bring in a ferret 2013-02-12T04:14:18 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Ping timeout: 245 seconds] 2013-02-12T04:19:16 < emeb> and you know how unreliable ferrets are... 2013-02-12T04:20:22 < upgrdman> always weaseling out of things 2013-02-12T04:33:27 < dongs> Simplify and automate your virtualization deployment, management and optimization with VMware Go Pro. 2013-02-12T04:33:31 < dongs> haha 2013-02-12T04:33:43 < gxti> ... 2013-02-12T04:33:57 < gxti> good job vmware marketing department 2013-02-12T04:43:17 < dongs> http://www.jst-mfg.com/product/detail_e.php?series=287 I wonder if tehy make the plug available in PCB mount style 2013-02-12T04:43:41 < dongs> guessing: no 2013-02-12T04:56:38 -!- cjbaird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-12T04:56:51 -!- baird [~cjb@ppp121-44-159-202.lns20.syd7.internode.on.net] has quit [Ping timeout: 276 seconds] 2013-02-12T05:24:58 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has joined ##stm32 2013-02-12T05:27:03 < emeb> so, got one of these for giggles: http://www.amazon.com/gp/product/B000HNHA12/ref=oh_details_o00_s00_i00 2013-02-12T05:27:13 < emeb> seems to work reasonably well. 2013-02-12T05:27:48 < emeb> si470x FM->USB chipset. 2013-02-12T05:28:10 -!- Gargantuasauce_ [~Gargantua@24.222.208.237] has joined ##stm32 2013-02-12T05:28:24 < emeb> and... supported under linux. :) 2013-02-12T05:29:52 < emeb> but looks like SI doesn't make the USB version of the chips any longer. 2013-02-12T05:31:30 -!- Viper168 [~Viper@node57.18.251.72.1dial.com] has joined ##stm32 2013-02-12T05:31:30 -!- Viper168 [~Viper@node57.18.251.72.1dial.com] has quit [Changing host] 2013-02-12T05:31:30 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-12T05:32:24 -!- Nutter [~Nutter@199.195.151.246] has quit [Disconnected by services] 2013-02-12T05:32:25 -!- Nutter` [~Nutter@199.195.151.246] has joined ##stm32 2013-02-12T05:32:40 -!- BJFreeman [~bjfree@228.sub-75-233-151.myvzw.com] has quit [Quit: had a good time] 2013-02-12T05:34:46 -!- CheBuzz- [~CheBuzz@76.164.192.124] has joined ##stm32 2013-02-12T05:34:51 < dongs> wtf is that 2013-02-12T05:35:10 < dongs> usb audio shiz that has fm tuner? 2013-02-12T05:35:12 -!- BusError_ [~michel@81.152.248.48] has joined ##stm32 2013-02-12T05:39:49 -!- Netsplit *.net <-> *.split quits: CheBuzz, |sark, BusError 2013-02-12T05:39:49 -!- CheBuzz- is now known as CheBuzz 2013-02-12T05:40:17 -!- Netsplit over, joins: |sark 2013-02-12T05:46:01 < emeb> ya 2013-02-12T05:56:42 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-12T05:57:28 < dirty_d> eheh, thats pretty neat http://www.youtube.com/watch?v=CWx5ji_sFUw 2013-02-12T05:57:45 < dirty_d> im gonna make a hard drive clock 2013-02-12T05:58:29 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-12T05:58:56 < dirty_d> you ever looked inside a hard drive? 2013-02-12T05:59:06 < dirty_d> and im supposed to believe that was made on this planet 2013-02-12T05:59:08 < dirty_d> right... 2013-02-12T06:00:27 < gxti> ghettofabulous 2013-02-12T06:01:01 < gxti> dunno, i think it would be more rewarding to make an actual motor controller, plus it would last longer than 5 minutes 2013-02-12T06:01:22 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-12T06:01:40 < dongs> wut 2013-02-12T06:01:49 < dongs> hd motor is just fucking brushless motor right 2013-02-12T06:01:54 < gxti> yes 2013-02-12T06:01:57 < dongs> wtf is that guy doing 2013-02-12T06:02:01 < gxti> and like comment guy says, he made a brush 2013-02-12T06:02:08 < gxti> a ghetto brush 2013-02-12T06:02:09 < dongs> haha.. wow retard 2013-02-12T06:02:14 < gxti> it's kind of clever 2013-02-12T06:02:32 < gxti> but if you make anything permanent from that, *then* you're a retard 2013-02-12T06:03:02 < dirty_d> sensorless motor control is pretty tricky 2013-02-12T06:03:12 < dongs> yes but oyu can use anyt random rc esc to drive a hdd motor 2013-02-12T06:03:18 < dirty_d> i wonder if i have everything i need to do it 2013-02-12T06:03:27 < dirty_d> dongs, yea but its gonna sound like shit 2013-02-12T06:03:35 < dirty_d> theyre usually 8kHz 2013-02-12T06:03:38 < dirty_d> the cheap ones 2013-02-12T06:03:55 < gxti> HDs are usually <8krpm :p 2013-02-12T06:04:06 < dirty_d> yea, no reasopn you cant drive it slower though 2013-02-12T06:04:19 < dongs> https://youtube.com/watch?v=ZfG-A5HlDq0 sounds ok to me 2013-02-12T06:04:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-12T06:04:21 < dirty_d> im gonna make a clock like i saw 2013-02-12T06:04:27 < dirty_d> cnc mill the numbers into a platter 2013-02-12T06:04:31 < dirty_d> put leds behind them 2013-02-12T06:04:34 < dirty_d> light them at the right time 2013-02-12T06:04:35 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 255 seconds] 2013-02-12T06:04:59 < dirty_d> cant hear the motor over the prop 2013-02-12T06:05:03 < dongs> right 2013-02-12T06:05:21 < dirty_d> if a clock makes even a slightly annoying sound, it will become extremely annoying very fast since its on all the time 2013-02-12T06:07:06 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 244 seconds] 2013-02-12T06:08:03 < dirty_d> i guess i could actually stick 3 optical switches behind the platter, and glue an encoder wheel to the back of it 2013-02-12T06:08:12 < dirty_d> that way its sensored 2013-02-12T06:20:56 < ds2> leds behind? wonder if it would look better using a presistant of vision setup 2013-02-12T06:21:01 < dongs> http://i.imgur.com/6nIxKOG.jpg 2013-02-12T06:21:37 < ds2> hehe 2013-02-12T06:22:43 < dirty_d> ds2, thats what it is 2013-02-12T06:22:49 < dirty_d> https://www.youtube.com/watch?v=kPdYpST_yoE 2013-02-12T06:25:22 < dongs> ds2: still no pics/ videos of F1 board running lunex? 2013-02-12T06:26:16 < ds2> dongs: sorry, no. debugging another board for MWC 2013-02-12T06:26:30 < dongs> that lunix conf? 2013-02-12T06:26:52 < ds2> no 2013-02-12T06:26:55 < ds2> something else 2013-02-12T06:26:59 < ds2> using a bigger TI chip 2013-02-12T06:27:25 < ds2> have to cut 24 traces and rewire it thanks to TI 2013-02-12T06:27:36 < dongs> awesome 2013-02-12T06:28:13 < ds2> lcd0..24 do not mean r: lcd0...7, g: lcd8...15, b: lcd16...23 2013-02-12T06:29:14 < ds2> good thing this is a 2 layer board :D 2013-02-12T06:30:29 < dongs> haha that digital pov clock is driven by obsolete as fuck 3phase driver chip 2013-02-12T06:30:36 < dongs> last datasheet update : 1994 2013-02-12T06:30:39 < dirty_d> wow 2013-02-12T06:30:52 < dongs> typical tarduino shit, reusing components that have been deader than *BSD 2013-02-12T06:31:17 < gxti> if it was good enough for the 80s it must be good enough forevar 2013-02-12T06:31:30 < esden> well the design decision was probably: "I have that chip I got as a sample somewhere in my drawer, let's use it" :) 2013-02-12T06:31:42 < zippe> Still waiting for a replacement for the CD4501B 2013-02-12T06:31:42 < dongs> for sure, 2013-02-12T06:31:52 < emeb_mac> recycling is good for the planet, mkay? 2013-02-12T06:32:03 < dongs> hey esden do you have highlights/beeps on '3phase', 'driver', 'brushless', etc? :) 2013-02-12T06:32:09 < zippe> TI bought Harris, scanned the datasheet as a PDF, added their standard package and disclaimer pages to the end. 2013-02-12T06:32:40 < esden> dongs: no not really :) 2013-02-12T06:33:08 < esden> was just randomly lurking here and thought I say something. :) 2013-02-12T06:36:26 < dongs> heh 2013-02-12T06:44:30 < dongs> arm/thumb have any good rotateleft instruction for 32bits? 2013-02-12T06:44:55 < dongs> and can I make gaycc emit it without having to resort to inline asm? 2013-02-12T06:55:00 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-12T06:55:10 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-12T07:02:26 < emeb_mac> dongs: rotate right 32-n is the same as rotate left n. 2013-02-12T07:04:25 < dongs> im aware of that 2013-02-12T07:09:50 < zippe> dongs, gcc will emit left/right rotations just fine 2013-02-12T07:10:10 < zippe> And rotating left 32-n is *not* the same as rotating right n 2013-02-12T07:10:18 < zippe> in any general sense. 2013-02-12T07:10:59 < dongs> well 2013-02-12T07:11:02 < dongs> i do return ((val << count) | (val >> (32-count))); 2013-02-12T07:11:04 < dongs> on intel 2013-02-12T07:11:11 < dongs> and it does it properly (with vstudio compiler) 2013-02-12T07:11:12 < gxti> one is part of doing the other in a bigger int iirc 2013-02-12T07:11:16 < dongs> for rotate left uint32 2013-02-12T07:11:19 < gxti> but they're not equivalent in any sense 2013-02-12T07:11:50 < dongs> will it emit proper single instrution for arm w/ ^ that code? 2013-02-12T07:11:50 < gxti> unless it's explicitly a barrel shift 2013-02-12T07:11:51 < zippe> Are you trying to do a circular rotation? 2013-02-12T07:11:55 < dongs> yes 2013-02-12T07:13:34 < zippe> And what's wrong with the output you're getting on ARM? 2013-02-12T07:13:57 < dongs> I haven't looked yet. i would expect it to be single instruction though 2013-02-12T07:14:08 < zippe> It'll probably be ROR 2013-02-12T07:14:32 < emeb_mac> since there is no ROL in ARM/Thumb-2 2013-02-12T07:14:33 < zippe> But the ARM instruction set allows shifts as part of other arithmetic 2013-02-12T07:14:51 < zippe> so you may find the compiler will break the operation up and fold it into other instructions 2013-02-12T07:15:53 < emeb_mac> zippe: when I asserted the equivalence of ROR & ROL above I explicitly meant a rotate, not a shift. 2013-02-12T07:16:19 < zippe> emeb_mac: different vendors mean different things by rotate 2013-02-12T07:16:31 < emeb_mac> orly? interesting. 2013-02-12T07:16:41 < emeb_mac> what are the alternatives? 2013-02-12T07:16:51 < zippe> well, for ARM, consider RRX 2013-02-12T07:17:10 < zippe> Other people would call that ASR 2013-02-12T07:17:54 < zippe> There are 8-bit processors that rotate through a set of registers (so you can do 16-bit rotations, for example) 2013-02-12T07:18:56 < dongs> haha 2013-02-12T07:18:57 < emeb_mac> zippe: you sure about RRX? looking at the ISA definitions it seems like that's rotate right including carry. 2013-02-12T07:19:02 < dongs> fucking gcc just optimized e verything away 2013-02-12T07:19:08 < dongs> since i was doing static stuff. 2013-02-12T07:19:18 < zippe> emeb_mac: no, that's RRX.S 2013-02-12T07:20:04 < emeb_mac> hmm. not seeing that. 2013-02-12T07:20:16 < zippe> If you don't specify S, C is not updated 2013-02-12T07:20:19 < zippe> dongs: 2013-02-12T07:20:19 < zippe> unsigned test(unsigned a, unsigned b) 2013-02-12T07:20:20 < zippe> { 2013-02-12T07:20:20 < zippe> return (a << b) | (a >> (32 - b)); 2013-02-12T07:20:20 < zippe> } 2013-02-12T07:20:23 < emeb_mac> RRX.S just updates N/Z 2013-02-12T07:20:33 < zippe> test: 2013-02-12T07:20:33 < zippe> @ args = 0, pretend = 0, frame = 0 2013-02-12T07:20:33 < zippe> @ frame_needed = 0, uses_anonymous_args = 0 2013-02-12T07:20:33 < zippe> @ link register save eliminated. 2013-02-12T07:20:33 < zippe> rsb r1, r1, #32 2013-02-12T07:20:34 < zippe> ror r0, r0, r1 2013-02-12T07:20:34 < zippe> bx lr 2013-02-12T07:20:46 < zippe> emeb_mac: "RRX can optionally update the condition flags based on the result. In that case, bit<0> is shifted into the carry flag." 2013-02-12T07:21:18 < emeb_mac> zippe: ok. 2013-02-12T07:21:53 < zippe> emeb_mac: my point wasn't "this is what rotate generally means" is wrong - just that there are lots of wrinkles 2013-02-12T07:22:17 < emeb_mac> zippe: no doubt. 2013-02-12T07:22:54 * emeb_mac remembers an architecture where the bit numbering was reversed - bit N was lsb and bit 0 was msb. 2013-02-12T07:23:02 < zippe> dongs: all happy now? 2013-02-12T07:23:06 < ds2> PPPPPPppPPPpppC? 2013-02-12T07:23:07 < ds2> :D 2013-02-12T07:23:35 < emeb_mac> ds2: I think it was the TI9900 2013-02-12T07:24:03 < ds2> emeb_mac: I think the PPCs are like that too or at least switcheable to that 2013-02-12T07:24:27 < emeb_mac> "let's make everything an option!" 2013-02-12T07:24:50 < emeb_mac> bad enough that ARM has changeable endianness 2013-02-12T07:25:15 < gxti> ^ 2013-02-12T07:25:48 < ds2> that apparently isn't always implemented 2013-02-12T07:26:26 < gxti> i thought all of cortex-m* didn't have it 2013-02-12T07:26:34 < emeb_mac> right 2013-02-12T07:26:53 < emeb_mac> small mercy 2013-02-12T07:26:54 < gxti> probably anything that runs arm (vs thumb only) 2013-02-12T07:27:17 < ds2> a m0 with that option would be a neat trick ;) 2013-02-12T07:33:15 < emeb_mac> heh. best korea nukes themselves. 2013-02-12T07:35:20 < dongs> news? 2013-02-12T07:37:31 < emeb_mac> http://www.usatoday.com/story/news/world/2013/02/11/earthquake-north-korea-nuclear-test/1911587/ 2013-02-12T07:37:47 < gxti> dear leader says radiation good for health 2013-02-12T07:37:49 < ds2> emeb_mac: have they confirmed it? 2013-02-12T07:38:09 < emeb_mac> nah - still just going on seismic readings 2013-02-12T07:38:17 < ds2> saw that on BBC 2013-02-12T07:52:39 < dongs> hmm 2013-02-12T07:52:52 < dongs> yeah its doing ror here too. 2013-02-12T07:52:59 < dongs> but with 32-n 2013-02-12T07:53:06 < dongs> yeah, infact exactly as you pasted 2013-02-12T07:53:45 < dongs> and that ends up being 3 instructions. 2013-02-12T08:01:07 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-12T08:05:48 -!- cjbaird is now known as baird 2013-02-12T08:20:21 < emeb_mac> handy phrases translated: http://i.imgur.com/88WcBT1l.jpg 2013-02-12T08:34:43 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-12T08:43:37 -!- Tekno01 is now known as TeknoJuce 2013-02-12T08:44:30 -!- TeknoJuce [~TeknkoJuc@69.158.165.12] has quit [Changing host] 2013-02-12T08:44:30 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-12T08:44:49 -!- fxd0h [~fx@186.18.172.240] has quit [Ping timeout: 246 seconds] 2013-02-12T08:46:10 < jpa-> somehow the 96-bit unique id is not so random as i would expect it to be 2013-02-12T08:46:10 -!- fxd0h [~fx@186.18.172.240] has joined ##stm32 2013-02-12T08:46:19 < jpa-> 0x002d0025 0x32314705 0x32314705 2013-02-12T08:46:25 < jpa-> almost ascii but not quite 2013-02-12T08:46:38 < jpa-> oops 2013-02-12T08:46:50 < jpa-> last one should be 0x32383435 2013-02-12T08:50:09 < dongs> unique ID is NOT random. 2013-02-12T08:50:15 < dongs> there's a part in datasheet that says waht it is 2013-02-12T08:50:22 < dongs> its like lot code + wafer position and shit 2013-02-12T08:50:37 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-12T08:52:32 -!- fxd0h [~fx@186.18.172.240] has quit [Ping timeout: 255 seconds] 2013-02-12T08:58:06 < jpa-> dongs: ok, i'll check 2013-02-12T09:09:33 -!- Nutter` is now known as Nutter 2013-02-12T09:26:27 < dongs> so uh... whoa... 2013-02-12T09:26:33 < dongs> this chinese 10port usb hub is fucking SLOW. 2013-02-12T09:26:42 < dongs> shit that was copying at 5meg/sec before now goes at 700k 2013-02-12T09:26:45 < dongs> consistently 2013-02-12T09:26:55 < dongs> damn... and it didnt look so bad 2013-02-12T09:27:07 < dongs> I wonder if its worth taking it apart, measuring pcb and making my own board with quality hub ICs 2013-02-12T09:27:15 < dongs> to fit into the casing 2013-02-12T09:27:17 < dongs> or im nuts. 2013-02-12T09:30:47 < dongs> whoa,, 4 port usb3 hub IC from TI is only 9 bucks 2013-02-12T09:30:59 < dongs> hmmm 2013-02-12T09:32:47 < dongs> Multi Transaction Translator (MTT) Hub: Four Transaction Translators, One Per Port 2013-02-12T09:32:50 < dongs> nice 2013-02-12T09:34:25 < dongs> retardedproj: on the way 2013-02-12T09:35:04 < emeb_mac> build it into the dp board. :P 2013-02-12T09:38:00 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-12T09:38:05 < dongs> .. no. 2013-02-12T09:39:26 -!- neuro_sys|off is now known as neuro_sys 2013-02-12T09:41:24 < dongs> hmpf 2013-02-12T09:41:29 < dongs> did microchip buy SMSC or something? 2013-02-12T09:41:34 < dongs> or is digikey nuts 2013-02-12T09:56:56 < zyp> they did 2013-02-12T09:57:19 < dongs> really? 2013-02-12T09:57:24 < zyp> http://www.smsc.com/Press/phoenix3bec.html 2013-02-12T09:57:47 < dongs> thats like last years shit 2013-02-12T09:57:52 < dongs> wtf 2013-02-12T09:58:21 < dongs> blah 2013-02-12T10:04:32 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 272 seconds] 2013-02-12T10:08:10 < dongs> i wonder if this chink trash is USB 2.0 FullSpeed lol 2013-02-12T10:08:18 < dongs> 700k sounds about right for that, right? 2013-02-12T10:08:29 < zyp> kilobytes? 2013-02-12T10:08:33 < zyp> yes 2013-02-12T10:09:22 < dongs> hmm 2013-02-12T10:09:26 < dongs> windows says hub is operating at high speed 2013-02-12T10:10:26 < dongs> OH! 2013-02-12T10:10:29 < dongs> hub is operating at full-speed 2013-02-12T10:10:30 < dongs> haha. 2013-02-12T10:11:19 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has quit [Quit: ggh] 2013-02-12T10:13:57 < dongs> http://dx.com/p/usb-2-0-high-speed-10-port-hub-black-65299 2013-02-12T10:14:23 < dongs> refund time 2013-02-12T10:17:47 -!- Laurenceb [~laurence@128.243.75.1] has joined ##stm32 2013-02-12T10:39:45 -!- jef79m_lurking is now known as jef79m 2013-02-12T10:42:44 < Simon--> dongs: this thing seems to be usb2: http://dx.com/p/quality-usb-2-0-4-port-hub-24cm-cable-41270 2013-02-12T10:44:57 < dongs> haha wow just wow this thing is made on phenol single sided pcb 2013-02-12T10:45:13 < dongs> http://i.imgur.com/5ulEMcd.jpg 2013-02-12T10:49:01 < zyp> so, those are four-port hubs, two controlling four ports each, the last controlling those two and the last two ports 2013-02-12T10:50:09 < baird> Looks very 1980s.. 2013-02-12T10:50:09 < zyp> doesn't exactly look like high speed routing 2013-02-12T10:50:25 < dongs> right 2013-02-12T10:50:42 < dongs> fucking chinese garbage lol 2013-02-12T10:54:42 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-12T10:58:53 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-12T11:01:21 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds] 2013-02-12T11:02:08 -!- claude is now known as Claude 2013-02-12T11:27:21 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 252 seconds] 2013-02-12T11:47:03 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-12T11:57:02 -!- Laurenceb_ [~Laurence@host86-179-248-139.range86-179.btcentralplus.com] has joined ##stm32 2013-02-12T12:17:32 -!- Laurenceb_ [~Laurence@host86-179-248-139.range86-179.btcentralplus.com] has quit [Ping timeout: 272 seconds] 2013-02-12T12:28:14 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-12T12:31:09 < dongs> https://youtube.com/watch?v=jq--vrqLNDE#t=160s this guy is like a weird version of Laurenceb 2013-02-12T12:37:06 -!- Netsplit *.net <-> *.split quits: trepidaciousMBR 2013-02-12T12:37:06 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-12T12:43:56 < baird> "So /that's/ Minecraft..?" What fucking shit. 2013-02-12T12:49:37 < karlp> so, I got into the blocking handler, and the bus fault status registers says that bus fault address is valid, and it's a precise error, 2013-02-12T12:49:53 < karlp> andthe bus fault address is 0x80023808 2013-02-12T12:49:54 < karlp> but that' 2013-02-12T12:50:00 < karlp> s way past the end of my program? 2013-02-12T13:00:35 < Thorn> https://www.youtube.com/watch?v=GiI9GIIPM30 "How to add microwave capability to a cat scarer" 2013-02-12T13:00:41 < Thorn> was in related videos 2013-02-12T13:04:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-12T13:08:03 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-12T13:12:58 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-12T13:22:48 < zyp> karlp, and thus the error 2013-02-12T13:23:16 < zyp> BFAR is the address which failed once attempted accessed 2013-02-12T13:23:51 < zyp> I assume this is a chip with 128k flash or less? 2013-02-12T13:24:20 < karlp> yeah, 32k. 2013-02-12T13:24:30 < karlp> just not sure how it was ever accessed 2013-02-12T13:24:38 < zyp> check the backtrace? 2013-02-12T13:24:50 < karlp> bt has setup code, signal handler, blocking handler 2013-02-12T13:24:57 < karlp> it seems tobe openocd rfelated, 2013-02-12T13:25:02 < karlp> if I just power it up, it never happens. 2013-02-12T13:25:15 < karlp> but sometimes monit reset halt, continue fails like this. 2013-02-12T13:25:30 < karlp> trying master oocd with better extended-remote support to use kill and run instead. 2013-02-12T13:25:41 < karlp> if i step through the startup code it always works. 2013-02-12T13:29:28 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-12T13:32:04 < baird> https://youtube.com/watch?v=jq--vrqLNDE#t=160s this guy is like a weird version of Laurenceb ... Reminds me of Kenny Everett.. 2013-02-12T13:36:00 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Ping timeout: 256 seconds] 2013-02-12T13:38:31 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-12T13:40:12 < baird> Awesome, I'm subscribing to his channel. 2013-02-12T13:42:32 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-12T13:46:46 -!- mrcan_ [~mrcan@78.173.84.143] has joined ##stm32 2013-02-12T13:46:46 -!- mrcan_ [~mrcan@78.173.84.143] has quit [Changing host] 2013-02-12T13:46:46 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32 2013-02-12T13:48:27 -!- DaKu_ [DaKu@dakus.dk] has joined ##stm32 2013-02-12T13:49:57 -!- daku [DaKu@dakus.dk] has quit [Ping timeout: 240 seconds] 2013-02-12T13:49:57 -!- DaKu_ is now known as DaKu 2013-02-12T13:50:19 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 240 seconds] 2013-02-12T13:52:36 -!- Claude is now known as claude 2013-02-12T13:54:23 < karlp> so with, "Invalid combination of EPSR and instruction, for reasons other than UNDEFINED 2013-02-12T13:54:26 < karlp> instruction. Return PC points to faulting instruction, with the invalid state. 2013-02-12T13:54:29 < karlp> is "return pc" lr? 2013-02-12T13:54:32 < karlp> or something else? 2013-02-12T13:54:46 -!- jef79m is now known as jef79m_lurking 2013-02-12T13:55:47 < jpa-> with hardfault? it is the stacked PC value 2013-02-12T13:55:52 < jpa-> SP+6 or so 2013-02-12T13:56:52 < zyp> keep in mind that it's 6 words 2013-02-12T13:57:23 < jpa-> every sane person has SP as uint32_t* anyway :) 2013-02-12T13:57:26 < zyp> «x/wx $sp+24» should do the trick 2013-02-12T13:57:38 < zyp> «has»? 2013-02-12T13:57:58 < karlp> 0x20003eb8:0x00000000 ? 2013-02-12T13:58:05 < karlp> wat? 2013-02-12T13:58:30 < jpa-> karlp: if you don't break at the exact start of hardfault handler, your SP may already have been modified 2013-02-12T13:58:31 < zyp> this is of course assuming your hardfault handler has not modified sp yet 2013-02-12T13:58:47 < karlp> I did break blocking handler so I'd try and catch it straight away 2013-02-12T13:58:51 < jpa-> zyp: i usually have the first instruction in hardfault handler take the SP to a uin32_t* variable :P 2013-02-12T13:59:25 < jpa-> karlp: you can also do x /32xw $sp to have a stack dump and look for 0x08xxx addresses :P 2013-02-12T13:59:44 < jpa-> (but really, it should just be visible in backtrace anyway) 2013-02-12T14:00:19 < jpa-> and the first function listed after that is where the fault occurred 2013-02-12T14:00:39 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection timed out] 2013-02-12T14:00:44 < karlp> http://fpaste.org/odr1/ 2013-02-12T14:00:49 < karlp> yeah, i get 0x0000000 there. 2013-02-12T14:01:03 < karlp> only seems to happen running while gdb's connected though :| 2013-02-12T14:01:46 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-12T14:02:06 < jpa-> karlp: do you perhaps call a null function pointer or something? 