#avr | Logs for 2012-04-25

Back
[00:00:06] <clever> ive got a 1wire bus snaking its way all over the house
[00:00:12] <Essobi_> where the heck do you live? heh..
[00:00:16] <clever> currently, with 5 sensors
[00:00:17] <Essobi_> clever: yea? Powered?
[00:00:19] <clever> NB canada
[00:00:29] <Essobi_> ah
[00:00:30] <clever> 6 wire technicaly, only 3 in use
[00:00:39] <clever> i couldnt get it to run in parasitic mode
[00:00:48] <clever> the sensors kept browning out or returning garbage
[00:01:01] <Essobi_> Hmm... I had mine in parasitic mode... maybe the length of the run?
[00:01:03] <clever> ah, i remember now, it would brownout durring measurement
[00:01:07] <clever> and return the power on default value
[00:01:08] <Essobi_> I heard that hurts it..
[00:01:22] <clever> i think its because i was missing the mosfet strong pullup
[00:01:29] <clever> which bypasses the pullup resistor
[00:01:30] <Essobi_> oh yea
[00:01:33] <Essobi_> you have to have that
[00:02:04] <clever> currently, the avr controls a single SSR, to turn one furnace zone on/off
[00:02:11] <clever> but the min/max temp are hard-coded in flash
[00:02:16] <clever> so you need a programmer to change the temps
[00:02:50] <clever> dad doesnt like it
[00:02:53] <Essobi_> oh right
[00:02:54] <Essobi_> hehe
[00:03:03] <clever> and one day, he actualy stuck an ice cube on the sensor to raise the heat
[00:03:10] <Essobi_> haha
[00:03:16] <Essobi_> that's funny
[00:03:21] <Essobi_> Doing any zoning?
[00:03:36] <Essobi_> I was considering doing some zoning in my house.
[00:03:36] <clever> the house already has 3 zones, and originaly had 3 seperate thermostats
[00:03:43] <Essobi_> Oh, interesting
[00:03:44] <clever> but the basement only has one zone
[00:03:54] <clever> and if you close the bedroom door, the bedroom is isolated from the thermostat
[00:04:04] <clever> hallway is cold
[00:04:07] <clever> heat comes up
[00:04:11] <clever> hallway stays cold
[00:04:16] <clever> bedroom cooks me alive
[00:04:56] <clever> my current 'fix' is rather simple
[00:05:04] <clever> punch a hole in the bedroom wall, move the thermostat
[00:05:30] <clever> but my end goal is to have several sensors in a zone, and turn the heat on if any get cold
[00:05:42] <clever> with a seperate time/temp setting for each sensor
[00:05:53] <clever> for example, ignore the hallway temp at night
[00:06:03] <clever> but ignore the bedroom durring the day
[00:07:04] <Essobi_> interesting
[00:07:26] <clever> its effectively hooking several programable stats to the same zone, in parallel
[00:07:31] <Essobi_> right
[00:07:39] <clever> but rather then fork out $30 per room, its $2 per room plus an avr
[00:07:50] <Essobi_> also people control those with electronic dampers on the vents too
[00:08:03] <Essobi_> so you can have mini-zones and seperate temps
[00:08:03] <clever> hot water heating
[00:08:08] <clever> 3 servo valves in the furnace room
[00:08:17] <Essobi_> :D
[00:08:29] <clever> 3 loops arround the house, which meet up in the furnace room
[00:08:37] <clever> cant rezone it without tearing half the walls apart
[00:08:44] <Essobi_> i've looked at some solar heaters too.
[00:08:51] <clever> oil burner currently
[00:08:53] <Essobi_> similar pid controllers run those as well
[00:08:56] <Essobi_> eww
[00:08:56] <clever> also does the hot water
[00:09:15] <Essobi_> well these suplicate hot water by raising the ingress temps of water
[00:10:21] <clever> the biggest problem is basicaly 'WAF', or in my case 'DAF'
[00:11:43] <Essobi_> DAF?
[00:11:50] <clever> dad acceptance factor
[00:11:53] <Essobi_> haha
[00:12:16] <clever> currently, he has no way to control it, and any method i make is either ugly or too complex :P
[00:12:54] <Essobi_> Umm.. LCD, and a few digital IOs for the 1-wires?
[00:13:11] <clever> cant get the lcd to work at all, and the project box wont look 'good'
[00:13:24] <Essobi_> does he have a phone?
[00:13:39] <clever> windows 7
[00:13:39] <clever> you need vista to even try to dev on it
[00:13:44] <clever> enless i make it html based
[00:13:51] <Essobi_> Or DTMF based.
[00:13:54] <Essobi_> :D
[00:14:06] <clever> and that would fall under 'too complex'
[00:14:34] <Essobi_> Google voice -> FS/unix boxen. You call the house and tell it what temperature. :D
[00:14:39] <Essobi_> Speech to text, even.
[00:14:52] <clever> still, he doesnt want to dig out his phone every time he gets cold
[00:15:03] <Essobi_> But he could do it from the store. :D
[00:15:05] <Essobi_> meh.
