Back
[00:35:19] * drgreenthumb attaches 2x http://www.digikey.com/product-detail/en/BU-102B-0/BU-102B-0-ND/2688281 and jump starts the channel. evenin.
[00:46:28] <mimcpher> I have a friend who has scars on his nipples because somebody thought it would be funny to put jumper cables on 'em
[01:17:15] <drgreenthumb> uh I wasn't suggesting anything quite so... horrific as that o:
[01:39:32] <DarkSector> so how is the AS6?
[01:39:49] <DarkSector> is it better than AS5 when it comes to burning the chip?
[01:45:11] <ziph> mimcpher: Are you sure "funny" was the real motive?
[03:32:29] <gidna> Hi
[03:32:35] <nevyn> hi
[03:36:25] <gidna> I've just finished my parallel programmer but I don't have an atmega now.. can I test it on this
http://www.engineersgarage.com/sites/default/files/LM324_1.jpg?1281349133 ?
[03:45:31] <Kevin`> gidna: what would the point be of connecting it?
[03:45:52] <Kevin`> gidna: if you ran the interface slow enough you could watch it on some leds. i'd just wait until you get a chip
[03:46:20] <gidna> Kevin`, I just want to see if the programmer works,,
[03:46:58] <Kevin`> and how will you test whether the programmer works using opamps?
[04:46:50] <Znaap> Hi, Im using a uc3c1512c in which i have configured the AD-converter to run at 3,2 kHz. Im using the sequence done irq to trigger an interrupt routine in which i have placed a busy-wait loop (for debug reason).
[04:47:32] <Znaap> When i run the busy/wait loop over a certain amount of time the entire microprocessor dies, it enters some weird state where noting is working
[04:47:50] <Znaap> Any ideas on what possible could be wrong?
[05:14:04] <specing_> Znaap: Never, ever do busy waiting in an interrupt handler
[05:14:34] <specing_> tobbor is CANUCKED ;P
[05:14:34] <tobbor> YANKIE.
[05:25:40] <Znaap> specing_: Im aware of that, its debug purpose, there seems to be some weird time dependency against something unknown
[06:15:52] <`Wedge_> stupid canucks
[06:16:46] <`Wedge_> so?
[06:16:48] <`Wedge_> tobbor: hello?
[06:16:51] <`Wedge_> tobbor: canucks?
[06:16:55] <OndraSter> stupid test, I missed an A by
[06:16:55] <`Wedge_> :(
[06:16:56] <OndraSter> wait for it
[06:16:59] <OndraSter> 0,05%
[06:17:00] <OndraSter> ..
[06:17:09] <`Wedge_> even a bot won't talk to me
[06:18:01] <DrLuke> what's a canuck
[06:18:43] <CapnKernel> It's what gets Yankees worked up.
[06:18:55] <OndraSter> anyway, I got an A from my long term project
[06:19:05] <`Wedge_> DrLuke: canadian
[06:20:20] <CapnKernel> OndraSter: 0.05%: The amount one missed out on an A by, because one hangs around in #avr :-)
[06:20:28] <OndraSter> :D
[06:20:37] <OndraSter> the maximum was 69 points
[06:20:39] <OndraSter> I had 60
[06:21:19] <CapnKernel> that's pretty good dude
[06:21:30] <OndraSter> yeah
[06:21:33] <OndraSter> still there is literature waiting :D
[06:21:38] <OndraSter> and I will be glad if I get a C there
[06:23:53] <OndraSter> and en essay
[06:23:59] <OndraSter> if it passes the requiements - I am going for A or B there
[06:24:04] <OndraSter> but if it doesn't, hopefuly D :)
[06:28:01] <specing_> OndraSter: I've got both literature and essays D
[06:28:04] <specing_> err C
[06:28:09] <OndraSter> :)
[06:28:24] <OndraSter> why are we using ABC anyway, it is mark 3!
[06:28:45] <specing_> Yes, its the damn CANUCKS and their imperial units
[06:28:46] <tobbor> Yankie!
[06:28:57] <OndraSter> I thought that CANUCKS use metric units?
[06:28:58] <tobbor> yankie, yankie, yankie.
[06:29:02] <OndraSter> americans use imperials
[06:29:25] <specing_> Yes, but I said CANUCKS so that the bot would scream the right ones :D
[06:29:26] <tobbor> Yankie.
[06:29:33] <`Wedge_> pam pam pam pam papam pam papam
[06:30:23] <specing_> OndraSter: Also, 87% for 5, what kind of a grading system is that?
[06:30:27] <OndraSter> wat
[06:30:31] <specing_> 90-100 % is for 5 here
[06:30:37] <OndraSter> 5?