2013-02-12T14:02:23 < jpa-> i.e. what does modbusinoc.c:333 do? 2013-02-12T14:02:31 < zyp> no 2013-02-12T14:02:48 < zyp> I also considered that, but that should give an INVSTATE usage fault 2013-02-12T14:03:00 < zyp> since that would attempt to switch from thumb to arm 2013-02-12T14:03:39 < jpa-> hm, true 2013-02-12T14:03:45 < karlp> it was an invstate, 2013-02-12T14:03:56 < zyp> I thought you said bus fault 2013-02-12T14:04:13 < karlp> that was a different one :) 2013-02-12T14:04:33 < karlp> that didn't giev me the "return pc" thing I didn't understand 2013-02-12T14:04:59 < zyp> ah, well, calling a zero pointer would do this 2013-02-12T14:05:37 < karlp> hmm, I'll have to look at that. 2013-02-12T14:05:59 < karlp> it's only been happening today, and only while runnign with the debugger. will poke it a bit harder 2013-02-12T14:06:47 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-12T14:07:13 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-12T14:08:14 < jpa-> you are very lucky if you just call a null pointer 2013-02-12T14:09:30 < jpa-> in nuttx there are a lot of 'device operation tables', which are basically structs of function pointers.. then when the pointer to the table is null, the pointers will actually be vector table values; so you get all sorts of crazy bugs like re-entry to reset vector, unexpected interrupt calls that return silently etc. 2013-02-12T14:09:49 < Laurenceb> http://www.google.co.uk/search?client=ubuntu&channel=fs&q=-4^%281%2F4%29&ie=utf-8&oe=utf-8&gl=uk&redir_esc=&ei=AzEaUa6SHeW00QXSqYDgBQ 2013-02-12T14:10:02 < jpa-> now i have set the MPU to forbid 0x0000000 + 1MB, helps a lot :P 2013-02-12T14:10:33 < zyp> jpa-, ouch, that sounds nasty 2013-02-12T14:10:42 < karlp> well, that modbus code is a struct of function pointers, so that might be why I see "signal ahndler" called in the stack trace 2013-02-12T14:11:04 < karlp> so, that sounds pretty good. 2013-02-12T14:11:11 < jpa-> just means "hardfault (or other interrupt) occurred" 2013-02-12T14:12:12 < jpa-> zyp: fortunately i had a relatively simple case of that first so i knew to look for it the next time it occurred :) 2013-02-12T14:20:10 < karlp> huh, yeah, the struct of pointers is getting zeroed out somehow 2013-02-12T14:20:15 < karlp> bad pointer write somewhere else I guess. 2013-02-12T14:20:44 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 252 seconds] 2013-02-12T14:21:00 < zyp> watchpoints to the rescue 2013-02-12T14:21:41 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-12T14:22:44 < karlp> I can use watchpoints on global vars? 2013-02-12T14:22:57 < karlp> I've set them up to break on function params matching, 2013-02-12T14:23:23 < karlp> oh, nvm, will just break at the right place before it tries to use them 2013-02-12T14:23:28 < zyp> hardware watchpoints monitors memory accesses 2013-02-12T14:23:54 < zyp> so unless it's overwritten by DMA, you'll trap whatever is writing it to 0 2013-02-12T14:24:57 < zyp> so global vars are really one of the things watchpoints are most suited for 2013-02-12T14:27:03 < karlp> watch struct.member never hits 2013-02-12T14:27:30 < zyp> umm 2013-02-12T14:27:38 < zyp> watchpoints take memory addrs 2013-02-12T14:27:45 < zyp> did you get that part right? 2013-02-12T14:27:49 < karlp> probably not. 2013-02-12T14:27:56 < zyp> try &struct.member 2013-02-12T14:41:30 < dongs> ARM Blogs 2013-02-12T14:43:51 < zyp> time for some LEG blogs 2013-02-12T14:43:52 < karlp> is that meant to make everything go reallllllly slowly? 2013-02-12T14:44:30 < dongs> if I have execution time in ms (~580ms), and data size (~60megs) is that ~90meg/sec? 2013-02-12T14:44:38 < Laurenceb> http://www.google.co.uk/search?client=ubuntu&channel=fs&q=-4^%281%2F4%29&ie=utf-8&oe=utf-8&gl=uk&redir_esc=&ei=AzEaUa6SHeW00QXSqYDgBQ 2013-02-12T14:44:41 < Laurenceb> i lolled 2013-02-12T14:44:56 < zyp> dongs, ~100, but yea 2013-02-12T14:45:24 < karlp> yeah, ok, openocd is single stepping the cpu. 2013-02-12T14:45:27 < karlp> awesome watch point 2013-02-12T14:45:47 < dongs> Laurenceb: wtf is it you linked that last time 2013-02-12T14:45:47 < dongs> wasnt funny the first time 2013-02-12T14:46:01 < Laurenceb> dongs: equations on google now give pron 2013-02-12T14:46:19 < zyp> karlp, umm, ok 2013-02-12T14:46:29 < zyp> because cortex-m has hardware watchpoints 2013-02-12T14:47:00 < dongs> Laurenceb: its cuz your search-source is ubuntu 2013-02-12T14:47:06 < dongs> google knows lunix faggots never get laid 2013-02-12T14:47:10 < dongs> so it's trying to help you out 2013-02-12T14:48:48 < zyp> dongs, status on my dp board? 2013-02-12T14:49:09 < dongs> zyp: assembled 2013-02-12T14:49:30 < zyp> yeah, saw the pics, just wondering when you'll have time to ship it 2013-02-12T14:49:37 < dongs> whenever 2013-02-12T14:49:52 < dongs> i guess y ou dont care if im still hacking on teh firmware 2013-02-12T14:49:56 < zyp> got addr? 2013-02-12T14:50:02 < dongs> since you're capable of fixing it yourself w/c++ crap 2013-02-12T14:50:13 < zyp> of course, just ship me a blank one 2013-02-12T14:50:13 < dongs> nope, need again 2013-02-12T14:50:20 < dongs> aLSo how much am i charging you for this shit? 2013-02-12T14:50:30 < dongs> ill test it though, to make sure it works 2013-02-12T14:51:00 < zyp> dunno, how about $15? 2013-02-12T14:51:21 < zyp> or what is BOM? 2013-02-12T14:52:01 < dongs> runtime: 2049ms, 26.388035 MiB/sec 2013-02-12T14:52:10 < dongs> ouch 2013-02-12T14:52:20 < dongs> runtime: 569ms, 95.024750 MiB/sec 2013-02-12T14:52:31 < dongs> ^ x86 @ 2ghz, 1st one = A8 @ 1.2ghz 2013-02-12T14:53:33 < zyp> so a bit over 3x difference 2013-02-12T14:53:37 < zyp> that's not too bad 2013-02-12T14:53:59 < dongs> it is, thats not the only thign the processor has to do 2013-02-12T14:55:20 < dongs> O3 didnt make much diff 2013-02-12T14:55:22 < zyp> I mean compared to x86 2013-02-12T14:57:04 < dongs> hmm 2013-02-12T14:57:26 < dongs> i wonder if i ca n somehow use armcc to build one of the files 2013-02-12T14:57:26 < dongs> the EABI shit specifies what? 2013-02-12T14:57:34 < dongs> calling convention? 2013-02-12T14:57:51 < zyp> yeah 2013-02-12T14:58:33 < dongs> one of the files that does all the work is just code, no stdlib functions or anything like that 2013-02-12T14:58:44 < dongs> just calculations, on ints only 2013-02-12T14:59:03 < zyp> as long as the ABI is the same, it should be fine 2013-02-12T14:59:21 < dongs> what is the ABI though 2013-02-12T14:59:49 < jpa-> it's the EABI :) 2013-02-12T14:59:55 < jpa-> so it should work just fine 2013-02-12T14:59:55 < donigs> i see only lunix shit references to it 2013-02-12T15:00:22 < jpa-> basically ARM thumb2 has only one calling convention, so there shouldn't be any problem between compilers for plain C code 2013-02-12T15:00:27 < jpa-> for C++ code i'm not 100% sure 2013-02-12T15:00:41 < dongs> so basically i just build a .o out of it and it'll wokr? 2013-02-12T15:00:46 < dongs> for ARMv7 or whatever 2013-02-12T15:00:49 < zyp> yes 2013-02-12T15:00:58 < dongs> mkay lets see if I can make armcc do that 2013-02-12T15:01:12 < dongs> what about r9 shit 2013-02-12T15:01:26 < zyp> r9 what? 2013-02-12T15:02:05 < donigs> "The role of register r9 is platform specific. A virtual platform may assign any role to this register and must document this usage. For example, it may designate it as the static base (SB) in a position-independent data model, or it may designate it as the thread register (TR) in an environment with thread-local storage. The usage of this register may require that the value held is persistent across all calls. A virtual platform that has no need for such a special reg 2013-02-12T15:02:09 < dongs> some retweet 2013-02-12T15:02:15 < donigs> useless? 2013-02-12T15:03:18 < zyp> dunno 2013-02-12T15:03:24 < zyp> just try 2013-02-12T15:03:51 < jpa-> never heard about r9 :P 2013-02-12T15:03:53 < dongs> hmm crap.. dunno how to make it emit ARMv7 stuff 2013-02-12T15:04:01 < jpa-> i guess it only matters for virtual platforms? 2013-02-12T15:04:10 < zyp> jpa-, it sits between r8 and r10 2013-02-12T15:04:21 < jpa-> zyp: what about r8.5? 2013-02-12T15:04:54 < zyp> is that the upper half of r8? 2013-02-12T15:04:55 < zyp> :p 2013-02-12T15:05:01 < dongs> hmm generate arm code 2013-02-12T15:05:04 < dongs> lets try that 2013-02-12T15:05:11 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-12T15:07:29 < dongs> error: mul.o: Object has vendor-specific contents that must be processed by the 'ARM' toolchain 2013-02-12T15:08:04 < dongs> gaaay 2013-02-12T15:08:04 < dongs> ^ gcc failing 2013-02-12T15:08:37 < dongs> error: main uses VFP register arguments, mul.o does not 2013-02-12T15:09:26 < zyp> hardfloat vs softfloat 2013-02-12T15:09:32 < dongs> --arm_linux 2013-02-12T15:12:05 < dongs> This option configures a set of other options with defaults that are suitable for ARM Linux compilation. 2013-02-12T15:12:06 < dongs> haha 2013-02-12T15:12:06 < dongs> maybe this. 2013-02-12T15:12:06 < dongs> hmm. not in MDK compiler :( 2013-02-12T15:12:23 < jpa-> why not compile everything with armcc? :P 2013-02-12T15:12:37 < jpa-> (hmm, btw, are the relocation stuffs in .o compatible across compilers?) 2013-02-12T15:14:40 < dongs> how do i make gcc output asm 2013-02-12T15:14:49 < jpa-> -S 2013-02-12T15:14:50 < donigs> jpa,cuz the shit is lunix 2013-02-12T15:14:53 -!- dongs_ [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-12T15:15:32 < jpa-> sounds like some open sauces, you should only use closed sauces 2013-02-12T15:15:37 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Read error: Connection reset by peer] 2013-02-12T15:15:43 < jpa-> like windows ce 2013-02-12T15:17:00 < zyp> jpa-, I think it's standardized in ELF 2013-02-12T15:17:14 < jpa-> zyp: ah, true 2013-02-12T15:17:21 < zyp> but that doesn't help if people make vendor specific extensions 2013-02-12T15:22:01 < dongs_> http://bcas.tv/paste/results/MVzWIA24.html 2013-02-12T15:22:19 -!- dongs_ is now known as dongs 2013-02-12T15:22:38 < dongs> is gcc emitting thumb or osmething? 2013-02-12T15:24:16 < zyp> well, the .thumb seems to indicate that 2013-02-12T15:24:27 < donigs> well hwo teh fock do i make it output arm? 2013-02-12T15:24:43 < zyp> but considering how this is UAL, it could possibly be either? 2013-02-12T15:24:56 < zyp> dunno? --arm maybe 2013-02-12T15:25:04 < donigs> thats on armcc. 2013-02-12T15:25:14 < zyp> but this looks like a function where thumb may be faster 2013-02-12T15:25:16 < donigs> arm-linux-gnueabihf-gcc: error: unrecognized command line option '--arm' 2013-02-12T15:25:25 < donigs> i dont know, i wanna try both and see 2013-02-12T15:25:51 < zyp> lots of simple instructions 2013-02-12T15:26:00 < donigs> -marm 2013-02-12T15:26:11 < zyp> ah, of course 2013-02-12T15:27:05 < dongs> runtime: 1937ms, 27.913828 MiB/sec 2013-02-12T15:27:53 < dongs> hurf, not a single difference 2013-02-12T15:27:53 < dongs> runtime: 2059ms, 26.259876 MiB/sec 2013-02-12T15:28:45 < zyp> as I would expect 2013-02-12T15:29:08 < zyp> so it makes sense to select thumb for that 2013-02-12T15:29:13 < donigs> i wonder how hard it would be to wrap armcc generated asm into __asm() for gcc 2013-02-12T15:29:45 < zyp> can't you just save it to an .s-file and input that into gcc? 2013-02-12T15:29:56 < donigs> i doubt it can read the shit properly? 2013-02-12T15:30:08 < donigs> isnt gnu-gas-wahtever shit syntax different from arm? 2013-02-12T15:30:21 < zyp> dunno 2013-02-12T15:31:19 < zyp> the topmost part of your paste is armcc output? 2013-02-12T15:31:29 < dongs> ya 2013-02-12T15:32:19 < zyp> just rewrite line 1 to «core_encrypt:», line 11 to «.L36:» and «BGT .L36» on line 68 2013-02-12T15:32:37 < zyp> maybe also grab the .thumb-stuff over the core_encrypt: 2013-02-12T15:32:56 < dongs> A8 does same set of thumb2 as cortex M3 right? 2013-02-12T15:33:50 < zyp> pretty much 2013-02-12T15:34:14 < zyp> if there are any differences, they shouldn't be relevant for userspace code anyway 2013-02-12T15:36:04 < dongs> http://bcas.tv/paste/results/oQR9qP57.html armcc generated thumb looks weird 2013-02-12T15:36:38 < dongs> also syntax is diff 2013-02-12T15:37:06 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!] 2013-02-12T15:37:34 < zyp> I don't see any significant differences, apart from the jump labels 2013-02-12T15:37:35 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-128.pptp.stw-bonn.de] has joined ##stm32 2013-02-12T15:38:14 < dongs> and STR->STMIA, LDR->LDMIA that sorta shit? 2013-02-12T15:38:23 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-12T15:38:23 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-12T15:38:23 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-12T15:38:26 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-12T15:38:38 < dongs> youre looking at 1st pastee 2013-02-12T15:38:48 < dongs> well actually doesnt matter 2013-02-12T15:39:00 < zyp> huh? 2013-02-12T15:39:05 < dongs> gaycc output is full of 4-letter instructions 2013-02-12T15:39:13 < dongs> and armcc are ll 3 letter. 2013-02-12T15:39:23 < dongs> MOVS, ADDS, etc 2013-02-12T15:39:44 < zyp> 3? 2013-02-12T15:39:54 < dongs> ... 2013-02-12T15:40:08 < dongs> anyway i doubt its gonn make much diff 2013-02-12T15:40:13 < dongs> wiht armcc or gcc 2013-02-12T15:40:16 < dongs> not enough to notice anyway. 2013-02-12T15:40:19 < zyp> agreed, probably won't 2013-02-12T15:40:40 < zyp> probably not worth wasting more time on 2013-02-12T15:40:53 < dongs> well thats pretty damn slow then 2013-02-12T15:40:54 < dongs> hmm 2013-02-12T15:41:13 < karlp> yay, found it, my dma was overrunning memory. 2013-02-12T15:43:00 < karlp> openocd watchpoints don't work at all in the 0.6.1 release, and in trunk, they don't work with gdb's watch command, only if you pass it through a monitor command to oocd itself. 2013-02-12T15:43:08 < karlp> another whopping black mark 2013-02-12T15:43:55 < dongs> would neon have any advantage for this kinda shit? 2013-02-12T15:44:21 < dongs> its just adds/shifts 2013-02-12T15:44:27 < zyp> maybe 2013-02-12T15:44:46 < zyp> I'm not an expert on simd stuff 2013-02-12T15:45:10 < dongs> Instructions perform the same operation in all lanes 2013-02-12T15:45:14 < dongs> i dont think that'll work here 2013-02-12T15:46:15 < zyp> unless you can reorder shit or otherwise work on stuff in parallel 2013-02-12T15:46:32 < dongs> cant 2013-02-12T15:47:31 < dongs> because its CBC stuff 2013-02-12T15:47:37 < dongs> so output of next block depends on current block 2013-02-12T15:49:18 < dongs> hm apparently cbc decryption can be parallelized 2013-02-12T15:49:36 -!- jef79m_lurking [~jef79m@124-168-171-28.dyn.iinet.net.au] has quit [Ping timeout: 245 seconds] 2013-02-12T15:53:32 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-12T15:54:09 -!- jef79m_lurking [~jef79m@124.171.5.86] has joined ##stm32 2013-02-12T15:54:15 -!- jef79m_lurking is now known as jef79m 2013-02-12T15:57:07 < Laurenceb> Canadian Broadcasting Corporation 2013-02-12T16:03:14 < jpa-> dongs: i would think there is parallelism inside every block 2013-02-12T16:04:13 < jpa-> but that depends on the cipher you are using 2013-02-12T16:05:20 < dongs> right, there is, its cbc encryption that cant parallelize because output depends on previous result but cbc decrypt is fine, you can do as many blocks as possible at once and then just xor results w/iv to get plaintext 2013-02-12T16:07:32 < jpa-> ah, right 2013-02-12T16:11:48 -!- jef79m [~jef79m@124.171.5.86] has quit [Ping timeout: 245 seconds] 2013-02-12T16:16:11 -!- alexn [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-12T16:16:40 -!- jef79m [~jef79m@124-169-157-222.dyn.iinet.net.au] has joined ##stm32 2013-02-12T16:17:27 -!- neuro_sys is now known as neuro_sys|off 2013-02-12T16:19:35 -!- Erlkoenig [~erlkoenig@pptp-194-94-199-128.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-12T16:20:00 < dongs> haha 2013-02-12T16:20:09 < dongs> -Os with gcc is fucking TERRIBLE 2013-02-12T16:20:09 < dongs> runtime: 3268ms, 16.545008 MiB/sec 2013-02-12T16:20:09 < dongs> runtime: 1687ms, 32.050434 MiB/sec overclocked to 1.2ghz 2013-02-12T16:20:27 < dongs> runtime: 2002ms, 27.007534 MiB/sec 2013-02-12T16:20:28 < dongs> at 1008 2013-02-12T16:20:28 < dongs> hmm 2013-02-12T16:21:04 < jpa-> is it smaller? :) 2013-02-12T16:21:14 < dongs> -Os? 2013-02-12T16:21:17 < dongs> probably 2013-02-12T16:21:20 < dongs> whats the -O for speed? 2013-02-12T16:21:23 < dongs> -O3? 2013-02-12T16:21:26 < jpa-> -O2 or -O3 2013-02-12T16:21:30 < dongs> with -fun-roll-loops? 2013-02-12T16:21:32 < dongs> right 2013-02-12T16:21:32 < jpa-> -O3 is sometimes slower than -O2 2013-02-12T16:21:50 < jpa-> (because larger code may cause slowness because of cache being full etc.) 2013-02-12T16:23:31 * Steffanx wonders why dongs cares about gcc 2013-02-12T16:24:00 < dongs> how the fuck do i make a graph in wolframalpha 2013-02-12T16:24:01 < dongs> to see mb/mhz 2013-02-12T16:24:01 < dongs> out of 27=1008mhz,32=1200mhz 2013-02-12T16:24:53 < jpa-> dongs: lol, use gnuplot 2013-02-12T16:25:12 < donigs> gnoooplot 2013-02-12T16:25:28 < dongs> i think ill sleep instead. 2013-02-12T16:27:44 <+Steffanx> gn 2013-02-12T16:29:05 -!- _BJFreeman [~bjfree@239.sub-75-196-6.myvzw.com] has joined ##stm32 2013-02-12T16:31:52 -!- _BJFreeman is now known as BJFreeman 2013-02-12T16:32:07 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-12T16:34:36 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-12T16:36:37 -!- mrcan_ is now known as mrcan 2013-02-12T16:46:41 < BJFreeman> http://www.ovt.com/products/category.php?id=26 anyone worked with this 2013-02-12T16:49:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-12T16:52:40 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-12T17:02:30 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-12T17:08:37 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-12T17:11:31 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-12T17:16:16 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Ping timeout: 245 seconds] 2013-02-12T17:16:28 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 272 seconds] 2013-02-12T17:22:54 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32 2013-02-12T17:24:28 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Disconnected by services] 2013-02-12T17:24:50 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-12T17:29:56 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-12T17:31:23 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds] 2013-02-12T17:31:51 <+Steffanx> That's a no BJFreeman 2013-02-12T17:31:59 <+Steffanx> What you want to use em for BJFreeman ? 2013-02-12T17:33:13 < gxti> -Omg-optimized 2013-02-12T17:33:17 < gxti> ^ what the pros use 2013-02-12T17:33:57 < Thorn> -funroll-infinite-lops 2013-02-12T17:34:03 < Thorn> loops 2013-02-12T17:34:21 <+Steffanx> Heh, gxti 2013-02-12T17:34:32 < jpa-> -funroll-whole-program-as-a-lookup-table 2013-02-12T17:34:56 <+Steffanx> -and-generate-the-hdl ? 2013-02-12T17:35:23 < jpa-> no need for HDL for a simple ROM chip 2013-02-12T17:36:04 <+Steffanx> Nah, the schematic for the actually hw with only nands would be ok 2013-02-12T17:36:08 < gxti> gcc tarduino.c -o /dev/3dprinter 2013-02-12T17:36:26 <+Steffanx> :) 2013-02-12T17:36:37 <+Steffanx> How amazing would that be :D 2013-02-12T17:37:04 <+Steffanx> Mechinical GCC outpit 2013-02-12T17:37:12 <+Steffanx> +a -i 2013-02-12T17:37:22 < gxti> probably not very, because it would be a 10mm process instead of 10nm 2013-02-12T17:37:31 < gxti> print to ASIC 2013-02-12T17:37:44 <+Steffanx> I mean, really mechanical. 2013-02-12T17:37:51 <+Steffanx> mm/cm scale 2013-02-12T17:38:16 < gxti> compile your code into a babbage machine? 2013-02-12T17:38:30 < gxti> hand-crank to tweet 2013-02-12T17:38:36 <+Steffanx> Yeah, something like that 2013-02-12T17:39:56 <+Steffanx> Too bad the average 3d printer cant print something like that 2013-02-12T17:41:09 <+Steffanx> I'm sure you get project on hackedgadgets when you make something like that gxti :P 2013-02-12T17:41:32 < gxti> i will never submit my work to hackabutts because my work is not a 'hack' 2013-02-12T17:41:54 <+Steffanx> isn't/was one of those hackaday posters here? 2013-02-12T17:41:59 <+Steffanx> *wasn't 2013-02-12T17:42:19 < gxti> amateurish and ill-advised yes but not a hack, no sir 2013-02-12T17:42:34 <+Steffanx> szczys iirc 2013-02-12T17:51:54 -!- neuro_sys|off is now known as neuro_sys 2013-02-12T17:52:12 < Thorn> it hasn't crashed for 24 hours 2013-02-12T17:57:24 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has joined ##stm32 2013-02-12T17:59:01 <+Steffanx> waow 2013-02-12T17:59:13 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-12T17:59:44 < Laurenceb> what hasnt? 2013-02-12T18:00:05 <+Steffanx> it 2013-02-12T18:00:42 -!- neuro_sys is now known as neuro_sys|off 2013-02-12T18:06:10 -!- emeb [~ericb@72.201.78.226] has joined ##stm32 2013-02-12T18:36:30 -!- barthess [~barthess@178.154.53.6] has joined ##stm32 2013-02-12T19:09:31 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-12T19:16:05 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-12T19:25:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-12T19:41:53 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection timed out] 2013-02-12T19:43:34 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-12T19:46:56 -!- barthess [~barthess@178.154.53.6] has quit [Quit: Leaving.] 2013-02-12T19:47:06 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-12T19:50:34 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 256 seconds] 2013-02-12T19:51:27 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has quit [Ping timeout: 256 seconds] 2013-02-12T19:54:15 -!- barthess [~barthess@178.154.53.6] has joined ##stm32 2013-02-12T19:57:53 -!- barthess [~barthess@178.154.53.6] has quit [Client Quit] 2013-02-12T19:59:42 -!- barthess [~barthess@178.154.53.6] has joined ##stm32 2013-02-12T19:59:52 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-12T20:06:00 < BJFreeman> Steffanx for visual recognition and 3d scanning 2013-02-12T20:07:42 < BJFreeman> want to interface to the camera frame work of the stm32f4 2013-02-12T20:10:15 <+Steffanx> Isn't the dcmi too limited for 16mp? 2013-02-12T20:10:43 < jpa-> ram is the limit 2013-02-12T20:10:53 <+Steffanx> That too 2013-02-12T20:12:06 <+Steffanx> but i meant the transfer rate of the dcmi 2013-02-12T20:12:38 <+Steffanx> 54MByte/s .. which is like 1 16MP frame per second? 2013-02-12T20:13:01 < jpa-> does that sensor even hook up to dcmi anyway? 2013-02-12T20:13:07 < zyp> hmm, my PWM outputs aren't outputting anything, this is using TIM8 on PC6-9 on F4 2013-02-12T20:13:19 < zyp> this is code that used to work before, with TIM2 2013-02-12T20:13:38 < zyp> and I have double checked the AF setup 2013-02-12T20:13:45 < zyp> has anyone seen similar behavior? 2013-02-12T20:14:23 < zyp> I can see the timer itself counting, but neither GPIOC_IDR nor LA on signals show them changing 2013-02-12T20:15:02 < jpa-> just all my stupid mistakes pop to mind, like CCER 2013-02-12T20:15:44 < zyp> CCER is set to 0x1111, like it used to be when running on TIM2 2013-02-12T20:15:47 < jpa-> ah, tim8 is advanced control 2013-02-12T20:15:56 < jpa-> i think there was some gotcha.. wish i remember what 2013-02-12T20:15:57 < zyp> yes, that's a difference 2013-02-12T20:16:19 < zyp> hmm, I could change to TIM3, which is present on same pins 2013-02-12T20:16:23 < zyp> but that would be cheating :p 2013-02-12T20:16:25 < jpa-> TIM8_BDTR maybe? 