#avr | Logs for 2013-02-05

Back
[04:45:41] <Malinuss> when I try to do single adc measurments from different sources, I get some very weired results.... Like if I try and measure from adc pin 0 first, and then from adc pin 1 the results somehow seems to be random. Here is my adc function - http://pastebin.com/cDVJNrmC
[05:56:48] <dunz0r> Malinuss: Iirc, the first results can't be trusted, try doing one initial conversion and then reading the result again.
[05:56:53] <dunz0r> I'm not entirely sure though, might be some specific sensor that is like that.
[05:57:47] <Malinuss> dunz0r, nah just two pots. I'll try multiple conversions on each channel first
[06:01:13] <dunz0r> Btw... can registers "get stuck"?
[06:01:39] <dunz0r> I've got a function which does AD-conversions, but I get an 8-bit result, even though I haven't specified it.
[06:02:14] <dunz0r> Here's the code http://git.hax0r.se/?p=AVR/.git;a=blob_plain;f=Minisumo2013/adc.c;h=6b586b3f4e2797e7a5a7f3a2c6d4532b94d38006;hb=HEAD
[06:04:44] <OndraSter_> dunz0r, and what returns the result?
[06:13:28] <Malinuss> okay so I figured out a way to read the VCC, using the internal refereance. yay... BUT - if I try to make any other kind of adc usage, both results will always be wrong ("vcc function" will return -1, and adc from a pin would return 0).. Here is the code for the functions: http://pastebin.com/ESfKcABr and my main loop: http://pastebin.com/3csCNxxh
[06:13:47] <Malinuss> also - who needs loops when you have ctrl+c and ctrl+v, haha
[06:26:52] <dunz0r> OndraSter_: It's triggered by an interrupt, so it's stored in ad_channel[channelnumber]
[06:31:22] <GuShH_Lap> Malinuss: huh? the compiler will unroll loops, depending on settings and constraints
[06:56:04] <Malinuss> GuShH_Lap, what do you mean?
[06:56:17] <GuShH_Lap> look up loop unrolling
[06:56:20] <GuShH_Lap> re your copy paste joke.
[06:56:34] <Malinuss> huh?
[06:57:01] <Malinuss> sorry I'm pretty thick-headed. What do you mean?
[06:57:38] <Malinuss> GuShH_Lap, can't really see what it has to do with my problem though
[06:58:42] <specing> -funroll-loops
[06:59:10] <GuShH_Lap> sigh.
[06:59:23] <GuShH_Lap> because everything everyone says has to be a solution for your problems.
[06:59:39] <GuShH_Lap> go back to tom cruise's closet!
[07:02:03] <Malinuss> GuShH_Lap, haha sorry. I get it now - I was just confused, don't get so mad over it... Of course when I read your text I read it with "he is responding to my question"-glasses. Especially if I don't get it..
[07:04:53] <theBear> surely you can comprehend "look up ...", being thickheaded isn't an excuse for that, and generally in technical forums you are better receieved when you don't make jokes about things you don't understand
[07:05:49] <Malinuss> theBear, when did I make a joke about something I did not understand?
[07:06:12] <theBear> "who needs loops when you have ctrl+c and ctrl+v, haha"
[07:06:21] <theBear> looks like a joke to me
[07:08:51] <Malinuss> theBear, in this context? Why in the world would you even use loop unrolling in a example like this... sigh, you guys.. Asperger team or what?
[07:09:57] <theBear> mmm, insults aren't a good way to get your questions answered
[07:12:09] <theBear> not to mention being a technical internet forum in this modern day and age, it's not unlikely one or several people here do suffer from aspergers, in which case that would be even more offensive
[07:12:21] <Malinuss> But seriously, this was just a little over my aspie level.. It's like if someone make a joke about "hey I used floats instead of integers because example, lol", and you would go "hey don't make jokes about things you don't understand"...
[07:12:53] <Malinuss> theBear, It's just a saying... Like "don't be such a jew" etc. (no offense for the minority)...
[07:13:49] <Malinuss> but fair enough, if it makes you feel more superior to tell me "not to make jokes about things you don't understand" - fair enough
[07:13:57] <theBear> DO you understand compilers and how they build/optimise code in detail ? it sure sounds like you don't... and that example just there, notice how the phrase used DOES insult jewish peoples ? it's pretty much the same thing
[07:14:27] <theBear> you mentioned someone getting so mad at you, i offered an explanation why...
[07:15:28] <theBear> we don't hang around here to make fun of people, we are here to learn and to help others do the same... noone was making fun of you, at least not initially, they may have resorted to it eventually tho
[07:16:14] <theBear> bored now... i'm gonna go do something
[07:21:36] <GuShH_Lap> holy shit he went through the aspie route and took a left turn to the hitler route, awesome.
[07:21:37] <GuShH_Lap> this guy!
[07:21:49] <GuShH_Lap> way to go buddy
[07:22:32] <GuShH_Lap> although he forgot to mention slaves
[07:22:39] <GuShH_Lap> so, one point deducted.
[07:22:50] <OndraSter_> Chewbacca time
[07:23:03] <OndraSter_> IT DOES NOT MAKE ANY SENSE!
[07:24:18] <theBear> heh
[07:28:51] <OndraSter_> (for the others... South Park)
[08:03:00] <Malinuss> OndraSter_, you mean I forgot to mention some of the steortypes about black people? heh... theBear I know you were not making "fun", it was merely a way to point out "hey you are a noob don't think you can act cool". I get it
[08:04:49] <theBear> no, more a way of pointing out you are immature and don't know how to interact with people
[08:05:20] <specing> hehe
[08:07:00] <GuShH_Lap> no one actually called you a noob
[08:07:06] <GuShH_Lap> but it is clear you feel like one
[08:07:10] <GuShH_Lap> otherwise why would you bring that up
[08:07:17] <GuShH_Lap> it's ok, freud knew his shit.
[08:07:26] <Malinuss> theBear, I was refering to your initial comment "don't joke about things you don't understand"
[08:08:09] <Malinuss> GuShH_Lap, I don't "feel" like a noob. I know I am a noob, I am a complete beginner in this hobby. I would never question that..
[08:08:39] <GuShH_Lap> It wasn't implied either
[08:08:56] <GuShH_Lap> If you want to be called a n00b you could try 4chan
[08:09:04] <GuShH_Lap> Or any online game
[08:10:25] <Malinuss> GuShH_Lap, I never said you were. You were just poining out that there is actually a thing in programming called loop unrolling. I think I called you out too, because I got mad at theBear, sorry about that, heh
[08:48:55] <theBear> admitting fault... perhaps now the healing can begin
[08:55:52] <Malinuss> Nah maybe I was just a bit touchy..
[08:58:45] <jadew> fight! fight! fight!
[08:58:59] <OndraSter> YEEEES
[08:59:01] <OndraSter> FIGHTTTT
[08:59:08] <OndraSter> <Malinuss> Nah maybe I was just a bit touchy..
[08:59:10] <OndraSter> that's what HE said
[08:59:28] <creep> h
[09:00:19] <Malinuss> b-but we just made friends?
[09:02:37] <rue_house> theBear, is australlian, you dont stand a chance
[09:03:20] <darknite> australians doesn't have friends, they just keep a short mental note of people not to beat up
[09:03:31] <OndraSter> haha
[09:03:34] <OndraSter> they do not have friends
[09:03:36] <OndraSter> they have mates
[09:03:38] <OndraSter> and they mate
[09:03:50] <OndraSter> instead of have sex
[09:03:51] <rue_house> he's been trampled by more ute's and squashed by more things,
[09:04:03] <rue_house> I dont think anything you can throw at him can phase him
[09:04:41] <rue_house> what are you making anyhow
[09:07:00] * RikusW wonders whether flyback is au of not...
[09:07:24] <rue_house> no, flyback is just an asshole
[09:07:43] <rue_house> but that not why we have him banned from here
[09:07:48] <rue_house> he's vulgar too
[09:08:04] <rue_house> Malinuss, what WAS your origional question, I cant find it
[09:09:31] <specing> flyback is a redneck I think
[09:09:43] <specing> and redneck have a grudge against canucks?