[00:15:08] <clever> he just wants the controls on the wall where they 'belong'
[00:15:11] <Essobi_> hehehe
[00:15:26] <Essobi_> sounds like my old man
[00:16:05] <clever> another idea i had, rfid based
[00:16:10] <Essobi_> nice
[00:16:12] <Essobi_> heh
[00:16:13] <clever> you have a wireless thermostat, with lcd+buttons
[00:16:18] <clever> and an rfid tag in the wall
[00:16:30] <clever> slap the thermostat on the wall, and it controls the current room
[00:16:45] <clever> move it to another room, slap it on the wall, and it magicaly controls a diff room
[00:16:55] <Essobi_> interesting concept
[00:17:04] <Essobi_> zigbee too... or bluetooth.. Hmm.
[00:17:17] <clever> already got an xbee mounted internaly on my netbook
[00:17:32] <clever> that was another fun project
[00:17:59] <Essobi_> Nice.
[00:18:03] <clever> the netbook has a 4 pin JST connector labled webcam
[00:18:05] <clever> its just plain usb
[00:18:33] <clever> grab a 4 pin JST cable from sparkfun
[00:18:39] <clever> hotair the usb conn off an ftdi
[00:18:46] <clever> and i have internaly connected ftdi
[00:19:05] <clever> set the ftdi to 3.3v, and hook up an xbee, done
[00:19:53] <Essobi_> Hehe.
[00:19:58] <Essobi_> Pretty slick.
[00:20:03] <clever> biggest surprise was the bios
[00:20:10] <clever> if the bios cant detect anything on the usb data lines
[00:20:13] <clever> it cuts usb power!
[00:20:22] <Essobi_> woops
[00:20:32] <clever> but once i put the ftdi on the data lines, it left power on
[00:20:41] <clever> and now i can even cut power manualy via /sys/
[00:20:42] <Essobi_> bet that made debugging a pain in the ass
[00:20:52] <clever> yeah, the volt meter saw +5v for about 2 seconds
[00:20:54] <clever> then it vanished
[00:23:12] <clever> d+ d- where labled well in a photo online, and vcc/gnd where easy to figure out
[00:26:23] <clever> crap, laptop locked up again
[00:26:29] <clever> need to keep a closer eye on free mem...
[00:26:42] <clever> ive got 0 swap
[00:26:46] <clever> and no hdd :P
[00:27:39] <clever> 2400mb free
[00:28:03] <clever> trying the fatal step...
[00:28:14] <clever> ah, kernel oops!
[00:30:43] <clever> Essobi_: oh.... oops....
[00:31:03] <clever> the instant i ran dmesg|less to read the oops
[00:31:06] <clever> it had a panic attack :P
[00:31:25] <CapnKernel> kernel anxiety. It happens.
[00:31:42] <clever> CapnKernel: i think i'm mixing kernel versions with aufs2
[00:31:57] <CapnKernel> Hmm
[00:32:25] <clever> another try!
[00:32:33] <clever> nope, oops'd
[00:33:34] <clever> nope
[00:33:39] <clever> it needs some chroot
[00:33:55] <clever> i'll get it in the morning, already 2am
[00:33:56] <clever> Essobi_: later
[04:48:30] <johntramp> CapnKernel: cheers for that
[05:12:14] <CapnKernel> johntramp: Haha, I forgot what for.
[05:13:57] <johntramp> the quote
[05:13:58] <johntramp> ravage
[05:14:09] <CapnKernel> Oh I see :-)
[05:14:16] <CapnKernel> For the PCBs.
[05:14:55] <johntramp> yes
[08:59:53] <Xata> got a question - why when i try to compile my program with optimizations other than -O3 (including -O0) - ADC stops to work?
[09:00:09] <Xata> specing: my synth already makes some sounds. yay!
[09:00:36] <specing> because gcc generates dumb code, no matter what -O you specify, next question?
[09:01:13] <specing> e.g. it could access ADCH before ADCL causing the ADC module to lock down
[09:01:53] <Xata> specing: how i can replace gcc? with what? or some trick to fix this?
[09:02:22] <Xata> or only hardcore asm wich i really don't know
[09:02:25] <Xata> ?
[09:12:43] <izua> how exactly does the adc "stop"
[09:12:58] <izua> the whole chip freezes, or further adc samples have wrong values?
[09:14:54] <Xata> izua: i have adsc controlling ocr0, and with optimizations other than o3 he output from chip is like when values are close to zero.
[09:15:17] <Xata> value in adch
[09:15:51] <izua> do you use _delay functions?
[09:15:52] <Xata> abd adch controlling ocr0. adsc is a new conversion bit.
[09:16:10] <Xata> izua: everything on interrupts from timer0
[09:16:20] <izua> multiple channels?
[09:17:04] <Xata> izua: you mean more than one adc input? or more than one DDRx?