[06:30:53] <OndraSter> are you in one of those backwards country where 5 = the best and 1 = the worst?
[06:30:57] <specing_> 60/69 pts is 86.95%
[06:31:00] <specing_> Yes
[06:31:04] <OndraSter> oh
[06:31:15] <specing_> NO YOU ARE BACKWARDS!
[06:31:16] <specing_> :D
[06:31:18] <OndraSter> noIamnot
[06:31:21] <OndraSter> aren't you German?
[06:31:24] <specing_> nope
[06:31:37] <OndraSter> oh..
[06:31:41] <OndraSter> it was amee2k !
[06:32:06] <specing_> Do you also have 6 grades?
[06:32:14] <OndraSter> no
[06:32:15] <OndraSter> 5
[06:32:19] <specing_> same here
[06:32:20] <OndraSter> 1 = the best, 5 = the worst (fail)
[06:32:28] <OndraSter> on uni only 4
[06:32:33] <OndraSter> 1 = the best, 4 = the worst (fail)
[06:32:37] <specing_> uni here is 1-10
[06:32:41] <specing_> 10 -- the best
[06:32:47] <OndraSter> or you could say it somewhat else
[06:32:50] <OndraSter> uni has 2 marks
[06:32:53] <OndraSter> "YES I PASSED!"
[06:32:57] <OndraSter> "ohgawd I did not pass"
[06:33:05] <OndraSter> :D
[06:33:18] <specing_> if you have 9+ grades, they let you pick classes at random
[06:33:25] <specing_> :D
[06:34:06] <specing_> the choosing part, that is
[06:34:19] <specing_> not sure how to name it, though
[06:38:24] * CapnKernel is suspecting tobbor is a fake - isn't something different supposed to happen the third time someone says CANUCKS?
[06:38:24] <tobbor> YANKIE.
[06:38:58] <specing_> It seems tobbor (as flyback would put it) is CANUCKED :P
[06:38:59] <tobbor> YANKIE!!!!
[06:39:13] <`Wedge_> where's flyback btw?
[06:39:17] <specing_> banned
[06:39:24] <specing_> :D
[06:39:55] <`Wedge_> :(
[06:41:28] <CapnKernel> It Is A Good Thing™
[06:41:48] <amee2k> i was what?
[06:43:17] <OndraSter> german
[06:43:38] <amee2k> oh only that
[06:43:44] <amee2k> i thought it was something interesting
[06:44:39] <CapnKernel> specing_ was accused of being German.
[07:00:23] <DrLuke> I am german
[07:01:37] <specing_> Zeee germans are everywhere!!!
[07:23:17] <CapnKernel> "Don't mention the war. I did once, but I think I got away with it"
[07:23:53] <CapnKernel> John Cleese, as Basil Fawlty
[09:21:42] <rue_mohr> hehehe
[09:23:27] <OndraSter> MEHHEHHHEHEHHEE
[09:25:27] * rue_mohr hits specing STOP TORMENTING THE BOT!
[09:44:21] <Toneloc> can a homebrew programmer be used to program a atmega16a-pu ?
[09:56:15] <OndraSter> sure
[09:56:22] <OndraSter> ISP is "open"
[09:56:25] <OndraSter> and documented
[09:56:52] <DrLuke> tom_itx: I've tried installing the Jungo drivers again, and FLIP could connect to the programmer now :D
[10:00:59] <DrLuke> and avr studio recognizes it =)
[10:02:27] <OndraSter> I need better regulated power supply. This 0 - 15V one won't make the 15W 33R resistor red :)
[10:02:38] <OndraSter> glow*
[10:02:41] <OndraSter> it is already painted red
[10:02:52] <OndraSter> (good old soviet russian resistors!)
[10:03:20] <DrLuke> hehe
[10:21:42] <Toneloc> OndraSter- yes, but it is supported by the same isp programmers you can build for the attiny2313?
[10:24:23] <Toneloc> this seems as simple as it can get>>
[10:24:24] <Toneloc> http://kartikmohta.com/tech/avr/programmer/avr_programmer.png
[10:24:55] <Toneloc> all those unconnected pins on the parallel port connector looks wrong
[10:25:04] <Toneloc> compared to other schematics
[10:29:36] <Tom_itx> so how many should there be??
[10:30:09] <Tom_itx> it's not the safest of programmers but it should work
[10:30:23] <Tom_itx> ghetto but work
[10:31:14] <Toneloc> well, there is pins that are supposed ot be tied together
[10:31:38] <Tom_itx> what for?
[10:31:46] <Tom_itx> to tell it it's a stk200?
