--- Log opened Mon Oct 01 00:00:43 2012
2012-10-01T00:19:40 -!- barthess [~barthess@178.154.13.227] has quit [Quit: Leaving.]
2012-10-01T00:24:51 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-10-01T00:25:15 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-01T00:25:16 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-01T00:25:16 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-01T00:25:31 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has joined ##stm32
2012-10-01T00:26:15 -!- ReggieUK [~quassel@5aca5630.bb.sky.com] has joined ##stm32
2012-10-01T00:42:26 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-01T00:42:39 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-01T00:47:47 -!- ReggieUK [~quassel@5aca5630.bb.sky.com] has quit [Read error: Connection reset by peer]
2012-10-01T00:50:44 -!- ReggieUK [~quassel@5aca5630.bb.sky.com] has joined ##stm32
2012-10-01T00:53:02 < dongs> sup dongs
2012-10-01T00:53:02 -!- izua_ [~izua@unaffiliated/izua] has quit [Quit: Pull me under]
2012-10-01T00:53:06 < dongs> too fucking early
2012-10-01T00:54:07 -!- izua [~izua@86.121.85.113] has joined ##stm32
2012-10-01T00:54:07 -!- izua [~izua@86.121.85.113] has quit [Changing host]
2012-10-01T00:54:07 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-01T00:54:10 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-01T00:54:55 -!- izua [~izua@unaffiliated/izua] has quit [Client Quit]
2012-10-01T01:00:35 <+Steffanx> Too f-ing late dong
2012-10-01T01:00:55 < Laurenceb_> suppppppppp
2012-10-01T01:00:58 -!- ReggieUK [~quassel@5aca5630.bb.sky.com] has quit [Remote host closed the connection]
2012-10-01T01:01:15 <+Steffanx> Tha sky
2012-10-01T01:01:16 < Laurenceb_> me is contemplating gps locked frequency sources
2012-10-01T01:01:22 < BrainDamage> dongs https://www.youtube.com/watch?v=d3CWLCoQu7c
2012-10-01T01:01:37 < dongs> BrainDamage: i saw it last time you linked it
2012-10-01T01:01:40 < Laurenceb_> didnt realise stm has spare plls
2012-10-01T01:01:41 < dongs> paypal'd several
2012-10-01T01:01:45 < Laurenceb_> very useful
2012-10-01T01:02:19 < BrainDamage> as long as you can find a gps with "raw" output, I don't see why not
2012-10-01T01:02:44 < dongs> what spare plls?
2012-10-01T01:02:51 < Laurenceb_> oh my god
2012-10-01T01:02:56  * Laurenceb_ just found video
2012-10-01T01:03:25 <+Steffanx> BrainDamage, that thing is REALLLY OLD
2012-10-01T01:03:52 < BrainDamage>  I saw it for the first time today
2012-10-01T01:04:00 < BrainDamage> and so seems Laurenceb and dongs
2012-10-01T01:04:03 < BrainDamage> so news to us
2012-10-01T01:04:20 < BrainDamage> also "Published on Sep 21, 2012 by     KickAsshTv"
2012-10-01T01:04:32  * Steffanx takes a look in his archive
2012-10-01T01:04:44 < dongs>  dongs> what spare plls?
2012-10-01T01:05:28 <+Steffanx> Hmm, can't find it
2012-10-01T01:05:40 <+Steffanx> but i bet i've seen that thing over a year ago
2012-10-01T01:06:04 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-01T01:10:09 <+Steffanx> http://www.youtube.com/watch?v=9jTM3klngWg BrainDamage .. april 2011 :P
2012-10-01T01:10:58 <+Steffanx> Oh, it's even the same videp
2012-10-01T01:10:59 <+Steffanx> o
2012-10-01T01:12:11 < dongs> lemme guess
2012-10-01T01:12:17 < dongs> some "blogger" "found" it again yesterday
2012-10-01T01:12:38 < dongs> and posted as news
2012-10-01T01:13:10 <+Steffanx> Probably and many  copied it
2012-10-01T01:18:52 -!- Mazingaro [~Tetsuja@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Quit: Leaving]
2012-10-01T01:20:06 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-01T01:20:34 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-01T01:21:53 < Laurenceb_> http://www.dailymail.co.uk/news/article-2206613/Chinese-hospitals-introduce-hands-free-automatic-sperm-extractor-donors-play-videos-help.html
2012-10-01T01:22:57 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-01T01:35:42 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-01T01:41:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-01T01:50:51 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-01T02:04:52 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 248 seconds]
2012-10-01T02:04:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-01T02:08:08 -!- BjoernC [~BjoernC@dyndsl-178-142-049-197.ewe-ip-backbone.de] has quit [Remote host closed the connection]
2012-10-01T02:11:16 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-01T02:53:57 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-01T03:03:05 < dongs> dailyfail
2012-10-01T03:09:46 < Thorn> DCD_EP_Tx() hangs until host actually reads the data (EP is interrupt IN)
2012-10-01T03:12:05 < zyp> blocks would be the proper term
2012-10-01T03:12:43 < Thorn> well, it blocks, and the mcu hangs :)
2012-10-01T03:12:45 < zyp> are you sure it blocks until host reads data and not until host reads the previous block of data (freeing up the tx fifo for the new block of data)
2012-10-01T03:14:06 < Thorn> I have a host app that calls libusb_interrupt_transfer() (blocking, timeout 1 sec) when a gui button is pressed
2012-10-01T03:14:49 < Thorn> I press the (physical) button on the device. LED stops blinking. I press the "receive" button (with a mouse) and the led immediately starts blinking again
2012-10-01T03:15:03 < Thorn> and the host app reports receied data
2012-10-01T03:16:34 < Thorn> same thing happens after  libusb_interrupt_transfer() timeouts so no previous data anywhere as far as I can tell
2012-10-01T03:16:52 < zyp> ah, right
2012-10-01T03:17:45 < Thorn> this is an extremely stupid way to receive data from an interrupt EP but the device shouldn't block in this manner whatever happens, right?
2012-10-01T03:18:25 < zyp> well, it's a question of how you handle flow control
2012-10-01T03:18:25 < Thorn> btw there's a way to hang the host app, too, if it send a control transfer with data and the device doesn't read this data
2012-10-01T03:19:55 < zyp> in my stack, writes are currently non-blocking, but you have to make sure not to start another write when one is already pending
2012-10-01T03:20:47 < Thorn> I don't really see where it could block in the driver code
2012-10-01T03:21:28 < zyp> I assume the write-code writes the data and then loops until it receives confirmation that the transfer is completed
2012-10-01T03:22:14 < Thorn> loops is what I'm not seeing
2012-10-01T03:22:22 < Thorn> only register writes
2012-10-01T03:22:25 < zyp> I suggest halting in the debugger while it's hanging and doing a backtrace to see where
2012-10-01T03:24:03 < Thorn> I tried that, no particular place. seems linke it's in the usb interrupt handler even though no usb events are happening and when one does happen it's processed normally
2012-10-01T03:24:17 < Thorn> will try again
2012-10-01T03:26:36 < Thorn> if it's really looping in the handler and another interrupt is generated, it wouldn't be serviced right? an interrupt can't preempt itself
2012-10-01T03:27:08 < zyp> well, is it really in handler mode?
2012-10-01T03:27:11 < Thorn> but I'm getting traces within OTG_FS_IRQHandler
2012-10-01T03:27:28 < zyp> info registers
2012-10-01T03:27:36 < zyp> what's the lowest byte of psr?
2012-10-01T03:27:57 < Thorn> cpsr = 0x81000073
2012-10-01T03:28:07 < zyp> right, it's in interrupt handler 0x73
2012-10-01T03:28:30 < Thorn> I wonder what it's doing there
2012-10-01T03:28:43 < Thorn> looks like usb interrupts are generated constantly
2012-10-01T03:28:45 < zyp> pastebin the backtrace?
2012-10-01T03:29:10 < zyp> well, usb-interrupts are generated constantly, at least if you have the SOF interrupt enabled :p
2012-10-01T03:29:42 < Thorn> http://pastebin.com/LcPhTxSp
2012-10-01T03:30:29 < Thorn> still it's statistically likely that a ctrl+c break will end up in the idle rtos task
2012-10-01T03:30:49 < Thorn> what I'm seeing is rtos tasks stop executing completely though
2012-10-01T03:31:12 < zyp> ah
2012-10-01T03:32:13 < zyp> https://gist.github.com/2282087 <- a google search found this, sounds relevant to your issue
2012-10-01T03:32:19 < Thorn> another backtrace is in CLEAR_IN_EP_INTR(epnum, emptyintr);
2012-10-01T03:33:26 < zyp> what seems to happen, if I understand it correctly, is that you get an interrupt telling you that the tx fifo is now empty
2012-10-01T03:34:00 < zyp> and since you don't write anything more, it remains empty, so you get interrupted again and again
2012-10-01T03:34:01 < Thorn> right, I see it's enabled by that DCD_EP_Tx() thing
2012-10-01T03:34:11 < Thorn> nice
2012-10-01T03:34:30 < zyp> so the solution is to do what it says and disable the interrupt until next time you need it
2012-10-01T03:34:50 < zyp> I'm just guessing now, by the way :p
2012-10-01T03:36:09 < Thorn> fifoemptymsk = 1 << ep->num; USB_OTG_MODIFY_REG32(&pdev->regs.DREGS->DIEPEMPMSK, 0, fifoemptymsk);
2012-10-01T03:36:17 < Thorn> this is what it does
2012-10-01T03:37:31 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 265 seconds]
2012-10-01T03:38:00 < zyp> well, try the solution from that gist
2012-10-01T03:39:07 < Thorn> works great. thanks much
2012-10-01T03:39:22 < Thorn> I wonder how many other surprises are hidden there
2012-10-01T03:42:50 < zyp> :)
2012-10-01T03:43:10 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-01T03:43:11 < zyp> probably less than you would have to deal with writing a stack from scratch :p
2012-10-01T03:48:24 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-01T04:14:46 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-01T04:14:46 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-01T04:14:46 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-01T04:33:53 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-10-01T05:00:15 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-01T05:42:06 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 245 seconds]
2012-10-01T06:00:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-01T06:21:31 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-01T06:21:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-01T06:34:08 -!- WorkerBuzz [~Buzz@75-167-129-50.bois.qwest.net] has joined ##stm32
2012-10-01T07:02:30 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-01T08:22:21 < dongs> useless fucking HK post.
2012-10-01T08:22:31 < dongs> sent collected on 26th: still sitting there
2012-10-01T08:22:36 < dongs> shit collected on 27th: departed
2012-10-01T08:22:51 < dongs> shit "shipped" on 28th still not registered
2012-10-01T08:23:39 < jpa-> gls: stuff shipped on 26th, stuck in denmark since 27th
2012-10-01T08:47:57 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-01T08:53:42 < dongs> cool. blagged 3 free feeders for dicknplace
2012-10-01T08:53:54 < dongs> need to figure out how to clone the fucking things
2012-10-01T08:53:58 < dongs> not paying 300$/ea
2012-10-01T08:54:09 < emeb_mac> got yer own picknplace machine?
2012-10-01T08:54:25 < dongs> had one for years
2012-10-01T08:54:31 < emeb_mac> wowzer.
2012-10-01T08:54:47 < emeb_mac> fun programming that
2012-10-01T08:54:51 < dongs> its acutally not bad
2012-10-01T08:55:09 < dongs> some csv output fuckery and some vi macro magic
2012-10-01T08:55:17 < dongs> and the file is almost good to  go into the import shit
2012-10-01T08:55:29 < emeb_mac> cool. How many different parts can you run at once?
2012-10-01T08:55:35 < dongs> the software is pure shit of course
2012-10-01T08:56:21 < emeb_mac> so I guess you've got a reflow machine too of course.
2012-10-01T08:56:27 < dongs> ive for like 20 cut tape and 10+34 reel
2012-10-01T08:56:34 < dongs> yea.
2012-10-01T08:56:54 < emeb_mac> don't need a CM
2012-10-01T08:57:14 < dongs> For low-volume, no:)
2012-10-01T08:57:29 < emeb_mac> old slow machines?
2012-10-01T08:57:46 < dongs> its not too bad. just annoying.
2012-10-01T08:58:05 < emeb_mac> rather be programming than feeding the machines?
2012-10-01T08:58:05 < dongs> i.e. 've done 100-200batch runs on teh shit but anything more than that I'll get someone else to do oit
2012-10-01T08:58:30 < emeb_mac> still - nice for small stuff
2012-10-01T08:58:41 < dongs> yea
2012-10-01T08:59:04 < emeb_mac> so is abusemark.com your outfit?
2012-10-01T08:59:11 < dongs> the troll part of it sure
2012-10-01T08:59:26 < emeb_mac> there's a non-troll part?
2012-10-01T08:59:34 < dongs> almost
2012-10-01T09:01:13 < emeb_mac> saw the schematic for the led matrix earlier. was still unfinished at that time.
2012-10-01T09:01:46 < dongs> yea. didnt progress since then cause i went to sleep. need to do input/output + buffer + power + some jumpers for things that are configurable
2012-10-01T09:01:49 < dongs> oh and usb vs can question
2012-10-01T09:02:19 < dongs> you want usb or can on it
2012-10-01T09:02:32 < emeb_mac> usb seems more widely applicable.
2012-10-01T09:02:44 < dongs> yeah. thats what i thought.
2012-10-01T09:04:35 < emeb_mac> If you were to get really clever then you could use usb to load up a display list for a whole chain.
2012-10-01T09:05:04 < emeb_mac> ie - host pc -> usb -> 1st node -> spi -> 2nd node -> spi -> etc
2012-10-01T09:05:37 < dongs> right
2012-10-01T09:06:38 -!- NonaSuomy is now known as TeknoJuce
2012-10-01T09:19:53 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-01T09:43:21 < dongs> emeb_mac: im gonna route RGB to each driver in best-fit way
2012-10-01T09:43:34 < dongs> so the order isnt gonna be rgbrgb etc for filling the line.
2012-10-01T09:43:45 < dongs> or else its gonna be one hell of a mess
2012-10-01T09:44:20 < dongs> its gonna be something like one driver doing all of blue + half red and another one doing half red + all of green
2012-10-01T09:44:37 < dongs> unless you have reasons why this shouldnt be done
2012-10-01T09:44:50 < zyp> are you sure running all colors at the same current is a good idea?
2012-10-01T09:44:58 < dongs> right uh,, i just thought of that
2012-10-01T09:45:11 < dongs> cause you have individual Rset for each r/g/b part
2012-10-01T09:45:15 < dongs> crap.. shitty routing it is then lol
2012-10-01T09:46:29 < dongs> and i was raging earlier, datasheet draws 15pins
2012-10-01T09:46:38 < dongs> actual schematic has 16 per row..
2012-10-01T09:46:48 < dongs> fucking china cant figure shit out
2012-10-01T09:47:22 < dongs> http://www.seeedstudio.com/depot/datasheet/2088RGBMatrix.pdf
2012-10-01T09:47:28 < dongs> 2.54*15
2012-10-01T09:49:01 < dongs> (thats 15 spaces, but  they should draw 16 <pins>)
2012-10-01T09:49:31 < zyp> wat
2012-10-01T09:49:39 < dongs> zyp, the drawing
2012-10-01T09:49:50 < zyp> yes, you are correct
2012-10-01T09:51:00 < dongs> haha, even their math is fucked
2012-10-01T09:51:04 < dongs> cuz 2.54*15 is NOT 35.56
2012-10-01T09:51:31 < dongs> looks like they got a off-by-one on both spaces and pins
2012-10-01T09:51:38 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 255 seconds]
2012-10-01T09:52:09 < zyp> well, it's obvious that there should be 16 pins per row since there is 32 in total
2012-10-01T09:52:31 < dongs> right
2012-10-01T09:53:00 < dongs> which is what I originaly said
2012-10-01T09:53:09 < dongs> schematic = 16, pins = fucked.
2012-10-01T09:53:36 < zyp> hmm, all blue cols first, then all reds, then top rows, then greens, then bottom rows
2012-10-01T09:53:48 < dongs> yeah
2012-10-01T09:53:52 < zyp> and your driver has them all interleaved?
2012-10-01T09:54:02 < dongs> its like
2012-10-01T09:54:05 < dongs> 8B | 8R
2012-10-01T09:54:09 < dongs> D | 8G | D
2012-10-01T09:54:16 < dongs> yeah, shits interleaved
2012-10-01T09:54:25 < dongs> outA[0], [1], [2]
2012-10-01T09:54:55 < zyp> can see how that would be a pain
2012-10-01T09:55:00 < dongs> lol
2012-10-01T09:55:24 < zyp> matrix is probably made for having an individual driver for each color instead
2012-10-01T09:56:20 < dongs> theres plenty of space, it shouldnt be THAT bad to route it "proper"
2012-10-01T09:56:52 < zyp> yeah
2012-10-01T09:57:49 < zyp> it's about twice as big as the ones I've got, and the pin positions is much more sane
2012-10-01T09:59:38 < zyp> http://bin.jvnv.net/f/xZq63.JPG <- I mean, look at the hole pattern, the square in the center is not large, you can't fit and route huge chips there
2012-10-01T10:04:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-01T10:04:30 < dongs> haha
2012-10-01T10:04:33 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-01T10:10:04 -!- dekar [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-10-01T10:11:37 < dongs> ok it doesnt look too bad.
2012-10-01T10:15:17 < dongs> piece of shit doesnt provide info on how long the pins are...
2012-10-01T10:15:42 < dongs> i should probably keep all components on the back side
2012-10-01T10:16:05 < dongs> and matrix on the clean side.
2012-10-01T10:16:31 < zyp> mine has some spacer pins, so there is a 2mm or so gap between board and matrix
2012-10-01T10:17:14 < dongs> not enouygh for usb socket
2012-10-01T10:17:26 < dongs> but the fucking china datasheet doesnt say anyway
2012-10-01T10:22:59 < dongs> haahha
2012-10-01T10:23:02 < dongs> someone must have bitched at yageo
2012-10-01T10:23:07 < dongs> for plastic tape
2012-10-01T10:23:13 < dongs> new reel of resistors i got is back to paper
2012-10-01T10:24:34 < dongs> wait, datecode is liek a week ago
2012-10-01T10:26:40 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-01T10:26:41 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-01T10:26:41 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-01T10:30:34 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-10-01T10:37:23 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-01T10:44:01 < dongs> hmm i dunno if I wanna do latch by solder jumper
2012-10-01T10:44:05 < dongs> or just run it through the chip anyway. hmm
2012-10-01T10:45:49 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-01T11:01:10 -!- dekar [~dekar@dslb-088-068-016-069.pools.arcor-ip.net] has joined ##stm32
2012-10-01T11:01:11 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-01T11:06:17 -!- dekar_ [~dekar@dslb-088-068-016-069.pools.arcor-ip.net] has joined ##stm32
2012-10-01T11:06:20 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-01T11:06:47 -!- dekar [~dekar@dslb-088-068-016-069.pools.arcor-ip.net] has quit [Ping timeout: 255 seconds]
2012-10-01T11:09:56 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-01T11:11:25 < cjbaird> golly, what have I accomplished here..? http://i.imgur.com/BFQCs.gif
2012-10-01T11:12:54 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-01T11:12:57 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-01T11:13:07 < cjbaird> I can get the kernel down to 304kB if had a 80387..
2012-10-01T11:13:30 < dongs> why theres nothing moving
2012-10-01T11:13:37 < dongs> or does it turn into goatse
2012-10-01T11:13:40 < dongs> afrer a few minutes
2012-10-01T11:13:50 <+Steffanx> Give it a try dong
2012-10-01T11:13:51 <+Steffanx> s
2012-10-01T11:15:46 < jpa-> cjbaird: next, put uclinux on cortex-m3
2012-10-01T11:16:40 < cjbaird> Bah. Sable not only goes multi-user, it's got emacs, gcc, bash, /uucp/, cnews, and the all-important zgv. :)
2012-10-01T11:16:51 < cjbaird> So.. Internets 1990 style.
2012-10-01T11:17:05 < dongs> RMS would be proud
2012-10-01T11:17:43 < cjbaird> That's a /real/ Linux kernel (and Debian system) too, not ucLinux.
2012-10-01T11:17:52 < dongs> do you only  dialup to use gnu fetchmail to grab your email, then read it offline in emacs, if you wanna view any webpages you email an address to a server which later emails you tarballs of the webpage you wanted to see right
2012-10-01T11:18:15 < cjbaird> ftpmail was awesome, and used it for years. :)
2012-10-01T11:18:31 < dongs> nutcase
2012-10-01T11:19:10 < cjbaird> To: ftpmail@uow.oz.au Subject: send ftp.apple.com /pub/tmp/samanthafox/* ... everything a growing boy in the 1980s could need..
2012-10-01T11:19:51 < cjbaird> There really was a pr0n direction on Apple's main ftp server for a while..
2012-10-01T11:20:04 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-01T11:20:04 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-01T11:20:04 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-01T11:22:44 < cjbaird> "I've got 720kB of porn on this bad boy."
2012-10-01T11:26:01 <+Steffanx> In ASCII art, YAY
2012-10-01T11:26:45 <+Steffanx> People in the 80's were really desperate :P
2012-10-01T12:02:37 < gnomad> ascii art goes back to the 1920s
2012-10-01T12:03:06 < gnomad> http://vimeo.com/7088524
2012-10-01T12:04:30 <+Steffanx> help
2012-10-01T12:05:24 <+Steffanx> A 1 hour video about THAT :S
2012-10-01T12:07:29 < gnomad> yes
2012-10-01T12:07:54 < gnomad> i happen to really like all of Jason Scott's talks/documentaties
2012-10-01T12:08:23 < gnomad> he did a feature length docu on text-based adventures
2012-10-01T12:10:15 -!- tavish3 [~tavish@120.59.44.211] has joined ##stm32
2012-10-01T12:19:05 < cjbaird> That's the same guy who played-up some 1990s BBS kiddies as E71TE ANSI ArTiSt5? Meh meh.
2012-10-01T12:20:41 < cjbaird> rage level: calling the Gameboy the original of all retro computer games
2012-10-01T12:21:23 < cjbaird> There's a shit weekly TV show here in Oz called "Good Game" that does that all the fucking time..
2012-10-01T12:23:26 -!- BrainDamage [~yaaic@217.200.200.234] has joined ##stm32
2012-10-01T12:23:50 < cjbaird> Good Game is also one of those failtopias that have a feminist-correctness rules about always calling a joystick a 'controller'. "Oh, you show writers have a problem with that? You're Fired! <replace him with a grrl with tatts>"
2012-10-01T12:27:18 < dongs> todays wank: http://i.imgur.com/adPlz.jpg
2012-10-01T12:27:45 < dongs> note a couple fucking tombstoned caps
2012-10-01T12:27:47 <+Steffanx> waow
2012-10-01T12:28:03 <+Steffanx> You failed dongs
2012-10-01T12:28:07 < dongs> i know :(
2012-10-01T12:28:53 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-01T12:29:05 < dongs> mac trash
2012-10-01T12:29:29 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-01T12:29:33 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-01T12:40:48 < zyp> is that 0603?
2012-10-01T12:41:18 < dongs> as in passives? sure
2012-10-01T12:41:38 -!- BrainDamage [~yaaic@217.200.200.234] has quit [Read error: Connection reset by peer]
2012-10-01T12:41:39 < karlp> usb OTG like on f107 and f2/f4 (and f4 disco board) should be usable with android accessory mode right?
2012-10-01T12:41:44 < zyp> I mean the tombstoned caps
2012-10-01T12:41:53 < dongs> zyp: yes
2012-10-01T12:42:00 < Thorn> did you pay for automatic assembly for a non-commercial project?
2012-10-01T12:42:01 < zyp> karlp, yes
2012-10-01T12:42:15 < dongs> Thorn: wut
2012-10-01T12:42:51 < zyp> dongs, how did you manage that? even with my shitty technique I've never had a 0603 tombstone
2012-10-01T12:42:53 -!- BrainDamage [~yaaic@217.200.200.234] has joined ##stm32
2012-10-01T12:43:00 < zyp> just a bunch of 0402s
2012-10-01T12:43:36 < Thorn> I see fiducial marks on your panels
2012-10-01T12:43:58 < dongs> zyp, perhaps the thermal/weird stuff, i dont usually have shit tombstone either, maybe not enough paste on the pad or something like that.
2012-10-01T12:44:06 < dongs> Thorn: yea so, they're on all my panels
2012-10-01T12:44:39 < dongs> the fun begins when im gonna solder them up into a long bar.
2012-10-01T12:44:44 < dongs> still deciding how im gonna do that
2012-10-01T12:45:59 < dongs> ive got a bunch of ribbon cable.. maybe that.. though tinning/stripping short pieces of that shit is gonna be annoying
2012-10-01T12:48:31 < cjbaird> Score a light cut with a scalpel across the ribbon cable 3mm from the end, wank the insulation off in a vice? I managed SCSI-1 50-way ribbon that way..
2012-10-01T12:53:35 -!- BrainDamage [~yaaic@217.200.200.234] has quit [Ping timeout: 256 seconds]
2012-10-01T12:54:19 -!- BrainDamage [~yaaic@217.200.200.234] has joined ##stm32
2012-10-01T12:56:30 < dongs> haha
2012-10-01T12:59:04 < cjbaird> http://i.imgur.com/ocjU9.jpg
2012-10-01T12:59:32 < dongs> is that tinned.
2012-10-01T13:00:13 < dongs> man i wish my time was worthless
2012-10-01T13:00:40 < dongs> 240 stipped wires/tin/solder
2012-10-01T13:01:04 -!- BrainDamage [~yaaic@217.200.200.234] has quit [Read error: Connection reset by peer]
2012-10-01T13:01:55 < cjbaird> You flip the vice onto its side, then run a tinned soldering tip across the wires all at once. Takes less than a minute to do the whole lot.
2012-10-01T13:02:17 < dongs> youre assuming i have a vice
2012-10-01T13:03:24 -!- BrainDamage [~yaaic@217.200.200.234] has joined ##stm32
2012-10-01T13:04:15 < cjbaird> Well you'll need one, little boy.
2012-10-01T13:04:29 < dongs> or this chink idc cable i got will work same way
2012-10-01T13:05:16 < BrainDamage> Or in laurenceb's case: you'll need a little boy
2012-10-01T13:05:23 < karlp> surely a little frying pan full of liquid solder's easier?
2012-10-01T13:06:08 < BrainDamage> I usually put a tomato can on the stove to make a soldering pot
2012-10-01T13:06:18 < dongs> its working but i dont think i wanna spend all night doing this
2012-10-01T13:06:19 < dongs> hmm
2012-10-01T13:06:22 < dongs> need a better solution
2012-10-01T13:06:23 < BrainDamage> Works perfectly
2012-10-01T13:07:20 < dongs> i save clippings from HC49 crystal legs.
2012-10-01T13:07:25 < dongs> finalyl theres a use for them.
2012-10-01T13:07:28 < dongs> and i have hundreds.
2012-10-01T13:07:29 < dongs> fuck yea.
2012-10-01T13:08:36 < BrainDamage> To make body piercings ? :p
2012-10-01T13:11:15 < dongs> fuck yea its working
2012-10-01T13:11:57 < dongs> http://i.imgur.com/ZYDEt.jpg
2012-10-01T13:16:31 < dongs> BrainDamage: i used htem as board to board GND bridge in some other product
2012-10-01T13:17:26 -!- BrainDamage [~yaaic@217.200.200.234] has quit [Ping timeout: 252 seconds]
2012-10-01T13:17:58 < dongs> cjbaird: whats your take on assange being labeled enemy of the state
2012-10-01T13:35:45 < cjbaird> It's his greatest fantasy come true...
2012-10-01T13:36:17 < dongs> what about the discovery that allegedly his DNA wasnt inside those sluts
2012-10-01T13:38:13 < cjbaird> Didn't hear of that. I didn't think any useful samples would've been available anyway, as the two girl waited about a week or so before reporting.
2012-10-01T13:39:31 < cjbaird> Funny how schizophrenics always say that they're the personal enemy of whoever the The Most Powerful Entity is at the time..
2012-10-01T13:40:08 < cjbaird> The psycho who Dad had running his motel claimed that Julia Gillard was personally out to get her..
2012-10-01T13:41:29 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-10-01T13:41:50 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-01T13:43:41 < dongs> 30/240: done
2012-10-01T13:45:25 < cjbaird> I love the part where she claims that Centerlink (i.e: the Dole Office) has secret undercover assassination agents: http://i.imgur.com/ciizx.gif
2012-10-01T13:49:28 < cjbaird> moar of the Ann saga: http://i.imgur.com/pAb9X.png
2012-10-01T13:53:35 < cjbaird> I'd estimate that she's probably cost my Dad about $500,000 in loses due to the dropping motel sale price.
2012-10-01T13:54:54 < dongs> who is ann? one of the chicks he slept with?
2012-10-01T13:56:34 < cjbaird> lol
2012-10-01T13:57:15 < dongs> dunno all that shit is tldr and its pngs so i cant ctrl+f the shit for anything funny
2012-10-01T13:58:55 < cjbaird> Dad owned a motel, and when he retired, he brought resident managers to run the place. The first one turned the place into a Cat&Dog kennel, and Dad barely missed being charged for public health&safety. Ann was the second attempt. After that, Dad had to go back and run the place by himself, which he's been doing since the start of the year.
2012-10-01T13:59:43 < cjbaird> red highlight: funniest parts. Hitler is mentioned
2012-10-01T14:01:21 -!- tavish3 [~tavish@120.59.44.211] has quit [Read error: Operation timed out]
2012-10-01T14:04:38 -!- tavish3 [~tavish@120.59.46.136] has joined ##stm32
2012-10-01T14:05:59 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-10-01T14:07:17 -!- Netsplit over, joins: @ChanServ
2012-10-01T14:07:18 < dongs> 60/240: done
2012-10-01T14:21:59 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-01T14:27:40 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-01T14:39:45 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-01T14:40:20 < dongs> Laurenceb_: what were you babbling about ppm
2012-10-01T14:40:38 < Laurenceb_> how do you capture it?
2012-10-01T14:40:46 < Laurenceb_> how often does the timer overflow?
2012-10-01T14:41:19 < dongs> overflow doesnt matter
2012-10-01T14:41:25 < dongs> timer runs at 1mhz
2012-10-01T14:41:30 < Laurenceb_> flyback: bubbas cock?
2012-10-01T14:41:31 < dongs> more than  enough to capture 22ms of crap
2012-10-01T14:42:13 < dongs> also it doesnt acutally matter if it overflows or not - http://code.google.com/p/afrodevices/source/browse/trunk/baseflight/src/drv_pwm.c#419
2012-10-01T14:42:34 < dongs> youre only interested in prev, now, and if prev>now is > 2700us
2012-10-01T14:56:11 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-01T14:57:11 < dongs> took a bath, continuing t o solder bridges
2012-10-01T15:00:31 < dongs> 90/240
2012-10-01T15:02:42 < dongs> http://i.imgur.com/jdQFs.jpg progress
2012-10-01T15:03:38 < BrainDamage> are you soldering them by hand?
2012-10-01T15:04:43 < dongs> BrainDamage: the bridges between them yes
2012-10-01T15:05:02 < dongs> the jumper wires between each block
2012-10-01T15:05:15 < BrainDamage> sounds painful
2012-10-01T15:06:05 < dongs> its not so bad.
2012-10-01T15:06:12 < dongs> i found out if i make a bunch of jumpers before hand
2012-10-01T15:06:15 < dongs> and just stick them in and solder
2012-10-01T15:06:16 < dongs> its faster
2012-10-01T15:06:19 < dongs> than doing one by one and soldering
2012-10-01T15:06:45 < dongs> jumper material: free from HC49 crystal legs after assembly
2012-10-01T15:14:56 < ABLomas> nice
2012-10-01T15:20:29 < Laurenceb_> cool
2012-10-01T15:20:40 < Laurenceb_> that pwm code doesnt look fun
2012-10-01T15:23:26 < dongs> pwm callback?
2012-10-01T15:25:01 < Laurenceb_> the whole time code
2012-10-01T15:25:10 < Laurenceb_> tons of it :S
2012-10-01T15:25:18 < Laurenceb_> but yeah the pppm capture looks simple
2012-10-01T15:29:07 < dongs> the code is fine
2012-10-01T15:29:15 < dongs> its very generic and can easily adapt to any setup
2012-10-01T15:29:43 < dongs> (within limits of rc-use anyway)
2012-10-01T15:33:08 < Laurenceb_> is it using coocox functions?
2012-10-01T15:33:20 < dongs> nope
2012-10-01T15:35:53 < dongs> its just clean code, you might not be used to seeing it :p
2012-10-01T15:37:14 < dongs> ledbar: almost done
2012-10-01T15:37:38 < Laurenceb_> ok.. where is the ppmcallback called from?
2012-10-01T15:38:27 < dongs> its registered
2012-10-01T15:38:30 < dongs> during init
2012-10-01T15:38:42 < dongs> pwmInConfig(port, callback, channel
2012-10-01T15:38:44 < Laurenceb_> ah
2012-10-01T15:39:15 < dongs> it sits on all timer irqs and calls stuff if any is registered
2012-10-01T15:40:04 < Laurenceb_> got it
2012-10-01T15:43:27 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-01T15:51:26 < dongs> time to deadbug a SOT23-6
2012-10-01T15:51:39 < Laurenceb_> easy
2012-10-01T15:51:42 < Laurenceb_> try a sc-70
2012-10-01T15:52:04 -!- tavish3 [~tavish@120.59.46.136] has quit [Quit: leaving]
2012-10-01T16:06:45 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-01T16:09:51 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-01T16:13:42 < dongs> ugh
2012-10-01T16:13:45 < dongs> leg broke off..
2012-10-01T16:13:46 < dongs> lame
2012-10-01T16:14:14 < Bird|lappy> quite.
2012-10-01T16:14:59 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-01T16:21:34 < dongs> ya bitches
2012-10-01T16:21:47 < dongs> http://i.imgur.com/VzAKW.jpg
2012-10-01T16:24:56 < Posterdati> hi
2012-10-01T16:25:26 < Posterdati> does arm-none-eabi-gcc from Codesourcery supports fp registers in maths operations?
2012-10-01T16:29:11 < jpa-> Posterdati: yes, but the newlib that comes with it does not
2012-10-01T16:29:18 < jpa-> (you have to recompile newlib)
2012-10-01T16:29:38 < Posterdati> I may use an external newlib
2012-10-01T16:29:41 < Posterdati> recompiled
2012-10-01T16:30:38 < Posterdati> so in the context switch I have to take care about fpx registers saving. Haven't I to?
2012-10-01T16:31:09 < dongs> yes
2012-10-01T16:31:36 < Posterdati> good
2012-10-01T16:31:54 < Posterdati> dongs: finally my context switch in PendSV exception handler works
2012-10-01T16:33:13 < dongs> excellent.
2012-10-01T16:33:18 < Posterdati> stmdb r12!, { f0-f7 }
2012-10-01T16:33:33 < Posterdati> may I save all with one instruction?
2012-10-01T16:33:48 < dongs> im surprised your job who doens't allow you to use opensores software isn't willing to provide you with proper tools
2012-10-01T16:33:49 < Posterdati> stmdb r12!, { r0, r4-r11, f0-f7 }
2012-10-01T16:34:40 < Posterdati> dongs: well I either don't won't to user thrd party, because I need to understand arm and cortex-M architecture first
2012-10-01T16:35:06 < Posterdati> this is a job but also a learning process too
2012-10-01T16:35:34 < dongs> you use fstmiad t o store fp shit.
2012-10-01T16:35:54 < Posterdati> so a different instruction
2012-10-01T16:36:09 < Posterdati> fstmdb r12!. { f0-f7 }
2012-10-01T16:36:10 < Posterdati> ???
2012-10-01T16:36:19 < Posterdati> fstmdb r12!, { f0-f7 }
2012-10-01T16:36:44 < Posterdati> load and store multiple instructions are amazing
2012-10-01T16:36:57 < Posterdati> so simple and yet powerfull
2012-10-01T16:38:14 < Posterdati> does in c/c++ a fp registers equals a double?
2012-10-01T16:38:28 < dongs> i think so
2012-10-01T16:38:36 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-01T16:38:36 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-01T16:38:36 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-01T16:38:43 < dongs> the code i have saves fpu regs to... sram
2012-10-01T16:39:50 < zyp> well, saving them to flash would be stupid
2012-10-01T16:40:10 < dongs> lkolol
2012-10-01T16:40:23 < dongs> zyp, could save htem to external SPI eeprom
2012-10-01T16:40:33 < zyp> save them to the cloud
2012-10-01T16:40:38 < dongs> perfect.
2012-10-01T16:40:44 < dongs> iCloud to be exact.
2012-10-01T16:40:58 < Posterdati> /tmp/cc1eQDss.s:224: Error: bad instruction `fstmia r12!,{f0-f7}'
2012-10-01T16:40:58 < dongs> http://www.youtube.com/watch?v=Y5EL3qCRfh4 ive spent several evenings on something I could have bought from china for $10
2012-10-01T16:41:13 < dongs> fstmiad
2012-10-01T16:43:09 < ABLomas> ql
2012-10-01T16:43:15 < ABLomas> wanna one
2012-10-01T16:43:22 < ABLomas> wait, wanna ten!
2012-10-01T16:43:47 < Posterdati> dongs: for load? fldmiad?
2012-10-01T16:43:59 < dongs> yeah
2012-10-01T16:44:20 < Posterdati> /tmp/ccTarcpt.s:224: Error: selected processor does not support Thumb mode `fstmiad r12!,{f0-f7}'
2012-10-01T16:44:25 < dongs> lo
2012-10-01T16:44:34 < dongs> wait what>?
2012-10-01T16:44:55 < dongs> look at google
2012-10-01T16:44:57 < dongs> for some lunix hits
2012-10-01T16:45:03 < Posterdati> ok
2012-10-01T16:45:04 < dongs> about dudes using that instruction
2012-10-01T16:45:10 < dongs> theres some more crap to do
2012-10-01T16:49:07 < Posterdati> :(
2012-10-01T16:51:27 < Posterdati> ...
2012-10-01T16:57:47 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 255 seconds]
2012-10-01T17:00:50 < Posterdati> dongs: it's a gcc problem
2012-10-01T17:01:02 < dongs> <shocking>
2012-10-01T17:01:57 < jpa-> Posterdati: what compiler flags do you pass to gcc?
2012-10-01T17:02:10 < dongs> -funrollloops
2012-10-01T17:02:27 < Posterdati> jpa-: eh
2012-10-01T17:02:29 < Posterdati> wait
2012-10-01T17:02:35 < jpa-> -fno-fun-trolls
2012-10-01T17:02:47 < Posterdati> -fno-working-executables
2012-10-01T17:03:19 < Posterdati> CPPFLAGS=-mthumb 		\
2012-10-01T17:03:19 < Posterdati> 	${CPU}			\
2012-10-01T17:03:19 < Posterdati> 	${FPU}			\
2012-10-01T17:03:19 < Posterdati> 	-ffunction-sections	\
2012-10-01T17:03:19 < Posterdati> 	-fdata-sections		\
2012-10-01T17:03:20 < Posterdati> 	-fno-unwind-tables	\
2012-10-01T17:03:22 < Posterdati> 	-fno-exceptions		\
2012-10-01T17:03:25 < Posterdati> 	-nostartfiles		\
2012-10-01T17:03:27 < Posterdati> 	-MD			\
2012-10-01T17:03:29 < Posterdati> 	-Wall			\
2012-10-01T17:03:31 < Posterdati> 	-DPART_${PART}		\
2012-10-01T17:03:33 < Posterdati> 	-std=gnu++0x	   	\
2012-10-01T17:03:35 < Posterdati> 	-c
2012-10-01T17:03:39 < dongs> http://bcas.tv/paste
2012-10-01T17:03:40 < Posterdati> CPU=cortex-m3
2012-10-01T17:03:42 < Posterdati> FPU=
2012-10-01T17:04:01 < Posterdati> sorry
2012-10-01T17:04:02 < jpa-> huh? and how exactly do you expect to be able to use FPU instructions if you specify cortex-m3?
2012-10-01T17:04:38 < Posterdati> because it is a cortex-m3
2012-10-01T17:04:46 < zyp> …
2012-10-01T17:04:55 < zyp> m3 doesn't have an FPU
2012-10-01T17:05:01 < Posterdati> LOL
2012-10-01T17:05:17 < zyp> m4f is the only cortex-m with FPU
2012-10-01T17:05:26 < Posterdati> aha
2012-10-01T17:05:42 < jpa-> (m4f is in stm32f4 and stm32f3)
2012-10-01T17:07:06 < Posterdati> no I've got stm32f10x
2012-10-01T17:07:29 < jpa-> so software float for you
2012-10-01T17:08:52 < Posterdati> ok
2012-10-01T17:09:06 < Posterdati> I was confusing with an old str912 I've got here
2012-10-01T17:10:07 < dongs> uh so on f10x you dont hae any FPU registers to save.
2012-10-01T17:10:11 < dongs> so theres nothing to carea bout
2012-10-01T17:10:21 < Posterdati> good :)
2012-10-01T17:10:34 < Posterdati> so Codesourcery newlib will fit
2012-10-01T17:11:04 < Thorn> dongs: note how yellow LEDs give off red and green reflections
2012-10-01T17:11:50 < Posterdati> dongs: I need to compile gsl library for bare metal application...I hope it will work :)
2012-10-01T17:12:47 < dongs> gnu..
2012-10-01T17:13:31 < dongs> i still havent fixed the fucking level shifter
2012-10-01T17:13:38 < dongs> so i have to run the shit at like 2.8V
2012-10-01T17:13:43 < dongs> so brightness is crap etc.
2012-10-01T17:13:56 < dongs> the drivers need 5V I/O
2012-10-01T17:14:07 < dongs> tried to deadbug a sot23 buffer and fucked it up
2012-10-01T17:29:50 < Thorn> what exactly does CDC-ACM send over its interrupt EP? the spec isn't exactly clear on that
2012-10-01T17:30:07 < dongs> ep0?
2012-10-01T17:30:10 < jpa-> raw data, mostly
2012-10-01T17:30:30 < Thorn> no, not ep0. control interface (one interrupt IN EP)
2012-10-01T17:30:46 < dongs> oh duno
2012-10-01T17:30:58 < Thorn> data is sent over 2 bulk EPs which are in the data interface afaik
2012-10-01T17:31:17 < jpa-> ah, interrupt
2012-10-01T17:32:17 < jpa-> no idea, actually :)
2012-10-01T17:32:23 < Thorn> it has 1) class-specific ontrol requests (ep0) - set baud rate etc. 2) data interface - a pair of bulk EPs and 3) control interface - one interrupt IN EP
2012-10-01T17:32:55 < jpa-> maybe RTS/CTS/DTR?
2012-10-01T17:33:02 < Thorn> there's an example in the ST stack but ut doesn't do anything meaningful with that EP as far as I can tell
2012-10-01T17:33:10 < Thorn> probably, but what's the format
2012-10-01T17:33:27 < dongs> the format is on a need-to-know basis
2012-10-01T17:33:32 < dongs> Thorn: here's your mission
2012-10-01T17:33:40 < dongs> Thorn: instead of fucking with cdcacm, emulate FTDI
2012-10-01T17:33:44 < dongs> say FT232RL or so.
2012-10-01T17:33:49 < dongs> then reuse their vid/pid
2012-10-01T17:33:52 < dongs> instant profit.
2012-10-01T17:33:58 < dongs> publish y our code on github for e veryone else.
2012-10-01T17:34:27 < Thorn> send me a hardware analyzer
2012-10-01T17:34:34 < zyp> Thorn, as far as I understand, inbound metadata
2012-10-01T17:34:45 < dongs> i've got one.
2012-10-01T17:34:51 < zyp> like changed state on the inbound extra lines
2012-10-01T17:35:37 < zyp> dongs, what would the benefit be?
2012-10-01T17:36:15 < dongs> zyp: should be obvious
2012-10-01T17:36:21 < dongs> driverless install on any real OS
2012-10-01T17:36:22 < Thorn> using ftdi drivers for non-ftdi devices?
2012-10-01T17:36:37 < dongs> Thorn: yes.
2012-10-01T17:36:51 < dongs> or rather not driverless but annoyingless.
2012-10-01T17:37:01 < Thorn> since when is linux a real os in your book? because windows downloads a driver when you first plug in an ftdi device
2012-10-01T17:37:19 < dongs> see my correction
2012-10-01T17:39:02 < Thorn> hehehe.
2012-10-01T17:40:57 < zyp> I wonder just what the rationale for windows not recognizing ACM devices just by class is
2012-10-01T17:42:01 < Thorn> http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/include/linux/usb/cdc.h#L251
2012-10-01T17:42:51 < Thorn> that looks like a setup packet.
2012-10-01T17:43:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-01T17:43:12 < Thorn> they send setup packets over an interrupt EP?
2012-10-01T17:43:20 < jpa-> does windows already recognize mass storage by class? i remember that back when USB was new, you needed separate drivers for each usb stick in windows
2012-10-01T17:43:21 < Thorn> how clever
2012-10-01T17:43:43 < Thorn> jpa-: it does for a long time
2012-10-01T17:44:16 < zyp> Thorn, same structure as a setup packet, but obviously not a setup packet since those can only be sent on control endpoints towards the device :p
2012-10-01T17:44:59 < Thorn> that's a setup *transaction*
2012-10-01T17:45:28 < Thorn> the packet itself isn't special, it has a PID of data1 or something
2012-10-01T17:46:05 < Thorn> that was their rationale I guess.
2012-10-01T17:46:19 < zyp> sure, but it follows an IN token, making it an IN transaction
2012-10-01T17:46:37 < zyp> setup transactions consists of a data packet following a SETUP token
2012-10-01T17:46:43 < zyp> those are distinct from IN and OUT
2012-10-01T17:49:04 < Thorn> here's the function that handles these on the host http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/drivers/usb/class/cdc-acm.c#L291
2012-10-01T17:50:29 < Thorn> ACM_CTRL_DCD, ACM_CTRL_DSR etc.
2012-10-01T17:50:54 < zyp> yeah, so the inbound lines
2012-10-01T17:50:58 < zyp> also break detection
2012-10-01T17:51:05 < zyp> and error reporting
2012-10-01T17:51:39 < Thorn> that could be useful but I don't have spare endpoints.
2012-10-01T17:51:59 < zyp> then don't use it?
2012-10-01T17:52:12 < Thorn> DCD/DSR make no sense on rs-485 but errors can still happen
2012-10-01T17:52:48 < zyp> weren't you talking about doing a vendor specific device?
2012-10-01T17:53:15 < Thorn> a composite device most likely, with 3 vendor-specific interfaces and 2 cdc-acm (for rs-485)
2012-10-01T17:53:47 < Thorn> what I'm doing now is testing the usb stack on a dev board (instead of finishing the PCB)
2012-10-01T17:56:44 < Thorn> does synopsys license their cores by number of endpoints? (like arm does with nvic interrupts)
2012-10-01T17:56:56 < Thorn> or is it a silicon saving measure
2012-10-01T17:57:04 < zyp> no idea, I imagine silicon
2012-10-01T18:01:00 < Thorn> I guess I'll add red LEDs to indicate errors. (for a total of ~20LEDs)
2012-10-01T18:07:43 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-01T18:09:06 -!- metaxa_ [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-10-01T18:11:29 -!- metaxa [metaxa@us.members.bombshellz.net] has quit [Ping timeout: 260 seconds]
2012-10-01T18:13:39 < Tectu> karlp, ping
2012-10-01T18:14:47 -!- dekar_ [~dekar@dslb-088-068-016-069.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-10-01T18:17:00 -!- metaxa_ [metaxa@us.members.bombshellz.net] has quit [Ping timeout: 252 seconds]
2012-10-01T18:20:34 -!- alcides [~alcides@unaffiliated/alcides] has joined ##stm32
2012-10-01T18:21:30 -!- alcides [~alcides@unaffiliated/alcides] has left ##stm32 []
2012-10-01T18:27:45 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-01T18:33:19 < karlp> Tectu: pong
2012-10-01T18:33:48 < Tectu> karlp, remember the "too long" package issue I had yesterday with oocd so I couldn't load the elf in GDB?
2012-10-01T18:33:59 < Tectu> karlp, ntfreak gave me a fix which works!
2012-10-01T18:34:00 < karlp> yes, did you see ntfreaks response with the xml file?
2012-10-01T18:34:04 < Tectu> karlp, http://www.mail-archive.com/openocd-development@lists.berlios.de/msg18182.html
2012-10-01T18:34:10 < karlp> yeah, that one.
2012-10-01T18:34:12 < Tectu> ah, you saw it :)
2012-10-01T18:34:15 < karlp> still an openocd problem in my mind :)
2012-10-01T18:35:10 < Tectu> I am not posting this here because I think it's an STM32 issue, I thought I let you know ;-)
2012-10-01T18:35:20 < Tectu> I always like when people tell me the solution for a problem I couldn't help them with
2012-10-01T18:40:44 < ntfreak> karlp: i agree, it need sorting on the OpenOCD side
2012-10-01T18:43:29 < zyp> ntfreak, you asked once why people weren't using openocd - part of it might be that this issue is still not fixed, since it seems to be the most common issue people run into
2012-10-01T18:44:28 < Tectu> zyp, what's exactly "that issue"?
2012-10-01T18:44:52 < zyp> openocd not providing that xml information to gdb directly
2012-10-01T18:46:11 -!- daku is now known as DaKu
2012-10-01T18:50:45 < ntfreak> it is an issue that needs sorting, it really only effects cortex_m targets (and certain gdb builds/versions).
2012-10-01T18:50:46 < ntfreak> the change to OpenOCD is not trivial as we need to fix for all targets, and the workaround is not that hard to apply.
2012-10-01T18:51:23 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-10-01T18:52:58 < Thorn> "it really only effects cortex_m targets" - these are what we here care about :)
2012-10-01T18:55:18 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-01T18:55:41 < karlp> is it in the release notes for 0.6.0?
2012-10-01T18:55:52 < karlp> given that gdb versions that run into this are very common these days...
2012-10-01T18:56:46 < cjbaird> I need to do something before I starting hacking up a 80386 emulator. :/
2012-10-01T18:57:07 < Thorn> cjbaird: worst architecture ever
2012-10-01T18:59:32 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 252 seconds]
2012-10-01T18:59:35 < cjbaird> By 'hacking up', that's 'ripping other people's existing code' and putting it together into something I can use for a project. Even that will be a big time-sink.
2012-10-01T19:02:13 < ntfreak> i agree it needs fixing - but unless we have devs step up and help it will go unfixed.
2012-10-01T19:02:13 < ntfreak> reports of the issue are quite rare - codesoucery is the most common version seen and this behaves ok.
2012-10-01T19:02:25 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-01T19:03:26 < cjbaird> I noticed that the raspberrypi jtag work has mainlined in openocd recently..
2012-10-01T19:05:00 < ntfreak> that is a generic sysfs driver, there is also another faster driver pending - http://openocd.zylin.com/758
2012-10-01T19:07:21 < cjbaird> Nuts, I thought it was the Ferster driver.
2012-10-01T19:08:22 < karlp> ntfreak: isn't it self selecting that code sourcery is the most common, when it's the only one that works? ;) (sorry, that's being deliberately mean now)
2012-10-01T19:12:42 < ntfreak> perhaps - mainline gdb was working aswell last time i checked. Not sure what gcc-arm-embedded base their gdb on.
2012-10-01T19:13:00 < karlp> that's the one that wasn't, iirc
2012-10-01T19:14:14 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-01T19:15:10 < ntfreak> yeah i know that one fails - just wondering where the gdb comes from.
2012-10-01T19:16:33 -!- barthess [~barthess@178.154.78.115] has joined ##stm32
2012-10-01T19:19:44 < Tectu> ntfreak, i can place that xml file at any place I want and give the absolut path in my ~/.gdbinit?
2012-10-01T19:19:51 < Tectu> or must .gdbinit be in the same dir as well?
2012-10-01T19:19:58 < karlp> either
2012-10-01T19:20:31 < karlp> if you're not working on any other target types, might as well put it in your ~/.gdbinit
2012-10-01T19:21:13 < Tectu> karlp, not sure if I understand?
2012-10-01T19:21:20 < Tectu> .gdbinit can be in ~ right?
2012-10-01T19:25:25 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-10-01T19:25:28 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-01T19:27:29 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-01T20:17:45 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-10-01T20:22:07 -!- metaxa [metaxa@us.members.bombshellz.net] has quit [Ping timeout: 240 seconds]
2012-10-01T20:33:00 < zyp> Tectu, yes, either mean both
2012-10-01T20:33:20 < Tectu> okay, thanks
2012-10-01T20:33:24 < zyp> gdb checks both ~ and . for .gdbinit and sources both if present
2012-10-01T20:33:47 < zyp> I usually use both, ~ for generic stuff and . for project specific stuff
2012-10-01T20:35:01 < Tectu> cool
2012-10-01T20:38:15 < Tectu> I am not sure what that XML actualy does
2012-10-01T20:38:45 < zyp> it describes for gdb what registers is present in the cpu
2012-10-01T20:38:59 < zyp> so gdb knows which registers to ask for/expect
2012-10-01T20:39:27 < zyp> without it, it defaults to some weird setting, which is incompatible with what openocd actually sends
2012-10-01T20:39:45 < Tectu> registers as in r0 to r15? o0
2012-10-01T20:39:50 < Tectu> or all registers?
2012-10-01T20:40:00 < Tectu> ah, wait
2012-10-01T20:40:03 < zyp> yes, core registers
2012-10-01T20:40:03 < Tectu> they are named there :D
2012-10-01T20:40:13 < zyp> not memory mapped registers :p
2012-10-01T20:40:28 < Tectu> :D
2012-10-01T20:40:29 < karlp> you can apparently use "set arm abi blah" too...
2012-10-01T20:40:36 < karlp> or so I've read on the ninterweb
2012-10-01T20:40:37 < zippe> openocd is trying to handle old gdb, which wants the ARMv4 FPU registers so it can ignore them
2012-10-01T20:40:42 < Tectu> <reg name="f1" bitsize="96" type="arm_fpa_ext"/>
2012-10-01T20:40:47 < Tectu> what's arm_fpa_ext?
2012-10-01T20:41:00 < karlp> do you care?
2012-10-01T20:41:03 < zippe> That's the v4 FPU aka floating point accelerator
2012-10-01T20:41:09 < Tectu> karlp, yes?
2012-10-01T20:41:21 < Tectu> zippe, dosen't my STM32 be have v7?
2012-10-01T20:41:21 < karlp> it's a magic string that gdb recognises
2012-10-01T20:41:30 < Tectu> is there a difference between FPU and FPA? o0
2012-10-01T20:41:38 < zippe> Tectu: There is
2012-10-01T20:41:46 < zippe> Tectu: you are discovering why people hate OpenOCD
2012-10-01T20:41:53 < zyp> Tectu, here is the code in texane/stlink that sends the target description xml to gdb automatically: https://github.com/texane/stlink/blob/master/gdbserver/gdb-server.c#L195
2012-10-01T20:41:55 < zippe> Tectu: i.e. it's stupid
2012-10-01T20:42:10 < zyp> if openocd also did that, you wouldn't have to care
2012-10-01T20:42:14 < zippe> And inexplicably nobody wants to fix it
2012-10-01T20:42:21 < Tectu> You all know, I am a newbie, so I guess I can ask: what's OpenOCDs problem to implement this?
2012-10-01T20:42:22 < karlp> they'd rather write more tcl...
2012-10-01T20:42:24 < Tectu> "the right way"?
2012-10-01T20:42:39 < zippe> Tectu: stuffed if I know, but I gave up on them a year and more ago
2012-10-01T20:42:45 < zyp> the lack of somebody who cares to actually write the code?
2012-10-01T20:42:47 < Tectu> should't it be posible to add this to the target/ script?
2012-10-01T20:43:01 < zippe> Tectu: Probably
2012-10-01T20:43:07 < karlp> probably, I'm sure ntfreak would love your contribtuions
2012-10-01T20:43:11 < Tectu> I always thought people like OpenOCD
2012-10-01T20:43:12 < zippe> Assuming there's some way to handle GDB remote commands at the script level
2012-10-01T20:43:24 < zippe> Tectu: You probably think people "like" Linux too
2012-10-01T20:43:37 < karlp> hehe
2012-10-01T20:43:48 < zippe> Stockholm syndrome is a powerful thing
2012-10-01T20:44:28 < Tectu> what are people using then?
2012-10-01T20:44:34 < Tectu> they also don't like texane/stlink
2012-10-01T20:44:52  * Tectu wonders that dongs didn't joing this discussion yet
2012-10-01T20:45:08 < BrainDamage> stlink and black magic probe seems to be the dominant stuff here
2012-10-01T20:47:41 < zippe> Most debuggers have their quirks
2012-10-01T20:47:48 < zippe> OCD's quirk is that it doesn't work
2012-10-01T20:47:52 < Bird|lappy> zippe, :P I'd actually rather have a *nix box than a Windows one
2012-10-01T20:48:08 < Bird|lappy> some very basic things are still frustrating on Windows
2012-10-01T20:48:15 < zippe> BMP kinda screws up when you connect/disconnect things unexpectedly
2012-10-01T20:48:31 < zippe> Bird|lappy: Don't confuse me with dongles here
2012-10-01T20:50:03 < Bird|lappy> well, as to the debugger debate: I doubt STlink/Texane supports the EFM32s at all, and BMP probably would work, but I already have a general-purpose JTAG adapter (Dangerous Prototypes BusBlaster)
2012-10-01T20:50:32 < Thorn> st-link only supports ST chips
2012-10-01T20:50:59 < Thorn> deliberate firmware limitation
2012-10-01T20:51:18 < zyp> I've tried a bunch of different stuff
2012-10-01T20:52:08 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-01T20:52:15 < zyp> openocd with jtagkey, st-link on discovery boards, BMP
2012-10-01T20:52:38 < zyp> and I also bought a clone jlink, but I bricked that before I got to try it out
2012-10-01T20:54:02 < zyp> it worked at first, but I didn't use it for anything meaningful before an update came along and bricked it
2012-10-01T20:54:38 < zyp> but I probably wouldn't like the hassle of having to use windows anyway
2012-10-01T20:55:30 < Thorn> j-link supports linux afaik
2012-10-01T20:55:51 < zyp> well, then I probably wouldn't like the hassle of having to use linux
2012-10-01T20:56:13 < Bird|lappy> yeah.  openocd supports the jlink, and I think the Segger tools are available for Linux too
2012-10-01T20:56:53 < Thorn> what's the point of using openocd with j-link? :)
2012-10-01T20:57:16 < BrainDamage> if you had the debugger hw already
2012-10-01T20:57:18 < zyp> well, the original reason I switched from openocd was lack of swd support
2012-10-01T20:58:11 < zyp> and yes, this was before swd support was added.
2012-10-01T21:00:46 < Thorn> I have an onboard j-link clone on a Chinese lpc1768 board, worked great iirc. openocd (+ olimex adapter) had problems with it though
2012-10-01T21:01:13 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-01T21:06:33 < emeb> openocd + olimex jtag-tiny works great for my SAM7 stuff.
2012-10-01T21:06:53 < emeb> never tried it w/ STM32 though. BMP is best for that so far...
2012-10-01T21:10:43 < zyp> well, SAM7 is about as outdated as openocd :p
2012-10-01T21:11:07 < emeb> yeah - I designed that board 7 yrs ago...
2012-10-01T21:11:15 < Bird|lappy> and I must be the only EFM32 user in the channel.
2012-10-01T21:11:43 < zyp> Bird|lappy, so far maybe
2012-10-01T21:12:21 < zyp> I'm planning to play with the EFR4D series in the future
2012-10-01T21:12:26 < Bird|lappy> ah.
2012-10-01T21:13:08 < emeb> oh look. EFM32 has M4F.
2012-10-01T21:13:29  * emeb thinks all processors should be M4F. (not really)
2012-10-01T21:13:33 < Bird|lappy> :P
2012-10-01T21:17:33 < Laurenceb_> whats the frequency ?
2012-10-01T21:17:42 < emeb> One thing I've got to say about BMP - I've learned a lot more about GDB since starting to use it.
2012-10-01T21:18:12 < emeb> Laurenceb_: kenneth?
2012-10-01T21:18:32 < Laurenceb_> of the energy micro m4f
2012-10-01T21:19:20 < emeb> 48MHz.
2012-10-01T21:19:25 < emeb> http://www.energymicro.com/products/
2012-10-01T21:19:49 < Laurenceb_> booo
2012-10-01T21:19:55 < emeb> kinda poky
2012-10-01T21:20:03 < emeb> and qfn/bga only
2012-10-01T21:20:18 < emeb> oh wait - qfp too
2012-10-01T21:20:54 < emeb> only 32kB RAM too
2012-10-01T21:21:34 < Bird|lappy> yeah, some of the low-energy peripherals on the EFM32s are pretty intriguing though
2012-10-01T21:21:55 < Bird|lappy> low-speed UART comms for 150nA standby? I'll take it.
2012-10-01T21:23:13 < emeb> M4F parts not stocked at the usual suspects.
2012-10-01T21:24:13 < zyp> energy micro is apparently norwegian, maybe I should just ask them for samples
2012-10-01T21:24:32 < Bird|lappy> zyp, yep, they are indeed norwegian
2012-10-01T21:25:09 < Laurenceb_>  "hello im norwegian, give me free stuff cuz i am too"
2012-10-01T21:25:14 < emeb> Probably just on the other side of the fjord...
2012-10-01T21:25:20 < Laurenceb_> *you are
2012-10-01T21:25:22 < zyp> :p
2012-10-01T21:25:24 < karlp> yay, got my new mrf24j40 modules today
2012-10-01T21:25:27 < zyp> http://www.energymicro.com/freekit
2012-10-01T21:25:51 < emeb> 100% success rate when country = Norway
2012-10-01T21:27:43 < karlp> the energy micro radio stuff looks cool, when it comes out maybe.
2012-10-01T21:28:13 < emeb> that was a highly qualified statement
2012-10-01T21:29:08 < zyp> so does nrf51
2012-10-01T21:29:13 < karlp> q3 2013....
2012-10-01T21:29:17 <+Steffanx> heh
2012-10-01T21:29:24 < karlp> that's for first samples...
2012-10-01T21:30:07 <+Steffanx> digikey stock: q2 2014
2012-10-01T21:30:45 < karlp> oh groovy, they have qfn24 sampling now.
2012-10-01T21:31:18 < karlp> if only batteries were getting that small.
2012-10-01T21:31:36 < Bird|lappy> karlp, well, if 50uAh is enough for your app, they are!
2012-10-01T21:31:47  * Bird|lappy points at the Cymbet EnerChips
2012-10-01T21:32:38 < karlp> I'm still looking at 2xAA :)
2012-10-01T21:32:57 < Bird|lappy> karlp, what are your requirements?
2012-10-01T21:33:56 < karlp> battery powered is hobby work, so, "easy to find, easy to use, easy to charge, easy to replace, run time ~2 years"
2012-10-01T21:34:16 < karlp> I'm not working at the extreme end of energy harvesting
2012-10-01T21:34:25 < karlp> more playing with how much can be done for how long with 2xaas.
2012-10-01T21:34:40 < karlp> those enerchips look pretty interesting
2012-10-01T21:35:19 < Bird|lappy> yeah, the MEC200s are an option in a different form factor for larger jobs
2012-10-01T21:36:05 < karlp> is that from cymbet as well?
2012-10-01T21:36:36 < Bird|lappy> both are available through DigiKey in onesies, are easy enough to charge (you can even get EnerChip+PMIC combos in a single package), and have long enough cycle times that replacement isn't a concern (provided they don't grossly fail to meet spec ofc)
2012-10-01T21:36:41 -!- DaKu is now known as daku
2012-10-01T21:36:43 < Bird|lappy> nope, Infinite Power Solutions
2012-10-01T21:36:49 -!- daku is now known as DaKu
2012-10-01T21:38:47 < Laurenceb_>  1/0 power
2012-10-01T21:38:59 < Bird|lappy> :p
2012-10-01T21:39:48 -!- WorkerBuzz [~Buzz@75-167-129-50.bois.qwest.net] has quit [Ping timeout: 248 seconds]
2012-10-01T21:45:39 < zyp> I tried submitting a request for one of these: http://www.energymicro.com/tools/efm32-giant-gecko-starter-kit-efm32gg-stk3700
2012-10-01T21:45:45 < zyp> seems nice, with usb and everything
2012-10-01T21:46:36 < zyp> I wonder if it's also using the dwg core :p
2012-10-01T21:47:01 < Bird|lappy> dwg?
2012-10-01T21:47:18 < zyp> synopsys designware
2012-10-01T21:47:21 < Laurenceb_> dawg
2012-10-01T21:47:22 < zyp> or something like that
2012-10-01T21:47:29 < zyp> it's the one used in F4
2012-10-01T21:47:30 < karlp> everybody's favourite usb core
2012-10-01T21:47:36 < Bird|lappy> oh. :)
2012-10-01T21:48:03 < Bird|lappy> they custom-design all their peripherals save for DMA, which is the PrimeCell uDMA
2012-10-01T21:48:38  * emeb remembers dealing w/ PrimeCell documentation from his ARM SoC ASIC days...
2012-10-01T21:49:09 < zyp> they are indeed using dwg_otg
2012-10-01T21:49:27 -!- prog` [prog@bornbrilliant.org] has quit [Excess Flood]
2012-10-01T21:49:34 -!- prog [prog@bornbrilliant.org] has joined ##stm32
2012-10-01T21:49:51 < zyp> configured for six endpoints in each direction in addition to EP0
2012-10-01T21:51:14 < zyp> found this in the reference manual: «Portions Copyright © 2010 Synopsys, Inc. Used with permission. Synopsys and DesignWare are registered trademarks of Synopsys, Inc.»
2012-10-01T21:51:21 < zyp> ;)
2012-10-01T21:52:11 -!- WorkerBuzz [~Buzz@75-167-139-152.bois.qwest.net] has joined ##stm32
2012-10-01T22:29:20 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-01T22:50:19 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-01T22:50:19 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-01T22:50:19 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-01T23:02:08 < Thorn> zyp: stm32 manuals say "copyright 2005-2006", do they use old versions of the cores or what?
2012-10-01T23:02:19 -!- mabl00 [~Thunderbi@2a01:198:71f:0:910b:62e2:1207:bd1e] has joined ##stm32
2012-10-01T23:03:27 < mabl00> Hi, if i do push {r0, r1} is it equivalent to push r0; push r1? Or reversed?
2012-10-01T23:03:52 < zyp> reversed I think
2012-10-01T23:04:09 < zyp> let me check
2012-10-01T23:04:47 < zyp> yes, reversed
2012-10-01T23:05:11 < mabl00> zyp: Thx, where did you find it?
2012-10-01T23:05:29 < zyp> I checked what I do in my own code which I know works :p
2012-10-01T23:05:50 < mabl00> zyp: Ah I see - thx again.
2012-10-01T23:06:02 < zyp> keep in mind that stack grows downwards
2012-10-01T23:06:19 < zyp> push {r0, r1} will put r0 followed by r1 in memory
2012-10-01T23:06:41 < zyp> push r1; push r0 will put r1 in memory and then put r0 in front of it
2012-10-01T23:06:47 < mabl00> zyp: Thx, I do. I
2012-10-01T23:08:38 -!- barthess [~barthess@178.154.78.115] has quit [Quit: Leaving.]
2012-10-01T23:09:07 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-10-01T23:34:30 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-01T23:37:41 < Laurenceb_> F3 cant run off 1.8v  :(
2012-10-01T23:37:46 < Laurenceb_> thats annoying
2012-10-01T23:39:13 < zyp> put a boost reg in front of it.
2012-10-01T23:40:48 < Laurenceb_> heh
2012-10-01T23:41:14 < Laurenceb_> meanwhile...
2012-10-01T23:41:15 < Laurenceb_> http://www.youtube.com/watch?v=Wi-yZsTbCYE
--- Day changed Tue Oct 02 2012
2012-10-02T00:08:04 < Laurenceb_> 1.2v is where its at
2012-10-02T00:26:34 < Thorn> anyone seen stm32f3-discovery available anywhere?
2012-10-02T00:33:02 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-02T00:33:37 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-02T00:36:20 -!- WorkerBuzz [~Buzz@75-167-139-152.bois.qwest.net] has quit [Ping timeout: 248 seconds]
2012-10-02T00:39:39 < Laurenceb_> what is it with flyback and mpus
2012-10-02T00:39:52 < Laurenceb_> maybe they protect him from bubba or something
2012-10-02T00:40:25 < emeb> manly protection unit
2012-10-02T00:41:04 < emeb> sorta like a sports cup I guess
2012-10-02T00:41:39  * emeb only uses insecure hardware
2012-10-02T00:42:01 < Laurenceb_> http://www.geeky-gadgets.com/wp-content/uploads/2009/01/nes-controller-jacket.jpg
2012-10-02T00:44:38 < emeb> total nerd-out.
2012-10-02T00:45:44 < Laurenceb_> http://pastebin.com/Y04SrpzR
2012-10-02T00:46:54 < emeb> because everyone knows cycling was common pre-14th cen.
2012-10-02T00:48:23 -!- WorkerBuzz [~Buzz@75-167-154-20.bois.qwest.net] has joined ##stm32
2012-10-02T01:14:10 -!- mabl00 [~Thunderbi@2a01:198:71f:0:910b:62e2:1207:bd1e] has quit [Ping timeout: 246 seconds]
2012-10-02T01:56:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-02T01:57:38 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-02T02:00:11 -!- DaKu is now known as daku
2012-10-02T02:30:33 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-02T02:41:08 < dongs> sup dongs
2012-10-02T03:04:42 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-02T03:29:12 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-02T03:32:34 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-02T03:37:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-02T03:51:43 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-02T03:52:20 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-02T04:09:43 < dongs> http://www.rcgroups.com/forums/showpost.php?p=22865834&postcount=115
2012-10-02T04:09:43 < dongs> wtf
2012-10-02T04:12:55 < dongs> huh, f3discovery is in stock somewehre?
2012-10-02T04:21:54 < cjbaird> E14 UK have finally got their stock of the Stellaris Launchpads..
2012-10-02T04:23:32 < dongs> f3disco ship date keeps going back on digikey
2012-10-02T04:23:41 < dongs> thinking just ordering it and forgetting
2012-10-02T04:31:04 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-02T04:31:04 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-02T04:31:04 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-02T04:54:15 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-02T04:57:54 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-02T05:02:10 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-02T05:06:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-02T05:17:54 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 256 seconds]
2012-10-02T05:21:11 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 255 seconds]
2012-10-02T05:44:56 < dongs> commence chats
2012-10-02T05:45:23 < dongs> talkin to hackkitten again?
2012-10-02T05:46:33 < Bird|lappy> \o/ finally got my test jig board layout done
2012-10-02T05:46:59 < emeb_mac> w00t. post pix
2012-10-02T05:52:05 < Bird|lappy> http://ompldr.org/vZnB5OQ <-- pix :)
2012-10-02T05:52:22 < dongs> wtf
2012-10-02T05:52:29 < dongs> what opensauce pcb cad exports shit as svg
2012-10-02T05:53:16 < Bird|lappy> I just did a SVG plot for viewing purposes
2012-10-02T05:53:23 < Bird|lappy> obviously I'd be sending RS-274-X in for fab
2012-10-02T05:54:12 < dongs> ya but i mean how/what
2012-10-02T05:55:43 < Bird|lappy> btw that was cranked out by pcbnew from KiCad, it has SVG as one of its plot output options (others include HPGL, PS, and RS-274-X, but people's browsers don't support those formats very well :p)
2012-10-02T05:55:56 < dongs> ok.
2012-10-02T05:56:02 < dongs> suspected it'd haveto be opensauce
2012-10-02T05:58:20 < Bird|lappy> I wonder if there's an Eagle ULP for SVG plotting
2012-10-02T05:58:45 < emeb_mac> how do you like kicad?
2012-10-02T05:58:56 < Bird|lappy> not bad.
2012-10-02T05:59:04 < Bird|lappy> the ERC in eeschema is actually usable
2012-10-02T05:59:15 < emeb_mac> it's not bad
2012-10-02T05:59:22 < Bird|lappy> making schematic symbols is pretty easy too
2012-10-02T05:59:27 < emeb_mac> pcbnew routing is a PITA sometimes.
2012-10-02T05:59:58 < Bird|lappy> the big problem is that pcbnew itself is OK (quirks and all) but the default libraries are kinda untrustworthy for anything more than a simple test-jig board
2012-10-02T06:00:33  * emeb_mac usually ends up building a lot of his own symbols / footprints.
2012-10-02T06:00:56 < Bird|lappy> emeb_mac, have you used the CycleRecorder footprint generator before?
2012-10-02T06:00:56 < dongs> i never trust library shit
2012-10-02T06:01:09 < dongs> if I didnt make it myself chances are it will be fucked
2012-10-02T06:01:16 < emeb_mac> Bird|lappy: nope
2012-10-02T06:01:28 < emeb_mac> dongs: word
2012-10-02T06:01:58 < dongs> of course the closedsores pcb cad I use doesn't have problem easily making patterns or components.
2012-10-02T06:02:39 < dongs> i remember eagle library editor is fucking horrible
2012-10-02T06:02:42 < emeb_mac> pcbnew is not too bad that way.
2012-10-02T06:03:05 < Bird|lappy> of course making a footprint for a 256BGA fully manually is going to be hard no matter what you're using.  IIRC, Altium and suchnot have semi-auto generators as well
2012-10-02T06:03:09 < emeb_mac> gEDA/PCB is pretty awful though - always end up in a text editor w/ a calculator
2012-10-02T06:03:52 < emeb_mac> Bird|lappy: there are a fair number of automated footprint & symbol generation scripts for regular structures like BGA, etc.
2012-10-02T06:04:02 < Bird|lappy> yeah.
2012-10-02T06:04:12 < dongs> Bird|lappy: yes, i use pads generator in altium its great
2012-10-02T06:04:19 < dongs> infact thats probably hte best part about it.
2012-10-02T06:04:31 < emeb_mac> it's the funky specialized parts that have irregular structures that require hand work.
2012-10-02T06:04:33 < dongs> you can enter the usual a/b/c/etc dimensions from datasheet and it spits out compliant pads pattern
2012-10-02T06:04:41 < emeb_mac> nice
2012-10-02T06:04:59 < emeb_mac> buddy of mine bought Altium a few years ago
2012-10-02T06:05:16 < emeb_mac> never bothered to learn it though - still uses his old PCAD.
2012-10-02T06:05:32 < emeb_mac> waste of $$$
2012-10-02T06:06:29 < Bird|lappy> from what I know so far, if there's one commercial EDA suite to learn, it's Altium...I've had some firsthand horror-show experiences with Cadence's stuff
2012-10-02T06:06:33 < emeb_mac> Bird|lappy: what does that board do?
2012-10-02T06:06:39 < Bird|lappy> I mean, "Save" should not crash your program
2012-10-02T06:06:41 < Bird|lappy> no matter what
2012-10-02T06:06:55 < Bird|lappy> as to the board? it's a 2-in-1 test jig, really
2012-10-02T06:06:59 < dongs> i bought http://www.toporouter.com/ and never really used that shit for anything serious
2012-10-02T06:07:16 < Bird|lappy> one part is 550-750mV out with switchable Zout (1.5 or 3.6Ohms) to simulate a thermopile
2012-10-02T06:07:18 < dongs> i prefer m anual routing shit anyway... more fun
2012-10-02T06:07:24 < Bird|lappy> o/\o dongs
2012-10-02T06:07:24 < dongs> maybe i should do the led board with it haha.
2012-10-02T06:07:27 < dongs> just let it rip
2012-10-02T06:07:55 < emeb_mac> you got cash just laying around? that's not cheap!
2012-10-02T06:07:56 < Bird|lappy> I doubt an autorouter would have been able to cram that board onto 1 layer with the fills and all
2012-10-02T06:08:12 < Bird|lappy> emeb_mac, what isn't cheap?
2012-10-02T06:08:23 < emeb_mac> toporouter
2012-10-02T06:08:32 < Bird|lappy> the other part is 4.1Vout at oh 1mA or so to actually power the DUT
2012-10-02T06:08:44 < dongs> i got the 4l no pin limit version, its not that expensive compared to other eda shit.
2012-10-02T06:08:50 < dongs> i think base altium license is like 5k?
2012-10-02T06:09:03 < emeb_mac> sounds about right
2012-10-02T06:09:15 < dongs> my pcb cad was something liek 600 for unlimited version
2012-10-02T06:09:30 < dongs> shit i make money with i dont mind buying
2012-10-02T06:10:02 < emeb_mac> if it saves time/$$/hassle in the long run.
2012-10-02T06:10:26 < emeb_mac> that's why I pay $2k/yr maintenance on Matlab.
2012-10-02T06:10:36 < dongs> im to dum for matlab
2012-10-02T06:10:59 < Bird|lappy> :p
2012-10-02T06:11:01 < emeb_mac> doubt that - just don't need it for what you do.
2012-10-02T06:11:06  * Bird|lappy hasn't needed Matlab
2012-10-02T06:11:10 < emeb_mac> I use it every day
2012-10-02T06:11:25 < Bird|lappy> I do have a somewhat-oldish Maple CD kicking around
2012-10-02T06:11:35 < Bird|lappy> although Sage is my current CAS of choice
2012-10-02T06:11:41 < dongs> probably why you have answers to all those analog questions within 5 seconds
2012-10-02T06:11:57 < Bird|lappy> and my stats prof and books used R so thats what I have there
2012-10-02T06:12:01 < emeb_mac> heh - that's just misspent youth.
2012-10-02T06:13:47 < Bird|lappy> but yeah.  would never have been able to do that design in the first place without a LT3080
2012-10-02T06:19:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-02T06:19:41 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-02T06:20:30 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-02T06:20:36 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-02T06:26:43 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-02T06:42:31 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-02T07:21:13 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-10-02T08:09:56 < dongs> cmsis DSP library has pId controller?
2012-10-02T08:18:09 < dongs> i read that as 'chokes on a pile of dicks'
2012-10-02T08:18:46 < dongs> great minds think alke
2012-10-02T08:18:48 < dongs> alike.
2012-10-02T08:40:05 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 255 seconds]
2012-10-02T08:54:54 < cjbaird> I've just watched a documentary about the 'ingredients' behind the "e=mc^2" equation.. You would think it was directed by the same people who did the Silent Hill movie, by the way it tries to make women responsible for everything.. (and rewrites/excluses a number of things like Davy's wife.. and attributing Laplace's ideas to Émilie du Châtelet.. and so on)
2012-10-02T08:56:36 -!- mabl00 [~Thunderbi@2a01:198:71f:0:75d0:a0cd:ccf9:fd46] has joined ##stm32
2012-10-02T08:58:01 < dongs> WITH BALLS
2012-10-02T08:58:06 < cjbaird> Can't wait for someone to do an anti-cis edition of scientific history. :P
2012-10-02T08:59:06 < cjbaird> wikitroll: https://en.wikipedia.org/wiki/Cisgender
2012-10-02T09:01:36 < dongs> da fuq
2012-10-02T09:02:19 < dongs> you know the Wachowski brother who had a sex change and changed his name to Lana?
2012-10-02T09:02:22 < dongs> i just realised that Lana is Anal backwards.
2012-10-02T09:20:03 < ABLomas> ...
2012-10-02T09:20:18 < ABLomas> ,--8<-
2012-10-02T09:20:19 < ABLomas> |RA208808920CN
2012-10-02T09:20:19 < ABLomas> |RA632816851CN
2012-10-02T09:20:19 < ABLomas> |RA524547642CN
2012-10-02T09:20:19 < ABLomas> `-->8-
2012-10-02T09:20:23 < ABLomas> RB478438206HK oops
2012-10-02T09:20:26 < ABLomas> dammit
2012-10-02T09:20:30 < ABLomas> stupid window
2012-10-02T09:20:40 < ABLomas> sorry for that
2012-10-02T09:23:44 < dongs> sweet, stealing your packages
2012-10-02T09:26:58 < ABLomas> gj
2012-10-02T09:38:03 -!- mabl00 [~Thunderbi@2a01:198:71f:0:75d0:a0cd:ccf9:fd46] has left ##stm32 []
2012-10-02T09:40:20 < dongs> http://bcas.tv/paste/results/PVJhfn98.html lawl
2012-10-02T09:56:55 < jpa-> has anyone used the STM32F4 built-in USB bootloader?
2012-10-02T09:57:12 < zippe> jpa-: yes
2012-10-02T09:57:19 < dongs> i think i tried it once
2012-10-02T09:57:25 < dongs> its DFU shit isnt it?
2012-10-02T09:57:29 < jpa-> zippe: does it work? :)
2012-10-02T09:57:31 < jpa-> dfu yeah
2012-10-02T09:57:35 < dongs> yes, it worked
2012-10-02T09:57:41 < dongs> st has some DFU-util shit.
2012-10-02T09:57:46 < dongs> i used that, it uploaded.
2012-10-02T09:57:48 < dongs> seemed to work.
2012-10-02T09:57:49 < zippe> jpa-: for some value of work
2012-10-02T09:57:53 < jpa-> it doesn't even enable the pull-up for me
2012-10-02T09:58:05 < jpa-> so no enumeration, the host doesn't even try
2012-10-02T09:58:34 < dongs> did you check the boot pdf
2012-10-02T09:58:36 < dongs> to see which usb pins it c hecks
2012-10-02T09:58:40 < jpa-> yes
2012-10-02T09:59:21 < dongs> do you have all other conditions met
2012-10-02T09:59:28 < dongs> i.e. if it sees high signal on uart it wont try usb
2012-10-02T09:59:29 < dongs> etc
2012-10-02T09:59:53 < jpa-> hm, it just says that other pins must be at stable "high or low level"
2012-10-02T10:00:09 < jpa-> but yeah.. i have CAN2_RX floating :F
2012-10-02T10:00:13 < jpa-> maybe it is that
2012-10-02T10:01:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-02T10:05:50 < zyp> I assume you know it only supports the FS port, not the HS?
2012-10-02T10:12:39 -!- dekar [~dekar@212.255.31.84] has quit [Quit: This computer has gone to sleep]
2012-10-02T10:12:45 < dongs> "oops"
2012-10-02T10:15:43 < dongs> k time to finalize ledcrap schematic
2012-10-02T10:22:23 < jpa-> aha, i quess it helps if you don't hold NRST low :D
2012-10-02T10:22:35 < jpa-> didn't realize that my stlink was doing that
2012-10-02T10:23:02 < jpa-> now it enumerates
2012-10-02T11:00:29 -!- dekar [~dekar@dslb-088-069-135-103.pools.arcor-ip.net] has joined ##stm32
2012-10-02T11:00:33 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-02T11:14:01 < dongs> which is faster APB2 or APB1
2012-10-02T11:14:13 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-02T11:14:13 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-02T11:14:13 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-02T11:15:11 < dongs> ok 72 on  apb2
2012-10-02T11:16:27 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-02T11:24:26 -!- cjbaird_300baud_ [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-02T11:24:58  * dongs msgs cjbaird_300baud_ html for wikipedia assange article
2012-10-02T11:25:02 < dongs> that should ping you out
2012-10-02T11:28:18 < cjbaird_300baud_> peenode wouldn't let me have 'bitches' in the nick. :/
2012-10-02T11:28:46 < dongs> racist
2012-10-02T11:29:13 -!- dongs is now known as dongs_bitches
2012-10-02T11:29:16 -!- dongs_bitches is now known as dongs
2012-10-02T11:29:20 < dongs> i think your nick was just too long.
2012-10-02T11:32:29 < cjbaird_300baud_> lol..So much shit in that wiki article. "Mendax" was a fiction for that Underground book-- he used "The Mad Proffessor", later "proff". And his fellow undergraduate students were doing DARPA research? Really?
2012-10-02T11:33:19 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-02T11:34:17 < cjbaird_300baud_> A lot of the citations given are people just reciting 'Underground'..
2012-10-02T11:35:54 < cjbaird_300baud_> Of course, if I were to attempt to correct that, a hundred gormless fuckers would revert it in an instant..
2012-10-02T11:39:09 < cjbaird_300baud_> No mention of his conviction in 1996..
2012-10-02T11:40:16 < dongs>  he was "Australia's most famous ethical computer hacker",
2012-10-02T11:41:29 < cjbaird_300baud_> lol. He only stole credit cards of criminials, I suppose.
2012-10-02T11:45:21 < dongs> seems legit.
2012-10-02T11:52:39 < zippe> "most famous"?  Maybe now.  I'd say Daryl Tester was probably more famous earlier
2012-10-02T11:57:45 < cjbaird_300baud_> making noodles again..
2012-10-02T11:59:46 -!- cjbaird_300baud_ is now known as cjbaird_300baud
2012-10-02T12:12:35 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-02T12:17:32 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-02T12:17:35 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-02T12:17:48 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-10-02T12:20:34 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-02T12:20:36 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-02T12:23:21 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Ping timeout: 245 seconds]
2012-10-02T12:28:38 < cjbaird_300baud> Going to enjoy an hour or 4 using Kermit at 300 baud...
2012-10-02T12:28:45 -!- cjbaird_300baud [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Leaving]
2012-10-02T12:29:05 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-02T12:29:46 < dongs> http://i.imgur.com/qOjhH.png did i miss anything
2012-10-02T12:35:40 <+Steffanx> Yes, no yellow dongs
2012-10-02T12:35:50 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has joined ##stm32
2012-10-02T12:35:51 -!- HTT-Bird [~Birdz0r@ip68-107-237-136.om.om.cox.net] has quit [Changing host]
2012-10-02T12:35:51 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-10-02T12:36:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-02T12:45:44 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-02T12:46:26 < dongs> troulenceb_
2012-10-02T12:47:57 <+Steffanx> and Ardongso?
2012-10-02T12:50:42 < Laurenceb_> http://pastebin.com/Y04SrpzR
2012-10-02T12:51:59 < dongs> Laurenceb_: go review my schematic for any obvious fail http://i.imgur.com/qOjhH.png
2012-10-02T12:52:36 < Laurenceb_> too big
2012-10-02T12:53:13 < dongs> crashes ur lunix png viewer?
2012-10-02T12:53:33 < Laurenceb_> lol
2012-10-02T12:53:33 <+Steffanx> About which bridge is that story Laurenceb? Which one is "the bridge" ?
2012-10-02T12:53:42 < Laurenceb_> in cambridge
2012-10-02T12:54:01 <+Steffanx> There are more than one in cambridge
2012-10-02T12:54:06 < Laurenceb_> dongs: im not sure you need a pot divider on vbus sense
2012-10-02T12:54:21 < dongs> i dont if I stick it on a FT pin i guess.
2012-10-02T12:54:28 <+Steffanx> *is
2012-10-02T12:55:10 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-02T12:55:11 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-02T12:55:11 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-02T12:55:22 < Laurenceb_> is that a usb version of the zyp led thingy?
2012-10-02T12:55:42 < dongs> its a ledthigny emeb/me are dicking with
2012-10-02T12:55:49 < dongs> for chink rgb matrix
2012-10-02T12:55:55 < Laurenceb_> ah cool
2012-10-02T12:56:02 < dongs> free pcbs for all if it works
2012-10-02T12:56:06 < Laurenceb_> rest of schematic looks sane to me
2012-10-02T12:56:06 <+Steffanx> Too bad that story is not funny because it fails terribly Laurenceb_
2012-10-02T12:56:58 < dongs> when the IC5 (single buffer) OE goes high, output is supposed to be high-impedance, so I can drive CLK_OUT with SPI_CLK right?
2012-10-02T12:57:19 < Laurenceb_> as opposed to?
2012-10-02T12:57:23 < dongs> well
2012-10-02T12:57:28 < dongs> jsut leaving it low
2012-10-02T12:57:38 < dongs> then clk_in is connected to clk_out
2012-10-02T12:57:45 < Laurenceb_> oh
2012-10-02T12:57:47 < dongs> but 1st device in the chain clk_in is floaiting
2012-10-02T12:58:11 < dongs> so if i kept OE permanently low, i dunno if spi_clk can drive it while buffer is pulling it low
2012-10-02T13:01:19  * Laurenceb_ off
2012-10-02T13:01:39 < dongs> dongs
2012-10-02T13:05:53 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-02T13:09:46 < zyp> schematic looks pretty sane
2012-10-02T13:11:25 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-02T13:11:55 < zyp> how are you planning to use data_in/data_out?
2012-10-02T13:12:39 < dongs> for loading next in line modules wiht shit
2012-10-02T13:12:47 < dongs> 1st one (with usb connected) only sends stuff out.
2012-10-02T13:13:15 < dongs> emeb babbled something about clock through mode or someshit
2012-10-02T13:13:32 < dongs>  but I guess in passthrough mode out has data that was on in 8 clocks ago
2012-10-02T13:15:31 < zyp> but you would still be having a full mcu on each board, right?
2012-10-02T13:15:43 < dongs> yes
2012-10-02T13:15:53 < dongs> (lol)
2012-10-02T13:16:02 < dongs> but shit is cheap so..
2012-10-02T13:16:08 < zyp> then don't be dumb about board-to-board communication
2012-10-02T13:16:35 < dongs> do tell
2012-10-02T13:17:21 < zyp> I'd do bidirectional ports, so you can communicate both down and up the chain
2012-10-02T13:17:50 < dongs> zyp, onyl 2 SPIs on that chip and one is used for drivers
2012-10-02T13:17:58 < dongs> whats the point of comms up the chain?
2012-10-02T13:18:10 < dongs> idea is master (usb) block clocks out data for everyone and latches
2012-10-02T13:18:18 < dongs> then next one will forward on latch when it arrives.. etc
2012-10-02T13:18:48 < dongs> what situation needed that nth block wants to talk to 1st block?
2012-10-02T13:18:48 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-02T13:19:10 < zyp> then you could do stuff like query chain length
2012-10-02T13:19:57 < dongs> lemme know how i can do that with current hardware restrictions. :)
2012-10-02T13:20:00 < dongs> dinner time.
2012-10-02T13:55:09 -!- mabl00 [~Thunderbi@2a01:198:71f:0:75d0:a0cd:ccf9:fd46] has joined ##stm32
2012-10-02T13:55:59 < mabl00> Hi, is there a flag which indicates if the m3 core is currently executing an interrupt?
2012-10-02T13:58:38 < zyp> ipsr contains the number of the currently executing handler, or 0 if you are in thread mode
2012-10-02T14:00:08 < mabl00> zyp: thx very useful. I've looked for a flag and not a counter - so i did not see it
2012-10-02T14:01:09 < zyp> there is probably a flag in the control register as well, but I don't remember the specifics
2012-10-02T14:02:14 < zyp> hmm, no
2012-10-02T14:02:27 < zyp> then ipsr is probably the way to go
2012-10-02T14:02:43 < mabl00> zyp: ok, thx!
2012-10-02T14:03:05 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-02T14:03:34 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-02T14:05:05 < Thorn> there're also NVIC registers that can tell you even more
2012-10-02T14:05:10 < dongs> zyp, so any great ideas
2012-10-02T14:05:17 < dongs> UART or something?
2012-10-02T14:05:23 < dongs> cant use CAN.
2012-10-02T14:05:36 < dongs> i dont think bidirectional communication is that important for a one-time setup.
2012-10-02T14:05:40 < zyp> clearly I2C is the answer :p
2012-10-02T14:05:44 < dongs> especially since the host (pc/1st block) would
2012-10-02T14:05:45 < dongs> lol
2012-10-02T14:05:52 < dongs> need to know the block layout etc aynway
2012-10-02T14:06:09 < zyp> doing a spi chain is probably good enough
2012-10-02T14:06:38 < dongs> zyp, then i'd have to buffer all teh shit
2012-10-02T14:10:33 < zyp> then the answer is obviously to throw an usb hub chip on the board, and just run usb to the next board :p
2012-10-02T14:10:52 < dongs> ahahha,.
2012-10-02T14:11:03 < dongs> zyp, hubs have a limit of like 7 devices or someshit.
2012-10-02T14:11:06 < dongs> i mean
2012-10-02T14:11:07 < zyp> sure
2012-10-02T14:11:08 < dongs> hub behind hub.
2012-10-02T14:12:01 < dongs> zyp, any long-distance/not fail 1wire shit that I can use as aux channel then
2012-10-02T14:12:08 < zyp> :p
2012-10-02T14:13:52 < dongs> (multidrop too)
2012-10-02T14:14:45 < dongs> it would be cool to hook shit together and power it up and they go through a count sequence each one lighting up its number in the chain
2012-10-02T14:15:41 < Thorn> dongs: http://www.nxp.com/documents/leaflet/75017050.pdf
2012-10-02T14:16:03 < dongs> Thorn: yes, im gonna put another cortex mcu on that shit
2012-10-02T14:16:07 < dongs> just for initial count down.
2012-10-02T14:16:33 < zyp> oh, that's nice
2012-10-02T14:18:02 < Thorn> also I got async libusb i/o working on windows with the help of the libusb developer
2012-10-02T14:18:30 < zyp> which libusb developer?
2012-10-02T14:19:01 < dongs> probably the only one
2012-10-02T14:19:56 < Thorn> Peter Stuge (CareBear in #libusb)
2012-10-02T14:20:11 < Thorn> 1.0.9 is buggy but git head works
2012-10-02T14:21:04 < Thorn> integrating with qt required a separate thread
2012-10-02T14:21:08 < jpa-> ah, explains why he is always trying to convince me that usb is easy and everyone should make their own usb interface instead of CDC or FTDI usart
2012-10-02T14:28:01 < Thorn> I find usb requires lots more code on both sides than say usart/serial port comms lol
2012-10-02T14:28:45 < jpa-> indeed
2012-10-02T14:28:47 < dongs> how is nonscrolling touchwheel done on jewpod? capsense?
2012-10-02T14:29:33  * dongs just found my old ipod nano from 2006
2012-10-02T14:29:48 < dongs> the only apple product I own,, and didnt evne pay for it (got free for buying some raritan warez)
2012-10-02T14:33:06 < Thorn> how do you transfer a megabyte over a bulk endpoint? does it require multiple application-initiated transfers?
2012-10-02T14:33:29 < dongs> just send it
2012-10-02T14:34:42 < Thorn> looks like it does
2012-10-02T14:35:48 < zyp> your question is unclear
2012-10-02T14:36:21 < zyp> a transfer larger than max packet size has to be splitted into packets of maximum size and sent one by one
2012-10-02T14:37:24 < Thorn> but afair a bulk transfer can consist of several transactions
2012-10-02T14:37:53 < zyp> exactly
2012-10-02T14:38:19 < zyp> when I said packets of maximum size, I meant transactions of maximum size
2012-10-02T14:38:29 < dongs> http://pics.vdott.com/us/13622_kiishop_1.jpg
2012-10-02T14:39:45 < dongs> https://www.youtube.com/watch?v=kgLPJnGGjzU this had to appear sooner or later.
2012-10-02T14:41:37 < Thorn> so at what level the data is split into transactions? is there a limit to the size of the transfer?
2012-10-02T14:41:53 < dongs> my japtv shit sends like 38k blocks
2012-10-02T14:42:01 < Thorn> and does the device know when a transfer ends and the next one begins?
2012-10-02T14:42:07 < dongs> er no, 48k... 188*256bytes
2012-10-02T14:43:01 < Thorn> "The USB 2.0 specification doesnÂ’t define a protocol for indicating the number of data bytes in a bulk transfer."
2012-10-02T14:43:31 < zyp> Thorn, the last transaction of a transfer is non-maxpacket sized
2012-10-02T14:43:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-02T14:44:06 < Thorn> at what level is this detected? stack or application?
2012-10-02T14:44:14 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-02T14:44:27 < zyp> stack, as far as I know
2012-10-02T14:45:40 < zyp> if you have a 64B endpoint and you are doing a 160B transfer, you'll get packets like this: 64, 64, 32
2012-10-02T14:45:57 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-02T14:46:17 < zyp> with a 128B transfer the transfer will be terminated by a zero-length packet to indicate the end
2012-10-02T14:46:20 < zyp> 64, 64, 0
2012-10-02T14:46:40 < Laurenceb> wtf
2012-10-02T14:46:40 < zyp> unless I've misunderstood something, that is.
2012-10-02T14:46:43 < Laurenceb>  p TIM2->CNT
2012-10-02T14:46:43 < Laurenceb> $1 = 2554464731
2012-10-02T14:46:49 < Laurenceb> ^how is that possible?
2012-10-02T14:47:08 < zyp> 32-bit timer? :p
2012-10-02T14:49:23 < jpa-> yeah, TIM2 is 32-bit
2012-10-02T14:49:42 < zyp> then I don't see the problem
2012-10-02T14:49:51 < jpa-> me neither
2012-10-02T14:50:49 <+Steffanx> He means 2554464731 can't be divided by 2
2012-10-02T14:51:26 < Laurenceb> CNT[15:0]: ounter value
2012-10-02T14:51:40 < jpa-> "16-bit (TIM3 and TIM4) or 32-bit (TIM2 and TIM5)
2012-10-02T14:51:51 < jpa-> assuming F4
2012-10-02T14:53:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-02T14:54:01 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-02T14:54:30 < Laurenceb> oh
2012-10-02T14:56:40 -!- phantoxe [~destroy@95.92.89.24] has joined ##stm32
2012-10-02T14:58:54 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-02T15:05:18 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-02T15:09:01 < Thorn> more lulz in the ST stack.
2012-10-02T15:10:13 < Thorn> if the host does an OUT xfer and you didn't call DCD_EP_PrepareRx() beforehand, your MCU crashes because the stack tries to copy received data to a random memory location
2012-10-02T15:11:00 < Thorn> well, *hopefully* crashes
2012-10-02T15:15:26 < Thorn> no attempt is made to e.g. check for a null pointer and STALL the xfer
2012-10-02T15:15:40 < Laurenceb> ARG WTF
2012-10-02T15:15:59 < zyp> nice
2012-10-02T15:16:01 < Laurenceb> https://github.com/Laurenceb/Chibi-Spectro/blob/master/utilities/Timer.c
2012-10-02T15:16:13 < Laurenceb> time2 starts up in 32bit mode
2012-10-02T15:16:28 < Laurenceb> and overflows past the arr value
2012-10-02T15:17:11 < Laurenceb> hmm
2012-10-02T15:17:13 < Laurenceb> 	TIM2->CR1|=TIM_CR1_ARPE;	//Preload enable
2012-10-02T15:17:28 < Laurenceb> ^meant the timer has to overflow once?
2012-10-02T15:18:31 < jpa-> yeah, but you can also cause update event manually
2012-10-02T15:19:28 < Laurenceb> or just turn off ARPE?
2012-10-02T15:19:38 < jpa-> yeah
2012-10-02T15:19:42 < Laurenceb> i only set the ARR during init, before timer enabled
2012-10-02T15:20:40 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-02T15:24:53 < dongs> Laurenceb: how do i un-8bit this shit. http://bcas.tv/paste/results/nkAIyD83.html
2012-10-02T15:25:34 < dongs> needs sin/cos somewehre in there.
2012-10-02T15:26:25 < Laurenceb> hmm
2012-10-02T15:26:34 < Laurenceb> im not in the mood for maths
2012-10-02T15:26:40 < Laurenceb> look up rotation matrices
2012-10-02T15:26:44 <+Steffanx> heh
2012-10-02T15:26:57 < dongs> thats 2 hard 4 me
2012-10-02T15:28:45 < mabl00> dongs: this should be straight forward linear algebra
2012-10-02T15:28:58 < mabl00> keyword is rotation matrices
2012-10-02T15:29:45 < mabl00> wikipedia has the rotation matrix around a unit vector
2012-10-02T15:29:53 < dongs> mabl00: i failed calculus 101 twice
2012-10-02T15:29:59 < dongs> sorry
2012-10-02T15:30:13 < mabl00> unfortunately only the german one
2012-10-02T15:30:26 < mabl00> http://de.wikipedia.org/wiki/Rotationsmatrix
2012-10-02T15:30:26 < mabl00> Search for "Drehung um eine Ursprungsgerade"
2012-10-02T15:30:52 < dongs> looks good but doesnt look like C
2012-10-02T15:31:00 < mabl00> ^^ true
2012-10-02T15:31:02 < Thorn> looks like synopsys controller has special "transfer complete" status & interrupt, this is what the stack uses
2012-10-02T15:31:30 < mabl00> dongs: maybe it's even easier than what I just gave you
2012-10-02T15:31:43 < mabl00> i assume delta is a pointer to euler angles right?
2012-10-02T15:32:00 < dongs> delta is rotation angle
2012-10-02T15:32:14 < dongs> yeah
2012-10-02T15:33:54 < mabl00> ok, so what you basically need to do is to calculate "RotationMatrix"*\vec{v}
2012-10-02T15:34:14 < mabl00> you will have o find the rotatio matrix first
2012-10-02T15:34:33 < mabl00> trouble is tahat there are many types of euler anges
2012-10-02T15:37:33 < claude> Thorn , the stm32 usb controler is from synopsis?
2012-10-02T15:37:41 -!- mabl00 [~Thunderbi@2a01:198:71f:0:75d0:a0cd:ccf9:fd46] has quit [Quit: mabl00]
2012-10-02T15:37:50 < claude> any chance nxp uses the same ip?
2012-10-02T15:37:52 < Thorn> yes, synopsys dwc
2012-10-02T15:38:09 < Thorn> I don't think so. energy micro apparently does
2012-10-02T15:38:10 < claude> asking because of lufa ...
2012-10-02T15:39:24 < claude> hm okay , just used lufa the first time some weeks ago . it's so much easier to use than the usblib from st ^^
2012-10-02T15:40:33 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-02T15:40:59 < claude> so the host part of the otg controler is like ohci?
2012-10-02T15:41:35 < Thorn> does avr use synopsys usb cores too?
2012-10-02T15:42:04 < zyp> I just had a look in the lpc17xx reference manual, it's not using dwc_otg
2012-10-02T15:42:27 < Thorn> they use something custom
2012-10-02T15:42:31 < Thorn> afaik
2012-10-02T15:48:57 < Laurenceb> ive got 4 timers running on F4 now...
2012-10-02T15:49:06 < Laurenceb> but tim10 isnt slaving tim9   :(
2012-10-02T15:49:12 < dongs> slaying
2012-10-02T15:49:14 < Laurenceb> its supposed to be able to
2012-10-02T15:49:30 < Laurenceb>  p TIM10->SMCR
2012-10-02T15:49:30 < Laurenceb> $27 = 0
2012-10-02T15:49:32 < Laurenceb> thats bad
2012-10-02T15:51:54 < dongs> <Laurenceb> I forgot to power it
2012-10-02T15:52:18 < dongs> or "powered" up APB2 peripheral by APB1PeriphClockCmd
2012-10-02T15:57:37 < Laurenceb> page 437 on f4 ref manual
2012-10-02T15:57:59 < Laurenceb> you can slave TIM9 off TIM10
2012-10-02T15:59:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-02T15:59:57 < Laurenceb> only you cant...
2012-10-02T16:00:02 < Laurenceb> or i cant work out how to
2012-10-02T16:00:13 < dongs> https://forum.sparkfun.com/viewtopic.php?f=14&t=33621
2012-10-02T16:01:38 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-02T16:10:25 < Laurenceb> http://www.youtube.com/watch?v=L2o9eBl_Gzw
2012-10-02T16:10:27 < Laurenceb> WTF
2012-10-02T16:12:55 < dongs> Laurenceb: livestream from ISS?
2012-10-02T16:14:21 <+Steffanx> live?
2012-10-02T16:26:20 < Thorn> claude: to be more correct, f105, f107, f2 and f4 use synopsys usb cores. f102, f103, f3 use a simpler proprietary device-only controller
2012-10-02T16:28:43 < claude> Thorn, thanks good to know
2012-10-02T16:35:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-02T16:36:16 < karlp> f3 doens't have hte otg core?
2012-10-02T16:36:19 < karlp> interesting.
2012-10-02T16:40:01 < Thorn> and synopsys controllers keep track of received vs. expected number of bytes for all transfers (DIEPTSIZ/DOEPTSIZ registers)
2012-10-02T16:40:35 < Thorn> quiz: there's a field called STUPCNT in those. what does it stand for?
2012-10-02T16:40:47 < claude> stupid count? ;)
2012-10-02T16:41:02 < claude> cunt even ..
2012-10-02T16:41:15 < Thorn> SETUP packet count.
2012-10-02T16:41:37 < Thorn> obviosuy enough
2012-10-02T16:41:54 -!- dekar [~dekar@dslb-088-069-135-103.pools.arcor-ip.net] has quit [Ping timeout: 260 seconds]
2012-10-02T16:42:51 < Thorn> and XFERSIZ is 19 bits long, so transfers are 512Kb max?
2012-10-02T16:43:01 < Thorn> that's much larger than internal SRAM anyway
2012-10-02T16:43:50 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-02T16:48:20 < Thorn> Values are in terms of 32-bit words: 0x0: Endpoint TxFIFO is full 0x1: 1 word available 0x2: 2 words available 0xn: n words available Others: Reserved
2012-10-02T16:49:04 < Thorn> what 'others'
2012-10-02T16:49:36 < dongs> lol
2012-10-02T16:49:40 -!- WorkerBuzz [~Buzz@75-167-154-20.bois.qwest.net] has quit [Ping timeout: 248 seconds]
2012-10-02T16:50:32 < Laurenceb> http://static.fjcdn.com/pictures/Vintage+Rape+Face_5b5e75_4094968.jpg
2012-10-02T16:50:56 <+Steffanx> Please give a warning next time Laurenceb
2012-10-02T16:51:13 < dongs> what else do you expect hosted on a site like 'fjcdn'
2012-10-02T16:51:20 < BrainDamage> isn't his nick a warning enough?
2012-10-02T16:51:29 < dongs> and jpeg name as "vintage rape face"
2012-10-02T16:51:39 < BrainDamage> like, laurenceb or dongs' link -> 90% chance of nsfw
2012-10-02T16:57:13 < cjbaird> I need to work on my percentage..
2012-10-02T17:03:36 < karlp> dongs was that schematic in http://i.imgur.com/qOjhH.png from diptrace? (is that what you use?) I like the different colours for different signal groups
2012-10-02T17:04:04 < dongs> yea dicktrace
2012-10-02T17:04:25 < dongs> right, i use net color stuff a lot, in layout as well
2012-10-02T17:04:34 < dongs> so I can quickly see what stuff should be going wehre.
2012-10-02T17:04:54 < dongs> oh, emeb_mac ^ beep that schematic might be relevant for you
2012-10-02T17:05:58 < emeb_mac> that's the 8x8 driver? I'll look...
2012-10-02T17:06:03 < dongs> yes
2012-10-02T17:11:03 < emeb_mac> I assume the todo list is for pins that can be hooked up anywhere during routing?
2012-10-02T17:11:06 -!- dekar [~dekar@88.69.135.103] has joined ##stm32
2012-10-02T17:11:08 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-02T17:15:32 < dongs> yes
2012-10-02T17:24:45 < Tectu> anyone ever did a DDS with an STM32?
2012-10-02T17:25:09 < jpa-> DDS of what?
2012-10-02T17:25:22 < jpa-> i have put a sine wave out of STM32 DAC, does that count?
2012-10-02T17:26:10 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-02T17:26:11 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-02T17:26:11 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-02T17:29:02 < Laurenceb> lol
2012-10-02T17:29:25  * Laurenceb is currently trying to generate orthogonal carriers with pwm
2012-10-02T17:30:54 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-02T17:45:29 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-02T17:48:27 < Laurenceb> http://www.youtube.com/watch?v=5E5u5xO8t68
2012-10-02T17:48:59 <+Steffanx> At least i understand why your project isn't finished Laurenceb
2012-10-02T17:49:17 <+Steffanx> You watch too much crap /  browse to often on the web
2012-10-02T17:50:01 < Thorn> so one 256-byte transfer from host turns into 8 32-byte transfers on the device (this is the buffer size). max packet size on the EP is 8 bytes.
2012-10-02T17:56:08 < karlp> Thorn: those nxp chips with the onboard CAN you linked to, what makes that different to the stm32's with can?
2012-10-02T17:56:33 < karlp> or is this a USB vs USB w/ PHY type thing again?
2012-10-02T17:56:43 < Thorn> do any stm32 chips have can PHY onboard?
2012-10-02T17:58:17 < karlp> I guess that answers my questions :)
2012-10-02T17:58:32 < karlp> I didn't even know that you needed a PHY as well
2012-10-02T17:59:07 < Thorn> same as with any interface, like rs-485, rs-232 etc.
2012-10-02T17:59:25 < dongs> karlp: yeah, they're saving you a $0.10 can rx/tx chip
2012-10-02T17:59:40 < dongs> and some board space
2012-10-02T17:59:43 < dongs> since those are usually so8
2012-10-02T17:59:52 < Thorn> $0.10?
2012-10-02T17:59:57 <+Steffanx> 0.11
2012-10-02T18:00:05 < Thorn> at least $0.12
2012-10-02T18:00:44 < karlp> well, not _every_ interface, spi and i2c are ok, which is why I was asking.
2012-10-02T18:01:23 < Thorn> those aren't designed to leave your device
2012-10-02T18:02:26 < Thorn> actually CAN has been quite expensive until relatively recently as I understand
2012-10-02T18:03:33 < zyp> 22:49:46 <+izua> and if some guy that branched long time ago wishes to update, it's his job to solve conflicts and not the developers?
2012-10-02T18:03:40 < zyp> uh
2012-10-02T18:03:58 < zyp> I didn't mean to paste that.
2012-10-02T18:04:03 <+Steffanx> Yes you did
2012-10-02T18:04:54 < zyp> I'm not even sure why it was on my clipboard
2012-10-02T18:05:15 < zyp> probably been there since last week
2012-10-02T18:07:25 < Thorn> so apparently you can collect data from a dwc FIFO either per packet or per transaction
2012-10-02T18:07:43 < Thorn> or per transfer even
2012-10-02T18:08:07 < zyp> sure
2012-10-02T18:08:34 < zyp> as long as the transfer fit into the fifo
2012-10-02T18:09:17 < zyp> you configure the core to interrupt you after receiving «at least this much or a complete transfer»
2012-10-02T18:09:39 < Thorn> and max transfer is half a megabyte
2012-10-02T18:10:10 < zyp> how so?
2012-10-02T18:10:51 < Thorn> XFERSIZ in DOEPTSIZ is 19 bits long
2012-10-02T18:12:42 < zyp> that's irrelevant
2012-10-02T18:13:11 < zyp> This field contains the transfer size in bytes for the current endpoint. The core only interrupts
2012-10-02T18:13:14 < zyp> the application after it has exhausted the transfer size amount of data. The transfer size can be set to the maximum packet size of the endpoint, to be interrupted at the end of each packet.
2012-10-02T18:13:40 -!- dekar [~dekar@88.69.135.103] has quit [Ping timeout: 246 seconds]
2012-10-02T18:14:03 < zyp> it just controls how much the core will receive before you get the interrupt
2012-10-02T18:18:13 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-02T18:21:29 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-10-02T18:21:55 < Laurenceb> http://hackaday.com/2012/10/02/whatever-a-fear-of-fingers-is-this-is-it/#comments
2012-10-02T18:21:59 < Laurenceb> lol @ comments
2012-10-02T18:23:24 < dongs> looks like a perfect anal invasion device
2012-10-02T18:24:26 <+Steffanx> And i'm sure they'll make a TV show about in Japan
2012-10-02T18:35:45 < Laurenceb> something for dongs  http://www.b3tards.com/u/27480baa08ab789588d7/bat.jpg
2012-10-02T18:36:46 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-02T18:41:47 < Thorn> well at least the ST stack uses that interrupt to signal when transfer is complete
2012-10-02T18:48:47 < dongs> Laurenceb: i dont get it
2012-10-02T18:48:54 < dongs> Laurenceb: is that something like "captain obvious"
2012-10-02T18:55:30 < dongs> cool, digikey accepted my RMA for some junk
2012-10-02T18:56:05 < dongs> i get liek $50 back
2012-10-02T19:00:09 -!- barthess [~barthess@178.154.61.18] has joined ##stm32
2012-10-02T19:02:16 -!- dekar [~dekar@212.255.31.84] has joined ##stm32
2012-10-02T19:02:17 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-02T19:06:52 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 246 seconds]
2012-10-02T19:08:03 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-02T19:10:38 < dongs> http://bcas.tv/paste/results/X2UVh640.html defucked it.
2012-10-02T19:10:57 <+Steffanx> You mean, fucked it
2012-10-02T19:11:21 < emeb> yay trigonometry!
2012-10-02T19:11:36 <+Steffanx> The approximation wasn't accurante enough dong?
2012-10-02T19:12:15 < emeb> sample period too long & the small angle approximation no longer holds?
2012-10-02T19:12:46 < dongs> ive got cycles to burn
2012-10-02T19:12:55 < dongs> so no sense fucking with approximations
2012-10-02T19:13:06 <+Steffanx> emeb true, but don't make that period too long :P
2012-10-02T19:13:44 < emeb> gets laggy.
2012-10-02T19:15:25 < emeb> yeah - if your sample rate is so low that the delta angles get into the range where you need real trig then you've got more than one problem.
2012-10-02T19:15:49 < Thorn> and of course there's no API to find out how much data has been transferred.
2012-10-02T19:15:53 <+Steffanx> And regexes and you're screwed :P
2012-10-02T19:15:56 <+Steffanx> *add
2012-10-02T19:16:00 < Thorn> uint32_t received = ((USB_OTG_CORE_HANDLE*)pdev)->dev.out_ep[epnum].xfer_count;
2012-10-02T19:16:45 <+Steffanx> Thorn is working on his own stack?
2012-10-02T19:17:04 < Thorn> worse. on the ST stack
2012-10-02T19:17:16 <+Steffanx> oops
2012-10-02T19:17:29 < emeb> might as well give up now. :)
2012-10-02T19:39:05  * Laurenceb has F4 timers working at last
2012-10-02T19:39:56 < Laurenceb> 5 orthogonal channels
2012-10-02T19:44:46 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-02T19:46:36 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-gcomyincbmqglsww] has quit [Ping timeout: 256 seconds]
2012-10-02T19:47:28 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ldsegujyfdswtxyz] has joined ##stm32
2012-10-02T19:48:47 -!- phantoxe [~destroy@95.92.89.24] has quit [Ping timeout: 246 seconds]
2012-10-02T19:48:47 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 246 seconds]
2012-10-02T19:48:48 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-02T19:50:52 <+Steffanx> incl. functional chibios ?
2012-10-02T19:51:10 < dongs> you wish
2012-10-02T19:51:30 <+Steffanx> You really want to know my wishes?
2012-10-02T19:51:51 < emeb> please, no.
2012-10-02T19:52:23 < dongs> emeb: anything else i missed in the schema?
2012-10-02T19:52:36 < dongs> or should I throw that shit into toporouter tomorrwo ^_^
2012-10-02T19:53:09 < dongs> any more jumper-configurable shit?
2012-10-02T19:53:21 < dongs> i did the gck and latchthrough stuff
2012-10-02T19:53:33 < emeb> dongs: other than the todo stuff I didn't see anything.
2012-10-02T19:54:24 < dongs> mmkay
2012-10-02T19:54:27 < dongs> oh yeah.. i was gonna ask
2012-10-02T19:54:42 < dongs> components on the side facing into the led case or on back side
2012-10-02T19:54:56 < dongs> iwas thinking back side cuz of usb cnnector but
2012-10-02T19:55:02 < emeb> yeah.
2012-10-02T19:55:11 < zyp> back, more airflow for parts that heat up
2012-10-02T19:55:19 < dongs> oh hm thats a good point too.
2012-10-02T19:55:22 < dongs> drivers might get warmey
2012-10-02T19:55:28 < emeb> mmm... toasty.
2012-10-02T19:58:24 < dongs> so i gotta figure out what the pattern for that chinaled looks like from the back of the board
2012-10-02T19:58:33 < dongs> cause i got a 50/50 chance il lfuck it up and invert it
2012-10-02T19:59:11 < zyp> what, can't your pcb software place components on either side and flip the pattern accordingly?
2012-10-02T19:59:24 < dongs> i didnt use a pattern i used 2 16pin headers
2012-10-02T19:59:30 < dongs> i can group and flip them sure
2012-10-02T19:59:38 < dongs> but i'll probably still fuck shit up :p
2012-10-02T20:02:52 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 252 seconds]
2012-10-02T20:09:47 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-10-02T20:25:50 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-02T20:30:20 < jpa-> shouldn't SCB->AIRCR software reset completely reset all peripherals, like power cycling?
2012-10-02T20:32:53 < Thorn> for one thing it doesn't reset SRAM afaik
2012-10-02T20:33:34 < zyp> neither do powercycling
2012-10-02T20:34:07 < Thorn> depends on how long it stays without power
2012-10-02T20:34:23 < jpa-> i'm having I2C problems after powerdown, but even software reset doesn't fix it
2012-10-02T20:34:28 < jpa-> but power cycling does
2012-10-02T20:34:48 < zyp> Thorn, you are supposed to treat sram contents at reset as undefined
2012-10-02T20:35:56 < zyp> jpa-, are you sure that's not a device keeping the bus busy by pulling down one of the lines?
2012-10-02T20:36:25 < zyp> so powercycling helps because it resets the slave
2012-10-02T20:37:23 < jpa-> yeah... apparently it is the slave after all
2012-10-02T20:37:41 < jpa-> LSM303DLHC, i want to award this chip for the worst datasheet ever
2012-10-02T20:37:45 < jpa-> trying to use the interrupt pin
2012-10-02T20:39:12 <+Steffanx> ST & I2c == trouble :P
2012-10-02T20:47:14 -!- barthess [~barthess@178.154.61.18] has quit [Quit: Leaving.]
2012-10-02T20:47:51 -!- barthess [~barthess@178.154.61.18] has joined ##stm32
2012-10-02T20:50:18 -!- daku is now known as DaKu
2012-10-02T20:53:51 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Remote host closed the connection]
2012-10-02T20:55:10 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-02T20:58:14 <+dekar> Steffanx, https://dl.dropbox.com/u/28467113/rtlsdr/hf_mod_transformer_H16105DF.jpg
2012-10-02T20:58:23 < jpa-> hmph, wtf
2012-10-02T20:58:41 <+Steffanx> dekar goes RF
2012-10-02T20:58:42 < jpa-> as soon as an interrupt occurs, LSM303DLHC starts holding the SCL low
2012-10-02T20:58:51 < jpa-> but only way to reset interrupt is through I2C
2012-10-02T20:58:59 <+dekar> Steffanx, HF, I receive stuff in weird languages - a ton of it
2012-10-02T20:59:37 <+dekar> took that transformer from some ADSL modem :P
2012-10-02T20:59:42 <+Steffanx> You were the one with that fancy antenna outside?
2012-10-02T20:59:45 < BrainDamage> weird languages = english
2012-10-02T20:59:46 <+Steffanx> or was that someone else?
2012-10-02T21:01:17 <+dekar> I built a new one for 20 meter band
2012-10-02T21:01:21 <+dekar> https://dl.dropbox.com/u/28467113/rtlsdr/hf_mod_antenna.jpg
2012-10-02T21:01:42 <+dekar> litz wire with sma connector :D
2012-10-02T21:02:00 < BrainDamage> what's the end connector? smb?
2012-10-02T21:02:18 <+dekar> sma to mcx adaptor
2012-10-02T21:03:38 < prog> beautiful noise trap:D
2012-10-02T21:04:45 <+dekar> noise trap? I use that transformer to get a differential input for my ADC
2012-10-02T21:06:12 < emeb> dekar: RTL SDR receiver w/ downconverter removed?
2012-10-02T21:06:53 <+dekar> yeah
2012-10-02T21:07:07 <+dekar> does 0-28mhz now
2012-10-02T21:07:19 < emeb> cool - I've heard that was possible.
2012-10-02T21:07:48 < emeb> amazing what folks have done with those even w/o documentation.
2012-10-02T21:08:14 <+Steffanx> trial and error?
2012-10-02T21:08:42 < emeb> or something. I think part of it has been studying FOSS driver code that was developed w/ docs.
2012-10-02T21:09:23 <+dekar> I think the datasheets leaked
2012-10-02T21:10:02 < emeb> Some. I grabbed copies of the E4000 a few weeks back, but those parts seem to be on the way out now that Elonics is dead.
2012-10-02T21:10:07 <+Steffanx> So now you can receive stuff, what's the plan dekar ?
2012-10-02T21:11:17 <+dekar> emeb, I have this spare e4000 left: https://dl.dropbox.com/u/28467113/rtlsdr/e4000.jpg
2012-10-02T21:11:20 < Thorn> on the subject of USB sniffing: http://www.usbmadesimple.co.uk/ums_3.htm
2012-10-02T21:11:43 < emeb> dekar: save it - might be worth something someday. :)
2012-10-02T21:11:48 <+dekar> Steffanx, I guess I'll receive some stuff
2012-10-02T21:11:53 <+Steffanx> 27MHz / citizen band is pretty dead, so that's useless
2012-10-02T21:11:59 <+dekar> emeb, there actually is a better tuner than the e4000 now
2012-10-02T21:12:02 < Thorn> you may be able to see packets from host but not from other devices
2012-10-02T21:12:29 < emeb> dekar: yeah. R820T?
2012-10-02T21:12:34 <+dekar> I think so yeah
2012-10-02T21:12:49 < emeb> Looks like they got the hooks for that going reliably now.
2012-10-02T21:13:15 < BrainDamage> I really liked the E4000 arch, zero-if ftw
2012-10-02T21:13:28 < BrainDamage> R820 is supposedly heterodyne
2012-10-02T21:13:39 < emeb> dekar: I thing prog is involved with some of the client software development.
2012-10-02T21:14:02 < emeb> wants to get an RTL running on an STM32F4
2012-10-02T21:14:20 < emeb> Might be difficult w/o a HS phy.
2012-10-02T21:14:24 < BrainDamage> both dekar and prog and me are both here and in ##rtlsdr
2012-10-02T21:14:33 < BrainDamage> why would you want to use an rtl chip?
2012-10-02T21:14:39 < BrainDamage> just use the stm32 adc
2012-10-02T21:14:50 < BrainDamage> esp when F3 are out
2012-10-02T21:15:02 <+Steffanx> Go for it BrainDamage
2012-10-02T21:15:05 < emeb> stm32 adc can't sample @ 32MHz?
2012-10-02T21:15:19 < BrainDamage> the F3 can go pretty close to it
2012-10-02T21:15:25 < prog> I'd use it for VHF/UHF
2012-10-02T21:16:39 < emeb> Instead of using an RTL over USB I'd get a fast ADC & FPGA, use the FSMC & do the initial tuning & downsample in HW.
2012-10-02T21:17:06 < BrainDamage> FSMC = ?
2012-10-02T21:17:18 < zyp> external memory interface
2012-10-02T21:17:24 < emeb> yep
2012-10-02T21:17:25 <+Steffanx> flexible s tatic memory controler
2012-10-02T21:17:40 <+Steffanx> +l - space
2012-10-02T21:17:45 < BrainDamage> ah thanks, I'm always awful with akronyms
2012-10-02T21:17:46 < jpa-> BrainDamage: close to 32MHz? huh, isn't F3 ADC limited at 1MSps?
2012-10-02T21:18:07 < emeb> for very loose definitions of "close" :)
2012-10-02T21:18:20 <+Steffanx> 5 jpa- ..
2012-10-02T21:18:21 < BrainDamage> the rtl chip will sustain 2.4MSps @ 8 bit after fir and cordic
2012-10-02T21:18:40 < BrainDamage> the stm32 F3 can do 5 @ 12 or 16 bit, forgot
2012-10-02T21:18:45 < prog> 8bit cordic?
2012-10-02T21:18:45 < emeb> BrainDamage: yeah - but it samples faster & does a DDC. I'd do that in an FPGA.
2012-10-02T21:18:49 <+Steffanx> 5 @ 12
2012-10-02T21:18:53 < prog> it's using a mere lut
2012-10-02T21:19:28 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-112.pptp.stw-bonn.de] has joined ##stm32
2012-10-02T21:19:44 < emeb> lut / cordic doesn't matter. Either works for coarse tuning.
2012-10-02T21:21:19 < emeb> but of course doing the frontend in an FPGA + ADC costs a lot more than an RTL chip.
2012-10-02T21:21:56 < emeb> so you're trading off ease of software dev for ease of hardware dev & cost.
2012-10-02T21:22:21 < BrainDamage> the question is what's considerable acceptable, for instance fpga could get a dac and you'd be able to tx too with small additional cost
2012-10-02T21:22:36 < emeb> true - I've got that working already.
2012-10-02T21:23:06 < BrainDamage> then again, there seems to be a boom of sdr boards on recent times
2012-10-02T21:23:51 < emeb> the hardware is getting sophisticated & cheap enough to be worthwhile and the host computing power can keep up.
2012-10-02T21:24:10 < emeb> 5 yrs ago we knew how to do this, but all levels of hardware were too $$$
2012-10-02T21:28:21 < Erlkoenig> http://minecraft.niklas.2g2s.de/software/make.rb ... if anyone is interested
2012-10-02T21:28:34 < Erlkoenig> (don't mind the domain)
2012-10-02T21:44:01 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-02T21:46:46 < zyp> some nice ideas there
2012-10-02T21:47:26 < Erlkoenig> really? :3
2012-10-02T21:47:40 < zyp> but you are setting yourself up for a lot of work by writing it from scratch rather than extending an existing framework :p
2012-10-02T21:48:31 < Erlkoenig> hm, the framework itself is not so much work...
2012-10-02T21:48:45 < Erlkoenig> also i didn't like the existing ones :D
2012-10-02T21:49:29 < Thorn> I still use gnu make. my encounter with scons wasn't very pleasant
2012-10-02T21:49:48 < zyp> Erlkoenig, have you seen how I am using SCons?
2012-10-02T21:50:27 < Laurenceb_> scones sound tasty
2012-10-02T21:51:10 < Erlkoenig> zyp: hm no
2012-10-02T21:52:00 < zyp> http://cgit.jvnv.net/laks_demo/tree/SConstruct
2012-10-02T21:52:55 < Erlkoenig> argh telephone... wait a few min
2012-10-02T21:53:05 < zyp> also, your rule command looks a bit verbose
2012-10-02T21:53:07 < Thorn> zyp: did you get out of tree build working?
2012-10-02T21:53:17 < zyp> Thorn, for what?
2012-10-02T21:53:41 < zyp> oh, you mean placing objects in a different directory from source files?
2012-10-02T21:53:52 < Thorn> yes
2012-10-02T21:54:10 < zyp> I did that in a work project last year, I didn't have any problems with that
2012-10-02T21:54:20 < Thorn> or even several directories, such as debug and release
2012-10-02T21:54:27 < zyp> yes, that's why I did it
2012-10-02T21:54:29 < Erlkoenig> excuse me, afk for a few min
2012-10-02T21:54:39 < Thorn> from what I understood scons has problems with that
2012-10-02T21:54:48 < zyp> I'm planning to do something similar with this demo project later
2012-10-02T21:55:21 < zyp> building the same demos for multiple microcontrollers
2012-10-02T21:56:00 < zyp> hmm, let me try it out, my ACM demo already supports both F1 and F4
2012-10-02T22:08:36 < Laurenceb_> https://encyclopediadramatica.se/Freenode
2012-10-02T22:08:38 < Laurenceb_> ^lol
2012-10-02T22:09:17 < Erlkoenig> re... hm make.rb shouldn't have problems with placing files into different directories... ^
2012-10-02T22:09:25 < BrainDamage> Laurenceb: https://encyclopediadramatica.se/File:Gnaa-wikipedia-assault.jpg
2012-10-02T22:10:05 < Erlkoenig> zyp: env.SelectMCU('stm32f407zg')  ah, how's defined which MCU gets what environment?
2012-10-02T22:10:26 < Laurenceb_> lawl
2012-10-02T22:11:12 < Erlkoenig> (08:53:05 PM) zyp: also, your rule command looks a bit verbose <- that's right, but allows for more flexibility... also, when having many files it balances out (there's one thing missing in the doc about shortening it)
2012-10-02T22:16:05 < emeb> hmmm... http://www.youtube.com/watch?v=hyGJBV1xnJI&feature=player_embedded
2012-10-02T22:16:39 < Laurenceb_> cheating
2012-10-02T22:16:46 < Laurenceb_> they have motion capture stuff
2012-10-02T22:17:08 < karlp> and?
2012-10-02T22:17:12 < karlp> why wouldn't you?
2012-10-02T22:17:23 < karlp> try and do it all by remembering where you flew and do it blind?
2012-10-02T22:18:12 < Laurenceb_> i mean they have loads of IR cameras around it
2012-10-02T22:18:49 < karlp> oh, you mean, they're telling the copters what to do from externally?
2012-10-02T22:21:08 < Laurenceb_> yes
2012-10-02T22:27:06 -!- barthess [~barthess@178.154.61.18] has quit [Quit: Leaving.]
2012-10-02T22:27:15 <+Steffanx> That sucks indeed
2012-10-02T22:27:26 < zyp> Thorn, hmm, it sort of works, but I didn't account for that in the implicit rules generated by my custom commands
2012-10-02T22:30:10 < emeb> Yeah - using fixed cameras to guide the copters isn't as impressive.
2012-10-02T22:30:29 < zyp> bullshit
2012-10-02T22:30:37 < Laurenceb_> they have synchronised IR LEDS
2012-10-02T22:30:37 <+Steffanx> not it's not bullshit
2012-10-02T22:30:38 < Laurenceb_> too
2012-10-02T22:30:38 < Erlkoenig> (08:53:05 PM) zyp: also, your rule command looks a bit verbose <-- I now documented it, now it's shorter: http://minecraft.niklas.2g2s.de/software/make.rb-doc/file.using.html#ddep
2012-10-02T22:31:02 < Laurenceb_> on the copters, and get position and attitude from each one out
2012-10-02T22:31:34 < Laurenceb_> so they can ID each one and get out x,y,z, quaternion
2012-10-02T22:31:56 < zyp> the cameras is merely a position sensing system, they are developing motion algorithms
2012-10-02T22:32:06 < Laurenceb_> its a commercial motion capture system
2012-10-02T22:32:22 < zyp> and what's wrong with it?
2012-10-02T22:32:44 < Laurenceb_> its useful in the real world
2012-10-02T22:33:52 < Laurenceb_> not much harder than flying a simulated copter in a games engine
2012-10-02T22:34:32 < zyp> how are the motion algorithms any less useful if they change to another position sensing system later?
2012-10-02T22:34:57 < Laurenceb_> you cant get that sort of accuracy easily
2012-10-02T22:35:22 < Laurenceb_> and you need to have synchronised comms and negotiation between the uavs
2012-10-02T22:35:55 < Laurenceb_> the motion capture is pretty cool, but they didnt make that part
2012-10-02T22:36:04 < Laurenceb_> and its useless in the real world
2012-10-02T22:36:06 < zyp> they didn't claim it either?
2012-10-02T22:36:12 < zyp> you are useless in the real world.
2012-10-02T22:36:36 < Laurenceb_> lol
2012-10-02T22:36:42 < Laurenceb_>  trololololol
2012-10-02T22:36:55 <+Steffanx> zyp tries to make it personal? :D
2012-10-02T22:38:17 < emeb> Good thing I don't have to exist in the real world :)
2012-10-02T22:38:38 < emeb> I sit here in my cozy office, talking to imaginary people over wires.
2012-10-02T22:39:59 <+Steffanx> At least i can kick you here :D :P
2012-10-02T22:40:43 < Laurenceb_> wait how did Steffanx get to be mod
2012-10-02T22:40:49 < Laurenceb_> hes never even made anything
2012-10-02T22:40:55 < Laurenceb_>  /troll
2012-10-02T22:41:00 <+Steffanx> sure i did
2012-10-02T22:41:12 -!- dos_fan] [~dos_fan@pdpc/supporter/active/dos-fan] has joined ##stm32
2012-10-02T22:41:51 < zyp> he made izua happy
2012-10-02T22:41:54 < zyp> or something like that
2012-10-02T22:42:07 <+Steffanx> or i was here before you two ? :P
2012-10-02T22:46:22 <+Steffanx> Anyway, i do the same thing you do Laurenceb
2012-10-02T22:46:28 <+Steffanx> I never finish anything :P
2012-10-02T23:04:45 < zyp> did you even start? :p
2012-10-02T23:05:35 < emeb> Nothing is ever finished. You just have to stop at some point.
2012-10-02T23:05:46 < emeb> there's alway one more tweak.
2012-10-02T23:07:43 <+Steffanx> For sure zyp
2012-10-02T23:13:24 <+Steffanx> You?
2012-10-02T23:15:26 < Tectu> I?
2012-10-02T23:28:52 <+Steffanx> He
2012-10-02T23:29:06 < zyp> me?
2012-10-02T23:29:10 < zyp> what about me?
2012-10-02T23:38:08 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-02T23:38:09 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-02T23:38:09 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-02T23:48:08 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-112.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-02T23:56:54 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
--- Day changed Wed Oct 03 2012
2012-10-03T00:32:51 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Read error: Operation timed out]
2012-10-03T00:36:25 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-03T00:49:18 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-03T01:09:59 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-10-03T01:15:32 -!- dos_fan] [~dos_fan@pdpc/supporter/active/dos-fan] has left ##stm32 ["Leaving"]
2012-10-03T01:20:37 < emeb> *crickets*
2012-10-03T02:11:49 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-03T02:16:42 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-03T02:47:27 < Thorn> why the hell am I using this stm32 crap
2012-10-03T02:48:12 < emeb> what's wrong with it?
2012-10-03T02:48:16 < Thorn> lpc1768 supports 6 pairs of bulk and 5 pairs of interrupt EPs, and has 2 CAN controllers too
2012-10-03T02:48:30 < emeb> oh - usb problems
2012-10-03T02:52:04 < Thorn> I can do 2x CDC-ACM and everything else I can think of
2012-10-03T02:53:11 < emeb> so what's the downside of LPC (there's got to be something).
2012-10-03T02:53:34 < emeb> price/speed/other features
2012-10-03T02:54:09 < Thorn> it's on par with stm32f405 costwise (at least where I buy parts) even though it's m3 and 64Kb RAM
2012-10-03T02:54:35 < Thorn> not sure about speed but I think ~100MHz
2012-10-03T02:54:37 < emeb> so a bit more expensive for the features.
2012-10-03T02:55:15 < Thorn> much smaller errata, sane i2c, etc.
2012-10-03T02:55:56 < emeb> I guess the thing that brings a lot of people to STM32 is the price and the cheap devboards.
2012-10-03T02:56:20 < emeb> but there are definitely warts.
2012-10-03T02:56:42 < emeb> I've only used I2C for writing and it's pretty easy to use for that. I understand reading is a PITA.
2012-10-03T02:57:22 < Thorn> most interesting i2c devices are bidirectional in my experience
2012-10-03T02:57:33 < emeb> depends.
2012-10-03T02:57:43 < emeb> a lot of audio codecs have write-only register files.
2012-10-03T02:57:55 < emeb> that's what I use it for.
2012-10-03T02:58:31 < emeb> and those are only accessed once during power-on setup in my applications
2012-10-03T02:58:47 < emeb> so I'm not a typical case.
2012-10-03T02:59:42 < Thorn> and I already have a lpc1768 dev board lying around. time to dig it out
2012-10-03T03:00:29 < emeb> "familiarity breeds contempt" - would be good to get your opinion of the LPC after you've used it for a while.
2012-10-03T03:00:51 < dongs> attn Laurenceb http://stop-irc-bullying.eu/news
2012-10-03T03:02:21 < Thorn> do they have an irc channel?
2012-10-03T03:02:33 < emeb> a very polite one probably.
2012-10-03T03:06:14 < dongs> Thorn: haha
2012-10-03T03:14:29 < emeb> http://gothamist.com/attachments/arts_jen/LGpreviewgooglemaps1012.jpeg
2012-10-03T03:14:49 -!- izua_ [~izua@79.115.170.139] has joined ##stm32
2012-10-03T03:15:13 -!- izua_ is now known as Guest93857
2012-10-03T03:17:24 -!- Guest93857 [~izua@79.115.170.139] has quit [Changing host]
2012-10-03T03:17:24 -!- Guest93857 [~izua@unaffiliated/izua] has joined ##stm32
2012-10-03T03:17:27 -!- mode/##stm32 [+v Guest93857] by ChanServ
2012-10-03T03:18:50 < feurig> emeb thats pretty much how nyc was for most of th people who brouht their ipoops to the opn hardware semmit / makerfaire last week
2012-10-03T03:18:55 -!- Guest93857 is now known as izua
2012-10-03T03:19:47 < emeb> feurig: a mass of confusion? fruitco needs to be soundly thrashed for that.
2012-10-03T03:40:08 < dongs> http://www.youtube.com/watch?v=yD94CV7kLHE link for Laurenceb
2012-10-03T03:40:19 < Thorn> aaand BMP doesn't support lpc17
2012-10-03T03:40:25 < dongs> huh?
2012-10-03T03:40:38 < HTT-Bird> waitwhat?
2012-10-03T03:40:39 < dongs> why would debug probe need to "support" a processor
2012-10-03T03:41:00 < Thorn> probably easy to hack some support using existing lpc11 code
2012-10-03T03:41:25 < HTT-Bird> Thorn: debugging should still work apparently
2012-10-03T03:41:25 < Thorn> because flash programming is all different on every cortex chip
2012-10-03T03:41:36 < HTT-Bird> you'll just have to find some other way to program the Flash
2012-10-03T03:41:48 < HTT-Bird> do the LPC1700s come with a bootloader?
2012-10-03T03:42:11 < Thorn> anyway my dev board comes with j-link, this problem can wait until hardware is ready
2012-10-03T03:42:27 < feurig> emeb: yeah
2012-10-03T03:42:43 < dongs> jlink is good.
2012-10-03T03:42:47 < dongs> just keep using that.
2012-10-03T03:43:10 < Thorn> HTT-Bird: yes, and you can interact with it using a terminal emulator (or if using usb it presents itself as a MSC drive iirc)
2012-10-03T03:43:31 < Thorn> much more advanced than stm32
2012-10-03T03:43:39 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-10-03T03:44:50 < Thorn> and it has real code read protection, too
2012-10-03T03:44:52 < HTT-Bird> yeah.  the EFM32 I'm working with comes with a UART bootloader as well
2012-10-03T03:44:58 < HTT-Bird> XMODEM based
2012-10-03T03:46:01 < dongs> don't let cjbeard know, or he might get offended its not kermit
2012-10-03T03:46:23 < Thorn> time to do yet another 100-pin component in diptrace x_x
2012-10-03T03:46:43 < dongs> its easier than you think.
2012-10-03T03:47:00 < Thorn> I like long pin names
2012-10-03T03:47:25 < dongs> i just call them PA0.. etc and then rename them task-specific per schematic
2012-10-03T03:47:36 < dongs> i.e. actually naming it the function im using.
2012-10-03T03:47:48 < dongs> once the component is on schematic you can just right click/rename pin
2012-10-03T03:49:22 < dongs> good
2012-10-03T03:50:26 < Thorn> http://i.imgur.com/vvGrP.png
2012-10-03T03:50:33 < dongs> LOL
2012-10-03T03:50:43 < dongs> ya fuck that man
2012-10-03T03:50:45 < Thorn> I new you're going to like it
2012-10-03T03:50:53 < Thorn> +k
2012-10-03T03:51:30 < Thorn> reduces probability of error.
2012-10-03T03:52:32 < dongs> i usually do pin planning in excel etc for something thats really complicated
2012-10-03T03:52:48 < Thorn> but I admit it does look somewhat unorthodox
2012-10-03T03:53:35 < dongs>  http://i.imgur.com/qOjhH.png f103 with task-named pins
2012-10-03T03:54:03 < Thorn> that way you can rename a wrong pin
2012-10-03T03:54:32 < dongs> no
2012-10-03T03:54:38 < dongs> why would I do that?
2012-10-03T03:56:00 < dongs> Thorn: i mean, give me an example when its possible to rename a wrong pin.
2012-10-03T03:56:20 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 248 seconds]
2012-10-03T03:56:22 < dongs> I have the datasheet + (not in this case, but in other cases) excel pin planning sheet open.
2012-10-03T03:56:48 < dongs> and I crosscheck pin# w/function etc.
2012-10-03T03:57:23 < Thorn> that's a soucre of errors, with multiple columns of pin numbers in ST datasheets
2012-10-03T03:57:51 < dongs> i really dont like how they did pin numbersr in F3 datasheets..
2012-10-03T03:58:07 < dongs> if they dont fix the layout for that section im gonna copypaste that part of datasheet into excel and make my own
2012-10-03T03:58:26 < dongs> but F1/F2/F4 is ok
2012-10-03T04:11:50 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-03T04:12:16 -!- DaKu is now known as daku
2012-10-03T04:14:42 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-03T04:16:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-03T04:18:32 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-10-03T04:43:00 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-03T05:07:50 < karlp> Thorn: I dunno, only reason I am using stm32 is that a) it was actually available, b) it was cheaper than the avrs I was using, and c) they had a demo board that included a debugger for so cheap it was absurd
2012-10-03T05:08:07 < karlp> doesn't now and never did mean that stm32 was the most awesome ever
2012-10-03T05:09:15 < Bird|lappy> is having pin1 in the lower left corner instead of the upper left corner OK in a footprint library?
2012-10-03T05:11:37 < dongs> why not
2012-10-03T05:14:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 256 seconds]
2012-10-03T05:15:10 <+dekar> izua, https://dl.dropbox.com/u/28467113/rtlsdr/hf_mod_transformer_H16105DF.jpg
2012-10-03T05:15:13 < Bird|lappy> yeah, as long as one is consistent. :)
2012-10-03T05:16:25 <+izua> dekar: that looks like a bunch of stuff whacked together to do epicness
2012-10-03T05:16:31 <+izua> SDR hacks?
2012-10-03T05:17:03 <+izua> i think that's the SDR where you removed the mlf chip and shorted the pads
2012-10-03T05:17:06 <+dekar> yeah, I added a transformer I stole from an ADSL modem
2012-10-03T05:17:10 < Thorn> looks like NXP doesn't have USB drivers for lpc17xx. they do have example which include some kind of usb stack (C) Keil 2005-2006, but it's not separated into a library or documented in any way
2012-10-03T05:17:26 <+dekar> I did that to generate an differential signal for the ADC input
2012-10-03T05:17:28 <+izua> ooh, that's a transformer?
2012-10-03T05:17:40 <+izua> given the size of the USB jack, i'd say that's pretty neat
2012-10-03T05:17:48 <+dekar> RIP ADSL modem :P
2012-10-03T05:18:18 <+izua> i'm trying to make a DCF77 emulator
2012-10-03T05:18:29 <+izua> that stupid frankfurt signal only comes here at night, and even then it's faily
2012-10-03T05:18:48 <+izua> or perhaps you can tell those guys to juice up the xmitter
2012-10-03T05:20:12 <+dekar> why are you using our german signal anyway? is the romanian one too far off? :P
2012-10-03T05:21:44 <+izua> yeah, stupid romanian one still doesn't emit into the past..
2012-10-03T05:22:34 < dongs> Thorn: lol
2012-10-03T05:22:38 <+dekar> now I feel like recording some DCF77 for you :P
2012-10-03T05:22:53 <+izua> i know a guy who manually made a rbu666 decoder
2012-10-03T05:23:07 <+izua> s/manually made/made a receiver
2012-10-03T05:23:20 <+izua> i don't think you can go that low
2012-10-03T05:23:32 <+izua> it's VLF i think, 77.5kHz
2012-10-03T05:23:40 <+dekar> I can do that
2012-10-03T05:23:58 <+dekar> https://dl.dropbox.com/u/28467113/rtlsdr/gqrx-20121002-170201.mp3
2012-10-03T05:24:00 <+izua> try to get rbu
2012-10-03T05:24:04 <+dekar> some weird language I recorded
2012-10-03T05:24:07 <+izua> everybody gets dcf
2012-10-03T05:24:14 <+izua> also
2012-10-03T05:24:18 <+izua> speaking of weird and radio stuff
2012-10-03T05:24:51 <+izua> try to get UVB-76 too, 4.625MHz iirc
2012-10-03T05:25:09 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-03T05:25:11 <+izua> it would probably help to orient your thing in the general direction of russia :P
2012-10-03T05:25:16 <+dekar> will try tomorrow
2012-10-03T05:25:20 <+dekar> somewhat sleepy atm
2012-10-03T05:26:21 <+izua> try to get the russian ones though
2012-10-03T05:26:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-03T05:26:26 <+dekar> my blogs visits pretty much exploded since I posted some mac software on reddit
2012-10-03T05:26:26 <+izua> everybody does DCF77
2012-10-03T05:26:49 <+dekar> I even got fan mail and stuff
2012-10-03T05:27:17 <+izua> the gqrx?
2012-10-03T05:27:34 <+dekar> yeah
2012-10-03T05:27:48 <+izua> set up a paypal thing, you know, aquire money, aquire females.
2012-10-03T05:28:45 <+izua> they'll flock towards you like radio waves in the focal of a parabolic mirror
2012-10-03T05:29:00 <+izua> and be like "show us your radio stuff elias" xD
2012-10-03T05:30:38 <+izua> so anyway
2012-10-03T05:30:42 <+dekar> I totally wanna see those females
2012-10-03T05:30:57 <+izua> so far i'm outputting a 77500 hz wave, and modulating it with a button
2012-10-03T05:31:15 <+izua> and the boxed receiver seems to enjoy it, more or less
2012-10-03T05:31:29 <+dekar> got a sunburn like sensation yet?
2012-10-03T05:31:53 <+izua> an antenna (read: bigass wire) seems to help, although the breadboard seems to act like a big one too
2012-10-03T05:32:08 <+izua> dekar: not yet, but i will bring on the MW range transmitter soon
2012-10-03T05:32:09 <+dekar> no need for an antenna, just up the power :D
2012-10-03T05:32:22 <+izua> antenna use the power more efficiently though :P
2012-10-03T05:32:57 <+izua> the frankfurt one has like 40% efficiency or so, since they need a huge one to match even 1/4 of the wavelength
2012-10-03T05:33:01 <+dekar> people like you are the reason why there is so much unemployment!
2012-10-03T05:33:07 <+izua> (not sure. i'm not good with radio stuff)
2012-10-03T05:33:13 <+izua> lol
2012-10-03T05:33:15 <+dekar> waste moar power, make moar jobs!
2012-10-03T05:33:30 <+izua> i'm not the one with lovemail from reddit
2012-10-03T05:33:34 <+izua> speaking of females
2012-10-03T05:33:55 <+dekar> my antenna > your antenna
2012-10-03T05:34:14 <+izua> they probably look like ladyada
2012-10-03T05:34:38 <+izua> and probably are as many as she's the only EE i know of, and that jessica chick that built her own console in a joystick
2012-10-03T05:34:54 <+dekar> izua, https://dl.dropbox.com/u/28467113/rtlsdr/hf_mod_antenna.jpg
2012-10-03T05:34:54 <+izua> yeah but my antenna has a higher frequency than your antenna
2012-10-03T05:35:23 <+dekar> I decided to solder 10 meters of litz to an sma connector
2012-10-03T05:39:18 <+izua> you know what you should record instead of dcf77
2012-10-03T05:39:21 <+izua> my dcf77 emitter
2012-10-03T05:39:26 <+izua> with its puny milliwatt range
2012-10-03T05:39:53 <+izua> since you're not anywhere near the transmitter, you can definitely pick it up
2012-10-03T05:43:43 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-03T05:46:33 < cjbaird> I've always wanted to see some milliwatt QRP digitial action..
2012-10-03T05:47:41 <+izua> cjbaird: iirc, the senior guy on the avrfreaks forums did something even more awesome - an I/Q decoder, purely software
2012-10-03T05:48:16 <+izua> it was incredibly selective too ><
2012-10-03T05:48:27 <+izua> but for even more QRP - olivia/mfsk :D
2012-10-03T05:50:07 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-03T05:50:08 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-03T05:50:08 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-03T05:53:42 < cjbaird> izua: can you recall offhand how much grunt the AVR hardware needed for that? I'e seen demodulators on MSP430 Launchpads before..
2012-10-03T05:54:32 < dongs> hamfags
2012-10-03T05:55:27 <+izua> hm. i'm not good with radio - but i think it was just the tank, some filter he wound up, and a delay network. not sure if that's it. here's the link though http://www.avrfreaks.net/index.php?name=PNphpBB2&file=viewtopic&p=216836 it's huge and incredibly ontopic.
2012-10-03T05:55:32 <+izua> dongs: hey dude
2012-10-03T05:55:54 < dongs> sup
2012-10-03T05:56:26 <+izua> i'm messing with a dcf77 transmitter for debugging
2012-10-03T05:56:33 <+izua> i just realised the protocol allows you to send :66 minutes
2012-10-03T05:57:16 < dongs> cool
2012-10-03T05:57:18 < dongs> oh, crap
2012-10-03T05:57:20 <+izua> you know.. if you'd broadcast that with even just a few watts, a lot of clocks could display 6:66 or 4:20 or such stuff.
2012-10-03T05:57:44 <+izua> one could troll a whole city like that.
2012-10-03T05:57:47 < dongs> emeb: i knew what I forgot to do... 5V/VBUS .. should I connect them? or not even bother? i.e. power MCU from LED_5V only and just ignore vbus
2012-10-03T05:58:03 < dongs> izua: haha
2012-10-03T05:58:13 < dongs> i think my clock here supports some rf timecrap
2012-10-03T05:58:22 < dongs> i dont think japan uses anything standard though
2012-10-03T05:58:31 <+izua> there is one in japan iirc
2012-10-03T05:59:05 < dongs> maybe for yank clocks
2012-10-03T05:59:10 < emeb> dongs: OH, you mean run the LEDs off the USB supply. Yeah, why not?
2012-10-03T05:59:19 < emeb> USB should be able to supply the load.
2012-10-03T05:59:24 < cjbaird> Years ago I picked up a cheap Bearcat scanner, and dsicovered the local shack were doing weekly RTTY and SSTV nights. Got it happening withNetBSD, even.
2012-10-03T05:59:40 < dongs> emeb: well, i am keeping 5V through board to board, so im not sure i wanna mix usb supply into that
2012-10-03T06:00:01 < dongs> my usual way of doing it is sticking a bat54c between vbus/5vcc, but thats not gonna be enough for hte current im dealign wiht
2012-10-03T06:00:34 < emeb> dongs: Yeah - that might be a problem. Maybe a jumper or solder bridge to allow just one to run off VBUS?
2012-10-03T06:00:58 < dongs> no good automatic way?
2012-10-03T06:01:01 < dongs> a fet or something?
2012-10-03T06:01:11 <+izua> dongs: you love arduino.
2012-10-03T06:01:17 <+izua> check their schematic, they have that
2012-10-03T06:01:29 < dongs> izua: their schematic needs an opamp and a bunch of other shit
2012-10-03T06:01:30 < dongs> no fuckin way.
2012-10-03T06:01:33 <+izua> it's not really 'good' or 'simple'
2012-10-03T06:01:38 < dongs> izua: i already used it in another product :)
2012-10-03T06:01:42 < dongs> a while ago.
2012-10-03T06:01:44 <+izua> lol
2012-10-03T06:02:03 <+izua> i had a solution with a micro relay
2012-10-03T06:02:29 <+izua> not sure if it was better than the arduino one, but still..
2012-10-03T06:04:12 < dongs> doubtful.
2012-10-03T06:05:37 < emeb> Might be able to fake something up with diodes, if you can stand the 0.7V drop.
2012-10-03T06:05:54 < emeb> or use schottkys for a 0.3V drop...
2012-10-03T06:06:09 < dongs> nah, thats ghetto.
2012-10-03T06:06:13 < dongs> ill just jumper it.
2012-10-03T06:06:55 < emeb> I've seen power switching chips that could probably do the job. Don't know what they cost though.
2012-10-03T06:08:58 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-10-03T06:11:09 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-03T06:11:26 < dongs> that guy is as rude as Laurenceb at /quitting
2012-10-03T06:11:39 < dongs> except he acutally quis the client instead of just smashing the power button in and ping timeouting
2012-10-03T06:13:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-03T06:19:28 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-03T06:19:34 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-03T06:21:46 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-03T06:49:50 -!- izua [~izua@unaffiliated/izua] has quit [Remote host closed the connection]
2012-10-03T08:11:26 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 252 seconds]
2012-10-03T09:39:56 < dongs> http://www.winningsolitaire.com/
2012-10-03T09:44:03 < emeb_mac> wheee!
2012-10-03T09:44:12 < emeb_mac> been a long time since I played that one.
2012-10-03T09:45:29 < dongs> click n drag for even more fun
2012-10-03T09:45:56 < emeb_mac> oh my!
2012-10-03T09:51:03 < ABLomas> yeeee
2012-10-03T10:11:21 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Read error: Connection reset by peer]
2012-10-03T10:11:34 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-03T10:21:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-03T11:06:36 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Quit: leaving]
2012-10-03T11:09:43 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-03T11:09:43 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-03T11:09:43 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-03T11:10:56 -!- |sark [~|sark@109.235.51.202] has quit [Read error: Operation timed out]
2012-10-03T11:14:04 < dongs> sup BLOGGERS
2012-10-03T11:14:16 < zyp> sup yourself
2012-10-03T11:14:48 < dongs> wanna get various shit done but lazy
2012-10-03T11:14:51 < jpa-> bah, antivirus is blocking winusb driver so i can't get openocd to work
2012-10-03T11:14:53 < jpa-> i hate windows
2012-10-03T11:15:09 < dongs> ???????????????
2012-10-03T11:15:23 < dongs> why use antivirus on windows?
2012-10-03T11:15:26 < dongs> you didn't seem particularly stupid
2012-10-03T11:15:36 < dongs> the only people who get viruses in 2012 are dumbfucks
2012-10-03T11:16:02 < dongs> HI MY NAME IS EMBAMBO MANUBUFUCK I HAVE $10000000 FOR YOU JUST CLICK HERE AND DOWNLOAD THIS EXE
2012-10-03T11:16:50 < jpa-> dongs: corporate policy
2012-10-03T11:16:58 < jpa-> also, can't add exceptions
2012-10-03T11:17:41 < dongs> terrible
2012-10-03T11:21:47 < zyp> hmm, my memory is not good enough
2012-10-03T11:21:53 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-03T11:22:41 < zyp> working on some issue and found shit in the logs I know I've seen before, but I don't remember where
2012-10-03T11:24:00 < zyp> so I've been looking through the shit I've worked on the last few months, but I can't fucking find it
2012-10-03T11:26:46 < dongs> last few months , eh
2012-10-03T11:26:52 < dongs> im lucky if i remember shit i worked on in the last few days
2012-10-03T11:27:11 < dongs> assburgers to the max
2012-10-03T11:27:42 < zyp> well, my memory is good enough that I remember that I've seen those symptoms before, but I don't remember what I was working on at the time
2012-10-03T12:01:58 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-03T12:01:59 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-03T12:01:59 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-03T12:15:33 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T12:15:34 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Excess Flood]
2012-10-03T12:15:52 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T12:20:07 < jpa-> if i order http://thequadshot.com/products/black-magic-probe-mini  what kind of cables does it include?
2012-10-03T12:20:48 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-03T12:33:24 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Remote host closed the connection]
2012-10-03T12:36:51 < dongs> jpa, by "cables" i assume its the 10pin idc.
2012-10-03T12:37:50 < dongs> also: why not just get jlinkob or something for fraction of the price
2012-10-03T12:41:49 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-03T12:47:41 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T12:50:43 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-03T12:51:23 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T12:52:08 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-03T13:00:32 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-03T13:05:39 < dongs> http://i.imgur.com/UC8u5.gif
2012-10-03T13:15:14 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-03T13:17:53 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T13:42:33 < dongs> http://i.imgur.com/xAmKX.jpg typical tarduino porj
2012-10-03T13:42:35 < dongs> proj
2012-10-03T13:43:21 < jpa-> typical arduino proj runs on z80?
2012-10-03T13:43:30 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has joined ##stm32
2012-10-03T13:54:42 -!- daku is now known as DaKu
2012-10-03T13:59:24 < cjbaird> i came
2012-10-03T13:59:45 < cjbaird> I'd say it's likely a NOP tester.
2012-10-03T14:00:26 < cjbaird> (Z80 NOP = 0x00; gnd all data lines = cpu cycles through entire address space)
2012-10-03T14:29:31 < dongs> cjbaird: wahts the point
2012-10-03T14:30:48 < cjbaird> Proves the Z80 werks
2012-10-03T14:30:58 < dongs> ohok.
2012-10-03T14:31:01 < dongs> how doyou see it working?
2012-10-03T14:32:31 < cjbaird> LEDs on the address lines, usually. http://i.imgur.com/TgZWO.gif
2012-10-03T14:32:44 < dongs> haha
2012-10-03T14:35:32 < Thorn> these days you need an expensive scope to see anything in address lines
2012-10-03T14:36:38 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-03T14:38:21 < cjbaird> A lot of those 1970s cpus could have their clocks running at any old rate..
2012-10-03T14:38:33 < dongs> well, same can be said about most mcu,s no?
2012-10-03T14:38:51 < dongs> i remember pulling a crystal off a sam7 board and it just paused, put the crystal back on, it went back to work
2012-10-03T14:42:13 < cjbaird> ha
2012-10-03T14:44:16 < dongs> ya, look at stm32 datasheet too
2012-10-03T14:44:22 < dongs> minimum fCLK shit is like 0
2012-10-03T14:50:10 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-03T14:53:35 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-03T15:07:54 < dongs> http://gizmodo.com/5947972/apple-acknowledges-iphone-5-camera-problem-says-youre-holding-it-wrong
2012-10-03T15:30:19 < dongs> http://www.alvolution.co.uk/safety/fall-detectors-and-fall-alarms/s1032ab-fall-panic-wander-alarm-badge-with-pager why is this so expensive
2012-10-03T15:31:08 < cjbaird> http://www.gizmodo.com.au/2012/09/laser-bike-light-creates-your-own-tron-like-virtual-lane/ --> http://boards.4chan.org/n/res/402307
2012-10-03T15:31:12 < dongs> whats the fall alarm made with, some random accelerometer?
2012-10-03T15:32:17 < karlp> it's expensive because it can be.
2012-10-03T15:32:35 < dongs> I need it for about 1/10th that cost.
2012-10-03T15:32:44 < dongs> should I use the new SAM4L ULTRA LOW POWER TARDUINO
2012-10-03T15:47:50 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-03T16:01:17 < Laurenceb> http://www.thisisbristol.co.uk/LIVE-STREAM-Bloodhound-rocket-test/story-17021454-detail/story.html
2012-10-03T16:01:26 < Laurenceb> lol @ "Cosworth" sign
2012-10-03T16:04:38 < Thorn> extremely fascinating http://www.ganssle.com/tem/tem229.html#article3
2012-10-03T16:10:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-03T16:36:13 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-03T16:46:54 < dongs>  'Brit patent specialists will be flown out around the world to give workshops and seminars, essentially educating nations on the benefits of obeying intellectual property law.'
2012-10-03T16:46:57 < dongs> just waht africa needs.
2012-10-03T16:47:08 < dongs> a bunch of brits coming there explaining htem about IP patents...
2012-10-03T16:47:43 < dongs> Britain's Intellectual Property Office will "promote the understanding of IP" in developing countries after signing a deal with the head of the UN's World Intellectual Property Organisation.
2012-10-03T16:52:43 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: leaving]
2012-10-03T16:57:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-03T17:26:17 < jpa-> how to disable IWDG during standby?
2012-10-03T17:26:35 < jpa-> hmm i mean in stop
2012-10-03T17:27:38 < jpa-> hmm.. apparently have to reset the processor to do that :F
2012-10-03T17:31:13 < zyp> yep, it's impossible to stop it once started
2012-10-03T17:32:44 < Tectu> I know how to use print to get the value of a variable in scope, but how to get it's address?
2012-10-03T17:34:35 < zyp> &
2012-10-03T17:34:57 < zyp> should be obvious if you know how to get the address in C :p
2012-10-03T17:35:00 < dongs> i was gonna suggest that, but that would be too obvious for GNU
2012-10-03T17:35:02 < Tectu> damn, that was pretty easy :D
2012-10-03T17:35:03 < Tectu> thanks zyp  :D
2012-10-03T17:35:10 < dongs> GNU/TOOHARD
2012-10-03T17:35:14 < jpa-> so no watchdog for me
2012-10-03T17:36:11 < jpa-> i'm too lazy to do some crazy hack that stores a bit saying "i really want to sleep" in some memory location (which i'd have to exclude from crt0 initialization) and then checking that after power on
2012-10-03T17:36:26 < dongs> uh
2012-10-03T17:36:30 < dongs> use BKP memory :)
2012-10-03T17:36:33 < dongs> thats waht its there for
2012-10-03T17:36:43 < dongs> no need to fuck wiht crt.
2012-10-03T17:37:36 < dongs> http://bcas.tv/paste/results/VPo3Mh13.html
2012-10-03T17:38:15 < Thorn> jpa-: just declare another section in your linker script and put it into ram
2012-10-03T17:38:21 < dongs> Thorn: no.
2012-10-03T17:38:25 < dongs> Thorn: ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2012-10-03T17:38:33 < dongs> BKP ram is for shit like this.
2012-10-03T17:38:41 < dongs> dont fucking gcc-ize crap when its not needed.
2012-10-03T17:39:01 < karlp> sections aren't gcc-izeing
2012-10-03T17:39:05 < dongs> yes it is.
2012-10-03T17:39:07 < Thorn> it's more portable than backup ram
2012-10-03T17:39:15 < dongs> Thorn: to what? every stm32 has it
2012-10-03T17:39:25 < dongs> linker scripts is gcc faggotry.
2012-10-03T17:39:34 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-03T17:39:41 < Thorn> lol stm32. there're more cortex chips than stm32
2012-10-03T17:40:01 < dongs> Thorn: hes working on stm32. hes asking about IWDG, which is stm32 specific. KTHX.
2012-10-03T17:40:23 < dongs> to do persistent ram between resets, he should be using bkp registers, as THATS WHAT THEY'RE FOR.
2012-10-03T17:41:13 < dongs> if youre saying some linker script trash is "portable" between x different cortex m mcus, yeaaaaaaaaaaaaaaaaah.
2012-10-03T17:41:27 < Thorn> how large is that backup ram anyway?
2012-10-03T17:41:35 < dongs> 42 *16bit
2012-10-03T17:41:44 < dongs> PLENTY FOR A ONE BIT FLAG
2012-10-03T17:42:05 < dongs> They are implemented in the backup domain that remains powered on by VBAT when the
2012-10-03T17:42:08 < dongs> VDD power is switched off. They are not reset when the device wakes up from Standby
2012-10-03T17:42:11 < dongs> mode or by a system reset or power reset.
2012-10-03T17:42:53 < Tectu> can I run a bashscript inside gdb to count a given pattern (0x55) when doing examine ?
2012-10-03T17:43:16 < zyp> you can probably do that with a python script
2012-10-03T17:43:23 < Thorn> you can use python in gdb, not bash
2012-10-03T17:43:32 < Tectu> damn
2012-10-03T17:43:52 < Thorn> and codesourcery gdb doesn't have python
2012-10-03T17:43:59 < Tectu> wtf?
2012-10-03T17:44:07 < Tectu> I use the GNU-ARM-GDB
2012-10-03T17:44:59 < zyp> well, does this work?: python print 'foo'
2012-10-03T17:45:13 < zyp> if not, it's built without python support
2012-10-03T17:45:25 < Tectu> I'll see tonight
2012-10-03T17:46:11 < Tectu> I just want a script which tells me how many stack I have still left in a thread in chibios. therefore, chibios does fill the stack (initializes it) with 0x55. so you can examine the given working are and count the 0x55 pattern
2012-10-03T17:46:27 < Tectu> the examine output does look like this (not from a chibios programm, so no 0x55):   http://pastebin.ca/2238930
2012-10-03T17:46:41 < Tectu> now it would be nice when there would be a script or something else which would count the 0x55
2012-10-03T17:47:57 < Thorn> there will be one if you write it
2012-10-03T17:48:54 < Tectu> I'd write a simple .py and then? I have to put it in the project directory where i start gdb from?
2012-10-03T17:53:45 < Thorn> http://yogiken.files.wordpress.com/2010/02/c-register-access.pdf
2012-10-03T17:53:48 < Thorn> my_organization::my_platform::periph::cr::en::write();
2012-10-03T17:56:00 < Tectu> o0
2012-10-03T18:01:50 < zyp> sounds like he's trying to turn it into java
2012-10-03T18:02:04 < zyp> they like namespaces in namespaces in namespaces like that
2012-10-03T18:04:33 < Tectu> namespace-ception
2012-10-03T18:04:45 < dongs> you gotta go deeper.r
2012-10-03T18:04:58 < dongs> also... bash scripts.. what?
2012-10-03T18:05:04 < dongs> isnt there automatic way you can do stack checking
2012-10-03T18:05:08 < dongs> without gdb.
2012-10-03T18:05:22 < dongs> have some stack checking task or someshit.
2012-10-03T18:05:29 < dongs> im to lazy to find the code I saw that did that.
2012-10-03T18:05:35 < dongs> it wasnt for kawaiiOS though.
2012-10-03T18:06:30 < dongs> sleep
2012-10-03T18:07:32 < Tectu> dongle, show code
2012-10-03T18:14:05 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-03T18:33:02 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-10-03T18:45:23 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-03T18:47:50 -!- nicodh [~grunbaum@mey38-3-88-164-118-144.fbx.proxad.net] has joined ##stm32
2012-10-03T19:17:43 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 272 seconds]
2012-10-03T19:19:39 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-03T19:23:35 -!- meloun [~meloun@188-192-251-120-dynip.superkabel.de] has joined ##stm32
2012-10-03T19:28:59 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-03T19:36:22 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-03T19:48:17 < Bird|lappy> besides, my EFM32TG222 doesn't even HAVE a backup domain :P
2012-10-03T19:48:24 < Bird|lappy> does. not. exist.
2012-10-03T19:50:28 < Bird|lappy> also, dongs: were you the one looking for a USB/5V switch?
2012-10-03T19:52:23 < zyp> I could also use a nice power source switch
2012-10-03T19:53:56 < Bird|lappy> http://www.linear.com/product/LTC4415
2012-10-03T19:59:51 -!- izua [~izua@79.115.170.139] has joined ##stm32
2012-10-03T19:59:51 -!- izua [~izua@79.115.170.139] has quit [Changing host]
2012-10-03T19:59:51 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-03T19:59:54 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-03T20:02:46 < emeb> Typical LTC - $$$
2012-10-03T20:04:11 < Bird|lappy> yeah, not that cheap, but capable
2012-10-03T20:10:22 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-03T20:11:56 < karlp> awesome, the hardware guys reckon it's a software problem, I reckon it's an analog problem.
2012-10-03T20:12:02 < karlp> here comes the data dumping code.
2012-10-03T20:12:16 < emeb> let the finger pointing begin.
2012-10-03T20:13:00 < karlp> same software works reliably when the voltage to sample is switched in with an SSR, fails abyssmally when it's on separate adc channels.
2012-10-03T20:13:05 < karlp> fun times.
2012-10-03T20:19:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-03T20:19:15 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-03T20:31:24 < zippe> zyp: TI have some that aren't bad cost-wise
2012-10-03T20:32:52 < Thorn> for dongs, stm32-specific == portable while gcc-specific is not lol
2012-10-03T20:34:26 < zyp> I'm more likely to use other MCU vendors than other compilers :p
2012-10-03T20:34:39 < Thorn> same
2012-10-03T20:35:43 < Thorn> besides, backup domain is great because it can keep data while main power is off if you have a battery. everything else is a disadvantage
2012-10-03T20:35:45 < Bird|lappy> besides, linker scripts are going to need to be ported anyway
2012-10-03T20:39:12 < Bird|lappy> btw: should I expect to be able to hand-solder a SOD-523 at all?
2012-10-03T20:39:21 < Bird|lappy> or should I go to a bigger package...?
2012-10-03T20:40:54 < Thorn> that thing looks really small, much smaller than sot-23
2012-10-03T20:41:27 < Bird|lappy> it is tiny.  doesn't look to be too problematic though, just put a little blob of solder on one pad, slide part in while heating blob, solder other pad
2012-10-03T20:41:51 < Bird|lappy> or am I crazed for thinking that way?
2012-10-03T20:42:51 < zyp> about the size of a 0603 if I'm not mistaken, should be easy
2012-10-03T20:43:15 < Bird|lappy> 1.6*0.8mm
2012-10-03T20:43:29 < zyp> yes, 0603 is 1.5x1 or so
2012-10-03T20:43:47 < Bird|lappy> yeah
2012-10-03T20:43:50 < Bird|lappy> it's about 0603 sized
2012-10-03T20:43:53 < Bird|lappy> so should be fine :)
2012-10-03T20:44:03 -!- tavish [~tavish@120.56.133.25] has joined ##stm32
2012-10-03T20:44:04 -!- tavish [~tavish@120.56.133.25] has quit [Changing host]
2012-10-03T20:44:04 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-03T20:46:15 < karlp> well, hardware doods put a 5v regulator on instead of a 3v3, that's first fail.
2012-10-03T20:46:28 < karlp> not the actual cause though :|
2012-10-03T20:49:20 < Bird|lappy> way to go :P
2012-10-03T20:49:21 -!- BrainDamage2 [~yaaic@217.200.185.44] has joined ##stm32
2012-10-03T20:49:21 -!- BrainDamage2 [~yaaic@217.200.185.44] has quit [Client Quit]
2012-10-03T20:55:28 < Thorn> how do I solder a qfp with a center pad (must be connected)? large PTH in the pad?
2012-10-03T20:56:08 <+Steffanx> Build a reflow oven :)
2012-10-03T20:56:29 < zyp> agreed
2012-10-03T20:56:47 <+Steffanx> or buy one of course
2012-10-03T20:56:52 < Thorn> and spend $$$ for stencils
2012-10-03T20:56:52 < emeb> Hotplate!
2012-10-03T20:57:16 <+Steffanx> You don't NEED stencils do you?
2012-10-03T20:57:22 < emeb> I've done center pads hand applying paste w/o stencils
2012-10-03T20:57:31 < emeb> then reflowing on a hotplate.
2012-10-03T20:57:59 < Thorn> well I tried solder paste + hot air -> lots of solder bridges on a qfp-0.5
2012-10-03T20:58:37 < emeb> too much paste?
2012-10-03T20:58:55 < zyp> likely
2012-10-03T20:59:02 < Thorn> if I use less paste, not all pins get soldered
2012-10-03T20:59:32  * emeb uses an iron on 0.5mm pitch lqfp
2012-10-03T20:59:59 < zyp> yeah, that's probably faster than dicking around with hot air
2012-10-03T21:00:06 < Thorn> I do too, but not with center pad of course
2012-10-03T21:00:28 < zyp> well, do center pad with paste and the rest by hand? :p
2012-10-03T21:00:44 < emeb> and hope the pins stay aligned during reflow.
2012-10-03T21:01:02 < zyp> add paste, solder other pins, reflow? :p
2012-10-03T21:01:09 < Thorn> solder center pad, find the chip is improperly positioned, profit
2012-10-03T21:01:38 < zyp> after pins are soldered, they'll keep it in position during reflow
2012-10-03T21:01:57 < emeb> surface tension is your friend
2012-10-03T21:02:04 < zyp> yep
2012-10-03T21:03:21 < emeb> I've heard of folks putting in vias, PTH on center pads and filling in solder afterwards. No idea how well it works.
2012-10-03T21:03:54 < emeb> ^if
2012-10-03T21:03:54 -!- izua [~izua@unaffiliated/izua] has quit [Quit: leaving]
2012-10-03T21:04:15 -!- izua [~izua@79.115.170.139] has joined ##stm32
2012-10-03T21:04:15 -!- izua [~izua@79.115.170.139] has quit [Changing host]
2012-10-03T21:04:15 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-03T21:04:16 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-03T21:04:16 <+Steffanx> Get the cheapest chip you can find and try it :)
2012-10-03T21:04:32 -!- barthess [~barthess@178.154.17.80] has joined ##stm32
2012-10-03T21:06:28 < zyp> I bet it's pretty hard to get a good solder connection that way
2012-10-03T21:07:24 < zyp> you need a thin tip and then the ground planes are eating up all the heat making it hard to heat the pad on the chip so the solder will connect properly
2012-10-03T21:07:58 < zyp> and then you try adding solder and end up having it fill the hole without connecting properly to the chip
2012-10-03T21:08:21 <+Steffanx> or not
2012-10-03T21:09:13 < zyp> well, I just imagine that's how it'll work out when you try stuffing both the tip and solder through a hole to solder a pad on the other side of the hole
2012-10-03T21:09:52 <+Steffanx> make two holes
2012-10-03T21:09:56 <+Steffanx> or one large enough holes
2012-10-03T21:10:04 <+Steffanx> -s
2012-10-03T21:10:19 < zyp> you can't make the hole larger than the footprint of the chip. :p
2012-10-03T21:10:27 <+Steffanx> but you have your oven, so you'll probably don't waste your time on it :)
2012-10-03T21:10:50 < zyp> we're not talking about me now :p
2012-10-03T21:11:13 <+Steffanx> Yes, but i wanted to say "give it a try"
2012-10-03T21:18:19 < emeb> anyone ever tried to use the DFU bootloader on an STM32F4xx?
2012-10-03T21:20:27 < zippe> emeb: yes
2012-10-03T21:20:36 < emeb> work?
2012-10-03T21:34:44 -!- izua [~izua@unaffiliated/izua] has quit [Remote host closed the connection]
2012-10-03T21:35:58 < zyp> why wouldn't it?
2012-10-03T21:36:31  * emeb shrugs
2012-10-03T21:37:16 < emeb> app note made some mention of the automatic HSE detection being a bit sensitive.
2012-10-03T21:43:10 < jpa-> Steffanx: spherebot complete  http://kapsi.fi/~jpa/stuff/pix/12100018.mp4
2012-10-03T21:44:44 <+Steffanx> complete or .. functional? :)
2012-10-03T21:44:57 <+Steffanx> And no fancy web format anymore?
2012-10-03T21:45:01 <+Steffanx> that google one
2012-10-03T21:45:16 < jpa-> didn't bother to convert
2012-10-03T21:45:45 < jpa-> and quite functional, yes
2012-10-03T21:46:56 < jpa-> http://kapsi.fi/~jpa/stuff/pix/sphere_and_controller.jpg
2012-10-03T21:48:44 <+Steffanx> It's not transparent anymore :(
2012-10-03T21:49:18 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has joined ##stm32
2012-10-03T21:49:27 <+Steffanx> but nice :)
2012-10-03T21:49:58 < jpa-> better grip :)
2012-10-03T21:50:24 <+Steffanx> Better grip, it's some rubber stuff?
2012-10-03T21:50:27 < jpa-> yeah
2012-10-03T21:50:29 < jpa-> plastidip
2012-10-03T21:50:37 < emeb> rabbit is not impressed
2012-10-03T21:50:37 <+Steffanx> Yeah, that stuff is great
2012-10-03T21:50:54 < jpa-> it wasn't impressed the first time either
2012-10-03T21:51:27 <+Steffanx> Rabbit? I thougth that creature in the background was a cat :)
2012-10-03T21:51:56 < emeb> bun.
2012-10-03T21:52:49 < jpa-> it runs past the camera at one point :P
2012-10-03T21:53:50 < emeb> our cats are rarely impressed with the stuff I make.
2012-10-03T21:54:05 < emeb> mostly they just stare at me like I'm out of my fscking mind.
2012-10-03T21:58:00 <+Steffanx> What's next jpa- ?
2012-10-03T21:58:12 <+Steffanx> One that can move on the water?
2012-10-03T21:58:47 < emeb> the jesus robot.
2012-10-03T22:02:37 < jpa-> Steffanx: nah, something totally else :P
2012-10-03T22:02:47 < jpa-> i'll give this as a present to my nephew
2012-10-03T22:23:47 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 245 seconds]
2012-10-03T22:26:17 < karlp> bah, eagle update all from library isn't updating the footprint
2012-10-03T22:27:28 < zyp> no, you need to update a specific library to do that
2012-10-03T22:29:08 < karlp> tried that to, with the path
2012-10-03T22:29:15 < karlp> what worked was addign another copy of the part,
2012-10-03T22:29:28 < karlp> it then went, "hey, another version in the schematic already, update?"
2012-10-03T22:29:40 < karlp> which fixed things, and then I could just remote the temporarily added part
2012-10-03T22:30:52 < emeb> eagle. meh.
2012-10-03T22:32:03 < karlp> don't feel like learning kicad from scratch
2012-10-03T22:32:05 < zippe> emeb: agreed
2012-10-03T22:32:11 < zippe> karlp: Proteus
2012-10-03T22:32:41 < zippe> US$500-1k depending on how big you want your designs to be
2012-10-03T22:32:45 < karlp> not my day job sorry,
2012-10-03T22:33:00 < zippe> karlp: how much is your sanity worth?
2012-10-03T22:33:07 < karlp> some, but not that much
2012-10-03T22:33:14 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-03T22:33:46 < zippe> mmk
2012-10-03T22:34:02 < Laurenceb_> sexy time
2012-10-03T22:34:34 < emeb> I've learned so many schematic / layout tools over the years. All blurring together...
2012-10-03T22:34:48 < karlp> I've got most of the things sorted out with eagle most of the time, and I don't really want to spend _any_ time learning any more right now.
2012-10-03T22:34:59 < karlp> doesn't mean I like eagle
2012-10-03T22:35:26  * emeb doesn't really _like_ any  of them. 
2012-10-03T22:36:34 < zyp> I'm quite happy with eagle for the boards I've done so far
2012-10-03T22:37:00 < Laurenceb_> people say eagle is hard to learn
2012-10-03T22:37:01 < karlp> the hw guys at work are usign eagle anyway, it works enough, and it measn I can get their help when I'm doing anything more than I know how to do anyway
2012-10-03T22:37:05 < Laurenceb_> i say try kicad
2012-10-03T22:37:13 < Laurenceb_> and then say its hard to learn
2012-10-03T22:37:55 < zyp> how hard a tool is to learn is irrelevant after you've learned it
2012-10-03T22:38:05 < zippe> zyp: not necessarily
2012-10-03T22:38:06 <+Steffanx> And it is non-functional on the os i use
2012-10-03T22:38:15 <+Steffanx> At least, it was the last time i tried it
2012-10-03T22:38:26 < zippe> For example, Eagle is so hard to learn that unless you use it regularly, you are always (re)learning how to use it
2012-10-03T22:38:40 < Laurenceb_> lol
2012-10-03T22:38:42 < emeb> that's how I feel about Perl. :P
2012-10-03T22:38:49 < Laurenceb_> zippe: try kicad
2012-10-03T22:38:53 <+Steffanx> That's how you should feel about per
2012-10-03T22:38:54 <+Steffanx> l
2012-10-03T22:38:54 < Laurenceb_> then say its hard
2012-10-03T22:39:01 < karlp> what's odb++ manufacturering output?
2012-10-03T22:39:01 < zippe> Laurenceb: I have
2012-10-03T22:39:05 -!- barthess [~barthess@178.154.17.80] has quit [Quit: Leaving.]
2012-10-03T22:39:20 < Laurenceb_> kicad makes eagle look like a tool for retards
2012-10-03T22:39:27 < Laurenceb_> in terms of ease of use
2012-10-03T22:39:30 < zippe> kicad has its moments, but the library support is still very bad
2012-10-03T22:39:42 < karlp> I'm sure whether that means it's better or worse laurence?
2012-10-03T22:39:49 < zippe> And it's a total crapshoot reliability wise
2012-10-03T22:39:51 < Laurenceb_> lol
2012-10-03T22:39:54 < karlp> kicad lets you define footprints in text thoguh right?
2012-10-03T22:40:00 < Laurenceb_> ...
2012-10-03T22:40:02 < Laurenceb_> somehow
2012-10-03T22:40:07 < zippe> karlp: geda used to let you do that
2012-10-03T22:40:10 < karlp> so theoretically you can script define them from datasheet defintions, rather than drawing all over the place?
2012-10-03T22:40:14 < Laurenceb_> yeah i might try geda
2012-10-03T22:40:24 < emeb> that moment when kicad has had enough of your shenanigans and just goes *poof* and disappears.
2012-10-03T22:40:28 < zippe> but apparently procedural generation of parts was "too hard" for some people, so they swapped to "draw it wrong and deal" like everyone else
2012-10-03T22:40:30 < Laurenceb_> hehe some chance
2012-10-03T22:41:01 < zippe> I hate working with the Open Hardware community
2012-10-03T22:41:01 < emeb> Oh yeah - PCB footprints written in M4. ugh!
2012-10-03T22:41:03 < karlp> I tried geda and kicad a couple of years ago, but it had been so long since I'd made pcbs I wanted some handholding from the ninterwebnet
2012-10-03T22:41:09 < zippe> They whine if it's not Eagle, and that sads me
2012-10-03T22:41:38 < emeb> zippe: yeah - what is it w/ Open HW & Eagle?
2012-10-03T22:42:02 < zyp> the affordability of eagle, I guess?
2012-10-03T22:42:13 < emeb> must be
2012-10-03T22:42:35 < Laurenceb_> its pretty funny
2012-10-03T22:42:41 < Laurenceb_> open source but closed tools
2012-10-03T22:42:52 < Laurenceb_> or rather
2012-10-03T22:42:58 < Laurenceb_> open hardware but closed source
2012-10-03T22:43:01 < emeb> Laurenceb_: I mostly use gEDA/PCB these days. Some kicad, but gEDA is in my fingers at this point...
2012-10-03T22:43:21 < Laurenceb_> yeah
2012-10-03T22:43:27 < Laurenceb_> i tried kicad and gave up
2012-10-03T22:43:35 < Laurenceb_> ill try geda stuff
2012-10-03T22:43:40 < zippe> emeb: "fwee"
2012-10-03T22:43:46 < emeb> also, I've got fairly good lib of symbols & footprints.
2012-10-03T22:43:52 < zippe> No workable kicad snapshots for OS X
2012-10-03T22:44:10 < emeb> zippe: no kidding - no idea why they can't do Mac OS X.
2012-10-03T22:44:28 < karlp> what's the difference between odb++ and "gerbers" ?
2012-10-03T22:44:29 < emeb> part of the reason I don't use it as much. gEDA/PCB works on Mac though.
2012-10-03T22:45:21 < zyp> so does eagle
2012-10-03T22:45:29 < emeb> Yep.
2012-10-03T22:45:34 -!- nicodh [~grunbaum@mey38-3-88-164-118-144.fbx.proxad.net] has quit []
2012-10-03T22:45:35 < zyp> and that's one of the reasons I like eagle
2012-10-03T22:45:45 <+Steffanx> Indeed, the reason why i don't use kicad .. :)
2012-10-03T22:45:54 < emeb> I got biased against eagle when I first tried it and all the databases were binary.
2012-10-03T22:46:12 <+Steffanx> Didn't the partially fix that in 6.x?
2012-10-03T22:46:14 < emeb> I guess they're XML now, so more tweakable.
2012-10-03T22:46:34 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-03T22:46:35 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-248-131.clienti.tiscali.it] has quit [Disconnected by services]
2012-10-03T22:46:36 < zyp> the change didn't matter to me at all
2012-10-03T22:47:06 <+Steffanx> No, it didn't change much for the 'eye'
2012-10-03T22:47:10 < emeb> I'd gotten used to tweaking text for things that the GUI couldn't do.
2012-10-03T22:47:22 <+Steffanx> lol
2012-10-03T22:47:30 < emeb> probably a bad habit. :P
2012-10-03T22:47:34 < karlp> and having any sort of idea on changes between files
2012-10-03T22:47:47 < emeb> there's that too.
2012-10-03T22:48:11 < emeb> the qi hardware folks have scripts that generate graphic diffs from kicad. Pretty cool.
2012-10-03T22:48:18 < karlp> jpa-: how do you control that ball again?
2012-10-03T22:48:27 < zippe> Eagle "works" on the Mac like wax and feathers "work" for transcontinental travel.
2012-10-03T22:48:40 <+Steffanx> uh?
2012-10-03T22:49:24 <+Steffanx> With KiCad you won't even make it zippe, so everyhing is better than that
2012-10-03T22:50:15 < BrainDamage1> there was a port but it was so buggy it was unusable
2012-10-03T22:50:24 < BrainDamage1> I didn't believe until I tried it on a friend's pc
2012-10-03T22:50:30 <+Steffanx> Isn't it based on wxwidgets? So why it works so terrible?
2012-10-03T22:50:48 < zippe> geda/gaf seems dead; no stable release in > 18mo
2012-10-03T22:50:54 < BrainDamage1> because wx got an implementation for each platform using the native toolkit
2012-10-03T22:51:05 < BrainDamage1> so it has it's own set of bugs on each platform
2012-10-03T22:51:13 < zippe> Steffanx: "it" being what? wx?
2012-10-03T22:51:18 <+Steffanx> kicad
2012-10-03T22:51:18 < zippe> Steffanx: or kicad?
2012-10-03T22:51:21 < BrainDamage1> so you still need some os specific code
2012-10-03T22:51:39 < zyp> zippe, are you claiming that eagle runs worse on os x than on windows or linux?
2012-10-03T22:51:42 < zippe> Steffanx: Oh, because most of what makes a CAD program usable is very fast canvas updates
2012-10-03T22:51:50 < zippe> zyp: yes
2012-10-03T22:51:59 < zippe> zyp: unless it's also amazingly slow on Windows
2012-10-03T22:52:17 <+Steffanx> I guess you use another version of EAGLE, zippe
2012-10-03T22:52:28 < BrainDamage1> the canvas isn't very usable on mac of kicad
2012-10-03T22:52:33 < BrainDamage1> periodic blanking
2012-10-03T22:52:46 < BrainDamage1> the library selector doesn't even work well either
2012-10-03T22:53:01 < zippe> Steffanx: I'm current with Eagle on a 4.5GHz hackintosh with a recent ATi card.
2012-10-03T22:53:18 < zippe> But the braindamage with Eagle runs deep
2012-10-03T22:53:27 < zippe> For example, why can't I click on something to select it?
2012-10-03T22:53:46 < zippe> To move something, why do I have to pick the "move" tool, then hunt for the reference point for an object?
2012-10-03T22:54:01 <+Steffanx> Yeah, that's weird
2012-10-03T22:54:03 < zippe> If I click and drag inside the boundary of a single object, why doesn't that just pick it up?
2012-10-03T22:54:17 < zippe> Why can't I shift or alt-click to select multiple things?
2012-10-03T22:54:18 <+Steffanx> *annoying
2012-10-03T22:54:32 < zippe> Why can't I group objects into ad-hoc groups?
2012-10-03T22:54:36 < emeb> zippe: where you getting that 18mo figure on gaf/geda?
2012-10-03T22:54:39 < zippe> I could go on and on and on
2012-10-03T22:54:47 < zippe> emeb: release date for 1.6.2 was 1/2011
2012-10-03T22:55:11 < zippe> emeb: straight off their website
2012-10-03T22:55:22 < zippe> Last I checked it was 9/2012
2012-10-03T22:55:58 < emeb> the website doesn't track the code base very closely.
2012-10-03T22:56:16 < emeb> the git repo is a lot fresher than that.
2012-10-03T22:56:17 < BrainDamage1> you guys should talk with bunder
2012-10-03T22:56:28 < BrainDamage1> in ##electronics
2012-10-03T22:56:32 < BrainDamage1> he's involved with geda dev
2012-10-03T22:56:43 < BrainDamage1> but he also says the project is almost dead
2012-10-03T22:56:54  * emeb got a patch to gschem accepted a couple months back
2012-10-03T22:57:02 < BrainDamage1> almost, not totally
2012-10-03T22:57:25 < emeb> Yeah - the ml is really quiet lately.
2012-10-03T22:57:45 < emeb> The primary maintainer for PCB seems to have drifted away.
2012-10-03T22:58:00 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-03T22:58:10 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-03T23:02:06 < zippe> Don't get me wrong; I'd love a community-developed package with good UI and library discipline
2012-10-03T23:02:08 < zippe> There just ain't one
2012-10-03T23:04:28 < emeb> all of 'em - FOSS & commercial have plenty of warts.
2012-10-03T23:05:10 < emeb> nice thing is there's so many to choose from :P
2012-10-03T23:06:54 < BrainDamage> that's the typical talk done with politicians
2012-10-03T23:07:03 < BrainDamage> they all suck but there's many to pick from ...
2012-10-03T23:08:29 < HTT-Bird> zippe: it's the library discipline that probably will kill you :P
2012-10-03T23:08:49 < HTT-Bird> it's one of those things where there's just too many different choices there to settle on one "perfect" one
2012-10-03T23:08:55 < HTT-Bird> IPC7351 has its warts, even
2012-10-03T23:08:56 < zippe> HTT-Bird: yeah; I did library maint for an employer a while back… it's very annoying
2012-10-03T23:09:19 < zippe> HTT-Bird: but the alternate practice - build your own from scratch - also sucks
2012-10-03T23:10:02 < HTT-Bird> I personally think that footprint generators are the way forward for the most part for IC packaging
2012-10-03T23:10:16 < HTT-Bird> connectors can be quite strange though, sadly :<
2012-10-03T23:11:09 < Laurenceb_> http://www.gedasymbols.org/scripts/search.cgi?key=stm32
2012-10-03T23:11:14 < Laurenceb_> says it all
2012-10-03T23:13:48 < HTT-Bird> Laurenceb_: :P nothing that you can't handle with djboxsym
2012-10-03T23:16:29 < emeb> funny thing is that most mfg provide drawings of the recommended footprint.
2012-10-03T23:16:55 < emeb> wonder why there can't be a standard CAD format for that and automatic tools to convert it to every PCB layout tool.
2012-10-03T23:17:36 < BrainDamage> nice joke
2012-10-03T23:18:02 < BrainDamage> why would you want your library to work with competitor's sw?
2012-10-03T23:18:33 < emeb> not the library
2012-10-03T23:18:51 < emeb> import mfgs recommended FP into your library.
2012-10-03T23:19:03 < BrainDamage> please don't use akronyms :/
2012-10-03T23:19:25 < BrainDamage> I can use them but I can never understand them when someone else says
2012-10-03T23:19:33 < emeb> component manufacturers provide drawings for recommended footprints.
2012-10-03T23:19:48 < emeb> in electronic form that can be downloaded, along with datasheets.
2012-10-03T23:20:06 < emeb> the footprint data can then be converted to any CAD tool library format.
2012-10-03T23:20:40 < BrainDamage> that would be lovely, but like this having a good lib is also an advantage for a sw
2012-10-03T23:20:52 < BrainDamage> a selling point
2012-10-03T23:20:55 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-10-03T23:21:10 < BrainDamage> for free software it would make much sense tough
2012-10-03T23:21:21 < BrainDamage> but the industry generally ignores free software
2012-10-03T23:21:37 < emeb> so you're saying that the CAD tool providers wouldn't want to import a common format because it would devalue their libraries?
2012-10-03T23:21:53 < BrainDamage> oh no, if it'd exists, they'd import it
2012-10-03T23:22:08 < BrainDamage> but I am skeptical they'd settle down and actually agree on a standard
2012-10-03T23:22:16 < emeb> There's that.
2012-10-03T23:22:41 < emeb> and if they did it would be "horse designed by a committee."
2012-10-03T23:22:47 < emeb> ie EDIF.
2012-10-03T23:22:49 < Tectu> I've a question to the cores internals: the PC does not always increment right? when you do calls and branches and stuff, it changes, right?
2012-10-03T23:23:14 < emeb> why does it matter?
2012-10-03T23:23:26 < BrainDamage> jump does reset the program counter
2012-10-03T23:23:34 < HTT-Bird> emeb: actually you could use a mechanical-CAD format
2012-10-03T23:23:39 < HTT-Bird> .DXF perhaps?
2012-10-03T23:23:54 < Tectu> BrainDamage, what happens when it overflows?
2012-10-03T23:23:54 < emeb> HTT-Bird: yeah - that's the first thing that came to mind.
2012-10-03T23:24:44 < BrainDamage> if the compiler is anyway sane, it'll never overflow
2012-10-03T23:24:53 < BrainDamage> there'll be a jump just before th overflow
2012-10-03T23:26:05 < Tectu> hmm
2012-10-03T23:26:12 < Tectu> so it is impossible to overflow it with a sane compiler?
2012-10-03T23:26:30 < Tectu> BrainDamage, but it does "reset" the PC to 0 when I call  foo()  etc?   or call in a deeper level?
2012-10-03T23:26:39 < Tectu> </noob_questions>
2012-10-03T23:27:02 < BrainDamage> do you even know what the program counter contains?
2012-10-03T23:27:21 < Tectu> i read about it, should do again
2012-10-03T23:27:59 < BrainDamage> it's the address of the instruction to load, different instructions will have different entry addresses in the code
2012-10-03T23:28:20 < BrainDamage> sorry, different functions
2012-10-03T23:28:54 < Tectu> wait.... instruciton to load?
2012-10-03T23:29:07 < Tectu> as the address of    mov   instruction or something?
2012-10-03T23:29:53 < BrainDamage> yes
2012-10-03T23:30:30 < BrainDamage> take the bytestream that's loaded in your memory
2012-10-03T23:30:36 < BrainDamage> which forms the program
2012-10-03T23:30:49 < Tectu> ah
2012-10-03T23:30:49 < BrainDamage> say each instruction is n bytes long
2012-10-03T23:30:59 < BrainDamage> you load the first instruction by setting the PC to that
2012-10-03T23:31:08 < BrainDamage> then increment the pc by n
2012-10-03T23:31:19 < BrainDamage> to point the next instruction to execute
2012-10-03T23:31:25 < Tectu> ah, it's the address i can read in the ELF?
2012-10-03T23:32:08 < BrainDamage> I'm not sure what you mean by that
2012-10-03T23:32:57 < Tectu> eh, not the EFL
2012-10-03T23:33:02 < Tectu> in the objdump
2012-10-03T23:33:39 < Laurenceb_> http://www.brownebagracing.com/pictures/hf_tool_sale.pdf
2012-10-03T23:35:36 < Tectu> Laurenceb_, wtf?
2012-10-03T23:37:59 < zippe> tectu: the PC is just a register in v7m
2012-10-03T23:38:16 < zippe> Tectu: the CPU fetches the next instruction from the address it points to
2012-10-03T23:38:42 < Tectu> zippe, i am not sure if i understand "fetching instruction"
2012-10-03T23:39:04 < zippe> Tectu: it reads bytes from that address and pushes them into the CPU
2012-10-03T23:39:17 < Tectu> ah, so the code that will make the instruction?
2012-10-03T23:39:19 < zippe> Tectu: the CPU interprets the bytes as an instruction
2012-10-03T23:39:23 < Tectu> so the CPU knows which instruction to make?
2012-10-03T23:39:29 < Tectu> okay, I see
2012-10-03T23:39:45 < zippe> Tectu: every instruction causes the PC to change
2012-10-03T23:39:53 < Tectu> so with 32-bit it's impossible to overflow?
2012-10-03T23:40:25 < zippe> Tectu: when you say "overflow", what do you mean?
2012-10-03T23:41:24 < Tectu> zippe, getting from 0xFFFFFFFF to 0x000000
2012-10-03T23:41:31 < Tectu> aw, to less digits :D
2012-10-03T23:42:03 < BrainDamage> again, as I said, won't happened with a sane compiler
2012-10-03T23:42:05 < Tectu> just reaching the maximum you can hold and add + 1
2012-10-03T23:42:11 < Tectu> yeah
2012-10-03T23:42:17 < zippe> Tectu: so there are a bunch of problems there
2012-10-03T23:42:22 < BrainDamage> if a section of code is near the end of PC it'll just insert an unconditional jump
2012-10-03T23:42:28 < zippe> First, the PC is normally never incremented by 1
2012-10-03T23:42:34 < zippe> though you can certainly do that by hand
2012-10-03T23:42:36 < BrainDamage> to another part of code
2012-10-03T23:42:55 < BrainDamage> zippe: unless the instructions are 1 byte wide!
2012-10-03T23:43:13 < zippe> second, for the PC to be 0xffffffff and then update, you would have to have loaded an instruction from 0xffffffff
2012-10-03T23:43:23 < zippe> BrainDamage: assuming stm32, they aren't
2012-10-03T23:43:40 < zippe> Tectu: however, in general you are looking at this from the wrong direction
2012-10-03T23:43:59 < zippe> Tectu: ARM define a bunch of rules that affect how the machine behaves for various PC values
2012-10-03T23:44:05 < zippe> Tectu: you should understand them
2012-10-03T23:44:29 < Tectu> okay
2012-10-03T23:44:35 < zippe> Tectu: if you add a value to the PC that causes it to overflow, then it will just wrap
2012-10-03T23:44:43 < zippe> Just like any other register
2012-10-03T23:45:07 < Laurenceb_> arg wtf
2012-10-03T23:45:08 < zippe> Tectu: However, you are subject to all the rules that apply to the PC
2012-10-03T23:45:20 < Laurenceb_> so ive made a datalogger that mounts as mass storage
2012-10-03T23:45:22 < zippe> Tectu: e.g. it must be suitably aligned, you must be able to read from the location it points to, etc.
2012-10-03T23:45:28 < Laurenceb_> i thought id be nice and simple to use
2012-10-03T23:45:34 < Laurenceb_> so utter n00bs could run it
2012-10-03T23:45:35 < Tectu> zippe, but the address in the PC is the one i see when i use objdump?
2012-10-03T23:45:55 < zippe> Tectu: objdump shows you what's in a file
2012-10-03T23:45:56 < Bird|lappy> ...but Windoze needs a .INF just to be able to use its built-in class driver?
2012-10-03T23:46:02 < Laurenceb_> im getting emails from people asking what a USB cable is, how to plug it in etc
2012-10-03T23:46:03 < zippe> Tectu: it has nothing to do with the PC
2012-10-03T23:46:05 < Bird|lappy> oh. :P
2012-10-03T23:46:10 < Bird|lappy> wow
2012-10-03T23:46:14 < Tectu> zippe, hmm
2012-10-03T23:46:23 < Laurenceb_> someone just returned one broken with a usb cable jammed in upside down
2012-10-03T23:46:51 < Bird|lappy> Laurenceb_, pix? because that takes a fair bit of doing
2012-10-03T23:46:59 < zippe> Tectu: assuming you are disassembling an object file, objdump will print the addresses at which instructions are located, and addresses that jumps reference, etc.
2012-10-03T23:47:17 < zippe> Tectu: when executing in the machine, the PC will normally track those values
2012-10-03T23:47:27 < Laurenceb_> Bird|lappy: im at home atm
2012-10-03T23:47:31 < Tectu> zippe, "addresses at which instructions are located"   <--- so it is a value that will be in PC ?
2012-10-03T23:47:34 < Laurenceb_> was returned to where i work
2012-10-03T23:47:45 < Laurenceb_> commercial medical datalogger
2012-10-03T23:47:53 < Bird|lappy> ah.
2012-10-03T23:47:54 < Laurenceb_> stm32 based
2012-10-03T23:47:55 < zippe> Tectu: may
2012-10-03T23:48:04 < Laurenceb_> i knew it had to be idiot proof
2012-10-03T23:48:08 < zippe> Tectu: assuming a) the code works and b) the code in question is ever executed
2012-10-03T23:48:17 < Laurenceb_> but i didnt realise itd encounter this level of idiocy
2012-10-03T23:48:26 < Tectu> zippe, hmm
2012-10-03T23:48:27 < zippe> Laurenceb: it's ok, bill them for a new one plus service charge
2012-10-03T23:48:38 < Laurenceb_> heh
2012-10-03T23:48:40 < zippe> Tectu: It's better to think about it the other way
2012-10-03T23:48:46 < Bird|lappy> just how do you insert a USB B connector upside down?
2012-10-03T23:48:50 < zippe> Tectu: If you see a value in the PC, you can look up the instructions
2012-10-03T23:48:58 < Laurenceb_> with great force
2012-10-03T23:49:07 < Bird|lappy> ...must be
2012-10-03T23:49:07 < Laurenceb_> the pcb header is all bent
2012-10-03T23:49:34 < Laurenceb_> - mini B
2012-10-03T23:49:52 < Bird|lappy> ah.
2012-10-03T23:50:08 < Laurenceb_> half in at 45 degrees
2012-10-03T23:50:36 < Tectu> zippe, where? in the objdump, right?
2012-10-03T23:53:43 < Tectu> zippe, sorry if I ask to stupid or obvious questions
2012-10-03T23:55:30 < Thorn> you can even lookup a C source line with addr2line if compiled with debugging information
2012-10-03T23:56:24 < zippe> Tectu: my point was more that IMO it helps to understand how the machine works and apply that to the software
2012-10-03T23:56:32 < zippe> too many people think "it works because magic"
2012-10-03T23:57:00 < Tectu> Thorn, -g is to compile with debug, right?
2012-10-03T23:57:02 < zippe> If you look at the ARM architecture manual, they have pseudocode for what each instruction really does
2012-10-03T23:57:06 < Laurenceb_> wait.. thats not true?!
2012-10-03T23:57:19 < Tectu> zippe, okay
2012-10-03T23:57:58 < Tectu> zippe, thanks for you explanations :)
--- Day changed Thu Oct 04 2012
2012-10-04T00:01:12 < Thorn> -g or even -g3
2012-10-04T00:04:29 < Tectu> what does the 3?
2012-10-04T00:08:02 < Tectu> anyways, i have to go to bed, cu guys
2012-10-04T00:18:45 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has quit [Ping timeout: 276 seconds]
2012-10-04T00:26:01 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-04T00:31:25 -!- meloun [~meloun@188-192-251-120-dynip.superkabel.de] has left ##stm32 []
2012-10-04T00:35:53 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-04T00:44:47 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-04T01:08:28 -!- Tunilame [29e28e51@gateway/web/freenode/ip.41.226.142.81] has joined ##stm32
2012-10-04T01:08:47 < Tunilame> hello
2012-10-04T01:16:15 < karlp> yeah, priv messages are not normally considered good form.
2012-10-04T01:17:37 <+Steffanx> but it did wake you up
2012-10-04T01:31:53 -!- DaKu is now known as daku
2012-10-04T01:36:41 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-10-04T01:41:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-04T01:42:37 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-04T02:11:29 -!- Laurenceb_ [~Laurence@host86-177-59-132.range86-177.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-04T02:29:17 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-04T02:37:15 -!- Tunilame [29e28e51@gateway/web/freenode/ip.41.226.142.81] has quit [Quit: Page closed]
2012-10-04T02:37:37 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-04T02:38:07 -!- Tunilame [29e28e51@gateway/web/freenode/ip.41.226.142.81] has joined ##stm32
2012-10-04T02:57:46 -!- Tunilame [29e28e51@gateway/web/freenode/ip.41.226.142.81] has quit [Quit: Page closed]
2012-10-04T04:15:41 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-04T04:25:08 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-04T04:28:00 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-04T04:37:39 -!- |sark [~|sark@109.235.51.202] has quit [Remote host closed the connection]
2012-10-04T04:39:01 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-04T04:52:22 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-04T04:52:50 < dongs> um
2012-10-04T04:52:58 < dongs> static float accVelScale = 9.80665f / 10000.0f / acc_1G ; static float accVelScale = 9.80665f / acc_1G / 10000.0f;
2012-10-04T04:53:02 < dongs> isnt that same?
2012-10-04T04:59:44 < qyx_> should be
2012-10-04T05:04:55 < dongs> some guy on the internet says it is nt
2012-10-04T05:05:38 < emeb> some guy is often wrong.
2012-10-04T05:06:57 < dongs> yeah it wouldnt make sense reallyt
2012-10-04T05:07:06 < dongs> BUT for some other compilers it can be an issue, if compiler start to parse from end to begin. E.g. for 2nd case (temp = acc_1G / 10000.0f) will be calculated at first and then 9.80665f / temp. As you see result will be fully incorrect...
2012-10-04T05:07:11 < dongs> ((9.80665f / acc_1G) / 10000.0f) != (9.80665f / (acc_1G / 10000.0f))
2012-10-04T05:07:13 < dongs> ?????????????
2012-10-04T05:07:22 < dongs> C has operator precedence rules etc why would different compiler do this differently??
2012-10-04T05:09:24 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 260 seconds]
2012-10-04T05:10:29 < dongs> woot. one of led reels showed up
2012-10-04T05:11:02 < dongs> no product label or anything lol.
2012-10-04T05:11:05 < dongs> some shit written in white marker.
2012-10-04T05:12:56 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-04T05:14:33 < qyx_> afaik every compiler parses expressions left to right
2012-10-04T05:14:45 < qyx_> maybe except some hipster stuff
2012-10-04T05:17:09 < dongs> right
2012-10-04T05:17:25 < dongs> what about if acc_1G was an int
2012-10-04T05:17:34 < dongs> it would implicitly cast because its being divided by floats right?
2012-10-04T05:21:02 < qyx_> yes
2012-10-04T05:21:22 < qyx_> if at least one operand is float then the division is fp also
2012-10-04T05:22:08 < dongs> right
2012-10-04T05:22:14 < dongs> so this guy must be a nutcase
2012-10-04T05:26:39 < cjbaird> Anonymous still lives: http://i.imgur.com/Tpg7D.jpg
2012-10-04T05:28:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-04T05:32:01 < qyx_> what cabling would be more suitable for ultrasound receiver? coax or twisted pair?
2012-10-04T05:32:29 < qyx_> the transducer can is connected to one pin
2012-10-04T05:33:59 < BrainDamage> coax is always superior to twisted pair
2012-10-04T05:34:28 < BrainDamage> is it mass produced or small amounts for which some extra $ don't matter?
2012-10-04T05:34:40 < BrainDamage> for the latter, go coax without a doubt
2012-10-04T05:35:08 < BrainDamage> for the first, consider how strong is the received signal and how much noise you can tollerate
2012-10-04T05:35:20 < qyx_> thats the matter, i need to connect ~6 of them and the prices of small coax connectors are fairly high
2012-10-04T05:35:45 < qyx_> i was considering smb or something similar
2012-10-04T05:35:52 < qyx_> until i saw prices
2012-10-04T05:35:53 < BrainDamage> really? sma are dirt cheap
2012-10-04T05:36:33 < qyx_> i'd like to have something smaller and more lightweight
2012-10-04T05:36:48 < BrainDamage> you can also just use a barrel jack
2012-10-04T05:36:53 < qyx_> ufl's are also relatively cheap
2012-10-04T05:36:55 < BrainDamage> it's not like you need constant impedance
2012-10-04T05:38:37 < qyx_> i was trying to do systhetic aperture sonar on a breadboard, but not usable
2012-10-04T05:38:48 < qyx_> as it catches too moch noise
2012-10-04T05:39:41 < BrainDamage> shitty coax + audio jacks
2012-10-04T05:39:47 < BrainDamage> should be enough for you
2012-10-04T05:39:57 < qyx_> yep, at least for the prototype
2012-10-04T05:40:39 < qyx_> in future it will be carried on uav
2012-10-04T05:40:52 < qyx_> hence the worries about connector weight
2012-10-04T05:43:55 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-04T05:44:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-04T05:52:24 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-04T06:15:26 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-04T06:18:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-04T06:18:42 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-04T06:37:51 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-04T06:56:28 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-04T07:02:43 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-04T07:21:01 < dongs> if opensores faggots would actually get paid for what they do.. they would remain poor fucks
2012-10-04T07:21:05 < dongs> because all the shit they make is fucking garbage.
2012-10-04T07:25:39 < upgrdman> lol
2012-10-04T07:25:47 < upgrdman> troll much? ;)
2012-10-04T07:27:10 < dongs> truth is not  trolling
2012-10-04T07:34:27 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-04T07:52:09 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-04T07:55:27 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-04T07:56:34 < dongs> gooked?
2012-10-04T08:03:26 < emeb_mac> getting really bored with all the "imaginative" covers of gotye's "somebody you used to know".
2012-10-04T08:03:51 < cjbaird> I don't even remember the song, and it supposedly was produced and released here.
2012-10-04T08:04:14 < emeb_mac> it's amusing the first few thousand times.
2012-10-04T08:06:15 < emeb_mac> so how would you route this: http://www.digikey.com/product-detail/en/MT45W1MW16BDGB-708%20AT%20TR/557-1419-1-ND/1828643
2012-10-04T08:07:12 < emeb_mac> looks like it's about the only thing available to use on the FSMC that's got more than 512kB
2012-10-04T08:07:58 < emeb_mac> but 0.75mm bga pitch sounds like it would need an expensive PCB process.
2012-10-04T08:08:49 < dongs> he SiM3L1xx Precision32 32-bit microcontroller family enables up to 65% longer battery life than competing solutions.
2012-10-04T08:08:52 < jpa-> karlp: movable counterweight (motor + servo)
2012-10-04T08:08:52 < dongs> - 175 .A/MHz active mode; < 250 nA sleep mode with real-time clock enabled at 3.6 V
2012-10-04T08:08:55 < dongs> oops
2012-10-04T08:08:58 < dongs> there goes SAM4L
2012-10-04T08:09:33 < dongs> emeb_mac: you mean hole in pad type shit?
2012-10-04T08:09:51 < dongs> er via in pad
2012-10-04T08:10:04 < emeb_mac> dongs:  yeah, or at the very least 10mil vias between pads
2012-10-04T08:10:58 < dongs> my standard vias are 10/20mil
2012-10-04T08:11:01 < dongs> 10hole 20ring
2012-10-04T08:11:17 < dongs> that doesnt fit does it?
2012-10-04T08:12:20 < emeb_mac> checking...
2012-10-04T08:13:11 < dongs> o shieet
2012-10-04T08:13:12 < dongs> no way
2012-10-04T08:13:16 < cjbaird> Rage Against the BGA
2012-10-04T08:13:28 < dongs> wait
2012-10-04T08:13:29 < dongs> it does
2012-10-04T08:13:48 < emeb_mac> the recommended pads are 0.37mm dia
2012-10-04T08:13:52 < dongs> no
2012-10-04T08:13:54 < emeb_mac> on 0.75mm pitch
2012-10-04T08:13:56 < dongs> recommended pads are 0.3
2012-10-04T08:14:01 < dongs> in teh datasheet
2012-10-04T08:14:08 < dongs> DIMENSIONS APPLY
2012-10-04T08:14:08 < dongs> TO SOLDER BALLS
2012-10-04T08:14:08 < dongs> POST REFLOW.
2012-10-04T08:14:08 < dongs> PRE-REFLOW BALL
2012-10-04T08:14:08 < dongs> DIAMETER IS 0.35
2012-10-04T08:14:11 < dongs> ON A 0.30 SMD
2012-10-04T08:14:13 < dongs> BALL PAD.
2012-10-04T08:14:17 < emeb_mac> so space between pads = 0.37mm = 15mil
2012-10-04T08:14:30 < dongs> 0.3 is that referring to pattern
2012-10-04T08:14:33 < dongs> or the shit on chipo
2012-10-04T08:14:35 < cjbaird> Make me a 4x16MB PSRAM breakout board, plz.
2012-10-04T08:14:53 < dongs> sure
2012-10-04T08:14:55 < dongs> are ytou gonna use it
2012-10-04T08:14:56 < emeb_mac> cjbaird: yeah - that would be handy
2012-10-04T08:15:27 < cjbaird> We were looking at psram a few months ago. would be nice..
2012-10-04T08:15:46 < emeb_mac> too bad they don't put them in nicer pkgs
2012-10-04T08:16:38 < dongs> http://i.imgur.com/QomIz.png
2012-10-04T08:16:42 < dongs> 8/16 via works
2012-10-04T08:16:49 < dongs> 10/16 might work too if factory doesnt bitch
2012-10-04T08:16:57 < cjbaird> I partially-recall the FSMC on the 'F4 has a 64MB banking limit
2012-10-04T08:17:09 < emeb_mac> cjbaird: sounds right
2012-10-04T08:17:17 < emeb_mac> dongs: not bad
2012-10-04T08:21:54 < emeb_mac> the fabs I've used wouldn't take that - all 6/6 and 13mil min drill, 7mil min ring.
2012-10-04T08:24:29 < dongs> heh
2012-10-04T08:25:09 < dongs> http://i.imgur.com/CaVpI.png i fabbed this shit few months ago
2012-10-04T08:25:51 < dongs> that had ot be via in pad though, but holes are 6/12mil;
2012-10-04T08:27:20 < emeb_mac> tight!
2012-10-04T08:29:26 < dongs> is that the biggest psram on dkey?
2012-10-04T08:29:36 < emeb_mac> nah - they go bigger.
2012-10-04T08:29:51 < emeb_mac> that one had smallest number of bga pads though
2012-10-04T08:30:09 < dongs> lol
2012-10-04T08:30:23 < dongs> as if that matters
2012-10-04T08:31:01 < dongs> most are nnc anyway
2012-10-04T08:33:30 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has joined ##stm32
2012-10-04T08:38:01 < emeb_mac> Micron used to make PSRAM in TSSOP pkgs - got one on a Digilent Nexsys 2 FPGA board.
2012-10-04T08:38:25 < emeb_mac> M45W8MW16 part number - no longer even cross-references on Micron's website though.
2012-10-04T08:38:37 < emeb_mac> "the secratary will disavow any knowledge..."
2012-10-04T08:43:55 < emeb_mac> I wish. So far nothing - all the folks who want that much RAM can either afford the high-end PCB process, or can settle for SDRAM in friendlier packages.
2012-10-04T08:44:51 < emeb_mac> or buy w/o going thru limited US distys
2012-10-04T08:48:25 -!- mrcan_ [~mrcan@78.173.75.48] has joined ##stm32
2012-10-04T08:48:25 -!- mrcan_ [~mrcan@78.173.75.48] has quit [Changing host]
2012-10-04T08:48:25 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-04T08:51:34 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 240 seconds]
2012-10-04T08:52:33 < emeb_mac> winbond: datasheet? "Contact us"
2012-10-04T08:52:53 < emeb_mac> package? "KGD only" wtf?
2012-10-04T08:53:02 < dongs> KGB only
2012-10-04T08:53:17 < emeb_mac> that's GRU to you.
2012-10-04T08:53:25 < emeb_mac> or was it FSB?
2012-10-04T08:53:46 < dongs> haha i dnno
2012-10-04T08:53:48 < dongs> what are those 2
2012-10-04T08:54:49 < emeb_mac> other three-letter-acronyms for Russian intelligence services.
2012-10-04T08:55:19 < emeb_mac> equiv to US FBI/CIA/NSA/TSA etc
2012-10-04T08:56:58 < dongs> kgb might be dead now
2012-10-04T08:57:17 < emeb_mac> prolly
2012-10-04T08:57:42 < emeb_mac> etron: datasheet? "Contact our sales directly"
2012-10-04T08:57:47 < emeb_mac> no pkg specified.
2012-10-04T08:58:16 < emeb_mac> or else TFBGA
2012-10-04T08:58:21 < dongs> how about issi shit?
2012-10-04T08:58:22 < dongs> http://www.issi.com/pdf/66WVE4M16BLL.pdf etc
2012-10-04T08:59:05 < emeb_mac> there you go.
2012-10-04T08:59:41 < emeb_mac> "The 48-pin TSOP-I package option is not yet available. Please contact SRAM Marketing at sram@issi.com
2012-10-04T08:59:41 < emeb_mac> for additional information. "
2012-10-04T08:59:46 < dongs> wat
2012-10-04T08:59:51 < dongs> places have it in stock
2012-10-04T08:59:58 < emeb_mac> must be old DS
2012-10-04T09:00:07 < emeb_mac> who stocks it?
2012-10-04T09:00:16 < dongs> http://components.arrow.com/part/detail/49697817S8745984N1393?region=na
2012-10-04T09:00:18 < dongs> arrow apparently
2012-10-04T09:00:21 < dongs> 28 iun stock
2012-10-04T09:00:30 < dongs> o wtf says fbga
2012-10-04T09:00:31 < dongs> hmm
2012-10-04T09:00:59 < dongs> shit never mind i suck dicks.
2012-10-04T09:02:38 < emeb_mac> d'oh!
2012-10-04T09:04:27 < emeb_mac> http://components.arrow.com/part/detail/50905582S8883481N1393
2012-10-04T09:04:40 < emeb_mac> ISSI PSRAM - 512k x 16.
2012-10-04T09:04:56 < dongs> isnt that real sdhitty
2012-10-04T09:04:57 < dongs> shitty
2012-10-04T09:04:59 < dongs> 1 meg?
2012-10-04T09:05:00 < emeb_mac> kinda small capacity, but in a tssop-ii
2012-10-04T09:05:27 < emeb_mac> cheap though - < $3
2012-10-04T09:05:52 < emeb_mac> but you can get real SRAM at about the same capacity for not much more $$
2012-10-04T09:07:05 < dongs> emeb_mac: those shits with more balls didnt have better pitch?
2012-10-04T09:07:58 < emeb_mac> viz: http://www.digikey.com/product-detail/en/R1LV0816ASB-7SI%23B0/R1LV0816ASB-7SI%23B0-ND/2772721
2012-10-04T09:08:46 < emeb_mac> flyback: yeah - I've used ISSI large SRAM before. Worked fine, good TSOP pkg.
2012-10-04T09:09:43 < emeb_mac> yeah: http://www.digikey.com/product-detail/en/IS62WV51216BLL-55TLI/706-1048-ND/1555421
2012-10-04T09:09:57 < emeb_mac> a bit more $$ than the Renesas parts tho
2012-10-04T09:11:40 < dongs> Amic, Chiplus, Etron, Fidelix, Hynix, Micron, Pointec, Winbond
2012-10-04T09:11:42 < dongs> psram makers.
2012-10-04T09:12:11 < emeb_mac> don't forget ISSI.
2012-10-04T09:12:23  * zyp forgets ISSI
2012-10-04T09:12:33 < emeb_mac> what?
2012-10-04T09:14:33 < dongs> lol, looks like everyone standardtized on that 6x8mm bga
2012-10-04T09:15:23 < dongs> all h ynix psram is eol
2012-10-04T09:15:31 < emeb_mac> just spotted that
2012-10-04T09:16:07 < emeb_mac> only old people use PSRAM
2012-10-04T09:16:27 < dongs> http://www.pointec.com.tw/en/en_Product.asp
2012-10-04T09:16:31 < dongs> but no info aout package type
2012-10-04T09:17:18 < emeb_mac> typical
2012-10-04T09:17:45 < zyp> so, what's up? today is PSRAM day?
2012-10-04T09:18:01 < emeb_mac> something to do before bedtime.
2012-10-04T09:18:28 < emeb_mac> since STM32 FSMC doesn't support DRAM, PSRAM is the largest capacity option.
2012-10-04T09:18:34 < zyp> I just woke up
2012-10-04T09:18:52 < emeb_mac> but all PSRAM is in tiny BGAs, so not feasible for small projects.
2012-10-04T09:19:35 < zyp> surely tiny BGAs would be the most feasible option for _small_ projects :p
2012-10-04T09:19:45 < emeb_mac> har - you made a funny.
2012-10-04T09:20:20 < dongs> haha
2012-10-04T09:20:52 < dongs> if theres any possibility of return on investment i have no problem running a small batch of psram breakout boards
2012-10-04T09:20:59 < dongs> but number of people who'd use that shit is probably quite low
2012-10-04T09:21:00 < emeb_mac> so to be more precise, mid-sized projects on small budgets.
2012-10-04T09:21:20 < dongs> maybe if I put it on an arduino shield...
2012-10-04T09:21:38 < dongs> "more ram than you can ever hope to address!"
2012-10-04T09:21:39 < emeb_mac> I've got a customer who would probably buy several  hundred.
2012-10-04T09:21:47 < cjbaird> I'
2012-10-04T09:21:54 < cjbaird> I'm only good for 2 boards. :)
2012-10-04T09:23:08 < zyp> so, default package is this TFBGA with 6x9 balls, 0.75mm pitch?
2012-10-04T09:23:14 < emeb_mac> yep
2012-10-04T09:23:50 < emeb_mac> or 6x8 for smaller capacity
2012-10-04T09:24:13 < zyp> shouldn't be that hard to do?
2012-10-04T09:24:21 < cjbaird> kickstarter this shit..
2012-10-04T09:24:23 < dongs> < dongs> http://i.imgur.com/QomIz.png
2012-10-04T09:24:27 < dongs> nope not hard at all
2012-10-04T09:24:37 < dongs> 0.75mm and 8/16 via
2012-10-04T09:24:48 < zyp> 6 cols means it's only three «rings», do the two outer in top layer and inner in bottom, shouldn't even need four layer
2012-10-04T09:24:52 < emeb_mac> zyp: can't do it on OSHPark rules
2012-10-04T09:25:17 < zyp> really?
2012-10-04T09:25:30 < dongs> um you surely dont need 4 layer
2012-10-04T09:25:36 < dongs> but hte problem is vias between pads
2012-10-04T09:25:37 < emeb_mac> min drill on OSHPark is 13
2012-10-04T09:26:13 < zyp> then use seeed, I've heard they are good at drilling :p
2012-10-04T09:26:15 < dongs> with 10/20 via, you have <5mil pad to pad
2012-10-04T09:26:18 < dongs> looool
2012-10-04T09:26:56 < dongs> http://i.imgur.com/aSlBy.jpg any idea waht are those connectors
2012-10-04T09:27:15 < dongs> o hey my camera has same
2012-10-04T09:27:23 < dongs> but still that d oesnt help
2012-10-04T09:27:37 < emeb_mac> new to me
2012-10-04T09:27:50 < dongs> the white ones obv
2012-10-04T09:27:57 < dongs> the connectors are flat
2012-10-04T09:27:58 < emeb_mac> yep
2012-10-04T09:28:00 < dongs> not pins
2012-10-04T09:28:05 < dongs> must be something common cuz it s chinashit
2012-10-04T09:28:20 < dongs> emeb_mac: hundreds of psram breakout boards?
2012-10-04T09:28:26 < dongs> you know i can make it happen :)
2012-10-04T09:28:31 < emeb_mac> dongs: yeah
2012-10-04T09:28:54 < emeb_mac> guy does small run audio stuff - would probably sell several hundred of these
2012-10-04T09:29:03 < dongs> waht is that shit for? your DSP stuff?
2012-10-04T09:29:07 < emeb_mac> talking about doing his own at the moment
2012-10-04T09:29:10 < emeb_mac> dongs: yeah
2012-10-04T09:29:44 < dongs> i can find out. that via in pad 0.5mm bga shit was like 30 boards run and it was really cheap
2012-10-04T09:30:54 < emeb_mac> nothing formal at the moment, but if you can wag the cost in 100qty that would be a good datapoint.
2012-10-04T09:31:12 < dongs> Right.
2012-10-04T09:31:16 < dongs> http://i.imgur.com/3sLac.jpg this really needs something to show the scale
2012-10-04T09:31:58 < emeb_mac> cute
2012-10-04T09:31:59 < dongs> but most passives are 0402.
2012-10-04T09:32:11 < dongs> and whoel board is 3x2.5cm
2012-10-04T09:32:27 < emeb_mac> the 8-pins regular SOIC?
2012-10-04T09:32:30 < dongs> ea
2012-10-04T09:32:32 < dongs> yea
2012-10-04T09:32:45 < emeb_mac> gives scale.
2012-10-04T09:33:33 < emeb_mac> how does it mount?
2012-10-04T09:34:02 < dongs> theres a long board to board socket on teh bottom
2012-10-04T09:34:08 < dongs> its evalboard shit, plugs into the green one y ou see below
2012-10-04T09:34:16 < emeb_mac> makes sense.
2012-10-04T09:35:03 < zyp> hmm, right, oshpark vias are too large to fit
2012-10-04T09:35:23 < emeb_mac> yeah - disappointing
2012-10-04T09:35:46 < dongs> dongs bloggery service.
2012-10-04T09:36:04 < zyp> I ran some numbers here, 0.75mm pitch means 41.75mil diagonal pitch
2012-10-04T09:36:23 < zyp> with 12mil pads, you have just under 30mil for the via and spacing on either side of it
2012-10-04T09:36:30 < emeb_mac> subtract 12 for clearance
2012-10-04T09:36:40 < emeb_mac> = 29.75
2012-10-04T09:36:59 < dongs> can you calculate probability of seeedstudio missing the drill hole
2012-10-04T09:37:08 < zyp> 250%?
2012-10-04T09:37:13 < emeb_mac> 13mil min drill + 2x7 min ring = 27
2012-10-04T09:37:22 < emeb_mac> might barely fit
2012-10-04T09:37:34 < dongs> whats 2x7 min ring shit
2012-10-04T09:37:37 < dongs> what does tha mean
2012-10-04T09:37:50 < zyp> emeb_mac, you're forgetting either spacing or pad
2012-10-04T09:37:56 < emeb_mac> annular ring around drill must be >= 7 mil
2012-10-04T09:38:10 < dongs> ok right so 10mil drill = 17mil outer?
2012-10-04T09:38:54 < emeb_mac> dongs: I think it's 2x7 - ring goes all the way around drill hole.
2012-10-04T09:39:19 < dongs> whats that in mils. i dont understand
2012-10-04T09:39:20 < emeb_mac> zyp: subtracted 12mil from 41.75 for 6 mil space on all sides
2012-10-04T09:39:41 < emeb_mac> dongs: all in mils
2012-10-04T09:39:47 < dongs> right so?
2012-10-04T09:39:49 < dongs> drill = 17
2012-10-04T09:39:50 < dongs> err
2012-10-04T09:39:52 < dongs> drill = 10
2012-10-04T09:39:55 < dongs> wahts minimum ring?
2012-10-04T09:39:58 < emeb_mac> drill = 10
2012-10-04T09:40:09 < dongs> 14? 17? 10+14?
2012-10-04T09:40:14 < emeb_mac> min ring = 7 on all sides, so 14 total
2012-10-04T09:40:17 < emeb_mac> yeah
2012-10-04T09:40:21 < dongs> ok. thats not bad.
2012-10-04T09:40:22 < emeb_mac> 24
2012-10-04T09:40:43 < dongs> so 10/24 min?
2012-10-04T09:40:45 < dongs> or 10/14
2012-10-04T09:40:51 < dongs> 10/14 is reasonable, 10/24 is like get the fuck out
2012-10-04T09:40:51 < emeb_mac> 10/24
2012-10-04T09:41:21 < emeb_mac> but that 7 mil is OSHPark rules - no idea what your fab's are.
2012-10-04T09:42:33 < dongs> well theyre not bitching about my standard 10/20 vias on default process,, and they did 6/12 for that bga board so
2012-10-04T09:42:55 < emeb_mac> so they must allow 3mil ring
2012-10-04T09:45:10 < zyp> 08:39:20 < emeb_mac> zyp: subtracted 12mil from 41.75 for 6 mil space on all sides
2012-10-04T09:45:19 < zyp> and then you have to subtract another 12 for the pad
2012-10-04T09:45:37 < zyp> unless you're going to have only vias and no pads :p
2012-10-04T09:46:15 < emeb_mac> zyp: 41.75 - 6 - 6 = 29.75 - 13 (drill) - 7 -7 = 2.75
2012-10-04T09:46:57 < emeb_mac> basically yeah - pads & vias
2012-10-04T09:47:15 < emeb_mac> but only need vias for the interior 2 rows
2012-10-04T09:47:22 < zyp> that's irrelevant
2012-10-04T09:47:31 < zyp> the pitch is the same everywhere
2012-10-04T09:47:31 < emeb_mac> 2 outer rows can route on top layer
2012-10-04T09:47:54 < emeb_mac> no vias needed there
2012-10-04T09:51:19 < emeb_mac> zyp: oh. derp.
2012-10-04T09:52:24 < emeb_mac> yeah - need to subtract the pad dia too. :P
2012-10-04T09:54:41 -!- dekar_ [~dekar@212.255.33.3] has joined ##stm32
2012-10-04T09:54:44 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-04T09:55:04 < zyp> hmm, I did some more math
2012-10-04T09:55:52 < zyp> if you pull the via closer to the pad, it's sort of possible to make it fit
2012-10-04T09:56:30 < zyp> except that the hole will eat up almost half the pad
2012-10-04T09:57:26 < emeb_mac> I've seen bga breakouts that looked like two merged circles with a via in one of them.
2012-10-04T09:58:08 -!- dekar [~dekar@212.255.31.84] has quit [Ping timeout: 240 seconds]
2012-10-04T09:59:22 < zyp> problem is that 13/27 is fucking huge vias
2012-10-04T10:00:59 < zyp> oshpark specs are good enough for 1mm pitch bgas, anything smaller will be a problem
2012-10-04T10:01:16 < emeb_mac> makes sense.
2012-10-04T10:08:45 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has quit [Quit: Quit]
2012-10-04T10:14:41 < dongs> sigh
2012-10-04T10:14:50 < dongs> so fucking busy all day didnt have time to throw chinkboard into autorouter
2012-10-04T10:15:01 < dongs> and todays pcbs are submiteted.
2012-10-04T10:16:21 < emeb_mac> always tomorrow.
2012-10-04T10:17:13 < emeb_mac> gn
2012-10-04T10:17:16 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-04T10:21:33 < dongs> wtf is this crap
2012-10-04T10:28:14 < dongs> ok i gotta figure out how to reverse this socket properly
2012-10-04T10:40:36 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-04T10:45:03 -!- Laurenceb_ [~Laurence@host86-136-118-159.range86-136.btcentralplus.com] has joined ##stm32
2012-10-04T10:48:43 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-04T10:48:56 -!- mrcan__ is now known as mrcan
2012-10-04T10:50:25 < dongs> looks like 8/16mil vias is just 20% extra on already cheapass pcb cost
2012-10-04T10:52:13 < zyp> that should fit nicely
2012-10-04T10:52:39 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 260 seconds]
2012-10-04T11:01:46 -!- Laurenceb_ [~Laurence@host86-136-118-159.range86-136.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-04T11:09:48 < dongs> time to see how these china leds compare to shit that was like $300/reel
2012-10-04T11:18:36 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-04T11:22:51 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-04T11:24:46 < dongs> http://www.uctronics.com/download/rgb_led.pdf ugh...
2012-10-04T11:26:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-04T11:26:27 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-04T11:27:38 < dongs> http://i.imgur.com/bTpIG.png is this correct?
2012-10-04T11:49:51 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-04T12:06:16 -!- claude is now known as claude_work
2012-10-04T12:45:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-04T12:45:55 < karlp> oh, tunilame actually had a real message for me, not just wildyly sending pm's asking for help.
2012-10-04T12:46:06 < karlp> now I feel a little bad.
2012-10-04T12:49:02 < zyp> huh?
2012-10-04T12:49:39 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-04T12:51:46 < karlp> jpa-: I meant, was it remote controlled? (with that rod in the picture?) I knew how the ball moved itself
2012-10-04T13:01:06 < Laurenceb> http://www.digikey.com/product-search/en/integrated-circuits-ics/embedded-microcontrollers/2556109?k=stm32f3
2012-10-04T13:01:10 < Laurenceb> they have prices
2012-10-04T13:02:57 < karlp> zyp he was trying to give me some feedback on http://false.ekta.is/2012/05/using-netbeans-for-stm32-development-with-stlink-texane/ for use with the newer netbeans
2012-10-04T13:03:29 < karlp> apparently now instead of "localhost:4242" you have to put "target localhost:4242" in a text box. or so he claims
2012-10-04T13:03:38 < karlp> what's the difference between psram and sram?
2012-10-04T13:04:05 < karlp> ahh, it's pseudo.
2012-10-04T13:04:24 < karlp> dram with onchip refresh
2012-10-04T13:04:27 < karlp> neat
2012-10-04T13:05:26 < dongs> huh
2012-10-04T13:05:29 < dongs> tehy removed 48lqfp version
2012-10-04T13:05:30 < dongs> cunts
2012-10-04T13:06:34 < dongs> http://www.digikey.com/scripts/dksearch/dksus.dll?vendor=0&keywords=stm32f3discovery
2012-10-04T13:06:37 < dongs> loll
2012-10-04T13:06:39 < dongs> "unable to get pricing"
2012-10-04T13:06:44 < dongs> shows ERROR for all price etc shits
2012-10-04T13:07:45 < dongs> http://i.imgur.com/gw7Tu.png
2012-10-04T13:08:28 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-04T13:08:30 < dongs> https://fbcdn-sphotos-a-a.akamaihd.net/hphotos-ak-prn1/46507_443051449067327_1154009182_n.jpg
2012-10-04T13:14:02 < cjbaird> Delicious Scones, from a recipe passed-down from a girlfriend who came from country Queensland :P .. http://images.4chan.org/ck/src/1349345582562.jpg
2012-10-04T13:21:15 < dongs> i made some apple cookie shits today t hat could be considered scones i guess. just not as thick.
2012-10-04T13:22:19 < dongs> http://i.imgur.com/Nlizq.jpg
2012-10-04T13:22:24 < dongs> wow, NIizq.jpg is nasty
2012-10-04T13:22:28 < dongs> fuck I and l looking same.
2012-10-04T13:22:41 < dongs> (was copypasting url from assdroid imgur uploader
2012-10-04T13:22:42 < zyp> nice reflow
2012-10-04T13:23:09 < dongs> lo
2012-10-04T13:26:54 < Laurenceb> why are the links to 4chan...
2012-10-04T13:27:06 < dongs> apparently thats cjbeard's image hosting service
2012-10-04T13:27:15 < cjbaird> because
2012-10-04T13:27:25 < dongs> he probably has a thread there
2012-10-04T13:27:25 < Laurenceb> its actually pron
2012-10-04T13:27:51 < Laurenceb> here are my "girlfriends" "scones"
2012-10-04T13:28:08 < cjbaird> with "cream"
2012-10-04T13:28:13 < cjbaird> and "jam"
2012-10-04T13:28:23 < cjbaird> ..and a nice hot cup of tea.
2012-10-04T13:28:58 < cjbaird> And I've just reminded myself of a video clip outtake of a Ron Jeremy porno. :P
2012-10-04T13:29:44 < cjbaird> Also, conveniently non-permanent links.
2012-10-04T13:30:02 < Laurenceb> his girlfriend is April Jones
2012-10-04T13:31:26 < karlp> cjbaird: care to share the scones recipe?
2012-10-04T13:31:43 < karlp> I'm quite happy with mine, using home made yogurt instead of milk,
2012-10-04T13:31:47 < karlp> should make some more again
2012-10-04T13:32:55 < Laurenceb>  home made yogurt
2012-10-04T13:33:02 < Laurenceb> i see what you did there
2012-10-04T13:33:39 < jpa-> karlp: ah, yeah, simple RF modules
2012-10-04T13:34:09 < cjbaird> karlp: this one was done more to test out the 'light fluffy scone' promise in the article: http://www.smh.com.au/lifestyle/homestyle/blogs/tried-and-tasted/how-to-bake-the-perfect-scone-20110504-1e7xn.html
2012-10-04T13:36:48 < cjbaird> ...which worked out. I went with mixing the dough with a knife.
2012-10-04T13:37:14 < cjbaird> Me making "yogurt", NSFW: http://i.imgur.com/AvGzp.jpg
2012-10-04T13:38:23 < jpa-> karlp: and yeah, LSM303 in the rod and in the ball, it tries to go to the same magnetic direction as where the rod is pointing
2012-10-04T13:38:32 < cjbaird> Definitely the way I'll be making bulk quantities of cream cheese in the future. Fuck paying $15/kilo for the supermarket stuff.
2012-10-04T13:39:20 < Laurenceb> lol @ NSFW
2012-10-04T13:40:41 < dongs> made Laurenceb click
2012-10-04T13:41:03 < Laurenceb> http://www.st.com/internet/com/MULTIMEDIA_RESOURCES/IMAGE/BOARD_PHOTO/stm32f3discovery.jpg
2012-10-04T13:41:07 < Laurenceb> high res photo now
2012-10-04T13:41:16 < Laurenceb> looks like the leds are some sort of compass thingy
2012-10-04T13:41:24 < dongs> yes..
2012-10-04T13:41:59 < Laurenceb> seems to run usb with internal osc?
2012-10-04T13:42:34 < karlp> cjbaird: so that was gloria's recipe in the 4chan picture?
2012-10-04T13:43:02 < karlp> jpa-: what rf modules are you using?
2012-10-04T13:43:31 < Laurenceb> http://www.youtube.com/watch?v=mpfZKYTQWk8
2012-10-04T13:43:32 < karlp> what'd you use powde3red milk for in your yogurt?
2012-10-04T13:43:32 < Laurenceb> omg
2012-10-04T13:43:57 < zyp> Laurenceb, interesting
2012-10-04T13:44:12 < zyp> though, might be that pic is just of a board without crystal fitted
2012-10-04T13:44:26 < Laurenceb> check the video
2012-10-04T13:44:33 < dongs> zyp, no
2012-10-04T13:44:39 < dongs> shipping version has no c rysal either
2012-10-04T13:45:54 < cjbaird> karlp: yup, the 3rd one. If the scones are large ("Devonshire tea sized"), they'll need ~19 minutes at 220C
2012-10-04T13:46:17 < zyp> Laurenceb, video doesn't show usb in use
2012-10-04T13:46:24 < Laurenceb> true
2012-10-04T13:46:29 < Laurenceb> but it should work
2012-10-04T13:46:34 < Laurenceb> given the rc tolerance
2012-10-04T13:47:58 < cjbaird> karlp: Re: powdered milk. I just had a small amount of it left over that wasn't enough for breadmaking, so I used it to prop up the milk solids.
2012-10-04T13:48:01 < dongs> video shows that it just came out of hte fucking package
2012-10-04T13:48:05 < karlp> huh, with an egg.
2012-10-04T13:48:15 < dongs> why does a fucking pole has f3discovery and i dont
2012-10-04T13:48:20 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-04T13:48:42 < cjbaird> http://pulkomandy.tk/_/_Electronique/_Discovering%20the%20STM32F3%20Discovery
2012-10-04T13:48:52 < Laurenceb> http://www.kamami.pl/index.php?ukey=product&productID=194378
2012-10-04T13:50:12 < cjbaird> When did video production tools make it behind the Iron Curtain? :P
2012-10-04T13:51:55 < zyp> Laurenceb, stm32f303 datasheet says «the clock source must use a HSE crystal oscillator
2012-10-04T13:52:14 < Laurenceb> that sucks
2012-10-04T13:52:19 < Laurenceb> i was reading F1 datasheet
2012-10-04T13:52:22 < jpa-> karlp: http://quasaruk.co.uk/acatalog/info_QAM_TX3_433_S.html  http://quasaruk.co.uk/acatalog/info_QAM_RX2_433.html    these
2012-10-04T13:52:31 < zyp> Laurenceb, doesn't it say the same?
2012-10-04T13:52:41 < Laurenceb> i didnt spot it
2012-10-04T13:53:30 < karlp> what are you doing to powerdown the ball? how long's it run? where is the ball code again?
2012-10-04T13:53:50 < jpa-> https://svn.kapsi.fi/jpa/sphere2/   is the code
2012-10-04T13:53:56 < karlp> sexy, thanks
2012-10-04T13:54:13 < dongs> ball code?
2012-10-04T13:54:17 < jpa-> powerdown is stm32 to sleep + everything off except accelerometer at 25Hz; then the accelerometer gives interrupt to wake it up
2012-10-04T13:54:46 < karlp> oh, nice
2012-10-04T13:55:00 < jpa-> it has 3000mAh (probably more like 2000mAh) li-ion cell; current usage is somewhere around 500mA when running & steering 
2012-10-04T13:55:03 < karlp> so src_tx is the code onthe bar?
2012-10-04T13:55:10 < jpa-> yeah
2012-10-04T13:55:14 < karlp> and src_rx is the ball
2012-10-04T13:55:16 < jpa-> and src_common is.. well common :)
2012-10-04T13:55:20 < karlp> right.
2012-10-04T13:55:20 < jpa-> yep
2012-10-04T13:55:37 < karlp> oh, so the bar has pwoerdown stuff too I guess,
2012-10-04T13:55:49 < karlp> because the ball is meant to just copythe bar in all ways
2012-10-04T13:55:53 < karlp> cool toy :)
2012-10-04T13:55:58 < Laurenceb> oh yeah
2012-10-04T13:56:01 < Laurenceb> that sucks
2012-10-04T13:56:10 < karlp> does it have variably speed basedon the degree of tilt of the bar?
2012-10-04T13:56:14 < Laurenceb> i still want to try running off internal RC :P
2012-10-04T13:56:17 < jpa-> karlp: yeah, the bar just wakes up on the button
2012-10-04T13:56:25 < jpa-> karlp: and yeah, speed is variable
2012-10-04T13:56:42 < zyp> Laurenceb, FS USB requires an accuracy of 0.25%
2012-10-04T13:56:52 < jpa-> though as is typical for humans, atleast i end up running it at full speed all the time :D
2012-10-04T13:56:57 < Laurenceb> yeah
2012-10-04T13:57:01 < karlp> 500mA when running? I didn't think motors needed that much.
2012-10-04T13:57:04 < Laurenceb> Rc might just work
2012-10-04T13:57:09 < Laurenceb> ball?
2012-10-04T13:57:10 < karlp> but I have amost zero motor experience
2012-10-04T13:57:11 < Laurenceb> what is this
2012-10-04T13:57:13 < zyp> and HSI seem to be user trimmable in 1% steps
2012-10-04T13:57:19 < jpa-> karlp: servo already takes 200mA
2012-10-04T13:57:20 < zyp> on F3
2012-10-04T13:57:26 < Laurenceb> i thought it was better than that..
2012-10-04T13:57:31 < jpa-> karlp: and it is a geared motor
2012-10-04T13:57:44 < dongs> its better thatn tarduino for sure
2012-10-04T13:57:57 < jpa-> but gtg now, be back in a few hours
2012-10-04T13:58:12 < zyp> so USB on HSI doesn't seem to be possible in a sane way .p
2012-10-04T13:58:37 < Laurenceb> the clock source must use a HSE crystal oscillator
2012-10-04T13:58:38 < karlp> jpa-: what's "swdflash" ? (when you get back)
2012-10-04T13:58:39 < Laurenceb> oops
2012-10-04T13:58:41 < zyp> of course, there is nothing stopping you from trying, if you are lucky you might hit the 0.25% window :p
2012-10-04T13:59:10 < Laurenceb> nope
2012-10-04T13:59:24 < Laurenceb> 0x10==1%
2012-10-04T13:59:45 < dongs> NO WAY
2012-10-04T13:59:46 < karlp> jpa-: and what's with appending blank spaces to the end of the bin file?
2012-10-04T13:59:48 < dongs> http://jp.mouser.com/ProductDetail/STMicroelectronics/STM32F373CCT6/?qs=%2fha2pyFadugVtlbo%2fWUwLMv5LGFGppCQBddAWnCJtXeg%252ba9EqyZlEg%3d%3d
2012-10-04T13:59:51 < dongs> !!!!
2012-10-04T14:00:13 < zyp> Laurenceb, I'm just quoting «HSI user trimming step» from page 75 in F303 datasheet
2012-10-04T14:00:16 < karlp> mouser, because digikey will always fuck you over
2012-10-04T14:00:24 < Laurenceb> oh sorry im reading F1
2012-10-04T14:00:28 < Laurenceb> F1 seems ok
2012-10-04T14:00:58 < dongs> fuck
2012-10-04T14:01:02 < dongs> mouser parametric search sucks cock
2012-10-04T14:01:07 < dongs> how cna i filter by pcakage type???
2012-10-04T14:01:08 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-04T14:01:13 < karlp> zyp: if HSI is specced to be 1% accurate, then what's the point in 1% steps in user trimming?
2012-10-04T14:01:26 < zyp> karlp, it's not
2012-10-04T14:02:51 < Laurenceb> sounds worse than F1 then, that sucks
2012-10-04T14:05:49 < dongs> ok ordered!
2012-10-04T14:07:03 < dongs> ok, count decreased on site
2012-10-04T14:07:08 < dongs> so its realtime unlike digikey
2012-10-04T14:07:12 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-04T14:07:15 < dongs> so im probably actually getting them.
2012-10-04T14:07:29 < zyp> nice
2012-10-04T14:07:43 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-04T14:07:49 < Laurenceb> http://uk.mouser.com/ProductDetail/STMicroelectronics/STM32F373CCT6/?qs=sGAEpiMZZMts0YcPTleK%252bfFyenw%2f%252boXO
2012-10-04T14:07:52 < Laurenceb> awesome
2012-10-04T14:08:00 < Laurenceb> but thats the boring 16bit adc one
2012-10-04T14:09:32 < dongs> boring?
2012-10-04T14:09:44 < dongs> 373 is the one wiht sdadc isnt it?
2012-10-04T14:09:49 < Laurenceb> yes
2012-10-04T14:09:51 < dongs> so?
2012-10-04T14:09:54 < dongs> and 303 is the one without
2012-10-04T14:10:02 < Laurenceb> well the fast adc and opamps is nicer
2012-10-04T14:10:08 < dongs> right but not for my application
2012-10-04T14:10:11 < dongs> this is drop-in for F1
2012-10-04T14:10:12 < Laurenceb> heh
2012-10-04T14:10:17 < dongs> on my trollboard
2012-10-04T14:10:33 < Laurenceb> ah, that should be exciting
2012-10-04T14:10:44 < dongs> apparently its code and api compatible!
2012-10-04T14:10:50 < dongs> so i should be able to just flash it
2012-10-04T14:10:52 < dongs> for MAX MAX
2012-10-04T14:10:53 < dongs> er
2012-10-04T14:10:54 < dongs> MAX HAX
2012-10-04T14:10:55 < Laurenceb> i2c fail compatible
2012-10-04T14:10:58 < dongs> lo
2012-10-04T14:10:59 < dongs> yes
2012-10-04T14:11:00 < Laurenceb> max heil
2012-10-04T14:11:01 < dongs> and USB fail compatible
2012-10-04T14:11:15 < Laurenceb> mad max
2012-10-04T14:12:57 < zyp> which is code compatible? pretty sure it's not
2012-10-04T14:13:43 < dongs> zyp, apparently it is
2012-10-04T14:13:51 < dongs> why would it not be?
2012-10-04T14:13:54 < dongs> M4 runs M3 code right?
2012-10-04T14:13:57 < dongs> just no fp
2012-10-04T14:14:02 < dongs> and peripherals are same
2012-10-04T14:14:05 < zyp> no
2012-10-04T14:14:09 < dongs> no?
2012-10-04T14:14:25 < zyp> both F303 and F373 are using the new sane AF mapping
2012-10-04T14:14:32 < dongs> uh
2012-10-04T14:14:35 < dongs> no way.
2012-10-04T14:14:39 < dongs> u be trollen
2012-10-04T14:14:42 < dongs> thats not "api compatible"
2012-10-04T14:15:06 < zyp> exactly
2012-10-04T14:15:39 < zyp> and that's a good thing, because F1 AF stuff is just bullshit
2012-10-04T14:15:46 < karlp> so what, f2, f3, f4, f0? and l1 all use the new af stuff
2012-10-04T14:15:53 < karlp> just f1 out in the cold?
2012-10-04T14:16:09 < dongs> zyp, thats fucking bullshit
2012-10-04T14:16:11 < zyp> yes, everything newer than F1 use the new AF stuff
2012-10-04T14:16:11 < dongs> wehre does it say this
2012-10-04T14:16:34 < dongs> RM313?
2012-10-04T14:16:36 < zyp> dongs, both F303 and F373 datasheets have AF tables with AF numbers and so on
2012-10-04T14:17:02 < dongs> teh AF0/AF1 etc shit?
2012-10-04T14:17:16 < dongs> o'AFRLy
2012-10-04T14:18:15 < dongs> fuck looks like it.
2012-10-04T14:18:17 < dongs> that sucks.
2012-10-04T14:18:39 < dongs> FUCK STM32F3; STM32F3 SUX; STM32F3 IS DYING; STM32F3 IS DEAD TO ME; STM32F3 HIT WTC
2012-10-04T14:19:19 < zyp> nah, it's F1 that sucks.
2012-10-04T14:37:50 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-04T14:44:15 < Laurenceb> lol
2012-10-04T14:52:32 < Bird|lappy> on the 10pin JTAG/SWD connector: is the keying notch/tab on the same side as or the opposite side from pin 1?
2012-10-04T14:59:08 < dongs> um i think same
2012-10-04T14:59:19 < dongs> i dont use the keyed connector
2012-10-04T14:59:32 < dongs> appy> on the 10pin JTAG/SWD connector: is the keying notch/tab on the same side as or the opposite side f
2012-10-04T14:59:36 < dongs> err
2012-10-04T14:59:36 < dongs> http://infocenter.arm.com/help/topic/com.arm.doc.faqs/attached/13634/cortex_debug_connectors.pdf
2012-10-04T14:59:48 < dongs> page 2
2012-10-04T15:03:33 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-04T15:06:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-04T15:13:24 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-04T15:34:08 < Laurenceb> http://www.silabs.com/products/mcu/lowpower/Pages/SiM3L1xx.aspx?mkt_tok=3RkMMJWWfF9wsRolsq%2FKZKXonjHpfsX56OwsUKC0lMI%2F0ER3fOvrPUfGjI4ATspmI%2FqLAzICFpZo2FFLE%2ByUfpM%3D
2012-10-04T15:34:12 < Laurenceb> built in dc-dc
2012-10-04T15:35:33 < zyp> well, so does F4
2012-10-04T15:35:55 < Laurenceb> huh
2012-10-04T15:36:03 < zyp> and 3.8V max isn't that impressive, is it?
2012-10-04T15:36:17 < Laurenceb> nope
2012-10-04T15:36:23 < Laurenceb> f4 has no dc-dc
2012-10-04T15:36:30 < Laurenceb> only ldo
2012-10-04T15:37:02 < zyp> does dc-dc imply switching?
2012-10-04T15:37:03 < BrainDamage> ldo is a dcdc, just not switching :p
2012-10-04T15:37:09 < jpa-> karlp: swdflash is just renamed stm32flash from stlink stuff.. i already had stm32flash the USART bootloader thing :P
2012-10-04T15:37:29 < zyp> BrainDamage, exactly
2012-10-04T15:38:09 < jpa-> karlp: and the spaces.. they were workaround for some crazy bug in stlink.. it wasn't writing the last partial word fully or something (e.g. if you had const char array it might not work).. not sure if it is fixed now
2012-10-04T15:38:50 < karlp> huh, interesting.
2012-10-04T15:39:05 < karlp> I don't remember ever even hearing about it.
2012-10-04T15:39:21 < jpa-> but it gave a warning about it so i figured it out quite quickly
2012-10-04T15:39:35 < jpa-> might be some stm32L or stlinkv1 related bug also
2012-10-04T15:40:50 < jpa-> (swdserver similarly renamed from stlink, back when it was still called just "stlink" or something)
2012-10-04T15:47:07 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-04T15:56:25 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-04T15:58:04 -!- izua [~izua@79.115.170.139] has joined ##stm32
2012-10-04T15:58:04 -!- izua [~izua@79.115.170.139] has quit [Changing host]
2012-10-04T15:58:04 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-04T15:58:07 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-04T16:02:57 < Laurenceb> if i have a ttyACM0 screen session it should have ~0 lag right?
2012-10-04T16:03:09 < Laurenceb> im getting about 2 seconds lag from somewhere...
2012-10-04T16:04:40 < zyp> lag how?
2012-10-04T16:04:58 < Laurenceb> i send something into the adc, it doesnt appear for a couple of seconds
2012-10-04T16:05:37 < zyp> you'll always have a millisecond or two due to usb framing
2012-10-04T16:05:41 < zyp> but not seconds.
2012-10-04T16:06:15 < Laurenceb> yeah
2012-10-04T16:06:28 < Laurenceb> some chibios buffer is filling up..
2012-10-04T16:12:29 < cjbaird> Laurenceb: is this on a *nix system? NetBSD has a stupid serial receive buffer interrupt setting that waits for several characters, rather than irq'ing after each char.
2012-10-04T16:13:48 < cjbaird> If you had a 2400 baud or faster connection, that was sending single characters only once in a while, you could get the same several-seconds-wait issue.
2012-10-04T16:15:46 < jpa-> Laurenceb: for me, it seems that if i write more than buffer size to chibios usb cdc, it won't flush the rest of the buffer until i write some more
2012-10-04T16:15:53 < jpa-> might be fixed in newer versions
2012-10-04T16:16:06 < karlp> cjbaird: yeah, but no-one uses netbsd, so who cares? ;)
2012-10-04T16:16:23 < Laurenceb> interesting
2012-10-04T16:16:26 < Laurenceb> this is ubuntu
2012-10-04T16:18:00 < Laurenceb> yeah about 2 second lag
2012-10-04T16:18:07 < cjbaird> http://dpaste.org/vSn8v/ ... lines14- is what I'm on able. I wonder if Linux has a similar trigger interrupt issue.
2012-10-04T16:18:28 < jpa-> Laurenceb: otherwise i'm getting 0 second lag when using the interactive shell
2012-10-04T16:18:46 < cjbaird> *what I'm on about
2012-10-04T16:19:05 < Laurenceb> hmm
2012-10-04T16:19:10 < Laurenceb> ill check the buffers
2012-10-04T16:19:28 < Laurenceb> the timing looks fine.. but it might be an issue on connect
2012-10-04T16:19:38 < Laurenceb> attach rather
2012-10-04T16:21:02 < jpa-> chibios usb cdc is a bit silly (well i'm not sure if there is a better way but still...)  when e.g. chprintf writes to it byte-by-byte, it sends the first byte as it's own packet, then the rest of the bytes are queued while it is sending, then the next time a byte arrives it sends all in the queue etc; so there are many very short packets
2012-10-04T16:21:07 < jpa-> not very bandwidth efficient
2012-10-04T16:22:12 < Laurenceb> lulwut
2012-10-04T16:22:29 < Laurenceb> im using chMBPostI ...
2012-10-04T16:22:34 < Laurenceb> i think thats wrong
2012-10-04T16:23:00 < Laurenceb> as its called from normal code
2012-10-04T16:23:07 < zyp> jpa-, well, that's a rather sane way of converting a byte stream to frames
2012-10-04T16:23:18 < jpa-> probably
2012-10-04T16:23:36 < zyp> the problem is that they are using a byte stream at all, rather than just using sprintf to a buffer and then sending the entire buffer at once
2012-10-04T16:37:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-04T16:38:05 -!- claude_work is now known as claude
2012-10-04T16:46:25 < Laurenceb> http://www.invensense.com/mems/gyro/mpu6500.html
2012-10-04T16:48:04 < dongs> old, no?
2012-10-04T16:56:24 < jpa-> not even available yet
2012-10-04T16:58:15 < dongs> ya but spec has been tehre for like half a year
2012-10-04T16:58:33 < dongs> PR was published in like july
2012-10-04T16:58:36 < dongs> fucking scammers
2012-10-04T16:58:44 < dongs> http://www.invensense.com/mems/gyro/documents/articles/071012-worlds-smallest-six-axis-motiontracking-device.html
2012-10-04T17:07:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-04T17:07:27 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-04T17:11:50 < Laurenceb> http://3.bp.blogspot.com/_v2pOgslN3m4/SiirI7mxUxI/AAAAAAAAYtE/syD3IHdwfrw/s400/LAZ_parking_meter_20090507_005.JPG
2012-10-04T17:17:24 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-04T17:18:28 -!- claude is now known as claude_work
2012-10-04T17:23:57 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-04T17:42:09 < cjbaird> I got my old KodakDC3200 serial camera working. Neat.
2012-10-04T17:43:13 < dongs> haha
2012-10-04T17:43:39 < dongs> cjbaird: i wrote some shit for kodak dc21x
2012-10-04T17:44:10  * Steffanx gets weird images in his head when dongs says "wrote some shit"
2012-10-04T17:44:50 < dongs> wow.. in like 1999.
2012-10-04T17:44:51 < cjbaird> The '3200 was like $500 when I got it..
2012-10-04T17:45:13 < cjbaird> I got mine in 2000, I think.
2012-10-04T17:45:22 <+Steffanx> The good old times
2012-10-04T17:45:46 < Laurenceb> in the days before pedobear
2012-10-04T17:46:17 < cjbaird> When alt.sex.teens was king.. -_-
2012-10-04T17:46:22 < Laurenceb> lol
2012-10-04T17:47:16 < Laurenceb> so how is April Jones?
2012-10-04T17:47:23 < Laurenceb> taking good care of her?
2012-10-04T17:48:14 < dongs> april who
2012-10-04T17:48:16 < cjbaird> Want my autograph?
2012-10-04T17:48:28 < Laurenceb> lurn 2 google
2012-10-04T18:01:35 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has joined ##stm32
2012-10-04T18:03:00 < cjbaird> Doubleneat, downloading from the camera still works. (95 seconds per image..)
2012-10-04T18:05:37 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-04T18:06:43 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-04T18:09:05 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-04T18:10:30 < ABLomas> bit offtopic but...
2012-10-04T18:10:42 < ABLomas> i have stepper motor (8 wires), missing wire labels
2012-10-04T18:10:46 < dongs> no lunix discussions
2012-10-04T18:10:52 < ABLomas> i could find ends of windings
2012-10-04T18:11:05 < ABLomas> but no idea how to find start/end of winding
2012-10-04T18:11:09 < ABLomas> any suggestions?
2012-10-04T18:11:28 < dongs> http://www.stepperworld.com/Tutorials/bidentify.gif
2012-10-04T18:11:31 < ABLomas> dongs: but UNIX is OK, right? =)
2012-10-04T18:11:34 < dongs> no
2012-10-04T18:11:51 < ABLomas> meh
2012-10-04T18:12:00 < dongs> what do you mean 'start end'
2012-10-04T18:12:07 < ABLomas> thanks for link, but that does not answer where's start and end
2012-10-04T18:12:07 < dongs> you can multimeter individual ends
2012-10-04T18:12:14 < dongs> sure it does
2012-10-04T18:12:22 < dongs> when you continuity check one
2012-10-04T18:12:27 < ABLomas> its not brushless motor where
2012-10-04T18:12:29 < dongs> it means you found one of the 4 coils
2012-10-04T18:12:31 < ABLomas> yes
2012-10-04T18:12:37 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-04T18:12:40 < ABLomas> i could find all coils easily
2012-10-04T18:12:47 < ABLomas> thats not a problem
2012-10-04T18:12:54 < ABLomas> problem is - where's "start" and "end"
2012-10-04T18:13:05 < ABLomas> it won't turn OK even with one winding reversed
2012-10-04T18:13:33 < ABLomas> and there's 12 combinations in total (maybe more), bit too much to check them all
2012-10-04T18:14:08 < ABLomas> i was thinking that i could apply voltage to any wind and check direction, but i was wrong, it could turn to both sides
2012-10-04T18:14:25 < ABLomas> so, still no idea how to check "start" and "end"
2012-10-04T18:15:37 < zyp> wtf is start and end?
2012-10-04T18:16:28 < ABLomas> start of winding, end...
2012-10-04T18:16:40 < ABLomas> at least, controller i used has such notation
2012-10-04T18:16:49 < dongs> you meanlike +, -? or wat
2012-10-04T18:16:53 < ABLomas> A_start, A_end, B_start, B_end
2012-10-04T18:17:21 < ABLomas> and all wires on motor are marked, like "A-", "A+"
2012-10-04T18:17:28 < ABLomas> so probably yes, "+" and "-"
2012-10-04T18:17:45 < zyp> identify each pair first, get four channel scope, hook one end of each coil to gnd, hook other end of each to a channel, spin the motor, observe pattern
2012-10-04T18:18:06 < dongs> thats a good idea
2012-10-04T18:18:20 < ABLomas> uh
2012-10-04T18:18:44 < ABLomas> spinning that 3kg stepper isn't really good idea, same for four channel scope
2012-10-04T18:18:47 < ABLomas> but thanks =)
2012-10-04T18:21:12 < cjbaird> I suppose it could even be done with a single moving-coil ammeter, using it to determine how two coils are sequenced relative to another.
2012-10-04T18:21:56 < ABLomas> 0yes, i'm currently doing that ;-)
2012-10-04T18:22:11 < dongs> 3kg stepper sounds hardcore
2012-10-04T18:22:13 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-04T18:22:38 < ABLomas> homemade cnc, used everything what got cheap
2012-10-04T18:23:17 < dongs> emeb: 20% extra of already dirtcheap pcb  cost to do 8/16 vias.
2012-10-04T18:23:38 < dongs> assembly/test is not an issue if a testjig exists.
2012-10-04T18:23:51 < cjbaird> A friend-of-a-friend has a source for ~50 motor-adjustable hospital beds.. The steppers in that are good for like a kN/metre of torque.
2012-10-04T18:23:59 < dongs> sweet
2012-10-04T18:24:07 < cjbaird> Make a Gundam! :D
2012-10-04T18:24:17 < dongs> or a Gangam
2012-10-04T18:24:25 < emeb> dongs: thx for checking. That's not too bad.
2012-10-04T18:38:07 < ABLomas> i guess no way to check start and end using only multimeter - i'm getting AC from stepper, does not matter rotation direction
2012-10-04T18:38:43 < ABLomas> going to try combos on CNC controller, would be nice if i won't fry it =)
2012-10-04T18:40:56 -!- daku is now known as DaKu
2012-10-04T18:42:14 -!- barthess [~barthess@164.177.220.3] has joined ##stm32
2012-10-04T18:45:49 -!- DaKu is now known as daku
2012-10-04T18:45:55 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-04T18:46:55 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-04T18:47:19 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-04T18:49:30 < karlp> huh, f050 is still more expensive than f100
2012-10-04T18:50:42 < emeb> odd that
2012-10-04T18:50:54 < emeb> is Pdiss lower?
2012-10-04T18:52:00 < karlp> not a concern in my application
2012-10-04T18:52:18 < karlp> where PDiss == power dissipation?
2012-10-04T18:52:35 < emeb> yeah. maybe ST think they can charge more for better efficiency.
2012-10-04T18:53:01  * karlp shrugs
2012-10-04T18:53:02 -!- daku is now known as DaKu
2012-10-04T18:53:21 < karlp> I thought it was twin selling points of cheaper _and_ lower power.
2012-10-04T18:53:49 < emeb> as time -> infinity I guess.
2012-10-04T18:55:17 < dongs> < karlp> where PDiss == power dissipation?
2012-10-04T18:55:23 < dongs> how much you get dissed for using it
2012-10-04T18:55:53 < karlp> well, I was only looking at it because I thought it would be cheaper than f100c4
2012-10-04T18:56:12 < emeb> should be cheaper - fewer instructions.
2012-10-04T18:56:36 < zippe> cost is a function of demand, mostly
2012-10-04T18:56:55 < karlp> which is why c4 is cheapest, but c8 is often cheaper than c6
2012-10-04T18:56:55 < zippe> more demand -> more opportunity to recoup NRE -> lower cost
2012-10-04T18:57:19 < dongs> http://www.digikey.com/product-detail/en/74ACT715SC/74ACT715SC-ND/211854 what is this kinda shit good for?
2012-10-04T18:57:52 < emeb> these days? not much.
2012-10-04T18:58:08 < emeb> considering the data sheet says it came out 24 years ago.
2012-10-04T18:58:17 < dongs> haha
2012-10-04T18:58:30 < dongs> i was wonderingwhy datasheet looked like some scanned garbage
2012-10-04T18:58:59 < emeb> DK should remainder all that stuff to Jameco.
2012-10-04T18:59:25 < emeb> Jameco is sitting on a warehouse full of parts from the 70s and wonders why people don't buy from them any more.
2012-10-04T19:00:28 < dongs> so how does a nigga in 2012 generate video sync
2012-10-04T19:00:35 < dongs> i guess even stm32 can do it?
2012-10-04T19:00:45 < emeb> bitbang?
2012-10-04T19:00:47 < emeb> fpga?
2012-10-04T19:00:54 < emeb> clever timer programming?
2012-10-04T19:01:30 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-04T19:01:30 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-04T19:01:30 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-04T19:01:31 < dongs> lets see if theres some timer AF on this shit.
2012-10-04T19:03:29 < dongs> TIM3_CH2!
2012-10-04T19:03:37 < dongs> gr8, maybe i can hax0r somethin up then.
2012-10-04T19:09:21 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-04T19:12:55 < Laurenceb> error: missing binary operator before token "("
2012-10-04T19:12:59 < Laurenceb> ^what causes that?
2012-10-04T19:13:10 < Laurenceb> #if (TARGET_ADC)>59000000UL
2012-10-04T19:14:12 < dongs> ..
2012-10-04T19:14:14 < dongs> )>
2012-10-04T19:14:26 < dongs> you should move the parenthesis at the end you know
2012-10-04T19:17:01 < Laurenceb> it doesnt like sqrt in preprocessor stuff :(
2012-10-04T19:17:02 < Laurenceb> error: floating constant in preprocessor expression
2012-10-04T19:17:19 < zyp> of course it doesn't, you're being stupid
2012-10-04T19:17:52 < Laurenceb> howso
2012-10-04T19:19:01 < zyp> it's just a dumb preprocessor
2012-10-04T19:19:13 < zyp> not a floating point calculator
2012-10-04T19:19:17 <+Steffanx> No, it also executes functions
2012-10-04T19:20:23 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-10-04T19:28:40 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-10-04T19:29:46 < Thorn> anyone done ethernet (with builtin MAC (stm32 or other cortex-m), not enc28j60 etc.)?
2012-10-04T19:30:05 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-04T19:30:05 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-04T19:30:05 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-04T19:30:42 < dongs> I heard TI parts have mac+phy on chip.
2012-10-04T19:31:00 < Thorn> yeah noticed that today while browsing
2012-10-04T19:35:19 < Thorn> akso found a thread on mbed forum where someone did a board and ethernet didn't work. lots of whining, no attempts to debug
2012-10-04T19:35:48 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-04T19:38:01 < emeb> Should just start out with one of these to get the SW going: https://www.olimex.com/Products/ARM/ST/STM32-E407/
2012-10-04T19:38:25 < emeb> Once that's sorted, do a custom board.
2012-10-04T19:38:32 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-04T19:38:52 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-04T19:38:55 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-04T19:39:07 < Thorn> I have an f107 board from them, maybe the MAC core is the same as in f4 (good old synopsys)
2012-10-04T19:40:08 < Thorn> >ARDUINO compatible platform connector
2012-10-04T19:40:17 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-04T19:40:40 <+Steffanx> What else you need?
2012-10-04T19:41:38 < Thorn> something like USB Complete but about ethernet, i guess
2012-10-04T19:41:59 < emeb> Suggest it to Jan.
2012-10-04T19:49:59 < zyp> I'm going to try out ethernet on my waveshare board when I've got time for it
2012-10-04T19:50:10 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Changing host]
2012-10-04T19:50:10 -!- dfletcher [~fletch@drupal.org/user/72475/view] has joined ##stm32
2012-10-04T19:50:25 < zyp> maybe I'll do CDC EEM or something like that
2012-10-04T19:51:17 -!- dfletcher is now known as KrytenSeries4000
2012-10-04T19:51:46 < Thorn> it seems at least as complicated as usb, I wonder why. all it does is send/receive packets. no enumeration, no endpoints etc. (you supply your own tcp/ip stack for all that)
2012-10-04T19:52:13 < zyp> it doesn't really seem very complicated to me
2012-10-04T19:53:00 < Thorn> well I actually need to read the docs through first tbh.
2012-10-04T19:54:32 < dongs> https://www.youtube.com/watch?v=FQm7k4riCoE ...
2012-10-04T19:58:16 < emeb> NY, LA only. :(
2012-10-04T19:59:25 < dongs> is it?
2012-10-04T19:59:44 < dongs> someone pick that shit up for me ill pay shipping
2012-10-04T20:00:06 < dongs> emeb: where area you getting this info
2012-10-04T20:00:12 -!- claude_work is now known as claude
2012-10-04T20:01:33 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-10-04T20:02:15 < Thorn> http://www.amazon.com/Embedded-Ethernet-Internet-Complete-Axelson/dp/1931448000/ref=sr_1_7?s=books&ie=UTF8&qid=1349369309&sr=1-7
2012-10-04T20:06:02 < emeb> dongs: http://www.theverge.com/2012/10/4/3452890/entertainment-weekly-cw-ad-smartphone-teardown
2012-10-04T20:07:26 < dongs> o, lame
2012-10-04T20:07:29 < dongs> only 1000.
2012-10-04T20:07:34 < emeb> yep
2012-10-04T20:07:42 < dongs> i guess they didnt wanna waste $.
2012-10-04T20:07:43 < dongs> hEh
2012-10-04T20:08:17 < emeb> yeah - surgically targeted ads. Not for us hackers, more for the industry wonks.
2012-10-04T20:08:46 < dongs> Que> Entertainment Weekly
2012-10-04T20:08:47 < dongs> Que> sounds like a magazine for new york homosexuals
2012-10-04T20:08:51 < dongs> he wasnt far off.
2012-10-04T20:13:49 -!- KrytenSeries4000 is now known as davelister
2012-10-04T20:13:58 -!- DaKu is now known as daku
2012-10-04T20:14:12 -!- davelister is now known as dlister
2012-10-04T20:15:03 -!- dlister is now known as dfletcher
2012-10-04T20:16:33 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-04T20:16:46 -!- dfletcher is now known as KrytenSeries4k
2012-10-04T20:23:06 -!- KrytenSeries4k is now known as acerimmer
2012-10-04T20:29:31 < emeb> dude's got a lotta nicks.
2012-10-04T20:29:42 < emeb> must be a red dwarf fan.
2012-10-04T20:33:03 < acerimmer> it wouldn't let me be lister so I was stuck with kryten until I finally settled :P
2012-10-04T20:33:21 < acerimmer> s/it/nickserv/g
2012-10-04T20:33:43 < acerimmer> and heh yes. today's the day! season X :)
2012-10-04T20:37:39 < acerimmer> derp probably should have used whois command for that. oh well. it's Red Dwarf day nothing could possibly be bad today.
2012-10-04T20:38:07 < emeb> I've got a season on dvd I haven't watched yet.
2012-10-04T20:38:42 < emeb> they used to run it on the local PBS station late saturday nights. That was long ago though.
2012-10-04T20:41:13 < acerimmer> heh nice that's how I'd found Monty Python and Fawlty Towers in the 80s
2012-10-04T20:41:20 < acerimmer> WTTW Chicago was mine
2012-10-04T20:41:41 < acerimmer> they played those before Doctor Who. PBS used to be so awesome.
2012-10-04T20:41:51 < emeb> used to
2012-10-04T20:41:55 < acerimmer> ;)
2012-10-04T20:42:26 <+Steffanx> Ha mr bumbleb
2012-10-04T20:43:41 < emeb> eric the half-a-bee?
2012-10-04T20:44:49 < emeb> Kemal Ataturk had an entire menagerie, all called Abdul!
2012-10-04T20:48:34 < Tectu> Steffanx
2012-10-04T20:50:32 <+Steffanx> Tectu,
2012-10-04T20:50:55 < dongs> kawaiiTectu
2012-10-04T20:56:47 < emeb> be one with the cute.
2012-10-04T20:57:15 < dongs> should really sleep. 3am.
2012-10-04T20:57:18 < dongs> gonna be fucked tomrorow.
2012-10-04T20:58:16 < emeb> dongs does more before 3am than most people do all day
2012-10-04T21:06:32 <+Steffanx> deja vu
2012-10-04T21:13:07 < Thorn> jpa-: your dso quad logic analyzer project has been mentioned in a russian magazine
2012-10-04T21:13:27 < emeb> starting to repeat myself. time to dial it back a few notches.
2012-10-04T21:35:00 < jpa-> Thorn: a paper magazine?
2012-10-04T21:35:20 < Thorn> yes. pretty good paper too
2012-10-04T21:35:24 < jpa-> nice
2012-10-04T21:35:37 < jpa-> a side mention in some dso quad review i assume?
2012-10-04T21:35:45 < Thorn> http://radio.ru/
2012-10-04T21:35:52 < Thorn> yes pretty much
2012-10-04T21:37:26 < jpa-> yeah, for some reason dso quad is getting quite much attention in russian forums etc.
2012-10-04T21:37:55 < Thorn> it's rather unusual material for them, they basically ignore cortex-m, seeedstudio and similar companies and other modern trends in amateur electronics (except arduino IIRC)
2012-10-04T21:40:15 < jpa-> no traffic spike yet :P
2012-10-04T21:40:38 < jpa-> but otherwise the dso quad stuff is giving me over thousand hits per month, more than any other content
2012-10-04T21:42:04 < Tectu> can I ask a question related to GDB/OpenOCD?
2012-10-04T21:42:34 < jpa-> why couldn't you?
2012-10-04T21:42:53 < jpa-> it's not like karlp is going to pop up and say "no! we only use texane stlink here!"
2012-10-04T21:44:00 < Tectu> thahaha
2012-10-04T21:44:12 < Tectu> I am not sure if I see through the logic which command does need monitor and which not...
2012-10-04T21:45:13 < jpa-> monitor is for commands passed through to openocd
2012-10-04T21:45:18 < jpa-> any normal gdb command is without it
2012-10-04T21:45:30 < Tectu> ah
2012-10-04T21:49:17 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has joined ##stm32
2012-10-04T21:50:02 < karlp> "no! we only use texane stlink here!"
2012-10-04T21:50:12 < karlp> (sorry, too tempting ;)
2012-10-04T21:50:53 < upgrdman_> im new to SPI. question: spi1 has an ISR, and there are flags you can check to determine why the ISR was called. one is RXNE (RX not empty) but wouldn't that always be the case after the mcu (master) transmits a byte since the communication is sync with a shared clock?
2012-10-04T21:51:56 < upgrdman_> in others words, after you write to SPI1->DR, to send a byte, RXNE would always get set, correct?
2012-10-04T21:52:47 < Thorn> not immediately though (and afair there is a transmit-only mode)
2012-10-04T21:53:00 < upgrdman_> ok
2012-10-04T21:53:05 < jpa-> yeah, there is a delay and also it could use slave mode
2012-10-04T21:54:03 < upgrdman_> ok, but to make sure i understand it: if the mcu is the master, and you are in tx/rx mode, shortly after any write to SPI1->DR (to send a byte), a new byte will always be received and RXNE will get set?
2012-10-04T21:54:26 < Thorn> you normally enable interrupts like this one only when you need them and disable when you don't
2012-10-04T21:55:37 < upgrdman_> ok. well im playing with the f4discov board, and it has a 3-axis accelerometer with SPI. i want to update 3 global variables that store the current acceleration in all three axes. i assume the most efficient way to do that is use ISRs?
2012-10-04T21:55:41 < jpa-> upgrdman_: yeah (but as a disclaimer i have to add that i haven't used the spi peripheral directly)
2012-10-04T21:56:00 < jpa-> upgrdman_: most efficient could be DMA
2012-10-04T21:56:01 < upgrdman_> ok
2012-10-04T21:56:06 < upgrdman_> hum
2012-10-04T21:56:16 < upgrdman_> ya, i need to read up on dma.
2012-10-04T21:56:26 < upgrdman_> short of DMA, ISRs would be best?
2012-10-04T21:56:45 < jpa-> start a 6-byte transfer, then you can transfer all the axes with CPU totally free to do other stuff in the meantime
2012-10-04T21:56:49 < jpa-> yeah
2012-10-04T21:57:43 < upgrdman_> jpa-: and that would be one byte at a time, waiting for TXE (TX buffer empty) in between each byte, correct?
2012-10-04T21:58:13 < jpa-> you mean ISR's? you could start the next transfer always in the ISR
2012-10-04T21:58:23 < jpa-> though that might mean a tiny pause between transfers
2012-10-04T21:58:35 < Thorn> there's an app note on how to toggle a direction pin for rs-485 using uart interrupts. it demonstrates a good interrupt technique (you have a state machine in an interrupt handler that enables/disables interrupts "on the fly"). I'm pretty sure you can do the same with spi, would be even easier
2012-10-04T22:00:08 < jpa-> upgrdman_: but if you are not going to do anything else, then you can just use a while() loop to wait on the flag
2012-10-04T22:06:19 -!- Laurenceb_ [~Laurence@host86-136-118-159.range86-136.btcentralplus.com] has joined ##stm32
2012-10-04T22:07:34 < Tectu> is there a way to write/use this as .c ?   https://github.com/mobyfab/ChibiOS-Examples/blob/master/STM32F4-LCD-GWEN/src/hacks.cpp
2012-10-04T22:09:44 < Thorn> why? does your C code call pure virtual functions?
2012-10-04T22:10:29 < jpa-> you could premangle the names and write it as C (well, for __cxa_pure_virtual() you don't even need to mangle).. but yeah, why would you want to do that?
2012-10-04T22:10:44 < Tectu> not sure :D
2012-10-04T22:10:47 < Tectu> to be honest
2012-10-04T22:11:39 < jpa-> here is another take on the same thing: https://svn.kapsi.fi/jpa/dsoquad/logic/cxxglue.cc
2012-10-04T22:11:52 < Tectu> what's that cxx stuff?
2012-10-04T22:11:59 < upgrdman_> jpa-: ya i was going to do that at first, and after i make sure my logic works i will move it into an ISR
2012-10-04T22:12:58 < jpa-> Tectu: c++ without using +-sign in filenames :P
2012-10-04T22:13:31 < Tectu> jpa-, lol
2012-10-04T22:13:34 < Tectu> jpa-, so like cpp ?
2012-10-04T22:13:37 < jpa-> yeah
2012-10-04T22:13:55 < jpa-> except purists would say that cpp means /usr/bin/cpp
2012-10-04T22:14:05 < Tectu> [tectu@TecTop vitfit]$ arm-none-eabi-size build/vitfit.elf
2012-10-04T22:14:05 < Tectu>    text	   data	    bss	    dec	    hex	filename
2012-10-04T22:14:06 < Tectu>   82916	    696	   7804	  91416	  16518	build/vitfit.elf
2012-10-04T22:14:18 < Tectu> wtf? is my "image" 100kB ?
2012-10-04T22:14:32 < Tectu> "image" -> the thing that gets into flash
2012-10-04T22:14:45 < jpa-> yes
2012-10-04T22:14:53 < Tectu> seriously?
2012-10-04T22:15:03 < Tectu> Start address 0x8000130, load size 83612
2012-10-04T22:15:03 < Tectu> Transfer rate: 22 KB/sec, 10451 bytes/write.
2012-10-04T22:15:07 < Tectu> and that?    -> 10kB ?
2012-10-04T22:15:25 < jpa-> no, "load size 83612"
2012-10-04T22:15:37 < Tectu> what's the bytes/write?