[09:17:14] <izua> adc channels, ofc
[09:17:27] <Xata> izua: no, only one
[09:18:17] <izua> i recall an odd bug on 90pwm3b, where if you'd write the counter top under the current duty cycle (count), it would have to overflow to enter a cycle again
[09:18:44] <izua> which is pretty normal behaviour, after all, but it isn't actually documented
[09:19:09] <izua> perhaps something similar happens? is your system messing with both duty cycle and frequency?
[09:19:43] <Xata> izua: strange, switched to release mode in avrstudio, that has -Os - it works ok, but when i add DDRA=0xff to my DDRD=0xff bug appears again
[09:19:45] <izua> make sure to reset the counter if you increase the frequency (lower the TOP value)
[09:20:09] <izua> DDRA has the analog section?\
[09:20:13] <izua> (analog pins)
[09:20:40] <Xata> izua: yes... lol.
[09:20:55] <izua> why would you do that, then?
[09:22:22] <Xata> izua: forgot that i need ddrb, not ddra. but this strange behavior with optimizations still has its place. but i can live with this. thanks for help, and for advice about counter resetting.
[09:25:05] <Xata> omg, that was dumb
[10:05:36] <taos> What are the instructions needed to set a larger stack size? currently I do this http://paste.pocoo.org/show/587098/
[10:05:40] <taos> but I need a much bigger stack
[10:19:08] <jacekowski> taos: you can't
[10:19:15] <jacekowski> taos: stack is limited by memory size
[10:20:15] <jacekowski> taos: what chip are you using
[10:20:26] <jacekowski> taos: because your RAMEND is set to 607bytes
[10:20:33] * specing inflates taos's head
[10:20:39] <specing> There, that should do it
[10:21:12] <taos> specing: what is your problem with me D:
[10:21:44] <jacekowski> taos: what chip are you using?
[10:21:56] <taos> Don't worry its not the stack im a **%" idiot
[10:23:05] <specing> I concur
[10:23:24] <taos> specing: if you have nothing useful to say then don't say anything
[10:26:08] <specing> hehe ;)
[10:30:32] <asteve> taos: you're missing a c
[10:30:42] <asteve> (tacos) :)
[10:30:51] <taos> Ah sorry I couldn't work out what you meant
[10:32:33] <specing> He is missing C
[10:32:58] <OndraSter> 680uF caps -> 220uF caps, soldered from top -> soldered from the bottom (they were too big for the top now lol, I couldn't get the same one as original); main big caps -- 1000uF -> 1000uF, but different bottom pin size => again soldered from bottom... and I am not even sure they are low esr :D
[10:33:09] <OndraSter> recapped motherboard
[10:33:13] <OndraSter> but hey, the board is worth like $10 - $20 tops
[10:33:19] <OndraSter> so caps worth of $10 are not the best idea :P
[10:33:24] <OndraSter> and it runs the lowest end CPU ever
[10:33:48] <OndraSter> since it was supposed to be able to run quadcore eating 90 - 130W, I don't think that it will cause any trouble running 25W tops CPU (about 1W in idle)
[10:33:52] <OndraSter> 1 - 5W in idle
[10:34:30] <OndraSter> actually the original caps didn't look like low esr
[10:34:39] <OndraSter> is there any way how to find it out?
[10:35:41] <OndraSter> they were some cheap Tk (??) manufacturer
[10:35:46] <OndraSter> ATWB 810a
[10:35:56] <OndraSter> can't find anything but one porn video, some hongkong inventory
[10:36:00] <OndraSter> and some russian page lol
[10:45:36] <clever> rue_house: was that your solder pot i just saw?
[11:00:11] <specing> OndraSter: Well atleast you found the video...
[11:08:04] <OndraSter> yep lol
[11:16:43] <Xata> Guys, advice me lpf filter scheme with q-factor? on op-amps
[11:18:01] <taos> http://paste.pocoo.org/show/587141/ this code
[11:18:07] <taos> produces this bug
[11:18:09] <taos> http://www.youtube.com/watch?v=lMT_QTpzJDQ
[11:18:52] <taos> Im entirely lost
[11:19:35] <Xata> asm. nah. i really need to learn it.
[11:20:04] <OndraSter> taos, why not just include your device's file?
[11:20:11] <OndraSter> instead all those PORT/PIN on the beginning
[11:20:15] <OndraSter> that's safer
[11:20:36] <OndraSter> also, what is it supposed to be doing?
[11:22:01] <taos> OndraSter: its supposed to clear the clock, then genreate a new number, display it and return to the main label.
[11:22:27] <OndraSter> I am not sure that helped :D
[11:22:59] <taos> :/
[11:23:00] <OndraSter> let's see
[11:23:16] <OndraSter> also, what is the problem? Can't figure it out from the video
[11:23:53] <taos> the problem is, when the "attempt" times out rather than just beeping the buzzer once and displaying the new number, it beeps forever cycling through numbers on the left side of the display
[11:24:03] <OndraSter> ldi r16, 0x05
[11:24:03] <OndraSter> out TCCR1B, r16
[11:24:07] <OndraSter> can't you do something like
[11:24:16] <OndraSter> LDI R16, (1<<CS00)|(1<<CS01) etc?