2013-02-12T20:16:36 < jpa-> and main output enable 2013-02-12T20:16:50 < zyp> ah, likely 2013-02-12T20:17:23 < zyp> yep, that was it 2013-02-12T20:17:24 < dirty_d> MOA and AOE 2013-02-12T20:17:25 < zyp> thanks 2013-02-12T20:17:29 < dirty_d> that got me too 2013-02-12T20:19:26 < zyp> http://bin.jvnv.net/f/mDLBc.png <- looking all nice now :) 2013-02-12T20:19:42 <+Steffanx> Whoaw 2013-02-12T20:19:44 < BJFreeman> ah looks like i have more research thanks for the tips 2013-02-12T20:23:16 < jpa-> perhaps we need to start keeping a list http://stm32.izua.ro/wiki/Assorted_STM32_Gotchas 2013-02-12T20:26:03 < zyp> http://bin.jvnv.net/f/FkIel.JPG <- got some cables hanging from it now 2013-02-12T20:26:04 < zyp> :p 2013-02-12T20:26:35 <+Steffanx> Everything for more throughput than one needs 2013-02-12T20:27:21 < zyp> hey, it's nice 2013-02-12T20:27:38 < zyp> http://bin.jvnv.net/f/d2rY7.png <- I was graphing gyro data earlier to check that everything still worked :p 2013-02-12T20:28:50 < jpa-> nice pendulum 2013-02-12T20:29:14 <+Steffanx> What is the blue line? 2013-02-12T20:29:19 < zyp> yaw 2013-02-12T20:29:36 < zyp> I pushed it, and then it slowed down by itself 2013-02-12T20:29:54 < zyp> x axis is ten seconds long 2013-02-12T20:52:24 < Tectu> BrainDamage, you're from como, right? 2013-02-12T20:53:56 < BrainDamage> yes 2013-02-12T20:59:18 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-12T21:18:36 -!- dongs_ [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-12T21:24:16 < dirty_d> zyp, you got it flying right? 2013-02-12T21:24:25 < zyp> huh? 2013-02-12T21:24:44 -!- drgreenthumb is now known as dfletcher 2013-02-12T21:24:49 < zyp> it's currently hanging from my ceiling here 2013-02-12T21:24:51 -!- CheBuzz [~CheBuzz@76.164.192.124] has quit [Ping timeout: 276 seconds] 2013-02-12T21:24:56 -!- Netsplit *.net <-> *.split quits: dongs 2013-02-12T21:24:56 < zyp> http://bin.jvnv.net/f/sUitf.JPG 2013-02-12T21:24:59 < zyp> for testing 2013-02-12T21:26:17 < zyp> I've done some hovering around in the garden with it before, with an older controller board, and the older frame I did previously was flying 2013-02-12T21:27:06 < zyp> but the code has been majorly reworked since then, to give a much better foundation 2013-02-12T21:27:53 < jpa-> the better the code becomes, the further it is from completion :) 2013-02-12T21:28:10 < zyp> yeah 2013-02-12T21:28:31 < zyp> the first flying one was hacked up in a very short time :p 2013-02-12T21:28:52 < zyp> it didn't fly good, but was still lots of fun :p 2013-02-12T21:29:03 <+Steffanx> cheater 2013-02-12T21:29:44 < zyp> now I know more about how everything works, so now I'm going to do it better ;) 2013-02-12T21:31:03 < dirty_d> sweet 2013-02-12T21:31:03 < zyp> now it's going to hang from the ceiling until it's stable enough to hover indoors by itself 2013-02-12T21:31:06 < zyp> :p 2013-02-12T21:31:48 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-12T21:32:00 < dirty_d> what gyro/accel did you use? 2013-02-12T21:32:03 < zyp> the old one was more «doesn't matter if it's a bit unstable, I'll just fly high enough to keep it off the ground» :p 2013-02-12T21:32:16 < zyp> this one is LSM303DLM and L3GD20 2013-02-12T21:32:50 < zyp> previous one used some atmel demo board with ITG3200 and some other chips 2013-02-12T21:33:19 < zyp> http://bin.jvnv.net/f/0nLzO.JPG <- second gen vs third gen controllers 2013-02-12T21:33:40 < zyp> first gen was a protoboard and a lot of wires :p 2013-02-12T21:33:51 < zyp> with same sensor board as second gen 2013-02-12T21:34:17 < zyp> http://bin.jvnv.net/f/PT6Z4.jpg 2013-02-12T21:37:41 < dirty_d> looks really nice 2013-02-12T21:38:29 < zyp> yeah, if I can just get it to fly stable, I'm going to have lots of fun this summer ;) 2013-02-12T21:40:00 < dirty_d> youre using regular ESCs right? 2013-02-12T21:40:10 < zyp> currently, yes 2013-02-12T21:40:25 < dirty_d> with a 400Hz servo signal? 2013-02-12T21:40:40 < zyp> yep 2013-02-12T21:40:43 < dirty_d> yea me too 2013-02-12T21:40:58 < dirty_d> i need to get off my ass and get this thing flying 2013-02-12T21:41:00 < zyp> http://bin.jvnv.net/f/Ftmrn.JPG <- currently got some experimental code for this as well, which I was going to control with can 2013-02-12T21:41:32 < zyp> but I don't have time to work on that now, need to prioritize getting the controller working instead 2013-02-12T21:41:37 < dirty_d> yea 2013-02-12T21:42:20 < dirty_d> i should be able to just plug everything into my smt32f3disco and give it a shot 2013-02-12T21:42:28 < dirty_d> easier to debug than my old board 2013-02-12T21:42:40 < dirty_d> and i dont really feel like re-learning all the avr stuff, lol 2013-02-12T21:43:44 < dirty_d> did you ever like perfectly balance your quad, set it level and adjust the board with the accel on it so its zeroed? 2013-02-12T21:43:56 < dirty_d> i haventreally done that, i wonder if thats one of the reasons i had a hard time 2013-02-12T21:46:47 < zyp> umm, don't remember 2013-02-12T21:51:50 < dirty_d> hmm i suppose i can just plug the servo connectors straight into the discovery board 2013-02-12T21:52:22 < dirty_d> jsut make sire the signal pin is to a usable pin, and the other pins can be driven high or low for VCC and gnd 2013-02-12T21:52:31 < zyp> wat 2013-02-12T21:52:45 < zyp> just forget vcc and gnd. 2013-02-12T21:53:07 < zyp> pwm is on one end, so it should be easy enough to connect only that one 2013-02-12T21:53:29 < dirty_d> then how will i ground it 2013-02-12T21:53:47 < zyp> how will you supply power to it? 2013-02-12T21:53:59 < dirty_d> yea 2013-02-12T21:54:06 < dirty_d> i dont have any male-female jumpers 2013-02-12T21:54:11 < dirty_d> wires 2013-02-12T21:54:27 < zyp> the ESCs have common power and ground anyway 2013-02-12T21:54:32 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 255 seconds] 2013-02-12T21:54:46 < zyp> and you'll be powering the discovery board from that source in some fashin too 2013-02-12T21:55:01 < zyp> so you already got a usable signal ground 2013-02-12T21:55:04 < dirty_d> this is true 2013-02-12T21:55:26 -!- neuro_sys|off is now known as neuro_sys 2013-02-12T21:55:45 < dirty_d> i have a BEC, im sure theres somewhere i cna plug that into right 2013-02-12T21:56:20 < zyp> make up a usb connector for it, so you can plug it into one of the usb plugs 2013-02-12T21:56:29 < zyp> sockets* 2013-02-12T21:57:30 < dirty_d> that would work 2013-02-12T21:57:38 < dirty_d> i dont have enough though 2013-02-12T21:58:50 < dirty_d> i think there are pins on the top for poer, not sure though 2013-02-12T21:59:04 < zyp> yep 2013-02-12T21:59:54 < dirty_d> 5V power supply (input/output) 2013-02-12T22:00:09 < dirty_d> i can stick my BEc right into that 2013-02-12T22:01:31 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-12T22:07:06 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 276 seconds] 2013-02-12T22:10:00 < dirty_d> the datasheet for LSM303DLHC is terrible 2013-02-12T22:10:19 < dirty_d> they give absolutely no explaination for what the DRDY1 and DRDY2 interrupts actually are 2013-02-12T22:11:01 < zyp> isn't that configurable? 2013-02-12T22:12:48 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-12T22:13:05 < dirty_d> it is 2013-02-12T22:13:14 < dirty_d> but it doesnt say what triggers those interruopts 2013-02-12T22:13:28 < dirty_d> im assuming when accel data is ready, but why are there 2? 2013-02-12T22:13:46 < zyp> hang on 2013-02-12T22:14:34 < zyp> you said dlhc 2013-02-12T22:14:36 < zyp> ? 2013-02-12T22:16:10 < dirty_d> yes 2013-02-12T22:16:30 < zyp> ah, right, now I get your question 2013-02-12T22:17:59 < dirty_d> maybe lo and high byte? 2013-02-12T22:18:04 < dirty_d> sounds stupid though 2013-02-12T22:18:51 < zyp> I suspect one is accel and one is coupled from mag die 2013-02-12T22:19:59 < dirty_d> zyp, the mag has its own 2013-02-12T22:20:02 < dirty_d> its just plain DRDY 2013-02-12T22:20:15 < zyp> yes, I suspect that one is coupled to the accel die 2013-02-12T22:20:15 < emeb> dumb question: do the motors put out enough field to affect the mag? 2013-02-12T22:20:29 < zyp> emeb, haven't measured 2013-02-12T22:20:39 < dirty_d> if its close enough definitely 2013-02-12T22:20:43 < zyp> also, these motors are utter shit 2013-02-12T22:20:58 < ds2> yes 2013-02-12T22:20:59 < dirty_d> i think if it were like a foot away, no way 2013-02-12T22:21:08 < ds2> so do batteries 2013-02-12T22:21:19 < zyp> one has a bad bearing, and another aren't fully good either 2013-02-12T22:21:34 < zyp> and I've already replaced one of those with a spare earlier 2013-02-12T22:21:47 < zyp> need to get some more decent ones 2013-02-12T22:23:03 < dirty_d> mine all seem ok 2013-02-12T22:23:16 < dirty_d> i got the cheapest ones i could find that were pwoerful enough 2013-02-12T22:23:16 -!- CheBuzz [~CheBuzz@76.164.192.124] has joined ##stm32 2013-02-12T22:23:37 < zyp> the ones I used on my first was better than this shit 2013-02-12T22:24:02 < dirty_d> mine are too big i think 2013-02-12T22:24:04 < dirty_d> 2Kw 2013-02-12T22:24:16 < dirty_d> but whatever 2013-02-12T22:25:07 < dirty_d> zyp, how are you timing your 400Hz loop? 2013-02-12T22:25:53 < dirty_d> on ly last board the accel had a data ready interrupt, so i just set the rate to 400Hz and used that 2013-02-12T22:26:07 < zyp> huh? 2013-02-12T22:26:24 < zyp> I just use a timer running at 1MHz, with a reload value of 2500 2013-02-12T22:26:25 < dirty_d> i used the accel's data ready interrupt to time my control loop 2013-02-12T22:26:39 < dirty_d> zyp, but if the data from the accel isnt ready you get an old sample 2013-02-12T22:27:10 < zyp> haven't dealt with that yet 2013-02-12T22:27:30 < dirty_d> you have the same one as me pretty much right? max rate is 400Hz on the accel? 2013-02-12T22:28:02 < dirty_d> if you run the controll loop at 400Hz you might only really be effectively getting a 200Hz loop since half the data is the same 2013-02-12T22:28:59 < dirty_d> im just gonna screw around with these interrupt settings and see what it does, i guess thats the only way 2013-02-12T22:32:04 < zyp> accel is pretty irrelevant, it's the gyro that has interesting data 2013-02-12T22:32:25 -!- barthess [~barthess@178.154.53.6] has quit [Quit: Leaving.] 2013-02-12T22:32:55 < dirty_d> what? 2013-02-12T22:33:12 < dirty_d> accel is more important for me 2013-02-12T22:33:21 < dirty_d> its what makes it hover by itself without me doing anything 2013-02-12T22:33:44 < dirty_d> gyro is just to help filter out accelerations other than gravity 2013-02-12T22:33:56 < zyp> don't be silly 2013-02-12T22:34:01 < dirty_d> i forgot what the hell its called 2013-02-12T22:34:06 < dirty_d> complimentary filter or something 2013-02-12T22:34:26 < dirty_d> accel is lowpassed and gyro highpassed and combined 2013-02-12T22:34:33 < zyp> exactly 2013-02-12T22:34:51 < zyp> that's why the gyro speed is more important than the accel speed 2013-02-12T22:35:02 < dirty_d> this is true 2013-02-12T22:35:10 < gxti> gyro is necessary to know how to interpret the accel 2013-02-12T22:35:33 < gxti> but that's just another way of restating what dirty_d said to begin with 2013-02-12T22:35:53 < zyp> it's the gyro that is the main source for the stabilization, the accel is just for cancelling out integration errors on the gyro data 2013-02-12T22:36:29 < zyp> you could easily have it reasonably stable with gyro only, just not over time due to integration errors 2013-02-12T22:36:49 < zyp> but you don't get it stable with accel only 2013-02-12T22:37:52 < zyp> so again, the accel is pretty irrelevant when it comes to the fast update loop 2013-02-12T22:38:08 < dirty_d> ok that makes sense 2013-02-12T22:45:51 < dirty_d> zyp, are you using the compas for this also? 2013-02-12T22:46:02 < dirty_d> wouldnt you be able to make yaw locked down with that? 2013-02-12T22:46:53 < zyp> only serves same role as accel in that respect 2013-02-12T22:47:01 < dirty_d> right 2013-02-12T22:47:20 < dirty_d> but youd be able to point in some directio and then stay there forever with no input right? 2013-02-12T22:49:04 < dirty_d> like when the commanded yaw rate reaches 0 save the current heading, and stay there until told to rotate again 2013-02-12T22:53:00 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-12T22:55:53 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-12T22:56:09 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 244 seconds] 2013-02-12T23:31:46 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-12T23:32:42 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-12T23:58:33 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Remote host closed the connection] --- Day changed Wed Feb 13 2013 2013-02-13T00:01:48 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-13T00:15:32 -!- scrts [~quassel@unaffiliated/scrts] has quit [Quit: No Ping reply in 180 seconds.] 2013-02-13T00:15:47 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32 2013-02-13T00:20:13 -!- scrts [~quassel@unaffiliated/scrts] has quit [Client Quit] 2013-02-13T00:20:32 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32 2013-02-13T00:32:25 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-13T00:59:28 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T01:00:42 -!- dongs_ is now known as dongs 2013-02-13T01:00:44 < dongs> http://www.dailymail.co.uk/news/article-2269094/North-Korean-parents-eat-children-driven-mad-hunger-famine-hit-pariah-state.html#axzz2KjEvEzVq 2013-02-13T01:00:48 < dongs> lol 2013-02-13T01:04:01 < zyp> dongs, http://hobbyking.com/hobbyking/store/__25556__AX_2810Q_750KV_Brushless_Quadcopter_Motor.html <- is this good shit? 2013-02-13T01:04:11 < zyp> I hope it's good shit, because I ordered a bunch 2013-02-13T01:05:24 < dongs> > weight 70g 2013-02-13T01:05:26 < dongs> what hte fuck? 2013-02-13T01:06:07 < zyp> xbox, etc… 2013-02-13T01:06:25 < dongs> right 2013-02-13T01:06:34 < dongs> personally for that price I'd have gotten sunnysky2212 2013-02-13T01:06:43 < dongs> http://www.himodel.com/electric/SUNNYSKY_X2212_KV980_Outrunner_Brushless_Motor_Multi-rotor_Version.html i.e. this 2013-02-13T01:06:52 < dongs> way less aids 2013-02-13T01:06:58 < zyp> hmm 2013-02-13T01:07:18 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-13T01:07:26 < dongs> seems you need 12" or bigger pr ops 2013-02-13T01:07:30 < dongs> to make any use of the ones you linked. 2013-02-13T01:07:38 < dongs> as in your shit would be xbox-huge 2013-02-13T01:07:42 < zyp> yes 2013-02-13T01:07:46 < zyp> already got that 2013-02-13T01:08:57 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-13T01:10:56 < dongs> hmm lost your microusb part# again 2013-02-13T01:11:11 -!- DaKu is now known as daku 2013-02-13T01:11:13 < zyp> hmm, maybe I'll cancel order and get those motors instead, looks nice 2013-02-13T01:12:42 < zyp> 609-4050-1-ND 2013-02-13T01:13:18 < dongs> http://i.minus.com/ibj7wm2hqVvdmw.gif 2013-02-13T01:13:42 -!- Laurenceb_ [~Laurence@host86-136-119-170.range86-136.btcentralplus.com] has joined ##stm32 2013-02-13T01:14:10 < gxti> needs more lift 2013-02-13T01:15:27 < zyp> dongs, those keda motors you are selling, are they nice? 2013-02-13T01:15:49 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-13T01:15:53 < dongs> they're nice for small stuff, since noone else makes a good small motor 2013-02-13T01:15:56 < dongs> (for cheap) 2013-02-13T01:16:54 < zyp> probably too small for mine then 2013-02-13T01:18:40 < zyp> hmm, quad is >1kg without battery :p 2013-02-13T01:18:46 < zyp> with current motors 2013-02-13T01:19:03 < dongs> right 2013-02-13T01:20:20 < zyp> should I get those sunnysky motors then? 2013-02-13T01:21:07 < dongs> i would think so... the HK shit just looks too big 2013-02-13T01:21:31 < dongs> depends what youre trying to make. 2013-02-13T01:23:09 < zyp> any suggestions on where to order? 2013-02-13T01:25:23 < dongs> whats wrong wiht himodel? 2013-02-13T01:25:44 < zyp> nothing, just haven't ordered there before 2013-02-13T01:25:51 < dongs> theres eurotrash dealers for the stuff, but its > $20 there for sure 2013-02-13T01:26:03 < dongs> himodel is legit, i got ~10k wasted with them 2013-02-13T01:26:16 < Simon--> they do charge paypal fee and more expensive shipping than hk though 2013-02-13T01:26:26 < dongs> the usual china holiday shit applies though, until 16th 2013-02-13T01:26:28 < dongs> or wah tever 2013-02-13T01:38:08 < zyp> might as well just wait and order it later then 2013-02-13T01:39:26 < zyp> cheaper to order from himodel with fast shipping than order from european dealers anyway 2013-02-13T01:49:43 < dongs> right 2013-02-13T01:51:21 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-72.pptp.stw-bonn.de] has joined ##stm32 2013-02-13T01:52:25 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out] 2013-02-13T01:52:42 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-13T01:53:48 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-13T02:09:51 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Remote host closed the connection] 2013-02-13T02:16:10 < dongs> poking about pink solder mask again 2013-02-13T02:16:15 < dongs> didnt get a proper ans last time 2013-02-13T02:26:40 -!- Laurenceb_ [~Laurence@host86-136-119-170.range86-136.btcentralplus.com] has quit [Ping timeout: 256 seconds] 2013-02-13T02:43:38 < Thorn> pink solder mask? for lady engineers? 2013-02-13T02:44:42 < emeb> Did you know that pink used to be for boy babies? http://www.straightdope.com/columns/read/2831/was-pink-originally-the-color-for-boys-and-blue-for-girls 2013-02-13T03:04:44 < Gargantuasauce_> http://raspberryhigh.wikia.com/wiki/Raspberryhigh_Wiki 2013-02-13T03:04:53 < Gargantuasauce_> dongs your tarduino tag line is fucking terrifyingly apt 2013-02-13T03:10:06 < dongs> what teh fucking fuck 2013-02-13T03:11:21 < dongs> wth 2013-02-13T03:11:23 < Thorn> what is that, something having to do with pot? 2013-02-13T03:11:25 < dongs> why is windows task manager using 23% cpu 2013-02-13T03:11:29 < dongs> Thorn: yeah 2013-02-13T03:11:32 < dongs> stoners 2013-02-13T03:11:34 < gxti> it's managing your tasks 2013-02-13T03:18:05 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-13T03:23:41 < emeb> wouldn't want those tasks to go unmanaged. 2013-02-13T03:25:15 < esden> one does not want to get stoned in an unmanaged manner. Use stoneduino to manage your stonage. 2013-02-13T03:37:20 < dongs> haha 2013-02-13T03:37:21 < dongs> found the problem 2013-02-13T03:37:25 < dongs> pagefile.sys: 50gigs 2013-02-13T03:37:33 < dongs> looks like some driver took a shat or something 2013-02-13T03:37:58 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-13T03:40:31 < Thorn> http://oglaf.com/media/comic/umbra.jpg 2013-02-13T03:40:33 < Erlkoenig> what about putting drugs into an implanted medicamentation pump? 2013-02-13T03:40:45 < Thorn> well at lrast it's not lunix right? 2013-02-13T03:41:12 < Erlkoenig> huehue oglaf 2013-02-13T03:51:37 -!- emeb [~ericb@72.201.78.226] has quit [Quit: Leaving.] 2013-02-13T04:03:26 < BJFreeman> dongs the pagefile.sys should not exceed you Ram 2013-02-13T04:03:34 < BJFreeman> in size 2013-02-13T04:03:41 < dongs> well, it was set to "windows managed". 2013-02-13T04:03:45 < dongs> and somethign fucked up . 2013-02-13T04:04:05 < BJFreeman> yup 2013-02-13T04:04:29 < dongs> set it to 1gb since i have 16gig ram. 2013-02-13T04:04:32 < gxti> heh, FE has KSZ8021 for half the cost of DK, to go with the stm32f107rc that is also half the cost 2013-02-13T04:04:54 < BJFreeman> set it to 16gigs 2013-02-13T04:04:56 < gxti> it's a shame their search fails or i might actually go there for reasons other than price matching 2013-02-13T04:05:24 < dongs> gxti: know of a cheap gige phy? 2013-02-13T04:05:30 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-13T04:05:40 < gxti> nah. i think micrel owns the bottom there too. 2013-02-13T04:06:04 < dongs> well, ok, forget cheap, how about one I can actually buy 2013-02-13T04:06:08 < dongs> margvell shit = nda + failware 2013-02-13T04:06:20 < dongs> digikey has almost nothing unless im looking in wrong places 2013-02-13T04:06:36 < gxti> micrel != marvell 2013-02-13T04:06:42 < dongs> i know. 2013-02-13T04:07:26 < dongs> http://www.digikey.com/product-search/en/integrated-circuits-ics/interface-drivers-receivers-transceivers/2556324?k=KSZ9021GN 2013-02-13T04:07:29 < dongs> cool 2013-02-13T04:07:59 < gxti> i know at least one person who has used those successfully 2013-02-13T04:09:59 < dongs> ok, nice 2013-02-13T04:10:07 < dongs> thinking for the spartan6 proj here 2013-02-13T04:10:24 < dongs> to use that shit. 2013-02-13T04:10:36 < gxti> that's what he was doing as well 2013-02-13T04:10:36 < dongs> but not tcpip, just raw physical layer 2013-02-13T04:10:54 < gxti> doing some research on implementing higher layers in hardware 2013-02-13T04:10:57 < dongs> not sure if worth doing tcpip just to dump data cross 2013-02-13T04:11:02 < gxti> nutty stuff 2013-02-13T04:11:04 < dongs> across 2013-02-13T04:11:29 < gxti> but yeah the lazy approach is spamming at layer 2 2013-02-13T04:12:02 < dongs> is it bad? 2013-02-13T04:12:07 < dongs> its just point to point, high speed link 2013-02-13T04:12:41 < gxti> well, you can do UDP without much more trouble, it's just a bigger header 2013-02-13T04:12:48 < dongs> but why? 2013-02-13T04:12:53 < dongs> source and receiver hardware is both fpga 2013-02-13T04:12:53 < gxti> easier to receive in software 2013-02-13T04:12:57 < dongs> i'd have to implement that shit over it etc 2013-02-13T04:12:58 < gxti> ah 2013-02-13T04:12:59 < gxti> never mind 2013-02-13T04:13:04 < gxti> do raw ethernet then 2013-02-13T04:13:06 < dongs> right 2013-02-13T04:14:07 < gxti> right then, 20 more ksz8021 in the cart just because they're so f'ing cheap 2013-02-13T04:14:16 < dongs> haha 2013-02-13T04:14:18 < dongs> at FE? 2013-02-13T04:14:21 < gxti> yeah 2013-02-13T04:14:27 < dongs> know any c heap fiber phy? 2013-02-13T04:15:06 < gxti> don't know much about fiber, i think there's a phy before the actual fiber module in which case again i'd check what micrel has 2013-02-13T04:16:29 < dongs> broadscum and their awesome "datasheets" 2013-02-13T04:16:47 < gxti> ah, 8021 is the one that comes up in 50mhz osc mode instead of 25mhz xtal mode, good thing i'm not using it to clock the mcu 2013-02-13T04:18:17 < gxti> although amusingly stm32 mac will fail to reset if the phy is not clocking 2013-02-13T04:18:26 < gxti> and chibios explicitly shuts down the phy before resetting mac 2013-02-13T04:18:36 < gxti> so it hangs until you fix that 2013-02-13T04:18:43 < dongs> > chibidongs 2013-02-13T04:20:33 < gxti> shit, stm32f107rc is out of stock now 2013-02-13T04:20:39 < gxti> fucking canadians 2013-02-13T04:21:18 < dongs> ha ha 2013-02-13T04:21:19 < dongs> that was me 2013-02-13T04:21:34 < gxti> you always say that 2013-02-13T04:22:39 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 252 seconds] 2013-02-13T04:29:43 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-13T04:32:34 < dongs> gonna ask bladerf fags 2013-02-13T04:32:35 < dongs> they might know 2013-02-13T04:32:57 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-13T04:53:20 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-13T05:10:32 < emeb_mac> Minecraft for rpi - now I gotta get one 2013-02-13T05:18:14 < Viper168> rpi? 2013-02-13T05:21:20 < emeb_mac> raspberry pi 2013-02-13T05:22:58 < gxti> client or server? 2013-02-13T05:23:31 < gxti> because i'm pretty sure someone made a server for a 16 bit micro already 2013-02-13T05:23:49 < Erlkoenig> server 2013-02-13T05:23:50 < gxti> which means a server for rpi would be even dumber than otherwise 2013-02-13T05:24:05 < Erlkoenig> it's more like a full server that does more of the fancy dynamics 2013-02-13T05:24:06 < gxti> durr i can run the javas on a linux computar 2013-02-13T05:36:23 < baird> "Hey, This year's Hot Indie Game is a more boring version of The Sentinel! :D" 2013-02-13T05:45:34 < baird> My Amazing Psychic Powers predict that within the next few days, Hackaday will blog about a Raspberry Pi Minecraft MAME cabinet. 2013-02-13T05:47:27 < gxti> each button will be an entire arduino 2013-02-13T05:47:33 < gxti> and the cabinet is 3d printed. 