2012-10-04T22:15:42 < jpa-> (ok 80kB not 100kB if you want to be exact, bss doesn't go to ram)
2012-10-04T22:15:54 < jpa-> it is bytes per write call
2012-10-04T22:16:05 < Tectu> what does that mean?
2012-10-04T22:16:06 < jpa-> arm-none-eabi-objdump -t build/vitfit.elf | sort -k5   or something to see what takes the space
2012-10-04T22:16:25 < jpa-> Tectu: just some internal stuff about what sizes of blocks it writes or something
2012-10-04T22:16:49 < Thorn> I'm guessing newlib or something
2012-10-04T22:17:01 < jpa-> newlib could be the culprit
2012-10-04T22:17:07 < jpa-> replace it with baselibc :)
2012-10-04T22:17:27 < Tectu> jpa-, so I have even more issues? :D
2012-10-04T22:17:32 < Tectu> jpa-, how much would I save? 5k?
2012-10-04T22:17:47 < jpa-> depends on how much stuff you use from newlib
2012-10-04T22:17:58 < Tectu> jpa-, I don't know... I just don't want more trouble
2012-10-04T22:18:03 < Tectu> also, my STM32 does have 1MB of flash
2012-10-04T22:18:04 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-04T22:18:07 < jpa-> i have seen 10-20kB savings if using newlib malloc etc.
2012-10-04T22:18:07 < Tectu> so...
2012-10-04T22:18:08 < Tectu> 10% :D
2012-10-04T22:18:16 < jpa-> yeah, maybe not so important on stm32f4
2012-10-04T22:18:22 < Tectu> ah, wait, 512kB
2012-10-04T22:18:24 < Tectu> I'm on F1
2012-10-04T22:18:37 < jpa-> but faster flash times are nice
2012-10-04T22:19:34 < Tectu> well, it takes 4 secounds or something
2012-10-04T22:19:41 < Tectu> 22kB/s
2012-10-04T22:19:46 < Tectu> can life with that
2012-10-04T22:20:11 < emeb> what are you flashing with (stlink, bmp, etc)?
2012-10-04T22:20:50 < Tectu> emeb, oocd
2012-10-04T22:21:32 < emeb> Tectu: what hardware are you driving w/ oocd?
2012-10-04T22:21:54 < Tectu> emeb, jtagkey2
2012-10-04T22:21:54 < emeb> ie - olimex jtag-tiny, jlink, etc?
2012-10-04T22:21:58 < Tectu> emeb, why are you asking?
2012-10-04T22:22:16 < Tectu> it's a selfmade jtag device based on jtagkey2
2012-10-04T22:22:22 < emeb> Tectu: guy I know is worried about programming times on an F4 project
2012-10-04T22:22:32 < emeb> wants a fast programmer.
2012-10-04T22:22:52 < jpa-> stlinkv2 gives me 10kB/s with F4
2012-10-04T22:23:03 < Tectu> emeb, i'd say build a good jtag out of jtagkey2
2012-10-04T22:23:07 < Tectu> yeah, stlink won't get over 10
2012-10-04T22:23:23 < emeb> He thinks that if he spends $LARGE_NUMBER for a programmer that he'll be able to flash faster.
2012-10-04T22:23:56 < emeb> I haven't timed how fast my BMP will go, but it's worth checking.
2012-10-04T22:24:26 < Tectu> emeb, afaik BMP has something like 22kB/s as well
2012-10-04T22:24:27 < Tectu> ask zyp
2012-10-04T22:24:37 < ntfreak> Tectu: i get about 38-40 with stlink and f4 - last time i checked
2012-10-04T22:24:37 < ntfreak> are you using the new ftdi driver, as that should also be quicker overall
2012-10-04T22:25:15 < emeb> At some point you run up against the speed of the flash memory itself. No idea what that limit is.
2012-10-04T22:25:51 < Tectu> ntfreak, that's kinda nice
2012-10-04T22:26:09 < Tectu> ntfreak, can you tell me about the differences building OpenOCD with and without libftdi?
2012-10-04T22:27:33 < ntfreak> Tectu: sorry do not follow, do you mean about using libftdi or ftd2xx etc ?
2012-10-04T22:27:43 < Tectu> second one
2012-10-04T22:27:55 < Tectu> there's the option when configuring OpenOCD if you take ftd stuff as well, right?
2012-10-04T22:28:05 < Tectu> like for my distribution, there is OpenOCD and OpenOCD-libftdi
2012-10-04T22:28:13 < Tectu> one is build with, one without
2012-10-04T22:28:21 < Tectu> but both work with my ftdi chip
2012-10-04T22:30:45 < ntfreak> ftd2xx (closed ftdi) is generally quicker under windoze, linux about the same - last time i checked.
2012-10-04T22:30:45 < ntfreak> however we now have a new ftdi driver that is faster than them all - however it requires libusb-1.0
2012-10-04T22:30:45 < ntfreak> this is enabled by using --enable-ftdi
2012-10-04T22:31:58 < karlp> emeb and co, the flash times are not always related to stlink vs other programmers, but the programming algorithm used.
2012-10-04T22:32:30 < karlp> oocd has a better flash loader than texane/stlink does right now, so they get faster flashes, even when using an stlinkv2 hardware.
2012-10-04T22:37:55 < ntfreak> karlp: very true - when we added async algorithm to openocd we saw over 10KiB/sec increase.
2012-10-04T22:38:38 < emeb> karlp: makes sense.
2012-10-04T22:39:28 < Tectu> ntfreak, so i should go with   --enable-ftdi     instead of    --enable-ft2232_libftdi ?
2012-10-04T22:39:28 < jpa-> i wonder how i could get openocd working for me.. antivirus forced by company policy is blocking winusb driver (when i try to install it with zadig)
2012-10-04T22:39:32 < emeb> ntfreak: so openocd now can use stlink v2 hardware?
2012-10-04T22:39:35 < Tectu> ntfreak, on linux with libusbx
2012-10-04T22:39:39 < Tectu> emeb, yes
2012-10-04T22:39:46 < emeb> sweet!
2012-10-04T22:39:47 < jpa-> it would also be nice to keep the original stlink tool working
2012-10-04T22:40:30 < emeb> need to try that out at some point. don't need it, but it's good to have all the options.
2012-10-04T22:41:02 < ntfreak> emeb: yes it supports v1 and v2 hardware
2012-10-04T22:43:15  * emeb goes looking for the source
2012-10-04T22:45:09 < ntfreak> Tectu: yes to enable the new driver, the only other change is to the adapter config, so if you were using
2012-10-04T22:45:09 < ntfreak> source [find interface/jtagekey2.cfg] change to source [find interface/ftdi/jtagekey2.cfg]
2012-10-04T22:46:23 < ntfreak> this is until we are happy all is ok with the new driver then we can swap to usng it by default.
2012-10-04T22:46:24 < Tectu> ntfreak, good to know, let's give it a try
2012-10-04T22:48:11 < Tectu> ntfreak, getting 2kB/s more :D
2012-10-04T22:49:22 < ntfreak> the biggest increase is with reading memory
2012-10-04T22:50:17 < Tectu> okay
2012-10-04T22:50:20 < Tectu> i am off for some sleep now
2012-10-04T22:50:21 < Tectu> cu guys
2012-10-04T22:50:43 < Tectu> ntfreak, it's indeed way faster when doing examine
2012-10-04T22:51:05 < Tectu> out
2012-10-04T22:51:32 < BrainDamage> ntfreak: does openocd have now target scan or do you still have to configure the target?
2012-10-04T22:52:46 < emeb> hmmm. ./configure --enable-ftdi worked, but that option isn't listed in the README. baroo?
2012-10-04T22:55:42  * ntfreak adds to the todo list
2012-10-04T22:57:49 < Laurenceb_> http://upload.wikimedia.org/wikipedia/en/e/e1/Pedobearad.jpg
2012-10-04T22:59:57 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has quit [Ping timeout: 240 seconds]
2012-10-04T23:01:31 <+Steffanx> Get a life ( and no, not life.zip )
2012-10-04T23:02:05 < Laurenceb_> life.exe segfaulted many years ago
2012-10-04T23:03:07 < Thorn> exe segfaulted? why do you name your linux binaries .exe
2012-10-04T23:03:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-04T23:03:37 -!- barthess [~barthess@164.177.220.3] has quit [Quit: Leaving.]
2012-10-04T23:04:24 < jpa-> Thorn: wine :)
2012-10-04T23:05:48 < Laurenceb_> after my life segfaulted i started using linux
2012-10-04T23:06:26 < BrainDamage> how much did your glider weight?
2012-10-04T23:08:56 < Laurenceb_> the balloon dropped thingy?
2012-10-04T23:09:11 < Laurenceb_> about 600 grams
2012-10-04T23:12:00 < Laurenceb_> http://www.zug.com/static_pages/herbal/ropex/050208_ropex_doll2.jpg
2012-10-04T23:12:20 < Laurenceb_> ^meanwhile... on the internets
2012-10-04T23:21:29 < Thorn> nice keyboard http://vimeo.com/47678872
2012-10-04T23:23:33 < prog> hey that's a TI-85
2012-10-04T23:34:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-04T23:43:16  * Laurenceb_ has a Ti-83+
2012-10-04T23:43:25 < Laurenceb_> i wrote fft in basic :P
2012-10-04T23:43:40 < ntfreak> BrainDamage: yes and no - it will auto discover the jtag chain - however a config is required for flash etc.
2012-10-04T23:43:40 < ntfreak> most people use a cfg so they can validate chain
2012-10-04T23:50:54 <+dekar_> Laurenceb_, was it slow? :P
2012-10-04T23:51:20 < Laurenceb_> just a little id help if basic didnt do everything in 64bit floating point
2012-10-04T23:51:31 < Laurenceb_> on a 2 mhz Z80
2012-10-04T23:52:03 < Laurenceb_> around a minute to do a 4096point fft iirc
2012-10-04T23:53:06 < Laurenceb_> - largest that would fit in memory
--- Log closed Thu Oct 04 23:58:31 2012
--- Log opened Thu Oct 04 23:58:40 2012
2012-10-04T23:58:40 -!- jpa-_ [jpa@hilla.kapsi.fi] has joined ##stm32
2012-10-04T23:58:40 -!- Irssi: ##stm32: Total of 67 nicks [1 ops, 0 halfops, 3 voices, 63 normal]
2012-10-04T23:58:41 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-04T23:58:54 -!- jpa- [jpa@hilla.kapsi.fi] has quit [Read error: Operation timed out]
2012-10-04T23:59:32 -!- Irssi: Join to ##stm32 was synced in 58 secs
--- Day changed Fri Oct 05 2012
2012-10-05T00:00:44 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection]
2012-10-05T00:02:12 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-05T00:14:07 < karlp> F3
2012-10-05T00:18:41 < emeb> Laurenceb_: I once wrote an FFT in 6502 assembly.
2012-10-05T00:19:17 < emeb> 8-bit data, 256 points IIRC. Not terribly useful.
2012-10-05T00:49:34 < karlp> efm support just landed in libopencm3
2012-10-05T00:49:39 < karlp> whee!
2012-10-05T01:03:56 < qyx_> what could be the problem with eth mac hanging at reset?
2012-10-05T01:04:23 < qyx_> working in chibi, it enables clocks and then wait for ETH_DMABMR_SR to be cleaer
2012-10-05T01:04:25 < qyx_> cleared
2012-10-05T01:05:16 < qyx_> refman says it should clear automatically when the reset operation is completed
2012-10-05T01:06:55 < qyx_> ok, it works if i just comment out the line which resets eth mac
2012-10-05T01:10:49 <+dekar_> I really like all those new CMSIS apis (like CMSIS-RTOS), yet I wish they'd provide some kind of HAL
2012-10-05T01:13:06 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-05T01:13:06 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-05T01:13:06 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-05T01:14:27 < Thorn> there's also CMSIS-DAP http://www.keil.com/support/man/docs/dapdebug/dapdebug_introduction.htm
2012-10-05T01:15:18 < Thorn> an attempt at a standard debugging interface if I understand it right
2012-10-05T01:16:52 < karlp> thats looking vood
2012-10-05T01:17:11 < Thorn> http://www.brc-electronics.nl/projects/119-projects/144-cmsisdap <- superior soldering sk1llz
2012-10-05T01:17:53 < karlp> theres a keol guy in #openocx from keil, with permission to be workinv on cmsi-dap support
2012-10-05T01:18:34 < karlp> forgive my typing, im on my phone tonight
2012-10-05T01:19:37 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: Lost terminal]
2012-10-05T01:24:19 < Thorn> that's nice
2012-10-05T01:24:30 < Thorn> hardware seems pretty trivial too
2012-10-05T01:32:07 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-10-05T01:40:17 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-05T01:41:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-10-05T02:07:04 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-10-05T02:08:20 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-05T02:08:53 -!- Laurenceb_ [~Laurence@host86-136-118-159.range86-136.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-05T02:10:34 < emeb> Almost as trivial as the BMP. :)
2012-10-05T02:33:57 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-05T03:18:13 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-05T03:23:23 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-05T03:43:53 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-05T03:48:15 < dongs> can someone wiht arm troll account download that cmsis dap shit for me
2012-10-05T03:49:28 <+dekar_> dongs, I think it also comes bundled with the ST library
2012-10-05T03:49:37 <+dekar_> so just download the stm32f4 lib
2012-10-05T03:49:55 < dongs> does it?
2012-10-05T03:49:58 < dongs> the NEW f4 lib?
2012-10-05T03:50:04 <+dekar_> oh my bad, dap - I read DSP -.-"
2012-10-05T03:50:06 < dongs> yeah
2012-10-05T03:50:07 < dongs> not dsp
2012-10-05T03:57:50 < emeb> dongs: there is DSP in the stm32f4 cmsis libs. Lotsa filters, transforms, vector ops, etc.
2012-10-05T03:58:06 < dongs> dammit dekar and his inability to read
2012-10-05T03:58:16 < dongs> now he confused 2 people
2012-10-05T03:58:23 < qyx_> dsp?
2012-10-05T03:58:25 < qyx_> :X
2012-10-05T03:59:52 < emeb> http://www.st.com/internet/com/SOFTWARE_RESOURCES/SW_COMPONENT/FIRMWARE/stm32f4_dsp_stdperiph_lib.zip
2012-10-05T04:00:30 < emeb> Down inside -> STM32F4xx_DSP_StdPeriph_Lib_V1.0.1/Libraries/CMSIS/DSP_Lib
2012-10-05T04:01:09 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-05T04:03:19 < dongs> i already have 101
2012-10-05T04:03:19 < dongs> hmm
2012-10-05T04:03:37 < dongs> i did look through that stuff before, but it just looks like generic code, what is arm-specific about it?
2012-10-05T04:05:07 < dongs> i.e. where's the shit that uses "dsp" instructins on M4
2012-10-05T04:08:35 < dongs> ah theres some asm in the .h file
2012-10-05T04:08:47 < emeb> dongs: I think that they're hoping that the compiler will handle that part.
2012-10-05T04:08:55 < dongs> yea heh
2012-10-05T04:11:31 < emeb> gcc 4.6.2 actually doesn't do too bad a job.
2012-10-05T04:12:04 < emeb> but I've heard that 4.7 has even better support for the saturating math, etc.
2012-10-05T04:12:24 < dongs> theres a bunch of SIMD intrinsics but i dont see the mathlib actually using any of htem
2012-10-05T04:12:32 < dongs> i.e. shit like core_cm4_simd.h
2012-10-05T04:13:06 < dongs> what the hell the number of F3 chips on mouser hasnt changed
2012-10-05T04:13:08 < dongs> still 208
2012-10-05T04:13:15 < dongs> why isnt anyone buying them :(
2012-10-05T04:13:30 < emeb> waiting for the discovery boards to test things out?
2012-10-05T04:13:37 < dongs> ya but shit's drop in
2012-10-05T04:14:08 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Read error: No route to host]
2012-10-05T04:14:17 < emeb> DK still shows 0.
2012-10-05T04:15:10 < emeb> mmm - lqfp48.
2012-10-05T04:15:20  * emeb likes little pkgs.
2012-10-05T04:15:34 < dongs> http://uk.mouser.com/ProductDetail/STMicroelectronics/STM32F373CCT6/?qs=sGAEpiMZZMts0YcPTleK%252bfFyenw%2f%252boXO
2012-10-05T04:15:38 < dongs> thats what I picked up yeah.
2012-10-05T04:16:11 < emeb> too bad it's only got 32kB SRAM.
2012-10-05T04:16:17 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-05T04:16:39 < emeb> not enough for my reverb algorithm..
2012-10-05T04:17:30 < dongs> emeb: more than F1.
2012-10-05T04:17:40 < dongs> i can do black & white video overlay in 32k I think.
2012-10-05T04:17:51 < dongs> it was *barely* fitting into 20k on F1.
2012-10-05T04:18:04 < dongs> 10k per backbuffer.
2012-10-05T04:18:20 < emeb> flyback: whut?
2012-10-05T04:18:50 < emeb> what will you do with it now?
2012-10-05T04:19:07 < dongs> poop on it
2012-10-05T04:19:21  * emeb is glad he doesn't have to use sparcstations any more.
2012-10-05T04:19:57 < dongs> ok so im pissed now . since F373 is not so "api compatible" with things like uart, GPIO/AF etc
2012-10-05T04:20:04 < dongs> why teh FUCK did tehy keep the idiot F1 DMA controller??
2012-10-05T04:20:13 < emeb> oh, yay.
2012-10-05T04:20:24 < emeb> new! improved!
2012-10-05T04:20:29 < dongs> apparently uart has separate read/write DRs
2012-10-05T04:20:32 < emeb> rewrite all yer code!
2012-10-05T04:20:56 < dongs> AF stuff is F2/F4 style
2012-10-05T04:21:12 < dongs> some 'expert' claims I2C is all fucked up but I wont believe that until the shit is on a board
2012-10-05T04:25:18 < dongs> emeb: < dongs> http://i.imgur.com/bTpIG.png is this correct?
2012-10-05T04:25:36 < dongs> thats the best china datasheet i have on them.
2012-10-05T04:32:16 < emeb> that's the LED array?
2012-10-05T04:32:20 < dongs> ya
2012-10-05T04:33:13 < Bird|lappy> what are these claims about fubar I2C handling all about anyway?
2012-10-05T04:33:16 < emeb> I think I have a better one on the other machine.
2012-10-05T04:33:19 < dongs> i drew it on a piece of paper
2012-10-05T04:33:29 < dongs> trying to reverse it
2012-10-05T04:33:45 < emeb> bbiab...
2012-10-05T04:35:00 < dongs> http://i.imgur.com/3EHNR.png
2012-10-05T04:35:09 < dongs> this *should* work if its plugged in from the BACK of the board.
2012-10-05T04:35:29 < dongs> oops
2012-10-05T04:35:30 < dongs> no
2012-10-05T04:35:33 < dongs> 17/32 is reversd
2012-10-05T04:43:34 < dongs> hm i wonder if sparkfun datasheet looks better
2012-10-05T04:43:43 < dongs> wondering what kinda height that keepout tab is
2012-10-05T04:45:40 < dongs> nope no info there either
2012-10-05T04:45:40 < dongs> great
2012-10-05T04:54:28 -!- izua [~izua@unaffiliated/izua] has quit [Remote host closed the connection]
2012-10-05T04:59:04 < dongs> https://www.olimex.com/Products/OLinuXino/A13/A13-OLinuXino-WIFI-DEV/images/A13-OLinuXino-01.jpg
2012-10-05T04:59:08 < dongs> haha
2012-10-05T04:59:08 < dongs> TQFP A13
2012-10-05T05:05:35 < dongs> hmm.. i should just have 5V on same socket that connects the blocks
2012-10-05T05:05:38 < dongs> instead of separate
2012-10-05T05:05:47 < dongs> like sparkfun did.
2012-10-05T05:05:50 < dongs> https://dlnmh9ip6v2uc.cloudfront.net/images/products/7/6/0/00760-03-L.jpg
2012-10-05T05:06:03 < dongs> y/n/mare
2012-10-05T05:07:22 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-05T05:09:04 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-05T05:16:08 < upgrdman> if i need to extract, for example, the third digit (hundreds place) of an integer, is there a more efficient way to do it other than num / 100 % 10 ?
2012-10-05T05:16:29 < upgrdman> no hardware support for div or mod sucks :( but stm32f0 is cheap :)
2012-10-05T05:25:05 < dongs> https://groups.google.com/forum/?hl=en&fromgroups=#!msg/comp.os.msdos.programmer/1rNPLe4Lxog/eLZCJy9VvtUJ
2012-10-05T06:00:12 -!- Netsplit *.net <-> *.split quits: Bird|lappy, jaeckel, ossifrage
2012-10-05T06:05:38 -!- Netsplit over, joins: jaeckel
2012-10-05T06:06:26 -!- Netsplit over, joins: ossifrage
2012-10-05T06:08:15 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-05T06:17:24 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-05T06:17:38 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-05T06:17:45 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-05T06:25:51 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-05T06:44:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-05T06:59:43 < dongs> someone wanna write me a software I2C slave for STM32, with events/callbacks on every step? ill give you like 50bucks
2012-10-05T07:08:53 <+dekar_> slave… polling SCL? or using an interrupt for that?
2012-10-05T07:09:04 < dongs> fully software
2012-10-05T07:09:06 < dongs> not using i2c peripheral
2012-10-05T07:10:00 < dongs> but i want full control over events
2012-10-05T07:10:04 < dongs> and being able to reply at any address
2012-10-05T07:10:36 < dongs> incoming shit (from master) looks like start addr <variable length data> restart anotheraddr <variablelengthdata> stop
2012-10-05T07:10:48 < dongs> i want to be able to hook all those restart+start to be able to answer at new address
2012-10-05T07:10:58 < dongs> cant do this shit with the hw peripheral
2012-10-05T07:11:13 < dongs> performance doesnt matter as thats all teh chip will be doing
2012-10-05T07:13:06 <+dekar_> *cough* <variable length data> *cough*
2012-10-05T07:13:17 <+dekar_> addresses can only be 7bit or 10bit
2012-10-05T07:14:51 <+dekar_> till when do you need that? I'd have use for a i2c slave emulator as well
2012-10-05T07:15:49 < dongs> whenever is fine
2012-10-05T07:15:57 < dongs> yeah VARIABLE LENGTH DATA
2012-10-05T07:15:59 < dongs> im not fucking joking
2012-10-05T07:16:02 < dongs> thats why its fucking aids.
2012-10-05T07:16:23 < dongs> as in it can be 2 bytes or 3 bytes etc.
2012-10-05T07:16:25 <+dekar_> oh I got you wrong, that <variable length data> is for addressing data in the slave right? like for those i2c-eeproms
2012-10-05T07:16:35 < dongs> right, but theres no subaddress
2012-10-05T07:16:41 < dongs> its just raw data, but it varies dependign on waht packet it is
2012-10-05T07:16:45 < dongs> but I dont actually KNOW the length of it
2012-10-05T07:16:49 < dongs> until I receive another restart!
2012-10-05T07:16:56 < dongs> so any hardware implementation of doing this is fucked
2012-10-05T07:17:12 < Bird|lappy> dongs: is there an upper bound on what the master can send?
2012-10-05T07:17:15 <+dekar_> a logic analyzer could do what you want :D
2012-10-05T07:17:22 <+dekar_> Bird|lappy, I don't think so
2012-10-05T07:17:24 < Bird|lappy> or can it keep spewing data at you until the cows all came home?
2012-10-05T07:17:33 < dongs> well, in my application it sjust a few bytes
2012-10-05T07:17:40 < dongs> but technically as far as I2c is concerned, no theres no limit
2012-10-05T07:17:46 < dongs> there's no length field or something liek tath
2012-10-05T07:17:52 < dongs> 'depends on wahtever device youre talking to.
2012-10-05T07:17:52 <+dekar_> I've only seen 1 or 2 bytes
2012-10-05T07:18:23 < dongs> a lot of devices I worked with that had subaddresses for registers and shit will just accept a long write
2012-10-05T07:18:27 < dongs> of 10+ bytes
2012-10-05T07:18:30 < dongs> and autoincrement the pointer register
2012-10-05T07:18:46 < dongs> saves some time filling up initial register stuff etc
2012-10-05T07:19:57 < Bird|lappy> dongs: just have it write to a (relatively) big circular buffer?
2012-10-05T07:20:05 < dongs> Bird|lappy: thats not the problem though?
2012-10-05T07:20:11 < dongs> i mean, why are  you asking about data length
2012-10-05T07:20:22 < dongs> the problem wiht my  requirement is multi-addressing and unknown length of data
2012-10-05T07:20:37 < dongs> i dont know the 1st address is no longer being addressed until i receive stop or restart for next address
2012-10-05T07:20:53 < dongs> and i dont know what address will be addressed *next* so I can't prepare hardware for it.
2012-10-05T07:20:54 <+dekar_> dongs, also earlier I was suggesting using EXTI instead of the i2c peripheral
2012-10-05T07:21:04 <+dekar_> interrupt on SCL change
2012-10-05T07:21:08 < dongs> dekar_: sure thing. doesnt matter how its implemented.
2012-10-05T07:21:17 < Bird|lappy> dongs, what the heck is sending this at you?
2012-10-05T07:21:19 < dongs> i just dont feel like writing it, if you want free 50 bucks have at it
2012-10-05T07:21:35 < dongs> Bird|lappy: some retarded shit
2012-10-05T07:21:38 < dongs> wasnt my idea
2012-10-05T07:21:52 < dongs> it thought it was cool to address each random device wiht its own address
2012-10-05T07:22:06 < dongs> and the protocol is dumb enough so that it cant be made to work
2012-10-05T07:22:13 < Bird|lappy> so you're trying to emulate a composite device with multiple I2C addresses?
2012-10-05T07:22:23 < dongs> yeah
2012-10-05T07:22:40 < Bird|lappy> I've seen multi-match functionality on some uC I2C peripherals...
2012-10-05T07:22:46 < dongs> not good enough for this
2012-10-05T07:23:04 < Bird|lappy> o.O how many different addresses is this system trying to refer to you by?
2012-10-05T07:23:08 < dongs> ive seen some trash on avr where you can "mask' addresses
2012-10-05T07:23:23 < dongs> Bird|lappy: 8-12 depending on setup, but they're not continous
2012-10-05T07:23:34 < dongs> but beacuse of noncontinous masking would work either
2012-10-05T07:24:40 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-05T07:24:40 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-05T07:24:40 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-05T07:25:34 < dongs> anyway since the chip isnt doing much else, it doesnt matter how heavy/ inefficient software iplementation is as long as it works
2012-10-05T07:26:28 < Bird|lappy> dongs: could you set the address mask wide open and then make the decision about what addresses to respond to in your start/repeated start ISR?
2012-10-05T07:26:42 < dongs> no
2012-10-05T07:26:50 < dongs> the onyl shit I know with mask is avr and im not touching that
2012-10-05T07:27:11 < dongs> and it was limited to like... 7bit/6bitmask type shit
2012-10-05T07:27:16 < dongs> it would auto-reply by hardware
2012-10-05T07:27:20 < dongs> to all those masked addresses
2012-10-05T07:27:25 < dongs> i dont have a choice to NOT reply
2012-10-05T07:27:31 < dongs> if I dont want that address active
2012-10-05T07:28:20 < dongs> i.e. i have addresses 0x10, 0x12, 0x20, 0x22, the master will address 10, 12, 14, bla blah, but if my peripheral is only doing 10,12,20 i'd have to nak the missing ones.
2012-10-05T07:28:25 < dongs> if shit was masked it wouldnt work.
2012-10-05T07:28:54 < dongs> dunno any hardware i2c that allows pre-listening on address
2012-10-05T07:28:57 < dongs> to decide ack/nak
2012-10-05T07:31:04 < dongs> ive dicked around looking for software i2c appnotes
2012-10-05T07:31:12 < dongs> but most are either master, or all slaves are for retarded 8bit micros with half the shit in asm
2012-10-05T07:31:20 < dongs> and i just dont feel like sifting through that shit
2012-10-05T07:32:10 < Bird|lappy> dongs, SiM32Cxx
2012-10-05T07:32:27 < Bird|lappy> the I2C module lets you send ACK/NAK from the start interrupt handler
2012-10-05T07:32:28 < dongs> wtf is that? the silabs shit?
2012-10-05T07:32:39 < Bird|lappy> err SiM3C1xx or w/e
2012-10-05T07:32:40 < Bird|lappy> yeah
2012-10-05T07:32:45 < Bird|lappy> SiLabs Precision32
2012-10-05T07:33:07 < dongs> youre gonna make me learn a new cortex arch for a one-off?
2012-10-05T07:33:22 < dongs> LGA, QFN and TQFP package options (40/64/80/92-pin
2012-10-05T07:33:23 < dongs> hmm
2012-10-05T07:34:23 <+dekar_> no promises, but I might do it this weekend
2012-10-05T07:35:31 < dongs> The firmware should read the address and R/W direction bit from the DATA register and determine if the
2012-10-05T07:35:34 < dongs> address matches the slave address. If not, the slave should send a NACK by clearing the ACK bit to 0 and clear
2012-10-05T07:35:37 < dongs> the start interrupt.
2012-10-05T07:35:40 < dongs> wow
2012-10-05T07:35:42 < dongs> that is amazing.
2012-10-05T07:35:45 < dongs> thats liek exactly what I need lol
2012-10-05T07:36:05 < dongs> oh!
2012-10-05T07:36:07 < dongs> but theres a problem
2012-10-05T07:36:11 <+dekar_> you found one?
2012-10-05T07:36:14 < dongs> If the slave address does match, the slave should ACK the address by setting ACK to 1, set the
2012-10-05T07:36:17 < dongs> byte count, clear the start bit, arm reception (RXARM = 1), and clear the start interrupt.
2012-10-05T07:36:22 < dongs> dekar_: no thats hardware i2c on silabs m3 parts
2012-10-05T07:36:24 < Bird|lappy> dongs, set the byte count to 1
2012-10-05T07:36:28 < dongs> and?
2012-10-05T07:36:31 < Bird|lappy> and then reset it when you get the rx interrupt
2012-10-05T07:36:42 < Bird|lappy> if you get stop instead of rx, turn off the rx :)
2012-10-05T07:36:54 < dongs> what about restart? also interrupt i guess?
2012-10-05T07:37:00 < Bird|lappy> let me double check that
2012-10-05T07:37:34 < dongs> anyway, still, chips are too expensive
2012-10-05T07:37:40 < dongs> 4 bucks for cheapest sim3c
2012-10-05T07:37:46 < dongs> f103 is half that
2012-10-05T07:37:53 < Bird|lappy> repeated start = start interrupt
2012-10-05T07:38:10 < Bird|lappy> (there's a status bit you can check from the ISR to distinguish the two)
2012-10-05T07:38:17 < dongs> dekar_: sure. no rush, if its something y ou can use later, the better.
2012-10-05T07:38:45 < dongs> Bird|lappy: and all small  package c1 series are not even stockable
2012-10-05T07:38:49 < dongs> so even if I wanted to buy one i couldnt.
2012-10-05T07:39:30 < dongs> http://www.digikey.com/product-detail/en/SIM3C134-B-GM/336-2218-ND/2795720
2012-10-05T07:39:32 < dongs> i.e. shit like this
2012-10-05T07:40:10 < dongs> Bird|lappy: i2c peripheral looks much more sane than STM32's haha.
2012-10-05T07:40:41 < dongs> hmm.. it also has masking.. by bit...
2012-10-05T07:43:34 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-10-05T07:49:54 <+dekar_> the f4 discovery only does USB FS? :/
2012-10-05T07:50:29 < dongs> yes
2012-10-05T07:50:37 < dongs> hs needs external phy
2012-10-05T07:50:59 <+dekar_> they should have added one
2012-10-05T07:51:08 < dongs> then it wouldnt be $134
2012-10-05T07:51:10 < dongs> er $13
2012-10-05T07:51:18 <+dekar_> I am fine with $15
2012-10-05T07:56:01 < Bird|lappy> dongs: DigiKey will take an order of 1 on them even though they're not in stock
2012-10-05T07:59:01 < Bird|lappy> leadtimes seem to be on the order of 5-6wks
2012-10-05T08:03:42 < dongs> sure but that means I ha ve to order chip, design breakout board for it, etc etc. or i could just use software slave on stm32.
2012-10-05T08:03:57 < dongs> it sure LOOKS nice, but amount of work seems more than its worth.
2012-10-05T08:04:08 < dongs> plus wiating 5-6 weeks to get them :)
2012-10-05T08:04:29 < HTT-Bird> http://www.digikey.com/product-detail/en/SIM3C157-B-GQ/336-2228-ND/2795730
2012-10-05T08:04:42 < dongs> yes, i have one of those
2012-10-05T08:04:43 < HTT-Bird> 80pin version is in stock. :)
2012-10-05T08:04:51 < HTT-Bird> ah.
2012-10-05T08:04:53 < dongs> that chip is bigger than the entier board this is for
2012-10-05T08:04:57 < HTT-Bird> oh, wow :P
2012-10-05T08:05:20 < dongs> I ordered one of the U1 series
2012-10-05T08:05:25 < dongs> the ones with gate drivers built in
2012-10-05T08:05:32 < dongs> was gonna play around with some motor cotnrol stuff
2012-10-05T08:05:37 < HTT-Bird> still, you could put together proof-of-concept code using the 80pin version
2012-10-05T08:05:38 < dongs> but got too lazy making evb
2012-10-05T08:05:59 < HTT-Bird> get it debugged and ready for when the smaller packages become available
2012-10-05T08:06:27 < dongs> sounds like it'd be too easy on the sim3
2012-10-05T08:06:37 < dongs> at least software is a bit more of a challenge :)
2012-10-05T08:07:10 < dongs> still not sure the address/masking stuff though
2012-10-05T08:07:17 < dongs> the slave receive interrupt would fire for ANY i2c address on teh bus?
2012-10-05T08:07:30 < dongs> so you dont set slave address in hardware, you always get to choose it in software?
2012-10-05T08:07:33 < dongs> then why have masking?
2012-10-05T08:08:01 < dongs> you get the interrupt anyway
2012-10-05T08:08:49 < HTT-Bird> dongs: there's a config bit you set to determine if you want to use HW or SW address detect
2012-10-05T08:09:01 < dongs> ah
2012-10-05T08:16:09 < dongs> yea that is really neat
2012-10-05T08:18:28 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 245 seconds]
2012-10-05T08:18:38 < dongs> http://export.farnell.com/_/dp/1864742RL how the fuck can farnell do this
2012-10-05T08:24:07 < emeb_mac> what - set up sub-distributors worldwide?
2012-10-05T08:30:50 < dongs> no
2012-10-05T08:30:55 < dongs> charge that much for that shit
2012-10-05T08:40:45 < jpa-_> oh, it's available now
2012-10-05T08:41:18 < jpa-_> dongs: 1864742  is cheaper because RL means it comes on a reel
2012-10-05T08:42:03 < dongs> jpa, i am not buying that shit from farnell, they cost me something like $4/ea and im getting ripped at that price as well
2012-10-05T08:43:05 < dongs> http://export.farnell.com/invensense/mpu-6050/gyro-accel-9-axis-fusion-prog-i2c/dp/1864742?Ntt=1864742 is just as expensive as RL version.
2012-10-05T08:44:12 < jpa-_> i guess they do different pricing in different regions
2012-10-05T08:44:17 < jpa-_> but where do you get those for $4?
2012-10-05T08:44:43 < dongs> in a box right next to em
2012-10-05T08:44:44 < dongs> me
2012-10-05T08:44:50 < dongs> got around 600, want some?
2012-10-05T08:45:23 < jpa-_> nah, no need for them right now
2012-10-05T08:45:34 < dongs> :)
2012-10-05T08:45:36 < jpa-_> i find it kind of boring that they didn't put a magnetometer on the same chip
2012-10-05T08:45:41 < dongs> they did
2012-10-05T08:45:44 < dongs> in 91xx
2012-10-05T08:45:54 < dongs> mpu9150/9100
2012-10-05T08:46:43 < jpa-_> ah, i keep messing up these numbers
2012-10-05T08:47:54 < dongs> flyback: hot
2012-10-05T08:55:53 < emeb_mac> goat milk
2012-10-05T09:57:08 -!- dekar_ [~dekar@212.255.33.3] has quit [Ping timeout: 246 seconds]
2012-10-05T10:04:07 -!- claude is now known as claude_work
2012-10-05T10:11:30 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-05T10:12:32 -!- acerimmer [~fletch@drupal.org/user/72475/view] has quit [Ping timeout: 246 seconds]
2012-10-05T10:14:17 -!- amitofu [~str@cannabis.dataforce.net] has quit [Read error: Operation timed out]
2012-10-05T10:14:26 -!- amitofu [~str@cannabis.dataforce.net] has joined ##stm32
2012-10-05T10:33:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-05T10:53:50 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-05T10:56:01 -!- Laurenceb_ [~Laurence@host86-143-188-224.range86-143.btcentralplus.com] has joined ##stm32
2012-10-05T11:03:53 -!- Laurenceb_ [~Laurence@host86-143-188-224.range86-143.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-05T11:17:02 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-05T11:31:05 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has quit [Quit: Quit]
2012-10-05T13:00:25 < Laurenceb> http://answers.yahoo.com/question/index?qid=20081014220206AAs1XDb
2012-10-05T13:01:08 < dongs> thanks, i totally needed how to do that
2012-10-05T13:01:20 < jpa-_> that's not what i needed to know
2012-10-05T13:01:34 -!- You're now known as jpa-
2012-10-05T13:04:56 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-05T13:08:03 < dongs>  you might need to take the ball out of your mouse if it has semen on it's balls or ball.
2012-10-05T13:09:21 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-05T13:13:01 < Laurenceb> wtf
2012-10-05T13:13:07 < Laurenceb> solid core wire is utter fail
2012-10-05T13:13:57 < Laurenceb> i think its this - http://www.maplin.co.uk/solid-core-wire-1-0.6-6187
2012-10-05T13:17:55 < Laurenceb> it peels stright off leaded and lead free solder
2012-10-05T13:20:00 < Laurenceb> wondered why my hacked together stripboard kept failing
2012-10-05T13:20:06 < Laurenceb> half the wires have fallen off
2012-10-05T13:36:23 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-05T13:43:45 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-05T13:43:51 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-05T13:44:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-05T13:47:01 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-05T13:55:43 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-05T13:57:24 < cjbaird> dongs: a Daiso store has opened near here. I can get that pinku bento box now!
2012-10-05T13:58:01 < dongs> cjbaird: haha awesome.
2012-10-05T14:00:08 < dongs> make sure to take pics next to your tadpole laptop
2012-10-05T14:31:31 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-10-05T14:36:46 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-05T14:42:17 -!- daku is now known as DaKu
2012-10-05T14:42:45 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-05T14:45:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-05T14:49:40 < Laurenceb> RAGE
2012-10-05T14:49:59 < Laurenceb> why the hell is the a diode on the vreg output on f4discovery
2012-10-05T14:50:13 < Laurenceb> the voltage rail is all over the place
2012-10-05T15:05:58 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
2012-10-05T15:08:37 < dongs> nice rage bro
2012-10-05T15:11:46 < karlp> better rage: why do all the disco boards have silk that says 3v3 but only output 3v?
2012-10-05T15:11:56 < Laurenceb> for the same reason
2012-10-05T15:12:16 < Laurenceb> ive just shorted the diode
2012-10-05T15:21:44 < ABLomas> banzaaaai!
2012-10-05T15:47:58 -!- DaKu is now known as daku
2012-10-05T15:51:00 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-05T15:51:03 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-05T15:53:16 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-05T16:02:35 -!- daku is now known as DaKu
2012-10-05T16:09:02 < karlp> fucking autotools bullshit
2012-10-05T16:14:42 < Laurenceb> fucking swearing
2012-10-05T16:14:53 < Laurenceb> fuck fucking bad pissing language
2012-10-05T16:26:40 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-05T16:28:35 < cjbaird> should be fucking banned from the bloody channel
2012-10-05T16:29:33 < dongs> who?
2012-10-05T16:32:17 < Laurenceb> http://i.imgur.com/TL155.jpg
2012-10-05T16:32:50 < karlp> hehe
2012-10-05T16:36:27 < BrainDamage> instructions unclear, got my dick stuck in the modem
2012-10-05T16:41:49 -!- dekar [~dekar@212.255.22.87] has joined ##stm32
2012-10-05T16:41:52 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-05T16:49:27 < claude_work> http://i.imgur.com/Sl4f6.jpg
2012-10-05T17:04:04 < Laurenceb> temperature sensor etc on F1 still connects to ADC1 only
2012-10-05T17:04:07 < Laurenceb> very sucky
2012-10-05T17:10:11 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-05T17:15:32 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-05T17:17:25 < karlp> why's that matter?
2012-10-05T17:18:58 < Laurenceb> well i need adc1
2012-10-05T17:20:40 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-05T17:20:42 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-05T17:22:09 < Laurenceb> oh - they fixed the dma tho
2012-10-05T17:22:17 < Laurenceb> so ADC3 can DMA nicely
2012-10-05T17:28:28 < karlp> how does the temp sensor being on adc1 stop you from uysing it?
2012-10-05T17:29:54 < Laurenceb> i was thinking i had to use adc1 and 2 to get decent dma
2012-10-05T17:29:59 < Laurenceb> but i can use 2 and 3
2012-10-05T17:30:01 < Laurenceb> http://piholder.com/raspberry-pi-cases/5-raspberry-pi-case.html
2012-10-05T17:30:03 < Laurenceb> ^nice
2012-10-05T17:30:22 <+Steffanx> Little expensive though
2012-10-05T17:30:40 < BrainDamage> half of the pi's cost for the case
2012-10-05T17:34:33 < cjbaird> Manchildren are ruining the Raspberry Pi... :/
2012-10-05T17:35:18 <+Steffanx> Uh what cjbaird ?
2012-10-05T17:35:25 < BrainDamage> the whole point of a pi is being cheap
2012-10-05T17:35:44 < BrainDamage> ie better than a router's hw for the same cost
2012-10-05T17:36:02 < BrainDamage> if you slap in a fancy case, it kinda loses it's pourpose
2012-10-05T17:36:13 <+Steffanx> I bought a case for mine
2012-10-05T17:36:24 < BrainDamage> did it cost 3$ or 15?
2012-10-05T17:36:29 <+Steffanx> < 10
2012-10-05T17:36:42 < BrainDamage> still a bit on the expensive side
2012-10-05T17:36:54 <+Steffanx> Don't care at all :)
2012-10-05T17:37:05 < cjbaird> $70 for a case? "We're trying to get-rick-quick off the blogosphere.."
2012-10-05T17:37:28 <+Steffanx> You should care less cjbaird
2012-10-05T17:37:38 <+Steffanx> Don't want it, don't buy it :)
2012-10-05T17:38:01 < Laurenceb> all the cases are made by noobs
2012-10-05T17:38:09 < Laurenceb> thats the first pro case ive seen
2012-10-05T17:38:31 <+Steffanx> Some people buy an old scooter/motorized thingy, cjbaird :P
2012-10-05T17:40:13 < cjbaird> Bah. "We need to do something about the shit quality of secondary-school computer education!" "Yeah! And let's put 99% of the effort into supporting the 20-30-somethings market!"
2012-10-05T17:42:32  * cjbaird rages every time the official blog carries on about being popular with the Maker® scene. "So.. Liz.. How many under-18s were there?"
2012-10-05T17:43:52 <+Steffanx> Your life i hard cjbaird ?
2012-10-05T17:44:00 <+Steffanx> You seem to be able to complain about EVERYTHING
2012-10-05T17:44:26 < cjbaird> "Liz: Oh! There were quite a few well-off private school kids there! They showed me how they accessed their Raspberry Pis from their iPads!" ... >_>
2012-10-05T17:45:54 < cjbaird> I'm 40-plus. I have 20+ years to bitch about over most of you.
2012-10-05T17:47:04 < dongs> liz is a fucking food blogger
2012-10-05T17:47:13 < dongs> fucking cunt
2012-10-05T17:47:49 <+Steffanx> At least we have a 'food blogger' here too ( or maybe even multiple )
2012-10-05T17:52:48 < dongs> haha
2012-10-05T17:53:40 <+Steffanx> Too bad dongs has to pay himself for it
2012-10-05T17:54:12 < dongs> eating out rice from kawaii pinku bento box
2012-10-05T17:54:18 < cjbaird> "Food bloggers" are like the 21st Century's equivalent of calling yourself a DJ..
2012-10-05T17:54:31 < Laurenceb> lol
2012-10-05T17:54:37 <+Steffanx> At least they make a shitload of money cjbaird
2012-10-05T17:54:44 <+Steffanx> At least the 'famous' dutch ones
2012-10-05T17:55:22 < Laurenceb> devvo still thinks its cool to be a dj
2012-10-05T17:55:25 < Laurenceb> http://www.youtube.com/watch?v=MvRgYjHNQxc
2012-10-05T17:55:27 <+Steffanx> Few 100k for 0.5-1 hour. Who doesn't say yes to that?
2012-10-05T17:55:29 < cjbaird> pointless money transferrence from point A to point B; not actually 'making' any.
2012-10-05T17:56:26 < BrainDamage> I'd like some pointless money
2012-10-05T17:56:40 < BrainDamage> even if I'm not generating actual wealth for the country
2012-10-05T17:57:31 <+Steffanx> "pointless money" that's what we ( north EU ) give you all the time :P
2012-10-05T17:57:31 < Laurenceb> wrong video
2012-10-05T17:57:36 < Laurenceb> http://www.youtube.com/watch?v=wfa8W5oMRoY&feature=related
2012-10-05T17:57:39 < Laurenceb> that even
2012-10-05T17:58:13 < BrainDamage> Steffanx: you should check who has the biggest eco here and who's the most taxed ...
2012-10-05T17:58:32 < BrainDamage> we actually pay penalties to the EU when we produce too many of certain goods
2012-10-05T17:58:35 < BrainDamage> like milk
2012-10-05T17:58:41 < cjbaird> Money from advertising.. not for improving people's physical or mental health, or Science, or practical knowledge That Does Important Shit..
2012-10-05T17:58:42 <+Steffanx> I bet we are
2012-10-05T17:58:49 < BrainDamage> "to keep it fair for northern markets"
2012-10-05T17:59:08 <+Steffanx>  as in pay the most taxes  :)
2012-10-05T17:59:11 < BrainDamage> because what's not consumed is usually exported to the EU market, and it'd cause a price crash
2012-10-05T17:59:35 <+Steffanx> Rules are rules :P
2012-10-05T18:00:01 < cjbaird> REvoke voting rights for pointless farmers. Needs to be done in Australia, too. :P
2012-10-05T18:00:04 < BrainDamage> cjbaird: again, money for services vs actual wealth
2012-10-05T18:00:41 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-05T18:02:11 < cjbaird> Meh. I think it's a bit of a stretch when classifying "giving eyeballs and customers to a poor managed doomed Cafe in a too-high rent area" as a 'service'. :P
2012-10-05T18:02:57 < BrainDamage> as long as people are willing to pay for it
2012-10-05T18:03:46 <+Steffanx> Hmm, isn't that where this 'discussion' sort of started with?
2012-10-05T18:03:53 < dongs> sucking dick is a service to.
2012-10-05T18:04:17 <+Steffanx> Give it a try dongs
2012-10-05T18:04:23 <+Steffanx> Make yourself useful
2012-10-05T18:04:38 < dongs> are you paying?
2012-10-05T18:05:04 <+Steffanx> No, thanks
2012-10-05T18:06:04 -!- claude_work is now known as claude
2012-10-05T18:06:47 < Laurenceb> loolll
2012-10-05T18:07:02 <+Steffanx> I have my hand, dongs
2012-10-05T18:07:04 < dongs> does i2c require oversampling when receiving, or do all the state transitions occur on scl edges?
2012-10-05T18:07:06 < Laurenceb> ewww
2012-10-05T18:07:20 < Laurenceb> dongs: i dont think its oversampled
2012-10-05T18:08:20  * Laurenceb is now watching all the early devvo videos
2012-10-05T18:08:25 < Laurenceb> grr no work going to get done
2012-10-05T18:08:35 < dongs> hmm.
2012-10-05T18:16:51 < Laurenceb> what does (v##U) mean in a macro?
2012-10-05T18:17:49 < cjbaird> concatenation, with C procession. #define foo(x,y) (x##y) -> "xy"
2012-10-05T18:17:59 < cjbaird> *preprocessing
2012-10-05T18:18:19 < karlp> or, in other words, "here be black magic"
2012-10-05T18:22:51 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-10-05T18:23:14 -!- Thorn [~Thorn@128-73-0-164.broadband.corbina.ru] has joined ##stm32
2012-10-05T18:23:15 -!- Thorn [~Thorn@128-73-0-164.broadband.corbina.ru] has quit [Changing host]
2012-10-05T18:23:15 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-05T18:31:04 < Laurenceb> #define PIN_AFIO_AF(n, v)           ((v##U) << ((n % 8) * 4))
2012-10-05T18:31:08 < Laurenceb> wtf is U  ?
2012-10-05T18:32:32 <+Steffanx> not me
2012-10-05T18:45:01 < Laurenceb> oh
2012-10-05T18:45:05 < Laurenceb> unsigned
2012-10-05T18:50:09 < Laurenceb> lol someone thinks im a girl
2012-10-05T18:50:57 < Laurenceb> <Solozerk> hi
2012-10-05T18:50:58 < Laurenceb> <Solozerk> are you, by any chance, available ?
2012-10-05T18:50:58 < Laurenceb> <Solozerk> this is gonna seems a little weird
2012-10-05T18:51:24 < Laurenceb> epic fail / 0
2012-10-05T18:52:04 < BrainDamage> "hey, I just met you ,and this is crazy but here's my number, so call me maybe"
2012-10-05T18:54:40 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Quit: Lost terminal]
2012-10-05T19:00:41 -!- claude is now known as Claude
2012-10-05T19:08:36 < cjbaird> You're assuming he's straight. And you just blew a chance.
2012-10-05T19:16:51 -!- barthess [~barthess@5.100.222.96] has joined ##stm32
2012-10-05T19:18:41 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-05T19:19:00 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-05T19:27:47 <+izua> http://i.imgur.com/BIb1h.jpg
2012-10-05T19:28:50 < emeb> telling it like it is.
2012-10-05T19:29:08 < cjbaird> There's another old-as-internet resume like that..
2012-10-05T19:30:48 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-05T19:39:57 < cjbaird> Ah, this guy: https://en.wikipedia.org/wiki/Brian_O%27Dea
2012-10-05T19:43:12 < emeb> http://dx.com/p/jy-mcu-3208-lattice-clock-ht1632c-driver-with-mcu-support-secondary-development-104306
2012-10-05T19:46:08 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has joined ##stm32
2012-10-05T19:46:41 < cjbaird> has this: http://www.freetronics.com/collections/display/products/dot-matrix-display-32x16-red
2012-10-05T19:47:31 < upgrdman_> stm32f0: MCO can be used to output the system clock to a GPIO pin. but GPIO frequencies are less than maximum system clock frequencies.... will the output actually be at the system clock speed, or truncated the the GPIO speed?
2012-10-05T19:47:58 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-05T19:51:53 < Laurenceb> try it and find out
2012-10-05T19:52:07 <+Steffanx> déjà vu
2012-10-05T19:52:13 < Laurenceb> id imagine at high frequencies the slew rate would be bad
2012-10-05T19:54:08 < upgrdman_> yes, will do. just don't have the mcu with me at the moment
2012-10-05T19:54:56 < upgrdman_> Steffanx: why deja vu?
2012-10-05T19:55:17 <+Steffanx> I say the same thing to Laurenceb once in a while upgrdman
2012-10-05T19:55:37 < upgrdman_> :)
2012-10-05T19:57:44 < emeb> It's good advice.
2012-10-05T19:58:16 < emeb> Too many times folks'll ask a question when they could learn more without too much effort on their own.
2012-10-05T20:01:35 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-05T20:06:32 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 244 seconds]
2012-10-05T20:06:39 < jpa-> upgrdman_: you can set a divider for the MCO, and if it exceeds the IO speed you will see some heavily attenuated signal
2012-10-05T20:06:45 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-10-05T20:09:49 < upgrdman_> jpa-: im looking through rm0091 rcc chapter... where's the mco divider?
2012-10-05T20:11:54 < jpa-> hmm i'm only familiar with F1 & F4.. let's see if F0 has it..
2012-10-05T20:13:15 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-05T20:13:17 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-05T20:21:18 < Thorn> do .ARM.attributes, .ARM.extab, .ARM.exidx sections need to be in the image? my linker scripts (derived from codesourcery and/or ST standard library) put all of that into flash
2012-10-05T20:23:09 <+dekar> don't think so
2012-10-05T20:26:11 < Thorn> googling for these section names finds lots of arm linker scripts with those, everybody probably copies the same stuff without thinking
2012-10-05T20:26:33 < Thorn> but why did original author(s) include those?
2012-10-05T20:27:43 < jpa-> they are for exception handling
2012-10-05T20:28:03 < jpa-> https://bugs.eclipse.org/bugs/show_bug.cgi?id=369466
2012-10-05T20:32:50 <+dekar> oh I only use C and my script lacks them :)
2012-10-05T20:34:02 < Thorn> "Names beginning .ARM.exidx name sections containing index entries for section unwinding. Names beginning .ARM.extab name sections containing exception unwinding information."
2012-10-05T20:34:07 < Thorn> http://infocenter.arm.com/help/topic/com.arm.doc.ihi0044d/IHI0044D_aaelf.pdf
2012-10-05T20:34:37 < Thorn> ".ARM.attributes names a section that contains build attributes. "
2012-10-05T20:38:28 < Thorn> not if it also said what these are used for...
2012-10-05T20:38:31 < Thorn> *now
2012-10-05T20:38:47 < jpa-> you mean build attributes?
2012-10-05T20:39:18 < Thorn> yes
2012-10-05T20:52:44 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-05T21:02:01 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-10-05T21:04:49 < Thorn> so rigol 1102e probes have only 6MHz bandwidth in 1:1 mode and 250MHz in 1:10 mode
2012-10-05T21:05:28 < BrainDamage> that's often the case
2012-10-05T21:05:33 < BrainDamage> shit bandwith in 1:1 mode
2012-10-05T21:08:39 < Thorn> and capacitance is also much larger in 1:1
2012-10-05T21:09:03 < Thorn> which might explain why I2C hates them so much
2012-10-05T22:14:51 < upgrdman_> anyone tried to overclock an stm32?
2012-10-05T22:15:06 <+Steffanx> dekar did
2012-10-05T22:15:15 < upgrdman_> i want to eek out a little more speed from an f0 ... instead of 48MHz, maybe 60?
2012-10-05T22:15:40 < upgrdman_> Steffanx: any details posted online? or did he just jack up the pll multiplier and it worked?
2012-10-05T22:16:04 <+Steffanx> Afaik he didn't posted any details online
2012-10-05T22:16:13 < upgrdman_> :( ok
2012-10-05T22:16:29 <+Steffanx> Just something like 72MHz => 128MHz => "Yay it works" and done
2012-10-05T22:16:56 < upgrdman_> do you know how though? pll multiplier? different crystal?
2012-10-05T22:17:10 <+Steffanx> Just a pll iirc
2012-10-05T22:17:18 <+Steffanx> *the
2012-10-05T22:17:30 <+Steffanx> dekar is around so maybe he can speak for himself
2012-10-05T22:18:15 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 246 seconds]
2012-10-05T22:19:01 < upgrdman_> dekar: hey, you here? any comments on the overclocking? :) :)
2012-10-05T22:25:16 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-05T22:25:34 <+izua> upgrdman_: iirc, he just jacked the pll up (and it was on a f1)
2012-10-05T22:25:41 <+izua> but the peripherals stopped working pretty fast
2012-10-05T22:29:23 <+Steffanx> Sure about the last part izua ?
2012-10-05T22:42:32 -!- Claude is now known as claude
2012-10-05T22:56:44 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 252 seconds]
2012-10-05T23:13:13 < Tectu> how can I store a struct in a text file to an SD card so I can read it back again and return it as a struct?
2012-10-05T23:13:58 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has quit [Ping timeout: 263 seconds]
2012-10-05T23:16:32 < Thorn> nanopb (binary, not text) or some minimal json library
2012-10-05T23:17:23 < Tectu> Thorn, can you explain?
2012-10-05T23:18:08 < Thorn> explain what exactly?
2012-10-05T23:18:15 < Tectu> what do you mean :)
2012-10-05T23:18:39 <+dekar> upgrdman, 8mhz xtal, upped the pll to the max -> 136mhz at 2 flash waitstates
2012-10-05T23:19:44 <+dekar> I think uart stopped working at 136mhz though, 128mhz still worked
2012-10-05T23:21:32 < Thorn> Tectu: I mean you can use these 2 methods among others to solve your problem
2012-10-05T23:21:44 <+dekar> http://dekar.wc3edit.net/flexoptix/arm-cortex-m3-OCed.jpg
2012-10-05T23:22:02 <+dekar> well at least my hd44780 driver was running fine at 128mhz :)
2012-10-05T23:22:12 < Tectu> dekar, can you provide the driver?
2012-10-05T23:22:44 <+dekar> haven't used those in ages, but I think I still have it somewhere
2012-10-05T23:23:05 < Tectu> Thorn, is nanopb some standard stuff or what?
2012-10-05T23:23:53 < Thorn> it's an implementation of google protocol buffers for small embedded systems
2012-10-05T23:24:03 < Thorn> speaking of which, GOOGLE.
2012-10-05T23:25:47 <+dekar> Tectu, https://dl.dropbox.com/u/28467113/hd44780.zip
2012-10-05T23:26:00 < Tectu> danke dekar :)
2012-10-05T23:26:16 <+dekar> np :) I think it was on the cd that came with my board
2012-10-05T23:26:29 <+dekar> I did modify it a bit, don't remember
2012-10-05T23:33:34 -!- barthess [~barthess@5.100.222.96] has quit [Quit: Leaving.]
2012-10-05T23:37:48 < Tectu> dekar, what are you building, btw
2012-10-05T23:42:02 <+dekar> http://www.flexoptix.net/flexbox-v2-transceiver-programmer.html?___store=english
2012-10-05T23:42:33  * Steffanx likes the case
2012-10-05T23:42:48 < emeb> dongs: got those RGB arrays in the mail today. faster than I thought.
2012-10-05T23:45:11 <+dekar> Tectu, well that's what I was building, now it is pretty much done :)
2012-10-05T23:47:26  * Tectu dislikes the price
2012-10-05T23:47:49 <+dekar> Tectu, write a review and get it free, that's not too bad
2012-10-05T23:48:08 < BrainDamage> how can you review it without owning?
2012-10-05T23:48:31 <+dekar> " After two weeks with the new flexBox V2 you send us some feedback regarding the flexBox V2, which we may publish on our website."
2012-10-05T23:49:18 < BrainDamage> oh cool, too bad I don't have anything to interface it with :/
2012-10-05T23:49:25 <+dekar> probably the most exclusive stm32 product out there :)
2012-10-05T23:49:33 <+Steffanx> You can get some transcievers for cheap iirc BrainDamage
2012-10-05T23:49:35 < Tectu> dekar, not sure what it exactly does?
2012-10-05T23:50:21 < Thorn> "I'm not sure what exactly it does, but I like the case." <- review
2012-10-05T23:50:38 <+dekar> Ever had one of those and wondered how to make it change the wavelength? http://www.flexoptix.net/10-gigabit-dwdm-tunable-xfp-transceiver-80km-ddm-dom.html
2012-10-05T23:50:39 < emeb> OSHPark boards: http://imagebin.org/231044
2012-10-05T23:50:56 <+dekar> well the flexbox does it for you :)
2012-10-05T23:51:39 <+dekar> Tectu, high end switches and routers have interchangeable physical interfaces (transceivers)
2012-10-05T23:52:00 < BrainDamage> I'll send the link to a friend that works with fiber optics, he might have an use for it
2012-10-05T23:52:01 <+dekar> using the flexbox you can reprogram those for vendor interoperability and more
2012-10-05T23:52:23 < Tectu> dekar, i am impressed
2012-10-05T23:52:27 < Tectu> eventhough I don't know any of that stuff
2012-10-05T23:52:39 < Tectu> wait... 80km with one fibre? o0
2012-10-05T23:52:57 < Tectu> isn't that.... much? o0
2012-10-05T23:52:59 < BrainDamage> it's not that crazy
2012-10-05T23:53:28 < BrainDamage> think of transoceanic links for instance, the repeaters are hundred of km spaced
2012-10-05T23:53:57 < Tectu> BrainDamage, but with optical, right?
2012-10-05T23:54:02 < Tectu> not?
2012-10-05T23:54:12 <+dekar> Tectu, 80km is still pretty common, everything above that gets freaky (some up to 260km)
2012-10-05T23:54:20 -!- Laurenceb_ [~Laurence@host86-143-188-224.range86-143.btcentralplus.com] has joined ##stm32
2012-10-05T23:54:41 <+dekar> Tectu, those things are optical transceivers
2012-10-05T23:55:03 < Tectu> ah
2012-10-05T23:55:04 < BrainDamage> cable would have too large attenuation yes
2012-10-05T23:56:01 <+dekar> those tunable 10G transceivers I linked, you can have up to 80 of them in parallel on the same pair of fibers
2012-10-05T23:56:03 < Tectu> dekar, where's the problem with going longer than 80km with optical?
2012-10-05T23:56:26 <+dekar> utilizing different wavelengths, insane throughput :)
2012-10-05T23:56:28 < Thorn> btw, are these SFP modules only for ethernet or also for SDH etc.?
2012-10-05T23:56:39 < Laurenceb_> Tectu: dispersion
2012-10-05T23:56:52 < Laurenceb_> also loss by absorption at that sort of distance
2012-10-05T23:57:11 < Laurenceb_> but unless you use single mode your bit rate drops with distance
2012-10-05T23:57:15 <+dekar> Tectu, the long range lasers generate a lot of heat, the XFP form factor can't deal with that
2012-10-05T23:57:34 < Tectu> dekar, I don't know ANYTHING about optics
2012-10-05T23:57:44 < Tectu> never heard of XFP, SDH DOM, DDM or anything before
2012-10-05T23:58:52 <+dekar> http://www.flexoptix.net/media/wysiwyg/flexbox/transceiver-pallette.jpg
2012-10-05T23:59:10 < Tectu> how big are these?
2012-10-05T23:59:12 <+dekar> the smaller ones can't get rid of the heat the high performance lasers generate
2012-10-05T23:59:27 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has joined ##stm32
2012-10-05T23:59:31 < upgrdman_> izua: still here? by "peripherals stopped working" so you mean permanently, or just while overclocked?
2012-10-05T23:59:54 < karlp> dongs: http://imgur.com/gallery/NvJGI
--- Day changed Sat Oct 06 2012
2012-10-06T00:00:02 <+dekar> Tectu, http://en.wikipedia.org/wiki/File:SFP-side.jpg
2012-10-06T00:00:05 < BrainDamage> Laurenceb_: you can use graded index to compensate for modal dispersion, chromatic dispersion is still an issue tough
2012-10-06T00:00:15 < Laurenceb_> you can use it a little
2012-10-06T00:00:15 <+dekar> upgrdman_, no permanent damage
2012-10-06T00:00:21 < Laurenceb_> it doesnt solve the problem
2012-10-06T00:00:32 < BrainDamage> yes, not a definitive solution
2012-10-06T00:01:13 < upgrdman_> dekar: do you know if upping the pll, then upping the ahb/aph prescalers (to slow them down) would work?
2012-10-06T00:01:29 < Tectu> dekar, and these little bastards cost up to 4k?
2012-10-06T00:02:17 <+izua> upgrdman_: did you let the magic smoke out?
2012-10-06T00:02:18 <+dekar> Tectu, yeah, some up to 5k
2012-10-06T00:02:37 <+izua> it shouldn't happen
2012-10-06T00:02:43 < upgrdman_> izua: no, im not even around the mcu. just planning to get some work done over the weekend
2012-10-06T00:03:05 <+dekar> upgrdman_, give it a try, I never intended on using that - only did it for fun
2012-10-06T00:03:20 < upgrdman_> dekar: ok. thanks for the input.
2012-10-06T00:04:01 < upgrdman_> dekar: im porting something from the f4 to the f0 and need a little more computation power::) but im going to look through my code some more, i think i can get rid of some overhead first
2012-10-06T00:04:06 < Tectu> dekar, wtf
2012-10-06T00:04:13 < Tectu> dekar, what's so expensive in these little things?
2012-10-06T00:04:27 <+dekar> I could just have used a beefier SoC for the flexbox if I actually had needed more speed
2012-10-06T00:04:46 < upgrdman_> dekar: how quickly did it take for the periphs to die? and did they just stop working, or fail in an unsafe manner?
2012-10-06T00:04:54 < Thorn> a cisco plugin card for routers can cost $20K+ iirc
2012-10-06T00:05:18 <+dekar> upgrdman_, I think UART just didn't work at 136mhz, didn't try much more than that
2012-10-06T00:05:25 < upgrdman_> ok
2012-10-06T00:05:49 < karlp> Thorn: I wored on one that was list price at 470k :)
2012-10-06T00:05:55 < karlp> only sold two ever though
2012-10-06T00:06:02 < Thorn> ...
2012-10-06T00:06:04 <+dekar> Tectu, actually there are much more expensive CFP transceivers as well - 100G
2012-10-06T00:06:10 < Thorn> what was the cost of parts?
2012-10-06T00:06:26 < karlp> no idea, but it wasn't cheap.
2012-10-06T00:06:36 < upgrdman_> dekar: the f1 is rated for 72MHz max, right?
2012-10-06T00:06:38 < karlp> it was about the size of a 1U server by itself.
2012-10-06T00:06:45 <+dekar> upgrdman_, yeah
2012-10-06T00:08:00 <+dekar> Tectu, http://en.wikipedia.org/wiki/C_Form-factor_Pluggable
2012-10-06T00:08:48 <+dekar> Tectu, it gets really expensive when you require them to have a really narrow wavelength so that you can multiplex several over the same fibre
2012-10-06T00:08:49 < karlp> it was the big brother of this: http://www.cisco.com/en/US/products/hw/routers/ps368/products_data_sheet09186a0080092250.html
2012-10-06T00:08:55 < karlp> not even listed anymore.
2012-10-06T00:09:16 < Tectu> dekar, impressive
2012-10-06T00:09:30 < Laurenceb_> i was thinking about changing core voltage on F4
2012-10-06T00:09:37 < Laurenceb_> over or undervolt it
2012-10-06T00:09:42 < Laurenceb_> for speed/low power
2012-10-06T00:10:12 < Thorn> karlp: are these things actually so insanely complicated, or is it simply carrier grade tax?
2012-10-06T00:10:20 < karlp> aww thanks Thorn, you 've made me go and look up old things I played on.
2012-10-06T00:10:25 < karlp> Thorn: bit of both
2012-10-06T00:10:30 < Laurenceb_> the optics is a pita
2012-10-06T00:10:45 < Laurenceb_> also fairly low volume
2012-10-06T00:10:50 < karlp> some of them are priced substantially higher than others based on how much they could be resold.
2012-10-06T00:10:56 < karlp> and how much equipment they would replace.
2012-10-06T00:11:04 < karlp> I did a whole pile of assembly on this one: http://www.cisco.com/en/US/prod/collateral/modules/ps2831/ps4372/product_data_sheet09186a008008876f_ps368_Products_Data_Sheet.html
2012-10-06T00:13:30 < Thorn> the only cisco I played with was a 2801 lol.
2012-10-06T00:13:55 < karlp> it was my first job,
2012-10-06T00:13:56 < Thorn> with an adsl card
2012-10-06T00:14:14 < karlp> assembly monkey on a custom network processor, straight out of uni.
2012-10-06T00:16:03 < Thorn> nice
2012-10-06T00:17:51 <+dekar> speaking of the flex box, I can highly recommend using USB-HID if you don't need much speed
2012-10-06T00:18:18 <+dekar> on Windows and OSX you just plug the flexbox in, allow the java plugin to run and it just works
2012-10-06T00:18:38 <+dekar> no installation, no admin permissions required
2012-10-06T00:18:41  * Thorn releases CareBear on dekar
2012-10-06T00:18:52 <+dekar> hehe :P
2012-10-06T00:19:27 <+dekar> yeah I had a few arguments with him about it, but I still think the superior user experience is worth it
2012-10-06T00:19:48 < karlp> seems like windows 8 starts to make it better to use other classes again.
2012-10-06T00:19:58 <+dekar> I even do firmware updates over USB-HID and it is reasonably fast, about 8s for 120kb of firmware
2012-10-06T00:20:08 <+Steffanx> Too bad windows 8 is probably a disaster
2012-10-06T00:20:30 <+dekar> I feel like windows8 will gain market share as slow as windows vista
2012-10-06T00:20:54 < karlp> eplaced windows on my work computer today. not looking back
2012-10-06T00:21:55 <+dekar> I see no reason to update my windows7 development VM
2012-10-06T00:23:18 < karlp> I've just inverted the vms, linux outside instead of inside :)
2012-10-06T00:23:25 <+dekar> I don't care too much about operating systems anyway, I guess the only OS where it really matters is OSX since they shamelessly break binary compatibility
2012-10-06T00:25:03 <+dekar> I recently ported some SDR software to OSX and people have weird problems, I am glad I don't have to support it :P http://www.reddit.com/r/RTLSDR/comments/10pv9j/gqrx_macos_x_port_packaged_ready_to_use/
2012-10-06T00:26:51 < upgrdman_> Laurenceb_: is "core voltage" vdd?
2012-10-06T00:28:14 < Laurenceb_> no
2012-10-06T00:28:21 < Laurenceb_> theres an internal regulator
2012-10-06T00:28:24 < Laurenceb_> down to 1.2v
2012-10-06T00:28:54 < upgrdman_> how would you adjust it? is it controlled through registers?
2012-10-06T00:29:12 <+Steffanx> Let me give it a try dekar
2012-10-06T00:29:31 <+dekar> Steffanx, you have an rtlsdr dongle?
2012-10-06T00:29:35 <+Steffanx> No :P
2012-10-06T00:29:55 <+Steffanx> Should get one
2012-10-06T00:29:57 <+dekar> it has no functionality without those, should start though
2012-10-06T00:30:18 <+dekar> but the main problem are people reporting it wouldn't detect the dongle
2012-10-06T00:30:34 <+dekar> even though I packaged like 60mb if libraries including libusb
2012-10-06T00:30:42 <+dekar> *of
2012-10-06T00:31:09 <+Steffanx> I thought you stopped using a mac..
2012-10-06T00:31:25 <+dekar> I plan on doing so, still using Lion though
2012-10-06T00:32:00  * emeb is still on Snow Leopard.
2012-10-06T00:32:01 < Laurenceb_> theres external pins
2012-10-06T00:32:07 < Laurenceb_> you can turn off the internal reg
2012-10-06T00:32:40 <+Steffanx> I guess i'm too up-to-date :P
2012-10-06T00:32:58 < emeb> dekar: I got the RTLSDR stuff running on beaglebone a while back. Seems to have no problems running max sample rate.
2012-10-06T00:33:36 <+dekar> emeb, 2.4mhz you mean?
2012-10-06T00:33:42 < emeb> dekar: yep
2012-10-06T00:34:12 <+dekar> that's nice, I haven't tried it on my panda board, but I assumed the USB was too crappy
2012-10-06T00:34:21 < emeb> Should be OK.
2012-10-06T00:34:41 <+Steffanx> At least it chrashes when you start the dsp without dongle dekar :)
2012-10-06T00:34:47 < emeb> I need to update my BBone OS though - want to get something that can support directly connected audio codec.
2012-10-06T00:35:01 <+dekar> Steffanx, that's unrelated to my port, the original does that as well
2012-10-06T00:35:16 <+Steffanx> No time to fix it?
2012-10-06T00:35:28 <+dekar> Steffanx, I don't care
2012-10-06T00:35:49 <+Steffanx> You sound like microsoft programmer
2012-10-06T00:36:05 <+dekar> Steffanx, I think my GQRX is the only out of the box SDR software there is on OSX
2012-10-06T00:36:10 <+dekar> so people use it anyway
2012-10-06T00:36:22 <+dekar> got a bazillion downloads on my blog
2012-10-06T00:36:54 <+Steffanx> You should add a counter
2012-10-06T00:38:21 <+dekar> don't really care
2012-10-06T00:38:52 <+Steffanx> me neither
2012-10-06T00:39:19 < emeb> so much apathy.
2012-10-06T00:40:33 <+dekar> Steffanx, btw, I am a microsoft programmer. I started to excessively use the registry to store data lately, I guess that should qualify me :D
2012-10-06T00:41:18 <+Steffanx> I guess it's time to get some sleep now
2012-10-06T00:41:38 <+dekar> Qt has this convenient functionality to serialize objects and store them in a mystery place with unlimited storage capacity
2012-10-06T00:41:54 <+izua> %appdata% on windows i think
2012-10-06T00:41:54 <+dekar> also known as the windows registry
2012-10-06T00:42:05 <+izua> wait wait wait
2012-10-06T00:42:10 <+izua> serialized objects on windows go into the registry?
2012-10-06T00:42:18 <+dekar> izua, yeah :D
2012-10-06T00:42:26 <+Steffanx> LOL
2012-10-06T00:42:53 <+dekar> my co-workers started ranting once I told them about my new feature, dunno why XD
2012-10-06T00:44:01 <+Steffanx> It's a good reason to get fired imho :P
2012-10-06T00:44:22 <+dekar> izua, on Linux they go into $HOME/.config and on OSX in $HOME/Library/AppData
2012-10-06T00:44:26 <+Steffanx> Especially when you say it out loud in a publicly logged channel
2012-10-06T00:44:45 <+dekar> Steffanx, it's just some internal test tool :)
2012-10-06T00:45:07 <+dekar> also it is indeed pretty convenient, it now remembers every single setting
2012-10-06T00:45:12 <+Steffanx> Until a customer likes to use that test tool
2012-10-06T00:45:48 <+dekar> our customers only get web applications, so we control updates etc
2012-10-06T00:46:24 <+dekar> we had problems with customers failing to extract zip archives etc, now we don't do that anymore
2012-10-06T00:46:57 <+Steffanx> heh
2012-10-06T00:47:07 <+Steffanx> but you do support IE6?
2012-10-06T00:47:14 <+dekar> I am not involved in that kind of software anyway, I write drivers and tools to test those
2012-10-06T00:47:41 <+dekar> Steffanx, we support java and flash, the browser doesn't matter
2012-10-06T00:48:02 <+Steffanx> web applications => java :S
2012-10-06T00:48:19 <+dekar> hey, it used to be activeX and IE6 only
2012-10-06T00:48:22  * HTT-Bird kicks IE6 straight to the curb
2012-10-06T00:48:34 <+izua> how do you talk with the boards though?
2012-10-06T00:48:35 <+dekar> actually I think it also supported newer IE as well ;)
2012-10-06T00:48:37 <+izua> if no more ocx?
2012-10-06T00:48:45 <+Steffanx> Java applets don't count as webbased imho
2012-10-06T00:48:47 <+dekar> izua, signed java can do
2012-10-06T00:49:46 <+dekar> Steffanx, "The flexBox V2 is part of our web application flexConnect which allows you to rapidly reconfigure pluggable transceivers." feel free to contact out marketing representative if you disagree ;)
2012-10-06T00:50:39 <+Steffanx> No, it's easier to complain here.. so you can read it
2012-10-06T00:51:29 <+dekar> Steffanx, can't help you with that, not my software
2012-10-06T00:51:37 <+dekar> Steffanx, https://demo.reconfigure.me/common/check_fb2_only.html
2012-10-06T00:51:57 <+Steffanx> Check not passed
2012-10-06T00:52:26 <+dekar> what's missing?
2012-10-06T00:52:47 <+Steffanx> flexConnect, didn't accept the applet :)
2012-10-06T00:53:12 <+dekar> well if you buy hardware for 1500€ I'd assume you'd accept it
2012-10-06T00:53:59 <+Steffanx> And that JS is scary
2012-10-06T00:54:28 <+Steffanx> Great combination of german, english and html
2012-10-06T00:55:16 <+Steffanx> </MYTWOCENTS>
2012-10-06T00:56:02 <+dekar> Steffanx, if it was up to me the web should die
2012-10-06T00:56:14 <+Steffanx> var doExtendedJavaCheck = isMac; // || true; // TODO: remove '|| true' :D :D
2012-10-06T00:56:35 <+dekar> I'd rather have Native Client like google is proposing
2012-10-06T00:56:54 <+dekar> mozilla is using a hacked LLVM to compile C++ to JS these days
2012-10-06T00:57:10 <+dekar> broken by design imo
2012-10-06T00:57:32 <+dekar> https://developer.mozilla.org/de/demos/detail/bananabread
2012-10-06T00:58:05 <+Steffanx> Oh, i've seen that yes
2012-10-06T00:58:16 <+dekar> the same mozilla that is proposing WebVibrator to be a standard HTML5 API
2012-10-06T00:59:24 <+Steffanx> Too bad they demontrate the capabilities with a crappy quake-ish game
2012-10-06T00:59:51 <+dekar> and I inherently disagree with mozillas claims that C++ compiled to JS would confirm the open source nature of the web
2012-10-06T01:02:13 <+dekar> I'd rather have googles portable LLVM bytecode, you could compile that into "open source" JS any day if you really wanted to :/
2012-10-06T01:03:24 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-06T01:16:16 <+dekar> izua, http://qt-project.org/doc/qt-4.8/qsettings.html#details
2012-10-06T01:22:43 -!- upgrdman_ [42a6d414@gateway/web/freenode/ip.66.166.212.20] has quit [Quit: Page closed]
2012-10-06T01:23:47 <+dekar> http://en.wikipedia.org/wiki/Windows_Registry#Rationale
2012-10-06T01:24:01 <+dekar> I am seriously a bit confused on how to use the registry on windows
2012-10-06T01:24:14 <+dekar> people always fear it getting too bloated
2012-10-06T01:24:26 <+dekar> but it is supposed to be used that way I think
2012-10-06T01:25:14 <+dekar> also being a database it should be able to cope with a lot of data
2012-10-06T01:26:09 < Bird|lappy> yeah, my take on the Registry is that it isn't a good DB design
2012-10-06T01:26:21 < Bird|lappy> that, and it desperately needs a comment function
2012-10-06T01:27:08 < BrainDamage> it's been used and abused for everything
2012-10-06T01:27:30 < BrainDamage> from inter process communication, to data storage
2012-10-06T01:27:39 < BrainDamage> to whatever people have tought of
2012-10-06T01:30:14 -!- dfletcher [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Changing host]
2012-10-06T01:30:14 -!- dfletcher [~fletch@drupal.org/user/72475/view] has joined ##stm32
2012-10-06T01:30:17 -!- dfletcher is now known as acerimmer
2012-10-06T01:30:48 < Bird|lappy> sysadmins who set up funky configs in Windows are at risk if someone else takes over and digs in
2012-10-06T01:31:02 < Bird|lappy> in the current system that is
2012-10-06T01:33:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-06T01:33:17 <+dekar> lol inter process communication, I guess it actually is a simple way to do that
2012-10-06T01:34:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Client Quit]
2012-10-06T01:34:19 <+dekar> I probably put 2kb of data into it, I guess that isn't much of a problem
2012-10-06T01:34:57 < BrainDamage> I think of it like a public forum, programs post their data that want to be accessible system wide
2012-10-06T01:35:10 < BrainDamage> sorta env vars on steroids
2012-10-06T01:35:41 <+dekar> it is as accessible as configuration files would be
2012-10-06T01:37:08 <+dekar> and according to wikipedia microsoft suggests using it for application configuration data, whether you want to share it, or not
2012-10-06T01:38:00 < BrainDamage> I think the public access as benefit, not restriction
2012-10-06T01:38:46 < BrainDamage> for instance, for programs that run folder portable, they can publish their location there
2012-10-06T01:39:12 < BrainDamage> sure, they could write to an hardcoded path a config file, but registry is the first place one would look for
2012-10-06T01:39:14 <+dekar> BrainDamage, I don't see how it is _different_ from $HOME/.config %appdata% etc when it comes to being accessible
2012-10-06T01:39:39 < BrainDamage> just psychological, not technical
2012-10-06T01:46:17 < Bird|lappy> dekar, yeah.  I agree that it should be accessible
2012-10-06T01:59:42 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-06T02:18:20 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-10-06T02:22:03 < dongs> karlp: haha
2012-10-06T02:27:05 -!- Laurenceb_ [~Laurence@host86-143-188-224.range86-143.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-06T02:34:44 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-06T02:35:15 < dongs> sup trolls
2012-10-06T02:40:42 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-06T02:40:42 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-06T02:40:42 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-06T02:43:34 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-06T02:43:53 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 245 seconds]
2012-10-06T02:45:41 < emeb> (01:42:48 PM) emeb: dongs: got those RGB arrays in the mail today. faster than I thought.
2012-10-06T02:49:48 < dongs> damn
2012-10-06T02:49:48 < dongs> wtf
2012-10-06T02:49:54 < dongs> mine still have no idea where they are
2012-10-06T02:50:11 < emeb> which one of the many on ebay did you order from?
2012-10-06T02:50:21 < dongs> same one as you
2012-10-06T02:50:26 < dongs> electronic lee or whatever shit
2012-10-06T02:50:40 < emeb> yep
2012-10-06T02:50:42 < dongs> http://www.ebay.com/itm/271056288881 right>?
2012-10-06T02:50:53 < emeb> that's the one
2012-10-06T02:50:54 < dongs> and you chose free shippin?
2012-10-06T02:50:58 < emeb> yes
2012-10-06T02:51:01 < dongs> fuckin assholes :D
2012-10-06T02:51:12 < dongs> i guess its here soon... i would have expected china>jp to be faster
2012-10-06T02:51:19 < dongs> than china>us
2012-10-06T02:51:30 < dongs> ok then you can confirm pinout for me :)
2012-10-06T02:51:37 < emeb> sure
2012-10-06T02:52:37 < dongs> http://i.imgur.com/35AwN.png i was also gonna ask if I should put 5V power on separate header (like screw type) or just make it part of the data header (like sparmfun serial back)
2012-10-06T02:52:51 < dongs> this is looking on "top" of pcb, components are on the bottom
2012-10-06T02:52:58 < dongs> RGB goes in from this side
2012-10-06T02:53:13 < dongs> i *think* this is correct
2012-10-06T02:54:13 < dongs> all the text is just for my reference cause for my pcb cad this is back of the board and I had to mirror it
2012-10-06T02:55:36 < emeb> OK - let me grab the datasheet and verify it vs the part I've got.
2012-10-06T02:57:42 < emeb> damn-near impossible to read the pin #s on the datasheet diagram.
2012-10-06T02:57:52 < emeb> all pixelated...
2012-10-06T02:58:02 -!- DaKu is now known as daku
2012-10-06T02:58:04 < dongs> haha
2012-10-06T02:58:07 < dongs> the sparkfun one was better
2012-10-06T02:58:10 < dongs> i think tis same device...
2012-10-06T02:58:20 < dongs> do yours have the snap in edges?
2012-10-06T02:58:25 < dongs> like jigsaw type shit
2012-10-06T02:58:43 < emeb> at least they marked pin 1 on the device...
2012-10-06T02:58:58 < dongs> oh did htey?
2012-10-06T02:59:01 < dongs> does it match the shit i pasted?
2012-10-06T03:00:14 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-06T03:01:58 < emeb> RGB cathodes match the data sheet
2012-10-06T03:02:36 < emeb> pin order matches the numbering on the diagram you just posted.
2012-10-06T03:02:58 < emeb> 1->16, 32<-17
2012-10-06T03:03:16 < emeb> just like an old DIP
2012-10-06T03:08:36 < emeb> I wonder if these are factory 2nds - blems.
2012-10-06T03:09:04 < emeb> all three of the ones I've got have various nicks & scratches on the front surface. Minor, but noticeable in the right light.
2012-10-06T03:12:32 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-06T03:28:45 -!- izua [~izua@unaffiliated/izua] has quit [Remote host closed the connection]
2012-10-06T03:29:15 -!- izua [~izua@79.115.170.139] has joined ##stm32
2012-10-06T03:29:15 -!- izua [~izua@79.115.170.139] has quit [Changing host]
2012-10-06T03:29:15 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-06T03:29:18 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-06T03:36:37 < dongs> hmm
2012-10-06T03:39:50 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Connection reset by peer]
2012-10-06T03:46:20 -!- izua [~izua@unaffiliated/izua] has quit [Quit: Segmentation fault]
2012-10-06T03:47:24 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-06T04:12:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-06T04:12:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-06T04:20:21 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-06T04:52:49 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 246 seconds]
2012-10-06T04:53:41 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-06T04:55:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-06T04:57:00 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-06T05:15:23 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Remote host closed the connection]
2012-10-06T05:16:13 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-10-06T05:31:29 < upgrdman> im using some timers that trigger on their TI1's. one is on APB1, the other on APB2. all settings are the same, but the APB1 timer seems to mis triggers or react slowly. its an f0, and SYSCLK = PCLK = HCLK = 48MHz, so any ideas why?
2012-10-06T05:37:47 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-10-06T05:39:01 < upgrdman> i've also noticed this on my f4, and the clocks for both APBs are set the same too
2012-10-06T05:42:23 < zippe> Hey dongs
2012-10-06T05:42:45 < zippe> this guy a friend of yours?
2012-10-06T05:42:46 < zippe> http://code.google.com/p/tmr/
2012-10-06T05:46:26 < emeb_mac> looks familiar
2012-10-06T06:13:25 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-06T06:14:18 <+dekar> Does the ST library license allow distribution of the source code?
2012-10-06T06:15:02 < dongs> zippe: no, but he messaged me asking if he can use my code
2012-10-06T06:15:08 < dongs> i said after he sends me hardware, and I never got any
2012-10-06T06:16:04 < dongs> haha he uses cpal
2012-10-06T06:16:04 < dongs> wtf
2012-10-06T06:16:15 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-06T06:16:21 < dongs> zippe: stopped taking his shit seriously few months ago when his final pcb mounted a buzzer right next to mag
2012-10-06T06:16:23 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-06T06:18:47 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 240 seconds]
2012-10-06T06:20:19 <+dekar> dongs, is that based on your code? what are you developing?
2012-10-06T06:21:21 < dongs> dekar: my f103-based flyingthing, been done a while ago. that dude asked if he could use some/parts of my code. my shit> http://code.google.com/p/afrodevices/wiki/AfroFlight32
2012-10-06T06:26:41 <+dekar> I want to play with those flying things :)
2012-10-06T06:27:34 < dongs> sure, you want one? i've got a bunch heh.
2012-10-06T06:28:21 <+dekar> they need a ton of stuff, don't they? I'd rather just fly one for a while, without the hassle of building it first
2012-10-06T06:28:35 < dongs> http://www.myrcmart.com/hj-toys-v997-mini-pet-4ch-24ghz-6axis-rtf-quadcopter-bee-p-4854.html here you go then
2012-10-06T06:29:54 <+dekar> wow that looks cool, and uber cheap
2012-10-06T06:29:58 < dongs> yes
2012-10-06T06:30:01 < dongs> includes everything
2012-10-06T06:30:02 < dongs> transmitter as well
2012-10-06T06:30:06 < dongs> just plug the shit in and go
2012-10-06T06:31:00 <+dekar> and they actually do 6-8 minutes? I don't think that's too bad given the size of that thing
2012-10-06T06:31:11 < dongs> no problem 6 mins
2012-10-06T06:31:14 < dongs> i think it'll do  < 10 easily
2012-10-06T06:31:26 < dongs> ive got one, its lot of fun.
2012-10-06T06:31:32 < dongs> just get some extra props for first time flying
2012-10-06T06:31:36 < dongs> they snap almost as soon as you hit anything
2012-10-06T06:31:50 < dongs> the rest of the model is really tough though.
2012-10-06T06:32:41 <+dekar> well thanks, I guess I'll get one or two - they're a great gift as well
2012-10-06T06:36:47 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 255 seconds]
2012-10-06T06:36:51 < dongs> ^_^
2012-10-06T06:38:21 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-06T06:54:39 < upgrdman> STUPID CANUCK
2012-10-06T06:59:01 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-06T07:02:04 <+dekar> dongs, http://www.ebay.de/itm/130746067673
2012-10-06T07:02:37 <+dekar> that one has weird gears attached to the motors, should I stay away?
2012-10-06T07:09:11 < dongs> nah just reduction gears for the brushed motors
2012-10-06T07:09:18 < dongs> i think ar.troll is same shit
2012-10-06T07:09:40 < dongs> its twice the size
2012-10-06T07:09:42 < dongs> of that one i linked
2012-10-06T07:09:44 < dongs> 20cm vs 10cm
2012-10-06T07:19:21 < emeb_mac> those are cute.
2012-10-06T07:19:32 < emeb_mac> how do they make them so cheap?
2012-10-06T07:20:21 < zippe> dongs: funny stuff, though the buzzer is a piezo so probably not a big deal
2012-10-06T07:20:40 < dongs> zippe: pretty sure its your typical 5V magnetic buzzer.
2012-10-06T07:20:47 < dongs> regardless of type it doesnt belong on the board.
2012-10-06T07:22:08 < zippe> dongs: looking at the schematic, it could go either way.  Certainly marked as a piezo.
2012-10-06T07:22:38 < dongs> oh well
2012-10-06T07:24:38 < zippe> Heh, looks like maybe he's lifted the PX4 code...
2012-10-06T07:24:39 < zippe> http://www.rcgroups.com/forums/showthread.php?t=1694086
2012-10-06T07:24:59 < zippe> Hmm, no, just vanilla NuttX
2012-10-06T07:31:28 < dongs> zippe: why the interest. it's just some fail chinese shit on a huge board
2012-10-06T07:31:37 < dongs> with donation buttons.
2012-10-06T07:32:50 < dongs> is he eating into px4  profits?
2012-10-06T07:43:36 < zippe> nah, I thought it might be fun to bring px4 up on the board
2012-10-06T07:43:45 < zippe> but it's all i2c for the sensors … yuck
2012-10-06T07:44:08 < zippe> As for profits?  bwahahaha
2012-10-06T07:46:52 < upgrdman> im using some timers that trigger on their TI1's. one is on APB1, the other on APB2. all settings are the same, but the APB1 timer seems to mis triggers or react slowly. its an f0, and SYSCLK = PCLK = HCLK = 48MHz, so any ideas why?
2012-10-06T08:17:17 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-06T08:20:35 < dongs> zippe: yeah, all i2c, failware.
2012-10-06T08:21:04 < dongs> zippe: http://i.imgur.com/EvncB.jpg ill assemble this with F3 tomorrow. mpu6000+baro on spi, mag on i2c_aux.
2012-10-06T08:22:58 < dongs> but px4 is even more obfuscated than openpilot :)
2012-10-06T08:28:16 < dongs> attn Laurenceb http://songle.jp/
2012-10-06T08:30:08 < upgrdman> your servo connectors are backwards
2012-10-06T08:30:24 < upgrdman> ground goes toward the edge! :(
2012-10-06T08:30:29 < dongs> thats what  you think.
2012-10-06T08:30:52 < upgrdman> it's as close to an industry standard as RC has
2012-10-06T08:31:34 < dongs> never heard of that.
2012-10-06T08:33:23 < upgrdman> look at any hobby receiver with vertical pins
2012-10-06T08:33:35 < upgrdman> i work in the industry
2012-10-06T08:34:07 < dongs> i dont, which is why I dont care :)
2012-10-06T08:34:13 < upgrdman> :)
2012-10-06T08:35:02 < emeb_mac> dongs - the free spirit.
2012-10-06T08:35:24 < dongs> oh right
2012-10-06T08:35:31 < emeb_mac> fuckin' hippie
2012-10-06T08:35:37 < dongs> did y ou ever say about 5V stuff on ledboard?
2012-10-06T08:35:42 < dongs> should I have it on same connector as data
2012-10-06T08:35:45 < dongs> or separate like i have it now
2012-10-06T08:35:49 < emeb_mac> Makes sense to me.
2012-10-06T08:35:59 < emeb_mac> to keep it separate that is
2012-10-06T08:36:13 < dongs> right, because im thinking hten say blocks of 10 or something could be pwoered from separate supply
2012-10-06T08:36:28 < cjbaird> fapfap: http://www.retrotechnology.com/memship/memship.html .. Might get one..
2012-10-06T08:36:28 < emeb_mac> you could end up with hellatious IR drops if you tried to chain it
2012-10-06T08:36:32 < dongs> yeah
2012-10-06T08:37:44 < dongs> what bout connector? 2.54 pin header type shit or should i do something heavier duty
2012-10-06T08:38:39 < emeb_mac> separate 2.54mm headers for 5V/Gnd seem OK
2012-10-06T08:38:54 < emeb_mac> all star-wired to a high current supply.
2012-10-06T08:39:06 < dongs> oh, so dont even bother chaining them?
2012-10-06T08:39:09 < dongs> just have a 5V connector?
2012-10-06T08:39:25 < emeb_mac> Depends on how many you do at once I guess.
2012-10-06T08:39:32 < dongs> right
2012-10-06T08:40:08 < emeb_mac> at 0.5A each it could add up to a lot, so if you did a daisy chain it could drop a lot.
2012-10-06T08:40:23 < dongs> yeah.
2012-10-06T08:40:31 < dongs> plus carrying all that across pcb traces
2012-10-06T08:40:35 < dongs> for the next block
2012-10-06T08:40:38 < emeb_mac> no kidding
2012-10-06T08:42:13 < dongs> ok, so only input 2.54 connector on one end.
2012-10-06T08:42:49 < dongs> or maybe on top? hmm.
2012-10-06T08:42:56 < dongs> like near anodes.
2012-10-06T08:43:42 < emeb_mac> never hurts to be close, but it's not _that_ critical.
2012-10-06T08:43:57 < dongs> well just trying to make the shit looks somewhat symmetrical
2012-10-06T08:45:23 < emeb_mac> yeah
2012-10-06T08:46:17 < emeb_mac> cjbaird: what is it w/ 1802 processors? Things are coming up on 40 years old now. Why do people like them so much?
2012-10-06T08:47:47 < dongs> http://i.imgur.com/GO2o0.png somethin like this?
2012-10-06T08:49:19 < cjbaird> emeb_mac: I've always got the impression that they were the alternative-6502, filling the same niche (up until the 6502 gots it big break with home micros)
2012-10-06T08:49:40 < emeb_mac> dongs: looks good
2012-10-06T08:49:57 < dongs> uart might move somewehre in teh center of the board depending on how ugly it gets to route
2012-10-06T08:50:09 < emeb_mac> flyback: yeah - know about the rad-hard aspect. We had them in some of the flight hardware I used to work on.
2012-10-06T08:50:24 < emeb_mac> Not usually an issue for most terrestrial apps though
2012-10-06T08:51:24 < emeb_mac> cjbaird: I started out on 6502 and 1802 was around before that IIRC. I think they're even more primitive.
2012-10-06T08:51:42 < emeb_mac> I had some buddies in college who had 1802 systems - big on forth too.
2012-10-06T08:52:05 < dongs> dude, youre only 13
2012-10-06T08:52:09 < dongs> being older than you isnt hard
2012-10-06T08:52:29 < dongs> when you reach to level of avrfreak though..
2012-10-06T08:52:52 < emeb_mac> heh. pipsqueak.
2012-10-06T08:53:18 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-06T08:53:19 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-06T08:53:19 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-06T08:54:03 < dongs> i was gonna say about the only thing flying in taht video is wahtever the fuck hes cooking
2012-10-06T08:55:26 < emeb_mac> is he on something?
2012-10-06T08:55:43 < emeb_mac> kids - don't smoke dope before doing a video.
2012-10-06T08:56:04 < dongs>  < flyback> I think he's canadian
2012-10-06T08:56:06 < dongs> so are you
2012-10-06T09:00:06 < emeb_mac> if you like forth check out Chuck Moore's site: http://www.colorforth.com/
2012-10-06T09:00:18 < emeb_mac> dude invented the language and is still working on it.
2012-10-06T09:00:50 < emeb_mac> designs chips that execute forth, using CAD tools built from the processors he designs.
2012-10-06T09:01:05 < dongs> yo dawg
2012-10-06T09:01:09 < emeb_mac> talk about drinking your own kool-aide
2012-10-06T09:03:15 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-10-06T09:05:31 < cjbaird> ..and paid for by a patent he owns about instruction pipelining..
2012-10-06T09:07:50 < emeb_mac> he's hardcore. Lives to work on forth.
2012-10-06T09:08:06 < dongs> isnt forth like dead
2012-10-06T09:08:36 < emeb_mac> Still used in some niche areas. Smaller and smaller niches all the time of course.
2012-10-06T09:09:06 < emeb_mac> takes a fairly orthogonal mindset to use it effectively.
2012-10-06T09:09:22 < emeb_mac> kinda like HP RPN calculators. :)
2012-10-06T09:12:22 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 252 seconds]
2012-10-06T09:12:43 < cjbaird> And lazy scabs like me who keep porting 1978-era FIG-Forth for a quick and dirty OS.
2012-10-06T09:19:54 < zippe> Well, actually
2012-10-06T09:20:15 < zippe> IIRC he decided in the end that using Forth to design chip tools was a bad idea
2012-10-06T09:20:23 < dongs> heh
2012-10-06T09:20:50 < zippe> For all that, there are some nice compact Forths that are great when you don't need a whole lot of stuff, and you don't expect anyone to ever help you
2012-10-06T09:21:12 < dongs> can safely say Forth is deader than *BSD
2012-10-06T09:21:51 < cjbaird> Having FORTH supported by a 'cottage industry' never helped it..
2012-10-06T09:22:52 < dongs> whats more useless, forth or focal? cuz i remember programming someshit in focal at some point.
2012-10-06T09:24:20 < cjbaird> What on earth did you use FOCAL on? I've only found it running on an old DEC RSTS system my Uni kept around in 1989..
2012-10-06T09:24:44 < emeb_mac> didn't Sun use forth for their boot roms?
2012-10-06T09:25:02 < emeb_mac> and of course Postscript is basically forth + graphics.
2012-10-06T09:25:06 < cjbaird> Yep. FreeBSD does as well. PPC Macs did.
2012-10-06T09:25:33 < cjbaird> The OLPC OX-1 also have a forth bootroom
2012-10-06T09:25:43 < emeb_mac> so forth is fairly widely distributed. folks just don't admit it...
2012-10-06T09:27:47 < dongs> olpc is still in business?
2012-10-06T09:28:49 < dongs> cjbaird: i dont fucking remember. yes, it was something PDP related
2012-10-06T09:30:44 < cjbaird> What people hung around after the Microsoft fiasco at OLPC, thinned out even further when the project announced the OX-3 was to be a tablet..
2012-10-06T09:31:34 < dongs> death to tablets
2012-10-06T09:32:10 < emeb_mac> craplets!
2012-10-06T09:32:19 < dongs> its fucking sickening that a $500 piece of shit with garbage processor and lunix has a 10" 1920x1200 display, while the cheapest laptop with a proper CPU and 1920x1200 is 15" and starts at $2500+
2012-10-06T09:32:24 < dongs> do not want.
2012-10-06T09:33:19 < emeb_mac> what _do_ you want?
2012-10-06T09:33:31 < dongs> a 10" laptop with a proper processor, proper keyboard, and 1920x1200 lcd.
2012-10-06T09:33:35 < dongs> WITHOUT FUCKING TOUCHSCREEN.
2012-10-06T09:33:58 < emeb_mac> small + hires + powerful.
2012-10-06T09:34:07 < emeb_mac> for a reasonable price...
2012-10-06T09:34:25 < dongs> price doesnt matter if it existed
2012-10-06T09:34:27 < dongs> but it doesnt.
2012-10-06T09:34:29 < dongs> and noone makes htem
2012-10-06T09:34:39 < dongs> i paid something like 2.5k for panasonic 10" wiht i7
2012-10-06T09:34:43 < dongs> but its a fucking 1024x768 panel
2012-10-06T09:35:19 < emeb_mac> death to 1024x768
2012-10-06T09:35:37 < dongs> all the 'ultrabook' shit is freakin 1366x768
2012-10-06T09:35:42 < dongs> in 2012!!!
2012-10-06T09:35:44 < dongs> DIEEEEEEEEEE
2012-10-06T09:36:01 < dongs> I had higher rez screens on my laptops almost a fucking decade ago
2012-10-06T09:36:09 < dongs> toshiba portege m400,, 1400x1050 in 12"
2012-10-06T09:36:25 < emeb_mac> I got a $500 I5 Win7 laptop a few months back w/ 1366xsomething
2012-10-06T09:36:32 < dongs> ya, garbage.
2012-10-06T09:36:35 < dongs> you can see the fucking pixels
2012-10-06T09:36:38 < dongs> like a mile away
2012-10-06T09:36:43 < dongs> HELLO I AM A PIXEL
2012-10-06T09:37:00 < emeb_mac> I have a real monitor hooked to it thru a KVM
2012-10-06T09:37:20 < emeb_mac> hardly ever use the laptop kbd/mouse/screen
2012-10-06T09:37:27 < dongs> then obviosuly you didnt need a laptop
2012-10-06T09:37:37 < emeb_mac> no, but it was cheap
2012-10-06T09:37:38 < dongs> and you coulda got equivalent ssmall factor desktop for like $300
2012-10-06T09:38:22 < emeb_mac> probably, but most of those are no-name hack jobs.
2012-10-06T09:38:37 < dongs> ya, like dell.
2012-10-06T09:39:01 < cjbaird> Where's my notebook with HUD glasses.. :/
2012-10-06T09:39:01 < emeb_mac> heh
2012-10-06T09:39:23 < ABLomas> use HMD
2012-10-06T09:39:32 < ABLomas> why keep monitor at all
2012-10-06T09:40:08 < cjbaird> Yep. I wants a HMD, with the
2012-10-06T09:40:17 < cjbaird> Yep. I wants a HMD, with the 'notebook' strapped to my arm.
2012-10-06T09:40:35 < emeb_mac> get blisters when the cpu ramps up. :)
2012-10-06T09:41:20 < cjbaird> CyberBlisters
2012-10-06T09:41:32 < emeb_mac> those cost extra.
2012-10-06T09:42:24 < emeb_mac> what kind of resolution do the HMDs have?
2012-10-06T09:42:40 < emeb_mac> enough to make dongs happy?
2012-10-06T09:43:14 < emeb_mac> most of the ones I've seen are like NTSC or something lame like that.
2012-10-06T09:44:02 < cjbaird> sweeeeet http://www.vrealities.com/prov035.jpg
2012-10-06T09:44:28 < cjbaird> 800 x 600 is about the best I recall seeing, although that was ~9 years ago when I was looking.
2012-10-06T09:44:45 < dongs> cinemizer oled is 720p
2012-10-06T09:44:57 < dongs> but thats not hud/hmd
2012-10-06T09:46:09 < dongs> wow
2012-10-06T09:46:16 < dongs> autorouter did entire board
2012-10-06T09:46:20 < dongs> so it should be totally doable by hand
2012-10-06T09:46:35 < cjbaird> who carses about the resolution when this is possible... http://www.vrealities.com/vrproar.jpg
2012-10-06T09:47:12 < cjbaird> hmm.. a number of the products coming up are just mounted OLED
2012-10-06T09:49:00 < emeb_mac> you'd think that something that's filling your visual field should probably have more pixels than that.
2012-10-06T09:59:31 -!- claude is now known as Claude
2012-10-06T10:05:35 < dongs> hmm
2012-10-06T10:06:32 < dongs> shit looks nasty autorouted.
2012-10-06T10:06:39 < dongs> i guess i better be not lazy and hand-do it
2012-10-06T10:17:32 < emeb_mac> doo eeet!
2012-10-06T10:26:56 < dongs> http://i.imgur.com/G3DcG.png it helped me to figure out where to route the missing lines anyway
2012-10-06T10:27:39 < dongs> but yeah looks terrible. maybe ill reuse some of hte ideas and properly do 5V lines
2012-10-06T10:33:28 < emeb_mac> pink and purple. so cute!
2012-10-06T10:35:04 < emeb_mac> yeah - some of those runs look a little longer than they really need to be.
2012-10-06T10:35:09 < dongs> yea heh
2012-10-06T10:37:23 < emeb_mac> gn
2012-10-06T10:37:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-06T10:55:22 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-06T10:55:26 < ABLomas> what's that board?
2012-10-06T10:57:33 < jpa-> it probably blinks leds
2012-10-06T10:57:39 < jpa-> that's all dongs usually does
2012-10-06T10:58:28 -!- acerimmer [~fletch@drupal.org/user/72475/view] has quit [Ping timeout: 245 seconds]
2012-10-06T11:04:56 < ABLomas> ...
2012-10-06T11:05:19 < ABLomas> he usually does trolling, so that's probably something like trollboard...
2012-10-06T11:22:36 -!- Claude is now known as claude
2012-10-06T12:09:26 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-06T12:09:29 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-06T12:18:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-06T12:34:37 -!- barthess [~barthess@164.177.211.124] has joined ##stm32
2012-10-06T12:50:04 < cjbaird> MY Anthem for setting up an ISP back in 1994: http://www.netfunny.com/rhf/jokes/95q2/imaluser.html
2012-10-06T12:50:55 <+Steffanx> Is it worth my time cjbaird ?
2012-10-06T12:51:01 < cjbaird> YES
2012-10-06T12:53:24 < dongs> glad I dont know what beck is
2012-10-06T12:55:09 <+Steffanx> I have to say: me neither
2012-10-06T12:55:57 < dongs> probably some animu
2012-10-06T12:56:18 <+Steffanx> It seems to be somethning they call "music"
2012-10-06T12:56:22 -!- daku is now known as DaKu
2012-10-06T12:58:05 < dongs> i only listen to oppa korea style
2012-10-06T13:01:12 < cjbaird> Pfft, cultureless internet kiddies.
2012-10-06T13:01:33 <+Steffanx> Yes, grandpa
2012-10-06T13:03:25 < cjbaird> Well I'm old enough to spend this evening getting wasted on vodka mixers, dunno about you lot..
2012-10-06T13:05:04 <+Steffanx> You know all you have to know
2012-10-06T13:06:16 < dongs> what the fuck
2012-10-06T13:06:26 < dongs> I update SystemCoreClock variable but it goes back to 72mhz
2012-10-06T13:07:01 <+Steffanx> There's all kind of magic in the stm32 lib ..
2012-10-06T13:08:21 < dongs> fuckkkkkkkkkk this doesnt make sense
2012-10-06T13:09:05 <+Steffanx> Set a breakpoint a memory write of the variable?
2012-10-06T13:09:13 <+Steffanx> *at
2012-10-06T13:14:18 < dongs> WTF
2012-10-06T13:14:25 < dongs> when im in startup asm
2012-10-06T13:14:28 < dongs> calling SystemInit()
2012-10-06T13:14:34 < dongs> SystemCoreClock is properly set to 64mhz
2012-10-06T13:14:41 < dongs> as soon as I enter to main()
2012-10-06T13:14:45 < dongs> its back at 72000000....
2012-10-06T13:14:47 < dongs> ?!?!?!?!
2012-10-06T13:14:50 < dongs> that doesnt make sense
2012-10-06T13:16:19 < dongs> something called __decompress fucks wiht it
2012-10-06T13:16:23 < dongs> waat
2012-10-06T13:21:44 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-06T13:22:13 < dongs> hmm
2012-10-06T13:22:28 < dongs> so i have to modify the shit inside main(), fuck.
2012-10-06T13:24:21 <+Steffanx> :S
2012-10-06T13:24:42 < dongs> works now.
2012-10-06T13:24:45 < dongs> lol @ HSI trimming
2012-10-06T13:24:51 < dongs> 400Hz loop = 401.3Hz
2012-10-06T13:28:46 < dongs> wow this is some avr-class jitter
2012-10-06T13:28:54 <+Steffanx> uh?
2012-10-06T13:29:39 < dongs> sec, screenshottin
2012-10-06T13:31:15 < jpa-> oh noes, promised 1% accuracy was only 0.3% accurate :((
2012-10-06T13:33:04 < dongs> http://i.imgur.com/iIEoq.png
2012-10-06T13:33:05 < dongs> check this shit out
2012-10-06T13:33:15 < dongs> though i guess avr was 400ns jitter
2012-10-06T13:33:24 < dongs> but this is pretty close
2012-10-06T13:35:36 < jpa-> internal RC is quite sensitive to Vcc bypass caps
2012-10-06T13:36:02 < dongs> i've got'em
2012-10-06T13:36:19 < jpa-> sure, but are they good enough
2012-10-06T13:36:27 < jpa-> and close enough
2012-10-06T13:36:28 < jpa-> etc.
2012-10-06T13:36:32 < dongs> yessir
2012-10-06T13:36:52 < dongs> anyway, this was just a test
2012-10-06T13:37:01 < dongs> i had to patch system_f1xxx to enable HSI to max freq
2012-10-06T13:37:14 < dongs> to still make hardware with broken crystla shit usable
2012-10-06T13:37:50 < jpa-> ah, so you are overclocking it to higher PLL multiplier than rated for?
2012-10-06T13:37:56 < dongs> no, still 16
2012-10-06T13:37:59 < dongs> should I try higher? :)
2012-10-06T13:38:12 < jpa-> then what did you patch?
2012-10-06T13:38:28 < dongs> it doesnt startup to 64mhz if HSE is failed
2012-10-06T13:38:38 < dongs> there's a if (hseworks) { } else { // add code here }
2012-10-06T13:39:03 < dongs> so i added switching to hsi + pll
2012-10-06T13:39:06 < dongs> default fallback is just hsi, no pll
2012-10-06T13:39:12 < jpa-> ah
2012-10-06T13:39:26 < dongs> but you give me an idea :)
2012-10-06T13:39:36 < dongs> lemme try 18.
2012-10-06T13:39:46 < dongs> fuck, i just soldered the xtal back on too.
2012-10-06T13:41:24 < dongs> o wait
2012-10-06T13:41:27 < dongs> its fixed bit value
2012-10-06T13:41:29 < dongs> cant go higher.
2012-10-06T13:41:41 < dongs> RCC_CFGR_PLLMULL16 :(
2012-10-06T13:44:22 < dongs> fails.
2012-10-06T13:45:21 < dongs> i think on F2/F4 you can specify arbitrary pll shit
2012-10-06T13:45:22 < dongs> not on F1 ;(
2012-10-06T13:45:37 < dongs> and since HSI is /2 first
2012-10-06T14:07:58 < ABLomas> ea... would that ulink2 clone work with keil compiler?
2012-10-06T14:07:59 < ABLomas> http://www.ebay.co.uk/itm/Ulink-2-USB-JTAG-Emulator-ARM9-Cortex-Keil-Ulink-II-GH2-/120980197645?pt=LH_DefaultDomain_0&hash=item1c2afb510d
2012-10-06T14:08:28 < ABLomas> (yes, i know it's written in title, but...)
2012-10-06T14:08:45 < dongs> yes
2012-10-06T14:08:47 < dongs> thats what im using
2012-10-06T14:08:51 < ABLomas> thanks
2012-10-06T14:08:52 < dongs> works great
2012-10-06T14:10:01 < ABLomas> i paid a lot... pfff, now prices are funny =)
2012-10-06T14:11:08 < dongs> ABLomas: that CD labeled MDK probably has a warezed copy of keil too
2012-10-06T14:11:09 < dongs> lol
2012-10-06T14:11:52 < ABLomas> hehe
2012-10-06T14:12:23 < dongs> jlinkOB might be a cheaper choice
2012-10-06T14:12:24 < dongs> onyl 15 bucks
2012-10-06T14:12:30 < dongs> http://www.hotmcu.com/usbminijtag-jlink-ob-jtagswd-emulator-with-usb-interface-p-29.html?cPath=3_25
2012-10-06T14:12:40 < dongs> not as nicely integrated into keil but still works
2012-10-06T14:13:05 < dongs> and you could probably reflash it with CMSIS-DAP once thats afvailable
2012-10-06T14:13:43 < ABLomas> it's cheap enough to buy without many questions... Will order one for friend probably
2012-10-06T14:13:57 < ABLomas> compared to 400$ original... ;-)
2012-10-06T14:14:06 < dongs> right.
2012-10-06T14:14:15 < dongs> its
2012-10-06T14:38:01 < dongs> attn zyp https://maps.google.com/maps?q=Kanda+Station,+Japan&hl=en&ll=35.691524,139.776112&spn=0.000339,0.000603&sll=38.997934,-105.550567&sspn=5.30981,9.876709&oq=Kanda&hnear=Kanda+Station,+Tokyo,+Japan&t=h&z=21&layer=c&cbll=35.691524,139.776112&panoid=eyPA3DB98GyoLxaHvuMFIQ&cbp=12,343.74,,0,14.22
2012-10-06T14:38:30 < zyp> huh?
2012-10-06T14:40:11 < dongs> lol
2012-10-06T14:42:22 < zyp> what about it?
2012-10-06T14:42:30 < dongs> one spot parking place?
2012-10-06T14:43:30 < zyp> it's Tokyo after all, why let the space go to waste? :p
2012-10-06T14:43:40 < zyp> but I like that it's even numbered :p
2012-10-06T14:48:41 < karlp> oh heh, that's the one I pasted last night isn't it?
2012-10-06T14:48:51 < dongs> ya
2012-10-06T14:49:12 < zyp> where I grew up, having to pay to park anywhere is unheard of
2012-10-06T14:50:51 < dongs> http://i.imgur.com/qHrN6.gif why the fuck woudl this rc shit be needed
2012-10-06T14:50:59 < dongs> noise right?
2012-10-06T14:52:05 < karlp> I've seen that before, but can't remember
2012-10-06T14:52:28 < karlp> isn't it to provide a discharge path when it's in the wrong state or something?
2012-10-06T14:52:31 < dongs> no
2012-10-06T14:52:34 < BrainDamage> it's a snubber network
2012-10-06T14:53:10 < BrainDamage> it absorbs the current peaks to avoid high deltaV
2012-10-06T14:54:00 < karlp> don't they normally use diodes for that?
2012-10-06T14:54:37 < BrainDamage> depends on the task, RC is just as common
2012-10-06T14:54:47 < BrainDamage> it's a combination of R, C and D
2012-10-06T15:04:57 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-06T15:04:57 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-06T15:04:57 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-06T15:09:00 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Quit: Leaving]
2012-10-06T15:23:47 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-06T15:30:33 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-06T15:30:36 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-06T15:38:32 <+Steffanx> "depends on the task" how that depends on the task BrainDamage1 ?
2012-10-06T15:39:31 < BrainDamage1> take this simple example: a diode in antiphase with a coil, like it's commonly used with relays
2012-10-06T15:39:57 < BrainDamage1> now to discharge the coil you're applying approx -0.6V
2012-10-06T15:40:04 < BrainDamage1> which means very slow discharge
2012-10-06T15:40:39 < BrainDamage1> you can add a resistor in series, but this means higher peak voltage, but also faster shutdown
2012-10-06T15:41:19 < BrainDamage1> you can inject the current in a cap, to discharge slowly, but then you need high power diodes that'll withstand very large surges
2012-10-06T15:41:45 < dongs> so do I need high current / high voltage caps/resistors here?
2012-10-06T15:41:48 < dongs> or just normal shit
2012-10-06T15:42:19 < BrainDamage1> low ESR caps and power resistor
2012-10-06T15:43:03 < dongs> yeah, define "power". like > 1W?
2012-10-06T15:43:43 < BrainDamage1> depends how much current would flow in and for how much time, I guess 1W is ok as ballpark
2012-10-06T15:54:15 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-06T15:54:24 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-06T16:11:21 -!- Laurenceb_ [~Laurence@host86-136-119-107.range86-136.btcentralplus.com] has joined ##stm32
2012-10-06T16:11:52 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-06T16:59:21 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-06T17:08:11 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-06T17:33:55 < emeb_mac> morning
2012-10-06T18:03:16 -!- izua [~izua@79.115.170.139] has joined ##stm32
2012-10-06T18:03:16 -!- izua [~izua@79.115.170.139] has quit [Changing host]
2012-10-06T18:03:16 -!- izua [~izua@unaffiliated/izua] has joined ##stm32
2012-10-06T18:03:19 -!- mode/##stm32 [+v izua] by ChanServ
2012-10-06T18:16:21 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32
2012-10-06T18:39:57 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-06T19:02:53 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-06T19:03:15 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-06T19:12:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-06T19:54:58 <+Steffanx> A simple 10uF cap for 0.50 euro :S
2012-10-06T19:55:25 <+Steffanx> ( the one that TI advices for this regulator )
2012-10-06T19:57:01 <+izua> it sounds as good as those golden guitar cables do
2012-10-06T19:59:40 <+Steffanx> And by digikey i have to order 100k
2012-10-06T20:00:00 <+Steffanx> oh, not
2012-10-06T20:01:31 <+Steffanx> but 10k is still too many
2012-10-06T20:02:39 < karlp> choose another cap?
2012-10-06T20:03:00 <+Steffanx> of course
2012-10-06T20:11:38 <+Steffanx> 10uF in 0603, didn't even know that exists
2012-10-06T20:12:05 < dongs> i have a ton of those
2012-10-06T20:12:11 < dongs> you just gotta buy the right type.
2012-10-06T20:12:12 <+Steffanx> I have none
2012-10-06T20:12:14 < dongs> they do exist.
2012-10-06T20:12:24 < dongs> im using them on some boards here too..
2012-10-06T20:13:29 < Robint91> Steffanx, I have a full reel of those
2012-10-06T20:13:46 <+Steffanx> Give me some :P
2012-10-06T20:15:10 <+Steffanx> Since when we can find you here Robint91 ?
2012-10-06T20:15:27 < Robint91> been a few times, ...
2012-10-06T20:17:30 <+Steffanx> ok, wb i guess
2012-10-06T20:17:40 < Robint91> Sun Sep 16 23:46:26 2012
2012-10-06T20:33:14 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-06T20:33:32 < TitanMKD> hi
2012-10-06T20:45:45 -!- claude is now known as Claude
2012-10-06T20:46:06 < upgrdman> 10uf in 0603, so that would have to be tantalum, right?
2012-10-06T20:46:25 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-06T20:46:36 < upgrdman> and fairly low-voltage to get that size
2012-10-06T20:47:03 < qyx_> ceramic
2012-10-06T20:47:29 < dongs> uh ceramic yeah
2012-10-06T20:47:32 < dongs> MLCC or whatever
2012-10-06T20:47:41 < dongs> i think mine are 6.3V
2012-10-06T20:48:17 < qyx_> o bought some 0805 10uF last time, 16V i think
2012-10-06T20:49:05 < dongs> I got a reel of them for like 40 bucks. pretty good deal.
2012-10-06T20:52:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-06T20:57:49 -!- barthess [~barthess@164.177.211.124] has quit [Quit: Leaving.]
2012-10-06T20:59:51 < Laurenceb_> ive got some 22uF, 30v ceramics in 1206
2012-10-06T20:59:56 <+Steffanx> A reel with one dongs? :P
2012-10-06T21:05:18 <+Steffanx> So how many dongs?
2012-10-06T21:05:30 < dongs> more dongs than you can shake a stick at
2012-10-06T21:06:14 <+Steffanx>  /kick dongs
2012-10-06T21:07:52 < emeb_mac> I think dongs linked a video of someone shaking their stick
2012-10-06T21:08:56  * emeb_mac needs brainwash
2012-10-06T21:12:25 <+Steffanx> Can't wash what you don't have emeb_mac :P
2012-10-06T21:13:34 < emeb_mac> I have MRIs that prove I had a brain at one time.
2012-10-06T21:13:55 < emeb_mac> They've ~20 yrs old though. No telling what's up there now.
2012-10-06T21:16:35 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-06T21:20:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-06T21:23:51 < Laurenceb_> sarah palin nanbots
2012-10-06T21:23:54 < Laurenceb_> *nano
2012-10-06T21:31:20 < Laurenceb_> are the brony boards ready DONGS?
2012-10-06T21:32:03 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-06T21:34:56 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-10-06T21:35:31 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-06T21:46:07 -!- barthess [~barthess@178.154.33.35] has joined ##stm32
2012-10-06T21:57:57 -!- DaKu is now known as daku
2012-10-06T21:59:06 -!- Claude is now known as claude
2012-10-06T22:02:10 < upgrdman> emeb_mac, this vid? http://www.youtube.com/watch?v=fS1uDnIPTvo&t=40
2012-10-06T22:08:49 -!- izua [~izua@unaffiliated/izua] has quit [Remote host closed the connection]
2012-10-06T22:11:52 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-10-06T22:12:27 < Laurenceb_> what the actual fuck
2012-10-06T22:13:45 < upgrdman> <emeb_mac> I think dongs linked a video of someone shaking their stick
2012-10-06T22:14:04 < upgrdman> i think dongs linked that one... not sure though
2012-10-06T22:18:46 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-06T22:22:31 -!- claude is now known as Claude
2012-10-06T22:23:12 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-06T22:26:12 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 248 seconds]
2012-10-06T22:26:13 -!- barthess [~barthess@178.154.33.35] has quit [Quit: Leaving.]
2012-10-06T22:51:05 -!- corecode_ [~2@0x2c.org] has joined ##stm32
2012-10-06T22:53:58 -!- njan- [~james@freenode/staff/njan] has joined ##stm32
2012-10-06T22:55:22 -!- Netsplit *.net <-> *.split quits: daku, amitofu, njan, corecode
2012-10-06T22:56:04 -!- Netsplit over, joins: daku
2012-10-06T22:56:07 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-06T22:56:07 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-06T22:56:07 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-06T22:56:19 -!- Netsplit over, joins: amitofu
2012-10-06T22:59:03 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 244 seconds]
2012-10-06T23:00:42 -!- Netsplit *.net <-> *.split quits: daku
2012-10-06T23:01:07 -!- Netsplit over, joins: daku
2012-10-06T23:24:33 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-06T23:35:10 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-06T23:35:29 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-06T23:39:48 -!- barthess [~barthess@164.177.206.190] has joined ##stm32
--- Day changed Sun Oct 07 2012
2012-10-07T00:07:58 < upgrdman> anyone tried pwm on the f0? i cant get it to work, but the registers look correct in gdb. same code works fine with the f4 (but different timer number)
2012-10-07T00:10:58 < Laurenceb_> wtf
2012-10-07T00:11:14 < Laurenceb_> F37x in lqfp48 is not pin compatible with f103
2012-10-07T00:11:22 < jpa-> upgrdman: does the timer count run properly when checked with gdb?
2012-10-07T00:11:27 < Laurenceb_> A7 is a VDD pin
2012-10-07T00:11:41 < jpa-> does F3 have vcap etc. like F4?
2012-10-07T00:12:10 < Laurenceb_> and theres a VREFSD+ pin on B12
2012-10-07T00:12:17 < Laurenceb_> then two new port F pins
2012-10-07T00:12:25 < Laurenceb_> so it has same number of io total
2012-10-07T00:13:03 < upgrdman> jpa-, yes
2012-10-07T00:14:34 < Laurenceb_> oh - its dedicated for the sigma delat adc
2012-10-07T00:14:42 < Laurenceb_> ill check the F30x
2012-10-07T00:18:09 < Laurenceb_> ah - that one is pin compatible
2012-10-07T00:22:47 < Laurenceb_> very nice
2012-10-07T00:41:53 -!- Claude is now known as claude
2012-10-07T00:46:21 -!- njan- is now known as njan
2012-10-07T00:46:33 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
2012-10-07T00:49:23 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-07T00:53:59 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Read error: Connection reset by peer]
2012-10-07T00:55:49 -!- barthess [~barthess@164.177.206.190] has quit [Read error: Connection reset by peer]
2012-10-07T02:08:46 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-07T02:13:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-07T02:25:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-07T02:28:15 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-07T02:51:18 -!- Laurenceb_ [~Laurence@host86-136-119-107.range86-136.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-07T03:02:49 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-07T03:23:17 < dongs> sup trolls
2012-10-07T03:23:29 < dongs> < Laurenceb_> F37x in lqfp48 is not pin compatible with f103
2012-10-07T03:23:31 < dongs> wat
2012-10-07T03:23:34 < dongs> you best be trollen
2012-10-07T03:26:19 < dongs> oh motherfucker
2012-10-07T03:26:29 < dongs> lol @ vdd2 being in the center of the chip
2012-10-07T03:27:05 < emeb_mac> doh!
2012-10-07T03:27:24 < dongs> of course I only checked F30x and was like ok, this shit looks fine
2012-10-07T03:27:38 < emeb_mac> was thinking about getting some of those with my latest mouser order
2012-10-07T03:27:40 < dongs> first ST lies about API compatibility..
2012-10-07T03:27:50 < dongs> now they lie about fucking pin compatibility too.
2012-10-07T03:27:59 < emeb_mac> grumble
2012-10-07T03:28:25 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-07T03:28:25 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-07T03:28:25 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-07T03:29:08 < dongs> well this is fucking stupid. I do have a lqfp socket for  STM32... but even THAT won't work because it's preconnecting VDD_*/VSS_ shit
2012-10-07T03:29:20 < dongs> shit's goin back to mouser, fuck ST.
2012-10-07T03:35:04 < dongs> ok i can just ghetto some shit up i guess...
2012-10-07T03:35:09 < dongs> and use it in a socket
2012-10-07T03:36:07 -!- Reggie__ [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-07T03:37:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Ping timeout: 245 seconds]
2012-10-07T03:41:23 < dongs> losing some pins.
2012-10-07T03:41:24 < dongs> or someshit
2012-10-07T03:47:16 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-10-07T03:49:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-07T03:50:18 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-07T03:50:28 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-07T03:54:00 -!- Reggie__ is now known as ReggieUK
2012-10-07T04:12:09 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Read error: Operation timed out]
2012-10-07T04:32:31 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-07T04:32:39 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-07T04:39:24 < dongs> hmm
2012-10-07T04:39:36 < dongs> why would const char *blah[] = { "FOO", "BAR", NULL }; end up in ram?
2012-10-07T04:39:54 < dongs> is tehre  way to rewrite that so it just stays in .rodata?
2012-10-07T04:46:43 < dongs> hmm since i just scan it sequentially maybe i should redo it as blah\0bar\0end\0\0 or osmething
2012-10-07T04:50:17 < zyp> try: const char const *blah
2012-10-07T04:51:13 < dongs> nope
2012-10-07T04:51:26 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 256 seconds]
2012-10-07T04:51:27 < dongs> warning:  #83-D: type qualifier specified more than once
2012-10-07T04:51:35 < zyp> oh, wait
2012-10-07T04:51:41 < dongs> prob c++ only
2012-10-07T04:51:42 < zyp> const char* const blah[]
2012-10-07T04:52:03 < dongs> wow, nice
2012-10-07T04:52:32 < zyp> const before * means pointer to const, const after * means const pointer
2012-10-07T04:52:45 < dongs> right, i figured the const would just apply to all
2012-10-07T04:52:48 < dongs> nice find.
2012-10-07T04:53:25 < zyp> that code you are making are creating both string literals going into rodata and a list of pointers to those, which may or may not go into rodata depending on whether they are mutable or not
2012-10-07T04:53:48 < dongs> featureNames                             0x0800a9cc   Data          52  cli.o(.constdata)
2012-10-07T04:53:52 < dongs> ya it went into constdata in flash nwo
2012-10-07T04:53:54 < dongs> before it was in ram
2012-10-07T04:54:01 < zyp> sounds right
2012-10-07T04:54:03 < dongs> featureNames                             0x20000050   Data          52  cli.o(.data)
2012-10-07T04:56:12 < dongs> found some more similar shit, cool
2012-10-07T04:56:25 < dongs> pow() on F1 takes up 2.5k, what the fuck
2012-10-07T04:56:44 < zyp> lookup tables?
2012-10-07T04:56:54 < dongs>     pow                                      0x08009d91   Thumb Code  2522  pow.o(i.pow)
2012-10-07T04:56:58 < dongs> nope
2012-10-07T04:57:17 < dongs> and 2k of lookup tables.
2012-10-07T04:57:34 < dongs> wait no, ~130bytes of crap
2012-10-07T04:57:40 < dongs> 2.5k of code.
2012-10-07T04:57:55 < dongs> even sin & friends are much less
2012-10-07T04:58:11 < dongs> should just replace it with (a / (b - a * b + a)); amirite??
2012-10-07T04:58:31 < zyp> replace what?
2012-10-07T04:58:41 < dongs> pow()
2012-10-07T04:58:41 < dongs> heh
2012-10-07T04:58:59 < dongs> its a ghetto approximation
2012-10-07T04:59:33 < qyx_> pow?
2012-10-07T04:59:54 < dongs> x ^ y
2012-10-07T04:59:58 < dongs> (not xor)
2012-10-07T05:00:10 < zyp> often written as x ** y
2012-10-07T05:00:17 < zyp> to avoid confusion with xor
2012-10-07T05:00:34 < qyx_> i understand, but that approximation confuses me
2012-10-07T05:00:50 < zyp> same, I don't get how it would work :p
2012-10-07T05:01:00 < zyp> but I'm maybe just too tired
2012-10-07T05:01:13 < zyp> 4 in the middle of the night here,
2012-10-07T05:01:23 < qyx_> i think it should be at least recursive
2012-10-07T05:02:03 < zyp> it's probably an approximation that's only usable around some specific values of a and b
2012-10-07T05:02:23 < qyx_> http://martin.ankerl.com/2007/10/04/optimized-pow-approximation-for-java-and-c-c/
2012-10-07T05:02:28 < qyx_> that seems to be better
2012-10-07T05:03:21 < qyx_> i knew similar code for sqrt but don't like that typecast stuff here
2012-10-07T05:03:27 < qyx_> float->int and back
2012-10-07T05:06:38 < zyp> well, duh, on platforms without an FPU, floats are always worked on as ints in FPU emulation functions
2012-10-07T05:07:52 < qyx_> yes but here you actually exploits the format of float and using integer bit shift you also change the exponent in float
2012-10-07T05:07:56 < zyp> binary, a float is just a sign bit, exponent and mantissa, and working on it as an int is how you can access those parts seperately/directly
2012-10-07T05:07:58 < qyx_> *exploit
2012-10-07T05:08:05 < qyx_> yep
2012-10-07T05:08:15 < zyp> qyx_, well, how do you think FPU emulation works?
2012-10-07T05:08:33 < qyx_> the same as it is the only way :)
2012-10-07T05:09:41 < zyp> what happens is pretty obvious when you look into what the «integer» operations are actually doing to the exponent and mantissa
2012-10-07T05:11:08 < qyx_> ok, offtopic question
2012-10-07T05:11:19 < qyx_> how do you store settings in i2c eeproms?
2012-10-07T05:11:41 < zyp> you just write to it?
2012-10-07T05:11:54 < qyx_> i am thinking about some nice key-value abstraction
2012-10-07T05:12:01 < zyp> addr+W, offset, data, data, data
2012-10-07T05:12:12 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-07T05:13:30 < qyx_> something that could be written as conf_set(CONF_DEBUG_LEVEL, 3);
2012-10-07T05:13:49 < qyx_> and it automatically manages addresses and keys
2012-10-07T05:15:40 < zyp> just make an enum of the keys, and use those for offset?
2012-10-07T05:16:08 < qyx_> hm, thats true
2012-10-07T05:16:16 < qyx_> i probably overcomplicate it
2012-10-07T05:16:19 < zyp> i2c eeproms are usually quite small, so you don't really want to store the keys in the eeprom, better to have them hardcoded in the code
2012-10-07T05:17:15 < qyx_> now i store all settings in struct in sram and write it to eeprom upon request
2012-10-07T05:17:21 < qyx_> and load it at startup
2012-10-07T05:17:45 < zyp> I've also considered doing it that way
2012-10-07T05:19:04 < dongs> zyp, here's a good one for you
2012-10-07T05:19:13 < dongs> yesterday I was debugging running off HSI if crystal is broken
2012-10-07T05:19:23 < dongs> and i patched f103_system stuff to init hsi at 64mhz etc.
2012-10-07T05:19:51 < dongs> after which I run SyscoreClockupate() stuff from same file, which sets SystemCoreClock int to the cpu speed.
2012-10-07T05:20:05 < dongs> but as soon as it stopped runing that code and went into main, SystemCoreClock would go back to 72mhz
2012-10-07T05:20:24 < dongs> i stepped through and some shit called "__decompress" was fucking with the variable before hitting main()
2012-10-07T05:20:28 < dongs> any ideas?
2012-10-07T05:20:49 < dongs> (SystemInit is called from startup asm
2012-10-07T05:20:55 < zyp> I'm not using dumb libraries.
2012-10-07T05:20:55 < zyp> :p
2012-10-07T05:21:00 < dongs> yaman. :(
2012-10-07T05:21:08 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-10-07T05:21:33 < dongs> oh well i fixed it by running clockupdate() stuff once it hits main.
2012-10-07T05:21:39 < dongs> still fucked up / annoying
2012-10-07T05:21:58 < dongs> i mean it calculates it wrong, too.
2012-10-07T05:22:00 < zyp> wait, SystemCodeClock is a variable, used for calulating dividers and shit?
2012-10-07T05:22:04 < zyp> Core*
2012-10-07T05:22:07 < dongs> yea
2012-10-07T05:22:15 < dongs> its global var from stdperiphlib
2012-10-07T05:22:41 < dongs> uint32_t SystemCoreClock         = SYSCLK_FREQ_72MHz;
2012-10-07T05:22:49 < dongs> its extern'd from teh header
2012-10-07T05:23:12 < dongs> but SystemCoreClockUpdate() will modify it
2012-10-07T05:23:16 < dongs> according to HSI/HSE/pll
2012-10-07T05:23:27 < dongs> (and it does modify it)...
2012-10-07T05:23:46 < dongs> just as soon as shit returns from SystemInit() and before jump to main() it gets put back to 72mhz.
2012-10-07T05:23:52 < zyp> well, I don't know about the code, but reading it should explain what it's doing
2012-10-07T05:23:56 < dongs> i guess asm calls __main() which then calls some keil crap
2012-10-07T05:24:11 < dongs> zyp, the function does its job, i stepped through it
2012-10-07T05:24:17 < dongs> then I put a write breakpoint on SystemCoreClock
2012-10-07T05:24:26 < zyp> what about the __decompress shit?
2012-10-07T05:24:28 < dongs> right
2012-10-07T05:24:33 < dongs> thats where it got hit
2012-10-07T05:24:36 < dongs> it modified it back to 72mhz.
2012-10-07T05:24:40 < dongs> before jumping to main.
2012-10-07T05:24:50 < dongs> but theres no sores for that section.
2012-10-07T05:24:50 < zyp> then that's the code you need to read.
2012-10-07T05:25:01 < zyp> oh, right
2012-10-07T05:25:15 < zyp> guess you have to call your closed sores vendor for support then
2012-10-07T05:25:21 < dongs> ^_^
2012-10-07T05:25:54 < zyp> it's not some compiler environment setting or something like that?
2012-10-07T05:26:10 < dongs> i dont see why it would be..
2012-10-07T05:26:22 < dongs> i mean the compiler would have to know its for ST processors using stdlib...
2012-10-07T05:26:32 < zyp> hmm, right
2012-10-07T05:27:18 < zyp> hmm, maybe that's the function that initializes .data?
2012-10-07T05:27:33 < dongs> possibly, but why would it change it back?
2012-10-07T05:28:00 < dongs> because its assigned in the system_stm32f103x.c?
2012-10-07T05:28:15 < zyp> yes, try changing that assignment to 64MHz
2012-10-07T05:28:55 < dongs> hmm, i see. yeah, that makes sense. because systeminit is called before this.
2012-10-07T05:29:43 < zyp> I'm also changing clocks before I initialize .data in my code, so initializing .data would be faster
2012-10-07T05:30:14 < zyp> but that means I can't use anything in .data when initially changing clocks
2012-10-07T05:30:29 < dongs> so how should I save this systemclock stuff then
2012-10-07T05:30:45 < dongs> not initialzie it and call systemcoreclockupdate at the end of Init?
2012-10-07T05:30:46 < zyp> when you are defining the global variable
2012-10-07T05:31:07 < zyp> oh, right, you are selecting speed dynamically?
2012-10-07T05:31:08 < dongs> zyp, its variable, thats the point
2012-10-07T05:31:10 < dongs> yes
2012-10-07T05:31:24 < dongs> if (crystalisbroken) { clockdownto64mhzusinghsi(); }
2012-10-07T05:31:35 < dongs> and then I use t hat for timer dividers and shit
2012-10-07T05:32:05 < zyp> just move the entire clock switching thing until after initializing .data and shit?
2012-10-07T05:32:17 < dongs> well yeah.
2012-10-07T05:32:23 < zyp> I'm considering to do that in my own code, it's not like it takes that much time anyway
2012-10-07T05:34:07 < zyp> I'd guess that a worst-case initialization of the full 20 kB of ram would still take less than 5ms
2012-10-07T05:35:54 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-07T05:37:04 < dongs> heh
2012-10-07T05:39:56 < zyp> well, time to sleep
2012-10-07T05:47:42 < dongs> indeed
2012-10-07T05:53:20 < dongs> http://webx.dk/rc/uhf-link3/rx402_006.jpg how can this claim to have 3.3V outputs for signals when its clearly just directly going from failmega running at 5V...
2012-10-07T05:53:39 < dongs> or you think they're running failmega at 3.3?
2012-10-07T05:54:37 < upgrdman> there are 3v arduinos, so probably a 3v mcu
2012-10-07T05:54:49 < dongs> Input voltage range: 4-6V (rx uses 134mA at 5V)
2012-10-07T05:54:54 < dongs> hm right
2012-10-07T05:55:02 < dongs> since low end is 4V
2012-10-07T05:55:47 < qyx_> waht is it?
2012-10-07T05:55:52 < qyx_> ah, uhf link
2012-10-07T05:55:55 < dongs> some rc trash, yeah
2012-10-07T05:58:14 < upgrdman> dongs, was it you that linked to this vid a few days ago?  http://www.youtube.com/watch?v=fS1uDnIPTvo&t=40 ... it was either someone here or in ##electronics
2012-10-07T05:59:12 < dongs> not me.
2012-10-07T05:59:16 < upgrdman> k
2012-10-07T05:59:18 < dongs> but sounds like something laurenceB would link to
2012-10-07T06:01:31 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-07T06:01:36 < dongs> ^ you offended him
2012-10-07T06:08:29 < dongs> now i have 3000 rgb leds
2012-10-07T06:08:31 < dongs> wat2do
2012-10-07T06:11:27 < upgrdman> why so many
2012-10-07T06:11:31 < dongs> dunno
2012-10-07T06:11:36 < dongs> they were chceap
2012-10-07T06:11:41 < upgrdman> ?$
2012-10-07T06:12:06 < dongs> <100
2012-10-07T06:12:19 < dongs> zyp: oh, interesting. keils shit does "compress" ZI-data.
2012-10-07T06:12:27 < dongs> er RW-data
2012-10-07T06:12:29 < upgrdman> zi?
2012-10-07T06:12:39 < dongs> 1k before compress, 164bytes after
2012-10-07T06:12:44 < dongs> so __decompress must be for that.
2012-10-07T06:13:00 < dongs> zi = zero initialized
2012-10-07T06:15:02 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-07T06:15:09 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-07T06:17:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-07T06:47:18 < dongs> hm why the fuck i get a TIM_Update interrupt even ebfore TIM_Cmd is called?!
2012-10-07T06:48:44 < dongs> ->SR has last bit set, so its def 'update' interrupt...
2012-10-07T07:19:35 < dongs> ah, helps to TIM_DeInit() first.
2012-10-07T07:19:42 < dongs> must have been some crap leftover after power up
2012-10-07T07:20:54 < emeb_mac> bizarre
2012-10-07T07:21:19 < emeb_mac> POR doesn't clear the state?
2012-10-07T07:21:32 < emeb_mac> or does the bootloader use it for something?
2012-10-07T07:21:39 < dongs> guess not, i dont know. update interrupt would fire as soon as I did ITConfig
2012-10-07T07:21:46 < dongs> once
2012-10-07T07:21:50 < dongs> not again...
2012-10-07T07:21:53 < emeb_mac> you using the  STM periph libs?
2012-10-07T07:21:56 < dongs> yea
2012-10-07T07:24:10 < emeb_mac> looked thru the source for TIM_DeInit() - just asserts/deasserts RCC reset for that timer.
2012-10-07T07:24:15 < emeb_mac> Odd.
2012-10-07T07:24:43 < dongs> http://bcas.tv/paste/results/scndz826.html
2012-10-07T07:24:45 < dongs> very simple shit here
2012-10-07T07:24:55 < dongs> TIM1_UP_IRQHandler would fire once after TIM_ITConfig
2012-10-07T07:25:00 < dongs> without TIM_DeInit
2012-10-07T07:25:51 < emeb_mac> seems like a simple solution
2012-10-07T07:28:37 < upgrdman> dongs, not sure if this helps, as i do not use the stdperiph, but when init'd a timer you should force an update, then clear the update flag, or the ISR can get called immediately after enabling the counter.
2012-10-07T07:29:03 < upgrdman> force update and clear flag BEFORE enabling the counter
2012-10-07T07:29:47 < dongs> why? thats what it seems to be doing, but why?
2012-10-07T07:30:26 < upgrdman> i think it's because the ARR or PSC is buffered, and your settings do not actally go into effect until the first update event
2012-10-07T07:30:48 < upgrdman> i discovered this the hard way when i wanted to have a one-shot timer fire after 3 seconds, and it fired immediately
2012-10-07T07:30:53 < upgrdman> here's my code:
2012-10-07T07:31:01 < dongs> bcas.tv/paste
2012-10-07T07:31:06 < upgrdman> http://www.farrellf.com/projects/hardware/2012-08-11_STM32F4_Basics:_Timers_%28Part_1%29/
2012-10-07T07:31:12 < dongs> tghat works too.
2012-10-07T07:31:21 < upgrdman> :) i use my blog as notes
2012-10-07T07:35:44 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 255 seconds]
2012-10-07T07:56:37 < emeb_mac> stop looking at flashing lights.
2012-10-07T07:57:15 < upgrdman> release some tension
2012-10-07T08:07:48 -!- oPossum5150 [~opossum51@adsl-76-253-108-184.dsl.klmzmi.sbcglobal.net] has left ##stm32 []
2012-10-07T08:11:12 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-07T08:11:20 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-07T08:13:58 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-07T08:53:38 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-07T09:29:23 < dongs> damn im getting the dumbest shit not working today
2012-10-07T09:29:26 < dongs> return GPIOA->ODR & GPIO_Pin_8 ? 1 : 0;
2012-10-07T09:29:30 < dongs> shit never goes low
2012-10-07T09:29:39 < dongs> while I can totally see pin going low on scope
2012-10-07T09:29:45 < dongs> yes, PA8 is set to IN_FLOATING
2012-10-07T09:34:26 < dongs> oh, dumbfuck, should be IDR
2012-10-07T09:39:43 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 244 seconds]
2012-10-07T09:40:02 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-07T09:40:03 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-07T09:40:03 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-07T09:56:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-07T09:58:54 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-07T10:52:35 -!- prog [prog@bornbrilliant.org] has quit [Ping timeout: 260 seconds]
2012-10-07T10:53:11 -!- prog [prog@bornbrilliant.org] has joined ##stm32
2012-10-07T10:55:02 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-07T11:00:52 -!- claude is now known as Claude
2012-10-07T12:13:30 -!- barthess [~barthess@164.177.220.61] has joined ##stm32
2012-10-07T12:50:36 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-07T12:55:09 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-07T12:55:12 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-07T12:57:33 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-07T13:05:39 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-07T13:11:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-07T13:21:04 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-07T13:25:23 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-07T13:25:26 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-07T13:27:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 255 seconds]
2012-10-07T13:39:31 -!- Claude is now known as claude
2012-10-07T13:42:58 -!- barthess [~barthess@164.177.220.61] has quit [Quit: Leaving.]
2012-10-07T13:44:15 -!- dekar_ [~dekar@212.255.19.185] has joined ##stm32
2012-10-07T13:44:18 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-07T13:46:09 -!- dekar [~dekar@212.255.22.87] has quit [Ping timeout: 256 seconds]
2012-10-07T13:47:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-07T13:48:35 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-07T13:49:07 < TitanMKD> hi
2012-10-07T13:49:47 < TitanMKD> for those interested in a cheap funny board to test USB current consumption/voltage or even check USB D+/D- and ID signal i have designed a new PCB
2012-10-07T13:50:07 < TitanMKD> see picture here http://postimage.org/image/6fc1438r5/
2012-10-07T13:50:34 < TitanMKD> It is 3D with XRay effect to see both side at once
2012-10-07T13:53:53 < TitanMKD> the primary idea was to check current consumption of a board connected on one side and power on the other side
2012-10-07T13:54:10 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-07T13:54:17 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-07T13:59:26 -!- Laurenceb_ [~Laurence@host86-177-214-10.range86-177.btcentralplus.com] has joined ##stm32
2012-10-07T14:02:17 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-10-07T14:09:35 < TitanMKD> if someone is interested see the DP forum http://dangerousprototypes.com/forum/viewtopic.php?f=19&t=4613
2012-10-07T14:10:51 < zyp> dongs, compression saves some flash, but probably increases initialization time since it has to decompress
2012-10-07T14:22:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-07T14:35:30 -!- Netsplit *.net <-> *.split quits: @ChanServ
2012-10-07T14:43:34 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-07T14:44:52 -!- Netsplit over, joins: @ChanServ
2012-10-07T15:03:10 < Thorn> looks like LPC17xx datasheet allows short circuit on gpio pins with current up to 50mA per pin
2012-10-07T15:14:33 < dongs> ya
2012-10-07T15:16:28 < dongs> so fedex  didnt show up today with my F3s
2012-10-07T15:16:33 < TitanMKD> Thorn but i imagine with a total maximum of 150mA
2012-10-07T15:16:35 < dongs> probably for hte better, since im still fucking raged
2012-10-07T15:16:41 < dongs> that teh fucking things rae not pin compatible
2012-10-07T15:16:51 < dongs> im debating waht to do, just reject the package and let mouser eat shipping or...
2012-10-07T15:18:48 < Thorn> TitanMKD: definitely something like that
2012-10-07T15:45:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
2012-10-07T15:51:30 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-07T15:51:30 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-07T15:51:30 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-07T16:10:36 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-10-07T16:14:38 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-07T17:16:22 -!- mrcan_ [~mrcan@78.173.72.192] has joined ##stm32
2012-10-07T17:16:22 -!- mrcan_ [~mrcan@78.173.72.192] has quit [Changing host]
2012-10-07T17:16:22 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-07T17:19:55 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 260 seconds]
2012-10-07T17:25:38 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-07T17:30:45 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-07T17:40:07 -!- barthess [~barthess@5.100.212.125] has joined ##stm32
2012-10-07T17:46:42 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-07T17:51:58 < Thorn> so the 10-pin debug connector lacks /TRST. is this ok to omit it for jtag?
2012-10-07T17:52:22 < Thorn> (especially from the viewpoint of openocd)
2012-10-07T17:52:38 < qyx_> you need only tdi, tdo, tms, tck, gnd
2012-10-07T17:52:47 < qyx_> i use it that way and it works
2012-10-07T17:56:59 < Thorn> interestingly, the 20-pin connector has trst but no reset, and the 10-pin one is the other way round
2012-10-07T17:58:30 -!- daku is now known as DaKu
2012-10-07T17:58:39 < Thorn> "Missing TRST is not a problem, since JTAG-level resets can be triggered using with TMS signaling." <- openocd guide 9.2
2012-10-07T17:58:54 < Thorn> (sic)
2012-10-07T17:59:44 < Thorn> "Many systems treat combined assertion of SRST and TRST as a trigger for a harder reset than SRST alone."
2012-10-07T17:59:48 < Thorn> wat
2012-10-07T18:00:26 < dongs> you gotta go deeper
2012-10-07T18:00:32 < dongs> its a reset within a reset.
2012-10-07T18:00:35 < TitanMKD> anyway it is really better to trig a real RESET
2012-10-07T18:00:57 < TitanMKD> to be sure the MCU has hw reset instead of a fake reset with for example peripherals stuff running ...
2012-10-07T18:01:06 < Bird|lappy> Thorn, yeah.  you can use TMS and TCK to clock the TAP into its reset state
2012-10-07T18:01:23 < Bird|lappy> SRST is kind of important though, yes
2012-10-07T18:02:04 < TitanMKD> even SWD require NRST
2012-10-07T18:02:19 < TitanMKD> and very useful also to reset all the chip behind ;)
2012-10-07T18:04:30 < Thorn> my debugger should feature a target reset button then, so that I don't have to design one on every board
2012-10-07T18:04:51 < TitanMKD> Thorn it is really better to connect it to NRST too ;)
2012-10-07T18:05:07 < TitanMKD> in order to to real hw reset on the MCU and other peripherals behind if any
2012-10-07T18:05:44 < TitanMKD> especially to debug startup code and use hw reset to synchronize at startup and stop all ...
2012-10-07T18:06:35 < Thorn> okay, let's approach it from another direction. do real world debuggers actually use TRST?
2012-10-07T18:06:59 < TitanMKD> yes all REAL debugger use it
2012-10-07T18:07:20 < TitanMKD> but some hw just have it not connected ;)
2012-10-07T18:07:42 < Thorn> well in this case the official recommended connector doesn't have this signal
2012-10-07T18:08:06 < TitanMKD> for me it is connected
2012-10-07T18:09:31 < dongs> i duno dude
2012-10-07T18:09:31 < TitanMKD> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka14540.html
2012-10-07T18:09:38 < dongs> i have 10pin swd port
2012-10-07T18:09:42 < TitanMKD> it is true 10pins connector does not have it
2012-10-07T18:09:49 < TitanMKD> ha no ;)
2012-10-07T18:09:52 < TitanMKD> pin 10 RESET
2012-10-07T18:10:00 < dongs> and pin10 goes to nrst
2012-10-07T18:10:07 < dongs> and i have no problem stepping through startup etc.
2012-10-07T18:10:13 < Thorn> that's system reset, not jtag tap reset
2012-10-07T18:10:16 < dongs> i do not connect jtrst or wahtever
2012-10-07T18:10:18 < dongs> yeah, shrugging
2012-10-07T18:10:21 < dongs> it works
2012-10-07T18:10:23 < dongs> and i dont care
2012-10-07T18:10:32 < Thorn> dongs: you're only using swd, aren't you?
2012-10-07T18:10:35 < dongs> of course
2012-10-07T18:10:50 < dongs> swd+swo
2012-10-07T18:10:52 < Thorn> my question if it's ok for jtag mode.
2012-10-07T18:10:55 < Thorn> +is
2012-10-07T18:10:57 < dongs> dunno ~
2012-10-07T18:11:04 < TitanMKD> dongs of course it work it is just not clean to reset the hw without it
2012-10-07T18:11:31 < TitanMKD> even if OpenOCD can write some instructions to do an internal reset in the MCU
2012-10-07T18:14:56 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32
2012-10-07T18:32:04 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-07T18:51:27 -!- barthess [~barthess@5.100.212.125] has quit [Quit: Leaving.]
2012-10-07T18:56:58 -!- barthess [~barthess@5.100.212.125] has joined ##stm32
2012-10-07T19:08:42 < Thorn> lpc17xx have an interesting pinout. one side is almost exclusively jtag/swd, crystals, vref, vbat etc. GPIO pins are on remaining 3 sides
2012-10-07T19:09:14 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-07T19:10:27 < dongs> Thorn: Silabs stuff is even better, you can reassign any pin to *
2012-10-07T19:16:11 < Thorn> how many cortex families are there anyway? stm32, lpc1xxx, lm3s/lm4s (stellaris),  efm32, kinetis, silabs, atmel sam3, ...?
2012-10-07T19:16:39 < Bird|lappy> Nuvoton has one...
2012-10-07T19:17:09 < Bird|lappy> I think Holtek is coming out with one too?
2012-10-07T19:17:39 < Thorn> actel smartfusion
2012-10-07T19:17:58 < Bird|lappy> oh yeah, and PSoC 5 too
2012-10-07T19:22:40 < Thorn> need to update the wiki
2012-10-07T19:25:27 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-07T19:26:42 < cjbaird> 8 BITS OF AWESOME: http://www.sunrise-ev.com/MembershipCard.htm .. http://www.retrotechnology.com/memship/memship.html
2012-10-07T19:29:34 < dongs> OLD
2012-10-07T19:29:38 < dongs> you pasted this like 24 hours ago
2012-10-07T19:30:49 < cjbaird> I must've been asleep then.
2012-10-07T19:31:02 < cjbaird> Kit Enigma machines are around too: https://secure.flickr.com/photos/ptdecker/5386483628/
2012-10-07T19:36:45 < dongs> lampenfeld. lol.
2012-10-07T19:45:58 < emeb_mac> next up - hacking the ENIGMA.
2012-10-07T19:46:42 < emeb_mac> using only woodworking tools...
2012-10-07T19:57:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-07T19:58:58 < cjbaird> Wikipedia just glitched on me..
2012-10-07T20:01:30 < Laurenceb_> the wiki page on Zeigler-Nichols seems to be wrong...
2012-10-07T20:08:06 -!- corecode_ is now known as corecode
2012-10-07T20:13:22 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-07T20:24:20 < Robint91> Laurenceb_ What is wrong?
2012-10-07T20:24:41 < Laurenceb_> the equations in the table seem to be incorrect
2012-10-07T20:26:17 < Robint91> Laurenceb_, In correct meaning, different scaling factors or just plain wrong equations
2012-10-07T20:26:34 < Laurenceb_> they seem to be plain wrong
2012-10-07T20:26:42 < Laurenceb_> like someone screwed up
2012-10-07T20:26:51 < Laurenceb_> the I term seems to be upside down
2012-10-07T20:30:39 < Robint91> Laurenceb_ It seems that it is correct.
2012-10-07T20:31:16 < Robint91> Laurenceb_, mind watch out if you got a series/parallell controller
2012-10-07T20:31:17 < cjbaird> Something I saw cool yesterday while shopping: there was an kid with a cochlear implant (much like http://www.hearingpocket.com/images/baby2.jpg) .. however the kid's was all chromed and styled-up like a SF cybernetic implant.. He looked a lot like Adam Jensen in Deus Ex..
2012-10-07T20:33:40 < dongs> this shit needs pics
2012-10-07T20:36:32 < cjbaird> The implant was a lot like this http://i.imgur.com/b3a7c.jpg ..expect with chrome edging.
2012-10-07T20:37:52 < dongs> nice
2012-10-07T20:38:12 < dongs> whats with the flat thing on teh back
2012-10-07T20:38:24 < dongs> why cant they just put all the shit into the thing around the ear
2012-10-07T20:38:24 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Read error: No route to host]
2012-10-07T20:38:40 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-07T20:39:08 < cjbaird> They did-- there's an inductive circuit into the ear that the flat thing powers through the skull.
2012-10-07T20:40:38 < dongs> awesome.
2012-10-07T20:43:25 < dongs> This is used to transmit the processed sound information over a radio frequency link to the internal portion of the device. Radio frequency is used so that no physical connection is needed, which reduces the chance of infection and pain. The transmitter attaches to the receiver using a magnet that holds through the skin.
2012-10-07T20:43:32 < dongs> sounds like shit's wide open for hax0ring
2012-10-07T20:43:46 < dongs> you could start playing "hey everybody im looking at gay porno" into every one of those guys ears
2012-10-07T20:43:48 < cjbaird> I gather they're rather crude and brute-force, just stimulating the cochlear nerves with 'buzz', rather than trying to select the particular nerves used for different frequences along the canal.
2012-10-07T20:44:36 < cjbaird> No more shitty earphones that won't still on your ears-- just bluetooth it directly into your skull! \o/
2012-10-07T20:44:44 < cjbaird> *won't sit in
2012-10-07T20:45:18 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-07T20:47:42 < cjbaird> Definitely a lot more useful to hack than the guy who got the source code for his pacemaker..
2012-10-07T21:19:34 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-07T21:20:02 < emeb> cjbaird: I thought the implants had a few frequency bands - not as fine resolution as natural hearing though
2012-10-07T21:26:07 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 240 seconds]
2012-10-07T21:28:30 <+dekar_> Does anyone know whether you're allowed to distribute the ST firmware library with open source projects?
2012-10-07T21:28:42 <+dekar_> Some of you are doing that, aren't you?
2012-10-07T21:29:03 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-07T21:31:36 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-07T21:31:54 < zippe> doing != allowed
2012-10-07T21:37:37 < Laurenceb_> i have an example project that compiles it separately and links
2012-10-07T21:37:58 < Thorn> WTF. lpc uarts don't have transmission complete interrupts.
2012-10-07T21:38:25 < Thorn> (except one which is fancier than others)
2012-10-07T21:39:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-07T21:41:47 < zyp> Thorn, but that's not very interesting, the THRE interrupt is a lot more useful
2012-10-07T21:42:21 < zyp> what you usually want to know is when you can write another byte
2012-10-07T21:42:29 < Thorn> I need a rs-485 direction signal basically
2012-10-07T21:42:42 < Thorn> which is not the same thing at all
2012-10-07T21:42:57 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-07T21:43:36 < Thorn> st has an appnote on exactly how to do that using a transmission complete interrupt. nxp has one uart which can do that automatically but I need more than one
2012-10-07T21:43:40 < zyp> that's why you have one with rs-485 mode
2012-10-07T21:44:59 < Thorn> one is not enough
2012-10-07T21:48:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-07T21:51:11 < Laurenceb_> http://www.pimpwar.com/
2012-10-07T21:51:13 < Laurenceb_> ^lol
2012-10-07T21:51:36 <+Steffann> No Laurenceb in the top 10?
2012-10-07T21:51:49 < Laurenceb_> not _yet_
2012-10-07T21:52:17 <+Steffann> You should start your own Criminals wannabee website
2012-10-07T22:03:07 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-07T22:03:07 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-07T22:03:07 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-07T22:07:48 <+dekar_> Laurenceb_, do you include it in your repo?
2012-10-07T22:07:55 < Thorn> and that UART1 doesn't have a transmission interrupt either. if you need to control driver and receiver separately (such as when you need to implement a sleep mode for the transceiver chip) you'd need external logic to turn receiver off
2012-10-07T22:08:00 < Laurenceb_> yes
2012-10-07T22:08:13 < Thorn> *transmission complete itnterrupt
2012-10-07T22:10:03 <+dekar_> Laurenceb_, their license sounds somewhat like an NDA to me
2012-10-07T22:10:17 <+dekar_> it speaks about protecting the software from disclosure etc
2012-10-07T22:10:27 < Laurenceb_> ...
2012-10-07T22:10:31 <+dekar_> that's why I'm asking
2012-10-07T22:10:35 < Laurenceb_> then why do they publish it
2012-10-07T22:10:52 <+dekar_> I plan to also distribute it with my repo, but I am not sure whether I am allowed
2012-10-07T22:11:26 <+dekar_> Laurenceb_, why does microsoft distribute windows? You aren't allowed to upload windows to your github repo either ;)
2012-10-07T22:14:42 <+dekar_> Laurenceb_, http://www.st.com/internet/com/LEGAL_RESOURCES/LEGAL_AGREEMENT/LICENSE_AGREEMENT/software_license_agreement_liberty_v2.pdf
2012-10-07T22:15:02 < Laurenceb_> eww
2012-10-07T22:15:20 <+dekar_> http://www.st.com/internet/com/LEGAL_RESOURCES/LEGAL_AGREEMENT/LICENSE_AGREEMENT/software_license_agreement_liberty_v2.pdf
2012-10-07T22:15:26 <+dekar_> wrong paste
2012-10-07T22:15:43 <+dekar_> can't copy paste from that pdf O.o
2012-10-07T22:16:20 <+dekar_> OWNERSHIP AND COPYRIGHT. Title to the Licensed Software, related documentation and all copies thereof remain with ST and/or its licensors. You may not remove the copyrights notices from the Licensed Software and to any copies of the Licensed Software. You agree to prevent any unauthorized copying of the Licensed Software and related documentation.
2012-10-07T22:16:48 < Laurenceb_> hmm
2012-10-07T22:16:57 < Laurenceb_> well ill wait for them to take me to court
2012-10-07T22:17:36 < Laurenceb_> http://hackaday.com/2012/10/02/weather-balloon-payload-that-almost-guides-itself-back-to-you/
2012-10-07T22:17:38 <+dekar_> I would assume they don't care and it's just the lawyers pretending to be important
2012-10-07T22:17:46 < Laurenceb_> yeah
2012-10-07T22:17:48 < Laurenceb_> "There seems to be a very large chasm between the lofty goals of the project and the reality of the video posted above."
2012-10-07T22:17:49 < Laurenceb_> hehe
2012-10-07T22:19:00 < zyp> that license seems to be intended for software that's only distributed to certain customers, not anyone
2012-10-07T22:19:19 <+dekar_> dongs, you're keeping the ST lib in your repo as well, aren't you?
2012-10-07T22:19:28 < zyp> I guess they just got lazy
2012-10-07T22:20:53 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-07T22:20:58 <+dekar_> zyp, DEFINITIONS. Licensed Software: means the enclosed SOFTWARE/FIRMWARE, EXAMPLES, PROJECT TEMPLATE (…)
2012-10-07T22:21:17 < zyp> dekar_, and?
2012-10-07T22:21:57 <+dekar_> zyp, I'd guess they just chose the license to stay in control of the software
2012-10-07T22:22:16 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 260 seconds]
2012-10-07T22:22:42 <+dekar_> they don't want people to port it, reuse it etc
2012-10-07T22:22:47 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-07T22:22:53 <+dekar_> and they want if they take it down to stay offline
2012-10-07T22:23:11 < zyp> right, that's going to happen :p
2012-10-07T22:23:55 <+dekar_> doesn't matter, if they don't want me to use it, then I won't use it
2012-10-07T22:24:19 <+dekar_> I guess I'll mail some people at ST, I think I still got some mail addresses
2012-10-07T22:24:20 < zyp> doesn't matter to me either, I'm not going to use it either way
2012-10-07T22:24:27 < Laurenceb_> http://www.engadget.com/2011/09/19/arduino-brings-the-new-goods-to-maker-faire-new-york-welcomes/
2012-10-07T22:24:29 < Laurenceb_> omg
2012-10-07T22:24:30 < Laurenceb_> omg
2012-10-07T22:24:36 < Laurenceb_> fapfapfap
2012-10-07T22:24:43 < Laurenceb_>  /sarc
2012-10-07T22:25:11 <+dekar_> Laurenceb_, what's so special about that? the ARMduino has been announced a while ago
2012-10-07T22:25:25 < Laurenceb_> im joking
2012-10-07T22:25:33 < Thorn> fuck NXP and their LPC crap. looks like I'll need some monostables
2012-10-07T22:26:53 < Laurenceb_> yeah, fuck their shit
2012-10-07T22:27:02 <+dekar_> does szczys still visit this channel? he distributes the ST lib as well and he seemed to care about licenses
2012-10-07T22:27:54 <+Steffann> Haven't seen him active for a while here
2012-10-07T22:28:03 < BrainDamage> do we have a !seen bot?
2012-10-07T22:28:06 <+Steffann> No
2012-10-07T22:28:18 <+Steffann> zlog doesn't have that feature
2012-10-07T22:28:18 < zlog> Steffann: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-10-07.html
2012-10-07T22:31:55 < zippe> The ARM arduino has been announced, yes, but it's something of a disaster by the looks of it
2012-10-07T22:34:42 < Laurenceb_> hehe
2012-10-07T22:34:44 < Laurenceb_> why?
2012-10-07T22:34:46 < zyp> how so? (not that I'm in any way surprised)
2012-10-07T22:35:27 <+Steffann> ARM arduino. That's that SAM3u something-ish arduino?
2012-10-07T22:35:33 < BrainDamage> yes
2012-10-07T22:35:41 < BrainDamage> atmel's arm
2012-10-07T22:36:15 < Thorn> olimex already makes some, including an "industrial" one
2012-10-07T22:36:28 < TitanMKD> atmel's arm are just 2 year late compared to STM32F4 or LPC43xx
2012-10-07T22:36:33 <+Steffann> Isn't the maple also sort of arduino compatible?
2012-10-07T22:36:39 < BrainDamage> minus the sort
2012-10-07T22:36:44 < BrainDamage> it's arduino compatible
2012-10-07T22:36:55 < BrainDamage> they made their lib like 3 years ago
2012-10-07T22:36:58 < BrainDamage> they simply got ignored
2012-10-07T22:37:07 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-10-07T22:37:11 <+Steffann> Arduino == $$
2012-10-07T22:37:12 <+dekar_> atmel had ARM like 10 years ago
2012-10-07T22:37:36 <+dekar_> at university I had to work with some old atmel ARM7 (ARMv4)
2012-10-07T22:38:01 <+dekar_> 32mhz iirc
2012-10-07T22:38:32 < BrainDamage> I didn't check, have they switched to 3V logic since they were there?
2012-10-07T22:39:02 <+dekar_> I think so
2012-10-07T22:39:11 -!- tavish [~tavish@59.177.6.54] has joined ##stm32
2012-10-07T22:39:11 -!- tavish [~tavish@59.177.6.54] has quit [Changing host]
2012-10-07T22:39:11 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-07T22:40:19 < TitanMKD> dekar_ anyway old ARMv4 architecture really suxx compared to new CortexM
2012-10-07T22:44:39 <+dekar_> I think it had an MMU :)
2012-10-07T22:45:07 <+dekar_> But I agree, I decided to write Thumb1 on it and it sucked big time
2012-10-07T22:45:17 <+Steffann> Does ST has something interesting upcoming devices?
2012-10-07T22:45:17 -!- tavish_ [~tavish@59.177.8.6] has joined ##stm32
2012-10-07T22:45:19 <+Steffann> Excetp F3?
2012-10-07T22:45:25 <+Steffann> -thing
2012-10-07T22:46:00 < TitanMKD> Steffann yes like an F6 @ 450Mhz cortex-M6 ;)
2012-10-07T22:46:15 <+Steffann> You wish
2012-10-07T22:46:42 < TitanMKD> it will be fun to see just an M4F with 22nm
2012-10-07T22:46:53 < TitanMKD> consumption and frequency should be crazy
2012-10-07T22:47:05 < Bird|lappy> I don't know why
2012-10-07T22:47:07 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-10-07T22:47:13 < Bird|lappy> but TI's 130nm low-leakage process has me excited
2012-10-07T22:47:13 < BrainDamage> does st even have a 22nm process?
2012-10-07T22:47:30 < TitanMKD> BrainDamage or 28nm ;)
2012-10-07T22:47:45 < TitanMKD> bye
2012-10-07T22:47:51 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-07T22:47:54 < Laurenceb_> there is F5 coming
2012-10-07T22:47:59 < Laurenceb_> but not for a while
2012-10-07T22:48:16 < BrainDamage> 32 nm seems their limit
2012-10-07T22:48:51 <+Steffann> LOL, google sort of only finds this channel when you google for "stm32f5" :D
2012-10-07T22:50:35 <+Steffann> And some pdf which also seems to show an STM32L4 ( Hi Google! )
2012-10-07T22:52:45 < Laurenceb_> nice
2012-10-07T22:56:55 < Laurenceb_> STM32F6  (F U ST)
2012-10-07T22:57:22 < Laurenceb_> now they are very trolled
2012-10-07T22:57:52 <+Steffann> And now your going to monitor google?
2012-10-07T22:58:07 < Laurenceb_> yup
2012-10-07T23:03:25 <+Steffann> Great
2012-10-07T23:04:39 < zippe> Still waiting for a CortexA5 SoC
2012-10-07T23:05:16 < zippe> In theory, Freescale have their "Vybrid" SoCs due "any day now"
2012-10-07T23:05:38 < Thorn> what package?
2012-10-07T23:05:50 < Thorn> and will it support sdr sdram?
2012-10-07T23:06:26 < zippe> The VF1xxR will be LQFP144
2012-10-07T23:06:36 < zippe> 1.5MiB of internal SRAM
2012-10-07T23:07:06 < Thorn> hm
2012-10-07T23:07:23 < zyp> interesting, M4+A5
2012-10-07T23:08:30 < zippe> The VF3xx might be interesting too
2012-10-07T23:08:35 < zippe> No M4
2012-10-07T23:09:27 < zippe> And by the looks of it, no SDR, only DDR
2012-10-07T23:11:21 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-07T23:14:31 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-07T23:16:05 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Client Quit]
2012-10-07T23:16:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-07T23:16:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Excess Flood]
2012-10-07T23:16:40 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-07T23:33:35 -!- tavish_ [~tavish@59.177.8.6] has quit [Quit: leaving]
2012-10-07T23:42:47 < Robint91> Laurenceb_ F4?
2012-10-07T23:42:51 < Robint91> *F5?
2012-10-07T23:42:53 <+dekar_> https://github.com/szczys/stm32f0-discovery-basic-template/issues/5
2012-10-07T23:58:52 < upgrdman> i've used that f0 collection. works well.
--- Day changed Mon Oct 08 2012
2012-10-08T00:07:36 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Read error: Connection reset by peer]
2012-10-08T00:11:10 -!- barthess [~barthess@5.100.212.125] has quit [Quit: Leaving.]
2012-10-08T00:24:23 <+dekar_> upgrdman, I wonder about the license of the ST lib, that's all. And he distributes them with that template
2012-10-08T00:25:24 < upgrdman> ya, legit q
2012-10-08T01:04:54 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-08T01:23:53 -!- Laurenceb_ [~Laurence@host86-177-214-10.range86-177.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-08T01:26:42 -!- DaKu is now known as daku
2012-10-08T01:34:50 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-08T01:46:29 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-08T02:03:12 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-08T02:17:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T02:43:43 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Disconnected by services]
2012-10-08T02:43:44 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-08T02:44:16 -!- Nutter` is now known as Nutter
2012-10-08T03:03:12 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-10-08T03:25:37 < upgrdman> f0: with TIM17 i could not get it to trigger and reset on TI1, (SMCR TS=101, SMCR SMS=100) but it worked with TIM2. RM0091 leads me to believe it should have worked with TIM17, any ideas why it didn't?
2012-10-08T04:01:15 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T04:04:48 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 260 seconds]
2012-10-08T04:14:48 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-08T04:15:04 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-08T04:15:20 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T04:16:42 < dongs> dekar_: yes
2012-10-08T04:23:08 < qyx_> hm, that mcp9808 worked at first try \o/
2012-10-08T04:25:07 < qyx_> on
2012-10-08T04:33:56 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 255 seconds]
2012-10-08T04:36:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Read error: Connection reset by peer]
2012-10-08T05:00:07 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-08T05:23:59 < upgrdman> any got pwm output working on the f0?
2012-10-08T05:24:09 < upgrdman> it works fine for me on the f4, but not on the f0
2012-10-08T05:41:43 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-08T05:54:06 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-08T06:06:32 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-08T06:29:18 < dongs> emeb_mac: beep
2012-10-08T06:29:31 < emeb_mac> whazzup?
2012-10-08T06:29:31 < dongs> emeb_mac: I can have a choice of 9221SS or TE.. SS = ssop24, TE = tsop24/exposed pad
2012-10-08T06:29:35 < dongs> which one should I go for?
2012-10-08T06:29:39 < dongs> price diff is neglibible
2012-10-08T06:30:00 < emeb_mac> exposed pad  is for heatsink, right?
2012-10-08T06:30:04 < dongs> (though SS seems eo bte more available)
2012-10-08T06:30:05 < dongs> yes, heatsink
2012-10-08T06:30:18 < emeb_mac> question is if it makes any difference to your mfg process.
2012-10-08T06:30:27 < dongs> no difference here
2012-10-08T06:30:31 < dongs> since its paste/reflow anyway
2012-10-08T06:30:34 < emeb_mac> yep
2012-10-08T06:30:44 < dongs> ive got other shit with EP parts.
2012-10-08T06:31:36 < emeb_mac> probably worth a look at the datasheet to see if the expected power dissipation will be an issue.
2012-10-08T06:31:58 < dongs> let me see what the distributor says for availability..
2012-10-08T06:32:07 < dongs> i dont wanna end up having to pay extra if TE is not widely used
2012-10-08T06:32:16 < emeb_mac> yep
2012-10-08T06:32:43 < dongs> 70.5 (SS:SSOP24-150mil-0.635mm)
2012-10-08T06:32:47 < dongs> 31 (TE:TSSOP24-173mil-0.65mm (EP))
2012-10-08T06:32:51 < dongs> C/W
2012-10-08T06:34:49 < dongs> page 22
2012-10-08T06:34:53 < dongs> has some graphs about power dissipation
2012-10-08T06:35:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-08T06:35:11 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-08T06:36:41 < dongs> so it'll be 70C at 1W..
2012-10-08T06:36:45 < dongs> but is that gonna be dissipating 1W?
2012-10-08T06:37:17 < dongs> hmm ssop24 is worst
2012-10-08T06:37:32 < dongs> only other problem is EP part will be even more annoying to route, though not by much
2012-10-08T06:38:46 < dongs> ok looks liek if im gonna do it right i better use TE part.
2012-10-08T06:39:33 < dongs> TE it is.
2012-10-08T06:39:44 < dongs> the guy says no problem getting it just takes an extra 2-3 days.
2012-10-08T06:42:24 < emeb_mac> with all 12 channels running it would be good to have the extra heatsinking.
2012-10-08T06:42:31 < dongs> right
2012-10-08T06:42:43 < dongs> [ buy it now ] on 100x TE
2012-10-08T06:43:23 < emeb_mac> the high-side PFETs have only a 1/8 duty cycle, but the 9221 runs 100%
2012-10-08T06:43:35 < dongs> yep
2012-10-08T07:11:33 < upgrdman> "TIL dead elephants are interesting"
2012-10-08T07:19:52 < upgrdman> lo
2012-10-08T07:19:53 < upgrdman> lol
2012-10-08T07:19:58 < upgrdman> link?
2012-10-08T07:20:10 < upgrdman> no tv
2012-10-08T07:26:04 < dongs> d o  n   g    s
2012-10-08T07:26:20 < dongs> flyback: ooooooooold
2012-10-08T07:26:30 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 264 seconds]
2012-10-08T07:27:17 < dongs> yep
2012-10-08T07:27:49 < dongs> http://www.youtube.com/watch?v=yqFDBLnRvDE this sorta shit
2012-10-08T07:39:42 < dongs> um
2012-10-08T07:39:45 < dongs> why are all 30x parts "preview"
2012-10-08T07:40:38 < dongs> except the fuckign 100lqfp
2012-10-08T07:40:39 < dongs> damn
2012-10-08T07:41:28 < upgrdman> what does that mean? silicon is not final?
2012-10-08T07:42:12 < zippe> upgrdman: not in mass production yet
2012-10-08T07:42:18 < zippe> Probably still engineering samples only
2012-10-08T07:49:32 < dongs> http://blog-imgs-51.fc2.com/y/a/r/yaraon/1349615828159.jpg attn zyp
2012-10-08T08:12:05 < gnomad> anime girls get me hot too
2012-10-08T08:15:55 < upgrdman> lol
2012-10-08T08:16:18 < upgrdman> dongs and flyback: you guys make irc fun
2012-10-08T08:17:49 < upgrdman> flyback, what channels do you regular?
2012-10-08T08:17:57 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 256 seconds]
2012-10-08T08:32:17 < dongs> http://www.kickstarter.com/projects/18182218/freesoc-and-freesoc-mini
2012-10-08T08:32:18 < dongs> lol
2012-10-08T09:10:30 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-08T09:21:35 < emeb_mac> it's the best microcontroller in the world!
2012-10-08T09:25:00 < dongs> freeSoC Mini offers 24 general purpose I/O pins and 8 special I/O pins. It's dimensioned exactly to fit in 40-pin DIP sockets, and can act as a drop-in replacement for an 8051 microcontroller.
2012-10-08T09:25:05 < dongs> totally a sales point right there
2012-10-08T09:25:11 < dongs> i'm gonna get all my 8051's replaced asapa
2012-10-08T09:27:32 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 246 seconds]
2012-10-08T09:40:11 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-08T09:42:35 < emeb_mac> I have so many to replace.
2012-10-08T09:56:56 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-10-08T10:00:07 < ABLomas> looks cool, but price...
2012-10-08T10:02:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-08T10:07:00 < Thorn> LPC17xx have separate power for core and i/o and you can disconnect i/o power and keep the core running
2012-10-08T10:12:40 < dongs> https://www.google.com/search?q=stm32f3%20discovery&sugexp=chrome,mod%3D9&um=1&ie=UTF-8&hl=en&tbm=isch&source=og&sa=N&tab=wi&authuser=0&ei=1XxyUML8EY_JmQXWy4GYBw&biw=1058&bih=662&sei=13xyUPu9I-bKmgXltoCQBg
2012-10-08T10:12:44 < dongs> haha
2012-10-08T10:14:54 < dongs> scroll down to page 3
2012-10-08T10:19:02 < zyp> also your pony on page 2
2012-10-08T10:19:23 < dongs> right
2012-10-08T10:19:59 < dongs> page 3 has happy 9/11 shit
2012-10-08T10:20:00 < dongs> for me
2012-10-08T10:26:36 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-08T10:26:36 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-08T10:26:36 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-08T10:42:28 -!- claude is now known as claude_work
2012-10-08T10:45:17 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 244 seconds]
2012-10-08T11:23:34 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-08T11:23:38 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-08T11:39:14 < Thorn> <scrts> our ST cortex-m3 microcontroller is capable of doing 97Mbps multicast streaming plus MPEG-TS PID filtering plus HTTP server plus respond to ping within <1ms
2012-10-08T11:39:20 < Thorn> ^^ from ##fpga
2012-10-08T11:39:53 <+Steffann> uh?!
2012-10-08T11:44:44 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T11:46:34 -!- dekar_ [~dekar@212.255.19.185] has quit [Quit: This computer has gone to sleep]
2012-10-08T11:47:39 -!- tavish3 [~tavish@59.177.69.211] has joined ##stm32
2012-10-08T11:57:48 < claude_work> and the cortex-m3's only job is to download the bitstream into the fpga? ;)
2012-10-08T11:59:49 < Thorn> apparently not lol. it receives the raw mpeg stream from it aiui
2012-10-08T11:59:59 < claude_work> cool
2012-10-08T12:21:53 -!- ntfreak_ is now known as ntfreak
2012-10-08T12:45:39 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-252-37.clienti.tiscali.it] has joined ##stm32
2012-10-08T12:49:33 -!- dekar [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has joined ##stm32
2012-10-08T12:49:34 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-08T12:49:54 <+Steffann> Oeh, dekar is from the fbi
2012-10-08T12:51:40 <+dekar> Steffann, how do you know? It's called Bundeskriminalamt btw :P
2012-10-08T12:52:07 <+Steffann> fbi.h-da.de <= fachbereich informatik :P
2012-10-08T12:52:08 <+dekar> oh my hostname :)
2012-10-08T12:59:23 -!- dekar_ [~dekar@studpool-wlan-74-252.fs.fbi.h-da.de] has joined ##stm32
2012-10-08T12:59:30 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-08T13:02:25 -!- dekar__ [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has joined ##stm32
2012-10-08T13:02:28 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-10-08T13:03:06 -!- dekar [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has quit [Ping timeout: 264 seconds]
2012-10-08T13:04:11 -!- dekar_ [~dekar@studpool-wlan-74-252.fs.fbi.h-da.de] has quit [Read error: Connection reset by peer]
2012-10-08T13:04:22 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T13:17:58 < dongs> china doesnt fuck around http://i.imgur.com/VYLPu.png
2012-10-08T13:18:50 < karlp> what is that?
2012-10-08T13:19:05 < dongs> the drawing for a cable they're makign for me
2012-10-08T13:21:45 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-08T13:24:29 < Thorn> what is the 100+-5 dimension? does it imply all the cable lenghts are actually different?
2012-10-08T13:24:43 < dongs> ya, lol
2012-10-08T13:24:49 < dongs> probably tolerance for crimping
2012-10-08T13:25:08 < Thorn> 100, not 10
2012-10-08T13:25:40 < Thorn> seems to be a horizontal projection of cable lengths
2012-10-08T13:26:07 < Thorn> which are laid out such that it's the same for all of them
2012-10-08T13:27:15 < dongs> 100mm is length of the shit
2012-10-08T13:36:53 -!- daku is now known as DaKu
2012-10-08T13:38:35 < Laurenceb> omg
2012-10-08T13:38:41 < Laurenceb> plastic electron factory
2012-10-08T13:38:48 < dongs> i kno rite
2012-10-08T13:39:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T13:39:20 < dongs> MANUFACTORY even.
2012-10-08T13:39:47 < dongs> Laurenceb: do you participate in dooking on oct31
2012-10-08T13:40:01 < Laurenceb> devils night?
2012-10-08T13:40:42 < Laurenceb> petrol is too expensive these days :(
2012-10-08T13:46:58 < Laurenceb> http://www.bbc.co.uk/newsbeat/19866191
2012-10-08T13:47:00 < Laurenceb> the hell
2012-10-08T13:50:33 < dongs> wat
2012-10-08T13:51:34 < karlp> fuck that shit!
2012-10-08T13:52:52 < dongs> Laurenceb: http://www.instructables.com/id/How-to-Make-Super-Cool-Frozen-Party-Cocktails-with/ here you go
2012-10-08T13:52:55 < dongs> DIY
2012-10-08T14:17:46 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Remote host closed the connection]
2012-10-08T14:18:13 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T14:18:32 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T14:21:14 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-10-08T14:23:58 -!- dekar__ [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-08T14:24:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T14:24:56 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-08T14:25:19 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-08T14:34:46 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-08T14:40:09 < Thorn> dongs: what grid do you use in diptrace when routing traces from qfp-0.5mm packages?
2012-10-08T14:40:39 < dongs> i dont usually use grid when routing.. dunno why, im weird
2012-10-08T14:40:55 < dongs> only when I route like dip stuff
2012-10-08T14:41:09 < dongs> usually my stuff is so small having a grid would just get in the way..
2012-10-08T14:41:17 < dongs> with like 6/6 traces
2012-10-08T14:49:31 < Thorn> almost done http://i.imgur.com/OIKvl.png
2012-10-08T14:50:11 < Thorn> 21 LEDs x_x
2012-10-08T14:50:16 < dongs> not bad
2012-10-08T14:50:30 < dongs> wai so many led
2012-10-08T14:51:25 < Thorn> I like LEDs
2012-10-08T14:51:30 < Thorn> I even collected pictures of telecom/network equipment for LED porn at one time
2012-10-08T14:51:46 < dongs> what hte hell are those ovals
2012-10-08T14:51:50 < dongs> is that assembly layer
2012-10-08T14:52:04 < dongs> hm no they dont look connected to patterns
2012-10-08T14:52:28 < dongs> I like your component patterns
2012-10-08T14:52:35 < Thorn> they are reminders to check these patterns
2012-10-08T14:53:46 < dongs> i dunno if i'd route that shit between so8 legs
2012-10-08T14:54:56 < Thorn> DRC doesn't complain (and I think I've done that before and it was ok)
2012-10-08T14:55:14 < zyp> external pullups on jtag signals?
2012-10-08T14:55:32 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-08T14:56:32 < dongs> ya who the fuck needs those
2012-10-08T14:56:37 < Thorn> why not
2012-10-08T14:57:01 < zyp> the question is why?
2012-10-08T14:57:02 < dongs> not needed on stm afaik..
2012-10-08T14:57:32 < Thorn> well lpc docs do say those pins have internal pullups
2012-10-08T14:57:54 < dongs> i might have to steal your diode pattern
2012-10-08T14:57:59 < dongs> i like the idea of polarity dot inside
2012-10-08T14:58:01 < dongs> isntead of outside.
2012-10-08T14:59:45 < Thorn> lpc was not the best choice for the project tbh. shady usb stack and stupid UARTs
2012-10-08T15:00:16 < Thorn> but I did like lots of endpoints and wasn't going to research yet another mcu
2012-10-08T15:00:34 < Thorn> hopefully I can make it work
2012-10-08T15:01:58 < dongs> teehee
2012-10-08T15:02:17 < HTT-Bird> dongs: why a dot and not just drawing a diode symbol in the silkscreen?
2012-10-08T15:02:40 < dongs> HTT-Bird: lol beacuse the parts are 0603/0805
2012-10-08T15:02:58 < HTT-Bird> dongs: ah.  what I did with the SOD-523 is split the diode symbol in two and put one half on each side
2012-10-08T15:03:17 < dongs> i dont re ally care for diode symbol there
2012-10-08T15:03:20 < dongs> but shit like leds dont need a dot
2012-10-08T15:05:45 < dongs> stolen: http://i.imgur.com/JLrA9.png
2012-10-08T15:05:52 < Laurenceb> Necrophilia: Better late than never.
2012-10-08T15:05:56 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-10-08T15:06:22 < Thorn> how do you tell A from K in a smd led reliably anyway?
2012-10-08T15:06:28 < dongs> dot inside
2012-10-08T15:06:39 < dongs> also pattern on the bottom
2012-10-08T15:06:46 < dongs> usually a triangle or a T or similar
2012-10-08T15:07:05 < Thorn> what side the dot is?
2012-10-08T15:07:36 < Thorn> in my experience it can be anywhere (chinese leds though)
2012-10-08T15:07:55 < zyp> the side specified in the datasheet.
2012-10-08T15:08:09 < dongs> Thorn: havent noticed any problem here.
2012-10-08T15:08:13 < dongs> but mine are on reel and machine-placed
2012-10-08T15:08:21 < dongs> but they always have the little green mark on them.
2012-10-08T15:08:24 -!- dekar [~dekar@2001:41b8:9bf:fe74:da1:6d70:272d:d928] has joined ##stm32
2012-10-08T15:08:27 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-08T15:08:29 < dongs> dekar: status
2012-10-08T15:08:44 < Thorn> I think mine were even placed randomly in their tape.
2012-10-08T15:08:57 < Thorn> datasheet? lol.
2012-10-08T15:09:27 < dongs> haha, thats fucked up
2012-10-08T15:09:28 < Thorn> ebay page with some random specs
2012-10-08T15:09:32 < dongs> i get quality shit..
2012-10-08T15:09:35 < dongs> everbright? some taiwan maker
2012-10-08T15:09:38 < dongs> like 12 bucks for 3k reel
2012-10-08T15:10:00 < dongs> the plastic probably costs mroe
2012-10-08T15:10:02 < dongs> and they're fucking bright
2012-10-08T15:10:19 < Thorn> I want that
2012-10-08T15:11:03 <+dekar> dongs, status? math lecture, university
2012-10-08T15:11:43 <+dekar> dongs, also ST granted me to freely distribute their firmware library :)
2012-10-08T15:12:42 < dongs> http://www.digikey.com/product-search/en?x=0&y=0&KeyWords=17-21SURC%2FS530-A3%2FTR8
2012-10-08T15:12:50 < dongs> dekar: yea i saw some blog about that
2012-10-08T15:12:56 < dongs> dekar: everyone and their mother includes a copy of st lib
2012-10-08T15:12:59 < dongs> i dont think tehy give a fuck
2012-10-08T15:13:02 < dongs> did you get it in writing or something?
2012-10-08T15:14:21 < dongs> LED BLUE	17-21/BHC/AN1P2/3T	3000
2012-10-08T15:14:21 < dongs> LED RED		17-21SURC/S530-A3/TR8	3000
2012-10-08T15:14:21 < dongs> LED GREEN	17-215-GHC-YR1S2-3T	3000
2012-10-08T15:14:24 < dongs> these are hte ones im using
2012-10-08T15:14:26 < dongs> fucking awesomes hit.
2012-10-08T15:14:40 < dongs> lol @ digikey price for them.
2012-10-08T15:14:45 < BrainDamage> dekar: but does it mean anyone can do it as well, or everyone has to request that individually?
2012-10-08T15:14:53 < BrainDamage> because the latter sounds counter producing
2012-10-08T15:15:21 <+dekar> BrainDamage, "We confirm that you can redistribute our library without any issue."
2012-10-08T15:15:40 < BrainDamage> thanks
2012-10-08T15:15:57 <+dekar> they seem to be quite interested in how the company I work for uses the STM32, thus I'll probably have another mail or two with them
2012-10-08T15:16:12 <+dekar> I'll ask more specific about whether that applies to everyone, but I think so
2012-10-08T15:16:20 < dongs> see if you can get some free STM32F3's out of that.
2012-10-08T15:16:32 <+dekar> dongs, I'd rather take F4
2012-10-08T15:16:43 < dongs> they dont make that in lqfp48
2012-10-08T15:17:01 <+dekar> I don't mind footprints
2012-10-08T15:19:30 < Thorn> should I join 3V3 in a loop? right now it goes forward then back without connecting to itself
2012-10-08T15:19:57 < zyp> join it in a loop for infinite power
2012-10-08T15:20:11 <+dekar> dongs, some blog about what exactly? that one? http://hackaday.com/
2012-10-08T15:20:32 < Thorn> won't it increase inductance?
2012-10-08T15:20:40 < dongs> dekar: no, i mean i saw part of the st lib licesngin discussion
2012-10-08T15:20:44 < dongs> cause you highlighted my nick in it
2012-10-08T15:22:43 <+dekar> oh on github you mean?
2012-10-08T15:23:17 <+dekar> well I just asked around whether someone got a permit, not I got one from ST and I don't care anymore :)
2012-10-08T15:24:32 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-08T15:24:39 < dongs> hehe
2012-10-08T15:29:05 <+dekar> hack a day recently stated their posts would often take the servers they link to down
2012-10-08T15:29:12 <+dekar> I wonder whether mine will stand the traffic
2012-10-08T15:30:02 < BrainDamage> I'd imagine many hacking web sites are hosted on home asymmetrical connections
2012-10-08T15:30:07 < BrainDamage> with crap upload capability
2012-10-08T15:30:17 < karlp> I'd imagine very few are.
2012-10-08T15:30:37 <+dekar> well they were suggesting to temporarily host at the amazon cloud
2012-10-08T15:31:05 < dongs> dekar, what is your shit youre gonna hackaday
2012-10-08T15:31:07 < HTT-Bird> Thorn: I wouldn't loop it
2012-10-08T15:31:27 < HTT-Bird> you'd be turning your 3v3 line into a magnetic pickup that way :P
2012-10-08T15:31:41 < HTT-Bird> (power loop, kind of like a ground loop, no?]
2012-10-08T15:31:49 < BrainDamage> minus the kind
2012-10-08T15:32:03 < dongs> YHBT-Bird
2012-10-08T15:32:10 < BrainDamage> it's the same thing
2012-10-08T15:32:25 < HTT-Bird> hehe, thought so, just wasn't sure :)
2012-10-08T15:33:56 < Thorn> thanks. my intuition didn't fail me then
2012-10-08T15:34:07 -!- DaKu is now known as daku
2012-10-08T15:37:01 <+dekar> dongs, It's already up, the topmost post
2012-10-08T15:37:07 <+dekar> OSX stuff
2012-10-08T15:37:39 < BrainDamage> do you have traffic metering?
2012-10-08T15:37:46 < BrainDamage> if so, did you notice any spike?
2012-10-08T15:38:14 <+dekar> BrainDamage, I basically killed my wordpress traffic metering by having my wordpress generate static HTML
2012-10-08T15:38:41 <+dekar> I guess I could check my access logs using webalizer or the like, but I don't really care to be honest
2012-10-08T15:39:28 -!- tavish3 [~tavish@59.177.69.211] has quit [Quit: leaving]
2012-10-08T15:42:34 <+dekar> BrainDamage, constant 3059.3 kb/s atm
2012-10-08T15:43:11 < BrainDamage> thanks, not too bad but enough to kill a home connection
2012-10-08T15:43:27 < BrainDamage> Laurenceb: http://www.seeedstudio.com/depot/nose-led-kit-p-1274.html?cPath=202
2012-10-08T15:43:50 < Laurenceb> is the asian guy free?
2012-10-08T15:44:12 <+dekar> https://dl.dropbox.com/u/28467113/traffic.png
2012-10-08T15:44:28 <+dekar> well I have 100mbit/s
2012-10-08T15:45:37 <+dekar> my nightly backups generate worse peaks ;)
2012-10-08T16:07:41  * Laurenceb facepalms
2012-10-08T16:08:05 < Laurenceb> boss just sent out email to tons of people discussing roadmaps etc
2012-10-08T16:08:20 < Laurenceb> the document referenced in the email wasnt attached
2012-10-08T16:08:24 < Laurenceb> epic fail
2012-10-08T16:09:45 < Laurenceb> couldnt look like a bigger retard
2012-10-08T16:12:16 <+dekar> well attaching the wrong file could be worse ;)
2012-10-08T16:13:02 < Laurenceb> hehe
2012-10-08T16:13:10 < Laurenceb> dongs has been at it again
2012-10-08T16:13:12 < Laurenceb> http://www.b3tards.com/u/dceb14b0c0ee9d13b69a/the-thing-jong-un.jpg
2012-10-08T16:13:27 < dongs> haha
2012-10-08T16:13:36 < dongs> linking to every korean i know
2012-10-08T16:14:56 < dongs> (which is like 2)
2012-10-08T16:18:21 < Laurenceb> http://s1.b3ta.com/host/creative/65077/1349350501/bambihumper.gif
--- Log closed Mon Oct 08 16:27:04 2012
--- Log opened Mon Oct 08 16:27:17 2012
2012-10-08T16:27:17 -!- jpa- [jpa@hilla.kapsi.fi] has joined ##stm32
2012-10-08T16:27:17 -!- Irssi: ##stm32: Total of 62 nicks [1 ops, 0 halfops, 2 voices, 59 normal]
2012-10-08T16:27:43 < dongs> some turdball already got it on a flyingthing
2012-10-08T16:27:44 <+dekar> dongs, what's that f1 board?
2012-10-08T16:27:48 <+dekar> oh
2012-10-08T16:27:52 < dongs> https://dl.dropbox.com/u/3873342/IMG_3470.jpg
2012-10-08T16:27:55 < dongs> dekar: just a breakout for  f103
2012-10-08T16:28:06 -!- Irssi: Join to ##stm32 was synced in 55 secs
2012-10-08T16:28:16 < dongs> should probably kickstarter it but its not tarduino compatible
2012-10-08T16:28:21 < dongs> so I'll get a total of $300 in donations
2012-10-08T16:29:30 < Thorn> http://www.everlight.com/  challenge: click 0805 smd leds in the menu.
2012-10-08T16:29:31 <+dekar> dongs, just flash the maple firmware
2012-10-08T16:29:36 <+dekar> makes it tarduino compatible
2012-10-08T16:29:45 < dongs> dekar: haha. ugh. such filth.
2012-10-08T16:30:05 <+dekar> I'd seriously try :P
2012-10-08T16:30:17 <+dekar> also state you can overclock it up to 136mhz :D
2012-10-08T16:30:25 < dongs> haha
2012-10-08T16:30:30 < dongs> i could never get it to do that again :(
2012-10-08T16:30:36 < BrainDamage> Thorn: http://www.everlight.com/index.php?option=com_k2&view=itemlist&layout=category&task=category&id=105&Itemid=188&lang=en
2012-10-08T16:30:36 < dongs> i swear it was working though..
2012-10-08T16:30:48 < dongs> BrainDamage: thats what I linked him
2012-10-08T16:31:00 < dongs> BrainDamage: are you using them?
2012-10-08T16:31:03 <+dekar> f1 can do 136mhz
2012-10-08T16:31:08 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-08T16:31:10 < dongs> dekar: ?
2012-10-08T16:31:13 < dongs> with what
2012-10-08T16:31:15 < BrainDamage> no he asked someone to reach the 0805
2012-10-08T16:31:23 < Thorn> thanks. I opened it by widening the browser window. if the menu opens on the left side, good luck
2012-10-08T16:31:30 < Thorn> web programming fail
2012-10-08T16:31:30 <+dekar> dongs, just up the PLL
2012-10-08T16:31:39 < BrainDamage> I have a bunch of no name chinese leds
2012-10-08T16:32:03 <+dekar> dongs, mine at 128mhz: http://dekar.wc3edit.net/flexoptix/arm-cortex-m3-OCed.jpg
2012-10-08T16:32:24 < dongs> off 8mhz xtal? or wat
2012-10-08T16:32:29 < dongs> or you have to use 24/25
2012-10-08T16:32:38 <+dekar> they can probably go even higher, though UART stopped working for me at 136mhz
2012-10-08T16:32:44 <+dekar> yeah I used 8mhz xtal
2012-10-08T16:33:03 < dongs> well shit
2012-10-08T16:33:03 <+dekar> 136 is the maximum you can do with 8mhz xtal though
2012-10-08T16:33:21 <+dekar> 128mhz seems to work fine though, I didn't spot any problems
2012-10-08T16:33:29 < dongs> ok gimme your PLL settings ill try them now on my flyingthing
2012-10-08T16:33:40 < dongs> that uses a bunch of peripherals
2012-10-08T16:33:44 < dongs> i2c, uart, pwm, dma, etc
2012-10-08T16:33:48 < dongs> would be a good test.
2012-10-08T16:33:49 <+dekar> I did that like 2 years ago, I don't think I still have any of that code
2012-10-08T16:34:24 < dongs> RCC_CFGR_PLLMULL9
2012-10-08T16:34:28 < dongs> i guess i have 8*9
2012-10-08T16:34:51 < dongs> 8*16 is the max you can do with pll multipliersr.
2012-10-08T16:35:56 <+dekar> "128Mhz is just the beginnigng . About two months ago i plugged 25Mhz crystal on board for Ethernet STM32F107 family when i was changing developmnet platform from STM32F103 at 8MHz clock. PLL was set to 72Mhz when 8MHz crystal was mounted. I did’nt change PLL’s configuration for STM32F107, and guess what? It was working ! at 225 MHz with FreeRTOS onboard. Led was blinking faster, PWM was goin faster, UART to !"
2012-10-08T16:35:57 < Thorn> dongs: and you're buying those leds for $14 / 3000??
2012-10-08T16:35:59 < dongs> its running
2012-10-08T16:36:05 < dongs> lol
2012-10-08T16:36:05 <+dekar> http://blog.tkjelectronics.dk/2010/02/stm32-overclocking/
2012-10-08T16:36:16 < dongs> CPU 128MHz, detected sensors: ACC BARO MAG ACCHW: MPU6050
2012-10-08T16:36:18 < dongs> cool
2012-10-08T16:36:34 <+dekar> dongs, I am pretty sure you can do 8*17
2012-10-08T16:36:43 <+dekar> it just lacks the define for it
2012-10-08T16:37:07 < dongs> right
2012-10-08T16:37:10 < dongs> and one next is for USBPER
2012-10-08T16:37:15 < dongs> i tried that off intrc and it didnt work :(
2012-10-08T16:37:29 <+dekar> at some point the flash will probably make problems
2012-10-08T16:37:42 <+dekar> since you can't increase the wait states further than 2
2012-10-08T16:37:49 < dongs> right.
2012-10-08T16:38:06 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-08T16:38:07 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-08T16:38:07 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-08T16:38:49 -!- BrainDamage [~BrainDama@62.10.5.141] has quit [Read error: No route to host]
2012-10-08T16:39:38 <+dekar> I have some boards with 12mhz xtals, I'll try to OC them later today as well :)
2012-10-08T16:40:20 < dongs> Thorn: sorry not THAT cheap but i just checked,, it was something like  0.01xx/led so like 30-40$/reel
2012-10-08T16:40:28 < dongs> definitely below digikey price of $270/reel...
2012-10-08T16:40:48 < Thorn> I checked my local suppliers, ~$100/reel
2012-10-08T16:41:41 < dongs> ya thats too much
2012-10-08T16:42:21 < Thorn> also they have "Brilliant Green" which I like much better than just "green"
2012-10-08T16:42:34 < dongs> the shit is really bright
2012-10-08T16:42:38 < dongs> i like it
2012-10-08T16:42:49 < Thorn> (525nm ws. 562)
2012-10-08T16:42:53 < Thorn> *vs
2012-10-08T16:43:07 < dongs> http://www.youtube.com/watch?v=-Axa7j3IBN0 of course video doesnt show anywhere near how awesome they are
2012-10-08T16:43:19 < Thorn> normal green is yellowish and this is cool green
2012-10-08T16:45:03 < dongs> now that I remember..
2012-10-08T16:45:10 < dongs> the 'weird' overclock was off 12mhz crystal
2012-10-08T16:45:15 < dongs> i was debugging USB Shit
2012-10-08T16:45:35 < dongs> and I think i got it running at something like 144mhz
2012-10-08T16:45:42 < dongs> according to how fast the led blinked
2012-10-08T16:46:16 < dongs> i was using the peripheral display window in keil for rcc/clock shit
2012-10-08T16:46:27 < dongs> and you can just choose stuff tehre directly and it gets modified realtime
2012-10-08T16:48:00 <+dekar> dongs, so change the xtal and advertise it at kickstarter as an 300mhz tarduino
2012-10-08T16:48:27 < dongs> i think 128 is plenty
2012-10-08T16:48:35 < dongs> plus its a cool round number
2012-10-08T16:48:41 <+dekar> ya
2012-10-08T16:48:52 < dongs> i wonder how much you can overclock F3 to..
2012-10-08T16:48:53 < dongs> from 72mhz
2012-10-08T16:49:14 <+dekar> I only have F1 and F4, so it's up to you to try
2012-10-08T16:49:17 < dongs> too lazy to check if the PLL shit is api-comaptible to F1
2012-10-08T16:49:20 < dongs> or F4
2012-10-08T16:49:30 < dongs> F4 can go to some crazy shit with p/q or whatever multipliers
2012-10-08T16:49:37 < dongs> but F1 only has fixed ones
2012-10-08T16:50:38 <+dekar> I OC'd my FPGA to do 14G throughput via XAUI
2012-10-08T16:56:10 < dongs> http://www.kitco.com/LFgif/au3650nys.gif aesome
2012-10-08T16:56:29 -!- BrainDamage [~BrainDama@62.10.5.141] has joined ##stm32
2012-10-08T16:57:16 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 256 seconds]
2012-10-08T17:06:14 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-08T17:17:21 <+dekar> why did the US give up on the gold standard?
2012-10-08T17:17:48 < karlp> you want a one line irc answer?
2012-10-08T17:18:14 < karlp> to a question like that?
2012-10-08T17:19:16 <+dekar> sure, but I don't understand the financial system so make it easy :)
2012-10-08T17:19:35 <+dekar> dongs, http://en.wikipedia.org/wiki/File:Gold_Index.png
2012-10-08T17:19:38 <+dekar> inflation adjusted
2012-10-08T17:21:04 < dongs> its still not bad.
2012-10-08T17:21:09 < dongs> ive got a bunch of gold.
2012-10-08T17:23:40 < dongs> http://i.imgur.com/AoNCa.jpg nice
2012-10-08T17:24:08 < karlp> what is that?
2012-10-08T17:25:08 <+dekar> it's dongs throwing around his gold in the dark :D
2012-10-08T17:25:13 < dongs> fireflies i guess.
2012-10-08T17:25:23 < dongs> was sorting through old pics and was like what is sthis shit
2012-10-08T17:31:20 < Laurenceb> jew gold
2012-10-08T17:39:24 -!- dekar_ [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has joined ##stm32
2012-10-08T17:39:29 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-08T17:42:44 -!- dekar [~dekar@2001:41b8:9bf:fe74:da1:6d70:272d:d928] has quit [Ping timeout: 268 seconds]
2012-10-08T17:45:04 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-08T17:59:16 -!- Tectu is now known as tectu
2012-10-08T17:59:30 -!- tectu is now known as Tectu
2012-10-08T17:59:33 -!- Tectu is now known as tectu
2012-10-08T17:59:56 -!- tectu is now known as Tectu
2012-10-08T18:03:09 <+Steffann> Are you sure Tectu ?
2012-10-08T18:03:14 <+Steffann> or tectu ?
2012-10-08T18:04:00 < Tectu> Steffann, hmm?
2012-10-08T18:04:05 < Tectu> Steffann, sure about?
2012-10-08T18:04:35 <+Steffann> You are your nick changes, from Tectu to tecto and back to Tectu and back to tectu and back to Tectu
2012-10-08T18:04:59 < Tectu> huh, reinstalled IRC client on new box, Steffann
2012-10-08T18:05:16 < zyp> 16:59:16 -!- Tectu is now known as tectu
2012-10-08T18:05:16 < zyp> 16:59:30 -!- tectu is now known as Tectu
2012-10-08T18:05:16 < zyp> 16:59:33 -!- Tectu is now known as tectu
2012-10-08T18:05:16 < zyp> 16:59:56 -!- tectu is now known as Tectu
2012-10-08T18:05:18 < zyp> that.
2012-10-08T18:05:24 < zyp> what the fuck was that?
2012-10-08T18:05:33 < Tectu> zyp, <3
2012-10-08T18:09:01 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-08T18:12:21 < dongs> cho-kawaii
2012-10-08T18:13:25 < Tectu> hallo, dongs
2012-10-08T18:14:40 <+Steffann> So far Japanese for dummies
2012-10-08T18:16:16 < dongs> need moar pinku bento
2012-10-08T18:18:30 < Tectu> dongs, so I guess you are gangnam style?
2012-10-08T18:18:39 <+Steffann> That's from korea
2012-10-08T18:19:03 <+Steffann> but for dongs that probably doesn't matter
2012-10-08T18:20:05 < Tectu> Steffann, as long as it has a proprietary license, it's okay for dongs ;)
2012-10-08T18:22:29 < dongs> Tectu: no, gundam style
2012-10-08T18:24:49 < Tectu> dongs, maybe HotDog-Condom-Style?   http://www.youtube.com/watch?v=fFxlZxjiSpQ
2012-10-08T18:24:56 < dongs> old link is old
2012-10-08T18:25:00 < dongs> infact I probably pasted it in here
2012-10-08T18:25:16 < dongs> and thats a rip of the original video
2012-10-08T18:25:57 -!- tavish [~tavish@120.56.137.0] has joined ##stm32
2012-10-08T18:25:57 -!- tavish [~tavish@120.56.137.0] has quit [Changing host]
2012-10-08T18:25:57 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-08T18:29:55 < Tectu> karlp, ping
2012-10-08T18:32:53 -!- dekar_ [~dekar@studpool-wlan-74-15.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-08T18:32:59 < karlp> pong
2012-10-08T18:33:32 < karlp> ping? ping! ping ping ping ping!
2012-10-08T18:33:49 < karlp> why have you summoned me from my slumber mortal!
2012-10-08T18:33:53 < Tectu> karlp, you know about the target.xml fix for GDB repling "too long package" with OpenOCD?
2012-10-08T18:34:15 < karlp> I have heard of this,yes.
2012-10-08T18:34:27 < Tectu> it worked pretty well, not it dosen't anymore 0o
2012-10-08T18:34:34 < karlp> remember that I don't actually use openocd.
2012-10-08T18:35:02 < Tectu> you should start with
2012-10-08T18:35:06 < Tectu> there's a lot to fix :D
2012-10-08T18:35:31 < karlp> I have zero desire to work on openocd
2012-10-08T18:36:04 < karlp> so, no more pinging me with openocd problems ok? there's #openocd for that
2012-10-08T18:36:34 < Tectu> wow... sorry... didn't want to disturb you or being offensive o.O
2012-10-08T18:36:44 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-08T18:37:03 -!- claude_work is now known as claude
2012-10-08T18:46:08 < Laurenceb> http://www.oxfordmartin.ox.ac.uk/event/1402
2012-10-08T18:46:11 < Laurenceb> sold out :(
2012-10-08T18:52:54 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 260 seconds]
2012-10-08T18:55:51 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-08T18:58:46 -!- daku is now known as DaKu
2012-10-08T19:00:17 -!- tavish_ [~tavish@120.59.3.171] has joined ##stm32
2012-10-08T19:01:16 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-10-08T19:05:34 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-08T19:11:40 -!- tavish_ [~tavish@120.59.3.171] has quit [Quit: leaving]
2012-10-08T19:14:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-08T19:14:26 -!- tavish [~tavish@120.59.3.171] has joined ##stm32
2012-10-08T19:14:26 -!- tavish [~tavish@120.59.3.171] has quit [Changing host]
2012-10-08T19:14:26 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-08T19:18:10 -!- barthess [~barthess@178.154.51.107] has joined ##stm32
2012-10-08T19:30:18 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T19:30:26 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-10-08T19:32:56 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-10-08T19:39:42 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-08T19:43:40 -!- meloun [~meloun@188-195-251-164-dynip.superkabel.de] has joined ##stm32
2012-10-08T19:44:51 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal]
2012-10-08T19:55:16 < zyp> todays challenge: getting OTG_HS to work with external transceiver
2012-10-08T19:55:49 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T19:56:03 < BrainDamage> todays challenge: try not to get maimed from the failure modes of my firmware controlled DCDC
2012-10-08T19:56:19 < zyp> :)
2012-10-08T20:03:25 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T20:05:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T20:06:32 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-08T20:07:11 < TitanMKD> hello
2012-10-08T20:09:53 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-08T20:11:14 < upgrdman_> f0 gpio: these's a BSRR register to set or clear a pin, and there's also a BRR register just to clear a pin....why? isn't BRR just a duplicate with half the functionality?
2012-10-08T20:12:47 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-08T20:19:28 -!- tavish_ [~tavish@120.59.3.171] has joined ##stm32
2012-10-08T20:23:52 -!- tavish_ [~tavish@120.59.3.171] has quit [Ping timeout: 260 seconds]
2012-10-08T20:24:13 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-08T20:37:46 < zyp> in F0?
2012-10-08T20:38:51 < zyp> F1 has both BSRR and BRR, F4 doesn't, I thought F0 used the same register layout as F4?
2012-10-08T20:43:28 < zyp> hmm, interesting
2012-10-08T20:43:50 < zyp> F0 is using the same register layout as F4, except that BRR is added at the end
2012-10-08T20:44:21 < zyp> I tried poking that addr on F4 here, but it does nothing, so it's not present in F4, i.e. not documentation error
2012-10-08T20:44:48 < zyp> so it apparently got added back
2012-10-08T20:46:52 < zyp> the rationale for BRR is simple, it's equivalent to the upper 16 bits of BSRR, but you avoid the need of the shift
2012-10-08T20:47:15 < Thorn> we need another table in wiki to document ST peripheral struggles (usb, gpio, dma etc.)
2012-10-08T20:48:06 < zyp> struggles?
2012-10-08T20:48:32 < zyp> if you don't need BRR, don't use it, no problem there?
2012-10-08T20:49:07 < Thorn> I mean differences between families
2012-10-08T20:49:58 < zyp> that's mostly F1 vs. the rest, and that's covered fairly good in the F1 to F2 migration document
2012-10-08T20:56:18 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-08T21:03:19 < TitanMKD> does anyone has noted the TI.com search is totally since more than 2weeks ?
2012-10-08T21:03:45 < TitanMKD> I tested it on 3 major browser Firefox/GoogleChrome/IE8 and no result for all search
2012-10-08T21:05:08 -!- dekar [~dekar@212.255.19.185] has joined ##stm32
2012-10-08T21:05:11 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-08T21:05:26 < Thorn> search is serverside, browsers shouldn't make a difference unless there's some weird ass javascript or something
2012-10-08T21:06:00 < TitanMKD> yes but server side is down since 2 weeks !!
2012-10-08T21:06:15 < Thorn> OTOH, web development and weird ass go hand in hand in my experience
2012-10-08T21:06:33 < TitanMKD> it is just crazy for TI to have such failure
2012-10-08T21:06:44 < TitanMKD> but i sent them a bug report ;)
2012-10-08T21:07:02 < TitanMKD> they replied me it is fixed but it is still the same no result at all
2012-10-08T21:08:28 < Thorn> I remember an outright error page (autogenerated 403 or something) when trying to search a few days ago
2012-10-08T21:08:58 < Thorn> atm it just doesn't seem to find anything
2012-10-08T21:09:07 < TitanMKD> hehe ok like me ;)
2012-10-08T21:09:28 < TitanMKD> I even cleared all cookies to check but now it is confirmed it is a problem on their server
2012-10-08T21:09:39 < Thorn> well, I enter msp430 and I get 3 "try these first" links. I enter dp83848 and get nothing at all
2012-10-08T21:09:52 < TitanMKD> hehe ;)
2012-10-08T21:10:19 < Thorn> that is a bug
2012-10-08T21:10:45 < TitanMKD> just very funny that no one reported it as it is that mess since 2 weeks or even more
2012-10-08T21:12:32 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-08T21:12:40 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-08T21:23:03 -!- Tectu is now known as tectu
2012-10-08T21:29:15 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Read error: Connection reset by peer]
2012-10-08T21:39:16 < tectu> can someone tell me why this dosen't work (gdb related)?    http://pastebin.com/KgaakfQh     OpenOCD server dosen't give any error output
2012-10-08T21:41:33 < TitanMKD> it seems an internal bug of gdb
2012-10-08T21:41:46 -!- Laurenceb_ [~Laurence@host86-136-116-190.range86-136.btcentralplus.com] has joined ##stm32
2012-10-08T21:41:49 < TitanMKD> A problem internal to GDB has been detected,
2012-10-08T21:41:49 < TitanMKD> further debugging may prove unreliable.
2012-10-08T21:42:04 < TitanMKD> home/build/work/jenkins-daily-build/src/gdb/gdb/thread.c:81: internal-error: inferior_thread: Assertion `tp' failed.
2012-10-08T21:42:30 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-08T21:42:50 < tectu> TitanMKD, should I report that somewhere?
2012-10-08T21:43:05 < TitanMKD> anyway you are not using an official build
2012-10-08T21:43:18 < TitanMKD> maybe try official build and not a daily build
2012-10-08T21:44:28 < TitanMKD> tectu also check latest openocd there's an important bug fix with openocd 0.6.1
2012-10-08T21:44:45 < tectu> TitanMKD, I got that from here:   https://launchpad.net/gcc-arm-embedded
2012-10-08T21:44:54 < tectu> TitanMKD, I am on openocd 0.7.x
2012-10-08T21:45:05 < zyp> my guess is that openocd sends something gdb doesn't like, and gdb instead of handling it gracefully lets it cause an internal error
2012-10-08T21:45:35 < TitanMKD> woo you are using a fully beta chain ;)
2012-10-08T21:46:24 < TitanMKD> so i will say like zyp your openocd version is probably buggy
2012-10-08T21:47:16 < tectu> TitanMKD, I am not pro enought to use ultra stable 10 years old never did crash not even during war toolchain version ;-)
2012-10-08T21:48:06 < TitanMKD> tectu just test the openocd 0.6.1
2012-10-08T21:48:41 < tectu> TitanMKD, someone just made ChibiOS RTOS integration to OpenOCD to debug threads properly and that's not in upstream yet ;-)
2012-10-08T21:48:59 < TitanMKD> ha ok it is for that feature ;)
2012-10-08T21:50:05 < tectu> TitanMKD, yep :)
2012-10-08T21:50:15 < tectu> TitanMKD, anyways, thanks for your help, I asked in #openOCD :)
2012-10-08T21:50:58 < TitanMKD> tectu if you find a fix i'm interested in it
2012-10-08T21:51:14 < TitanMKD> tectu as Chibios integration in OpenOCD is what i will love to see
2012-10-08T21:52:44 < tectu> TitanMKD, the user 'mabl' made it, it's just there for a couple of days but works very very very nice:    https://github.com/mabl/openocd_chibios
2012-10-08T21:53:56 < tectu> TitanMKD, just made a GDB copy'n'paste for you:    ny
2012-10-08T21:54:00 < tectu> http://pastebin.com/7FwfHd5n
2012-10-08T21:54:29 < tectu> first, showing all threads, then doing up to 8 backtraces for all threads ;)
2012-10-08T21:54:53 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-10-08T21:55:07 < TitanMKD> ha yes very nice
2012-10-08T21:55:20 < tectu> TitanMKD, indeed :)
2012-10-08T21:55:24 < tectu> TitanMKD, mabl did a very nice job :)
2012-10-08T21:56:16 < TitanMKD> and your problem is because you have done a merge between mabl version and openocd 0.6.1 trunk ?
2012-10-08T21:56:30 < tectu> TitanMKD, jep
2012-10-08T21:56:40 < tectu> TitanMKD, I guess that could be done easily
2012-10-08T21:56:42 < TitanMKD> ha ok ;)
2012-10-08T21:56:44 < tectu> TitanMKD, ist just a few files
2012-10-08T21:57:44 < TitanMKD> have you tested openocd 0.6.1 without those file to know if the famous GDB thread error happen too ?
2012-10-08T21:58:11 < tectu> TitanMKD, nope, I don't need it that hardly :)
2012-10-08T21:58:37 < Laurenceb_> http://hackaday.com/2012/10/08/scary-putin-guards-your-stash/
2012-10-08T21:58:42 < Laurenceb_> thats... pretty scary
2012-10-08T21:59:06 <+dekar> I just had webalizer parse my logs and so far I got around 800 visits from hackaday
2012-10-08T21:59:16 <+dekar> I don't see how that is able to crash anyones server
2012-10-08T21:59:35 < Laurenceb_> wait you have a project on there?
2012-10-08T21:59:49 <+dekar> yeah, but that's pretty depressing
2012-10-08T21:59:52 < zyp> you're probably not cool enough
2012-10-08T21:59:55 <+dekar> my forum has more daily traffic
2012-10-08T22:00:18 <+dekar> Laurenceb_, http://hackaday.com/2012/10/08/os-x-port-of-gqrx-is-the-easiest-way-to-get-into-software-defined-radio/
2012-10-08T22:00:52 < TitanMKD> hehe cool that SDR stuff
2012-10-08T22:01:08 < TitanMKD> dekar maybe you will be interested by hackRF hw
2012-10-08T22:01:28 < TitanMKD> dekar mossmann has done a very impressive hw
2012-10-08T22:01:33 < Laurenceb_> oh
2012-10-08T22:01:43 < Laurenceb_> thats kind of obvious
2012-10-08T22:01:46 < Thorn> certain forums frequently have posts like 'website xxx.com has security holes, check it out'. 800 visits from such a forum are certain to cause some trouble with your server lol
2012-10-08T22:01:57 < Laurenceb_> i had it running in 5 minutes 2 months ago
2012-10-08T22:02:20 < Laurenceb_>  :P
2012-10-08T22:02:31 < Laurenceb_> you need a better/proper hack
2012-10-08T22:02:36 < zyp> bullshit, you had it running in 2 months, 5 minutes ago
2012-10-08T22:03:15 < Laurenceb_> lol
2012-10-08T22:04:06 < Laurenceb_> i mean i opened in qtcreator and hit build
2012-10-08T22:04:16 < Laurenceb_> and it was running within 5 minutes
2012-10-08T22:04:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T22:04:29 < TitanMKD> dekar it was the same problem for me with webanalyse
2012-10-08T22:04:45 < TitanMKD> dekar free.fr has just removed the database
2012-10-08T22:05:08 < TitanMKD> and tell me to do not use any traffic logger/analyzer
2012-10-08T22:06:43 <+dekar> free.fr? what? O.o
2012-10-08T22:07:13 <+dekar> I have my own server and webalizer runs fine
2012-10-08T22:07:14 < TitanMKD> it's a web site provider ;)
2012-10-08T22:07:32 < TitanMKD> they support php and mysql database at start
2012-10-08T22:07:54 <+dekar> Laurenceb_, yet it had weird sound bugs and depended on a ton of other stuff most people fail to build
2012-10-08T22:08:05 < Laurenceb_> oh
2012-10-08T22:08:11 <+dekar> my version works, also switching between demodulations works
2012-10-08T22:08:15 < Laurenceb_> worked first time for me
2012-10-08T22:08:18 < Laurenceb_> on ubuntu
2012-10-08T22:08:24 <+dekar> ubuntu yeah
2012-10-08T22:08:38 <+dekar> mine is for OSX
2012-10-08T22:09:02 <+dekar> I had to replace the pulseaudio backend since there is none for OSX
2012-10-08T22:09:30 < BrainDamage> dekar: do you plan to merge upstream, or are the changes too radical?
2012-10-08T22:09:42 <+dekar> they're going upstream
2012-10-08T22:11:33 <+dekar> I surely didn't do any magic, just some programming
2012-10-08T22:11:55 <+dekar> but I don't think there was a properly working version for OSX before
2012-10-08T22:12:11 < BrainDamage> changing completely audio backend is something that can meet contrast with some devs
2012-10-08T22:14:59 <+dekar> BrainDamage, I had some drop-in replacement, I switched the gnuradio-pulseaudio backend for the gr-audio (which is the normal one that uses alsa on linux and supports osx)
2012-10-08T22:15:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T22:16:03 <+dekar> that resulted in some other weird soundbugs that I fixed with some hack
2012-10-08T22:16:21 <+dekar> but overall, not too special - yet it works, and it didn't before :)
2012-10-08T22:21:14 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-08T22:23:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T22:23:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Client Quit]
2012-10-08T22:25:37 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-08T22:26:20 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-08T22:39:37 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-10-08T22:49:59 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-08T23:03:24 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-08T23:08:47 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 240 seconds]
2012-10-08T23:08:54 < zyp> there, got it working: http://bin.jvnv.net/f/LktBj.png
2012-10-08T23:08:59 < Laurenceb_> oh noes
2012-10-08T23:09:05 < Laurenceb_> sikipedia has shut down
2012-10-08T23:09:47 < Laurenceb_> "Set Line Coding"?!
2012-10-08T23:09:58 < Laurenceb_> are you making a stm32 usb - serial?
2012-10-08T23:10:11 < zyp> nah, it's just my CDC ACM echo demo
2012-10-08T23:10:22 < zyp> modified to use OTG_HS with external phy in HS mode
2012-10-08T23:10:38 < Laurenceb_> hmm
2012-10-08T23:10:50 < Laurenceb_> im sure the chibios version doesnt have line coding
2012-10-08T23:10:57 < Laurenceb_> i though you could strip it out
2012-10-08T23:11:41 < Laurenceb_> guess its not much more traffic to send :P
2012-10-08T23:11:48 < zyp> yeah, availability is specified by a flag in one of the descriptors
2012-10-08T23:12:14 < zyp> I think it was some talk about some host stacks not being happy if it's omitted, I haven't tried without
2012-10-08T23:12:40 < Laurenceb_> ah
2012-10-08T23:12:51 < Laurenceb_> yeah - i think this is supported in chibios
2012-10-08T23:13:14 < Laurenceb_> i can swap it and it mounts as /dev/ttyUSB0 or /dev/ttyACM0 iirc
2012-10-08T23:13:38 < Laurenceb_> i think i read ACM is commonly used on phones
2012-10-08T23:13:44 < Laurenceb_> for modem
2012-10-08T23:13:51 < Laurenceb_>  - AT command stuff
2012-10-08T23:14:20 < Thorn> zyp: nice, was it that chinese board? what phy does it have?
2012-10-08T23:14:38 < Laurenceb_> if you could use OTG host on F4 to support ACM you could have phone modem support
2012-10-08T23:15:09 < zyp> Thorn, usb3300
2012-10-08T23:15:20 < zyp> hmm
2012-10-08T23:15:43 < zyp> works nicely on OS X without the capability bit set
2012-10-08T23:18:01 < zyp> it's apparently windows having issues with it: https://github.com/gsmcmullin/blackmagic/commit/9663274
2012-10-08T23:18:48 < Laurenceb_> doesnt windows need a .init file for this stuff?
2012-10-08T23:19:09 < zyp> no idea, I'm not testing on windows
2012-10-08T23:19:14 < Laurenceb_> hehe
2012-10-08T23:19:21 < zyp> not yet at least
2012-10-08T23:19:25 < Laurenceb_>  "winbdws?! who runs that"
2012-10-08T23:20:17 < zyp> I've considered setting up some test boxes with different OSes for testing
2012-10-08T23:23:54 < zyp> oh, and apparently on HS bulk endpoints are only allowed to have a max packet size of 512
2012-10-08T23:24:07 < zyp> so the host stack complains about mine being 64
2012-10-08T23:24:17 < zyp> USBF:	247682.559	Endpoint 0x81 of the USB device "CompositeDevice" at location 0x1d152000:  converting Bulk MPS from 64 to 512 (USB 2.0 Spec section 5.8.3)
2012-10-08T23:24:20 < zyp> USBF:	247682.559	Endpoint 0x1 of the USB device "CompositeDevice" at location 0x1d152000:  converting Bulk MPS from 64 to 512 (USB 2.0 Spec section 5.8.3)
2012-10-08T23:25:24 < Thorn> my notes say *up to* 512 bytes for HS
2012-10-08T23:25:27 < zyp> section 5.8.3 reads: «The USB defines the allowable maximum bulk data payload sizes to be only 8, 16, 32, or 64 bytes for full-speed endpoints and 512 bytes for high-speed endpoints.»
2012-10-08T23:26:33 < Laurenceb_> on windows?
2012-10-08T23:26:59 < zyp> no, section 5.8.3 of the actual usb 2.0 spec
2012-10-08T23:27:13 < Laurenceb_> oh
2012-10-08T23:27:15 < zyp> «All Host Controllers are required to have support for 8-, 16-, 32-, and 64-byte maximum packet sizes for full-speed bulk endpoints and 512 bytes for high-speed bulk endpoints. No Host Controller is required to support larger or smaller maximum packet sizes.»
2012-10-08T23:27:37 < Laurenceb_> yeah
2012-10-08T23:27:48 < zyp> I didn't know it were that strict
2012-10-08T23:27:54 < Laurenceb_> your " shows as « to me
2012-10-08T23:28:04 < Laurenceb_> character encoding...
2012-10-08T23:28:11 < zyp> that's because I'm writing « and not "
2012-10-08T23:28:25 < tectu> zyp, that again? :D
2012-10-08T23:28:32 < zyp> :p
2012-10-08T23:28:34 < Laurenceb_> heh
2012-10-08T23:28:45 < zyp> I like «», they are nesting nicely unlike ""
2012-10-08T23:29:22 < zyp> and when enclosing code in it, it makes it pretty clear what is code and what is not
2012-10-08T23:29:26 < Laurenceb_> fair enough
2012-10-08T23:29:27 < Thorn> okay, fixed my notes
2012-10-08T23:29:30 < Laurenceb_> perfectionist
2012-10-08T23:29:47 < tectu> zyp, i should start using « » as well
2012-10-08T23:29:49 < Laurenceb_> im just happy when my PID controller works
2012-10-08T23:30:46 < Thorn> how did it "convert" MPS btw?
2012-10-08T23:30:51 < tectu> I only used to build PID controller with bare metal analog electronics
2012-10-08T23:31:17 < zyp> Thorn, no idea, I assume it's assuming 512 instead of 64
2012-10-08T23:31:43 < zyp> that would probably screw up the device if it tries to write more than 64 bytes
2012-10-08T23:31:44 < Thorn> your device probably isn't going to like that very much
2012-10-08T23:31:53 < Thorn> yeah
2012-10-08T23:32:07 < Thorn> IIRC your trace showed 61 byte transfers
2012-10-08T23:32:11 < Laurenceb_> i can never make PID behave well
2012-10-08T23:32:24 < zyp> Thorn, not on bulk
2012-10-08T23:32:24 < Laurenceb_> its always oscillating
2012-10-08T23:32:40 < tectu> wtf is this for a board?
2012-10-08T23:32:52 < tectu> http://mcuoneclipse.files.wordpress.com/2012/07/front-of-tower-system.png?w=571&h=380
2012-10-08T23:32:57 < tectu> is that some KEIL board?
2012-10-08T23:33:12 <+dekar> my mirror of the old st site has insane traffic
2012-10-08T23:33:30 < tectu> dekar, link????????? *__*
2012-10-08T23:33:38 <+dekar> 864 downloads of the virtual com port driver for windows this month so far
2012-10-08T23:34:04 <+dekar> tectu, http://dekar.wc3edit.net/st%20website/STM32%20(ARM%20Cortex%20M3)%20-%2032-bit%20Microcontrollers%20-%20Technical%20Literature%20and%20Support%20Files.html
2012-10-08T23:34:19 <+dekar> it has been linked to from the official ST forum
2012-10-08T23:34:27 <+dekar> I wonder how people know about this?
2012-10-08T23:34:30 <+dekar> XD
2012-10-08T23:34:36 < BrainDamage> ahhh back when st's site made sense
2012-10-08T23:34:44 < tectu> lol
2012-10-08T23:35:08 < Laurenceb_> heh
2012-10-08T23:35:46 <+dekar> https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=%2Fpublic%2FSTe2ecommunities%2Fmcu%2FLists%2Fcortex_mx_stm32%2FVirtual%20COM%20port%20in%20Win7%20or%20Vista%2032-bit&currentviews=5561
2012-10-08T23:35:52 <+dekar> who is this clive1 guy?
2012-10-08T23:36:45 <+dekar> it's funny how the official link they have in this thread is 404
2012-10-08T23:36:51 < zyp> heh, writing 63 bytes worked nicely, writing 64 bytes sort of worked, writing 65 bytes broke everything :p
2012-10-08T23:36:54 <+dekar> but my mirror still works
2012-10-08T23:37:53 <+dekar> is clive1 affiliated with ST?
2012-10-08T23:38:31 < Laurenceb_> http://www.aliexpress.com/item/2012-New-Hot-sale-Mele-A1000-TV-Box-Android-4-0-hd-media-player-1080p-allwinner/603395842.html
2012-10-08T23:38:31 <+dekar> I should charge him for using my private mirror, I'd like an USB PID from ST in exchange
2012-10-08T23:38:39 < Laurenceb_> dekar: no
2012-10-08T23:38:44 < Laurenceb_> and that thread is funny
2012-10-08T23:39:25 < Laurenceb_> oh i thought you linked the rage thread
2012-10-08T23:39:35 < Laurenceb_> where everyone is moaning about ST support
2012-10-08T23:40:11 < zyp> ah
2012-10-08T23:40:47 < Laurenceb_> STM32-One or whatever comes in and writes a ton of bullshit
2012-10-08T23:41:16 < zyp> device controller just drops packets > 64 bytes, resulting in no ACK for the host, which makes host stuck on retrying
2012-10-08T23:41:23 < Laurenceb_> oops
2012-10-08T23:41:58 -!- tectu is now known as Tectu
2012-10-08T23:42:14 <+dekar> Laurenceb_, I'd stay away from ARM/Linux till they get EFI
2012-10-08T23:42:37 < Laurenceb_> mele a1000/2000 can boot ubuntu aiui
2012-10-08T23:42:44 < Laurenceb_> pwns rpi
2012-10-08T23:44:19 < zyp> not sure how sane assuming 512 bytes is, but if host controller doesn't support anything else (because it doesn't have to), I guess it's the only way of getting a sort-of working device
2012-10-08T23:44:24 <+dekar> Laurenceb_, I'd rather take an rpi if I had the choice
2012-10-08T23:45:19 < Laurenceb_> get both XD
2012-10-08T23:45:32 < Laurenceb_> oh you even get a case and power supply :P
2012-10-08T23:46:11 < Laurenceb_> dunno what the differenc ebetween 1000 and 2000
2012-10-08T23:46:21 <+dekar> Laurenceb_, and shitty linux support, no video acceleration…
2012-10-08T23:47:01 < Thorn> do ULPI PHYs support a listen only mode?
2012-10-08T23:47:13 <+dekar> they both don't have an EFI, yet the rpi probably has a proper flattened device tree blob
2012-10-08T23:47:21 <+dekar> which is the next best thing
2012-10-08T23:47:24 < zyp> Thorn, probably
2012-10-08T23:48:13 < zyp> if I wanted an ARM board for linux now I'd consider the odroid-x
2012-10-08T23:48:44 <+Steffann> No, you want the zedboard
2012-10-08T23:48:54 <+Steffann> *that
2012-10-08T23:49:38 < zyp> well, to be fair, the odroid-x has more cpu power, and I said ARM board, not FPGA board :p
2012-10-08T23:49:54 <+Steffann> You don't have to use the ftp
2012-10-08T23:49:56 <+Steffann> fpga lol
2012-10-08T23:49:58 -!- barthess [~barthess@178.154.51.107] has quit [Quit: Leaving.]
2012-10-08T23:50:34 < zyp> Steffann, no, and at that point it becomes a three times as expensive board with less than half the computing power
2012-10-08T23:51:10 <+Steffann> but indeed that odroid-x looked pretty impressive
2012-10-08T23:51:12 < Bird|lappy> thing with a FPGA though is it's also the ultimate peripheral device ;)
2012-10-08T23:51:31 <+dekar> odroid x also has no video acceleration
2012-10-08T23:51:35 < zyp> Steffann, yeah, it's like 16 times the cpu power compared to rpi for less than four times the price
2012-10-08T23:51:44 <+dekar> I'd really take the rpi over those things
2012-10-08T23:52:11 -!- meloun [~meloun@188-195-251-164-dynip.superkabel.de] has left ##stm32 []
2012-10-08T23:52:15 < zyp> dekar, [citation needed]
2012-10-08T23:52:36 <+dekar> http://www.cnx-software.com/2012/08/08/ubuntu-12-04-on-odroid-x-development-board/
2012-10-08T23:54:16 <+dekar> if they still rely on the old boardfiles then the support will probably never go upstream
2012-10-08T23:54:42 <+dekar> in case of FDT blobs they might get it working with an upstream kernel, but I doubt it
2012-10-08T23:54:57 <+dekar> all the arm boards I have seen so far had hundreds of kernel patches
2012-10-08T23:55:07 < zyp> I don't see any statements about video decoding in your link?
2012-10-08T23:55:26 <+dekar> well it says no 3d
2012-10-08T23:55:27 < Laurenceb_> https://gitorious.org/lima
2012-10-08T23:56:06 <+dekar> Laurenceb_, yeah, show me that on working hardware
2012-10-08T23:56:09 < Laurenceb_> eww ubuntu 12 is horrible
2012-10-08T23:56:17  * Laurenceb_ runs 10.04LTS
2012-10-08T23:56:19 < zyp> dekar, I see a comment there though: «AFAIK, ODroid-X only support H.264 …»
2012-10-08T23:56:24 < zyp> so I don't know
2012-10-08T23:56:27 < Laurenceb_> last version before they went insane
2012-10-08T23:56:41 <+dekar> zyp, it probably supports that on _Android_ but surely not on GNU/Linux
2012-10-08T23:56:53 <+dekar> we have a ton of arm SoC boards at work and I hate every single one of them
2012-10-08T23:58:02 < Laurenceb_> heh
2012-10-08T23:58:19 <+dekar> it's just a mess, I spend hours backporting drivers to outdated buggy kernels
2012-10-08T23:58:20 < Laurenceb_> http://limadriver.org/
2012-10-08T23:58:22 < zyp> Android is not that different from GNU/Linux
2012-10-08T23:58:24 <+dekar> *spent
2012-10-08T23:58:26 < Laurenceb_> ways it supports A1000
2012-10-08T23:58:54 < Bird|lappy> dekar, :P that seems to be the way of ARM SoCs
2012-10-08T23:59:01 <+dekar> zyp, it is - the kernel is the same and there is binary compatibility, but audio and video is completely different
2012-10-08T23:59:43 <+dekar> zyp, android drivers won't work on a non android GNU/Linux
--- Day changed Tue Oct 09 2012
2012-10-09T00:00:07 <+dekar> you could probably write some wrappers and whatnot, you don't want to :/
2012-10-09T00:00:26 < zyp> it can't really be that hard.
2012-10-09T00:00:53 <+dekar> zyp, buy one of those things and show me how awesome it is
2012-10-09T00:01:08 < Bird|lappy> IIRC: they are working on merging a bunch of the currently-Android-specific stuff back into the mainline Linux kernel
2012-10-09T00:01:21 < BrainDamage> they already did since kernel 3
2012-10-09T00:01:38 < zyp> dekar, I almost would out of spite :p
2012-10-09T00:01:39 < Bird|lappy> yeah, they did a bunch of big merges already
2012-10-09T00:01:47 < Bird|lappy> mostly power mgmt stuff, which makes sense
2012-10-09T00:01:48 <+dekar> zyp, do so
2012-10-09T00:02:12 < BrainDamage> you can chroot gnu/linux within android
2012-10-09T00:02:26 <+dekar> they even replaced fricking alsa on android, I really don't get it
2012-10-09T00:02:30 < BrainDamage> so you can start with a known framework and see if it can access
2012-10-09T00:02:49 <+dekar> BrainDamage, you can do that, and then you have no sound and no video
2012-10-09T00:03:16 < BrainDamage> I am doing it, but I have video
2012-10-09T00:03:22 < BrainDamage> ( altough cheating )
2012-10-09T00:03:30 < BrainDamage> virtual X server and vnc
2012-10-09T00:03:57 <+dekar> BrainDamage, you're trolling right?
2012-10-09T00:04:02 < BrainDamage> no
2012-10-09T00:04:09 <+dekar> there is an xserver for android
2012-10-09T00:04:12 <+dekar> just check the market
2012-10-09T00:04:15 <+dekar> no need for VNC
2012-10-09T00:04:37 <+dekar> and VNC can't fluently play video to begin with
2012-10-09T00:05:02 < BrainDamage> I am running the x server within the gnu environment, the external wouldn't be accessible without tons of linking
2012-10-09T00:05:24 <+dekar> I'd really like you guys to show me booting up a vanilla linux kernel on any of your weird chinese toys
2012-10-09T00:05:42 < BrainDamage> I never claimed that
2012-10-09T00:05:54 < BrainDamage> I only said I chrooted gnu linux in android
2012-10-09T00:05:56 <+dekar> BrainDamage, you can run the android xserver app and use it from within the chroot
2012-10-09T00:05:58 <+dekar> I have done it
2012-10-09T00:06:07 < BrainDamage> and got a 70% working desktop
2012-10-09T00:06:13 <+dekar> but that has no acceleration at all
2012-10-09T00:06:44 <+dekar> you don't have sound and your visuals lag, an rpi would beat that easily
2012-10-09T00:07:14 < BrainDamage> I didn't say either that it's a good solution atm, but there's room for improvements
2012-10-09T00:08:28 <+dekar> well I can boot windows 95 in dosbox on my phone
2012-10-09T00:08:31 <+dekar> so what
2012-10-09T00:08:53  * dekar is waiting for EFI
2012-10-09T00:09:13 < Bird|lappy> I wonder if you could make accelerated X work on the Android kernel
2012-10-09T00:09:23 < Bird|lappy> or if that's been done already :)
2012-10-09T00:10:18 <+dekar> and then comes android and kills it, cause android likes killing applications when it thinks it needs more ram
2012-10-09T00:10:57 < BrainDamage> I hate that stupid automated ram management system
2012-10-09T00:11:05 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-09T00:11:11 <+dekar> hey you agreed with me :P
2012-10-09T00:11:19 <+dekar> it was much better on android 1.6
2012-10-09T00:11:36 <+dekar> I was using swap on 1.6 and I could easily run open office in background on my phone
2012-10-09T00:11:39 < Laurenceb_> http://www.ghielectronics.com/catalog/product/351
2012-10-09T00:11:39 < Laurenceb_>  ".NET Micro Framework 4.2"
2012-10-09T00:11:39  * Laurenceb_ vomit
2012-10-09T00:12:10 <+dekar> these days it seems to ignore swap for some reason, it still kills your xserver and other stuff
2012-10-09T00:12:39 <+dekar> and killing the xserver results in killing your whole desktop
2012-10-09T00:13:39 < zyp> doesn't the memory management stuff only apply to stuff running under dalvik?
2012-10-09T00:13:44 <+dekar> Bird|lappy, so what once you got an xserver? write an alsa emulation layer to get sound?
2012-10-09T00:13:59 <+dekar> zyp, xserver is an android app
2012-10-09T00:14:34 < zyp> technically it doesn't have to be.
2012-10-09T00:14:46 <+dekar> zyp, how is it supposed to render then?
2012-10-09T00:15:04 <+dekar> hijack the framebuffer (ab)using root privileges?
2012-10-09T00:15:31 <+dekar> it's not worth hacking android, just get a proper linux and be done with it
2012-10-09T00:15:34 < zyp> sure, get rid of the entire dalvik stack, we're not interested in that are we?
2012-10-09T00:16:00 <+dekar> zyp, replace the android kernel - oh wait, you can't get a vanilla kernel to run on that crap :/
2012-10-09T00:16:05 < zyp> and really, what prevents you from compiling an android kernel with alsa?
2012-10-09T00:16:31 <+dekar> zyp, do so, it might work in some cases
2012-10-09T00:17:36 < BrainDamage> I'm curious what canonical is doing for this: http://www.ubuntu.com/devices/android
2012-10-09T00:18:01 < BrainDamage> and I mean it in a totally neutral way
2012-10-09T00:18:08 < BrainDamage> not facetious at all
2012-10-09T00:18:32 < BrainDamage> if they are writing say android interfaces for pulseaudio
2012-10-09T00:18:41 < BrainDamage> and gstreamer
2012-10-09T00:19:06 <+dekar> is there any proof of that stuff actually running?
2012-10-09T00:19:39 < zyp> doesn't the docking stuff for motorola atrix use something like that?
2012-10-09T00:19:45 < BrainDamage> no, until it's announced a release, it's to be considered vapourware
2012-10-09T00:19:55 < BrainDamage> since afaik no public progress infos
2012-10-09T00:23:39 <+dekar> BrainDamage, with enough manpower you could pull that off, they could switch the whole graphics and sound backend while running
2012-10-09T00:23:59 <+dekar> like for those nvidia optimus things that switch the GPU
2012-10-09T00:24:39 < BrainDamage> actually both PA and gstreamer are capable of "live" stream switching
2012-10-09T00:24:42 <+dekar> it's just that I personally got better things to do with my time then waste it on annoying ARM problems that probably will be gone soon
2012-10-09T00:24:45 < BrainDamage> provided there's a backend available
2012-10-09T00:24:59 <+dekar> BrainDamage, you could do that at kernel level if you wanted
2012-10-09T00:25:19 <+dekar> less overhead, no wrappers
2012-10-09T00:25:29 < BrainDamage> but requires a custom kernel
2012-10-09T00:25:42 <+dekar> it's ubuntu, they're leading ARM Linux
2012-10-09T00:26:10 <+dekar> they're the center of linaro
2012-10-09T00:26:23 < zyp> anyway, the reason I'm not getting an ARM board for my TV yet is that I'm waiting for something that would have enough grunt to do video decoding on the cpu, so I don't have to care about hardware decoder limitations
2012-10-09T00:26:23 < BrainDamage> I'm not sure if that's sarcasm or not
2012-10-09T00:26:42 <+dekar> BrainDamage, I honestly believe ubuntu is leading when it comes to ARM support
2012-10-09T00:27:21 <+dekar> they have a kernel that boots on like 7 different devboards already, which is revolutionary for ARM
2012-10-09T00:27:33 <+dekar> you used to compile custom kernels for everything
2012-10-09T00:28:14 <+dekar> so I don't say those flattened device tree blobs are bad, I just prefer to wait for EFI
2012-10-09T00:28:39 <+dekar> ubuntu is definitively making progress
2012-10-09T00:31:32 < Thorn> ARGH should read datasheet *before* laying out PCB
2012-10-09T00:32:20 < zyp> that's probably a good idea
2012-10-09T00:32:47 <+dekar> That's the best linux on arm has to offer so far: https://wiki.linaro.org/Boards/
2012-10-09T00:33:02 <+dekar> a single kernel that runs (somewhat) on some of these boards
2012-10-09T00:33:30 <+dekar> on x86 you can just boot your ubuntu CD on any PC - god bless the BIOS
2012-10-09T00:33:36 < Thorn> looks like most LPC pins have internal pullups
2012-10-09T00:33:59 <+dekar> also for that linaro kernel you still need the fdt blobs
2012-10-09T00:34:08 <+dekar> poor mans EFI
2012-10-09T00:36:57 < Bird|lappy> btw: when a boardhouse says say "7mil annular ring" is that 7mil radius or 7mil diameter?
2012-10-09T00:37:25 < Thorn> that sounds really small even for diameter
2012-10-09T00:37:38 < Thorn> *radius.
2012-10-09T00:37:41 < zyp> varies
2012-10-09T00:37:52 < Bird|lappy> zyp, it's the dorkbotpdx/oshpark order
2012-10-09T00:38:05 < zyp> radius outside drill radius
2012-10-09T00:38:43 < Bird|lappy> ok.
2012-10-09T00:41:09 < zyp> and it's not really small
2012-10-09T00:42:18 < zyp> oshpark drill specs doesn't allow for anything denser than 1mm pitch BGA, otherwise vias won't fit between the pads
2012-10-09T00:43:13 < Bird|lappy> what's a decent size for power/gnd vias in a low power circuit?
2012-10-09T00:43:37 < Bird|lappy> (i.e. impedance is the size limiting factor, not ampacity)
2012-10-09T00:44:38 < zyp> I don't think larger vias have any benefit at all, so I just went with the smallest
2012-10-09T00:44:44 < Bird|lappy> ok
2012-10-09T00:45:17 < zyp> and when it comes to current, multiple small ones are as good as one larger
2012-10-09T00:48:24 < Rickta59> does the linker with arm-none-eabi-gcc produce .axf files? if so what makes them different than .elf files
2012-10-09T00:48:45 < Rickta59> i've looked at the arm site and it seems like .axf files are a keil thing
2012-10-09T00:49:26 < zyp> http://forums.arm.com/index.php?/topic/12786-what-exactly-an-axf-file/
2012-10-09T00:49:43 < zyp> according to this, a file with the .afx ending can be in either ELF or AIF format
2012-10-09T00:49:57 < Rickta59> gcc isn't going to produce aif files is it?
2012-10-09T00:50:37 < Rickta59> btw i found that and the link that the comment comes from is dead
2012-10-09T00:50:44 < zyp> AIF seems to be some outdated format
2012-10-09T00:51:00 < zyp> http://en.wikipedia.org/wiki/ARM_Image_Format
2012-10-09T00:51:22 < zyp> so I'd guess that most .axf files you'll encounter are in fact ELF files
2012-10-09T00:51:35 < Rickta59> yeah they are ..
2012-10-09T00:51:38 < Rickta59> produced by gcc linker
2012-10-09T00:51:59 < Rickta59> i was just trying to figure out why someone would name a file .axf if it was just .elf
2012-10-09T00:52:10 < Rickta59> * these are coming from the makefile setup for stellarisware
2012-10-09T00:52:24 < zyp> probably legacy reasons
2012-10-09T00:52:49 < Thorn> Vdda: analog 3.3 V pad supply voltage: This should be nominally the same voltage as VDD(3V3) but should be isolated to minimize noise and error.
2012-10-09T00:52:49 < Thorn> This voltage is used to power the ADC and DAC. This pin should be tied to 3.3 V if the ADC and DAC are not used.
2012-10-09T00:52:56 < Rickta59> thanks .. that was my conclusion also .. just thought i'd bounce it off arm-none-eabi-gcc users
2012-10-09T00:53:01 < Thorn> doesn't it also power PLLs?
2012-10-09T00:53:39 < Thorn> it looks like if you aren't using  ADC/DAC you shouldn't concern yourself with analog power at all (lpc17xx)
2012-10-09T00:53:52 < Thorn> unlike stm32
2012-10-09T00:54:03 < zyp> huh?
2012-10-09T00:54:09 < zyp> how so?
2012-10-09T00:55:15 < zyp> I'm reading what you pasted as «give it clean 3.3V if you use ADC/DAC, otherwise just tie it to the normal 3.3V supply»
2012-10-09T00:55:29 < zyp> which is how it's on stm32 as well, or what?
2012-10-09T00:55:33 < Thorn> well with stm32 afair you need to filter Vdda even if not using analog components
2012-10-09T00:55:44 < zyp> I'm not doing that.
2012-10-09T00:56:00 < zyp> why should I?
2012-10-09T00:58:29 < upgrdman_> hold do you filter it? a small ceramic and a big tant/elect cap?
2012-10-09T01:00:39 < Thorn> they recommend 10nF+1uF on Vdda (fig. 13 in f103 datasheet)
2012-10-09T01:00:56 < Thorn> separate from 100nF+4.7uF for Vdd
2012-10-09T01:02:15 < upgrdman_> both vdd and vdds would lead to the same power supply right? just extra decoupling caps near vdda, correct?
2012-10-09T01:02:29 < upgrdman_> vdd and vdda
2012-10-09T01:02:52 < Thorn> yes they have to according to the datasheet
2012-10-09T01:02:59 < upgrdman_> k
2012-10-09T01:09:17 -!- DaKu is now known as daku
2012-10-09T01:09:40 < Bird|lappy> in kicad, how does one insert vias manually? (need to 'stitch' thermal-pickup ground planes together on two sides of the board)
2012-10-09T01:12:30 < Thorn> no idea. diptrace has commands to add static vias and specify nets for them too
2012-10-09T01:15:44 < zyp> so do eagle :p
2012-10-09T01:20:57 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-09T01:23:22 < Bird|lappy> apparently you hack it in using a dummy trace that gets "swallowed" by the plane
2012-10-09T01:24:01 < zyp> ew
2012-10-09T01:24:13 < Thorn> but if you don't specify the net and then update copper pours you will get vias that don't stitch anything. my previous board has many such vias (and the Chinese didn't complain lol)
2012-10-09T01:24:39 < zyp> of course they don't complain, they just make what you tell them to.
2012-10-09T01:26:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-09T01:26:52 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T01:27:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Client Quit]
2012-10-09T01:27:31 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T01:27:45 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-09T01:28:11 < Thorn> I should've just moved the mcu ~1cm to the right. but if I do it now lots of traces will turn into noodle soup
2012-10-09T01:29:43 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Client Quit]
2012-10-09T01:30:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T01:30:35 < zyp> then redo them?
2012-10-09T01:30:45 < Thorn> time to WFI
2012-10-09T01:30:55 < zyp> or move them along with the MCU
2012-10-09T01:35:32 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 248 seconds]
2012-10-09T01:43:31 <+dekar> use the autorouter :P
2012-10-09T01:50:34 -!- Laurenceb_ [~Laurence@host86-136-116-190.range86-136.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-09T01:57:24 -!- WorkerBuzz [~Buzz@134.20.83.19] has joined ##stm32
2012-10-09T03:11:33 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-09T03:19:24 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-09T03:53:21 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-09T04:50:17 < gnomad> Has anyone here used the CMSIS DSP libraries on ant of the ST chips?
2012-10-09T04:50:42 < gnomad> or know of a good DSP library for the STs?
2012-10-09T05:00:25 < emeb_mac> gnomad: I've used the DSP libs on an ST405
2012-10-09T05:01:42 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-09T05:01:43 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-09T05:01:43 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-09T05:10:33 < gnomad> which is the ST405?  I'm not familiar with that...
2012-10-09T05:13:39 < emeb_mac> that's me mis-typing.
2012-10-09T05:13:45 < emeb_mac> I meant stm32f405
2012-10-09T05:20:36 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 248 seconds]
2012-10-09T05:23:38 < dongs> where the FUCK ARE MY LED MATRIXES
2012-10-09T05:24:40 < emeb_mac> still not
2012-10-09T05:24:42 < emeb_mac> ?
2012-10-09T05:24:55 < emeb_mac> dude shipped 'em to me toot-suite.
2012-10-09T05:25:07 < emeb_mac> probably that whole islands thing.
2012-10-09T05:25:13 < dongs> lulz
2012-10-09T05:25:28 < dongs> lemme see waht 17track says
2012-10-09T05:25:37 < emeb_mac> all commerce between CN/JP is shut down.
2012-10-09T05:25:52 < dongs> i would believe that if I didnt receive liek 5 other things from cn/hk meanwhile
2012-10-09T05:26:21 < dongs> 2012-09-29 total export package interoperability seal sealing of the Nanjing International, Japan
2012-10-09T05:26:41 < emeb_mac> damn - lost in the mails
2012-10-09T05:27:25 < emeb_mac> how many did you order?
2012-10-09T05:27:32 < dongs> i think 10 or 8
2012-10-09T05:27:44 < dongs> i doubt the amount is hte problem..
2012-10-09T05:28:06 < emeb_mac> nah - the box mine came is is big enough to hold that many.
2012-10-09T05:28:24 < dongs> haha ebay send mail box doesnt support unicode
2012-10-09T05:28:36 < dongs> i pasted chink tracking info into it and it came back to me as ??? ???
2012-10-09T05:28:47 < emeb_mac> FAIL
2012-10-09T05:29:48 < dongs> also i need to redo the patterns for tsop24-EP
2012-10-09T05:30:16 < emeb_mac> footprints don't match up?
2012-10-09T05:30:26 < dongs> no, i mean make them
2012-10-09T05:30:40 < dongs> since i originally made ones for SOP24
2012-10-09T05:30:48 < emeb_mac> ah - fun.
2012-10-09T05:32:26 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-09T05:33:17 < emeb_mac> I did an EP footprint for this in gEDA/PCB : http://www.ti.com/product/tps70345&lpos=Middle_Container&lid=Alternative_Devices
2012-10-09T05:34:25 < dongs> probably similar
2012-10-09T05:34:30 < dongs> its not hard, i have other EP parts
2012-10-09T05:37:31 < dongs> does F3 have DFU USB bootloader in ROM?
2012-10-09T05:37:43 < emeb_mac> think so.
2012-10-09T05:38:00 < dongs> even with F1-compatible USB peripheral?
2012-10-09T05:38:04 < dongs> i better check this.
2012-10-09T05:38:52 < emeb_mac> The boot loader is located in system memory. It is used to reprogram the Flash memory by using USART1, USART2 or USB.
2012-10-09T05:38:56 < dongs> i wonder if AN2606 got updated for F3
2012-10-09T05:39:04 < dongs> where are you reading that from?
2012-10-09T05:39:32 < emeb_mac> section 2.8, pg 13/120 of the datasheet
2012-10-09T05:39:51 < emeb_mac> for the f37x
2012-10-09T05:40:19 < dongs> i see it. clever of them not to update the boot appnote...
2012-10-09T05:40:39 < emeb_mac> yeah - an2606 doesn't mention F3
2012-10-09T05:40:40 < dongs> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/APPLICATION_NOTE/CD00167594.pdf
2012-10-09T05:41:05 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-09T05:41:44 < dongs> they did update it in july t o add F0 shit..
2012-10-09T05:41:46 < dongs> hm
2012-10-09T05:41:49 < dongs> maybe they're still working on it.
2012-10-09T05:42:04 -!- BrainDamage [~BrainDama@62.10.5.141] has quit [Read error: Connection reset by peer]
2012-10-09T05:42:12 < emeb_mac> hope so. could be that it's the same as the f4 though
2012-10-09T05:44:37 < dongs> http://moeller.io/sixteenbit.html the freesoc guy
2012-10-09T05:45:50 < emeb_mac> oooohhkaaaaayyy
2012-10-09T05:46:08 < emeb_mac> no details.
2012-10-09T05:46:27 < emeb_mac> butt-ugly case design tho
2012-10-09T05:52:12 < Bird|lappy> emeb_mac, btw: that's a TSSOP not a TSOP :P
2012-10-09T05:52:26 < Bird|lappy> (a TSOP is a totally different package design)
2012-10-09T05:52:52 < emeb_mac> Bird|lappy: err - that TI regulator?
2012-10-09T05:52:56 < Bird|lappy> yeah
2012-10-09T05:53:09 < cjbaird> dongs: all I can see is a Hipster that I want to kill.
2012-10-09T05:53:19 < emeb_mac> well, terminological pedantry aside, my footprint worked.
2012-10-09T05:54:28 < Bird|lappy> only reason I was getting on your case is because confusing a TSOP and a TSSOP would leave you scratching your head when you went to solder the part down
2012-10-09T05:54:50 < gnomad> emeb_mac: so the stm43f405 and stm43f407 are very similar, no?
2012-10-09T05:54:59 < dongs> tyes
2012-10-09T05:55:00 < emeb_mac> It would - if I was dumb enough to release the board to fab without checking the footprints against real parts.
2012-10-09T05:55:02 < dongs> 407 adds ethernet i think
2012-10-09T05:55:05 < gnomad> ok
2012-10-09T05:55:18 < emeb_mac> gnomad: yep. Just the Ethernet MAC IIRC
2012-10-09T05:55:19 < gnomad> so what wan your experience with the CMSIS DSP libraries?
2012-10-09T05:55:22 < dongs> cjbaird: haha, i had similar thoughts
2012-10-09T05:55:29 < gnomad> what did you use to compile them?
2012-10-09T05:55:40 < dongs> they have binaries
2012-10-09T05:55:48 < dongs> for the only build system that matters (keil/mdk-arm)
2012-10-09T05:55:49 < emeb_mac> gnomad: I used the floating point real FFT/IFFTs - they worked fine. Seemed pretty fast.
2012-10-09T05:55:52 < gnomad> oh, you get binaries from your vendor?
2012-10-09T05:56:13 < dongs> f4 stdperiphlib has dsp lib, and it has binaries for gcc and keil
2012-10-09T05:56:24 < dongs> but I suspect you can just add them to build and it should work the same
2012-10-09T05:56:27 < emeb_mac> gnomad: err - no. ST distributes the source on their website.
2012-10-09T05:56:34 < gnomad> I'm trying to find out what the options are for using those libraries on the ST ARM chips....
2012-10-09T05:56:49 < emeb_mac> you just include them in your build & go.
2012-10-09T05:56:56 < gnomad> emeb_mac: did you use gcc or some other compiler?
2012-10-09T05:57:09 < emeb_mac> gnomad: GCC all the way. worked fine.
2012-10-09T05:57:16 < gnomad> ok
2012-10-09T05:57:43 < gnomad> I'm just getting started with ARM devel, in case you can't tell...
2012-10-09T05:58:02 < emeb_mac> Bird|lappy: but good point on the TSSOP vs TSOP issue. It's a problem for newbs.
2012-10-09T05:58:18 < emeb_mac> gnomad: It's a long learning curve.
2012-10-09T05:58:29 < gnomad> I've been using the msp430 almost exclusively for the past 18 months.
2012-10-09T05:58:34 < dongs> since I always make my own patterns i don't have this problem as much but still I usually print out the layout and fit stuff if im concerned
2012-10-09T05:58:58 < emeb_mac> yep
2012-10-09T05:59:01 < gnomad> but I have been lured into ARM by all the cheap devkits and chips...
2012-10-09T05:59:04 < cjbaird> gnomad: was much the same when I first showed up here..
2012-10-09T05:59:21 < gnomad> still not sure which ARM I want to work with....
2012-10-09T05:59:24 < emeb_mac> there are enough minor variations on footprints that it's a good practice to check vs a 1:1 plot
2012-10-09T05:59:33 < gnomad> LPC, ST, or stellaris
2012-10-09T05:59:55 < emeb_mac> gnomad: if you hang out here you use ST. Or else! :)
2012-10-09T06:00:01 < gnomad> leaning towards the ST now, seems to have the biggest community
2012-10-09T06:00:19 < emeb_mac> we are legion.
2012-10-09T06:00:45 < emeb_mac> and the parts + devboards are _really_ cheap.
2012-10-09T06:00:55 < gnomad> I picked up the Discovery F1 and F4 boards when they were free....
2012-10-09T06:01:06 < gnomad> but they have just been gathering dust until the right time.
2012-10-09T06:01:19 < emeb_mac> Break out the F4 and get to work. F1 - meh.
2012-10-09T06:01:35 < gnomad> other thing is... I don't like GUI toolchains
2012-10-09T06:01:52 < emeb_mac> gnomad: I don't either. I use make + gcc + gdb.
2012-10-09T06:01:52 < gnomad> I am a big emacs+gcc+make+gdb person.....
2012-10-09T06:01:57 < gnomad> sike!
2012-10-09T06:02:01 < cjbaird> Also FOSS-friendly, too. I've got LPC doorstops here that have that Code Red programmer shit in them.
2012-10-09T06:02:12 < emeb_mac> but emacs bites. :P
2012-10-09T06:02:18 < gnomad> i've been playing with the LPC1114x....
2012-10-09T06:02:29 < gnomad> it programs easily from the serial bootloader.
2012-10-09T06:03:23 < gnomad> but definitely weakest on community and documentation.
2012-10-09T06:03:43 < Bird|lappy> and I'm off in the land of the EFM32TG222 heh
2012-10-09T06:03:51 < gnomad> ok.
2012-10-09T06:04:06 < gnomad> I've seen a bunch of getting started pages for the Discovery F4.
2012-10-09T06:04:18 < gnomad> I think that's the one I'll really dive in with.
2012-10-09T06:04:25 < emeb_mac> F4 rocks.
2012-10-09T06:04:41 < dongs> cjbaird: what code red shit.
2012-10-09T06:05:15 < gnomad> the lpcxpresso boards are a proprietary protocol.
2012-10-09T06:05:23 < gnomad> they *only* speak to code-red.
2012-10-09T06:05:26 < dongs> ? they dont have jtag/swd?
2012-10-09T06:05:34 < cjbaird> d: The LPCXpresso boards. Proprietry closed-spec jtag interface.
2012-10-09T06:05:47 < gnomad> the xpresso is the jtag (serial sbw) interface
2012-10-09T06:06:30 < cjbaird> You can get to the programmer parts on the other side of the interface, but it's needless hacking
2012-10-09T06:06:43 < emeb_mac> blech.
2012-10-09T06:06:46 < gnomad> There only seems to be one SBW debug option for the LPCs -- Black Voodoo board, I believe is the name
2012-10-09T06:07:05 < emeb_mac> *plonk*
2012-10-09T06:08:04 < gnomad> anyways, I dig the concept of a an ARM in DIP for $1.26, but imma hold off for a bit....
2012-10-09T06:08:38  * emeb_mac couldn't care less about DIP packages these days.
2012-10-09T06:09:29 < dongs> ok so lemme get this right, LPC doesn't have standard CMSIS-DP or whatever shit on it, but some proprietary NXP debug port ONLY?
2012-10-09T06:09:58 < dongs> they didnt wanna pay arm for the license to use SWD or something?
2012-10-09T06:11:27 < gnomad> not exactly...
2012-10-09T06:11:38 < gnomad> the chips use the standard ARM SWD.
2012-10-09T06:11:59 < gnomad> the xpresso dev board speaks that standard SWD...
2012-10-09T06:12:11 < dongs> o, so theres no problem?
2012-10-09T06:12:14 < gnomad> but the host-api spec is closed.
2012-10-09T06:12:18 -!- esden [esden@repl.esden.net] has quit [Read error: Operation timed out]
2012-10-09T06:12:22 < emeb_mac> so they've got some on-board USB->SWD that uses a weird protocol
2012-10-09T06:12:23 < dongs> doesnt matter.
2012-10-09T06:12:26 < gnomad> so you can *only* use the xpresso boards with code-red
2012-10-09T06:12:29 < cjbaird> With the xpressos you need to get at the traces; they haven't broken-out the interface like nice people would.
2012-10-09T06:12:36 < emeb_mac> bastards
2012-10-09T06:12:38 < dongs> cjbaird: ok, yeah thats fail.
2012-10-09T06:12:44 < gnomad> yes
2012-10-09T06:13:03 < dongs> oh well as long as chips haveSWD im not worried
2012-10-09T06:13:08 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-10-09T06:13:13 < gnomad> and as far as I know, there is only one tool that does SWD that is supported by OpenOCD.
2012-10-09T06:13:31 < gnomad> and I think it only does it with a patchset.
2012-10-09T06:13:35 < emeb_mac> err - whut?
2012-10-09T06:13:52 < gnomad> that is what I am led to believe...
2012-10-09T06:14:14 < emeb_mac> latest OpenOCD supports STLINK V1/V2
2012-10-09T06:14:38 < gnomad> Yes, that is a big reason why I am learning towards working with the F4 board right now.
2012-10-09T06:14:58 < gnomad> so I've got my gcc support
2012-10-09T06:15:04 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Read error: Connection reset by peer]
2012-10-09T06:15:04 < gnomad> i've got my openocd support....
2012-10-09T06:15:14 < gnomad> CMSIS dsp libraries compile and work...
2012-10-09T06:15:20 < gnomad> with gcc...
2012-10-09T06:15:30 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-09T06:15:43 < Bird|lappy> emeb_mac, could you fill me in on just what is supported for SWD H/W with OpenOCD?
2012-10-09T06:16:16 < dongs> < emeb_mac> latest OpenOCD supports STLINK V1/V2
2012-10-09T06:16:20 < dongs> stlink is kinda fail though :)
2012-10-09T06:16:24 < emeb_mac> Bird|lappy: I'm just going from what's in the README for OpenOCD git
2012-10-09T06:16:48 < emeb_mac> I've built it with stlink support and that worked. Haven't tried it yet though.
2012-10-09T06:17:44 < emeb_mac> but one of the OpenOCD devs was on this chl a few days ago & claimed very speedy downloads.
2012-10-09T06:19:28 < cjbaird> Going to drop by the local neighbourhood Farnell warehouse and use sexual extortion to find out where the hell my Stellaris and KL25 boards are..
2012-10-09T06:20:38 < emeb_mac> good luck with that.
2012-10-09T06:21:50 < dongs> I ordered that 4.99 stellaris shite
2012-10-09T06:21:59 < dongs> havent heard from them about it yet either
2012-10-09T06:22:41 -!- cjbaird is now known as cjbaird_away
2012-10-09T06:22:58 < emeb_mac> oh yeah - I ordered one of those too. I think zyp ordered them also.
2012-10-09T06:23:17 < emeb_mac> (where's zyp been lately?)
2012-10-09T06:23:20 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-09T06:23:30 < dongs> hes bloggin
2012-10-09T06:23:52 < emeb_mac> is that code?
2012-10-09T06:24:18 < dongs> he talked a lot in the last 48 hours
2012-10-09T06:24:36 < dongs> about 6 hours ago today.
2012-10-09T06:25:43 < emeb_mac> yep - see the logs. I was out of town for a while.
2012-10-09T06:25:44 < dongs> apparently fedex delivered my mouser order to next door neighbor..
2012-10-09T06:25:50 < emeb_mac> AHA!
2012-10-09T06:25:52 < dongs> tahts why I havent seen my F3 shit
2012-10-09T06:26:05 < emeb_mac> and your neighbor probably has your LEDs too.
2012-10-09T06:26:09 < dongs> haha
2012-10-09T06:26:17 < dongs> thats less likely
2012-10-09T06:26:40 < emeb_mac> FedEx used to mis-deliver to my neighbors so often that they stopped telling me.
2012-10-09T06:27:00 < emeb_mac> I'd just walk to their front door and see it sitting there on a bench.
2012-10-09T06:27:10 < dongs> haha
2012-10-09T06:27:36 < emeb_mac> I complained to FedEx and the problem seems to have stopped.
2012-10-09T06:31:55 < emeb_mac> dongs: does Mouser have a warehouse in JP, or do they ship to you from US?
2012-10-09T06:32:36 < dongs> all from us. digikey and mouser
2012-10-09T06:32:43 < dongs> noone has any fucking warehouse in JP except RS
2012-10-09T06:32:48 < dongs> and their selection is pitiful as fuck
2012-10-09T06:33:16 < dongs> every company i know orders prototype parts from digikey..
2012-10-09T06:33:27 < emeb_mac> that's so screwed up.
2012-10-09T06:33:53 < dongs> delivery is fast though.. i order friday evening its here monday morning
2012-10-09T06:33:57 < dongs> so its not bad,..
2012-10-09T06:34:05 < dongs> i usually just queue shit up during the week and do an order on friday
2012-10-09T06:34:07 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-09T06:34:13 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-09T06:34:27 < emeb_mac> cool. hopefully shipping isnt' too steep.
2012-10-09T06:34:31 < dongs> if i wasnt in teh middle of nowehre it would be even faster
2012-10-09T06:34:33 < dongs> shipping is free
2012-10-09T06:34:37 < dongs> so its fine
2012-10-09T06:34:40 < emeb_mac> yeow!
2012-10-09T06:35:39 < emeb_mac> your orders must be fairly large.
2012-10-09T06:35:58 < emeb_mac> what's it, free shipping for orders > $200 or something?
2012-10-09T06:36:02 < dongs> >80 gets free shipping
2012-10-09T06:36:05 < dongs> so not bad at all.
2012-10-09T06:38:51 < dongs> i could always order shit i dont need, then RMA it when it arrives..
2012-10-09T06:38:57 < dongs> mailing it back costs about $3
2012-10-09T06:40:07 < dongs> like I added those ST drivers to an order few weeks ago then found teh chinese shit..
2012-10-09T06:40:14 < dongs> back they went..
2012-10-09T06:42:09 < emeb_mac> I've never sent stuff back.
2012-10-09T06:42:40 < dongs> i have enough shit i'm using to keep shit I won't use around :)
2012-10-09T06:43:06 < emeb_mac> I hear that. bags and boxes of parts from old projects.
2012-10-09T06:43:40 < emeb_mac> always buy a few extra just in case I wreck something, so it piles up.
2012-10-09T06:55:39 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 260 seconds]
2012-10-09T07:14:53 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-09T07:15:15 -!- mrcan_ [~mrcan@78.173.72.192] has joined ##stm32
2012-10-09T07:15:15 -!- mrcan_ [~mrcan@78.173.72.192] has quit [Changing host]
2012-10-09T07:15:15 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-09T07:17:17 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-09T07:17:17 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-09T07:17:17 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-09T07:58:45 < dongs> dixels
2012-10-09T07:59:34 < zippe> PHRS
2012-10-09T08:00:19 < zippe> Trying to make the case for ordering a FORM1
2012-10-09T08:02:48 -!- cjbaird_away is now known as cjbaird
2012-10-09T08:02:58 < cjbaird> I got lucky! :D
2012-10-09T08:03:40 < cjbaird> After a bit of SNAFU at the E14 warehouse, I've got both orders (KL25 Freedom, TI Stell.Launchpad)
2012-10-09T08:04:10 < cjbaird> Also score some significantly marked-down 'shop soiled' compounent storage cabinates..
2012-10-09T08:06:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 240 seconds]
2012-10-09T08:06:59 < dongs> pics or it didnt happen
2012-10-09T08:11:42 < cjbaird> http://i.imgur.com/UETgx.jpg
2012-10-09T08:12:35 < dongs> wth
2012-10-09T08:12:39 < dongs> left is that 4.99 thing?
2012-10-09T08:17:37 < cjbaird> The Stellaris.
2012-10-09T08:18:23 < dongs>  Pre-orders received before 25-Sept are now being processed and shipped. Please allow a minimum of 2 to 8 weeks for delivery.
2012-10-09T08:18:29 < dongs> I ordered m ine way back
2012-10-09T08:18:48 < dongs> sept 2nd... :|
2012-10-09T08:22:04 < cjbaird> http://i.imgur.com/9N4LM.jpg
2012-10-09T08:22:20 < dongs> nice
2012-10-09T08:23:09 < dongs> intersting, individual chip fiducials
2012-10-09T08:23:30 < cjbaird> I won't have any time today to fire them up and play, though. :/
2012-10-09T08:27:22 < cjbaird> Hmm, the Launchpad comes with a micro-usb connector this time. (I've got about a dozen waste mini-usb cables from the other LPs I've bought.)
2012-10-09T08:29:56 < dongs> too bad its SMD
2012-10-09T08:37:51 < cjbaird> Comparing the three launchpads: http://i.imgur.com/k0ZJQ.jpg
2012-10-09T08:38:59 < dongs> lol dip
2012-10-09T08:59:24 -!- Tectu is now known as Tectu|fork
2012-10-09T09:00:00 < dongs> more like Tectu|kawaii
2012-10-09T09:09:40 < dongs> damn the chinese are awesome
2012-10-09T09:09:56 < dongs> the guys who make my logic analyzer, about a month ago i sent them some DMX512 capture samples
2012-10-09T09:10:00 < dongs> ebcause it was tripping up their uart decoder
2012-10-09T09:10:17 < dongs> few days ago they email me a complete dmx512 protocol analyzer dll
2012-10-09T09:10:46 < dongs> and today just send update that uses brightness to visually show value of different channels
2012-10-09T09:11:22 < dongs> opensores dudes never do this sorta shit, unless they need to scratch some itch, you aint getting a protocol decoder implemented
2012-10-09T09:11:47 < dongs> or unless you do it yourself lol lol
2012-10-09T09:16:35 < claude> yeah these lazy opensource coders , doing the stuff only for fun and/or reputation... always party but no 'customer' support :/
2012-10-09T09:20:58 -!- Tectu|fork is now known as Tectu
2012-10-09T09:21:15 -!- NonaSuomy is now known as TeknoJuce
2012-10-09T09:26:54 < dongs> Version 2.2.9.1 (October 5, 2012)
2012-10-09T09:26:55 < dongs> ooo
2012-10-09T09:26:56 < dongs> new dicktrace
2012-10-09T09:27:02 < dongs>  "Manual Route" mode hotkey (~).
2012-10-09T09:27:18 < dongs> Blocking file by the program while editing it. Overwriting by second/network appearance is prohibited.
2012-10-09T09:27:21 < dongs> fuck yeah
2012-10-09T09:27:22 < dongs> i bitched enough about that
2012-10-09T09:28:23 < cjbaird> Those guys have done this before for you, you've mentioned.
2012-10-09T09:28:44 < dongs> cjbaird: dmx? yeah, but today they send a new version with brightness to visualize data
2012-10-09T09:28:47 < dongs> thats new
2012-10-09T09:29:06 < dongs> and usually turn around time to questions/bugs is within same workday, which is great.
2012-10-09T09:29:38 < dongs> I didnt *ask* for a dmx decoder at all. but I gave them enough data and got them interested I guess to implement it.
2012-10-09T09:30:39 < dongs> updating dicktrace.
2012-10-09T09:31:52 < dongs> ah, great. ~ to start manual route is awesome.
2012-10-09T09:31:55 < dongs> now i dont need to clik anymore.
2012-10-09T09:33:35 < emeb_mac> dicktrace - sounds like an old comics character.
2012-10-09T09:34:53 < dongs> i let it autoroute the ledshit and now just cleaning it up a bit
2012-10-09T09:35:19 < dongs> is 12mil enough for 5V traces? I calculated shit for 500mA and it only needed like 10mil for 10C raise
2012-10-09T09:37:09 < emeb_mac> I'd give it more than that
2012-10-09T09:37:17 < dongs> rly? like wat
2012-10-09T09:38:59 < dongs> looks like others use 32mil
2012-10-09T09:39:02 < dongs> will do that.
2012-10-09T09:41:53 < cjbaird> Looks like you can stack the Stellaris boards onto each other, like the old newmicros sbcs (like: http://images.4chan.org/g/src/1349764892773.jpg)
2012-10-09T09:42:09 < emeb_mac> 32 is good. I usually use 25 mil for power routes.
2012-10-09T09:45:17 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-09T10:01:48 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-09T10:19:55 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-10-09T10:22:00 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-09T10:32:26 < dongs> > Uploaded by boazal on Jun 27, 2011
2012-10-09T10:32:27 < dongs> old
2012-10-09T10:32:53 < dongs> still old
2012-10-09T10:33:13 < dongs> http://www.youtube.com/watch?v=Zqmdv5iyIOY much better
2012-10-09T10:42:04 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-09T11:02:40 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-09T11:06:24 -!- claude is now known as claude_work
2012-10-09T11:10:42 < dongs> ok cleaned up driver a bit. should be able to submit it sometime this week.
2012-10-09T11:13:33 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-09T11:13:34 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-09T11:13:34 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-09T11:32:37 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 246 seconds]
2012-10-09T12:16:23 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-09T12:17:07 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-09T12:19:10 < karlp> wouldn't the xpresso interface be just like the stlink interface? that wasn't exactly open either originally.
2012-10-09T12:25:06 < dongs> karlp: you'd think so
2012-10-09T12:25:12 < dongs> karlp: but this is opensauce dudes youre talkin wiht
2012-10-09T12:25:17 < dongs> they want free or nothign
2012-10-09T12:29:44 < karlp> something like that yeah
2012-10-09T12:30:17 < karlp> also, flyback, I'm with dongs on this one, I've seen quantum levitation with magnets in person probably 10 years ago, it's not even remotely new, no matter how much classical music you play with it.
2012-10-09T12:31:09 < karlp> there was some other reason I never got the xpresso board when I was evalling an arm choice, can't remember anymore though now.
2012-10-09T12:35:33 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-09T12:35:34 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-09T12:36:59 < Thorn> dongs: while you're at that, please bitch about the lack of a hotkey editor, no support for standard shortcuts in input fields (like ctrl+a to select all), multilevel right click menus where they're the last thing that is needed, etc.
2012-10-09T12:37:24 < dongs> Thorn: they have a bugtracker you can complain on
2012-10-09T12:37:38 < dongs> yeah, lack of standard keys is annoying as fuck but its i think because they capture them globally
2012-10-09T12:37:45 < dongs> i.e. ctrl+a inside a textbox selects all shit in schematic
2012-10-09T12:38:00 < dongs> hotkey editor is on the todo list
2012-10-09T12:38:01 < dongs> for 2.2
2012-10-09T12:38:05 <+Steffann> bad excuse
2012-10-09T12:38:06 < dongs> er
2012-10-09T12:38:07 < dongs> 2.5 or so
2012-10-09T12:38:13 < dongs> Steffann: i agree.
2012-10-09T12:38:18 < Thorn> if they used a sane gui framework like qt that would work out of the box.
2012-10-09T12:38:27 < dongs> you dont wanna know what they're using
2012-10-09T12:38:33 < dongs> its like delphi6 or someshit.
2012-10-09T12:38:37 < dongs> something utterly nasty.
2012-10-09T12:38:52 < Thorn> and a hotkey editor is 200 lines of code if they used QAction
2012-10-09T12:38:54 < Thorn> etc.
2012-10-09T12:39:36 < Thorn> delphi has been very popular in russia, no idea why. never used it myself
2012-10-09T12:39:41 < dongs> yeah
2012-10-09T12:40:06 < karlp> I used it briefly, it was really nice at the itme, much nicer than vb
2012-10-09T12:40:17 < karlp> but ms subsidized and heavily promoted vb to western devs
2012-10-09T12:40:36 < karlp> that's my theory anyway
2012-10-09T12:40:51 < zyp> I remember dicking around with delphi as a kid
2012-10-09T12:41:16 < dongs> http://i.imgur.com/WkdiQ.png
2012-10-09T12:41:22 < dongs> semi-autorouted led driver of doom
2012-10-09T12:42:05 < zyp> whoa, going with direct usb now? that's hardcore
2012-10-09T12:42:25 < dongs> well, USB power is a jumper
2012-10-09T12:42:36 < dongs> it could be useful for 1st block in the series to get loaded from USB though
2012-10-09T12:42:43 < zyp> I mean without usb-ttl-adapter :p
2012-10-09T12:42:47 < dongs> haha..
2012-10-09T12:42:59 < Thorn> looking at BMP schematics, target reset is driven directly by a mcu pin. I have a reset button on my board that shorts it to ground. won't it create a short circuit in a debugger?
