Back
[04:09:33] <hdas> I have atmega168a I am trying to output a basic rectangular wave by set 1 .. delay .. set 0 .. delay
http://pastebin.com/JHtSTZuW but I see such weird signal output on an oscilloscope
http://oi50.tinypic.com/2r3i4p2.jpg .. what am I doing wrong in the code?
[04:15:47] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/how_to/atmega168/mega168_led_blink_delay_index.php
[04:15:50] <Tom_itx> try that one
[04:17:09] <hdas> Tom_itx: thank you
[04:49:41] <hdas> Tom_itx: why do I get the signal on ping 14 (PB0) but I cannot get any output on pin 15 (PB1) with your code?
[05:19:40] <hdas> I am using #include <util/delay.h> library and using _delay_us(462); function and toggle the output. however, the time between the toggles is 454 usec, where I set 462 usec. how can I get precise delay?
[05:21:23] <RifRaf> fussy
[05:25:09] <hdas> RifRaf: here is the code that I am using
http://pastebin.com/JHtSTZuW on atmega168a. When I measure the output on an oscilloscope the time between the transaction is 454.0 usec. but I set 462 usec delay. do you know if _delay_us() fcn should provide very precise output? If it is the case, do I need to set something else on fuse or somewhere else? or is it possible to get a more precise delay by using a different method?
[05:26:10] <RifRaf> i have always just trusted that delay_us would do as it should, and was always close enough for wat i needed, no no
[05:26:15] <RifRaf> so no*
[05:27:27] <RifRaf> i am not in the most intelectual state right now
[05:27:35] <RifRaf> but i did make a thing today
[05:27:38] <hdas> RifRaf: so do you think that 8 usec inaccuracy is normal in my case?
[05:27:55] <RifRaf> this thing
http://imagebin.org/248530
[05:28:18] <RifRaf> well does 8us really matter in the big scheme of things?
[05:28:58] <RifRaf> if it does then you have work to do
[05:29:45] <RifRaf> 20000uS means something to me
[05:31:54] <hdas> I am planning to output some data, 8 usec might cause a synchronization problem. I am very new on microcontroller programming, I wanted to ask experience people if it is normal to have such inaccuracy
[05:32:48] <RifRaf> hdas then you will have to be patient, cause i am just a hacker, but others will be able to help you better in time
[05:35:12] <hdas> thanks RifRaf, I guess I need to take a look at timers and interrupts. do you know any good source to study for atmega168a programming for newbies?
[05:37:10] <turutk> hdas: i am also a newbie and i have been messing with timers and counters for the last week
[05:37:38] <RifRaf> hdas this may be a decent start point
http://www.societyofrobots.com/programming_timers.shtml
[05:37:49] <turutk> hdas: my guess is that your problem is about clock accuracy
[05:38:14] <hdas> thanks RifRaf
[05:38:17] <hdas> turutk: right
[05:39:19] <turutk> hdas: i was trying to make a real time clock. it had an inaccuracy of 10 seconds a day
[05:39:23] <Malinuss> hdas, you don't want to use delay for such a thing, you propably want a CTC timer, with toggel on compare
[05:39:48] <Malinuss> I'm pretty sure it has nothing to do with clock accuracy
[05:40:00] <Malinuss> his problem is using delays
[05:40:31] <turutk> hdas: changing the crystal decreased the inaccuracy to 1 second per day
[05:41:09] <Malinuss> turutk, did you use the internal oscilator in the beginning, or did you just have a shitty crystal?
[05:41:18] <hdas> I am using atmel studio 6 and SUT_CKSEL is INTRCOSC_8MHZ_6CK_14CK_65MS, do you suggest me to another setting?
[05:41:19] <turutk> Malinuss: shitty crystal
[05:41:51] <turutk> http://www.fourwalledcubicle.com/AVRArticles.php
[05:42:02] <turutk> this page is a nice guide for beginners
[05:42:34] <Malinuss> hdas, if you want better precision you propably want a crystal yeah. That isn't the issue in this setup, you just need to use timers instead of the delay function
[05:43:29] <turutk> Malinuss: at first i made the clock using delays which made an inaccuracy of 1 minute per hour. timers and interrupts are way better solution as you also say
[05:44:31] <hdas> thank you turutk. Malinuss, do you know any example code or instruction that I can take a look at to get a precise delay?
