Back
[02:56:33] <CapnKernel> OndraSter: Hi
[02:56:46] <OndraSter> hi
[02:56:47] <tobbor> Hello OndraSter
[02:56:50] <OndraSter> hi
[03:06:37] <da78> Moin.
[03:07:08] <OndraSter> ey
[03:07:21] <OndraSter> http://a4.sphotos.ak.fbcdn.net/hphotos-ak-snc7/s720x720/398874_353499291376805_103308439729226_986932_753542747_n.jpg good morning :P
[04:09:18] <DarkSector> random chicks' photo
[04:09:37] <DarkSector> anyways, I am trying to connect ISPMKII with Mac OS X and it doesn't seem to light up
[04:09:52] <DarkSector> anyone who's used the ISP MKII with the mac?
[05:39:45] <buhman> is there there something similar to, for example, usbtinyisp, only that does jtag? I found the aquaticus jtag; but that one seems to require I use rs232
[05:41:19] <buhman> OndraSter: nasty; dyed hair
[05:41:42] <OndraSter> why not buy full AVR Dragon?
[05:41:44] <OndraSter> it is not that expensive
[05:41:55] <buhman> well, I'd rather make whatever it is I use
[05:56:14] <Kevin`> buhman: RikusW reverse-engineered most of the jtag stuff and has custom firmware for it. key part is reverse-engineered though, it's not normally possible to use a generic jtag adapter with avr stuff since the specifics of the protocol are secret
[06:00:13] <buhman> why would atmel want to do that?
[06:00:30] <buhman> wouldn't it be a good thing if everyone knew how their jtag protocol worked?
[06:02:27] <WormFood> buhman, not good if you want to be the only one selling jtag programmers
[06:07:02] <buhman> does RikusW also have custom a jtag hardware design?
[06:15:49] <abcminiuser_> exDM69, what name do you want in the changelog for the endpoint poll patch?
[06:16:58] <exDM69> abcminiuser_: you want to credit me in the commit message
[06:17:02] <exDM69> ?
[06:17:32] <abcminiuser_> Sure, I always add a "thanks to" for any commits that were a result of outside help
[06:17:50] <abcminiuser_> I don't apply actual code patches directly due to the copyright/license issue, but I don't see why I can't at least put in the helper's name
[06:18:28] <exDM69> abcminiuser_: Riku Salminen, that's my given name
[06:19:07] <exDM69> with Git, applying patches easy. But I don't see where the license problem is. Is LUFA using some funny license or are you just afraid of lawyers?
[06:20:01] <abcminiuser_> Bit of both - since I resell LUFA occasionally, having third party code in the core is a no-no
[06:20:59] <abcminiuser_> OK, updating trunk now
[06:22:07] <exDM69> cool
[06:34:56] <abcminiuser_> Hrm
[06:35:07] <abcminiuser_> http://www.olimex.com/dev/DUINO/OLIMEXINO-32U4/OLIMEXINO-32U4-A-SCH.pdf
[06:35:09] <abcminiuser_> Am I lacking coffee, or is the schematic of the HWB button a bad idea?
[06:35:28] <abcminiuser_> Adding a cap on HWB would mean it would stay low for longer on startup while the cap charges, right?
[06:37:00] <CapnKernel> abcminiuser_: You could include third-party code if contributors expressly assigned copyright to you. You can then do with it what you like.
[06:37:20] <abcminiuser_> CapnKernel, I know, I just like to stay away from the legal stuff as far as possible
[06:37:26] <abcminiuser_> I like to code, not to be sued
[06:37:57] <CapnKernel> How sage
[06:42:09] <abcminiuser_> How's the business going?
[06:43:52] <buhman> abcminiuser_: I really like your interrupt and timer tutorials by the way; thank you.
[06:47:51] <abcminiuser_> buhman, glad to hear it :)
[06:48:09] <abcminiuser_> Did you read the old AVRFreaks versions, or the new PDF versions from my site?
[06:50:42] <buhman> well, I skimmed the avrfreaks version on the timers; then I noticed at the top you wrote that the stuff on your site is better, so I read those
[06:52:58] <buhman> so both; but I do like the PDF better
[06:53:10] <buhman> latex output is always nice to look at
[07:41:53] <abcminiuser_> Hrm, any make/shell script monkeys in here?
[07:41:53] <abcminiuser_> I have the output of a grep, for each matching line I want to find the corresponding line in another file, and pull out data from that corresponding line
[07:42:28] <CapnKernel> I'm pretty handy with make
[07:42:48] <CapnKernel> Can you put some examples on pastebin?