2012-10-09T12:43:31 < dongs> push it and find out
2012-10-09T12:43:33 < zyp> Thorn, no, pin would be configured in OD mode
2012-10-09T12:43:41 < zyp> i.e. only driving low, not high
2012-10-09T12:43:47 < dongs> what if its not?
2012-10-09T12:43:47 < Thorn> I kind of hope so
2012-10-09T12:43:55 < karlp> hey guess what? opensores!
2012-10-09T12:43:56 < Thorn> need to check BMP sources
2012-10-09T12:43:57 < karlp> you can check!
2012-10-09T12:44:00 < Thorn> yes
2012-10-09T12:44:22 < Thorn> but there are other debuggers, too/
2012-10-09T12:44:23 < dongs> hmm
2012-10-09T12:44:25 < dongs> no leds on this board
2012-10-09T12:44:34 < dongs> I guess the fact that there are 64 of htem on backside is OK.
2012-10-09T12:44:56 < zyp> :)
2012-10-09T12:46:15 < dongs> o well
2012-10-09T12:46:17 < dongs> prob good enough
2012-10-09T12:46:22 < dongs> too bad i was assburgering all week
2012-10-09T12:46:28 < dongs> missed like 3 orders where I could have paneled this shit
2012-10-09T12:47:04 < Thorn> say with FTDI based debuggers you have to rely on e.g. openocd to do it right. there's no open drain mode in ftdi iirc, you have to reconfigure it to input every time you need to deassert reset
2012-10-09T12:47:05 < Laurenceb> assburgerzzz
2012-10-09T12:47:23 < dongs> i wish ftdi would just die
2012-10-09T12:47:27 < dongs> overpriced tarduino shit
2012-10-09T12:48:01 <+Steffann> dongs prefers prolific
2012-10-09T12:48:02 < Thorn> it works and that is a plus.
2012-10-09T12:48:45 < Laurenceb> can anyone help me unclusterfuck git?
2012-10-09T12:49:00 < Laurenceb> i need to merge from a github repository
2012-10-09T12:49:34 < zyp> what's the situation?
2012-10-09T12:51:17 < Laurenceb> i changed local files without pulling from github
2012-10-09T12:53:55 < Laurenceb> and the github repo had been changed from elsewhere
2012-10-09T12:54:38 < dongs> sounds liek a typical opensores scenario, no?
2012-10-09T12:54:55 < dongs> im sure git has some mile-long command line to fix this
2012-10-09T12:56:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T12:56:53 < Laurenceb> http://www.sickipedia.org/
2012-10-09T12:56:58 < Laurenceb> hehe its still down
2012-10-09T12:57:17 < dongs> what was there
2012-10-09T12:57:24 < Laurenceb>  "whats the difference between April Jones and the Costa Concordia"
2012-10-09T12:57:53 < Laurenceb> "nothing, they are both at the bottom of the sea with a torn gash leaking seamen"
2012-10-09T12:58:02 <+Steffann> The only difference for me is.. i know what the Costa Concordia was, not who April Jones is
2012-10-09T12:58:19 < Laurenceb> someone got 1 year in jail for posting a link to that on farcebook
2012-10-09T12:58:30 < dongs> link to what
2012-10-09T12:58:31 < dongs> sikipedia?
2012-10-09T12:58:32 < Laurenceb> now sikipedia are shitting themselves
2012-10-09T12:58:42 < Laurenceb> yeah - the April Jones joke competition
2012-10-09T12:58:56 < dongs>  idont get it
2012-10-09T12:58:57 <+Steffann> You believe everything people say Laurenceb ?
2012-10-09T12:59:15 < Laurenceb> Steffanx: no its true
2012-10-09T12:59:52 < karlp> april jones was the ninja turtles journalist?
2012-10-09T12:59:57 <+Steffann> Could be possible.. GB is the little pet of yankieland of course :P
2012-10-09T13:00:58 < Laurenceb> http://www.telegraph.co.uk/news/uknews/crime/9595394/April-Jones-lanterns-light-sky-as-police-admit-we-may-never-find-body.html
2012-10-09T13:01:21 < Laurenceb> whats the difference between mark bridger and santa claus?
2012-10-09T13:01:39 < Laurenceb> mark bridger comes in april
2012-10-09T13:02:27 <+Steffann> Yes, you should be proud Laurenceb
2012-10-09T13:03:08 <+Steffann> Proud of yourself, that is
2012-10-09T13:03:39 < dongs> > 5 yo
2012-10-09T13:03:42 < dongs> > parents 40 and 43
2012-10-09T13:03:45 < dongs> wtf
2012-10-09T13:03:58 < Laurenceb> so
2012-10-09T13:04:42 < dongs> i dont get it
2012-10-09T13:04:42 < Laurenceb> too young?
2012-10-09T13:04:44 < dongs> why dont tehy just ask the guy
2012-10-09T13:04:48 < dongs> where he fucked her
2012-10-09T13:04:50 < dongs> or hwatever.
2012-10-09T13:04:57 < dongs> or is that too obvious
2012-10-09T13:05:19 < Laurenceb> presumably the sea
2012-10-09T13:05:25 < dongs> o
2012-10-09T13:05:50 < dongs> Laurenceb knows this all too well..
2012-10-09T13:05:58 < Laurenceb> yup
2012-10-09T13:06:22 < Laurenceb> somewhere i have the aquatic underage gore page bookmarked
2012-10-09T13:06:53 < zyp> Laurenceb, did you fix your shit?
2012-10-09T13:06:57 < Laurenceb> but ill spare you
2012-10-09T13:07:02 < Laurenceb> zyp: nope
2012-10-09T13:07:11 < Laurenceb> so i need to merge the changes from github
2012-10-09T13:07:16 < zyp> right
2012-10-09T13:07:29 < zyp> first of all, are the local changes committed or uncommitted?
2012-10-09T13:07:38 < Laurenceb> uncommitted
2012-10-09T13:07:51 < zyp> so you have no new local commits?
2012-10-09T13:08:03 <+Steffann> I'm glad everything you say is publicly logged and you use your own name with it, Laurenceb or should is say Laurence Blaxter?
2012-10-09T13:08:11 < dongs> haha
2012-10-09T13:08:30 < zyp> just stash the changes, pull and then merge the stash
2012-10-09T13:08:33 < zyp> git stash, git pull, git stash pop
2012-10-09T13:10:14 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-09T13:10:28 < Laurenceb> its not my real name
2012-10-09T13:11:03 <+Steffann> You wish
2012-10-09T13:11:04 < dongs> my real name is dongs
2012-10-09T13:11:24 < Laurenceb> loljews
2012-10-09T13:12:15 < Laurenceb> apparently
2012-10-09T13:14:43 < Laurenceb> ok it seemed to work
2012-10-09T13:15:10 < zyp> of course it did, that's why I told you to do it.
2012-10-09T13:15:12 < Laurenceb> but i have some conflicts with build files
2012-10-09T13:15:15 < Laurenceb> hehe
2012-10-09T13:15:20 < Laurenceb> can those be ignored
2012-10-09T13:15:33 < zyp> build files?
2012-10-09T13:15:51 < karlp> built files perhaps?
2012-10-09T13:15:54 < Laurenceb> CONFLICT (delete/modify): build/lst/usb_lld.lst deleted in Updated upstream and modified in Stashed changes.
2012-10-09T13:16:01 < Laurenceb> built then
2012-10-09T13:16:02 < karlp> we've told you before not to put those files in git in the first place
2012-10-09T13:16:07 < zyp> …
2012-10-09T13:16:23 < karlp> so yes, sure, ignore them, accept whatever ones you want to make the messages go away
2012-10-09T13:16:30 < karlp> and then go and properly delete them from the repo
2012-10-09T13:19:16 -!- daku is now known as DaKu
2012-10-09T13:20:38 < zyp> hmm, using SD cards seems pretty simple
2012-10-09T13:21:06 < zyp> and using the SDIO peripheral looks almost simpler than using SPI to access SD cards
2012-10-09T13:21:15 < dongs> right!
2012-10-09T13:21:26 < dongs> and its the correct way
2012-10-09T13:22:08 < zyp> sure
2012-10-09T13:22:08 < dongs> attn zyp http://www.ebay.com/itm/150471487665
2012-10-09T13:22:57 < zyp> I figure it's not much work to write an SD driver supporting both
2012-10-09T13:28:44 < dongs> looks like shitpal fixed my north korea status
2012-10-09T13:34:41 < Laurenceb> lol
2012-10-09T13:35:56 < dongs> In accordance with applicable regulations, PayPal is applying a cumulative sending limit of ¥100,000 JPY to all PayPal Japan accounts. We do so to prevent identity theft, money laundering, and the financing of terrorism.
2012-10-09T13:36:09 < dongs> i no longer get that warning.
2012-10-09T13:36:50 < Laurenceb> so you are ok to fund the jihad?
2012-10-09T13:36:53 < Laurenceb> awesome
2012-10-09T13:38:20 < dongs> ya
2012-10-09T13:40:47 -!- BrainDamage [~BrainDama@62.10.5.141] has joined ##stm32
2012-10-09T13:41:35 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-09T13:48:03 < Tectu> regarding to GDB and backtracing, what's exactly "a frame"?
2012-10-09T13:48:48 < dongs> stack frame, no?
2012-10-09T13:48:54 < dongs> kawaii frame
2012-10-09T13:49:02 < Tectu> yeah, but what's "a frame"?
2012-10-09T13:49:05 < Tectu> one elemnt of the stack?
2012-10-09T13:49:21 < dongs> no i think entier stack
2012-10-09T13:49:44 < Tectu> «A backtrace is a summary of how your program got where it is. It shows one line per frame, for many frames, starting with the currently executing frame (frame zero), followed by its caller (frame one), and on up the stack.»
2012-10-09T13:49:57 < zyp> each function has a stack frame
2012-10-09T13:50:17 < zyp> i.e. frame contains everything the corresponding function needs to store
2012-10-09T13:50:38 < zyp> when it calls another function, you get another frame on top of it
2012-10-09T13:50:44 < Tectu> ah
2012-10-09T13:50:45 < Tectu> i see
2012-10-09T13:51:00 < Tectu> so a frame is just some symbolic thing we make for better understanding?
2012-10-09T13:51:24 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-09T13:51:32 < Tectu> so the frame contains the entire stuff the "active" function put on the stack?
2012-10-09T13:51:32 < dongs> zyp: amazing new projecct http://wiki.open-drone.org/doku.php?id=flymaple
2012-10-09T13:51:56 <+Steffann> What's amazing about it?
2012-10-09T13:52:04 < dongs> </sarcasm>
2012-10-09T13:52:40 < Tectu> damn, aren't there enought drones out there?
2012-10-09T13:52:41 <+Steffann> Gmpf
2012-10-09T13:52:47 < dongs> you program it with tarduino
2012-10-09T13:52:56 < Tectu> wtf is tarduino=
2012-10-09T13:53:08 <+Steffann> dongs version of arduino
2012-10-09T13:53:13 < Laurenceb> i should stick a stm32f3 on one of my dactyl boards
2012-10-09T13:53:54 < dongs> Laurenceb: yeah the 303 version tahts not avialble.
2012-10-09T13:54:08 < Tectu> dongs, 404?
2012-10-09T13:54:36 < Laurenceb> annoying
2012-10-09T13:55:02 < Laurenceb> when this crazy austrian going to jump?
2012-10-09T13:55:07 < Laurenceb> i cant find ustream of it
2012-10-09T13:55:34 < Tectu> Laurenceb, it was in the local news saturday night here
2012-10-09T13:55:39 < Tectu> so i guess it will be soon :D
2012-10-09T13:56:00 < Laurenceb> yes its scheduled to liftoff now
2012-10-09T13:56:17 < Tectu> oh
2012-10-09T13:56:18 < Tectu> xD
2012-10-09T13:56:25 < Tectu> Laurenceb, got live stream?
2012-10-09T13:56:25 < dongs> the guy who gonna jump off a helium baloon in a spacesuit?
2012-10-09T13:56:32 < Tectu> dongle, yes
2012-10-09T13:56:36 < dongs> stream plz
2012-10-09T13:56:36 < Tectu> dongs, -.-
2012-10-09T13:56:44 < dongs> its ok dongle is me too
2012-10-09T13:56:50 < Tectu> cool
2012-10-09T13:56:50 < Laurenceb> nope
2012-10-09T13:56:54 < Laurenceb> searching for it
2012-10-09T13:56:56 < Tectu> Laurenceb, make one
2012-10-09T13:57:04 < Tectu> Laurenceb, please share as soon as you have any
2012-10-09T13:57:07 < Laurenceb> ...
2012-10-09T13:57:19 < Laurenceb> well i am in a 25 story building
2012-10-09T13:57:30 < Tectu> wtf
2012-10-09T13:57:39 < Laurenceb> guess i could stick a cam up there and jump out of the window
2012-10-09T13:57:46 < Tectu> Laurenceb, yes please
2012-10-09T13:57:53 <+Steffann> http://www.youtube.com/watch?v=vkJ5ItzEq3M&feature=player_embedded
2012-10-09T13:57:54 < Tectu> Laurenceb, can you open the windows?
2012-10-09T13:57:55 < Laurenceb> k
2012-10-09T13:58:04 <+Steffann> ^^ live on youtube
2012-10-09T13:58:15 < Tectu> Steffann, will that be live
2012-10-09T13:58:16 < Tectu> aw
2012-10-09T13:58:21 < Laurenceb> i didnt know that was possible
2012-10-09T13:58:25 <+Steffann> A delay of 20 sec. i read somewhere :P
2012-10-09T13:58:29 < Laurenceb> booo
2012-10-09T13:58:33 < dongs> i hope he dies
2012-10-09T13:58:36 < Laurenceb> more like 1hour 30 min
2012-10-09T13:58:39 < Tectu> Laurenceb, youtube does live stuff since a year
2012-10-09T13:58:40 < Laurenceb> loldongs
2012-10-09T13:58:42 < Laurenceb> ok
2012-10-09T13:58:48 < Tectu> dongle, you're an ass :D
2012-10-09T13:58:49 <+Steffann> The delay is just because they don't want us to see him die
2012-10-09T13:58:51 < cjbaird> Quadcopters are too mainsteam. https://en.wikipedia.org/wiki/Avro_Canada_VZ-9_Avrocar
2012-10-09T13:59:19 <+Steffann> Wouldn't be surprised, just like they filmed the man who shot himself ( somewhere in the US )
2012-10-09T13:59:26 <+Steffann> *surprised if they fail
2012-10-09T13:59:28 < Tectu> Steffann, wtf?
2012-10-09T13:59:33 <+Steffann> Didn't see that?
2012-10-09T13:59:39 < Tectu> no?
2012-10-09T13:59:42 < Tectu> when and where and WHY?
2012-10-09T13:59:53 <+Steffann> When, last week? Where the US
2012-10-09T13:59:57 < dongs> lol
2012-10-09T13:59:59 <+Steffann> Why? The man didn't tell me
2012-10-09T14:00:00 < dongs> it was some nigga
2012-10-09T14:00:38 < Laurenceb> at least he wasnt a jew
2012-10-09T14:00:46 < Laurenceb> or a dongs
2012-10-09T14:00:54 <+Steffann> It was live on TV, it had a delay, but they forgot to turn the live stream off when the man killed himself
2012-10-09T14:00:56 < Tectu> wtf happend to this channel
2012-10-09T14:01:04 < Tectu> Steffann, lol fail
2012-10-09T14:01:05 <+Steffann> forgot as in failed to switch
2012-10-09T14:01:08 < Tectu> Steffann, what did he try to do?
2012-10-09T14:01:21 < dongs> he was some nig who stole a car
2012-10-09T14:01:26 < dongs> so he killed himself
2012-10-09T14:01:27 <+Steffann> It looks very weird, and you can't see much
2012-10-09T14:01:29 < dongs> wehn they cornered him
2012-10-09T14:01:37 < dongs> and nothing of value was lost
2012-10-09T14:01:58 < Tectu> Steffann, seriously, what was it about?
2012-10-09T14:02:07 < dongs> I just told you FUCKER
2012-10-09T14:02:11 <+Steffann> Have to look if i can find the video
2012-10-09T14:03:22 < Tectu> Steffann, @work
2012-10-09T14:03:26 < Tectu> dongle, OMG fuck yourself :)
2012-10-09T14:03:48 <+Steffann> http://entertainment.time.com/2012/09/28/fox-news-apologizes-after-airing-live-footage-of-suicide/ the story in text Tectu
2012-10-09T14:04:58 < Tectu> «...apologized to his audience this afternoon after his program inadvertently aired live video of a man killing himself.»
2012-10-09T14:05:02 < Tectu> lol, that sounds so weird
2012-10-09T14:05:57 < cjbaird> Snuff News. All
2012-10-09T14:06:03 < cjbaird> Snuff News. All Snuff, all the time!
2012-10-09T14:10:10 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-09T14:19:04 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-09T14:20:14 < Tectu> dongs,
2012-10-09T14:20:40 < dongs> sup
2012-10-09T14:21:03 < Tectu> how are you?
2012-10-09T14:21:08 < dongs> wth
2012-10-09T14:21:09 < dongs> fine.
2012-10-09T14:21:38 < Tectu> that's good to hear :)
2012-10-09T14:26:53 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-09T14:31:14 < Tectu> dongs, and did you sleep well?
2012-10-09T14:31:33 < dongs> Tectu: you're entering the "wtf is this shit" territory.
2012-10-09T14:32:46 < Tectu> dongs, 1. would that be new?    2. would that be bad?
2012-10-09T14:32:58 < dongs> yes to both
2012-10-09T14:33:07 < BrainDamage> http://i.imgur.com/yGiOW.jpg
2012-10-09T14:33:23 < Tectu> dongs, explain the second
2012-10-09T14:34:07 <+Steffann> I think Tectu thinks dongs needs a psychologist
2012-10-09T14:34:14 < Tectu> Steffann, no
2012-10-09T14:34:27 < Tectu> Steffann, I just think that he needs someone ho loves him and take care of him
2012-10-09T14:34:32 < Tectu> :)
2012-10-09T14:34:53 < Laurenceb> called "bubba"
2012-10-09T14:35:12 <+Steffann> You want him to go into jail?
2012-10-09T14:35:30 < dongs> tectu is not kawaii enough to fill that role though
2012-10-09T14:35:34 < Laurenceb> well.. he can keep flyback company
2012-10-09T14:35:38 < dongs> haha
2012-10-09T14:35:53 < Laurenceb> flyback obviously has bubba issues
2012-10-09T14:35:59 <+Steffann> Laurenceb, 54 minutes remaining
2012-10-09T14:36:06 < Laurenceb> cool
2012-10-09T14:36:36 < dongs> 53.25
2012-10-09T14:36:43 < dongs> i should take a bath first.
2012-10-09T14:36:45 <+Steffann> 15
2012-10-09T14:36:48 < dongs> then I can troll as that guy plummets to the ground
2012-10-09T14:36:51 <+Steffann> Yes, you smell awful
2012-10-09T14:37:06 < dongs> no my smell is kawaii
2012-10-09T14:37:08 < dongs> ok bbl
2012-10-09T14:38:30 < cjbaird> ##stm32-hugs
2012-10-09T14:38:33  * Tectu strokes dongs 
2012-10-09T14:40:27 < cjbaird> Tectu: no yiffing
2012-10-09T14:40:28 < Laurenceb> fag
2012-10-09T14:43:54 < Laurenceb> i found a photo of dongs
2012-10-09T14:43:56 < Laurenceb> http://media.heavy.com/media/2010/03/king.jpg
2012-10-09T14:43:58 < Laurenceb> nsfw
2012-10-09T14:47:37 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-09T14:48:34 < Laurenceb> actually i was wrong
2012-10-09T14:48:46 < Laurenceb> http://poorlydressed.files.wordpress.com/2012/06/fashion-fail-thats-not-kawaii-not-at-all.jpg
2012-10-09T14:50:09 < zyp> http://www.redbullstratos.com/live/ so yeah
2012-10-09T14:50:23 < Laurenceb> oh its shopped
2012-10-09T14:50:25 < Laurenceb> fail
2012-10-09T14:50:30 < Laurenceb> didnt spot the feet :(
2012-10-09T14:51:02 <+Steffann> Yes, that's the youtube one zyp :)
2012-10-09T14:51:25 < zyp> oh, right
2012-10-09T14:51:34 < zyp> I just saw it linked on another channel
2012-10-09T14:51:55 <+Steffann> but this one shows some more info
2012-10-09T14:56:13 < karlp> who was asking about ST's process capabilities here the other day?
2012-10-09T14:56:36 <+Steffann> Uhm
2012-10-09T14:56:39 < Bird|lappy> what's the smallest package (footprint-wise) one can get a 32.768kHz crystal in?
2012-10-09T14:56:46 < karlp> 20nm test chips have been made apparently, but not yet in production.
2012-10-09T14:58:22 < Laurenceb> yeah i read that
2012-10-09T14:58:29 < Laurenceb> sounds pretty awesome
2012-10-09T15:00:11 < zyp> karlp, source?
2012-10-09T15:00:31 < Laurenceb> http://imagebin.org/231372
2012-10-09T15:00:33 < Laurenceb> working
2012-10-09T15:00:56 < karlp> zyp: my friend at mentor in grenoble.
2012-10-09T15:01:02 < zyp> ah
2012-10-09T15:01:04 < karlp> he worked for st in crolle before.
2012-10-09T15:01:13 < karlp> he says you st-e guys are the biggest consumers of 28nm parts.
2012-10-09T15:01:19 < karlp> won't say who the test chips are being made for.
2012-10-09T15:01:28 < zyp> :)
2012-10-09T15:03:04 < dongs> on weather hold?
2012-10-09T15:04:05 < Laurenceb> waiting for wind to die down
2012-10-09T15:04:09 <+Steffann> meh
2012-10-09T15:05:11 < dongs> wtf does that mean
2012-10-09T15:05:24 < karlp> it means he's scared shitless
2012-10-09T15:05:30 <+Steffann> 13:30PM .. Why not 1:30PM ?!
2012-10-09T15:05:48 < Laurenceb> lol
2012-10-09T15:06:07 < BrainDamage> hah, I have the same shitty chinese psu
2012-10-09T15:06:19 <+Steffann> Mine is shittier
2012-10-09T15:06:32 < Laurenceb> stm32f4discovery is shittier
2012-10-09T15:06:41 < Laurenceb> as far as power supply goes
2012-10-09T15:06:43 <+Steffann> In your hands maybe
2012-10-09T15:06:46 < BrainDamage> btw, can't you just move the CE mark from the thing on the right side to your product?
2012-10-09T15:06:51 < Laurenceb> hehe
2012-10-09T15:08:19 < dongs> so when the fuck is it 1.30pm gmt
2012-10-09T15:08:24 < dongs> what time is in eurotrash now
2012-10-09T15:08:26 < karlp> in 90min
2012-10-09T15:08:26 < Laurenceb> 20 minutes
2012-10-09T15:08:35 < Laurenceb> or that
2012-10-09T15:08:42 < karlp> 80min utc, 20min bst
2012-10-09T15:08:42 < Laurenceb> im on BST here
2012-10-09T15:08:53 <+Steffann> I'm GMT+2 :P
2012-10-09T15:09:09 < karlp> I live in a proper country, where there is UTC, and only UTC, all the time, all the year, no matter where the sun may or may not be.
2012-10-09T15:09:31 < BrainDamage> do you mean the utc shift?
2012-10-09T15:09:34 < dongs> proper time is WTC
2012-10-09T15:09:42 < BrainDamage> or that you're in utc time slot?
2012-10-09T15:09:51 < Laurenceb> lol
2012-10-09T15:09:53 < Laurenceb> WTC7
2012-10-09T15:10:13 < Laurenceb> what are those things in the sky?
2012-10-09T15:10:24 < Laurenceb> looks like tethered balloons or something
2012-10-09T15:10:29 < Laurenceb> or the greys?
2012-10-09T15:10:49 <+Steffann> Helicopters? They move...
2012-10-09T15:10:56 < BrainDamage> the video doesn't load :/
2012-10-09T15:10:56 <+Steffann> I've no idea
2012-10-09T15:11:00 < BrainDamage> "please stand by"
2012-10-09T15:11:28 < dongs> reload
2012-10-09T15:11:31 < dongs> they siwtched from countdown
2012-10-09T15:11:32 < dongs> to live feed
2012-10-09T15:11:40 <+Steffann> Too bad even in 1080p the quality sucks
2012-10-09T15:11:58 < Laurenceb> F5
2012-10-09T15:12:18 <+Steffann> Doesn't work
2012-10-09T15:12:27 < dongs> ctrl+r
2012-10-09T15:12:29 < dongs> who the fuck uses f5??
2012-10-09T15:12:44 <+Steffann> I won't say that doesn't work either
2012-10-09T15:13:46 < karlp> who fuck uses ctrl-r?
2012-10-09T15:13:46 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T15:14:05 <+Steffann> cmd+r it is :P
2012-10-09T15:14:10 < karlp> BrainDamage: I happen to live in a timezone that is UTC, with no daylight savings.
2012-10-09T15:14:34 < dongs> mac faggots
2012-10-09T15:14:55 < karlp> and we're far enough west that we shouldn't be by most uses of the 15 deg / hour thing, so we're just used to times being an idea, not anything specifically related to the position of the sun
2012-10-09T15:15:24 < karlp> and I (naively perhaps) think that if we all just used the same time, no matter where in the world we are, we'd quickly get used to whatever times made sense locally,
2012-10-09T15:15:43 < BrainDamage> I agree
2012-10-09T15:15:57 < karlp> makes for some rather arbitrary change of day defintions though
2012-10-09T15:16:21 < Bird|lappy> do you folks use guard rings on your watch-crystal oscillators?
2012-10-09T15:17:20 < Thorn> why does BMP have both a switched 1K5 resistor on usb d+ and a Vbus detection circuit? isn't it a bus powered device?
2012-10-09T15:17:35 < dongs> ?
2012-10-09T15:17:48 < dongs> probably so they can renumerate
2012-10-09T15:17:56 < dongs> and they were too lazy to steal the proper renumerate sores from openpilot :)
2012-10-09T15:18:01 < dongs> that does it without external components
2012-10-09T15:18:48 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 260 seconds]
2012-10-09T15:20:03 < zyp> karlp, I agree, I've made the same argument myself before
2012-10-09T15:21:06 < dongs> hey its getting brighter
2012-10-09T15:21:31 <+Steffann> heh
2012-10-09T15:22:14 <+Steffann> It'll be a nice and romantic sun rise dongs
2012-10-09T15:22:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T15:25:02 -!- corecode [~2@0x2c.org] has quit [Quit: switch]
2012-10-09T15:33:22 < Thorn> dongs: you were selecting mosfets for your led project, what did you end up using?
2012-10-09T15:42:12 < dongs> FUCK THEY MOVED IT
2012-10-09T15:42:16 < dongs> Thorn: lets see
2012-10-09T15:42:23 < dongs> something in SOT23
2012-10-09T15:42:27 < dongs> i have it in my digikey bag
2012-10-09T15:44:20 < dongs> umn.. or not.
2012-10-09T15:44:38 < dongs> but I think what happened is that once I switched to using 3:8 mux for fet driver
2012-10-09T15:44:39 <+Steffann> Yet another 0.5 hour postponed
2012-10-09T15:44:41 < dongs> any logic level fet will work
2012-10-09T15:44:50 < dongs> since mux will pull gate to 5V
2012-10-09T15:45:11 < Thorn> did you use that chinese led driver?
2012-10-09T15:45:15 < dongs> yes
2012-10-09T15:45:19 < dongs> dd you see the board?
2012-10-09T15:45:26 < dongs> http://i.imgur.com/WkdiQ.png MY9221
2012-10-09T15:45:27 < Thorn> and what's the circuit, is it a matrix or something?
2012-10-09T15:45:42 <+Steffann> why is it so large dongs?
2012-10-09T15:45:50 <+Steffann> I'm sure you can make it smaller
2012-10-09T15:45:51 < dongs> Steffann: it fits behind the matrix.
2012-10-09T15:46:07 <+Steffann> Oh, you don't have to make it smaller
2012-10-09T15:46:09 < dongs> its kinda silly making it smaller since.. the matrix will be in the way anyway
2012-10-09T15:46:09 < Thorn> ah so it is a matrix. I see
2012-10-09T15:47:15 < Thorn> I'm thinking about a small rgb led strip cnontroller with CAN
2012-10-09T15:47:29 < Thorn> using LPC11C22
2012-10-09T15:47:29 < dongs> you and your can stuff.
2012-10-09T15:47:44 < dongs> i was like emeb, should we do usb or can on this ledmatrix shit
2012-10-09T15:47:47 < dongs> and hes ilke duhh, usb
2012-10-09T15:48:03 <+Steffann> and bluetooth!
2012-10-09T15:48:05 < Thorn> you can't route usb around the room can you?
2012-10-09T15:48:16 < dongs> only 1st block has usb
2012-10-09T15:48:19 < dongs> the rest is spi-like shit.
2012-10-09T15:48:33 < Thorn> without microcontrollers?
2012-10-09T15:48:36 < dongs> wiht
2012-10-09T15:48:42 < dongs> cause it has to multiplex rows.
2012-10-09T15:48:49 < Thorn> why not CAN then.
2012-10-09T15:49:15 < dongs> because its a chain
2012-10-09T15:49:19 < dongs> not multidrop.
2012-10-09T15:49:24 < Thorn> how do I drop 0.1A from 12V to 3.3V?
2012-10-09T15:49:36 < dongs> switcher
2012-10-09T15:50:00 < Thorn> probably
2012-10-09T15:51:23 < zyp> Thorn, 87 ohm series resistor
2012-10-09T15:51:37 < zyp> and if that's not what you wanted, you need to state the question better :p
2012-10-09T15:52:38 < Thorn> how do I drop 0.1A from 12V to 3.3V without creating a mini room heater?
2012-10-09T15:52:40 < dongs> what does that do zyp
2012-10-09T15:52:43 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Read error: Connection reset by peer]
2012-10-09T15:52:47 -!- BjoernC_ [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T15:57:35 -!- BjoernC_ [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 260 seconds]
2012-10-09T15:59:12 < dongs> whats up with the launch
2012-10-09T15:59:18 < dongs> delayed forever?
2012-10-09T15:59:34 <+Steffann> I guess so
2012-10-09T15:59:45 < qyx_> porbably
2012-10-09T15:59:48 <+Steffann> We now can look at the nice sun rise
2012-10-09T16:01:22 < dongs> ..
2012-10-09T16:01:23 < dongs> delayed again
2012-10-09T16:01:49 <+Steffann> Hmp
2012-10-09T16:01:50 <+Steffann> f
2012-10-09T16:03:41 <+Steffann> And then it has to get up  36KM how long does THAT take?
2012-10-09T16:05:06 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-09T16:05:12 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Client Quit]
2012-10-09T16:05:33 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-09T16:05:46 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-09T16:07:03 -!- mrcan__ [~mrcan@78.173.72.192] has joined ##stm32
2012-10-09T16:07:04 -!- mrcan__ [~mrcan@78.173.72.192] has quit [Changing host]
2012-10-09T16:07:04 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-09T16:08:33 < dongs> lol
2012-10-09T16:08:36 < dongs> the dorks are changing nd filters
2012-10-09T16:08:46 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Client Quit]
2012-10-09T16:09:13 <+Steffann> Going to be a long night for dongs?
2012-10-09T16:09:21 < dongs> im not staying up
2012-10-09T16:09:26 < dongs> i gotta drive to pickup shit tomrow at 9am
2012-10-09T16:09:32 <+Steffann> That's not kawaii
2012-10-09T16:09:34 < dongs> i slept in until like 11.30 today
2012-10-09T16:09:39 < dongs> fucking fail
2012-10-09T16:10:43 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 256 seconds]
2012-10-09T16:14:57 -!- BjoernC_ [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T16:18:09 < dongs> < Bruno`> what are the possible reasons for writting garbage on the serial port ?
2012-10-09T16:18:14 < dongs> < Yotson> em pulse from nearby nuclear explosion.
2012-10-09T16:24:38 -!- BjoernC__ [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T16:27:02 <+Steffann> Meeh, 17:30 now :S
2012-10-09T16:27:10 <+Steffann> GMT
2012-10-09T16:28:07 < dongs> raped
2012-10-09T16:28:20 <+Steffann> Have a nice sleep :P
2012-10-09T16:28:29 -!- BjoernC_ [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 260 seconds]
2012-10-09T16:29:00 < Laurenceb> boo
2012-10-09T16:33:59 -!- dekar [~dekar@212.255.19.185] has quit [Quit: This computer has gone to sleep]
2012-10-09T16:34:57 < Tectu> Steffann, still nothing?
2012-10-09T16:35:04 <+Steffann> Still nothing what?
2012-10-09T16:35:09 < Tectu> Steffann, the live stream
2012-10-09T16:35:30 <+Steffann> The live stream is there, but all you seen is the sun
2012-10-09T16:37:43 < zyp> Steffann, you're not making sense
2012-10-09T16:37:47 < zyp> 15:27:02 <+Steffann> Meeh, 17:30 now :S
2012-10-09T16:37:59 < zyp> I'm in +2, so it was 13:27 when you said it.
2012-10-09T16:38:38 <+Steffann> I was refering to the "earliest launch" time of that red bull stuff
2012-10-09T16:38:55 < zyp> so in four hours?
2012-10-09T16:39:04 <+Steffann> Probably not
2012-10-09T16:39:09 < zyp> :p
2012-10-09T16:39:43 <+Steffann> Next update in 2 hours, earliest possible launch 4 hours
2012-10-09T16:39:49 <+Steffann> So => not going to happen today :P
2012-10-09T16:39:56 < zyp> smells like it'll be scrubbed, yeah
2012-10-09T16:42:32 < Tectu> but dude, he jumps from 38'000km, right?
2012-10-09T16:42:47 < Tectu> aw, m, not km
2012-10-09T16:43:11 <+Steffann> 38k km would be nice :P
2012-10-09T16:43:27 < BrainDamage> 38000 km would be a geostationary satellite
2012-10-09T16:43:41 < BrainDamage> unfortunately, balloons can't go much past 40km
2012-10-09T16:44:39 -!- kerwood_wk [80f40908@gateway/web/freenode/ip.128.244.9.8] has joined ##stm32
2012-10-09T17:02:32 < cjbaird> 100kms is the generally accepted point where you enter 'space'..
2012-10-09T17:05:25 -!- dekar [~dekar@studpool-wlan-75-90.fs.fbi.h-da.de] has joined ##stm32
2012-10-09T17:05:27 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-09T17:05:27 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-09T17:07:31 < cjbaird> They're finally getting around to scooping marsdirt now: http://mars.jpl.nasa.gov/msl/multimedia/raw/?rawid=0061ML0315000180M0_DXXX&s=61
2012-10-09T17:09:46 < dongs> emeb_mac: http://i.imgur.com/WkdiQ.png autoroute fixup done
2012-10-09T17:10:56 < emeb_mac> dongs: looks good.
2012-10-09T17:11:00 < emeb_mac> cleaner
2012-10-09T17:12:13 < emeb_mac> like the fat +5V routes to the PFETs
2012-10-09T17:12:15 < dongs> yeah, no more clock trace going around half hte board
2012-10-09T17:12:37 <+dekar> "The new Java machine is designed for full headless operation and is capable of running on devices with as little as 130 kB of RAM and 350 kB of ROM."
2012-10-09T17:12:47 < zyp> dongs, no chaining 5V from board to board?
2012-10-09T17:12:51 < dongs> zyp: nope
2012-10-09T17:13:16 < zyp> why not?
2012-10-09T17:13:22 < dongs> zyp, voltage drop
2012-10-09T17:13:37 < zyp> so what, you're using constant current drivers
2012-10-09T17:13:39 < dongs> and more and more current that needs to be handled by the board in front
2012-10-09T17:13:51 < dongs> zyp, 0.5A at first board + 3 boards behind it = 2A on 1st board
2012-10-09T17:14:19 < zyp> you obviously hook input power to middle board
2012-10-09T17:14:24 <+Steffann> Sure dekar :P
2012-10-09T17:14:29 < jpa-> dongs: you could make the top layer groundplane more connected by just moving around a few vias (like under the "DATA IN" text)
2012-10-09T17:15:07 < zyp> and you've got enough free area to run a beefy trace from one end to the other to chain it
2012-10-09T17:16:19 < dongs> zyp, was emebs idea.
2012-10-09T17:16:26 < dongs> i can certainly fix it
2012-10-09T17:16:35 < dongs> jpa, i dont think that would work.
2012-10-09T17:16:41 <+dekar> I wonder who would prefer Java over C on deeply embedded targets
2012-10-09T17:16:44 < dongs> i see what youre saying but the backside under drivers is pretty limited
2012-10-09T17:17:34 <+dekar> 350kb rom is huge, and that's just the runtime environment
2012-10-09T17:18:00 < dongs> who cares about java
2012-10-09T17:18:36 < zyp> dongs, how thick was that power trace? I think you said 35mil earlier?
2012-10-09T17:18:46 < dongs> yeah but that was huge
2012-10-09T17:18:50 < dongs> so i moved to 24mil
2012-10-09T17:19:17 < zyp> and the board will be done with 35um cu?
2012-10-09T17:19:21 < dongs> ya
2012-10-09T17:19:29 < emeb_mac> yeah - blame it on me. :P
2012-10-09T17:19:59 < zyp> right, so at 2A through that trace you'll still have less than 0.1V drop across a board
2012-10-09T17:20:43 < dongs> fine, so Ill add 5V through
2012-10-09T17:20:48 < dongs> move uart inward
2012-10-09T17:20:53 < dongs> and put it there
2012-10-09T17:21:04 < dongs> with like 36mil trace.
2012-10-09T17:21:08 < jpa-> dongs: i mean like this, just to make the ground plane more connected around the signals to avoid EMI: http://kapsi.fi/~jpa/stuff/pix/dongspcb.png
2012-10-09T17:21:14 < emeb_mac> dongs: add it in for short chains
2012-10-09T17:21:32 < emeb_mac> leave the direct connect for larger ones.
2012-10-09T17:21:42 < BrainDamage> not the right channel for it, but does  anyone have articles on propagation infos for 2.4GHz ISM band on atmospheric layers?
2012-10-09T17:21:50 < zyp> either way, for longer chains you can still use it to run shorter segments of the chain
2012-10-09T17:22:54 < dongs> jpa-: noted
2012-10-09T17:23:59 < emeb_mac> man - tough crowd. design reviews on IRC - I like it.
2012-10-09T17:24:31 < zyp> :)
2012-10-09T17:24:40 < dongs> haha
2012-10-09T17:25:32 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-09T17:25:47 < zyp> it looks a bit interesting, I wonder how fast/large chains you can build with it
2012-10-09T17:26:18 < cjbaird> Could there be just one nude model on Flickr who doesn't have tattoos? :/
2012-10-09T17:27:08 <+Steffann> Should I quote you cjbaird? :P
2012-10-09T17:27:27 <+Steffann> Let me find what you exactly said
2012-10-09T17:34:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-09T17:35:01 < dongs> hmm
2012-10-09T17:35:06 < dongs> i could rotate that 5V connector on top
2012-10-09T17:35:08 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-09T17:35:18 < dongs> so its vertical
2012-10-09T17:35:26 < dongs> i dont wanna remove it cuz... hooking shit up  to sides is annoying
2012-10-09T17:38:57 < jpa-> there are some multidrop connectors that might be useful.. like http://library.delphi.com/images/library/loResJpegDownload/IDC0503-1170_1.jpeg
2012-10-09T17:41:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-09T17:44:39 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 276 seconds]
2012-10-09T17:47:52 < dongs> nah im just using 2.54mm spaced shit
2012-10-09T17:48:01 < dongs> that does look cool but probably costs more than the board
2012-10-09T17:50:44 -!- claude_work is now known as claude
2012-10-09T17:50:45 < jpa-> maybe
2012-10-09T17:57:21 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-09T17:58:55 < cjbaird> Hmm, the F3Disco won't arrive at E14/AU until the 21st of December. :/
2012-10-09T17:59:03 < dongs> funtimes.
2012-10-09T18:02:38 < cjbaird> It looks like Renesas is finally getting in on the cheap-eval-board sexiness.. http://am.renesas.com/products/tools/introductory_evaluation_tools/cpu_board/index.jsp .. http://au.element14.com/renesas/qb-78k0lg3m-tb/cpu-board-78k0-lg3-m/dp/2137159
2012-10-09T18:04:09 < cjbaird> I know bugger-all about the 78K series at this point, though.
2012-10-09T18:04:32 <+Steffann> Whoa, they are finally doing something. "Balloon layout has begun"
2012-10-09T18:05:36 -!- dekar_ [~dekar@studpool-wlan-75-90.fs.fbi.h-da.de] has joined ##stm32
2012-10-09T18:05:39 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-09T18:05:55 <+Steffann> Renesas doesn't need that cjbaird
2012-10-09T18:06:19 <+Steffann> According to some presentation from ST Renesas is the largest in the business
2012-10-09T18:06:53 <+Steffann> In revenues that is
2012-10-09T18:07:33 <+Steffann> ( and only in the general purpose uc business )
2012-10-09T18:07:35 < jpa-> largest in the business of manufacturing outdated 8051 clones
2012-10-09T18:07:57 <+Steffann> but still
2012-10-09T18:08:12 -!- dekar [~dekar@studpool-wlan-75-90.fs.fbi.h-da.de] has quit [Ping timeout: 245 seconds]
2012-10-09T18:08:20 <+Steffann> Almost 2 billion vs ~400million for ST
2012-10-09T18:08:24 < jpa-> sure, there are still a lot of them in use.. but not popular for new designs
2012-10-09T18:08:37 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 245 seconds]
2012-10-09T18:08:59 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-09T18:09:40 <+Steffann> According to this chart the revenues of Atmel are higher than those of microchip
2012-10-09T18:09:49 <+Steffann> I didn't expect that
2012-10-09T18:10:15 < cjbaird> The IBM of MCU..
2012-10-09T18:10:29 < BrainDamage> atmel of microchip?
2012-10-09T18:10:34 < BrainDamage> or*
2012-10-09T18:10:38 < jpa-> crazy expensive board and 8 bit :F
2012-10-09T18:10:52 < jpa-> renesas
2012-10-09T18:12:12 < cjbaird> tiiiiny rams, too.
2012-10-09T18:12:35 < karlp> Steffann: atmel makes lots of money from flash,
2012-10-09T18:13:18 <+Steffann> True, but this is a "WW GP & Secure MCUs Revenues" comparison, so i'm not sure if that's included
2012-10-09T18:13:18 -!- BjoernC__ [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 246 seconds]
2012-10-09T18:14:26 <+Steffann> ( this chart http://www.naffets.nl/share//a-20121009-171415.png )
2012-10-09T18:17:03 < BrainDamage> meh, burn experiment was a partial failure :/
2012-10-09T18:17:39 < BrainDamage> anything to have a decent combustion with NaOx leftovers?
2012-10-09T18:18:06 -!- BjoernC__ [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T18:19:45 -!- dekar_ [~dekar@studpool-wlan-75-90.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-09T18:19:45 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-09T18:24:30 < zyp> guy I know apparently got his Teensy 3.0 today
2012-10-09T18:25:57 < dongs> sad.
2012-10-09T18:26:08 < zyp> must be the first kickstarter project I've heard about that didn't get delayed
2012-10-09T18:36:05 < karlp> pjrc didn't put up the kickstart til it was almost finished
2012-10-09T18:36:12 < karlp> he's a pretty solid guy.
2012-10-09T18:36:38 < karlp> I'm a little sad he didn't encourage people to start using debug but teensy3 will probably be pretty great
2012-10-09T18:38:15 <+Steffann> The question is.. does that pjrc guy really need ks?
2012-10-09T18:38:40 < karlp> not really, but it gets him to an audience he didn't nromally have
2012-10-09T18:38:44 < karlp> so... moah moneh?
2012-10-09T18:38:49 < zyp> karlp, the bootloader stuff is just weird
2012-10-09T18:38:51 < karlp> why not use the damn thing if you were going to be using it anyway
2012-10-09T18:38:53 < BrainDamage> most people use kickstater as marketing
2012-10-09T18:39:05 < BrainDamage> not so much as store as for exposure
2012-10-09T18:39:07 < karlp> zyp: his bootloader on teensy 2 is awesome.
2012-10-09T18:39:23 < zyp> aren't tennsy 2 avr?
2012-10-09T18:39:26 <+Steffann> Yes
2012-10-09T18:39:27 < BrainDamage> yes
2012-10-09T18:39:27 < karlp> yup
2012-10-09T18:39:27 < zyp> teensy*
2012-10-09T18:40:09 < zyp> usb avrs already ship with dfu bootloaders, so that's kind of a moot point
2012-10-09T18:40:32 < BrainDamage> arduino compatibility probably
2012-10-09T18:40:48 < dongs> zyp, bootloader stuff + jtag is not connected
2012-10-09T18:40:52 < dongs> zyp: plus its all closed sores.
2012-10-09T18:40:59 < dongs> fuck pjrc and fuck all of teensy shit
2012-10-09T18:41:02 < dongs> fucking scammer.
2012-10-09T18:42:30 < karlp> I thought'd you lvoe it.
2012-10-09T18:42:41 < karlp> closed sores and a guy who actually promptly responds to fixes and problems.
2012-10-09T18:42:47 < karlp> unlike arduino core
2012-10-09T18:43:26 <+Steffann> dongs i remember you have a scam website as well
2012-10-09T18:44:16 <+Steffann> don't remember the url though
2012-10-09T18:52:13 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-09T18:58:35 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-09T19:16:40 -!- BjoernC__ [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-09T19:20:04 -!- WorkerBuzz [~Buzz@134.20.83.19] has quit [Ping timeout: 248 seconds]
2012-10-09T19:22:52 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-09T19:24:59 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-09T19:25:03 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-09T19:31:09 < dongs> http://i.imgur.com/kcdRb.png dunno
2012-10-09T19:31:55 <+Steffanx> 30 minutes remaining ( to get some new stream )
2012-10-09T19:32:06 <+Steffanx> *boardcast start
2012-10-09T19:32:16 < emeb> dongs: what's wrong with that?
2012-10-09T19:32:22 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-09T19:32:27 < dongs> emeb: nothing as far as I know
2012-10-09T19:32:45 -!- mrcan__ [~mrcan@78.173.72.192] has joined ##stm32
2012-10-09T19:32:46 -!- mrcan__ [~mrcan@78.173.72.192] has quit [Changing host]
2012-10-09T19:32:46 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-09T19:32:54 < emeb> Someone will probably complain that the 5V chain isn't on the same header as the data.
2012-10-09T19:33:02 < dongs> right, fuck them
2012-10-09T19:33:11 -!- dekar [~dekar@212.255.19.185] has joined ##stm32
2012-10-09T19:33:12 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-09T19:33:12 < emeb> vigorously
2012-10-09T19:33:14 <+Steffanx> jpa- you linked that nice connector. Something like that has a special name?
2012-10-09T19:42:37 < zippe> Steffanx: It's just an IDC connector
2012-10-09T19:42:50 < zippe> Steffanx: with a top shroud set up for dual entry
2012-10-09T19:43:12 < dongs> dual entry, sounds kinky
2012-10-09T19:43:21 <+Steffanx> Only in your country
2012-10-09T19:45:51 <+Steffanx> ARGH, another 1/2 hour added to the countdown
2012-10-09T19:46:04 < zyp> nice
2012-10-09T19:46:25 < dongs> ha ha
2012-10-09T19:46:27 < zyp> I'm just waiting for somebody to mention on irc that stuff is actually happening
2012-10-09T19:46:36 < dongs> i bet i'll wake up and the fucking thing isn't even gonna l aunch
2012-10-09T19:46:40 <+Steffanx> heh
2012-10-09T19:46:54 < dongs> <Funklord> is it normal for 0805 caps to get slightly damaged when soldering by hand?
2012-10-09T19:47:05 < dongs> Yaaaaaaaaah.
2012-10-09T19:47:06 <+Steffanx> It also takes at least 2 hours to get up to 36km or so, so .. good night dongs :P
2012-10-09T19:50:26 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T20:02:52 -!- barthess [~barthess@178.154.21.145] has joined ##stm32
2012-10-09T20:05:12 < Posterdati> hi
2012-10-09T20:05:59 <+Steffanx> Lo
2012-10-09T20:06:31 < Posterdati> finally my rtos begin to work thanks to your hints
2012-10-09T20:08:30 < Posterdati> now I'm facing another problem: I need i2c routines to not be preempted by any IRQ, I try to use __disable_irq() and __enable_irq(), but I've got HardFault during PendSV_Handler execution
2012-10-09T20:24:41 < emeb> cute: http://vimeo.com/49676399
2012-10-09T20:24:53 < emeb> (this one is for the copter fans)
2012-10-09T20:25:02 <+Steffanx> Who here isn't a copter fan?
2012-10-09T20:25:13 < emeb> my point
2012-10-09T20:25:20 <+Steffanx> Oh, i just saw that video somewhere else
2012-10-09T20:25:36 < emeb> any you didn't link it here immediately. shame.
2012-10-09T20:25:40 < BrainDamage> I'm not, airplanes > n-copters
2012-10-09T20:25:50 <+Steffanx> It's nice how stabilized the camera is
2012-10-09T20:25:54 < BrainDamage> unless you consider tilt rotors n-copters
2012-10-09T20:26:03 < emeb> yeah - very solid.
2012-10-09T20:26:12 <+Steffanx> Don't like to link to this dutch many random videos website, emeb
2012-10-09T20:26:49 <+Steffanx> ' "I am strapped into the capsule, and I am ready to go." Felix Baumgartner to Joe Kittinger in Mission Control ' … YAY
2012-10-09T20:26:54 < emeb> Oh look - Castle Urqhardt.
2012-10-09T20:27:27 < emeb> Still nothing on the Youtube stream. Looks like the envelope hasn't been inflated yet.
2012-10-09T20:27:35 <+Steffanx> Nothing?
2012-10-09T20:27:43 <+Steffanx> I see the capsule + some people
2012-10-09T20:28:01 <+Steffanx> Broad cast will start within ~10 minutes
2012-10-09T20:28:29 < emeb> this stream: http://www.youtube.com/watch?feature=player_embedded&v=vkJ5ItzEq3M
2012-10-09T20:28:42 <+Steffanx> Yes, that one
2012-10-09T20:28:57 <+Steffanx> That's the same one as here http://www.redbullstratos.com/live/ except this one has more info
2012-10-09T20:31:50 < zyp> hmm, weird, I have SDIO peripheral turning on and off clock when it's supposed to write commands, but it's not driving command pin
2012-10-09T20:32:09 < zyp> i.e. it's driving clock properly, but not command
2012-10-09T20:32:17 <+Steffanx> Countdown started: 4 min remaining
2012-10-09T20:32:47 < emeb> oh - spacey countdown.
2012-10-09T20:33:03 <+Steffanx> No, it's not T minus crap
2012-10-09T20:33:06 < emeb> wonder if that's a live shot. :P
2012-10-09T20:33:18 <+Steffanx> It's not. It's delayed by 20 seconds :P
2012-10-09T20:33:30 <+Steffanx> So we won't see the big disaster ( if there is one )
2012-10-09T20:33:51 < karlp> what's the point in live if you don't get to see things as they happen
2012-10-09T20:33:56 < emeb> in case something awful happens. Like that guy who shot himself on Fox last week. (That was < 20 miles from me)
2012-10-09T20:34:02 < karlp> yeah yeah
2012-10-09T20:34:11 < karlp> still, that's the whole damn point of live
2012-10-09T20:34:36 <+Steffanx> emeb, like they don't show the dead bodies in the news when something bad happens in iraq etc. ?
2012-10-09T20:34:46 <+Steffanx> Here they do, without a warning
2012-10-09T20:34:57 < emeb> Steffanx: right. It's OK to see the aftermath, just not the event.
2012-10-09T20:35:11 < emeb> hypocrites.
2012-10-09T20:35:11 <+Steffanx> Yes people are weird
2012-10-09T20:35:59 < emeb> 20
2012-10-09T20:36:05 < zyp> 10
2012-10-09T20:36:09 <+Steffanx> 20 :P
2012-10-09T20:36:14 < zyp> 0
2012-10-09T20:36:20 < emeb> lets hear it for sync
2012-10-09T20:36:25 < zyp> :p
2012-10-09T20:36:46 < emeb> seemed like about a +/-10 second variance.
2012-10-09T20:36:49 <+Steffanx> I don't want to see a stupid inotr, come on
2012-10-09T20:36:52 <+Steffanx> *intro
2012-10-09T20:37:09 < emeb> Roswell. Hah.
2012-10-09T20:37:27 <+Steffanx> Hmmpfs, what;s new?
2012-10-09T20:37:35 < zyp> it's inflated
2012-10-09T20:37:51 < zyp> or in the process of being
2012-10-09T20:38:07 < emeb> He goes back in time to 1947 - that explains everything.
2012-10-09T20:38:28 <+Steffanx> That guy must be nervous
2012-10-09T20:38:46 < emeb> Maybe. He's done enough practice runs from lower altitude to know what's coming.
2012-10-09T20:40:46 <+Steffanx> I wonder if Redbull is going to sponsor me too when i want to do something stupid
2012-10-09T20:41:12 < emeb> Send 'em a proposal. I'm sure they're not getting thousands of them as we speak.
2012-10-09T20:41:26 <+Steffanx> I don't have anthing stupid in mind yet
2012-10-09T20:42:08 < emeb> s/stupid/stupid enough/
2012-10-09T20:42:21 <+Steffanx> Abort :(
2012-10-09T20:42:48 < zyp> no wonder in that wind
2012-10-09T20:42:50 < zyp> :p
2012-10-09T20:43:02 < emeb> Felix Baumgartner's blustery day.
2012-10-09T20:43:20 < emeb> Should have done it here in AZ - dead calm.
2012-10-09T20:44:36 < emeb> The problem with continual delays & aborts is the audience has ADD and will stop caring.
2012-10-09T20:47:03 < karlp> it's a scam anyway, he's only what, targetting 3km higher than that usaf dude did in the 50s or something?
2012-10-09T20:47:09 < karlp> who the fuck cares?
2012-10-09T20:47:27 < karlp> woo! I beat your record by 0.04%!
2012-10-09T20:47:27 < Thorn> my take on Black Magic Probe: http://i.imgur.com/YjMix.png
2012-10-09T20:47:30 <+Steffanx> "party pooper"
2012-10-09T20:47:40 < Thorn> what to add / change?
2012-10-09T20:47:49 < karlp> Thorn: is that your own board to run the bmp firmware?
2012-10-09T20:47:55 < Thorn> yes
2012-10-09T20:47:59 < Thorn> future board
2012-10-09T20:48:05 <+Steffanx> F102 ?
2012-10-09T20:48:14 < Thorn> same as original
2012-10-09T20:48:26 < karlp> cheapest with usb Steffanx :)
2012-10-09T20:48:36 < Thorn> (and I happen to have 2 of them already)
2012-10-09T20:48:50 <+Steffanx> Not as the original i have :)
2012-10-09T20:48:57 <+Steffanx> Mine has 103
2012-10-09T20:49:06 <+Steffanx> *a
2012-10-09T20:49:17 <+Steffanx> Not like that really matters
2012-10-09T20:49:45 < zyp> so do mine
2012-10-09T20:49:46 < Thorn> probably first hardware version. this one uses BMP mini
2012-10-09T20:49:56 < zyp> and I have first version.
2012-10-09T20:51:29  * emeb hasn't used 102/103 - forgot that the USB needed external 1.5k PU.
2012-10-09T20:51:34 < Thorn> I only added 2 sockets, a target reset button, usb esd protection and debug connectors for the device itself
2012-10-09T20:51:54 < emeb> yeah - noticed the USB ESD. Nice.
2012-10-09T20:52:09 < zyp> debug connectors for the device itself seems sort of pointless, firmware seems pretty mature
2012-10-09T20:52:35 < Thorn> what if I want to hack it
2012-10-09T20:52:41 < emeb> belt + braces?
2012-10-09T20:53:02 < emeb> Write your own firmware. BMP++
2012-10-09T20:53:21 < Thorn> it doesn't support a whole lot of mcu families, I wanted to add some and SWV too
2012-10-09T20:53:30 < zyp> I've been thinking about doing one myself, I'd use a female connector for the 20-pin one and mount it at one edge, so I can plug it directly into boards
2012-10-09T20:53:38 < Thorn> I thought about a separate Qt app to display SWV data
2012-10-09T20:53:47 < emeb> zyp: nice idea.
2012-10-09T20:54:09 <+Steffanx> Can't you use Qt and Python as plugin for GDB?
2012-10-09T20:54:16 <+Steffanx> There is already some swv stuff in it
2012-10-09T20:54:41 < Thorn> a GUI plugin for a console program?
2012-10-09T20:54:44 <+Steffanx> *Qt binding for python
2012-10-09T20:54:46 <+Steffanx> Yeah, why not :P
2012-10-09T20:54:57 < Thorn> I'd never think of that.
2012-10-09T20:55:49 < zyp> I'm thinking about female 20-pin, male 0.05" 10-pin and also the discovery-ish pinout with pins clearly marked in silk for flying lead hookups
2012-10-09T20:56:06 < Thorn> zyp: you've been using BMP, right? any input on desired hardware changes for the future?
2012-10-09T20:56:18 < Thorn> or something
2012-10-09T20:56:20 < zyp> see what I just mentioned
2012-10-09T20:56:26 < zyp> mostly connector changes
2012-10-09T20:56:49 < Thorn> I'll add discovery-style connector, seems like a good idea
2012-10-09T20:56:54 <+Steffanx> 100% oxygen.. that guy must have been high
2012-10-09T20:57:08  * emeb <3 his BMPM - only problem is that it's so tiny it seems fragile.
2012-10-09T20:57:09 < Thorn> (and make device debug one same pinout)
2012-10-09T20:57:12 < karlp> or stupid, didn't learn anything fromapollo 9?
2012-10-09T20:57:15 < zyp> it's a bit unnecessary if you already have male 0.1" pins
2012-10-09T20:57:52 < Thorn> do female pcb mount edge IDC connectors even exist?
2012-10-09T20:58:05 < zyp> uh
2012-10-09T20:58:20 < zyp> just take regular pth ones and solder them at the edge?
2012-10-09T20:58:28 < emeb> I've done that
2012-10-09T20:58:30 < emeb> works fine.
2012-10-09T20:58:37 < zyp> 1.6mm pcb fits nicely between 0.1" pitch pins
2012-10-09T20:58:40 < emeb> just need a footprint w/ pads on both sides.
2012-10-09T20:59:23 < zyp> http://bin.jvnv.net/f/sTW8U.JPG <- like I did here, just female
2012-10-09T20:59:58 < Thorn> why does it have all this vbus detect circuitry?
2012-10-09T21:02:40  * emeb is somewhat annoyed by the T_PWR stuff on BMPM
2012-10-09T21:02:55 < Thorn> yes and that too
2012-10-09T21:03:05 < emeb> it tries to power my board but fails - board sits there slowly cycling on/off.
2012-10-09T21:04:18 < Thorn> btw, I didn't connect 3.3V in the debug connector on my last f103 board. is it important for BMP? does it detect target by sensing its supply voltage?
2012-10-09T21:04:27 <+Steffanx> Remove the jumper emeb ?
2012-10-09T21:04:34 <+Steffanx> Or doesn't have the mini that one?
2012-10-09T21:04:43 <+Steffanx> *the mini have
2012-10-09T21:04:54 < emeb> Steffanx: correct - mini doesn't have the jumper.
2012-10-09T21:04:55 < Thorn> no jumper on the schematics
2012-10-09T21:05:05 < zyp> Thorn, old BMP has voltage translators which require device power to work
2012-10-09T21:05:34 <+Steffanx> It also has a jumper so the BMP can power the target ..
2012-10-09T21:05:39 < Thorn> this one still has a target power detection circuit
2012-10-09T21:05:46 < zyp> Steffanx, or just device side of translator
2012-10-09T21:06:00 < Thorn> and a mosfet most likely to power the target
2012-10-09T21:06:10 < Thorn> how is it controlled btw?
2012-10-09T21:06:20 < Thorn> a 'monitor' command or something?
2012-10-09T21:06:29 < zyp> probably
2012-10-09T21:06:33 < zyp> try monitor help
2012-10-09T21:06:54 < Thorn> will try once I get the boards lol
2012-10-09T21:07:44 -!- TitanMKD [Titan@88.183.104.56] has joined ##stm32
2012-10-09T21:07:57 < TitanMKD> hello
2012-10-09T21:08:06 <+Steffanx> Lo
2012-10-09T21:15:06 < emeb> monitor help -> http://pastebin.com/k34ExX0b
2012-10-09T21:16:23 < emeb> nothing in there about target power.
2012-10-09T21:23:18 < zyp> oh, I'm dumb.
2012-10-09T21:23:39 < emeb> ?
2012-10-09T21:24:04 < zyp> 19:31:50 < zyp> hmm, weird, I have SDIO peripheral turning on and off clock when it's supposed to write commands, but it's not driving command pin
2012-10-09T21:24:16 < zyp> CMD pin is PD2, I forgot to enable GPIOD
2012-10-09T21:24:40 < emeb> simple fix
2012-10-09T21:24:52 < zyp> clock is PC12, so that's of course fine
2012-10-09T21:27:58 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-10-09T21:30:26 < qyx_> leave the rx/tx center taps unconnected :S
2012-10-09T21:30:31 < qyx_> argh that micrel thing
2012-10-09T21:40:43 < emeb> ether?
2012-10-09T21:49:35 < qyx_> yep, ksz8051mll
2012-10-09T21:49:50 < qyx_> i can't find jack with integrated magnetics for poe
2012-10-09T21:50:08 < qyx_> with CT which are not connected together
2012-10-09T21:50:56 < emeb> Odd - poe is not _that_ uncommon.
2012-10-09T21:51:42 < qyx_> 8051 is uncommon in that it has rx and tx common voltages different
2012-10-09T21:53:59 < emeb> but that's on the local side windings of the pulse transfomers. The line side is where the POE is tapped.
2012-10-09T21:54:16 < emeb> so you need magnetics with separate commons on both sides, and that's hard to find?
2012-10-09T21:54:21 < Thorn> what does the pushbutton on BMP do?
2012-10-09T21:54:44 < emeb> Thorn: seems to cause the morse-code LED to cycle a msg.
2012-10-09T21:54:49 < qyx_> yes, but for some reason most of them have also phy side connected together
2012-10-09T21:54:56 < qyx_> maybe to save some pins
2012-10-09T21:55:01 < Thorn> wat
2012-10-09T21:55:04 < emeb> probably.
2012-10-09T21:55:11 < qyx_> http://www.digikey.com/product-detail/en/1-6605310-1/A99644-ND/1893235
2012-10-09T21:55:15 < zyp> Thorn, it can be used to force it into bootloader mode
2012-10-09T21:55:16 < qyx_> this one is good
2012-10-09T21:55:20 < qyx_> except the price
2012-10-09T21:55:33 < emeb> zyp: Thorn: oh yeah.
2012-10-09T21:55:49 < emeb> hold button & plug in to load new code
2012-10-09T21:55:54 < emeb> (I did that once)
2012-10-09T21:56:07 < Thorn> it's not on a boot pin
2012-10-09T21:56:12 < emeb> no
2012-10-09T21:56:17 < Thorn> does it enable DFU?
2012-10-09T21:56:25 < emeb> this isn't the ROM bootloader.
2012-10-09T21:56:49 < emeb> It's a user flash bootloader provided by the firmware.
2012-10-09T21:57:24 < emeb> gsmcmullin's source has a python script to talk to it.
2012-10-09T21:57:29 < zyp> normally it's not needed, as long as application code works it'll change to DFU mode automatically
2012-10-09T21:57:57 < zyp> but if you flash a broken firmware, you can still force it into DFU mode with the button
2012-10-09T21:58:59 < emeb> qyx_: yeah - that's kinda steep.
2012-10-09T21:59:12 < LeelooMinai> Are there some "traps" with I2C on F0?  I think I configured everything and I see correct values in hardware registers, but writing the START bit does nothing (it sits at 0)
2012-10-09T21:59:20 < Thorn> I see thanks.
2012-10-09T21:59:43 < zyp> LeelooMinai, did you set the enable bit? :p
2012-10-09T21:59:58 < LeelooMinai> Yes, PE ia 1
2012-10-09T22:00:19 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-09T22:00:22 < LeelooMinai> I have scope on SDA and nothing happens there
2012-10-09T22:00:46 < Thorn> SDA sits at 0?
2012-10-09T22:00:59 < qyx_> it should be at H
2012-10-09T22:01:01 < Thorn> it's 1 when idle
2012-10-09T22:01:08 < Thorn> 0 = busy bus
2012-10-09T22:01:32 < zyp> LeelooMinai, have you configured both pins correctly?
2012-10-09T22:01:35 < LeelooMinai> Hmm, but BUSY flag is 0
2012-10-09T22:01:49 < zyp> AF, correct AF number, opendrain?
2012-10-09T22:01:56 < LeelooMinai> zyp, Yes, as open drain, and I have pull-up resistors on sda and scl
2012-10-09T22:01:58 < zyp> do you have external pullups on the bus?
2012-10-09T22:02:07 < LeelooMinai> AS AF_1
2012-10-09T22:02:35 < LeelooMinai> Te strange thing is I write to START bit and step on it in the debugger and it is 0
2012-10-09T22:02:58 < zyp> did you enable the clock for the I2C peripheral?
2012-10-09T22:03:12 < LeelooMinai> Yes
2012-10-09T22:03:41 < upgrdman> enable clock for gpio channel?
2012-10-09T22:04:00 < LeelooMinai> Yes, port b is enabled
2012-10-09T22:04:20 < Thorn> pin remapping?
2012-10-09T22:04:48 < LeelooMinai> I have this: GPIO_PinAFConfig(i2c_GPIO_Port, i2c == I2C1 ? i2c1_GPIO_SCL_PinSource : i2c2_GPIO_SCL_PinSource,       GPIO_AF_1 );
2012-10-09T22:04:59 < LeelooMinai> ^ if that is "pin remapping"
2012-10-09T22:05:42 < zyp> how are you configuring the I2C peripheral?
2012-10-09T22:06:23 < Thorn> and what is the bus doing, is sda 0 or 1 after all
2012-10-09T22:06:54 < zyp> by the way, it's normal that you wouldn't read the START bit as 1 after writing it to 1
2012-10-09T22:07:30 < zyp> when you write it to 1, it causes the hardware to send a start condition, and then clear it again, and that probably happens faster than you can read it back
2012-10-09T22:09:05 < zyp> but when start is sent, you should get SB in SR1 as 1
2012-10-09T22:09:06 < Thorn> should I add ESD protection for debug connectors?
2012-10-09T22:09:34 -!- Tectu is now known as tectu
2012-10-09T22:09:36 < zyp> if you are worried about stuff like that
2012-10-09T22:10:15 < LeelooMinai> O, btw, sda is high  - I had AC coupling on, sorry
2012-10-09T22:10:15 < LeelooMinai> Duh...
2012-10-09T22:10:16 < LeelooMinai> Damn, I spent 2 hours poking in the code, and my scope was on AC - no wonder it did not trigger
2012-10-09T22:10:16 < LeelooMinai> Ok, I see the I2C "taking", thx:)
2012-10-09T22:11:07 -!- tavish_ [~tavish@59.180.220.142] has joined ##stm32
2012-10-09T22:12:25 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-09T22:12:46 < DLPeterson> is anybody here comfortable with running a program from ram using OpenOCD, and know what needs to change in the linker script and/or build process to make it work?
2012-10-09T22:13:17 < DLPeterson> Ideally without having to change physical boot pin jumpers.
2012-10-09T22:13:47 < zyp> I'm not very comfortable with openocd, but I don't think that part would matter :)
2012-10-09T22:14:00 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-10-09T22:14:12 < TitanMKD> DLPeterson you can just change 1 things with same mapping
2012-10-09T22:14:24 < zyp> anyway, there is basically two things you'll need to change; linker script and startup code
2012-10-09T22:14:33 < Thorn> why would boot pins matter
2012-10-09T22:14:43 < TitanMKD> DLPeterson by debugger change the M4MAP to RAM and load the code ;)
2012-10-09T22:15:02 -!- tavish_ [~tavish@59.180.220.142] has quit [Client Quit]
2012-10-09T22:15:04 < zyp> in the linker script you just need to place all segments into ram
2012-10-09T22:15:14 < TitanMKD> DLPeterson but you will need the linker script i have developed for hackrf and adapt it to STM32F4
2012-10-09T22:15:20 < zyp> and then disable the part of the startup code that copies .data from flash to ram
2012-10-09T22:15:21 < DLPeterson> zyp, do you have an example linker script?
2012-10-09T22:16:03 < DLPeterson> TitanMKD, by M4MAP do you mean SYSCFG_MEMRMP?
2012-10-09T22:16:09 < zyp> DLPeterson, http://cgit.jvnv.net/laks/tree/ld_scripts/arm_flash_ram.ld <- this is my normal one
2012-10-09T22:16:27 < zyp> to adapt it to run from ram I'd just have to change all > flash lines to read > ram
2012-10-09T22:16:27 < TitanMKD> DLPeterson maybe i do not remember the name on STM32F4 but it is similar
2012-10-09T22:17:10 < TitanMKD> zyp it is crappy it can be done without changing any mapping ;) and load always at addr 0 and remap it to flash or ram using the MEM Remap feature
2012-10-09T22:17:37 < TitanMKD> it is what the boot loader pin exec from Flash or SRAM do in fact
2012-10-09T22:18:32 < zyp> TitanMKD, it's not that simple
2012-10-09T22:18:42 < TitanMKD> zyp i have do the trick on LPC43xx
2012-10-09T22:18:53 < TitanMKD> yes it was not so easy but it work now ;)
2012-10-09T22:19:33 < TitanMKD> the linker script in fact load code from ROM to RAM and execute it from RAM or start code from ROM directly
2012-10-09T22:20:14 -!- BrainDamage [~BrainDama@62.10.5.141] has quit [Ping timeout: 240 seconds]
2012-10-09T22:21:31 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has joined ##stm32
2012-10-09T22:21:32 < DLPeterson> TitanMKD, on the STM32F4 SYSCFG_MEMRMP has 2 bits which are derived from sampling the BO and B1 pins.  They are software writable, is that what the M4MAP does?
2012-10-09T22:22:04 < DLPeterson> TitanMKD, do you have example code that shows what you mean?
2012-10-09T22:23:12 < TitanMKD> you can check here
2012-10-09T22:23:12 < TitanMKD> https://github.com/TitanMKD/libopencm3/tree/master/lib/lpc43xx
2012-10-09T22:23:20 < TitanMKD> .ld files
2012-10-09T22:23:43 < TitanMKD> the advantage is it copy code from ROM to RAM and execute from RAM so 0 wait states
2012-10-09T22:24:07 < TitanMKD> for your case it is even simpler as maybe you just want to load it with debugger so just load in RAM by debugger and execute in RAM
2012-10-09T22:25:04 < zyp> TitanMKD, code running from flash does still have some segments located in ram, which would normally have a different layout from when you also place .text in ram
2012-10-09T22:25:20 < zyp> (it would be terribly wasteful and dumb otherwise)
2012-10-09T22:25:33 < TitanMKD> zyp no the hint is they have same addr for ram or flash
2012-10-09T22:25:59 < zyp> that's what I'm saying they don't have.
2012-10-09T22:26:10 < TitanMKD> zyp the hack is done just on 2 linker exported vars ;)
2012-10-09T22:26:47 < TitanMKD> zyp you can just compare the 2 files from https://github.com/TitanMKD/libopencm3/blob/master/lib/lpc43xx/
2012-10-09T22:27:01 < TitanMKD> libopencm3_lpc43xx.ld vs libopencm3_lpc43xx_rom_to_ram.ld
2012-10-09T22:27:27 < TitanMKD> you will understand map will be the same except _etext_ram & _etext_rom
2012-10-09T22:28:11 < TitanMKD> and the hint is in the vector code
2012-10-09T22:28:12 < TitanMKD> https://github.com/TitanMKD/libopencm3/blob/master/lib/lpc43xx/vector.c
2012-10-09T22:28:26 < TitanMKD> if( (&_etext_ram-&_text_ram) > 0 ) copy code from ROM to RAM ;)
2012-10-09T22:28:35 < TitanMKD> and change before the copy CREG_M4MEMMAP = (unsigned long)src;
2012-10-09T22:28:36 < tectu> why isn't this working?
2012-10-09T22:28:37 < tectu> (gdb) break lcd_lld.c:75
2012-10-09T22:28:39 < tectu> No line 75 in file "src/lcd_lld.c".
2012-10-09T22:28:51 < tectu> there is an src/lcd_lld.c and there is a line number 75
2012-10-09T22:29:16 < TitanMKD> zyp and at end of copy CREG_M4MEMMAP = (unsigned long)&_text_ram; to Change Shadow memory to Real RAM
2012-10-09T22:29:47 < zyp> TitanMKD, wait a moment, what are you actually claiming?
2012-10-09T22:29:56 < TitanMKD> zyp so by changing just 2 constants you can change code executed from ROM or from RAM without changing any addr of code or data
2012-10-09T22:31:42 < TitanMKD> zyp it is just a hint which could also work on other MCU with the MEM REMAP feature
2012-10-09T22:32:38 < zyp> TitanMKD, can you give me an objdump -h output of the elf files produced with each of the linker scripts?
2012-10-09T22:32:53 < TitanMKD> zyp of course
2012-10-09T22:33:27 < TitanMKD> example with minblink ;)
2012-10-09T22:33:37 < zyp> and why do you have two? I thought your point was that you didn't need any linker script changes
2012-10-09T22:35:09 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-09T22:35:39 < TitanMKD> in this example in fact i moved a bit data in an other data section
2012-10-09T22:36:07 < TitanMKD> but anyway it can be exactly the same except for _etext_ram & _etext_rom
2012-10-09T22:36:49 < TitanMKD> zyp in fact there's 2 linker but final elf code addr is same on both
2012-10-09T22:37:55 < zyp> TitanMKD, code addr as in VMA for .text? that part is fine
2012-10-09T22:38:13 < zyp> I'm just claiming that using same VMA for .data and .bss in both cases would be stupid.
2012-10-09T22:38:35 < TitanMKD> http://pastie.org/5025891 => miniblink.map
2012-10-09T22:38:44 < TitanMKD> execute from ROM
2012-10-09T22:40:00 < TitanMKD> http://pastie.org/5025901 => miniblink_rom_to_ram.map copy code from ROM to RAM and execute in RAM
2012-10-09T22:40:10 < zyp> I'm not interested in the .map, I'm interested in the section listing from objdump -h, with VMA and LMA for each section
2012-10-09T22:40:19 < TitanMKD> ha ok ;)
2012-10-09T22:41:43 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-09T22:43:18 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-09T22:43:29 < TitanMKD> zyp see the both files content
2012-10-09T22:43:30 < TitanMKD> http://pastie.org/5025929
2012-10-09T22:44:24 < zyp> heh, no data?
2012-10-09T22:44:35 < TitanMKD> yes in fact there's only bss in this example
2012-10-09T22:44:43 < TitanMKD> it is just a basic led blink test
2012-10-09T22:44:56 < zyp> well, VMA of .bss is different
2012-10-09T22:45:01 < TitanMKD> of course
2012-10-09T22:45:22 < TitanMKD> i tell you i moved it to 10080000 on ROM to RAM version
2012-10-09T22:45:37 < TitanMKD> but i can do the same for the ROM only
2012-10-09T22:45:55 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-09T22:46:30 < TitanMKD> zyp the important stuff is base code from ROM or RAM is equal 0x0
2012-10-09T22:46:47 < zyp> and when you do, you're wasting all the ram you are saving by not having .text there, that's been my point all along :p
2012-10-09T22:47:18 < TitanMKD> anyway there's 128KB for code ;)
2012-10-09T22:47:28 < TitanMKD> in Thumb2 it is a lot of code
2012-10-09T22:47:42 < TitanMKD> and there's 2*64KB other section for data ;)
2012-10-09T22:47:49 < TitanMKD> total 256KB
2012-10-09T22:48:11 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-09T22:52:42 < DLPeterson> TitanMKD, zyp, so if I understand correctly, the only difference in the output binary is VMA for the .bss section?
2012-10-09T22:52:49 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-10-09T22:53:03 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-09T22:53:16 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has left ##stm32 []
2012-10-09T22:55:32 < TitanMKD> but in fact i will fix that to have absolutely no difference ;)
2012-10-09T23:04:17 < TitanMKD> ok fixed now they are exactly the same except in code => 3*32 are modified between 2 version
2012-10-09T23:04:50 < tectu> four floats on CM3 are pretty much 16 bytes, right?
2012-10-09T23:05:11 < zyp> right
2012-10-09T23:05:44 < tectu> it works \o/
2012-10-09T23:07:49 < zyp> so does my SD card
2012-10-09T23:08:07 < zyp> but I'm too tired to write a driver tonight
2012-10-09T23:10:10 < tectu> zyp, go to bed :)
2012-10-09T23:12:34 < emeb> zyp: that's good news.
2012-10-09T23:19:08 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ldsegujyfdswtxyz] has quit [Ping timeout: 256 seconds]
2012-10-09T23:19:15 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-yurcjvhaaoafkjok] has joined ##stm32
2012-10-09T23:28:55 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-09T23:31:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-09T23:37:23 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-09T23:37:24 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-09T23:37:24 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-09T23:45:52 < cjbaird> Looking like Stellaris non-blob linux programming isn't there yet..
2012-10-09T23:56:16 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Ping timeout: 245 seconds]
2012-10-09T23:58:00 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
--- Day changed Wed Oct 10 2012
2012-10-10T00:03:08 -!- tectu is now known as Tectu
2012-10-10T00:09:23 -!- DaKu is now known as daku
2012-10-10T00:15:03 -!- kerwood_wk [80f40908@gateway/web/freenode/ip.128.244.9.8] has quit [Ping timeout: 245 seconds]
2012-10-10T00:20:54 < TitanMKD> bye
2012-10-10T00:21:01 -!- TitanMKD [Titan@88.183.104.56] has quit []
2012-10-10T00:22:24 < zyp> I'm intending to just hook up my BMP to the stellaris launchpad when I get it instead of dicking around with the onboard programmer
2012-10-10T00:22:42 < zyp> I already do it with the discovery board since I'm tired of dicking around with stlink
2012-10-10T00:28:37 < emeb> Haven't reflashed your discovery w/ BMP firmware?
2012-10-10T00:31:11 <+dekar> emeb, there is a BMP firmware for that?
2012-10-10T00:32:08 < emeb> dekar: apparently so - a build-time flag for the BMP firmware lets you target it to the stlink hardware on the discovery boards.
2012-10-10T00:34:21 <+Steffanx> Yes there is
2012-10-10T00:34:40 <+dekar> nice
2012-10-10T00:34:54 <+Steffanx> dekar just use your versaloon to get the fw on the stlink  :)
2012-10-10T00:35:06 <+dekar> I already have a BMP anyway ;)
2012-10-10T00:35:10 <+Steffanx> Ah
2012-10-10T00:35:19 <+Steffanx> And a versaloon and everything
2012-10-10T00:35:36 <+Steffanx> It's probably easy to port the bmp fw to the versaloon hw
2012-10-10T00:35:38 <+dekar> my versaloon setup was working great, then I updated the firmware and everything went SNAFU
2012-10-10T00:36:03 <+dekar> the openOCD I had wasn't compatible anymore
2012-10-10T00:36:32 <+dekar> and the latest openOCD also wasn't (since that versaloon guy broke compatibility on purpose :/)
2012-10-10T00:36:42 <+Steffanx> huh :S
2012-10-10T00:36:44 <+dekar> so I was supposed to use weird patches
2012-10-10T00:36:57 <+dekar> at that point I ordered the BMP
2012-10-10T00:37:10 <+Steffanx> The min or original one?
2012-10-10T00:37:13 <+Steffanx> *mino
2012-10-10T00:37:16 <+Steffanx> *mini lol
2012-10-10T00:37:50 <+dekar> mini
2012-10-10T00:39:23 <+dekar> I guess I could have used my f4 discovery instead ;)
2012-10-10T00:39:53 -!- barthess [~barthess@178.154.21.145] has quit [Quit: Leaving.]
2012-10-10T00:41:43 <+Steffanx> Your boss paid :P
2012-10-10T00:42:18 <+dekar> ya
2012-10-10T00:42:28 < zyp> emeb, I've been thinking about it, but it's always faster to just flywire the BMP to the discovery board :p
2012-10-10T00:44:29 < emeb> Yup.
2012-10-10T00:49:23 < zyp> and I haven't had to do it enough times to be annoyed by it yet :p
2012-10-10T01:13:58 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-10T01:31:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-10T01:37:08 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-10-10T01:38:15 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-10T01:41:28 -!- BjoernC_ [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-10T01:44:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 264 seconds]
2012-10-10T01:44:38 -!- BjoernC_ is now known as BjoernC
2012-10-10T01:49:59 < karlp> damn, it's been so long since I got some time to work on my own projects I can't even remember what was working and what wasn't
2012-10-10T01:50:03 < karlp> let alone where I put the code
2012-10-10T01:51:31 < zyp> assume nothing was working
2012-10-10T01:51:40 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-10T01:51:50 < zyp> then if you were wrong, you're in for a pleasant surprise
2012-10-10T01:52:06 < karlp> yeah, that was pretty much where I was at.
2012-10-10T01:52:13 < karlp> started cleaning up some stuff
2012-10-10T01:52:21 < karlp> god it's been a while for my own toys
2012-10-10T01:52:55 < zyp> I'm trying to commit stuff more often now
2012-10-10T01:55:09 < zyp> especially when stuff is working
2012-10-10T01:55:49 < Rickta59> what is BMP? I see mention of the stellarpad
2012-10-10T01:55:51 < zyp> it makes it so much easier to backtrack when I break something again when I clean it up later :p
2012-10-10T01:55:58 < zyp> black magic probe
2012-10-10T01:56:05 < Rickta59> hardware?
2012-10-10T01:56:13 < zyp> jtag/swd dongle
2012-10-10T01:56:20 < Rickta59> how much?
2012-10-10T01:56:35 < zyp> google it
2012-10-10T01:56:49 < Rickta59> thought you might know ..
2012-10-10T01:57:03 < Rickta59> i've been using my stellarpad with CCS on linux
2012-10-10T01:57:11 < Rickta59> but i'd rather use it with openocd
2012-10-10T01:57:38 < Rickta59> but I'm not going to spend any money ... the thing only cost $5
2012-10-10T01:58:48 < Rickta59> yeah .. $75 .. not happening
2012-10-10T02:01:22 < karlp> zyp: yeah, it was never workign in my memory though
2012-10-10T02:02:00 < zyp> Rickta59, not my problem, I already got one so I'm going to use it.
2012-10-10T02:02:00 < karlp> stellarpad?
2012-10-10T02:02:08 < zyp> stellaris launchpad
2012-10-10T02:02:14 < karlp> yeah, just not a name I'd heard :)
2012-10-10T02:02:21 < karlp> I was kinda guessing it must be the stellaris launchpad
2012-10-10T02:02:57 < karlp> must be time for texane/icdi?  lots of bugs but maybe might one day perhaps works a little bit? ;)
2012-10-10T02:03:29 < zyp> :p
2012-10-10T02:04:52 < Rickta59> staunchpad spad stellarpad spad spudnik
2012-10-10T02:05:01 < zyp> huh, http://processors.wiki.ti.com/index.php/Stellaris-ICDI_Programming
2012-10-10T02:05:11 < karlp> heh, didnt' even have the right pin definitions in my config file
2012-10-10T02:05:19 < karlp> no point dragging high a pin that the device isn' tlooking at.
2012-10-10T02:07:18 < Rickta59> that looks like it is for stellaris boards that have an ftdi chip .. the stellarpad has 2 identical chips .. one is used as the icdi
2012-10-10T02:07:30 < zyp> yeah
2012-10-10T02:07:45 < zyp> wonder if it's firmware locked
2012-10-10T02:08:16 < zyp> otherwise porting some other firmware to it shouldn't be that hard
2012-10-10T02:08:22 < Rickta59> there seems to be a small connection specifically for theh
2012-10-10T02:08:25 < Rickta59>  chip
2012-10-10T02:08:57 < Rickta59> http://www.tag-connect.com/Tag-Connect-TC2050IDC_NL_430 .. something with a footprint that could be used by a cable like this
2012-10-10T02:09:14 < karlp> BMP alllll the chips!
2012-10-10T02:11:20 < karlp> bleh, now I've lost my conection to th eboard.
2012-10-10T02:11:24 < karlp> I guess it's in reset somewhere
2012-10-10T02:12:46 < karlp> there we go, drag nrst low manually for a bit and we're back in action
2012-10-10T02:20:19 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-10T02:21:22 < karlp> heh, that was easy
2012-10-10T02:21:30 < karlp> actually _read_ the datasheet mofo
2012-10-10T02:40:03 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has joined ##stm32
2012-10-10T02:40:04 -!- R0b0t1` [~Dev@64-136-219-55.dyn.everestkc.net] has quit [Changing host]
2012-10-10T02:40:04 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has joined ##stm32
2012-10-10T02:47:46 < dongs> Red Bull Stratos - Today's mission aborted due to gusty winds
2012-10-10T02:47:48 < dongs> fucking useless
2012-10-10T02:51:27 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Read error: Connection reset by peer]
2012-10-10T02:53:36 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-10T03:10:26 -!- R0b0t1` [~Dev@unaffiliated/r0b0t1] has quit [Read error: Connection reset by peer]
2012-10-10T03:12:03 < cjbaird> Sugar-water SCIENCE yyyyeeeeeaaahhhhhhhhhh...no.
2012-10-10T03:15:23 < karlp> damnit,
2012-10-10T03:15:34 < karlp> dodgy soldering on this milled pcb has fallen apart
2012-10-10T03:17:12 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-10T03:18:35 < emeb> cjbaird: lulz
2012-10-10T04:37:43 < dongs> ok where the fuck are the chats
2012-10-10T04:41:50 < cjbaird> Is my face red?
2012-10-10T04:43:56 < esden> dongs: bored? :)
2012-10-10T04:45:43 < emeb> who let the chats out.
2012-10-10T04:47:48 < cjbaird> I saw the movie "Looper" last night-- "I'm from the future. Go to China." :)
2012-10-10T04:48:18 < cjbaird> It's set about 20 years into the future. America has gone to shit.
2012-10-10T04:49:01 < dongs> soundsa bout right
2012-10-10T04:49:59 < emeb> don't have to 20+ years for that.
2012-10-10T04:50:38 < emeb> ^wait
2012-10-10T04:50:59 < esden> cjbaird: that sounds like a reasonable assessment doesn't it? :)
2012-10-10T04:52:03 < emeb> note: Americans don't mind saying we've gone to hell, but we hate it when furriners say it. :)
2012-10-10T04:52:31 < esden> emeb: what about foreigners that live in the US?
2012-10-10T04:52:49 < emeb> esden: That's worst of all! :D
2012-10-10T04:52:50 < cjbaird> It was quite funny to see all the cars in future-USA are junkers that've had engine conversions done to them.
2012-10-10T04:52:57 < esden> emeb: damn!
2012-10-10T04:53:31 < emeb> visitors are supposed to argue with us and tell us that things are actually OK.
2012-10-10T04:53:48 < emeb> (because it's worse where they're from, etc)
2012-10-10T04:54:16 < cjbaird> "We'll have to live like it's the 1970s! baaaawwwwww"
2012-10-10T04:54:18 < dongs> looper sounds good, is that out on bd yet
2012-10-10T04:55:04 < emeb> cjbaird: but we want a new smartphone every 6mo.
2012-10-10T04:55:26 < dongs> jesus christ. Japan / 12 January 2013
2012-10-10T04:55:27 < cjbaird> "No! Not wood panelling..."
2012-10-10T04:55:33 < dongs> shits gonna be out on BD by the time the movie arrives to jp
2012-10-10T04:55:34 < dongs> fucking trash.
2012-10-10T04:55:45 < emeb> cjbaird: and orange shag carpeting!
2012-10-10T04:55:48 < esden> emeb: well it is similar from where I am from, also everything going down :D
2012-10-10T04:55:54 < esden> so does not really matter
2012-10-10T04:56:21 < BrainDamage> I am not going on anyone's movie
2012-10-10T04:56:24 < emeb> esden: we'll all go down together.
2012-10-10T04:56:29 < esden> beside the fact that some politicians here are even worse at hiding their retiredness :)
2012-10-10T04:57:37 < cjbaird> Parts of the movie are certain subtle references to a few classics SFs, like The Terminator, etc. Thankfully, the movie doesn't dumb things down to the usual blockbuster cliches.
2012-10-10T04:57:39 < esden> emeb: sounds like a slip and slide ride doesn't it? :)
2012-10-10T04:58:07 < emeb> esden: could be worse: https://www.youtube.com/watch?feature=player_embedded&v=R6MlUcmOul8
2012-10-10T04:58:47 < esden> emeb: yeah awesome short! fucking robots are coming for us! :)
2012-10-10T04:59:30 < emeb> That's what I'm thinking everytime I see one of these university research videos with swarms of ncopters flying in close formation.
2012-10-10T05:00:11 < emeb> One of these days they're going to stop playing with ping-pong balls and go for the eyes.
2012-10-10T05:00:29 < esden> yeah well … true, but university demos are to a big degree fake anyways, the only one that actually impressed me recently is the one from MIT flying a fixed wing in the garage
2012-10-10T05:00:37 < esden> the rest is cheated with vicon
2012-10-10T05:00:42 < dongs> haha
2012-10-10T05:00:44 < esden> the upen stuff
2012-10-10T05:00:51 < emeb> "Any sufficiently advanced technology is indistinguishable from a rigged demo"?
2012-10-10T05:01:21 < esden> emeb: thanks … I will put that one in my quote book! :)
2012-10-10T05:01:55 < emeb> with apologies to Arthur C. Clarke of course. :)
2012-10-10T05:02:29 < esden> emeb: of course
2012-10-10T05:03:16  * esden likes magic
2012-10-10T05:04:12 < esden> it is the amount of leverage you get … I think most people going into engineering get into that because of that… small input huge effect … just like magic :)
2012-10-10T05:05:22 < emeb> Yeah - there's a bit of "spell casting" and alchemy involved.
2012-10-10T05:11:25 < dongs> hmm i never watched sintel  either
2012-10-10T05:11:29 < dongs> i remember some adverts for it
2012-10-10T05:11:36 < emeb> sintel?
2012-10-10T05:11:44 < dongs> the 3rd shit from blenderguys
2012-10-10T05:11:59 < dongs> i think its all cg though
2012-10-10T05:12:11 < emeb> yeah - looks like no live-action actors.
2012-10-10T05:18:16 < dongs> apparently "better than new one" according to 'tube comments.
2012-10-10T05:20:39 < emeb> longer
2012-10-10T05:25:51 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-10T05:33:11 < emeb> why are they such bummers?
2012-10-10T05:33:51 < dongs> who
2012-10-10T05:34:24 < emeb> those movies.
2012-10-10T05:35:38 < dongs> haha
2012-10-10T05:35:45 < dongs> emo opensauce guys , what do you expect
2012-10-10T05:35:59 < dongs> its in their blood
2012-10-10T05:41:58 < cjbaird> Blade Runner envy
2012-10-10T05:46:24 < emeb> hmpf. BR isn't that depressing.
2012-10-10T05:47:09  * emeb dreads the day some Hollywood hack decides they need to remake BR.
2012-10-10T05:47:50 < emeb> w/ more CGI and asplosions!!
2012-10-10T05:48:07 < emeb> and a happier ending!
2012-10-10T05:50:08 < dongs> maybe they'll make han solo shoot first
2012-10-10T05:51:38 < emeb> he _did_ shoot first. GL just can't leave well enough alone.
2012-10-10T05:59:46  * emeb researches how to hand solder lqfps on an ENIG board. blech.
2012-10-10T06:01:25 < emeb> http://www.howardelectronics.com/jbc/dragsoldering.html <- looks like the ticket
2012-10-10T06:04:10 < dongs> wat? why would it be any different on enig
2012-10-10T06:05:18 < emeb> I usually get boards HASL. Those are pre-tinned so I just reflow the pins to pads.
2012-10-10T06:05:34 < emeb> ENIG has no solder on it, so I have to apply some.
2012-10-10T06:05:44 < dongs> right, but it doesnt change the overall process
2012-10-10T06:05:54 < dongs> drop a solder on one pad, hold + f ix the chip, solder it
2012-10-10T06:06:02 < dongs> then use thin solder and just do each pad at a time
2012-10-10T06:06:10 < emeb> Sure it does. With HASL I don't apply any fresh solder - just use what's already there.
2012-10-10T06:06:31 < dongs> http://www.youtube.com/watch?v=4cxoKJFAyT8
2012-10-10T06:06:35 < dongs> ^ enig
2012-10-10T06:07:16 < dongs> no bridges but its 0.65mm pitch, though I can do 0.5mm just as well without bridges usually
2012-10-10T06:07:20 < emeb> yeah, but that's not a 0.5mm pitch part.
2012-10-10T06:07:28 < dongs> not much diff
2012-10-10T06:07:36 < emeb> I suppose.
2012-10-10T06:09:05 < emeb> Oh lookie - Sparkfun has the right tip for my iron: https://www.sparkfun.com/products/10719
2012-10-10T06:09:28 < emeb> Time for a sparkfun order...
2012-10-10T06:09:36 < dongs> i just buy aoyue  tips that are I think ripoffs/clones of hakko tips
2012-10-10T06:09:53 < dongs> and/or maybe just hakko tips with a diufferent label on them
2012-10-10T06:09:59 < dongs> except you get liek 10 for the price of 1 hakko
2012-10-10T06:10:36 < emeb> Cool.
2012-10-10T06:11:30 < dongs> im submitting the ledtrash sometime this week,,, i missed like 3 panels that went out cuz i was too lazy
2012-10-10T06:11:38 < dongs> depends if theres another order i can panel it on to
2012-10-10T06:12:56 < emeb> Nice. Will be interesting to see how that turns out.
2012-10-10T06:13:50 < dongs> gonna ha ve to handsolder the shit unless there's also stencil I can tack it on to
2012-10-10T06:14:24 < emeb> better get a stencil - you don't want to hand solder a gross of those. :)
2012-10-10T06:14:48 < dongs> yeah, but i'd have to wait until theres enough stencil stuff to submit that I can panel it
2012-10-10T06:14:54 < dongs> i will definitely get one though
2012-10-10T06:15:28 < emeb> trying to find some other stuff to get from SF. Why is it all so boring lately?
2012-10-10T06:15:45 < emeb> all arduino stuff. <yawn>
2012-10-10T06:16:03 < dongs> blame all the MAKE:Rs
2012-10-10T06:16:18 < emeb> meh. there goes the neighborhood.
2012-10-10T06:17:03 < emeb> t-shirts! patches! blech.
2012-10-10T06:17:17 < dongs>  It's an Arduino-controlled ouija board!
2012-10-10T06:18:10 < emeb> is it just me or is most of their stuff way overpriced too?
2012-10-10T06:18:20 < dongs> yeah, duh
2012-10-10T06:18:22 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-10T06:18:34 < emeb> $6.95 for an ATMega32U4?
2012-10-10T06:18:37 < dongs> i mean shit like https://www.sparkfun.com/products/9801
2012-10-10T06:18:44 < dongs> thats a $2 chip on a $0.01 pcb
2012-10-10T06:18:49 < emeb> *cough*
2012-10-10T06:19:14 < emeb> guess they gotta do something to afford all those hipsters that work for them.
2012-10-10T06:25:12 < emeb> Man - it's been like 2 years since I last ordered from SF.
2012-10-10T06:25:29 < emeb> Probably the last free day.
2012-10-10T06:25:36 < emeb> what a circus that was.
2012-10-10T06:26:23 < dongs> i buy BNC connectors from them
2012-10-10T06:26:48 < dongs> https://www.sparkfun.com/products/10550
2012-10-10T06:26:50 < dongs> shit is filthy cheap
2012-10-10T06:27:06 < emeb> not bad.
2012-10-10T06:27:15 < emeb> prolly cheaper than Pasternack.
2012-10-10T06:32:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-10T06:32:58 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-10T06:33:20 < dongs> still seaching through their shit for bnc right angle
2012-10-10T06:33:33 < emeb> didn't see any PCB mount.
2012-10-10T06:33:38 < dongs> ditto
2012-10-10T06:33:57 < emeb> they're known for high prices - not worth buying from them if you can get it anywhere else.
2012-10-10T06:37:08 < emeb> switching machines...
2012-10-10T06:37:12 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-10T06:39:34 < dongs> http://www.youtube.com/watch?v=vN2WzQzxuoA haha
2012-10-10T06:42:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-10T06:54:00 < dongs> would be cool if you didnt announce it
2012-10-10T06:54:04 < dongs> and just quietly disappeared
2012-10-10T07:07:45 < emeb_mac> ouch!
2012-10-10T07:56:48 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 240 seconds]
2012-10-10T08:13:17 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 244 seconds]
2012-10-10T08:48:44 -!- claude is now known as Claude
2012-10-10T08:56:43 < jpa-> has someone used USB OTG FS on stm32f4 without the dedicated VBUS pin (PA9)?
2012-10-10T08:57:48 < jpa-> for some reason it is not activating the pull-up resistor, even though i have enabled NOVBUSSENS in OTG_FS_GCCFG
2012-10-10T09:06:06 < zippe> What do you mean by "not activating the pull-up resistor"?
2012-10-10T09:06:26 < zippe> You mean not pulling D+ ?
2012-10-10T09:08:05 < dongs> thats what it sounds like.
2012-10-10T09:08:05 < jpa-> yeah
2012-10-10T09:08:15 < jpa-> so host does not start enumeration and nothing happens
2012-10-10T09:08:26 < jpa-> i'm trying to use the usb driver that comes with nuttx
2012-10-10T09:09:59 < zippe> You set FDMOD ?
2012-10-10T09:10:20 < jpa-> yes
2012-10-10T09:10:55 < jpa-> the usb bootloader in ROM works.. i'm now going to dump the otg fs registers and compare what it does differently
2012-10-10T09:17:19 < jpa-> aha, the bootloader disables SWD
2012-10-10T09:17:22 < jpa-> so much for that
2012-10-10T09:20:03 < emeb_mac> denied!
2012-10-10T09:20:07 < dongs> uh
2012-10-10T09:20:08 < dongs> it does?
2012-10-10T09:20:14 < dongs> why would it do that lol
2012-10-10T09:20:31 < jpa-> maybe in case some nutjob has used those pins for other purpose in his design
2012-10-10T09:20:51 < dongs> maybe its so that you cant patch out ROP via swd/jtag
2012-10-10T09:21:17 < jpa-> could be something else, but my debugger cannot read memory when the bootloader is running
2012-10-10T09:25:16 < zyp> jpa-, what does BSVLD in GOTGCTL read?
2012-10-10T09:26:44 < jpa-> register 0x50000000 = 0x000d0000   so 1.. not sure why it says also A session valid
2012-10-10T09:27:43 < zyp> do you only have NOVBUSSENS set in GCCFG, or do you also have VBUSBSEN? if not, try it
2012-10-10T09:27:55 < jpa-> i'm not sure what i should do with VBUSBSEN and VBUSASEN in GCCFG, but i've tried various combinations
2012-10-10T09:28:46 < zyp> also, I've always seen ASVLD at the same value as BSVLD, so that's normal
2012-10-10T09:29:10 < jpa-> ok
2012-10-10T09:29:13 < zyp> what about HNPCAP and SRPCAP in GUSBCFG?
2012-10-10T09:29:56 < zyp> if you have them set, I suggest trying without them
2012-10-10T09:31:07 < jpa-> GUSBCFG 0x5000000c = 0x40001440  so they are not set
2012-10-10T09:31:45 < zippe> hrm, I had no trouble using JTAG in the bootloader
2012-10-10T09:31:58 < zippe> It's how I flashed my early bootloaders on the F4
2012-10-10T09:32:09 < zyp> jpa-, and you've verified that your code works when you hook vbus to PA9?
2012-10-10T09:43:32 < dongs> attn zyp/tectu http://i.imgur.com/ScrDb.jpg
2012-10-10T09:43:33 < jpa-> not yet
2012-10-10T09:43:52 < jpa-> but actually, for some reason, SDIS was set in DCTL.. that resolved the pull-up issue
2012-10-10T09:44:07 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-10T09:44:07 < Tectu> dongs, what's that for a nice thing
2012-10-10T09:44:18 < jpa-> it still does not enumerate, but i'll debug further
2012-10-10T09:44:18 < dongs> wut
2012-10-10T09:45:18 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Client Quit]
2012-10-10T09:48:19 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-10T09:55:08 < jpa-> aha, i have to call the class level driver setup function myself
2012-10-10T09:57:38 -!- Claude is now known as claude
2012-10-10T09:57:48 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-10-10T10:02:56 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Ping timeout: 245 seconds]
2012-10-10T10:18:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-10T10:33:12 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-10T10:42:00 -!- claude is now known as claude_work
2012-10-10T10:47:36 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-10T10:50:52 -!- dekar [~dekar@212.255.19.185] has quit [Quit: This computer has gone to sleep]
2012-10-10T10:57:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-10T11:27:43 -!- dekar [~dekar@studpool-wlan-74-34.fs.fbi.h-da.de] has joined ##stm32
2012-10-10T11:27:46 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-10T12:06:26 < dongs> great news for zyp and tectu http://www.japantoday.com/category/lifestyle/view/japans-first-cuddle-cafe-lets-you-sleep-with-a-stranger-for-y6000-an-hour
2012-10-10T12:06:58 < zyp> isn't that a couple of weeks old?
2012-10-10T12:07:34 < dongs> sorry just got pasted to me today
2012-10-10T12:15:22 < zyp> I wonder who has money to spend on shit like that though
2012-10-10T12:15:50 < zyp> on the other hand, being paid 3500 yen per hour just to sleep doesn't sound like a bad job
2012-10-10T12:28:22 < dongs> ya, im applying.
2012-10-10T12:28:59 -!- Steffanx [~Steffanx@77.167.38.174] has joined ##stm32
2012-10-10T12:29:02 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-10T12:36:49 -!- dekar [~dekar@studpool-wlan-74-34.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-10T12:39:55 -!- Steffanx [~Steffanx@77.167.38.174] has quit [Quit: This computer has gone to sleep]
2012-10-10T12:47:31 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-10T13:02:01 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has joined ##stm32
2012-10-10T13:03:40 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-10T13:05:33 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-10T13:26:02 < Tectu> dongs, sounds more like good news for you
2012-10-10T13:27:07 < dongs> how so
2012-10-10T13:29:08 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 256 seconds]
2012-10-10T13:30:26 < dongs> ahha, fuck yea, in related articles http://www.japantoday.com/category/national/view/tepco-couldnt-get-battery-during-fukushima-meltdown-went-to-hardware-store-the-next-day
2012-10-10T13:41:20 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-10T13:52:19 < dongs> attn lunixfgts http://shop.embedded-projects.net/index.php?module=artikel&action=artikel&id=1694
2012-10-10T13:53:38 < zyp> more outdated crap?
2012-10-10T13:53:43 < zyp> ARM9
2012-10-10T13:53:45 < dongs> but its GNU
2012-10-10T13:54:42 < zyp> pretty cheap and small too
2012-10-10T13:54:53 < dongs> right.
2012-10-10T13:54:55 < zyp> not _that_ bad
2012-10-10T13:54:58 < qyx_> dongs: good links
2012-10-10T13:54:58 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-10T13:55:22 < dongs> thats including like 50% vat
2012-10-10T13:55:31 < dongs> so if i was gonna buy it its actually affordable
2012-10-10T13:56:14 < dongs> zyp, just incase you missed it < dongs> attn zyp/tectu http://i.imgur.com/ScrDb.jpg
2012-10-10T13:57:22 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-10T13:57:55 < karlp> so it needs a sdcard, then a breakout to put a usb-eth device there or something?
2012-10-10T13:58:06 < dongs> it has a sdcard
2012-10-10T13:58:07 < karlp> I mean, carambola is cheaper than taht and inclues flash and wifi
2012-10-10T13:58:19 < karlp> it has an sdcard slot,
2012-10-10T13:58:24 < karlp> there's no flash onboard.
2012-10-10T13:58:52 < dongs> carambola looks like its about the size of an xbox
2012-10-10T13:59:06 < karlp> 35x45mm?
2012-10-10T13:59:17 < karlp> ok, bigger than the gnublin sure.
2012-10-10T13:59:39 < karlp> wifi on board is kinda handy
2012-10-10T13:59:58 < dongs> sure, and are you saying its < $30?
2012-10-10T14:00:07 < karlp> 22 euro
2012-10-10T14:00:10 < karlp> http://shop.8devices.com/
2012-10-10T14:00:56 < dongs> cute
2012-10-10T14:01:12 < dongs> too bad its mips/lunix/etc..
2012-10-10T14:01:39 < karlp> mips/arm/whocards,
2012-10-10T14:01:52 < karlp> you can write C/python/lua whatever
2012-10-10T14:02:08 < karlp> that's the appeal of something that already runs linux
2012-10-10T14:02:21 < dongs> http://www.kickstarter.com/projects/2027072188/plasma-jet-electric-thrusters-for-spacecraft?ref=category lol wat
2012-10-10T14:03:59 < dongs> http://www.kickstarter.com/projects/kenburns/tinyduino-the-tiny-arduino-compatible-platform-w-s?ref=category more fail
2012-10-10T14:05:15 < dongs> http://www.kickstarter.com/projects/818225433/arduino-style-battery-shield holy fuck
2012-10-10T14:14:31 < jpa-> gah, this nuttx build process if totally messed up on windows
2012-10-10T14:14:59 < jpa-> it relies so heavily on symlinks & gcc dependency list generation that i basically have to make distclean before every build
2012-10-10T14:16:33 < dongs> haha
2012-10-10T14:18:43 < zyp> I've heard that make is such a modern and nice tool
2012-10-10T14:18:43 < dongs> http://www.kickstarter.com/projects/minioncity/minion-the-wireless-enabled-dev-board
2012-10-10T14:19:00 < dongs> had me interested until I scrolled down and saw they were ust using regular tqfp atmega
2012-10-10T14:19:09 < dongs> i thought the guys were gonna do COB manufacture for $2.5k
2012-10-10T14:19:22 < dongs> but looks like cob shit is just some ghetto chink rf module
2012-10-10T14:19:40 < dongs> i was like "atmel sells atmega32u4 dies?"
2012-10-10T14:22:02 < dongs> i guess wlcsp is the best tehy offer.
2012-10-10T14:27:10 < Laurenceb> http://pinterest.com/pin/11610911510160745/
2012-10-10T14:28:12 < Laurenceb>  "Mechapolypse"
2012-10-10T14:28:14 < Laurenceb> pmsl
2012-10-10T14:28:25 < karlp> couldn't you have given us the actual link, instead of the pinterest link?
2012-10-10T14:28:48 < zyp> I wonder if they realize that they are operating it out of spec
2012-10-10T14:29:05 < zyp> «We are using the Atmega32u4 clocked at 16mhz» «3.3V Power Regulator»
2012-10-10T14:29:11 < Laurenceb> lol
2012-10-10T14:29:18 < Laurenceb>  "particular passion for experimental garments and latex."
2012-10-10T14:29:19 < zyp> it's only speced to 8 MHz at 3.3V, 16MHz is for 5V
2012-10-10T14:29:21 < Laurenceb> tehhee
2012-10-10T14:29:36 < Laurenceb> iirc you can fun at 12mhz on 3.3v
2012-10-10T14:29:49 < dongs> zyp, just like my 128mhz overclocked F103 devboard
2012-10-10T14:30:59 < zyp> ah, maybe it was 12
2012-10-10T14:31:02 < zyp> still not 16
2012-10-10T14:32:23 < Laurenceb> zyp: theres an operating range table in the datasheet
2012-10-10T14:32:44 < zyp> sure
2012-10-10T14:32:55 < zyp> but I don't remember it 100% :p
2012-10-10T14:33:14 < zyp> I just remember that 16 is not valid at 3.3V, so I ran 8 in my last avr project.
2012-10-10T14:33:21 < Laurenceb> 13.5Mhz at 3.3v or something
2012-10-10T14:34:15 < dongs> whatever.
2012-10-10T14:34:18 < dongs> its fucking fail arduino.
2012-10-10T14:34:26 < dongs> should just fucking die
2012-10-10T14:36:01 < dongs> http://www.kickstarter.com/projects/bzcore/bizzybee-breadboard-with-built-in-arduino
2012-10-10T14:37:49 < Laurenceb> Eliminating the middle man.
2012-10-10T14:38:23 < dongs> http://www.kickstarter.com/projects/323660810/640-1280-2560tiny ahh that poor guy
2012-10-10T14:38:26 < dongs> he had this shit up for a month
2012-10-10T14:38:42 < dongs> hes almost like tom66
2012-10-10T14:39:06 < dongs> http://www.kickstarter.com/projects/2135449066/quickscanit-your-portable-smart-phone-document-sys scambox, revisited
2012-10-10T14:39:12 < dongs> you can only scam people once wiht this sorta trash.
2012-10-10T14:39:47 < karlp> yeah, but there's new people all the time :)
2012-10-10T14:41:03 < dongs> http://www.kickstarter.com/projects/1048071893/introducing-the-best-desk-top-laser-on-the-market?ref=category wut
2012-10-10T14:41:27 < dongs> http://www.kickstarter.com/projects/74441894/gigastrand-operating-system hahaha.
2012-10-10T14:42:19 < dongs> The second version is the Premium Version. This will have some really cool, kick-butt stuff that you would normally have to purchase separate from your OS and figure out how to install yourself. This will be sold at a premium price through Gigastrand pre-loaded on its PCs and through Linux4Everyone as a standalone retail product.
2012-10-10T14:42:24 < dongs> fuck yea
2012-10-10T14:42:27 < dongs> exactly waht I want.
2012-10-10T14:42:36 < Laurenceb> lol
2012-10-10T14:43:10 < dongs> The Gigastrand Operating System is a project worthy of your support. Everyone who donates will get a credit mention on the OS page. Even if you can only donate a small amount, we will make sure everyone will benefit in some way. After all, our mission is Making Linux Possible for Everyone.
2012-10-10T14:43:20 < dongs> im gonna throw my money to be listed on  opensauce page
2012-10-10T15:07:17 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-10T15:07:42 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-10T15:13:01 -!- dekar [~dekar@212.255.19.185] has joined ##stm32
2012-10-10T15:13:04 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-10T15:30:21 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-10T15:32:25 < Laurenceb> http://spacenear.us/tracker/
2012-10-10T15:32:28 < Laurenceb> heading for tom66
2012-10-10T15:34:25 < Laurenceb> in the piss poor area of town
2012-10-10T15:37:51 < dongs> haha
2012-10-10T15:37:56 < dongs> Laurenceb: we were just talking about tom66
2012-10-10T15:37:58 < dongs> in another channel
2012-10-10T15:38:05 < dongs> how the fuck did you knwo
2012-10-10T15:38:22 < dongs> <tom66> Well, had a problem this summer. :/ Btw, I'm now at the University of Leeds doing my MEng Electronics Engineering. I was going to make it at BitBox in my last two weeks but the reflow oven died - the clutch broke on the drive belt. They got it working again luckily but there was too little time left with their back log needing to be done, of course.  I'm going to see if I've got enough time to do it over winter.
2012-10-10T15:39:22 < dongs> http://www.kickstarter.com/projects/739076784/high-performance-32-bit-through-hole-arduinotm-com
2012-10-10T15:41:09 < dongs> Laurenceb: whats wrong wiht that baloon, is it falling?
2012-10-10T15:41:12 < Laurenceb> omfg
2012-10-10T15:41:16 < Laurenceb> it failed
2012-10-10T15:41:25 < dongs> blame tom66
2012-10-10T15:41:30 < dongs> maybe you can call him up and ask for extraction
2012-10-10T15:41:36 < Laurenceb> the balloon noobs are convinced they can make foil balloons superpressurize
2012-10-10T15:41:42 < Laurenceb> more like fool balloons
2012-10-10T15:41:47 < dongs> haha
2012-10-10T15:41:50 < dongs> its gonna land ain a lake
2012-10-10T15:42:05 < dongs> hm no
2012-10-10T15:42:13 < Laurenceb> they need to be much stronger to pressurize
2012-10-10T15:42:33 < Laurenceb> cheap balloon shop foil balloons just burst like that one
2012-10-10T15:43:01 < dongs> is this shit landing or wat
2012-10-10T15:43:13 < Laurenceb> it got run over by a lorry
2012-10-10T15:43:17 < Laurenceb> hopefully
2012-10-10T15:43:46 < dongs> on that highway?
2012-10-10T15:43:57 < dongs> is there pics
2012-10-10T15:43:58 < Laurenceb> oh its crossed
2012-10-10T15:44:00 < Laurenceb> boring
2012-10-10T15:45:08 < Laurenceb> hehe stuck in the treeline
2012-10-10T15:45:38 < dongs> how you know
2012-10-10T15:47:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-10T15:47:38 < Laurenceb> maybe not
2012-10-10T15:47:43 < Laurenceb> they lost radio
2012-10-10T15:48:00 < Laurenceb> meanwhile at tom66 house
2012-10-10T15:48:02 < Laurenceb> http://4.bp.blogspot.com/-mOeBw-YHuLU/T7jbmsRqVyI/AAAAAAAAAw8/U3mAUFxzLvM/s1600/narwhal-cosplay.jpg
2012-10-10T16:00:04 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-10T16:01:34 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has joined ##stm32
2012-10-10T16:29:01 -!- daku is now known as DaKu
2012-10-10T16:35:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-10T16:38:06 < dongs> http://www.kickstarter.com/projects/671510748/superlf398-low-parts-count-sample-and-hold hahahaha "PCBs are from www.iteadstudio.com. They are in China, have wonderful customer service, and consistently provide excellent quality. The PCBs are ROHS compliant with solder mask and silkscreen."
2012-10-10T16:38:40 < dongs> attn Laurenceb / etc
2012-10-10T16:38:44 < Laurenceb> http://i.imgur.com/faoiy.jpg
2012-10-10T16:38:54 < Laurenceb> found in my attic
2012-10-10T16:39:09 < Laurenceb> heh
2012-10-10T16:39:17 < Laurenceb> writing is too blurred actually
2012-10-10T16:40:13 < Laurenceb> so you have to guess which one is related to me
2012-10-10T16:41:21 < zyp> I'm guessing the leftmost guy with white jacket
2012-10-10T16:41:38 < dongs> looks sorta like hitler
2012-10-10T16:43:16 < Laurenceb> in b4 "paki guy"
2012-10-10T16:43:38 < BrainDamage> http://i.imgur.com/OQ1Dk.jpg
2012-10-10T16:43:49 < cjbaird> It's either one of those guys who must sit on the ground.
2012-10-10T16:44:25 < BrainDamage> I'd say it's the tree on the background
2012-10-10T16:44:34 < Laurenceb> its actually top right guy
2012-10-10T16:44:35 < BrainDamage> since there's lack of an idiot photobombing
2012-10-10T16:45:18 < Laurenceb> the preist made me laugh
2012-10-10T16:45:44 < Laurenceb>  "team chaplain"
2012-10-10T16:49:43 < Thorn> wtf sucktrace
2012-10-10T16:49:47 < dongs> what no we
2012-10-10T16:49:49 < dongs> now
2012-10-10T16:49:57 < dongs> did you upgrade to latest beta?A?
2012-10-10T16:50:08 < Thorn> doesn't keep static via net assignment with copy/paste
2012-10-10T16:50:17 < Thorn> I think it appeared in the beta.
2012-10-10T16:50:19 < dongs> thats expected
2012-10-10T16:50:20 < dongs> no
2012-10-10T16:50:23 < dongs> thats laways been that
2012-10-10T16:50:32 < dongs> is this for gnd vias?
2012-10-10T16:50:38 < dongs> you have to manually run traces to them
2012-10-10T16:50:45 < Thorn> wat
2012-10-10T16:50:46 < dongs> under the copper fill
2012-10-10T16:50:49 < Thorn> never had to do that
2012-10-10T16:50:59 < dongs> or else when you update from schematic  they'll all lose the nets...
2012-10-10T16:51:03 < Thorn> but I have to select GND for all of them
2012-10-10T16:51:14 < dongs> well, seelcting all vias is easy
2012-10-10T16:51:17 < dongs> edit->selection
2012-10-10T16:51:52 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Quit: Leaving]
2012-10-10T16:52:13 < Thorn> just tried updating from shematics, nothing bad happened
2012-10-10T16:52:14 < dongs> Thorn: do you use sync to schematic shit (ctrl+alt-u) once you make schema changes to reflect htem on pcb?
2012-10-10T16:52:28 < dongs> i always had to run traces to all gnd grid vias
2012-10-10T16:52:30 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-10T16:52:31 < dongs> if I wanted htem to stay in gnd
2012-10-10T16:52:37 < Thorn> I use 'by component'
2012-10-10T16:52:44 < Thorn> from the same menu
2012-10-10T16:53:07 < zyp> dongs, that sounds pretty dumb
2012-10-10T16:53:17 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-10T16:53:23 < dongs> zyp, why?
2012-10-10T16:53:32 < dongs> it works and i dont mind.
2012-10-10T16:54:23 < zyp> so when you have finished the board and want to stitch up the ground planes with a few more vias, you have to run a trace to all of them?
2012-10-10T16:54:45 < Thorn> that'd be a disaster
2012-10-10T16:56:34 < dongs> zyp: yea, thats what i do?
2012-10-10T16:56:39 < dongs> its gnd plane anyway, you dont see the traces
2012-10-10T16:57:04 < dongs> i dunno apparently hte non-idiotic way works for thorn but ive been doing it this way for years so shrug
2012-10-10T16:57:44 < zyp> so why do you even need them if you don't see them?
2012-10-10T16:58:35 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-10T16:58:52 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-10T16:59:38 < Thorn> http://i.imgur.com/IMAz5.png
2012-10-10T17:00:03 < zyp> what about when you want to add a signal later, and then hit one of those traces? sounds like a bunch of pointless bother
2012-10-10T17:00:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-10T17:00:14 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-10T17:00:19 < karlp> what's that six pin header on the bottom?
2012-10-10T17:00:31 < dongs> jtag?>
2012-10-10T17:00:31 < Thorn> debugger debug
2012-10-10T17:00:46 < dongs> wait this is different shit
2012-10-10T17:00:49 < dongs> from your can-scam thing
2012-10-10T17:00:55 < zyp> Thorn, no edge connector for direct plugging into boards?
2012-10-10T17:01:01 < dongs> does anyone here have a f3discovery board they can sell me
2012-10-10T17:01:10 < zyp> and why a huge B connector on such a small board?
2012-10-10T17:01:29 < karlp> werenðt you putting a discovery style header on it?
2012-10-10T17:01:31 < dongs> he probably couldnt find any miniB in dip
2012-10-10T17:01:59 < Thorn> zyp: nope I thought that connection style wouldn't be too reliable on a workbench
2012-10-10T17:02:17 < Thorn> and I HATE mini-B.
2012-10-10T17:02:23 < zyp> and I dislike lack of 0.05"-header
2012-10-10T17:02:54 < Thorn> well even if I add a pattern I'm unlikely to populate it
2012-10-10T17:02:58 < dongs> Thorn: your gnd vias are kinda too close to caps etc
2012-10-10T17:03:05 < dongs> (just personal preference though i guess
2012-10-10T17:03:12 < Thorn> although it could be done to future-proof the thing
2012-10-10T17:03:39 < dongs> also I sually put vias next to every gnd pad
2012-10-10T17:04:45 < dongs> also wtf
2012-10-10T17:04:51 < dongs> you made it 2 layer assembly just for 2 caps?
2012-10-10T17:04:57 < dongs> route those fuckers on top
2012-10-10T17:04:59 < dongs> dont be lazy
2012-10-10T17:05:07 < Thorn> it's for manual assembly
2012-10-10T17:05:11 < dongs> doesnt matter
2012-10-10T17:05:16 < Thorn> about 3-4 in total
2012-10-10T17:05:20 < zyp> oh well, guess I'll just have to do my own to get it the way I want :p
2012-10-10T17:06:18 < dongs> zyp, can you steal a f3discovery from somewehre
2012-10-10T17:06:38 < zyp> hard to steal one when I haven't even seen one
2012-10-10T17:08:18 < Thorn> well I guess I can place the 2 bottom parts on top.
2012-10-10T17:08:51 < Thorn> but NO mini-B.
2012-10-10T17:09:07 < zyp> I'd go with micro-b
2012-10-10T17:09:16 < dongs> no way
2012-10-10T17:09:18 < dongs> fuck SMD usb
2012-10-10T17:09:31 < zyp> the ones I got for my F4 board is pretty decent
2012-10-10T17:09:32 < Thorn> I only have one micro-b cable, it's forever jammed into my f4discovery
2012-10-10T17:09:50 < Thorn> don't think I can remove it now
2012-10-10T17:09:57 < zyp> they are PTH shroud, so it's not going to rip off easily
2012-10-10T17:10:07 < zyp> similar to the micro-ab on the f4discovery
2012-10-10T17:10:08 < dongs> WM17115-ND
2012-10-10T17:10:13 < dongs> is the only USB connector you should ever use
2012-10-10T17:10:13 < dongs> EVER
2012-10-10T17:10:30 < Laurenceb> http://uk.mouser.com/ProductDetail/Maxim-Integrated-Products/MAX9938TEBS+TG45/?qs=sGAEpiMZZMvEn2pkGav3bAvIVRe3YtizUtqFdcrVKnk%3d
2012-10-10T17:10:37 < Laurenceb> a little stock there
2012-10-10T17:10:52 < dongs> aahahh.
2012-10-10T17:11:07 < dongs> well its ucsp
2012-10-10T17:11:09 < dongs> so i guess 10k/reel
2012-10-10T17:11:14 < dongs> so tehy have 124 reels..
2012-10-10T17:11:15 < dongs> yeah
2012-10-10T17:11:57 <+Steffanx> Should be enough to survive one day
2012-10-10T17:12:13 < zyp> the only usb connectors I've ever ripped off a board was from my avr usb project, and that's because the footprint for the connector was dumb.
2012-10-10T17:13:18 < zyp> read: I used the footprint in the sparkfun eagle library
2012-10-10T17:13:33 < dongs> thats your problem rihgt thre
2012-10-10T17:13:57 <+Steffanx> I think it was because you didn't pay the hobby-tax zyp :P
2012-10-10T17:14:26 < zyp> http://bin.jvnv.net/f/DheuF.jpg <- turned out like this (bottom right)
2012-10-10T17:14:53 < zyp> only two small pads holding small tabs of the shroud, pretty much no mechanical strength
2012-10-10T17:17:41 < Thorn> it's going to be hanging halfway in air with this pattern, no?
2012-10-10T17:17:49 < zyp> the ones I'm using now have both tabs going into holes, tabs on pads and pads underneath the shroud itself
2012-10-10T17:18:09 < zyp> using the footprint recommended in the connector datasheet
2012-10-10T17:18:25 < zyp> http://bin.jvnv.net/f/g17eb.JPG
2012-10-10T17:18:59 < karlp> what's that rf board zyp?
2012-10-10T17:19:35 < zyp> uni project
2012-10-10T17:19:54 < zyp> for my bachelor thesis
2012-10-10T17:23:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-10T17:24:10 < BrainDamage> did you pick the japanise animu name to irritate dongs?
2012-10-10T17:25:07 <+Steffanx> He is a japan fanboy BrainDamage :P
2012-10-10T17:26:14 < Thorn> does diptrace have 0.05" patterns?
2012-10-10T17:26:22 < dongs> the fuck is a 0.05"  pattern
2012-10-10T17:26:26 < Thorn> search for FTSH finds nothing
2012-10-10T17:26:28 < dongs> you mean connector?
2012-10-10T17:26:33 < Thorn> that connector
2012-10-10T17:26:34 < dongs> Thorn: what hte fuck dude make your own
2012-10-10T17:26:39 < dongs> from teh datasheet
2012-10-10T17:26:40 < dongs> takes 5 sec
2012-10-10T17:27:03 < Thorn> too lazy
2012-10-10T17:27:13 < Thorn> search for FTSH finds nothing
2012-10-10T17:27:18 < dongs> of course it wont
2012-10-10T17:27:24 < dongs> i dont think i ever searched in any default libs
2012-10-10T17:28:18 < Thorn> does the ISP pin for LPC1768 need to have an external pullup so it doesn't boot into bootloader?
2012-10-10T17:28:29 < dongs> timecop.mine.nu/ftsh.lib
2012-10-10T17:29:57 < dongs> is what i use so..
2012-10-10T17:31:55 < Thorn> thanks
2012-10-10T17:32:11 < Thorn> have to install the latest latest beta
2012-10-10T17:32:15 < dongs> im to lazy to remake it but i'd proably put a more obvious pin1 mark on it
2012-10-10T17:32:18 < dongs> like a triangle or something.
2012-10-10T17:32:28 < dongs> now pin1 is thicker silk on left side.
2012-10-10T17:37:54 -!- Rickta59 [~kimballr@rrcs-70-61-73-166.midsouth.biz.rr.com] has quit [Ping timeout: 240 seconds]
2012-10-10T17:42:29 -!- claude_work is now known as claude
2012-10-10T17:44:39 < zyp> BrainDamage, of course
2012-10-10T17:44:40 -!- nashpa_ [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-10T17:44:51 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-10T17:45:26 -!- Netsplit *.net <-> *.split quits: nashpa
2012-10-10T17:45:38 -!- nashpa_ is now known as nashpa
2012-10-10T17:46:58 < Thorn> I want an 'unpour all pours' button
2012-10-10T17:47:40 < zyp> in dicktrace?
2012-10-10T17:49:20 < Thorn> in any sane package you could do write a script to do that yourself
2012-10-10T17:49:27 < Thorn> -do
2012-10-10T17:54:52 < Thorn> ok added that thing
2012-10-10T17:55:43 < Thorn> no room for copyright now
2012-10-10T17:56:36 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 256 seconds]
2012-10-10T18:00:10 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-10T18:05:06 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-10-10T18:06:00 < Thorn> http://i.imgur.com/qc4Lb.png
2012-10-10T18:07:09 < Thorn> ah, need to emphasize pin 1 too
2012-10-10T18:07:52 < zyp> the shrouded connectors I'm using would cover your mounting hole
2012-10-10T18:08:06 < zyp> by the way, why do you have mounting holes? :p
2012-10-10T18:08:33 < Thorn> I had them because of components on the bottom
2012-10-10T18:08:53 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-10T18:08:53 < Thorn> now that they're gone, the utility of mounting holes is somewhat debatable
2012-10-10T18:09:04 < zyp> heh
2012-10-10T18:09:35 < zyp> you're still going to have a bunch of pins protruding through the bottom, wouldn't they be more annoying anyway?
2012-10-10T18:10:35 < Thorn> yeah. although I notice boards like this one tend to end up hanging in the air somehow
2012-10-10T18:10:54 < Thorn> certainly true looking at my ft2232h board
2012-10-10T18:11:16 < Thorn> it's hanging by the usb cable from my table lamp.
2012-10-10T18:11:23 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-10T18:11:32 < zyp> that's why I'd like to just plug the board directly into the target board
2012-10-10T18:12:56 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-10T18:14:02 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-10T18:16:04 < zyp> what's the physical measurements of your board?
2012-10-10T18:17:31 < Thorn> 85x30mm
2012-10-10T18:21:47 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-10T18:31:06 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-10T18:35:42 < Thorn> is this enough clearance for the 0.05" connector? http://i.imgur.com/6NVM4.png
2012-10-10T18:40:35 < zyp> depends on which you use
2012-10-10T18:40:57 < zyp> the shrouded one I'm using will cover hole and some of the silk
2012-10-10T18:42:38 < Thorn> found a forum thread where someone says that lpc17xx performs much better than stm32f1xx on mp3 decoding, stm32 can only do one channel @ 320Kbps while lpc can do 2 with 25% cpu load. wtf
2012-10-10T18:43:18 < zyp> right, because everything that's written in forum threads is true
2012-10-10T18:43:49 < Thorn> that someone tested both himself
2012-10-10T18:44:40 < zyp> in a very scientific way, I'm sure
2012-10-10T18:47:50 < emeb> how much difference can there be between various flavors M3?
2012-10-10T18:48:23 < emeb> clock speed, instruction cache, peripheral performance.
2012-10-10T18:48:42 < Thorn> bus topology
2012-10-10T18:49:02 < Laurenceb> lol
2012-10-10T18:49:16 < Laurenceb> maybe it means whoever wrote the stm32 code fails at coding
2012-10-10T18:49:19 < Thorn> lpc17xx have 2 kinds of ram like stm32f4 and some kind of flash accelerator
2012-10-10T18:49:40 < Thorn> he says code is exactly the same
2012-10-10T18:50:03 < Thorn> it does sound fishy but I've no idea where he could fail
2012-10-10T18:50:18 < Laurenceb> i see
2012-10-10T18:50:36 < Laurenceb> Thorn: nice bmp clone
2012-10-10T18:50:49 < Laurenceb> i tried to clone on maple but i failed miserably
2012-10-10T18:51:04 < Laurenceb> something wrong with my libopencm3 linker
2012-10-10T18:51:17 < Laurenceb> are gerbers released?
2012-10-10T18:52:41 < Thorn> what gerbers? original bmp?
2012-10-10T18:54:33 < Laurenceb> no, your board
2012-10-10T18:55:09 < Thorn> they don't exist yet lol but I can release both diptrace files and gerbers if there's any interest
2012-10-10T18:56:11 < Thorn> (and gsmcmullin doesn't mind)
2012-10-10T18:57:08 < Thorn> hopefully I'll be able to order boards from a local manufacturer who does them in a week with delivery instead of a month (seeedstudio)
2012-10-10T18:57:31 < Thorn> it's expensive though so I try to submit 3 projects at once
2012-10-10T18:57:43 < Thorn> then I'll know if it works
2012-10-10T18:59:34 -!- Tectu is now known as tectu
2012-10-10T18:59:37 < Laurenceb> cool
2012-10-10T19:03:04 < cjbaird> Was there anything more said about the possibility of hacking the Ti Stellaris to run a BMP port?
2012-10-10T19:04:52 <+Steffanx> zlog :P
2012-10-10T19:04:52 < zlog> Steffanx: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-10-10.html
2012-10-10T19:05:00 < Thorn> bmp uses libopencm3, I don't think it has any code for stellaris yet
2012-10-10T19:05:18 < Thorn> ah it does https://github.com/libopencm3/libopencm3/tree/master/lib/lm3s
2012-10-10T19:05:21 -!- tectu is now known as Tectu
2012-10-10T19:05:26 < Thorn> GPIO will work :D
2012-10-10T19:07:37 <+Steffanx> All you need is standard gpio, just like the ftdi port :P
2012-10-10T19:07:38 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-10T19:10:18 < Thorn> usb is also marginally useful
2012-10-10T19:10:28 <+Steffanx> Oh, of course
2012-10-10T19:11:11 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-10T19:11:22 <+Steffanx> TI has no CMSIS-ish lib you can use?
2012-10-10T19:11:36 < cjbaird> I believe it does.
2012-10-10T19:11:38 <+Steffanx> *peripheral lib
2012-10-10T19:11:56 < Thorn> Stellarisware
2012-10-10T19:12:07 < Thorn> I think it even comes in ROM or something
2012-10-10T19:12:07 < zyp> adding lm3s support to the libopencm3 usb stack shouldn't be that hard
2012-10-10T19:12:31 < Thorn> or some poarts of it
2012-10-10T19:12:35 < Thorn> -o
2012-10-10T19:12:40 <+Steffanx> I hear that's zyp's next project
2012-10-10T19:13:04 < zyp> I didn't spend much time adding F1 support to my stack, and it's entirely different from F4 USB
2012-10-10T19:13:28 < zyp> so should be around same amount of effort
2012-10-10T19:13:53 < Thorn> btw there's an opensource usb stack for lpc2368 called lpcusb. it's been ported to lpc1768. (fyi)
2012-10-10T19:14:18 <+Steffanx> Isn't that a port of lufa?
2012-10-10T19:14:56 < Thorn> lufa is for synopsys dwc no?
2012-10-10T19:15:47 < zyp> lufa is for avr
2012-10-10T19:15:56 <+Steffanx> lpcusb = lufa + modifications
2012-10-10T19:16:36 <+Steffanx> or isn't nxpUSBlib the same als pcusb ?
2012-10-10T19:16:53 <+Steffanx> *lpcusb
2012-10-10T19:17:04 <+Steffanx> No, nevermind :P
2012-10-10T19:17:37  * Steffanx got confused
2012-10-10T19:21:55 <+Steffanx> and i'm very sorry i confused you too :P
2012-10-10T19:23:55 < karlp> nxp embedded lufa+mods into rom right?
2012-10-10T19:24:43 < Thorn> there're some lpc11xx chihps with usb code in rom afair
2012-10-10T19:24:50 < Thorn> lpc17xx doesn't have that
2012-10-10T19:39:51 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-10T19:43:18 -!- barthess [~barthess@178.154.94.18] has joined ##stm32
2012-10-10T19:51:55 -!- phantone [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-10-10T19:52:16 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-10T20:15:05 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-10T20:45:34 -!- tavish_ [~tavish@120.56.138.237] has joined ##stm32
2012-10-10T20:48:04 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-10-10T20:53:08 -!- tavish_ [~tavish@120.56.138.237] has quit [Ping timeout: 255 seconds]
2012-10-10T21:12:27 < Thorn> most if not all development boards I've seen use RMII instead of MII, why? is it just to save pins? it uses higher frequency and requires an exernal oscillator
2012-10-10T21:15:44 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-10T21:15:52 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has joined ##stm32
2012-10-10T21:16:52 < zyp> Thorn, I would assume so
2012-10-10T21:17:19 < zyp> not just pins, but routing in general
2012-10-10T21:23:57 < Thorn> olimex seems to have switched to chibios for demo code for recent boards but is still using old 20-pin jtag connector.
2012-10-10T21:28:11 < Thorn> I'm not the only one who has external pullups on jtag signals. http://www.keil.com/mcbstm32f400/mcbstm32f400-schematics.pdf
2012-10-10T21:37:42 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-10T21:42:59 < zippe> Thorn: also, small, low-cost transceivers tend to be RMII
2012-10-10T21:43:10 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-10T21:51:21 -!- tavish_ [~tavish@59.177.10.101] has joined ##stm32
2012-10-10T21:53:26 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-10T21:59:21 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-10T21:59:29 < TitanMKD> hi
2012-10-10T22:06:17 < zyp> hi
2012-10-10T22:08:29 < emeb> re JTAG pullups - did they do that because it's needed for the STM32, or because "we've always done it that way"?
2012-10-10T22:08:39 < emeb> IIRC STM32 has built-in JTAG pulls.
2012-10-10T22:10:38 < Thorn> the very first book on microcontrollers I read taught me to never trust internal pullups. it was about avr and the author used his industrial designs as examples but still.
2012-10-10T22:11:46 < Thorn> especially on reset, ISP pins etc.
2012-10-10T22:12:10 < Thorn> not sure how universally applicable it is
2012-10-10T22:12:22 < emeb> Internal pulls tend to be fairly weak and vary widely w/ process. But if the were completely untrustworthy then there would be no point in providing them at all.
2012-10-10T22:15:19 < zyp> so, what's the risk if you don't have it?
2012-10-10T22:16:05 < emeb> I guess it depends a lot on what the expected environment is. If you're in a high EMI situation, or uncontrolled impedances on the I/O pins (say, someone handling the device during operation) then strong, controlled pulls become more critical.
2012-10-10T22:16:16 -!- tavish_ [~tavish@59.177.10.101] has quit [Ping timeout: 252 seconds]
2012-10-10T22:17:22 < Thorn> iirc avr are easy to put into programming mode (a single glitch is enough), it's not as easy with cortex-m
2012-10-10T22:17:25 < TitanMKD> zyp i will say risk on forgetting pullup is bus contention
2012-10-10T22:17:29 < TitanMKD> and burn IO
2012-10-10T22:17:29 < emeb> For industrial controllers, yeah it makes sense. For consumer electronics, probably yes.
2012-10-10T22:17:51 < zyp> bus contention on jtag/swd?
2012-10-10T22:17:56 < emeb> for hobby/lab equipment, not as big a deal.
2012-10-10T22:18:06 < TitanMKD> zyp i will say for any GPIO in fact
2012-10-10T22:19:01 < Thorn> another risk (for gpio) is indeterminate behavoir when the mcu is in reset / not yet initialized and GPIOs are tristated
2012-10-10T22:19:28 < Thorn> I think a satellite mission failed because of something like that once
2012-10-10T22:19:42 < TitanMKD> yes to have know state at startup
2012-10-10T22:20:04 <+dekar> they they saved a resistor, that should pay for another satellite mission! :D
2012-10-10T22:20:19 < zyp> you are getting very far off track now
2012-10-10T22:20:19 < Thorn> e.g. if you leave a DE input of a rs-485 driver floating you could disturb the whole bus
2012-10-10T22:20:31 < zyp> we were talking about jtag pullups, not pullups in general
2012-10-10T22:21:26 <+dekar> concerning RMII, couldn't you output 50mhz from the stm32?
2012-10-10T22:22:10 <+dekar> aren't the pins internally running at 50mhz already on an f1?
2012-10-10T22:22:12 < TitanMKD> for info on JTAG pull up see the official ARM details here http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka11418.html
2012-10-10T22:23:21 < TitanMKD> the most important is to use pullup to avoid floating JTAG pin
2012-10-10T22:23:37 < TitanMKD> (when they are not connected)
2012-10-10T22:24:24 <+dekar> also if you need a 50mhz clock for RMII, couldn't you use that to clock the PHY as well as the stm32?
2012-10-10T22:24:30 < zyp> yes, and 1. that's achieved by the internal pullups, 2. what's the risk if they are floating?
2012-10-10T22:24:58 < TitanMKD> zyp the risk is JTAG could understand random command and execute them and crash the CPU ;)
2012-10-10T22:26:22 < zyp> and what's the probability of that?
2012-10-10T22:26:41 < zyp> now, take that probability and multiply it with the probability of the internal pullups not being sufficient
2012-10-10T22:26:57 < zyp> that's how unlikely it is that you'll have any problems.
2012-10-10T22:26:59 < TitanMKD> it is big as floating pin change state all the time or is fixed temporarily to a state (depending on pin state of pin around sometimes)
2012-10-10T22:27:51 < Thorn> speaking of that satellite, they had a pyro emulator which did trigger during power-up of the pyro control box, but it was dismissed as the emulator being too sensitive. so when that box was powered up shortly after launch...
2012-10-10T22:28:01 < TitanMKD> zyp it is like SEU or MBU
2012-10-10T22:28:24 < TitanMKD> you have very low probability at ground but 1 each hour at 10000m alt
2012-10-10T22:28:39 < TitanMKD> and 1 each sec in space ;)
2012-10-10T22:29:08 < TitanMKD> it is just the difference between a good hw and a bad one ;)
2012-10-10T22:29:37 < TitanMKD> zyp i'm sure you also avoid using ESD chipset on USB port for the same reason
2012-10-10T22:29:52 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Remote host closed the connection]
2012-10-10T22:30:04 < zyp> yep
2012-10-10T22:30:23 < TitanMKD> it's a big mistake ;)
2012-10-10T22:30:33 < karlp> depends whether you're makign 1 or 10000
2012-10-10T22:30:46 < karlp> and depends whether you'd rather pay for replacements or not.
2012-10-10T22:31:00 < karlp> it's a design tradeoff you should maybe be aware of, but it's hardly a "big mistake"
2012-10-10T22:31:16 < TitanMKD> yes it will do a real difference on lot of hw when you will see hw problems will exceed 10% ;) instead of less than 2%
2012-10-10T22:31:27 < zyp> I'm not using ESD chipset on the USB port because it seems silly to only protect it and not all the other connection points which are just as vulnerable
2012-10-10T22:31:37 < karlp> sure 10% might be an issue
2012-10-10T22:32:07 < karlp> but vodafone here in iceland has a 28% failure rate on STBs, and it's still working for them.
2012-10-10T22:32:12 < karlp> they're not happy,
2012-10-10T22:32:18 < karlp> and they chose a different STB vendor,
2012-10-10T22:32:24 < karlp> but it's workable.
2012-10-10T22:32:31 < TitanMKD> karlp woo 28% failure it's a total fail for hw
2012-10-10T22:32:41 < karlp> depends where the money's coming from, what damage is it doing, and how easy it is to replace
2012-10-10T22:32:48 < karlp> oh yeah, 28% is epic :)
2012-10-10T22:33:01 < zyp> if I made an enclosed product where the usb port would be the only connection points that users would come in contact with, ESD protection on USB lines would make sense
2012-10-10T22:33:30 < TitanMKD> zyp i always plan USB protection on PCB even if user does not want to solder it ;)
2012-10-10T22:34:21 < zyp> but why only usb?
2012-10-10T22:34:34 < TitanMKD> the protection is also on +VCC
2012-10-10T22:34:39 < zyp> if you are that obsessed about esd protection, why only USB?
2012-10-10T22:34:44 -!- ChanServ [ChanServ@services.] has quit [shutting down]
2012-10-10T22:34:53 < TitanMKD> and IP4220CZ6
2012-10-10T22:35:02 < TitanMKD> is not so expensive and even easy to solder by hand
2012-10-10T22:35:25 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-10-10T22:35:25 -!- ServerMode/##stm32 [+o ChanServ] by lindbohm.freenode.net
2012-10-10T22:35:27 < TitanMKD> zyp because for other power source they enter in DC/DC converter protected against that ;)
2012-10-10T22:35:49 < zyp> and you have no other IO?
2012-10-10T22:36:13 < TitanMKD> in fact it is more common to have ESD problem on USB than on other IO ;)
2012-10-10T22:36:43 < TitanMKD> especially when you do not have direct contact with IO pin
2012-10-10T22:37:24 < zyp> but the boards I make is full of IO, and I don't see any reason why USB should deserve special treatment.
2012-10-10T22:38:02 < TitanMKD> hehe ;) there's just more probability on USB as it is more sensitive/critical
2012-10-10T22:38:36 -!- os-app92 [~ftw@12.170.195.77.rev.sfr.net] has joined ##stm32
2012-10-10T22:38:42 < karlp> hardly,
2012-10-10T22:38:54 < karlp> billions of usb devices out there with no protection at all.
2012-10-10T22:39:00 < TitanMKD> and IIRC standard IO are more robust they even give KVolt supported on some datasheet
2012-10-10T22:39:49 < TitanMKD> karlp yes maybe it is also a bit marketing to sell ESD chipset ;)
2012-10-10T22:40:16 < TitanMKD> but for debug/test board i just prefer to add a tiny USB ESD chipset ;)
2012-10-10T22:40:44 -!- os-app92 [~ftw@12.170.195.77.rev.sfr.net] has quit [Client Quit]
2012-10-10T22:44:03 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 276 seconds]
2012-10-10T22:45:20 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-10T22:46:25 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-10T22:49:50 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-10T22:53:30 < zippe> zyp: any I/O where you expect the user to connect/disconnect cables on any sort of regular basis should have ESD protection
2012-10-10T22:53:45 < zippe> zyp: this is why real products cost more money than "development boards"
2012-10-10T22:54:57 < zyp> agreed
2012-10-10T22:58:15 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-10T22:59:37 < emeb> the essence of good engineering is to build for purpose. Overdesigning a circuit is as much a failure as underdesigning.
2012-10-10T23:01:41 <+Steffanx> Overdesign but don't solder the components to it.. the big guys do it all the time :P
2012-10-10T23:01:54 <+Steffanx> *on it
2012-10-10T23:02:04 < Thorn> pcb houses will love you
2012-10-10T23:02:10 < emeb> Nothing wrong with that - if it doesn't cost you a lot extra for real-estate.
2012-10-10T23:02:12 < emeb> Building to standards worthy of flight hardware when it's a one-off for personal use shows that you don't understand the problem space and are just following rote rules of thumb.
2012-10-10T23:10:34 < Laurenceb_> usb cables are odd
2012-10-10T23:10:59 < Laurenceb_> they seem to build up static charges by themselves
2012-10-10T23:11:44 < Laurenceb_> sometimes i can plug a long usb cable into my datalogger with no host connected to the other end and it will boot up
2012-10-10T23:11:59 < Laurenceb_> due to high voltage static buildup on the cable
2012-10-10T23:12:27 < Laurenceb_> discharging into the wakeup circuit
2012-10-10T23:13:10 < Laurenceb_> im guessing its plastic sheeth on the wires rubbing against them
2012-10-10T23:13:23 < emeb> made out of cat fur + amber.
2012-10-10T23:16:25 < zippe> All cables do this
2012-10-10T23:16:56 < zippe> Also, many things connected to the other end of a USB cable (e.g. plastic laptops) will shift ground by tens or hundreds of volts.  ESD protection is essential in this case.
2012-10-10T23:19:47 < Laurenceb_> surely you cant fry the ground ?!
2012-10-10T23:20:06 < Laurenceb_>  - i mean on usb ground and power connect first
2012-10-10T23:21:06 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 264 seconds]
2012-10-10T23:22:31 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-10T23:26:45 < zippe> You can absolutely kill things if ground is level-shifted due to capacitive coupling inside the cable
2012-10-10T23:28:04 < zippe> On the device side, what matters most is keeping the signal:ground potential difference within the allowed range
2012-10-10T23:30:10 < Laurenceb_> hmm
2012-10-10T23:30:25 < Laurenceb_> wont the 22ohm resistors help protect the usb?
2012-10-10T23:32:45 < zyp> aren't ESD rather high voltages and low currents? the voltage drop across the diode won't be very significant
2012-10-10T23:32:53 < zyp> resistor*
2012-10-10T23:33:57 < Laurenceb_> yes, but the resistor will limit current
2012-10-10T23:34:13 < Laurenceb_> so preventing the esd protection diodes in the stm32 for blowing maybe
2012-10-10T23:35:37 < TitanMKD> 22ohm does not protect against ESD
2012-10-10T23:35:43 < TitanMKD> as it is more than 5KV
2012-10-10T23:37:31  * emeb used to design ESD pads for chips. watches this with amusement.
2012-10-10T23:37:39 < zyp> series resistors doesn't limit current, they drop voltage depending on current, but if the voltage drop is not significant, they don't make any difference
2012-10-10T23:37:49 < Laurenceb_> if 22ohm is greater than diode impedance its going to help
2012-10-10T23:39:20 < Laurenceb_> the problem is blowing the stm32 esd diode
2012-10-10T23:39:26 < TitanMKD> it is even recommended to add 45 Ohm resistor on D+ and D- terminated to GND
2012-10-10T23:39:58 < TitanMKD> it is only required for HS ;)
2012-10-10T23:41:45 -!- barthess [~barthess@178.154.94.18] has quit [Quit: Leaving.]
2012-10-10T23:45:17 < LeelooMinai> Does autoend work with >1 bytes in send mode?  I have nbytes=3 and the IC generates stop immediatelly after one byte is sent
2012-10-10T23:45:29 < LeelooMinai> (in I2C)
2012-10-10T23:59:30 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 264 seconds]
--- Day changed Thu Oct 11 2012
2012-10-11T00:00:43 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-11T00:18:17 < TitanMKD> does anyone tested -Os
2012-10-11T00:18:36 < TitanMKD> it is just amazing how the code is better with this option compared to -O2 or -O3
2012-10-11T00:19:01 < TitanMKD> it avoid lot of ref base addr to be loaded again and again
2012-10-11T00:19:02 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-11T00:19:17 < TitanMKD> the only drawback is it does not unroll loops ;)
2012-10-11T00:20:13 < zyp> -Os is -O2 without size-increasing optimizations
2012-10-11T00:20:46 < TitanMKD> yes maybe but the result is really amazing ;)
2012-10-11T00:21:11 < TitanMKD> just look that example
2012-10-11T00:21:12 < TitanMKD> https://gist.github.com/3868342
2012-10-11T00:21:59 < TitanMKD> ok O2 in the example is unrolled but it use tons of MOV to reload again and again the base addr
2012-10-11T00:22:32 < TitanMKD> and the MOV asm is a pseudo instruction in fact it use 2 MOV ;)
2012-10-11T00:22:44 < TitanMKD> like 41 F2 4C 13+                MOV     R3, 0x4010114C
2012-10-11T00:22:53 < TitanMKD> use 2 instructions
2012-10-11T00:23:40 < TitanMKD> and here an other example ;)
2012-10-11T00:23:41 < TitanMKD> https://gist.github.com/3868386
2012-10-11T00:23:54 < TitanMKD> where -O2 unroll the loop with ugly code
2012-10-11T00:24:18 < zyp> so, which is faster?
2012-10-11T00:24:34 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-11T00:25:44 < TitanMKD> i have not tested but in both case i will say -Os ;)
2012-10-11T00:26:07 < TitanMKD> even for the loop as it can be predicted very easily
2012-10-11T00:28:03 < TitanMKD> and i'm using arm-none-eabi-gcc.exe (GNU Tools for ARM Embedded Processors) 4.6.2 20120316 (release) [ARM/embedded-4_6-branch revision 185452]
2012-10-11T00:30:05 < emeb> I just tried -O2 vs -Os on my M4F audio app.
2012-10-11T00:30:27 < emeb> -O2 gives a binary that's 16k and with a processing loop that takes 180us
2012-10-11T00:30:29 < TitanMKD> and you have also better code for -Os ?
2012-10-11T00:30:44 < emeb> -Os gives a binary that's 13k and takes 168us
2012-10-11T00:30:55 < TitanMKD> hehe yes faster on -Os ;)
2012-10-11T00:31:13 < emeb> not much, but it could make a difference if things are tight.
2012-10-11T00:31:24 < TitanMKD> and for specialized stuff like audio as for my case it's more standard register access
2012-10-11T00:31:55 < TitanMKD> anyway for me -O2 is buggy on ARM ;)
2012-10-11T00:32:17 < TitanMKD> it should do like -Os and keep 1 reference as start for common addr
2012-10-11T00:32:38 <+dekar> what kind of bugs?
2012-10-11T00:32:58 < TitanMKD> dekar in fact it is just the code is badly optimized ;)
2012-10-11T00:33:04 < TitanMKD> dekar as -Os code is better
2012-10-11T00:33:28 < TitanMKD> dekar look here https://gist.github.com/3868342
2012-10-11T00:34:13 < TitanMKD> how it is possible to generate such ch***t in -O2 or -O3 it is same
2012-10-11T00:35:30 < zyp> TitanMKD, which compiler version are you using?
2012-10-11T00:36:18 < emeb> I ran my tests w/ gcc version 4.6.2 20120613
2012-10-11T00:36:21 < TitanMKD> i'm using arm-none-eabi-gcc.exe (GNU Tools for ARM Embedded Processors) 4.6.2 20120316 (release) [ARM/embedded-4_6-branch revision 185452]
2012-10-11T00:36:31 < emeb> same
2012-10-11T00:36:40 < TitanMKD> and it is just cheat on some case ;)
2012-10-11T00:36:54 < TitanMKD> I think I will keep -Os for all
2012-10-11T00:36:56 < emeb> actually - ARM/embedded-4_6-branch revision 188521
2012-10-11T00:37:03 <+dekar> I only use -Os anyway
2012-10-11T00:37:17 < TitanMKD> hehe
2012-10-11T00:37:35 < TitanMKD> I will have loved -Os code but with unroll loop ;)
2012-10-11T00:37:40 < TitanMKD> but it is not possible
2012-10-11T00:37:47 < TitanMKD> or also inline small function
2012-10-11T00:38:13 <+dekar> TitanMKD, just check which options -Os sets and add them manually
2012-10-11T00:38:29 < TitanMKD> dekar unroll loop just do nothing with -Os :(
2012-10-11T00:38:48 < TitanMKD> like -funroll-all-loops
2012-10-11T00:40:04 <+dekar> TitanMKD, I think -Os just disables some features of -O2, thus you can do that manually
2012-10-11T00:40:06 < zyp> hmm, interesting, compiling my code with -Os makes gcc complain about float abi incompatibilities with my newlib
2012-10-11T00:40:21 < zyp> that's a bit weird, I'm not using float anywhere in the code
2012-10-11T00:40:52 <+dekar> zyp, maybe sprintf anywhere?
2012-10-11T00:40:58 <+dekar> that pulls float in
2012-10-11T00:41:02 < zyp> no
2012-10-11T00:41:24 < zyp> I think the only newlib header I'm including at all is stdint.h
2012-10-11T00:41:56 <+dekar> have the linker generate a map and check it
2012-10-11T00:43:55 < TitanMKD> zyp add -mfloat-abi=hard
2012-10-11T00:45:29 <+dekar> TitanMKD, I actually think hardfloat is zyps problem
2012-10-11T00:45:35 < zyp> -Os makes the compiler emit calls to memset here and there in the code, so it's complaining about memset abi
2012-10-11T00:46:01 < zyp> /usr/local/yagarto-4.7.1/bin/../lib/gcc/arm-none-eabi/4.7.1/../../../../arm-none-eabi/bin/ld: error: demo.elf uses VFP register arguments, /usr/local/yagarto-4.7.1/bin/../lib/gcc/arm-none-eabi/4.7.1/../../../../arm-none-eabi/lib/thumb/v7m/libc.a(lib_a-memset.o) does not
2012-10-11T00:46:10 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-11T00:46:43 < zyp> hmm, actually, it seems like it's just a metadata conflict
2012-10-11T00:47:19 < TitanMKD> anyone know why when linking with lib function with 2 instruction(including bx lr) why this lib function is not inlined with -O2 or -O3 ?
2012-10-11T00:47:32 < TitanMKD> as it should be inlined
2012-10-11T00:47:37 < zyp> so it refuses because some objects are compiled with hardfloat ABI and some are compiled with softfloat ABI, even though I'm not calling any functions using float at all
2012-10-11T00:48:13 < zyp> TitanMKD, it can't be inlined if the compiler can't see the source
2012-10-11T00:48:43 < TitanMKD> I was thinking it could inline such simple case ;)
2012-10-11T00:49:17 < zyp> when you put a function in a .c file and call it from another .c file, the compiler doesn't know anything about the contents of the former when compiling the latter
2012-10-11T00:49:29 < zyp> it only knows the function signature you declare in the .h file
2012-10-11T00:49:55 < zyp> (and this is stuff that LTO is supposed to fix)
2012-10-11T00:52:15 < emeb> -Os makes the output from objdump -dS much more readable.
2012-10-11T00:52:30 < TitanMKD> emeb yes it's natural code ;)
2012-10-11T00:52:37 < TitanMKD> emeb not far from C ;)
2012-10-11T00:53:33 < emeb> I was trying to makes sense of the -dS output from -O2 a while back. Made me feel pretty dumb.
2012-10-11T00:54:11 < TitanMKD> emeb do not search -O2/-O3 is just like buggy code even if the code is fully functional it is ugly
2012-10-11T00:54:26 < TitanMKD> it is not even ordered ;)
2012-10-11T00:55:07 < emeb> sort of like looking at a layout from an autorouter. Makes no sense, but it works.
2012-10-11T00:55:48 < TitanMKD> i'm sure -Os is better to debug in step by step too
2012-10-11T00:55:54 < TitanMKD> as it is more close to C source
2012-10-11T00:56:35 < zyp> http://paste.jvnv.net/view/kPOG2 http://paste.jvnv.net/view/lRnPt
2012-10-11T00:56:54 < zyp> doesn't look that much more sane.
2012-10-11T00:57:13 < zyp> it still shows main() { twice :p
2012-10-11T00:58:14 < TitanMKD> yes -O2 step by step is pratically impossible ;
2012-10-11T00:58:40 < zyp> nah, it's fine
2012-10-11T00:58:54 < zyp> you just need to keep in mind that you are debugging instructions, not source lines.
2012-10-11T01:00:28 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-11T01:04:04 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-11T01:04:53 < zyp> http://paste.jvnv.net/view/lRnPt#line-242 <- and this looks awfully like an unrolled loop to me
2012-10-11T01:05:06 < zyp> in -Os, without any other special flags
2012-10-11T01:08:27 < zyp> oh, no, wtf
2012-10-11T01:10:09 < emeb> ?
2012-10-11T01:11:01 < zyp> I have no idea what that code is doing, but it doesn't look sane
2012-10-11T01:12:22 < emeb> I've often thought that looking at arm gcc output. I can only assume it knows things I don't.
2012-10-11T01:12:50 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-11T01:14:58 < zyp> hmm, it's generating objects that should be optimized out
2012-10-11T01:15:08 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Read error: Connection reset by peer]
2012-10-11T01:15:38 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-11T01:16:01 < emeb> being stupid in other words.
2012-10-11T01:16:16 < emeb> so much for machine intelligence.
2012-10-11T01:17:33 < zyp> it's doing shit with the Pin objects I'm using to abstract gpio
2012-10-11T01:19:14 < emeb> shuffling around data that has no effect in the real world?
2012-10-11T01:21:22 < zyp> it's emitting a lot of objects that's supposed to optimize out
2012-10-11T01:22:09 < emeb> too bad there's no such thing as "synthesis translate off" like we have in HDL.
2012-10-11T01:23:22 < emeb> I wonder if changing compiler versions would alter the results?
2012-10-11T01:24:22 < zyp> quite possibly
2012-10-11T01:32:33 < zyp> heh,
2012-10-11T01:32:45 < emeb> something changed?
2012-10-11T01:32:59 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-10-11T01:34:59 < zyp> yep, gcc is being weird
2012-10-11T01:35:08 < emeb> *news flash*
2012-10-11T01:35:56 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-11T01:36:28 < zyp> it doesn't optimize out the transparent constructor of the Pin object at compile time like it should
2012-10-11T01:37:04 < zyp> by simply removing it (and marking the members as public) it's emitting static objects in .data instead
2012-10-11T01:37:07 < zyp> and that's fine
2012-10-11T01:37:47 < emeb> as long as you've got memory to waste.
2012-10-11T01:37:51 < zyp> but without the constructor it can't have private members, and that's not fine
2012-10-11T01:38:55 < zyp> and either way, the objects should be optimized out so they don't get emitted at all
2012-10-11T01:39:05 < zyp> this worked when I first wrote that code
2012-10-11T01:40:02 < emeb> have you changed that code at all since the initial testing?
2012-10-11T01:40:30 < zyp> no
2012-10-11T01:41:11 < emeb> is it the compiler version, or is it other objects for additional peripherals that might change the overall behavior?
2012-10-11T01:41:52 < zyp> might be compiler version, might be that I'm now using C++11 mode
2012-10-11T01:43:08 < zyp> I need to make a reduced testcase of this tomorrow and track down why it doesn't optimize the code like it should
2012-10-11T01:43:43 < emeb> been there. big fun.
2012-10-11T01:58:51 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-11T01:59:18 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 245 seconds]
2012-10-11T02:22:38 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-11T02:24:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-11T02:28:22 < emeb> this is cool: http://dangerousprototypes.com/forum/viewtopic.php?f=2&t=4622&p=45011#p45000
2012-10-11T02:29:00 < emeb> like that he generates the HV with a boost converter that's controlled by the MCU.
2012-10-11T02:41:12 -!- DaKu is now known as daku
2012-10-11T02:47:32 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-11T02:55:01 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-11T03:00:00 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-251-238.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-11T03:22:18 < emeb> dongs: http://www.logos-electro.com/rgb-led-matrix-backpack-1
2012-10-11T03:27:04 <+dekar> I made a small example project: https://github.com/EliasOenal/TNT_Example
2012-10-11T03:27:11 <+dekar> should work for all GCC toolchains
2012-10-11T03:27:53 <+dekar> the F1 discovery probably needs some ifdefs for the correct LED pins though, F4 works fine
2012-10-11T03:28:36 < dongs> emeb: uh, arduino?
2012-10-11T03:29:12 < dongs> wait even worse, a fucking tiny2313
2012-10-11T03:29:26 < dongs> and what the fuck is that driver chip
2012-10-11T03:29:55 < dongs> haha is that i2c io expander
2012-10-11T03:30:06 < emeb> right - presented for amusement.
2012-10-11T03:31:42 < emeb> tiny2313 are cheap though - got a bunch from BGMicro for less than $1 a while back.
2012-10-11T03:32:08 < dongs> oh no, the tqfp is an actual led driver, but a real shitty one
2012-10-11T03:32:37 < dongs> http://arduino.cc/playground/Main/HT1632C
2012-10-11T03:33:35 < dongs> like 16grayscale lol.
2012-10-11T03:57:37 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-11T04:39:20 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-11T05:03:08 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-11T05:21:38 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 255 seconds]
2012-10-11T05:48:52 < emeb_mac> <crickets>
2012-10-11T05:52:07 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 260 seconds]
2012-10-11T05:54:40 < dongs> yeah
2012-10-11T05:54:42 < dongs> no chats at all
2012-10-11T05:54:51 < dongs> blame flyback
2012-10-11T06:04:47 -!- cjbaird_ [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-11T06:07:47 < cjbaird_> Seems like all the bandwidth from Mars at the moment is being used to send a 'video' of the scooper in action from Sol 61..
2012-10-11T06:08:49 < dongs> did that stratos guy jump yet
2012-10-11T06:09:05 < emeb_mac> haven't heard anything
2012-10-11T06:09:28 < dongs> holy fuck delayed for 3 more days
2012-10-11T06:09:37 < cjbaird_> Red Bull: The energy drink of choice for Flake No-shows!
2012-10-11T06:10:32 < cjbaird_> Red Bull: Naaa, mate; can't be bothered doing it today...
2012-10-11T06:11:31 < emeb_mac> Red Bull: Wind? Run away!!!!
2012-10-11T06:13:24 -!- RL [~RL@114-38-84-34.dynamic.hinet.net] has joined ##stm32
2012-10-11T06:15:56 -!- RL [~RL@114-38-84-34.dynamic.hinet.net] has quit [Client Quit]
2012-10-11T06:16:13 < dongs> chinese hackers
2012-10-11T06:19:35 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-11T06:31:43 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-11T06:31:48 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-11T06:40:36 < dongs> time to dicknplace more shit
2012-10-11T06:43:56 < dongs> noone, i paypal'd it
2012-10-11T06:44:24 < dongs> in several installments since shitpal doesnt allow sending >$10k at a time LOL
2012-10-11T06:46:18 < HTT-Bird> dongs: I think its some goofy piece of US treasury dept. anti-money-laundering paperwork...
2012-10-11T06:46:34 < HTT-Bird> but yeah, paypal is lazy and doesn't want to deal with that
2012-10-11T06:48:56 < dongs> yeah cuz you really fucking need one.
2012-10-11T06:48:57 < dongs> NOT.
2012-10-11T06:49:16 < dongs> ?!
2012-10-11T06:49:25 < dongs> all more useless than current u.s. president
2012-10-11T06:53:16 < dongs> flyback: here's a full cup. now shut the full cup.
2012-10-11T06:56:26 < emeb_mac> people who worship musical keyboards are hilarious
2012-10-11T06:56:55 < emeb_mac> they gotta have just that particular Moog/Arp/Roland/etc
2012-10-11T06:57:33 < emeb_mac> meanwhile we can may new stuff that outperforms the old. But they worship the old so they pay thru the nose to maintain it.
2012-10-11T06:58:46 < cjbaird_> Fairlights were Fucking Awesome[tm].
2012-10-11T06:59:42 < emeb_mac> emphasis on "were".
2012-10-11T07:00:23 < emeb_mac> but TBH, we all lusted after them back in the day.
2012-10-11T07:00:45 < ossifrage> anyone familiar with stm32 one pulse mode?
2012-10-11T07:04:48 < dongs> coblogger is workin on F373 component so i dont have to ~ http://i.imgur.com/HMqCv.jpg
2012-10-11T07:05:35 < emeb_mac> err - that's hard? Takes me about 10 minutes to make a symbol.
2012-10-11T07:05:49 < dongs> well thats 10 minutes saved for me
2012-10-11T07:05:52 < dongs> no its not hard at all
2012-10-11T07:06:22 < dongs> infact in dicktrace is probably easier than in any other pcb cad
2012-10-11T07:06:27 < emeb_mac> esp if you've got any other STM32 symbol
2012-10-11T07:06:38 < emeb_mac> edit a few pin names. done.
2012-10-11T07:06:43 < dongs> ya
2012-10-11T07:07:00 < dongs> but im loading dicknplace so thats more important ;d
2012-10-11T07:07:15 < emeb_mac> so, what you going to do with a 373 that you can't do with others?
2012-10-11T07:07:53 < dongs> he needs it for fast(er) adc and comparators
2012-10-11T07:08:03 < dongs> current sense stuff in like 600W led drivers
2012-10-11T07:08:04 < emeb_mac> there's that.
2012-10-11T07:08:30 < dongs> his current shit runs off STM8, barely haha.
2012-10-11T07:08:40 < emeb_mac> ssssslooooowwww.
2012-10-11T07:08:41 < dongs> still faster ADC than avrtrash though.
2012-10-11T07:08:56 < Thorn> why do you need a fast adc for that?
2012-10-11T07:09:09 < dongs> because theres no comparator on stm8
2012-10-11T07:09:13 < dongs> so he uses analog watchdog
2012-10-11T07:09:31 < dongs> but still, thats a bit too slow. it works, but not optimal
2012-10-11T07:10:00 < dongs> duno not my proj i m just bloggin about it
2012-10-11T07:10:07 < emeb_mac> nice thing about f373 is 16-bit ADCs. Would be interesting to see how many real bits those have.
2012-10-11T07:11:39 < dongs> stm8 is $1/250 tray
2012-10-11T07:11:53 < dongs> for qfp32
2012-10-11T07:12:14 < dongs> no, 1/ea per tray of 250, fuckwit
2012-10-11T07:13:04 < dongs> im sure they are, but with f103 being $2 it doesnt matter.
2012-10-11T07:13:45 < dongs> i cant believe ST moved 2 fucking pins on F373 just to make it not pin compatible
2012-10-11T07:13:52 < dongs> fucking assholes
2012-10-11T07:14:36 < emeb_mac> odd - wonder what the motivation for that was.
2012-10-11T07:15:52 < dongs> looks like 9221s are ready to be shipped to me
2012-10-11T07:16:00 < dongs> but im waiting for some other shit to arrive tehre so they can send it at once
2012-10-11T07:16:06 < emeb_mac> nice
2012-10-11T07:16:34 < emeb_mac> led arrays still MIA?
2012-10-11T07:18:13 < dongs> yep
2012-10-11T07:18:19 < dongs> guy swears he sent them
2012-10-11T07:20:02 < emeb_mac> mysterious mail
2012-10-11T07:20:52 < dongs> Departure from outward office of exchange
2012-10-11T07:20:52 < dongs> Oct-09-12, 11:52 AM, SHANGHAI
2012-10-11T07:20:55 < dongs> jesus christ
2012-10-11T07:21:03 < dongs> posted: sept 28
2012-10-11T07:21:39 < dongs> was your shit shipped by RAxxxCN?
2012-10-11T07:21:42 < dongs> i.e. registered airmail
2012-10-11T07:22:05 < dongs> at least that means im geting teh trahs sometime  this week
2012-10-11T07:22:14 < dongs> once its in jp its <= 3 days to me.
2012-10-11T07:22:17 < dongs> depending on where it lands.
2012-10-11T07:24:59 < dongs> hm need another reel of 10uf shit, gonna run out in a week
2012-10-11T07:25:02 < emeb_mac> ordered them on 9-26
2012-10-11T07:25:48 < emeb_mac> Processed Through Sort Facility
2012-10-11T07:25:49 < emeb_mac> Sep-28-12, 20:51 PM, NANJING
2012-10-11T07:26:09 < emeb_mac> Delivered Oct-05-12
2012-10-11T07:26:14 < dongs> fuu
2012-10-11T07:26:51 < emeb_mac> looks like china mail was sloooow for yours.
2012-10-11T07:40:15 < dongs> if i paid anythign for shipping, ill be getting arefund for sure
2012-10-11T07:40:27 < dongs> if I order more than 10 from him next time ill just contact him off ebay and get him to pay for SF express
2012-10-11T07:40:30 < dongs> that shit is here in 2 days
2012-10-11T07:40:33 < dongs> for liek 30 bucks
2012-10-11T08:11:50 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-11T08:11:50 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-11T08:11:50 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-11T08:38:49 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-11T08:42:41 < dongs> commence chats
2012-10-11T09:01:45 < cjbaird_> commencing trolling of the local raspberry pi 'meetup'...
2012-10-11T09:03:43 < cjbaird_> damage done: http://www.ozberrypi.org/events/78094552/
2012-10-11T09:03:55 < dongs> haha.
2012-10-11T09:05:11 < dongs> haha are those shitberrypies
2012-10-11T09:05:14 < dongs> what the hell are you doing wiht htem
2012-10-11T09:06:14 < cjbaird_> I've noticed 4 other people have bailed on the meet in the past 2 days.
2012-10-11T09:06:27 < cjbaird_> (since the non-Rpi stuff was posted)
2012-10-11T09:06:46 < dongs> I'd also be happy to talk a bit about getting started with Arduino and perhaps also the gcc-avr toolchain (basic).
2012-10-11T09:07:18 < cjbaird_> If I /were/ going, I'd give a run-down on the 6-wire jtag work..
2012-10-11T09:07:42 < dongs> haha
2012-10-11T09:07:43 < cjbaird_> And bully people who bought a Gertboard, if any. (Haven't seen anyone..)
2012-10-11T09:08:00 < dongs> is gertboard that trash that was some protopcb on top of rpi?
2012-10-11T09:08:18 < dongs> ya looks like it
2012-10-11T09:08:20 < cjbaird_> The stuff I've gone over before about using the Rpi as a devel+programmer, rather than an embedded board in itself.
2012-10-11T09:08:39 < dongs> http://www.raspberrypi.org/wp-content/uploads/2012/08/image002.png ???????????????????
2012-10-11T09:08:54 < cjbaird_> ..and costs almost 3 times more than the RPi itself. And you have to do SMD assembly yourself. "Gee, someone thought that through.."
2012-10-11T09:10:35 < dongs> looks utterly useless to me
2012-10-11T09:10:46 < dongs> they did get rid of all smd parts from it in later revision it seems
2012-10-11T09:11:16 < dongs> but hm gertboard wasnt waht i was thinking of
2012-10-11T09:11:19 < dongs> there wasw something even worse looking
2012-10-11T09:12:22 < dongs> http://1.bp.blogspot.com/-QDzbWm0V67w/UDnevEuIByI/AAAAAAAABFM/rPR9YGShH08/s1600/HumblePi.jpg something along the lines of this
2012-10-11T09:12:55 < dongs> http://www.amazon.com/Programming-Raspberry-Pi-Getting-Started/dp/0071807837/ WHY???
2012-10-11T09:18:34 < cjbaird_> "Build professional-quality GUIs using Tkinter" -- hey, lets recycle that book you wrote back in 1994!
2012-10-11T09:20:59 < emeb_mac> rpi - meh.
2012-10-11T09:21:14 < emeb_mac> hype-pi
2012-10-11T09:21:44 -!- claude is now known as Claude
2012-10-11T09:22:56 < cjbaird_> I wonder if they'll get busted for trying to use a charity for 'unfair commercial advantage'..
2012-10-11T09:26:15 < emeb_mac> doubt it. what is "charity" in the UK anyway?
2012-10-11T09:29:34 < emeb_mac> hawt.
2012-10-11T10:01:43 -!- tavish3 [~tavish@120.59.43.197] has joined ##stm32
2012-10-11T10:02:50 -!- Claude is now known as claude
2012-10-11T10:14:21 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-11T10:15:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-11T10:38:09 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-11T10:50:34 < dongs> funny level: dr who
2012-10-11T10:50:37 < dongs> (which means FUCKING UNFNNY)
2012-10-11T10:51:00 < zyp> ok
2012-10-11T10:53:20 < dongs> what the hell is it with all teh 3d printer trash on kickstarter
2012-10-11T10:54:36 < Laurenceb_> kawaii desu
2012-10-11T10:55:18 < Laurenceb_>  	Weabo
2012-10-11T10:55:49 < dongs> http://printrbot.com/wp-content/uploads/2012/03/IMG_1089.jpg IS THIS REALLY MADE OUT OF WOOD
2012-10-11T10:56:12 < Laurenceb_> lol dr who
2012-10-11T10:56:44 < Laurenceb_> lol @ printerbot
2012-10-11T10:56:53 < dongs> $699 for that shit.
2012-10-11T10:56:55 < Laurenceb_> someone needs to learn 2 tirangulation
2012-10-11T10:57:13 < Laurenceb_> its going to shake everywhere
2012-10-11T10:58:06 < Laurenceb_> bbl
2012-10-11T10:58:13 < dongs> LEARNM TO PRIOERLY /QUIT FUCKER
2012-10-11T11:02:24 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-11T11:02:38 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-11T11:16:13 < dongs> http://www.youtube.com/watch?v=EPNI-51j008#t=8s lol
2012-10-11T11:29:57 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-11T11:30:56 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 256 seconds]
2012-10-11T11:31:08 < dongs> http://i.imgur.com/fXvLa.png ordering
2012-10-11T11:38:57 < Tectu> dongle, very bad choise
2012-10-11T11:38:59 < Tectu> choice*
2012-10-11T11:39:28 < Tectu> well, after I couldn't find any printer out there which is good, i decided to build mine up from scratch and it works very very well
2012-10-11T11:39:32 < Tectu> no comparing to reprap etc
2012-10-11T11:39:42 < Tectu> and your printrbot has big problems with the Z axis
2012-10-11T11:40:07 < Tectu> also, why the hell do you order 99 ?
2012-10-11T11:40:13 < dongs> 999.
2012-10-11T11:40:25 < dongs> im not buying any 3d print trash
2012-10-11T11:40:56 < dongs> its hilarious the shit actually allowed me to put 999 in a cart AND calculated shipping on it.
2012-10-11T11:41:01 < Tectu> how comes that it only costs 80$ to ship
2012-10-11T11:41:12 < Tectu> oh, wait, it's 80'000
2012-10-11T11:41:21 < Tectu> well, here,   ,  is decimal point -.-
2012-10-11T11:41:43 < dongs> haha
2012-10-11T11:57:16 -!- tavish3 [~tavish@120.59.43.197] has quit [Ping timeout: 246 seconds]
2012-10-11T11:59:15 -!- tavish3 [~tavish@120.56.169.199] has joined ##stm32
2012-10-11T12:01:00 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-11T12:01:00 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-11T12:01:00 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-11T12:01:07 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away]
2012-10-11T12:01:29 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-11T12:04:18 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Client Quit]
2012-10-11T12:04:30 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-11T12:29:26 < cjbaird_> Adventure level: riding a motorbike in the rain without wet-weather gear. :/
2012-10-11T12:29:40 < tavish3> i know that a bus fault precise error occured, and SCB->BFAR is 0x20020000. Any way to know where this might have occured?
2012-10-11T12:32:47 < zyp> cjbaird_, also known as swimming
2012-10-11T12:33:35 < zyp> tavish3, at the code trying to access address 0x20020000
2012-10-11T12:34:41 < tavish3> zyp: :P ofcourse
2012-10-11T12:34:58 < zyp> 0x20020000 is start of ram + 128k
2012-10-11T12:35:28 < zyp> is this F4?
2012-10-11T12:35:31 < tavish3> yeah
2012-10-11T12:35:43 < zyp> if so, it's just at the end of ram
2012-10-11T12:36:08 < zyp> trying to pop too much off the stack?
2012-10-11T12:36:36 < tavish3> but i think it has 192KB ram
2012-10-11T12:36:39 < dongs> uh
2012-10-11T12:36:40 < zyp> no
2012-10-11T12:36:41 < dongs> you think wrong
2012-10-11T12:36:43 < dongs> its 128k+64k ccm
2012-10-11T12:36:49 < zyp> exactly
2012-10-11T12:36:50 < dongs> which is not even continuous
2012-10-11T12:36:56 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-11T12:37:00 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-11T12:37:04 < zyp> there is a 128k block at 0x20000000 and a 64k block at 0x10000000
2012-10-11T12:37:09 < tavish3> ohh
2012-10-11T12:37:41 < dongs> also, while youre listening, you can't dma to/from 64k.
2012-10-11T12:37:44 < zyp> so 0x20020000 is just after the end of the 128k block
2012-10-11T12:39:30 < tavish3> didnt know that, thanks.
2012-10-11T12:54:55 < Laurenceb> http://pastebin.com/QFVdgpax
2012-10-11T12:56:05 < dongs> hahaha.
2012-10-11T12:56:14 < dongs> tectu being kawaii
2012-10-11T12:56:17 < dongs> what channel is this in
2012-10-11T12:56:19 < dongs> this one?
2012-10-11T12:56:56 <+Steffanx> Here is no one with the nick mobysomething
2012-10-11T12:57:02 < dongs> right
2012-10-11T12:57:06 < dongs> probably #chibios
2012-10-11T12:57:08 < dongs> or some other faggotry
2012-10-11T12:57:22 < Laurenceb> yes
2012-10-11T12:57:25 <+Steffanx> Looks like it
2012-10-11T12:57:27 <+Steffanx> oh, yes
2012-10-11T12:58:00 < Laurenceb> #chibios
2012-10-11T13:00:10 < Laurenceb> hes raging in a PM now
2012-10-11T13:00:22 <+Steffanx> Tectu you failed
2012-10-11T13:00:36 < karlp> remember, we only tease you because we love you :)
2012-10-11T13:04:53 < cjbaird_> You're never Forever Alone if people hang around to make fun of you!
2012-10-11T13:04:57 -!- cjbaird_ is now known as cjbaird
2012-10-11T13:05:21 < dongs> i want those pics
2012-10-11T13:06:09 <+Steffanx> Google: kawaii girl with scars
2012-10-11T13:06:21 <+Steffanx> fs
2012-10-11T13:06:24 < dongs> going to images.google.com asap
2012-10-11T13:06:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-11T13:06:46 < dongs> http://masamania.com/wp-content/uploads/300x400xwrist_cut099.jpg.pagespeed.ic.BD3LryxeSf.jpg
2012-10-11T13:07:16 < Laurenceb> dongs idea of a hot girl
2012-10-11T13:07:20 < cjbaird> She applying to blog for you?
2012-10-11T13:08:07 < dongs> http://topjpteen.com/kawaiigirls/?p=4572 haha
2012-10-11T13:08:50 < Thorn> People prefer a badly designed, expressiveless, macro assembler, "hacker" language like C because otherwise they feel their algorithmic creativity restrained. the real reason is that they never learnt to write good code, and were only taught the "hacker" part of computing.
2012-10-11T13:08:50 < Thorn> People use the right tool for the right job, but when their definition of good is bad, they end up using the bad tool.
2012-10-11T13:09:21 < dongs> Thorn: sounds like this trash was written by some python or php "programmer"
2012-10-11T13:09:27 < Laurenceb> lolwut
2012-10-11T13:09:37 < cjbaird> LISP weenie, more like.
2012-10-11T13:09:40 < Laurenceb> python jerkcirlce
2012-10-11T13:10:20 < Laurenceb> " definition of good is bad"
2012-10-11T13:10:30 < Laurenceb> yeah they need to be reeducated
2012-10-11T13:10:41 < Laurenceb> with nerve gas
2012-10-11T13:18:16 < Laurenceb> http://i3.kym-cdn.com/photos/images/newsfeed/000/067/554/tumblr_l7q3a8ppiy1qdolauo1_500.jpg
2012-10-11T13:19:55 <+Steffanx> LOL
2012-10-11T13:24:16 <+Steffanx> Laurenceb likes those "no GF" pictures ...
2012-10-11T13:26:23 < Laurenceb> http://weknowmemes.com/wp-content/uploads/2012/08/forever-alone-brofist.jpg
2012-10-11T13:34:13 -!- BrainDamage [~BrainDama@62.10.7.162] has joined ##stm32
2012-10-11T13:37:26 < dongs> lo'
2012-10-11T13:54:14 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-11T14:18:14 <+Steffanx> Man, i guess i'm blacklisted
2012-10-11T14:18:21 <+Steffanx> Don't receive my small chinese crap anymore
2012-10-11T14:20:07 < zyp> oh no
2012-10-11T14:20:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-11T14:37:52 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-10-11T14:42:15 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Read error: Connection reset by peer]
2012-10-11T14:42:57 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-11T14:43:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-11T14:52:27 -!- dekar_ [~dekar@212.255.33.22] has joined ##stm32
2012-10-11T14:52:30 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-11T14:54:41 -!- dekar [~dekar@212.255.19.185] has quit [Ping timeout: 244 seconds]
2012-10-11T14:58:49 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-11T15:11:43 -!- claude is now known as claude_work
2012-10-11T15:27:51 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has joined ##stm32
2012-10-11T15:29:03 < Abhishek_> Hi everyone
2012-10-11T15:29:16 <+Steffanx> Lo
2012-10-11T15:30:23 < Abhishek_> Is it possible to run OpenCV on STM32F4?
2012-10-11T15:30:59 < Abhishek_> with a camera module?
2012-10-11T15:31:03 < dongs> depends what you wanna CV.
2012-10-11T15:31:35 < dongs> but last i checked, cuz its called "open" it means its written in C++, ugly as shit code, super bloated, probably needs a gig of ram just to detect a feature.
2012-10-11T15:31:49 < Abhishek_> for basic shape detection, edge detection, etc?
2012-10-11T15:32:19 < dongs> i think you'd want to at least buffer the video image in sram for that...
2012-10-11T15:32:25 < dongs> and there's only 128k of that available.
2012-10-11T15:33:15 < Abhishek_> um, yes
2012-10-11T15:33:18 < zyp> unless you throw on some external ram
2012-10-11T15:33:39 < Abhishek_> yeah. That was exactly I was thinking
2012-10-11T15:36:16 < dongs> at that point you might as well just buy a hypeberrypi, install lunix, and knock yourself out
2012-10-11T15:36:37 < ReggieUK> good luck with that, depending on the choice of camera
2012-10-11T15:36:58 < dongs> cuz USB on hypeberry sucks?
2012-10-11T15:37:24 < ReggieUK> donkey balls through carbon nanotubes
2012-10-11T15:37:53 < ReggieUK> it's had various fixes so it's much better than it was
2012-10-11T15:38:10 < ReggieUK> but it uses the same dwc_otg usb IP that is on the stm boards apparently
2012-10-11T15:38:24 < ReggieUK> I died inside a little bit when I found that out
2012-10-11T15:38:34 < dongs> oh right  i remember someone in here complaining about it.
2012-10-11T15:38:38 < dongs> that it needs more realtime than lunix can provide
2012-10-11T15:38:50 < dongs> becausae the core needs callback to cpu that has to reply fast or something.
2012-10-11T15:38:57 < ReggieUK> 8000 interrupts per second
2012-10-11T15:39:41 < dongs> funtimes.
2012-10-11T15:39:50 < ReggieUK> some stuff plays nicely, some stuff doesn't
2012-10-11T15:40:05 < Laurenceb> look who i found
2012-10-11T15:40:06 < zyp> it's indeed the same usb core
2012-10-11T15:40:06 < Laurenceb> https://picasaweb.google.com/115325807780508608997/ProfilePhotos
2012-10-11T15:40:14 < zyp> it's apparently used in a lot of stuff
2012-10-11T15:40:35 < ReggieUK> don't get me wrong, I'm not criticising the pi as such, it is what it is and it does what it's meant to do for the most part and *everyone* was aware that it was essentially a beta board when it came out
2012-10-11T15:40:45 < ReggieUK> if they didn't, shame on them for not reading the small print
2012-10-11T15:41:09 < BrainDamage> the small?
2012-10-11T15:41:28 < BrainDamage> 27pt font small print
2012-10-11T15:41:34 < ReggieUK> 5pt
2012-10-11T15:41:51 < ReggieUK> small print was probably the wrong choice of words
2012-10-11T15:41:58 -!- tavish3 [~tavish@120.56.169.199] has quit [Quit: Lost terminal]
2012-10-11T15:42:59 < Laurenceb> https://picasaweb.google.com/115325807780508608997/ProfilePhotos
2012-10-11T15:43:04 < Laurenceb> ^ is that Tectu?
2012-10-11T15:43:08 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has left ##stm32 []
2012-10-11T15:43:37 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has joined ##stm32
2012-10-11T15:44:45 < dongs> not kawaii enough
2012-10-11T15:45:31 <+Steffanx> Seriously, Laurenceb WTF do you browse all day long?
2012-10-11T15:45:38 < dongs> ha ha
2012-10-11T15:45:38 < Laurenceb> apparently it is
2012-10-11T15:45:49 < Laurenceb> i think i found him
2012-10-11T15:46:15 < zyp> ReggieUK, you may not like to hear this, but apparently dwc_otg is also used in the OMAP chips
2012-10-11T15:47:04 < Laurenceb> teeheee
2012-10-11T15:47:36 < dongs> Laurenceb: breaking news
2012-10-11T15:47:37 < dongs> <tom66> Well, had a problem this summer. :/ Btw, I'm now at the University of Leeds doing my MEng Electronics Engineering. I was going to make it at BitBox in my last two weeks but the reflow oven died - the clutch broke on the drive belt. They got it working again luckily but there was too little time left with their back log needing to be done, of course.  I'm going to see if I've got enough time to do it over winter.
2012-10-11T15:48:20 < dongs> <timecop> you should call up bitbox and ask if they fired any assburgers recently
2012-10-11T15:48:23 < dongs> <causality> i'd rather call them and say some child on the internet is making them look bad
2012-10-11T15:48:26 < dongs> <causality> and that i was going to use them until i read his junk
2012-10-11T15:48:39 < Laurenceb> lol
2012-10-11T15:48:41 < Laurenceb> http://s1.b3ta.com/host/creative/13/1349814329/shortcircuitoffencepolice.jpg
2012-10-11T15:49:48 < dongs> that film is available on BD. they could have at least used HD screenshots for that shit, Laurenceb
2012-10-11T15:50:06 < Laurenceb> heh
2012-10-11T15:50:25 < dongs> also i'm also curiosu  to know what the fuck do you browse all day
2012-10-11T15:50:32 < dongs> Tectu: why is Laurenceb e-stalking you
2012-10-11T15:50:54 < Laurenceb> cuz he needs reap repayment
2012-10-11T15:51:03 < Laurenceb> conservation of rape
2012-10-11T15:51:34 < dongs> openscad more like openscat
2012-10-11T15:51:54 < dongs> im glad none of my real info is on the internets
2012-10-11T15:52:02 < dongs> or else fags like Laurenceb would be pasting funnies on irc
2012-10-11T15:53:44 < Tectu> because I'm sexy and he know it
2012-10-11T15:54:05 < dongs> saving tectu's piccies to d:\pron
2012-10-11T15:54:32 < zyp> oh, wait, looks like OMAP4 is using Mentor IP for the OTG controller
2012-10-11T15:55:38 < zyp> and then some host-only synposys stuff as well
2012-10-11T15:58:39 < cjbaird> omg, just google'd a pic of an old ex-gf (terrible photo, though).
2012-10-11T16:00:27 < Laurenceb> http://www.b3tards.com/u/d03ec72263e23ed4c51e/twathat.gif
2012-10-11T16:00:32 < Laurenceb> ^ found it already
2012-10-11T16:03:35 < dongs> why hte fuck linkedin doesn't have a 'trolling' industry
2012-10-11T16:03:40 < Laurenceb> hehe
2012-10-11T16:04:15 < zyp> do you need to hire more trolls?
2012-10-11T16:04:24 < dongs> yes
2012-10-11T16:06:36 < dongs> entertainment it is.
2012-10-11T16:07:26 < dongs> i need zipcode for gaza strip
2012-10-11T16:07:47 < dongs> 00972 it is.
2012-10-11T16:07:57 < Laurenceb> lulwut
2012-10-11T16:08:15 < Laurenceb> launching drones again?
2012-10-11T16:24:27 < BrainDamage> http://www.funroll-loops.info
2012-10-11T16:28:39 < dongs> so fucking old
2012-10-11T16:30:41 < dongs> The single biggest performance boost I have yet to see-now I can have gnome2.2 running using gnome-terminal to compile the latest j2sdk from source (nice -n 19)while browsing with mozilla while running e17 in a seperate login with two eterms and run Unreal Tournament at full speed (this with an apache webserver running for my dyndns pseudo-domain and a mysql for my answering-machine software for my isdn card-which keeps track of all incomming phonec
2012-10-11T16:30:48 < dongs> fuck yeah
2012-10-11T16:30:53 < Thorn> anybody used NAND flash with f4 yet?
2012-10-11T16:31:03 < dongs> i've used it with F2
2012-10-11T16:31:08 < dongs> finally got the shit to read
2012-10-11T16:31:13 < dongs> after dicking with fsmc junk for like a week
2012-10-11T16:31:23 < Thorn> wat.
2012-10-11T16:31:33 < dongs> you know, stm32F2
2012-10-11T16:32:00 < Thorn> I got that. why was it so hard to make work? timing problems?
2012-10-11T16:32:16 < dongs> yes
2012-10-11T16:32:28 < dongs> i had no idea what im doing so i just kept trying rnadom combinations for the 3 or 4 variable timing shits
2012-10-11T16:33:17 < Thorn> keil boards with nand should come with examples on how to configure the fsmc
2012-10-11T16:34:10 < dongs> i made a nand socket and was using it with teh waveshare board i got
2012-10-11T16:36:00 < Thorn> what read speed did you get?
2012-10-11T16:36:36 < dongs> fucking terrible
2012-10-11T16:36:51 < dongs> took overnight to dump like 128meg nand and it wasnt even done
2012-10-11T16:37:06 < zyp> I also got a nand board with my waveshare board, but I haven't had time to play with it yet
2012-10-11T16:39:05 < dongs> http://www.linkedin.com/skills/skill/IRC
2012-10-11T16:39:23 < karlp> dongs: what's this "BD" place you keep talking about?
2012-10-11T16:39:31 < dongs> karlp: ?
2012-10-11T16:39:42 < dongs> karlp: referring to shortcircuit?
2012-10-11T16:40:10 < dongs> shortcircuit (1986) is available on bluray.
2012-10-11T16:40:17 < dongs> since like 2008.
2012-10-11T16:41:13 < dongs> http://i.imgur.com/N3uav.jpg doesnt look bad either.
2012-10-11T16:41:56 <+Steffanx> yes it does
2012-10-11T16:44:20 < Abhishek_> I guess you require a wear-levelling algo with NAND flash, right?
2012-10-11T16:45:18 < Thorn> if it's parallel, you do
2012-10-11T16:45:25 < karlp> is BD Blueray Disc for you?
2012-10-11T16:45:33 < Thorn> serial flash has a controller afaik
2012-10-11T16:45:51 < dongs> karlp: yea?
2012-10-11T16:46:02 < karlp> ok :)
2012-10-11T16:46:04 < dongs> karlp: isnt it anywhere else?
2012-10-11T16:46:14  * karlp shrugs
2012-10-11T16:46:22 < dongs> i mean the stack of bluray recordable disks on my shelf are labeeled BD-R
2012-10-11T16:46:33 < karlp> piratebay often has BRRIP or something.
2012-10-11T16:46:43 < BrainDamage> stop highlighting me :(
2012-10-11T16:46:52 < karlp> when you've talked about seomthing "being available on BD" I wasn
2012-10-11T16:46:53 < BrainDamage> I think you mean BDRIP
2012-10-11T16:46:56 < karlp> t sure if that was a place
2012-10-11T16:47:03 < karlp> or a thing.
2012-10-11T16:47:11 < karlp> get better highlighting BD :)
2012-10-11T16:50:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-11T16:50:53 < Thorn> To access these documents, you must first read and accept Micron's electronic non-disclosure agreement.
2012-10-11T16:51:29 < Abhishek_> oh
2012-10-11T16:51:35 < Abhishek_> is that chargeable?
2012-10-11T16:54:30 < Thorn> I clicked accept and it loaded the datasheet.
2012-10-11T16:54:47 < Thorn> apparently 4Gb is 4 gigabytes, not 4 gigabits.
2012-10-11T16:55:09 < zyp> really?
2012-10-11T16:55:29 < zyp> micron sounds like ram, ram chips is usually measured in bits
2012-10-11T16:56:02 < Thorn> no, it's bits. I calculated size for a 32Gb device and got 4.3 gigabytes
2012-10-11T16:56:15 < Thorn> looking at nand flash atm.
2012-10-11T16:56:45 -!- Abhishek_ [cb6ef7dd@gateway/web/freenode/ip.203.110.247.221] has quit [Quit: Page closed]
2012-10-11T16:58:56 < Thorn> they only use like 20 of 56 pins?
2012-10-11T17:01:15 < dongs> yes
2012-10-11T17:01:25 < dongs> and luckily a stanadrd pinout
2012-10-11T17:01:31 < Thorn> or 48
2012-10-11T17:04:17 < Laurenceb> RAGEEEEE
2012-10-11T17:04:39 < Laurenceb> so i order leds from roithner laser
2012-10-11T17:04:48 < Laurenceb> the pin labelling is completely random
2012-10-11T17:04:54 < Laurenceb> half my leds are reversed
2012-10-11T17:06:18 <+Steffanx> Exactly half
2012-10-11T17:06:19 <+Steffanx> ?
2012-10-11T17:07:44 < Thorn> oh lol, pin pitch on the package drawing is "0.50 TYP for reference only"
2012-10-11T17:11:04 < jpa-> i always wonder what "for reference only" on a CAD drawing is supposed to mean
2012-10-11T17:11:38 < zyp> not accurate?
2012-10-11T17:11:55 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 246 seconds]
2012-10-11T17:12:00 < jpa-> so in what situation should you refer to non-accurate measurements? :P
2012-10-11T17:12:34 < Thorn> wht if I, say, actually want to make a pattern for the package.
2012-10-11T17:12:59 < BrainDamage> maybe the schematic is copyrighted or anyway restricted from being implemented?
2012-10-11T17:13:17 < BrainDamage> but you're allowed to do derivative work from it
2012-10-11T17:18:15 -!- claude_work is now known as claude
2012-10-11T17:23:57 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-11T17:32:30 < Laurenceb> http://dieinafire.com/
2012-10-11T17:35:29 <+Steffanx> No sound Laurenceb ?!
2012-10-11T17:39:11 < Laurenceb> im bored
2012-10-11T17:39:24 < Laurenceb> can be bothered to do CE paperwork crap
2012-10-11T17:39:45 <+Steffanx> Isn't it nice you can be bothered with that?
2012-10-11T17:39:57 < Laurenceb> cant
2012-10-11T17:41:43 < Laurenceb> http://diydrones.com/forum/topics/arduplane-2-50-released
2012-10-11T17:41:50 < Laurenceb> but... its on arduino
2012-10-11T17:41:53 < Laurenceb> instant fail
2012-10-11T17:43:12 < cjbaird> Ha, the samba and rsync guy. He should know better.. >_> (There were guys doing early-ish MSP430 work around him..)
2012-10-11T17:44:01 < jpa-> so arduino here is a fail because it would be easier to do with a higher power platform?
2012-10-11T17:44:37 < Laurenceb> it would be much better
2012-10-11T17:44:50 < Laurenceb> imo F1 is underpowered for autopilots
2012-10-11T17:45:15 < Laurenceb> im going to stick an F3 on one of my dactyl boards and see if it will run a bit more nicely
2012-10-11T17:46:12 < Laurenceb> atm it only manages 100hz
2012-10-11T17:47:35 < karlp> still, betting against tridge takes some nerves
2012-10-11T17:49:22 < cjbaird> Maybe he's currently based in Canberra again-- Jon Oxer et.al. are based there, explaining the Arduino activity.
2012-10-11T17:50:26 < Laurenceb> explaining?
2012-10-11T17:50:33 -!- claude is now known as claude_work
2012-10-11T17:50:54 < Laurenceb> so its a fail center
2012-10-11T17:51:05 < cjbaird> One way of putting it. :)
2012-10-11T17:51:10 <+Steffanx> Yes, and ##stm32 too
2012-10-11T17:51:54 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-11T17:52:23 < jpa-> well-tuned simple algorithms can be surprisingly competitive against poorly tuned advanced algorithms.. it may be a good approach to start with a well-established hardware platform and spend no time on that
2012-10-11T17:53:04 < dongs> needs more chat
2012-10-11T17:53:21 < zyp> then chat more
2012-10-11T17:53:33 < Laurenceb> i like turtles
2012-10-11T17:53:39 < zyp> not you
2012-10-11T17:57:55 <+Steffanx> Don't be so hard for mr Laurenceb
2012-10-11T17:58:16 < Laurenceb> you're hard for me?
2012-10-11T17:58:25 -!- claude_work is now known as claude
2012-10-11T17:58:30 < jpa-> uh.. yeah we are
2012-10-11T17:58:35 < Tectu> using arm-none-eabi-size, what exactly is the RAM size that will be used?   data +bss ?
2012-10-11T17:58:41 < Laurenceb> ill feltch you one of these then
2012-10-11T17:58:51 < Laurenceb> http://www.b3tards.com/u/d03ec72263e23ed4c51e/twathat.gif
2012-10-11T17:59:19 < jpa-> Tectu: data + bss + whatever you allocate dynamically
2012-10-11T17:59:20 <+Steffanx> Laurenceb and his weird hobbies
2012-10-11T17:59:22 < jpa-> Tectu: + stack
2012-10-11T17:59:27 <+Steffanx> hobbys or whatever
2012-10-11T17:59:38 < Tectu> jpa-, isn't stack in data/bss ?
2012-10-11T17:59:47 < dongs> Laurenceb: thats a retweet
2012-10-11T17:59:50 < jpa-> Tectu: no
2012-10-11T17:59:53 < cjbaird> What's the highest rpm a fixed-pitch (& ducted?) propeller for a UAV-type project can acheive? Was thinking about a 'flying-saucer' idea..
2012-10-11T18:00:01 < jpa-> Tectu: or depends how you do things, but usually no
2012-10-11T18:00:05 < Laurenceb> very high
2012-10-11T18:00:12 < Laurenceb> like 120000rpm
2012-10-11T18:00:37 < Laurenceb> but you lose efficiency
2012-10-11T18:00:38 < cjbaird> Coanda-effect was what I was going to mention.
2012-10-11T18:00:42 < Laurenceb> hehe
2012-10-11T18:01:07 < Laurenceb> http://jlnlabs.online.fr/gfsuav/index.htm
2012-10-11T18:01:13 < cjbaird> 'efficiency smenficiency, as long as it flies..' :)
2012-10-11T18:02:41 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-11T18:03:43 < Laurenceb> http://jlnlabs.online.fr/steorn/indexen.htm
2012-10-11T18:04:48 < BrainDamage> how much money does he have to waste on stupid shit?
2012-10-11T18:05:37 < dongs> haha
2012-10-11T18:05:40 < dongs> is that the overunity guy
2012-10-11T18:06:15 < Laurenceb> yes
2012-10-11T18:06:23 < Laurenceb> *the*  guy
2012-10-11T18:06:31 < Laurenceb> who comes up with about 90% of it
2012-10-11T18:06:41 < dongs> http://jlnlabs.online.fr/steorn/images/orboscope.jpg wow a 3 channel scope
2012-10-11T18:06:55 < dongs> hes better equipped than most of trolls
2012-10-11T18:07:01 < Laurenceb> i like how all his projects get abandoned
2012-10-11T18:07:02 < jpa-> probably just multiplied the two channels
2012-10-11T18:07:10 < jpa-> it says "MATH"
2012-10-11T18:07:11 < Laurenceb> he "proves"  it works
2012-10-11T18:07:11 < dongs> o
2012-10-11T18:07:12 < dongs> hurite
2012-10-11T18:07:12 < dongs> yeah
2012-10-11T18:07:14 < dongs> sorry, :(
2012-10-11T18:07:19 < Laurenceb> then nothing happens
2012-10-11T18:07:28 < Laurenceb> he never builds anything with it
2012-10-11T18:08:08 < Laurenceb> hes big on rcgroups
2012-10-11T18:09:09 < cjbaird> vid: https://www.youtube.com/watch?v=KXVtUCABiv8
2012-10-11T18:10:16 < Laurenceb> http://jlnlabs.online.fr/2SGen/images/2Sgenl.jpg
2012-10-11T18:10:16 < Laurenceb> wtf does that prove
2012-10-11T18:11:12 <+Steffanx> It proves he has at least two dmms
2012-10-11T18:11:19 <+Steffanx> make it 4
2012-10-11T18:11:35 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-11T18:12:05 < BrainDamage> just that he built an incredibly inefficient dcdc
2012-10-11T18:12:19 < BrainDamage> and when he connects a load, efficiency goes slightly up
2012-10-11T18:12:46 < Laurenceb> yeah
2012-10-11T18:29:16 < zyp> Tectu, data and bss are static allocations (globals)
2012-10-11T18:29:32 < zyp> you usually use what is left over after that for heap and stack
2012-10-11T18:31:04 < Laurenceb> http://news.nationalgeographic.com/news/2005/08/0816_050816_gas_theft.html
2012-10-11T18:31:15 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-11T18:44:35 -!- tavish [~tavish@120.56.129.77] has joined ##stm32
2012-10-11T18:44:36 -!- tavish [~tavish@120.56.129.77] has quit [Changing host]
2012-10-11T18:44:36 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-11T18:45:16 < cjbaird> lol. And if he's apprehended, he can just detach the balloon from the bike, and float away to freedom.
2012-10-11T18:52:54 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-11T18:53:27 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-11T18:54:16 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-11T18:57:53 <+Steffanx> lol
2012-10-11T18:58:39 < emeb> mfm
2012-10-11T18:59:08 -!- tavish_ [~tavish@59.177.9.213] has joined ##stm32
2012-10-11T18:59:54 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-10-11T19:08:06 -!- tavish_ [~tavish@59.177.9.213] has quit [Ping timeout: 265 seconds]
2012-10-11T19:10:12 -!- barthess [~barthess@178.154.67.222] has joined ##stm32
2012-10-11T19:27:36 -!- tavish_ [~tavish@120.56.131.186] has joined ##stm32
2012-10-11T19:29:33 -!- Tectu is now known as tectu
2012-10-11T19:35:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-11T19:36:04 -!- tavish_ [~tavish@120.56.131.186] has quit [Ping timeout: 248 seconds]
2012-10-11T19:36:54 -!- claude is now known as Claude
2012-10-11T19:38:03 -!- tavish_ [~tavish@120.59.1.75] has joined ##stm32
2012-10-11T19:43:12 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Read error: Connection reset by peer]
2012-10-11T19:47:02 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-11T19:47:15 -!- Claude is now known as claude
2012-10-11T19:55:25 -!- tavish_ [~tavish@120.59.1.75] has quit [Ping timeout: 272 seconds]
2012-10-11T19:56:29 -!- tavish_ [~tavish@59.177.11.5] has joined ##stm32
2012-10-11T20:05:26 -!- tavish_ [~tavish@59.177.11.5] has quit [Ping timeout: 255 seconds]
2012-10-11T20:06:24 -!- tavish [~tavish@120.56.130.81] has joined ##stm32
2012-10-11T20:06:25 -!- tavish [~tavish@120.56.130.81] has quit [Changing host]
2012-10-11T20:06:25 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-11T20:31:42 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-11T20:35:22 < TitanMKD> hi
2012-10-11T20:36:41 <+Steffanx> lo
2012-10-11T20:37:50 < TitanMKD> anyone know a good hint to do prime factorization very quickly ? with for example D = x*y*z in order to find value for x, y and z whith a D value ?
2012-10-11T20:40:08 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-11T20:45:51 -!- claude is now known as Claude
2012-10-11T20:54:42 -!- Claude is now known as claude
2012-10-11T21:00:01 < jpa-> TitanMKD: for what range of values (less or over 2^32, for example)
2012-10-11T21:08:34 < TitanMKD> less than 2^32 ;)
2012-10-11T21:08:44 < TitanMKD> or just from 0 to 2^32 in fact
2012-10-11T21:08:53 < TitanMKD> but in a determinist way ;)
2012-10-11T21:13:47 -!- tectu is now known as Tectu
2012-10-11T21:15:54 < jpa-> http://en.wikipedia.org/wiki/Shanks%27_square_forms_factorization  might be easiest to implement
2012-10-11T21:16:15 < jpa-> but there are probably internet full of premade implementations of any of http://en.wikipedia.org/wiki/Integer_factorization#General-purpose
2012-10-11T21:30:26 < emeb> O(n^(1/4)) - hmmm. So does that mean roughly 256 operations for a 32-bit number?
2012-10-11T21:30:40  * emeb doesn't know this kind of math well...
2012-10-11T21:31:17 < zyp> n^(1/4) means fourth-root of n
2012-10-11T21:31:22 < zyp> or square root of square root
2012-10-11T21:32:24 < zyp> and square root of 2**32 is 256, so that sounds reasonable
2012-10-11T21:33:14 < zyp> it's not about number of operations though, it's just how the algorithms scales with input
2012-10-11T21:33:41 < emeb> yeah - that's where I'm not clueful.
2012-10-11T21:35:46 < emeb> so it's proportionality, not absolute.
2012-10-11T21:36:17 < emeb> but the example for 11111 shows a reasonable number of iterations, so perhaps 2^32 wouldn't be too awful.
2012-10-11T21:37:53 <+dekar_> Linkerscript without any copyleft: https://github.com/EliasOenal/TNT_Example/blob/master/linker_script/simple.ld
2012-10-11T21:38:56 < TitanMKD> to be sure i will like to see the worst case ;)
2012-10-11T21:40:30 < emeb> so try all 4 billion :)
2012-10-11T21:41:25 < emeb> dekar_: "supports only C and will lead to weird behavior when used for C++"
2012-10-11T21:41:45 <+dekar_> emeb, feel free to add the sections for C++
2012-10-11T21:41:49 <+dekar_> I only use C
2012-10-11T21:42:08 < emeb> me too. Don't know what's needed for C++.
2012-10-11T21:42:23 <+dekar_> emeb, it's like 3 other sections
2012-10-11T21:42:37 <+dekar_> one for constructing global objects at the start
2012-10-11T21:42:48 < emeb> for F4 need to add the ccm section too.
2012-10-11T21:43:01 <+dekar_> ccm?
2012-10-11T21:43:12 < emeb> 64kB core-coupled SRAM
2012-10-11T21:43:19 <+dekar_> oh yeah well
2012-10-11T21:43:35 <+dekar_> my blinky example doesn't need that
2012-10-11T21:43:37  * emeb uses that for audio delay buffers
2012-10-11T21:43:51 <+dekar_> I agree that it is pretty useful
2012-10-11T21:44:50 <+dekar_> also if any of you has an f1 discovery, please fix the pins for the leds in my example and give me a patch
2012-10-11T21:45:45 <+dekar_> I've only tested it on a custom f1 board and I didn't feel like checking the pins of the f1 discovery ;)
2012-10-11T21:45:57 < zyp> ew, stdperiphlib :p
2012-10-11T21:46:12 <+dekar_> zyp, feel free to add any other lib to the example
2012-10-11T21:46:32 <+dekar_> I just want a good example for beginners, everything welcome
2012-10-11T21:47:00 <+dekar_> I filled most of the stubs, malloc is working and printf is easy to fill in as well
2012-10-11T21:47:47 < zyp> I believe my stuff when cleaned up should be rather easy to get started with
2012-10-11T21:47:56 < zyp> I'm just afraid of having to support beginners :)
2012-10-11T21:48:59  * emeb raises hand. <- C++ beginner.
2012-10-11T21:49:19  * zyp hides
2012-10-11T21:49:20  * dekar_ <- LUA beginner
2012-10-11T21:50:44 <+dekar_> building the example with TNT results in a firmware 8k smaller than with yagarto - just saying
2012-10-11T21:51:14 < TitanMKD> dekar_ what's that TNT version ?
2012-10-11T21:51:38 <+Steffanx> That is 1k vs 9k? dekar_  :P
2012-10-11T21:51:45 <+dekar_> the GCC? linaro 4.7
2012-10-11T21:51:57 < jpa-> dekar_: just make sure that it also works :P
2012-10-11T21:52:05 < jpa-> LTO can sometimes optimize a bit too much ,)
2012-10-11T21:52:10 <+dekar_> jpa-, it works, try it
2012-10-11T21:52:19 <+dekar_> that's without LTO btw
2012-10-11T21:52:32 <+dekar_> yagarto:
2012-10-11T21:52:33 <+dekar_>    text	   data	    bss	    dec	    hex	filename
2012-10-11T21:52:33 <+dekar_>   27900	   2240	     60	  30200	   75f8	build/firmware.elf
2012-10-11T21:52:33 <+dekar_> TNT:
2012-10-11T21:52:33 <+dekar_>    text	   data	    bss	    dec	    hex	filename
2012-10-11T21:52:33 <+dekar_>   20184	   1380	     60	  21624	   5478	build/firmware.elf
2012-10-11T21:52:50 < zyp> how does the example get that big?
2012-10-11T21:53:03 <+dekar_> it has printf("Hello World!");
2012-10-11T21:53:10 < zyp> oh, right
2012-10-11T21:53:24 < jpa-> lol
2012-10-11T21:53:29 < jpa-> newlib sucks :)
2012-10-11T21:53:40 <+dekar_> jpa-, why?
2012-10-11T21:53:54 <+dekar_> you can use iprintf(); instead for example
2012-10-11T21:54:00 < jpa-> true
2012-10-11T21:54:04 < jpa-> how big is that btw?
2012-10-11T21:54:11 <+dekar_> also yagarto configured it wrong
2012-10-11T21:54:16 -!- claude is now known as Claude
2012-10-11T21:54:17 <+dekar_> thus it is so huge
2012-10-11T21:54:24 <+dekar_> jpa-, let me try
2012-10-11T21:55:51 <+dekar_>    text	   data	    bss	    dec	    hex	filename
2012-10-11T21:55:52 <+dekar_>    8688	   1324	     60	  10072	   2758	build/firmware.elf
2012-10-11T21:56:06 <+dekar_> iprintf("Hello World"); using TNT
2012-10-11T21:56:28 <+Steffanx> It should be iPrintf !
2012-10-11T21:56:38 <+dekar_> they'd get sued instantly
2012-10-11T21:56:58 <+dekar_> you can also use _write("Hello World"); if you only use strings
2012-10-11T21:57:19 < emeb> TIL what TNT stands for. Looks handy.
2012-10-11T21:57:29 <+dekar_>    text	   data	    bss	    dec	    hex	filename
2012-10-11T21:57:30 <+dekar_>     980	      0	      0	    980	    3d4	build/firmware.elf
2012-10-11T21:57:46 <+dekar_> that's _write("Hello World");
2012-10-11T22:04:10 <+dekar_> TNT isn't just Thumb2 anymore, Thumb1 is now also supported
2012-10-11T22:04:49 < jpa-> 1.5kB with baselibc (so tinyprintf)
2012-10-11T22:06:08 < TitanMKD> dekar_ anyway Thumb1 is really not interesting and obsolete with bad performances for old ARM CPU
2012-10-11T22:11:50 <+dekar_> TitanMKD, stm32f0?
2012-10-11T22:14:04 < TitanMKD> any CortexMx are Thumb2 only in fact
2012-10-11T22:14:14 < TitanMKD> Thumb1 is obsolete from old architecture
2012-10-11T22:17:14 < zyp> huh?
2012-10-11T22:17:24 < zyp> thumb2 extends thumb1, not replaces
2012-10-11T22:17:51 < zyp> and cortex-m0 lacks most of the thumb2 instructions, leaving it with almost only thumb1
2012-10-11T22:17:55 < TitanMKD> not exactly in fact
2012-10-11T22:18:04 < TitanMKD> Thumb2 replace some part of old Thumb1 too
2012-10-11T22:18:25 < TitanMKD> because Thumb1 was not complete instruction set requiring ARM instruction
2012-10-11T22:18:39 <+dekar_> stm32f0 support thumb2? didn't know that
2012-10-11T22:18:50 <+dekar_> I wonder where I picked that misconception up...
2012-10-11T22:18:57 < TitanMKD> i do not know F0 but CortexM are all Thumb2 ONLY
2012-10-11T22:19:01 < zyp> dekar_, it doesn't
2012-10-11T22:19:20 < jpa-> F0 is Cortex M0
2012-10-11T22:19:27 < TitanMKD> M0 shall be Thumb2 too
2012-10-11T22:19:31 < zyp> TitanMKD, you are full of shit
2012-10-11T22:19:36 < TitanMKD> zyp ?
2012-10-11T22:20:07 < TitanMKD> look there
2012-10-11T22:20:07 < TitanMKD> http://www.arm.com/products/processors/cortex-m/cortex-m0.php
2012-10-11T22:20:13 < jpa-> it is "full thumb / thumb2 subset"
2012-10-11T22:20:14 < TitanMKD> it's wrote THumb2 !!!
2012-10-11T22:20:57 < zyp> TitanMKD, subset
2012-10-11T22:21:38 < jpa-> The Cortex-M0 / M0+ / M1 include a minor subset of Thumb-2 instructions (BL, DMB, DSB, ISB, MRS, MSR).   so you have a total of 6 thumb2 instructions that are supported :)
2012-10-11T22:21:48 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Ping timeout: 245 seconds]
2012-10-11T22:22:02 < TitanMKD> yes but it's not like old pre
2012-10-11T22:22:06 < zyp> yep, and the rest is the full thumb1 set
2012-10-11T22:22:06 < TitanMKD> v5 ARM
2012-10-11T22:22:12 < jpa-> and actually M0 doesn't have even the full thumb1
2012-10-11T22:22:21 < jpa-> http://en.wikipedia.org/wiki/ARM_Cortex-M#Instruction_sets
2012-10-11T22:22:55 <+dekar_> http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0432c/CHDCICDF.html
2012-10-11T22:23:17 <+dekar_> zyp, thx :)
2012-10-11T22:25:24 <+dekar_> so it definitively supports armv6-m which is what I put in my makefile
2012-10-11T22:25:40 < zyp> F0? yes.
2012-10-11T22:26:18 -!- barthess [~barthess@178.154.67.222] has quit [Quit: Leaving.]
2012-10-11T22:26:36 < TitanMKD> anyway on GCC we cannot specify thumb1 or 2 it is automaticly selected by arch IIRC
2012-10-11T22:26:56 < zyp> correct
2012-10-11T22:27:24 < TitanMKD> I was wrong only because i was thinking M0 support full Thumb2 and it is partial ;)
2012-10-11T22:27:50 < zyp> https://launchpadlibrarian.net/109296007/readme.txt <- this readme has a good overview of the flags to use for each arch
2012-10-11T22:28:15 < TitanMKD> and the worst is
2012-10-11T22:28:15 < TitanMKD> Thumb (most), missing CBZ, CBNZ, IT.
2012-10-11T22:28:30 < TitanMKD> IT is very useful it is a very bad things to do not have them
2012-10-11T22:28:33 < zyp> yeah, it's so horrible, better not use F0
2012-10-11T22:29:29 < TitanMKD> yes only M3 and M4 support full Thumb/Thumb2 ;)
2012-10-11T22:29:44 < TitanMKD> and with branch speculation
2012-10-11T22:30:29 <+dekar_> https://github.com/EliasOenal/TNT_Example/blob/master/Makefile
2012-10-11T22:31:15 <+dekar_> ITEEE, programming assembly never was so much fun :)
2012-10-11T22:31:49 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-11T22:31:59 < zyp> do cortex-m0 even have branch prediction?
2012-10-11T22:32:30 <+dekar_> idk, but having a 3 stage pipeline it shouldn't hurt that much
2012-10-11T22:32:38 < zyp> exactly
2012-10-11T22:33:27 <+dekar_> I think you could configure M0 (at hardware "compile" time) to have faster or slower multiplications
2012-10-11T22:33:40 < zyp> yep
2012-10-11T22:33:49 <+dekar_> which is pretty odd imo
2012-10-11T22:33:57 < zyp> why?
2012-10-11T22:34:08 < zyp> it's just a matter of having a fully unrolled multiplier or not
2012-10-11T22:34:09 <+dekar_> I get why they do it, but it can have weird peformance impact
2012-10-11T22:38:00 <+dekar_> that readme on the launchpad site has a linkerscript that seems to support C++ :)
2012-10-11T22:38:01 < zyp> it's just that the optimizer should know
2012-10-11T22:38:18 < zyp> so do mine, it's not hard
2012-10-11T22:38:40 < Thorn> what do 50MHz oscillators normally cost (single units)?
2012-10-11T22:39:14 < zyp> http://cgit.jvnv.net/laks/tree/ld_scripts/arm_flash_ram.ld <- C++ support is mostly about the .init_array section, which contain function pointers to global constructors
2012-10-11T22:39:33 <+dekar_> Thorn, less than a euro I'd assume
2012-10-11T22:39:47 < zyp> the .fini_array are global destructors, but you never get there when main() never returns :p
2012-10-11T22:40:33 < zyp> http://cgit.jvnv.net/laks/tree/startup/entry.cpp <- then it's just a matter of calling all the constructors before calling main()
2012-10-11T22:40:47 <+dekar_> Thorn, http://www.digikey.com/product-detail/en/9C-50.000MEEJ-T/887-1291-1-ND/2207873
2012-10-11T22:41:04 < Thorn> I mean active ones with 4 pins
2012-10-11T22:41:32 <+dekar_> maybe a bit more
2012-10-11T22:42:09 < zyp> http://www.digikey.com/product-detail/en/FXO-HC735-50/631-1060-1-ND/1024765 <- like this?
2012-10-11T22:42:15 < Thorn> I'm looking at ~2.5 euro
2012-10-11T22:44:29 < Thorn> pretty much yes
2012-10-11T22:44:40 < zyp> well, it says $1.31
2012-10-11T22:44:46 < zyp> in single quantity
2012-10-11T22:47:41 <+dekar_> zyp, do you know whether you can tell GCC how fast a multiplication is?
2012-10-11T22:47:51 < zyp> no idea
2012-10-11T22:48:39 <+dekar_> I'm glad I don't use M0 :)
2012-10-11T22:48:54 -!- Claude is now known as claude
2012-10-11T22:55:28 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-11T22:55:44 -!- daku is now known as DaKu
2012-10-11T22:57:18 -!- DaKu is now known as daku
2012-10-11T22:57:47 <+dekar_> should I offer binary distributions of TNT?
2012-10-11T22:58:00 < Laurenceb_> yes
2012-10-11T22:58:13 < Laurenceb_> but full of trojans
2012-10-11T22:58:34 <+dekar_> cbf writing trojans for OSX and Linux, no one uses those anyway ;)
2012-10-11T22:58:41 -!- BrainDamage [~BrainDama@62.10.7.162] has quit [Ping timeout: 255 seconds]
2012-10-11T23:00:08 -!- daku is now known as DaKu
2012-10-11T23:02:17 -!- BrainDamage [~BrainDama@62.10.7.162] has joined ##stm32
2012-10-11T23:02:47 -!- BrainDamage [~BrainDama@62.10.7.162] has quit [Remote host closed the connection]
2012-10-11T23:02:58 -!- BrainDamage [~BrainDama@62.10.7.162] has joined ##stm32
2012-10-11T23:14:23 < Laurenceb_> http://pwnage.ro/wp-content/uploads/2010/11/the-fail-element.jpg
2012-10-11T23:15:31 < emeb> my eyes...
2012-10-11T23:16:12 < Tectu> kinda hot
2012-10-11T23:16:22 < zyp> is that you?
2012-10-11T23:16:31 < Laurenceb_> lol Tectu swings both ways
2012-10-11T23:17:48 < emeb> unicorn chaser: http://www.blackmilkclothing.com/products/artoo-swimsuit
2012-10-11T23:19:06 < Laurenceb_> http://media.heavy.com/media/2010/03/pink-catgirl.jpg
2012-10-11T23:19:22 < Laurenceb_>  *sounds of channel fapping*
2012-10-11T23:19:36  * emeb screams, claws at face.
2012-10-11T23:19:41 < Tectu> http://img.pr0gramm.com/2012/10/1349756718054.gif
2012-10-11T23:20:15 < BrainDamage> aaaannnd that's enough internet for me today
2012-10-11T23:20:35 -!- BrainDamage [~BrainDama@62.10.7.162] has quit [Quit: Putting the "sensual" back into "nonconsensual"]
2012-10-11T23:21:13 < emeb> he was serious I guess...
2012-10-11T23:21:31 < zyp> sometimes I'm glad you guys haven't found any of the pictures of me that's floating around out there somewhere :)
2012-10-11T23:21:32 < Tectu> http://img.pr0gramm.com/2012/10/qm4hd.jpg
2012-10-11T23:21:48 < Tectu> zyp, which one do you mean?
2012-10-11T23:22:02 < emeb> not clicking...
2012-10-11T23:22:12 < Laurenceb_> haha
2012-10-11T23:22:22 < Tectu> emeb, that's not that ugly
2012-10-11T23:22:56 < Laurenceb_> ...
2012-10-11T23:43:03 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 246 seconds]
2012-10-11T23:43:59 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-10-11T23:44:18 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-11T23:44:53 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-11T23:51:37 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-11T23:53:15 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
--- Day changed Fri Oct 12 2012
2012-10-12T00:03:09 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-12T00:32:38 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 256 seconds]
2012-10-12T00:54:39 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-12T01:08:43 -!- DaKu is now known as daku
2012-10-12T01:20:39 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-12T01:28:28 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-12T01:32:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-12T01:36:35 -!- BrainDamage [~BrainDama@62.10.7.162] has joined ##stm32
2012-10-12T01:38:22 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-10-12T01:41:06 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-12T01:46:30 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-12T01:58:04 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Read error: Connection reset by peer]
2012-10-12T02:04:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-12T02:06:33 -!- Blinda [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-12T02:07:23 -!- Posterdati [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-10-12T02:08:24 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-12T02:09:54 -!- Blinda is now known as Posterdati
2012-10-12T02:13:10 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-12T02:15:02 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has joined ##stm32
2012-10-12T02:15:16 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/]
2012-10-12T02:15:52 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-12T02:16:55 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Client Quit]
2012-10-12T02:17:12 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-12T02:18:06 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Client Quit]
2012-10-12T02:18:35 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-12T02:21:55 -!- Laurenceb_ [~Laurence@host86-129-128-213.range86-129.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-12T02:22:51 -!- Posterdati_ [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-12T02:23:05 -!- Posterdati_ [~quassel@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-10-12T02:58:46 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-10-12T03:00:49 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-12T03:04:36 < dongs> sup kawaii dongs
2012-10-12T03:07:51 < karlp> status: failing to count usecs.
2012-10-12T03:08:02 < karlp> status: just generally failing I guess :)
2012-10-12T03:49:50 < karlp> holy shit, wfi totally screws up texane/stlink doesn't it
2012-10-12T03:50:00 < karlp> I heard you say it "didn't work well"
2012-10-12T03:50:08 < karlp> but it's a nightmare if you use wfi anywhere.
2012-10-12T03:50:39 < dongs> I should try it and see how it works with proper tools :)
2012-10-12T03:54:52 < emeb> I'm using wfi in one of my projects - seems to work OK, but I haven't tried doing any serious debugging on that one.
2012-10-12T03:55:16 < emeb> just used texane/stlink for flash programming.
2012-10-12T03:56:29 < karlp> I nromally have st-util running, and just ctrl-c in gdb and "load;run" over and over again
2012-10-12T03:56:44 < karlp> st-flash might work, but st-util and rejoining gdb breaks terribly.
2012-10-12T03:56:53 < emeb> oh - I was prolly using st-flash for that. It's been a while.
2012-10-12T04:00:03 < emeb> I usually put the flash command in my Makefile, so all I need to do is 'make flash' from cmd line.
2012-10-12T04:04:47 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-12T04:11:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-12T04:12:27 < dongs> dekar_: any luck wiht softi2c?
2012-10-12T04:13:54 <+dekar_> dongs, didn't have time, talked to my boss though and he said we could use i2c slave emulation and I'd be free to share it
2012-10-12T04:14:11 < dongs> cooldeal. offer still stands
2012-10-12T04:47:13 < upgrdman> what is softi2c? big banged?
2012-10-12T04:47:39 < dongs> slave, yes
2012-10-12T04:51:45 -!- BrainDamage [~BrainDama@62.10.7.162] has quit [Quit: Leaving.]
2012-10-12T05:24:29 < GargantuaSauce> http://workman-industries.net/images/f4_cube.jpg
2012-10-12T05:24:38 < GargantuaSauce> that took way too long
2012-10-12T05:24:44 < GargantuaSauce> perhaps i should not have rolled my own linear math
2012-10-12T05:24:48 < GargantuaSauce> \(O_o)/
2012-10-12T05:26:17 < dongs> does it spin
2012-10-12T05:26:23 < GargantuaSauce> of course it does
2012-10-12T05:26:33 < dongs> well who knwos maybe youre just displaying a static .png
2012-10-12T05:27:12 < dongs> http://www.youtube.com/watch?v=smz2cU3FfFk seen mine?
2012-10-12T05:27:38 < GargantuaSauce> oh that's you? cool
2012-10-12T05:28:43 < dongs> ha, its got more views than most of my videos haha
2012-10-12T05:29:02 < GargantuaSauce> i see you actually bothered with a framebuffer
2012-10-12T05:29:10 < GargantuaSauce> i did not :D
2012-10-12T05:29:16 < dongs> oh, s oyours flickers?
2012-10-12T05:29:22 < dongs> but yours is fsmc so it should be faster to update
2012-10-12T05:29:27 < GargantuaSauce> no it's not
2012-10-12T05:29:46 < dongs> how do you 'draw' into single buffer without flicker?
2012-10-12T05:29:51 < dongs> single buffer being htelcd
2012-10-12T05:30:06 < GargantuaSauce> it flickers
2012-10-12T05:31:05 < dongs> ok, i thought you had some magic :)
2012-10-12T05:31:40 < GargantuaSauce> i can think of a couple ways to magic it up but probably will not bother
2012-10-12T05:32:44 < GargantuaSauce> i have a fleet of robots to build before i put time into fancy stuff :P
2012-10-12T05:37:16 < dongs> heh
2012-10-12T05:54:38 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-12T06:30:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-12T06:30:42 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-12T06:58:13 < dongs> zerochats
2012-10-12T06:59:36 < emeb_mac> too quiet
2012-10-12T07:01:27 < emeb_mac> heh DK finally got some F3s
2012-10-12T07:03:07 < dongs> which ones?
2012-10-12T07:03:22 < dongs> dont see any.
2012-10-12T07:03:25 < dongs> al 100lqfp shit
2012-10-12T07:03:37 < dongs> STM32F373VCT6 FAIL
2012-10-12T07:03:50 < emeb_mac> yep
2012-10-12T07:03:56 < emeb_mac> one tray of 200
2012-10-12T07:04:17 < emeb_mac> still no F3 discoverys
2012-10-12T07:04:49 < dongs> if f4dfiscovery is any indicator, they wont be in stock for a while
2012-10-12T07:04:51 < dongs> just backorder and wait.
2012-10-12T07:06:48 < emeb_mac> what did you get - CCT6?
2012-10-12T07:07:07 < dongs> yea
2012-10-12T07:07:20 < dongs> http://uk.mouser.com/ProductDetail/STMicroelectronics/STM32F373CCT6/?qs=sGAEpiMZZMts0YcPTleK%252bfFyenw%2f%252boXO still in stock lol
2012-10-12T07:08:05 < emeb_mac> yup
2012-10-12T07:08:06 < dongs> wow mouser has a ton of F3 shit in stock
2012-10-12T07:08:13 < emeb_mac> yes they do.
2012-10-12T07:08:46 < dongs> no 30x though.
2012-10-12T07:09:43 -!- lawrenceseattle [~anonymous@216.243.49.130] has joined ##stm32
2012-10-12T07:10:20 < emeb_mac> someone should make an F373CCT6 DIP breakout w/ a USB connector.
2012-10-12T07:10:40 < emeb_mac> Hmmm - prooooject.
2012-10-12T07:10:59 < dongs> on eagle?
2012-10-12T07:11:01 < dongs> on->in
2012-10-12T07:11:07 < emeb_mac> don't care.
2012-10-12T07:11:22 < emeb_mac> If I did it would be in opensores :P
2012-10-12T07:11:39 < dongs> ugh
2012-10-12T07:11:42 < emeb_mac> heh
2012-10-12T07:11:44 < dongs> you mean shit like kikcad?
2012-10-12T07:11:56 < emeb_mac> nah - gEDA/PCB is my go-to.
2012-10-12T07:12:28 < dongs> o rite.
2012-10-12T07:12:54 < emeb_mac> I've already got one for F100 in a 48lqfp : http://ebrombaugh.studionebula.com/embedded/stm32breakout/index.html
2012-10-12T07:13:22 < emeb_mac> I'd probably tweak that for a 10-pin mini JTAG
2012-10-12T07:13:41 < emeb_mac> add USB
2012-10-12T07:13:47 < dongs> that is HUGE.
2012-10-12T07:13:58 < emeb_mac> heh - go easy - my first one.
2012-10-12T07:14:30 < emeb_mac> I could make it a lot tighter now.
2012-10-12T07:15:02 < emeb_mac> I did that before I knew about SWD or mini JTAG.
2012-10-12T07:15:24 < emeb_mac> plus I used a huge rst button and left a lot of space.
2012-10-12T07:18:40 < dongs> http://i.imgur.com/Jc9v4.jpg i could prob modify my minishits by trashing usb>uart
2012-10-12T07:18:55 < dongs> wait yours didnt have a xtal?
2012-10-12T07:19:48 < emeb_mac> nope
2012-10-12T07:19:56 < dongs> need one for usb :p
2012-10-12T07:20:10 < dongs> I got some nice 16mhz in 3225
2012-10-12T07:20:12 < emeb_mac> Indeed - that was an F100 w/o USB
2012-10-12T07:20:12 < dongs> and 12mhz.
2012-10-12T07:20:28 < emeb_mac> yours is nice & tiny.
2012-10-12T07:22:27 < emeb_mac> I see the xtal - that little black bar on the bottom w/ 6 side pins?
2012-10-12T07:22:51 < dongs> yea
2012-10-12T07:22:55 < dongs> resonator w/builtin caps
2012-10-12T07:23:03 < emeb_mac> aha
2012-10-12T07:23:35 < emeb_mac> what's the accuracy on resonators? good enough for USB?
2012-10-12T07:24:02 < dongs> unsure, didnt try hooking up USB pins on that board.
2012-10-12T07:24:15 < dongs> it should be better htan intrc
2012-10-12T07:24:16 < dongs> heh
2012-10-12T07:25:45 < emeb_mac> yep
2012-10-12T07:25:52 < emeb_mac> like this: http://www.digikey.com/product-detail/en/PRQC8.00CR5010X000/478-4584-1-ND/1645895
2012-10-12T07:26:09 < dongs> i believe thats waht im using
2012-10-12T07:26:12 < dongs> yes.
2012-10-12T07:26:27 < emeb_mac> +/- 0.5%
2012-10-12T07:26:59 < emeb_mac> not 100ppm like xtal, but better than internal
2012-10-12T07:31:14 < emeb_mac> what you using for regulator? is that the SOT-23-5 part?
2012-10-12T07:32:00 < dongs> yeah
2012-10-12T07:32:05 < dongs> richtek clone of LP2992
2012-10-12T07:32:08 < dongs> very common pinout.
2012-10-12T07:32:12 < dongs> RT9391 or so
2012-10-12T07:32:21 < dongs> like 0.01$/ea in reel
2012-10-12T07:32:47 < dongs> RF9193
2012-10-12T07:32:50 < dongs> err RT9193
2012-10-12T07:34:03 < dongs> but there is a ton in that pinout
2012-10-12T07:34:07 < dongs> MCP1802 etc
2012-10-12T07:34:25 < emeb_mac> true lp2992 is $0.87/qty1
2012-10-12T07:36:17 < emeb_mac> mcp1802 is a bit cheaper.
2012-10-12T07:36:42 < emeb_mac> not bad. don't need more than 100mA anyway.
2012-10-12T07:38:04 < emeb_mac> (my F405 project running full-blast w/ codec takes ~93ma)
2012-10-12T07:38:16 < dongs> nice
2012-10-12T07:39:18 < emeb_mac> need to check what kind of freq accuracy is needed for USB...
2012-10-12T07:44:29 < emeb_mac> USB spec for FS devices is 0.25%
2012-10-12T07:44:49 < emeb_mac> so those AVX parts aren't quite good enough.
2012-10-12T07:45:52 < dongs> thats too bad.
2012-10-12T07:45:58 < dongs> smallest 3225 xtal is 12mhz.
2012-10-12T07:46:02 < dongs> 8mhz is huge.
2012-10-12T07:47:53 -!- claude is now known as Claude
2012-10-12T07:55:02 < emeb_mac> neither DK nor Mouser have resonators that are much better than 0.5%
2012-10-12T07:55:30 < emeb_mac> and as you noted, true xtals are large - 7x5mm pkgs look about the smallest
2012-10-12T07:55:57 < dongs> i am using 12mhz xtal on f103w/usb board.
2012-10-12T07:56:15 < dongs> nothing wrong with 12, still plls nicely to 48mhz etc.
2012-10-12T07:56:26 < dongs> chip runs at full 72mhz
2012-10-12T07:56:31 < dongs> can overdong it more too :p
2012-10-12T07:56:58 < emeb_mac> 12 is fine.
2012-10-12T07:57:24 < emeb_mac> HSE runs to 25MHz, right?
2012-10-12T07:57:31 < dongs> i belive so
2012-10-12T08:08:43 < emeb_mac> something like this? http://www.mouser.com/ProductDetail/AVX/CX3225GB12000D0HPQZ1/?qs=sGAEpiMZZMsBj6bBr9Q9aT8H41%252bCVaoKI0MDqdbvrX4%3d
2012-10-12T08:09:57 < dongs> yeah, thats too expensive.. i use NDK.. sec
2012-10-12T08:10:11 < dongs> 644-1047-1-ND
2012-10-12T08:11:00 < emeb_mac> nice
2012-10-12T08:18:27 < emeb_mac> this the USB conn you like: http://www.digikey.com/product-detail/en/0548190519/WM17115-ND/773802
2012-10-12T08:18:55 < dongs> yeah, the only dip usb socket worth using ever
2012-10-12T08:19:21 < dongs> of course i dont use molex version
2012-10-12T08:20:05 < emeb_mac> $1.49 seems a bit high
2012-10-12T08:20:23 < dongs> yeah i get them in trays of 100 for a lot less than that.
2012-10-12T08:22:33 < dongs> believe it or not thats the che apest through hole usb connector on digikey.
2012-10-12T08:24:04 < dongs> mouser has some more choices, none are cheaper
2012-10-12T08:25:03 < upgrdman> wow, why so expensive? low volume compared to SMD?
2012-10-12T08:25:03 < emeb_mac> maybe this one? http://www.mouser.com/ProductDetail/Molex/51387-0578/?qs=sGAEpiMZZMulM8LPOQ%252bykzAp4yt8IxVbM5hlI3P3qys%3d
2012-10-12T08:25:12 < emeb_mac> 0.11 cheaper
2012-10-12T08:25:38 < dongs> thats a nonstandard pad pattern
2012-10-12T08:25:43 < emeb_mac> true
2012-10-12T08:26:03 < emeb_mac> upgrdman: good an explanation as any.
2012-10-12T08:26:10 < dongs> oh
2012-10-12T08:26:13 < dongs> its not even all through hole
2012-10-12T08:26:20 < dongs> its the fake smt + shell tabs
2012-10-12T08:26:20 < emeb_mac> ORLY?
2012-10-12T08:26:22 < dongs> yeah
2012-10-12T08:26:25 < dongs> the pins are smt
2012-10-12T08:26:30 < dongs> USB On-The-Go (OTG) Mini-B Receptacle, Right Angle, Short Body, SMT Solder Tails
2012-10-12T08:26:31 < dongs> and Through Hole Shell Tabs, with Cover Tape, Lead-Free
2012-10-12T08:26:35 < emeb_mac> aha
2012-10-12T08:26:36 < dongs> why the fuck the datasheet doesnt have pattern/pinout
2012-10-12T08:26:39 < dongs> fucking molex
2012-10-12T08:26:44 < emeb_mac> grmbl
2012-10-12T08:26:51 < dongs> http://www.mouser.com/ds/1/276/0513870578_IO_CONNECTORS-115766.pdf
2012-10-12T08:26:56 < dongs> why the hell did mouser cache this shit
2012-10-12T08:27:56 < dongs> http://www.molex.com/pdm_docs/sd/513870530_sd.pdf
2012-10-12T08:27:58 < dongs> theres the right shit
2012-10-12T08:28:09 < dongs> o wait
2012-10-12T08:28:11 < dongs> tahts different series
2012-10-12T08:28:12 < dongs> anyway, still fail
2012-10-12T08:28:15 < dongs> smt + tabs
2012-10-12T08:29:00 < dongs> http://www.molex.com/pdm_docs/sd/513870578_sd.pdf
2012-10-12T08:29:02 < dongs> this is the real one
2012-10-12T08:30:57 < emeb_mac> ah - not the same as the 1st one I linked - this has 4 tabs, other had only 2
2012-10-12T08:31:47 < dongs> right, the last pdf is 4 tabs
2012-10-12T08:31:54 < emeb_mac> main problem with that footprint for me is that most of the cheapo PCB vendors I use don't do oval plated holes
2012-10-12T08:32:23 < dongs> http://www.ebay.com/itm/130748829351 rofl
2012-10-12T08:32:54 < emeb_mac> one born every minute
2012-10-12T08:33:59 < dongs> oval holes is standard shit here
2012-10-12T08:34:02 < dongs> can gEDA even do those?
2012-10-12T08:34:03 < dongs> :p
2012-10-12T08:34:24 < upgrdman> why would playing ovals be harder than circles?
2012-10-12T08:34:32 < upgrdman> plating
2012-10-12T08:34:37 < dongs> I ordered a fr4 panel of some shit, told the dorks NO GUIDES. got boards; http://i.imgur.com/1jWY1.jpg
2012-10-12T08:36:22 < upgrdman> what are guides? the strips top and bottom?
2012-10-12T08:47:22 < dongs> yeah
2012-10-12T08:52:40 -!- lawrenceseattle [~anonymous@216.243.49.130] has quit [Remote host closed the connection]
2012-10-12T08:53:00 -!- lawrenceseattle [~anonymous@216.243.49.130] has joined ##stm32
2012-10-12T09:02:08 -!- dekar__ [~dekar@212.255.43.16] has joined ##stm32
2012-10-12T09:02:11 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-10-12T09:03:32 -!- lawrenceseattle_ [~anonymous@216.243.49.130] has joined ##stm32
2012-10-12T09:05:51 -!- lawrenceseattle [~anonymous@216.243.49.130] has quit [Ping timeout: 276 seconds]
2012-10-12T09:06:11 -!- dekar_ [~dekar@212.255.33.22] has quit [Ping timeout: 255 seconds]
2012-10-12T09:07:59 -!- lawrenceseattle_ [~anonymous@216.243.49.130] has quit [Ping timeout: 255 seconds]
2012-10-12T09:12:57 < jpa-> upgrdman: because if they only do milling after plating, and only normal drilling before it
2012-10-12T09:13:20 < upgrdman> o
2012-10-12T09:14:36 < dongs> does oval hole require milling?
2012-10-12T09:14:43 < dongs> i thought they just run drill close together
2012-10-12T09:14:58 < Tectu> o0
2012-10-12T09:15:02 < jpa-> not sure
2012-10-12T09:15:07 < Tectu> I'd mill it for sure
2012-10-12T09:15:16 < dongs> not if the hole is like 1mm
2012-10-12T09:15:19 < Tectu> not depending if any fast'n'dirty stuff or not
2012-10-12T09:15:22 < dongs> how yo gonna get a tool in tehre
2012-10-12T09:15:31 < Tectu> dongs, nano mill?
2012-10-12T09:15:39 < Tectu> mills with 0.1mm tools
2012-10-12T09:15:43 < dongs> dunno
2012-10-12T09:15:53 < Tectu> dongs, or water-jet / laser
2012-10-12T09:15:56 < dongs> k now im assburgering again with nothing to do
2012-10-12T09:15:59 < Tectu> if possible
2012-10-12T09:16:34 < Tectu> dongs, is it an oval hole you need once or do you need a few of them? what for a material? in a plate or something like a box?
2012-10-12T09:16:48 < dongs> Tectu: discussion was about oval holes in pcb
2012-10-12T09:17:02 < dongs> i dunno how its made either. i just guessed they ram the drill repeatedly with small offset.
2012-10-12T09:17:08 < Tectu> dongs, oh, you want your PCB manufacturer do it?
2012-10-12T09:17:23 < dongs> they already do. but emeb_mac said none of his cheap fabs did oval holes
2012-10-12T09:17:29 < Tectu> dongle, no, a PCB manufacturer can do oval holes with 1mm without any problem
2012-10-12T09:17:31 < dongs> which I find kinda hard to believe, it shouldnt be THAT much extra work
2012-10-12T09:17:43 < Tectu> yes, a serious manufacturer, of course
2012-10-12T09:17:43 < dongs> yeah mine doesnt even ask any questions about oval shit
2012-10-12T09:17:51 < Tectu> yeah
2012-10-12T09:18:10 < Tectu> i guess everything above 0.3mm shouldn't be ANY problem
2012-10-12T09:18:17 < dongs> emeb needs to hang out wiht the pros
2012-10-12T09:18:32 < Tectu> yeah
2012-10-12T09:18:33 < emeb_mac> that's why I'm here. ;)
2012-10-12T09:18:44 < Tectu> emeb_mac, where do you life? europa?
2012-10-12T09:18:50 < emeb_mac> AZ USA
2012-10-12T09:18:52 < dongs> i think amerikkka
2012-10-12T09:18:56 < Tectu> aw
2012-10-12T09:19:02 < Tectu> emeb_mac, do you have a company?
2012-10-12T09:19:22 < Tectu> emeb_mac,   try   eurocircuit     they are very very very very very nice and professional, but only deliver to companies
2012-10-12T09:19:23 < emeb_mac> Tectu: I do contract work.
2012-10-12T09:19:47 < Tectu> also, who the hell nees a 1mm oval hole in a PCB?
2012-10-12T09:19:53 < emeb_mac> Mostly DSP design for software radio. Embedded coding for audio applications.
2012-10-12T09:19:54 < dongs> eruoricruit sounds liek a ripoff already
2012-10-12T09:19:58 < dongs> cuz it has 'euro' in the name
2012-10-12T09:20:08 < emeb_mac> All my PCB designs are personal use only. Hobby stuff.
2012-10-12T09:20:12 < Tectu> dongs, most professional PCB manufacturer ever
2012-10-12T09:20:20 < Tectu> emeb_mac, then try   leit-on.de
2012-10-12T09:20:26 < Tectu> they'll do it for sure
2012-10-12T09:20:32 < Tectu> but they are very expensive when you need single PCBs
2012-10-12T09:20:41 < Tectu> emeb_mac, might I guess? you asked seeed to do the oval stuff? :D
2012-10-12T09:20:59 < emeb_mac> I gotta be honest - I don't know for sure if ovals are excluded from all the outfits I've used.
2012-10-12T09:21:26 < emeb_mac> Most of my PCBs have been done thru SparkFun's BatchPCB service though and I'm pretty sure they only drill round holes.
2012-10-12T09:21:30 < Tectu> there shouldn't be any limitations of the shape they have to mill, emeb_mac
2012-10-12T09:21:46 < Tectu> it's really just the smalles diameter thing, which whould be definitley over 0.5mm and mostly over 0.3mm
2012-10-12T09:21:55 < dongs> STANDARD pool6.028 EUR301.40 EUR30.47 EUR0.00 EUR331.87 EUR
2012-10-12T09:21:58 < dongs> thats pretty ripoffish
2012-10-12T09:22:11 < Tectu> dongs, euro, as ya said
2012-10-12T09:22:30 < Tectu> dongs, dosen't matter, when the average salary of your country is 5'000€
2012-10-12T09:23:12 < dongs> also they dont really have a way to order a fr4 panel
2012-10-12T09:23:18 < dongs> like 1m^2 of shit to fit x panelized boards
2012-10-12T09:23:21 < dongs> so i kinda guessed
2012-10-12T09:23:24 < Tectu> dongs, I guess you're doing it wrong
2012-10-12T09:23:30 < Tectu> dongs, no
2012-10-12T09:23:35 < Tectu> dongs, they even just make you one PCB
2012-10-12T09:23:39 < dongs> i dont need one pcb.
2012-10-12T09:23:44 < Tectu> dongs, made one PCB, 52x62mm
2012-10-12T09:23:56 < dongs> i have a local fab that does the job just fine.
2012-10-12T09:25:06 < Tectu> dongs, you know the difference between STANDARD and TECH pool?
2012-10-12T09:25:09 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 260 seconds]
2012-10-12T09:25:13 < Tectu> beside it's twice the price
2012-10-12T09:25:27 < dongs> finer drill/trace tolerance i guess
2012-10-12T09:25:37 < Tectu> oval holes support
2012-10-12T09:25:37 < Tectu> xxD
2012-10-12T09:28:58 < dongs> http://0x.ca/sim/tmp/ltspice_100nf_between_motor_phases.png this is pretty hax0r shit
2012-10-12T09:33:58 < Tectu> yeah, that's what I am learning at school
2012-10-12T09:33:59 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-12T09:34:27 < Tectu> also building PID regulator out of analog circuits
2012-10-12T09:34:41 < Tectu> but useful things, not only demonstrations which don't work
2012-10-12T09:38:52 -!- james [~james@bas1-toronto05-1177963637.dsl.bell.ca] has joined ##stm32
2012-10-12T09:39:16 -!- james is now known as Guest7186
2012-10-12T09:40:21 -!- Guest7186 [~james@bas1-toronto05-1177963637.dsl.bell.ca] has quit [Client Quit]
2012-10-12T09:40:58 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-12T09:43:26 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-12T09:46:23 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T09:47:10 -!- Claude is now known as claude
2012-10-12T09:57:46 < emeb_mac> Here's what BatchPCB says: "All drill hits are plated through. No internal routes, no v-scoring, only drill files are sent to the fab house."
2012-10-12T09:58:32 < emeb_mac> No internal routes suggests that holes must be round. I don't think they like to do overlapping drills.
2012-10-12T09:59:45 -!- dekar [~dekar@212.255.45.64] has joined ##stm32
2012-10-12T09:59:48 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-12T09:59:59 -!- dekar [~dekar@212.255.45.64] has quit [Remote host closed the connection]
2012-10-12T10:01:04 < dongs> no internal routes just means dont make slots in boards
2012-10-12T10:01:15 < dongs> thats different from overlapping drills I'd guess
2012-10-12T10:03:04 < emeb_mac> makes sense.
2012-10-12T10:03:19 -!- dekar__ [~dekar@212.255.43.16] has quit [Ping timeout: 260 seconds]
2012-10-12T10:05:13 -!- tavish_ [~tavish@120.59.39.108] has joined ##stm32
2012-10-12T10:07:17 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-12T10:09:13 -!- dekar [~dekar@212.255.45.64] has joined ##stm32
2012-10-12T10:09:16 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-12T10:10:09 -!- dekar [~dekar@212.255.45.64] has quit [Client Quit]
2012-10-12T10:11:35 < emeb_mac> it's probably no biggie - most oval holes can be replaced with round holes of sufficient diameter.
2012-10-12T10:11:43 < emeb_mac> if you're willing to use a lot of solder...
2012-10-12T10:18:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-12T10:19:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-12T10:23:45 < Tectu> what's the difference between   DMA    and  DMA2?
2012-10-12T10:25:01 < Tectu> just another DMA controller like you can have two ADCs?
2012-10-12T10:29:14 < dongs> different peripherals on them if you notice.
2012-10-12T10:30:42 < zyp> yep
2012-10-12T10:30:51 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-10-12T10:30:53 < zyp> i.e. different sources for dma requests
2012-10-12T10:33:28 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-12T10:38:42 -!- Laurenceb_ [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has joined ##stm32
2012-10-12T10:39:09 < dongs> trolluenceb_ any more Tectu slipups
2012-10-12T10:50:34 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-12T10:50:37 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-12T10:50:53 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-10-12T10:51:58 < Tectu> does an F1 have two DMAs as well? o0
2012-10-12T11:03:25 -!- Laurenceb_ [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-12T11:07:23 < dongs> Tectu: they all do..
2012-10-12T11:13:40 < zyp> no, there are some F1s with only one DMA
2012-10-12T11:15:17 < zyp> The DMA2 controller is available only in high-density and XL-density devices.
2012-10-12T11:15:37 < zyp> so low and medium-density only has DMA1
2012-10-12T11:18:54 -!- dekar [~dekar@dslb-088-069-147-184.pools.arcor-ip.net] has joined ##stm32
2012-10-12T11:18:58 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-12T11:46:40 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-10-12T11:55:34 < jpa-> and for some reason really many STM32F parts have errata about the two DMA controllers conflicting, causing hardfaults or lost data
2012-10-12T11:55:54 < jpa-> so sometimes you can't use DMA2 anyway
2012-10-12T12:02:44 -!- |sark [~|sark@109.235.51.202] has quit [Remote host closed the connection]
2012-10-12T12:16:47 < cjbaird> THROUGH EARLY MORNING FOG I SEE
2012-10-12T12:16:56 < zyp> ok
2012-10-12T12:19:36 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 272 seconds]
2012-10-12T12:20:37 < dongs> attn Laurenceb http://i.imgur.com/N1R8i.jpg
2012-10-12T12:22:13 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-12T12:29:35 -!- BrainDamage [~BrainDama@62.10.7.162] has joined ##stm32
2012-10-12T13:02:56 < dongs> no blog
2012-10-12T13:05:44 < dongs> http://www.gamingjobsonline.com/images/banner/banner300x250.jpg
2012-10-12T13:10:59 < Tectu> "never before seen"
2012-10-12T13:11:40 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-12T13:14:15 < Laurenceb> looks legit
2012-10-12T13:17:50 < dongs> so Laurenceb is that your post < dongs> attn Laurenceb http://i.imgur.com/N1R8i.jpg
2012-10-12T13:18:16 < Laurenceb> lol
2012-10-12T13:21:45 < dongs> fuck, so bored
2012-10-12T13:21:47 < dongs> gonna go to sauna bbl
2012-10-12T13:22:08 < BrainDamage> going to get his daily dose of dongs
2012-10-12T13:23:41 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-10-12T13:24:58 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has quit [Quit: No Ping reply in 180 seconds.]
2012-10-12T13:25:42 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-12T13:29:44 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-12T13:41:14 -!- Netsplit *.net <-> *.split quits: Nutter, @ChanServ, phantoxeD
2012-10-12T13:41:14 -!- Netsplit *.net <-> *.split quits: ntfreak, jaeckel, TeknoJuce
2012-10-12T13:41:22 -!- Netsplit *.net <-> *.split quits: zlog, LeelooMinai, GargantuaSauce_, nashpa, esden, Laurenceb, dongs, feurig, gsmcmullin, ossifrage,  (+1 more, use /NETSPLIT to show all of them)
2012-10-12T13:44:58 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-12T13:44:58 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-12T13:44:58 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-12T13:44:58 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-12T13:44:58 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-12T13:44:58 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-12T13:44:58 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-12T13:44:58 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-10-12T13:44:58 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-12T13:44:58 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-12T13:44:58 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-yurcjvhaaoafkjok] has joined ##stm32
2012-10-12T13:44:58 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-10-12T13:44:58 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-12T13:44:58 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-12T13:44:58 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-10-12T13:44:58 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-10-12T13:44:58 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-12T13:44:58 -!- feurig [~don@198.202.31.180] has joined ##stm32
2012-10-12T13:44:58 -!- ServerMode/##stm32 [+o ChanServ] by lindbohm.freenode.net
2012-10-12T13:45:03 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection timed out]
2012-10-12T13:45:27 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-12T13:59:20 < Laurenceb> http://store.diydrones.com/APM_2_5_Kit_p/br-ardupilotmega-05.htm
2012-10-12T13:59:28 < Laurenceb> omg its so awesome
2012-10-12T13:59:32 < Laurenceb> dual core bro
2012-10-12T14:05:24 < Laurenceb> "the PPM encoder (Atmega32-U2) "
2012-10-12T14:05:26 < Laurenceb> lolling
2012-10-12T14:09:36 -!- tavish_ [~tavish@120.59.39.108] has quit [Ping timeout: 245 seconds]
2012-10-12T14:11:44 -!- tavish [~tavish@120.59.41.40] has joined ##stm32
2012-10-12T14:12:07 -!- tavish is now known as Guest68122
2012-10-12T14:17:28 < dongs> irc from sauna... wantanypics Laurenceb?
2012-10-12T14:17:52 < dongs> bbl
2012-10-12T14:18:29 < Laurenceb> ewww
2012-10-12T14:18:44 < cjbaird> yaranaika?
2012-10-12T14:18:59 < Laurenceb> www.gay-sauna.com/
2012-10-12T14:23:11 -!- Guest68122 is now known as tavish3
2012-10-12T14:29:59 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-12T14:30:12 -!- claude is now known as claude_work
2012-10-12T14:32:13 < cjbaird> must be enjoying himself...
2012-10-12T14:44:43 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-12T14:53:26 < dongs> lol
2012-10-12T14:57:53 -!- kerwood_wk [80f40908@gateway/web/freenode/ip.128.244.9.8] has joined ##stm32
2012-10-12T15:09:16 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-12T15:17:10 < dongs> http://www.gay-sauna.com/images/rightadbanner.gif
2012-10-12T15:17:16 < dongs> im not surprised Laurenceb knows this
2012-10-12T15:18:30 < Laurenceb> dongs is a dolphin?
2012-10-12T15:18:32 < Laurenceb> http://www.youtube.com/watch?v=xp7nkImBL6g
2012-10-12T15:21:12 < Tectu> so, it's  «share your DAC code»  time
2012-10-12T15:21:25 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-12T15:22:02 < dongs> <not for chibios>
2012-10-12T15:29:24 -!- tavish3 [~tavish@120.59.41.40] has quit [Quit: leaving]
2012-10-12T15:38:56 < Tectu> is it just me or does the reference manual not say anything about the maximumg DAC output frequency?
2012-10-12T15:41:14 < dongs> ..
2012-10-12T15:41:23 < dongs> what's a "maximum dac output frequency"
2012-10-12T15:41:35 < Tectu> yes, I just got that
2012-10-12T15:41:50 < Tectu> it's more the maximum conversion time
2012-10-12T15:42:30 < Laurenceb> depends what resolution you want
2012-10-12T15:42:31 < claude_work> Tectu, look for 'settling time' in the datasheet ... the DAC itself is fast but don't expect too much from it ;)
2012-10-12T15:42:51 < Laurenceb> if you want ~12bit you are limited to <100khz or so
2012-10-12T15:43:14 < claude_work> iirc ~4uS on F1 parts
2012-10-12T15:43:28 < Laurenceb> its good enough for audio
2012-10-12T15:44:39 < Tectu> so no 10MHz sine output with 10bit?
2012-10-12T15:45:38 < dongs> haha.
2012-10-12T15:45:39 < dongs> no.
2012-10-12T15:46:03 < Laurenceb> Felix Baumgartner must be embarrassed ....all those people watching ... and the poor chap cant get it up.
2012-10-12T15:46:13 < Laurenceb> which means sikipedia is back
2012-10-12T15:46:16 < Laurenceb> wohoo
2012-10-12T15:46:47 < claude_work> Tectu, yeah like Laurenceb said , around 100khz-200khz maybe
2012-10-12T15:46:58 < Tectu> thanks claude_work and Laurenceb
2012-10-12T15:47:30 < Laurenceb> use a resistor ladder
2012-10-12T15:47:33 < Tectu> is there some easy to use external DDS our there?
2012-10-12T15:47:44 < Tectu> Laurenceb, and GPIO?
2012-10-12T15:47:49 < Tectu> R2R??
2012-10-12T15:47:57 < Laurenceb> it might just work
2012-10-12T15:48:41 < Tectu> I'd need a sine from <10Hz to 100MHz
2012-10-12T15:48:46 < Tectu> controllable over USB
2012-10-12T15:50:48 < qyx_> external dac with gpio/dma
2012-10-12T15:51:10 < karlp> 100meg sine? good times
2012-10-12T15:51:28 < karlp> function generator was our second year uni project.
2012-10-12T15:51:43 < Tectu> qyx_, can you recommend any DAC?
2012-10-12T15:51:48 < karlp> we did it all analog, it was awesome
2012-10-12T15:51:55 < Tectu> qyx_, also, afaik F1 cannot access GPIO with DMA, right?
2012-10-12T15:51:59 < Tectu> karlp, got something to share?
2012-10-12T15:52:02 < dongs> http://www.youtube.com/watch?v=7JlhMAJfDMo
2012-10-12T15:52:04 < dongs> great stuff
2012-10-12T15:52:05 < Tectu> karlp, I am quite good with analogs ^^
2012-10-12T15:52:17 < karlp> nah, it went to maybe 2meg with THD falling off the whole time
2012-10-12T15:52:27 < karlp> if I was doing it today I'd do DDS all the way
2012-10-12T15:52:39 < karlp> I'm not the analog dude anyway
2012-10-12T15:52:41 < Tectu> karlp, yeah, that's what I just asked
2012-10-12T15:53:05 < karlp> I just found the secret appnote from natsemi on their laser matched pnp pairs for building wein bridges
2012-10-12T15:53:38 < Tectu> secret?
2012-10-12T15:53:45 < karlp> well, we were all only second year,
2012-10-12T15:54:03 < karlp> itw asn't really secret, just that not all of us had learnt the ways of the interet
2012-10-12T15:54:07 < karlp> also, pre-google
2012-10-12T15:54:51 < Tectu> ah, right, you're old
2012-10-12T15:54:56 < Tectu> ;-)
2012-10-12T15:54:59 < karlp> not that old, just old enough
2012-10-12T15:55:05 < qyx_> Tectu: but you cannot do 100M sine even with ext dac, you would require at least 200 mhz dac which is beyond the capabilities of dma
2012-10-12T15:55:08 < karlp> google beta'd in 4th year I think
2012-10-12T15:56:13 < Tectu> DDS just seems to a bit heavy just for some hobby projects ^^
2012-10-12T15:56:27 < Tectu> is even 10MHz possible?
2012-10-12T15:56:57 < dongs> someone make a nice signal generator with a external dac
2012-10-12T15:57:02 < dongs> DDS or whatever
2012-10-12T15:57:23 < qyx_> Tectu: it depends
2012-10-12T15:57:25 < karlp> AD have a DDS chip
2012-10-12T15:57:39 < Tectu> karlp, they do have DDS chips up to 135$ :D
2012-10-12T15:57:41 < qyx_> if you want nice looking sine you would require faster dma & dac
2012-10-12T15:57:42 < karlp> but I don't think it goes to 10meg
2012-10-12T15:57:49 < Tectu> with 1GHz output
2012-10-12T15:57:54 < Tectu> karlp, sure!
2012-10-12T15:58:01 <+dekar> I made a carmount for my phone out of office supplies
2012-10-12T15:58:01 < Tectu> karlp, http://www.analog.com/en/rfif-components/direct-digital-synthesis-dds/products/index.html#Direct_Digital_Synthesis
2012-10-12T15:58:04 <+dekar> https://dl.dropbox.com/u/28467113/phone%20car%20mount/IMG_20121012_144734.jpg
2012-10-12T15:58:05 < karlp> I was talking about things in the <$2 range
2012-10-12T15:58:10 <+dekar> https://dl.dropbox.com/u/28467113/phone%20car%20mount/IMG_20121012_144744.jpg
2012-10-12T15:58:14 < karlp> sure, if I want to spend a pile of money I can get a pile of features
2012-10-12T15:58:29 < karlp> but at that price, I might as well just buy a fancy off the shelf function generator
2012-10-12T15:58:41 < Tectu> dekar, congratu-fucky-lations :D
2012-10-12T15:58:49 < karlp> ti's AN263 is the sine generation techniques app note
2012-10-12T15:59:34 < Tectu> karlp, well, what I want to do is some simple signal/function generator for prototyping which can be configured over USB. having sine, rect, triangle output and also noise
2012-10-12T16:00:00 < qyx_> you can use long wire for noise generation
2012-10-12T16:00:05 < karlp> and as I think we've all come to, it's all going to depend on how high a freq you want to go to :)
2012-10-12T16:00:31 < karlp> you're original goal of 100meg sine will be quite an adventure :)
2012-10-12T16:01:29 < zyp> the thing about lab equipment is that you need to be able to trust it
2012-10-12T16:01:33 < Tectu> karlp, as far as I can tell after a few minutes of google, everything >1meg is adventure :D
2012-10-12T16:01:48 < karlp> zyp: true dat
2012-10-12T16:02:10 < qyx_> zyp: don't you trust your own projects? :X
2012-10-12T16:02:14 < Tectu> that as well
2012-10-12T16:02:20 < claude_work> Tectu, using a R2R on a FPGA to 'bitbang' PAL Video , so yes 10MHz is perfectly possible with R2R.
2012-10-12T16:02:49 < zyp> qyx_, no, do you?
2012-10-12T16:02:54 < claude_work> R2R + stm32f1 fsmc should do the job
2012-10-12T16:03:15 < Tectu> claude_work, is fsmc faster than GPIO? isn't FSMC just about CPU time?
2012-10-12T16:03:15 < dongs> isnt r2r fucking ghetto
2012-10-12T16:03:18 < qyx_> zyp: me neither
2012-10-12T16:05:39 < qyx_> yesterday i ordered 15 pcbs with f107 & ethernet for ip thermometers and i think it just won't work
2012-10-12T16:05:51 < zyp> :)
2012-10-12T16:05:58 < dongs> assembled?
2012-10-12T16:05:58 < zyp> sounds fun
2012-10-12T16:06:01 < qyx_> despite the fact that prototype worked fine
2012-10-12T16:06:04 < qyx_> not assembled
2012-10-12T16:06:18 < dongs> my shit workrs on first try usually with minimal or no rework
2012-10-12T16:06:33 < qyx_> prototype worked also
2012-10-12T16:06:39 < dongs> then youre probably ok.
2012-10-12T16:06:39 < qyx_> except the magjack connector
2012-10-12T16:06:42 < qyx_> wrong lib in eagle
2012-10-12T16:06:54 < dongs> lol, eagle...
2012-10-12T16:06:57 < qyx_> pst
2012-10-12T16:06:59 < dongs> never trust opensauce bullshit
2012-10-12T16:07:12 < dongs> 'sparkfun eagle library' is so utterly full of fail
2012-10-12T16:07:17 < qyx_> i use to draw my own libs, but that time i was too lazy
2012-10-12T16:07:23 < dongs> anyone using anything from there is automatically doomed to failure
2012-10-12T16:07:24 < qyx_> yes, it was from sparkfun :D
2012-10-12T16:08:45 < Tectu> lol
2012-10-12T16:13:57 < claude_work> Tectu, cpu time , because you can DMA to the fsmc
2012-10-12T16:14:07 < dongs> you can dma to GPIO too..
2012-10-12T16:14:12 < dongs> can you dma to bitband regions?
2012-10-12T16:14:15 < Tectu> dongle, on f1 as well?
2012-10-12T16:14:19 < dongs> yes
2012-10-12T16:14:27 < claude_work> oh okay didn't thought of that
2012-10-12T16:14:30 < Tectu> okay
2012-10-12T16:14:36 < Tectu> claude_work, could be easier then :)
2012-10-12T16:15:15 < dongs> setup would probably be about samea..
2012-10-12T16:15:16 < zyp> dongs, no
2012-10-12T16:15:21 < dongs> no -> bitband?
2012-10-12T16:15:25 < zyp> yep
2012-10-12T16:15:35 < dongs> well, for his purposes just dma to ODR or wahtever should be enough
2012-10-12T16:15:39 < zyp> bitbanding happens in cpu core, not visible to other bus masters
2012-10-12T16:16:26 < dongs> mkay
2012-10-12T16:16:46 < dongs> plus with gpio DMA he could have used a smaller package...
2012-10-12T16:16:55 < dongs> I think even 48lqfp has PA0..15 all available
2012-10-12T16:17:00 < dongs> so you could do 16bit ghettoDAC
2012-10-12T16:17:23 < dongs> fsmc is 100+pin only?
2012-10-12T16:17:35 < Tectu> f4 max GPIO frequency is the same as for F1?
2012-10-12T16:17:38 < Tectu> 50MHz?
2012-10-12T16:17:43 < dongs> double that.
2012-10-12T16:18:21 < Tectu> f1 = 50Mhz,  f4 = 100Mhz?
2012-10-12T16:18:24 < dongs> y
2012-10-12T16:18:29 < Tectu> hmm
2012-10-12T16:18:51 < Tectu> so with 100Mhz and 10bit, I could do 10Mhz Sine or is that wrong?
2012-10-12T16:19:19 < dongs> maybe someone who actually knows shit can confirm/deny.
2012-10-12T16:21:15 < claude_work> Tectu, thats the PAL video output with an R2R DAC (9bit , 1 bit for sync , 8 bits for video) . look at the rise and fall times . way faster than then 1uS , didn't measured it in deep but i guess 10MHz sine is no problem at all . R is around 400 Ohms : https://docs.google.com/open?id=0B_5UXaEUPGdZN290NTZyMHJXS2c
2012-10-12T16:21:43 < Tectu> what's    PAL video?
2012-10-12T16:22:03 < claude_work> composite video , the yellow socket on your tv
2012-10-12T16:22:12 < Tectu> aah
2012-10-12T16:22:18 < Tectu> claude_work, kinda nice, is it f1?
2012-10-12T16:22:51 < dongs> claude_work: 10us/div?
2012-10-12T16:22:52 < claude_work> nah FPGA , but only with 12MHz on the 'gpios'
2012-10-12T16:23:34 < Tectu> claude_work, so with F4 and 100Mhz GPIO and let's say 12bit, how fast can I go? not sure how to calulate that right
2012-10-12T16:24:57 < claude_work> um 100MHz is RF black magic for me :) but well you get 100MSPs per second then . so 100MHz 12 bit sine ?
2012-10-12T16:25:28 < claude_work> um wait forget that
2012-10-12T16:25:33 < Tectu> so doing 10Mhz sine with 12Bit R2R on f1 is really no problem?
2012-10-12T16:25:46 < dongs> why you wanna do 10mhz sine
2012-10-12T16:25:50 < dongs> YOU CANT EVEN HEAR IT!!!
2012-10-12T16:26:21 < qyx_> nodgs pls
2012-10-12T16:26:58 < qyx_> Tectu: it depends on how much samples you want/require for one sine cycle
2012-10-12T16:27:36 < qyx_> you need at least double amount of samples for given sine freq
2012-10-12T16:27:37 < claude_work> well you get 100 sampling points on your 10MHz sine curve then ...
2012-10-12T16:27:51 < claude_work> with 100MHz sampling rate..
2012-10-12T16:28:13 < claude_work> don't know how fast the F1 gpio or fsmc can go
2012-10-12T16:28:14 < qyx_> with 100msps you can theoretically do 50MHz wave
2012-10-12T16:28:31 < qyx_> but it will hardly be nice sine
2012-10-12T16:28:44 < claude_work> -_-  <---- 50MHz sine at 100msps ;)
2012-10-12T16:28:55 < Tectu> claude_work, f1 GPIO 50Mhz
2012-10-12T16:28:57 < dongs> more like ^_^
2012-10-12T16:29:27 < claude_work> ლ(ಠ益ಠლ) !
2012-10-12T16:30:10 < dongs> no unicode bro
2012-10-12T16:30:12 < claude_work> Tectu, 5MHz ? with 10 sampling points
2012-10-12T16:30:40 < Tectu> claude_work,   10 sampling points per cycle????
2012-10-12T16:30:49 < qyx_> at least
2012-10-12T16:30:52 < claude_work> as rule of thumb yeah
2012-10-12T16:30:59 < Tectu> you can even feel with your finger that that isn't a nice sine :D
2012-10-12T16:32:02 < qyx_> maybe you can do 32bit dma at 100Mhz and use some fancy delay logic
2012-10-12T16:32:07 < qyx_> with 400MHz dac
2012-10-12T16:32:11 < qyx_> at 8bit
2012-10-12T16:32:47 < qyx_> but i think you can do dma at maximum of sysclk/4 rate
2012-10-12T16:32:53 < qyx_> someone mentioned that here last time
2012-10-12T16:35:33 -!- daku is now known as DaKu
2012-10-12T16:38:13 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Read error: Connection reset by peer]
2012-10-12T16:38:34 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-12T16:38:35 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-12T16:38:35 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-12T16:47:50 < Tectu> dongs, are you sure that F1 can access GPIO over DMA?
2012-10-12T16:48:08 < Tectu> the drawing in the reference manual dosen't show the GPIO
2012-10-12T16:50:09 < zyp> huh?
2012-10-12T16:50:55 < dongs> tectu, which drawing?
2012-10-12T16:51:01 < dongs> Figure 48 clearly shows GPIOx there
2012-10-12T16:51:38 < Tectu> where? i only see ADC; USART, TIM and stuff
2012-10-12T16:51:49 < Tectu> in Figure 48
2012-10-12T16:52:56 < Tectu> dongs, are you sure that you are in F1 ref manual?
2012-10-12T16:53:11 < dongs> yase
2012-10-12T16:53:42 < Tectu> where the hell do you see GPIOx
2012-10-12T16:54:13 < dongs> hm, shitty. i think GPIO DMA is only on connectivity line devices :(
2012-10-12T16:54:19 < dongs> for f1.
2012-10-12T16:54:30 <+dekar> how do you drive GPIO at 100mhz if DMA can't do it?
2012-10-12T16:54:45 < dongs> well, you don't do it at 100mhz on F1 anyway.
2012-10-12T16:55:20 < zyp> dekar, speed limits apply to both GPIO and AF mode
2012-10-12T16:55:29 < dongs> anyway, wouldnt GPIO dma just be memory to memory
2012-10-12T16:55:34 < Tectu> dongs, i don't see GPIOx in fig 48 for the connectivity line as well...
2012-10-12T16:55:39 < dongs> i.e. from ram block to GPIOx->ODR
2012-10-12T16:56:40 < dongs> Tectu: one sec uploading kawaii screenshot
2012-10-12T16:56:41 <+dekar> dongs, well can you toggle at 50mhz on F1?
2012-10-12T16:56:49 < dongs> yea
2012-10-12T16:56:57 < dongs> http://i.imgur.com/jH2KD.png
2012-10-12T16:56:58 <+dekar> how do you do it? DMA?
2012-10-12T16:57:30 < Tectu> dongs, we have a different datasheet
2012-10-12T16:57:38 < Tectu> RM0008 dosen't show any GPIO in figure 48
2012-10-12T16:57:41 < dongs> i dunno, how about M2M from sram to GPIO register?
2012-10-12T16:57:51 < dongs> would that work?
2012-10-12T16:57:53 < dongs> or fale
2012-10-12T16:58:10 < dongs> wait
2012-10-12T16:58:13 < dongs> wasnt i doing this
2012-10-12T16:58:16 < dongs> for my 16uart breakout
2012-10-12T16:58:18 < dongs> lemme check the code
2012-10-12T16:58:42 < Tectu> 16 uart? dafuq?
2012-10-12T16:58:55 < dongs> yes, i did and it workoed
2012-10-12T16:59:07 < dongs> http://bcas.tv/paste/results/7C3Jut83.html
2012-10-12T16:59:49 < dongs> memory to peripheral
2012-10-12T17:00:06 < Tectu> dude, 16 uart?
2012-10-12T17:00:18 < dongs> ya
2012-10-12T17:00:23 < dongs> output only, so its easy
2012-10-12T17:00:29 < dongs> and syncronized.
2012-10-12T17:02:10 < Tectu> yes, but 16?
2012-10-12T17:02:21 < dongs> dmxfags wanna control a lot of lights
2012-10-12T17:02:43 < Tectu> so give them light switches
2012-10-12T17:02:57 < dongs> that'd be like 8000 switches
2012-10-12T17:03:05 < Tectu> exactly!
2012-10-12T17:03:09 < Tectu> still better than 16 uart
2012-10-12T17:03:10 < dongs> almost over 9000.
2012-10-12T17:03:23 < Tectu> level over 9990
2012-10-12T17:12:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-12T17:43:19 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T17:44:00 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-12T17:44:03 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-12T17:47:46 -!- tavish_ [~tavish@120.59.133.252] has joined ##stm32
2012-10-12T17:48:45 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-10-12T17:53:12 -!- claude_work is now known as claude
2012-10-12T17:55:25 -!- tavish_ [~tavish@120.59.133.252] has quit [Ping timeout: 265 seconds]
2012-10-12T17:55:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-12T18:07:14 -!- dekar [~dekar@dslb-088-069-147-184.pools.arcor-ip.net] has quit [Ping timeout: 260 seconds]
2012-10-12T18:11:04 -!- dekar [~dekar@dslb-088-069-147-184.pools.arcor-ip.net] has joined ##stm32
2012-10-12T18:11:05 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-12T18:25:38 -!- kerwood_wk [80f40908@gateway/web/freenode/ip.128.244.9.8] has left ##stm32 []
2012-10-12T18:25:39 <+dekar> today I killed my first tantal
2012-10-12T18:25:46 <+dekar> dangerous things
2012-10-12T18:26:35 < jpa-> did it bite you?
2012-10-12T18:26:35 <+Steffanx> With tantal you mean a tantalum capacitor i guess?
2012-10-12T18:26:58 <+Steffanx> Someone from the FBI shouldn't mess with electronics anyway, dekar
2012-10-12T18:27:23 < Laurenceb> dekar is from the FNI/
2012-10-12T18:27:27 < Laurenceb> *FBI
2012-10-12T18:27:44 <+Steffanx> Fachbereich informatik :P
2012-10-12T18:27:53 < Laurenceb> wtf is that
2012-10-12T18:28:10 <+Steffanx> Google it
2012-10-12T18:28:14 < Laurenceb> Technische Universität Darmstadt
2012-10-12T18:28:16 < Laurenceb> i see
2012-10-12T18:28:42 < Laurenceb> university of techno
2012-10-12T18:29:27 <+Steffanx> computer science stuff ..
2012-10-12T18:30:20 < Laurenceb> s/computer science/spotty loser basement dwelling nerd stuff
2012-10-12T18:30:25 < jpa-> hi Steffanx :))
2012-10-12T18:30:48 <+Steffanx> Hi mr jpa-
2012-10-12T18:30:54 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T18:33:30 < jpa-> Steffanx: so, any weekend plans?
2012-10-12T18:33:45 <+Steffanx> I'm going to visit you this weekend
2012-10-12T18:34:00 < jpa-> nice
2012-10-12T18:34:06 < dongs> a stm32 meat-up
2012-10-12T18:34:06 < jpa-> i can cook for you
2012-10-12T18:34:32 < jpa-> some meat food so that we can have a non-sexual meaning for the meeting's official name
2012-10-12T18:34:48 <+Steffanx> You have some good skillz?
2012-10-12T18:35:10 < jpa-> better than my stm32 skillzors.. i hope
2012-10-12T18:35:44 <+Steffanx> I had some terrible food the last time i was in finnishland so you better make something good
2012-10-12T18:36:10 < jpa-> what did you have?
2012-10-12T18:36:19 <+Steffanx> I've no idea WHAT it was
2012-10-12T18:36:53 < jpa-> maybe it's just that you are used to 50% cannabis content
2012-10-12T18:37:00 <+Steffanx> Sure
2012-10-12T18:37:20 <+Steffanx> NL != everybody uses that crap
2012-10-12T18:38:17 < jpa-> ah, just like FI != everywhere is cold.. wait no
2012-10-12T18:39:17 -!- dekar [~dekar@dslb-088-069-147-184.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-10-12T18:50:18 < Thorn> >1Gbyte nand flash seems to be hard to find.
2012-10-12T18:50:33 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-10-12T18:51:12 < Thorn> digikey has some mt29f
2012-10-12T18:51:18 < Rickta59> does the summon toolchain have a hw floating point libm.a for cortex-m4f?
2012-10-12T18:52:06 < jpa-> Thorn: yeah, for some reason microsd cards are the easiest type of large flash memory to find
2012-10-12T18:52:31 < Rickta59> i haven't compiled it but will if that is true
2012-10-12T18:53:59 < jpa-> TNT has it
2012-10-12T18:54:26 < Rickta59> is it just a matter of compiling newlib with the proper flags?
2012-10-12T18:54:36 < jpa-> yes
2012-10-12T18:54:42 < jpa-> and libgcc also
2012-10-12T18:54:47 < Rickta59> k thanks
2012-10-12T18:56:21 < Rickta59> https://github.com/EliasOenal/TNT that one?
2012-10-12T18:56:30 < jpa-> yeah
2012-10-12T18:56:36 < Rickta59> great thanks much
2012-10-12T18:57:55 -!- tavish [~tavish@120.56.233.137] has joined ##stm32
2012-10-12T18:57:55 -!- tavish [~tavish@120.56.233.137] has quit [Changing host]
2012-10-12T18:57:55 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T18:58:22 < jpa-> wtf http://www.etsy.com/listing/63188812/circuit-board-keychain-with-robot-lights?
2012-10-12T18:58:48 < jpa-> Amaze your friends and neighbors with this circuit board keychain that lights up when you plug it into a USB port.
2012-10-12T18:59:41 -!- barthess [~barthess@178.154.15.156] has joined ##stm32
2012-10-12T19:00:03 -!- zippe1 [~Adium@173.11.99.161] has joined ##stm32
2012-10-12T19:00:43 -!- zippe [~Adium@173.11.99.161] has quit [Ping timeout: 268 seconds]
2012-10-12T19:00:51 < qyx_> jpa-: :D
2012-10-12T19:00:57 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-12T19:01:08 < Thorn> why so many traces?
2012-10-12T19:03:52 < dongs> wow thats truly amazing
2012-10-12T19:06:39 < qyx_> to amaze people of course
2012-10-12T19:07:48 < dongs> would be more amazing if they used PCB traces to drop the voltage
2012-10-12T19:07:51 < dongs> instead of resistors
2012-10-12T19:07:53 < dongs> now THAT would be amazing
2012-10-12T19:08:14 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-10-12T19:09:43 < dongs> man am the only one who actually made a PCB with chip mounted inverted?
2012-10-12T19:10:28 < Laurenceb> as in ic upside down?
2012-10-12T19:10:41 < dongs> yes
2012-10-12T19:10:42 < qyx_> why would someone do that?
2012-10-12T19:11:10 < dongs> saving space?
