#avr | Logs for 2012-02-01

Back
[01:19:49] <Sgt_Lemming> sigh, f'ing landlord is apparently trying to dodge fixing the problem properly >_<
[01:24:48] <ziph> It's probably hard to keep up on the maintenance for their 7 maximally leveraged properties.
[02:23:00] <pepsi> dean
[03:25:40] <skorket> hey, probably a stupid question, but on the ATTiny13, 'dec r14' can expect to loop back to 255 if r14 had 0 in it, yes?
[03:27:26] <pepsi> yes
[03:32:37] <skorket> pepsi, thanks
[03:47:56] <GeorgeJ> skorket: That's how unsigned integers works. They overflows! It doesn't really matter what AVR you're using.
[03:48:20] <amee2k> signed integers pretty much overflow the same >_>
[03:48:36] <skorket> right...like I said, stupid question.
[03:49:03] <amee2k> modular arithmetic vs. saturation arithmetic
[03:49:36] <amee2k> saturation arithmetic is popular with the DSP folks iirc
[03:49:46] <skorket> 'saturation' meaning 0 or MAX is absorbing?
[03:50:26] <amee2k> yeah... any operation that would yield a result >MAX returns MAX, and any result <0 yields 0
[03:51:35] <amee2k> sounds straight forward, but is kinda tricky to use sometimes because things like associativity and distributivity don't hold anymore
[03:52:46] <amee2k> "20 - 100 + 123" is the expected 43 in modular arithmetic, but depending on order of evaluation it may yield 123 in saturation arithmetic
[03:53:21] <ziph> Normal intN_t's aren't a field either.
[03:53:28] <skorket> I mean, you get the same thing in floating point arithmetic
[03:54:17] <amee2k> ziph: hmm because integer division is wonky?
[03:54:55] <ziph> amee2k: No, because they're modulo a non-prime number.
[03:55:20] <amee2k> ooh right. forgot about that one >_>
[03:56:24] <ziph> So algebra isn't that much more broken with saturating arithmetic. :)
[03:56:54] <ziph> The main reason it gets used is because the error from an overflow is less than it would be otherwise.
[03:56:56] <amee2k> well, with saturation the order of evaluation is distinctly more important i'd say
[03:57:51] <amee2k> depends on what kind of error is more acceptable i suppose. for DSP stuff it makes sense because it "just" starts to clip somewhere
[03:59:20] <ziph> Having an intermediate value overflow and then getting a correct result at the end in normal modulo arithmetic is more of a special case, usually saturation will work better.
[03:59:37] <ziph> Particularly since the overflows that do happen tend to be just outside the range.
[04:00:53] <amee2k> hmm interresting
[04:01:18] <amee2k> never thought about it that way
[04:01:48] <edboogie2011> hey guys.. a little off topic.. I am using a Cortex ARM M3.. it says it has "In system programming" ISP, and IAP, In application programming. to flash a rom image to the chip.. what the hell is the difference?
[04:01:59] <edboogie2011> I know IAP is when you flash in user code..
[04:02:02] <edboogie2011> what is ISP?
[04:02:13] <ziph> amee2k: Think about audio mixing.
[04:02:30] <amee2k> yeah, thats what i thought about when i said it would just clip
[04:02:38] <ziph> amee2k: To ensure the resulting signal is in range you have to scale each by 1/N.
[04:02:58] <edboogie2011> I mean.. I know IAP is when you flash the chip using code, like a firmware update routine..
[04:03:02] <edboogie2011> what is ISP?
[04:03:05] <ziph> amee2k: Or you can just make sure most of the resulting PDF is in range.
[04:03:15] <amee2k> edboogie2011: ISP means "in-system programming"
[04:03:33] <edboogie2011> amee2k: yeah I know what it stands for.. what does ISP mean?
[04:03:42] <edboogie2011> like how is it different to IAP>?
[04:03:54] <amee2k> and generally entails being able to read and write all on-die memories like flash, eeprom, fuses and sometimes even RAM for debugging
[04:04:32] <amee2k> i have no idea what "IAP" is supposed to be because i can't find a likely expansion for the acronym
[04:04:43] <edboogie2011> IAP is in application programming
[04:05:06] <edboogie2011> thats when you flash memory using code during execution of your code..
[04:05:14] <raek> edboogie2011: I think the biggest difference is that ISP (in AVR circles) is a programming method and protocol specifically for AVRs
[04:05:26] <edboogie2011> raek: ??
[04:05:28] <amee2k> hmm so the equivalent of atmel's self-flashing capability?
[04:05:52] <edboogie2011> raek: I don't think it is AVR specific
[04:05:59] <edboogie2011> most uC have ISP
[04:06:06] <edboogie2011> some don't have IAP
[04:06:09] <raek> not the concept, but the implementation also called ISP
[04:06:20] <amee2k> imo the difference is not exactly the functionality, but the interface
[04:06:45] <amee2k> ISP means stick a programming adapter on the board and upload it and everything will just work
[04:07:20] <amee2k> for IAP / self-flashing you need to write and flash a bootloader first and arrange for a physical interface to make the data to be flashed on available
[04:07:30] <raek> for AVRs, you use ISP to flash a ROM image onto an AVR from a programmer (usually connected to a computer)
[04:08:03] <raek> if an AVR reprograms itself, it's not using ISP
[04:08:24] <edboogie2011> yeah I wrote a bootloader.. so for production deployment.. I use ISP to deploy my bootloader on all my hardware..
[04:08:38] <edboogie2011> then use the IAP features on my bootloader to load the default firmware
[04:08:40] <edboogie2011> right?
[04:08:50] <amee2k> depends
[04:09:10] <edboogie2011> well is that the route that you guys would take to deploy a commercial app?
[04:09:24] <amee2k> if you're a good coder and don't expect to be distributing bugfixes on a regular basis like some other manufacturers, then just adding an ISP header will be fine
[04:09:28] <edboogie2011> my bootloader cannot update itself.. that is my only concern
[04:10:03] <edboogie2011> I want to lock my hardware from being able to use ISP after deployment
[04:10:11] <edboogie2011> which my chip can do
[04:10:18] <amee2k> if your code sucks and you can't afford time to properly debug it you should arrange for something more convenient because you're essentially shipping a beta-quality product. thats what people normally do these days
[04:10:31] <amee2k> except that the anticipated bugfix updates rarely actually happen
[04:10:49] <edboogie2011> amee2k: so how would I do that if I ship a beta-quality product
[04:10:55] <edboogie2011> load a bootloader
[04:11:00] <edboogie2011> right?
[04:11:13] <edboogie2011> what would you recommend?
[04:11:29] <amee2k> disable the ISP interface of the chip and write a boot loader that requires some kind of authentication before it allows flashing
[04:11:40] <edboogie2011> yes I did that
[04:11:50] <amee2k> well, i would recommend only giving realistic bids so you can finish the product before deployment ;)
[04:11:50] <edboogie2011> authentication and encryption
[04:11:59] <edboogie2011> lol
[04:12:27] <edboogie2011> the code is good.. like I said.. my only concern is that the bootloader can't be updated.. I just couldn't get that code to work
[04:12:36] <amee2k> encryption is probably not so important if you have authentication. just make sure your field technicians reset the device properly and not leave it in unlocked state
[04:12:37] <edboogie2011> its tricky to write a self updating bootloader
[04:13:33] <amee2k> but the bootloader is relatively small compared to the main firmware and has a very straight forward task
[04:13:56] <edboogie2011> should a copy of the firmware be hidden away in a corner of the flash.. so that the customer can "reset to factory default"?
[04:14:01] <amee2k> chances are you can modify the code of an existing one too
[04:14:01] <edboogie2011> is this commonly done?
[04:14:37] <amee2k> since the bootloader can't overwrite itself, once you have a working bootloader the device becomes impossible to brick accidentially
[04:14:50] <edboogie2011> yeah right..
[04:14:53] <amee2k> even if an IAP flash update screws up, reset the device and try again
[04:15:01] <edboogie2011> right
[04:15:14] <edboogie2011> so a good bootloader will essentially make it brick-proof
[04:15:25] <amee2k> pretty much, yeah
[04:15:42] <edboogie2011> then how do some commercial apps get bricked by customers sometimes?
[04:15:44] <amee2k> at least what bricking from software is concerned
[04:15:59] <amee2k> stuff like pulling the xtal off the board will still disable it of course
[04:16:01] <edboogie2011> I don't buy or hack a lot of commercial goods.
[04:16:23] <edboogie2011> how do customers typically brick their goods?
[04:16:37] <Valen> most commercial stuff doesn't have bootloaders
[04:16:49] <edboogie2011> Valen: really?
[04:16:50] <Valen> they will use smaller chips and buy them pre-programmed
[04:16:53] <amee2k> commercial stuff usually either gets bricked because of hardware failure like broken solder joints. most types of casual failure are easy to fix actually
[04:16:53] <edboogie2011> wow.. i didn't know that
[04:16:56] <ziph> If you make your bootloader small enough you can just put two on there.