[09:09:49] <specing> +s
[09:10:16] <Malinuss> okay I've just redone my question... I have a general problem when using the adc, and trying to either change the channel, or the reference in the code, or do both.. If I don't change it everything is fine. I've made a easy-to-read (and short) pastebin visualising the problem in code: http://pastebin.com/iEfEEt6m
[09:10:42] * specing wont open pastebin.com links
[09:11:01] <RikusW> whats wrong with pastebin specing ?
[09:11:25] <specing> its just ... wrong
[09:11:53] <Malinuss> just for you specing: http://pastie.org/6060230
[09:12:01] <rue_house> pff
[09:12:04] <Malinuss> haha
[09:12:43] <rue_house> you can make channel a static int IN the read function if you want
[09:13:23] <rue_house> wow, your polling it!
[09:13:38] <Malinuss> rue_house, explain pleas?
[09:14:08] <rue_house> I set the adc up in an interrupt loop that puts the values of the channels in an array
[09:14:20] <rue_house> once the interrupt is going, you just read the values from the array
[09:14:23] <rue_house> :)
[09:14:27] <rue_house> but whatever
[09:14:41] <specing> That code will be so sluggish I don't wanna look at it again
[09:14:59] <Malinuss> specing, I've made all the comments so people won'
[09:15:00] <Malinuss> t
[09:15:04] <Malinuss> have to look in the datasheet
[09:15:10] <Malinuss> nothing else sluggish about it
[09:15:11] <rue_house> huh, interesting problem
[09:15:31] <Malinuss> rue_house, yeah that is clever if you don't want to wait. This will not be an issue though, also right now I just want to understand why this doesn't work
[09:15:31] <rue_house> specing, dosn't hurt to help with the problem
[09:15:45] <rue_house> yea
[09:15:59] <rue_house> I bet I know the problem
[09:16:03] <rue_house> I bet its hardware
[09:16:05] <rue_house> :)
[09:16:12] <rue_house> Malinuss, try this
[09:16:28] <rue_house> put a pulldown resistor on EVERY ADC channel, then do your test
[09:16:32] <Malinuss> specing, but at least give me some constructive feedback, so I can learn from it?
[09:16:40] <specing> interrupts.
[09:16:41] <rue_house> I think you will find the channels your selecting are not the channels you think they are
[09:16:53] <jadew> what's the issue?
[09:17:03] <rue_house> specing, you havn't said anything to help with the acual issue
[09:17:15] <rue_house> Malinuss, catch what I'm saying?
[09:17:21] <specing> Ok
[09:18:00] <rue_house> specing, many times it SEEMS like your reading a channel fine becuase the values change, but it turns out its just electrostatic coupling between the pins
[09:18:24] <rue_house> specing, if a person puts a pulldown/up resistor on all the pins, they find out that their channel numbers are out
[09:18:41] <rue_house> but once you know that, its easy to fix/compensate for
[09:18:46] <Malinuss> rue_house, one thing: then why is the channel the one I think it is, when I only choose it once, and stick to it? I'm using pots so I'm pretty sure the numbers are tight
[09:19:13] <specing> I still don't know what the exact problem is
[09:19:14] <jadew> Malinuss
[09:19:21] <specing> just "general problem"
[09:19:22] <jadew> why _delay_ms(10);
[09:19:24] <rue_house> Malinuss, electrostatic coupling, trust me, if thats the problem, it can fool you well
[09:19:32] <specing> Also this works for me: http://sprunge.us/ZhML
[09:19:46] <rue_house> jadew, the delay has NOTHING do to with the problem
[09:19:58] <jadew> rue_house, I know
[09:20:02] <rue_house> it could be a 10 second delay and it woulnd't matter
[09:20:06] <jadew> it shouldn't even be there
[09:20:15] <Malinuss> jadew, "when changing the reference or the channel it make take up to 125usec to settle"- datasheet
[09:20:42] <rue_house> Malinuss, pull all the other adc channels up or down with a resistor and I'd bet you your first two loops dont work anymore
[09:20:44] <Malinuss> jadew, is should, just not that long.. or toss the first number, doesn't matter though - not part of the problem like rue_house said
[09:20:54] <jadew> Malinuss, I was expecting the hardware to handle that
[09:21:10] <Malinuss> Special care should be taken when changing differential channels. Once a differential channel
[09:21:10] <Malinuss> has been selected, the stage may take as much as 125 µs to stabilize to the new value. Thus
[09:21:10] <Malinuss> conversions should not be started within the first 125 µs after selecting a new differential channel. Alternatively, conversion results obtained within this period should be discarded.
[09:21:11] <Malinuss> The same settling time should be observed for the first differential conversion after changing
[09:21:11] <Malinuss> ADC reference (by changing the REFS1:0 bits in ADMUX).
[09:21:15] <Malinuss> sorry
[09:21:27] <rue_house> obviously, the first reading might be off cause of the reference change, but thats it
[09:21:34] <Malinuss> ^
[09:21:50] <rue_house> Malinuss, you gonna put those resistors in and try again
[09:21:51] <rue_house> ?
[09:22:09] <jadew> Malinuss, good to know, didn't know that about the settling time
[09:22:48] <rue_house> I'v had PIC's that the datasheet had the ADC channel order ALL WRONG
[09:22:59] <rue_house> makes a good mess of an already-etched pcb
[09:23:29] <Malinuss> rue_house, of course, just wanna make sure I do what you want me to do. So I put a let's say 10k resistor between the adc pins and the ground?
[09:23:40] <rue_house> yup
[09:23:58] <rue_house> all the channels, not just the ones your using
[09:24:11] <jadew> Malinuss, are you running all the 3 functions at once?
[09:24:15] <jadew> try that please
[09:24:16] <rue_house> no
[09:24:24] <jadew> loop1() loop2() loop3()
[09:24:35] <rue_house> he gave 3 examples of the loops that do and do not work
[09:24:44] <rue_house> instead of posting code 3 times
[09:24:59] <Malinuss> ^
[09:25:17] * rue_house hits jadew with a wiffle bat "duh!"
[09:25:26] <jadew> he should run all 3 of them in one go
[09:25:52] <jadew> eventually with a 4th loop in which he displays adc 1 before adc 0
[09:26:32] <rue_house> no I think this is a hardware issue with channel identity
[09:26:51] <rue_house> so I'm betting your solution would not help in the least
[09:27:13] <jadew> the reason why I want the results of that test is to rule out hardware issue
[09:27:19] <jadew> I'd much rather blame gcc
[09:27:21] <rue_house> I say this based on experiance, lots of "why the hell is that channel going down when this one goes down?"
[09:28:15] <theBear> darknite, hehe, i like that summing up of me and 'my peoples'
[09:28:47] <rue_house> theBear, are you in one peice latley?
[09:29:06] <Malinuss> rue_house, okay with some additional testing I can tell you this: "loop 1" and "loop 2" works just as I said. The pulldown resistor didn't seem to make any change. I'm using a pot so I am very confident that the numbers are not fooling me... So I am afraid that the problem isn't with the MUX setting vs which pin is actually chosen...
[09:29:24] <theBear> and as a sidenote, i'm not fond of the term 'mate' .. these days when english or australians use it, it is often like calling someone 'sir' ... kind of a formal "i'm trying to be nice here, but you're in trouble" term ....
[09:29:34] <rue_house> you said 'resistor' did you put a resistor on EVERY channel at once when you did the test?
[09:29:39] <theBear> rue_house, well, i haven't gotten any more crippled :) and i had a good couple days :)
[09:30:06] <rue_house> I'm telling you that if your reading a floating channel BESIDE the pin the pot is connected to, you will get "good looking values"
[09:31:02] <rue_house> theBear, so your maybe a bit closer to getting back on your feet?