[07:45:33] <abcminiuser_> I want to make a build test for LUFA
[07:46:00] <CapnKernel> Ok
[07:46:01] <abcminiuser_> Where it will grep the board define map in a header and produce an output file, BoardDefs.txt
[07:46:12] <CapnKernel> Yep
[07:46:27] <abcminiuser_> This contains newline seperated board names (BOARD_NONE, BOARD_USBKEY, etc)
[07:46:40] <abcminiuser_> I will then have a second file, BoardMap.cgf which will contain the compile settings, (BOARD_NONE = AVR8 | at90usb1287)
[07:46:45] <CapnKernel> People often do it the opposite way: Have a txt file, which is used to generate a .h
[07:47:17] <abcminiuser_> I want to go through each line in BoardDefs.txt, pick out the corresponding line in BoardMap.cfg and then use those values
[07:47:24] <CapnKernel> Yep
[07:47:33] <abcminiuser_> I can generate both files, but no idea how to cross-reference them
[07:47:49] <abcminiuser_> Methinks I should just use Python, since the end user won't be using the tests
[07:48:03] <CapnKernel> This is not for end users?
[07:48:25] <CapnKernel> I'd be inclined to use awk, as it does this sort of thing as a one-liner. Can also be embedded in makefiles
[07:48:41] <abcminiuser_> No, for my own testing before release
[07:48:46] <CapnKernel> Ok.
[07:48:50] <abcminiuser_> To try various compile options and configurations to ensure everything works
[07:49:03] <CapnKernel> I'd recommend generating the .h from the .txt and .cfg files though
[07:58:36] <abcminiuser_> Oh wait, xargs
[08:00:25] <CapnKernel> xargs rocks. Look out for spaces in names.
[08:00:41] <CapnKernel> I'm happy to help you Dean!
[08:12:15] <_abc_> actually bash programming is a little more complicated than just using xargs
[08:12:33] <_abc_> joining #bash and perusing the wiki at wooledge.org is highly recommended
[08:13:01] <grummund> actually xargs isn't bash :p
[08:14:41] <grummund> abcminiuser_: use the -f option to grep
[08:21:32] <gmarsh> hey, i'm banging my head against a desk here trying to make WDT interrupts reliably work... I'm using the WDT on a Mega328 to wake up at a ~60ms rate, do some processing and go back to sleep.
[08:22:59] <gmarsh> got code running, debugging using debugwire on an avr dragon... anyway it seems that once the WDT fires, the WDIE bit clears
[08:23:04] <CapnKernel> gmarsh: I'm curious why you're using the WDT specifically.
[08:23:16] <gmarsh> so my "wake up from sleep" thing works once.. then never wakes up
[08:23:49] <gmarsh> power consumption - I'm running the AVR off a battery, using a regular system timer means running the core and a few mA of power draw. waking up from sleep with the WDT means uA.
[08:24:20] <grummund> gmarsh: so it wakes up on interrupt or it resets?
[08:25:08] <gmarsh> hmm... found a clue, WDRF is being set in MCUSR
[08:25:16] <OndraSter> that's reset
[08:25:52] <gmarsh> I'm writing (1<<WDIE)|(0<<WDE)|(other stuff) to the watchdog control register... hmm
[08:26:32] <grummund> there is some strict timing & sequence required or it will not work
[08:26:54] <OndraSter> and sure WDIE clears after WDT fires
[08:26:56] <OndraSter> WHOLE MCU RESETS
[08:26:57] <OndraSter> :)
[08:27:07] <grummund> which is why avr-libc provides a macro which implements it using inline assembler
[08:27:26] <OndraSter> it is like pulling reset line to ground for few usecs/msecs
[08:27:53] <OndraSter> and I am somewhat sure you can go to uA with one timer running
[08:28:17] <gmarsh> yeah, I was using a pair of C writes, compiled with -Os, which worked... then I turned off -Os to make code easier to debug, realized "hey it's not working anymore", turned on -Os again and it still wasn't working.
[08:28:17] <OndraSter> or is it only on xmegas possible with ULP clock source and advanced power management..
[08:28:38] <grummund> gmarsh: do you want it to reset or interrupt?
[08:28:39] <gmarsh> the avr-libc WDT functions don't support the 'interrupt only' mode so I just banged out some inline asm
[08:28:45] <gmarsh> interrupt, not reset.
[08:29:11] <grummund> ok so it works now?
[08:29:19] <OndraSter> there are some WDT interrupts?
