Back
[03:35:11] <Bushman> woohoo! i've done it!
[03:35:13] <Bushman> http://bushman.pl/misc/success.pdf
[03:35:34] <Bushman> the data values are inverted but that's not a problem. forgot to negate them for print
[03:36:44] <Bushman> now all what's left is yet again smooth it out a bit and voila!
[03:39:49] <twnqx> congratulations
[06:27:58] <Lambda_Aurigae> http://imgur.com/amMBh4b
[06:28:20] <Thrashbarg> jeez what caused that?
[06:29:20] <Lambda_Aurigae> at a guess, 24V or so.
[06:29:44] <Lambda_Aurigae> that or someone put the wrong magic blue smoke in it.
[06:29:56] <antto> ouch
[07:11:23] <Tom_itx> do they run cooler with the lid off?
[07:13:56] <Lambda_Aurigae> apparently they do.
[07:14:18] <twnqx> what arre we talking about?
[07:14:24] <Lambda_Aurigae> http://imgur.com/amMBh4b
[07:15:44] <twnqx> heh
[07:40:14] <Jartza> I have plenty of 24V powers at home, maybe I should try that with one attiny :)
[07:57:47] <twnqx> DON'T
[07:57:52] <twnqx> noone misses a PIC
[07:57:58] <twnqx> but people cry after AVRs!
[07:58:08] <Jartza> :)
[07:58:19] <Jartza> I don't have any PICs
[07:58:31] <twnqx> then use some intel
[07:58:39] <Jartza> none of those either
[07:58:43] <twnqx> what
[07:58:48] <Jartza> only arms and avrs
[07:58:57] <twnqx> what are you chatting on :S
[07:59:09] <Jartza> amd :P
[07:59:18] <twnqx> an AMD will be fine, too
[07:59:27] <abcminiuser> Anyone know how to inject a #define'd value into inline ASM (GCC)?
[08:01:15] <Jartza> I got some NXP LPCXpresso -boards, but don't have an idea what to do with those :)
[08:01:33] <Jartza> lpc1114, lpc1200 and lpc1769
[08:03:35] <Jartza> I guess most of them are already obsolete as such
[08:05:10] <Jartza> but as those cards have the "LPC-Link", I think that can be used as a standalone JTAG
[09:58:37] <Bushman> damn... why always the simplest of things turn out to be the hardest to figure out?
[09:58:49] <Bushman> i need a variable that will change it's value to the opposite. if 0, make it 1, if 1 make it 0.
[09:58:51] <Bushman> is there a way to do that WITHOUT 2 if statements?
[10:00:10] <Bushman> i'm stupid... var = !var; works
[10:07:04] <Jartza> wuut
[10:07:19] <Jartza> why on earth my timer0 overflow interrupt on attiny85 is too slow
[10:10:33] <Jartza> I'm running attiny85 on 8Mhz, setting the TCCR0B bit CS02 (clk/256) and using overflow interrupt
[10:10:43] <Jartza> and just switching one pin on that ISR
[10:11:01] <Jartza> and I get about 61.48Hz output
[10:11:28] <Jartza> argh
[10:11:32] <Jartza> of course I do
[10:12:07] <_abc_> Anyone familiar with gpsim internals here?
[10:12:21] <Jartza> the isr is being called ~122 times per sec
[10:12:31] <Jartza> that divided by 2 makes it 61.48
[10:12:35] <twnqx> ~123!
[10:12:39] <Jartza> <facepalm>
[10:13:09] <Jartza> </facepalm>
[10:17:44] <_abc_> Why divided by 2?
[10:17:48] <twnqx> on one interrupt, turn on. on next, turn off.
[10:17:51] <_abc_> Ah
[10:18:04] <twnqx> i... made that mistake before.
[10:21:33] <Jartza> yep
[10:21:48] <Jartza> 1Hz needs to go up, and down :)
[10:21:55] <Jartza> so interrupt called twice
[10:22:09] <_abc_> So set the divider to clk/128?
