Back
[00:13:59] <splud> beh. Compiling some rather trivial code for an ATTiny13A.
[00:14:12] <splud> Using the Arduino IDE with Tiny core installed.
[00:14:49] <splud> Code is all of 280 bytes or so and then I simply add a second call to a function (with one param), and the code balloons to over 1KB.
[00:15:45] <splud> This is a call ALREADY being made (basically a blinker, and I want to blink 3 times for instance). I can do it from a for loop, or just repeat 'realblink(iLED)' on three consecutive lines.
[00:16:51] <splud> 296 -> 1120, from repeating one line of code.
[00:17:43] <splud> and if I comment out the first occurrence entirely (not calling the function), then the binary is 158 bytes.
[00:20:40] <Hexum064_2> and now splud knows why we shy away from arduino
[00:23:29] <splud> Thanks for the helpful input.
[00:25:50] <Hexum064_2> sry man. snarky remarks all around.
[00:26:22] <Hexum064_2> The point is, that's probably a quirk of the arduino framework, and you may not be able to get around it.
[00:42:02] <splud> I expect to dig up proper toolchain for the AVR, as I also want to program the devices in asm.
[00:42:41] <splud> Just odd as heck if I call the same function twice in a row, the code size goes up by 5 fold.
[00:43:18] <splud> Have the code down to 170 bytes.
[00:48:15] <Casper> splud: you said the call is already there? you mean outside the main? or in the main?
[00:58:22] <DKordic> splud: paste?
[01:48:41] <rue_house> splud, you know how the optimier works right?
[01:48:56] <rue_house> ^z
[02:04:13] <splud> I was away from console doing tests.
[02:04:37] <splud> Managed to get it to work, still unclear what the heck the deal was.
[02:04:54] <splud> Possibly cacheing of the intermediate files.
[02:06:14] <splud> I've done several projects using ATTiny85, 84 and 24V. This is a first for 13A.
[02:06:27] <splud> the core was sourced from elsewhere.
[02:07:26] <splud> (as versus the cores for the other 'tinys).
[02:08:44] <redspl> hi
[02:10:31] <splud> Im unfamiliar with the specifics of the avr-gcc optimizer. Was looking at some commandline options earlier, address wrap and some consecutive returns and such. Decided none of that had any bearing on the result I was seeing.
[02:12:14] <splud> Anway, have it at 182 bytes. Not nearly as optimized as it should be, and I still need to put in some low power sleep code (I've done that on ATTiny85 for an average of like 7uA), plus reading of an analog pin to sense daylight.
[02:14:22] <splud> This particular project is some stroboscopic stakes to ward off deer.
[02:14:26] <splud> Hopefully <g>
[02:15:02] <splud> More an exercise in doing it and making light dispersion via edge-lit acrylic.
[02:17:04] <DKordic> redspl: Hello.
[02:17:32] <splud> First made a PCB with 4 LEDs, Tiny randomy flashes one, delays for 'x' plus random additional delay, then repeats the process of selecting an LED to flash. Realized if I laser-cut an acrylic disc, along with a hole in the center, affix an LED in there and line the top and bottom with foil, I'd get good dispersion out 360 degrees from the sides.
[02:20:24] <splud> Received 50 devices in the post over the weekend, and was dissappointed to find that I couldn't program any of the samples. Came to realize that despite being in tape (as in tape reel), they must have been pre-programmed and had the reset disabled. Sure enough after hitting them with an HVP I had to cobble up to reset the fuse bits, they were then capable of being programmed.
[05:55:28] <Lambda_Aurigae> "Please submit to this security check. You are reminded that you DON'T have to take part in this check, in the same way we don't HAVE to shoot you if you try to pass without one."
[09:21:20] <rue_house> but in the fine print it says that all shootings are done with a 25MP SLR...
[09:29:14] <redspl> GUYS I GOT THE SONG NAME FROM "Extreme prototyping board wiring techniques"
[09:29:24] <redspl> クローバー・クラブ - ゆうゆ
[09:30:38] <redspl> if anyone wants :PP
https://www.youtube.com/watch?v=QWT0y2YE8W8
[09:31:24] <Strangework> hehh this is awesome
[09:56:01] <m3chanical> Wow. I've never seen (to be fair I'm new to this realm) prototyping done so fabulously on a board like that. Here I am struggling with through-hole components on those!
[11:06:31] <Hexum064> Any know if it's possible to set the source or destination address for a DMA channel to a port? (like PORTC.IN for the source or PORTC.OUT for the destination)?
[11:06:41] <Hexum064> I haven't tried it yet but it would be good to know.
[12:56:24] <bss36504> Andybody know of either a modern, speaker-independent speech recognition IC or a piece of stand alone software that could be compiled onto an ARM or some such processor (without relying on "the cloud")?