2013-02-13T05:49:07 < baird> ..by Jeri Ellsworth 2013-02-13T05:52:10 < dongs> haha 2013-02-13T05:54:10 < emeb_mac> the perfect storm - maybe it'll be sold on adafruit.com 2013-02-13T05:56:29 < Erlkoenig> my friend loves ordered places so much he ordered parts of my resistor collection... 2013-02-13T05:56:45 < dongs> wat 2013-02-13T05:58:30 < dongs> is there a tarduino ISP for stm32 2013-02-13T05:59:55 < dongs> http://www.jnutthailand.com/hobby/image/DSC04719B.JPG 2013-02-13T05:59:56 < dongs> fuck yeah 2013-02-13T06:00:25 < gxti> what peripherals does one typically use for sd? 2013-02-13T06:00:31 < dongs> SDIO? 2013-02-13T06:00:35 < dongs> if you want speed 2013-02-13T06:00:36 < dongs> or SPI 2013-02-13T06:00:40 < dongs> if you want aids 2013-02-13T06:00:42 < gxti> cheap and slow is cool 2013-02-13T06:00:53 < Erlkoenig> SDIO? 2013-02-13T06:00:54 < dongs> what do you mean peripherals. on STM32? 2013-02-13T06:00:56 < gxti> i'm running low on pins, let's see what i can do 2013-02-13T06:01:00 < gxti> yes, stm32 2013-02-13T06:01:13 < dongs> well, im sure you can bitbang it if you really wanted to 2013-02-13T06:01:21 < gxti> that might be a little too shitty 2013-02-13T06:01:27 < gxti> so let's go with spi 2013-02-13T06:01:43 < GargantuaSauce> use an MCP23017 2013-02-13T06:01:50 < gxti> but i probably don't have one free now so i'll have to shuffle crap around 2013-02-13T06:02:10 < dongs> haha 2013-02-13T06:02:11 < dongs> @ mcp 2013-02-13T06:02:23 < GargantuaSauce> whats wrong with 3 orders of magnitude slowdown anyway 2013-02-13T06:02:31 < gxti> also lol "Eease AVR" 2013-02-13T06:03:55 < dongs> haha 2013-02-13T06:04:49 < Thorn> Dave Jones finally got himself a cesium clock http://www.eevblog.com/2013/02/13/eevblog-423-hp5061a-atomic-clock-cesium-beam-frequency-standard/ 2013-02-13T06:05:11 < Thorn> he only had rubidium ones before 2013-02-13T06:06:49 < gxti> not his though 2013-02-13T06:07:35 < dongs> what hte hell 2013-02-13T06:07:47 < dongs> noone made a avr-isp out of *discovery boards 2013-02-13T06:07:49 < dongs> ?!?!!? 2013-02-13T06:08:10 < gxti> waste of time dongs, everybody who has learned stm32 has kicked that shit to the curb 2013-02-13T06:08:16 < dongs> lol yes im sure 2013-02-13T06:08:16 < gxti> even for the lulz 2013-02-13T06:08:27 < dongs> but i blew my avr-isp (again) and i need working programmer asap 2013-02-13T06:08:42 < gxti> sux2bu 2013-02-13T06:08:45 < dongs> indeed 2013-02-13T06:08:46 < GargantuaSauce> i have a few atmega8's i plan on using for slaves...will be using my f4 to program them i imagine 2013-02-13T06:08:48 < GargantuaSauce> but havent done it yet 2013-02-13T06:09:00 < gxti> i use pics for cheap shit, pickit3 hasn't died yet 2013-02-13T06:09:24 < emeb_mac> ICD3 FTW 2013-02-13T06:09:27 < gxti> last thing i did was like 6 months ago and it was literally a clock divider 2013-02-13T06:09:58 < dongs> i might have a tarduino kicking around somewehre 2013-02-13T06:10:02 < dongs> might have to go tarduinoISP route 2013-02-13T06:16:51 < gxti> ah good, i have at least two SPIs. just have to shove some LEDs somewhere else. 2013-02-13T06:18:52 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-13T06:21:04 < emeb_mac> whatcha spi-ing? 2013-02-13T06:21:11 < gxti> dem sd cards 2013-02-13T06:21:25 < emeb_mac> oooo kewl 2013-02-13T06:22:04 < emeb_mac> that ended up being a lot easier to do than I first imagined. fatfs FTW. 2013-02-13T06:22:23 < ds2> run Linux and don't worry about it ;) 2013-02-13T06:22:48 < emeb_mac> meh. linux don't fit into 2"^2 2013-02-13T06:23:07 < ds2> what's '2"^2'? 2013-02-13T06:23:12 < emeb_mac> 2sqin 2013-02-13T06:23:20 < gxti> "'2\"^2'" 2013-02-13T06:23:23 < ds2> my board is 2 square inches 2013-02-13T06:23:24 < ds2> :) 2013-02-13T06:23:29 < ds2> less actually 2013-02-13T06:23:35 < ds2> and I got a battery charger and sensors on it too 2013-02-13T06:23:43 < gxti> mine isn't linux and is 4sqin right now :p 2013-02-13T06:23:53 < gxti> but it has to actually do some stuff. 2013-02-13T06:23:57 < emeb_mac> that's because ds2 is superhuman. 2013-02-13T06:24:21 < emeb_mac> and I didn't use dongspcb 2013-02-13T06:24:25 < ds2> emeb_mac: you can even find OMAP3 based boards under 2x2inches 2013-02-13T06:24:35 < ds2> wait... 2 square inches...ops.. blah 2013-02-13T06:24:59 < ds2> sqrt(2)xsqrt(2)...hmmmm I can do that if I get rid of the 2nd display connector 2013-02-13T06:25:50 < emeb_mac> ds2 did you actually get uclinux working on your system? 2013-02-13T06:26:18 < ds2> emeb_mac: yep. doing a talk about that sort of stuff at ELC too 2013-02-13T06:26:26 < emeb_mac> nice 2013-02-13T06:26:28 < ds2> I'll post pictures when I get a chance 2013-02-13T06:26:38 < emeb_mac> looking fwd to that. 2013-02-13T06:26:57 < ds2> rushing through another demo right now....had to rewire 24 traces cuz of the %@#$#@$@#%$@#$@#$#@$#@ TI docs 2013-02-13T06:27:03 < ds2> <--- not happy about the TI docs 2013-02-13T06:27:07 < emeb_mac> thanks TI 2013-02-13T06:27:46 < ds2> 24bpp for the LCDC doesn't really mean 24bpp...it means 16bpp + 8bpp (from a wiring standpoint) 2013-02-13T06:28:26 < emeb_mac> joy 2013-02-13T06:29:17 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-72.pptp.stw-bonn.de] has quit [Ping timeout: 240 seconds] 2013-02-13T06:30:06 < ds2> this has been a bad few months for vendor info 2013-02-13T06:30:30 < ds2> the ST app note uCLinux port had its issues 2013-02-13T06:31:31 < emeb_mac> not surprising. I'd take anyone's discussion of linux implementation w/ NaCl - if only because linux is a moving target. 2013-02-13T06:31:43 < ds2> no 2013-02-13T06:31:51 < ds2> that isn't the problem 2013-02-13T06:32:07 < ds2> the exact config/source tree they used does not work and cannot work. 2013-02-13T06:32:17 < emeb_mac> oh fun! 2013-02-13T06:32:27 -!- TeknoJuce [~TeknkoJuc@xbmc/staff/TeknoJuce] has quit [Ping timeout: 252 seconds] 2013-02-13T06:32:45 < ds2> using their config/source results in the memory management table placed outside of valid memory space 2013-02-13T06:32:58 < emeb_mac> ah yes - saw that discussion the other day. 2013-02-13T06:33:45 < ds2> but this explains the error messages that flew by during a demo at ARM Techcon a few years ago 2013-02-13T06:34:51 < gxti> sometimes i really wonder about the pinout for stm32 2013-02-13T06:35:01 < ds2> in what way? 2013-02-13T06:35:04 < gxti> why is there a usart3 alt on the exact same pins as uart4? 2013-02-13T06:35:19 < ds2> F3? 2013-02-13T06:35:19 < emeb_mac> so you can swap them? 2013-02-13T06:35:24 < gxti> f1 2013-02-13T06:35:48 < ds2> does USART3 support exactly the same features as USART4? (i.e. flow control) 2013-02-13T06:35:49 < gxti> i guess because it's usart vs uart there's some use case 2013-02-13T06:35:55 < dongs> i was gonna say 2013-02-13T06:35:59 < dongs> usart4 is shittier than usart3 2013-02-13T06:36:35 < gxti> the annoying thing about f107r is that ethernet overlaps with everything 2013-02-13T06:37:05 < gxti> but i'm not moving to f107v, no sir 2013-02-13T06:37:39 < ds2> what are people using ethernet for thesedays? 2013-02-13T06:37:50 < gxti> porn 2013-02-13T06:37:55 < emeb_mac> ethernet phy? 2013-02-13T06:38:12 < ds2> why not wireless? 2013-02-13T06:38:14 < dongs> lol, porn 2013-02-13T06:38:28 < gxti> price for starters 2013-02-13T06:38:38 < dongs> wireless sucks 2013-02-13T06:38:39 < gxti> this is a ntp server so it's latency sensitive 2013-02-13T06:38:40 < emeb_mac> reliability? 2013-02-13T06:38:45 < ds2> oh 2013-02-13T06:38:51 < dongs> need to figure out what else to order fro digikey 2013-02-13T06:38:55 < dongs> olny buying atmega32s 2013-02-13T06:38:56 < gxti> not to mention wireless sucks 2013-02-13T06:39:00 < dongs> seems kind a awaste 2013-02-13T06:39:17 < ds2> dongs: passives? 2013-02-13T06:39:18 < emeb_mac> get beaglebone. that'll use up some bux 2013-02-13T06:40:44 < ds2> get a few more reels of passives 2013-02-13T06:40:58 < dongs> set for now 2013-02-13T06:41:40 < emeb_mac> what did you get? 2013-02-13T06:41:49 < dongs> i mean 2013-02-13T06:41:52 < dongs> w/passives 2013-02-13T06:41:54 < emeb_mac> ya 2013-02-13T06:41:57 < dongs> dont need any extra for now. 2013-02-13T06:42:04 < dongs> the chinease $15/10uF reel was a nice scam 2013-02-13T06:42:10 < emeb_mac> heh 2013-02-13T06:42:17 < ds2> sure they are really 10uF? 2013-02-13T06:42:25 < dongs> ya they're legit murata 2013-02-13T06:42:28 < dongs> made in japan even 2013-02-13T06:42:34 < ds2> maybe they are just 0.1uF with a bigger case 2013-02-13T06:42:39 < emeb_mac> pfft 2013-02-13T06:42:46 < ds2> did they counterfiet the voltage rating? 2013-02-13T06:42:52 < dongs> i only wanted 6.3V 2013-02-13T06:42:54 < dongs> so thats waht i got 2013-02-13T06:42:57 < ds2> remarked 3.3V's? ;) 2013-02-13T06:42:58 -!- BJFreeman [~bjfree@239.sub-75-196-6.myvzw.com] has quit [Quit: had a good time] 2013-02-13T06:45:43 < R2COM> so...capacitors from Murata but power supply from.. whom? AOS as far as I remember you told 2013-02-13T06:45:57 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-219.pptp.stw-bonn.de] has joined ##stm32 2013-02-13T06:47:14 < R2COM> its like buying Lexus, but then feeding it with chinese oil every 10000 miles. 2013-02-13T06:50:21 < dongs> Heh 2013-02-13T06:54:03 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-13T06:54:07 < dongs> hm 2013-02-13T06:54:12 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-13T06:59:47 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has quit [Remote host closed the connection] 2013-02-13T07:00:07 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has joined ##stm32 2013-02-13T07:01:21 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 248 seconds] 2013-02-13T07:02:21 -!- Viper168 [~Viper@node97.19.251.72.1dial.com] has joined ##stm32 2013-02-13T07:02:21 -!- Viper168 [~Viper@node97.19.251.72.1dial.com] has quit [Changing host] 2013-02-13T07:02:21 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-13T07:22:25 < emeb_mac> here ya go: http://www.youtube.com/watch?feature=player_embedded&v=FKdu01LqXdI 2013-02-13T07:22:48 < emeb_mac> it's tarduino controlled! 2013-02-13T07:25:16 < Erlkoenig> why is this on youtube= 2013-02-13T07:25:19 < Erlkoenig> ? 2013-02-13T07:25:52 < emeb_mac> are you offended? 2013-02-13T07:25:58 < Erlkoenig> no, just wondering 2013-02-13T07:26:15 < emeb_mac> she's a Dr. so it's for SCIENCE! 2013-02-13T07:26:30 < Erlkoenig> because they ban videos just because they're gay and maybe show a bit of skin 2013-02-13T07:26:40 < emeb_mac> no skin there... 2013-02-13T07:27:03 < Erlkoenig> yeah, great 2013-02-13T07:27:13 < Erlkoenig> 2013-02-13T07:31:25 < baird> http://www.toymakerproject.com/hammer/ .. Make them in 'dragon' shapes, and she'll make a fortune off of Furries.. 2013-02-13T07:35:44 < Erlkoenig> http://www.cracked.com/blog/the-6-geekiest-sex-toys/ .."They've fetishized the "frozen in carbonite" scene, which is how you tell Rule 34 to bow down and suck it. Especially since these vac-beds can come with sockets." ... I don't think vac beds come from Star Wars... 2013-02-13T07:36:24 < baird> Can't imagine the "Ph.D" is actually in Engineering, since everything is DIP-level components. 2013-02-13T07:37:05 < ds2> could be a ME 2013-02-13T07:37:11 < ds2> or a CE 2013-02-13T07:38:49 < Erlkoenig> but sex toys using arduinos aren't new P: 2013-02-13T07:39:43 < Erlkoenig> integrating a tiny radio module, a µC and a FET into a vibrator might be funny... :> 2013-02-13T07:42:03 < baird> Erlkoenig: RC vibes have been a standard in Japanese hentai and kink for a good 10 years.. 2013-02-13T07:42:40 < Erlkoenig> ah, would have been strange if noone invented that before 2013-02-13T07:42:43 < baird> I wouldn't be surprised if dongs has actually built some. 2013-02-13T07:43:09 < Erlkoenig> if i where more able with the soldering iron... :-> 2013-02-13T07:44:29 < baird> A ex wanted to start a Hitachi Magic Wand import business a while back. The start-up costs which includes getting all the safety certs killed the idea (thankfully..) 2013-02-13T07:46:10 < Erlkoenig> i wonder how to certify a sex toys safety :D 2013-02-13T07:46:30 < Erlkoenig> a friend of a friend imports high heels for men from china... :D 2013-02-13T07:47:23 < baird> The first MW she got came from the States. "How do I use a 110VAC product in Australia, I wonder..." http://i.imgur.com/Ru0Ntrd.jpg 2013-02-13T07:48:10 < Erlkoenig> i'd certainly prefer using a battery-powered toy :D 2013-02-13T07:48:10 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-13T07:48:28 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-13T07:49:50 < ds2> I 2013-02-13T07:51:56 < dongs> baird: is that ellisworthless 2013-02-13T07:59:55 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-219.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-13T08:03:40 < baird> ha, no. It's the ex who went psycho. 2013-02-13T08:06:18 -!- dekar [~dekar@212.255.45.48] has joined ##stm32 2013-02-13T08:06:21 -!- mode/##stm32 [+v dekar] by ChanServ 2013-02-13T08:07:54 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has quit [Quit: ggh] 2013-02-13T08:13:34 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-13T08:29:02 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 255 seconds] 2013-02-13T08:31:12 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-13T08:36:44 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-13T08:42:59 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 255 seconds] 2013-02-13T08:57:40 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T08:59:43 -!- Gargantuasauce_ [~Gargantua@24.222.208.237] has quit [Ping timeout: 245 seconds] 2013-02-13T09:20:05 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T09:21:23 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T09:32:29 -!- dekar_ [~dekar@212.255.34.180] has joined ##stm32 2013-02-13T09:32:31 -!- mode/##stm32 [+v dekar_] by ChanServ 2013-02-13T09:33:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 272 seconds] 2013-02-13T09:33:22 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 256 seconds] 2013-02-13T09:33:22 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-13T09:36:17 -!- dekar [~dekar@212.255.45.48] has quit [Ping timeout: 252 seconds] 2013-02-13T09:36:38 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32 2013-02-13T09:37:24 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-13T09:50:13 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-13T09:53:52 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-13T09:56:42 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-13T10:01:06 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 252 seconds] 2013-02-13T10:08:19 < R2COM> okay..finally 2013-02-13T10:08:28 < R2COM> in this board will use Kemet+Murata 2013-02-13T10:08:37 < R2COM> capacitors.. 2013-02-13T10:09:00 < R2COM> after tons of simulations and hand derivations for each capacitor almost... 2013-02-13T10:10:26 < dongs> fuckit ill just order a reel of 0.1uF to get freeship 2013-02-13T10:11:12 < dongs> http://www.digikey.com/product-detail/en/GRM188R71C104KA01D/490-1532-2-ND/587060 2013-02-13T10:11:29 < R2COM> to use where? 2013-02-13T10:11:43 < dongs> anywhere. 2013-02-13T10:11:43 < dongs> generic bypass shit. 2013-02-13T10:11:55 < R2COM> bypass on a power? 2013-02-13T10:12:01 < dongs> bypass on fucking anything 2013-02-13T10:12:05 < R2COM> what is frequency 2013-02-13T10:12:18 < dongs> D/C as in Dont_Care 2013-02-13T10:12:31 < R2COM> ... 2013-02-13T10:12:33 < R2COM> you are good. 2013-02-13T10:13:23 < R2COM> actually, people (engineers) care about what type of capacitor used in what type of bypassing 2013-02-13T10:13:33 < dongs> yeah, sorry I'm neither 2013-02-13T10:13:44 < dongs> (people or engineer) 2013-02-13T10:14:31 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-13T10:14:52 < R2COM> each cap has its own self resonant frequency, and beyond that its capacitance degrades, it becomes more inductive, as a result not as good in bypassing bus line, especially if bus is powered from switcher 2013-02-13T10:15:38 < dongs> yes, i realize all that shit 2013-02-13T10:15:54 < dongs> however I'm not stocking 10 different types of 0.1uF caps 2013-02-13T10:16:07 < dongs> nor does this matter in any of the shit i make 2013-02-13T10:19:10 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 260 seconds] 2013-02-13T10:19:13 < dongs> R2COM: here's how my purchasing for 0.1UF cap went 2013-02-13T10:19:22 < dongs> tpye in 0.1uF 0603 X7R 2013-02-13T10:19:25 < dongs> click 'sort by price' 2013-02-13T10:19:27 < dongs> enter 4000 2013-02-13T10:19:31 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32 2013-02-13T10:19:33 < dongs> click 'ok' 2013-02-13T10:19:34 < dongs> buy 1st thing that was on top of hte list 2013-02-13T10:19:59 < dongs> that usually goes for most other shit i get as well 2013-02-13T10:21:12 < R2COM> if it works, it works... for non mission critical things I guess its fine then 2013-02-13T10:21:50 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-13T10:23:59 < R2COM> that cap 2013-02-13T10:24:01 < R2COM> ok 2013-02-13T10:24:26 < R2COM> I think I used it before 2013-02-13T10:24:28 < R2COM> lets see... 2013-02-13T10:24:38 < R2COM> popping up some notes with that 2013-02-13T10:25:24 < R2COM> ok that looks like nice capacitor to me, its inductive effects do not predominate on below 2MHz 2013-02-13T10:25:55 < R2COM> and capacitance is a bit more than 0.1uF closer to 2MHz... which is its self resonant frequency 2013-02-13T10:26:21 < R2COM> so yes, I guess its a good cap then for applications below that range 2013-02-13T10:27:18 < R2COM> with standard derating recommended by Navair for example, one can use this cap on 9.6V DC biased applications 2013-02-13T10:27:22 < R2COM> because its rating is 16V 2013-02-13T10:27:34 < dongs> all my DC this cap would be used in is <= 3.3V 2013-02-13T10:32:34 < zyp> some nice logic here tonight 2013-02-13T10:32:49 < zyp> «running low on pins, better use SPI for SD card» 2013-02-13T10:32:55 < dongs> haha 2013-02-13T10:33:05 < zyp> re. gxti 2013-02-13T10:33:06 < dongs> SDIO is less pins for 1bit isnt it? 2013-02-13T10:33:15 < dongs> or same 2013-02-13T10:33:15 < zyp> yes, 1-bit SDIO is three pins 2013-02-13T10:34:09 < zyp> SPI is 4 2013-02-13T10:34:20 < zyp> and 4-bit SDIO is 6 2013-02-13T10:34:28 < dongs> im googling for usb3/2 combo hubs 2013-02-13T10:34:33 < dongs> and I dont feel so bad I wanna make my own anymore 2013-02-13T10:34:58 < dongs> all I see is utter shit designs 2013-02-13T10:47:04 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T10:49:35 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.] 2013-02-13T10:49:47 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T10:52:12 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Quit: trepidaciousMBR] 2013-02-13T10:54:50 < baird> I'm going to need both hands for this retroboner: https://github.com/billbudge/PCS_Atari800 2013-02-13T10:55:45 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32 2013-02-13T10:55:46 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host] 2013-02-13T10:55:46 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32 2013-02-13T11:06:59 < dongs> You may have recently purchased one of the following products – System Centre 2012, SQL Server 2012 or Windows Azure – from Microsoft and we want to know what you think! This is your invitation to participate in the Microsoft product survey and provide us feedback on your experience. Your feedback is an integral part of helping us improve your experience with our products. 2013-02-13T11:07:05 < dongs> i did no such thing 2013-02-13T11:07:06 < dongs> why are tehy spamming me 2013-02-13T11:26:06 < zyp> blame linux 2013-02-13T11:29:38 < baird> blame OS2 2013-02-13T11:30:14 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T11:32:24 < trepidaciousMBR> Yup no one makes nice hubs 2013-02-13T11:43:21 < dongs> ill be sure to post my design under CRETIN COMMONS LICENSE 3.0 on tarduino.cc 2013-02-13T11:54:24 <+dekar_> dongs, how about that one? http://www.wtfpl.net/txt/copying/ 2013-02-13T11:54:39 < dongs> dekar_: is that sam hocevars license 2013-02-13T11:54:45 < dongs> eyah it is. 2013-02-13T12:18:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-13T12:32:41 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-13T12:35:02 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 252 seconds] 2013-02-13T13:24:44 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-13T13:24:47 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-13T13:24:47 < jpa-> hmph, looks like STM32F4 RTC resets itself if the fall time of Vdd is too slow on shut down 2013-02-13T13:33:06 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has quit [Ping timeout: 272 seconds] 2013-02-13T13:33:15 -!- Laurenceb_ [~Laurence@host86-136-119-170.range86-136.btcentralplus.com] has joined ##stm32 2013-02-13T13:38:47 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-13T13:38:47 < baird> That Flea86 guy is getting up to something.. http://www.fleasystems.com/ 2013-02-13T13:43:22 < Laurenceb_> http://www.b3tards.com/u/037e58b9e054b3c8dd7b/topgea3.gif 2013-02-13T13:45:23 < Laurenceb_> http://www.theregister.co.uk/2013/02/13/pi_carbot/ <- is that a CC3D or something? 2013-02-13T13:47:12 <+Steffanx> We know that Flea guy baird ? 2013-02-13T13:49:54 < baird> We has poking around here a few months ago. 2013-02-13T13:50:00 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T13:50:12 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Read error: Connection reset by peer] 2013-02-13T13:50:47 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T13:50:57 < baird> He made a rather impressive 8086 IBM PC clone simulator out of a 100MHz 8052. 2013-02-13T13:51:29 < baird> Did all the video modes, etc. At the time he was adding an ISA slot bridge. 2013-02-13T13:51:30 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T13:56:00 <+Steffanx> 8052 yay 2013-02-13T13:58:34 -!- BrainDamage [~BrainDama@62.10.6.67] has joined ##stm32 2013-02-13T14:06:43 < dongs> attn Laurenceb_ https://youtube.com/watch?v=m5zf_iFrr1M 2013-02-13T14:07:37 <+Steffanx> "For Jappies by jappies" 2013-02-13T14:09:58 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection] 2013-02-13T14:10:14 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T14:10:44 -!- Laurenceb_ [~Laurence@host86-136-119-170.range86-136.btcentralplus.com] has quit [Ping timeout: 260 seconds] 2013-02-13T14:30:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-13T14:32:39 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 256 seconds] 2013-02-13T14:34:09 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32 2013-02-13T14:38:53 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T14:44:49 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32 2013-02-13T14:44:57 < Robint91> hi all 2013-02-13T14:45:11 < Robint91> what is this 2013-02-13T14:45:12 < Robint91> www.stm32java.com 2013-02-13T14:45:34 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T14:46:32 < Robint91> W00T 2013-02-13T14:46:38 < Robint91> http://www.stm32.eu/node/237/page/0/2 2013-02-13T14:46:42 < Robint91> THIS IS AWESOME 2013-02-13T14:46:50 < Robint91> SDRAM interface 2013-02-13T14:46:51 < Robint91> finally 2013-02-13T14:48:32 < dongs> its trash 2013-02-13T14:49:00 < Thorn> nxp and stellaris had sdram for some time now 2013-02-13T14:49:25 < Robint91> Thorn, yeah,but ST is so much better than nxp and stellaris 2013-02-13T14:49:39 < Robint91> dongs, the java part of sdram? 