[05:45:47] <hdas> btw, when I build my code, I see this warning "Warning 1 #warning "F_CPU not defined for <util/delay.h>" [-Wcpp]" how can I define F_CPU?
[05:46:47] <OndraSter> #define F_CPU 8000000
[05:46:51] <OndraSter> if you have 8MHz clock
[05:47:05] <OndraSter> put that before #include <util/delay.h>
[05:47:20] <turutk> OndraSter: sometimes they add UL to the end. what does that mean?
[05:47:26] <OndraSter> unsigned long
[05:47:40] <OndraSter> I have never had to do that
[05:47:45] <OndraSter> on the other hand I am not using delay.h :)
[05:47:48] <OndraSter> anymore
[05:49:07] <hdas> thank you OndraSter
[05:49:14] <Malinuss> hdas, you don't want "precise delays"... you want "precise timing", and not use delays ;D
[05:49:52] <turutk> mine didn't work without the UL
[05:50:39] <hdas> Malinuss: thanks for your advise, I will take a look at timers now
[05:51:50] <Malinuss> hdas, if you want to swtich something on and on at x Hz, you want timer, yes. take your time to understand them, and all their functions. Then move on to PWM
[05:53:52] <hdas> if I don't add #define F_CPU 8000000, what is the default value for F_CPU?
[05:57:34] <hdas> Malinuss: I will learn timers, as a newbie it is good to get advise, thanks
[06:00:51] <Malinuss> hdas, the default is 1M
[06:01:18] <hdas> thank you
[06:01:33] <Malinuss> hdas, you can also see that in the delay.h
[06:03:01] <hdas> thanks
[06:56:05] <megal0maniac> hdas: Check out
http://deans-avr-tutorials.googlecode.com/svn/trunk/Timers/Output/Timers.pdf
[07:27:51] <Tom_itx> i wonder if he had F_CPU defined
[07:28:29] <Malinuss> Tom_itx, he didn't
[07:28:37] <Tom_itx> hdas, it has no default
[07:29:01] <Malinuss> Tom_itx, I think it defaults to 1M in delay.h if not defined already. Almost 100% sure
[07:29:11] <Tom_itx> k
[07:37:57] <hdas> what library includes CLI(), SEI() functions?
[07:38:10] <Malinuss> hdas, interrupts.h
[07:38:30] <hdas> Malinuss: thank you
[07:38:56] <Malinuss> hdas,
http://www.nongnu.org/avr-libc/user-manual/modules.html
[07:39:14] <hdas> that's great, thank you
[07:40:29] <Malinuss> it is ;D
[08:45:09] <hdas> I am reading an example code on
http://www.protostack.com/blog/2010/09/timer-interrupts-on-an-atmega168/ it says " TCCR0B = _BV(CS02) | _BV(CS00); // Clock/1024, 0.001024 seconds per tick" the text above the code says "An ATmega168 with default fuses runs at 8MHz with the system clock prescaler enabled. This means that the timer can be updated, up to 8,000,000 times per second." but it is calculated as 1MHz to find 0.001024.
[08:45:21] <hdas> when I need to prescale the clock, do I need to refer 8 MHz or 1 MHz
[08:46:55] <hdas> on atmel studio: Tools > Device programming > Fuses > SUT_CKSEL shows INTRCOSC_8MHZ_6CK_14CK_0MS
[08:52:16] <Malinuss> hdas, yes your F_CPU is set to 8MHz, but by default there is a prescalar set to F_CPU/8 so it actually runs at 1MHz.. So the F_CPU will be running at 1MHz, and so will all the timers
[08:53:31] <hdas> thank you Malinuss
[08:53:38] <Malinuss> hdas, if you search the datasheet for the F_CPU prescalar, it shows you how to set it.
[08:54:25] <Malinuss> hdas, yeah it's kinda confusing. but it just gives you even more flexibility to change the clock speed. So you can actually change the speed of the clock while running by prescaling it
[08:56:12] <hdas> Thank you Malinuss, I understand it now. do you know what does F stands on F_CPU?