2012-10-12T19:12:18 < dongs> http://i.imgur.com/6D4MC.jpg
2012-10-12T19:12:30 < dongs> on 1.6mm FR4 the chip doesnt stick out at all
2012-10-12T19:13:29 < qyx_> the resistors are nearly the same height as the chip
2012-10-12T19:13:34 < dongs> sure
2012-10-12T19:13:38 < dongs> but thats besides the point.
2012-10-12T19:13:44 < qyx_> if it would be mounted the same way as 99.9% people do
2012-10-12T19:13:55 < dongs> I think you meant 99.999%
2012-10-12T19:14:01 < dongs> cuz when I tried to look for a similar pic, i failed
2012-10-12T19:14:32 < Thorn> dongs is trolling his assembly house again?
2012-10-12T19:14:46 < Laurenceb> http://www.badreputation.org.uk/wp-content/uploads/2012/09/Gentlemens-Baby-Changing.jpg
2012-10-12T19:14:46 < qyx_> maybe the others don't make pics of such fail!
2012-10-12T19:14:48 < zyp> dongs is his own assembly house
2012-10-12T19:14:50 < zyp> :p
2012-10-12T19:15:37 < qyx_> but to be honest, i did mistake once and i had to reverse pins on dip8
2012-10-12T19:15:51 < qyx_> and solder it upside down
2012-10-12T19:15:52 < dongs> I did on SO8 once, fucked up pad numbering
2012-10-12T19:16:02 < dongs> i had it like...
2012-10-12T19:16:05 < dongs> 5 6 7 8
2012-10-12T19:16:06 < dongs> 1 2 3 4
2012-10-12T19:16:08 < dongs> heh
2012-10-12T19:16:14 < zyp> the only chips I've mounted upside down are deadbugs
2012-10-12T19:16:33 < qyx_> dongs: that wont work too
2012-10-12T19:16:36 <+Steffanx> lol @ pcb dong
2012-10-12T19:16:39 < qyx_> you would have to twist it
2012-10-12T19:16:49 < dongs> qyx_: right
2012-10-12T19:16:53 < dongs> i had to desolder one side
2012-10-12T19:16:56 < dongs> and get small wires
2012-10-12T19:16:59 < qyx_> ah
2012-10-12T19:17:06 < dongs> so it was kinda half lifted
2012-10-12T19:17:08 < qyx_> that twisted chip would be awesome
2012-10-12T19:17:39 < zyp> http://bin.jvnv.net/f/qHoay.JPG <- upside down cpld
2012-10-12T19:17:59 < dongs> ya nice fail there zyp
2012-10-12T19:18:05 < dongs> not anywhere as awesome as mine ;p
2012-10-12T19:18:14 <+Steffanx> duh
2012-10-12T19:18:58 < dongs> i got the clearance perfect on it too
2012-10-12T19:19:07 < zyp> well, I only did it because I wanted to test my awesome jtag code while waiting for the boards to arrive
2012-10-12T19:19:12 < dongs> the legs actually kinda springload into the pcb
2012-10-12T19:19:28 < qyx_> hm, it must be easier to align and solder
2012-10-12T19:19:39 < zyp> and because I didn't have a qfp breakout with correct pitch for the cpld to solder it to
2012-10-12T19:19:52 < dongs> qyx_: well yeah, it just sits there already aligned
2012-10-12T19:19:56 < zyp> I had both larger and smaller pitch, but not the exact
2012-10-12T19:20:12 < qyx_> hm, breakouts!
2012-10-12T19:20:17 < qyx_> dont you have goot source of them?
2012-10-12T19:20:33 < qyx_> some qfp/soic/ssop stuff
2012-10-12T19:20:36 < qyx_> *good
2012-10-12T19:20:45 < dongs> i should do stm32f3 breakout like this
2012-10-12T19:20:52 < zyp> I'm talking about having shit on hand
2012-10-12T19:21:03 < dongs> just to troll
2012-10-12T19:21:04 <+Steffanx> Make one for BGAs dong:P
2012-10-12T19:21:16 < dongs> so funny
2012-10-12T19:21:48 < zyp> hmm, it's weekend now
2012-10-12T19:21:57 < dongs> i should be sleeping
2012-10-12T19:21:58 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T19:21:58 < zyp> maybe I should troll up some pcb designs during the weekend
2012-10-12T19:22:08 < dongs> holy shit. my led matrixes have landed in japan.
2012-10-12T19:22:08 < zyp> I've got a few that I want to do
2012-10-12T19:22:13 <+Steffanx> dongs and his bad influence
2012-10-12T19:22:18 < zyp> like bmp-clone
2012-10-12T19:22:27 < zyp> and lpc4300 bga breakout
2012-10-12T19:22:55 < qyx_> but i like that dongs way of mounting chips
2012-10-12T19:23:14 < dongs> see, you gotta try it out.
2012-10-12T19:23:21 < Laurenceb> you cant route under them like that
2012-10-12T19:23:27 < qyx_> now it doestn seem to be that trolling than as on first sigh
2012-10-12T19:23:31 < dongs> you cant route under QFN either
2012-10-12T19:23:42 < zyp> dongs, sure you can, on other layers
2012-10-12T19:24:02 < dongs> implying people in here can afford > 2l pcbs
2012-10-12T19:24:02 < qyx_> but without vias from top layer
2012-10-12T19:24:07 <+Steffanx> And what about the mechanical strength qyx_ ?
2012-10-12T19:24:22 < zyp> dongs, you've still milled away the bottom layer :p
2012-10-12T19:24:29 < qyx_> Steffanx: it doesnt matter much if the design is awesome
2012-10-12T19:24:58 <+Steffanx> Don't forget to mount the leds upside down as well
2012-10-12T19:25:05 < dongs> Steffanx: well, tehy make those
2012-10-12T19:25:09 <+Steffanx> I know
2012-10-12T19:25:10 < zyp> dongs, by the way, how cheap can you do four layer boards?
2012-10-12T19:25:11 < dongs> they have "in" shining smd leds.
2012-10-12T19:25:13 -!- kerwood_wk [80f40907@gateway/web/freenode/ip.128.244.9.7] has joined ##stm32
2012-10-12T19:25:17 < dongs> zyp: cheap, depending
2012-10-12T19:25:48 < qyx_> actually you can also do passives that way
2012-10-12T19:25:50 < qyx_> not only leds
2012-10-12T19:25:54 <+Steffanx> Compared to $100k everything below 100 is cheap
2012-10-12T19:25:59 < zyp> want to do the lpc4300 bga board? :p
2012-10-12T19:26:26 < qyx_> 4300 is not avalilable in something qfp-ish?
2012-10-12T19:26:34 < zyp> sure it is
2012-10-12T19:26:40 < dongs> zyp, possibly. depends who i can panel it with etc.
2012-10-12T19:26:51 < zyp> but I just want to make some shit to see how hard it is to do bga :p
2012-10-12T19:27:11 < zyp> and if it goes ok, I get some lpc boards to play with afterwards
2012-10-12T19:27:24 < dongs> lpc4300 is the dualcore shit?
2012-10-12T19:27:26 < zyp> yes
2012-10-12T19:27:28 < dongs> m0 + M4 or whatever?
2012-10-12T19:28:33 <+Steffanx> When ST will come with their first dualcore-ish STM32?!
2012-10-12T19:29:40 < qyx_> should do quad
2012-10-12T19:29:54 < dongs> i could also probably do 2layer panels for peoples who want protos of shit for not much more than failseeedstudio but with my usual awsm quality.
2012-10-12T19:29:59 <+Steffanx> Get the release date right and win a meet and greet with dongs
2012-10-12T19:31:33 < qyx_> dongs: how cheap are your japan pcbs?
2012-10-12T19:31:48 < qyx_> 1dm^2, minimal order, 2L, 2 silkscreens, 2 soldermasks
2012-10-12T19:31:51 < qyx_> hal
2012-10-12T19:32:05 <+Steffanx> Watch out.. you'll get pink kawaii pcbs
2012-10-12T19:32:21 < Thorn> 0.2mm traces, 0.3/0.8mm vias
2012-10-12T19:32:25 < dongs> you mean 1 fr4 panel? cheap, i'd have to check.
2012-10-12T19:32:37 < qyx_> yep, fr4 1.4mm
2012-10-12T19:32:37 < Thorn> (smaller annular ring is better of course)
2012-10-12T19:32:39 < dongs> yeah, spec is 10/20 vias, 6/6 trace etc
2012-10-12T19:32:40 < qyx_> 1.5
2012-10-12T19:33:24 < zyp> hmm
2012-10-12T19:34:05 <+Steffanx> And of course non-plated holes when you want non-plated holes
2012-10-12T19:34:17 < dongs> heh yeah... fucking seeedstudio plates everythigtn
2012-10-12T19:34:24 < zyp> oshpark's 13/27 vias would be a bit problematic even under a 1mm pitch bga
2012-10-12T19:34:56 < Thorn> wtf. someone published a letter from digikey saying that they don't accept private orders from my country anymore.
2012-10-12T19:35:09 < zyp> sounds fun
2012-10-12T19:35:17 < karlp> that's rad dongs (the upsdie down chip mounting)
2012-10-12T19:35:19 < dongs> what  country is that? israel?
2012-10-12T19:35:30 < Thorn> not quite
2012-10-12T19:36:02 <+Steffanx> Time to have a good talk with mother russia Thorn
2012-10-12T19:36:28 < Thorn> she's not gonna lilsten
2012-10-12T19:36:35 <+Steffanx> Aw
2012-10-12T19:37:30 < zyp> hmm, 10/20 are just small enough for 1mm pitch, too large for 0.8mm pitch
2012-10-12T19:37:49 <+Steffanx> And you need 0.8?
2012-10-12T19:38:21 < jpa-> dongs: that upside-down IC could be nice for those electronic business cards that are ah-so-popular nowadays
2012-10-12T19:38:38 < zyp> LPC4300 is available in TFBGA100, TFBGA180 and LBGA256, TFBGA is 0.8, LBGA is 1mm
2012-10-12T19:38:39 < dongs> zyp, 20%+ for 8/16 vias since i chcked few days ago
2012-10-12T19:38:48 <+Steffanx> No, then you should use an AVR wit the Arduino stuff jpa-
2012-10-12T19:38:48 < jpa-> i wonder if you could get just the ends of a hole plated.. that way you could install also the passive components inside holes
2012-10-12T19:39:02 < dongs> jpa-: yea, except it'd have to be a attiny or something retareded that doesnt require passives
2012-10-12T19:39:16 < dongs> so you dont ruin it by adding shit sticking out
2012-10-12T19:39:16 < jpa-> Steffanx: i'm sure that dongs mounted the chip upside down just so that we wouldn't see that it is really an atmega
2012-10-12T19:39:23 <+Steffanx> Ha
2012-10-12T19:39:23 < zyp> dongs, still to large to route a trace between with 6/6 for the trace
2012-10-12T19:39:37 < dongs> zyp, 8/16 process probably does 4mil traces
2012-10-12T19:39:40 < dongs> i'd have to check though
2012-10-12T19:39:59 < zyp> 0.8mm is 31 mil
2012-10-12T19:40:09 < dongs> i've done some 5mil edge cases on 6/6 boards and they didnt bitch
2012-10-12T19:40:34 < zyp> but doesn't matter, guess I'll do the LBGA256 anyway
2012-10-12T19:40:57 < dongs> jpa-: atmega upside down looks totally different
2012-10-12T19:41:56 < dongs> http://www.electronics-lab.com/blog/wp-content/uploads/2011/08/XMEGA-USB-chip-TQFP64-001.jpg lol
2012-10-12T19:42:23 <+Steffanx> xmega ...
2012-10-12T19:42:31 < dongs> WHY NOONE TAKES PICS OF UPSIDE DOWN CHIPS FUCK
2012-10-12T19:43:10 < dongs> http://i.ebayimg.com/t/20PCS-ATmega8A-AU-16MH-Z-mega8A-TQFP-8-bit-MCU-16MH-/00/$(KGrHqEOKiME5Z3TGY(3BObiZSEmvg~~_35.JPG
2012-10-12T19:43:13 < dongs> finally, fucker
2012-10-12T19:43:20 < dongs> see al that crap on back of atmega chip
2012-10-12T19:44:31 < jpa-> i bet that you could get an stm32 working and running usb without any external components.. way outside specs of course, but working for a while :P
2012-10-12T19:45:00 < dongs> you could mount a sot23-5 vreg upside down.. but you need caps for it at least..
2012-10-12T19:45:11 < Thorn> usb requires a crystal doesn't it
2012-10-12T19:45:23 < jpa-> Thorn: only to be within spec
2012-10-12T19:45:25 < dongs> you can sorta do smd passives the same way.. just route the hole for htem and put pads on the edge
2012-10-12T19:45:39 < dongs> then carefully solder.
2012-10-12T19:45:51 < jpa-> i've verified that usb on stm32l1 works with HSI.. though i did have bypass caps when i tried that :P
2012-10-12T19:46:39 < jpa-> of course with attiny USB without external components would be trivial.. but that's only because it is a superior platform
2012-10-12T19:47:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-12T19:47:59 < dongs> yeah superior for sure.
2012-10-12T19:48:15 < dongs> keep busy  bitbanging usb
2012-10-12T19:49:17 < TitanMKD> hello
2012-10-12T19:51:14 < dongs> sup
2012-10-12T19:51:27 < karlp> blinkenlights
2012-10-12T19:52:02 < zyp> jpa-, it's surely not because LS USB requres less timing accuracy ;)
2012-10-12T19:52:55 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-12T19:54:46 < dongs> can you run the usb peripheral on stm in lowspeed mode?
2012-10-12T19:54:48 < dongs> no right>?
2012-10-12T19:55:10 < dongs> how come noone ported a bitbanging usb device from tarduino to stm32
2012-10-12T19:55:22 < dongs> sounds like a job for zyp
2012-10-12T19:56:53 < jpa-> zyp: well it just proves how awesome low performance can be!
2012-10-12T19:58:56 < dongs> http://code.google.com/p/ardupilot-mega/source/browse/libraries/AP_InertialSensor/AP_InertialSensor_MPU6000.cpp cool, dmp warez
2012-10-12T19:59:00 < dongs> they cleaned it up a bit
2012-10-12T20:01:19 < Laurenceb> i should try F1 doing usb with HSI
2012-10-12T20:01:22 < Laurenceb> ... sometime
2012-10-12T20:01:43 < zyp> dongs, right, they are FS only (except for HS)
2012-10-12T20:01:51 < dongs> failware.
2012-10-12T20:03:00 < Laurenceb> is dmp any good?
2012-10-12T20:03:05 < dongs> no
2012-10-12T20:03:08 < dongs> max 200hz output
2012-10-12T20:03:32 < Laurenceb> faster than my F1
2012-10-12T20:03:35 < Laurenceb> .. just
2012-10-12T20:03:41 < dongs> and that code doesn't integrate mag at all..
2012-10-12T20:03:52 < Laurenceb> doh
2012-10-12T20:04:12 < dongs> how come tardupilot is at 2.5 already
2012-10-12T20:04:22 < dongs> what was wrong wiht 2.0? i remember them talking about awesome  separateboard wiht gps and compass
2012-10-12T20:04:26 < dongs> and now compass is on main pcb again..
2012-10-12T20:06:49 < Laurenceb> heh
2012-10-12T20:07:01 < Laurenceb> they dont know what to do
2012-10-12T20:07:10 < Laurenceb> so they respin the same atmega fail
2012-10-12T20:07:18 < dongs> right
2012-10-12T20:07:25 < dongs> waiting for tarduino due, i sure
2012-10-12T20:07:46 < dongs> http://www.diydrones.com/profiles/blogs/visit-to-new-3d-robotics-tijuana-plant haha
2012-10-12T20:08:33 < dongs> selling atmega shit must be profitable
2012-10-12T20:09:43 < Laurenceb> The default value is 16, which, when added to the HSICAL value, should trim the HSI to 8
2012-10-12T20:09:43 < Laurenceb> MHz ± 1%. The trimming step (Fhsitrim) is around 40 kHz between two consecutive HSICAL
2012-10-12T20:09:43 < Laurenceb> steps.
2012-10-12T20:09:58 < Laurenceb> ^ F302/3 looks like it will run USB off HSI
2012-10-12T20:10:01 < Laurenceb> when trimmed
2012-10-12T20:10:17 < Laurenceb> thats 0.5%, so +-0.25% error
2012-10-12T20:11:31 < dongs> i was seeing something like 100ns jitter off hsi
2012-10-12T20:13:40 < Laurenceb> did you put it through the pll?
2012-10-12T20:13:43 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-12T20:15:17 -!- BrainDamage [~BrainDama@62.10.7.162] has quit [Quit: Leaving.]
2012-10-12T20:15:30 < dongs> Laurenceb: of course
2012-10-12T20:15:38 < dongs> to get 64mhz
2012-10-12T20:16:44 < Laurenceb> wait 100ns at 64mhz?
2012-10-12T20:16:59 < dongs> no i was running a timer  at 1mhz
2012-10-12T20:17:09 < dongs> and set duty to wahtever
2012-10-12T20:17:14 < dongs> and it was jittering around that point
2012-10-12T20:17:16 < dongs> +- 100ns
2012-10-12T20:17:32 < dongs> with crystal, its perfect
2012-10-12T20:18:07 < Laurenceb> i see
2012-10-12T20:18:28 < Laurenceb> so overflow every 64ms?
2012-10-12T20:18:37 < dongs> or something? why is that relevant
2012-10-12T20:18:56 < Laurenceb> allen variance
2012-10-12T20:19:16 < Laurenceb> i dont think that means usb is totally screwed
2012-10-12T20:19:23 < Laurenceb> ill try it when i get an F3
2012-10-12T20:19:52 -!- tavish_ [~tavish@59.177.15.240] has joined ##stm32
2012-10-12T20:20:13 < dongs> i have an F3 and im still pissed 373 is 2-pin off from being pin compatible to F1
2012-10-12T20:20:21 < dongs> FUCK ST; ST SUX; ST IS DYING; ST IS DEAD TO ME; ST HIT WTC
2012-10-12T20:20:31 < zyp> :)
2012-10-12T20:20:34 < dongs> i mean what hte fuck were tehy thinking?
2012-10-12T20:20:38 < dongs> it was probably zyp
2012-10-12T20:21:38 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-12T20:25:04 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-12T20:27:20 < Laurenceb> 302/3 is compatible
2012-10-12T20:28:56 < Thorn> btw, has anyone (outside of US) tried buying something with hardware crypto like stm32f41x from digikey? did you run into any export restrictions?
2012-10-12T20:29:06 < Thorn> or mouser for that matter
2012-10-12T20:29:47 < dongs> > Pin compatible with the STM32 F1 series, the new STM32 F3 series enlarges the STM32Â’s Cortex-M4 portfolio which now offers both entry-level cost with the F3 series and highest performance with the F4 series.
2012-10-12T20:29:57 < dongs> Thorn: yes, digikey blocks
2012-10-12T20:30:00 < dongs> mouser doesnt care though
2012-10-12T20:30:30 < dongs> Laurenceb: Pin compatible with the STM32 F1 series, the new STM32 F3 series enlarges the STM32Â’s Cortex-M4 portfolio which now offers both entry-level cost with the F3 series and highest performance with the F4 series.
2012-10-12T20:30:39 < dongs> Laurenceb: right off the front page of http://www.st.com/internet/mcu/subclass/1605.jsp
2012-10-12T20:30:46 < dongs> fucking lying scumbags
2012-10-12T20:31:20 < Laurenceb> lol
2012-10-12T20:35:00 < Laurenceb> http://www.jewtube.co.uk/
2012-10-12T20:35:19 -!- tavish [~tavish@59.177.15.13] has joined ##stm32
2012-10-12T20:35:20 -!- tavish [~tavish@59.177.15.13] has quit [Changing host]
2012-10-12T20:35:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T20:37:44 -!- tavish_ [~tavish@59.177.15.240] has quit [Ping timeout: 260 seconds]
2012-10-12T20:42:08 -!- dekar [~dekar@212.255.45.64] has joined ##stm32
2012-10-12T20:42:10 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-12T20:42:41 < Laurenceb> anyone tried driving flourescent tubes from low voltage?
2012-10-12T20:42:52 < zyp> heh, NXP are skipping some letters when naming bga rows
2012-10-12T20:42:59  * Laurenceb has some 12mm tubes
2012-10-12T20:43:01 < zyp> ABCDEFGHJKLMNPRT
2012-10-12T20:43:03 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 256 seconds]
2012-10-12T20:43:29 < Thorn> I, O are skipped
2012-10-12T20:43:46 < Thorn> afair
2012-10-12T20:43:58 < Thorn> so you don't confuse them with 1 & 0
2012-10-12T20:43:59 < zyp> and Q and S
2012-10-12T20:44:28 < Thorn> S = 5 probably
2012-10-12T20:45:01 < zyp> this is silly
2012-10-12T20:45:15 < zyp> but oh well
2012-10-12T20:45:39 -!- DaKu is now known as daku
2012-10-12T20:45:59 < jpa-> Laurenceb: cold cathode tube inverters work fine
2012-10-12T20:46:04 < zyp> made LBGA256 footprint, not going to enjoy making schematic symbol
2012-10-12T20:46:21 < Laurenceb> ok
2012-10-12T20:46:40 < Thorn> how are you going to solder it?
2012-10-12T20:47:04 < zyp> reflow oven of course
2012-10-12T20:48:21 < Laurenceb> bbl
2012-10-12T20:48:38 -!- tavish [~tavish@59.177.15.13] has joined ##stm32
2012-10-12T20:48:38 -!- tavish [~tavish@59.177.15.13] has quit [Changing host]
2012-10-12T20:48:38 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-12T20:49:34 < zyp> NXP can't count
2012-10-12T20:49:50 < zyp> «On the LPC4350/30/20/10, digital pins are grouped into 16 ports, named P0 to P9 and PA
2012-10-12T20:49:53 < zyp> to PF, with up to 20 pins used per port.»
2012-10-12T20:50:07 < zyp> and then they start listing pins
2012-10-12T20:50:21 < zyp> they have pins P1_0 - P1_20
2012-10-12T20:50:27 < zyp> which are fucking 21 pins.
2012-10-12T20:51:04 < Thorn> they never have all 21 pins in any port in my experience
2012-10-12T20:51:09 < Thorn> some are missing
2012-10-12T20:51:22 <+Steffanx> Dutch are cheap remember zyp ?
2012-10-12T20:51:29 < zyp> Thorn, I checked that, all are present.
2012-10-12T20:51:53 < Thorn> lol then
2012-10-12T20:53:52 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 272 seconds]
2012-10-12T20:54:32 < zyp> hmm, maybe this is sort of sane
2012-10-12T20:55:07 < zyp> in that a package either haves or lacks an entire group
2012-10-12T21:06:04 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-10-12T21:11:05 < Rickta59> jpa- i see that TNT uses linaro, is the gcc toolchain stuff all GPL?
2012-10-12T21:11:50 < zyp> I believe some parts may be less restrictive licensed
2012-10-12T21:13:33 < zyp> heh, there is also listed a VDD pin that is not present in any package
2012-10-12T21:18:07 < Thorn> what small and cheap character/raster displays are there except winstar & newhaven? hobbyists seem to use Chinese noname or nokia LCDs but these are hardly usable in a commercial product (no docs, no part number, no reliable sources, etc.)
2012-10-12T21:20:08 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 255 seconds]
2012-10-12T21:21:01 < zyp> oh, both usb controllers are HS capable
2012-10-12T21:21:27 < zyp> one has built in HS phy, one requires external HS phy
2012-10-12T21:23:04 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-12T21:30:21 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-12T21:37:25 < jpa-> Rickta59: who cares?
2012-10-12T21:37:34  * Steffanx cares
2012-10-12T21:37:38 < jpa-> Steffanx: why?
2012-10-12T21:37:44 <+Steffanx> because dekar cares
2012-10-12T21:37:50 < jpa-> why does dekar care?
2012-10-12T21:38:00 <+Steffanx> He is our "i care about licenses' guy
2012-10-12T21:38:40 < jpa-> but what does the license of GCC matter?
2012-10-12T21:38:54 < jpa-> he wants to publish his own propietary GCC fork?
2012-10-12T21:39:31 < jpa-> Thorn: for character displays, HD44780 is a sure bet, they'll never go away :P
2012-10-12T21:39:44 < Thorn> I wish they did
2012-10-12T21:39:51 < jpa-> i also
2012-10-12T21:40:21 < Thorn> how do you translate 3.3V to 5V for them in a compact way? (one if the signals is bidirectional iirc)
2012-10-12T21:40:42 < jpa-> the TFT market has the same problem, only single suppliers and poor docs unless you buy a million
2012-10-12T21:41:46 < Thorn> I noticed electronic scales in a store today with a seriously big character lcd. not sure winstar makes such large ones
2012-10-12T21:42:15 < jpa-> many of the modern HD44780 clones work fine with 3.3V (and also within spec)
2012-10-12T21:42:52 < HTT-Bird> Thorn: you can use an IC for that
2012-10-12T21:43:26 < Thorn> I imagine you can use an IC for anything if it's a right IC :D
2012-10-12T21:43:40 < HTT-Bird> :P voltage translator chips are rather common these days
2012-10-12T21:45:05 < Thorn> I remember some of these displays even had pads to solder some IC onto them
2012-10-12T21:45:39 < Thorn> not sure for what purpose though. some charge pump I think
2012-10-12T21:45:53 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-12T21:45:53 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-12T21:47:10 < Thorn> there's even a russian manufacturer of winstar clones http://www.melt.com.ru/shop/indikatory-zhk-3.html
2012-10-12T21:47:56 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 260 seconds]
2012-10-12T22:04:03 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-12T22:08:54 < zyp> hmm, making a sane symbol for this shit is a lot of work
2012-10-12T22:16:09 <+Steffann> 2^8 pins?
2012-10-12T22:17:12 < zyp> yes
2012-10-12T22:17:57 < zyp> I'm splitting the groups into individual sections
2012-10-12T22:18:14 < zyp> of course, each group has a different size
2012-10-12T22:18:52 < zyp> http://bin.jvnv.net/f/NFMbn.png <- so it looks all messy :p
2012-10-12T22:19:05 < karlp> hmm. stlinkv2 on my L1 disco board provides a serial number, but it's binary/unprintable chars.
2012-10-12T22:19:12 < karlp> f0 disco presents a serial number of 0000000000001
2012-10-12T22:19:30 < karlp> neither of them get presented as udev attributes to be able to give ethem sane names.
2012-10-12T22:20:44 < zyp> does names matter? doesn't texane/stlink look up through libusb anyway?
2012-10-12T22:20:54 < zyp> you could filter on serial there
2012-10-12T22:21:24 < zyp> I'm assuming you want to be able to address several at once
2012-10-12T22:31:18 < karlp> yeah
2012-10-12T22:31:39 < karlp> want to have multiple plugged in at once to help when testing out libopencm3 changes that affect multiple targets
2012-10-12T22:32:02 < jpa-> couldn't you just assign them in udev by port?
2012-10-12T22:32:10 < karlp> yeah, udev gives them a name ok,
2012-10-12T22:32:16 < karlp> I've got that, it's in the texane repo,
2012-10-12T22:32:27 < karlp> but taht still changes as they plug/unplug
2012-10-12T22:32:40 < karlp> was hopign to mount them as /dev/stlink-XAYYASERIALASDASD
2012-10-12T22:32:51 < karlp> not just /dev/stlinkv2-3 and /dev/stlinkv2-4
2012-10-12T22:32:57 < karlp> no biggie.
2012-10-12T22:33:08 < zyp> again, why not just specify serial to texane-tool and have it look up through libusb?
2012-10-12T22:33:23 < karlp> because, as far as I can tell, the f0 doesn't have a unique serial
2012-10-12T22:33:35 < karlp> and the l1 and vl have nonprintable serials.
2012-10-12T22:33:41 < karlp> just exploring what data is there.
2012-10-12T22:33:57 < jpa-> hex to the rescue
2012-10-12T22:34:17 < karlp> sure, but there has to be a way for someone to find it out to pass as the arg saying, "I want this one"
2012-10-12T22:34:25 < karlp> will probably make st-util show all the stlinks found on the system
2012-10-12T22:34:28 < karlp> or something like that.
2012-10-12T22:34:51 < zyp> you could flash them with BMP firmware, they got unique serials :p
2012-10-12T22:35:15 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-10-12T22:35:23 < zyp> (and on OS X, that's used for naming ACM device nodes)
2012-10-12T22:42:08 -!- kerwood_wk [80f40907@gateway/web/freenode/ip.128.244.9.7] has quit [Ping timeout: 245 seconds]
2012-10-12T22:43:44 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-12T22:50:16 -!- M-Kat [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has joined ##stm32
2012-10-12T22:51:53 < karlp> on osx, does it make up a serial or does it just use the serial that the device provides?
2012-10-12T22:54:20 < zyp> yes
2012-10-12T22:54:42 < zyp> as in either
2012-10-12T22:54:58 < zyp> if the device provides a serial, it's used, otherwise one is generated
2012-10-12T22:55:13 < zyp> and then the devicenode is named /dev/tty.usbmodem<serial>
2012-10-12T22:55:14 < karlp> how do they know if it's the same device again later?
2012-10-12T22:55:29 < karlp> I'd see osx do that serial appending thing, I like it,
2012-10-12T22:56:04 < zyp> I think the generated serial is based on where the device is plugged in
2012-10-12T22:56:07 < M-Kat> sup
2012-10-12T22:56:33 < karlp> oh yeah, linux has this sort of thing: /dev/serial/by-id/usb-Silicon_Labs_CP2102_USB_to_UART_Bridge_Controller_0001-if00-port0
2012-10-12T22:56:40 < karlp> but they sure don't make that common knowledge
2012-10-12T22:57:05 < karlp> in fact, one library I was working on that expected to get a /dev/ttyXXXX type name only even allowed 16chars
2012-10-12T22:57:11 < karlp> so it didn't even work on osx!
2012-10-12T22:58:00 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-12T22:59:27 < zyp> right, my own ACM demo doesn't provide a serial, so OS X is using the «location ID» instead
2012-10-12T22:59:37 < Thorn> btw windows has limitations on the format of usb serial numbers
2012-10-12T22:59:49 < zyp> which is 0x1d142000, so the device ends up being named /dev/tty.usbmodem1d1521
2012-10-12T23:00:57 < karlp> Thorn: don't I know it :) it got very upset trying to pass the stlink2 with the unprintable serial number via virtualbox :)
2012-10-12T23:01:43 < Thorn> time to consult my notes
2012-10-12T23:01:46 < Thorn> http://msdn.microsoft.com/en-us/windows/hardware/gg487321
2012-10-12T23:02:02 < M-Kat> does linux do that by default?
2012-10-12T23:02:08 < zyp> right, on another port of the same hub, it's called /dev/tty.usbmodem1d1531
2012-10-12T23:02:10 < M-Kat>  - provide /dev/serial
2012-10-12T23:02:59 < zyp> and on a different hub it ends up being named /dev/tty.usbmodem1d16331
2012-10-12T23:04:19 < zyp> Thorn, it's not specific to windows
2012-10-12T23:04:45 < zyp> the usb spec itself says that the serial number should be ascii-encoded hex
2012-10-12T23:07:00 < karlp> yeah, cos everyone follows specs :)
2012-10-12T23:07:05 < zyp> :)
2012-10-12T23:07:13 < karlp> M-Kat: no idea sorry.
2012-10-12T23:07:22 < karlp> this is on an old ubuntu 10.04 system though, so maybe?
2012-10-12T23:07:28 < M-Kat> yes
2012-10-12T23:07:43 < zyp> I believe that's an udev thing
2012-10-12T23:08:03 < zyp> so it depends on your default udev configuration, I guess
2012-10-12T23:08:24 < M-Kat> ah
2012-10-12T23:08:56 < Thorn> http://xkcd.com/1119/
2012-10-12T23:09:05 < M-Kat> http://www.turnkeylinux.org/forum/support/20120925/udev-does-not-create-devserial
2012-10-12T23:09:07 < M-Kat>   :-/
2012-10-12T23:09:37 < zyp> hmm
2012-10-12T23:10:02 < zyp> http://bin.jvnv.net/f/ftuFw.png <- lpc4300 have all grounds in center, surrounded by vdd
2012-10-12T23:10:35 < zyp> I don't really want to put caps on the bottom :/
2012-10-12T23:10:44 < karlp> sure you do!
2012-10-12T23:10:52 < karlp> even old pentiums and so on stuck all the caps on the bottom
2012-10-12T23:11:05 < zyp> yes, but I didn't have to solder them :p
2012-10-12T23:11:20 < Thorn> how else would you do it?
2012-10-12T23:11:45 < Thorn> I've never done bga but I understand they're made this way to put caps under them
2012-10-12T23:11:56 < zyp> well how would you do it for qfp? :p
2012-10-12T23:12:30 < zyp> if it's ok to put caps around a qfp it should be ok to put caps around a bga containing the same die :p
2012-10-12T23:13:02 < Thorn> with qfp you at least don't have to route from the center around all the balls
2012-10-12T23:14:19 < zyp> I'm tempted to just dump them directly into vdd/gnd planes and then just hook up caps to the planes around the chip
2012-10-12T23:14:26 < zyp> can't be that bad, can it? :p
2012-10-12T23:15:52 < zyp> oh, and since I'm not doing buried vias, all vias would be protruding through to the bottom, interfering with any footprints there :p
2012-10-12T23:16:16 < Thorn> well altera docs say capacitors are ineffective above 100MHz anyway (pcb planes are), and if there's no thin traces between them and che chip, maybe they can be located a little farther away
2012-10-12T23:17:57 < Thorn> something similar here http://mirror.altium.com/community/newsletters/may-08/en/intuitive-interactive-routing.html
2012-10-12T23:18:04 < zyp> either way, it'll just be a toy board to see if I'm even capable of soldering it, some noise won't kill me :p
2012-10-12T23:18:52 < Thorn> but I imagine that guy has no signals on the bottom layer
2012-10-12T23:19:22 < zyp> and a sane way to do double-sided assembly
2012-10-12T23:19:30 < zyp> I'd rather not.
2012-10-12T23:23:07 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 246 seconds]
2012-10-12T23:23:46 < upgrdman_> anyone tried to do an LQFP with toner transfer for etch resist? is that do-able?
2012-10-12T23:23:56 < karlp> anyone got any binaries for the f0 disco? or sane ready to build projects?
2012-10-12T23:24:01 < upgrdman_> they're what, 0.2mm track width?
2012-10-12T23:24:11 < karlp> I just need some different code to test that it works,
2012-10-12T23:24:25 < upgrdman_> karlp: i found some files for the f0 from szczy or whatever his name is
2012-10-12T23:24:38 < upgrdman_> on his github ..
2012-10-12T23:24:42 < karlp> yeah, he has a single template that I might be able to hack a bit
2012-10-12T23:24:51 < karlp> it has a led blinker at least.
2012-10-12T23:25:04 < upgrdman_> i've some some stuff with the f0, using slighly modified versions of his makefile
2012-10-12T23:25:40 < upgrdman_> and i know the stm stdperiph pwm example code will compile, upload and work with szczy's makefile/linker script
2012-10-12T23:25:53 < upgrdman_> (i use linux)
2012-10-12T23:26:35 < Thorn> upgrdman_: http://we.easyelectronics.ru/uploads/images/00/04/59/2011/07/10/eb2a67.jpg
2012-10-12T23:26:50 < Thorn> for example
2012-10-12T23:27:00 < zyp> awh, I wish I had F0 support in my lib now, then I could trick karlp into start using it :p
2012-10-12T23:27:02 < Thorn> (but i never tried it)
2012-10-12T23:27:16 < upgrdman_> looks good... will try with my new laser printer over the weekend
2012-10-12T23:28:06 < upgrdman_> Thorn: any idea why the ground plane is checkered in that guy's design?
2012-10-12T23:28:46 < Thorn> no idea, it's not like he's worriyng about board warping or something
2012-10-12T23:29:05 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-12T23:29:08 < upgrdman_> warping? how would that affect it?
2012-10-12T23:29:59 < karlp> huh, I thought f0 support had been added to stlink.
2012-10-12T23:30:15 < upgrdman_> i used openocd to get my f0discov working
2012-10-12T23:30:36 < upgrdman_> st-flash didnt work for the f0, but st-util lets me debug the f0 with gdb
2012-10-12T23:30:50 < karlp> once it's already flashed?
2012-10-12T23:30:51 < zyp> I think it's just flashing that's not implemented
2012-10-12T23:30:52 < M-Kat> but can you load  ?
2012-10-12T23:30:53 < karlp> taht sounds far too complicated.
2012-10-12T23:30:56 < upgrdman_> i couldn't get openocd to play friends with gdb, but i didnt try hard either
2012-10-12T23:31:00 < zyp> i.e. lack of flashloader
2012-10-12T23:31:10 < upgrdman_> i flash with openocd, debug with st-util
2012-10-12T23:31:21 < Thorn> some more http://one1ove.ru/?p=308
2012-10-12T23:31:22 < karlp> what a crazy absurd world we lie in
2012-10-12T23:31:28 < upgrdman_> and when im not too lazy i will look into getting openocd to work with gdb
2012-10-12T23:31:54 < zyp> karlp, it should be easy to add if you're up to the task
2012-10-12T23:32:00 < karlp> haha
2012-10-12T23:32:10 < karlp> I was planning on sitting down and adding support for specifying the device
2012-10-12T23:32:14 < karlp> not f0 flash loaders :)
2012-10-12T23:32:18 < zyp> :p
2012-10-12T23:32:20 < upgrdman_> so much flux...
2012-10-12T23:33:22 < karlp> I was hoping to do something small this evening, feel good after a frustrating unsuccessful week :)
2012-10-12T23:33:27 < zyp> I could do it, but I don't have any F0 hardware
2012-10-12T23:33:31 < karlp> would hav ebeen if I hadn't left the VL board at work.
2012-10-12T23:33:38 < zyp> nor any real interest since I'm not using st-link anymore :p
2012-10-12T23:33:49 < karlp> yeah, no stress, i wasn't asking you to :)
2012-10-12T23:34:13 < zyp> just flash bmp to every discovery you own and be done with it? :p
2012-10-12T23:34:44 < upgrdman_> bmp == bootloader?
2012-10-12T23:34:55 < zyp> black magic probe firmware
2012-10-12T23:35:12 < upgrdman_> whats special/helpful about it
2012-10-12T23:35:31 < zyp> built in gdbserver, no middleware needed
2012-10-12T23:35:32 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 248 seconds]
2012-10-12T23:36:05 < karlp> zyp, I've thought about it, but that's somethign I can do, but not that everyone can or should have to do.
2012-10-12T23:36:27 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-12T23:36:37 < karlp> does bmp include an f0 flash loader?
2012-10-12T23:39:46 < zyp> I thought so, but looking through the code it seems to only support F1 and F4
2012-10-12T23:39:55 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
--- Day changed Sat Oct 13 2012
2012-10-13T00:00:16 < karlp> well, there's a branch that can flash, but it's super flaky
2012-10-13T00:02:53 < karlp> goddamn I'd forgotten how ugly the flashing code was. it's like I block it out o fmy brain everytime I see it.
2012-10-13T00:05:53 < zyp> hmm, splitting the symbol was not dumb, now I can place the parts on different sheets: http://bin.jvnv.net/f/mpHWx.png
2012-10-13T00:10:16 < karlp> whee, now I have binaries I can flash
2012-10-13T00:11:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-13T00:14:49 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-10-13T00:32:43 < M-Kat> youth offending team: meeting on a street corner near you
2012-10-13T00:35:55 <+Steffann> Are you sure M-Kat ?
2012-10-13T00:36:09 < M-Kat> apparently
2012-10-13T00:40:41 < zyp> http://bin.jvnv.net/f/bKazF.png <- next to 64-pin F4 for scale
2012-10-13T00:41:20 < zyp> not really that small even considering it has four times the IO
2012-10-13T00:43:54 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-13T00:47:08 < M-Kat> haha
2012-10-13T00:47:26 < M-Kat> im not sure that going to happen
2012-10-13T00:55:40 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-13T00:58:04 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-13T01:04:18 -!- barthess [~barthess@178.154.15.156] has quit [Read error: Connection reset by peer]
2012-10-13T01:05:43 < karlp> yeah, even libusb's get_device_string_ascii does't make any sense of it when it's garbage.
2012-10-13T01:05:48 < karlp> http://pastebin.com/wYdsV3q1
2012-10-13T01:06:05 < karlp> libusb's "get ascii" just means you don't have to specify the language id :)
2012-10-13T01:07:46 < zyp> :p
2012-10-13T01:08:34 < karlp> I'm not sure this is helpful.  the serial's are bogus, and you have to run this once to see what serials there are.
2012-10-13T01:08:45 < karlp> or you ahve to know what the libusb magic internal bus/id numbers are.
2012-10-13T01:28:11 < M-Kat> arent the descriptors 16bit characters
2012-10-13T01:28:22 < M-Kat> is it unicode?
2012-10-13T01:28:37 < karlp> they're _meant_ to be ascii hex.
2012-10-13T01:29:13 < M-Kat> https://github.com/Laurenceb/STM32-Logger/blob/master/Util/USB/usb_desc.c#L105
2012-10-13T01:29:32 < zyp> M-Kat, yes, they are unicode
2012-10-13T01:29:39 < zyp> utf16le if I'm not wrong
2012-10-13T01:30:35 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-13T01:30:58 < karlp> oh, sorry, I was still thinking about the serial
2012-10-13T01:33:05 < emeb> Get your mind out of the gutter.
2012-10-13T01:34:32 < zyp> hmm, SDRAM chips are pretty cheap, maybe I should just throw on one because I can
2012-10-13T01:36:27 < emeb> who has an SDRAM interface?
2012-10-13T01:36:35 < zyp> LPC4300
2012-10-13T01:36:41 < emeb> oh yeah.
2012-10-13T01:37:06 < emeb> M4F + M0 but no on-chip flash.
2012-10-13T01:37:23 < zyp> I'm making a toyboard just to try see how hard BGA is :p
2012-10-13T01:37:39 < zyp> actually flashed variants are available now
2012-10-13T01:37:52 < emeb> an idle mind is the devil's workshop. or something.
2012-10-13T01:37:57 < zyp> http://www.digikey.com/product-detail/en/LPC4357FET256,551/568-9603-ND/3477068 <- this is in stock at digikey
2012-10-13T01:38:01 < emeb> no kidding?
2012-10-13T01:38:34 < emeb> Nice. 1MB Flash. EEPROM too.
2012-10-13T01:39:39 < zyp> hmm, I wonder if it has any footprint changes
2012-10-13T01:39:48 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-13T01:39:52 < zyp> I doubt it, but it has a seperate datasheet
2012-10-13T01:40:21 < emeb> Too bad it's only BGA.
2012-10-13T01:40:51 < zyp> the qfp variants are in the pipeline
2012-10-13T01:40:55 < zyp> apparently
2012-10-13T01:41:13 < emeb> I've got an application that could use a boatload of memory.
2012-10-13T01:41:21 < emeb> (audio sampler/looper)
2012-10-13T01:41:48 < emeb> but the customer doesn't want to deal w/ BGAs.
2012-10-13T01:42:07 < zyp> oh, found a footprint change
2012-10-13T01:42:42 < zyp> the single pin listed as NC in LPC4350 now apparently has a function
2012-10-13T01:43:11 < qyx_> hm, 1mm pitch
2012-10-13T01:43:16 < zyp> yep
2012-10-13T01:43:19 < emeb> that's not bad
2012-10-13T01:43:19 < qyx_> for lbga256
2012-10-13T01:43:27 < zyp> that's why I chose that
2012-10-13T01:43:28 < qyx_> but 256 is too much
2012-10-13T01:43:37 < emeb> might even be able to do it on OSHPark rules. :)
2012-10-13T01:43:41 < zyp> no.
2012-10-13T01:43:49 < zyp> well
2012-10-13T01:43:53 < zyp> depends
2012-10-13T01:44:14 < emeb> on...?
2012-10-13T01:44:31 < zyp> on which pins you need to route out
2012-10-13T01:45:00 < qyx_> uuu
2012-10-13T01:45:03 < zyp> oshpark via spec is not good enough that you can place vias with 1mm pitch and still route traces between them
2012-10-13T01:45:05 < qyx_> 144lqfp available at farnell
2012-10-13T01:45:30 < emeb> ah
2012-10-13T01:45:31 < qyx_> http://sk.farnell.com/nxp/lpc4310fbd144-551/mcu-32bit-arm-cortex-m4-144lqfp/dp/2094322
2012-10-13T01:46:04 < emeb> so where you going for this board then?
2012-10-13T01:46:21 < zyp> 1mm is 39 mil, oshpark vias are 27 mil, leaving only 12 mil for trace and clearances on each side :p
2012-10-13T01:46:46 < zyp> dunno
2012-10-13T01:48:01 < zyp> dongs mentioned something about being able to panelize it along with other four layer stuff, and I'm not really in a hurry to get it done, so I can wait until he does a four layer panel
2012-10-13T01:49:48 < emeb> His board house seems to have pretty tight rules.
2012-10-13T02:06:06 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-13T02:08:48 < M-Kat> http://sourceforge.net/apps/mediawiki/openecosys/index.php?title=VUE32
2012-10-13T02:08:51 < M-Kat> interesting
2012-10-13T02:08:58 < M-Kat> but PIC32 do not want
2012-10-13T02:10:52 < emeb> MIPS MEH
2012-10-13T02:16:50 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-13T02:33:31 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 246 seconds]
2012-10-13T02:35:33 -!- M-Kat [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-13T02:53:00 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-13T03:38:59 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 256 seconds]
2012-10-13T03:39:23 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-13T03:39:53 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 246 seconds]
2012-10-13T03:40:07 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-13T03:41:43 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-yurcjvhaaoafkjok] has quit [Ping timeout: 260 seconds]
2012-10-13T03:43:43 <+dekar> jpa-, Rickta59 when did I fork GCC?
2012-10-13T03:44:16 <+dekar> My parts of TNT (buildscripts, patches) are all public domain
2012-10-13T03:44:43 <+dekar> GCC is indeed GPL with linking exception and Newlib is under various BSD style licenses
2012-10-13T03:45:13 <+dekar> so unless you use C++ the linked firmware contains no GPL code at all
2012-10-13T03:46:49 -!- feurig [~don@198.202.31.180] has quit [Remote host closed the connection]
2012-10-13T03:47:14 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-yyutmyvnyryerwpn] has joined ##stm32
2012-10-13T03:47:29 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 256 seconds]
2012-10-13T03:50:46 -!- Netsplit *.net <-> *.split quits: Niedar
2012-10-13T03:51:15 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-13T03:51:15 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-13T03:51:15 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-13T03:57:10 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-yyutmyvnyryerwpn] has quit [Ping timeout: 265 seconds]
2012-10-13T03:59:11 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-cyppeviepedcjqah] has joined ##stm32
2012-10-13T04:01:31 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-13T04:26:25 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-13T04:30:16 < dongs> emeb_mac: breaking news, my leds have landed in jp
2012-10-13T04:30:55 < emeb_mac> w00t!
2012-10-13T04:31:06 < emeb_mac> They do exist.
2012-10-13T04:31:38 < dongs> so that means I should order the pcbs soon, though led drivers are still on hold while tehy're waiting to send them with other pars I need
2012-10-13T04:33:29 < emeb_mac> yep - annoying when you've got some but not all parts sitting around.
2012-10-13T04:44:19 < emeb_mac> I got shipments from DK & Mouser to assemble this over the last few days: http://specialcomp.com/beaglebone/BeagleBone_FPGA.html
2012-10-13T04:45:47 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-cyppeviepedcjqah] has quit [Ping timeout: 240 seconds]
2012-10-13T04:52:01 -!- Oldboy [~Oldboy@gateway/shell/sh3lls.net/x-bggkfzapnjazjncs] has joined ##stm32
2012-10-13T04:55:41 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-13T04:56:36 -!- Oldboy [~Oldboy@gateway/shell/sh3lls.net/x-bggkfzapnjazjncs] has quit [Ping timeout: 248 seconds]
2012-10-13T04:57:18 < cjbaird> The Beagle is handling the fpga programming as well? (how..?)
2012-10-13T05:02:06 -!- Oldboy [~Oldboy@gateway/shell/sh3lls.net/x-pqbrwefctpifwtgi] has joined ##stm32
2012-10-13T05:03:40 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 246 seconds]
2012-10-13T05:12:28 < dongs> silk around tssop package looks so ugly
2012-10-13T05:12:35 < dongs> but thats the default GEDA WAY
2012-10-13T05:15:13 < dongs> cjbaird: looks like no, 'jtag port for development and debugging'
2012-10-13T05:15:53 < dongs> if you read further down the page tehres a mode you can set to load it over SPI.
2012-10-13T05:15:57 < dongs> (from beagleboard)
2012-10-13T05:21:40 < emeb_mac> cjbaird: Yeah - there are 3 ways to configure the FPGA - via JTAG, from a SPI Flash chip or via SPI from the Beaglebone host.
2012-10-13T05:22:04 < emeb_mac> Plus, the beaglebone can reprogram the SPI flash of course.
2012-10-13T05:22:29 < emeb_mac> Default @ power up is to load from the SPI flash.
2012-10-13T05:24:37 < emeb_mac> dongs: there are a lot of different TSSOP footprints out there for gEDA. Some have the outline like that, others don
2012-10-13T05:24:42 < emeb_mac> 't.
2012-10-13T05:24:59 < emeb_mac> I'm not too picky - this one fit, so I used it.
2012-10-13T05:26:38 < dongs> wehn I think gEDA, i see this: http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a4074291-106-DSCN5649.JPG or this http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a4171184-243-DSCN5840.JPG or this http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a4306099-136-DSCN6500.JPG or this http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a4437543-121-DSCN6724.jpg
2012-10-13T05:28:07 < emeb_mac> Hmmm.
2012-10-13T05:29:09 < emeb_mac> Are those ugly boards the fault of the design tool, the fab or the way they were soldered?
2012-10-13T05:29:26 < dongs> I think the fault started with the design tool :p
2012-10-13T05:30:30 < dongs> http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a3698692-157-pcb-v3-lite.jpg
2012-10-13T05:30:53 < dongs> at least you ha ve borders around R/C etcv
2012-10-13T05:30:55 < dongs> lols
2012-10-13T05:30:59 < dongs> this shit is just hideous
2012-10-13T05:31:55 < dongs> it shouldnt be allowed to have passive patterns without silk
2012-10-13T05:32:20 < emeb_mac> It's a fairly congested board, refdes overlapping the outlines and on vias. Not pretty, but that's up to the designer.
2012-10-13T05:32:41 < emeb_mac> Unless you've got a tool that automatically pretties it up for you.
2012-10-13T05:33:06 < dongs> well i dont even bother with refdes on something this packed
2012-10-13T05:33:25 < dongs> but all those unparked R/C where you cant even guess teh direction sometime
2012-10-13T05:33:33 < dongs> i mean look at R16/R37
2012-10-13T05:33:38 < dongs> between the yes/down buttons
2012-10-13T05:33:41 < dongs> bottom part of pcb
2012-10-13T05:33:48 < dongs> which way are those? up or horizontal?
2012-10-13T05:33:50 < dongs> noone knows.
2012-10-13T05:34:07 < emeb_mac> yeah. You need an assembly drawing for that.
2012-10-13T05:34:29 < dongs> if there was proper silk outline you wouldn't
2012-10-13T05:34:39 < emeb_mac> but that all depends on where the designer got the footprints.
2012-10-13T05:35:00 < Bird|lappy> dongs: what do you think of "inside-out" silk (i.e. the silk doesn't form a courtyard around the part, but sits between the pads?)\
2012-10-13T05:35:16 < dongs> yeah, thats what it should be for *sop like stuff
2012-10-13T05:35:28 < dongs> silk outside of entire pattern looks ugly and is just dumb because it actually wastes space
2012-10-13T05:35:36 < dongs> since you can't put components too close around it
2012-10-13T05:35:58 < Bird|lappy> yeah, the whole point of silk border around pattern is to serve as a poor man's keepout
2012-10-13T05:36:14 < Bird|lappy> I meant the 'inside out' style on passives
2012-10-13T05:36:19 < dongs> oh
2012-10-13T05:36:23 < dongs> no, passives need them
2012-10-13T05:36:28 < dongs> for orientation
2012-10-13T05:36:30 < dongs> etc
2012-10-13T05:36:46 < dongs> mine are drawn so that I can stack them very closely overlapping their silk
2012-10-13T05:37:01 < dongs> and it looks like one line separating them
2012-10-13T05:37:39 < emeb_mac> that's what I do on tight stuff.
2012-10-13T05:37:53 < dongs> thorn (?) I think had some nice pics of his boards with very clever passive outlines
2012-10-13T05:37:54 < emeb_mac> for very loose definition of "tight"
2012-10-13T05:38:02 < dongs> different border style for caps/resistors/diodes
2012-10-13T05:38:19 < dongs> usually shit like altium resistors are square, caps are oval
2012-10-13T05:38:41 < emeb_mac> oval? blech.
2012-10-13T05:38:49 < dongs> look at the back of your video card :)
2012-10-13T05:39:01 < dongs> it was prob designed in altium, and has  same style silk :)
2012-10-13T05:39:20 < Bird|lappy> dongs, I mean: your orientation lines sit under the passive part
2012-10-13T05:39:39 < Bird|lappy> although diodes should have a diode symbol in the silk.  no other way to do it
2012-10-13T05:39:57 < dongs> well passives d ont have orientation.
2012-10-13T05:40:00 < dongs> except leds/diodes.
2012-10-13T05:40:08 < dongs> for polarized caps.. i have a dot
2012-10-13T05:40:09 < dongs> outside.
2012-10-13T05:40:22 < dongs> same pad as regular cap.
2012-10-13T05:40:31 < dongs> i think it would look better to have it inside the pattern.
2012-10-13T05:40:35 < dongs> too lazy to redraw though.
2012-10-13T05:40:43 < Bird|lappy> I use + and - in the silk for polarized caps, although i don't have any whatsoever on my current board
2012-10-13T05:40:59 < emeb_mac> Bird|lappy: you use kicad?
2012-10-13T05:41:02 < Bird|lappy> yeah
2012-10-13T05:41:09 < Bird|lappy> doing my own footprints tho
2012-10-13T05:41:24 < emeb_mac> I've done a couple boards with that - a lot of my own fps.
2012-10-13T05:41:38 < emeb_mac> fps are easier to do in kicad than gEDA
2012-10-13T05:41:44 < Bird|lappy> agreed.
2012-10-13T05:42:03 < emeb_mac> gEDA is pretty much a text thing. kicad has a reasonable GUI approach
2012-10-13T05:42:39 < dongs> altium has the best footprint designer.
2012-10-13T05:50:33 < Bird|lappy> yeah, if I was to blow the big bucks on an EDA suite, it'd be Altium
2012-10-13T05:50:55 < Bird|lappy> I still recall OrCAD blowing up when I tried to use the Save function in the schematic capture tool
2012-10-13T05:53:08 < emeb_mac> OrCAD both sucks and blows.
2012-10-13T05:53:26  * emeb_mac started using OrCAD on DOS in 1990.
2012-10-13T05:53:30 < emeb_mac> it was better then.
2012-10-13T05:54:13 < dongs> at least altium is in the 21st century
2012-10-13T05:54:17 < dongs> with using directx/opengl for rendering
2012-10-13T05:54:23 < emeb_mac> Buddy of mine has a full seat of Altium that he's never used. Dope.
2012-10-13T05:54:25 < dongs> orcad still fucking flickers like its running on win3.1 with GDI
2012-10-13T05:54:40 < emeb_mac> probably true.
2012-10-13T05:55:16 < emeb_mac> dude still uses OrCAD + PCAD + hand-tweaked netlists. Blech.
2012-10-13T05:55:25 < upgrdman> is there a slashdot equiv for mcu/fgpa people?
2012-10-13T05:55:47 < emeb_mac> not centrally located as far as I know.
2012-10-13T05:56:33 < emeb_mac> depends heavily on what MCU and FPGA families you're interested in.
2012-10-13T05:56:46 < dongs> i'm sure theres hot news on arduino.cc
2012-10-13T05:56:47 < dongs> daily
2012-10-13T05:56:55 < upgrdman> stm32, xilinix spartan, maybe avr.
2012-10-13T05:57:07 < dongs> http://www.kickstarter.com/projects/818225433/arduino-style-battery-shield i mean this is so awesome
2012-10-13T05:57:08 < upgrdman> dongs, choke on a dick
2012-10-13T05:57:24 < upgrdman> lol
2012-10-13T05:57:54 < Bird|lappy> so: I have this panel meter (Modutec 920051) that I pulled out of a flowmeter/totalizer
2012-10-13T05:58:04 < dongs> http://s3.amazonaws.com/ksr/assets/000/134/557/bc76b5a511b83e614fa0d3d48e048ada_large.png?1346165978
2012-10-13T05:58:08 < dongs> fuck yeah.
2012-10-13T05:58:28 < Bird|lappy> thing is, I can't tell if it's a voltmeter or an ammeter :P ohmmeter across it reads 2.2-2.3kOhms, and causes the panel meter to go to quarter-scale deflection
2012-10-13T05:58:49 < upgrdman> kohms? probably voltmeter
2012-10-13T05:59:03 < Bird|lappy> yeah, would make sense
2012-10-13T05:59:07 < upgrdman> should ammeters appear as ~0ohm
2012-10-13T05:59:20 < dongs> yes, arent tehy just basically a short
2012-10-13T05:59:28 < dongs> and measures potential across shunt or wahtever
2012-10-13T05:59:33 < Bird|lappy> well, the 77 III is not known for having a low burden voltage :P
2012-10-13T05:59:36 < upgrdman> Bird|lappy, keep in mind that voltmeters often report a much high voltage, so you can use a voltage divider to calibrate it
2012-10-13T05:59:38 < upgrdman> for example
2012-10-13T06:00:08 < Bird|lappy> upgrdman, this one is 0-100% of flow in some pipe apparently :P (or at least it was)
2012-10-13T06:00:10 < upgrdman> i have maybe 20 analog voltmeters with marking from 0 - 12v, but 1.4V puts the needle in the 12V area....
2012-10-13T06:00:25 < upgrdman> and you use a pot + resistor to calibrate it as needed
2012-10-13T06:01:40 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-13T06:01:58 < Bird|lappy> upgrdman, chinese cheapies?
2012-10-13T06:02:06 < upgrdman> no, good jap ones
2012-10-13T06:02:16 < upgrdman> its common, and for a good reason
2012-10-13T06:02:16 < Bird|lappy> ah.
2012-10-13T06:02:32 < upgrdman> it lets you calibrate them properly
2012-10-13T06:03:20 < upgrdman> i've also seen this in consumer electronics from the 80's (when analog meters were common)
2012-10-13T06:03:32 < Bird|lappy> yeah, could be
2012-10-13T06:03:34 < upgrdman> for example, almost all rc hobby radios from that error were like that
2012-10-13T06:03:41 < upgrdman> era, not error
2012-10-13T06:03:51 < Bird|lappy> I plan to turn it into the meter movement for my power supply
2012-10-13T06:04:20 < upgrdman> i have an old kikusui power supply... bought it specifically for the meters. love analog meters for psu's
2012-10-13T06:07:56 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-13T06:10:29 < dongs> http://www.kicktraq.com/projects/2135449066/quickscanit-your-portable-smart-phone-document-sys/ scambox2.0 isnt doing so well
2012-10-13T06:10:51 < upgrdman> anyone have a single layer pcb images with 0.2mm tracks that I can have a high-res copy of? i want to see if my printer can print with good enough quality, and i havn't setup any CAD stuff yet
2012-10-13T06:11:39 < dongs> whats that 8mil?
2012-10-13T06:11:54 < upgrdman> 5mil
2012-10-13T06:12:00 < dongs> 0.2mm is not 5 mil
2012-10-13T06:12:08 < dongs> 0.2mm is 7.8mil.
2012-10-13T06:12:47 < dongs> i dont think youre gonna etch 5mil traces
2012-10-13T06:12:49 < upgrdman> shit, yes. dumb fingers
2012-10-13T06:12:50 < dongs> at h ome
2012-10-13T06:12:54 < upgrdman> 8mil
2012-10-13T06:13:15 < dongs> lemme see if this dc drive i made was 8mil
2012-10-13T06:13:21 < dongs> thats the biggest trace proj i had i na while
2012-10-13T06:14:04 < dongs> 10mil and wider heh
2012-10-13T06:14:46 < dongs> but im sure you can just scale it.
2012-10-13T06:16:09 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-13T06:16:17 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-13T06:18:03 < upgrdman> sure
2012-10-13T06:19:21 < dongs> haha the image is gonna blow up your browser
2012-10-13T06:19:22 < dongs> http://i.imgur.com/J3H1e.png
2012-10-13T06:24:29 < dongs> the thinnest traces are 10mil. you can prob scale as needed.
2012-10-13T06:25:27 < upgrdman> thanks!
2012-10-13T06:26:21 < upgrdman> what size is that? 5" wide?
2012-10-13T06:26:30 < upgrdman> mmm no, 3"?
2012-10-13T06:26:36 < dongs> lets see ill get dimension
2012-10-13T06:26:50 < dongs> 104.14 x 53.98mm
2012-10-13T06:27:03 < dongs> or 4100x2125mil
2012-10-13T06:27:07 < upgrdman> thanks, i'll aim for 70mm wide print
2012-10-13T06:30:19 < upgrdman> what software do you use for pcbs?
2012-10-13T06:32:18 < dongs> http://diptrace.com/screenshots.php
2012-10-13T06:34:42 < upgrdman> i wonder, the freeware is non-profit use only... how would they know?
2012-10-13T06:35:09 < dongs> shrug. i paid for full/unlimited version, it paid for itself wiht first 4layer board I routed in it.
2012-10-13T06:53:07 < emeb_mac> man - they really rearranged the pins on the 48-pin F373.
2012-10-13T07:00:46 < dongs> yes, fuckers
2012-10-13T07:00:57 < dongs> and left 303 same...
2012-10-13T07:18:20 < dongs> http://i.imgur.com/xAowh.jpg I wonder whats going on here
2012-10-13T07:22:42 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-13T07:25:20 < zippe1> Anyone played with circuit.io?
2012-10-13T07:25:28 < zippe1> I feel a bit weird about having my designs in the cloud
2012-10-13T07:26:39 < dongs> zippe, looks targeted towards tarduino crowd
2012-10-13T07:33:29 < emeb_mac> hmmm - all I get is a French "this site available" page.
2012-10-13T07:34:01 < emeb_mac> oh - circuit_s_.io
2012-10-13T07:35:02 < emeb_mac> all the kids these days love stuff to be in the cloud.
2012-10-13T07:35:38 < dongs> http://i.imgur.com/xAoOJ.jpg LOLL
2012-10-13T07:36:10 < emeb_mac> a mite dusty there.
2012-10-13T07:36:24 < emeb_mac> oddly, the rest of the machine looks fairly clean.
2012-10-13T07:37:07 < emeb_mac> now see - if they had a liquid cooled system that wouldn't happen.
2012-10-13T07:45:36 < zippe1> My girlfriend has two parrots
2012-10-13T07:45:50 < zippe1> I regularly have to strip and rebuild her machines to get the dust out
2012-10-13T07:46:05 < zippe1> And that's with an active scrubbing filter in the room
2012-10-13T07:46:18 < emeb_mac> birds = dusty?
2012-10-13T07:46:43 < zippe1> I just grab a slab of air duster cans from work every year or two
2012-10-13T07:46:53 < zippe1> emeb_mac: yes, very; they shed a lot of little stuff
2012-10-13T07:47:19 < zippe1> emeb_mac: they are also loud and annoying when it suits them
2012-10-13T07:47:31 < emeb_mac> zippe1: what's the upside?
2012-10-13T07:47:34 < zippe1> But I figure that the first meal after the big quake here is going to be bird kebab
2012-10-13T07:47:37 < emeb_mac> (of birds)
2012-10-13T07:47:40 < zippe1> zippe1: amazing sex
2012-10-13T07:47:49 < zippe1> oh, birds?  screwed if I know
2012-10-13T07:47:56 < zippe1> I don't understand women, really
2012-10-13T07:47:58 < emeb_mac> uhh... oh. not the birds.
2012-10-13T07:48:15 -!- zippe1 is now known as zippe
2012-10-13T07:48:37 < emeb_mac> :)
2012-10-13T07:48:39 < zippe> The way I see it, they're an emotional maintenance tool for her, like hacking on my car is for me
2012-10-13T07:49:01 < emeb_mac> well, cheaper than having babies I guess.
2012-10-13T07:49:37 < emeb_mac> We just have cats, which also fill the air with debris.
2012-10-13T07:49:40 < zippe> Don't have to send them to college, for sure
2012-10-13T08:08:30 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has joined ##stm32
2012-10-13T08:09:01 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-13T08:09:04 < erik-k> If I flip a port's bit twice, is it guaranteed that the output will actually toggle (for however short a time)?
2012-10-13T08:11:42 < zippe> Yes
2012-10-13T08:12:05 < erik-k> Thanks
2012-10-13T08:12:14 < zippe> Assuming that whatever runtime you're using has marked the relevant interfaces correctly.
2012-10-13T08:12:29 < zippe> If you are using your own port accessors, then Results May Varyâ„¢
2012-10-13T08:12:58 < erik-k> ... I'm not using an OS or anything
2012-10-13T08:13:44 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-13T08:13:52 < mrnuke|laptop> flyback: whaddup?
2012-10-13T08:15:15 < akaWolf> can you recommend some simple RTOS?
2012-10-13T08:15:37 < mrnuke|laptop> akaWolf: linux-rt ?
2012-10-13T08:15:45 < akaWolf> mrnuke|laptop: no =)
2012-10-13T08:16:10 < akaWolf> flyback: maybe scmRTOS?
2012-10-13T08:16:11 < mrnuke|laptop> it might cut the stm32
2012-10-13T08:16:16 < mrnuke|laptop> in two pieces that is
2012-10-13T08:16:46 < akaWolf> flyback: or RTX?
2012-10-13T08:18:34 < akaWolf> and which one is easier to understand basic RTOS?
2012-10-13T08:18:34 < mrnuke|laptop> BTW, what's the criteria to decide wheter you need an OS, or just some plain 'ole hardware?
2012-10-13T08:19:04 < mrnuke|laptop> s/hardware/firmware/
2012-10-13T08:19:18 < akaWolf> stm32l151
2012-10-13T08:19:43 < akaWolf> monochromatic display
2012-10-13T08:19:56 < akaWolf> keyboard
2012-10-13T08:22:29 < akaWolf> I've never used OS. a device and program. need to rewrite the program under the RTOS.
2012-10-13T08:24:13 < akaWolf> and? =)
2012-10-13T08:35:08 < emeb_mac> dongs: I'm I seeing this right - F373 has no USB VBUS input?
2012-10-13T08:35:11 < akaWolf> free? =)
2012-10-13T08:41:23 < akaWolf> flyback: good documentation and examples?
2012-10-13T08:43:11 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 255 seconds]
2012-10-13T08:46:56 < akaWolf> ok
2012-10-13T08:47:12 < akaWolf> thanks =)
2012-10-13T09:00:58 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-13T09:06:54 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-13T09:11:41 < zippe> erik-k: then you had better pray your accessors are correctly volatile
2012-10-13T09:13:32 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-13T09:31:54 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 240 seconds]
2012-10-13T09:41:59 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 260 seconds]
2012-10-13T09:44:58 < erik-k> zippe: What do you mean accessors? I'm literally just writing GPIOx->ODR ^= stuff twice.
2012-10-13T09:45:24 < zippe> As long as the pointer you're dereferencing is volatile, you'll be ok
2012-10-13T09:45:48 < zippe> Otherwise, not so much
2012-10-13T09:46:15 < erik-k> So if the flip is physically written twice (i.e. not taken out by the compiler) it flips twice?
2012-10-13T09:46:24 < emeb_mac> isn't that the sort of problem that would show up pretty quickly?
2012-10-13T09:46:33 < zippe> Depends on your compile options
2012-10-13T09:46:40 < erik-k> -O0
2012-10-13T09:47:00 < zippe> At 0, even non-volatile dereferences will probably go to memory
2012-10-13T09:47:24 < zippe> But if you try -O2, you will probably discover that everything stops working, and blame the compiler
2012-10-13T09:47:28 < zippe> Which would be incorrect.
2012-10-13T09:47:31 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-13T09:47:43 < zippe> What's your definition for GPIOx ?
2012-10-13T09:47:59 < erik-k> I'd have to go look that up, it's buried somewhere in ST's headers
2012-10-13T09:48:27 < zippe> Oh, you *are* using a runtime
2012-10-13T09:48:44 < zippe> In which case you'll be OK
2012-10-13T09:48:56 < zippe> I mean, ST's libraries are junk, but they get at least that much right
2012-10-13T09:49:29 < erik-k> I thought by runtime you meant an OS
2012-10-13T09:50:37 < erik-k> TBH just about all I use of them is their convenient #DEFINEs of all the addresses and bits one pokes at to set the hardware up
2012-10-13T09:52:36 < cjbaird> lifestyle blog: back from a party for an aquaintance who finally completed her Ph.D (in Arts, alas)
2012-10-13T09:52:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-13T09:53:03 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-13T09:53:29 < erik-k> would sticking an asm("nop"); in between a pair of toggles keep the compiler from optimizing them out at higher -Os?
2012-10-13T09:53:30 < cjbaird> there was like one person there who wasn't a wanker^H^H^H^H^H^Hwriter :P
2012-10-13T09:53:54 < zippe> erik-k: no
2012-10-13T09:54:09 < cjbaird> lol: http://www.news.com.au/entertainment/music/gangnam-style-beats-out-one-direction-in-australian-music-chart/story-e6frfn09-1226490602781
2012-10-13T09:54:38 < zippe> erik-k: but if the pointer is pointer-to-volatile the compiler is required to perform each store to the target in program order
2012-10-13T09:55:50 < jpa-> a harder problem is when you want to force a register read
2012-10-13T09:55:59 < jpa-> but volatile helps there also
2012-10-13T09:57:27 < erik-k> zippe: In the headers the GPIO is ((GPIO_TypeDef *) GPIOA_BASE), and GPIOx_BASE is just an 0xADDRESS with no other keywords
2012-10-13T09:59:20 -!- dekar_ [~dekar@212.255.46.253] has joined ##stm32
2012-10-13T09:59:24 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-13T10:00:26 < zippe> erik-k: hmm, looks like at least in my copy of the F4 stdperiph lib I have that tagged volatile
2012-10-13T10:00:42 < zippe> erik-k: But it's been a while since I synced and I may have added that
2012-10-13T10:00:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-13T10:00:52 < zippe> At any rate, if it's not volatile, you're screwed
2012-10-13T10:01:27 < zippe> Hmm, no, that's the same as the vanilla stuff
2012-10-13T10:01:34 < zippe> Look at the __IO define.
2012-10-13T10:01:50 < erik-k> zippe: I'm digging in the ST peripheral lib v3.3.0... so I should just do something like "volatile GPIO_TypeDef *x = GPIOA_Base"?
2012-10-13T10:01:55 < zippe> No no
2012-10-13T10:02:01 < zippe> This is a pretty good example of how retarded ST are
2012-10-13T10:02:13 < zippe> __IO is just a #define for 'volatile'
2012-10-13T10:02:25 -!- dekar [~dekar@212.255.45.64] has quit [Ping timeout: 256 seconds]
2012-10-13T10:02:56 < zippe> And their comment is "defines 'read / write' permissions"
2012-10-13T10:03:01 < zippe> Which of course it doesn't.
2012-10-13T10:03:45 < erik-k> Yet the STM32 series of controllers seem quite impressive hardwarewise...
2012-10-13T10:04:08 < zippe> The hardware and the peripheral library are, probably fortunately, developed by entirely separate teams
2012-10-13T10:09:12 < erik-k> Oh, wait, their GPIO_TypeDef has __IO before the types
2012-10-13T10:09:18 < erik-k> So I'm safe then?
2012-10-13T10:20:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-13T10:23:07 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-13T10:26:07 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-13T10:28:26 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-13T10:29:16 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Client Quit]
2012-10-13T10:29:36 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-13T10:56:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-13T11:23:27 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-13T11:23:31 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-13T11:23:45 -!- |sark [~sark@109.235.51.202] has joined ##stm32
2012-10-13T12:04:47 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-13T12:38:09 -!- barthess [~barthess@178.154.61.60] has joined ##stm32
2012-10-13T13:28:26 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has quit [Disconnected by services]
2012-10-13T13:28:28 -!- Nutter` [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-13T13:34:42 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-13T14:01:20 -!- barthess [~barthess@178.154.61.60] has quit [Ping timeout: 252 seconds]
2012-10-13T14:43:36 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-13T14:49:45 -!- Tectu is now known as tectu
2012-10-13T15:10:06 -!- daku is now known as DaKu
2012-10-13T15:10:10 -!- tavish [~tavish@59.177.11.57] has joined ##stm32
2012-10-13T15:10:10 -!- tavish [~tavish@59.177.11.57] has quit [Changing host]
2012-10-13T15:10:11 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-13T15:11:09 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-13T15:21:23 -!- M-Kat [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has joined ##stm32
2012-10-13T15:24:50 -!- M-Kat is now known as Laurenceb_
2012-10-13T15:32:39 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Remote host closed the connection]
2012-10-13T15:39:28 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Ping timeout: 246 seconds]
2012-10-13T15:39:45 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-13T16:09:13 -!- Laurenceb_ [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-13T16:10:59 -!- Laurenceb_ [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has joined ##stm32
2012-10-13T16:17:00 < zyp> zippe, would it be sane to use systick as a deadline timer?
2012-10-13T16:17:20 < dongs> deadline?
2012-10-13T16:17:40 < zyp> instead of ticks at a fixed rate
2012-10-13T16:17:59 < dongs> constantly changing its tick rate?
2012-10-13T16:18:23 < dongs> zyp, you might as well just use any regular timer for this.
2012-10-13T16:18:34 < dongs> probably easier too.
2012-10-13T16:18:43 < jpa-> i think systick should work fine
2012-10-13T16:18:50 < dongs> i mean it'll work but whats the point?
2012-10-13T16:18:58 < zyp> dongs, yes, that's what zippe is doing
2012-10-13T16:19:09 < jpa-> well if your kernel is deadline-based or something
2012-10-13T16:19:12 < zyp> systick doesn't have compare stuff
2012-10-13T16:19:15 < dongs> yeah, and me too.. i hacked together some alarm/callback type shit
2012-10-13T16:19:29 < dongs> set it for x ticks in teh futar, call back when it hits.
2012-10-13T16:19:43 < zyp> but I'm looking at using this for my OS scheduler, that I would like to be able to use on non-stm32
2012-10-13T16:26:03 < zyp> the problem with systick is that there is no sane way of changing the time until the next tick while still accurately counting the time elapsed since last tick
2012-10-13T16:26:42  * Laurenceb_ is raging
2012-10-13T16:27:00 < Laurenceb_> is there a pdf to doc convertor that actually works?
2012-10-13T16:27:04 < jpa-> no.
2012-10-13T16:27:17 < Laurenceb_> argggg
2012-10-13T16:27:26 < Laurenceb_> i overwrote a word document
2012-10-13T16:27:34 < Laurenceb_> and only have a pdf backup
2012-10-13T16:27:58 < Laurenceb_> fail/0
2012-10-13T16:27:59 < jpa-> undelete
2012-10-13T16:28:13 < jpa-> and look for it in the temp folders
2012-10-13T16:28:36 < Laurenceb_> hmm
2012-10-13T16:28:51 < Laurenceb_> this is openoffice
2012-10-13T16:28:58 < Laurenceb_> on ubuntu
2012-10-13T16:29:21 < Laurenceb_> i think i must have copied an old version over the original file
2012-10-13T16:30:01 < jpa-> just copy-paste the text and fix the formatting
2012-10-13T16:30:09 < Laurenceb_> hehe
2012-10-13T16:30:12 < Laurenceb_>   "just"
2012-10-13T16:30:22 <+Steffann> this is openoffice: problem 1
2012-10-13T16:30:24 < Laurenceb_> its a 100 page document with a shiton of graphics
2012-10-13T16:30:27 < dongs> bwhah
2012-10-13T16:30:30 < dongs> openoffice = fucked
2012-10-13T16:30:38 < dongs> (so is ubuntu)
2012-10-13T16:30:40 < jpa-> but openoffice tends to make backups under ~/.openoffice, look for recently changed files
2012-10-13T16:30:44 < dongs> 2012 and lunix has no undelete
2012-10-13T16:30:45 < Laurenceb_> ok
2012-10-13T16:30:48 < Laurenceb_> lol
2012-10-13T16:30:49 < dongs> hilarious.
2012-10-13T16:31:03 < jpa-> (and remember that ubuntu nowadays uses libreoffice, so ~/.libreoffice)
2012-10-13T16:31:23 < dongs> how do you call a shitty fork of shit?
2012-10-13T16:31:59 < jpa-> Laurenceb_: 100 page documents and you make no backups? wtf
2012-10-13T16:32:10 < Laurenceb_> i did
2012-10-13T16:32:22 < Laurenceb_> but somehow they got overwritten with a crappy old version
2012-10-13T16:32:24 < dongs> what was wrong with openoffice that needed a fork?
2012-10-13T16:32:27 < dongs> wasn't  free enough?
2012-10-13T16:32:36 <+Steffann> SUN ..
2012-10-13T16:32:37 < jpa-> dongs: oracle
2012-10-13T16:32:41 <+Steffann> => oracle
2012-10-13T16:32:47 <+Steffann> :)
2012-10-13T16:32:57 < jpa-> Laurenceb_: huh, all of them?
2012-10-13T16:33:00 < dongs> this just in... Laurenceb_ fails school because his  thesis was being written in opensores.
2012-10-13T16:33:17 < jpa-> no, Laurenceb_ fails because Laurenceb_ fails
2012-10-13T16:33:17 < Laurenceb_> lol
2012-10-13T16:33:22 < Laurenceb_> hehe
2012-10-13T16:33:29 < jpa-> it's a self-contained failing solution, it doesn't need platform support
2012-10-13T16:33:57 < BrainDamage> jokes aside, can't you set automatic incremental backups?
2012-10-13T16:34:12 < BrainDamage> even if software doesn't fail, hw still can
2012-10-13T16:34:16 < BrainDamage> or accidents
2012-10-13T16:35:08 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-13T16:36:38 < Laurenceb_> theres no ~./openoffice
2012-10-13T16:36:44 < Laurenceb_> hmm ok
2012-10-13T16:36:55 < Laurenceb_> usually i use latex with version control
2012-10-13T16:37:00 < Laurenceb_> but this is for CE crap
2012-10-13T16:37:54 < Laurenceb_> could any temporary files still be hiding somewhere?
2012-10-13T16:39:58 < jpa-> not outside /tmp and your homedir usually
2012-10-13T16:40:59 < jpa-> find ~ /tmp -ctime 0 | less
2012-10-13T16:41:15 <+dekar_> I have nightly incremental backups as well, I can recommend http://en.wikipedia.org/wiki/BackupPC
2012-10-13T16:41:57 <+dekar_> Laurenceb_, try photorec
2012-10-13T16:42:16 < Laurenceb_> hmm
2012-10-13T16:42:28 < jpa-> yeah, or the ext3 undelete programs etc.
2012-10-13T16:42:32 <+dekar_> since you probably saved the file quite often there should be tons of copies all over the disk
2012-10-13T16:42:46 < Laurenceb_> yeah ill have hundereds of saves
2012-10-13T16:43:05 <+dekar_> first thing to do, plug the power from that PC
2012-10-13T16:43:08 < Laurenceb_> i had a script to distribute it over a load of pcs
2012-10-13T16:43:19 <+dekar_> prevent all further writes to the disk
2012-10-13T16:43:22 < Laurenceb_> which went wrong with one of the files and overwrote it with an old version
2012-10-13T16:43:28 < Laurenceb_> including all the backups
2012-10-13T16:43:30 < Laurenceb_> fail
2012-10-13T16:43:31 < jpa-> :DD
2012-10-13T16:43:47 < jpa-> only Laurenceb_ can fail a backup solution so that it ruins all copies :)
2012-10-13T16:44:03 <+Steffann> Laurenceb_ you really should blog about your adventures ..
2012-10-13T16:44:09 <+Steffann> on a real blog
2012-10-13T16:44:53 <+dekar_> Steffann, what was that comment on my license yesterday?
2012-10-13T16:44:59 < dongs> i should add laurenceb section on lordpil.com
2012-10-13T16:45:28 <+dekar_> my TNT buildscript as well as the patches are public domain (aka do whatever you want with it, no copyleft needed)
2012-10-13T16:45:29 <+Steffann> That indian(??) guy had a question about it
2012-10-13T16:45:51 <+dekar_> GCC is GPL and Newlib under various BSD style licenses
2012-10-13T16:46:12 <+Steffann> Oh, not even close to india :)
2012-10-13T16:46:20 <+Steffann> His nick is similar to one who is
2012-10-13T16:46:37 <+Steffann> Rickta59 ^^
2012-10-13T16:47:06 <+dekar_> Steffann, I posted precompiled TNT versions btw: http://dekar.wc3edit.net/
2012-10-13T16:50:37 < Rickta59> thanks dekar_ I had come across a linaro copyright page and then checked the files in gcc and came to the same conclusion
2012-10-13T16:51:25 < Rickta59> linaro gcc vs fsf .. i had searched on that in google
2012-10-13T16:52:12 < Rickta59> https://wiki.linaro.org/Copyright ... that was the first link
2012-10-13T16:52:27 < TitanMKD> dekar_ I want TNT for Windows to test it ;) (i'm lazy to build all with Cygwin)
2012-10-13T16:52:56 < jpa-> TNT for windows could be nice, now i have to use codesourcery at work
2012-10-13T16:53:13 < TitanMKD> anyway i really like Linaro Bare Metal GCC
2012-10-13T16:53:36 < TitanMKD> the install is really easy all in one and just need to add path to bin
2012-10-13T16:54:00 < zyp> do they have 4.7 yet?
2012-10-13T16:54:13 < TitanMKD> no Linaro Bare Metal is 4.6.x
2012-10-13T16:54:33 < TitanMKD> what are the imrpovement of 4.7 for ARM Mx ?
2012-10-13T16:54:44 < zyp> better C++11 support
2012-10-13T16:54:55 < zyp> I'm using C++11
2012-10-13T16:55:14 < dongs> i guess zyp is the only user of that shit on embedded
2012-10-13T16:55:15 < TitanMKD> C++11 for embedded is not really what I want ;)
2012-10-13T16:55:29 <+Steffann> Yes, you do
2012-10-13T16:55:32 < zyp> I don't really care what you want, I do.
2012-10-13T16:55:44 <+Steffann> Look at his fancy usb code TitanMKD
2012-10-13T16:55:47 < TitanMKD> C++ for embedded stuff is a bad practice ;)
2012-10-13T16:55:48 < jpa-> C++11 for embedded is nice
2012-10-13T16:55:53 < zyp> TitanMKD, bullshit
2012-10-13T16:55:57 < TitanMKD> except for GUI stuff maybe ;)
2012-10-13T16:56:02 < jpa-> TitanMKD: not true
2012-10-13T16:56:06 < zyp> you have no idea what you are talking about
2012-10-13T16:56:29 < TitanMKD> I speak about low level apps of course
2012-10-13T16:56:37 < zyp> C++11 makes my code smaller and more efficient, so it's very suited for embedded.
2012-10-13T16:56:44 < TitanMKD> where class and C++ are not really a must
2012-10-13T16:57:06 < jpa-> but of course you need to code it differently than you would do C++ for desktop computers.. but same stands for C.. suddenly it is not ok to allocate big buffers on stack etc.
2012-10-13T16:57:15 < zyp> a lot of the improvements in C++11 are being able to do more at compile time
2012-10-13T16:57:29 < zyp> which is exactly why I'm using it
2012-10-13T16:58:11 < TitanMKD> C++ template are very powerful too
2012-10-13T16:58:23 < zyp> yes, that's part of it
2012-10-13T16:58:36 < TitanMKD> but it's like macro (Define) you need to check the generated code at end to be sure it's the things you want ;)
2012-10-13T16:59:08 < jpa-> not really
2012-10-13T16:59:28 < jpa-> but it is more powerful than macros
2012-10-13T16:59:42 < zyp> macros are horrible compared to templates :p
2012-10-13T16:59:46 < TitanMKD> of course macro are very simple with no type checking ...
2012-10-13T17:00:13 < TitanMKD> I only like strong type checking features in fact ;)
2012-10-13T17:00:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-13T17:00:31 < TitanMKD> also for enum where checks are done only in C++ IIRC
2012-10-13T17:01:34 < TitanMKD> but the things i would love is a mix between -Os including inline of small function ;)
2012-10-13T17:02:12 < TitanMKD> because -O2/-O3 inline is good but code is ugly and huge on GCC 4.6.x for ARM
2012-10-13T17:02:32 < zyp> -Os -finline-small-functions
2012-10-13T17:02:35 < zyp> problem solved.
2012-10-13T17:02:43 < TitanMKD> are you sure it really works ;)
2012-10-13T17:02:56 <+Steffann> ;) ;) ;) ;)
2012-10-13T17:02:59 < TitanMKD> i will check because last time i tested inline stuff it was not working
2012-10-13T17:04:31 < zyp> well, a «small» function is a function that's so small that it's smaller to inline than call :p
2012-10-13T17:05:05 < TitanMKD> yes containing just 1 instruction ;)
2012-10-13T17:05:26 < zyp> not necessarily
2012-10-13T17:06:02 < zyp> a function call may be larger than just the call instruction itself if you need to prepare the arguments for the functions
2012-10-13T17:06:31 < zyp> or store caller-save-registers
2012-10-13T17:06:38 <+dekar_> jpa-, you can never allocate big buffers on stack, most operating systems have ridiculously small stack sizes
2012-10-13T17:06:44 < TitanMKD> of course especially on arm where only r0 to r3 are not saved
2012-10-13T17:06:58 <+dekar_> I think it was just 1 or 2mb on windows
2012-10-13T17:07:26 < TitanMKD> on power pc it was a bit better with 2 times more registers ;)
2012-10-13T17:07:36 < jpa-> dekar_: google says 8 MB
2012-10-13T17:07:38 < jpa-> not so small
2012-10-13T17:07:40 < TitanMKD> but it is a compromise between x86 & ppc
2012-10-13T17:09:17 <+dekar_> jpa-, "The default stack reservation size used by the linker is 1 MB."
2012-10-13T17:09:21 <+dekar_> http://msdn.microsoft.com/en-us/library/windows/desktop/ms686774(v=vs.85).aspx
2012-10-13T17:10:06 < jpa-> ah
2012-10-13T17:10:31 < jpa-> still enough to allocate any string you might want to store :P
2012-10-13T17:10:48 < jpa-> a bit different than working with 1kB stacks
2012-10-13T17:11:25 < zyp> on a system with MMU, stacks can be dynamically grown
2012-10-13T17:12:06 <+dekar_> zyp, on windows it doesn't grow
2012-10-13T17:12:20 < zyp> I said can, not will
2012-10-13T17:12:21 < zyp> :p
2012-10-13T17:12:35 <+dekar_> zyp, stack can always grow, on my stm32 it can grow to 64kb
2012-10-13T17:12:54 <+dekar_> I actually think heap use is the bigger problem on embedded systems
2012-10-13T17:12:57 < jpa-> actually it does grow, but 1MB is the maximum size (the allocated amount of virtual memory)
2012-10-13T17:12:57 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-13T17:13:15 <+dekar_> without mmu heap corruption gets a problem at some point, you don't have that with stack
2012-10-13T17:13:30 < jpa-> corruption? you mean fragmentation?
2012-10-13T17:13:37 <+dekar_> oh yeah my bad
2012-10-13T17:13:40 <+dekar_> fragmentation
2012-10-13T17:13:53 <+dekar_> writing while eating ;)
2012-10-13T17:15:00 -!- barthess [~barthess@178.154.21.183] has joined ##stm32
2012-10-13T17:15:02 <+dekar_> I also don't think C++ is a problem on embedded, it's just that the standard libraries are wasteful
2012-10-13T17:15:44 < zyp> noone is forcing you to use them
2012-10-13T17:15:47 < zyp> I don't
2012-10-13T17:15:59 <+dekar_> yeah I wouldn't either, thus I don't see a problem
2012-10-13T17:16:06 < jpa-> the most popular standard libraries for C are also very wasteful
2012-10-13T17:16:06 <+dekar_> isn't that arduino stuff using C++ as well?
2012-10-13T17:16:22 <+dekar_> printf :)
2012-10-13T17:16:26 <+dekar_> I use that though
2012-10-13T17:16:30 < zyp> C++ with some dumb preprocessor if I'm not mistaken
2012-10-13T17:17:03 <+dekar_> if they use C++ on arduino then it can't be a problem on stm32
2012-10-13T17:17:46 < zyp> of course it's not a problem, C++ itself is essentially zero-overhed over C
2012-10-13T17:18:09 < prog> not if you use rtti
2012-10-13T17:18:50 < zyp> then don't.
2012-10-13T17:19:33 < zyp> C doesn't have rtti, thus writing equivalent code in C++ shouldn't need rtti
2012-10-13T17:20:58 < zyp> using certain features of C++ may add overhead, but then you get features that is not present in C, so unless you count the overhead of implementing something equivalent in C it's not really a fair comparison
2012-10-13T17:25:23 <+dekar_> why do all linkerscripts use align(4) instead of align(2)? I don't see how thumb would need that and my linkerscript for the TNT_Example works great with align(2)
2012-10-13T17:28:17 < jpa-> stack needs 4-byte 
2012-10-13T17:28:20 < jpa-> align atleast
2012-10-13T17:28:25 < jpa-> but code shouldn't care
2012-10-13T17:28:35 < dongs> so much chat
2012-10-13T17:32:46 < jpa-> hmm.. i want a usb class driver for filesystem-level access.. to manage concurrency between host and device
2012-10-13T17:35:17 < zyp> mtp?
2012-10-13T17:35:53 < jpa-> yeah, but standard :P
2012-10-13T17:35:56 < zyp> newer android devices use mtp instead of msc for that reason
2012-10-13T17:36:16 < zyp> so it's the most standard class there is
2012-10-13T17:36:20 < jpa-> ok
2012-10-13T17:37:10 < dongs> mtp is huge fail
2012-10-13T17:37:30 < zyp> well, yeah
2012-10-13T17:37:38 < jpa-> how does it fail?
2012-10-13T17:37:49 < dongs> its a usermode driver on windows.
2012-10-13T17:37:59 < dongs> and all the folder/etc shit is faked
2012-10-13T17:38:04 < dongs> you cant access it as normal files.
2012-10-13T17:38:15 < jpa-> so the windows driver sucks?
2012-10-13T17:38:20 < zyp> I've never tried mtp myself, but it doesn't seem very good
2012-10-13T17:38:22 < dongs> the only driver that matters, yes
2012-10-13T17:40:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-13T17:41:06 < karlp> huh, my new work linux laptop doesn't have any problems with the VL disco board, shows up just fine.
2012-10-13T17:44:15 < Laurenceb_> wtf
2012-10-13T17:44:21 < Laurenceb_> never use doc
2012-10-13T17:44:33 < Laurenceb_> i found a backup of my file, but its corrupted
2012-10-13T17:44:43 < Laurenceb_> due to idiotic macros and crap
2012-10-13T17:44:57 < Laurenceb_> s/corrupted/got screwy fond weirdness
2012-10-13T17:51:24 < dongs> lunix laptop, haha.
2012-10-13T17:51:28 < dongs> still laughing.
2012-10-13T17:54:52 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32
2012-10-13T17:55:02 < qyx_> dongs: don't
2012-10-13T17:55:25 < scrts> hello ;)
2012-10-13T17:55:39 < jpa-> hello
2012-10-13T18:04:18 <+Steffann> Hi ;)
2012-10-13T18:05:01 < emeb_mac> Ho
2012-10-13T18:07:34 < scrts> anyone using atollic for stm32? :)
2012-10-13T18:08:08 < Laurenceb_> http://www.omgsoysauce.com/wp-content/uploads/2010/08/TamponCosplay.jpg
2012-10-13T18:11:53 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-13T18:11:58 < cjbaird> oldasinternets
2012-10-13T18:19:48 -!- barthess [~barthess@178.154.21.183] has quit [Quit: Leaving.]
2012-10-13T18:24:54 < Laurenceb_> http://www.afro-squad.com/blog/wp-content/plugins/rss-poster/cache/0ebdf_a98181_t12.jpg
2012-10-13T18:25:44 < Laurenceb_> http://www.afro-squad.com/blog/wp-content/plugins/rss-poster/cache/18405_a98181_T15.jpg
2012-10-13T18:25:47 < Laurenceb_> hehe iPoo
2012-10-13T18:27:07 <+Steffann> That first one is a linux one, for sure
2012-10-13T18:27:46 -!- barthess [~barthess@178.154.7.181] has joined ##stm32
2012-10-13T18:28:31 -!- tectu is now known as Tectu
2012-10-13T18:33:56 -!- Tectu is now known as tectu
2012-10-13T18:50:37 < emeb_mac> neither looks particularly comfortable.
2012-10-13T18:51:12 < Laurenceb_> windows toilet would be a half trashed sticky public toilet
2012-10-13T18:54:18 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-13T18:58:00 < cjbaird> Japanese squat toilet
2012-10-13T19:06:09 -!- tectu is now known as Tectu
2012-10-13T19:32:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-13T19:36:56 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-13T19:39:10 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 265 seconds]
2012-10-13T19:40:42 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-13T19:52:47 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-13T20:10:42 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-13T20:18:02 < emeb> wtf - DK packed smd tantalums in moisture-proof sealed bag w/ sensor card.
2012-10-13T20:18:20 -!- tavish [~tavish@120.59.128.67] has joined ##stm32
2012-10-13T20:18:20 -!- tavish [~tavish@120.59.128.67] has quit [Changing host]
2012-10-13T20:18:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-13T20:18:46 < jpa-> emeb: what makes that a wtf?
2012-10-13T20:19:10 < jpa-> they explode just like any other component package if they get moist and go into reflow oven
2012-10-13T20:19:18 < emeb> since when are tantalums moisture sensitive?
2012-10-13T20:19:52 < emeb> just haven't seen this before.
2012-10-13T20:20:20 < karlp> are there any tricks to setting the timer prescaler? I had it working fine on tim6 for ms,
2012-10-13T20:20:31 < karlp> but setting the timer prescaler for tim7 to anything at all is just ignored
2012-10-13T20:20:44 < karlp> it's always clocking at 24mhz (which is sysclk on this f100)
2012-10-13T20:21:02 < jpa-> emeb: hmm, google search suggests that it actually depends on material
2012-10-13T20:21:19 < karlp> I ws getting strange numbers using it to coutn times on an exti interrupt, and when I divided tehm all by 24 I got exactly the numbers I expected.
2012-10-13T20:21:27 < jpa-> karlp: do you have register preloading enabled on the timer?
2012-10-13T20:21:36 < karlp> not explicitly
2012-10-13T20:21:40 < jpa-> ok, so no
2012-10-13T20:21:45 < karlp> it should still take effect on the next update anyway
2012-10-13T20:21:49 -!- barthess [~barthess@178.154.7.181] has quit [Remote host closed the connection]
2012-10-13T20:21:50 < jpa-> and anyway it should roll over and update qui.. yeah
2012-10-13T20:21:50 < karlp> and the same code works on tim6
2012-10-13T20:22:01 < karlp> it sets up ms coutns nicely there
2012-10-13T20:22:06 -!- barthess [~barthess@178.154.7.181] has joined ##stm32
2012-10-13T20:22:06 < emeb> jpa-: these are just normal yellow AVX 1206 SMDs.
2012-10-13T20:22:18 < jpa-> karlp: check the registers using a debugger?
2012-10-13T20:22:31 < karlp> will try that next I guess.
2012-10-13T20:23:35 < karlp> yeah, gdb says it's got the right number in it.
2012-10-13T20:23:39 < karlp> ie, not zero.
2012-10-13T20:27:52 < jpa-> x /16xw 0x40001400  for me :)
2012-10-13T20:36:20 < karlp> I guess yeah, it has the right things in it.
2012-10-13T20:37:08 < karlp> I'll try it later with a different timer, see what the hell's going on.
2012-10-13T20:47:17 < scrts> are there any tools with eclipse as IDE for STM32, except atollic truestudio?
2012-10-13T20:47:50 < karlp> you mean, are there other packaged and rebranded tools on the clipse platform?
2012-10-13T20:49:28 <+Steffann> Isn't that CooCox IDE eclipse based?
2012-10-13T20:50:41 < Thorn> you can use eclipse itself with right plugins if you can configure it all
2012-10-13T20:53:11 < scrts> hmm, well I'd like to use eclipse ide plus I want debugger to be fully working too
2012-10-13T20:53:18 < scrts> CooCox... let's google that
2012-10-13T20:53:55 < scrts> I actually liked IAR, but it doesn't work with ST-Link afaik
2012-10-13T20:54:38 < Thorn> latest openocd supports st-link
2012-10-13T20:55:00 < scrts> and what about compiler then?
2012-10-13T20:55:56 < jpa-> gcc
2012-10-13T20:56:03 <+Steffann> Oh, yes for that CooCox thing you have to install your own toolchain
2012-10-13T20:56:28 < Thorn> see toolchains here http://stm32.izua.ro/wiki/Main_Page
2012-10-13T20:57:55 <+Steffann> You really have to register to get that sourcery one?!
2012-10-13T20:58:14 < Thorn> afair not, maybe they changed it
2012-10-13T20:59:36 <+Steffann> So just go with yagarto :P
2012-10-13T21:00:35 < jpa-> Steffann: i just put a direct link there :P
2012-10-13T21:01:44 <+Steffann> Yagarto is a little more up to date :P
2012-10-13T21:04:28 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-13T21:04:45 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-13T21:11:50 < emeb> one of my clients wants to use CooCox. Haven't quite figured it out yet.
2012-10-13T21:12:20 < scrts> hmm, google dropped out raisonance ride7
2012-10-13T21:12:30 < scrts> let's check what's that
2012-10-13T21:17:24 < Bird|lappy> board layout question: what do I need to do to my footprints for a NPTH (vs PTH) board?
2012-10-13T21:17:52 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-10-13T21:18:02 < karlp> use more vias
2012-10-13T21:18:02 < jpa-> Bird|lappy: for smd, nothing :)
2012-10-13T21:18:08 < karlp> no, he means non-plated
2012-10-13T21:18:18 < jpa-> for through-hole, only run traces on the opposite side of the component
2012-10-13T21:18:31 < karlp> use vias on the tracks instead of assuming that you can make connections via the component leads
2012-10-13T21:18:36 < karlp> it's often hard to solder under the compnent
2012-10-13T21:18:54 < Bird|lappy> karlp, all my tracks are on one side anyway :)
2012-10-13T21:19:15 < jpa-> so only one layer?
2012-10-13T21:19:23 < Bird|lappy> yeah
2012-10-13T21:19:33 < Bird|lappy> single sided mixed-technology
2012-10-13T21:19:41 < jpa-> so nothing special needed
2012-10-13T21:20:23 < Bird|lappy> I heard somewhere that NPTH through-hole footprints need bigger pads?
2012-10-13T21:20:46 < karlp> you can get away with smaller drills for npth
2012-10-13T21:20:50 < karlp> but that's all I've heard
2012-10-13T21:21:13 < jpa-> plated-through pads are a bit stronger mechanically
2012-10-13T21:21:18 < jpa-> but that only matters for connectors etc.
2012-10-13T21:21:21 < karlp> mostly the problem is the other way, people moving from npth to pth finding that their drills weren't big enough
2012-10-13T21:21:52 < karlp> jpa-: are you meaning if the solder in the plating causes it to  be better bonded to the insides of the via?
2012-10-13T21:23:20 < jpa-> i mean that when the hole is plated through, the risk of the pad lifting off the PCB is smaller
2012-10-13T21:24:13 < karlp> oh, right.
2012-10-13T21:24:57 < Bird|lappy> also: I have a 1"x1" or so 'slab' of copper that's directly connected (no thermal relief) to a SMD land (it's the tab-land on a SOT223, and the 'slab' is intended for heatsinking
2012-10-13T21:25:17 < Bird|lappy> is that going to pose any unusual issues? (I'm going to be hand-soldering this thing together btw)
2012-10-13T21:25:59 < jpa-> depends on your soldering iron.. with any decent >40W iron it should be fine
2012-10-13T21:26:08 < Bird|lappy> *nods*
2012-10-13T21:35:41 < zippe> ha
2012-10-13T21:35:47 < Thorn> looking at winstar datasheets. character displays have Vil = 0.7Vdd, raster (wg12232e) have Vil = 2V. I wonder if that actually means the latter work with 3.3V signal levels
2012-10-13T21:35:53 < zippe> I just tried the circiuts.io component designer
2012-10-13T21:36:05 < zippe> Wow.  That's some terribad stuff
2012-10-13T21:36:21 < emeb> zippe: but it's in the cloud! It must be good.
2012-10-13T21:36:36 < zyp> zippe, did you see my question about systick and deadline timer?
2012-10-13T21:37:17 < jpa-> Thorn: Vih you mean?
2012-10-13T21:37:40 < jpa-> Thorn: you could also do the open collector output + pull-up to 5V trick
2012-10-13T21:37:41 < Thorn> yes Vih sorry.
2012-10-13T21:38:32 < zippe> zyp: sorry, I didn't
2012-10-13T21:38:47 < zippe> just scrolled back
2012-10-13T21:38:57 < zippe> zyp: and yes, systick as a deadline timer makes perfect sense
2012-10-13T21:39:32 < zippe> emeb: you can either 1) specify a list of terminal names, and it draws a rectangle with the terminals arranged anti-clockwise around the rectangle (sides only)
2012-10-13T21:40:00 < zippe> emeb: or 2) specify the shape as SVG and then pixel co-ordinates for each of the terminals (which get no text)
2012-10-13T21:40:18 < emeb> so, better put your text in the SVG
2012-10-13T21:40:20 < zyp> zippe, but how do I update it in an accurate way?
2012-10-13T21:40:24 < emeb> sounds hideous.
2012-10-13T21:40:39 < zippe> zyp: when you say "update" it
2012-10-13T21:40:43 < Thorn> jpa-: data lines are bidirectional so you can read graphics memory, no idea if I'll need it, reading controller datasheet
2012-10-13T21:40:51 < zippe> zyp: you mean "how do you keep track of time"?
2012-10-13T21:41:29 < zyp> zippe, sort of
2012-10-13T21:42:16 < jpa-> Thorn: doesn't matter if you use 5V tolerant IO pins
2012-10-13T21:43:30 < zyp> zippe, I've looked through your code and what you are doing makes sense, just moving the compare value doesn't interfere with the counter for timekeeping
2012-10-13T21:43:44 < zippe> zyp: right; I'm cheating there
2012-10-13T21:43:46 < zyp> but systick is downcounting without any compare possibility so that's not possible
2012-10-13T21:43:56 < zippe> zyp: what I usually do is turn on the cycle counter and use that to track time
2012-10-13T21:44:13 < zyp> ah
2012-10-13T21:45:01 < jpa-> cycle counter?
2012-10-13T21:45:07 < zyp> DWT_CYCCNT
2012-10-13T21:45:19 < zippe> Like using the TSC on x86
2012-10-13T21:46:14 < zippe> zyp: you can also play games with wrap and the previous deadline, but that requires you to time the reload of the systick CVR and to do math every time you want the current time.
2012-10-13T21:46:36 < zippe> zyp: so the cycle counter is always the first and best option
2012-10-13T21:47:21 < zippe> Compare is just a convenience
2012-10-13T21:47:54 < zippe> Since you can change the current value to get the same effect with a fixed compare at zero
2012-10-13T21:48:18 < zippe> (obviously you need compares when you have more than one compare interrupt per timer)
2012-10-13T21:51:06 < zyp> but changing the value might introduce inaccuracy
2012-10-13T21:51:11 < Thorn> jpa-: but I'll need 2 MCU pins per line then (one for output, other for input)?
2012-10-13T21:51:27 < jpa-> Thorn: no?
2012-10-13T21:51:35 < jpa-> why would you?
2012-10-13T21:52:05 < zippe> zyp: if you turn off interrupts, you can do it pretty precisely (a couple cycles of uncertainty, yes, may or may not matter depending on the application)
2012-10-13T21:52:23 < Thorn> ah wait, I for some reason thought I'd need external transistors. right.
2012-10-13T21:52:41 < zyp> it's the couple of cycles I'm thinking of :p
2012-10-13T22:02:44 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 248 seconds]
2012-10-13T22:11:21 < Laurenceb_> argggg
2012-10-13T22:11:25 < Laurenceb_> openoffice is so fucked
2012-10-13T22:11:29 < jpa-> no, you are
2012-10-13T22:11:39 < Laurenceb_> is there any format it handles ok?
2012-10-13T22:11:50 < Laurenceb_> so i can store archives and they wont get corrupted
2012-10-13T22:17:34 -!- |sark [~sark@109.235.51.202] has quit [Quit: Verlassend]
2012-10-13T22:41:00 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Quit: Leaving.]
2012-10-13T22:42:10 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-13T22:43:23 < karlp> well, that's odd.  replaced tim7 with tim6, same behaviour.
2012-10-13T22:43:41 < jpa-> show us your code :)
2012-10-13T22:43:41 < karlp> but before I had tim6 working with interrupts and a prescaled clock and everything.
2012-10-13T22:43:55 < karlp> yeah, was still thinking it was something I'd obviously done differentl
2012-10-13T22:44:48 < karlp> https://github.com/karlp/ocm3_playground/blob/master/exti-rht03/main.c#L124
2012-10-13T22:44:54 < karlp> prescaler is ignored.
2012-10-13T22:46:12 < karlp> I'm not sure how to simulate that for anything else,
2012-10-13T22:46:34 < karlp> but it should basically print out uSec timings between down edges every now and again.
2012-10-13T22:48:37 < karlp> when I was using tim6 as ms interrupts, I just had the rpescaler and period set differently, nothing else
2012-10-13T22:51:02 < jpa-> off-topic, but i still don't understand the point of functions like timer_set_prescaler
2012-10-13T22:51:23 < karlp> no, neither did I when I looked at it's impl
2012-10-13T22:51:30 < karlp> it can make the text a little easier to read,
2012-10-13T22:51:42 < karlp> ie, makes C look more like pseudocode and hope the compiler works it all out
2012-10-13T22:52:00 < karlp> though the timers stuff in locm3 has far far far too many methods, not all of which do anything for certain timers
2012-10-13T22:52:14 < jpa-> timer_set_repetition_counter is especially crap, as using TIMx_RCR directly would atleast give a compilation error if you use wrong timer
2012-10-13T22:52:27 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-13T22:52:33 < karlp> indeed.
2012-10-13T22:52:50 < jpa-> and compiler will optimize none of those unless LTO is used
2012-10-13T22:52:56 < karlp> anyway, I have had TIM7_PSC = blah too, and so on,
2012-10-13T22:52:59 < karlp> made no difference
2012-10-13T22:53:17 < karlp> gdb shows that the reg has what I expect
2012-10-13T22:53:20 < karlp> code size be damned.
2012-10-13T22:53:49 < karlp> the code size bloat is from using vfprintf
2012-10-13T22:53:50 < jpa-> if you set a ridicuously large prescaler (0xFFFF), does the counter still run at 24MHz when you check it with gdb?
2012-10-13T22:54:49 < karlp> well, I tried 2300, I can try further, but the printed times are always the same
2012-10-13T22:54:59 < jpa-> not the times printed by your app
2012-10-13T22:55:03 < karlp> the bits coming in are pretty stable on the scope
2012-10-13T22:55:13 < jpa-> i don't trust your app, i only trust your gdbserver
2012-10-13T22:55:40 < karlp> well, if I'm reading the TIM7_CNT, what's wrong with my app?
2012-10-13T22:55:53 < karlp> also, how would I realistically watch the timer at 24mhz with gb?
2012-10-13T22:55:58 < zyp> http://paste.jvnv.net/view/8JPwn <- can somebody try compiling this? my current gdb version fails to optimize it correctly
2012-10-13T22:56:11 < zyp> http://paste.jvnv.net/view/UflVL <- looks like this here
2012-10-13T22:56:22 < zyp> it's supposed to optimize out completely
2012-10-13T22:56:28 < karlp> even if my code is werd, changing the timer scalar should still be giving me _different_ values, even if they're "wrong"
2012-10-13T22:56:38 < jpa-> karlp: well if you set the prescaler to 0xffff, the overflow should take 200 seconds
2012-10-13T22:56:57 < jpa-> karlp: i'm just trying to bisect the problem space
2012-10-13T22:57:05 < karlp> yeah, and I get aprint when it interrupts, every time,
2012-10-13T22:57:10 < karlp> and that is always just about instant
2012-10-13T22:57:29 < karlp> run
2012-10-13T22:57:39 < jpa-> it's not like typing   set TIM7_PSC=0xFFFF   p TIM7_CNT   p TIM7_CNT   in gdb takes that much time
2012-10-13T22:59:52 < karlp> (well, with PS is 23000, and between typing in p TIm6_CNT and again, it had wrapped from 3683 back to 145
2012-10-13T22:59:57 < jpa-> zyp: http://paste.dy.fi/W2t/plain
2012-10-13T23:00:08 < karlp> swapping tim6 for tim7 clearly wasn't my problem, but it's on tim6 now :)
2012-10-13T23:00:55 < jpa-> karlp: ok
2012-10-13T23:00:59 < karlp> thanks for the ideas though, I'd forgotten I could "monitor resume" and then print from gdb while it was running.
2012-10-13T23:01:10 < zyp> jpa-, thanks
2012-10-13T23:01:19 < karlp> I'll turn off the rest of the app, and just leave the counter running until I can work out the tick rate properly
2012-10-13T23:01:31 < zyp> though, I'm a bit surprised by the results
2012-10-13T23:04:14 < karlp> zyp: here's mine: http://paste.jvnv.net/view/vuqoP
2012-10-13T23:06:06 -!- DaKu is now known as daku
2012-10-13T23:07:54 < jpa-> zyp: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=4131
2012-10-13T23:07:54 < zyp> hmm, yagarto-4.6.0 and 4.6.2 both yields the same results as 4.7.1
2012-10-13T23:08:17 < jpa-> could be related
2012-10-13T23:09:38 < zyp> possibly
2012-10-13T23:13:27 < zyp> oh
2012-10-13T23:15:31 < Laurenceb_> karlp: where do you get vfprintf from?
2012-10-13T23:15:34 < Laurenceb_> and how big is it?
2012-10-13T23:17:07 < jpa-> and do you need float support?
2012-10-13T23:18:13 < Tectu> dongs, status
2012-10-13T23:18:30 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-13T23:19:00 < jpa-> status: up
2012-10-13T23:19:36 < zyp> ok, this is silly
2012-10-13T23:19:48 < emeb> latest project -> http://imagebin.org/231854
2012-10-13T23:21:46 < zyp> http://paste.jvnv.net/view/OPSWM <- for some reason using a const pointer optimizes correctly
2012-10-13T23:22:00 < cjbaird> emeb: that question I asked before about FPGA on the Beagle was supposed to be asking about the FPGA designing-- I suppose it still needs that x86-binary-blob software (..on another machine)
2012-10-13T23:22:09 < zyp> (but only if constructor is marked constexpr)
2012-10-13T23:22:46 < emeb> cjbaird: FPGA design tools run under x86 Linux.
2012-10-13T23:23:06 < emeb> result is a "bitstream" blob that the ARM processor can load.
2012-10-13T23:23:28 < cjbaird> ..so no generating cores on-the-fly on the Beagle..
2012-10-13T23:23:44 < zyp> emeb, looks nice
2012-10-13T23:23:58 < zyp> I almost want one :p
2012-10-13T23:24:53 < jpa-> emeb: what kind of interface is there between fpga and beagle? just raw gpio port or does it support DMA?
2012-10-13T23:26:21 < zyp> emeb, did you hook up the jtag pins to the beaglebone too, so you can run a software jtag interface?
2012-10-13T23:26:35 < zyp> i.e. http://debugmo.de/2012/02/xvcd-the-xilinx-virtual-cable-daemon/
2012-10-13T23:26:58 < karlp> Laurenceb_: vfprintf just general newlib, and yes, it's enormous, but this is on the vl disco board with lots of flash, and just for debugging.
2012-10-13T23:27:11 < karlp> vfiprintf is quite a bit smaller
2012-10-13T23:27:13 < Laurenceb_> ok
2012-10-13T23:27:18 < Laurenceb_> i use rfprintf
2012-10-13T23:27:23 < Laurenceb_> prom pascal stang
2012-10-13T23:27:26 < Laurenceb_> about 2KB
2012-10-13T23:27:35 < Laurenceb_> *from
2012-10-13T23:28:13 < Laurenceb_> theres a small printf in chibios
2012-10-13T23:28:26 < Laurenceb_> but it seems to be buggy - ive had issues where it spews junk all over ram
2012-10-13T23:29:21 < Laurenceb_>  "procyon avrlib"  or whatever its called
2012-10-13T23:29:29 < Laurenceb_> some quite nice stuff in there
2012-10-13T23:29:32 < Laurenceb_>  *some*
2012-10-13T23:30:08 < karlp> yeah, I've not looked for small printfs much,
2012-10-13T23:30:09 < jpa-> maybe i'll advertise baselibc once again, it has a small printf ,)
2012-10-13T23:30:19 < karlp> most of the small ones are gpl though
2012-10-13T23:30:29 < karlp> what license is your baselibc jpa?
2012-10-13T23:31:40 < jpa-> bsd
2012-10-13T23:32:20 < karlp> hmm, I guess the last time I looked at your stuff was so long ago I couldn't get anything working and didn't want to try subbing out libcs just yet
2012-10-13T23:32:29 < jpa-> i finally bothered to check that no gpl stuff came from klibc & asked tinyprintf author to relicense
2012-10-13T23:32:47 < karlp> oh, might have been that when I first looked it wasn't bsd :)
2012-10-13T23:33:12 < jpa-> i should make a proper example of it's use
2012-10-13T23:35:41 < Laurenceb_> so how big is the baselibc printf?
2012-10-13T23:35:53 < emeb> cjbaird: no - can't currently design FPGA bitstreams outside of the Xilinx tools on x86.
2012-10-13T23:36:14 < emeb> there was a stab at that using Icarus Verilog a while back, but I think they gave up.
2012-10-13T23:36:22 < cjbaird> (..been waiting 10 years for that to change :/)
2012-10-13T23:37:19 < emeb> jpa-: interface between FPGA & Beagle is through one of the SPI ports. FPGA takes it's bitstream in "slave serial" mode using clock & data. I have a userspace app that reads the bitstream from a file and sends using linux spidev device.
2012-10-13T23:37:46 < emeb> the spi port can be DMA, so it configure takes about a second.
2012-10-13T23:37:48 < jpa-> Laurenceb_: 824 bytes
2012-10-13T23:37:59 < Laurenceb_> wow
2012-10-13T23:38:04 < emeb> zyp: JTAG pins are not hooked to the beagle. Just to the test header.
2012-10-13T23:38:06 < Laurenceb_> does it support floating point?
2012-10-13T23:38:15 < jpa-> http://paste.dy.fi/e0j/plain
2012-10-13T23:38:17 < jpa-> nope
2012-10-13T23:38:24 < Laurenceb_> oh :(
2012-10-13T23:38:39 < jpa-> it would be nice to add primitive float support though
2012-10-13T23:38:45 < jpa-> maybe also fix16_t support
2012-10-13T23:39:05 < jpa-> emeb: i mean data transfer, not configuring the FPGA
2012-10-13T23:58:35 < emeb> jpa-: on the beaglebone the FPGA is hooked up to both the SPI port and also to the GPMC - 16-bits muxed address/data.
2012-10-13T23:59:01 < emeb> Don't know how fast the GPMC can run - haven't played with it yet, but it should be pretty fast.
2012-10-13T23:59:32 < jpa-> ok,nice
--- Day changed Sun Oct 14 2012
2012-10-14T00:04:36 -!- barthess [~barthess@178.154.7.181] has quit [Quit: Leaving.]
2012-10-14T00:05:34 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds]
2012-10-14T00:06:50 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-14T00:14:59 < zyp> heh, wrapping up the register block in a container class instead of making a direct reference to an address apparently solves the problem
2012-10-14T00:15:36 < zyp> because then the reference is not towards the address but towards the containing class, and that optimize out
2012-10-14T00:16:01 < zyp> hmm, sort of
2012-10-14T00:31:03  * emeb is learning to drag solder 0.5mm pitch LQFP
2012-10-14T00:33:37 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-14T00:46:23 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-14T00:48:01 < Laurenceb_> http://www.rcgroups.com/forums/showpost.php?p=22990488&postcount=7164
2012-10-14T00:48:04 < Laurenceb_> omg dramaz
2012-10-14T00:50:26 <+Steffann> Oh, i guess that's the same peebody as the one who joins here once in a while?
2012-10-14T00:50:49 < Laurenceb_> yes
2012-10-14T00:51:12 < jpa-> dekar_: btw, about align(4) vs. align(2), const data in .text can be accessed faster when it is aligned on a word boundary
2012-10-14T00:59:09 < upgrdman> i've been using lcds with the f0 and f4, and im still learning so i have been bit-banging. i know the FSMC would be a better way to do it, but the f0 does not have an FSMC. does that mean bit-banging is the best/only way to do it with the f0?
2012-10-14T00:59:57 < jpa-> upgrdman: yeah, but bitbanging is not that bad if you allocate a full port for it
2012-10-14T01:00:05 < upgrdman> sure
2012-10-14T01:00:28 < upgrdman> i dont mind bit banging anyway :)
2012-10-14T01:00:49 < jpa-> FSMC is really nice for TFT interfacing when done properly
2012-10-14T01:00:50 < upgrdman> just got the waveshare kit with a color LCD and plan to finally learn the FSMC
2012-10-14T01:01:21 < zyp> nice, same as I have?
2012-10-14T01:01:48 < jpa-> it's worse than bitbanging when done wrong (e.g. R/S line on regular GPIO => you need to insert syncs & waits to make sure FSMC has completed operations before switching its state)
2012-10-14T01:01:57 < upgrdman> i have the open407v-d with 14 accessories
2012-10-14T01:02:34 < zyp> ah, the one for the discovery board?
2012-10-14T01:02:59 < upgrdman> the lcd w/ touchscreen, transcievers and ext flash will probably the first things i play with
2012-10-14T01:03:15 < upgrdman> yes, i wanted to go with the discov version
2012-10-14T01:04:41 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-14T01:07:25 < zyp> largest difference from the one I have seems to be lack of NOR flash and SRAM, and the port they connect to
2012-10-14T01:08:18 < zyp> some of the pins required for that are probably in use already
2012-10-14T01:09:08 < upgrdman> i really justed wanted and lcd wire wireless com's but figured i should learn some of the other stuff in the kit to broaden my horizons.
2012-10-14T01:09:24 < upgrdman> wtf... i mean "an lcd and wireless"
2012-10-14T01:10:03 < upgrdman> why the hell do my mind and key press differ so much sometimes ;(
2012-10-14T01:11:11 < zyp> :)
2012-10-14T01:11:31 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-14T01:12:41 < jpa-> upgrdman: i think it is in the errata for your brains, it has something to do with DMA and spurious transfers
2012-10-14T01:12:55 < upgrdman> ya, something like that
2012-10-14T01:13:16 < upgrdman> the weird things is that i look at the screen while i type, and don't always catch it
2012-10-14T01:13:36 < upgrdman> and when i type my grammar is often horrible
2012-10-14T01:13:52 < upgrdman> but i came with no warranty :)
2012-10-14T01:31:06 < upgrdman> zyp, ever playing with the lcd? 320x240x18bpp makes me wonder how i might try to implement a pixel buffer.
2012-10-14T01:31:10 < upgrdman> played*
2012-10-14T01:31:14 < zyp> nah
2012-10-14T01:31:57 < upgrdman> i guess i might have to use that ram accessory board :)
2012-10-14T01:32:08 < jpa-> x16bit (RGB565) is much nicer to work with
2012-10-14T01:32:26 < zyp> upgrdman, huh? which one?
2012-10-14T01:32:31 < jpa-> and just render directly to the TFT memory, don't bother with a framebuffer on the STM32
2012-10-14T01:33:18 < zyp> upgrdman, as far as I can see, your kit doesn't include a ram board
2012-10-14T01:33:20 < upgrdman> well to implement a gui i think i would need to use a pixel buffer to do it right
2012-10-14T01:33:22 < zyp> nor a port for it
2012-10-14T01:34:04 < upgrdman> "FM24CLXX FRAM Board (I2C) x 1"
2012-10-14T01:34:11 < zyp> ha
2012-10-14T01:34:23 < upgrdman> ?
2012-10-14T01:35:07 < jpa-> upgrdman: no you don't
2012-10-14T01:35:39 < zyp> i2c fram is pretty much an i2c eeprom except it's based on fram and not eeprom :p
2012-10-14T01:36:02 < jpa-> upgrdman: also, writing to I2C FRAM is slower than writing to the TFT directly
2012-10-14T01:36:06 < zyp> it's completely unsuitable as a framebuffer, as it's both too slow and small
2012-10-14T01:36:10 < upgrdman> jpa-, i havn't reading this LCD's datasheet yet, but it seems like things usually need to be 8pixel aligned, and i dont necessarily want to do that with my gui elements
2012-10-14T01:36:25 < upgrdman> o ok
2012-10-14T01:36:27 < jpa-> upgrdman: hmm, some strange controller it has then
2012-10-14T01:36:38 < jpa-> upgrdman: most important is whether you can read back the data
2012-10-14T01:36:51 < upgrdman> sure, but wouldn't that waste time?
2012-10-14T01:37:00 < jpa-> https://code.google.com/p/gwen/   is something that people on ##chibios have been trying for GUI stuff
2012-10-14T01:37:17 < jpa-> upgrdman: not really, with FSMC it is really fast and even with bitbanging it is fast enough
2012-10-14T01:37:40 < upgrdman> ok
2012-10-14T01:38:16 < jpa-> (and you only need to read data back to do translucent stuff & scrolling)
2012-10-14T01:38:46 < jpa-> only problem is that you cannot easily do double buffering
2012-10-14T01:38:46 < upgrdman> k
2012-10-14T01:39:10 < upgrdman> double buf is to prevent tearing motion, right?
2012-10-14T01:39:23 < jpa-> but if you have some animation element, just write the code in order to render it in one pass and you are fine
2012-10-14T01:39:25 < jpa-> yeah
2012-10-14T01:39:37 < zyp> no
2012-10-14T01:39:55 < jpa-> ah no
2012-10-14T01:39:58 < jpa-> not just tearing
2012-10-14T01:40:03 < jpa-> but also the flicker
2012-10-14T01:40:17 < zyp> double buf is to prevent showing the drawing
2012-10-14T01:40:20 < jpa-> if you e.g. first paint the background and then e.g. some text on the foreground
2012-10-14T01:40:24 < upgrdman> why would it flicker?
2012-10-14T01:40:29 < upgrdman> o
2012-10-14T01:40:40 < zyp> because you draw directly onto the screen and not a background buffer
2012-10-14T01:40:54 < upgrdman> the waveshare demo for the lcd has massive flicker... guess that's why :)
2012-10-14T01:40:55 < jpa-> one way to do one-pass drawing is to render a scanline at a time.. i do that here: http://essentialscrap.com/dsoquad/logic.html
2012-10-14T01:41:57 < jpa-> i basically have a 240 pixel framebuffer in memory, i render that one and let DMA transfer it while rendering the next buffer
2012-10-14T01:43:35 < zyp> my kit included a 256k external sram, so I can get away with having a full framebuffer :)
2012-10-14T01:43:45 < upgrdman> neat
2012-10-14T01:44:11 < upgrdman> is it hard to interface pc RAM with an mcu? i have lots of old laptop and 486 ram
2012-10-14T01:44:12 < zyp> that's why I didn't buy the discovery package, because it lacks that port :p
2012-10-14T01:44:28 < jpa-> upgrdman: it's DRAM, so a bit difficult
2012-10-14T01:44:54 < upgrdman> i wanted the discov version because i often want to make my designs semi-permanent and using a discov board is cheap
2012-10-14T01:45:10 < upgrdman> so i can buy a handful of boards
2012-10-14T01:45:19 < upgrdman> locally, too
2012-10-14T01:45:47 < zyp> sure, I've got discovery boards too
2012-10-14T01:46:16 < upgrdman> but i may try to do my own pcb's now that i have a laser printer. not sure how well it can do lqfp's but i'll find out
2012-10-14T01:47:46 < zyp> I don't see how that's an argument for not buying the waveshare kit with onboard mcu instead though :p
2012-10-14T01:47:55 < upgrdman> maybe i could use a uSD card for a pixel buffer </sarcasm> :)
2012-10-14T01:48:34 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-14T01:48:42 < upgrdman> zyp, if i get something working on the ws board, i can pull the discov board, wire up my shit, and then just buy a new $14 discov board locally.
2012-10-14T01:48:57 < zyp> yes?
2012-10-14T01:49:07 < upgrdman> why the ?
2012-10-14T01:49:42 < zyp> because I don't see why you can't just get it working on the ws board with onboard MCU, then grab a discovery and wire up your shit?
2012-10-14T01:50:19 < upgrdman> ok, you're right
2012-10-14T01:51:20 < upgrdman> whats the point of the power jack on the ws board? the discov powers the ws board.
2012-10-14T01:52:03 < zyp> only when you have usb connected to the stlink port
2012-10-14T02:03:47 < Laurenceb_> http://www.stericsson.com/products/cw1250-wlan.jsp
2012-10-14T02:03:51 < Laurenceb_> ^looks impressive
2012-10-14T02:03:59 < Laurenceb_> presumably what zyp does for a living
2012-10-14T02:06:41 < jpa-> http://www.bluegiga.com/wf121-wi-fi-module
2012-10-14T02:06:49 < zyp> blogtime: I just snapped a pic of my desk for your amusement: http://bin.jvnv.net/f/tvvVf.JPG
2012-10-14T02:07:33 < Laurenceb_> jpa-: not single chip tho
2012-10-14T02:08:09 < jpa-> true, but on the other hand includes antenna and protocol support.. more tarduino compatible!
2012-10-14T02:09:14 < Laurenceb_> zyp: presumably that stuff in the background is the 2m thick insulation
2012-10-14T02:10:06 < zyp> huh?
2012-10-14T02:10:20 < Laurenceb_> due to the -30C outside
2012-10-14T02:10:32 < zyp> ha
2012-10-14T02:10:33 < zyp> :p
2012-10-14T02:10:41 < Laurenceb_> also
2012-10-14T02:10:42 < Laurenceb_>  /join #pepsi
2012-10-14T02:11:08 < Laurenceb_> http://hackaday.com/2009/02/01/eye-fi-teardown/
2012-10-14T02:11:18 < Laurenceb_>  :( I was hoping it had a single ic thingy
2012-10-14T02:11:26 < zyp> hmm, I got one of those
2012-10-14T02:11:28 < zyp> they are nice
2012-10-14T02:11:39 < Laurenceb_> was looking for a simple way to add wifi to a super small pcb
2012-10-14T02:12:43 < zyp> also, I've never had anything to do with CW1250 :p
2012-10-14T02:13:57 <+dekar_> jpa-, why would align(2) be slower, are there thumb2 opcodes that only operate on 32bit aligned data?
2012-10-14T02:14:54 < upgrdman> hobby king
2012-10-14T02:14:57 < upgrdman> :)
2012-10-14T02:15:02 < Laurenceb_> hobo king
2012-10-14T02:15:14 < zyp> yeah, they sent me some spam the other day
2012-10-14T02:15:25 < Laurenceb_> they send me sms spam :(
2012-10-14T02:15:47 < Laurenceb_> looks like wifi is not easy to add
2012-10-14T02:15:54 < Laurenceb_> guess a module is required
2012-10-14T02:22:52 < Laurenceb_> http://www.dfrobot.com/index.php?route=product/product&product_id=548#.UHn2rfLyM5Y
2012-10-14T02:22:54 < Laurenceb_> looks nice
2012-10-14T02:23:12 < zyp> just get an attiny and bitbang wlan, I hear that's what all the cool kids do
2012-10-14T02:24:39 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-14T02:26:02 < Laurenceb_> teeehee
2012-10-14T02:36:13 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-14T02:36:13 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-14T02:36:13 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-14T02:39:32 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 248 seconds]
2012-10-14T02:44:47 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-14T02:45:09  * Laurenceb_ zzz
2012-10-14T02:49:43 -!- Laurenceb_ [~Laurence@host81-151-162-100.range81-151.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-14T02:50:47 < BrainDamage> zyp: what is the white pad on the right side of your desk?
2012-10-14T02:51:03 < zyp> wireless charger for the mouse
2012-10-14T02:51:24 < BrainDamage> ah
2012-10-14T03:01:10 < upgrdman> mouse charger? why? how long does the battery last?
2012-10-14T03:01:27 < upgrdman> my wireless mice last >2yrs without charging
2012-10-14T03:06:02 < dongs> sup dongs
2012-10-14T03:06:07 < dongs> Tectu: sleeping
2012-10-14T03:08:05 < zyp> upgrdman, some weeks or so
2012-10-14T03:08:28 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-14T03:08:49 < zyp> sure your mouse lasts longer, but it probably use some stripped down radio protocol and not bluetooth
2012-10-14T03:09:32 < zyp> and this mouse contains a multitouch trackpad, I guess that also consumes some power
2012-10-14T03:10:17 < dongs> zyp promoting his apple scam again
2012-10-14T03:10:27 < zyp> not really
2012-10-14T03:11:07 < zyp> I'm just saying it's not reasonable to expect two years of battery life out of this shit :p
2012-10-14T03:13:01 < zyp> and I'm not sure how long a charge lasts, I tend to put it on the charger before it's empty
2012-10-14T03:13:59 < zyp> when I used alkalines I had to change every couple of weeks, maybe two months or so
2012-10-14T03:14:33 < dongs> i have some microsoft bt mouse
2012-10-14T03:14:48 < dongs> hm no not bt i guess.
2012-10-14T03:15:06 < dongs> anyway, it lasts a while and i never turn it off so it sits powered 24/7
2012-10-14T03:15:12 < dongs> lasts a few months  on 2AA
2012-10-14T03:15:38 < zyp> bt is nice, eliminates need for having a custom dong
2012-10-14T03:15:58 < dongs> yea
2012-10-14T03:16:37 < qyx_> custom dongs
2012-10-14T03:17:47 < zyp> and wireless charging makes charging essentially effortless; nothing to plug in, you just leave it on top of the charger
2012-10-14T03:20:13 < zyp> mousepad with built in charger would probably eliminate the need for a battery at all
2012-10-14T03:20:16 < zyp> just run from a supercap
2012-10-14T03:21:23 < zyp> though, if you need to keep the mouse on the mouse pad, it could just be wired :p
2012-10-14T03:27:29 < upgrdman> why would a bt mouse be better?
2012-10-14T03:27:37 < zyp> 02:15:38 < zyp> bt is nice, eliminates need for having a custom dong
2012-10-14T03:27:49 < upgrdman> dong is small
2012-10-14T03:28:09 < upgrdman> worth the battery life :)
2012-10-14T03:30:49 < zyp> pft, I bet you're just jealous because you don't have any wireless chargers :p
2012-10-14T03:33:13 < zippe> anyone played with the electric imp stuff yet?
2012-10-14T03:34:08 < zyp> I had a brief look at it once, then decided I weren't interested
2012-10-14T03:36:16 < dongs> Our patent-pending Blinkup scheme ensures easy network configuration, and supports WEP, WPA and WPA2 encryption schemes as well as WPS setup.
2012-10-14T03:36:31 < zyp> custom high level language, cloud stuff to program it, not compatible with SD signalling even though it's in SD form factor…
2012-10-14T03:40:30 < zyp> the hardware could be fun enough, but I understood that the software stack was locked down, so I'll rather pass
2012-10-14T03:53:44 < BrainDamage> sorry for the bad offtopic, but my bash-fu is a bit weak, can anyone spot the mistake in my script? find . -iname *.pdf -print0 | for i in `xargs -0`; do echo "$i"; pdftotext "$i" - | grep "capacit"; done   it's supposed to find text inside all the pdf in a folder I tried using -print0 and the matching -0 in xargs, but it's still splitting every space in the path instead of using the null terminator
2012-10-14T03:54:49 < zyp> first of all, you didn't escape the * in *.pdf
2012-10-14T03:55:42 < BrainDamage> that actually works even without escaping
2012-10-14T03:55:45 < zyp> and I don't get what you're trying to do with xargs
2012-10-14T03:55:55 < BrainDamage> escape the spaces in the path
2012-10-14T03:56:02 < zyp> only if you don't have any pdf files in your current dir
2012-10-14T03:56:14 < BrainDamage> they are all in subdirs
2012-10-14T03:57:01 < BrainDamage> and there's spaces in the paths and filenames
2012-10-14T03:58:00 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-14T03:58:06 < zyp> try this: find . iname '*.pdf' -exec pdftotext {} - \; grep capacit
2012-10-14T03:58:20 < zyp> sorry
2012-10-14T03:58:26 < zyp> find . iname '*.pdf' -exec pdftotext {} - \; |  grep capacit
2012-10-14T03:59:05 < BrainDamage> then I won't know which file contains the strings
2012-10-14T03:59:38 < BrainDamage> I need to both echo and run pdftotext
2012-10-14T03:59:52 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-14T04:02:32 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-14T04:04:01 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Client Quit]
2012-10-14T04:04:01 < zyp> then I suggest writing a script to do both conversion and matching and invoke that with -exec
2012-10-14T04:04:19 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-14T04:06:15 < BrainDamage> probably the best
2012-10-14T04:08:19 < BrainDamage> with the script it worked, I just had to escape {}
2012-10-14T04:08:23 < BrainDamage> thanks
2012-10-14T04:09:14 < zyp> shouldn't be any need to escape {}, find will pass it as a single argument
2012-10-14T04:09:41 < BrainDamage> it actually doesn't run without escapes
2012-10-14T04:09:50 < BrainDamage> it complains exec has no arguments
2012-10-14T04:10:08 < zyp> your shell must be weird :)
2012-10-14T04:10:42 < BrainDamage> just bash
2012-10-14T04:10:47 < BrainDamage> [cloud@cloud-laptop Unknown]$ find . iname '*.pdf' -exec pdfsearch {}\;
2012-10-14T04:10:47 < BrainDamage> find: missing argument to `-exec'
2012-10-14T04:11:04 < zyp> you forgot the space after }
2012-10-14T04:11:18 < zyp> also, it should be -iname
2012-10-14T04:11:33 < BrainDamage> yeah, I fixed both
2012-10-14T04:11:42 < BrainDamage> accidentally copied a failed attempt
2012-10-14T04:19:54 < dongs> this lunix shit sounds like first world problems
2012-10-14T04:20:00 < dongs> creating solutions to nonproblems, daily.
2012-10-14T04:22:12 < zyp> :p
2012-10-14T04:23:48 < zyp> so you're basically saying that there are so few problems with linux that people have to invent their own to have something to solve?
2012-10-14T04:24:27 < dongs> no. lunix way implies creating a solution to something normal people find perfectly functional.
2012-10-14T04:24:42 < dongs> i.e. lets copy the shittiest parts of macos and iwndows into our own destkop environment
2012-10-14T04:25:02 < dongs> or like this PDF search shit.
2012-10-14T04:25:12 < dongs> you know windows indexing service probably searches inside PDFs right?
2012-10-14T04:25:33 < zyp> well, yeah
2012-10-14T04:25:36 < zyp> so does OS X'
2012-10-14T04:25:49 < BrainDamage> before it indexes 4GB pdf in a flash drive, it'll take me less to do it manually
2012-10-14T04:25:56 < BrainDamage> and I know it has auto indexing
2012-10-14T04:26:36 < BrainDamage> anyway, it seems to be working decently
2012-10-14T04:26:41 < zyp> BrainDamage, does it really take you less?
2012-10-14T04:27:13 < zyp> aren't you just doing essentially the same?
2012-10-14T04:27:30 < dongs> he does it in a hacker/fucktarded way
2012-10-14T04:27:34 < dongs> so of course its faster
2012-10-14T04:27:36 < BrainDamage> I'd have to copy it first on hd to get indexing services to run on it
2012-10-14T04:27:39 < dongs> but he had to type a mile-long command line first
2012-10-14T04:28:02 < BrainDamage> I just want the file names and uplug, that's it
2012-10-14T04:28:08 < BrainDamage> it's done now
2012-10-14T04:28:33 < BrainDamage> so 2 mins to run the command, now 3 hours trolling, go!
2012-10-14T06:14:51 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-14T06:14:59 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-14T06:43:02 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-14T06:44:08 -!- NonaSuomy is now known as TeknoJuce
2012-10-14T06:44:19 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-14T06:46:34 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-10-14T06:46:51 < dongs> http://i.imgur.com/ww930.jpg lol'd
2012-10-14T06:49:20 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-14T06:49:51 < cjbaird> What happens on Reddit, stays on Reddit.
2012-10-14T06:56:02 < dongs> sorry, i dont read faggit
2012-10-14T07:02:38 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-14T07:26:35 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-14T07:30:06 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-14T07:31:07 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Client Quit]
2012-10-14T07:31:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-14T08:04:49 -!- Oldboy [~Oldboy@gateway/shell/sh3lls.net/x-pqbrwefctpifwtgi] has quit [Ping timeout: 244 seconds]
2012-10-14T08:05:43 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xhojwlvjozmxwest] has joined ##stm32
2012-10-14T08:23:23 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-14T08:32:38 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 272 seconds]
2012-10-14T08:47:44 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-14T08:49:26 < cjbaird> My F4 mandelbrot is now drawing after than xterm can scroll with the current libopencm3 hardfloat work..
2012-10-14T08:53:52 < dongs> hard at float
2012-10-14T08:53:53 < dongs> bbl
2012-10-14T08:58:48 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-14T09:20:14 < cjbaird> *drawing faster
2012-10-14T09:29:11 -!- CheBuzz [~CheBuzz@76.164.192.124] has joined ##stm32
2012-10-14T09:56:26 -!- claude is now known as Claude
2012-10-14T09:57:50 -!- dekar [~dekar@212.255.38.46] has joined ##stm32
2012-10-14T09:57:50 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-14T10:01:25 -!- dekar_ [~dekar@212.255.46.253] has quit [Ping timeout: 256 seconds]
2012-10-14T10:20:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-14T10:27:01 -!- Claude is now known as claude
2012-10-14T10:37:19 -!- claude is now known as Claude
2012-10-14T11:09:49 < jpa-> dekar: unaligned accesses are slower as the CPU converts them to two aligned accesses
2012-10-14T11:40:32 -!- Claude is now known as claude
2012-10-14T11:41:59 -!- claude is now known as Claude
2012-10-14T11:48:35 -!- Claude is now known as claude
2012-10-14T11:52:05 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-14T11:56:37 < upgrdman> fun times reading lib comments.... "/* importance */"     :(
2012-10-14T11:59:02 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-14T12:11:59 < upgrdman> why are some software dev's afraid of enums ;(
2012-10-14T12:12:19 < jpa-> in C or in C++?
2012-10-14T12:12:25 < upgrdman> this damn code would be some much more readable with some enums insteading of magic numbers
2012-10-14T12:12:26 < upgrdman> c
2012-10-14T12:12:48 < jpa-> some stupid compilers complain when you do anything with enums
2012-10-14T12:13:19 < jpa-> like bitwise operations
2012-10-14T12:13:27 < upgrdman> really? i only use gcc/g++ and a while ago used msvc++ for school work... enums seemed fine
2012-10-14T12:13:40 < jpa-> sure
2012-10-14T12:13:47 < upgrdman> o, never tried bitwise stuff with enums
2012-10-14T12:14:10 < jpa-> ok, yeah, in the most basic case enums work :P
2012-10-14T12:14:23 < jpa-> but in my opinion also this is clearer with enums than #defines: https://code.google.com/p/nanopb/source/browse/pb.h#50
2012-10-14T12:14:36 < upgrdman> wouldn't casting the enum as an int help?
2012-10-14T12:14:45 < jpa-> it would, but it would be equally ugly
2012-10-14T12:18:24 < upgrdman> well the code im going through isn't doing bitwise stuff... i just see lots of if(n == 0x1234) ... where an enum would make it more readable
2012-10-14T12:20:01 < upgrdman> or even a bunch of #defines would help
2012-10-14T12:20:22 < upgrdman> anyway... i'm justing whining.
2012-10-14T12:27:13 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 246 seconds]
2012-10-14T12:32:04 -!- cjbaird_ [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-14T12:32:19 -!- cjbaird [~cjb@ppp121-44-204-61.lns20.syd7.internode.on.net] has quit [Ping timeout: 260 seconds]
2012-10-14T12:55:41 -!- barthess [~barthess@178.154.51.64] has joined ##stm32
2012-10-14T13:09:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-14T13:11:12 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-14T13:17:51 < dongs> hello dongs
2012-10-14T13:21:22 < dongs> http://i.imgur.com/PIeH7.jpg wow that is annoying.
2012-10-14T13:34:08 -!- |sark [~|sark@109.235.51.202] has quit [Quit: Verlassend]
2012-10-14T13:38:04 < zyp> sounds fun
2012-10-14T14:08:10 < dongs> anything better for temperature/industrial temps than aluminum caps?
2012-10-14T14:08:20 < dongs> i need like 200-600uf capacitance
2012-10-14T14:08:33 < dongs> and im not paralleling 20 ceramic caps
2012-10-14T14:09:49 -!- daku is now known as DaKu
2012-10-14T14:14:10 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-14T14:14:13 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-14T14:48:11 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-14T14:50:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-14T14:52:43 <+dekar> zyp, you said you were using C++11, have you tried using std::thread on the stm32? I wonder how the c++ library handles that
2012-10-14T14:56:39 < dongs> Laurenceb: where's your stm32logger schematic
2012-10-14T15:11:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-14T15:16:48 < zyp> dekar, no, I haven't
2012-10-14T15:17:28 < dongs> so zyp, what;ll happen when I try running usb off a 0.5% resonator
2012-10-14T15:17:44 < Thorn> dekar: it has to use OS threads, and you're quite unlikely to have something like posix threads running on your stm32
2012-10-14T15:19:18 < prog> oh! the std c++ aims to support some essential OS! isn't that wonderful
2012-10-14T15:19:34 < prog> next step is tu support a unified GUI model
2012-10-14T15:19:35 < zyp> apparently my library doesn't have std::thread
2012-10-14T15:19:41 < zyp> main.cpp:7:1: error: 'thread' in namespace 'std' does not name a type
2012-10-14T15:21:14 <+dekar> I just read C11 has threads.h as well
2012-10-14T15:21:36 <+dekar> I don't think newlib caught up with that though
2012-10-14T15:22:00 < Thorn> rtos tasks != threads anyway
2012-10-14T15:22:47 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Ping timeout: 240 seconds]
2012-10-14T15:22:48 <+dekar> well I was just curious ;)
2012-10-14T15:23:39 < Thorn> there was some "small" os with a posix api, but I'm not sure things like that belong on a microcontroller
2012-10-14T15:24:10 < zyp> dongs, that's kind of undefined, since you are operating out of spec :p
2012-10-14T15:24:45 < dongs> i guess ill slap a 12mhz xtal pads on the board then
2012-10-14T15:24:47 < zyp> it will probably work, but it may not be reliable
2012-10-14T15:24:52 < dongs> and will try with  8mhz resonator first.
2012-10-14T15:25:21 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-14T15:25:37 < zyp> maybe work for a while, then fail sporadicly
2012-10-14T15:25:54 < zyp> or just work with some hosts and not other
2012-10-14T15:26:23 < zyp> like the ANT radio shit I was doing once
2012-10-14T15:26:55 <+dekar> Thorn, I don't think you can do posix, afaik it needs an MMU
2012-10-14T15:27:20 <+dekar> but those C(++)11 threads seem to have way lower requirements
2012-10-14T15:27:23 < zyp> I had wrong loads on the caps causing a 0.01% error or so in frequency
2012-10-14T15:28:03 < karlp> huh, Cannot insert hardware breakpoint -1.\nCould not insert hardware breakpoints:\nYou may have requested too many...
2012-10-14T15:28:07 < karlp> that's new.
2012-10-14T15:28:07 < karlp> good job netbeans
2012-10-14T15:28:10 < Thorn> once you start porting e.g. linux multithreaded apps you'll find they use stack generously, etc.
2012-10-14T15:28:13 < zyp> it worked perfectly with other nRF based nodes, but TI based nodes weren't able to receive from it
2012-10-14T15:29:17 < zyp> because the nRF chips apparently got a larger tolerance for frequency error than the TI chips
2012-10-14T15:30:44 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-14T15:34:42 < Thorn> how much flash would be required for lwip + fatfs + usb on an f4? is 512Kb enough?
2012-10-14T15:35:21 < dongs> ok 12mhz rapemod complete
2012-10-14T15:35:25 < dongs> Thorn: uhhhhhhhhhhhhhhh yes
2012-10-14T15:35:29 < dongs> 128k is enoughfor that
2012-10-14T15:35:45 < Thorn> good, 1M parts seem to be twice as expensive
2012-10-14T15:37:00 < zyp> Thorn, planning to do usb ethernet?
2012-10-14T15:38:02 < Thorn> usb device, usb host for flash drives, parallel flash, ethernet, audio via i2s, 122x32 display
2012-10-14T15:38:21 < Thorn> I've got 2 projects done, need a 3rd one lol
2012-10-14T15:38:49 < Thorn> and microsd card
2012-10-14T15:39:00 < Thorn> (and CAN of course)
2012-10-14T15:40:15 < Thorn> hopefully a 100-pin f4 is enough for all that
2012-10-14T15:50:49 < Thorn> also what idiot came up with a 32 pixel high dislpay. can only fit 4 lines of 6x8 characters with no space between them which looks ugly as hell
2012-10-14T15:58:00 < dongs> what kinda dumfuk display is that
2012-10-14T15:58:06 < dongs> just use one of $15 chink 320x240 touchscreen TFTs
2012-10-14T15:59:30 < jpa-> Thorn: maybe a person who only needed two lines of text? :)
2012-10-14T15:59:56 < Thorn> I wonder if I have any 4.87K 1% resistors left.
2012-10-14T16:00:53 < jpa-> dongs: too bad if you want to do anything commercial, those TFTs change types every few months
2012-10-14T16:01:49 < dongs> jpa-: ? how so
2012-10-14T16:02:10 < dongs> jpa-: you can get them directly from the vendor with any requested spec
2012-10-14T16:02:19 < dongs> i.e. thaoyu electronis or wahtever that makes those cheapass 3.2" shits
2012-10-14T16:02:24 < dongs> will custom make a batch for you of any qty
2012-10-14T16:02:28 < dongs> chip/pin available/etc
2012-10-14T16:02:57 < dongs> like they only had 8/16bit version, so I asked them to make me one with SPI enabled by default
2012-10-14T16:03:05 < dongs> 2 days later they shipped me 2 displays with SPI
2012-10-14T16:03:26 < jpa-> so they just switched the solder jumpers on a premade board? sure
2012-10-14T16:03:38 < jpa-> but a year from now they might use a totally different controller on those boards
2012-10-14T16:04:34 < dongs> well, i ordered some expensive as fuck I2C lcd displays from newhaven
2012-10-14T16:04:37 < dongs> like 3-4 years ago
2012-10-14T16:04:45 < dongs> few months ago they sent me EOL notice for htem...
2012-10-14T16:04:52 < dongs> big difference??
2012-10-14T16:04:58 < jpa-> atleast you got a notice
2012-10-14T16:05:00 < dongs> thats a "expensive" "commercial" manufacturer
2012-10-14T16:05:27 < dongs> im saying bitching about something not available in a year is kinda lame
2012-10-14T16:05:36 < jpa-> but yeah, the same problem seems to exist also with more expensive vendors
2012-10-14T16:05:55 < jpa-> it's a problem nevertheless
2012-10-14T16:06:36 < zyp> it's rather easy abstracting out the various differences
2012-10-14T16:06:45 < jpa-> zyp: not the mechanical differences
2012-10-14T16:06:58 < zyp> sure, but you're talking about different controller
2012-10-14T16:06:58 < dongs> hes not talking about mechanical
2012-10-14T16:07:14 < dongs> and yeah thats easy enough to do as well
2012-10-14T16:07:38 < dongs> just incldue both drivers.. and if theres no way to autodetect (id register or so).. just add a jumper or something on new hardware revision...
2012-10-14T16:07:45 < dongs> no jumper = old driver, jumper = new one..
2012-10-14T16:09:10 < jpa-> on a different note, i'm currently testing some displaytech TFT modules (INT035TFT-TS).. they have marked the backlight power supply pin NC :)
2012-10-14T16:09:22 < dongs> here you go
2012-10-14T16:09:29 < dongs> some shitty overpriced vendor selling you chink shit
2012-10-14T16:09:33 < dongs> with improperly marked stuff.
2012-10-14T16:09:43 < jpa-> no, it's chinese vendor selling cheap chink shit
2012-10-14T16:09:51 < dongs> awesmn
2012-10-14T16:47:50 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-14T17:02:22 < dongs> are they fucking with stratos countdown again
2012-10-14T17:02:23 < dongs> ugh
2012-10-14T17:28:00 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-14T17:33:46 < zyp> whoops
2012-10-14T17:34:41 < dongs> seems oK now
2012-10-14T17:34:43 < dongs> 40mins
2012-10-14T17:35:04 < zyp> I put the cpu in deep sleep
2012-10-14T17:35:11 < zyp> killing the debugger connection
2012-10-14T17:39:05 < zyp> well, more like suspending than killing
2012-10-14T17:39:15 < zyp> it recovered correctly when I hit reset
2012-10-14T17:42:49 < dongs> wheres is Blagurenceb
2012-10-14T17:52:03 < emeb_mac> morning all...
2012-10-14T17:53:05 < zyp> no, just you
2012-10-14T17:53:16 < zyp> I've got afternoon and dongs got midnight in a few minutes
2012-10-14T17:56:12 < zyp> emeb_mac, so, what are you up to today?
2012-10-14T17:56:50 < emeb_mac> finishing assembly of that Beaglebone FPGA board.
2012-10-14T17:57:15 < scrts> of what?
2012-10-14T17:57:22 < emeb_mac> Probably try to bring my BB OS SD card up to date and see what I need to do to start testing it.
2012-10-14T17:58:40 < zyp> :)
2012-10-14T17:58:45 < emeb_mac> scrts: an expansion board for Beaglebone with a Xilinx Spartan S3A -> http://imagebin.org/231854
2012-10-14T17:59:18 < dongs> countdown 1hr again
2012-10-14T17:59:19 < dongs>  fuck you
2012-10-14T17:59:20 < zyp> I almost want one for my own BB
2012-10-14T17:59:21 < scrts> what's the point of such board? what do You want to do with it?
2012-10-14T17:59:30 < zyp> but I've got no idea what I'd do with it
2012-10-14T18:00:27 < scrts> I'd say some serdes or fast adc/dac.. but this fpga board is empty, so what's the point?
2012-10-14T18:00:53 < emeb_mac> there's uncommitted I/O on the right-side connector.
2012-10-14T18:01:23 < emeb_mac> the guy I'm working with wants it to build a smart motor controller for 3-d printing.
2012-10-14T18:02:12 < emeb_mac> (stepper drivers with accel control, etc to ease realtime requirements on OS)
2012-10-14T18:02:42 < emeb_mac> I've got some software-defined radio stuff I'd like to try (tuning, resampling, demodulation, etc)
2012-10-14T18:03:29 < zyp> how is the connection towards the BB? is SPI the fastest way to push data?
2012-10-14T18:03:59 < emeb_mac> zyp: there's spi for initial config, but the 16-bit wide GPMC is also available.
2012-10-14T18:04:32 < zyp> the speed of initial config is not very interesting once it's configured :p
2012-10-14T18:05:33 < emeb_mac> no, but if you _don't_ need GPMC then spi is also available for after config and is easier to set up in the OS.
2012-10-14T18:06:31 < emeb_mac> other guy is hacking the Linux/ARM NAND flash driver (which uses GPMC) to be used for this board.
2012-10-14T18:08:00 < zyp> ah
2012-10-14T18:08:06 < dongs> lunix nand
2012-10-14T18:08:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-14T18:08:51 < emeb_mac> dongs: IIRC there's a Windows CE build for Beaglebone. You can use that one. :P
2012-10-14T18:09:05 < emeb_mac> bring $$$
2012-10-14T18:09:08 < dongs> not win8rt?
2012-10-14T18:09:10 < dongs> forget it.
2012-10-14T18:10:24 < emeb_mac> there's also been some discussion of using the FPGA as a coprocessor to offload some compute tasks that can be done efficiently in hardware.
2012-10-14T18:10:37 < dongs> > buttcoin mining
2012-10-14T18:10:48 < zyp> heh
2012-10-14T18:10:52 < emeb_mac> yeah - crypto type stuff comes to mind.
2012-10-14T18:11:35 < Thorn> already being done
2012-10-14T18:11:46 < emeb_mac> Thorn: naturally.
2012-10-14T18:12:04 < emeb_mac> nothing new under the sun.
2012-10-14T18:12:07 < zyp> bitcoin mining on a spartan-3 is not going to be very efficient
2012-10-14T18:12:09 < Thorn> there're even asic bitcoin miners already
2012-10-14T18:12:17 < emeb_mac> no doubt.
2012-10-14T18:13:18 < emeb_mac> but, as noted above this is uncommitted hardware - bring your own ideas.
2012-10-14T18:14:26 < emeb_mac> If I were to redesign it I'd probably use S6, add RF-capable ADC/DAC and make something specifically for SDR.
2012-10-14T18:16:11 < zyp> sign me up for one of those boards if you make some ;)
2012-10-14T18:16:48 < emeb_mac> zyp: what would you think about an STM32F4xx + FPGA?
2012-10-14T18:17:11 < zyp> probably cool
2012-10-14T18:17:12 < dongs> emeb_mac: only if FPGA is doing ddr2 controller
2012-10-14T18:17:25 < dongs> (among other shit)
2012-10-14T18:17:38 < emeb_mac> dongs: that has been on my mind.
2012-10-14T18:18:00 < emeb_mac> although maybe not DDR2 - probably just SDRAM to keep things simpler.
2012-10-14T18:18:30 < zyp> at that point you could just use lpc4300 :p
2012-10-14T18:18:58 < emeb_mac> yeah - that's the thing. better alternatives if that's the only reason.
2012-10-14T18:19:49 < emeb_mac> so the question is what sort of additional features on an F4+FPGA would make it more useful.
2012-10-14T18:20:48 < emeb_mac> thinking 100-pin pkg F4 so muxed FSMC is available for the interface between them.
2012-10-14T18:22:31 < Thorn> I've been thinking about something like that a few months ago, code name for the project was "an ARM and a leg"
2012-10-14T18:23:54 < jpa-> stm32f4xx + fpga is a quite standard solution, but i don't think it is so useful as a generic board
2012-10-14T18:25:15 < scrts> we use fpga with stm32f2xx connected through DCMI
2012-10-14T18:25:22 < scrts> that interface is damn good :>
2012-10-14T18:26:51 < jpa-> huh, fpga connected dcmi? that's something new :)
2012-10-14T18:26:54 < emeb_mac> scrts: interesting. Is DCMI bi-directional?
2012-10-14T18:27:10 < Thorn> and is there addressing?
2012-10-14T18:27:56 <+Steffann> So, try 2 ..
2012-10-14T18:28:59 < zyp> emeb_mac, no, it's in-only
2012-10-14T18:29:15 < emeb_mac> thought so.
2012-10-14T18:29:30 < zyp> it's designed to receive camera data
2012-10-14T18:29:44 < emeb_mac> yup. that's why I was suspicious.
2012-10-14T18:30:21 < dongs> http://www.youtube.com/watch?v=MrIxH6DToXQ
2012-10-14T18:30:23 < dongs> troll in progress
2012-10-14T18:30:31 <+Steffann> Hmm, streams seems to be a little overloaded??
2012-10-14T18:30:48 <+Steffann> http://www.redbullstratos.com/live that page is 'nicer' gives you some more info
2012-10-14T18:30:54 <+Steffann> bye bye
2012-10-14T18:31:32 < Thorn> finally someone thought about live telemetry over the net
2012-10-14T18:32:02 < dongs> where's laurenceb
2012-10-14T18:32:12 <+Steffann> in the sky now?
2012-10-14T18:32:31 <+Steffann> His real name is Felix baumgartner
2012-10-14T18:33:00 < dongs> haha
2012-10-14T18:33:23 <+Steffann> It means the mission is going to fail
2012-10-14T18:34:35 < emeb_mac> whole lotta people gonna have stiff necks today.
2012-10-14T18:35:30 < dongs> oh no, buffering
2012-10-14T18:35:36 < Thorn> same here
2012-10-14T18:36:41 < dongs> watch him get sniped by a 747
2012-10-14T18:38:53 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
2012-10-14T18:39:34 < dongs> 4.x = meters/second?
2012-10-14T18:41:33 < dongs> dont they have laurenceb-style telemetry
2012-10-14T18:41:35 < dongs> why the fuck is he asking him for this shit
2012-10-14T18:42:22 < zyp> because it's probably fucking boring to just sit there
2012-10-14T18:42:22 -!- DaKu is now known as daku
2012-10-14T18:43:28 < dongs> they should just play some porn
2012-10-14T18:52:42 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-14T19:09:20 < cjbaird_> Have the crew constantly keep telling the guy how he's going to die.
2012-10-14T19:09:26 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-14T19:09:30 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-14T19:11:41 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 252 seconds]
2012-10-14T19:24:22 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-14T19:50:09 -!- daku is now known as DaKu
2012-10-14T19:56:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-14T20:03:52 < emeb_mac> think they'll abort because he can't defog his faceplate?
2012-10-14T20:04:51 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-14T20:06:50 -!- DaKu is now known as daku
2012-10-14T20:09:55 < Thorn> oh wow, it gets warmer in the stratosphere because you *reduce your distance to the sun*
2012-10-14T20:10:11 < Thorn> like halfway there already
2012-10-14T20:11:24 < dongs> haha
2012-10-14T20:13:19 < emeb_mac> *eyerolls*
2012-10-14T20:14:29 < Thorn> I drew an f407v symbol
2012-10-14T20:14:45 < Thorn> won't show it or dongs may have a heart attack
2012-10-14T20:16:03 < emeb_mac> in which cad tool?
2012-10-14T20:16:27 < cjbaird_> http://i.imgur.com/rwiCd.jpg
2012-10-14T20:17:04 < Thorn> diptrace
2012-10-14T20:20:17 < emeb_mac> the crowd goes mild
2012-10-14T20:23:52 < Thorn> omg qfp-464 are you serious
2012-10-14T20:24:18 < emeb_mac> dafuq?
2012-10-14T20:24:34 < zyp> sounds huge
2012-10-14T20:24:52 < emeb_mac> 0.1mm pitch ;)
2012-10-14T20:25:00 < Thorn> diptrace has this pattern and several in the 200-300 range too
2012-10-14T20:25:05 < Thorn> 0.3mm actually
2012-10-14T20:25:23 < dongs> Thorn: ST has recommended land patterns for their chips.
2012-10-14T20:25:28 < emeb_mac> try and hand-solder that.
2012-10-14T20:25:42 < dongs> just draw it in pattern editor. takes like 30 seconds
2012-10-14T20:26:01 < Thorn> it also has 3 qfp-100 patterns, 16mm, 16x6mm, 17.2mm
2012-10-14T20:26:21 < Thorn> I already used one of them and it worked well with f103v
2012-10-14T20:26:32 < Thorn> just need to find out which one
2012-10-14T20:45:34 < dongs> thorn in the time you spent masturbating looking for a pattern you could have drawn your own from datasheet
2012-10-14T20:47:44 -!- mra^ [~mra@184.22.81.68] has joined ##stm32
2012-10-14T20:49:24 < Thorn> I used a recommended pattern for a supercap for the f103 board. ended up filing down supercap pins
2012-10-14T20:52:37 < Thorn> >Also, note that the CLK line doesn't have a pull-up.
2012-10-14T20:52:43 < Thorn> is this true (sdio)?
2012-10-14T20:56:41 < dongs> hisssssssssssss
2012-10-14T20:56:56 < Thorn> *cough* *cough* *cough*
2012-10-14T20:57:06 < jpa-> why wouldn't it be true?
2012-10-14T21:00:01 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-14T21:00:57 < Thorn> jpa-: I don't remember any authoritative source that would say that
2012-10-14T21:01:05 -!- Tectu is now known as tectu
2012-10-14T21:04:25 < jpa-> so what source did say that?
2012-10-14T21:04:48 < jpa-> and does some source say that it does have a pull-up (apart from the weak pull-up that is on every pin)
2012-10-14T21:04:57 < dongs> lo
2012-10-14T21:05:42 < Thorn> I quoted a thread on the st forum. and I've recently seen a schematic where there was a pullup on clk. but it could be some arduino shield or something
2012-10-14T21:14:40 < emeb_mac> so he broke the altitude record and the speed record, but not the freefall time record?
2012-10-14T21:16:23 -!- claude is now known as Claude
2012-10-14T21:16:35 < Thorn> can f4 usb work if Vbus is not connected?
2012-10-14T21:16:48 < jpa-> Thorn: yes
2012-10-14T21:18:35 < jpa-> Thorn: you need to set NOVBUSSENS in OTG_FS_GCCFG
2012-10-14T21:19:58 < Thorn> do I need to sense Vbus then? or can the core still sense host connection? (device isn't bus-powered and is not always connected)
2012-10-14T21:20:38 < jpa-> it can sense when the host sends usb reset
2012-10-14T21:21:08 < TitanMKD> IIRC VBUS sense is only required to switch from Host to Device no ? (or maybe also to enter LowPower/sleep)
2012-10-14T21:21:10 < jpa-> but i'm not sure how safe it is to keep the usb controller enabled if usb host might be powered off.. someone suggested that it might be bad
2012-10-14T21:21:11 < Thorn> good, thanks
2012-10-14T21:21:24 < jpa-> Thorn: ID is used for host/device selection
2012-10-14T21:21:32 < jpa-> vbus is more for low power AFAIK
2012-10-14T21:21:40 < Thorn> TitanMKD: synopsys cores have internal pullup resistors
2012-10-14T21:21:41 < TitanMKD> jpa- only ID they do not also check vbus ?
2012-10-14T21:21:59 < jpa-> TitanMKD: not sure
2012-10-14T21:22:22 < jpa-> but just vbus is not enough if i have understood correctly
2012-10-14T21:22:25 < scrts> emeb_mac -> no, DCMI is input only in stm32
2012-10-14T21:24:34 < dongs> 3.30am, sleep time
2012-10-14T21:25:39 < zyp> jpa-, shouldn't be any problems
2012-10-14T21:26:36 < Thorn> the OTG_HS_VBUS pin is busy if you use ethernet
2012-10-14T21:27:03 < Thorn> and there's no remapping
2012-10-14T21:32:44 -!- Claude is now known as claude
2012-10-14T21:33:29 < zyp> if you don't need OTG, leaving out vbus shouldn't be any problems
2012-10-14T21:35:43 < Thorn> I don't in this case. but it looks like you can't use usb HS in otg mode togeter with ethernet because that pin is shared
2012-10-14T21:36:57 < zyp> unless you use ULPI of course
2012-10-14T21:37:20 < zyp> oh, wait, ULPI also shares pins with ethernet?
2012-10-14T21:37:50 < Thorn> most likely
2012-10-14T21:39:05 < zyp> oh, wait, ethernet is remappable to some PG pins
2012-10-14T21:39:46 < zyp> on a large enough package :p
2012-10-14T21:40:25 < Thorn> I'm talking about 100-pin packages
2012-10-14T21:41:21 < zyp> also, see appendix A.1 in datasheet
2012-10-14T21:41:44 < jpa-> i wonder if you could do the OTG selection in software and just set force device or force host bits accordingly
2012-10-14T21:41:53 < jpa-> then you could just use any EXTI pin
2012-10-14T21:41:59 < zyp> yep
2012-10-14T21:42:34 < zyp> but then you also have to consider HNP/SRP
2012-10-14T21:43:43 < Thorn> okay looks like config 2 is acceptable for me
2012-10-14T21:44:17 < zippe> flyback: uh, fauxnews you are reading because?
2012-10-14T21:45:40 -!- CheBuzz [~CheBuzz@76.164.192.124] has quit [Ping timeout: 246 seconds]
2012-10-14T21:48:45 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-14T21:53:01 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-14T22:01:58 -!- mra^ [~mra@184.22.81.68] has left ##stm32 []
2012-10-14T22:16:32 -!- daku is now known as DaKu
2012-10-14T22:17:09 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-14T22:32:39 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-14T22:35:08 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 245 seconds]
2012-10-14T22:45:26 < emeb> Fully assembled board -> http://imagebin.org/231942
2012-10-14T22:45:50 <+Steffanx> :)
2012-10-14T22:46:22 <+Steffanx> You're a little slow though
2012-10-14T22:46:24 <+Steffanx> 2011 ..
2012-10-14T22:46:55 < emeb> Yeah - originally laid out back in Dec 2011. For another guy who's been dragging his feet.
2012-10-14T22:47:16 < emeb> Finally got tired of waiting for him, so I did it myself.
2012-10-14T22:48:03 < Thorn> 2 layers?
2012-10-14T22:48:05 < emeb> 4
2012-10-14T22:48:55 < emeb> I have an alternate layout that's just 2, but the supply & gnd routing gets... iffy.
2012-10-14T22:49:49 < emeb> There's a bit of routing and a bunch of supply bypass caps on the backside.
2012-10-14T22:50:42 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-14T23:14:35 < barthess>  emeb: what is the board?
2012-10-14T23:14:50 < Thorn> what i2s codecs and opamps (if any) do I choose so I can connect 120 Ohm headphones and get acceptable audio quality? do I need to isolate analog power from digital, etc.? not trying to do hi-fi but it would be nice if it sounded well
2012-10-14T23:15:46 < emeb> barthess: It's an expansion board for Beaglebone with a Xilinx FPGA.
2012-10-14T23:16:04 < zyp> Thorn, looked at the f4discovery?
2012-10-14T23:16:22 < emeb> Thorn: I've had good luck with the Wolfson WM8731 I2S codec. It has a built-in headphone amp, so you could hook up directly.
2012-10-14T23:16:56 < Thorn> zyp: I'd really prefer something non-QFN
2012-10-14T23:17:35 < emeb> Thorn: -> http://ebrombaugh.studionebula.com/synth/stm32f4_codec/index.html
2012-10-14T23:17:37 < Thorn> also I've really no idea if I should bother with bipolar +/-15V power on such a pcb
2012-10-14T23:17:59 < emeb> U3 is codec. Runs off +3.3V only.
2012-10-14T23:18:27 < Thorn> yeah I've been looking at this board yesterday
2012-10-14T23:18:51 < emeb> I didn't hook up the headphone outputs, just the line outputs.
2012-10-14T23:19:02 < emeb> but the DS suggests it's pretty simple.
2012-10-14T23:19:06 -!- tavish_ [~tavish@120.56.140.134] has joined ##stm32
2012-10-14T23:20:01 < Thorn> is it enough for high impedance headphones?
2012-10-14T23:20:30 < emeb> Should be. Let's check the DS to make sure though.
2012-10-14T23:21:05 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-10-14T23:21:35 < emeb> DS says it's optimized for 16 & 32 ohm headphones.
2012-10-14T23:22:09 < emeb> One would think 120 ohm headphones would be less of a load and would also be OK.
2012-10-14T23:24:15 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-14T23:24:17 < emeb> Based on my experience with this board on the bench, I'd suggest providing a separate regulator for the analog supplies. I can hear ripple from the MCU getting through at low levels.
2012-10-14T23:24:32 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-14T23:25:24 < Thorn> right. I'm thinking about a dc/dc so even more incentive for a separate ldo
2012-10-14T23:25:36 < emeb> good plan.
2012-10-14T23:26:36 < emeb> although - most modern smps run at very high rates - well beyond audio. Still good to provide a clean analog rail.
2012-10-14T23:33:43 < Thorn> is there any advantage in an onboard cmoy-like circuit (i.e. an op amp)?
2012-10-14T23:35:05 < emeb> since there's already a headphone driver on-chip, I wouldn't think so.
2012-10-14T23:35:47 < emeb> unless the cmoy ckt has higher quality driver than the built-in one (and that matters to you).
2012-10-14T23:36:14 < Thorn> that's what I'm basically talking about
2012-10-14T23:36:50 < Thorn> but it seems to require bipolar power
2012-10-14T23:37:42 < emeb> Well, given the CMOY stuff I've seen recommends fairly high-spec op-amps, and higher voltage rails, I'd guess there's a good chance that it might sound better to those with "golden ears".
2012-10-14T23:38:23 < emeb> but I'd be hard pressed to tell you for certain just by looking at the datasheets.
2012-10-14T23:39:10 < emeb> Perhaps if you have room you could build up both alternatives. Report back which is better. :)
2012-10-14T23:39:24 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-14T23:41:22 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-14T23:50:14 < Thorn> emeb: have you bothered with a low noise LDO for analog power?
--- Day changed Mon Oct 15 2012
2012-10-15T00:02:44 -!- tavish_ [~tavish@120.56.140.134] has quit [Quit: leaving]
2012-10-15T00:06:48 < Bird|lappy> emeb, you'll want to make sure that high frequency noise stays off analog rails
2012-10-15T00:06:50 < emeb> Thorn: I've done it once, on a USB Audio codec. I couldn't tell the difference, but the mfgs recommendations suggested it bought a few extra dB of SNR.
2012-10-15T00:07:03 < Bird|lappy> op amp CMRR falls off linearly with frequency past a certain (rather low) point
2012-10-15T00:07:09 < Bird|lappy> (and PSRR, too)
2012-10-15T00:07:17 < emeb> Bird|lappy: yup.
2012-10-15T00:07:37 < Bird|lappy> one idea would be to use some ferrite beads and/or small chokes to help filter the analog rails
2012-10-15T00:08:12 < emeb> Bird|lappy: I've got ferrite bead on mine. Of course it doesn't do diddly at audio frequencies.
2012-10-15T00:08:41 < Bird|lappy> it isn't audio feedthrough I'd worry about, but high frequency noises that can then get rectified
2012-10-15T00:09:28 < emeb> Yeah - nonlinearities can bring them back down into the audible range.
2012-10-15T00:09:37 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-15T00:10:04 < Bird|lappy> hence, beads n' chokes
2012-10-15T00:10:45 < Bird|lappy> and yeah, the thing with driving a load from an opamp with 3.3V and 0 as the supply rails would be a shortage of headroom
2012-10-15T00:18:51 < Thorn> found some recommendations which are: isolated power, buffer DAC output with a BB op amp with -2 gain, good clock source (apparently crystal is not good enough??)
2012-10-15T00:19:37 < Bird|lappy> one idea for a headphone driver would be to repurpose a DSL line driver IC
2012-10-15T00:19:41 < Bird|lappy> but that might be overkill :P
2012-10-15T00:19:54 < emeb> Bear in mind that the MCLK to the codec is probably coming from a PLL (in my board it is).
2012-10-15T00:20:03 < emeb> PLL jitter freaks some audiophiles out.
2012-10-15T00:20:14 < Bird|lappy> (a decent DSL line driver is as close to "flat wire with gain" as you can get in the audio band)
2012-10-15T00:21:32 < tectu> is  «Developper's»  right, or  «Developer's»  =
2012-10-15T00:21:33 < tectu> ?*
2012-10-15T00:21:43 < Bird|lappy> one p
2012-10-15T00:22:13 < tectu> thanks :)
2012-10-15T00:22:20 < Thorn> what pll jitter, I'm not doing 200MHz ADC sampling here. shoot all audiophiles
2012-10-15T00:22:37 < emeb> Thorn: that's pretty much how I feel about it.
2012-10-15T00:22:57 < Thorn> and give me their gear
2012-10-15T00:23:05 < emeb> Audiophiles != Engineers (in general) and many don't seem to understand "good enough".
2012-10-15T00:23:32 < emeb> Where "good enough" means "there's no fucking way you could possible hear it."
2012-10-15T00:24:01 < emeb> s/ble/bly/
2012-10-15T00:24:12 < Thorn> one cmoy clone kit I googled an hour ago included a gold-plated dip socket for the amp
2012-10-15T00:24:39 < emeb> because that's really going to help. :)
2012-10-15T00:26:48 < Thorn> and polypropylene caps only, of course
2012-10-15T00:27:01 < emeb> well, DUHHH!
2012-10-15T00:27:13 < emeb> for the bypass caps to naturally.
2012-10-15T00:27:20 < emeb> s/to/too/
2012-10-15T00:27:50 < Bird|lappy> I'd rather have MLCCs for bypassing
2012-10-15T00:28:04 < Bird|lappy> and coupling caps should flatly DIAF
2012-10-15T00:28:31 < emeb> Well, you can't always be fully DC coupled. They have their place.
2012-10-15T00:30:59 < Thorn> and 1% resistors only!
2012-10-15T00:31:14 < emeb> Wolfson has a pretty good appnote on choosing caps for audio applications: http://www.wolfsonmicro.com/documents/uploads/misc/en/WAN0176.pdf
2012-10-15T00:32:54 < emeb> Going AFK - bbiaw
2012-10-15T00:32:59 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-15T00:34:22 -!- tectu is now known as Tectu
2012-10-15T01:00:31 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-15T01:05:58 -!- barthess [~barthess@178.154.51.64] has quit [Ping timeout: 245 seconds]
2012-10-15T01:30:14 -!- barthess [~barthess@178.154.5.176] has joined ##stm32
2012-10-15T01:31:29 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-15T01:39:15 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 252 seconds]
2012-10-15T01:43:24 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-15T01:57:28 -!- barthess [~barthess@178.154.5.176] has quit [Quit: Leaving.]
2012-10-15T01:59:05 -!- DaKu is now known as daku
2012-10-15T02:04:58 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 246 seconds]
2012-10-15T02:24:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-15T02:34:17 < cjbaird_> Hey guise... What if.. that made a laptop that could be recharged through its USB ports..
2012-10-15T02:35:03 < BrainDamage> if usb could have sw selectable voltage up to 25V and current up to 5A, should be fine
2012-10-15T02:35:21 < BrainDamage> altough I guess 5A trough the smallest connectors might have contact resistance issues
2012-10-15T02:36:27 < cjbaird_> I'm thinking its recharger would do the trickle-charge thing, drawing the ~2.5watts
2012-10-15T02:37:05 < BrainDamage> that wouldn't even work with an atom
2012-10-15T02:37:31 < zyp> haven't you read about the high power usb stuff?
2012-10-15T02:37:35 < zyp> http://www.usb.org/developers/powerdelivery/
2012-10-15T02:38:39 < cjbaird_> Leaving it like that for a day would probably completely recharge a flat laptop battery.. 'Emergency recharge', etc.
2012-10-15T02:39:41 < BrainDamage> more like 2 days for an average 350kJ battery
2012-10-15T02:39:47 < BrainDamage> assuming 100% efficiency
2012-10-15T02:40:39 < cjbaird_> Hmm, USB3.0 allows 1.5A @ 5VDC, so that's 7.5watts recharging..
2012-10-15T02:40:44 < zyp> hmm, goal for usb power delivery seems to be 100W
2012-10-15T02:40:45 < BrainDamage> zyp: premises sound great
2012-10-15T02:41:02 < zyp> 20V 5A
2012-10-15T02:41:12 < cjbaird_> Wait, USB3 max current seems to be 5A..
2012-10-15T02:41:23 < zyp> micro connectors are apparently limited to 3A
2012-10-15T02:41:25 < zyp> so 60W
2012-10-15T02:41:36 < zyp> I suggest reading this: http://www.usb.org/developers/powerdelivery/PD_1.0_Introduction.pdf
2012-10-15T02:41:50 < cjbaird_> My EeePC701 draws ~22W, and the Lemote is 9-16W..
2012-10-15T02:42:16 < BrainDamage> I was just readin that
2012-10-15T02:42:49 < zyp> hmm
2012-10-15T02:44:25 < cjbaird_> Neat. So I'm not a genius that is way ahead of the Industry. :)
2012-10-15T02:45:59 < BrainDamage> you could at least wait for dongs or Laurenceb to start the trolling
2012-10-15T02:47:35 < zyp> judging by how long my batteries lasts, my macbook seems to draw slightly less than 15W during normal load (bright display, video playback)
2012-10-15T02:48:45 < cjbaird_> The OLPC OX-1 ran at 2 watts..
2012-10-15T02:52:50 < zyp> well, yeah, but then it was designed as a low power laptop
2012-10-15T02:53:03 < zyp> didn't some of the OLPC stuff have a crank to charge it? :p
2012-10-15T02:55:18 < dongs> sup trolls
2012-10-15T03:00:24 < dongs> haha
2012-10-15T03:00:30 < dongs> http://i.imgur.com/7XkUC.gif
2012-10-15T03:01:43 < qyx_> like xmas tree \o/
2012-10-15T03:01:53 < qyx_> never enough diagnostic leds
2012-10-15T03:03:12 < Thorn> qyx_: I have 30 on a small (80x35mm) board, is it enough?
2012-10-15T03:03:28 < zyp> pff
2012-10-15T03:03:29 < Thorn> *20
2012-10-15T03:03:37 < zyp> I have 256 on a 50x32mm board
2012-10-15T03:03:38 < qyx_> i have 9 on 70x80
2012-10-15T03:03:56 < Thorn> zyp: all of them diagnostic?
2012-10-15T03:04:01 < zyp> of course
2012-10-15T03:04:10 < qyx_> propably drawing nice icon representing error
2012-10-15T03:04:10 < Thorn> each with specific meaning?
2012-10-15T03:04:27 < Thorn> that doesn't count
2012-10-15T03:05:04 < qyx_> inspired by morse code and bios beep codes to signal different error states
2012-10-15T03:05:06 < qyx_> looks nice
2012-10-15T03:07:16 < zyp> http://bin.jvnv.net/f/ldMgR.JPG <- each led indicates that both it's corresponding row and col is powered at the same time, don't come here and tell me that it's not diagnostic :p
2012-10-15T03:07:45 < qyx_> lol
2012-10-15T03:08:16 < Thorn> ...
2012-10-15T03:08:28 < Thorn> you can also say each led indicates that it is lit
2012-10-15T03:08:34 < zyp> exactly :D
2012-10-15T03:11:51 < zyp> so, who's going to pull out one of those tiny oleds and claim «I've got even more leds in even less area, so I win!»?
2012-10-15T03:20:36 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 248 seconds]
2012-10-15T03:29:05 < dongs> just picked up leds from post
2012-10-15T03:39:16 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-15T03:46:14 < emeb_mac> finally showed up?
2012-10-15T03:51:35 < dongs> ya.
2012-10-15T03:52:27 < dongs> picked them up on the way to town
2012-10-15T03:58:26 < emeb_mac> nice
2012-10-15T04:26:09 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Read error: No route to host]
2012-10-15T04:26:11 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-15T04:26:40 < emeb_mac> good times.
2012-10-15T04:41:01 < dongs> shit is well packed
2012-10-15T04:41:18 < dongs> i dont see any blems on mine bro
2012-10-15T04:41:34 < dongs> my shit is smooth as china can make it
2012-10-15T04:44:39 < dongs> mine don't have the puzzle border.
2012-10-15T04:46:55 < dongs> you didnt mention the senders name was XuDong
2012-10-15T04:48:04 < emeb_mac> heh.
2012-10-15T04:48:16 < emeb_mac> didn't notice the similarity.
2012-10-15T04:48:21  * emeb_mac is oblivious.
2012-10-15T04:49:00 < emeb_mac> an mine don't have the puzzle border either.
2012-10-15T04:49:12 < dongs> pins fit to my printed board.
2012-10-15T04:49:12 < emeb_mac> (like the Sparkfun ones did)
2012-10-15T04:49:15 < dongs> so i think im good to go
2012-10-15T04:49:42 < emeb_mac> got your boards already?
2012-10-15T04:49:57 < dongs> no, i printed the shit out on paper. gonna submit today,  get wednesday or somethign
2012-10-15T04:50:07 < dongs> no point since 9221 drivers are still waiting to be shipped today.
2012-10-15T04:50:53 < emeb_mac> there's that.
2012-10-15T04:51:31 < dongs> but by end of week i should have all.
2012-10-15T04:52:28 < emeb_mac> then it's just code
2012-10-15T04:52:55 < dongs> yeah lighting a single row up shouldnt be hard.
2012-10-15T04:54:39 < emeb_mac> caek. ish.
2012-10-15T04:55:09 < dongs> how many boards do you want
2012-10-15T04:55:14 < dongs> did you say you got like 3 matrixes?
2012-10-15T04:55:27 < emeb_mac> yeah - I've got 3
2012-10-15T04:56:37 < dongs> ill send 4 or something so you can fuck one up
2012-10-15T04:56:43 < emeb_mac> cool.
2012-10-15T04:56:55 < emeb_mac> bare boards or assembled?
2012-10-15T04:57:24 < dongs> depends how you want them, i was gonna send board + driver chips but could be whatever
2012-10-15T04:57:46 < emeb_mac> boards + 9221 chips would be great. I can get the rest of the stuff easy.
2012-10-15T04:58:53 < emeb_mac> can I paypal you for costs? postage, etc?
2012-10-15T04:59:01 < dongs> ya, after its done
2012-10-15T04:59:10 < emeb_mac> great.
2012-10-15T05:01:29 < dongs> yesterday i realized the 8mhz resonator might be too shitty for USB so I added pads for my other crystal I use, 12mhz 3225
2012-10-15T05:01:34 < dongs> the shit I linked you few days ago
2012-10-15T05:02:26 < emeb_mac> yeah.
2012-10-15T05:02:50 < emeb_mac> I know some AVR stuff uses resonators for USB and they get by. But it's best to do it right.
2012-10-15T05:04:47 < dongs> fucking Laurenceb didnt show up all day
2012-10-15T05:05:00 < dongs> i need his stm32 2logger schematic :(
2012-10-15T05:05:11 < dongs> so i could effortlessly clone it
2012-10-15T05:05:58 < emeb_mac> got a use for it?
2012-10-15T05:06:13 < dongs> no, but im making something wiht F103 + USB + SDCard
2012-10-15T05:06:18 < dongs> might as well reuse half of his shit
2012-10-15T05:06:22 < dongs> since he seems to have got it to work somewhat
2012-10-15T05:06:54 < emeb_mac> SPI-based SD I assume - no SDIO on F103
2012-10-15T05:06:58 < dongs> right
2012-10-15T05:07:05 < dongs> there is SDIO but in 100+ pin package i think
2012-10-15T05:08:01 < emeb_mac> if you don't need the speed though it's good to go with the smaller pkg
2012-10-15T05:08:20 < dongs> nope, and i dont need to stock 100pin shits either
2012-10-15T05:08:27 < dongs> and i think he used 48lqfp
2012-10-15T05:08:35 < emeb_mac> like those little pkgs.
2012-10-15T05:09:10 < emeb_mac> F373 in 48lqfp is so cool - full floating point in a tiny space.
2012-10-15T05:09:24 < dongs> apparently kinetis shit still dosnt have M4F
2012-10-15T05:09:32 < dongs> the teensy3 trash is just M4
2012-10-15T05:09:46 < emeb_mac> lame
2012-10-15T05:09:47 < dongs> I remembver getting a hardon for kinetis shit cuz they were first to have M4 stuff and then like "SORRY FLOATING POINT OPTIONAL LOL"
2012-10-15T05:10:06 < emeb_mac> yeah - you have to go to the high-end parts for the FPU
2012-10-15T05:10:09 < dongs> then one look at their codewarrior shit and I was like forget it.
2012-10-15T05:10:23 < cjbaird_> dongs strikes me more as a M2F type
2012-10-15T05:10:32 < dongs> cjbaird_ is more like A2M
2012-10-15T05:11:13 < emeb_mac> b2b?
2012-10-15T05:11:34 < dongs> dong2dong
2012-10-15T05:12:10 < cjbaird_> Okay.. I had to google it.
2012-10-15T05:12:15 < dongs> haha
2012-10-15T05:12:44 < emeb_mac> I learn so much here.
2012-10-15T05:13:51 < dongs> to be fair I had to google m2f as well.
2012-10-15T05:13:58 < dongs> but lol'd when found.
2012-10-15T05:14:08 < dongs> also 1st link is some thai site, hilarious.
2012-10-15T05:14:28  * emeb_mac inferred that one correctly.
2012-10-15T05:15:06 < cjbaird_> http://i.imgur.com/E00nl.gif
2012-10-15T05:15:22 < dongs> haha
2012-10-15T05:15:47 < emeb_mac> and still landed on its feet.
2012-10-15T05:16:46 -!- cjbaird_ is now known as cjbaird
2012-10-15T05:25:40 -!- BrainDamage1 [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-15T05:25:50 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-15T05:39:06 < dongs> damn electronics_lee has been busy
2012-10-15T05:39:11 < dongs> someone bought 40 of those displays
2012-10-15T05:39:30 < dongs> and price went up
2012-10-15T05:39:31 < dongs> lol
2012-10-15T05:41:30 < emeb_mac> to what?
2012-10-15T05:41:51 < dongs> 6.98
2012-10-15T05:42:07 < emeb_mac> mouse nuts
2012-10-15T05:45:08 < dongs> HCT is faster than HC right?
2012-10-15T05:45:23 < emeb_mac> no
2012-10-15T05:45:27 < dongs> oh?
2012-10-15T05:45:44 < emeb_mac> hct has TTL compatible logic thresholds on input.
2012-10-15T05:46:09 < emeb_mac> so Vil < 0.8, Vih > 2.4 etc
2012-10-15T05:46:19 < dongs> oh.
2012-10-15T05:46:23 < dongs> yeah, thats what i need. hmm
2012-10-15T05:46:31 < emeb_mac> as opposed to HC which has CMOS levels - threshold is like VDD/2
2012-10-15T05:46:34 < dongs> looking for  a quad buffer with TTL inputs but 5V vcc
2012-10-15T05:46:51 < dongs> so i'm assuking 74xx126-type shit
2012-10-15T05:46:53 < dongs> assuming
2012-10-15T05:47:13 < dongs> but ccant be too slow
2012-10-15T05:49:44 < emeb_mac> 12ns seems fast
2012-10-15T05:49:53 < dongs> sure. which part is that?
2012-10-15T05:50:15 < emeb_mac> 74hct126
2012-10-15T05:50:30 < emeb_mac> http://www.datasheetcatalog.org/datasheet/SGSThomsonMicroelectronics/mXurtvu.pdf
2012-10-15T05:50:31 < dongs> oh wtf NOW it shows up
2012-10-15T05:50:41 < dongs> i did hct before and digishit showed nothing available
2012-10-15T05:50:57 < emeb_mac> server malf
2012-10-15T05:51:00 < dongs> 0oh.. eyah they'er all nonstock
2012-10-15T05:51:01 < dongs> wth
2012-10-15T05:51:18 < dongs> anything else? AHC about same?
2012-10-15T05:51:29 < dongs> i wonder why quad buffers arent popular
2012-10-15T05:51:34 < dongs> peopel buy hex instead?
2012-10-15T05:52:13 < dongs> wtfak
2012-10-15T05:52:30 < dongs> even less hits for hex... hmm
2012-10-15T05:54:13 < dongs> VHCT seems fast as well
2012-10-15T05:55:25 < dongs> http://www.digikey.com/product-detail/en/MC74VHCT125ADR2G/MC74VHCT125ADR2GOSCT-ND/1967121
2012-10-15T05:55:29 < dongs> will go with that
2012-10-15T05:59:26 < emeb_mac> sry - afk
2012-10-15T06:00:01 < emeb_mac> yeah - hex is prolly most popular. if they want tri-state then they probably get hc/t/244
2012-10-15T06:00:14 < dongs> i dont actually need tristate or individaul enable
2012-10-15T06:00:16 < dongs> just buffers
2012-10-15T06:00:22 < dongs> there are less hex buffers on digikey.
2012-10-15T06:00:25 < emeb_mac> just 4?
2012-10-15T06:00:39 < dongs> in my last proj I needed to buffer 3 shits, so I just used a hex inverter and tied them end to end for 3 outputs
2012-10-15T06:00:53 < dongs> yea just 4
2012-10-15T06:00:58 < emeb_mac> have you ever used the little single-gate parts?
2012-10-15T06:01:12 < emeb_mac> they're cheap & easy to place-n-route.
2012-10-15T06:01:12 < dongs> they'll be used in the led driver
2012-10-15T06:01:28 < dongs> i dont think they're that cheap... quad buffer is like 12c in qty
2012-10-15T06:01:40 < emeb_mac> if you can get 'em
2012-10-15T06:01:47 < dongs> sure i can
2012-10-15T06:02:31 < dongs> is tehre some 74 series buffer without individual enables?
2012-10-15T06:02:35 < dongs> ill be just wasting t hose pins
2012-10-15T06:02:59 < dongs> looks like no..
2012-10-15T06:03:30 < emeb_mac> looking...
2012-10-15T06:03:58 < dongs> 74x210..
2012-10-15T06:04:36 < emeb_mac> 74x07?
2012-10-15T06:04:52 < dongs> hex buffer/driveR? hmm
2012-10-15T06:05:15 < dongs> open collecotr wat
2012-10-15T06:05:49 < dongs> seems more expensive
2012-10-15T06:05:51 < dongs> what the hell
2012-10-15T06:06:38 < emeb_mac> oc? oops.
2012-10-15T06:07:41 < emeb_mac> 74x367?
2012-10-15T06:08:41 < emeb_mac> meh - still tristate in 4/2 configuration.
2012-10-15T06:08:45 < dongs> yeah
2012-10-15T06:08:49 < dongs> even more pins
2012-10-15T06:08:52 < dongs> and still same shit with output enable
2012-10-15T06:12:16 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-15T06:13:42 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-15T06:13:49 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-15T06:14:03 < dongs> looks like quad it is
2012-10-15T06:15:21 < emeb_mac> boo
2012-10-15T06:16:48 < dongs> itll do
2012-10-15T06:16:52 < dongs> cheap enough for sure
2012-10-15T06:17:05 < dongs> one proj needs it for signal buffer to fet gate drivers
2012-10-15T06:17:07 < dongs> cause they need 5V
2012-10-15T06:17:15 < dongs> and another one is for 4 outputs to led drivers
2012-10-15T06:17:46 < dongs> shitty routing though, have they ever thought of making it pass-through??
2012-10-15T06:17:56 < dongs> lol @ input + output right next to each other
2012-10-15T06:18:15 < emeb_mac> http://www.digikey.com/product-detail/en/SN74LVC2G240DCTR/296-15588-1-ND/571169
2012-10-15T06:18:38 < dongs> uh yeaH?
2012-10-15T06:18:41 < dongs> i'd need 4 of htem
2012-10-15T06:20:09 < emeb_mac> small, fast, hella current.
2012-10-15T06:25:15 < dongs> not sure if that'll replace hte gate drivers.
2012-10-15T06:25:26 < dongs> not my proj so im not too worried
2012-10-15T06:25:36 < dongs> but its good if I can reuse same part for 2 projects.
2012-10-15T06:25:41 < emeb_mac> sounds like the 125 is right
2012-10-15T06:25:50 < dongs> ya
2012-10-15T06:29:21 < dongs> http://i.imgur.com/5JDoE.jpg Laurenceb is disappointed
2012-10-15T06:32:35 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-15T06:33:18 < emeb_mac> looks like muckraking to me.
2012-10-15T07:03:20 < dongs> cjbaird would be excited http://armpit.sourceforge.net/
2012-10-15T07:04:41 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-15T07:17:27 < cjbaird> Woo!
2012-10-15T07:17:46 < cjbaird> I think I've run across that before, actually.
2012-10-15T07:18:17 < cjbaird> Bare-metal LISP-esque machines. Sweet.
2012-10-15T07:23:01 < cjbaird> A number of such projects were started about 10 years ago (when the LispM story did the rounds), but hardly any got to the self-hosting stage.
2012-10-15T07:25:06 < emeb_mac> cons car cdr cadr... repeat
2012-10-15T07:28:46 < cjbaird> People had a programmerboner for the LispM mostly because the system (and other) objects were immediately accessible to the user. I've got to wonder how hairy a lisp system would be with access privledge controls on everything..
2012-10-15T07:38:55 < dongs> http://api.ning.com/files/NfnxxURv9Vtb5vQHYN-0FSvqYgxY4ScRql8P01BGJlimk6POxCinBkPdkzf0KvYxkGx6ECdRvCyQ6urv43IYTQ__/IMAG0646.jpg
2012-10-15T07:45:56 < zippe> dongle: yet another 802.11 thing?
2012-10-15T07:46:54 < dongs> ya
2012-10-15T07:47:01 < dongs> some chink shit.
2012-10-15T07:47:06 < dongs> never heard of the airoha shit.
2012-10-15T07:47:42 < dongs> weird, its abg shit only
2012-10-15T07:47:59 < dongs> does n imply all of abg or is n only n
2012-10-15T07:54:24 < zippe> n is n
2012-10-15T08:11:04 -!- claude is now known as Claude
2012-10-15T08:12:11 < upgrdman> regarding positive uv masking, when i developed the mask it only took ~20 seconds to dissolve the exposed film, and within seconds it started to dissolve the rest. is that normal, or did i expose for too long?
2012-10-15T08:14:19 < cjbaird> probably tonight's dinner :) .. http://i.imgur.com/kOB5y.jpg (Apple Pie)
2012-10-15T08:14:41 < upgrdman> om nom nom
2012-10-15T08:25:23 < dongs> emeb_mac: i should probably move one of the in/out connectors and use female headers
2012-10-15T08:25:38 < dongs> move = move further in
2012-10-15T08:25:52 < emeb_mac> dongs: f on one side, m on the other?
2012-10-15T08:25:57 < dongs> ya
2012-10-15T08:26:04 < emeb_mac> good idea
2012-10-15T08:26:10 < emeb_mac> easier chaining.
2012-10-15T08:26:28 < dongs> and that got me thinking I should bring 5V signal there too....
2012-10-15T08:26:33 < dongs> haha.
2012-10-15T08:26:54 < emeb_mac> wait, what? don't you already have that?
2012-10-15T08:27:03 < dongs> no i mean into same connector
2012-10-15T08:27:07 < dongs> 5V crap GND
2012-10-15T08:27:11 < emeb_mac> oh.
2012-10-15T08:27:14 < dongs> instead of 5v GND                     space             crap GND
2012-10-15T08:27:21 < emeb_mac> instead of on separate 2-pin above/
2012-10-15T08:27:31 < dongs> right
2012-10-15T08:27:42 < emeb_mac> yeah - why not?
2012-10-15T08:32:03 < dongs> then i'd need 5pin header :(
2012-10-15T08:32:05 < dongs> uncommon shit
2012-10-15T08:32:09 < dongs> make 2 pins for 5V?
2012-10-15T08:33:29 < cjbaird> I'll have to make sure no teenagers sneak in and put their dick in the apple pie while its cooling down..
2012-10-15T08:34:22 < dongs> sounds ilek a common activity at your house
2012-10-15T08:37:15 < emeb_mac> dongs: or 2 pins for GND. 1 between 5V and SPI, one on end: Gnd, spi, spi, spi, Gnd, 5V
2012-10-15T08:37:36 < dongs> ok, extra GND sounds better.
2012-10-15T08:39:24 < emeb_mac> hmm - that pinout is compatible w/ Digilent PMOD interface.
2012-10-15T08:39:59 < emeb_mac> http://digilentinc.com/Products/Catalog.cfm?NavPath=2,401&Cat=9
2012-10-15T08:40:16 < dongs> 5v/gnd/spix3/gnd?
2012-10-15T08:40:22 < emeb_mac> yup
2012-10-15T08:40:57 < emeb_mac> they have 5v/gnd/4xgpio
2012-10-15T08:41:08 < emeb_mac> so last gnd is gpio - still works.
2012-10-15T08:58:58 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-15T09:34:14 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 240 seconds]
2012-10-15T09:36:08 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-15T09:44:36 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 248 seconds]
2012-10-15T09:56:17 -!- dekar_ [~dekar@212.255.35.17] has joined ##stm32
2012-10-15T09:56:20 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-15T09:57:03 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-15T09:57:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-15T09:58:40 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Client Quit]
2012-10-15T09:59:43 -!- dekar [~dekar@212.255.38.46] has quit [Ping timeout: 245 seconds]
2012-10-15T10:04:11 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-15T10:06:06 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has quit [Quit: Lost terminal]
2012-10-15T10:09:01 -!- ABL [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-15T10:14:21 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Remote host closed the connection]
2012-10-15T10:25:00 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-15T10:27:24 -!- Claude is now known as claude
2012-10-15T10:37:05 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-15T10:51:55 < zyp> hey, dongs
2012-10-15T10:51:56 < zyp> http://stop-irc-bullying.eu/stop
2012-10-15T10:52:01 < dongs> jesus fucking odl
2012-10-15T10:52:02 < dongs> old
2012-10-15T10:52:08 < dongs> saw + lold liek a week ago
2012-10-15T10:52:32 < zyp> hmm, I should have figured :p
2012-10-15T10:54:15 < dongs> argh i keep finding small shit i wanna fix in teh ledriver
2012-10-15T10:59:40 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-15T11:07:27 -!- claude is now known as claude_work
2012-10-15T11:26:55 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-15T11:39:40 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-15T12:14:18 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-15T12:14:21 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-15T12:21:41 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-15T12:26:16 -!- Laurenceb is now known as _FELIX_
2012-10-15T12:27:02 < _FELIX_> WIND FROM THE EAST?
2012-10-15T12:27:27 < dongs> o hai
2012-10-15T12:27:36 < dongs> _FELIX_: gimme schematic for your stmlogger shiz
2012-10-15T12:27:45 < dongs> NOT IN EAGLE FORMAT PLOX
2012-10-15T12:27:52 < _FELIX_> IS THE FIND TOWARDS THE RIDGE?
2012-10-15T12:27:58 < _FELIX_> *WIND
2012-10-15T12:28:03 < dongs> MY RIDGE OR UR RIDGE
2012-10-15T12:28:08 < _FELIX_> I DUNNO
2012-10-15T12:28:11 < _FELIX_> WHATS THE FREQUENCY
2012-10-15T12:28:17 < _FELIX_> MY VISOR IS MISTING UP
2012-10-15T12:28:27 < dongs> THERE"S THESE THINGS CALLED BALLAST
2012-10-15T12:28:29 < dongs> THEY'RE LIEK HEAVY.
2012-10-15T12:28:48 < dongs> _FELIX_: admit it, you feared the camera jobs though.
2012-10-15T12:28:49 <+Steffanx> laucenceb .. you are not Felix
2012-10-15T12:28:54 <+Steffanx> The mission didn't fail :P
2012-10-15T12:28:57 -!- _FELIX_ is now known as Laurenceb
2012-10-15T12:29:33 < Laurenceb> dongs: ok just a sec
2012-10-15T12:29:39 < dongs> yay thanks
2012-10-15T12:31:48 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-15T12:33:35 <+Steffanx> On the other hand, it could be Laurenceb .. the mission took 5 years of preparation (aka messing around) :P
2012-10-15T12:33:56 < Laurenceb> wtf imgur died
2012-10-15T12:34:01 < dongs> watr
2012-10-15T12:34:36 < Laurenceb> ah back
2012-10-15T12:34:37 < Laurenceb> http://i.imgur.com/MGHVZ.png
2012-10-15T12:34:42 <+Steffanx> Finally, a nice good thing on this borign day
2012-10-15T12:36:01 <+Steffanx> Just curious, why you use those weird horizontal busses Laurenceb ?
2012-10-15T12:36:08 <+Steffanx> Why horizontal?
2012-10-15T12:36:15 < Laurenceb> shrug
2012-10-15T12:36:41 < Laurenceb> random chance :P
2012-10-15T12:37:23 < zyp> http://techcrunch.com/2012/10/15/raspberry-pi-mini-computer-now-shipping-with-ram-upgraded-to-512mb-from-256mb-same-tiny-35-price-tag/
2012-10-15T12:38:07 <+Steffanx> I guess i'm the only one who changes the grid so the labels align better with the wires
2012-10-15T12:38:19 < dongs> Laurenceb: thanks
2012-10-15T12:38:23 <+Steffanx> 512.. not bad
2012-10-15T12:39:40 <+Steffanx> More RAM for the GPU :P
2012-10-15T12:39:49 < Laurenceb> most of that is random stuff like pressure sensor and LEd drivers
2012-10-15T12:40:23 < Laurenceb> also firmware update over bluetooth doesnt work with that schematic :(
2012-10-15T12:40:40 <+Steffanx> I guess your life is completely random...
2012-10-15T12:40:44 < Laurenceb> it needs a resistor off one of the gpios to the NPWREN
2012-10-15T12:41:14 < zyp> that schematic looks as crowded as your code :p
2012-10-15T12:41:54 <+Steffanx> It does have some more space
2012-10-15T12:42:08 <+Steffanx> Laurenceb has no license for EAGLE?
2012-10-15T12:42:17 < zyp> http://bin.jvnv.net/f/EF6TY.png <- I feel that this is still more readable :p
2012-10-15T12:42:32 <+Steffanx> For sure
2012-10-15T12:42:44 <+Steffanx> less = more
2012-10-15T12:43:19 < dongs> zyp: how the fuck did you generate tht schema of yours
2012-10-15T12:44:00 < dongs> the opamp part in Laurenceb schematic is hilarious
2012-10-15T12:44:03 <+Steffanx> Just attach the components move and rotate the groups :)
2012-10-15T12:44:07 < dongs> its just as clusterfucked as his code
2012-10-15T12:44:12 <+Steffanx> *attach pins togother
2012-10-15T12:44:16 < zyp> pinswapping when routing the pcb
2012-10-15T12:44:29 <+Steffanx> oh, that's even more complicated
2012-10-15T12:44:43 < zyp> I cleaned up the sheet after I was done pinswapping, pic is from before cleanup
2012-10-15T12:46:16 < zyp> I don't seem to have a pic of that sheet after cleanup, but I have one of another sheet: http://bin.jvnv.net/f/Kgy4D.png
2012-10-15T12:59:11 < Thorn> interesting statistical analysis of adc data here http://mbed.org/forum/bugs-suggestions/topic/1514/?page=1
2012-10-15T12:59:23 < Thorn> how does thaat colorbar diagram work?
2012-10-15T13:20:47 < Posterdati> hi
2012-10-15T13:21:10 < Posterdati> is there a source driver for stm32f107 + ste101p ? tx
2012-10-15T13:23:11 < Thorn> Posterdati: see example code here https://www.olimex.com/Products/ARM/ST/STM32-P107/
2012-10-15T13:23:38 < Posterdati> tx
2012-10-15T13:24:43 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 245 seconds]
2012-10-15T13:30:32 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-15T13:32:21 < Laurenceb> http://www.bbc.co.uk/news/uk-england-beds-bucks-herts-19940942
2012-10-15T13:32:24 < Laurenceb> lol genius
2012-10-15T13:33:24 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-15T13:41:09 < dongs> ok i think im done fucking with the layout for the 100th time http://i.imgur.com/Ic49j.png
2012-10-15T13:42:24 < zyp> nice
2012-10-15T13:43:33 < Laurenceb> omg blickenlights
2012-10-15T13:49:12 < dongs> ok printed paper version matches
2012-10-15T14:03:45 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-15T14:07:04 -!- claude_work is now known as claude
2012-10-15T14:09:03 < dongs> http://bcas.tv/paste/results/ayc2tT79.html
2012-10-15T14:09:06 < dongs> the story of Laurenceb
2012-10-15T14:09:16 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-15T14:09:58 < Laurenceb> but...
2012-10-15T14:10:08 < Laurenceb> so the feminazis are wrong
2012-10-15T14:10:25 < dongs> http://webcache.googleusercontent.com/search?q=cache:http://brainz.org/10-psychological-experiments-went-horribly-wrong/&hl=en&prmd=imvns&strip=1
2012-10-15T14:10:28 < dongs> original site dead
2012-10-15T14:10:31 < dongs> but its full of hilarity
2012-10-15T14:11:15 < Laurenceb> too bored to read
2012-10-15T14:11:24 < dongs> just start fapping and reading from 1.
2012-10-15T14:11:53 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection]
2012-10-15T14:13:17 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-15T14:15:18 < Thorn> lol. large flash isn't only hard to find. it's also freaking expensive for SLC
2012-10-15T14:16:14 < dongs> blame apple
2012-10-15T14:16:23 < dongs> time to head out to sauna
2012-10-15T14:16:42 < zyp> why SLC?
2012-10-15T14:16:49 < jpa-> dongs: how hot is it in sauna there?
2012-10-15T14:17:02 < dongs> jpa, depending on day of the week, either 45 or 95
2012-10-15T14:17:19 < dongs> hoping its 95 today. 45 was fucking trash.
2012-10-15T14:17:45 < jpa-> yeah, 45 C is no sauna at all
2012-10-15T14:17:47 < Thorn> case in point: MT29F32G08CBACAWP is $8.4 for 1, MT29F32G08AFABAWP is $41.5 for 1000
2012-10-15T14:17:55 < Thorn> same 32Gbit capacity
2012-10-15T14:18:13 < dongs> MT29F32G08AFABAWP seems quite cheap
2012-10-15T14:18:15 < dongs> just buy 1000
2012-10-15T14:18:17 < dongs> figure out what to do with it later.
2012-10-15T14:18:34 < zyp> I had a finnish friend with me last time I went to japan, she complained that the saunas were too cold
2012-10-15T14:19:02 < Thorn> do I want SLC or not?
2012-10-15T14:20:29 < zyp> so, SLC is faster and more durable? is MLC good enough for your application?
2012-10-15T14:22:27 < Thorn> how do I know.
2012-10-15T14:22:37 < Thorn> I read MLC requires deeper ECC
2012-10-15T14:22:52 < Thorn> need to check if fsmc supports that
2012-10-15T14:23:24 -!- BusError_ [~michel@host86-145-72-200.range86-145.btcentralplus.com] has joined ##stm32
2012-10-15T14:24:44 < BusError_> anyone can tell me whats the minimal pinout needed to program a stm32 ? does the 4 pin VCC/SWIM/GND/RSET is enough to flash/debug it?
2012-10-15T14:25:04 < zyp> swim sounds like stm8
2012-10-15T14:25:23 < zyp> minimum required for debugging stm32 is swdio, swdclk and gnd
2012-10-15T14:25:26 < BusError_> looked too good to be true. been googling and getting confusing info
2012-10-15T14:31:40 < Laurenceb> rage
2012-10-15T14:31:50  * Laurenceb has a jammed $200 pump
2012-10-15T14:36:32 < Posterdati> is there anyone that make lwip works on stm32-p107? tx
2012-10-15T14:37:03 < zyp> what are you having problems with?
2012-10-15T14:49:03 < jpa-> BusError_: you can program & debug with SWDIO, GND, SWDCLK; with the built-in bootloader, you can program also if you have USART1 and BOOT0 pin is high
2012-10-15T14:50:21 < BusError_> jpa-, very cool thanks!
2012-10-15T14:50:58 -!- dekar_ [~dekar@212.255.35.17] has quit [Quit: This computer has gone to sleep]
2012-10-15T14:51:54 < Thorn> #!*^%* diptrace
2012-10-15T14:52:50 < karlp> what's SLC vs MLC?
2012-10-15T14:53:08 < karlp> also, that part of thorns wsa $41 each in 1000 qty, not 41 for 1000
2012-10-15T14:54:35 < zyp> MLC is using multiple levels to store several bits in each cell
2012-10-15T14:54:37 < Thorn> sure, that was my point
2012-10-15T14:54:43 < zyp> i.e. not binary storage like SLC
2012-10-15T14:55:20 < karlp> Thorn: yeah, but dongs's reply was "buy 1000, work it out later"
2012-10-15T14:55:37 < karlp> with a comment that it seemed quite cheap :)
2012-10-15T14:55:39 < zyp> more bits per cell means fewer cells for the same amount of bits, i.e. cheaper
2012-10-15T14:56:04 < zyp> and I also read it as 41 for 1000 first
2012-10-15T14:56:19 < Thorn> MLC has circuitry to measure level of chqarge of flash transistors and to charge them to different levels
2012-10-15T14:56:49 < Thorn> $41 for 1000 4gig flash chips = instant rich lol
2012-10-15T14:57:04 < Thorn> stm32f4 pinout is officially retarded
2012-10-15T14:57:18 < Thorn> RMII pins are on 3 sides of the chip. FSMC pins, same
2012-10-15T14:57:46 < Thorn> routing is going to be fun
2012-10-15T14:58:30 < Thorn> it doesn't help that I absolutely can't rotate the phy so that most paths don't cross
2012-10-15T14:58:57 < Posterdati> zyp: hi
2012-10-15T14:59:05 < scrts> Thorn: MII too?
2012-10-15T14:59:18 < zyp> RMII pins are a subset of MII, so yeah
2012-10-15T14:59:19 < Posterdati> zyp: https://my.st.com/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Flat.aspx?RootFolder=/public/STe2ecommunities/mcu/Lists/cortex_mx_stm32/Stm32F2%20%20Ethernet%20start%20up%20infinite%20Loop&FolderCTID=0x01200200770978C69A1141439FE559EB459D7580009C4E14902C3CDE46A77F0FFD06506F5B&currentviews=355
2012-10-15T14:59:26 < scrts> also some pins have switchable configurations, check if You can switch another pins to do the same
2012-10-15T14:59:28 < Thorn> MII = lots of conflicts, can't use it at all in my case
2012-10-15T14:59:32 < Posterdati> zyp: I can't configure the ethernet at all
2012-10-15T14:59:40 < scrts> what conflicts?
2012-10-15T14:59:46 < scrts> reference design?
2012-10-15T15:00:20 < Thorn> pin conflicts with other peripherals
2012-10-15T15:00:22 < zyp> Thorn just wants to use a lot of peripherals on a low pin-count package
2012-10-15T15:00:48 < Thorn> I want to use *several* peripherals on a *pretty large* package
2012-10-15T15:01:24 < zyp> 100-pin is not pretty large when you want to use several high speed parallel interfaces
2012-10-15T15:04:46 < Thorn> I'll need a bag of vias for this.
2012-10-15T15:07:51 < Thorn> also one of I2S peripherals (I2S3) conflicts with jtag/swd. nice
2012-10-15T15:09:24 < Laurenceb> http://s.funnymama.com/store/120511/84807_v0_600x.jpg
2012-10-15T15:11:51 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-15T15:14:22 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-10-15T15:17:13 < Thorn> fsmc routes are going to cross rmii routes a lot.
2012-10-15T15:27:03 < jpa-> tqfp144 is nothing to fear
2012-10-15T15:30:28 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-10-15T15:32:33 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-15T15:49:24 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Quit: Rebooting...]
2012-10-15T15:50:52 < Laurenceb> http://www.nokilli.com/sacto/karen-greenlee.htm
2012-10-15T15:53:33 < dongs> That's the reason I haven't tried anything lately. I'm sure I'd have found a way to get into one of those funeral homes by now, but the group I find attractive-- young men in their twenties-- are the ones who are dying of AIDS.
2012-10-15T15:53:38 < dongs> LOL
2012-10-15T15:53:51 < Laurenceb> When you're on top of a body it tends to purge blood out of its mouth, while you're making passionate love .. You'd have to be there, I guess.
2012-10-15T15:54:28 < dongs> exciting
2012-10-15T15:55:48 < dongs> Laurenceb: why do you have ressitors on all spi lines
2012-10-15T15:56:12 < Laurenceb> to reduce risk of frying stuff with piss poor code
2012-10-15T15:56:17 < dongs> haha
2012-10-15T15:56:24 < dongs> sounds like im removing them then
2012-10-15T15:57:19 < dongs> you dont do card insertion detection? heh
2012-10-15T15:57:25 < dongs> and uSd is poweerd from same reg?
2012-10-15T15:57:28 < dongs> fail
2012-10-15T15:57:43 < Laurenceb> shrug
2012-10-15T15:57:53 < Laurenceb> its permanently installed
2012-10-15T15:58:03 < dongs> mkay
2012-10-15T15:59:27 < Thorn> what's the problem with uSD being powered from the same reg?
2012-10-15T15:59:39 < dongs> well, it draws some current :)
2012-10-15T16:00:00 < dongs> my regs are only good to like 200mA
2012-10-15T16:00:03 < dongs> i dont wanna overlaod htem
2012-10-15T16:00:06 < dongs> so ill do separate.
2012-10-15T16:00:27 < Thorn> ah
2012-10-15T16:01:58 < Laurenceb> its actually on a different reg on ym board
2012-10-15T16:02:04 < Laurenceb> same reg as the nluetooth
2012-10-15T16:02:14 < dongs> sorry i couldnt tell
2012-10-15T16:02:22 < dongs> in the fucking mess of shit
2012-10-15T16:02:24 < dongs> :p
2012-10-15T16:04:29 < dongs> however I do remember needing pullups on CMD and DAT0 lines...
2012-10-15T16:04:44 < dongs> no?
2012-10-15T16:06:03 < Laurenceb> yes
2012-10-15T16:06:35 < Laurenceb> R22 and stuff
2012-10-15T16:08:57 < dongs> DO doesnt have pull up
2012-10-15T16:15:47 < Laurenceb> oh
2012-10-15T16:15:56 < Laurenceb> i use pullup on the smt
2012-10-15T16:15:59 < Laurenceb> *stm
2012-10-15T16:16:08 < Laurenceb> before turning on the power to the card
2012-10-15T16:17:04 < dongs> oic
2012-10-15T16:17:19 < dongs> thats enough?
2012-10-15T16:17:44 < dongs> where the hell is PIN1 connected to
2012-10-15T16:17:49 < dongs> i see R17 and??
2012-10-15T16:18:21 < dongs> er R16
2012-10-15T16:19:55 < Laurenceb> http://tech.slashdot.org/comments.pl?sid=3186059&cid=41656349
2012-10-15T16:19:57 < Laurenceb> i lolled
2012-10-15T16:20:40 < Laurenceb> oops missing label
2012-10-15T16:20:43 < Laurenceb> VDD
2012-10-15T16:20:46 < Laurenceb> no
2012-10-15T16:20:50 < Laurenceb> oh who cares
2012-10-15T16:21:02 < dongs> haha
2012-10-15T16:21:04 < dongs> its not connected naywehre?
2012-10-15T16:21:11 < Laurenceb> yeah VDD
2012-10-15T16:21:17 < Laurenceb> just checked in eagle
2012-10-15T16:21:34 < dongs> o ok
2012-10-15T16:21:42 < dongs> right so all reserved shit is pulled up
2012-10-15T16:21:48 < dongs> and other shit is internal pullup by stm
2012-10-15T16:21:49 < Laurenceb> R17 is 10K iirc
2012-10-15T16:22:00 < dongs> oh i know i wasnt ognna use anything else
2012-10-15T16:22:05 < Laurenceb> C32 is 2.2uF
2012-10-15T16:22:10 < Laurenceb> thats for power on/off
2012-10-15T16:24:34 < Laurenceb> http://anti-joke.com/anti-joke/popular/page/1
2012-10-15T16:27:14 < Laurenceb> "Knock Knock Who's there? The police, your entire family died in a car accident "
2012-10-15T16:28:09 < dongs> i dont even know where C32 iws
2012-10-15T16:28:56 < Laurenceb> at the top
2012-10-15T16:29:27 < dongs> oh.
2012-10-15T16:29:33 < dongs> i dont think i have use for that.
2012-10-15T16:29:39 < dongs> this hsit isnt battery powered
2012-10-15T16:50:07 -!- dekar [~dekar@studpool-wlan-74-65.fs.fbi.h-da.de] has joined ##stm32
2012-10-15T16:50:10 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-15T16:56:50 < Laurenceb> http://t.qkme.me/3cvr.jpg
2012-10-15T17:00:38 < Thorn> >meteor strikes
2012-10-15T17:02:01 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-15T17:02:34 < Laurenceb> http://static.quickmeme.com/media/social/qm.gif
2012-10-15T17:05:30 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-15T17:07:14 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 255 seconds]
2012-10-15T17:08:18 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-15T17:14:38 < emeb_mac> haha 1pixel
2012-10-15T17:15:52 <+Steffanx> w00t
2012-10-15T17:18:04 < dongs> emeb_mac: i think im finally done dicking with the layout http://i.imgur.com/Ic49j.png
2012-10-15T17:18:13 < dongs> any final complaints?
2012-10-15T17:18:37 < dongs> in: female, out: male. totally Laurenceb style
2012-10-15T17:19:12 < Laurenceb> dick and ass style
2012-10-15T17:19:30 < dongs> xactly
2012-10-15T17:19:59 < emeb_mac> wouldn't  have been my choice for metaphors, but to each his own.
2012-10-15T17:20:27 < Laurenceb> mantrain pcb
2012-10-15T17:20:36 < zyp> dongs, are you going to populate all boards with all connectors?
2012-10-15T17:20:37 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has joined ##stm32
2012-10-15T17:20:40 < emeb_mac> dongs: looks good.
2012-10-15T17:20:51 < dongs> zyp: the power, yes
2012-10-15T17:20:59 < dongs> i dunno if I wanna bother with usb socket on the next next ones.
2012-10-15T17:21:12 < dongs> I measured and the right angle header should fit flush next to each other
2012-10-15T17:21:12 < zyp> usb is what I was thinking of
2012-10-15T17:21:25 < dongs> USB should work though, it will overlay a bit to ne xt board
2012-10-15T17:21:28 < dongs> but wont prevent it from connector
2012-10-15T17:21:31 < dongs> err connecting
2012-10-15T17:22:17 < zyp> I'm just thinking that you might not want the chain connector sticking out on the last board in the chain
2012-10-15T17:23:04 < zyp> so if you are going to assemble multiple variants, the sanest way would be swapping the in/out connectors and then populating either usb or data in
2012-10-15T17:23:37 < dongs> you mean style? female for out?
2012-10-15T17:23:50 < zyp> yes, because it doesn't stick out
2012-10-15T17:24:22 < emeb_mac> could also just move the pads right on both.
2012-10-15T17:25:01 < emeb_mac> make female stick out, male pin tips stop at board edge.
2012-10-15T17:25:17 < dongs> well at that point i mgiht as well just swap connector styles though
2012-10-15T17:25:21 < zyp> emeb_mac, well, layout-wise male/female doesn't matter
2012-10-15T17:25:24 < dongs> it should fit, i can move the clock buffer up
2012-10-15T17:27:41 < dongs> http://i.imgur.com/qL98m.png should work
2012-10-15T17:28:38 < jpa-> a lot of chips just to drive a few leds :o
2012-10-15T17:29:02 <+Steffanx> :)
2012-10-15T17:29:08 < zyp> just 192 leds
2012-10-15T17:29:33 < jpa-> but they are multiplexed
2012-10-15T17:30:07 < jpa-> but yeah, probably a fine design; chip's don't cost much
2012-10-15T17:30:12 < jpa-> -'
2012-10-15T17:32:09 < emeb_mac> dongs: any chance you could bring a few unused MCU pins to thru-hole testpoints?
2012-10-15T17:32:33 < emeb_mac> thinking it would be interesting to drive analog inputs to the ADC and make a voltage controlled LED array.
2012-10-15T17:32:40 < dongs> there isnt that much unused, hmm
2012-10-15T17:32:44 < dongs> do you still have the schema?
2012-10-15T17:32:54 < emeb_mac> don't have a current one.
2012-10-15T17:33:55 < emeb_mac> looks like pins 10-14 are unused tho - those are analog inputs.
2012-10-15T17:34:15 < emeb_mac> with clear routes out to an empty area.
2012-10-15T17:34:43 < emeb_mac> 14 may be a power pin, but 10-13 are PAx
2012-10-15T17:36:26 < dongs> http://i.imgur.com/VYzaD.png
2012-10-15T17:37:21 < dongs> yeah 10,11,12 looks clear to use
2012-10-15T17:37:32 < Laurenceb> does everything use p substrate cmos these days?
2012-10-15T17:37:47 < Laurenceb> where everything means stm32
2012-10-15T17:38:47 < emeb_mac> dongs: yup. just route 10-12 out to pads in case. could do some crazy xyz effects w/ analog signals.
2012-10-15T17:42:59 < emeb_mac> afk - morning bike ride.
2012-10-15T17:43:13 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-15T17:50:07 < Posterdati> please help, I've got PSP stack corruption after few context switching, the code for my RTOS is at http://pastebin.com/YJ85DFyJ
2012-10-15T17:50:10 < Posterdati> thanks
2012-10-15T17:54:26 -!- Tectu is now known as tectu
2012-10-15T17:54:28 < jpa-> Posterdati: step it through in a debugger, check manually after each step, see where it goes wrong
2012-10-15T17:54:49 < jpa-> it takes time but it is very unlikely anyone could spot the mistake from the code
2012-10-15T17:57:26 < Posterdati> jpa-: I did it
2012-10-15T17:57:45 -!- tectu is now known as Tectu
2012-10-15T17:57:56 < Posterdati> jpa-: but I cannot understand why the stack is wrong
2012-10-15T17:58:24 < Posterdati> jpa-: LOAD_CONTEXT restore a wrong PSP
2012-10-15T17:58:43 < Posterdati> jpa-: so PC will contain a wrong return address
2012-10-15T17:59:05 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-15T18:00:03 < jpa-> so you have a correct PSP, wrong PSP and you see the instruction which restores it? doesn't sound too difficult to work out where the wrong value comes from
2012-10-15T18:00:45 < Posterdati> yes
2012-10-15T18:01:26 < Posterdati> jpa-: the problem is in psp loading via LOAD_CONTEXT
2012-10-15T18:16:27 < dongs> ok rotated the connectors around.
2012-10-15T18:21:27 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-15T18:22:38 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-15T18:25:21 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-15T18:26:27 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has quit [Ping timeout: 240 seconds]
2012-10-15T18:28:27 -!- dekar [~dekar@studpool-wlan-74-65.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-15T18:37:26 < dongs> ha
2012-10-15T18:37:37 < dongs> some spartan3 breakout got submitted for assembly
2012-10-15T18:37:50 < dongs> eagle though :(
2012-10-15T18:41:20 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has joined ##stm32
2012-10-15T18:42:15 < Laurenceb> on your pcb service?
2012-10-15T18:42:19 < dongs> ya
2012-10-15T18:43:03 < Laurenceb> i should try that
2012-10-15T18:43:07 < Laurenceb> how much does it cost?
2012-10-15T18:43:22 < dongs> you probably cant afford it :( its for rich japs only
2012-10-15T18:43:30 < Laurenceb> boo
2012-10-15T18:43:43 < Posterdati> jpa-: maybe I'm short of stack space: 1024K
2012-10-15T18:44:16 < Laurenceb> wohoo
2012-10-15T18:44:20 < Laurenceb> fixed my pump
2012-10-15T18:44:32 < Laurenceb> had injested something
2012-10-15T18:45:04 < Laurenceb> looks like dried spunk or something.. maybe its too close to my pc
2012-10-15T18:46:10 < Laurenceb> http://abusemark.com/store/images/stm32_dev.jpg
2012-10-15T18:46:12 < Laurenceb> lolling
2012-10-15T18:46:17 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-15T18:47:20 < dongs> http://i.imgur.com/87dac.png moved shit around.
2012-10-15T18:47:29 < dongs> no moar changes
2012-10-15T18:47:36 < zyp> Laurenceb, seen that before
2012-10-15T18:47:41 < emeb> w00t!
2012-10-15T18:47:43 < zyp> maybe I also need to take pics like that
2012-10-15T18:47:48 < dongs> zyp: lo
2012-10-15T18:48:13 < zyp> I think I got a figure in a box somewhere that I can dig up
2012-10-15T18:49:29 < jpa-> Posterdati: 1024K is a bit much ,)
2012-10-15T18:53:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-15T18:55:59 -!- barthess [~barthess@5.100.202.124] has joined ##stm32
2012-10-15T18:57:46 < zyp> http://bin.jvnv.net/f/xlA5N.JPG
2012-10-15T18:57:51 < emeb> lol -> http://www.slashgear.com/amazon-weighing-ti-mobile-chip-business-grab-tip-insiders-15251828/
2012-10-15T18:58:28 < emeb> zyp: what's the rectangular black/yellow thing on the top of the board?
2012-10-15T18:58:57 < zyp> gps with antenna
2012-10-15T18:59:18 < dongs> zyp: laolling
2012-10-15T18:59:24 < emeb> interesting - never seen that formfactor before.
2012-10-15T19:02:45 < zyp> http://www.antenova.com/Product%20Specs/M10382_11MD-0043-3-PS.pdf
2012-10-15T19:03:03 < qyx_> zyp: what's that
2012-10-15T19:04:58 < emeb> wonder why they put the antenna match off-module?
2012-10-15T19:05:07 < Laurenceb> i spot lsm303 issues and lack of antenova impedance match
2012-10-15T19:05:43 < zyp> emeb, so it can be picked and placed along with the rest of the components
2012-10-15T19:06:00 < zyp> antenna match is environment specific, depends on stuff like ground planes
2012-10-15T19:06:08 < zyp> and enclosure
2012-10-15T19:06:18 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-15T19:06:20 < emeb> right
2012-10-15T19:06:20 < zyp> Laurenceb, yep
2012-10-15T19:07:25 < zyp> that board is the spare one without lsm303
2012-10-15T19:07:58 < zyp> the fully working one currently has wires soldered to it :p
2012-10-15T19:08:54 < jpa-> issues? with lsm303? no wai
2012-10-15T19:09:11 < zyp> more like issues with dumb footprint for the qfn
2012-10-15T19:09:28 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-10-15T19:09:38 < jpa-> i wish the hardest part was getting it soldered
2012-10-15T19:10:04 < jpa-> heck, it freezes the I2C bus when it gets an interrupt.. so if you set threshold wrong you can no longer disable the interrupt :)
2012-10-15T19:10:19 < zyp> oh
2012-10-15T19:10:27 < qyx_> i was using lsm303dlhc on my imu and it worked on first try
2012-10-15T19:10:31 < qyx_> wioth no issues
2012-10-15T19:10:44 < qyx_> but i didn't use interrupts
2012-10-15T19:10:46 < zyp> I forgot to hook up the interrupt lines when routing the board :p
2012-10-15T19:11:07 < dongs> http://upload.wikimedia.org/wikipedia/commons/7/7e/Big_Bang_LG.jpg spot laurenceb
2012-10-15T19:11:33 < jpa-> qyx_: did you have any issues with magnetometer calibration?
2012-10-15T19:11:37 < zyp> I was thinking «I'll hook them up to some leftover pins later», and then I completely forgot about it
2012-10-15T19:11:45 < jpa-> the bias on mine seemed to change around a lot
2012-10-15T19:11:46 < Laurenceb> ive got one working nicely
2012-10-15T19:11:46 < zyp> I even forgot to hook up ad0
2012-10-15T19:11:50 < Laurenceb> eith interrupts
2012-10-15T19:11:56 < Laurenceb> - i2c interrupt
2012-10-15T19:12:06 < Laurenceb> but you _have_ to calibrate them
2012-10-15T19:12:19 < zyp> also, this is lsm303dlm
2012-10-15T19:12:39 < jpa-> of course you have to calibrate.. i just wish the calibration remained valid for more than a few hours
2012-10-15T19:13:09 < zyp> seems to work fine once I got it soldered correctly though
2012-10-15T19:13:52 < qyx_> jpa-: there were so much metal parts around that i was unable to get some useful data
2012-10-15T19:13:58 < qyx_> jpa-: old industry building :)
2012-10-15T19:14:22 < jpa-> yep, i had that also
2012-10-15T19:14:32 < jpa-> it really doesn't help that magnetic inclination is like 70° around here
2012-10-15T19:14:38 < Laurenceb> heh
2012-10-15T19:14:49 < qyx_> here it is ~3°
2012-10-15T19:14:56 < jpa-> but in the end i got it working somewhat.. some silly function approximating the biases in runtime
2012-10-15T19:15:06 < jpa-> lol, you have it easy
2012-10-15T19:15:29 <+Steffanx> You should build your gadgets remotely jpa-
2012-10-15T19:15:33 < jpa-> qyx_: inclination, not declination, btw.
2012-10-15T19:15:37 < qyx_> hm, ah
2012-10-15T19:15:41 < qyx_> wait
2012-10-15T19:15:44 < zyp> doesn't lsm303 just have a hmc die bonded on top of the st accel die?
2012-10-15T19:15:50 < qyx_> i had some nice tables for that
2012-10-15T19:16:04 < jpa-> zyp: it seems so
2012-10-15T19:16:21 < jpa-> qyx_: http://en.wikipedia.org/w/index.php?title=File:World_Magnetic_Inclination_2010.pdf&page=1
2012-10-15T19:16:27 < zyp> so it's pretty much two seperate chips in one package
2012-10-15T19:17:13 < zyp> I did some calibration on a board with a seperate hmc chip once
2012-10-15T19:17:13 < qyx_> so inclination is around 65
2012-10-15T19:17:29 < zyp> http://bin.jvnv.net/f/unBCL.png
2012-10-15T19:17:53 -!- claude is now known as Claude
2012-10-15T19:18:10 < Laurenceb> qyx_: 3degrees? where are you?
2012-10-15T19:18:23  * Laurenceb uses the worldmagneticmodel code from openpilot
2012-10-15T19:18:29 < Laurenceb> using gps to initialise i
2012-10-15T19:18:31 < qyx_> wait what
2012-10-15T19:18:46 < qyx_> i doesnt understand that
2012-10-15T19:18:48 < qyx_> Laurenceb: SK
2012-10-15T19:18:56 < qyx_> or sr, slovakia
2012-10-15T19:19:05 < Laurenceb> hmm
2012-10-15T19:19:08 < qyx_> but last time i got some ~3° values
2012-10-15T19:19:11 < Laurenceb> its not 3 degrees
2012-10-15T19:19:14 < Laurenceb> oh
2012-10-15T19:19:19 < jpa-> http://kapsi.fi/~jpa/stuff/pix/magnetometer2.png
2012-10-15T19:19:21 < Laurenceb> you are getting confused
2012-10-15T19:19:23 < qyx_> it should be more
2012-10-15T19:19:41 < qyx_> ok, maybe i failed totally
2012-10-15T19:19:45 < Laurenceb> jpa-: looks about right
2012-10-15T19:20:54 < jpa-> yeah
2012-10-15T19:21:04 < jpa-> this is from the RX; motor and servo cause quite a bias: http://kapsi.fi/~jpa/stuff/pix/magnetometer3.png
2012-10-15T19:23:30 < Laurenceb> not suprising really
2012-10-15T19:23:58 < jpa-> not in retrospect
2012-10-15T19:24:12 < jpa-> quess who initially mounted the servo on the swinging arm
2012-10-15T19:24:30 < jpa-> guess then who after that still had the li-ion batteries on the swinging arm, with their iron shells
2012-10-15T19:25:23 -!- Claude is now known as claude
2012-10-15T19:27:24 -!- claude is now known as Claude
2012-10-15T19:28:38 -!- Claude is now known as claude
2012-10-15T19:28:42 <+Steffanx> lol claude ?
2012-10-15T19:29:14 -!- claude is now known as Claude
2012-10-15T19:30:17 < jpa-> Steffanx: he's just trying to get your attention
2012-10-15T19:30:31 <+Steffanx> There are easier ways
2012-10-15T19:31:04 < Laurenceb> what do you get if you cross a lion with a centipede?
2012-10-15T19:31:38 -!- dekar [~dekar@212.255.35.17] has joined ##stm32
2012-10-15T19:31:41 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-15T19:32:10 <+Steffanx> You Laurenceb
2012-10-15T19:32:54 < Laurenceb> a debate on the ethics of genetic engineering
2012-10-15T19:33:09 <+Steffanx> Same thing
2012-10-15T19:35:37 < dongs> haha
2012-10-15T19:39:13 -!- Claude is now known as claude
2012-10-15T19:39:48 -!- claude is now known as Claude
2012-10-15T19:42:45 -!- Claude is now known as claude
2012-10-15T19:43:45 -!- claude is now known as Claude
2012-10-15T19:49:00 -!- Claude is now known as claude
2012-10-15T19:50:54 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has quit [Ping timeout: 240 seconds]
2012-10-15T19:53:45 -!- claude is now known as Claude
2012-10-15T19:55:25 < karlp> hehe: http://imgur.com/iKt8S
2012-10-15T19:55:56 < jpa-> it only lacks arduino shield pinout
2012-10-15T19:56:16 < zyp> is that a haircomb?
2012-10-15T19:56:52 <+Steffanx> Heh, I knew what the title was before i read it :)
2012-10-15T19:58:24 < dongs> lol Laurenceb retweeted my stm32_dev in #highaltitude
2012-10-15T19:58:50 < zyp> he always does that
2012-10-15T19:58:51 -!- Claude is now known as claude
2012-10-15T20:01:16 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has joined ##stm32
2012-10-15T20:02:06 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-15T20:05:26 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 245 seconds]
2012-10-15T20:07:04 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has quit [Ping timeout: 244 seconds]
2012-10-15T20:07:46 < dongs> http://d24w6bsrhbeh9d.cloudfront.net/photo/5604170_460s.jpg
2012-10-15T20:46:19 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-15T20:46:30 < dongs> WHY SO DEAD????
2012-10-15T20:46:49 < dongs> i think ill goto sleep and layout this usb/sd/shit tomrorow
2012-10-15T20:53:56 < scrts> :)
2012-10-15T20:54:17 < zyp> *yawn*
2012-10-15T21:10:30 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 252 seconds]
2012-10-15T21:12:26 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-15T21:13:48 < Tectu> jpa-, http://www.stevebate.net/chibios-rpi/GettingStarted.html
2012-10-15T21:21:56 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-15T21:24:25 < dongs> loll
2012-10-15T21:24:28 < emeb> OK - that's cute.
2012-10-15T21:27:02 < karlp> vi _and_ emacs?
2012-10-15T21:27:08 < karlp> the dude's clearly got screws loose
2012-10-15T21:28:32 < emeb> two flavors, both bad.
2012-10-15T21:32:18 -!- tech2077 [~tech2077@207.80.127.250] has joined ##stm32
2012-10-15T21:48:16 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 252 seconds]
2012-10-15T21:49:58 -!- barthess [~barthess@5.100.202.124] has quit [Quit: Leaving.]
2012-10-15T22:04:49 -!- tech2077 [~tech2077@207.80.127.250] has quit [Ping timeout: 260 seconds]
2012-10-15T22:16:19 < Rickta59> has anyone made noise about a win32 version of TNT?
2012-10-15T22:18:41 <+Steffanx> Someone, but that someone was too lazy to mess with cygwin iirc
2012-10-15T22:19:03 -!- daku is now known as DaKu
2012-10-15T22:37:56 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has joined ##stm32
2012-10-15T22:44:11 < zyp> emeb, are you trying to start a flamewar? :p
2012-10-15T22:45:25 <+Steffanx> The question is: Do you want one?
2012-10-15T22:47:17 < emeb> zyp: heh, I guess that is dangerous territory.
2012-10-15T22:47:31 < emeb> and no. I don't want one.
2012-10-15T22:48:21  * emeb uses both vi and emacs. hates both.
2012-10-15T22:49:07 < emeb> so in that sense I'm neutral I suppose.
2012-10-15T22:52:20 < zyp> I think either seems rather good :p
2012-10-15T22:53:19 < zyp> though, the only experience I have with emacs is watching coworkers use it
2012-10-15T22:53:51 < emeb> I'll admit that dedicated emacs users can be very productive with it.
2012-10-15T22:54:10 < emeb> but it takes a while to get the key commands committed to spinal reflex.
2012-10-15T22:54:14 < BrainDamage> wait, why do you use them?
2012-10-15T22:54:21 < emeb> Me?
2012-10-15T22:54:24 < BrainDamage> yes
2012-10-15T22:54:37 < zyp> emeb, you use mac from time to time, ever used textmate?
2012-10-15T22:54:43 < emeb> I use vi for quick edits to *nix text files when a GUI isn't around.
2012-10-15T22:54:56 < emeb> I use emacs because that's the default editor for Matlab in *nix.
2012-10-15T22:55:24 < emeb> but I prefer a simple GUI text editor w/ standard ctrl+zxcv cut & paste.
2012-10-15T22:55:28 < BrainDamage> mmm editor should be changeable for matlab
2012-10-15T22:55:46 < BrainDamage> and wrt vi in console, there's other simpler console editors
2012-10-15T22:56:23 < zyp> I used to be quite fond of textmate
2012-10-15T22:56:24 < emeb> BrainDamage: but vi is the one I know. :) esc : q! is in my fingers.
2012-10-15T22:57:12 < zyp> but it's OS X only, so when I started working last year I needed another editor
2012-10-15T22:57:14 < emeb> I sometimes find myself typing vi commands in GUIs. A form of stockholm syndrome.
2012-10-15T22:57:36 < emeb> zyp: Don't think I've used textmate.
2012-10-15T22:57:46 < BrainDamage> actually some gui text editors allow vi commands trough plugins
2012-10-15T22:57:54 < emeb> I started out w/ BBEdit on mac. Textwrangler now I think.
2012-10-15T22:58:02 < emeb> But I also use scite.
2012-10-15T22:58:21 < zyp> and after using vim on a daily basis for a while, I started to miss the commands in textmate
2012-10-15T22:58:23 < BrainDamage> while I appreciate the possibility to do everything using keyb, I prefer having also the option to navigate trough menus
2012-10-15T22:58:34 < emeb> yep.
2012-10-15T22:58:59 < emeb> It's my belief that refusing to do everything via kbd has prevented me from having RSI problems.
2012-10-15T22:59:00 < BrainDamage> because for rarely used functions, it takes me less time to navigate trough a menu than check several manual pages to check a shortcut I'll quickly forget due underuse
2012-10-15T22:59:16 < zyp> then came textmate 2 with a lot of changes, so I figured that rather than getting used to textmate 2, I could just switch to macvim instead :)
2012-10-15T22:59:17 < emeb> ie - mousing may slow me down, but keeps my hands happier.
2012-10-15T22:59:49 < emeb> of course I could just be lucky and not be prone to RSI.
2012-10-15T23:00:34 < BrainDamage> wtf of the day: https://dx.com/p/car-vehicle-emergency-urine-bags-blue-4-pcs-159599
2012-10-15T23:00:45 < zyp> isn't RSI usually a result of too much mouse usage, rather than too little? :p
2012-10-15T23:01:09 < emeb> got me.
2012-10-15T23:01:35 < emeb> guys I know who got RSI were "must do everything w/ kbd cmds" types.
2012-10-15T23:02:34 < emeb> but then I'm also a notorious goof-off, so maybe I just don't spend enough of the day working to have problems. :P
2012-10-15T23:03:22 < zyp> oh well
2012-10-15T23:03:42 < zyp> personally I prefer a good mix of both keyboard and mouse usage
2012-10-15T23:03:48 < emeb> *nod*
2012-10-15T23:04:42 < zyp> that's kind of why I like OS X; it's as mouse friendly as windows while being as keyboard friendly as linux
2012-10-15T23:05:16 < emeb> agree. a nice mix. Too bad Apple is slowly killing it.
2012-10-15T23:05:21 < zyp> if I'm allowed to say it that way :p
2012-10-15T23:05:48 < BrainDamage> what is killing?
2012-10-15T23:05:53 < BrainDamage> the desktop os?
2012-10-15T23:05:58 < emeb> yeah
2012-10-15T23:06:09 < zyp> I'm not sure I agree
2012-10-15T23:06:26 < zyp> I don't feel it's becoming worse
2012-10-15T23:07:06 < BrainDamage> I didn't read that as becoming worse, I read that as receiving less attention
2012-10-15T23:07:10 < emeb> seems to me that they're gradually chopping off the stuff I like and adding layers of gui dumbness.
2012-10-15T23:07:26 < BrainDamage> ok I read it wrong
2012-10-15T23:07:31 < zyp> what's being chopped off?
2012-10-15T23:08:26 < emeb> good question. I'm still on Snow Leopard, so I'm talking out my ass.
2012-10-15T23:09:01 < emeb> but I've got friends who keep up w/ the latest and they're constantly carping about the new versions.
2012-10-15T23:09:30 < zyp> I'm using 10.8, I feel it's a nice improvement over 10.6
2012-10-15T23:09:51 < zyp> and I can't really point out anything I miss
2012-10-15T23:09:56 < emeb> text shell still works? Do they still include an X server?
2012-10-15T23:10:42 < emeb> One complaint that I see over and over is the "feature" that restarts all your running apps after a crash.
2012-10-15T23:10:45 < zyp> text shell is as good as always, X server is now a seperate download
2012-10-15T23:11:23 < zyp> and the auto-restart feature is rather nice
2012-10-15T23:11:42 < zyp> I came home from work today to find there had been a power outage
2012-10-15T23:11:55 < emeb> buddy of mine keeps dozens of apps open. When he has a crash it can take 15 min for the machine to restart.
2012-10-15T23:11:59 < zyp> just powered up computer and had most apps as I left them yesterday
2012-10-15T23:12:10 < zyp> oh
2012-10-15T23:12:32 < emeb> And since he's a software developer, often the crashed app gets restarted and crashes again -> endless loop.
2012-10-15T23:13:03 < zyp> http://bin.jvnv.net/f/c4ZbV.png <- my machine is not underpowered :)
2012-10-15T23:13:11 < zyp> (just a bit confused about what it is)
2012-10-15T23:13:27 < emeb> picture fail.
2012-10-15T23:13:33 <+Steffanx> Uh?
2012-10-15T23:13:36 < emeb> or text fail?
2012-10-15T23:13:46 <+Steffanx> both
2012-10-15T23:14:13 < zyp> translation, most likely
2012-10-15T23:14:41 < emeb> so is it a box or a laptop?
2012-10-15T23:15:02 <+Steffanx> That picture on the left is the new macbook pro.. not released yet
2012-10-15T23:15:15 <+Steffanx> It looks a little like the mac pro
2012-10-15T23:15:17 < zyp> hackintosh, configured to identify as a macpro
2012-10-15T23:15:52 < zyp> so box
2012-10-15T23:16:18 < emeb> Ah. I've heard mixed results from building hackintosh. Sounds like you were successful.
2012-10-15T23:16:30 <+Steffanx> So the GPU and CPU are correct zyp ?
2012-10-15T23:16:36 < zyp> yes
2012-10-15T23:17:10 <+Steffanx> emeb isn't it 'easy' .. use compatible ( listed somewhere ) hardware and you are done
2012-10-15T23:17:42 < Posterdati> jpa-: but still not enough when vectors and matrix calculations are involved
2012-10-15T23:17:45 < zyp> http://bin.jvnv.net/f/r4wGD.JPG
2012-10-15T23:19:49 < emeb> impressive case
2012-10-15T23:20:21 < zyp> I decided it was time to replace my old imac, so I compared the 27" imac to what it would cost to build this and get a Dell U2711
2012-10-15T23:20:37 < zyp> before I added the gpu, they were at around the same price point
2012-10-15T23:20:43 < emeb> nice big fans - probably pretty quiet.
2012-10-15T23:20:57 < zyp> yeah
2012-10-15T23:21:08 < zyp> but not virtually noiseless like my old imac was
2012-10-15T23:21:29 < zyp> and the case is neat
2012-10-15T23:21:34 < emeb> the Intel-supplied CPU fan on my I5 is the loudest thing in my box.
2012-10-15T23:21:45 < zyp> big inside, tiny on the desk
2012-10-15T23:21:48 < emeb> has an oddly metallic whine to it.
2012-10-15T23:21:58 < emeb> yeah - not a common form-factor.
2012-10-15T23:22:04 <+Steffanx> Today the mac(book pro)s aren't 'noiseless' at all anymore, so that's not a downside for using your own hw zyp
2012-10-15T23:22:12 < zyp> http://bin.jvnv.net/f/tvvVf.JPG <- fits nicely in the corner behind the display
2012-10-15T23:22:24 < zyp> and all connectors are readily available under the top cover
2012-10-15T23:22:44 < emeb> nicely organized bench.
2012-10-15T23:22:52 < zyp> Steffanx, bullshit, my MBP is also virtually noiseless unless it's running at 100%
2012-10-15T23:23:07 <+Steffanx> At least, my mbp 2010 can be VERY noisy. Now with an SSD it became a little less noisy though
2012-10-15T23:23:19 <+Steffanx> At i read the same about the newer ones
2012-10-15T23:23:22 <+Steffanx> *and
2012-10-15T23:23:55 < zyp> the only time I hear the drive in my MBP is the «clack» it makes when it parks the head due to detected motion :p
2012-10-15T23:24:13 <+Steffanx> I guess it's too quiet here
2012-10-15T23:24:20 <+Steffanx> Mine doesn't like it to run at 100% for a longer time.. so i never do that
2012-10-15T23:24:32 <+Steffanx> A cpu of > 105C isn't very nice
2012-10-15T23:24:34 -!- mrnuke|laptop [~mrnuke@129.7.255.77] has quit [Ping timeout: 240 seconds]
2012-10-15T23:24:46 <+Steffanx> *cpu temperature
2012-10-15T23:24:48 < emeb> can use it to make coffee!
2012-10-15T23:25:38 < emeb> I've got an ancient Macbook - white plastic case, core 2 duo. Fan on that guy cranks up just watching Youtube videos.
2012-10-15T23:26:00 < zyp> mine is a 2011 model
2012-10-15T23:26:04 < zyp> first with sandy bridge
2012-10-15T23:26:07 < zyp> and thunderbolt
2012-10-15T23:26:07 < emeb> 2008 I think.
2012-10-15T23:26:24 < zyp> ah
2012-10-15T23:26:37 < emeb> heh - 4 years old is ancient.
2012-10-15T23:26:48 < zyp> my mother have one of those, got it in 2009
2012-10-15T23:27:20 < emeb> plastic case is poorly engineered. Mine has a bunch of cracks.
2012-10-15T23:27:27 <+Steffanx> Mine does that too, and it's 2 years older :P
2012-10-15T23:27:37 < zyp> she's complaining about it being slow, so I'm thinking about putting an SSD and some more memory in it :p
2012-10-15T23:27:49 < emeb> My wife's 2010 version has the rubberized bottom peeling off.
2012-10-15T23:28:16 < emeb> there was a recall on those, but her's is just outside the window that they'll repair for free.
2012-10-15T23:28:29 < zyp> ouch
2012-10-15T23:28:52 <+Steffanx> Apple <3 ( not )
2012-10-15T23:29:23 < zyp> there is probably a reason they discontinued the plastic ones :p
2012-10-15T23:29:32 < emeb> yep.
2012-10-15T23:30:03 < emeb> And now the latest HW is not user serviceable. Cases glued shut.
2012-10-15T23:30:08 < zyp> the alu cases seems pretty sturdy
2012-10-15T23:30:15 < zyp> really?
2012-10-15T23:30:32 < emeb> yeah. Throw it away when it needs repair.
2012-10-15T23:31:44 < zyp> I'm not sure I believe that, but ok
2012-10-15T23:32:46 < zyp> first thing I did when I got mine was opening it up and putting in an ssd :p
2012-10-15T23:33:27 <+Steffanx> And it's the first thing that died :P
2012-10-15T23:33:45 < Posterdati> jpa-: fixed! It wasn't the code, but the stack size
2012-10-15T23:34:00 < zyp> well, the ssd weren't bought from apple, can't really blame them :p
2012-10-15T23:39:50 < emeb> re: glue - it's not that they're glued shut, but the internal batteries are glued in place.
2012-10-15T23:40:17 < emeb> plus on the latest MB Air models they've gone to proprietary connectors on the SSDs so you can't put in ordinary ones.
2012-10-15T23:42:05 < emeb> http://www.macnewsworld.com/story/75378.html
2012-10-15T23:42:05 < zyp> well, yeah
2012-10-15T23:43:04 <+Steffanx> oh fancy new feature
2012-10-15T23:43:20 < zyp> similar to msata but not compatible, if I've understood it correctly
2012-10-15T23:43:49 <+Steffanx> I bet you can buy a chinese clone/converter emeb
2012-10-15T23:44:13 < emeb> no doubt - with all the quality that goes along with such things. :)
2012-10-15T23:44:39 <+Steffanx> It'll be the cheapest part ( when you calculated it by price/components ) in the entire notebook
2012-10-15T23:44:43 <+Steffanx> :P
2012-10-15T23:44:56 < emeb> :)
2012-10-15T23:57:35 < Posterdati> hi
2012-10-15T23:58:12 <+Steffanx> lo
2012-10-15T23:59:21 < Posterdati> any LWIP expert here? I've got the LWIP running on my RTOS, eth green led is lit, yellow is blinky (sporadically), net interface config is entered, but connection is 10 MBps not 100 MBps (the board won't answer to ping)
--- Day changed Tue Oct 16 2012
2012-10-16T00:02:16 < Thorn> does it receive pings (or see any frames)?
2012-10-16T00:04:27 < Posterdati> no it doesn't
2012-10-16T00:04:38 < Posterdati> wait
2012-10-16T00:04:52 < Posterdati> if I ping it, the yellow led blinks
2012-10-16T00:05:27 < Posterdati> but I've got "Destination Host Unreachable"
2012-10-16T00:07:09 < Thorn> in order for it to respond to pings, the whole stack must work right
2012-10-16T00:07:34 < Thorn> so it's time to start diagnosing layer by layer.
2012-10-16T00:07:53 < Thorn> (do you even have code that receives ethernet frames?)
2012-10-16T00:08:32 < Posterdati> and which is?
2012-10-16T00:08:56 < Posterdati> I'm a newbie in tcp/ip stack on micros
2012-10-16T00:09:31 < Thorn> what about ethernet on micros
2012-10-16T00:09:44 < Thorn> that ought to work before tcp/ip becomes useful
2012-10-16T00:09:54 < Posterdati> ok
2012-10-16T00:10:09 < Posterdati> how can I check for receiving frames?
2012-10-16T00:12:14 < Thorn> set a breakpoint in your frame receive code. or print something to usart from there. or blink a led, etc.to see if it's even triggered
2012-10-16T00:14:13 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has joined ##stm32
2012-10-16T00:14:36 < Posterdati> Thorn: do you mean LwIP_Periodic_Handle ?
2012-10-16T00:15:34 < Thorn> hardly
2012-10-16T00:18:26 < Posterdati> LwIP_Pkt_Handle();
2012-10-16T00:20:24 < Posterdati> ETH_IRQHandler is never executed
2012-10-16T00:24:27 < Posterdati> Thorn: this is NVIC config -> http://pastebin.com/9KLfMZYV
2012-10-16T00:43:48 < Posterdati> why ETH_IRQ is never triggered?
2012-10-16T01:10:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-16T01:11:54 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-16T01:11:54 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Client Quit]
2012-10-16T01:17:08 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-16T01:31:20 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-16T02:10:10 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-10-16T02:11:49 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Read error: Connection reset by peer]
2012-10-16T02:12:26 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-16T02:13:20 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-16T02:13:56 < Posterdati> please how can I configure ETH irq? The st LWIP demo won't work!
2012-10-16T02:19:38 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Quit: Going away]
2012-10-16T02:19:53 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-16T02:19:56 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-16T02:33:03 -!- DaKu is now known as daku
2012-10-16T02:36:39 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-16T02:44:57 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-16T02:52:17 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Read error: Connection reset by peer]
2012-10-16T02:53:00 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-16T02:55:10 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T02:56:30 < dongs> http://bcas.tv/paste/results/UEly4F50.html lol
2012-10-16T02:56:59 < dongs> Posterdati: just a random guess, did you power up ethernet peripheral?
2012-10-16T03:00:00 < qyx_> :D
2012-10-16T03:11:43 < emeb> Mr. Dongs, professional troll.
2012-10-16T03:12:42 < dongs> thats what i have on my linkedin profile!
2012-10-16T03:13:08 < emeb> Also works as a pickup line in dive bars.
2012-10-16T03:13:26 < emeb> Helllooo ladies! I'm a _professional_ troll.
2012-10-16T03:13:38 < dongs> whats a dive bar.
2012-10-16T03:13:41 < dongs> sorry I dont get out much.
2012-10-16T03:14:11 < emeb> A sleazy bar where loose women and angry men hang out and frequently commit mayhem.
2012-10-16T03:15:03 < emeb> http://www.urbandictionary.com/define.php?term=Dive%20bar
2012-10-16T03:16:07 < emeb> I'm guessing you don't have a lot of that sort of thing in rural JP.
2012-10-16T03:16:51 < dongs> either that or I dont go to bars.
2012-10-16T03:17:34 < emeb> I did that a few times in college. Not something I got into the habit of.
2012-10-16T03:19:20 < dongs> annd, ledshit sent off to pcbplace, so if something's fucked too bad.
2012-10-16T03:19:44 < emeb> how many boards you getting on the 1st run?
2012-10-16T03:19:55 < dongs> i dunno whatevcer their sample shit is,, > 10 i guess
2012-10-16T03:20:03 < dongs> im paneling it with some other guys order.
2012-10-16T03:20:26 < emeb> eso si que es. It is what it is.
2012-10-16T03:21:07 < dongs> definitely on schedule to have everything by this weekend
2012-10-16T03:21:38 < emeb> expecting Youtube timecop videos promptly.
2012-10-16T03:42:41 < emeb> Online pizza orders w/ googled coupon codes = win.
2012-10-16T03:43:37 < qyx_> lol
2012-10-16T03:44:45 < emeb> onoz - twitter is over capacity.
2012-10-16T03:47:16 < BrainDamage> einstein was wrong! http://imgur.com/59rw6 ... or not
2012-10-16T03:48:59 < emeb> they said it on telebishun - it must be true.
2012-10-16T03:59:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-16T04:01:31 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-16T04:05:42 -!- heryMcDerpykins [~herplolde@c-98-245-252-134.hsd1.co.comcast.net] has joined ##stm32
2012-10-16T04:11:03 -!- heryMcDerpykins [~herplolde@c-98-245-252-134.hsd1.co.comcast.net] has quit [Ping timeout: 276 seconds]
2012-10-16T04:39:15 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-16T04:41:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-16T04:59:07 < dongs> what the fucking hell
2012-10-16T04:59:24 < dongs> some dude submits a pcb to do... as dxf files... and asks... can we try different via density...
2012-10-16T04:59:39 < dongs> via diameter: 0.15mm... density: 1/mm^2 2/mm^2 3/mm^2...
2012-10-16T04:59:54 < dongs> is he trying to make a sieve or something.
2012-10-16T05:00:07 < dongs> or is that some nutty antenna shit
2012-10-16T05:00:14 < dongs> 4 layers of that
2012-10-16T05:00:17 < dongs> with all layers solid copper.
2012-10-16T05:01:45 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-16T05:02:04 < emeb> can't imagine that would be useful for antenna
2012-10-16T05:02:15 < dongs> i cant imagine that would be useful for ANYTHING
2012-10-16T05:02:36 < emeb> maybe some sort of thermal conductivity experiments?
2012-10-16T05:03:01 < dongs> 3 holes per mm^2 ...
2012-10-16T05:03:26 < emeb> is that even manufacturable?
2012-10-16T05:04:08 < dongs> min is 0.2mm
2012-10-16T05:04:11 < dongs> (via)
2012-10-16T05:04:29 < emeb> even if it were - you'd burn through a lot of drills & time doing it on a continuous grid.
2012-10-16T05:06:01 < dongs> 0.2mm shit is laser only
2012-10-16T05:08:07 < emeb> fricken lazers.
2012-10-16T05:08:22 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-16T05:09:26 < dongs> if thats really waht they want.. im guessing hes here as last resort because everyone else either told them to fuck off or asked thousands.
2012-10-16T05:10:10 < emeb> sounds right.
2012-10-16T05:11:25 < dongs> stencil guys should be able to make a good sieve this way :)
2012-10-16T05:11:38 < dongs> i think 0.15mm holes in a tight grid should be no problem there
2012-10-16T05:16:16 < dongs> haha, its a heatsink
2012-10-16T05:16:28 < dongs> just all copper, no electrical connections, and as many vias as possible
2012-10-16T05:25:20 < emeb> bingo
2012-10-16T05:40:58 < emeb> D'oh! Part installed backwards.
2012-10-16T05:41:06 < emeb> Luckily just an SOIC-8
2012-10-16T06:05:45 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-16T06:12:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-16T06:12:46 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-16T06:26:59 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-16T06:32:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-16T06:58:29 < emeb_mac> quiet here... too quiet.
2012-10-16T07:05:38 < mrnuke> emeb: It's 'cause you have a mac
2012-10-16T07:07:00 < dongs> agreed
2012-10-16T07:07:02 < emeb_mac> ark ark
2012-10-16T07:07:08 < dongs> youre missing all the exciting chats cuz youre on a mac
2012-10-16T07:07:21 < emeb_mac> special mac filter on the IRC channel.
2012-10-16T07:07:25 < dongs> ya
2012-10-16T07:07:28 < dongs> +k no_macs
2012-10-16T07:08:37 < emeb_mac> awesome.
2012-10-16T07:09:13 < mrnuke> it took me 5 hours to install gnuplot on a mac
2012-10-16T07:09:34 < mrnuke> but maaan, those graphs are pretty
2012-10-16T07:09:37 < emeb_mac> err - "sudo port install gnuplot"
2012-10-16T07:10:01 < mrnuke> on a vanilla mac?
2012-10-16T07:10:29 < emeb_mac> well, it does presuppose you've got macports installed already.
2012-10-16T07:11:25 < mrnuke> see, there's the rub. I would have spent 4 hours 58 minutes and 2 seconds installing that
2012-10-16T07:11:30 < dongs> http://abcnews.go.com/US/wireStory/police-release-names-maine-zumba-sex-case-17485548
2012-10-16T07:11:33 < dongs> america
2012-10-16T07:11:41 < mrnuke> the other 1 minute and 58 seconds would have meen spent opening up a terminal
2012-10-16T07:13:24 < emeb_mac> your gnuplot using aquaterm?
2012-10-16T07:15:52 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 246 seconds]
2012-10-16T07:16:15 < mrnuke> nope, I had to install X11 to get it running
2012-10-16T07:21:52 < emeb_mac> does the x11 version give you zoom/pan options as the latest linux version does?
2012-10-16T07:25:10 < mrnuke> I have no idea. It wasn't my mac, so i didn't care if I messed it up :p
2012-10-16T07:25:31 < mrnuke> saw gnuplot plotting, and gave it back to the owner
2012-10-16T07:26:45 < emeb_mac> ah - not your thing.
2012-10-16T07:27:04 < mrnuke> macs slow me down
2012-10-16T07:27:15 < emeb_mac> what speeds you up?
2012-10-16T07:27:44 < mrnuke> 18gigs of RAM usually does the trick
2012-10-16T07:28:34 < mrnuke> but you'd be surprised how well KDE and I get a long
2012-10-16T07:28:51 < emeb_mac> ah.
2012-10-16T07:29:07 < emeb_mac> more of a gnome guy myself
2012-10-16T07:29:44 < mrnuke> every once in a while, I like gnome
2012-10-16T07:30:11 < emeb_mac> I'm not really liking gnome 3
2012-10-16T07:30:57 < mrnuke> I don't think I seriously tried gnome3
2012-10-16T07:31:34 < emeb_mac> avoid if possible.
2012-10-16T07:32:07 < mrnuke> is that why you bought a MAC. Favorite desktop started sucking?
2012-10-16T07:35:26 < mrnuke> I remember having thi eeepc, which was not particularly fast, so I was running startkde from within fluxbox
2012-10-16T07:36:10 < emeb_mac> Nah - I use a mac for casual surfing while watching tv.
2012-10-16T07:36:24 < emeb_mac> I use Fedora 17 for work.
2012-10-16T07:36:45 < mrnuke> hehe! I use Fedora 17 for everything here
2012-10-16T07:40:02 < emeb_mac> got an eee pc netbook w/ Ubuntu 12.04 for taking on the road. Not running unity tho.
2012-10-16T07:41:39 < mrnuke> at least it's not SUSE :p
2012-10-16T07:42:38 < mrnuke> Well, I gotta go
2012-10-16T07:42:45 < mrnuke> time for one more smoke and sleep
2012-10-16T07:43:58 < emeb_mac> later
2012-10-16T07:44:05 < mrnuke> later
2012-10-16T08:20:50 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-16T08:42:42 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xhojwlvjozmxwest] has quit [Quit: leaving]
2012-10-16T08:43:58 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-eptzipaykxztrihb] has joined ##stm32
2012-10-16T08:54:34 -!- Tectu is now known as tectu
2012-10-16T09:07:43 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-16T09:12:51 < zippe> yay for rewriting .S files as __asm__ so that I can use llvm
2012-10-16T09:12:57 < zippe> or clang, really
2012-10-16T09:13:35 < emeb_mac> what's involved in that?
2012-10-16T09:14:42 < zippe> Mostly just futzing around with getting arguments into the right registers
2012-10-16T09:14:57 < zippe> There aren't explicit constraints for the individual ARM registers
2012-10-16T09:15:19 < zippe> So you have to use __asm__("r0") etc. on the local variables and then the "r" constraint
2012-10-16T09:17:13 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-16T09:20:47 -!- tectu is now known as Tectu
2012-10-16T09:24:57 < emeb_mac> wait - this is hand-coded assembly and you can't specify the registers?
2012-10-16T09:25:14 < emeb_mac> or is it inline assy inside C files?
2012-10-16T09:28:29 < emeb_mac> OK - figured out what you're talking about. Converting hand-coded .S files to inline assy.
2012-10-16T09:28:47 < zippe> Yes, sorry
2012-10-16T09:28:48 < emeb_mac> and trying to control how named vars get allocated into regs.
2012-10-16T09:29:31 < emeb_mac> I was playing with that a while back.
2012-10-16T09:29:48 < emeb_mac> trying to use the new M4 saturating math instructions
2012-10-16T09:30:22 -!- dekar [~dekar@212.255.35.17] has quit [Quit: This computer has gone to sleep]
2012-10-16T09:31:13 < emeb_mac> ended up doing this to saturate a var called "sum":
2012-10-16T09:31:15 < emeb_mac> 		asm("ssat %[dst], #16, %[src]" : [dst] "=r" (sum) : [src] "r" (sum));
2012-10-16T09:31:23 < emeb_mac> seemed to work ok.
2012-10-16T09:32:33 < emeb_mac> but honestly, the compiler was doing such a good job optimizing that explicit assembly wasn't making much difference.
2012-10-16T09:32:58 < emeb_mac> I guess I'd have to really work to find a place where it'd be worthwhile.
2012-10-16T09:34:12 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T09:34:14 < zippe> Context save/restore code
2012-10-16T09:34:25 < zippe>   register uint32_t opcode __asm__("r0") = SYS_switch_context;
2012-10-16T09:34:26 < zippe>   register uint32_t sregs __asm__("r1") = (uint32_t)saveregs;
2012-10-16T09:34:26 < zippe>   register uint32_t rregs __asm__("r2") = (uint32_t)restoreregs;
2012-10-16T09:34:26 < zippe>   __asm__ volatile("svc 0" :: "r" (opcode), "r" (sregs), "r" (rregs));
2012-10-16T09:34:34 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T09:37:02 < emeb_mac> seems fairly tidy.
2012-10-16T09:37:08 < emeb_mac> did it do what you wanted?
2012-10-16T09:38:34 < zippe> At -O0, there is some screwing around with scratch registers that makes me nervous
2012-10-16T09:39:09 < zippe> gcc seems to really like moving arguments into r3 first
2012-10-16T09:39:11 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T09:42:27 < zippe> Hmm, actually, seems to be related to the order that variables are declared in
2012-10-16T09:42:40 < zippe> If I move the assignment to r0 after the others, it doesn't need r3
2012-10-16T09:44:45 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T09:46:40 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T09:47:57 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T09:49:43 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 245 seconds]
2012-10-16T09:51:10 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T09:52:02 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T09:54:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-16T09:54:41 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T09:56:25 < zippe> The toplevel exception handler will be … moar exciting
2012-10-16T10:00:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-16T10:02:20 < emeb_mac> take yer excitements where you can get 'em.
2012-10-16T10:04:06 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-16T10:06:17 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 246 seconds]
2012-10-16T10:07:28 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T10:10:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-16T10:12:10 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-16T10:12:43 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has joined ##stm32
2012-10-16T10:12:46 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T10:13:58 < dongs> instead of working i spent half a day fixing broken lunix shit
2012-10-16T10:14:00 < dongs> good job
2012-10-16T10:17:58 -!- emeb_mac [~ericb@72.201.79.123] has joined ##stm32
2012-10-16T10:21:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-16T10:24:12 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T10:26:26 < BusError_> anyone know the typical power consumption for a stm32f1/2/4 in lqfp64 ? I'm sizing an LDO here, and I wonder if the handy, tiny 250mA is enough
2012-10-16T10:27:44 < dongs> i think emeb had some numbers, something liek 100mA when doing a bunch of audio DSP shit
2012-10-16T10:27:49 < dongs> so youre prob ok
2012-10-16T10:27:59 < dongs> i'm using equivalent of LP2992
2012-10-16T10:28:17 < dongs> 100mA for F4, f1 is much less
2012-10-16T10:29:42 < BusError_> ohh cool. I have these mcp1700 in sot23, so it fit the bill easily
2012-10-16T10:30:19 < BusError_> thanks for that, trying to find that rating from the datasheet or the interweb leads nowhere
2012-10-16T10:31:00 < dongs> ive used LP2992, MCP1802Ts
2012-10-16T10:31:03 < dongs> all same pinout
2012-10-16T10:31:08 < dongs> then switched to cheap richtek parts
2012-10-16T10:32:06 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-16T10:32:39 < BusError_> can I run a couple of thing by you quickly just to make sure I don't do something stupid ? ;-) making my first stm32 board here, so I don't want to beeep it up :>
2012-10-16T10:32:41 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T10:33:29 < BusError_> BOOT0 is tied to GND, and I have a pushbutton to tie it up, I assume pressing the button, and reset will trigger the DFU bootloader. Am I correct?
2012-10-16T10:34:53 < BusError_> seesm ti make the footprint work with F1/F2/F4, I just need two 2.2uF on pins 31&47. I fit 0R there for F1, 2.2uF instead for F2/F4..
2012-10-16T10:35:03 < dongs> BusError_: uh no
2012-10-16T10:35:20 < dongs> by "tied" of course you mean like 10k pulldown right?
2012-10-16T10:35:34 < dongs> and up -> vcc -> switch -> BOOT0
2012-10-16T10:37:11 < emeb_mac> BusError_: yeah - re current draw - I'm seeing ~90ma for an F405 running full speed and a few I/O toggling.
2012-10-16T10:37:23 < emeb_mac> zzzzzzzz
2012-10-16T10:37:27 -!- emeb_mac [~ericb@72.201.79.123] has quit [Quit: emeb_mac]
2012-10-16T10:38:12 < BusError_> yeah there is a pulldown... So basically the "pull it up" button will work... the application note is not entirely clear; it;s like they are trying NOT to say anything useful in there
2012-10-16T10:38:35 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-16T10:39:09 < dongs> haha
2012-10-16T10:39:40 < dongs> i just have pads and short them with tweezers.
2012-10-16T10:39:46 < dongs> no need to waste $ on a button
2012-10-16T10:39:50 < dongs> short, power up, remove
2012-10-16T10:40:58 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 272 seconds]
2012-10-16T10:41:21 < BusError_> trying ti make something for a group who think thru hole is small and hard,,, not sure the tweezer trick would go down well :>
2012-10-16T10:44:05 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Read error: Connection reset by peer]
2012-10-16T10:45:19 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-16T10:45:27 -!- nashpa [~nashpa@dliviu.plus.com] has quit [Ping timeout: 256 seconds]
2012-10-16T10:46:16 -!- nashpa [~nashpa@dliviu.plus.com] has joined ##stm32
2012-10-16T10:46:51 < cjbaird> Spent the day visiting my mother in Trololoville http://i.imgur.com/kquQD.jpg
2012-10-16T10:58:40 <+dekar> zyp, Laurenceb "Framebuffer only - no VPU/GPU in use yet. VPU/GPU support in GNU/Linux hopefully coming soon." http://a10linux.org/
2012-10-16T10:59:07 < zyp> huh?
2012-10-16T11:01:49 <+dekar> allwinner crap, randomly read that in some forum
2012-10-16T11:02:27 < zyp> well, I got that part
2012-10-16T11:02:40 < zyp> but why me?
2012-10-16T11:03:09 <+dekar> didn't you argue the allwinner was so much better than raspi?
2012-10-16T11:03:50 < zyp> no, allwinner is singlecore A8, not much better. :p
2012-10-16T11:04:02 <+dekar> anyway, gtg ;)
2012-10-16T11:04:04 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-16T11:06:06 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T11:13:17 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has joined ##stm32
2012-10-16T11:13:21 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T11:19:31 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T11:24:44 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T11:28:39 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T11:36:30 < zippe> Er, A8 with L2 is *much* better than 1176
2012-10-16T11:41:05 < dongs> sup dongs
2012-10-16T11:41:29 < Posterdati> hi
2012-10-16T11:41:38 < dongs> Posterdati: did you power up ETH peripheral?
2012-10-16T11:41:39 < dongs> rcc_blah
2012-10-16T11:41:43 < zyp> zippe, «much» is relative
2012-10-16T11:41:55 < Posterdati> dongs: no
2012-10-16T11:42:06 < dongs> thats probably why its not firing interrupts.
2012-10-16T11:42:06 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T11:42:31 < Posterdati> dongs: leds go on, but no ETH_IRQ is triggered
2012-10-16T11:42:54 < dongs> no shit, >did you power on the ethernet peripheral clocks<
2012-10-16T11:42:59 < zyp> zippe, I believe the discussion he refers to was rpi vs odroid-x, i.e. quad A9
2012-10-16T11:43:19 < Posterdati> dongs: using RCC?
2012-10-16T11:43:23 < dongs> yea
2012-10-16T11:43:31 < Posterdati> dongs: yes
2012-10-16T11:44:33 < Posterdati> RCC_APB1PeriphClockCmd(RCC_APB1Periph_USART2, ENABLE);
2012-10-16T11:44:46 < Posterdati> RCC_AHBPeriphClockCmd(RCC_AHBPeriph_ETH_MAC | RCC_AHBPeriph_ETH_MAC_Tx |                         RCC_AHBPeriph_ETH_MAC_Rx, ENABLE);
2012-10-16T11:45:30 < dongs> okay.
2012-10-16T11:45:51 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-16T11:45:55 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-16T11:46:06 < Posterdati> ETH_IRQHandler is never triggered
2012-10-16T11:46:53 < Posterdati> maybe a NVIC misconfiguration
2012-10-16T11:46:59 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-16T11:47:22 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-16T11:49:00 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T11:52:01 < Posterdati> dongs: nvic configuration -> http://pastebin.com/wYwz3K01
2012-10-16T11:53:33 < dongs> i saw it last time
2012-10-16T11:54:05 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T11:54:45 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Max SendQ exceeded]
2012-10-16T11:54:50 < Posterdati> then I chnaged PendSV and SysTick too
2012-10-16T11:55:31 < Posterdati> NVIC_SetPriority(SysTick_IRQn, 0x01);
2012-10-16T11:55:38 < Posterdati> NVIC_SetPriority(PendSV_IRQn, 0xff);
2012-10-16T11:57:04 < Posterdati> dongs: PE14?
2012-10-16T11:58:05 < Posterdati> PE14 = ETH_RMII_MDINT
2012-10-16T11:59:01 < Posterdati> this is the config -> http://pastebin.com/fK36xFpg
2012-10-16T12:04:53 < Posterdati> should it be AF_OD?
2012-10-16T12:09:55 < dongs> did you try looking at st samples for ethernet
2012-10-16T12:13:37 < Posterdati> I found one with mdint
2012-10-16T12:13:46 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-16T12:28:12  * Laurenceb is watching from dusk till dawn
2012-10-16T12:28:29 < dongs> isnt that gay
2012-10-16T12:28:37  * dongs is laying out clone of Laurenceb stm32logger
2012-10-16T12:29:43 < Laurenceb> psycho guy reminds me of my dad <3
2012-10-16T12:33:35 < Laurenceb> dongs idea of a bbq  http://t3.gstatic.com/images?q=tbn:ANd9GcTLBUgkeWgY789r5R1KZEZdodAK1YiKMRoyrfqHQdZQ-H6Vj0zP-g
2012-10-16T12:33:39 < Laurenceb> nsfw
2012-10-16T12:33:50 < dongs> wut
2012-10-16T12:33:55 < dongs> what w is it nsf at?
2012-10-16T12:34:19 < Laurenceb> someone might get the wrong idea
2012-10-16T12:34:41 < Laurenceb> e.g. if you work at a petrol/gas station
2012-10-16T12:38:50 <+Steffanx> You are weird Laurenceb
2012-10-16T12:40:37 < Laurenceb> http://anti-joke.com/anti-joke/popular/2510-what-did-the-blind-deaf-quadriplegic-boy-get-for-christmas-cancer
2012-10-16T12:41:03 < Thorn> dongs: did you see this in diptrace: you make a component with lots of pins, place it into schematics, connect the pins, turn pin numbers on, and they're all 4 or 8
2012-10-16T12:41:46 < Thorn> when you replace part with itself it becomes normal but all connections are fucked up
2012-10-16T12:41:48 < dongs> uh no. but you can edit pin number and name in component editor
2012-10-16T12:42:03 < dongs> how did you "make" the pins?
2012-10-16T12:42:07 < dongs> copypaste same pin over and over?
2012-10-16T12:42:20 < Posterdati> dongs: no way, it won't work :(
2012-10-16T12:42:43 < Thorn> yes but I normally paste a pin, change its number, copy/paste it again, etc.
2012-10-16T12:43:11 < dongs> wrong
2012-10-16T12:43:13 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-16T12:43:14 < dongs> there's name and number
2012-10-16T12:43:15 < dongs> per pin
2012-10-16T12:43:20 < dongs> you probably have a bunch of pins called 4 and 8 :)
2012-10-16T12:43:33 < dongs> right click a pin
2012-10-16T12:43:35 < Thorn> no I change names too of course.
2012-10-16T12:43:37 < dongs> pin properties
2012-10-16T12:43:39 < dongs> oh.
2012-10-16T12:43:42 < dongs> how about pattern then
2012-10-16T12:43:46 < dongs> is that properly assigned?
2012-10-16T12:43:54 < dongs> if you have same named pins on pattern that';ll break sthi too
2012-10-16T12:43:56 < Thorn> pattern seems to have no effect in this
2012-10-16T12:44:14 < dongs> running latest beta from  this month?
2012-10-16T12:44:20 < Thorn> 2.2.9.1
2012-10-16T12:45:56 < dongs> isolate the problem / make it repeatable and submit a bug on their tracker.
2012-10-16T12:46:02 < dongs> but ive personally not seen this
2012-10-16T12:48:39 < Thorn> diptrace. draw your circuit with your dong.
2012-10-16T12:50:09 <+Steffanx> *dicktrace
2012-10-16T12:51:06 < zyp> zippe, quick question: https://github.com/PX4/Firmware/blob/master/nuttx/arch/arm/include/armv7-m/irq.h#L174 <- any idea why it does test/branch/cpsie instead of just msr primask, %0?
2012-10-16T12:58:24 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has joined ##stm32
2012-10-16T12:58:27 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T13:00:43 < Posterdati> zyp: hi
2012-10-16T13:02:50 < Posterdati> dongs: I programmed ETH_RMII_MDINT --> PE14 as GPIO_Mode_IN_FLOATING, but no luck
2012-10-16T13:03:17 < Laurenceb> hehe pixhawk
2012-10-16T13:03:32 < Laurenceb> do they have sane code now?
2012-10-16T13:05:05 < dongs> no, it got even more insane
2012-10-16T13:05:59 < Laurenceb> https://github.com/PX4/Firmware/blob/master/apps/attitude_estimator_ekf/attitude_estimator_ekf_main.c
2012-10-16T13:06:08 < Laurenceb> their kalman filter looks a bit weird
2012-10-16T13:06:39 <+Steffanx> No, yours looks great ..
2012-10-16T13:06:48 < Posterdati> :(
2012-10-16T13:07:50 <+Steffanx> The code looks clear to me Laurenceb ...
2012-10-16T13:08:17 < Laurenceb> https://github.com/PX4/Firmware/blob/master/apps/attitude_estimator_ekf/codegen/attitudeKalmanfilter.c#L54
2012-10-16T13:08:54 <+Steffanx> Oh, yeah that's a little less clear :)
2012-10-16T13:09:42 <+Steffanx> but that's generated code?
2012-10-16T13:10:14 < jpa-> why don't people use matrix libraries when they write those filters.. it's not like the overhead is so big
2012-10-16T13:10:59 < jpa-> ah, generated.. i wonder by what
2012-10-16T13:11:30 < dongs> matlabia probably
2012-10-16T13:12:14 < Laurenceb>  state vector x has the following entries [ax,ay,az||mx,my,mz||wox,woy,woz||wx,wy,wz]' */
2012-10-16T13:12:19 < Laurenceb> thats interesting
2012-10-16T13:14:11 < dongs> steve woz vector
2012-10-16T13:14:23 < dongs> this one gives you cancer.
2012-10-16T13:14:27 < Laurenceb> lol
2012-10-16T13:14:38 < Laurenceb> but the state is the same as the sensors
2012-10-16T13:14:48 < dongs> i wish I knew even 1% of whwat the fuck youre talkin about.
2012-10-16T13:15:34 < dongs> to me, matrix is a movie made by lana "anal" wachowsky
2012-10-16T13:15:38 < Laurenceb> https://store.diydrones.com/ProductDetails.asp?ProductCode=BR-PXFMU
2012-10-16T13:15:55 < dongs> ya, total sellouts
2012-10-16T13:17:31 < dongs> ive been told the productio nhas been a total mess
2012-10-16T13:17:32 < Laurenceb> https://pixhawk.ethz.ch/px4/
2012-10-16T13:17:39 < Laurenceb> looks overcomplex to me
2012-10-16T13:17:40 < dongs> caps in wrong places
2012-10-16T13:17:41 < dongs> missing docs
2012-10-16T13:17:42 < dongs> etc
2012-10-16T13:17:48 < Laurenceb> they could integrate on one pcb
2012-10-16T13:17:59 < dongs> its just a way to make you pay $150+100
2012-10-16T13:18:04 < Laurenceb> wonder if the firmware would port
2012-10-16T13:18:47 < Laurenceb> or if they still have the horrific mavlink code
2012-10-16T13:19:04 < dongs> port to waht
2012-10-16T13:20:43 < qyx_> posix compliant rtos?