2013-02-13T14:49:45 < Thorn> is it? 2013-02-13T14:51:11 < baird> My Postie bike (Honda CT110) now has a USB port. Suck on that, BMW. 2013-02-13T14:51:37 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Remote host closed the connection] 2013-02-13T14:52:03 < Robint91> Thorn, what do you find? keep in mind the channel name :) 2013-02-13T14:56:36 < donigs> Robint91: the java 2013-02-13T14:56:46 < donigs> Robint91: also i heard (in this channel) that datasheet actually lies 2013-02-13T14:56:59 < donigs> and thers no sdram controller 2013-02-13T14:57:16 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T14:58:19 < Robint91> donigs, NOOOOO 2013-02-13T14:58:36 < Robint91> donigs, I had hope that there would be an sdram controller 2013-02-13T14:59:05 < Robint91> donigs, but if they want to run java they better should have the memory to do it 2013-02-13T14:59:45 < Thorn> lol 2013-02-13T14:59:57 < Thorn> everybody else just licenses the EMC from ARM 2013-02-13T15:00:04 < Thorn> but not ST 2013-02-13T15:00:38 < Robint91> Thorn, isn't that because they have the ART? 2013-02-13T15:01:10 < Robint91> Thorn, but yeah still strange 2013-02-13T15:01:19 < Thorn> what's the connection, that one sits in front of the flash iirc? 2013-02-13T15:02:40 < Robint91> Thorn, dunno 2013-02-13T15:08:56 < ntfreak_> parts ending with 9 will have sdram/tft, eg. stm32f429x, stm32f439x 2013-02-13T15:09:49 < Robint91> ntfreak_, mhh, so it will become easier to run something like uclinux on those things 2013-02-13T15:10:06 < Robint91> ntfreak_, opens a lot of new possibilties 2013-02-13T15:11:17 -!- trepidaciousMBR_ [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T15:11:35 < ntfreak_> if you feel that way inclined, i still think uclinux is too much - ecos has a better footprint 2013-02-13T15:14:43 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 244 seconds] 2013-02-13T15:14:43 -!- trepidaciousMBR_ is now known as trepidaciousMBR 2013-02-13T15:16:17 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 240 seconds] 2013-02-13T15:17:31 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T15:18:39 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T15:23:16 -!- NonaSuomy [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-13T15:23:16 -!- NonaSuomy [~NonaSuomy@69.158.165.12] has quit [Changing host] 2013-02-13T15:23:16 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has joined ##stm32 2013-02-13T15:24:30 < Robint91> ntfreak_, I don't know ecos that well, but how much does it difference as linux, eg do I have a shell where I can log into? 2013-02-13T15:25:30 < donigs> no 2013-02-13T15:25:35 < donigs> onl useless shit has "shells" 2013-02-13T15:25:42 < donigs> ecos is for getting work done 2013-02-13T15:25:46 < donigs> without wasting time. 2013-02-13T15:26:41 <+Steffanx> "Real non-useless shit" has closes sores tools to manage everything 2013-02-13T15:27:00 < jpa-> nuttx does have a (limited) shell 2013-02-13T15:27:01 < Robint91> donigs, does it have a good tcp/ip stack? 2013-02-13T15:27:11 < donigs> of course. 2013-02-13T15:27:19 < donigs> any serious RTOS will have tcpip 2013-02-13T15:27:22 < donigs> shit, even keil's RTX has it 2013-02-13T15:27:23 < jpa-> it comes useful when debugging, but not much else 2013-02-13T15:27:27 < donigs> or most of the freetarded shit 2013-02-13T15:27:34 < donigs> or you can just use lwip 2013-02-13T15:27:35 < donigs> or wahtever 2013-02-13T15:27:45 < Robint91> not lwip 2013-02-13T15:27:52 < Robint91> lwip == root of all evil 2013-02-13T15:27:54 < donigs> bad experience? 2013-02-13T15:27:55 < donigs> ha ha 2013-02-13T15:27:58 < donigs> yeah, welcome to opensauce. 2013-02-13T15:28:03 < Robint91> donigs, yes 2013-02-13T15:28:24 < jpa-> i like nuttx better in that it implements the standard posix interfaces (pthreads etc.) instead of inventing its own 2013-02-13T15:28:26 < Robint91> doesn't ecos use the freebsd stack? 2013-02-13T15:29:22 < Robint91> jpa-, yeah, but I like a normal shell, you get a free configure/maintenance interface for free 2013-02-13T15:29:46 < jpa-> you won't be running a text editor on it anyway 2013-02-13T15:30:02 < gxti> zyp: i don't get it 2013-02-13T15:30:04 < Robint91> like shit no connection, log in via rs232, run ifconfig, run ping, ... 2013-02-13T15:30:08 < jpa-> so not that much "for free", you need to make your own commands for management 2013-02-13T15:30:51 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 245 seconds] 2013-02-13T15:31:08 < gxti> zyp: i wasn't saying spi was less pins, but there are more of them and nearly everything is occupied already so better chance of success. turns out f107 doesn't even have sdio so yeah. 2013-02-13T15:31:10 < Robint91> jpa-, dunno, I like that I have vi or nano, to edit config files 2013-02-13T15:31:28 < donigs> ... 2013-02-13T15:31:32 < jpa-> Robint91: but you won't have that on any RTOS 2013-02-13T15:31:41 < donigs> vi or nano, still fucking laughing 2013-02-13T15:31:44 < donigs> editing config files 2013-02-13T15:31:46 < Robint91> who said that I need a rtos 2013-02-13T15:31:48 < donigs> jesus christ dude is this 1985 or what 2013-02-13T15:32:02 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T15:32:05 < jpa-> Robint91: on any os running on stm32, because it is just too much bloat 2013-02-13T15:32:17 < jpa-> (sure, it would fit - but who cares, it's stupid) 2013-02-13T15:32:31 < Robint91> donigs, config files are WAY more modular than any configuration utility that you write 2013-02-13T15:32:36 < Thorn> ECOS = Expensive Commercial Operating System? (otherwise dongs would never recommend it right?) 2013-02-13T15:32:43 < donigs> Thorn: urite 2013-02-13T15:33:10 < jpa-> vxworks would be the expensive commercial solution :) 2013-02-13T15:33:19 < gxti> "eCos deployed on Chibis-M microsatellite" - lol chibis 2013-02-13T15:33:23 < jpa-> but i guess they won't be having cortex-m3 port anytime soon 2013-02-13T15:33:42 < jpa-> gxti: they should have used chibios, obviously 2013-02-13T15:33:48 < gxti> exactly 2013-02-13T15:35:12 < karlp> isn't ecos free? 2013-02-13T15:35:16 < gxti> doesn't look very interesting 2013-02-13T15:35:27 < jpa-> karlp: yes 2013-02-13T15:35:52 < gxti> no description of how fat it is 2013-02-13T15:36:02 < karlp> it's pretty old, 2013-02-13T15:36:07 < jpa-> if you don't mind getting it from cvs :P 2013-02-13T15:36:13 < karlp> reboot is part of ecos tree, 2013-02-13T15:36:18 < karlp> redboot sorry 2013-02-13T15:36:20 < jpa-> eww.. who uses cvs anymore 2013-02-13T15:36:23 < gxti> there are only so many ways to write a rtos jpa-, eventually somebody must get it right 2013-02-13T15:36:36 < jpa-> gxti: aren't all of them right? 2013-02-13T15:36:56 < gxti> only in socialist europe 2013-02-13T15:36:57 <+Steffanx> stupid question? 2013-02-13T15:37:16 < jpa-> and that is the problem, figuring out which one is best when all of them are superficially "ok" 2013-02-13T15:37:18 <+Steffanx> gxti our favourite yankeelander 2013-02-13T15:42:31 < karlp> wasn't someone talking about problems with nvic priority groups yesterday? 2013-02-13T15:42:43 < karlp> my own logs don't show it 2013-02-13T15:42:55 < zyp> karlp, only problems with understanding them 2013-02-13T15:43:06 < zyp> and it was a few days ago, I think 2013-02-13T15:43:09 < zyp> time flies fast :p 2013-02-13T15:43:42 < karlp> welll, it's probably understanding anyway, higher numbers are lower priority right? 2013-02-13T15:43:51 < karlp> if I don't do anything at all with groups, they're all in the same group? 2013-02-13T15:44:06 < Thorn> www.freertos.org/RTOS-Cortex-M3-M4.html 2013-02-13T15:44:18 < dongs> freertos SUCKS 2013-02-13T15:44:25 < dongs> such assholey syntax, damn 2013-02-13T15:44:53 <+Steffanx> blabla dongs :P 2013-02-13T15:44:54 < Laurenceb> asshole syntax? 2013-02-13T15:44:58 < Thorn> that page doesn't have much to do with freertos 2013-02-13T15:44:59 < Laurenceb> os.goatse ? 2013-02-13T15:45:06 < gxti> i'm looking to try coos when i get a chance, worst case it's ugly and bad. not using coide again though. 2013-02-13T15:46:04 < dongs> coos is cute and easy 2013-02-13T15:46:05 < dongs> and no aids 2013-02-13T15:46:14 < dongs> dongs approved! 2013-02-13T15:46:16 < Laurenceb> coo cox clan right? 2013-02-13T15:46:16 < Thorn> and pointer only queues 2013-02-13T15:46:18 < gxti> i'm a fan of not having aids 2013-02-13T15:46:43 < dongs> your point being? 2013-02-13T15:46:43 < dongs> just because you can't use provided tools doesn't mean they're bad 2013-02-13T15:47:07 < Laurenceb> #stm32flamewarz 2013-02-13T15:47:34 < gxti> GNU/AIDS 2013-02-13T15:48:26 < Laurenceb> no 2013-02-13T15:48:33 < Laurenceb> RMS cant have AIDS 2013-02-13T15:48:38 < Laurenceb> cuz hes a virgin 2013-02-13T15:48:47 <+Steffanx> like i care? 2013-02-13T15:48:53 < gxti> GNU/AIDS is transmitted by software, otherwise it would be HIV/AIDS 2013-02-13T15:49:07 <+Steffanx> like that is the only way to get that Laurenceb... 2013-02-13T15:52:11 <+Steffanx> How much did you pay for your first time Laurenceb ? 2013-02-13T15:52:15 < Thorn> btw, why don't small RTOSes specify worst case execution time for system calls? afaik qnx / vxworks docs have that 2013-02-13T15:53:01 < zyp> soft real time vs hard real time 2013-02-13T15:53:17 < zyp> or lazyness vs commercial stuff with real requirements 2013-02-13T15:54:20 < Thorn> most of them have commercial licensing and support, like freertos, kawaiios etc. 2013-02-13T15:55:47 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has quit [Ping timeout: 240 seconds] 2013-02-13T15:56:05 < Robint91> -_- 2013-02-13T15:56:14 < zyp> quite different price ranges, I imagine 2013-02-13T15:56:37 -!- dongs [~dongs@l212168.ppp.asahi-net.or.jp] has joined ##stm32 2013-02-13T16:00:34 < Robint91> this channel is great 2013-02-13T16:01:18 < Viper168_> hey now guys it's not nice to say things about dongs when he's not here 2013-02-13T16:01:37 < Viper168_> I'm ashamed 2013-02-13T16:01:40 <+Steffanx> It IS nice 2013-02-13T16:01:48 <+Steffanx> just because donigs is also here 2013-02-13T16:02:18 < donigs> Viper168_: cool trollin bro 2013-02-13T16:02:18 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has joined ##stm32 2013-02-13T16:03:35 < Laurenceb> https://twitter.com/infiltrateproxy 2013-02-13T16:03:38 < Laurenceb> i lolled 2013-02-13T16:03:49 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32 2013-02-13T16:04:00 < Laurenceb> was expecting hacking stuff 2013-02-13T16:04:58 < dirty_d> damnit man, im at work 2013-02-13T16:05:05 < dirty_d> can a motherfucker get a NSFW? 2013-02-13T16:05:06 < dirty_d> lol 2013-02-13T16:05:52 < Laurenceb> there i was thinkignt witter was meant to be sfw 2013-02-13T16:06:07 < dirty_d> me too 2013-02-13T16:07:32 < karlp> Thorn: that page was helpful thanks, I've got this http://pastebay.net/1182602 2013-02-13T16:07:48 < karlp> so I've done the same as on that page, and set NVIC_PRIORITY_Group 4, 2013-02-13T16:08:08 < karlp> and then only used three priority groups, but the dma irq always fires, and the timer and uart irqs never fire. 2013-02-13T16:08:26 -!- BrainDamage [~BrainDama@62.10.6.67] has quit [Remote host closed the connection] 2013-02-13T16:09:16 < Thorn> if they never fire and there's no interrupt flood I'd expect the problem is not in priorities 2013-02-13T16:09:24 < karlp> yeah, I think I just reaslied it's flooding. 2013-02-13T16:09:33 < karlp> and the timer irq never re-enables 2013-02-13T16:09:41 < karlp> will check. 2013-02-13T16:09:55 < dirty_d> fun 2013-02-13T16:10:35 < karlp> yup, seems to be. 2013-02-13T16:10:49 < karlp> ok, priorities work as expected again :) 2013-02-13T16:13:26 < dirty_d> karlp, youre using tht open something or anotehr library? 2013-02-13T16:15:06 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has left ##stm32 ["Konversation terminated!"] 2013-02-13T16:17:41 < karlp> libopencm3 yes. 2013-02-13T16:23:40 < dirty_d> i dont think it supports f3 yet 2013-02-13T16:24:02 -!- _BJFreeman [~bjfree@190.sub-75-196-115.myvzw.com] has joined ##stm32 2013-02-13T16:24:39 -!- _BJFreeman is now known as BJFreeman 2013-02-13T16:28:15 < dirty_d> i freakin hate monthly reports 2013-02-13T16:28:22 < dirty_d> i never keep track of anything i do, lol 2013-02-13T16:28:26 < karlp> dirty_d: there's a pull request adding a lot of it, it's not substantially different. 2013-02-13T16:29:14 < dirty_d> karlp, yea, im kinda content with no library 2013-02-13T16:29:36 < dirty_d> simple enough just to use the regs directly 2013-02-13T16:30:30 < karlp> sometimes I like the library, sometimes I don't... 2013-02-13T16:30:43 < karlp> one thing locm3 has it doxygen on the .c files, and nothing in the .h files. 2013-02-13T16:31:02 < karlp> so netbeans doesn't give me any real documentation inline, which is a pity 2013-02-13T16:31:10 < dirty_d> ahh 2013-02-13T16:31:31 < dirty_d> i like not using a lib, because then all i have to read is the datasheet 2013-02-13T16:32:36 < zyp> fewer layers to check for bugs when you discover errant behavior too :p 2013-02-13T16:33:12 < dirty_d> yea 2013-02-13T16:33:26 < dirty_d> plus you learn how the mcu actually works at a low level 2013-02-13T16:35:05 < dirty_d> i always through SPI was more complex than I2C 2013-02-13T16:35:09 < dirty_d> its actually really simple 2013-02-13T16:35:17 < dirty_d> ive never used it before 2013-02-13T16:36:03 < dirty_d> the manual says to enable the MSTR bit while the internal NSS is high 2013-02-13T16:36:10 < dirty_d> then enable the peripheral 2013-02-13T16:36:16 < dirty_d> if i do that MSTR is low afterward 2013-02-13T16:36:23 < dirty_d> if i set NSS, enable, then set MSTR 2013-02-13T16:36:28 < dirty_d> MSTR is high after 2013-02-13T16:36:54 < dirty_d> is the manual wrong? 2013-02-13T16:36:59 < zyp> huh? 2013-02-13T16:37:39 < zyp> enable how? 2013-02-13T16:38:24 < jpa-> i never bothered to use the NSS.. GPIO for the chip select is way easier for most cases 2013-02-13T16:38:25 < dirty_d> http://pastebin.com/dBhzPdTi 2013-02-13T16:38:27 < dirty_d> number 12 2013-02-13T16:38:46 < dirty_d> zyp, SPI1->CR1 |= SPI_CR1_SPE 2013-02-13T16:39:00 < dirty_d> jpa-, that is what im doing 2013-02-13T16:39:05 < dirty_d> but there is an internal NSS state 2013-02-13T16:39:15 < dirty_d> that needs to be high for the SPI master to work i guess 2013-02-13T16:39:47 < dirty_d> In master mode, the internal NSS signal must stay at a high level during the complete sequence 2013-02-13T16:40:35 < jpa-> i have just done this and it has worked fine for me SPI1->CR1 = SPI_CR1_SPE | (0 << 3) | SPI_CR1_MSTR | SPI_CR1_SSM | SPI_CR1_SSI; 2013-02-13T16:40:45 < dirty_d> i set the internal NSS high by enabling SSM and SSI 2013-02-13T16:41:09 < dirty_d> jpa-, if i do that, then MSTR is not set afterward if i check it 2013-02-13T16:41:22 < dirty_d> but if i set MSTR after i set SPE, it is 2013-02-13T16:41:24 < karlp> just don't even try and get it to use NSS for regualr chip select 2013-02-13T16:41:31 < karlp> it's only there for some stupid multi master mode. 2013-02-13T16:41:35 < dirty_d> right 2013-02-13T16:42:27 < zyp> hmm, I checked my code 2013-02-13T16:42:38 < zyp> I just set SSM, SSI, SPE and MSTR all at once 2013-02-13T16:42:45 < dirty_d> hmmm 2013-02-13T16:42:46 < jpa-> so same as i 2013-02-13T16:42:54 < dirty_d> does MSTR stay high? 2013-02-13T16:43:05 < dirty_d> oh wait 2013-02-13T16:43:13 < dirty_d> im setting SPE after everything else 2013-02-13T16:43:26 < dirty_d> because it said all that stuff has to be set before you enable it 2013-02-13T16:43:35 < dirty_d> but at the same time is ok? 2013-02-13T16:43:40 < zyp> works for me 2013-02-13T16:43:43 < dirty_d> ok 2013-02-13T16:43:57 < jpa-> works for me also, i never read that initialization sequence stuff anyway :) 2013-02-13T16:44:15 < dirty_d> then how the hell do you know how to activate it? lol 2013-02-13T16:44:48 < zyp> just by reading the register description? 2013-02-13T16:44:59 < jpa-> yeah 2013-02-13T16:45:09 < zyp> it's like two control registers, and the second one only does interrupt/dma stuff 2013-02-13T16:45:10 < dirty_d> there are all kinda of quirks that the reg descriptions dont always point out though 2013-02-13T16:45:24 < dirty_d> like this has to bappen before this etc 2013-02-13T16:45:30 < jpa-> it is true that there are 2013-02-13T16:45:40 < jpa-> but if it doesn't work i'll go and read more :) 2013-02-13T16:45:44 < dirty_d> yea 2013-02-13T16:45:53 < zyp> same here 2013-02-13T16:46:05 < zyp> I just tend to poke the registers directly in the debugger first 2013-02-13T16:46:16 < zyp> then type out the code when I've figured out what I need to write 2013-02-13T16:46:19 < dirty_d> thats a good idea 2013-02-13T16:48:37 < Laurenceb> in b4 dongs 2013-02-13T16:49:01 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-13T16:50:37 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has joined ##stm32 2013-02-13T16:57:48 -!- Sundiver_ [~angel@174-124-17-235.dyn.centurytel.net] has joined ##stm32 2013-02-13T16:58:12 < Sundiver_> Hello! 2013-02-13T16:58:26 < zyp> hi 2013-02-13T17:08:14 < emeb_mac> interesting news on STM32F42x - sdram + sai. 2013-02-13T17:08:30 < zyp> sai? 2013-02-13T17:08:41 < emeb_mac> serial audio interface 2013-02-13T17:08:55 < emeb_mac> enhanced I2S + others 2013-02-13T17:08:56 < zyp> oh 2013-02-13T17:09:03 < zyp> how many channels does that support? 2013-02-13T17:09:07 < jpa-> i wonder if they will have cache for the sdram 2013-02-13T17:09:11 < emeb_mac> no idea yet 2013-02-13T17:09:24 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [] 2013-02-13T17:13:13 -!- Niedar [~nnscript@ip68-110-154-15.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )] 2013-02-13T17:18:21 < Laurenceb> wow tons of F4s now 2013-02-13T17:18:21 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has quit [Quit: Ex-Chat] 2013-02-13T17:19:21 < Thorn> f4discovery v2 with sdram for $20? 2013-02-13T17:19:21 <+Steffanx> TONS 2013-02-13T17:19:32 < Laurenceb> wut? 2013-02-13T17:19:48 <+Steffanx> that was a question.. nothing else Laurenceb 2013-02-13T17:19:54 < Laurenceb> :( 2013-02-13T17:20:45 <+Steffanx> Wonder when they'll come with this L4 2013-02-13T17:22:12 <+Steffanx> Or the stm32F6. google mentions it :P 2013-02-13T17:24:24 < Laurenceb> http://translate.google.co.uk/translate?hl=en&sl=zh-CN&u=http://bbs.eeworld.com.cn/thread-340133-1-1.html&prev=/search%3Fq%3Dstm32f5%26hl%3Den%26safe%3Doff%26client%3Dubuntu%26hs%3DqXf%26tbo%3Dd%26channel%3Dfs%26gl%3Duk&sa=X&ei=tq8bUdGCIdKV0QWHooGQCA&ved=0CHkQ7gEwCA 2013-02-13T17:24:31 < Laurenceb> "The next year will be listed stm32F5 series, clocked really soared to 250MHz, the core should remain cortex m4" 2013-02-13T17:24:46 < dongs> totally legit 2013-02-13T17:24:56 < Laurenceb> seems legit 2013-02-13T17:25:14 <+Steffanx> Yeah, these jappies talk about it too: https://twitter.com/jshimada3256/statuses/270981201567825921 2013-02-13T17:25:51 < Laurenceb> lol @ avatars 2013-02-13T17:26:13 <+Steffanx> dongs's secret account 2013-02-13T17:26:18 -!- barthess [~barthess@86.57.155.106] has quit [Quit: Leaving.] 2013-02-13T17:27:15 < Robint91> STM32F5 O_0 2013-02-13T17:27:30 < dongs> lol animu trash 2013-02-13T17:27:47 <+Steffanx> oh, ntfreak_ is our guy for info about new stuff i see 2013-02-13T17:27:59 <+Steffanx> He mentioned the stm32f439 back in july 2013-02-13T17:28:43 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-62.pptp.stw-bonn.de] has joined ##stm32 2013-02-13T17:28:48 < ntfreak_> it has taken ST that long to mess it up some more :) 2013-02-13T17:29:23 < Robint91> ntfreak_, what is about that F5? 2013-02-13T17:29:35 <+Steffanx> NDA, NDA, NDA :P 2013-02-13T17:30:05 < ntfreak_> all i can say is soon 2013-02-13T17:30:12 < Robint91> Steffanx, I have signed enough of NDA's 2013-02-13T17:30:20 < Robint91> ntfreak_, shall we trade NDAs 2013-02-13T17:30:27 < Robint91> I have a few from omnivision 2013-02-13T17:30:28 < Robint91> :p 2013-02-13T17:30:38 < dongs> omnivision is usless though 2013-02-13T17:30:45 < dongs> i have a NDA w/DTLA 2013-02-13T17:30:56 < Robint91> dongs, omnivision> aptina 2013-02-13T17:31:00 < dongs> really? 2013-02-13T17:31:05 < dongs> i'd have to disagree 2013-02-13T17:31:46 < Laurenceb> omgivision 2013-02-13T17:31:55 < dongs> omgnigvision 2013-02-13T17:32:00 < dongs> do they have anythign 4k @ 30fps? 2013-02-13T17:32:11 < Robint91> dongs, nope 2013-02-13T17:32:11 < dongs> orrrr anything > 240fps? 2013-02-13T17:32:13 < dongs> well then. 2013-02-13T17:32:19 < dongs> omtivision < aptina 2013-02-13T17:32:36 < Laurenceb> wtf 2013-02-13T17:32:43 < Laurenceb> 8 uarts on F437 2013-02-13T17:32:56 < Laurenceb> what ardutards fap to 2013-02-13T17:33:16 <+Steffanx> And you fap to the bare metal, what's the difference? 2013-02-13T17:33:25 < Robint91> dongs, they have 4k @ 60fps 2013-02-13T17:33:37 < Robint91> or 13M @30fps 2013-02-13T17:33:42 < Laurenceb> wtf 2013-02-13T17:33:46 < Laurenceb> madness 2013-02-13T17:33:56 < Thorn> what's the use for chips like these? they're not SoCs and they're not microcontrollers anymore 2013-02-13T17:34:07 < gxti> why not 2013-02-13T17:34:12 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T17:34:38 < Thorn> 250MHz and SDRAM assume something more advanced than freertos 2013-02-13T17:34:42 < Robint91> Thorn, they are a hybrid of soc and microcontrollers that only run bloated OS systems 2013-02-13T17:34:47 < gxti> does it have a mmu? 2013-02-13T17:34:51 < Robint91> Thorn, JAVA 2013-02-13T17:35:02 <+Steffanx> LOL 2013-02-13T17:35:12 < Thorn> wat, java only? 2013-02-13T17:35:14 < Robint91> Thorn, I hope that on the F5 it is atleast DDR SDRAM 2013-02-13T17:35:19 <+Steffanx> ( Robint91 tries to start a flamewar, again? ) 2013-02-13T17:35:36 < Robint91> Thorn, ST is going to push to run java on those I think 2013-02-13T17:35:36 < Laurenceb> on here theres a flamewar by default 2013-02-13T17:35:45 < Laurenceb> eww 2013-02-13T17:35:49 <+Steffanx> only when you are around Laurenceb 2013-02-13T17:38:19 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-13T17:38:47 < Robint91> what process node would be used? 2013-02-13T17:38:51 < Robint91> 90nm or 65nm 2013-02-13T17:38:55 < Robint91> or even 40nm 2013-02-13T17:39:06 < Robint91> I know that ST uses now mostly 90nm 2013-02-13T17:39:15 < Robint91> and TI and others 65nm 2013-02-13T17:39:24 < Robint91> (I was told by Avnet guys) 2013-02-13T17:40:34 <+Steffanx> Hmpf, why some chinese are so slow? 2013-02-13T17:40:54 < Robint91> Is TI the only one with cortex-R mcus? 2013-02-13T17:42:38 < Laurenceb> what features does cortex r have? 2013-02-13T17:42:44 < zyp> probably, cortex-r has a different target audience 2013-02-13T17:43:07 < Robint91> it is more for realtime things 2013-02-13T17:43:12 < zyp> yep 2013-02-13T17:43:21 < Robint91> and binary compatible with ARM9 and ARM11 cpu's 2013-02-13T17:44:01 < zyp> we're using cortex-r in stuff at work 2013-02-13T17:44:12 < Robint91> zyp, TI parts? 2013-02-13T17:44:26 < zyp> no, our parts 2013-02-13T17:44:33 < Robint91> custom 2013-02-13T17:44:34 < Robint91> NICE 2013-02-13T17:44:59 < zyp> well, I work for a company producing chips :p 2013-02-13T17:45:24 < Robint91> zyp, may we know which one? 2013-02-13T17:45:32 < Laurenceb> ST 2013-02-13T17:45:33 < zyp> ST-Ericsson 2013-02-13T17:45:45 < Robint91> O_0 2013-02-13T17:46:15 < Robint91> zyp, are those used in the communication modems? 2013-02-13T17:46:27 < zyp> correct 2013-02-13T17:46:52 < Robint91> nice 2013-02-13T17:48:32 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Quit: pokðt] 2013-02-13T17:48:35 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-13T17:51:48 -!- alexn [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-13T17:55:37 < Erlkoenig> when exactly is a company "industry-leading"? :D 2013-02-13T17:55:53 < gxti> when it hires a marketing person 2013-02-13T17:55:59 < zyp> yeah :p 2013-02-13T17:56:10 < Erlkoenig> sounds like a strategy 2013-02-13T17:57:34 <+Steffanx> Ha Erlkoenig :P 2013-02-13T17:58:24 -!