[04:17:12] <amee2k> or people try to hack into the firmware and accidentially kill the bootloader while doing so
[04:17:32] <ziph> Then a first stage bootloader can pick the first bootloaded with a valid CRC.
[04:17:40] <edboogie2011> ziph: yeah I know that trick.. the two bootloader trick.. so one can update the other.. the trick I programmed, was the bootloader that runs in RAM,, so it can update itself
[04:17:45] <edboogie2011> couldn't get the code to work though
[04:17:48] <Valen> I wonder when the arduino will use a usb avr
[04:18:01] <ziph> And the first stage is trivial enough that you'd never need to replace it.
[04:18:17] <amee2k> Valen: never. it would break compatibility with existing stock of narrow minds
[04:18:22] <ziph> edboogie2011: Does it brick the device if you lose power or reset for some other reason while it is running?
[04:19:02] <edboogie2011> ziph: no it shouldn't be able to brick.. since the bootloader can always flash a new firmware at anytime.. as long as the bootloader doesn't get corrupted
[04:19:19] <edboogie2011> ziph: is that what you are asking?
[04:19:27] <ziph> edboogie2011: What happens if it resets after the bootloaded is erased but before you put the good one on?
[04:20:02] <edboogie2011> well currently my bootloader can't update the bootloader..
[04:20:21] <edboogie2011> if it erases the firmware.. its okay.. you can always flash a firmware again
[04:20:39] <ziph> Ah, right, I was talking about with bootloader replacement.
[04:21:13] <edboogie2011> yeah I don't see a good reason to allow the bootloader to update itself.. if you think about it. its just too dangerous to allow customers to mess with the bootloader
[04:21:27] <edboogie2011> thats the only real way to brick my machine
[04:21:35] <edboogie2011> F up the bootloader..
[04:21:54] <ziph> edboogie2011: Yet another trick is to actually upload the bootloader just before you do the update.
[04:22:08] <edboogie2011> ??
[04:22:14] <edboogie2011> ziph: I don't understand
[04:22:16] <edboogie2011> can you explain
[04:22:43] <ziph> The code that parses the firmware and writes it to flash gets uploaded in to RAM and then jumped into.
[04:23:35] <edboogie2011> yeah i think I mentioned that suggestion.. the execute in RAM trick
[04:23:53] <ziph> The difference is you don't copy it from flash.
[04:24:02] <ziph> You copy it from the PC that is doing the upgrade.
[04:25:01] <edboogie2011> i have another problem.. my application has only one interface: ethernet web client.. no LEDs or LCDs.. problem is.. the way the firmware updates is this.. the bootloader looks for a firmware file in SD card, if it exists.. it flashes.. but the customer will never know when the flash is finished.. any suggestions?
[04:26:00] <edboogie2011> I guess I can make it so that when the flash is finished. the web client boots with a success page
[04:30:25] <amee2k> http://www.bosch-professional.com/de/de/ocs/werkzeuge/101300/26171/entfernungsmesser/glm-80/ hm..... /me ponders
[04:30:29] <amee2k> any thoughts?
[04:31:16] <edboogie2011> amee2k: what is that?
[04:31:20] <edboogie2011> a type of meter?
[04:31:24] <amee2k> laser range finder
[04:31:35] <ziph> Can you set it to kill?
[04:31:35] <edboogie2011> oh cool
[04:31:36] <amee2k> 40 meters long folding rule + digital water level
[04:31:52] <amee2k> err 80* meters
[04:31:56] <edboogie2011> is there a reason why you want a german one?
[04:32:07] <edboogie2011> there are no good american laser range finders?
[04:32:08] <amee2k> because i'm sitting in the middle of germany
[04:32:13] <edboogie2011> lol.. oh okay
[04:32:15] <edboogie2011> good reason'
[04:32:30] <amee2k> and bosch stuff is usually quite decent from my exp
[04:32:59] <edboogie2011> is that the same bosch, that is part of auto union? .. AKA audi?
[04:33:21] <amee2k> yeah, there is only one bosch that i know of
[04:33:33] <amee2k> my old polo had a bosch TBI system >_<
[04:33:40] <edboogie2011> it seems like most industrial goods companies in germany are in some way linked to the auto industry
[04:34:11] <amee2k> because auto is a big industry in germany
[04:34:16] <edboogie2011> i know
[04:34:36] <amee2k> VW group, merc and BMW are big shots in the european auto industry
[04:35:17] <amee2k> so i'd say it is kinda natural that getting contracts with them is a valuable opportunity
[04:36:26] <edboogie2011> don't you feel that germany is being reduced to just auto makers, or concentrating too much on auto,, when pre war, it made planes and so much cooler stuff
[04:37:02] <edboogie2011> like what happened to messerschmitt and heinkel// why do those firms not exist anymore
[04:37:24] <amee2k> well, imo they are concentrating too much on cars. doesn't mean that i can't like the byproducts that result like decent quality tools that are readily available :)
[04:38:27] <amee2k> edboogie2011: do you want my honest guess on that?
[04:38:53] <edboogie2011> yes
[04:39:24] <amee2k> has something to do with cars not having a certain image that rebuilding an extensive military-aligned industry would
[04:39:56] <amee2k> and a few generations back it was kind of important to avoid that image for a certain reason that everyone knows but noone wants to talk about
[04:40:39] <amee2k> if you get my drift...
[04:40:47] <edboogie2011> yes i do.. same thing goes on in japan
[04:41:05] <amee2k> :)
[04:43:15] <edboogie2011> one day germany will lose its iron grip on the auto industry.. only a matter of time
[04:43:52] <edboogie2011> just like how japan lost out with sony to the koreans.. maybe same thing will happen with hyundais
[04:43:54] <amee2k> one day the auto industry will lose grip, but noone wants to hear that either >_>
[04:44:13] <karlp> germany has an iron grip on the auto industry?
[04:44:15] <karlp> news to me.
[04:44:30] <edboogie2011> i wouldn't mind.. hyundai already makes a car that is very competitive to the S500 merc
[04:44:30] <amee2k> i'd say it is more the other way around, but still
[04:44:39] <amee2k> someone is groping...err gripping avidly here
[04:44:43] <edboogie2011> for half the pri ce
[04:45:28] * amee2k <3 E36 M3
[04:45:52] <edboogie2011> my friend just got an M3, insane for $70k. very good value
[04:45:55] <edboogie2011> for a car sooo fast
[04:46:05] <amee2k> the E90 one?
[04:46:11] <edboogie2011> let me check
[04:46:16] <edboogie2011> its 2008
[04:46:18] <edboogie2011> so latets one
[04:46:24] <amee2k> yeah, sounds like it
[04:46:43] <amee2k> i found it quite interresting, but tbh not very european >_>
[04:47:18] <amee2k> V8 says american style, all wheel drive says japan
[04:47:44] <amee2k> after using inline-6 and rear wheel drive for hmm 4 generations? 5?
[04:48:34] <edboogie2011> why is it not european?? how much more german could you get than a 4 seat coupe sporty but still luxurious
[04:48:43] <edboogie2011> wait.. its all wheel drive?
[04:49:59] <edboogie2011> actually maybe germany is losing its grip.. since the Caddilac CTS-V is faster than the M3
[04:50:14] <edboogie2011> american comeback would be great
[04:54:31] <amee2k> mmh, can't find a reference for it but AWD is definitely lister for current 3-series
[04:55:09] <edboogie2011> that doesn't surprise me. since thats what lambo gallardo, porsche turbo, and audi R8 are all going
[04:55:29] <amee2k> yeah
[04:56:03] <amee2k> either way changing from I6 to V8 wasn't so hot but at least they picked a good breed for the new engine
[04:56:32] * amee2k <3 high revving NA engines + well-balanced rear-wheel drive layout
[04:56:32] <edboogie2011> what is the benefit of using an inline engine as opposed to a V8?
[04:57:09] <edboogie2011> I know boxer engines have a lower center of gravity.. but I fail to see the benefit of having an inline engine
[04:57:44] <amee2k> they don't neccessarily have an advantage, but it is definitely a change in style imo
[04:58:14] <amee2k> actually the S65 is lighter than the previous generation S54 iirc
[04:59:02] <amee2k> ooh and another change that i find very asian-aligned is dropping the sequential gearbox in favour of semi-automatic twin clutch
[05:01:48] <amee2k> sequential gearboxes have a long history in racing because they combine high power handling capability and good shifting times in a relatively light and compact package at the expense of some ride quality
[05:03:33] <edboogie2011> yeah but double clutch systems have insanely fast gear shifts, no point in going backwards with that
[05:04:17] <amee2k> i know
[05:04:51] <amee2k> i don't doubt that they're effective systems
[05:05:01] <edboogie2011> are there any disadvantages to having a double clutch system? other than initial cost?