[10:31:56] <Tom_itx> that's about it i think
[10:32:18] <Toneloc> http://wiredworld.tripod.com/tronics/atmel_isp.html
[10:32:36] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/avr/ispdongle.pdf
[10:32:45] <Toneloc> ^I built that and it had a heap of wires to connect
[10:32:52] <Toneloc> and its operation was flaky at best
[10:33:00] <Toneloc> not yours, mine
[10:33:11] <Tom_itx> so get a good one and be done with it
[10:33:54] <Toneloc> Im hoping to build a good one:)
[10:33:55] <Tom_itx> nice little USB one that supports all the protocols :)
[10:34:08] <Toneloc> hmm, cheap chinese one
[10:34:21] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/boards/USBTiny_Mkii/USBTiny_Mkii_index.php
[10:34:28] <Tom_itx> sry, i'm not chinese
[10:35:46] <Toneloc> haha, I didnt say you were
[10:35:52] <Tom_itx> works with studio and avrdude
[10:35:57] <Tom_itx> windows osx and linux
[10:35:59] <Toneloc> I will buy one of those chepa chinese ones from ebay some time
[10:36:34] <Tom_itx> the parport one will work for isp
[10:37:18] <Toneloc> thats good
[10:37:33] <Toneloc> another guy is telling me i need to use a programmer
[10:37:44] <Toneloc> witha HC244 buffer in it
[10:38:06] <Toneloc> with many pins connected together
[10:38:23] <Toneloc> I guess that gives it more power to drive the pins?
[10:41:59] <Tom_itx> protects the parallel port
[10:42:42] <Toneloc> and you dont have to remove the programmer either because it tri-states its outputs?
[10:42:53] <Tom_itx> not sure
[10:43:23] <Tom_itx> you're better off removing it to prevent problems
[10:43:40] <Toneloc> ok
[10:43:49] <Toneloc> I want something quick and dirty for noe:)
[10:43:51] <Toneloc> *now
[10:43:54] <Tom_itx> at this point i would seriously consider a usb one as there are very few pc's that have parallel ports anymore
[10:44:18] <Toneloc> i have loads of paralllel port machines
[10:44:23] <Tom_itx> ok
[10:44:26] <Toneloc> I dont 'do' modern :P
[10:44:35] <Toneloc> I live out of salvage
[10:44:43] <Tom_itx> me too
[10:44:57] <Toneloc> this would work?.>
[10:44:58] <Toneloc> http://kartikmohta.com/tech/avr/programmer/avr_programmer.png
[10:45:25] <Tom_itx> cheap enough to find out... huh?
[10:45:39] <Toneloc> true!
[10:45:47] <Toneloc> lets do it:)
[10:45:47] <Tom_itx> 2 resistors
[10:45:53] <Tom_itx> i'm outta here
[10:45:54] <Toneloc> quick is what i want
[10:45:57] <Toneloc> okay
[10:46:00] <Toneloc> thanks for the help
[10:46:16] <Tom_itx> select stk200 in avrdude
[10:46:20] <Tom_itx> i think
[10:46:25] <Toneloc> ok
[10:46:30] <Tom_itx> later
[10:46:40] <Toneloc> thanks, later
[11:00:16] <DrLuke> Toneloc, I can only recommend tom's programmer, it's quite simple to use
[11:22:49] <inflex> +1 for Tom_itx 's programmer.
[11:22:54] * inflex has two of them, love them.
[11:50:49] <Toneloc> yeha, im sure it is
[11:50:54] <Toneloc> I have mine built now
[11:51:18] <Toneloc> now I have got to get a code example for the atmega16a
[11:51:23] <Toneloc> led blinky
[11:51:39] <Toneloc> I will try and change the one for teh ATTINY2313
[11:51:41] <Toneloc> *the
[11:53:48] <Toneloc> I asusme this attiny2313 code
[11:53:56] <Toneloc> should work on the atmega16a
[11:53:58] <Toneloc> http://pastebin.com/BEEUMCVC
[12:13:28] <Steffanx> Except for the wrong comments, yes Toneloc
[12:15:29] <specing> Steffanx: Maybe he is doing sinkinf
[12:15:34] <specing> g*
[12:15:46] * RikusW is wondering if 74LSxxx chips will work with AVR ?
[12:16:30] <Steffanx> Could be possible specing :P
[12:18:03] <specing> Toneloc: there are macros like sbi() and cbi() out there to make your code more readable, use them
[12:18:18] <Steffanx> Noo
[12:18:20] <Steffanx> Don't :P
[12:20:01] <RikusW> use asm instead :-P
[12:20:18] <specing> RikusW: Why use asm when you can use machine code?
