Back
[00:24:16] <Addy> anyone alive in here?
[00:41:36] <geckosenator> me
[00:45:34] <crazy_pete> not much up here addy
[00:45:45] <Addy> =\
[00:45:57] <Addy> I was going to ask you guys a question because #electronics was dead too
[00:46:01] <Addy> but I got it answered
[01:27:26] * tlvb is alive but has a cold
[03:15:04] <inflex> mmm... strange problems with some of my latest PCBs
[03:15:17] <inflex> if I put them through reflow the "bare pads" become almost unsolderable
[03:22:16] <ziph> Bare pads?
[03:25:06] <RikusW> oxidization ?
[03:34:29] <RikusW> inflex: what plating was used on it ?
[03:35:18] <ziph> Must be that new faux gold they're using. ;)
[03:36:32] <inflex> supposed to be HAL
[03:36:48] <RikusW> HASL ?
[03:47:49] <RikusW> Steffann: Fixed my jtag mki to support flash programming on newer avrs too :)
[03:51:23] <inflex> HAL = hot air levelling
[03:51:43] <inflex> the boards are getting some a dark-grey hue over them, some even before I put them in the reflow
[03:52:00] <inflex> the reflow/solder seems to work fine (in the oven) it's near impossible to solder with normal station + solder
[03:52:40] <RikusW> apply flux to all pads ?
[03:52:55] <inflex> not prior to reflow. Never have had to before
[03:55:25] <Mithrandir> hmm, so how likely is it that I've managed to brick my stk500? It seems to refuse talking to avrdude and just goes: "stk500_program_enable(): failed to enter programming mode".
[03:55:40] <Mithrandir> and, is there a known workaround? Try updating the firmware?
[04:02:03] <mrfrenzy> well then you just need to update the builtin firmware
[04:02:07] <mrfrenzy> I think it has a bootloader
[04:02:24] <mrfrenzy> just follow instructions in the manual and use avrstudio
[04:11:44] <Mithrandir> ok, I'll try that. Thanks.
[04:13:17] <RikusW> Mithrandir: did it happen after trying to update fw ?
[04:13:48] <RikusW> could be that you didn't connect the target avr properly
[04:16:26] <Mithrandir> RikusW: it happened when I was flashing an attiny, yes.
[04:17:51] <RikusW> but not updating the actual stk500 fw...
[04:17:55] <RikusW> so that should be fine
[04:18:15] <RikusW> more likely that the connection is not right
[04:18:27] <RikusW> do you use offboard target ?
[04:19:43] <Mithrandir> I've reseated the attiny85, tried with another one and with an atmega32l as well as redone all connections about five times, so I doubt the connections are off.
[04:20:30] <Mithrandir> also, it happened between two flashings where I did not touch the board.
[04:20:42] <mrfrenzy> ok, so first of all, does avrstudio identify your stk?
[04:20:58] <mrfrenzy> from your description the firmware should be unharmed
[04:21:22] <RikusW> Mithrandir: did you change the tiny85's fuses ?
[04:21:28] <Mithrandir> RikusW: not at that time, no.
[04:21:37] <Mithrandir> I'm running linux, so I haven't tried with avrstudio yet.
[04:22:02] <Mithrandir> (I do have a windows box, I just need to move a power supply and the stk itself and wire it all up to test with avrstudio)
[04:22:20] <RikusW> Mithrandir: do you use ribbon cables for connecting the isp headers ?
[04:22:27] <Mithrandir> yes
[04:22:35] <RikusW> and are you sure you connected the right colored one ?
[04:23:43] <Mithrandir> yes. (And I did change which one was connected when I tried the atmega32l)
[04:24:04] <Mithrandir> https://gist.github.com/1395416 is the output from avrdude, where it looks like it identifies stuff correctly.
[04:24:33] <RikusW> vtg jumper connected ?
[04:25:57] <RikusW> and what PSU do you connect to the stk500 ?
[04:26:05] <Steffann> tiny85 :)
[04:26:22] <Mithrandir> RikusW: vtarget, aref, reset, xtal1 are all connected.
[04:26:52] <Mithrandir> I use a 500W PC power supply where I've pulled the 12V line (and GND) into a connector that fits the STK500
[04:27:02] <Mithrandir> so I should have plenty of juice. :-)
[04:27:50] <Steffann> Some people say you shouldn't use a pc power supply for things like this..
[04:30:29] <Mithrandir> hm, why is that?
[04:31:08] <RikusW> Mithrandir: from stk docs for t85
[04:31:16] <RikusW> SPROG1.
[04:31:16] <RikusW> Connect RST on PORTE to PB5 on PORTB.
[04:31:17] <RikusW> Connect XT1 on PORTE to PB3(XTAL1 on 2323) on PORTB.
[04:31:22] <Mithrandir> yup, done that.
[04:31:57] <Mithrandir> what I find odd is it worked, then suddenly broke without me changing anything on the board.
[04:33:16] <RikusW> Is PJUMP and BSEL2 jumpers removed ?
[04:34:07] <ziph> A plug pack works fine for the STK500.
[04:34:36] <ziph> And you should get a lab supply anyhow so that you can test things without having them completely obliterated in the event of a problem.
[04:35:09] <Mithrandir> BSEL2 is connected, but it's been connected the whole time.
[04:35:14] <mrfrenzy> no problem in this case since the stk500 has internal regulator
[04:36:01] <RikusW> Mithrandir: remove it, its only for some avrs during HVPP
[04:36:07] <ziph> That won't help when all the tracks burn off. :)
[04:37:37] <RikusW> Mithrandir: try connecting ISP directly to the t85 on breadboard ?
[04:37:46] <mrfrenzy> the regulator limits current to 1A
[04:38:51] <Mithrandir> RikusW: yeah, that's my next step, basically. Or try to find out where I've put my jtagice.
[04:39:04] <Mithrandir> I just wonder if I've fried the stk or not.
[04:39:12] <Mithrandir> so maybe try with avr studio too
[04:39:39] <RikusW> jtag mkii ?
[04:40:39] <ziph> mrfrenzy: Until you cook it.
[04:41:06] <Mithrandir> yeah, JTAGICE mkii.
[04:41:21] <Mithrandir> but it's either in a box or I've lent it to somebody :-/
[04:42:04] <RikusW> you have any other avrs around ?
[04:42:41] <Mithrandir> I've got a tube with 10 attiny85s in it, and an mega32l
[04:43:21] <RikusW> and the other t85's wont work either ?
[04:43:51] <Mithrandir> I've only tried one other, without luck.
[04:44:12] <RikusW> what programmer type do you use with avrdude ? -c stk500 ?
[04:44:25] <Mithrandir> -c stk500v1
[04:44:40] <Mithrandir> I haven't ever upgraded the stk since I got it in, uhm, 2003 or 2004?
[04:45:21] <RikusW> and you used SCKT3400D1 (blue) ?
[04:45:41] <Mithrandir> yes
[04:48:04] <RikusW> I do have the actual stk500 around (decrypted) for both at90s1200 (bootloader) and 8535, so if you ever need to repair I can help
[04:48:12] <RikusW> lets hope its not broken
[04:48:40] <RikusW> actually only the 1200 fw is required
[04:49:10] <Mithrandir> ok, thanks.
[04:50:09] <RikusW> but the 8535 will be nearly impossible to find...
[05:28:17] <ElMonkey> RikusW, problem resolved by attaching an external powersupply to the butterfly
[05:28:27] <ElMonkey> apparently the battery is too weak for ISP
[05:34:32] <ben1066> Say I have two avrs that I want to run both at 16mhz
[05:34:42] <ben1066> Is there anyway I can do that with one clock source?
[05:35:30] <ben1066> they will be doing usart so it has to be fairly accurate
[05:35:58] <Steffanx> Enable clock out of one avr?
[05:36:07] <tlvb> hey! I was going to say that
[05:36:28] <ben1066> can I input that straight into the other avr? would I put it to XTAL1 or XTAL2?
[05:36:33] <ben1066> or both?
[05:36:41] <tlvb> use it as an external clock
[05:36:56] <ben1066> Ah ok
[05:37:00] <RikusW> ben1066: enable CKOUT on one
[05:37:08] <RikusW> and select external clock on the other
[05:37:17] <RikusW> using the fuses
[05:37:27] <RikusW> then connect ckout -> xt1
[05:37:48] <ben1066> I cant double the clock can I?
[05:38:00] <RikusW> no
[05:38:20] <RikusW> but you might be able to divide it by 2 on the receiving avr
[05:38:28] <RikusW> using the clock prescaler
[05:38:36] <ben1066> Hm ok, cause itd be out of spec otherwise :P
[05:38:56] <ben1066> Albeit, its hardly as if Im going to sell this
[05:39:01] <ben1066> or use eeprom
[05:39:25] <RikusW> eeprom use the internal RC afaik
[05:39:57] <ben1066> Oh ok, I heard using out spec clocks can affect eeprom
[05:40:29] <ben1066> Im currently using 20mhz on my 168, which is the max specced, although 24mhz should be ok, it used to be specified, and swinSID uses 32mhz and seems to be ok
[05:42:30] <ben1066> I need to buy some stuff before I can do any more on my project anyway
[05:42:43] <ben1066> I need a ps/2 keyboard, mini din socket phono socket and some ram :P
[05:46:27] <RikusW> what are you building ? an overclocked PC ? :-P
[05:47:12] <ben1066> Keyboard is most important, I need some input device :p
[05:47:45] <ben1066> and the only spare keyboard I have is USB
[05:50:15] <Steffanx> Go ARM if you need more speed ben1066 :P
[05:50:17] <Steffanx> or maybe xmega
[05:51:16] <ben1066> I quite like the DIL parts though, I dont really want to have to make PCBs
[05:51:29] <Steffanx> Why not?
[05:51:33] <Steffanx> Seeed is cheap
[05:51:43] <ben1066> Yes, but takes forever
[05:51:57] <ben1066> + if you make a mistake, FUUUU
[05:52:12] <Steffanx> Then you lost 10$ .. who cares :P
[05:52:36] <ben1066> Yea, its not the money
[05:52:39] <ben1066> Its the time I have to wait
[05:52:56] <ben1066> I can get perfboard/stripboard and parts next day from farnell
[05:52:56] <Steffanx> And you can make/buy some DIL board for the chips too ..
[05:53:17] <ben1066> Steffanx: 100 pin DIL doesnt sound fun :p
[05:54:16] <Steffanx> It should better than overclocking an AVR
[05:54:47] <Steffanx> sounds :P
[05:54:50] <Steffanx> not should
[05:55:36] <ben1066> :S
[05:56:19] <Steffanx> No not :S
[05:56:32] <ben1066> cheapest avr about 70p
[05:57:47] <Steffanx> Yes?