[05:05:12] <amee2k> its probably just a personal preference thing for me because i grew up on manual gearboxes and i like the comparative simplicity
[05:05:14] <edboogie2011> weight maybe?
[05:05:26] <amee2k> edboogie2011: from what i've seen definitely weight
[05:05:26] <edboogie2011> of course amee2k, manual gearbox is more fun
[05:05:38] <amee2k> although they're getting that under control too
[05:05:41] <edboogie2011> amee2k: how much weight.. about?
[05:05:52] <amee2k> but you're still effective driving around with two complete gearboxes side by side
[05:06:02] <Valen> inline 6 is a naturally balanced engine
[05:06:05] <amee2k> well, more inside of each other. the clutches are concentric, but still
[05:06:07] <Valen> smoooother
[05:06:18] <Valen> 's why jag use V12s
[05:06:35] <amee2k> definitely smoother than an I-4 >_>
[05:06:46] <Valen> my ass is smoother than an I4
[05:06:50] <edboogie2011> lol
[05:06:56] <edboogie2011> i don't doubt that
[05:07:03] <amee2k> I4 is smoother than a vespa :P
[05:07:20] <Valen> http://www.vapourforge.com/jake/general/hotness_touched.jpg inline 6
[05:07:30] <Valen> 4 liters worth ;->
[05:07:43] <amee2k> sweet
[05:08:04] <edboogie2011> Valen: what kind of car is that? we don't have those in usa..
[05:08:18] <edboogie2011> its actually pretty nice looking
[05:08:26] <Valen> ford falcon
[05:08:30] <Valen> AU CR6
[05:08:32] <Valen> bah
[05:08:34] <Valen> XR6
[05:08:36] <amee2k> a friend of mine has an E36 M3 with the 3.2l on the track ... <3
[05:08:40] <edboogie2011> cool
[05:08:41] <Valen> http://www.fordmods.com/ford-4l-and-6-cylinder-f1/lifter-replacement-on-au-falcon-howto-with-lots-of-pictures-t95385.html
[05:08:54] <Valen> is me replacing bits in the top of the engine
[05:09:37] <edboogie2011> amee2k: in america, especially in new york.. M3s are common as hell. i would think in germany too. M3s and 911s must be as common as seeing hot blondes on the road :P
[05:09:52] <amee2k> lol
[05:10:10] <amee2k> all the hot blondes around here look mildly slutty
[05:10:18] <amee2k> or a lot of them anyway
[05:10:33] <edboogie2011> yeah but there must be enough for everyone
[05:10:39] <amee2k> sportscars aren't too common in germany due to the ridiculous fuel prices
[05:10:57] <edboogie2011> amee2k: oh yeah... lol. I forgot about european fuel charges
[05:11:00] <edboogie2011> :(
[05:11:26] <amee2k> if you say "3 liter car" everyone here thinks "3l/100km" which is about 78mpg. not 3 liter engine displacement >_>
[05:12:00] <Valen> I get ~10L to the hundred
[05:12:08] <amee2k> nice valvetrain you got there :)
[05:12:51] <Valen> the later ones have more valves i think
[05:12:59] <Valen> mine is a 1999 model
[05:13:51] <Valen> 4000km from hitting 300,000 and its still purring along
[05:14:11] <Valen> I see them come up as ex taxi's with 800,000+k's with the origional engine
[05:14:33] <amee2k> http://ompldr.org/vY2tncA/101_3995b.jpeg << camshaft and valvetrain on my old '92 polo mk.2F... 1043cc VW AAU series engine
[05:15:01] <amee2k> during head gasket replacement at ~122k km the year before i had to give it away :'(
[05:15:02] <edboogie2011> yeah wait.. if gas is so expensive in europe.. how do taxis operate? do they cost waaaaay too much..
[05:15:18] <edboogie2011> sorry for being an ignorant americans
[05:15:27] <karlp> how do taxis in the states work?
[05:15:35] <karlp> different cities have vastly dfifferent prices for taxis
[05:15:35] <edboogie2011> gas is chepa\
[05:15:42] <karlp> it's clearly not related to petrol prices.
[05:15:42] <edboogie2011> cheap*
[05:15:49] <amee2k> i rarely ever go by taxi but from what i remember i pretty much always paid ~20EUR per trip around munich
[05:15:55] <amee2k> no matter how far i went lol
[05:16:00] <edboogie2011> lol
[05:16:01] <Valen> I want to get a LPG conversion for mine
[05:16:07] <Valen> its like 30% the price
[05:16:21] <amee2k> Valen: per l or kg, or per km?
[05:16:38] <Valen> roughly speaking all of the above
[05:17:04] <Valen> liquid injection LPG takes away no power and has about the same economy in terms of l/100
[05:17:10] <amee2k> the LPB/autogas prices on the gas station signs are like half that of gasoline, but i have no idea how far you get on it
[05:17:26] <amee2k> LPG*
[05:17:29] <Valen> generally a bit less than petrol with an injection system
[05:17:53] <Valen> the venturi systems are pretty craptacular and you get like 60-70% the range
[05:18:08] <Valen> but its like going from carburetter to injection all over again
[05:18:26] <amee2k> what very few people know is that LPG cars aren't allowed in parking garages or multi-story car parks here
[05:18:42] <Valen> lol
[05:18:46] <Valen> thats silly
[05:18:52] <Valen> if your tanks are that dangerous
[05:18:53] <amee2k> i know
[05:19:05] <amee2k> and its not like gasoline doesn't burn too
[05:19:43] <amee2k> welcome to a country where the car industry is a small government too
[05:20:13] <karlp> amee2k: is that where, germany?
[05:20:17] <karlp> no lpg in car parks?
[05:20:19] <amee2k> karlp: exactly
[05:20:24] * karlp chortles
[05:20:27] <amee2k> how did you guess that :P
[05:20:39] <karlp> something you said earlier about germany car industry
[05:21:08] <amee2k> actually the conversation started with a link to a Bosch laser range finder >_>
[05:21:25] <amee2k> that i'm considering buying and wanted to ask for comments on it
[05:22:06] <edboogie2011> lol.. yeah this conversation all started with a laser range finder
[05:22:19] <edboogie2011> talk about covnersational tangents
[05:22:21] <edboogie2011> ..
[05:23:22] <amee2k> >_>
[05:23:49] <amee2k> haahaa huge wtf
[05:23:58] <amee2k> Conrad is selling "blue" bosch stuff too now
[05:24:21] <amee2k> http://www.conrad.at/ce/de/product/821251/LASER-ENTFERNUNGSMESSER-GLM-150/1510012&ref=list
[05:40:50] <amee2k> mmh, what i don't like is trading off one feature for another that is not a subset of the other :/
[05:41:22] <amee2k> GLM80 has an angle sensor so it can measure indirect heights with only a single measurement because it knows the incination
[05:42:27] <amee2k> but that only works vertically, and it doesn't have the general trigonometric indirect measurement function anymore
[05:42:54] <amee2k> where you take one reading at a right angle to the target surface, and the second to the point you want to measure
[05:43:22] <amee2k> the latter is not a strict subset because it can't just measure heights but distances in arbitrary orientation
[05:44:01] <amee2k> GLM50 can do classic indirect measurements but that one doesn't have the angle sensor anymore :(
[05:48:24] <amee2k> e.g. the 50 could indirectly measure the horizontal length of a wall just as well as height
[07:01:13] <vectory> what keyword do i need to search for, to learn more about batteries loosing/dropping voltage over time?
[07:02:40] <Tom_itx> http://www.altenergy.org/renewables/battery-storage.html
[07:04:47] <vectory> ok
[07:05:09] <Tom_itx> http://www.mpoweruk.com/life.htm
[07:06:22] <Tom_itx> http://www.mpoweruk.com/storage.htm
[07:07:34] <vectory> first link didnt help much
[07:07:50] <Tom_itx> it was free
[07:08:01] <vectory> no offence
[07:08:11] <Tom_itx> i know
[07:08:58] <eruif> self discharge ?
[07:09:12] <vectory> nono
[07:09:23] <eruif> power curve ?
[07:09:59] <vectory> i mean, a 1,5 volt mignon will supply ~2000mAh, but will it have 1,5V up to the last 1,Ah?
[07:10:13] <Tom_itx> i doubt it
[07:10:15] <eruif> the mfr's datasheet will tell you that
[07:10:57] <vectory> yeah, but whats the keyword i need to be looking for? thought there was a common name for that, like voltage drop or sth
[07:11:08] <vectory> nvm then, ill look for ds
[07:11:14] <vectory> later x)
[07:11:49] <Tom_itx> those other 2 links may help some
[07:12:10] <RikusW> leakage ?