[08:56:25] <Malinuss> frequency ;D
[08:56:31] <hdas> thanks :)
[08:56:43] <Malinuss> I'm just guessing - it would make sense though, haha
[08:57:11] <hdas> it is reasonable :)
[09:31:36] <osfd> Hi there. I can't find this piece of information. Do I need to burn the fuses when I FIRST program an atmega or attiny ?
[09:31:50] <osfd> please
[09:32:21] <specing> only if you don't like them
[09:33:39] <osfd> specing, should I understand my question is properly stated ?
[09:34:30] <Malinuss> lol specing
[09:34:45] <specing> KILL IT WITH FIRE :D
[09:35:22] <Malinuss> osfd, yes specing is trolling you - you don't need change the fuses no. Normally they are set to internal oscilator, and no BOD, and some other stuff that you propably want to leav alone
[09:36:05] <specing> hence if you "don't like them"
[09:36:24] <Malinuss> osfd, you can easliy find "this piece of information" if you look at what the fuses do in your specific uC, in its datasheet
[09:36:36] <specing> you don't manage to provide a good answer and trolling in the same sentence every day
[09:36:51] <osfd> Malinuss, so, for a first flash, no need to change the fuses... but if I need, for example, external osc, then I should check the fuse state and change it if needed
[09:37:08] <specing> just don't disable SPI
[09:37:17] <specing> never ever disable SPI
[09:38:24] <Malinuss> osfd, yes. the default setup is normally pretty sane
[09:38:55] <Malinuss> also what specing said on which I want to elaborate and say : DON'T TURN OFF THE RESET PIN ;D
[09:39:09] <Malinuss> that will disable SPI ;D
[09:39:18] <osfd> Malinuss, ok
[09:39:48] <osfd> I'll try to remember that
[11:07:22] <creep> h
[11:08:14] <creep> which one do you prefer? the clip-style/self-locking, or the 2 metal screw style 230V cable connectors ?
[11:26:50] <theBear> all mains connections should be screwed
[12:31:26] <Posterdati> hi
[12:33:10] <Steffanx> lo
[12:33:22] <Posterdati> please I need help programming an arduino uno without using arduino ide: I was able to compile an arduino project using arduino libraries and a custom made makefile, but programming it make the atmega328 freezes
[12:34:58] <RifRaf> unplug other hardware that may share the isp pins
[12:35:40] <RifRaf> maybe try erase before flash, avrdude -p m328p -c yourprogrammer -e
[12:39:02] <Posterdati> RifRaf: isp? The arduino uno is alone
[13:19:36] <r00t^home> creep: "clip-style/self-locking"?
[13:20:07] <r00t^home> theBear: actually screws aren't good, they can become loose and all... spring-loaded connections are better
[13:20:56] <creep> rootyeah those have some wings that self-lock in a tight fitting hole for the connector
[13:22:21] <creep> it is more common in printers
[13:39:37] <OndraSter> that was a quick method!
[13:39:40] * abcminiuser obligatory "That's what she said"
[13:39:44] <OndraSter> :)
[13:39:48] <OndraSter> that was what I was pointing towards
[13:40:17] <OndraSter> abcminiuser, "Start without debugging" option = upload the code only, I presume?
[13:40:23] <OndraSter> I searched for that button since... ever :D
[13:40:25] <OndraSter> (AS6)
[13:43:58] <specing> Posterdati: #arduino
[13:45:27] <abcminiuser> OndraSter, yes
[13:45:33] <abcminiuser> It's always been there :P
[13:46:25] <OndraSter> yeah :P
[13:47:24] <r00t^home> creep: _printers_?
[13:48:17] <OndraSter> there we go
[13:48:21] <OndraSter> F6 now does upload code
[13:51:38] <abcminiuser> Everyone install the new beta already :P
[13:52:03] <OndraSter> nope
[13:54:36] <creep> r00t^home<< printer power input.
[14:16:35] <hdas> I need to output 1s and 0s (e.g. 11011010011) from a pin of atmega168a, but first I need to store the value. I am confused how I should store the value, I would like to store it as unsigned data[18] = "1001010...10"; however unsigned int is stored in 2 byte. Is there a better way to store the value (18 bit)?