2012-10-16T13:21:04 < Laurenceb> https://github.com/PX4/Firmware/blob/master/mavlink/share/mavlink/src/v1.0/pixhawk/pixhawk.pb.cc
2012-10-16T13:21:06 < Laurenceb> the hell
2012-10-16T13:21:18 < dongs> protobuffers
2012-10-16T13:21:19 < dongs> lol
2012-10-16T13:21:34 < Laurenceb> wtf is all the google stuff
2012-10-16T13:21:42 < dongs> protobuffers is some google trash
2012-10-16T13:21:52 < dongs> https://developers.google.com/protocol-buffers/docs/overview
2012-10-16T13:22:03 < dongs> serdes for ur codez!!
2012-10-16T13:22:10 < qyx_> nuttx?
2012-10-16T13:22:11 < qyx_> hm
2012-10-16T13:22:45 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-16T13:23:54 < Laurenceb> omg wut
2012-10-16T13:24:05 < Laurenceb> so they did something even more insane with their mavlink code
2012-10-16T13:25:13 < dongs> http://nuttx.sourceforge.net/NuttxUserGuide.html
2012-10-16T13:25:14 < dongs> reat
2012-10-16T13:25:16 < dongs> great
2012-10-16T13:25:16 < dongs> 404
2012-10-16T13:25:17 < qyx_> have anyone used that nuttx?
2012-10-16T13:25:24 < dongs> i think zippe wrote most of it
2012-10-16T13:25:34 < qyx_> dongs: moved
2012-10-16T13:25:35 < qyx_> http://nuttx.org/doku.php?id=documentation:userguide
2012-10-16T13:25:46 < dongs> does anyone use sorurceforge anymore
2012-10-16T13:26:52 < dongs> wtf
2012-10-16T13:26:54 < dongs> pthread?
2012-10-16T13:27:02 <+Steffanx> I could say "yes" but then you'll say i'm annoying :P
2012-10-16T13:27:18 < dongs> link to your warez on sourceforge
2012-10-16T13:27:39 <+Steffanx> I don't say I use it, but i'm sure "someone" uses it :)
2012-10-16T13:29:00 < qyx_> hm, execution of binaries directly from filesystem
2012-10-16T13:29:06 < qyx_> mountable filesystems, vfs
2012-10-16T13:29:13 < qyx_> nice
2012-10-16T13:29:17 < dongs> But standard compliance is more important than small footprint. Surely a smaller RTOS could be produced by ignoring standards. Think of NuttX is a tiny Linux work-alike with a much reduced feature set.
2012-10-16T13:29:28 < dongs> closing page.
2012-10-16T13:29:36 < qyx_> why
2012-10-16T13:29:43 < dongs> I dont need a lunix look-alike
2012-10-16T13:29:49 < dongs> if I wanted the fail of lunix, I'd just run lunix.
2012-10-16T13:30:09 < qyx_> think of vxworks
2012-10-16T13:30:17 < qyx_> it got most of mars rovers to mars
2012-10-16T13:30:22 < qyx_> and it looks like lunix
2012-10-16T13:31:09 < Laurenceb> what? stuck on another planet?
2012-10-16T13:31:37 < dongs> Because of this standards conformance, software developed under other standard OSs (such as Linux) should port easily to NuttX.
2012-10-16T13:32:56 < dongs> to me this basicalyl smells "our shit is only compileable by GCC, and only using makefiles. if you use any proper IDE or compiler, youore fucked"
2012-10-16T13:33:22 < qyx_> hater
2012-10-16T13:33:31 <+Steffanx> Like it should be dongs
2012-10-16T13:33:43 <+Steffanx> and what qyx_ said
2012-10-16T13:34:16 < dongs> :)
2012-10-16T13:36:02 < Laurenceb> dongs hates everything
2012-10-16T13:36:06 < Laurenceb> even the jews
2012-10-16T13:38:19 < dongs> sleep time
2012-10-16T13:38:19 < dongs> bbl
2012-10-16T13:38:36 < Posterdati> hi
2012-10-16T13:39:44 < Posterdati> please I need help on stm32 + ste101p, I'm running lwip, but ETH_IRQHandler is never triggered. I copied an stm32-p107 rev. a demo, but isn't working! thanks
2012-10-16T13:41:46 < Laurenceb> On my first day in prison I was taking a shower when I noticed a bar of Dove lying on the floor.
2012-10-16T13:41:46 < Laurenceb> Some guy said, "Hey, where's the soap?"
2012-10-16T13:41:46 < Laurenceb> I said, "I'll be fucked if I know."
2012-10-16T13:43:11 < Thorn> Posterdati: what revision is your board? because they changed the PHY to KS8721 in rev. b
2012-10-16T13:43:15 < Posterdati> you'll br fucked even if you don't know
2012-10-16T13:43:22 < Posterdati> Thorn: rev. a.
2012-10-16T13:43:56 < Posterdati> Thorn: rev. a = ste101p, there's also an ETH_RMII_MDINT <--> PE14
2012-10-16T13:44:31 < Posterdati> I configured it as GPIO_Mode_IN_FLOATING
2012-10-16T13:45:00 < Thorn> why?
2012-10-16T13:48:46 < Posterdati> MDINT is that an interrupt?
2012-10-16T13:50:40 < Thorn> no idea, it's not part of MII or RMII
2012-10-16T13:51:00 < Posterdati> yes I know, but on the board it is connected
2012-10-16T13:52:28 < Posterdati> Thorn: anyway... ETH_IRQHandler is not called so no packet handling at all
2012-10-16T13:53:22 < Thorn> it's connected to a gpio pin which is not used by the MAC core at all. you configured that pin as an input and... you expect it to trigger ETH_IRQHandler?
2012-10-16T13:55:18 < Posterdati> no
2012-10-16T13:55:30 < Posterdati> ETH_IRQHandler is DMA related
2012-10-16T13:56:09 < Posterdati> ah the PHY_ADDRESS
2012-10-16T13:57:24 < Posterdati> maybe it is wrong let me check
2012-10-16T14:05:39 < Laurenceb> will usb work with a pullup to 2.4v on L1?
2012-10-16T14:08:47 < scrts> btw, anyone running LwIP on STM32F4xx?
2012-10-16T14:09:43 < Posterdati> Thorn: can't find it
2012-10-16T14:10:22 < Thorn> scrts: I'm planning to... eventually
2012-10-16T14:11:23 < scrts> the reference design readme says, that DSP stuff is disabled there... Are You planning to use DSP?
2012-10-16T14:12:30 < Thorn> you mean STM32F4x7_ETH_LwIP_V1.0.0/Release_Notes.html ?
2012-10-16T14:13:04 < Thorn> they use and old version of freertos, that's the reason aiui
2012-10-16T14:16:08 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has joined ##stm32
2012-10-16T14:16:29 < Posterdati> :(
2012-10-16T14:16:58 < Posterdati> Thorn: ETH_Init(&ETH_InitStructure, PHY_ADDRESS);
2012-10-16T14:17:21 < Posterdati> where I can found PHY_ADDRESS value for ste101p? The datasheet is vague
2012-10-16T14:17:52 < Thorn> in the vague datasheet
2012-10-16T14:18:07 < Thorn> it should say how it's configured and what's the default
2012-10-16T14:18:17 < Thorn> for dp83848 it's 0001
2012-10-16T14:18:25 < Thorn> or even 00001
2012-10-16T14:18:40 < Posterdati> I tried it, not working even 0x00
2012-10-16T14:19:24 < Thorn> 31 more values to try...
2012-10-16T14:19:45 < Thorn> is there anything on interface pins? mii/rmii/smi
2012-10-16T14:19:50 < Thorn> scope them?
2012-10-16T14:21:41 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-16T14:32:36 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-16T14:33:00 < scrts> Thorn: is there an update version demo? or maybe You've ported a new one?
2012-10-16T14:33:07 < scrts> *updated
2012-10-16T14:33:28 < scrts> also what tools do You use for development? I mean ide and compiler
2012-10-16T14:34:13 < Thorn> I most likely will once I get to coding, need to do the hardware first
2012-10-16T14:34:32 < Thorn> I'm not using an ide, just a big makefile
2012-10-16T14:34:46 < scrts> what about debug then?
2012-10-16T14:34:56 < Thorn> make gdb :)
2012-10-16T14:35:00 < scrts> :>
2012-10-16T14:35:25 < Posterdati> The STE101P PAD[4:0] inputs provide up to 32 unique PHY address options. An address selection of all zeros (00000) will result in a PHY isolation condition as a result of power- on/reset, as documented for Register PR00 bit 10. (See Section 7 for more detailed descriptions of device operation.)
2012-10-16T14:35:55 < Posterdati> Thorn: but there aren't PAD pins!!
2012-10-16T14:37:09 < Thorn> what are their numbers.
2012-10-16T14:37:34 < Thorn> I see MF[4:0] pins with resistors to 3.3V and ground on them in the schematics.
2012-10-16T14:38:06 < Posterdati> MF is a configuration
2012-10-16T14:38:20 < Thorn> also, these kinds of pins mols likely have internall pullups/pulldowns which effectively provide defaults
2012-10-16T14:39:18 < Posterdati> MF0 = Auto Negotiation
2012-10-16T14:46:04 < Thorn> okay, looked at the datasheet. the datasheet is FUCKING HORRIBLE
2012-10-16T14:46:17 < Thorn> phy address is in PR14[7:3]
2012-10-16T14:46:56 < Thorn> searching for PR14. e.g. pin 38, ledr10. The pull-up/pull-down status of this pin is latched into the PR14 bit 7 during power up/reset.
2012-10-16T14:47:02 < Thorn> etc.
2012-10-16T14:47:49 < Thorn> i.e. pins 34...38 set phy address at reset.
2012-10-16T14:47:54 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-16T14:50:01 < Posterdati> Thorn: but they are for the led?
2012-10-16T14:50:03 < Thorn> looking at schematic, which is not much better. pins 38, 36, 35, 34 have pullups, pin 37 has a pulldown
2012-10-16T14:50:26 < Thorn> so the address is 10111 unless I'm wrong
2012-10-16T14:51:01 < scrts> isn't the address is automatically found?
2012-10-16T14:53:08 < Posterdati> 10011
2012-10-16T14:53:11 < Thorn> I don't see anything about that in the MAC documentation
2012-10-16T14:53:39 < Posterdati> 0x13
2012-10-16T14:54:26 < Thorn> why 10011?
2012-10-16T14:54:41 < Posterdati> pin 38 = +3.3V
2012-10-16T14:54:49 < Posterdati> pin37 = gnd
2012-10-16T14:54:54 < Posterdati> pin36 = gnd
2012-10-16T14:55:12 < Posterdati> pin35 =gnd
2012-10-16T14:55:15 < Thorn> pin36 = gnd?
2012-10-16T14:55:20 < Thorn> I see a pullup
2012-10-16T14:55:30 < Posterdati> you're right
2012-10-16T14:55:38 < Posterdati> I didn't follow the wire :)
2012-10-16T14:55:43 < Thorn> pin 35 has a pulldown though
2012-10-16T14:55:55 < Thorn> so 10101
2012-10-16T14:56:41 < Posterdati> 0x15 then
2012-10-16T14:57:13 < Posterdati> let's see
2012-10-16T14:57:17 < Thorn> try reading some register from the phy
2012-10-16T14:57:22 < Posterdati> programming with new config..
2012-10-16T14:57:26 < Posterdati> done!
2012-10-16T14:57:34 < Thorn> like identification which is in the datasheet
2012-10-16T14:57:49 < Posterdati> still not working
2012-10-16T14:58:50 < Thorn> what is not working, ping is not working?
2012-10-16T14:59:01 < Posterdati> yes
2012-10-16T14:59:13 < Posterdati> Destination Host Unreachable
2012-10-16T14:59:23 < Thorn> >try reading some register from the phy
2012-10-16T14:59:33 < Thorn> to see if it's even alive
2012-10-16T14:59:34 < Posterdati> and still no ETH_IRQHandler call
2012-10-16T14:59:44 < Posterdati> but how?
2012-10-16T14:59:55 -!- tech2077 [~tech2077@108-249-45-175.lightspeed.hstntx.sbcglobal.net] has quit [Quit: Leaving]
2012-10-16T15:00:58 < Posterdati> ETH_DMAITConfig(ETH_DMA_IT_NIS | ETH_DMA_IT_R, ENABLE);
2012-10-16T15:01:18 < Thorn> with ETH_ReadPHYRegister() I presume
2012-10-16T15:01:28 < Thorn> forget about dma, lwip etc. for now
2012-10-16T15:01:57 < Thorn> you need better debugging skills. there're some good books about it.
2012-10-16T15:02:45 < Posterdati> I don't need books on debugging I need the damned thing to go now! :)
2012-10-16T15:03:30 < Thorn> then you have to hire a consultant
2012-10-16T15:04:32 < Thorn> you have two peripherals in an unknown state and you expect things to magically work
2012-10-16T15:05:40 < scrts> and check mdio line with the scope
2012-10-16T15:05:45 < scrts> if it's working at all
2012-10-16T15:06:09 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has joined ##stm32
2012-10-16T15:06:12 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T15:14:13 -!- |sark [~|sark@109.235.51.202] has quit [Remote host closed the connection]
2012-10-16T15:15:43 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-16T15:21:02 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-16T15:32:35 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-16T15:32:35 -!- BrainDamage [~BrainDama@dynamic-adsl-78-12-228-215.clienti.tiscali.it] has quit [Disconnected by services]
2012-10-16T15:37:48 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-16T15:37:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-16T16:06:55 < Posterdati> registers are read and wrote correctly and even mdio is active
2012-10-16T16:14:20 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-16T16:19:19 < scrts> then You're using correct address
2012-10-16T16:19:24 < scrts> problem is somewhere else
2012-10-16T16:19:57 < Thorn> now check rmii rxd pins while sending packets to it
2012-10-16T16:20:28 < Posterdati> I'm checking stm32_eth.c
2012-10-16T16:21:38 < Posterdati> there are some difference for ste101p, the file has got only dp83848 config
2012-10-16T16:24:07 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T16:38:28 < scrts> usually the stack doesn't start if the register locations of link-up/down, link speed, duplex differs from the one in example
2012-10-16T16:39:35 < scrts> e.g. I had to add special initialization procedure for the low level driver to initialize Micrel KSZ9021 PHY instead of other ones, that are supported by default
2012-10-16T16:39:47 < scrts> it also had different OUI (or how's that called..)
2012-10-16T16:42:50 -!- dekar [~dekar@studpool-wlan-74-158.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-16T16:44:40 -!- nashpa [~nashpa@dliviu.plus.com] has left ##stm32 []
2012-10-16T16:50:04 < Posterdati> scrts: very strange: phy sr register = 0xffff
2012-10-16T16:50:19 < scrts> sr?
2012-10-16T16:50:51 < Posterdati> status register
2012-10-16T16:51:40 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Ping timeout: 260 seconds]
2012-10-16T16:53:21 < scrts> probably read wrong...
2012-10-16T16:54:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-16T16:54:30 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-16T16:54:34 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-16T16:54:54 < Thorn> do ID registers read correctly?
2012-10-16T16:56:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-16T16:57:16 < Posterdati> testing it
2012-10-16T17:04:42 < Posterdati> Thorn: always read 0xffff on every registers, odd
2012-10-16T17:05:19 < Thorn> I asked you to read out ID registers whose values are in the datasheed. you said it worked
2012-10-16T17:05:53 < Thorn> is therer activity on mdc/mdio?
2012-10-16T17:07:54 < Posterdati> yes
2012-10-16T17:08:05 < Posterdati> odd
2012-10-16T17:08:16 < Thorn> see PR02/PR03 in the datasheet
2012-10-16T17:08:55 < Posterdati> yes they read 0xffff and 0xffff
2012-10-16T17:08:57 -!- CheBuzz [~CheBuzz@76.164.192.124] has joined ##stm32
2012-10-16T17:09:10 < Posterdati> not the values on the datasheet
2012-10-16T17:09:21 -!- CheBuzz is now known as Guest27481
2012-10-16T17:10:30 < Thorn> that means your phy doesn't respond to SMI commands
2012-10-16T17:10:30 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-16T17:11:40 < Posterdati> Thorn: wrong address?
2012-10-16T17:12:11 < Posterdati> Thorn: I'm going to perform search for address reading ID answer
2012-10-16T17:17:22 -!- dekar [~dekar@212.255.39.68] has joined ##stm32
2012-10-16T17:17:26 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T17:34:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-16T17:39:22 < Posterdati> Thorn: no PHY in address space 0-255
2012-10-16T17:49:52 < Posterdati> Thorn: I configured TXD0 and TXD1 as GPIO_Mode_AF_PP
2012-10-16T17:51:40 < Thorn> they have nothing to do with reading registers. mdc/mdio are used for that
2012-10-16T17:52:09 < Posterdati> mdc = clock and mdio = data
2012-10-16T17:52:22 < Posterdati> isn't that true?
2012-10-16T17:52:48 < Posterdati> it's like an i2c protocol
2012-10-16T17:53:02 < Posterdati> clock and bidirectional data
2012-10-16T17:56:12 < Posterdati> are they like I2C?
2012-10-16T17:57:53 < Thorn> yes pretty much
2012-10-16T17:58:28 < Thorn> http://en.wikipedia.org/wiki/Management_Data_Input/Output
2012-10-16T18:02:40 < Posterdati> but data is bidirectional, so what about mdio?
2012-10-16T18:03:54 < Thorn> what about it?
2012-10-16T18:04:27 < Posterdati> mdio is connected to a port that is is GPIO_Mode_AF_PP
2012-10-16T18:05:32 < Laurenceb> https://i.chzbgr.com/completestore/12/10/11/Q2QFcb-JskGgfEC5lLNg4Q2.jpg
2012-10-16T18:06:48 -!- dekar_ [~dekar@212.255.38.65] has joined ##stm32
2012-10-16T18:06:52 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-16T18:09:48 -!- dekar__ [~dekar@212.255.41.103] has joined ##stm32
2012-10-16T18:09:52 -!- mode/##stm32 [+v dekar__] by ChanServ
2012-10-16T18:10:14 -!- dekar [~dekar@212.255.39.68] has quit [Ping timeout: 240 seconds]
2012-10-16T18:10:35 -!- dekar [~dekar@212.255.42.8] has joined ##stm32
2012-10-16T18:10:38 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-16T18:13:44 -!- dekar [~dekar@212.255.42.8] has quit [Read error: Operation timed out]
2012-10-16T18:13:49 -!- dekar_ [~dekar@212.255.38.65] has quit [Ping timeout: 272 seconds]
2012-10-16T18:15:04 -!- dekar__ [~dekar@212.255.41.103] has quit [Ping timeout: 272 seconds]
2012-10-16T18:29:21 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T18:39:25 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-16T18:48:36 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Ping timeout: 248 seconds]
2012-10-16T18:50:55 -!- barthess [~barthess@178.154.41.127] has joined ##stm32
2012-10-16T18:54:05 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-16T19:14:58 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Ping timeout: 246 seconds]
2012-10-16T19:17:52 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T19:34:31 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Ping timeout: 246 seconds]
2012-10-16T20:21:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-16T20:45:30 < Posterdati> hi
2012-10-16T20:46:01 < Posterdati> is there anyone that make lwip works with ste101p? tx
2012-10-16T20:57:40 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T21:12:26 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-16T21:12:48 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-16T21:18:33 < TzacksT> hey guys, any idea or a piece of advice how to convert pcm data for sigma delta dac?
2012-10-16T21:18:33 -!- daku is now known as DaKu
2012-10-16T21:23:02 < jpa-> TzacksT: i would expect it to be just http://en.wikipedia.org/w/index.php?title=File:Block_Diagram_Delta-Sigma.svg&page=1  that block diagram implemented in software
2012-10-16T21:25:08 < TzacksT> i've seen many of these diagrams, i just dont get it somehow, how to convert 16bit word for the dac
2012-10-16T21:26:05 < TzacksT> http://www.beis.de/Elektronik/DeltaSigma/DeltaSigmaDBlockDiagram.GIF
2012-10-16T21:26:34 < TzacksT> that one is a bit better, it just says a bit stream out, i need 16bit words out, :/
2012-10-16T21:28:07 < zyp> huh?
2012-10-16T21:28:11 < emeb> TzacksT: what are you asking for? Most S/D DACs have some sort of standard parallel or serial interface.
2012-10-16T21:28:32 < zyp> a delta-sigma signal is by definition a bitstream, no?
2012-10-16T21:29:00 < emeb> if you're talking about the output.
2012-10-16T21:29:08 < emeb> the input is usually not bitstream.
2012-10-16T21:29:29 < TzacksT> this is lm4500 ac97 codec, it accepts data in frames, but all it says it uses sigma delta dacs, i was making NCO i need to use and the output was everything but not a sine wave and the code was ok
2012-10-16T21:29:31 < emeb> usually there are interpolation filters and the S/D modulator in the DAC.
2012-10-16T21:29:47 < zyp> emeb would know, he's the channel's resident audio guy
2012-10-16T21:30:00 < emeb> TzacksT: then you'll need to create an AC97 compliant serial data stream.
2012-10-16T21:30:14 < jpa-> TzacksT: i would expect the DAC to handle the conversion, your code was probably broken in some other way
2012-10-16T21:30:24 < emeb> some MCUs have hardware that can do that. STM32 doesn't.
2012-10-16T21:31:03 < emeb> It _may_ be possible to bit-bang an AC97 interface, but it's fairly fast and would likely take most of the processor bandwidth to do so.
2012-10-16T21:31:12 < TzacksT> i know but this is actually pretty bac because al the datasheet for lm4550 says is that is using sigma delta dacs/adcs and when i looked at the ac97 datasheet it says it uses 2's complement pcm audio
2012-10-16T21:31:50 < emeb> TzacksT: forget about sigma-delta. That's an implementation detail that's far removed from what your processor is dealing with.
2012-10-16T21:32:11 < TzacksT> jpa well i believe there was nothing to break on reading rom and sending the data to dac :)
2012-10-16T21:32:55 < TzacksT> emeb:  so you are saying that when i say eg max or min coded in 2's complement as it say i dont have to bother with sigma delta?
2012-10-16T21:33:22 < jpa-> TzacksT: how did you implement the AC97 serial format without hardware support?
2012-10-16T21:33:57 < jpa-> i guess one could use SPI if the processor was clocked from a N * 12.288 MHz crystal
2012-10-16T21:34:25 < TzacksT> i do this on fpga, i had "talkthrough" software i used, it was sampling data from mic and then lathicng it back do dacs, i used only the input part of it
2012-10-16T21:34:56 < emeb> TzacksT: you using a Digilent FPGA board w/ AC97 codec?
2012-10-16T21:35:06 < TzacksT> then i made numerically controlled oscillator that was simply reading data from LUT but the output was anything but sine :/
2012-10-16T21:35:10 < TzacksT> yes
2012-10-16T21:35:36 < jpa-> TzacksT: so did the loopback work properly?
2012-10-16T21:35:40 < TzacksT> yup
2012-10-16T21:35:45 < TzacksT> no prob at all
2012-10-16T21:35:53 < jpa-> and what did the data from mic look like?
2012-10-16T21:36:14 < TzacksT> i didnt sample it
2012-10-16T21:36:22 < TzacksT> i just used it with mic and speaker
2012-10-16T21:36:32 < jpa-> well do; it's the obvious way to debug your system
2012-10-16T21:36:37 < TzacksT> then i simply rerouted the inputs to my nco
2012-10-16T21:38:28 < TzacksT> jpa well i was expecting that when i send data from LUT to that codec it will work, but today i hooked it to scilloscope and it is totaly bad and i have no idea whats wrong, i will definitelly work on it as i have to but i needed to ask about that sigma delta if this was problem
2012-10-16T21:56:18 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Quit: Konversation terminated!]
2012-10-16T21:57:37 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T21:59:14 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-16T22:12:54 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Ping timeout: 245 seconds]
2012-10-16T22:16:25 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 246 seconds]
2012-10-16T22:18:39 < emeb> TzacksT: Things to check: 1) do you have signed/unsigned data format correctly set?
2012-10-16T22:19:09 < emeb> 2) Do you have your data properly framed (channel/bit alignment)?
2012-10-16T22:26:12 < TzacksT> i believe so, i will try to work on it tomorrow or so and will try to keep the latch that was in original design and will see what will happen
2012-10-16T22:29:18 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has joined ##stm32
2012-10-16T22:59:49 -!- barthess [~barthess@178.154.41.127] has quit [Quit: Leaving.]
2012-10-16T23:06:49 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Ping timeout: 256 seconds]
2012-10-16T23:08:46 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Connection reset by peer]
2012-10-16T23:15:03 -!- meloun [~meloun@188-192-251-37-dynip.superkabel.de] has joined ##stm32
2012-10-16T23:15:03 -!- meloun [~meloun@188-192-251-37-dynip.superkabel.de] has left ##stm32 []
2012-10-16T23:17:41 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-16T23:21:05 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-16T23:36:08 -!- DaKu is now known as daku
2012-10-16T23:37:52 < zyp> http://infocenter.arm.com/help/topic/com.arm.doc.dai0234a/index.html <- I find this a bit interesting
2012-10-16T23:38:06 < zyp> «Migrating from PIC Microcontrollers to Cortex-M3»
2012-10-16T23:38:50 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-16T23:39:21 < zyp> «It is common, when programming for PIC devices, to use NOP instructions as a way of consuming time. The execution time of NOP instructions is easily determined from the system clock speed. When developing for Cortex-M3 devices this cannot be relied on as the pipeline is free to “fold out” NOP instructions from the instruction stream. When this happens, they do not consume time at all.
2012-10-16T23:39:27 < zyp> Deterministic delays in Cortex-M3 systems must therefore make use of a timer peripheral.b
2012-10-16T23:39:35 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-10-16T23:40:17 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-10-16T23:40:50 < zyp> this should probably be linked to everybody still insisting on using nop-delays
2012-10-16T23:41:58 -!- mrnuke|laptop [~mrnuke@129.7.255.110] has quit [Ping timeout: 246 seconds]
2012-10-16T23:54:51 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
--- Day changed Wed Oct 17 2012
2012-10-17T00:06:05 < emeb> interesting.
2012-10-17T00:08:29 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-17T00:16:40 < emeb> I generally rely on DMA IRQs to keep me in sync w/ hardware.
2012-10-17T00:16:58 < emeb> about the only place I use NOP is in dumb time-wasters for test code (blinky lights, etc)
2012-10-17T00:18:52 < zyp> I already have a sleep function based on systick for stuff like that
2012-10-17T00:19:12 < zyp> which allows me to run other threads meanwhile
2012-10-17T00:20:28 < emeb> all part of laks?
2012-10-17T00:20:40 < zyp> sure
2012-10-17T00:21:00 < zyp> http://cgit.jvnv.net/laks/tree/os/time.h
2012-10-17T00:21:54 < zyp> it's a bit crude, but it works
2012-10-17T00:22:08 < zyp> I'm currently working out some better scheduling code to replace it
2012-10-17T00:22:11 < emeb> all you need for simple stuff.
2012-10-17T00:26:33 < zyp> precision is not so good since it only operates with milliseconds
2012-10-17T00:30:14 < emeb> have you run into situations where that hurt you?
2012-10-17T00:35:54 < zyp> not directly hurt, but I've had situations where a millisecond would be too long
2012-10-17T00:36:04 < zyp> http://cgit.jvnv.net/suzumebachi/tree/main.cpp#n75 <- like bitbanging jtag
2012-10-17T00:58:34 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-17T01:09:36 < emeb> I suppose for finer granularity you could sit & spin on a faster timer dedicated to the task.
2012-10-17T01:12:44 < zyp> yep
2012-10-17T01:15:09 < zyp> I've concluded that 32 bits is too low resolution for absolute timestamps, and by going with 64-bit timestamps instead, I can use cpu frequency as time base without rollover being an issue ever
2012-10-17T01:16:23 < zyp> even at 168 MHz, a 64-bit time base will spend 3500 years or so before rolling over
2012-10-17T01:19:31 < zyp> so I'm intending to have a simple function for just busywaiting an amount of cycles, for small delays that doesn't justify switching to another thread, and a sleep-until function that puts the thread to sleep until the time arrives
2012-10-17T01:20:42 < zyp> which again can be used to implement normal sleeps (now + delay) or intervals (last + interval)
2012-10-17T01:20:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-17T01:23:35 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Quit: TzacksT]
2012-10-17T01:46:01 < dongs> 12h of sleep is nice
2012-10-17T01:46:33 < qyx_> sleepy dongs
2012-10-17T01:46:52 < emeb> busywait or yield?
2012-10-17T01:54:04 < zyp> wish I had time for that
2012-10-17T01:54:21 < zyp> I'm planning to be at work within 8
2012-10-17T02:09:02 < dongs> heh
2012-10-17T02:25:30 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 276 seconds]
2012-10-17T02:31:04 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
2012-10-17T02:31:26 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T02:32:20 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T02:36:33 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 276 seconds]
2012-10-17T02:39:35 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T02:39:39 < Posterdati> hi
2012-10-17T02:40:00 < Posterdati> is there anyone that was able to make stm32 works with ste101p?
2012-10-17T02:40:14 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 255 seconds]
2012-10-17T03:03:21 < dongs> is that a ethernet phy
2012-10-17T03:03:35 < dongs> oh
2012-10-17T03:04:41 -!- mrcan_ [~mrcan@78.173.243.11] has joined ##stm32
2012-10-17T03:04:42 -!- mrcan_ [~mrcan@78.173.243.11] has quit [Changing host]
2012-10-17T03:04:42 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-17T03:06:11 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 255 seconds]
2012-10-17T03:09:08 < qyx_> hm, i am just configuring that nuttx thingie
2012-10-17T03:09:30 < qyx_> seems that i will like it
2012-10-17T03:18:36 < Posterdati> dongs: yes a damned phy :)
2012-10-17T03:19:07 < dongs> cool. i didnt notice m scope had firmware from 2012
2012-10-17T03:19:10 < dongs> upgraded, shits a lot faster now
2012-10-17T03:19:17 < dongs> they must have switched to lunix 3.0
2012-10-17T03:20:13 < Posterdati> dongs: I've got an stm32-p107 with lwip on it, but I'm not able to make the ste101p works with it :(
2012-10-17T03:20:43 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-17T03:20:55 < dongs> surely if they sell it it works
2012-10-17T03:21:34 < Posterdati> yes
2012-10-17T03:21:45 < Posterdati> maybe they could tell me how it works
2012-10-17T03:22:17 < dongs> did ytou look at schematic?
2012-10-17T03:22:22 < dongs> maybe theres some gpio to power it up / enable it?
2012-10-17T03:22:29 < dongs> i.e. something like nRESET or whatever.
2012-10-17T03:22:41 < Posterdati> not all sold things worked, think about pentium 5
2012-10-17T03:23:12 < Posterdati> dongs: it seems that mdio/mdc interface won't work
2012-10-17T03:24:09 < qyx_> Posterdati: stm32-p107 worked out-of-box
2012-10-17T03:24:10 < qyx_> rev B
2012-10-17T03:24:19 < Posterdati> rev A
2012-10-17T03:24:19 < qyx_> but with chibios
2012-10-17T03:24:28 < qyx_> i don't have revA
2012-10-17T03:24:33 < Posterdati> rev B has got a different PHY
2012-10-17T03:24:46 < qyx_> actually chibios has example for rev A
2012-10-17T03:24:58 < qyx_> i changed some defines for new phy
2012-10-17T03:25:07 < qyx_> and also some remaps
2012-10-17T03:25:09 < Posterdati> where?
2012-10-17T03:25:32 < qyx_> http://www.chibios.org/dokuwiki/doku.php?id=chibios:download
2012-10-17T03:25:59 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-17T03:26:14 < qyx_> https://chibios.svn.sourceforge.net/svnroot/chibios/branches/stable_2.4.x/demos/ARMCM3-STM32F107-LWIP/
2012-10-17T03:27:13 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-17T03:32:19 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 260 seconds]
2012-10-17T03:33:23 < Posterdati> qyx_: is not simple to understand the code in chibios...
2012-10-17T03:33:48 < qyx_> you have to use its HAL + mac drivers
2012-10-17T03:33:58 < qyx_> and there are prepared bindings for lwIP
2012-10-17T03:34:31 < dongs> http://imgur.com/a/wBsn8 gaming rigs collection
2012-10-17T03:34:53 < qyx_> the whole chibi kernel is smaller than stdperiph lib
2012-10-17T03:35:45 < dongs> qyx_: too bad the API / syntax is fucking horrible
2012-10-17T03:35:47 < Posterdati> I've got my RTOS
2012-10-17T03:36:31 < qyx_> the api is uhm.. unusual
2012-10-17T03:36:35 < qyx_> but working
2012-10-17T03:37:11 < qyx_> and personally i hate camelCase
2012-10-17T03:43:16 < BrainDamage> dongs' home, third pic from the top http://imgur.com/a/wBsn8#GN76C
2012-10-17T03:44:25 < emeb> those are some serious hacker caves.
2012-10-17T03:45:50 < dongs> almost as good as http://i.imgur.com/dubTU.jpg
2012-10-17T04:16:32 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-17T04:16:32 -!- mrnuke|laptop [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Client Quit]
2012-10-17T04:20:29 -!- Netsplit *.net <-> *.split quits: Count_Niedar, esden
2012-10-17T04:21:10 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T04:21:48 -!- Netsplit over, joins: esden, Count_Niedar
2012-10-17T04:22:24 < emeb> Yes, Keanu?
2012-10-17T04:23:27 < emeb> lotta folks here seem to like it.
2012-10-17T04:23:53 < dongs> FUCK CHIBIOS; CHIBIOS SUX; CHIBIOS IS DYING; CHIBIOS IS DEAD TO ME; CHIBIOS HIT WTC
2012-10-17T04:24:34 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-10-17T04:30:15 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-17T04:32:10 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-17T04:36:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-17T04:50:45 < gnomad> dongs: so what RTOS *do* you like for the ST chips?
2012-10-17T04:51:12 < dongs> gnomad: coos seems ok, API is not aids, and it stays out of hte way
2012-10-17T04:52:11 < gnomad> how is coos peripheral support for the ST chips?
2012-10-17T04:52:22 < dongs> gnomad: thats the "stays out of the way" partr
2012-10-17T04:52:29 < dongs> i dont WANT my OS to support peripherals
2012-10-17T04:52:42 < dongs> so im not forced to their ridiculous hal/whatever
2012-10-17T04:52:42 < gnomad> As in, you talk to the peripherals directly?
2012-10-17T04:52:44 < dongs> and just do what i want
2012-10-17T04:52:45 < dongs> yes
2012-10-17T04:53:07 < gnomad> So you go right to the hardware and not use any libraries?
2012-10-17T04:53:25 < dongs> stdperiphlib, syre
2012-10-17T04:53:25 < dongs> sure
2012-10-17T04:53:30 < gnomad> That's how I do things in the MSP430 realm...
2012-10-17T04:54:01 < gnomad> I don't know if I'd mind a decent, non-buggy library, but those seem not to exist...
2012-10-17T04:54:33 < gnomad> OTOH, I have had difficulty locating code samples to talk to any of the ARM peripherals directly.
2012-10-17T04:55:25 < gnomad> TI has a really nice set of sample code for all the MSP430 peripherals... I long for something similar for any ARM chip...
2012-10-17T04:55:47 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-17T04:57:11 < dongs> stdperiph is fine
2012-10-17T04:57:32 < zippe> stdperiph is a festering heap of post-research monkeys
2012-10-17T04:58:02 < dongs> yes but it works
2012-10-17T04:58:06 < dongs> and stays out of hte way
2012-10-17T05:00:18 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-17T05:09:31 < dongs> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/DM00048919.pdf
2012-10-17T05:10:46 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has quit [Read error: Connection reset by peer]
2012-10-17T05:12:12 < dongs> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/DM00051031.pdf totally useful
2012-10-17T05:14:33 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has joined ##stm32
2012-10-17T05:22:20 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-17T05:22:21 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-17T05:22:21 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-17T05:57:49 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-17T06:11:35 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-17T06:11:41 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-17T06:54:41 < dongs> http://blipbox.org/blog/ ooOOooo
2012-10-17T06:58:39 < cjbaird> The words "art installation" was omitted. I'll have to email the author and inform gim.
2012-10-17T07:00:52 < dongs> oh BORING
2012-10-17T07:00:57 < dongs> they'rte using resistive touchscreen
2012-10-17T07:01:08 < dongs> I thought they were doing the hax where you use a led as light sensor
2012-10-17T07:01:20 < dongs> and scan it while off-duty
2012-10-17T07:04:53 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-17T07:21:55 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 246 seconds]
2012-10-17T07:32:24 < dongs> http://www.silabs.com/products/mcu/Pages/Web-Drawing-for-SiM3L1xx-Software-Tools-Flipcard.aspx
2012-10-17T07:33:18 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-17T08:23:19 < emeb_mac> dongs: you 'sposed to put one of those LED bypasses per LED? seems like that could get painful.
2012-10-17T08:23:28 < dongs> emeb_mac: yeah lol
2012-10-17T08:23:36 < dongs> more like EXPENSIVE
2012-10-17T08:23:40 < dongs> not sure what they're targeting
2012-10-17T08:24:05 < emeb_mac> datasheet said tail lights, traffic lights.
2012-10-17T08:24:30 < emeb_mac> makes sense for hi-rel, but you'd think it would make more sense to bypass multiple LEDs.
2012-10-17T08:25:01 < emeb_mac> so on bypass per N LEDs - you don't lose the whole chain, but you don't need one-per.
2012-10-17T08:26:19 < dongs> well, in high brightness leds, usually you do have a chain
2012-10-17T08:26:30 < dongs> like 6 leds for 24V system etc
2012-10-17T08:28:53 < emeb_mac> we have LED traffic lights - many seem to have reliability issues in the heat (120F sometimes)
2012-10-17T08:29:17 < emeb_mac> whole sections of the array go dark, or even more interesting start to flash intermittently.
2012-10-17T08:31:38 < dongs> flash = maybe power supply fail out?
2012-10-17T08:31:41 < dongs> overheat + trip or whatever
2012-10-17T08:32:54 < emeb_mac> yeah - no idea exactly what's happening.
2012-10-17T08:33:23 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-17T08:33:25 < emeb_mac> thinking some sort of heat-related relaxation oscillation. heats up, disconnects, cools down reconnects, repeat.
2012-10-17T09:02:44 < dongs> http://i.imgur.com/ZT3Ry.png WHY THE FUCK NOONE TO)LD ME PAL BLACK IS FUCKING 0V
2012-10-17T09:06:51 < emeb_mac> wait, what?
2012-10-17T09:08:41 < dongs> fuck.
2012-10-17T09:08:52 < dongs> i waste all this time getting black level to 50mv or wahtever
2012-10-17T09:09:08 < dongs> and apparently pal black is at 0v
2012-10-17T09:09:10 < dongs> just like sync
2012-10-17T09:09:25 < emeb_mac> ah - europeans gotta be different.
2012-10-17T09:10:51 < emeb_mac> screw all this analog video crap. Just do HDMI. :P
2012-10-17T09:11:56 < dongs> ugh now i gotta figure out how to switch that shit
2012-10-17T09:12:34 < emeb_mac> how you generating the levels?
2012-10-17T09:12:44 < cjbaird> I bought tea bags instead of leaf tea. rage.
2012-10-17T09:12:57 < dongs> emeb_mac: a mega haxy circuit
2012-10-17T09:13:21 < emeb_mac> cjbaird: I feel bad for you son.
2012-10-17T09:13:54 < dongs> http://i.imgur.com/R4weO.png
2012-10-17T09:14:21 < dongs> im surprised it even works
2012-10-17T09:14:57 < dongs> now i need to poke my analol'g guy to see how to fix this shit
2012-10-17T09:15:25 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 256 seconds]
2012-10-17T09:15:31 < emeb_mac> dongs: so OSD_BW is a 0v/3.3v signal?
2012-10-17T09:15:51 < dongs> yea
2012-10-17T09:16:41 < cjbaird> Hmm. I'm seriously going to return to the supermarket and bludge an exchange..
2012-10-17T09:16:47 < cjbaird> tea bags. ffs.
2012-10-17T09:17:11 < emeb_mac> ok - so when 0V, the D5 drop turns that to ~+0.7 and when 3.3V the D4+RA1||R11 gives a white level tunable w/ RA11
2012-10-17T09:17:23 < dongs> the diode does somethingwith black
2012-10-17T09:18:26 < dongs> something about reverse voltage or someshit maybe  i forget
2012-10-17T09:18:33 < emeb_mac> dongs: yeah - D5 fwd drop sets the black level.
2012-10-17T09:20:02 < emeb_mac> if you want the black level to be lower, then use a different diode (schottky) or get rid of it entirely.
2012-10-17T09:20:36 < dongs> just short it?
2012-10-17T09:20:43 < dongs> or take out
2012-10-17T09:21:08 < emeb_mac> yeah - short it would pull the black level down to whatever the Vol of the OSD_BW source is.
2012-10-17T09:22:07 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-17T09:22:07 < dongs> ugh
2012-10-17T09:22:21 < emeb_mac> but cutting it out (so the ckt is open when OSD_BW = 0v) wouldn't work - R16 would pull up to VCC then.
2012-10-17T09:25:02 < dongs> ill have the palfag try taking the diode and shorting it.
2012-10-17T09:25:15 < dongs> diode is already schottky
2012-10-17T09:25:56 < emeb_mac> ah - yeah, should work when shorted. black level will be a bit lower.
2012-10-17T09:26:11 < dongs> well, what did you mean by "get rid of it entirely"
2012-10-17T09:26:20 < dongs> that meant = shorting it?
2012-10-17T09:26:23 < emeb_mac> shorting it out.
2012-10-17T09:26:25 < dongs> right.
2012-10-17T09:27:41 < emeb_mac> so you were getting gray background instead of black?
2012-10-17T09:28:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-17T09:28:35 < enots> yep
2012-10-17T09:29:18 < dongs> before I added that hax video switcher I was just pulling video signal high or low
2012-10-17T09:29:21 < dongs> the ghetto way
2012-10-17T09:30:05 < dongs> so doing too much black
2012-10-17T09:30:10 < dongs> would darken entier monitor
2012-10-17T09:30:14 < emeb_mac> probably hitting the white level pretty hard.
2012-10-17T09:30:20 < dongs> and white yeah
2012-10-17T09:31:45 < emeb_mac> well, shorting D5 will revert to the same black level you had prior.
2012-10-17T09:32:03 < emeb_mac> but maybe that won't darken the whole monitor w/ pal signal.
2012-10-17T09:32:15 < dongs> well, since pals black is 0 right,
2012-10-17T09:32:20 < emeb_mac> yeah.
2012-10-17T09:33:13 < emeb_mac> the "darken entire monitor" may have been the strong white driving the video agc down though.
2012-10-17T09:33:35 < dongs> yeah, it was. sorry, i just remembered that was the reason for pot/etc
2012-10-17T09:33:51 < dongs> and analog switch
2012-10-17T09:34:01 < dongs> cause if i turn brightness pot too high it will do same shit now too
2012-10-17T09:37:43 < emeb_mac> makes sense.
2012-10-17T09:37:59 < emeb_mac> so try shorting D5 - maybe you'll get a better black.
2012-10-17T09:38:39 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 260 seconds]
2012-10-17T09:43:50 < dongs> that's racist
2012-10-17T09:44:08 < dongs> goprpo hero3 info released in 16 minutes
2012-10-17T09:45:03 < scrts> moar overprice!
2012-10-17T09:45:24 < scrts> they could finally release uHD resolution camera... :>
2012-10-17T09:45:26 < dongs> yeah, its like $399 now
2012-10-17T09:45:35 < dongs> scrts: 4k @ 15fps
2012-10-17T09:45:39 < dongs> appareantly
2012-10-17T09:45:41 < scrts> announced?
2012-10-17T09:45:48 < dongs> according to some presentatio nslide
2012-10-17T09:45:56 < scrts> kinda slow... but still nice
2012-10-17T09:46:07 < dongs> 12mp @ 30fps or someshit.
2012-10-17T09:46:16 < dongs> but i think only for 1 s econds.
2012-10-17T09:46:17 < scrts> next summer - definitely skydiving one!
2012-10-17T09:46:18 < scrts> :)
2012-10-17T09:46:36 < scrts> I believe the slowest part becomes writing to flash card
2012-10-17T10:02:00 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-17T10:05:03 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-10-17T10:07:39 < dongs> http://gopro.com/cameras/hd-hero3-black-edition/#specs
2012-10-17T10:08:06 < dongs> ohh 240fps
2012-10-17T10:09:05 < emeb_mac> did you see the one where the gopro came off the guy's helmet during a jump & fell all the way, landed and still worked?
2012-10-17T10:11:17 < dongs> yeah, totally bullshit
2012-10-17T10:11:31 < emeb_mac> oh?
2012-10-17T10:11:45 < dongs> i've got a bettery story with gopro falling out of hte sky with my quad from about 100 meters up  and landing flat on its lens, the BGA inside ripped itself off the pcb
2012-10-17T10:11:58 < dongs> bettery=better
2012-10-17T10:12:20 < emeb_mac> what did it land on?
2012-10-17T10:12:22 < dongs> of course, it landed on pavement.
2012-10-17T10:12:41 < emeb_mac> ah - g-shock a little harsher than landing in tall grass.
2012-10-17T10:13:09 < emeb_mac> shoulda put rubber bumpers on or something.
2012-10-17T10:13:53 < zyp> sounds strange that the bga would even have enough mass to rip itself off
2012-10-17T10:14:13 < emeb_mac> strange things happen at high-g
2012-10-17T10:14:44  * emeb_mac used to work on hi-rel flight hardware.
2012-10-17T10:15:05 < emeb_mac> shock & vibe tests were scary times.
2012-10-17T10:17:00 < scrts> 400USD....
2012-10-17T10:17:25 < dongs> i need to offload my new hero2 that i havent evne taken out of hte box yet
2012-10-17T10:26:21 < emeb_mac> neophilia.
2012-10-17T10:29:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-17T10:30:27 -!- claude is now known as claude_work
2012-10-17T10:39:53 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-17T10:44:03 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 276 seconds]
2012-10-17T10:44:27 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-17T10:47:34 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-17T10:57:14 < jpa-> stupid ld is linking non-thumb libgcc.a for me
2012-10-17T10:57:33 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-17T10:59:37 < jpa-> hmm.. or where the heck does __aeabi_ldivmod come from
2012-10-17T11:02:41 < zyp> it does come from libgcc
2012-10-17T11:03:23 < jpa-> ah bah, my bad
2012-10-17T11:04:23 < jpa-> dealing with codesourcery lacking hardfp libs, i had set -mfpu=.. with -mfloat-abi=soft, and arm gcc built lacked libgcc for that particular configuration so it just used the default
2012-10-17T11:05:50 < zyp> «can't find a valid library variant, so let's just silently substitute an invalid one and hope noone notices»
2012-10-17T11:08:33 < jpa-> yeah :)
2012-10-17T11:12:41 < scrts> ooooh flybacknuck here!
2012-10-17T11:12:44 < scrts> :)))
2012-10-17T11:41:23 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-17T11:45:40 -!- SuicideFunky [~randy@vps-212-153-70-76.twenty-five.nl] has quit [Ping timeout: 260 seconds]
2012-10-17T11:57:56 < Posterdati> hi
2012-10-17T12:06:34 < dongs> accelerated dongs
2012-10-17T12:14:50 -!- Tectu is now known as tectu
2012-10-17T12:16:13 < Posterdati> dongs: hi!
2012-10-17T12:16:25 < dongs> sup
2012-10-17T12:16:30 < dongs> tectu: stop being kawaii with all those nick changes
2012-10-17T12:17:07 < tectu> dongle, hu?
2012-10-17T12:20:39 < dongs> relevant: https://sphotos-a.xx.fbcdn.net/hphotos-ash3/577052_10151495146148298_1885388048_n.jpg
2012-10-17T12:47:38 < Laurenceb> jpa-: huh
2012-10-17T12:47:48 < Laurenceb> codesourcery works for me with hardfloat
2012-10-17T12:50:40 < dongs> i'm hard for your float
2012-10-17T12:52:06 < cjbaird> Mantissa.. more like Man-Tissa..
2012-10-17T12:53:31 < Laurenceb> no results for electroejaculation on instructibles
2012-10-17T12:53:33 < Laurenceb> wtf
2012-10-17T12:54:18 < cjbaird> That Armpit Scheme could probably work on the stm32l-disco. I wouldn't care to do much Original Work for it, though..
2012-10-17T12:54:32 < Laurenceb> armpit scheme?
2012-10-17T12:55:03 < cjbaird> dongs' little discovery the other day: http://armpit.sourceforge.net/
2012-10-17T12:55:40 < Laurenceb> wtf
2012-10-17T12:56:00 < Laurenceb> WHYYY
2012-10-17T12:58:05 < cjbaird> Functional programming on bare-metal is almost a free after-hours pass to the Paul Graham boudoir
2012-10-17T12:58:36 < cjbaird> "I HAS THIS NEW WEBSIGHT" "IT RUNS ON LISP!" "I NEEDS 150 MILLION!
2012-10-17T12:58:44 < cjbaird> ""I'LL BE BIGGER THAN TWITTER!"
2012-10-17T12:59:58 < cjbaird> Always amusing when those Y-Comp startups become plain C/C++/Java codemonkey operations when they actually take off..
2012-10-17T13:09:57 < dongs> ruby on fales
2012-10-17T13:10:51 < jpa-> Laurenceb: but does it now come with hardfloat newlib?
2012-10-17T13:11:08 < Laurenceb> ive forgotten sorry
2012-10-17T13:11:16 < jpa-> yeah.. i think mine did not :)
2012-10-17T13:27:06 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-17T13:30:20 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 255 seconds]
2012-10-17T13:43:13 -!- SuicideFunky [~randy@vps-212-153-70-8.twenty-five.nl] has joined ##stm32
2012-10-17T13:46:23 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-17T14:13:19 < Laurenceb> attn dongs
2012-10-17T14:13:22 < Laurenceb> http://racerelations.about.com/b/2011/08/11/new-research-finds-jews-have-african-ancestry.htm
2012-10-17T14:22:56 < dongs> lo
2012-10-17T14:28:05 < cjbaird> "Bad News"
2012-10-17T14:31:45 < Posterdati> cjbaird: are you involved in porting Common Lisp on ARM?
2012-10-17T14:32:06 < dongs> ahah.
2012-10-17T14:32:41 < dongs> we got a live one!
2012-10-17T14:34:47 < Posterdati> where?
2012-10-17T14:35:08 < cjbaird> Posterdati: no, just an interest in it.
2012-10-17T14:35:31 < dongs> sorry just laughign cuz i thought cjbeard would be the only loon interested in lisp in 2012
2012-10-17T14:36:37 < Posterdati> dongs: Common Lisp is the most advanced programming language
2012-10-17T14:36:52 < dongs> in 1989 maybe
2012-10-17T14:36:57 < cjbaird> I'm a FORTHfag, not lisp.
2012-10-17T14:38:10 < zyp> dongs, I like how you are bashing both old stuff like lisp and new stuff like python
2012-10-17T14:38:25 < dongs> ya, in my view they're both just as worthless
2012-10-17T14:38:27 < dongs> python even more so
2012-10-17T14:38:40 < dongs> whlie lisp could have had a purpose 2 decades ago, python serves absolutely NO purpose.
2012-10-17T14:38:50 < dongs> it does nothing that perl or any existing interpreted language have done
2012-10-17T14:39:01 < Posterdati> dongs: you're wrong, python has got less features of Lisp
2012-10-17T14:39:02 < dongs> but its just another newfag shit that noone cares about
2012-10-17T14:39:07 < cjbaird> Python's purpose: to kill of the last of BASIC.
2012-10-17T14:39:23 < dongs> cjbaird: i'm sorry, i'd take basic over any python shit.
2012-10-17T14:39:33 < dongs> if its goal is to kill off basic, its failing hard.
2012-10-17T14:47:50 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-17T14:48:40  * HTT-Bird drops dongs into a Prolog interpreter
2012-10-17T14:48:49 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-17T15:05:59 < dongs> Laurenceb stm32logger clone complete
2012-10-17T15:08:50 < dongs> http://i.imgur.com/vTIaw.png also stolen thorn's diode symbols
2012-10-17T15:08:57 < dongs> er, led rather
2012-10-17T15:10:28 < zyp> that's so hardcore
2012-10-17T15:10:40 < dongs> :( i kno rite.
2012-10-17T15:11:54 < dongs> bedtime
2012-10-17T15:12:24 < zyp> sounds early
2012-10-17T15:12:46 < zyp> no wonder you get 12 hours of sleep
2012-10-17T15:46:31 < Laurenceb> nice
2012-10-17T15:46:42 < Laurenceb> whats the soic thingy for?
2012-10-17T15:46:47 < Laurenceb> oh hes gone
2012-10-17T15:52:45 < Laurenceb> and why all the 3 pin headers
2012-10-17T16:02:48 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-17T16:10:58 < Laurenceb> some kind of crazy rc pwm logger?
2012-10-17T16:13:54 < Laurenceb> http://cdn.memegenerator.net/instances/400x/10639358.jpg
2012-10-17T16:24:42 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-17T16:39:01 < Thorn> 144-pin f4 package doesn't make routing any easier
2012-10-17T16:39:28 < Thorn> FSMC and RMII pins are still all over the chip and there's almost no remapping
2012-10-17T16:44:42 < jpa-> yeah, no remapping for the big peripherals
2012-10-17T16:45:17 < scrts> found very cheap ethernet PHY
2012-10-17T16:45:22 < scrts> KSZ9031
2012-10-17T16:45:28 < scrts> wrong, KSZ8031
2012-10-17T16:45:36 < scrts> will try to use with STM32F207 :>
2012-10-17T16:46:21 < scrts> 24QFN is also nice
2012-10-17T16:47:26 < Thorn> why not f407?
2012-10-17T16:51:12 < zyp> jpa-, I think it has more to do with fast than big
2012-10-17T16:53:11 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-10-17T16:53:30 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-17T16:55:38 < Thorn> so I have lots and lots of crossing traces, especially fsmc
2012-10-17T16:55:59 < jpa-> zyp: true
2012-10-17T16:56:22 < zyp> Thorn, got a pic of how bad it is?
2012-10-17T16:57:13 < jpa-> bah.. atollic gdb server gives SIGTRAP on some important interrupts.. i can use "handle SIGTRAP noprint" in gdb, but it causes context switches to slow down 1000x
2012-10-17T16:58:02 < zyp> ouch
2012-10-17T16:59:12 < scrts> Thorn: not needed, plus 207 is cheaper afaik
2012-10-17T16:59:24 < scrts> we'll also use 176BGA
2012-10-17T17:05:16 < Laurenceb> https://github.com/steve-bate/ChibiOS-RPi
2012-10-17T17:05:36 < zyp> Laurenceb, old news
2012-10-17T17:06:15 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-17T17:17:29 < Thorn> zyp: will post picture in 10 mins
2012-10-17T17:26:19 < Thorn> http://i.imgur.com/dMMqx.png
2012-10-17T17:27:48 < Thorn> not everything is yet connected (as usual I'm doing schematics and layout in parallel lol). there should be a usb A connector under the B one, ~12 signals for the display etc. 4-way pin header plays a Wolfson I2C codec.
2012-10-17T17:27:57 < Thorn> *I2S
2012-10-17T17:29:24 < zyp> what's U6? NAND?
2012-10-17T17:29:47 < Thorn> nand, connected to fsmc
2012-10-17T17:30:13 < zyp> yeah, that's what I figured
2012-10-17T17:30:46 < Thorn> placing the ethernet PHY on a differemt side ws a bright idea, made routing much better
2012-10-17T17:31:02 < zyp> probably makes soldering fun
2012-10-17T17:31:03 < zyp> :p
2012-10-17T17:31:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-17T17:31:20 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-17T17:32:04 < Thorn> so the question is, will this crap work?
2012-10-17T17:32:39 < Thorn> more crap will be added, so I don't even know if I manage to stitch together something resembling a ground plane
2012-10-17T17:33:04 < zyp> you should stop being cheap and do four layers :p
2012-10-17T17:34:28 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-17T17:34:40 -!- |sark [~|sark@109.235.51.202] has quit [Remote host closed the connection]
2012-10-17T17:34:42 < Thorn> ...
2012-10-17T17:35:08 < Thorn> the idea was to do this board together with two previous ones to save setup costs
2012-10-17T17:35:16 < Thorn> in 2 layers of course
2012-10-17T17:36:31 < zyp> your routing seems very top-heavy
2012-10-17T17:36:38 < zyp> long tracks on top, short tracks on bottom
2012-10-17T17:37:09 < emeb_mac> Thorn: what's this do? some sort of net-connected audio player?
2012-10-17T17:37:25 < Thorn> zyp: deliberately
2012-10-17T17:37:35 < Thorn> emeb_mac: yes something like that
2012-10-17T17:37:52 < emeb_mac> cool
2012-10-17T17:38:08 < zyp> Thorn, well yeah, I figure less routing on bottom makes for a better ground plane
2012-10-17T17:38:12 < Thorn> will be cool if it works
2012-10-17T17:38:25 < Thorn> possibly makes for a ground plane lol
2012-10-17T17:38:45 < zyp> I still prefer not having to care about that :p
2012-10-17T17:39:02 < emeb_mac> Main thing I'd suggest is keep the analog vdd on a separate regulator.
2012-10-17T17:39:22 < emeb_mac> supply ripple blows through the codec, gets on the audio.
2012-10-17T17:39:42 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-17T17:39:53 < Thorn> that's being planned. also separate analog ground (somewhere under the display)
2012-10-17T17:40:13 < Thorn> I haven't done analog schematics yet though
2012-10-17T17:40:33 < emeb_mac> did you decide on the output buffers?
2012-10-17T17:40:51 < emeb_mac> going w/ Cmoy or just datasheet circuit?
2012-10-17T17:40:59 < Thorn> probably some dual supply op amp that audioph00ls love
2012-10-17T17:41:05 < Thorn> I think I'll power this thing with +/- 9 or so volts
2012-10-17T17:41:09 < emeb_mac> heh
2012-10-17T17:41:24 < Thorn> not sure if good idea
2012-10-17T17:41:55 < emeb_mac> leave the door open for the simpler approach.
2012-10-17T17:42:59 < Thorn> one question is volume control. I looked at cmoy and it has gain of 10 or so, with pots on the input
2012-10-17T17:43:36 < Thorn> not sure if pots are a good idea here (no ability to control volume remotely)
2012-10-17T17:44:12 < Thorn> some codecs seem to have digital volume control, there's also a possibility to do it in software
2012-10-17T17:45:12 < emeb_mac> it's a philosophical choice.
2012-10-17T17:45:45 < emeb_mac> some folks like analog pots for volume. No zipper noise, works even if the MCU locks up.
2012-10-17T17:45:50 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T17:46:34 < emeb_mac> using the codec gain controls is fairly easy, but usually kind of coarse resolution.
2012-10-17T17:47:08 < emeb_mac> and yeah - scaling the data in the output path is also easy, very controllable.
2012-10-17T17:47:10 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-10-17T17:47:18 < emeb_mac> personally I wouldn't bother with pots.
2012-10-17T17:47:50 < Thorn> mcu locks up = no source for audio data
2012-10-17T17:48:52 < emeb_mac> ah - but DMA can still be running, sending I2S data (usually a short loop)
2012-10-17T17:49:01 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-10-17T17:49:05 < emeb_mac> sounds bad - ask me how I know. :)
2012-10-17T17:49:33 < Thorn> heh, I think I had that on my PC before
2012-10-17T17:50:30 < emeb_mac> bbiab...
2012-10-17T17:50:35 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-17T18:09:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-17T18:15:23 < zyp> hmm, lpc4300 doesn't seem to be much better with regards to rmii locations
2012-10-17T18:15:37 < zyp> at least on lbga256 they are all over the chip
2012-10-17T18:21:45 -!- claude_work is now known as claude
2012-10-17T18:26:55 < Thorn> who said layout for fpga was easier.
2012-10-17T18:27:32 <+Steffanx> Who
2012-10-17T18:27:52 < Thorn> s/easier/more difficult/
2012-10-17T18:29:21 < jpa-> i think it is easier?
2012-10-17T18:29:51 < zyp> you're much more free to rearrange pins
2012-10-17T18:39:49 -!- barthess [~barthess@178.154.19.206] has joined ##stm32
2012-10-17T18:42:04 < Thorn> I'm selecting 10uF tantalum caps and ESR of most is 2-3 Ohm. isn't that a lot?
2012-10-17T18:42:26 < zyp> any ideas what I should put on this lpc4300 board?
2012-10-17T18:42:28 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-17T18:43:08 < zyp> or break out
2012-10-17T18:43:34 < Thorn> zyp: EVERYTHING
2012-10-17T18:44:24 < zyp> well, I'm thinking «as little as possible, because I'm not sure I'm capable of soldering this without fucking up, and I don't want to waste a ton of parts if I do :p
2012-10-17T18:45:37 < emeb> Thorn: why tantalums for 10uf - you can get decent 10uf MLCCs
2012-10-17T18:46:08 < Thorn> yea thinking about that too
2012-10-17T18:46:26 < zyp> http://bin.jvnv.net/f/NtfrV.png <- so far I've put on swd connector and usb for internal hs phy, and also connector for waveshare rmii phy board
2012-10-17T18:46:55 < zyp> so I've got hs usb and ethernet covered with minimal cost
2012-10-17T18:47:28 < BrainDamage> a peltier below the main board
2012-10-17T18:47:37 < BrainDamage> to reflow the balls when it warps
2012-10-17T18:48:02 < jpa-> Thorn: hmm.. the last tantalums i put on a project were 470µF 0.1ohm
2012-10-17T18:49:18 < Thorn> that's probably in the same ballpark looking at a datasheet
2012-10-17T19:05:34 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-17T19:26:28 < Laurenceb> lol braindamage
2012-10-17T19:28:35 < scrts> zyp -> I like to have devkit loaded as much as possible :)
2012-10-17T19:28:42  * scrts too lazy to solder something...
2012-10-17T19:28:59 < scrts> especially if that addon required 16bit data, 12bits address, etc...
2012-10-17T19:29:06 < scrts> *requires
2012-10-17T19:30:22 < zyp> well, this is mostly an experiment to see if I'm capable of soldering bga :p
2012-10-17T19:30:35 < zyp> and if it works afterwards then I have a nice toy
2012-10-17T19:31:14 < BrainDamage> I'm probably stating the obvious, but make sure to include a diagnostic to strobe the pins to check for contacts
2012-10-17T19:31:57 < BrainDamage> getting all balls to touch at the same time can be problematic, warped boards, thermal stresses, etc
2012-10-17T19:32:47 < BrainDamage> and can your process do tented vias? because vias next to balls will get them wicked in
2012-10-17T19:33:21 < zyp> what boardhouse doesn't do tented vias?
2012-10-17T19:33:31 < BrainDamage> some cheap chinese
2012-10-17T19:38:55 -!- barthess [~barthess@178.154.19.206] has quit [Quit: Leaving.]
2012-10-17T19:42:36 < Laurenceb> lots of places have issues with exposed copper at the corners
2012-10-17T19:42:40 < Laurenceb> even pcbtrain
2012-10-17T19:42:43 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T19:43:25 < Laurenceb> bbl
2012-10-17T19:43:58 < jpa-> the board house we use at work doesn't do tented vias for proto boards, only for production runs
2012-10-17T19:44:14 < jpa-> which kind of sucks for bga :P
2012-10-17T19:44:45 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds]
2012-10-17T19:45:20 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-17T19:49:49 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T19:50:57 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 244 seconds]
2012-10-17T20:11:19 -!- kerwood_wk [80f40907@gateway/web/freenode/ip.128.244.9.7] has joined ##stm32
2012-10-17T20:31:51 -!- barthess [~barthess@178.154.19.206] has joined ##stm32
2012-10-17T20:32:07 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-17T20:35:35 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-17T20:36:07 -!- barthess [~barthess@178.154.19.206] has quit [Client Quit]
2012-10-17T20:36:09 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-17T20:36:10 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-17T20:39:58 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-17T20:48:00 -!- barthess [~barthess@178.154.88.75] has joined ##stm32
2012-10-17T20:48:51 -!- barthess [~barthess@178.154.88.75] has quit [Client Quit]
2012-10-17T20:49:05 -!- barthess [~barthess@178.154.88.75] has joined ##stm32
2012-10-17T20:53:53 < zyp> http://bin.jvnv.net/f/x6nGA.png <- starting to look like fun now
2012-10-17T20:54:29 < emeb> starting to look like more than 4 layers.
2012-10-17T20:54:45 < zyp> nah, it's going to be fine
2012-10-17T20:55:10 < zyp> I'll pull this off in two plus ground and vcc planes :p
2012-10-17T20:55:28 < emeb> I await this with interest. :)
2012-10-17T20:55:57 <+Steffanx> I'm more interested in his solder skillz :)
2012-10-17T20:56:03 <+Steffanx> *paste applying skillz
2012-10-17T20:56:26 < emeb> toaster oven skillz
2012-10-17T20:56:29 < zyp> shouldn't be that hard, remember it's 1mm pitch
2012-10-17T20:56:36 -!- daku is now known as DaKu
2012-10-17T20:56:51 < zyp> should be way easier than dense QFN
2012-10-17T20:56:59  * emeb looks at his metric ruler with raise eyebrows.
2012-10-17T20:57:11 <+Steffanx> Should :)
2012-10-17T20:57:21 < emeb> +d
2012-10-17T20:57:30 <+Steffanx> emeb has a metric ruler?
2012-10-17T20:57:51 < emeb> Steffanx: not all americans are stuck in the stone-age.
2012-10-17T20:58:31 < emeb> I even have metric/english digital calipers!
2012-10-17T20:59:47 <+Steffanx> Who's going to do the board zyp ?
2012-10-17T20:59:49 <+Steffanx> Dongs mfg?
2012-10-17T21:00:26 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 255 seconds]
2012-10-17T21:00:28 < zyp> probably, if he bothers doing it
2012-10-17T21:00:41 < zyp> havend thought much about that part yet :p
2012-10-17T21:00:45 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-10-17T21:04:42 <+Steffanx> Not important of course
2012-10-17T21:08:53 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-17T21:09:27 < zyp> oshpark could be an alternative, but they don't have good enough via specs
2012-10-17T21:13:07 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 240 seconds]
2012-10-17T21:13:23 < emeb> I've got a whole tray of MT48LC8M16A2 that I'd like to find a use for.
2012-10-17T21:15:08 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 272 seconds]
2012-10-17T21:15:23 <+Steffanx> A tray-fetish isn't good for you emeb
2012-10-17T21:15:30 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-17T21:15:53 < emeb> Steffanx: whoosh - over my head there.
2012-10-17T21:16:20 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-17T21:18:00  * emeb reads latest ADI newsletter. coughs - $131 ADCs.
2012-10-17T21:18:47 < zyp> emeb, should be a direct fit
2012-10-17T21:19:04 < emeb> zyp: yeah - footprint looked familiar.
2012-10-17T21:19:34 < zyp> it seems to be pretty standard across manufacturers
2012-10-17T21:20:08 < emeb> Oh look - ADI has Cortex M3 devices: http://www.analog.com/en/processors-dsp/analog-microcontrollers/aducm360/products/product.html
2012-10-17T21:20:41 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 246 seconds]
2012-10-17T21:20:49 < emeb> with reaaaallllyy sssssloooowwww Sigma/Delta converters. 3.5kHz! w00t!
2012-10-17T21:20:54 <+Steffanx> lo
2012-10-17T21:20:57 <+Steffanx> l
2012-10-17T21:21:20 < zyp> I just picked the one for MT48LC16M16A2 in eagle standard library because it has all the lines available in the TSOP-54 footprint
2012-10-17T21:21:21 -!- claude is now known as Claude
2012-10-17T21:21:41 < zyp> MT48LC8M16A2 is same except it lacks one of the address lines
2012-10-17T21:21:49 < emeb> Yup.
2012-10-17T21:21:58 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T21:22:25 < emeb> Odd - ADI part has 32kHz xtal built-in?
2012-10-17T21:23:04 < zyp> sure it's not RC?
2012-10-17T21:23:18 < zyp> like LSI in stm32
2012-10-17T21:23:21 < emeb> "Internal watch crystal for wake-up timer"
2012-10-17T21:23:25 < zyp> oh
2012-10-17T21:23:27 < zyp> neat
2012-10-17T21:24:23 < emeb> Very low power it seems - 4uA on wake-up timer only.
2012-10-17T21:25:27 <+Steffanx> isn't that something the l1 can do to?
2012-10-17T21:25:28 <+Steffanx> too
2012-10-17T21:26:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 245 seconds]
2012-10-17T21:27:08 < emeb> haven't paid much attention to power before this. don't do any battery apps.
2012-10-17T21:29:19 < zyp> why?
2012-10-17T21:29:22 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-17T21:29:38 < zyp> oh, wait
2012-10-17T21:29:42 < zyp> I misread
2012-10-17T21:30:01 < zyp> I read the last sentence as a suggestion
2012-10-17T21:30:20 < emeb> heh. _I_ don't do any battery apps. sry.
2012-10-17T21:30:50 < emeb> not "no one should do ..."
2012-10-17T21:32:14 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-17T21:32:43 < TitanMKD> hello
2012-10-17T21:33:01 < emeb> hola
2012-10-17T21:37:10 < zyp> TitanMKD, you were the lpc4300 guy, right?
2012-10-17T21:37:54 < TitanMKD> why the lpc43 guy ?
2012-10-17T21:38:01 < TitanMKD> I hope i'm not the only one to use it ;)
2012-10-17T21:38:14 < TitanMKD> But i like also the STM32F4 ;)
2012-10-17T21:38:17 < zyp> well, I'm doing a board now, just for fun
2012-10-17T21:38:26 < zyp> what cool stuff should I put on it?
2012-10-17T21:38:35 < TitanMKD> 2 USB ;)
2012-10-17T21:38:39 < TitanMKD> all SGPIO
2012-10-17T21:38:50 < TitanMKD> SCT is also a must for some stuff
2012-10-17T21:38:57 < zyp> sct?
2012-10-17T21:39:04 < TitanMKD> and of course all peripherals like SSP/SPI/I2C
2012-10-17T21:39:16  * emeb lobbies for I2S also.
2012-10-17T21:39:19 < TitanMKD> SCT is like SGPIO but more for traffic light managed by hw ;)
2012-10-17T21:39:28 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-17T21:39:29 < zyp> ah, right, it was the state machine stuff
2012-10-17T21:39:42 < TitanMKD> anyway SCT are really less interesting than SGPIO
2012-10-17T21:39:53 < TitanMKD> also add SDRAM ;)
2012-10-17T21:39:58 < zyp> already done
2012-10-17T21:39:59 < TitanMKD> 166MHz or more ;)
2012-10-17T21:40:02 < zyp> http://bin.jvnv.net/f/x6nGA.png
2012-10-17T21:40:15 < TitanMKD> the best is 200MHz SDRAM to have pratically 0 wait state ;)
2012-10-17T21:40:29 < TitanMKD> hehe LBGA version
2012-10-17T21:40:38 < TitanMKD> I have 3 LBGA here
2012-10-17T21:40:52 < zyp> yeah, I'm really just doing the board to see how hard it is to solder :p
2012-10-17T21:40:57 < zyp> (and probably route too)
2012-10-17T21:41:24 < TitanMKD> If you want I have already done a LBGA board but the routing is crap ;)
2012-10-17T21:41:48 < TitanMKD> like USB 2.0 design rules are not respected ...
2012-10-17T21:42:11 < zyp> then you'd take away all my fun, and thus the point for doing it :p
2012-10-17T21:42:17 < TitanMKD> but the most interesting is the SDRAM stuff
2012-10-17T21:42:31 < emeb> TitanMKD: did you do matched lengths?
2012-10-17T21:42:40 < TitanMKD> emeb no nothing match ;)
2012-10-17T21:42:56 < TitanMKD> emeb in fact it should work only in USB FS i think
2012-10-17T21:42:56 < emeb> and it still worked @ 200MHz?
2012-10-17T21:43:03 < zyp> I don't find SDRAM very interesting, but it could be nice to have
2012-10-17T21:43:11 < TitanMKD> emeb 204MHz is internal clock
2012-10-17T21:43:25 < emeb> TitanMKD: I meant for the SDRAM
2012-10-17T21:43:29 < TitanMKD> and i never connected a SDRAM as the board was 5cm*5cm dual layer
2012-10-17T21:43:49 < emeb> Ah. simple stuff.
2012-10-17T21:44:01 < TitanMKD> it's a crash board to test if soldering is possible with standard hot plate ;)
2012-10-17T21:44:08 < emeb> was it?
2012-10-17T21:44:17 < TitanMKD> I have not tested to solder it ;)
2012-10-17T21:44:22 < TitanMKD> but i will ;)
2012-10-17T21:44:42 < emeb> Have a stiff drink, grit your teeth and go to it!
2012-10-17T21:45:00 < zyp> hey, my board is also 5x5
2012-10-17T21:45:08 < zyp> got lots of room so far
2012-10-17T21:45:14 < TitanMKD> I will try adding some solder flux and just heat the hot plate with right ramp up/ramp down
2012-10-17T21:45:29 < zyp> but all the connectors looks so huge compared to the chips :p
2012-10-17T21:45:31 < emeb> Is 5x5 some sort of area limit on Eagle?
2012-10-17T21:45:39 < TitanMKD> anyway 5cm*5cm dual layer with LBGA is impossible to route a SDRAM ;)
2012-10-17T21:45:44 < zyp> not at all
2012-10-17T21:45:56 < TitanMKD> it requires at least 4 layers
2012-10-17T21:46:19 < TitanMKD> as it requires 16bits data + at least 24bits addr
2012-10-17T21:46:24 < TitanMKD> + synchro stuff
2012-10-17T21:46:27 < zyp> huh?
2012-10-17T21:46:33 < zyp> no it doesn't.
2012-10-17T21:46:39 < emeb> TitanMKD: muxed address/data
2012-10-17T21:46:44 < zyp> emeb, no
2012-10-17T21:46:53 < TitanMKD> muxed addr/data on DDR SDRAM ???
2012-10-17T21:46:56 < zyp> data is dedicated, but row/col addressing is muxed
2012-10-17T21:46:57 < emeb> zyp: right ras/cas muxed address
2012-10-17T21:47:08 -!- kerwood_wk [80f40907@gateway/web/freenode/ip.128.244.9.7] has quit [Ping timeout: 245 seconds]
2012-10-17T21:47:15 < zyp> TitanMKD, it doesn't support ddr, only sdr
2012-10-17T21:47:27 < TitanMKD> yes sdr ;)
2012-10-17T21:47:42 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 255 seconds]
2012-10-17T21:47:46 < TitanMKD> maybe you should try with LQFP208
2012-10-17T21:47:54 < TitanMKD> at least you are sure to solder it correctly ;)
2012-10-17T21:48:05 < TitanMKD> because 1 chip cost is 10euros ;)
2012-10-17T21:48:11 < zyp> and the controller supports at most 13 address lines for sdram, so your figure of 24 is way wrong :p
2012-10-17T21:48:26 < TitanMKD> zyp on LBGA version it support more
2012-10-17T21:48:29 < emeb> is the lqfp208 version available?
2012-10-17T21:48:35 < zyp> TitanMKD, no it doesn't.
2012-10-17T21:48:35 < TitanMKD> wait i check the datasheet IIRC it's 24bits addr
2012-10-17T21:48:44 < zyp> not for sdram.
2012-10-17T21:48:56 < zyp> for sram sure, but not sdram
2012-10-17T21:49:21 < TitanMKD> you are wrong
2012-10-17T21:49:22 < TitanMKD> EMC_A[22:0
2012-10-17T21:49:25 < TitanMKD> 23 bit addr
2012-10-17T21:49:35 < TitanMKD> even on LQFP208
2012-10-17T21:49:49 < TitanMKD> also LBGA256 support even 32bits data
2012-10-17T21:49:55 < zyp> are you unable to comprehend what I am saying?
2012-10-17T21:49:58 < TitanMKD> but LQFP208 only 16bits data
2012-10-17T21:50:34 < zyp> the sdram controller in lpc4300 is unable to use more than 13 bits of addressing for sdram, plus two lines for bank select
2012-10-17T21:50:58 < TitanMKD> hmm
2012-10-17T21:51:34 < zyp> also, I'm not sure there exists sdram chips with more than 13 address lines either, so having more would be pretty pointless.
2012-10-17T21:51:58 < TitanMKD> ok maybe ;)
2012-10-17T21:52:29 < TitanMKD> it support up to 256MB SDRAM
2012-10-17T21:52:41 < TitanMKD> but anyway 32 or 64MB is enough ;)
2012-10-17T21:52:54 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 264 seconds]
2012-10-17T21:53:05 < TitanMKD> but trace length is vital at 166 or 200MHz
2012-10-17T21:53:17 < TitanMKD> and even 4 layers i will say
2012-10-17T21:53:49 < TitanMKD> to avoid crossing other signal and to have some GND and VCC layer dedicated
2012-10-17T21:53:56 < Claude> pardon for my 2 cents , but yes up to A12 on a sdram is enough for even 512MB
2012-10-17T21:54:15 < Claude> RAS/CAS ..
2012-10-17T21:54:46 < TitanMKD> does cheap SDRAM support 32bits data ?
2012-10-17T21:54:51 < TitanMKD> it will be a must to have that ;)
2012-10-17T21:55:10 < TitanMKD> 32bits even at 133MHz will just a must
2012-10-17T21:55:19 < Claude> yes of couse , but only in an uncommon package .. TSOP 66 iirc
2012-10-17T21:55:32 < TitanMKD> common package are 16bits data ?
2012-10-17T21:55:37 < Claude> yes
2012-10-17T21:55:38 < zyp> TitanMKD, you can just hook up two in parallel with D0-15 on one and D16-31 on the other
2012-10-17T21:55:56 < TitanMKD> zyp yes but it is a bit complex and take lot of space
2012-10-17T21:56:14 < TitanMKD> the must is SDRAM @ 200MHZ 16bits
2012-10-17T21:56:34 < zyp> «must»
2012-10-17T21:59:20 -!- tavish [~tavish@120.56.133.215] has joined ##stm32
2012-10-17T21:59:20 -!- tavish [~tavish@120.56.133.215] has quit [Changing host]
2012-10-17T21:59:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-17T21:59:48 < TitanMKD> zyp have you see my ugly LBGA test crash board
2012-10-17T21:59:49 < TitanMKD> http://postimage.org/image/vrzq6xm4h/
2012-10-17T22:00:06 < Claude> i doubt an cortex m4 , even at 200MHz , needs that much memory bandwidth you asking for . also keep in mind that sdram is quite slow when used not right . sdram only gets decent memory bandwidth when used as burst memory
2012-10-17T22:00:07 < TitanMKD> it was a real pain to output all those GPIO... ;)
2012-10-17T22:00:24 < TitanMKD> Claude of course it need especially when you use DMA ;)
2012-10-17T22:00:24 < zyp> «all»
2012-10-17T22:00:43 < TitanMKD> Claude or for peripherals to memory direct access with DMA
2012-10-17T22:00:49 < TitanMKD> Claude can replace an FPGA ;)
2012-10-17T22:01:05 <+Steffanx> You forget «;)» zyp  :P
2012-10-17T22:01:22 < Claude> as long as your DMA really does burst transfers to the memory
2012-10-17T22:01:37 < zyp> TitanMKD, hooking together both vbus lines like that doesn't look very sane
2012-10-17T22:02:05 < zyp> where did you have the board made?
2012-10-17T22:02:18 < TitanMKD> zyp have made it at Itead
2012-10-17T22:02:37 < zyp> what's their via specs?
2012-10-17T22:02:44 < TitanMKD> look the bottom
2012-10-17T22:02:45 < TitanMKD> http://postimage.org/image/pnfuclq9p/
2012-10-17T22:02:56 < TitanMKD> their specs are very good
2012-10-17T22:03:02 < TitanMKD> but i paid 7us$ for BGA
2012-10-17T22:03:41 < Claude> cool , BGA on 2 layers ? neat
2012-10-17T22:03:54 < zyp> I didn't ask you for an opinion of their via specs, I'm asking what their specs are.
2012-10-17T22:04:08 < TitanMKD> drill shall be 12mil min
2012-10-17T22:04:18 < zyp> and annular?
2012-10-17T22:04:35 < TitanMKD> no spec on annular
2012-10-17T22:04:53 < TitanMKD> wait i give you the online specs
2012-10-17T22:05:12 < zyp> surely they need to have a spec on that, otherwise you could make the pad for the via also 12 mil :p
2012-10-17T22:05:20 < TitanMKD> I use design rule check in fact to be sure ;)
2012-10-17T22:05:39 < TitanMKD> wire 6mil min too
2012-10-17T22:05:57 < TitanMKD> pad 6mil min and via 6mil min
2012-10-17T22:06:14 < zyp> so that's 6 mil on either side of the drill?
2012-10-17T22:06:23 < zyp> so a via is minimum 24mil in diameter?
2012-10-17T22:06:35 < TitanMKD> no diamter min is 12
2012-10-17T22:06:47 < zyp> I'm talking about the pad, not the hole
2012-10-17T22:07:03 < zyp> you know, a via is a pad with a hole through.
2012-10-17T22:07:25 < TitanMKD> look here
2012-10-17T22:07:26 < TitanMKD> http://imall.iteadstudio.com/open-pcb/pcb-prototyping/im120418001.html
2012-10-17T22:07:46 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 246 seconds]
2012-10-17T22:08:11 < TitanMKD> and only 9.9US$ for 10board 5cm*5cm
2012-10-17T22:08:16 < TitanMKD> and 100% ETest
2012-10-17T22:08:20 -!- tavish [~tavish@59.177.14.15] has joined ##stm32
2012-10-17T22:08:20 -!- tavish [~tavish@59.177.14.15] has quit [Changing host]
2012-10-17T22:08:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-17T22:08:45 < TitanMKD> but warning for LBGA they ask +7US$
2012-10-17T22:08:53 < TitanMKD> anyway it is very cheap
2012-10-17T22:09:15 < TitanMKD> it is why i have done that in hurry
2012-10-17T22:09:52 < zyp> ah, annular ring is 0.15mm
2012-10-17T22:10:01 < zyp> so vias are 12/24
2012-10-17T22:10:19 < zyp> and trace/space is 6/6
2012-10-17T22:11:42 < zyp> that's not good enough
2012-10-17T22:12:02 < zyp> with those specs there is no space to route between two vias located 1mm from each other
2012-10-17T22:12:52 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 260 seconds]
2012-10-17T22:14:06 < TitanMKD> ha yes only 1 wire
2012-10-17T22:14:26 < TitanMKD> it is why i tell you it is impossible to route that on 2 layers
2012-10-17T22:14:29 < zyp> no. no wires
2012-10-17T22:14:48 < TitanMKD> between 2 via i set 1 trace
2012-10-17T22:15:04 < TitanMKD> it is why they asked me additional 7US$
2012-10-17T22:15:10 < TitanMKD> because it is less than 6mil
2012-10-17T22:15:41 < zyp> 1mm is approx. 39mil, using 24 mil for via + 6 for trace + 2 * 6 for spacing around trace adds up to 42 mil
2012-10-17T22:17:20 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-17T22:26:57 < TitanMKD> zyp anyway for LBGA Itead is maybe not the best
2012-10-17T22:27:19 < zyp> don't worry, I'm not going to use itead
2012-10-17T22:27:34 < TitanMKD> especially silkscreen are really awfull at Itead
2012-10-17T22:27:45 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-17T22:28:22 < TitanMKD> even if it is not the most important stuff it is hard to align components like QFN/LBGA with wrong silkscreen (with offset in x and y)
2012-10-17T22:30:08 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 260 seconds]
2012-10-17T22:41:06 -!- dekar [~dekar@212.255.46.139] has joined ##stm32
2012-10-17T22:41:09 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-17T22:45:45 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-17T22:50:41 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-17T22:53:51 < Posterdati> zoccole italiane?
2012-10-17T22:54:39 < Posterdati> lol
2012-10-17T22:55:29 < BrainDamage> almost tempted to submit that to bash.org but very few would understand the "ooops wrong chan"
2012-10-17T22:55:39 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-17T22:59:44 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-10-17T23:00:03 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-17T23:01:24 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 248 seconds]
2012-10-17T23:03:23 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has joined ##stm32
2012-10-17T23:06:08 < Posterdati> no, no wrong channel!
2012-10-17T23:12:11 < emeb> italian prostitutes? apparently nothing is OT here. :)
2012-10-17T23:13:06 < Posterdati> yes
2012-10-17T23:13:51 < zyp> http://bin.jvnv.net/f/Fml9t.png <- sdram routed without any conflicts
2012-10-17T23:14:07 < emeb> not bad.
2012-10-17T23:14:37 < emeb> Like the 45deg placement -> http://ebrombaugh.studionebula.com/synth/armfpga/index.html
2012-10-17T23:14:43 < TitanMKD> woo very fast
2012-10-17T23:14:54 < zyp> emeb, nice :D
2012-10-17T23:15:04 < TitanMKD> but the signal are crossing
2012-10-17T23:15:13 -!- Claude is now known as claude
2012-10-17T23:15:34 < zyp> TitanMKD, there are no drc violations anywhere
2012-10-17T23:16:07 -!- tectu is now known as Tectu
2012-10-17T23:16:23 < TitanMKD> zyp yes but check trace length and also crossing is not good
2012-10-17T23:17:37 < zyp> don't be silly, it's just the first step, getting everything routed without conflicts
2012-10-17T23:18:19 < TitanMKD> hehe yes i joke
2012-10-17T23:18:31 < TitanMKD> it seems you do that with freeroute and kicad i'm right ?
2012-10-17T23:18:38 < zyp> eagle
2012-10-17T23:18:44 < TitanMKD> ha eagle ok
2012-10-17T23:18:52 < TitanMKD> you have used bga escape in that case
2012-10-17T23:19:25 < zyp> nah, it doesn't look very useful
2012-10-17T23:20:38 -!- barthess [~barthess@178.154.88.75] has quit [Quit: Leaving.]
2012-10-17T23:22:10 < zyp> I've got some in-progress pictures of the last board I did: http://bin.jvnv.net/f/ZNoth.png http://bin.jvnv.net/f/BXU4o.png http://bin.jvnv.net/f/nGjGf.png
2012-10-17T23:22:34 < zyp> it's fun to look through and see the board taking form :)
2012-10-17T23:23:15 < TitanMKD> very good
2012-10-17T23:24:16 < TitanMKD> bye
2012-10-17T23:24:19 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-17T23:26:17 < TzacksT> you guys, just a little offtopic, if you stop doing great electronics stuff you do i'll throw a giangt replica of sputnic at you :-)
2012-10-17T23:26:47 < zyp> ouch? :p
2012-10-17T23:26:53 < TzacksT> :))
2012-10-17T23:26:54 < emeb> sounds pointy
2012-10-17T23:27:28 < TzacksT> just wanted to say gn in some non boring way :O
2012-10-17T23:28:25 < zyp> emeb, do you happen to know how much length deviation that is acceptable for this kind of sdram?
2012-10-17T23:29:10 < emeb> zyp: not offhand.
2012-10-17T23:29:24 < emeb> Depends on what clock speed you're running I'd think.
2012-10-17T23:30:39 < zyp> sure
2012-10-17T23:30:51 < zyp> what clock speed would I be running?
2012-10-17T23:31:32 < emeb> the ds I've got for the above mentioned part says either 100 or 133 MHz.
2012-10-17T23:32:03 < emeb> 1.5ns setup, 0.8ns hold, so you've got about a 7.5ns - 1.5 -0.8 = 5.2ns window for skew
2012-10-17T23:32:55 < emeb> prop delay in PCB traces is ~ 1ns/foot, so assuming no skew on the MCU pins the traces could deviate by quite a lot.
2012-10-17T23:33:04 < zyp> shortest trace is currently 11.3mm, longest is 52.9
2012-10-17T23:33:05 < emeb> but that's an unreasonable assumption.
2012-10-17T23:33:11 < zyp> so, 41.6mm difference
2012-10-17T23:33:31 < emeb> you need to know the skew @ the LPC4300 balls too and derate from there.
2012-10-17T23:33:33 < zyp> and that's just going to become less as I clean it up
2012-10-17T23:34:38 < emeb> yeah. I'm guessing you don't have to worry too much.
2012-10-17T23:47:22 < zyp> http://www.nxp.com/documents/application_note/AN10935.pdf <- I'm not sure how relevant this is, or what speed this is based on, but it suggests 0.3-1" depending on which signals
2012-10-17T23:49:13 < zyp> anyway, that just gives me a reason to try out eagle's meander tool :p
--- Day changed Thu Oct 18 2012
2012-10-18T00:05:38 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-10-18T00:08:31 < upgrdman_> the f4discov board has a 2.5V regulator that seems to only be used by the audio section... why does it need 2.5V instead of 3V?
2012-10-18T00:08:46 < upgrdman_> (seems kind of weird)
2012-10-18T00:10:25 < zyp> check the datasheet for the CS43L22
2012-10-18T00:10:48 < zyp> it requires a seperate analog supply between 1.65-2.63V
2012-10-18T00:11:47 < zyp> (and regardless of value, seperate regulator for analog stuff is good anyway, to avoid noise from digital power rail to get into your audio)
2012-10-18T00:12:15 < zyp> oh, wait, it requires a digital supply of the same value
2012-10-18T00:12:55 < zyp> it's only io supply that's allowed to be >3V
2012-10-18T00:13:33 < upgrdman_> seems kind of weird? is it common for audio chips to want <3V?
2012-10-18T00:14:35 < zyp> ask emeb, he has touched more audio chips than me :p
2012-10-18T00:14:46 < upgrdman_> fair enough
2012-10-18T00:17:05 < upgrdman_> what is newlib/multilib?
2012-10-18T00:17:20 < emeb> upgrdman_: I've seen a few audio chips with lower core supply voltages. Not that unusual, especially with newer parts.
2012-10-18T00:17:42 < upgrdman_> o ok
2012-10-18T00:18:12 < emeb> in general you'll get better audio performance from higher supply voltages. There are some codecs that have 3.3V (or lower) digital supplies and allow up to 5V for the analog section.
2012-10-18T00:18:13 < zyp> newlib is a libc implementation and multilib is a gcc framework that allows it to keep several variants of standard library for different architectures and select between them
2012-10-18T00:21:39 < upgrdman_> ya, i googled it and found that out too. but why? efficiency? license issues?
2012-10-18T00:25:12 < zyp> with regards to multilib?
2012-10-18T00:26:22 < zyp> there is a lot of variation between different arm cpus
2012-10-18T00:27:44 < zyp> cortex-m0, m3, m4 and m4f all support a different number of instructions, and having one library variant be compatible with all means limiting it to only instructions available in all
2012-10-18T00:28:31 < zyp> to have the library make use of, say, the fpu in m4f, you need to compile it with fpu support
2012-10-18T00:28:35 < upgrdman_> newlib, not multilib
2012-10-18T00:29:16 < zyp> what about newlib?
2012-10-18T00:29:58 < upgrdman_> why newlib? whats wrong with libc? just not optimized for various  arm cores, or other reasons?
2012-10-18T00:30:38 < zyp> which libc?
2012-10-18T00:30:46 < cjbaird> Once upon a time, glibc had a reputation of being 'bloated'..
2012-10-18T00:30:55 < upgrdman_> the ones that gcc normally comes with (i assume??)
2012-10-18T00:31:09 < upgrdman_> glibc == gnu's libc?
2012-10-18T00:31:14 < zyp> yes
2012-10-18T00:31:24 < upgrdman_> o, i meant glibc then
2012-10-18T00:31:59 < zyp> newlib is written to be more independent from operating systems
2012-10-18T00:32:15 < upgrdman_> oh, that makes sense. thanks!
2012-10-18T00:32:38 < zyp> remember that glibc is mainly written to run on top of linux
2012-10-18T00:33:24 < upgrdman_> ok
2012-10-18T00:33:28 < cjbaird> At one time, glibc was intended for full multiuser networking systems, and newlib et.al. where for embedded work. The change in the embedded sphere has seen that become less necessary, but the developers have stuch with it.
2012-10-18T00:34:21 < zyp> newlib is fully freestanding and only requires that you provide your own variants of some system calls it needs (which pretty much can be ignored if you don't use that particular functionality it's needed for)
2012-10-18T00:35:19 < upgrdman_> do you know if gcc-arm-embedded comes with newlib?
2012-10-18T00:36:11 < zyp> pretty much any «gcc for cortex-m» package you'll find comes with newlib
2012-10-18T00:36:23 < upgrdman_> ok, good :)
2012-10-18T00:37:16 < cjbaird> glibc also had a period where the primary developers went through an ivory-tower phase that lasted for about ~8 years.
2012-10-18T00:38:36 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Quit: TzacksT]
2012-10-18T00:39:07 < cjbaird> A number of mini-libc projects started because the glibc czars weren't interested in making the mainline configure for limited systems.
2012-10-18T00:41:26 < cjbaird> I haven't seen eglibc used much?
2012-10-18T00:45:22 -!- DaKu is now known as daku
2012-10-18T00:47:04 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-18T00:48:47 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-18T00:56:33 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-18T01:00:05 < karlp> huh, he left.
2012-10-18T01:00:32 < karlp> from what I've seen in the openwrt forums and lists, eglibc's size optimizations are gradually being absorbed back into glibc
2012-10-18T01:00:49 < karlp> apparently because glibc people are more prepared to talk to other people these days  or something
2012-10-18T01:01:17 < karlp> also, on owrt at least, it's either uclibc or glibc if you want it to work. eglibc still breaks lots of things.
2012-10-18T01:14:44 < upgrdman_> this mcu stuff has really opened my eyes to many things i took for granted :)
2012-10-18T01:16:57 < upgrdman_> and i somehow managed to burn up (literally) the 2.5V regulator on the f4discov board... and im not even using the audio stuff. but the power supply I used sucks, and the voltage dips/surges if you draw a 1-2A transient load :(
2012-10-18T01:17:56 < upgrdman_> and the 3v regulator seems to have taken a dump too. small speck on it, maybe just flux splatter, or maybe it burned too
2012-10-18T01:18:29 < upgrdman_> luckily the f4 and my lcd seems to be unharmed. providing 3V on a 3V pin powers it up fine
2012-10-18T01:23:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-18T01:23:53 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-18T01:28:28 < dongs> sup dongs
2012-10-18T01:28:39 < dongs> Laurenceb: not rc shit.
2012-10-18T01:32:22 < emeb> been sleeping again?
2012-10-18T01:33:15 < zyp> dongs, what do you think about my super awesome sdram routing? http://bin.jvnv.net/f/A4Ep7.png :p
2012-10-18T01:33:50 < dongs> all I can say is youre lucky you dont need to keep trace length same
2012-10-18T01:33:55 < dongs> cause that looks horrible ;p
2012-10-18T01:34:10 < zyp> :p
2012-10-18T01:34:18 < qyx_> fuh
2012-10-18T01:34:35 < dongs> i would have expected the pinout on BGA to match teh common sdram pinout
2012-10-18T01:34:47 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-18T01:34:59 < dongs> so you wouldn't have to do a lot of crazy stuff with vias etc
2012-10-18T01:35:09 < dongs> but looks like its all over the place hum.
2012-10-18T01:35:37 < qyx_> that 90° edges look frightening
2012-10-18T01:35:38 < zyp> well, it's not PoP
2012-10-18T01:35:45 < zyp> qyx_, it's not done
2012-10-18T01:35:48 < qyx_> ah
2012-10-18T01:36:43 < qyx_> why don't you use qfp?
2012-10-18T01:36:54 < qyx_> is that the lpc43xx?
2012-10-18T01:36:58 < zyp> yes
2012-10-18T01:37:13 < zyp> and because I want to try bga to see how hard I can fail
2012-10-18T01:37:46 < qyx_> k, understand :)
2012-10-18T01:38:10 < qyx_> how big sdram does it support?
2012-10-18T01:38:24 < qyx_> 32M/64M at least?
2012-10-18T01:38:31 < qyx_> i haven't read the datasheet yet
2012-10-18T01:38:57 < zyp> as large as you can get in TSOP-54
2012-10-18T01:39:03 < zyp> I've looked at 32
2012-10-18T01:39:22 < qyx_> i meant lpc, not board
2012-10-18T01:39:38 < zyp> oh, right
2012-10-18T01:40:06 < qyx_> hm, and it seems to be cheaper than F4
2012-10-18T01:40:45 < dongs> doubtful
2012-10-18T01:41:18 < qyx_> NXP - LPC4310FBD144,551 - MCU, 32BIT, ARM CORTEX M4, 144LQFP
2012-10-18T01:41:29 < qyx_> 7.72e @ 10pcs
2012-10-18T01:42:29 < qyx_> the cheapest F4 is 9.32e @ 10pcs
2012-10-18T01:43:03 < zyp> seems like the largest the controller will support is 256MB, i.e. 64Mx32
2012-10-18T01:43:17 < zyp> using four 64Mx8 chips in parallel
2012-10-18T01:43:28 < qyx_> not so bad
2012-10-18T01:43:37 < zyp> and then you can have four of those banks
2012-10-18T01:43:58 < zyp> hmm, but apparently one bank only supports 128M
2012-10-18T01:44:16 < zyp> so, 896 MB or so
2012-10-18T01:44:22 < dongs> zyp: http://files.chumby.com/hackerboard/chumby_hackerboard_gerbers.zip maybe some ideas for sdram routing there.
2012-10-18T01:44:49 < zippe> re: DDR, most vendors will have example escape routing
2012-10-18T01:45:01 < qyx_> altera had some good docs about high freq pcb design
2012-10-18T01:45:58 < dongs> i think its only 4 layer... opening now to confirm
2012-10-18T01:46:03 < qyx_> an2582 freescale
2012-10-18T01:46:06 < qyx_> also
2012-10-18T01:46:23 < dongs> hah, wow, its all trace length matched
2012-10-18T01:46:54 < qyx_> i did that on 2 layer, matched to +-2mm, but never tested
2012-10-18T01:46:59 < qyx_> ddr with imx233
2012-10-18T01:47:12 < zyp> this is sdr
2012-10-18T01:47:20 < qyx_> should be easier
2012-10-18T01:47:25 < zyp> exactly
2012-10-18T01:47:32 < dongs> http://i.imgur.com/bltlJ.png
2012-10-18T01:47:40 < dongs> this is what I meant by "sane" sdram controller layout...
2012-10-18T01:47:53 < dongs> see how all sdr shit is just corner 2 balls
2012-10-18T01:48:29 < karlp> what are those things in the lines?
2012-10-18T01:48:32 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-18T01:48:34 < zyp> resistors
2012-10-18T01:48:35 < karlp> buffers/series resistors or what?
2012-10-18T01:48:56 < karlp> what are they for?
2012-10-18T01:49:08 < dongs> ya looks like resistor networks
2012-10-18T01:49:18 < qyx_> btw good pin layout
2012-10-18T01:49:23 < qyx_> on that bga package
2012-10-18T01:49:30 < dongs> yes.
2012-10-18T01:59:58 < zyp> http://bin.jvnv.net/f/Q3vyK.png <- some more layers
2012-10-18T02:01:15 < dongs> does NXP have a evalboard with sdram and that lpc?
2012-10-18T02:01:28 < dongs> looking at those layout/gerbers would be not a bad idea
2012-10-18T02:01:35 < dongs> which one is that? lpc4300?
2012-10-18T02:01:56 < zyp> http://www.keil.com/mcb4300/ <- keil does
2012-10-18T02:02:21 < zyp> http://www.nxp.com/demoboard/OM13031.html <- also this
2012-10-18T02:02:29 < dongs> wehres the sores
2012-10-18T02:02:55 < dongs> lol schematic only
2012-10-18T02:02:57 < dongs> terrible
2012-10-18T02:03:06 < dongs> one of the reasons ill never buy nxp i guess!
2012-10-18T02:04:45 < dongs> keil has schematics too
2012-10-18T02:05:25 < zyp> by the way, it's not so strange that it's not as nicely laid out as a dedicated sdram controller, considering that the emc stuff is similar to fsmc in that it can be used for sram and nor flash too
2012-10-18T02:05:34 < dongs> ah
2012-10-18T02:06:11 < zyp> looking at the picture of the keil board, the sdram is placed on the opposite side of the flash chip
2012-10-18T02:06:23 < zyp> so it's obvious that it's serving both with the same lines
2012-10-18T02:09:13 < dongs> still looks pretty nice on imx
2012-10-18T02:09:37 < dongs> not to mention that altium in general produces nice looking layouts.
2012-10-18T02:09:51 < dongs> those teardropped vias are  really nice lookin.
2012-10-18T02:10:38 < zyp> true
2012-10-18T02:10:42 < dongs> in that imx gerber balls are 15mil
2012-10-18T02:10:45 < dongs> vias are 18mil
2012-10-18T02:11:10 < dongs> traces are 5 mil
2012-10-18T02:12:31 < zyp> it's a bit interesting how outer layers are used as planes and only routing on inner
2012-10-18T02:14:04 < dongs> huh where do you see that? i dont see any plane layers. maybe the bottom one only
2012-10-18T02:14:08 < dongs> top is certainly used for routing.
2012-10-18T02:14:43 < zyp> yes, but free areas are filled
2012-10-18T02:14:54 < dongs> right, but thats normal?
2012-10-18T02:15:45 < zyp> not always when you have dedicated plane layers
2012-10-18T02:15:50 < dongs> im sure most of the shit was autorouted anyway.
2012-10-18T02:16:45 < dongs> i cant believe not a single lpc4330 evalboard maker has gerbers
2012-10-18T02:16:50 < dongs> does olimex make one?
2012-10-18T02:17:11 < zyp> no idea
2012-10-18T02:17:36 < dongs> http://olimex.wordpress.com/2012/10/17/new-product-arm-jtag-coocox/ lo
2012-10-18T02:18:34 < upgrdman> am i the only one that finds the name coocox absurd (in a funny way) ?
2012-10-18T02:18:50 < dongs> upgrdman: even better, their HAL library is called COX...
2012-10-18T02:19:40 < dongs> doesnt look like olimex has ant 1800/4300 parts
2012-10-18T02:19:54 < dongs> man their devboards are fucking cheap though
2012-10-18T02:22:38 < zyp> that's why they come with so good support
2012-10-18T02:23:44 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 260 seconds]
2012-10-18T02:24:00 < dongs> haha, what do you mean
2012-10-18T02:24:08 < dongs> how much suport do you need with a devboard
2012-10-18T02:25:24 < zyp> I don't need any, I'm just referring to the guys turning up here and there and asking for help because they can't get their cheap shit working :p
2012-10-18T02:25:49 < zyp> maybe that's more a problem with the buyer than the seller though
2012-10-18T02:42:31 < qyx_> hm, that power consumption in lcp43xx datasheet is only for M4 core running?
2012-10-18T02:42:36 < qyx_> or both?
2012-10-18T03:07:39 < zippe> Beats the hell out of my old days cleaning boards with MEK
2012-10-18T03:32:31 < cjbaird> flyback: pirate my ideas, why don't you.. http://i.imgur.com/6lBfH.jpg
2012-10-18T03:33:01 < cjbaird> That'd been stored in a roof for ~30 years..
2012-10-18T03:33:18 < dongs> what hte hell is that shit
2012-10-18T03:35:08 < cjbaird> You people have shit long-term memories. It's the BigBoardII I was poking at ~9 months ago. Still haven't found a stable PSU for it. (But the short-circuit problem was solved..)
2012-10-18T03:36:59 < cjbaird> The limonine solvent I used that time was a fabric stain remover.
2012-10-18T03:44:06 < cjbaird> moar good shit: http://www.starringthecomputer.com/computers.html
2012-10-18T03:54:54 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-18T03:55:00 < dongs> ok where are the chats
2012-10-18T03:59:36 < cjbaird> I just had to regrease the cpu heatsink on this machine.. There's quite a bit of difference now-- 100% load temp of 67-72C is now 43C :D ..
2012-10-18T04:04:48 < cjbaird> Whatever the silicone stuff was that came with the system.
2012-10-18T04:05:43 < emeb> time to start using water-soluble flux?
2012-10-18T04:06:32 < cjbaird> I've still got some gold leaf here...
2012-10-18T04:11:33 < dongs> 9221 shipped, should be here friday/sat
2012-10-18T04:11:55 < emeb> \o/
2012-10-18T04:15:53 < dongs> and PCBs maybe sat/sun.
2012-10-18T04:16:02 < dongs> i think theres some stencils to panel so maybe that too
2012-10-18T04:50:13 -!- Guest27481 [~CheBuzz@76.164.192.124] has quit [Ping timeout: 246 seconds]
2012-10-18T06:10:24 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-18T06:10:31 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-18T06:14:26 < Bird|lappy> apparently DigiKey will not mail lithium cells even though they are allowed under certain conditions by the USPS...o.O
2012-10-18T06:15:05 < Bird|lappy> might have something to do with mail rates or the likes?
2012-10-18T06:16:24 < dongs> obviosuly they dont want you huffing on lithium
2012-10-18T06:20:17 < Bird|lappy> :P
2012-10-18T06:28:02 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has joined ##stm32
2012-10-18T06:32:21 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-18T06:51:49 < emeb> or maybe they want the option to airmail & lithium isn't allowed on planes?
2012-10-18T07:20:34 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-18T07:24:55 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-18T08:03:08 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-18T08:08:52 < dongs> jap post banned lithium years ago
2012-10-18T08:08:55 < dongs> hell not evne jap post
2012-10-18T08:09:00 < dongs> even fedex/etc outgfoing from JP
2012-10-18T08:09:48 < emeb_mac> so who ships lithium now?
2012-10-18T08:10:32 < emeb_mac> that's ok - johnson (heh heh heh) bought them at the fire sale.
2012-10-18T08:10:52 < emeb_mac> I see what you did there.
2012-10-18T08:11:51 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-18T08:11:54 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-18T08:11:55 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-18T08:14:30 < emeb_mac> you tried to do business with a123 and were denied?
2012-10-18T08:14:52 < emeb_mac> ah
2012-10-18T08:15:06 < emeb_mac> hilarious.
2012-10-18T08:16:19 < Bird|lappy> yeah, that's just silly
2012-10-18T08:16:23 < Bird|lappy> down-right silly
2012-10-18T08:16:27 < emeb_mac> yeah - basically they'd set their sights on the automotive market and weren't considering anyone else.
2012-10-18T08:18:06 < emeb_mac> probably lessons to be learned from chip mfgs who support smaller customers & hobbyists via internet.
2012-10-18T08:18:59 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 252 seconds]
2012-10-18T08:19:05 < Bird|lappy> emeb_mac, yeah.  don't stop people from coming up with ingenious applications for your product. :)
2012-10-18T08:19:11 < emeb_mac> well - tech marketing has changed a lot over the last few decades.
2012-10-18T08:19:44 < Bird|lappy> who would have thought that a humble thermostat could be a possible app for a thin-film Li-Ion battery?
2012-10-18T08:19:58 < emeb_mac> it doesn't take huge sales force to service many small customers via the web.
2012-10-18T08:20:21 < emeb_mac> Bird|lappy: like what - Nest?
2012-10-18T08:20:48 < Bird|lappy> not even at that level
2012-10-18T08:20:53 < emeb_mac> that too - got some good distys
2012-10-18T08:20:56 < Bird|lappy> yeah.
2012-10-18T08:21:12 < Bird|lappy> IPS for instance has their full line available through DigiKey
2012-10-18T08:21:54 < emeb_mac> well - sounds like they fell to bad management, not bad tech.
2012-10-18T08:22:11 < emeb_mac> hopefully johnson can do better with their ip.
2012-10-18T08:23:03 < ds2> just use NiMH
2012-10-18T08:23:24 < emeb_mac> nice
2012-10-18T08:23:38 < emeb_mac> so, good for long life, lower capacity apps?
2012-10-18T08:24:10 < emeb_mac> ds2: dude.
2012-10-18T08:25:11 < cjbaird> A123 sounds like an Australian company..
2012-10-18T08:26:25 < emeb_mac> cjbaird: sounds like, but was US - had gov't green energy loans, etc.
2012-10-18T08:28:01 < dongs> emeb_blog: did you start on your F373 breakout yet?
2012-10-18T08:28:33 < ds2> hey emeb_mac
2012-10-18T08:28:41 < emeb_mac> dongs: got a start on it. Schematic is about 70% but other priorities, etc.
2012-10-18T08:28:47 < ds2> F373?
2012-10-18T08:29:02 < emeb_mac> ds2: yeah stm32f373
2012-10-18T08:29:25 < cjbaird> When most of the aussie companies I've personally been involved/associated/etc. with had a similar /extreme misguidedness/ and failed HARD, I'm now someone who thinks it's a good thing for global multinationals 'taking over' the place...
2012-10-18T08:29:42 < ds2> OOooooh that
2012-10-18T08:29:43 < dongs> A123 sounds like some kinda battery
2012-10-18T08:29:49 < ds2> emeb_mac: do you have a F3 discovery  board?
2012-10-18T08:30:00 < emeb_mac> ds2: they're not shipping yet.
2012-10-18T08:30:09 < emeb_mac> I'll have one when they do though.
2012-10-18T08:30:18 < ds2> really? ST's website claims RS/EU has 45 in stock already
2012-10-18T08:30:29 < dongs> wat
2012-10-18T08:30:38 < emeb_mac> ds2: but Mouser/DK in US don't have any yet.
2012-10-18T08:30:59 < ds2> you checked DK? their site was down this afternoon :(
2012-10-18T08:31:10 < emeb_mac> ds2: not in the last few days.
2012-10-18T08:31:16 < dongs> i did, still 10/31
2012-10-18T08:31:18 < dongs> i haev 2 on order.
2012-10-18T08:31:33 < ds2> emeb_mac: are you laying it out for bare metal use?
2012-10-18T08:31:47 < emeb_mac> ds2: yeah.
2012-10-18T08:31:59 < ds2> emeb_mac: how big are you targeting?
2012-10-18T08:32:02 < emeb_mac> just USB and xtal dedicated.
2012-10-18T08:32:17 < dongs> apparently RS does have it in stock in UK heh
2012-10-18T08:32:24 < emeb_mac> something small - DIP headers kinda thing.
2012-10-18T08:32:37 < emeb_mac> use the 48lqfp
2012-10-18T08:32:38 < ds2> ohh you must be using the small package?
2012-10-18T08:32:44 < ds2> ah...
2012-10-18T08:32:49 < ds2> guess I will have to make my own board then
2012-10-18T08:32:59 < emeb_mac> ds2: what do you need?
2012-10-18T08:33:24 < ds2> a tiny 1.5x1.5" board w/8Mbyte of RAM and 32G of flash :)
2012-10-18T08:33:29 < emeb_mac> just checked DK again - still no F3 discovery.
2012-10-18T08:33:35 < emeb_mac> har
2012-10-18T08:33:41 < ds2> not joking
2012-10-18T08:33:51 < ds2> I am 95% done laying that out
2012-10-18T08:33:59 < emeb_mac> you're not getting there with an f373
2012-10-18T08:34:01 < ds2> F103 board for uClinux
2012-10-18T08:34:17 < emeb_mac> wait, what?
2012-10-18T08:34:27 < emeb_mac> what you using for RAM?
2012-10-18T08:34:29 < ds2> STM32F103 with the above specs for Linux
2012-10-18T08:34:32 < ds2> SRAM
2012-10-18T08:34:37 < emeb_mac> FSMC + what parts?
2012-10-18T08:35:01 < ds2> haven't finalized the P/N but it'll either be the one from the ST/EVAL board or a micron part
2012-10-18T08:35:14 < ds2> seems to take the same BGA but I think micro is 1.8VddIO only :(
2012-10-18T08:35:17 < emeb_mac> those BGA PSRAMs?
2012-10-18T08:35:21 < ds2> yeah
2012-10-18T08:35:30 < cjbaird> Someone just said the magic word again.
2012-10-18T08:35:37 < emeb_mac> yep
2012-10-18T08:35:51 < ds2> hence the 0201 complaints about size
2012-10-18T08:35:52 < emeb_mac> we've been lamenting that PSRAMs require fancy PCB processes here.
2012-10-18T08:36:04 < ds2> uh... fancy?
2012-10-18T08:36:18 < emeb_mac> as in not hobbyist quality
2012-10-18T08:36:20 < cjbaird> But certainly something a lot of people here have been checking out.
2012-10-18T08:36:22 < ds2> does anyone here need a F103 board capable of uCLinux?
2012-10-18T08:36:33 < ds2> be happy to run more of those
2012-10-18T08:36:47 < emeb_mac> at this point I'm not too interested in anything but F4xx
2012-10-18T08:36:55 < emeb_mac> need speed + FPU
2012-10-18T08:37:17 < ds2> the F[234]xx have a bigger package
2012-10-18T08:37:45 < ds2> given that, the K70 looks more attractive since it can do DRAM
2012-10-18T08:37:47 < emeb_mac> but one of my clients just got boards back for an F407 w/ PSRAM so I'll be working on that soon.
2012-10-18T08:38:16 < emeb_mac> yeah - no *DRAM on STM32 is a big hole in their capability.
2012-10-18T08:39:05 < ds2> K70 seem to have it all
2012-10-18T08:39:23 < emeb_mac> what's the clock speed on those?
2012-10-18T08:39:33 < dongs> do tehy even have any K70 wiht hardware fp
2012-10-18T08:39:42 < ds2> emeb_mac: do you know what is the smallest package you can get with the F4xx's that have FSMC exposed?
2012-10-18T08:39:52 < ds2> the K70 is a M4 so it should have hardware FP
2012-10-18T08:40:03 < emeb_mac> ds2: the lqfp100 has muxed A/D FSMC
2012-10-18T08:40:06 < ds2> emeb_mac: not sure but over 100, IIRC
2012-10-18T08:40:21 < emeb_mac> but if you need demuxed A/D you have to go to 144
2012-10-18T08:40:22 < ds2> but muxed A/D PSRAMs are a pain to source
2012-10-18T08:40:42 < ds2> isn't there a bigger package then the 144 for the F4xx's?
2012-10-18T08:40:53 < emeb_mac> ds2: I haven't found the muxed PSRAM at the usual distys.
2012-10-18T08:41:04 < ds2> yep
2012-10-18T08:41:06 < emeb_mac> ds2 maybe a 176?
2012-10-18T08:41:32 < emeb_mac> but I think that's BGA
2012-10-18T08:41:34 < ds2> If I had room, I'd do a 245 latch
2012-10-18T08:41:42 < ds2> BGA is fine and at this point preferred
2012-10-18T08:42:27 < emeb_mac> there you go: http://www.digikey.com/product-detail/en/STM32F407IGH6/497-11765-ND/2754209
2012-10-18T08:42:42 < ds2> less pain on the size
2012-10-18T08:43:11 < ds2> now you have my interest piqued... I wonder if the BGA144 is backwards compatible with the F103's BGA144
2012-10-18T08:44:42 < emeb_mac> I think the 144 is an lqfp
2012-10-18T08:44:58 < ds2> !!??
2012-10-18T08:45:05 < emeb_mac> but the F4 is _mostly_ pin compatible.
2012-10-18T08:45:24 < emeb_mac> there are a couple core supply bypass pins not found on the lower-end parts.
2012-10-18T08:45:40 < ds2> that'd be odd... the is a BGA144 for the F103
2012-10-18T08:46:00 < ds2> time to hold my nose and look at the st site again
2012-10-18T08:46:12 < emeb_mac> hmm
2012-10-18T08:46:35 < emeb_mac> DK only sells LQFP144 and UFBGA176
2012-10-18T08:47:38 < ds2> you're right... no BGA144
2012-10-18T08:47:46 < ds2> so that is a nonstarter
2012-10-18T08:48:38 < ds2> emeb_mac: for your apps, I am guessing power is not an issue?
2012-10-18T08:49:25 < emeb_mac> ds2: yeah - I'm working stuff that sits in a cabinet hooked to AC line.
2012-10-18T08:49:58 < ds2> ahhhh
2012-10-18T08:50:11 < emeb_mac> F4 isn't too bad on power though - I've got a '405 running full speed that draws less than 90ma
2012-10-18T08:50:15 < dongs> ds2: picture of this supposed 1.5" lunix board
2012-10-18T08:50:23 < ds2> the 100MHz+ on the F4 is worse then F1xx
2012-10-18T08:50:31 < ds2> dongs: still laying it out so no pictures yet
2012-10-18T08:50:40 < emeb_mac> you can run f1xx @ 100?
2012-10-18T08:50:49 < emeb_mac> thought those topped out at 72
2012-10-18T08:50:51 < ds2> dongs: it'll also have displays and sensors on it too
2012-10-18T08:51:05 < ds2> that is right... 72MHz, clocked down as slow as possible
2012-10-18T08:51:20 < ds2> using epaper style displays to cut power down more and allow for a long sleep
2012-10-18T08:52:23 < dongs> emeb_mac: you can technically run f103 at 144mhz.
2012-10-18T08:52:27 < dongs> i did 128 just few days ago
2012-10-18T08:52:31 < dongs> it seemed to work
2012-10-18T08:52:45 < emeb_mac> dongs: overclocking?
2012-10-18T08:52:46 < dongs> yea
2012-10-18T08:53:11 < emeb_mac> yeah - I've run the f100 parts @ 48 (accidentally) and they worked fine.
2012-10-18T08:53:34 < dongs> 48mhz?
2012-10-18T08:53:37 < dongs> of course you can run it slower
2012-10-18T08:53:39 < dongs> as slow as you want
2012-10-18T08:53:43 < emeb_mac> 48MHz
2012-10-18T08:53:44 -!- claude is now known as Claude
2012-10-18T08:53:56 < dongs> you can just click if off external oscillator and manually toggle clock bit.
2012-10-18T08:54:01 < dongs> click=run
2012-10-18T08:54:01 < emeb_mac> wonder how fast you can overclock an f4?
2012-10-18T08:54:12 < ds2> Hmmmm
2012-10-18T08:54:27 < ds2> dongs: what speed xtal are you using?
2012-10-18T08:54:34 < dongs> ds2: for 128?
2012-10-18T08:54:36 < dongs> 8mhz.
2012-10-18T08:54:39 < dongs> thats hte max pll mult
2012-10-18T08:54:45 < dongs> with 12mhz you can go to 144.
2012-10-18T08:55:37 < ds2> ah
2012-10-18T08:56:44 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-18T08:58:06 < ds2> isn't there an internal high speed RC oscillator too?
2012-10-18T08:58:57 < emeb_mac> yeah
2012-10-18T08:59:01 < emeb_mac> 8MHz nominal
2012-10-18T08:59:05 < dongs> but you have to /2 it
2012-10-18T08:59:10 < dongs> so max with intrc is 64mhz
2012-10-18T08:59:13 < ds2> Ohhhh
2012-10-18T08:59:19 < dongs> 8/2 into pll -> x 16 max = 64
2012-10-18T08:59:23 < ds2> on a side note - is that one good enough for USB?
2012-10-18T08:59:23 < emeb_mac> yup
2012-10-18T08:59:26 < emeb_mac> no
2012-10-18T08:59:35 < emeb_mac> usb needs +/-0.25%
2012-10-18T08:59:44 < emeb_mac> and HSI is 2.5%
2012-10-18T08:59:50 < ds2> but don't the discoverys have USB?
2012-10-18T09:00:00 < emeb_mac> Yeah - but they have xtal.
2012-10-18T09:00:12 < ds2> really? I see bare pads
2012-10-18T09:00:19 < ds2> or is that the wrong xtal?
2012-10-18T09:00:25 < emeb_mac> which discovery?
2012-10-18T09:00:27 < dongs> they have pads for it
2012-10-18T09:00:28 < dongs> f3.
2012-10-18T09:00:32 < dongs> no crystal mounted.
2012-10-18T09:00:38 < ds2> F0, F3, F1
2012-10-18T09:01:13 < dongs>         + PLL_SOURCE_HSE
2012-10-18T09:01:13 < dongs>            - HSE (8 MHz) used to clock the PLL, and the PLL is used as system
2012-10-18T09:01:13 < dongs>              clock source.
2012-10-18T09:01:13 < dongs>            - The HSE crystal is not provided with the Discovery Kit, some
2012-10-18T09:01:13 < dongs>              hardware modification are needed in manner to connect this crystal.
2012-10-18T09:01:15 < dongs>              For more details, refer to section "4.10 OSC clock" in "STM32F3 discovery kit User manual (UM1570)"
2012-10-18T09:01:19 < dongs> readme for USB examples on f3discovery
2012-10-18T09:01:38 < dongs> no magic
2012-10-18T09:01:40 < ds2> so that USB connector is not useable as is?
2012-10-18T09:01:40 < emeb_mac> whaddaya want for $15
2012-10-18T09:01:45 < dongs> ds2: correct
2012-10-18T09:02:17 < emeb_mac> f4 discovery has usb otg connector, but has xtal installed.
2012-10-18T09:02:26 < ds2> nice to have it confirmed
2012-10-18T09:02:52 < dongs> oh!
2012-10-18T09:02:58 < ds2> same goes with the 32Khz source?
2012-10-18T09:02:59 < dongs> there's a way to use MCO from the stlink part
2012-10-18T09:03:02 < dongs> at 8mhz
2012-10-18T09:03:07 < dongs> you have to jumper some shit
2012-10-18T09:03:09 < emeb_mac> Aha - cute
2012-10-18T09:03:13 < dongs> stlink part is crystal'd so
2012-10-18T09:03:15 < ds2> nice
2012-10-18T09:03:17 < dongs> that should be good
2012-10-18T09:03:22 < dongs> but really a HC48 crystal is checap
2012-10-18T09:03:27 < dongs> er hc49
2012-10-18T09:03:29 < dongs> 8mhz
2012-10-18T09:03:34 < emeb_mac> but doesn't help ds2 on a custom board.
2012-10-18T09:04:02 < dongs> how so
2012-10-18T09:04:07 < dongs> did he forget to stick a crystal on it?
2012-10-18T09:04:37 < emeb_mac> no - he's still designing, but will have to put xtal on since he has no stlink onboard.
2012-10-18T09:04:37 < ds2> no, I am trying to shrink it
2012-10-18T09:04:49 < ds2> the 32KHz smaller then the 8MHz xtal
2012-10-18T09:04:49 < dongs> well, just use 12mhz xtal. you can get them in 3225 size.
2012-10-18T09:04:51 < dongs> done deal.
2012-10-18T09:05:00 < dongs> or go for 24mhz
2012-10-18T09:05:06 < dongs> that can be even smaller
2012-10-18T09:05:13 < ds2> does any crystal work with USB?
2012-10-18T09:05:24 < dongs> as long as it can pll to 48mhz i think
2012-10-18T09:05:34 < emeb_mac> yup
2012-10-18T09:05:37 < dongs> i know 12mhz works because thats what I used on usb-capable boards.
2012-10-18T09:05:50 < dongs> and im using 12mhz in other shit so i just reused crystals/pads
2012-10-18T09:05:54 < emeb_mac> even 25 works with proper divisors
2012-10-18T09:05:57 < ds2> Ooh I see.
2012-10-18T09:06:08 < ds2> is there anyway I can use the 32KHz one and PLL it to 48MHz?
2012-10-18T09:06:25 < dongs> i dont think you can use rtc crystal for usb/clock..
2012-10-18T09:06:30 < emeb_mac> hmmm - don't think that the 32k hooks to the HS stuff
2012-10-18T09:06:39 < ds2> the 32KHz is only for RTC?
2012-10-18T09:06:47 < emeb_mac> that would need a hella big multiplier
2012-10-18T09:07:07 < emeb_mac> probably push the PLL loop params off the edge.
2012-10-18T09:07:15 < ds2> use 3 stages
2012-10-18T09:07:17 < emeb_mac> (phase noise, etc)
2012-10-18T09:07:29 < emeb_mac> if you had 'em. :)
2012-10-18T09:07:38 < ds2> true
2012-10-18T09:08:38 -!- Claude is now known as claude
2012-10-18T09:09:33 < emeb_mac> checked the f103 datasheet - the 32k osc does not hook to the main tree.
2012-10-18T09:09:41 < dongs> thats what i figured
2012-10-18T09:12:03 -!- claude is now known as Claude
2012-10-18T09:16:56 < dongs> how do I set SWOENA in ITM registers?
2012-10-18T09:17:35 < dongs> ITM TCR: SWOENA
2012-10-18T09:21:02 < dongs> holy shit waht the fuck
2012-10-18T09:21:05 < dongs> http://www.segger2.com/index.php?page=Thread&threadID=608 ???
2012-10-18T09:21:41 < emeb_mac> ?
2012-10-18T09:22:50 < dongs> just wanna know how to enable SWO
2012-10-18T09:23:04 < dongs> this seems idiotic, core_cm3.h has all defines but these fucknuts are touching direct registers
2012-10-18T09:23:22 < emeb_mac> some folks don't like using the libs I guess.
2012-10-18T09:23:24 -!- dekar_ [~dekar@212.255.38.104] has joined ##stm32
2012-10-18T09:23:27 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-18T09:24:05 < emeb_mac> I did an experiment while doing one of my F100 products - coded it up w/ STM32 StdPeriph and also w/ direct writes.
2012-10-18T09:24:25 < emeb_mac> lib version was like 8-12k code, direct write version was < 1k.
2012-10-18T09:24:32 < ds2> anyone here tried uCLinux on the STM32s?
2012-10-18T09:24:53 < emeb_mac> not me. sounds fun tho.
2012-10-18T09:25:28 < emeb_mac> I think most of the folks here using OS end up w/ chibios or nuttx
2012-10-18T09:25:47 < ds2> I see. not familiar with those
2012-10-18T09:26:21 < emeb_mac> chibios already has drivers for most STM32 peripherals so it's easy to get going quick.
2012-10-18T09:26:49 < ds2> does that work with the libraries from ST?
2012-10-18T09:27:14 < dongs> i think nuttx has their own shit
2012-10-18T09:27:17 -!- dekar [~dekar@212.255.46.139] has quit [Ping timeout: 272 seconds]
2012-10-18T09:27:20 < dongs> acutally i think they all so
2012-10-18T09:27:21 < dongs> do
2012-10-18T09:27:26 < dongs> reinventin the fucking wheel
2012-10-18T09:27:43 < emeb_mac> http://chibios.org/dokuwiki/doku.php
2012-10-18T09:28:11 < emeb_mac> yeah - hardly anyone uses the ST libs
2012-10-18T09:28:53 < ds2> isn't that the only way to get stuff like capsense?
2012-10-18T09:29:28 < emeb_mac> maybe. that's specialized enough that the FOSS OSes probably don't support it.
2012-10-18T09:29:51 < dongs> are capsense libs closedsores or something?
2012-10-18T09:29:57 < ds2> think so
2012-10-18T09:29:57 < dongs> dudes like libopencm3 rewrote fucking everything
2012-10-18T09:30:02 < dongs> and gcc-ified all teh shit
2012-10-18T09:30:16 < ds2> but it is unclear if they are available for all STM32s
2012-10-18T09:32:48 < emeb_mac> capsense often requires special I/O structures.
2012-10-18T09:32:50 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Read error: Connection reset by peer]
2012-10-18T09:33:08 < emeb_mac> usually the algorithms are "sekrit sauce"
2012-10-18T09:33:35 < ds2> *nod*
2012-10-18T09:33:50 < emeb_mac> and they may be timing sensitive, so be careful what else is running in the background / priorities.
2012-10-18T09:46:22 < dongs> ok so
2012-10-18T09:52:24 < zyp> *yawn*
2012-10-18T09:52:38 < zyp> I should start going to bed earlier
2012-10-18T09:53:45 < dongs> zyp, halp me with this swo shit
2012-10-18T09:56:13 < zyp> what are you trying to do?
2012-10-18T09:56:55 -!- Claude is now known as claude
2012-10-18T09:58:57 < dongs> enable it
2012-10-18T09:59:02 < dongs> so i can use swv
2012-10-18T10:00:43 -!- tavish [7838a931@gateway/web/freenode/ip.120.56.169.49] has joined ##stm32
2012-10-18T10:01:09 < zyp> use it for what?
2012-10-18T10:02:20 < emeb_mac> debuggin?
2012-10-18T10:02:36 < dongs> dump realtime data
2012-10-18T10:02:51 < emeb_mac> sending naughty messages to the host pc.
2012-10-18T10:07:43 < zyp> swo is just the output part of it, it can be used to output a bunch of different shit from the DWT and ITM modules in a bunch of different formats
2012-10-18T10:08:54 < zyp> it's a bit hard to tell you how to enable it without knowing what you want to send and in which format it is expected
2012-10-18T10:09:32 < zyp> ideally your debugger that decodes it knows what it expects and should do all the enabling for you
2012-10-18T10:12:11 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-18T10:12:22 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-18T10:21:20 < dongs> hmm.. i think its working.. at least shit shjows up in logic analyzer disply.. http://i.imgur.com/TfiDp.png but it still says trace: no sync
2012-10-18T10:21:25 < dongs> i saw it say once trace: hw overun
2012-10-18T10:21:26 < dongs> hmm
2012-10-18T10:22:02 < dongs> and update is kinda blocky... so I think its just pulling it repeatedly out of debugger like normal shit in watch window
2012-10-18T10:34:13 < zyp> hmm, it's not strange the lines on that chumby stuff are length matched, it's ddr
2012-10-18T10:35:37 < emeb_mac> yep
2012-10-18T10:37:13 < dongs> zyp: yea, right, i noticed that a few hours after pasting it
2012-10-18T10:40:52 < emeb_mac> did you find the skew specs on the lpc outputs?
2012-10-18T10:41:22 < zyp> I didn't look, but I assume that part would be in the datasheet
2012-10-18T10:49:35 < emeb_mac> looks like about 2.5ns typ-max variation. well within the window.
2012-10-18T10:49:52 < emeb_mac> gn...
2012-10-18T10:49:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-18T10:52:43 -!- claude is now known as claude_work
2012-10-18T11:00:14 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 240 seconds]
2012-10-18T11:14:12 < dongs> teh window
2012-10-18T11:15:16 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-18T11:18:03 < Tectu> dongs, so today there was a gil in the train, wearing a shirt, labled "kawaii"
2012-10-18T11:18:53 < zyp> that's so awesome
2012-10-18T11:21:18 < cjbaird> Not as awesome as the girl I saw in a full-body pikachu halloween costume back in July..
2012-10-18T11:21:31 < cjbaird> two bricks: http://i.imgur.com/1dMjV.jpg
2012-10-18T11:35:48 < jpa-> yay, got texane stlink working in a ubuntu virtualbox
2012-10-18T11:36:02 < jpa-> after the atollic crappy gdbserver this is heaven
2012-10-18T11:36:09 < jpa-> karlp <3
2012-10-18T11:38:00 -!- mrcan_ [~mrcan@78.173.40.143] has joined ##stm32
2012-10-18T11:38:00 -!- mrcan_ [~mrcan@78.173.40.143] has quit [Changing host]
2012-10-18T11:38:00 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-18T11:40:59 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-18T12:14:32 < karlp> jpa-: heh
2012-10-18T12:14:58 < karlp> oh, jpa- I got my timers working.  seemed to have been some specific ordering of setting and clearing and enabling and resetting I didn't have spot on.
2012-10-18T12:15:21 < karlp> I want to do some more experiments yet to work out what bit I wasn't understanding.
2012-10-18T12:15:46 < karlp> I feel I'll need to learn some arm assembler to fully work it out, but it's working at least now.
2012-10-18T12:18:02 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-18T12:18:04 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-18T12:21:15 < Posterdati> hi
2012-10-18T12:33:06 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-18T12:33:47 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds]
2012-10-18T12:34:29 <+Steffanx> lo
2012-10-18T12:38:54 < BusError_> Hey Steffanx ;-)
2012-10-18T12:57:01 < jpa-> bah.. i wish STM32 would just crash on null pointer
2012-10-18T13:03:20 < zyp> just use the MPU and protect the area?
2012-10-18T13:05:56 < jpa-> yeah, i guess that would work
2012-10-18T13:06:17 < jpa-> it could have been done by default
2012-10-18T13:06:31 < dongs> commence chats
2012-10-18T13:06:36 < dongs> Tectu: pics or it didnt happen
2012-10-18T13:06:50 < dongs> cjbaird: old
2012-10-18T13:11:12 < Posterdati> hi
2012-10-18T13:12:14 < Posterdati> please I need help on ste101p driver: I'm using lwip, but seems that ste101p cannot comunicate over mdio bus (olimex stm32-p107 board)
2012-10-18T13:23:48 -!- tavish [7838a931@gateway/web/freenode/ip.120.56.169.49] has quit [Ping timeout: 245 seconds]
2012-10-18T13:39:14 < scrts> olimex has no examples?
2012-10-18T13:45:44 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-18T13:46:06 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-18T13:46:41 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-18T13:47:10 < jpa-> for some reason i always find it really exciting to wait if my git rebase -i goes through without problems
2012-10-18T13:48:23 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Client Quit]
2012-10-18T13:48:44 <+Steffanx> lol
2012-10-18T13:54:56 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-18T13:55:44 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Client Quit]
2012-10-18T13:57:08 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-18T13:59:37 < Tectu> so the F4 has a maximum GPIO frequence of 100MHz.... When I need to toggle 16 bits on the same port, does it take 16*(1/100MHz)  ?
2012-10-18T14:00:00 < jpa-> no
2012-10-18T14:00:34 < jpa-> same port can toggle at once, and even different ports is only limited by the CPU frequency
2012-10-18T14:00:43 < Tectu> okay
2012-10-18T14:00:53 < Tectu> it's about an R2R network
2012-10-18T14:00:59 < jpa-> the GPIO max frequency simply means that if you toggle the signal faster than 100MHz, the external voltage will not change fast enough
2012-10-18T14:01:10 < Tectu> ah
2012-10-18T14:01:34 < jpa-> with R2R, you will have some crazy values during the switching time, but some lowpass filtering might get rid of them
2012-10-18T14:01:35 < Tectu> I am not sure how to determine the maximum frequency of a sine with a given resolution and my 100MHz GPIO
2012-10-18T14:02:21 < claude_work> Tectu, http://en.wikipedia.org/wiki/Quantization_(signal_processing)
2012-10-18T14:02:45 < jpa-> Tectu: only by testing, I'm afraid
2012-10-18T14:02:52 < jpa-> don't expect to get over 10MHz sine
2012-10-18T14:02:54 < Tectu> jpa-, huh??
2012-10-18T14:03:07 < Tectu> jpa-, dosen't that depend on my resolution?
2012-10-18T14:03:14 < Tectu> I thought I'd go with 12 bit
2012-10-18T14:03:29 < Tectu> but with R2R network, 16 bit could be done easily... but will you see a difference?
2012-10-18T14:03:47 < claude_work> not with your 8bit scope ;)
2012-10-18T14:03:56 < Tectu> hrhr
2012-10-18T14:04:01 < jpa-> Tectu: do you have 0.001% accurate resistors?
2012-10-18T14:04:09 < Tectu> jpa-, 0.1%
2012-10-18T14:04:25 < jpa-> that's good only up to 9 bit
2012-10-18T14:04:44 < Tectu> I guess farnell dosen't have any problems with 0.01% ?
2012-10-18T14:04:58 < jpa-> the accurate resistor part is the hardest thing about doing your own R2R dacs
2012-10-18T14:05:05 < jpa-> real DACs have laser-trimmed resistors
2012-10-18T14:05:24 < Tectu> jpa-, yes, But I 've seen only go up to a few hundret KHz
2012-10-18T14:06:34 < Tectu> 0.01% resistors cost like 100 times more :x
2012-10-18T14:06:51 < Tectu> 20bucks for 100R 0.25W 0.01% wtf
2012-10-18T14:07:44 < claude_work> Tectu, how about this DAC? http://www.ti.com/product/ths8136
2012-10-18T14:08:16 < Tectu> 180MHz wtf
2012-10-18T14:08:25 < claude_work> yeah and cheap too!
2012-10-18T14:08:26 < Tectu> claude_work, I want to build some signal generator, btw
2012-10-18T14:08:34 < Tectu> in case of this matters ;-)
2012-10-18T14:08:55 < Tectu> I thought about something like 0.1Hz to 10MHz sine, triangle, rectangle
2012-10-18T14:09:05 < Tectu> with and without noise :)
2012-10-18T14:09:31 < claude_work> these THS video DAC can be (ab)used as generic DACs too . Plus they allready have build in output buffers for 75R loads
2012-10-18T14:09:50 < Tectu> jep, have seen that
2012-10-18T14:10:05 < Tectu> and quite hard to get
2012-10-18T14:10:05 < Tectu> http://ch.farnell.com/texas-instruments/ths8136iphp/10bit-180msps-dac-vid-48htqfp/dp/1755236?Ntt=THS8136
2012-10-18T14:10:19 < claude_work> ah TI samples aren't hard to get ;)
2012-10-18T14:10:33 < claude_work> i see , .ch ?
2012-10-18T14:11:03 < Tectu> switzerland
2012-10-18T14:11:13 < claude_work> nachbarland hier ;)
2012-10-18T14:11:22 < Tectu> deutsches reich?
2012-10-18T14:11:46 < claude_work> sagen wir deutschland , reich hat so einen negativen beigeschmack ^^
2012-10-18T14:11:56 < Tectu> musste sein, sorry ^^
2012-10-18T14:12:00 < Tectu> bist du ja sicher gewohnt, oder? ;-)
2012-10-18T14:12:07  * BusError_ checks his modem for problems as the discussion turns all gibberish again :>
2012-10-18T14:12:34  * BusError_ think it might be a parity error or something
2012-10-18T14:13:15 < claude_work> haha ja schon ok . but let's stick to english ... BusError_ is a bit picky with foreign languages ;)
2012-10-18T14:13:21 < Tectu> claude_work, that chips looks very very good
2012-10-18T14:13:24 < claude_work> 7bit encoding
2012-10-18T14:13:31 < Tectu> huh?
2012-10-18T14:13:33 < BusError_> bloody foreigners! :-)
2012-10-18T14:13:48 < Tectu> i am not forgein, I am swiss :i
2012-10-18T14:13:48 < Tectu> :o
2012-10-18T14:13:56 < claude_work> you have to run the lines trough an enigma BusError_
2012-10-18T14:14:19 < claude_work> lampenfeld for today : 007
2012-10-18T14:14:57 < BusError_> https://plus.google.com/111387094029238541867/posts/jGTNqXJ8SXP <- need to order the pcbs, now!
2012-10-18T14:15:13 <+Steffanx> Whoa, someone really uses g+ ?!
2012-10-18T14:15:43 < BusError_> well, cheap bloggy thing really
2012-10-18T14:15:47 < Tectu> BusError_, want!
2012-10-18T14:15:58 < Tectu> (the board, not the g+)
2012-10-18T14:17:03 < BusError_> well I'll have spares. gonna order 20 (sweet spot it seems) -- if they work that is :-)
2012-10-18T14:17:21 < Tectu> BusError_, how much would it cost?
2012-10-18T14:18:04 < Tectu> claude_work, http://ch.farnell.com/texas-instruments/dac3162irgzt/ic-dac-12bit-500msps-vqfn-48/dp/2084397
2012-10-18T14:18:16 < BusError_> ni idea, populated. mostly depends on what STM I stick on it... probably like 20 GBP, with a F4. possibly less
2012-10-18T14:18:21 < claude_work> phew hefty price imo
2012-10-18T14:18:28 < Tectu> WTF is GBP...
2012-10-18T14:18:35 < claude_work> erhmm the DAC not BusError_ PCB !
2012-10-18T14:18:35 < Tectu> claude_work, indeed :D
2012-10-18T14:19:19 < BusError_> the STM32 are not cheap at farnell :/ well considering the price of the disco, and that I don't want to buy 500 of them :>
2012-10-18T14:19:28 <+Steffanx> "WTF is GBP..." <= WTF Tectu
2012-10-18T14:19:46 < Tectu> GreatBritainPound?
2012-10-18T14:19:52 < zyp> yep
2012-10-18T14:19:59 < Tectu> seriously?
2012-10-18T14:20:01 < Posterdati> scrts: no example from olimex for rev. a
2012-10-18T14:20:02 < zyp> yes
2012-10-18T14:20:38 < Tectu> Steffanx, of course i know GBP, it stand on each window of the 12 mio banks in my 750ppl village.
2012-10-18T14:20:42 < scrts> Posterdati: do You see data coming on MDIO using a scope?
2012-10-18T14:21:14 < Posterdati> wait
2012-10-18T14:21:15 <+Steffanx> Tectu, your other half didn't know that
2012-10-18T14:22:03 < claude_work> Tectu, don't know how digikey handles shipping to .ch , but for .de it is like 60euros and shipping is free. i don't order from farnell anymore , digikey is way cheaper , it really up to date with the parts and stocks almost everything
2012-10-18T14:22:38 < Tectu> claude_work, well, I can places order for farnell over my fachhochschule, so no charges at all \o/
2012-10-18T14:22:48 < claude_work> ah okay
2012-10-18T14:23:06 < claude_work> go for the DAC then ;)
2012-10-18T14:23:12 <+Steffanx> Tectu, the only thing a "fachhochschule" is useful for :)
2012-10-18T14:23:19 < Tectu> Steffanx, indeed
2012-10-18T14:23:22 < claude_work> +mensa food
2012-10-18T14:23:34 < BusError_> claude_work, no customs on the digikey orders?
2012-10-18T14:23:51 < Tectu> also, I really don't know what's «fachhochschule» in english. all the netherland guys talking to me actually always say «fachhochschule»
2012-10-18T14:24:04 < Tectu> claude_work, look this one:   http://www.ti.com/product/tlv5637
2012-10-18T14:24:12 < Posterdati> scrts: nothing
2012-10-18T14:24:31 < claude_work> BusError_, digikey pays for the customs iirc , so far never had to pay the ups guy. even with 1keuro+ orders
2012-10-18T14:24:40 < scrts> then something is bad with the low level driver or lwip
2012-10-18T14:24:50 <+Steffanx> Tectu, that's because we/I sort of know what you mean :)
2012-10-18T14:25:02 < Posterdati> scrts: yes, I know
2012-10-18T14:25:10 < scrts> OR You haven't switched the I/O matrix to enable special purpose function on the pin
2012-10-18T14:25:11 < Tectu> Steffanx, i guess so, yes
2012-10-18T14:25:12 < scrts> did You do that?
2012-10-18T14:25:19 < Tectu> Steffanx, but explain it to BusError_
2012-10-18T14:25:24 < Tectu> or an american guy
2012-10-18T14:25:39 < Tectu> BusError_, btw... GBP, don't you guys over there use EUR as well?
2012-10-18T14:25:39 <+Steffanx> Their system is weird, so you can't explain :)
2012-10-18T14:25:42 < BusError_> claude_work, maybe I'll try again.. I ordered /once/ from them and was hammered with customs/vat etc
2012-10-18T14:25:59 < Tectu> Steffanx, hah, you think their is weird? never heard of the swiss system, i guess
2012-10-18T14:26:03 < zyp> Tectu, GB is using GBP
2012-10-18T14:26:06 < claude_work> yeah but at least for .de they changed that 3-4 years ago or so
2012-10-18T14:26:24 <+Steffanx> BusError_, here you can sort of order stuff including tax. As long as i order from nl.digikey.com.
2012-10-18T14:26:27 < BusError_> actually I'n not british, just live here. Thats why I can make fun of bloody foreigners and non-english gibberish :-)
2012-10-18T14:26:28 <+Steffanx> Never tried it myself though
2012-10-18T14:26:33 < zyp> several EU countries kept their original currency instead of switching to EUR
2012-10-18T14:26:36 < Tectu> I see
2012-10-18T14:28:18 <+Steffanx> That country in the middle of EU was smart enough to not even join the EU :)
2012-10-18T14:28:29 < Posterdati> scrts: it seems that phy isn't mapped on the bus, I try different addresses
2012-10-18T14:28:56 < scrts> Posterdati: no, if You see nothing coming out of the stm32, then what answer do You expect?
2012-10-18T14:29:07 < scrts> if nothing knock on the door, do You go to check?
2012-10-18T14:29:25 < Tectu> Steffanx, ;-)
2012-10-18T14:29:29 < Posterdati> scrts: the phy won't answer
2012-10-18T14:30:06 < Tectu> claude_work, okay, made my choice:)
2012-10-18T14:30:12 < Tectu> http://ch.farnell.com/texas-instruments/tlv5637cd/10bit-dac-smd-5637-soic8/dp/8455813
2012-10-18T14:31:08 < scrts> Posterdati: again: what do You see using a scope on stm32 pins, where MDC/MDIO pins are?
2012-10-18T14:33:18 < Posterdati> mdc = 0v
2012-10-18T14:33:34 < Posterdati> mdio = 0v
2012-10-18T14:33:51 < scrts> reset the stm32 and wait some amount of time until the software starts
2012-10-18T14:34:04 < Posterdati> yes I did so
2012-10-18T14:34:07 < scrts> check if You can see anything there?
2012-10-18T14:34:10 < scrts> nothing right?
2012-10-18T14:34:28 < Posterdati> yes nothing
2012-10-18T14:34:50 < scrts> then how come the PHY will answer if STM32 is not asking for data?
2012-10-18T14:35:02 < scrts> it's not the PHY not answering, it's STM32 not asking
2012-10-18T14:35:15 < Posterdati> yes, but stm32 is configured!
2012-10-18T14:35:26 < zyp> obviously not.
2012-10-18T14:35:30 < scrts> it's configured wrong then
2012-10-18T14:35:30 < Posterdati> ...
2012-10-18T14:35:39 < zyp> if it's not sending any data, it's not configured correctly.
2012-10-18T14:36:09 < scrts> the FIRST thing to check: did You enable MDC/MDIO special function on those STM32 pins?
2012-10-18T14:36:34 < claude_work> Tectu, but that DAC is far away from 10MHz sine
2012-10-18T14:36:47 < claude_work> more like 100kHz sine
2012-10-18T14:37:20 < dongs> Abtastrate: 278kSPS
2012-10-18T14:37:22 < dongs> trash dac
2012-10-18T14:37:34 < Tectu> claude_work, huh? how did you get that from?
2012-10-18T14:37:52 < Posterdati> scrts: yes, but there's a 166.6 MHz signal on ref clock
2012-10-18T14:38:02 < Posterdati> it should be 25 MHz
2012-10-18T14:38:03 < Tectu> claude_work, it has seattling time from 800ns  -> 1.25MHz with 10 bit sine -> 125KhZ, i guess?
2012-10-18T14:38:38 < Tectu> claude_work, maybe I should stick with R2R and 0.1% ?
2012-10-18T14:38:52 < scrts> Posterdati: then You've also set the output clock wrong
2012-10-18T14:39:01 < scrts> and the whole configuration is probably wrong for the MII setup
2012-10-18T14:39:03 < Posterdati> seems so
2012-10-18T14:39:19 < scrts> read the docu how to enable MII on STM32
2012-10-18T14:39:27 < scrts> plus read the docu how to set up clock output
2012-10-18T14:39:36 < scrts> plus read the docu how to set up MDC/MDIO
2012-10-18T14:40:02 -!- cjbaird [~cjb@ppp121-45-173-137.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-18T14:41:56 < Tectu> claude_work, bit better, eh?   http://ch.farnell.com/texas-instruments/dac2902-250/dual-12-bit-d-a-konv-48-tqfp/dp/1754790
2012-10-18T14:42:02 < Tectu> also three times more expensive :D
2012-10-18T14:43:05 < Tectu> claude_work, damn, that's still far away from 10MHz sine -.-
2012-10-18T14:43:11 < dongs> 1 Jahr
2012-10-18T14:43:17 < claude_work> Tectu, http://www.ti.com/product/tlc7524 , used this DAC in an vector arcade thingy a while ago..
2012-10-18T14:43:28 < Tectu> vector arcade?
2012-10-18T14:43:39 < claude_work> yeah vectrex/atari like
2012-10-18T14:43:50 < Tectu> ah
2012-10-18T14:43:54 < zyp> it's like a matrix arcade, minus one dimension
2012-10-18T14:44:19 < claude_work> DAC is easy to use , just 8 bits of data , write strobe signal and ready
2012-10-18T14:44:42 < claude_work> fits quite good the stm32 fsmc interface
2012-10-18T14:46:46 < claude_work> just tie the DAC data lines to the fsmc's lower 8 bits (D0..D7) and the DAC write strobe to the fsmc's WR line . ready for DMAing to the DAC with no ugly glitches and stuff
2012-10-18T14:47:13 -!- dekar_ [~dekar@212.255.38.104] has quit [Quit: This computer has gone to sleep]
2012-10-18T14:48:07 < zyp> apart from bus contention
2012-10-18T14:49:30 < claude_work> yeah as long as you don't try to read from the dac ... ;)
2012-10-18T14:49:43 < claude_work> and the dac is the only part on the bus
2012-10-18T14:52:26 < jpa-> Tectu: for a signal generator, i might go with a fast 8 bit dac and a separate ~12 bit digital potentiometer for setting a gain; so you can control the signal amplitude over a wider range
2012-10-18T14:55:19 < claude_work> good idea
2012-10-18T14:55:35 < claude_work> dc offset is handy too
2012-10-18T14:58:38 < zyp> claude_work, you still need to fetch the data to write to the dac from somewhere
2012-10-18T14:58:48 < zyp> somewhere which may have bus contention.
2012-10-18T15:01:14 < BusError_> claude_work, checking digikey.co.uk here, they also seem to have a free shipping over 50 GBP (30 more than farnell).. a lot of parts, some are massively cheaper (stm32f4) and some are more expensive than farnell.. not a full on replacement. I'll make an order anyway to try them out...
2012-10-18T15:01:41 < Posterdati> scrts: not I've got a 50 MHz clock on PHY sclk pin, I presume it is correct (see datasheet at page 11) because I want RMII protocol
2012-10-18T15:01:58 < claude_work> zyp yes you are probably right , so WR plus and AND gate with A19 for example
2012-10-18T15:02:09 < Posterdati> scrts: still no signals on mdio port
2012-10-18T15:03:12 < zyp> still wrong settings for mdio pins then.
2012-10-18T15:03:37 < Posterdati> yes
2012-10-18T15:04:16 -!- amitofu [~str@cannabis.dataforce.net] has quit [Ping timeout: 252 seconds]
2012-10-18T15:04:26 < Tectu> jpa-, interesting
2012-10-18T15:04:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-18T15:05:11 < Tectu> jpa-, is a fast 8-bit DAC still better than R2R?
2012-10-18T15:05:34 < BusError_> claude_work, don;t you need a NOT on the WR for the AND to work with the A19 line ? thats what I had figured out when making my other board...
2012-10-18T15:05:53 < BusError_> A is acrive high, WR is active low...
2012-10-18T15:06:12 < claude_work> um yeah ..
2012-10-18T15:06:59 < claude_work> or use A19 with an inverter as chip select signal
2012-10-18T15:10:42 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-18T15:12:06 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-18T15:14:12 < jpa-> Tectu: yeah, i'd say it is
2012-10-18T15:19:27 < Tectu> maybe I should take some serious DDS...
2012-10-18T15:20:35 < jpa-> maybe, though if you are interested in generating real time (low-latency) signals, a simple DAC may be better
2012-10-18T15:23:43 < Tectu> jpa-, well, it's just for some homebrew signal generator, but something that should be useful
2012-10-18T15:23:44 < Tectu> if a DDS has a bandwich of 200kHz, that's the maximum sine freq I could do?
2012-10-18T15:23:52 < Tectu> http://ch.farnell.com/analog-devices/ad9834bruz/10bit-dds-smd-9834-tssop20/dp/8398330
2012-10-18T15:24:14 < BusError_> half that
2012-10-18T15:24:25 < Tectu> how comes?
2012-10-18T15:24:27 < Tectu> ah
2012-10-18T15:24:28 < Tectu> :D
2012-10-18T15:24:33 < Tectu> the somehow-theorem
2012-10-18T15:25:04 < BusError_> http://en.wikipedia.org/wiki/Nyquist_frequency
2012-10-18T15:25:52 < Tectu> jpa-, well, what would you do for a sine to 10Mhz?   8bit DAC ?
2012-10-18T15:26:09 < jpa-> maybe a premade DDS
2012-10-18T15:26:26 < jpa-> i have my own signal generator with analog stuff, MAX038
2012-10-18T15:26:33 < jpa-> finding high-speed opamps is hard by itself
2012-10-18T15:28:14 < Tectu> jpa-, MAX038 seems hard to get as well
2012-10-18T15:29:09 <+Steffanx> ebay Tectu :P
2012-10-18T15:33:32 < Posterdati> zyp: config seems correct
2012-10-18T15:34:37 < dongs> RS in UK has F3discovery in stock!
2012-10-18T15:38:45 < Laurenceb> sup
2012-10-18T15:38:48 < Laurenceb> wut
2012-10-18T15:38:50 < Laurenceb> OMG
2012-10-18T15:40:41 < dongs> https://youtube.com/watch?v=Vlyc2wEsak4#t=237s
2012-10-18T15:40:42 < dongs> loll
2012-10-18T15:40:48 < dongs> Laurenceb: yea  laready ordered 3
2012-10-18T15:40:50 < dongs> I HOPE OYU GET NONE
2012-10-18T15:40:53 < Laurenceb> ordered
2012-10-18T15:40:56 < dongs> fu
2012-10-18T15:41:00 < Laurenceb> will pick up from depot
2012-10-18T15:41:03 < dongs> Ya i bet you will
2012-10-18T15:41:06 < Laurenceb> after some food
2012-10-18T15:41:18 < zyp> ordered
2012-10-18T15:41:20 < dongs> o shit
2012-10-18T15:41:21 < dongs> only 30 left
2012-10-18T15:41:26 < Laurenceb> i work on the same industrial estate
2012-10-18T15:41:29 < dongs> RS Stock No.: 769-2091
2012-10-18T15:41:29 < dongs> 30 in stock for next day delivery - back order for despatch 11/01/2013 for delivery within 1 working day(s)
2012-10-18T15:41:46 < zyp> ordered one F3 and one F0
2012-10-18T15:42:37 < zyp> but I suspect they'll bitch because I didn't order it as a company
2012-10-18T15:43:00 < HTT-Bird> zyp: :<
2012-10-18T15:45:11 < Laurenceb> got 2
2012-10-18T15:45:18 < Laurenceb> pickup from 8am tomorrow
2012-10-18T15:45:58 <+Steffanx> heh
2012-10-18T15:46:15 < Laurenceb> eek thats a long delay
2012-10-18T15:46:45 <+Steffanx> zyp, yeah that's why i never order there
2012-10-18T15:47:29 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-18T15:49:58 < zyp> hmm, stellaris launchpad order still has status as pending
2012-10-18T15:50:07 < zyp> estimated shipping is 16nov
2012-10-18T15:50:11 < Laurenceb> lol they says its F0
2012-10-18T15:50:15 < zyp> still a month
2012-10-18T15:50:27 <+Steffanx> uh Laurenceb ?
2012-10-18T15:50:28 < zyp> Laurenceb, pic and order code says f3
2012-10-18T15:50:33 < zyp> but title says f0
2012-10-18T15:50:37 < Laurenceb> yeah
2012-10-18T15:50:50 < Laurenceb> http://uk.farnell.com/stmicroelectronics/stm32f3discovery/eval-stm32f3-cortex-m4-discovery/dp/2215108
2012-10-18T15:50:52 < Laurenceb> more
2012-10-18T15:51:03 < Laurenceb> but they have annoying order conditions in uk
2012-10-18T15:51:30 < Tectu> not sure how to get the best value for R in an R2R network?
2012-10-18T15:54:23 < Laurenceb> now we just need some libraries for it...
2012-10-18T15:54:40 < Laurenceb> and texane etc
2012-10-18T15:54:49 < Tectu> use openocd
2012-10-18T15:55:01 < Laurenceb> it supports F3?
2012-10-18T15:55:25 < Tectu> yes
2012-10-18T15:55:33 < Tectu> http://hackaday.com/2012/09/27/stm32-f4-discovery-tutorial-using-open-source-tools/
2012-10-18T15:56:21 < Laurenceb> fail
2012-10-18T15:56:30 < Laurenceb> stm32-f4-discovery
2012-10-18T15:56:46 < Laurenceb> wait what
2012-10-18T15:56:47 < zyp> if you read the article, it says f3
2012-10-18T15:57:00 < Tectu> lawl
2012-10-18T15:57:02 <+Steffanx> picture is f3 too :)
2012-10-18T15:57:06 < Laurenceb> is everyone dyslexic or something
2012-10-18T15:57:19 < zyp> apparently
2012-10-18T15:57:20 <+Steffanx> or something, yes
2012-10-18T15:57:38 <+Steffanx> At least "or something" can be anything
2012-10-18T15:57:49 < karlp> it uses the same flash loader as f4 right?
2012-10-18T15:58:04 < dongs> <randomlunixfgt> It took me less than a few hours (mostly due to pebkac) to build a toolchain from scratch to program a chibios demo into my disco F4. Thanks gentoo for crossdev.
2012-10-18T15:58:07 < dongs> haha
2012-10-18T15:58:10 < zyp> I would have guessed same as F1
2012-10-18T15:58:16 < BusError_> wahts the differences between f3 and f4 ? haven't seen the f3 brief yet
2012-10-18T15:58:22 < karlp> it's an m4f zyp,
2012-10-18T15:58:36 < zyp> yes, but flash interface is not part of the cpu core
2012-10-18T15:58:44 < karlp> fair enough,
2012-10-18T15:58:45 < zyp> flash interface is an st peripheral
2012-10-18T15:58:47 < BusError_> guys, you can download the linaro toolchain, thats what I use on the f4 here
2012-10-18T15:58:57 < dongs> i just use tools that work
2012-10-18T15:59:01 <+Steffanx> ARMCC :P
2012-10-18T15:59:03 < dongs> ^
2012-10-18T15:59:19 < Laurenceb> its has either 16bit adc
2012-10-18T15:59:24 < dongs> <freetard> Yay, I have one of these and I really donÂ’t want to use a bunch of proprietary tools to develop on it.
2012-10-18T15:59:25 < Laurenceb> or opamps and fast adc
2012-10-18T15:59:45 < Laurenceb> sounds like my philosophy
2012-10-18T15:59:45 < Tectu> dongs, which channel is that
2012-10-18T15:59:53 < dongs> Tectu: that hackaday link
2012-10-18T15:59:57 < Laurenceb> dongs crazy mind
2012-10-18T16:00:04 < Tectu> dongs, ah
2012-10-18T16:00:17 < dongs> <freetard> I picked up an older discovery board thinking it was such a steal at $10, but I wasnÂ’t too impressed that I had to register in order to get the dev tools.
2012-10-18T16:00:21 < dongs> haha
2012-10-18T16:00:23 < dongs> I AHVE TO REGISTER!!!111
2012-10-18T16:00:25 < dongs> OHY NOESE
2012-10-18T16:02:01 < zyp> LPC4300 apparently won't play very well along with my Pin abstraction for configuring pins and controlling gpio
2012-10-18T16:02:25 < zyp> because on LPC4300, pin configuration registers and gpio registers are pretty unrelated
2012-10-18T16:02:50 < zyp> gpio is just another function block that can be selected on some of the pins
2012-10-18T16:04:24 < Laurenceb> i want the 48 pin f303 stuff
2012-10-18T16:04:28 < zyp> so I'll need to rethink that part when I port my shit to lpc4300
2012-10-18T16:04:29 < Laurenceb> stil;l no stock :(
2012-10-18T16:04:47 < dongs> you and me both.
2012-10-18T16:18:17 < Laurenceb> down to 19 stock now
2012-10-18T16:21:50 < dongs> heh
2012-10-18T16:30:01 -!- amitofu [~str@cannabis.dataforce.net] has joined ##stm32
2012-10-18T16:51:52 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-18T17:06:41 < Laurenceb> haha WTF
2012-10-18T17:06:53 < Laurenceb> i asked someone for somw photos of hardware
2012-10-18T17:06:56 < Laurenceb> *some
2012-10-18T17:07:06 < Laurenceb> they have just sent me 65 emails
2012-10-18T17:07:12 < Laurenceb> one for each photo
2012-10-18T17:07:38 < dongs> well, you got hte pics right
2012-10-18T17:08:40 < Laurenceb> lurn 2 attach
2012-10-18T17:09:10 < Laurenceb> http://i.imgur.com/x8E3u.jpg
2012-10-18T17:09:12 < Laurenceb> i lolled
2012-10-18T17:09:42 < dongs> Laurenceb's dead hardware
2012-10-18T17:10:18 < emeb_mac> that's not a little creepy
2012-10-18T17:11:15 < Laurenceb> its my stm32autozombifier
2012-10-18T17:11:43 < Tectu> damn, there are so many OpAmps out there... which one is the right...
2012-10-18T17:11:49 < emeb_mac> lol
2012-10-18T17:11:54 < dongs> Tectu: LMV321
2012-10-18T17:11:59 < dongs> you can thank me later.
2012-10-18T17:12:05 < emeb_mac> lol^2
2012-10-18T17:12:06 < Tectu> no, it's an LMH or an THS
2012-10-18T17:12:15 < Tectu> wtf is LMV at all
2012-10-18T17:12:22 < Tectu> at all -> anyways -.-'
2012-10-18T17:12:23 < dongs> no fucking idea
2012-10-18T17:12:28 < dongs> its hte opamp I use when i need an opamp
2012-10-18T17:12:39 < Tectu> then I use 741
2012-10-18T17:12:47 < emeb_mac> lol^3
2012-10-18T17:12:49 < Tectu> but this needs to go up to 30MHz rail to rail
2012-10-18T17:13:21 < emeb_mac> what you driving?
2012-10-18T17:13:26 < dongs> wtf, you think i have a ny idea what im doing?
2012-10-18T17:14:02 < Tectu> emeb_mac, output driver for signal generator
2012-10-18T17:14:14 < emeb_mac> what kind of signals?
2012-10-18T17:14:26 < Tectu> sine, triangle, rectangle, noise
2012-10-18T17:14:40 < dongs> plz implement pink noise 1st
2012-10-18T17:14:43 < Tectu> with let's say 100mA output and +-12V
2012-10-18T17:14:47 < emeb_mac> so just general purpose lab equipment?
2012-10-18T17:14:54 < Tectu> emeb_mac, jep
2012-10-18T17:15:11 < dongs> emeb_mac: pics of ledpcbs tomrow
2012-10-18T17:15:17 < emeb_mac> dongs: nice
2012-10-18T17:15:19 < Tectu> emeb_mac, sadly TI dosen't have 'general purpose lab equipement OpAmp' category :D
2012-10-18T17:15:44 < Tectu> I guess I'll go with LMH6639
2012-10-18T17:15:55 < emeb_mac> Tectu: someone here had a pretty good recommendation the other day:
2012-10-18T17:16:09 < emeb_mac> use a DSL line driver.
2012-10-18T17:16:26 < Tectu> wtf is «a DSL line driver»
2012-10-18T17:16:26 < emeb_mac> nice, broad bandwidth, good immunity do nasty loads.
2012-10-18T17:16:38 < dongs> http://liftoffexpress.com/product/stellaris/ if someone is dying for a launchscam, this guy claims to have some in stock
2012-10-18T17:16:41 < dongs> http://liftoffexpress.com/product/stellaris/
2012-10-18T17:16:42 < Laurenceb> lurn 2 parametric product search
2012-10-18T17:16:42 < dongs> oops, doublespam.
2012-10-18T17:17:01 < dongs> (at double the price and in USA, so probably scam for sure)
2012-10-18T17:17:08 < emeb_mac> Tectu: imagine what a DSL modem uses to send a signal out onto a phone line.
2012-10-18T17:17:09 < Laurenceb> hibernation module.
2012-10-18T17:17:11 < Laurenceb> lulwut
2012-10-18T17:19:10 < Laurenceb> so my friend just called me, hes just out of jail
2012-10-18T17:19:11 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 245 seconds]
2012-10-18T17:19:24 < Laurenceb>  "dude in knee deep in pussy, want to come round"
2012-10-18T17:19:54 < Laurenceb> apparently he has a load of whores from "teh internet"
2012-10-18T17:20:00 < Tectu> emeb_mac, hmmm
2012-10-18T17:20:09 < Laurenceb> excessive risk of aids?
2012-10-18T17:20:24 < emeb_mac> solution: whole-body condom.
2012-10-18T17:20:33 < dongs> gas him
2012-10-18T17:20:33 < Laurenceb> lawl
2012-10-18T17:20:56 < Laurenceb> im not sure if its gay to screw a whore together with another guy
2012-10-18T17:21:37 < Laurenceb> i think ill leave him to it
2012-10-18T17:21:42 < dongs> this might be your onlychance
2012-10-18T17:21:46 < Laurenceb> lol
2012-10-18T17:21:50 < dongs> to be near a femail  creature
2012-10-18T17:22:10 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-18T17:22:29 < dongs> none of hte exciting Laurenceb blogs ever gets logged
2012-10-18T17:22:40 < Laurenceb> http://thumbs.dreamstime.com/thumblarge_429/1250515601KqMx9N.jpg
2012-10-18T17:22:52 < dongs> ~ kawaii ~
2012-10-18T17:23:32 < Laurenceb> weeaboo
2012-10-18T17:24:27 < Laurenceb> http://en.wikipedia.org/wiki/File:I-to-v_building_1000.jpg
2012-10-18T17:24:29 < Laurenceb> pmsl
2012-10-18T17:24:44 < dongs> Ya
2012-10-18T17:24:48 < dongs> that guy draw a shitload of schematics
2012-10-18T17:24:49 < dongs> like this
2012-10-18T17:25:16 < emeb_mac> so cute
2012-10-18T17:25:44 < dongs> luckily i have no fucking idea what that does
2012-10-18T17:25:50 < dongs> beyond having a resistor i it
2012-10-18T17:26:11  * emeb_mac weeps at the state of electrical engineering today.
2012-10-18T17:26:35 < dongs> i went to school for >trolling< bro.
2012-10-18T17:26:38 < dongs> i didnt touch anything EE.
2012-10-18T17:26:49 < emeb_mac> sounds lucrative
2012-10-18T17:26:50 < dongs> if I did, I'd probably know even less.
2012-10-18T17:26:58 <+Steffanx> lol
2012-10-18T17:27:25 < Laurenceb> born2troll
2012-10-18T17:27:35 <+Steffanx> Back in the 80ties they didn't know what a "troll" was dongs
2012-10-18T17:27:46 <+Steffanx> So how could you go to school for "trolling" ?!
2012-10-18T17:27:54 < emeb_mac> he had a dream
2012-10-18T17:27:59 < emeb_mac> an awesome dream
2012-10-18T17:29:02 < emeb_mac> saw a much-needed void before it even existed
2012-10-18T17:29:15 < dongs> thats lunix.
2012-10-18T17:29:21 < dongs> creating solutions for non-problems.
2012-10-18T17:29:42 < dongs> (and failing to fix giant holes/problems plaguing it today)
2012-10-18T17:29:50 < dongs> but HEY you can choose from 200 window managers.
2012-10-18T17:29:54 < dongs> and like 10 office suites
2012-10-18T17:29:56 < emeb_mac> fixing problems is haaarrd.
2012-10-18T17:30:08 < dongs> maybe one fucking day these lazy fucks will have a browser that isnt total shit
2012-10-18T17:30:14 < dongs> or maybe some day their fonts will render properly
2012-10-18T17:30:19 < dongs> but, that day is not today
2012-10-18T17:31:48 <+Steffanx> You ever get tired of yourself dongs?
2012-10-18T17:32:03 < dongs> no, but im going to sleep now
2012-10-18T17:32:09 < dongs> cuz im tired of YO U
2012-10-18T17:32:19 <+Steffanx> YAY, finally
2012-10-18T17:32:41 < Posterdati> zyp: which are the correct RCC_Clocks for stm32f107 @ 72MHz ?
2012-10-18T17:32:45 <+Steffanx> My life is complete
2012-10-18T17:33:15 < Posterdati> zyp: I have: SYS_CLK_Frequency = 225000000
2012-10-18T17:34:19 < Posterdati> HCLK_Frequency = 225000000, PCLK1_Frequency = 112500000, PCLK2_Frequency = 225000000, ADCCLK_Frequency = 112500000
2012-10-18T17:35:30 < dongs> attn Laurenceb https://www.youtube.com/watch?v=k8yKTuvRmPE
2012-10-18T17:40:29 < emeb_mac> even that can't get rid of my current earworm
2012-10-18T17:40:55 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Remote host closed the connection]
2012-10-18T17:43:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-18T17:43:44 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-18T17:49:19 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has joined ##stm32
2012-10-18T17:50:23 < upgrdman_> if there's only one slave on an SPI bus, you can just tie its CS line low (if active low) to keep it always selected, right? (you don't have to toggle it after each byte?)
2012-10-18T17:53:11 < dongs> you could, some actually need the cs going low at start, depends on what youre using
2012-10-18T17:53:20 < dongs> why the hell am i still awake
2012-10-18T17:53:23 <+Steffanx> YEAH
2012-10-18T17:53:50 < dongs> https://www.youtube.com/watch?v=avP5d16wEp0
2012-10-18T17:54:07 < upgrdman_> going low as in must by high for some period before going low?
2012-10-18T17:54:14 < karlp> upgrdman_: depends on the slave device, like he said.
2012-10-18T17:54:16 < dongs> yeah
2012-10-18T17:54:19 < upgrdman_> ok
2012-10-18T17:54:38 < upgrdman_> i'll have to read the datasheet on that 3axis accelerometer that's on my f4discov board
2012-10-18T17:55:30 < dongs> pretty sure its accessible either spi or i2c
2012-10-18T17:55:50 < dongs> and if its on f4disco its cs is already connected so just add the 2 extra lines of code to toggle it when addressing it
2012-10-18T17:55:53 < dongs> and declare victory
2012-10-18T17:56:23 < upgrdman_> i hope thats my problem. my SCK line works as expected, with periodic bursts. the MISO line is always low, and the MOSI line is always high... i have both pins configured as AF, open-drain with pull-ups. :(
2012-10-18T17:56:42 < dongs> pretty sure thats not how you configure SPI
2012-10-18T17:56:47 < dongs> have you tried looking at f4discovery sample code
2012-10-18T17:56:50 < qyx_> spi is not open drain
2012-10-18T17:56:54 < upgrdman_> the cs line is connected to e2 or e3... a regular gpio
2012-10-18T17:56:54 < dongs> specifically the mousesample app
2012-10-18T17:57:01 < dongs> which uses hte accel
2012-10-18T17:57:19 < upgrdman_> shit :) ... is it i2c that uses open-drain?
2012-10-18T17:57:23 < qyx_> yes
2012-10-18T17:57:54 < upgrdman_> spi is push-pull?
2012-10-18T17:57:58 < dongs> yeah.
2012-10-18T17:58:02 < dongs> spi is just.. normal.
2012-10-18T17:58:09 < upgrdman_> oh great. maybe it will be a trivial fix.
2012-10-18T17:58:46 < upgrdman_> so does that mean spi does not handle bus contention well?
2012-10-18T17:58:51 < dongs> ?!
2012-10-18T17:59:05 < upgrdman_> two spi slaves trying to talk at once
2012-10-18T17:59:09 < dongs> they wont
2012-10-18T17:59:09 < upgrdman_> on the same lines
2012-10-18T17:59:21 < dongs> slave doesnt talk unless talked to
2012-10-18T17:59:21 < qyx_> no, thats the purpose of cs line
2012-10-18T17:59:26 < dongs> and wont respond unless you pull its cs low
2012-10-18T17:59:28 < qyx_> master selects only one slave to talk to
2012-10-18T17:59:36 < upgrdman_> k
2012-10-18T17:59:55 < dongs> you can select both, but results will likely be very lulzy
2012-10-18T18:00:18 < dongs> o shit midnight
2012-10-18T18:00:20 < dongs> gone!
2012-10-18T18:00:25 < upgrdman_> so on the mcu, setup as an spi master, the MOSI needs to be push-pull, and the MISO needs to be a floating input?
2012-10-18T18:01:19 < qyx_> yes
2012-10-18T18:01:26 < upgrdman_> ok. thanks!
2012-10-18T18:01:27 < qyx_> and cs is push-pull low
2012-10-18T18:01:30 < upgrdman_> sure
2012-10-18T18:03:41 < upgrdman_> hum. GPIO->MODER would need to be set to AF obviously, but then the only options are push-pull and open-drain. does that mean the spi peripheral would ignore the contents of GPIO->OTYPER?
2012-10-18T18:04:26 < qyx_> on F4?
2012-10-18T18:04:54 < upgrdman_> yes
2012-10-18T18:04:55 < qyx_> look up some example code for spi
2012-10-18T18:05:27 -!- claude_work is now known as claude
2012-10-18T18:05:33 < upgrdman_> all the examples i see use the stdperiphlib, which i don't like. but ok, i'll try one of those and poke the registers in GDB
2012-10-18T18:07:44 < qyx_> not sure now, but i think all have to be set to push-pull and the peripheral takes care of inputs also
2012-10-18T18:11:49 < qyx_> in examples they set all spi pins to AF mode and PP
2012-10-18T18:18:32 < zyp> yes, you need to set all to AF
2012-10-18T18:24:49 < Posterdati> zyp: I found the stm32f has clock misconfiguration
2012-10-18T18:25:12 < Posterdati> zyp: is it possible to achieve 72 MHz with HSE clock @ 25 MHz?
2012-10-18T18:26:33 < zyp> you have f107?
2012-10-18T18:26:36 < upgrdman_> 75mhz maybe
2012-10-18T18:27:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-18T18:30:12 < Posterdati> zyp: yes a connectivity line device
2012-10-18T18:31:16 < Posterdati> zyp: I found that clock init sets 72 MHz, but consider input 8 Mhz HSE, on board there's a 25 MHz crystal, so I achieve 225 MHz :)
2012-10-18T18:31:25 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-18T18:34:52 < zyp> hmm, apparently there are two plls with a divisor each that can be chained
2012-10-18T18:35:22 < Posterdati> zyp: see page 107 of RM0008
2012-10-18T18:36:03 < zyp> the clock tree figure?
2012-10-18T18:36:12 < Posterdati> yes
2012-10-18T18:36:18 < Posterdati> I assume a 25 MHz HSE
2012-10-18T18:36:22 < zyp> it's on page 115 in my revision
2012-10-18T18:36:37 < Posterdati> I've got Rev 9
2012-10-18T18:37:15 < qyx_> see chibios clock init, i have 72MHz on 25MHz xtal
2012-10-18T18:37:21 < zyp> anyway, you can probably run 25 MHz HSE, configure PREDIV2 = /5, PLL2MUL = *8, PREDIV1SCR = PLL2, PREDIV1 = /5, PLLMUL = *9
2012-10-18T18:37:33 < qyx_> i think two plls are chained as zyp said
2012-10-18T18:37:35 < zyp> then you get 25 / 5 * 8 / 5 * 9 = 72
2012-10-18T18:37:52 < qyx_> yep
2012-10-18T18:40:03 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has quit [Ping timeout: 245 seconds]
2012-10-18T18:43:58 < Posterdati> sorry but sysclk is routed from prediv1 and pllmul only
2012-10-18T18:44:34 < zyp> of course, but prediv1 can run from pll2 output
2012-10-18T18:44:58 < zyp> also, I have rev 11, so you might want to get a newer revision
2012-10-18T18:45:06 < Posterdati> ah ok
2012-10-18T18:45:14 < Posterdati> where did you get the rev 11?
2012-10-18T18:45:25 < zyp> rev 11 is probably pretty old too, I downloaded it last year
2012-10-18T18:45:26 < Posterdati> I downloaded the Rev 9 from st site
2012-10-18T18:45:42 < zyp> huh?
2012-10-18T18:45:54 < zyp> rev 14 is current
2012-10-18T18:45:56 < zyp> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/REFERENCE_MANUAL/CD00171190.pdf
2012-10-18T18:48:46 < jpa-> dongs: why didn't you just use http://blog.makezine.com/2012/10/18/rgb-matrix-backpack/  :)
2012-10-18T18:49:54 < Posterdati> zyp: I'm downloading version 14.0
2012-10-18T18:50:18 < zyp> it's moved to page 123 in rev 14
2012-10-18T18:53:25 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-18T18:55:12 < Posterdati> zyp: yes, I'm on it
2012-10-18T18:55:43 < Posterdati> zyp: I have to manipulate the RCC registers there aren't st library functions for that
2012-10-18T19:00:21 -!- barthess [~barthess@5.100.192.144] has joined ##stm32
2012-10-18T19:01:03 < jpa-> who needs such functions?
2012-10-18T19:01:07 < jpa-> just write to the register
2012-10-18T19:01:53 -!- barthess [~barthess@5.100.192.144] has quit [Client Quit]
2012-10-18T19:13:48 < Laurenceb> you can have 25mhz hse
2012-10-18T19:13:55 < Laurenceb> its max 18 or something
2012-10-18T19:14:31 < Posterdati> zyp: done! Let's see if it works!
2012-10-18T19:15:12 < zyp> Laurenceb, f107 supports up to 25 MHz
2012-10-18T19:15:19 < Laurenceb> oh
2012-10-18T19:15:32 < zyp> also f105
2012-10-18T19:15:39 < zyp> the rest supports up to 16 MHz
2012-10-18T19:16:14 < Posterdati> zyp: working!
2012-10-18T19:18:13 < karlp> so this ethernet thing was all because the clock on your board didn't match what your bootups cripts thought they were running at?
2012-10-18T19:18:55 < Posterdati> zyp: now clocks are correct, I've got 50 MHz on ETH_RMII_REF_CLK, but no signals on MDIO and MDC
2012-10-18T19:19:12 < Posterdati> karlp: ETH still not working
2012-10-18T19:19:26 < Laurenceb> http://www.dumpaday.com/wp-content/uploads/2010/09/wrong-2.jpg
2012-10-18T19:19:37 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-18T19:19:53 < karlp> well, you didn't have a clock before, so yo're getting somewhere at least
2012-10-18T19:20:47 < zyp> Posterdati, which pins are MDIO and MDC connected to?
2012-10-18T19:21:34 < Posterdati> karlp: no the 50 MHz signal was wrong, then I changed it because it belong to PLL3 that is not involved in the CPU/peripherals functions
2012-10-18T19:22:04 < Laurenceb> hmm this was posted on chibios forumz
2012-10-18T19:22:14 < zyp> mm
2012-10-18T19:22:18 < Laurenceb> supposedly mco wont drive ethernet phy
2012-10-18T19:22:21 < zyp> PA2 and PC1
2012-10-18T19:22:25 < Laurenceb> i suspected n00bs at the time
2012-10-18T19:22:36 < Laurenceb> but someone said there was too much jitter
2012-10-18T19:23:09 < emeb> baroo?
2012-10-18T19:23:22 < zyp> Posterdati, verify that clocks are enabled for both AFIO, IOPA and IOPC
2012-10-18T19:23:29 < emeb> wouldn't that depend on the source, or do the switching threshold vary?
2012-10-18T19:25:05 < Posterdati> zyp IOPA? Do you mean GPIOA and GPIOC?
2012-10-18T19:25:30 < emeb> oh, who knows. They changed the nomenclature.
2012-10-18T19:25:48 < zyp> Posterdati, yes, they are named IOPA and IOPC in the datasheet
2012-10-18T19:26:01 < Posterdati> yes they are
2012-10-18T19:28:29 < Posterdati> they aren't even remapped
2012-10-18T19:30:03 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has joined ##stm32
2012-10-18T19:37:37 < Posterdati> :(
2012-10-18T19:41:47 < Posterdati> port A seems dead
2012-10-18T19:42:31 < zyp> even in gpio mode?
2012-10-18T19:42:31 < Posterdati> I configured as GPIO_Mode_Out_PP
2012-10-18T19:42:36 < Posterdati> zyp: yes
2012-10-18T19:42:44 < Posterdati> I did a WriteBit
2012-10-18T19:42:51 < zyp> check the rcc bits again, check that the enable bit is set and that the reset bit is not set
2012-10-18T19:42:58 < Posterdati> to Bit_SET but no signal on it
2012-10-18T19:43:55 < zyp> (I mean reset and enable bits for IOPA)
2012-10-18T19:44:19 < Posterdati> zyp: in the project there aren't RCC Reset commands
2012-10-18T19:44:47 < zyp> just read the register values and check what they are actually configured to
2012-10-18T19:45:04 < zyp> if everything is ok, then it's ok, if not then you start looking for what caused it.
2012-10-18T19:46:50 < Posterdati> peripheral is not reset
2012-10-18T19:47:15 < zyp> good, is it enabled?
2012-10-18T19:47:17 -!- dekar [~dekar@212.255.38.104] has joined ##stm32
2012-10-18T19:47:20 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-18T19:47:23 < Posterdati> clock is provided
2012-10-18T19:47:41 < Posterdati> RCC_APB2PeriphClockCmd(RCC_APB2Periph_GPIOA | RCC_APB2Periph_GPIOB | RCC_APB2Periph_GPIOC | RCC_APB2Periph_GPIOD | RCC_APB2Periph_GPIOE | RCC_APB2Periph_AFIO | RCC_APB2Periph_ADC1, ENABLE);
2012-10-18T19:50:57 < Posterdati> :(
2012-10-18T20:05:47 -!- daku is now known as DaKu
2012-10-18T20:13:25 < Posterdati> zyp: very strange
2012-10-18T20:14:16 < Laurenceb> try same code on another port?
2012-10-18T20:14:27 < Posterdati> ?
2012-10-18T20:17:03 < Posterdati> ...
2012-10-18T20:19:32 < Laurenceb> nvm
2012-10-18T20:19:39 < Laurenceb> throught porta was dead
2012-10-18T20:19:46 < Laurenceb> so you could try same code on portb
2012-10-18T20:20:23 < Posterdati> this board is pretty new, I never used it
2012-10-18T20:20:34 < Posterdati> I don't think it's broken
2012-10-18T20:27:16 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-18T20:27:26 < TitanMKD> hello
2012-10-18T20:28:20 < emeb> greetings
2012-10-18T20:43:06 <+Steffanx> Salute
2012-10-18T20:43:26 < zyp> god dag
2012-10-18T20:46:18 <+Steffanx> That's not in TitanMKD's language
2012-10-18T20:47:14 < Laurenceb> fo sho
2012-10-18T20:47:20 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-18T20:47:53 < emeb> it's the language of lutefisk and lefse.
2012-10-18T20:47:59 < zyp> :D
2012-10-18T20:48:28 < emeb> lotta Norwegian descendants in the place I came from.
2012-10-18T21:08:44 -!- Tectu is now known as tectu
2012-10-18T21:08:55 -!- tectu is now known as Tectu
2012-10-18T21:25:05 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 255 seconds]
2012-10-18T21:29:21 < Posterdati> zyp: I see a signal on MDIO (PA2), but MDC (PC1) is dead
2012-10-18T21:30:21 < zyp> make it be non-dead then.
2012-10-18T21:52:05 -!- barthess [~barthess@178.154.41.122] has joined ##stm32
2012-10-18T21:54:50 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-18T21:56:23 < Posterdati> zyp: undead a pin
2012-10-18T21:58:07 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds]
2012-10-18T21:59:24 < Posterdati> working!
2012-10-18T21:59:49 < Posterdati> GPIO_Init(GPIOE, &GPIO_InitStructure);
2012-10-18T21:59:57 <+Steffanx> :D
2012-10-18T22:00:01 < Posterdati> but the port to set was the GPIOC
2012-10-18T22:00:03 < Posterdati> lol
2012-10-18T22:00:54 < Posterdati> ping isn't working, but now connection is 100 MB/s on the switch
2012-10-18T22:01:21 < Posterdati> I see the clock now
2012-10-18T22:01:57 < Posterdati> now how can I determine the phy address?
2012-10-18T22:07:23 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 255 seconds]
2012-10-18T22:10:41 < Posterdati> zyp: every address I read from the phy is filled with 0xffff
2012-10-18T22:12:46 < qyx_> so something is not working!
2012-10-18T22:12:54 < qyx_> check the mdc/mdio lines
2012-10-18T22:13:03 < qyx_> also i think mdio needs strong pullup
2012-10-18T22:13:12 < Posterdati> AF_PP
2012-10-18T22:13:51 < qyx_> i mean check with oscope or logic analyzer
2012-10-18T22:14:06 < qyx_> but i assume the hw is ok
2012-10-18T22:14:26 < Posterdati> yes, the board is pretty new
2012-10-18T22:14:33 < qyx_> and olimex provides some examples with sources
2012-10-18T22:15:00 < Posterdati> qyx_: not for rev. a
2012-10-18T22:15:21 < qyx_> port settings will be the same
2012-10-18T22:15:34 < qyx_> only phy manufacturer/model changed
2012-10-18T22:15:44 < qyx_> and thats only means changed id
2012-10-18T22:15:45 < Posterdati> port is ok now
2012-10-18T22:15:51 < Posterdati> maybe is the phy config
2012-10-18T22:16:03 < Posterdati> difference in SR and CR register
2012-10-18T22:16:50 < qyx_> i used code for that ste101p also for ksz8051mll and it worked on first try
2012-10-18T22:17:04 < qyx_> so you can use any code you find as a reference
2012-10-18T22:24:51 < emeb> lol australia: http://arbroath.blogspot.co.uk/2012/10/police-arrest-man-armed-with-sawfish.html
2012-10-18T22:26:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-18T22:29:50 <+Steffanx> Australia: Almost as paranoid as the US
2012-10-18T22:32:39 < BrainDamage> https://encrypted.google.com/search?q=do+a+barrel+roll
2012-10-18T22:33:38 <+Steffanx> BrainDamage is also paranoid .. encrypted.google.com :D
2012-10-18T22:34:19 < BrainDamage> I use https everywhere plugin
2012-10-18T22:34:22 < emeb> Steffanx: Almost - slightly different areas of concern though :)
2012-10-18T22:34:33 < BrainDamage> it automatically switches to https version if available
2012-10-18T22:34:45 < BrainDamage> because sometimes I forget to use vpn when I roam wifi
2012-10-18T22:34:53  * Steffanx downloads that login
2012-10-18T22:36:09 <+Steffanx> Hmm, it seems to break some websites i visit :D
2012-10-18T22:36:24 <+Steffanx> nervermind
2012-10-18T22:43:44 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-18T22:46:29 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-18T22:46:30 < Thorn> rmii requires 50MHz with 50ppm accuracy, microcontroller PLLs can't do that (reliably anyway)
2012-10-18T22:47:10 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-18T22:53:41 < Posterdati> zyp: now connecting the cable hard fault the board!
2012-10-18T22:54:22 <+Steffanx> heh
2012-10-18T22:54:44 < Posterdati> zyp: well the ping actually does it
2012-10-18T22:56:02 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-18T22:57:54 < zyp> hmm, I did some length optimization of the longest tracks to the sdram, now the difference is <24mm
2012-10-18T22:58:00 < zyp> i.e. less than an inch
2012-10-18T22:58:33 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-18T22:58:36 < Laurenceb_> hi
2012-10-18T22:58:40 < zyp> hi
2012-10-18T22:58:50 < Laurenceb_> so who is going to add F3 support to texane?
2012-10-18T22:58:52 < Laurenceb_>  :P
2012-10-18T22:59:04 < Posterdati> zyp: a quantum leap forward
2012-10-18T22:59:35 < karlp> if you read the docs for me and work out whether it's the same as f1, f4 or l1, or somethign else, I can maybe build you a test build
2012-10-18T22:59:52 < karlp> but if it's a different flash load routine you'r eon your own :)
2012-10-18T23:00:10 < Laurenceb_> i will do later
2012-10-18T23:00:21 < Laurenceb_> going to pick mine up tomorrow morning
2012-10-18T23:01:06 < zyp> I put my money on flash interface being the same as in F0/F1/L1
2012-10-18T23:01:53 < jpa-> i bet the same as F4
2012-10-18T23:02:08 < karlp> I'm with jpa :)
2012-10-18T23:02:14 < karlp> L1 is not the same as f1 or f0!
2012-10-18T23:02:20 < karlp> nor is f1 like f0!
2012-10-18T23:02:20 < zyp> isn't it?
2012-10-18T23:02:26 < karlp> no, not at all
2012-10-18T23:02:34 < jpa-> f1 is "special"
2012-10-18T23:02:38 < zyp> f0=l1 then?
2012-10-18T23:02:40 < jpa-> like Laurenceb
2012-10-18T23:02:42 < karlp> nope :)
2012-10-18T23:02:49 < karlp> all different except f2 and f4 so far
2012-10-18T23:03:10 < zyp> you say different, but different how?
2012-10-18T23:03:48 < Laurenceb_> i see what you did there
2012-10-18T23:08:00 < Laurenceb_> its not the same as F4
2012-10-18T23:08:02 < zyp> first of all, it's not equal to F4
2012-10-18T23:08:05 < Laurenceb_> yes
2012-10-18T23:08:06 < Laurenceb_> heh
2012-10-18T23:09:09 < zyp> not F0 either
2012-10-18T23:09:36 < Laurenceb_> looks like F1 XL density and CL
2012-10-18T23:09:40 < Laurenceb_> to me
2012-10-18T23:09:55 < zyp> nor L1
2012-10-18T23:10:11 < Laurenceb_> a load of 2KB pages?
2012-10-18T23:11:05 < zyp> looks very similar to F1, yeah
2012-10-18T23:11:30 < Laurenceb_> looks like a copy of F1  CL to me
2012-10-18T23:12:43 < zyp> main difference is OBR
2012-10-18T23:12:54 < zyp> but that should be irrelevant to flashing
2012-10-18T23:13:07 < zyp> so it will likely work fine with F1 flashloader
2012-10-18T23:14:14 < zyp> registers are at exact same addresses as F1 also
2012-10-18T23:14:45 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-18T23:15:18 < zyp> and keys are the same as F1 too
2012-10-18T23:15:28 < Laurenceb_> its 2K pages tho
2012-10-18T23:15:37 < Laurenceb_> so needs to be CL/XL-Density
2012-10-18T23:16:10 < Laurenceb_> i cant find OBR description on F1 ref manual
2012-10-18T23:16:36 < Laurenceb_> how does the bootloader get its data over SWD?
2012-10-18T23:16:47 < zyp> huh?
2012-10-18T23:17:16 < zyp> the flash register block is in the flash programming manual for F1
2012-10-18T23:17:18 < zyp> seperate pdf
2012-10-18T23:17:24 < Laurenceb_> i see
2012-10-18T23:17:36 < zyp> PM0068/PM0075
2012-10-18T23:17:47 < Laurenceb_> so the texane stuff loads the flashloader
2012-10-18T23:17:52 < Laurenceb_> then what?
2012-10-18T23:18:04 < Laurenceb_> its actually sending data ove rSWD?
2012-10-18T23:18:50 < zyp> flashloader is just a simple routine that copies a word from ram to flash and wait for a flag before it copies the next word
2012-10-18T23:19:03 < zyp> think like dma
2012-10-18T23:19:15 -!- mrcan__ [~mrcan@78.173.40.143] has joined ##stm32
2012-10-18T23:19:15 -!- mrcan__ [~mrcan@78.173.40.143] has quit [Changing host]
2012-10-18T23:19:15 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-18T23:19:38 < zyp> so stlink writes a block of data into ram, then executes flashloader to write it to flash
2012-10-18T23:19:45 < Laurenceb_> oh
2012-10-18T23:19:48 < Laurenceb_> i see
2012-10-18T23:19:59 < Laurenceb_> ok thats kind of obvious
2012-10-18T23:20:36 < Laurenceb_> F1 flashing is slow :(
2012-10-18T23:20:40 < Laurenceb_> 7KB/s
2012-10-18T23:22:44 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-18T23:23:09 < zyp> https://github.com/texane/stlink/blob/master/flashloaders/stm32f4.s <- you can see the source for the f4 flashloader I wrote here, it's not very complex
2012-10-18T23:24:27 < zyp> lines 17-20 is pretty much while(FLASH_SR & BSY);
2012-10-18T23:24:28 < Laurenceb_> so why is F1 so slow?
2012-10-18T23:24:42 < zyp> no idea, I haven't touched anything regarding that :p
2012-10-18T23:24:48 < Laurenceb_> heh
2012-10-18T23:25:02 < zyp> maybe the flash just is slower? :p
2012-10-18T23:25:23 < Laurenceb_> dongs claims to get several times the speed
2012-10-18T23:26:24 < zyp> a more efficient approach would be to double buffer in ram and transfer one block via swd while the flashloader is busy writing the other block to flash
2012-10-18T23:26:53 < zyp> neither texane/stlink nor BMP does that
2012-10-18T23:27:03 < Laurenceb_> fail
2012-10-18T23:27:35 < Posterdati> zyp: now I've to debug more...
2012-10-18T23:27:48 < zyp> I've considered trying to implement it to see how much speed increase I can get, but I havent had time to yet
2012-10-18T23:28:11 -!- barthess [~barthess@178.154.41.122] has quit [Quit: Leaving.]
2012-10-18T23:30:06 < Laurenceb_> farnell is so annoying
2012-10-18T23:30:13 < Laurenceb_> i dont want to buy a rpi
2012-10-18T23:32:15 < Laurenceb_> nice
2012-10-18T23:32:28 < Laurenceb_> my F3discoveries are ready for collection
2012-10-18T23:32:34 < Laurenceb_> just got the email
2012-10-18T23:34:04 < ds2> you must be in europe?
2012-10-18T23:34:12 < Laurenceb_> for some reason the RS warehouse staff are nocturnal
2012-10-18T23:34:18 < Laurenceb_> in in england
2012-10-18T23:34:26 < Laurenceb_> about 1 mile from the RS depot :P
2012-10-18T23:34:38 < karlp> zyp: the oocd flash loader for f1 does that double buffering thing.
2012-10-18T23:34:54 < Laurenceb_> annoyingly its only got about 2 staff during the day
2012-10-18T23:35:03 < Laurenceb_> so you can only pick up pre ordered stuff
2012-10-18T23:35:12 < karlp> Laurenceb_: you're lycky to get 7kb on f1 wtih texane :) I get 7kb at work (presumably newer usb drivers?) but only about 3k at home.
2012-10-18T23:35:52 < Laurenceb_> ubuntu?
2012-10-18T23:36:02 < emeb> better ac power at work. :P
2012-10-18T23:36:04 < karlp> fed17 at work, old ubu10.04 at home.
2012-10-18T23:36:11 < karlp> emeb: hehe
2012-10-18T23:36:24 < Laurenceb_> interesting
2012-10-18T23:36:30 < Laurenceb_> i use 10.04
2012-10-18T23:36:37 < Laurenceb_> no unity crap
2012-10-18T23:36:50 < zyp> I've never really had to care much about flashing speed, my projects are so small :p
2012-10-18T23:37:27 < Laurenceb_> for some reason my logger project is 119K
2012-10-18T23:38:19 < Laurenceb_> but i compile with -O0 so debugging is nice
2012-10-18T23:38:21 < zyp> that's because you copy paste all kinds of shit
2012-10-18T23:38:30 < zyp> :p
2012-10-18T23:39:06 < Laurenceb_> with -Os and link time optimisation its closer to 40
2012-10-18T23:41:31 < karlp> libopencm3 needs a good dose of LTO I reckon. so it can make better decisions about inlining all the tiny syntactical sugar methods.+
2012-10-18T23:42:22 < scrts> anyone ever tried to attach wifi module to stm32?
2012-10-18T23:42:44 < karlp> are you the guy that was talking about running uclinux on stm32?
2012-10-18T23:43:12 < emeb> no - that's ds2
2012-10-18T23:43:27 < zyp> better change to lpc4300, I hear it supports up to 896MB of SDRAM
2012-10-18T23:43:28 < ds2> that'd be me
2012-10-18T23:43:32 < zyp> :p
2012-10-18T23:43:58 < ds2> no room for that big of a battery!
2012-10-18T23:44:40 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-18T23:45:02 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-18T23:45:25 < emeb> get that man a wallwart - stat!
2012-10-18T23:46:11 < ds2> i go mobile
2012-10-18T23:46:31 < ds2> defeats having epaper displays
2012-10-18T23:47:23 < emeb> what sort of RAM footprint does uclinux have?
2012-10-18T23:48:16 < qyx_> i think 2-4M is minimum
2012-10-18T23:48:36 < qyx_> but there is also independent cm3 port of linux kernel
2012-10-18T23:48:55 < qyx_> http://www.linux-arm.org/LinuxKernel/LinuxM3
2012-10-18T23:49:24 < ds2> 2M RAM is what some folks claim
2012-10-18T23:49:42 < ds2> but I am doing 8M just in case
2012-10-18T23:49:51 < karlp> I wouldn't want to try 2M,
2012-10-18T23:49:56 < ds2> and 32G better be sufficient ;)
2012-10-18T23:50:00 < qyx_> but there is no reason in running just the kernel without something useful
2012-10-18T23:50:01 < karlp> 2M is for, "look I booted uclinux"
2012-10-18T23:50:11 < karlp> yeah, exactly.
2012-10-18T23:50:13 < ds2> there is
2012-10-18T23:50:21 < ds2> the kernel can route w/just kernel
2012-10-18T23:50:23  * emeb has an old MIPS-based Agenda VR3 PDA that runs full linux.
2012-10-18T23:50:29 < zyp> what's the appeal of using linux on cortex-m?
2012-10-18T23:50:36 < emeb> has 8MB sdram
2012-10-18T23:50:38 < ds2> userland is less needed then you'd think
2012-10-18T23:50:38 < karlp> what's the idea of using a cm3 for a router?
2012-10-18T23:50:45 < karlp> depends what you want to do.
2012-10-18T23:50:50 < ds2> easier build
2012-10-18T23:50:59 < karlp> easier build?
2012-10-18T23:51:07 < zyp> more like bloated
2012-10-18T23:51:14 < ds2> there aren't many small MMU enabled SoC around
2012-10-18T23:51:30 < ds2> for same foot print, you're looking at a POP setup requiring microvias
2012-10-18T23:51:43 < karlp> what?
2012-10-18T23:51:51 < ds2> not sw build, hw build
2012-10-18T23:51:56 < karlp> ralink?
2012-10-18T23:52:04 < karlp> look at all the tiny openwrt boards
2012-10-18T23:52:16 < karlp> they run full linux, so it actually does count as easier build.
2012-10-18T23:52:17 < ds2> the TI CortexA parts
2012-10-18T23:52:23 < ds2> I am going smaller then that
2012-10-18T23:52:40 < ds2> what's the pitch of the ralink parts?
2012-10-18T23:52:51 < karlp> I don't know, I don't bother building my own for that sort of thing.
2012-10-18T23:53:09 < zyp> ah, I have some stuff based on that
2012-10-18T23:53:15 < qyx_> just see how many of mikrotik routers are built
2012-10-18T23:53:16 < scrts> it depends.. but where do You buy ralink chips?
2012-10-18T23:53:24 < karlp> I say, "give me a linux board with x meg ram and flash that fits into this case"
2012-10-18T23:53:25 < scrts> they sell minimum 100k/year...
2012-10-18T23:53:28 < qyx_> actually adm1520 or something like that was in qfp
2012-10-18T23:53:35 < ds2> the 0.8mm on the F103 is doable w/o toomuch pain
2012-10-18T23:53:39 < qyx_> and has mmu
2012-10-18T23:53:50 < qyx_> but most of the soc's in routers are mips-es
2012-10-18T23:53:51 < ds2> ewwww QFP. those are huge
2012-10-18T23:53:55 < scrts> maybe it's better to use cheap modules? ;)
2012-10-18T23:53:56 < zyp> I bought a couple of travel-routers when I was in japan this summer
2012-10-18T23:54:00 < scrts> like carambola?
2012-10-18T23:54:20 < ds2> tplink has some nice small routers but it is still huge
2012-10-18T23:54:20 < zyp> the size of a small matchbox, takes power from microusb, got two rj45 ports and wlan with 2x2 mimo
2012-10-18T23:54:32 < karlp> yeah, those are awesome
2012-10-18T23:54:44 < karlp> with the price of those, I don't see a lot of motivation for building my own uclinux board
2012-10-18T23:54:50 < scrts> http://www.8devices.com/product/3/carambola
2012-10-18T23:54:52 < zyp> agreed
2012-10-18T23:55:02 < zyp> I bought several of those
2012-10-18T23:55:09 < karlp> wifi onboard, all the board support already done, already runs full linux.
2012-10-18T23:55:17 < ds2> matchbox? you can't put 2 RJ45's in a match box
2012-10-18T23:55:30 < qyx_> why
2012-10-18T23:55:38 < qyx_> just one example
2012-10-18T23:55:47 < ds2> cuz it won't fit
2012-10-18T23:56:03 < karlp> the tiny routers he's talking about are the size of the short side of a creditcard, cquare
2012-10-18T23:56:07 < qyx_> hm, i need to look up the name :>
2012-10-18T23:56:13 < karlp> so, the long side of a matchbox, square
2012-10-18T23:56:22 < ds2> match boxes are like 1.25x2x0.5
2012-10-18T23:56:23 < karlp> plenty of room for two rj45s
2012-10-18T23:56:24 < qyx_> http://www.lantronix.com/device-networking/embedded-device-servers/xport.html
2012-10-18T23:56:25 < zyp> I measured one: 14x42x56mm
2012-10-18T23:56:27 < qyx_> that thing runs linux
2012-10-18T23:56:33 < qyx_> and has 8MB sdram i think
2012-10-18T23:56:46 < qyx_> hm, no
2012-10-18T23:56:48 < qyx_> not this one
2012-10-18T23:56:54 < ds2> that's about 2.25x1.8x0.75
2012-10-18T23:56:55 < qyx_> but similar one
2012-10-18T23:56:57 < ds2> big
2012-10-18T23:57:22 < ds2> the xports are uCLinux, IIRC
2012-10-18T23:57:27 < zyp> http://oclab.pl/wp-content/uploads/2012/05/logitec-lan-w300nr-2.jpg
2012-10-18T23:57:28 < karlp> creditcard short side square, 50-55 square.
2012-10-18T23:58:06 < zyp> I can fit two of these in the same area as a credit card
2012-10-18T23:58:37 < karlp> tplink 703n is 57x57, 4meg flash, 32meg ram, wireless, one eth, one wifi, one usb host port
2012-10-18T23:59:36 < karlp> that's a nice one zyp, I hadn't seen that one before
--- Day changed Fri Oct 19 2012
2012-10-19T00:02:34 < zyp> http://www.planex.co.jp/product/product_photo/300/mzk-rp150n.jpg <- other one is same size, different vendor and doesn't seem to have mimo
2012-10-19T00:06:58 < ds2> does that run off batteries?
2012-10-19T00:08:02 < Laurenceb_> btw, ST have stm32f1 based bluetooth modules
2012-10-19T00:08:15 < Laurenceb_> wonder if the firmware is avaliable
2012-10-19T00:08:18 < karlp> yeah, they look cool.
2012-10-19T00:08:21 < ds2> they do?
2012-10-19T00:08:22 < karlp> decent price too
2012-10-19T00:08:29 < karlp> was in the latest newsletter
2012-10-19T00:09:11 < ds2> oh
2012-10-19T00:09:39 < Laurenceb_> http://www.st.com/internet/imag_video/product/253471.jsp
2012-10-19T00:09:42 <+Steffanx> Laurenceb_ linky linky: https://www.youtube.com/watch?v=VARUtWxDg_g
2012-10-19T00:09:52 < zyp> ds2, they take power from microusb, so any of these battery packs with 5V output can power them
2012-10-19T00:10:03 < Laurenceb_> if there was firmware for it..
2012-10-19T00:10:13 < Laurenceb_> then it'd make a good f1 breakout too
2012-10-19T00:10:19 < zyp> or any other 5V source
2012-10-19T00:11:08 < zyp> I'd be surprised if it's not locked
2012-10-19T00:11:48 < ds2> zyp: 5V is too in efficient -> 3.7nominal boost to 5V then it takes down
2012-10-19T00:12:04 < karlp> ds2: some of those type of micro rotuers come with a built in battery,
2012-10-19T00:12:08 < karlp> tplink sells some like that.
2012-10-19T00:12:35 < ds2> karlp: really? know models off hand? I been looking for a mobile router like that and been coming up blank
2012-10-19T00:12:51 < zyp> ds2, what's the point of running the router off a battery anyway?
2012-10-19T00:13:04 < ds2> zyp: mobile use
2012-10-19T00:13:16 < karlp> one less thing to plug in...
2012-10-19T00:13:33 < zyp> you still have to have ethernet hooked up to something?
2012-10-19T00:13:43 < karlp> no, usb 3g
2012-10-19T00:13:47 < ds2> no, wifi/bluetooth
2012-10-19T00:13:56 < ds2> bluetooth tether to a phone. wifi out to other devices
2012-10-19T00:14:03 < karlp> well, the wifi for the lan side, usb/bt to phone or 3g
2012-10-19T00:14:21 < ds2> exactly.... mobile/field datacenter
2012-10-19T00:15:05 < zyp> there are a lot of wifi/cellular routers around if that's what you are talking about?
2012-10-19T00:15:22 < ds2> same idea but don't want the cellular radio on there
2012-10-19T00:15:26 < zyp> and all those are of course battery powered, since they are designed for carrying with you
2012-10-19T00:15:38 < ds2> no reason to be locked down to a carrier
2012-10-19T00:15:56 < zyp> don't they just take sim cards?
2012-10-19T00:16:18 < ds2> SIM is mainly for the GSM based carriers
2012-10-19T00:16:25 < karlp> ds2 looks like only in china so far: http://www.tplink.cn/pages/product-detail.asp?d=233
2012-10-19T00:16:32 < zyp> in my part of the world, we only have GSM based carriers
2012-10-19T00:16:50 < karlp> they have one that has an eth port too, but I can't find it right now.
2012-10-19T00:17:04 < zyp> anyway, what is bluetooth supposed to connect to? a phone?
2012-10-19T00:17:07 < ds2> ah... got too many different technology here
2012-10-19T00:17:16 < ds2> bt to phone.
2012-10-19T00:17:22 < zyp> why not just get a phone with wifi?
2012-10-19T00:17:37 < zyp> every modern phone can share their connectivity via wifi
2012-10-19T00:17:53 < ds2> not really
2012-10-19T00:18:00 < ds2> @#%@%$@@$@$#@$@ carriers like to lock out features like that
2012-10-19T00:18:13 < zyp> oh, right
2012-10-19T00:18:46 < zyp> in my part of the world, we don't have to buy phones from carriers themselves
2012-10-19T00:18:46 < ds2> they don't like people to be able to do useful things besides consume ads and write them checks
2012-10-19T00:19:02 < zyp> we just get the phone we want and put in a sim card from the carrier we want
2012-10-19T00:19:36 < ds2> in this part of the world, carriers lock sims to types of devices... so a sim will work in a phone or a tablet or a mifi but not any of them
2012-10-19T00:19:45 < karlp> here: http://www.tp-link.com/en/products/details/?model=TL-MR3040
2012-10-19T00:19:58 < karlp> usb, eth, wifi, and abattery
2012-10-19T00:20:10 < karlp> ds2: where's that?
2012-10-19T00:20:11 < zyp> ds2, I don't get why you put up with that shit
2012-10-19T00:20:21 < karlp> and how is that legally allowed?
2012-10-19T00:20:55 < ds2> karlp: US
2012-10-19T00:21:00 < karlp> oh, land of the free.
2012-10-19T00:21:02 < ds2> it is theor ToS
2012-10-19T00:21:07 < ds2> their
2012-10-19T00:21:07 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-19T00:21:46 < ds2> karlp: oooh nice...
2012-10-19T00:21:54 < ds2> I think that one is supported by openwrt too
2012-10-19T00:23:26 < karlp> it is.
2012-10-19T00:23:42 < karlp> all the tplinks are ar7xx, but some of the china release models are very tight on flash
2012-10-19T00:24:28 < karlp> this one too, http://wiki.openwrt.org/toh/tp-link/tl-mr11u
2012-10-19T00:24:37 < ds2> just need enough space to put in some version of bluez
2012-10-19T00:24:44 < karlp> shouldn't be a problem.
2012-10-19T00:25:03 < ds2> made the mistake of getting the asus one w/o the battery
2012-10-19T00:25:04 < karlp> you can drop it on the usb
2012-10-19T00:25:41 < ds2> *nod*
2012-10-19T00:25:59 < ds2> but these things don't replace the smaller size of something like the stm32 running uClinux
2012-10-19T00:26:06  * Laurenceb_ pmsl
2012-10-19T00:26:24 < karlp> except, you don't have to do all the hardwar development yourself,
2012-10-19T00:26:26 < Laurenceb_> my albanian friend has been told he cant live in the air conditioning at work
2012-10-19T00:26:44 < ds2> but these are still bigger and don't have what I want
2012-10-19T00:26:54 < ds2> the end goal is basically a watch
2012-10-19T00:27:06 < ds2> and all these are too big
2012-10-19T00:27:34 < karlp> you want a uclinux watch?
2012-10-19T00:27:50 < karlp> "doood my watch runs linux!" you're lucky dongs has gone to bed :)
2012-10-19T00:27:54 < ds2> either that or Linux
2012-10-19T00:29:18 < Laurenceb_> he had a sleeping bag in one of the ducts
2012-10-19T00:29:35 < ds2> hahah
2012-10-19T00:29:53 < ds2> body scent discharges must be interesting
2012-10-19T00:30:34 < Laurenceb_> apparently its a "breech of fire safety"
2012-10-19T00:35:02 < qyx_> yesterday i said that lpc43xx has 1mm tfbga100 package
2012-10-19T00:35:05 < qyx_> now i cannot find it
2012-10-19T00:36:54 < qyx_> but it seems logical that bga100 with rm=1mm wont fit in 9x9mm package
2012-10-19T00:36:59 < qyx_> :S
2012-10-19T00:42:02 < karlp> is there a "right" way of zeroing a volatile struct?
2012-10-19T00:42:31 < karlp> memset doesn't approve.
2012-10-19T00:49:17 < upgrdman_> what not manually after declaration?
2012-10-19T00:49:23 < upgrdman_> why not*
2012-10-19T00:49:51 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-19T00:51:26 < Laurenceb_>   {}
2012-10-19T00:51:39 < Laurenceb_> oh at runtime
2012-10-19T00:52:14 < Laurenceb_> flag at one end?
2012-10-19T00:52:40 < zyp> http://paste.jvnv.net/view/Grbuo
2012-10-19T00:52:47 < zyp> compiles for me
2012-10-19T00:52:54 < zyp> although assembly is a bit funny
2012-10-19T00:53:21 < zyp> http://paste.jvnv.net/view/NE0Ig
2012-10-19T00:53:22 < Laurenceb_> its going to loop tho?
2012-10-19T00:54:10 < zyp> it ends up allocating two words from stack, storing 0 in both, then reading them back and writing them to the volatile struct :p
2012-10-19T00:54:44 < zyp> I'm not sure why it doesn't optimize out the first step and just goes on to store 0 in the volatile directly
2012-10-19T00:55:04 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-19T01:02:36 < karlp> zyp that's not the same as "struct state_t volatile state;"
2012-10-19T01:03:04 < karlp> at least, if I understood what I was reading yesterday abotu volatile poitns to structs and pointers to volatile structs
2012-10-19T01:03:26 < karlp> I have multiple entries in the struct that I wanted to treat as volatile, and wanted to just make the whole thing volatile
2012-10-19T01:03:33 < zyp> this is not a pointer
2012-10-19T01:04:16 < zyp> neither is your line
2012-10-19T01:04:30 < karlp> no, but memset(&state, 0, sizeof (state)); is,
2012-10-19T01:04:43 < zyp> volatile struct: volatile struct foo_t foo;
2012-10-19T01:04:58 < zyp> also volatile struct: struct foo_t volatile foo;
2012-10-19T01:05:11 < karlp> you don't get "missing initializer [-Wmissing-field-initializers]" from your {} trick?
2012-10-19T01:05:17 < zyp> pointer to volatile struct: volatile struct foo_t* foo;
2012-10-19T01:05:31 < zyp> aslo pointer to volatile struct: struct foo_t volatile *foo;
2012-10-19T01:05:50 < zyp> volatile pointer to non-volatile struct: struct foo_t* volatile foo;
2012-10-19T01:06:02 < zyp> volatile pointer to volatile struct: volatile struct foo_t* volatile foo;
2012-10-19T01:06:24 < zyp> hmm
2012-10-19T01:06:28 < zyp> with -Wextra I do
2012-10-19T01:06:33 < zyp> it's not enabled by -Wall
2012-10-19T01:07:02 < zyp> that's easy to solve, just put {0}
2012-10-19T01:08:02 < zyp> ah
2012-10-19T01:08:11 < zyp> I found a much saner variant
2012-10-19T01:08:46 < zyp> http://paste.jvnv.net/view/2wy0T
2012-10-19T01:08:59 < zyp> having it global makes it optimize in a sane way
2012-10-19T01:09:00 < karlp> ooh tricky.
2012-10-19T01:09:22 < zyp> and global objects are always default-initialized to zero if nothing is specified
2012-10-19T01:10:07 < zyp> this is with -O2 of course
2012-10-19T01:10:22 < zyp> with -O0 it actually emits the zero-object and copies it as requested :p
2012-10-19T01:10:51 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-19T01:11:21 < upgrdman_> i need a cheap 2-port ethernet hub but can't find any locally. a y with male ends would be fine... do you think it would be acceptable to just cut up some cat6 and wire three ends in parallel? keeping the un-twisted length ot a minimum....
2012-10-19T01:11:28 -!- cjbaird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-19T01:11:30 < ds2> what would be the reason to ever have a zero object? isn't there a constant provided in the ARM instructions?
2012-10-19T01:11:40 < zyp> wait, I misread the -O2 variant
2012-10-19T01:11:45 < zyp> it also copies it
2012-10-19T01:11:45 < Posterdati> zyp: now the phy's got configured, but app hardfault when it receive a packet and still no IRQ is triggered for ETH_IRQn
2012-10-19T01:11:55 < zyp> but still saner than constructing it on stack first
2012-10-19T01:12:29 < zyp> Posterdati, sounds fun
2012-10-19T01:12:44 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-19T01:12:44 < zyp> seems like you're making progress, keep going :)
2012-10-19T01:14:03 < Posterdati> but I don't know how debug that
2012-10-19T01:15:09 < karlp> zyp: so really, if my state struct is static global, I shouldn't need to do any zeroing anyway then
2012-10-19T01:15:42 < zyp> karlp, not unless you need to zero it after initialization
2012-10-19T01:16:17 < zyp> if you instance it without any initializer it would be put in .bss which is zeroed as part of the system startup
2012-10-19T01:16:50 < zyp> if you have a nonzero initializer it's put in .data which is initialized with a copy from flash during system startup
2012-10-19T01:16:56 < karlp> well, yay, I've finally got my RHT03 one wire thing reading. main.c:INFO: Temp: 213 *100 C, RH = 376 * 100 %
2012-10-19T01:17:03 < karlp> that only took far far too long.
2012-10-19T01:17:05 < zyp> and the flash copy will still contain 0 for all uninitialized fields
2012-10-19T01:17:18 < karlp> yay, let bss do it's job.
2012-10-19T01:22:42 -!- DaKu is now known as daku
2012-10-19T01:24:54 < karlp> ok, reviewed all  uses of memset, found none of them were actually needed.
2012-10-19T01:25:38 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-19T01:35:08 < zyp> looks like it's a bit more sane about the optimizations in C++ mode
2012-10-19T01:37:27 < zyp> except it doesn't want to assign a volatile struct at all
2012-10-19T01:38:27 < karlp> well, timer reset, and then unreset just before reinitalizing all the params seems to totally destroy my counters.
2012-10-19T01:38:53 < karlp> let's read the docs again to see what that's meant ot be used for
2012-10-19T01:41:41 < karlp> what are you meant to use things like "APB1 peripheral reset register
2012-10-19T01:41:54 < karlp> all I can see is "reset PERIPH interface"
2012-10-19T01:48:33 < dongs> haha
2012-10-19T01:48:37 < dongs>  < jpa-> dongs: why didn't you just use http://blog.makezine.com/2012/10/18/rgb-matrix-backpack/  :)
2012-10-19T01:48:51 < dongs> jpa-: that display is shit, you have to do software pwm etc
2012-10-19T01:48:56 < dongs> plus, tarduino?
2012-10-19T01:54:39 < cjbaird> That's the same DMD 512 display I've got, but with an Arduino thrown in. The original board came with a 'shield' in the first place..
2012-10-19T01:57:11 < cjbaird> ah, rgb, not plain.
2012-10-19T02:10:03 < emeb> love the power wires
2012-10-19T02:11:38 < qyx_> i was just about to ask why they are so thick
2012-10-19T02:16:48 < emeb> 16x32x3 = 1536 LEDs total. @ 20mA each that's like 30A if they were all on 100%.
2012-10-19T02:16:56 < emeb> It's muxed so they won't be of course.
2012-10-19T02:17:05 < emeb> but the current can still be pretty high.
2012-10-19T02:17:08 < qyx_> hm
2012-10-19T02:17:45 < dongs> i think its 1/8 or 1/16
2012-10-19T02:17:52 < emeb> yep
2012-10-19T02:18:14 < emeb> so 2 or 4 amps. -> thick wires.
2012-10-19T02:18:40 < emeb> although, the walwart they're powering it with is much thinner gauge wire.
2012-10-19T02:18:53 < emeb> so it's mostly for show.
2012-10-19T02:21:52 < ds2> nice 1536 LEDs
2012-10-19T02:22:24 < ds2> that can get toasty
2012-10-19T02:28:09 < dongs> so basically the only reason to own it is gone
2012-10-19T02:28:09 < Posterdati> zyp: I found the problem! I used an hd startup file instead of cl startup
2012-10-19T02:28:29 < dongs> so fucking what
2012-10-19T02:28:38 < dongs> lcd gone = whats the point of using it?
2012-10-19T02:29:05 < dongs> yea, good luck wiht that
2012-10-19T02:29:53 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Read error: Operation timed out]
2012-10-19T02:30:53 < Posterdati> zyp: but still get no ping answer
2012-10-19T02:31:14 < zyp> then you have more to fix, keep on going
2012-10-19T02:31:22 < Posterdati> lol
2012-10-19T02:32:19 < Posterdati> zyp: now the interrupt is correctly triggered
2012-10-19T02:33:15 < zyp> good
2012-10-19T02:33:21 < zyp> time to move up a layer then
2012-10-19T02:37:40 < emeb> flyback: kewl. many hours happy hacking.
2012-10-19T02:46:29 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-19T03:06:12 < upgrdman> what temp and duration?
2012-10-19T03:06:28 < upgrdman> o
2012-10-19T03:06:32 < upgrdman> :)
2012-10-19T03:06:58 < upgrdman> damn it. just burned another regulator
2012-10-19T03:07:16 < upgrdman> i wish the f4discov board came with a reg that could handle a few hundred mA
2012-10-19T03:08:24 < upgrdman> lol
2012-10-19T03:08:57 < upgrdman> sure, no problem with that
2012-10-19T03:09:48 < qyx_> flyback: lol
2012-10-19T03:09:52 < qyx_> the third picture
2012-10-19T04:15:42 < dongs> what the hell i think my ADC shit is all fucked
2012-10-19T04:24:25 < dongs> hm ok fixed that.. but now its not changing value ;d
2012-10-19T04:46:33 < upgrdman> my spi clk line works (i see bursts of square waves) but miso/mosi are always low. any common causes?
2012-10-19T05:06:20 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-19T05:13:28 < dongs> try sending something to spi.
2012-10-19T05:22:49 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-19T06:09:14 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-19T06:09:20 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-19T06:22:29 < emeb_mac> it's like a graveyard here.
2012-10-19T06:24:22  * ds2 turns the reflow oven til it incandences
2012-10-19T06:24:41 < emeb_mac> cooking burgers?
2012-10-19T06:25:06 < dongs> deader than lunix on desktop
2012-10-19T06:25:27 < emeb_mac> what's trollin' dongs?
2012-10-19T06:25:35 < ds2> even graveyards need light
2012-10-19T06:25:36 < dongs> not much. waiting for pcb pics
2012-10-19T06:25:41 < cjbaird> Brewing yogurt in a reflow oven should be attempted..
2012-10-19T06:26:11 < emeb_mac> more likely to succeed than making ice cream I suppose
2012-10-19T06:26:21 < dongs> my ledfaggot is failing wiht his overengineering F373 breakout board so no stencil for me this time, somtime next week
2012-10-19T06:26:24 < ds2> Mmmmm reflowed yogurt
2012-10-19T06:26:29 < cjbaird> Well, you just need a constant 42-43C
2012-10-19T06:26:47 < emeb_mac> hah - like an Arizona summer.
2012-10-19T06:27:04 < emeb_mac> at night
2012-10-19T06:28:13 < cjbaird> Unfortunately I didn't try the yogurt brewing thing on the bike trip down the middle of Australia-- some fresh culture, and a bottle of milk on the back..
2012-10-19T06:28:37 < ds2> how'd you keep the culture fresh?
2012-10-19T06:29:08 < cjbaird> Just buy 'real yogurt' from somewhere, and keep a bit from each brew for the next.
2012-10-19T06:30:17 < cjbaird> (only good for about 3-4 iterations though, by that time contaminating strains build up)
2012-10-19T06:30:23 < ds2> does real yogurt survive in the fridge?
2012-10-19T06:30:33 < emeb_mac> sure
2012-10-19T06:30:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-19T06:30:52 < emeb_mac> my dad used to do that.
2012-10-19T06:31:22 < emeb_mac> mixed up milk & starter at night, warmed it overnight - fresh yogurt in the morning. save a bit for next evening.
2012-10-19T06:31:35 < emeb_mac> keep it in the fridge in between.
2012-10-19T06:32:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-19T06:33:33 < ds2> an Ontopic question - anyone here do STM32 stuff w/batteries?
2012-10-19T06:35:34 < emeb_mac> I think most do.
2012-10-19T06:35:57 < ds2> what do you do for power management?
2012-10-19T06:38:17 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 268 seconds]
2012-10-19T06:38:45 < dongs> some shit i made used a lipo charger IC and some TI buckboost reg
2012-10-19T06:38:52 < dongs> tps61000 or someshit.
2012-10-19T06:39:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-19T06:40:04 < ds2> I mean on the STM32
2012-10-19T06:40:15 < ds2> just sleep? clocking down? both? other stuff
2012-10-19T06:40:16 < ds2> ?
2012-10-19T06:44:13 < dongs> nah this shit was constatntly writing to SD so it didnt matter
2012-10-19T06:45:00 < ds2> ah, you must have big batteries
2012-10-19T06:58:50 < ds2> what about folks using tiny batteries?
2012-10-19T06:58:55 -!- zippe [~Adium@173.11.99.161] has quit [Quit: Leaving.]
2012-10-19T06:59:19 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-10-19T07:04:16 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-19T07:08:00 < emeb_mac> dongs: you got a picture of your little DIP STM32 breakout handy?
2012-10-19T07:08:14 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 240 seconds]
2012-10-19T07:08:53 < dongs> haha uh sec
2012-10-19T07:09:34 < dongs> http://i.imgur.com/Jc9v4.jpg the only one in my imgur cache
2012-10-19T07:10:15 < emeb_mac> thx
2012-10-19T07:10:28 < emeb_mac> thinking about layout for the 373 breakout...
2012-10-19T07:16:59 < ds2> when you say 373 and breakout, I think of the 74LS373
2012-10-19T07:17:00 < ds2> :)
2012-10-19T07:24:52 < dongs> isnt  that already dip
2012-10-19T07:25:13 < dongs> a breakout for dip is awesome
2012-10-19T07:25:15 < dongs> super pro
2012-10-19T07:31:22 < ds2> think it does come in SOIC and such...
2012-10-19T07:39:52 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-19T07:49:11 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 252 seconds]
2012-10-19T08:13:36 < dongs> enjoy your pentium M in 2012
2012-10-19T08:14:39 < dongs> great
2012-10-19T08:28:45 -!- claude is now known as Claude
2012-10-19T08:51:46 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-19T09:05:24 -!- jkent [~jkent@173-23-182-202.client.mchsi.com] has quit [Ping timeout: 260 seconds]
2012-10-19T09:16:43 < emeb_mac> dongs:-> http://imagebin.org/232460
2012-10-19T09:16:56 < emeb_mac> schematic for f373 breakout
2012-10-19T09:17:37 < dongs> switcher?
2012-10-19T09:17:37 < dongs> haha.
2012-10-19T09:17:48 < dongs> wait wtf
2012-10-19T09:17:51 < dongs> ah. avcc shit
2012-10-19T09:18:49 < emeb_mac> yep
2012-10-19T09:19:06 < emeb_mac> in case one wants separate analog supply. bead if not.
2012-10-19T09:19:33 < emeb_mac> flyback: saw your comments
2012-10-19T09:20:03 < emeb_mac> hold flashlight behind it? :P
2012-10-19T09:20:26 < emeb_mac> ya think?
2012-10-19T09:21:10 < emeb_mac> dongs: went with 40 pins total. couldn't think which pins to leave off...
2012-10-19T09:21:48 < dongs> emeb_mac: well, all pins is good
2012-10-19T09:22:02 -!- dekar_ [~dekar@212.255.39.10] has joined ##stm32
2012-10-19T09:22:03 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-19T09:22:03 < dongs> waht did you end up for the xtal?
2012-10-19T09:22:05 < dongs> i onyl see 2 legs
2012-10-19T09:22:38 < emeb_mac> placeholder for now. Probably go one of the smaller 12 or 16 MHz parts.
2012-10-19T09:23:08 < emeb_mac> still need to nail down all the footprints.
2012-10-19T09:23:26 < dongs> boot0 perm to gnd?
2012-10-19T09:23:29 < dongs> how are you gonna bootloader
2012-10-19T09:23:36 < ds2> DOHHHHH
2012-10-19T09:23:51 < emeb_mac> no - boot0 will have a TP on it to pull up.
2012-10-19T09:24:05 < ds2> @;2S$%@#$@!$!@ microsd sockets
2012-10-19T09:24:18 < emeb_mac> ds2: something bite you?
2012-10-19T09:24:29 < dongs> i bet he didnt print it out before submitting boards
2012-10-19T09:24:41 < emeb_mac> RAGE
2012-10-19T09:25:14 < dongs> where is the LED connected to
2012-10-19T09:25:19 < dongs> o there
2012-10-19T09:25:20 < dongs> PF7
2012-10-19T09:25:21 < dongs> yeah k
2012-10-19T09:25:23 < dongs> uselss port
2012-10-19T09:25:35 < dongs> if I were you i'd put it on like one of useless jtag signals
2012-10-19T09:25:37 < ds2> I caught it during review
2012-10-19T09:25:38 < dongs> like TMS or someshit
2012-10-19T09:25:45 < ds2> damn sockets open the opposite way then I expected
2012-10-19T09:25:57 < emeb_mac> ds2: yeah - been there.
2012-10-19T09:26:01 -!- dekar [~dekar@212.255.38.104] has quit [Ping timeout: 268 seconds]
2012-10-19T09:26:31 < ds2> back to layout
2012-10-19T09:26:42 < emeb_mac> dongs: I'll check. Isn't TMS also SWDIO?
2012-10-19T09:27:33 < dongs> there one thats uselss in swd mode
2012-10-19T09:27:58 < dongs> JTDI i think
2012-10-19T09:28:11 < ds2> you guys prefer to bring out jtag instead of swd?
2012-10-19T09:28:22 < dongs> i bring out the full connector but not connect jtag signals
2012-10-19T09:29:07 < dongs>   *     @arg GPIO_Remap_SWJ_NoJTRST      : Full SWJ Enabled (JTAG-DP + SW-DP) but without JTRST
2012-10-19T09:29:10 < dongs>   *     @arg GPIO_Remap_SWJ_JTAGDisable  : JTAG-DP Disabled and SW-DP Enabled
2012-10-19T09:29:13 < dongs>   *     @arg GPIO_Remap_SWJ_Disable      : Full SWJ Disabled (JTAG-DP + SW-DP)
2012-10-19T09:29:16 < dongs> GPIO_Remap_SWJ_JTAGDisable will free JTDI shitz
2012-10-19T09:34:22 < emeb_mac> Yeah - really not much point in hooking up the full JTAG - SWD works just as well (if not better)
2012-10-19T09:35:07 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-10-19T09:35:44 < emeb_mac> Put a reset button on - I use that more often than I care to admit. :P
2012-10-19T09:37:20 < dongs> does freetard tools not have a debug->reset cpu or somethign menu?
2012-10-19T09:38:36 < emeb_mac> BMPM is controlled entirely through gdb - reset command appears to work.
2012-10-19T09:42:14 < zyp> I'm not putting reset buttons on my stuff
2012-10-19T09:42:43 < jpa-> i use the reset button on the stlink
2012-10-19T09:44:31 < emeb_mac> bringing RST to the debug connector probably does reduce the need for that.
2012-10-19T09:44:42 < emeb_mac> I hadn't been bringing it out prior.
2012-10-19T09:45:47 < zyp> I'll admit that the reset button on the waveshare board was kind of neat to have when I put it in deep sleep which killed the debugger connection
2012-10-19T09:50:42 < jpa-> been there, done that, used a piece of wire :)
2012-10-19T09:51:37 < emeb_mac> buttons I've been using lately are pretty small
2012-10-19T09:53:33 -!- jkent [~jkent@173-26-157-254.client.mchsi.com] has joined ##stm32
2012-10-19T10:00:06 -!- Claude is now known as claude
2012-10-19T10:01:03 < dongs> https://www.youtube.com/watch?v=mPHHr75qWAA
2012-10-19T10:01:04 < dongs> lol
2012-10-19T10:11:08 < cjbaird> neat
2012-10-19T10:13:07 < emeb_mac> burn baby burn
2012-10-19T10:13:22 < emeb_mac> manhole inferno...
2012-10-19T10:13:42 < cjbaird> lol
2012-10-19T10:14:16 < dongs> these are not the manholes youre thikning of
2012-10-19T10:15:55 -!- dekar_ [~dekar@212.255.39.10] has quit [Quit: This computer has gone to sleep]
2012-10-19T10:16:41 < cjbaird> My brother is involved with a Veolia landfill refuse gas recovery project..
2012-10-19T10:18:52 < cjbaird> Veolia being one of the multinational companies I referred to yesterday that're 'taking over' AU... because of a simple thing called 'running a company properly'..
2012-10-19T10:19:21 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-19T10:25:13 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-19T10:45:15 -!- claude is now known as claude_work
2012-10-19T10:49:31 < dongs> https://youtube.com/watch?v=OYuplf0WmmM
2012-10-19T10:49:34 < dongs> yay
2012-10-19T10:49:39 < dongs> now i can enjoy clean urls
2012-10-19T10:49:41 < dongs> again
2012-10-19T11:03:25 -!- dekar_ [~dekar@dslb-088-069-153-042.pools.arcor-ip.net] has joined ##stm32
2012-10-19T11:03:28 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-19T11:11:49 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-19T11:34:46 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 272 seconds]
2012-10-19T11:59:51 < dongs> attn Tectu / zyp http://neurogadget.com/2012/09/22/meet-the-next-necomimi-a-fluffy-tail-that-wags-with-your-mood-and-shares-your-location/4860
2012-10-19T12:01:08 < cjbaird> I'd prefer a Skunk version that maces people... -_-
2012-10-19T12:03:05 < dongs> but that wouldnt be kawaii
2012-10-19T12:07:44 < Laurenceb_> where the hell is the f3 header file?>
2012-10-19T12:07:44 < Laurenceb_> #
2012-10-19T12:08:47 < Laurenceb_> theres a load of example code on the st site now
2012-10-19T12:08:51 < Laurenceb_> but no header
2012-10-19T12:10:20 < dongs> huh?
2012-10-19T12:10:25 < dongs> in cmsis shit as usual?
2012-10-19T12:10:31 < Laurenceb_> oh
2012-10-19T12:10:44 < dongs> Libraries\CMSIS\Device\ST\STM32F30x\Include
2012-10-19T12:10:49 < dongs> 477k
2012-10-19T12:10:55 < Laurenceb_> ah
2012-10-19T12:10:57 < Laurenceb_> thanx
2012-10-19T12:11:03 < Laurenceb_> rage
2012-10-19T12:11:08 < Laurenceb_> f37x.h
2012-10-19T12:11:39 < Laurenceb_> i dont care about that
2012-10-19T12:12:18 < dongs> huh then way
2012-10-19T12:12:24 < dongs> wat?
2012-10-19T12:12:25 < dongs> i haev stm32f30x.h
2012-10-19T12:12:27 < dongs> fGt
2012-10-19T12:12:35 < Laurenceb_> oh its in the AN4046 code
2012-10-19T12:12:37 < Laurenceb_> i see
2012-10-19T12:12:46 < dongs> in f3discovery codez too
2012-10-19T12:12:47 < Laurenceb_> bbl going to RS to pick up my F3
2012-10-19T12:12:50 < dongs> how many comparators
2012-10-19T12:12:52 < dongs> on f3disco?
2012-10-19T12:12:53 < dongs> 2 or 3
2012-10-19T12:12:54 < Laurenceb_> 4
2012-10-19T12:13:00 < dongs> neat
2012-10-19T12:13:05 < Laurenceb_> bbl
2012-10-19T12:13:09 < dongs> pics whe nyou back
2012-10-19T12:14:39 < dongs> rogramming target (35464 bytes, 1 range) ... - RAM tested O.K. - Target programmed successfully - Completed after 1.731 sec
2012-10-19T12:14:47 < qyx_> mcp9808 breakout boards just arrived \o/
2012-10-19T12:17:17 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-19T12:22:13 < zyp> RS is apparently working on shipping mine, the card I used for the order was charged earlier today
2012-10-19T12:24:17 < cjbaird> I should be getting an F3Disco from E14 next week..
2012-10-19T12:48:18 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-19T12:50:48 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 244 seconds]
2012-10-19T13:05:45 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T13:08:36 < Tectu> -> 16bit :d
2012-10-19T13:10:40 < dongs> https://youtube.com/watch?v=qkXtMqmPFTg
2012-10-19T13:36:47 < zyp> wat
2012-10-19T13:36:58 < zyp> apparently my f3discovery is shipped and f0discovery is not
2012-10-19T13:39:00 < dongs> ahha
2012-10-19T13:49:09 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T13:53:02 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 268 seconds]
2012-10-19T14:01:26 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T14:01:59 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-19T14:02:51 < zyp> heh
2012-10-19T14:02:55 < zyp> «Generic message to confirm that any outstanding items will be in a separate delivery and notifications will be sent when they are despatched»
2012-10-19T14:03:35 < zyp> also, looks like they fucked up my addr
2012-10-19T14:04:13 < zyp> seems to be shipped to company name at addr, and I didn't provide a company name, so there is no name in the addr now
2012-10-19T14:04:18 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 264 seconds]
2012-10-19T14:04:36 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T14:06:14 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds]
2012-10-19T14:10:32 < Laurenceb> http://i.imgur.com/HWA9z.jpg
2012-10-19T14:10:48 < dongs> fgt
2012-10-19T14:11:38 < Laurenceb> its mien
2012-10-19T14:11:42 < zyp> I see olimex maple clone
2012-10-19T14:12:23 < dongs> and his faillogger
2012-10-19T14:13:56 < Laurenceb> and a df3120
2012-10-19T14:15:58 < Laurenceb> http://hackaday.com/tag/df3120/
2012-10-19T14:17:33 < Laurenceb> mine has 64MB ram
2012-10-19T14:17:34 < claude_work> Laurenceb, df3120 ;)
2012-10-19T14:18:11 < claude_work> ah yeah i spoted the df3120!
2012-10-19T14:21:55 < Laurenceb> lol discovery magno is uncalibrated
2012-10-19T14:22:35 < dongs> their magno sores is trash
2012-10-19T14:22:48 < dongs> its just atan2 or whatever
2012-10-19T14:22:55 < Laurenceb> oh i fail
2012-10-19T14:22:56 < dongs> of x/y
2012-10-19T14:23:00 < Laurenceb> its my steel desk
2012-10-19T14:23:05 < dongs> you probably got it next to your steel dong though
2012-10-19T14:23:06 < dongs> yueah
2012-10-19T14:23:27 < Laurenceb> ok off my desk its not too bad
2012-10-19T14:23:35 < Laurenceb> hard to say with just 8 leds
2012-10-19T14:23:36 < dongs> just keep it level
2012-10-19T14:23:41 < dongs> no acc compensation
2012-10-19T14:23:42 < Laurenceb> but its not horribly wrong
2012-10-19T14:23:54 < Laurenceb> hmm either the firmware is crap
2012-10-19T14:23:58 < Laurenceb> or my usb connector is
2012-10-19T14:24:05 < Laurenceb> it keeps hanging/rebooting
2012-10-19T14:24:12 < dongs> im going for the latter.
2012-10-19T14:24:12 < Laurenceb> probably my usb
2012-10-19T14:24:26 < Laurenceb> yeah this cables has given me issues before
2012-10-19T14:24:37 < Laurenceb> the acc code is weird
2012-10-19T14:24:51 < Laurenceb> seems to be high pass filtered or something
2012-10-19T14:25:35 < Laurenceb> yeah i think high pass 2hz or something
2012-10-19T14:25:38 < Laurenceb> very odd
2012-10-19T14:26:22 < Laurenceb> compass works well in middle of room using longer cable
2012-10-19T14:28:26 < Laurenceb> wonder if the accel is broken?
2012-10-19T14:28:45 < Laurenceb> i guess its more likely they used the inbuilt high pass
2012-10-19T14:28:57 < Laurenceb> oh
2012-10-19T14:29:03  * Laurenceb EPIC FAIL
2012-10-19T14:29:19 < Laurenceb> "press USER button to enable gyroscope"
2012-10-19T14:30:05 < Laurenceb> no wonder it looks like high pass
2012-10-19T14:30:43 < Laurenceb> oh and the "reboot/hang" is the accel induced mode switch thingy
2012-10-19T14:31:04 < Laurenceb> seems to work 100% then
2012-10-19T14:36:27 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-19T14:48:28 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-19T14:53:41 < Laurenceb> 2012-10-19T12:56:02 INFO src/stlink-common.c: Loading device parameters....
2012-10-19T14:53:42 < Laurenceb> 2012-10-19T12:56:02 WARN src/stlink-common.c: unknown chip id! 0x10016422
2012-10-19T14:53:48 < Laurenceb> hmm this will take some work
2012-10-19T14:56:21 -!- daku is now known as DaKu
2012-10-19T14:56:44 -!- claude_work is now known as claude
2012-10-19T14:58:12 < zyp> https://github.com/texane/stlink/blob/master/src/stlink-common.h#L136 <- first change you need is adding it here
2012-10-19T14:59:39 < zyp> once that is done, I'm not sure if you really have to do anything more
2012-10-19T15:01:37 < zyp> ah, some flashing functions will fail on unknown chip id
2012-10-19T15:09:04 < Laurenceb> wont the flash config stuff fix it?
2012-10-19T15:09:17 < Laurenceb>      .flash_size_reg = 0x1ffff7e0,
2012-10-19T15:09:17 < Laurenceb>                     .flash_pagesize = 0x800,
2012-10-19T15:09:28 < Laurenceb> ^isnt that all thats needed?
2012-10-19T15:09:52 < zyp> some functions behave differently based on chip id
2012-10-19T15:10:04 < zyp> most are special cases for F2/F4, so they are fine
2012-10-19T15:10:10 < zyp> but some have no default behavior
2012-10-19T15:10:39 < zyp> but anyway, just add the chip id first, then see where it complains next and fix it
2012-10-19T15:11:09 < zyp> https://github.com/texane/stlink/blob/master/src/stlink-common.c#L1290 <- I mean stuff like this
2012-10-19T15:14:20 < Laurenceb> i see
2012-10-19T15:14:25 < Laurenceb> theres no F1 there
2012-10-19T15:15:11 < zyp> STM32VL_CORE_ID
2012-10-19T15:15:17 < Laurenceb> oh
2012-10-19T15:15:22 < zyp> (I believe)
2012-10-19T15:15:32 < zyp> you can probably just add the F3 to that case
2012-10-19T15:15:47 < zyp> when needed
2012-10-19T15:15:47 < Laurenceb> yeah id tend to agree
2012-10-19T15:16:00 < Laurenceb> i might try this.. in a bit
2012-10-19T15:16:43 < zyp> support for different chips should be cleaned up a bit
2012-10-19T15:17:08 < zyp> it seems to just have been hacked on for each new variant, by different people with different styles
2012-10-19T15:20:30 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-19T15:21:17 < Laurenceb> http://2.bp.blogspot.com/--tE-vK8lwQ4/Tzz8t3wYk5I/AAAAAAAAAN8/QVTO32CG-pM/s1600/loneboner_Forever_Alone_Comics-s697x622-102197-580.jpg
2012-10-19T15:21:24 < Laurenceb> the story of Tectu
2012-10-19T15:22:18 < Tectu> or not
2012-10-19T15:42:22 < karlp> that vlcore id is the ugliest part,
2012-10-19T15:42:33 < karlp> all the rest is on chip id, then it has an else if core id..
2012-10-19T15:42:46 < karlp> it should all be chipids really.
2012-10-19T15:48:00 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-19T15:51:06 < Laurenceb> wtf
2012-10-19T15:51:21 < Laurenceb> the "user" usb is connecting as a HID device
2012-10-19T15:52:39 < Laurenceb> Bus 005 Device 090: ID 0483:5710 SGS Thomson Microelectronics
2012-10-19T15:53:37 < dongs> yeah?
2012-10-19T15:53:40 < dongs> why so surprised
2012-10-19T15:54:42 < Laurenceb> theres no xtal
2012-10-19T15:54:57 < Laurenceb> i cant work out what its supposed to be
2012-10-19T15:54:58 < zyp> does it emulate a mouse like the the default f4 demo did?
2012-10-19T15:55:11 < Laurenceb> not at the moment
2012-10-19T15:55:22 < Laurenceb> maybe i need to click user
2012-10-19T15:55:28 -!- cjbaird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-19T15:55:31 < zyp> yeah, you had to do that on F4
2012-10-19T15:56:09 < Laurenceb> nothing
2012-10-19T15:56:13 < Laurenceb> this is odd
2012-10-19T15:56:29 < zyp> it's a bit interesting that it ships without crystal with usb enabled, when datasheet claims that usb requires external crystal to be precise enough
2012-10-19T15:57:10 < zyp> either datasheet is wrong or they built it violating the spec
2012-10-19T15:58:59 < Laurenceb> The STM32F303VCT6 MCU is also used to drive the second USB mini-B connector (USB
2012-10-19T15:58:59 < Laurenceb> USER) which allows the board to be used as a USB Device. The STM32F3DISCOVERY
2012-10-19T15:58:59 < Laurenceb> can then act as a USB joystick, mouse, or other similar device
2012-10-19T16:01:13 < Laurenceb> ah found it
2012-10-19T16:01:18 < Laurenceb> /dev/hidraw3
2012-10-19T16:02:43 < Laurenceb> but no data out
2012-10-19T16:04:56 < Laurenceb> wait wtf - theres a /dev/input/mouse2
2012-10-19T16:06:11 < Laurenceb> but no data from it
2012-10-19T16:08:52 < Laurenceb> very odd - how do you get mouse to work on f4?
2012-10-19T16:08:53 < dongs> plug it into a windows desktop
2012-10-19T16:09:01 < Laurenceb> i did
2012-10-19T16:09:10 < Laurenceb> "new device detected"
2012-10-19T16:09:18 < Laurenceb> "new HID device connected and ready"
2012-10-19T16:10:07 < dongs> great
2012-10-19T16:16:40 < Laurenceb> then nothing
2012-10-19T16:16:55 < Laurenceb> looks like its a mouse that sends no data
2012-10-19T16:17:40 < zyp> did you try waving it around?
2012-10-19T16:19:45 < Laurenceb> yes
2012-10-19T16:19:54 < Laurenceb> and pressing user button
2012-10-19T16:20:56 < Laurenceb> attn dongs
2012-10-19T16:20:57 < Laurenceb> http://i.imgur.com/5BLOO.jpg
2012-10-19T16:21:17 < dongs> wahts the funny
2012-10-19T16:22:05 < Laurenceb> http://www.thomasandfriends.com/uk/Thomas.mvc/Home
2012-10-19T16:34:45 < Laurenceb> nope theres no data
2012-10-19T16:34:57 < Laurenceb> usb mouse but nothing coming out
2012-10-19T16:45:25 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-19T16:45:29 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-19T16:46:07 < dongs> install HSE.
2012-10-19T16:50:53 < Laurenceb> health and safety executive?
2012-10-19T16:53:45 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-19T16:54:10 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-19T17:03:15 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-19T17:06:17 -!- Nutter` is now known as Nutter
2012-10-19T17:10:00 < Laurenceb> this is odd
2012-10-19T17:10:14 <+Steffanx> even
2012-10-19T17:10:15 < Laurenceb> f3discovery has resistors between the f3 and the hse xtal
2012-10-19T17:11:30 < qyx_> why odd?
2012-10-19T17:11:40 < zyp> populated?
2012-10-19T17:14:27 < Laurenceb> no
2012-10-19T17:14:35 < Laurenceb> theres footprints for them
2012-10-19T17:14:59 < jpa-> sometimes people want crazy resistors with their crystals
2012-10-19T17:15:03 < jpa-> i think it improves the taste
2012-10-19T17:15:04 < dongs> Laurenceb: read the fucking manual
2012-10-19T17:15:06 < dongs> for f3 board
2012-10-19T17:15:14 < Laurenceb> i have done
2012-10-19T17:15:16 < dongs> you can run HSE off a c rystal or by 8mhz MCO from stlink part
2012-10-19T17:15:34 < Laurenceb> so
2012-10-19T17:15:41 < dongs> so thats hwo you get working usb
2012-10-19T17:15:48 < dongs> internal oscialltor cant do usb
2012-10-19T17:15:53 < dongs> fi you read discovery firmware readme
2012-10-19T17:15:55 < dongs> for usb-related shit
2012-10-19T17:16:00 < dongs> it says you gotta hookup a crystal
2012-10-19T17:16:01 < dongs> to use it
2012-10-19T17:16:03 < Laurenceb> do i need to short a jumper?
2012-10-19T17:16:16 < dongs> read the manual, you need to remove some jumpers and add some and solder in 8mzh x tal
2012-10-19T17:16:31 < Laurenceb> isnt mco simpler
2012-10-19T17:16:36 < dongs> about same
2012-10-19T17:16:39 < dongs> you still need to remove/solder
2012-10-19T17:17:37 < dongs> page 19 of manual
2012-10-19T17:17:42 < dongs> OSC clock supply
2012-10-19T17:18:44 < Laurenceb> If SB12 is ON, SB17 must be OFF and R33 removed to provide the clock source from MCO
2012-10-19T17:18:44 < Laurenceb> correctly.
2012-10-19T17:18:53 < Laurenceb> it comes with that set up
2012-10-19T17:18:59 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-19T17:19:16 < dongs> you are reading some old manual
2012-10-19T17:19:20 < dongs> my shit has different wording
2012-10-19T17:19:21 < dongs> but yeah.
2012-10-19T17:19:26 < dongs> it might be setup for MCO by dfeault
2012-10-19T17:19:31 < dongs> which is how they get away with working usb
2012-10-19T17:20:15 < Laurenceb> SB12 closed, SB17 open, R33 removed
2012-10-19T17:20:39 < Laurenceb> thats how my board is setup
2012-10-19T17:21:17 < dongs> ok. so youre good to go for stable osc.
2012-10-19T17:21:26 < dongs> just make sure stlink part is powered up.
2012-10-19T17:21:41 < Laurenceb> oh
2012-10-19T17:21:59 < zyp> ooh
2012-10-19T17:22:09 < dongs> zyp, pretty clever huh?
2012-10-19T17:22:09 < zyp> so that's why it ships without crystal
2012-10-19T17:22:12 < dongs> ya
2012-10-19T17:24:46 < Laurenceb> still nothing
2012-10-19T17:25:09 < Laurenceb> with two usb cables connected
2012-10-19T17:26:19 < zyp> well, a bad usb clock would surely cause the host to report spurious errors, not just dropp all traffic
2012-10-19T17:26:40 < Laurenceb> yeah
2012-10-19T17:26:55 < Laurenceb> i suspect its either disabled in the stock firmware
2012-10-19T17:27:07 < Laurenceb> or theres some sort of button trick to make it run
2012-10-19T17:27:17 < dongs> guru meditation
2012-10-19T17:29:11 < Laurenceb> anyway i need to do work
2012-10-19T17:29:21 < dongs> <laurenceb> opening 4chan
2012-10-19T17:29:21 < Laurenceb> will try fixing texane l8trs
2012-10-19T17:29:27 < Laurenceb> fapfapfap
2012-10-19T17:29:42 < Laurenceb> ghetto booty tgp
2012-10-19T17:29:46 < Laurenceb> oops wrong window
2012-10-19T17:29:54 < dongs> what the hell is tgp
2012-10-19T17:30:18 < Laurenceb> A thumbnail gallery post (TGP) is a common type of ad-driven website that provides links to free Internet pornography
2012-10-19T17:30:51 < dongs> Circle jerk (CJ) TGP
2012-10-19T17:30:51 < dongs> These sites look like TGP sites but end up sending users on a 'circle jerk' of blind links. When users click on a gallery link provided by a TGP, they will be unwillingly redirected to another TGP instead.
2012-10-19T17:31:03 < dongs> sounds like a typical day on the internet
2012-10-19T17:31:08 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-19T17:31:34 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-19T17:34:52 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Quit: Life is too short]
2012-10-19T17:36:20 < Laurenceb>  "Tutorial about TGPs and galleries for adult webmasters, by Cozy Academy"
2012-10-19T17:36:23 -!- Vutral [ss@vutral.net] has joined ##stm32
2012-10-19T17:36:25 -!- Vutral [ss@vutral.net] has quit [Changing host]
2012-10-19T17:36:25 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-19T17:36:32 < Laurenceb> wikipedians know what too much about that sort of stuff
2012-10-19T17:36:39 < dongs> haha
2012-10-19T17:37:15 < dongs> yea i bet the page on child pornography on wikipedo has been very well researched, and nominated as " good article " more than once
2012-10-19T17:38:14 < Laurenceb> http://en.wikipedia.org/wiki/Pornographic_actress
2012-10-19T17:38:39 < Laurenceb> wayyy too many links to "actress" pages
2012-10-19T17:38:47 < dongs> http://en.wikipedia.org/wiki/File:Bobby%26Flex2.jpg
2012-10-19T17:39:47 < Laurenceb> eye bleach
2012-10-19T17:46:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-19T17:56:16 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T17:56:40 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 272 seconds]
2012-10-19T18:14:18 <+Steffanx> You as ex-gnaa must like that photo dongs
2012-10-19T18:15:18 <+Steffanx> Laurenceb, google: STM32F6  ( and also try STM32L4 ) :D
2012-10-19T18:16:36 < Laurenceb> lol
2012-10-19T18:18:13 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-19T18:24:59 -!- dekar_ [~dekar@dslb-088-069-153-042.pools.arcor-ip.net] has quit [Ping timeout: 260 seconds]
2012-10-19T18:33:23 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T18:34:22 < Laurenceb> https://www.sparkfun.com/products/11451
2012-10-19T18:34:23 < Laurenceb> omg
2012-10-19T18:35:27 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Read error: Operation timed out]
2012-10-19T18:35:39 < BrainDamage> only 0.75?
2012-10-19T18:35:48 < BrainDamage> that's cheap for sparkfun's standard
2012-10-19T18:35:57 < BrainDamage> they should've sold it for 7.5$
2012-10-19T18:35:58 < Laurenceb> i was trolling
2012-10-19T18:36:03 < BrainDamage> I am too
2012-10-19T18:36:07 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Quit: No Ping reply in 180 seconds.]
2012-10-19T18:36:13 < Laurenceb> trollercoaster
2012-10-19T18:36:29 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-19T18:57:05 < Laurenceb> 2012-10-19T16:59:30 INFO src/stlink-common.c: Device connected is: F3 device, id 0x10016422
2012-10-19T18:57:05 < Laurenceb> 2012-10-19T16:59:30 INFO src/stlink-common.c: SRAM size: 0xc000 bytes (48 KiB), Flash: 0x3fffc00 bytes (65535 KiB) in pages of 2048 bytes
2012-10-19T18:57:05 < Laurenceb> Chip ID is 00000422, Core ID is  2ba01477.
2012-10-19T18:57:07 < Laurenceb> hell yeah
2012-10-19T18:58:13 < jpa-> :)
2012-10-19T18:58:17 < jpa-> does it work?
2012-10-19T18:58:55 < Laurenceb> seems to
2012-10-19T18:59:04 < Laurenceb> i havent tried compiling any code yet
2012-10-19T18:59:13 < Laurenceb> bet the flashing goes wrong somewhere
2012-10-19T18:59:22 < jpa-> just flash some random bin to test
2012-10-19T18:59:25 < Laurenceb> i can halt and dump registers and stuff
2012-10-19T18:59:36 < Laurenceb> id like to backup the example
2012-10-19T18:59:52 < Laurenceb> can you read flash with gdb?
2012-10-19T19:00:07 < jpa-> yeah, though it is easier with the st-flash tool
2012-10-19T19:00:26 < jpa-> but in gdb, dump memory 0x0000000 ... something
2012-10-19T19:04:19 < Laurenceb> hmm stflash dumps a massive file
2012-10-19T19:04:25 < Laurenceb> and never completes
2012-10-19T19:04:42 < jpa-> did you give it the size?
2012-10-19T19:05:04 < Laurenceb> nope
2012-10-19T19:05:06 < Laurenceb> lol
2012-10-19T19:06:42 < Laurenceb> worked this time
2012-10-19T19:06:46 < Laurenceb> looking at dissasm
2012-10-19T19:08:09 < Laurenceb> objdump should do that right?
2012-10-19T19:10:09 -!- barthess [~barthess@178.154.17.125] has joined ##stm32
2012-10-19T19:10:15 < Laurenceb> odd - objdump doesnt like it
2012-10-19T19:10:26 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-19T19:10:32 < Laurenceb> i see "STM32F3"  in a hex editor
2012-10-19T19:10:39 < Laurenceb> looks like usb config strings
2012-10-19T19:11:16 < Laurenceb> arm-none-eabi-objdump: f3dis.bin: File format not recognized
2012-10-19T19:11:18 < zyp> you have to tell objdump about the format when it's looking at raw binary without any metadata
2012-10-19T19:11:24 < Laurenceb> oh
2012-10-19T19:11:39 < Laurenceb> how?
2012-10-19T19:11:47 < zyp> man objdump
2012-10-19T19:11:58 < zyp> :p
2012-10-19T19:12:49 < zyp> I don't remember the exact ones, but you have to specify that it's little endian arm and that it should assume everything is thumb
2012-10-19T19:14:05 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 256 seconds]
2012-10-19T19:16:34 < Laurenceb> ok i kind of see
2012-10-19T19:16:40 < Laurenceb> its asking for a cpu now
2012-10-19T19:19:39 < Laurenceb> how the hell do i get a list of architectures?
2012-10-19T19:25:20 < Laurenceb> binary
2012-10-19T19:25:21 < Laurenceb>  (header endianness unknown, data endianness unknown)
2012-10-19T19:25:21 < Laurenceb>   arm
2012-10-19T19:31:23 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-19T19:31:43 < Laurenceb> oh finally
2012-10-19T19:31:44 < Laurenceb> arm-none-eabi-objdump --endian=little -b binary -m arm force-thumb -D f3dis.bin
2012-10-19T19:32:09 < zyp> that looks about right, yeah
2012-10-19T19:35:12 < Laurenceb> http://pastebin.com/8Zw84H3M
2012-10-19T19:37:21 < Laurenceb> arm-none-eabi-objdump --endian=little -b binary -m arm -M force-thumb -D f3dis.bin
2012-10-19T19:40:04 < Laurenceb> Flash: 0x3fffc00 bytes (65535 KiB)
2012-10-19T19:40:11 < Laurenceb> ^thats not right is it?
2012-10-19T19:41:51 < zyp> no, that's about 64 MB
2012-10-19T19:42:04 < Laurenceb> is shifted by 8 bits
2012-10-19T19:42:51 < Laurenceb> for some reason i tried to read 64K of flash and the dump is only 18
2012-10-19T19:43:00 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-19T19:43:07 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-19T19:43:08 < Laurenceb> it can determine filled flash size?
2012-10-19T19:43:42 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-19T19:43:45 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-19T19:43:47 < zyp> talking about the file or objdump?
2012-10-19T19:44:52 < zyp> objdump will usually skip consecutive regions of all ones or all zeros
2012-10-19T19:44:54 < Laurenceb> file
2012-10-19T19:45:09 < zyp> that sounds more like you messed up something
2012-10-19T19:45:45 < Laurenceb> hmm
2012-10-19T19:46:21 < Laurenceb> ./st-flash read ./f3dis.bin 0x08000000 65535
2012-10-19T19:46:51 < zyp> that last number seems to be kilobytes :p
2012-10-19T19:46:58 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-19T19:47:03 < zyp> 17:57:05 < Laurenceb> 2012-10-19T16:59:30 INFO src/stlink-common.c: SRAM size: 0xc000 bytes (48 KiB), Flash: 0x3fffc00 bytes (65535 KiB) in pages of 2048 bytes
2012-10-19T19:47:58 < Laurenceb> oh it expects hex
2012-10-19T19:48:30 < zyp> no it doesn't, it expects you to say the number of kilobytes you want
2012-10-19T19:48:34 < zyp> not the number of bytes
2012-10-19T19:49:13 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-19T19:49:37 < TitanMKD> hi
2012-10-19T19:49:46 < zyp> good evening
2012-10-19T19:49:57 < TitanMKD> you too
2012-10-19T19:50:13 <+Steffanx> Lo
2012-10-19T19:50:13 < TitanMKD> any news on the LPC43xx SDRAM board ?
2012-10-19T19:50:23 < zyp> not yet
2012-10-19T19:50:23 < Laurenceb> def bytes
2012-10-19T19:50:27 < Laurenceb> just tested it
2012-10-19T19:50:37 < zyp> I did some cleanup yesterday, but nothing major
2012-10-19T19:50:53 < Laurenceb> it maxes out at 17224bytes
2012-10-19T19:51:11 < TitanMKD> zyp just an idea it will be a must to use DP SOB case
2012-10-19T19:51:15 < zyp> Laurenceb, oh, ok
2012-10-19T19:51:35 < Laurenceb> i get 256 bytes with 0x100
2012-10-19T19:51:51 < Laurenceb> i get more as i increase up to 17224bytes
2012-10-19T19:51:56 < Laurenceb> any idea why?
2012-10-19T19:52:05 < zyp> TitanMKD, you should probably stop using the term «a must» like that :)
2012-10-19T19:52:42 < TitanMKD> hehe
2012-10-19T19:52:59 < Laurenceb> http://pastebin.com/8Zw84H3M
2012-10-19T19:53:04 < zyp> I'm not sure what you mean by it, but I read it as «a requirement», and you shouldn't go stating requirements on other people's projects ;)
2012-10-19T19:53:07 < Laurenceb> it doesnt look right to me?
2012-10-19T19:53:26 < TitanMKD> if anyone has implemented Dual CDC for chibios i will be really interested in it
2012-10-19T19:53:34 < TitanMKD> especially for STM32F4
2012-10-19T19:53:48 < zyp> Laurenceb, starts out right with the vectors at least
2012-10-19T19:53:53 < TitanMKD> Else i will start using only simple USB CDC
2012-10-19T19:54:04 < Laurenceb> really
2012-10-19T19:54:09 < zyp> stack is 0x20000648, reset is 0x080040b5
2012-10-19T19:54:16 < zyp> so that part looks sane
2012-10-19T19:54:22 < Laurenceb> why the 648?
2012-10-19T19:54:40 < zyp> no idea, maybe they place more shit above it
2012-10-19T19:54:52 < Laurenceb> hmm
2012-10-19T19:55:00 < Laurenceb> maybe ill try flashing back the dump then
2012-10-19T19:56:08 < zyp> I suspect it's incomplete
2012-10-19T19:56:22 < zyp> it still has data at the end
2012-10-19T19:56:50 < zyp> a complete dump would normally show all ones for the unused area at the end
2012-10-19T19:56:51 < Laurenceb> bad idea to flash back then?
2012-10-19T19:56:56 < zyp> since unprogrammed flash is high
2012-10-19T19:57:25 < zyp> that's up to you
2012-10-19T19:57:46 < zyp> it shouldn't harm anything, but you shouldn't be surprised if it crashes
2012-10-19T19:57:53 < Laurenceb> hmm ill try and fix the size stuff
2012-10-19T19:58:11 < Laurenceb> yeah i dont want to lose the demo
2012-10-19T19:59:20 < zyp> then it would be a bad idea :p
2012-10-19T20:05:20 < Laurenceb> looks like size register is a little different
2012-10-19T20:07:00 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has quit [Quit: Leaving]
2012-10-19T20:09:33 < Laurenceb> 2012-10-19T18:12:11 INFO src/stlink-common.c: SRAM size: 0xc000 bytes (48 KiB), Flash: 0x40000 bytes (256 KiB) in pages of 2048 bytes
2012-10-19T20:09:36 < Laurenceb> looks more sane
2012-10-19T20:09:48 < Laurenceb> .flash_size_reg = 0x1ffff7cc,
2012-10-19T20:09:55 < Posterdati> hi people
2012-10-19T20:11:14 -!- HTT-Bird [~Birdz0r@unaffiliated/htt-bird] has joined ##stm32
2012-10-19T20:15:17 < Laurenceb> well this is odd
2012-10-19T20:15:51 < Laurenceb> https://github.com/texane/stlink/blob/master/flash/main.c#L163
2012-10-19T20:15:59 < Laurenceb> i break there and the size is 256KiB
2012-10-19T20:16:09 < Laurenceb> yet the read file is only 17224 bytes
2012-10-19T20:18:32 < Posterdati> zyp: arm is a fantastic arch I love the way I implemented the context switcher
2012-10-19T20:27:48 -!- prog [prog@bornbrilliant.org] has quit [Quit: That's all folks!]
2012-10-19T20:28:00 < Laurenceb> oh found it
2012-10-19T20:28:13 < Laurenceb> theres a trim operation as part of st-flash
2012-10-19T20:28:25 < Laurenceb> it removes whatever it thinks is null on the end
2012-10-19T20:30:30 -!- dekar [~dekar@212.255.39.10] has joined ##stm32
2012-10-19T20:30:31 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-19T20:38:13 -!- prog [prog@bornbrilliant.org] has joined ##stm32
2012-10-19T20:40:00 < Laurenceb> 2012-10-19T18:42:38 INFO src/stlink-common.c: Flash written and verified! jolly good!
2012-10-19T20:40:04 < Laurenceb> HELL YEAH
2012-10-19T20:40:07 < Laurenceb> whos the boss
2012-10-19T20:45:08 < karlp> score
2012-10-19T20:45:52 < karlp> there's probably some bugs in properly detecting it's fpu registers and stuff.
2012-10-19T20:45:59 < Laurenceb> only i didnt follow the correct naming scheme
2012-10-19T20:46:07 < karlp> is it just using the f1 flash loader?
2012-10-19T20:46:10 < Laurenceb> core versus chipid
2012-10-19T20:46:13 < Laurenceb> yes
2012-10-19T20:46:18 < Laurenceb> fl-cl
2012-10-19T20:46:32 < Laurenceb> ill github it in a sec
2012-10-19T20:46:50 < Laurenceb> its fast too
2012-10-19T20:49:41 < Laurenceb> hmm whats the best way to do this?
2012-10-19T20:49:50 < Laurenceb> just setup a new github repor
2012-10-19T20:50:02 < Laurenceb> or create a fork?
2012-10-19T20:50:24 < karlp> probably easiest is to click fork on github,
2012-10-19T20:50:47 < karlp> clone that, put your changes there, commit and push to github, then click on pull requests
2012-10-19T20:52:17 < Laurenceb> hmm i already have my changes on a local copy
2012-10-19T20:52:23 < Laurenceb> ok so ive forked it
2012-10-19T20:52:32 < Laurenceb> and i already have the changes on my machine
2012-10-19T20:52:36 < Laurenceb> whjat to do now?
2012-10-19T20:53:33 < Laurenceb> ive gtg :(
2012-10-19T20:53:39 < Laurenceb> will upload later this evening
2012-10-19T20:53:39 < TitanMKD> what are the news on stlink ?
2012-10-19T20:54:13 < karlp> Laurenceb: just added flash support for the f3
2012-10-19T20:57:16 < TitanMKD> does anyone know an open source tool for Serial Wire Viewer ?
2012-10-19T20:57:23 < jpa-> Laurenceb: cp -r new_version/* old_version/   .. they are just files anyway, just don't copy .git :)
2012-10-19T21:04:06 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-19T21:14:57 < karlp> TitanMKD: gsmcmullin's BMP repo has some python plugin for gdb, but there's nothing polished.
2012-10-19T21:15:19 < karlp> I'd like to merge that work into stlink one day maybe
2012-10-19T21:15:20 < karlp> or you can
2012-10-19T21:15:28 < TitanMKD> karlp ha ok
2012-10-19T21:16:09 < karlp> Laurenceb seems to be on a roll, you can ask him.
2012-10-19T21:16:34 < TitanMKD> karlp does there's possibility to use Serial Wire Viewer with just a JATG (instead of a Trace) ?
2012-10-19T21:17:20 < TitanMKD> karlp especially to dump events
2012-10-19T21:21:57 < Thorn> TitanMKD: not with jtag, with swd. it uses a single additional signal
2012-10-19T21:21:58 -!- DaKu is now known as daku
2012-10-19T21:33:14 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-19T21:34:52 < TitanMKD> Thorn anyway SWD is good but i must check which is this additional wire
2012-10-19T21:36:34 < Thorn> tck = swclk, tms = swdio, tdo = swo iirc
2012-10-19T21:37:42 < TitanMKD> so just the standard swd connector
2012-10-19T21:57:58 < karlp> if you mean the mini 10 pin, then yes,
2012-10-19T21:58:11 < karlp> as long as you routed the swo/swv line to the connector,
2012-10-19T21:58:35 < karlp> it's a line in addition to the swclk and swdio
2012-10-19T22:01:01 < TitanMKD> ok thanks
2012-10-19T22:05:46 < Posterdati> hi
2012-10-19T22:09:05 < zyp> the nice part about jtag is that routing out jtag means that you also get swd+swo available for free on the same pins
2012-10-19T22:11:22 < zyp> Laurenceb, to upload your changes to github you just make a fork on github, then change the remote on your local repo to point to your fork instead of texane upstream and push
2012-10-19T22:11:28 < zyp> then it gets pushed to your fork
2012-10-19T22:11:36 < zyp> and then you can submit a github pull request
2012-10-19T22:13:21 < zyp> karlp, magictpa doesn't really seem to need to be merged into texane/stlink
2012-10-19T22:14:05 < zyp> magictpa runs as a gdb plugin and accesses the swo endpoint of the bmp directly through libusb
2012-10-19T22:14:36 < zyp> so you just need to fork it and adapt it to hook to stlink's swo endpoint instead
2012-10-19T22:25:17 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-19T22:38:21 < karlp> yeah, but to use magictpa, the stlink side needs to output it in the same format as bmp, which may or may not be the same as what comes natively out of the stlinkv2s endpoint
2012-10-19T22:39:24 < zyp> that's what I meant by fork and adapt
2012-10-19T22:40:34 < zyp> it's simpler to just adapt magictpa to whatever stlinkv2 outputs than writing a translator and then adapting magictpa to read from your translator instead of a usb ep
2012-10-19T22:42:03 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-19T22:42:13 <+Steffanx> afaik they both ouput the raw stream..
2012-10-19T22:43:22 < Laurenceb_> sup
2012-10-19T22:43:28 <+Steffanx> Tha sky bro
2012-10-19T22:44:40 < zyp> 21:11:21 < zyp> Laurenceb, to upload your changes to github you just make a fork on github, then change the remote on your local repo to point to your fork instead of texane upstream and push
2012-10-19T22:46:22 < Laurenceb_> ok
2012-10-19T22:46:30 < Laurenceb_> coming up in a few minutes then
2012-10-19T22:49:19 < Thorn> If you use 3.3v,please JP1.If you use 5v ,you must JP1 , or LCD will broken.
2012-10-19T22:50:08 <+Steffanx> I agree Thorn
2012-10-19T22:50:33 < zyp> ok
2012-10-19T23:01:26 < Laurenceb_> git remote add -m master https://github.com/laurenceb/stlink
2012-10-19T23:01:27 < Laurenceb_> usage: git remote add [<options>] <name> <url>
2012-10-19T23:01:28 < Laurenceb_> huh
2012-10-19T23:02:28 <+Steffanx> Hahaha, git
2012-10-19T23:10:02 < BrainDamage> wrong name
2012-10-19T23:10:16 < BrainDamage> git remote add reponame url
2012-10-19T23:10:24 < BrainDamage> reponame != branchname
2012-10-19T23:12:00 < Laurenceb_> got it
2012-10-19T23:12:01 < Laurenceb_> https://github.com/Laurenceb/stlink/
2012-10-19T23:13:20 < Laurenceb_> i probably havent followed the right "protocol" for adding the chipid/coreid stuff
2012-10-19T23:30:09 < karlp> no, you did the right thing.
2012-10-19T23:30:22 < karlp> v100 was always wrong,
2012-10-19T23:30:34 < karlp> f0 followed f100 instead of L1/f2/f4
--- Day changed Sat Oct 20 2012
2012-10-20T00:03:09 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-20T00:06:35 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 260 seconds]
2012-10-20T00:10:34 -!- CheBuzz [~CheBuzz@76.164.192.124] has joined ##stm32
2012-10-20T00:10:58 -!- CheBuzz is now known as Guest56399
2012-10-20T00:13:57 -!- Guest56399 [~CheBuzz@76.164.192.124] has quit [Client Quit]
2012-10-20T00:14:53 -!- CheBuzz_ [~CheBuzz@76.164.192.124] has joined ##stm32
2012-10-20T00:20:18 -!- CheBuzz_ is now known as CheBuzz_Home
2012-10-20T00:30:20 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has joined ##stm32
2012-10-20T00:31:03 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-20T01:02:41 -!- barthess [~barthess@178.154.17.125] has quit [Ping timeout: 256 seconds]
2012-10-20T01:14:23 < Laurenceb_> http://people.howstuffworks.com/friday-thirteenth.htm
2012-10-20T01:14:26 < Laurenceb_> what the hell
2012-10-20T01:24:05 < BrainDamage> I've seen "instructions" how to boil water
2012-10-20T01:24:21 < Rickta59> anyone familiar with gdb rsp protocol ?
2012-10-20T01:25:13 -!- upgrdman_ [42a6d414@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-20T01:25:21 < Rickta59> i'm having some limited success but the breakpoints don't seem to be triggering
2012-10-20T01:25:25 < Rickta59> http://paste.debian.net/201965/
2012-10-20T01:25:41 < Rickta59> debug output for those with rsp knowledge
2012-10-20T01:43:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-20T01:58:31 < dongs> hello trolls
2012-10-20T02:01:08 < HTT-Bird> dongs: ding
2012-10-20T02:01:18 < Laurenceb_> http://uk.mouser.com/ProductDetail/NXP-Semiconductors/LD6806CX4-30H315/?qs=sGAEpiMZZMuLLNXTG1MZaoRhM8gSaH4BL%252bFj4GxDdJ0%3d
2012-10-20T02:01:24  * Laurenceb_ wonders if he can solder ^
2012-10-20T02:02:01 < HTT-Bird> Laurenceb: hot air or a toaster oven would do the trick
2012-10-20T02:02:54 < dongs> Laurenceb_: lol
2012-10-20T02:02:56 < dongs> nice
2012-10-20T02:03:05 < Laurenceb_> 0.75mm square
2012-10-20T02:03:22 < Laurenceb_> dongs: I got F3 flashing to work with texane
2012-10-20T02:03:31 < dongs> nice.
2012-10-20T02:03:41 < dongs> im sure by the time it arrives here I'll just download keil/crosworks update and it will work
2012-10-20T02:03:45 < dongs> if it doesnt already.
2012-10-20T02:03:46 < Laurenceb_> lol
2012-10-20T02:06:08 -!- cjbaird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-20T02:07:10 < cjbaird> Hell No.
2012-10-20T02:13:01 < dongs> dicktrace 2.3 is out
2012-10-20T02:39:45 < Laurenceb_> flyback is obsessed with his pcs
2012-10-20T02:45:14 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Quit: TzacksT]
2012-10-20T02:45:49 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-20T03:53:25 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 244 seconds]
2012-10-20T04:20:01 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-20T04:57:57 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-20T05:01:56 < dongs> attn Laurenceb http://www.sharp.co.jp/fappy/
2012-10-20T05:02:44 < zyp> hah
2012-10-20T05:07:38 < dongs> Laurenceb is screwed  http://www.animenewsnetwork.com/news/2012-10-18/missouri-man-pleads-guilty-to-possession-of-cartoon-child-pornography
2012-10-20T05:11:21 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-20T05:14:43 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 252 seconds]
2012-10-20T05:17:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-20T05:24:54 < zippe> dongs: welcome to the new world order
2012-10-20T05:27:42 < cjbaird> We _warned_ everyone shit like this would happen when the fgts brought the Internet to the mainstream...
2012-10-20T05:29:08 < cjbaird> All the cybercrime, forum morons, government and legal intervention, etc., were all foreseen... "But NOOO! We has to make big bux by selling Internets to average joe mouthbreather!1"
2012-10-20T05:29:11 < zippe> note to self: make sure partner is cool with guro before opening the front door
2012-10-20T05:41:54 < dongs> lol guro
2012-10-20T05:42:08 < dongs> youre now entering laurenceb territory
2012-10-20T05:43:29 < cjbaird> dongy: that recently JP loli ban was just a Tokyo-area thing?
2012-10-20T05:43:50 < dongs> not sure, i dont pay attention to that sorta shit usually. links?
2012-10-20T05:49:56 < dongs> woot
2012-10-20T05:49:58 < dongs> 9221 is here
2012-10-20T05:52:07 < cjbaird> Hmm. A total mess of links while trying to search on it-- seems there's been several attempts to pass anti Loli and Hentai bills in Tokyo.
2012-10-20T05:57:31 < zyp> first time I was in japan, one of the guys that travelled with me tried to buy some loli-dvd
2012-10-20T05:59:47 < zyp> he was quite disappointed afterwards when he tried watching it and realized that it weren't about girls but boys
2012-10-20T06:00:04 < dongs> ugh. got sop 3:8 decoders, not tssop
2012-10-20T06:00:13 < zyp> we gave him a lot of crap for that for the rest of the trip :p
2012-10-20T06:00:27 < dongs> lilo dvd is still pixelated
2012-10-20T06:00:29 < dongs> so its worthless.
2012-10-20T06:01:08 < zyp> his defense was something along the lines of «they didn't show the dicks on the dvd cover»
2012-10-20T06:02:52 < zyp> if memory serves me correctly it was one of these: http://en.wikipedia.org/wiki/Boku_no_Pico
2012-10-20T06:04:13 < cjbaird> HAHA!
2012-10-20T06:04:47 < dongs> leave it to wikipedo to document this sorta shit
2012-10-20T06:05:39 < cjbaird> There was a Kenny Everett sketch about a seedy Soho dealer in a trenchcoat approuching a guy, and offering to sell him an unlabelled video.. about Nuns.. and children.. and goats...
2012-10-20T06:06:03 < cjbaird> ..the guy unwraps the tape after he's bought it, and it's The Sound of Music.
2012-10-20T06:07:05 < zyp> could be worse, could be what he actually thought he was getting
2012-10-20T06:08:02 < cjbaird> This is it: https://www.youtube.com/watch?v=x4J7B0VL7PU
2012-10-20T06:08:04 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-20T06:08:12 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-20T06:09:22 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-20T06:15:05 < dongs> well onsemi samples are worthless now
2012-10-20T06:15:09 < dongs> $21 shipping
2012-10-20T06:15:22 < dongs> time to call local place
2012-10-20T06:15:25 < emeb_mac> *cough*
2012-10-20T06:16:06 < emeb_mac> that's some heavy postage. wonder why it's so high?
2012-10-20T06:16:16 < dongs> and they dont take paypal either
2012-10-20T06:17:19 < emeb_mac> I guess that doesn't surprise me.
2012-10-20T06:17:29 < dongs> freescale takes paypal
2012-10-20T06:17:32 < dongs> they're fucking awesome.
2012-10-20T06:17:39 < emeb_mac> no kidding?
2012-10-20T06:17:42 < dongs> yep
2012-10-20T06:17:45 < dongs> for real orders too
2012-10-20T06:17:51 < dongs> i bought a tray of those mmasomeshit accels from them wiht paypal
2012-10-20T06:17:57 < emeb_mac> how... forward thinking...
2012-10-20T06:19:09 < dongs> i wonder how i ended up with sop 3:8's ordered and tssop on pcb
2012-10-20T06:19:49 < HTT-Bird> dongs: ding-dong. :p although don't feel too bad
2012-10-20T06:19:56 < HTT-Bird> footprint mismatches happen to the best of us
2012-10-20T06:20:09 < zyp> your sop's doesn't have enough ts
2012-10-20T06:20:15 < dongs> wait no
2012-10-20T06:20:16 < dongs> wth
2012-10-20T06:20:28 < dongs> thats not 3:8 thats the quad buffer
2012-10-20T06:20:37 < dongs> ok. im good.
2012-10-20T06:20:40 < emeb_mac> at least you can use sop on something. I've got a batch of QFN codecs I can't use.
2012-10-20T06:20:43 < dongs> no
2012-10-20T06:20:55 < dongs> all is fine, the sop shits were quad buffers for the laurenceb logger clone
2012-10-20T06:21:03 < dongs> i ordered 3:8s in tssop long time ago
2012-10-20T06:21:08 < zyp> emeb_mac, because you can't do qfn or because you aren't doing another board with codecs?
2012-10-20T06:21:52 < emeb_mac> zyp: haven't ever tried QFN. I hand solder most everything, so I don't think it'd work.