[09:31:07] <Malinuss> rue_house, no I did not. Will it be enough with 0-7 or should I get the 7-11 too? Also I can promise you there are no floating numbers, the numbers follow even the slightest turn of the pot, being 0 and 1023 at each end and not floating at all. Neverless I do like you say
[09:31:11] <rue_house> hows the washing machine :)
[09:31:37] <rue_house> Malinuss, floating pins, pins that do not have any connections
[09:31:56] <rue_house> Malinuss, yes, if its electrostatic carryover they WILL do that
[09:32:07] <rue_house> untill you put a pulldown resistor on the channel its actually reading
[09:32:23] <theBear> rue_house, maybe physically, pretty sure whatever happens i'm still STRONGLY advised not to 'do anything' fer another err, 10 weeks or so, for my likelihood of not relapsing again... more should develop in a week or so after my next doc visit
[09:32:33] <rue_house> Malinuss, every adc channel, connect to something
[09:33:30] <theBear> washing machine ? awesome ! still quiet, no false fullness triggering, minimum vibration etc etc... i'm sure 90% of it is the lubed suspension.. amazing how much difference it makes... not like it was siezed before, just not supersmooth/a bit sticky i spose
[09:33:33] <rue_house> theBear, you should make side money from canada recording tel messages with your accent :)
[09:33:40] <Malinuss> rue_bed, roger. one thing though: I tried reading (one compile at the time), all the channel but the one that was connected, and the number floated, expect the right one. But neverless I will try this
[09:33:47] <theBear> heh, i'm surprised they still do that... all robots here
[09:33:56] <rue_house> theBear, did you get a new machine?
[09:34:11] <theBear> nah, same one, just serviced it a few weeks ago
[09:34:21] <rue_house> cool
[09:34:25] <rue_house> ok I'm late for work
[09:34:46] <theBear> was gonna be a minor service, but curiosity.. pulled a plate off and oil went everywhere, so it got gearbox service/refill and most everything i could think of :)
[09:34:55] <theBear> hehe, go son, can't always blame us for making yer late
[09:35:27] <rue_house> Malinuss, I know it seems like a pointless hoop, but its an easy one, and I'm 90% sure I'm right
[09:36:44] <Malinuss> I'm doing it, wish I had actual breadboard cabel right now, hah
[09:37:42] <rue_house> want 50' of 24 guage solid copper wire?
[09:38:22] <Malinuss> haha ' is that feet or inches? I can't remember
[09:38:29] <theBear> hehe, feets
[09:38:33] <OndraSter> " is inch
[09:38:44] <Malinuss> oh snap
[09:38:48] <theBear> for reference most cat? network cable and old phone cabling is good for breadboards
[09:38:54] <Malinuss> that is one long copper wire
[09:40:03] <theBear> i'd like to see a wire factory one day... supposedly wire is pulled, tho they probly got a fancy trick these days, plus, however you make it you gotta be fancy to be able to make a single 1000' reel of network cable/whatever
[09:40:05] <OndraSter> 50 ft = 50*30 = 1500cm = 15m. USE PROPER UNITS damnit!
[09:40:11] <Malinuss> yeah all I have are some old wires from a broken pcu.. all of them are not solid, so I have to twist, and twist, and twist some more, hehe
[09:40:42] <theBear> mm, not solid will make you sad, and all those little broken bits in the bottom of the breadboard will make you confused and sad eventually
[09:40:57] <Malinuss> yep
[09:41:27] <Malinuss> ordered some "breadboard wire set" for 3$ on ebay, heh
[09:41:35] <theBear> nothing quite as confusing as the first few times yer breadboard goes unreliable... not real good once you know it's got problems, but REAL confusing at before you work that out
[09:42:30] <Malinuss> theBear, I can't really see how small bit's inside can change anything...? Can they actually connect somethow to a other row?
[09:43:29] <theBear> depending on the breadboard they often can... tiny scraps of wire can go all sorts of places given enough time and vibration (aka bumps)
[09:44:07] <Malinuss> I kinda hoped there would be solid plastic between rows
[09:44:58] <OndraSter> there is still open back
[09:45:08] <OndraSter> otherwise they couldn't fit in the metal inserts
[09:45:25] <Malinuss> rue_house, I should have asked eaerlier, but it will be fine if they all go to the same pulldown, right?
[09:46:03] <jadew> obviously not
[09:46:17] <jadew> going to the same pulldown means they are shorted together
[09:47:12] <Malinuss> im sad
[09:47:14] <Malinuss> hehe
[10:01:54] <Malinuss> rue_bed, okay I did it. And after some testing I can tell you this. For adc0 and adc1 everything works as I said. Too add to the confusion: if I choose any other channels (2-7) I keep getting 1023, even though I have pulldowns on them all... And I'm pretty sure that channel=channel & 0b00000111; and then simply ADMUX|=(1<<REFS0)|channel; should work for the selection
[10:02:28] <Malinuss> but adc 0 and 1 works just as expected, even with pulldowns on all the channels.. hmm
[10:10:58] <theBear> you checked that you are actually measuring adc0 and 1 ? that looks/sounds a lot like your channel increment is a little wrong
[10:11:25] <Malinuss> theBear, I have a pot on adc0 and 1, and they follow it *perfectly*
[10:12:39] <theBear> ok, that's a good start
[10:13:14] <theBear> you sure you set the port modes and a-d registers right ? (haven't done multiport ad for years, i forget the details)
[10:16:25] <Malinuss> what do you mean with setting the a-d registers right? The I/0 registers? I don't set those at all. If you mean the control register for adc, I am confident I set it right, since the values from adc0 and adc1 are right... I don't understand what happens when I try to change the channel though (from 1 to 0 or 0 to 1)
[10:18:32] <theBear> generally for something like a-d taking over general io ports you need to both set the ports to the 'right' values, i would guess input in this case, AND tell some register that those ports are gonna be a-d ports and not normal io
[10:27:24] <abcminiuser> Hey all
[10:27:30] * abcminiuser has a new apartment
[10:33:51] <DagoRed> hello
[10:33:51] <tobbor> DagoRed! like, totally tell us about the project!
[11:04:53] <megal0maniac> hello
[11:04:54] <tobbor> Hello megal0maniac
[11:04:56] <megal0maniac> :)
[11:29:33] <creep> http://www.androidinabox.com/mini-joystick-capacitive-touch-screen-joystick-playing-games.html
[11:34:21] <creep> Horologium ! :) how about a capacitive joystick ?
[11:34:55] <creep> ofc this is more like a force-sensing stick
[11:35:03] <creep> not much movement
[11:35:58] <megal0maniac> Like laptop trackpoint?
[11:56:12] <Malinuss> abcminiuser, maybe you know this (you proapbly do, the question is if you have time for it).. Well when I change the adc channel or refereance I get some weired bahviour... this nice little example shows 3 example loops each with explanation about the results. Hope you have a minute: http://pastebin.com/iEfEEt6m
[11:56:35] <Malinuss> I'm at the point of thinking that I have somehow damaged the adc, heh
[11:57:07] <abcminiuser> ADMUX|=(1<<REFS0)|channel;
[11:57:28] <abcminiuser> Once those bits are set you can't clear them with your code
[11:57:39] <abcminiuser> So if you make a channel selection, those mux bits are "stuck"
[11:57:39] <Malinuss> oh shit
[11:57:44] <Malinuss> yeah I get it
[11:57:53] <Malinuss> thanks.. sigh I can't belive it's just that...
[11:58:05] <Malinuss> really, thanks
[11:58:11] <abcminiuser> :)
[12:08:48] <Malinuss> abcminiuser, now one little other thing - in the datasheet the internal V referance is said to be 2.56V.. Only in the mux it actually says 1.1V, and my "calculations", shows that the internal referance is actually 1.1V and not the 2.56V... (atmega32u4) Maybe you could clear this up :)?
[12:09:27] <abcminiuser> Sounds like you're using Bandgap
[12:09:32] * specing hasn't coded for more than half a year now
[12:09:41] <specing> GaaH
[12:10:33] <Malinuss> abcminiuser, the unit I'm using is actually the small teensy devboard.. Could the Bandgap be on it?