[12:20:29] <Steffanx> Why use an AVR?!
[12:20:47] <RikusW> because asm is easier to remember than hex codes
[12:20:49] <specing> RikusW: btw, using asm instead of those silly C bitwise ops is actually more readable
[12:21:05] <Steffanx> Don't forget "imho" specing
[12:21:19] <RikusW> sbi/cbi only works on the low registers btw
[12:21:32] <specing> RikusW: I know
[12:21:47] <specing> RikusW: the cpi(), sbi() macros take care of that
[12:22:07] <specing> I only use asm when gcc poops itself
[12:22:28] <Toneloc> RikusW - I used it for my first programmer0- 74LSXXX series, worked fine
[12:22:30] <specing> like 8*8 multiplication when I know that the result will never be >2^8
[12:22:32] <Toneloc> as an octal buffer
[12:22:39] <Steffanx> cbi/sbi are deprecated anyway specing
[12:22:43] <Toneloc> well, programmer was actually crummy
[12:22:44] <Steffanx> _BV it is
[12:22:52] <specing> Or when it decides to use a lot of movs
[12:23:01] <specing> Steffanx: Not in my headers they arent.
[12:23:12] <Toneloc> anyway , this ultra simple programmer
[12:23:12] <Steffanx> Hmpf
[12:23:26] <Toneloc> isnt worth the bandwidth it was delivered to me on
[12:23:45] <Toneloc> it never writes to the mcu
[12:24:49] <specing> Toneloc: what crap did you get?
[12:25:16] <Toneloc> http://kartikmohta.com/tech/avr/programmer/avr_programmer.png
[12:25:34] <Toneloc> ^6a waste of bandwidth
[12:28:16] <specing> bandwidth?
[12:28:37] * RikusW is contemplating making a shift register with a 74LS374....
[12:29:48] <Toneloc> yes, the bandwidth I used from the image host
[12:29:53] <specing> RikusW: Take a PLD :)
[12:29:59] <Toneloc> it was a waste, because that circuit is useless
[12:30:05] * RikusW don't have any
[12:30:10] <specing> Toneloc: Are you on a dialup or what that you care?
[12:30:27] <RikusW> then connecting 3 together to make a 24bit address register for reading ROM chips
[12:30:35] <Toneloc> Im just joking, it was a way to say how worthless that circuit is
[12:30:46] <RikusW> I could easily use SPI to clock out the address...
[12:31:40] * RikusW is converting a rom based 2 way radio to and m8 controlled radio :)
[12:31:42] <specing> RikusW: but if you take a PLD you can reuse it for other purpuses after you are done with reading that rom
[12:32:19] <RikusW> specing: using that address register I could connect to ISA cards too
[12:32:19] <specing> RikusW: Also, I have an AT91SAM9260 that has 90 IOs exposed ;)
[12:32:50] <RikusW> maybe get an old CGA card and connect that to an CGA screen I have lying about..
[12:33:22] <RikusW> specing: That ARM is 3.3v right ?
[12:35:05] <specing> yeah
[12:35:22] <specing> Maybe it has 5V tolerant IOs, but Im not sure
[12:35:47] <specing> Also, many are connected to DRAMs
[12:36:02] <specing> Or not...
[12:36:05] <specing> Hmm, Im not sure
[12:36:20] <specing> they probably aren't
[12:36:42] <specing> exposing DRAM pins would be somehow pointless...
[12:38:08] <RikusW> I need 5V IO for that ROM and for ISA
[12:38:35] <RikusW> seems like MC14599 or 4099 might make nice port expanders
[12:38:40] <RikusW> * 4099B
[12:39:36] <RikusW> could make 1 port into 4
[14:16:03] <drgreenthumb> nice okw sent me some sample enclosures :)
[14:16:19] <drgreenthumb> didn't get the one I really wanted but these are still pretty neat
[15:03:11] <drgreenthumb> perfboard!
http://imgur.com/Kwn3P
[15:03:27] <drgreenthumb> bleh that stuff is hard to work with, that little thing took me about 3 hours.
[15:04:18] <drgreenthumb> but now least I have a nice permanent 802.15.4 controller that works over USB :)
[15:05:22] <OndraSter> congrats lol
[15:05:38] <drgreenthumb> hmm is it possible to get the USB AVRs into programming mode (HWB/RESET trick) in software? then I could just stick this in my closet ;)
[15:06:53] <drgreenthumb> probably need to build a delay circuit to make that happen.
[15:18:12] <drgreenthumb> heh this controller board would be a lot slicker if I just ran a board for it. it's got like 2 big parts, half dozen passives and connectors. it'd be really small.