2012-10-20T06:22:16 < zyp> I've heard that hand soldering QFN is possible
2012-10-20T06:23:03 < dongs> ok, so falsealarm
2012-10-20T06:23:04 < zyp> just make sure the pads extend enough that you can heat and apply solder to them with the chip in place
2012-10-20T06:23:16 < dongs> when pcbs arrive in the morning i have all the parts for led of doom
2012-10-20T06:23:43 < emeb_mac> i've had fairly good luck with those leadless ceramic oscillator packages. Kinda like QFN
2012-10-20T06:24:18 < HTT-Bird> emeb_mac: you might be able to do it with an iron, although possibly best bet is hot air/skillet/toaster oven reflow
2012-10-20T06:25:01 < emeb_mac> HTT-Bird: yep - I've done a bit of reflow. Mainly for stuff with exposed pads underneath
2012-10-20T06:25:14 < zyp> after assembling some boards with stencil and reflow oven I wouldn't want to go back to hand soldering them
2012-10-20T06:26:49 < zyp> the main time consuming part with that is finding the components and getting them out of whatever they are packed in
2012-10-20T06:28:32 < emeb_mac> what do you do for paste?
2012-10-20T06:29:04 < zyp> what do you mean?
2012-10-20T06:29:08 < emeb_mac> I've heard that paste "spoils"
2012-10-20T06:29:24 < emeb_mac> use-by dates. has to be kept refrigerated, etc.
2012-10-20T06:30:29 < zyp> I got some small jars of cheap shit from dx, works nicely
2012-10-20T06:30:50 < emeb_mac> heh - I bought some of those little jars from dx a while back too.
2012-10-20T06:31:00 < zyp> I keep them in my refrigerator, so far I've only used the same jar for the boards I've done at home so far
2012-10-20T06:31:12 < zyp> probably the same then
2012-10-20T06:31:40 < dongs> http://i.imgur.com/quSJG.jpg
2012-10-20T06:31:42 < zyp> even if the jars are small, they last enough for a bunch of boards
2012-10-20T06:31:50 < dongs> my9221te + some 6pin female headers
2012-10-20T06:32:03 < zyp> and they are cheap enough that I don't mind throwing them away before they are empty either
2012-10-20T06:32:19 < HTT-Bird> zyp: dx is useless, the site fubared my registration when I first tried to use it IIRC
2012-10-20T06:32:29 < HTT-Bird> besides: they got bapped on the head by the FCC not too long ago
2012-10-20T06:32:37 < dongs> that DX paste is utterly worthless
2012-10-20T06:32:40 < dongs> its dry before it even gets to you
2012-10-20T06:33:12 < dongs> http://www.zeph.com/zephpaste.htm this is the shit i use
2012-10-20T06:33:17 < dongs> leadfree vrsion of course
2012-10-20T06:33:55 < zyp> the one jar I've been using so far has been ok
2012-10-20T06:34:17 < zyp> at least just as good as the expensive shit we used for the uni project
2012-10-20T06:36:03 < dongs> i support america by paypalling that shit
2012-10-20T06:36:36 < emeb_mac> zyp: what do you do for stencils?
2012-10-20T06:37:25 < zyp> so far I've been using kapton stencils from that ohara guy
2012-10-20T06:37:35 < zyp> or whatever he's called
2012-10-20T06:37:36 < dongs> horrible
2012-10-20T06:37:39 -!- baird_110baud [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-20T06:37:48 < dongs> i bought those, i didnt even bother putting them on pcb
2012-10-20T06:38:07 < dongs> just straight to the bin
2012-10-20T06:39:15 < zyp> well, I just picked the cheapest choice, has worked for me so far :p
2012-10-20T06:39:25 < emeb_mac> http://ohararp.com/Stencils.html <- these?
2012-10-20T06:39:28 < zyp> yep
2012-10-20T06:40:14 < zyp> but I will certainly open to try other stuff in the same price range
2012-10-20T06:40:41 < dongs> metal stencils here run around 60bucks and i can panel a bunch of designs on one.
2012-10-20T06:40:49 < dongs> then they go in my stencil printer that stretches the metal
2012-10-20T06:41:06 < dongs> totalyl worth it considering its not shit that'll fall apart
2012-10-20T06:41:52 < zyp> ohara is 25 for a sheet and will also panel designs
2012-10-20T06:42:29 < zyp> but then I'm not mass producing like you are
2012-10-20T06:42:30 < dongs> and how many tiems you can reuse it wiht a metal paste spreader?
2012-10-20T06:42:33 < dongs> im gonna guess: 0
2012-10-20T06:42:56 < emeb_mac> paste spreader was the next question :)
2012-10-20T06:43:10 < dongs> i got this good shit from america
2012-10-20T06:43:39 < dongs> http://www.krafttool.com/catalog.aspx?cat=91&subcat=111
2012-10-20T06:43:46 < dongs> DW533
2012-10-20T06:44:09 < zyp> I use a putty knife or wtf it's called in english, pretty much the same shit and available in every hardware store
2012-10-20T06:44:16 < dongs> yea
2012-10-20T06:44:29 < emeb_mac> I was going to say - looks like something from the hardware store.
2012-10-20T06:44:30 < dongs> excepte japan
2012-10-20T06:44:46 < emeb_mac> you don't do drywall in JP?
2012-10-20T06:45:07 < zyp> and I assembled ten of those uni boards with a single kapton stencil, it held up fine
2012-10-20T06:45:36 < zyp> and I don't really have any plans to assemble larger series than ten or so myself anyway
2012-10-20T06:50:20 < zyp> by the way, I was playing with the idea of getting one of those cheap paste dispensers and mounting it some cnc shit to apply paste directly to the board in an automated fashion
2012-10-20T06:50:31 < emeb_mac> most of the time I build 2 or 3 of anything, so it's not that big a deal.
2012-10-20T06:50:49 < zyp> should eliminate the need for a stencil
2012-10-20T06:50:49 < emeb_mac> I did build ~30 of one small board a few years ago though.
2012-10-20T06:51:08 < emeb_mac> that would be pretty cool.
2012-10-20T06:51:28 < emeb_mac> just convert the stencil gerber to g-code and off you go.
2012-10-20T06:51:57 < zyp> yeah, something like that
2012-10-20T06:52:28 < emeb_mac> trick would be figuring out how to regulate the paste applicator.
2012-10-20T06:52:44 < zyp> it would not be as fast as stencils for mass production, but for prototypes it removes the entire need for the stencil
2012-10-20T06:52:57 < dongs> zyp: doesnt work
2012-10-20T06:53:08 < dongs> zyp: to dispense proper amount you need auger-style head
2012-10-20T06:53:12 < dongs> that shit is expensive for paste.
2012-10-20T06:53:23 < dongs> pressure doesnt work
2012-10-20T06:53:28 < dongs> you cant regulate it continously properly
2012-10-20T06:53:49 < dongs> https://youtube.com/watch?v=tPjFQ15Zdp0 shit like this
2012-10-20T06:54:20 < dongs> you cant do same kind of repeated results with pressure-based head
2012-10-20T06:54:33 < zyp> ah, makes sense
2012-10-20T06:55:09 < dongs> my dicknplace originally came wiht pressure dispenser
2012-10-20T06:55:11 < dongs> totally worthless
2012-10-20T06:55:13 < zyp> I guess the pressure based shit is just meant for larger pads where less precision is needed then?
2012-10-20T06:55:15 < dongs> dismounted it after a week and sent it back
2012-10-20T06:55:20 < dongs> yeah
2012-10-20T06:55:26 < dongs> but it was useless except for like huge shit
2012-10-20T06:55:31 < dongs> 0.5mm? forget it
2012-10-20T06:55:51 < zyp> right, I pretty much suspected that it wouldn't be precise enough
2012-10-20T06:56:06 < zyp> or rather, expensive to get precise enough
2012-10-20T06:56:34 < dongs> plus that kinda shit is still wasteful for one-offs
2012-10-20T06:56:38 < dongs> because you need to clean the head
2012-10-20T06:56:42 < dongs> and thus waste all that paste
2012-10-20T06:56:48 < dongs> so unless its continuously running al lday
2012-10-20T06:56:54 < dongs> and only cleaned once, then it doesnt make sense either
2012-10-20T06:57:29 < zyp> makes sense
2012-10-20T06:57:42 < dongs> the shit I got had a preheater under the nozzle
2012-10-20T06:57:52 < dongs> to make paste sorta liquidish,, kept it at liek 34C or something
2012-10-20T06:57:59 < dongs> that was supposed to help wiht consistent dispensing
2012-10-20T06:58:09 < dongs> of course, it didnt do fuckall for any kind repeat accuracy
2012-10-20T06:58:22 < dongs> as the paste amount decreased in the tube, the air pressure would need to be regulated differently
2012-10-20T06:58:26 < dongs> and it wasnt
2012-10-20T06:59:59 < emeb_mac> wonder if it could work w/ a syringe piston driven by a helical stepper?
2012-10-20T07:00:25 < dongs> well, auger is sorta like that but it pushes the paste out by rotation instead of applying pressure to paste
2012-10-20T07:00:38 < dongs> i dunno if that makes difference as paste volume decreases
2012-10-20T07:00:40 < emeb_mac> sounds really fun to keep clean.
2012-10-20T07:00:46 < dongs> dont you ened to apply less pressure?
2012-10-20T07:00:59 -!- baird_110baud is now known as baird_300baud
2012-10-20T07:01:02 < dongs> yeah, im sure it is, and im sure thats part of the reason teh heads run like $3k+/ea
2012-10-20T07:01:15 < emeb_mac> maybe - viscosity * volume?
2012-10-20T07:01:24 < dongs> shit probably needs to be machined very tight tolerances to make sure paste balls dont get squeezed between teh helix
2012-10-20T07:01:39 < emeb_mac> hence the cost
2012-10-20T07:02:11 < emeb_mac> think I'll stick to hand soldering. :)
2012-10-20T07:02:15 < dongs> lo
2012-10-20T07:02:23 < dongs> im gonna handsolder ledtrash tomorrow :(
2012-10-20T07:02:32 < emeb_mac> nice
2012-10-20T07:02:46 < dongs> but ill get stencil for it eventually once theres enough designs to panel
2012-10-20T07:02:51 < dongs> not doing more than one as proto
2012-10-20T07:03:02 < emeb_mac> makes sense
2012-10-20T07:03:15 < dongs> and once it uh, works
2012-10-20T07:03:16 < dongs> heh
2012-10-20T07:03:24 < dongs> though I dont expect too many problems
2012-10-20T07:03:35 < emeb_mac> yep - if you don't need to spin the board.
2012-10-20T07:06:16 < zyp> you mean flip it?
2012-10-20T07:06:52 < zyp> should be no need for that, matrix is symmetric with regards to connectors so worst case you just have to reorder the outputs in software
2012-10-20T07:07:23 < dongs> no, he meant like respin as in redo layout/whatever because of some shitty / dumb bug
2012-10-20T07:07:38 < dongs> im pretty sure 99% that led layout is correct
2012-10-20T07:07:48 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-20T07:09:57 < zyp> wtf, clock just passed 6
2012-10-20T07:10:00 < zyp> in the morning
2012-10-20T07:10:06 < dongs> why are you awake
2012-10-20T07:10:14  * emeb_mac was wondering
2012-10-20T07:10:15 < zyp> yes, why the fuck am I still awake?
2012-10-20T07:10:39 < zyp> I didn't pay any attention to the time
2012-10-20T07:11:45 < emeb_mac> too much caffeine late in the day.
2012-10-20T07:12:23 < zyp> I don't drink coffee
2012-10-20T07:12:48 < emeb_mac> red bull? monster? tea?
2012-10-20T07:13:39 < dongs> me neither
2012-10-20T07:14:33  * emeb_mac has a big bag of caffeinated penguin mints.
2012-10-20T07:15:58 < baird_300baud> The Adrenalin Rush of focussed coding late at night...
2012-10-20T07:16:18 < dongs> heh
2012-10-20T07:16:21 < zyp> I haven't been coding tonight
2012-10-20T07:16:28 < zyp> haven't been doing anything useful really
2012-10-20T07:16:36 < baird_300baud> Duh. You've been on IRC.
2012-10-20T07:16:42 < zyp> exactly :p
2012-10-20T07:16:43 < dongs> lo
2012-10-20T07:16:44 < emeb_mac> internet pr0n
2012-10-20T07:16:50 < dongs> fappy
2012-10-20T07:17:46 < emeb_mac> isn't that some sort of phone?
2012-10-20T07:17:47 < baird_300baud> My Mum handed back a ~300g bag of chocolate-coasted coffee beans that I'd given her about 3 years ago. I ate the lot in one sitting-- probably ingesting the equivalent of 10 cups of coffe. :)
2012-10-20T07:18:16 < emeb_mac> buzzin'
2012-10-20T07:18:33 < zyp> emeb_mac, with fax, apparently
2012-10-20T07:18:47 < baird_300baud> No problems on the motorbike ride home..
2012-10-20T07:19:06 < emeb_mac> except speeding ticket.
2012-10-20T07:19:41 < baird_300baud> Not that I know of-- the collection group here like to wait 2-3 months before mailing out the fines.
2012-10-20T07:20:26 < zyp> nah, I'll head to bed now before it becomes even later, have a good whatever it is in your timezones
2012-10-20T07:21:25 < baird_300baud> 37degC here today. Can't be bothered doing anything productive like I was supposed to (buying topo maps..)
2012-10-20T07:21:27 < emeb_mac> sleep tight.
2012-10-20T07:28:44 -!- cjbaird is now known as baird
2012-10-20T07:58:37 < emeb_mac> sounds ominous
2012-10-20T08:03:04 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-20T08:03:04 -!- NonaSuomy [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-20T08:03:04 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-20T08:06:04 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 244 seconds]
2012-10-20T08:56:00 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 272 seconds]
2012-10-20T08:58:59 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 255 seconds]
2012-10-20T09:21:00 -!- dekar_ [~dekar@212.255.27.61] has joined ##stm32
2012-10-20T09:21:03 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-20T09:22:36 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-20T09:24:15 -!- dekar [~dekar@212.255.39.10] has quit [Ping timeout: 260 seconds]
2012-10-20T09:53:10 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 245 seconds]
2012-10-20T10:30:24 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-20T11:34:40 -!- ABL [~ABLomas@78-60-198-200.static.zebra.lt] has quit [Ping timeout: 246 seconds]
2012-10-20T11:50:56 < dongs> commence chats
2012-10-20T11:52:18 -!- K1AFR [~fsdf@p3EE2E8F2.dip.t-dialin.net] has joined ##stm32
2012-10-20T12:05:04 < baird_300baud> nu
2012-10-20T12:07:21 -!- K1AFR [~fsdf@p3EE2E8F2.dip.t-dialin.net] has quit [Quit: Verlassend]
2012-10-20T12:17:09 -!- baird_300baud [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Leaving]
2012-10-20T12:28:31 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-20T12:28:34 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-20T12:50:03 -!- barthess [~barthess@178.154.30.238] has joined ##stm32
2012-10-20T12:55:57 -!- ABL [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-20T12:56:27 -!- ABL is now known as ABLomas
2012-10-20T13:08:58 < dongs> no u
2012-10-20T13:11:16 <+Steffanx> Not today
2012-10-20T13:13:29 < dongs> 5. To prevent receiver damage, please turn off
2012-10-20T13:13:29 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-20T13:13:30 < dongs> power during electrical storms.
2012-10-20T13:14:06 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-20T13:14:46 -!- daku is now known as DaKu
2012-10-20T13:18:08 -!- TzacksT [~zz@089-101-122023.ntlworld.ie] has quit [Ping timeout: 256 seconds]
2012-10-20T13:32:39 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-20T13:49:38 -!- barthess [~barthess@178.154.30.238] has quit [Quit: Leaving.]
2012-10-20T13:56:18 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-20T13:56:22 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-20T13:59:22 -!- barthess [~barthess@164.177.194.64] has joined ##stm32
2012-10-20T14:26:19 -!- |sark [~sark@109.235.51.202] has joined ##stm32
2012-10-20T14:36:26 < dongs> hello blogs
2012-10-20T14:37:53 < jpa-> hello nuts
2012-10-20T14:39:54 < baird> Ask a man who has located every public toilet in Australia anything!
2012-10-20T14:40:05 < baird> http://imgur.com/uqFTG
2012-10-20T14:40:30 < jpa-> did you put them on openstreetmap?
2012-10-20T14:40:59 < jpa-> and, more importantly, did you take a dump in each?
2012-10-20T14:41:14 < baird> Mostly likely is already on there-- I ripped the dataset from data.gov.au
2012-10-20T14:42:06 < jpa-> lazy
2012-10-20T14:45:56 < dongs> http://1.standaardcdn.be/Assets/Images_Upload/2012/10/15/site_1610_femen.jpg attn Laurenceb
2012-10-20T14:46:39 < baird> No nude Nun? Boo.
2012-10-20T14:46:53 < jpa-> they are all nuns, just one is not naked
2012-10-20T14:53:50 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-20T14:56:02 < Laurenceb_> sup
2012-10-20T14:59:15 < jpa-> per
2012-10-20T15:01:44 < dongs> 20:45 < dongs> http://1.standaardcdn.be/Assets/Images_Upload/2012/10/15/site_1610_femen.jpg attn Laurenceb
2012-10-20T15:04:29 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has left ##stm32 []
2012-10-20T15:04:54 < Laurenceb_> what the fuuuu
2012-10-20T15:05:26 < Laurenceb_> rapists are killers and they want to be raped
2012-10-20T15:05:31 < Laurenceb_> some basically they are emo
2012-10-20T15:06:27  * Laurenceb_ -> get a life
2012-10-20T15:06:30 < Laurenceb_> bbl
2012-10-20T15:09:35 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-20T15:10:30 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-10-20T15:11:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-20T15:20:19 < dongs> bored
2012-10-20T15:20:40 <+Steffanx> There are enough websites to unbore you
2012-10-20T15:20:47 < dongs> unlikely
2012-10-20T15:22:34 <+Steffanx> Ask Laurenceb_
2012-10-20T15:46:01 < zyp> *yawn*
2012-10-20T15:48:35 <+Steffanx> Good morning
2012-10-20T15:51:28 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-20T16:00:31 -!- claude is now known as Claude
2012-10-20T16:14:04 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-20T16:20:48 -!- Devilholk [~banned@h-201-147.a357.priv.bahnhof.se] has joined ##stm32
2012-10-20T16:21:30 < Devilholk> Hi, I'm reading about how to interface quadrature encoders, docs refer to Timer Inputs (like TI3 for instance). Do they by that mean the timers channel 3 input? Like, TI3 applied to TIM2 would be TIM2_CH3? I am all confused >.<
2012-10-20T16:22:24 < zyp> which family are you using?
2012-10-20T16:22:58 < Devilholk> This is for STM32F100
2012-10-20T16:23:37 < Devilholk> I am reading in the reference manual (RM0008)
2012-10-20T16:23:44 < zyp> then I suggest having a look at page 284 in the reference manual (RM0008 rev 14)
2012-10-20T16:23:47 < jpa-> Devilholk: yeah, like that
2012-10-20T16:24:25 < Devilholk> Why do I have rev 9 >.<
2012-10-20T16:24:46 < zyp> :p
2012-10-20T16:25:00 < zyp> just included revision number because page numbers doesn't match between them
2012-10-20T16:25:43 < zyp> anyway, I'm referring to the block diagram which is in the start of each timer chapter
2012-10-20T16:25:51 < zyp> third page of the chapter or so
2012-10-20T16:26:38 < Devilholk> That would be in chapter General-purpose timer (TIMx) ?
2012-10-20T16:27:22 < zyp> I guess so
2012-10-20T16:27:48 < zyp> my revision have two General-purpose timer chapters, both have one
2012-10-20T16:28:33 < Devilholk> Oh, there I see it, thanks
2012-10-20T16:28:47 < Devilholk> I was looking in all the wrong places I guess, at remapping tables and stuff
2012-10-20T16:28:55 < zyp> the gist of it is that TIx is usually corresponding to CHx
2012-10-20T16:29:12 < zyp> except for TI1 which can be switched to CH1^CH2^CH3
2012-10-20T16:30:04 < zyp> and then both IC1 and IC2 can be either TI1 or TI2 and so on
2012-10-20T16:31:18 < Devilholk> Thanks a bunch
2012-10-20T16:31:35 < zyp> no problem
2012-10-20T16:58:10 -!- dekar_ [~dekar@212.255.27.61] has quit [Quit: Leaving]
2012-10-20T17:06:35 < dongs> http://www.ebay.com/itm/300798864416 [ buy it now ]
2012-10-20T17:07:06 -!- barthess [~barthess@164.177.194.64] has quit [Read error: Connection reset by peer]
2012-10-20T17:11:07 <+Steffanx> WTF WTF WTF dong
2012-10-20T17:11:07 <+Steffanx> s
2012-10-20T17:15:39 < dongs> if you think that's wtf... try THIS: https://youtube.com/watch?v=Pd2pSLFjOas
2012-10-20T17:16:19 <+Steffanx> dongs opens his box with WTFs
2012-10-20T17:41:17 -!- barthess [~barthess@5.100.197.133] has joined ##stm32
2012-10-20T18:00:28 < zyp> oh fuck
2012-10-20T18:00:39 < zyp> I've hooked some of the sdram signals to wrong pins
2012-10-20T18:04:10 <+Steffanx> Poor uou
2012-10-20T18:04:11 <+Steffanx> you
2012-10-20T18:04:37 < zyp> better spot it now than after submitting the board :p
2012-10-20T18:08:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-20T18:14:23 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-20T18:25:02 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has quit [Remote host closed the connection]
2012-10-20T18:25:20 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has joined ##stm32
2012-10-20T18:26:27 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-20T18:34:42 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-20T18:37:29 -!- elektrinis [~circuit@78.60.169.125] has joined ##stm32
2012-10-20T18:53:54 < zyp> huh
2012-10-20T18:53:56 < zyp> «The core frequency must be 120 MHz or lower to use the SWO function.»
2012-10-20T18:54:04 < zyp> (lpc4300 reference manual)
2012-10-20T18:54:59 < zyp> «Remark: The SWD mode is supported for the ARM Cortex-M4 only.»
2012-10-20T18:55:31 < zyp> that's a bit weird
2012-10-20T18:55:43 < jpa-> it sounds weird
2012-10-20T18:56:06 < dongs> wut
2012-10-20T18:56:11 < dongs> so how do oyu debug the m0 part?
2012-10-20T18:56:15 < zyp> jtag
2012-10-20T18:56:17 < dongs> o
2012-10-20T18:56:24 < dongs> oh, then it kinda makes sense
2012-10-20T18:56:33 < dongs> can swd support several devices like jtag can?
2012-10-20T18:56:38 < zyp> yes
2012-10-20T18:56:47 < dongs> how?
2012-10-20T18:56:54 < dongs> you specify number of taps to ignore or whatever?
2012-10-20T18:57:08 < dongs> to get to next device in chain
2012-10-20T18:57:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-20T18:57:24 < zyp> remember that SWDIO/SWDCLK is TMS/TCK, so they are connected in parallel to all
2012-10-20T18:57:29 < zyp> so there is no chain
2012-10-20T18:57:46 < zyp> think sort of like i2c
2012-10-20T18:59:29 < zyp> http://www.arm.com/about/newsroom/25872.php <- but multi-drop swd is newer than the swd spec itself
2012-10-20T18:59:42 < zyp> so I suspect lpc4300 is using swd cores without multi-drop support
2012-10-20T19:03:09 < dongs> https://youtube.com/watch?v=KS3IS1WWpPU furry style
2012-10-20T19:04:03 < dongs> attn cjbeard/etc
2012-10-20T19:05:47 < Bird|lappy> so...how do I tell if my board layout is too small from an assembly standpoint?
2012-10-20T19:08:21 < dongs> question does not compute
2012-10-20T19:08:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-20T19:09:06 < Thorn> if you pwm led lighting, what frequency can be considered safe for eyesight?
2012-10-20T19:09:28 < karlp> there's a frequency that's not safe?
2012-10-20T19:09:58 < jpa-> i'd go over 1000Hz
2012-10-20T19:10:07 < jpa-> otherwise the flicker is visible when you turn your head
2012-10-20T19:11:40 < Thorn> also photo/video cameras hate low frequencies if I'm not mistaken
2012-10-20T19:11:59 < jpa-> yeah, but for those problem is mostly < 200 Hz
2012-10-20T19:13:20 < Thorn> also, wth is TIM_OC[1-4]PreloadConfig() and TIM_ARRPreloadConfig()?
2012-10-20T19:15:15 < jpa-> preloading means that the new register values don't apply immediately, but only when the timer rolls over the next time
2012-10-20T19:15:17 < Thorn> the manual says you must enable preload registers in pwm mode but it works for me without that
2012-10-20T19:15:30 < jpa-> in that way you can modify the values on the fly without causing strange spikes
2012-10-20T19:15:46 < Thorn> aha I see
2012-10-20T19:16:17 < Thorn> I should scope the waveforms I have now, they probably look pretty weird
2012-10-20T19:16:20 < Thorn> thanks
2012-10-20T19:37:32 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-eptzipaykxztrihb] has quit [Ping timeout: 255 seconds]
2012-10-20T19:38:10 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-wvvpmxdkmnmqirvg] has joined ##stm32
2012-10-20T19:50:37 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/]
2012-10-20T19:54:07 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-20T20:20:54 < zyp> http://bin.jvnv.net/f/U88MX.png
2012-10-20T20:21:34 < zyp> sdram, hs usb, jtag/swd, sdio, rmii, i2s in/out
2012-10-20T20:22:49 < zyp> wonder what more I should put on it
2012-10-20T20:23:32 < jpa-> leds
2012-10-20T20:24:31 < zyp> yeah, am planning to
2012-10-20T20:24:39 < zyp> also some gpio/sgpio
2012-10-20T20:24:52 < zyp> and probably boot selection pins
2012-10-20T20:34:53 < karlp> heh, just replacing ; in while(1) with __WFI(); and current consumption drops from 11mA to 3.5mA.
2012-10-20T20:35:28 < zyp> sounds expected, no?
2012-10-20T20:35:36 < karlp> yeah, just hadn't actually measured it.
2012-10-20T20:35:41 < karlp> it's nice that it was that easy.
2012-10-20T20:35:59 < zyp> it's only that easy when you are actually waiting for an interrupt :p
2012-10-20T20:36:08 < karlp> well, yes, I did write it behave like that first :)
2012-10-20T20:36:30 < karlp> I'm finally happy with the RHT03 one wire decoding thingy.
2012-10-20T20:41:42 < jpa-> for some reason i didn't have such large saving when enabling WFI in chibios
2012-10-20T20:42:14 < jpa-> just some 15mA -> 13mA or similar.. but i guess it wasn't that much idle
2012-10-20T20:45:05 < zyp> my code doesn't have any concept of thread state yet
2012-10-20T20:45:09 < zyp> but I'm working on it
2012-10-20T20:45:31 < zyp> I mean, currently all threads are always ready to run
2012-10-20T20:45:39 < karlp> well, this is only active for ~25ms every 3 seconds, plus to print "tick" every second.
2012-10-20T20:47:41 < zyp> in singlethreaded applications I can of course use wfi and it works like expected because there is no background threads that would also be blocked
2012-10-20T20:48:20 -!- olasd [~olasd@pdpc/supporter/active/olasd] has quit [Ping timeout: 260 seconds]
2012-10-20T20:49:19 -!- olasd [~olasd@leibniz.olasd.eu] has joined ##stm32
2012-10-20T20:49:20 -!- olasd [~olasd@leibniz.olasd.eu] has quit [Changing host]
2012-10-20T20:49:20 -!- olasd [~olasd@pdpc/supporter/active/olasd] has joined ##stm32
2012-10-20T20:49:20 < zyp> but I'm working on a scheduler that would keep track of thread state and set SLEEPONEXIT when there are no ready threads
2012-10-20T20:50:59 < zyp> SLEEPONEXIT instructs the cpu to do the equivalent of a wfi when the ISR returns
2012-10-20T20:52:22 < jpa-> you have no idle thread?
2012-10-20T20:53:06 < zyp> what's the point of having an idle thread?
2012-10-20T20:55:02 < jpa-> not so much point, just that it appears to be a quite common practise
2012-10-20T20:55:38 < jpa-> i wonder if you can disable SLEEPONEXIT somehow so that it retains the same behaviour but does not mess up debugger.. i think there was some setting for that in DBGMCU
2012-10-20T20:56:23 < zyp> you mean don't disable clock to debugger interface while sleeping
2012-10-20T20:57:41 < jpa-> yeah
2012-10-20T20:57:49 < zyp> also, when I was testing, it didn't have any conflict with the debugger interface unless I set the SLEEPDEEP bit
2012-10-20T20:58:05 < jpa-> hmm, could be actually
2012-10-20T20:58:13 < zyp> so that part shouldn't be a problem
2012-10-20T21:13:46 -!- karlp [~karl@palmtree.beeroclock.net] has quit [Ping timeout: 245 seconds]
2012-10-20T21:21:38 < upgrdman> spi q: even if the slave is not communicating, the MOSI line should still work, correct? my sclk line works, but MOSI and MISO always stay low
2012-10-20T21:22:04 < zyp> yes, MOSI is driven by host
2012-10-20T21:22:10 < upgrdman> ok
2012-10-20T21:23:07 < upgrdman> weird thing is i loaded the f4discov demo firmware, which uses the accelerometer, and compared its GPIOA and SPI1 registers, and they are identical to my non-working firmware :(
2012-10-20T21:23:33 < upgrdman> and i have the gpioa and spi1 clocks enabled
2012-10-20T21:23:55 < zyp> can I see your code?
2012-10-20T21:24:44 -!- mase [~mase@kemp-5d8609af.pool.mediaWays.net] has joined ##stm32
2012-10-20T21:25:23 < upgrdman> sure, hold on
2012-10-20T21:25:42 < upgrdman> im just using reg def's.... not the stdperiph lib
2012-10-20T21:27:21 < mase> hello, has somebody got CAN working on an stm32f4? i seem to have problems with the bit-timing. I'm googling for two days now
2012-10-20T21:28:43 < upgrdman> zyp, http://pastebin.com/93zUz4L1
2012-10-20T21:29:14 < upgrdman> its a mess, sorry.
2012-10-20T21:29:15 < zyp> mase, yep, CAN works fine for me
2012-10-20T21:29:43 < zyp> mase, what's your problem?
2012-10-20T21:30:25 < mase> zyp cool, what values have you used for BS1/BS2/prescaler?
2012-10-20T21:31:39 < mase> i tried different values/bitrates. measurements with an oszi showed very different frame widths compared to the frames of an USB-Can Adapter
2012-10-20T21:32:47 < zyp> http://cgit.jvnv.net/laks/tree/can/can.h#n59
2012-10-20T21:34:26 < mase> ah, ok , i'm using the stm library
2012-10-20T21:36:33 < mase> for what bitrate are these values?
2012-10-20T21:36:38 < zyp> 1MHz
2012-10-20T21:36:56 < zyp> hang on, I'm working out how I did this
2012-10-20T21:40:05 < zyp> ah, yeah, BS2 is 2, BS1 is 3, that means that bittime is (1 + 2 + 3) * t_q
2012-10-20T21:40:33 < zyp> i.e. 6 * t_q
2012-10-20T21:41:16 < upgrdman> zyp, any ideas?
2012-10-20T21:41:18 < zyp> which means that frequency have to be six times the desired baudrate
2012-10-20T21:41:18 < mase> thank you very much, i'll give that a try...
2012-10-20T21:41:52 < zyp> so the prescaler have to be set to give busfreq / 6 / baudrate
2012-10-20T21:42:09 < zyp> which is where the (clk / 6 / 1000000 - 1) part in my code comes in
2012-10-20T21:42:50 < zyp> upgrdman, sorry, I have to go pick up a pizza, back in some minutes
2012-10-20T21:42:56 < zyp> will look at it when I'm back
2012-10-20T21:43:02 < upgrdman> ok thanks
2012-10-20T21:44:00 < mase> so "1000000" is the 1Mhz-speed of the controller? my knowledge about microcontrollers is not very excelent yet
2012-10-20T21:55:30 -!- Guest28776 [~karl@palmtree.beeroclock.net] has joined ##stm32
2012-10-20T21:55:44 -!- Guest28776 is now known as karlp
2012-10-20T21:56:18 < mase> with BS1=3, BS2=2; prescale=6 i can measure a frame-width of 112us
2012-10-20T21:56:44 < zippe> zyp, shouldn't the −1 be outside the parens?
2012-10-20T21:58:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-20T21:58:49 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-20T22:00:39 -!- barthess [~barthess@5.100.197.133] has quit [Read error: Connection reset by peer]
2012-10-20T22:03:23 < zyp> zippe, why? division have higher precedence than subtraction
2012-10-20T22:04:08 < zyp> the parens are enclosing this expression from the other values getting or-ed together into the register
2012-10-20T22:04:19 < zyp> i.e. reg.BTR = 0x01120000 | (clk / 6 / 1000000 - 1);
2012-10-20T22:12:57 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-20T22:12:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Disconnected by services]
2012-10-20T22:14:41 -!- barthess [~barthess@5.100.200.146] has joined ##stm32
2012-10-20T22:20:38 < zippe> zyp: depending on precedence rules is bad style 8(
2012-10-20T22:22:23 < zippe> zyp: for example, | has lower precedence than -
2012-10-20T22:22:32 < zippe> So if you think you need the parens for |, you need them for - as well.
2012-10-20T22:23:11 < Laurenceb_> stm32f1 startup code should run on f3 right?
2012-10-20T22:23:51 < zippe> Depends what you mean by "startup
2012-10-20T22:23:56 < zyp> zippe, I was almost expecting you to reply with that :p
2012-10-20T22:24:29 < zippe> zyp: sorry to be so predictable. 8)  personally, I think that operator precedence is a bad idea outside of the unary operators
2012-10-20T22:24:42 < zippe> zyp: I prefer to be explicit about the order of evaluation
2012-10-20T22:26:28 < zyp> I don't know, I learned in elementary school that multiplication and division has a higher order of precedence than addition and subtraction, and I can't really imagine any large pitfalls around that
2012-10-20T22:26:55 < Laurenceb_> hmm
2012-10-20T22:27:08 < Laurenceb_> the bootloader runs ok with texane
2012-10-20T22:27:20 < Laurenceb_> so it will probably be ok?
2012-10-20T22:27:32 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-20T22:27:47 < zyp> but I prefer using parens to enclose an arithmetic subexpression in an otherwise bitwise expression, mostly to make it obvious when reading it
2012-10-20T22:28:09 < zyp> Laurenceb_, huh?
2012-10-20T22:28:33 < zyp> you mean flashloader? how is that at all related to startup code?
2012-10-20T22:28:45 < Laurenceb_> well it starts up the micro?
2012-10-20T22:28:55 < zyp> no it doesn't
2012-10-20T22:29:42 < zyp> startup code initializes stuff like data and bss segments that other code depends on, the flashloader is fully self-contained and has no such dependencies
2012-10-20T22:29:53 < Laurenceb_> hmm ok
2012-10-20T22:30:05 < Laurenceb_> well ill try it tomorrow
2012-10-20T22:30:14 < Laurenceb_> once i have some example code put together
2012-10-20T22:30:18 < zyp> but yeah, generic data/bss startup code will work fine
2012-10-20T22:30:30 < zyp> no differences there
2012-10-20T22:31:05 < zyp> if you have rcc stuff to enable PLL or whatever you might want to check that
2012-10-20T22:32:01 < Laurenceb_> but thats done later
2012-10-20T22:32:07 < Laurenceb_> using periph lib functions
2012-10-20T22:32:19 < zyp> then no changes needed
2012-10-20T22:32:22 < Laurenceb_> and theres a new periph lib for the f30x/f37x
2012-10-20T22:32:42 < Laurenceb_> just reading the opamp code..
2012-10-20T22:32:51 < upgrdman> zyp, any ideas?
2012-10-20T22:33:21 < zyp> I'm eating :p
2012-10-20T22:33:28 < upgrdman> o :)
2012-10-20T22:38:06 -!- GargantuaSauce_ [~sauce@blk-222-208-237.eastlink.ca] has quit [Remote host closed the connection]
2012-10-20T22:38:15 < Laurenceb_> epic lol
2012-10-20T22:38:18 < Laurenceb_> "Bit 1 FORCE_VP
2012-10-20T22:38:18 < Laurenceb_> This description will be given in a future version of this document.
2012-10-20T22:38:18 < Laurenceb_> "
2012-10-20T22:38:40 < Laurenceb_> it connects the "reference voltage" to the non inverting input of the opamp
2012-10-20T22:39:01 < Laurenceb_> a feature that isnt even shown in the diagram
2012-10-20T22:39:36 < Laurenceb_> luckily there is periph lib code :P
2012-10-20T22:40:54 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds]
2012-10-20T22:43:14 < zyp> upgrdman, can't see anything wrong
2012-10-20T22:43:30 < zyp> apart from the fact that you don't read/clear the rx buffer after the first write
2012-10-20T22:43:49 < zyp> but I can't see how that would cause clk to be output but not data
2012-10-20T22:44:48 < upgrdman> thats what rm0090 said to do. i will try reading after the first byte.
2012-10-20T22:44:59 < upgrdman> thanks for lookin
2012-10-20T22:45:24 < Laurenceb_> hmm hopefully ill have the opamp characterised soon
2012-10-20T22:45:32 < Laurenceb_> want to see how much noise
2012-10-20T22:46:22 < zyp> upgrdman, also, why pulldown?
2012-10-20T22:46:40 < zyp> SPI is driven in a push-pull fashion, no pullups or pulldowns should be required
2012-10-20T22:46:59 < zyp> oh, wait
2012-10-20T22:47:07 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-219.pptp.stw-bonn.de] has joined ##stm32
2012-10-20T22:47:18 -!- NonaSuomy is now known as TeknoJuce
2012-10-20T22:47:27 < zyp> GPIO_MODER_MODER5_1 <- is that (1 << 5*2) ?
2012-10-20T22:47:30 < upgrdman> i know, but thats what the working stm demo had in its PUPDR register
2012-10-20T22:47:41 < zyp> 1 is GPIO, AF is 3
2012-10-20T22:47:53 < zyp> no, sorry, AF is 2
2012-10-20T22:48:34 < upgrdman> #define GPIO_MODER_MODER5_1                  ((uint32_t)0x00000800)
2012-10-20T22:48:39 < upgrdman> bit 1
2012-10-20T22:48:48 < zyp> oh
2012-10-20T22:49:00 < zyp> wtf is up with those define names
2012-10-20T22:49:11 < upgrdman> so moder = 10 for pin5
2012-10-20T22:49:26 < upgrdman> ya...stm code is weird
2012-10-20T22:50:33 -!- |sark [~sark@109.235.51.202] has quit [Ping timeout: 256 seconds]
2012-10-20T22:50:44 < zyp> also, your fashion of setting that assumes that all bits are 0 in the first place
2012-10-20T22:50:53 < zyp> since you are not masking out original contents
2012-10-20T22:51:03 < upgrdman> this is fresh after POR
2012-10-20T22:51:17 < upgrdman> i also tried spi reset
2012-10-20T22:51:23 < upgrdman> no changes
2012-10-20T22:51:42 < zyp> did you try reading those registers from the debugger?
2012-10-20T22:51:42 < upgrdman> and gpio reset
2012-10-20T22:51:49 < upgrdman> yes gdb
2012-10-20T22:52:54 < zyp> pastebin the output of «p/x *GPIOA», will you?
2012-10-20T22:53:04 < zyp> and also p/x *SPI1
2012-10-20T22:53:14 < upgrdman> wtf is p/x?
2012-10-20T22:53:19 < zyp> print hex
2012-10-20T22:53:42 < upgrdman> k
2012-10-20T22:53:44 < zyp> and if you don't have it already, do «set print pretty on» first
2012-10-20T22:53:58 < upgrdman> hold on while i gather it all
2012-10-20T22:54:20 < zyp> I'll grab another slice of pizza meanwhile :p
2012-10-20T22:55:45 < upgrdman> (gdb) p/x *(unsigned long*)0x40020000
2012-10-20T22:55:45 < upgrdman> $2 = 0xa800a800
2012-10-20T22:55:45 < upgrdman> (gdb) p/t *(unsigned long*)0x40020000
2012-10-20T22:55:45 < upgrdman> $3 = 10101000000000001010100000000000
2012-10-20T22:56:03 < upgrdman> thats GPIOA + 0x00
2012-10-20T22:56:22 < upgrdman> hum, i need to this this for each reg :) hold on
2012-10-20T22:56:26 < zyp> wait
2012-10-20T22:56:35 < zyp> use the name GPIOA
2012-10-20T22:56:51 < zyp> and you will get the entire structure with named fields
2012-10-20T22:56:56 < jpa-> the name GPIOA doesn't work if it doesn't happen to be in scope or if -gdwarf-2 is not enabled
2012-10-20T22:57:25 < zyp> it is in scope, and I assume that every sane person compiles with debugging information enabled :p
2012-10-20T22:57:34 < jpa-> :P
2012-10-20T22:58:19 < upgrdman> (gdb) p/t *GPIOA
2012-10-20T22:58:19 < upgrdman> No symbol "GPIOA" in current context.
2012-10-20T22:58:32 < upgrdman> GPIOA is a #define... no symbol in the ELF/BIN
2012-10-20T22:58:47 < zyp> ok
2012-10-20T22:59:00 < zyp> then use whatever name the gpio struct is called instead of unsigned long then
2012-10-20T22:59:00 < upgrdman> and yes, i compiled for debugging
2012-10-20T22:59:12 < upgrdman> there isn't one... it's all #defines
2012-10-20T22:59:24 < jpa-> upgrdman: you need -gdwarf-2 so that it includes debug info about #defines
2012-10-20T22:59:37 < upgrdman> hum ok
2012-10-20T22:59:38 < zyp> there must be one, otherwise you wouldn't be able to use -> in your code
2012-10-20T23:00:04 < upgrdman> zyp, no... its #defines that resolve to pointers iirc
2012-10-20T23:00:18 < zyp> yes, to struct pointers
2012-10-20T23:00:23 < jpa-> upgrdman: you can do p/x *(GPIO_TypeDef*)0x40020000
2012-10-20T23:01:05 < upgrdman> oh sweet!
2012-10-20T23:01:20 < jpa-> i wish there was a way to load the .h just directly to gdb.. i often want to check peripherals while i'm some other scope and then i need to use hex addresses
2012-10-20T23:02:10 < upgrdman> http://pastebin.com/FN9ZxCrH
2012-10-20T23:02:39 < Laurenceb_> nice - they fixed i2c of F3
2012-10-20T23:02:42 < Laurenceb_> its now sane
2012-10-20T23:03:06 < zyp> sane how?
2012-10-20T23:03:11 <+Steffanx> That's want you to believe Laurenceb_
2012-10-20T23:03:42 < Laurenceb_> simplified interrupt handling
2012-10-20T23:04:15 < zyp> can you be more specific?
2012-10-20T23:04:33 < zyp> will I have to change my I2C code to make it work on F3?
2012-10-20T23:06:27 <+Steffanx> Got complains from Farnell yet zyp ?
2012-10-20T23:07:18 < zippe> jpa-: why not write gdb macros or python for that?
2012-10-20T23:07:34 < zippe> jpa-: of course, that would require ST to supply their data in a sensible format...
2012-10-20T23:07:43 < zyp> Steffanx, no, just shipment notifications
2012-10-20T23:07:51 <+Steffanx> :D
2012-10-20T23:08:08 < zippe> jpa-: for "real" work, I convinced the silicon guys to deliver YAML containing the address and field details for every register in the device
2012-10-20T23:08:23 < zyp> also, they are apparently shipping them to my addr, without my name, since I left the company name field blank
2012-10-20T23:08:42 < zyp> that part seems a bit dumb :p
2012-10-20T23:08:45 < zippe> jpa-: so our JTAG debugger knows every register in the SoC, and it can decode many of the fields in them
2012-10-20T23:09:10 < zippe> jpa-: It'd be kinda seksi if other folks would do the same
2012-10-20T23:10:55 < Laurenceb_> zyp: yes
2012-10-20T23:11:00 < Laurenceb_> it has a transfer size register
2012-10-20T23:11:07 < Laurenceb_> all the other registers have changed
2012-10-20T23:11:10 < zyp> oh
2012-10-20T23:11:16 < Laurenceb_> theres oversampling with a digital filter and stuff
2012-10-20T23:11:26 < Laurenceb_> you "set up" the transfer
2012-10-20T23:11:31 < Laurenceb_> and leave it alone
2012-10-20T23:11:38 < jpa-> zippe: yeah.. standard format for such a stuff would be nice; also for pinouts etc. (i think there is some format for pinouts)
2012-10-20T23:11:42 < zyp> sounds good
2012-10-20T23:11:43 < Laurenceb_> just put data in/out inside the isr
2012-10-20T23:12:29 < Laurenceb_> for >=256 bytes you have to reset mid way but other than that its simple
2012-10-20T23:15:14 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-10-20T23:15:15 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-20T23:15:23 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-20T23:15:46 < Laurenceb_> it can even finish with repeated start
2012-10-20T23:18:43 < upgrdman> to get spi working all you need is to setup the gpios (gpio regs + rcc clock enable) and the spi (spi reg + rcc clock enable) right? this is so frustrating for something that seem so simple. :(
2012-10-20T23:19:41 < zyp> yep
2012-10-20T23:19:54 < zyp> seems a bit weird, it works fine for me
2012-10-20T23:20:11 < upgrdman> do you have an f4discov board?
2012-10-20T23:20:21 < zyp> yes
2012-10-20T23:20:35 < upgrdman> and its updating xaccel etc with real values?
2012-10-20T23:20:56 < zyp> oh, I haven't tested that part
2012-10-20T23:21:01 < upgrdman> o ok
2012-10-20T23:21:10 < upgrdman> if i give you a complete main.c would you try it?
2012-10-20T23:21:15 < zyp> but I'm using SPI on my board which works
2012-10-20T23:21:20 < upgrdman> o
2012-10-20T23:22:04 < zyp> sure I can test it, if you give me a complete package without any external dependencies
2012-10-20T23:22:16 < upgrdman> sure. do you use gcc?
2012-10-20T23:22:19 < zyp> yes
2012-10-20T23:25:05 -!- GargantuaSauce [~sauce@blk-222-208-237.eastlink.ca] has joined ##stm32
2012-10-20T23:25:54 < upgrdman> zyp, http://www.rcheliguy.com/temp/open407vd_lcd.zip ... it's setup for use with an lcd, but it should work fine regardless. all of the pertinent code is in main.c. i did not write the lcd parts.
2012-10-20T23:26:28 < upgrdman> (the lcd is interfaced with fsmc, not spi)
2012-10-20T23:33:04 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-20T23:33:41 < upgrdman> i find that xaccel, yaccel and zaccel always get assigned 255... and probing the MISO/MOSI pins shows they are always low, with a big of noise corresponding the SCLK (parasitic inductance?)
2012-10-20T23:35:35 < upgrdman> s/big/bit
2012-10-20T23:40:04 -!- |sark [~sark@p54B646B9.dip.t-dialin.net] has joined ##stm32
2012-10-20T23:44:35 < zyp> fatal error: stm32f4xx.h: No such file or directory
2012-10-20T23:45:39 < zyp> I tested the compiled .elf you included though, it is actually creating activity on SPI lines here
2012-10-20T23:46:08 < zyp> but I want to test with some changes to the code, and then I need that file to be able to compile:p
2012-10-20T23:47:23 < zyp> found it myself
2012-10-20T23:49:03 < zyp> no
2012-10-20T23:51:38 < Tzacks> hey guys is there anyone with exp in digitl sound?
2012-10-20T23:52:32 < zyp> emeb seem to have some, but he's apparently not around now
2012-10-20T23:52:58 < Tzacks> yup i know about him but hes gone somwhere
2012-10-20T23:53:20 < zyp> upgrdman, ok, I give up for now, I've got other things to do
2012-10-20T23:53:34 < zyp> upgrdman, but MOSI works here
2012-10-20T23:54:20 < zyp> http://bin.jvnv.net/f/Ba7gT.png
2012-10-20T23:54:41 < zyp> you're running it so fast that my analyzer can't keep up, so don't trust the values there
2012-10-20T23:54:47 < upgrdman>  ok
2012-10-20T23:54:51 < zyp> but there is activity there
2012-10-20T23:54:52 < upgrdman> mosi = a7, right?
2012-10-20T23:54:58 < zyp> yep
2012-10-20T23:55:11 < upgrdman> and you didnt modify the code?
2012-10-20T23:55:31 < zyp> no, it's missing lots of stdlib dependencies that I don't have so I'm not able to compile it
2012-10-20T23:55:37 < zyp> I just flashed your firmware.elf
2012-10-20T23:55:43 < upgrdman> maybe my board has a bad solder joint or something. i'll probe some more.
2012-10-20T23:55:44 < upgrdman> ok
2012-10-20T23:56:15 < upgrdman> well thanks for the effort! :)
--- Day changed Sun Oct 21 2012
2012-10-21T00:06:22 -!- Claude is now known as claude
2012-10-21T00:08:10 < upgrdman> zyp, any chance you should share some code snippets of your working spi stuff?
2012-10-21T00:08:59 < zyp> http://cgit.jvnv.net/suzumebachi/tree/main.cpp
2012-10-21T00:10:56 -!- barthess [~barthess@5.100.200.146] has quit [Quit: Leaving.]
2012-10-21T00:19:15 < Laurenceb_> http://hackaday.com/2012/10/20/the-arduino-due-is-finally-here/
2012-10-21T00:19:21 < Laurenceb_> OHMYGOD
2012-10-21T00:20:00 < Laurenceb_> wait it has two usb microb headers now
2012-10-21T00:20:01 < Laurenceb_> wtf
2012-10-21T00:28:40 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-21T00:47:29 < zippe> Laurenceb: Might be the solution to their retarded bootloader story
2012-10-21T00:47:52 < zippe> Laurenceb: though still not clear how they think the host is going to reset the board
2012-10-21T01:06:55 -!- elektrinis [~circuit@78.60.169.125] has quit [Ping timeout: 260 seconds]
2012-10-21T01:09:02 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-21T01:47:00 -!- barthess [~barthess@5.100.214.200] has joined ##stm32
2012-10-21T02:14:39 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-21T02:22:09 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-21T02:27:04 -!- barthess [~barthess@5.100.214.200] has quit [Quit: Leaving.]
2012-10-21T02:45:49 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-219.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-21T03:06:55 < upgrdman> the f4discov manual lists the accelerometer CS line as PE2 in the table, and PE3 in the schematic. wtf.
2012-10-21T03:24:51 < dongs> dongs
2012-10-21T03:25:34 < baird> I expect the Due will have the usual Open Arduino libraries..
2012-10-21T03:27:13 -!- DaKu is now known as daku
2012-10-21T03:31:11 < dongs> http://i.imgur.com/Gvado.jpg attn emeb_mac
2012-10-21T03:33:15 < baird> Hmm, the Due looks like it could run my simulator. And it's got 256kB Flash onboard.
2012-10-21T03:34:15 < dongs> is due out or something.
2012-10-21T03:34:49 < dongs> rofl, first complaint
2012-10-21T03:34:50 < dongs> October 20, 2012 at 8:06 am
2012-10-21T03:34:51 < dongs> 3.3v operating voltage.. Does this mean ‘HIGH’ is 3.3v? :(
2012-10-21T03:34:59 < dongs> yes you dumbfuck, likehas been for the last decade
2012-10-21T03:38:15 < dongs> ledpcb fits well
2012-10-21T03:39:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-21T03:39:15 < upgrdman> lol
2012-10-21T04:04:44 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Quit: Tzacks]
2012-10-21T04:13:30 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-89.pptp.stw-bonn.de] has joined ##stm32
2012-10-21T04:28:54 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 264 seconds]
2012-10-21T04:30:53 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-21T04:31:55 < emeb_mac> yo dongs - nice boards.
2012-10-21T04:32:53 < dongs> soldering one up right now
2012-10-21T04:33:18 < emeb_mac> sweet
2012-10-21T04:35:15 -!- Ademan [~dan@adsl-70-231-140-183.dsl.snfc21.sbcglobal.net] has joined ##stm32
2012-10-21T04:38:36 < Ademan> http://www.st.com/internet/evalboard/product/254044.jsp Do any of the other boards match the sensors on this one? (gyro, acc, magnetometer)
2012-10-21T04:38:51 < dongs> "match"?
2012-10-21T04:39:57 < Ademan> Sorry, I mean, do any of the other boards have gyro, acc, magnetometer sensors as good as or equal to those found on the STM32F3 discovery board
2012-10-21T04:40:11 < zyp> of the discovery boards? no.
2012-10-21T04:40:25 < zyp> f3discovery is the only one with a gyro
2012-10-21T04:40:41 < zyp> f4discovery has an accelerometer but no gyro
2012-10-21T04:40:43 < dongs> i was under impression st sensors were fairly shitty which is probably why tehy're giving them away.
2012-10-21T04:41:27 < zyp> hmm, f3discovery is using approx. the same sensors as I use on my own f4 board
2012-10-21T04:41:38 < Ademan> Oh? Hrm. Well that's unfortunate, for the price I'll probably pick one up anyways, but do you know of a better set of sensors?
2012-10-21T04:41:50 < dongs> for WHAT
2012-10-21T04:42:00 < zyp> sensing of course
2012-10-21T04:42:04 < dongs> usualyl when you are looking for sensors you know what you want
2012-10-21T04:42:06 < Ademan> I almost said that :-/
2012-10-21T04:42:18 < dongs> and dont just join a random irc channel and go "HAI GUISE I NEED SUM SENSORS"
2012-10-21T04:43:33 < Ademan> As I said before I want a gyro, accelerometer and magnetometer. I want to do attitude sensing.
2012-10-21T04:43:56 < dongs> so, just get f3disco and done.
2012-10-21T04:44:22 < dongs> only stm32 left to solder up
2012-10-21T04:45:58 < Ademan> You just said you'd heard the sensors were crap though
2012-10-21T04:46:25 < zyp> crap is relative
2012-10-21T04:46:48 < zyp> they are probably good enough for whatever you are going to do
2012-10-21T04:46:56 < zyp> even though they are not the best on the market
2012-10-21T04:47:02 < Ademan> that's probably true
2012-10-21T04:47:13 < Ademan> Still it makes me want to know what else is out there that's better
2012-10-21T04:47:40 < dongs> by the time you will need to know that information, you will know it
2012-10-21T04:47:49 < Ademan> s/probably //
2012-10-21T04:48:11 < zyp> well, rumor has it that invensense has the best gyros, and was it freescale that has the best accelerometers?
2012-10-21T04:48:57 < zyp> no idea about magno, my impression is that external disturbances is a much larger problem than magno accuracy anyway
2012-10-21T04:49:19 < Ademan> thanks zyp
2012-10-21T04:49:28 < zyp> but you shouldn't really care about that before you're doing a board of your own
2012-10-21T04:49:50 < zyp> for a premade board, the f3discovery is probably the best bang for the buck you are going to get
2012-10-21T04:50:52 < dongs> and stm32 soldered
2012-10-21T04:51:10 < Ademan> Yeah, I certainly won't be designing my own board for a while (if ever) but that's good information, thanks
2012-10-21T04:51:17 < zyp> hmm
2012-10-21T04:51:20 < dongs> time to reflow xtal and led drivers
2012-10-21T04:51:37 < emeb_mac> was wondering how you'd do those
2012-10-21T04:51:49 < zyp> I should compare the f3discovery to my own board
2012-10-21T04:52:30 < dongs> compare in what? arent you using same sensors?
2012-10-21T04:52:35 < emeb_mac> wish I could get an f3 discovery.
2012-10-21T04:52:37 < zyp> when I tested my own board, gyro was noticeably more noisy than itg3200
2012-10-21T04:52:53 < dongs> and itg3200 is terrible
2012-10-21T04:52:54 < zyp> but I'm not sure that my vcc is clean enough
2012-10-21T04:53:45 < emeb_mac> separate reg for the vcc?
2012-10-21T04:54:26 < zyp> no, it's hooked to the output of the same buck regulator that powers everything else
2012-10-21T04:54:35 < dongs> lol buck
2012-10-21T04:55:38 < zyp> looks clean on the scope, but I suspect that my scope doesn't have enough resolution to show noise that would still get into the measurements :p
2012-10-21T04:56:22 < upgrdman> decoupling caps near sensors?
2012-10-21T04:56:26 < zyp> of course
2012-10-21T05:00:04 < zyp> next revision I'm probably going to do buck to 3.6V or something, and then hang several LDOs off that to different stuff, so each get a clean 3.3V
2012-10-21T05:00:25 < emeb_mac> maybe a ferrite bead between buck & sensitive loads?
2012-10-21T05:00:36 < zyp> probably a good idea
2012-10-21T05:01:32 < zyp> also, LDOs with enable inputs could be handy, then I can control power to other modules from software
2012-10-21T05:01:59 < HTT-Bird> emeb_mac: ferrite bead = "megahurts to millihurts converter" ;)
2012-10-21T05:02:27 < emeb_mac> heh
2012-10-21T05:05:35 < dongs> woohoo, analog jokes
2012-10-21T05:11:02 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-21T05:12:34 < dongs> http://i.imgur.com/8E03B.jpg time to try testing.
2012-10-21T05:13:00 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-10-21T05:14:37 < emeb_mac> nice soldering job - you did all but the xtal and drivers by hand, right?
2012-10-21T05:14:46 < dongs> yea
2012-10-21T05:14:54 < dongs> well drivers only the bottom pad.
2012-10-21T05:15:06 < emeb_mac> makes sense
2012-10-21T05:15:15 < dongs> blobbed some paste with one leg fixed and heated it up.
2012-10-21T05:15:15 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-21T05:15:55 < emeb_mac> i've done xtals like that with an iron too
2012-10-21T05:16:22 < dongs> yeah, it doesnt need paste.
2012-10-21T05:16:25 < zyp> by the way, what's the pwm frequency of the drivers?
2012-10-21T05:17:28 < dongs> 1khz @ 16bit it seems unless im reading the ds wrong
2012-10-21T05:17:36 < dongs> >1000Hz@16-bit grayscale, >256KHz@8-bit grayscale
2012-10-21T05:18:42 < zyp> 1 kHz sounds a bit low
2012-10-21T05:19:20 -!- |sark [~sark@p54B646B9.dip.t-dialin.net] has quit [Ping timeout: 260 seconds]
2012-10-21T05:19:37 < zyp> you'll want the pwm frequency to be at least an order of magnitude higher than the muxing frequency
2012-10-21T05:19:57 < emeb_mac> oh darn. can't do 16-bit.
2012-10-21T05:20:27 < emeb_mac> like anyone could see the difference on an 8x8 led.
2012-10-21T05:20:37 < zyp> but really, 8-bit will give you 24-bit colors, should be more than good enough :p
2012-10-21T05:20:46 < emeb_mac> exactly
2012-10-21T05:20:55 < zyp> and 256kHz is certainly fine
2012-10-21T05:21:19 < dongs> 5v applied, nothing smoked.
2012-10-21T05:21:21 < dongs> so its prob ok.
2012-10-21T05:21:36 < emeb_mac> time for some firmware.
2012-10-21T05:23:00 < dongs> swd works
2012-10-21T05:24:19 < emeb_mac> good sign
2012-10-21T05:27:59 < emeb_mac> hmmm - not finding a full data sheet for the my9221
2012-10-21T05:28:05 < HTT-Bird> emeb_mac: how close together can you stick parts before they start causing assembly issues when you are doing reflow processing?
2012-10-21T05:28:30 < emeb_mac> HTT-Bird: don't ask me - I haven't done any automated assy.
2012-10-21T05:28:37 < emeb_mac> dongs: probably knows.
2012-10-21T05:28:39 < HTT-Bird> emeb_mac: heh.  manual place + toaster oven
2012-10-21T05:28:51 < HTT-Bird> aka reflow@home
2012-10-21T05:29:18 < emeb_mac> HTT-Bird: I'm even worse - all hand soldered except for a few hotplate reflows.
2012-10-21T05:29:27 < dongs> HTT-Bird: ive placed passives next to ecah other on automated assembled shit like you see in that last pic near USB sockets and its fine
2012-10-21T05:29:48 < zyp> why wouldn't it be?
2012-10-21T05:29:49 < emeb_mac> so in other words - do you your worst human.
2012-10-21T05:31:29 < emeb_mac> ah - seeeeed had the ds.
2012-10-21T05:35:50 -!- CheBuzz_Home [~CheBuzz@76.164.192.124] has quit [Ping timeout: 268 seconds]
2012-10-21T05:43:55 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-21T05:53:54 -!- CheBuzz_Home [~CheBuzz@76.164.192.124] has joined ##stm32
2012-10-21T06:16:24 < dongs> scanning works.. now to dump some spi data
2012-10-21T06:19:59 < emeb_mac> so the 3:8 decoder and PFETs are working?
2012-10-21T06:20:49 < dongs> ya
2012-10-21T06:21:04 < dongs> i suppose i should check rise time
2012-10-21T06:21:09 < dongs> but i just used random pfets i ha d
2012-10-21T06:21:41 < emeb_mac> got the array in?
2012-10-21T06:21:55 < dongs> no not yet
2012-10-21T06:22:03 < emeb_mac> how are you loading the fets?
2012-10-21T06:22:12 < dongs> im not. just stickign a scope on them.
2012-10-21T06:22:42 < emeb_mac> until you load them down with some current the risetime measurements won't mean much.
2012-10-21T06:22:49 < dongs> ok, then i wont bother
2012-10-21T06:24:25 < emeb_mac> yep - you'd need about 10 ohms to gnd to match what happens worst-case in the real system.
2012-10-21T06:25:23 < emeb_mac> which is roughly 1.25W, so it'd have to be a big resistor.
2012-10-21T06:26:19 < dongs> looks like I can use 16bit SPI mode to stick stuff into the driver
2012-10-21T06:28:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-21T06:28:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-21T06:31:07 < emeb_mac> seems the 9221 needs 17 x 16-bit transfers to load all regs.
2012-10-21T06:31:38 < dongs> i thought it was 13
2012-10-21T06:31:41 < dongs> 208/16
2012-10-21T06:33:15 < emeb_mac> yep
2012-10-21T06:33:24 < emeb_mac> but...
2012-10-21T06:33:55 < emeb_mac> notice that it clocks data on both edges of dclk!
2012-10-21T06:34:45 < dongs> oh, dont fuck with me now.
2012-10-21T06:34:49 < dongs> are you saying i cant use hardware spi? heh
2012-10-21T06:35:04 < emeb_mac> looks that way.
2012-10-21T06:35:21 < emeb_mac> bitbang!
2012-10-21T06:35:30 < dongs> lol chinaDDR
2012-10-21T06:35:38 < emeb_mac> heh
2012-10-21T06:36:56 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-21T06:37:03 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-21T06:37:52 < emeb_mac> plus, there's a special pulse sequence needed on di to latch the data in
2012-10-21T06:39:24 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Ping timeout: 260 seconds]
2012-10-21T06:46:02 < HTT-Bird> bah!
2012-10-21T06:46:24 < HTT-Bird> silly people who think they need to reinvent the square wheel :P
2012-10-21T06:47:52 < emeb_mac> yeah - why not just use spi?
2012-10-21T06:49:09 < emeb_mac> this allows lower clock freq tho - good for long strings
2012-10-21T06:52:01 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-21T07:06:29 < dongs> time to go enjoy weekend. looks like everything should be lighting up this evening.
2012-10-21T07:37:21 < Ademan> zyp: what's your board?
2012-10-21T07:49:57 < emeb_mac> wat? dongs has a life away from irc? away from boards/chips/soldering? *eyeroll*
2012-10-21T08:07:01 -!- Erlkoenig [~erlkoenig@pptp-212-201-77-89.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-21T08:11:05 < dongs> 1 hour for weekend enjoyment is enough.
2012-10-21T08:19:10 < emeb_mac> couldn't stay away.
2012-10-21T08:23:02 < dongs> looks ok
2012-10-21T08:23:06 < dongs> i got it drawing random ass pattern
2012-10-21T08:23:24 < dongs> need to put scanning on a timer
2012-10-21T08:23:35 < dongs> https://youtube.com/watch?v=UFaSo-FMHiI
2012-10-21T08:24:11 < dongs> bbl
2012-10-21T08:24:28 < emeb_mac> sweepy scanny!
2012-10-21T08:26:02 < baird> Needs more epilepsy.
2012-10-21T08:31:09 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Remote host closed the connection]
2012-10-21T08:39:02 < emeb_mac> interesting - F373 doesn't have BOOT1 pin.
2012-10-21T08:39:19 < emeb_mac> does have 'nBOOT1' bit in a system register though.
2012-10-21T08:46:25 -!- claude is now known as Claude
2012-10-21T08:56:46 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-21T09:05:13 < dongs> emeb_mac: right
2012-10-21T09:05:33 < dongs> boot1 was just wasting a gpio
2012-10-21T09:05:44 < dongs> utterly useless on 48lqfp parts too
2012-10-21T09:05:49 < emeb_mac> yeah - i like it.
2012-10-21T09:06:05 < dongs> i mean boot1 toggles booting from 'external sram' right
2012-10-21T09:06:10 < dongs> cant even hook that up to lqfp48.
2012-10-21T09:06:13 < dongs> so shrug.
2012-10-21T09:06:29 < emeb_mac> was it external, or just SRAM?
2012-10-21T09:06:39 < dongs> oh, maybe sram
2012-10-21T09:06:54 < dongs> i think some combo of 1/0 made it external
2012-10-21T09:07:04 < dongs> you can boot shit from sram over debugger anyway, no?
2012-10-21T09:07:12 < emeb_mac> think you can load up code into RAM, hit reset w/ Boot 1/0 set right and execute volatile mem.
2012-10-21T09:07:38 < emeb_mac> yep - debugger lets you run from RAM w/o fiddling the boot pins.
2012-10-21T09:08:24 < emeb_mac> anyway - only BOOT0 is good.
2012-10-21T09:09:51 < dongs> so what should i ahve scan set at
2012-10-21T09:09:54 < dongs> 1khz?
2012-10-21T09:10:16 < emeb_mac> 1khz for the whole array?
2012-10-21T09:10:28 < dongs> no i mean, im gonna run a timer
2012-10-21T09:10:33 < dongs> with period of whatever
2012-10-21T09:10:39 < dongs> and toggle scanline in it
2012-10-21T09:10:42 < emeb_mac> ok so 1ms / row.
2012-10-21T09:11:19 < emeb_mac> that will update the whole array @ 125Hz
2012-10-21T09:12:02 < emeb_mac> which is roughly LCD TV raster rate. Might be a bit on the low side.
2012-10-21T09:12:12 < emeb_mac> but usable for starters.
2012-10-21T09:13:17 < emeb_mac> is this the xtal you used: http://www.digikey.com/product-detail/en/NX3225SA-12.000000MHZ-STD-CSR-1/644-1047-1-ND/1128919
2012-10-21T09:13:22 < dongs> yea
2012-10-21T09:15:17 < emeb_mac> and the jtag connector: http://www.digikey.com/product-detail/en/20021121-00010T4LF/609-3729-ND/2209075
2012-10-21T09:15:30 < dongs> Ya
2012-10-21T09:15:39 < dongs> thats one of them, chepest one I belive
2012-10-21T09:15:52 < emeb_mac> yup. no need for shroud
2012-10-21T09:15:52 < dongs> the original is some FTSH series.
2012-10-21T09:16:02 < emeb_mac> unobtainium from samtec
2012-10-21T09:16:27 < dongs> SN74HCT138PWR for the 3:8
2012-10-21T09:16:40 < dongs> WM17115-ND for usb
2012-10-21T09:16:58 < dongs> 74HCT1G125 for buffer
2012-10-21T09:18:01 < emeb_mac> thx - putting together an order...
2012-10-21T09:19:34 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 265 seconds]
2012-10-21T09:21:27 < emeb_mac> stm32f103c ???
2012-10-21T09:21:44 < HTT-Bird> emeb_mac: samtec makes unobtainium?
2012-10-21T09:22:00 < HTT-Bird> I thought DK was better about carrying their stuff man...
2012-10-21T09:22:10 < emeb_mac> HTT-Bird: that one is non-stock
2012-10-21T09:22:48 < emeb_mac> some stuff is in stock, but Samtec's big on custom orders.
2012-10-21T09:23:29 < dongs> wth
2012-10-21T09:24:51 < emeb_mac> bugs?
2012-10-21T09:25:05 < dongs> no, forgot to enable TIM_ITConfig
2012-10-21T09:25:09 < dongs> and was like "its not firing"
2012-10-21T09:25:36 < emeb_mac> simple fix.
2012-10-21T09:29:35 < emeb_mac> dongs: which STM32F103 are you using?
2012-10-21T09:29:47 < dongs> theres only 2, C8T6 or CBT6
2012-10-21T09:29:54 < dongs> 64/128k flash "officially" but both have 128k
2012-10-21T09:31:42 < emeb_mac> kk
2012-10-21T09:31:51 < HTT-Bird> dongs: helps to turn it on :)
2012-10-21T09:32:25 < emeb_mac> embedded coding version of "is it plugged in?"
2012-10-21T09:37:28 < dongs> driver gettin toasty @ FF output
2012-10-21T09:37:34 < dongs> 680mA draw
2012-10-21T09:37:35 < emeb_mac> not surprising.
2012-10-21T09:37:42 < emeb_mac> 680 is a bit high.
2012-10-21T09:38:21 < emeb_mac> that's like 200ma more than I'd expect for all on
2012-10-21T09:38:36 < emeb_mac> (not including the mcu etc)
2012-10-21T09:39:10 < dongs> all on is very nice.
2012-10-21T09:39:28 < dongs> it did flicker pretty terribly at 1khz
2012-10-21T09:39:32 < dongs> so i changed it to /10
2012-10-21T09:39:34 < dongs> wahtever that makes it
2012-10-21T09:39:41 < dongs> timer period = 100 @ 1mhz
2012-10-21T09:40:30 < dongs> lets try some tarduino hsv demos
2012-10-21T09:41:50 < emeb_mac> flyback-: the lcd?
2012-10-21T09:44:51 < emeb_mac> dongs: you used 2.2k Rext -> gives about 26.5ma / LED. x24 = 636ma. Add in MCU & other parts and that's about right.
2012-10-21T09:45:46 < dongs> ya.
2012-10-21T09:45:48 < dongs> i didnt have 2k
2012-10-21T09:46:10 < dongs> emeb_mac: /msg me your info so i can get 3 sets of shit shipped over tomororw
2012-10-21T09:48:09 < dongs> noted
2012-10-21T09:48:25 < emeb_mac> ty
2012-10-21T09:49:37 < emeb_mac> anything special about the 6-pin in/out connectors?
2012-10-21T09:51:04 < dongs> nope
2012-10-21T09:51:07 < dongs> just regular shits
2012-10-21T09:51:11 < dongs> like angled 2.54mm
2012-10-21T09:51:23 < dongs> i had the male ones and female  i got 100 from china
2012-10-21T09:51:31 < dongs> i think the sizes are standard
2012-10-21T09:52:05 < emeb_mac> ok
2012-10-21T09:53:11 < emeb_mac> heh - need to buy a load of 0603 resistors. All I stock are 0805.
2012-10-21T09:53:36 < dongs> lucky all you need for my shit is 10k, 4.7k, 1.5k, 22R
2012-10-21T09:53:45 < dongs> and whatever for Rset
2012-10-21T09:53:50 < emeb_mac> 2.2k
2012-10-21T09:53:50 < dongs> 2k i guess.
2012-10-21T09:54:02 < dongs> well, i spec'd 2k on schematic but i didnt have any.
2012-10-21T09:54:02 < emeb_mac> get 100x ea
2012-10-21T09:54:25 < emeb_mac> not much diff
2012-10-21T09:58:16 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-21T10:00:41 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-21T10:01:07 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-21T10:09:24 < emeb_mac> time for zzz
2012-10-21T10:11:26 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-21T10:32:14 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 240 seconds]
2012-10-21T10:45:59 -!- tech2077 [~tech2077@75-148-212-157-Houston.hfc.comcastbusiness.net] has joined ##stm32
2012-10-21T11:11:40 -!- tech2077 [~tech2077@75-148-212-157-Houston.hfc.comcastbusiness.net] has quit [Quit: Leaving]
2012-10-21T11:21:55 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 246 seconds]
2012-10-21T11:23:47 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-21T11:28:13 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-21T11:46:16 < dongs> aww yea
2012-10-21T11:52:47 < upgrdman> ?
2012-10-21T11:52:57 < dongs> it blinks
2012-10-21T11:53:21 < dongs> https://youtube.com/watch?v=dMO_7iHXYhg
2012-10-21T12:00:55 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-21T12:09:46 -!- Claude is now known as claude
2012-10-21T12:23:23 -!- barthess [~barthess@164.177.197.245] has joined ##stm32
2012-10-21T12:32:20 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-21T12:37:16 -!- elektrinis [~circuit@78.60.169.125] has joined ##stm32
2012-10-21T12:42:42 -!- daku is now known as DaKu
2012-10-21T12:59:35 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-21T13:16:01 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 246 seconds]
2012-10-21T13:40:56 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T14:00:44 < dongs> where the hell are the chats
2012-10-21T14:09:35 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-21T14:10:35 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-21T14:52:39 < baird> IRL chatting about Mass Effect 3 pornfics
2012-10-21T14:52:45 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 256 seconds]
2012-10-21T14:53:22 < dongs> sounds like the kinda shit cjbeard would be into
2012-10-21T14:53:24 < dongs> oh shit, thats you
2012-10-21T14:54:48 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T14:55:58 < baird> The housemate is trying to do 'serious' writing, yet I'm trolling her with that. :D
2012-10-21T14:59:02 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-21T15:03:23 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-21T15:03:26 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-21T15:06:11 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-21T15:10:40 -!- tavish [~tavish@120.56.131.69] has joined ##stm32
2012-10-21T15:10:40 -!- tavish [~tavish@120.56.131.69] has quit [Changing host]
2012-10-21T15:10:40 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T15:13:26 < Thorn> how do I connect an encoder to a stm32? should outputs be active high or low or it doesn't matter? should I add r/c for debouncing? timer inputs seem to have some kind of filter internally
2012-10-21T15:14:07 < dongs> encoder? you mean like knob shit that outputs AB?
2012-10-21T15:14:11 < dongs> you dont even need timer for it man
2012-10-21T15:14:27 < Thorn> yes that's what I mean
2012-10-21T15:14:39 < Thorn> but timers do support those things
2012-10-21T15:14:44 < dongs> http://www.arduino.cc/playground/Main/RotaryEncoders
2012-10-21T15:14:51 < dongs> its like 5 lines of code
2012-10-21T15:15:38 <+Steffanx> Ardongso powa!
2012-10-21T15:16:00 < Thorn> I'm going to have better things to do than poll encoder inputs
2012-10-21T15:17:02 < dongs> tru
2012-10-21T15:17:06 < dongs> just sayin.
2012-10-21T15:17:14 < dongs> you could exti one
2012-10-21T15:17:18 < dongs> and read the other one..
2012-10-21T15:17:21 < dongs> same principle.
2012-10-21T15:17:28 < dongs>  i thikn that swhat I did on my f4 proj that had a knob
2012-10-21T15:17:58 < Thorn> exti = 1244 interrupts per click if not debounced
2012-10-21T15:18:15 < dongs> i was only getting 1243 :(
2012-10-21T15:20:32 < jpa-> Thorn: quadrature encoder doesn't care about contact bouncing
2012-10-21T15:20:42 < jpa-> just use the timer inputs
2012-10-21T15:21:07 < Thorn> does it care about polarity?
2012-10-21T15:21:44 < Thorn> is it ok if I put pullups on outputs and tie the common pin to ground?
2012-10-21T15:22:00 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
2012-10-21T15:22:05 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 244 seconds]
2012-10-21T15:22:29 < zyp> polarity just affects rotation direction if anything
2012-10-21T15:23:55 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T15:24:01 < Thorn> okay thanks.
2012-10-21T15:29:24 < Thorn> look at that! I even have 2 unused pins mapped to tim3 channels
2012-10-21T15:29:45 < Thorn> no free SPI though so will have to bitbang the display :<
2012-10-21T15:29:55 < Laurenceb_> suppp
2012-10-21T15:30:16  * Laurenceb_ is trying to work out how to characterise the F3 opamps
2012-10-21T15:30:38 < Laurenceb_> think if i set them to use internal ref then use external feedback network
2012-10-21T15:31:01 < Laurenceb_> and connect output pin to my tectronix scope
2012-10-21T15:31:31 < Laurenceb_> and get noise spectrum with two different feedback networks, then its possible to work out voltage and current noise spectrum
2012-10-21T15:38:12 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-21T16:02:45 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-21T16:04:35 -!- tavish [~tavish@120.59.134.194] has joined ##stm32
2012-10-21T16:04:35 -!- tavish [~tavish@120.59.134.194] has quit [Changing host]
2012-10-21T16:04:35 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T16:08:56 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-21T16:16:10 -!- badger [~tom@cxkg.net] has joined ##stm32
2012-10-21T16:17:55 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-10-21T16:43:30 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-21T16:51:02 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-165.pptp.stw-bonn.de] has joined ##stm32
2012-10-21T16:56:24 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-21T16:56:51 < TitanMKD> hello
2012-10-21T16:57:10 < zyp> hi
2012-10-21T16:58:08 < TitanMKD> any news on LPC43xx board ;)
2012-10-21T16:58:25 < zyp> not really
2012-10-21T16:58:32 < zyp> I was breaking out some more stuff yesterday
2012-10-21T16:59:41 < TitanMKD> ha
2012-10-21T17:00:07 < zyp> http://bin.jvnv.net/f/U88MX.png
2012-10-21T17:01:25 < dongs> zyp: chinaled works
2012-10-21T17:02:28 < TitanMKD> zyp at your place i will keep place on bottom for decoupling capacitor
2012-10-21T17:02:37 < TitanMKD> just under the LBGA
2012-10-21T17:03:46 < zyp> I would like to avoid two-sided assembly
2012-10-21T17:03:51 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-21T17:11:40 < zyp> http://bin.jvnv.net/f/q1xHk.png <- also, it's not much room for capacitors underneath the chip anyway
2012-10-21T17:19:23 < TitanMKD> it is just an advise on LBGA as they really need decoupling capacitor as closer as possible ...
2012-10-21T17:19:57 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Quit: Leaving]
2012-10-21T17:21:44 < TitanMKD> byt yes at end it requires capacitor 0402 or even 0201 but 0201 seems just impossible to solder by hand even if I have never tried
2012-10-21T17:23:12 < TitanMKD> it seems there's also embedded capacitor buried on PCB for some case but I imagine the cost shall be really high
2012-10-21T17:28:36 -!- tavish_ [~tavish@120.56.129.40] has joined ##stm32
2012-10-21T17:30:50 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-21T17:31:05 -!- tavish__ [~tavish@59.177.9.193] has joined ##stm32
2012-10-21T17:33:38 -!- tavish_ [~tavish@120.56.129.40] has quit [Ping timeout: 268 seconds]
2012-10-21T17:35:30 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-21T17:38:44 -!- tavish__ [~tavish@59.177.9.193] has quit [Ping timeout: 256 seconds]
2012-10-21T17:40:27 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-21T17:40:30 -!- tavish_ [~tavish@120.56.142.25] has joined ##stm32
2012-10-21T17:46:11 -!- tavish__ [~tavish@120.59.130.15] has joined ##stm32
2012-10-21T17:46:18 -!- tavish_ [~tavish@120.56.142.25] has quit [Ping timeout: 272 seconds]
2012-10-21T17:46:23 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Read error: Connection reset by peer]
2012-10-21T17:51:15 -!- tavish__ [~tavish@120.59.130.15] has quit [Ping timeout: 256 seconds]
2012-10-21T17:52:30 < emeb_mac> morning trollers and trollees
2012-10-21T17:54:21 < dongs> only trolee is TitanMKD
2012-10-21T17:54:32 <+Steffanx> Hi McDonalds food
2012-10-21T17:54:39 < dongs> emeb_mac: https://youtube.com/watch?v=7_0DMUjTb-c
2012-10-21T17:54:54 < dongs> and https://youtube.com/watch?v=dMO_7iHXYhg (retweet)
2012-10-21T17:55:18 <+Steffanx> Fancy "shift in from right" feature dong
2012-10-21T17:55:23 < dongs> i kno rite
2012-10-21T17:55:33 < dongs> i was goin for something like 'd o  n  g   s'
2012-10-21T17:56:17 < emeb_mac> very nice
2012-10-21T17:56:26 < dongs> emeb_mac: i did find one problem though
2012-10-21T17:56:31 < emeb_mac> ?
2012-10-21T17:56:38 < dongs> Input Voltage “H” Level VIH CMOS logic level 0.7VDD . VDD
2012-10-21T17:57:27 < dongs> so gotta run array at like 4.5V
2012-10-21T17:57:37 < dongs> or else it dongs up @ 5V exactly
2012-10-21T17:57:50 -!- tavish_ [~tavish@120.59.128.70] has joined ##stm32
2012-10-21T17:58:07 < emeb_mac> I noticed on the photo you posted that the 3:8 decoder isn't HCT
2012-10-21T17:58:15 < emeb_mac> wonder if that is the reason.
2012-10-21T17:58:23 < dongs> no 3:8 works fine
2012-10-21T17:58:29 < dongs> its the drivers that dont work
2012-10-21T17:58:42 < emeb_mac> Oh - the 9221?
2012-10-21T17:58:55 < dongs> yes
2012-10-21T17:59:18 < emeb_mac> wonder if those could be run @ 3.3V even with +5 to the LEDs?
2012-10-21T17:59:41 < emeb_mac> the 9221 should never see +5V due to the drop of the LEDs.
2012-10-21T17:59:48 < dongs> yes
2012-10-21T17:59:56 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-21T17:59:59 < dongs> VDD and VLED are different.
2012-10-21T18:00:44 < dongs> Operating Supply Voltage VDD 3.0 ~ 5.5 V
2012-10-21T18:00:45 < dongs> ya
2012-10-21T18:00:46 < dongs> silly me.
2012-10-21T18:00:54 < dongs> anyhow, its plenty bright at 4.5V
2012-10-21T18:01:04 < emeb_mac> so wait - the 9221 is already running @ VDD = 3.3V, but if VLED > 4.5V then things screw up?
2012-10-21T18:01:14 < dongs> no
2012-10-21T18:01:19 < dongs> 9221 is running at vled
2012-10-21T18:01:22 < dongs> (on my hardware.
2012-10-21T18:01:24 < dongs> i routed 5V there.
2012-10-21T18:01:26 < emeb_mac> OK.
2012-10-21T18:01:32 < dongs> i didnt want to use the stm32's 3.3V reg for anythign else.
2012-10-21T18:01:47 < emeb_mac> understood.
2012-10-21T18:01:54 < dongs> theres just one trace going to them for 5V and doesnt loop anywhere else
2012-10-21T18:02:08 < dongs> ill try cutting it on my board and wire it to 3.3 and it should run at vled=5 as well
2012-10-21T18:02:18 < emeb_mac> yep
2012-10-21T18:02:38 < dongs> anyway, that 1st video i posted with epileptic-like drawing
2012-10-21T18:02:41 < dongs> is what it does when vled=5
2012-10-21T18:02:50 < dongs> it mis-latches/mis-clocks stuff
2012-10-21T18:02:52 < emeb_mac> OK - I was wondering about that.
2012-10-21T18:03:08 < dongs> as it goes over like 4.9V or so, weird shit starts to happen.
2012-10-21T18:03:13 < dongs>  isaw exactly same effecto n WS2811's
2012-10-21T18:03:16 -!- tavish_ [~tavish@120.59.128.70] has quit [Ping timeout: 260 seconds]
2012-10-21T18:03:22 < dongs> which are also cmos/0.7VCC high type shit
2012-10-21T18:04:34 < emeb_mac> you got the link to the schematic pix handy?
2012-10-21T18:04:56 < emeb_mac> otherwise I gotta trawl the logs...
2012-10-21T18:05:54 < dongs> should be in my imgur src
2012-10-21T18:06:01 < dongs> http://i.imgur.com/VYzaD.png
2012-10-21T18:06:40 < emeb_mac> ty
2012-10-21T18:09:36 < zyp> I had the same problem with my stuff
2012-10-21T18:09:54 < zyp> when I ran it at 5V it wouldn't respond properly to 3.3V IO
2012-10-21T18:10:00 < emeb_mac> just looked at the my9221 ds again - the outputs are HV tolerant, so there shouldn't be a big deal with VLED != VDD
2012-10-21T18:10:27 < zyp> but since I have only red leds, running everything at 3.3V is no problem
2012-10-21T18:10:32 < emeb_mac> stupid logic input design.
2012-10-21T18:11:03 < emeb_mac> need to set the vih/vil thresholds differently.
2012-10-21T18:11:57 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-21T18:12:21 < emeb_mac> so the main question on the 9221 parts is if they generate trash on VDD that would hose up the MCU.
2012-10-21T18:13:06 < emeb_mac> but you've got them nicely bypassed so that shouldn't be a huge deal.
2012-10-21T18:16:39 < TitanMKD> i soldered my first LBGA 256 with hot plate
2012-10-21T18:16:44 < TitanMKD> result seems really not bad
2012-10-21T18:17:03 < TitanMKD> i can clearly see the balls are soldered on each side
2012-10-21T18:19:08 < ossifrage> TitanMKD, did you have to use weights?
2012-10-21T18:20:17 < TitanMKD> i pushed a bit on the top during soldering process with a tweezer
2012-10-21T18:20:56 < TitanMKD> before soldering i have also added flux on the pcb
2012-10-21T18:21:43 < TitanMKD> but the real test will be when i will power on the board ;) because there is no way to check all ball are soldered correctly as i have no xray to check
2012-10-21T18:22:15 < emeb_mac> TitanMKD: can't do a quick impedance test w/ DMM to check for opens/shorts?
2012-10-21T18:23:14 < TitanMKD> emeb_mac yes i will do that especially on power pins vs gnd to be sure it's > than 1KOhm
2012-10-21T18:24:10 < emeb_mac> on lqfp I test continuity between adjacent pins to avoid bridges. Might be good to spot-check that too.
2012-10-21T18:24:37 < TitanMKD> yes i also do that on all available pins
2012-10-21T18:24:56 < TitanMKD> anyway i have no 3.3V regulator anymore to do the final power on test ;)
2012-10-21T18:25:13 < TitanMKD> so i will do maximum check with voltmeter short/open vdd/vss
2012-10-21T18:25:54 < emeb_mac> When powering on a new board recently I ran the supply through a DMM current meter to check for large draw.
2012-10-21T18:26:10 < TitanMKD> anyway Itead have done a very good job on this PCB
2012-10-21T18:26:22 < emeb_mac> when I saw > 400ma I shut down quickly and found a SOIC in backwards.
2012-10-21T18:26:45 < TitanMKD> emeb_mac yes it is why i have built a USB PCB to check current consumption ;)
2012-10-21T18:27:19 < TitanMKD> yes or i touch with finger and if the MCU is hot it's very bad
2012-10-21T18:27:29 < emeb_mac> by that time it's usally too late.
2012-10-21T18:27:38 < TitanMKD> i have burnt 2 STM32F4 because of a mistake on my PCB last time ;)
2012-10-21T18:27:54 < emeb_mac> A current-limited bench supply would be a good thing for these situations.
2012-10-21T18:28:19 < TitanMKD> in fact i was using TI DC/DC and they detected the short/problem
2012-10-21T18:28:34 < emeb_mac> smart supply!
2012-10-21T18:28:50 < TitanMKD> yes they regulate current
2012-10-21T18:28:52 < emeb_mac> power up and it shouts "UR DOIN IT RONG"
2012-10-21T18:29:06 < TitanMKD> really better than LM1117 which burn everything behind ;)
2012-10-21T18:29:21 < TitanMKD> even if it also protect against big short ;) like A
2012-10-21T18:29:24 < TitanMKD> 1A
2012-10-21T18:29:46 < emeb_mac> mmm... toasty.
2012-10-21T18:30:13 < TitanMKD> in fact it does not really shout but it try to keep the voltage ;) and i was thinking this was the problem on the DC components badly soldered ;)
2012-10-21T18:30:40 < TitanMKD> and when i replaced it by the LM1117 glurps ;) the chipset was very very hot in 5s
2012-10-21T18:31:33 < emeb_mac> time for new chips.
2012-10-21T18:32:35 < TitanMKD> yep ;)
2012-10-21T18:33:05 < TitanMKD> ok 9.5MegaOhm between GND and GPIO maybe it is not bad
2012-10-21T18:34:29 < TitanMKD> (whithout any capacitor connected or anything else except the LBGA LPC43xx soldered)
2012-10-21T18:45:29 < TitanMKD> anyway xray check will be a must to validate quality of the soldering process with a simple hot plate and a correct ramp up/down
2012-10-21T18:45:42 < dongs> lets put it this way
2012-10-21T18:45:50 < dongs> anyone "simple hot plate" soldering BGA isnt gonna ahve xray
2012-10-21T18:46:00 < dongs> i solder bga and i have xray, but I'm not soldering it on a fucking hotplate.
2012-10-21T18:46:43 < TitanMKD> dongs hehe of course but just to know if for small batch/prototype it's good enough
2012-10-21T18:47:40 < TitanMKD> and whith cheap hot plate the temperature is very homogeneous especially for small board like that 5cm*5cm
2012-10-21T18:47:53 < dongs> seedstudio special
2012-10-21T18:47:58 < TitanMKD> yes ;)
2012-10-21T18:48:29 < TitanMKD> but in fact i used itead especially for LBGA
2012-10-21T18:50:11 < dongs> since they're actually worse than seed, im surprised you even bothered
2012-10-21T18:51:46 < TitanMKD> in fact for LBGA you pay extra US$ and the result is really better
2012-10-21T18:53:55 < TitanMKD> so 9.9USD+8.4USD (for LBGA tight space) for 10boards is still very cheap
2012-10-21T18:54:07 < dongs> pics or it didnt fucking happen.
2012-10-21T18:54:14 < dongs> last time i ordered their trash, it was fucking horrible.
2012-10-21T18:54:20 < dongs> drills all over the place, etc
2012-10-21T18:56:16 < TitanMKD> dongs look here http://www.postimage.org/image/vrzq6xm4h/
2012-10-21T18:56:58 < TitanMKD> silkscreen are not very good because i used too little vector font
2012-10-21T18:57:35 < dongs> glad to see their drills hasnt improved.
2012-10-21T18:57:38 < dongs> yeah, this is fucking garbage.
2012-10-21T18:57:56 < dongs> check out the 3 vias you ahve under bga
2012-10-21T18:58:03 < dongs> none of htem are drilled right
2012-10-21T18:58:12 < dongs> esp the one in bottom right and one above it
2012-10-21T18:58:13 -!- tavish [~tavish@120.59.13.48] has joined ##stm32
2012-10-21T18:58:13 -!- tavish [~tavish@120.59.13.48] has quit [Changing host]
2012-10-21T18:58:13 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T18:59:18 < zyp> uh, wat
2012-10-21T18:59:25 < zyp> SW-DP detected IDCODE: 0x2BA01477
2012-10-21T18:59:25 < zyp> Available Targets:
2012-10-21T18:59:25 < zyp> No. Att Driver
2012-10-21T18:59:25 < zyp>  1      LuminaryMicro Stellaris
2012-10-21T18:59:44 < TitanMKD> but for those via i have just used via with no stop it is why you do not see the via
2012-10-21T18:59:45 < zyp> this is L1
2012-10-21T19:01:29 < TitanMKD> dongs anyway for the price i think it is good (especially for this crappy designed board and just for test purpose)
2012-10-21T19:01:48 < qyx_> 18usd for 10 boards?
2012-10-21T19:02:09 < TitanMKD> yes
2012-10-21T19:02:55 < TitanMKD> anyone test that http://www.wedirekt.de/index.php/pcb/2
2012-10-21T19:03:27 < TitanMKD> they are really more expensive than Itead/Seed but it is for complex multi layer boards and the quality seems very good
2012-10-21T19:04:11 < TitanMKD> It is Würth Elektronik Group
2012-10-21T19:05:25 < qyx_> the vias looks really bad
2012-10-21T19:07:39 < TitanMKD> qyx_ via contact are hidden in fact because i used via without stop layer
2012-10-21T19:08:05 < TitanMKD> but yes via size is not respected too
2012-10-21T19:08:35 < dongs> wtf are ytou talkign about
2012-10-21T19:08:37 < dongs> via without stop layer?
2012-10-21T19:08:50 -!- barthess [~barthess@164.177.197.245] has quit [Quit: Leaving.]
2012-10-21T19:08:54 < zyp> tented via, I guess
2012-10-21T19:08:59 < dongs> zyp: no way.
2012-10-21T19:09:04 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 260 seconds]
2012-10-21T19:09:36 < dongs> or is tenting here just filling up with mask
2012-10-21T19:09:40 < dongs> which is like default on my shit.
2012-10-21T19:09:55 < TitanMKD> dongs yes there's mask on each via
2012-10-21T19:10:02 < dongs> yeah thats like.. normal
2012-10-21T19:10:05 < TitanMKD> but yes they are far from perfect too ;)
2012-10-21T19:10:57 < TitanMKD> but i just want no short for such proto board
2012-10-21T19:15:19 < qyx_> and whats the shipping time in seeed?
2012-10-21T19:15:57 < qyx_> they only mention 4-6 days to manufacture
2012-10-21T19:16:21 < dongs> typical china trash
2012-10-21T19:16:26 < dongs> 2-3 weeks or whatever
2012-10-21T19:16:53 < qyx_> ah
2012-10-21T19:16:57 < TitanMKD> no it was 6days + 3days shipping
2012-10-21T19:17:12 < TitanMKD> but i paid for that because i was buying other stuff
2012-10-21T19:17:13 < qyx_> i will rather pay 30e here to have it in 5 days
2012-10-21T19:17:36 < TitanMKD> qyx_ but what the price of PCB ?
2012-10-21T19:17:38 < dongs> and have proper vias too
2012-10-21T19:18:09 < TitanMKD> anyway if you know cheap PCB manufacturer with good quality i'm interested ;)
2012-10-21T19:18:17 < qyx_> 30e for 1dm^2 surface
2012-10-21T19:18:27 < qyx_> for prototyping
2012-10-21T19:18:34 < TitanMKD> and 2 layers ?
2012-10-21T19:18:47 < qyx_> 2 layers, 2 silkscreens, 2 solder masks
2012-10-21T19:18:58 < TitanMKD> but it is for USA only ?
2012-10-21T19:19:03 < qyx_> central EU
2012-10-21T19:19:25 < TitanMKD> as for Europe i avoid USA manufacturer they are too expensive for export VAT about 30% !!!
2012-10-21T19:19:32 < TitanMKD> qyx_ what's the URL ?
2012-10-21T19:20:12 < qyx_> http://www.pcb-top.com/
2012-10-21T19:20:22 < qyx_> but that page looks weird
2012-10-21T19:20:37 < TitanMKD> thanks
2012-10-21T19:20:45 < TitanMKD> does they provide also 100% Etest ?
2012-10-21T19:21:23 < qyx_> their capabilities http://www.pcb-top.com/capabilities.php
2012-10-21T19:22:32 < TitanMKD> strange there's no online form to order stuff ?
2012-10-21T19:23:17 < qyx_> i place orders by email
2012-10-21T19:24:08 < TitanMKD> qyx_ have you some photo of pcb ?
2012-10-21T19:24:44 < qyx_> mmt
2012-10-21T19:24:56 < qyx_> also look at http://www.pragoboard.cz/en/o_firme
2012-10-21T19:25:05 < qyx_> but there are plenty of pcb manufacturers
2012-10-21T19:25:12 < qyx_> in eu
2012-10-21T19:25:20 < qyx_> you just have to google
2012-10-21T19:25:34 < TitanMKD> yes plenty but 10times more expensive than Seeed/Itead ;)
2012-10-21T19:25:58 < TitanMKD> I have even some manufacturer at less than 5km of my home but they just ask 300euros for 3 PCB !!
2012-10-21T19:26:52 < qyx_> last time i ordered 15pcs 80x50mm for ~200e incl vat & UPS shipping
2012-10-21T19:26:52 < dongs> yeah, i found one like that here too
2012-10-21T19:27:03 < dongs> they quoted me something like $4500 for a bunch of prototype boards.
2012-10-21T19:27:12 < dongs> i was gonna call them back and ask how long of a runway do I need next ot my house
2012-10-21T19:27:19 < dongs> and  what kinda private jet its being delivered by
2012-10-21T19:28:01 < TitanMKD> dongs hehe yes it is the problem for prototype local manufacturer are often not interested and give crazy price
2012-10-21T19:28:14 < TitanMKD> and it is only interesting for a batch of 500 PCB
2012-10-21T19:29:14 < TitanMKD> For my part i have found a good SMT stencil (laser) here http://www.smtstencil.co.uk
2012-10-21T19:29:22 < TitanMKD> fast and very good quality
2012-10-21T19:29:49 < TitanMKD> for prototype only too ;)
2012-10-21T19:30:53 < qyx_> $18 per 21x22cm sheet
2012-10-21T19:31:54 < Laurenceb_> TitanMKD: thats shruldu- from ##electronics
2012-10-21T19:33:35 < TitanMKD> qyx_ if you have better price for sheet i want ;)
2012-10-21T19:33:35 < dongs> that polymer shit?
2012-10-21T19:33:41 < dongs> that was another "cheap" stencil service I tried
2012-10-21T19:33:42 < TitanMKD> yes polymer of course
2012-10-21T19:33:46 < dongs> went straight to the bin
2012-10-21T19:33:48 < dongs> just like the kapton shit
2012-10-21T19:33:50 < dongs> fucking useless.
2012-10-21T19:33:54 < dongs> didnt even bother pasting it.
2012-10-21T19:34:14 < TitanMKD> like i said it's for small prototypes
2012-10-21T19:34:28 < TitanMKD> if you plan to build more than 20 board with that it's not the right stuff ;)
2012-10-21T19:34:44 < qyx_> http://imgur.com/TSpRU
2012-10-21T19:35:17 < TitanMKD> qyx_ has yes really not bad
2012-10-21T19:35:30 < TitanMKD> qyx_ and the via are center and seems very clean
2012-10-21T19:36:21 < qyx_> vias are ok, these are 0.4mm drill
2012-10-21T19:36:26 < qyx_> they are able to do 0.2mm
2012-10-21T19:36:36 < qyx_> http://imgur.com/Pnk6W
2012-10-21T19:37:04 < qyx_> mcp9808 breakout boards for soldering to cable
2012-10-21T19:37:26 < TitanMKD> it's panelling stuff ?
2012-10-21T19:41:23 < qyx_> yes, thats the only way to do such small pcbs
2012-10-21T19:49:27 < TitanMKD> qyx_ anyway for stuff like that http://i.imgur.com/Pnk6W.jpg ITead/Seeed do the job very well ;)
2012-10-21T19:53:20 < emeb_mac> dongs: parts ordered. backordered an F3 discovery while I was at it...
2012-10-21T19:56:46 < dongs> and told them to ship f3 separately?
2012-10-21T20:01:04 < zyp> whoops, fuck
2012-10-21T20:01:11 < zyp> bricked my BMP
2012-10-21T20:02:15 < Laurenceb_> unbrick it with a discovery?
2012-10-21T20:02:33 < zyp> yeah
2012-10-21T20:02:36 < Laurenceb_> emeb_mac: there will be a long wait for ne f3discoveries
2012-10-21T20:02:58 < zyp> it's just a bother
2012-10-21T20:03:07 < zyp> since I have it heatshrinked and all
2012-10-21T20:03:28 < emeb_mac> dongs: DK said they'd have them on 10-31
2012-10-21T20:03:35 < emeb_mac> so prolly 2 weeks?
2012-10-21T20:04:18 < emeb_mac> Laurenceb_: why long wait?
2012-10-21T20:04:40 < Laurenceb_> oh
2012-10-21T20:04:54 < Laurenceb_> farnell/RS were saying several months
2012-10-21T20:04:58 < Laurenceb_> who is DK?
2012-10-21T20:05:05 < emeb_mac> Digi-Key
2012-10-21T20:05:33 < emeb_mac> The Borg of Thief River Falls.
2012-10-21T20:05:47 < zyp> https://github.com/gsmcmullin/blackmagic/commit/241bf76 <- right, my BMP is older than this commit, so I guess this is what fucked up
2012-10-21T20:06:23 < zyp> (and also the fact that I tried to write an .elf directly to flash and not the extracted .bin)
2012-10-21T20:06:51 < Laurenceb_> http://uk.rs-online.com/web/p/processor-microcontroller-development-kits/7692091/?searchTerm=stm32f3discovery&relevancy-data=636F3D3226696E3D4931384E4B6E6F776E41734D504E266C753D656E266D6D3D6D61746368616C6C7061727469616C26706D3D5E5C772B2426706F3D313326736E3D592673743D4D414E5F504152545F4E554D424552267573743D73746D33326633646973636F766572792677633D424F544826
2012-10-21T20:06:57 < Laurenceb_> still stock
2012-10-21T20:07:02 < zyp> I guess the file turned out too large and something wrapped, causing it to overwrite the start of the flash
2012-10-21T20:07:11 -!- mase [~mase@kemp-5d8609af.pool.mediaWays.net] has quit [Ping timeout: 268 seconds]
2012-10-21T20:07:22 < Laurenceb_> wow 1 left
2012-10-21T20:07:26 < Laurenceb_> order fast
2012-10-21T20:07:51 < Laurenceb_> dongs: did you get your pink boards?
2012-10-21T20:09:22 < emeb_mac> website derp derp: header says F0Discovery, fine print says F3.
2012-10-21T20:09:45 < emeb_mac> zyp: so that's what killed the bootloader?
2012-10-21T20:10:35 < zyp> well, I was flashing some different stuff, trying to figure out why the newest version wouldn't enumerate
2012-10-21T20:10:55 < zyp> and then I flashed an elf instead of a bin by mistake
2012-10-21T20:11:15 < zyp> it hit me when it took much longer to flash than the other stuff I tried
2012-10-21T20:11:34 < zyp> and then the dfu stuff suddenly failed
2012-10-21T20:11:43 < zyp> and now bootloader is dead :p
2012-10-21T20:16:43 < Laurenceb_> wtf is CPAL  ?
2012-10-21T20:17:07 < dongs> some bloated i2c driver from ST
2012-10-21T20:17:16 < dongs> with about 4 layers of abstraction
2012-10-21T20:17:52 < Laurenceb_> ah
2012-10-21T20:18:09 < Laurenceb_> where is the proper download location for st periph lib?
2012-10-21T20:18:23 < dongs> any of the 'resources' shits off the chip you wanna get it for
2012-10-21T20:18:36 < dongs> the link is asme i believe, filename doesnt change with new version
2012-10-21T20:19:04 < Laurenceb_> http://www.st.com/stonline/stappl/resourceSelector/app?page=resourceSelectorPage&doctype=st_software_rsrc&SubClassID=1605
2012-10-21T20:19:15 < Laurenceb_> the periph lib is in there
2012-10-21T20:19:26 < Laurenceb_> but thats a seriously disorganised way to release it
2012-10-21T20:19:42 < dongs> http://www.st.com/internet/com/SOFTWARE_RESOURCES/SW_COMPONENT/FIRMWARE/stm32f37x_dsp_stdperiph_lib.zip yeah
2012-10-21T20:19:42 < Laurenceb_> AN4046 has the periph lib
2012-10-21T20:19:49 < Laurenceb_> ok thanx
2012-10-21T20:19:55 < dongs> its right on that page you fgt
2012-10-21T20:20:15 < dongs> ohh they have F3 clock tool
2012-10-21T20:24:14 < Laurenceb_> thats 37x
2012-10-21T20:24:21 < Laurenceb_> im interested in 30x
2012-10-21T20:24:33 < dongs> http://www.st.com/internet/com/SOFTWARE_RESOURCES/SW_COMPONENT/FIRMWARE/stm32f30x_dsp_stdperiph_lib.zip
2012-10-21T20:24:42 < dongs> sorry, couldnt resist
2012-10-21T20:24:45 < dongs> and no its not there.e
2012-10-21T20:24:50 < dongs> its included in f3discovery firmware.
2012-10-21T20:31:03 -!- barthess [~barthess@178.154.63.178] has joined ##stm32
2012-10-21T20:33:35 < emeb_mac> dongs: ordered up some F373CC.
2012-10-21T20:33:56 < emeb_mac> will see about getting this breakout going.
2012-10-21T20:40:05 < Laurenceb_> yeah i have it
2012-10-21T20:40:12 < Laurenceb_> it just seems disorganised
2012-10-21T20:46:04 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-21T20:50:19 < Laurenceb_> doe sF3 have DFU?
2012-10-21T20:50:30 < Laurenceb_> i see it in the ref manual
2012-10-21T20:50:32 < Laurenceb_> but no proper info
2012-10-21T20:54:23 < emeb_mac> I assume it does, but AN2606 is older than F3 so it's not specifically described.
2012-10-21T20:55:28 < emeb_mac> easy way to check though - hook up USB and try reset w/ BOOT0 = VDD
2012-10-21T20:58:01 < Laurenceb_> yeah
2012-10-21T20:58:09 < Laurenceb_> i presume it needs a clock tho
2012-10-21T20:59:08 < emeb_mac> HFE xtal @ 8, 12, 25 IIRC - it uses HSI to measure the HFE speed so you don't need a specific frequency.
2012-10-21T20:59:16 < Laurenceb_> neat
2012-10-21T20:59:29 < Laurenceb_> but dicovery has clk input
2012-10-21T20:59:39 < Laurenceb_> ill probably need to stick an xtal on
2012-10-21T21:00:19 < dongs> just plug in usb to stlink
2012-10-21T21:00:25 < dongs> by default its wired to mco from that
2012-10-21T21:01:15 -!- tavish [~tavish@59.180.220.129] has joined ##stm32
2012-10-21T21:01:15 -!- tavish [~tavish@59.180.220.129] has quit [Changing host]
2012-10-21T21:01:15 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-21T21:02:13 < emeb_mac> cute.
2012-10-21T21:03:41 < Laurenceb_> yeah but exernal clk != xtal?
2012-10-21T21:03:55 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 260 seconds]
2012-10-21T21:04:47 < Laurenceb_> http://phx.dk/pico.html
2012-10-21T21:04:53 < Laurenceb_> looks like a slow sort of hell
2012-10-21T21:11:09 < Laurenceb_> is there an example project with TNT anywhere?
2012-10-21T21:13:37 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 246 seconds]
2012-10-21T21:15:27 < emeb_mac> HSE input can't tell if the signal comes from an xtal or another oscillator. clk ~= xtal.
2012-10-21T21:20:40 < emeb_mac> why is it whenever I see "DFU" my brain substitutes "STFU"?
2012-10-21T21:22:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-21T21:24:08 -!- barthess [~barthess@178.154.63.178] has quit [Read error: Connection reset by peer]
2012-10-21T21:24:19 -!- barthess [~barthess@178.154.69.53] has joined ##stm32
2012-10-21T21:30:43 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-165.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-21T21:38:57 <+Steffanx> I guess that's called dongs-disease
2012-10-21T21:57:06 < Laurenceb_> /tmp/ccgFoOEX.o: In function `LoopFillZerobss':
2012-10-21T21:57:06 < Laurenceb_> (.text.Reset_Handler+0x26): undefined reference to `SystemInit'
2012-10-21T21:57:06 < Laurenceb_> /tmp/ccgFoOEX.o: In function `LoopFillZerobss':
2012-10-21T21:57:06 < Laurenceb_> (.text.Reset_Handler+0x2a): undefined reference to `main'
2012-10-21T21:57:39 < Laurenceb_> what have i done wrong?
2012-10-21T22:02:29 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Quit: Leaving.]
2012-10-21T22:03:25 < Laurenceb_> ok its the ride7 start code
2012-10-21T22:03:36 < Laurenceb_> it cant see the other headers
2012-10-21T22:03:42 < Laurenceb_> now do i fix this?
2012-10-21T22:08:50 < Laurenceb_> after all your time lurking in your mums basement
2012-10-21T22:09:21 < Tectu> so has anyone of you guy seen the movie "Symbol" from 2009?
2012-10-21T22:11:05 < Laurenceb_> so, how do i #include in a .s asm file?
2012-10-21T22:11:28 < Tectu> Laurenceb, you don't
2012-10-21T22:11:32 < Tectu> Laurenceb, you give it to you linker
2012-10-21T22:11:41 < Tectu> compiler.
2012-10-21T22:11:45 < Tectu> assembler
2012-10-21T22:11:46 < Tectu> or what ever
2012-10-21T22:12:09 < Tectu> if you need a symbol in a .s asm file, you use «extern»
2012-10-21T22:13:26 < Laurenceb_> hmm
2012-10-21T22:13:57 < Laurenceb_> yeah my F1 startup code doesnt have any main or systeminit extern
2012-10-21T22:14:10 < Laurenceb_> its looks almost the same as this startup code
2012-10-21T22:15:28 < jpa-> why would you use assembler anyway?
2012-10-21T22:15:46 < Laurenceb_> reuse ride7 code
2012-10-21T22:15:48 < Laurenceb_> EXTERN(__cs3_start_c main __cs3_stack __cs3_heap_end)
2012-10-21T22:15:55 < Laurenceb_> i have that in my F1 project
2012-10-21T22:17:25 < Laurenceb_> that didnt fix it
2012-10-21T22:21:11 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-21T22:26:25 -!- tavish_ [~tavish@59.177.12.76] has joined ##stm32
2012-10-21T22:27:48 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 248 seconds]
2012-10-21T22:34:47 < Laurenceb_> nvm
2012-10-21T22:34:56 < Laurenceb_> looks like i just need assert param adding now
2012-10-21T22:40:23 -!- tavish_ [~tavish@59.177.12.76] has quit [Quit: leaving]
2012-10-21T22:46:22 < Laurenceb_> but why is my startup file deleted when i make clean?
2012-10-21T22:48:30 < Laurenceb_> nvm
2012-10-21T22:48:42 < Laurenceb_> lib/STM32F30x_StdPeriph_Driver/src/stm32f30x_flash.o: In function `FLASH_ErasePage':
2012-10-21T22:48:42 < Laurenceb_> stm32f30x_flash.c:(.text.FLASH_ErasePage+0x12): undefined reference to `assert_param'
2012-10-21T22:48:45 < Laurenceb_> grumble
2012-10-21T23:14:14 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-21T23:14:30 -!- elektrinis [~circuit@78.60.169.125] has quit [Ping timeout: 246 seconds]
2012-10-21T23:16:58 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 246 seconds]
--- Day changed Mon Oct 22 2012
2012-10-22T00:36:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-22T00:45:18 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-165.pptp.stw-bonn.de] has joined ##stm32
2012-10-22T00:53:55 < upgrdman> with the f4, trying out spi, im getting some really weird waveforms on MOSI and MISO: http://www.farrellf.com/temp/stm32f4_spi_weird_mosi_miso.png     any ideas?
2012-10-22T00:57:43 < Thorn> at first I though the host was farnell.com
2012-10-22T00:58:41 < upgrdman> :)
2012-10-22T01:00:23 < emeb_mac> sorry - not a clue what you're doing there. You sure you have the clk/mosi pins right?
2012-10-22T01:01:12 < upgrdman> tripple checked. and since SCLK looks the most periodic, it makes sense?
2012-10-22T01:01:29 < Thorn> not with this frequency
2012-10-22T01:01:47 < Thorn> you sure it's not aliasing or something?
2012-10-22T01:02:18 < upgrdman> i went to the smalled timescale and it looks the same
2012-10-22T01:02:18 <+Steffanx> emeb_mac, "why is it whenever I see "DFU" my brain substitutes "STFU"?" <= dongs-disease
2012-10-22T01:02:26 < upgrdman> whats dfu
2012-10-22T01:02:43 < Thorn> dong firmware upgrade
2012-10-22T01:03:37 < BrainDamage> a protocol to upgrade firmwares trough usb
2012-10-22T01:04:40 < emeb_mac> dongs disease sounds like an STD.
2012-10-22T01:04:42 <+Steffanx> but ignore me and continue the debugging upgrdman :)
2012-10-22T01:04:42 < Thorn> how essential is a usb power control IC for a host interface? I'm not going to plug anything other than flash sticks there (probably)
2012-10-22T01:04:53 < upgrdman> the weird thing about the spi problem is that MISO is around 2V ... not 3V or 0V
2012-10-22T01:04:53 < emeb_mac> *there I go again*
2012-10-22T01:05:55 -!- barthess [~barthess@178.154.69.53] has quit [Ping timeout: 244 seconds]
2012-10-22T01:06:50 < BrainDamage> meh if you want to troll people, you can link my h bridge masterpiece, it doubles as abstract art http://dl.dropbox.com/u/40949413/20121021_223901.jpg
2012-10-22T01:07:20 <+Steffanx> like a pro
2012-10-22T01:07:56 <+Steffanx> BrainDamage plays gitar?
2012-10-22T01:08:03 < BrainDamage> yes
2012-10-22T01:10:06 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has quit [Excess Flood]
2012-10-22T01:10:30 < Thorn> I made a led strip driver on a stripboard yesterday that doesn't look much better
2012-10-22T01:10:52 < BrainDamage> does it have to withstand 1.5kW too?
2012-10-22T01:11:02 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has joined ##stm32
2012-10-22T01:12:23 <+Steffanx> Of course, one LED/driver
2012-10-22T01:12:26 < Thorn> 200mA per channel for the length of led strip that I currently have.
2012-10-22T01:44:11 < Laurenceb_> http://www.ebay.co.uk/itm/19-Little-Folk-1982-TEDDY-BEAR-Jointed-w-Chain-G-McBride-England-/271023955215?pt=US_Dolls_Bears_Toys&hash=item3f1a495d0f
2012-10-22T01:44:13 < Laurenceb_> bro bear
2012-10-22T01:46:16 < zyp> played a bit around with L1discovery, I see that it also is shipped without HSE crystal, with option to get 8MHz from stlink instead
2012-10-22T01:46:28 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-22T01:47:24 < zyp> and whoever claimed that L1discovery doesn't have an usb connector due to those pins being used for lcd were wrong
2012-10-22T01:47:50 < zyp> according to schematics, usb pins are unused and broken out
2012-10-22T01:49:30 < Laurenceb_> i cant get my F3 code to compile
2012-10-22T01:50:22 < Laurenceb_> stm32f30x_flash.c:(.text.FLASH_ErasePage+0x12): undefined reference to `assert_param'
2012-10-22T01:54:03 < Laurenceb_> which is defined in the conf.h file
2012-10-22T01:54:20 < Laurenceb_> all the other perip lib headers seem to link ok
2012-10-22T02:08:54 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-22T02:09:00 < Laurenceb_> arg wtf
2012-10-22T02:09:01 < Laurenceb_> lib/STM32F30x_StdPeriph_Driver/inc/stm32f30x_usart.h:587:48: error: expected ')' before '*' token
2012-10-22T02:09:07 < Laurenceb_> time a bizillion
2012-10-22T02:09:11 < Laurenceb_> *times
2012-10-22T02:14:59 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-22T02:34:14 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Quit: Tzacks]
2012-10-22T02:44:18 -!- DaKu is now known as daku
2012-10-22T03:05:06 -!- Erlkoenig [~erlkoenig@pptp-212-201-79-165.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-22T03:11:33 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-22T03:30:02 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-22T03:35:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-22T04:17:50 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-22T04:22:30 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has left ##stm32 []
2012-10-22T04:34:04 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-22T04:41:34 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-22T05:17:03 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-22T05:54:14 < dongs> emeb_mac: your trash shipped
2012-10-22T05:54:52 < emeb_mac> dongs: sweet - what do I owe you?
2012-10-22T05:55:38 < dongs> ill figure it out once the bill from post comes back
2012-10-22T05:55:57 < emeb_mac> ok - let me know.
2012-10-22T05:55:57 -!- upgrdman_ [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-22T05:56:17 < emeb_mac> got the rest of the parts ordered this morning.
2012-10-22T06:35:47 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-22T06:35:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-22T06:41:48 < emeb_mac> except for the pfets. grrrr.
2012-10-22T06:42:12 < dongs> did they s hip yet?
2012-10-22T06:42:16 < emeb_mac> no
2012-10-22T06:42:20 < dongs> well, then you can just add.
2012-10-22T06:42:25 < dongs> chat up their livehelp stuff.
2012-10-22T06:42:29 < emeb_mac> probably not until tomorrow
2012-10-22T06:42:29 < dongs> super easy.
2012-10-22T06:42:37 < dongs> live help is 24h
2012-10-22T06:42:50 < emeb_mac> got a PN for the pfets?
2012-10-22T06:43:03 < dongs> lemme see if i wrote it
2012-10-22T06:43:13 < dongs> DMG3415U according to my schematic
2012-10-22T06:43:21 < dongs> lets seeif thats even waht we ended upw ith
2012-10-22T06:43:39 < dongs> o, ya
2012-10-22T06:43:41 < dongs> that soubnds right
2012-10-22T06:43:47 < dongs> DMG3415UDICT-ND
2012-10-22T06:44:02 < dongs> i just used my lefotver NDS356AP
2012-10-22T06:44:21 < dongs> and that semeed to work
2012-10-22T06:44:24 < dongs> they're cheaper as well. shrug
2012-10-22T06:46:43 -!- upgrdman_ [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Quit: Ex-Chat]
2012-10-22T06:55:00 < emeb_mac> OK - 30x of DMG3415UDICT-ND added - they're cheaper than the other at qty > 10
2012-10-22T06:55:33 < emeb_mac> by a few cents. whoopee...
2012-10-22T06:56:30 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-22T06:58:31 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-22T07:02:30 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-22T07:05:52 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-22T07:44:49 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 256 seconds]
2012-10-22T08:28:05 -!- claude is now known as Claude
2012-10-22T08:29:06 < emeb_mac> having big fun bringing my bare-bones developed code into CooCox.
2012-10-22T08:29:55 < emeb_mac> It's not bad, but their default CMSIS_BOOT files don't deal with all the PLL setup that I need, so it has to be augmented.
2012-10-22T08:30:13 < emeb_mac> otherwise just a straight copy.
2012-10-22T08:32:29 < dongs> wut
2012-10-22T08:32:38 < dongs> oh the IDE?
2012-10-22T08:32:41 < dongs> never used it
2012-10-22T08:32:58 < dongs> i just use thier OS from keil
2012-10-22T08:34:07 < ds2> everyone here uses the IDE?
2012-10-22T08:35:05 < emeb_mac> no
2012-10-22T08:35:37 < emeb_mac> I've been developing using straight gcc/editor/gdb on Linux.
2012-10-22T08:35:55 < emeb_mac> got a client who wants to use CooCox though, so setting that up. Works OK
2012-10-22T08:36:00 < ds2> plain Linero GCC?
2012-10-22T08:36:22 < ds2> or you using the GCC based setup on the ST's tool list?
2012-10-22T08:36:30 < emeb_mac> started with CS GCC, now using launchpad/ARM
2012-10-22T08:37:48 < emeb_mac> CooCox is just your basic Eclipse+GCC for Windows. Not sure who's maintaining it.
2012-10-22T08:42:20 < dongs> my time is usually not worthless so i prefer to get work done wiht IDE
2012-10-22T09:25:59 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32
2012-10-22T09:27:55 < dongs> http://www.navilock.de/produkte/F_812_u-blox-6_61846/merkmale/2.html huh? ublox makes chips now?
2012-10-22T09:28:01 < dongs> how'd these germans get nonmodule version
2012-10-22T09:31:46 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Quit: pokðt]
2012-10-22T09:38:46 < zyp> of course ublox makes chips
2012-10-22T09:39:14 < zyp> you're not going to get any phone manufacturers to make room for a huge module nowadays
2012-10-22T09:39:48 < dongs> ah so.
2012-10-22T09:40:17 < dongs> what do they do anyway? buy some atmel shit and laser their label on it?
2012-10-22T09:40:22 < dongs> surely its not thier own silicon right?
2012-10-22T09:45:09 < zyp> I was of the understanding that they have their own silicon, why wouldn't they?
2012-10-22T09:45:42 < dongs> duno
2012-10-22T09:45:50 < R2COM> ...
2012-10-22T09:45:56 < dongs> shit like xbee just uses atmel sam7s
2012-10-22T09:46:00 < dongs> wiht sekret codez
2012-10-22T09:46:10 < dongs> and they dont evne bother lasering the label off
2012-10-22T09:46:19 < zyp> http://www.u-blox.com/en/gps-chips/stand-alone-gps-chips.html
2012-10-22T09:47:56 < dongs> ya i know
2012-10-22T09:49:33 < dongs> but didnt know until about 10mins ago though
2012-10-22T10:07:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-22T10:13:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 244 seconds]
2012-10-22T10:26:38 -!- Claude is now known as claude
2012-10-22T10:40:09 < zyp> also, how would antenova be able to make ublox based modules without ublox chips?
2012-10-22T10:41:29 -!- claude is now known as claude_work
2012-10-22T10:44:35 < dongs> ya ya
2012-10-22T11:14:18 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-22T11:20:16 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-22T11:20:23 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-22T11:20:47 < jpa-> systick timer increments the nuttx tick count every 1 ms. I want to get 64-bit timestamps with 1us precision. Basically, just tickcount * 1000 + (uint64_t)NVIC->SYSTICK_CURRENT * 1000000/ HCLK_FREQ.
2012-10-22T11:20:52 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-10-22T11:21:18 < jpa-> but any good ideas on how to handle timer overflows when in the middle of that computation?
2012-10-22T11:23:41 < zyp> don't
2012-10-22T11:24:50 < jpa-> so, what shall i do instead?
2012-10-22T11:25:14 < zyp> enable DWT_CYCCNT and use it as a freerunning counter, then do += (cyccnt - last_cyccnt) * 1000000 / HCLK_FREQ;
2012-10-22T11:26:09 < jpa-> hmm, isn't that reserved for some profiling tools etc?
2012-10-22T11:26:59 < zyp> not really
2012-10-22T11:27:41 < jpa-> but yeah, that'll be easier
2012-10-22T11:28:09 < zyp> I'm planning to do my timekeeping this way, except that instead of dictating 1us precision, I'll just count cycles with the 64-bit timestamp
2012-10-22T11:28:29 < zyp> even at 168 MHz, it won't overflow for 3500 years or so
2012-10-22T11:28:37 < jpa-> could use a regular timer also, except i have no 32 bit timers free
2012-10-22T11:29:16 < jpa-> yeah, i'm using microseconds just because these timestamps are exposed as sample times; better not tie those to clock frequency
2012-10-22T11:29:34 < zyp> my argument is that I would like to keep it portable to other cortex-m chips
2012-10-22T11:30:33 < zyp> well, just have utility functions to convert to/from microseconds?
2012-10-22T11:33:35 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-22T11:33:38 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-22T11:35:31 < jpa-> zyp: yeah, sure, except no point when i'm only interested in microseconds :)
2012-10-22T12:08:19 < zyp> I'd like to avoid unnecessary divisions (by HCLK), so unless I need to know the time in microseconds I'd rather just keep it in cycles
2012-10-22T12:08:55 < zyp> and since I mainly will use it for stuff like sleep timing, I can rather just multiply the number of microseconds I need to sleep with how many cycles I have per microsecond
2012-10-22T12:18:17 -!- cjbaird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-22T12:18:17 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Read error: Connection reset by peer]
2012-10-22T12:25:17 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-22T12:25:17 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-22T12:25:17 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-22T12:30:33 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-22T12:33:25 < dongs> lol, fucking TI
2012-10-22T12:33:40 < dongs> since they ate up natsemi, tehy dont include their parts into their parametric search anymore
2012-10-22T12:42:53 < dongs> damn this proj needs 1.2A on 1.5V I was gonna use a LDO but that sounds like a bad idea
2012-10-22T12:43:22 < dongs> but at same time I dont wanna deal wiht filtering shitty switching supply outputs
2012-10-22T12:45:50 < dongs> maybe I'll get a high current 3.3V switcher and LDO shit off that.
2012-10-22T12:47:58 <+Steffanx> What's the problem dongs? Use the par. search from TI ?
2012-10-22T12:48:10 < dongs> Steffanx: it doesnt show Natsemi parts, lol
2012-10-22T12:48:12 < dongs> thats the problem
2012-10-22T12:48:18 <+Steffanx> It does afaik..
2012-10-22T12:48:29 < dongs> not to me.. I have a LP3962 I used in other proj
2012-10-22T12:48:34 < dongs> and I looked for similar specs, ldo etc
2012-10-22T12:48:38 < dongs> nothign only lame TPS parts
2012-10-22T12:49:02 < dongs> then I tried some other params.. still cant make any LP* parts to appear.
2012-10-22T12:49:12 < Laurenceb_> https://github.com/Laurenceb/STM32F30x-SkeleProj
2012-10-22T12:49:18 < Laurenceb_> a little something...
2012-10-22T12:49:48 < dongs> stop committing .o files fgt
2012-10-22T12:49:58 <+Steffanx> Hmm, you can make the lpv511/521 appear, not that's useful for you though :P
2012-10-22T12:50:27 < dongs> thats not even a natsmi part
2012-10-22T12:50:35 <+Steffanx> It is
2012-10-22T12:50:48 < dongs> oh it is
2012-10-22T12:50:57 <+Steffanx> 'rebranded' datasheet :)
2012-10-22T12:51:06 < dongs> ruby to rails
2012-10-22T12:51:45 < dongs> looks like another proj I just used LM1117-ADJ's...
2012-10-22T12:51:52 < dongs> maybe ill jus do that lol
2012-10-22T12:52:26 < dongs> then I can just hookup a bunch of htem in parallel
2012-10-22T12:52:27 < dongs> and be like
2012-10-22T12:52:33 < dongs> HANDLE THIS CURRENT PLZ
2012-10-22T12:53:08 < dongs> 5V	980mA / 1.5V	1260mA /  3.3V	21mA / 2.5V	280mA
2012-10-22T12:53:11 < dongs> shitty fucking requirements.
2012-10-22T12:53:25 < Laurenceb_> thats a lot of power
2012-10-22T12:53:38 < zyp> well, most of the power is on 5V
2012-10-22T12:53:49 <+Steffanx> 1.5V ..
2012-10-22T12:53:59 <+Steffanx> nvm :)
2012-10-22T12:54:06 <+Steffanx> fail fail fail
2012-10-22T12:54:13 <+Steffanx> but it's monday morning
2012-10-22T12:54:27 < zyp> no it's lunchtime
2012-10-22T12:54:35 <+Steffanx> It's still morning :)
2012-10-22T12:55:09  * Laurenceb_ off
2012-10-22T12:55:22 <+Steffanx> gn
2012-10-22T12:56:42 < jpa-> hmm.. i want to use C++ static constructors to register a bunch of objects to a list; i have the cxxinitialize() stuff ok, but GCC optimizes away my classes if i never refer to them from another file
2012-10-22T12:56:56 < jpa-> even though there is a static instance defined in the same file
2012-10-22T12:57:11 < dongs> http://www.mbl.is/frimg/6/39/639812.jpg attn Laurenceb_
2012-10-22T12:59:12 < jpa-> ah, http://stackoverflow.com/questions/1229430/how-do-i-prevent-my-unused-global-variables-being-compiled-out
2012-10-22T13:00:11 <+Steffanx> stackoverflow .. the source of all your solutions :)
2012-10-22T13:00:22 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 246 seconds]
2012-10-22T13:01:19 < karlp> something odd... https://github.com/xatier/qemu
2012-10-22T13:01:36 < karlp> (qemu support for stm32)
2012-10-22T13:02:19 < karlp> what was that doing on mbl.is dongs?
2012-10-22T13:05:29 < qyx_> virtualised stm32 running on stm32?
2012-10-22T13:07:00 < karlp> I'd assume on x86 actually,
2012-10-22T13:19:27 < dongs> finally, you'll be able to run lunix on stm32.
2012-10-22T13:20:35 < qyx_> it should be possible now, there is cm3 port
2012-10-22T13:20:56 <+Steffanx> Good project to waste your time on
2012-10-22T13:21:56 < zyp> jpa-, sounds like a bad linker script
2012-10-22T13:21:58 < dongs> http://en.wikipedia.org/wiki/Cave_of_the_Negroes_incident
2012-10-22T13:22:44 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Read error: Connection reset by peer]
2012-10-22T13:22:47 <+Steffanx> dongs ..........
2012-10-22T13:23:04 < zyp> karlp, qemu already have support for cortex-m3, so I guess they just added stm32 peripherals?
2012-10-22T13:23:06 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-22T13:32:25 -!- Tectu is now known as tectu
2012-10-22T13:32:36 < jpa-> zyp: actually it was the -whole-archive stuff
2012-10-22T13:32:47 < dongs> tectu: stop changing your nick
2012-10-22T13:32:59 < tectu> dongle, okay
2012-10-22T13:33:01 -!- tectu is now known as Tectu
2012-10-22T13:33:24 < zyp> ah
2012-10-22T13:33:38 < zyp> oh
2012-10-22T13:34:22 < zyp> -whole-archive would probably have solved the issue I had with using archives and weak symbols
2012-10-22T13:34:23 <+Steffanx> Tectu, just curious.. why you do that?
2012-10-22T13:35:17 < dongs> what teh fuck is TO-xxx for DPAK
2012-10-22T13:35:22 < dongs> TO-252 right?
2012-10-22T13:36:13 < Tectu> Steffanx, what exactly? I just logged in to my bouncer, I don't see any nickchange beside the last  tectu -> Tectu  here
2012-10-22T13:36:17 < Tectu> and dongs said that before
2012-10-22T13:36:27 < dongs> 19:32 -!- Tectu is now known as tectu
2012-10-22T13:36:30 < dongs> yo0u keep doing this shit
2012-10-22T13:36:37 < dongs> sometimes like several times per day
2012-10-22T13:36:42 < Tectu> wtf
2012-10-22T13:36:46 <+Steffanx> I just wonder what the idea behind that t-T-t-T is ( claude does that too )
2012-10-22T13:36:54 <+Steffanx> c-C-c-C-c etc.
2012-10-22T13:37:04 < dongs> it must be some luniex feature
2012-10-22T13:37:37 < Tectu> well, most people do it this way:   if nick is in written with first letter capital, their online, otherwise offline. saidly, some do it exactly the other way arround
2012-10-22T13:37:37 < Tectu> let me fix
2012-10-22T13:37:54 < dongs> what thte hell
2012-10-22T13:37:55 -!- Tectu is now known as tectu
2012-10-22T13:38:04 -!- tectu is now known as Tectu
2012-10-22T13:38:04 < dongs> ive been on irc since before you were born
2012-10-22T13:38:05 < dongs> and I've Never heard of this
2012-10-22T13:38:05 < Tectu> better?
2012-10-22T13:38:14 <+Steffanx> Ha, Dongs
2012-10-22T13:38:47 < zyp> Tectu, that's just dumb
2012-10-22T13:38:59 <+Steffanx> My client supports the away-feature :P
2012-10-22T13:39:26 < zyp> I don't get why it would matter either way
2012-10-22T13:39:42 < Tectu> zyp, i don't try to do that
2012-10-22T13:40:00 < Tectu> zyp, it looks like this only happens when i log in from my workstation, not from my notebook
2012-10-22T13:42:02 < dongs> fapstation more like
2012-10-22T13:42:28 <+Steffanx> We'll hate your forever for this Tectu
2012-10-22T13:43:02 < zyp> also, lol bouncer
2012-10-22T13:43:07 < Tectu> wtf
2012-10-22T13:44:16 < dongs> man so many LM1117  clones and NONE wiht enable
2012-10-22T13:44:20 < dongs> what the hell is this shit
2012-10-22T13:44:25 < cjbaird> Reconnection with a different nick. It happened here with cjbaird/cjbaird_/cjbaird__/baird
2012-10-22T13:44:37 -!- cjbaird is now known as baird
2012-10-22T13:44:39 < dongs> next try "cjbeard"
2012-10-22T13:46:55 <+Steffanx> ^^ dongs-disease
2012-10-22T13:46:56 < baird> I was about 5 when I discovered a neat little feature of my surname :) .. http://i.imgur.com/zNlgW.png
2012-10-22T13:47:13 <+Steffanx> lol
2012-10-22T13:47:50 < baird> Apparently I'm some mid-distant relative of John Logie..
2012-10-22T13:47:51 < dongs> wow.
2012-10-22T13:49:46 <+Steffanx> I bet it was a coincidence .. your handwriting skillz weren't good enough, so you wrote that
2012-10-22T13:51:18 < dongs> why teh fuck is LM317 in ldo section
2012-10-22T13:51:20 < dongs> on digikey
2012-10-22T13:51:25 < dongs> doesnt it have like 2-3V dropout
2012-10-22T13:51:51 < dongs> http://www.digikey.com/product-detail/en/LM317MDT%2FNOPB/LM317MDTNS-ND/363904
2012-10-22T13:51:53 <+Steffanx> Yay, this piece of software gives me a fake dead pixel :D
2012-10-22T13:52:26 < claude_work> Steffanx, CcCcCc ... thats my bouncer and my lazyness to proper configure it
2012-10-22T13:52:57 < dongs> I wonder what opensores retard came up wiht this >  < Tectu> well, most people do it this way:   if nick is in written with first letter capital, their online, otherwise offline.
2012-10-22T13:53:27 -!- claude_work is now known as Claude_work
2012-10-22T13:53:29 < Claude_work> ;)
2012-10-22T13:53:42 <+Steffanx> It must have been Mr. lunix himself, dongs
2012-10-22T13:54:16 < baird> IRC predates Linux by a few years..
2012-10-22T13:54:48 < dongs> i was ircing on VMS before most of you (except cjbeard) knew what irc was
2012-10-22T13:54:52 <+Steffanx> Then it was you who invented that AaAaAaA-feature
2012-10-22T13:55:08 < baird> must've
2012-10-22T13:55:18 <+Steffanx> We know you two are OLD
2012-10-22T13:55:30 < baird> lol VMS. I started on that too..
2012-10-22T13:55:30 < Tectu> dongs, dunno, I don't like it that much as well
2012-10-22T13:57:00 < dongs> http://www.micrel.com/_PDF/mic35152.pdf this looks like a winner
2012-10-22T13:57:04 < dongs> assuming micrel stays in business long enough
2012-10-22T13:57:41 < baird> I favourite VMS hack was a logout.com that put the VT230(?) terminal into graphics mode, displayed a GIF of a Mac desktop, and I left a borrowed mouse on the desk. I sat at another terminal and watched 14-15 people walk up over the rest of the afternoon and try to use it. :D
2012-10-22T13:58:08 < dongs> more like ripterm
2012-10-22T14:00:20 < zyp> dongs, http://www.globalconnectortechnology.com/connector/?series=USB3130 <- what do you think about this? :p
2012-10-22T14:00:57 < dongs> zyp: amazing.
2012-10-22T14:01:22 < dongs> next question: why the fuck is it on tape/reel
2012-10-22T14:01:36 < zyp> though, being vertical I think it will break easier than horizontal ones :p
2012-10-22T14:01:43 < dongs> right.
2012-10-22T14:01:54 < dongs> i dont think i ever seen a fully through hole microusb
2012-10-22T14:02:01 < dongs> seen one wiht through-hole tabs
2012-10-22T14:02:04 < dongs> but the pins were smt.
2012-10-22T14:02:12 < zyp> yeah, that's the kind I'm using
2012-10-22T14:02:14 < dongs> these guys dont make a right angle version?
2012-10-22T14:02:22 < zyp> apparently not
2012-10-22T14:02:25 < dongs> the pitch on micro socket is so fucked up though
2012-10-22T14:02:28 < dongs> its like 0.5mm or someshit
2012-10-22T14:02:43 < dongs> bitch to hansdsolder and rework
2012-10-22T14:03:20 < baird> Got NetBSD running on a RPi. Now doing a "make world" of /usr/src .. this'll take a while
2012-10-22T14:03:39 < dongs> i should cancel my shitberrypi order from RS japan.
2012-10-22T14:03:46 < dongs> last i checked they had ship date at like sometime in 20123.
2012-10-22T14:03:48 < dongs> 2013.
2012-10-22T14:03:48 < zyp> dongs, especially when you stick a QFN right behind it
2012-10-22T14:03:57 < dongs> zyp, yes.
2012-10-22T14:04:09 <+Steffanx> Overclocked too 1GHz first baird ?
2012-10-22T14:04:11 < zyp> and use a footprint without proper solder pads for the shroud
2012-10-22T14:04:12 <+Steffanx> *-o
2012-10-22T14:05:53 < baird> Steffanx: for some reason I've had no lucky getting that going, even on the Linux images and latest git-pulled firmware blobs.. :/
2012-10-22T14:06:25 <+Steffanx> Hmm, here the fw update worked .. it did screw up the bootfile though
2012-10-22T14:06:58 < zyp> my freebsd guy has my rpi
2012-10-22T14:07:00 < dongs> what happens when I have 2 SPI buses DMA'ing shit at once
2012-10-22T14:07:03 < dongs> is that evne gonna wokr? no right
2012-10-22T14:07:10 < zyp> he borrowed it to fix usb support
2012-10-22T14:07:23 < zyp> dongs, on seperate dma channels? should work fine
2012-10-22T14:07:35 < dongs> well obviosuly separate
2012-10-22T14:07:39 < dongs> but how is it gonna share?
2012-10-22T14:08:12 < zyp> round robin/priorities/whatever?
2012-10-22T14:08:31 < zyp> as long as there is enough capacity on the bus for both, it'll be fine
2012-10-22T14:09:45 < baird> NetBSD/rpi has only support for serial and SDIO-- no USB, Ethernet, etc. The molestation here is all done through a serial connection. :)
2012-10-22T14:10:08 < dongs> where do people order thier shitberrypi these days
2012-10-22T14:10:12 < dongs> are tehy actually in stock anywhere?
2012-10-22T14:11:18 <+Steffanx> YAY baird aka useless
2012-10-22T14:11:28 < baird> E14-- 2-3 weeks at most, now.
2012-10-22T14:11:45 <+Steffanx> And you'll get the 512mb version
2012-10-22T14:12:02 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-22T14:12:09 < dongs> E14 doesnt ship to scatland
2012-10-22T14:13:02 < Claude_work> RS?
2012-10-22T14:13:09 < dongs> RS says ill get it in 2013.
2012-10-22T14:13:18 < Claude_work> umm
2012-10-22T14:13:50 <+Steffanx> I thought you were anti-rpi dongs
2012-10-22T14:14:06 < dongs> I heard they're coming up with Win8RT BSP for it.
2012-10-22T14:14:08 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-22T14:14:10 <+Steffanx> overhyped etc.
2012-10-22T14:14:52 < baird> bait for attracting 16 year-old boys..
2012-10-22T14:16:08 < zyp> so win8rt will run on old non-cortex shit?
2012-10-22T14:16:36 < dongs> just looekd a the trash i got from RS
2012-10-22T14:16:41 < dongs> delivery date: 13/01/04
2012-10-22T14:16:42 < Claude_work> received my rpi in april , connected it once to my tv, ran a 3d demo and now it's gathering dust on the ol' devboard pile
2012-10-22T14:16:52 < dongs> thats y/m/d
2012-10-22T14:16:58 < zyp> that's more than I did with mine
2012-10-22T14:17:22 < zyp> I unpacket mine, looked at it, laughed a bit and packed it back up
2012-10-22T14:17:30 < Claude_work> hehe
2012-10-22T14:18:06 < zyp> then it sat for some weeks until my freebsd guy said that he was looking to get one, so I let him borrow mine, and that's where it is now
2012-10-22T14:18:21 < zyp> can't say I've missed it
2012-10-22T14:18:22 < zyp> :p
2012-10-22T14:18:25 < baird> I made a 5-node SUPERCOMPUTER of out mine, and still running it..
2012-10-22T14:18:36 < zyp> yeah, real super
2012-10-22T14:18:52 < dongs> an atom netbook probably beats your 5node opensores circlejerk
2012-10-22T14:19:05 < dongs> (while running windows 7 starter)
2012-10-22T14:19:23 < zyp> you can get the exynos 4 odroid-x for less than four times the price of a rpi, and it packs 16 times the performance
2012-10-22T14:19:34 < dongs> at
2012-10-22T14:19:34 < dongs> wat
2012-10-22T14:19:41 < dongs> 4x CHEAPER than shitberry pi?
2012-10-22T14:19:46 < dongs> isnt that like 5GBP
2012-10-22T14:20:15 < zyp> that's not what I wrote
2012-10-22T14:20:16 < BrainDamage> no, price < 4x rpi price
2012-10-22T14:20:20 < zyp> exactly
2012-10-22T14:20:25 < zyp> $129 or something
2012-10-22T14:20:29 < dongs> um
2012-10-22T14:20:39 < dongs> thats fucked up english.
2012-10-22T14:20:49 <+Steffanx> Try japanese
2012-10-22T14:21:14 -!- Rious [~Rious@ec2-50-17-240-121.compute-1.amazonaws.com] has quit [Ping timeout: 252 seconds]
2012-10-22T14:24:50 -!- Rious [~Rious@ec2-50-17-240-121.compute-1.amazonaws.com] has joined ##stm32
2012-10-22T14:26:47 < dongs> ok so, is tehre any way to trigger more than one DMA channel   ON at same time?>
2012-10-22T14:27:05 < dongs> i.e. spi1/spi2, enable DMA on both at same time.
2012-10-22T14:27:58 < Laurenceb> <Steffanx> I thought you were anti-rpi dongs
2012-10-22T14:27:58 < Laurenceb> <dongs> I heard they're coming up with Win8RT BSP for it.
2012-10-22T14:28:00 < Laurenceb> lolled
2012-10-22T14:33:16 < dongs> ^_^
2012-10-22T14:38:30 < dongs> http://www.digikey.com/product-detail/en/OKI-78SR-3.3%2F1.5-W36-C/811-2195-5-ND/2259780 i wonder how much noise is on something like this
2012-10-22T14:41:42 -!- badger [~tom@cxkg.net] has quit [Ping timeout: 246 seconds]
2012-10-22T14:45:15 < dongs> datasheet claims 19mV ripple...
2012-10-22T14:46:11 < dongs> but my 5V needs < 7mV
2012-10-22T14:46:15 < dongs> cocks.
2012-10-22T14:48:32 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-22T14:49:31 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-22T14:55:35 < dongs> http://blog.frankvh.com/2012/01/13/stm32f2xx-stm32f4xx-dma-maximum-transactions/ is this old
2012-10-22T15:00:48 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-22T15:02:05 < Laurenceb> looks like the discovery example is a joystick
2012-10-22T15:02:14 < Laurenceb> ive got one on those in my pants
2012-10-22T15:03:16 -!- badger [~tom@cxkg.net] has joined ##stm32
2012-10-22T15:04:07 < dongs> http://www.digikey.com/product-detail/en/OKL-T%2F3-W12N-C/811-2072-1-ND/2063142 looks like this is a nice cheap way to do like 5.2V, then LM1117 to 5.0V, and another murata switcher to go 3.3V (3.3V can have some ripple since its vccio only), then those micrel regs to do 1.5V and 2.5V
2012-10-22T15:04:25 < Claude_work> disco(very)stick then?
2012-10-22T15:04:46 < Laurenceb> lol
2012-10-22T15:05:00 < Laurenceb> hmm looks like maybe they didnt complete the demo
2012-10-22T15:05:07 < Laurenceb> i cant see where the data goes in
2012-10-22T15:06:38 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-22T15:09:50 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-22T15:10:53 < dongs> whats the voltage on gpio when its 'low'?
2012-10-22T15:10:57 < dongs> is it perfect 0/gnd?
2012-10-22T15:11:00 < Laurenceb> this is very simple
2012-10-22T15:11:02 < Laurenceb>      /* Copy mouse position info in ENDP1 Tx Packet Memory Area*/
2012-10-22T15:11:02 < Laurenceb>       USB_SIL_Write(EP1_IN, Mouse_Buffer, 4);
2012-10-22T15:11:03 < dongs> or is tehre some leaking shit
2012-10-22T15:11:09 < Laurenceb> looks like its just missing thast
2012-10-22T15:11:16 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-22T15:12:20 < Laurenceb> wtf
2012-10-22T15:12:24 < Laurenceb> st fail epically
2012-10-22T15:13:00 < Laurenceb> presumably the demo code was written by a n00b
2012-10-22T15:15:32 <+Steffanx> Too many spaces Laurenceb ?
2012-10-22T15:16:10 < Laurenceb> host sends 0x00 to the mouse and it replys with x,y as int8_t
2012-10-22T15:19:11 < Laurenceb> looks like it should be easy to add the mouse function to discovery firmware
2012-10-22T15:19:25 < Laurenceb> just copy over the missing lines from the mouse example
2012-10-22T15:20:52 < zyp> heh
2012-10-22T15:21:13 < zyp> remember that whatever sent in the report must match the format described in the HID report descriptor
2012-10-22T15:21:48 < Laurenceb> im not sure yet
2012-10-22T15:22:04 < Laurenceb> but it looks like they moved the usb files over from the usb example
2012-10-22T15:22:19 < Laurenceb> but didnt know how to add the "loop"  code into main
2012-10-22T15:22:36 < Laurenceb> so usb is initialised but no data is sent
2012-10-22T15:22:41 < zyp> got a call from DHL, delivery guy is apparently standing outside my home with F3discovery now
2012-10-22T15:27:49 < Laurenceb> doh
2012-10-22T15:27:58 < Laurenceb> use your work address
2012-10-22T15:28:27 < BrainDamage> leave him there
2012-10-22T15:28:33 < BrainDamage> tell him to wait for you
2012-10-22T15:28:57 < zyp> nah, he said he could just leave it in my mailbox if I was okay with it
2012-10-22T15:29:05 < zyp> so it's all good
2012-10-22T15:38:05 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-22T15:44:28 < dongs> enots: http://i.imgur.com/fmWGP.png this is my black... ~40mV
2012-10-22T15:44:34 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-22T15:44:39 < dongs> not perfect black on pal but its barely visible
2012-10-22T15:45:06 < dongs> this is with diode..
2012-10-22T15:50:31 < Laurenceb> http://hardware.slashdot.org/story/12/10/21/182216/new-arduino-due-brings-more-power-to-the-table
2012-10-22T15:50:39 < Laurenceb> 6 years?! pmsl
2012-10-22T15:51:18 < Laurenceb> at some point today ill try and get f3discovery mouse demo online....
2012-10-22T15:51:22 < Laurenceb>  <- lazy
2012-10-22T15:51:56 < Laurenceb> using some of the openpilot vector lib functions to get orientation from magno and accel would give a proper mouse
2012-10-22T15:52:33 <+Steffanx> No work for your boss you should do Laurenceb ? :P
2012-10-22T15:52:51 < BrainDamage> he's paid to irc
2012-10-22T15:52:55 < Laurenceb> suck cock ?
2012-10-22T15:53:05 <+Steffanx> I bet you ahve something to finish
2012-10-22T15:53:12 < dongs> The theoretical sampling rate has gone from the 15 ksps (kilosamples per second) of the existing boards, the Arduino Uno, Leonardo, and Mega 2560, to a whopping 1,000 ksps.
2012-10-22T15:53:15 < dongs> rofl
2012-10-22T15:53:15 < dongs> 1ksps
2012-10-22T15:53:35 <+Steffanx> 1,000 ..
2012-10-22T15:53:41 <+Steffanx> Yankie-style
2012-10-22T15:53:44 < BrainDamage> maybe the use , as thousand separator
2012-10-22T15:53:50 < BrainDamage> so 1MHz
2012-10-22T15:54:02 < Laurenceb> 1000000000millihz
2012-10-22T15:54:25 < BrainDamage> I always hated thousand separators
2012-10-22T15:54:31 < BrainDamage> so much needless confusion
2012-10-22T15:54:39 < BrainDamage> esp if there's only 1
2012-10-22T15:54:55 < Laurenceb> http://hardware.slashdot.org/comments.pl?sid=3200043&cid=41725411
2012-10-22T15:54:57 <+Steffanx> Blame the americans :P
2012-10-22T15:57:13 <+Steffanx> True, but for a numbers like 230000000 is 'neccessary' BrainDamage
2012-10-22T15:57:27 <+Steffanx> Or people should just write 230 million :P
2012-10-22T15:58:34 < BrainDamage> I just cover digits with my fingers to count them
2012-10-22T15:58:44 < BrainDamage> but with such large numbers there's no confusion anyway
2012-10-22T15:58:53 < BrainDamage> because thousand separator is repeated > once
2012-10-22T15:59:00 < BrainDamage> while decimal separator only once
2012-10-22T15:59:12 < BrainDamage> so as long as there's at least 2, there's no ambiguity
2012-10-22T16:01:48 < baird> I got into trouble in high-school for always writing numbers down like 2.3E8, instead of the traditional way with superscripts. :p
2012-10-22T16:02:46 < BrainDamage> I think anyone who got out of high school knows that high school teachears are usually cretins, with some rare exceptions
2012-10-22T16:03:47 < baird> Our Head Maths Teacher got into Politics after we/I left.. which freaked a lot of people out.
2012-10-22T16:04:45 <+Steffanx> in aussieland politicians  are .. freaky yes
2012-10-22T16:06:13 < HTT-Bird> BrainDamage: heh.  I know a good chunk of mine weren't, but then again, my high school is not a normal case
2012-10-22T16:12:42 < baird> My school had an English teacher who became a successful TV comedian after giving up the schoolie gig.
2012-10-22T16:13:13 < dongs> has anyone ever remapped SPI1 on F103
2012-10-22T16:13:19 < dongs> does it affect any other peripherals
2012-10-22T16:16:16 < HTT-Bird> baird: not too bad :)
2012-10-22T16:18:37 < baird> https://en.wikipedia.org/wiki/Club_Buggery (John Doyle was our teacher)
2012-10-22T16:22:41 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-22T16:31:35 < dongs> argh
2012-10-22T16:31:40 < dongs> my software i2c project is NOT WORKING.
2012-10-22T16:31:54 < dongs> but since youve got rtos
2012-10-22T16:32:00 < dongs> some interrupt could happen while youre nop'ing
2012-10-22T16:32:03 <+Steffanx> mr. Laurenceb will fix it
2012-10-22T16:32:36 < dongs> he will just remove spaces
2012-10-22T16:32:39 < dongs> that doesnt fix shit
2012-10-22T16:33:43 < Laurenceb> https://en.wikipedia.org/wiki/File:%C3%89douard-Henri_Avril_%2828%29.jpg
2012-10-22T16:33:45 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-22T16:35:41 < dongs> Laurenceb: i got a better one http://i.imgur.com/T9tEE.jpg
2012-10-22T16:36:46 < Laurenceb> http://s2.b3ta.com/host/creative/71320/1350847614/CLITORISALLSORTSFIREWORKS.jpg
2012-10-22T16:37:27 <+Steffanx> Is that funny?
2012-10-22T16:37:38 < zyp> I don't think I want to open those at work
2012-10-22T16:37:50 <+Steffanx> You don't see ANYTHING
2012-10-22T16:38:28 <+Steffanx> but they're not funny, so need to click zyp :)
2012-10-22T16:38:47 < dongs> no but this might be https://en.wikipedia.org/wiki/File:Tako_to_ama_retouched.jpg
2012-10-22T16:39:18 <+Steffanx> No, that's just some drawing of someone with a crazy mind
2012-10-22T16:39:20 <+Steffanx> aka japanese
2012-10-22T16:39:56 < BrainDamage> Laurenceb: http://i.imgur.com/rhPlU.jpg
2012-10-22T16:39:57 <+Steffanx> *from
2012-10-22T16:41:41 < BrainDamage> is that paint recent?
2012-10-22T16:41:57 < BrainDamage> otherwise it means that japanese were really ahead with comic porn
2012-10-22T16:43:07 < dongs> http://arduino.cc/en/uploads/Main/ArduinoDue_Back.jpg
2012-10-22T16:43:07 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-22T16:43:08 < dongs> ????????????????????????
2012-10-22T16:43:57 <+Steffanx> !!!
2012-10-22T16:44:24 < dongs> how the fuck do you do different color mask?
2012-10-22T16:45:01 <+Steffanx> Uhm, i guess it's the mask vs silk screen
2012-10-22T16:45:09 < dongs> oh duh
2012-10-22T16:45:17 <+Steffanx> Good night dong
2012-10-22T16:45:19 <+Steffanx> s
2012-10-22T16:45:24 < dongs> i need to stop smoking Laurenceb shit
2012-10-22T16:45:43 < Thorn> is it really FCC and CE certified?
2012-10-22T16:45:48 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-22T16:46:10 < zyp> who the fuck decided it was sane to cover most of the underside with silk?
2012-10-22T16:46:14 < zyp> seems like a waste
2012-10-22T16:47:19 -!- mrcan__ [~mrcan@78.173.69.222] has joined ##stm32
2012-10-22T16:47:21 -!- mrcan__ [~mrcan@78.173.69.222] has quit [Changing host]
2012-10-22T16:47:21 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-22T16:47:31 < dongs> they confused me!
2012-10-22T16:47:37 < zyp> ok, worth it
2012-10-22T16:47:47 <+Steffanx> :D
2012-10-22T16:50:41 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-22T16:52:38 < Laurenceb> whats the smps for?!
2012-10-22T16:52:51 < dongs> wut
2012-10-22T16:52:58 < dongs> on tardruino due?
2012-10-22T16:53:09 <+Steffanx> 5v => 3.3 ?
2012-10-22T16:53:19 < Laurenceb> maybe
2012-10-22T16:53:58 <+Steffanx> maybe leafs are purple, maybe grass is red
2012-10-22T16:55:46 < Thorn> what are those huge mlcc?
2012-10-22T16:56:07 < Thorn> 1210 or whatever they are
2012-10-22T16:56:09 < Laurenceb> for huge fail
2012-10-22T16:56:26 < dongs> just looks like failcaps
2012-10-22T16:56:27 < Thorn> I ordered some 0805 10uF ones
2012-10-22T16:58:05 < Thorn> and why fiducials on the board itself?
2012-10-22T16:58:12 < Thorn> looks like complete crap
2012-10-22T16:58:47 < Thorn> what's that atmel qfn thing near usb connectors?
2012-10-22T16:59:04 < dongs> maybe some usb?>serhai shit?
2012-10-22T16:59:08 < dongs> serial
2012-10-22T16:59:09 < dongs> wtf i cant type
2012-10-22T16:59:19 < Thorn> that's retarded^2
2012-10-22T16:59:32 < Thorn> as if that mcu can't do proprer usb
2012-10-22T16:59:41 < Laurenceb> it might be debugger
2012-10-22T17:00:22 < Thorn> arduinofags know how to debug?
2012-10-22T17:00:46 < dongs> Thorn: thats what the other usb socket is for
2012-10-22T17:02:25 < Thorn> they have some serpentine tracks there
2012-10-22T17:02:29 < dongs> yeah
2012-10-22T17:02:32 < Thorn> wtf, HS usb?
2012-10-22T17:02:41 < dongs> on MY tarduino?
2012-10-22T17:02:44 < dongs> yeah I think sam3 has highspeed
2012-10-22T17:03:58 < Thorn> and the dc/dc as far as possible from the power socket...
2012-10-22T17:04:15 < BrainDamage> the annoying thing is people thinking it's advanced or new somehow
2012-10-22T17:04:29 < zyp> Thorn, isn't that a sensible choice?
2012-10-22T17:04:57 < dongs> for what
2012-10-22T17:05:11 < zyp> higher voltage -> less current, so there are less losses upstream of the reg than downstream
2012-10-22T17:05:28 < zyp> so moving the reg towards the load rather than the source makes sort of sense
2012-10-22T17:05:43 < zyp> not that there should be any load for it to matter in this case anyway
2012-10-22T17:06:15 < zyp> but there is no reason why it should be located as close to the power input as possible
2012-10-22T17:06:17 < Thorn> the load is going to be motors and stuff connected directly to mcu pins anyway
2012-10-22T17:06:43 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-22T17:06:59 < dongs> lol is that some faggot level shifter for the 5V>3V uart
2012-10-22T17:07:02 < dongs> on teh due
2012-10-22T17:07:05 < dongs> better look at schematic
2012-10-22T17:07:37 < Thorn> how are they going to deal with all the 5V "shields" btw?
2012-10-22T17:07:42 < dongs> they dont
2012-10-22T17:07:48 < Thorn> lol
2012-10-22T17:07:49 < dongs> Warning: Unlike other Arduino boards, the Arduino Due board runs at 3.3V. The maximum voltage that the I/O pins can tolerate is 3.3V. Providing higher voltages, like 5V to an I/O pin could damage the board.
2012-10-22T17:07:58 < Thorn> LOL
2012-10-22T17:08:20 < Thorn> so the chip doesn't even have 5V toperant GPIO?
2012-10-22T17:08:25 < Thorn> *tolerant
2012-10-22T17:08:31 < dongs> apparently not
2012-10-22T17:08:34 < zyp> that's pretty nice actually
2012-10-22T17:09:00 < zyp> finally people will stop going on about level conversion to interface 3.3V stuff with 5V AVRs
2012-10-22T17:09:13 < BrainDamage> Steffanx: what was the lowest speed your fancy anemometer could measure?
2012-10-22T17:09:20 < zyp> since pretty much everything else is at 3.3V nowadays
2012-10-22T17:11:07 < Thorn> Arduino Due Front  Arduino Due Back  Arduino Due In 2013
2012-10-22T17:11:53 < zyp> ha
2012-10-22T17:11:56 < Laurenceb>  Arduino Due In 2100
2012-10-22T17:12:05 < zyp> nice pun
2012-10-22T17:12:11 <+Steffanx> Cant tell you
2012-10-22T17:12:17 <+Steffanx> never calibrated it BrainDamage
2012-10-22T17:12:56 < BrainDamage> aw, someone in ##electronics was searching for anemometers for small airflow, and yours sounded easily dia-ble
2012-10-22T17:13:08 <+Steffanx> It was a uni project and dont own it anymore
2012-10-22T17:13:17 <+Steffanx> small as in ..
2012-10-22T17:13:34 < BrainDamage> room convection / gusts
2012-10-22T17:13:47 < zyp> Steffanx, you didn't get to keep it?
2012-10-22T17:13:52 <+Steffanx> No
2012-10-22T17:14:03 < zyp> I still have a pile of trash I made at uni
2012-10-22T17:14:15 <+Steffanx> I found this one while i was working on mine BrainDamage .. http://www.technik.dhbw-ravensburg.de/~lau/ultrasonic-anemometer.html
2012-10-22T17:15:10 < BrainDamage> ok, magnitude worse than hot wire
2012-10-22T17:16:50 <+Steffanx> 0.05m/s isnt enough? :)
2012-10-22T17:17:15 < BrainDamage> resolution isn't accuracy
2012-10-22T17:17:41 < BrainDamage> you can have 1000 digits, and 999 of them be garbage
2012-10-22T17:17:55 <+Steffanx> yes yes
2012-10-22T17:18:21 < Thorn> okay when am I going to finish this damn project.
2012-10-22T17:22:14 < emeb_mac> when the cows come home?
2012-10-22T17:22:31 <+Steffanx> When daddy calls them
2012-10-22T17:23:03 < Thorn> the cows emigrated to new zealand for greener pastures.
2012-10-22T17:24:18 < Thorn> what if I make an external amplifier later and leave only the codec on this board.
2012-10-22T17:25:14 -!- jkent [~jkent@173-26-157-254.client.mchsi.com] has quit [Ping timeout: 255 seconds]
2012-10-22T17:28:41 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-22T17:29:49 < emeb_mac> that's what I would do.
2012-10-22T17:30:45 < emeb_mac> which codec?
2012-10-22T17:30:58 < Thorn> wm8731
2012-10-22T17:31:04 < emeb_mac> heh.
2012-10-22T17:31:18 < Thorn> I looked at wm8778 but it's not available anywhere
2012-10-22T17:31:31 < emeb_mac> do you have room to bring out both the line and hp outputs?
2012-10-22T17:31:31 < Laurenceb> * Now talking on ##fapping
2012-10-22T17:31:43 < Thorn> planning to do just that
2012-10-22T17:32:09 < emeb_mac> good idea - then you can compare the hp output to line + cmoy.
2012-10-22T17:32:09 < Thorn> it also has both line in and mic inputs, I'd like to at least use mic
2012-10-22T17:32:17 < dongs> USB clocking mode can generate all common MP3 sample rates (incl. 441.kHz) directly from USB clock
2012-10-22T17:32:20 < dongs> ???????????????
2012-10-22T17:32:33 < emeb_mac> someone misplaced a decimal
2012-10-22T17:32:34 < dongs> (from WM8731 spec)
2012-10-22T17:32:50 < dongs> well, not evne tha,t, USB clocking?
2012-10-22T17:32:56 < dongs> does it evne have USB
2012-10-22T17:33:05 < jpa-> maybe it means the start-of-frame signal
2012-10-22T17:33:05 < Thorn> do standard computer headset mics require bias or something of that sort?
2012-10-22T17:33:17 < Thorn> dongs: it has no usb interface
2012-10-22T17:33:21 < emeb_mac> it's designed to be used with an external clock generator
2012-10-22T17:33:36 < jpa-> that's the common way to synchronize usb isochronous transfers so that you don't get buffer under/overflows
2012-10-22T17:33:43 < BrainDamage> Thorn: yes
2012-10-22T17:33:54 < emeb_mac> so they have dividers that can generate sample rates from the usb / mcu clock.
2012-10-22T17:34:05 < BrainDamage> they have an integrated transistor that is controlled by the microphone
2012-10-22T17:34:11 < dongs> lol cant copypaste from their fucking PDF
2012-10-22T17:34:12 < emeb_mac> but I've never used that mode - it's not as flexible as it sounds.
2012-10-22T17:34:14 < BrainDamage> and you need to bias the drain
2012-10-22T17:34:16 < dongs> anyway it just means it runs of 12mhz clock
2012-10-22T17:34:29 < emeb_mac> yep
2012-10-22T17:34:37 -!- mrcan_ [~mrcan@78.173.69.222] has joined ##stm32
2012-10-22T17:34:37 -!- mrcan_ [~mrcan@78.173.69.222] has quit [Changing host]
2012-10-22T17:34:37 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-22T17:34:49 < BrainDamage> a simple pullup R and inline C as highpass filter is the usual choice
2012-10-22T17:35:09 < dongs> < emeb_mac> it's designed to be used with an external clock generator
2012-10-22T17:35:12 < dongs> actually it has on inside
2012-10-22T17:35:17 < dongs> but can use external
2012-10-22T17:35:30 < emeb_mac> Thorn - common mics are electrets that need a bias. The WM8731 can generate that bias.
2012-10-22T17:35:54 < emeb_mac> dongs:  yes - control register chooses which mode.
2012-10-22T17:36:21 < emeb_mac> but if you don't want it then you can leave the bias pin disconnected.
2012-10-22T17:37:14 < zyp> ha, RS actually did ship F0 and F3 discovery seperately
2012-10-22T17:37:25 < emeb_mac> more postage!
2012-10-22T17:38:17 -!- mrcan__ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 246 seconds]
2012-10-22T17:38:52 < dongs> http://i.imgur.com/OlqtH.png wtf did they REALLY make those pins like 1.7mm pitch
2012-10-22T17:39:38 < jpa-> why not? :P
2012-10-22T17:40:25 < jpa-> also 0.81 mm diameter pins so that you need 0.9 or 1.0 mm holes
2012-10-22T17:40:40 < emeb_mac> unbiased - not an integer number in any measuring system.
2012-10-22T17:40:44 < jpa-> it's too easy if you can route nets between the pins :P
2012-10-22T17:41:19 < dongs> and no recommended hole pattern
2012-10-22T17:41:21 < dongs> assholes!
2012-10-22T17:42:07 < zyp> http://bin.jvnv.net/f/mu3oa.JPG
2012-10-22T17:42:14 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Read error: Connection reset by peer]
2012-10-22T17:42:46 < dongs> fuuu
2012-10-22T17:42:49 < emeb_mac> family portrait
2012-10-22T17:42:53 < jpa-> gotta catch 'em all?
2012-10-22T17:42:53 < zyp> :D
2012-10-22T17:42:57 < zyp> yep
2012-10-22T17:43:13 < zyp> I almost regret I didn't buy the VL discovery when I saw it in japan
2012-10-22T17:43:16 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-22T17:43:23 < BrainDamage> nice collection
2012-10-22T17:43:27 < zyp> but I already have enough F1 boards
2012-10-22T17:43:30 < emeb_mac> I noticed the VL missing
2012-10-22T17:43:46 < jpa-> what is one supposed to do with the circle in the F3 board?
2012-10-22T17:43:57 < BrainDamage> gyro?
2012-10-22T17:44:03 < emeb_mac> make a compass
2012-10-22T17:44:04 < BrainDamage> or magnetometer
2012-10-22T17:44:25 < BrainDamage> I think I am seeing "N S E W" on it
2012-10-22T17:44:33 < jpa-> ok
2012-10-22T17:44:33 < BrainDamage> using a stupid font
2012-10-22T17:44:38 < jpa-> ah, yeah
2012-10-22T17:44:51 < dongs> i have vldiscover  i think
2012-10-22T17:44:59 < BrainDamage> I have vl and f4
2012-10-22T17:45:19 < dongs> huh whihc one is the 2nd one?
2012-10-22T17:45:21 < Thorn> what SDIO pin do I put an activity led on?
2012-10-22T17:45:22 < dongs> thats not F0
2012-10-22T17:45:25 < emeb_mac> Q: Which direction are we going? A: LD7
2012-10-22T17:45:34 < BrainDamage> second from left is L
2012-10-22T17:45:34 < dongs> Thorn: waT? sdio has no activity led?
2012-10-22T17:45:50 < Thorn> wat?
2012-10-22T17:45:58 < BrainDamage> from left to right: F0, L1, F3, F4
2012-10-22T17:46:04 < zyp> dongs, on lpc4300 it does
2012-10-22T17:46:10 < dongs> o
2012-10-22T17:46:11 < jpa-> hmm, what is the JP4 "TX" "RX" connector on the F3 board?
2012-10-22T17:46:19 < dongs> bedtime
2012-10-22T17:46:30 < Thorn> this is because NXP loves hardware LED pins for usb and other things
2012-10-22T17:46:38 < Thorn> but not ST
2012-10-22T17:47:33 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-22T17:49:42 < Laurenceb>   21048	is discovery demo size with codesourcery
2012-10-22T17:49:51 < Laurenceb> and code off my github
2012-10-22T17:49:58 < Laurenceb> stock is 17244
2012-10-22T17:50:17 < Laurenceb> ill try adding usb mouse now
2012-10-22T17:51:31 < Laurenceb> hmm st code calculates euler angles
2012-10-22T17:51:38 < Laurenceb> i could impliment a proper mouse
2012-10-22T17:55:53 <+Steffanx> could
2012-10-22T17:56:03 < Laurenceb> am doing atm
2012-10-22T17:58:35 -!- Claude_work is now known as claude
2012-10-22T17:58:45 -!- daku is now known as DaKu
2012-10-22T18:02:48 < zyp> does anyone happen to know whether stm32 are supporting multi-drop swd?
2012-10-22T18:12:46 < Laurenceb> nope
2012-10-22T18:13:03 < Laurenceb> usb mouse is very basic
2012-10-22T18:13:29 < Laurenceb> 4 bytes of data
2012-10-22T18:18:34 < Laurenceb> http://www.poopreport.com/phpBB/viewtopic.php?f=1&t=7476
2012-10-22T18:24:40 -!- claude is now known as Claude
2012-10-22T18:28:19 < Laurenceb> http://store.arduino.cc/eu/index.php?main_page=product_info&cPath=11_12&products_id=243
2012-10-22T18:28:23 < Laurenceb> its actually in stock
2012-10-22T18:29:54 < Laurenceb> The Programming port is connected to an ATmega16U2, which provides a virtual COM port to software on a connected computer (To recognize the device, Windows machines will need a .inf file, but OSX and Linux machines will recognize the board as a COM port automatically.). The 16U2 is also connected to the SAM3X hardware UART. Serial on pins RX0 and TX0 provides Serial-to-USB communication for programming the board through the ATmega16U2 micr
2012-10-22T18:29:55 < Laurenceb> ocontroller
2012-10-22T18:29:57 < Laurenceb> epic fail
2012-10-22T18:30:12 < Laurenceb> i thought it might be a gdbserver
2012-10-22T18:35:03 < BrainDamage> wait, no debugger on board? for 50$?
2012-10-22T18:35:17 < BrainDamage> extra debugger available for extra 50$ I imagine ...
2012-10-22T18:36:30 < Thorn> arduino users don't know what gdbserver is anyway
2012-10-22T18:37:26 < BrainDamage> how would you call a ferrule which has one end closed with a flat disk?
2012-10-22T18:37:36 < BrainDamage> sorta cap shape
2012-10-22T18:45:36 -!- utzig [~utzig@177.143.196.215] has joined ##stm32
2012-10-22T18:53:35 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-22T18:56:08 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-22T18:56:33 < Laurenceb> hell yeah
2012-10-22T18:56:37 < Laurenceb> mouse is working
2012-10-22T18:56:47 < Laurenceb> its a little goofy atm
2012-10-22T18:56:55 < Thorn> are dp/dm pins enough for a host-only interface on f4?
2012-10-22T18:56:57 < Laurenceb> need to fix the auler angle mapping
2012-10-22T18:57:27 < zyp> Thorn, I believe so
2012-10-22T18:57:36 < Laurenceb> ok wut
2012-10-22T18:57:44 < Laurenceb> F3discovery has 22ohm Rs on usb
2012-10-22T18:57:56 < Laurenceb> datasheet says usb is impedance matched
2012-10-22T18:58:00 < Laurenceb> so nothing needed
2012-10-22T18:59:25 < zyp> fucking stlink shit
2012-10-22T19:00:15 < zyp> 2012-10-22T17:50:29 INFO src/stlink-common.c: Starting Flash write for VL/F0 core id
2012-10-22T19:00:19 < zyp> 2012-10-22T17:50:29 INFO src/stlink-common.c: Successfully loaded flash loader in sram
2012-10-22T19:00:22 < zyp> 2012-10-22T17:50:29 FATAL src/stlink-common.c: flash loader run error
2012-10-22T19:00:37 < zyp> is this normal for F103?
2012-10-22T19:00:57 < Thorn> Laurenceb: if F3 has a F103-style usb core then you do need resistors afaict
2012-10-22T19:01:01 < emeb> there's a reason I spent the $$$ on a BMPM
2012-10-22T19:01:15 < Laurenceb> Thorn: according to datasheet its now impedance matched
2012-10-22T19:01:18 < zyp> I suspect cpu is stuck in lockup and stlink doesn't know how to get it unstuck
2012-10-22T19:01:32 < zyp> weren't there a issue like that?
2012-10-22T19:01:35 < Laurenceb> zyp: stlink works fine on f103 for me
2012-10-22T19:01:49 < emeb> zyp: do you have access to a Win machine?
2012-10-22T19:01:53 < Laurenceb> and F3 too XD
2012-10-22T19:02:08 < zyp> emeb, yes, I'm almost considering booting windows just to get this shit flashed
2012-10-22T19:02:16 < emeb> the Windows STLINK driver & app have always been able to rescue for me.
2012-10-22T19:02:18 < zyp> it's my BMP that I'm attempting to flash back
2012-10-22T19:02:28 < zyp> sounds good
2012-10-22T19:02:34 < bsdfox_> anyone happen to use vsprog with stm8s?
2012-10-22T19:02:36 < zyp> so there is a seperate stlink app for windows?
2012-10-22T19:02:55 < emeb> zyp: yes - a single-window gui
2012-10-22T19:02:56 < zyp> i.e. standalone from any IDE shit
2012-10-22T19:03:00 < zyp> good
2012-10-22T19:03:03 < emeb> should be on their website.
2012-10-22T19:03:16 < emeb> just loads .bin files. (maybe hex too??)
2012-10-22T19:03:37 < zyp> that's fine, I just need to get the blackmagic bootloader flashed back
2012-10-22T19:05:16 < emeb> interesting conundrum though - the SWD pins for the 103 on the BMPM aren't brought out.
2012-10-22T19:05:33 < emeb> I wonder how you'd unbrick that one.
2012-10-22T19:06:02 < zyp> oh
2012-10-22T19:06:06 < zyp> they were on mine
2012-10-22T19:06:21 < emeb> you've got the older/larger BMP though, right?
2012-10-22T19:06:26 < emeb> not the mini.
2012-10-22T19:06:29 < zyp> anyway, I guess serial bootloader is as feasible as swd
2012-10-22T19:06:52 < emeb> ah yes - the serial pins are on a connector.
2012-10-22T19:07:00 < emeb> or maybe DFU?
2012-10-22T19:07:12 < zyp> but I think the bootloader pins are on some of the gnd pins of the 20-pin connector
2012-10-22T19:07:27 < zyp> and I've soldered an adapter board to that
2012-10-22T19:07:28 < emeb> again - no 20-pin connector on the mini
2012-10-22T19:07:36 < zyp> I'm talking about mine now :p
2012-10-22T19:08:13 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-22T19:08:22 < emeb> Oh - cute.
2012-10-22T19:08:48 < emeb> looking at the BMPM schematic - they hooked SWD/SWCK up to the TXD/RXD pins of the serial conn.
2012-10-22T19:08:59 < emeb> via resistors.
2012-10-22T19:09:44 < zyp> I guess that's also a way to do it
2012-10-22T19:09:47 < emeb> so it can be SWD'ed
2012-10-22T19:10:09 < jpa-> zyp: just time your reset button well and you might be able to write a functional program that doesn't immediately lock up
2012-10-22T19:11:03 < emeb> f103 doesn't have USB DFU on the system bootloader does it.
2012-10-22T19:11:16 < zyp> no, it doesn't
2012-10-22T19:11:32 < zyp> that's why it has a DFU bootloader in the start of the flash
2012-10-22T19:11:50 < zyp> which I somehow managed to destroy
2012-10-22T19:11:59 < zyp> through that same bootloader
2012-10-22T19:12:30 < emeb> so having the BOOT0 permanently tied to GND really only loses you the serial bootloader. No biggie.
2012-10-22T19:13:12 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 252 seconds]
2012-10-22T19:15:29 < emeb> bootloader overwrites itself. sw suicide.
2012-10-22T19:15:57 < emeb> Dave... I.. can... feel... my... mind... going...
2012-10-22T19:22:09 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 276 seconds]
2012-10-22T19:25:34 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-22T19:28:25 < Laurenceb> https://github.com/Laurenceb/STM32F3Discovery
2012-10-22T19:28:27 < Laurenceb> added
2012-10-22T19:30:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-22T19:32:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-22T19:33:04 <+Steffanx> You didn't read what someone said before Laurenceb ?
2012-10-22T19:33:25 <+Steffanx> Dont commit the .o-files
2012-10-22T19:34:40 < jpa-> but the .o's deserve to be loved also
2012-10-22T19:34:52 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 272 seconds]
2012-10-22T19:35:05 < Thorn> echo *.o >> .gitignore
2012-10-22T19:35:12 < zyp> ha
2012-10-22T19:35:22 < zyp> that sounds like a bad idea
2012-10-22T19:35:33 < Thorn> echo "*.o" >> .gitignore
2012-10-22T19:35:35 < zyp> (the unescaped * part of it)
2012-10-22T19:35:36 < zyp> :)
2012-10-22T19:36:45 < Thorn> I'm going to use at least 85 pins of the f4.
2012-10-22T19:36:55 < zyp> ok
2012-10-22T19:37:30 < zyp> does that include ground/power pins?
2012-10-22T19:37:44 < Thorn> yes
2012-10-22T19:37:47 < Thorn> even BOOT0
2012-10-22T19:37:50 < Thorn> and Vcap
2012-10-22T19:38:00 < emeb> why so many?
2012-10-22T19:38:23 < Laurenceb> okok
2012-10-22T19:38:33 < karlp> fukcing stlink shit, yeah, the detect and recover from lockup is known and no-ones fixed it yet.
2012-10-22T19:38:37 < Laurenceb> looks like ST fail at quaternions
2012-10-22T19:38:47 < Laurenceb> this 3D heading is utter drap
2012-10-22T19:38:52 < Laurenceb> *crap
2012-10-22T19:39:29 < emeb> either that, or their magno bites (like most seem to)
2012-10-22T19:39:48 < Thorn> this should work well on microcontrollers http://sourceforge.net/projects/ogl-math/
2012-10-22T19:40:03 < Laurenceb> bites?
2012-10-22T19:40:32 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-22T19:40:36 < emeb> munches/sucks/blows/is_lame/etc
2012-10-22T19:40:44 < zyp> I'm currently using 115 pins of the lpc4300 I'm breaking out
2012-10-22T19:40:57 < Laurenceb> bbl
2012-10-22T19:42:13 < zyp> of course, that includes 42 power/ground pins
2012-10-22T19:43:45 < Thorn> they are not pins
2012-10-22T19:43:58 < zyp> balls then
2012-10-22T19:44:02 < zyp> same shit
2012-10-22T19:45:22 < emeb> balls... how appropriate.
2012-10-22T19:45:56 -!- barthess [~barthess@5.100.203.34] has joined ##stm32
2012-10-22T20:01:20 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Read error: Connection reset by peer]
2012-10-22T20:01:38 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-22T20:02:16 -!- |sark [~|sark@109.235.51.202] has quit [Ping timeout: 246 seconds]
2012-10-22T20:02:40 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-22T20:03:03 -!- |sark [~sark@109.235.51.202] has joined ##stm32
2012-10-22T20:03:57 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-22T20:34:18 -!- utzig [~utzig@177.143.196.215] has quit [Ping timeout: 264 seconds]
2012-10-22T20:37:19 -!- utzig [~utzig@177.143.196.215] has joined ##stm32
2012-10-22T20:40:05 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-22T20:42:28 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-22T21:11:32 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-22T21:17:46 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-22T21:18:03 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Read error: Connection reset by peer]
2012-10-22T21:20:15 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-22T21:20:17 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-22T21:21:21 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has joined ##stm32
2012-10-22T21:28:01 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-22T21:34:00 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has joined ##stm32
2012-10-22T21:35:04 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 246 seconds]
2012-10-22T21:37:23 < Laurenceb_> SE PARECE A LA SOBRINA DE OBAMA
2012-10-22T21:38:22 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has quit [Ping timeout: 268 seconds]
2012-10-22T21:41:47 -!- Erlkoenig1 [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-22T21:42:04 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has joined ##stm32
2012-10-22T21:43:12 < emeb> who looks like Obama's niece?
2012-10-22T21:48:13 <+Steffanx> His mother
2012-10-22T21:54:33 < Erlkoenig> me if i shaved my epic beard
2012-10-22T21:57:15 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 260 seconds]
2012-10-22T21:58:23 <+Steffanx> Hmm, now you look like santa?
2012-10-22T21:59:15 < Laurenceb_> i thought Erlkeonig was a grrrl
2012-10-22T21:59:42 < Erlkoenig> not yet, i think i need a few years for my gandalf-look... and it's not white yet
2012-10-22T21:59:42 <+Steffanx> ( me too )
2012-10-22T21:59:47 < Erlkoenig> wat :D
2012-10-22T22:00:00 < Erlkoenig> why? :D
2012-10-22T22:00:19 <+Steffanx> apparently you gave us that impression
2012-10-22T22:00:49 < Erlkoenig> hmm funny
2012-10-22T22:01:06 <+Steffanx> I don't know why.. maybe Laurenceb knows
2012-10-22T22:01:08 < Laurenceb_> obsession with knitting
2012-10-22T22:01:13 < Erlkoenig> i might be a beardy gurl
2012-10-22T22:01:32 < Erlkoenig> knitting wires :D
2012-10-22T22:09:32 < Laurenceb_> https://github.com/Laurenceb/STM32F3Discovery/blob/master/src/main.c#L310
2012-10-22T22:09:41 < Laurenceb_> im unconvinced by that ST code
2012-10-22T22:09:49 < Laurenceb_> what do people here think?
2012-10-22T22:11:21 < Laurenceb_> seem to be trying to use "pitch" and "roll" from accel to rotate the magno
2012-10-22T22:11:37 -!- barthess [~barthess@5.100.203.34] has quit [Quit: Leaving.]
2012-10-22T22:12:41 < Laurenceb_> they seem to think they can rotate magno into world space leaving out the yaw
2012-10-22T22:12:52 < Laurenceb_> but its not working at all on my f3discovery
2012-10-22T22:15:37 < Laurenceb_> anyone got an F3discovery running?
2012-10-22T22:27:07 < Laurenceb_> http://cdn02.animenewsnetwork.com/images/cms/interest/56122/win8_you.png
2012-10-22T22:36:47 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-22T22:37:43 < Tectu> wtf is that?
2012-10-22T22:40:47 <+Steffanx> The reason why dongs want a rpi
2012-10-22T22:40:51 <+Steffanx> wants
2012-10-22T22:44:58 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-22T22:46:36 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
2012-10-22T23:09:55 < Tectu> jpa-, didn' you had problems that your MCU did go to standby/sleep/powerdown mode right after reset a while ago?
2012-10-22T23:17:49 -!- Titan___ [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-22T23:26:47 -!- Claude is now known as claude
2012-10-22T23:32:21 -!- bgamari [~bgamari@pool-108-8-243-8.spfdma.east.verizon.net] has quit [Ping timeout: 245 seconds]
2012-10-22T23:32:33 -!- utzig [~utzig@177.143.196.215] has left ##stm32 []
2012-10-22T23:34:51 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-22T23:46:07 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
--- Day changed Tue Oct 23 2012
2012-10-23T00:16:01 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-23T00:19:00 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-23T00:19:03 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-23T00:19:09 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Remote host closed the connection]
2012-10-23T00:27:57 -!- DaKu is now known as daku
2012-10-23T00:55:23 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-23T01:15:45 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-23T01:22:52 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Quit: Tzacks]
2012-10-23T01:56:04 < dongs> http://i.imgur.com/QIYQR.jpg
2012-10-23T01:56:14 < zyp> ok
2012-10-23T01:56:57 < dongs> http://i.imgur.com/xuq29.gif heres a better one for you zyp
2012-10-23T01:57:26 < zyp> old
2012-10-23T01:57:30 < dongs> sorry.
2012-10-23T01:58:18 < Laurenceb_> been reading your fan club on multiwii
2012-10-23T02:01:24 < Laurenceb_> and they are all gathering around like kawaii schoolgirls around jimmy savile in #multiwii now
2012-10-23T02:01:54 < Laurenceb_> #cultofdongs
2012-10-23T02:04:36 < dongs> yaman
2012-10-23T02:04:51 < dongs> btw, that heading code in stmdisco, lol
2012-10-23T02:05:27 < Laurenceb_> yeah
2012-10-23T02:05:37 < Laurenceb_> doesnt work for me
2012-10-23T02:05:49 < Laurenceb_> the leds are move like a tilt sensor than magno
2012-10-23T02:09:58 < dongs> even if the board is level?
2012-10-23T02:10:03 < dongs> i mean, and you rotate it
2012-10-23T02:10:19 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Ping timeout: 268 seconds]
2012-10-23T02:10:32 < Laurenceb_> if its tilting a little
2012-10-23T02:10:38 < Laurenceb_> like 30 degrees
2012-10-23T02:11:01 < dongs> i dont think that tilt compensation works for large tilt angles
2012-10-23T02:11:22 < Laurenceb_> you dont say
2012-10-23T02:11:30 < Laurenceb_>  _theirs_  doesnt
2012-10-23T02:11:37 < Laurenceb_> im trying openpilot code
2012-10-23T02:11:41 < dongs> heH
2012-10-23T02:12:17 < Laurenceb_> http://25.media.tumblr.com/tumblr_mavnz9yYZ41qfm239o1_500.jpg
2012-10-23T02:12:32 < dongs> http://www.johndcook.com/standard_deviation.html
2012-10-23T02:12:55 < dongs> if i run 1000 gyro values through that, will std_dev output give me what I expect?
2012-10-23T02:13:51 < Laurenceb_> itll give you a measurement of gyro noise
2012-10-23T02:13:55 < Laurenceb_> rms noise
2012-10-23T02:14:38 < Laurenceb_> thats not what you get if we put you in a room with richard m stallman
2012-10-23T02:16:06 < emeb> groannnnnnnn
2012-10-23T02:16:36 < emeb> get thee to a punnery
2012-10-23T02:18:14 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-23T02:23:08 < Laurenceb_> http://www.youtube.com/watch?v=bOD2-mRphTI
2012-10-23T02:23:11 < Laurenceb_> lolled
2012-10-23T02:25:09 < dongs> haha
2012-10-23T02:31:21 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-10-23T02:34:30 < emeb> useless machines - with attitude.
2012-10-23T02:35:10 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 246 seconds]
2012-10-23T02:35:11 -!- Erlkoenig [~erlkoenig@pptp-212-201-70-31.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-23T02:36:05 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-23T03:18:41 < emeb> Arduino + Raspberry Pi + Beer = lol: http://vimeo.com/51526148
2012-10-23T03:19:58 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-23T03:24:27 < dongs> http://japandailypress.com/police-arrest-73-year-old-prostitute-working-the-yokohama-streets-2216928 typical japan activity attn Laurenceb / zyp
2012-10-23T03:27:06 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit [Ping timeout: 264 seconds]
2012-10-23T03:27:40 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-23T03:28:42 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-23T03:29:01 -!- bgamari [~bgamari@pool-96-240-252-23.spfdma.east.verizon.net] has joined ##stm32
2012-10-23T03:33:36 -!- bgamari [~bgamari@pool-96-240-252-23.spfdma.east.verizon.net] has quit [Ping timeout: 272 seconds]
2012-10-23T03:37:54 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-23T03:57:39 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-23T03:59:26 < Tectu> emeb: LOL
2012-10-23T04:35:59 < dongs> attn cjbeard http://i.imgur.com/8aR5C.jpg
2012-10-23T04:41:02 < gnomad> i swear I know her!
2012-10-23T04:41:43 < dongs> is it hackkitten's friend
2012-10-23T04:46:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-23T04:49:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-23T04:58:13 < baird> doggy: just another emo swedish girl: https://www.youtube.com/watch?v=jyTpk6xyzzk
2012-10-23T05:25:44 < gnomad> needs english subs
2012-10-23T05:26:51 -!- bgamari_ [~bgamari@pool-96-240-253-253.spfdma.east.verizon.net] has joined ##stm32
2012-10-23T05:29:52 < dongs> gnomad: http://www.reddit.com/r/WTF/comments/y06u9/no_but_for_real_how_does_it_feel_to_be_a_cat/
2012-10-23T05:30:02 < gnomad> heh, reddit.
2012-10-23T05:33:14 < gnomad> oh, I love this comment in the thread:
2012-10-23T05:33:33 < gnomad> Most are harmless furries or people into role-playing, but I've seen it go deep places. In my locale they meet up with the 'real vampires' at the Panera Bread.
2012-10-23T05:38:09 < emeb_mac> hint: if you're meeting up at Panera, you're not a real vampire.
2012-10-23T05:39:05 < gnomad> iknowrite?
2012-10-23T05:39:39 < emeb_mac> suburbanvampireposers
2012-10-23T05:43:41 -!- bgamari_ [~bgamari@pool-96-240-253-253.spfdma.east.verizon.net] has quit [Ping timeout: 268 seconds]
2012-10-23T05:47:21 -!- bgamari [~bgamari@pool-108-8-228-141.spfdma.east.verizon.net] has joined ##stm32
2012-10-23T05:51:25 < baird> An old friend of mine is from a Hungarian expat family.. When The Chauchescu regime fell, the Hungarian and Romanian governments were figuring out who the land belonged too, and the friend (or rather is Dad) got a notice that they had land /and a family crypt/ in Transylvania.
2012-10-23T06:01:00 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-23T06:01:08 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-23T06:12:23 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 260 seconds]
2012-10-23T06:22:59 < emeb_mac> amazing how many inquiries I get for info on my STM32F4 + codec board. Every couple days someone wants to buy one.
2012-10-23T06:26:17 < gnomad> do you have a web page for it?
2012-10-23T06:27:36 < dongs> does it actually do anythign
2012-10-23T06:34:24 < emeb_mac> webpage: http://ebrombaugh.studionebula.com/synth/stm32f4_codec/index.html
2012-10-23T06:35:05 < emeb_mac> dongle: yes it does something - I've got demo code that does a simple audio delay w/ regeneration.
2012-10-23T06:35:19 < emeb_mac> dongs not dongle
2012-10-23T06:35:28 < dongs> same guy anywy
2012-10-23T06:36:01 < emeb_mac> but I've been converting a lot of my old audio FX code over to it over the last few weeks, plus trying out some new stuff.
2012-10-23T06:36:34 < emeb_mac> Mostly it's an R&D project though. A couple of guys are trying to make guitar stomp-box effects with it.
2012-10-23T06:39:17 < emeb_mac> I guess some day I need to try out the USB and SD interfaces :D
2012-10-23T06:40:28 < gnomad> emeb_mac: how about a FLAC player for it?
2012-10-23T06:40:53 < emeb_mac> gnomad: could probably do that. The M4F has plenty of horsepower for audio decoding.
2012-10-23T06:41:42 < emeb_mac> load up FLACs on a huge SD card and play 'em. Get a display and some buttons to navigate. Should be fairly easy.
2012-10-23T06:41:46 < gnomad> oh hey... what do you think of the black magic probe?
2012-10-23T06:41:54 < emeb_mac> really like it.
2012-10-23T06:42:02 < gnomad> i've been looking at that...
2012-10-23T06:42:07 < emeb_mac> recommended.
2012-10-23T06:42:41 < gnomad> seems to be the only tool that does both openocd and serial wire debug.
2012-10-23T06:42:43 < dongs> i dunno how thats any better than $15 jlinkOB
2012-10-23T06:42:57 < dongs> which also does swv
2012-10-23T06:44:28 < gnomad> dongs: this http://www.segger.com/jlink-ob.html ???
2012-10-23T06:44:43 < dongs> ya
2012-10-23T06:44:54 < dongs> http://www.hotmcu.com/usbminijtag-jlink-ob-jtagswd-emulator-with-usb-interface-p-29.html?cPath=3_25 shit like this from china
2012-10-23T06:45:50 < dongs> i dunno how it works with freetard tools but jlink has a gdbserver (so theoretically you dont need openocd at all?)
2012-10-23T06:46:10 < Thorn> dongs: correct
2012-10-23T06:46:47 < Thorn> can't really say how well that gdbserver works though
2012-10-23T06:47:17 < dongs> i would suspect beign a commercial product, better than openocd.
2012-10-23T06:47:50 < dongs> those guys are probably still arguing if they should use binary version of FTDI libs or not
2012-10-23T06:48:05 < dongs> even if it makes jtag 10x faster, FUCK CLOSED SOURES right??
2012-10-23T06:48:19 < dongs> free as in beer before free as in GET MY FUCKING WORK DONE
2012-10-23T06:48:53 < Thorn> gdb is not very easy to deal with as I understand. not sure if segger bothered to polish its gdbserver
2012-10-23T06:49:04 < dongs> they've had gdbserver support for years.
2012-10-23T06:49:16 < dongs> im sure if tehre's any freetards using their shit, tehy keep them on thier toes wrt new features/making shit work.
2012-10-23T06:49:27 < gnomad> I actually rather like gdb in emacs....
2012-10-23T06:49:44 < gnomad> anyways, that is worth $15 to find out....
2012-10-23T06:49:59 < emeb_mac> looks OK
2012-10-23T06:50:04 < gnomad> I am really looking for this for the LPC1114....
2012-10-23T06:50:11 < dongs> well I have it and it works fine
2012-10-23T06:50:19 < dongs> but i just use keil/crossworks ide
2012-10-23T06:50:21 < gnomad> with an LPC?
2012-10-23T06:50:21 < dongs> which natively support it
2012-10-23T06:50:28 < dongs> no, but jtag is jtag
2012-10-23T06:50:30 < emeb_mac> just need to get yourself an adapter to convert the old-school 20-pin JTAG
2012-10-23T06:50:36 < dongs> thre's no reason wy it wouldnt work
2012-10-23T06:50:43 < emeb_mac> to SWD or 10-pin mini
2012-10-23T06:50:48 < dongs> yeah
2012-10-23T06:50:54 < dongs> i should sell my adapters.
2012-10-23T06:50:58 < emeb_mac> you should
2012-10-23T06:50:59 < dongs> i still have a shitlaod of blank pcbs left.
2012-10-23T06:51:08 < dongs> maybe even preassembled with that shitty 1.27mm conector
2012-10-23T06:51:11 < gnomad> dongs: do you use a paid version of keil/crossworks?
2012-10-23T06:51:14 < emeb_mac> throw 'em up on your website.
2012-10-23T06:51:28 < dongs> gnomad: correct. though freetard keil should work for you below 32kbytes.
2012-10-23T06:51:39 < dongs> paying for crossworks was the best thing ever.
2012-10-23T06:51:47 < dongs> even though its just gcc but.. somehow it feels less shit.
2012-10-23T06:51:54 < emeb_mac> dongs:  you use crossworks?
2012-10-23T06:51:58 < dongs> yea
2012-10-23T06:52:05 < dongs> for some stuff.
2012-10-23T06:52:09 < dongs> that has too many gccisms to edit out
2012-10-23T06:52:20 < emeb_mac> I bought a personal license of that a few years ago, never use it though
2012-10-23T06:52:24 < dongs> shit like named struct fills, __attribute__ shit etc
2012-10-23T06:52:26 < gnomad> I do like how there is a crossworks guy that hangs out on all the mailing lists for chips they support.
2012-10-23T06:52:30 < dongs> really? you should ugprade 2.2 or 2.3 or so is not so bad
2012-10-23T06:52:36 < emeb_mac> they've got all sorts of restrictions on what they want you to do with it.
2012-10-23T06:52:42 < dongs> emeb_mac: like whaT?
2012-10-23T06:52:47 < dongs> you mean personal license?
2012-10-23T06:52:51 < emeb_mac> "no commercial, no subcontract"
2012-10-23T06:52:53 < emeb_mac> yeah.
2012-10-23T06:52:58 < dongs> well, shrug.
2012-10-23T06:52:59 < emeb_mac> the $150 version.
2012-10-23T06:53:03 < dongs> i dont sell anything I build in it
2012-10-23T06:53:26 < dongs> people that do, just rebuild with freetard gcc :)
2012-10-23T06:53:39 < dongs> i dont use their ctl or any other shit
2012-10-23T06:53:40 < emeb_mac> Actually, been using CooCox for the last few days in Win7
2012-10-23T06:53:47 < emeb_mac> works rather nicely.
2012-10-23T06:53:52 < dongs> how does it flash/debug?
2012-10-23T06:53:55 < dongs> over gdbserver?
2012-10-23T06:53:57 < emeb_mac> seamless w/ the Discovery STLINK
2012-10-23T06:54:05 < dongs> hmm stlink
2012-10-23T06:54:06 < emeb_mac> yes - gdb debug is no problem
2012-10-23T06:54:08 < dongs> so that must be native?
2012-10-23T06:54:25 < dongs> i didnt wanna dick wiht it cuz >eclipse< ya kno but
2012-10-23T06:54:47 < emeb_mac> They skinned the Eclipse pretty well so it's not to annoying.
2012-10-23T06:54:55 < emeb_mac> s/to/too/
2012-10-23T06:55:32 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-23T06:55:59 < dongs> and rumor has it their entire  team are girls
2012-10-23T06:56:05 < dongs> paradise for Laurenceb
2012-10-23T06:56:24 < emeb_mac> who the heck are they? Why do they do it?
2012-10-23T06:56:46 < dongs> their "busienss plan" is make freetard shit and get paid from chip makers
2012-10-23T06:56:54 < dongs> apparneetly its working, at leat acording to last time i chatted with one of htem
2012-10-23T06:57:00 < emeb_mac> aha.
2012-10-23T06:57:29 < gnomad> emeb_mac: who are the people querying you about your codec board -- people in the synth community?
2012-10-23T06:58:21 < dongs> maybe they're just trolls
2012-10-23T06:58:21 < emeb_mac> gnomad: some synth geeks, some college guys doing projects, some folks just want a simple DSP platform.
2012-10-23T06:59:05 < gnomad> I have been thinking about doing an analog synth as an on-going project....
2012-10-23T06:59:08 < emeb_mac> some folks just glad I posted code for the I2S codec interface. Apparently getting that running is hard.
2012-10-23T06:59:19 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-23T06:59:25 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-23T06:59:25 < emeb_mac> gnomad: a lot of that going around these days.
2012-10-23T06:59:27 < gnomad> however, one of the real problems of a true analog VCO is stability...
2012-10-23T06:59:43 < gnomad> so I was thinking about cheating there and going the digital route.
2012-10-23T06:59:56 < gnomad> with something pretty much exactly as you designed.
2012-10-23T07:00:10 < emeb_mac> gnomad: heh - I make a fair bit of change on digital VCOs for synths. :D
2012-10-23T07:00:29 < emeb_mac> keeps me in Digi-Key/Mouser parts.
2012-10-23T07:00:36 < gnomad> oh really?
2012-10-23T07:00:51 < gnomad> you sell them as boards/kits/assembled devices?
2012-10-23T07:01:20 < emeb_mac> I work with a couple modular synth mfgs - I design the basic HW & SW - they productize it and sell to the musicians.
2012-10-23T07:01:29 < emeb_mac> I get royalties per-unit.
2012-10-23T07:02:12 < emeb_mac> I've got products sold to the likes of NIN, Vince Clarke, Daft Punk, deadmau5, etc.
2012-10-23T07:02:36 < gnomad> nice!
2012-10-23T07:03:31 < gnomad> I want to get into it to learn more about analog signal design
2012-10-23T07:03:37 < emeb_mac> it's a strange market - barely existed 10 yrs ago. Now it's exploding.
2012-10-23T07:04:05 < gnomad> yeah, there seems to be a huge resurgance in modular analog synths.
2012-10-23T07:04:15 < dongs> digital filters? pfft. everyone knows unless its analog and rackmount, youre just wasting time
2012-10-23T07:04:30 < emeb_mac> gnomad: check out the SDIY mail list - been around for years and has a lot of info about synth design.
2012-10-23T07:04:39 < dongs> either that or everything must go through roland tape echo
2012-10-23T07:04:45 < emeb_mac> heh
2012-10-23T07:04:57 < gnomad> yeah, i've found that
2012-10-23T07:05:10 < emeb_mac> dongs: yeah - there's a strong bias towards analog circuits
2012-10-23T07:05:30 < emeb_mac> but digital stuff is getting a foothold - can do things analog can't.
2012-10-23T07:05:38 < gnomad> yeah, there is no question that analog filters make a bigger sound
2012-10-23T07:06:01 < gnomad> i think a lot of people are realizing that even the best VSTs are no match for analog.
2012-10-23T07:06:14 < emeb_mac> well, a lot of the problem is that the digital implementations were too clean. Analog stuff has nasties that people like.
2012-10-23T07:06:47 < gnomad> I learned that when I was reading up on electric guitar design
2012-10-23T07:06:49 < emeb_mac> now the digital guys are figuring out how to model the whackiness that analog brings and it's getting less differentiated.
2012-10-23T07:07:13 < gnomad> *everything* in the signal chain is designed to color it in some way, shape or form.
2012-10-23T07:07:33 < gnomad> all the way from the response of the pickups to the use of an open-back speaker cabinet
2012-10-23T07:07:37 < emeb_mac> err - "designed" may be too strong a word.
2012-10-23T07:07:56 < gnomad> well, all the effects boxes are clearly "designed"
2012-10-23T07:08:05 < emeb_mac> it _does_ color it, but much is unintentional, but people have come to expect that color.
2012-10-23T07:08:19 < gnomad> perhaps
2012-10-23T07:08:26 < dongs> you just need a spool of dongs special oxygen-free speaker cable
2012-10-23T07:08:33 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-23T07:08:40 < emeb_mac> cheap at 1/2 the price, right? :)
2012-10-23T07:08:55 < gnomad> but in any case, I am not really a music geek.
2012-10-23T07:09:12 < gnomad> I just want something better than an atari-punk homebrew synth.
2012-10-23T07:09:19 < emeb_mac> heh.
2012-10-23T07:09:42 < emeb_mac> If you can figure out a fun way to generate waveforms there's a market for it.
2012-10-23T07:09:42 < gnomad> so, as you can see, my bar is pretty low.
2012-10-23T07:09:56 < gnomad> i've noticed that
2012-10-23T07:10:05 < emeb_mac> you'd be amazed at what kind of shit sells out there.
2012-10-23T07:10:09 < gnomad> I also like improvised group music.
2012-10-23T07:10:33 < gnomad> I'd like to be able to make a *lot* of modules on the cheap, and then let people play with them like legos.
2012-10-23T07:11:02 < emeb_mac> that's pretty much what the Eurorack modular market is all about.
2012-10-23T07:11:23 < gnomad> heh.  ok.
2012-10-23T07:11:37 < emeb_mac> see here: http://analoghaven.com/
2012-10-23T07:11:59 < gnomad> so are you a music maker as well?
2012-10-23T07:12:29 < dongs> make:R
2012-10-23T07:12:43 < emeb_mac> unfortunately no. I like listening and noodling around, but have no training or real skill at that.
2012-10-23T07:14:20 < gnomad> what really started my interest was a general sequencer to fire physical controls....
2012-10-23T07:15:03 < gnomad> the particular physical controls I need to use are *quite* expensive...
2012-10-23T07:15:22 < emeb_mac> like what?
2012-10-23T07:15:30 < gnomad> so I started thinking about making cheap electronic substitutes.
2012-10-23T07:15:49 < gnomad> flamable-gas safe solenoid valves...
2012-10-23T07:15:55 < gnomad> and HV spark initiators.
2012-10-23T07:16:12 < gnomad> I have the HV spark bit down.
2012-10-23T07:16:33 < gnomad> cheap gas-save valves, not so much
2012-10-23T07:17:12 < gnomad> I may also need oxygen-safe valves, too.
2012-10-23T07:18:17 < emeb_mac> doing some sort of pyro effects for shows?
2012-10-23T07:18:41 < gnomad> something like that.
2012-10-23T07:18:54 < gnomad> still in the developmental stages.
2012-10-23T07:19:29 < emeb_mac> glad it's your eyebrows and not mine ;)
2012-10-23T07:20:08 < gnomad> oh, I'm one of those people who dances with fire on the end of chains...
2012-10-23T07:20:18 < gnomad> and on the ends of stick...
2012-10-23T07:20:26 < gnomad> and the ends of hoops...
2012-10-23T07:21:19 < gnomad> anyways, all stems from a particularly crazy 4th of july party which started with oxy-acetlyne filled balloons and cigarettes...
2012-10-23T07:21:28 < gnomad> and then progressed upwards...
2012-10-23T07:22:21 < gnomad> a 50 gallon trash bag is *most* spectacular.
2012-10-23T07:22:46 < gnomad> BTW, it is your ears you need to worry about, not your eyebrows.
2012-10-23T07:25:43 < emeb_mac> WHAT?
2012-10-23T07:26:16 < zippe> SRL are local
2012-10-23T07:26:34 < emeb_mac> I wish I could see SRL
2012-10-23T07:26:49 < zippe> Minimum Safe Distance… bah
2012-10-23T07:27:00 < emeb_mac> They were scheduled to do a show here in PHX a few years back and the FD shut 'em down.
2012-10-23T07:27:52 < gnomad> get out to burningman.  ;-)
2012-10-23T07:28:43 < baird> My brother and his high-school friends made the news once with an acetlyne-powered UFO :)
2012-10-23T07:29:30 < gnomad> SRF is SF, though, right?
2012-10-23T07:29:35 < gnomad> s/srf/srl/
2012-10-23T07:30:15 < zippe> Oakland/Emereyville, last time I looked
2012-10-23T07:30:29 < zippe> Ah, Petaluma these days (further north)
2012-10-23T07:30:32 < gnomad> yeah, east bay
2012-10-23T07:30:55 < zippe> That's right, they got kicked out of E'ville due to the yuppy invasion
2012-10-23T07:30:55 < gnomad> for an east coaster, it's all just SF.
2012-10-23T07:31:06 < zippe> Or more precisely, the property speculators
2012-10-23T07:31:49 < gnomad> **warning ** graphic **  http://www.srl.org/yard/misc/injury.html
2012-10-23T07:32:12 < emeb_mac> zippe: yeah - I get the feeling that every toxic waste-dump in the Bay area is a gold mine.
2012-10-23T07:32:13 < gnomad> making rocket motors is one of my other hobbies.
2012-10-23T07:32:41 < emeb_mac> no matter what remediation costs, the upside outweighs it.
2012-10-23T07:32:47 < gnomad> been making my own propellant loads for a while now...
2012-10-23T07:32:54 < gnomad> just started turning my own casings.
2012-10-23T07:33:14 < gnomad> looking to get into liquid-hybrids next year.
2012-10-23T07:34:11 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-23T07:37:01 < baird> Always wanted to make Goddard replicas, personally. :)
2012-10-23T07:37:43 < gnomad> that would be an interesting project.
2012-10-23T07:38:02 < gnomad> liquid hybrids can be made pretty easily.
2012-10-23T07:38:26 < gnomad> and NOX is much nicer to work with than *any* other liquid oxidizer
2012-10-23T07:39:12 < gnomad> liquid/liquid adds a whole new dimension in difficulty.
2012-10-23T07:45:06 < emeb_mac> lolapple -> http://appleinsider.com/articles/12/10/22/unauthorized-third-party-lightning-authentication-chips-reportedly-in-production
2012-10-23T07:45:22 < emeb_mac> that didn't take long
2012-10-23T07:53:07 < gnomad> if i had an iphone5 i'd be wanting to get some of those...
2012-10-23T07:53:32 < zippe> It still doesn't do IAP
2012-10-23T07:53:56 < zippe> All that those parts will do is let you negotiate for the interfaces muxed onto the connector
2012-10-23T07:57:03 < emeb_mac> did you see the iPhone5 teardown? Notice how the flex circuit to the Lightning connector is serpentine within the case?
2012-10-23T07:57:33 < emeb_mac> like whatever phy chip they have driving it expects a certain stub length or no worky.
2012-10-23T08:12:35 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Quit: Leaving]
2012-10-23T08:32:38 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-23T08:32:38 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-23T08:32:38 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-23T08:35:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-23T08:36:22 < jpa-> Tectu: i did
2012-10-23T08:36:40 < jpa-> resolved with reset wire timing
2012-10-23T09:37:23 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-23T09:42:50 -!- ntfreak_ [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-23T09:43:02 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 246 seconds]
2012-10-23T10:00:27 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-23T10:03:26 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T10:13:28 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-23T10:13:53 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-23T10:20:16 -!- Steffanx [~Steffanx@client-215-040.flexnet2.rug.nl] has joined ##stm32
2012-10-23T10:20:17 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-23T10:38:12 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-23T10:40:09 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Quit: leaving]
2012-10-23T10:51:07 < dongs> The compiler accepts ISO C 1990 and 1999 source as input.whoaaaaaaa
2012-10-23T10:51:12 < dongs> i think keil just upgraded their  compiler
2012-10-23T10:52:18 < zyp> it just took them 13 years to support C99?
2012-10-23T10:52:36 < dongs> fuck, its impossible to find acutal changelog on arms gay site
2012-10-23T10:52:38 < dongs> ive seen it before
2012-10-23T10:52:41 < scrts> does keil support st-link?
2012-10-23T10:53:15 < dongs> http://www.keil.com/update/whatsnew.asp?p=RVMDK&v=4.60 yes finalyl found
2012-10-23T10:53:17 < dongs> scrts: of course
2012-10-23T10:54:17 < scrts> and maybe their interface is more like eclipse now? I really like that feature, when You push ctrl+ first mouse pedal to go to function declaration
2012-10-23T10:54:43 < dongs> no, but they do have go to definition/declaration in submenus
2012-10-23T10:55:01 -!- claude is now known as claude_work
2012-10-23T10:55:48 -!- claude_work is now known as Claude_work
2012-10-23T10:56:09 < scrts> I don't see STM32F4 added to support list :/
2012-10-23T10:56:58 < dongs> uh
2012-10-23T10:57:04 < dongs> ive been using keil with F4 for ever
2012-10-23T11:10:52 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-23T11:12:06 < scrts> hm, cool :>
2012-10-23T11:23:21 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Read error: Connection reset by peer]
2012-10-23T11:25:20 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T11:28:23 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Remote host closed the connection]
2012-10-23T11:35:10 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T11:37:19 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-23T12:14:36 < dongs> http://i.imgur.com/kXmhh.jpg  gotta turn this into a ebike
2012-10-23T12:17:35 < zyp> hmm, the pins lpc4300 are using for boot selection are same pins that are used for sdram interfacing
2012-10-23T12:18:06 < zyp> that's pretty dumb
2012-10-23T12:18:22 <+Steffanx> dongs found some old stuff in the basement?
2012-10-23T12:18:50 < Thorn> how does that steering work.
2012-10-23T12:19:05 < dongs> Thorn: pretty well apparently
2012-10-23T12:20:04 < baird> Better get some lights and indicators in it before going out onto the road.
2012-10-23T12:20:29 < Thorn> one wheel will lift into the air if you try to turn
2012-10-23T12:23:55 < karlp> heh, download datasheet on this guy: http://www.triadsemi.com/services/product-catalog/tsx1001/ just gives you a pdf of the page you were on.
2012-10-23T12:25:57 < dongs> awesome
2012-10-23T12:28:51 < karlp> seen this guy yet? http://www.youtube.com/watch?feature=player_embedded&v=P5j619mdesM
2012-10-23T12:29:40 < dongs> fake
2012-10-23T12:30:45 < karlp> only hard part is getting on/off the wire.
2012-10-23T12:31:25 < dongs> well apparently its not even a wire
2012-10-23T12:31:26 < dongs> more liek a rod
2012-10-23T12:31:40 < karlp> a very tight tightwire :)
2012-10-23T12:32:21 < karlp> On-Chip Memories - 16 KB boot ROM - 16 KB high-speed program memory - 32 KB high speed data memory - 32 KB high-speed communication - 512 KB Flash Memory with 4 KB instruction cache Communication Peripherals
2012-10-23T12:32:25 < karlp> too many memories!
2012-10-23T12:32:54 < baird> You probably would've hated the Amiga..
2012-10-23T12:37:27 < Claude_work> zyp, thats common on many SoC . normally just a pull up/down is required . after init the sdram controller 'overrides' the pull up/down
2012-10-23T12:38:12 < Thorn> sounds like ethernet phy config
2012-10-23T12:38:37 < Thorn> only those use way less than 10K for some reason
2012-10-23T12:39:21 < zyp> Claude_work, I don't really want any pullups/pulldowns on high speed signal lines
2012-10-23T12:39:50 < Claude_work> it's just SDR , or?
2012-10-23T12:39:57 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-23T12:41:56 < Claude_work> anyhow , 100k pull up/down vs. some pF parasitic pin capacity. guess which one does more harm to signal integrity/EMI
2012-10-23T12:42:15 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-23T12:44:08 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T12:46:30 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Client Quit]
2012-10-23T13:03:45 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-10-23T13:12:51 < dongs> PC13, PC14 and PC15 are supplied through the power switch. Since the switch only sinks a limited amount of current
2012-10-23T13:12:54 < dongs> (3 mA), the use of GPIOs PC13 to PC15 in output mode is limited: the speed should not exceed 2 MHz with a maximum
2012-10-23T13:12:57 < dongs> how about sinking a led?
2012-10-23T13:13:00 < dongs> load of 30 pF and these IOs must not be used as a current source (e.g. to drive an LED).
2012-10-23T13:13:32 -!- Steffanx [~Steffanx@client-215-040.flexnet2.rug.nl] has quit [Quit: This computer has gone to sleep]
2012-10-23T13:28:15 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-23T13:34:09 < zyp> hmm, interesting
2012-10-23T13:34:25 < zyp> lpc4300 datasheet has listed typical power consumption of each individual peripheral
2012-10-23T13:37:39 < dongs> um
2012-10-23T13:37:43 < dongs> so did st's?
2012-10-23T13:37:49 < dongs> i remember seeing the tables just yesterday
2012-10-23T13:37:52 < zyp> did it?
2012-10-23T13:37:54 < dongs> yes..
2012-10-23T13:37:57 < zyp> can't remember having seen it, but ok
2012-10-23T13:38:05 < dongs> there;'s a table,, lists each TIMx,, SPI, ADC, etc
2012-10-23T13:38:09 < dongs> they're all pretty tiny.. like 2-3mA
2012-10-23T13:38:31 < dongs> On-chip peripheral current consumption
2012-10-23T13:38:35 < dongs> in datasheet
2012-10-23T13:38:39 < dongs> page 48 in rev13
2012-10-23T13:38:41 < dongs> table 19
2012-10-23T13:38:45 < dongs> for F103
2012-10-23T13:38:50 < zyp> ok
2012-10-23T13:39:01 < dongs> TIM1 rapes a whole 1.6mA
2012-10-23T13:39:11 < dongs> ADC1 is worst.. 1.8m1
2012-10-23T13:39:13 < dongs> mA
2012-10-23T13:39:23 < zyp> timers in lpc4300 draws 0.15mA or so
2012-10-23T13:39:40 < zyp> ADC draws 0.07mA
2012-10-23T13:39:41 < dongs> looks like around 1.3mA average for ST timers
2012-10-23T13:39:43 < dongs> they must be amazing
2012-10-23T13:39:46 < dongs> LPC probably sucks
2012-10-23T13:39:52 < zyp> :p
2012-10-23T13:40:11 < zyp> worst is EMC at 7.95mA
2012-10-23T13:40:48 < dongs> wat a surprise
2012-10-23T13:41:16 < zyp> not really
2012-10-23T13:41:24 < dongs> </sarcasm>
2012-10-23T13:41:27 < zyp> the usb controllers also draws a lot
2012-10-23T13:41:39 < dongs> sdram needs constant refresh right?
2012-10-23T13:41:44 < zyp> yeah
2012-10-23T13:57:41 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T14:21:12 < dongs> zyp, what do you think about PC13-15 sinking leds
2012-10-23T14:21:14 < dongs> bad idea?
2012-10-23T14:28:07 < zyp> probably
2012-10-23T14:28:17 < dongs> gay.
2012-10-23T14:28:20 < dongs> those pins are fucking useless otherwise.
2012-10-23T14:29:05 < dongs> still trying to see how to make armcc C99 compliant
2012-10-23T14:32:01 < baird> use it to build gcc, duh.
2012-10-23T14:32:11 < dongs> lawl
2012-10-23T14:32:29 < Laurenceb> run gcc on an F4Discovery
2012-10-23T14:33:05  * Laurenceb fails at tilt corrected compass
2012-10-23T14:33:19 < dongs> Laurenceb: http://n0m1.com/2012/02/27/6dof-arduino-compass-accelerometer/
2012-10-23T14:33:21 < dongs> steal code
2012-10-23T14:33:55 < Laurenceb> hmm i dont trust arduino
2012-10-23T14:34:01 < dongs> ok, then http://www.freescale.com/files/sensors/doc/app_note/AN4248.pdf
2012-10-23T14:34:02 < Laurenceb> think ive worked it out now
2012-10-23T14:34:11 < Laurenceb> ah that looks trustable
2012-10-23T14:35:01 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-23T14:35:36 < Laurenceb> wtf
2012-10-23T14:35:42 < Laurenceb> thats the same as ST code
2012-10-23T14:35:45 < Laurenceb> hmf
2012-10-23T14:35:55 < Laurenceb> ok going to try this on my f3disco
2012-10-23T14:36:54 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T14:37:36 < dongs> Added: RTX_Blinky and LCD_Blinky examples for TI (Luminary) EK-LM4F232 evaluation board, based on LM4F232H5QD device, in folder ..\ARM\Boards\Luminary\EK-LM4F232\.
2012-10-23T14:37:39 < dongs> is that the stellaris?
2012-10-23T14:37:42 < dongs> ohh
2012-10-23T14:37:55 < dongs> oh, no
2012-10-23T14:37:57 < dongs> osme bigger shit
2012-10-23T14:41:29 < Laurenceb>       Dot_Norm = MagBuffer[2]/AccBuffer[2];
2012-10-23T14:41:42 < Laurenceb>       fTiltedX = MagBuffer[0] - AccBuffer[0] * (Dot_Norm);
2012-10-23T14:41:42 < Laurenceb>       fTiltedY = MagBuffer[1] - AccBuffer[1] * (Dot_Norm);
2012-10-23T14:41:51 < Laurenceb>       HeadingValue = (float) ((atan2f((float)fTiltedY,(float)fTiltedX))*180)/PI;
2012-10-23T14:41:58 < Laurenceb> surely it cant be that easy
2012-10-23T14:42:51 < badger> Laurenceb: that assumes the only forces acting on the accelerometer are due to gravity
2012-10-23T14:42:59 < Laurenceb> i know
2012-10-23T14:43:16 < Laurenceb> tilt corrected compass is lame by definition
2012-10-23T14:43:32 < Laurenceb> but i cant even get anything sane out of it atm
2012-10-23T14:44:08 < badger> do the raw values look roughly correct?
2012-10-23T14:44:09 < Laurenceb> im thinking use accel vector to cancel the z component in board coordinates
2012-10-23T14:44:15 < Laurenceb> i think so
2012-10-23T14:44:26 < baird> It's 2012, and I'm using kermit to transfer files... :/
2012-10-23T14:44:28 < Laurenceb> then atan2 the x and y in board space
2012-10-23T14:45:04 < badger> are you saying that the mag values are not sane?
2012-10-23T14:47:13 < badger> you can try this to see how skewed your axis are: https://dev.qu.tu-berlin.de/projects/sf-razor-9dof-ahrs/wiki/Tutorial#Extended-magnetometer-calibration
2012-10-23T14:48:22 < Laurenceb> yeah
2012-10-23T14:48:30 < Laurenceb> 6 point cal helps a lot
2012-10-23T14:55:00 < Laurenceb> very odd
2012-10-23T14:55:06 < Laurenceb> i can only ever move right
2012-10-23T14:55:50 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-23T14:58:53 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-23T14:58:55 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-23T14:59:28 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-10-23T14:59:46 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-23T15:10:50 < Laurenceb> nope its utterly random now
2012-10-23T15:11:44 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-23T15:11:54 < Laurenceb> static uint8_t HID_Buffer[4] = {0};
2012-10-23T15:12:08 < Laurenceb> i dont understnad why its unsigned
2012-10-23T15:12:35 < dongs> ??
2012-10-23T15:12:36 < zyp> because it's a buffer to be passed to the usb stack?
2012-10-23T15:13:41 < Laurenceb> yeah but they subtract from it
2012-10-23T15:13:56 < Laurenceb> HID_Buffer[1] += CURSOR_STEP;
2012-10-23T15:13:56 < zyp> so?
2012-10-23T15:14:00 < Laurenceb> HID_Buffer[1] -= CURSOR_STEP;
2012-10-23T15:14:05 < Laurenceb> hmf
2012-10-23T15:14:30 < Laurenceb> something is badly screwed
2012-10-23T15:14:39 < Laurenceb> if i use uint8_t i get no movement at all
2012-10-23T15:14:57 < zyp> because of two's complement, addition and subtraction is exactly the same
2012-10-23T15:15:02 < Laurenceb> indicator leds are working correctly now
2012-10-23T15:15:27 < Laurenceb> but the mouse is still screwed
2012-10-23T15:16:13 < Laurenceb> http://pastebin.com/g2YDYbkK
2012-10-23T15:16:16 < Laurenceb> current code
2012-10-23T15:16:23 < Laurenceb> that produces no motion
2012-10-23T15:17:05 < zyp> so, what is calling USBD_HID_GetPos?
2012-10-23T15:17:45 < zyp> when I tested with analyzer yesterday I observed that no reports were sent at all
2012-10-23T15:17:58 < Laurenceb> Mouse_Buffer = USBD_HID_GetPos( HeadingValue-128, PitchAng-128);
2012-10-23T15:18:08 < Laurenceb> yeash i added that function
2012-10-23T15:18:15 < Laurenceb> from usb example code
2012-10-23T15:18:27 < Laurenceb> its in the main loop
2012-10-23T15:18:31 < Laurenceb> same as the example
2012-10-23T15:18:56 < Laurenceb> if i change to int8_t it works.. badly
2012-10-23T15:19:04 < Laurenceb> only ever moves right
2012-10-23T15:20:27 < zyp> wtf are those ifs in the start of the code you pasted?
2012-10-23T15:21:20 < Laurenceb> ok ill make that more sane
2012-10-23T15:26:08 < Laurenceb> ok wtf
2012-10-23T15:26:12 < Laurenceb> this makes no sense
2012-10-23T15:26:25 < Laurenceb> im going to have to grab a load of attitude data out of it i think
2012-10-23T15:26:36 < Laurenceb>  ... yet leds look sane
2012-10-23T15:26:56 < Laurenceb> its got to be in that function somewhere that the issue is
2012-10-23T15:27:47 < Laurenceb> http://pastebin.com/Xyt6CwU5
2012-10-23T15:28:41 < Laurenceb> not it only ever moves right
2012-10-23T15:34:32 < Laurenceb>   if((Delta_Heading) < -10)
2012-10-23T15:34:32 < Laurenceb>   {
2012-10-23T15:34:32 < Laurenceb>    HID_Buffer[1] += (Delta_Heading+10)/CURSOR_STEP;
2012-10-23T15:34:38 < Laurenceb> ^looks like that doesnt work
2012-10-23T15:34:47 < Laurenceb> HID_Buffer[1] is left as 0
2012-10-23T15:35:07 < zyp> try hardcoding it to 255
2012-10-23T15:35:10 < zyp> (-1)
2012-10-23T15:35:21 < zyp> should move it left
2012-10-23T15:38:20 < scrts> Laurenceb: whats the type os Delta_Heading?
2012-10-23T15:38:24 < scrts> *of
2012-10-23T15:38:27 < scrts> how it is declared?
2012-10-23T15:38:54 < zyp> float
2012-10-23T15:39:24 < scrts> hm.. and it is really float? not signed int?
2012-10-23T15:39:41 < zyp> pasted code says float.
2012-10-23T15:40:04 < Laurenceb> zyp ok ill try hardcoding
2012-10-23T15:45:13 < Laurenceb> hmm that works
2012-10-23T15:45:55 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-23T15:47:51 < zyp> as expected, then you know that the problem is that the code is somehow only stuffing values of 0-127 into that byte
2012-10-23T15:50:38 < BrainDamage> http://www.urbandictionary.com/define.php?term=flyback
2012-10-23T15:50:50 < zyp> old
2012-10-23T15:51:04 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-23T15:53:46 -!- Oldboy is now known as `````
2012-10-23T15:53:52 -!- ````` is now known as Oldboy
2012-10-23T15:55:12 < baird> | buy flyback mugs & shirts |
2012-10-23T16:13:02 < dongs> http://bcas.tv/paste/results/XUZyFx42.html any suggestions on this setup?
2012-10-23T16:14:55 < Tectu> has anyone any sourcecode to display a mandelbrot ?
2012-10-23T16:14:59 < Laurenceb> https://github.com/Laurenceb/STM32F3Discovery
2012-10-23T16:15:03 < Laurenceb> ok works for me
2012-10-23T16:15:13 < Laurenceb> had to use atan2 on the pitch
2012-10-23T16:15:35 < karlp> there's enough headroom to do 1.6v->1.5V reliably on teh LDO?
2012-10-23T16:15:36 < Laurenceb> its a little goofy - needs calibrated magno ideally, and the loop sped up a bit
2012-10-23T16:15:55 < dongs> Tectu: http://bcas.tv/paste/results/4BWGmL51.html
2012-10-23T16:16:13 < Tectu> thanks dongs
2012-10-23T16:16:14 < dongs> karlp: yeah, i mean even shit like LM1117 should do it, no?
2012-10-23T16:16:32 < Tectu> dongs, anything I should know about it? anything special how to use it?
2012-10-23T16:16:36 < dongs> i've got lm1117s in 4.5V>1.2V stuff and they're working, a bit warmer than i would like though
2012-10-23T16:16:44 < dongs> Tectu: just run loop() in a loop i guess.
2012-10-23T16:16:48  * karlp shrugs, I know a lot of them say 70mv drop out, but I wouldn't hve normally tried to run it that way all the time though.
2012-10-23T16:18:05 < zyp> wouldn't it be better to just put some filtering after each switcher instead of a LDO?
2012-10-23T16:18:31 < dongs> zyp, 5V and 1.5V needs like < 7mV ripple
2012-10-23T16:19:58 < Tectu> dongs, can I use the code in a demo for ChibiOS/GFX ?
2012-10-23T16:20:09 < dongs> Tectu: why not, i forgot where i ripped it from
2012-10-23T16:20:57 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-23T16:23:06 < Laurenceb> http://www.bbc.co.uk/newsbeat/20019519
2012-10-23T16:23:22 < Laurenceb>  "After 47 hours of logging"
2012-10-23T16:23:26 < Laurenceb> more like fapping
2012-10-23T16:23:57 -!- mrcan_ [~mrcan@unaffiliated/mrcan] has quit [Ping timeout: 252 seconds]
2012-10-23T16:24:30 < baird> Tectu: something I used to test the hardfloat support in libopemcm3: http://kildall.apana.org.au/~cjb/mandel.c
2012-10-23T16:26:05 < Tectu> baird, PRINTCHAR(mandel())   ?
2012-10-23T16:26:16 < Tectu> it does return chars?
2012-10-23T16:27:30 < baird> A legacy artifact from ~20 year-old code of mine :) .. int, rather. The escape codes are for xterm, too.
2012-10-23T16:27:31 < Tectu> dongs, got entire project to that code? what's the variable "mag" ?
2012-10-23T16:27:48 < Tectu> hrhr
2012-10-23T16:28:01 < zyp> magnification
2012-10-23T16:28:07 < dongs> float mag = 1;
2012-10-23T16:28:07 < dongs> float offsetx = -0.756423894274328;
2012-10-23T16:28:07 < dongs> float offsety = 0.064179410646170;
2012-10-23T16:28:14 < dongs> sorry.
2012-10-23T16:28:16 < dongs> was at top of file
2012-10-23T16:28:24 < Tectu> ah, no problem :)
2012-10-23T16:28:24 < Tectu> thanks
2012-10-23T16:28:55 < Tectu> dongle, are there more declarations? FrameBuffer for example?
2012-10-23T16:29:15 < dongs> thats just y our own shit man
2012-10-23T16:29:18 < dongs> wahtever youre drawing to
2012-10-23T16:29:30 < zyp> mandelbrot is pretty easy to implement yourself
2012-10-23T16:29:30 < baird> zooming in on (-0.55, -0.55) is rather funky up until you hit the limits of floats.
2012-10-23T16:29:34 < dongs> you get pixel data, mine was 16bit 565 framebuffer
2012-10-23T16:31:01 < Tectu> dongs, one more thing...    count   ?
2012-10-23T16:31:06 < dongs> int = 0
2012-10-23T16:31:07 < zyp> animation
2012-10-23T16:31:37 < zyp> see how it is panning and moving each time it's drawn
2012-10-23T16:31:45 < zyp> panning and zooming*
2012-10-23T16:31:52 < zyp> except panning is commented out
2012-10-23T16:34:19 < baird> cellular automata are another neat thing for screenhacks.. life, hodgepodge, etc.
2012-10-23T16:34:20 < zyp> http://paste.jvnv.net/view/1XvKX <- here's some shit I wrote once
2012-10-23T16:34:40 < dongs> cool python bro
2012-10-23T16:34:52 < zyp> no, it's GLSL
2012-10-23T16:35:02 < zyp> i.e. GPU code
2012-10-23T16:35:03 < dongs> wat
2012-10-23T16:35:28 < zyp> it's a fragment shader
2012-10-23T16:37:10 < zyp> http://bin.jvnv.net/f/4ztUH.png <- here's some early version of it running in a shader testing software
2012-10-23T16:37:39 < baird> http://hunter.apana.org.au/~cjb/Code/hodge.c .. I used to use this as a text console hack.. up until machines hit ~50MHz and it animated too quickly to follow :)
2012-10-23T16:39:14 < Tectu> dongs, got it working, thanks!
2012-10-23T16:39:46 < Tectu> dongs, kinda slow on F4 + FSMC display....
2012-10-23T16:39:55 < dongs> ./hodge
2012-10-23T16:39:55 < dongs> Segmentation fault
2012-10-23T16:40:03 < dongs> Tectu: compile wiht hardfloat.
2012-10-23T16:40:11 < Tectu> renders an image every 10s
2012-10-23T16:40:22 < dongs> you're definitely doing it wrong
2012-10-23T16:40:43 < Tectu> dongle, turned FPU on, 1000 times faster
2012-10-23T16:41:00 < dongs> the rest of the slowness must be blamed on kawaiiOS
2012-10-23T16:42:49 < dongs> http://news.bbcimg.co.uk/media/images/63652000/jpg/_63652159_hi004614637.jpg found a pic of cjbeard
2012-10-23T16:43:13 < Tectu> dongs, how do i compile with hardfloat? o0
2012-10-23T16:43:21 < Tectu> what's   "hardfloat" at all? FPU support?
2012-10-23T16:43:58 < zyp> yes
2012-10-23T16:44:09 < dongs> if you dont know at this point, maybe give up
2012-10-23T16:44:37 < zyp> people call it hardfloat to differentiate it from softfloat, which is software FPU emulation
2012-10-23T16:48:05 < Laurenceb> i fap to hardcore float
2012-10-23T16:48:29 < Tectu> dongs, i am running USB, FATFS, Ethernet and LWIP the same time >.<
2012-10-23T16:48:49 < Laurenceb> your mum
2012-10-23T16:50:54 < scrts> Tectu: with RTOS? ;)
2012-10-23T16:51:12 < Tectu> scrts, ChibiOS
2012-10-23T16:51:14 < Tectu> scrts, #chibios
2012-10-23T16:57:56 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-23T17:04:46 < Tectu> dongs, what exactly does this line do, and are that 512 relative to the 128?   for(t=0;t<512;t++) {
2012-10-23T17:05:42 < zyp> no
2012-10-23T17:05:51 < zyp> 512 is the max number of iterations to do
2012-10-23T17:06:04 < zyp> you can choose it arbitrarily
2012-10-23T17:06:37 < zyp> the more iterations you do, the more details your fractal will have
2012-10-23T17:06:52 < zyp> and of course, the longer time it will take
2012-10-23T17:07:45 < Laurenceb> http://en.wikipedia.org/wiki/National_Masturbation_Day
2012-10-23T17:07:53 < zyp> it's the same as i in my GLSL code
2012-10-23T17:08:59 < Laurenceb> ^lol
2012-10-23T17:21:22 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-23T17:23:39 < Tectu> thanks zyp
2012-10-23T17:26:26 < dongs> beard: still waiting for instructions how to make hodge NOT CRASH
2012-10-23T17:27:20 < Laurenceb> how do i add hardfloat in a makefile?
2012-10-23T17:30:25 < zyp> https://launchpadlibrarian.net/109296007/readme.txt has a good overview on compiler flags
2012-10-23T17:30:38  * Steffanx bookmarks
2012-10-23T17:31:20 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Read error: Connection reset by peer]
2012-10-23T17:31:41 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-23T17:33:23 < karlp> does m0 let you have C only startup code?
2012-10-23T17:33:33 < zyp> sure
2012-10-23T17:33:52 < karlp> so, all of those devices in that table have C only startup as a feature, yet they provide asm startup files?
2012-10-23T17:34:16 < Laurenceb> error: lib/STM32F30x_StdPeriph_Driver/src/stm32f30x_adc.o uses VFP register arguments, main.elf does not
2012-10-23T17:34:20 < Laurenceb> the hell
2012-10-23T17:34:34 < zyp> armv6-m is pretty much armv7-m with fewer instructions
2012-10-23T17:35:07 < zyp> Laurenceb, your toolchain are missing a library variant with hardfloat-abi
2012-10-23T17:35:20 < Laurenceb> hmm
2012-10-23T17:35:21 < zyp> use -mfloat-abi=softfp
2012-10-23T17:35:26 < zyp> or get a better toolchain
2012-10-23T17:35:36 < Laurenceb> codesourcery is supposed to support fpu
2012-10-23T17:35:45 < Laurenceb> maybe i need a newer version?
2012-10-23T17:35:59 < zyp> sure, softfloat-abi still allows fpu usage
2012-10-23T17:36:03 < Laurenceb> oh
2012-10-23T17:36:13 < dongs> fuck, he /quit
2012-10-23T17:36:13 < zyp> but retains compatibility with code compiled for non-fpu
2012-10-23T17:36:17 < Laurenceb> oh, thats the math functions?
2012-10-23T17:36:49 < zyp> it's irrelevant, the error is due to the metadata of the file itself
2012-10-23T17:36:54 < Laurenceb> right
2012-10-23T17:37:15 < Laurenceb> ok it compiled
2012-10-23T17:37:19 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-23T17:37:21 < Laurenceb> but binary is slightly larger
2012-10-23T17:37:40 < zyp> than?
2012-10-23T17:38:25 < Tectu> dongs, can I improve the performance somehow?
2012-10-23T17:38:30 < Tectu> I get about 2fpos
2012-10-23T17:38:32 < Tectu> fps*
2012-10-23T17:41:12 < Laurenceb> than with pure soft float
2012-10-23T17:43:43 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 244 seconds]
2012-10-23T17:46:50 < qyx_> is anyone familiar with F4 dsp implementation?
2012-10-23T17:47:31 < jpa-> you mean the ARM's dsp library?
2012-10-23T17:47:39 < qyx_> are there any extended instructions? another dsp coprocessor? do i need toolchain support? or some libs?
2012-10-23T17:47:58 < jpa-> it has some 8-bit and 16-bit packed SIMD instructions, but nothing too fancy
2012-10-23T17:48:00 < qyx_> ah, it is somehow standardised in arm/cm4?
2012-10-23T17:48:13 < qyx_> so it is not stm specific?
2012-10-23T17:48:13 < jpa-> yeah, just the normal cortex-m4f stuff
2012-10-23T17:48:19 < qyx_> hm, okay
2012-10-23T17:48:46 < jpa-> i think they think that anything with multiply-accumulate qualifies as "DSP" ,)
2012-10-23T17:49:23 < qyx_> i was mainly thinking of some matrix manipulations
2012-10-23T17:49:32 < dongs> there's the dsp lib w/sores
2012-10-23T17:49:40 < dongs> in v4 stdperiphlib.
2012-10-23T17:49:44 < dongs> er M4
2012-10-23T17:49:45 -!- Claude_work is now known as claude
2012-10-23T17:49:51 < dongs> and preubuilt shit
2012-10-23T17:49:55 < jpa-> w/sores?
2012-10-23T17:49:57 < dongs> for arm and gcc compilers as a.lib you can link
2012-10-23T17:49:58 < dongs> yes
2012-10-23T17:50:04 < qyx_> sources in donglang
2012-10-23T17:50:04 < jpa-> what the heck does that mean?
2012-10-23T17:50:07 < jpa-> aha
2012-10-23T17:50:07 < jpa-> :)
2012-10-23T17:50:13 < dongs> yes, opensores.
2012-10-23T17:50:18 < jpa-> yeah, i guess it is just the DSP lib from ARM?
2012-10-23T17:50:28 < dongs> more or less, maybe with extra ST headers.
2012-10-23T17:51:16 < jpa-> qyx_: if you want float matrices, no, there is not much special instructions; and you are better off using some random matrix library than the dsp lib (because the dsp lib doesn't have much matrix stuff)
2012-10-23T17:51:32 < qyx_> actually
2012-10-23T17:51:40 < qyx_> i was looking at some analog devices parts
2012-10-23T17:51:57 < qyx_> some blackfin or whatever 400MHz dsp
2012-10-23T17:52:07 < qyx_> they do matrix multiply in 15ns
2012-10-23T17:52:46 < jpa-> for what size of matrix?
2012-10-23T17:52:52 < zyp> 4x4?
2012-10-23T17:52:54 < qyx_> 3x3
2012-10-23T17:53:07 < zyp> that's just nine multiplications?
2012-10-23T17:53:11 < qyx_> 3x3 multiplied with 1x3
2012-10-23T17:53:14 < zyp> and six additions
2012-10-23T17:53:43 < qyx_> hm, wait
2012-10-23T17:54:53 < zyp> 1/15ns is 67 MHz or so
2012-10-23T17:55:09 < jpa-> so 6 clocks
2012-10-23T17:55:14 < zyp> yeah
2012-10-23T17:56:32 < zyp> how many multipliers does it use?
2012-10-23T17:56:38 < qyx_> http://www.analog.com/en/processors-dsp/sharc/adsp-21483/products/product.html
2012-10-23T17:56:41 < qyx_> see
2012-10-23T17:56:52 < qyx_> not blackfin, sharc
2012-10-23T17:57:45 < qyx_> there are some benchmarks on 3rd page
2012-10-23T17:58:03 < jpa-> hmm, i wonder why http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0439c/BEHJADED.html  says that multiply is 1 cycle, addition is 1 cycle, but fuse mul+add is 3 cycles
2012-10-23T17:58:40 < zyp> pipelining
2012-10-23T17:59:06 < zyp> those numbers for seperate operations likely assume operating on seperate registers
2012-10-23T17:59:11 < jpa-> but why would anyone use the fused instruction, instead of interleaving mul & adds
2012-10-23T17:59:16 < zyp> so addition doesn't have to wait for multiplication to be ready
2012-10-23T17:59:32 < jpa-> i like the intel manuals that list separately "throughput" and "latency"
2012-10-23T18:00:00 < jpa-> but anyway, would be about 20 clocks on STM32F4.. so 4x faster for the adsp
2012-10-23T18:00:36 < zyp> interleaved muls and adds would probably be faster as long as you have operations on independent datasets that you can interleave
2012-10-23T18:01:02 < jpa-> yeah, quite easy to do in the example of 3x3 * 1x3 matrix mul
2012-10-23T18:01:14 < zyp> did you read the notes at the bottom?
2012-10-23T18:01:15 < zyp> Floating-point arithmetic data processing instructions, such as add, subtract, multiply, divide, square-root, all forms of multiply with accumulate, as well as conversions of all types take one cycle longer if their result is consumed by the following instruction.
2012-10-23T18:01:47 < jpa-> ah, yeah, just noticed it
2012-10-23T18:01:54 < qyx_> floating point is also 1 cycle?ň
2012-10-23T18:02:05 < qyx_> add/sub
2012-10-23T18:02:13 < zyp> this is kind of neat:
2012-10-23T18:02:14 < zyp> Both fused and chained multiply with accumulate instructions consume their addend one cycle later, so the result of an arithmetic instruction that is followed by a multiply with accumulate instruction is consumed as the addend of the MAC instruction.
2012-10-23T18:02:44 < jpa-> so fused instructions are basically just as fast but take less code space
2012-10-23T18:02:49 < jpa-> qyx_: yeah
2012-10-23T18:02:50 < zyp> exactly
2012-10-23T18:02:55 < jpa-> qyx_: with fpu, of course
2012-10-23T18:03:50 < jpa-> the code size thing may be quite important if flash wait states are considered
2012-10-23T18:04:25 < zyp> remember that F4 has an instruction cache
2012-10-23T18:04:48 < Laurenceb> f3discovery mouse demo is working now
2012-10-23T18:04:50 < Laurenceb> more or less
2012-10-23T18:04:53 < Laurenceb> its on github
2012-10-23T18:05:40 <+Steffanx> I'm glad not every programmer thinks like that Laurenceb "more or less"
2012-10-23T18:05:47 < zyp> :)
2012-10-23T18:05:56 < qyx_> square root 14, hm
2012-10-23T18:06:13 <+Steffanx> pi + something :P
2012-10-23T18:06:38 < Laurenceb> http://spacenear.us/tracker/
2012-10-23T18:06:40 < jpa-> zyp: yeah, though i have seen people claim significant wait states despite that (don't know if they just measured wrong)
2012-10-23T18:06:42 < Laurenceb> looks a bit gay
2012-10-23T18:07:47 <+Steffanx> I think you have some malware on your pc, i don't see anything that looks gay
2012-10-23T18:08:20 < zyp> well, it's landing close to Gaydon
2012-10-23T18:08:34 < zyp> inb4 «moar like Gaydong»
2012-10-23T18:08:43 < qyx_> wat
2012-10-23T18:08:43 < jpa-> zyp: btw, do you know how large is the F4 instruction cache?
2012-10-23T18:08:47 < qyx_> two baloons?
2012-10-23T18:08:56 < zyp> jpa-, hang on
2012-10-23T18:09:05 < zyp> qyx_, two payloads on one
2012-10-23T18:09:48 < zyp> jpa-, 1k
2012-10-23T18:10:22 < zyp> flash is 128-bit wide, instruction cache has 64 cache lines and data cache has 8 cache lines
2012-10-23T18:10:44 <+Steffanx> jpa-, did you use the comp. of the L1 to wake it up?
2012-10-23T18:11:10 < jpa-> Steffanx: i did not
2012-10-23T18:11:25 < jpa-> zyp: thanks
2012-10-23T18:11:38 <+Steffanx> Hmm, i guess the DAC doesn't work in stop mode, so that means in need a external reference voltage..
2012-10-23T18:11:43 < qyx_> so it is not worth it to try that adsp thing probably
2012-10-23T18:12:01 <+Steffanx> */in/i
2012-10-23T18:13:24 <+Steffanx> and skip that "I guess"
2012-10-23T18:14:07 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-23T18:14:46 < jpa-> Steffanx: may be easier just to wake up e.g. once a second and do one ADC conversion before going back to sleep
2012-10-23T18:15:34 <+Steffanx> Yeah, I exploring my option ( and that was one of them )
2012-10-23T18:15:37 <+Steffanx> *I'm
2012-10-23T18:15:50 < jpa-> am i one of your options?
2012-10-23T18:16:16 <+Steffanx> You want to be one of them?
2012-10-23T18:16:29 < jpa-> no
2012-10-23T18:16:41 <+Steffanx> Ok, then you're not on the list
2012-10-23T18:16:44 < jpa-> :(
2012-10-23T18:16:56 <+Steffanx> You only want to be the first choice, not?
2012-10-23T18:19:58 < zyp> Steffanx, are you finally making something cool?
2012-10-23T18:20:20 <+Steffanx> I have no idea if it's cool iyho
2012-10-23T18:21:23 < jpa-> everything Steffanx does is cool
2012-10-23T18:21:48 < zyp> sure, but he never does anything :p
2012-10-23T18:22:24 <+Steffanx> You have no idea WHAT i do Mr. zyp
2012-10-23T18:22:30 <+Steffanx> Anyway, thank you ST
2012-10-23T18:23:13 < zyp> anyway, time to boot windows
2012-10-23T18:23:20 <+Steffanx> Whaat?
2012-10-23T18:23:49 <+Steffanx> Thank you ST for no ADC input on the same pin as the comp2 input :(
2012-10-23T18:26:07 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-23T18:26:40 < Laurenceb> ST hit WTC
2012-10-23T18:26:57 < jpa-> maybe you should have gone with the retarded little brother comp1 instead
2012-10-23T18:27:16 <+Steffanx> I could do that jpa-
2012-10-23T18:27:32 <+Steffanx> but that makes no sense :)
2012-10-23T18:27:56 < jpa-> you could tie the pins together externally, though
2012-10-23T18:28:27 < Laurenceb> http://www.media-criticism.com/New_York_City_Nuked_2006_911.html
2012-10-23T18:28:59 < jpa-> or have a capacitor high-pass there so that you don't need to care about the reference voltage for comparator, as long as your changes are sharp enough
2012-10-23T18:37:09 <+Steffanx> I'm here to babysit you and jpa-, zyp
2012-10-23T18:37:20 < zyp> hmm
2012-10-23T18:37:52 < zyp> i flashed new bootloader to my BMP, apparently core still runs into lockup
2012-10-23T18:40:55 < Laurenceb> http://www.chiaki.cc/Pyxis2010/images/pyxis2010-fpgasol2.jpg
2012-10-23T18:41:00 < Laurenceb> ^insane
2012-10-23T18:41:20 < Laurenceb> zyp: interesting, thats what happens on my maple clone
2012-10-23T18:41:33 < jpa-> Laurenceb: looks a bit like your code
2012-10-23T18:42:21 < Laurenceb> http://www.chiaki.cc/Pyxis2010/images/pyxis2010-fpgasol1.jpg
2012-10-23T18:44:36 -!- barthess [~barthess@178.154.58.185] has joined ##stm32
2012-10-23T18:44:46 < zyp> maybe it has some weird incompatibilities with my toolchain
2012-10-23T18:45:03 < zyp> can anybody else build the latest BMP firmware for me?
2012-10-23T18:45:46 < karlp> just top of tree?
2012-10-23T18:46:00 < zyp> yes
2012-10-23T18:47:17 < karlp> ugh, that git submodule for libopecm3 just cloned as an empty local directory
2012-10-23T18:47:32 < zyp> run git submodule update
2012-10-23T18:48:14 < karlp> did nothing
2012-10-23T18:48:39 < zyp> oh, you have to run git submodule init the first time
2012-10-23T18:48:53 < zyp> then perhaps git submodule sync
2012-10-23T18:48:56 < zyp> then update
2012-10-23T18:49:45 < zyp> I forgot the exact order you are supposed to do them in, but it goes something like that
2012-10-23T18:50:03 < karlp> init was the important one :=)
2012-10-23T18:50:17 < zyp> yeah, it's the one creating the submodule repo
2012-10-23T18:50:33 < zyp> and I believe sync will add the remote to that repo from the current .gitmodules
2012-10-23T18:50:42 < karlp> what files do you want where?
2012-10-23T18:50:59 < zyp> blackmagic_dfu.bin, you can put it at bin.jvnv.net
2012-10-23T18:52:18 < karlp> http://palmtree.beeroclock.net/~karlp/bmp/
2012-10-23T18:52:32 < zyp> thanks
2012-10-23T18:52:42 < karlp> that's with arm-none-eabi-gcc (GNU Tools for ARM Embedded Processors) 4.6.2 20120613 (release) [ARM/embedded-4_6-branch revision 188521]
2012-10-23T18:52:49 < karlp> no guarantees it's any better than yours :)
2012-10-23T18:52:51 < Laurenceb> http://palmtree.beeroclock.net/~karlp/35821e.jpg
2012-10-23T18:53:05 < karlp> and?
2012-10-23T18:53:14 < zyp> ah, thanks, it did the trick
2012-10-23T18:54:12 < Tectu> dongs, why does it restart after those 80 loops? shouldn't it go to inifinity?
2012-10-23T18:54:15 < Laurenceb> karlp fap material
2012-10-23T18:54:21 < Tectu> Laurenceb, wtf
2012-10-23T18:54:33 < Laurenceb> on karlp server
2012-10-23T18:54:52 < karlp> oh noes!
2012-10-23T18:55:05 < Tectu> Laurenceb, who cares?
2012-10-23T18:55:26 <+Steffanx> Your gf does
2012-10-23T18:55:35 < Tectu> na, she's bisexual
2012-10-23T18:55:44 <+Steffanx> heh
2012-10-23T18:56:26 < Tectu> so, here's the mandelbrot, running really really slow:   http://youtu.be/NlqijOg_TJg
2012-10-23T18:57:22 <+Steffanx> 0.5fps ?
2012-10-23T18:57:39 < Tectu> yeah
2012-10-23T18:57:47 < Tectu> it's getting faster after 40 and 50s
2012-10-23T18:57:49 < Tectu> no idea why
2012-10-23T18:57:50 <+Steffanx> The auto focus of your cam doesn't like it
2012-10-23T18:58:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-23T18:59:08 < Tectu> Steffanx, yeah, as mentioned in description, it's shitty smartphone camera
2012-10-23T18:59:14 < Tectu> it dosen't look that bad on the screen
2012-10-23T19:00:17 < zyp> hmm, right, the same issue that broke the bootloader apparently also breaks the firmware itself
2012-10-23T19:00:36 < zyp> that's why I was having problems when I was trying to build it myself
2012-10-23T19:00:43 < zyp> karlp's build works nicely
2012-10-23T19:00:48 < zyp> so thanks again
2012-10-23T19:01:32 < karlp> how did that affect the build though?
2012-10-23T19:01:45 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-23T19:01:47 < zyp> I have no idea
2012-10-23T19:02:25 < zyp> the most probable reason I can imagine is that the newlib shipped with yagarto is having some issues with libopencm3
2012-10-23T19:02:58 < zyp> or perhaps compilation flags, maybe it's somehow not selecting the correct multilib variant
2012-10-23T19:03:54 < zyp> either way, it's probably not an issue just affecting me, so it might be worth looking into
2012-10-23T19:04:31 < zyp> especially since I was planning to work a bit on BMP firmware to add F3 support and so on
2012-10-23T19:04:40 < zyp> :p
2012-10-23T19:05:47 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-23T19:05:49 <+Steffanx> jpa-, that is cap. isn't an option
2012-10-23T19:05:55 <+Steffanx> the changes are VERY slow
2012-10-23T19:07:11 <+Steffanx> So external connection it is
2012-10-23T19:32:04 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-23T19:34:10 < zyp> 17:57:47 < Tectu> it's getting faster after 40 and 50s
2012-10-23T19:34:10 < zyp> 17:57:49 < Tectu> no idea why
2012-10-23T19:35:18 < zyp> because the only pixels that take the full 512 iterations are the black pixels (or whatever the color you have in the middle of the fractal is)
2012-10-23T19:35:30 < zyp> so the more black pixels you see, the longer time it takes to render
2012-10-23T19:36:18 < zyp> t is actually the value that decides color, so the color of a pixel is directly related to how long time it takes to render
2012-10-23T19:36:55 < zyp> so rendering time depends on which portion of the fractal you are rendering
2012-10-23T19:37:27 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-23T19:37:51 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-23T19:37:59 -!- ohama [ohama@92.240.248.100] has joined ##stm32
2012-10-23T19:39:46 < zyp> http://upload.wikimedia.org/wikipedia/commons/f/f4/Animation_of_the_growth_of_the_Mandelbrot_set_as_you_iterate_towards_infinity.gif <- I suggest you have a look at this
2012-10-23T19:40:33 < zyp> if you set the limit for t to 1, you will get just the circle, and so on
2012-10-23T19:57:45 < zyp> hmm, building the launchpad toolchain took around 25 minutes here, not too bad
2012-10-23T20:01:11 < emeb> zyp: you build it on your own to turn on features not found in the binary distribution?
2012-10-23T20:01:24 < emeb> (like gdb python scripting?)
2012-10-23T20:01:35 < zyp> there is no binary distribution for OS X
2012-10-23T20:01:42 < zyp> or did I miss it?
2012-10-23T20:01:52 < emeb> No - thought you were on linux.
2012-10-23T20:02:51 < emeb> looks like the Yagarto site has a binary of 4.7.x for Mac OS X
2012-10-23T20:02:52 < zyp> I didn't do any changes, and it appears to have python support by default here
2012-10-23T20:02:59 < zyp> yes, that's what I've been using
2012-10-23T20:03:09 < zyp> but it's not building BMP correctly
2012-10-23T20:03:20 < zyp> so I'm going to test with launchpad toolchain now
2012-10-23T20:03:36 < emeb> Interesting - I'm using the binary for Linux from launchpad and its gdb doesn't have python turned on.
2012-10-23T20:04:16 < emeb> I was able to build the BMPM w/ launchpad.
2012-10-23T20:04:22 < zyp> ok, works
2012-10-23T20:04:26 < emeb> nice
2012-10-23T20:04:46 < zyp> so there is some incompatibility between yagarto and either BMP firmware or libopencm3
2012-10-23T20:04:57 < emeb> very strange.
2012-10-23T20:05:16 < karlp> zyp: could you not even build it? I thought you could built it but it didn't work after flashing it?
2012-10-23T20:05:44 < zyp> it built just fine, but it didn't work
2012-10-23T20:06:03 < zyp> anyway, enough time wasted on that, now I'm going to add F3 support to BMP
2012-10-23T20:09:06 < emeb> oh, good!
2012-10-23T20:09:24 < emeb> I was wondering if that was in there already or not.
2012-10-23T20:10:18 < zyp> it pretty much is, I believe I just have to detect it and treat it like F1
2012-10-23T20:10:42 < emeb> so  no new loader stub needed hopefully.
2012-10-23T20:10:56 < zyp> exactly
2012-10-23T20:13:55 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-23T20:14:07  * emeb reads the liveblog of the Apple event. More new goodies I won't be buying...
2012-10-23T20:15:02 <+Steffanx> You should be using your mac for that
2012-10-23T20:15:28 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-23T20:15:32 < emeb> heh - I think there's something iconoclastic about doing apple stuff on a linux box.
2012-10-23T20:15:54 < emeb> since apple itself is not longer particularly iconoclastic
2012-10-23T20:16:02 < emeb> -t
2012-10-23T20:16:12 < zyp> (gdb) mon swdp_scan
2012-10-23T20:16:12 < zyp> Target voltage: OK
2012-10-23T20:16:12 < zyp> Available Targets:
2012-10-23T20:16:13 < zyp> No. Att Driver
2012-10-23T20:16:13 < zyp>  1      STM32F3xx
2012-10-23T20:16:15 < zyp> looks fine so far
2012-10-23T20:16:30 < zyp> just need to test flashing something
2012-10-23T20:16:46 < karlp> well, that'ðsa ll laurence had to do for stlink, can'timagine it would be more complicated on bmp :)
2012-10-23T20:16:56 < zyp> ;)
2012-10-23T20:17:18 < zyp> rather way easier, since the code has a sane structure
2012-10-23T20:17:43 < karlp> less authors and someone reviewing before merge tends to do that.
2012-10-23T20:17:52 < emeb> zing!
2012-10-23T20:18:41 < zyp> wtf
2012-10-23T20:18:42 < zyp> arm-none-eabi-gcc: error: unrecognized command line option '-fpromote-loop-indices'
2012-10-23T20:18:48 < zyp> what has Laurenceb been up to?
2012-10-23T20:20:14 < jpa-> funroll-loops wasn't good enough for him
2012-10-23T20:20:36 < emeb> this stuff he did in his branch of stlink?
2012-10-23T20:21:39 < zyp> no, f3 demo project
2012-10-23T20:21:55 < emeb> aha.
2012-10-23T20:22:28 < zyp> well, it's moving my mouse pointer, so it appears it is working
2012-10-23T20:22:37 < emeb> \o/
2012-10-23T20:24:08 < emeb> $1.7k for the new 13" MBP
2012-10-23T20:24:21 < zyp> retina?
2012-10-23T20:24:25 < emeb> yup
2012-10-23T20:24:37 < zyp> what's your price for the non-retina one?
2012-10-23T20:25:18 < emeb> I'd have to go to the apple site... just a sec.
2012-10-23T20:25:57 < emeb> $1.2k
2012-10-23T20:26:30 < zyp> so 40% or so more expensive
2012-10-23T20:27:03 < emeb> how much is that extra resolution worth to you?
2012-10-23T20:27:17 <+Steffanx> nothng
2012-10-23T20:27:32 < emeb> can't say it's a big deal to me either.
2012-10-23T20:27:44 < zyp> I'd probably choose a retina MBP at 14k NOK over a normal MBP at 10k nok
2012-10-23T20:28:15 <+Steffanx> Who chooses a macbook pro? :P
2012-10-23T20:28:41 < zyp> but considering my current MBP is just over a year old, I'm not looking to replace it anytime soon
2012-10-23T20:29:25 < gnomad> I've been using the same macbook pro for nearly 7 years now...
2012-10-23T20:30:21 < gnomad> only time speed is a problem is if I'm compiling something like gcc...
2012-10-23T20:30:56 < zyp> for laptops, performance is not that important, IMO
2012-10-23T20:31:07 < zyp> battery life is more interesting
2012-10-23T20:31:15 < zyp> how is your battery after seven years? :p
2012-10-23T20:31:17 < emeb> new iMac has no optical drive. Audience applauds.
2012-10-23T20:31:32 < zyp> :)
2012-10-23T20:31:49 < zyp> my MBP doesn't either
2012-10-23T20:31:51 <+Steffanx> Apple crowd is weird
2012-10-23T20:31:53 < zyp> I replaced it with an SSD
2012-10-23T20:32:07 <+Steffanx> ( samer hre )
2012-10-23T20:32:40 < zyp> I was planning to put it into my new workstation since the case have a slot that fits it, but I haven't had any use for it yet
2012-10-23T20:33:00 < zyp> I mean, who uses optical drives nowadays?
2012-10-23T20:33:15  * emeb watches dvds on his old MB sometimes.
2012-10-23T20:34:45 < karlp> I need an optical drive long enough to rip the cds I buy from bands at small shows
2012-10-23T20:34:46 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-23T20:34:59 < karlp> if they start selling me a small flash drive instead, that need will go away too
2012-10-23T20:35:07 < zyp> :p
2012-10-23T20:36:21 < zyp> I put my entire CD collection in my car a few years ago
2012-10-23T20:36:26 < zyp> I mean, all eight discs or so
2012-10-23T20:36:39 < emeb> heh
2012-10-23T20:36:48 < zyp> of course, I never listen to them
2012-10-23T20:36:59 < zyp> my car player reads SD cards, so that's all I ever listen to
2012-10-23T20:37:01 < zyp> in the car
2012-10-23T20:37:33 < karlp> you have music on sd cards? where did you get that?
2012-10-23T20:37:47 < zyp> I got an SD card and put music on it.
2012-10-23T20:41:30 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-212.pptp.stw-bonn.de] has joined ##stm32
2012-10-23T20:44:33 <+Steffanx> pirated zyp ? :P
2012-10-23T20:44:51 < zyp> the music?
2012-10-23T20:45:06 < zyp> probably
2012-10-23T20:45:21 < emeb> arrrr!
2012-10-23T20:46:34 < Erlkoenig> Hi, the ARMv7 Architecture Reference Manual states that upon startup, the core initializes the "VTOR" register to some implementation specific value, and then executes the code found at the adress *(VTOR+4) (which thereby is the reset handler).
2012-10-23T20:46:35 < emeb> Average 1.267549 iPads sold *per second* for past 30 months.
2012-10-23T20:46:35 < Erlkoenig> However the STM32F4 Reference Manual doesn't say anything about VTOR's initial value and where the interrupt vector should be placed (and how it could be moved). Where is the documentation about that?
2012-10-23T20:47:00 < zyp> huh?
2012-10-23T20:47:15 < zyp> doesn't VTOR always have an initial value of 0?
2012-10-23T20:48:06 < zyp> yep
2012-10-23T20:48:19 < Erlkoenig> from the Arch. Manual: "On reset, the processor initializes the vector table base address to an IMPLEMENTATION DEFINED address. Software can find the current location of the table, or relocate the table, using the VTOR, ..."
2012-10-23T20:48:25 < zyp> PM0214, rev 3, page 39:
2012-10-23T20:48:27 < zyp> On system reset, the vector table is fixed at address 0x00000000. Privileged software can write to the VTOR to relocate the vector table start address to a different memory location, in the range 0x00000080 to 0x3FFFFF80
2012-10-23T20:48:31 < Erlkoenig> Page 634
2012-10-23T20:49:29 < zyp> PM0214 is the documentation for ST's cortex-m4 implementation
2012-10-23T20:50:17 < Erlkoenig> aaaah thanks that's what i wanted to know... thanks :)
2012-10-23T20:50:23 < zyp> no problem
2012-10-23T20:51:25 < zyp> having a default value of 0 means it'll either point to start of flash, start of ram or start of rom-bootloader depending on value of boot0/boot1 pins
2012-10-23T20:51:50 < zyp> since the lowest area of the adress space is remapped to either of those
2012-10-23T20:52:05 < Erlkoenig> ah okay that's what i guessed ;-)
2012-10-23T20:52:33 < Erlkoenig> so when normally booting via Flash, i can put an "alternative" vector table somewhere into the SRAM (0x2000 0000) and point VTOR there?
2012-10-23T20:52:52 < zyp> yes
2012-10-23T20:53:08 < zyp> if you want to be able to change the vectors at runtime, you could
2012-10-23T20:54:20 < Erlkoenig> yup, exactly...
2012-10-23T20:55:36 < emeb> heh - gen4 iPad & iPad mini announced. Fanbois lament there old iPads are all crappy now.
2012-10-23T20:55:46 < emeb> s/there/their/
2012-10-23T20:56:09 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-23T20:56:35 < Erlkoenig> hu? did they think that apple would never release a never version?
2012-10-23T20:57:02 < zyp> what changed? I assume they use the same A15-ish custom-arm-cpu as the new iphone got, but anything else?
2012-10-23T20:57:06 < emeb> ya - it always seems like a surprise that the treadmill keeps rolling.
2012-10-23T20:57:29 < zyp> probably same new usb connector too?
2012-10-23T20:57:49 < emeb> gen4 iPad uses the A6x
2012-10-23T20:58:02 < emeb> same as iPhone 5
2012-10-23T20:58:04 < Erlkoenig> ... the bit in the PRIMASK register acts like the "I" bit in the SREG of the AVR - global interrupt enable/disable - right?
2012-10-23T20:58:27 < Erlkoenig> (except for the faults)
2012-10-23T20:58:30 < zyp> yep
2012-10-23T21:01:16 < Erlkoenig> ah that programming manual is somewhat more readable than the architecture reference... using the "BASEPRI" register i can give the main() "task" a priority, preventing the execution of lower priority interrupts, right?
2012-10-23T21:02:16 < zyp> you pretty much disable all lower priority interrupts, yeah
2012-10-23T21:02:34 < zyp> hmm, that's an interesting way of putting it
2012-10-23T21:03:17 < zyp> when I think about it, I could give all my threads a priority and change the value of primask on context switch
2012-10-23T21:03:19 < Erlkoenig> ah even when a higher priority interrupt than BASEPRI occurs and exits, lower-priority ISR's wont ever be executed?
2012-10-23T21:03:45 < zyp> but I can't really imagine any case where an interrupt should have lower priority than a thread
2012-10-23T21:04:10 < zyp> s/primask/basepri/
2012-10-23T21:05:02 < Erlkoenig> maybe the main thread does some time critical stuff on it's own... well i was just wondering ;)
2012-10-23T21:07:33 -!- BusError_ is now known as buserror
2012-10-23T21:39:30 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Quit: my exit]
2012-10-23T21:53:21 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-23T21:56:13 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-23T21:56:44 < TitanMKD> hi
2012-10-23T21:56:54 < emeb> hola
2012-10-23T21:58:26 <+Steffanx> Salute
2012-10-23T22:09:46 -!- daku is now known as DaKu
2012-10-23T22:16:06 < Tectu> salut
2012-10-23T22:20:21 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-23T22:21:17 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-23T22:21:22 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-23T22:48:38 -!- ntfreak_ is now known as ntfreak
2012-10-23T23:05:40 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-23T23:05:41 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Disconnected by services]
2012-10-23T23:06:19 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Client Quit]
2012-10-23T23:06:28 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-23T23:21:19 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Read error: Connection reset by peer]
2012-10-23T23:25:20 -!- Vutral [ss@vutral.net] has joined ##stm32
2012-10-23T23:25:24 -!- Vutral [ss@vutral.net] has quit [Changing host]
2012-10-23T23:25:24 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-23T23:29:44 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-23T23:32:04 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-23T23:38:02 -!- barthess [~barthess@178.154.58.185] has quit [Quit: Leaving.]
2012-10-23T23:51:40 < badger> anyone know of some stm32f4 libraries for kicad?
2012-10-23T23:57:33 < Laurenceb_> Randomskk on github/ #highaltitude would
2012-10-23T23:58:26 < Laurenceb_> ill see if hes around
2012-10-23T23:58:54 < Laurenceb_> doesnt look like he is :-/
2012-10-23T23:59:18  * Laurenceb_ is trying to work out a simple way to get magno cal working onboard the f3discovery
2012-10-23T23:59:46 < zyp> Laurenceb_, I tested your code, it's so awesome, it's moving my mouse pointer
--- Day changed Wed Oct 24 2012
2012-10-24T00:00:39 < Laurenceb_> https://github.com/adamgreig/wombat/tree/master/pcb/libs
2012-10-24T00:00:46 < Laurenceb_> lol :P
2012-10-24T00:00:58 < Laurenceb_> did the mouse point move for you
2012-10-24T00:01:24 < zyp> both left and right
2012-10-24T00:01:27 < Laurenceb_> <Randomskk> schematic/pcb footprint
2012-10-24T00:01:27 < Laurenceb_> <Randomskk> https://github.com/adamgreig/wombat/blob/master/pcb/libs/stm32f405rgt6.lib https://github.com/adamgreig/wombat/blob/master/pcb/libs/stm32f405rgt6.dcm
2012-10-24T00:01:27 < Laurenceb_> <Randomskk> only any good if they're using the same package though
2012-10-24T00:01:29 < zyp> and a bit up and down
2012-10-24T00:01:39 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-24T00:01:42 < zyp> but I'm not sure what the logic behind it is supposed to b
2012-10-24T00:01:43 < zyp> e
2012-10-24T00:02:41 < Laurenceb_> pitch == up down
2012-10-24T00:02:50 < Laurenceb_> heading/yaw == left right
2012-10-24T00:02:56 < badger> Laurenceb_: thanks for that!
2012-10-24T00:03:56 < Laurenceb_> i think <10degrees hysteresis on heading will help
2012-10-24T00:04:06 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-24T00:04:10 < Laurenceb_> try 5 like on the pitch
2012-10-24T00:04:27 < zyp> I don't have it powered up right now
2012-10-24T00:04:42 < Laurenceb_> also it needs calibrated magno
2012-10-24T00:04:51 < Laurenceb_> mine is a bit weird and nonlinear
2012-10-24T00:04:54 < zyp> I just needed something to flash to test that F3 support in BMP was working so I grabbed it :p
2012-10-24T00:05:27 < Laurenceb_> so you can flash the f103 on the f3discovery with bmp and it works now?
2012-10-24T00:05:46 < Erlkoenig> /home/erlkoenig/programs/gcc-arm-embedded/gcc-arm-none-eabi-4_6-2012q1/bin/../lib/gcc/arm-none-eabi/4.6.2/../../../../arm-none-eabi/lib/armv7e-m/fpu/libc.a(lib_a-abort.o): In function `abort': abort.c:(.text.abort+0xa): undefined reference to `_exit'   ...   why the heck is the libc referring to _exit :S
2012-10-24T00:05:47 < zyp> I haven't done that
2012-10-24T00:06:16 < zyp> I just disconnected onboard stlink and hooked my BMP to PA13/PA14
2012-10-24T00:06:23 < Laurenceb_> ah
2012-10-24T00:06:43 < zyp> but I'll probably flash all my discovery boards with BMP firmware later
2012-10-24T00:06:59 < zyp> I'm not sure it has MCO support, so I might need to add that too
2012-10-24T00:07:12 < Laurenceb_> yeah, 8mhz
2012-10-24T00:07:21 < Laurenceb_> is needed
2012-10-24T00:07:48 < Laurenceb_> i need to test usb dfu
2012-10-24T00:07:52 < Laurenceb_> or someone does...
2012-10-24T00:08:46 < zyp> test how?
2012-10-24T00:09:23 < Laurenceb_> see if the BOOT0 bootloader supports DFU
2012-10-24T00:10:05 < zyp> is it supposed to?
2012-10-24T00:10:16 < Laurenceb_> datasheet seems to say so
2012-10-24T00:10:23 < zyp> oh
2012-10-24T00:10:24 < Laurenceb_> bu no proper details
2012-10-24T00:10:34 < Laurenceb_> no app note etc
2012-10-24T00:10:41 < zyp> ok, testing
2012-10-24T00:12:56 < zyp> yep
2012-10-24T00:13:25 < zyp> http://bin.jvnv.net/f/hhzfA.png
2012-10-24T00:14:16 < Laurenceb_> sweet
2012-10-24T00:14:32 < Laurenceb_> looks good for a autopilot
2012-10-24T00:15:55 < zyp> works
2012-10-24T00:15:56 < zyp> http://paste.jvnv.net/view/SB24j
2012-10-24T00:16:42 < Laurenceb_> nice
2012-10-24T00:17:55 < zyp> conveniently enough boot0 is right next to vdd, so just jumper those before power up and it will go into bootloader mode
2012-10-24T00:23:48 < zyp> wonder if L1 does that
2012-10-24T00:33:49 < Laurenceb_> L1 discovery has no usb does it?
2012-10-24T00:34:51 < zyp> no connector but DM/DP pins are free gpio and should be able to be hooked up easily
2012-10-24T00:35:42 < Laurenceb_> according to F3 datasheet the usb is impedance matched
2012-10-24T00:35:52 < Laurenceb_> yet theres 22ohm resistors on the discovery
2012-10-24T00:38:09 < zyp> hmm, true
2012-10-24T00:38:11 < Laurenceb_> maybe there was a misunderstanding at ST
2012-10-24T00:38:22 < Laurenceb_> they cant write tilt compensated compass...
2012-10-24T00:38:58 < Laurenceb_> i still need to characterise the opamps
2012-10-24T00:39:28 < Laurenceb_> was planning on a tissue spectrometer with usb interface
2012-10-24T00:39:38 < Laurenceb_> on 14mm diameter pcb
2012-10-24T00:41:45 < Erlkoenig> do you guys know what's prettier than a STM32F4 Discovery?
2012-10-24T00:42:09 < zyp> http://bin.jvnv.net/f/mu3oa.JPG <- the whole family?
2012-10-24T00:42:10 < Laurenceb_> your dick?
2012-10-24T00:42:29 < Erlkoenig> meh :D
2012-10-24T00:42:37 < Erlkoenig> Two F4 Discoveries :3
2012-10-24T00:42:40 < Laurenceb_> no F1, F2, W
2012-10-24T00:42:52 < Erlkoenig> gotta catch them all? :D
2012-10-24T00:42:55 < zyp> is there any F2 discovery?
2012-10-24T00:43:10 < Laurenceb_> dunno
2012-10-24T00:43:28 < zyp> Erlkoenig, I'm writing a lib supporting several of them, it's nice to be able to test stuff :p
2012-10-24T00:43:42 < Erlkoenig> thats of course a reason :D
2012-10-24T00:43:43 < Laurenceb_> nope
2012-10-24T00:43:44 < Erlkoenig> what does the lib do?
2012-10-24T00:43:45 < Laurenceb_> http://www.st.com/internet/com/MULTIMEDIA_RESOURCES/IMAGE/BOARD_PHOTO/board_stm32vl_discovery.jpg
2012-10-24T00:43:48 < Laurenceb_> thats that
2012-10-24T00:44:33 < zyp> Erlkoenig, it let's me type out all my insane ideas for how stuff could be done differently from everybody else :p
2012-10-24T00:44:53 < Laurenceb_> heh
2012-10-24T00:44:57 < Erlkoenig> that's great :D
2012-10-24T00:45:16 < Erlkoenig> i need to write my own "rtos" so just my code will be incompatible with other oses
2012-10-24T00:45:44 < zyp> I have multithreading as well
2012-10-24T00:45:57 < zyp> http://cgit.jvnv.net/laks/tree/
2012-10-24T00:46:03 < zyp> in case you wanted to look
2012-10-24T00:46:05 < Laurenceb_> but i have a bbc model b
2012-10-24T00:46:11 < Laurenceb_> used to design the arm1
2012-10-24T00:46:41 < Laurenceb_> and an acorn with arm2
2012-10-24T00:47:21 < Laurenceb_> more often than you get laid
2012-10-24T00:49:22 < zyp> no thanks
2012-10-24T00:49:42 < Laurenceb_> hehe
2012-10-24T00:50:02 < Erlkoenig> Hm, the ARMv7 Architecture Reference, on page 652, chapter B1.5.8., says that to return from an ISR, e.g. by using "bx lr", the LR needs some weird 0xFXX... value. Is it correct that upon entry to the ISR the core automatically puts the correct weird value into LR (the pseudocode at the end of page 644 looks like that), so i don't need to worry about all that and can just have a "bx lr"?
2012-10-24T00:50:14 < zyp> Erlkoenig, yep
2012-10-24T00:50:44 < Erlkoenig> okay... ARM is weird :D
2012-10-24T00:51:00 < zyp> armv7-m interrupt handling is fully eabi compliant so any normal eabi function can be used as an interrupt handler
2012-10-24T00:51:18 < Erlkoenig> oh nice
2012-10-24T00:51:45 < Laurenceb_> wtf is eabi?
2012-10-24T00:52:17 < zyp> the standard abi for modern arm stuff
2012-10-24T00:52:24 < Laurenceb_> ah
2012-10-24T00:52:30 < emeb> Application Binary Interface?
2012-10-24T00:52:41 < Erlkoenig> could an ISR modify the SP to point to some completely different stack so when returning, the core is basically performing a context(thread) switch? (one will probably still need some PUSHs and POPs)
2012-10-24T00:52:42 < zyp> you never noticed that your toolchain is named arm-none-eabi- ? :p
2012-10-24T00:52:46 < zyp> emeb, yes
2012-10-24T00:52:55 < zyp> the abi defines the interface between functions
2012-10-24T00:52:58 < emeb> no idea what the E is for
2012-10-24T00:53:19 < zyp> Erlkoenig, that's exactly how context switches work
2012-10-24T00:53:29 < emeb> Extended/Enhanced/Euphonious/Erlkoenig/etc
2012-10-24T00:53:29 < Erlkoenig> yeah? oh nice :D
2012-10-24T00:53:39 < zyp> http://cgit.jvnv.net/laks/tree/interrupt/fault.cpp <- here's how I'm currently doing it
2012-10-24T00:54:03 < Erlkoenig> cool i'll pirate that code :3
2012-10-24T00:54:23 < Laurenceb_> i think chibios can do the same thing?
2012-10-24T00:54:27 < emeb> it's patented and copyright. See you in court, in East Texas.
2012-10-24T00:54:43 < zyp> it's due to be replaced when I get around to writing the improvements
2012-10-24T00:54:48 < zyp> but it works as is
2012-10-24T00:55:05 < zyp> I've just learned a bit since I wrote it ;)
2012-10-24T00:55:08 < emeb> haha - then Erlkoenig will be old-n-busted while zyp is new hotness.
2012-10-24T00:55:11 < Erlkoenig> what if a thread voluntarily wanted to switch to some other context? it would call a function that nearly does the same things as your code, right?
2012-10-24T00:55:30 < zyp> that's currently all I'm doing
2012-10-24T00:55:38 < zyp> I'm not doing preemptive multitasking
2012-10-24T00:56:10 < Laurenceb_> its very useful for handing hardware stuff
2012-10-24T00:56:16 < Erlkoenig> yeah assuming i wanted to write an OS that does preemptive m.t.
2012-10-24T00:56:28 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- so every thread is calling Thread::yield() when they are waiting for something and want to let other threads run
2012-10-24T00:56:46 < Erlkoenig> aaah svc
2012-10-24T00:56:50 < zyp> http://cgit.jvnv.net/laks/tree/os/mutex.h <- see Mutex::lock() for an example
2012-10-24T00:56:56 < Erlkoenig> and the svc ISR does the context switch?
2012-10-24T00:57:04 < zyp> currently, yes
2012-10-24T00:57:27 < zyp> it's better to trigger pendsv and have pendsv do the actual switch, so that's what I'm going to rewrite to
2012-10-24T00:57:32 < Erlkoenig> could yield() also be replaced by switch_context() resulting in a "direct" context switch?
2012-10-24T00:58:05 < zyp> no, you are not allowed to return to exc_return from thread mode, only handler mode
2012-10-24T00:58:43 < zyp> also, my threading code is missing floating point support, I haven't had any need for it yet
2012-10-24T00:59:04 < Erlkoenig> perhaps not returning... but just doing some sort of jump into the new thread?
2012-10-24T00:59:19 < Erlkoenig> what's the advantage of PendSV?
2012-10-24T01:01:17 < zyp> you can trigger a context switch from any ISR by triggering pendsv, and by setting pendsv lowest priority it will run after all other pending ISRs are processed and do the actual context switch
2012-10-24T01:02:00 < Erlkoenig> this is just for improving the latency for possibly timing critical interrupts?
2012-10-24T01:02:17 < zyp> no, it's so you can handle it in a standardized manner
2012-10-24T01:04:04 < Erlkoenig> hmm
2012-10-24T01:04:40 < Erlkoenig> funny in how many ways a Cortex-M4F can cause faults... while e.g. an AVR doesn't have any support for handling faults :D
2012-10-24T01:05:24 < zyp> well, it's not without reason that we consider AVR being outdated
2012-10-24T01:05:42 < Erlkoenig> hehe okay :D
2012-10-24T01:05:48 < Erlkoenig> well it's nice as a learning platform i think
2012-10-24T01:06:39 < zyp> depends on what you are thinking of
2012-10-24T01:07:05 -!- DaKu is now known as daku
2012-10-24T01:07:10 < zyp> it's not bad for introduction to asm, the instruction set is nice and simple and easy to learn
2012-10-24T01:07:30 < Erlkoenig> well the datasheets are *much* shorter and easier to understand... it's easier to get some LED-blinking running and get a nice result
2012-10-24T01:08:43 < zyp> they are shorter because there is less peripherals to document
2012-10-24T01:09:55 < Laurenceb_> and the peripherals are lame
2012-10-24T01:09:57 < zyp> even though stm32 has more chapters in the reference manual, it doesn't mean that anyone have to read them all :p
2012-10-24T01:10:35 < zyp> limited is probably a more objective wording
2012-10-24T01:11:24 < zyp> AVR I2C peripheral doesn't implement a state machine in hardware
2012-10-24T01:11:29 < Erlkoenig> yeah but to just get an 1 Hz LED Blinker using a Timer, you still have to read much more pages that aren't easy to read... because the STm32 Timers are so powerful, but also complicated
2012-10-24T01:11:44 < zyp> IIRC it has three registers; control, status and data
2012-10-24T01:11:57 < zyp> and it's very simple to get working reliably
2012-10-24T01:12:01 < Laurenceb_> yeah something like that
2012-10-24T01:12:10 < Laurenceb_> havent made anything with avr for a couple of years
2012-10-24T01:12:35 < zyp> I used I2C for sensors on my uni project, never had any problems with it
2012-10-24T01:13:48 < Erlkoenig> i wrote a completely asynchroneous (all code in interrupts) code for AVR which uses RFM12 modules to do reliable communication (checksums, re-transmit) - unfortunately somewhere in the ~6000 ASM Lines there must be a tiny bug :D
2012-10-24T01:14:46 < ds2> you need RAID!!!