[05:57:49] <ben1066> cheapest arm in a solderable package
[05:58:00] <ben1066> £1.80
[05:58:27] <ben1066> and thats only a 24mhz arm :P
[05:59:19] <Steffanx> True, but now you compare devices which can't be really compared
[05:59:54] <ben1066> True
[06:00:07] <ben1066> I also dont have a programmer for ARM or AVR32
[06:00:43] <ben1066> and xmegas seem massively overpriced
[06:00:58] <karlp> what do you consider to be a "solderable package" for arm?
[06:00:59] <Steffanx> The stm32 discovery board are nice and pretty cheap :)
[06:01:06] <Steffanx> *boards
[06:01:07] <ben1066> £10
[06:01:29] <ben1066> that programmer only supports ST arms no?
[06:01:48] <Steffanx> I don't know
[06:01:53] <karlp> so they say, yes
[06:02:06] <ben1066> karlp: not BGA or something where all the pads are underneath the package
[06:02:34] <karlp> when you said, "cheapest arm in a solderable package: 1.80" what package was that? (and what part?)
[06:02:49] <ben1066> STM32F100R6T6B
[06:02:53] <ben1066> TQFP
[06:02:59] <ben1066> LQFP*
[06:03:01] <karlp> cheapest avr at 70p can be a hell of a lot less than an arm.
[06:03:07] <karlp> that's only just solderable,
[06:03:12] <karlp> it's tqfp at 0.5mm pitch,
[06:03:21] <karlp> not 0.8mm pitch like an atmega328 in tqfp
[06:03:29] <ben1066> and not STM32F100 blah STM32F100R4T6B
[06:04:15] <ben1066> karlp: your just proving my point further, that arms are a PITA to solder :p
[06:04:25] <karlp> yeah, but they do more :)
[06:04:30] <Steffanx> No, it's not that hard
[06:04:39] <karlp> I don't intend to do much more dev work with avrs
[06:04:45] <karlp> they're not cost effective anymore
[06:04:54] <karlp> the soldering is just a technique,
[06:04:55] <Steffanx> I do.. for things that don't need a big ass arm :P
[06:04:58] <ben1066> If I could find something thats as accessible I would consider switching
[06:05:08] <inflex> I guess cost-effective is all relative
[06:05:30] <ben1066> But there arent nice IDEs (hint I HATE ECLIPSE), no universal programmers and they are harder to prototype
[06:05:41] <ben1066> Especially when you cant etch your own PCBs
[06:05:42] <karlp> looking forward to the soic20 arms from NXP
[06:05:51] <karlp> there are universal programmers,
[06:06:00] <karlp> dozens and dozens and dozens of them
[06:06:09] <karlp> IDEs are relative,
[06:06:13] <karlp> what IDE do you want?
[06:06:18] <karlp> it's all just C/C++ normally,
[06:06:23] <karlp> the target is irrelevant
[06:06:35] <Steffanx> And you can use that buggy st-link util tool from texane for the stm32 discovery boards :P
[06:07:10] <ben1066> Something that doesnt use Eclipse, I like AVRStudio since it uses visual studio, which im used to, it all, from my perspective, fits together well and it just works
[06:07:13] * karlp spanks Steffanx
[06:07:27] <ben1066> I mean, I click compile, and I can simulate, or program from the same place
[06:07:32] <Steffanx> Don't say it's not buggy karlp
[06:07:38] <karlp> I'm not :)
[06:07:44] <karlp> but bugs are getting fixed pretty well I'd say,
[06:07:45] <Steffanx> ok ok
[06:07:52] <karlp> it's got plenty of warts,
[06:07:58] <karlp> but it's a hell of a lot better than nothing.
[06:08:02] <Steffanx> I still have to find a way to use it on os x
[06:08:15] <Steffanx> I wouldn't be surprised if that really a pita
[06:08:37] <karlp> well, I did all I could by getting rid of the sg-utils stuff, but the rest of it needs someone actually on osx
[06:08:47] <karlp> and it seems not many people care about the VL board much now.
[06:09:04] <karlp> it is cheap though :)
[06:09:10] <ben1066> and ok then. what programmers are there that dont cost £100 :P
[06:09:23] <ben1066> or a stupid amount
[06:09:24] <Steffanx> The versaloon
[06:09:33] <karlp> dangerous prototyps bus blaster, versaloon, any of the olimex usb jtag things
[06:09:43] <Steffanx> The shipping costs of the versaloon suck though
[06:09:56] <karlp> versaloon rubs me the wrong way, and I can't really explain it.
[06:10:05] <Steffanx> ?
[06:10:07] <karlp> I don't know why I don't really like it.
[06:11:07] <Steffanx> You have a versaloon?
[06:11:10] <karlp> no
[06:11:18] <karlp> maybe I just don't like their website
[06:11:42] <ben1066> development packages: GCC C compiler, openOCD debugger and Eclipse IDE.
[06:11:43] <Steffanx> The website is … bad, true
[06:11:48] <ben1066> oh look at that, eclipse >_>
[06:12:01] <Steffanx> ben1066 ?!
[06:12:03] <karlp> what are you looking at ben1066 ?
[06:12:17] <ben1066> USB-TINY-H omilex
[06:12:27] <ben1066> its the cheapest one I can find in the UK
[06:12:34] <karlp> the programmer has nothing to do with using eclipse
[06:12:34] <ben1066> £33, still more than the AVRISP mk2
[06:12:47] <ben1066> its the IDE that it has support with aparently
[06:12:51] <karlp> yeah, and how often do you buy a programmer?
[06:13:12] <Steffanx> 15 times/year
[06:13:17] <ben1066> once until you find it cant program something cause its annoying and has its own special thing
[06:14:18] <ben1066> i wish less people used eclipse, everything seems to use it or be based on it
[06:14:23] <ben1066> I really cant stand it :S
[06:14:36] <Steffanx> Eclipse isn't that bad imho
[06:14:57] <ben1066> I dont like the UI, it feels clunky
[06:17:13] <crazy_pete> i don't like IDE's in general but that makes me a dying breed
[06:17:37] <karlp> crazy_pete: cscope/ctags and vim/emacs?
[06:17:38] <ben1066> id rather use notepad++ than eclipse :S
[06:17:59] <karlp> I wish jetbrains made a C tool.
[06:18:06] <karlp> or that netbeans still supported python...
[06:18:13] <ben1066> What the hell
[06:18:15] <ben1066> error deleting "C:/ti/ccsv5\eclipse\configuration\org.eclipse.osgi\bundles\136": file already exists
[06:18:15] * Steffanx waits for the next rant
[06:18:30] <ben1066> Yea right code composer doesnt like uninstalling
[06:18:51] <karlp> Steffanx: no, I think I've finished what I was doing on the net, time to spend some quiet time actually tryign to _use_ one of these chips for a change...
[06:19:41] <crazy_pete> vim karlp :)
[06:19:50] <crazy_pete> gvim actually :)
[06:19:50] <Steffanx> emacs!
[06:20:06] * crazy_pete loves all the emacs/vi jokes
[06:20:26] <crazy_pete> (my favourite is still VI Editor = Very Irritating Editor) :)
[06:20:36] <inflex> vim++
[06:20:55] <tlvb> eight megabytes and constantly swapping
[06:23:17] <Steffanx> That's irsi not?
[06:23:32] <Steffanx> irssi
[06:24:14] <ben1066> I use irssi for IRC :D
[06:24:19] <k-man> anyone tried out the sparkfun quadstepper driver board?
[06:24:47] <k-man> me too
[06:24:52] <k-man> irssi is good
[06:33:53] <inflex> heh
[06:34:07] * inflex admits to using Xchat... but I use TTYtter :D
[06:35:02] <RikusW> sounds almost like twytter ;)
[06:36:29] <Steffanx> My client is even worse inflex .. xchat aqua :P
[06:41:58] <inflex> oooh, so sick
[06:43:24] <Valen> pidgin ftw!
[06:43:37] <RikusW> gaim ftw ;)
[06:44:07] <crazy_pete> you sick or just commenting on xchat inflex ?
[06:44:07] <Valen> isnt gaim ~ pidgin or something?
[06:44:22] <RikusW> Valen: old pidgin yes
[06:44:27] <RikusW> very old
[06:44:38] <Valen> beh they havent made it work better
[06:44:43] <Valen> still no msn file transfers
[06:44:45] <Valen> even slow ones
[06:45:46] <RikusW> mine occasionally segfaults when doing nickserv commands
[06:45:59] <RikusW> sometimes even when starting up...
[06:46:25] <inflex> crazy_pete: referring to Xchat Aqua :p
[06:48:23] <k-man> where can I read up on timer interrupt routines in avr?
[06:48:42] <crazy_pete> the data sheets ?
[06:49:06] <crazy_pete> what AVR are you looking to program k-man and in what language? (C, C++, ASM, etc?)
[06:49:25] <RikusW> k-man:
http://tom-itx.dyndns.org:81/~webpage/
[06:49:29] <k-man> crazy_pete, atmega 1280 in c
[06:50:03] <k-man> I want to write a proper library to drive the sparkfun quadstepper board
[06:50:04] <inflex> k-man: at the end of each datasheet, there's a section on all the instructions and timings
[06:50:15] <crazy_pete> which IIRC is basically a an atmega128
[06:50:26] <k-man> inflex, what were you saying are the difficulties with interrupts in AVR?
[06:50:35] <crazy_pete> got a url for that board k-man? (I would like to see it)
[06:51:09] <crazy_pete> basically you should read the intterupt and the timer section of the data sheet for the avr128
[06:51:09] <k-man> crazy_pete,
http://www.sparkfun.com/products/10507
[06:51:14] <crazy_pete> ty k-man :)
[06:51:21] <inflex> k-man: basically don't assume interrupts won't mess with timings elsewhere. If you have to be precise with your pulses/timings, then use the timer's / PWM / overflows etc
[06:51:55] <crazy_pete> that is cool k-man :)
[06:51:59] <k-man> crazy_pete, I also uploaded their library code to my github:
https://github.com/jasonblewis/quadstep
[06:52:08] <k-man> crazy_pete, the library is not cool howerver
[06:52:12] <crazy_pete> thanks k-man :)
[06:52:15] <k-man> at least, doesn;'t look cool to me
[06:52:26] <crazy_pete> yeah actually interrupts WILL always mess with your timings
[06:52:36] * inflex isn't much of a fan of C++ for uC programming
[06:52:38] <k-man> crazy_pete, which timings?