[15:18:30] * drgreenthumb should do it
[15:20:27] <drgreenthumb> good thing I put the expensive parts of that perfboard in F->M headers so they just come right off :)
[15:20:35] <OndraSter> :)
[16:07:20] <drgreenthumb> heh I feel so organized right now
http://imgur.com/gnPTc
[16:07:37] <drgreenthumb> I can make anything! whatcha need. I'll make it.
[16:08:02] <OndraSter> breakfast?
[16:08:15] <drgreenthumb> ok one waffle bot coming up
[16:08:40] <Landon> can you make a catbot suitable for taking funny cat pictures?
[16:08:43] <specing> When having to pick between repeating 4 years of highschool and watching the X files, what would you choose?
[16:09:10] <Landon> specing: you mean that wasn't what you did for 4 years of highschool?
[16:09:14] <Landon> s/X files/futurama/
[16:21:25] <drgreenthumb> OndraSter, here's your breakfast bot.
http://imgur.com/YTJCM
[16:27:58] <OndraSter> lolthanks
[16:28:00] <OndraSter> CANNUCKS
[16:28:03] <OndraSter> CANNUCKED
[16:28:05] <OndraSter> tobbor, dead?
[16:40:16] <OndraSter> http://blog.xamarin.com/2012/05/01/android-in-c-sharp/
[16:40:16] <OndraSter> FYI
[16:40:22] <OndraSter> (to all android users :P)
[16:44:40] <Kevin`> interesting
[16:45:23] <Kevin`> i'd have to call that a good thing, it's unlikely they will get bound up in microsoft's incompatible libraries
[16:45:36] <Kevin`> although there's nothing wrong with native linux/c code either ;)
[16:46:41] <OndraSter> being MS fan I say that .NET is superior to everything when you _can_ do managed code
[16:48:10] <Kevin`> OndraSter: most .net programmers don't know how to code stuff to be cross-platform
[16:48:29] <jacekowski> .net is for morons that can't manage their own memory
[16:48:31] <jacekowski> like java
[16:48:32] <OndraSter> because .NET isn't cross platform really, MONO isn't MS's task
[16:48:51] <OndraSter> it was built with cross platform in their minds of course
[16:48:55] <jacekowski> well, microsoft was "selling" .net as portable
[16:48:58] <Kevin`> OndraSter: so now you should see why there could be a small problem using it =p
[16:49:07] <OndraSter> nobody cares about linux
[16:49:08] <OndraSter> :)
[16:49:11] <OndraSter> well, I don't :D
[16:49:14] <jacekowski> die
[16:49:17] <Kevin`> OndraSter: no, YOU don't care. plenty of other people do
[16:49:19] <OndraSter> will do
[16:49:27] <jacekowski> die quickly
[16:49:43] <OndraSter> even Atmel is now using Visual Studio which is (I suppose) built partially on .NET :P
[16:49:50] <Kevin`> OndraSter: almost all phones and higher end embedded devices (stuff that wouldn't use an avr) run linux or unix
[16:50:15] <OndraSter> my phone disagrees, the embedded stuff in our subway disagrees too
[16:50:25] <Landon> the oscope behind me agrees :P
[16:50:27] <OndraSter> :D
[16:50:29] <Kevin`> OndraSter: you are in the minority.
[16:50:39] <OndraSter> because windows is minority :(
[16:50:45] <Kevin`> OndraSter: for this, yes
[16:50:46] <specing> Landon: repeating as in learning for the finals
[16:50:53] <specing> Landon: I watched star trek!
[16:50:56] <Landon> haha
[16:51:01] <OndraSter> give me PROPER IDE for Java and linux
[16:51:03] <OndraSter> I dare you.
[16:51:16] <OndraSter> nothing can beat Visual Studio and C#
[16:51:16] <specing> OndraSter: Java? Have you gone mad?
[16:51:17] <Landon> I never did get into star trek, probably because the only times I tried watching were 10 or 11pm on saturdays
[16:51:22] <OndraSter> okay, Linux and C++
[16:51:35] <specing> vim :P
[16:51:39] <Kevin`> OndraSter: what, you can't use visual studio for linux projects?
[16:51:47] <OndraSter> when you talk cross platform, .NET is more cross platform than C++
[16:51:55] <Landon> specing: eheehhhhhhh, vim as an ide :\
[16:52:01] <OndraSter> lolvimasanidea
[16:52:03] <OndraSter> ide*
[16:52:12] <Landon> it _can_ be done, but I've never managed to get everything working right
[16:52:16] <OndraSter> kevin, sure you can, but you need to setup cygwin etc
[16:52:20] <Kevin`> OndraSter: except microsoft is behind .net and doesn't want it to be cross-platform. c has that support.