[08:29:29] <OndraSter> duh
[08:29:39] <OndraSter> it wasn't on my previous mega, sorry :)
[08:29:42] <OndraSter> didn't know about that
[08:29:51] <gmarsh> ain't working yet... just adding the code now to clear WDRF
[08:30:59] <grummund> gmarsh: i have a macro similar to the avr-libc one but for interrupts
[08:31:13] <grummund> *somewhere* ;)
[08:31:13] <gmarsh> loading 30K of code over debugwire is sloooow, loading now
[08:31:19] <gmarsh> hah
[08:32:00] <gmarsh> I did a boatload of AVR programming at a previous job, which included a functioning WDT sleep and lots of other handy functions... part of me wishes I walked out with a thumbdrive full of code, alas that would have been theft.
[08:32:25] <grummund> it would
[08:32:35] <grummund> but also "smart"
[08:35:53] <gmarsh> YES! working now. Seems you can't reconfigure the watchdog with WDRF set.
[08:38:50] <abcminiuser_> CapnKernel, nearly got it
[08:38:56] <CapnKernel> Yay
[08:39:00] <abcminiuser_> Just need to know how to test if a bash variable is empty
[08:43:22] <CapnKernel> [ -z $var ]
[08:43:52] <CapnKernel> Or [ "$var" = "" ]
[08:44:10] <CapnKernel> You must have double quotes on the latter form.
[08:44:30] <grummund> also the first form
[08:47:15] <CapnKernel> why?
[08:47:38] <grummund> try it: var="foo bar"; if [ -z $var ]; then echo z; fi
[08:48:00] <CapnKernel> Ahh I see.
[08:48:58] <grummund> rule of thumb: *always* quote variables and *always* use ${...} form
[08:49:15] <grummund> e.g. "${myvar}"
[08:49:36] <CapnKernel> grummund: Why the latter? (genuine query)
[08:53:03] <grummund> prefix=foo; suffix=bar; ext=baz; filename="$prefix_$suffix.$ext"; echo "filename=$filename"
[08:53:18] <grummund> without testing, what would you expect that to print?
[08:54:27] <abcminiuser_> CapnKernel:
http://pastebin.com/qpRT4SmY
[08:54:35] <abcminiuser_> That almost works, it echos the right command
[08:54:42] <abcminiuser_> But when I run it it goes skitzo
[08:54:50] <CapnKernel> filename=bar.baz
[08:54:58] <CapnKernel> prefix_ will be taken to be the name of the variable
[08:55:05] <CapnKernel> Now to try it
[08:55:35] <grummund> yep exactly. but obviously was intended to be foo_bar.baz
[08:59:48] <abcminiuser_> The call to make in that script seems to ignore the parameters, causing it to recursively run
[09:01:50] <grummund> that's going to break in so many ways
[09:02:14] <CapnKernel> testboards should have BoardList.txt as a dependency, to force generation
[09:05:38] <abcminiuser_> Got it!
[09:05:52] <abcminiuser_> Science!
[09:12:47] <CapnKernel> As opposed to the waving of dead chickens! :-)
[09:20:47] <gmarsh> aaah
[09:21:08] <gmarsh> mega328 doesn't seem to have any means to tell you if a TWI STOP has been sent
[09:21:18] <gmarsh> making back to back writes fail.
[09:29:38] <OndraSter> anybody has extra $26k5? For 800x600 RGB LED wall :D :D
[09:29:50] <OndraSter> or $17k for 640x480
[09:29:57] <gmarsh> sounds like a kickstarter project!
[09:30:09] <OndraSter> or $126k for 1920x1200 D:
[09:30:21] <gmarsh> if you contribute 100 dollars or more, i'll send a picture of your name displayed on the screen!
[09:30:31] <OndraSter> :D
[09:31:08] <gmarsh> (sadly I'm considering using the site for a project I'm doing..)
[09:32:27] <OndraSter> I suppose that through this amount and through CapnKernel it would be noticably less
[09:32:34] <OndraSter> by 1/4 - 1/3 I suppose
[09:33:19] <gmarsh> I do wonder if there's someone out there building such a thing
[09:33:33] <OndraSter> if I had the money support - I would be :)
[09:33:58] <OndraSter> I built small 32x48 one, just RG colors
[09:34:19] <CapnKernel> Some of the buildings in Shenzhen have full colour LED screens that are 12 stories high.