[06:52:47] <crazy_pete> any timing you do outside of an interrupt
[06:53:02] <inflex> k-man: consider the worst case, where you're servicing an interrupt... and another interrupt interrupts :D
[06:53:06] <crazy_pete> for example if you have a delay loop that is calculated to delay for 100uS
[06:53:16] <inflex> (you can of course disable interrupts within)
[06:53:18] <crazy_pete> and an interrupt fires and takes 10uS to complete
[06:53:22] <k-man> inflex, yeah, strange choice to use c++, and if you look at the code afaict, it doesn't really do anything that you couldn't do in c
[06:53:28] <crazy_pete> your delay loop will actually run for 110uS before it returns
[06:53:34] <crazy_pete> see what i mean k-man?
[06:53:44] <crazy_pete> lol C++ is such a boon and godsend for uC development
[06:53:51] <k-man> crazy_pete, yeah, makes sense
[06:53:57] <inflex> crazy_pete: how do you see that?
[06:54:05] <inflex> crazy_pete: other than pushing the need for larger/faster uC's ?
[06:54:21] <crazy_pete> well for starts, in general C++ results in slightly smaller binaries
[06:54:57] <crazy_pete> although that really isn't a big reason to do it (an avr-g++ program is about 100 bytes smaller than the same C program
[06:55:20] <inflex> crazy_pete: mm... got any references for that?
[06:55:35] <inflex> I mean, did we go from 16.1K to 16.0K ?
[06:55:42] <crazy_pete> the real reason i like C++ is the IO facilities (iostreams beat printf to hell, i know a guy how is about 1000x smarter than me but uses C and is always running into printf problems
[06:55:52] <crazy_pete> something like that inflex it is not a big difference
[06:56:14] <inflex> well, sounds more like a personal choice thing, never had problems with printf's or such - I suppose it depends a lot on what people are doing
[06:56:24] <crazy_pete> but compile "hello.c" as "hello.cxx" and see which is smaller binary :)
[06:56:29] <inflex> I find a lot of people come from PC programming and try applying similar principles to uC's... just doesn't work out
[06:56:32] <k-man> inflex, have you ever heard of a book called Deep C Secrets?
[06:56:39] <crazy_pete> lol if you haven't had problems with printf, you haven't used it enough
[06:56:53] <inflex> crazy_pete: no, I think I an a good anecdotal point against that.
[06:57:22] <crazy_pete> my friend is always converting a float to a double in his code or converting an int to an unsigned and forgetting to change the format specifier
[06:57:33] <crazy_pete> i don't know what your last sentence means inflex
[06:57:43] <inflex> yeah, see, that's precisely what I mean... it's PC people coming into a uC world
[06:57:45] <crazy_pete> with C++ you don't have that problem
[06:57:53] <crazy_pete> i have no idea what you are talking about inflex
[06:58:05] <crazy_pete> my friend is a total uC programmer with 10 years experience
[06:58:14] <crazy_pete> never programmed a PC in his life
[06:58:36] <inflex> Then why is he using floats and doubles... rather than doubles outright or just going 32bit fixed point int?
[06:58:41] <crazy_pete> still i don't see what you are saying
[06:58:41] <inflex> you just don't mix floats and doubles
[06:58:57] <crazy_pete> if you need to save space and later need the precision you do
[06:59:19] <crazy_pete> to say "this shouldn't change" is generally bad , but C enforces that sort of thinking
[06:59:36] <crazy_pete> or changing a signed to an unsigned
[06:59:46] <crazy_pete> and forgetting to change the %d to %u
[06:59:57] <crazy_pete> that nasty bug goes away in C++
[07:00:08] <inflex> Ah, so C++ is good for people who aren't good for remembering to be strict.
[07:00:15] <crazy_pete> still don't know what :no, I think I an a good anecdotal point against that." means
[07:00:27] <crazy_pete> C++ is good for people who know what programming is about :)
[07:00:33] <crazy_pete> which is defensive coding
[07:00:35] <inflex> I'd say a lot of people lose those 100bytes gain or "better binary size" with their lax programming
[07:01:00] <crazy_pete> yeah well if C++ generates smaller binaries to begin with, i will take advantage of that
[07:01:08] <inflex> It means that I have been doing C a very long time and I've been using printf for all manner of things... and I just don't run into those "problems".
[07:01:25] <crazy_pete> given the huge number of bugs that C++ eliminates i will also take it over C
[07:01:37] <inflex> Sure, fair enough; select C++.
[07:01:42] <crazy_pete> no offence , inflex but every C programmer says that, and then you look at their code and it is full of just the bugs they say they avoid :)
[07:01:53] <inflex> However a lot of the time people end up coding C in C++ thinking that C is a subset of C++.
[07:01:53] <crazy_pete> printf being just the tip of the iceberg
[07:02:41] <inflex> crazy_pete: sure - I'm sure there's bugs in my code, I know there is; feel free to take a look at the stuff I've done.
[07:02:46] <crazy_pete> i am having trouble following your sentences inflex, i should warn you that i am unusually sick (cough, etc) something i am not used to
[07:02:52] <crazy_pete> sure inflex
[07:03:02] <inflex> http://pldaniels.com/opensource-projects.php
[07:03:35] <inflex> What I was saying before (re C and C++) is that a lot of people are actually coding a mangled mess of C and C++ and thinking that it's "C++"
[07:04:09] <crazy_pete> nice inflex
[07:04:11] <karlp> bad code exists in any language,
[07:04:14] <karlp> can we move on now?
[07:04:17] <crazy_pete> yeah i have seen that a lot inflex
[07:04:20] <inflex> indeed
[07:04:32] <crazy_pete> something i like to do a lot, especially with uC is mix C++ and assembler
[07:04:37] <specing> crazy_pete!
[07:04:47] <crazy_pete> what specing?
[07:05:13] <crazy_pete> since asm functions look like C functions to C/C++ code
[07:05:16] <specing> I see you are still alive
[07:05:32] <crazy_pete> ( me thinks that avr assembler is way too cool and very underappreciated :) )
[07:05:36] <inflex> At least when you're putting asm into C or C++ you know you're doing it.
[07:05:44] <k-man> inflex, check out this book:
http://goo.gl/IoVTm imho, it is very interesting, and a great read on its own, but also very informative
[07:05:49] <crazy_pete> yes specing
[07:06:04] <specing> crazy_pete: are you coming around to social any time soon?
[07:06:10] <crazy_pete> nope
[07:06:18] <specing> Why not?
[07:06:23] <crazy_pete> i promised blank i wouldn't return due to his homophobia
[07:06:34] <crazy_pete> or whatever his problem is with me
[07:07:01] <crazy_pete> (he was really mad about something one day, and as a joke, somebody told me he had just been dumped by his boyfriend)
[07:07:17] <inflex> k-man: haven't read that one, though my shelves are full of lots of older C books
[07:07:35] <crazy_pete> not knowing the person was teasing us i tried to comfort blank and now blank hates me forever because "you accused me of being gay"
[07:07:37] <crazy_pete> sheesh !!!!
[07:07:52] <crazy_pete> but back to avr assembler ..... :)
[07:08:11] <crazy_pete> i am trying to put together some avr classes in san francisco at one of the free schools
[07:08:15] <inflex> surprisingly one of the nicer C books I read was by an unknown Australian lecturer "Comprehensive C" by David Spuler... though admittingly the C from that was from the late 1980s - but still very nice read
[07:08:18] <crazy_pete> anybody in the area ?
[07:08:32] <ziph> inflex: Don't you have to explicitly re-enable interrupts to get them nested on an AVR?
[07:08:35] <k-man> inflex, interesting
[07:09:01] <crazy_pete> ooh good question ziph, i think you're right
[07:09:40] <inflex> http://www.amazon.com/Comprehensive-C-David-Spuler/dp/0131565141 <=- *sigh* hard to get it now
[07:10:01] <inflex> ziph: yerp, I do believe you're right
[07:10:30] <ziph> inflex: The Cortex M3 interrupt support is very, very nifty. You have interrupt groups and then priorities within those groups.
[07:10:50] <ziph> inflex: An interrupt of a higher group will interrupt if a lower group interrupt is running.
[07:10:52] <inflex> ziph: that is nice having prioritised interrupts
[07:11:17] <ziph> inflex: An within a group the priority just sets which one will run first when more than one is armed.
[07:11:31] <crazy_pete> that is arm or avr ziph?
[07:11:39] <ziph> ARM Cortex M3.
[07:11:40] <inflex> though I suppose with the AVR you can always just monitor the flags on things like the timer-overflows rather than letting them interrupt... but yeah
[07:11:56] <crazy_pete> (i know cortex is arm but i don't remember what the nesting is on an AVR, i thought the int enable flag was cleared on interrupt entry ?
[07:12:11] <specing> crazy_pete: blank is not there
[07:12:17] <crazy_pete> its his channel
[07:12:23] <crazy_pete> and i promised him i would never return
[07:12:29] <ziph> inflex: And also, they optimise saving registers; if you've just finished servicing and interrupt and another needs to be handled, the registers aren't restored between.
[07:12:43] <specing> crazy_pete: oh come on
[07:12:47] <crazy_pete> ok
[07:12:47] <inflex> Ugh, why do the people you want to find most never join Facebook!?
[07:13:08] <ziph> inflex: Because only people you wouldn't want to associate with use Facebook? ;)
[07:13:10] <k-man> oh great, I think my Mac book pro's power supply just died
[07:13:11] <Kevin`> inflex: to me that's obvious
[07:13:34] <inflex> yeah, they know I'm STALKING THEM!!!!
[07:13:42] * crazy_pete bristles at "I find a lot of people come from PC programming and try applying similar principles to uC's... just doesn't work out"
[07:13:52] <inflex> One day they'll make a mistake and I'll be cyber-stalking their brains out
[07:14:05] * crazy_pete finds that bad programmers can fake it on a PC but can't get away with their crap on a uC :)
[07:14:06] <inflex> crazy_pete: hey, feel free.
[07:14:31] <inflex> lots of programmers just are too used to shooting fat and fast with their coding
[07:14:45] * crazy_pete started out doing uC in C++/Assembler and moved to PCs later
[07:15:02] <inflex> Too used to "Meh, there'll be a module that'll do that for me and the system will cope"
[07:15:09] <crazy_pete> yeah
[07:15:26] <crazy_pete> although honestly i am surprised that hardware programmers don't take more advantage of C++
[07:15:57] <crazy_pete> (for example there is some initialisation/release code needed to switch ADC channels on an avr)
[07:16:10] <inflex> maybe because a lot of us just don't have the -need-. Really, in a 1K flash, if you can't do a nice job in C or ASM, then you probably should just pack up and go sell papers.