[14:17:44] <krphop> uint8_t data[18] ?
[14:17:58] <tzanger> store it as a 32-bit int and shift it out
[14:18:02] <tzanger> why store it as a string?
[14:20:50] <specing> Oh my, people don't know how to handle bits
[14:20:54] * specing facepalms
[14:22:02] <hdas> specing, I am new in microcontroller programming, I'm learning it. normally I do linux programming.
[14:22:41] <hdas> thank you tzanger, shifting is easier than storing data in array
[14:23:30] <specing> the facepalm was for krphop
[14:23:51] <krphop> heh
[14:24:20] <krphop> maybe he needed the individual bits as a string
[14:24:45] * Tom_itx gives specing a bowl of bits
[14:25:00] <abcminiuser> Uh oh, Stargate's bringing out the dramatic tuba
[14:25:05] <specing> hdas: you should know how to store bits regardless of the architecture you are programming for
[14:25:07] <abcminiuser> SHITS GOING DOWN
[14:25:24] <specing> and if you store each bit in a seperate char, then FUCK YOU
[14:27:19] <hdas> specing, I assume you knew these things by birth
[14:31:42] <specing> Well I certainly knew it after some time "normally I do linux programming."
[14:34:19] <Malinuss> hdas,
http://playground.arduino.cc/Code/BitMath
[14:34:27] <Malinuss> hdas, read that, then try again
[14:34:35] <specing> what do I see here?
[14:34:38] <specing> WHAT?
[14:34:42] <specing> A LINK TO ARDUINO?
[14:34:45] <Malinuss> haha
[14:34:48] <specing> AAAAAAAAAARRRRRRRRRRRRRRRRGHHHJHHHHHHHHHHHHHHHHH
[14:34:57] <Malinuss> well atmel never made a dev node about bit math ;D
[14:35:08] <Malinuss> as far not one I've seen so far
[14:35:14] <Malinuss> as far
[14:35:53] <Malinuss> damn, I even made bram1 quit haha and made specing go crazy ;D
[14:36:03] <hdas> thanks Malinuss
[14:36:10] <specing> http://www.hackersdelight.org/
[14:36:53] <Malinuss> specing, eh?
[14:37:09] <specing> bit tricks
[14:37:17] <specing> download the sample chapter
[14:37:20] <specing> :D
[14:38:09] <Malinuss> http://www.hackersdelight.org/basics2.pdf
[14:38:10] <Malinuss> I see
[14:38:42] <Malinuss> propably good to read if you already do know something about bit math ;D
[14:39:22] <Malinuss> specing, I can already see that the article expects you know what a bit is etc...
[14:51:22] <specing> Malinuss: If you don't know what a bit is then may I kindly refer you to #arduino?
[14:53:15] <Malinuss> specing, haha ;D. 1) I do know what a bit is 2) I'm already in arduino, stop acting all superior 3) I meant - that is a good tutorial if you already know something, I just linked to a tutorial that assumed a person doesn't have any basic knowledge, which, no offence, it sounded hdas didn't have
[14:56:31] <specing> "normally I do linux programming." -- no basic knowledge? Hmmm...
[14:57:28] <specing> unless said programming is limited to doing ":(){:|:
[14:57:40] <specing> ;}" in a terminal
[14:57:55] <specing> sorry, Im not used to this keyboard, my laptop died
[15:01:19] <hdas> specing: I studied electronics engineering in my undergraduate, I can give you a lecture, it also says you stfu.. what I mean in linux programming, I never needed bit wise programming where I am new,
[15:03:45] <specing> Ohoho and he studied EE!
[15:03:47] <hdas> specing, if you are bored, go and play tetris or something instead of being arrogant
[15:04:02] <specing> I just broke the minesweeper record
[15:14:34] <tzanger> doing "linux stuff" isn't quite the same as embedded programming... even if "linux stuff" is programming
[15:15:36] <tzanger> hdas: how long is your message?