[11:24:19] <OndraSter> so it is more readable
[11:24:37] <taos> OndraSter: Can I leave it as it is, I only just got the timer working :(
[11:25:12] <Xata> taos: why assembly?
[11:25:26] <taos> Xata: course requirement. I get no choice in the fact.
[11:25:39] <taos> Xata: and believe me, when this unit is done im never coming back.
[11:25:41] <OndraSter> hmm
[11:26:26] <Xata> taos: so you don't like avr... :3
[11:26:38] <OndraSter> :D
[11:26:49] <taos> Its okay, but this unit is making me want to carve my eyes out with a rusty spoon to escape it
[11:28:04] <taos> Ill stay far away from you nutters up in higher level programming langauages
[11:28:13] <Xata> taos: well, i wish i could help you, but i don't know asm, only C.
[11:28:42] <taos> Xata: mhm
[11:30:05] <OndraSter> Xata, grab an emulator and step through it
[11:30:18] <OndraSter> easiest solution, if you can't find the error quickly
[11:30:51] <Xata> OndraSter: you look like smart one. have you ever tried making low-pass filter with resonance on op-amp? i just can't find any schematics - all 'bout bypass filters.
[11:31:03] <OndraSter> nope
[11:31:05] <Xata> OndraSter: i think that your message was for taos
[11:31:11] <OndraSter> err, yes
[11:31:14] <OndraSter> it was for taos
[11:31:24] <OndraSter> no, I have not tried doing LPF
[11:31:27] <OndraSter> with opamp
[11:31:35] <OndraSter> only with LC
[11:31:38] <OndraSter> and I used online calculator for it
[11:33:02] <taos> OndraSter: I stepped through it, the emulator shows no error
[11:33:20] <OndraSter> well then, you need to simulate exactly what you do :P
[11:33:32] <OndraSter> I see you have there some input
[11:33:52] <Xata> OndraSter: nah, i already have lpf on op-amp, but i really need resonance as this has to be synthesizer. and what is synthesizer when you can not get that hard resonant sound?
[11:33:53] <taos> I simulate the input on the simulator
[11:34:22] <OndraSter> if it works in the emulator, it must work in the device
[11:34:32] <OndraSter> if it is the same input/output simulation
[11:34:35] <Xata> key word _must_
[11:34:38] <taos> OndraSter: well its not
[11:34:40] <OndraSter> Xata, oh
[11:34:47] <OndraSter> taos, then you are forgetting something :)
[11:34:50] <taos> OndraSter: im not
[11:34:56] <taos> the emulator shows it doing it just right
[11:35:00] <taos> the hardware screws itself
[11:35:56] <mrfrenzy> then something is incorrectly wired on the hardware, or emulator is not properly configured
[11:36:52] <taos> The emulator is AVR Studio 4 and its set to the EXACT hardware model, same clock speed, everything
[11:37:11] <Xata> taos: well, i thought that i am right 100% too, but then i found out that i am using same port for adc and data input. that was really dumb, maybe you recheck it, or give a blood for your blood god, or whatever that crazy people, that use asm, usually do in such situations.
[11:37:48] <taos> Xata: this is the last task on the entire unit T_T I came so close
[11:37:49] <OndraSter> Xata, lol, we use real debug option (JTAG)... I know, I am repeating myself, but yes, JTAG IS quite useful thing to have
[11:39:42] <Xata> taos: you can *cough*ste*cough*al proteus and simulate chip with whole circuit.
[11:41:11] <OndraSter> btw, question for broad public: How much "random" are the poweron register values?
[11:41:16] <Xata> OndraSter: but one needs to have that stk600 thingy for jtag, amirite? i mean no way to connect it to com/lpt/usb?
[11:41:16] <OndraSter> R0-R31
[11:41:23] <OndraSter> Xata, what chip?
[11:41:39] <OndraSter> well, for the old chips there is JTAG ICE
[11:41:40] <Xata> OndraSter: a32
[11:41:43] <Xata> in dip40
[11:41:44] <OndraSter> which is like $10
[11:41:51] <OndraSter> it has USB
[11:42:00] <OndraSter> and it sports small mega168 I think, it can connect to the older chips
[11:42:24] <OndraSter> then there is AVR Dragon, which sports mega2560+128a, which can connect to pretty much anything, throguh JTAG, ISP, dW, PDI, ..
[11:43:52] <Xata> OndraSter: 100$? no way, at least as soon, as i start make some monies with atmegas (obv., if i ever will)
[11:44:16] <OndraSter> $50 for Dragon
[11:44:35] <OndraSter> but ye, if you are not going to do anything with megas laters, it is not worth it :D
[11:47:27] <Xata> OndraSter: i think i'm in love with atmegas, but i am not ready at a time to buy something except atmegas themself. i already have precious 2$ wasted for burned a16! i could buy so much useless capacitors and resistances for that two bucks! :D
[11:48:32] <OndraSter> :D
[11:49:33] <OndraSter> mega88: 1.25€, mega168: 1.44€, mega328: 1.63€
[11:49:49] <OndraSter> 88 is too small, 168 is... "I can always make use of the extra storage" :P
[11:49:59] <OndraSter> no mega648
[11:51:47] <karlp> OndraSter: what do you mean register power on values? the power on reset values for all registers is in the datasheet
[11:51:57] <OndraSter> really?