[12:10:47] <abcminiuser> Bandgap is an internal reference
[12:10:57] <abcminiuser> There's two references you need to know:
[12:11:03] <Malinuss> ah yes: 1.1V (VBand Gap
[12:11:03] <Malinuss> )
[12:11:12] <abcminiuser> 2.56V internal, which can be used as the reference for ADC_MAX
[12:11:35] <OndraSter> abcminiuser, y u no TPI support on Dragon :(
[12:11:48] <abcminiuser> And the 1.1V bandgap, which can be used as a reference for differential mode
[12:11:58] <abcminiuser> OndraSter, if you'd ever seen the firmware you'd know
[12:12:05] <OndraSter> huh?
[12:12:08] <OndraSter> is it a mess?
[12:12:11] <abcminiuser> Of the Dragon
[12:12:12] <Malinuss> ah I see. Thanks for clearing this up!
[12:12:17] <abcminiuser> It is apparently not pretty
[12:12:19] <Malinuss> *thanks for making this clear
[12:12:22] <OndraSter> haha you have seen it?
[12:13:10] <megal0maniac> Wait, what? No TPI?
[12:13:17] <OndraSter> nope
[12:13:22] <OndraSter> TPI = attiny4/5/9/10
[12:13:28] <megal0maniac> Yeah I know
[12:13:31] <OndraSter> and it sucks tbh
[12:13:32] <Malinuss> is it not possible to porgram those tinys on it then?
[12:13:58] <OndraSter> nope
[12:14:03] <OndraSter> not these tiniest tinies
[12:14:21] <megal0maniac> 25/45/85 :)
[12:14:28] <OndraSter> those have got regular SPI ISP
[12:14:33] <OndraSter> or debugWire for debug
[12:14:39] <OndraSter> iirc
[12:14:44] <megal0maniac> Exactly my point
[12:14:47] <megal0maniac> They do
[12:14:58] <abcminiuser> I haven't seen it, just asked the guys that have
[12:15:01] <OndraSter> well then you can use them with Dragon
[12:15:06] <OndraSter> ah abcminiuser
[12:15:16] <megal0maniac> OndraSter: So get those instead :)
[12:15:22] <OndraSter> :P
[12:15:28] <OndraSter> but they are big compared to those SOT23-6
[12:15:37] <OndraSter> don't worry, I am even using attiny13a
[12:15:39] <megal0maniac> Ah, true
[12:15:41] <OndraSter> just because it is cheaper than attiny85v
[12:15:48] <OndraSter> and I need maybe 128B of code :P
[12:15:50] <megal0maniac> What for??
[12:15:53] <megal0maniac> -?
[12:15:53] <OndraSter> small dice :D
[12:16:00] <megal0maniac> Oh, that :)
[12:16:08] <OndraSter> probably will be sold as a kit for the school
[12:16:09] <OndraSter> preprogrammed
[12:16:17] <OndraSter> but let the students solder it
[12:16:24] <megal0maniac> How's the USB stack / xide / xboard mini coming along?
[12:16:26] <OndraSter> instead buying those $10 - $20 stuff from regular places
[12:16:30] <OndraSter> school dude
[12:16:51] <OndraSter> the USB CDC works
[12:16:53] <megal0maniac> I understand
[12:16:55] <OndraSter> I can rx and tx serial data
[12:17:24] <megal0maniac> Nice
[12:17:50] <megal0maniac> First class of digital systems today. We learnt about logic gates :P
[12:18:07] <OndraSter> hehe
[12:18:55] <megal0maniac> My main goal at the moment is to set up a pocket router to deal with the proxy and authentication and pass on nice, clean internet
[12:19:31] <megal0maniac> Because most programs don't know how to authenticate :/
[12:19:35] <megal0maniac> Stupid Windows
[12:19:46] <Malinuss> abcminiuser, okay, okay. Last thing - I promise! So as far as I can tell the channel=channel & 0b00000111; ADMUX|=(1<<REFS0)|channel; should work for channel= 0-7... Neverless if channel is anything else then 0 or 1, I just get the value 1023, no matter what.. Like the numbers are not even floating when nothing is connected..
[12:20:15] <OndraSter> aren't the pins set as output?
[12:20:59] <Malinuss> They should be input as default, and I don't change them. Neverless I set them up as input first...
[12:23:25] <abcminiuser> There isn't a channel 2 or 3 in those devices
[12:23:33] <abcminiuser> 0,1,4,5...
[12:26:55] <Malinuss> oh god.. Well at least, no matter how simple a thing I do with a avr you will be impressed abcminiuser... How could I be this retarded.. Really. haha.. Well thank you again. really, thanks.. sigh..
[12:27:27] <abcminiuser> No problem, 90% of my job is just referencing the datasheet :P
[12:28:12] <kobsu> abcminiuser: does Eric Weddington use irc and what might be his nick?
[12:29:12] <abcminiuser> Not sure about that, I did meet him once
[12:29:17] <abcminiuser> He's a very private person
[12:29:22] <abcminiuser> Hates photos
[12:29:29] <abcminiuser> I do too, but not to THAT extent
[12:33:57] <Malinuss> abcminiuser, anyone ever recognized you on the street? "Hey you'r the avr man, right?"
[12:34:21] <kobsu> abcminiuser: ok, maybe i'll email him
[12:34:21] <abcminiuser> Bwahahahaaha no I'm not that popular
[12:34:33] <Essobi> abcminiuser: You are to us. :D
[12:34:36] <abcminiuser> I'm just a dude that seems to pop up in strange places online :P
[12:34:37] <DagoRed> I bet to differ abcminiuser
[12:34:39] <Essobi> *squeeee*
[12:34:51] <Essobi> Hehe.
[12:34:53] <abcminiuser> I really hope one of y'all is female :S
[12:34:54] <megal0maniac> DagoRed: We /are/ on an AVR IRC channel :P
[12:35:08] <abcminiuser> kobsu, what do you need him for?
[12:35:17] <kobsu> abcminiuser: btw. i was looking forward your theses... it pop up when i was googling bluetooth stuff :D
[12:35:20] <Malinuss> I've seen your photo so many times reading your tutorial, I would have recognized you...
[12:35:31] <Essobi> abcminiuser: I was being facetious. :D </sarcasm>
[12:35:40] <abcminiuser> I know, I kid, I kid
[12:35:55] <abcminiuser> Malinuss, heh that's basically where I got my fame
[12:36:03] <Essobi> I do probably owe you a beer or something thou.
[12:36:10] <abcminiuser> Me at 15(ish): "Hey, I should write this down."
[12:36:16] <kobsu> abcminiuser: i would like to ask if it would be possible to distribute avr32 toolchain with dfu-programmer + inf-file and st2
[12:36:33] <abcminiuser> st2?
[12:36:37] <Malinuss> abcminiuser, you made those tutorials at a age of 15? Not bad..
[12:36:39] <kobsu> abcminiuser: looking forward == through :/
[12:36:43] <kobsu> abcminiuser: sublime text 2
[12:36:55] <abcminiuser> Jesus I love Sublime, I switched to it in Jan
[12:37:19] <abcminiuser> Fantastic but I was waiting for an update before I bought since it was inactive since September
[12:37:27] <abcminiuser> Then he jacked up the price
[12:37:35] <abcminiuser> Still, I guess I do use it a lot...
[12:37:42] <abcminiuser> kobsu, do you have his work email?
[12:37:51] <kobsu> i don't care the price... i can pay double to upgrade st3 ;)
[12:38:07] <kobsu> abcminiuser: nope, can you give it to me?
[12:38:34] <kobsu> abcminiuser: i found some from google, but don't know if he is reading those or if those are correct
[12:38:40] <abcminiuser> One sec I'll dig it up
[12:38:41] <abcminiuser> PM me
[12:41:35] <megal0maniac> abcminiuser: Looks slick, but owie. The price...