[07:12:36] <Tom_itx> power curve
[07:12:58] <Tom_itx> performance charachteristics
[07:13:11] <eruif> droopy thingy
[07:13:28] <Tom_itx> peukert curve
[07:13:47] <Tom_itx> The Peukert equation is a convenient way of characterising cell behaviour and of quantifying the capacity offset in mathematical terms
[07:14:17] <Tom_itx> This is an empirical formula which approximates how the available capacity of a battery changes according to the rate of discharge. C = I n T where "C" is the theoretical capacity of the battery expressed in amp hours, "I" is the current, "T" is time, and "n" is the Peukert Number, a constant for the given battery.
[07:14:28] <Tom_itx> but you didn't follow my links
[07:15:37] <eruif> that peukert thing appears to be lead-acid specific
[07:15:44] <Tom_itx> http://www.mpoweruk.com/performance.htm#life
[07:16:39] <Tom_itx> The nominal voltage of a galvanic cell is fixed by the electrochemical characteristics of the active chemicals used in the cell, the so called cell chemistry. The actual voltage appearing at the terminals at any particular time, as with any cell, depends on the load current and the internal impedance of the cell and this varies with, temperature, the state of charge and with the age of the cell.
[07:17:32] <vectory> Tom_itx: i ment too, but the text is way too long for me to be finished by now :)
[07:18:04] * Tom_itx pastes a table showing discharge rates of liion lead acid nicd himh zn/mnO2 batteries
[07:18:47] <Tom_itx> use that last link i pasted
[07:18:55] <vectory> kthx
[07:19:10] <vectory> will make my head hurt, tho
[07:19:36] <Tom_itx> you asked
[07:21:23] <eruif> paste an ascii art table!
[07:21:42] <Tom_itx> -----\_____
[07:23:05] <vectory> whats that, a shovel?
[07:23:26] <Tom_itx> a blast shield deflecting a ray gun blast
[07:23:51] <vectory> :'D
[07:23:57] <Studioeng> evening all
[07:24:10] <vectory> well, discharge rate seems to be what i was looking for, thx Tom
[07:25:14] <Tom_itx> temp has an effect too
[07:26:39] <Studioeng> ive got a couple of issues ... i don't even know where to start
[07:26:50] <Tom_itx> at the beginning
[07:27:00] <Tom_itx> easier to follow that way
[07:27:16] <Studioeng> i built a programmer using a t85, http://www.simpleavr.com/avr/vusbtiny. it worked perfectly
[07:27:39] * Tom_itx senses 'worked' is the operative word here
[07:27:47] <Studioeng> then I was working on creating a USB -> LCD controller, http://www.obdev.at/products/vusb/prjdetail.php?pid=28.
[07:28:00] <Studioeng> using a t2313
[07:28:19] <Studioeng> i wired it all up, made a quick test hex file and it all worked great
[07:28:33] <Studioeng> using the usbtiny t85 programmer ofc
[07:29:09] <Studioeng> so next step was to modify the usblcd source to work on the t2313 and get the correct pins etc
[07:29:26] <Studioeng> again, flashed perfectly, screen fired up and said 1.10... ready.
[07:29:30] <Studioeng> all good so far
[07:29:46] <Studioeng> then i started making up the vusb part to actually communicate with the usb (
[07:29:52] <Studioeng> (ofc)
[07:30:51] <Studioeng> spending a few hours and not getting any joy from the USB/drivers and decided to lift the known working t85, and rig it upto the vusb from the t2313 (the t2313 was not connected at all)
[07:31:07] <Studioeng> it worked, it detected the usbtiny and appeared in the hardware list
[07:31:19] <Studioeng> (sorry should of mentioned im using win7 at the start)
[07:31:42] <Studioeng> okay, so everything back to where it was before, tried to flash a different hex file to the t2313 ..... nothing
[07:31:47] <Studioeng> just random errors
[07:32:44] * Tom_itx leaves for work
[07:33:49] <Tom_itx> consider a usb avr and lufa
[07:34:16] <Tom_itx> bitbanged usb can be timing critical i believe. never actually used it
[07:35:24] <Tom_itx> http://www.fourwalledcubicle.com/LUFA.php
[07:37:31] <Studioeng> when i try using avrdude with the t85 programmer, windows makes the device detach sound, and then a moment later the device attach sound
[07:38:06] <Studioeng> so im wondering if my connecting it to another vusb to test, ive buggered it up ....
[07:38:08] <Studioeng> :/
[08:09:48] <amee2k> hahaa... this is hilarious
[08:10:37] <amee2k> the GLM150's datasheet specifies battery runtime of 5 hours, battery life 30k readings and the time it takes to get a reading as <0.5 seconds
[08:11:02] <amee2k> 5 hours / 0.5 seconds = 36k readings XD
[08:11:29] <amee2k> so, err. 30k readings if you turn it on then just hold down the button
[08:12:21] <vectory> ~5 hours
[08:12:29] <vectory> because fractions are ugly
[08:14:15] <vectory> with generous rounding -_-
[08:15:23] <amee2k> 10 minute coffee break after an hour of button holding and it fits again >_>
[08:16:17] <Studioeng> Aimee060608
[08:16:49] <vectory> if your 10 minute break extends to 50 mins, then yes :)
[08:17:19] <amee2k> vectory: 10 minute break every hour for 5 hours
[08:17:25] <vectory> oh ok
[08:18:43] <amee2k> what i don't like about the 80 is that it only says 1250mAh li-ion and not what battery type it needs
[08:18:52] <amee2k> and only micro-usb for charging
[08:20:34] <amee2k> smells like the pack is proprietary and/or non-replacable
[08:20:58] <HKCMD> set alarm with DS1307 ?
[08:21:37] <HKCMD> can we set like if(hr>=6) { portd1 =1} ?
[08:25:47] <amee2k> yep, non-replacable battery pack :X
[08:26:25] <amee2k> both the 50 and 150 use regular AAA cells, why not the 80 >_<
[08:50:39] <amee2k> OndraSter: made some pics of the shitty "trimmed" shunts in my 5$ DMMs
[08:50:48] <OndraSter> :)
[08:50:50] <amee2k> OndraSter: http://ompldr.org/vY2treg/CIMG8953b.jpeg http://ompldr.org/vY2tsMA/2012-02-01-153858.jpg http://ompldr.org/vY2tsMQ/2012-02-01-153838.jpg
[08:51:02] <amee2k> i think i promised showing some a couple days ago, no?
[08:51:14] <OndraSter> o_O
[08:51:21] <OndraSter> probably
[08:51:47] <HKCMD> > set alarm with DS1307 ? can we set like if(hr>=6) { portd1 =1} ?
[08:51:51] <amee2k> hmm someone was asking for a dirt cheap strategy to implement an over-current limiter with extremely low voltage drop iirc
[08:52:21] <amee2k> and i said the 5$ DMMs use just a piece of wire as shunt and "trim" it with a wire cutter
[08:53:00] <amee2k> anyway, here they are. have fun with an industry quality example of how to save money >_>
[09:00:44] <vectory> ಠ_ಠ
[09:01:27] <amee2k> is that edible?
[09:02:01] <vectory> huh
[09:02:14] <amee2k> KANNADA LETTER TTHA
[09:02:30] <vectory> http://www.youtube.com/watch?v=YQIMGV5vtd4&wadsworth=1
[09:02:38] <amee2k> i only see "0C/A0" in a box >_>
[09:03:08] <amee2k> haha awesome
[09:03:37] <amee2k> makes me wonder how they manage to maintain the formation that precisely
[09:03:52] <amee2k> unless they just have preprogrammed flight paths and absolute navigation
[09:04:12] <vectory> makes me wonder all kind of things
[09:05:40] <vectory> like what to use it for
[09:05:41] <amee2k> like if they can go pick up some pizza for you?
[09:05:42] <amee2k> >_>
[09:05:46] <vectory> and if it would work in wind
[09:06:20] <amee2k> apparently bees like quadcopters
[09:06:36] <amee2k> http://www.youtube.com/watch?v=S2NuDqKXExw&feature=related << skip to 1:40 or so for the interresting part
[09:09:16] <amee2k> http://www.youtube.com/watch?v=4kFpG_xfymg&feature=related << haahaa hilarious. it doesn't seem to get anywhere but it looks cool doing this
[09:13:35] <keenerd> amee2k: Re shunts, why is that so bad? At least they are trimmed.
[09:16:22] <vectory> because you want real fuses, i guess?
[09:17:41] <amee2k> the 10+A ranges are almost never fused
[09:18:14] <amee2k> keenerd: even with that kind if trimming i seriously doubt they can manage anything better than 10%
[09:20:06] <keenerd> From benchmarking my own cheap meters, they area bit better than 10%.
[09:20:30] <amee2k> how much better?