[11:52:08] <OndraSter> R0-R31 seemed always different after each power cycle
[11:52:11] <OndraSter> as far as I remember
[11:52:16] <karlp> (but make sure to read the notes on brownout and watchdogs)
[11:52:17] <OndraSter> NOT the memory mapped IO SFRs
[11:53:12] <karlp> I'm going to go out on a limb and say there's a line in the datasheet somewhere saying, "don't assume the registers have anything"
[11:53:28] <Martyn> that is what the docs say, yes
[11:53:31] <OndraSter> well yes
[11:53:35] <Martyn> NEVER assume power on register values
[11:53:39] <OndraSter> yes
[11:53:43] <Martyn> always initialize on reset
[11:53:46] <OndraSter> yes
[11:53:48] <Martyn> (in fact, it's in the sample code)
[11:53:56] <OndraSter> but if one would want to use them as a random generator seed values...
[11:54:03] <OndraSter> that's why I am asking how much random are they
[11:54:16] <Martyn> then you'd be picking the WORST seeds, short of just picking "7" all the time
[11:54:30] <Martyn> They aren't sufficiently random to pass FIPS
[11:54:47] <OndraSter> still more random than fixed value
[11:55:06] <Martyn> Reverse bias a diode, and take advantage of tunnelling to get a good RND generator
[11:55:33] <Martyn> no, because I absolutely know I can influence the value of the registers on boot
[11:55:38] <karlp> there's no such thing as "more random than a fixed value"
[11:55:51] <karlp> as martyn says, they're too predictable, and therefore _not_ random at all.
[11:55:53] <Martyn> In fact, I can proveably get them to go to 0 with a simple hardware hack
[11:56:20] <OndraSter> hmm
[11:56:22] <Martyn> So, look .. there are lots of great ways to get a seed, cheaply
[11:56:37] <OndraSter> it was just an idea about afterthought, nothing about security related
[11:56:46] <OndraSter> just for random poweron image generation.
[11:57:15] <Martyn> Look up : or Zener breakdown noise from a reverse-biased Zener diode
[11:57:33] <Martyn> Simple, easy, and as random as the universe can be
[11:57:54] <Xata> how do i overcome adc value instability? i meanif i have resistance on potentiometer that is between, for example, 34 and 45 - it jumps between that values. making adc slower is not an option - i really need that speed. maybe some lpf function? (not real lpf, but a code)
[11:58:12] <Xata> *34 and 35
[11:58:17] <OndraSter> <OndraSter> it was just an idea about afterthought, nothing about security related
[11:58:17] <OndraSter> <OndraSter> just for random poweron image generation.
[11:58:53] <OndraSter> all zeros = still fine with me
[11:58:57] <OndraSter> all ones = still fine with me
[11:58:59] <karlp> xata *=2 & >> 1
[11:59:32] <karlp> or, why does you have problems with it being 34 vs 35?
[12:00:31] <Xata> karlp: because 34 and 35 is a seperate notes, and now imagine sound jumping beween 2 notes at adc conversion speed. sounds crappy.
[12:00:52] <OndraSter> if the value is 34.49999 or 34.500001
[12:00:57] <Xata> *are the separate
[12:01:17] <OndraSter> the value is so close to both that it will be jumping around
[12:01:40] <OndraSter> => either make it digital pot or just grab 4 samples and average them.
[12:01:45] <OndraSter> not much you can do really
[12:01:52] <OndraSter> maybe decouple it
[12:01:53] <OndraSter> a bit
[12:01:59] <OndraSter> (or w/e you call it with the caps)
[12:02:12] <Xata> OndraSter: but values in ADCH are int, aren't they?
[12:02:17] <OndraSter> yes
[12:02:27] <Xata> oh, understood.
[12:02:47] <Xata> so lpf. god give me enough cpu cycles for that.
[12:04:54] <karlp> if you're using the raw adc value from a pot as being distinct notes, then you deserve what you get :)
[12:10:09] <Xata> karlp: this will be converted midi info or buttons in release, but this is ok for testing purposes. on he other side i want everything to be ideal right now, so everything will be ok when i do next stuff
[13:10:25] <sabesto> xata concider rotary encoders
[13:12:59] <asteve> yes, call all of the cider from the rotary encoders!
[13:13:08] <asteve> NEVER LET THEM RELAX WHILE DRINKING CIDER!!!!!!1
[13:13:21] <asteve> and by call I meant con....fuck
[13:30:39] <abcminiuser> tobbor,
[13:30:41] <abcminiuser> Tom_itx,
[13:30:51] <abcminiuser> What are the LED colors of the USBTINY MKII again?
[13:33:22] <OndraSter> I have just figured that atmega328 vs atxmega32d4 is 0.05€ difference
[13:34:01] <specing> d
[13:34:23] <OndraSter> so?