- |sark [~|sark@37.247.88.155] has quit [Read error: Connection reset by peer] 2013-02-13T17:59:37 -!- |sark [~|sark@37.247.88.155] has joined ##stm32 2013-02-13T18:01:34 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Quit: trepidaciousMBR] 2013-02-13T18:03:10 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T18:36:26 -!- Viper168_ [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-13T18:36:59 < Laurenceb> whats the M4f fpu multiply asm? 2013-02-13T18:37:08 < Laurenceb> fmuls or something? 2013-02-13T18:41:49 < Erlkoenig> vmul ? 2013-02-13T18:42:39 < Erlkoenig> A7.7.241 VMUL Floating-point Multiply multiplies two single-precision register values, and places the result in the destination single-precision register. 2013-02-13T18:43:47 < Erlkoenig> page 195 for more FPU instructions 2013-02-13T18:46:19 < Laurenceb> ah thanks 2013-02-13T18:46:24 < Laurenceb> just debugging asm output 2013-02-13T18:47:01 -!- barthess [~barthess@5.100.212.152] has joined ##stm32 2013-02-13T18:48:08 < dirty_d> hmm 2013-02-13T18:48:11 < dirty_d> im confused 2013-02-13T18:48:19 < dirty_d> (gdb) x SPI1->SR 0x403: 0x80f44142 2013-02-13T18:48:28 < dirty_d> thats the wrong address isnt it 2013-02-13T18:49:04 < dirty_d> the offset of that reg is 8 2013-02-13T18:52:04 < dirty_d> gdb is being weird 2013-02-13T18:52:20 < dirty_d> why is it not stepping into the body of a while loop? 2013-02-13T18:52:37 < dirty_d> stepi just keeps showing the while part 2013-02-13T18:54:17 < Erlkoenig> instruction reordering because of optimization? step through assembly lines 2013-02-13T18:55:42 < dirty_d> is there a different stepp command for that? 2013-02-13T18:56:55 < dirty_d> what the hell, its not even executing the body of the loop 2013-02-13T18:57:17 < Erlkoenig> "layout asm" 2013-02-13T18:57:48 < Erlkoenig> and then probably stepi 2013-02-13T18:58:32 < dirty_d> how do you go back to normal from layout asm? 2013-02-13T18:58:57 < Erlkoenig> help layout 2013-02-13T18:58:58 < Erlkoenig> :D 2013-02-13T18:59:55 < dirty_d> i dont get how my loop body is getting optimized out 2013-02-13T19:00:02 < dirty_d> i have an asm volatile("nop") in there 2013-02-13T19:00:13 < dirty_d> i thought thats not supposed to get messed with 2013-02-13T19:00:49 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Quit: trepidaciousMBR] 2013-02-13T19:01:06 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-13T19:01:09 < Erlkoenig> look at the disassembly via objdump -d yourprog.elf 2013-02-13T19:01:44 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-13T19:02:15 < Erlkoenig> are you making a delay loop? evil 2013-02-13T19:02:22 < dirty_d> no 2013-02-13T19:02:27 < dirty_d> waiting for a status register to change 2013-02-13T19:03:00 < dirty_d> the body of the loop would normally be empty 2013-02-13T19:03:07 < Erlkoenig> ah. 2013-02-13T19:03:11 < dirty_d> but i added somethign to it and i noticed its not being executed 2013-02-13T19:03:27 < Erlkoenig> status register declared volatile? :) 2013-02-13T19:03:39 < dirty_d> and im not getting the right value back when i `print SPI1->SR` 2013-02-13T19:03:42 < dirty_d> yup 2013-02-13T19:03:48 < dirty_d> im using STs header 2013-02-13T19:04:33 < Erlkoenig> does the disassembly look ok? 2013-02-13T19:04:46 < dirty_d> hard to tell, i dont relaly know assembly 2013-02-13T19:04:53 < dirty_d> 0x80005de b.n 0x80005de 2013-02-13T19:04:54 < dirty_d> stuck there 2013-02-13T19:05:15 < Erlkoenig> duh 2013-02-13T19:05:22 < Erlkoenig> pastebin the C code and disassembly :) 2013-02-13T19:05:39 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T19:09:25 < trepidaciousMBR> Does anyone know of any code for using STM32F4 as a usb host, controlling a mass storge device? 2013-02-13T19:10:17 < Erlkoenig> ST's example codes? 2013-02-13T19:11:06 < dirty_d> Erlkoenig, ok 2013-02-13T19:12:19 < dirty_d> Erlkoenig, http://pastebin.com/w9CwS8K4 2013-02-13T19:14:00 < dirty_d> print SPI1->SR give me 0x403, print sr gives me 0x1082 2013-02-13T19:15:39 < trepidaciousMBR> Erlkoenig: Ah yeah, there is one here as well: http://vedder.se/2012/12/stm32f4-discovery-usb-host-and-mp3-player/ 2013-02-13T19:16:24 < Erlkoenig> dirty_d: ah funny 2013-02-13T19:16:28 < Laurenceb> sweet 2013-02-13T19:16:36 < Erlkoenig> dirty_d: GCC "unrolled" the while-loop, to 2x the body 2013-02-13T19:16:59 < Erlkoenig> i.e. something like while(1) { if (SPI->SR ...) break; if (SPI->SR...) break; } 2013-02-13T19:17:13 <+Steffanx> l 2013-02-13T19:17:14 <+Steffanx> o 2013-02-13T19:17:14 <+Steffanx> o 2013-02-13T19:17:14 <+Steffanx> ks 2013-02-13T19:17:14 <+Steffanx> 2013-02-13T19:17:14 <+Steffanx> l 2013-02-13T19:17:16 <+Steffanx> o 2013-02-13T19:17:18 <+Steffanx> o 2013-02-13T19:17:20 <+Steffanx> p 2013-02-13T19:17:22 <+Steffanx> s 2013-02-13T19:17:24 <+Steffanx> f 2013-02-13T19:17:25 < Laurenceb> FAIL 2013-02-13T19:17:26 <+Steffanx> a 2013-02-13T19:17:28 <+Steffanx> OOPS 2013-02-13T19:17:30 <+Steffanx> :P 2013-02-13T19:17:32 < Laurenceb> die in a fire 2013-02-13T19:17:32 <+Steffanx> ENTER stucked 2013-02-13T19:17:35 < Erlkoenig> that's why gdb does funny stuff 2013-02-13T19:17:48 < dirty_d> Erlkoenig, why the hell did it do that? 2013-02-13T19:17:53 < Erlkoenig> it's probably faster :D 2013-02-13T19:18:02 < Erlkoenig> which does not matter in a wait-loop, but GCC doesn't know that :D 2013-02-13T19:18:23 < dirty_d> so it unrolls 2 iterations then sits in the real loop? 2013-02-13T19:18:33 < Erlkoenig> yes 2013-02-13T19:18:47 < Erlkoenig> try turning off loop unrolling via that compiler switch and look what happens :) 2013-02-13T19:18:48 < dirty_d> hmm 2013-02-13T19:18:49 < dirty_d> wtf 2013-02-13T19:18:59 < dirty_d> i compiled with O0 and now its not stuck in thta while loop 2013-02-13T19:19:18 < Erlkoenig> just set a breakpoint to 0x80005d2 which is "just after the loop" 2013-02-13T19:19:44 < Erlkoenig> oh wait, not just 2 but 3 iterations :) 2013-02-13T19:20:57 < dirty_d> Erlkoenig, yea but why is it working with -O0, but not -O2? 2013-02-13T19:21:06 < Erlkoenig> you can see 4x "[r3, #8]" which is SPI1->SR in that case... (probably r3 = &SPI1) one access to it is that sr = SPI1->SR, and the other three are the loop 2013-02-13T19:21:20 < dirty_d> as in the data gets sent and the TX fifo empty flag gets set 2013-02-13T19:21:32 < Erlkoenig> with -O2 no data is sent? 2013-02-13T19:21:49 < dirty_d> theoretically 2013-02-13T19:21:49 < Erlkoenig> (sending is triggered by writing SPI1->DR, right?) 2013-02-13T19:21:57 < Erlkoenig> verified with scope? :) 2013-02-13T19:22:01 < dirty_d> i have no way to know, but the TXE flag never gets set 2013-02-13T19:22:03 < dirty_d> i dont have one 2013-02-13T19:22:14 < Erlkoenig> wah :D 2013-02-13T19:22:18 < dirty_d> yea i want one 2013-02-13T19:22:19 < dirty_d> lol 2013-02-13T19:22:20 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T19:22:29 < Erlkoenig> don't have a micro at the other side receiving the bytes? 2013-02-13T19:22:40 < dirty_d> Erlkoenig, i do its the stmf3disco gyro 2013-02-13T19:22:43 < Erlkoenig> like an arduino, they're handy for quick'n'dirty debugging 2013-02-13T19:23:35 < Erlkoenig> theoretically optimization might cause an error before that whole thing, resulting in nothing being sent at all, and as a result the loop never finishes 2013-02-13T19:23:40 < dirty_d> Erlkoenig, the value i get from gdb when i do `print SPI1->SR` is the same for -O0 and -O2 2013-02-13T19:23:55 < dirty_d> 0x403 2013-02-13T19:23:58 < Erlkoenig> i dunno whether accessing periphal registers from gdb works as desired ;) 2013-02-13T19:23:59 < dirty_d> i guess thats right 2013-02-13T19:24:05 < dirty_d> 0x403 would be that TXE is set 2013-02-13T19:24:08 < dirty_d> and the loop should end 2013-02-13T19:24:09 < Erlkoenig> ah 2013-02-13T19:26:16 < dirty_d> im gonna check with that mdw command to see if i get the same 2013-02-13T19:26:20 < dirty_d> gotta look up the damn address 2013-02-13T19:27:01 < Erlkoenig> but your loop runs from 0x80005c0b0 to 0x80005d0 endlessly? 2013-02-13T19:27:28 -!- Ranewen [~Ranewen4@82.132.1.100] has joined ##stm32 2013-02-13T19:27:55 < Laurenceb> Disassembly of section .ccmram: 2013-02-13T19:27:55 < Laurenceb> 08011ec8 2013-02-13T19:28:02 < Laurenceb> ^ thats wrong ? 2013-02-13T19:28:28 < Erlkoenig>  80005d2:       e7fe            b.n     80005d2 --- this looks strange... an endless loop o.o 2013-02-13T19:28:45 < dirty_d> there are some endless loops 2013-02-13T19:28:50 < dirty_d> but not the one it was stuck in 2013-02-13T19:29:00 -!- Ranewen [~Ranewen4@82.132.1.100] has quit [Client Quit] 2013-02-13T19:29:06 < Erlkoenig> where (assembly lines/addresses) exactly is it stuck? 2013-02-13T19:29:15 < dirty_d> one sec 2013-02-13T19:29:33 < dirty_d> while(!(SPI1->SR & SPI_SR_TXE_gm)) 2013-02-13T19:29:44 < dirty_d> thats where gdb shows when i do `stepi` 2013-02-13T19:29:59 < dirty_d> 80005d2: e7fe b.n 80005d2 2013-02-13T19:30:07 < dirty_d> is where it did when i did `layout asm` 2013-02-13T19:30:14 < Erlkoenig> yeah 2013-02-13T19:30:14 < Erlkoenig> weird 2013-02-13T19:30:40 < Erlkoenig> that means TXE bit is actually set, and then the core enters that weird endless loop 2013-02-13T19:30:47 < dirty_d> hmm 2013-02-13T19:30:49 < Erlkoenig> maybe show the code after that 2013-02-13T19:30:53 < dirty_d> 0x48003008: bf000080 2013-02-13T19:31:01 < dirty_d> thats what mdw shows me for SPI1->SR 2013-02-13T19:31:10 < dirty_d> i think i got the address right 2013-02-13T19:31:28 < Erlkoenig> what's mdw? 2013-02-13T19:31:40 < dirty_d> its an openocd command 2013-02-13T19:31:48 < dirty_d> gets data at memory address 2013-02-13T19:32:21 < Erlkoenig> ah. 2013-02-13T19:32:22 < dirty_d> hmm 2013-02-13T19:32:25 < dirty_d> thats def wrong 2013-02-13T19:32:31 < dirty_d> not the right data 2013-02-13T19:32:35 < dirty_d> or wrong address 2013-02-13T19:32:42 < Erlkoenig> does OpenOCD work well with STLink/V2 on F4 Discovery currently? 2013-02-13T19:33:01 < dirty_d> i think so 2013-02-13T19:33:04 < dirty_d> im using it with f3 2013-02-13T19:33:06 < dirty_d> never used f4 2013-02-13T19:33:50 < Erlkoenig> i should try it... texane st-link seems a bit buggy/unstable 2013-02-13T19:34:51 < dirty_d> i havent had any problems 2013-02-13T19:35:23 < Erlkoenig> it has a gdb server? 2013-02-13T19:36:36 < dirty_d> yea 2013-02-13T19:38:04 < Laurenceb> 10000158 <__memset_veneer>: 2013-02-13T19:38:08 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Quit: trepidaciousMBR] 2013-02-13T19:38:10 < Laurenceb> wtf is a veneer function? 2013-02-13T19:39:47 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-13T19:40:09 < Laurenceb> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0203h/Babedcjj.html 2013-02-13T19:40:15 < Laurenceb> interesting 2013-02-13T19:40:53 < Laurenceb> https://raw.github.com/Laurenceb/Chibi-Spectro/master/build/dis.asm 2013-02-13T19:41:00 < Laurenceb> im getting them in my ccmram 2013-02-13T19:42:45 < Laurenceb> bbl 2013-02-13T19:44:25 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has quit [Ping timeout: 256 seconds] 2013-02-13T19:47:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep] 2013-02-13T19:51:31 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Read error: Connection reset by peer] 2013-02-13T19:53:21 < dirty_d> Erlkoenig, i dont know what the hell i did, but its working now 2013-02-13T19:53:26 < dirty_d> i actually didnt do anything 2013-02-13T19:53:52 < Erlkoenig> bwahaha 2013-02-13T19:54:01 < Erlkoenig>  80005d2:       e7fe            b.n     80005d2 is this line still present? 2013-02-13T19:54:59 < gxti> Erlkoenig: yes openocd does stlinkv2 and f4disco 2013-02-13T19:56:35 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T19:57:51 < dirty_d> Erlkoenig, i havent check it out yet 2013-02-13T19:58:20 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-13T19:58:23 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-13T19:58:36 -!- Ranewen [~Ranewen4@82.132.1.100] has joined ##stm32 2013-02-13T19:59:09 < dirty_d> Erlkoenig, looks like its not there 2013-02-13T20:00:25 < Erlkoenig> strange... 2013-02-13T20:00:33 < Erlkoenig> with no change at the code and compiler options? 2013-02-13T20:01:38 < dirty_d> i thik the only thing i changes was taking asm volatile("nop") out of the loop body 2013-02-13T20:02:24 < dirty_d> put it back and it still works 2013-02-13T20:02:28 < dirty_d> weird 2013-02-13T20:03:38 < dirty_d> i recieve something from the gyro too 2013-02-13T20:03:44 < dirty_d> but not what im suposed to 2013-02-13T20:03:54 < dirty_d> i get 0xffff 2013-02-13T20:04:10 < dirty_d> which i bet means its just not listening and MISO is just low 2013-02-13T20:04:13 < dirty_d> i mean high 2013-02-13T20:04:17 < dirty_d> or something 2013-02-13T20:04:24 < dirty_d> but otherwise SPI is working on my side 2013-02-13T20:04:27 -!- Simon-- [~sim@2606:6a00:0:28:5604:a6ff:fe02:702b] has quit [Read error: Operation timed out] 2013-02-13T20:05:00 -!- fergusnoble [fergusnobl@repl.esden.net] has quit [Read error: Operation timed out] 2013-02-13T20:05:51 -!- Simon-- [~sim@2606:6a00:0:28:5604:a6ff:fe02:702b] has joined ##stm32 2013-02-13T20:06:49 < Erlkoenig> get a scope... :-) 2013-02-13T20:07:05 < dirty_d> yea i really should 2013-02-13T20:08:08 -!- fergusnoble [fergusnobl@repl.esden.net] has joined ##stm32 2013-02-13T20:08:12 < Erlkoenig> OwonSDS: 400-500€, 100MHz, 1 GS/s, 10 MByte, Ethernet, USB... ;) 2013-02-13T20:13:04 < jpa-> do i really need that sync_synchronize there? because it doesn't work without http://paste.dy.fi/RUn/plain 2013-02-13T20:13:18 < dirty_d> Erlkoenig, isnt the rigon better? 2013-02-13T20:13:23 < jpa-> only thing that should matter is that DMA1->IFCR is set before TIM4->DIER 2013-02-13T20:13:40 < Erlkoenig> dirty_d: yes, and is twice as expensive :P 2013-02-13T20:13:49 < dirty_d> twice? 2013-02-13T20:13:54 < dirty_d> the rigol is $400 right 2013-02-13T20:13:56 < dirty_d> that is $200? 2013-02-13T20:14:00 < Erlkoenig> what "the" rigol :D 2013-02-13T20:14:05 < dirty_d> the basic one 2013-02-13T20:14:05 < Erlkoenig> it probably has much lower specs 2013-02-13T20:14:09 < dirty_d> 50MHz hackableto 100 2013-02-13T20:15:11 -!- Ranewen [~Ranewen4@82.132.1.100] has quit [Quit: Leaving] 2013-02-13T20:15:35 < Erlkoenig> Rigol DS1052E ? 2013-02-13T20:16:14 < Erlkoenig> my Owon SDS has 100MHz and a 800x600 TFT compared to the 320x234 of the Rigol 2013-02-13T20:16:39 < dirty_d> yea i like the screen size 2013-02-13T20:16:47 < dirty_d> Erlkoenig, soyou have no complaints? 2013-02-13T20:17:21 < Erlkoenig> but if you just want to do SPI debugging and stuff, get a $10 USB Saleae logic analyzer clone :D 2013-02-13T20:17:38 < dirty_d> which model do you have? 2013-02-13T20:17:49 < Erlkoenig> Owon SDS 7102 2013-02-13T20:18:09 < dirty_d> $430? 2013-02-13T20:18:20 <+Steffanx> euros here and there :P 2013-02-13T20:18:36 < Erlkoenig> hmm was 450€ from a german dealer... more expensive but with warranty and stuff ^^ 2013-02-13T20:18:43 < dirty_d> ok 2013-02-13T20:18:47 <+Steffanx> euro stuff is always more expensive 2013-02-13T20:19:05 <+Steffanx> replace the $ with € and add a few euro :) 2013-02-13T20:19:31 < Erlkoenig> yeah probably... but you won't need to worry about customs... 2013-02-13T20:22:03 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T20:25:09 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T20:27:11 -!- neuro_sys|off is now known as neuro_sys 2013-02-13T20:38:17 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-13T20:39:47 < dirty_d> any of you actually done SPI yourself to the gyro on the f3disco? 2013-02-13T20:40:12 < zyp> is it using the L3GD20? 2013-02-13T20:40:31 < dirty_d> ye 2013-02-13T20:40:42 < zyp> I'm using that on my own board 2013-02-13T20:40:44 < zyp> http://cgit.jvnv.net/suzumebachi/tree/drivers/l3gd20.h 2013-02-13T20:41:02 < Erlkoenig> i have SPI code for the accelerometer on the F4... ^^ 2013-02-13T20:43:09 < dirty_d> zyp, the second clock transition is the data capture edge? 2013-02-13T20:43:30 < dirty_d> maybe thats why my code isnt working 2013-02-13T20:43:34 < dirty_d> i hace CPHA = 0 2013-02-13T20:43:53 < zyp> I don't remember the specifics, but my code works on my board 2013-02-13T20:44:17 < dirty_d> hey! 2013-02-13T20:44:20 < dirty_d> i think that fixed it 2013-02-13T20:44:33 < dirty_d> recieved 0xffd4 for WHO_AM_I 2013-02-13T20:45:19 < dirty_d> yup 2013-02-13T20:45:21 < dirty_d> thats right 2013-02-13T20:45:22 < dirty_d> sweet 2013-02-13T20:45:56 < dirty_d> so high bits are whats sent on SPI when there is nothing to send i guess 2013-02-13T20:46:12 < dirty_d> or is it not even standard 2013-02-13T20:46:47 < zyp> it's undefined 2013-02-13T20:47:09 < dirty_d> zyp, in your cs.on()/cs.off() are you just turning the GPIO pin on with no delay or anything? 2013-02-13T20:47:15 < zyp> yes 2013-02-13T20:47:23 < dirty_d> with 50MHz setting 2013-02-13T20:47:24 < dirty_d> ? 2013-02-13T20:47:32 < zyp> uh, I don't remember the speed 2013-02-13T20:49:04 < zyp> that code was just quickly thrown together, I'm intending to write a nicer SPI abstraction, but I haven't figured out a nice API for it yet 2013-02-13T20:49:36 < dirty_d> ok 2013-02-13T20:50:22 < zyp> I2C is easier to abstract since each transaction only can be either read or write, not both at the same time :p 2013-02-13T20:51:11 < dirty_d> yea 2013-02-13T20:51:22 < dirty_d> you ever get this after using load in gdb? WWDG_IRQHandler () at startup_stm32f30x.s:119 119 b Infinite_Loop 2013-02-13T20:51:39 < dirty_d> i have to do a monitor reset halt again to get it to run normally 2013-02-13T20:52:04 < zyp> you mean just after typing load, no other commands? 2013-02-13T20:52:07 < zyp> or what? 2013-02-13T20:53:11 < zyp> it's pretty obvious that you have to reset the cpu after flashing new contents, you can't just continue from the previous state :p 2013-02-13T20:53:31 < dirty_d> no 2013-02-13T20:53:33 < dirty_d> i do load 2013-02-13T20:53:34 < dirty_d> c 2013-02-13T20:53:38 < dirty_d> ctrl-c 2013-02-13T20:53:40 < dirty_d> then i see that 2013-02-13T20:53:50 < zyp> exactly 2013-02-13T20:53:54 < zyp> c means continue 2013-02-13T20:53:54 < dirty_d> but only about half the time 2013-02-13T20:54:11 < dirty_d> so halt, load, halt, c 2013-02-13T20:54:12 < dirty_d> ? 2013-02-13T20:54:31 < zyp> um, I don't do openocd 2013-02-13T20:54:35 < zyp> but I do load, then run 2013-02-13T20:54:45 < zyp> and run does a full system reset 2013-02-13T20:54:54 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32 2013-02-13T20:55:09 < zyp> I hear that it doesn't work with openocd currently though 2013-02-13T20:56:12 < dirty_d> what do you use? 2013-02-13T20:56:20 < dirty_d> im using gdb through openocd 2013-02-13T20:56:33 < zyp> I'm using black magic probe 2013-02-13T20:58:34 < dirty_d> oh 2013-02-13T21:03:03 < gxti> what i used before, which isn't necessarily indicative of correctness, was 'monitor reset halt' 2013-02-13T21:03:17 < gxti> but i also used openocd config to flash instead of gdb 2013-02-13T21:03:21 < gxti> so shrug 2013-02-13T21:07:03 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Ping timeout: 276 seconds] 2013-02-13T21:23:54 < karlp> you need a master build of oocd to have proper extended remote 2013-02-13T21:24:04 < karlp> otherwise you need monitor reset halt; load; continue 2013-02-13T21:24:30 < karlp> for exaactly hwat zyp was saying. 2013-02-13T21:25:46 < gxti> zyp: so re: SD and SPI, is that a sensible thing to do? f107 doesn't have sdio 2013-02-13T21:26:02 < zyp> are you sure? 2013-02-13T21:26:02 < emeb> bitbang it. 2013-02-13T21:26:46 < gxti> unless it's a secret peripheral 2013-02-13T21:27:16 < zyp> oh, according to reference manual, it's only available in high- and XL-density F1 devices, not connectivity line 2013-02-13T21:27:32 < zyp> no luck then 2013-02-13T21:28:07 < gxti> but is spi a useful alternative or should i just bitbang? 2013-02-13T21:28:22 < emeb> KIDDING! 2013-02-13T21:28:42 < gxti> well, whatever works. although the bitbang code obviously will be at least 8x uglier 2013-02-13T21:28:48 < gxti> it's not like i need speed 2013-02-13T21:28:51 < emeb> and slower 2013-02-13T21:31:11 < zyp> forget about bitbanging. 2013-02-13T21:31:42 < gxti> k, does spi work y/n 2013-02-13T21:31:52 < zyp> sure 2013-02-13T21:32:00 < gxti> kthx 2013-02-13T21:32:17 < emeb> "works for me" 2013-02-13T21:40:45 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Read error: Connection reset by peer] 2013-02-13T21:41:23 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32 2013-02-13T21:42:57 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: ….] 2013-02-13T21:43:14 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-13T21:43:14 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-13T21:43:14 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-13T21:43:17 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-13T22:02:47 -!- daku is now known as DaKu 2013-02-13T22:04:36 -!- Laurenceb_ [~Laurence@host86-177-210-6.range86-177.btcentralplus.com] has joined ##stm32 2013-02-13T22:05:56 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-13T22:16:54 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32 2013-02-13T22:16:54 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host] 2013-02-13T22:16:54 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32 2013-02-13T22:17:58 < Laurenceb_> if ive got routines in CCM, I need a loop to copy them to flash in startup right? 2013-02-13T22:18:12 < zyp> yes 2013-02-13T22:18:23 < zyp> well, from flash 2013-02-13T22:19:02 <+Steffanx> Actually, you don't NEED to. Can't you load it whenever you want? 2013-02-13T22:19:24 < zyp> sure, as long as it's done before you call it 2013-02-13T22:20:52 < Laurenceb_> this is weird 2013-02-13T22:21:03 <+Steffanx> Yes 2013-02-13T22:21:07 < Laurenceb_> the "CCM" ld script from chibios is just putting heap and stack in ccm 2013-02-13T22:22:19 -!- evil_dan2wik [~evil_dan2@unaffiliated/evil-dan2wik/x-0106201] has quit [Quit: Cheese Burger] 2013-02-13T22:23:21 -!- evil_dan2wik [~evil_dan2@unaffiliated/evil-dan2wik/x-0106201] has joined ##stm32 2013-02-13T22:23:41 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/STM32F407xG_CCMRAM.ld#L147 2013-02-13T22:24:13 < Laurenceb_> atm ive got that, but ill need something in startup and > ccmram AT > flash ? 2013-02-13T22:26:16 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has joined ##stm32 2013-02-13T22:27:40 < Laurenceb_> " core coupled memory? wtf? I've never heard of that before" 2013-02-13T22:27:46 < Laurenceb_> chibifail 2013-02-13T22:28:53 < gxti> ldscripts have nothing to do with the copying 2013-02-13T22:28:56 < emeb> Laurenceb_: I've got the right linker stuff and startup stuff to copy code from flash to ccm in that f303_sdr project. 2013-02-13T22:28:58 < Tectu> just because I haven't heard of it before dosen't mean that it's chibi fail 2013-02-13T22:29:10 <+Steffanx> IGNORE, ignore him Tectu 2013-02-13T22:29:11 < Laurenceb_> emeb: k ill check your project 2013-02-13T22:29:15 < gxti> but chibi ldscripts do sort of fail 2013-02-13T22:29:37 < Tectu> Steffanx, kay 2013-02-13T22:29:47 < gxti> all that fake align shit 2013-02-13T22:29:54 < emeb> Laurenceb_: here -> https://github.com/emeb/f303_sdr 2013-02-13T22:29:58 < gxti> caused a wrong-data-loaded issue for me 2013-02-13T22:30:11 -!- barthess [~barthess@5.100.212.152] has quit [Quit: Leaving.] 2013-02-13T22:30:16 < gxti> that was fun, especially since the wrong data was a pointer. hello hard fault. 2013-02-13T22:31:55 < Laurenceb_> emeb: so you have stuff in your startup? 