[12:41:46] <abcminiuser> Ja
[12:41:53] <abcminiuser> BUT IT WORKS SO GODDAM WELL
[12:42:03] <abcminiuser> CTRL+D and multiple cursors is like digital bacon
[12:42:04] * megal0maniac clicks download
[12:42:58] <OndraSter> lol
[12:43:08] * OndraSter does not really like bacon
[12:43:14] <OndraSter> if I know that it is a bacon
[12:43:21] <OndraSter> if I don't know it and there are eg eggs on top of it I don't mind it
[12:43:58] <abcminiuser> OndraSter, you are now dead to me
[12:44:01] <OndraSter> :D
[12:44:23] <kobsu> abcminiuser: do you use sublimeclang in st2?
[12:44:36] <abcminiuser> Nei, couldn't get it to index the header files correctly
[12:44:52] <abcminiuser> So just one of the jankier plugins that makes it auto-suggest from open files
[12:45:15] <abcminiuser> Also added the Modific package to get indicators of removes/modified/added lines in SVN/GIT
[12:45:25] <abcminiuser> And CFeather for usable C syntax support
[12:45:30] <abcminiuser> And a bunch of others
[12:45:36] <kobsu> you can sneak correct settings from my plugin https://github.com/aery32/sublime-aery32/blob/master/Aery32.py#L139
[12:47:20] <abcminiuser> Yoink
[12:48:14] <megal0maniac> WOAH
[12:48:20] <megal0maniac> MULTIPLE CURSORS
[12:48:24] <megal0maniac> Now I see
[12:48:49] <kobsu> abcminiuser: here's another example https://pastee.org/45hp3
[12:49:00] <kobsu> abcminiuser: that script makes that
[12:49:16] <abcminiuser> Holy crap
[12:52:04] <megal0maniac> Meh. "Educational discounts are not presently offered."
[12:52:28] <kobsu> megal0maniac: for sublime text?
[12:52:34] * megal0maniac nods
[12:53:05] <kobsu> abcminiuser: this is how the auto complete works for me with sublimeclang, http://kobsu.kapsi.fi/example/example.html
[12:55:07] <abcminiuser> Sweet merciful tapdancing monkey christ
[13:03:23] <kobsu> abcminiuser: do you know if it's possible to reprogram dfu bootloader (UC3A1) with AS6?
[13:04:28] <abcminiuser> Sure, but you need an external programmer
[13:04:47] <abcminiuser> Or do you mean program in an application via the bootloader using AS6?
[13:04:55] <kobsu> i am afraid that i have managed to brick my aery32 when i was trying to do that after doing some in-system debugging... my old school method did not work anymore for some reason
[13:05:10] <kobsu> i want to put dfu back
[13:05:21] <kobsu> it was erased when i was doing in-system debugging
[13:06:29] <kobsu> this is my method that didn't work, most likely because virtualbox betrayed me... http://devzone.aery32.com/2012/06/17/how-to-reprogram-the-dfu-bootloader-for-uc3-chips/
[13:06:48] <Essobi> kobsu: What'd you make that animation with?
[13:07:41] <abcminiuser> Well you can just attach a JTAG to it and reprogram it in
[13:07:48] <abcminiuser> The HEX files are on the Atmel website IIRC
[13:10:26] <kobsu> yep but thos hex files are for C-series only
[13:10:33] <kobsu> v1.1.0 IIRC
[13:10:48] <abcminiuser> Hrm, then they might have it inside AS6 as an ASF demo
[13:11:02] <abcminiuser> Seach for DFU in the new example wizard
[13:11:14] <kobsu> Essobi: https://github.com/sublimehq/anim_encoder
[13:11:23] <kobsu> abcminiuser: did that no dfu for a-series
[13:12:36] <Tom_itx> abcminiuser, did you see my question about the at90a4414 and at90a2313?
[13:12:44] <abcminiuser> Tom_itx, the what now?
[13:12:45] <Tom_itx> does studio still support those?
[13:12:55] <Tom_itx> if so do you know what tools
[13:12:57] <kobsu> and i managed to erase user page as well and now there's no device id... seems like a-series doesn't have separate factory page :/
[13:13:13] <abcminiuser> Tom_itx, no, they are too old I would think
[13:13:41] <abcminiuser> TINY2313 yes, but AT90 is ANCIENT
[13:13:48] <abcminiuser> OLDY OLDY OLDY
[13:14:07] <abcminiuser> kobsu, my trunk build of Studio has crapped out so I can't look for you
[13:14:26] <abcminiuser> But you might be able to make whatever default series UC3 DFU is in there and just change the device
[13:14:33] <abcminiuser> If it uses ASF, it should auto-adapt
[13:15:06] <OndraSter> UC3 is not like xmega?
[13:15:09] <OndraSter> all registers on the same positions..
[13:15:18] <abcminiuser> No
[13:15:25] <abcminiuser> They were....organically designed
[13:15:40] <abcminiuser> And no one but the IC designers seem to know exactly what is in each silicon :P
[13:15:48] <kobsu> abcminiuser: yeah no problem... i'm not using asf... i guess that i just accept that this board has been bricked and so be it
[13:16:07] <abcminiuser> kobsu, if you email me at work I can find the HEX for you
[13:16:12] <abcminiuser> dean_patrick.camera
[13:17:20] <kobsu> ok thanks... i'm just still wondering if it will work as the jtag cannot recognize the device ide anymore... and that's because i erased user page
[13:17:45] <abcminiuser> You can't erase the chip signature
[13:18:07] <kobsu> within uc3a1 it seems to be stored in user page
[13:18:32] <kobsu> C-series has a separate factory page for that i guess
[13:18:45] <abcminiuser> It shouldn't be user-writeable, but I don't know enough about it to be certain
[13:19:25] <Tom_itx> abcminiuser, do you know on those 4414 if they're ISP or something else?
[13:19:47] <RikusW> abcminiuser: the at90s4414 got just TWO fuse bits... SPIEN + FSTART (startup time)
[13:19:54] <RikusW> that is old...
[13:19:55] <Tom_itx> yes
[13:20:28] <RikusW> whats more it can only be written using HVPP
[13:20:35] <RikusW> (the fuses)
[13:20:39] <abcminiuser> Apparently it's so old it's not even in my datasheet archive under mature
[13:20:40] <Tom_itx> the .hex or the fuses?
[13:21:06] <RikusW> abcminiuser: I got at90s2313 on my 2005 cd
[13:21:11] <RikusW> but not the 4414
[13:21:32] <RikusW> Tom_itx: the fuses
[13:21:34] <abcminiuser> Yeah I've got a big-ass Techlib of datasheets
[13:21:39] <abcminiuser> And no 4414
[13:21:52] <RikusW> I got the new techlib DVD too
[13:21:58] <RikusW> almost no at90s in there
[13:22:01] <Essobi> RikusW: How you been man? :D
[13:22:03] <RikusW> only PWM and USB
[13:22:18] <RikusW> hi Essobi, been good, how about you ?
[13:22:25] <Essobi> RikusW: doing okay... Keepin that cattle inline?
[13:22:28] <Essobi> :D
[13:22:29] <Tom_itx> funny, i searched the site for 4414 and the 8515 popped up
[13:22:37] <RikusW> sort of :-P
[13:22:48] <Essobi> RikusW: Ever make that geophone?
[13:23:00] <Tom_itx> the 8515 must have replaced it
[13:23:13] <RikusW> Essobi: busy on a SIMCom SIM900 and GlobalTop PA6H PCB
[13:23:25] <RikusW> (GSM modem + GPS)
[13:23:28] <Essobi> RikusW: neat.
[13:23:44] <Tom_itx> abcminiuser, do you think there's any way you could get me a data sheet for it?
[13:23:45] <kobsu> abcminiuser: btw. where did you put the code you made in your thesis? it was said to be open source?:)
[13:23:47] <Tom_itx> the 4414?
[13:23:53] <RikusW> the geophone is too short range to be practical :(
[13:24:29] <abcminiuser> kobsu, it's on google code and my website
[13:24:36] <abcminiuser> Find it through my website, that's easier :)
[13:24:37] <RikusW> Essobi: I did use a m644 for the geofencing, but seems I can load it straigth onto the SIM900
[13:24:42] <abcminiuser> Tom_itx, IT BELONGS IN A MUSEUM
[13:24:49] <abcminiuser> Also email me and I'll see what I can do
[13:24:51] <Tom_itx> ok
[13:24:53] <Essobi> RikusW: Oh oh.. /me googles.