[13:34:45] <Kevin`> xmega is pretty nice
[13:34:48] <OndraSter> yeah
[13:34:50] <OndraSter> and for this price..
[13:35:00] <OndraSter> more RAM is always good :P
[13:35:35] <Kevin`> the d series doesn't have all the fanciest peripherals of the line, but it's still nice, and you get the nice register layout too
[13:35:42] <OndraSter> yeah
[13:35:53] <OndraSter> I don't care about the fanciest peripherals, it is just "futureproof" :P
[13:36:26] <OndraSter> and for my design the mega328 would be fine... but if there is really 0.05€ difference, I will go with bigger chip
[13:37:28] <specing> OndraSter: mouser is indeed cheap, but they have a 40 Eur shipping fee to here...
[13:37:51] <OndraSter> I can get $5 shipping throughout czech republic or go pick it up myself not far from Prague for free :P
[13:38:07] <specing> unless you order 150$+
[13:38:21] <OndraSter> you need some reseller or w/e they call themselves :P
[13:38:39] <specing> Nope, czech republic is 40 Eur, too
[13:39:22] <OndraSter> mausel.eu :P
[13:41:56] <specing> so, farnell is cheaper :)
[13:42:04] <OndraSter> for you
[13:42:06] <OndraSter> not for me :)
[13:43:29] <specing> why do they all have only 1-3 priority mails?
[13:43:34] <specing> 1-3 day*
[13:43:45] <OndraSter> you would be fine with 1 week I presume :P
[13:43:48] <OndraSter> for half the price
[13:43:51] <specing> I'd be happy to wait a week if it would cost only 5$
[13:44:05] <specing> but 40 Eur? No way
[13:45:43] <OndraSter> I have different issue
[13:45:52] <OndraSter> the part that I am using is not available on mouser :(
[13:45:56] <OndraSter> only on digikey and farnell
[13:46:09] <OndraSter> and well, either it is quite expensive shipping (digikey) or expensive part (farnell)
[13:46:18] <specing> heh
[13:50:18] <RikusW> seems shipping is always a problem :(
[13:50:36] <RikusW> I have a local Farnell reseller here, www.mantech.co.za
[13:50:54] <RikusW> but the sell stuff from many other companies too
[13:51:11] <RikusW> trouble is they also only ship on overnight.... :(
[13:51:29] * RikusW tried asking for slower cheaper options to no avail
[13:52:27] <RikusW> There is one part I have trouble locating, KA1H0165R.... used for 5Vsb in pc smps....
[13:52:52] <RikusW> seems I took the last one in mantech stock and now there is no more
[13:55:48] <Xata> i just had a thought. this is not so common, thus behold: how much will cost atmega32a from your local reseller?
[13:56:14] <Xata> in dip
[13:57:25] <RikusW> mantech only have a few old avr's and those are expensive
[13:57:34] <RikusW> RS might have it though
[13:58:09] <RikusW> I got a atmega324A for about $4
[13:58:39] <RikusW> in dip
[14:00:24] <Xata> RikusW: with delivery?
[14:00:56] <RikusW> RS deliver for free in SA
[14:01:30] <Xata> RikusW: hmm.. SA is some abr. for usa city?
[14:01:36] <RikusW> South Africa
[14:01:43] <Xata> oh
[14:01:58] <specing> Haha
[14:02:56] <Xata> i got mine for 3.4$, and they are ready to sell it for 3$ if i buy 100+. i mean - what is the point for reseller to sell a32 for 3 bucks?
[14:03:16] <Xata> mine = atmega32a
[14:03:37] <RikusW> mine was ZAR32
[14:04:07] <RikusW> a while ago is was like R10/$ now its more like R7.6/$
[14:04:56] <Xata> RikusW: what stands for zar32?
[14:06:17] <RikusW> South African Rand
[14:06:52] <Xata> also - there is other resellers and they sell it from 5$ to 10$ for a piece. i mean - what's wrong with them, why so damn cheap?
[14:07:22] <RikusW> don't ask whats wrong just get the cheap ones ;)
[14:07:36] <RikusW> maybe they bought it in bulk and got a HUGE discount
[14:08:13] <RikusW> Buying a few thousand drops the price dramatically
[14:09:29] <Xata> RikusW: yeah. we here have such thing as radio stuff fair, and i literally can buy a bunch of this atmegas, walk 500m and sell them there for double price.
[14:10:12] <Xata> tricky plan, lol
[14:14:14] <RikusW> and if nobody wants to buy ?
[14:18:09] <Xata> RikusW: why would nobody buy it from me if they buy from others?
[14:22:10] <Xata> also don't take this russian business too serious. i think if everything was that simple, there would be enough "tricksters" like me
[14:24:40] <RikusW> heh ;)
[14:25:06] <RikusW> They won't buy because I'll tip them off, to the cheaper source :-P
[14:27:01] <Kevin`> there's a cheaper source you can buy from in single quantities?