[09:34:26] <OndraSter> hehe
[09:34:33] <CapnKernel> full-colour as-good-as-a-TV video
[09:34:35] <gmarsh> I built a little 16x16 thing using four RGB 8x8 panels a couple years back... thoguht it was cool, doesn't see any use nowadays
[09:35:09] <gmarsh> in hindsight I should have done 32x8 so I could at least use it as a little scrolling sign
[09:35:24] <CapnKernel> Yes that would be useful
[09:35:24] <OndraSter> :)
[09:35:46] <gmarsh> but hey, gotta spend a hundred bucks on a dumb project once in a while
[09:35:58] <OndraSter> exactly
[09:36:47] <OndraSter> repeating my last weeks' question: anybody messed with analog switches? And knows stuff I should take care of etc
[09:37:02] <gmarsh> used analog switches plenty of times... what's the application?
[09:37:19] <OndraSter> waveform generator output switching (sine/triangle/square)
[09:37:28] <OndraSter> ADG704 from ADI sounds quite fine
[09:38:38] <gmarsh> only thing to watch out for is impedance... their on-resistance varies a bit with signal voltage, make sure whatever circuit you're putting it in can handle that
[09:39:08] <OndraSter> the current through the analog switch will be something near zero
[09:39:13] <OndraSter> ebcause it is fed to opamp
[09:39:15] <OndraSter> because*
[09:39:17] <gmarsh> ok
[09:39:28] <OndraSter> that's why Ron doesn't matter for me... Z does a little
[09:40:14] <gmarsh> often times they'll have a bit of parastic capacitance across the switches
[09:40:31] <OndraSter> I want to software limit the output to 20MHz
[09:40:37] <OndraSter> maybe a bit higher
[09:40:49] <OndraSter> we are not talking hundreds of MHz luckily :)
[09:41:29] <OndraSter> http://dl.dropbox.com/u/3936936/dds_brd.png
[09:41:34] <gmarsh> one thing I can tell you - you can't switch broadcast FM with an analog switch, there isn't much difference between on and off with analog switches... better off using a physical relay for that
[09:41:58] <OndraSter> isn't FM bcast 88 - 110MHz?
[09:42:16] <CapnKernel> How do cellphones switch frequency ranges?
[09:43:02] <gmarsh> passively
[09:45:01] <gmarsh> one comment on the schematic... how fast are you clocking the DDS chip, and how critical is jitter on the square wave output?
[09:45:14] <OndraSter> the DDS is clocked @ 75MHz
[09:45:38] <OndraSter> jitter on the square wave... it is somewhat expected in this price range I am targetting that the perfomance on high frequencies won't be that good
[09:45:51] <gmarsh> ok, as long as that's known
[09:45:59] <OndraSter> try building awesome waveform generator for $30 - $40 :)
[09:46:33] <gmarsh> yeah, using an AD9852 or something fast with a comparator would blow half your budget :D
[09:46:38] <OndraSter> yep
[09:46:56] <OndraSter> this has comparator as well, but I found out that you can internally connect the DAC's MSb to the output pin directly
[09:47:00] <OndraSter> saves wiring trouble
[09:47:59] <gmarsh> hmm
[09:48:16] <OndraSter> basically, the analog signal is fed to opamp, that makes it into the same (maximum) voltage level as the square wave (3v3), then it is split to two LC filters - one for <1MHz, the other one <30MHz (might put it a bit lower) and thirdly for no LC filter (triangle), these three plus square are fed to the analog switch
[09:48:28] <gmarsh> i've never worked with that exact DDS chip and I'm too busy with something here to pull out a datasheet... what's the output voltage compliance range of IOUTx?
[09:48:40] <OndraSter> 0-900mV they say
[09:48:46] <OndraSter> I found actually one problem
[09:48:48] <gmarsh> normally I see current output DDSes feeding an op-amp in an inverting configuration
[09:49:00] <OndraSter> I can't set the maximum voltage on square wave..
[09:49:30] * Tom_itx squarely waves at OndraSter
[09:50:09] <OndraSter> and yes, it is current output, that's why there is the 200R load, as specified by datasheet
[09:50:45] <gmarsh> and I'm assuming 100n is not the actual value of the load cap
[09:50:57] <OndraSter> their demo board has 100n load cap
[09:52:01] <gmarsh> you won't get MHz out of it with that sized load cap
[09:52:08] <OndraSter> :o)
[09:52:58] <OndraSter> http://clip2net.com/s/1T7Xn
[09:53:06] <OndraSter> looking at the evaluation board schematics..
[09:53:24] <gmarsh> DNI
[09:53:33] <OndraSter> Current Output. This is a high impedance current source. A load resistor of nominally 200 Ω should be connected between IOUT and AGND. IOUTB should preferably be tied through an external load resistor of 200 Ω to AGND, but it can be tied directly to AGND. A 20 pF capacitor to AGND is also recommended to prevent clock feedthrough.