[13:25:18] <Tom_itx> i found one for the 8515 and it says it is backward compatible with a few new features
[13:25:20] <RikusW> I did have to ask really nicely to get the ARM C compiler and docs to do it though
[13:25:22] <Essobi> RikusW: wait... using a 644 for fencing? How?
[13:25:26] <kobsu> ok found it, explorer bot?
[13:25:27] <Tom_itx> seems atmel has been doing that a long long time
[13:25:33] <RikusW> and I have no customer support on that...
[13:25:37] <abcminiuser> kobsu, ja
[13:25:43] <Essobi> RikusW: Oh.. I see.. GSM huh?
[13:26:02] <RikusW> Essobi: geofencing, to use GPS to see it you're inside or outside an area
[13:26:12] <Essobi> RikusW: that's pretty neat..
[13:26:33] <Essobi> RikusW: I see what you're doing... GPS.. check lat,long && sms if outside..
[13:26:45] <RikusW> or GPRS
[13:26:50] <Essobi> Or that..
[13:27:23] <RikusW> turned out lat,long calculations isn't so difficult after all :)
[13:27:53] <RikusW> my algo use a triangle fan
[13:28:12] <RikusW> it can do quite complex shapes
[13:28:24] <RikusW> even triangle overlaps :)
[13:29:00] <RikusW> the number to triangles tested must be odd, then you're inside
[13:30:11] <RikusW> Essobi: so the new design is GPS serial port straight onto the GPRS serial port
[13:30:20] <RikusW> the GPRS got a debug serial port too :)
[13:50:01] * RikusW is actually online using a SIM900 EVB right now
[13:52:34] * RikusW pushes TheBigRedButton
[13:53:20] <TheBigRedButton> RIKUSW YOU THINK YOU'RE MAN ENOUGH?
[13:53:22] <TheBigRedButton> DO YOU?
[13:53:25] <TheBigRedButton> I SAID DO YOU?!?!
[13:55:33] <RikusW> man enough for what ?
[14:26:54] <OndraSter> for the big red button, Rious
[14:26:55] <OndraSter> oh
[14:26:56] <OndraSter> he is gone
[14:26:58] <OndraSter> sorry Rious
[14:27:18] <megal0maniac> OndraSter: Apparently he wasn't :P
[14:28:02] <OndraSter> :P
[15:06:35] <Malinuss> abcminiuser, eh if you got antoher min. : so if I change the measuring to the band gap, while the VCC is the referance, I should in theory (according to the datasheet) only have to wait max. 125usec... Now if I wait less then ~5000+ usec, the value simply swings very much (25%+).. if I wait a couple of msec though, all seems fine..
[15:08:39] <abcminiuser> Malinuss, how quickly are you switching? The ADC is pipelined
[15:08:53] <abcminiuser> So it takes something like 15 (slow) ADC clock cycles for the first conversion to be ready
[15:09:55] <Malinuss> abcminiuser, well the functions has a loop that waits for ADIF in ADCSRA to clear before even giving me the value... ?
[15:11:14] <abcminiuser> Hrm then it should be fine
[15:11:25] <abcminiuser> Perhaps the bandgap takes time to stabalize
[15:11:32] <abcminiuser> On some devices it's activated on demand
[15:13:31] <Malinuss> yeah it seems to me like changing from "normal" channels to the gap or from the gap to channel actually takes longer the otherwise...
[15:18:18] <abcminiuser> I'm off to bed
[15:19:14] <abcminiuser> Night all, talk tomorrow
[15:21:12] <Malinuss> night
[15:45:41] <creep> which is better, a "force sensing" capacitive, or a hall senor joystick :?
[15:48:37] <creep> *sensor
[16:13:45] <Tom_itx> hall
[16:23:15] <Essobi> sup
[17:38:21] <creep> Tom_itx<< i'm in doubt because they say new fighter jets have force sensing joysticks, they do not have to be moved, just forced
[17:39:25] <creep> though the logitech review page i read recently was lame, and it stated force sensing is better because it has no deadzone... a good hall joystick has 0 deadzone.
[17:40:16] <creep> but force sensing is definitely simpler
[17:44:57] <Malinuss> creep, so are those: http://www.ebay.com/itm/Analogue-Thumb-Joystick-for-Arduino-PIC-Electronics-Project-/130787977428?pt=UK_Computing_Other_Computing_Networking&hash=item1e739228d4 hall sensors or force sensing?
[17:47:25] <creep> Malinuss<< what is with that crap ?
[17:48:40] <Malinuss> haha is that your auto response?
[17:51:00] <Tom_itx> creep, i was comparing to ones on wheel chairs
[17:51:03] <Tom_itx> they're hall
[17:51:53] <creep> Malinuss<< could be haha
[17:52:24] <creep> Tom_itx<< hmm, well force sensing can be integrated if you have a shaking hand
[17:52:44] <Tom_itx> force could be better i suppose
[17:52:54] <Tom_itx> i haven't seen any
[17:53:13] <creep> it is almost the same, but force sensing does not require you to move the joystick 60 deg for full scale
[17:53:25] <creep> so, it stays in place
[17:53:36] <Tom_itx> 3 sensors?
[17:53:39] <Tom_itx> 4?
[17:53:39] <creep> interesting to think about
[17:53:44] <creep> depends on solution
[17:53:55] <creep> i'd do it with 3 sensors and get R and theta
[17:54:19] <creep> 3 capacitors is all it needs
[17:54:23] <creep> very cheap
[17:54:26] <creep> ;/
[17:54:42] <creep> (or force sensors)
[17:55:25] <creep> http://www.androidinabox.com/mini-joystick-capacitive-touch-screen-joystick-playing-games.html
[17:55:49] <creep> this is a capacitive joystick for touchscreen, interesting
[17:55:51] <specing> Been google for a new computer to buy when: Hello parallel port mobo!
[17:55:56] <Tom_itx> i'd think force or hall would be better than capacitave
[17:56:29] <creep> by capacitive i mean force
[17:57:05] <creep> you have some dielectric, and a minimal deflection, so force creates capacity change
[17:57:29] <creep> *close to linear
[18:04:38] <creep> Tom_itx<< you have this doorknob shaped stuff that you try to move like a joystick, the resulting force will compress/expand the dielectric between the capacitor plates, so their capacity changes linearly, C = epsilon * Area / d
[18:05:21] <Tom_itx> is that how weight measurement sensors work?
[18:05:29] <creep> epsilon_0 = 8.854 * 10 ^ -12
[18:05:29] <Tom_itx> force sensors
[18:05:45] <Tom_itx> ie in scales
[18:05:48] <creep> well, there are resistive force sensors too
[18:06:02] <creep> and piezoelectric
[18:06:23] <creep> the resistive sensor is the simplest i think
[18:07:10] <creep> and possibly still far more stable and longer life than any sliding contact potentiometer
[18:07:57] <theBear> most scales i seen or know about are load cells... wheatstone bridge arranged with physical resistances on either side of a <===> shaped bit of metal as i recall, tho i did just wakeup
[18:08:27] <Tom_itx> theBear awakes!