[10:22:25] <Jartza> nah, I just did set the compare value lower
[10:22:35] <Jartza> I actually need around 140-150Hz
[10:22:44] <Jartza> doesn't need to be that accurate
[10:52:06] <Jartza> also my software-pwm for lcd-backlight seems to work
[10:52:42] <Jartza> although, now each color can be adjusted between 0-255, but I need to implement some kind of calculation to make it more simple
[10:53:07] <Jartza> like, just one "color" value between 0-255, so that the backlight intensity would stay about the same
[10:53:16] <Jartza> didn't find any ready made code for that either
[11:06:51] <_abc_> triplets. RGB
[11:07:11] <_abc_> Also 0-255 should be lograrythmic. 0 1 2 4 .. 128 255
[11:26:18] <N1njAway> Jartza: A reasonably close approximation to the log response correction is to simply square the value and shift it back. i.e. value = (value * value) >> 8; -- it helps quite a bit with correcting the low-end, though this admittedly works better if you have a 16-bit intensity resolution :)
[11:56:19] <_abc_> 0*0->0 1*1->0 ... 16*16->1 ... what am I missing? You want more resolution at the low end, not less?!
[11:56:48] <_abc_> Did you use a spreadsheet for this thing?
[11:56:52] <_abc_> N1njaneer:
[12:01:32] <N1njaneer> _abc_: Again this is more applicable to higher-resolution drive to the LEDs (ideally 16-bit), but this scheme will significantly linearize the visually-percived output of the LEDs vs linear number input drive.
[12:02:25] <N1njaneer> _abc_: This scheme is used in thousands of our LED fixtures which are used for high-end architectural installations, and which we can dim down to less than 0.05% brightness with complete linearity. :)
[12:02:59] <_abc_> "complete logarythmness" you mean? :)
[12:03:09] <_abc_> Also someone must have patented it already if it is so good
[12:04:09] <N1njaneer> Yes, Philips / Color Kinetics did, which is why we license over 350 of their patents so we don't have to worry about any legal contention.
[12:05:20] <_abc_> They patented an algorithm... outside Europe I hope... since I am IN Europe and you can't do that here.
[12:05:27] <N1njaneer> Many of their patents are quite sketchy, but they are recognized in multiple countries. Much easier to pay the piper than is it to pay the lawyer. But this isn't significantly different from simple gamma-curve corrections - same principles. This scheme is just particularly good to use on microcontrollers with limited ability to do intensive math.
[12:05:44] <_abc_> blah
[12:05:53] <_abc_> >>8 is making a mess at the low end
[12:05:58] <N1njaneer> Yes, which is why some are recognized in some countries.
[12:06:00] <_abc_> anyway yes it is like log
[12:06:23] * _abc_ suggests >>4 is a better idea for 8 bits >;)
[12:07:55] <N1njaneer> _abc_: It depends on what your final resolution mapping is. 8 bit * 8 bit = 16 bit result. Scale as is appropriate.
[12:08:29] <N1njaneer> We use a proprietary spread-spectrum PWM plus current control to give us better than 16-bit dimming control, so we map the result in to that when that feature is switched on.
[12:09:27] <N1njaneer> We can either do it in the control software or the fixtures. We generally do it in the fixtures because we also upsample the updates-per-second over 10x the control rate, which means we get absolutely smooth dimming responses far above what DMX512 normally allows :)
[12:09:59] <N1njaneer> We can even take a crappy 10fps, 8-bit fader box and make it looks immensely smooth. The approach works well for us, but depends on what your needs are.
[12:10:27] <N1njaneer> But again, your mileage may vary.
[12:12:40] <_abc_> Just small panel indicators with 'cool' fading and sometimes backlights.
[12:12:59] <N1njaneer> Yeah, not nearly as major of requirements then :)
[12:13:59] <_abc_> I am going to plug that into a spreadsheet and look at it later.
[12:14:24] <_abc_> I use bit shifting as is and linear low end up to a threshhold.
[12:14:34] <_abc_> Usually it is okay.
[12:14:47] <_abc_> 10bit pwm can be too little for very smooth low end even
[12:15:52] <N1njaneer> Yes, low-end is always a challenge to correct for. One of the things you can do if you are doing software-based PWM is to actually vary the frequency on the low-end in order to get more temporal resolution to adjust against.
[12:16:44] <N1njaneer> You will just end up competing with preceived flicker if you frequency is too low if you aren't doing some clever spread-spectrum aspects.