[09:54:14] <OndraSter> when I read DNI I can only think about direct neural interface :X
[09:55:25] <gmarsh> next comment: watch the voltage range on op-amp inputs/outputs
[09:55:42] <OndraSter> as in so they do not get into saturation?
[09:55:55] <gmarsh> yeah... IOUTx are fairly close to ground
[09:56:06] <OndraSter> don't worry about that
[09:56:15] <OndraSter> the opamp has offset of 600uV
[09:57:21] <gmarsh> just thinking, your minimum op-amp gain is 3, ground referenced
[09:57:40] <gmarsh> oh right, IOUTx will always be positive, you're good :D
[09:58:03] <gmarsh> make sure it's a rail-rail in/out op-amp
[09:58:07] <OndraSter> it is
[09:58:10] <OndraSter> ADA4851
[09:58:14] <OndraSter> it is all from ADI :D
[09:58:17] <OndraSter> I might (should!) check out PGAs
[09:58:30] <OndraSter> or VGAs as ADI calls them
[09:59:07] <gmarsh> for higher frequency stuff, I'm a fan of fixed gain and switchable attenuators
[09:59:19] <gmarsh> trying to remember a part number, 1sec
[10:00:05] <gmarsh> PE4302
[10:00:39] <OndraSter> 4GHz, that won't be cheap :D
[10:00:56] <gmarsh> heh, right
[10:01:16] <gmarsh> at my last job, I designed FM radio broadcast exciters
[10:01:25] <OndraSter> :)
[10:01:32] <OndraSter> I am still studying
[10:01:33] <gmarsh> stupidly high end stuff, so I'm not used to working on a hobby budget
[10:01:38] <OndraSter> finishing middle (high for US) school
[10:02:28] <gmarsh> highest end one I made used a AD9779A DAC, transformer output, PE4302 attenuator, about 50 bucks worth of analog filtering, 50 dollar clock oscillator...
[10:03:06] <gmarsh> awesome you're in high school and designing a DDS card
[10:03:25] <gmarsh> planning on doing EE eventually?
[10:03:29] <OndraSter> it is just one of those "I like this, I could use it, many people could want it, so why not at least think about it"
[10:03:31] <OndraSter> sure
[10:03:40] <gmarsh> let me tell you this: you'll kick ass.
[10:03:43] <Tom_itx> OndraSter i forgot where you were located...
[10:03:47] <OndraSter> but I am more MCU than electronics
[10:03:48] <OndraSter> Tom_itx, CZE
[10:03:55] <Tom_itx> ahh ok
[10:03:56] <OndraSter> more MCU programming*
[10:04:18] <gmarsh> I was one of only a few people in engineering school that went into the program knowing how to design things, program microcontrollers, etc
[10:04:24] <OndraSter> :)
[10:05:03] <gmarsh> and desinging a DDS card, experience with Eagle, etc... you're far ahead of where I was in high school
[10:05:29] <OndraSter> :)
[10:05:44] <OndraSter> my fellow schoolmates are doing tomorrow "practical" part of final exams
[10:05:48] <OndraSter> I did long term project instead
[10:05:51] <OndraSter> let me find some pictures
[10:06:22] <OndraSter> https://skydrive.live.com/redir.aspx?cid=40bf7833586103ab&resid=40BF7833586103AB!499&parid=40BF7833586103AB!111&authkey=!AM9BlG3lYBLJUqs
[10:08:07] <gmarsh> awesome
[10:08:12] <gmarsh> what are you using for drivers?
[10:08:33] <OndraSter> my teacher had bunch ("bunch" = 24 + another 24 waiting for another one) of Allegro A6275
[10:08:40] <OndraSter> serial in, parallel out with constant current mode
[10:08:47] <OndraSter> (constant current limit?)
[10:09:59] <gmarsh> cool
[10:10:07] <gmarsh> I gotta run... good work with the sign, i'm impressed, keep at it.
[10:11:19] <OndraSter> yay
[10:11:27] <OndraSter> remind me to say him thanks, when he gets back :D
[10:13:46] <Tom_itx> tell him thanks
[10:15:09] <Tom_itx> good stuff OndraSter
[10:15:17] <OndraSter> thanks :D
[10:15:20] <Tom_itx> abcminiuser are your tutorials on my page outdated?
[10:31:04] <abcminiuser> Tom_itx, yes, in favor of the updated PDFs
[10:44:50] <OndraSter_> why do PGAs have so many wires?
[10:44:53] <OndraSter_> I want just:
[10:44:55] <OndraSter_> signal in
[10:44:56] <OndraSter_> signal out
[10:45:05] <Tom_itx> abcminiuser should i remove them?