2013-02-13T22:32:15 < gxti> if ccmram is expected to contain data then something will have to copy it 2013-02-13T22:32:29 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: Connection reset by peer] 2013-02-13T22:32:31 < gxti> so look at the startup you have and if it doesn't say ccmram then you might have to write something 2013-02-13T22:32:35 < Laurenceb_> ah i see 2013-02-13T22:32:43 < Erlkoenig> emeb: no license information... 2013-02-13T22:32:50 < gxti> linker scripts don't do anything at runtime, they just put the crap in flash 2013-02-13T22:32:52 < Laurenceb_> i cant even find the chibios startup 2013-02-13T22:32:57 < gxti> something has to take it out of flash and it in ram 2013-02-13T22:32:59 < Laurenceb_> is it in c or something 2013-02-13T22:33:03 < gxti> yes 2013-02-13T22:33:07 < gxti> asm is for jerks 2013-02-13T22:33:07 < Laurenceb_> ah 2013-02-13T22:33:09 < Laurenceb_> lol 2013-02-13T22:33:12 < Laurenceb_> where? 2013-02-13T22:33:14 < gxti> it's under ports/GCC/ARMCMx 2013-02-13T22:34:43 < emeb> Laurenceb_: yeah - https://github.com/emeb/f303_sdr/blob/master/CMSIS/startup_stm32f30x.s#L96 2013-02-13T22:34:50 < Laurenceb_> i see 2013-02-13T22:34:57 < Laurenceb_> ill have to faff with chibios 2013-02-13T22:35:03 < Laurenceb_> and use my own startup 2013-02-13T22:35:22 < gxti> not rly, you can just stick it at the top of main 2013-02-13T22:35:24 < gxti> or whatever 2013-02-13T22:35:31 < emeb> Erlkoenig: It's in main.c -> https://github.com/emeb/f303_sdr/blob/master/main.c#L7 2013-02-13T22:35:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32 2013-02-13T22:35:43 < gxti> you should fix the linker script though, _textdata = LOADADDR(.data); 2013-02-13T22:36:14 < Erlkoenig> emeb: ah. maybe a LICENSE.txt would be easier/faster to find 2013-02-13T22:36:19 < Laurenceb_> my simple .ccmram section thingy seems to work 2013-02-13T22:36:21 < gxti> Laurenceb_: what are you loading into ccmram? 2013-02-13T22:36:24 < Laurenceb_> just need to load it 2013-02-13T22:36:29 < Laurenceb_> some dsp functions 2013-02-13T22:36:45 < gxti> i don't actually know what ccmram is so i guess i should rtfm 2013-02-13T22:36:55 < gxti> haven't graduated to f3/f4 yet 2013-02-13T22:38:12 < Laurenceb_> so crt0.c ? 2013-02-13T22:38:54 < Laurenceb_> ill need to move that out of chibios 2013-02-13T22:38:56 < gxti> looks like you can define a __late_init() 2013-02-13T22:39:01 < gxti> and it'll call that before constructors 2013-02-13T22:39:06 < Laurenceb_> hmm 2013-02-13T22:39:13 < Laurenceb_> so i can have that elsewhere? 2013-02-13T22:39:15 < gxti> but again i doubt you'd have any issue just doing it in main() 2013-02-13T22:39:34 < gxti> yeah it's a weak symbol, just define one in your code 2013-02-13T22:40:00 < Laurenceb_> ok cool 2013-02-13T22:40:03 < gxti> and copy how it does data load 2013-02-13T22:40:08 < Laurenceb_> yeah 2013-02-13T22:40:10 < gxti> except fix the linker script like i said 2013-02-13T22:40:25 < Laurenceb_> ill need a pointer and a size for the CCM stuff 2013-02-13T22:41:01 < gxti> declare two externs and make the linker script define them before and after the load addr 2013-02-13T22:41:01 < emeb> Laurenceb_: those are global symbols that come from the linker script. 2013-02-13T22:41:08 < Laurenceb_> yeah 2013-02-13T22:41:10 < Laurenceb_> _siccmram = LOADADDR(.ccmram); 2013-02-13T22:41:13 < Laurenceb_> wtf is that? 2013-02-13T22:41:29 < gxti> that defines _siccmram to the start of where .ccmram is loaded in flash 2013-02-13T22:41:41 < Laurenceb_> ah, in flash i see 2013-02-13T22:41:59 < gxti> when you do > ram AT > flash, .LOADADDR gives where the AT > flash went 2013-02-13T22:42:32 < gxti> i'm assuming there is a way to sanely define the end as well 2013-02-13T22:42:34 < Laurenceb_> yeah 2013-02-13T22:42:44 < Laurenceb_> *(.ccmram*) < whats that for? 2013-02-13T22:42:55 < gxti> linker script 101 Laurenceb_ 2013-02-13T22:43:03 < Laurenceb_> as opposed to *(.ccmram) 2013-02-13T22:43:15 < gxti> what do you think it might do? 2013-02-13T22:43:26 < Laurenceb_> some sort of pointer stuff? 2013-02-13T22:43:59 < Laurenceb_> i just have *(.ccmram) 2013-02-13T22:44:10 < gxti> it's a pattern match, silly goose 2013-02-13T22:44:12 < Laurenceb_> and the code is placed at the correct address in the elf 2013-02-13T22:44:17 < Laurenceb_> oh 2013-02-13T22:44:23 < Laurenceb_> i see now 2013-02-13T22:44:40 < gxti> mapping symbols to sections 2013-02-13T23:00:16 < Laurenceb_> silly goose?! what is it the middle ages? 2013-02-13T23:01:10 < gxti> i'm sorry 2013-02-13T23:01:14 < gxti> it's a pattern match, you fucking moron 2013-02-13T23:01:18 < gxti> happy? 2013-02-13T23:01:31 < Laurenceb_> :P 2013-02-13T23:01:56 < Laurenceb_> https://github.com/Laurenceb/Chibi-Spectro/blob/master/late_init.c 2013-02-13T23:02:00 < Laurenceb_> looks sane? 2013-02-13T23:02:21 < Laurenceb_> apart from wrong colon 2013-02-13T23:02:27 < gxti> eugh 2013-02-13T23:02:32 -!- alexn [~alexn@178-27-128-46-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-13T23:02:57 < gxti> it might work yes but dear god that is some ugly code 2013-02-13T23:03:20 < Erlkoenig> use memcpy 2013-02-13T23:03:52 < gxti> and just declare those as pointers (AT THE TOP OF THE FUNCTION) 2013-02-13T23:04:10 < gxti> s/and/or/ 2013-02-13T23:04:27 < gxti> since you wouldn't need them for memcpy. but personally i probably wouldn't but i don't have a particularly good reason why. 2013-02-13T23:05:19 < gxti> uint32_t *addr, *addr_flash; for (addr = _sccmram, addr_flash=_siccmram; addr < _eccmram; addr++, addr_flash++) { *addr = *addr_flash; } or something to that effect 2013-02-13T23:05:57 < gxti> note < not <= 2013-02-13T23:06:07 < Laurenceb_> ah, i wondered about that 2013-02-13T23:06:13 < gxti> and ++ since pointers increment by dereferenced size 2013-02-13T23:06:28 < Laurenceb_> yeah ok 2013-02-13T23:07:14 < Erlkoenig> why so complicated? for(uint32_t i = 0; i < (_eccmram-_sccmram); i+=4) { *(_sccmram+i) = *(addr_flash+i); } is more readably and thanks to offset addressing not slower 2013-02-13T23:07:15 < gxti> or use a while loop like crt0.c does 2013-02-13T23:07:56 < gxti> come to think of it why did you not just copy what crt0.c does? 2013-02-13T23:08:25 * gxti goes home 2013-02-13T23:10:18 < Laurenceb_> dp = &_data; 2013-02-13T23:10:32 < Laurenceb_> wut? i thought itd be a start address? 2013-02-13T23:13:42 < Laurenceb_> ok redone the crt0.c way 2013-02-13T23:15:03 < Laurenceb_> so does the ldscript code actually create a variable at the _sccmram/whatever location? 2013-02-13T23:20:52 -!- jef79m is now known as jef79m_lurking 2013-02-13T23:22:50 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-13T23:23:57 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-13T23:24:23 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-13T23:25:07 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Client Quit] 2013-02-13T23:25:24 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-13T23:26:22 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 256 seconds] 2013-02-13T23:28:58 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-13T23:31:48 -!- dirty_d [~adowning@209-213-71-70.meganet.net] has quit [Quit: Leaving] 2013-02-13T23:33:06 -!- phantoneD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Ping timeout: 264 seconds] 2013-02-13T23:37:08 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has quit [Remote host closed the connection] 2013-02-13T23:38:01 -!- neuro_sys is now known as neuro_sys|off 2013-02-13T23:39:11 -!- zlog [~zlog@ip68-102-205-27.ks.ok.cox.net] has joined ##stm32 2013-02-13T23:46:59 < gxti> Laurenceb_: that's why they are extern, they are defined by ld 2013-02-13T23:48:24 < gxti> hence PROVIDE(_data = .) at the start of the data section and PROVIDE(_edata = .) at the end 2013-02-13T23:49:45 < gxti> or i guess you meant you expected it to be a pointer already, to which the answer is no. a pointer is a variable that holds an address, these are variables whose location in memory is the value of interest. it's a subtle distinction. 2013-02-13T23:50:21 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Remote host closed the connection] 2013-02-13T23:50:31 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-13T23:50:31 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-13T23:50:31 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-13T23:50:34 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-13T23:50:38 < gxti> the type is irrelevant, it could be a huge struct but obviously that would be silly. you just have to make a symbol that both C and ld can see, and ld defines it to be in a particular place, and C takes the pointer of it 2013-02-13T23:58:10 <+Steffanx> meh, internet... 2013-02-13T23:59:02 < Laurenceb_> ok 2013-02-13T23:59:03 < emeb> Steffanx: internets getting you down? 2013-02-13T23:59:09 <+Steffanx> Yeah 2013-02-13T23:59:28 < Laurenceb_> emeb: in your sdr code, i dont see where you put the downconvertion in CCM --- Day changed Thu Feb 14 2013 2013-02-14T00:00:09 < Laurenceb_> https://github.com/emeb/f303_sdr/blob/master/amrx.c#L89 2013-02-14T00:00:13 < Laurenceb_> i see the c code is 2013-02-14T00:00:15 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: ….] 2013-02-14T00:01:14 < emeb> https://github.com/emeb/f303_sdr/blob/master/ddc.s#L17 2013-02-14T00:01:34 < emeb> Laurenceb ^^ 2013-02-14T00:03:05 < emeb> Laurenceb_: https://github.com/emeb/f303_sdr/blob/master/ddc.s#L17 2013-02-14T00:04:15 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32 2013-02-14T00:04:49 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 244 seconds] 2013-02-14T00:11:17 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has joined ##stm32 2013-02-14T00:11:17 -!- Steffanx [~Steffanx@524834A0.cm-4-1a.dynamic.ziggo.nl] has quit [Changing host] 2013-02-14T00:11:18 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-14T00:11:18 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-14T00:16:40 -!- neuro_sys|off is now known as neuro_sys 2013-02-14T00:20:55 -!- gavin__ [814ff42e@gateway/web/freenode/ip.129.79.244.46] has joined ##stm32 2013-02-14T00:33:12 -!- neuro_sys is now known as neuro_sys|off 2013-02-14T00:36:33 -!- ShiftPlusOne [~Shift@unaffiliated/shiftplusone] has joined ##stm32 2013-02-14T00:43:49 < Laurenceb_> ah i see 2013-02-14T00:53:13 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [] 2013-02-14T01:06:48 < Robint91> emeb, that is non floating point? 2013-02-14T01:07:01 < emeb> Robint91: what? 2013-02-14T01:07:10 < Robint91> emeb, your SDR code? 2013-02-14T01:07:32 < emeb> Robint91: most of it. The IIR filter is floating point. 2013-02-14T01:08:22 < emeb> Robint91: https://github.com/emeb/f303_sdr/blob/master/amrx.c#L158 2013-02-14T01:09:11 < Robint91> emeb, interesting project, I wanted to make a SDR with the F4 2013-02-14T01:09:27 < emeb> Robint91: Might work. 2013-02-14T01:10:00 < emeb> The big problem with the F303 is that the ADC is fast enough to sample RF, but the M4 CPU can process it in real time. 2013-02-14T01:10:31 < Robint91> emeb, use two minicircuit mixers making I and Q, an IF bandwitdh around 24khz 2013-02-14T01:10:34 < emeb> s/can/can't/ 2013-02-14T01:10:40 < Robint91> emeb, direct sampling? 2013-02-14T01:11:23 < emeb> Robint91: Right - the idea was to try and do as much in the F303 as possible, and a close analysis shows that RF sampling isn't going to work. 2013-02-14T01:11:38 < gxti> i really gotta find a good resource on learning how RF processing works. 2013-02-14T01:11:42 < Robint91> emeb, I think that also 2013-02-14T01:11:58 < emeb> So, any SDR in an F303 would require an external front-end & downconversion step as you mentioned. 2013-02-14T01:12:35 < Robint91> emeb, yeah, even if it work to direct sample, you still need to demodulate 2013-02-14T01:13:11 < emeb> Robint91: I believe that if the sample rate is low enough (sub 100kHz) that the F303 would be able to handle demodulation for the more common audio formats. 2013-02-14T01:13:41 < Robint91> emeb, yeah, doing AM is easy if you got I and Q 2013-02-14T01:13:45 < emeb> The problem is that first tune/downsample step for getting from RF rates down to a manageable rate. 2013-02-14T01:13:56 < Robint91> ssb is getting the hilbert filter at work 2013-02-14T01:13:59 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds] 2013-02-14T01:14:07 < Robint91> FM is also easy 2013-02-14T01:14:19 < Robint91> normally SSB is the most difficult of them all 2013-02-14T01:14:26 < Robint91> requires the most CPU power 2013-02-14T01:14:43 < emeb> If you've got an analytic signal all of that stuff just falls out with simple math. 2013-02-14T01:14:47 < Robint91> but bonus points for who does something like OFDM 2013-02-14T01:14:52 < Robint91> emeb, yeah 2013-02-14T01:15:41 < ShiftPlusOne> Hello. New to the stm32 boards, but was wondering if all of the libraries and examples provided by st work fine with the gcc toolchains like summon-arm? 2013-02-14T01:16:12 < emeb> ShiftPlusOne: "works for me" 2013-02-14T01:16:45 < gxti> ShiftPlusOne: i recommend https://launchpad.net/gcc-arm-embedded intead of summon, i had major problems with openocd using summon 2013-02-14T01:16:46 <+Steffanx> emeb's new favourite sentence 2013-02-14T01:17:02 < gxti> ShiftPlusOne: but as far as building, both work obviously 2013-02-14T01:17:15 < ShiftPlusOne> excellent, thank you. 2013-02-14T01:17:16 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has joined ##stm32 2013-02-14T01:17:43 < emeb> Steffanx: :) 2013-02-14T01:18:21 < dirty_d> zlog 2013-02-14T01:18:22 < zlog> dirty_d: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2013-02-13.html 2013-02-14T01:18:33 < ShiftPlusOne> Just wanted to make sure there is no closed proprietary crap to worry about. 2013-02-14T01:21:28 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-14T01:22:11 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Quit: Ik ga weg] 2013-02-14T01:24:13 < gavin__> Hi. I had a question about esden's Summon Arm Toolchain and it's fpu support. More specifically, I have some code doing floating math which is generating some inefficient assembly. 2013-02-14T01:25:26 < esden> did you try to use gcc-arm-embedded? 2013-02-14T01:26:09 < gavin__> No, I have not. Does it have better fpu support? 2013-02-14T01:27:14 < esden> it is a better polished embedded arm toolchain overall so I would hope so. :) 2013-02-14T01:27:44 < esden> beside some of the drawbacks that people mention, like not having some libraries built with it and such but that is not what you are looking for at the moment 2013-02-14T01:28:00 < ShiftPlusOne> Any point building it from source or are the binaries fine? 2013-02-14T01:28:04 < gavin__> What sort of libraries? 2013-02-14T01:28:31 < esden> libopencm3 for example 2013-02-14T01:28:39 < esden> or having python support in gdb 2013-02-14T01:29:13 < esden> the only point building it from source is if you need something the binaries don't have built in or you need/want to tweaks some parameters 2013-02-14T01:29:24 < ShiftPlusOne> fair enough, thanks 2013-02-14T01:29:43 -!- DaKu is now known as daku 2013-02-14T01:30:10 < gavin__> I would be inclined to stick with summon-arm if I could. It's what people at the school I study at have been using along with codesourcery. 2013-02-14T01:30:11 < esden> but from my testing, the gcc-arm-embedded is a very good solution, the only thing I personally miss in it is python support in the gdb provided in the linux binary 2013-02-14T01:30:31 < ShiftPlusOne> Righto then... time to blink some LEDs >.> 2013-02-14T01:31:01 < gavin__> That would be quite nice to keep as well. Strangly however I have had some trouble with debugging at all on my current situation. Breakpoints haven't been behaiving correctly. 2013-02-14T01:32:31 < esden> what tool do you use for debugging? 2013-02-14T01:33:12 < gavin__> st-util (stlink) to connect to the discovery board I'm using, then summon arm's gdb connecting to the stlink server 2013-02-14T01:33:44 < dirty_d> what hte shit 2013-02-14T01:34:01 < gavin__> hmm? 2013-02-14T01:35:41 < gxti> gavin__: try the ARM one i just linked to instead of summon 2013-02-14T01:35:57 < gxti> i never got stlink to work at all, but when i was using openocd i was having breakpoint issues with summon 2013-02-14T01:37:28 < gavin__> Interesting. I believe I've had it working on a different system with the f4's, but only having trouble currently. Alright, I'll probably build that then. Thank you. 2013-02-14T01:38:03 < dirty_d> external interrupt insanity 2013-02-14T01:38:24 < dirty_d> serentiy now! serenity now! 2013-02-14T01:43:42 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-14T01:46:13 < dirty_d> this is madness 2013-02-14T01:46:28 < dirty_d> i enabled EXT1 on PE1 2013-02-14T01:46:35 < dirty_d> but its getting triggered on PA1 instead 2013-02-14T01:49:49 < dirty_d> am i crazy or is this wrong? SYSCFG->EXTICR[0] = 0x40; 2013-02-14T01:55:11 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-14T01:55:31 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-14T01:55:38 < gxti> seems superficially correct, the chibios macro says m1 << 4 and GPIOE is 4 2013-02-14T01:56:27 < dirty_d> YEA 2013-02-14T01:56:34 < dirty_d> but PA1 is triggereint the interrupt 2013-02-14T01:56:41 < dirty_d> and PE1 doesnt 2013-02-14T01:56:50 < dirty_d> i hope the discovery board isnt labled wrong 2013-02-14T01:58:25 < gxti> and your clocks are kosher? 2013-02-14T01:58:53 < dirty_d> should be, im using f3disco 2013-02-14T01:59:01 < dirty_d> left the default clock settings 2013-02-14T01:59:14 < gxti> i mean clock gating 2013-02-14T01:59:56 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-14T02:00:14 < dirty_d> what? 2013-02-14T02:00:30 < gxti> that thing you have to do before a peripheral works 2013-02-14T02:00:41 < dirty_d> oh, yup 2013-02-14T02:00:59 < gxti> ok, then try to make other banks or pins work and see if you discover anything exciting 2013-02-14T02:03:23 < dirty_d> that sounds so exciting, lol 2013-02-14T02:03:41 < gxti> hey you're the one with the broken code, it's not my responsibility to motivate you 2013-02-14T02:03:45 < gxti> i can't even motivate myself! 2013-02-14T02:04:14 < dirty_d> lol me neither 2013-02-14T02:09:41 -!- timemob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has joined ##stm32 2013-02-14T02:09:50 < timemob> Suuup 2013-02-14T02:10:30 < Thorn> http://www.youtube.com/watch?v=KlgbKIswpzI#! 2013-02-14T02:13:58 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Remote host closed the connection] 2013-02-14T02:15:07 < gxti> well that escalated quickly 2013-02-14T02:16:41 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-14T02:18:04 < timemob> Baird music video? 2013-02-14T02:18:40 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-62.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-14T02:21:06 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-62.pptp.stw-bonn.de] has joined ##stm32 2013-02-14T02:29:24 -!- timemob [~dongs@g1-27-253-251-109.bmobile.ne.jp] has quit [Quit: Bye] 2013-02-14T02:31:06 -!- Laurenceb_ [~Laurence@host86-177-210-6.range86-177.btcentralplus.com] has quit [Ping timeout: 256 seconds] 2013-02-14T02:51:14 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-14T02:59:25 < dongs> sup 2013-02-14T02:59:53 < ds2> Hmmmmm 2013-02-14T03:00:16 < ds2> sub 100KHz RF... so does this mean the STM32 discovery can work as a digital RF for an audio receiver? 2013-02-14T03:01:34 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32 2013-02-14T03:02:25 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-14T03:04:50 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 260 seconds] 2013-02-14T03:14:57 < dirty_d> gxti, remember way back when i said i enabled my clock 2013-02-14T03:14:59 < dirty_d> yea i didnt 2013-02-14T03:15:02 < dirty_d> for SYSCFG 2013-02-14T03:15:19 < gxti> nailed it 2013-02-14T03:15:45 < gxti> also somehow i motivated myself to write some documentation for the hardware i sold 2013-02-14T03:15:50 -!- baird [~cjb@ppp121-44-193-6.lns20.syd7.internode.on.net] has joined ##stm32 2013-02-14T03:16:00 < gxti> so there's that 2013-02-14T03:16:13 < gxti> not done yet though, plenty of time to fail 2013-02-14T03:16:44 < dirty_d> what did you sell? 2013-02-14T03:17:11 < gxti> http://partiallystapled.com/2013/01/laureline-gps-ntp-server/ 2013-02-14T03:17:27 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer] 2013-02-14T03:18:48 < dirty_d> are you making money off this stuff? 2013-02-14T03:19:13 < gxti> only if you don't count how much time i spend working on it 2013-02-14T03:19:21 < gxti> but i do it because i like it :p 2013-02-14T03:19:42 < dirty_d> yea 2013-02-14T03:19:55 < dirty_d> seems hard for an individual to make money with this type of thing 2013-02-14T03:20:03 < dirty_d> unless you make something that doesnt exist 2013-02-14T03:20:23 < gxti> this exists but not at a price point where an individual would consider 2013-02-14T03:21:00 < Erlkoenig> your site looks as lazy-made as mine :D 2013-02-14T03:21:17 < gxti> hey, i like it 2013-02-14T03:21:20 < dongs> what site 2013-02-14T03:21:21 < gxti> not a lot, but i like it 2013-02-14T03:21:23 < dongs> isnt it just some shitty blog engine 2013-02-14T03:21:28 < dongs> even tarduino.cc looks better 2013-02-14T03:21:37 < Erlkoenig> dongs, stfu 2013-02-14T03:22:13 < gxti> right now i don't feel like wasting my time dicking with web stuff 2013-02-14T03:22:46 < Erlkoenig> yeah dito 2013-02-14T03:23:12 < Erlkoenig> http://games.2g2s.de/?page_id=196 <-- also Wordpress \o/ 2013-02-14T03:23:17 < emeb> better than my ancient collection of hand-edited static html 2013-02-14T03:23:50 < gxti> i actually did look at themes but the default one ended up being the least shitty one that i saw 2013-02-14T03:30:04 -!- jef79m_lurking is now known as jef79m 2013-02-14T03:30:14 < gavin__> /leave 2013-02-14T03:30:22 < gxti> quitter 2013-02-14T03:30:30 < gxti> :p 2013-02-14T03:30:34 < gavin__> Lol, so bad at IRC 2013-02-14T03:30:42 < gavin__> Ya, But I'll be back 2013-02-14T03:30:54 -!- gavin__ [814ff42e@gateway/web/freenode/ip.129.79.244.46] has quit [] 2013-02-14T03:33:00 < dirty_d> why the hell isnt this gyro generating interrupts 2013-02-14T03:33:40 < dongs> in soviet russia, gyro interrupts you 2013-02-14T03:35:25 < dirty_d> thats the ideal situation 2013-02-14T03:35:50 < dirty_d> in capitalist america gyro doesnt work 2013-02-14T03:36:15 < ds2> rotations are overrated 2013-02-14T03:36:22 < ds2> dirty_d: which gyro, btw? 2013-02-14T03:37:13 < dirty_d> the one on the f3disco 2013-02-14T03:37:22 < ds2> oh that cheapass one 2013-02-14T03:37:28 < dirty_d> L3GD20 2013-02-14T03:37:36 < ds2> isn't the interrupt pin on that thing programmable? 2013-02-14T03:37:40 < dirty_d> yea 2013-02-14T03:37:46 < ds2> checked to see if it is configured right? 2013-02-14T03:37:54 < ds2> i.e PP vs OD 2013-02-14T03:37:57 < dirty_d> yea i checked like 7 times 2013-02-14T03:37:59 < dongs> not that 2013-02-14T03:38:04 < dirty_d> its PP by default 2013-02-14T03:38:07 < dirty_d> well no 2013-02-14T03:38:08 < dirty_d> thats INT1 2013-02-14T03:38:09 < dongs> isnt there a register in the gyro 2013-02-14T03:38:11 < dongs> for enabling teh shit 2013-02-14T03:38:14 < dirty_d> the data ready interrupt is INT2 2013-02-14T03:38:17 < dirty_d> yea 2013-02-14T03:38:19 < dirty_d> i enabled it 2013-02-14T03:38:21 < dongs> you cant remap one to another? 2013-02-14T03:38:35 < dirty_d> i dont think so 2013-02-14T03:38:41 < dirty_d> INT1 is for like advanced stuff 2013-02-14T03:39:07 < dirty_d> i only see 1 bit for INT2 in 1 reg 2013-02-14T03:39:09 < dirty_d> nowhere else 2013-02-14T03:39:23 < dongs> and whats it connected to on stm end 2013-02-14T03:39:25 < dongs> exti trigger? 