[18:08:33] * Tom_itx feeds theBear quickly
[18:08:49] <theBear> kinda, sleep was less than successful, and yeah, might have to go and get me some bacons and eggs
[18:09:00] <creep> http://blogs.cio.com/al_sacco/touch_screen_blackberry_storm_ui_takes_cues_from_google_phone
[18:09:01] <creep> :)
[18:09:14] <Tom_itx> make yerself a good omlet
[18:09:55] <theBear> i suppose that's an option, been a while, and i do like a nice french omellete :) damn, that is hard to spell
[18:10:12] <Tom_itx> i used the short version
[18:10:13] <Tom_itx> :)
[18:10:57] <Tom_itx> or breakfast burritos
[18:11:52] <theBear> fwooo i dunno, i do believe in beer, it's what for breakfast/lunch/dinner, and i usually feel pretty robust in the mornings, but the idea of a breakfast burrito always sounds kinda heavy duty, heck, kinda burritos i like are heavy for dinner :)
[18:12:24] <theBear> then again, i suppose a big greasy plate of eggs and bacons and tomatoes and fried bread and toast and butter is a bit heavy too
[18:13:03] <theBear> i thought it was supposed to be hot today, but right now it seems just a little chilly to be riding to the shops
[18:13:06] <Tom_itx> with a side of pancakes
[18:13:38] <creep> digital scales use this http://green-library.org/2012/09/16/force-sensor-or-strain-sensor-instruments-2/
[18:16:35] <creep> as the conductive meander line is stretched, it increases resistance
[18:16:55] <theBear> yeah that, most sources seem to call them load cells
[18:17:15] <theBear> trivial to auto-calibrate within a large tolerance, and fairly accurate
[18:17:45] <creep> in fact an RTD can be made from a copper wire too, good up to 150C
[18:18:19] <creep> just wound up 100 ohms at 0C from it, and you have a cheap RTD
[18:20:34] <theBear> mmm, interesting
[18:24:55] <creep> hhaha
[18:25:00] <creep> theBear<< http://en.wikipedia.org/wiki/Electrical_resistivity_and_conductivity
[18:25:27] <creep> one could replace a platinum RTD with a piece of copper wire for low temperatures
[18:25:41] <creep> tempco is close
[18:33:39] <Horologium> atmel samples order should be here tomorrow.
[18:34:39] <specing> Holy****
[18:34:50] <specing> you managed to get samples from them?
[18:35:21] <Horologium> yup.
[18:35:25] <Horologium> 8 chips coming.
[18:35:29] <Horologium> 3 different kinds.
[18:35:33] <Tom_itx> CSS555 programmable 555 timer
[18:35:36] <specing> what $deity did you pray to?
[18:35:36] <Tom_itx> http://www.customsiliconsolutions.com/downloads/Revised%20Standard%20products/CSS555_App_Note1_Serial_Interface.pdf
[18:35:40] <Horologium> ordered sunday...should be here tomorrow.
[18:35:49] <Horologium> with their new samples system it's as simple as microchip.
[18:36:04] <Tom_itx> ordered from their website?
[18:36:14] <Horologium> as for my diety,,,the FSM...I sacrificed an entire pot of sketty and meatballs.
[18:36:17] <Horologium> Tom_itx, yes.
[18:36:22] <specing> Suddenly a surge of orders from #avr
[18:36:33] <Tom_itx> naw not em
[18:36:34] <Tom_itx> me
[18:36:38] * amee2woof prays to the furry goddess of cuteness overload :3
[18:36:38] <specing> Horologium: Haha, nice to meet a fellow FMSer
[18:36:50] <specing> FSMer* :)
[18:37:00] <Horologium> used to follow the IPU but could never find her.
[18:37:19] <creep> hm, so 43.975m of 0.1mm copper wire = 100 ohm at room temp
[18:47:30] <Horologium> one wonders if one could use an attiny to create something like a programmable 555 timer.
[18:50:15] <theBear> my best buddy is a official fsm err, priest analogue dude...
[18:52:00] <Horologium> I would be an atheist but that's a very non-prophet way of living.
[18:52:08] <theBear> hehe
[18:53:20] <creep> Horologium<< yes
[18:54:11] <creep> http://blog.customsiliconsolutions.com/uncategorized/an-advancement-in-micro-power-555-timers
[18:54:20] <Horologium> I need to combine a 555 and a couple of digital pots and see what I can come up with.
[18:54:36] <creep> for discharge you can make an active low/tristate pin
[18:54:37] <creep> ;>
[18:55:21] <Horologium> yeah.
[18:55:30] <Horologium> the 555 is still one of the greatest designs ever.
[18:55:47] <creep> what is your destiny?
[18:55:57] <Horologium> my destiny?
[18:56:00] <creep> to use an attiny ?
[18:56:01] <theBear> to wander the earth, miserable for eternity
[18:56:02] <creep> :)
[18:56:27] <Horologium> oh, no clue...the attiny as a 555 was just a passing thought.
[18:57:12] <creep> analog joystick interface ?
[18:57:26] <creep> why not wire as voltage divider an duse adc?
[18:58:01] <Horologium> have made analog joystick interface with an atmega8 and vusb.
[18:59:25] <creep> i have 2 atmega8 too for that :)
[19:00:34] <creep> genius and logitech and the noname joysticks should be ashamed
[19:01:06] <creep> they sell useless junk
[19:01:28] <Horologium> logitech used to be good junk.
[19:01:43] <creep> the $40 stuff is still crap
[19:01:58] <theBear> i don't mind their wireless keyboard/mouses... they got a brandname on them, and it aint ms
[19:05:29] <creep> why does a useful joystick need to be $200 ?
[19:07:35] <Horologium> because people will pay that for it.
[19:09:19] <Tom_itx> not this people
[19:10:59] <Posterdati> hi
[19:11:00] <tobbor> Hello Posterdati
[19:11:33] <Posterdati> please I need help with avr-gcc, assigning a uint16_t variable to a uint16_t function is not working!
[19:12:02] <Tom_itx> post your code somewhere
[19:12:16] <Tom_itx> pastebin
[19:13:35] <Posterdati> http://pastebin.com/0DZ3ibN4
[19:15:30] <Posterdati> Tom_itx: ok?
[19:17:23] <Tom_itx> stick around maybe somebody will help in a bit
[19:17:49] <Tom_itx> is that ardweenie code?
[19:18:26] <Posterdati> what?
[19:19:21] <Tom_itx> what error do you get?
[19:19:46] <Posterdati> nTemp = 0
[19:27:51] <rue_house> Posterdati, you just want to limit the value?
[19:28:07] <rue_house> I know why
[19:28:12] <rue_house> you have to RETRUN the value
[19:28:23] <rue_house> return nRetVal;
[19:28:40] <Posterdati> rue_house: doesn't work either
[19:29:29] <rue_house> Posterdati, you just want to limit the value?
[19:29:31] <Tom_itx> what about uint16_t &nMin etc?
[19:29:39] <Tom_itx> in the parameter list
[19:29:45] <Posterdati> rue_house: no
[19:29:57] <rue_house> you also have to prototype the funciton
[19:30:02] <rue_house> http://paste.debian.net/232103/
[19:30:05] <rue_house> try that code
[19:30:35] <Posterdati> rue_house: function is prototyped
[19:30:39] <rue_house> if you do not have 'return nRetVal;' the value will ALWAYS be 0
[19:30:48] <rue_house> ok
[19:31:09] <rue_house> the compiler should give you a warning that you didn't exit the function with a return statement
[19:31:14] <Posterdati> rue_house: I forgot to paste it, but the function has got the return :)
[19:31:21] <Horologium> must have a return value or the thing won't return anything...in fact, the compiler should throw a warning or even an error if you have a function with a return value and no return.
[19:31:36] <rue_house> Posterdati, want to give us a whole new paste?
[19:34:38] <Posterdati> rue_house: I changed the function, I put "return nRetVal" alone
[19:35:09] <rue_house> yes, lets see
[19:35:16] <rue_house> repost so we can see what you have
[19:35:29] <rue_house> because I think your still misunderstanding it
[19:35:38] <Tom_itx> i had this pinout all figured out and now i can't find my note
[19:35:44] <rue_house> I also suspect this is arduino code, so maybe #arduino
[19:36:23] <rue_house> but I will continue to entertail
[19:36:33] <Tom_itx> i suppose i'll have to do it all over
[19:37:08] <Posterdati> http://paste.debian.net/232104/
[19:37:29] <Malinuss> rue_house, which part of the code looks like arduino to you?