[07:16:24] <ziph> C++ has some annoying runtime requirements.
[07:16:24] <crazy_pete> by putting that code in the constructor/destructor you get it called automatically
[07:16:33] <ziph> A C++ without any of the additional runtime requirements would be nice.
[07:16:40] <crazy_pete> i dunno i keep seeing C++ programs being smaller than C ziph
[07:17:05] <ziph> Smaller? Faster maybe, but I can't imagine smaller in general.
[07:17:13] <inflex> crazy_pete: I want to see some good references on that - sure there'll be some corner cases, but I'm not thinking it's overall.
[07:17:15] <Kevin`> crazy_pete: only with the same code
[07:17:45] <crazy_pete> well with the same code is close enough for me
[07:17:46] <Kevin`> crazy_pete: and that constructor stuff isn't gonna help you when you want to efficiently switch channels instead ;p
[07:18:00] <crazy_pete> how the hell is a constructor not going to help when you want to switch channels?
[07:18:31] <crazy_pete> if nothing else you need to reinit the admux register when switching channels and the constructor can do that for you
[07:18:32] <Kevin`> because it implies blocking polling and switching the channels on demand instead
[07:18:44] <crazy_pete> huh ?
[07:19:00] <crazy_pete> in some cases that is appropriate
[07:19:18] <inflex> but that's the crux isn't it - you're now letting the compile assume what is appropriate more and more often
[07:19:23] <crazy_pete> when would you switch the channles
[07:19:28] <crazy_pete> what are you talking about
[07:19:35] <crazy_pete> when wouldn't it be appropriate
[07:20:11] <crazy_pete> i wouldn't call a constructor from within an interrupt or do anything else non deterministic from within an interrupt (for that matter i generally write ISR in asm , not C, especially not C++)
[07:20:35] <inflex> why not in C++ ?
[07:20:46] <inflex> why write your ISR's in asm at all?
[07:21:18] <ziph> Bah, the only difference between an interrupt and normal code is that an interrupt runs with slightly different status register flags. ;)
[07:21:34] <crazy_pete> and timing can be extremely critical in an ISR
[07:21:50] <ziph> Why? :)
[07:21:55] <crazy_pete> lol ziph
[07:21:56] <Kevin`> can be, but normally c is good enough if you don't do things that take indeterminite time
[07:22:26] <ziph> I'm serious, timing isn't critical in an ISR. :)
[07:22:33] <inflex> Not all ISR's are timing critical
[07:22:40] <inflex> very few truly are.
[07:23:59] <inflex> But I find it interesting that you'd like to see us use C++, yet you yourself want to stick with ASM for various parts *shrug* What about all those poor bastards that start using C++ wholesale and then wonder wtf things just don't work quite as right?
[07:24:37] <inflex> So what it comes down to in the end is that you have to actually know what you're doing first.
[07:26:53] <tlvb> One ISR I have made had to run constantly in less than X cycles in the worst case scenario, so it was very much time critical
[07:28:16] <inflex> for sure, there will be some ISRs that genuinely have to run within a hard real-time limit; for most they're merely a nice convenience that saves us doing annoying busy-wait loops :D
[07:29:19] <tlvb> of course, I was just providing one counter example to the statement that isr's are not timing critical
[07:29:22] <inflex> mmm... 1.5hrs before the launch
[07:29:33] <tlvb> launch?
[07:29:57] <inflex> Yes, NASA's Curiosity mars rover
[07:30:03] <inflex> http://www.ustream.tv/nasahdtv
[07:30:19] <tlvb> ah
[07:30:48] <inflex> it looks kind of small doesn't it.... on the launch pad... but I'm sure if you put a human there at the base it'd make it a lot bigger in perception
[07:31:28] <tlvb> yeah, the scale is kind of lost
[07:32:37] <inflex> interesting that they have grass around the launch site... you'd think it'd get fried
[07:33:36] <RikusW> http://ca.isohunt.com/download/152232381/expert+c+programming.torrent
[07:33:47] <RikusW> seems a bit dead at the moment
[07:34:07] <ziph> inflex: Did you ever see the mercury launch videos? They charred everything. With new launches they seem to have good control of what gets hot though. :)
[07:34:12] <tlvb> the grass is part of another sience experiment investigating how to develop draught hardier crops
[07:34:41] <RikusW> ooh :-P
[07:35:13] <Valen> i wonder if the mars defence screen will be active
[07:36:16] <karlp> what's their track record again at shooting down incomings?
[07:36:22] <karlp> 3/8 or something?
[07:48:57] <inflex> well, they had their long-range going well the other week, knocking off the Russian grunt probe
[07:49:59] <RikusW> they ? the aliens ? :-P
[07:50:37] <Steffanx> The Americans are weird guys RikusW
[07:51:15] <crazy_pete> sorry i wasn't paying attention inflex
[07:51:52] <crazy_pete> i don't "want" you to use C++, i just think the fact that you don't use it indicates you don't really know what is going on, (no offence but that is what i think)
[07:52:17] <crazy_pete> what doesn't "work out right" on C++ anyway ?
[07:52:24] <RikusW> C++ can be irritating at times, particularly the constructors...
[07:52:34] <inflex> crazy_pete: well, by your own example... ISRs in C++.
[07:52:41] <RikusW> when combined with virtual functions
[07:52:54] <crazy_pete> lol @ virtual functions in an ISR
[07:53:03] <crazy_pete> i was always taught that you get in and out of an ISR ASAP
[07:53:12] <crazy_pete> which is why i like to write them in asm
[07:53:58] <inflex> crazy_pete: you're welcome to your views - no point saying "no offence", because that only ever makes things harder to get your point across. No, C++ doesn't offer any compelling advantage in terms of development that wouldn't be eroded away by having to retool/relearn everything again. Why change tools when the existing one is working appropriately?
[07:54:26] <crazy_pete> lol that is complete nonsense inflex
[07:54:35] <crazy_pete> 'scuse me
[07:54:41] <RikusW> crazy_pete: I did and stk500 and jtag mki and bootloader too in about 8kb in asm
[07:54:59] <inflex> crazy_pete: *shrug*
[07:55:02] <crazy_pete> i guess it depends on your "relearn" everything curve so that statement is just too subjective for rebuttal
[07:55:07] <crazy_pete> kewl RikusW
[07:55:47] <crazy_pete> and to say that "c++ offers no compelling advantage over c" is religious nonsense
[07:55:49] <inflex> crazy_pete: best go tell all those folks who still use FORTRAN, COBOL and other such languages that they should shift to something newer/better.
[07:55:59] <crazy_pete> er why would i do that ?
[07:56:15] <inflex> crazy_pete: given that I've not given you my criterion you are actually unable to make an evaluation of the validity of my stance.
[07:56:26] <inflex> crazy_pete: exactly, why would I do that.
[07:56:31] <crazy_pete> which is exactly what i said about 4 sentences ago inflex
[07:56:45] <crazy_pete> that your position is vague and subjective and not rebuttable
[07:56:49] <inflex> crazy_pete: but you're happy to still make responses on your own personal conjecture.
[07:57:00] <crazy_pete> i am not making any conjecture inflex
[07:57:16] <inflex> you must be making some to have induced your delightful responses.
[07:57:22] <Steffanx> => #crazypeteandinflexsdiscussion <=
[07:57:26] <crazy_pete> i stated facts you stated vague subjective opinions
[07:58:06] <crazy_pete> which is usually what happens when a scientist engages a religious fanatic. I state facts and you hide from facts behind meaningless subjective judgements
[07:58:09] <crazy_pete> to each their own :)
[07:58:22] <inflex> crazy_pete: some people you just don't bother going down that path with.
[07:58:29] <Tom_itx> and neither will convince the other
[07:58:34] <Tom_itx> end of discussion
[07:58:39] <crazy_pete> and did i ever try to tell you that you "MUST" use C++? I think you are putting words in my mouth
[07:58:40] <inflex> well said.
[07:58:47] <Steffanx> Ban Ban Ban Tom_itx :P
[07:58:54] <crazy_pete> i don't think i was ever trying to convince you
[07:59:08] <Steffanx> lol Tom_itx :P
[07:59:24] <crazy_pete> to Tom_itx if you want me to leave i will it is not a big deal
[07:59:30] <Steffanx> How rude to stop a discussion by abusing your powers
[07:59:43] <crazy_pete> i didn't even know it was a discussion
[07:59:49] <Steffanx> Not?
[08:00:02] <Steffanx> It was just a "throw in your arguments"-talk?
[08:00:48] <crazy_pete> the way i saw it, i was making smalltalk and i made a statement of fact and someone went off on a religious defensive rant
[08:00:53] <inflex> crazy_pete: really, there is no scientific discussion here - never was.
[08:01:16] <crazy_pete> if that is socially unacceptable in this room i apologise and will humbly leave
[08:01:48] * Tom_itx reads log, figures it can carry on a while
[08:01:53] <Steffanx> Don't leave
[08:02:03] * Tom_itx at least until one eye is open
[08:02:16] <crazy_pete> i never said there was any scientific discussion, inflex nor did i ever (AFAIK) try to compel you to use C++, i just stated that as a matter of fact it has a lot of advantages
[08:03:06] <inflex> crazy_pete: At the same time alas though you moved to deride people who chose not to migrate or accept.
[08:03:55] <crazy_pete> well since you were putting words in my mouth i don't believe i uttered i am not sure i was deriding you either inflex
[08:04:18] <crazy_pete> but if you want to discuss it personally we probably SHOULD take that to personal message?
[08:04:21] <inflex> """<crazy_pete> i don't "want" you to use C++, i just think the fact that you don't use it indicates you don't really know what is going on, (no offence but that is what i think)"""
[08:04:21] <Tom_itx> nothing like waking up to an argument that will go nowhere
[08:04:35] <inflex> Tom_itx: it's like coffee for the sourl.
[08:04:36] <inflex> soul.
[08:04:55] <crazy_pete> what did Isaak Denison say about coffee and the soul?
[08:05:05] <inflex> probably that it's poison - devil's broth.