[12:19:36] <megal0maniac> Today we started the subject where they teach us about microcontrollers. The lecturer isn't stupid, but he's bloody ignorant. The reasoning for using PIC instead of Arduino (no mention of AVR, he doesn't know what that means) was because PICs are used in industry and Arduinos wouldn't hold up to the extreme conditions because PICs are more robust
[12:19:42] <megal0maniac> I nearly shouted at him
[12:20:20] <megal0maniac> Since the stuff we need to do is VERY basic, I'm going to do it twice. Once on PIC and once on AVR
[12:20:40] <megal0maniac> And lead him out of his ignorant hole
[12:21:03] <megal0maniac> </rant>
[12:25:03] <Keev> I wouldn't want arduinos in a production environment either
[12:25:25] <megal0maniac> Keev: Yeah, but compare apples and apples ffs :/
[12:27:16] <megal0maniac> The other comparison was rpi, which is a totally different thing
[12:30:21] <megal0maniac> *grumbles*
[12:32:49] <N1njaneer> megal0maniac: Tell him PIC has more errata after four revisions of a single chip than AVRs have after one.
[12:33:53] <N1njaneer> There are some particularly notorious silicon bugs in PICs
[12:34:00] <N1njaneer> Like, to absurd levels.
[12:35:10] <megal0maniac> N1njaneer: I'd have to teach him what an AVR is first
[12:35:19] <megal0maniac> But we'll get there
[12:35:20] <N1njaneer> Keev: Most of the Arduino hardware design actually isn't half-bad, and obviously it's all GCC on the back-end, but yeah... Not exactly production-worth stuff, though it happens ALLLL the time. You should try the fun of getting something with an Arduino through UL certs. Been there. :)
[12:36:02] <megal0maniac> Right now I'm looking for a bag with 18 PICs in it and I can't find it :/
[12:36:34] <N1njaneer> megal0maniac: PIC it up out of the garbage?
[12:37:24] <megal0maniac> N1njaneer: Har har
[12:37:44] <megal0maniac> I don't hate PIC. But I do prefer AVR. And I hate ignorance
[12:38:00] <Keev> N1njaneer, i started using avrs since before arduino was a thing. I never used an arduino in my life and i can't find any reasonable reason to change that now
[12:39:26] <Keev> it's a great product to get people in the hobby though, but they're most likely never going to be able to use anything else afterwards (unless they actually want to learn)
[12:40:02] <N1njaneer> Keev: Yes, I've been doing AVR +12 years now. Arduino pisses me off partially because it's another profitable idea I could have easily done but had no idea there would be a demand for that stuff, plus it's all hobbiest-level stuff.
[12:40:40] <N1njaneer> Keev: The GOOD part is that if you can't do your own PCBs, you can always take Arduino hardware and simply hit the AVRs on them through ISP and just erase the bootloader and Do It The Real Way :)
[12:40:59] <megal0maniac> Keev: My first ever purchase was an Arduino Uno and a programmer. I'd do the same thing again.
[12:41:38] <N1njaneer> The thing that blows my mind is that they are intended for hobbiests... and yet they don't have the capabilities of even doing JTAG with deep debugging - stuff you really NEED if you are first learning how the stuff works. And for $35 and free Atmel Studio you get all of that.
[12:42:24] <N1njaneer> Plus an IDE that doesn't blow. The damn thing doesn't even have the ability to do block tab-indent/remove
[12:42:27] <Keev> i never used jtag
[12:43:30] <N1njaneer> Keev: I rarely do (on AVRs) but that's because I started using them before JTAG was really prominent, then when introduced the JTAG ICE used to cost about $400. But JTAG/SWD is extremely handly when you need it, and is necessity when doing stuff more complex than AVR.
[12:44:03] <N1njaneer> But now the new Atmel programmer does AVR and ARM for ISP/SWD/JTAG all in one box, for all of $35
[12:44:12] <N1njaneer> So no reason not to take advantage :)
[12:45:00] <megal0maniac> Wow. Which one is that?
[12:45:25] <Keev> well i still use my $2 usbasp :p
[12:46:52] <N1njaneer> megal0maniac: Newly released one, ISPMK2 has been EoL'd -- P/N# ATATMEL-ICE -- DigiKey and Mouser have them.