2013-02-14T03:40:50 < dirty_d> yea 2013-02-14T03:40:51 < dirty_d> it works 2013-02-14T03:40:54 < dirty_d> i tested it with a wire 2013-02-14T03:40:59 < dirty_d> its to PE1 2013-02-14T03:43:15 < ds2> is the L3GD20 setup to generate interrupts regardless? i.e. it won't block waiting for the previous one to be acked before generating more 2013-02-14T03:45:11 < dirty_d> im not sure, i dont see anything in the datasheet that says anything about it 2013-02-14T03:45:24 < dirty_d> it doesnt really explain DRDY/INT2 at all 2013-02-14T03:47:37 < ds2> do you really need that interrupt? 2013-02-14T03:48:54 < dirty_d> kinda 2013-02-14T03:49:18 < dirty_d> i wont be able to rest until i know why its not working though 2013-02-14T03:49:50 < dirty_d> it might only work in FIFo mode 2013-02-14T03:49:56 < dirty_d> but it doesnt say that 2013-02-14T03:50:15 < dongs> Earlier this week you may have received an obsolete product notification email from Digi-Key Corporation regarding part LPC4357FET256,551 from NXP Semiconductors. This message was sent to you in error as this part does not have an obsolete status. We apologize for this mistake and any confusion that it may have caused. 2013-02-14T03:50:21 < dongs> haha 2013-02-14T03:52:08 < dongs> moc d:/TauLabs/ground/gcs/src/libs/qtconcurrent/multitask.h 2013-02-14T03:52:09 < dongs> compiling debug/moc_multitask.cpp 2013-02-14T03:52:09 < dongs> /usr/bin/sh: /c/mingw/bin/g++: Invalid argument 2013-02-14T03:52:10 < dongs> mingw32-make[5]: *** [debug/moc_multitask.o] Error 126 2013-02-14T03:52:12 < dongs> lol opensauce 2013-02-14T03:53:48 < gxti> nothing with 'ming' in the name can be good 2013-02-14T03:55:11 < dongs> or gnu 2013-02-14T04:07:09 < emeb> ming the merciless! 2013-02-14T04:08:16 < dongs> check out this hilarious opensauce failure 2013-02-14T04:08:22 < dongs> just d oing up-enter results in different errors! http://bcas.tv/paste/results/dNo0NY27.html 2013-02-14T04:08:49 < dongs> shift-del'd that shit, what a fucking joke 2013-02-14T04:09:03 -!- emeb [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: Leaving.] 2013-02-14T04:10:03 < Erlkoenig> http://arduino.cc/forum/index.php?topic=138643.0 RTOS for Arduino, integrating with Arduino libs 2013-02-14T04:10:38 < dongs> I think they've got an extra 'T' in the name 2013-02-14T04:11:37 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 256 seconds] 2013-02-14T04:14:28 < baird> You would think that someone that can measure battery cell internal resistance wouldn't start at the $800 mark.. :/ 2013-02-14T04:15:36 < Erlkoenig> is that so difficult to measure? ... just connect the battery to two different loads and calculate a bit? 2013-02-14T04:16:53 < baird> A $60 device, tops. But noooo... 2013-02-14T04:18:09 < Erlkoenig> there's even a version of the Arduino RTOS for ARM arduinos - and ports of ChibiOS to Arduino 2013-02-14T04:24:49 < gxti> spread the suck 2013-02-14T04:26:22 -!- BJFreeman [~bjfree@190.sub-75-196-115.myvzw.com] has quit [Quit: had a good time] 2013-02-14T04:27:19 < dongs> retweet https://youtube.com/watch?v=FKdu01LqXdI 2013-02-14T04:29:40 -!- Gargantuasauce [~Gargantua@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-14T04:34:10 < Erlkoenig> STM32 microcontrollers are now moving into the Java world. Microcontroller applications have been traditionally developed in C with some C++. Now STM32-Java allows embedded developers to take advantage of Java Object-Oriented Programming (OOP), which is inherently better suited to GUI development than traditional structured programming languages. 2013-02-14T04:34:11 < Erlkoenig> The object-oriented nature also simplifies support and maintenance of Java applications. 2013-02-14T04:34:23 < Simon--> ugh 2013-02-14T04:34:55 < dongs> old 2013-02-14T04:35:04 < dongs> thats why they're making one with SDRAM 2013-02-14T04:35:09 < dongs> so it has enouhg memory to run javashit 2013-02-14T04:37:15 < Erlkoenig> $2600 ... nice 2013-02-14T04:40:40 < dirty_d> ds2, fixed it 2013-02-14T04:41:13 < dirty_d> i wasnt reading from the data register after writing to the gyro regs 2013-02-14T04:41:26 < dirty_d> so it was overflowing i geuss 2013-02-14T04:41:34 < dongs> right 2013-02-14T04:41:39 < dongs> makes sense 2013-02-14T04:41:51 < dongs> it probably wouldnt trigger again until you read data and cleared it 2013-02-14T04:42:23 < dirty_d> yea something like that 2013-02-14T04:42:35 < dirty_d> i dont think it was even sending anything after the first write 2013-02-14T04:42:53 < dirty_d> i dunno 2013-02-14T04:54:44 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-14T04:55:03 < dongs> attn emeb http://wiki.linuxaudio.org/apps/screenshots 2013-02-14T04:55:14 < R2COM> so what was relatively huge current consuption someone could see for F4 chip 2013-02-14T04:55:30 < dongs> R2COM: datasheet said something like 150mA max w/all shit on 2013-02-14T04:55:36 < R2COM> lets say, running some loop, at full speed cpu 168 2013-02-14T04:55:39 < R2COM> hm 2013-02-14T04:55:51 < dongs> i think emeb saw something similar when running DSP stuff 2013-02-14T04:56:03 < R2COM> thats not that bad 2013-02-14T04:56:37 -!- Viper168 [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-14T05:02:58 < ds2> ah... my theory was close... have to remember that 2013-02-14T05:10:11 < R2COM> discovery f4 board with some while loop at 168MHz, some PWM and stuff measured now around 100mA max (97-98mA) 2013-02-14T05:10:26 < R2COM> with all ports enabled 2013-02-14T05:10:32 < dongs> sounds about right 2013-02-14T05:10:34 < R2COM> and 4 running pwm channels 2013-02-14T05:14:42 < dongs> while loop doing nothing is probably pretty useles for load measurement 2013-02-14T05:14:50 < R2COM> yeah I'm just saying 2013-02-14T05:14:54 < Erlkoenig> the st-link on the F4 Discovery consumes ~ 50mA just during idle... by forcing it to reset one can avoid that 2013-02-14T05:15:33 < gxti> Erlkoenig: there's a jumper for current measurement 2013-02-14T05:15:54 < Erlkoenig> yup, but it doesn't stop the st-link from draining batteries :) 2013-02-14T05:15:54 < gxti> i can't recall but i presume only the f4 is on it 2013-02-14T05:16:03 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has joined ##stm32 2013-02-14T05:16:05 < gxti> what's a battery? 2013-02-14T05:16:16 < Erlkoenig> a limited power source :D 2013-02-14T05:16:25 < gxti> sounds useless 2013-02-14T05:16:31 < Erlkoenig> and wireless 2013-02-14T05:45:26 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 245 seconds] 2013-02-14T06:12:17 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has quit [Read error: Connection reset by peer] 2013-02-14T06:16:30 < emeb_mac> dongs: what about those linux screenshots? 2013-02-14T06:17:29 < dongs> emeb_mac: they're fucking funny 2013-02-14T06:17:31 < dongs> shit apps on shit OS 2013-02-14T06:17:35 < dongs> doing useless shit. 2013-02-14T06:17:43 < emeb_mac> lol 2013-02-14T06:17:45 < dongs> all different worthless GUI toolkits 2013-02-14T06:18:05 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has joined ##stm32 2013-02-14T06:18:08 < emeb_mac> dongs: diversity! 2013-02-14T06:18:20 < emeb_mac> a riot of competing ecosystems! 2013-02-14T06:18:30 < emeb_mac> something for everyone! 2013-02-14T06:18:36 < emeb_mac> blah blah blah 2013-02-14T06:19:19 < dongs> incorrect. 2013-02-14T06:19:21 < dongs> just aids. 2013-02-14T06:19:41 < dongs> i wonder why is blender e ven in the list 2013-02-14T06:21:08 < emeb_mac> whadaya mean? 2013-02-14T06:21:12 < dongs> well 2013-02-14T06:21:15 < dongs> its alist of lunix audio apps 2013-02-14T06:21:18 < dongs> i thought blender was 3dshite 2013-02-14T06:21:24 < emeb_mac> right 2013-02-14T06:21:56 -!- jef79m is now known as jef79m_lurking 2013-02-14T06:22:10 < ggh> Hi, I'm new to micro controllers and got an f4 discovery board to learn on. I wrote a device driver for an optical rpm sensor, but when I try to load nuttx onto the board with stlink, i get this error: (gdb) load 2013-02-14T06:22:11 < ggh> Loading section .text, size 0xef9c lma 0x8000000 2013-02-14T06:22:16 < ggh> any ideas? 2013-02-14T06:22:46 -!- phantoxeD [~destroy@a89-155-23-52.cpe.netcabo.pt] has joined ##stm32 2013-02-14T06:24:42 < emeb_mac> not a clue - don't use nuttx 2013-02-14T06:24:50 < dongs> ask zippe 2013-02-14T06:24:54 < dongs> hes the nuttx product managwer 2013-02-14T06:25:27 < ggh> ok, will do - thanks guys 2013-02-14T06:27:58 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 252 seconds] 2013-02-14T06:28:27 < R2COM> real time operating system.. heh 2013-02-14T06:29:45 < emeb_mac> I prefer fake time operating systems 2013-02-14T06:29:59 < emeb_mac> or even imaginary time operating systems 2013-02-14T06:31:43 < R2COM> I prefer to get away with as shitload of software written by someone as possible and get done just what has to get done 2013-02-14T06:32:00 < R2COM> as less* 2013-02-14T06:32:09 < R2COM> or actually, without. 2013-02-14T06:35:13 < baird> Delivering Results is all that matters when you're making ManToys... 2013-02-14T06:35:46 < emeb_mac> Grunt! Man Toys! 2013-02-14T06:36:08 < R2COM> yeah, and ManToys are not supposed to contain boycode 2013-02-14T06:38:39 < Erlkoenig> ggh: this is not an error, this is the usual message of GDB when flashing... 2013-02-14T06:39:36 < Erlkoenig> why is it possible to turn of Instruction Prefetch, Instruction Cache and Data management for the Flash access of the F4 core? To save power? 2013-02-14T06:41:19 < ggh> Erlkoenig: i guess i forgot to paste the next line: Load failed 2013-02-14T06:41:40 < ggh> nothing happens in stlink when I try to load it 2013-02-14T06:41:53 < Erlkoenig> please poste you whole gdb session 2013-02-14T06:41:59 < Erlkoenig> *paste 2013-02-14T06:44:26 < ggh> http://pastebin.com/j7Q72t6V 2013-02-14T06:45:12 < ggh> i have in my .gdbinit 2013-02-14T06:45:12 < ggh> target remote localhost:4242 2013-02-14T06:45:46 < Erlkoenig> 2013-02-13T20:42:11 INFO src/stlink-common.c: Loading device parameters....2013-02-13T20:42:11 WARN src/stlink-common.c: unknown chip id! 0 2013-02-14T06:46:11 < ggh> hmm, i built from the master on github 2013-02-14T06:46:23 < Erlkoenig> disconnect the board from the PC, hold the reset button on the board, connect the board to the PC, start st-link, release reset button 2013-02-14T06:46:33 < Erlkoenig> then continue with gdb 2013-02-14T06:46:57 < dirty_d> about damn time i got this gyro working 2013-02-14T06:46:59 < dirty_d> now i can sleep 2013-02-14T06:47:00 < dirty_d> lol 2013-02-14T06:49:30 < ggh> Erlkoenig: sweet! it flashed. thanks for your help 2013-02-14T06:49:57 < Erlkoenig> hehe... st-link is a bit buggy it seems... 2013-02-14T06:50:53 < dirty_d> Erlkoenig, is it? 2013-02-14T06:50:57 < dirty_d> what happened? 2013-02-14T06:51:12 < Erlkoenig> if the core is not in reset state and you start st-link, you get that error 2013-02-14T06:51:33 < Erlkoenig> not always, of course 2013-02-14T06:52:19 < Erlkoenig> also sometimes you just can't start the program from within gdb, it just hangs at the very first instruction 2013-02-14T06:52:45 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services] 2013-02-14T06:52:50 < Erlkoenig> re-running a program does not work either 2013-02-14T06:52:54 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32 2013-02-14T06:53:38 < dirty_d> hmm, i dont think ive seen that yet 2013-02-14T06:53:45 < dirty_d> im out, later 2013-02-14T06:53:51 -!- dirty_d [~andrew@c-76-118-112-158.hsd1.ma.comcast.net] has quit [Quit: Leaving] 2013-02-14T06:56:07 < ds2> always start from reset.. that seems to solve most STlink problems for me 2013-02-14T06:56:16 < ShiftPlusOne> Are there any guides out there on how to actually flash new firmware using stlink v2? When I run st-util, I get 2013-02-14T15:55:52 WARN src/stlink-usb.c: Couldn't find any ST-Link/V2 devices 2013-02-14T06:56:46 < gxti> try running it as root? 2013-02-14T06:57:14 < ShiftPlusOne> ah >.> I copied the udev rules and reloaded it, but I guess I missed something 2013-02-14T06:57:14 < Erlkoenig> the guide is "run st-link" ;) 2013-02-14T06:57:15 < ShiftPlusOne> thanks 2013-02-14T06:58:22 < Erlkoenig> ds2: yeah... the ST-Link should automatically reset the µC or something... but this probably wouldn't work because hardware bugs prevent complete&clean reset 2013-02-14T06:59:10 < ds2> I just tap the reset button 2013-02-14T07:04:46 < R2COM> linux debugging... 2013-02-14T07:05:17 < R2COM> I'v done that a little bit before, then set up my toolchain under win7, with Eclipse popping out all stuff and showing things nicely 2013-02-14T07:05:38 < gxti> that's good for you 2013-02-14T07:05:49 < R2COM> sure it is 2013-02-14T07:07:21 < Erlkoenig> what gdb server are you using? 2013-02-14T07:09:34 < ds2> st-util works fine for me 2013-02-14T07:19:07 -!- NonaSuomy_ [~NonaSuomy@69.158.165.12] has joined ##stm32 2013-02-14T07:22:16 -!- NonaSuomy [~NonaSuomy@xbmc/staff/TeknoJuce] has quit [Ping timeout: 240 seconds] 2013-02-14T07:26:01 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-62.pptp.stw-bonn.de] has quit [Quit: Leaving.] 2013-02-14T07:31:23 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has quit [Read error: Operation timed out] 2013-02-14T07:32:30 -!- Gargantuasauce [~Gargantua@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 264 seconds] 2013-02-14T07:36:09 -!- Gargantuasauce [~Gargantua@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-14T07:36:24 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32 2013-02-14T07:48:18 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer] 2013-02-14T07:48:37 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32 2013-02-14T08:18:36 -!- ReggieUK [ReggieUK@5aca56e7.bb.sky.com] has quit [] 2013-02-14T08:29:05 -!- solid_liq [~solidly@unaffiliated/solidliq] has quit [Ping timeout: 252 seconds] 2013-02-14T08:43:29 -!- emeb_mac [~ericb@ip72-201-78-226.ph.ph.cox.net] has quit [Quit: emeb_mac] 2013-02-14T08:47:56 -!- solid_liq [~solidly@173-23-250-226.client.mchsi.com] has joined ##stm32 2013-02-14T08:48:02 -!- solid_liq [~solidly@173-23-250-226.client.mchsi.com] has quit [Changing host] 2013-02-14T08:48:02 -!- solid_liq [~solidly@unaffiliated/solidliq] has joined ##stm32 2013-02-14T09:10:54 < dongs> zyp, your trash shipped 2013-02-14T09:14:34 -!- _BJFreeman [~bjfree@112.sub-75-233-119.myvzw.com] has joined ##stm32 2013-02-14T09:15:40 -!- _BJFreeman is now known as BJFreeman 2013-02-14T09:20:42 < ds2> l 2013-02-14T09:25:52 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-14T09:26:22 -!- barthess [~barthess@86.57.155.106] has joined ##stm32 2013-02-14T09:31:30 -!- dekar__ [~dekar@212.255.45.157] has joined ##stm32 2013-02-14T09:31:33 -!- mode/##stm32 [+v dekar__] by ChanServ 2013-02-14T09:34:15 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds] 2013-02-14T09:34:58 -!- dekar_ [~dekar@212.255.34.180] has quit [Ping timeout: 252 seconds] 2013-02-14T09:42:32 < zyp> dongs, great, thanks 2013-02-14T09:42:37 < zyp> how do you want payment? 2013-02-14T09:42:50 < R2COM> with gold bars 2013-02-14T09:45:15 < dongs> zyp: shitpal will work i guess 2013-02-14T09:45:18 < dongs> or buttcoin 2013-02-14T09:48:44 < zyp> paypal is fine, how much and where to? 2013-02-14T09:49:56 < dongs> you havent shitpal'd me before? 2013-02-14T09:50:07 < dongs> whatever amount you figure out ^ that address. 2013-02-14T09:50:27 < zyp> $15 would cover BOM and everything? 2013-02-14T09:50:30 < dongs> preferably in israeli shekels since thats my main currencly probably. but hten baird would want one at same price too 2013-02-14T09:51:06 < dongs> i suppose i can cut you a deal in exchange for internet hugs. 2013-02-14T09:53:01 < zyp> ok, sent 2013-02-14T09:55:14 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32 2013-02-14T09:55:48 -!- alexn [~alexn@93.104.205.194] has quit [Remote host closed the connection] 2013-02-14T09:56:03 -!- alexn [~alexn@93.104.205.194] has joined ##stm32 2013-02-14T09:59:02 -!- ggh [~ggh@173-160-154-169-Washington.hfc.comcastbusiness.net] has quit [Quit: ggh] 2013-02-14T10:05:30 < dongs> cool 2013-02-14T10:05:33 < dongs> finally got my code signing cert shit done 2013-02-14T10:05:44 < dongs> signed my logic analyzer driver 2013-02-14T10:05:48 < dongs> now need to do cubieboard 2013-02-14T10:12:52 < dongs> err,, smoothieboard 2013-02-14T10:16:13 < zyp> you put your cubieboard in a blender? 2013-02-14T10:16:20 < dongs> heh 2013-02-14T10:27:20 -!- BJFreeman [~bjfree@112.sub-75-233-119.myvzw.com] has quit [Quit: had a good time] 2013-02-14T10:30:38 -!- neuro_sys|off is now known as neuro_sys 2013-02-14T10:31:45 -!- dekar__ [~dekar@212.255.45.157] has quit [Quit: This computer has gone to sleep] 2013-02-14T10:39:57 -!- claude is now known as Claude 2013-02-14T10:45:59 -!- Laurenceb_ [~Laurence@host86-177-210-6.range86-177.btcentralplus.com] has joined ##stm32 2013-02-14T10:46:45 -!- rob_w [~bob@unaffiliated/rob-w/x-1112029] has joined ##stm32 2013-02-14T10:52:02 -!- neuro_sys is now known as neuro_sys|off 2013-02-14T10:54:07 -!- neuro_sys|off is now known as neuro_sys 2013-02-14T10:59:32 -!- Viper168_ [~Viper@unaffiliated/viper168] has joined ##stm32 2013-02-14T10:59:57 -!- Viper168 [~Viper@unaffiliated/viper168] has quit [Ping timeout: 244 seconds] 2013-02-14T11:01:53 -!- Laurenceb_ [~Laurence@host86-177-210-6.range86-177.btcentralplus.com] has quit [Ping timeout: 248 seconds] 2013-02-14T11:03:54 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has left ##stm32 [] 2013-02-14T11:08:39 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has joined ##stm32 2013-02-14T11:08:42 -!- trepidaciousMBR_ [~trepidaci@217.155.204.2] has quit [Client Quit] 2013-02-14T11:12:30 -!- trepidaciousMBR [~trepidaci@217.155.204.2] has quit [Ping timeout: 260 seconds] 2013-02-14T11:38:47 -!- zippe [~Adium@173.11.99.161] has quit [Quit: Leaving.] 2013-02-14T11:45:40 -!- trepidaciousMBR [~trepidaci@176.62.208.227] has joined ##stm32 2013-02-14T12:14:33 -!- neuro_sys is now known as neuro_sys|off 2013-02-14T12:16:36 -!- Claude is now known as claude 2013-02-14T12:16:54 -!- claude is now known as Claude 2013-02-14T12:51:46 < Laurenceb> this is odd 2013-02-14T12:51:59 < Laurenceb> i seem to have sane binaries and code in CCM 2013-02-14T12:52:10 < Laurenceb> but arm-none-eabi-size doesnt seem sane 2013-02-14T12:52:23 < Laurenceb> text data bss dec hex filename 2013-02-14T12:52:23 < Laurenceb> 73464 2664 62760 138888 21e88 build/ch.elf 2013-02-14T12:52:39 < Laurenceb> yet the data section seems to be 2268 bytes 2013-02-14T12:54:06 < Laurenceb> http://www.raspberrypi.org/ 2013-02-14T12:54:13 < Laurenceb> lady ada looks more like lady gaga 2013-02-14T12:56:23 -!- Steffanx [~Steffanx@client-212-146.flexnet2.rug.nl] has joined ##stm32 2013-02-14T12:56:23 -!- Steffanx [~Steffanx@client-212-146.flexnet2.rug.nl] has quit [Changing host] 2013-02-14T12:56:23 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32 2013-02-14T12:56:25 -!- mode/##stm32 [+v Steffanx] by ChanServ 2013-02-14T13:01:08 < Laurenceb> i.e. a retarded attention whore 2013-02-14T13:04:42 <+Steffanx> Good description of Laurenceb 2013-02-14T13:05:57 < dongs> ladyaids 2013-02-14T13:06:07 < dongs> they should get together with jerri ellisworthless 2013-02-14T13:06:11 < dongs> and do some MAKEring 2013-02-14T13:07:51 <+Steffanx> dongs why they didn't invite you to talk with mr. obama himself ? 2013-02-14T13:08:22 <+Steffanx> using a g+ hangout 2013-02-14T13:08:50 < baird> Considering how much that the LadyAda group is fucking up the mission of the RPi Foundation... 2013-02-14T13:09:10 <+Steffanx> That's something they did to themself baird 2013-02-14T13:09:25 <+Steffanx> They choose something with a overpowered gpu 2013-02-14T13:09:28 <+Steffanx> -o? 2013-02-14T13:09:47 <+Steffanx> By doing that, they asked for stuff like xmbc on the rpi 2013-02-14T13:09:49 < baird> Still, contributing to the fail. 2013-02-14T13:09:55 <+Steffanx> + all the hackers around that 2013-02-14T13:10:00 <+Steffanx> oh, yes 2013-02-14T13:10:21 <+Steffanx> fail.. i remember you had a cluster of rpis 2013-02-14T13:10:58 < baird> I was Learnings about MPI and clustering. 2013-02-14T13:11:36 < baird> No longer having a Maspar to play with, it's the next best thing. :) 2013-02-14T13:11:51 < Laurenceb> why is ladyaids fucking up Rpi any more than it already was? 2013-02-14T13:12:25 <+Steffanx> Hmm, why i even care? 2013-02-14T13:12:32 < Laurenceb> also ladyaids sounds like a rampant rabbit or something 2013-02-14T13:12:47 < baird> There was also the embedded tooling work. 2013-02-14T13:13:46 < baird> Go to the Ladyada youtube channel and post "Why isn't the suicide girl doing vids n e moar???" 2013-02-14T13:13:56 < Laurenceb> lol 2013-02-14T13:14:25 < Laurenceb> as in suicidegirls.com? 2013-02-14T13:14:33 < Laurenceb> bet Tectu knows about that site 2013-02-14T13:15:29 < baird> SG jumped the shark when they knocked-back Keiko Lune.. -_- 2013-02-14T13:16:13 < Laurenceb> see i bet even Tectu doesnt know _that_ much detail 2013-02-14T13:17:09 < baird> http://i.imgur.com/6KK1z.png 2013-02-14T13:18:21 < Laurenceb> i lolled 2013-02-14T13:18:58 <+Steffanx> suitable for school Laurenceb ? 2013-02-14T13:19:05 < Laurenceb> yup 2013-02-14T13:19:08 <+Steffanx> *uni 2013-02-14T13:32:30 -!- rigid [~daniel@178-26-71-67-dynip.superkabel.de] has quit [Remote host closed the connection] 2013-02-14T13:35:20 < zyp> Laurenceb, you can't trust size to interpret custom sections correctly, use objdump -h 2013-02-14T13:35:42 < Laurenceb> yeah thats why i said it didnt seem correct 2013-02-14T13:35:51 < Laurenceb> objdump was different 2013-02-14T13:37:24 < Laurenceb> way too many veneer functions atm 2013-02-14T13:37:30 < Laurenceb> thats going to slow it down 2013-02-14T13:38:20 < Laurenceb> annoyingly logf and other math functions arent using the fpu 2013-02-14T13:38:32 < Laurenceb> they call eabi_fmul and equivalent 2013-02-14T13:39:26 < zyp> ah, ccm is probably outside relative jump range 2013-02-14T13:39:37 < zyp> that's why the linker has to make veneers 2013-02-14T13:39:44 < Laurenceb> i see 2013-02-14T13:39:51 < Laurenceb> is there a way to fix that? 2013-02-14T13:40:08 < Laurenceb> or is it a hardware limit? 2013-02-14T13:41:13 < zyp> the thing is that the compiler makes relative branch instructions for everything, but it's the linker's job to fill in the target addr 2013-02-14T13:42:12 < zyp> so if the target is outside the range, the linker will add a veneer within range and use that to jump to an absolute addr 2013-02-14T13:42:51 < Laurenceb> so where is the range limit set? 2013-02-14T13:43:12 < zyp> it's the number of bits available in the branch instruction encoding 2013-02-14T13:43:17 < zyp> I don't remember 2013-02-14T13:43:22 < zyp> but it's not changeable 2013-02-14T13:43:33 < Laurenceb> ah 2013-02-14T13:45:53 < Laurenceb> bbl 2013-02-14T13:47:56 <+Steffanx> [12:19:04] yup <= that's a no :P 2013-02-14T13:50:46 < zyp> range is ±16M 2013-02-14T13:51:13 < karlp> who would ever need more than 16meg of program space?! ;) 2013-02-14T13:51:28 <+Steffanx> microsoft 2013-02-14T13:51:34 < zyp> but there's 128M between flash and ccm ;) 2013-02-14T13:52:20 < karlp> wasn't there an f4 tech training sheet we found that said you probably wouldn't get any speedups with ccm for a lot of uses anyway? 2013-02-14T13:52:28 < karlp> something about contention on ibus and blah blah blah 2013-02-14T13:52:43 < zyp> aren't we talking about f3 now? 2013-02-14T13:53:10 < zyp> f4 has a nice cache between flash and ibus, f3 doesn't 2013-02-14T13:55:51 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: zzzzZZzzz] 2013-02-14T13:56:59 < zyp> anyway, bl