[10:45:07] <OndraSter_> and MOSI+SCK or SDA+SCL
[10:46:34] <specing> OndraSter_: Because it rhymes with BGA!
[10:46:47] <OndraSter_> :D
[11:13:29] <abcminiuser> Tom_itx, if you think so, the updated ones aren't vastly different except for the EEPROM one
[11:13:39] <abcminiuser> Probably best to just mirror the PDFs or links I guess
[11:13:55] <Tom_itx> k
[11:14:02] <Tom_itx> i'll address it in due time
[11:15:10] <exDM69> abcminiuser: woohoo, I have my box working as a joystick for the first time
[11:15:20] <exDM69> it's no longer a very expensive brick \o/
[11:16:34] <abcminiuser> exDM69, Awesome!
[11:16:34] <abcminiuser> What kind of joystick is it?
[11:17:02] <OndraSter> I am thinking how to solve the square wave output voltage ... :/
[11:17:11] <exDM69> abcminiuser: it's not a joystick at all. it's a huge box with lots of knobs and buttons
[11:17:27] <exDM69> abcminiuser: I intend to use it for radios and navigational instruments in flight simulation
[11:17:52] <exDM69> also some of the buttons are nice backlit buttons with 2 color leds inside, I can use them eg. to indicate autopilot state
[11:18:45] <exDM69> I don't think I have a picture at hand. but it's basically built from heavy duty buttons and toggle switches to make it look like an aircraft part. housed in a water proof junction box
[11:20:39] <abcminiuser> Sounds terrifyingly awesome
[11:21:08] <exDM69> I might add a hard disk motor as a trim wheel later
[11:21:32] <OndraSter> you should also add the rest of the cockpit
[11:21:33] <OndraSter> and rent it
[11:21:37] <OndraSter> I think I would go there lol
[11:22:07] <exDM69> it's also very usable for music. when I have the time to I'll write another LUFA firmware to make it work as a midi controller
[11:22:21] <OndraSter> Launchpad? :)
[11:22:24] <exDM69> OndraSter: I have a huge surround display with 3 monitors and I use a head tracking
[11:22:38] <exDM69> I only want an input with good tactile feedback
[11:22:39] <OndraSter> Ableton Novation Launchpad*
[11:22:58] <abcminiuser> BREAKING NEWS: Plane crashes as pilot gets confused and tries to lay down a sick drum beat on the autopilot console
[11:23:09] <OndraSter> :D :D
[11:23:57] <buhman> the timing on my microcontroller scares me; I have
http://sprunge.us/TTgN . I'm using a 20mHz crystal. When I attach just my multimeter to PA0, it says "1.000 Hz". However, when I attach an LED and the multimeter, the LED is /clearly/ not blinking regularly, and the value on the multimeter starts jumping around.
[11:24:07] <buhman> eventually the LED stops blinking completely
[11:24:17] <buhman> and the DMM says 0.000 Hz
[11:24:22] <buhman> MHz*
[11:24:44] <buhman> (that is, 20 mega-hertz, not milli-hertz)
[11:25:49] <buhman> watching Vcc very near where it connects to the controller, it does bounce around a bit from 5.15<->5.17 V
[11:27:29] <buhman> my crystal resonates at 18pF, and I have two 18pF caps connected between ground and each pin of the controller
[11:40:22] <OndraSter> hmm high bandwidth digital pot
[11:40:24] <OndraSter> http://www.maxim-ic.com/app-notes/index.mvp/id/3081
[11:40:28] <OndraSter> good appnote
[11:40:31] <OndraSter> thumbs up to Dallas!
[11:51:01] <OndraSter> actually
[11:51:05] <OndraSter> there might be another solution
[11:54:07] <OndraSter> bah I will think about it after dinner
[12:20:10] <rue_house> buhman,
[12:22:50] <OndraSter> ping
[12:24:51] <OndraSter> so
[12:24:51] <OndraSter> http://clip2net.com/s/1T9MT
[12:24:52] <OndraSter> thi
[12:24:53] <OndraSter> s
[12:25:05] <OndraSter> feeding 0 - 3V3 signal from some DAC to the negative input of opamp
[12:25:19] <OndraSter> the higher voltage on the DAC, the smaller output voltage of the opamp
[12:25:30] <OndraSter> and since it is without any feedback, it is in saturation mode
[12:25:33] <OndraSter> wait
[12:25:35] <OndraSter> yes
[12:25:41] <OndraSter> at least at 0V it is saturated
[12:25:43] <OndraSter> not in the top
[12:25:49] <OndraSter> or is it
[12:27:55] <Casper> if it even work as you expect
[12:28:03] <buhman> rue_house: pong
[12:28:04] <Casper> an opamp without feedback will react as a comparator
[12:28:48] <rue_house> buhman, what did you do with the reset line, do you have a pullup resistor?