[08:05:18] <crazy_pete> naw she was really complimentary about it
[08:06:43] <RikusW> lets see what Expert C Programming Deep C Secrets.pdf is about :)
[08:08:11] <RikusW> thegeeks.bz
[08:09:12] <tlvb> Secrets of the deep C
[08:09:18] <inflex> C of Peril
[08:11:56] <crazy_pete> well i am off peace all
[08:12:36] <Tom_itx> back to devcon 1
[08:12:54] <inflex> Meh
[08:13:06] <inflex> I was wondering why the temperature in this room was up so high today
[08:13:56] <inflex> Sorry about that k-man
[08:14:05] <Kevin`> lower defcon numbers are more critical
[08:14:39] <inflex> You should use the Australian bush-fire system... we had to push it beyond "Extreme" last year to add a new level "Catastrophic"
[08:14:49] <inflex> Waiting for them to add "Biblical" next.
[08:15:17] <Tom_itx> funny you have those but you set your cane off on purpose
[08:15:43] <inflex> Tom_itx: ja, completely different areas of the country in this case.
[08:16:00] <inflex> but I see the humour there.
[08:16:40] <inflex> Don't get me wrong, the fires that tore up the country last year were horrific (260 people or so died I think?), but I don't see the point of adding another word on top of an already existing scale.
[08:16:56] <inflex> all that'll happen is people will get used to "catastrophic" and we'll need to notch it up again
[08:54:54] <amee2k> are there any downsides to using a motor starting cap as general purpose (medium voltage, high capacity, non-polarized) cap?
[08:55:13] <Kevin`> amee2k: they aren't high capacity
[08:55:54] <amee2k> compared to foil caps, they aren't too bad
[08:56:05] <amee2k> (emphasis on non-polarized)
[08:56:34] <amee2k> looking at a 2u / 450Vac starting cap right now
[09:09:25] <Casper> amee2k: possibly have ESR issue
[09:10:25] <amee2k> should be fine if its less than an ohm or so
[09:10:42] <amee2k> any higher and dissipation will probably pop it
[09:11:17] <tlvb> liftoff
[09:15:41] <amee2k> mmh note to self... i need to buy a big glass jar when i go shopping later
[10:33:55] <errordeveloper>
[10:45:29] <inflex> wel, that wasgood
[10:45:32] <inflex> now to wait 18 months
[10:45:34] <inflex> 8 even
[10:46:41] <amee2k> 8 months? o.O
[10:50:16] <Landon> amee2k: _everyone_ has to east ;)
[10:50:18] <Landon> eat*
[10:50:24] <soul-d> :)once again amazed by chineese schipping was a suprise even delivery on saturday
[10:50:36] <inflex> amee2k: yes, 8 months before it gets to Mars
[10:50:43] <soul-d> http://i.imgur.com/jEcAW.jpg stocking up for the winter
[10:51:05] <amee2k> soul-d: when i order stuff from china i'm usually surprised if it arrives at all
[10:51:16] <Landon> inflex: ah nooo, I forgot to watch that
[10:51:19] <soul-d> i was almost thinking you got some girl pregnant but that would take 9 :P
[10:53:02] <inflex> Landon: well, youtube is full of videos alrady
[10:53:15] <inflex> Landon: but at least you know it got away successfully
[10:53:15] <soul-d> amee2k, true and this is like 3rth shipment in row to
[10:53:23] <soul-d> all under 10 day's from tracking code
[10:53:43] <amee2k> nice
[10:54:20] <Landon> inflex: true, but spoilers ;)
[10:54:23] <soul-d> but still need to check list against recevied stuff but looks correct
[10:56:15] <amee2k> whats a cheap and easy to use off-line SMP controller ic? any recommendations?
[10:57:44] <jakllsch> SMP?
[10:57:57] <amee2k> switch mode power supply
[11:01:35] <ben1066> SRAM is only reset when power is removed right? if so, could you have an avr write a magic number then the firmware, then in the bootloader, check the sram, if magic number is found, flash the avr
[11:01:39] <ben1066> then the avr could update itself
[11:02:20] <ben1066> amee2k: there is a quite old one, but very cheap
[11:02:21] <HKCMD> where can i get atmega16-8pu bootloader for arduino ?
[11:02:24] <ben1066> Ill see if I can find it
[11:02:37] <ben1066> HKCMD: There isnt one...168, 8 and 328, not 16
[11:02:45] <ben1066> there may be an unofficial one...
[11:02:55] <HKCMD> ok
[11:03:41] <ben1066> MC34063 amee2k, its not great, but its very cheap
[11:03:54] <amee2k> mmh, that works for off-line supplies too?
[11:04:17] <ben1066> off-line?
[11:04:25] <amee2k> straight off mains
[11:04:39] <ben1066> Not sure what its rated for....
[11:04:53] <amee2k> i've allready got some of these somewhere here. i don't particularly like them for low voltage stuff due to the built in darlington
[11:05:11] <ben1066> Nope, seems 40v max
[11:05:11] <amee2k> makes adding an external power switch somewhat awkward
[11:05:55] <amee2k> well off-line ICs don't see the full line voltage either but they have a say to safely bleed some power off the transformer to supply it
[11:06:22] <ben1066> how much voltage out?
[11:06:35] <amee2k> plus low start current and BOD so they can reliably start off a high valued resistor off the mains rectifier
[11:07:11] <ben1066> Hm, your on your own then :P Im not good with very specific supplies, Ive only ever had to use them to step up :p
[11:07:17] <amee2k> some 10-20V, if thats important. 50 probably tops
[11:07:50] <amee2k> i'd wire the feedback path so it is current sensitive anyway
[11:16:13] <ben1066> Whats the pitch on SOJ?
[11:47:38] <rue_house> ben1066, pls wait
[12:22:24] <RikusW> hi vectory
[12:23:34] <ben1066> rue_house: eh?
[12:35:07] <RikusW> RavrProg + Dragon + jtag programs m324a :)
[12:35:39] <RikusW> Steffanx: you want the newer version ?
[12:35:46] <Tom_itx> that yours?
[12:35:54] <RikusW> yes
[12:35:57] <Tom_itx> nice
[12:36:01] <RikusW> Qt4 gui
[12:36:05] <Tom_itx> you've been working on it for some time
[12:36:13] <RikusW> on linux :)
[12:36:15] <RikusW> yes
[12:36:25] <RikusW> starting to show results now
[12:36:57] <RikusW> left the flash writing for last, thats usually the most annoying part to get working
[12:37:30] <RikusW> Tom_itx: interested in getting the source ?
[12:38:24] <Tom_itx> not at the moment. i probably wouldn't know what i was looking at anyway
[12:38:47] <RikusW> It almost looks like the AS4 programmer gui
[12:39:59] <RikusW> Tom_itx: works with AVRISP mkII too
[12:40:06] <Tom_itx> i recall that
[12:40:09] <RikusW> only ISP mode for now
[12:40:27] <RikusW> got some t10's will add TPI sometime
[12:40:30] <Tom_itx> i don't know that i have the tool chain to compile it
[12:41:08] <RikusW> gcc libusb-0.1.12 Qt4-dev
[12:41:13] <RikusW> and make
[12:41:30] <RikusW> on linux easily available
[12:50:01] <RikusW> One feature that works great is the AVR autodetect
[12:50:19] * RikusW don't like looking around a looong list in AS4
[12:52:26] <ben1066> whats this RikusW ?
[12:53:01] <RikusW> avrdude replacement
[12:53:12] <RikusW> with qt4 gui
[12:53:27] <ben1066> Sounds nice, what devices does it support, programmers?
[12:53:33] <RikusW> I started on a commandline interface, but didn't pay much attention to it
[12:53:40] <ben1066> Though, I use AS5 so I dont really need it :p
[12:53:41] <RikusW> all atmel ones
[12:53:50] <RikusW> except jtag3 and avrone
[12:54:09] <ben1066> why except those?
[12:54:19] <RikusW> interface is better than AS5 :-P
[12:54:28] <RikusW> don't have any docs on their protocols
[12:54:56] <RikusW> stk600 is untested, don't have one
[12:54:57] <RikusW> but it might work
[12:55:13] <RikusW> since its similar to AVRISP mkII
[12:55:25] <RikusW> and I have abcminiusers clone to test that
[12:55:45] <Tom_itx> you got one from him?
[12:56:00] <RikusW> loaded LUFA on one of my 32u2 boards
[12:56:10] <Tom_itx> that works
[12:56:15] <RikusW> 2 actually with different serial
[12:57:35] <RikusW> so I now have AVRISP mkII - JTAG ICE mkI clone on m16 - AVRDragon - and my own U2S around, thats a lot of programmers to have :-P
[12:58:14] <RikusW> Had a borrowed stk500 too
[12:58:45] <Tom_itx> i've got a few but most are the same
[12:59:34] <RikusW> what do you have apart from the ISP mkII clone ?
[13:00:01] <Tom_itx> jtagice, dragon, parport
[13:00:21] <RikusW> jtag mkii ?
[13:00:54] <Tom_itx> i don't remember, it's a clone
[13:01:04] <RikusW> then its mki
[13:01:16] <Tom_itx> supposed to be mkii
[13:01:31] <RikusW> ah there is mkii clones too
[13:01:40] <RikusW> wonder how that was done
[13:02:11] <RikusW> since stk600 and jtag mkii fw is encrypted
[13:02:16] <Tom_itx> it doesn't do pdi which is why i got it originally
[13:02:23] <Tom_itx> so it sits
[13:02:52] <RikusW> dragon doesn't like pdi much either
[13:03:06] <RikusW> its supposed to work on xmA4
[13:03:17] <RikusW> but not at all on xm128A1
[13:03:20] <RikusW> only jtag
[13:07:14] <ben1066> There arent any jtagice mk2 clones are there?
[13:07:18] <ben1066> If not why not :p
[13:07:25] <Steffanx> There are
[13:07:32] <ben1066> There are?
[13:07:33] <Steffanx> I have a jtagice mkii clone
[13:07:34] <ben1066> Link?
[13:07:42] <Steffanx> avrvi has them
[13:08:09] <RikusW> about halfprice iirc ?
[13:08:15] <ben1066> No open source ones though?
[13:08:17] <Steffanx> Something like that
[13:08:24] <Steffanx> Some.. buggy onces
[13:08:37] <RikusW> ben1066: you're curios about the internals ?
[13:09:08] <RikusW> ben1066:
http://sites.google.com/site/megau2s/home/jtag
[13:09:33] <RikusW> there is the jtag mki disassembly and my rewritten one too
[13:09:40] <ben1066> ooh nice
[13:09:47] <RikusW> AS5 doesn't support mki tough
[13:09:50] <RikusW> though
[13:09:52] <ben1066> Yea, I know :S
[13:10:06] <ben1066> And the mk2 clones arent worthwhile at only 50%, 50$ more you get official mk3....so
[13:10:08] <RikusW> took me like 6 months to make sense of that fw
[13:11:07] <RikusW> The debugging scanchains is almost simpler than the programming ones ;)
[13:11:18] <ben1066> Lol
[13:11:43] <RikusW> just enter IR=8 to break
[13:11:47] <RikusW> and IR=9 to GO
[13:11:58] <RikusW> 0xA is to execute instructions on the avr
[13:12:09] <RikusW> and 0xB is the OCD registers
[13:13:08] <RikusW> ben1066: you know anything about jtag internals ? like the TAP ?