[16:52:29] <Landon> I wouldn't mind embedding vim in something like code::blocks though, that would be sweet
[16:52:35] <mimcpher> WHY BACK IN MY DAY, I WROTE FORTH IN VI ON A PDP-8
[16:52:59] <OndraSter> Kevin, MS doesn't want or doesn't care about nonMS platforms?
[16:53:16] <OndraSter> what the hell is "empty command" in C++? Some kind of NOP? :D
[16:53:42] <Landon> empty command?
[16:53:48] <Kevin`> OndraSter: ms cares a lot about non-ms platforms, they do their best to make it hard to use them if you stick with microsoft solutions
[16:54:08] <OndraSter> heh
[16:54:20] <OndraSter> they wrote HyperV code for linux kernel
[16:54:29] <OndraSter> but the kernel managers removed it or what
[16:54:35] <OndraSter> (no idea how you call them)
[16:54:57] <Kevin`> OndraSter: it's been cleaned up and put back in since. that's just driver code anyway, stuff you need to run linux guests on your hyper-v systems with reasonable performance
[16:55:20] <OndraSter> it's back already? yay
[16:55:22] <Landon> heh, wikipedia says they were caught with gpl code in there wrt hyperv drivers
[16:55:30] <OndraSter> eh
[16:55:45] <Kevin`> microsoft couldn't really compete with vmware,kvm,xen if guests were getting 10Mbytes/sec disk/network io
[16:56:59] <Kevin`> although you have to be a microsoft partner or newbie startup to benefit from hyper-v anyway, so maybe they could
[17:38:40] <iR0b0t1> What would be the best way to receive IR data? Like, it hits the optical receiver, then I get pulses out. I am doing 2400us header, 600us spacer, and 1200us hi/600us lo.
[17:39:02] <iR0b0t1> I'm using an atmega328 at this moment, but I hope to port this code over to an attiny of some variety.
[17:39:21] <iR0b0t1> (which would be dedicated to the task - so if the best way is a busy loop and some counting, it might be a viable option)
[17:39:39] <iR0b0t1> (but the attiny would be expected to maintain i2c connectivity)
[17:40:44] <Kevin`> iR0b0t1: attiny is quite capable of receiving and decoding ir
[17:41:15] <Kevin`> at least demodulated ir, not the raw analog signal.. that's what the receiver is for
[17:41:20] <iR0b0t1> I didn't think I wouldn't, but my previous experience with signal decoding involved heavy use of timers. Which is something I would expect to vary between the mega/tiny families.
[17:41:31] <iR0b0t1> Yeah, demodulated IR
[17:42:48] <Kevin`> so yours is pwm based, not manchester or something?
[17:43:19] <iR0b0t1> yeah. I would call it "pulse length"
[17:44:07] <iR0b0t1> I would have honestly picked something else, like ... one where the presence of a rising/falling edge in a time window denotes bit value
[17:44:12] <iR0b0t1> but I need to maintain compatibility
[17:44:48] <Kevin`> pulse length is pretty simple to decode though. just time pulses and compare to a threshold value
[17:46:34] <iR0b0t1> time with... timer?
[17:46:55] <iR0b0t1> What kind of interrupts are available for pin changes or timer compares?
[17:47:21] <iR0b0t1> I'm asking before I explicitly look through the datasheets because if someone here knows a attiny w/ or w/o certain features, I would like to know about it.
[17:48:55] <Kevin`> iR0b0t1: there's an input capture mode on some timers that would let you use non-time-critical code and get very accurate timing. not that the latter is necessary
[17:49:48] <Kevin`> iR0b0t1: most timers will have an overflow and arbitrary compare match interrupt
[17:49:59] <iR0b0t1> Hmm okay
[17:50:04] <Kevin`> all the modern chips have pin-change (non-timer function)
[17:50:39] <iR0b0t1> Yeah. I could have the timer running in the background, and use a pin-change interrupt. That might be the most convenient.
[17:50:47] <Kevin`> be careful using external interrupts, such that your code would still respond to i2c if someone stuck a 4mhz square wave into it
[17:51:03] <Kevin`> should be possible if done carefully
[17:51:45] <iR0b0t1> Well. Worst case the data decoding can be "reduced" (like actions recorded) to be gathered later if the interrupts do really take away from CPU time
[17:55:36] <Kevin`> you should always be doing that.
[18:11:54] <iR0b0t1> Kevin`: Know of a bit-bang I2C that might work on the likes of the attiny13?