[15:17:24] <hdas> tzanger: it is 18 bits, but I need to convert the message to Manchester coding where it would be 36 bits
[15:23:39] <tzanger> this sounds pretty straightforward to do
[15:23:49] <tzanger> start in the simple case. don't try to optimize
[15:24:04] <tzanger> you have an 18 bit message. that'll fit into a 32-bit int
[15:24:08] <tzanger> int message;
[15:24:45] <tzanger> for (i=0; i<18; i++) { if (message & 0x1) tx(1); else tx(0); }
[15:25:10] <tzanger> er I missed a step
[15:25:17] <tzanger> for (i=0; i<18; i++) { if (message & 0x1) tx(1); else tx(0); message >>= 1; }
[15:25:40] <specing> Now look at the disassembly
[15:25:41] <tzanger> if you need MSB instead of LSB test for (1 << 17) and <<= instead of >>=
[15:25:47] <specing> Im sure you'll like it
[15:26:20] <tzanger> I'll leave it up to you to implement it in a more statemachine-like manner so you aren't blocking everything else, or you want better control over your bit timing
[15:26:26] <tzanger> but that's the start. get it working, then get it better.
[15:26:37] <tzanger> specing: I know, this is an academic example to get him started
[15:26:55] <hdas> tzanger, thank you very much
[15:44:29] <creep> do you think an SSR board is cool if the led terminals are directly driven ? i think it is better to put a current limiter in series, so accidental overcurrent will not kill it
[15:45:00] <creep> so current limiter as part of SSR board, not the things that drive it
[15:45:18] <Malinuss> hdas, how god is your understanding of timers? You will need that for manchester coding
[15:46:09] <Malinuss> hdas, to give you something specific to work with: have a timer run at the frequency you want to transmit with. and have a interrupt on overflow. and a toggel in the mid of the timer on the pin
[15:48:11] <creep> Tom_itx ? Horologium ?
[15:48:44] <creep> r00t^home ?
[15:50:33] <hdas> thank you Malinuss
[15:54:48] <Malinuss> np
[16:28:17] <Santh> Oui :D
[16:28:22] <Santh> Oops
[16:37:38] <Steffanx> non
[17:01:37] <theBear> r00t^home, after many years connecting and repairing and using them for a living, in all kinds of conditions and uses, i guarantee a well terminated screw connection is almost guaranteed never to loosen, and any pressure based/non screw connections will screw up sooner or later and start arcing and melt themselves or set something on fire...
[17:03:22] <theBear> and stop being a pack of big girly bitches, if you don't know what bits and bytes are, and the basic details of most variable types and how they are stored, you don't know how to program anything and should learn the basics that EVERYONE needs
[17:03:53] <theBear> and not learn them from some stupid arduino aimed bullshit
[17:12:31] <krphop> well, any suggestions on books or reading material for that?
[17:16:06] <theBear> any basic/beginners programming guide will have it, probably in the first chapter, but being that i learned it before the web existed as we now know it, and from books that are out of print for well over 20 years now, i can't recommend anything
[17:16:52] <krphop> most of the stuff i've seen on the net is garbage, or very pointed towards 'arduino' stuff
[17:20:04] <theBear> the net has literally millions of programming help sites and books, and had the large majority of them long before arduinos OR avrs even existed
[17:20:43] <theBear> there's a good one for c, err, i think it's art of something, been a while.. it's a 'real' book, but easy to find in electronic formats
[17:21:00] <krphop> ok cool, thanks!
[18:01:47] <tjtr33> using arduino uno (atmega 328p with 16Mhz clock), i tried timer0 with CTC and prescale of 1,8,64,256,1024,
[18:02:09] <tjtr33> got odd size pulse widths of 16, 1030, 1030, 16500,16500uS resp. is chip borqd or is it my code?
http://pastebin.com/2N6NZXc7
[18:03:18] <tjtr33> i expected the pulses to be 1,8,64,256 or 1024 time the 1/16000000 seconds
[18:08:34] <creep> anybody know of a usable electronic schematic drawing software for linux ?