2012-10-24T01:14:53 < zyp> I also had an AVR assembly course at uni, had to write asm on paper at the exam :)
2012-10-24T01:15:08 < zyp> that's fun
2012-10-24T01:15:56 < Erlkoenig> wah... like multiplying matrices in exams (without calculator of course) :S
2012-10-24T01:16:10 < zyp> that was also the same day I was going on christmas vacation that year, so I had my car packed and ready outside
2012-10-24T01:16:29 < Erlkoenig> the AVR Instruction set is burned into my brain :D
2012-10-24T01:16:37 < zyp> I've never written so much in so short time on an exam ever before or since :p
2012-10-24T01:16:57 < zyp> asm is pretty verbose :p
2012-10-24T01:17:22 < zyp> so most of the time was spent for the physical writing
2012-10-24T01:18:16 < zyp> the tasks given was pretty much identical to some of the tasks we had done during the course, so it was almost just about remembering and writing down :p
2012-10-24T01:19:13 < Erlkoenig> today in an exercise lesson at uni i did the tasks (which where homework) in the first 30min or something, including some that nobody managed to do... then pointed out an error in the assignment to the instructor, he's going to as the prof... :D
2012-10-24T01:19:26 < Erlkoenig> now everybody's gonna think i'm a ubernerd
2012-10-24T01:19:38 < Laurenceb_> our prof was lurking in the exam room during my finals
2012-10-24T01:19:50 < Laurenceb_> peering over our shoulders
2012-10-24T01:19:53 < Laurenceb_> thats didnt help
2012-10-24T01:20:05 < zyp> don't they always do that?
2012-10-24T01:20:24 < Laurenceb_> dunno
2012-10-24T01:20:29 < Laurenceb_> you only do finals once
2012-10-24T01:20:31 < Erlkoenig> it's better when they stand in the front and smiling "ah this is all so easy why didn't i put harder questions"
2012-10-24T01:20:43 < Laurenceb_> there was a guy with a log periodic yagi and icom ic1000
2012-10-24T01:20:44 < zyp> then they smile and snicker a bit and walks on
2012-10-24T01:20:55 < Laurenceb_> looking for phones etc
2012-10-24T01:21:02 < Erlkoenig> wat oO
2012-10-24T01:21:04 < zyp> ha
2012-10-24T01:21:36 < Erlkoenig> all the nerds get best marks "ah i was not sure about question 4.b) but i think i got it" - and you just fail miserably :3
2012-10-24T01:21:50 < Laurenceb_> i got 70% in one of my exams
2012-10-24T01:22:06 < Erlkoenig> if, when leaving the room you know "it's absolutely impossible i have reached 50%", you're happy :3
2012-10-24T01:22:10 < Laurenceb_> by drawing a probability distribution as a rectange rather than elipse
2012-10-24T01:22:16 < Laurenceb_> arg spelling
2012-10-24T01:22:22 < Laurenceb_> epic fail
2012-10-24T01:22:34 < zyp> Laurenceb_, only once? as in once per course?
2012-10-24T01:22:50 < Laurenceb_> theres only one set of _final_  exams
2012-10-24T01:23:07 < Laurenceb_> but overall score is a crazy weighted thing of all the exams
2012-10-24T01:23:20 < zyp> I'm not sure how your education system works
2012-10-24T01:23:28 < Laurenceb_> i got a first and came 31st out of 140 at oxford
2012-10-24T01:23:34 < Erlkoenig> the whole exam system ist just sh*t and extremele unfair -.-
2012-10-24T01:23:34 < Laurenceb_> so im not complaining XD
2012-10-24T01:23:36 < zyp> here we have a final exam in each course
2012-10-24T01:23:51 < Laurenceb_> i was aiming to get better than 35th
2012-10-24T01:23:57 < Laurenceb_> top 35 or so get 1st class
2012-10-24T01:24:03 < zyp> and courses last for half a year, so we have two exam periods per year
2012-10-24T01:24:11 < Erlkoenig> ... dito
2012-10-24T01:24:12 < Laurenceb_> so i optimised the amount of revision perfectly
2012-10-24T01:24:51 < Laurenceb_> but nearly messed it up by drawing a rectangular probability density field
2012-10-24T01:25:09 < Laurenceb_> if id got a 2nd class due to that itd be kind of embarrassing
2012-10-24T01:25:14 < Laurenceb_> luckly that didnt happen
2012-10-24T01:25:34 < zyp> heh
2012-10-24T01:25:34 < Erlkoenig> snob :D
2012-10-24T01:25:48 < zyp> we only get graded A-E per exam (or F for fail)
2012-10-24T01:26:10 < Erlkoenig> we get 1.0 (best) - 4.0 (worst), and 5.0 (fail)
2012-10-24T01:26:13 < Laurenceb_> lost 25% of my marks for that exam straight away
2012-10-24T01:26:22 < Erlkoenig> 4.0 is my favourite :3
2012-10-24T01:26:34 < zyp> I never got an E
2012-10-24T01:26:42 < Laurenceb_> and realised 10 seconds before the end of the exam, but it was too late :P
2012-10-24T01:27:08 < Erlkoenig> Laurenceb_: also nice: forget one "A" in one formula, so 2 assignments are completely wrong, fail the exam
2012-10-24T01:27:19 < Laurenceb_> yeah
2012-10-24T01:27:25 < zyp> I got a D in physics though, the prof was a dick :p
2012-10-24T01:27:45 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-24T01:27:52 < zyp> at first I was like «wtf?», and then I checked the statistics
2012-10-24T01:28:12 < zyp> 2/3rds of the students got D or worse
2012-10-24T01:28:14 < Erlkoenig> if less than 50% of the students failed the test, the prof is a pussy
2012-10-24T01:28:31 < Erlkoenig> so the tests are designed accordingly
2012-10-24T01:29:01 < zyp> I've only once used all the allotted time
2012-10-24T01:29:07 < Erlkoenig> lol oO
2012-10-24T01:29:15 < zyp> because I didn't manage to finish on time
2012-10-24T01:29:26 < zyp> got an A on that one
2012-10-24T01:29:45 < emeb> Glad I've forgotten all that stuff.
2012-10-24T01:29:51 < Erlkoenig> my tests where all made in a way so it's barely possible to look at all assignments
2012-10-24T01:29:53 < zyp> it was designed to take up all time and then some, so people not managing to do everything was expected
2012-10-24T01:30:23 < Erlkoenig> all tests are made this way...
2012-10-24T01:30:35 < zyp> I've usually used less than half the time allotted, I believe
2012-10-24T01:30:42 < Erlkoenig> wow
2012-10-24T01:30:51 < Erlkoenig> either you are a genius or your uni is easy :D
2012-10-24T01:31:03 < zyp> they introduced a new rule when I was a year in
2012-10-24T01:31:12 < zyp> nobody could leave before an hour had passed
2012-10-24T01:31:22 < zyp> I ran into that several times
2012-10-24T01:31:45 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-24T01:31:51 < zyp> allotted time was 3-5 hours depending on the exam
2012-10-24T01:33:01 < zyp> I'm not sure I could call myself a genious, I'm just lazy
2012-10-24T01:33:38 < zyp> I did everything I could and didn't bother spending time on stuff I weren't going to get anyway, so I just handed in and left
2012-10-24T01:34:07 < Erlkoenig> hmm
2012-10-24T01:34:32 < Erlkoenig> after failing maths, i'm now in CS, maybe i'll do the exams better here
2012-10-24T01:35:00 < zyp> I was in EE
2012-10-24T01:35:24 < zyp> it of course helps to be interested in whatever you are studying, because then you soak up the knowledge without even trying :p
2012-10-24T01:36:03 < zyp> I did some CS courses as well
2012-10-24T01:36:19 < Erlkoenig> well i am still very much interested in maths... but it's just damn difficult
2012-10-24T01:36:41 < Erlkoenig> maths for the CS people is kindergarten in comparison
2012-10-24T01:37:11 < zyp> after an exam in C++, the prof told me I was the only guy that got a specific question correct, so he asked if I tought it was an unfair question
2012-10-24T01:37:48 < Erlkoenig> wut :D
2012-10-24T01:38:02 < Erlkoenig> in maths i was rather the only one who got it wrong :3
2012-10-24T01:40:52 < zyp> http://paste.jvnv.net/view/iM62x <- it went something like this
2012-10-24T01:41:24 < zyp> multiple choice, four choices, I forgot how the fourth went
2012-10-24T01:42:24 < zyp> and I forgot the exact code too, but it was something that would cause a print of a value from an uninitialized stack variable somewhere
2012-10-24T01:43:38 < zyp> everyone must have noticed that it was something wrong with the code, because I was the only one that answered «The program will compile and run.»
2012-10-24T01:44:27 < zyp> because that's what it will do, even though it will print some random garbage :)
2012-10-24T01:45:08 < Erlkoenig> well uhm i might have been able to answer that :D
2012-10-24T01:46:43 < zyp> I was probably the only one at that course that had studied the relationship between C/C++ source and the assembly it will generate
2012-10-24T01:47:26 < zyp> knowing how to write assembly is honestly not so useful
2012-10-24T01:47:33 < Erlkoenig> the best C "error" i saw was: i told some beginner "you can assume all variables have random values by default, so you *have* to initialize them" ... the result was: "i wanted to get some random numbers by not initializing this variable, but they're always zero" :D and the second best: double* ptr; *ptr = 27; "do i really NEED to initialize ptr? on my home computer it worked! ah, it's better to initialize them?"
2012-10-24T01:47:54 < zyp> knowing what assembly you can expect the compiler to produce from a given source snippet however is very useful to know
2012-10-24T01:48:09 < Erlkoenig> well perhaps not the exact assembly code but what it does ;)
2012-10-24T01:48:23 < zyp> random is a bad word, undefined is better ;)
2012-10-24T01:48:36 < emeb> also, having compiled a lot of stuff, you'd know that it just throws a warning for an uninitialized var.
2012-10-24T01:49:08 < emeb> (using an)
2012-10-24T01:49:14 < Erlkoenig> zyp: i learned it that day :D or: "random numbers with an unknown distribution" ;-)
2012-10-24T01:49:27 < zyp> emeb, I think it jumped through some more hoops
2012-10-24T01:49:54 < zyp> returning an undefined variable from a function that did nothing more than declare it and return it, or something like that
2012-10-24T01:50:17 < Erlkoenig> i once said "implementation specific detail"... that phrase followed me a few terms :D
2012-10-24T01:50:19 < emeb> I suppose those who said it wouldn't compile could argue that they had -Werr switch set.
2012-10-24T01:50:55 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-24T01:50:56 < zyp> because I remember I later demonstrated for some of the other guys that it could be affected by what was called before that function was called
2012-10-24T01:51:34 < zyp> how setting that stack slot in the previous function let us choose that undefined value
2012-10-24T01:51:51 < zyp> and how that assumption went out of the window when we turned on optimizations :p
2012-10-24T01:52:05 < emeb> and set them on the long road to hacking code vulns.
2012-10-24T01:53:50 < emeb> I remember a midterm in my first college programming course consisted of lines of bad code - the student had to write word-for-word the error that the compiler would throw.
2012-10-24T01:54:02 < zyp> a friend of mine were one of those guys, he's now working night shift at a network security monitoring center while doing his master ;)
2012-10-24T01:55:06 < Erlkoenig> hmm in a few weeks, we will get our first Hello World, in Java
2012-10-24T01:55:11 < zyp> :p
2012-10-24T01:55:59 < Erlkoenig> after one term, only one-page-long programs, in CS...
2012-10-24T01:56:37 < ds2> as a bonus task, do the same Hello World program in bytecode
2012-10-24T01:56:58 < Erlkoenig> uuhm no :D
2012-10-24T01:57:10 < Erlkoenig> the student's lanparty got cancelled :(
2012-10-24T01:57:10 < zyp> we had to write 6502-bytecode in high school
2012-10-24T01:57:16 < ds2> you did assembly before....
2012-10-24T01:57:49 < gnomad> what year was this?
2012-10-24T01:58:07 < zyp> we had a 6502 trainer board with a rom bootloader that let you key in the bytecode on a hex keypad into ram, and then execute it
2012-10-24T01:58:22 < zyp> so we wrote assembly on paper, and then assembled it for hand, before we keyed it in
2012-10-24T01:58:24 < gnomad> I did 6502 asm on my apple 2 when I was in high school... but that was back in the mid 80s.
2012-10-24T01:58:49 < Erlkoenig> we got an assignment to reverse-engineer the communication between two ATmega32-boards... using nice expensive DSO's :-> ... and thursday we have to do a small presentation about "what is a mikrocontroller"
2012-10-24T01:58:56 < gnomad> learning asm on your own before the internet was a bit of a challenge...
2012-10-24T01:58:59 < zyp> let me think, it was second year of vocational education here, electronics class
2012-10-24T01:59:17 < zyp> so it must have been 2005-2006
2012-10-24T01:59:25 < Erlkoenig> before the internet? when was that? :D
2012-10-24T01:59:40 < ds2> zyp: an actual 6502?
2012-10-24T01:59:46 < ds2> not the HC11's HC16's, etc?
2012-10-24T01:59:57 < zyp> yes, it was some old stuff that they had had for years
2012-10-24T02:00:12 < ds2> 6502's would be quite old
2012-10-24T02:00:13 < emeb> gnomad is almost as old as me. :P
2012-10-24T02:00:20 < ds2> at least with HC11's, it is a SoC
2012-10-24T02:00:26 < zyp> yeah, the documentation was pretty old
2012-10-24T02:00:27 < gnomad> this is the book I learned from:
2012-10-24T02:00:28 < gnomad> http://www.ebay.com/itm/Excellent-6502-Assembly-Language-Programming-Book-1979-/221045232997?pt=US_Vintage_Computers_Mainframes&hash=item3377529565
2012-10-24T02:00:39 < gnomad> i *think* i still have that in a box somewhere
2012-10-24T02:00:49 < Erlkoenig> what is the difference of SoC and mikrocontroller?
2012-10-24T02:01:17 < dongs> so whats blogging
2012-10-24T02:01:27 < emeb> war stories
2012-10-24T02:01:37 < zyp> ha
2012-10-24T02:01:38 < zyp> http://www.ee.siue.edu/~smuren/ece282/labs/manuals/emma2_manual.pdf
2012-10-24T02:01:41 < ds2> microcontroller can be bare
2012-10-24T02:01:41 < zyp> it was that stuff
2012-10-24T02:01:57 < zyp> user manual is dated 1986 :D
2012-10-24T02:02:06 < zyp> so it's older than me :p
2012-10-24T02:02:20 < zyp> Erlkoenig, mostly the scope
2012-10-24T02:02:27 < zyp> a microcontroller is a small SoC
2012-10-24T02:02:58 < emeb> interesting question. I'd say it varies with the task.
2012-10-24T02:03:08 < gnomad> in many ways, "microcontroller" and "system on a chip" are synonomous
2012-10-24T02:03:14 < zyp> I wouldn't call say OMAP4 for a microcontroller, but that's a SoC
2012-10-24T02:03:21 < emeb> sometimes, you can do the whole task on one MCU - then it might be an SoC.
2012-10-24T02:03:34 < ds2> I could imagine the OMAP4 being called a microcontroller much to TI's chargin
2012-10-24T02:03:38 < gnomad> depends on your definition of "system"
2012-10-24T02:03:42 < zyp> :)
2012-10-24T02:03:48 < emeb> Amazon's chagrin now, :)
2012-10-24T02:04:55 < zyp> dongs, you might be happy to hear that F3's rom bootloader supports DFU
2012-10-24T02:05:07 < ds2> that reminds me -
2012-10-24T02:05:10 < zyp> so now you have one less reason to keep around the serial converter chip
2012-10-24T02:05:24 < dongs> zyp, so i heard yea
2012-10-24T02:05:29 < dongs> have they updated the boot appnote yet?
2012-10-24T02:05:36 < ds2> if I wire up USB, a UART, and SWD on a new board - I should be reasonably sure I have at least 1 working way of downloading code to the STM32F103 right?
2012-10-24T02:05:52 < zyp> no idea, I just pulled boot0 high and flashed it with dfu-util
2012-10-24T02:05:58 < dongs> zyp, lol ok.
2012-10-24T02:05:58 < emeb> usually SWD is all you need.
2012-10-24T02:06:04 < ds2> oh and Boot1 and Boot0 is both exposed
2012-10-24T02:06:06 < zyp> since dfu-util already supports ST's dfuse-extension
2012-10-24T02:06:21 < zyp> emeb, not for user upgrades
2012-10-24T02:06:23 < Erlkoenig> /bin/brainctrl shutdown
2012-10-24T02:06:25 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-212.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-24T02:06:26 < ds2> emeb: this is for new board bring up with the BGA package so I want redundancy
2012-10-24T02:06:44 < emeb> ds2: redundancy is good. redundancy is good.
2012-10-24T02:06:50 < ds2> trying to make sure I break out enough things to not have to trash my board
2012-10-24T02:06:51 < zyp> ds2, F1 does not have DFU support in bootloader
2012-10-24T02:07:08 < emeb> ds2: an2606 is your friend.
2012-10-24T02:07:09 < ds2> zyp: really? the docs suggests I can USB boot
2012-10-24T02:07:24 < ds2> what's the CDxxxxxxx number for an2606?
2012-10-24T02:07:35 < dongs> .google an2606 stmicro
2012-10-24T02:07:43 < ds2> trying not to crash my browser openign 234234092840932804932890423840923809420 windows to download a document from st
2012-10-24T02:08:02 < ds2> I have them saved under the CDxxxxxxx.pdf name
2012-10-24T02:08:11 < zyp> can't you just search?
2012-10-24T02:08:12 < dongs> CD00167594.pdf
2012-10-24T02:08:20 < dongs> fucking st needs to stop this bullshit file naming
2012-10-24T02:08:27 < ds2> thanks.
2012-10-24T02:08:34 < ds2> ST's search opens up 2340240921482384901238490328904238094 windows
2012-10-24T02:08:42 < zyp> I'm talking about file search
2012-10-24T02:08:49 < ds2> and is slower then molassses in the northpole
2012-10-24T02:08:58 < ds2> searching for things on google is a waste of time
2012-10-24T02:09:05 < zyp> if you already have it saved
2012-10-24T02:09:06 < ds2> it'll probally return me links to cds
2012-10-24T02:09:22 < ds2> dongs: thanks
2012-10-24T02:10:03 < Laurenceb_> https://github.com/Laurenceb/STM32F3Discovery/blob/master/lib/STM32F30x_StdPeriph_Driver/src/stm32f30x_i2c.c#L861
2012-10-24T02:10:05 < zyp> whenever I need to open some datasheet I have saved I pull up the spotlight search stuff in OS X and type whatever I'm searching for
2012-10-24T02:10:09 < Laurenceb_> looks actually usable
2012-10-24T02:11:06 < dongs> in what situation?
2012-10-24T02:11:24 < Laurenceb_> actually no
2012-10-24T02:11:25 < zyp> I2C_CR2_SADD, is that some sad-disable bit, so i2c doesn't make you sad?
2012-10-24T02:11:25 < Laurenceb_> hmm
2012-10-24T02:11:29 < Laurenceb_> lol
2012-10-24T02:11:40 < ds2> how criticial is the 100nF per Vdd pin recommendation?
2012-10-24T02:11:42 < Laurenceb_> im trying to understand the f3 code
2012-10-24T02:11:44 < dongs> i2c is fucking always sad
2012-10-24T02:11:46 < Laurenceb_> hehe
2012-10-24T02:11:47 < dongs> ds2: yes
2012-10-24T02:11:54 < Laurenceb_> its more complex that it looked
2012-10-24T02:11:56 < dongs> because its a shit bus
2012-10-24T02:12:02 < Laurenceb_> still, more organised than f1
2012-10-24T02:12:03 < Laurenceb_> http://i.imgur.com/1LlNY.gif
2012-10-24T02:12:10 < dongs> what the fuck happened to all this "api compatible" wank with F3???
2012-10-24T02:12:25 < Laurenceb_> hehe
2012-10-24T02:12:31 < dongs> the pin compatible shit is already out in the garbage, now half hte peripherals are also not compatible to F1?
2012-10-24T02:12:35 < Laurenceb_> hey they fixed the i2c
2012-10-24T02:12:35 < ds2> dongs: any experience with how long of a trace is acceptable for the 100nF?
2012-10-24T02:12:43 < Laurenceb_> it has dfu too
2012-10-24T02:13:03 < Laurenceb_> http://paste.jvnv.net/view/SB24j
2012-10-24T02:13:22 < Laurenceb_> http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=FXOS8700CQ&webpageId=129288320677672732B37A&nodeId=011269B37A&fromPage=tax
2012-10-24T02:13:25 < Laurenceb_> ^looks decent
2012-10-24T02:13:33 < Laurenceb_> maybe a little better than the st stuff
2012-10-24T02:13:52 < dongs> uh, one adc muxed???
2012-10-24T02:14:02 < dongs> is invenshit the only people with separate adc for each axis?
2012-10-24T02:14:15 < Laurenceb_> hmm
2012-10-24T02:14:15 < zyp> doesn't they also just mux?
2012-10-24T02:14:21 < dongs> nope
2012-10-24T02:14:25 < dongs> per axis adc.
2012-10-24T02:14:27 < dongs> look at datasheet
2012-10-24T02:14:39 < Laurenceb_> http://cache.freescale.com/files/sensors/doc/app_note/AN4459.pdf?fpsp=1&WT_TYPE=Application%20Notes&WT_VENDOR=FREESCALE&WT_FILE_FORMAT=pdf&WT_ASSET=Documentation
2012-10-24T02:14:42 < Laurenceb_> interesting
2012-10-24T02:14:43 < zyp> checking
2012-10-24T02:15:02 < zyp> yeah, you're right
2012-10-24T02:15:03 < dongs> 7.5 Block diagram
2012-10-24T02:16:44 < dongs> Laurenceb_: its preproduction but you can sample and order them?
2012-10-24T02:16:55 < dongs> oh, available feb 2013
2012-10-24T02:16:57 < Laurenceb_> ill try
2012-10-24T02:16:58 < ds2> hmmm I am almost sure I ran across usb dfu on the f103
2012-10-24T02:17:06 < Laurenceb_> hehe they have example python code
2012-10-24T02:17:08 < dongs> ds2, sure, with your own bootloader.
2012-10-24T02:17:10 < ds2> but uart works
2012-10-24T02:17:25 < ds2> dongs: is that what the F1 discovery has?
2012-10-24T02:17:57 < dongs> f1 discovery has no usb?? as far as i know
2012-10-24T02:18:14 < ds2> there is a USB connector on the board
2012-10-24T02:18:24 < dongs> yeah its for stlink...
2012-10-24T02:18:29 < ds2> ohhhhh
2012-10-24T02:18:30 < ds2> blah
2012-10-24T02:18:39  * Laurenceb_ is trying to work out simple ways to calibrate F3discovery magno
2012-10-24T02:19:03 < ds2> Laurenceb_: that's the mag in the LIS303DLHC, isn't it?
2012-10-24T02:19:12 < zyp> just steal some calibration implementation from somewhere like you always do? :p
2012-10-24T02:19:19 < Laurenceb_> hehe
2012-10-24T02:19:29 < Laurenceb_> newton raphson  isnt very practical
2012-10-24T02:19:35 < Laurenceb_> i want to run it on the board
2012-10-24T02:19:36 < ds2> isn't the an app note for that
2012-10-24T02:19:44 < Laurenceb_> yes thats what im trying to find
2012-10-24T02:19:50 < Laurenceb_> isnt it from freescale?
2012-10-24T02:19:53 < dongs> Laurenceb_: woot i can sample
2012-10-24T02:19:58 < ds2> the DLHC is the same mag as the HMC part, IIRC
2012-10-24T02:20:04 < Laurenceb_> ooh ill grab some samples
2012-10-24T02:20:06 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 240 seconds]
2012-10-24T02:20:07 < zyp> how does the «wave the phone in a figure eight pattern» algorithms work?
2012-10-24T02:20:24 < Laurenceb_> i think newton raphson
2012-10-24T02:20:25 < ds2> IIRC, the app note is under the HMC part number
2012-10-24T02:20:45 < ds2> zyp: the goal is to turn the device and try to fit the data into a sphere
2012-10-24T02:21:19 < ds2> the adjustments to meet a sphere is the "cal"
2012-10-24T02:22:55 < Laurenceb_> yes
2012-10-24T02:23:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-24T02:23:14 < Laurenceb_> the ideal approach is a newton raphson fitter
2012-10-24T02:23:26 < Laurenceb_> but with a ton of data points that gets tricky
2012-10-24T02:23:36 < Laurenceb_> maybe you could do some tricks
2012-10-24T02:23:43 < Laurenceb_> take combinations of 6 points
2012-10-24T02:23:54 < Laurenceb_> run newton raphson and get goodness of fit
2012-10-24T02:24:02 < Laurenceb_> then do weighted average of all the fits
2012-10-24T02:24:05 < dongs> freescale taking paypal in their sample store= fuck yeah
2012-10-24T02:24:13 < dongs> SYSDR01: Sorry, we could not process your request please contact Freescale representative.
2012-10-24T02:24:16 < dongs> fuuck!
2012-10-24T02:24:26 < Laurenceb_> that was you dont eat all the ram and have to crunch massive matrices
2012-10-24T02:24:27 < Laurenceb_> oops
2012-10-24T02:24:33 < ds2> freescale is a pain to sampel from
2012-10-24T02:24:34 < Laurenceb_> must be cuz they lkike pytohn
2012-10-24T02:24:47 < ds2> Lauranceb_: the app note I recall has a simplified cal
2012-10-24T02:24:52 < Laurenceb_> yes
2012-10-24T02:24:56 < Laurenceb_> its least squares fit
2012-10-24T02:25:06 < Laurenceb_> least squares is basically newton raphson
2012-10-24T02:25:17 < Laurenceb_> i need to get some sleep
2012-10-24T02:25:42 < Laurenceb_> i may have a go at the interleaved 6 point fitter and weighted average idea tomorrow
2012-10-24T02:25:48 < zyp> I broke freescales site
2012-10-24T02:25:50 < zyp> Exception Type: Freescale eBusiness Exception (major: 33060 minor: 110 details: null).
2012-10-24T02:25:56 < Laurenceb_> that could stand a chance of running on F3
2012-10-24T02:26:02 < Laurenceb_> cya all
2012-10-24T02:26:03 < ds2> exceptional business!
2012-10-24T02:26:04 < dongs> well its only $5 i guess. looks like their paypal gateway is broken :(
2012-10-24T02:26:05  * Laurenceb_ zzz
2012-10-24T02:26:10 < dongs> ordered.
2012-10-24T02:26:40 < zyp> oh well
2012-10-24T02:26:52 < zyp> I don't need sensors before I'm going to put them on a board anyway
2012-10-24T02:27:12 < ds2> zyp: is the freescale mag available now?
2012-10-24T02:27:25 < zyp> and I'm not going to do another board revision before I have some working code
2012-10-24T02:27:25 < ds2> last I looked at it, it wasn't available yet
2012-10-24T02:28:02 < zyp> don't ask me
2012-10-24T02:28:16 < ds2> zyp: isn't that what you are trying to order?
2012-10-24T02:28:44 < zyp> I just saw the sample button Laurenceb_ mentioned and clicked it, I didn't put any more tought into that :p
2012-10-24T02:29:02 < ds2> oh blah
2012-10-24T02:29:06 < ds2> nevermind
2012-10-24T02:29:15 < zyp> sorry for disappointing you :)
2012-10-24T02:29:28 < ds2> was hoping to find out if they gave you a near future ship date or not
2012-10-24T02:29:43 < zyp> ask dongs or Laurenceb_, they apparently ordered
2012-10-24T02:30:14 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-24T02:32:47 < dongs> i saw the sample button, clicked it, selected 5 units, paid, ordered.
2012-10-24T02:32:54 < dongs> i think Laurenceb just clicked it, fapped, then went to sleep
2012-10-24T02:33:00 < dongs> yanking his ethernet cord out wiht rage
2012-10-24T02:33:04 < ds2> dongs: this is the mag, right?
2012-10-24T02:33:08 < dongs> so isntead of /quitting he just ping timeouts
2012-10-24T02:33:15 < dongs> ds2: the combo acc/mag he pasted.
2012-10-24T02:33:57 < dongs> its also filthy cheap
2012-10-24T02:34:03 < dongs> like < $2
2012-10-24T02:34:13 < ds2> dongs: did you get a ship date on it?
2012-10-24T02:34:45 < ds2> I tried chasing the LSM303DLHC last year and not sure if I want to try that with another part
2012-10-24T02:34:56 < dongs> nope. reel order said it'd ship in feb 2013.
2012-10-24T02:35:01 < dongs> no idea on sample.
2012-10-24T02:35:13 < dongs> the problem wiht ST acc/mag parts is..
2012-10-24T02:35:15 < dongs> tehy're FUCKING GAY.
2012-10-24T02:35:30 < dongs> <st> lets put 2 separate sensors inside same package, and dont even bother integrating them
2012-10-24T02:36:14 < dongs> if I rememver correctly you had to have 2 separate SPI buses going to the acc/mag parts
2012-10-24T02:36:21 < ds2> no no
2012-10-24T02:36:21 < dongs> and of course totally different register set etc
2012-10-24T02:36:27 < ds2> that's the LSM303DLH
2012-10-24T02:36:39 < ds2> there is a slightly better integrated one...the LSM303DLHC
2012-10-24T02:36:49 < ds2> smaller package, single I2C bus, single supply
2012-10-24T02:37:18 < zyp> I have the LSM303DLM, two buses, two supplys
2012-10-24T02:37:44 < zyp> but to me it doesn't really matter whether they are internally or externally connected, seperate dies are seperate dies
2012-10-24T02:37:49 < ds2> that's the older part...POS
2012-10-24T02:38:01 < ds2> the DLHC seems to address that
2012-10-24T02:38:08 < zyp> I heard it was better than DLH
2012-10-24T02:38:23 < zyp> not sure if DLHC was available when I designed my board
2012-10-24T02:38:37 < ds2> not defending them... but I am short of room for seperate accel and mags
2012-10-24T02:38:41 < ds2> zyp: when did you design the board?
2012-10-24T02:38:52 < zyp> hmm
2012-10-24T02:38:56 < zyp> february perhaps?
2012-10-24T02:39:05 < qyx_> hm, i am just calibrating my lis331
2012-10-24T02:39:13 < qyx_> it is too much temperature dependent
2012-10-24T02:39:22 < ds2> don't think it was available for low volume guys at that time
2012-10-24T02:39:31 < ds2> I was chasing it last Sept
2012-10-24T02:39:32 < qyx_> +- 10mg if i just dont let the air to flow around
2012-10-24T02:39:35 < zyp> yep, I sent the design to fab in the middle of february
2012-10-24T02:40:06 < zyp> and yeah, I remember Laurenceb were going on about DLHC a while later and said I should have chose that instead since it was better
2012-10-24T02:40:46 < ds2> all the distributors was telling me to use the older DLH part
2012-10-24T02:41:03 < ds2> but it requires different voltages on the rails and 2 buses... I was like WTF
2012-10-24T02:41:14 < zyp> DLM doesn't
2012-10-24T02:41:24 < qyx_> zyp: i ordered dlhc's from farnell in march, my new board has footprint for dlhc
2012-10-24T02:41:29 < qyx_> also replacet dlh with dlhc
2012-10-24T02:41:56 < zyp> oh well, it's not that important
2012-10-24T02:42:09 < zyp> will pick better sensors for next board revision
2012-10-24T02:42:22 < zyp> no point in having good sensors before there is any code using them, really
2012-10-24T02:42:24 < qyx_> but i was also confused with that two power supplies and two i2c buses
2012-10-24T02:42:38 < zyp> did they have to be seperate?
2012-10-24T02:42:39 < ds2> I see
2012-10-24T02:42:46 < qyx_> no
2012-10-24T02:42:58 < qyx_> sensors have different addresses
2012-10-24T02:43:06 < ds2> i think the 2 power supplies didn't have the same voltage ranges :(
2012-10-24T02:43:17 < zyp> because IIRC DLM has seperate core and IO supplies, but both support 3.3V
2012-10-24T02:43:33 < ds2> ah DLM solves that too
2012-10-24T02:43:35 < qyx_> ah dlm is another part
2012-10-24T02:43:40 < qyx_> meh, mess
2012-10-24T02:44:10 < zyp> I just realized I'm talking to two guys and not one
2012-10-24T02:44:12 < zyp> :p
2012-10-24T02:44:36 < qyx_> :D
2012-10-24T02:44:57 < qyx_> i am the intruder
2012-10-24T02:46:26 < dongs> two dongs
2012-10-24T02:54:00 -!- CheBuzz_Home [~CheBuzz@76.164.192.124] has quit [Ping timeout: 252 seconds]
2012-10-24T03:05:34 < ds2> is there a STM32F103 reference layout for the BGA144 package?
2012-10-24T03:09:59 < dongs> dunno look at one of -eval boards?
2012-10-24T03:11:18 < dongs> i dont understand why  youre makign a new design for lunix wiht M3 core
2012-10-24T03:11:33 < dongs> do you have a bag of bga f103s or something?
2012-10-24T04:04:03 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-24T04:31:25 < ds2> all the eval boards use tfp
2012-10-24T04:32:56 < ds2> find a small enough and purchaseable SoC
2012-10-24T04:36:41 < dongs> http://www.cnn.com/2012/10/23/world/europe/italy-quake-scientists-guilty/index.html lol wat
2012-10-24T04:36:57 < dongs> ds2, im talking about stm32f4xx
2012-10-24T04:37:54 < ds2> oh
2012-10-24T04:38:01 < ds2> it all comes down to size
2012-10-24T04:38:07 < ds2> and yes, we been through it the other day
2012-10-24T04:43:40 < dongs> https://youtube.com/watch?v=LK3QmqDREgw german beatboxing
2012-10-24T04:47:09 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T05:02:21 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Quit: Leaving.]
2012-10-24T05:08:00 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-24T05:08:41 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-24T05:29:09 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-24T05:31:54 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-10-24T05:32:51 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )]
2012-10-24T05:34:27 < dongs> flyback just discovered arduino wifi shield
2012-10-24T05:37:15 < baird> AnGrY pEoPlE1! .. http://www.element14.com/community/message/62495#62495/l/re-512mb-deliberate-deception
2012-10-24T05:40:12 < dongs> lol shitberry pi drama
2012-10-24T05:42:34 < baird> Knew you'd like..
2012-10-24T05:44:37 < dongs> do you think the one RS is shipping me in march is gonnabe 256 or 512meg
2012-10-24T05:44:39 < dongs> (not that I care)
2012-10-24T05:45:11 < dongs> another major difference beteen PL2303 and any other USB>uart: the others dont crash when running at > 9600 baud
2012-10-24T05:45:50 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T05:46:13 < emeb_mac> whazzat?
2012-10-24T05:46:30 < dongs> flyback: if you paste your ufcking switch uart shit again i will kill you
2012-10-24T05:46:34 < dongs> bcas.tv/paste that shit.
2012-10-24T05:51:48 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Quit: Tzacks]
2012-10-24T05:54:44 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Ping timeout: 248 seconds]
2012-10-24T05:54:50 -!- cjbaird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-24T05:56:22 -!- cjbaird is now known as baird
2012-10-24T05:59:47 -!- txf [~tiago@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-24T06:01:15 < dongs> flyback discovers IDE raid
2012-10-24T06:01:23 < dongs> in 2012
2012-10-24T06:02:02 < emeb_mac> yo dongs
2012-10-24T06:03:29 < dongs> dongs indeed
2012-10-24T06:47:27 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-24T06:56:05 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-24T06:57:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T06:58:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-24T06:58:24 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-24T07:00:35 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T07:13:40 -!- baird is now known as cjbaird_away
2012-10-24T07:13:57 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-24T07:17:22 < emeb> whee - making footprints for leetle xtals.
2012-10-24T07:18:38 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 272 seconds]
2012-10-24T07:18:56 < dongs> dongle crystals
2012-10-24T07:19:17 < emeb> yeah - the little 3x2mm 12MHz ones
2012-10-24T07:19:34 < emeb> 3.2x2.5 to be exact
2012-10-24T07:19:58 < emeb> so tiny, so cute, so squee
2012-10-24T07:20:11 < dongs> I think you forgot "kawaii"
2012-10-24T07:20:28 < emeb> I guess that's japaneseish for all that.
2012-10-24T07:21:03 < emeb> unfortunately I associate Kawaii with a brand of cheesy electronic organs from years ago.
2012-10-24T07:27:41 -!- txf [~tiago@bl8-122-218.dsl.telepac.pt] has quit [Quit: txf]
2012-10-24T07:32:13 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-24T07:34:51 < ds2> emeb: sounds like the ones I spec'ed
2012-10-24T07:35:24 < emeb> ds2: could be : http://www.digikey.com/product-detail/en/NX3225SA-12.000000MHZ-STD-CSR-1/644-1047-1-ND/1128919
2012-10-24T07:37:26 < ds2> looks like same as mine but I am using an 8MHz xtal
2012-10-24T07:37:59 < emeb> you found an 8 that size?
2012-10-24T07:38:10 < ds2> yep
2012-10-24T07:38:17 < ds2> I have size requirements
2012-10-24T07:38:17 < emeb> All the 8MHz I found were 5x3
2012-10-24T07:38:20 < bsdfox> is newlib the way to go?
2012-10-24T07:38:29 < ds2> hence I cannot do anything newer then a F103
2012-10-24T07:38:31 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-24T07:38:45 < bsdfox> I got a toolchain built yesterday and compiling some demo code required a number of changes
2012-10-24T07:39:02 < ds2> let me see if I can find my P/N
2012-10-24T07:39:13 < dongs> uh
2012-10-24T07:39:17 < dongs> 8mhz in 3225 doesnt e xist.
2012-10-24T07:40:05 < ds2> oops, you are right
2012-10-24T07:40:12 < ds2> just checked my datasheets
2012-10-24T07:40:17 < ds2> it is the 32Khz one that is smaller
2012-10-24T07:40:23 < emeb> ah
2012-10-24T07:40:32 < dongs> http://www.ndk.com/images/products/catalog/c_NX3225GD-STD-CRA-3_e.pdf this is 3225 size but its not standard 4 pad layout.
2012-10-24T07:40:43 < emeb> so did you use 12 in a 3225, or 8 in a 5x3?
2012-10-24T07:41:08 < ds2> ABMM2 series in 8MHz
2012-10-24T07:41:12 < dongs> yueha
2012-10-24T07:41:18 < dongs> abmm sounds like 5x3 or maybe even bigger
2012-10-24T07:41:46 < dongs> 6036 looks like.
2012-10-24T07:41:50 < ds2> 3.6x6.0
2012-10-24T07:41:56 < emeb> ds2: you can run the f103 w/ 12MHz HSE
2012-10-24T07:41:58 < ds2> is there a smaller size?
2012-10-24T07:42:08 < emeb> use the smaller 3.2x2.5 pkg
2012-10-24T07:42:13 < dongs> I just linked to 3225 8mhz if you dont mind nonstandard pad pattern
2012-10-24T07:42:13 < ds2> emeb: really?
2012-10-24T07:42:18 < emeb> ds2: yeah
2012-10-24T07:42:20 < dongs> < ds2> emeb: really?
2012-10-24T07:42:25 < dongs> yes, im running F103 off 12mhz, with USB
2012-10-24T07:42:32 < ds2> does that mean I ahve to recalculate the stuff for my drivers?
2012-10-24T07:42:33 < emeb> the HSE is good for like 4-25MHz IIRC
2012-10-24T07:42:35 < dongs> no
2012-10-24T07:42:37 < ds2> or is it just one master PLL I need to reset?
2012-10-24T07:42:41 < dongs> just pll.
2012-10-24T07:42:47 < emeb> ds2: just change the pll setup in the startup code.
2012-10-24T07:42:57 < ds2> 1 PLL or many PLLs?
2012-10-24T07:42:59 < dongs> I have 12mhz code for F103, it was like one line change
2012-10-24T07:43:04 < emeb> ds2: 1 PLL
2012-10-24T07:43:10 < ds2> haven't looked if the STM32's have one centeral PLL or PLLs for each block
2012-10-24T07:43:14 < emeb> the one that boosts HSE to SYSCLK
2012-10-24T07:43:16 < ds2> ah... seems reasonable
2012-10-24T07:43:23 < ds2> I might switch then
2012-10-24T07:43:27 < dongs> and 12mhz is still oK for USB etc.
2012-10-24T07:43:31 < emeb> yup
2012-10-24T07:43:58 < ds2> emeb: same basic 10pF to ground?
2012-10-24T07:44:33 < emeb> ds2: the one I linked at DK is actually 8pf - but should work fine.
2012-10-24T07:44:36 < dongs> i think i use 18pF but i dont really calculate the shit
2012-10-24T07:44:56 < emeb> ds2: dongs: yeah - those values aren't too critical.
2012-10-24T07:45:06 < emeb> likely swamped out by stray cap anyway.
2012-10-24T07:45:15 < ds2> ST has a nice app note on calculating it
2012-10-24T07:45:24 < dongs> http://www.digikey.com/product-detail/en/NX3225GD-8MHZ-STD-CRA-3/644-1178-1-ND/3125567 anyway , 8mhz in 3225.
2012-10-24T07:45:29 < ds2> as long as it will start with 10pF, I'm happy
2012-10-24T07:45:30 < dongs> have not seen that 2 years ago when I last checked.
2012-10-24T07:45:34 < dongs> but nonstandard pattern
2012-10-24T07:45:43 < dongs> still wouldnt use it
2012-10-24T07:46:03 < emeb> and a few more cents if you're watching prices.
2012-10-24T07:46:40 < ds2> if it is only 1 PLL, I'll just use 12MHz
2012-10-24T07:46:44 < dongs> yeah
2012-10-24T07:46:52 < ds2> I assume nothing will break, it'll just run at the wrong speed if I don't reprogram it
2012-10-24T07:46:54 < dongs> and with 12mhz you can overclock
2012-10-24T07:47:04 < dongs> to 144mhz!
2012-10-24T07:47:22 < dongs> i ran my shit at 128mhz, everything worked (including uart)
2012-10-24T07:47:56 < emeb> I ran an STM32F100 that's supposed to top out at 24MHz all the way to 48MHz by accident. Worked fine.
2012-10-24T07:47:58 < ds2> 128MHz on a F1?
2012-10-24T07:48:02 -!- cjbaird_away is now known as baird
2012-10-24T07:48:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-24T07:50:31 < dongs> ds2: ya
2012-10-24T07:52:08 < ds2> wow
2012-10-24T07:52:12 < ds2> dongs: did USB still work?
2012-10-24T07:53:23 < dongs> that board did not have native usb.
2012-10-24T07:53:28 < baird> Ha-- I've just been to buy a copy of Doom3:BFG/PS3, and it was like GenX-thining-hair-Bros 4 Lyfe in the game store with 3 of us (myself, store clerk, and a customer) went Nostaglia Goggles about the Doom games. :)
2012-10-24T07:53:29 < dongs> was uart>usb separate IC.
2012-10-24T07:53:32 < ds2> Oh
2012-10-24T07:53:51 < dongs> i think USB pll is separate though
2012-10-24T07:53:54 < dongs> but  dunno.
2012-10-24T07:53:59 < ds2> oh I see
2012-10-24T07:54:18 < ds2> btw, anyone actually tried power LEDs directly from a GPIO?
2012-10-24T07:54:28 < dongs> i sink them, works fine.
2012-10-24T07:54:31 < ds2> I know ST does it on their boards but in practice, does that work well?
2012-10-24T07:54:40 < ds2> hmmm
2012-10-24T07:54:42 < dongs> apparently PC13-PC15 cant sink or drive
2012-10-24T07:54:56 < dongs> anyway, driving i wiouldn't do it but sink should be ok.
2012-10-24T07:55:46 < ds2> I see
2012-10-24T07:56:31 < ds2> currently I have it driving, let me change that
2012-10-24T07:57:04 < ds2> there is no tricks to let the STM32 side acts an open collector sink, is there? i.e. LED on 5V and STM32 sinking it
2012-10-24T07:57:34 < bsdfox> aren't they all fet outputs?
2012-10-24T07:57:34 < dongs> no, youll need a pnp or wahtever.
2012-10-24T07:57:57 < ds2> oh okay... 3V3 it is
2012-10-24T07:59:16 < ds2> anyone tried driving a piezo disc directly from a GPIO?
2012-10-24T08:15:14 < ds2> wow... what a difference the 3225 part makes
2012-10-24T08:15:15 < emeb> ds2: I've got a bunch of ckts w/ STM32 driving LED anode. Works fine - nice & bright.
2012-10-24T08:15:34 < emeb> But - didn't do the analysis for long-term. Sink is prolly better.
2012-10-24T08:15:52 < ds2> emeb: besides the ST eval boards/discovery stuff?
2012-10-24T08:16:01 < emeb> Yeah - my own boards
2012-10-24T08:16:51 < ds2> hmmm
2012-10-24T08:16:59 < ds2> emeb: how many mA/
2012-10-24T08:17:30 < dongs> just sink them and dotn care
2012-10-24T08:18:41 < emeb> ds2: Just looked at the datasheet - GPIO is rated to sink/source up to 20ma.
2012-10-24T08:18:49 < emeb> I usually set my LEDs for 10ma.
2012-10-24T08:18:54 < ds2> emeb: i mean on your boards
2012-10-24T08:18:57 < ds2> ah
2012-10-24T08:42:02 -!- claude is now known as Claude
2012-10-24T09:11:08 < bsdfox> High-current source output driver (20 mA) on one pin.
2012-10-24T09:11:09 < bsdfox>  High-current sink driver (20 mA) on true open-drain pins.
2012-10-24T09:11:34 < bsdfox> doesn't seem like that's a valid blanket statement emeb
2012-10-24T09:20:54 < ds2> wheeeeeeeeeeeeee PCB gerbers passes DFM
2012-10-24T09:25:25 < emeb> bsdfox: From the datasheet for the F3xx: "The GPIOs (general purpose input/outputs) can sink or source up to +/-8 mA, and sink or
2012-10-24T09:25:26 < emeb> source up to +/- 20 mA (with a relaxed VOL/VOH)."
2012-10-24T09:25:43 < emeb> for driving an LED the VOH/VOL can be relaxed.
2012-10-24T09:25:52 < emeb> ds2: yay!
2012-10-24T09:26:43 < ds2> sigh.. quoting system is broke
2012-10-24T09:26:56 < emeb> who you quoting from?
2012-10-24T09:27:08 < ds2> 4pcb
2012-10-24T09:27:23 < ds2> it thinks I want 0.001 dia holes even though I said no microvias
2012-10-24T09:27:39 < ds2> so that gets kicked out to a 'call us'
2012-10-24T09:27:46 < emeb> I imagine that costs extra.
2012-10-24T09:28:56 < emeb> bbiab...
2012-10-24T09:29:00 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-24T09:32:17 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T09:54:38 -!- Claude is now known as claude
2012-10-24T10:04:36 < baird> \o/ .. http://www.microbeetechnology.com.au/
2012-10-24T10:11:02 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-24T10:24:29 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-24T10:26:55 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-10-24T10:37:03 -!- claude is now known as claude_work
2012-10-24T10:37:06 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-24T10:37:55 -!- claude_work is now known as Claude_work
2012-10-24T10:39:01 < dongs> $399 gtfo
2012-10-24T10:47:15 < baird> Was seriously thinking about getting one of the Bees.. but unfortunately all the software is still being 'blackmailed' by a moderator of the 'Bee webforum. Fuck webforums.
2012-10-24T10:48:08 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-24T10:48:42 < baird> There's this 'Microbee Software Preservation Project' he runs, but to access it, you need to suck-up to the before-mentioned moderator, and then it's only if he cares to like you. Classic abuse of mod power-tripping trying to ensure he's got the only site on the net, making it rather difficult to mirror Other People's Software elsewhere that's more free..
2012-10-24T10:50:32 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Remote host closed the connection]
2012-10-24T10:50:37 < baird> dongs: https://github.com/norinori2222/boyfriend_require/blob/master/README-en.md
2012-10-24T10:51:08 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-24T10:51:22 < baird> I failed the grade at condition 3 ;_;
2012-10-24T10:51:56 < dongs> PC: MacBook Pro Retina, Mid 2012 (finally bought a new one after 3 and 1/2 years!)
2012-10-24T10:52:00 < dongs> get the fuck out.
2012-10-24T10:54:05 < baird> I did have 5 friends.. but two of them are dead now. :/
2012-10-24T11:03:54 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-24T11:04:03 < zyp> hmm
2012-10-24T11:04:20 < zyp> looking at F0 RM, it's indeed using same flash interface as F1/F3
2012-10-24T11:06:06 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 264 seconds]
2012-10-24T11:11:04 < baird> serious lol: http://images.h4ck.in/_img/00014_ceefax_final_words.jpg
2012-10-24T11:23:04 < dongs> http://www.dayvmattt.com/dm/wp-content/uploads/2012/09/09-05-31-9.jpg unrelated
2012-10-24T11:24:12 < baird> http://www.crunchyroll.com/anime-news/2012/10/23/video-japanese-idol-makes-music-with-hacked-furby-toys ... Chew on that, Jeri Ellsworth. :P
2012-10-24T11:25:18 < dongs> 404
2012-10-24T11:26:02 < baird> the inside link: https://www.youtube.com/watch?v=cu7HQWHnzp8
2012-10-24T11:26:27 < dongs> hey thats my dicknplace
2012-10-24T11:27:14 < dongs> hm no
2012-10-24T11:27:22 < dongs> i think old version of mine
2012-10-24T11:31:27 < baird> the 'competition' is now onto girls wearing maid outfits. Keep that in mind for your store blog.
2012-10-24T11:35:43 < baird> wut .. https://www.youtube.com/watch?v=kVgLLELRXYM
2012-10-24T11:44:30 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-24T12:00:40 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-24T12:10:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-24T12:10:40 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-24T12:11:10 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-24T12:19:28 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Read error: Connection reset by peer]
2012-10-24T12:19:54 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-24T12:31:04 < dongs> haha...
2012-10-24T12:31:12 < dongs> got some pcb submission wiht animu girls on it
2012-10-24T12:31:47 < dongs> http://i.imgur.com/0z4dO.png attn zyp/etc
2012-10-24T12:32:31 < zyp> don't you have mostly japanese customers? not surprising then :p
2012-10-24T12:37:09 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Remote host closed the connection]
2012-10-24T12:40:10 <+Steffanx> lol dongs
2012-10-24T12:40:36 <+Steffanx> No line about no anime crap in your terms of use?
2012-10-24T12:41:24 <+Steffanx> Oh, not that rules out 90% of your customers
2012-10-24T12:42:30  * Laurenceb is wondering if a kalman filter can track magnetometer calibration
2012-10-24T12:42:59 < Laurenceb> i was thinking of trying DCM for attitude and a kalman to track magno calibration
2012-10-24T12:43:04 < Laurenceb> a little different
2012-10-24T12:47:25 < dongs> isnt DCM tarducopter failcity
2012-10-24T12:47:34 < Laurenceb> it seems to work quite well
2012-10-24T12:52:43 < dongs> lemme know if you figure it out.
2012-10-24T12:52:58 < dongs> or you can just run that ripped openpilot shit on f3disco
2012-10-24T12:53:26 < dongs> https://github.com/lilvinz/OpenPilot/wiki/Discovery-F3-what-this-is-about
2012-10-24T12:58:20 < Laurenceb> wut
2012-10-24T12:58:32 < dongs> no wuts
2012-10-24T12:58:37 < Laurenceb> its fucking big
2012-10-24T12:58:42 < Laurenceb> that was ported fast
2012-10-24T12:59:02 < dongs> yea that guy's gonna get paid by some french clonerz
2012-10-24T12:59:08 < dongs> who are gonna make some F4 hardweare with shit sensors on it
2012-10-24T12:59:12 < dongs> so he's quick
2012-10-24T12:59:33 < Laurenceb> oh he got a prerelease f3dfisco
2012-10-24T13:03:08 < Laurenceb> http://en.wikipedia.org/wiki/Anal_masturbation
2012-10-24T13:03:16 < Laurenceb> theres way too much info on there
2012-10-24T13:05:12 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-24T13:05:13 <+Steffanx> I bet it's nothing new for you
2012-10-24T13:09:09 < karlp> zyp what package/tool did this come from? http://bin.jvnv.net/f/hhzfA.png
2012-10-24T13:09:26 < dongs> some usb sniffer shit
2012-10-24T13:12:47 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-24T13:21:52 < zyp> it's from the beagle stuff
2012-10-24T13:23:33 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-24T13:35:07 < Laurenceb> hmm
2012-10-24T13:35:22 < Laurenceb> looks like you could add an ekf to track magno gain and bias
2012-10-24T13:35:38 < Laurenceb> just using the magnetic field magnitude data
2012-10-24T13:36:13 < Laurenceb> so it'd also be possible to add the tracking onto the openpilot ekf code
2012-10-24T13:36:33 < Laurenceb> as it normalises magno prior to running the filter
2012-10-24T13:53:50 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-24T13:55:13 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 256 seconds]
2012-10-24T14:00:19 < dongs> https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-ash4/420901_298871520221212_294343486_n.jpg
2012-10-24T14:02:04 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-24T14:09:04 < dongs> http://i.imgur.com/P23X8.png added i2c to trolllogger just in case
2012-10-24T14:09:43 < BrainDamage> why there are no dicks in the silkscreen?
2012-10-24T14:09:51 < dongs> thats next
2012-10-24T14:11:13 < Laurenceb> what are all the 3 pin headers for?
2012-10-24T14:11:33 < dongs> secret stuffz
2012-10-24T14:13:42 < Laurenceb> http://www.december212012.com/articles/news/2012_Olympic_mascot_reigns_in_New_World_Order.htm
2012-10-24T14:15:50 < dongs> wat
2012-10-24T14:18:16 < Laurenceb> http://i.imgur.com/L23qM.jpg
2012-10-24T14:18:18 < Laurenceb> thats what
2012-10-24T14:19:41 < dongs> more Laurenceb rape fantasies
2012-10-24T14:20:07 < dongs> http://i.imgur.com/42xLS.jpg
2012-10-24T14:26:29 < Laurenceb> http://i.imgur.com/1W7Vc.jpg
2012-10-24T14:26:38 < Laurenceb> yo dawg we heard you liked monitors
2012-10-24T14:27:39 < BrainDamage> looks like medical equipment
2012-10-24T14:27:45 < BrainDamage> from the shape / colour
2012-10-24T14:27:47 < Laurenceb> you think
2012-10-24T14:27:54 < baird> <keenu> Whoa: http://www.raspberrypi.org/archives/2221
2012-10-24T14:29:00 < Laurenceb> yeah but does it have accelerated X windows yet
2012-10-24T14:29:08 < BrainDamage> impressive considering broadcom has been notorious for doing the opposite
2012-10-24T14:30:12 < baird> "All these projects could now potentially port these libraries and make use of the full hardware accelerated graphics facilities of the Raspberry Pi [..]"
2012-10-24T14:30:25 < Laurenceb> ItÂ’s not going to have an immediate impact on X acceleration,
2012-10-24T14:30:26 < Laurenceb> lol
2012-10-24T14:32:14 < dongs> 2012: Xwindows still sucks cock
2012-10-24T14:34:18 < baird> But it's good at it.
2012-10-24T14:37:34 < dongs> attn Laurenceb http://www.goodluckbuy.com/captian-stm32-mpu6050-ms5611-control-board-for-quadcopter-robot.html
2012-10-24T14:37:42 < dongs> Therefore: if you want to buy this flight control board,you should:
2012-10-24T14:37:43 < dongs> - can program the board,with C programming foundation
2012-10-24T14:38:15 < Laurenceb> lol
2012-10-24T14:39:03 < Laurenceb> does it love you long time?
2012-10-24T14:39:09 < dongs> doubtful
2012-10-24T14:39:41 < dongs> http://www.goodluckbuy.com/images/detailed_images2/image/2(224).jpg
2012-10-24T14:40:08 < dongs> Autoploit
2012-10-24T14:41:41 < Laurenceb> http://www.goodluckbuy.com/sige-gn3s-sampler-v3-gps-data-acquisition-board-.html
2012-10-24T14:41:56 < dongs> GN3S more like GNAA
2012-10-24T14:42:08 < dongs> free shipping, what a deal
2012-10-24T14:42:46 < baird> And you need to pipe its data through MATLAB
2012-10-24T14:43:11 < Laurenceb> i have several
2012-10-24T14:43:18 < Laurenceb> from when it was $200
2012-10-24T14:43:23 < Laurenceb> maybe i should sell
2012-10-24T14:43:34 < baird> I gather it's a grad-student hardware thing
2012-10-24T14:45:34 < dongs> http://googleblog.blogspot.jp/2011/10/fall-sweep.html oh fuck
2012-10-24T14:45:38 < dongs> google code search is gone now?
2012-10-24T14:52:21 < Laurenceb> in b4 "google hit wtc"
2012-10-24T14:52:36 < dongs> yaman.
2012-10-24T14:57:13 < dongs> found original place
2012-10-24T14:57:14 < dongs> http://item.taobao.com/item.htm?spm=a1z10.1.w11.3.We9egV&id=15229749913
2012-10-24T14:57:18 < dongs> lol @ shitty chinese GUI
2012-10-24T15:01:15 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-24T15:06:19 < dongs> http://img01.taobaocdn.com/imgextra/i1/27996160/T22QpxXm8MXXXXXXXX_!!27996160.jpg
2012-10-24T15:06:22 < dongs> fail!
2012-10-24T15:06:23 < dongs> diodes on rx/tx lnes
2012-10-24T15:13:31 < qyx_> whats failly about that
2012-10-24T15:15:45 < dongs> uh...???
2012-10-24T15:15:46 < zyp> heh, that broadcom driver contains a bunch of #ifdef WIN32 stuff
2012-10-24T15:15:54 < dongs> qyx_: isnt it obvious?
2012-10-24T15:16:42 < zyp> qyx_, diodes on push-pull data lines is not very sane
2012-10-24T15:17:28 < qyx_> meh
2012-10-24T15:17:47 < zyp> signal current is flowing in both directions depending on whether the signal is high or low, blocking it in one direction is not sane
2012-10-24T15:18:14 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-24T15:18:22 < dongs> http://img04.taobaocdn.com/imgextra/i4/27996160/T24XVNXf8bXXXXXXXX_!!27996160.jpg another china fail
2012-10-24T15:18:43 < dongs> spot the fail
2012-10-24T15:18:43 < zyp> oh, leds
2012-10-24T15:18:49 < zyp> I didn't see the pic at first
2012-10-24T15:18:57 < zyp> I thought you meant diodes as in diodes
2012-10-24T15:18:58 < qyx_> zyp: leds
2012-10-24T15:19:03 < dongs> leds are diodes.
2012-10-24T15:19:13 < zyp> but they are not connected in series
2012-10-24T15:19:16 < zyp> as I thought
2012-10-24T15:19:28 < BrainDamage> leds are shittier diodes than "normal" diodes even
2012-10-24T15:19:30 < zyp> leds are fine, apart from the extra load they add
2012-10-24T15:19:50 < karlp> how much load do they add?
2012-10-24T15:20:03 < dongs> enough that no proper circuit should ever use htem like that
2012-10-24T15:20:04 < qyx_> but dongs can make that obvious part even more obvious
2012-10-24T15:20:13 < karlp> I thought it was fine to put leds on the rx/tx lines to get free blinkies on data transfer
2012-10-24T15:20:26 < BrainDamage> 10-100pF or so and few mA
2012-10-24T15:20:30 < zyp> sure, for slow signals it is
2012-10-24T15:20:35 < karlp> hmmm.
2012-10-24T15:20:36 < karlp> interesting.
2012-10-24T15:20:40 < zyp> slow/short
2012-10-24T15:20:48 < karlp> we've had interrmittent corruptions turn up.
2012-10-24T15:20:58 < karlp> and leds were recently added to the rx/tx lines....
2012-10-24T15:21:01 < qyx_> i understand it is not good idea to put them on some high freq lines
2012-10-24T15:21:11 < qyx_> but that serial is at most 115200 :S
2012-10-24T15:21:16 < karlp> I think I will have to go and have a tlak to the hardware people.
2012-10-24T15:21:18 < BrainDamage> you should buffer them with a transistor if it's high speed
2012-10-24T15:21:26 < karlp> we're only at 19200 though,
2012-10-24T15:21:37 < zyp> BrainDamage, or rather weak drivers
2012-10-24T15:22:36 < dongs> http://i.imgur.com/GWVIh.png this worked for me.
2012-10-24T15:22:58 < dongs> nice and bright even at 115k baud
2012-10-24T15:23:18 < qyx_> dongs: i have nice and bright leds on rx/tx lines on reference board from ftdi
2012-10-24T15:23:29 < qyx_> without transistors
2012-10-24T15:23:31 < dongs> > ftdi
2012-10-24T15:23:34 < dongs> found your problem
2012-10-24T15:23:41 < qyx_> dont
2012-10-24T15:23:44 < qyx_> do that
2012-10-24T15:23:44 < BrainDamage> at high speed like 10 MHz, 100pF is ~150 Ohm, so you really should buffer them
2012-10-24T15:23:55 < BrainDamage> even with weak drives
2012-10-24T15:24:55 < dongs> qyx, ftdi has rx/tx led lines.
2012-10-24T15:25:19 < dongs> as in, they have gpio lines that blink on rx/tx specifically for leds.
2012-10-24T15:25:30 < dongs> with internal buffer/transistor/wahtever.
2012-10-24T15:25:34 < qyx_> hm, okay, maybe they are connected to them :S
2012-10-24T15:25:40 < dongs> i would imagine so
2012-10-24T15:26:25 < zyp> using a ghetto solution when a proper is provided would be dumb
2012-10-24T15:27:16 < dongs> yeah.
2012-10-24T15:27:29 < dongs> and lunix-like
2012-10-24T15:27:50 < zyp> by the way, do you have anything against ftdi apart from the price?
2012-10-24T15:27:58 < dongs> just that
2012-10-24T15:28:29 < zyp> right, that's what I'm also thinking
2012-10-24T15:29:32 < dongs> still waiting for someone to do ftdi emulation for stm32 though
2012-10-24T15:30:04 < dongs> got a couple projects where I can use it, but not critical. masquerading as tarduino 2560 is also fine (but thier drivers are unsigned)
2012-10-24T15:30:28 < qyx_> last time i saw pl2303 emulation in nuttx
2012-10-24T15:30:35 < dongs> lol.
2012-10-24T15:30:39 < dongs> how fucking terrible.
2012-10-24T15:30:42 < dongs> that chip is garbage.
2012-10-24T15:31:02 < qyx_> o\
2012-10-24T15:32:01 < dongs> http://code.google.com/p/picoc/ hhhm
2012-10-24T15:32:33 < qyx_> old
2012-10-24T15:32:38 < dongs> new2me
2012-10-24T15:32:48 < dongs> source code: tar.bz2
2012-10-24T15:32:50 < dongs> closing page
2012-10-24T15:35:28  * Laurenceb runs his printerscript
2012-10-24T15:35:43 < Laurenceb> finds all printers on lan, and trolls them
2012-10-24T15:50:19 < baird> Huh. zik@zikzak.net is an old Bro of mine from APANA..
2012-10-24T15:55:27 < baird> Being a Melbourne guy, he would've had to deal with Assange's shit as well..
2012-10-24T16:04:13 < Laurenceb> but asshenge is awesome right
2012-10-24T16:07:41 -!- txf [~txf@193.136.166.125] has joined ##stm32
2012-10-24T16:10:38 < Laurenceb> http://www.shevibe.com/squirt-the-realistic-ejaculating-dildo-by-doc-johnson.aspx
2012-10-24T16:10:51 < Laurenceb> looks like an invention we could get behind
2012-10-24T16:32:22 < dongs> pasted to my german math pal
2012-10-24T16:32:38 < dongs> the pdf, not the dildo
2012-10-24T16:36:27 < Laurenceb> we do not recommend this as sugar can cause yeast infections. A preferred substance would be plain yogurt. The PH balance more closely resembles that inside the body
2012-10-24T16:36:41 < Laurenceb> good to know
2012-10-24T16:39:00 <+Steffanx> Laurenceb isn't it time to talk in another channel about your sexual preferences?
2012-10-24T16:39:16 < Laurenceb> #yogurt
2012-10-24T16:39:28 < dongs> which masturbation handbook is that out of
2012-10-24T16:39:39 < Laurenceb> the shevibe
2012-10-24T16:40:05 <+Steffanx> dongs instead of that c interpreter you should run tinycc :P
2012-10-24T16:40:17 <+Steffanx> *TCC
2012-10-24T16:40:23 < dongs> Steffanx: old, and useless
2012-10-24T16:40:41 <+Steffanx> Old yes, but you can make it useful :)
2012-10-24T16:40:56 < Laurenceb> dongs only runs microvision
2012-10-24T16:44:56 < dongs> attn Laurenceb http://imageshack.us/a/img853/590/photowcz.jpg
2012-10-24T16:52:04 -!- dekar [~dekar@dslb-188-107-175-168.pools.arcor-ip.net] has joined ##stm32
2012-10-24T16:52:06 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-24T17:06:13 -!- dekar_ [~dekar@95.143.165.210] has joined ##stm32
2012-10-24T17:06:16 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-24T17:06:25 < Tectu> demo_graph1.png (1 MB) exceeds the maximum allowed size of 1024 KB
2012-10-24T17:06:35 < Laurenceb> chibios?
2012-10-24T17:09:03 -!- dekar [~dekar@dslb-188-107-175-168.pools.arcor-ip.net] has quit [Ping timeout: 276 seconds]
2012-10-24T17:09:52 < Tectu> no
2012-10-24T17:13:25 < Laurenceb> haha wtf
2012-10-24T17:13:36 < Laurenceb> just got an email invite to "laptand uk"
2012-10-24T17:13:41 < Laurenceb> *lapland
2012-10-24T17:15:15 -!- txf [~txf@193.136.166.125] has quit [Remote host closed the connection]
2012-10-24T17:25:06 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T17:28:12 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-24T17:28:53 < Laurenceb> http://habitat.habhub.org/
2012-10-24T17:28:54 < Laurenceb> wtf
2012-10-24T17:29:02 < Laurenceb> i just want to launch a balloon
2012-10-24T17:31:33 < dongs> its on ghettohub, it must be amazing
2012-10-24T17:33:18 < qyx_> are there any constrains or regulations for small high altitude baloons?
2012-10-24T17:33:31 < qyx_> small i mean something <1m
2012-10-24T17:33:31 < dongs> you bet there is
2012-10-24T17:33:45 < dongs> qyx, <1m on teh ground becomes like 10m up high
2012-10-24T17:33:54 < qyx_> hm, thats true
2012-10-24T17:34:10 < qyx_> but there are no palnes here
2012-10-24T17:34:11 < qyx_> planes
2012-10-24T17:34:15 < dongs> you totally want that shit getting sucked into commercial airliner turbine intake
2012-10-24T17:34:23 < dongs> thereis on the  way there.
2012-10-24T17:34:25 < qyx_> :>
2012-10-24T17:35:31 < qyx_> here i mean >10km where the baloon is high
2012-10-24T17:35:33 < Laurenceb>  <1m is fine
2012-10-24T17:35:41 < qyx_> but of course i wasnt considering turbines
2012-10-24T17:35:52 < Laurenceb> an inflated balloon will go straight through a turbine
2012-10-24T17:37:14 < qyx_> around 7-8 small standard baloons are needed to lift 100g
2012-10-24T17:37:21 < qyx_> i think i am visiting circus next time
2012-10-24T17:44:38 <+Steffanx> dongs .. they made the gpu crap off the rpi opensores. Isn't get bad news for you?
2012-10-24T17:44:54 <+Steffanx> gpu crap = VideoCore driver
2012-10-24T17:45:11 < dongs> Steffanx: already mirrored incase they change their mind
2012-10-24T17:48:30 < dongs> bedtime
2012-10-24T17:48:45 <+Steffanx> gn
2012-10-24T17:55:35 < zyp> well, «gpu crap»
2012-10-24T17:55:58 < zyp> it's apparently just a thin RPC client for the gpu driver running on the videocore itself
2012-10-24T17:58:19 < jpa-> hmm, .net micro framework is ported to STM32F4; and it is under apache license
2012-10-24T17:58:25 < zyp> hmm
2012-10-24T17:58:37 < jpa-> of course it is interpreted only, no JIT
2012-10-24T17:58:50 < zyp> F1 blackmagic flashloader is apparently not working on F0
2012-10-24T17:59:01 < jpa-> but seems like a viable alternative to squirrel & lua, if one likes C# better
2012-10-24T17:59:04 < zyp> here I was hoping adding F0 support would be as easy as F3 :p
2012-10-24T17:59:44 < Laurenceb> F3 is working with bmp?
2012-10-24T17:59:49 <+Steffanx> of course
2012-10-24T17:59:52 < zyp> sure, I added support yesterday
2012-10-24T17:59:53 < Laurenceb> cool
2012-10-24T18:00:31 < zyp> https://github.com/gsmcmullin/blackmagic/commit/cb41855 <- it was really hard
2012-10-24T18:01:08 <+Steffanx> heh
2012-10-24T18:03:01 < Laurenceb> nice
2012-10-24T18:03:16 < Laurenceb> way less disorganised than texane
2012-10-24T18:03:21 < zyp> yep
2012-10-24T18:04:52 -!- Claude_work is now known as claude
2012-10-24T18:07:20 < Laurenceb> http://www.sanslogic.co.uk/images/warning-xp.png
2012-10-24T18:11:12 < emeb_mac> guess I'll have to pull and reflash
2012-10-24T18:11:55 < zyp> I'm having a go at F0 support now
2012-10-24T18:12:04 < zyp> detection works fine, but it hangs while flashing
2012-10-24T18:14:50 < emeb_mac> loader stub overwriting something critical?
2012-10-24T18:15:23 < zyp> no, flash shows as empty after reattaching
2012-10-24T18:15:35 < emeb_mac> trashing sram?
2012-10-24T18:16:07 < emeb_mac> or else flashing doesn't work the way you think it does.
2012-10-24T18:16:09 < zyp> it might be that flashloader is not v6-m compatible, but it doesn't appear to use any thumb-2 instructions at least
2012-10-24T18:22:17 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-24T18:28:17 < karlp> https://github.com/jpallister/stlink/commit/d31557df1ac28bfa6773d0decaba1b6087da9417 seems to support it to?
2012-10-24T18:30:37 < zyp> I'm talking about BMP, not stlink
2012-10-24T18:31:21 < karlp> I was only suggesting it might show something that would help you work out what was missing
2012-10-24T18:32:14 < zyp> oh, wait
2012-10-24T18:32:17 < zyp> I just realized
2012-10-24T18:32:35 < zyp> v6-m doesn't have cbz
2012-10-24T18:32:45 < zyp> so that's where it would be locking up
2012-10-24T18:37:29 -!- barthess [~barthess@178.154.90.70] has joined ##stm32
2012-10-24T18:50:08 -!- claude is now known as Claude
2012-10-24T18:50:16 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-24T18:58:01 < zyp> yep, that was it
2012-10-24T19:08:34 -!- Netsplit *.net <-> *.split quits: Tectu, Ademan, mrcan, bsdfox_
2012-10-24T19:08:38 -!- Tectu_ [tectu@kunsmann.eu] has joined ##stm32
2012-10-24T19:08:47 -!- Netsplit over, joins: Ademan
2012-10-24T19:08:56 -!- Netsplit over, joins: mrcan
2012-10-24T19:08:56 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-24T19:08:57 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-24T19:08:57 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-24T19:10:28 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-24T19:12:12 -!- txf [~txf@193.136.166.125] has joined ##stm32
2012-10-24T19:16:20 -!- dekar_ [~dekar@95.143.165.210] has quit [Read error: Connection reset by peer]
2012-10-24T19:19:17 -!- dekar_ [~dekar@dslb-178-004-068-151.pools.arcor-ip.net] has joined ##stm32
2012-10-24T19:19:20 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-24T19:33:24 < Laurenceb> http://yaleaerospace.com/live-4/
2012-10-24T19:33:28 < Laurenceb> epic fail live
2012-10-24T19:34:53 < Tectu_> how much does an  abs()  cost?   I guess not that much since it only "removes" the signing?
2012-10-24T19:35:14 -!- txf [~txf@193.136.166.125] has quit [Remote host closed the connection]
2012-10-24T19:35:31 -!- txf [~txf@193.136.166.125] has joined ##stm32
2012-10-24T19:36:20 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-24T19:37:26 -!- dekar_ [~dekar@dslb-178-004-068-151.pools.arcor-ip.net] has quit [Read error: Operation timed out]
2012-10-24T19:38:26 < BrainDamage> depends on the format
2012-10-24T19:38:38 < BrainDamage> in 2s complement, you need to flip bits too
2012-10-24T19:39:15 < Laurenceb> like the yale guys probably need to
2012-10-24T19:40:17 < BrainDamage> I quite like highaltitude "gentlemen" suggestion to run the predictor on the landing and tell them since they are gonna need it :p
2012-10-24T19:55:11 -!- tavish_ [~tavish@120.56.238.130] has joined ##stm32
2012-10-24T19:57:25 -!- tavish_ [~tavish@120.56.238.130] has quit [Client Quit]
2012-10-24T19:57:42 -!- tavish_ [~tavish@120.56.238.130] has joined ##stm32
2012-10-24T19:59:54 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 240 seconds]
2012-10-24T20:02:49 -!- tavish [~tavish@120.56.142.230] has joined ##stm32
2012-10-24T20:02:50 -!- tavish [~tavish@120.56.142.230] has quit [Changing host]
2012-10-24T20:02:50 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-24T20:03:53 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-24T20:03:56 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-24T20:05:03 -!- tavish_ [~tavish@120.56.238.130] has quit [Ping timeout: 260 seconds]
2012-10-24T20:33:03 -!- txf [~txf@193.136.166.125] has quit [Remote host closed the connection]
2012-10-24T20:43:02 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-24T20:43:20 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-24T20:43:23 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-24T20:46:06 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 240 seconds]
2012-10-24T21:03:03 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-63.pptp.stw-bonn.de] has joined ##stm32
2012-10-24T21:08:45 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 252 seconds]
2012-10-24T21:20:45 -!- txf [~txf@193.136.166.125] has joined ##stm32
2012-10-24T21:22:00 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-10-24T21:30:33 -!- txf [~txf@193.136.166.125] has quit [Remote host closed the connection]
2012-10-24T21:30:35 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-24T21:33:07 -!- daku is now known as DaKu
2012-10-24T21:41:57 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-24T21:43:40 < Tectu_> http://www.abload.de/img/20121024_203635udufl.jpg
2012-10-24T21:43:54 < Tectu_> 320x240
2012-10-24T21:46:50 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-24T21:52:53 < Erlkoenig> i see 3,264px × 2,448px
2012-10-24T21:55:09 < Tectu_> lol
2012-10-24T22:04:47 < zyp> 18:25:26 < bliblok> Hva er blackmagic?
2012-10-24T22:04:47 < zyp> 18:26:18 < zyp> jtag/swd-greia jeg har
2012-10-24T22:04:47 < zyp> 18:26:35 < zyp> driver og legger til støtte for flere targets for den
2012-10-24T22:04:54 < zyp> woops
2012-10-24T22:05:21 < zyp> disregard that.
2012-10-24T22:05:31 < jpa-> okej
2012-10-24T22:11:16 < Erlkoenig> wouldn't have been able to read it anyways? :D
2012-10-24T22:21:33 < emeb> some kinda strange moon-man language. :)
2012-10-24T22:22:29 < Laurenceb_> agga wagga woo
2012-10-24T22:22:44 < Erlkoenig> probably danish or norwegian or something like that :D
2012-10-24T22:23:10 < Laurenceb_> failish
2012-10-24T22:23:42 < jpa-> it's just mistyped swedish
2012-10-24T22:24:00 < emeb> swedish on a broken keyboard?
2012-10-24T22:24:11 < jpa-> ja
2012-10-24T22:25:11 < jpa-> människör borje minst skriva der svenska rätt
2012-10-24T22:25:29 < jpa-> inte fel så att som jag skriver
2012-10-24T22:25:31 < BrainDamage> utf-8 fail?
2012-10-24T22:25:48 < emeb> Well - you wouldn't be laughing with a boatload of Vikings heading your way. But they didn't have keyboards.
2012-10-24T22:26:00 < BrainDamage> I see "?" in your sentence, looks like you're not using utf-8 as encoding to send
2012-10-24T22:26:20 < zyp> looks right to me
2012-10-24T22:26:44 < jpa-> dom bra svenska människör har iso8859-15 som deras charset
2012-10-24T22:26:47 < Erlkoenig> emeb: i would if i had a strategic LAZ0R
2012-10-24T22:27:04 < zyp> but I might have fallback to iso-8859-1 when it's not valid utf-8
2012-10-24T22:27:46 < BrainDamage> much better, thanks
2012-10-24T22:27:59 < BrainDamage> I intentionally leave the fallback so I can bitch people about it
2012-10-24T22:28:17 < emeb> Erlkoenig: Vikings with sunglasses and polished shields scoff at your puny LAZOR.
2012-10-24T22:29:04 < Erlkoenig> emeb: a reichsflugscheibe-load of nazis equipped with flamethrowers, machine guns and tactical missiles laugh at your sunglasses.
2012-10-24T22:29:40 < emeb> Erlkoenig: well there's that.
2012-10-24T22:29:41 < Erlkoenig> also, submarines
2012-10-24T22:29:44 < Erlkoenig> :3
2012-10-24T22:29:56 < emeb> I call Godwin.
2012-10-24T22:30:40 < BrainDamage> when he's gonna get there?
2012-10-24T22:31:08 < emeb> He already did, about 5 lines back.
2012-10-24T22:31:16 < Laurenceb_> godwinner A10
2012-10-24T22:32:40 < BrainDamage> http://imgur.com/gallery/1NjM7
2012-10-24T22:34:07 < emeb> funny/sad
2012-10-24T22:34:58 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: leaving]
2012-10-24T22:38:46 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 245 seconds]
2012-10-24T22:53:22 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-24T22:58:56 -!- dekar [~dekar@212.255.27.61] has joined ##stm32
2012-10-24T22:59:00 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-24T23:05:14 < scrts> daaamn, I need a chip already with wifi :/
2012-10-24T23:11:32 < Erlkoenig> and i need a digital de/modulator for 27MHz
2012-10-24T23:12:00 < Erlkoenig> or any lother lower frequency, which can be used without license
2012-10-24T23:13:14 -!- Claude is now known as claude
2012-10-24T23:30:42 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
--- Day changed Thu Oct 25 2012
2012-10-25T00:01:52 -!- barthess [~barthess@178.154.90.70] has quit [Ping timeout: 244 seconds]
2012-10-25T00:06:00 < emeb> DK shipment arrived. Accidentally ordered 100x 0201 caps. So tiny! So useless!
2012-10-25T00:06:24 -!- Tectu_ is now known as Tectu
2012-10-25T00:06:56 < karlp> small enough to pass through the system, just use them as cake decorations.
2012-10-25T00:07:10 < emeb> hopefully ROHS
2012-10-25T00:07:24 < zyp> emeb, cute
2012-10-25T00:07:50 < Erlkoenig> DK? DildoKing?
2012-10-25T00:08:02 < emeb> lol
2012-10-25T00:08:32 < emeb> Damn Kraken
2012-10-25T00:09:01 < emeb> I wonder if I could actually hand-solder these things.
2012-10-25T00:09:30 < Erlkoenig> use laser vision
2012-10-25T00:10:39 < emeb> nope - they're about the same size as the tips of my tweezers. Doubt that the 6/6 PCB rules would provide proper pad dimensions.
2012-10-25T00:20:29 < BrainDamage> be careful not to inhale them
2012-10-25T00:21:05 < Erlkoenig> throw them in a blob of glue :>
2012-10-25T00:26:59 < Laurenceb_> i was planning un building a stm32 board with 0201/0402 for everything
2012-10-25T00:27:02 < Laurenceb_> fun project
2012-10-25T00:27:36 < Erlkoenig> hm what size are the small R's on the F4 Disco?
2012-10-25T00:27:53 < Laurenceb_> 0402
2012-10-25T00:28:04 < Laurenceb_> wish there was flip chip F3
2012-10-25T00:28:35 < Erlkoenig> hmm they already look very small :D
2012-10-25T00:28:43 < Erlkoenig> flip?
2012-10-25T00:31:45 < zyp> WLCSP
2012-10-25T00:32:31 < zyp> it's basicly a die that you attach some solder point on top of, flip over and solder to the board
2012-10-25T00:32:47 < zyp> therefore the term flip-chip
2012-10-25T00:33:15 < Erlkoenig> wat
2012-10-25T00:33:18 < Erlkoenig> black magic
2012-10-25T00:33:21 < zyp> look at the esd protection on the F4discovery (U8)
2012-10-25T00:33:25 < zyp> it's a flip-chip
2012-10-25T00:34:31 < Laurenceb_> i wanted to make a disposable pulse oximeter with usb
2012-10-25T00:34:38 < Laurenceb_> to plug into a smartphone
2012-10-25T00:34:56 < zyp> got L1 support halfway implemented on L1 now
2012-10-25T00:34:57 < Laurenceb_> but lqfp-48 F3 is too big
2012-10-25T00:35:04 < zyp> supports erase but not write
2012-10-25T00:35:14 < zyp> on BMP*
2012-10-25T00:50:11 <+Steffanx> Hmm, the L1 needs a different flash loader than the F1 zyp ?
2012-10-25T00:55:03 < zyp> I'm not sure it needs one
2012-10-25T00:55:21 < zyp> I'm trying without first to see how fast it will be
2012-10-25T00:55:56 < zyp> it has a half-page mode where you load up 128 bytes and it writes them all at once
2012-10-25T00:56:13 < zyp> and no need for checking any busy flag in between
2012-10-25T01:04:04 < zyp> hmm
2012-10-25T01:04:15 < zyp> almost working no
2012-10-25T01:04:15 < zyp> w
2012-10-25T01:04:17 < zyp> (gdb) compare-sections
2012-10-25T01:04:17 < zyp> Section .vectors, range 0x8000000 -- 0x8000130: matched.
2012-10-25T01:04:17 < zyp> Section .init_array, range 0x8000130 -- 0x8000140: matched.
2012-10-25T01:04:20 < zyp> Section .text, range 0x8000140 -- 0x8000478: MIS-MATCHED!
2012-10-25T01:04:22 < zyp> Section .data, range 0x8000478 -- 0x8000484: matched.
2012-10-25T01:05:11 < Erlkoenig> are you using texane st-link?
2012-10-25T01:05:24 < zyp> no, BMP
2012-10-25T01:05:46 < zyp> I believe texane stuff already has L1-support
2012-10-25T01:06:03 < Erlkoenig> what's BMP?
2012-10-25T01:06:18 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-25T01:06:54 < zyp> blackmagic probe
2012-10-25T01:07:55 < Erlkoenig> ah
2012-10-25T01:10:12 < Erlkoenig> wah... for my F4 disco, on linux, i started a (texane) st-util session, then an arm-gdb session, and was able to debug my prog. then i modified the code, recompiled it, did "load" in gdb to re-flash the F4, did "kill" and "run", but the code won't start (breakpoint at main() isn't reached, LED doesn't blink). Replugging the USB however starts the program (LED blinks)
2012-10-25T01:10:52 < zyp> tried not doing kill?
2012-10-25T01:11:00 < zyp> I usually don't
2012-10-25T01:11:09 < Erlkoenig> doesn't help...
2012-10-25T01:11:19 < zyp> also, set breakpoint at your entry symbol instead
2012-10-25T01:11:23 < Erlkoenig> "run" does kill implicitly
2012-10-25T01:11:59 < zyp> sounds like it get stuck in lockup for some reason
2012-10-25T01:12:15 < zyp> can you still halt and inspect registers after typing run?
2012-10-25T01:13:40 < Erlkoenig> ah, i can set a breakpoint at Reset_Handler, and then do step-by-step execution
2012-10-25T01:14:14 < Erlkoenig> "break main" and "c" however doesnt return
2012-10-25T01:14:55 < Erlkoenig> hmm, Ctrl+C reveals it's stuck at an endless loop that gets executed when an interrupt occurs which has no implementation
2012-10-25T01:15:12 < Erlkoenig> but i only enabled the Timer Interrupt, and wrote a handler for that
2012-10-25T01:16:43 < Erlkoenig> can i read IPSR from GDB to determine what exception occured?
2012-10-25T01:17:07 < zyp> info registers
2012-10-25T01:17:17 < zyp> lowest byte of xpsr
2012-10-25T01:17:36 < emeb> progress on an F373 breakout: http://imagebin.org/233156
2012-10-25T01:17:57 < emeb> still missing the mini 10-pin JTAG/SWD connector.
2012-10-25T01:18:20 < Erlkoenig> xpsr           0x61000003  ... would mean HardFault
2012-10-25T01:18:25 < zyp> yep
2012-10-25T01:18:41 < zyp> SCB registers would have information about what caused it
2012-10-25T01:19:11 < zyp> by the way, what's the value of pc?
2012-10-25T01:19:15 < Erlkoenig> hmm how to retrieve that?
2012-10-25T01:20:16 < zyp> if pc is 0xfffffffe, it's stuck in lockup
2012-10-25T01:20:35 < Erlkoenig> pc             0x8000234
2012-10-25T01:21:12 < zyp> then it's good
2012-10-25T01:21:15 < zyp> just a normal hardfault
2012-10-25T01:21:30 < Erlkoenig> that's the adress of that endless loop
2012-10-25T01:21:37 < zyp> yep
2012-10-25T01:22:13 < Erlkoenig> how comes that the HardFault only occurs upon that "warm start" - and the startup code is the example code by ST
2012-10-25T01:22:36 < zyp> do: x/wx 0xe000ed28
2012-10-25T01:22:47 < zyp> it will read out SCB_CFSR
2012-10-25T01:23:10 < zyp> http://infocenter.arm.com/help/topic/com.arm.doc.dui0553a/Cihcfefj.html
2012-10-25T01:23:27 < zyp> it will contain the reason for the hardfault
2012-10-25T01:23:57 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-25T01:25:43 < Erlkoenig> ah very nice. the bus fault status register is 0b1000 0010   while all others are zero
2012-10-25T01:26:19 < zyp> huh?
2012-10-25T01:26:40 < zyp> check the value of CFSR before checking any other registers
2012-10-25T01:26:44 -!- DaKu is now known as daku
2012-10-25T01:27:01 < zyp> oh, wait, I misread you
2012-10-25T01:27:07 < Erlkoenig> yes that's what i did ;)
2012-10-25T01:27:09 < zyp> I didn't see the binary, I read it as hex
2012-10-25T01:27:19 < Erlkoenig> it consists of 4 regs, of which 3 are zero
2012-10-25T01:27:28 < Erlkoenig> argh, 2/3
2012-10-25T01:27:46 < zyp> so, PRECISERR, BFARVALID
2012-10-25T01:27:58 < Erlkoenig> 1 = a data bus error has occurred, and the PC value stacked for the exception return points to the instruction that caused the fault.
2012-10-25T01:28:18 < zyp> sounds simple enough
2012-10-25T01:28:31 < Erlkoenig> that means the first dword on the stack is the code adress that caused the fault?
2012-10-25T01:28:42 < zyp> no
2012-10-25T01:29:18 -!- kerwood1 [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-25T01:29:21 < zyp> try: x/8wx $sp
2012-10-25T01:29:46 < zyp> it will print the stacked exception frame, the seventh word should be the pc value
2012-10-25T01:30:02 < Erlkoenig> 0x2001ffe0:	0xe000ed88	0x00f00000	0x08000000	0xe000ed00
2012-10-25T01:30:02 < Erlkoenig> 0x2001fff0:	0x00000000	0x0800040f	0x0800041c	0x61000000
2012-10-25T01:30:11 < zyp> so 0x8000f1c
2012-10-25T01:30:21 < zyp> 0x800041c*
2012-10-25T01:30:35 < zyp> now just look it up in the assembly listing
2012-10-25T01:31:12 < Erlkoenig> hm. it's the SystemInit function by ST.:
2012-10-25T01:31:15 < Erlkoenig> 800041a:       609a            str     r2, [r3, #8]
2012-10-25T01:31:15 < Erlkoenig>  800041c:       bd80            pop     {r7, pc}
2012-10-25T01:31:15 < Erlkoenig>  800041e:       bf00            nop
2012-10-25T01:31:26 < Erlkoenig> these are the last 3 instructions in that function
2012-10-25T01:31:46 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 244 seconds]
2012-10-25T01:32:16 < zyp> ah
2012-10-25T01:33:23 < zyp> sounds like it tries to pop an invalid value into pc then
2012-10-25T01:33:49 < zyp> hmm, no
2012-10-25T01:34:03 < Erlkoenig> hm this is BFAR, the address causing the fault:   (gdb) x 0xE000ED38
2012-10-25T01:34:03 < Erlkoenig> 0xe000ed38:	0x20020000
2012-10-25T01:35:01 < Erlkoenig> thats a reserver address oO
2012-10-25T01:35:01 < zyp> right, it's trying to pop from an empty stack
2012-10-25T01:35:06 < zyp> this is F4, right?
2012-10-25T01:35:09 < Erlkoenig> just above the SRAM
2012-10-25T01:35:13 < BrainDamage> FBAR? more like FUBAR!
2012-10-25T01:35:14 < zyp> exactly
2012-10-25T01:35:19 < Erlkoenig> yep, the F4, on the F4 Disco
2012-10-25T01:35:52 < zyp> so it initializes the stack to end of ram, then somehow tries to pop from the empty stack
2012-10-25T01:36:33 < zyp> possibly a bug in texane stlink, possibly a bug in the code
2012-10-25T01:36:55 < Erlkoenig> http://pastebin.com/wAXWmVdY   this is the failing code
2012-10-25T01:37:22 < Erlkoenig> it does "push    {r7, lr}"   and   "pop     {r7, pc}" ...
2012-10-25T01:37:23 < zyp> you say it starts fine on powerup
2012-10-25T01:37:35 < Erlkoenig> yup exactly
2012-10-25T01:37:40 < zyp> what about hitting the reset button without debugger attached?
2012-10-25T01:37:49 < zyp> that would also be a sort of warm restart, does it matter?
2012-10-25T01:38:05 < Erlkoenig> shall i try? i'll loose the debugging information ;)
2012-10-25T01:38:11 < zyp> yes, try
2012-10-25T01:38:18 < Erlkoenig> okay...
2012-10-25T01:38:28 < zyp> debugging information can always be reproduced ;)
2012-10-25T01:41:13 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-25T01:42:41 < Erlkoenig> oO ... powercycling starts the program properly. pressing reset restarts it normally. starting st-link, gdb, doing "c", killig st-link leaves the program running - when THEN hitting reset, the program does NOT start...
2012-10-25T01:43:11 < Thorn> Erlkoenig: is the code you're debugging compiled with -O0?
2012-10-25T01:43:18 < Erlkoenig> without any -O
2012-10-25T01:43:27 < zyp> same as -O0 then
2012-10-25T01:43:27 < Erlkoenig> that probably means -O2
2012-10-25T01:43:30 < Erlkoenig> wah
2012-10-25T01:43:31 < Thorn> if not, debuggers go haywire
2012-10-25T01:43:35 < Erlkoenig> yes, -O0
2012-10-25T01:43:41 < Thorn> as I've noticed
2012-10-25T01:43:41 < zyp> Thorn, bullshit
2012-10-25T01:44:08 < zyp> debugging optimized code is no problem if you understand what you are doing
2012-10-25T01:44:31 < Thorn> the problem is, gdb doesn't.
2012-10-25T01:45:14 < Erlkoenig> how can gdb/st-link mess up the F4 in a way that it can't properly reset without power-cycling... oO
2012-10-25T01:45:16 < zyp> sure it does, gdb understand perfectly well that you are executing instructions and not source lines
2012-10-25T01:45:36 < zyp> so as long as you're not confusing yourself by thinking about executing source lines, there is no problem
2012-10-25T01:46:00 < Erlkoenig> btw, it's fascinating how the ARM provides information about the crash... ;-)
2012-10-25T01:46:19 < HTT-Bird> zyp: and modern debug info formats can provide the debugger with enough information to understand optimized code
2012-10-25T01:46:50 < zyp> I've been thinking about writing a python script to read out the SCB values and decode the error
2012-10-25T01:47:09 < zyp> but I haven't done it yet
2012-10-25T01:47:12 < zippe> Erlkoenig: there are a number of ways that you can screw the F4 up
2012-10-25T01:47:16 < Erlkoenig> zyp: better would be an eclipse plugin ;-)
2012-10-25T01:47:30 < zyp> except I'm not using eclipse
2012-10-25T01:47:35 < Erlkoenig> meh :D
2012-10-25T01:47:58 < Erlkoenig> zippe: that are stronger than the allmighty reset button... annoying
2012-10-25T01:51:04 -!- kerwood1 [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Quit: Leaving.]
2012-10-25T01:54:08 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-25T01:59:59 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-25T02:05:30 < dongs> sup dongs/laurencb
2012-10-25T02:15:52 < emeb> dongs: -> f3 breakout in progress: http://imagebin.org/233159
2012-10-25T02:15:59 < zippe> Erlkoenig: yes; I'm pretty sure that they're fundamentally chip bugs
2012-10-25T02:17:15 < Erlkoenig> zippe: oh great. didn't run across any while using AVR... ;-)
2012-10-25T02:20:02 < dongs> emeb: so huge! ;d
2012-10-25T02:20:29 < emeb> terrifyingly ginormous.
2012-10-25T02:20:44 < dongs> hm, thts lqfp48? looks bigger than that heh
2012-10-25T02:20:48 < dongs> is that pattern from datasheet?
2012-10-25T02:21:02 < dongs> because it looks a bit shorter pads than waht it should be
2012-10-25T02:21:06 < emeb> that's the pattern I have in my library.
2012-10-25T02:21:16 < emeb> pads are a tad short, but they work.
2012-10-25T02:21:38 < emeb> pins don't hang off the edges at least. :)
2012-10-25T02:22:16 < dongs> well figure 52/53 in f103x datasheet has the lqfp4 pad pattern.
2012-10-25T02:22:31 < dongs> the pads should be like 1.2mm long
2012-10-25T02:22:32  * emeb looks
2012-10-25T02:22:53 < dongs> yours look even less than 1mm maybe
2012-10-25T02:22:54 < dongs> cant tell
2012-10-25T02:24:35 < emeb> I've used that fp on stm32f100x and it worked OK. Probably worth checking it over.
2012-10-25T02:25:24 < dongs> it looks like an opensores footprint from here.
2012-10-25T02:25:27 < dongs> so you better fix it.
2012-10-25T02:25:51 < emeb> Don't recall where I got it.
2012-10-25T02:25:56 < dongs> also im not quite sure how that xtal is working for you
2012-10-25T02:26:01 < dongs> why are caps before the crystal
2012-10-25T02:26:15 < dongs> are the 2 signals routed between them?
2012-10-25T02:26:34 < dongs> oh i see one between one outside
2012-10-25T02:26:45 < emeb> just grounds for the caps
2012-10-25T02:27:20 < dongs> wtf why 0805 R for led?
2012-10-25T02:27:39 < dongs> oh wait.. some caps are 0402?
2012-10-25T02:27:43 < dongs> and the rest of teh shit is 0603?
2012-10-25T02:27:50 < emeb> all caps are 0603
2012-10-25T02:28:10 < emeb> the LED resistors are 0805 because I didn't update them.
2012-10-25T02:28:12 < dongs> whats up with R2 pattern then
2012-10-25T02:28:13 < dongs> oh
2012-10-25T02:28:38 < emeb> (usually use 0805 resistors, but not on this board)
2012-10-25T02:28:47 < zyp> C3 and C4 has different sizes
2012-10-25T02:28:51 < emeb> Could get 0603 LEDs too I guess.
2012-10-25T02:29:08 < emeb> Yeah - one is 0805 - 10uf MLCC
2012-10-25T02:29:39 < zyp> and L1 looks uncomfortably close to the chip
2012-10-25T02:29:40 < emeb> same as C1, C2
2012-10-25T02:29:55 < emeb> yep - need to pull it back a bit.
2012-10-25T02:30:10 < emeb> plenty of room to J2
2012-10-25T02:31:17 < emeb> need to substitute fp for LEDs w/ polarized silk.
2012-10-25T02:36:04 < Erlkoenig> __WFE ();
2012-10-25T02:36:04 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-63.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-25T02:39:42 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-25T02:45:17 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Quit: Ex-Chat]
2012-10-25T02:45:34 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-25T02:47:22 < emeb> dongs: heh - had the wrong fp for the F373. Was a 0.8mm pitch.
2012-10-25T02:47:45 < emeb> don't know how that got in there - the board I forked this from had the right one.
2012-10-25T02:48:32 < dongs> geez.
2012-10-25T02:48:36 < dongs> thats opensauce for you
2012-10-25T02:49:38 < Laurenceb_> trollercoaster
2012-10-25T02:51:49 < emeb> heh - more like me not paying attention. pilot error.
2012-10-25T02:59:45 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Quit: Leaving.]
2012-10-25T03:00:38 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 244 seconds]
2012-10-25T03:11:07 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-25T03:11:29 < baird> Woo. the F3disco is awaiting for me...
2012-10-25T03:17:17 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-25T03:18:54 < emeb> sweet!
2012-10-25T03:40:34 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-25T03:43:41 < Tectu> http://www.youtube.com/watch?v=QcqgMXY2c6Y&feature=g-crec-u
2012-10-25T03:44:52 < Tectu> definitley need to do that on my F4 with FSMC instead of F1 with GPIO
2012-10-25T04:26:57 < GargantuaSauce> spoilers: there are no good gui irc client
2012-10-25T04:29:06 < GargantuaSauce> Tectu: you find chibios tolerable?
2012-10-25T04:29:27 < GargantuaSauce> I got sick of it the second I learned you can't do IO in an interrupt
2012-10-25T04:29:28 < Tectu> GargantuaSauce, yep
2012-10-25T04:29:38 < Tectu> GargantuaSauce, not sure what you mean
2012-10-25T04:29:45 < Tectu> when have you checked it out last time either?
2012-10-25T04:29:52 < GargantuaSauce> like a month ago
2012-10-25T04:29:57 < GargantuaSauce> also it fucks up if you use the fpu
2012-10-25T04:30:03 < GargantuaSauce> which makes me a sad panda
2012-10-25T04:30:10 < GargantuaSauce> (on the f4 i mean)
2012-10-25T04:30:48 < Tectu> GargantuaSauce, huh? I used FPU yesterday and it worked very well
2012-10-25T04:31:00 < Tectu> http://www.youtube.com/watch?v=NlqijOg_TJg&feature=g-crec-u
2012-10-25T04:31:13 < GargantuaSauce> i think it is hit and miss...there's a comment somewhere saying "you're on your own"
2012-10-25T04:31:39 < GargantuaSauce> are you sure that's hard fpu? seems a little slow
2012-10-25T04:31:59 -!- txf_ [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-25T04:32:38 < Tectu> it takes 20 times longer per image if FPU is off
2012-10-25T04:32:43 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit []
2012-10-25T04:32:43 -!- txf_ is now known as txf
2012-10-25T04:33:37 < GargantuaSauce> mine is about the same framerate using a bitbanged interface instead of fsmc, at 320x240 instead of 180x180
2012-10-25T04:33:41 < GargantuaSauce> same number of iteration
2012-10-25T04:33:57 < ossifrage> I've been having this problem with the STM32 timers are 'on' but not counting.
2012-10-25T04:34:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-25T04:34:54 < Tectu> GargantuaSauce, can you share your sources?
2012-10-25T04:35:01 < Tectu> GargantuaSauce, I only have that size because I keep everything in ram
2012-10-25T04:35:15 < Tectu> GargantuaSauce, https://github.com/Tectu/ChibiOS-GFX/blob/master/demos/mandelbrot/mandelbrot.c
2012-10-25T04:37:14 < GargantuaSauce> nothing special about my sauce, only real difference is no blitting
2012-10-25T04:37:25 < GargantuaSauce> which could explain the speed difference i guess
2012-10-25T04:39:14 < Tectu> GargantuaSauce, how are you doing it
2012-10-25T04:39:42 < Tectu> I guess you refresh pixel-wise instead of picture-wise?
2012-10-25T04:39:53 < GargantuaSauce> yeah basically
2012-10-25T04:41:27 < Tectu> GargantuaSauce, you think that this will be faster?
2012-10-25T04:41:36 < GargantuaSauce> well there's one less copy involved
2012-10-25T04:41:59 < GargantuaSauce> i mean i'd think fsmc would be faster than bitbanging it
2012-10-25T04:42:07 < GargantuaSauce> but i dunno \(O_o)/
2012-10-25T04:42:14 < Tectu> GargantuaSauce, it should, yes
2012-10-25T04:42:23 < Tectu> GargantuaSauce, I am pretty sure that the FSMC interface isn't the bottleneck here
2012-10-25T04:42:27 < Tectu> GargantuaSauce, how fast are you?
2012-10-25T04:42:34 < Tectu> way faster or just 0.00001ns faster? :D
2012-10-25T04:43:08 < GargantuaSauce> not an order of magnitude or anything, just a bit
2012-10-25T04:43:33 < Tectu> so you don't get like 30FPS?
2012-10-25T04:43:57 < GargantuaSauce> it might be the math too...i don't have function calls in the loop, nor redundant multiplications
2012-10-25T04:44:09 < GargantuaSauce> which will only make a difference if you have optimizations off i guess
2012-10-25T04:44:16 < BrainDamage> just pastebin the sauce
2012-10-25T04:44:22 < Tectu> GargantuaSauce, just give the sauce :)
2012-10-25T04:44:23 < Tectu> if you can
2012-10-25T04:44:32 < Tectu> and I'll tell you in 5minutes
2012-10-25T04:44:52 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 248 seconds]
2012-10-25T04:45:29 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-25T04:46:44 < Tectu> GargantuaSauce, can I expect something comming? got everything ready to try
2012-10-25T04:46:53 < GargantuaSauce> yeah just a sec
2012-10-25T04:46:59 < GargantuaSauce> cleaning it up a bit but not a lot
2012-10-25T04:48:00 < Tectu> k
2012-10-25T04:48:05 < Tectu> what optimization did you take?
2012-10-25T04:48:16 < GargantuaSauce> none, it doesnt work at all with them on :V
2012-10-25T04:48:40 < Tectu> so -O0 i guess?
2012-10-25T04:48:45 < GargantuaSauce> yup
2012-10-25T04:48:51 < GargantuaSauce> even -Os breaks ;_;
2012-10-25T04:49:05 < BrainDamage> sometimes I should just blackmail you with #peet logs
2012-10-25T04:49:06 < Tectu> mine works nice with all optimizations
2012-10-25T04:49:17 < Tectu> BrainDamage, what's #peet?
2012-10-25T04:49:25 < GargantuaSauce> <_<
2012-10-25T04:49:31 < GargantuaSauce> a place i have said plenty of stupid shit
2012-10-25T04:49:49 < Tectu> GargantuaSauce, well, let's say it this way: «I know how you feel»
2012-10-25T04:49:52 < GargantuaSauce> don't even want to know which gem you're brandishing
2012-10-25T04:51:24 < Tectu> GargantuaSauce, lol, with -O0 it takes like 10s per image
2012-10-25T04:51:29 < Tectu> with -Os not
2012-10-25T04:58:35 < Tectu> GargantuaSauce, sauce? -.-
2012-10-25T04:59:34 < GargantuaSauce> i probably should have verified that you're using a similar lcd chipset
2012-10-25T04:59:37 < GargantuaSauce> 8080 interface?
2012-10-25T05:00:36 < Tectu> currently, my library supports:   SSD1289, SSD1963, S6D1121, Nokia6610
2012-10-25T05:00:44 < Tectu> adding a new driver is quite easy
2012-10-25T05:00:58 < Tectu> wait, that wasn't the question, eh?
2012-10-25T05:00:59 < GargantuaSauce> oh yeah you're on top of this crap
2012-10-25T05:01:02 < GargantuaSauce> it's the first one
2012-10-25T05:01:14 < Tectu> i tought so
2012-10-25T05:03:26 < GargantuaSauce> http://xn--d-bga.su/lcd.c
2012-10-25T05:03:35 < Tectu> wtf  @ URL
2012-10-25T05:03:49 < GargantuaSauce> are you not enamored with my soviet union domain
2012-10-25T05:05:08 < Tectu> well
2012-10-25T05:05:09 < Tectu> that URL
2012-10-25T05:05:13 < Tectu> that URL well
2012-10-25T05:05:17 < Tectu> it just looks wrong
2012-10-25T05:05:56 < BrainDamage> utf-8 fail?
2012-10-25T05:06:17 < GargantuaSauce> it's actually dé.su but basically every irc client fails at punycode so i just give the ascii version
2012-10-25T05:06:42 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 264 seconds]
2012-10-25T05:06:59 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-25T05:07:21 < Tectu> GargantuaSauce, no output \o/
2012-10-25T05:07:44 < GargantuaSauce> ha ha ha now i get to say the foss mantra
2012-10-25T05:07:48 < GargantuaSauce> WORKS ON MY BUILD!
2012-10-25T05:07:53 < Tectu> thahaha :D
2012-10-25T05:07:58 < Tectu> wait, you said -Os
2012-10-25T05:08:00 < Tectu> -O0
2012-10-25T05:08:04 < GargantuaSauce> no -Os
2012-10-25T05:08:08 < Tectu> still no
2012-10-25T05:08:12 < Tectu> well, then no output :D
2012-10-25T05:08:45 < GargantuaSauce> did you actually change your physical configuration
2012-10-25T05:08:51 < Tectu> http://codepad.org/6zWluXmx
2012-10-25T05:09:22 < Tectu> gdispGetWidth() returns 320, gdispGetHeight() 240
2012-10-25T05:09:31 < GargantuaSauce> oh you reaaallly gutted it
2012-10-25T05:10:02 < GargantuaSauce> oh
2012-10-25T05:10:04 < Tectu> I what?
2012-10-25T05:10:18 < Tectu> wtf is gutted in this context
2012-10-25T05:10:30 < GargantuaSauce> my beautiful lcd interfacing
2012-10-25T05:10:39 < BrainDamage> disembowel
2012-10-25T05:10:47 < GargantuaSauce> give the function -1.0,-1.0,1.0,1.0
2012-10-25T05:10:49 < BrainDamage> or if you prefer, you removed the internals
2012-10-25T05:11:08 < GargantuaSauce> it's in map coords not screen
2012-10-25T05:11:30 < GargantuaSauce> and what does drawpixel do exactly, set the address register and then the data?
2012-10-25T05:11:55 < GargantuaSauce> because i only set the address once
2012-10-25T05:12:00 < Tectu> oh, you stream
2012-10-25T05:12:19 < Tectu> gdispDrawPixel() does take x and y coordinates of the screen
2012-10-25T05:12:38 < GargantuaSauce> i hope your blit function doesn't call that repeatedly
2012-10-25T05:13:36  * GargantuaSauce blits all over flyback's back porch
2012-10-25T05:16:57 < dongs> got some more tape feeders, yay
2012-10-25T05:16:59 < dongs> less work
2012-10-25T05:17:09 < dongs> i need to send one off to china to get it cloned
2012-10-25T05:19:20 < Tectu> GargantuaSauce, isn't it possible to make the code in the same performance using gdispDrawPixel() '
2012-10-25T05:19:47 < GargantuaSauce> i don't believe so
2012-10-25T05:20:06 < GargantuaSauce> since you have to set the address for each pixel
2012-10-25T05:20:27 < GargantuaSauce> which is at least one command
2012-10-25T05:22:14 < Tectu> couln't get any output yet
2012-10-25T05:22:38 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-25T05:23:04 < GargantuaSauce> parameters are around the map's domain now?
2012-10-25T05:23:04 < Tectu> GargantuaSauce, are you sure that this is right?   lcd_mandelbrot(-1.0, -1.0, 1.0, 1.0);
2012-10-25T05:23:05 < upgrdman> stm32f4 has "Up to 192+4 Kbytes of SRAM including 64-Kbyte of CCM (core coupled memory) data RAM" ... why CCM? what does it offer/why use it?
2012-10-25T05:23:08 < GargantuaSauce> no :P
2012-10-25T05:23:15 < GargantuaSauce> but i think so!
2012-10-25T05:24:18 < GargantuaSauce> i assume your big init functions turn on the fpu
2012-10-25T05:25:13 < GargantuaSauce> (and yes now that i double checked i am sure those params are right)
2012-10-25T05:29:38 < GargantuaSauce> wait i should have noticed this earlier
2012-10-25T05:30:07 < GargantuaSauce> you pass x and y (which are real and complex coefficients) to the drawpixel function
2012-10-25T05:30:27 < GargantuaSauce> that should be i and j :V
2012-10-25T05:31:39 < Tectu> ah, and then this should work=
2012-10-25T05:32:11 < GargantuaSauce> should is one of those words that has little traction among the gods of sauce
2012-10-25T05:33:19 < Tectu> getting output  \o/
2012-10-25T05:33:21 < Tectu> quite fast o0
2012-10-25T05:33:27 < Tectu> okay, let's zoom in
2012-10-25T05:33:34 < Tectu> what for parameters did you use to zoom
2012-10-25T05:34:46 < GargantuaSauce> smaller
2012-10-25T05:35:01 < GargantuaSauce> they're the boundaries of the viewport
2012-10-25T05:35:05 < Tectu> hmm
2012-10-25T05:35:15 < Tectu> so you do panning and zooming over those four parameters
2012-10-25T05:37:08 < Tectu> GargantuaSauce, can you show me your usage code?
2012-10-25T05:38:02 < GargantuaSauce> i havent bothered with version control yet and that stuff was just a quick test so i don't have the full deal
2012-10-25T05:38:17 < GargantuaSauce> one line remains, lcd_mandelbrot(-2.0f*zoom+cx, -1.5f*zoom+cy, 2.0f*zoom+cx, 1.5f*zoom+cy);
2012-10-25T05:38:28 < GargantuaSauce> cx and cy are centre point, zoom decrease to zoom in
2012-10-25T05:40:05 < Tectu> in what range is zoom=
2012-10-25T05:40:44 < GargantuaSauce> oh you can probably go down to .01 or maybe a tenth of that before the iterations become too prominent
2012-10-25T05:40:56 < GargantuaSauce> depends on where you're looking of course
2012-10-25T05:41:03 < dongs> dongirations
2012-10-25T05:41:32 < GargantuaSauce> of the dongelbrot complex map?
2012-10-25T05:41:44 < dongs> http://i.imgur.com/kMMLk.jpg
2012-10-25T05:41:45 < dongs> lool
2012-10-25T05:42:03 < GargantuaSauce> >celeron
2012-10-25T05:42:09 < GargantuaSauce> arguably it was obsolete on day zero
2012-10-25T05:42:52 < GargantuaSauce> i'm sure that floppy drive till performs beautifully though
2012-10-25T05:43:50 < BrainDamage> http://imgur.com/gallery/TdxBv
2012-10-25T05:44:25 < GargantuaSauce> speaking of old stuff i am now the proud owner of no less than THREE commodore 64s
2012-10-25T05:44:36 < BrainDamage> is any of them working?
2012-10-25T05:44:53 < GargantuaSauce> i am pretty sure at least one is, haven't jury rigged a ps yet
2012-10-25T05:45:01 < GargantuaSauce> 9v ac :/
2012-10-25T05:45:07 < BrainDamage> the next question is: are they worth more than their weight as scrap?
2012-10-25T05:45:23 < GargantuaSauce> doubtful, but they're a piece of history man!
2012-10-25T05:45:30 < GargantuaSauce> plus in any case i am going to fuck with the SID
2012-10-25T05:45:33 < GargantuaSauce> because how could i not
2012-10-25T05:47:20 < Tectu> GargantuaSauce, okay, looks good now
2012-10-25T05:47:34 < Tectu> just making like 10 to 11 "zoom" now
2012-10-25T05:47:48 < Tectu> and I am not yet sure why xD
2012-10-25T05:48:02 < GargantuaSauce> the real question is why not
2012-10-25T05:48:42 < Tectu> did you limit your code?
2012-10-25T05:48:44 < Tectu> i guess not
2012-10-25T05:48:58 < Tectu> maybe I miss something... zoom is float:
2012-10-25T05:48:58 < Tectu>         zoom -= 0.1;
2012-10-25T05:48:59 < Tectu>         if(zoom <= 0.00001)
2012-10-25T05:48:59 < Tectu>             zoom = 1;
2012-10-25T05:49:17 < GargantuaSauce> i made it zoom in and out
2012-10-25T05:49:34 < Tectu> I don't get deep enough -.-
2012-10-25T05:49:36 < Tectu> seriously :;D
2012-10-25T05:49:44 < GargantuaSauce> with 64 iterations still?
2012-10-25T05:50:02 < GargantuaSauce> guess you found a good region, i was too lazy to look hard
2012-10-25T05:50:15 < GargantuaSauce> i've spent many an hour perusing the set previously <_<
2012-10-25T05:50:22 < GargantuaSauce> you should find some regions that look cool, save their parameters and do cubic interpolation between them for the demo
2012-10-25T05:50:36 < Tectu> cubic whaaat :D
2012-10-25T05:50:47 < GargantuaSauce> linear would be fine too :P
2012-10-25T05:52:51 < Tectu> GargantuaSauce, I don't get it... i can zoom out like 1000 zooms, but I cannot soom in more than 10. when I has done 10, it resets and starts all over
2012-10-25T05:53:46 < GargantuaSauce> oh
2012-10-25T05:53:55 < GargantuaSauce> check what you're decrementing it by :P
2012-10-25T05:54:23 < Tectu> well, that's just the steop of the zooming
2012-10-25T05:54:29 < Tectu> how far I zoom in per step
2012-10-25T05:54:33 < Tectu> the amount of steps is way too low
2012-10-25T05:55:01 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-25T05:55:25 < GargantuaSauce> <_<
2012-10-25T05:55:43 < Tectu> i should be able to zoom it like 100 steps, right?
2012-10-25T05:55:55 < Tectu> and I don't get it :D
2012-10-25T05:56:02 < Tectu> i can zoom out from the "default" view for 100 steps
2012-10-25T05:56:07 < Tectu> but not in for 100 steps
2012-10-25T05:57:03 < Tectu> and I still don't see my fault
2012-10-25T05:57:32 < GargantuaSauce> multiply instead of subtracting
2012-10-25T05:59:33 < Tectu> wtf?
2012-10-25T06:00:49 < GargantuaSauce> math and stuff
2012-10-25T06:00:59 < Tectu> well
2012-10-25T06:01:02 < Tectu> not better at all
2012-10-25T06:02:41 < Tectu> GargantuaSauce, i must be something doing horribly wrong
2012-10-25T06:02:42 < Tectu>         zoom *= 0.1f;
2012-10-25T06:02:42 < Tectu>         if(zoom <= 0.00000000001f)
2012-10-25T06:02:42 < Tectu>             zoom = 1.0f;
2012-10-25T06:02:59 < Tectu> but very very fast
2012-10-25T06:03:17 < Tectu> and also only a few times
2012-10-25T06:03:25 < GargantuaSauce> so increase the 0.1 a bit
2012-10-25T06:03:34 < GargantuaSauce> say, 0.6
2012-10-25T06:04:28 < Tectu> i've been stupid today \o/
2012-10-25T06:04:49 < Tectu> GargantuaSauce, your code is way faster
2012-10-25T06:05:03 < Tectu> but I only get blue/black
2012-10-25T06:05:26 < GargantuaSauce> i assume you know how 16 bit colour works
2012-10-25T06:05:46 < GargantuaSauce> note that iterations only goes up to 64, 2^6
2012-10-25T06:07:15 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-25T06:07:55 < Tectu> okay, no sleep since 14 hours, I have to take a nap -.-
2012-10-25T06:08:02 < Tectu> thanks for your help, GargantuaSauce !
2012-10-25T06:09:00 < GargantuaSauce> yeah no problem
2012-10-25T06:09:10 < GargantuaSauce> fuck with the numbers until you understand what's going on once you're rested :P
2012-10-25T06:09:21 < Tectu> GargantuaSauce, will do!
2012-10-25T06:09:30 < Tectu> GargantuaSauce, can I add the code as a demo to my library?
2012-10-25T06:10:00 < Tectu> GargantuaSauce, but even with 512 iteration, your code is way faster
2012-10-25T06:10:21 < GargantuaSauce> it is basically a trivial/canonical implementation of a well known algorithm, so you don't reaaaally need to ask
2012-10-25T06:10:25 < GargantuaSauce> but thanks and go ahead :P
2012-10-25T06:11:28 < Tectu> thanks sir :)
2012-10-25T06:11:29 < Tectu> g'night
2012-10-25T06:14:00 < dongs> spam subject line: Develop your possibilities in bed
2012-10-25T06:14:32 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Read error: Operation timed out]
2012-10-25T06:41:38 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Ping timeout: 256 seconds]
2012-10-25T06:54:00 < emeb> dongs: more better? -> http://imagebin.org/233183
2012-10-25T06:56:27 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-25T06:57:04 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-25T06:57:10 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-25T07:06:54 < dongs> attn Laurenceb
2012-10-25T07:06:55 < dongs> PART #              QTY       STATUS        ANTICIPATED SHIP DATE
2012-10-25T07:06:55 < dongs> --------            -----     ------        ----------------------
2012-10-25T07:06:55 < dongs> FXOS8700CQR1        5         Scheduled     12-FEB-2013
2012-10-25T07:06:57 < dongs> lol
2012-10-25T07:13:09 < emeb> that's some backorder
2012-10-25T07:16:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-25T07:33:36 < ds2> I have seen worse from TI
2012-10-25T07:33:41 < ds2> 9 months ahead!
2012-10-25T07:33:53 < emeb> "don't call us, we'll call you"
2012-10-25T07:35:17 < emeb> ds2: how's your e-reader-ish thing coming along?
2012-10-25T07:36:26 < ds2> e-reader-ish?
2012-10-25T07:36:47 < ds2> beagle bone stuff?
2012-10-25T07:36:53 < emeb> the f103 board you're working on with the e-ink display
2012-10-25T07:37:09 < ds2> Ohhhh that... layout is done... looking at fab options
2012-10-25T07:37:20 < ds2> btw, anyone used goldphoenixpcb directly?
2012-10-25T07:37:24 < ds2> pricing looks attractive
2012-10-25T07:38:14 < emeb> nope - only ever gone thru Sparkfun.
2012-10-25T07:38:34 < ds2> sparkfun pools, right?
2012-10-25T07:38:46 < emeb> pools to goldphoenix IIRC
2012-10-25T07:38:56 < ds2> so they won't do the 5/5 rules w/10mil holes, IIRC
2012-10-25T07:39:07 < emeb> nope. 8/8 last I checked
2012-10-25T07:39:17 < ds2> goldphoenix directly will do 5/5
2012-10-25T07:39:22 < emeb> nice.
2012-10-25T07:39:34 < ds2> despite all the extra charges, they look to be 200-300 cheaper then advance pcb
2012-10-25T07:39:51 < emeb> big difference.
2012-10-25T07:39:55 < dongs> whats your spec?
2012-10-25T07:40:05 < dongs> i mean , what spec you need
2012-10-25T07:40:14 < dongs> 5/5 + 10mil holes?
2012-10-25T07:40:50 < ds2> yep
2012-10-25T07:40:55 < dongs> i can do that.
2012-10-25T07:41:01 < dongs> whats hte board size?
2012-10-25T07:41:04 < ds2> 1.8"x1.8"
2012-10-25T07:41:09 < ds2> dongs: you a fab house?
2012-10-25T07:41:12 < dongs> fyucking faggot yank units
2012-10-25T07:41:20 < ds2> one sec
2012-10-25T07:41:27 < dongs> 46x46mm
2012-10-25T07:41:31 < dongs> lemme see.
2012-10-25T07:41:39 < ds2> yeah, 46x46mm
2012-10-25T07:41:52 < dongs> 4 layer I guess?
2012-10-25T07:41:56 < ds2> yep
2012-10-25T07:42:17 < ds2> it is all cuz the PSRAM is BGA only and the F103 is BGA
2012-10-25T07:45:11 < baird> I haz an F3 \o/
2012-10-25T07:46:07 < ds2> baird: what board?
2012-10-25T07:47:11 < baird> The F3 Disco. The one Dongs isn't allowed to have..
2012-10-25T07:47:36 < dongs> ds2: no via in pad right?
2012-10-25T07:47:46 < dongs> dont need plugged/pad via?
2012-10-25T07:47:52 < emeb> I got some loose F373CCT6 parts today - does that count?
2012-10-25T07:48:11 < ds2> dong: yes, no via in pad. Tenting would be good otherwise, I can't read the silk screen
2012-10-25T07:48:20 < dongs> yeah thats default
2012-10-25T07:48:35 < dongs> i fucking hate fabs that mask out vias
2012-10-25T07:49:02 < ds2> never needed that til I this. usually my projects are bigger
2012-10-25T07:49:50 < ds2> just realize, I am just 1 gyro short of turning this into a quadcopter brain
2012-10-25T07:57:19 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-25T07:57:42 < ds2> blah...
2012-10-25T07:57:51 < dongs> ds2: looks like around 230 or so for ~10 boards
2012-10-25T07:57:53 < ds2> anyone know off hand what's the recommended load capacitance of the 32KHz xtal?
2012-10-25T07:58:09 < dongs> 32khz needs caps?
2012-10-25T07:58:27 < ds2> dongs: wow...
2012-10-25T07:58:31 < ds2> 32KHz, yep
2012-10-25T07:58:54 < ds2> I forgot to note the load capacitance so now there is 6 xtals on digikey that fit the foot print
2012-10-25T08:02:20 < emeb> probably depends on the mfg. Looks like the f4 discovery used 6.8pf though.
2012-10-25T08:02:44 < ds2> found it... AN2867 has a list
2012-10-25T08:07:56 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-25T08:20:16 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-25T08:26:12 -!- claude is now known as Claude
2012-10-25T08:27:02 < dongs> from work email: "We did tested with your code and feel very useful."
2012-10-25T08:27:54 < emeb> engrish!
2012-10-25T08:27:57 < jpa-> our sales person once got a reply "thank you for your corporation" and wondered what exactly did he sell..
2012-10-25T08:32:48 < dongs> teehee
2012-10-25T08:33:58 < emeb> dongs: made a fp for the thru-hole mini-usb. Looks like it's narrower than the SMD version.
2012-10-25T08:34:08 < dongs> of course it is.
2012-10-25T08:34:11 < emeb> saves a little real-estate.
2012-10-25T08:38:51 < dongs> there are zero disadvantages to it and all the advantages.
2012-10-25T08:46:58 < ds2> might make it a tiny bit more expensive
2012-10-25T08:51:24 < emeb> more drills?
2012-10-25T08:53:01 < emeb> updated render -> http://imagebin.org/233192
2012-10-25T08:53:11 < ds2> no, assembly - they will either need to mask out the SMT parts to run a wave (if you have stuff in the bottom) or they have to hand mount it
2012-10-25T08:53:36 < ds2> emeb: discovery jr?
2012-10-25T08:53:48 < emeb> ds2: heh - pretty much.
2012-10-25T08:55:50 < baird> Someone needs to produce a 'Duck' discovery board. -> Disco Duck. (and if dongs does it, Afro Duck)
2012-10-25T08:57:47 < emeb> here's what OSHpark thinks it'll look like: http://imagebin.org/233193
2012-10-25T08:59:24 < emeb> their next panel goes out on the 26th so I have some time to check if it's all correct.
2012-10-25T09:04:27 < dongs> haha, i linked that vcut cutter shit to a dude
2012-10-25T09:04:28 < dongs> http://serviceapi.nmv.naver.com/flash/convertIframeTag.nhn?vid=8CBB85FBDBE13DF62D3A5D817C8BE1AC2415&outKey=V128c1128e20e756f75d5613fc220ef5f5af5449bba6b4e44d341613fc220ef5f5af5&width=720&height=438
2012-10-25T09:04:33 < dongs> he made same thing
2012-10-25T09:07:00 < dongs> emeb: the pad spacing on caps/Rs is terrible
2012-10-25T09:07:03 < dongs> on that oshpark pic
2012-10-25T09:07:32 < emeb> dongs: it looks bad - I think it's cause of the way they render the solder mask.
2012-10-25T09:08:46 < emeb> go back an look at the green render and it looks better
2012-10-25T09:08:49 < dongs> right
2012-10-25T09:11:01 < dongs> do i need an account
2012-10-25T09:11:02 < dongs> to use that hsit?
2012-10-25T09:11:03 < dongs> shit
2012-10-25T09:11:06 < dongs> to see preview
2012-10-25T09:11:24 < emeb> which?
2012-10-25T09:11:31 < dongs> oshpark gerber preview
2012-10-25T09:12:11 < emeb> I think you can upload w/o having an account.
2012-10-25T09:12:39 < emeb> http://oshpark.com/ <- should be the 1st page of their DRC upload process.
2012-10-25T09:14:39 < dongs> processing file
2012-10-25T09:14:42 < dongs> does render show up there?
2012-10-25T09:15:05 < emeb> after it passes their checks it should give you a page with all the layers and the combined render.
2012-10-25T09:15:11 < dongs> ok
2012-10-25T09:15:32 < dongs> how did you make it zoomed?
2012-10-25T09:15:37 < dongs> mine is rendered 1:1
2012-10-25T09:15:52 < dongs> ah next page
2012-10-25T09:16:42 < dongs> fuck, that looks great
2012-10-25T09:16:49 < dongs> http://i.imgur.com/g70hU.png http://i.imgur.com/E9gcX.png
2012-10-25T09:17:03 < dongs> except for oval usb
2012-10-25T09:17:19 < emeb> all purple and pink!
2012-10-25T09:17:29 < emeb> it did render that slot kinda funny.
2012-10-25T09:21:33 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-wvvpmxdkmnmqirvg] has quit [Ping timeout: 245 seconds]
2012-10-25T09:22:05 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-bmnjepctiroazkki] has joined ##stm32
2012-10-25T09:24:59 < emeb> Hmmm... GPIO pin names on the back silk seems like a good idea...
2012-10-25T09:25:17 < emeb> esp. since I don't have anything on the back silk right now.
2012-10-25T09:28:47  * emeb -> emeb_mac
2012-10-25T09:28:51 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-25T09:34:51 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-25T09:37:22 < ds2> Hmmmm
2012-10-25T09:37:36 < ds2> red LEDs have a higher MCD rating
2012-10-25T09:40:36 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Ping timeout: 240 seconds]
2012-10-25T09:41:13 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-25T09:41:13 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-25T09:41:13 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-25T09:46:24 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 265 seconds]
2012-10-25T09:47:35 < emeb_mac> red is generally more efficient.
2012-10-25T10:22:14 < dongs> The fab will automatically remove any silkscreen that crosses drilled holes or exposed metal.
2012-10-25T10:22:17 < dongs> gay
2012-10-25T10:22:18 < dongs> ^ oshpark
2012-10-25T10:28:46 < zyp> that's why my awesome drawing turned non-awesome on my F4 board :(
2012-10-25T10:29:27 < zyp> even though the vias it crossed was tented
2012-10-25T10:29:33 < dongs> you shoulda have got it done wiht dongs special trollpcb service
2012-10-25T10:30:07 < dongs> well, you'd have to pay >$5 for it though.
2012-10-25T10:31:24 < zyp> I paid $45 for three of those, shipped
2012-10-25T10:31:29 < zyp> 50x50mm, four layers
2012-10-25T10:31:42 < dongs> my minimum is like 10
2012-10-25T10:31:51 < dongs> the price difference for 3 vs 10 isnt significant enough
2012-10-25T10:32:02 < zyp> that's not a problem
2012-10-25T10:34:08 < zyp> I figure that the lpc4300 stuff will end up around those measurements too, maybe a bit larger to have room for connectors
2012-10-25T10:35:20 < dongs> you should just clone TitanMDKs lpc design
2012-10-25T10:35:26 < dongs> and get t hem done @ iteadstudio
2012-10-25T10:35:46 < zyp> he has broken out like nothing :p
2012-10-25T10:35:52 < dongs> yea
2012-10-25T10:37:36 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-25T10:38:11 < zyp> anyway, how much do you figure a 50x50mm four layer design will cost to do?
2012-10-25T10:38:48 -!- Claude is now known as claude
2012-10-25T10:39:30 < dongs> ds2 was asking for similar
2012-10-25T10:39:36 < dongs> his shit was 46x46 4 layer 5/5mil + 10mil vias
2012-10-25T10:39:39 < dongs> that was aroudn 230+shipping
2012-10-25T10:40:13 < zyp> I'm designing for 6/6mil 10/20 vias
2012-10-25T10:40:23 < dongs> yeah thats stanadrd spec.
2012-10-25T10:40:25 < dongs> should be cheaper.
2012-10-25T10:40:41 < zyp> and since it's just a toy project, I don't mind waiting until you got other shit to panelize it with
2012-10-25T10:40:47 < dongs> right
2012-10-25T10:46:46 < ds2> 6/6? geez that is huge :D
2012-10-25T10:48:28 < zyp> sure, it's just 1mm pitch lbga :p
2012-10-25T10:48:53 < ds2> Ohhhh no wonder
2012-10-25T10:49:10 < ds2> zyp: how do you plan to assemble that?
2012-10-25T10:49:24 < zyp> reflow oven
2012-10-25T10:50:34 < ds2> u got inspection facilities?
2012-10-25T10:51:32 < zyp> nah, I'll just solder it, power it up and see if it smokes
2012-10-25T10:51:39 < dongs> tap tap, does it fall off? no? ok, we're good
2012-10-25T10:51:43 < zyp> :)
2012-10-25T10:52:33 < zyp> being 1mm pitch, I don't really expect much problems
2012-10-25T10:52:39 -!- claude is now known as claude_work
2012-10-25T10:52:39 < dongs> right
2012-10-25T10:52:51 -!- claude_work is now known as Claude_work
2012-10-25T10:55:40 < zyp> anyway, the whole point of the project is to see just how hard it is to do bga, that's why I'm designing it cheap so it's not a huge waste if/when I fuck up
2012-10-25T10:56:11 < zyp> but come on, if TitanMKD can do it with a hotplate, it surely can't be that hard
2012-10-25T10:56:31 < dongs> with hotplate on shitty fucking iteastdudio boards
2012-10-25T10:56:34 < dongs> with half hte drills wrong
2012-10-25T11:11:02 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-25T12:42:03 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Read error: Operation timed out]
2012-10-25T12:45:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-25T12:54:47 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-25T13:12:06 -!- BjoernC [~BjoernC@guest124.ibr.cs.tu-bs.de] has joined ##stm32
2012-10-25T13:15:42 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Quit: Leaving.]
2012-10-25T13:16:28 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-25T13:23:28 -!- BjoernC [~BjoernC@guest124.ibr.cs.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-25T13:30:11 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-25T13:30:40 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-25T14:08:41 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 244 seconds]
2012-10-25T14:10:23 < Tectu> GargantuaSauce, https://www.youtube.com/watch?v=P1ihO3br15Q
2012-10-25T14:31:23 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Quit: ZNC - http://znc.in]
2012-10-25T14:32:05 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-25T14:34:05 < Tectu> have you guys seen this?   http://www.youtube.com/watch?v=rg9sCIHA6DM
2012-10-25T14:36:18 < zyp> what about it?
2012-10-25T14:36:56 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 244 seconds]
2012-10-25T14:39:04 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-25T14:44:19 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-25T14:48:48 < dongs> ohhh
2012-10-25T14:48:51 < dongs> bike project might be suceeding
2012-10-25T14:48:59 < dongs> http://tncscooters.com/index.php?route=product/product&product_id=168
2012-10-25T14:49:03 < dongs>  i found the motor controller
2012-10-25T14:49:09 < dongs> and the motor, in case i need a new one
2012-10-25T14:49:20 < dongs> i took it apart and oiled everything today, mechanically it seems OK
2012-10-25T14:49:29 < Tectu> oh dear
2012-10-25T14:49:31 < BrainDamage> hub motor?
2012-10-25T14:49:44 < dongs> no, motor is http://tncscooters.com/index.php?route=product/product&path=52_55&product_id=51
2012-10-25T14:49:51 < dongs> now i need to measure + figure out what belt i need
2012-10-25T14:49:54 < Tectu> 32$  ? wtf?
2012-10-25T14:49:58 < dongs> its that weird flat vbelt wihtout the V
2012-10-25T14:50:14 < dongs> but with teeth on the inside
2012-10-25T14:50:23 < dongs> Tectu: wat?
2012-10-25T14:50:41 < BrainDamage> electric motors have stupidly high powers for cheap
2012-10-25T14:50:43 < Tectu> what are you doing doings
2012-10-25T14:51:09 < dongs> tectu, http://i.imgur.com/MyPB1.jpg fixing this bike
2012-10-25T14:51:48 < BrainDamage> Li battery or Pb ?
2012-10-25T14:51:48 < Tectu> holy-wtf-crap
2012-10-25T14:53:09 < dongs> pb. 2 12V slas
2012-10-25T14:53:21 < dongs> once it works and i feel like dickign wiht it i might do a brushless conversion
2012-10-25T14:53:21 < dongs> heh
2012-10-25T14:53:50 < dongs> http://i.imgur.com/QrrFc.jpg shit's a little rusted
2012-10-25T14:54:36 < Tectu> what's slas?
2012-10-25T14:54:53 < dongs> SLA battery.
2012-10-25T14:55:10 < dongs> those 7Ah are common shits
2012-10-25T14:55:12 < dongs> in UPS etc.
2012-10-25T14:55:25 < dongs> http://www.amazon.com/Yuasa-NP7-12-12V-7AH-Battery/dp/B002MA6FYG
2012-10-25T14:56:42 < Tectu> ah
2012-10-25T14:56:48 < Tectu> yeah have two of those
2012-10-25T14:57:04 < Tectu> my UPS can drive my backupserver for 62minutes
2012-10-25T14:57:07 < BrainDamage> sla = sealed lead acid
2012-10-25T14:57:31 < dongs> righto.
2012-10-25T14:57:34 < dongs> hmm
2012-10-25T14:57:39 < dongs> well i feel better this controlelr is known.
2012-10-25T14:57:47 < dongs> i could at least get a replacement for 15bucks.
2012-10-25T14:57:47 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-25T14:59:47 < BrainDamage> ugh @ inverted fork bolt
2012-10-25T15:00:01 < BrainDamage> ah wait, it's lateral
2012-10-25T15:00:04 < BrainDamage> nvm
2012-10-25T15:00:54 < dongs> inverted what
2012-10-25T15:01:29 < BrainDamage> I tought the frame hole slot mounts the wheel was pointing up
2012-10-25T15:01:37 < BrainDamage> instead of sideways
2012-10-25T15:01:51 < dongs> no. the wheel is loose, cause i took it out to clean the brakes/etc parts
2012-10-25T15:01:56 < dongs> its just regular bicycle style
2012-10-25T15:02:04 < dongs> a U-slot and a thing to keep it in there
2012-10-25T15:02:46 < dongs> the tube fillup thing is really fucked up, its behind the one-way bearing thing
2012-10-25T15:02:51 < dongs> and theres just no way to get a compressor in there..
2012-10-25T15:03:09 < dongs> trying to find who the fuck makes this thing, theres some chink shit written on the brake handle but I dont think thats the maker
2012-10-25T15:03:35 < dongs> no company or model name or anything anywehre else, the only thing i found was a label on that controller
2012-10-25T15:05:44 < BrainDamage> I wouldn't even know where to start, maybe some local minibike racing organization?
2012-10-25T15:05:55 < dongs> im sure its  chinq shit
2012-10-25T15:06:03 < dongs> http://ep.yimg.com/ca/I/scootercatalog_2234_10240374 looks similar design
2012-10-25T15:06:07 < dongs> that one has some hardcore brakes though
2012-10-25T15:06:16 < dongs> mine is brake inside the rear wheel only
2012-10-25T15:06:18 < dongs> nothing in front
2012-10-25T15:06:52 < Tectu> well, i've an SPI slave here having DIN and OUT instead of MOSI and MISO... does that matter?
2012-10-25T15:07:02 < dongs> of course not
2012-10-25T15:07:05 < BrainDamage> that one also have a frame label
2012-10-25T15:07:10 < dongs> yeah that one does
2012-10-25T15:07:13 < dongs> mine does not :D
2012-10-25T15:07:20 < BrainDamage> stolen
2012-10-25T15:07:32 < Tectu> well is DIN MOSI and OUT MISO I guess?
2012-10-25T15:08:30 < dongs> maybe anoher way around
2012-10-25T15:08:33 < dongs> hook it up and swap if its wrong
2012-10-25T15:09:44 < dongs> blah, shengdi isnt the maker.
2012-10-25T15:11:08 < dongs> http://ep.yimg.com/ca/I/scootercatalog_2229_15720329 hmmm
2012-10-25T15:11:10 < dongs> that looks very similar
2012-10-25T15:11:14 < dongs> hmm no
2012-10-25T15:14:23 < dongs> man so much china shit
2012-10-25T15:19:55 -!- dekar [~dekar@212.255.27.61] has quit [Quit: This computer has gone to sleep]
2012-10-25T15:28:19 -!- Devilholk [~banned@h-201-147.a357.priv.bahnhof.se] has quit [Ping timeout: 260 seconds]
2012-10-25T15:31:16 < badger> anyone know of an external ram breakout module I could use with FSMC? looking for about 512KB module
2012-10-25T15:31:43 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-25T15:32:20 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-25T15:32:23 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-25T15:33:01 < zyp> ram breakout?
2012-10-25T15:36:53 < badger> a breakout board with some form of ram chip
2012-10-25T15:37:44 < dongs> i guess sram/psram
2012-10-25T15:37:57 < badger> yes that is what i was thinking
2012-10-25T15:38:09 < dongs> we were gonna make a breakout for those bga psram chips
2012-10-25T15:38:14 < badger> but trying to find a pre-made module I can experiment with is proving difficult!
2012-10-25T15:38:15 < dongs> but i got lazy.
2012-10-25T15:38:18 < dongs> yes.
2012-10-25T15:38:33 < dongs> 512k is not so bad, im sure you can find some sram in tssop for that
2012-10-25T15:38:42 < dongs> but like 1-8meg stuff, is annoying
2012-10-25T15:39:00 < dongs> lets se e if waveshare has somehting
2012-10-25T15:40:10 < dongs> http://www.wvshare.com/product/IS62WV12816BLL-SRAM-Board.htm
2012-10-25T15:40:11 < dongs> here you go
2012-10-25T15:40:31 < dongs> http://www.wvshare.com/product/IS62WV51216BLL-SRAM-Board.htm
2012-10-25T15:40:32 < dongs> 8mbit
2012-10-25T15:41:00 < badger> dongs: great! thank you
2012-10-25T15:41:37 < dongs> yep.
2012-10-25T15:41:43 < dongs> i actually have teh 2mbit version i think
2012-10-25T15:41:46 < dongs> came with one of their kits.
2012-10-25T15:46:00 < zyp> I was about to suggest waveshare, but I got interrupted
2012-10-25T15:46:23 < dongs> hehs
2012-10-25T15:48:26 <+Steffanx> zyp also has stocks/shares in WS now..
2012-10-25T15:50:01 < Laurenceb> sup
2012-10-25T15:50:09 <+Steffanx> Tha sky bro
2012-10-25T15:50:30  * Laurenceb is working on a magno calibration EKF in matlab
2012-10-25T15:50:38 < Laurenceb> seems the idea is workable
2012-10-25T15:50:53 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal]
2012-10-25T15:50:56 < Laurenceb> i might code it in c and get it running on the f3discovery tomorrow or something
2012-10-25T16:26:09 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-25T16:26:55 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Excess Flood]
2012-10-25T16:28:31 < Tectu> we should create a blog
2012-10-25T16:28:40 < Tectu> about the differences of the qualities of whiteboard markers
2012-10-25T16:30:16 <+Steffanx> We?
2012-10-25T16:32:06 < dongs> i think Tectu is talkin about him and his kawaii friends
2012-10-25T16:34:03 <+Steffanx> Are you kawaii ?
2012-10-25T16:35:25 < baird> uguu~
2012-10-25T16:40:44 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-25T16:48:11 < Tectu> dongs, is the master of kawaiiness
2012-10-25T16:56:16 < Laurenceb> CMUCam v4. This is an Arduino shield with a small camera mounted on it designed to allow development of computer vision applications. On itÂ’s own, an Arduino would struggle with the number crunching needed for motion and colour detection, but this board has an on-board processor to do the heavy liftin
2012-10-25T16:56:22 < Laurenceb> so much fail
2012-10-25T16:57:26 < Claude_work> more fail ? http://www.maxwell.com/products/microelectronics/docs/hsn1000_rev3.pdf
2012-10-25T16:59:26 < Laurenceb> haha
2012-10-25T16:59:28 < Laurenceb> pmsl
2012-10-25T16:59:43 < Laurenceb> ive seen them in keyrings
2012-10-25T17:00:34 < Laurenceb> The active low Nuclear Event Detection signal (NED) is
2012-10-25T17:00:34 < Laurenceb> used to initiate a wide variety of circumvention functions, thus
2012-10-25T17:00:34 < Laurenceb> preventing upset and burnout of electronic components.
2012-10-25T17:00:40 < Laurenceb> lol
2012-10-25T17:01:35 < Claude_work> hehe : engage lead melting pot , spill lead over circuit , circuit protected from nuclear explosion ?
2012-10-25T17:02:12 < Laurenceb> https://dl.dropbox.com/u/63720513/Images/pico.jpg
2012-10-25T17:03:17 < Claude_work> nice , but the CPU is missing!
2012-10-25T17:03:36 < zyp> Claude_work, it would draw too much power, better leave it on the ground
2012-10-25T17:03:45 < Claude_work> ahhh clever!
2012-10-25T17:03:54 < Claude_work> precompute all
2012-10-25T17:10:12 < Laurenceb> http://hackaday.com/2012/10/24/launching-a-balloon-and-not-landing-in-the-ocean/#comments
2012-10-25T17:10:23 < Laurenceb>  "I think the word “not” needs to be removed from the title of this article."
2012-10-25T17:12:19 < baird> hackaday's editors are shittier than slashdot's..
2012-10-25T17:13:13 < baird> Although I'd say it's more the case of "WE NEEDS 500,000 VISITS A DAY, MAN! DUMP ANY SHIT YOU CAN FIND!"
2012-10-25T17:14:30 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-25T17:14:33 < Claude_work> like arduinos and howtos for bios flashing with arduinos?
2012-10-25T17:14:51 < zyp> it's called catering to the masses
2012-10-25T17:16:40 < baird> catering for the AdSense..
2012-10-25T17:20:24 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Ping timeout: 260 seconds]
2012-10-25T17:22:52 < baird> http://hackaday.com/2012/10/24/kim-kardashian-circuit-bender/
2012-10-25T17:23:14 <+Steffanx> Oh noes! I cannot find your stuff! ..
2012-10-25T17:24:19 < dongs> how does  NSS work in SPI slave mode
2012-10-25T17:24:28 < dongs> is tehre a mode where it just works as soon as it gets clock?
2012-10-25T17:24:39 < dongs> is thajt the nss_soft shit?
2012-10-25T17:24:45 < zyp> probably
2012-10-25T17:25:24 < zyp> the problem with «it just works» is «where the fuck is the boundary between bytes»?
2012-10-25T17:25:55 < dongs> not using it for that
2012-10-25T17:26:08 < dongs> im trolling SPI into arbitrary prescaler by clocking it from a TIMx
2012-10-25T17:26:14 < dongs> and making it output data
2012-10-25T17:26:18 < dongs> (while ignoring input)
2012-10-25T17:27:56 < zyp> ah, right
2012-10-25T17:29:43 < dongs> can i still dump shit out of MOSI in slave mode?
2012-10-25T17:30:02 < dongs> i remember i fucked shit up on F4 and miso/mosi got swapped
2012-10-25T17:30:13 < Claude_work> dongs , sounds like generating video with the spi unit ?
2012-10-25T17:30:17 < dongs> yeah.
2012-10-25T17:30:26 < dongs> i did it using SPI c lock / master, but that was fail
2012-10-25T17:30:37 < dongs> cause only ^2 prescalers.
2012-10-25T17:31:32 < Claude_work> hmmm doing/trying to do something similar with the carambola (http://www.8devices.com/product/3/carambola) i2s port atm
2012-10-25T17:32:19 < Claude_work> feeding a 'video friendly' external clock into the i2s shift register
2012-10-25T17:36:25 < Claude_work> dongs , so you want to use NSS/CS as sync then? if not , at least on other micros you have to pull down NSS for spi slave to work
2012-10-25T17:37:04 < zyp> going with soft slave selection should do the trick
2012-10-25T17:37:17 < zyp> just make it believe it's always a slave
2012-10-25T17:37:30 < dongs> right
2012-10-25T17:37:37 < dongs> im worried i cant output shit  over MOSI in slave mode
2012-10-25T17:38:12 < zyp> maybe in bidirectional mode
2012-10-25T17:38:17 < baird> It's always "The Internet of Things", never "the UUCP of things", or "the Zmodem of things"...
2012-10-25T17:38:19 < dongs> thats what im loking at
2012-10-25T17:38:29 < zyp> bidirectional mode only uses one of the lines, but I don't remember which
2012-10-25T17:38:34 < dongs> In this configuration the MOSI pin is a data input and the MISO pin is a data output.
2012-10-25T17:38:38 < dongs> standard slave. lame.
2012-10-25T17:39:28 < dongs> Transmit-only mode is similar to full-duplex mode (BIDIMODE=0, RXONLY=0): the data
2012-10-25T17:39:31 < dongs> are transmitted on the transmit pin (MOSI in master mode or MISO in slave mode) and
2012-10-25T17:39:34 < dongs> the receive pin (MISO in master mode or MOSI in slave mode) can be used as a
2012-10-25T17:39:37 < dongs> general-purpose IO.
2012-10-25T17:41:22 < dongs> fuck
2012-10-25T17:41:25 < dongs> looks like no way huh
2012-10-25T17:41:32 < dongs> was this different on F4?
2012-10-25T17:54:56 < zyp> likely not
2012-10-25T18:02:10 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-25T18:02:35 -!- Claude_work is now known as claude
2012-10-25T18:04:48 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-25T18:29:41 < dongs> awwm, attn cjbeard etc http://www.phoronix.com/scan.php?page=news_item&px=MTIxNDk
2012-10-25T18:30:31 < Laurenceb> http://1.bp.blogspot.com/-XIMhAAFCPJo/T3pOQeyrVSI/AAAAAAAADaY/iE8PPuOCB3c/s1600/epic-fail-reading-material-fail.jpg
2012-10-25T18:30:45 < Laurenceb> loldongs
2012-10-25T18:34:03 < baird> Reminds of when friends of a brother broke into a "Life Education Centre" (Drugs are Baaaad, m'kay) and made off with the Giraffe mascot costume.. and a few weeks later, photos started circulating of the mascot doing bongs and buckets. :D
2012-10-25T18:34:41 < baird> The LEC didn't use that mascot again..
2012-10-25T18:47:20 < zyp> dongs, that's what everybody has been saying since it was announced
2012-10-25T18:48:35 < dongs> zyp, i dont actaully care. i figured some assburgerberrypi rage would be nice to read
2012-10-25T18:49:10 < zyp> I don't care either, I've just read it already :)
2012-10-25T18:49:11 < Laurenceb> lol
2012-10-25T18:49:18 < baird> There isn't anything really USB related in the code either, which makes it useless for the non-Linux ports.
2012-10-25T18:49:19 < Laurenceb>  assburgerberrypi
2012-10-25T18:49:38 < zyp> baird, what do you mean?
2012-10-25T18:49:40 < dongs> does anyone without assburgers own a Pi?
2012-10-25T18:49:45 < baird> assberrypi works better
2012-10-25T18:50:16 < BrainDamage> I'm still waiting for min to ship, I ordered in june
2012-10-25T18:50:18 < BrainDamage> fucking rs
2012-10-25T18:50:23 < dongs> or aspieberrypi
2012-10-25T18:50:29 < baird> zyp: The big hold-up with the non-Linux ports (Plan9, NetBSD) is getting to the USB system, and the just-released code has nothing of that in it, just graphics.
2012-10-25T18:50:49 < dongs> closed sores USB, how fucking wonderful
2012-10-25T18:50:53 < dongs> ok bedtime
2012-10-25T18:51:12 < BrainDamage> baird: the just released code isn't even a real driver, it's just a bunch of header wrappers
2012-10-25T18:51:27 < zyp> the usb is a standard dwc_otg core and competely unrelated to video stuff
2012-10-25T18:54:00 < baird> a 'shim' as the dude inthe phronix post called it
2012-10-25T18:54:27 < zyp> sure, but what is a driver really?
2012-10-25T18:56:43 < zyp> I'm not intending to defend anyone here, but I would call the released code a thin driver
2012-10-25T18:58:16 -!- barthess [~barthess@5.100.207.77] has joined ##stm32
2012-10-25T18:58:19 < zyp> it's everything that is required on the host side to use the videocore, so I would consider it a driver, but since it's just an RPC client translating function calls into messages passed to the videocore firmware, it can be made quite thin
2012-10-25T18:59:22 < zyp> it's not a complete graphics stack, but it's the driver required to interface with the videocore
2012-10-25T19:00:18 < zyp> and since people are talking about porting to other operating systems, isn't it exactly what is required?
2012-10-25T19:01:30 < zyp> not being able to modify the entire graphics stack is a completely different issue from not being able to interface with it
2012-10-25T19:02:05 < emeb> zyp: this is true. The main agument against it I've seen is that w/o access to the video core source there's no way for the community to deal with bugs & extensions.
2012-10-25T19:02:47 < emeb> so they end up as tenants on Broadcom's farm as it were.
2012-10-25T19:03:53 < emeb> although how that differs from just using a locked-down hardware design is beyond me.
2012-10-25T19:05:12 < emeb> reminds me a bit of the qi-hardware folks who won't use any processor architecture that they can't compile in an FPGA.
2012-10-25T19:06:00 <+Steffanx> An quad-core i7 inside an FPGA, :D :D
2012-10-25T19:06:04 <+Steffanx> -n
2012-10-25T19:06:28 < zyp> sure, it just wouldn't be very fast :p
2012-10-25T19:08:02 < emeb> like the guy who emulated an ARM with an AVR?
2012-10-25T19:08:23 < zyp> something like that :p
2012-10-25T19:09:29 <+Steffanx> Best project EVER
2012-10-25T19:10:45 < zyp> well, it becomes sort of a microcoded ARM, using the AVR as a microcode interpreter, I guess
2012-10-25T19:10:48 < zyp> :p
2012-10-25T19:11:24 < zyp> microcode is a bit interesting actually
2012-10-25T19:19:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-25T19:38:31 < emeb> next up - an AVR emulating an FPGA running a quad-core i7.
2012-10-25T19:40:43 -!- claude is now known as Claude
2012-10-25T19:41:53 < baird> ..running FORTH
2012-10-25T19:42:05 < BrainDamage> MUMPS
2012-10-25T19:42:31 < BrainDamage> http://thedailywtf.com/Articles/A_Case_of_the_MUMPS.aspx
2012-10-25T19:43:34 < baird> I've got a copy of it for the CP/M system..
2012-10-25T20:09:21 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-25T20:13:10 < zyp> the flash in L1 is a bit odd
2012-10-25T20:13:53 < zyp> it's the first flash I've seen that reads 0 when unprogrammed and 1 when programmed
2012-10-25T20:14:30 < karlp> I think they even mention that explicitly in the flash programming manual don't they?
2012-10-25T20:14:56 < zyp> that might be, I haven't read it from start to end
2012-10-25T20:27:41 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-25T20:39:35 < TitanMKD> hello
2012-10-25T20:41:50 <+Steffanx> Salut!
2012-10-25T20:42:06 <+Steffanx> Ça va? :)
2012-10-25T20:42:35 < emeb> moshi moshi
2012-10-25T20:44:55 <+Steffanx> Whatever that means.. my japanese isn't very good
2012-10-25T20:46:13 < emeb> IIRC it's how many folks in JP answer the telephone. Sort of like "hello". Dongs will probably correct me...
2012-10-25T20:46:41 <+Steffanx> Dongs sings songs in dongsland atm
2012-10-25T21:08:52 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 265 seconds]
2012-10-25T21:32:46 < zyp> does anyone happen to have numbers on texane flashing speed on L1?
2012-10-25T21:36:02 < zyp> or any other flashing tool for that matter
2012-10-25T21:36:49 < zyp> I'm currently getting 5 kB/s with my BMP implementation, is the flash really that slow?
2012-10-25T21:40:14 < zyp> also, I had some weird issues with the BMP losing connection with the target now and then
2012-10-25T21:40:34 < zyp> turns out I had forgot to remove the jumpers connecting the onboard stlink, so it was interfering
2012-10-25T21:40:50 <+Steffanx> heh
2012-10-25T21:49:22 < karlp> 5kb's ok.
2012-10-25T21:50:51 < karlp> I probably only get 2kb or so wtih stlink here?
2012-10-25T21:51:05 < karlp> but this has some usb issues I believe, I get 3kb on f100, and I get 7kb at work.
2012-10-25T21:51:09 < zyp> ok is subjective, I want to know what I should expect
2012-10-25T21:51:25 < zyp> how large image? I'm testing with around 20k now
2012-10-25T21:51:50 < karlp> that was only an 8k image, and that was using time st-flash.
2012-10-25T21:52:00 < karlp> I don't have any big images lying around handy
2012-10-25T21:52:05 < zyp> hmm, I sped it up a bit, I get seven now
2012-10-25T21:52:18 < zyp> and this is going by gdb's measurement
2012-10-25T21:52:29 < karlp> how much of bmp can just be used with the stlink hardware as is?
2012-10-25T21:52:38 < karlp> it supports ft2232 hardware doesn't it?
2012-10-25T21:53:06 < zyp> unsure, I've been thinking that thought myself
2012-10-25T21:53:15 < zyp> could be an interesting project
2012-10-25T21:53:16 < karlp> gdb for me here gives 1k/sec,
2012-10-25T21:53:29 < karlp> but part of that's becaus eof gdb's insane way of asking each page to be done separately.
2012-10-25T21:53:40 < karlp> it's way slower than using flash externally
2012-10-25T21:54:24 < zyp> IIRC texane stuff has some slowness in the gdbserver
2012-10-25T21:54:49  * karlp shrugs
2012-10-25T21:54:59 < karlp> gdbserver is th ebit I have looked at the least,
2012-10-25T21:55:08 < karlp> and my images are still small enough it's not a hgue deal for me.
2012-10-25T21:55:21 < zyp> I seem to recall it using 10s or so from I typed load in gdb until it started actually sending stuff on usb when I was testing the flashloader I wrote with 1M images
2012-10-25T21:55:54 < zyp> so I suspect it's transforming the data between hex and binary or something in a very inefficient manner
2012-10-25T21:56:43 < karlp> yeah, it writes each piece one at a time, but last I looked gdb was _asking_ the gdbserver to do that.
2012-10-25T21:56:43 < zyp> I also believe texane gdbserver is buffering up the entire image from gdb before flashing, so gdb transfer size shouldn't matter
2012-10-25T21:56:59 < karlp> you can see it reloads the flash loader writes a single page then returns dozens of times over
2012-10-25T21:57:03 < karlp> it's clearly stupid,
2012-10-25T21:57:18 < karlp> but the one time I looked, gdb was sending those commands, not the texane gdbserver choosing to do so.
2012-10-25T21:57:41 < karlp> but that may hav e been in reponse to some "capabilities" that the gdb server was sending.
2012-10-25T21:58:47 < zyp> I don't know and I'm not interested in looking at it now either
2012-10-25T21:59:07 < zyp> but I believe texane gdbserver could be seriously improved wrt to speed
2012-10-25T21:59:13 < zyp> -to
2012-10-25T21:59:21 < karlp> yeah, if you're getting >5k/sec, you're off to a good start.
2012-10-25T21:59:21 -!- Claude is now known as claude
2012-10-25T21:59:53 < karlp> adding stlink hardware support would be awesome :)
2012-10-25T21:59:53 < zyp> I'm up to seven now, I'll see if I can push out some more
2012-10-25T22:00:09 < zyp> oh, and this is without flashloader
2012-10-25T22:00:25 < karlp> yeah, the half page flashing on l1 means it shouldn't need a flash loader
2012-10-25T22:00:28 < zyp> there doesn't seem to be any point in using a flashloader on L1
2012-10-25T22:00:37 < zyp> well
2012-10-25T22:00:55 < zyp> even in single word flashing it's stalling the bus while busy, so there is no need to check the busy flasg
2012-10-25T22:03:45 <+Steffanx> "adding stlink hardware support would be awesome" afaik there is already support for that
2012-10-25T22:03:58 < zyp> no
2012-10-25T22:04:03 < zyp> you're misunderstanding
2012-10-25T22:04:13 <+Steffanx> I guess so :P
2012-10-25T22:04:19 < zyp> he's talking about running it as middleware against stlink firmware
2012-10-25T22:04:25 < zyp> like how texane works
2012-10-25T22:04:35 <+Steffanx> Ah, like that :)
2012-10-25T22:05:18 < jpa-> karlp: i get about 9kB/s with stlink v2 and STM32F4
2012-10-25T22:06:11 < zyp> jpa-, must be a small image, when I tested F4 with full flash I got 30-40 or something
2012-10-25T22:06:20 < jpa-> ok
2012-10-25T22:07:18 < jpa-> i think 5kB/s is about the same as i get with L1 and stlink
2012-10-25T22:07:52 < TitanMKD> why only 5KB/s ?
2012-10-25T22:08:21 < TitanMKD> the internal flash is so slow to program even using big page ?
2012-10-25T22:08:25 < zyp> presumably because L1 flash is slow
2012-10-25T22:08:43 < TitanMKD> but 9kb/s on STM32F4 is also very slow
2012-10-25T22:09:20 < TitanMKD> IIRC it's a bit faster with SWD and Attolic TrueStudio
2012-10-25T22:09:23 < zyp> sure, but that's usually attributed to measurement inaccuracy of a small image
2012-10-25T22:09:41 < jpa-> yeah, the pages at the beginning of STM32F4 flash are smaller than the rest
2012-10-25T22:09:50 < jpa-> btw, it's a 170kB image
2012-10-25T22:10:12 < zyp> ouch, then it sounds a bit slow
2012-10-25T22:10:34 < jpa-> ubuntu under virtualbox under windows, i guess that may add some overhead
2012-10-25T22:10:44 < jpa-> or then it just plain simple is slow
2012-10-25T22:12:12 < TitanMKD> the only bottleneck should be the internal flash
2012-10-25T22:12:36 < TitanMKD> else USB is 1MB/s in best case or 100KB/s in worst i will say especially for USB 1.1 FS
2012-10-25T22:15:17 < zyp> you forget to consider the speed of the swd connection
2012-10-25T22:19:11 < TitanMKD> yes really true ;)
2012-10-25T22:19:25 < TitanMKD> 1 or 2MHz IIRC
2012-10-25T22:20:56 < zyp> hmm, looks like 7k is the best I can do, even with full flash
2012-10-25T22:29:09 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-25T22:37:22 -!- piele is now known as piele_gone
2012-10-25T22:40:17 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Read error: Connection reset by peer]
2012-10-25T22:50:10 -!- mrcan [~mrcan@unaffiliated/mrcan] has joined ##stm32
2012-10-25T22:51:10 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-25T22:53:46 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-25T23:06:16 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-25T23:10:15 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-25T23:13:37 -!- dekar [~dekar@212.255.27.61] has joined ##stm32
2012-10-25T23:13:39 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-25T23:47:00 < Laurenceb_> https://raw.github.com/mossmann/hackrf/master/doc/jawbreaker.jpeg
2012-10-25T23:47:04 < Laurenceb_> that looks epic
2012-10-25T23:47:22 < Laurenceb_> is it possible to do sgpio with an stm32?
2012-10-25T23:47:42 < Laurenceb_> using timer input -> dma trigger and dma from gpio?
2012-10-25T23:52:14 < zyp> it's not as deterministic, as latency depends on bus contention
2012-10-25T23:53:47 < Laurenceb_> ah yeah of course
2012-10-25T23:54:22 < Laurenceb_> having said that, for clked parallel adc data like that it wouldnt matter
2012-10-25T23:54:28 < Laurenceb_> looks like F2 has sgpio
2012-10-25T23:55:02 < zyp> isn't sgpio an lpc thing?
2012-10-25T23:57:29 < Laurenceb_> wait no it doesnt
2012-10-25T23:57:37 < Laurenceb_> just eading for forumz
--- Day changed Fri Oct 26 2012
2012-10-26T00:00:20 < Laurenceb_> they were getting confused about camera interface
2012-10-26T00:00:25 < Laurenceb_> but thats only 14bit
2012-10-26T00:00:31 < Laurenceb_> that sucks
2012-10-26T00:01:26 < Laurenceb_> looks like hackrf uses 16bit parrallel
2012-10-26T00:02:23 < TitanMKD> Laurenceb hackrf use 8bits parallel
2012-10-26T00:04:36 -!- barthess [~barthess@5.100.207.77] has quit [Quit: Leaving.]
2012-10-26T00:10:13 < TitanMKD> Laurenceb if you prefer ADC = 8bits and DAC=10bits
2012-10-26T00:10:55 < TitanMKD> other SGPIO signals are used for codec clock, host clk, transfer direction ...
2012-10-26T00:12:08 < Laurenceb_> yeah
2012-10-26T00:12:09 < Laurenceb_> http://www.ifixit.com/Teardown/MacBook+Pro+15-Inch+Retina+Display+Mid+2012+Teardown/9462/3
2012-10-26T00:12:13 < Laurenceb_> hehe a stellaris
2012-10-26T00:13:47 < emeb> always with the glue and foam and flex circuits. meh.
2012-10-26T00:14:57 < emeb> I get the feeling that Jony Ive has to have a gang of MEs and EEs following up after him to figure out how to shoehorn working circuits into the industrial designs he dreams up.
2012-10-26T00:15:36 < emeb> and lately they're scrambling so hard that every problem gets solved with something sticky.
2012-10-26T00:16:21 < Tectu> dongs, what scope did you get?
2012-10-26T00:16:31 < Laurenceb_> Renesas R4F2113 H8S series CISC MCU
2012-10-26T00:16:32 < Laurenceb_> odd
2012-10-26T00:17:14 < emeb> no kidding. wtf is that for?
2012-10-26T00:18:36 < TitanMKD> emeb to manage led and I2C maybe ;)
2012-10-26T00:19:59 < emeb> go to the previous pg - the other side of the mainboard is about 50% caps & inductors for all the smps.
2012-10-26T00:24:44 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-26T00:35:01 < ds2> Mmmmmmm flex
2012-10-26T00:35:08 < ds2> the next big frontier
2012-10-26T00:36:31 < emeb> I want to see flex silicon.
2012-10-26T00:36:47 < ds2> MMmm doped silicone
2012-10-26T00:37:05 < emeb> I think there's enough of that in LA already.
2012-10-26T00:38:05 < ds2> but those are merely 2 terminal devices
2012-10-26T00:38:25 < emeb> It's not how many terminals you have, it's what you do with them.
2012-10-26T00:46:55 < Laurenceb_> it isnt that hard to get custom flex pcb made
2012-10-26T00:47:26 < Laurenceb_> i paid about 300euros for 20 lots of 4cm square 2layer with adhesive backing
2012-10-26T00:47:38 < Laurenceb_> and 6mil/6mil tolerances
2012-10-26T00:50:30 < ds2> Laurenceb_: where?
2012-10-26T00:53:10 < Laurenceb_> leiton.de
2012-10-26T00:54:14 < Laurenceb_> i might try making a flexi board stm32 project
2012-10-26T00:54:41 < Laurenceb_> it can be bent around the inside of an enclosure then just stuck on its its adhesive backed
2012-10-26T00:54:54 < Laurenceb_> *if
2012-10-26T00:56:42 < Laurenceb_> this was for a work project
2012-10-26T00:56:56 < Laurenceb_> the previous uk supplier we used was UKP6K MOQ
2012-10-26T01:03:36 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-26T01:06:07 < ds2> $600 isn't bad for a flex
2012-10-26T01:06:22 < ds2> i been ball parked at $2000 and up for a basic cable
2012-10-26T01:20:25 < karlp> we got a bunch of flexi's made up somewhere in china.
2012-10-26T01:20:43 < karlp> some proto company I was looking at recently offered them too.
2012-10-26T01:25:04 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-26T01:25:27 < Laurenceb_> you can buy all sorts of flex cable off the shelf
2012-10-26T01:25:35 < Laurenceb_> i have a load from farnell
2012-10-26T01:25:52 < Laurenceb_> karlp: got any links to the chinese places?
2012-10-26T01:27:21 < ds2> yes but they often aren't exactly what you want
2012-10-26T01:27:27 < Laurenceb_> yeah
2012-10-26T01:27:33 < Laurenceb_> mine arent ideal
2012-10-26T01:27:43 < Laurenceb_> i seem to have fatigued one installing it
2012-10-26T01:28:02 < ds2> like you can't get a 1" long flex cable off the shelf...they seem to start at like 3-6 inches
2012-10-26T01:28:03 < Laurenceb_> the resistance of some of the tracks changes as it moves
2012-10-26T01:28:09 < Laurenceb_> yeah
2012-10-26T01:28:12 < ds2> cool, you have a sensor!
2012-10-26T01:28:20 < Laurenceb_> lol
2012-10-26T01:28:21 < ds2> characterize and sell it
2012-10-26T01:28:33 < Laurenceb_> its pretty random
2012-10-26T01:28:43 < Laurenceb_> but then again... its soo badddd
2012-10-26T01:28:51 < ds2> then it is a random number generator
2012-10-26T01:28:53 < ds2> :)
2012-10-26T01:29:07 < Laurenceb_> nintendo powerglove more like
2012-10-26T01:29:21 < ds2> would be fun to get my hands on one of those
2012-10-26T01:29:44 < karlp> Laurenceb_: ask me tomorrow day time, I'll find out at work.
2012-10-26T01:29:48 < Laurenceb_> thanx
2012-10-26T01:31:38 < ds2> wonder what's the going price for a dead simple, 1 sided PCB, w/o any plated through holes
2012-10-26T01:32:32 < Laurenceb_> moq seems to be around 270 euros
2012-10-26T01:32:53 < Laurenceb_> you can get about 10 x 5cm^2 2layer without much increase in price
2012-10-26T01:33:02  * ds2 converts
2012-10-26T01:33:12 < Laurenceb_>  (ten lots of)
2012-10-26T01:33:20 < ds2> ah, 10 sq inch
2012-10-26T01:33:25 < ds2> 10 x 2 sq inch
2012-10-26T01:33:43 < qyx_> Laurenceb_: 10 pcs of 50cm^2 for 270?
2012-10-26T01:34:01 < ds2> that's a lot to pay for someone to route out boards
2012-10-26T01:34:04 < Laurenceb_> no, 25cm^2
2012-10-26T01:34:16 < Laurenceb_> i meant 5 by 5 cm
2012-10-26T01:34:19 < qyx_> ah
2012-10-26T01:34:29 < ds2> oh 4 sq inches
2012-10-26T01:36:28 < zyp> seeed will do that size for $10, then when you get them just put them in a vice and bend them until they are flexible to your liking
2012-10-26T01:36:59 < Laurenceb_> lol
2012-10-26T01:37:04 < ds2> thinking of an art type project
2012-10-26T01:48:50 < dongs>  < Tectu> dongs, what scope did you get?
2012-10-26T01:48:51 < dongs> wat
2012-10-26T01:49:16 < karlp> is there any cheap protoplace that does plated slots?
2012-10-26T01:49:21 < karlp> I've got a whol heap of these: http://eu.mouser.com/Search/ProductDetail.aspx?R=STX-3150-5N-701Cvirtualkey51700000virtualkey806-STX-31505N701C
2012-10-26T01:49:35 < karlp> and by the time I've made normal drills big enough to fit them, they start to overlap
2012-10-26T01:50:38 < Tectu> dongs, what scope do you have?
2012-10-26T01:50:41 < Tectu> oszillosocpe
2012-10-26T01:50:53 < Tectu> oscilloscope*
2012-10-26T01:51:04 < dongs> dpo4034
2012-10-26T01:58:58 < karlp> asdf
2012-10-26T02:04:53  * GargantuaSauce wishes he had a $9k scope
2012-10-26T02:06:32 < Thorn> >an as adult
2012-10-26T02:11:11 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-26T02:36:54 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-26T02:45:27 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-115.pptp.stw-bonn.de] has joined ##stm32
2012-10-26T03:07:03 < dongs> attn flyback http://www.retronaut.com/2012/10/30-ways-to-die-by-electrocution/
2012-10-26T03:13:44 < GargantuaSauce> how do you fry a southbridge <_<
2012-10-26T03:13:45 < zyp> not strange everything you have is broken when all you do is collect broken stuff
2012-10-26T03:14:31 < Erlkoenig> http://minecraft.niklas.2g2s.de/zeug/fh-labor.jpg  <-- laboratory at "my" uni, we have the assignment to reverse-engineer a communication, that's where we were looking at it with an ancient scope :3
2012-10-26T03:14:46 < zyp> Erlkoenig, I found your problem from yesterday
2012-10-26T03:15:01 < zyp> Erlkoenig, at least, I found a problem that was very similar
2012-10-26T03:15:15 < Erlkoenig> oh tell me
2012-10-26T03:15:31 < Erlkoenig> ah and i got the idea i just might have to initialize the stack pointer
2012-10-26T03:15:35 < zyp> ok, first of all, do you happen to use gcc-4.7?
2012-10-26T03:15:38 < Erlkoenig> just like you have on AVR
2012-10-26T03:15:53 < Erlkoenig> gcc version 4.6.2 20120316 (release) [ARM/embedded-4_6-branch revision 185452] (GNU Tools for ARM Embedded Processors)
2012-10-26T03:15:58 < zyp> no, ok
2012-10-26T03:16:16 < zyp> pastebin your SystemInit function or whatever it was called
2012-10-26T03:16:33 < Erlkoenig> jup mom
2012-10-26T03:18:39 < Erlkoenig> zyp: http://pastebin.com/6q1SvZXu
2012-10-26T03:18:51 < Erlkoenig> it's just from the ST example
2012-10-26T03:19:48 < GargantuaSauce> i guess that would be the obvious way a cmos ic would fail
2012-10-26T03:20:00 < zyp> Erlkoenig, hmm, oh, it looks fine
2012-10-26T03:20:05 < dongs> http://i.imgur.com/Jb3Fv.gif
2012-10-26T03:20:21 < Erlkoenig> zyp: that's what i thought :D
2012-10-26T03:20:30 < zyp> Erlkoenig, or rather, it seems like this is not your problem
2012-10-26T03:20:44 < Erlkoenig> it would be a workaround for a hardware bug
2012-10-26T03:20:47 < zyp> where's your reset vector?
2012-10-26T03:20:55 < zyp> i.e. the code that calls this?
2012-10-26T03:21:41 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-26T03:22:13 < Erlkoenig> zyp: http://pastebin.com/9jBEpjrc <- again, just example code
2012-10-26T03:22:18 < dongs> < Erlkoenig> http://minecraft.niklas.2g2s.de/zeug/fh-labor.jpg  <-- laboratory at "my" uni, we have the assignment to
2012-10-26T03:22:21 < dongs> i spot a f4discovery
2012-10-26T03:22:31 < zyp> oh, reset handler in assembly
2012-10-26T03:22:49 < zyp> then it's a different matter
2012-10-26T03:23:01 < Erlkoenig> i think that code (its "memcpy" thingy) is very uneffective... i might try improving it some time
2012-10-26T03:23:23 < Erlkoenig> (02:22:21) dongs: i spot a f4discovery <- correct, used it to generate test UART signals ;-)
2012-10-26T03:23:39 < Erlkoenig> zyp: one can do that in C? ah
2012-10-26T03:24:03 < zyp> http://paste.jvnv.net/view/MLtA4 <- I was looking at this a few hours ago, it was crashing at the ldmia.w instruction, I thought it looked very similar to your problem
2012-10-26T03:24:23 < dongs> GNU ARM
2012-10-26T03:24:39 < zyp> problem is that it's initializing stack pointer after c compiler has started using stack
2012-10-26T03:24:44 < zyp> for some gcc versions
2012-10-26T03:24:58 < zyp> I thought you might happen to have done the same mistake
2012-10-26T03:25:00 < Erlkoenig> yeah that's why you use ASM startup code :->
2012-10-26T03:25:40 < zyp> stack pointer is loaded from vector 0 by hardware and therefore does not need to be initialized from software
2012-10-26T03:26:20 < Erlkoenig> aaah of course
2012-10-26T03:26:22 < Erlkoenig> forgot it
2012-10-26T03:26:49 < zyp> in your case I can only suggest looking at vector 0, check that the value is sane
2012-10-26T03:28:43 < Erlkoenig> Linker script:     _estack = 0x20020000;    /* end of 128K RAM */
2012-10-26T03:28:43 < Erlkoenig> Startup code:    g_pfnVectors: // Begin of Vector
2012-10-26T03:28:43 < Erlkoenig>   .word  _estack
2012-10-26T03:28:56 < zyp> looks fine
2012-10-26T03:29:08 < Erlkoenig> yup
2012-10-26T03:29:23 < Erlkoenig> and it works fine when not using the debugger ;)
2012-10-26T03:29:33 < zyp> ah, right, forgot about that part
2012-10-26T03:29:53 < zyp> which debugger? texane stlink?
2012-10-26T03:30:04 < Erlkoenig> yup, and the arm-gdb
2012-10-26T03:30:18 < Erlkoenig> hmm might try it with atollic gdb server
2012-10-26T03:30:22 < zyp> interesting
2012-10-26T03:30:28 < zyp> possible bug in texane then
2012-10-26T03:31:33 < Erlkoenig> wouldn't it be a hardware bug if the controller isn't able to reset without power-cycling?
2012-10-26T03:31:43 < zyp> well, bedtime for me, sorry I hadn't found your problem anyway :)
2012-10-26T03:32:03 < Erlkoenig> it's okay ;-)
2012-10-26T03:32:10 < zyp> nah, it might be the st-link that's stuck while pulling reset low
2012-10-26T03:32:19 < Erlkoenig> should go to sleep as well..
2012-10-26T03:32:21 < zyp> or something like that
2012-10-26T03:32:24 < Erlkoenig> well no
2012-10-26T03:32:33 < Erlkoenig> the controller is starting
2012-10-26T03:32:38 < Erlkoenig> but just getting the weird stack problem
2012-10-26T03:32:48 < Erlkoenig> i.e. the MemFault
2012-10-26T03:37:36 < karlp> dongs: that bike crash looks suspiciosly like deadmans bend on the way up to skyline in san jose.
2012-10-26T03:38:04 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Quit: ZNC - http://znc.in]
2012-10-26T03:38:30 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Ping timeout: 264 seconds]
2012-10-26T03:38:35 < dongs> karlp: its US129 whatever that is
2012-10-26T03:38:49 < dongs> original here http://xtremesportsphotography.photoreflect.com/store/Orderpage.aspx?pi=0QAK01AY080000&po=0&pc=28
2012-10-26T03:38:51 < karlp> nah, somewhere else.
2012-10-26T03:38:56 < karlp> never mind then
2012-10-26T03:39:09 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-26T03:39:20 < karlp> not even same coast
2012-10-26T03:39:38 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Client Quit]
2012-10-26T03:41:51 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-26T03:42:09 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 260 seconds]
2012-10-26T03:44:35 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-26T03:47:35 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 256 seconds]
2012-10-26T03:48:09 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-26T03:49:47 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-26T03:52:46 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 244 seconds]
2012-10-26T03:52:58 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has joined ##stm32
2012-10-26T04:09:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-26T04:10:25 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-26T04:37:13 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-26T05:02:51 -!- Erlkoenig [~erlkoenig@pptp-194-94-198-115.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-26T05:23:34 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-26T05:40:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-26T05:53:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-26T05:57:11 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Quit: Tzacks]
2012-10-26T06:46:17 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-26T06:51:16 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 265 seconds]
2012-10-26T06:56:08 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-26T06:56:18 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-26T06:58:08 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 272 seconds]
2012-10-26T07:02:37 < emeb_mac> don't stick your tongue in the jacob's ladder.
2012-10-26T07:03:42 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-26T07:03:54 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Client Quit]
2012-10-26T07:04:57 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-26T07:05:04 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Client Quit]
2012-10-26T07:06:11 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-26T07:07:00 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-26T07:07:51 < ds2> Hmmmm... anyone one have opinions about account angles between features on a board?
2012-10-26T07:07:53 < ds2> accute
2012-10-26T07:07:55 < ds2> blah
2012-10-26T07:08:11 < ds2> acute
2012-10-26T07:09:51 < emeb_mac> I don't worry about them too much.
2012-10-26T07:10:22 < emeb_mac> concerns about acute angles are rooted in the non-idealities of old PCB fab processes.
2012-10-26T07:10:54 < emeb_mac> used to be that etchant would be attracted to the angles by surface tension, pool there and over-etch.
2012-10-26T07:11:04 < emeb_mac> not really a big problem now.
2012-10-26T07:11:22 < ds2> I see
2012-10-26T07:12:03 < emeb_mac> that said, if it's easy to avoid acute angles then I will. Just superstitious behavior I guess. :)
2012-10-26T07:12:26 < ds2> trying to see if I need to clean them up or not before commiting to it
2012-10-26T07:14:11 < emeb_mac> I'd probably do all the easy ones and all the ones on min-width traces.
2012-10-26T07:14:36 < ds2> lots of min width traces
2012-10-26T07:16:07 < emeb_mac> with a lot of acute angles?
2012-10-26T07:16:26 < ds2> kind of
2012-10-26T07:16:35 < ds2> been straightening it out to be at 90degrees
2012-10-26T07:17:05 < emeb_mac> probably a good idea. esp if they're on min width traces @ 5/5 (IIRC)
2012-10-26T07:17:49 < emeb_mac> what layout tool are you using?
2012-10-26T07:18:02 < ds2> eagle
2012-10-26T07:24:10 < emeb_mac> 90deg should be safe in all cases
2012-10-26T07:25:02 < emeb_mac> considering they're almost impossible to avoid when hitting a normal circular via annulus.
2012-10-26T07:25:26 < ds2> yep
2012-10-26T07:36:22 < dongs> attn zyp http://bijo-linux.com/
2012-10-26T07:37:28 < emeb_mac> hilarious
2012-10-26T07:41:05 < emeb_mac> dongs: cleaned up the mask relief on some of the footprints -> http://imagebin.org/233290
2012-10-26T07:41:17 < ds2> dongs: nice
2012-10-26T07:44:48 < baird> dongs: awesome
2012-10-26T07:59:19 < dongs> emeb_mac: almost looks like it wasnt made wiht opensauce
2012-10-26T08:06:17 < emeb_mac> dongs: lols
2012-10-26T08:06:39 < emeb_mac> that's RAW SKILL you're seeing there. :P
2012-10-26T08:07:39 < dongs> yeah im sure it took a lot of raw editing with emacs
2012-10-26T08:07:42 < dongs> to fix the footprints
2012-10-26T08:08:29 < emeb_mac> not a lot, and not emacs, but... well... ;)
2012-10-26T08:09:09 < emeb_mac> I really ought to fix those fps in the library so I don't have to keep doing this.
2012-10-26T08:14:02 < dongs> crystal pads still look wrong
2012-10-26T08:15:22 < dongs> im not sure identifyingf pin1 by a oval pad which wont be sen when soldered is a good idea
2012-10-26T08:15:27 < dongs> i'd put a triangle or osmething there as pin1 mark.
2012-10-26T08:15:38 < dongs> or like my footprint for it, pin1 side line is thicker
2012-10-26T08:16:49 < emeb_mac> Good point on the jtag.
2012-10-26T08:18:44 < emeb_mac> I'll look at the xtal pads again.
2012-10-26T08:25:14 < emeb_mac> xtal pads measure out just like the recommended land pattern in the ds.
2012-10-26T08:26:37 < ds2> emeb_mac: what layout sw are you using?
2012-10-26T08:26:50 < dongs> freakin geda.
2012-10-26T08:26:56 < emeb_mac> that
2012-10-26T08:27:15 < ds2> oh
2012-10-26T08:27:18 < ds2> is that usable yet?
2012-10-26T08:27:29 < emeb_mac> ds2: I've been using it for years
2012-10-26T08:27:37 < emeb_mac> just gets better all the time.
2012-10-26T08:27:58 < ds2> so it has enough libraries to make it useable?
2012-10-26T08:28:29 < emeb_mac> ds2: Enough for what I do. I end up building a lot of fps by hand though.
2012-10-26T08:28:48 < emeb_mac> don't think that's too unusual no matter what sw you use.
2012-10-26T08:28:51 < ds2> I see
2012-10-26T08:29:03 < ds2> I find i get a lot of the basics from eagle already
2012-10-26T08:29:12 < emeb_mac> that's good
2012-10-26T08:29:25 < emeb_mac> a lot of folks use eagle so there are a lot of libs out there.
2012-10-26T08:29:31 < ds2> can you directly export or import LTspice netlists?
2012-10-26T08:29:37 < ds2> (or schematics)
2012-10-26T08:30:00 < emeb_mac> there is a spice netlist conversion available. I've used it with the FOSS spice simulators.
2012-10-26T08:30:20 < dongs> i never trust supplied footprints, especially opensauce footprints
2012-10-26T08:30:28 < dongs> sparkfail eagle "library" is a great example of huge fail
2012-10-26T08:30:31 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-26T08:30:34 < emeb_mac> what he said - always check them yourself
2012-10-26T08:30:40 < dongs> with proper pcb cad, drawing a footprint should be trivial
2012-10-26T08:30:58 < dongs> so i just have my own custom libs with stuff I use and never rely on shit others make
2012-10-26T08:31:11 < emeb_mac> that's the one big gap in geda/pcb - doing fps graphically is a non-starter.
2012-10-26T08:31:39 < emeb_mac> I always need a sheet of paper, calculator and text editor.
2012-10-26T08:31:45 < dongs> loll..
2012-10-26T08:31:56 < emeb_mac> doing a complex fp, like an SMD SD card socket can take an hour or so.
2012-10-26T08:32:38 < ds2> what about doing one for those wireless modules?
2012-10-26T08:33:14 < emeb_mac> the ones with the crazy little pcb edge soldered pads?
2012-10-26T08:33:39 < ds2> that and pads underneath
2012-10-26T08:34:07 < dongs> http://i.imgur.com/VTDr9.png i made this 2-3 years ago, took prob around 5mins
2012-10-26T08:34:37 < emeb_mac> wouldn't be too bad, but depends on how well the datasheet drawing is annotated.
2012-10-26T08:34:54 < emeb_mac> dongs: that would take a lot more than 5min in geda/pcb
2012-10-26T08:35:18 < dongs> yea im sure
2012-10-26T08:35:30 < dongs> i used to chat in thneir channel until that faggot DJ banned me
2012-10-26T08:35:39 < dongs> cuz he couldnt handle "swear words" on irc
2012-10-26T08:35:46 < emeb_mac> heh
2012-10-26T08:35:58 < emeb_mac> he's fairly straight-laced.
2012-10-26T08:37:32 < emeb_mac> main complaint I have about the fp definitions is that the shape of pads is strangely defined.
2012-10-26T08:38:44 < emeb_mac> not just corner to corner but you specify width and then a distance between squares of that width. Nonintuitive.
2012-10-26T08:41:30 < emeb_mac> dongs: you use diptrace, right?
2012-10-26T08:44:28 < dongs> yea
2012-10-26T08:44:38 < dongs> you should give it a try. youll never go back to the garbage that is geda
2012-10-26T08:45:20 < emeb_mac> mebbe, but then I'd have to hold my nose and use winders.
2012-10-26T08:45:22 < emeb_mac> :P
2012-10-26T08:45:50 < dongs> they do have some gay macos build of it.
2012-10-26T08:46:00 < dongs> some other macfag I know apparently uses it.
2012-10-26T08:46:02 < emeb_mac> I've got a buddy that uses it - he's very happy with it. Esp after the latest update.
2012-10-26T08:46:07 < dongs> according to some lunix guy it also runs fine in wine.
2012-10-26T08:46:27 < emeb_mac> wine. meh. I've got a winders box.
2012-10-26T08:47:07 < emeb_mac> I like FOSS, but don't piss in the wind.
2012-10-26T08:51:21 -!- claude is now known as Claude
2012-10-26T08:56:35 < dongs> http://i.imgur.com/gH5do.jpg
2012-10-26T08:58:08 < emeb_mac> they don't want to deal with old feet I guess.
2012-10-26T09:05:13 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-26T09:11:03 < ds2> do any of these packages generate 3D models in IGES format?
2012-10-26T09:13:09 < dongs> dicktrace makes vrml
2012-10-26T09:13:49 < ds2> how accurate is that for cad work?
2012-10-26T09:14:11 < ds2> Protel was almost tempting enough just for the 3D modeling
2012-10-26T09:15:35 < dongs> um, i assume it uses proper units since pcb is proper.
2012-10-26T09:15:39 < dongs> i can export something and you can try.
2012-10-26T09:15:45 < dongs> i dont have anything that imports vrml heh
2012-10-26T09:16:43 < dongs> http://timecop.mine.nu/trash.zip
2012-10-26T09:18:15 < dongs> im pretty sure dicktrace guyts can be convinced to do iges instead of vrml
2012-10-26T09:18:18 < dongs> cuz who the fuck uses that??
2012-10-26T09:18:19 < dongs> in 2012
2012-10-26T09:42:27 < ds2> what would you use as a CAD import format then?
2012-10-26T09:42:45 < ds2> IIRC, VRML isn't very precise
2012-10-26T09:49:21 < dongs> tehy look to be using doubles in vrml
2012-10-26T09:49:22 < dongs> in mm
2012-10-26T09:49:26 < dongs> (optionduring export
2012-10-26T09:49:40 < ds2> I'll try converting it later
2012-10-26T09:58:38 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 272 seconds]
2012-10-26T10:09:09 < ds2> anyone tried mounting a QFN part using hot air and wire solder (tin pads, cover with flux then hotair it on)?
2012-10-26T10:09:22 < jpa-> yes
2012-10-26T10:09:26 < jpa-> works fine
2012-10-26T10:09:38 < ds2> does that self center well?
2012-10-26T10:10:01 < jpa-> quite well; problem is mostly getting the solder to melt evenly on all sides without blowing away the part
2012-10-26T10:10:37 < ds2> can't I just hold it down with tweezers on top while it melts?
2012-10-26T10:11:11 < jpa-> i guess you can
2012-10-26T10:11:32 < jpa-> i just keep the hot air tool a bit further away
2012-10-26T10:11:40 < ds2> Ohhh is that all
2012-10-26T10:11:40 < dongs> that'll cool that section down because tweezers touch it.
2012-10-26T10:11:49 < ds2> gotcha
2012-10-26T10:12:04 < ds2> thinking of trying to mount some sensor chips myself
2012-10-26T10:12:59 < ds2> I'll just plan on destorying one chip as learning then
2012-10-26T10:13:43 -!- Claude is now known as claude
2012-10-26T10:14:14 < jpa-> it's not too difficult with just soldering iron either, though that way i've ruined a chip once
2012-10-26T10:15:39 < ds2> what's the trick to get a QFN to mount with an iron?
2012-10-26T10:15:53 < ds2> I'd expect tinning to cause it to not be planar
2012-10-26T10:16:37 < jpa-> yeah, tin both the board and the chip
2012-10-26T10:17:16 < jpa-> then just heat one pad at the time.. but you have to make the pads on the footprint a bit longer in order to be able to heat them
2012-10-26T10:18:13 < ds2> doesn't the chip sit on the cold solder of the other pads when I just melt one at a time?
2012-10-26T10:19:04 < jpa-> yeah, but the solder on the pads will still heat enough to join together.. probably not the best solder joint ever, but good enough for protos
2012-10-26T10:20:03 < ds2> ah I see
2012-10-26T10:20:44 < dongs> sounds like a great way to get sensors mounted not exactly level/perpendiciular
2012-10-26T10:21:00 < jpa-> dongs: hehe, probably :)
2012-10-26T10:21:20 < ds2> makes for fun accelerometers :D
2012-10-26T10:26:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-26T10:37:47 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has left ##stm32 []
2012-10-26T10:40:26 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-26T10:44:38 < dongs> sup trolls
2012-10-26T10:45:35 -!- dekar [~dekar@212.255.27.61] has quit [Quit: This computer has gone to sleep]
2012-10-26T10:46:52 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-26T11:00:26 -!- claude is now known as claude_work
2012-10-26T11:02:54 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-26T11:14:13 < baird> foodblog: bear scones for dinner
2012-10-26T11:15:34 < baird> *beer
2012-10-26T11:31:21 -!- claude_work is now known as claude
2012-10-26T11:33:18 -!- claude is now known as claude_work
2012-10-26T11:36:36 -!- claude_work is now known as Claude_work
2012-10-26T11:40:10 -!- Claude_work is now known as claude
2012-10-26T11:48:41 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-26T11:48:41 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-26T12:05:42 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-26T12:08:26 < jpa-> hmm, using texane stlink, i started getting verify errors yesterday
2012-10-26T12:11:07 < jpa-> at 0006B4, 0086B4, 0106B4, 0186B4, 0206B4, 0286B4   there are 12 bytes of zeroes when i read the data back, and the program doesn't work
2012-10-26T12:11:18 < jpa-> when i use the ST's own windows tool, the same device programs fine
2012-10-26T12:15:16 < zyp> interesting
2012-10-26T12:15:19 < zyp> what did you change?
2012-10-26T12:15:32 < zyp> I mean, anything changed recently?
2012-10-26T12:22:53 < jpa-> not really anything.. i was debugging and loading code from GDB, it all worked fine; then i quit gdb and made some small changes to my program, tried programming with the st-flash instead of gdb, didn't work
2012-10-26T12:23:03 < jpa-> now it doesn't work with gdb nor with st-flash
2012-10-26T12:24:24 < jpa-> first i thought i would have weared out my flash, but it's specified to 10k erase cycles and i've only had about 300 so far
2012-10-26T12:29:53 -!- mrcan [~mrcan@unaffiliated/mrcan] has quit [Quit: Leaving]
2012-10-26T12:30:54 < jpa-> wtf
2012-10-26T12:31:00 < jpa-> reboot of ubuntu helped
2012-10-26T12:31:14 < jpa-> disconnecting the st-link wasn't enough
2012-10-26T12:32:43 < karlp> odd.
2012-10-26T12:32:52 <+Steffanx> And i thought "I'm not going to be silly this time and say: Try a reboot" :P
2012-10-26T12:32:57 < karlp> what ubuntu is this?
2012-10-26T12:33:09 < karlp> that's some crazy libusb shit
2012-10-26T12:33:27 < jpa-> 12.04
2012-10-26T12:33:41 < jpa-> but extra factor is that it is running under virtualbox
2012-10-26T12:33:59 < karlp> shouldn't matter, I've used stlink lots under virtualbox
2012-10-26T12:34:30 < jpa-> yeah it works fine for me too, until now
2012-10-26T12:34:47 < jpa-> also, rebooting host windows didn't help, rebooting virtual ubuntu helped
2012-10-26T12:47:24 < Laurenceb> haha 12.04
2012-10-26T12:47:36 < Laurenceb> you deserve everything you get
2012-10-26T12:47:46 <+Steffanx> So do you
2012-10-26T12:48:11 < Laurenceb> i get trolled
2012-10-26T12:49:17 <+Steffanx> You want to say you don't deserve it?
2012-10-26T12:53:12 < dongs> ebike is alive
2012-10-26T12:53:30 <+Steffanx> video or it didn't happen
2012-10-26T12:53:38 < dongs> still in pieces
2012-10-26T12:53:41 < dongs> only tested electric
2012-10-26T12:53:51 < dongs> now need to measure + find belt
2012-10-26T12:54:33 < Laurenceb> hmm
2012-10-26T12:54:40 < Laurenceb> what batteries does it have?
2012-10-26T12:54:57 < dongs> SLA, but they're probably trash
2012-10-26T12:55:02 < dongs> im gonna get a 6S lipo from HK
2012-10-26T12:55:07 < dongs> for future use
2012-10-26T12:56:18 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Ping timeout: 244 seconds]
2012-10-26T12:56:54 < dongs> when SLA is in series, what happens to Ah rating?
2012-10-26T12:56:57 < dongs> double or same?
2012-10-26T12:57:04 < dongs> (i.e. they're 7000mAh
2012-10-26T12:57:14 < karlp> same right? double voltage
2012-10-26T12:57:18 < jpa-> same, obviously
2012-10-26T12:57:29 < dongs> fuck if i know, im stupid
2012-10-26T12:57:30 < dongs> thanks
2012-10-26T12:57:36 < dongs> so i dont need 14000mAh lipo
2012-10-26T12:57:39 < dongs> cuz this shit looks expensive ;p
2012-10-26T12:59:12 < dongs> hmm it might be cheaper to do 2x 3S in series
2012-10-26T12:59:56 < dongs> http://www.hobbyking.com/hobbyking/store/__18507__B_Grade_5800mAh_3S_25C_Lipoly_Battery.html 2 of these
2012-10-26T13:00:24 < dongs> B-Grade Lithium Polymer packs are great for those not-so-important models or testing. They have a high internal resistance (thats bad), normally wont last past 60 full high-discharge cycles and may become puffy over time.
2012-10-26T13:00:28 < dongs> lol.
2012-10-26T13:02:56 < zyp> sounds like a nice way to have the ebike suddenly burst into flames
2012-10-26T13:03:02 < dongs> indeed
2012-10-26T13:03:30 < dongs> http://www.hobbyking.com/hobbyking/store/__10286__Turnigy_5800mAh_3S_25C_Lipo_Pack.html but equivalent non-shit one is like double the price
2012-10-26T13:04:28 < dongs> http://www.hobbyking.com/hobbyking/store/__9184__Turnigy_5000mAh_3S_20C_Lipo_Pack.html might have to settle for this
2012-10-26T13:05:23 < karlp> what's the discharge rate on SLAs?
2012-10-26T13:05:43 < dongs> if it s 5Ah,, and motor is 250W,, so thats liek 24V@250W, ~10A, so it should run for like half an hour? well ~20min safely
2012-10-26T13:05:49 < dongs> karlp: way less i'd imagine
2012-10-26T13:06:23 < karlp> you see home welder rigs bodged up with SLAs, would have thought it was decent enough?
2012-10-26T13:06:38 < dongs> actually never mind, cause car starter takes a fuckton of amps
2012-10-26T13:06:38 < dongs> so yeah
2012-10-26T13:07:35 < dongs> the motor controller in this thing is 30A peak, so thats only 5C..
2012-10-26T13:07:37 < dongs> 20C should be plenty.
2012-10-26T13:09:16 < karlp> isn't a coulomb an amp second?
2012-10-26T13:09:34 < dongs> thats not what C means in lipo shit
2012-10-26T13:09:44 < dongs> 1C = whatever mAh the battery is rated for
2012-10-26T13:10:02 < dongs> so 20C 5Ah battery can deliver 20*5 amps
2012-10-26T13:10:15 < dongs> http://www.commonsenserc.com/page.php?page=c_ratings_explained.html
2012-10-26T13:10:28 < zyp> «common sense rc»
2012-10-26T13:10:29 < karlp> ok, thanks :)
2012-10-26T13:10:32 < dongs> lo
2012-10-26T13:12:01 < dongs> aww,, there was a cheapass 5000/15C batt but its back order :(
2012-10-26T13:16:05 < dongs> Baterie namáa predepsanou kapacitu 5000mAh.Koupil jsme jednu na zkousku a i po 6ti nabiti je kapacita pouhych 3000mAh!!!!!Polepsete se HK!!!!!
2012-10-26T13:16:25 < dongs> i dont read polish but im gonna guess: I bought this fucking thing that said 5000mAh. but its fucking 3000. what a piece of shit hk!!!
2012-10-26T13:20:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-26T13:36:13 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-26T13:37:57 < Laurenceb> https://www.facebook.com/media/set/?set=a.531010533579662.138394.259791880701530&type=3
2012-10-26T13:40:25 <+Steffanx> nice
2012-10-26T13:54:36 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-26T13:57:10 < dongs> < Harekiet> https://youtube.com/watch?v=FFGfq0pRczY i for one welcome our drunk robot overlords
2012-10-26T14:03:57 < dongs> http://www.flickmylife.com/wp-content/uploads/2012/10/velagtn1.jpg appropriate license plate
2012-10-26T14:10:14 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-26T14:10:41 < Laurenceb> ooh
2012-10-26T14:10:48 < Laurenceb> ASSMANN has left me some cables
2012-10-26T14:12:49 <+Steffanx> Whoever that may be
2012-10-26T14:17:43 < Laurenceb> http://www.assmannusa.com/
2012-10-26T14:17:55 < dongs> i used to buy assman usb cables
2012-10-26T14:17:59 < dongs> until I got a better deal on them in korea
2012-10-26T14:25:34 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 240 seconds]
2012-10-26T14:26:45 < dongs> http://i.ebayimg.com/t/Powder-Coated-Shure-SM58-Microphone-Grille-in-Hot-Pink-NEW-/00/s/MTU1NlgxNjAw/$T2eC16F,!)8E9s4l5-(tBQPis+Qo0w~~60_57.JPG do want
2012-10-26T14:28:55 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-26T14:33:14 <+Steffanx> Just because it's pink dongs?
2012-10-26T14:33:31 < Laurenceb> my dong is pink
2012-10-26T14:35:37 <+Steffanx> At least we also have a famous pink microphonish think https://upload.wikimedia.org/wikipedia/commons/thumb/3/31/Powned_microfoon.jpg/800px-Powned_microfoon.jpg ..
2012-10-26T14:35:49 <+Steffanx> When you see that in dutchland ( when you're there ) RUN :P
2012-10-26T14:39:15 <+Steffanx> but Laurenceb would probably like it.. it's a sort of croud funded troll TV show here in dutchland
2012-10-26T14:41:02 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Remote host closed the connection]
2012-10-26T14:41:21 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has joined ##stm32
2012-10-26T15:07:43 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-26T15:32:46 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-26T15:32:46 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-26T15:32:46 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-26T15:40:31 < Laurenceb> odd
2012-10-26T15:40:36 < Laurenceb> fatfs is giving me FR_NOT_READY
2012-10-26T15:46:41 -!- BjoernC_ [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-26T15:46:59 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Read error: Connection reset by peer]
2012-10-26T15:51:24 < dongs> fartfs
2012-10-26T15:51:37 < dongs> faggots delayed my submission of stm32clonelogger until monday
2012-10-26T15:51:40 < dongs> so i wont see pcbs until wed-thu
2012-10-26T15:52:12 < HTT-Bird> ugh.  PCB fab difficulties?
2012-10-26T15:53:04 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-26T16:01:31 < Laurenceb> aha
2012-10-26T16:01:39 < Laurenceb> clk and miso shorted
2012-10-26T16:02:03 < HTT-Bird> heh, that'd screw things up alright
2012-10-26T16:02:30 < Laurenceb> works now
2012-10-26T16:02:58 < dongs> no, no difficulties. im freescamming the logger clones so i need to wait for enough shit to panel them into something.
2012-10-26T16:03:08 < dongs> was supposed to happen today but the fag who merges designs was golfing or somethign
2012-10-26T16:03:23 < Laurenceb> hehe
2012-10-26T16:03:24 < dongs> first world problems.
2012-10-26T16:03:34 < Laurenceb> sounds like where i work
2012-10-26T16:03:53 < dongs> i  didnt think you worked
2012-10-26T16:04:07 < Laurenceb> lol
2012-10-26T16:04:10 < Laurenceb> i troll
2012-10-26T16:13:01 < Laurenceb> hehe someone asked for links to ballooning blogz
2012-10-26T16:13:05 < Laurenceb> http://www.daveakerman.com/
2012-10-26T16:13:30 < Laurenceb> http://www.daveakerman.com/wp-content/uploads/2012/09/IMG_4049-e1347360316599-768x1024.jpg
2012-10-26T16:13:39 < Laurenceb> steriotypical bloggerzzz
2012-10-26T16:15:17 < karlp> Laurenceb: apparently our flexipcbs were organized by some dude we know who lives in shenzen, who went and "sorted it out"
2012-10-26T16:15:34 < karlp> so no web shop or quotes or anything for you sorry
2012-10-26T16:15:35 < Laurenceb> i see
2012-10-26T16:15:38 < Laurenceb> nvm
2012-10-26T16:16:04 < Laurenceb> thanks for the info
2012-10-26T16:16:48 < karlp> http://www.linkedin.com/pub/dir/%C3%81sgeir/Bjarnason
2012-10-26T16:17:06 < karlp> you can probably just tlak to him and say we sent you? he probably wouldn't object to more business?
2012-10-26T16:22:55 < Laurenceb> i dont have any boards to submit at the moment :P
2012-10-26T16:25:18 < zyp> Laurenceb just wanted to look at the specs so he could say «nice.. but it's shit»
2012-10-26T16:27:49 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-111.pptp.stw-bonn.de] has joined ##stm32
2012-10-26T16:28:50 < Erlkoenig> why do the arm-gcc toolchain and the CMSIS/StdPeriph library not provide the NULL
2012-10-26T16:28:54 < Erlkoenig> macro?
2012-10-26T16:30:26 < zyp> huh?
2012-10-26T16:30:28 < Erlkoenig> and why is 0x0 a valid address ... design fail :S
2012-10-26T16:31:31 < Erlkoenig> is there any reserved address on the STM32F4 that will surely result in a MemFault (or something like that) when trying to access it? as a NULL-replacement
2012-10-26T16:32:49 < zyp> just use the MPU to deny access to addr 0?
2012-10-26T16:33:04 < Erlkoenig> seems like to complicated? :D
2012-10-26T16:33:20 < zyp> dunno, haven't used the MPU yet
2012-10-26T16:33:23 < dongs> if you got code accessing 0, youve got other issues
2012-10-26T16:33:29 < zyp> yep
2012-10-26T16:33:46 < Erlkoenig> yes, and to find these issues, i'd like to get a MemFault or something
2012-10-26T16:34:13 < Erlkoenig> just like it's on x86 - initialize pointers with NULL=0x0000 , try to acess them => SIGSEGV
2012-10-26T16:34:32 < dongs> youre gonna get it anyway...
2012-10-26T16:35:09 < dongs> when you have a pointer to a struct or something goign to 0, it'll be filled with crap
2012-10-26T16:35:13 < dongs> and will surely cause a hardfault
2012-10-26T16:35:20 < zyp> I'm pretty sure that 0 is a valid addr on x86, until MMU is enabled and it's configured to be an unmapped page
2012-10-26T16:35:29 < Thorn> that SIGSEGV isn't magically generated by hardware either
2012-10-26T16:35:34 < zyp> so that's comparable to using the MPU on cortex-m
2012-10-26T16:35:37 < Thorn> exactly
2012-10-26T16:36:23 < Erlkoenig> hmm okay, the OS does it
2012-10-26T16:36:36 < zyp> yep, and so can yours on cortex-m
2012-10-26T16:36:49 < Erlkoenig> i have none
2012-10-26T16:37:18 < Erlkoenig> #define NULL ((void*) 0x20020000)    ... since that was the address that crashed the core ...
2012-10-26T16:37:40 < zyp> don't do that.
2012-10-26T16:37:46 < zyp> seriously, don't do that.
2012-10-26T16:38:40 < Erlkoenig> can you explain? it would be helpful to find errors where i coded wrong and the program tries to access uninitialized pointers
2012-10-26T16:38:52 < zyp> well
2012-10-26T16:38:55 < zyp> first of all
2012-10-26T16:39:11 < zyp> bool(NULL) is supposed to be false, bool(0x20020000) is not
2012-10-26T16:39:36 < zyp> so you can't use a simple if(ptr) foo(ptr->x); or whatever
2012-10-26T16:39:58 < Erlkoenig> hmm
2012-10-26T16:40:25 < zyp> it's going to cause more issues than it solves
2012-10-26T16:41:00 < zyp> it'll also render the compiler's null-pointer optimizations useless
2012-10-26T16:41:48 < Erlkoenig> hm okay... than i'll go with NULL and try the MPU trick later
2012-10-26T16:42:35 < Erlkoenig> aeh, "0", not "NULL" ;-)
2012-10-26T16:43:55 -!- dekar [~dekar@dslb-178-004-019-143.pools.arcor-ip.net] has joined ##stm32
2012-10-26T16:43:58 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-26T17:00:21 < Erlkoenig> hmm, can compilers optimize memory accesses for that weird bit-banding stuff?
2012-10-26T17:01:02 < zyp> not as far as I know
2012-10-26T17:01:14 < baird> Besides from giving you a few macros, no.
2012-10-26T17:01:32 < Erlkoenig> hm. i could imagine this would speed up a few things
2012-10-26T17:07:09 < Erlkoenig> why does the Cortex-M4F automatically push the R0-R3, R12 registers upon exception entry? so small ISR's don't need to any pushes, but why doesn't the core push the remaining registers, r5-r11? because that would be too slow for most cases?
2012-10-26T17:07:40 < zyp> because it's eabi compliant
2012-10-26T17:07:56 < zyp> eabi defines that some registers are caller-save and some are callee-save
2012-10-26T17:08:28 < Erlkoenig> ah
2012-10-26T17:08:30 < Erlkoenig> hmm
2012-10-26T17:08:44 < Erlkoenig> could you provide the disassembly of your switch_context () function?
2012-10-26T17:08:45 < zyp> caller-save registers are registers that functions are free to overwrite as they like, callee save registers have to be preserved by the function that is called if it is using them
2012-10-26T17:09:32 < Erlkoenig> hm okay, registers r5-r11 aren't caller-save because of hardware limitations?
2012-10-26T17:10:09 < karlp> sounds more like a tradeoff to not push everything all the time for the small cases like you suggested.
2012-10-26T17:10:26 < Erlkoenig> ah okay ;-)
2012-10-26T17:10:26 < zyp> Erlkoenig, no, because of convention
2012-10-26T17:11:35 < zyp> having callee-save registers means that only those registers that will be used need to be pushed
2012-10-26T17:12:38 < zyp> but registers r0-r3 are used for passing function arguments, which the calling function usually don't need after the function call
2012-10-26T17:12:49 < zyp> so it's no problem if the called function overwrites them
2012-10-26T17:13:03 < zyp> so the calling function only needs to save them if it needs them afterwards
2012-10-26T17:14:25 < Erlkoenig> ah, the callING function pushes r0-r3, and the callED function pushes r5-r11 (both only if needed)?
2012-10-26T17:14:48 < Erlkoenig> when talking about normal function calls, not exception entry
2012-10-26T17:15:37 < zyp> exactly
2012-10-26T17:15:52 < zyp> and on exception entry, the exception hardware takes the role of the calling function
2012-10-26T17:16:09 < Erlkoenig> hmm fascinating:D
2012-10-26T17:16:17 < Erlkoenig> how fast is the Cortex-M4F in the automatic pushing of r0-r3, r12, etc? one cycle per register?
2012-10-26T17:16:24 < zyp> it must save all the registers that the ISR is allowed to just overwrite, so the ISR won't interfere with other stuff
2012-10-26T17:16:35 < zyp> I believe so
2012-10-26T17:17:18 < Erlkoenig> this is all so much more powerful than the AVR :D
2012-10-26T17:19:05 < Erlkoenig> btw. yesterday i heard a small speech by students about introduction to mikrocontrollers... the first statement was "mikrocontrollers can deal with individual bits", followd by "mikrocontrollers have all periphals integrated on one chip"
2012-10-26T17:19:45 < dongs> https://youtube.com/watch?v=aigWkZHcqpA
2012-10-26T17:19:56 < zyp> the first statement is just bullshit
2012-10-26T17:20:13 < zyp> it was probably intended as «microcontrollers has GPIO»
2012-10-26T17:20:32 < Erlkoenig> yep that's what i thought... but why did they do it as the first statement in ther presentation oO
2012-10-26T17:20:55 < zyp> the other statement is pretty correct, microcontrollers usually do have both flash and ram in one chip
2012-10-26T17:20:55 <+Steffanx> The first one that was actually unexpected and funny dongs :D
2012-10-26T17:21:25 < Erlkoenig> well chip as in "plastic thingy with copper foots" not as in "die" ;-)
2012-10-26T17:21:32 < zyp> Erlkoenig, probably to distinguish it from huge computers without GPIO
2012-10-26T17:21:53 < Erlkoenig> yes but when phrasing it this way it's all screwed up :D
2012-10-26T17:22:02 < zyp> go tell them.
2012-10-26T17:22:12 < Erlkoenig> the prof did ^^
2012-10-26T17:22:24 < Erlkoenig> also nice: "oscilloscopes have different modi of displaying input: sinus, rectangle, sawtooth" oO
2012-10-26T17:24:07 < zyp> mine doesn't :(
2012-10-26T17:24:16 < zyp> that's what I get for buying a cheap scope :(
2012-10-26T17:24:28 < zyp> dongs' scope is so awesome, it probably has that
2012-10-26T17:24:29 < Erlkoenig> well uhm... none has? they just display the input, whatever form it has?
2012-10-26T17:24:54 < Erlkoenig> how you gonna display a weird noise signal as a sawtooth? :D
2012-10-26T17:25:26 < zyp> hey, I'm joking :p
2012-10-26T17:25:32 < Erlkoenig> ah good :D
2012-10-26T17:25:41  * Erlkoenig justifies his joke-detector
2012-10-26T17:26:12 < Erlkoenig> wait. that's the wrong word.
2012-10-26T17:26:19 < Erlkoenig> "adjust"
2012-10-26T17:27:08 < BrainDamage> actually the sampling error shape for an unifomingly distributed input value on an adc operation is a sawtooth which is equal to a noise!
2012-10-26T17:27:30 < BrainDamage> ( quantization noise )
2012-10-26T17:28:02 < Erlkoenig> "gcc function attribute interrupt: Use this attribute on the ARM, ...  ports to indicate that the specified function is an interrupt handler.  The compiler will generate function entry and exit sequences suitable for use in an interrupt handler when this attribute is present. On ARMv7-M the interrupt type is ignored."  ... aaaah :D
2012-10-26T17:28:21 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-26T17:28:27 <+Steffanx> BrainDamage really has a theoretical explanation for everything :)
2012-10-26T17:29:21 < Erlkoenig> zyp: could you show me the disassembly of your switch_context() function? i'd like to see what happens with the line     asm volatile ("str sp, [%0]" :: "r" (&Thread::active_thread->sp));
2012-10-26T17:29:31 < zyp> sure
2012-10-26T17:30:08 < zyp> http://paste.jvnv.net/view/PHPyn
2012-10-26T17:30:27 < zyp> turns into str.w   sp, [r2]
2012-10-26T17:31:10 < zyp> and before that, r2 gets initialized with the address in the C expression
2012-10-26T17:32:14 < Erlkoenig> aah neat, the compiler generates code to calculate that address
2012-10-26T17:33:05 < Erlkoenig> what would happen if your inline assembly code did something with the r* registers - they would interfere with the compiler generated code`
2012-10-26T17:33:06 < Erlkoenig> ?
2012-10-26T17:34:48 < Laurenceb> what the fuuuuuu
2012-10-26T17:35:03 < Laurenceb> my lipo charger ic only works if i set breakpoints
2012-10-26T17:35:07 < Laurenceb> this makes no sense
2012-10-26T17:35:28 < Laurenceb> ive got a stm32logger board, same design as others that work fine
2012-10-26T17:35:30 < Erlkoenig> heisenbug? :D
2012-10-26T17:35:36 <+Steffanx> Laurduino will fix it all, Laurenceb
2012-10-26T17:35:41 < Laurenceb> only the charger en pin goes to 0v at usb connection
2012-10-26T17:35:56 < Laurenceb> well - at some point during init
2012-10-26T17:36:06 < Erlkoenig> btw: http://minecraft.niklas.2g2s.de/elektronik/BEC.jpg <-- these superfine cables are supposed to carry 5A... oO
2012-10-26T17:36:07 < Laurenceb> if i connect swd to take a look the problem goes
2012-10-26T17:37:12 < BrainDamage> Erlkoenig: they are not superfine, compare to mains cabling which handle 16A, also, they look short
2012-10-26T17:38:13 <+Steffanx> Erlkoenig is going to build a quadrocopter in minecraft?
2012-10-26T17:38:16 < Erlkoenig> hmm... i think i'll measure the drop voltage under load...
2012-10-26T17:38:38 < Erlkoenig> Steffanx: no, a submarine in real life :3
2012-10-26T17:38:44 < Erlkoenig> http://www.youtube.com/watch?v=h2cA3pv0ONs&feature=youtu.be
2012-10-26T17:38:47 <+Steffanx> Also nice
2012-10-26T17:39:32 < Laurenceb> ok... with -Os it works
2012-10-26T17:39:41 < BrainDamage> where's the hull cannon?
2012-10-26T17:40:34 < Erlkoenig> whats that?
2012-10-26T17:40:45 < Laurenceb> fires you to Hull
2012-10-26T17:40:55 < Erlkoenig> my crazy friends have an idea of torpedos. :D
2012-10-26T17:41:09 < Erlkoenig> CO² cartridges... the firing mechanism is problably somewhat delicate
2012-10-26T17:41:19 <+Steffanx> Once I had plans for something similar, but my plans always end with "nah .. impossible to do" :(
2012-10-26T17:41:40 < Erlkoenig> most importantly, it's gonna go with a F4 Discovery :3
2012-10-26T17:42:35 < BrainDamage> https://lockthewelderdotcom.files.wordpress.com/2011/02/u-boat-2.jpg
2012-10-26T17:42:39 < BrainDamage> the front gun
2012-10-26T17:43:19 < Erlkoenig> it's probably a FLAK
2012-10-26T17:43:57 < Erlkoenig> hmmm hats
2012-10-26T17:44:07 < karlp> old school uboats were hardcore
2012-10-26T17:44:34 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-26T17:44:42 < BrainDamage> Erlkoenig: CO2 like a rocket, or co2 moving a propeller?
2012-10-26T17:44:45 < Erlkoenig> karlp: hardcore for the crew :D
2012-10-26T17:45:01 < Erlkoenig> BrainDamage: like a rocket... it'll probably move about 10cm and then get empty
2012-10-26T17:45:12 < karlp> that
2012-10-26T17:45:34 < Erlkoenig> of course the german submarines are still the best
2012-10-26T17:45:37 <+Steffanx> There are a few videos about miniature torpedos on youtube .. they looked pretty neat
2012-10-26T17:46:38 < Erlkoenig> but they sponsored a submarine for israel... for assuring it could fire strategic missiles in case of annihilation if isreal... thats somewhat insane
2012-10-26T17:46:53 < BrainDamage> people are insane
2012-10-26T17:47:23 <+Steffanx> Of course we are
2012-10-26T17:47:25 < BrainDamage> btw you should make you test trip to england with the u boat
2012-10-26T17:47:43 < BrainDamage> make sure to paint it black and use WWII german flags
2012-10-26T17:47:52 < baird> We had Japan attacking Sydney with mini manned subs; a lot of them got away with it, too-- they also found the one that didn't escape a few years ago off hte coast.
2012-10-26T17:47:52 < Erlkoenig>  people would die laughing? :D
2012-10-26T17:48:27 < Erlkoenig> i don't think it's even strong enough for crawling through a river... :D
2012-10-26T17:48:37 <+Steffanx> You don't need large subs anyway to fit a few japanese baird :P
2012-10-26T17:49:04 < baird> I wonder how deep a amateur submersible could go..
2012-10-26T17:49:24 <+Steffanx> To the bottom of the ocean, for sure
2012-10-26T17:49:38 < BrainDamage> yeah, going down isn't an issue
2012-10-26T17:49:41 < BrainDamage> it's going back up
2012-10-26T17:50:02 < Erlkoenig> our drain pipes are surprisingly stable :D
2012-10-26T17:51:19 < Erlkoenig> they are plugged together with some rubber rings that are lubed with some fat... with three persons we needed a lot of force to pull them apart
2012-10-26T17:53:42 < baird> I almost got into scuba-- have the C-card, but no-one to dive with-- I could think of a few reasons why a handmade robot sub wouldn't be that straight-forward. You would have to forget any telemetry..
2012-10-26T17:54:36 < zyp> Erlkoenig, no, the compiler handles register allocation, that's why I'm using the clobber lists
2012-10-26T17:55:02 < Erlkoenig> zyp: clobber lists? whats that?
2012-10-26T17:55:21 < BrainDamage> baird: it can be atonomous during immersion
2012-10-26T17:55:27 < Erlkoenig> baird: unfortunately we don't have a factory for producing subs automatically, so it's handmade
2012-10-26T17:55:29 < BrainDamage> and resurface to exchange data
2012-10-26T17:55:37 < zyp> Erlkoenig, the part after the assembly expression in the inline assembly statements
2012-10-26T17:55:54 < BrainDamage> I don't know what's a c-card, but I got a diving license myself
2012-10-26T17:56:15 < Erlkoenig> zyp: uhm probably at another code snippet, not at switch_context() ?
2012-10-26T17:56:35 < Erlkoenig> because there you aren't using any R* registers
2012-10-26T17:56:42 < zyp> I mean this expression: asm volatile ("str sp, [%0]" :: "r" (&Thread::active_thread->sp));
2012-10-26T17:57:13 < zyp> I just tell the compiler that I need a register and refer to it by %0, and then the compiler allocates one and substitutes it
2012-10-26T17:57:29 < zyp> that's why it turn into r2 in the compiled code
2012-10-26T17:57:32 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-26T17:57:35 < Erlkoenig> aah okay
2012-10-26T17:57:37 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-26T17:57:40 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-26T17:57:43 < Erlkoenig> i think i should read some docs about that...
2012-10-26T17:57:45 < zyp> the compiler is free to choose any register there
2012-10-26T17:58:02 < Erlkoenig> you could also tell the compiler "i need 3 registers for internal work, give me 3"?
2012-10-26T17:58:09 < zyp> yes
2012-10-26T17:58:27 < Erlkoenig> what would happen if you request 13 registers? compiler error? ^^
2012-10-26T17:58:34 < zyp> you can say «I need three registers and I'm going to overwrite them»
2012-10-26T17:58:46 < zyp> and then the compiler will be aware of that
2012-10-26T17:58:52 < baird> BrainDamage: who with? I'm just PADI Open Water.
2012-10-26T17:59:06 < Erlkoenig> this statement then is only valid inside that very line of "asm volatile" right?
2012-10-26T17:59:19 < zyp> yes
2012-10-26T17:59:26 < BrainDamage> ESA, it's splitered off PADI in europe and has equivalent licenses
2012-10-26T17:59:35 < zyp> you can have several instructions inside one if you want
2012-10-26T17:59:39 < BrainDamage> got an open water license myself, last year
2012-10-26T17:59:48 < BrainDamage> ops, sorry for disruping the convo
2012-10-26T18:00:17 <+Steffann> zlog
2012-10-26T18:00:18 < zlog> Steffann: Log stored at http://tom-itx.dyndns.org:81/~tom-itx/irc/logs/%23%23stm32/2012-10-26.html
2012-10-26T18:00:24 < baird> For a robot sub, you'd have motion-detecting camera/video for flashing the fishes.
2012-10-26T18:01:35 < Erlkoenig> i know better: we need an EMP for disrupting the remote control units of other rc ships, so only our sub can operate, because it can run autonomeously
2012-10-26T18:03:31 <+Steffann> yay
2012-10-26T18:04:18 <+Steffann> That's why I don't like fully autonomous robots.. at least I don't like them when i can't see what they 'see'
2012-10-26T18:04:31 < Laurenceb> http://i3.kym-cdn.com/photos/images/original/000/167/130/tumblr_lhlk2uTbQJ1qemdfvo1_1280.jpg
2012-10-26T18:04:35 < Erlkoenig> then we just need to turn the motor to full power and bump and crash oder rc ships :3
2012-10-26T18:04:39 < Erlkoenig> *other
2012-10-26T18:04:58 < zyp> Steffann, do you get jealous when they can see something you can't? :p
2012-10-26T18:05:03 < BrainDamage> Laurenceb: it's middle east, what do you espect?
2012-10-26T18:05:12 < baird> Evil Dead 3: "Ramming Speed!"
2012-10-26T18:05:16 < Laurenceb> lol
2012-10-26T18:05:20 <+Steffann> Sure zyp
2012-10-26T18:05:30 < Laurenceb> more like death race 2000
2012-10-26T18:05:37 < Laurenceb> greatest film ever
2012-10-26T18:05:40 <+Steffann> No, I want to see live 'footage' zyp :)
2012-10-26T18:05:51 < Laurenceb>  - original not the stupid remake
2012-10-26T18:06:21 < Erlkoenig> we have already tiny supercheap camera modules that are said to be accessible by the DCIM. we could record the pictures or superslowly transmit them
2012-10-26T18:06:51 < baird> Radio transmission underwater is a lot of pain (well, money..)
2012-10-26T18:06:59 < karlp> wel, power :)
2012-10-26T18:07:26 < BrainDamage> you want studpidly low freqs to avoid high attenuation
2012-10-26T18:07:32 < Erlkoenig> it's said to work when only going a few meters deep
2012-10-26T18:07:40 < Laurenceb> http://www.youtube.com/watch?v=o6LahtrxVhg
2012-10-26T18:09:07 < baird> I was thinking a goal would be the 100 metre mark.. Something that could go further than the Advance Deep Diver ex-girlfriend. :P :)
2012-10-26T18:09:39 < karlp> not that you're having any isses getting over her or anything...
2012-10-26T18:09:44 <+Steffann> char baird ?
2012-10-26T18:10:09  * karlp hi5s Laurenceb, deathrace 2000 was fucking awesome
2012-10-26T18:10:10 < baird> Charl*otte
2012-10-26T18:10:17 < Erlkoenig> baird: well... our plastic pipes won't withstand the pressure of 100m :D
2012-10-26T18:10:17 <+Steffann> Yes, Char
2012-10-26T18:10:29 < Erlkoenig> we're going for like 2-3m
2012-10-26T18:10:35 < BrainDamage> you have 1atm per m
2012-10-26T18:10:52 < BrainDamage> err per 10m
2012-10-26T18:10:54 < BrainDamage> assuming the internal is pressurized at 1atm
2012-10-26T18:11:01 < BrainDamage> you'll have only 10 atm
2012-10-26T18:11:45 < BrainDamage> consider that a thick plastic pet bottle for soft drinks can hold 30 or so before rupturing
2012-10-26T18:12:16 < BrainDamage> it's not an unrelistic goal
2012-10-26T18:12:29 < Erlkoenig> hm. but there are weaker points, like the connection point of the pipes... well i guess we just have to test it :D
2012-10-26T18:12:44 < baird> Simpler to make it floodable; would make boyancy a lot easier to control.
2012-10-26T18:12:44 <+Steffann> Trial and error :)
2012-10-26T18:13:42 < Erlkoenig> well uhm we're planning for some flooadable tank
2012-10-26T18:14:06 < baird> Erlkoenig: what amount of power is generally needed to propel the uboot?
2012-10-26T18:14:23 < BrainDamage> depends on the speed
2012-10-26T18:15:29 < baird> On the scale of watts, 1/10 watts, ..?
2012-10-26T18:15:49 < Erlkoenig> corret :D ... we're using a 12V DC Motor, it uses 2.5A in "Idle" and 12.5A under max.load
2012-10-26T18:16:00 < Erlkoenig> in the video it was taking like ~6A
2012-10-26T18:16:25 < Erlkoenig> you see the waves it generates... probably enough propulsion to give it some speed
2012-10-26T18:16:50 < BrainDamage> do you have depth controls?
2012-10-26T18:17:10 < BrainDamage> because getting something to be neutrally buoyant at various depths is surprsingly difficult
2012-10-26T18:17:26 < BrainDamage> so you'll need both thrust vectoring and buoyancy control
2012-10-26T18:17:54 < baird> An alternative to flooding is to fill the sub with something like mineral oil..
2012-10-26T18:18:24 < baird> Re: bouyancy -- there's these things called MEMS now, maybe you've heard of them. :)
2012-10-26T18:19:14 < Erlkoenig> we have 4 rudders at the end, these should be good for dynamic depth controls... at least when we got the accelerometer filtering working... and then, for the static depth, we have that floodable tank
2012-10-26T18:19:25 < BrainDamage> sounds good
2012-10-26T18:19:54 < Erlkoenig> at first we'll try to control everything manually to see how difficult it is... then write some algorithms to heat the F4's FPU a bit ;)
2012-10-26T18:20:16 < BrainDamage> baird: I was asking about actuators, not sensors
2012-10-26T18:21:35 < Erlkoenig> with the new switching regulator we should finally be able to control the servos properly without them going totally crazy...
2012-10-26T18:24:45 < Erlkoenig> d'oh. the bit banding stuff could be used to write semaphores for multithreading systems on the STM32, right?
2012-10-26T18:25:19 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-26T18:25:23 < zyp> depends
2012-10-26T18:25:26 < BrainDamage> so you want to reinvent an RTOS ?
2012-10-26T18:26:06 < zyp> actually, bitbanding doesn't help at all
2012-10-26T18:26:08 < Erlkoenig> BrainDamage: i just want to fiddle around a bit
2012-10-26T18:26:36 < Erlkoenig> zyp: because you can't read+conditional execution - modify - write?
2012-10-26T18:26:58 < zyp> exactly, it's no different from writing a normal word in an atomic fashion
2012-10-26T18:27:35 < zyp> and we already have instructions for helping with stuff like semaphores
2012-10-26T18:27:39 < Erlkoenig> hm. for locks one probably always needs if (!locked) { locked = true; ... } .... and need to be sure that this never gets interrupted
2012-10-26T18:27:54 < zyp> see http://cgit.jvnv.net/laks/tree/os/mutex.h for instance
2012-10-26T18:29:12 < Erlkoenig> ah, nice
2012-10-26T18:30:24 < jpa-> you can do locks with bit banding, but it's easier to just use the exclusive access instructions
2012-10-26T18:31:13 < Erlkoenig> jep, and probably more portable...
2012-10-26T18:31:21 < Erlkoenig> because the bitbanding is STM32-specific
2012-10-26T18:31:29 < jpa-> it's not STM32-specific
2012-10-26T18:32:03 < jpa-> but yeah, for portability using the exclusive access instructions through GCC atomic stuff is the best
2012-10-26T18:32:46 < jpa-> (and CCM region in STM32F4 doesn't have bitband, so it gets a bit annoying if your stack/heap is there)
2012-10-26T18:40:40 < Erlkoenig> get functional programming :3
2012-10-26T18:41:31 < jpa-> does it solve all my problems?
2012-10-26T18:42:02 < Erlkoenig> of course, even world peace
2012-10-26T18:42:21 < jpa-> nice
2012-10-26T18:42:28 < jpa-> will it make Steffann my friend?
2012-10-26T18:43:05 < Erlkoenig> if correctly used, it will make everyone your friend - or just kill everybode
2012-10-26T18:43:44 < Erlkoenig> where the acutal outcome is determined by your mathematical wizardness
2012-10-26T18:45:19 < jpa-> the second outcome might be quite acute
2012-10-26T18:51:31 -!- BjoernC_ [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-26T18:54:28 < Erlkoenig> why isn't there a __SVC() C Funtion like there is a __WFE() and __WFI() function...
2012-10-26T18:58:01 < jpa-> who cares when you can just type asm("svc")?
2012-10-26T18:58:48 < jpa-> besides, usually you want an argument to the svc instruction
2012-10-26T18:59:02 < jpa-> which is not so easy with C function, as IIRC svc instruction only takes literal args
2012-10-26T18:59:47 < Erlkoenig> yes it does... okay that explains it
2012-10-26T19:00:21 < Erlkoenig> the argument is encoded in the instruction but completely ignored by the core - the SVCall handler can read out the svc instruction from the memory and decode the argument right?
2012-10-26T19:00:30 < jpa-> yeah
2012-10-26T19:00:52 < jpa-> which leaves the registers free for the "real" system call args
2012-10-26T19:01:17 < Erlkoenig> oh and why does my GCC refuse "asm" but accepts "__asm__"...
2012-10-26T19:02:57 < jpa-> i think depends on which -std= setting you are using
2012-10-26T19:02:59 < jpa-> +it
2012-10-26T19:05:20 -!- baird [~cjb@ppp121-44-232-241.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-26T19:05:33 < Erlkoenig> oh, -std=c99
2012-10-26T19:06:03 < jpa-> -std=gnu99 probably allows asm(), and other gcc-specific extensions
2012-10-26T19:06:51 < HTT-Bird> yeah
2012-10-26T19:06:53 < HTT-Bird> that'd be it
2012-10-26T19:07:01 < Erlkoenig> ah so __asm__ is "standard" and asm is GNU-specific?
2012-10-26T19:08:58 < HTT-Bird> Erlkoenig: asm is GNU specific
2012-10-26T19:09:22 < HTT-Bird> __asm__ is an 'escape' provided to allow it to be used in code compiled in a standards-conforming mode
2012-10-26T19:10:39 < Erlkoenig> ah okay, so other compilers probably don't have inline assembly at all?
2012-10-26T19:18:19 < HTT-Bird> some do (I've used it in MSVC) but the syntax is different
2012-10-26T19:20:58 < Erlkoenig> okay so i can abandon compatibility at all and just use gnu99  and "asm"...
2012-10-26T19:21:17 -!- barthess [~barthess@185.6.25.219] has joined ##stm32
2012-10-26T19:36:55 -!- quitte [~quitte@tplink.ghb.hs-furtwangen.de] has joined ##stm32
2012-10-26T19:38:17 < quitte> hi. i'm a little confused about tbhe stm32 uC range. is there one with sdio and high speed usb for less than 7€ ?
2012-10-26T19:38:50 < jpa-> probably not, as you need an external PHY for high-speed USB
2012-10-26T19:38:59 < jpa-> (which immediately raises the price)
2012-10-26T19:40:33 < quitte> true. but so far the uCs themswelves I found where more in the 15€ range
2012-10-26T19:40:45 < jpa-> probably true
2012-10-26T19:41:34 < quitte> sigh. always the choosing of 2
2012-10-26T19:43:58 <+Steffann> The F2 seems to have HS (ext. phy) + SDIO .. for ~6.80
2012-10-26T19:44:01 <+Steffann> euro that is
2012-10-26T19:44:14 <+Steffann> ext. vat and crap
2012-10-26T19:44:19 <+Steffann> *excl.
2012-10-26T19:45:56 <+Steffann> STM32F205RB the smallest and probably cheapest you can get ..
2012-10-26T19:49:03 < quitte> Steffann: that sounds reasonable. where did you find that price?
2012-10-26T19:49:04 < zyp> you need to check that both ULPI and SDIO is available at the same time for that package too
2012-10-26T19:49:24 < quitte> heh. thanks for reminding me.
2012-10-26T19:49:37 <+Steffann> farnell
2012-10-26T19:49:48 < zyp> ULPI consumes 12 pins
2012-10-26T19:49:48 < quitte> .de ?
2012-10-26T19:49:53 <+Steffann> nl
2012-10-26T19:50:00 < jpa-> nerdlandia
2012-10-26T19:50:02 <+Steffann> digikey is 7.13$ which is probably less that 6.81 in euro
2012-10-26T19:50:17 < quitte> Steffann: per 100pc ?
2012-10-26T19:50:28 <+Steffann> 1
2012-10-26T19:50:35 < quitte> i'm looking for singles. but that price might still be okay
2012-10-26T19:50:48 < quitte> Steffann: how do i find the phy?
2012-10-26T19:50:54 < zyp> if highspeed usb is a deciding feature, you might want to choose another vendor with built in highspeed phy
2012-10-26T19:51:11 <+Steffann> zyp is more familier with stuff like that
2012-10-26T19:51:24 < quitte> zyp: it's both sdio and high speed
2012-10-26T19:51:36 < quitte> zyp: i'd be glad for suggestions
2012-10-26T19:51:44 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-26T19:52:14 < quitte> ti and microchip don't have a match
2012-10-26T19:52:30 < zyp> I don't have experience with anything else yet, but I believe both sam3u and the lpc range of microcontrollers have that
2012-10-26T19:52:41 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has joined ##stm32
2012-10-26T19:54:03 < jpa-> quitte: are you building a usb card reader? :P
2012-10-26T19:54:21 < quitte> jpa-: pretty much, yes.
2012-10-26T19:54:53 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-26T19:55:25 -!- dekar [~dekar@dslb-178-004-019-143.pools.arcor-ip.net] has quit [Quit: This computer has gone to sleep]
2012-10-26T19:55:29 < zyp> my lpc4300 toy project will feature both sdio and hs usb
2012-10-26T19:55:58 < zyp> but that's the upper end of the lpc range, I believe the cheaper lpc17xx microcontrollers also have some with built in hs phy
2012-10-26T19:56:14 < emeb> how's that lpc project coming along?
2012-10-26T19:56:16 < upgrdman_> what is a phy?
2012-10-26T19:56:23 < zyp> physical interface
2012-10-26T19:56:29 < upgrdman_> as in a port?
2012-10-26T19:56:31 < zyp> it handles usb analog levels
2012-10-26T19:56:32 < zyp> yes
2012-10-26T19:56:50 < upgrdman_> why would a uc have a port built onto it
2012-10-26T19:57:02 < zyp> stm32f2 and f4 only has ULPI for hs usb, which is a 12-signal interface towards a phy
2012-10-26T19:57:12 < zyp> sorry, I read part
2012-10-26T19:57:36 < upgrdman_> o. so a phy is a middleman that makes it easier for the uc to utilize the port?
2012-10-26T19:57:55 < zyp> the phy translates between the usb analog line signalling and the 12-signal ulpi bus, yes
2012-10-26T19:58:22 < zyp> you can compare it a bit to max232 for rs232 signalling, if you are familiar with that
2012-10-26T19:58:30 < zyp> you can call max232 a rs232 phy
2012-10-26T19:59:35 < zyp> http://www.smsc.com/index.php?pid=28&tid=143 <- my stm32f4 devboard is using this part
2012-10-26T19:59:42 < zyp> it's $2 or something
2012-10-26T19:59:52 < upgrdman_> k
2012-10-26T20:00:29 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-26T20:01:43 < emeb> hmmm... Parallella kickstarter has 29 hours to go and is $107k short.
2012-10-26T20:01:51 < emeb> can't say I'm surprised.
2012-10-26T20:02:32 < Erlkoenig> para what :D
2012-10-26T20:02:39 < emeb> http://www.kickstarter.com/projects/adapteva/parallella-a-supercomputer-for-everyone
2012-10-26T20:03:11 < Erlkoenig> ah hm
2012-10-26T20:03:23 < emeb> Basically a fabless IP outfit trying to get KS funding for a custom maskset.
2012-10-26T20:05:56 < zyp> I'm a bit tempted to throw $99 at them
2012-10-26T20:06:06 < zyp> as I understand it, the board will include a zynq
2012-10-26T20:06:10 < Erlkoenig> wah. why does the STM32 usb (host/device) library contain the CMSIS and the StdPeriph drivers... aren't these supposed to ONLY be in individual packages oO
2012-10-26T20:06:19 < Erlkoenig> zynq :o
2012-10-26T20:07:08 < Erlkoenig> well zynq is an argument :D
2012-10-26T20:07:39 < zyp> or are they redoing that part from the prototype perhaps
2012-10-26T20:07:42 < emeb> cheaper than a digilent zedboard
2012-10-26T20:07:42 < jpa-> Erlkoenig: you seem to have unrealistic hopes for ST's libraries
2012-10-26T20:07:51 < Erlkoenig> meh
2012-10-26T20:07:57 < upgrdman_> whats a zynq
2012-10-26T20:08:06 < Erlkoenig> upgrdman_: most awesome chip ;)
2012-10-26T20:08:10 < Erlkoenig> didn't they learn from the windows library mess...
2012-10-26T20:08:15 < emeb> xilinx fpga w/ cortex Ax on-chip
2012-10-26T20:08:16 < zyp> arm-soc with fpga
2012-10-26T20:08:20 < upgrdman_> mmmm
2012-10-26T20:08:26 < zyp> or fpga with arm hardcores
2012-10-26T20:08:40 < zyp> depending on how you look at it :p
2012-10-26T20:08:41 < TitanMKD> hmm yes this zynq low cost board seems to be a must ;)
2012-10-26T20:08:54 < upgrdman_> how low cost
2012-10-26T20:09:01 < emeb> $99
2012-10-26T20:09:22 < emeb> it's a raspberry pi killer. :P
2012-10-26T20:09:32 < upgrdman_> lol
2012-10-26T20:09:45 < TitanMKD> yes for 99USD ;)
2012-10-26T20:09:49 < zyp> oh, it will indeed include a zynq
2012-10-26T20:09:51 < zyp> ok, pledging
2012-10-26T20:09:52 < TitanMKD> i only want the zynq for 99USD
2012-10-26T20:09:54 < upgrdman_> im tempted to buy a r pi and have it replace my web server
2012-10-26T20:10:08 < upgrdman_> and hookup a QVGA lcd to display random info
2012-10-26T20:10:16 < TitanMKD> i do not really need the Epiphany stuff
2012-10-26T20:10:18 < upgrdman_> then put it all in a project box
2012-10-26T20:10:37 < TitanMKD> zyp are you sure it's the zynq ?
2012-10-26T20:10:42 < zyp> says so
2012-10-26T20:10:43 < TitanMKD> zyp it's not very clear on the page
2012-10-26T20:10:50 < emeb> can you pledge KS w/ paypal?
2012-10-26T20:11:04 < zyp> no, amazon payments
2012-10-26T20:11:26 < emeb> well. that decides it for me.
2012-10-26T20:11:35 < TitanMKD> ha yes for 99USD 2 ARM-A9 cores and 16 Epiphany cores.
2012-10-26T20:12:13 < zyp> ok, pledged
2012-10-26T20:12:29 < TitanMKD> but the question is does it will include USB PHY and Ethernet PHY ?
2012-10-26T20:12:52 < TitanMKD> i know without PHY it's not usable but maybe it's not integrated
2012-10-26T20:13:09 < zyp> read the planned specs for the board
2012-10-26T20:13:21 < zyp> both usb and gigabit ethernet is planned
2012-10-26T20:14:03 < TitanMKD> zyp me to ordered for 119USD for international shipping
2012-10-26T20:15:02 < emeb> they've got $103k to go and 28 hours to do it. It's gone up 4k in the last 15 min I've been watching. About $16k/h, so at that rate they'll make it.
2012-10-26T20:15:13 < zyp> no, it's less than $100k now
2012-10-26T20:15:45 < emeb> quibbles
2012-10-26T20:15:47 < zyp> $92k left or so according to my count here
2012-10-26T20:16:11 < emeb> mine still says $646k total & I just refreshed. wth?
2012-10-26T20:16:33 < zyp> cache I guess
2012-10-26T20:17:03 < emeb> yep: shift-refresh - now it's $658
2012-10-26T20:17:31 < emeb> looks like they're on track to squeak past the goal.
2012-10-26T20:18:32 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-26T20:18:38  * Steffann wants that board
2012-10-26T20:19:52 < emeb> so pledge $99 and it's yours!
2012-10-26T20:21:31 <+Steffann> Don't have a CC
2012-10-26T20:22:33 < emeb> See - once again, if they had more payment options then they'd open up more pledges.
2012-10-26T20:22:41 -!- daku is now known as DaKu
2012-10-26T20:22:52 <+Steffann> ks doesnt do paypal does it?
2012-10-26T20:23:00 < emeb> apparently not
2012-10-26T20:23:07 < emeb> amazon payments only
2012-10-26T20:25:21 < emeb> $662k now - only $88k to go
2012-10-26T20:25:37 <+Steffann> You pledged?
2012-10-26T20:25:50 < emeb> nope - only if they had paypal.
2012-10-26T20:26:06 <+Steffann> Thought you were from yankieland
2012-10-26T20:26:08 <+Steffann> aka CC land
2012-10-26T20:26:42 < emeb> I am. It's just that stuff like that gets paid out of the pp acct because that's all hobby "mad money".
2012-10-26T20:27:16 < zyp> do you guys store money in your paypal accounts?
2012-10-26T20:27:23 < emeb> I do.
2012-10-26T20:27:47 < emeb> I get paid for some of my side work thru pp. That money stays there for other hobby / side project expenses.
2012-10-26T20:27:53 < Erlkoenig> and why-the-heck are there EXAMPLE CODES buried in the library stdperiph directory structure oO
2012-10-26T20:28:03 < emeb> why not?
2012-10-26T20:28:07 < zyp> emeb, right, that makes sense
2012-10-26T20:28:34 < Erlkoenig> they should be shipped seperately
2012-10-26T20:28:51 < Erlkoenig> so automagic build scripts don't compile them into the application using the lib
2012-10-26T20:29:07 < emeb> whose build scripts do that?
2012-10-26T20:29:41 < zyp> «hey my dear IDE, have all my sources and do some magic for me!»
2012-10-26T20:29:44 < Erlkoenig> these who don't automatically recognize dir names "examples"
2012-10-26T20:29:58 < Erlkoenig> i.e. IDE's
2012-10-26T20:30:12 < emeb> I miss out on so much by not using IDEs.
2012-10-26T20:30:14 < Thorn> LOL
2012-10-26T20:30:24 < zyp> emeb, so do I, and I like it
2012-10-26T20:30:26 < zyp> :p
2012-10-26T20:30:56 < Erlkoenig> emeb: you typed all the 500 filenames of the CMSIS lib into your makefile manually?
2012-10-26T20:31:14 < zyp> Erlkoenig, if you feel my lib is better, you're free to use it
2012-10-26T20:31:15 < emeb> Erlkoenig: no - I just pull in the files I need.
2012-10-26T20:31:44 < Erlkoenig> emeb: these are already quite a few... seems more convenient to just pull in all
2012-10-26T20:32:33 < emeb> Erlkoenig: you end up doing some work either way. You spend some time figuring out what you need, or you hassle with all the stuff you don't need when the auto tools screw up.
2012-10-26T20:33:06 < Erlkoenig> the auto tools would work if the library was organized in an useful way... or it hat some makefiles or similar -.-
2012-10-26T20:33:21 -!- barthess [~barthess@185.6.25.219] has quit [Read error: Connection reset by peer]
2012-10-26T20:33:30 < emeb> adapt or die. :P
2012-10-26T20:33:34 < jpa-> Erlkoenig: i agree that the directory structure in the ST libs is quite annoying
2012-10-26T20:33:39 < jpa-> but it's not the worst of it
2012-10-26T20:33:56 < jpa-> you'll just might find that the code is much worse :)
2012-10-26T20:33:59 < zyp> my approach is having the lib integrate with the build system to select required files for the desired setup, so that the library user don't have to think about it
2012-10-26T20:34:29 < Erlkoenig> zyp: seems similar to just include everything and have the "unused function optimizer" sort it out
2012-10-26T20:35:08 < zyp> until you have one library that can be compiled for different targets requiring different source files
2012-10-26T20:35:36 < zyp> I'm already supporting both F1 and F4, and just started adding F3 support
2012-10-26T20:35:55 < zyp> my goal was to have the CDC demo run on F3 discovery tonight, but then I started wasting time
2012-10-26T20:36:11 < Erlkoenig> okay that will screw up everything
2012-10-26T20:36:24 < Erlkoenig> if there were a makefile....
2012-10-26T20:37:15 < emeb> Erlkoenig: why not just move the "Projects" dir out of the library hierarchy? Isn't that where all the examples are?
2012-10-26T20:37:41 < zyp> Erlkoenig, I've solved it like this: http://cgit.jvnv.net/laks_demo/tree/SConstruct
2012-10-26T20:38:05 < Erlkoenig> emeb: no, in the CMSIS/DSP dir are more examples
2012-10-26T20:38:13 < zyp> the SelectMCU rule pulls in the required files for the selected mcu
2012-10-26T20:38:16 < Erlkoenig> emeb: i've removed them now... but its annoying nevertheless
2012-10-26T20:38:19 < emeb> Erlkoenig: ah. messy.
2012-10-26T20:38:32 < Erlkoenig> zyp: ah nice
2012-10-26T20:38:55 < Erlkoenig> zyp: alternative would be #ifdef USING_CPU_STM32F407   at the beginning of the respective files
2012-10-26T20:38:55 < zyp> and also sets flags so it's compiled for the right architecture with or without fpu and so on
2012-10-26T20:39:19 < zyp> and defines like that
2012-10-26T20:40:06 < zyp> http://cgit.jvnv.net/laks/tree/usb/usb.h <- in the headers I'm using defines to distinguish between them
2012-10-26T20:40:45 < zyp> but I can do either, depending on what is best suited for the particular files
2012-10-26T20:40:58 < Erlkoenig> your lib replaces stdperiph and the usb lib?
2012-10-26T20:41:02 < zyp> and the user application doesn't have to care about it, because it's all done in the build system
2012-10-26T20:41:06 < zyp> yep
2012-10-26T20:42:46 < zyp> it was originally part of my quadrotor controller firmware, but kind of grew out of it, so I forked it out so it can be used for different stuff as well
2012-10-26T20:43:11 < Erlkoenig> hmm
2012-10-26T20:43:21 < Erlkoenig> is it complete, i.e. covers all the F4's periphals and stuff?
2012-10-26T20:43:35 < zyp> no, just the ones I've cared to implement yet
2012-10-26T20:44:38 < Erlkoenig> okay. :D
2012-10-26T20:45:18 < zyp> the usb stuff is probably the largest part
2012-10-26T20:45:52 < zyp> and to be honest it's still rather experimental
2012-10-26T20:46:09 < Erlkoenig> but why reimplement the whole USB stuff when there are the ST libs for that?
2012-10-26T20:46:33 < zyp> I haven't fully decided what the best design for the entire API is, so I'm doing the demos to get some more experience with that
2012-10-26T20:46:38 < zyp> because it's fun
2012-10-26T20:46:53 < zyp> and because I like working with nice code
2012-10-26T20:47:02  * emeb needs to try laks at some point.
2012-10-26T20:47:09 < Erlkoenig> ah. :D
2012-10-26T20:47:15 < emeb> but that means becoming more proficient w/ C++.
2012-10-26T20:48:16 < emeb> Std Periph is the path of least resistance. But it's uuuuugly.
2012-10-26T20:48:19 < zyp> I hate having to write a whole novel's worth of code just to be able to achieve something simple
2012-10-26T20:48:36 < zyp> because «write» in that case is more like cut and paste
2012-10-26T20:50:07 < zyp> if it's a common thing to do, I want to have a nice simple function to do it, not ten pages of boilerplate
2012-10-26T20:50:29 < Laurenceb> interesting
2012-10-26T20:50:36 < Laurenceb> just tested my logger on a new pcb
2012-10-26T20:50:43 < Laurenceb> that works with 18mhz spi
2012-10-26T20:50:54 < Laurenceb> the speed is still stuck at ~750KB/s
2012-10-26T20:51:06 < Laurenceb> looks like the usb is bottlenecking
2012-10-26T20:51:23 < Erlkoenig> zyp: maybe the ST stdperiph libs are autogenerated :D
2012-10-26T20:51:26 < emeb> zyp: makes sense to me.
2012-10-26T20:51:31 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has joined ##stm32
2012-10-26T20:51:32 < Erlkoenig> zyp: or made by robots
2012-10-26T20:51:54 < zyp> the usb descriptor stuff is kind of neat in that regard… considering how I've written it now, I can make a simple function that would instance all the descriptors required to make a CDC ACM interface
2012-10-26T20:52:07 < zyp> and if I want two of those, I just call it twice
2012-10-26T20:54:31 < zyp> but I'm not sure I'm happy with how the class driver and the descriptors are not connected at all in the code currently, so I might do something about that part later
2012-10-26T20:59:16 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 248 seconds]
2012-10-26T21:10:40 < emeb> just checked pricing of Xilinx Zynq chips: 1000pc qty price of the low-end part is $227/ea.
2012-10-26T21:11:07 < emeb> So that $99 Parallella devboard is a steal. If they actually get around to making them. :)
2012-10-26T21:11:53 < zyp> :p
2012-10-26T21:13:40 < emeb> New Xilinx chips are always expensive.
2012-10-26T21:13:46 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Quit: my exit]
2012-10-26T21:13:50 -!- barthess [~barthess@5.100.211.45] has joined ##stm32
2012-10-26T21:15:06 < emeb> I've got a couple wireless transmitter boards here that we built a few years ago with 2x Xilinx Virtex5 parts on them. The FPGAs cost $2500/ea.
2012-10-26T21:15:19 < zyp> :)
2012-10-26T21:15:38 < zyp> so, you're rather experienced with FPGA stuff?
2012-10-26T21:15:51 < emeb> Been doing FPGAs for > 20 years.
2012-10-26T21:16:12 < HTT-Bird> o/ emeb
2012-10-26T21:16:25 < emeb> :)
2012-10-26T21:16:49 < zyp> nice
2012-10-26T21:17:36 < TitanMKD> emeb hehe so this Zynq seems a beast
2012-10-26T21:17:44 < TitanMKD> especially for 99USD
2012-10-26T21:17:44 < emeb> It's impressive.
2012-10-26T21:18:09 < zyp> I've been able to read for > 20 years
2012-10-26T21:18:10 < TitanMKD> I hope they will not take money and run ;)
2012-10-26T21:18:21 < zyp> and I'm pretty good at reading, so you must be pretty good at FPGAs then
2012-10-26T21:18:41 < emeb> zyp: Well enough that it's what keeps dinner on the table.
2012-10-26T21:19:02 < zyp> what's your opinion on verilog vs vhdl?
2012-10-26T21:19:11 < emeb> zyp: I use both.
2012-10-26T21:19:22 < zyp> you're still allowed to have an opinion
2012-10-26T21:19:25 < emeb> I prefer verilog because it's less verbose.
2012-10-26T21:19:32 < zyp> :)
2012-10-26T21:19:42 < emeb> the old C vs Ada arguement.
2012-10-26T21:19:47 < emeb> -e
2012-10-26T21:19:56 < zyp> that's why I also picked verilog when I toyed with that
2012-10-26T21:20:07 < emeb> VHDL's strong typing is serious PITA.
2012-10-26T21:20:08 < Thorn> how come you aren't in ##fpga?
2012-10-26T21:20:30 < emeb> Thorn: just never got around to it.
2012-10-26T21:20:40 < HTT-Bird> emeb: yeah, Pascal (and by extension Ada/VHDL) are strong typing done wrong
2012-10-26T21:20:46 < HTT-Bird> if you ask me
2012-10-26T21:21:22 < HTT-Bird> if you want to do strong typing right, give us programmers the ability to transform and compose types efficiently
2012-10-26T21:22:22 < emeb> VHDL suffers from depending on a lot of 3rd-party libraries that are not well documented.
2012-10-26T21:23:14 < emeb> makes designs less portable IMHO.
2012-10-26T21:28:58 < emeb> Zynq would be a lot more attractive if the price weren't so high. Dual 800MHz A9 processors w/ all the usual peripherals + 400k-5.3M gates of FPGA would be nice, but they're charging a high premium for the uniqueness.
2012-10-26T21:29:30 < emeb> You could throw down a mid-range ARM SoC + mid-sized FPGA on a board for a lot less than $200.
2012-10-26T21:30:18 < zyp> yeah
2012-10-26T21:31:46 < Thorn> doesn't altera have something similar?
2012-10-26T21:32:25 < Thorn> http://www.altera.com/devices/processor/arm/cortex-a9/m-arm-cortex-a9.html
2012-10-26T21:33:16 < emeb> Of course they do - Xilinx & Altera are the classic tit-for-tat arms race. (ooh - a pun!)
2012-10-26T21:33:34 < zyp> I like puns
2012-10-26T21:33:35 < zyp> :D
2012-10-26T21:34:27 < Erlkoenig> why is there NVIC stuff in the StdPeriph library - doesn't that belong to the CMSIS?
2012-10-26T21:41:40 < emeb> NVIC stuff is all in misc.c/h
2012-10-26T21:42:54 < emeb> Thorn: Looks like Altera's A9+FPGA SoCs are the Cyclone V SE/SX series. AFAICT those are not in distribution yet.
2012-10-26T21:43:28 < Erlkoenig> emeb: which is in the stdperiph dir...
2012-10-26T21:44:29 < emeb> yup
2012-10-26T22:05:06 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-26T22:05:06 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Disconnected by services]
2012-10-26T22:09:11 < karlp> Erlkoenig: nvic also has multiple optional parts, like how many bits to use and so on
2012-10-26T22:09:55 < Erlkoenig> well uhm they could define the registers in the CMSIS, if they don't exist in a specific implementation and you try to access them, you get a MemFault or something
2012-10-26T22:10:57 < jpa-> but the stdperiph way is to use poorly documented wrapper functions instead of the registers
2012-10-26T22:11:13 < Erlkoenig> thats correct :D
2012-10-26T22:11:23 < Erlkoenig> but shouldn't ARM put that NVIC stuff into the CMSIS?
2012-10-26T22:11:29 < Erlkoenig> not ST
2012-10-26T22:12:16 < jpa-> the registers are there already
2012-10-26T22:13:33 < Erlkoenig> yeah? where e.g. is the NVIC_IPR0 ?
2012-10-26T22:14:55 -!- DaKu is now known as daku
2012-10-26T22:15:30 < jpa-> NVIC->IP[0]
2012-10-26T22:15:49 < Erlkoenig> ah, a different name than in the doc, how convenient
2012-10-26T22:16:41 < jpa-> well, with CMSIS, just always go to XXX_Type in the header files to find what members it has 
2012-10-26T22:17:17 < jpa-> it's quite reasonable to have the interrupt priority registers defined as an array
2012-10-26T22:17:33 < Erlkoenig> they might have used an union or something...
2012-10-26T22:18:05 < jpa-> and typed uint32_t IPR0; uint32_t IPR1; uint32_t IPR2; ... uint32_t IPR240;?
2012-10-26T22:18:10 < jpa-> sounds smart
2012-10-26T22:18:34 < Erlkoenig> codegeneration :D
2012-10-26T22:18:39 < jpa-> "yay"
2012-10-26T22:18:57 < Erlkoenig> well then at least the doc could have a hint about how to find the register.
2012-10-26T22:19:12 < zyp> I have volatile uint8_t IPR[2816];
2012-10-26T22:19:15 < zyp> :D
2012-10-26T22:19:34 < jpa-> Erlkoenig: doesn't your fancy GUI autocomplete? :P
2012-10-26T22:20:25 < Erlkoenig> for NVIC->IP ? yes, but how am i supposed to know what array subscript i need... i'd have to do calculations with the adresses and stuff
2012-10-26T22:20:30 < jpa-> zyp: good catch, it's uint8_t actually :)
2012-10-26T22:21:24 < jpa-> Erlkoenig: uh, doesn't need much imagination to conclude that IP[0] is the priority for interrupt.. 0 :)
2012-10-26T22:21:47 < zyp> Erlkoenig, why does it matter?
2012-10-26T22:22:06 < zyp> I'm not accessing IPR outside the file it's defined in
2012-10-26T22:22:38 < Erlkoenig> jpa-: yeah but just writing that without verifying it's actually that easy?
2012-10-26T22:22:54 < zyp> http://cgit.jvnv.net/laks/tree/interrupt/interrupt.h <- I have a set_priority function for setting priority
2012-10-26T22:22:58 < Erlkoenig> hmm, what happens if you try to enable the SVC Interrupt?
2012-10-26T22:23:28 < zyp> enable how? it's enabled by default
2012-10-26T22:23:43 < Erlkoenig> via the appriopriate register, via the StdPeriph NVIC_Init function
2012-10-26T22:23:59 < Erlkoenig> yeah that's exactly the problem :D
2012-10-26T22:24:23 < zyp> remember that vectors <16 are treated special
2012-10-26T22:25:02 < Erlkoenig> yeah i know
2012-10-26T22:25:32 < zyp> what exactly are you asking about?
2012-10-26T22:25:35 < Erlkoenig> if i wanted to set the priority of the SVC exception, i'd use the NVIC_Init function, wich sets the priority - and enables the interrupt
2012-10-26T22:25:52 < Erlkoenig> would that enabling be just a "NOP" or a MemFault?
2012-10-26T22:26:09 < zyp> SVC priority is not set in NVIC_IPR, it's set in SCB_SHPR
2012-10-26T22:27:28 < zyp> note how my file has two set_priority() functions
2012-10-26T22:27:34 < Erlkoenig> oh. uhm
2012-10-26T22:27:59 < Erlkoenig> what happens when i try to set priority of exception 11 = SVC?
2012-10-26T22:28:27 < zyp> with my stuff? it sets the SVC priority
2012-10-26T22:28:35 < Erlkoenig> no, with StdLib
2012-10-26T22:28:38 < Erlkoenig> argh
2012-10-26T22:28:41 < Erlkoenig> StdPeriph-lib
2012-10-26T22:28:50 < Erlkoenig> it will try to use NVIC->IP[11]
2012-10-26T22:29:06 < zyp> then you are using the wrong function for the purpose
2012-10-26T22:29:21 < zyp> IRQ 11 != Exception 11
2012-10-26T22:29:37 < Erlkoenig> really? wah
2012-10-26T22:31:55 < zyp> IP[11] will set the priority of handler 27
2012-10-26T22:32:40 < Erlkoenig> ah. makes sense.
2012-10-26T22:33:24 < zyp> that's why I have two seperate enums in my code
2012-10-26T22:35:38 < jpa-> Erlkoenig: actually CMSIS defines negative numbers for system handlers, like  SVCall_IRQn = -5
2012-10-26T22:36:00 < zyp> that's also a way of doing it
2012-10-26T22:36:08 < jpa-> but some functions like NVIC_Init only work with interrupts, while others like NVIC_SetPriority work with both
2012-10-26T22:37:20 < Erlkoenig> ah ohmygod
2012-10-26T22:37:38 < Erlkoenig> the CMSIS core_m4.h file does it exactly that way
2012-10-26T22:38:25 < Erlkoenig> so "NVIC_SetPriority(SVCall_IRQn, 15);" should just do it?
2012-10-26T22:39:22 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-26T22:39:25 < jpa-> maybe
2012-10-26T22:39:32 < jpa-> check the code to be sure
2012-10-26T22:40:03 < jpa-> or just use something else, those wrapper functions are poorly documented bloatcrap
2012-10-26T22:41:07 < Erlkoenig> http://pastebin.com/NKWBiHVR
2012-10-26T22:41:16 < Erlkoenig> looks like it should do
2012-10-26T22:41:22 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has joined ##stm32
2012-10-26T22:41:30 < Erlkoenig> so then "svc 0" should trigger the interrupt right?
2012-10-26T22:41:33 < Laurenceb_> sup
2012-10-26T22:46:28 -!- zlog [~zlog@ip68-102-196-2.ks.ok.cox.net] has quit [Ping timeout: 248 seconds]
2012-10-26T22:50:15 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has joined ##stm32
2012-10-26T22:50:59 < karlp> man, proper function overloading is awesome.
2012-10-26T22:51:06 < karlp> I'd like to use more c++ I think.
2012-10-26T22:52:38 < Erlkoenig> hahaha... yesterday someone was mourning that overloading is stupid :D
2012-10-26T22:53:11 < Erlkoenig> he also said overloading the [] operator in strings is stupid - std::string is an object, not an array, so he wants to call function on it, not access it like an array
2012-10-26T22:54:38 < HTT-Bird> Erlkoenig: :P
2012-10-26T22:55:05 < Erlkoenig> not like 1000 programming languages use [] on array objects, string objects, etc. to access elements
2012-10-26T23:05:57 < zyp> just you wait until I make an uint24_t with a packed struct with overloaded operators to act as an int :p
2012-10-26T23:06:14 < Erlkoenig> wah. :D
2012-10-26T23:06:47 < zyp> for some reason usb has a type like that, so I need it for packing into my descriptors :p
2012-10-26T23:07:35 < Thorn> but you don't need to actually use it in computations, do you
2012-10-26T23:07:36 < Erlkoenig> hm, when running in a main(), PRIMASK = 0, BASEPRI = 0, the "current priority" is like 16, so any interrupt has higher priority? ... and when BASEPRI=15, only interrupts with priority < 15 will be executed?
2012-10-26T23:10:08 -!- dekar [~dekar@212.255.27.61] has joined ##stm32
2012-10-26T23:10:09 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-26T23:17:45 < Erlkoenig> zyp: is there a reason that your "CPSIE i" is not at the end of your switch_context (), but before the POP?
2012-10-26T23:19:02 < zyp> because the pop is the return
2012-10-26T23:19:14 < Erlkoenig> aah because of "pc"
2012-10-26T23:19:19 < zyp> exactly
2012-10-26T23:19:47 < TitanMKD> ok i changed my pledge to 139USD to have the case ;)
2012-10-26T23:33:16 < emeb> whoa - only $44k to go
2012-10-26T23:33:35 < TitanMKD> yes it is possible
2012-10-26T23:33:44 < emeb> probable at this point.
2012-10-26T23:33:58 < emeb> kiss that $139 goodbye. :D
2012-10-26T23:34:05 < TitanMKD> i was disapointed by my pandaboard but with this one it seems very interesting
2012-10-26T23:34:17 < emeb> what didn't you like about the panda?
2012-10-26T23:34:28 < TitanMKD> 2D GFX acceleration suxx
2012-10-26T23:34:42 < TitanMKD> the system is just not usable too slow
2012-10-26T23:35:01 < TitanMKD> and each time they port it on new kernel and the 2D/3D GFX driver does not work anymore ...
2012-10-26T23:35:44 < TitanMKD> and it lacks stuff like there's on parallela ;)
2012-10-26T23:35:54 < emeb> Heh - been there. I've got a couple Beagleboards and a Beaglebone. The software churn is sometimes hard to keep up with.
2012-10-26T23:35:59 < TitanMKD> like Gigabits ethernet
2012-10-26T23:36:32 < TitanMKD> emeb it's the major problem with that dev board there's no stable version with all in one working immediatly
2012-10-26T23:36:55 < TitanMKD> emeb the worst was when i tried latest stable version of Pandaboard and Ethernet was just doing big crash
2012-10-26T23:37:09 < TitanMKD> Pandaboard Ubuntu
2012-10-26T23:37:34 < TitanMKD> the Linaro version of course
2012-10-26T23:37:59 < emeb> Never tried the various ARM Ubuntu flavors. I always use Angstrom.
2012-10-26T23:38:28 < TitanMKD> hehe i'm sure Angstrom is better and at least with shell it's fast ;)
2012-10-26T23:38:48 < TitanMKD> and the worst of all Full HD was never supported correctly on Pandaboard ;)
2012-10-26T23:39:40 < TitanMKD> but i doubt of Parallela power for GFX stuff
2012-10-26T23:40:10 < TitanMKD> they never speak about any GFX chipset for the display at least to have 2D acceleration
2012-10-26T23:40:46 < zyp> there is none
2012-10-26T23:41:01 < zyp> IIRC you have to do GPU in FPGA part of zynq if you want one
2012-10-26T23:41:08 < TitanMKD> yes ;)
2012-10-26T23:41:18 < TitanMKD> it seems it's what they are doing
2012-10-26T23:41:35 < TitanMKD> there's also no info on the Dual A9 speed
2012-10-26T23:41:36 < zyp> I saw some company doing IP for that
2012-10-26T23:41:45 < TitanMKD> yes but such IP are not free
2012-10-26T23:42:14 < zyp> http://www.logicbricks.com/Products/logi3D.aspx
2012-10-26T23:42:48 < zyp> The license fees offered through Xylon's Low-Volume IP Program (LVIP) start at €6,900 (< $9,900).
2012-10-26T23:42:51 < zyp> :D
2012-10-26T23:42:51 < TitanMKD> I also like the possibility to have 21GPIO with 1GB/s ;)
2012-10-26T23:43:10 < TitanMKD> or have an other batch of 21GPIO if you do not use HDMI
2012-10-26T23:44:18 < TitanMKD> zyp only one missing stuff is SATA
2012-10-26T23:44:38 < TitanMKD> but Gb Ethernet could save the stuff ;)
2012-10-26T23:50:26 < emeb> TitanMKD: Xilinx specs the Zynq A9 cores for 800MHz.
2012-10-26T23:51:26 < TitanMKD> ha ok
2012-10-26T23:54:06 < emeb> I'd love to play around with a Zedboard: http://www.digilentinc.com/Products/Detail.cfm?NavPath=2,400,1028&Prod=ZEDBOARD
2012-10-26T23:54:29 < emeb> But $319 is a bit steep for the privilege.
2012-10-26T23:56:16 < emeb> Plus, I've already got an Atlys floating around that I've never used :/
2012-10-26T23:57:57 -!- BrainDamage1 [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Read error: Connection reset by peer]
2012-10-26T23:58:02 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-26T23:58:43 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-26T23:59:16 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
--- Day changed Sat Oct 27 2012
2012-10-27T00:08:03 < zyp> I almost bought an atlys once
2012-10-27T00:08:33 < zyp> then I figured that it would be pointless considering how little I've used my s3eboard
2012-10-27T00:09:17 < Erlkoenig> i had an S3 board, plugged the wrong power supply, and poof
2012-10-27T00:09:23 < Erlkoenig> that was depressing
2012-10-27T00:09:31 < zyp> ouch
2012-10-27T00:09:53 < emeb> sadness
2012-10-27T00:10:24 < Erlkoenig> probably the only fpga board without a voltage regulator or fuse
2012-10-27T00:12:49 < emeb> which one was that?
2012-10-27T00:13:03 < zyp> s3eboard
2012-10-27T00:13:16 < zyp> the board doesn't like higher than 5V on the input
2012-10-27T00:13:25 < emeb> huh. Didn't know that.
2012-10-27T00:13:43 < emeb> I've got one but never looked at the supply design too closely.
2012-10-27T00:13:53 < Erlkoenig> ah, the S3A-Starter too
2012-10-27T00:14:10 < Erlkoenig> that's what i had... for 205 €
2012-10-27T00:14:21 < zyp> aren't they basicly the same board except different fpga?
2012-10-27T00:14:46 < emeb> only s3a board I ever had was the little red one from Avnet.
2012-10-27T00:14:51 < emeb> not the same at all.
2012-10-27T00:16:22 < karlp> I kinda wish I'd taken the fpga course in uni.
2012-10-27T00:16:29 < emeb> they're fun
2012-10-27T00:16:35 < karlp> took control systems instead, which was also fun.
2012-10-27T00:16:41 < emeb> indeed.
2012-10-27T00:16:52 < emeb> probably more generally useful too.
2012-10-27T00:17:09 < karlp> I'm glad I learnt a lot of it, but I don't reemember any of the maths
2012-10-27T00:17:16 < emeb> stability! bode plots! closed loops!
2012-10-27T00:17:37 < karlp> turn it all into s-space. solve matrix, turn back into normal space. *magic*
2012-10-27T00:17:48 < zyp> I had an fpga course in uni
2012-10-27T00:17:54 < zyp> weren't that interesting
2012-10-27T00:18:30 < Erlkoenig> i learnt it all by myself. :>
2012-10-27T00:18:43 < emeb> It helps if you have something you want to do with them.
2012-10-27T00:19:02 < TitanMKD> woo 713KUSD
2012-10-27T00:19:04 < TitanMKD> not far
2012-10-27T00:19:06 < karlp> that too :)
2012-10-27T00:19:17 < TitanMKD> just 750KUSD ;)
2012-10-27T00:19:24 < karlp> I haven't yet gotten to a use for a full fpga.
2012-10-27T00:19:41 < Erlkoenig> coolness factor :>
2012-10-27T00:19:41 < karlp> kinda thought I could use a cpld once or twice,
2012-10-27T00:19:53 < karlp> but even then just because
2012-10-27T00:20:40 < zyp> http://bin.jvnv.net/f/wNLwd.jpg <- found my old project docs
2012-10-27T00:20:45 < zyp> from the fpga course
2012-10-27T00:20:48 < emeb> In my day job we build DSP stuff for software radios. You can't do this kind of processing in real-time on most portable compute architectures.
2012-10-27T00:21:13 < Erlkoenig> http://minecraft.niklas.2g2s.de/elektronik/vga.jpg    S3A Starter in action ;-)
2012-10-27T00:21:41 < Erlkoenig> you clearly can see there was an error with the vga waveform :D
2012-10-27T00:21:54 < zyp> ha
2012-10-27T00:21:56 < emeb> zyp: some sort of color analyzer?
2012-10-27T00:22:01 < zyp> emeb, yep
2012-10-27T00:22:17 < emeb> sounds like Laurenceb's project...
2012-10-27T00:22:24 < zyp> photodiode and red and blue leds
2012-10-27T00:22:55 < zyp> and some analog circuit to filter that to analog red/blue outputs
2012-10-27T00:23:11 < zyp> and then fpga stuff to read it with an adc and discriminate colors
2012-10-27T00:23:33 < TitanMKD> emeb and what do you think about the hw of HackRF (Jawbreaker) ?
2012-10-27T00:24:19 < emeb> TitanMKD: It's interesting.
2012-10-27T00:24:43 < emeb> I've looked over their schematics a while back. Seemed a bit limited.
2012-10-27T00:25:40 < TitanMKD> emeb of course it's less faster than a pure FPGA design but it's more flexible to do code on dual core MCU ;)
2012-10-27T00:26:12 < TitanMKD> emeb and also very less expensive than a fast FPGA which also requires often BGA ...
2012-10-27T00:27:38 < karlp> on the topic of old projects: http://palmtree.beeroclock.net/~karlp/uni2ndyear-circa97.jpg
2012-10-27T00:27:44 < karlp> got corrosion? ;)
2012-10-27T00:27:53 < zyp> :)
2012-10-27T00:29:09 < HTT-Bird> bigtime
2012-10-27T00:33:32 <+Steffann> Wha what did you do to it karlp ?
2012-10-27T00:33:44 < karlp> cheap batteries.  had to come in under budget, but meet power specs.
2012-10-27T00:33:58 < karlp> we claimed the lifetime of nice batteries, put cheap ones in for the demo.
2012-10-27T00:34:04 <+Steffann> smile :)
2012-10-27T00:34:24 < karlp> stayed in it's box in my parent's basement since about 98, never opened again.
2012-10-27T00:35:12 < karlp> did all the code in assembly to use a 512byte program ram hc11 instead of the 2k one most other teams used.
2012-10-27T00:35:13 < Erlkoenig> what does it do?
2012-10-27T00:35:44 < karlp> light/temperature logger, to go in a fridge, "wire free" comms to excel/gui for analysis later.
2012-10-27T00:35:51 < emeb> TitanMKD: I like the LPC4300 they're using. Thought it was a bit sad that they're using a CPLD instead of a real FPGA.
2012-10-27T00:36:04 < karlp> Laurenceb_: eat your data logger heart out ;)
2012-10-27T00:36:48 < TitanMKD> emeb hehe yes could have open more features
2012-10-27T00:37:10 < TitanMKD> emeb anyway the schematics are open and you can change that for your own version
2012-10-27T00:37:36 < Erlkoenig> http://ccan.de/cgi-bin/ccan/ccan-view.pl?a=view&i=3132  this is one of my ancient projects... i even forgot that i ever did it, and couldn't remember what the program did and how it looked like after i saw the page again...
2012-10-27T00:37:41 < emeb> TitanMKD: One can get capable FPGAs for not much more than the CPLD they used, in the same pkg that would allow more flexibility.
2012-10-27T00:37:54 < karlp> it was back then I got interested in battery powered projects.
2012-10-27T00:38:18 < TitanMKD> emeb you can tell it to michael
2012-10-27T00:38:31 < emeb> Heh - I'm sure he's heard it already. :)
2012-10-27T00:38:34 < BrainDamage> karlp: excel gui? does it have a vba excel script that communicates over serial port? :p
2012-10-27T00:39:01 < karlp> yeah, if I remember rightly.
2012-10-27T00:39:05 < karlp> I didn't do that side.
2012-10-27T00:39:35 < zyp> wtf :p
2012-10-27T00:40:03 < karlp> it was actually surprisingly easy to use really.
2012-10-27T00:40:29 < karlp> or so I ws told.
2012-10-27T00:41:21 < Laurenceb_> has anyone tried writing stuff to some of the cal registers on mems sensors?
2012-10-27T00:41:36 < karlp> we even used some surface mount parts on that, must have been a cool move at the time.
2012-10-27T00:41:47 < Laurenceb_> some of them seem to have registers that you might be able to write to
2012-10-27T00:42:03 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-27T00:45:41 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 244 seconds]
2012-10-27T00:45:51 < ds2> Laurenceb_: what kind of mems sensor?
2012-10-27T00:46:08 < Laurenceb_> lsm330 or similar
2012-10-27T00:46:22 < ds2> IIRC, those aren't persistant
2012-10-27T00:47:21 < ds2> Laurenceb_: why would you want to use them? never found a good reason to those
2012-10-27T00:47:41 < Laurenceb_> to id different plug in sensors
2012-10-27T00:47:51 < Laurenceb_> and load the correct calibration
2012-10-27T00:48:11 < ds2> but they aren't EEPROM so you'll have to reload them each time
2012-10-27T00:48:53 < Laurenceb_> ok
2012-10-27T00:49:02 < Laurenceb_> i wondered if any of the sensors use eeprom
2012-10-27T00:49:09 < Laurenceb_> e.g. for the WHO_AM_I
2012-10-27T00:49:25 < ds2> haven't seen one that does
2012-10-27T00:49:34 < Laurenceb_> ok
2012-10-27T00:49:40 < ds2> cal values can change for board, temp, and environment reasons
2012-10-27T00:50:00 < ds2> don't want to tack a EEPROM on it?
2012-10-27T00:51:28 < zyp> that would be too easy
2012-10-27T00:52:01 < Laurenceb_> requires dead bugging
2012-10-27T00:52:08 < Laurenceb_> im using sparkfun 9dof boards
2012-10-27T00:52:16 < ds2> Ohhhhh
2012-10-27T00:52:40 < Laurenceb_> _lots_  of them
2012-10-27T00:52:41 < ds2> should wait for my boards ;) it has room for DIP parts ;)
2012-10-27T00:52:45 < Laurenceb_> on plug in cables
2012-10-27T00:53:02 < ds2> what's the project?
2012-10-27T00:53:14 < zyp> I would just use lots of f3discovery boards
2012-10-27T00:53:16 < zyp> cheaper :p
2012-10-27T00:55:23 < upgrdman_> aren't stm sensors known for being mediocre or less
2012-10-27T00:55:45 < zyp> nah
2012-10-27T00:55:57 < zyp> I believe they are just known to not be the best
2012-10-27T00:56:24 < zyp> but some people only distinguish between the best and shit
2012-10-27T00:56:30 < upgrdman_> ok
2012-10-27T00:56:53 < upgrdman_> i kinda wish the accelerometer on the f4discov board had >8bits of resolution
2012-10-27T00:57:00 < upgrdman_> but for the price im not whining :)
2012-10-27T00:58:02 < Laurenceb_> its a vehicle ride quality thingy
2012-10-27T00:58:06 < Laurenceb_> not my project really
2012-10-27T00:58:12 < Laurenceb_> i just helped them out
2012-10-27T01:00:33 -!- barthess [~barthess@5.100.211.45] has quit [Quit: Leaving.]
2012-10-27T01:01:00 -!- barthess [~barthess@5.100.206.106] has joined ##stm32
2012-10-27T01:01:52 < Laurenceb_> not clicking then
2012-10-27T01:03:56 < Erlkoenig> woah
2012-10-27T01:04:20 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has left ##stm32 []
2012-10-27T01:06:30 < Erlkoenig> hmm my favourite nightmare is when light doesn't work properly, it's dark and the lamps don't emit enough light to see clearly
2012-10-27T01:07:29 < Erlkoenig> oh, nice
2012-10-27T01:07:57 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-27T01:08:02 -!- barthess [~barthess@5.100.206.106] has quit [Quit: Leaving.]
2012-10-27T01:08:09 < Erlkoenig> that's why i have a very bright lamp and my room looks like an operating room in a hospital :D
2012-10-27T01:09:04 < zyp> I tend to walk around here in complete darkness
2012-10-27T01:09:47 < zyp> once I closed a door that's usually open
2012-10-27T01:10:01 < zyp> which resulted in me walking face first into it a few hours later
2012-10-27T01:10:47 < Erlkoenig> walking around in darkness in my room would result entangling your legs with wires and falling into a mess of electronics stuff. :D
2012-10-27T01:11:07 < zyp> I have safe walking paths
2012-10-27T01:11:47 < ds2> sounds like an RPG
2012-10-27T01:12:01 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-27T01:14:34 < emeb> you are likely to be eaten by a grue.
2012-10-27T01:15:55 < emeb> w00t - my little F373 breakout is off to the fab.
2012-10-27T01:28:04 -!- upgrdman_ [429f3cfe@blender/artist/upgrdman] has quit [Quit: Page closed]
2012-10-27T01:30:34 < Erlkoenig> hm tonight ima gonna dream of ARM Assembler
2012-10-27T01:38:49 < ds2> that sounds like a bad neighborhood if you got assemblers that need to be armed
2012-10-27T01:40:25 < Erlkoenig> wah. "ARM assembly code" if you are picky :P
2012-10-27T01:43:34 < ds2> that's some bad instructions for assembling limbs ;)
2012-10-27T01:44:21 < Erlkoenig>  80002b2:	4a06      	ldr	r2, [pc, #24]	; (80002cc <SVC_Handler+0x34>)
2012-10-27T01:44:27 < Erlkoenig> thats my ARM
2012-10-27T02:01:17 < emeb> only $22k left to go on Parallella.
2012-10-27T02:08:41 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Remote host closed the connection]
2012-10-27T02:09:08 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-27T02:12:45 < Erlkoenig> GCC is funny... inserts an instruction that's never to be reached, and if it were to, would cause and endless loop
2012-10-27T02:18:03 < Laurenceb_> http://www.st.com/internet/com/home/home.jsp
2012-10-27T02:18:06 < Laurenceb_> oh dear
2012-10-27T02:18:12 < Laurenceb_> at the heart of fail
2012-10-27T02:18:20 < Erlkoenig> the ST webpage
2012-10-27T02:18:25 < Erlkoenig> where's the problem?
2012-10-27T02:20:26 < Laurenceb_> windows8
2012-10-27T02:21:49 < Erlkoenig> yes
2012-10-27T02:22:30 < Erlkoenig> they're probably gonna get a good opportunity selling stuff made for win8, so why not?
2012-10-27T02:25:08 -!- Laurenceb_ [~Laurence@host86-178-238-123.range86-178.btcentralplus.com] has quit [Ping timeout: 245 seconds]
2012-10-27T02:27:48 -!- buserror [~michel@host86-145-72-200.range86-145.btcentralplus.com] has quit [Ping timeout: 248 seconds]
2012-10-27T02:27:56 < zyp> Erlkoenig, where?
2012-10-27T02:28:09 < Erlkoenig> when?