[09:21:36] <amee2k> http://spectrum.ieee.org/automaton/robotics/military-robots/could-domestic-surveillance-drones-spur-tougher-privacy-laws << lol, something for the paranoid androids to worry about
[09:21:58] <amee2k> that is why you should shut the blinds when you fap in the penthouse
[09:24:08] <keenerd> I think it was 2%-5%, but don't quote me on that. Did those tests years ago. But nothing as bad as 10%.
[09:25:45] <keenerd> The bigger problem is whether it can still sustain accuracy at 10 amps for long periods. Those pinch points lead to hot spots.
[09:26:22] <amee2k> i suspect it probably is the reason why these cheapos are only rated for 10 seconds at a time at full current
[09:26:31] <OndraSter> amee2k, seriously, where did you hide my NRF24L01 chips and the bunch of dead ethernet cards :(
[09:26:41] <amee2k> even if the dissipation suggested by the shunt drop would still be largely safe
[09:26:54] <amee2k> OndraSter: they tasted like chicken
[09:26:56] * amee2k runs
[09:27:03] <OndraSter> I WANT THE CHICKEN!
[09:27:10] <OndraSter> it's been over a week now
[09:27:13] <OndraSter> and I can't find it
[09:27:25] <OndraSter> if I tidy up the room, I am afraid I will start missing more things :P
[09:27:26] <amee2k> http://www.youtube.com/watch?v=nNbj2G3GmAo << hahaa this one is hilarious
[09:48:01] <vectory> amee2k: that looks rendered :/
[09:50:27] <amee2k> probably because it is lol
[09:50:42] <amee2k> http://www.youtube.com/watch?v=MD48SLxwezg&feature=related << this one look staged too, but at least not edited
[09:51:54] <vectory> i want my own marvin :3
[09:52:25] <amee2k> lol
[10:15:43] <buzzsaw> good morning
[10:16:06] <OndraSter> good good
[10:17:18] <buzzsaw> I am building a project that I want to be able to control remotely, does anyone know of an inexpensive wireless remote control?
[10:17:22] <OndraSter> http://1.bp.blogspot.com/-5nUaJR6ZLas/TyiRbiKy_0I/AAAAAAAAaj0/gzh-YEW-D4I/s1600/0aaaa3v.gif
[10:17:47] <buzzsaw> I kind of want a pre-made remote fancy looking package.
[10:18:53] <buzzsaw> I found a few remotes that look like the keyless entry for a car but not quite what I am looking for.
[10:20:13] <amee2k> http://www.youtube.com/watch?v=-K1GD1V6af0&feature=related XD
[10:26:42] <vectory> OndraSter: wtf, eating a spoonfull of cinamon?
[10:31:31] <amee2k> i want one of these NAOs
[10:31:48] <amee2k> but i wonder how much of the demo videos are actually dynamic behaviour
[10:32:04] <amee2k> and how much is just staged with preprogrammed movements
[11:32:26] <Tom_itx> abcminiuser you're welcome to stay
[11:32:37] <abcminiuser> Sorry, was eating dinner :P
[11:32:49] <abcminiuser> Also, wireless here likes to die every so often for fun
[11:33:05] <Tom_itx> every 250 sec or so
[11:33:16] <amee2k> thats what wireless usually does, yes
[11:37:09] <amee2k> hmm i've got a stack of old floppy drives i want to put on ebay for lulz
[11:37:26] <amee2k> should i put them up individually or the whole stack at once?
[11:37:38] <amee2k> its like 8 or 10 drives or so
[11:37:56] <Tom_itx> full height IBM?
[11:38:04] <Tom_itx> sell em by the pound
[11:38:11] <amee2k> generic 3.5" drives
[11:38:20] <karlp> "hard disk" drives :)
[11:38:32] <amee2k> "hard floppy"
[11:38:46] <amee2k> that have been collecting in the junk box over the years... they should work but i'm not gonna test them
[11:39:05] <amee2k> i'm having a "weeding out the junk box" week this week
[11:39:19] <Tom_itx> you will regret that
[11:39:41] <amee2k> already sorted out a bunch of old video cards and a pair of video capture cards too
[11:40:14] <amee2k> Tom_itx: i still have a floppy drive in my computer that works, and that i haven't needed in ages
[11:40:33] <Tom_itx> i still have a floppy drive in mine and use it
[11:40:47] <amee2k> and that i'll keep there but i have no idea what i'd possibly want to use over half a dozen of them
[11:40:52] <karlp> I left it too long in the last comptuer that had them.
[11:41:04] <karlp> went to try and archive all the floppies, too many of them had decayed.
[11:41:24] <amee2k> i don't have any significant floppy stock anymore either
[11:41:45] <karlp> I've had zero floppy stock now for probably 10 years
[11:41:54] <amee2k> a half-full box for bios updates and small time crap like that
[11:42:02] <karlp> I was trying to archive and store stuff at my dad's place after he died.
[11:42:29] <karlp> what do people still use them for?
[11:42:35] <karlp> old pcs connected to cnc mills and stuff?
[11:42:49] <Tom_itx> i still run dos 6.22 on a pc
[11:42:56] <karlp> the, "software is no longer being made, do NOT touch the working thing!"
[11:42:56] <amee2k> until a few years ago they were still common for bios updates and running diagnostics tools
[11:43:04] <Tom_itx> and use the floppy to transfer data to another pc
[11:43:32] <karlp> I'm more curious in the reasons for the existance of the pc :)
[11:43:44] <amee2k> but bios flash images and tools are getting too large for floppys these days, so people use cds for these now
[11:43:50] <karlp> clearly if it only had floppies, you use floppies.
[11:44:09] <amee2k> retro gaming machine?
[11:44:21] <amee2k> i've got an old P3 box for that >_>
[11:44:40] <amee2k> with all the 9x and most of the dos drivers
[11:45:58] <karlp> from trying to flog off some old stuff, I reckon you're likely to wait a while for someone who'll pay anything useful for them,
[11:46:07] <karlp> and youðre better off offering them on freecycle or something
[11:46:45] <amee2k> if noone buys them offering them for sale was for free
[11:47:21] <amee2k> and if someone does i'd just drive them to the post office instead of the recycling yard. so no real loss in trying imo
[11:48:07] <amee2k> i also want to put up one of these pretty cool ZIP drives that were popular like 10 years ago
[11:48:37] <amee2k> apparently quite a bunch of people still uses them. i found a zip750 in the original unopened package lol
[11:53:19] <amee2k> no idea how to ship it yet, but i've got the old board from my server too that i want to put up for sale
[11:54:16] <OndraSter> hmm
[11:54:19] <OndraSter> what board? :)
[11:54:24] <amee2k> mainboard
[11:54:42] <amee2k> by now its probably just old junk though
[11:54:57] <amee2k> conventional PCI... some old athlong and only a couple gigs of ram
[11:58:38] <OndraSter> when you said "old"
[11:58:43] <OndraSter> I was expecting Pentium I SMP system lol
[12:00:22] <amee2k> lol
[12:00:35] <amee2k> okay, not /that/ old >_>
[12:03:47] <OndraSter> guys, I am looking for ideas for mobile apps for WP7 marketplace :P
[12:03:56] <OndraSter> I've got like two or three
[12:04:37] <ziph> With that many ideas you'll be doubling the size of the marketplace. :)
[12:05:08] <OndraSter> pssh
[12:05:11] <OndraSter> I need five :D
[12:05:16] <OndraSter> to enter competition :P
[12:05:25] <ziph> I'm doing work on a Windows Mobile 6.5 application at the moment.
[12:05:31] <OndraSter> WinMo is dead
[12:05:39] <OndraSter> (I know, I <3 it)
[12:05:39] <ziph> Not in enterprise. :)
[12:05:49] <OndraSter> I thoguht that enterprise switched to iphones and androids by now
[12:05:57] <ziph> Not for actual stuff.
[12:06:15] <OndraSter> wow
[12:06:18] <ziph> Only the managers who need them to store pictures of themselves to masturbate to have switched.
[12:06:38] <dirty_d> I have a dilemma
[12:06:48] <dirty_d> and i seek guidance
[12:06:59] <ziph> There aren't any droppable Android devices with actual laser barcode scanners that I know of.
[12:07:12] <amee2k> you have no compulsive urge to masturbate uncontrollably?
[12:07:22] <ziph> Not at pictures of myself, no.
[12:07:36] <amee2k> not even photoshopped ones?
[12:07:58] <dirty_d> i have an interrupt triggered loop running at 400Hz that takes about 1ms to complete and i need to output 4 400Hz PWM signals
[12:08:13] <ziph> I suppose if I photographed my earlobe on to Elle I could manage it.
[12:08:30] <dirty_d> i can use timer0 and timer2 and live with 93 LSBs of resolution
[12:08:38] <dirty_d> or someone can suggest something better
[12:09:01] <OndraSter> hmm there are few OC registers per timer
[12:09:02] <amee2k> what IC?
[12:09:05] <OndraSter> or do you need them out of sync?