[19:37:53] <Tom_itx> the Serial.print stuff
[19:38:02] <Tom_itx> that or c++
[19:38:13] <Posterdati> yes it is an arduino project
[19:38:16] <rue_house> it has a loop() and no main()
[19:38:19] <Malinuss> ah of course
[19:38:26] <rue_house> real C has main()
[19:38:29] <Posterdati> an AC poer controller
[19:38:33] <Posterdati> an AC power controller
[19:38:45] <rue_house> (uint16_t) ((nMax + nMin) >> 1) + (uint16_t) (((float) ((nMax - nMin) >> 1)) * nT);
[19:39:01] <Malinuss> well I always make a "loop()" function anyway... but yeah the serial.print pretty much gives it away
[19:39:15] <rue_house> why are you shifting nMax + nMin?
[19:40:04] <rue_house> are you just trying to limit the value nT between two values?
[19:41:07] <Posterdati> not only, it's a clamp function
[19:41:30] <rue_house> so the answer is yes
[19:41:33] <rue_house> I have a macro for you
[19:41:49] <Posterdati> -1 corresponds to TIMER1 OCR1A bottom
[19:42:02] <Posterdati> 1 corresponds to TIMER1 OCR1A bottom
[19:42:19] <rue_house> #define Limit(v,l,h) ((v) > (h)) ? (h) : ((v) < (l)) ? (l) : (v)
[19:42:56] <rue_house> the return value will be v, or l or h if v is outside those limits
[19:44:00] <rue_house> I dont understand (uint16_t) ((nMax + nMin) >> 1) + (uint16_t) (((float) ((nMax - nMin) >> 1)) * nT);
[19:44:05] <rue_house> what thats trying to do
[19:45:09] <rue_house> but it looks like yes, the return value will always be between 1 and 0
[19:45:30] <rue_house> uint16_t is an integer value from 0 to 65535
[19:45:46] <rue_house> your taking the input value and making it between 1 and -1
[19:46:00] <rue_house> of which the result returnable can be 0 or 1
[19:46:31] <rue_house> your shifting the math result right, which divides, it, which excludes 1 from the possabilities
[19:46:36] <rue_house> so you would always get 0
[19:46:48] <rue_house> Posterdati, so WHAT are you TRYING to DO?
[19:47:48] <Posterdati> rue_house: I told you
[19:48:03] <Posterdati> it's a linear function
[19:48:27] <rue_house> no, you did not tell me what your trying to do, you said that 1 and -1 are placeholder values for the timer limits
[19:48:56] <Posterdati> to map [ -1, 1 ] set to [ OCR_MIN, OCR_MAX ] set
[19:49:07] <rue_house> Posterdati, if I said, what are the integer values of -0.49 to 0.49 what would the answer be?
[19:49:12] <rue_house> oh
[19:49:22] <rue_house> ok, now I know WHAT your trying to do
[19:49:38] <Posterdati> a PI controller
[19:50:13] <rue_house> to map [ -1, 1 ] set to [ OCR_MIN, OCR_MAX ] set <-- that is the answer I needed
[19:50:43] <rue_house> I'm working, give me a minute
[19:50:55] <Posterdati> for what?
[19:51:06] <rue_house> some code that does what you want
[19:51:09] <Tom_itx> rue's minutes are 3600 sec long
[19:51:21] <Posterdati> but I solved my problem
[19:51:31] <Tom_itx> what fixed it?
[19:51:34] <Tom_itx> the return?
[19:51:41] <Posterdati> return nRetVal alone
[19:51:46] <rue_house> Posterdati, you said it always returns zero
[19:51:51] <Posterdati> yes
[19:52:01] <Posterdati> because I returned the expression value
[19:52:16] <rue_house> you dont need help then, good!
[19:52:21] <Posterdati> tx
[19:52:30] <Tom_itx> rue_house i need help now
[19:52:38] <Tom_itx> where did i put that note for the pinout?
[19:52:50] <rue_house> pinout for what?
[19:53:31] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/temp/led/7segtop.jpg
[19:53:32] <Tom_itx> that
[19:55:39] <Tom_itx> i got 4 2n2222 to drive it too all in a bag so i must have figured it all out once
[19:55:46] <Tom_itx> and the bcd chip
[19:55:47] <rue_house> ooooo
[19:56:07] <rue_house> bcd?
[19:56:13] <rue_house> its 9 segment
[19:56:14] <Tom_itx> 7447
[19:56:28] <rue_house> but if you use an avr you can make it say so much more
[19:56:30] <Tom_itx> 7 are wired
[19:56:38] <rue_house> awe :(
[19:56:40] <rue_house> want code?
[19:56:41] <Tom_itx> not all 9
[19:56:52] <rue_house> m32 code
[19:56:54] <Tom_itx> i still need to drive it with the transistors don't i?
[19:56:59] <rue_house> na
[19:57:02] <rue_house> avr can do it
[19:57:16] <Tom_itx> i gotta figure out the segments all over again
[19:57:29] <Tom_itx> and it's got 4 anodes or cathodes too
[19:57:36] <Tom_itx> gotta figure that out again too
[19:59:12] <rue_house> http://paste.debian.net/232106/
[20:00:16] <rue_house> I put vfd drivers on mine, but its all the same
[20:00:25] <rue_house> clean? understandable?
[20:00:40] <rue_house> REALLY easy to change to use the bcd chip
[20:01:28] <Horologium> I use ULN2803 chips for drivers for LED displays.
[20:01:34] <rue_house> also really easy to add AbCdEFgHiJLnoP
[20:01:58] <rue_house> qrStu
[20:02:46] <rue_house> |[]
[20:03:02] <rue_house> -
[20:03:04] <rue_house> =
[20:10:01] <Tom_itx> bak
[20:10:07] <Tom_itx> had to get some ice cream
[21:22:06] <Grievre> So the datasheet says max current per io pin of 40 mA, I take that to mean at 5V I can easily use a 1k pullup on my i2c bus and not damage the AVRs at least?
[21:41:13] <Casper> Grievre: the current per pin is regardless of the voltage
[21:41:36] <Casper> and do the math on your 1k pullup... you'll realise that it's quite negligeable
[21:42:32] <Casper> also, the datasheet should make a mention on the 100mA limit per port and 200mA limit per VCC/GND pin... don't forget to include the internal of the avr on the port and pin limit
[21:45:51] <Grievre> Casper: Er, the current is regardless of the voltage? but the current pulled by a pullup resistor is proportional to the rail it's attached to isn't it?
[21:46:11] <Casper> yes, but not the current that the avr can handle
[21:46:49] <Grievre> right I never said that
[21:47:55] <Casper> Grievre: your initial question seemed to say so
[23:51:17] <rue_house> !time
[23:51:18] <tobbor> My watch says its 09:43PM Tue Feb 05 2013
[23:51:43] <rue_house> how can none of over 100 people not say anything for about 2 hours?
[23:51:55] <Essobi> tada!
[23:52:03] <Essobi> rue_house: how's it going?
[23:52:15] <rue_house> ita quiet, tooo quiet
[23:52:19] <Casper> rue_house: because we're all boring
[23:52:22] <Casper> or
[23:52:27] <Casper> that you bore us :D j/k
[23:52:54] <rue_house> nobody have a project?
[23:53:10] <Casper> yes
[23:53:20] <Casper> but lack time and money and parts
[23:53:21] <rue_house> you gonna tell us about it?
[23:53:39] <Casper> and some priority pushed it aside
[23:54:53] <rue_house> you can make up for the lack of time with money, you can make up for the lack of money with parts, and you can make up for hte lack of parts with time
[23:54:57] <rue_house> so whats the problem?
[23:55:21] <Casper> lack of sleep
[23:55:26] <Casper> I'm again almost an hour late
[23:55:47] <rue_house> well, pff
[23:55:47] <Casper> ohhhh I'll take those pills, will help me to sleep tonight...
[23:56:03] <rue_house> there is your problem, screw the sleep and you will have time for the project
[23:56:20] <Casper> ps3 move, the gladiator game is fun, but hard on the arm
[23:56:48] <rue_house> you stressed out your arm moving a ps3?
[23:58:49] <rue_house> every used a tiny85?
[23:59:08] <Casper> moving the ps3 stick things
[23:59:25] <Casper> swinging around can hurt your arm :D
[23:59:39] <rue_house> oh I was writing a http server, forgot about that