[12:48:08] <Keev> http://store.atmel.com/PartDetail.aspx?q=p:10500253#tc:description ?
[12:48:23] <Keev> woops not that one
[12:49:05] <N1njaneer> megal0maniac: It's nice because this new one does both AVR and ARM stuff - jtag for both, ISP and SingleWire for AVR. So several hundreds dollars of programmers all rolled in to one that is less than $100 for the whole thing, and as low as $35 if you just buy the raw PCBs.
[12:49:21] <N1njaneer> Local Atmel FAE swung us a couple :)
[12:50:53] <megal0maniac> Aaaah! Yes, I remember hearing about it
[12:51:14] <megal0maniac> I have the jtagice3
[12:51:25] <megal0maniac> http://fourwalledcubicle.com/blog/2014/03/instant-reactions/
[12:56:43] <N1njaneer> Yeah, pretty much. I have ICE3's here as well, I think it will be longer before those go EoL.
[12:59:46] <N1njaneer> The newer ATMEL ICE may not be perfect, but the price point is surely nice.
[13:00:19] <N1njaneer> If you need something more robust, you have the good old SAM ICE, or just go to Segger and spend $$$
[13:03:06] <megal0maniac> Woah! This is really cool!
http://store.atmel.com/PartDetail.aspx?q=p:10500376;c:100112
[13:04:14] <megal0maniac> Goodbye, Dragon
[13:06:18] <N1njaneer> Exactly!
[13:07:37] <Keev> http://store.atmel.com/PartDetail.aspx?q=p:10500379#tc:description
[13:07:50] <Keev> on the other hand this is very expensive
[13:08:02] <Keev> $39 for 3 cables
[13:08:32] <Jartza> okay, got the rgb working with simple function
[13:08:37] <Jartza> giving value between 0-223
[13:08:40] <N1njaneer> Yep, so just make your own if your time and materials costs less than $40. In my case it doesn't so I'd prefer to just have them out of the box :)
[13:13:29] <Jartza> but it doesn't have hvsp?
[13:18:51] <N1njaneer> Jartza: Not sure if it does, but I've never used or needed HVSP in over a decade of doing AVR stuff :)
[13:19:03] <megal0maniac> It won't
[13:19:17] <megal0maniac> Strangely, the JTAGICE3 doesn't do TPI either
[13:20:52] <Jartza> N1njaneer: I do as I use a lot of attinys and it's nice to have an extra pin instead of reset ;)
[13:21:26] <Jartza> well, I still use the one that I built, then... :)
[13:26:55] <N1njaneer> Indeed!
[13:28:22] <Jartza> ...as long as it works
[13:36:11] <megal0maniac> I still want to see more of these in the wild:
https://sites.google.com/site/megau2s/
[13:59:58] <megal0maniac> N1njaneer: Probably the most annoying thing about MPLAB X is that it doesn't recognise __delay_ms (and various macro functions) so I have red squiggly lines all over my perfectly valid code.
[14:00:40] <megal0maniac> It's small, but I just can't deal.
[14:39:02] <rewolff1> Has anybody seen this: I flash a certain program into my 't44, and then the t44 stops responding before I download the eeprom contents. If I wait a while, it will react ONCE. So if I use that opportunity to program a different binary, it works again, if it is my script probing for a connected AVR it will detect and then fail-to-program....
[14:40:12] <Casper> it may be a too fast clock source for your programmer
[14:40:17] <Casper> common with internal RC oscillator
[14:44:05] <rewolff1> Nope: 1) I program it very slow. 2) I normally start out with unflashed (i.e. internal 1MHz) processors and that works. 3) .... scratch that..... BIngo. I think you're right.
[14:44:18] <rewolff1> Thanks.
[14:48:06] <Casper> the default clock rate is for 1MHz I beleive
[14:48:14] <Casper> but many parts actually run a bit slower than that
[14:48:22] <Casper> due to the wide tolerance...
[14:48:42] <rewolff1> Yeah, but Atmel specifies "clock / 4" for max SPI clock,
[14:49:04] <rewolff1> which I THINK from an engineering standpoint should be "clock / 2".