[12:09:18] <amee2k> i'd say two 8 bit timers with four channels should do fine
[12:09:22] <dirty_d> im using an atmega324a
[12:09:23] <OndraSter> ziph, but I need ideas for WP7 now, not any kinky thoughts! :P
[12:09:56] <dirty_d> i can use phase correct mode and double that cant it? if i clobber the counter and stuff at the beginning of each iteration of the main loop?
[12:09:57] <Tom_itx> OndraSter, card scanner ap for commerce
[12:10:10] <ziph> OndraSter: Rdio has a nice API, you could so some music thing.
[12:10:12] <OndraSter> Tom_itx, there are many card scanners...
[12:10:20] <OndraSter> music things -> Zune :)
[12:10:24] <amee2k> just for clarification, are you thinking software PWM now?
[12:10:25] <dirty_d> set the counter to 0 set the pin to high, so its already high on the first match, and on the downcount it will set the pin low?
[12:10:28] <Tom_itx> pocket size?
[12:10:29] <ziph> Isn't Zune dead?
[12:10:31] <dirty_d> amee2k, no
[12:11:02] <OndraSter> ziph, Zune is now built into WP7... with all its glory
[12:11:10] <OndraSter> plus there is bazilion of music apps already :P
[12:11:54] <dirty_d> amee2k, i have an imu that generates an interrupt at 400Hz which triggers the start of my loop
[12:11:55] <Tom_itx> wireless code grabber for garage doors
[12:12:02] <OndraSter> uh?
[12:12:13] <dirty_d> amee2k, it busy waits on a status change set by the interrupt
[12:12:45] <Tom_itx> well you didn't like my first suggestion :)
[12:13:07] <dirty_d> amee2k, at the beginning of the loop after the status has changed, im considerint setting the timer 1 and 2 counters which are in phase correct mode to 0, and setting the 4 output pins to high
[12:13:14] <amee2k> brb phone
[12:13:18] <dirty_d> amee2k, so that on the downcount on compare match it will set the pins to low
[12:13:22] <ziph> OndraSter: A port of spice. ;)
[12:13:33] <OndraSter> spice is the electronics designer right?
[12:13:40] <dirty_d> effectively giving me double the resolution
[12:13:44] <ziph> Sure you're in the right channel? ;)
[12:14:02] <OndraSter> was just making sure, we are talking phones with capacitive touchscreens :D
[12:14:03] <dirty_d> Tom_itx, i think i just ordered a usbtiny mkII from you
[12:29:50] <Tom_itx> dirty_d, nothing showed up here yet
[12:30:00] <Tom_itx> when did you order?
[12:30:10] <dirty_d> i thought it said shipped, let me check
[12:30:17] <Tom_itx> oh ok
[12:30:25] <Tom_itx> i sent that out already
[12:30:46] <Tom_itx> didn't associate a nick / name
[12:31:38] <dirty_d> ok yea shipped, ordered the 27th
[12:31:53] <dirty_d> do you sell a lot of those?
[12:31:58] <Tom_itx> i was gone or it would have shipped the 27th
[12:32:06] <Tom_itx> quite a few
[12:32:09] <dirty_d> cool
[12:32:28] <Tom_itx> i just got a board batch of the original ones back in
[12:32:39] <Tom_itx> gonna solder em up soon so they'll be available again
[12:33:01] <Tom_itx> you're in Ma
[12:33:52] <dirty_d> yup
[12:34:17] <Tom_itx> you should get it tomorrow or so i'd guess
[12:37:06] <dirty_d> yea, maybe even today
[12:37:12] <dirty_d> what process do you use to solder them?
[12:37:53] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/boards/reflow/cooking.JPG
[12:38:02] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/boards/reflow/cooling.jpg
[12:38:41] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/toaster_oven/toaster_oven_index.php
[12:39:15] <dirty_d> nice
[12:39:24] <dirty_d> have you ever tried the hot plate method?
[12:39:32] <Tom_itx> no
[12:39:39] <Tom_itx> i wanted more control of the heat profile
[12:39:56] <Tom_itx> i've desoldered with a clothes iron though
[12:42:11] <Kevin`> Tom_itx: what type of controller are you using there?
[12:42:42] <dirty_d> Tom_itx, a hot plate with a temperature controller wouldnt be as good?
[12:42:44] <Kevin`> does it do pwm for the heating elements?
[12:43:07] <Tom_itx> it's a mega 168 but a 48 would be plenty big enough
[12:43:15] <Tom_itx> no pwm needed really
[12:43:22] <Tom_itx> heating elements react rather slow
[12:43:29] <dirty_d> schmidt trigger?
[12:43:41] <Tom_itx> for what?
[12:43:53] <Tom_itx> triac for the elements
[12:44:03] <Tom_itx> opto coupled
[12:44:07] <Tom_itx> the schematic is there
[12:44:21] <dirty_d> i mean is it just like turn on elements of the temp is too low, and turn off if too high
[12:44:24] <dirty_d> with a deadzone
[12:44:43] <Tom_itx> yeah
[12:44:57] <Tom_itx> it turns off before it gets too high though
[12:44:57] <Kevin`> how are you attaching the temperature sensor such that it doesn't desolder itself?
[12:44:58] <dirty_d> and the avr is just for the temperature ramp?
[12:45:26] <Tom_itx> Kevin`, i used silver solder and the board part is siliconed to the outside of the oven
[12:46:33] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/toaster_oven/sensor3.jpg
[12:46:44] <Tom_itx> tape is just holding it until it sets
[12:48:20] <Kevin`> I have my own toaster oven that needs a mod like this
[12:48:43] <Tom_itx> you should do it
[12:49:19] <Tom_itx> the PT100 sensors are linear so for this app just 1.1v ref worked fine
[12:49:29] <Kevin`> where can I get those nice temperature sensors from? mouser doesn't have a "dm-508"
[12:49:32] <Tom_itx> the 48 etc have an internal 1.1v ref
[12:49:44] <Tom_itx> no, those came from element 14
[12:50:10] <Tom_itx> whatever it used to be called
[12:50:19] <Tom_itx> farnell
[12:50:58] <Tom_itx> i had inflex send me a couple
[12:51:03] <Tom_itx> it was cheaper that way
[12:51:30] <Tom_itx> he and I made them at the same time
[12:51:40] <Tom_itx> it's mostly his code too
[12:58:41] <Kevin`> what's the max temperature you use? is 300c enough?
[12:59:39] <Tom_itx> 235 iirc
[12:59:55] <Tom_itx> you'd need more for lead free
[13:00:11] <Tom_itx> we use that past from DX
[13:00:15] <Tom_itx> paste
[13:00:56] <Kevin`> i'm having a hard time finding a cheap temperature sensor that can take the high temperature. does inflex have any more of those left? :)
[13:01:46] <Tom_itx> http://www.newark.com/labfacility/dm-508/sensor-pt100-thin-film-2x5mm-cl/dp/24M1444?Ntt=DM-508
[13:02:32] <OndraSter> which solder paste from DX? There are many, Tom_itx :)
[13:03:23] <Kevin`> Tom_itx: I suspect shipping from UK to US would not be optimal
[13:03:40] <Tom_itx> no probably not that's why i got em from inflex
[13:03:53] <Tom_itx> OndraSter, http://www.dealextreme.com/p/lodestar-soldering-paste-50g-4711
[13:04:03] <eruif_> is there any brazing alloy that can be used at home ? over a gas burner e.g.
[13:04:04] <Tom_itx> free shipping
[13:04:24] <OndraSter> I looked on this one this morning
[13:04:28] <OndraSter> I suspect it is good? :)
[13:04:34] <Tom_itx> it comes on a sailboat from HK though
[13:04:39] <Kevin`> eruif_: I have an acetylene torch at home.. more specific? :D
[13:04:45] <OndraSter> I ordered from DX many times
[13:04:52] <eruif_> Kevin`: a cooking range
[13:05:02] <Kevin`> eruif_: for brazing steel?
[13:05:16] <eruif_> for joining thermocouple wires
[13:05:32] <eruif_> to be used in a soldering oven
[13:05:46] <Kevin`> eruif_: weld it with a capacitor
[13:06:03] <Tom_itx> eruif_, i used silver solder
[13:06:04] <Tom_itx> hasn't come off yet
[13:06:21] <OndraSter> Tom_itx, and do you also have some syringe for it? Couldn't find any usable on DX
[13:06:29] <eruif_> Tom_itx: shouldn't it have melted ?
[13:06:39] <Tom_itx> the solder is on the outside of the oven
[13:06:56] <eruif_> how did you join the hot end of the thermocouple ?
[13:06:58] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/toaster_oven/sensor3.jpg
[13:07:09] <Tom_itx> mine is a PT100 sensor
[13:07:11] <eruif_> or did you buy one
[13:07:13] <Tom_itx> not a thermocouple
[13:07:25] <eruif_> I have to join the hot end. so I don't want to use solder.