[13:24:18] <ben1066> RikusW: No :(
[13:24:36] <ben1066> How AVRs work interests me but I dont really know much :P
[13:24:52] <RikusW> There is some info on jtag in the AVR datasheets
[13:25:14] <RikusW> thats where I saw the TAP the first time
[13:25:50] <ben1066> I might get a JTAGICE mk3 some time
[13:25:56] <ben1066> AKA, When i can buy one in the UK :p
[13:27:50] <ben1066> And when I need one, atm £150 for a debugger is a bit much for me
[13:28:32] <RikusW> dragon is $50
[13:28:52] <RikusW> or load my fw onto a m328
[13:28:55] <RikusW> or m168
[13:29:14] <RikusW> but you'll only be able to debug the older avr's with AS4
[13:29:41] <RikusW> like m16 / m32
[13:29:48] <ben1066> I can do yours easily, but as you said only older avrs which I dont use mostly, most of my avrs are 168s and 8s
[13:29:48] <RikusW> m64 m128
[13:30:17] <ben1066> and the dragon, albeit it works, ive heard toomany breaking :P
[13:30:31] <RikusW> just be carefull with it
[13:31:04] <ben1066> Then again, it be solely for JTAG, maybe HVSP/HVPP since I have my avrisp mk2
[13:31:07] <RikusW> once I have dW working on my board I'll only use my dragon for AVR32 or xmega
[13:31:09] <ben1066> <3
[13:31:19] <ben1066> dW? debug wire?
[13:31:27] <RikusW> yes
[13:31:44] <RikusW> already know the commands
[13:32:13] <RikusW> http://www.ruemohr.org/docs/debugwire.html
[13:33:22] <RikusW> Will probably add a Debug_Interface class to my SW as well, later
[13:33:50] <RikusW> first want to finish this and clean it up a bit
[13:39:01] <ben1066> I prefer using official programmers and such
[13:39:06] <ben1066> they just work better :p
[13:40:09] <RikusW> my board is an "official" stk500 ;)
[13:40:13] <RikusW> nearly
[13:40:20] <RikusW> I got hold of the stk500 fw too
[13:43:07] <vectory> hi, i just heard usb2rs232 converters were unreliable, can you confirm this?
[13:45:11] <Steffanx> unreliable as in ...
[13:45:43] <RikusW> the flow control could mess up...
[13:46:00] <RikusW> or cause syncing problems
[13:46:24] <RikusW> vectory: so you tried that x64 inf ?
[13:47:10] <vectory> nope, cant get windows to work realy
[13:47:27] <RikusW> Steffanx: AS4 somehow "supports" programming all avrs with jtag mki, its not greyed out, in the debug menu it is... weird
[13:47:28] <vectory> msdnaa keeps giving me crc errors after 3 gb downloads of the image
[13:47:34] <vectory> had to reinstall, ye know
[13:47:55] <RikusW> Steffanx: I actually programmed m324a with my jtag clone in AS4...
[13:48:06] <Steffanx> ok :)
[13:48:07] <ben1066> RikusW: Tut Tut :P Rikus, cant you just pull the FW from the AVR, or is it protected?
[13:48:20] <ben1066> If so, get it when your PC updates?
[13:48:39] <Steffanx> the fw is encrypted ben1066
[13:48:52] <Steffanx> It's decrypts it inside the programmer/avr
[13:48:55] <Steffanx> -'s
[13:49:06] <Steffanx> And the AVR's are protected :)
[13:49:06] <RikusW> ben1066: and Steffanx found out that the dragons lockbits is set too :(
[13:49:15] <Steffanx> Ofcourse
[13:49:24] <RikusW> the dragon actually have jtag ports on the underside...
[13:50:05] <RikusW> Steffanx: how about cryptanalysis ?
[13:50:11] <Steffanx> Good luck
[13:50:40] <RikusW> AES is supposedly impossible to crack
[13:51:05] <RikusW> was it not for mixcolumns i would have tried
[13:51:19] <RikusW> that mess up everything
[13:51:59] <RikusW> could have significantly simplified AES and collapsed all rounds into 1 were it not for mixcolumns
[13:52:31] <RikusW> but there always are side channel attacks ;)
[13:52:48] <RikusW> measure the current the dragon draws while updating...
[13:53:58] * RikusW wonders if the slave fw is decrypted on the master or slave ?
[13:54:46] <vectory> if it was encrypted on the master you could check signals of communication with your osci
[13:54:47] <Steffanx> I think it's decrypted on the slave
[13:54:52] <vectory> decrypted*
[13:55:03] <Steffanx> Or they are REALLY stupid
[13:56:32] <RikusW> or lazy
[13:57:38] <RikusW> vectory: seems I have flash writing working in my gui
[13:57:45] <RikusW> for ISP and JTAG
[13:57:50] <vectory> great
[13:57:58] <vectory> _seems_
[13:58:05] <vectory> ;D
[13:58:08] <Steffanx> The JTAGICE3 only has one uc :)
[13:58:22] <RikusW> and I have modified my jtag mki clone to support newer avr's too
[13:58:32] <Steffanx> The hw of the JTAGICE3 doesn't seem to be that advanced
[13:58:32] <RikusW> Steffanx: yeah the UC3A3 one
[13:58:45] <RikusW> hi speed usb 2
[13:59:14] <RikusW> vectory: so once its stable and if you need it I'll send you an update
[14:00:10] <RikusW> vectory: there is one thing with that jtag mki you have, don't use automatic chip erase....
[14:00:12] <Steffanx> http://wallyoz.smugmug.com/photos/1232039987_JmRBx-X2.jpg the inside of a jtagice3
[14:00:24] <RikusW> vectory: erase first then program flash
[14:00:42] <Steffanx> 490kb fyi RikusW :P
[14:00:53] <Steffanx> *KB
[14:01:24] <RikusW> I have about 35MB till end of the month so don't really care now :-P
[14:03:37] <RikusW> vectory: the erase part didn't wait for the avr to finish on your version....
[14:04:39] * RikusW wonders if that BLDR pads are to prevent bricking the jtag3 ?
[14:06:41] <vectory> oooh, you talk about the fw
[14:06:42] <Steffanx> :)
[14:07:00] <vectory> i thought the qt-app
[14:07:15] <Steffanx> Ha
[14:07:17] <Steffanx> :P
[14:07:34] <RikusW> vectory: yes the jtag mki fw on your U2S
[14:08:23] <ben1066> I want a christmassy project
[14:08:32] <Steffanx> me too
[14:08:57] <RikusW> lots of blinky leds ? ;)
[14:09:12] <Steffanx> I still have to continue my word clock
[14:09:28] <Steffanx> And toaster/reflow oven
[14:10:03] <ben1066> blinky leds to do what?
[14:10:14] <Steffanx> For your christmas tree
[14:10:27] <RikusW> seems there is even a jtag connector on the mk3.,
[14:10:28] <ben1066> Pah :p
[14:10:38] <ben1066> RikusW: What else would there be/
[14:10:58] <RikusW> ben1066: another one for debugging the mcu on the pcb
[14:11:06] <ben1066> Oh...
[14:11:13] <RikusW> look on the upper left
[14:12:15] <ben1066> the avrispmk 2 has one too
[14:13:04] <RikusW> lockbits are probably set like on the dragon
[14:13:51] <ben1066> its jtag or 10 pin ISP
[14:15:03] <ben1066> and it has a mega 128a
[14:15:14] <RikusW> check pins 2 and 10 for gnd then its probably jtag
[14:17:19] <ben1066> pin2 and 10
[14:17:21] <ben1066> are ground
[14:17:23] <ben1066> seems to be jtag
[14:18:04] <ben1066> it also has 4 test points
[14:18:06] <RikusW> probably
[14:18:17] <RikusW> atmel would have wanted to debug it...
[14:30:02] <RikusW> wc output of my SW
http://codepad.org/pNuS7x9j
[14:30:14] <RikusW> 15000 lines... oooh
[14:30:23] <RikusW> didn't realize it was that much
[14:35:37] <ben1066> The test points are odd
[14:35:40] <ben1066> 1 is GND
[14:35:46] <ben1066> two are near 5v
[14:35:50] <ben1066> one is nearer 2v
[14:36:32] <ben1066> Im guessing 1 is USB Voltage, 1 is regulated 5v, the other, no idea
[14:42:27] <Steffanx> That's incl. qextserial stuff RikusW ..
[14:43:01] <Steffanx> And rtk seems to be quite large too
[14:47:22] <ben1066> Did anyone see that the avr butterfly schematic got released?
[14:48:12] <RikusW> I have them
[14:48:40] <RikusW> on the AVR2005 library cd
[14:50:06] <RikusW> want them ?
[14:52:48] <ben1066> the what?
[14:53:22] <ben1066> I have the april 2011 one
[15:01:46] <RikusW> AVR2005/tools/developer/butterfly
[15:01:50] <ben1066> ah
[15:01:55] <ben1066> They are in the datasheet
[15:02:00] <RikusW> in docs
[15:02:14] <RikusW> is it actually the same on the 2011 one ?
[15:02:35] <ben1066> I dont know...
[15:03:32] <RikusW> I mean is that path on the 2011 dvd ?
[15:05:11] <ben1066> atmel avr technical library yes?
[15:06:12] <ben1066> No its
[15:06:13] <ben1066> not
[15:06:16] <ben1066> Its at F:\techlib\techlib8\tools\evk\butterflyF:\techlib\techlib8\tools\evk\butterfly
[15:07:11] <RikusW> and the schematic is there ?
[15:07:40] <ben1066> Under doc yes
[15:07:57] <ben1066> These havent been edited since 03
[15:08:00] <ben1066> So I think its the same :p
[15:10:44] * RikusW wonders if anyone here ever used a real AVR ICE ?
[15:12:33] <ben1066> i want one, does that count :p
[15:13:57] <RikusW> those are probably really expensive
[15:14:10] <RikusW> will make the AVRONE look cheap
[15:14:31] <RikusW> AVRONE == $600
[15:17:14] <ben1066> Oh not JTAGICE, just ICE
[15:18:16] <RikusW> yes
[15:18:25] <RikusW> emulating the IC itself
[15:18:42] <RikusW> like the AVR ICE 10
[15:18:50] <ben1066> surely you could do that yoruself with an FPGA
[15:20:34] <RikusW> with a lot of free time yes
[15:21:15] <ben1066> There are already AVR cores avaliable
[15:23:07] <RikusW> ICE40 List Price $995. > Same user interface for all AVR development tools. >
[15:23:08] <RikusW> Accurate emulation of tiny26 and mega8 over entire voltage and speed range ...