[14:27:30] <Kevin`> anyone can order a boatload of parts from china if they want. retail resellers are for people who only want a few
[14:29:43] <Xata> Kevin`: this is retarded, but i here have a situation when cost of single atmega32a chip rises twice in 500 meters. not joke, just singularity of price for electronic stuff in ukraine.
[14:32:04] <RikusW> Xata: want a free debugger for you m32a ?
[14:32:28] <RikusW> I have a jtagice mki clone
[14:32:48] <Xata> i mean - if you say that you can sell a chip $0.25 cheaper than thaaaat guy for some victim of your jewishness - you literally can go 500m buy it and sell to dat poor guy. cons - you have to go 1km
[14:34:05] <Xata> RikusW: free you say? hmmm... schematics?
[14:34:20] <RikusW> http://sites.google.com/site/megau2s/home -> jtag link for the firmware
[14:34:35] <RikusW> you can load it onto sa m328p or any other avr
[14:34:41] <RikusW> *say
[14:35:16] <RikusW> or 8/16/32u2 usb avr too
[14:35:34] <RikusW> Xata: I reverse engineered the original mki fw
[14:36:29] <RikusW> thats on there too if you want a peek
[14:38:07] <Xata> RikusW: but 25 euros.
[14:38:22] <RikusW> the mki fw is free
[14:38:27] <RikusW> my programmer is not
[14:38:38] <RikusW> download the zip files
[14:38:58] <RikusW> You'll need AVRStudio 4 to compile it
[14:39:09] <RikusW> its already compiled to m328p
[14:40:21] <Xata> RikusW: oh, i see. scheme is really simple, i can even try to predict the scheme by that photo you have on home. thanks, bookmarked for better times, when i'll need it
[14:41:04] <RikusW> when using the m328p you only need a rs232 level translator
[14:41:11] <RikusW> and the pinout for jtag is mentioned on that site
[14:42:12] <RikusW> Xata: ever used jtag before ?
[14:42:58] <Xata> RikusW: no. got no need for it yet.
[14:43:23] <RikusW> debugging with it is quite nice
[14:44:15] <RikusW> once you used it you'll be hooked ;)
[14:45:56] <RikusW> cheater -> http://www.hmsprojects.com/xboxbypc.html
[14:46:02] <Xata> RikusW: that's why i stay avay from gurls :D
[14:47:32] <RikusW> lol
[14:48:41] <Xata> and boys, if you'll ask me. not gay, just nerd
[14:49:24] <RikusW> ooh
[14:50:25] * RikusW don't see much of either around here....
[14:50:53] * RikusW lives on a farm
[14:51:01] <Xata> well, about all that xbox stuff... i think if you play the games this way, you already lost your life, and there is nothing to loose. so yaaay, pc-controlled bot for xbox, yaaay
[14:53:16] <RikusW> I'll try hacking the USB protocol, but then again I have no motivation to do such a thing
[14:53:54] <RikusW> Xata: hacking the mki fw took six months.... not full time though.
[14:54:01] <Kevin`> there's a linux driver for the xbox controllers
[14:54:18] <RikusW> nice
[14:54:30] <RikusW> is it plain HID ?
[14:54:45] <Kevin`> the newer one is, somewhat. the original xbox is a bit messed up though
[14:54:46] <RikusW> then it should be fairly straightforward
[14:55:14] <Xata> RikusW: i live in center of 1m+ city, but i don't like most local societies and this is mutual. well, i have a bunch of friends, but they are or coders, or alcoholics. so yes, there is girls, but i better fuck my 5-fingered sweaty lady then have something common with most of them.
[14:56:40] <Xata> RikusW: so promote your stuff. when you'll have some profit from your work -> huge motivation
[14:58:00] <RikusW> marketing is the hard part ;)
[14:58:11] <RikusW> coding is usually rather easy
[14:59:10] <Xata> after al this depressive solitude stuff - avr, yeah! fuck everything, i better overload my time with avr and forget that precious Real Life they have there
[14:59:56] <RikusW> heh
[15:00:04] <RikusW> or get on IRC ;)
[15:03:15] <Xata> RikusW: yup. pity that i have to crawl out from my basement in near time to get some work - dying from starvation is not a death for real chair heroes.
[15:04:48] <RikusW> API's -> http://ceklog.kindel.com/2012/04/18/dont-build-apis/
[15:05:56] <sabesto> OndraSter: check the mega128 and xmega128 on farnell
[15:06:11] <OndraSter> mega128a is a joke :P
[15:06:12] <OndraSter> even on mouser
[15:06:21] <OndraSter> haven't checked xmega128 there for my safety
[15:06:24] <sabesto> last time i checked the mega128 cost more then the xmega
[15:06:29] <OndraSter> yeah
[15:06:32] <OndraSter> it was like 20€ on mouser lol
[15:06:53] <RikusW> m128 is expensive m128a not so much
[15:07:02] <OndraSter> huh
[15:07:05] <OndraSter> and 128a is UPDATED 128
[15:07:18] <RikusW> but it is cheaper
[15:07:35] <RikusW> it is fresh stock
[15:07:38] <sabesto> got a reel of mega128l
[15:08:00] <RikusW> I have 3 m128 's here...