[14:49:04] <Casper> yeah
[14:49:11] <rewolff1> They specify /4 to have some margin.
[14:49:30] <Casper> possibly
[14:49:32] <rewolff1> So at 999kHz, i't work below 499 kHz,
[14:49:37] <Casper> bbl
[14:49:41] <rewolff1> but they specify 250kHz just to be sure.
[14:49:44] <rewolff1> ANyway....
[14:49:59] <rewolff1> My code was configuring the processor for clock / 256.
[14:50:05] <rewolff1> Duh!
[14:50:15] <rewolff1> So way slower than RC/8 that's the default....
[14:52:14] <N1njaneer> That would do it, yes :)
[14:52:31] <rewolff1> Grrrr... not funny.... .... :-)
[14:52:51] <N1njaneer> You can have the internal clock as fast as you want - it just needs to be at least greater than 1/8th the rate of the SPI clock for ISP in order to drive the state machine fast enough :)
[14:52:52] <rewolff1> Been banging my head for a long time.....
[14:53:29] <rewolff1> So how do I specfiy a slower clock on AVRDUDE?
[14:53:42] <rewolff1> I thought it was -b or -B ....
[14:53:56] <rewolff1> but both give me in hte kilohz range clocks...
[14:54:09] <rewolff1> while I specify -b 50 now....
[14:54:28] <rewolff1> (-b 1200 didn't get the avr's attenttion..... )
[14:54:39] <N1njaneer> Not sure as I only use Atmel's tools
[14:54:52] <N1njaneer> Command line options are probably different.
[14:55:56] <rewolff1> Hmm. -B specifies microseconds per clock instead of bits-per-second.
[14:56:04] <rewolff1> Lets try a bigger number. :-)
[14:56:44] <rewolff1> No change.
[14:56:58] <rewolff1> But apparently the -B option only applies to ONE programmer . :-(
[14:58:13] <rewolff1> I'm calling it a day. Thanks for the help!
[15:49:26] <Jartza> ahh
[15:49:34] <Jartza> I just love these chinese datasheets
[15:51:01] <Jartza> "IIC interface: It just only could write Data or Instruction to ST7036 by the IIC Interface. It could not read Data or Instruction from ST7036 (except Acknowledge signal)."
[15:51:11] <Jartza> :D
[16:16:24] <vsync_> wow. i just figured out someone sings danny boy at the end of "one after 909", on let it be
[16:16:42] <vsync_> just saying. keep going ~
[16:28:54] <A124> Hello. How can I ASM without Atmel Studio, please? .D
[16:30:04] <vsync_> notepad
[16:30:20] <vsync_> and high voltage programming (~400V)
[16:30:45] <kastein> I recommend the tesla coil approach
[16:32:45] <vsync_> you need a 1 farad cap
[16:32:53] <vsync_> rated at 300 volts
[16:32:58] <vsync_> a 1 kv transformer
[16:33:52] <A124> Ha, ha, ha.
[16:33:52] <vsync_> 2 girls 1 cap
[16:34:08] <A124> Well I meant the compilation.
[16:34:37] <A124> Found the ASM part of the studio. Is there any alternative?
[16:36:45] <vsync_> wtf. avrgcc?
[16:37:34] <acecase> I got my Atmel-ICE last week (thanks for the tip N1njaneer), and I keep reading about a condition where I can turn on DWEN (enable debugwire) and leave my part in a state where I may have to go to extreams to connect ISP again. Can someone explain how I get to that state?
[16:37:54] <myself> I was gonna suggest a pad of paper and a printout of the instruction set reference
[16:38:39] <A124> myself: Not the fastest, haha.
[16:38:58] <myself> you did say ANY alternative
[16:39:01] <acecase> I have been going Debug->Disable DebugWire and Close every time I stop a debug session. Every time.
[16:39:03] <myself> if you wanted fast you'd use avrgcc
[16:44:08] <A124> myself: Can you refer me to any page? Also I do not see the need for gcc if ASM
[16:44:18] <Casper> A124: joke appart, any text editor and avr-gcc, use avrdude to program with "any" programmer. If you don't have one and you have a parallel port you can make one with 3 resistors
[16:44:53] <A124> Casper: Thanks. So one does need gcc in all cases?