[18:14:04] <jacekowski> DIY
[18:15:52] <iR0b0t1> Yeah, looks like I'm going to have to.
[18:19:18] <dfletcher_> iR0b0t1, you can use peter fleury's i2c master library
http://homepage.hispeed.ch/peterfleury/avr-software.html
[18:19:39] <dfletcher_> pretty tight, written in asm
[18:22:52] <Kevin`> iR0b0t1: master or slave?
[18:23:35] <drgreenthumb> yeah good question, slave is a different beast altogether.
[18:26:30] <CapnKernel> today's the day!
[18:26:59] <drgreenthumb> that you lose your virginity? ;p day for what?
[18:28:05] <CapnKernel> big tin fish of transcontinental relocation
[18:30:51] <specing> lol
[18:31:35] <specing> happy riddin'
[18:34:39] <iR0b0t1> Kevin`: slave
[18:34:42] <Toneloc> CapnKernel- what far flug regiona re you for?
[18:34:50] <iR0b0t1> df
[18:34:58] <iR0b0t1> guess he's not here, but thanks for the link to the master one
[18:35:01] <Toneloc> *flung region are
[18:35:10] <drgreenthumb> yeah iR0b0t1 I'm here (dfletcher). this my norm nick.
[18:35:20] <Toneloc> got to get better light in here..
[18:35:42] <drgreenthumb> slaves are hard to do in software if I remember rightly.
[18:35:54] <drgreenthumb> Kevin`, you have a different idea bout that?
[18:37:53] <iR0b0t1> drgreenthumb: Why is that? Must be vigilant?
[18:38:58] <drgreenthumb> yeah the master drives the clock. you must obey.
[18:39:08] <iR0b0t1> Well... hmm...
[18:39:12] <iR0b0t1> Does one exist ANYWHERE?
[18:39:24] <iR0b0t1> If I have to make my own I can probably make it a bit nonstandard
[18:42:42] <iR0b0t1> Well hrrmm...
[18:44:47] <Kevin`> iR0b0t1: why not use the usi interface (or better, twi, if your chip has it) ? it's designed to do this
[18:45:14] <Kevin`> doing it entirely in software is certainly possible, but it's hard real-time and would limit your maximum clock rate
[18:45:53] <Kevin`> (hard as in guaranteed timing, not hard as in difficult, but it could be that too)
[18:46:35] <iR0b0t1> Kevin`: ATtiny13 doesn't have one... I might have to move up to ATtiny45 though, so yeah
[18:48:38] <drgreenthumb> heh ok so now I have a low level wireless driver and Dean's CDC serial both working in the same program and prototype hw. and I realize I have no idea how to forward packets into the PC and route replies =) I'll keep it simple. request type, target addr, seq num, data.
[18:48:52] <OndraSter> CapnKernel, enjoy the ride :)
[18:49:49] <Kevin`> drgreenthumb: I have some code that does an ethernet link with usb-cdc-ethernet and some rf modules, if you want it, might be useful to look at
[18:51:55] <drgreenthumb> Kevin`, cdc ethernet? is that based on the RNDIS code in LUFA?
[18:52:08] <drgreenthumb> I think my AVR is too small for that, demo won't load.
[18:52:25] <drgreenthumb> wait let me try it again.
[18:52:35] <Kevin`> drgreenthumb: the demo code in lufa includes an ip stack and webserver, that may contribute somewhat to being large ;)
[18:53:56] <Kevin`> drgreenthumb:
http://kwzs.be/~kevin/usb-rfm.tgz
[18:55:05] <Kevin`> it's wonderfully nasty but it works quite well
[18:55:32] <Kevin`> comes from not planning the code beforehand. although that wouldn't have been reasonable to do
[18:56:41] <Kevin`> drgreenthumb: oh, the code there has two interfaces, cdc serial (for control/monitoring) and rndis ethernet (for data). apparently not cdc ethernet, although they are similar
[18:56:52] <Kevin`> dean is a windows guy, probably why he chose rndis ;p
[18:57:32] <drgreenthumb> heh yeah: Data: 4895 (478.0% Full)
[18:57:48] <drgreenthumb> ok I'll take a look thanks Kevin`
[18:58:25] <Kevin`> drgreenthumb: if you decide to build MY code, be aware that it includes stuff like stdio and floating point math, used entirely for debugging
[18:59:00] <drgreenthumb> heh well I might just cherry pick some ideas out of it. I don't mind writing tons of my own code ;)
[18:59:39] <Kevin`> http://pastebin.com/eCTfDQVi
[18:59:40] <Kevin`> :D
[19:00:31] <drgreenthumb> heh I think even 3659 is over my limit
[19:00:33] <Kevin`> there's one thing you may not be able to avoid though, you need at least one copy of an ethernet frame in ram, if you are using high level usb library
[19:00:35] <drgreenthumb> 2k IIRC
[19:00:43] <Kevin`> I have two copies in ram, one for sending, one for recieving
[19:00:53] <Kevin`> plus two of the physical layer packets at 64b
[19:01:20] <Kevin`> if you can't fit at least ONE 1500b packet in ram, your code will end up being "interesting" :)
[19:01:34] <Kevin`> *48b
[19:01:42] <drgreenthumb> well, it's actually 802.15.4 packets
[19:01:47] <drgreenthumb> which I think are a bit trimmer.