[15:08:19] <RikusW> got it for free, almost
[15:08:31] <RikusW> 2 is on pcbs too
[15:08:54] <sabesto> speaking of free
[15:09:53] <sabesto> need to figure out which of the xmegas that is the most versatile
[15:10:43] <RikusW> why not use AVR32 ?
[15:10:55] <RikusW> or the usb xmega's ?
[15:11:35] <Xata> for me - because they are not dip, and i have no welding station
[15:12:29] <RikusW> are there any DIP xmegas ?
[15:12:35] <Xata> no
[15:12:49] <Xata> that's why no xmegas for /me
[15:14:00] <Xata> even they have internal dac, i had to build on r/2r for my atmega32 synth
[15:14:05] <sabesto> Xata: buy those breakout boards
[15:14:47] <sabesto> why not just use an external dac
[15:14:57] <sabesto> SPI
[15:15:28] <rue_house> clever, yes
[15:15:53] <sabesto> but building it yourself is cooler :P
[15:16:01] <Xata> sabesto: r/2r is dac, and it is enough for my <30khz output. plus lpf for noise cancellation
[15:16:38] <sabesto> Xata: yes, but obviosly i meant a chip
[15:19:30] <Xata> sabesto: curiosity is that it is far not easy to buy one here. not too common demand - no dacs. only high-cost analog devices stuff, i really don't need
[15:31:48] <uv_> re
[15:53:38] <Xata> guys, here is the audio sample of sound of synth i am make. notice glitch on high freq. this is what i was talking about. halp? (i know that sound is shityy, but hey - this is pre-alpha)
[15:54:19] <Xata> http://rghost.ru/private/37771816/efcf9cc9a39988d73334a5c3e9b1b907
[15:54:41] <Xata> *i am making
[16:05:40] <sabesto> i really dont know, havent tried anything similar before
[16:06:11] <sabesto> closest thing ive done is playing prerecorded sound from an xmega
[16:20:07] <Fleck> http://img31.imageshack.us/img31/8186/ssr.png << my load is bulb, any ideas - why bulb blinks?
[16:24:52] <LoRez> why do you have a snubber on a light bulb?
[16:24:55] <Xata> Fleck: fading blinking or on/off blinking? also i have no idea what moc3061 is
[16:25:13] <LoRez> it's a optoisolator as show by the drawing.
[16:25:17] <Fleck> on/off blinking but also fading in the same time
[16:25:53] <LoRez> Fleck: tie pin 4 to pin 6 and see if your issue goes away
[16:26:27] <Fleck> LoRez when moc if full on - blinking stops, also when off...
[16:26:35] <Fleck> but inbetween - blinks
[16:26:55] <LoRez> you're blinking the LED and wondering why your load is blinking?
[16:27:07] <Fleck> i man not blinking the led
[16:27:17] <LoRez> what does "full on" mean to you then?
[16:27:20] <Fleck> *i am
[16:27:46] <LoRez> if you're pwming the LED, you're blinking it.
[16:27:49] <Fleck> i regulate voltage on pins 1 and 2
[16:27:59] <Fleck> i use regulated voltage supply
[16:28:03] <LoRez> and what exactly do you think that'll do?
[16:28:19] <LoRez> that isolated triac has two states, on and off.
[16:28:45] <Fleck> ohh, i whant to fade light :(
[16:29:46] <LoRez> then you need to get rid of the zero cross optoisolator and use a normal one. feed a zero cross signal to the circuit controlling the LED in the opto and time your turn-ons until after the zero crossing by some amount of time
[16:29:55] <Xata> i am not good electrician yet, but i think you have some sort of loop between moc and bt. also capacitor. that's all i can say :D
[16:31:45] <Fleck> LoRez MOC3021 will do?
[16:32:31] <LoRez> should
[16:32:38] <Fleck> ok thx guys! :)
[16:32:58] <LoRez> Fleck: http://sound.westhost.com/appnotes/an005.htm
[16:33:54] <Fleck> yeah, stupid me :D
[16:34:53] <Fleck> didnt know that zero crossing means on or off :D
[16:36:43] <Xata> me when i see question about electricity http://25.media.tumblr.com/tumblr_ln3t7w7zfo1qb5gkjo1_400.jpg
[16:39:08] <LoRez> zero crossing means it will only turn on when the AC crosses the 0V point, to save the components/load from the impulse of near instantenous current when you turn it on in the middle of the waveform
[16:39:39] <Fleck> LoRez yeah
[16:39:52] <Fleck> ill buy moc3020 then! :)
[17:07:48] <Xata> bye
[21:38:27] <Essobi> Soooo...
[21:38:47] <Essobi> TWI/i2c slaves hard?
[21:41:45] <Essobi> Tom_itx: sup
[22:44:41] <Essobi> Anyone played with i2c slaves?
[22:44:54] <Essobi> err... AVRs as an i2c slave