2012-10-27T02:28:25 < zyp> 01:12:45 < Erlkoenig> GCC is funny... inserts an instruction that's never to be reached, and if it were to, would cause and endless loop
2012-10-27T02:28:28 < Erlkoenig> ah
2012-10-27T02:28:28 < Erlkoenig> mom
2012-10-27T02:29:45 < zyp> hmm
2012-10-27T02:29:47 < Erlkoenig> http://pastebin.com/4VLrApEs that b.n at the end should never be reached, and would jump to the first line...
2012-10-27T02:29:55 < zyp> didn't get anything done tonight
2012-10-27T02:30:38 < Erlkoenig> i just got the thought that the whole approach for this function is bullshit. have to do it some other way...
2012-10-27T02:31:01 < Erlkoenig> at least i know GCC deals with inline assembly the way i want it to
2012-10-27T02:31:04 < zyp> do what?
2012-10-27T02:31:33 < Erlkoenig> start a new task...
2012-10-27T02:32:02 < zyp> you mean context switch?
2012-10-27T02:32:30 < zyp> I'd put the context switch code directly in the pendsv handler and mark it naked
2012-10-27T02:32:42 < Erlkoenig> yes i copypasta'd that approach
2012-10-27T02:32:53 < Erlkoenig> but i also somehow need to stast a task - i.e. initialize its stack
2012-10-27T02:32:58 < zyp> my current code doesn't do that
2012-10-27T02:33:04 < zyp> I mean, use pendsv
2012-10-27T02:33:12 < Erlkoenig> well oohps
2012-10-27T02:33:16 < zyp> my current code is using svcall, in a semi-broken approach
2012-10-27T02:33:21 < Erlkoenig> yes, i do that in the SVCall_Handler
2012-10-27T02:33:28 < Erlkoenig> via the SVC interrupt
2012-10-27T02:34:05 < zyp> anyway, starting a new task is easy, you just need to fill the stack frame with some default values
2012-10-27T02:35:03 < zyp> http://cgit.jvnv.net/laks/tree/os/thread.h <- if you haven't found it yet, the constructor here is what does it in my code
2012-10-27T02:35:42 < Erlkoenig> hmm i need to think about it
2012-10-27T02:36:08 < zyp> i.e. just put the thread main addr into the pc field in the stack frame you are constructing, then put the exc_return magic value into the upper lr slot
2012-10-27T02:36:56 < zyp> then when your context switch happens, it will pop the lr value into pc, making hardware do a return which pulls the thread main value into the pc register
2012-10-27T02:37:28 < zyp> if you like, you can even fill r0 with a variable if you would like to pass a variable to your thread main
2012-10-27T02:37:37 < Erlkoenig> main value = address of thread's main function?
2012-10-27T02:37:40 < zyp> yes
2012-10-27T02:38:08 < Erlkoenig> i thought of that, but i had the idea it could be easier... but that wont work, i need to think a bit about it :>
2012-10-27T02:38:20 < zyp> it's 01:38 here now, might be a bit sleepy so excuse me being a bit confusing
2012-10-27T02:38:26 < zyp> easier?
2012-10-27T02:38:37 < zyp> this is seriously easy
2012-10-27T02:38:57 < Erlkoenig> well just lemme think :D
2012-10-27T02:39:05 < zyp> think of the stack as the stored state of a frozen thread
2012-10-27T02:39:15 < Erlkoenig> i know man :D
2012-10-27T02:39:35 < zyp> when you are creating a new thread you're creating the state of a thread that has not started executing
2012-10-27T02:39:52 < zyp> most registers contain undefined values, so you don't have to care about them
2012-10-27T02:39:54 < Erlkoenig> what about just calling the thread's main from my super-main() function, and instantly interrupt it via SVC
2012-10-27T02:40:02 < Erlkoenig> than i get the pre-initialized stack frame "for free"
2012-10-27T02:40:22 < Erlkoenig> the problem is "just" returning to the super-main
2012-10-27T02:40:24 < zyp> on which stack?
2012-10-27T02:40:44 < zyp> you need to create a new stack to hold the new state
2012-10-27T02:40:54 < Erlkoenig> before starting the new thread i'd switch to a new stack
2012-10-27T02:41:05 < Erlkoenig> i.e. ldr sp, ...
2012-10-27T02:41:37 < zyp> you're overthinking it
2012-10-27T02:41:50 < zyp> again, look at my Thread() constructor
2012-10-27T02:41:54 < Erlkoenig> man :D
2012-10-27T02:41:55 < zyp> it's four lines of code.
2012-10-27T02:42:11 < zyp> do you really think you could do it any simpler than that?
2012-10-27T02:42:12 < Erlkoenig> i try to learn something from this okay? so i'd like to think of multiple possibilities
2012-10-27T02:42:29 < Erlkoenig> if i just wanted to "do it" i could just use any ready OS
2012-10-27T02:42:50 < zyp> your model sounds a bit like fork()
2012-10-27T02:44:39 < zyp> but fork relies on having an mmu
2012-10-27T02:46:42 < Erlkoenig> ah. i got the problem - interrupts are "symmetric" in building/unwinding the stackframe on the stack, but this generally requires an asymmetric behaviour - which can be done by manually initializing the stack fram
2012-10-27T02:46:43 < Erlkoenig> e
2012-10-27T02:47:02 < Erlkoenig> or by manually leaving the interrupt
2012-10-27T02:47:14 < Erlkoenig> which would be more of an evil hack than just manually initializing the stack frame
2012-10-27T02:49:28 < zyp> I'm not sure you can just leave handler mode without doing a normal interrupt return
2012-10-27T02:49:45 < zyp> which means that it needs a stack frame to pop either way
2012-10-27T02:50:23 < Erlkoenig> yup thats the problem
2012-10-27T02:50:53 < Erlkoenig> on avr you can just jump out of the ISR, if you do the required "sei" and "pop"... but an ARM it might be different... anyways, it looks unclean and evil
2012-10-27T02:52:57 < zyp> actually, I thought of a way to start threads without preinitializing the stacks for them
2012-10-27T02:53:53 < dongs> lotsa chats
2012-10-27T02:54:04 < emeb> 'lo dongs
2012-10-27T02:55:11 < zyp> you can construct a read only stack bit that gets popped, which instead of executing the thread main, executes a fixed function that takes the new sp and thread main addr in callee-saved registers, sets the new sp and jumps to the thread main
2012-10-27T02:56:34 < zyp> then you can make a service call where you just save your old stack like first half of thread switch, then put the new sp and main into r4 and r5, the stack stub addr into sp, then do an interrupt return
2012-10-27T02:59:15 < zyp> morning, dongs
2012-10-27T03:00:18 < Erlkoenig> hm, do i need 8byte-aligned stack frames?
2012-10-27T03:00:38 < zyp> probably
2012-10-27T03:23:31 < dongs> http://il.linkedin.com/pub/gary-niger/52/862/775
2012-10-27T03:26:37 < zyp> hmm, interesting, F3 can clock some peripherals from other clock sources than syscloc
2012-10-27T03:26:50 < dongs> isnt it?
2012-10-27T03:26:55 < dongs> i thought you could run TIMx at 144mhz
2012-10-27T03:26:56 < dongs> or someshit
2012-10-27T03:26:58 < dongs> because of that
2012-10-27T03:27:01 < emeb> cool.
2012-10-27T03:27:06 < zyp> which I assumes means allowing them to run while the core sleeps
2012-10-27T03:27:58 < zyp> yep, TIM1/8 kan run directly from PLL output
2012-10-27T03:31:06 < Erlkoenig> why can't GCC do bitwise operations, especially &, on pointers...
2012-10-27T03:31:21 < HTT-Bird> Erlkoenig: because the C standard says no?
2012-10-27T03:31:31 < Erlkoenig> meh.
2012-10-27T03:33:33 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-27T03:34:47 < Erlkoenig> here come the casts
2012-10-27T03:35:11 < upgrdman> why would you want/need bitwise ops for pointers
2012-10-27T03:36:30 < Erlkoenig> to do 8/4 byte alignment of addresses in software
2012-10-27T03:40:35 < zyp> what
2012-10-27T03:40:41 < zyp> why do you need to do that?
2012-10-27T03:40:51 < Erlkoenig> to align the SP correctly
2012-10-27T03:41:09 < zyp> how are you allocating stack memory?
2012-10-27T03:41:23 < Erlkoenig> if (SCB->CCR & SCB_CCR_STKALIGN_Msk) {    // ... need to allign 8bytewise
2012-10-27T03:41:53 < Erlkoenig> currently just as global variables :> ... later i'll do that probably via linker script or something. of course one should make sure they are correctly aligned already...
2012-10-27T03:48:15 < Erlkoenig> uhm i can be sure that a "str r*, [...]" is atomic (i.e. can't be interrupted while only partly completed), right?
2012-10-27T03:48:43 < zyp> atomic how?
2012-10-27T03:49:15 < zippe> Erlkoenig: just use uintptr_t for your pointer math
2012-10-27T03:49:35 < zippe> Erlkoenig: exceptions are normally only fired on instruction boundaries
2012-10-27T03:49:44 < Erlkoenig> oh goot
2012-10-27T03:49:48 < zippe> Erlkoenig: multiple load/store operations can be interruptible
2012-10-27T03:49:51 < ds2> wow.... ESRs
2012-10-27T03:50:08 < Erlkoenig> zippe: yep, that's okay
2012-10-27T03:50:28 < Erlkoenig> ah, uintptr_t is nice, didn't know about that
2012-10-27T03:50:43 < Erlkoenig> zyp: atomic as in not to be interrupted, only BEFORE or AFTER
2012-10-27T04:02:52 < zyp> and there I got my ACM demo running on F3
2012-10-27T04:03:41 < Erlkoenig> nice
2012-10-27T04:04:15 < zyp> that was not really hard work
2012-10-27T04:04:40 < zyp> most of it is F1 compatible, so I just had to add F3 to a bunch of ifdefs
2012-10-27T04:04:56 < zyp> and the part that's not F1 compatible (GPIO) is F4 compatible
2012-10-27T04:05:10 < Erlkoenig> do you happen to hava an ACM demo for F4 which gets auto-registered on both windows and linux as a virtual TTY? so one could use it as a virtual UART without a driver? if yes, upload it to http://www.mikrocontroller.net/  or http://embdev.net/  people will love it ;-)
2012-10-27T04:05:44 < upgrdman> wtf is an acm demo?
2012-10-27T04:05:44 < zyp> dunno about windows, works on everything else
2012-10-27T04:05:50 < zyp> USB CDC ACM
2012-10-27T04:05:54 < zyp> i.e. virtual serial
2012-10-27T04:06:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-27T04:11:16 < Erlkoenig> could GCC generate PUSH/POP *within* a function, apart from the ones found at the beginning/end?
2012-10-27T04:11:24 < Erlkoenig> or something else that modifies SP?
2012-10-27T04:14:08 < zyp> of course
2012-10-27T04:14:47 < dongs> Erlkoenig: the best course of action for driverless would be emulating some FTDI shite
2012-10-27T04:15:09 < Erlkoenig> dongs: probably complicated because undocumented
2012-10-27T04:15:14 < dongs> hardly
2012-10-27T04:15:18 < dongs> theres opensauce lunix shit.
2012-10-27T04:15:24 < Erlkoenig> ah of course :D
2012-10-27T04:15:39 < zyp> doesn't look very complicated
2012-10-27T04:16:00 < zyp> I had a look at the descriptors for the four port chip once
2012-10-27T04:16:14 < zyp> it has four bulk pairs, one per chip
2012-10-27T04:16:30 < zyp> and the rest is probably just control requests
2012-10-27T04:17:09 < zyp> those should be easy to figure out by looking at a trace from a real ftdi
2012-10-27T04:17:57 < dongs> or just looking at the lunix driver for it
2012-10-27T04:18:40 < zyp> there is no guarantee that the windows driver is not issuing some extra secret commands that is strictly not required
2012-10-27T04:19:01 < zyp> and who knows what the windows driver will do if they are not answered like a real ftdi would?
2012-10-27T04:19:05 < Erlkoenig> hm, how to reverse engineer usb datastreams on windows without hardware that costs more than a ferrari?
2012-10-27T04:19:17 < zyp> borrow that hardwar
2012-10-27T04:19:18 < zyp> e
2012-10-27T04:19:22 < zyp> that's what I do ;)
2012-10-27T04:19:29 < dongs> um, hhd usb stuff works fine
2012-10-27T04:19:40 < Erlkoenig> "borrow"? what?
2012-10-27T04:19:40 < dongs> more than enough for ftdi sniffing.
2012-10-27T04:19:42 < Erlkoenig> aah okay
2012-10-27T04:19:54 < dongs> i think they even have a 30 day trial so you have like a month to get it to work
2012-10-27T04:19:57 < dongs> shouldtn take more than a day thou
2012-10-27T04:20:15 < dongs> you can have it setup to monitor on device connect
2012-10-27T04:20:36 < Erlkoenig> ah very nice
2012-10-27T04:21:20 < Erlkoenig> somebody should reverse engineer the myDMX protocol, build a clone with an STM32 that can do more than the original, an sell it for 1/5 of the price of the original :3
2012-10-27T04:21:29 < dongs> wut
2012-10-27T04:21:42 < dongs> what is there to "figure out" about DMX
2012-10-27T04:21:51 < dongs> its just uart basically
2012-10-27T04:22:00 < Erlkoenig> the protocol of the myDMX USB->myDMX adapter
2012-10-27T04:22:10 < dongs> um, its probaly just usb>serial.
2012-10-27T04:22:17 < Erlkoenig> maybe
2012-10-27T04:22:24 < Erlkoenig> so one just needs the correct VID/PID
2012-10-27T04:22:27 < Erlkoenig> so the clone would be compatible with the myDMX software
2012-10-27T04:23:26 < dongs> I made 16 channel usb>dmx out few months ago
2012-10-27T04:23:55 < dongs> thier 1 channel shit is simpleaids.
2012-10-27T04:24:08 < Erlkoenig> yes i made something like that too... but they're all useless without a superpowerful PC software
2012-10-27T04:24:24 < Erlkoenig> and the myDMX *has* that superpowerful software, just the hardware is ridiculously overpriced
2012-10-27T04:24:30 < dongs> wherei s their shitty softwarE?
2012-10-27T04:24:32 < dongs> is it downloadable?
2012-10-27T04:24:35 < Erlkoenig> yup freeware
2012-10-27T04:24:48 < dongs> 55 megs.
2012-10-27T04:24:53 < dongs> furiosuly downlading .
2012-10-27T04:24:58 < Erlkoenig> hehehe
2012-10-27T04:25:05 < Erlkoenig> someone sensed an opportunity? :>
2012-10-27T04:25:23 < dongs> no. but other people here are doing DMX shit so it could be useful for them.
2012-10-27T04:25:56 < Erlkoenig> if you implement that, include some mapping of: some special DMX channels => GPIO   -   so that you could transparently control lights with simple parallel (LPT) input from the myDMX software
2012-10-27T04:28:20 -!- DLPeterson [~hazelnuss@ims1065d.engr.ucdavis.edu] has quit [Ping timeout: 260 seconds]
2012-10-27T04:31:03 < dongs> unsigned installer...........
2012-10-27T04:31:06 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-27T04:31:07 < dongs> in 2012
2012-10-27T04:31:09 < dongs> these guys are pro
2012-10-27T04:32:06 < Erlkoenig> or you got a mitm :>
2012-10-27T04:32:35 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-bmnjepctiroazkki] has quit [Remote host closed the connection]
2012-10-27T04:32:44 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ejppulyjkoohahyl] has joined ##stm32
2012-10-27T04:32:50 < dongs> time to snapshot and install
2012-10-27T04:34:34 -!- txf [~txf@bl8-122-218.dsl.telepac.pt] has quit [Ping timeout: 244 seconds]
2012-10-27T04:36:03 -!- txf [~txf@bl13-176-136.dsl.telepac.pt] has joined ##stm32
2012-10-27T04:40:58 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ejppulyjkoohahyl] has quit [Ping timeout: 246 seconds]
2012-10-27T04:41:56 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xgonqyoqwxqdfexu] has joined ##stm32
2012-10-27T04:44:54 < dongs> ugly delphi shit.
2012-10-27T04:47:40 < dongs> looks like they're using cypress generic USB driver, technically if you're a for-profit company you can only use that with their hardware
2012-10-27T04:47:52 < dongs> so inside is probably cypress fx2 or similar shite
2012-10-27T04:48:31 < dongs> oh haha they even have fx2 firmware in there
2012-10-27T04:48:35 < dongs> so yeah, that answers that.
2012-10-27T04:48:59 < zyp> so now you can just grab an fx2 and flash it? :p
2012-10-27T04:49:02 < dongs> yeah
2012-10-27T04:49:09 < Erlkoenig> what.
2012-10-27T04:49:33 < Erlkoenig> sounds like to easy :D
2012-10-27T04:49:37 < zyp> they must love their china cloners
2012-10-27T04:50:58 < Erlkoenig> maybe the developer's revenge on the marketing guys :>
2012-10-27T04:51:17 < dongs> http://bcas.tv/paste/results/EPWTKy71.html
2012-10-27T04:51:21 < dongs> are those the harwdares theyt sell?
2012-10-27T04:51:35 < Erlkoenig> ooh it's intelligent
2012-10-27T04:51:44 < dongs> firmware downlaod = means fx2 has no large eepropm, so tehy just put vid+pid in there then softload
2012-10-27T04:51:58 < dongs> and firmware is  in the clear in a cypress script file you can extract with some lame python t ool
2012-10-27T04:52:09 < Erlkoenig> http://www.americandj.com/ProductDetails.aspx?Category=&ItemNumber=1773  this is the hardware. i don't know more...
2012-10-27T04:52:46 < Erlkoenig> at least the case looks solid
2012-10-27T05:06:33 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-27T05:12:51 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-27T05:19:32 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 272 seconds]
2012-10-27T05:24:18 < dongs> http://toronto.en.craigslist.ca/tor/mob/3366418989.html
2012-10-27T05:26:16 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Ping timeout: 255 seconds]
2012-10-27T05:28:34 < dongs> my thoughts exactly
2012-10-27T05:34:47 < Erlkoenig> if i could resolder the smd mini-usb on my ancient somewhat-smart-phone i would be happy with it...
2012-10-27T05:38:20 < bsdfox_> if it uses mini-usb it's time for an upgrade (assuming you pay for a data package)
2012-10-27T05:39:14 < Erlkoenig> whatwhy?
2012-10-27T05:39:25 < emeb_mac> Erlkoenig: what's stopping you?
2012-10-27T05:39:25 < Erlkoenig> and i have prepaid
2012-10-27T05:39:46 < Erlkoenig> emeb_mac: the lack of a smd soldering station and of my ability to use something like that :D
2012-10-27T05:40:37 < emeb_mac> what is this "smd soldering station" of which you speak? Just get a cheap temp-controlled iron with the tiniest tip you can find.
2012-10-27T05:41:13 < dongs> or just pay jewbinstein for aoyue 968+
2012-10-27T05:41:15 < dongs> its only like $120
2012-10-27T05:41:17 < Erlkoenig> does that suffice?
2012-10-27T05:41:32 < emeb_mac> I like my Hakko 936 that cost ~$70
2012-10-27T05:41:34 < Erlkoenig> my uni has a temp controlled iron with a fine tip... maybe i can use that
2012-10-27T05:41:35 < dongs> http://sra-solder.com/product.php/6267/1/968/668c4501454a36e504b867453b36920e
2012-10-27T05:41:49 < emeb_mac> Erlkoenig: that would be just great.
2012-10-27T05:42:05 < Erlkoenig> okay, i could try it
2012-10-27T05:43:46 < emeb_mac> dongs: that's a nice outfit.
2012-10-27T05:44:16 < dongs> yeah, thats what i got
2012-10-27T05:44:18 < emeb_mac> btw - I got an aoyue hoof tip for my Hakko - worked great.
2012-10-27T05:44:35 < dongs> right
2012-10-27T05:45:02 < dongs> saame factory just slap diff label on it
2012-10-27T05:49:05 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-27T05:55:54 < GargantuaSauce> i have a hakko 936 as well...seems to do the trick though i completely destroyed the upper half of the tip
2012-10-27T05:57:54 < GargantuaSauce> why wouldn't it? think about solder braid :P
2012-10-27T05:58:07 < GargantuaSauce> what would you guys recommend by way of hot air by the way?
2012-10-27T05:58:11 < GargantuaSauce> other than listening to flyback :V
2012-10-27T06:01:34 < GargantuaSauce> cool
2012-10-27T06:02:44 < GargantuaSauce> i used a butane torch when assembling my reprap and it did nooot work very well :P
2012-10-27T06:03:22 < GargantuaSauce> yeah...high heat low temperature
2012-10-27T06:03:25 < GargantuaSauce> a opposed to the opposite
2012-10-27T06:05:30 < GargantuaSauce> i tried reseating a tssop with the same torch a few days ago
2012-10-27T06:05:33 < GargantuaSauce> that did not go so well
2012-10-27T06:06:21 < Erlkoenig> *holy shit it works*
2012-10-27T06:12:33 < Erlkoenig> hwahwahwa
2012-10-27T06:12:42 < Erlkoenig> well that was not referring to resoldering that usb
2012-10-27T06:14:02 < Erlkoenig> well... not now...
2012-10-27T06:14:22 < GargantuaSauce> dx is your friend
2012-10-27T06:18:36 < Erlkoenig> aaah everyone wants to force me to SMD
2012-10-27T06:18:47 < Erlkoenig> the more they do it the less i want to do
2012-10-27T06:19:06 < GargantuaSauce> i have been pretty resistant to the idea too
2012-10-27T06:19:32 < GargantuaSauce> but sticking with dip ICs only is kiiiiind of limiting :P
2012-10-27T06:19:52 < Erlkoenig> luckily there are ready made boards with STM32 on them
2012-10-27T06:20:14 < GargantuaSauce> that cost as much as the micros themselves
2012-10-27T06:20:54 < emeb_mac> Erlkoenig: SMD is eaaaaasyyyy.
2012-10-27T06:21:05 < GargantuaSauce> as a result i have four f4-discos
2012-10-27T06:21:05 < emeb_mac> and faster to assemble than TH
2012-10-27T06:21:11 < GargantuaSauce> just because.
2012-10-27T06:34:47 < Erlkoenig> __WFE ();
2012-10-27T06:34:48 -!- Erlkoenig [~erlkoenig@pptp-212-201-72-111.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-27T06:39:11 < upgrdman> i aim for all smd, single layer designs myself :) glutton for punishment
2012-10-27T06:40:43 < emeb_mac> why single-layer?
2012-10-27T06:40:56 < upgrdman> one-off, at home pcbs,
2012-10-27T06:41:01 < emeb_mac> ah.
2012-10-27T06:41:02 < upgrdman> and im lazy
2012-10-27T06:41:17 < upgrdman> so alignment and drilling are not preferred
2012-10-27T06:41:22 < emeb_mac> probably doable, with some wiring for jumpers.
2012-10-27T06:41:40 < upgrdman> fuck wires. zero ohm resistors!
2012-10-27T06:41:50 < emeb_mac> that too
2012-10-27T06:42:06 < upgrdman> but yes, i like magnet wire for hacks
2012-10-27T06:43:28 < emeb_mac> 30-ga wire-wrap wire myself.
2012-10-27T06:43:48 < upgrdman> that solid core, right? (never see it in person)
2012-10-27T06:44:24 < emeb_mac> yep - silver plated, kynar insulation. works well.
2012-10-27T06:44:42 < upgrdman> is kynar better than silicone?
2012-10-27T06:45:24 < emeb_mac> beats me - never used silcone.
2012-10-27T06:45:42 < upgrdman> k
2012-10-27T06:56:22 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Read error: Operation timed out]
2012-10-27T06:57:59 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-27T07:16:09 < dongs> ok who here know smath
2012-10-27T07:16:16 < dongs> knows math even
2012-10-27T07:16:23 < dongs> http://dongle.imgur.com/all/
2012-10-27T07:16:25 < dongs> err
2012-10-27T07:16:42 < dongs> http://i.imgur.com/RpfbI.jpg
2012-10-27T07:17:09 < dongs> wat length beltdo i need
2012-10-27T07:22:32 < emeb_mac> 24 + 24 + pi/2*11 + pi/2 *2
2012-10-27T07:22:40 < emeb_mac> that's pretty close.
2012-10-27T07:23:16 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-27T07:23:54 < emeb_mac> 68.42cm
2012-10-27T07:25:56 < dongs> http://tncscooters.com/index.php?route=product/product&path=33&product_id=453 looks like tis this belt hten
2012-10-27T07:26:00 < dongs> width fits too.
2012-10-27T07:26:03 < dongs> ill give it a try.
2012-10-27T07:26:32 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Client Quit]
2012-10-27T07:26:34 < dongs> hmm.. places are selling it as CVT belt. i would have expected to see ebike hits.
2012-10-27T07:28:13 < emeb_mac> close. is it stretchy?
2012-10-27T07:28:23 < dongs> wlel, 24cm measurement might not be precise.
2012-10-27T07:28:33 < dongs> i figure that shop only hs 4 belts, and deals with ebikes/scooters
2012-10-27T07:28:45 < dongs> so maybe a roughly fitting size would work +- measurement mistake
2012-10-27T07:28:55 < dongs> width fits... i dont understand waht hte fuck 30mm tooth parameter is
2012-10-27T07:29:10 < dongs> Tooth Spacing: 1.18'' / 30mm
2012-10-27T07:29:22 < dongs> the teeth on the drive here are much finer than that
2012-10-27T07:30:15 < dongs> hmm
2012-10-27T07:30:20 < emeb_mac> yeah - the picture looks finer too.
2012-10-27T07:30:23 < dongs> right..
2012-10-27T07:30:30 < dongs> 3cm is like..
2012-10-27T07:30:31 < dongs> huge.
2012-10-27T07:30:41 < emeb_mac> yup
2012-10-27T07:31:00 < emeb_mac> maybe they mean 3mm - that'd be about right for the picture.
2012-10-27T07:32:19 < dongs> just measured
2012-10-27T07:32:21 < dongs> my shit is 10mm
2012-10-27T07:32:30 < dongs> so each tooth/drop is 5mm
2012-10-27T07:32:39 < dongs> of course i dont have original belt
2012-10-27T07:32:54 < emeb_mac> what's it for?
2012-10-27T07:33:00 < dongs> that bike
2012-10-27T07:33:06 < dongs> that im fixing
2012-10-27T07:33:24 < emeb_mac> orig mfg has spare parts?
2012-10-27T07:33:44 < dongs> i don eeven know who mfg is
2012-10-27T07:34:03 < emeb_mac> no-name. yay.
2012-10-27T07:34:05 < dongs> http://i.imgur.com/MyPB1.jpg
2012-10-27T07:34:06 < dongs> yeah
2012-10-27T07:34:12 < dongs> there are some chinese 'Razor' brand shits
2012-10-27T07:34:14 < dongs> that look different
2012-10-27T07:34:17 < dongs> but similar spec
2012-10-27T07:34:22 < dongs> this might be older version or soemthign
2012-10-27T07:34:41 < dongs> theres no maker sticker or label or anything anywhre on this thing lo
2012-10-27T07:34:54 < emeb_mac> wonderful
2012-10-27T07:35:35 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-27T07:35:42 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-27T07:37:18 < dongs> http://www.scooterpart.net/products.php?id=1828 hmm that looks similar too, writing is correct too
2012-10-27T07:37:23 < dongs> 5M =5mm spacing
2012-10-27T07:38:09 < dongs> would be really nice if shit was labeled properly.
2012-10-27T07:41:53 < emeb_mac> oooh - 670mm is just right too.
2012-10-27T07:42:11 < dongs> right. +- some measurement error or someshit.
2012-10-27T07:42:21 < emeb_mac> yup
2012-10-27T07:42:30 < dongs> only thing is the only hits for those are some gas scooters
2012-10-27T07:42:55 < emeb_mac> yours electric?
2012-10-27T07:42:58 < dongs> eha
2012-10-27T07:43:01 < dongs> yeas
2012-10-27T07:43:56 < emeb_mac> get one - it's only money.
2012-10-27T07:43:59 < dongs> right.
2012-10-27T07:44:04 < dongs> i dont see anything else in similar spec.
2012-10-27T07:44:11 < dongs> We only sell to USA & Puerto Rico Street address.  We no longer take international order.
2012-10-27T07:44:15 < dongs> time to find places that actually ship it to me.
2012-10-27T07:45:01 < emeb_mac> well - you've got a part number.
2012-10-27T07:45:18 < emeb_mac> if you can't find anyone I could relay.
2012-10-27T07:45:43 < dongs> im sure some chinaman on ebay has them
2012-10-27T07:45:54 < emeb_mac> safe bet
2012-10-27T07:46:53 < emeb_mac> why look: http://www.ebay.com/itm/670-5M-18-COBRA-GAS-SCOOTER-DRIVE-TIMING-BELT-/400197488137
2012-10-27T07:46:56 < dongs> http://www.ebay.com/itm/170675896535 we are the belt experts.
2012-10-27T07:47:02 < dongs> this polace is ch eapest.
2012-10-27T07:47:33 < emeb_mac> there you go.
2012-10-27T07:49:41 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has quit [Ping timeout: 255 seconds]
2012-10-27T07:55:02 < dongs> alrite shit ordered and i sent them a message wiht my aawesome drawingpic
2012-10-27T07:55:11 < dongs> bbl
2012-10-27T07:55:18 < emeb_mac> l8rs
2012-10-27T07:56:04 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-27T08:03:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-27T08:11:27 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-27T08:51:33 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-27T09:32:13 < baird> Classic aussie roadtrollercycling .. https://www.youtube.com/watch?v=9WmhMKWt8DI
2012-10-27T09:44:53 -!- tavish [7d3f6ea0@gateway/web/freenode/ip.125.63.110.160] has joined ##stm32
2012-10-27T09:46:58 -!- claude is now known as Claude
2012-10-27T09:47:51 < tavish> has anyone encountered "Core held in reset" error with ST-Link utility? (Also i was unable to finish writing flash with texane/stlink on linux too)
2012-10-27T10:02:00 -!- |sark [~sark@109.235.51.202] has quit [Ping timeout: 265 seconds]
2012-10-27T10:07:17 < emeb_mac> I think I saw that once when I had a part wire up funny.
2012-10-27T10:07:45 < emeb_mac> didn't have all the pins soldered down well.
2012-10-27T10:10:48 < tavish> i think i have a problem with my nRST pin
2012-10-27T10:11:47 < tavish> it resets when it is low, and the NRST pin of st-link seems to be always low
2012-10-27T10:21:06 < dongs> got bike sorta hacked
2012-10-27T10:21:11 < dongs> with a similar fitting vbelt
2012-10-27T10:21:14 < dongs> it slips of course, under load.
2012-10-27T10:21:18 < dongs> but starting slwoly works
2012-10-27T10:21:27 < dongs> will see how quickly the shit will wear out heh
2012-10-27T10:22:32 < emeb_mac> dongs: cut notches in the belt. It'll only take a few hours. :P
2012-10-27T10:22:48 < dongs> heh
2012-10-27T10:22:50 < dongs> right
2012-10-27T10:26:03 < baird> Loosen the rear axel and move the wheel back a bit?
2012-10-27T10:29:51 < emeb_mac> zzz
2012-10-27T10:29:54 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-27T10:30:16 < baird> I once did that on the around-AU trip on the Postie, having to adjust things after getting a new chain fitted. I rode about about 300 kms before realizing I forgot to tighten the axel nuts :)
2012-10-27T11:00:38 < tavish> :( nrst on the chip is not the nrst that st-link wants
2012-10-27T11:01:09 < jpa-> tavish: huh?
2012-10-27T11:01:43 < tavish> st-link/jtag needs NJTRST
2012-10-27T11:01:57 < jpa-> i'm pretty sure it wants NRST
2012-10-27T11:01:57 < tavish> nice naming
2012-10-27T11:02:14 < jpa-> atleast that's what i've always used
2012-10-27T11:02:42 < tavish> jpa-: it isnt working for my pcb, and also for olimex STM32F4-Hxx
2012-10-27T11:02:56 < jpa-> tavish: so what st-link do you have?
2012-10-27T11:03:13 < tavish> the one on stm32f4-discovery
2012-10-27T11:03:29 < jpa-> is the main processor on the discovery powered on, leds blinking etc?
2012-10-27T11:04:02 < tavish> nope, only st-link is working
2012-10-27T11:04:22 < jpa-> the reset lines are in parallel, so if you remove the jumper to cut power from main processor, it will pull down the reset line
2012-10-27T11:15:21 -!- txf [~txf@bl13-176-136.dsl.telepac.pt] has quit [Ping timeout: 276 seconds]
2012-10-27T11:15:36 -!- txf [~txf@bl13-176-136.dsl.telepac.pt] has joined ##stm32
2012-10-27T11:18:53 < tavish> jpa-: using the trst pin worked fine, I spent 2 days trying to figure out why a simple circuit wasnt working
2012-10-27T11:19:34 -!- tavish [7d3f6ea0@gateway/web/freenode/ip.125.63.110.160] has quit []
2012-10-27T11:23:44 < jpa-> njtrst isn't used for SWD, i guess, so he basically disconnected nrst (which usually isn't strictly necessary anyway)
2012-10-27T11:56:34 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-27T12:08:57 -!- daku is now known as DaKu
2012-10-27T12:13:54 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-27T12:25:07 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-27T12:38:39 -!- DaKu is now known as daku
2012-10-27T12:41:02 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has joined ##stm32
2012-10-27T12:42:49 -!- daku is now known as DaKu
2012-10-27T12:49:34 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-27T12:49:37 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-27T12:54:44 -!- dekar [~dekar@212.255.27.61] has quit [Quit: This computer has gone to sleep]
2012-10-27T12:56:30 -!- DaKu is now known as daku
2012-10-27T12:56:59 -!- dekar [~dekar@212.255.27.61] has joined ##stm32
2012-10-27T12:57:02 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-27T13:00:22 <+Steffann> Still have some doubts if i'll troll my self by backing that paralella KS project
2012-10-27T13:00:45 <+Steffann> They reached their goal, but ..
2012-10-27T13:02:28 -!- dekar_ [~dekar@212.255.27.61] has joined ##stm32
2012-10-27T13:02:31 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-27T13:03:04 -!- dekar [~dekar@212.255.27.61] has quit [Ping timeout: 260 seconds]
2012-10-27T13:03:21 < baird> My housemate/landlord admitted the other day that all the KS project she'd backed have been flakes.. lol. "I /told/ you..."
2012-10-27T13:03:58 <+Steffann> The question is what kind of projects she backed?
2012-10-27T13:04:59 < baird> I was going to say, they're all self-pub fiction book shit. But, still, the usual kickstarter delusion.
2012-10-27T13:07:07 <+Steffann> Yeah, and 120$ is a little to much to get … nothing
2012-10-27T13:08:38 <+Steffann> like that openvizsla project. That project reached its goal a little more than 1.5 years ago
2012-10-27T13:09:38 < baird> OMG! .. found a flickr photography model who /isn't covered in tattoos!/ .. https://secure.flickr.com/search/?w=62009901@N00&q=bonnie#page=0
2012-10-27T13:09:59 < baird> (..and is cute :3)
2012-10-27T13:10:19 <+Steffann> Tattoos photoshopped away
2012-10-27T13:10:47 <+Steffann> You really search for "bonnie" on flickr? :)
2012-10-27T13:11:48 < baird> Well, I was scanning the guy's photostream, and noticed the one good model's name was mentioned-- so did a grep.
2012-10-27T13:12:09 < baird> also, freckles <3
2012-10-27T13:25:30 -!- barthess [~barthess@178.154.8.131] has joined ##stm32
2012-10-27T13:49:47 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has joined ##stm32
2012-10-27T14:03:46 -!- Laurenceb_ [~Laurence@host86-178-234-158.range86-178.btcentralplus.com] has joined ##stm32
2012-10-27T14:09:40 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-27T14:19:03 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-27T14:25:01 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-39.pptp.stw-bonn.de] has joined ##stm32
2012-10-27T14:27:51 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-27T14:30:19 < dongs> hello dongs
2012-10-27T14:30:43 < Erlkoenig> why are you always greeting yourself?
2012-10-27T14:30:56 < qyx_> hello qyx_
2012-10-27T14:31:13 < jpa-> hello Steffann
2012-10-27T14:31:38 < jpa-> oops
2012-10-27T14:31:39 < jpa-> i mean
2012-10-27T14:31:41 < jpa-> hello jpa-
2012-10-27T14:31:45 <+Steffann> hello jpa- how is life?
2012-10-27T14:32:10 < jpa-> Steffann: fine fine.. just accidentally revealed that we two are really the same person
2012-10-27T14:32:22 < Laurenceb_> http://www.youtube.com/watch?v=ak3Q1cHrbFY
2012-10-27T14:32:27 < Laurenceb_> epic lol
2012-10-27T14:32:28 <+Steffann> One half in finland one half in dutchland?
2012-10-27T14:32:33 < jpa-> yeah
2012-10-27T14:33:22 <+Steffann> hehe Laurenceb_
2012-10-27T14:34:17 < dongs> http://i.imgur.com/EhPar.jpg
2012-10-27T14:34:20 < Erlkoenig> superposition!
2012-10-27T14:34:22 < dongs> awesome board from waveshare
2012-10-27T14:34:25 < Erlkoenig> or colocation
2012-10-27T14:34:27 <+Steffann> but they are right Laurenceb_ .. i don't think you want that on your car while driving 120km/h or more
2012-10-27T14:35:44 < Erlkoenig> an UFO :D
2012-10-27T14:35:53 < Laurenceb_> lol the boss
2012-10-27T14:36:06 < Laurenceb_>  "so it landed on the roof and recorded you two fucking then?"
2012-10-27T14:36:35 <+Steffann> If somethings like that lands on my head I would troll the owners.. just don't know how yet :P
2012-10-27T14:36:36 < dongs> LOL
2012-10-27T14:37:41 < baird> Xmas present time! http://www.welovefine.com/3787-rainbow-dash-hug-me-backpack.html
2012-10-27T14:38:54 < dongs> "proably still recording" haha
2012-10-27T14:39:19 < Erlkoenig> baird: there's neither a STM32 or a Zynq on it :/
2012-10-27T14:39:29 <+Steffann> jpa-, I found another couple of halfs who live away from each other, but are actually the same
2012-10-27T14:40:11 < dongs> Laurenceb_: retweeted nova vid everywhere
2012-10-27T14:40:30 < Erlkoenig> a friend found an art history student who looks exactly like me. disturbing
2012-10-27T14:40:38 <+Steffann> Laurenceb_ and baird. Both post the most random links in ##stm32 he/they can find.
2012-10-27T14:40:48 < dongs> what about me
2012-10-27T14:40:53 < dongs> mine are not random enough??
2012-10-27T14:41:20 <+Steffann> They are, but you are dongs
2012-10-27T14:41:41 < Erlkoenig> links!! http://s11.radikal.ru/i183/1208/9f/f72c6bb4f16d.gif
2012-10-27T14:41:53 <+Steffann> You are already more than one, so no space for Laurenceb_ and baird
2012-10-27T14:42:08 -!- tavish [~tavish@unaffiliated/tavish] has quit [Quit: Lost terminal]
2012-10-27T14:42:39 < baird> Someone must be voted off the island.
2012-10-27T14:44:13 < BrainDamage> this should satisfy both baird' furry spirit and general trolling mood then: http://i.imgur.com/VOjKP.jpg
2012-10-27T14:44:13 <+Steffann> Only you, dongs and Laurenceb_ live on an island :P
2012-10-27T14:44:45 < BrainDamage> well, we also do, it's just a bigger one
2012-10-27T14:44:49 < Erlkoenig> (13:34:17) dongs: http://i.imgur.com/EhPar.jpg <- what's that? ZIF for STM32?
2012-10-27T14:45:04 < dongs> Erlkoenig: yeah
2012-10-27T14:45:07 < Erlkoenig> very nice
2012-10-27T14:45:09 < dongs> with uart, xtal, etc connected
2012-10-27T14:45:14 < BrainDamage> they cost 60$ and up
2012-10-27T14:45:24 < dongs> that entire board was 50-somethign
2012-10-27T14:45:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-27T14:45:43 <+Steffann> Too expensive :(
2012-10-27T14:45:50 < dongs> not for whay out get.
2012-10-27T14:45:51 < Erlkoenig> yes sounds lik
2012-10-27T14:45:52 < dongs> out=you
2012-10-27T14:46:11 < BrainDamage> they are low volume and not easy to make
2012-10-27T14:46:34 < dongs> http://www.wvshare.com/column/STM32_Adapter.htm
2012-10-27T14:46:38 < zyp> also limited usability
2012-10-27T14:46:48 < zyp> only fits one footprint
2012-10-27T14:47:01 < dongs> the only footprint I care about.
2012-10-27T14:47:05 < dongs> they do have standalone sockets like that.
2012-10-27T14:47:08 < dongs> wihtout pcb.
2012-10-27T14:47:09 < baird> BrainDamage: cool, except for the gauged earlobes :/
2012-10-27T14:47:11 < zyp> I know
2012-10-27T14:47:32 < Erlkoenig> this socket is insane :D
2012-10-27T14:47:54 < zyp> but anyway, when is running the chips through that faster than just flashing them on the boards where they will be sitting?
2012-10-27T14:48:29 < dongs> its not
2012-10-27T14:48:34 < dongs> im gonna test some code on it.
2012-10-27T14:48:45 < dongs> without having to tear up some other board I got to get to pins i need.
2012-10-27T14:49:18 < zyp> right, so it's about as usable as a breakout board
2012-10-27T14:49:37 < Erlkoenig> maybe one could just get a discovery, test the code there, if it works flash it on the "production" board...
2012-10-27T14:49:41 < zyp> except it's more expensive because of the soxket
2012-10-27T14:49:48 < Laurenceb_> hehe
2012-10-27T14:49:56 < Laurenceb_> yeah discovery double pins is a massive pita
2012-10-27T14:50:00 < BrainDamage> you can swap the test chip easily
2012-10-27T14:50:18 < Laurenceb_> when you fry it
2012-10-27T14:50:29 < BrainDamage> altough for the cost, I guess you could simply buy one breakout board and 1 target chip for every
2012-10-27T14:50:41 < zyp> that was my point, yeah
2012-10-27T14:53:29 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-27T14:56:15 < zyp> if I was into mass producing shit and needed a cheap/fast way to flash boards, I'd probably look into the tag-connect cables
2012-10-27T14:56:25 < zyp> http://www.tag-connect.com/node/52 <- they have a variant for cortex-m
2012-10-27T14:57:37 < Laurenceb_>  Plug-of-Nails
2012-10-27T14:57:40 < Laurenceb_> uh on
2012-10-27T15:00:36 < Laurenceb_> i dont understand
2012-10-27T15:00:45 < Laurenceb_> do you have to hold it down all the time its used?
2012-10-27T15:00:52 < Erlkoenig> http://2g2s.de/elektronik/Die_Mutter_aller_fliegenden_Verdrahtungen_2.JPG
2012-10-27T15:01:14 < Laurenceb_> thats.. tidy
2012-10-27T15:01:35 < Erlkoenig> a bit :D
2012-10-27T15:01:42 < Erlkoenig> didn't work too well :>
2012-10-27T15:03:12 <+Steffann> Laurenceb_ didn't spot the arduino?
2012-10-27T15:03:23 <+Steffann> 2 actually
2012-10-27T15:04:20 < BrainDamage> is the fan to cool the arduino?
2012-10-27T15:04:41 < BrainDamage> mmm looks like a rfm module
2012-10-27T15:04:59 <+Steffann> 12 indeed
2012-10-27T15:05:28 <+Steffann> LOL @ Tape Erlkoenig :)
2012-10-27T15:05:41 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-27T15:05:48 < dongs> http://www.indiegogo.com/cubieboard
2012-10-27T15:05:51 < dongs> expensive shitberry pi
2012-10-27T15:06:24 < Laurenceb_> interesting - apple are stikcing stellaris on all their motherboards
2012-10-27T15:06:24 < zyp> Laurenceb_, in some cases, yes
2012-10-27T15:06:43 < Laurenceb_> maybe it does uefi or something
2012-10-27T15:06:53 < zyp> Laurenceb_, and if you are just flashing the boards during production, what's the problem with that?
2012-10-27T15:07:27 <+Steffann> Indeed dongs.. yet another allwinner 10 board
2012-10-27T15:07:40 < dongs> lemme guess ,allwinner sucks?
2012-10-27T15:08:12 <+Steffann> I think you can find at least 1000 other hobby bobby we want your money cround funding projects
2012-10-27T15:08:12 < Erlkoenig> (14:05:28) Steffann: LOL @ Tape Erlkoenig :) <-- tape is important :P
2012-10-27T15:08:17 <+Steffann> with that chip
2012-10-27T15:08:25 < dongs> heh
2012-10-27T15:09:04 < Laurenceb_> datasheeet is published
2012-10-27T15:09:10 <+Steffann> Since when?
2012-10-27T15:09:11 < Laurenceb_> you can make your own
2012-10-27T15:09:20 < Laurenceb_> i thought olimex had it?
2012-10-27T15:09:25 < Laurenceb_> or at least the basics
2012-10-27T15:09:29 < Laurenceb_> maybe not full
2012-10-27T15:09:36 < zyp> it's a singlecore A8, not a whole lot of performance compared to quad-A9 chips
2012-10-27T15:11:03 < Laurenceb_> but beats 700mhz arm11
2012-10-27T15:11:12 < zyp> yep
2012-10-27T15:11:52 <+Steffann> Hey, you can make the 700 1Ghz now :P
2012-10-27T15:11:57 <+Steffann> 700MHz
2012-10-27T15:28:25 < Laurenceb_> that thing has sata, IR, RTC and line in
2012-10-27T15:28:34 < Laurenceb_> and 4GB onboard flash
2012-10-27T15:28:43 < Laurenceb_> useful features
2012-10-27T15:29:10 < Laurenceb_> its pretty obviously designed as a media center :P
2012-10-27T15:37:35 < zyp> well, that's pretty much what the A10 was designed for as far as I've understood
2012-10-27T15:37:59 < Laurenceb_> http://en.wikipedia.org/wiki/Jimmy_Wales
2012-10-27T15:38:07 < Laurenceb_> In 1996, he and two partners founded Bomis, a male-oriented web portal featuring entertainment and adult content. The company would provide the initial funding for the peer-reviewed free encyclopedia Nupedia (2000–2003) and its successor, Wikipedia.
2012-10-27T15:38:12 < Laurenceb_> hahaha i didnt know that
2012-10-27T15:38:20 < dongs> ooold
2012-10-27T15:38:29 < Laurenceb_> BROmis
2012-10-27T15:38:30 < Erlkoenig> ha great :D
2012-10-27T15:40:50 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Ping timeout: 255 seconds]
2012-10-27T15:41:09 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-27T15:43:42 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-27T15:43:43 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-27T15:45:00 < baird> TI is getting Red Hat to support the mspgcc toolchain (i.e: the actual free one) .. http://blog.gmane.org/gmane.comp.hardware.texas-instruments.msp430.gcc.user/day=20121026
2012-10-27T15:46:37 < Erlkoenig> isn't it rather "Red Had getting TI" ;-)
2012-10-27T15:48:21 < Robint91> hi all
2012-10-27T15:48:46 < baird> dekar_: GO FOR THE MONEY! :)  (although the primary mspgcc developer seems to have been left out-- not that he's too upset about it)
2012-10-27T15:48:47 <+Steffann> All just left
2012-10-27T15:49:10 < Robint91> ...
2012-10-27T15:49:20 < Robint91> what IDE do you guys use? CooCox?
2012-10-27T15:49:35 < zyp> vim
2012-10-27T15:49:55 < baird> emacs
2012-10-27T15:50:13 < zyp> I don't think there is a concensus on what to use, so use whatever you like
2012-10-27T15:50:56 <+Steffann> What to do if there is nothing you really like?
2012-10-27T15:50:59 < baird> (A lot of us use the FOSS toolchains, which isn't tied the better-know IDEs)
2012-10-27T15:51:36 < zyp> Steffann, then use what you dislike the least
2012-10-27T15:52:24 < Erlkoenig> eclipse + arm-gcc
2012-10-27T15:52:30 < dongs> keil 4 lyfe
2012-10-27T15:52:31 < Erlkoenig> + texane st-util
2012-10-27T15:54:17 < baird> http://i.imgur.com/xhaha.jpg
2012-10-27T15:54:24 <+Steffann> What did you use for your vmarker Robint91 ?
2012-10-27T15:54:47 < Laurenceb_> i just use gedit
2012-10-27T15:55:02 < Laurenceb_> xhaha
2012-10-27T15:55:12 < Robint91> Steffann, atollic first, moved to CooCox, but it isn't stable enough
2012-10-27T15:55:15 < dongs> if youre using geidt.. might as well just use pico
2012-10-27T15:55:18 < Robint91> or my ST-Links
2012-10-27T15:55:33 < dongs> Robint91: pickup keil ulink2 + mdk-arm
2012-10-27T15:55:39 < Laurenceb_> gedit has a gui
2012-10-27T15:55:40 < dongs> you will finally stop wasting time on broken tools
2012-10-27T15:55:43 < dongs> and focus on getting work done
2012-10-27T15:55:45 < Laurenceb_> i have command line editors
2012-10-27T15:56:04 < Laurenceb_> *hate
2012-10-27T15:56:39 < BrainDamage> the most annoying thing of gedit is lack of gdb integration
2012-10-27T15:56:44 <+Steffann> keil stuff is awful dongs
2012-10-27T15:56:51 < dongs> stefann, i disagree.
2012-10-27T15:56:53 < BrainDamage> if you install rabbitvcs you can have git integrtion
2012-10-27T15:58:00 < Laurenceb_> interesting - msp430 current/mhz is about the same as L1
2012-10-27T15:58:11 < Robint91> dongs, no M4 support
2012-10-27T15:58:18 < dongs> you fucking shiting me?
2012-10-27T15:58:31 < dongs> i must have been using keil wiht M4 for the last 4+ months in my imagination
2012-10-27T15:58:33 < zyp> Laurenceb_, so L1 has better performance per current?
2012-10-27T15:58:41 < Laurenceb_> not quite
2012-10-27T15:58:55 < Laurenceb_> msp430 at 4mhz would be 720ua from datasheet
2012-10-27T15:58:57 < zyp> doesn't L1 have better performance per mhz?
2012-10-27T15:59:08 < Laurenceb_> i measure my L1 as 780ua
2012-10-27T15:59:10 < Laurenceb_> yeah
2012-10-27T15:59:13 < Laurenceb_> much better
2012-10-27T15:59:15 < zyp> exactly
2012-10-27T15:59:29 < Laurenceb_> so msp430 is dead XD
2012-10-27T15:59:40 < Laurenceb_> its not even good at low power
2012-10-27T15:59:46 < Robint91> dongs, mhh
2012-10-27T15:59:46 < zyp> so for a given application that would spend most of it's time in sleep, L1 should use less power
2012-10-27T15:59:53 < dongs> Robint91: where the fuck did you get the idea it doesnt support M4?
2012-10-27T16:00:04 < Robint91> dongs, I have seen it somewhere
2012-10-27T16:00:06 < dongs> Robint91: it supports stm32f4 (since months ago) and latest update added support for stm32f3
2012-10-27T16:00:11 < Robint91> dongs,  okay
2012-10-27T16:00:20 < dongs> and of course all the other M4 cores out there
2012-10-27T16:00:23 < dongs> like kinetis shit
2012-10-27T16:00:24 < dongs> and lpc
2012-10-27T16:00:24 < dongs> etc
2012-10-27T16:00:33 < Robint91> dongs, how is the IDE?
2012-10-27T16:00:48 < dongs> works just fine, considering F/OSS alternatives (lol).
2012-10-27T16:01:07 <+Steffann> Keep in mind dongs is really anit-opensores Robint91
2012-10-27T16:01:11 <+Steffann> *anti
2012-10-27T16:01:11 < Robint91> dongs, some code completion
2012-10-27T16:01:53 < dongs> Robint91: nope.
2012-10-27T16:02:30 < Laurenceb_> WHY NO flip chip F30x
2012-10-27T16:02:33 < Laurenceb_> so annyoing
2012-10-27T16:02:44 < dongs> I got used to not having code complete,, it does have code navigation at lest
2012-10-27T16:02:45 < dongs> least
2012-10-27T16:02:51 < dongs> go to reference, go to declaration etc.
2012-10-27T16:02:56 < dongs> enough to crossreference/find stuff.
2012-10-27T16:03:08 < Robint91> dongs, I'm spoiled with working with visual studio
2012-10-27T16:03:16 < dongs> yeah, for sure.
2012-10-27T16:04:02 < Laurenceb_> eww
2012-10-27T16:05:41 < dongs> if laurenceTROLL would write his code in vstudio it would have more spaces at least
2012-10-27T16:06:19 <+Steffann> laurduino doesn't need that
2012-10-27T16:06:29 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-27T16:09:30 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-27T16:10:06 < Robint91> dongs, and for C++?
2012-10-27T16:10:16 < dongs> armcc does C++.
2012-10-27T16:10:40 < dongs> probably not C++11 or wahtever teh fuck zyp uses but from what I know it works. never used any C++ shit for embedded.
2012-10-27T16:12:41 < Robint91> dongs, what is your opinion on running .NET MF
2012-10-27T16:12:56 < dongs> Robint91: trash, interrupt latency is like useless.
2012-10-27T16:13:02 < dongs> i dont think its useful for anything srs.
2012-10-27T16:14:11 < Laurenceb_> wow
2012-10-27T16:14:16 < Laurenceb_> something i agree with
2012-10-27T16:18:20 < Laurenceb_> hehe my loser friend is on github
2012-10-27T16:18:31 < Laurenceb_> https://github.com/offchooffcho/Quake-III-Arena
2012-10-27T16:20:25 < Laurenceb_> there should be a "productivity ratio" thingy on there
2012-10-27T16:20:43 < Laurenceb_> amount of useful stuff / useless crap
2012-10-27T16:21:43 < Erlkoenig> linux would have been rated "useless" in its beginnings
2012-10-27T16:21:48 < BrainDamage> so we can have competitions?
2012-10-27T16:21:58 <+Steffann> At least your loser friend is married/has a gf Laurenceb_ :P
2012-10-27T16:22:03 < Laurenceb_> what
2012-10-27T16:22:14 < Laurenceb_> no way in hell
2012-10-27T16:22:35 <+Steffann> His website is a lie?
2012-10-27T16:22:36 < Laurenceb_> he has a dandruff/pizza box/body odor trail behind him
2012-10-27T16:23:38 < Laurenceb_> where?
2012-10-27T16:24:01 <+Steffann> http://www.tbradshaw.net
2012-10-27T16:24:43 < Laurenceb_> wutr
2012-10-27T16:24:53 < Laurenceb_> where was that on github?
2012-10-27T16:25:02 <+Steffann> his profile
2012-10-27T16:25:14 <+Steffann> No fail me :P
2012-10-27T16:25:28 <+Steffann> Wrong profile
2012-10-27T16:25:36 < Laurenceb_> yeah i dont know that guy
2012-10-27T16:26:00 <+Steffann> Would be funny though.. if he got married without inviting you :P
2012-10-27T16:27:57 < Laurenceb_> his life consists purely of all night gaming
2012-10-27T16:29:19 < dongs> nice, my 2 SPI running off timer troll is successful http://i.imgur.com/Rovp8.png
2012-10-27T16:29:26 < dongs> SPI1 is outputting 0xAA, SPI2 is 0x55
2012-10-27T16:29:31 < dongs> perfect timing
2012-10-27T16:31:08 < Robint91> http://i.imgur.com/WcyRo.gif
2012-10-27T16:31:20 < Robint91> Tektronix
2012-10-27T16:32:04 < Laurenceb_> how do you run them off timer?
2012-10-27T16:32:38 < dongs> Laurenceb_: slave mode :D
2012-10-27T16:32:44 < dongs> http://bcas.tv/paste/results/meA9jE16.html
2012-10-27T16:32:57 < Erlkoenig> can i get the F4 to continuously shift in bits in a SPI-slave-like manner, and to compare the shifted in byte with a constant after every bitshift, and fire an interrupt upon match?
2012-10-27T16:33:05 < dongs> no
2012-10-27T16:34:13 < zyp> but sgpio in lpc4300 apparently can do just that
2012-10-27T16:34:25 < dongs> well isnt that FUCKING GREAT
2012-10-27T16:34:26 < dongs> bedtime
2012-10-27T16:35:38 <+Steffann> No, not really dongs
2012-10-27T16:36:03 <+Steffann> 22:36 ..
2012-10-27T16:36:14 <+Steffann> It's WEEKEND
2012-10-27T16:36:22 <+Steffann> Even in japland
2012-10-27T16:36:38 < Erlkoenig> doesn't the CMSIS or StdPeriph include a constant for the numer of present IRQ Lines?
2012-10-27T16:37:31 < Laurenceb_> i dont see where the spi is slaved off the timer?
2012-10-27T16:37:49 < Erlkoenig> like max(IRQn_Type) or something :> ... not counting the fault handlers
2012-10-27T16:44:12 < Laurenceb_> oh it has external clk input
2012-10-27T16:44:14 < Laurenceb_> i see
2012-10-27T16:50:16 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-27T16:56:15 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Ping timeout: 256 seconds]
2012-10-27T17:20:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-27T17:55:01 <+dekar_> Laurenceb_, so he just forks random stuff?
2012-10-27T17:55:16 < Laurenceb_> that guy?
2012-10-27T17:55:20 <+dekar_> yeah
2012-10-27T17:55:29 < Laurenceb_> yeah hes forked a lot of stuff and made 0 contributions
2012-10-27T17:55:39 <+dekar_> why would one do that?
2012-10-27T17:56:45 < Laurenceb_> because your life consists of eating pizza, popping your spots and all night gaming
2012-10-27T17:57:01 < Laurenceb_> hopefully he will be fired soon
2012-10-27T17:58:13 < jpa-> there is one interesting case on github where someone copied my library, renamed it and every part of docs & source that referred to the name and then did nothing else
2012-10-27T17:58:50 -!- Claude is now known as claude
2012-10-27T17:59:01 < jpa-> (yeah, the license allows that and i have nothing against it, but nevertheless quite curious :)
2012-10-27T18:01:09 -!- claude is now known as Claude
2012-10-27T18:02:50 < Laurenceb_> my boss actually realised this guy has done 0 work in the two years hes been employed
2012-10-27T18:03:05 < Laurenceb_> but it took 2 years for that to sink in O_o
2012-10-27T18:04:42 < BrainDamage> and you're wondering how many will it take for you?
2012-10-27T18:04:58 < Laurenceb_> i see what you did thar
2012-10-27T18:05:47 < jpa-> nah Laurenceb_ works hard and then loses it all to a homebrew backup solution gone wrong :)
2012-10-27T18:29:20 < Erlkoenig> might it be possible that clearing the TIM10->SR.UIF flag has no effect when the IRQ is disabled?
2012-10-27T18:37:03 < Erlkoenig> f*ck, seems so
2012-10-27T18:40:41 -!- piele_gone is now known as piele
2012-10-27T19:05:22 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-27T19:11:19 < Erlkoenig> wtf. calling SVC while PRIMASK=1 causes HardFaults?
2012-10-27T19:11:41 <+Steffann> Lots of trouble today Erlkoenig ?
2012-10-27T19:12:04 < Erlkoenig> seems so :S
2012-10-27T19:12:09 < Erlkoenig> nothing works
2012-10-27T19:33:33 <+Steffann> At least this channel isn't very yankielandish flyback
2012-10-27T19:33:34 < Erlkoenig> argh... i'd like to do context switches when interrupts arrive. i have a generic ISR that decides based in IPSR to which task to switch, disables the IRQn in the NVIC, and does the context switch. The task then should clean up the interrupt flags (e.g. the UIF flag for Timers). Unfortunately, as long as the IRQ is disabled, the UIF is always '1' and refuses to be reset,
2012-10-27T19:33:35 < Erlkoenig>  which means that the timer interrupt executes instantly again after the task decides to wait for the timer IRQ... it would be nice if the "scheduler/contextswitcher" doesn't know anything about specific timer interrupts, and just sees NVIC IRQ's, but this seems impossible...
2012-10-27T19:34:56 < zyp> interesting
2012-10-27T19:35:21 < zyp> I've been thinking about taking the same approach for the usb handling thread
2012-10-27T19:37:06 < Erlkoenig> maybe i should try to reset the UIF flag right after re-enabling the NVIC IRQ... but this is very unclean, assuming i have a interrupt source that might "fire" fast, and the flag then indicates that the event has been triggered again
2012-10-27T19:59:09 -!- BusError [~michel@host86-171-235-55.range86-171.btcentralplus.com] has joined ##stm32
2012-10-27T19:59:35 < Erlkoenig> hmmm no hints in the errata sheet
2012-10-27T20:01:01 < zyp> sure you're not just doing it wrong?
2012-10-27T20:01:46 < Erlkoenig> well uhm without the NVIC-IRQ disabling/enabling magic, the timer works as expected
2012-10-27T20:02:38 < Erlkoenig> i analyzed everything very close with the debugger... i'm quite sure that the "UIF" flag causes the problem
2012-10-27T20:04:54 <+Steffann> We hear "i'm quite sure" even without the "quite" pretty often here. And most of the times it turns out that person was wrong ( wont tell you who that person is )
2012-10-27T20:05:08 <+Steffann> LO
2012-10-27T20:05:11 <+Steffann> *:P
2012-10-27T20:05:13 < Erlkoenig> hmmm
2012-10-27T20:05:30 < Erlkoenig> i try to collect more evidence :>
2012-10-27T20:05:45 <+Steffann> ** it's Laurenceb_ **
2012-10-27T20:10:02 < Erlkoenig> commented out just the Enabling/Disabling magic, left all context switching stuff intact - works.
2012-10-27T20:10:36 < Erlkoenig> this would force me to write very unelegant code :S
2012-10-27T20:10:44 < Erlkoenig> especially copy&pasta code
2012-10-27T20:11:09 < zyp> sounds like you're doing something wrong
2012-10-27T20:11:24 < Erlkoenig> or the timer hardware...
2012-10-27T20:14:16 < Erlkoenig> this is probably the timer's revenge for not disabling the interrupt via the timer register, but via the NVIC register
2012-10-27T20:15:30 <+Steffann> STM32 wars: Episode V - The timer strikes back
2012-10-27T20:16:13 < Erlkoenig> i have the bigger weapons - nice inductors
2012-10-27T20:16:16 < Robint91> +1 for Steffann
2012-10-27T20:23:37 <+Steffann> Have fun
2012-10-27T20:26:45 < Erlkoenig> ... akf
2012-10-27T20:26:46 < Erlkoenig> afk
2012-10-27T20:26:54 -!- Erlkoenig [~erlkoenig@pptp-212-201-76-39.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-27T20:30:20 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-27T20:38:46 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 245 seconds]
2012-10-27T20:40:44 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-27T20:46:03 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 260 seconds]
2012-10-27T20:47:40 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-27T20:52:55 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-27T20:54:14 -!- Erlkoenig [~erlkoenig@pptp-212-201-69-221.pptp.stw-bonn.de] has joined ##stm32
2012-10-27T20:54:29 < Erlkoenig> holy shit it's fucking cold
2012-10-27T20:55:17 <+Steffann> According to google it's 2 °C here outside
2012-10-27T20:55:27 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 265 seconds]
2012-10-27T20:55:32 <+Steffann> And sunny
2012-10-27T20:55:35 <+Steffann> Hmm
2012-10-27T20:55:36 < BrainDamage> 16 here
2012-10-27T20:55:44 < BrainDamage> but it's raining :/
2012-10-27T20:56:08 <+Steffann> Yeah, buy sunny .. sunny in the dark :)
2012-10-27T20:56:25 <+Steffann> *but
2012-10-27T20:56:33 < Erlkoenig> better not leave the house
2012-10-27T20:56:41 <+Steffann> How cold its there?
2012-10-27T20:56:44 < BrainDamage> it could be sunny, just transparent ground
2012-10-27T20:57:07 < Erlkoenig> probably around 0°C
2012-10-27T20:57:15 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-27T20:58:35 <+Steffann> If the sun moves around by an unlimited speed and the speed of light was unlimited .. would it light as wel
2012-10-27T20:58:37 <+Steffann> l
2012-10-27T20:59:43 < Erlkoenig> if i every start doing ham radio i definitely need a beautiful morse button with brass decorations and stuff
2012-10-27T20:59:56 <+Steffann> Too bad that's not the case, so it's dark
2012-10-27T21:00:26 <+Steffann> You start to understand why we thought you are female Erlkoenig ? :P
2012-10-27T21:00:59 < Erlkoenig> wat. if it's technology and serves an acutal purpose, it is allowed to look nice :>
2012-10-27T21:01:35 <+Steffann> but it's JUST A button
2012-10-27T21:01:58 < BrainDamage> will you knit a cover for the button too?
2012-10-27T21:02:04 < Erlkoenig> but the historical reference
2012-10-27T21:06:47 < emeb> Don't forget - the physical construction of a tool like a morse key affects it's usability. Things like angle, mass, throw all contribute to the ergonomics of the tool and make a big difference to someone who uses it often. It's more than just a button.
2012-10-27T21:07:43 < Erlkoenig> probably... i guess the old construction was somewhat ergonomic
2012-10-27T21:10:43 < BrainDamage> but in order for those to be valuable, he'd have to use it constantly
2012-10-27T21:10:57 < BrainDamage> which means he'd be off irc, woo!
2012-10-27T21:11:26 < emeb> unless he uses morse code instead of a keyboard on his computer.
2012-10-27T21:11:40 < emeb> what's morse for @?
2012-10-27T21:12:00 < BrainDamage> you run out in problems to do spaces
2012-10-27T21:12:09 < BrainDamage> no such symbol afaik
2012-10-27T21:12:31 < emeb> keep a real keyboard around and hit the space bar with your other hand?
2012-10-27T21:13:06 < Erlkoenig> wah. i'd not have it for constant use, just because it would be awesome to have such a selfmade thing standing around
2012-10-27T21:13:12 < BrainDamage> good thing we were taking about comfort ...
2012-10-27T21:13:22 < emeb> ergonomics fail!
2012-10-27T21:13:48 < Erlkoenig> for comfort, i'd plug a PS2 keyboard to a mikrocontroller and have it generate the morse sequences and vice versa with an LCD for receiving
2012-10-27T21:13:53 < emeb> Benutzerfreundlichkeit!
2012-10-27T21:13:59 < Erlkoenig> genau
2012-10-27T21:14:24 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-27T21:20:47 -!- daku is now known as DaKu
2012-10-27T21:22:45 -!- Claude is now known as claude
2012-10-27T21:33:58 < Erlkoenig> "src/main.c:188:25: error: local frame unavailable (naked function?)"  ... what O_o ... yes it is a naked function, but what exactly is the problem oO
2012-10-27T21:34:19 < emeb> naked functions don't have code to set up a local frame.
2012-10-27T21:34:30 < emeb> ?
2012-10-27T21:34:34 < jpa-> naked functions just write to random points in memory
2012-10-27T21:34:49 < Erlkoenig> what, not relative to SP?!
2012-10-27T21:35:29 < jpa-> "The only statements that can be safely included in naked functions are asm statements that do not have operands."
2012-10-27T21:35:33 < Erlkoenig> declaring the function as "naked" stops the compiler from reserving space for local struct on the stack or what?
2012-10-27T21:35:49 < emeb> pretty much
2012-10-27T21:35:54 < Erlkoenig> hm
2012-10-27T21:36:41 < Erlkoenig> how to tell the compiler to not do any PUSH/POP at the beginning/end of the function, and just assume all registers are free to use, but still use the Stack normally?
2012-10-27T21:37:01 < jpa-> Erlkoenig: you can say "noreturn" if that is what you mean
2012-10-27T21:37:06 < emeb> use naked and write your own prolog/epilog?
2012-10-27T21:37:07 < jpa-> why else would any register be free to use?
2012-10-27T21:38:33 < Erlkoenig> the function is a task started by my "scheduler", and the context switcher makes sure that the function can just use all registers
2012-10-27T21:38:45 < Erlkoenig> well technically the function will never return
2012-10-27T21:38:56 < Erlkoenig> (or should... need to take care of that :o) )
2012-10-27T21:39:18 < Erlkoenig> but noreturn looks good...
2012-10-27T21:42:46 < Erlkoenig> meh, still generates "push {lr}" when compiling with -O2, and "push {r7, lr}" without optimization
2012-10-27T21:43:18 < zyp> that's why you should only use pendsv for the context switch, and give it the naked attribute
2012-10-27T21:43:43 < jpa-> horrible 4 bytes of waste
2012-10-27T21:43:46 < Erlkoenig> hu? i'm using the SVCall Interrupt for the context switch, and it is "naked"
2012-10-27T21:43:57 < emeb> oh, the humanity!
2012-10-27T21:43:59 < zyp> or maybe I misunderstood
2012-10-27T21:44:11 < jpa-> zyp: i guess he means the thread bodies, which are often declared as just normal functions and unnecessarily save the registers when they start
2012-10-27T21:44:19 < Erlkoenig> exactly
2012-10-27T21:44:42 < zyp> oh
2012-10-27T21:44:46 < Erlkoenig> prior to doing the first context switch to the newly created thread, all registers are saved
2012-10-27T21:45:01 < Erlkoenig> also, the thread function won't ever return, so there are no registers that need to be saved
2012-10-27T21:45:18 < zyp> thread functions might well return
2012-10-27T21:45:31 < Erlkoenig> well not in my case :>
2012-10-27T21:45:46 < Erlkoenig> but if they did, they don't need to restore registers...
2012-10-27T21:46:10 < Erlkoenig> hmm... even the main () has a "Push" at the beginning, even though that one should never return
2012-10-27T21:48:11 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-27T21:50:03 -!- alexn [~alexn@dslb-094-216-058-053.pools.arcor-ip.net] has joined ##stm32
2012-10-27T21:52:48 < emeb> how does the compiler know that?
2012-10-27T21:53:03 < Erlkoenig> could theoretically have been hardcoded or something
2012-10-27T21:53:18 < jpa-> lol
2012-10-27T21:53:40 < jpa-> main() often returns in all stuff except microcontrollers :)
2012-10-27T21:53:44 < zyp> I suggest stop trying to solve non-problems
2012-10-27T21:53:45 < Thorn> gcc has attributes to turn it off iirc
2012-10-27T21:54:07 < jpa-> but yeah, don't worry about a few bytes of waste here and there
2012-10-27T21:54:22 < Erlkoenig> yes normally main() returns... but even then the registers don't matter
2012-10-27T21:54:30 < emeb> I spent  some time fooling with hand-optimized assembly. Then I looked at the difference it made in execution speed and gave up.
2012-10-27T21:54:47 < Erlkoenig> i wonder what avr-gcc does - on such tiny devices single bytes matter
2012-10-27T21:55:01 < zyp> avr has a completely different ABI
2012-10-27T21:56:43 < jpa-> http://www.tty1.net/blog/2008-04-29-avr-gcc-optimisations_en.html  they use __attribute__ ((noreturn))
2012-10-27T21:57:16 < emeb> "Premature optimization is the root of all evil" - Donald Knuth
2012-10-27T21:57:33 < Erlkoenig> i disagree
2012-10-27T21:58:21 < Erlkoenig> the (data) structure(s) of programs have the largest influence on performance... if you don't care about performance at their design, you have to rewrite everything once you want to optimize it
2012-10-27T21:58:59 < emeb> Then it's not premature.
2012-10-27T21:59:39 < jpa-> your stuff, however, is not about data structures and wide performance implications; it's about saving 4 bytes on devices that have 4096 bytes or more
2012-10-27T21:59:42 < emeb> Note Knuth is not saying that you should never optimize. But if you try to optimize before you fully understand the problem then you're wasting time and creating problems you'll be dealing with again later.
2012-10-27T22:00:16 < Erlkoenig> i think i understand my problem here
2012-10-27T22:00:42 < Erlkoenig> jpa-: that's correct... however i find it somewhat unclean... whatever i'll leave it as it is
2012-10-27T22:02:04 < jpa-> Erlkoenig: when you code in C, you should not obsess about the assembler output too much
2012-10-27T22:02:19 < Erlkoenig> hmm probably
2012-10-27T22:02:25 < jpa-> when you write C compilers, you can do that :)
2012-10-27T22:03:53 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 245 seconds]
2012-10-27T22:06:50 < emeb> Worth reading on the topic: http://c2.com/cgi/wiki?PrematureOptimization
2012-10-27T22:10:36 -!- barthess [~barthess@178.154.8.131] has quit [Quit: Leaving.]
2012-10-27T22:26:48 -!- txf [~txf@bl13-176-136.dsl.telepac.pt] has quit [Ping timeout: 245 seconds]
2012-10-27T22:34:32 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-27T22:38:39 -!- txf [~txf@bl12-89-55.dsl.telepac.pt] has joined ##stm32
2012-10-27T22:42:11 -!- txf [~txf@bl12-89-55.dsl.telepac.pt] has quit [Read error: Connection reset by peer]
2012-10-27T22:55:03  * emeb has run out of motivation to work on things.
2012-10-27T22:55:09 -!- barthess [~barthess@178.154.8.131] has joined ##stm32
2012-10-27T22:55:20 < Erlkoenig> i could provide you with huge amounts of negative motivation
2012-10-27T22:55:25  * jpa- has not had motivation for weeks
2012-10-27T22:55:54 < emeb> flyback: opposite end of the country. Weather here is lovely.
2012-10-27T22:57:04 <+dekar_> my 75w uber high efficiency HTPC PSU has caps with bulges, one of my four drives kept losing the sata link and the 12V rail was down to 10.2V :/
2012-10-27T22:57:22 < emeb> whee...
2012-10-27T22:58:58 < emeb> I kept hearing strange snapping noises over a period of years. Then one day I was cleaning dust out of the PC and spotted this: https://twitter.com/i/#!/ebrombaugh/media/slideshow?url=pic.twitter.com%2FPFHTLfrS
2012-10-27T22:59:01 < Erlkoenig> it's funny how the american way to build structures always hits them back
2012-10-27T22:59:21 < emeb> Erlkoenig: I doubt that's uniquely American.
2012-10-27T22:59:40 < Erlkoenig> well at least the americans had the tec and money to do it the right wayâ„¢
2012-10-27T22:59:58 <+dekar_> I don't think I have ever seen those closed capacitors burst open
2012-10-27T23:00:04 < emeb> It's human nature to go with the lowest bidder.
2012-10-27T23:00:12 < emeb> dekar_:  Iknowright?
2012-10-27T23:00:19 < Erlkoenig> ... unless law require you to do it the right wayâ„¢
2012-10-27T23:02:47 < emeb> I replaced the whole video card. It was getting ancient anyway.
2012-10-27T23:04:33 <+dekar_> flyback, I am thinking about dumping the whole thing, one of the connectors looks charred
2012-10-27T23:05:14 <+dekar_> ZHC brand
2012-10-27T23:06:15 <+dekar_> I wouldn't have expected it to fail though, given that it's just a small switching DC/DC converter getting 12V input
2012-10-27T23:06:36 <+dekar_> it has an external wall wart doing the rest
2012-10-27T23:07:40 < emeb> I suppose it's not about the stresses the caps see, but the materials they're made from.
2012-10-27T23:07:46 -!- txf [~txf@bl13-160-172.dsl.telepac.pt] has joined ##stm32
2012-10-27T23:08:09 < emeb> bad electrolytes -> eventual failure no matter what the voltages.
2012-10-27T23:09:45 <+dekar_> the PSU I use now draws 10w more :/
2012-10-27T23:12:14 <+dekar_> so yeah I'll probably fix it
2012-10-27T23:12:17 < Laurenceb_> http://www.fsf.org/blogs/community/gnus-trick-or-treat-at-windows-8-launch
2012-10-27T23:12:20 < Laurenceb_> trolled
2012-10-27T23:12:31 <+dekar_> energy is expensive in Germany, that'd be about 20€ per year
2012-10-27T23:13:46 < emeb> use leyden jars.
2012-10-27T23:16:10 <+dekar_> jars… I'll just get two huge metal plates :)
2012-10-27T23:17:23 < emeb> can't find a more efficient replacement?
2012-10-27T23:17:29 < emeb> (ps)
2012-10-27T23:17:45 <+dekar_> those picoPSU things are pretty expensive
2012-10-27T23:18:01 <+dekar_> don't feel like buying another one, I'd rather fix the caps
2012-10-27T23:18:03 < emeb> ah - that changes the equation.
2012-10-27T23:18:28 < emeb> recap the whole thing then, with parts from a trusted mfg.
2012-10-27T23:19:09 < prog> "Special thanks to our volunteers, Linnea the GNU and Tristan the photographer, for helping make this action a huge success!"
2012-10-27T23:19:16 < prog> what a huge success, indeed!
2012-10-27T23:19:36 < emeb> yo prog
2012-10-27T23:19:36 <+dekar_> actually they're starting at 20€ on ebay these days, they were thrice that price two years ago
2012-10-27T23:19:57 <+dekar_> I'll probably fix it anyway, just two caps
2012-10-27T23:20:19 < prog> sup emeb
2012-10-27T23:20:40 < emeb> prog: not much. How's the SDR project?
2012-10-27T23:20:53 < prog> going fine
2012-10-27T23:20:56 < quitte> Does anyone happen to have one of those: http://www.logilink.org/showproduct/CR0015.htm ? I asked earlier but still would like to know if it has an epoxy blob or a cased chip.
2012-10-27T23:20:58 -!- BrainDamage [~BrainDama@dynamic-adsl-94-36-228-233.clienti.tiscali.it] has quit [Ping timeout: 246 seconds]
2012-10-27T23:21:04 < prog> I haven't had the time to test on the F4 though
2012-10-27T23:21:14 <+dekar_> flyback, I guess you're right, meh :/
2012-10-27T23:21:16 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-27T23:21:33 <+dekar_> flyback, ZHC
2012-10-27T23:22:29 < emeb> prog: SDR on an F4 would be cool. Probably needs some careful hardware design.
2012-10-27T23:22:47 < prog> absolutely. too much work
2012-10-27T23:23:25 < prog> I had a few things to finalize in the desktop version for the FCD+ launch
2012-10-27T23:23:57 < prog> but hey, FCD+ can do fine with a F4 ;)
2012-10-27T23:24:07 < emeb> Yeah - it's just audio rate.
2012-10-27T23:24:33 < jpa-> Laurenceb did software GPS on F4
2012-10-27T23:24:36 < emeb> You could probably hook the FCD rf front-end up straight to an F4 via I2S instead of dealing with USB.
2012-10-27T23:25:17 < emeb> get that crazy PIC24 out of the picture completely. :)
2012-10-27T23:25:29 <+dekar_> flyback, yes they are
2012-10-27T23:25:49 <+dekar_> and I can't find anything about that company either
2012-10-27T23:26:35 <+dekar_> 2200µF 16V rated -40-105°C
2012-10-27T23:27:01 <+dekar_> flyback, that bulge on top agrees with you :)
2012-10-27T23:27:30 <+dekar_> just ZHC, I can make a pic if you care
2012-10-27T23:29:00 <+dekar_> flyback, I can always go for low ESR caps, can't I?
2012-10-27T23:29:30 -!- txf [~txf@bl13-160-172.dsl.telepac.pt] has quit [Ping timeout: 264 seconds]
2012-10-27T23:30:10 <+dekar_> flyback, well it isn't an actual picoPSU brand PSU, but a clone that came with some HTPC case I bought a few years ago, the case I already discarded
2012-10-27T23:30:17 -!- txf [~txf@bl13-160-172.dsl.telepac.pt] has joined ##stm32
2012-10-27T23:31:37 <+dekar_> well the AC adapter says LC75ITX
2012-10-27T23:31:47 <+dekar_> the DC/DC thing is just a PCB without markings
2012-10-27T23:34:28 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-27T23:36:39 <+dekar_> AC Adaptor, RoSH compliant ;)
2012-10-27T23:37:12 <+dekar_> silkscreen only has weird markings
2012-10-27T23:37:25 <+dekar_> YEHE A-AD12VC-R22
2012-10-27T23:37:46 <+dekar_> THE FILES 23TH/MAY,07
2012-10-27T23:38:41 < emeb> http://translate.googleusercontent.com/translate_c?spm=a230r.1.10.16.Ll3Q6F&depth=1&hl=en&prev=/search%3Fq%3DYE%2BHE%2BA-AD12VC-R22%26hl%3Den%26sa%3DX%26biw%3D1159%26bih%3D897%26prmd%3Dimvns&rurl=translate.google.com&sl=zh-CN&twu=1&u=http://item.taobao.com/item.htm%3Fid%3D13926536102&usg=ALkJrhiJq_mfhc3CmrhJMl0pJ1xyxlmC9w
2012-10-27T23:39:23 <+dekar_> similar, but mine is a different revision I think
2012-10-27T23:39:33 <+dekar_> only rated 75W
2012-10-27T23:40:01 <+dekar_> also not really worth the hassle I think, thanks nonetheless :)
2012-10-27T23:40:17 < emeb> heh - lmgtfy, etc.
2012-10-27T23:41:04 <+dekar_> is there a problem with just replacing the caps with low ESR versions?
2012-10-27T23:41:45 < emeb> i suppose it's possible that low ESR could lead to higher inrush currents.
2012-10-27T23:41:58 < emeb> seems unlikely that would be a problem though.
2012-10-27T23:42:28 < GargantuaSauce> if it's on the output side (which is what i assume failed) that shouldn't be a problem
2012-10-27T23:42:52 < GargantuaSauce> since there's a conveniently placed inductor :P
2012-10-27T23:43:13 <+dekar_> yeah it's the output side
2012-10-27T23:44:03 <+dekar_> also I wouldn't care too much if some other part of it failed, I just learned there are 20€ replacements on ebay
2012-10-27T23:59:00 -!- |sark [~|sark@109.235.51.202] has joined ##stm32
--- Day changed Sun Oct 28 2012
2012-10-28T00:10:28 -!- barthess [~barthess@178.154.8.131] has quit [Quit: Leaving.]
2012-10-28T00:21:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T00:33:20 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-28T00:39:51 < Erlkoenig> this timer is driving me mad. i start a 2nd thread, it  does  TIM10->DIER |= 1;  NVIC_EnableIRQ (TIM1_UP_TIM10_IRQn);  to enable the timer's interrupt, and then "svc", to trigger a context switch to the main() thread. after the context switch, the timer is disabled again (indicated via LED's in the main). if i put these two lines into the SVC interrupt, the interrupt stays enabled. looks as if SVC clears the flags O_o
2012-10-28T00:41:30 < Erlkoenig> between these two lines and the svc are a __set_BASEPRI (0);, and a "cpsie i"
2012-10-28T00:53:35 < upgrdman> spi question: with the LS302DL accelerometer in SPI mode, i find that i can't leave CS low (it's active low) all the time, and I MUST make CS high in between transactions. and I MUST NOT make CS between bytes in the same transaction. is that normal? i thought some people permanently tied CS low at the hardware level if there was only one SPI device on the bus.
2012-10-28T00:54:00 < upgrdman> *MUST NOT make CS high between bytes...
2012-10-28T00:55:24 < Laurenceb_> <SpeedEvil> touch it to something, get the Wikipedia page
2012-10-28T00:55:24 < Laurenceb_> <SpeedEvil> 'that is a girl'
2012-10-28T00:55:48 < Laurenceb_> <SpeedEvil> I also want a rapid DNA sequencer
2012-10-28T00:55:59 < Laurenceb_> this seems like an unlikely thing to happen
2012-10-28T00:56:16 < Laurenceb_> hed have to encounter a girl for a start
2012-10-28T00:56:58 < Robint91> Erlkoenig, What RTOS are you using? or did you wrote your own?
2012-10-28T00:57:47 < Erlkoenig> Robint91: i'm trying to write my own
2012-10-28T00:58:12 < Erlkoenig> the context switching seems to work... but the integration with the timer refuses to work
2012-10-28T01:02:29 < Laurenceb_> http://www.b3tards.com/u/dceb14b0c0ee9d13b69a/maddy-monitor.jpg
2012-10-28T01:02:58 < Robint91> Erlkoenig, Is it also disabled in the Timer10 register
2012-10-28T01:03:18 < Erlkoenig> what's disabled?
2012-10-28T01:03:28 < Robint91> Erlkoenig,  TIM10->DIER |= 1;
2012-10-28T01:03:39 < Erlkoenig> that should enable the update interrupt.
2012-10-28T01:04:32 < Robint91> Erlkoenig, It stays enabled
2012-10-28T01:04:39 < Erlkoenig> no it doesn't
2012-10-28T01:04:45 < Erlkoenig> after the "svc" the bit is reset
2012-10-28T01:04:49 < Erlkoenig> i have no idea why
2012-10-28T01:07:59 < Robint91> Erlkoenig, strange
2012-10-28T01:09:09 -!- barthess [~barthess@164.177.203.127] has joined ##stm32
2012-10-28T01:14:20 <+Steffann> Only with certain devices that allow them to do that i guess upgrdman :P
2012-10-28T01:14:36 < upgrdman> ok
2012-10-28T01:14:48 < upgrdman> so you have see some that require CS toggling?
2012-10-28T01:15:12 -!- Robint91 [~Robin@dD5776E9E.access.telenet.be] has quit [Read error: Connection reset by peer]
2012-10-28T01:15:20 <+Steffann> I've never used/seen one that doesn't "require" it
2012-10-28T01:16:36 < upgrdman> o
2012-10-28T01:16:38 < upgrdman> ok
2012-10-28T01:17:04 < upgrdman> i've used an spi lcd that didn't require toggling, but it may have been a write-only spi device
2012-10-28T01:20:41 <+Steffann> Just read the datasheet and when you 200% sure wire it up .. it's only one extra wire :)
2012-10-28T01:20:52 <+Steffann> *when you want to be 200% sure
2012-10-28T01:28:48 < upgrdman> ya, im not short on gpios so no problem, but im surprised.
2012-10-28T01:30:33 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-28T01:31:35 < GargantuaSauce> hurrrrr
2012-10-28T01:31:50 < GargantuaSauce> just spent half an hour trying to figure out why a thing wasn't working
2012-10-28T01:32:01 < emeb> wait for it...
2012-10-28T01:32:05 < GargantuaSauce> just introduced the heap to my code/environment
2012-10-28T01:32:12 < GargantuaSauce> so i was barking up that tree a lot
2012-10-28T01:32:20 < GargantuaSauce> turns out it was a sin instead of sinf >:|
2012-10-28T01:33:36 < emeb> so the math lib you're using doesn't have sinf?
2012-10-28T01:34:02 < GargantuaSauce> i'm using the gcc stdlib, and sin is for doubles
2012-10-28T01:34:08 < emeb> yup
2012-10-28T01:34:52 < GargantuaSauce> it has sinf i was just calling the wrong one :P
2012-10-28T01:35:01 < GargantuaSauce> in one place
2012-10-28T01:35:02 < emeb> ah - you needed doubles, not floats.
2012-10-28T01:35:40 < GargantuaSauce> no i just forgot
2012-10-28T01:35:57  * emeb wonders how my CPU cycles are burned computing doubles where singles would do.
2012-10-28T01:36:28 < GargantuaSauce> for desktop stuff i just use doubles because why not
2012-10-28T01:36:46 < emeb> why not? indeed.
2012-10-28T01:37:19 < emeb> it's only a few more milliseconds out of your life. :)
2012-10-28T01:38:48 < emeb> web page for my F373 breakout: http://ebrombaugh.studionebula.com/embedded/stm32f373breakout/index.html
2012-10-28T01:39:33 < GargantuaSauce> not served by the board itself? disappointing
2012-10-28T01:40:03 < emeb> That'll have to wait until I actually have boards back from the fab.
2012-10-28T01:40:08 < emeb> and even then...
2012-10-28T01:51:00 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Remote host closed the connection]
2012-10-28T01:51:11 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-28T01:55:20 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-28T02:03:39 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T02:28:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-28T02:33:39 -!- alexn [~alexn@dslb-094-216-058-053.pools.arcor-ip.net] has quit [Remote host closed the connection]
2012-10-28T02:38:58 -!- lawrenceseattle [~anonymous@216.243.49.130] has joined ##stm32
2012-10-28T02:39:03 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-28T02:42:00 -!- barthess [~barthess@164.177.203.127] has quit [Quit: Leaving.]
2012-10-28T02:54:09 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T02:57:24 < Erlkoenig> i have spread lots of DSB and ISB aroudn the code, especially between the interrupt enabling and the svc. still svc seems to turn off the interrupt. -.-
2012-10-28T02:58:11 < zyp> you're doing something wrong
2012-10-28T02:59:05 < Erlkoenig> but what :(
2012-10-28T02:59:14 < zyp> pastebin some code?
2012-10-28T02:59:53 < Erlkoenig> yeah wait... i will clean it a bit
2012-10-28T03:01:58 < emeb> that'll probably fix it, or break it further. :P
2012-10-28T03:05:41 < Erlkoenig> i feared that ;) ... still the same behaviour. Code: http://pastebin.com/hKYmCPd7  Look at the lines with ****
2012-10-28T03:07:54 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Read error: Connection reset by peer]
2012-10-28T03:09:49 < Erlkoenig> it's experimental and thus a bit messy, perhaps... SynipCSwitchISR is the ISR supposed to be called by SVC Interrupt and all other interrupts that should cause a context switch, currently TIM1_UP_TIM10_IRQHandler. it does the context switch. SynipWaitIRQ is called by a thread to signal that it wants to sleep until that interrupt arrives.
2012-10-28T03:09:49 < Erlkoenig> It enables the interrupt and does svc to return to the main thread ( main () ) but the interrupt flags seem to get lost when the SVC Interrupt is executed... If i forcefully re-enable them in the SVC Handler , line 107 it works, but that's not how it's supposed to be done
2012-10-28T03:09:59 < Erlkoenig> @ zyp
2012-10-28T03:10:31 < zyp> hmm
2012-10-28T03:13:19 < Erlkoenig> the main() contains an endless loop wich uses the LED's to indicate whether the unterrupt flags are set... if i leave out the re-enabling in the svc handler (which is the supposed way) the red LED (D14) turns off, indicating that the interrupt is disabled, although the orange one is on (D13) indicating that the interrupt is pending
2012-10-28T03:14:16 < zyp> hang on, I'm going to write up a test with my stuff to see whether I'll have the same problems
2012-10-28T03:14:51 < Erlkoenig> hmm yeah
2012-10-28T03:15:12 < Erlkoenig> oh if you have the F4 disco, i could upload the binary, you could test whether it works for you
2012-10-28T03:15:49 -!- DaKu is now known as daku
2012-10-28T03:17:05 < zyp> I assume it would run just like on yours
2012-10-28T03:17:11 < zyp> but that's not relevant
2012-10-28T03:17:16 < zyp> just hang on
2012-10-28T03:17:38 < Erlkoenig> http://minecraft.niklas.2g2s.de/zeug/test-with-lines-108_109.elf      http://minecraft.niklas.2g2s.de/zeug/test-without-lines-108_109.elf          the first one works, the second one should, but doesn't
2012-10-28T03:18:19 < Erlkoenig> the aim is to have the blue LED blink...
2012-10-28T03:28:34 -!- Laurenceb_ [~Laurence@host86-178-234-158.range86-178.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-28T03:29:12 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has joined ##stm32
2012-10-28T03:30:23 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-28T03:30:44 -!- Tzacks [~zz@089-101-122023.ntlworld.ie] has quit [Client Quit]
2012-10-28T03:47:49 < Erlkoenig> zyp: did you find something?
2012-10-28T03:48:42 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:49:02 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:49:03 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:49:03 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:52:32 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-28T03:52:54 < Erlkoenig> he died of horror
2012-10-28T03:53:00 < zyp> nah
2012-10-28T03:53:15 < zyp> I rewrote my own threading code to use pendsv instead of svcall
2012-10-28T03:55:42 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:55:57 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:55:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-28T03:57:56 < zyp> ok, now I have a ready-flag on each of my thread so threads can be stopped
2012-10-28T03:59:10 < zyp> then I guess I have to configure a timer that can fire interrupts to restart it
2012-10-28T03:00:23 < Erlkoenig> yup... preferably TIM10 to comparable wiht my code ^^
2012-10-28T03:02:16 < zyp> ok
2012-10-28T03:02:41 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:03:00 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:03:00 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:03:00 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:07:55 -!- lawrenceseattle [~anonymous@216.243.49.130] has quit [Quit: lawrenceseattle]
2012-10-28T03:09:40 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:09:57 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:09:57 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:09:57 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:16:40 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:17:02 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:17:02 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:17:02 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:23:39 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T03:23:54 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:23:54 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:23:54 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:24:01 < dongs> asm volatile ("cpsie i");
2012-10-28T03:24:05 < dongs> jesus christ stop using gccism trash
2012-10-28T03:24:10 < dongs> there are intrinsics for this shit in CMSIS
2012-10-28T03:24:41 < dongs> opensauce 101: how ot make sure noone will want to use my shit, ever
2012-10-28T03:25:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-28T03:25:39 < zyp> hmm, wtf
2012-10-28T03:26:06 < dongs> sup?
2012-10-28T03:26:07 < dongs> blogging?
2012-10-28T03:26:16 < zyp> rblogging
2012-10-28T03:26:27 < Erlkoenig> dongs: how is the intrinsic called?
2012-10-28T03:26:38 < zyp> http://paste.jvnv.net/view/8qQ8q
2012-10-28T03:26:40 < dongs> dunno look at core_cm3 or someshit
2012-10-28T03:27:15 < emeb> 'lo dongs
2012-10-28T03:27:21 < dongs> emeb, my shit show up yet?
2012-10-28T03:27:22 < dongs> should be soon
2012-10-28T03:27:36 < zyp> looks like a race condition and shit
2012-10-28T03:27:52 < emeb> dongs: not yet. I've been watching for it tho
2012-10-28T03:28:14 < emeb> got all the parts except sip sockets for the LED arrays
2012-10-28T03:29:07 < dongs> ah, youre gonna make htem as sockets, i shoulda done that too heh. did you get the round milled type?
2012-10-28T03:29:15 < dongs> or just square shit
2012-10-28T03:29:33 < emeb> haven't picked what I'll use yet.
2012-10-28T03:29:51 < emeb> did you just solder the array in?
2012-10-28T03:30:06 < dongs> yeah lol. now that yo u mention i shoulda used a socket :)
2012-10-28T03:30:49 < emeb> Time to start searching on DK/Mouser. Need to order some other goodies too. 330ohm 0603 resistors for the breakout too...
2012-10-28T03:30:52 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:31:01 < Erlkoenig> This function enables FIQ interrupts by clearing the F-bit in the CPSR.   ---->   ARMv7 Arch Manual: Ctrl+F "CPSR" -> No Results
2012-10-28T03:31:45 < dongs> what does 330R do?
2012-10-28T03:31:54 < emeb> current limiter for the LEDs
2012-10-28T03:32:02 < dongs> eh i just use 100R
2012-10-28T03:32:04 < dongs> at 3.3V
2012-10-28T03:32:26 < dongs> isnt 330 fucking dim
2012-10-28T03:32:35 < emeb> Not with the LEDs I use.
2012-10-28T03:32:46 < emeb> that's about 4ma
2012-10-28T03:32:58 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 268 seconds]
2012-10-28T03:33:33 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T03:33:33 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T03:33:33 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T03:35:16 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Ping timeout: 246 seconds]
2012-10-28T03:36:02 < emeb> I use these -> http://www.mouser.com/ProductDetail/Lite-On/LTST-C170KGKT/?qs=%2fha2pyFadugfIWBwna%252bhOTRySO2zTctMqpVTxG8Y2sLgPWOd1RdOVg%3d%3d
2012-10-28T03:36:45 < emeb> also in Blue, Red
2012-10-28T03:37:07 < dongs> http://www.digikey.com/product-search/en?x=0&y=0&KeyWords=17-21SURC%2FS530-A3%2FTR8 i use these
2012-10-28T03:37:21 < emeb> a lot cheaper
2012-10-28T03:38:48 < dongs> the red ones are like 45/reel of 3000.
2012-10-28T03:39:12 < emeb> wonder what the 100pc price is.
2012-10-28T03:40:28 < zyp> Erlkoenig, I'm observing similar problems to what you are seeing, investigating cause
2012-10-28T03:40:41 < Erlkoenig> oha
2012-10-28T03:40:44 < Erlkoenig> okay
2012-10-28T03:40:48 < Erlkoenig> at least ima not the only one
2012-10-28T03:41:02 < Erlkoenig> i'm trying to debug where exactly the flags get screwed
2012-10-28T03:41:56 < zyp> I see that if I clear the interrupt in the timer in the ISR, everything works like expected
2012-10-28T03:42:49 < Erlkoenig> hu? that means the timer ISR gets fired?
2012-10-28T03:42:51 < zyp> if I disable the interrupt in the ISR, it's fired twice for every timer interrupt, and after a few it hits some sort of a race condition and gets stuck with both interrupt disabled and thread stopped
2012-10-28T03:43:09 < Erlkoenig> yes that looks like the first problem i ran into
2012-10-28T03:43:17 < zyp> ref. http://paste.jvnv.net/view/8qQ8q
2012-10-28T03:43:28 < Erlkoenig> my current issue arised when trying to do a workaround
2012-10-28T03:43:39 < zyp> first is systick in ms resolution, second is cycle counter
2012-10-28T03:43:55 < zyp> timer is set to interrupt at 1s intervals
2012-10-28T03:47:46 < zyp> hmm, wtf
2012-10-28T03:48:24 < Erlkoenig> i'm trying to find out when exactly the interrupt flags get reset. now it looks as if they actualle don't get reset, but: the update interrupt is enabled in the timer, it is "pending" in the timer registers, the interrupt is enabled in the timer - but the interrupt doesn't get the "pending" state in the NVIC register, and the ISR doesn't get called oO
2012-10-28T03:49:30 < emeb> dongs: array socket? -> http://www.digikey.com/product-detail/en/SFH11-PBPC-D08-ST-BK/S9196-ND/1990089
2012-10-28T03:50:13 < dongs> well thats the regular one. i was thinking http://i994.photobucket.com/albums/af68/tytzd/40Pin254mmSingleRowRoundSocket.jpg thjese sorta shits
2012-10-28T03:50:16 < dongs> i dunno if it will fit
2012-10-28T03:50:39 < Erlkoenig> how often do you say "shit"
2012-10-28T03:50:40 < emeb> oh yeah - basic IC pins.
2012-10-28T03:50:43 < dongs> trying (as i got one
2012-10-28T03:50:53 < dongs> fits.
2012-10-28T03:51:09 < emeb> sweet - I think I can get those locally at Frys
2012-10-28T03:51:40 < dongs> you could even break apart a DIP32 socket
2012-10-28T03:51:45 < dongs> heh
2012-10-28T03:51:45 < Erlkoenig> the superweird part is - i'm debugging the "not-working" version. after terminating GDB, the LED starts to blink O_O - when powercycling the board the LED doesn't blink
2012-10-28T03:51:51 < emeb> true.
2012-10-28T03:52:00 < Erlkoenig> (not i didn't change any memory or register from within GDB)
2012-10-28T03:52:38 < dongs> > gdb
2012-10-28T03:52:40 < dongs> found your problem
2012-10-28T03:52:51 < dongs> how can you guys develop/debug anything wihtout proper tools
2012-10-28T03:54:06 < GargantuaSauce> stockholm syndrome?
2012-10-28T03:54:17 < emeb> that's what I keep telling myself :)
2012-10-28T03:54:27 < emeb> love your oppressor.
2012-10-28T03:54:48 < GargantuaSauce> b-but closed source software is the real oppressor
2012-10-28T03:55:11 < emeb> takin' away yer fundamental freedoms 'n shit.
2012-10-28T03:55:24 < Erlkoenig> what the problem about GDB that's better in other debuggers?
2012-10-28T03:56:00 < Erlkoenig> what would happen if 'shit' were a badword here? sudden silence?
2012-10-28T03:56:10 < emeb> lol
2012-10-28T03:57:08 < zyp> ok, fixed it
2012-10-28T03:57:09 < zyp> http://paste.jvnv.net/view/U7Ggc
2012-10-28T03:57:32 < dongs> whats a rblog
2012-10-28T03:57:34 < Erlkoenig> fixed your code's race condition?
2012-10-28T03:57:44 < zyp> dongs, ringbuffer-log
2012-10-28T03:58:25 < zyp> some shit I wrote for logging with minimal overhead
2012-10-28T04:04:09 < Erlkoenig> what. IRQ's can't get 'pending' when doing singlestepping?
2012-10-28T04:07:33 < zyp> oh, got it
2012-10-28T04:08:01 < zyp> hang on, if this works I've figured out everything
2012-10-28T04:08:23 < Erlkoenig> as if i would quit now :D
2012-10-28T04:10:45 < zyp> right
2012-10-28T04:11:25 < zyp> check chapter 4.3.9 in PM0214
2012-10-28T04:11:55 < zyp> «For a level-sensitive interrupt, when the processor returns from the ISR, the NVIC samples the interrupt signal. If the signal is asserted, the state of the interrupt changes to pending, which might cause the processor to immediately re-enter the ISR.»
2012-10-28T04:12:14 < zyp> this happens even though it's disabled
2012-10-28T04:12:42 < Erlkoenig> wut oO
2012-10-28T04:12:45 < zyp> so it remains pending until the next time it's enabled, and then fires immediately
2012-10-28T04:12:53 < Erlkoenig> aah okay
2012-10-28T04:13:28 < Erlkoenig> that means to properly do that, i have to disable the timer interrupt in the timer control register
2012-10-28T04:13:35 < Erlkoenig> so the interrupt state doesn't get pending
2012-10-28T04:13:38 < zyp> no
2012-10-28T04:13:51 < zyp> it means you have to clear pending state after clearing the timer control register
2012-10-28T04:13:54 < zyp> before reenabling it
2012-10-28T04:15:10 < Erlkoenig> ah. if the timer has it's "interrupt occured" flag set, the interrupt will instantly become "pending" after reenabling the NVIC IRQ?
2012-10-28T04:15:14 < zyp> http://paste.jvnv.net/view/dAVzA <- it's a bit messy, but this works for me
2012-10-28T04:16:39 < Erlkoenig> but this won't fix the problem that the IRQ doesn't get into pending state after enabling it from within the thread, will it?
2012-10-28T04:16:56 < zyp> what do you mean?
2012-10-28T04:17:02 < Erlkoenig> well my current problem
2012-10-28T04:17:18 < Erlkoenig> the thread wants to sleep until the timer IRQ occurs
2012-10-28T04:17:27 < Erlkoenig> so enables the IRQ and does svc
2012-10-28T04:17:34 < zyp> yes, that's what my code does
2012-10-28T04:17:38 < Erlkoenig> the svc handler context-switchs into the main()
2012-10-28T04:17:58 < Erlkoenig> and suddenly, the NVIC forgets about the IRQ settings, and the Timer-Handler never gets called
2012-10-28T04:18:07 < zyp> Thread::yield() trigges pendsv to do a context switch
2012-10-28T04:19:48 < Erlkoenig> oh and my last gdb session has an "UsageFault", with "Invalid PC", triggered by a "svc" oO
2012-10-28T04:20:31 < zyp> you're not allowed to invoke svc in handler mode, did you do that?
2012-10-28T04:20:42 < Erlkoenig> no, that was from thread mode
2012-10-28T04:25:54 < emeb> lol - my cat loves to chew bubble pack. hopped up on the bench and started popping away...
2012-10-28T04:26:05 < zyp> hmm, right, this all makes sense now
2012-10-28T04:26:35 < zyp> you can also clear pending right before re-enabling the interrupt
2012-10-28T04:27:02 < emeb> that sounds like a really good idea.
2012-10-28T04:27:30 < zyp> if the interrupt source is still asserted, it will re-pend it immediately anyway, so there is no risk of losing valid interrupts
2012-10-28T04:27:45 < Erlkoenig> yeah thats important. okay...
2012-10-28T04:29:14 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Read error: Connection reset by peer]
2012-10-28T04:29:20 -!- BrainDamage1 [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-28T04:34:29 < Erlkoenig> oh noes, a flaw with my logic
2012-10-28T04:52:36 < dongs> Nazi Coins for Sale
2012-10-28T04:52:36 < dongs> Wholesale lots of Nazi coins. Authenticity guaranteed.
2012-10-28T04:52:36 < dongs> www.germancoins.com/
2012-10-28T04:53:18 < emeb> uh, yay?
2012-10-28T04:57:15 < Erlkoenig> nazigold \o/
2012-10-28T05:06:44 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-28T05:09:44 < emeb> Hmmm... wonder if one could use the DCMI + external FPGA & ADC to receive downconverted RF for SDR purposes?
2012-10-28T05:10:31 < emeb> someone here was talking about using DCMI as an input for non-camera data a while back... Could work.
2012-10-28T05:12:28 < dongs> emeb: um yeah, but why not just use fsmc?
2012-10-28T05:12:31 < dongs> is dcmi somehow faster?
2012-10-28T05:12:50 < zyp> dcmi is ahb master, fsmc is slave
2012-10-28T05:12:59 < emeb> dcmi can run up to 54MSPS. Don't know if FSMC can do that fast.
2012-10-28T05:13:45 < zyp> and on the other side, dcmi is slave of external data bus, fsmc is master of external data bus
2012-10-28T05:13:54 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has joined ##stm32
2012-10-28T05:14:28 < emeb> that's fine though - DCMI is input only, fine for receivers.
2012-10-28T05:14:36 < dongs> what would you do wiht 54msps ADC anyway
2012-10-28T05:14:45 < dongs> is there even enough cpu power to process itr
2012-10-28T05:14:56 < emeb> Probably not.
2012-10-28T05:15:28 < dongs> if you were gonna do this just stick a cypress fx2 or fx3 as 16bit parallel fifo on output of adc and dump the whole thing over usb2/3 and process it on pc side.
2012-10-28T05:15:29 < emeb> practical input bw is probably more like 1MSPS or less
2012-10-28T05:15:58 < emeb> which is essentially what those RTL SDR dongles do.
2012-10-28T05:16:03 < dongs> yeah.
2012-10-28T05:16:08 < emeb> (w/o cypress parts of course)
2012-10-28T05:16:33 < emeb> But I'm thinking about stand-alone devices here. USB would just muddle the issue.
2012-10-28T05:16:46 < emeb> and DCMI is a fairly simple parallel input channel.
2012-10-28T05:17:03 < dongs> standalone to do what
2012-10-28T05:17:16 < dongs> isnt the whole point of sdr is being able to visualize/dick with stuff on pc
2012-10-28T05:17:26 < emeb> Bzzzt - wrong.
2012-10-28T05:17:39 < emeb> For my day job I work on handheld SDR apps.
2012-10-28T05:18:03 < emeb> little portables capable of handling a wide range of RF waveforms.
2012-10-28T05:18:04 < dongs> "Self-effacing, portraying music with poise and intimacy"; "the PiTracer displayed remarkable proficiency at reproducing low-level and inner detail" and it "heightened the contrast between transients and the musical mix from which they emerged"; but more than anything else, the PiTracer "excels at presenting music as a complete entity, not merely an amalgam of sonic qualities."
2012-10-28T05:18:21 < emeb> audiophool.
2012-10-28T05:21:19 < dongs> thk or bwc?
2012-10-28T05:21:53 < emeb> ?
2012-10-28T05:24:14 < emeb> whoa - tsunami warning due to a 7.1 quake off the coast of B.C.
2012-10-28T05:25:44 < zyp> B.C.?
2012-10-28T05:26:17 < emeb> British Columbia - west coast of Canada.
2012-10-28T05:26:23 < zyp> ah
2012-10-28T05:27:25 < Erlkoenig> so quit your beach party...
2012-10-28T05:27:40 < dongs> noone lives in canada anyway
2012-10-28T05:28:39 < GargantuaSauce> uh huh
2012-10-28T05:30:09 < dongs> emeb: sauce plz
2012-10-28T05:31:01 -!- BrainDamage1 [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-28T05:37:19 < zyp> dongs, so, I'm apparently going to japan again, know anything cool I should plan visiting?
2012-10-28T05:37:47 < dongs> zyp, when?
2012-10-28T05:37:53 < zyp> easter
2012-10-28T05:38:00 < dongs> what the hell is that in normal dates
2012-10-28T05:38:09 < dongs> like march?
2012-10-28T05:38:19 < dongs> shit i dunno.. thats too far ahead.
2012-10-28T05:38:20 < zyp> end of march to start of april
2012-10-28T05:40:00 < dongs> i dont think theres any animu shows in march
2012-10-28T05:40:03 < dongs> :p
2012-10-28T05:40:06 < zyp> got an offer for some cheap tickets, and I've wanted to see some other season than the fucking hot summer
2012-10-28T05:41:55 < zyp> I read that Kanamara Matsuri will be held april 7. next year, I'm assuming you are going to be there :p
2012-10-28T05:45:26 < emeb> dongs: tsunami -> http://www.google.org/publicalerts/alert?aid=e03a6e895721fff1&hl=en&gl=US&source=maps
2012-10-28T05:45:43 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-28T05:56:42 < Erlkoenig> wah. what restrictions exist for writing to BASEPRI? http://pastebin.com/AecfiN5D  i try to set it to 1, but reading it out again tells me it's still 0 (error loop is entered). the debugger also says it's 0, and subsequently i get low priority (15) interrupts although the current execution priority of 1 should prevent that.
2012-10-28T05:58:17 < zyp> well
2012-10-28T05:59:14 < zyp> there is no priority 1, so that's not so strange
2012-10-28T05:59:48 < Erlkoenig> wut? as far is i understand, on the F4, i have 16 priorities, 0-15?
2012-10-28T06:00:10 < zyp> no, 0,16,32,48…
2012-10-28T06:00:12 < Erlkoenig> oh wait they're in the upper 4 bits right?
2012-10-28T06:00:14 < Erlkoenig> noooo fail
2012-10-28T06:00:16 < Erlkoenig> okay
2012-10-28T06:00:16 < zyp> exactly ;)
2012-10-28T06:00:22 < Erlkoenig> so when using 16 it should work?
2012-10-28T06:00:27 < zyp> yep
2012-10-28T06:00:33 < zyp> well, I imagine so
2012-10-28T06:01:13 < dongs> emeb: just saw on japnews
2012-10-28T06:01:16 < dongs> apparently its 7.7 now
2012-10-28T06:01:31 < dongs> japan r worried that fucking canucks are gonna set them up teh tsunami bomb
2012-10-28T06:01:31 < emeb> yoiks!
2012-10-28T06:01:59 < emeb> pacific tsunami center says no danger to Hawaii, so likely JP is safe.
2012-10-28T06:02:37 < zyp> plz don't be destroying japan now that I've bought tickets there
2012-10-28T06:02:40 < Erlkoenig> uhm the priority field for IRQ's also has only the upper 4 bits?
2012-10-28T06:02:47 < zyp> Erlkoenig, yes
2012-10-28T06:02:52 < Erlkoenig> argh.
2012-10-28T06:03:00 < zyp> PM0214, chapter 4.3.7
2012-10-28T06:03:20 < Erlkoenig> the ARMv7 ARch Manual says that too... forgot it :S
2012-10-28T06:03:28 < Erlkoenig> brain input buffer overflow
2012-10-28T06:05:13 < zyp> well, it's not so strange they're removing the lower bits rather than the higher bits
2012-10-28T06:05:37 < Erlkoenig> maybe would have been more convenient to remove the upper bits
2012-10-28T06:05:43 < zyp> no.
2012-10-28T06:05:52 < Erlkoenig> :/
2012-10-28T06:06:03 < zyp> consider priorities 127 and 128
2012-10-28T06:06:10 < zyp> those turn into 112 and 128
2012-10-28T06:06:17 < zyp> which is more sane than 15 and 0
2012-10-28T06:07:24 < zyp> the number of bits is implementation dependant
2012-10-28T06:07:46 < zyp> by removing the lower bits, software essentially doesn't have to care
2012-10-28T06:08:17 < zyp> consider software using 16 levels
2012-10-28T06:08:26 < zyp> consider it running on a cpu only implementing 8
2012-10-28T06:08:40 < zyp> now two and two levels would be merged, and software should still work fine
2012-10-28T06:10:34 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 272 seconds]
2012-10-28T06:11:07 < zyp> instead of considering it as a number from 0-255, you should consider it a binary decimal
2012-10-28T06:11:45 < Erlkoenig> yeah i got it
2012-10-28T06:14:53 < Erlkoenig> ah. NVIC_SetPriority automatically does the bitshift. __set_BASEPRI doesnt
2012-10-28T06:16:58 < bsdfox_> what lib do most people use in linux? libopencm3 or the proprietary ones?
2012-10-28T06:17:24 < Erlkoenig> i use the proprietary StdPeriph
2012-10-28T06:18:35 < zyp> I write my own
2012-10-28T06:18:45 < zyp> but I'm clearly insane
2012-10-28T06:18:53 < zyp> I would suggest libopencm3 for sane people
2012-10-28T06:29:27 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-28T06:30:44 < Erlkoenig> hmmm got it to work partially thanks to your hints. but the blinking is somewhat random, looks like some interrupts are completely missed oO
2012-10-28T06:31:55 < zyp> then I suggest logging timestamps
2012-10-28T06:32:58 < Erlkoenig> oOo inserted 1 line into ISR: "GPIOD->ODR ^= (1 << 13);" now it works
2012-10-28T06:33:15 < Erlkoenig> how to do that logging? using the USB_OTG Device Core?
2012-10-28T06:34:12 < zyp> printing shit is way too slow for tracing shit like this
2012-10-28T06:34:13 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-28T06:34:19 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-28T06:34:47 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-28T06:35:14 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-28T06:35:33 < zyp> http://cgit.jvnv.net/laks/tree/util/rblog.h <- I'm just stuffing timestamps, string pointers and optional arguments into an array, and then extract the contents of the array later
2012-10-28T06:36:13 < Erlkoenig> how do you do the examining?
2012-10-28T06:36:24 < zyp> wrote a python plugin for gdb
2012-10-28T06:36:41 < Erlkoenig> why the heck does it work when i insert that simple line oO
2012-10-28T06:36:57 < zyp> http://paste.jvnv.net/view/FqdsT
2012-10-28T06:39:32 < zyp> but of course, if you have a way to print stuff, it would be easy to write a function to dump and print the array later too
2012-10-28T06:39:54 < dongs> zyp, why not use swv/swo, o wait, ur using gdb and opensauce debug probes.
2012-10-28T06:40:15 < zyp> dongs, the too slow part
2012-10-28T06:42:10 < Erlkoenig> hmm maybe the duration of the ISR influences the behaviour
2012-10-28T06:42:27 < zyp> and also the fact that I can leave this in because it's low overhead, and then when something fucks up I can go look at the logs that was collected
2012-10-28T06:42:55 < zyp> Erlkoenig, or hardware races, try with some barriers
2012-10-28T06:43:29 < Erlkoenig> zyp: i wrote a similar thing for AVR... except that the ringbuffer contents was transferred via the UART, integrated into the regular applications data stream
2012-10-28T06:43:45 < Erlkoenig> already seeded a lot of barriers there... dont help much
2012-10-28T06:44:24 < Erlkoenig> when inserting a lot of NOP's instead of the GPIO access exactly every 2nd IRQ is missed
2012-10-28T06:44:53 < zyp> I don't have any uart hooked up, and I don't bother hooking up one when I already have access to everything I need with gdb
2012-10-28T06:45:19 < Erlkoenig> yup... my AVR needed UART for the application so it was already in place ;)
2012-10-28T06:45:46 < zyp> well, and I bet you didn't have a debugger for it :p
2012-10-28T06:46:21 < Erlkoenig> the micro didn't have debugging functionality :D
2012-10-28T06:46:32 < dongs> tarduino debugging
2012-10-28T06:46:38 < zyp> dongs, by the way, how did the swo stuff that you were talking about recently work out for you?
2012-10-28T06:46:45 < dongs> yeah, worked
2012-10-28T06:46:58 < dongs> i got keil logic analyzer shit to display some vars realtime
2012-10-28T06:47:54 < zyp> hmm, maybe I should try magictpa, now that I've upgraded my BMP to have swo support
2012-10-28T06:50:01 < Erlkoenig> hm enough for today... gnite
2012-10-28T06:50:11 -!- Erlkoenig [~erlkoenig@pptp-212-201-69-221.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-28T06:52:40 < emeb_mac> zyp: I saw that BPM was supposed to support swo now.
2012-10-28T06:53:03 < emeb_mac> buddy was using atollic w/ STLINK in windows and said swo stuff worked fine.
2012-10-28T06:55:22 < zyp> yes, magictpa is a gdb extension to use the swo stuff in bmp
2012-10-28T06:55:48 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-28T06:56:59 < emeb_mac> OK so it needs python support in gdb
2012-10-28T06:57:34 < emeb_mac> so I'd have to build the toolchain from scratch since the binary distro doesn't have python enabled.
2012-10-28T06:57:55 < emeb_mac> (insert appropriate RAGE here)
2012-10-28T07:23:43 < emeb_mac> lol australia: http://www.youtube.com/watch?feature=player_embedded&v=dr_cn66sYc8
2012-10-28T07:35:48 < baird> H.R. Geiger designed a penis and gave it to the Rhino:  http://img.chan4chan.com/img/2012-03-24/1332625447839.jpg
2012-10-28T07:36:59 < emeb_mac> ok - that's out there.
2012-10-28T07:37:04 < emeb_mac> looks like an orchid
2012-10-28T07:40:11 < baird> A John Wyndham Triffid..
2012-10-28T07:43:12 < emeb_mac> I vaguely recall triffids. Giant pod flowers IIRC.
2012-10-28T07:47:19 < baird> They had a poisonous stamen that came out and slapped you about the face. Maybe preferable to being dickslapped by a Rhino.
2012-10-28T08:02:47 -!- Netsplit *.net <-> *.split quits: kerwood, bsdfox_, ntfreak, +dekar_, esden, ossifrage, BusError, txf
2012-10-28T08:19:52 < dongs> negative ebay feedback: "Didn't have a chance to change my order, they shipped the order too quickly."
2012-10-28T08:20:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-28T08:20:23 < emeb_mac> oh noes!
2012-10-28T08:20:42 < emeb_mac> service too sudden.
2012-10-28T08:32:44 < baird> Had nigiri at a japanese resturant for lunch (with a bunch of other weebs) ..most agreeable, except for the squid.
2012-10-28T08:43:09 -!- Netsplit *.net <-> *.split quits: erik-k, zippe, jaeckel, ohama, Ademan, ABLomas, @ChanServ, Niedar, dongs, Posterdati,  (+7 more, use /NETSPLIT to show all of them)
2012-10-28T08:43:58 < emeb_mac> should try the eel sushi
2012-10-28T08:47:55 -!- prog [prog@bornbrilliant.org] has joined ##stm32
2012-10-28T08:48:35 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xgonqyoqwxqdfexu] has joined ##stm32
2012-10-28T08:48:35 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-28T08:48:35 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-28T08:48:35 -!- ohama [ohama@92.240.248.100] has joined ##stm32
2012-10-28T08:48:35 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-28T08:48:35 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has joined ##stm32
2012-10-28T08:48:35 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-28T08:48:35 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-10-28T08:48:35 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has joined ##stm32
2012-10-28T08:48:35 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-28T08:48:35 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-10-28T08:48:35 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-10-28T08:48:35 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-10-28T08:48:35 -!- ServerMode/##stm32 [+o ChanServ] by lindbohm.freenode.net
2012-10-28T08:59:36 -!- prog is now known as Guest38205
2012-10-28T09:16:58 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-28T09:42:57 -!- Netsplit *.net <-> *.split quits: erik-k, zippe, ABLomas, jaeckel, ohama, dongs, @ChanServ, Niedar, Posterdati, upgrdman,  (+3 more, use /NETSPLIT to show all of them)
2012-10-28T10:29:54 -!- erik-k [~erik-k@c-50-137-59-216.hsd1.or.comcast.net] has joined ##stm32
2012-10-28T10:29:54 -!- Posterdati [~antani@host45-237-dynamic.6-87-r.retail.telecomitalia.it] has joined ##stm32
2012-10-28T10:29:54 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xgonqyoqwxqdfexu] has joined ##stm32
2012-10-28T10:29:54 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-28T10:29:54 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-28T10:29:54 -!- ohama [ohama@92.240.248.100] has joined ##stm32
2012-10-28T10:29:54 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-28T10:29:54 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-10-28T10:29:54 -!- ChanServ [ChanServ@services.] has joined ##stm32
2012-10-28T10:29:54 -!- dongs [1000@l212168.ppp.asahi-net.or.jp] has joined ##stm32
2012-10-28T10:29:54 -!- ServerMode/##stm32 [+o ChanServ] by lindbohm.freenode.net
2012-10-28T10:31:23 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-28T10:31:23 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-10-28T10:46:43 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-xgonqyoqwxqdfexu] has quit [Ping timeout: 260 seconds]
2012-10-28T10:47:21 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-hlareesrhuqrfrlv] has joined ##stm32
2012-10-28T10:59:53 -!- ds2 [noinf@netblock-66-245-251-24.dslextreme.com] has joined ##stm32
2012-10-28T10:59:53 -!- Nutter [Nutter@64-46-18-151.dyn.novuscom.net] has joined ##stm32
2012-10-28T10:59:53 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-28T10:59:53 -!- Ademan [~dan@70.231.140.183] has joined ##stm32
2012-10-28T10:59:53 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T10:59:53 -!- txf [~txf@bl13-160-172.dsl.telepac.pt] has joined ##stm32
2012-10-28T10:59:53 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-28T10:59:53 -!- BusError [~michel@host86-171-235-55.range86-171.btcentralplus.com] has joined ##stm32
2012-10-28T10:59:53 -!- dekar_ [~dekar@212.255.27.61] has joined ##stm32
2012-10-28T10:59:53 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has joined ##stm32
2012-10-28T10:59:53 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-28T10:59:53 -!- esden [esden@repl.esden.net] has joined ##stm32
2012-10-28T10:59:53 -!- ServerMode/##stm32 [+v dekar_] by lindbohm.freenode.net
2012-10-28T11:03:58 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-28T11:05:27 -!- daku is now known as DaKu
2012-10-28T11:08:32 -!- Netsplit *.net <-> *.split quits: ds2, ossifrage, BusError, txf, dfletcher__, Ademan, +dekar_, Nutter, kerwood, esden,  (+2 more, use /NETSPLIT to show all of them)
2012-10-28T11:14:00 -!- barthess [~barthess@178.154.27.159] has joined ##stm32
2012-10-28T11:14:01 -!- Netsplit over, joins: ntfreak, esden, ds2, Nutter, dfletcher__, Ademan, bsdfox_, txf, BusError, +dekar_ (+2 more)
2012-10-28T11:21:41 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-28T11:34:26 -!- Netsplit *.net <-> *.split quits: Ademan, ntfreak, txf, Nutter, zippe, ohama, Niedar, jaeckel, bsdfox_, BjoernC,  (+15 more, use /NETSPLIT to show all of them)
2012-10-28T11:38:32 -!- Netsplit over, joins: @ChanServ, ohama, BjoernC, esden, ossifrage, kerwood, +dekar_, BusError, ntfreak, txf (+15 more)
2012-10-28T11:39:26 < qyx_> o\
2012-10-28T11:39:27 -!- ohama [ohama@92.240.248.100] has quit [Ping timeout: 260 seconds]
2012-10-28T11:40:36 -!- ohama [ohama@92.240.248.100] has joined ##stm32
2012-10-28T11:44:34 -!- Netsplit *.net <-> *.split quits: Posterdati, erik-k
2012-10-28T11:45:20 -!- Netsplit *.net <-> *.split quits: jaeckel, Laurenceb
2012-10-28T11:54:48 -!- Netsplit over, joins: Posterdati, erik-k
2012-10-28T11:56:06 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-28T11:56:06 -!- jaeckel [~jaeckel@unaffiliated/jaeckel] has joined ##stm32
2012-10-28T12:00:13 -!- ohama [ohama@92.240.248.100] has quit [Disconnected by services]
2012-10-28T12:00:30 -!- ohama [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-10-28T12:00:47 -!- ohama [metaxa@us.members.bombshellz.net] has quit [Client Quit]
2012-10-28T12:04:57 -!- ohama [ohama@92.240.248.100] has joined ##stm32
2012-10-28T12:08:06 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-10-28T12:16:56 -!- DaKu [DaKu@dakus.dk] has quit [Remote host closed the connection]
2012-10-28T12:19:49 -!- Guest38205 is now known as prog
2012-10-28T12:21:02 -!- piele is now known as piele_gone
2012-10-28T12:26:40 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-28T12:31:37 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-28T12:33:02 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 255 seconds]
2012-10-28T12:50:52 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-28T13:08:53 -!- txf [~txf@bl13-160-172.dsl.telepac.pt] has quit [Ping timeout: 245 seconds]
2012-10-28T13:44:18 -!- piele_gone is now known as piele
2012-10-28T13:50:28 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-28T14:05:20 -!- daku [DaKu@dakus.dk] has joined ##stm32
2012-10-28T14:16:16 -!- daku is now known as DaKu
2012-10-28T14:18:56 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-28T14:18:59 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-28T14:36:15 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-28T14:53:36 < zyp> it's awfully silent here today
2012-10-28T14:55:47 < dongs> feenode cant figure out how to keep their shit linked
2012-10-28T15:20:57 -!- barthess [~barthess@178.154.27.159] has quit [Quit: Leaving.]
2012-10-28T15:45:55 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-28T16:08:35 < Laurenceb_> sup
2012-10-28T16:09:24 < zyp> tha sky bro
2012-10-28T16:09:42 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-219.pptp.stw-bonn.de] has joined ##stm32
2012-10-28T16:10:44 < baird> sup   sup   sup  sup  sup  sup  sup sup sup supsupsupsupsupspuspuspuspusps
2012-10-28T16:11:42 < qyx_> here "sup" means vulture (bird)
2012-10-28T16:12:29 < Laurenceb_> http://www.disinfo.com/2010/10/are-redheads-descended-from-neanderthals/
2012-10-28T16:12:38 < Laurenceb_> suddenly it all makes sense
2012-10-28T16:12:47 < Laurenceb_> i always thought they looked a bit mutant
2012-10-28T16:34:16 -!- barthess [~barthess@178.154.27.159] has joined ##stm32
2012-10-28T16:43:22 <+Steffann> https://www.youtube.com/watch?v=m9unXDArPDY&feature=player_embedded The most luckiest bastard of the year
2012-10-28T16:43:50 <+Steffann> Old video though, so probably you've seen it already :)
2012-10-28T16:44:03 < Erlkoenig> wtf
2012-10-28T16:44:05 <+Steffann> Oh, not old :)
2012-10-28T16:44:23 <+Steffann> Thought it said 2010 in the video, but it says 2012
2012-10-28T16:45:07 < Erlkoenig> why is the car "sliding" to the left? loosing grip?
2012-10-28T16:45:30 < Erlkoenig> wet road perhaps?
2012-10-28T16:46:00 <+Steffann> Loosing grip, driving too fast, crap car? :)
2012-10-28T16:46:19 < zyp> bad driving
2012-10-28T16:46:29 <+Steffann> And that
2012-10-28T16:47:40 < zyp> grip seems to be good enough considering how the car gets up on two wheels when he realizes and tries to get back
2012-10-28T16:47:59 < zyp> so my guess is that the driver was just unattentive
2012-10-28T16:48:12 <+Steffann> Not only that
2012-10-28T16:48:22 <+Steffann> What's that can at the end of the video? Beer? :)
2012-10-28T16:48:29 < zyp> well, in addition to driving like a dick :p
2012-10-28T16:49:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-28T16:53:03 < Erlkoenig> i think the STM32 have a Non-Deterministic-Behaviour-Unit (NDBH) to facilate automatically evolving design
2012-10-28T16:53:22 < zyp> ok
2012-10-28T16:53:35 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-28T16:56:01 < baird> I think I'd exempt the car from the 'crap' classification for mostly withstanding a both-side impact like that. :)
2012-10-28T16:57:12 < Erlkoenig> aah i got it... it's for hardware accelerated execution of non-deterministic finite state machines
2012-10-28T16:57:49 -!- DaKu is now known as daku
2012-10-28T17:02:39 -!- daku is now known as DaKu
2012-10-28T17:09:16 < baird> lol .. https://www.youtube.com/watch?v=QCAntD1-DIk "Guy Walks Into Giant Bear While Texting"
2012-10-28T17:15:37 < bsdfox_> wtf does CONFMASK(7) do? I'm trying to port some code to libopencm3
2012-10-28T17:17:06 < bsdfox_> I would beat the shit out of that kid Steffann
2012-10-28T17:43:05 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-28T17:49:15 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-28T17:50:03 -!- barthess [~barthess@178.154.27.159] has quit [Quit: Leaving.]
2012-10-28T18:06:57 -!- metaxa [metaxa@us.members.bombshellz.net] has quit [Quit: leaving]
2012-10-28T18:07:44 -!- metaxa [metaxa@us.members.bombshellz.net] has joined ##stm32
2012-10-28T18:18:24 -!- barthess [~barthess@5.100.208.231] has joined ##stm32
2012-10-28T18:24:47 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-28T18:24:57 < TitanMKD> hi
2012-10-28T18:28:53 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-28T18:49:20 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-28T18:51:03 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-28T18:54:45 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-28T18:54:50 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-10-28T19:07:30 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-28T19:10:27 < karlp> awesome, the part I was using is now minimum order 5950 units.
2012-10-28T19:11:01 < zyp> time to make a lot then
2012-10-28T19:13:28 < karlp> time to find something with the same footprint and pinout I think actually.
2012-10-28T19:19:29 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-10-28T19:19:34 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-28T19:19:56 -!- Thorn [~Thorn@128-73-41-139.broadband.corbina.ru] has joined ##stm32
2012-10-28T19:19:57 -!- Thorn [~Thorn@128-73-41-139.broadband.corbina.ru] has quit [Changing host]
2012-10-28T19:19:57 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-28T19:20:08 -!- txf [~txf@bl14-151-32.dsl.telepac.pt] has joined ##stm32
2012-10-28T19:22:04 -!- txf [~txf@bl14-151-32.dsl.telepac.pt] has quit [Client Quit]
2012-10-28T19:22:35 -!- txf [~txf@bl14-151-32.dsl.telepac.pt] has joined ##stm32
2012-10-28T19:24:04 -!- tavish [~tavish@unaffiliated/tavish] has joined ##stm32
2012-10-28T19:31:49 < Laurenceb_> https://groups.google.com/forum/?fromgroups=#!topic/ukhas/uXUxd4rWjEk
2012-10-28T19:31:51 < Laurenceb_> pmsl
2012-10-28T19:32:22 < bsdfox_> anyone got a bin file I can flash on stm32vl-discovery? I can't figure out if I'm having a hardware or a toolchain issue.. can't even get simple led blinker example code to work (though it compiles and flashed fine). just constant activity on LD1
2012-10-28T19:34:02 < quitte> bsdfox_: did you set the peripheral bus clock?
2012-10-28T19:35:42 < quitte> bsdfox_: http://www.st.com/internet/evalboard/product/250863.jsp an3268 and the corresponding zip file seem to be what you are asking for
2012-10-28T19:36:36 -!- tavish [~tavish@unaffiliated/tavish] has quit [Ping timeout: 240 seconds]
2012-10-28T19:38:34 < bsdfox_> quitte, I don't think any of those are pre compiled
2012-10-28T19:40:19 < quitte> okay. my guess is still that the gpio module has no clock
2012-10-28T19:44:36 < karlp> you want a blink bin for 32vl tha tworks?
2012-10-28T19:46:08 < karlp> http://palmtree.beeroclock.net/~karlp/fancyblink.bin
2012-10-28T19:46:09 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T19:46:15 < karlp> that's from the libopencm3 examples.
2012-10-28T19:47:00 < jpa-> how fancy is it?
2012-10-28T19:53:27 < Erlkoenig> be careful, it might contain virus
2012-10-28T19:54:13 < Erlkoenig> better run antivir on it
2012-10-28T19:56:10 < karlp> not very fancy at all.
2012-10-28T20:00:14 < quitte> explain the fanciness already. the suspense is killing me.
2012-10-28T20:04:09 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-28T20:13:12 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T20:17:13 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 245 seconds]
2012-10-28T20:17:54 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-28T20:19:55 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Write error: Broken pipe]
2012-10-28T20:20:12 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T20:20:12 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T20:20:12 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T20:25:13 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds]
2012-10-28T20:26:55 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Write error: Connection reset by peer]
2012-10-28T20:27:12 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T20:27:13 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T20:27:13 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T20:31:51 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-28T20:32:17 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Client Quit]
2012-10-28T20:32:47 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 256 seconds]
2012-10-28T20:32:50 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-28T20:33:54 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-28T20:34:11 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T20:35:11 < bsdfox_> thanks karlp
2012-10-28T20:35:15 < bsdfox_> that works
2012-10-28T20:35:56 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Client Quit]
2012-10-28T20:36:20 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-28T20:41:06 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T20:45:27 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 260 seconds]
2012-10-28T20:48:53 -!- bsdfox\ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-28T21:01:01 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-28T21:01:01 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-28T21:01:01 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-28T21:09:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T22:00:51 < upgrdman> i dont have a hot air station and i would like to remove an LQFP 100 device from a circuit board, and be able to reuse the device afterward. would putting the board in an oven work? (get it hot, then try to pick up the part with tweezers?)
2012-10-28T22:02:52 < quitte> if you don't want to reuse the board a hotplate works pretty good
2012-10-28T22:11:09 -!- emeb [~ericb@75-171-83-235.phnx.qwest.net] has joined ##stm32
2012-10-28T22:11:35 < emeb> sup y'all
2012-10-28T22:17:02 -!- txf [~txf@bl14-151-32.dsl.telepac.pt] has quit [Ping timeout: 272 seconds]
2012-10-28T22:24:57 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-28T22:26:44 -!- txf [~txf@bl12-90-46.dsl.telepac.pt] has joined ##stm32
2012-10-28T22:28:14 < BrainDamage> broth to you too
2012-10-28T22:37:03 -!- Vutral [~ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T22:40:31 < TitanMKD> upgrdman i will advise you to buy ChipQuik SMD Removal Kit
2012-10-28T22:41:22 < TitanMKD> upgrdman it is just magic you can desolder easily even bigger than LQFP100 with just a soldering iron and without heating too much the component
2012-10-28T22:42:03 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-28T22:43:03 -!- Vutral [~ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-28T22:43:54 -!- barthess [~barthess@5.100.208.231] has quit [Quit: Leaving.]
2012-10-28T22:46:38 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-28T23:05:50 <+dekar_> that ChipQuik kit looks pretty awesome :)
2012-10-28T23:08:32 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-28T23:09:52 < BrainDamage> I have some, it works rather well, but for large chips requires multiple reheating passes
2012-10-28T23:10:13 -!- emeb [~ericb@75-171-83-235.phnx.qwest.net] has quit [Ping timeout: 252 seconds]
2012-10-28T23:10:33 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-28T23:11:16 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 245 seconds]
2012-10-28T23:13:39 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-28T23:13:55 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-28T23:15:14 <+dekar_> I would assume a second soldering iron helps
2012-10-28T23:15:53 < bsdfox_> upgrdman, I've used a blow torch on the back side of pcbs.. burns the shit out of it and smells bad but all the stuff on the opposite side comes right off in good shape
2012-10-28T23:17:39 < BrainDamage> dekar_: yeah, but not fundamental, it remains molten for tens of seconds
2012-10-28T23:19:33 < upgrdman> bsdfox_, that sounds good. i dont care about the other stuff!
2012-10-28T23:20:01 <+dekar_> I managed to burn a pcb using a proper hot air station :/
2012-10-28T23:20:11 < bsdfox_> upgrdman, wear safety glasses and get a pie tin or something to tap the board on every few seconds
2012-10-28T23:20:17 < upgrdman> do you torch it hot (up close) or just try to heat it up from a little further back?
2012-10-28T23:20:19 < bsdfox_> also, get a hot air station
2012-10-28T23:20:35 < upgrdman> ya
2012-10-28T23:20:38 < bsdfox_> upgrdman, in the middle.. the pcb will bubble and catch fire
2012-10-28T23:20:48 < upgrdman> ooooo :)O
2012-10-28T23:20:51 < bsdfox_> find a nice middle ground :)
2012-10-28T23:21:20 < BrainDamage> if you don't care about the rest, you can just trim most of the pcb with a saw if you're careful about not flexing it
2012-10-28T23:21:27 -!- gsmcmullin [~gareth@122-62-149-197.jetstream.xtra.co.nz] has joined ##stm32
2012-10-28T23:21:34 < BrainDamage> it'll give you a more comfortable working area
2012-10-28T23:21:43 < upgrdman> k
2012-10-28T23:21:57 < Erlkoenig> http://pastebin.com/S1QYydX7   without Line 38 (__set_BASPRI)  the program works fine (LED blinks with 1Hz), with that line the blinking is irregular (as if some interrupts are missed). But WHY?!?!
2012-10-28T23:22:00 <+dekar_> I don't really like sawing fr4
2012-10-28T23:22:42 < BrainDamage> I use a nibbling cutter
2012-10-28T23:22:50 < upgrdman> ?
2012-10-28T23:29:18 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 268 seconds]
2012-10-28T23:41:30 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
--- Day changed Mon Oct 29 2012
2012-10-29T00:12:47 < bsdfox_> anyone used summon-arm-toolchain and had problems compiling working binaries? I compile fine and upload fine but when I compile example code it doesn't seem to do anything but binaries of the same code from other sources seem to work
2012-10-29T00:29:33 -!- emeb [~ericb@75-171-83-235.phnx.qwest.net] has joined ##stm32
2012-10-29T00:51:43 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T00:55:25 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has quit [Ping timeout: 265 seconds]
2012-10-29T00:56:52 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has joined ##stm32
2012-10-29T00:59:19 -!- DaKu is now known as daku
2012-10-29T01:00:45 < dongs> http://openpilot.org/countdown_data/images/2b.jpg what a shitty fab
2012-10-29T01:00:56 < dongs> looks like they're using seedstudio or something look at those awsm drills
2012-10-29T01:02:34 < zyp> no, seeed doesn't do silk with that high resolution
2012-10-29T01:02:54 < dongs> did you see titans board? their 4 layer shit has better looking silk but still horrible drills
2012-10-29T01:02:58 < dongs> maybe its itead then
2012-10-29T01:04:10 < Laurenceb_> whats that?
2012-10-29T01:04:40 < dongs> Laurenceb_: i think some (?) openpilot revolution shits have a rf modem on them
2012-10-29T01:04:46 < Laurenceb_> oh
2012-10-29T01:04:47 < Laurenceb_> lol
2012-10-29T01:04:51 < Laurenceb_> thats an rfm22b then
2012-10-29T01:04:57 < dongs> right
2012-10-29T01:05:01 < Laurenceb_> so the board is made is shenzhen somewhere
2012-10-29T01:05:06 < Laurenceb_> but "HopeRF"
2012-10-29T01:05:11 < Laurenceb_> *by
2012-10-29T01:05:30 < Laurenceb_> its a relazered si4432 with crappy rf
2012-10-29T01:05:33 < Laurenceb_> do not want
2012-10-29T01:05:49 < Laurenceb_>  #highaltitude use them a lot, but they are really piss poor
2012-10-29T01:05:54 < dongs> orly
2012-10-29T01:05:57 < Laurenceb_> poor tolerance xtals
2012-10-29T01:06:10 < Laurenceb_> then it uses varicaps on the ic for the xtal osc
2012-10-29T01:06:25 < Laurenceb_> that drift massively with what the si4432 is doing
2012-10-29T01:07:02 < Laurenceb_> they have been experimenting with clk input
2012-10-29T01:07:08 < Laurenceb_> seems to way to go
2012-10-29T01:07:18 < dongs> isntead of using xtal?
2012-10-29T01:07:23 < Laurenceb_> also F4 has a "spare" pll that could be used to give a 30mhz signal
2012-10-29T01:07:25 < Laurenceb_> yeah
2012-10-29T01:07:37 < Laurenceb_> but on the +ive side at least pipxtreme should be made usable
2012-10-29T01:07:53 < dongs> wouldnt that be just same as drifty30mhz xtal?
2012-10-29T01:07:59 < Laurenceb_> nope
2012-10-29T01:08:00 < dongs> vs ??mhz+pll+mco to 30mhz?
2012-10-29T01:08:02 < dongs> why not
2012-10-29T01:08:04 < Laurenceb_> yes
2012-10-29T01:08:06 < upgrdman> does the brand or other markings on this RF pcb look familiar? nothing comes up on google... http://www.farrellf.com/temp/unknown_rf_pcb.jpg
2012-10-29T01:08:10 < Laurenceb_> as it wont drift
2012-10-29T01:08:41 < Laurenceb_> itll be clocked off stm xtal
2012-10-29T01:08:55 < zyp> upgrdman, what's it from?
2012-10-29T01:08:58 < Laurenceb_> if they get si4432 code into mainline openpilot will actually be decent
2012-10-29T01:09:14 < upgrdman> zyp, toy heli radio
2012-10-29T01:09:15 < dongs> Laurenceb_: so stm crystal drifts less?
2012-10-29T01:09:23 < Laurenceb_> of course
2012-10-29T01:09:23 < dongs> or do you mean less vs the shit thats on hoperf board
2012-10-29T01:09:29 < Laurenceb_> its a proper xtal oscillator
2012-10-29T01:09:33 < Laurenceb_> with a proper xtal
2012-10-29T01:09:50 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has quit [Remote host closed the connection]
2012-10-29T01:10:12 < Laurenceb_> then with custom rf matching on si4432 you should improve performance
2012-10-29T01:10:20 < Laurenceb_> use decent murata stuff
2012-10-29T01:10:30 < Laurenceb_> like on dactyl
2012-10-29T01:11:39 < dongs> well will see
2012-10-29T01:12:11 < zyp> Laurenceb_, what's your thoughts on the coming cortex-m radio socs?
2012-10-29T01:12:23 < zyp> i.e. efr4d and nrf51822
2012-10-29T01:12:27 < dongs> OP has bene hushhush about whole radio moddem shit
2012-10-29T01:12:32 < dongs> apparently theres something called op-link
2012-10-29T01:12:34 < dongs> thats receiver end
2012-10-29T01:12:35 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has joined ##stm32
2012-10-29T01:12:50 < dongs> and some revo shit has builtin modem, besides that i dunno
2012-10-29T01:13:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-29T01:13:58 < Laurenceb_> zyp: havent read datasheets
2012-10-29T01:14:43 < Laurenceb_> oh actually yeah
2012-10-29T01:14:49 < Laurenceb_> efr4d looks epic
2012-10-29T01:14:54 < Laurenceb_> but not avaliable
2012-10-29T01:15:01 < zyp> I have «custom radio stack» on my list of «shit I want to try writing»
2012-10-29T01:15:09 < Laurenceb_> but M3
2012-10-29T01:15:21 < zyp> and nrf51 is M0
2012-10-29T01:15:24 < zyp> so what?
2012-10-29T01:15:58 < zyp> what I'm wondering is whether they will really have any huge benefit over using a spi transceiver with a standalone mcu
2012-10-29T01:16:19 < Laurenceb_> i was hoping for rf built into main cpu
2012-10-29T01:16:26 < Laurenceb_> why not just use silabs stuff
2012-10-29T01:16:36 < Laurenceb_> it has everything youd ever want modem wise
2012-10-29T01:18:54 < upgrdman> what is "RU" ... the logo has the R backwards. some sort of certification org?
2012-10-29T01:18:57 < zyp> the ones with built in cpus are 8051 based, I'd rather not touch that :p
2012-10-29T01:19:29 < Laurenceb_> russian  :P
2012-10-29T01:19:55 < zyp> but I dunno, just using spi would probably be fine
2012-10-29T01:19:58 < upgrdman> but i've seen the logo near FCC and other logos
2012-10-29T01:20:25 < zyp> http://en.wikipedia.org/wiki/Recognized_Component_Mark
2012-10-29T01:23:17 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: This computer has gone to sleep]
2012-10-29T01:24:54 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 240 seconds]
2012-10-29T01:27:54 < upgrdman> is the address of the "WHO AM I" register the same for most SPI devices?
2012-10-29T01:28:20 < zyp> ha, optimist
2012-10-29T01:29:01 < upgrdman> :)
2012-10-29T01:30:46 < upgrdman> AWESOME! i think i found my rf pcb on google image search.
2012-10-29T01:31:51 < upgrdman> http://www.aliexpress.com/store/product/CC2500-Wireless-Module-Small-size-2-4GHz/107662_342715593.html
2012-10-29T01:33:21 < zyp> ah, the cheap rc stuff I bought once were also using CC2500
2012-10-29T01:34:04 < upgrdman> ya. the toy heli radio i pulled it from has a range of maybe 200 feet
2012-10-29T01:34:25 < upgrdman> but thats good enough for my projects... and i pull tens of theses out of trash radios
2012-10-29T01:40:15 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-29T01:52:38 -!- dekar_ [~dekar@212.255.27.61] has quit [Ping timeout: 245 seconds]
2012-10-29T02:02:13 -!- dekar [~dekar@212.255.224.68] has joined ##stm32
2012-10-29T02:02:17 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-29T02:06:48 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-29T02:07:29 < dongs> trollforge
2012-10-29T02:38:31 -!- emeb [~ericb@75-171-83-235.phnx.qwest.net] has left ##stm32 []
2012-10-29T02:48:53 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-29T02:52:41 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-29T02:55:55 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-29T02:56:01 -!- BrainDamage1 [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-29T03:10:40 < HTT-Bird> so: SnPb 63/37 or SAC305 for toaster-oven-reflow?
2012-10-29T03:11:11 < HTT-Bird> upgrdman: that "RU" is the logo for a UL-recognized component intended to go into a larger assembly
2012-10-29T03:11:38 < HTT-Bird> as opposed to something that is listed by UL for stand-alone application of some sort
2012-10-29T03:12:12 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-219.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-29T03:27:34 -!- dekar [~dekar@212.255.224.68] has quit [Ping timeout: 240 seconds]
2012-10-29T03:28:02 -!- dekar [~dekar@212.255.226.180] has joined ##stm32
2012-10-29T03:28:06 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-29T03:36:58 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Read error: Connection reset by peer]
2012-10-29T03:37:31 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-29T03:40:44 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-29T03:50:13 -!- dekar [~dekar@212.255.226.180] has quit [Ping timeout: 246 seconds]
2012-10-29T04:07:29 -!- BrainDamage1 [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-29T04:07:41 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-29T04:08:40 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-29T04:37:44 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-29T04:39:37 < bsdfox_> HTT-Bird, isn't PbSn 63/37 solid? you want paste for a reflow oven
2012-10-29T04:40:24 < HTT-Bird> bsdfox_: 63/37 comes in paste form too
2012-10-29T04:40:44 < dongs> yeah.. theres leadfree/leaded paste too
2012-10-29T04:40:49 < bsdfox_> it seems to be the magic formula
2012-10-29T04:40:58 < dongs> i use zephpaste.. best shit ever
2012-10-29T04:41:01 < bsdfox_> dongs, yeah.. ever used the silver stuff? fuck that
2012-10-29T04:41:23 < dongs> http://www.zeph.com/zephpaste.htm t his shit
2012-10-29T04:41:35 < dongs> no, i use leadfree paste,  do use leadfree paste though
2012-10-29T04:42:37 < dongs> err
2012-10-29T04:42:39 < dongs> what the fuck did i jsut say
2012-10-29T04:43:01 < dongs> i meant, never tried silver shit (dunno what that is), but I do use leadfree only when it comes to paste
2012-10-29T04:43:35 < dongs> hand solder is mostly leadfree too
2012-10-29T04:50:22 < bsdfox_> silver solder
2012-10-29T05:18:40 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-29T05:28:51 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-29T05:47:41 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-29T05:51:17 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-10-29T06:33:07 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-29T06:33:12 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-29T07:20:05 -!- txf [~txf@bl12-90-46.dsl.telepac.pt] has quit [Ping timeout: 244 seconds]
2012-10-29T07:32:08 < emeb_mac> Always wondered where ol' Bob got his hair done.
2012-10-29T07:34:23 < emeb_mac> yup
2012-10-29T07:34:30 < emeb_mac> still do
2012-10-29T07:35:08 < emeb_mac> died in '95
2012-10-29T07:36:48 < dongs> nothing on my google.
2012-10-29T07:51:24 < emeb_mac> maybe US only as Bob Ross was primarily on PBS here.
2012-10-29T07:52:34 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-10-29T07:54:20 < emeb_mac> here's what you're missing: http://imagebin.org/233730
2012-10-29T08:01:34 < dongs> haha, just helped some "system engineer wiht 15 years of experience" at irdeto to solve an obvious-to-fix winxp driver issue with some smartcard reader shite
2012-10-29T08:01:47 < dongs> i think i better put this on my linkedin profile asap
2012-10-29T08:02:15 < emeb_mac> I guess he had the wrong kind of experience.
2012-10-29T08:02:49 < dongs> well, irdeto deals with smartcards and DRM etc.
2012-10-29T08:02:56 < dongs> you'd kind of expect peopel working there for 15 years
2012-10-29T08:02:59 < dongs> to have a bit of a clue about this shit.
2012-10-29T08:03:56 -!- txf [~txf@bl14-145-209.dsl.telepac.pt] has joined ##stm32
2012-10-29T08:32:51 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-29T08:54:47 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-hlareesrhuqrfrlv] has quit [Ping timeout: 268 seconds]
2012-10-29T08:55:10 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-wvpuaqtrgfdwmvlh] has joined ##stm32
2012-10-29T09:03:36 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 240 seconds]
2012-10-29T09:05:52 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-29T09:05:52 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Client Quit]
2012-10-29T09:06:47 -!- txf [~txf@bl14-145-209.dsl.telepac.pt] has quit [Ping timeout: 252 seconds]
2012-10-29T09:16:58 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has joined ##stm32
2012-10-29T09:18:46 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Ping timeout: 245 seconds]
2012-10-29T09:25:29 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-29T10:01:43 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-29T10:03:02 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-29T10:03:02 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-29T10:03:02 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-29T10:38:36 -!- claude is now known as claude_work
2012-10-29T10:44:22 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-29T11:07:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-29T11:15:38 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T11:29:32 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T11:58:44 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-29T12:05:59 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-29T12:22:59 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-10-29T13:11:35 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-29T13:11:38 -!- mode/##stm32 [+v Steffann] by ChanServ
2012-10-29T13:14:55 <+Steffann> nu
2012-10-29T13:15:45 < zyp> gnu
2012-10-29T13:16:37 <+Steffann> I'll make you gnu
2012-10-29T13:24:40 < prog> http://www.portalplanetasedna.com.ar/archivos_varios/hippie.jpg
2012-10-29T13:24:50 < baird> I regreat that I have only one gnu to give for my country.
2012-10-29T13:32:18 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-29T13:36:54 -!- daku is now known as DaKu
2012-10-29T13:45:06 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T14:06:33 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T14:31:44 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-29T14:32:58 < Laurenceb> The Revolution is a full 10DoF board, it contains the following components:
2012-10-29T14:32:58 < Laurenceb>     3 Axis Gyro
2012-10-29T14:32:58 < Laurenceb>     3 Axis Accelerometer
2012-10-29T14:32:59 < Laurenceb>     3 Axis Magnetometer
2012-10-29T14:32:59 < Laurenceb>     Barometric pressure sensor
2012-10-29T14:33:00 < Laurenceb> omg
2012-10-29T14:33:04 < Laurenceb> whod have thunk it
2012-10-29T14:34:08 < Laurenceb> OH MY GOD
2012-10-29T14:34:19 <+Steffann> Since when you hare a yankielander Laurenceb ?
2012-10-29T14:34:19 < Laurenceb> http://boards.4chan.org/b/res/433583908
2012-10-29T14:35:57 < Laurenceb> nsfw
2012-10-29T14:36:06 <+Steffann> A little late, not?
2012-10-29T14:37:30 <+Steffann> When you'll introduce the blaxter-ratio, Laurenceb ?
2012-10-29T14:39:40 <+Steffann> Blaxter ratio: Crap/NoCrap .. comparable to SNR
2012-10-29T14:41:27 <+Steffann> My Blaxter ratio is pretty high, i know
2012-10-29T14:42:06 < Laurenceb> http://franciswheat.com/
2012-10-29T14:42:30 <+Steffann> Uh what Laurenceb ?
2012-10-29T14:42:44 < Laurenceb> thats an actual guy from irc
2012-10-29T14:42:53 <+Steffann> His website looks better than yours :P
2012-10-29T14:43:06 <+Steffann> http://users.ox.ac.uk/~space/aims.html :D
2012-10-29T14:43:13 < dongs> 404
2012-10-29T14:43:32 <+Steffann> The chinese firewall is messing with your dongs
2012-10-29T14:43:36 < Laurenceb> http://freeonlinesurveys.com/app/rendersurvey.asp?sid=er4ndrpie9srdcp138124
2012-10-29T14:43:45 < Laurenceb> ^that got 4channed
2012-10-29T14:45:02 <+Steffann> Too bad 4chan isn't funny at all
2012-10-29T14:45:50 <+Steffann> Just a bunch of silly kids with a high blaxter ratio
2012-10-29T14:49:15 < dongs> hey blaxter, what do you know about coplanar waveguide shits
2012-10-29T14:49:23 < dongs> does that apply to usb transmission lines?
2012-10-29T14:49:35 < zyp> what?
2012-10-29T14:49:46 < dongs> im seeing some trollish layouts for hs usb on 2 layer shit that have it surrounded by gnd
2012-10-29T14:51:14 < Laurenceb> yeah theres standard designs
2012-10-29T14:51:31 < Laurenceb> dont bother trying to actually work out the impedance
2012-10-29T14:53:31 < Laurenceb> hot from my indox:  I have actually dropped it off? Do you want to call him? he is on floor A ? We can go together? I can meet you there? let me know?
2012-10-29T14:53:40 < Laurenceb> ???
2012-10-29T14:53:48 < zyp> profit
2012-10-29T14:54:12 < Laurenceb> i might send that as a reply
2012-10-29T14:57:13 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-29T14:57:17 -!- DaKu is now known as daku
2012-10-29T14:59:33 < dongs> http://www.focusembedded.com/blog/high-speed-usb-in-a-two-layer-pcb/ looks like i was right
2012-10-29T15:01:08 -!- daku is now known as DaKu
2012-10-29T15:04:27 -!- Darkside [~Darkside@compsci.adl/committee/darkside] has joined ##stm32
2012-10-29T15:04:52 < Laurenceb> dongs: DarkSide says hello
2012-10-29T15:04:56 < Darkside> lol
2012-10-29T15:05:05 < Darkside> i just had to join to see if this was legit
2012-10-29T15:05:07 < Laurenceb> dongs: DarkSide is ->  http://franciswheat.com/
2012-10-29T15:05:21 < Darkside> fffffuuuuu
2012-10-29T15:05:27 <+Steffann> A a friend from Laurenceb
2012-10-29T15:05:28 < Laurenceb> hahaha
2012-10-29T15:05:30 <+Steffann> *Ah
2012-10-29T15:05:31 < Darkside> my secret is out
2012-10-29T15:05:31 < Laurenceb> trollllleeeddd
2012-10-29T15:05:43 < dongs> this WAS pretty legit
2012-10-29T15:05:46 < dongs> until Laurenceb and his pals joined
2012-10-29T15:05:51 <+Steffann> You have more underaged friends Laurenceb ?
2012-10-29T15:05:56 < Laurenceb> yes
2012-10-29T15:06:04 <+Steffann> pedobear
2012-10-29T15:06:13 < Darkside> note: that picture is quite old now
2012-10-29T15:06:23 <+Steffann> quite as in .. 2 years?
2012-10-29T15:06:27 < Darkside> probably at least 6
2012-10-29T15:06:57 < Darkside> 2005 i think
2012-10-29T15:06:59 < Darkside> not 100%
2012-10-29T15:08:01 <+Steffann> The DSC-H5 is from 2006 :P
2012-10-29T15:08:10 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T15:08:19 < dongs> so you're 13 now?
2012-10-29T15:08:21 < Darkside> hmm
2012-10-29T15:08:25 < Darkside> must be 2006 then
2012-10-29T15:08:34 <+Steffann> 19-12-2006 – 13:55
2012-10-29T15:08:38 < Darkside> actually that makes sense!
2012-10-29T15:08:39 < Darkside> aha!
2012-10-29T15:08:43 < Darkside> the EXIF data is still in the image i take it
2012-10-29T15:09:02 < baird> D'oh. I've just remembered I've had one of these wiki.openmoko.org/wiki/Debug_Board_v3 the whole time..
2012-10-29T15:09:19 <+Steffann> You should be more like baird, Darkside … he removes all exif data from photos
2012-10-29T15:09:35 < Darkside> heh
2012-10-29T15:09:45 < Darkside> anyway, i won't ruin this channel any more than Laurenceb already has
2012-10-29T15:09:48 < Darkside> peace out
2012-10-29T15:09:49 -!- Darkside [~Darkside@compsci.adl/committee/darkside] has left ##stm32 []
2012-10-29T15:10:12 < baird> Mostly because checking the exif and geolocation data is something I always do. :P
2012-10-29T15:10:21 < baird> (..with other people's pics)
2012-10-29T15:10:51 <+Steffann> Lets all join #compsci.adl :P
2012-10-29T15:11:02 < Laurenceb> hehe
2012-10-29T15:11:39 < dongs> #compsci.adl -> this adl? http://www.adl.org/
2012-10-29T15:11:54 < Laurenceb> haha
2012-10-29T15:12:14 < baird> #################compsci.adl lol
2012-10-29T15:12:17 < Laurenceb> http://www.adelaide.edu.au/
2012-10-29T15:12:48 <+Steffann> Oh, he is an aussielander?
2012-10-29T15:12:55 < dongs> attn baird
2012-10-29T15:13:25 < baird> attn dongs!
2012-10-29T15:13:42 < dongs> baird: how many crapberrypies you got now
2012-10-29T15:14:11 < baird> Still the 5.
2012-10-29T15:15:41 < baird> I'm thinking of picking up another this week to dedicate for the NetBSD port.
2012-10-29T15:16:19 < baird> That'll involve checking if I'm poor or not.
2012-10-29T15:16:23 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-29T15:16:34 < zyp> you have to be really poor if you can't afford one
2012-10-29T15:17:42 < baird> Yeah... I've been there. I'd recently just had to fork over ~$1300 in 250cc scooter services and rego in the past month. ://
2012-10-29T15:17:57 <+Steffann> No vegemite for baird this month
2012-10-29T15:18:15 <+Steffann> and next month
2012-10-29T15:18:54 < baird> sucks
2012-10-29T15:19:19 < baird> I'll also have to drink homebrew instead of buying anything.
2012-10-29T15:19:34 <+Steffann> What's wrong with water? :)
2012-10-29T15:19:43 < zyp> it's too wet
2012-10-29T15:20:09 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T15:20:14 < karlp> too susceptible to things growing in it.
2012-10-29T15:20:22 < karlp> adding a few percent alcohol can help with that
2012-10-29T15:20:59 < Laurenceb> /dev/sdc:
2012-10-29T15:20:59 < Laurenceb>  Timing buffered disk reads:    4 MB in  5.00 seconds = 819.43 kB/sec
2012-10-29T15:21:01 < baird> When was the last time anyone got wasted on plain water?
2012-10-29T15:21:01 < Laurenceb> niceeee
2012-10-29T15:21:28 < zyp> hardcore speed yo
2012-10-29T15:21:38 < Laurenceb> yup
2012-10-29T15:22:04 < baird> What's that on?
2012-10-29T15:22:11 < Laurenceb> F1 with spi2
2012-10-29T15:22:22 < Laurenceb> connected to microsd
2012-10-29T15:22:48 < Laurenceb> then usb with interleaved dma to transfer the data
2012-10-29T15:23:08 < Laurenceb> theoretical is about 850/900KB/s
2012-10-29T15:23:11 < Laurenceb> so im happy
2012-10-29T15:23:19 < Laurenceb>  - due to usb protocol overhead
2012-10-29T15:23:36 < dongs> Laurenceb: pcbs for loggerclone submitted
2012-10-29T15:24:18 < Laurenceb> cool
2012-10-29T15:24:31 < Laurenceb> theoretical is obviously 1.5MB/s
2012-10-29T15:24:38 < Laurenceb> but usb mass storage is lame
2012-10-29T15:24:57 < Laurenceb> especially with small usb packets
2012-10-29T15:26:16 < zyp> well, if you need speed, don't use fs usb
2012-10-29T15:27:44 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T15:33:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-29T15:33:15 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Remote host closed the connection]
2012-10-29T15:33:26 < dongs> if you need speed your obvious choice is arduino due
2012-10-29T15:34:06 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-29T15:35:44 < claude_work> picaxe with highly optimized basic code (many gosubs !!)
2012-10-29T15:45:26 < Laurenceb> hehe picaxe
2012-10-29T15:46:27 <+Steffann> No need to run flyback ? Run for Sandy?
2012-10-29T15:48:19 <+Steffann> Not going to leave the area? :)
2012-10-29T15:48:45 <+Steffann> Oh, none in yours case :P
2012-10-29T15:48:50 <+Steffann> -s
2012-10-29T15:52:57 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T15:57:54 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T16:03:10 < Laurenceb> https://groups.google.com/forum/?fromgroups=#!topic/ukhas/uXUxd4rWjEk
2012-10-29T16:03:19 < Laurenceb> ^origin of 4chan thread
2012-10-29T16:03:42 -!- BrainDamage1 [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-29T16:03:42 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Disconnected by services]
2012-10-29T16:03:56 -!- BrainDamage1 [~BrainDama@62.10.6.74] has quit [Client Quit]
2012-10-29T16:04:22 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-29T16:16:44 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-29T16:20:25 < dongs> so how did they solve due stuff wiht tarduino
2012-10-29T16:20:43 < dongs> i dont see a new tarduino for donwload
2012-10-29T16:20:46 < dongs> wiht arm compilers etc.
2012-10-29T16:20:56 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-29T16:23:15 <+Steffann> http://arduino.cc/en/Main/ArduinoDUE HAH
2012-10-29T16:23:43 < dongs> haha
2012-10-29T16:23:47 < zyp> when is it due?
2012-10-29T16:24:13 < dongs> i see no due software at all
2012-10-29T16:24:26 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T16:26:33 < baird> weeb joke: http://i.imgur.com/lXt9K.jpg
2012-10-29T16:28:31 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-29T16:28:40 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-29T16:30:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-29T16:38:55 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T16:39:09 < emeb_mac> "zyp: when is it due?" I see what you did there.
2012-10-29T16:43:29 <+Steffann> emeb_mac's eyes still work
2012-10-29T16:43:57 < emeb_mac> :P not a subtle joke.
2012-10-29T16:44:43 < emeb_mac> so, presumably for due they need to have the ard. IDE pull in armgcc?
2012-10-29T16:45:27 -!- Thorn [~Thorn@unaffiliated/thorn] has quit [Read error: Connection reset by peer]
2012-10-29T16:45:30 <+Steffann> i guess so
2012-10-29T16:45:33 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-29T16:45:53 -!- Thorn [~Thorn@128-73-41-139.broadband.corbina.ru] has joined ##stm32
2012-10-29T16:45:53 -!- Thorn [~Thorn@128-73-41-139.broadband.corbina.ru] has quit [Changing host]
2012-10-29T16:45:53 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-29T16:46:53 < zyp> emeb_mac, well, they could just ship it with an avr emulator
2012-10-29T16:47:07 < emeb_mac> ;)
2012-10-29T16:47:18 < emeb_mac> wonder what the big holdup is?
2012-10-29T16:47:43 < dongs> its pretty hard to dumb down a proper chip into "upload and pray" IDE
2012-10-29T16:49:06 < emeb_mac> Wonder if they're still using an FTDI for the download process...
2012-10-29T16:49:28 < dongs> no, they're using some failmega 8u2
2012-10-29T16:49:31 < dongs> for usb>serial
2012-10-29T16:49:37 <+Steffann> Also for this new board?
2012-10-29T16:49:40 < dongs> yes
2012-10-29T16:49:57 < emeb_mac> ah - kinda like on the Uno.
2012-10-29T16:51:12 < emeb_mac> I suppose that's better than using the native USB on the M3 part - in case you brick the main MCU the programmer MCU still works...
2012-10-29T16:52:41 < zyp> but still relies on a uart bootloader? I don't see how it's better
2012-10-29T16:53:27 < emeb_mac> I suppose not. Forgot that bit.
2012-10-29T16:53:50 < zyp> the advantage with a seperate programmer port is that it keeps its connection across program/run cycles and doesn't need to be reenumerated
2012-10-29T16:54:02 < zyp> well, the advantage I see.
2012-10-29T16:54:17 < emeb_mac> that's a pretty small advantage.
2012-10-29T16:54:34 < zyp> not when you consider that you can issue a mcu reset through it
2012-10-29T16:54:54 < zyp> which you wouldn't be able to do otherwise without application support
2012-10-29T16:54:56 <+Steffann> Too bad it doesn't do swd or something.. just some uart crap
2012-10-29T16:55:09 < emeb_mac> zyp: best point so far.
2012-10-29T16:55:15 < emeb_mac> Steffann: this
2012-10-29T16:55:29 <+Steffann> emeb_mac: that
2012-10-29T16:55:38 < emeb_mac> Steffann: the other
2012-10-29T16:55:40 < dongs> who cares about due anyway
2012-10-29T16:55:43 < dongs> they kept the moron pin spacing
2012-10-29T16:56:01 < emeb_mac> 0.05" offset between top & bottom rows?
2012-10-29T16:56:02 < dongs> whos gonna read the 3V/5V warning?
2012-10-29T16:56:12 <+Steffann> warning?
2012-10-29T16:56:17 < dongs> apparently none of sam3 pins are even 5V tolerant
2012-10-29T16:56:21 < dongs> 3.3V max, or smoke is out
2012-10-29T16:56:27 < dongs> and yet it has same shitty tarduino shield pinout
2012-10-29T16:56:31 < dongs> that fits all the current failshields.
2012-10-29T16:56:52 < emeb_mac> help increase sales
2012-10-29T16:57:34 <+Steffann> "The Due is priced at $49 " :D
2012-10-29T16:57:53 < zyp> oh well
2012-10-29T16:58:15 < zyp> by the way, are there any cheap decent sam3u devboards?
2012-10-29T16:58:21 < emeb_mac> more expensive than Raspberry Pi. Forget that. :P
2012-10-29T16:58:59 <+Steffann> olimex zyp ?
2012-10-29T16:59:03 < dongs> http://www.xtreamermobile.com/ haha these guys are surely jokers
2012-10-29T16:59:18 < dongs> Resolution: 480 x 800 WVGA on 5"???? for 200 EUR???
2012-10-29T16:59:27 < dongs> i have 480x800 in 2.4" on my cellphone 5 years ago
2012-10-29T16:59:37 < emeb_mac> https://www.olimex.com/Products/ARM/Atmel/
2012-10-29T17:00:04 < dongs> thosea re mega old
2012-10-29T17:00:06 < dongs> none for sam3u
2012-10-29T17:00:26 < dongs> and sam7 shit is ancient
2012-10-29T17:00:46 < dongs> i remember cursing at sam7 in like 2008, it sucked back then, probably still sucks now
2012-10-29T17:01:33  * emeb_mac did a board w/ sam7. It's OK, not great.
2012-10-29T17:01:38 < zyp> I'm not going to dick around with anything older than cortex-m
2012-10-29T17:02:52 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T17:04:52 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Remote host closed the connection]
2012-10-29T17:08:08 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-29T17:09:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T17:09:33 < bsdfox_> I ported this (crudely) to openlibcm3 but don't get any action on my LEDs on vldiscovery. anyone want to take a look and point me in the right direction? https://gist.github.com/3974066
2012-10-29T17:10:53 < dongs> jesus christ. whats the point of using libopencm3 if youre gonna dick with regs directly??
2012-10-29T17:12:09 < zyp> your lines 30-40 doesn't make sense
2012-10-29T17:12:58 < bsdfox_> zyp, yeah I know it;s just left over from the old code
2012-10-29T17:13:04 < dongs> why tim3 remap?
2012-10-29T17:13:20 < bsdfox_> GPIOC_CRH |= (1 << 3) | (1 << 7); is what it's doing
2012-10-29T17:14:17 < Laurenceb> nice - F3 has seperate dfu  interfaces for flash and option bytes
2012-10-29T17:14:25 < baird> otherwise known as (17 << 3)
2012-10-29T17:14:26 < bsdfox_> dongs, original code did that. I'm thinking that might be an area of concern but if I look at the TIM3 output compare ch3/4 the remap looks like it puts the outputs on the LEDs
2012-10-29T17:14:45 < Laurenceb> F3 is the shit
2012-10-29T17:17:25 < zyp> bsdfox_, still doesn't make sense, that sets the pins to inputs
2012-10-29T17:18:17 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-29T17:18:53 < zyp> try with CRH = (CRH & ~0xff) | 0x11
2012-10-29T17:19:40 < bsdfox_> yeah oops I see that now
2012-10-29T17:21:54 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T17:27:29 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-29T17:32:40 < bsdfox_> made that change but still not getting the output I'd expect
2012-10-29T17:35:22 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-29T17:36:14 < bsdfox_> is 0x11 really what I want though?
2012-10-29T17:37:03 < bsdfox_> I'm thinking I want 0xAA
2012-10-29T17:37:11 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T17:37:35 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-29T17:42:47 < zyp> oh, sorry, I missed the timer part
2012-10-29T17:43:57 -!- barthess [~barthess@164.177.212.190] has joined ##stm32
2012-10-29T17:44:43 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-29T17:46:54 -!- claude_work is now known as claude
2012-10-29T17:51:46 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T17:55:11 < Laurenceb> http://s1.b3ta.com/host/creative/79547/1351186309/eaudehumanity.gif
2012-10-29T17:55:46 <+Steffann> Blaxter ratio += 1
2012-10-29T18:09:48 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 256 seconds]
2012-10-29T18:15:20 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-29T18:17:17 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-29T18:29:05 -!- dekar [~dekar@212.255.250.138] has joined ##stm32
2012-10-29T18:29:09 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-29T18:29:42 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-29T19:17:39 -!- barthess [~barthess@164.177.212.190] has quit [Quit: Leaving.]
2012-10-29T19:21:28 < emeb> *crickets*
2012-10-29T19:38:36 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-29T19:49:56 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 252 seconds]
2012-10-29T19:50:26 <+Steffann> Calm before the storm, emeb
2012-10-29T19:56:12 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-29T20:03:58 <+Steffann> Salut
2012-10-29T20:07:28 < emeb> Yá'át'ééh
2012-10-29T20:07:33 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-29T20:07:51 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-29T20:09:38 < karlp> eat my þoð's
2012-10-29T20:09:53 <+Steffann> You really have 1000 and 1 ways to say hello? emeb
2012-10-29T20:10:47 < emeb> There _are_ more than that. Just drawing on the local culture. ;)
2012-10-29T20:12:11 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-29T20:13:50 < emeb> Now that we have interwebs the value of code talkers has likely decreased markedly.
2012-10-29T20:14:53 -!- dekar_ [~dekar@212.255.228.139] has joined ##stm32
2012-10-29T20:14:56 -!- mode/##stm32 [+v dekar_] by ChanServ
2012-10-29T20:16:06 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-10-29T20:18:16 -!- barthess [~barthess@164.177.212.190] has joined ##stm32
2012-10-29T20:18:28 -!- dekar [~dekar@212.255.250.138] has quit [Ping timeout: 248 seconds]
2012-10-29T20:20:25 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-29T20:23:31 < BrainDamage> emeb: http://www.xkcd.com/257/
2012-10-29T20:24:30 < emeb> :D
2012-10-29T20:43:12 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-10-29T20:43:49 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 260 seconds]
2012-10-29T20:45:49 -!- claude is now known as Claude
2012-10-29T20:47:17 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has joined ##stm32
2012-10-29T20:47:17 -!- TeknoJuce [~TeknoJuce@bas1-london15-1168024643.dsl.bell.ca] has quit [Changing host]
2012-10-29T20:47:17 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-29T20:50:32 -!- NonaSuomy [~TeknoJuce@69.158.167.190] has joined ##stm32
2012-10-29T20:50:33 -!- NonaSuomy [~TeknoJuce@69.158.167.190] has quit [Changing host]
2012-10-29T20:50:33 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-29T20:53:49 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 268 seconds]
2012-10-29T21:17:28 < jpa-> hmm nice http://www.seeedstudio.com/blog/2012/10/19/dso-quad-application-software-competition/
2012-10-29T21:18:30 -!- phantoxe [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-29T21:21:09 < TitanMKD> jpa- would be more fun with an STM32F4 ;)
2012-10-29T21:21:40 < emeb> everyone's a critic. :)
2012-10-29T21:21:43 < jpa-> you are allowed to modify the hardware ,)
2012-10-29T21:23:44 < emeb> hmm... ad9288 is relatively inexpensive.
2012-10-29T21:24:15 < TitanMKD> jpa- I have a real Oscilloscope Rigole DS1102E for that job
2012-10-29T21:24:21 < emeb> might make a nice I/Q sampler for SDR.
2012-10-29T21:24:27 < jpa-> TitanMKD: can you write your own software for it?
2012-10-29T21:24:30 < TitanMKD> jpa- not portable but it is not limited like DSO Quad
2012-10-29T21:24:53 < jpa-> and do you have the schematics for it?
2012-10-29T21:24:57 < TitanMKD> jpa- I do not need to write sw for my scope as it have everything i need for a scope ;)
2012-10-29T21:25:15 < TitanMKD> even USB connection with sw to drive it through windows ;)
2012-10-29T21:26:02 < jpa-> do you have a battery capacity meter for it? or automatically sweeping frequency response analyzer?
2012-10-29T21:27:54 < jpa-> a portable scope like dso quad is a bit different thing than a desktop model; sure, it doesn't sample at 1GSps, but it does other useful stuff
2012-10-29T21:28:56 < TitanMKD> jpa- of course it's not portable like DSO
2012-10-29T21:29:22  * emeb studies the qdso power/charging ckt.
2012-10-29T21:29:33 < jpa-> emeb: qdso != dso quad
2012-10-29T21:29:39 < jpa-> qdso is crap :D
2012-10-29T21:29:52 < emeb> oops - dso quad. sry!
2012-10-29T21:29:55 < jpa-> :)
2012-10-29T21:30:09 < TitanMKD> jpa- dso quad is just a bit limited by the CPU they used a little STM32F1 :(
2012-10-29T21:30:16 < emeb> too bad the names are so similar.
2012-10-29T21:30:18 < jpa-> TitanMKD: i agree
2012-10-29T21:30:20 < TitanMKD> jpa- too few memory and FFT is slow ...
2012-10-29T21:30:29 < TitanMKD> also display is not really fasty
2012-10-29T21:30:36 < TitanMKD> limited by STM32F1 too
2012-10-29T21:30:43 < jpa-> emeb: yeah, qdso seems to be a chinese knock off of dso quad.. which is itself chinese :P
2012-10-29T21:31:12 < emeb> ever tightening circles, speeding up. aaaahhhhh......
2012-10-29T21:31:17 < jpa-> TitanMKD: well 30 fps 512-point FFT is not that bad.. not great but not unusably slow
2012-10-29T21:31:34 < TitanMKD> jpa- with STM32F4 and the dsp they could have a must ;)
2012-10-29T21:31:40 < emeb> 30fps is very nice.
2012-10-29T21:31:44 < jpa-> TitanMKD: display could be much faster with better software
2012-10-29T21:32:10 < TitanMKD> jpa- of course sw could be also optimized because 72MHz MCU is not so bad
2012-10-29T21:32:13 < jpa-> TitanMKD: STM32F4 is just as useless if the software would be crap like it is now :)
2012-10-29T21:32:43 < TitanMKD> jpa- yes but crap @ 168MHz instead of 72Mhz is still at least 2 or 3 time faster ;)
2012-10-29T21:32:45 < jpa-> i hope those $5600 of prizes will put some interest to the software development :)
2012-10-29T21:32:48 < emeb> TitanMKD: use an F3xx & run 72, cheaper.
2012-10-29T21:33:11 < emeb> get fpu & dsp
2012-10-29T21:33:26 < TitanMKD> but F3 has a M4 or M3  ?
2012-10-29T21:33:34 < TitanMKD> because M4 is really better too
2012-10-29T21:33:44 < TitanMKD> with branch predictions
2012-10-29T21:33:54 < jpa-> TitanMKD: remember that STM32F4 didn't exist when dso quad came to market
2012-10-29T21:34:21 < TitanMKD> jpa- of course it was a must when it was out but they could upgrade it for few USD
2012-10-29T21:34:31 < jpa-> and break all software
2012-10-29T21:34:31 < TitanMKD> IIRC F4 is pin compatible with some F1
2012-10-29T21:34:34 < jpa-> is not
2012-10-29T21:34:37 < jpa-> needs Vcap
2012-10-29T21:34:49 < TitanMKD> ha yes just for some Vcap ;)
2012-10-29T21:34:51 < emeb> TitanMKD: F3 has M4F
2012-10-29T21:34:51 < jpa-> and e.g. DMA and GPIO peripherals are totally different
2012-10-29T21:35:06 < jpa-> so software would need large changes
2012-10-29T21:35:07 < TitanMKD> yes of course specific stuff need rewrite
2012-10-29T21:35:16 < TitanMKD> but they could move to chibios for that ;)
2012-10-29T21:35:22 < TitanMKD> or using libopencm3
2012-10-29T21:35:42 < jpa-> true
2012-10-29T21:36:09 < TitanMKD> but yes all that sw was not existing during dso quad development
2012-10-29T21:36:15 < jpa-> but wouldn't help me, i already have the F1 version :)
2012-10-29T21:56:58 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 272 seconds]
2012-10-29T21:57:32 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:01:52 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 246 seconds]
2012-10-29T22:02:49 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:07:43 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-10-29T22:13:51 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:18:34 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 244 seconds]
2012-10-29T22:19:15 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:24:20 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-10-29T22:24:35 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:29:06 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 240 seconds]
2012-10-29T22:33:50 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-29T22:34:57 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:39:16 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 248 seconds]
2012-10-29T22:39:18 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-29T22:40:11 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:44:36 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 245 seconds]
2012-10-29T22:45:28 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:45:53 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-29T22:49:59 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds]
2012-10-29T22:50:50 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:55:50 -!- grummund [~user@unaffiliated/grummund] has quit [Remote host closed the connection]
2012-10-29T22:56:14 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T22:56:45 -!- Steffann [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Quit: Ik ga weg]
2012-10-29T22:57:57 -!- DLPeterson [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has quit [Ping timeout: 276 seconds]
2012-10-29T22:59:34 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-29T22:59:34 < emeb> OK - buddy of mine is trying to build the BMP firmware for ST-LINK for one of his discovery boards.
2012-10-29T23:00:04 < emeb> Can't seem to get the libopencm3 to compile - all kinds of complaints from python about yaml.
2012-10-29T23:00:51 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 268 seconds]
2012-10-29T23:01:41 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-29T23:04:07 < GargantuaSauce> there is probably a python yaml package missing?
2012-10-29T23:05:48 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-29T23:05:49 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-29T23:06:11 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 255 seconds]
2012-10-29T23:07:23 < emeb> yeah. Recently added to the prereqs for libopencm3 - it's not in the makefile for my build which I cloned a month ago.
2012-10-29T23:13:25 -!- Claude is now known as claude
2012-10-29T23:15:14 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Changing host]
2012-10-29T23:15:14 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32
2012-10-29T23:15:14 -!- ServerMode/##stm32 [+v Steffanx] by lindbohm.freenode.net
2012-10-29T23:19:03 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-151.pptp.stw-bonn.de] has joined ##stm32
2012-10-29T23:22:36 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: Ik ga weg]
2012-10-29T23:22:38 < Erlkoenig> CFSR = *(0xE000ED28) = 00040000    =>    "An integrity check error has occurred on EXC_RETURN." ... verified with the debugger that LR didn't change during the ISR, and is still = 0xfffffff9.   still i get a HardFault. but why :'(
2012-10-29T23:24:31 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-29T23:24:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-29T23:24:38 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Changing host]
2012-10-29T23:24:38 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32
2012-10-29T23:24:41 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-29T23:25:08 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-29T23:28:19 -!- barthess [~barthess@164.177.212.190] has quit [Ping timeout: 260 seconds]
2012-10-29T23:57:50 -!- DaKu is now known as daku
2012-10-29T23:58:17 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 256 seconds]
--- Day changed Tue Oct 30 2012
2012-10-30T00:08:27 -!- NonaSuomy is now known as TeknoJuce
2012-10-30T00:11:18 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-30T00:16:07 < Erlkoenig> meh got it. LR was of course correct, but the stack a bit corrupt. whyever CFSR mourns about EXC_RETURN.
2012-10-30T00:29:34 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-30T00:29:35 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-30T00:29:35 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-30T00:42:43 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: This computer has gone to sleep]
2012-10-30T00:43:01 < emeb> OK - got libopencm3 built.
2012-10-30T00:43:17 < emeb> Now looks like the blackmagic source is out of sync with libopencm3.
2012-10-30T00:43:53 < emeb> blackmagic is #including stuff from libopencm3 but has the file locations wrong.
2012-10-30T00:43:54 < zyp> are you building the latest with libopencm3 as a submodule?
2012-10-30T00:43:58 < emeb> yes
2012-10-30T00:44:09 < zyp> and you checked out the submodule correctly?
2012-10-30T00:44:22 < emeb> used the link in the blackmagic github
2012-10-30T00:44:57 < zyp> huh?
2012-10-30T00:45:06 < emeb> error -> platforms/stlink/platform.c:26:38: fatal error: libopencm3/stm32/systick.h: No such file or directory
2012-10-30T00:45:24 < emeb> systick.h is actually in libopencm3/include/stm32/systick.h
2012-10-30T00:45:29 < zyp> git is supposed to check out the right commit automatically
2012-10-30T00:45:38 < karlp> is the submodule correctly pulling a fixed commit? or against HEAD?
2012-10-30T00:45:50 < zyp> karlp, the former
2012-10-30T00:45:57 < zyp> that's the whole point of submodules
2012-10-30T00:46:06 < karlp> well yeah, if it's working properly :)
2012-10-30T00:46:14 < zyp> you just have to run «git submodule update» to do that
2012-10-30T00:46:45 < zyp> after having done init and sync to initialize the submodule, of course
2012-10-30T00:47:17 < emeb> ok - didn't do that stuff. just cloned blackmagic in one area and cloned libopencm3 in another area
2012-10-30T00:47:39 < karlp> https://github.com/gsmcmullin/blackmagic/blob/master/HACKING could probably do with an update....
2012-10-30T00:48:00 < emeb> so the right procedure is to clone blackmagic, then do some git foolishness down in the libopencm3 dir?
2012-10-30T00:48:26 < zyp> no, from blackmagic dir
2012-10-30T00:48:37 < emeb> trying again...
2012-10-30T00:48:39 < zyp> git submodule init, git submodule sync, git submodule update
2012-10-30T00:49:01 < zyp> and then you can just type make from blackmagic dir and everything will be built
2012-10-30T00:49:26 < zyp> I tested here, building newest rev for stlink works fine
2012-10-30T00:50:16 < Erlkoenig> is that probe better than texane stlink?
2012-10-30T00:50:23 < zyp> yes.
2012-10-30T00:50:31 < emeb> OK, building...
2012-10-30T00:50:48 < Erlkoenig> hmm
2012-10-30T00:51:04 < emeb> yay - seems to have built
2012-10-30T00:51:20 < emeb> didn't know about the extra stuff needed for git submodules.
2012-10-30T00:51:30 < emeb> as you say, HACKING needs an update.
2012-10-30T00:52:45 < zyp> submodules adds a bit complexity, but is very handy when you know how to use them
2012-10-30T00:54:05 < emeb> indeed.
2012-10-30T00:54:22 < zyp> it ensures that you always can go back and build old revisions without having to deal with library api differences, because it also stores the reference to the library revision it was written against
2012-10-30T00:54:36 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 240 seconds]
2012-10-30T00:54:55 < zyp> or even that you can build the newest version after the library have done api changes
2012-10-30T00:55:08 < emeb> it's pretty slick *IF YOU KNOW IT'S THERE*
2012-10-30T00:55:11 < zyp> like what seems to have happened now that they merged some new architectures into libopencm3
2012-10-30T00:55:15 < zyp> :)
2012-10-30T00:55:22 < emeb> yeah
2012-10-30T00:55:23 < zyp> well, now you know
2012-10-30T00:55:46 < emeb> need to write that down in my notebook. Otherwise it'll leak out of me brain.
2012-10-30T00:55:59 < zyp> I'd submit a change to HACKING, but I'm heading to bed now
2012-10-30T00:56:11 < emeb> thanks for the help.
2012-10-30T00:56:13 < zyp> HACKING is obviously outdated since it still refers to libopenstm32
2012-10-30T00:56:21 < zyp> :p
2012-10-30T00:56:22 < dongs> hello HACKERS
2012-10-30T00:56:27 < zyp> hello DONGS
2012-10-30T00:56:31 < emeb> hello dongler
2012-10-30T01:02:00 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-30T01:08:31 -!- NonaSuomy [~TeknoJuce@69.158.167.190] has joined ##stm32
2012-10-30T01:08:32 -!- NonaSuomy [~TeknoJuce@69.158.167.190] has quit [Changing host]
2012-10-30T01:08:32 -!- NonaSuomy [~TeknoJuce@xbmc/staff/TeknoJuce] has joined ##stm32
2012-10-30T01:10:25 < emeb> dongs: no sign of boards in the mail today. Ordered the 16-pin sockets from DK though - should be here in a few days.
2012-10-30T01:10:37 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-30T01:11:28 < dongs> meh
2012-10-30T01:11:34 < dongs> lame usps
2012-10-30T01:12:04 -!- TeknoJuce [~TeknoJuce@xbmc/staff/TeknoJuce] has quit [Ping timeout: 256 seconds]
2012-10-30T01:13:17 < emeb> it can be variable.
2012-10-30T01:13:33 < emeb> generally pretty happy with it - faster than UPS/FedEx for some things.
2012-10-30T01:18:03 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-30T01:18:03 -!- bsdfox_ [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-30T01:18:03 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-30T01:18:08 -!- dfletcher__ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T01:21:30 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 255 seconds]
2012-10-30T01:21:42 -!- dfletcher_ [~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 264 seconds]
2012-10-30T01:27:27 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 260 seconds]
2012-10-30T01:37:20 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-30T01:45:05 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: No route to host]
2012-10-30T01:45:45 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has joined ##stm32
2012-10-30T01:46:15 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has joined ##stm32
2012-10-30T01:55:00 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-30T01:57:38 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has quit [Quit: Ex-Chat]
2012-10-30T02:04:46 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 252 seconds]
2012-10-30T02:26:13 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-30T02:36:23 -!- ossifrage [~ossifrage@c-71-224-59-2.hsd1.pa.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-30T02:36:58 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-30T02:49:15 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-30T03:05:28 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-30T03:06:25 -!- Laurenceb_ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 256 seconds]
2012-10-30T03:11:40 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-30T03:16:20 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-30T03:23:19 -!- Erlkoenig [~erlkoenig@pptp-212-201-71-151.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-30T03:24:48 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-30T03:55:13 < baird> https://www.youtube.com/watch?v=4ZwYlj4ym1Y "STM32F4 - Julia Set Fractal Demo (FPU)"
2012-10-30T03:56:19 < Bird|lappy> do any of you KiCad users know why a multi-unit part (a 4069 over here) has power and ground connections on all units, not just 1 or 2?
2012-10-30T03:58:41 < BrainDamage> that's weird, I have used several libs that only had the power pins once
2012-10-30T04:06:00 < Bird|lappy> so just connect 1 power and 1 gnd pin?
2012-10-30T04:06:07 < emeb> yup
2012-10-30T04:08:14 < Bird|lappy> what happens if you try to make a part sink more current than the Iol specified on the datasheet (besides the "obvious" of Vol and Pd rising?)
2012-10-30T04:09:37 < emeb> those are the principle effects. You may damage the device if you exceed the maximums.
2012-10-30T04:10:07 < emeb> note that most data sheets spec a max and also max that still meets Vol.
2012-10-30T04:10:41 < emeb> eg - STM32F1xx specs like 4ma while still hitting Vol, but 24 max.
2012-10-30T04:11:38 < Bird|lappy> I'm dealing with a 4069 (yes, I know, old tech :P)
2012-10-30T04:11:54 < Bird|lappy> and the TI datasheet for it gives a maximum dissipation but not a maximum Io
2012-10-30T04:12:56 < emeb> well, those have much larger geometries and lower current densities. Harder to hurt the transistors with mere current.
2012-10-30T04:13:48 < emeb> the Rds is probably high enough that they'll self-limit, so it makes sense that Pd is the limiting factor.
2012-10-30T04:24:12 < dongs> https://youtube.com/watch?v=ZAqYZ433TeQ#t=3m
2012-10-30T04:29:33 < emeb> almost as good as that WWII UXB detonating in DE a few months back.
2012-10-30T04:30:25 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T04:57:33 -!- bsdfox_ [~Bob@unaffiliated/bsdfox] has quit [Read error: Connection reset by peer]
2012-10-30T05:03:30 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-30T05:14:49 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has joined ##stm32
2012-10-30T05:14:50 -!- upgrdman [~upgrdman@71-92-66-171.static.mtpk.ca.charter.com] has quit [Changing host]
2012-10-30T05:14:50 -!- upgrdman [~upgrdman@blender/artist/upgrdman] has joined ##stm32
2012-10-30T05:14:59 < upgrdman> anyone here use geda pcb?
2012-10-30T05:15:20 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-30T05:15:23 < dongs> yeah, emeb does
2012-10-30T05:15:35 < dongs> probably the first and last user!
2012-10-30T05:15:52 < upgrdman> why last? you don't like it?
2012-10-30T05:16:14 < Bird|lappy> upgrdman, yeah, I work with it
2012-10-30T05:17:00 < emeb> right - I use geda/pcb
2012-10-30T05:17:35 < ds2> emeb: have you looked at KiCAD?
2012-10-30T05:17:51 < emeb> ds2:  yes. I've done a couple projects with it.
2012-10-30T05:17:54 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit [Ping timeout: 240 seconds]
2012-10-30T05:17:55 < Bird|lappy> upgrdman, if you have a question, fire away
2012-10-30T05:18:13 < ds2> emeb: how does it compare to geda?
2012-10-30T05:18:25 < emeb> ds2: it's better integrated
2012-10-30T05:18:42 < emeb> the PCB design tool is a littler nicer.
2012-10-30T05:19:08 < upgrdman> im learning to use it. i'm doing single layer pcb's ... at home. just copper and fiberglass. i want a ground plane, and i want to connect some LQFP pads, and some SMD cap/resistor pads to the ground plane. 1. can i put the ground plane on the component layer, or should i put it in the GND layer? 2. how do i connect stuff to it
2012-10-30T05:19:26 < Bird|lappy> upgrdman, you mean a ground fill?
2012-10-30T05:19:30 < dongs> I *highly* dislike kicad's main guy, dick hollenbeck
2012-10-30T05:19:31 < upgrdman> ok, yes
2012-10-30T05:19:33 < dongs> he's.. a fucking dick.
2012-10-30T05:19:46 < dongs> i thnk he's probably the last/the only/remaining developer on that shit.
2012-10-30T05:20:15 < ds2> hmmm I see
2012-10-30T05:20:23 < ds2> diptrace's 3D ability sounds so tempting
2012-10-30T05:20:26 < dongs> so he might as well just rename it to DickCAD
2012-10-30T05:20:30 -!- zippe [~Adium@173.11.99.161] has quit [Quit: Leaving.]
2012-10-30T05:20:36 < dongs> ds2: did you convert that vrml shite I linked the other day?
2012-10-30T05:20:40 < dongs> im curious to see what it looks like
2012-10-30T05:20:41 -!- zippe [~Adium@173.11.99.161] has joined ##stm32
2012-10-30T05:20:55 < ds2> dongs: didn't have a chance to yet. burned with other stuff and this week is ARM TechCON
2012-10-30T05:21:39 < dongs> there was some ST conference in sydney today i wonder if baird got free shit
2012-10-30T05:22:01 < ds2> someone is going to present on uCLinux there too
2012-10-30T05:22:45 < upgrdman> Bird|lappy, emeb, any ideas?
2012-10-30T05:23:39 < Bird|lappy> upgrdman, I'd put it as a fill in the same layer as your traces
2012-10-30T05:23:51 < Bird|lappy> and then you'll have to use traces to manually thermal to the SMT pads
2012-10-30T05:24:01 < upgrdman> ok. is it a bad idea to do my traces, and everything else, on the component layer?
2012-10-30T05:24:36 < Bird|lappy> not particularly, as long as you remember what's going on, an "upside down" single sided board can work just fine, and may be more convenient if you have a SMT-centric worldview
2012-10-30T05:24:47 < upgrdman> so theres no way to thermal an SMD pad to a ground fill other than manually drawing the traces?
2012-10-30T05:25:03 < upgrdman> yes, almost everything will be SMD.
2012-10-30T05:27:13 < upgrdman> also, how do i adjust the gap between pads and the ground fill?
2012-10-30T05:27:28 < Bird|lappy> upgrdman, can't remember how to adjust the pad-to-fill gap
2012-10-30T05:27:46 < Bird|lappy> and yeah, the gEDA/pcb thermal tool only works on pins, not pads
2012-10-30T05:28:45 < upgrdman> :( ok
2012-10-30T05:28:53 < ds2> anyone have thoughts on multiple ground planes in parallel?
2012-10-30T05:29:36 < Bird|lappy> ds2, how many layers do you have?
2012-10-30T05:29:46 < ds2> 4
2012-10-30T05:29:52 < ds2> trying to decide if I should flood them all
2012-10-30T05:30:21 < Bird|lappy> ds2, normally, if you have a dedicated ground plane on a 4 layer board, you should be fine
2012-10-30T05:30:39 < baird> doggy: which conference is this?
2012-10-30T05:30:43 < ds2> I don't have a dedicated plane
2012-10-30T05:30:50 < ds2> not enough routing resources
2012-10-30T05:30:59 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-30T05:31:28 < Bird|lappy> ds2, ugh
2012-10-30T05:31:36 < Bird|lappy> what are you trying to route? o.O
2012-10-30T05:31:54 < dongs> baird:  "STMicroelectronics Technology Day" seminar in Sydney
2012-10-30T05:31:55 < ds2> BGA144 version of the STM32F103
2012-10-30T05:32:16 < dongs> ohj that was yeasterday
2012-10-30T05:32:29 < Bird|lappy> ds2, oh, fun, BGA.
2012-10-30T05:32:41 < Bird|lappy> what pitch?
2012-10-30T05:32:50 < ds2> 0.8mm
2012-10-30T05:32:52 < Bird|lappy> hrm.
2012-10-30T05:32:56 < Bird|lappy> and rules?
2012-10-30T05:33:03 < ds2> 5/5
2012-10-30T05:33:05 < ds2> 10mil via
2012-10-30T05:33:14 < ds2> 10mil via hole
2012-10-30T05:33:38 < Bird|lappy> ds2, your power and ground vias are direct connects, right?
2012-10-30T05:33:54 < ds2> define "direct connects"
2012-10-30T05:34:04 < ds2> they are nice fat 16mil traces
2012-10-30T05:34:08 < Bird|lappy> ds2, i.e. they connect directly to the planes with no extra trace length?
2012-10-30T05:34:14 < ds2> nope
2012-10-30T05:34:38 < Bird|lappy> how much of that 144 is occupied by pwr/gnd?
2012-10-30T05:34:42 < ds2> I can't dedicate a plane, otherwise, I need a 6L board
2012-10-30T05:34:55 < ds2> maybe 20balls
2012-10-30T05:35:06 < Bird|lappy> hrm.
2012-10-30T05:35:58 < ds2> yep
2012-10-30T05:37:01 < Bird|lappy> that's...screwy
2012-10-30T05:37:40 < ds2> there is a PSRAM in there so a lot of lines get routed to another bga
2012-10-30T05:37:53 < Bird|lappy> you could try popping into ##electronics and pinging azonenberg? he's done some BGA stuff before, I'm just sort of familiar with how one'd go about it
2012-10-30T05:38:11 < ds2> I might just try it and see
2012-10-30T05:38:21 < dongs> beard: y ou coulda had a chance to snag a free f3discovery, probably
2012-10-30T05:38:34 < ds2> just wasnt sure what the conventional wisdom is for multiple planes
2012-10-30T05:38:49 < dongs> haha, azonenberg.
2012-10-30T05:39:04 < dongs> i remember that guy being very full of himself.
2012-10-30T05:41:18 < ds2> if they are doing it like at techcon, it might be less painful to pay the $10 and buy it
2012-10-30T05:41:57 < dongs> well, if i could actually pay $10 for a f3disco I'd be all set.
2012-10-30T05:42:07 < dongs> its ok though i got my pal in UK to grab some from farnell and they're now on the way to me
2012-10-30T05:42:16 < dongs> still faster than digikey ship date!
2012-10-30T05:42:36 < Bird|lappy> well, if nothing else, what I'd do is use a fill on each inner layer, then bring the inner layer breakout traces up to the surface ASAP
2012-10-30T05:43:30 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has quit [Remote host closed the connection]
2012-10-30T05:43:45 < ds2> i just don't get that excited about the M4's... M0's and M1 maybe
2012-10-30T05:45:16 < ds2> other then the higher clock speed, is there really anything else that interests a lot of folks?
2012-10-30T05:45:49 < dongs> F3 is interesteind because of M4F in 48pin package.
2012-10-30T05:46:42 < ds2> that's a QFP...still pretty big
2012-10-30T05:47:05 < dongs> noone atually stocks BGA versions of F1/whatever at reasonable prices.
2012-10-30T05:47:17 < ds2> what are you calling reasonable prices?
2012-10-30T05:47:31 < dongs> normal prices.
2012-10-30T05:47:33 < dongs> not digikey prices
2012-10-30T05:47:37 < ds2> what's that?
2012-10-30T05:47:41 < dongs> i.e. STM32F103CB for <$2
2012-10-30T05:47:42 < ds2> $10?$15?
2012-10-30T05:47:47 < ds2> ohhhh
2012-10-30T05:47:53 < ds2> is the CB the small memory one?
2012-10-30T05:48:05 < dongs> cb is 128k flash
2012-10-30T05:48:08 < dongs> the biggest in 48pin
2012-10-30T05:48:15 < dongs> http://www.digikey.com/product-detail/en/STM32F103CBT6/497-6288-ND/1754420
2012-10-30T05:48:20 < dongs> 20k ram / 128k flash.
2012-10-30T05:48:22 < ds2> ah
2012-10-30T05:48:23 < Bird|lappy> btw: can one use a kitchen-store vacuum sealer gizmo as an impulse sealer, for one?
2012-10-30T05:48:26 < upgrdman> have you checked mouser?
2012-10-30T05:48:28 < dongs> F3 is 40k ram/256k
2012-10-30T05:48:37 < ds2> I was looking only in the 1M Flash versions
2012-10-30T05:48:42 < ds2> XL ones I think
2012-10-30T05:57:23 < upgrdman> lol
2012-10-30T06:32:08 -!- [7] [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-30T06:32:14 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-30T06:37:09 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-30T07:01:24 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 252 seconds]
2012-10-30T07:20:36 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has joined ##stm32
2012-10-30T07:38:28 -!- Laurenceb_ [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-30T07:39:06 -!- Laurenceb [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 246 seconds]
2012-10-30T07:41:05 < emeb> 'sup devs?
2012-10-30T07:42:32 < R2COM> looks like firewire protocol license doesnt cost anything
2012-10-30T07:43:07 < emeb> who's doing FW?
2012-10-30T07:43:32 < R2COM> i was just looking at it
2012-10-30T07:47:50 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-30T07:48:14 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Read error: Operation timed out]
2012-10-30T07:51:03 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-30T08:10:49 -!- gurgalof [~gurgalof@c80-216-122-131.bredband.comhem.se] has joined ##stm32
2012-10-30T08:30:20 -!- R2COM [~blackops@c-98-230-217-221.hsd1.nm.comcast.net] has quit [Quit: Leaving.]
2012-10-30T08:32:23 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-30T08:44:15 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Ping timeout: 260 seconds]
2012-10-30T09:17:33 -!- dekar_ [~dekar@212.255.228.139] has quit [Quit: This computer has gone to sleep]
2012-10-30T09:17:59 -!- claude is now known as Claude
2012-10-30T09:18:32 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has joined ##stm32
2012-10-30T09:23:04 -!- Claude is now known as claude
2012-10-30T09:37:24 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-30T09:53:00 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-wvpuaqtrgfdwmvlh] has quit [Remote host closed the connection]
2012-10-30T09:53:07 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-itztantrcgzpqiee] has joined ##stm32
2012-10-30T09:59:51 -!- dekar [~dekar@studpool-wlan-75-151.fs.fbi.h-da.de] has joined ##stm32
2012-10-30T09:59:53 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T10:00:57 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-itztantrcgzpqiee] has quit [Ping timeout: 276 seconds]
2012-10-30T10:06:30 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-30T10:07:50 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ossvnzjdqfmjsxdj] has joined ##stm32
2012-10-30T10:40:38 -!- dekar [~dekar@studpool-wlan-75-151.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-30T10:46:57 < dongs> open source?
2012-10-30T10:51:58 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-30T11:15:15 -!- claude is now known as claude_work
2012-10-30T11:23:44 < karlp> dongs donging again, claims < $2, pastes link for 3.37 in 5k units :)
2012-10-30T11:24:07 < karlp> oh, never mind, /me reads context properly.
2012-10-30T11:43:01 -!- dekar [~dekar@studpool-wlan-75-166.fs.fbi.h-da.de] has joined ##stm32
2012-10-30T11:43:05 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T11:52:13 < dongs> karlp: thats what I pay, not what digikey sells them for
2012-10-30T11:54:40 < karlp> yeah, my bad reading irclogs backwards, sorry.
2012-10-30T12:41:09 -!- dekar [~dekar@studpool-wlan-75-166.fs.fbi.h-da.de] has quit [Quit: This computer has gone to sleep]
2012-10-30T13:48:18 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-30T13:57:14 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-30T14:01:37 -!- tavish3 [~tavish@120.59.161.72] has joined ##stm32
2012-10-30T14:05:23 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has quit [Quit: ZNC - http://znc.in]
2012-10-30T14:06:26 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has joined ##stm32
2012-10-30T14:22:05 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has quit [Quit: ZNC - http://znc.in]
2012-10-30T14:23:42 < tavish3> i am using a timer overflow interrupt to implement a delay function, and i am using that Delay inside SysTick interrupt handler, but the timer interrupt seems to be not preempting systick even though i think i have set the priorities OK. NVIC_SetPriority(TIM7_IRQn, NVIC_EncodePriority(NVIC_PriorityGroup_1, 0, 0)); NVIC_SetPriority(SysTick_IRQn, NVIC_EncodePriority(NVIC_PriorityGroup_1, 0, 2));
2012-10-30T14:28:59 -!- tavish3 [~tavish@120.59.161.72] has quit [Ping timeout: 260 seconds]
2012-10-30T14:36:32 -!- Laurenceb__ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-30T15:06:19 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-30T15:20:09 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-30T15:52:52 < BrainDamage> attn dongs http://www.rcgroups.com/forums/showpost.php?p=17755481&postcount=1045
2012-10-30T15:55:16 < dongs> clicking
2012-10-30T15:55:23 < dongs> BrainDamage: mega fucking old
2012-10-30T15:55:29 < dongs> i've been picking on that guy fore ver
2012-10-30T15:55:48 < BrainDamage> sorry, was new to me
2012-10-30T15:56:15 < dongs> http://www.rcgroups.com/forums/thumbgallery.php?t=1290963&do=threadgallery&page=2
2012-10-30T15:56:18 < dongs> BrainDamage: ^ more lulz
2012-10-30T15:56:40 < BrainDamage> thanks
2012-10-30T15:56:58 < dongs> http://static.rcgroups.net/forums/attachments/3/2/4/0/5/7/a4064277-55-uyXlG.jpg < the only decent soldering in that thread, cuz thats mine
2012-10-30T16:02:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-30T16:02:11 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Changing host]
2012-10-30T16:02:11 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32
2012-10-30T16:02:14 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-30T16:15:36 <+Steffanx> This is how you HELP people Laurenceb_ : https://www.youtube.com/watch?v=ffEYqGGYXRk&noredirect=1
2012-10-30T16:17:58 -!- mode/##stm32 [+o Steffanx] by ChanServ
2012-10-30T16:18:03 -!- Steffanx changed the topic of ##stm32 to: Prepare for the two minutes of /hate/love. This channel is publicly logged, ask zlog for more (no free z80s anymore). | Join us building the ##stm32 toolchain: https://github.com/EliasOenal/TNT (now with multilib) |  We have a wiki http://stm32.izua.ro | How engineers buy parts http://tim.cexx.org/?p=919
2012-10-30T16:18:12 -!- mode/##stm32 [-o Steffanx] by ChanServ
2012-10-30T16:23:42 < dongs> Steffanx: old
2012-10-30T16:23:44 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-30T16:24:03 <+Steffanx> Only 50k views, so new enough dongs :P
2012-10-30T16:24:19 < dongs> http://www.trollaxor.com/2012/10/apple-to-merge-ios-os-x.html this is new
2012-10-30T16:24:40 <+Steffanx> but that's nothing interesting
2012-10-30T16:26:19 <+Steffanx> And that domain. Sounds like a source one can trust
2012-10-30T16:28:45 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Read error: Connection reset by peer]
2012-10-30T16:29:05 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-30T16:32:37 <+Steffanx> Sure you didn't write that blog dongs?
2012-10-30T16:38:05 < dongs> yes
2012-10-30T16:38:06 < dongs> i dont blog
2012-10-30T16:41:47 < emeb_mac> dongs doesn't blog / blogs doesn't dong
2012-10-30T16:43:09 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-30T16:46:59 < dongs> Item location:
2012-10-30T16:47:00 < dongs> Clifton, New Jersey, United States
2012-10-30T16:47:00 < dongs> oh fuck
2012-10-30T16:47:08 < dongs> so i shouldnt be asking the guy 'wheres my order' right
2012-10-30T16:47:09 < dongs> LOL
2012-10-30T16:50:01 -!- Oldboy is now known as KingoftheJuice
2012-10-30T16:53:03 -!- KingoftheJuice is now known as qua
2012-10-30T16:53:14 < Laurenceb__> under water
2012-10-30T17:04:57 < dongs> http://www.northjersey.com/news/176280281_Clifton_Office_of_Emergency_Management_activated__shelter_opened.html
2012-10-30T17:05:00 < dongs> lols
2012-10-30T17:18:23 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-30T17:22:03 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-10-30T17:22:16 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T17:23:08 < Laurenceb__> pmsl
2012-10-30T17:23:14 < Laurenceb__> they built a apple boat
2012-10-30T17:23:15 < Laurenceb__> http://www.theregister.co.uk/2012/10/29/steve_jobs_yacht_venus/
2012-10-30T17:23:23 <+Steffanx> Yeah, in dutchland
2012-10-30T17:23:29 < Laurenceb__> "The design suggests this isn't really an ocean-going craft, since those hull windows would be a liability in very rough seas and the doors and windows of the superstructure aren't slanted but run at a 90 degree angle from the deck. This make the ship trickier to steer in high winds and much less resistant to wave damage."
2012-10-30T17:23:36 < Laurenceb__> obviously an apple boat
2012-10-30T17:24:19 <+Steffanx> "obviously an apple boat" you care about that Laurenceb__ ____ ?
2012-10-30T17:25:05 < Laurenceb__> " it looks to me like someone has just designed the first sea-going garden shed."
2012-10-30T17:25:06 < Laurenceb__> lol
2012-10-30T17:25:39 <+Steffanx> Blaxter ratio →∞
2012-10-30T17:26:42 < Laurenceb__> but the comment are genius
2012-10-30T17:28:13 <+Steffanx> Wonder how long it takes to get the blaxter ratio into google
2012-10-30T17:29:58 < Laurenceb__> forever
2012-10-30T17:30:43 <+Steffanx> I bet your life is complete when it's there
2012-10-30T17:39:12 -!- elektrinis [~circuit@82-135-241-134.static.zebra.lt] has quit [Quit: pokðt]
2012-10-30T17:53:02 -!- qua is now known as Oldboy
2012-10-30T18:06:02 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-10-30T18:06:12 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-30T18:11:23 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T18:11:49 -!- claude_work is now known as claude
2012-10-30T18:16:39 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has joined ##stm32
2012-10-30T18:16:39 -!- bsdfox [~Bob@c-71-197-83-152.hsd1.ca.comcast.net] has quit [Changing host]
2012-10-30T18:16:39 -!- bsdfox [~Bob@unaffiliated/bsdfox] has joined ##stm32
2012-10-30T18:17:53 -!- bsdfox [~Bob@unaffiliated/bsdfox] has quit [Client Quit]
2012-10-30T18:20:01 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-30T18:33:53 -!- barthess [~barthess@178.154.7.229] has joined ##stm32
2012-10-30T18:44:21 -!- dekar [~dekar@drms-590ed688.pool.mediaWays.net] has joined ##stm32
2012-10-30T18:44:25 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T18:49:10 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T18:57:46 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Read error: Operation timed out]
2012-10-30T19:02:51 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-30T19:05:29 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T19:09:59 -!- Vutral_ [ss@p5B2A62D1.dip.t-dialin.net] has joined ##stm32
2012-10-30T19:12:13 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T19:22:48 -!- Vutral_ [ss@p5B2A62D1.dip.t-dialin.net] has quit [Ping timeout: 255 seconds]
2012-10-30T19:27:08 -!- scrts [~quassel@unaffiliated/scrts] has quit [Read error: Connection reset by peer]
2012-10-30T19:29:29 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T19:30:10 <+Steffanx>  Vutral get yourself a better connect
2012-10-30T19:30:13 <+Steffanx> ion
2012-10-30T19:32:56 -!- dekar [~dekar@drms-590ed688.pool.mediaWays.net] has quit [Quit: This computer has gone to sleep]
2012-10-30T19:33:36 -!- dekar [~dekar@2002:590e:d688:0:9158:510b:221e:df2e] has joined ##stm32
2012-10-30T19:33:39 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T19:34:09 -!- dekar [~dekar@2002:590e:d688:0:9158:510b:221e:df2e] has quit [Client Quit]
2012-10-30T19:37:38 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T19:42:59 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T19:47:23 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: Leaving]
2012-10-30T19:49:35 -!- dekar [~dekar@drms-590ed688.pool.mediaWays.net] has joined ##stm32
2012-10-30T19:49:41 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T19:55:49 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-30T19:57:29 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-30T20:08:51 -!- dekar [~dekar@drms-590ed688.pool.mediaWays.net] has quit [Quit: This computer has gone to sleep]
2012-10-30T20:11:41 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-30T20:11:59 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Ping timeout: 260 seconds]
2012-10-30T20:14:27 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 260 seconds]
2012-10-30T20:15:01 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T20:16:37 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Ping timeout: 256 seconds]
2012-10-30T20:24:14 -!- dekar [~dekar@2002:590e:d688:0:2168:dc4a:21e4:85fe] has joined ##stm32
2012-10-30T20:24:18 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-30T20:26:24 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-30T20:27:07 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T20:37:38 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T20:45:46 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T20:50:08 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T20:50:54 < DLPeterson_> anybody here have experience with timers + optical encoders?
2012-10-30T20:50:57 -!- dekar [~dekar@2002:590e:d688:0:2168:dc4a:21e4:85fe] has quit [Quit: Leaving]
2012-10-30T20:51:11 < jpa-> kind of
2012-10-30T20:51:46 < DLPeterson_> jpa-, I've configured one timer to operate in encoder mode, so the CNT goes up or down appropriately
2012-10-30T20:52:09 < DLPeterson_> but what I want to do now is use another timer in input capture to time each quadrature count
2012-10-30T20:52:46 < DLPeterson_> I've figured out how to time each rising/falling edge on one or the other channels (A or B), so that is like half quadrature counts
2012-10-30T20:53:38 < jpa-> hm, not sure how i would do that
2012-10-30T20:53:54 < jpa-> i ended up doing the quadrature stuff using EXTI anyway, because i needed the timer for other stuff
2012-10-30T20:54:19 < DLPeterson_> but I can't figure out how to time quadrature counts, I am thinking I need to either live with half quadrature timing or to make sure each A/B channel is on a dedicated timer
2012-10-30T20:54:50 < DLPeterson_> currently I have channel A from three different encoders going to three channels on the same timer
2012-10-30T20:57:19 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T21:03:04 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-30T21:03:11 -!- prog [prog@bornbrilliant.org] has quit [Ping timeout: 255 seconds]
2012-10-30T21:04:38 -!- prog [~prog@bornbrilliant.org] has joined ##stm32
2012-10-30T21:05:53 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T21:06:56 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-69.pptp.stw-bonn.de] has joined ##stm32
2012-10-30T21:10:30 -!- prog [~prog@bornbrilliant.org] has quit [Quit: That's all folks!]
2012-10-30T21:11:17 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T21:15:32 -!- prog [~prog@bornbrilliant.org] has joined ##stm32
2012-10-30T21:23:03 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T21:28:34 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T21:35:27 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-30T21:35:36 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T21:41:27 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T21:48:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T21:50:52 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T22:04:52 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T22:05:35 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-30T22:17:39 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-10-30T22:18:11 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T22:18:42 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 276 seconds]
2012-10-30T22:19:50 -!- tkoskine [tkoskine@kapsi.fi] has joined ##stm32
2012-10-30T22:22:47 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 260 seconds]
2012-10-30T22:24:30 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-30T22:29:27 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T22:29:51 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T22:40:26 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-10-30T22:40:42 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-30T22:45:33 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T22:54:40 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T22:55:05 -!- barthess [~barthess@178.154.7.229] has quit [Quit: Leaving.]
2012-10-30T23:03:26 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has quit [Quit: No Ping reply in 180 seconds.]
2012-10-30T23:03:50 -!- ntfreak [~ntfreak@unaffiliated/ntfreak] has joined ##stm32
2012-10-30T23:08:20 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-30T23:13:57 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-30T23:24:08 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has quit [Ping timeout: 244 seconds]
2012-10-30T23:24:15 -!- ABLomas [~ABLomas@78-60-198-200.static.zebra.lt] has joined ##stm32
2012-10-30T23:42:23 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: No route to host]
2012-10-30T23:43:35 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-30T23:45:33 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-30T23:55:33 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
--- Day changed Wed Oct 31 2012
2012-10-31T00:00:28 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Read error: No route to host]
2012-10-31T00:07:10 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-31T00:13:03 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-31T00:15:57 < ds2> Hmmm
2012-10-31T00:22:28 < emeb> what's humming ds2?
2012-10-31T00:23:03 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T00:24:08 < TitanMKD> anyone have done some speed test with USB CDC ?
2012-10-31T00:24:22 < TitanMKD> I obtain about 200KB/s with Realterm
2012-10-31T00:24:32 < TitanMKD> in Full Speed and with EP 64bytes
2012-10-31T00:24:51 < TitanMKD> on LPC4330
2012-10-31T00:27:22 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 246 seconds]
2012-10-31T00:28:04 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-10-31T00:28:09 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has quit [Quit: This computer has gone to sleep]
2012-10-31T00:28:40 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T00:28:43 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T00:29:32 < Laurenceb__> TitanMKD: topic is ##stm32
2012-10-31T00:29:54 < Erlkoenig> µC-Racist? :D
2012-10-31T00:30:33 < emeb> ##stm32 and "surprising URLs" :)
2012-10-31T00:30:58 < emeb> although not such a surprise any more. :\
2012-10-31T00:31:24 < Laurenceb__> filthy lpc crap
2012-10-31T00:31:30 < Laurenceb__> id gas the lot of them
2012-10-31T00:31:44 < emeb> not an NXP fan?
2012-10-31T00:34:04 < BrainDamage> jokes aside, this channel could change to generic cortex-m and not much would change I bet
2012-10-31T00:35:32 < ds2> emeb: trying to finalize the fab vendor
2012-10-31T00:35:38 < ds2> board looks good enough to release to fab
2012-10-31T00:36:01 < Erlkoenig> still didn't manage to get my F4 to "safely" generate timer interrupts every 500ms. when changing some unrelated part of the programm, the pattern of which timer interrupts are missed change...
2012-10-31T00:36:30 < emeb> ds2: nice.
2012-10-31T00:37:09 < ds2> emeb: know what time dongs gets on?
2012-10-31T00:37:25 < emeb> ds2: usually about now. It's early morn in his locale.
2012-10-31T00:37:46  * emeb is getting a board w/ STM32F4 and PSRAM soon.
2012-10-31T00:37:55 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-31T00:37:56 < emeb> I'll be having fun getting the FSMC going.
2012-10-31T00:38:13 < Erlkoenig> http://www.mikrocontroller.net/attachment/158910/blink.c   the code in question... has anyone ever succeeded in getting a timer interrupt working reliably?
2012-10-31T00:39:26 < ds2> emeb: is the FSMC the same on the F4 and F1?
2012-10-31T00:39:58 < emeb> ds2: No idea. Probably not _too_ different, but there are some API changes - registers aren't identical on all peripherals.
2012-10-31T00:40:31 < ds2> emeb: ah I see. have to do that on my board
2012-10-31T00:40:42 < ds2> emeb: what's the PSRAM for in your board?
2012-10-31T00:41:27 < emeb> ds2: I've got a MT45W2MW16PGA
2012-10-31T00:42:42 < upgrdman> Erlkoenig, i think you need to toggle in the isr
2012-10-31T00:43:01 < upgrdman> Erlkoenig, for example, here's my notes:
2012-10-31T00:43:12 < upgrdman> http://www.farrellf.com/projects/hardware/2012-08-11_STM32F4_Basics:_Timers_%28Part_1%29/
2012-10-31T00:43:21 < ds2> emeb: that's a 1.8V part, isn't it?
2012-10-31T00:43:37 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T00:43:59 < Erlkoenig> upgrdman: i AM toggling in the ISR
2012-10-31T00:44:40 < upgrdman> o shit...overlooked that. nm
2012-10-31T00:44:45 < emeb> ds2:  yes 1.8V core, 3.3V IO
2012-10-31T00:45:04 < ds2> emeb: so you have the luxury of multiple LDOs on that board?
2012-10-31T00:45:21 -!- Thorn [~Thorn@unaffiliated/thorn] has quit []
2012-10-31T00:45:38 < emeb> ds2: yeah - this is actually a stack of 3 boards - UI/IO, MCU, memory.
2012-10-31T00:45:41 < upgrdman> Erlkoenig, looks like you're trying to toggle D12 and D14, right?
2012-10-31T00:46:00 < Erlkoenig> yep exactly
2012-10-31T00:46:01 < emeb> the PSRAM is on a separate daughter card that can be swapped out for larger/smaller as needed.
2012-10-31T00:46:20 < Erlkoenig> D12 is only to simulate "work" being done in main()
2012-10-31T00:47:14 < upgrdman> sure.
2012-10-31T00:47:29 < upgrdman> checked waveforms with scope/analyzer?
2012-10-31T00:47:53 < Laurenceb__> whats PSRAM?
2012-10-31T00:48:27 < emeb> pseudo-static
2012-10-31T00:48:28 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-31T00:48:37 < emeb> DRAM with self refresh and simple I/O bus.
2012-10-31T00:49:59 < Erlkoenig> my scope is not available right now :S ... well the timer is very slow so i can just see the frequency
2012-10-31T00:51:25 < ds2> emeb: I see... how expensive is the micron option?
2012-10-31T00:51:26 < upgrdman> Erlkoenig, and whats it doing wrong? inconsistent? wrong period?
2012-10-31T00:52:06 < Erlkoenig> well the timer seems to miss interrupts... it's like 3 interrupts occur normally, one missed
2012-10-31T00:52:14 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-31T00:52:55 < Erlkoenig> but change one line in main() - like more led blinking - one interrupt occured, two missed
2012-10-31T00:52:57 < upgrdman> Erlkoenig, never seen that. comment out the D12 toggling and WFI. see if anything changes wrt to misses
2012-10-31T00:52:59 < Erlkoenig> nondeterministic
2012-10-31T00:53:13 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T00:53:58 < Erlkoenig> oh sorry forgot to mention... WITH WFI it works (in this case), WITHOUT  it doesn't
2012-10-31T00:54:36 < Erlkoenig> ah without D12 and without WFI works again
2012-10-31T00:54:38 < upgrdman> hum. try my isr code, maybe your isr is triggered by multiple flags.
2012-10-31T00:54:40 < Erlkoenig> like a chaotic system
2012-10-31T00:54:59 < upgrdman> if(TIM6->SR & TIM_SR_UIF != 0) do something;
2012-10-31T00:55:05 < upgrdman> then
2012-10-31T00:55:07 < upgrdman> TIM6->SR &= ~TIM_SR_UIF;
2012-10-31T00:55:14 < upgrdman> well TIM10 for you
2012-10-31T00:55:41 < emeb> ds2: Those parts are fairly inexpensive @ DK - only about $5/qty100. But the daughterboards require a slightly more complex PCB fab so the price goes up a bit.
2012-10-31T00:56:31 < Erlkoenig> nope, doesn't work
2012-10-31T00:57:31 < upgrdman> Erlkoenig, fire up a bugger, break on your isr, then examine the contents of the TIM10->SR
2012-10-31T00:57:49 < Erlkoenig> it's always 1, tried it already
2012-10-31T00:58:33 < upgrdman> then why "TIM10->SR = (uint16_t)~TIM_IT_Update;" ? why not =0;
2012-10-31T00:59:08 < Erlkoenig> well after that it's 0 ;)
2012-10-31T00:59:33 < Erlkoenig> to not clear the other flags... well probably not needed, since they aren't needed in that application
2012-10-31T01:00:00 < Erlkoenig> TIM10->SR = (uint16_t)~TIM_IT_Update;    is the way StdPerip does it
2012-10-31T01:00:02 < upgrdman> that DOES clear the others iirc. you use =, not |= or &=
2012-10-31T01:00:20 < Erlkoenig> no, the bits are insensitive for writing '1'
2012-10-31T01:00:26 < upgrdman> o
2012-10-31T01:00:38 < Erlkoenig> but &= ~1 doesn't work either, i like tried everything :D
2012-10-31T01:00:44 < upgrdman> :)
2012-10-31T01:00:52 < Erlkoenig> sometimes under some conditions works, add a NOP somewehre, doesn't work
2012-10-31T01:01:58 < upgrdman> Erlkoenig, not sure then. i would find some working code, test it, adapt it, test it, and factor out until you find the problem
2012-10-31T01:02:25 < Erlkoenig> i'll probably wear out the flash before finding the actual problem :S
2012-10-31T01:02:29 < upgrdman> maybe try my code, remove OPM, and dick with the PSC/ARR as needed
2012-10-31T01:02:51 < Erlkoenig> i can't develop when ANY change to ANY part of the program can cause the timer to malfunction!
2012-10-31T01:03:04 < upgrdman> Erlkoenig, i spend almost a month figuring out why my spi code didnt work :) maybe 300 flashes...
2012-10-31T01:03:09 < dongs> lunix talk??
2012-10-31T01:03:35 < Erlkoenig> oh i already got SPI working ^^
2012-10-31T01:03:46 < dongs> < Erlkoenig> http://www.mikrocontroller.net/attachment/158910/blink.c   the code in question... has anyone ever succeeded in
2012-10-31T01:03:50 < dongs>                    getting a timer interrupt working reliably?
2012-10-31T01:03:52 < dongs> um yes
2012-10-31T01:04:00 < Erlkoenig> cool
2012-10-31T01:04:02 < Erlkoenig> how did you do it?
2012-10-31T01:04:24 < dongs> prety much exactly what you got there, minus WFI part
2012-10-31T01:04:30 < upgrdman> my problems were two fold: typo in shitty stm docs, and i though you could leave CS low all the time, when it had to be high between transactions
2012-10-31T01:04:36 < Erlkoenig> with WFI works, without doesnt work
2012-10-31T01:04:46 < dongs> um
2012-10-31T01:05:02 < Erlkoenig> upgrdman: ah very nice
2012-10-31T01:05:14 < dongs> Erlkoenig: while (1); works fine
2012-10-31T01:05:17 < Erlkoenig> no
2012-10-31T01:05:20 < dongs> yes it does
2012-10-31T01:05:57 < dongs> what is it "not doing"?
2012-10-31T01:06:05 < dongs> your timer update ISR toggles PD14 led
2012-10-31T01:06:09 < dongs> your while (1) toggles PD12
2012-10-31T01:06:15 < Erlkoenig> ah you're right, while(1); does work, while (1)  GPIOD->ODR ^= (1 << 12);     ddoesnt
2012-10-31T01:06:20 < dongs> of course not
2012-10-31T01:06:23 < Erlkoenig> why?
2012-10-31T01:06:27 < dongs> while (1) ->ODR blinks it real fucking fast
2012-10-31T01:06:30 < Erlkoenig> yep
2012-10-31T01:06:32 < dongs> you probably just see it dim.
2012-10-31T01:06:35 < Erlkoenig> exactly
2012-10-31T01:06:41 < dongs> so it IS working as expected.
2012-10-31T01:06:48 < Erlkoenig> lemme explain
2012-10-31T01:07:35 < dongs> just use the fucking TIM_ClearInterruptFlag (or wahtver) stuff from stdperiphlib.
2012-10-31T01:07:39 < dongs> stop trying to reinvent the hweel
2012-10-31T01:07:44 < dongs> or copy the code from that function
2012-10-31T01:07:48 < Erlkoenig> D12 is dimly lit. D14 blinks with 1Hz when doing while(1); or when while (1)  { GPIOD->ODR ^= (1 << 12); __WFI (); } .   BUT when doing while (1)  GPIOD->ODR ^= (1 << 12);,   D14 is blinking irregularly
2012-10-31T01:07:48 < dongs> TIM10->SR = (uint16_t)~TIM_IT_Update;
2012-10-31T01:07:50 < dongs> instead of this
2012-10-31T01:07:56 < Erlkoenig> aaalready tried that
2012-10-31T01:08:01 < Erlkoenig> it's exactly the same
2012-10-31T01:08:18 < Erlkoenig> that line is actually copied from the StdPeriph to see whether the function call causes problems
2012-10-31T01:10:43 < Erlkoenig> i am toggling D12 just to simulate "work" being done in the mainloop (want to do that later) and to see wether the work is being done by looking at the LED ad D12
2012-10-31T01:11:23 < dongs> thats not work
2012-10-31T01:11:37 < dongs> youre actually moving hardware (gpio).
2012-10-31T01:11:40 < Erlkoenig> i will add "useful" work later
2012-10-31T01:11:42 < dongs> no.
2012-10-31T01:11:48 < ds2> emeb: 'slightly more'? in what way?
2012-10-31T01:12:05 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T01:12:07 < Erlkoenig> this is just to see whether the f*cking mainloop runs!
2012-10-31T01:12:15 < Erlkoenig> or whether it froze or something
2012-10-31T01:12:22 < dongs> then dont use gpio for it
2012-10-31T01:12:31 < Erlkoenig> what should i use instead?
2012-10-31T01:12:31 < upgrdman> why not gpio?
2012-10-31T01:13:34 < emeb> ds2: most of the pcbs in this product are done on a cheaper 6/6 process. The PSRAM is done with a finer one to handle the BGA.
2012-10-31T01:14:03 < upgrdman> 6/6?
2012-10-31T01:14:17 < emeb> 6 mil lines/spaces.
2012-10-31T01:17:07 < Erlkoenig> dongs: what now?
2012-10-31T01:17:11 < ds2> emeb: ohhhh heheheehhe... gotcha... my PCB needed 5/5 to handle that foot print
2012-10-31T01:17:17 < dongs> Erlkoenig: i told you my opinion
2012-10-31T01:17:22 < ds2> I could switch that that part but it would cost me another LDO
2012-10-31T01:17:32 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has quit [Quit: Konversation terminated!]
2012-10-31T01:17:36 < Erlkoenig> which is, use something unknown in the mainloop to indicate it's working?
2012-10-31T01:17:36 < dongs> you are not simulating real work in the loop with constantly toggling gpio
2012-10-31T01:17:37 < upgrdman> dongs:     why not gpio?
2012-10-31T01:18:00 < upgrdman> why would gpio toggling cause the problem?
2012-10-31T01:18:03 < Erlkoenig> okay i will write a loop that does "work" and toggles GPIO so i can see the work being done
2012-10-31T01:18:12 < dongs> can ^= be interrupted, since it's  read , xor, write?
2012-10-31T01:18:15 < Erlkoenig> that won't fix anything
2012-10-31T01:18:23 < Erlkoenig> yes so what?
2012-10-31T01:18:27 < dongs> think about it
2012-10-31T01:18:34 < upgrdman> ooooooooooooooooooooo
2012-10-31T01:18:36 < dongs> your main loop is doing read
2012-10-31T01:18:40 < dongs> gets interrupted
2012-10-31T01:18:46 < Erlkoenig> that means the LED might blink funny for 10ns
2012-10-31T01:18:49 < dongs> the other shit does read, xor , write
2012-10-31T01:18:49 < Erlkoenig> so what
2012-10-31T01:19:03 < dongs> shrug
2012-10-31T01:19:04 < Erlkoenig> oh whait
2012-10-31T01:19:05 < Erlkoenig> wat
2012-10-31T01:19:07 < Erlkoenig> uhoh
2012-10-31T01:19:21 < dongs> then your main loop kills the shit interrupt wrote
2012-10-31T01:19:22 < upgrdman> Erlkoenig, maybe disable interrupts, toggle 12, enable interrupts, wfi
2012-10-31T01:19:25 < dongs> cuz it finishes xor + write
2012-10-31T01:19:27 < ds2> dongs: hey, what's the best way to dfm against ur design rules?
2012-10-31T01:19:53 < dongs> ds2: just teh spec I mentioned before, dont need to do anything special
2012-10-31T01:20:10 < dongs> x/x trace+space and whatever minsize drill
2012-10-31T01:20:14 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-31T01:20:41 < ds2> dongs: Ohh okay... some companies want to run a DFM
2012-10-31T01:20:56 < dongs> yeah
2012-10-31T01:21:02 < dongs> thats because they're lazy
2012-10-31T01:21:28 < emeb> does the "M" refer just to the PCB fab, or also to the picknplace?
2012-10-31T01:21:46 < dongs> i think he means drc or something, dfm is generally assembly term
2012-10-31T01:21:47 < upgrdman> dfm?
2012-10-31T01:21:54 < emeb> that's what I thought
2012-10-31T01:21:59 < ds2> oh crap
2012-10-31T01:22:02 < emeb> design for manufacturability.
2012-10-31T01:22:08 < ds2> my vendor is out of the PSRAM
2012-10-31T01:22:11 < emeb> ie - can you actually build it.
2012-10-31T01:22:15 < upgrdman> k
2012-10-31T01:22:23 < emeb> ds2: yoiks!
2012-10-31T01:23:20 < ds2> they had stock as of 2 wks ago :(
2012-10-31T01:24:56 < emeb> hate it when that happens, and it does.
2012-10-31T01:25:12 -!- Laurenceb__ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 244 seconds]
2012-10-31T01:26:58 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T01:27:38 -!- bgamari [~bgamari@pool-108-8-228-141.spfdma.east.verizon.net] has quit [Ping timeout: 252 seconds]
2012-10-31T01:28:00 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit []
2012-10-31T01:28:10 < upgrdman> why not go to another vendor?
2012-10-31T01:28:26 < dongs> Erlkoenig: do you see the problem finally.
2012-10-31T01:28:30 -!- bgamari [~bgamari@pool-108-8-241-91.spfdma.east.verizon.net] has joined ##stm32
2012-10-31T01:28:32 < Erlkoenig> yep
2012-10-31T01:28:38 < Erlkoenig> fixed the minimal example
2012-10-31T01:28:46 < Erlkoenig> not the main program, though
2012-10-31T01:28:51 < dongs> just toggle some other GPIO in while(1)
2012-10-31T01:28:55 < dongs> like GPIOE or somethign
2012-10-31T01:29:04 < dongs> or actually make it do SOMETHING USEFUL
2012-10-31T01:29:07 < Erlkoenig> i just used __enable_irq (); and __disable_irq ();
2012-10-31T01:29:11 < dongs> ...
2012-10-31T01:29:26 < Erlkoenig> the algorithms for "useful" aren't written yet, unfortunately
2012-10-31T01:29:49 < Erlkoenig> i thought i'd start with the more important and general "framework"
2012-10-31T01:30:34 < upgrdman> volatile int i = 0; while(i++ < 999999);
2012-10-31T01:31:04 < upgrdman> poke i in gdb to verify work being done
2012-10-31T01:34:18 < ds2> maybe I should put in a 1.8V option
2012-10-31T01:37:15 < emeb> ds2: got room for the little LDO?
2012-10-31T01:38:54 < ds2> this is irritating
2012-10-31T01:39:05 < ds2> all of a sudden NO ONE has stock
2012-10-31T01:39:06 < emeb> ?
2012-10-31T01:39:23 < emeb> someone out there probably had a new product and bought up everything.
2012-10-31T01:39:39 < ds2> I guess I can squeeze in an LDO
2012-10-31T01:39:44 < ds2> it'll just be painful batterylife wise
2012-10-31T01:39:59 < emeb> are the 1.8V parts the same size as the one you were using?
2012-10-31T01:40:18 < ds2> emeb: are these things standardized by JEDEC?
2012-10-31T01:40:29 < emeb> wouldn't count on it.
2012-10-31T01:40:52 < emeb> what P/N you have in there now?
2012-10-31T01:40:53 < ds2> I think they are
2012-10-31T01:41:00 < ds2> 0.75mm BGA
2012-10-31T01:41:02 < ds2> IS66WVE4M16BLL
2012-10-31T01:41:14 < ds2> it is the closest to the STEVAL part
2012-10-31T01:42:27 < dongs> ds2: verical has 480:)
2012-10-31T01:42:31 < dongs> bbl
2012-10-31T01:42:34 < dongs> time to drive 400km
2012-10-31T01:42:44 < emeb> o/
2012-10-31T01:43:20 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has joined ##stm32
2012-10-31T01:43:34 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Remote host closed the connection]
2012-10-31T01:43:38 < emeb> ds2: looks like the pinouts are the same.
2012-10-31T01:43:49 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T01:43:55 < ds2> dongs: verical?
2012-10-31T01:44:22 < ds2> emeb: If your part is what i looked at before, the main difference is spec for vddio
2012-10-31T01:44:41 < ds2> only good thing is I haven't pull the trigger on the fab :)
2012-10-31T01:45:07 < emeb> ds2: yeah
2012-10-31T01:45:23 < emeb> ds2: the Micron 1.8V part has a lower operating current.
2012-10-31T01:45:41 < emeb> IS part specs 30ma, MT specs 20ma
2012-10-31T01:45:58 < emeb> so MT + LDO = lower total power. Better for batteries.
2012-10-31T01:46:12 < ds2> unless I am bad at the LDO and burn it up in quiescent current
2012-10-31T01:46:35 < emeb> pick the right LDO and your Iddq will be in the nA.
2012-10-31T01:46:39 < ds2> wozers...
2012-10-31T01:46:51 < ds2> the micro part is more expensive
2012-10-31T01:47:29 < Erlkoenig> wow, i can't believe that was the problem
2012-10-31T01:47:33 < Erlkoenig> actually looks like it works niw
2012-10-31T01:48:44 < Erlkoenig> nobody was able to figure that out for a week now :S
2012-10-31T01:49:37 < ds2> emeb: what speed are you using? -70?
2012-10-31T01:49:58 < emeb> yeah
2012-10-31T01:50:37 < emeb> ds2: you've got more than one PSRAM on the board, right?
2012-10-31T01:50:40 < ds2> looks like you're right... it is a netwin. the LDOs I am using aren't too horrible
2012-10-31T01:50:43 < ds2> emeb: nope, just 1
2012-10-31T01:50:53 < emeb> Cool.
2012-10-31T01:51:20 < emeb> you could go with teeny QFN LDOs if needed.
2012-10-31T01:51:53 < ds2> LDOs I want a tad bit bigger. easier to bring up that way. and if I add enough TP's, it might as well be a SOT-23
2012-10-31T01:52:05 < emeb> true.
2012-10-31T01:52:20 < emeb> the little SOT-23-5 parts are easy to use.
2012-10-31T01:52:37 < ds2> paying a bit extra to go with a cap free part
2012-10-31T01:53:00 < emeb> *nod*
2012-10-31T01:53:08 < emeb> save some real-estate.
2012-10-31T01:54:13 < emeb> is the LDO Vdo low enough you can use 3.3V input?
2012-10-31T01:55:07 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 260 seconds]
2012-10-31T01:55:36 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:03:09 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Remote host closed the connection]
2012-10-31T02:03:23 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:03:25 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Remote host closed the connection]
2012-10-31T02:03:42 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:06:48 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-31T02:10:41 -!- SuicideFunky [~randy@vps-212-153-70-8.twenty-five.nl] has quit [Read error: Operation timed out]
2012-10-31T02:10:57 -!- SuicideFunky [~randy@vps-212-153-70-8.twenty-five.nl] has joined ##stm32
2012-10-31T02:17:42 < ds2> yes
2012-10-31T02:17:48 < ds2> trying to fit in now
2012-10-31T02:17:51 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-10-31T02:18:20 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-10-31T02:19:00 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:19:17 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:23:42 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T02:28:40 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-31T02:35:06 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T02:36:46 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
2012-10-31T02:38:31 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Remote host closed the connection]
2012-10-31T02:38:48 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:39:55 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
2012-10-31T02:43:30 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T02:46:17 -!- Erlkoenig [~erlkoenig@pptp-212-201-75-69.pptp.stw-bonn.de] has quit [Quit: Leaving.]
2012-10-31T03:02:56 < ds2> i hate these last minute bom changes :(
2012-10-31T03:07:04 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T03:45:48 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has joined ##stm32
2012-10-31T04:31:59 < ds2> dongs: do you guys do E-test?
2012-10-31T05:00:06 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-31T05:09:44 -!- quitte [~quitte@tplink.ghb.hs-furtwangen.de] has quit [Quit: Lost terminal]
2012-10-31T05:10:41 < dongs> yes,  100%
2012-10-31T05:16:52 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: Leaving.]
2012-10-31T05:19:37 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-31T05:31:42 < Bird|lappy> o/ emeb_mac
2012-10-31T05:32:00 < emeb_mac> hi Bird|lappy
2012-10-31T05:33:09 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has joined ##stm32
2012-10-31T05:37:43 < Bird|lappy> emeb_mac, in kicad: 1) for things like AA and 9V batteries that use a battery holder or battery clips, do you place a battery symbol in the schematic, or use say a connector symbol instead?
2012-10-31T05:40:16 -!- qyx_ [~qyx@krtko.org] has quit [Ping timeout: 246 seconds]
2012-10-31T05:40:23 -!- qyx_ [~qyx@krtko.org] has joined ##stm32
2012-10-31T05:45:15 -!- Bird|lappy [~Bird|ub3r@unaffiliated/htt-bird] has quit [Ping timeout: 276 seconds]
2012-10-31T05:45:17 < emeb_mac> Bird|lappy: it hasn't come up for me - I always have some other sort of power
2012-10-31T05:47:13 -!- kerwood [~Marshall@c-76-21-166-25.hsd1.md.comcast.net] has quit [Ping timeout: 245 seconds]
2012-10-31T06:00:35 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-10-31T06:01:13 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T06:01:20 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 268 seconds]
2012-10-31T06:02:14 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T06:30:00 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-31T06:30:53 -!- TheSeven [~quassel@rockbox/developer/TheSeven] has quit [Disconnected by services]
2012-10-31T06:31:00 -!- [7] [~quassel@rockbox/developer/TheSeven] has joined ##stm32
2012-10-31T06:51:08 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has joined ##stm32
2012-10-31T07:04:48 < baird> lol times infinity: http://www.businessinsider.com/disney-buys-lucasfilm-for-4-billion-2012-10
2012-10-31T07:05:42 < dongs> old
2012-10-31T07:06:11 < ds2> mouse ear shaped light sabres :)
2012-10-31T07:08:17 < baird> Star Wars Princesses..
2012-10-31T07:15:33 -!- Ademan [~dan@70.231.140.183] has quit [Quit: leaving]
2012-10-31T07:21:57 < baird> Nuts. I hate it when I dream up a new idea in Physics, only to wake up and discover someone's beaten me to it :P .. http://arxiv.org/abs/1001.5445v2 "Gravity from Quantum Information"
2012-10-31T07:33:59 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has quit []
2012-10-31T07:39:18 < baird> You've only had 20 fucking years to do it.
2012-10-31T07:46:36 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-31T07:46:36 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-31T07:47:42 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T07:48:07 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T08:00:42 -!- LeelooMinai [~leeloo@24-52-246-177.cable.teksavvy.com] has quit [Ping timeout: 264 seconds]
2012-10-31T08:20:13 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Ping timeout: 246 seconds]
2012-10-31T08:26:51 -!- LeelooMinai [~leeloo@198.84.185.212] has joined ##stm32
2012-10-31T08:27:36 < ds2> okay, sent.
2012-10-31T08:36:07 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 268 seconds]
2012-10-31T08:36:23 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T08:41:45 < emeb_mac> sent the gerbers?
2012-10-31T08:41:58 < emeb_mac> ds2 ^
2012-10-31T08:44:22 < ds2> yeah
2012-10-31T08:47:40 < emeb_mac> fun! what's the turn-around time?
2012-10-31T08:49:57 < ds2> emeb_mac: donno yet
2012-10-31T08:50:13 < ds2> trying to see if dongs' fab will work
2012-10-31T08:50:18 < emeb_mac> aha.
2012-10-31T08:50:41 < emeb_mac> he usually gets like 1wk turns.
2012-10-31T08:50:46 < ds2> I hope the timing for the Micron part is the same as my original part
2012-10-31T08:50:59 < ds2> so I should count on 2wk in reality
2012-10-31T08:51:08 < ds2> i have terrible luck with customs
2012-10-31T08:51:22 < emeb_mac> hold ups & slowdowns?
2012-10-31T08:51:40 < ds2> things seems to sit in the customs queue for a few days
2012-10-31T08:52:13 < emeb_mac> I've got a bunch of stuff coming - been waiting for a while.
2012-10-31T08:52:27 < ds2> who's doing your assembly?
2012-10-31T08:52:38 < emeb_mac> nothing like that.
2012-10-31T08:52:50 < ds2> oooh not pcbs
2012-10-31T08:52:51 < emeb_mac> just some stuff I ordered from DX, plus some stuff dongs sent
2012-10-31T08:52:59 < ds2> oooh DX...hehehe
2012-10-31T08:53:08 < emeb_mac> yep. cheep crap
2012-10-31T08:53:09 < ds2> just assume a 1-2 month turn around time for them
2012-10-31T08:53:15 < emeb_mac> pretty much
2012-10-31T08:53:42 < emeb_mac> you never know when it will show up.
2012-10-31T08:53:54 < ds2> yep
2012-10-31T08:53:56 < emeb_mac> it's like sending a present to your future self
2012-10-31T08:54:13 < ds2> there are folks showing up on amazon with prices similar to DX
2012-10-31T08:54:19 < ds2> and they seem to get better shipping times
2012-10-31T08:54:33 < emeb_mac> yeah. ebay has been pretty responsive lately too.
2012-10-31T08:54:47 < ds2> I don't deal with ebay. too much of a time sink
2012-10-31T08:55:10 < emeb_mac> this is just "buy it now". generally receive in < 2wks.
2012-10-31T08:55:25 < ds2> you got to watch shipping on those
2012-10-31T08:55:32 < ds2> and any issues, it is a pain to dispute it
2012-10-31T08:55:44 < ds2> amazon is less pain
2012-10-31T08:56:10 < emeb_mac> true. for ebay I keep the expenses low & just forget it if things go south.
2012-10-31T08:56:30 < emeb_mac> don't spend more than $20 at a shot.
2012-10-31T08:56:46 < ds2> I guess... still rather stick with amazon
2012-10-31T08:56:53 < emeb_mac> makes sense.
2012-10-31T08:57:05 < emeb_mac> got some good soldering stuff via amazon a few weeks back
2012-10-31T08:57:18 < ds2> you have a decent local place for soldering stuff
2012-10-31T08:57:39 < emeb_mac> I assume you mean Circuit Specialists?
2012-10-31T08:57:45 < ds2> yeah
2012-10-31T08:58:06 < emeb_mac> I did check them for this, but it was too specialized and they didn't have it.
2012-10-31T08:58:17 < ds2> oooh
2012-10-31T08:58:27 < emeb_mac> needed a 3mm "hoof" tip for my Hakko 936.
2012-10-31T08:58:45 < emeb_mac> for doing drag soldering on lqfps
2012-10-31T08:58:49 < ds2> ahhhh
2012-10-31T08:59:42 < ds2> might be redundant but has anyone looked at uboot on the stm32?
2012-10-31T09:00:21 < emeb_mac> not that I know of
2012-10-31T09:00:54 < ds2> guess I get to have fun with that
2012-10-31T09:01:00 < emeb_mac> How would that work? u-boot in flash loading an OS kernel into external SRAM?
2012-10-31T09:01:17 < ds2> u-boot in internal STM32 flash loading a OS kernel off of SD card
2012-10-31T09:01:27 < emeb_mac> right
2012-10-31T09:01:29 < ds2> then rootfs off the SD card
2012-10-31T09:01:56 < emeb_mac> Don't think anyone here uses any OS large enough to require that.
2012-10-31T09:02:04 < emeb_mac> nor external RAM.
2012-10-31T09:02:18 < emeb_mac> You're blazing the trail :)
2012-10-31T09:02:25 < ds2> great...
2012-10-31T09:02:43 < ds2> good thing is one of the guys that worked on uCLinux on the cortex-m3 will be in town giving a talk thu ;)
2012-10-31T09:03:04 < emeb_mac> pick his brain
2012-10-31T09:03:25 < ds2> yep
2012-10-31T09:03:40 < ds2> but he's talking about it on the fpga platform
2012-10-31T09:03:53 < emeb_mac> oh yay.
2012-10-31T09:04:09  * emeb_mac is not convinced that running processors in FPGAs is a great idea.
2012-10-31T09:04:15 < ds2> why not?
2012-10-31T09:04:17 < ds2> this is a hard core
2012-10-31T09:04:30 < emeb_mac> that's a little better.
2012-10-31T09:04:38 -!- cjbaird [~cjb@ppp121-44-1-182.lns20.syd6.internode.on.net] has joined ##stm32
2012-10-31T09:04:55 < ds2> you can only get M1's as a softcore, AFAIK
2012-10-31T09:05:10 < emeb_mac> but still - an FPGA + hardcore is almost always more expensive than a commodity SoC + external FPGA.
2012-10-31T09:05:16 < ds2> softcores are handy for quick and dirty statemachines
2012-10-31T09:05:20 < emeb_mac> and usually lower performance
2012-10-31T09:05:22 -!- baird [~cjb@ppp121-44-164-147.lns20.syd7.internode.on.net] has quit [Ping timeout: 246 seconds]
2012-10-31T09:05:41 < ds2> spacewise, the hardcore option wins
2012-10-31T09:05:56 < emeb_mac> only advantage that you get from putting the FPGA & processor on the same die is that the path between the CPU and FPGA can be very wide/fast.
2012-10-31T09:06:01 < ds2> wonder if the PSoCs bring out the memory bus
2012-10-31T09:06:18 < ds2> yes and it can potentially cut down on EMI
2012-10-31T09:06:53 < emeb_mac> there's that.
2012-10-31T09:07:14 < ds2> the other thing is what are you looking to get out of the FPGA
2012-10-31T09:08:05 < emeb_mac> I guess.
2012-10-31T09:08:26 < emeb_mac> Main thing I notice is that the price premium for these FPGA SoCs is pretty steep.
2012-10-31T09:09:03 < emeb_mac> Low-end Xilinx Zinq parts are ~$200 in 1000qty
2012-10-31T09:09:13 < ds2> I can see hard core can be good for combining different low volume projects to bring cogs down
2012-10-31T09:09:32 < emeb_mac> compare that to an OMAP + midrange FPGA for about 1/5th that.
2012-10-31T09:09:49 < ds2> Oooh you mean the big ones
2012-10-31T09:10:02 < ds2> I am thinking small... M3 hardcore would be big for what I have in mind
2012-10-31T09:10:10 < emeb_mac> Yeah - low volume with specialized requirements and cost no barrier.
2012-10-31T09:10:24 < emeb_mac> Who's got M3?
2012-10-31T09:10:35 < ds2> Microsemi
2012-10-31T09:10:38 < ds2> Cypress
2012-10-31T09:10:47 < ds2> didn't look at the others
2012-10-31T09:10:57 < ds2> both offer analog stuff on the chip
2012-10-31T09:10:59 < emeb_mac> Cypress really isn't an FPGA (if you're talking PSoC)
2012-10-31T09:11:07 < zyp> emeb_mac, how is the price of artix-7 compared to zynq?
2012-10-31T09:11:12 < ds2> yes, fair enough
2012-10-31T09:11:29 < ds2> cypress is really more of bunch of crap + cpld to link them
2012-10-31T09:11:36 < ds2> but it is on the order of my interests
2012-10-31T09:11:41 < emeb_mac> yeah
2012-10-31T09:11:46 < zyp> lattice have some cortex-m3 stuff
2012-10-31T09:12:02 < zyp> and horrible tools from what I've heard
2012-10-31T09:12:06 < emeb_mac> zyp: haven't looked at artix
2012-10-31T09:12:25 < zyp> it would be the most fair comparison, considering that zynq is based on artix fabric
2012-10-31T09:12:53 < ds2> don't they all license cadence tools ?
2012-10-31T09:14:52 < emeb_mac> beats me. I use Xilinx ISE and Altera Quartus.
2012-10-31T09:15:02 < emeb_mac> no Cadence stuff there AFAIK.
2012-10-31T09:15:19 < cjbaird> Sucks that getting this Stellaris Launchpad up and going (on free hippystallmanapproved linux) is dragging on.. It really is a cute little thing. :)
2012-10-31T09:16:39 < zyp> hmm, I still haven't seen any sign of mine
2012-10-31T09:17:13 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-31T09:17:21 < emeb_mac> haven't heard any word on mine.
2012-10-31T09:18:03 < cjbaird> And.. it's exactly half the size of the F3Disco.
2012-10-31T09:18:22 < emeb_mac> only 1 artix part on the Avnet site - $361/qty1000
2012-10-31T09:19:00 < cjbaird> Hmm, the F3Disco weighs 36 grammes, the Stell.LP is 20g
2012-10-31T09:19:24 < cjbaird> The Freescale KL25: 17g
2012-10-31T09:19:51 < zyp> sounds pretty irrelevant
2012-10-31T09:20:18 < cjbaird> There was a Trendsetter a few days ago carrying on about using the F3Disco as a Quadcopter controller.
2012-10-31T09:20:34 < zyp> the stellaris launchpad only have 40 IO-pins
2012-10-31T09:20:44 < zyp> no wonder they can make the board smaller
2012-10-31T09:21:47 < zyp> the stm32f3discovery have 100
2012-10-31T09:27:10 < emeb_mac> l8rs
2012-10-31T09:27:14 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-31T09:34:39 -!- Laurenceb_ [~laurence@vlsi1.eee.nottingham.ac.uk] has quit [Ping timeout: 260 seconds]
2012-10-31T09:46:35 -!- Laurenceb_ [~laurence@vlsi1.eee.nottingham.ac.uk] has joined ##stm32
2012-10-31T10:07:53 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Quit: This computer has gone to sleep]
2012-10-31T10:09:18 -!- |akaWolf| [~akaWolf@188.134.9.161] has joined ##stm32
2012-10-31T10:10:36 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds]
2012-10-31T10:37:52 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-10-31T10:38:22 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-10-31T10:40:08 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T10:41:00 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-31T10:44:39 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T10:50:46 < cjbaird> Groan, kiddie trick-or-treaters.. I gave them some troll-brand red bean mochi.
2012-10-31T10:52:43 < dongs> ds2: stuff received, will let you know tomorrow morning about manufactureability+cost
2012-10-31T10:59:17 < zyp> cjbaird, ew :p
2012-10-31T11:03:47 < cjbaird> zyp: yes, stupid idea trying to market Halloween in Australia.
2012-10-31T11:04:33 < cjbaird> The thing that /makes/ Halloween in the States is the adult parties... which no-one has the clue to try pushing here.
2012-10-31T11:04:48 -!- AaronM [~Aaron@pdpc/supporter/student/aaronm] has quit [Quit: Quit]
2012-10-31T11:09:02 < zyp> I was referring to the red bean mochi
2012-10-31T11:09:34 < cjbaird> I did say troll-brand.. (plus it was like the only confectionary in the house still in an unopened package)
2012-10-31T11:11:01 < zyp> troll-brand doesn't help, the thing is that I'm not a fan of neither red beans nor mochi :p
2012-10-31T11:11:02 < cjbaird> The 15yo older sister in the Little Black Dress made up for the interruption. :P :)
2012-10-31T11:11:41 < cjbaird> OH SHIT ZLOG
2012-10-31T11:12:05 < zyp> ha
2012-10-31T11:14:20 < cjbaird> Hopefully the brats won't be back next year.
2012-10-31T11:18:38 -!- claude is now known as claude_work
2012-10-31T11:19:38 < dongs> ...
2012-10-31T11:25:46 < cjbaird> dongs: those without an F3 Discovery board are not allowed to have an opinion.
2012-10-31T11:27:21 < cjbaird> Neat, got lm4flash working with the Stellaris.
2012-10-31T11:35:10 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has quit [Ping timeout: 246 seconds]
2012-10-31T11:38:04 < dongs> cjbaird: good job packeting off zlob
2012-10-31T11:45:01 < dongs> ds2: looks like you get free enig
2012-10-31T11:45:32 < dongs> got one of these free: http://www.camonetec.com/
2012-10-31T11:45:38 < dongs> shit looks bettar than godong
2012-10-31T11:50:03 -!- zlog [~zlog@ip24-255-189-172.ks.ks.cox.net] has joined ##stm32
2012-10-31T11:51:14 -!- Laurenceb__ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-31T12:17:59 -!- Laurenceb__ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has quit [Ping timeout: 260 seconds]
2012-10-31T12:20:09 < claude_work> pro dongs , of course this is better than godongs , this cam is made by germans! think of it as the oktoberfest of all action cams
2012-10-31T12:21:26 < dongs> i was pretty sure its some chink shit
2012-10-31T12:21:29 < dongs> where do yo usee its german?
2012-10-31T12:22:44 < claude_work> "copyright © 2012 by CamOneTec a Division of ACME the game company GmbH Home" GmbH == Ltd. in german
2012-10-31T12:22:53 < dongs> ohh,  neat
2012-10-31T12:23:34 < claude_work> but yeah probably chinese , just a german company labeling them
2012-10-31T12:30:52 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has joined ##stm32
2012-10-31T12:30:52 -!- jon1012 [~jon@pha75-14-88-162-134-157.fbx.proxad.net] has quit [Changing host]
2012-10-31T12:30:52 -!- jon1012 [~jon@foresight/developer/jon1012] has joined ##stm32
2012-10-31T12:45:00 < dongs> http://img.dxcdn.com/productimages/sku_120604_1.jpg haha nice font
2012-10-31T12:47:20 < dongs> http://www.st.com/internet/analog/product/253162.jsp "in-system ceramic resonator". Flexible interfaces: CAN, USART, SPI and I2C serial interfaces; full-speed USB 2.0
2012-10-31T12:47:29 < dongs> trollen?
2012-10-31T13:02:41 -!- daku is now known as DaKu
2012-10-31T13:21:34 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-31T13:23:07 -!- DaKu is now known as daku
2012-10-31T13:23:19 -!- daku is now known as DaKu
2012-10-31T13:24:36 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 248 seconds]
2012-10-31T13:24:53 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T13:24:54 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-31T13:25:08 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T13:27:54 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-31T13:28:10 < zyp> they probably used one that is accurate enough for FS
2012-10-31T13:29:52 < dongs> its the one they recommend in crystals appnote
2012-10-31T13:29:59 < dongs> hwen me and emeb looked, it wasnt
2012-10-31T13:31:18 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Ping timeout: 255 seconds]
2012-10-31T13:33:01 < zyp> oh
2012-10-31T13:57:50 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 252 seconds]
2012-10-31T13:58:17 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T14:29:25 -!- scrts [~quassel@2001:778:200:23a0:202:b3ff:febd:5f93] has joined ##stm32
2012-10-31T14:29:25 -!- scrts [~quassel@2001:778:200:23a0:202:b3ff:febd:5f93] has quit [Changing host]
2012-10-31T14:29:25 -!- scrts [~quassel@unaffiliated/scrts] has joined ##stm32
2012-10-31T14:29:48 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has joined ##stm32
2012-10-31T14:33:26 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-31T14:34:15 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has joined ##stm32
2012-10-31T14:34:16 -!- Steffanx [~Steffanx@ip4da726ae.direct-adsl.nl] has quit [Changing host]
2012-10-31T14:34:16 -!- Steffanx [~Steffanx@unaffiliated/steffanx] has joined ##stm32
2012-10-31T14:34:19 -!- mode/##stm32 [+v Steffanx] by ChanServ
2012-10-31T14:34:29 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T14:34:37 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Read error: Connection reset by peer]
2012-10-31T14:37:39 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T14:38:06 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Remote host closed the connection]
2012-10-31T14:39:16 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T14:42:08 -!- prog [~prog@bornbrilliant.org] has quit [Read error: Operation timed out]
2012-10-31T14:44:42 -!- prog [prog@94.100.30.150] has joined ##stm32
2012-10-31T15:08:07 -!- ReggieUK [~ReggieUK@5aca5630.bb.sky.com] has joined ##stm32
2012-10-31T15:23:06 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-31T15:23:20 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T15:28:25 -!- |akaWolf| [~akaWolf@188.134.9.161] has quit [Quit: my exit]
2012-10-31T16:14:12 < Laurenceb_> http://www.b3tards.com/u/a1e65f7200ce1e8fdc59/hijabitat.jpg
2012-10-31T16:15:16 < dongs> am i looking for a picture of hidden dong
2012-10-31T16:20:54 <+Steffanx> Look down dongs
2012-10-31T16:28:50 < dongs> thats not hidden
2012-10-31T16:37:11 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has quit [Quit: ZNC - http://znc.in]
2012-10-31T16:38:21 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has joined ##stm32
2012-10-31T16:38:46 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-31T16:43:05 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has quit [Client Quit]
2012-10-31T16:45:10 -!- txf [~txf@bl7-150-117.dsl.telepac.pt] has joined ##stm32
2012-10-31T17:03:36 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has joined ##stm32
2012-10-31T17:04:05 -!- gnomad [~gnomad@c-71-203-29-67.hsd1.fl.comcast.net] has quit [Client Quit]
2012-10-31T17:17:18 -!- emeb_mac [~ericb@ip72-201-79-123.ph.ph.cox.net] has quit [Quit: emeb_mac]
2012-10-31T17:36:09 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
2012-10-31T17:36:37 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-31T17:40:40 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 244 seconds]
2012-10-31T17:41:09 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T17:42:45 -!- Thorn [~Thorn@unaffiliated/thorn] has joined ##stm32
2012-10-31T17:57:49 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has quit [Quit: leaving]
2012-10-31T17:59:46 -!- mrnuke [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has quit [Quit: Konversation terminated!]
2012-10-31T17:59:49 -!- mrnuke_ [~mrnuke@c-98-196-107-247.hsd1.tx.comcast.net] has joined ##stm32
2012-10-31T18:06:09 < Laurenceb_> why am i finding caps with 	Lifetime:	2000h
2012-10-31T18:06:14 < Laurenceb_> thats  <100days
2012-10-31T18:06:37 < Laurenceb_> how do they specify lifetime
2012-10-31T18:07:09 < karlp> linky?
2012-10-31T18:07:58 < Laurenceb_> http://uk.rs-online.com/web/p/aluminium-capacitors/7296251/
2012-10-31T18:08:39 -!- barthess [~barthess@178.154.40.227] has joined ##stm32
2012-10-31T18:10:58 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Quit: leaving]
2012-10-31T18:11:02 -!- DLPeterson_ [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Quit: leaving]
2012-10-31T18:11:18 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has joined ##stm32
2012-10-31T18:15:11 -!- claude_work is now known as claude
2012-10-31T18:21:25 < karlp> 2000h at 105C Laurenceb_
2012-10-31T18:23:33 -!- Rickta59 [~kimballr@cpe-174-106-144-066.ec.res.rr.com] has joined ##stm32
2012-10-31T18:32:30 -!- emeb [~ericb@ip72-201-79-123.ph.ph.cox.net] has joined ##stm32
2012-10-31T18:38:59 < Laurenceb_> oh
2012-10-31T18:39:00 < Laurenceb_> i see
2012-10-31T18:39:39 < dongs> Laurenceb_: tarduino due software is out
2012-10-31T18:40:29 < emeb> so is it even larger now that it has both avr and arm compilers included?
2012-10-31T18:40:55 < dongs> yes
2012-10-31T18:40:57 < dongs> 160megs
2012-10-31T18:41:04 < emeb> heh
2012-10-31T18:41:07 < dongs> also: arm-gcc is sourcery 2010q1 edition
2012-10-31T18:41:08 < BrainDamage> meh, it's not a big deal compared to say an fpga ide
2012-10-31T18:41:19 < dongs> BrainDamage: at least FPGA IDE does something useful
2012-10-31T18:41:19 < BrainDamage> where the ide itself without hw info is 1-2GB
2012-10-31T18:41:19 < emeb> that
2012-10-31T18:41:32 < BrainDamage> heh
2012-10-31T18:41:49  * emeb tends not to use the IDE for FPGA stuff
2012-10-31T18:41:52 < BrainDamage> I guess it's sorta offensive to call "IDE" their text editor
2012-10-31T18:42:13 < emeb> invoking the cmd line tools from a makefile works so much more nicely.
2012-10-31T18:42:25 < dongs> http://i.imgur.com/sxCEd.png < arduino due IDE
2012-10-31T18:42:45 < emeb> lol
2012-10-31T18:43:52 < emeb> wonder how long until some hacks arm arduino to target other kinds of arm.
2012-10-31T18:44:07 < emeb> in 3...2...1...
2012-10-31T18:44:08 < dongs> why the fuck would anyone want to?
2012-10-31T18:44:30 < dongs> arm tarduino is just worthless
2012-10-31T18:44:35 < emeb> dongs: you're forgetting - this is arduino community. they don't need a reason.
2012-10-31T18:44:48 < dongs> everything is designed to be called in a loop()
2012-10-31T18:45:02 < BrainDamage> I guess it's like cancer
2012-10-31T18:45:03 < dongs> shit like ADC sampling just does start_conv(); while(1); read_conv()
2012-10-31T18:45:06 < BrainDamage> it wants to self spread
2012-10-31T18:45:39 < dongs> 2010 sourcery G++ is fucked up though
2012-10-31T18:45:42 < dongs> thats older than shit
2012-10-31T18:45:52 < dongs> i guess no surprise, since they ship with some 10 years old avr-gcc as well...
2012-10-31T18:45:52 < emeb> yep - getting a bit crusty.
2012-10-31T18:46:20 < BrainDamage> it's "stable" in the debian sense
2012-10-31T18:46:23 < BrainDamage> as in useless
2012-10-31T18:46:26 < dongs> their driver infs are still not signed either...
2012-10-31T18:46:48 < emeb> stable as in "so old it's no longer changing because it's no longer maintained" ?
2012-10-31T18:47:24 < BrainDamage> correct
2012-10-31T18:48:18 < emeb> Oh well - not too turned on by the Atmel Mx parts anyway, and the Due is kinda $$$
2012-10-31T18:48:29 < dongs> and its 3.3V only
2012-10-31T18:48:35 < dongs> a deal breaker for 99% of arduino users.
2012-10-31T18:48:48 < emeb> not even 5V tolerant - I predict lots of blown I/O pins.
2012-10-31T18:48:48 < BrainDamage> not even 5V tollerant?
2012-10-31T18:48:52 < BrainDamage> ouch
2012-10-31T18:49:11 < dongs> apparently not
2012-10-31T18:49:30 < karlp> such an odd decision.
2012-10-31T18:49:52 < dongs> they shoulda used silabs M3 parts.
2012-10-31T18:50:02 < dongs> those fucking things re not only 5V tolerant, they have like 5V drive capable pins.
2012-10-31T18:50:12 < dongs> (while shit runs off 3.3V
2012-10-31T18:50:20 < emeb> hard-core.
2012-10-31T18:50:28 < dongs> plus you can reroute the i/o pins matrix any way you want
2012-10-31T18:50:35 < BrainDamage> I'd imagine arduino "devs" are ass-deep into atmel's dick anyway
2012-10-31T18:50:43 < dongs> like 3/4 of the chip pins can be arbitrarily assigned to anything.
2012-10-31T18:50:45 < BrainDamage> so they are rather bound in producer choice
2012-10-31T18:50:48 < dongs> of course
2012-10-31T18:50:55 < emeb> that i/o matrix is sweet. newer PICs do that and it's really nice.
2012-10-31T18:51:40 < BrainDamage> while that somehow made sense since atmel was the only one that decently supported gcc which was one of their requirements, for arm it's completely useless
2012-10-31T18:51:52 < jpa-> i think some arduino devs are employed at atmel
2012-10-31T18:52:21 < emeb> sword of damocles :)
2012-10-31T18:52:52 < jpa-> IMO it is surprising though that they went with ARM and not AVR32
2012-10-31T18:52:56 < zyp> dongs, I hear that the io matrix stuff they are using have some weird limitations in practice
2012-10-31T18:53:47 < zyp> jpa-, why? avr32 has as little in common with avr as arm
2012-10-31T18:53:53 < dongs> zyp, it looks pretty simple on paper,
2012-10-31T18:54:04 < jpa-> zyp: sure, ARM is the smarter choice.. but this is arduino :)
2012-10-31T18:54:45 < dongs> pretty much e very due newspost i see those chipkit uno faggots spamming it up wiht PIC32
2012-10-31T18:54:50 < dongs> "omg its 5v t olerant
2012-10-31T18:54:58 < dongs> like seriously pic32?
2012-10-31T18:55:10 < dongs> if youre gonna pick a dead platform, at least pick something less dead
2012-10-31T18:55:17 < emeb> *cough*MIPS*cough*
2012-10-31T18:55:30  * karlp peers at the mips on his desk.
2012-10-31T18:55:44  * emeb knows the guy who was lead designer on the PIC32.
2012-10-31T18:55:47 < karlp> though that one's labelled atheros, not pic.
2012-10-31T18:56:08 < jpa-> yeah, mips is not that dead
2012-10-31T18:56:15 < Laurenceb_> http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32&cmpt=q
2012-10-31T18:56:24 < karlp> when do they get out their new mcu targetted mips?
2012-10-31T18:56:29 < karlp> with the strange name again?
2012-10-31T18:56:53 < BrainDamage> fuck yeah, google charts make failfox to crash
2012-10-31T18:57:24 < karlp> so the russians like avr32, the indians like pic32, and the chinese like stm32
2012-10-31T18:57:26 < emeb> Laurenceb_: cool chart
2012-10-31T18:57:31 < jpa-> Laurenceb_: http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32%2C%20arduino&cmpt=q
2012-10-31T18:57:44 < emeb> surprising that STM32 is mostly CN.
2012-10-31T18:57:45 < dongs> haha
2012-10-31T18:57:45 < karlp> You have reached your quota limit. Please try again later.
2012-10-31T18:58:24 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-31T18:58:42 < dongs> http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32%2C%20lunix&cmpt=q
2012-10-31T18:59:00 < Laurenceb_> lollll
2012-10-31T19:00:03 < Laurenceb_> http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32%2C%20risc%20os&cmpt=q
2012-10-31T19:00:03 < dongs> what other worthless 32bit mcus are there?
2012-10-31T19:00:05 < jpa-> karlp: log in to get by that quota limit
2012-10-31T19:00:25 < dongs> riscos? deader than *bsd
2012-10-31T19:00:38 < karlp> huh, took the .uk out of the url and it worked
2012-10-31T19:01:30 < dongs> http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32%2C%20parallax%20propeller&cmpt=q ahh yes
2012-10-31T19:02:30 < Laurenceb_> wut
2012-10-31T19:02:35 < Laurenceb_> people actually use that
2012-10-31T19:02:39 < Laurenceb_> also xmos
2012-10-31T19:02:50 < dongs> propeller? negrplz.
2012-10-31T19:03:02 < dongs> actually xmos is about as dead as prop
2012-10-31T19:03:05 < dongs> according to google
2012-10-31T19:03:10 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T19:03:13 < dongs> acutally MORE volume for xmos... by like 0.1%
2012-10-31T19:03:23 < Laurenceb_> xmos is epic fail
2012-10-31T19:03:25 < jpa-> http://www.google.co.uk/trends/explore#q=raspberry%2C%20strawberry%2C%20blueberry&cmpt=q   putting end to seasonal treats, raspberry is the official 2012 berry!
2012-10-31T19:03:31 < Laurenceb_> gpio only
2012-10-31T19:05:49 < emeb> interesting - raspberry pi has parity w/ arduino: http://www.google.co.uk/trends/explore#q=raspberry%20pi%2C%20beagleboard%2C%20beaglebone%2C%20arduino&cmpt=q
2012-10-31T19:06:37 < dongs> http://www.google.co.uk/trends/explore#q=raspberry%20pi%2C%20beagleboard%2C%20beaglebone%2C%20arduino%2C%20gay%20porn&cmpt=q
2012-10-31T19:06:49 < emeb> ROFL
2012-10-31T19:07:16 < dongs> thats probably a Laurenceb_ -worthy statistic
2012-10-31T19:07:19 < Laurenceb_> haha
2012-10-31T19:07:35 < Laurenceb_> papua new guinea, usa and uk are gayest
2012-10-31T19:08:36 < Laurenceb_> http://www.google.co.uk/trends/explore#q=%22free+porn+gay%22&cmpt=q
2012-10-31T19:08:42 < Laurenceb_> recession is biting in greece
2012-10-31T19:08:57 < jpa-> Laurenceb_: hmm, so what exactly is wrong with xmos?
2012-10-31T19:09:08 < jpa-> it seems quite cool IMO
2012-10-31T19:09:14 < Laurenceb_> theres only gpio
2012-10-31T19:09:25 < dongs> kinda like prop
2012-10-31T19:09:27 < Laurenceb_> so you have to have tarduino style looping code
2012-10-31T19:09:44 < dongs> except you dont need to program it in spin.gif language.
2012-10-31T19:09:57 < Laurenceb_> meatspin
2012-10-31T19:09:57  * emeb remembers the precursor to XMOS - Inmos Transputers.
2012-10-31T19:10:02 < karlp> no, trinidate and tobago were the gayest.
2012-10-31T19:10:07 < emeb> I've got a rail of those sitting around.
2012-10-31T19:10:08 < karlp> and fiji
2012-10-31T19:11:16 < jpa-> Laurenceb_: i thought they contain programmable logic elements to implement peripherals
2012-10-31T19:11:49 < dongs> http://www.google.co.uk/trends/explore#q=stm32%2C%20pic32%2C%20avr32%2C%20efm32%2C%20superH&cmpt=q
2012-10-31T19:11:52 < dongs> wghat the fuck
2012-10-31T19:11:57 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-ossvnzjdqfmjsxdj] has quit [Ping timeout: 244 seconds]
2012-10-31T19:11:59 < dongs> superH: japan and ... slovakia
2012-10-31T19:12:16 < dongs> the 1st one is obvious since they wank over that shit but slovakia?
2012-10-31T19:12:29 < jpa-> though.. no ADC sucks for many applications
2012-10-31T19:13:09 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-hpeugwzwpsgzpbxo] has joined ##stm32
2012-10-31T19:13:29 -!- BjoernC [~BjoernC@2001:638:602:1183:223:8bff:fe86:1627] has quit [Quit: Konversation terminated!]
2012-10-31T19:13:55 < dongs> https://youtube.com/watch?v=H7arGk8GDtA retweeting
2012-10-31T19:15:00 < emeb> kitty is rightly suspicious.
2012-10-31T19:16:21 < cjbaird> Completed Doom2 without cheating \o/ (..after 20 years)
2012-10-31T19:23:29 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-hpeugwzwpsgzpbxo] has quit [Ping timeout: 244 seconds]
2012-10-31T19:24:28 -!- Oldboy [myircshell@gateway/shell/sh3lls.net/x-uoozamclcugvzcea] has joined ##stm32
2012-10-31T19:38:03 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-31T19:41:22 -!- cjbaird [~cjb@ppp121-44-1-182.lns20.syd6.internode.on.net] has quit [Quit: Konversation terminated!]
2012-10-31T19:47:47 < prog> !life
2012-10-31T19:48:05 < qyx_> what is it?
2012-10-31T19:50:50 < emeb> conway in inverse video
2012-10-31T20:09:25 -!- DLPeterson_ [~hazelnuss@c-71-198-192-33.hsd1.ca.comcast.net] has joined ##stm32
2012-10-31T20:10:34 -!- DLPeterson [~hazelnuss@107-197-105-35.lightspeed.sntcca.sbcglobal.net] has quit [Ping timeout: 240 seconds]
2012-10-31T20:17:27 -!- barthess [~barthess@178.154.40.227] has quit [Quit: Leaving.]
2012-10-31T20:31:22 < ds2> hmmmm
2012-10-31T20:33:13 -!- BrainDamage1 [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T20:33:14 -!- BrainDamage [~BrainDama@62.10.6.74] has quit [Disconnected by services]
2012-10-31T20:40:06 -!- jon1012 [~jon@foresight/developer/jon1012] has quit [Quit: Quitte]
2012-10-31T20:42:32 -!- BrainDamage1 [~BrainDama@62.10.6.74] has quit [Quit: Leaving.]
2012-10-31T20:42:41 -!- BrainDamage [~BrainDama@62.10.6.74] has joined ##stm32
2012-10-31T20:42:41 < emeb> morning ds2
2012-10-31T20:45:36 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Ping timeout: 240 seconds]
2012-10-31T20:46:04 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has joined ##stm32
2012-10-31T20:47:47 -!- barthess [~barthess@178.154.40.227] has joined ##stm32
2012-10-31T20:52:17 < ds2> morning
2012-10-31T20:56:13 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-31T21:25:28 -!- BjoernC [~BjoernC@b071.apm.etc.tu-bs.de] has joined ##stm32
2012-10-31T21:34:37 -!- barthess [~barthess@178.154.40.227] has quit [Quit: Leaving.]
2012-10-31T21:35:10 -!- barthess [~barthess@178.154.40.227] has joined ##stm32
2012-10-31T21:39:12 -!- dekar [~dekar@212.255.245.205] has joined ##stm32
2012-10-31T21:39:15 -!- mode/##stm32 [+v dekar] by ChanServ
2012-10-31T21:47:00 <+dekar> new caps and everything is working again
2012-10-31T21:47:57 <+dekar> also my mainboard now powers up with the press of up button, I used to have to hold down the power button for 2-3s until the caps on the mainboard charged up sufficiently to start it up
2012-10-31T21:51:22 <+Steffanx> party!
2012-10-31T21:51:51 < karlp> what did you change?
2012-10-31T21:51:55 <+dekar> all caps
2012-10-31T21:52:00 < karlp> fromw hat to what?
2012-10-31T21:52:09 <+dekar> from those with bulges
2012-10-31T21:52:13 -!- akaWolf [~akaWolf@unaffiliated/akawolf] has quit [Quit: my exit]
2012-10-31T21:52:18 <+dekar> to some other chinese brand
2012-10-31T21:52:21 < karlp> ah.
2012-10-31T21:52:49 <+dekar> my 12V rail was down to 10.2V and one drive kept losing the sata link
2012-10-31T21:53:19 <+dekar> some high efficiency 75W HTPC PSU
2012-10-31T22:03:39 -!- grummund [~user@unaffiliated/grummund] has quit [Ping timeout: 260 seconds]
2012-10-31T22:09:29 -!- barthess [~barthess@178.154.40.227] has quit [Ping timeout: 260 seconds]
2012-10-31T22:10:50 -!- Laurenceb__ [~Laurence@host86-177-213-79.range86-177.btcentralplus.com] has joined ##stm32
2012-10-31T22:14:00 -!- Laurenceb__ is now known as Laurenceb
2012-10-31T22:14:31 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has quit [Remote host closed the connection]
2012-10-31T22:14:40 -!- TitanMKD [Titan@chl26-1-88-183-104-56.fbx.proxad.net] has joined ##stm32
2012-10-31T22:17:19 -!- barthess [~barthess@164.177.214.243] has joined ##stm32
2012-10-31T22:22:29 < Laurenceb> http://i.imgur.com/GG0Cf.jpg
2012-10-31T22:24:37 <+Steffanx> Blaxter ratio ++
2012-10-31T22:24:40 <+Steffanx> :P
2012-10-31T22:26:01 -!- gurgalof [~gurgalof@c80-216-122-131.bredband.comhem.se] has quit [Quit: Leaving]
2012-10-31T22:26:28 -!- barthess [~barthess@164.177.214.243] has quit [Quit: Leaving.]
2012-10-31T22:27:59 -!- phantoneD [~destroy@a95-92-89-24.cpe.netcabo.pt] has quit []
2012-10-31T22:28:22 -!- phantoxeD [~destroy@a95-92-89-24.cpe.netcabo.pt] has joined ##stm32
2012-10-31T22:53:22 -!- Count_Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has quit [Read error: Connection reset by peer]
2012-10-31T22:53:43 -!- Niedar [~nnscript@ip68-231-248-182.hr.hr.cox.net] has joined ##stm32
2012-10-31T23:02:38 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 245 seconds]
2012-10-31T23:19:28 -!- mrnuke_ is now known as mrnuke
2012-10-31T23:36:24 -!- Vutral [ss@mirbsd/special/Vutral] has joined ##stm32
2012-10-31T23:37:30 -!- grummund [~user@unaffiliated/grummund] has joined ##stm32
2012-10-31T23:55:14 -!- Vutral [ss@mirbsd/special/Vutral] has quit [Ping timeout: 255 seconds]
--- Log closed Thu Nov 01 00:00:54 2012