[19:01:57] <drgreenthumb> infact I think the max might be 255
[19:02:51] <drgreenthumb> PC side I really don't care about addressing or sequencing or whatever. that will just get copied out for reply.
[19:04:27] <drgreenthumb> heh yeah I should just have my AVR be very "blind" and just forward things.
[19:04:36] <drgreenthumb> no need to get all funky here :)
[21:16:38] <drgreenthumb> hrmf. dfu-programmer package on my ubuntu too old, no atmega32u2. latest source package refuses to build. upgrading ubuntu to 12.04 to see if it solves this. yay, yak shaving \o/
[21:28:30] <karlp> upgrading ubuntu to a whole new LTS, just to get a new dfu-programmer?!
[21:28:32] <karlp> you crazy mang
[21:28:41] <JoeLlama> mang (:
[21:43:56] <Tom_itx> drgreenthumb you remember the patch for that right?
[21:44:00] <Tom_itx> no need to update it
[21:44:31] <Tom_itx> iirc it was a one liner to the conf file
[21:45:04] <drgreenthumb> oh heh right
[21:45:22] <drgreenthumb> anyway, it's just my laptop. nothing's on there.
[22:12:32] <inflex> lo there Tom_itx
[22:13:22] <Tom_itx> hey
[22:13:38] <inflex> nothing exciting, just saying hello
[22:13:46] <Tom_itx> yup
[22:13:57] <Tom_itx> same here
[22:21:40] <inflex> Spent a few days going insane repairing a PC
[22:21:59] <inflex> had a fall, did some damage to the chassis... but the electronics damage was harder to pin down
[22:22:19] <inflex> in the end, a MXM connector (about 200 contacts, like an oversize SODIMM connector) was a big culprit
[22:22:42] <inflex> had to go through and manually readjust each contact to bring it back down to give it some spring-contact travel
[22:22:54] <Tom_itx> fun
[22:22:56] <inflex> (MXM is for the GPU/GFX cards)
[22:23:11] <inflex> http://dxp.me/i/mxm.jpg
[22:24:09] <inflex> There's contacts on both top and bottom... the bottom ones were easy, just an Xacto No.11 into the tiny hole under each pin
[22:24:15] <inflex> the topside was a fair bit harder
[23:02:13] <drgreenthumb> heh the latest ubuntu does have the newest dfu-programmer. nice.
[23:14:20] <drgreenthumb> hrm but I'm getting "no device present". it works fine in windows / flip so it's not the device. bleh.
[23:41:04] <iR0b0t1> dfletcher_: You are greenthumb right? Under linux I found I must run some things as root.
[23:41:10] <inflex> O_o
[23:41:14] <iR0b0t1> For example to use an AVRISP 2
[23:41:59] <Kevin`> alternately you can add udev rules so certain usb devices are usable by users
[23:42:20] <Casper> or suid it if you are lazy and understand the risk associated with it
[23:42:50] <iR0b0t1> Kevin`: I tried to do that but never found the group
[23:42:51] <Kevin`> I have mine to set up to work with a dragon and several of the usb bootloader ids
[23:42:59] <iR0b0t1> oh... by id
[23:43:18] <Kevin`> found the group?
[23:43:41] <dfletcher_> iR0b0t1, yeah tried it with sudo
[23:43:42] <iR0b0t1> Why would USB ever be permission restricted? I could understand having the ability to, but in practice wanting something restricted is the *exception*.
[23:44:00] <dfletcher_> yeah I've had to fiddle the udev settings for things.
[23:44:09] <dfletcher_> Tom_itx's programmer for one ;)
[23:44:17] <Kevin`> iR0b0t1: if you want to, it wouldn't be any harder to make the rule apply to ALL usb devices
[23:46:55] <Casper> iR0b0t1: actually under linux everything is restricted until given permission
[23:47:01] <Casper> which is why linux is so secure
[23:49:13] <iR0b0t1> then people like me get pissed off and use sudo too much
[23:50:50] <Casper> su -