[12:28:56] <buhman> erm nope
[12:29:02] <rue_house> well there ya go
[12:29:21] <OndraSter> Casper, but since it is +3v3-2v, on the output it should saturate to 1v3?
[12:29:26] <OndraSter> or am I doing bad logic here
[12:29:34] <buhman> so the reset line without a pullup resistor causes the microcontroller to act bizzarely like this?
[12:29:59] <rue_house> as it randomly resets with the varrying static level in the air? sure
[12:30:03] <OndraSter> there is internal pullup, 100k I think... but with this resistance it picks up noise
[12:30:26] <buhman> what if I just hold it at Vcc?
[12:30:40] <rue_house> try it with a pullup on reset, I use 1K, then tell me if you still have a problem
[12:30:56] <rue_house> its reccomended to always use a resistor
[12:31:18] <OndraSter> Casper, so? :P
[12:31:20] <OndraSter> any ideas?
[12:31:21] <rue_house> order a resistor from the store, and call me back in 2 weeks when you have recieved and isntalled it
[12:31:22] <OndraSter> if I am thinking right
[12:31:29] <Casper> OndraSter: (3.3-2)*gain = 1.3*over a million = full output
[12:31:34] <OndraSter> hmm
[12:31:39] <OndraSter> right
[12:31:40] <OndraSter> then I need gain of 1 :)
[12:31:55] <Casper> so you need feedback
[12:31:56] <rue_house> or, if your not above using a recycled part, pull one out of an old computer power supply
[12:32:04] <buhman> rue_house: still fails
[12:32:13] <Casper> but you might want to look up differential input (it's 4 resistors)
[12:32:18] <buhman> 1k between reset and Vcc right?
[12:32:23] <rue_house> just a sec
[12:32:26] <OndraSter> right, I didn't think about differential input
[12:33:03] <rue_house> yup, vcc and reset
[12:33:19] <rue_house> ok, you have the caps on the crystal...
[12:33:30] <rue_house> do you have the WDT off in the fuses?
[12:36:34] <buhman> nope
[12:37:30] <buhman> iirc I'm 0xFF low 0x99 high
[12:37:39] <buhman> 0xFF extended
[12:38:18] <buhman> erm
[12:38:21] <buhman> sorry
[12:38:23] <buhman> yes it is off
[13:30:16] <JoeLlama> Gate A20 Problem:
http://www.pagetable.com/docs/gate_a20.jpg
[13:32:59] <buhman> erm what problem?
[13:34:06] <OndraSter> gate a20
[13:34:09] <OndraSter> reminds me of.. 386
[13:39:52] <Steffanx> 386 isn't that "before your time" OndraSter ?
[13:41:09] <Casper> I never knew what the gate a20 prob was, but I read about that lots of time...
[13:41:09] <OndraSter> hmm
[13:41:13] <OndraSter> is it <92? :D
[13:41:16] <Casper> ahh the good old dos days
[13:41:35] <Casper> and memmaker... and emm386... and himem.sys and al....
[13:41:37] <OndraSter> the times when overclocking or hardware modding was DIYable :/
[13:42:00] <Casper> where overclocking was actually meaningfull
[13:42:04] <Steffanx> Or there was just a 'turbo' button OndraSter :)
[13:42:08] <OndraSter> or turbo
[13:42:15] <Casper> turbo button wasn't overclocking
[13:42:23] <Casper> it was really a downclock button
[13:42:41] <Casper> should have been called the turtle button
[13:42:51] <Steffanx> :)
[13:43:19] * OndraSter is using gear that is older than himself! scope + wave gen
[13:43:28] <OndraSter> wave gen is from 80 or 81, scope is from 86 or 87
[13:43:32] <Steffanx> I think a 386 was my first computer :)
[13:43:43] <OndraSter> our first PC at home was 386 I think as well
[13:43:55] <OndraSter> then 486, P II (Celeron actually), P III, P IV
[13:44:13] <Steffanx> At least those old computers still work
[13:44:26] <OndraSter> then our dad moved out and we were left with the P III (me and my brother, after that I got my own computer (Athlon 1.2GHz - that was around the time that first dualcores started appearing)
[13:44:27] <OndraSter> yep
[13:44:37] <OndraSter> and then I went for E2200, E5200 and now i5-2500k :)
[14:30:58] <exDM69> abcminiuser: damnit, seems like I can't use this as a joystick
[14:31:24] <exDM69> gotta make it a keyboard or something so it will stop missing clicks
[14:32:18] <exDM69> the software I'm trying to use this with does not recognize the 5 ms short clicks
[14:32:29] <exDM69> hmmmh, maybe if I bump it up to 10 ms
[14:40:44] <Toneloc> anyone see the hack where you can use a TI launchpad to program AVR ?