[18:08:49] <creep> xfig is retarded :(
[18:09:21] <tjtr33> eagle & geda
[18:10:20] <creep> hm geda? haven't heard of that, must try
[18:12:32] <tzanger> eagle is awesome. it's my tool of choice
[18:13:06] <tzanger> it's no altium but it's also not windows-only nor does it have altium's price tag
[18:13:48] <tzanger> I bought eagle and also elektra although I haven't used the latter in years, I hand route almost everything
[18:18:14] <creep> http://sourceforge.net/projects/geda/?source=directory i think i'll try this
[18:24:57] <specing> creep: I've built some boards with gEDA
[18:25:04] <specing> its good
[18:25:10] <creep> how do i start it lol ?
[18:25:27] <specing> though the parts library is somewhat lacking and making new parts requires running scripts :)
[18:25:32] <creep> tried geda and gEDA, and now searching for executables in the src dir
[18:25:37] <specing> creep: $ gschem
[18:25:44] <specing> and $ pcb
[18:26:20] <creep> then something is not good, i don't have these
[18:26:44] <creep> /usr/bin/ld: cannot find -lltdl
[18:26:47] <creep> what is this?
[18:27:07] <specing> Did you install through package manager?
[18:27:31] <creep> from source
[18:27:42] <tjtr33> and built it?
[18:27:49] <specing> well then use the pkg manager to install dependencies
[18:27:54] <creep> ./configure make, make check, make install
[18:28:14] <tjtr33> on ubu, sudo make install
[18:28:41] <specing> tjtr33: Maybe he doesen't want it system-wide?
[18:29:43] <tjtr33> on ubuntu?
[18:29:58] <creep> wat is /usr/bin/ld: cannot find -lltdl ?
[18:34:02] <specing> It cannot find libltdl.*
[18:34:12] <specing> install dependencies.
[18:34:51] <tjtr33> creep, dont know what that err is, but after an install that has libraries, they are cfg'd by ldconfig, and it got handed garbage (afaict)
[18:35:07] <creep> http://www.linuxquestions.org/questions/linux-software-2/couldnt-find-ltdl-library-but-all-libtools-etc-installed-499940/
[18:35:42] <specing> *pokerface*
[18:36:02] <creep> /usr/lib64/libltdl.so.7
[18:37:52] <specing> Can you people ever test stable stuff from the repositories before going all bleeding-edge, cutting yourself in the progress and then whining about said software being shit because it cant load/crashes/makes you go insane/cause WW3
[18:47:29] <tjtr33> creep: you might well try a .deb/rpm whatever to avoid the build deps
[18:48:56] <tjtr33> i got an arduino q on timer0. this code gives me larger that expected pulses in CTC mode
http://pastebin.com/2N6NZXc7
[18:49:27] <tjtr33> it kinda avr ;) and the #avr channel is full of non-arduino/avr talk
[19:01:11] <Tom_itx> as it should be
[19:01:36] <vsync_> arduino is not avr
[19:01:45] <vsync_> or well, is, but isn't. it's cancer nonetheless
[19:12:13] <creep> o yea, i made it "work" , here's what i did: installing geda on Scientific Linux 6 (RHEL 6)
http://pastebin.com/aLb8G1dt
[19:18:07] <creep> how to make geda black and white?
[19:22:58] <Malinuss> anyone ever played around with any of the arm chips?
[19:23:32] <Malinuss> now don't tell me arm is shit like you just did about arduino vsync_ xD
[19:25:24] <OndraSter> ARM is not AVR :)
[19:25:28] <OndraSter> many people did
[19:25:30] <OndraSter> ask specing
[19:25:35] <OndraSter> he is making love to them secretly I think
[19:25:44] <OndraSter> I have got here two cortex m4f boards
[19:25:50] <OndraSter> but I have not opened the packages :D
[19:26:05] <OndraSter> (I am lying, I opened one to check out the RGB LED)
[19:26:11] <Malinuss> lol
[19:26:35] <Malinuss> did you actually buy two boards, only to open one, and check its RGB LED? I hope you sampled it OndraSter ;D
[19:26:48] <OndraSter> haha
[19:26:51] <OndraSter> they were $5 from TI
[19:26:54] <OndraSter> TI Stellaris
[19:26:56] <OndraSter> they are now $13
[19:28:16] <Malinuss> OndraSter, is it the devboard with the small screen on?