[13:07:28] <amee2k> so, back
[13:07:56] <amee2k> and too late for the pwm party lol
[13:08:00] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/toaster_oven/PT100_a.jpg
[13:08:19] <dirty_d> amee2k, its never too late
[13:08:28] <amee2k> also, wtf kind of backbirth package is DRQFN?? >_<
[13:08:44] <Tom_itx> probably a thin qfn
[13:08:51] <eruif_> Durr Retarded Quad Flat Noleads.
[13:08:51] <Tom_itx> or something to do with the height of it
[13:09:01] <amee2k> looks like normal QFN but with staggered pads
[13:09:13] <Tom_itx> well there you go
[13:09:17] <Kevin`> eruif_: you can get solders with a rather high melting point, but i'm not sure what the most efficient source would be. maybe somewhere for jewelry
[13:09:31] <amee2k> i.e. a second row of pads further inwards like a ghetto BGA kind of thing
[13:09:47] <eruif_> Kevin`: good idea. I'll go drop in on a jewelry shop.
[13:10:10] <amee2k> looks pretty yuck to handle manually
[13:10:11] <eruif_> Dual Row QFN
[13:10:11] <Tom_itx> the pt100 works rather well and goes straight to a adc
[13:10:34] <eruif_> I've already bought a MAX6675 :(
[13:11:02] <Kevin`> eruif_: use it if you want. it will certainly work
[13:11:06] <amee2k> dirty_d: hehe. did you get your question answered while i was holding other people's hands on the phone?
[13:11:14] <dirty_d> amee2k, no lol
[13:11:29] <amee2k> what did i miss?
[13:11:38] <amee2k> something with PWM on megaX4 iirc
[13:11:38] <dirty_d> nothing i think, i stopped talking when you left
[13:11:45] <dirty_d> atmega324
[13:11:58] <dirty_d> oh 4 signals yea
[13:12:14] <amee2k> right. four 8-bit PWM channels on timers 0 and 2, but i don't remember the question
[13:12:34] <dirty_d> if i use fast pwm mode, my output only has a resolution of 93
[13:12:44] <dirty_d> because its a servo signal, varies from 1 - 2ms
[13:12:45] <eruif_> freenode's servers are burping a lot today
[13:13:04] <dirty_d> with my clock the closest to 400Hz i cna get is 366.something
[13:13:24] <dirty_d> so im left with 93 LSBs to vary the pulse length between 1 and 2 ms
[13:13:28] <amee2k> whats the MCU clock?
[13:13:32] <dirty_d> 12MHz
[13:14:02] <amee2k> yeah, 366Hz sounds about right with 1/128 prescaler
[13:14:05] <dirty_d> yup
[13:14:22] <dirty_d> but if i use phase correct mode i should be able to double that
[13:14:48] <amee2k> i don't think so
[13:15:05] <dirty_d> i cnat remember if you cna use both compare units to generate 2 outputs in pase correct mode
[13:15:16] <dirty_d> why?
[13:15:37] <amee2k> the only diff between phase correct and fast pwm is that the register settings are buffered and only updated every two PWM cycles at the start of the cycle
[13:15:57] <dirty_d> hmm? in phase correct it counts up to TOP then down to BOTTOM
[13:16:16] <dirty_d> on compare match going up it clears the output, on the way down sets it
[13:16:18] <amee2k> so you can't get weird output that would screw with a synchronous motor drive when your code changes the timer registers in the middle of a PWM cycle
[13:16:32] <dirty_d> but i have a plan
[13:16:37] <amee2k> yes, but the total resolution doesn't change. you've still got only 8 bit
[13:16:59] <dirty_d> its really 9 bits
[13:17:03] <dirty_d> in that mdoe
[13:17:05] <amee2k> every other cycle is mirrored so the on-periods are back to back and you effectively get half the frequency
[13:17:13] <dirty_d> right
[13:17:15] <dirty_d> but
[13:17:32] <dirty_d> if at the start of my 400Hz loop i set the counter to 0, and the output pin to high
[13:17:37] <amee2k> i don't see where the 9th bit comes from
[13:17:43] <dirty_d> then the compare match on upcount has no effect
[13:17:52] <dirty_d> and the downcount match will set the pin to low
[13:17:54] <amee2k> hmm i see
[13:18:10] <dirty_d> amee2k, well its not really a 9th bit, but it distinguishes between upcounting and down
[13:18:20] <dirty_d> so its really like counting from 0 to 512
[13:18:23] <dirty_d> 511
[13:18:28] <dirty_d> or whatever
[13:18:35] <amee2k> that could work i think
[13:18:43] <dirty_d> yea i dont see why not
[13:18:48] <amee2k> you'd only get duty cycles <50% this way though
[13:18:57] <dirty_d> i know
[13:18:58] <Kevin`> no 16bit timers on his chip?
[13:19:06] <dirty_d> but a servo pulse is always a minimum of 1ms
[13:19:13] <amee2k> Kevin`: megaX4 but he needs 4 channels
[13:19:21] <amee2k> and potential issues at settings close to zero or close to 50%
[13:19:24] <dirty_d> amee2k, wait no, over 50%
[13:19:45] <dirty_d> i also need to use the 16 bit timer as a rtc sorta deal
[13:19:58] <amee2k> over or under 50% depends on whether you force the pin high or low during the cycle where the timer is disabled
[13:20:05] <Kevin`> servo control can be done in semi-software with 8 (or is it 10?) channels per timer
[13:20:10] <dirty_d> i can only vary the last half of the signal anyway, so it works out
[13:20:13] <Kevin`> in a rather simple way
[13:20:34] <Kevin`> since your duty cycle is never over 1/8 (1/10?) with a standard servo
[13:20:42] <dirty_d> Kevin`, i know but i dont have the time to do it, i have a lot of processing to do that would interfere with any software pwm
[13:21:02] <dirty_d> over 1ms of work at 400Hz
[13:21:20] <amee2k> hmm i think you could do the disable timer trick with fast pwm as well
[13:21:28] <dirty_d> this isnt standard servo gignal btw
[13:21:35] <dirty_d> its the same 1-2ms pulse but at 400Hz instead of 50
[13:22:09] <dirty_d> i got 5 interrupts to deal with too
[13:22:26] <dirty_d> one i2c int, and 4 from an rc reciever
[13:22:56] <Kevin`> you only need 4 channels, right? i'm suprised you can't do that with 2 16-bit timers
[13:22:58] <dirty_d> that all makes software pwm impossible, unless i used a second avr
[13:23:11] <dirty_d> Kevin`, therse only 1 16 bit counter
[13:23:22] <Kevin`> also you have interrupt priority levels, so you COULD do it in software ;p
[13:23:50] <dirty_d> Kevin`, but i have at least 1ms of calculations i need to do at 400Hz
[13:24:04] <Kevin`> dirty_d: toggling a pin isn't going to take much away from that
[13:25:01] <dirty_d> hmm
[13:25:12] <Kevin`> this is one of those fancy helicopter things, isn't it? :)
[13:25:15] <dirty_d> yes, lol
[13:25:35] <dirty_d> i might be able to do that
[13:25:38] <Kevin`> I want to make one of those, with tilting wings
[13:25:48] <dirty_d> this is just 4 motors and a brain
[13:26:17] <dirty_d> 4 1500W motors
[13:26:45] <dirty_d> hopefully 20 pound lift capacity
[13:26:56] <dirty_d> if i cna find some 18" props
[13:27:11] <amee2k> thats a pretty stiff heli
[13:27:14] <dirty_d> yea
[13:27:23] <dirty_d> if its worth doing, its worth overdoing
[13:27:28] <amee2k> be careful not to lose your head >_>
[13:27:53] <amee2k> thats how real men do their projects *nod*
[13:29:09] <dirty_d> Kevin`, id need to have an interrupt at like 102kHz to even match the hardware 8-bit timers though
[13:29:15] <dirty_d> i dont think i have the cpu time left to do that
[13:29:52] <Kevin`> dirty_d: the timer can generate interrupts
[13:29:57] <dirty_d> Kevin`, i know
[13:30:18] <Kevin`> 102khz is NOT a lot of cpu time
[13:30:36] <Kevin`> especially when you aren't doing floating point math in it
[13:30:43] <amee2k> are you sure?