[14:15:40] <redspl> what's up, guys? :)
[15:08:40] <Jartza> https://drive.google.com/file/d/0B2dTzW9TMeBxSzI3X3c0aURqTWs/view
[15:08:49] <Jartza> this' up
[15:09:06] <Jartza> fresh from factory
[15:11:38] <Tom_itx> so what does it do?
[15:12:36] <Jartza> http://tagsu.io/about
[15:13:03] <Tom_itx> just what everybody needs
[15:13:34] <Jartza> nobody needs it :)
[15:13:54] <Jartza> but still people bought them, it's just fun
[15:16:42] <Strangework> hehhh, damn cute!
[15:20:15] <Jartza> https://drive.google.com/file/d/0B2dTzW9TMeBxMGZRWHZXSUxqbmM/edit?usp=docslist_api
[15:24:09] <Jartza> and screens can be updated with cellphone, tablet, laptop, pc.. almost anything that has browser and headphone connector
[15:24:38] <Jartza> or you can save your screens as mp3 and even play them from mp3-player :)
[15:24:57] <Jartza> and update the firmware too by playing audio
[15:47:00] <Strangework> whoa :o I've never heard of headphone jacks being utilized in such a way. Clever touch
[15:48:16] <Tom_itx> you never had a pc with cassette tape storage?
[15:48:39] <Tom_itx> old tech with a coat of wax
[15:49:38] <Strangework> No - I am not sure if I've seen one, actually!
[15:51:25] <Tom_itx> https://en.wikipedia.org/wiki/Magnetic_tape_data_storage
[15:51:35] <Strangework> Curious how closely I associate specific tech with certain applications. I likely wouldn't haven't been so impressed by the headphone trick if I had more contact with older/other techniques
[15:52:45] <Tom_itx> iirc it may have been commodore that used standard audio tape
[15:52:55] <Tom_itx> not very reliable
[15:53:32] <Tom_itx> https://en.wikipedia.org/wiki/Commodore_Datasette
[15:55:01] <Strangework> Magnetic tape must be fun to watch. Sometimes I find that lack of mechanical change in my electronics to be a bit disassociating
[15:55:23] <m3chanical> somewhere i heard that magnetic tape storage can hold data for a ridiculously long time (compared to hard drives, ssd, etc.) is that true?
[15:55:24] <Strangework> I just kinda sit there and stare at my silent, unmoving chunks of tech - they flicker lots, though, so there's that
[16:00:59] <Jartza> Strangework: yea, nothing "magical" in the audio-data transfer. except maybe the simplicity of the tech.
[16:01:25] <Jartza> Strangework:
http://labs.rakettitiede.com/?p=87
[16:01:40] <Jartza> there's the data transfer explained.
[16:05:21] <Strangework> Fuck. This is cool. Thanks for sharing, Jartza!!
[16:05:53] <Jartza> np
[16:16:47] <Jartza> I'm happy if someone finds that useful
[16:33:44] <Hexum064> Now, Jartza, if you could only help me find a way to communicate between to xmegas at 1MByte/s
[16:34:18] <Hexum064> Apparently, just because I can send out 8mbits/s over spi does not mean I can receive that fast.
[16:36:07] <Strangework> Hexum064, progress update? :)
[16:37:28] <Hexum064> Well, I got the chip receiving USB data at about 850kbytes/s, which is going to be just fast enough for 4 screens at 75x45 pixels at 20fps.
[16:37:53] <Hexum064> I even got it to pump out data at that rate as well via SPI
[16:38:03] <Hexum064> Thank you atmel for DMA on the xmegas!
[16:38:11] <Hexum064> Then I hit a road block.
[16:38:33] <Hexum064> I cannot get my other xmega, also running at 32mhz, to receive data that fast.
[16:39:01] <Hexum064> Soooo, I'm going to see if I can just switch over to parallel communication
[16:39:37] <Hexum064> If I can get close to 1mbyte that way, I'll be happy.
[16:40:20] <Jartza> hmmh
[16:40:36] <Jartza> I have some xmegas at home
[16:40:44] <Jartza> but I'm still on holiday trip
[16:40:54] <Jartza> so I can't really help currently
[16:41:18] <Hexum064> np, I appreciate that though
[16:41:52] <Jartza> ferry home leaves tomorrow :)
[16:41:55] <Strangework> Why is that reception is much slower?
[16:42:29] <Hexum064> I'm not actually sure.
[16:42:46] <Hexum064> I can't think of anything else to check either.
[17:06:13] <redspl> hi Hexum064!
[17:06:21] <redspl> remember that video with cool soundtrack?
[17:06:41] <redspl> https://www.youtube.com/watch?v=QWT0y2YE8W8
[17:22:57] <Hexum064> hey redspl
[22:02:25] <tis_me> On a ATtiny85, I'm trying to get USI 3-wire + timer as the clock, does anyone know if the OCFIE0A flag/interrupt clocks the data register in hardware or I need to add a software USICLK in the interrupt.