[14:43:39] <specing> Toneloc: there is no *hack*, programming AVRs only requires normal SPI
[14:45:46] <Toneloc> hmm, thats interesting
[14:46:14] <Toneloc> I built a parallel port based programmer for avr once
[14:46:42] <Toneloc> its wasnt very reliable, perhaps I could devise some sort of SPI device
[14:48:03] <Toneloc> I think Realterm has an SPI add-on device you can build
[14:56:29] <buhman> I have a simple light-blinking program I put on my atmega1284p which is on a breadboard connected to 5V power, on a 20MHz crystal (with capacitors going to ground on each pin), 100nF decoupling caps, and a 1K pullup resistor for the reset pin
[14:58:42] <buhman> my code
http://sprunge.us/TTgN ; when I have nothing by my DMM on PA0, I read a steady "1.000 Hz"
[14:59:05] <buhman> when I add a LED, the loop dies after about 8 blinks or so
[15:21:13] <amee2k> what letter (as variable name) does attenuation (of a signal) typically have in formulas?
[15:22:40] <alexh> same as gain, as it's the same thing
[15:23:06] <alexh> e.g. A
[15:23:16] <amee2k> hmm okay. makes sense i suppose
[15:23:18] <alexh> sometimes very un-mathematically att
[15:23:18] <amee2k> thanks :)
[18:34:44] <tekniq> ne1 know how to get power out of pin 4 of a Nokia pop-port connection?
[19:42:45] <WormFood> other than Eagle, what schematic design software is available for Linux?
[19:43:06] <WormFood> CapnKernel, 早上好
[19:44:15] <CapnKernel> WormFood: Hi
[19:44:32] <CapnKernel> There are several, but the two most popular are gEDA and KiCad
[19:44:34] <CapnKernel> I use KiCad
[19:44:41] <CapnKernel> Atommann uses gEDA
[19:44:46] <CapnKernel> Tully uses KiCad
[19:44:52] <WormFood> ok
[19:45:00] <CapnKernel> (avr: They're people we both know, don't fret)
[19:45:17] <WormFood> I've always used eagle in the past, but never needed it for anything serious, just little things
[19:45:28] <WormFood> people we know, in person (as opposed to just online)
[19:46:11] <WormFood> I want to do something a little more complex, and don't want to get clawed by the eagle, because I don't want to pay for it (most of the best software is free anyways)
[19:46:49] <nevyn> bdale uses gEDA
[19:46:56] <CapnKernel> It's worthy
[19:47:03] <CapnKernel> KiCad is a little easier to learn.
[19:47:12] <CapnKernel> Try #kicad for help
[19:47:23] <WormFood> at a glance of their screen shots, they both look nice.
[19:47:48] <CapnKernel> KiCad is a little more integrated than gEDA
[19:47:55] <WormFood> but it seems that the development of kikad is a little more active.
[19:47:57] <CapnKernel> gEDA's symbols are probably better than KiCad's.
[19:48:11] <CapnKernel> There's also another interesting one called Fritzing:
http://fritzing.org/
[19:49:59] <WormFood> it is a shame, that these different projects can't pool their resources, to work for a common goal, instead of splitting up their efforts.
[19:55:28] <Roklobsta> I use Mentor Graphics.
[19:55:34] <Casper> WormFood: or atleast find a way to have common symbols
[19:55:43] <Roklobsta> and Veribest
[22:01:49] <learningc> any knows any good book(s) on designing circuits?
[22:02:50] <Tom_itx> umm, join #electronics and read their topic
[22:02:54] <Tom_itx> they are listed there
[22:03:08] <Tom_itx> then get the hell outta that channel before you get poluted
[22:03:25] <Casper> LOL
[22:03:32] <CapnKernel> That place is toxic
[22:04:00] <Tom_itx> they used to be anyway...
[22:08:15] <Casper> if only the ops would make a cleanup of the trolls...
[22:08:22] <Casper> that would help so much
[23:11:19] <Landon> woo, tiny computers = standing desk for avr work
[23:12:54] <Landon> I should probably get an ESD mat though, just a plastic storage shelf