[16:45:11] <Casper> there is an assembler that come with avr-gcc, but I never used it so I do not know it's name
[16:45:41] <A124> Oh. Ok. Thought it was little uneccessary to have whole toolkit for just ASM.
[16:46:16] <A124> Now looking onto simulation without the Studio, if anyone has any hint, will be glad.
[16:46:45] <ambro718> Hey. In avr-gcc inline asm, how do I store a label address to Z pointer?
[16:46:57] <Casper> friendly advice: forget about simulators, they are inacurate
[16:47:32] <Casper> you are better to use other mean of debbuging, like serial output, leds, lcd, jtag or debugwire
[16:47:36] <acecase> A124, I looked quite a bit a weak ago, and found no free ones worth messing with
[16:47:39] <N1njaneer> Yep.
[16:48:18] <acecase> Be careful with debugWire :) ^
[16:48:33] <A124> acecase: Have you find any "unworthy"? .. Or at least code (and register memory) simulation without HW?
[16:49:27] <A124> Casper: Do you have any advice for small, compact and if possible low cost debugwire debugger?
[16:49:37] <acecase> SimulAVR was a free open source one that I found. Supposedly not great as they are saying though
[16:49:43] <Casper> I do not use debugwire
[16:49:48] <Casper> I use serial out and leds
[16:49:52] <A124> ok :)
[16:50:22] <acecase> I just got a ATATMEL-ICE for like $50 that does debugWire on AVR.
[16:50:31] <A124> acecase: Thanks will check that. Not looking for almighty. Aka peripherals, just code and internals plus basics like ports and interrupts.
[16:50:40] <acecase> N1njaneer, recommended it. It's pretty nice for the price range
[16:51:08] * Casper uses a ponyprog serial that I made myself
[16:51:12] <A124> acecase: Little cotly for me at the time, but seems ok. Pirate can do that also right? And both are Atmel products if not wrong?
[16:51:30] <acecase> If you do use DebugWire, remember to be careful though. There is some way that you can leave an avr in a state where you have to go to extreams to connect ISP again
[16:52:32] <Casper> debugwire disable isp and reset
[16:54:10] <A124> I am aware of that, thanks.
[16:58:34] <A124> acecase: I found ICE for $85 though only and UP.
[16:58:40] <acecase> I keep reading how if you do something without unsetting the DWEN bit it is hard to get back to it to enable ISP. Can't find anything clear about it though, so I just unset it every time I stop debug execution
[17:00:06] <acecase> Search digikey for ATATMEL-ICE
[17:00:16] <acecase> first AT is actually there
[17:01:13] <acecase> The big kit is around 80. You can get it with the basic cable and a case for 50(ish) and without the cable or case for 35
[17:01:29] <A124> I am Europe resident though. Digikey is NA
[17:01:44] <A124> Oh. Had no idea about that.
[17:01:54] <acecase> I'm sorry. <- typical american. <- thinks everyone on the internets is american
[17:01:56] <A124> Thanks
[17:03:37] <acecase> I do agree with these guys though. I'm a complete newb. Been playing with them since Friday. And today I figured out how to hook up and write to an LCD, and will probably never connect via debugWire again
[17:04:19] <A124> Why so?
[17:04:57] <A124> Myself what I find the most standing in the way is access to the information how to do this and this.
[17:05:09] <acecase> I just don't see the need unless I can't figure out why a value is what it is or something. Output to LCD is faster/easier
[17:05:56] <A124> acecase: And you are doing C++ or ASM?
[17:06:15] <acecase> It's a large barrier to entry, to me. A lot of new terminology and so much information that isn't relivant to fish through. Once you get the basics and learn how to find things as you need them, the information is out there
[17:06:20] <acecase> C
[17:06:40] <acecase> ASM is hard and C++ is "risky" <- or so I read
[17:06:56] <A124> Yeah every has its specific.
[17:07:17] <A124> acecase: Now you might get the use of debugger, right?
[17:07:52] <acecase> Maybe. Better to have and not need than to need and not have. Still glas N1njaneer pointed me to this instead of the MKii
[17:08:33] <acecase> Another nice bonus is the SAM support
[17:09:45] <A124> Yeah, MKII is good when you want to program. But.. thats it. :D
[17:15:05] <N1njaneer> Which is also why the MKII has been EoL'd -- costs too much and does too little with the newer programmer options becoming avaliable :)
[19:14:17] <dkordic> A124: For example there is #simavr channel. Algorith Builder (freeware) is a bit interesting, but I wouldn't use it.