[19:28:21] <OndraSter> no
[19:28:26] <OndraSter> just the micro, button, RGB LED
[19:28:33] <OndraSter> + programmer/debugger on the top part of the board
[19:28:51] <Malinuss> OndraSter, you got it for 5$? wow...
[19:28:58] <OndraSter> they raised the pricing
[19:29:04] <OndraSter> after it went from pre-orders to regular orders
[19:29:21] <Malinuss> OndraSter, so why are avr's better ;D?
[19:29:34] <OndraSter> 1000 times simpler really
[19:30:19] <OndraSter> cheaper
[19:30:21] <Malinuss> OndraSter, I see, but does that matter if you are using a higher lever language (not asm but C)? Or is setting the peripherals up just really retarded and hard?
[19:30:26] <OndraSter> yes
[19:30:30] <Malinuss> yes?
[19:30:43] <OndraSter> the chips themselves are also much more expensive compared to (x)mega
[19:30:58] <OndraSter> and compared to xmega they offer the same (or less) peripherals, only higher core clock + FPU
[19:30:59] <Malinuss> well they are much much much faster?
[19:31:05] <Malinuss> I see
[19:31:12] <OndraSter> core? yes, these ones are ... 80? MHz I think
[19:31:29] <OndraSter> xmega runs at 32MHz, I overclocked the core with IO peripheral to 80MHz though :P
[19:32:06] <Malinuss> OndraSter, still no FPU :D. Guess I would need to take a look at some code/datasheet to be abel to tell anything
[19:32:41] <OndraSter> if you have to ask what it is good for - you don't need it ;)
[19:32:45] <creep> ok nice, i got that too
[19:32:47] <OndraSter> that's the rule of the world
[19:33:08] <OndraSter> (same for "if you have to ask how much it costs you can't afford it"
[19:33:33] <OndraSter> ((for stuff you expect to cost a lot))
[19:33:37] <Malinuss> OndraSter, well I know what a fpu is good for :)... but a STM32 is like 1$>
[19:34:12] <OndraSter> do you know how I solved FPU issue? Instead of doing x/y*100 I did (x*100)/y
[19:34:36] <OndraSter> (I needed to draw some line on some display at school where we have got PICs)
[19:36:15] <Malinuss> yeah that's a pretty common practice
[19:36:39] <Malinuss> still you can't get the same precision when making math on data etc.
[20:27:39] <OndraSter> it is PI time
[20:27:40] <OndraSter> gn
[20:35:17] <Malinuss> night
[20:37:03] <Tom_itx> evening
[21:08:39] <creep> awesome, now i can start drawing schematics on computer at last
[21:09:08] <Malinuss> creep, welcome to the 80s... oh wait it's 2013 ;D
[21:09:15] <creep> :)
[21:26:05] <tjtr33> creep what was the fix?
[21:26:42] <tjtr33> btw its pretty easy to make new parts
[21:28:02] <creep> did this in 15 mins with geda :)
http://tinypic.com/view.php?pic=2iu2sgl&s=6
[21:28:19] <creep> tjtr33<< yes, i already did that too
[21:29:16] <creep> tjtr33<< here's what i did: installing geda on Scientific Linux 6 (RHEL 6)
http://pastebin.com/aLb8G1dt
[21:32:13] <tjtr33> cool, i read about SL, fermilabs nearby. i use an RTAI vrsn of Ubuntu for Linuxcnc.
[21:32:43] <tjtr33> and i think thats a schematic of a florida voting machine :)
[21:33:09] <creep> ^^
[23:09:05] <prpplague> anyone using DipTrace ?
[23:37:28] <Casper> has anyone succeded to make an avr without build in usb support to actually work as a host and allow multiple devices thru an hub?
[23:37:38] <Casper> i.e. to make a multi-device charger
[23:56:30] <Funklord> Casper: you can charge devices by putting 5V on the + - pins
[23:56:37] <Casper> nope
[23:56:43] <Casper> it's those sony ps3 controllers
[23:56:55] <Casper> it mandate that the host authorise the 500mA
[23:57:09] <Casper> so also do not work throught a passive host
[23:57:16] <Casper> err hub
[23:58:18] <Funklord> well, the usb standard is retarded, not everyone follows it slavishly, those things you can charga