[15:23:30] <RikusW> I know about those
[15:23:50] <RikusW> but don't have any fpga's around
[15:23:52] <ben1066> just tiny26 and mega8?
[15:25:18] <RikusW> and t13 too
[15:26:03] <RikusW> I'd rather use the dragon ;)
[15:28:17] <ben1066> Lol
[15:28:23] <ben1066> I might get a dragon, IDK
[15:28:28] <ben1066> im scared ill break it :p
[15:30:17] <RikusW> just don't feed it 15V :-P
[15:31:50] <RikusW> and don't touch the regulator while its on... (near the ferrite coils)
[15:32:25] <RikusW> I connect mine to a USB port directly on the PC MB
[15:33:28] <RikusW> and use 4 feet usb cables
[15:33:58] <ben1066> See this is why I like the jtagice, it has a case :p
[15:34:09] <ben1066> and its faster but meh
[15:34:18] <ben1066> but the dragon is 1/4 the price
[15:34:31] <RikusW> my board use a mega32u2 and is repairable...
[15:34:56] <ben1066> The only reason I want a dragon is for the JTAG
[15:34:56] <ben1066> :P
[15:35:47] <RikusW> for mega xmega or AVR32 ?
[15:36:07] <ben1066> Debugging and AVR32 sometime
[15:36:10] <ben1066> Id like to try them
[15:36:21] <ben1066> Mega, the new megas,
[15:36:31] <ben1066> 88/168/328
[15:36:48] <RikusW> why not try mega324a ?
[15:36:58] <RikusW> was cheaper here than m328
[15:36:59] <ben1066> I dont have any, I have the others, and I like them
[15:37:03] <RikusW> and have 40 pins
[15:37:06] <ben1066> They are really cheap for me
[15:37:17] <RikusW> 324 was cheaper for me
[15:37:25] <RikusW> in 40 pin dip
[15:37:30] <ben1066> i cant get 324s from farnell, I order most stuff there
[15:37:35] <RikusW> have a look at RS
[15:37:46] <ben1066> Oh nope, i lied
[15:38:00] <RikusW> http://rs-online.com/
[15:38:05] <ben1066> £3.67
[15:38:11] <RikusW> http://uk.rs-online.com/ maybe ?
[15:38:33] <ben1066> vs £4.98 well shush :p
[15:38:36] <ben1066> I use 168s :p
[15:38:43] <ben1066> thats all I have, and some atmega8s
[15:38:47] <ben1066> and an attiny45
[15:38:55] <ben1066> and a 164
[15:38:58] <RikusW> I got m328p for R43 and m324a was R31
[15:39:15] <RikusW> R8.44 / USD
[15:39:17] <ben1066> I like the new avrs, they run faster with lower power too
[15:39:36] <RikusW> mxx4 and mxx8 go to 20MHz
[15:39:36] <ben1066> The only other type of avr I want is one with XMEM so 8515 or 162
[15:39:53] <ben1066> Oh whats the difference between mxx4 and mxx8 then?
[15:39:58] * RikusW have 1 m162-au
[15:40:11] <RikusW> mxx4 is 40 pin
[15:40:21] <RikusW> mxx8 28 pin like m8
[15:41:03] <RikusW> m64 and m128 also have XMEM
[15:41:16] <ben1066> They are smd and harad to protoype :p
[15:41:37] <RikusW> m162 DIP is EXPENSIVE
[15:41:40] <ben1066> ja
[15:41:44] <ben1066> thats why 8515 :p
[15:42:14] <ben1066> wouldnt call the 128 cheap
[15:42:37] <RikusW> you'll need a 8 bit latch to use XMEM too
[15:42:39] <ben1066> Nor the 64
[15:42:41] <ben1066> Yea I know
[15:42:44] <ben1066> But thats what, 70p
[15:42:45] <RikusW> like 573 or 373...
[15:43:13] <RikusW> you might get ram from old 386 motherboards ;)
[15:43:29] <RikusW> I have quite a few 32kb sram chips here
[15:43:33] <RikusW> some 64k too
[15:43:34] <ben1066> sram is cheap
[15:43:35] <ben1066> so yea
[15:45:13] <RikusW> the UC3-A3 xplain board is like $29
[15:45:22] <RikusW> and have 8MB dram on it
[15:45:26] <ben1066> Yea :
[15:45:27] <ben1066> :p
[15:45:32] <ben1066> And, thats not an atmega D:
[15:46:36] <RikusW> what do you want XMEM for ?
[15:46:43] <ben1066> More ram :p
[15:48:21] <RikusW> UC3A3 got 128kb ram
[15:48:35] <ben1066> Your trying to convince me to switch to AVR32 D:
[15:48:44] <ben1066> albeit its probly what ill move on to
[15:48:46] <RikusW> nah
[15:48:59] <ben1066> I prefer micros from one vendor
[15:49:01] <RikusW> but they have nice specs
[15:49:21] <RikusW> but are very different from AVR8
[15:49:33] <ben1066> How so
[15:49:36] <RikusW> the block diagram looks more like an arm
[15:49:52] <ben1066> But the toolchain is similar and the way the programmer uses them is similar yes?
[15:50:07] <RikusW> http://sites.google.com/site/megau2s/home/datasheets
[15:50:14] <RikusW> have a look at the datasheets
[15:50:26] <RikusW> oh, but you have it on that dvd too...
[15:50:51] <RikusW> the AVR32 jtag interface is TOTALLY different
[15:51:29] <RikusW> but its actually open, debugging as well, nexus spec
[15:51:32] <powderhound> Does AVR Studio 5 have a define for CPP on the Xmegas? I can't seem to find it anywhere
[15:52:47] <RikusW> hey my UC3-A3 got AES onboard :-P no import licence crap on that one :)
[15:53:15] <RikusW> powderhound: avr-gcc docs ?
[15:53:22] <RikusW> or avr-libc maybe
[15:54:26] <powderhound> RikusW: Ahh, good call
[16:03:42] <Steffanx> Huh why not license crap for that one RikusW ?
[16:04:24] <RikusW> Different company imported it
[16:04:39] <RikusW> ArrowAltech, official atmel rep over here
[16:05:16] <RikusW> didn't even know there was AES onboard
[16:05:28] <RikusW> saw it now...
[16:07:49] <RikusW> http://opend.co.za/hardware/avride/avride.htm
[16:07:54] <RikusW> m32 -> hdd
[16:13:09] <ben1066> :O
[16:13:18] <ben1066> HDD or CF
[16:14:06] <RikusW> sd card seems like a better idea ...
[16:14:10] <ben1066> ja
[16:15:07] <RikusW> 00:13
[16:15:16] <ben1066> Eh?
[16:15:33] <RikusW> its getting early here
[16:15:49] <Steffanx> It's only 23:07 here :)
[16:15:52] <RikusW> better get some sleep
[16:16:09] <RikusW> Steffanx: hows that possible ?!
[16:16:09] <Steffanx> It's sunday..
[16:16:31] <Steffanx> summer/winter time RikusW
[16:16:35] <ben1066> 22:08 here
[16:16:45] <ben1066> RikusW: your clock is wrong
[16:17:07] <mrfrenzy> lör nov 26 23:08:40 CET 2011
[16:18:13] * RikusW is gmt+2
[16:18:52] <mrfrenzy> gmt+2+7 minutes
[16:35:29] <ben1066> someone should make an AVR Dragon case
[17:05:18] <ben1066> why is studio 5 hated so much?
[17:14:59] <Tom_itx> it's buggy and very bloated
[17:17:39] <Steffanx> Tom_itx thinks it's bloated ben1066
[17:19:40] <soul-d> http://imgur.com/a/82QeU almost looks like it works :P
[17:20:39] <soul-d> logic analyzer thingy now wait on 8 more cables to arrive in mail :P
[17:25:14] <grummund> i see lots of wires
[17:26:48] <ben1066> heh, im just going to buy a saleae logic, also, nxp are releasing dil arms
[17:27:50] <soul-d> thre 2 pics though also a capture of the counter on the leds 0 to 10 and back down :P
[17:29:03] <soul-d> but now the fun part are the errors inthe logic analyzer or in the breadboard setup or in code :P
[17:32:01] <Steffanx> Stop with your dil orientated mind ben1066 :P
[17:34:30] <ben1066> steffanx, its so much easier
[17:34:56] <Steffanx> It's easier, not so much easier :P
[17:35:19] <ben1066> i dont have to order pcbs
[17:35:34] <ben1066> or buy a good solering iron ;p
[17:36:36] <ben1066> i have some unbranded iron, totally fine for pth, dont think smd would work
[17:37:22] <ben1066> stencils are a repeating cost
[17:38:24] <ben1066> and to find a decent iron itd cost £100+ for something i dont use often
[17:39:22] <ben1066> so yea
[17:40:16] <soul-d> you tend to use stuff that works well more often versus stuff you know does'n't work on anything and you'r happy if you could solder dip with it
[17:41:15] <ben1066> the tip would be too fat for smd..probly bad temp too
[17:42:40] <ben1066> if anyone knows of a decent iron in the uk thatll leave me some money i might consider it
[17:42:57] <soul-d> basic ayoue station and tip ?
[17:43:44] <ben1066> ive never found one at a trustworthy place, uk compatible
[17:44:02] <soul-d> http://www.pcb-soldering.co.uk/aoyue-968-pcb-work-station.html
[17:44:26] <soul-d> if thats to expensive
[17:44:33] <soul-d> pick only iron and tip station
[17:45:25] <soul-d> http://www.pcb-soldering.co.uk/aoyue-937-soldering-station.html
[17:46:05] <soul-d> get few tips with it and your set :P plenty of options if you want
[17:46:08] <ben1066> are they any good?
[17:46:24] <soul-d> if as you said don't use it often it probably does good
[17:46:41] <soul-d> i have abn 968 station since few weeks i like it sofar
[17:47:05] <soul-d> also has vacume pum to suck up air near tip
[17:47:17] <keenerd> My GA144 arrived. Now I just have to make a programmer for it and re-learn Forth...
[17:48:20] <rue_shop2> anyone have a flashing led hex for a 2313
[17:48:31] <rue_shop2> ?