[19:18:20] <dkordic> I was searching quite a bit for assemblers, but found nothing useful enough. Now I am making my own. It is much easyer than I thought.
[19:19:03] <Lambda_Aurigae> avra
[19:19:15] <Lambda_Aurigae> good old reliable avr assembler.
[19:20:14] <dkordic> A124: Check this out
http://www.modk.it/micro/editor/
[19:31:54] <Lambda_Aurigae> dkordic, what the heck is that?
[19:32:42] <dkordic> JavaScript (SaSS) simulator.
[19:33:12] <Lambda_Aurigae> SaSS?
[19:33:21] <Lambda_Aurigae> it looks like an arduino simulator.
[19:33:26] <dkordic> There is also this
http://mbeddr.com/ bloatware. I have no idea what it is doing.
[19:33:56] <dkordic> Yes, Arduino simulator. Service as Software Substitute.
[19:34:16] <Lambda_Aurigae> yeah, no thanks.
[19:34:26] <Lambda_Aurigae> if I need a simulator there's always simulavr.
[21:42:24] <eexp> hi, anyone help. I can not reset clock fuse after change lfuse, chip is attiny13a and use usbasp cable.
[21:42:49] <Tom_itx> what did you set it to?
[21:43:26] <eexp> $(AVRDUDE) -U lfuse:w:0x69:m this is in my makefile
[21:43:38] <eexp> i change osc to 4.8M.
[21:44:15] <eexp> now i can not progam it anymore.
[21:44:33] <Tom_itx> default is 0x6A
[21:44:35] <eexp> AVRDUDE=avrdude -p t13 -c usbasp
[21:44:48] <eexp> yes, i change to 0x6b
[21:45:07] <eexp> yes, i change to 0x69. sorry
[21:45:16] <Tom_itx> that's internal osc
[21:45:52] <eexp> so it works well with 4.8M. but i can not reprogame it to 9.6M and can not download new hex file.
[21:47:05] <Tom_itx> maybe try -B32
[21:47:19] <Tom_itx> on the avrdude line
[21:47:20] <eexp> fpaste.org/121573/
[21:47:32] <eexp> ok, i had try -B250
[21:48:08] <eexp> avrdude: set SCK frequency to 16000 Hz
[21:48:10] <eexp> avrdude: warning: cannot set sck period. please check for usbasp firmware update.
[21:48:27] <eexp> when use -B32, sck is so high.
[21:48:53] <eexp> and i got "error: programm enable: target doesn't answer. 1 "
[21:49:19] <eexp> i think i need use low frequency?
[21:50:01] <Tom_itx> i dunno much about usbasp except it's slow
[21:51:06] <eexp> the only way i find is reduce sck frequency. but it not work with usbasp, maybe?
[21:51:45] <Tom_itx> it was worth a try
[21:52:09] <Tom_itx> still 4.8Mhz should be fast enough for a default spi sck
[21:52:25] <Tom_itx> should work up to 1Mhz sck
[21:52:40] <Tom_itx> see if avrdude has some other type of delay
[21:52:46] <eexp> yes, i think so.
[21:53:41] <eexp> -i <delay> ISP Clock Delay [in microseconds]
[21:53:45] <eexp> perhaps i need try this?
[21:55:09] <eexp> seems this usbasp "cannot set sck period".. :(
[21:56:02] <Tom_itx> it can't because it can't communicate with the chip now
[22:00:53] <eexp> sad, another PC run winxp, can reprogame lfuse...
[22:07:17] * Getty waves to brabo
[22:07:38] <brabo> hey Getty
[22:19:58] <brabo> hi, i'm having a rather bizare issue.. coding on a atmega8 and if i change the content of one char array i.e. have 'efs10' in the string is messes up the whole string i'm making but if i say use 'kty81' it works fine, even if i do not run the subroutine it is in.. any ideas?
[22:20:37] <brabo> i'm using different routines to construct a string, and the communication is over usart