[17:48:56] <soul-d> not quite sure what a led hex is
[17:49:34] <ben1066> hex file to flash a led
[17:49:53] <ben1066> also 968..bit pricy
[17:50:38] <soul-d> yeah thats if you don't need hoty air and don't mind nasty fumes
[17:50:44] <soul-d> get just a digital station
[17:51:17] <ben1066> hmm..atten or aoyue
[17:52:40] <rue_shop2> so?
[17:52:50] <soul-d> depends on what port ?
[17:52:58] <rue_shop2> does anyone have a known good .hex for a led flasher for a 2313?
[17:53:01] <rue_shop2> dosnt matter
[17:53:11] <rue_shop2> someone must have made a flashing led for a 2313
[17:54:02] <ben1066> soul-d the atten 938 or aoyue 937?
[17:54:31] <rue_shop2> nobody!?
[17:54:43] <rue_shop2> nobody has a hex for a led flasher on a 2313?
[17:54:44] <soul-d> donno only have had ayoue
[17:55:24] <ben1066> the atten one seems to be better equipped
[17:55:47] <ben1066> esd safe, shows cur temp set temp and iron power
[17:56:17] <soul-d> rue_shop2, moment
[17:58:01] <ben1066> and what type of tip
[17:58:45] <rue_shop2> this isn't something I was trying to start a long conversation about, I found one via google
[17:59:15] <rue_shop2> do you know what a flashing led program is!? do you know what a 2313 is? do you know what a hex file is? gizzard
[17:59:34] <rue_shop2> anyho, I was just wondering if someone had one already
[18:00:18] <soul-d> who you asking ?
[18:05:23] <soul-d> you got something already rue_shop2 ?
[18:08:00] <soul-d> http://www.webfilehost.com/?mode=viewupload&id=6820877 you could try that hex otherwise should flash on port B0 for about asecond
[18:08:06] <soul-d> if it runs on 8mhz
[18:17:24] <rue_shop> hehe
[18:17:31] <rue_shop> two kinds of 2313 :)
[18:17:38] <rue_shop> whats with that?
[18:27:52] <soul-d> you have A or somthing ?
[18:40:05] <Tom_itx> rue_shop
[18:40:09] <Tom_itx> i should have one
[18:40:49] <Tom_itx> why? is there something odd about it over the other tutorial ones?
[20:26:47] <SianaGearz> helloes.
[20:28:22] <SianaGearz> i have been wondering - as far as i see in-system-programming interface is SPI... how do those Arduino boards program themselves through a serial port - do they use use some extra status pins in serial port for bit-banging the SPI, or do they have something like their own bootloader and need to be bootstrapped?
[20:35:33] <SianaGearz> also i can see a serial cable schematic, documentation of which says that CTS, DTR, RTS need to be supported, and i'm wondering whether Prolific USB->Serial chips do that, but cannot find...
[20:37:26] <Tom_itx> maybe #arduino can help you better
[20:38:50] <SianaGearz> ok, pop-quiz, what kind of programmer hardware or cable do people around here use to program their AVRs?
[20:39:04] <Tom_itx> i use my own
[20:39:31] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/boards/USBTiny_Mkii/USBTiny_Mkii_index.php
[20:39:41] <SianaGearz> thankies.
[20:39:45] <Tom_itx> did i pass/
[20:39:46] <Tom_itx> ?
[20:41:06] <SianaGearz> oh excellent, thank you. but it was a survey, not a competition.
[20:41:23] <SianaGearz> quite an adorable device really.
[20:41:26] <Tom_itx> it was a quiz
[20:41:56] <Tom_itx> you can make one your very own
[20:41:59] <SianaGearz> well i suppose you won, pow pow pow pow!
[20:42:43] <Tom_itx> i have a jtagicemkii clone, dragon, parport programmer and quite a few of my own
[20:42:57] <Tom_itx> oh and some bitbanged usb ones as well
[20:45:26] <SianaGearz> mhm, it appears it has a common problem - it needs a programmer to make a programmer.
[20:45:58] <SianaGearz> though it sure is nice enough to be worth making a parallel port one and hunting down a pc with a parallel port for a few minutes.
[20:56:42] <SianaGearz> mhm i'm totally new to all this... i'm wondering, can i realistically have a timer interrupt fire at 1MHz with an AVR running at say 20 MHz or so? how far does safe overclocking usually stretch - can i do 25 or 30 MHz too? how many useful cycles would i get in the interrupt routine invoked that often - i.e. what is the overhead?
[21:06:00] <SianaGearz> ah i found the answer to my overclock question, and it's satisfactory and very voltage dependent...
[21:22:33] <Casper> SianaGearz: why 1MHz?
[21:23:07] <Casper> iirc, the interrupt is 4 cycle, the return is 4 cycles, so you have 8 cycles left...
[21:33:17] <SianaGearz> Casper: a crazy idea of emulating 6502 on AVR cycle precisely, while doing some slow, low priority work in the background. though considering 6502 has an internal frequency doubler and two distinct states per MHz, i might need a 2MHz interrupt... which ok, would be ouch.
[21:34:18] <SianaGearz> though it looks like an atmega 168 can be overclocked to 40MHz easily @ 5v
[22:23:50] <Tom_itx> if you're totally new to this why are you taking off wanting to overclock a chip?
[23:08:52] <rue_house> hmm,
[23:10:06] <rue_house> so jester ended up finding led flasher code for his 2313 his 90s2313, he had a time trying to figure out why the chip wasn't working, we worked out he had to give it a clock signal :)
[23:10:21] <rue_house> but he got his m32 and 2313 going with his programmer
[23:11:21] <Casper> how good would be an avr be at smps?
[23:12:14] <eadthem> the delay loop will be long
[23:12:19] <eadthem> err
[23:12:25] <eadthem> the propgation delay will be long
[23:12:30] <Casper> I was thinking of differential adc with max gain
[23:13:01] <Casper> 7 bits precision
[23:13:06] <eadthem> pepole have done it
[23:13:08] <Casper> and basically do timer-adc
[23:13:28] <rue_house> dont use an avr for an smps
[23:13:32] <Casper> sample at the start of the cycle, set the timer for the pulse at the end of the cycle...
[23:13:46] <rue_house> electrical stuff changes too fast
[23:13:55] <eadthem> what if the load drops at the beginning of the cycle
[23:14:00] <eadthem> then you will overshoot
[23:14:04] <rue_house> use a TL494 and control the setpoint with an avr
[23:14:12] <eadthem> thats why i say resonant mode analog servo loop is the best
[23:14:31] <eadthem> adjust your settings and controls with the avr
[23:14:55] <Casper> the smps chip I have here seems to be a total failure, or I fried it yet it still work
[23:15:13] <eadthem> ie current min current max and voltage
[23:15:28] <eadthem> have you tryed makeing just the basic current loop in spice
[23:15:32] <Casper> do you have a schematics for your resonant one?
[23:15:56] <Casper> I don't have anything but pspice... and pspice ain't that good at all
[23:16:04] <eadthem> let me post the most recent one but
http://final.servegame.com/swcad/buck/ is current
[23:16:10] <eadthem> you must get ltspice
[23:16:31] <eadthem> http://final.servegame.com/swcad/lib/ and mitaswell add my lib and cmp extentions on to it as well
[23:16:45] <eadthem> it will add several parts
[23:17:37] <eadthem> the biggest problem pepole have with AVR today is they want to use them for everything
[23:17:57] <Casper> it would have helped me understand stuff atleast
[23:18:10] <eadthem> there not ment to be your data path there micro CONTROLLERS not micro dooers they are the sit back and supervise manager position
[23:18:38] <Casper> yeah
[23:18:46] <eadthem> that realy is the best annalogy for a atmel the manager not the worker
[23:19:08] <Casper> however I know how they behave. so easier to understand when stuff ain't working as expected
[23:19:14] <eadthem> yep
[23:19:19] <Casper> like it is the case for the sg3524
[23:19:23] <eadthem> you can make a smps with them
[23:19:42] <eadthem> but it will never be more than just functional
[23:20:03] <eadthem> not unless you move the servo mech to analog and make the uC the manager
[23:20:23] <Casper> I'm not intending to make the avr based one more than functional
[23:20:47] <eadthem> you want to do a buck or forward design?
[23:22:46] <Casper> not important for now, I'm going forward/flyback because I have nothing good in pnp or p-chan
[23:23:06] <Casper> the local store have nothing good too (they USED to have something... their part are going down)
[23:23:24] <Casper> and digikey... well, I'm considering an order soon
[23:23:27] <eadthem> well you pulled apart some DEER's right
[23:23:34] <Casper> ya
[23:23:41] <Casper> but don't know where I put all the Ic
[23:24:08] <Casper> I moved, I can't even find my 3 soldering rolls! I had to buy new one...
[23:26:17] <Casper> sadly, I'm kinda stuck to thru-hole for now, still hasn't figured out how to do good transfert of toner... barelly can do thru-hole somehow
[23:31:05] <eadthem> so do you ahve both ltspice and my lib and my cmp installed>
[23:31:06] <eadthem> ?
[23:32:35] <Casper> not yet
[23:33:02] <Casper> I'm half asleep, wondering if I stay asleep a bit more or go to bed
[23:38:52] <soul-d> casper go wake up or sleep with a smile :)
http://www.youtube.com/watch?v=u1EX--vdxh4&feature=player_embedded
[23:40:57] <Casper> soul-d: . . . ok then, check that!
http://www.youtube.com/watch?v=9kF2WoF17Z4
[23:44:48] <soul-d> not into pink unicorns though
[23:46:41] <Casper> I like the cover attempt :D
[23:47:27] <soul-d> don't know the origional though
[23:47:51] <eadthem> k casper i fixed up mptifc1.thy.asc
[23:48:01] <eadthem> it requires ltspice my cmp and my lib
[23:48:18] <eadthem> it shuld help ya out
[23:49:43] <eadthem> you will note its made for offline use
[23:50:27] <Casper> I'll check out tomorrow, but I think I'll go with coding an avr as a simple converter... hmmm I do see a problem already... bahh timing detail...
[23:50:53] <eadthem> this shows you the logic and opamps needed
[23:55:52] <inflex> flippen heck it's nasty weather here
[23:56:04] <Casper> inflex: snow?
[23:56:14] <inflex> Completely the opposite
[23:56:21] <inflex> blasting heat and humidity
[23:56:41] <Casper> oh yeah you'Re aussie
[23:57:55] <inflex> yes
[23:58:04] <Casper> ... and best buy just did a sale fraud
[23:58:34] <Casper> the tv that was last week at 329 regular price is now at 369 - 40$ rebate