[13:30:48] <Tom_itx> bbl
[13:31:01] <dirty_d> yea but thats not even enough
[13:31:15] <amee2k> the ISR for my servo tracking thing fires at ~200kHz and accounts for over 40% of the total CPU load
[13:31:49] <amee2k> thats why i want to rewrite it in assembly sometime, i think i can halve the execution time with manual optimization
[13:32:06] <Kevin`> dirty_d: oh, you don't fire an interrupt continuously, you fire it when you next need to flip a pin
[13:32:10] <dirty_d> if im going to do software pwm then id at least want 1024LSBs of resolution which would require and interrupt at 409kHz
[13:32:30] <Kevin`> and update the timer for the next interval
[13:32:58] <dirty_d> thats what the hardware pwm already does
[13:33:43] <amee2k> ooh... i just found a HIP6021 on an old mainboard 0.0
[13:33:54] <Kevin`> the hardware pwm will keep running without having to update the time once per cycle
[13:34:10] <dirty_d> yea
[13:34:37] <amee2k> three error amps for linear regs and one synchronous buck controller, all with external power transistors, in one IC
[13:35:43] <Kevin`> lots of pc mainboards have something similar to that
[13:37:20] <amee2k> these buggers are dirt cheap too
[13:37:37] <amee2k> 2.72EUR for a HIP6012
[13:39:07] <_abc_> Can anyone explain in 10 words or fewer how much overhead a timer interrupt has on atmega?
[13:39:24] <_abc_> Assuming I run an atmega at 20MHz and I program it in C?
[13:39:50] <amee2k> you mean, in general? or compared to a non-timer interrupt
[13:40:09] <_abc_> I mean, in general. As in, usec, nsec, instructions, whatever
[13:40:12] <eruif_> I think he wants a cycle count
[13:40:31] <_abc_> An approximate number (I am not going to want nsec precision)
[13:40:52] <_abc_> Yeah a cycle count would be nice
[13:41:11] <_abc_> Most instructions are 1 cycle on avr, no? Exceptig jumps and calls and rets which are 2?
[13:42:05] <dirty_d> i dunno, if i just use phase correct ill get about 186 LSBs of resolution for my four PWM signals
[13:42:10] <dirty_d> that should be enough
[13:42:13] <_abc_> Well I guess I will have to dig out a listing and calculate
[13:42:37] <_abc_> dirty_d: that was not to me, right?
[13:42:45] <dirty_d> _abc_, i think its 4 cycles to start the interrupt
[13:42:51] <dirty_d> _abc_, no not the previous one
[13:42:53] <amee2k> hmm i don't know about the time between timer flag is set and the ISR execution starts, but i found writing ISRs in C has quite a bit of overhead due to all the pushing and popping going on
[13:43:10] <_abc_> dirty_d: hmm 4 cycles made up of a push and a call. Makes sense.
[13:43:21] <dirty_d> _abc_, no i think its 4 cycles before that even happens
[13:43:22] <_abc_> dirty_d: And max jitter is likely 2 cycles?
[13:43:42] <_abc_> hmm then max jitter is more like 4 cycles?
[13:43:46] <dirty_d> like 4 cycles until it has already jumped to the address of your handler
[13:43:50] <_abc_> Damn byzantine datasheets
[13:43:54] <Kevin`> why would there be jitter?
[13:44:05] <dirty_d> it might not be exactly 4
[13:44:14] <_abc_> Kevin`: because an interrupt coming in the middle of a 2 or more cycle instruction cannot be executed immediately
[13:44:46] <amee2k> if your code has stretches with interrupts disabled, jitter can be more or less arbitrary
[13:44:49] <dirty_d> _abc_, you can make the whole thing a lot faster if you can set aside some registers that the rest of the program promises not to use
[13:44:53] <_abc_> Normally you have an irq asserted and the cpu will service it in the next m1 cycle.
[13:45:00] <dirty_d> i dunno how to do that but i read about it
[13:45:06] <_abc_> dirty_d: Oh I will do that for sure
[13:45:23] <amee2k> also, i don't think the push for the return address is part of the overhead of the interrupt itself
[13:45:30] <_abc_> amee2k: sure it is
[13:46:12] <_abc_> amee2k: an interrupt works like this: assert(irq); while(!(machine_cycle_1)); push(pc); jmp(isr);
[13:46:45] <_abc_> amee2k: the assert is instant, the while waits untilt the cpu is in an interruptible state, then push takes 2 cycles and jmp another 2
[13:46:50] <amee2k> check out how an interrupt vector table looks in asembly, it is a sequence of RJMP instructions
[13:46:58] <_abc_> yes!
[13:47:05] <_abc_> but the rjmp only occurs after the push
[13:47:18] <amee2k> hmm
[13:47:29] <eruif_> if you don't push pc how will you know where to return to ?
[13:47:30] <dirty_d> _abc_, what are you making?
[13:47:52] <_abc_> dirty_d: nothing right now but I have something in mind which will max out the cpu and I need to know these details.
[13:47:53] <amee2k> okay, i think you're right then
[13:48:02] <dirty_d> ahh
[13:48:12] <amee2k> but if the push is implicit, why does the table have RJMP instructions then?
[13:48:16] <_abc_> amee2k: if the core is not optimized then the lookup of the ivt takes 2 more cycles
[13:48:29] <amee2k> wouldn't it be easier to just store the address of the ISR?
[13:49:02] <amee2k> and make the jump implicit
[13:49:02] <_abc_> amee2k: assert... while... push(pc); temp=ivt[vector_address]; jmp(pc);
[13:49:15] <_abc_> sure, but it just execs there
[13:49:23] <eruif_> that would require the location to be hardcoded
[13:50:04] <amee2k> no, i mean like other architectures only have a list of ISR addresses as IVT instead of jump instructions to save IVT space
[13:50:24] <_abc_> amee2k: well yes but they don't really really save space
[13:50:33] <_abc_> amee2k: it's a simplification by complication
[13:50:51] <_abc_> I think all risc cpus execute at the vector location (no double indirection)
[13:51:01] <_abc_> pic certainly does so
[13:51:02] <amee2k> if i wanted to simplify it like that, i would not automatically push the return address and use RCALL instructions instead
[13:51:24] <amee2k> this way single instruction ISRs would stay in the IVT
[13:51:32] <eruif_> how long does an RCALL take ?
[13:51:36] <_abc_> it is no big deal to code for it
[13:51:40] <amee2k> 3 cycles
[13:51:44] <_abc_> eruif_: 2 cycles?
[13:51:46] <eruif_> :)
[13:51:47] <_abc_> oh 3?
[13:51:55] <eruif_> because RCALL has to push PC
[13:51:59] <amee2k> RJMP is 2, RCALL is 3 due to the push
[13:52:07] <_abc_> ah so 3 not 4?
[13:52:16] <_abc_> and return is 2 right?
[13:52:20] <eruif_> you always have to push PC
[13:52:24] <eruif_> there is no escape
[13:52:46] <_abc_> eruif_: you could have several pcs and swap between them. 2 are enough for starters...
[13:52:46] <dirty_d> what if you saved the pc in a register
[13:52:54] <amee2k> i mean with RJMP even if i only wanted to do say an INC on some variable i'd have to do the RJMP, then the INC and then a RETI
[13:53:05] <_abc_> eruif_: cdp1802 did that (1976 or so...)
[13:53:11] <amee2k> thats like 6 extra cycles
[13:53:13] <eruif_> ARM does something like that
[13:53:20] <_abc_> yes it is common
[13:53:24] <_abc_> Sparc too I think
[13:53:53] <_abc_> Sparc has a whole bunch of registers assigned per task and switches register blocks when it switches tasks
[13:54:06] <_abc_> And Alpha also did that I think
[13:54:16] <amee2k> mmh switching register banks is awesome
[13:54:19] <_abc_> You can switch context in 1 clock cycle with those cpus
[13:54:23] <amee2k> yep
[13:54:31] <_abc_> (and that will be a 4GHz clock cycle)
[13:54:50] <_abc_> Well they are dead now. RIP.
[13:54:51] <amee2k> they made alphas that run up to 4GHz? 0.0
[13:54:57] <_abc_> x86 crapola über alles
[13:55:15] <_abc_> amee2k: I think they got to 2GHz and they had faster ones in the pipeline when they stopped.
[13:55:42] <eruif_> nobody is developing SPARC anymore ?
[13:55:53] <_abc_> http://shopper.cnet.com/processors/dec-alpha-21264-1/4014-3086_9-31298776.html
[13:56:01] <amee2k> spark kind of died even before sun did
[13:56:12] <_abc_> http://insidehpc.com/2011/12/27/chinese-sunway-bluelight-super-is-dec-alpha-one-decade-removed/ haha not dead
[13:56:38] <_abc_> amee2k: check out the price $6,189.95
[13:56:39] <_abc_> ...
[13:56:42] <amee2k> from WP, EV8 was supposed to do 2GHz in 2003 but apparently never happened
[13:56:47] <_abc_> yes
[13:56:47] <Essobi_> I've got a 32 core sparc in my basement.
[13:56:53] <_abc_> eruif_: hehe
[13:57:13] <amee2k> i've got a sun ultra60 with the 400MHz cpu
[13:57:24] <amee2k> only 4 of the 16 memory slots are popualted though :(
[13:57:37] <amee2k> and only one of the two CPU slots
[13:58:08] <eruif_> that's what happens when you get your computer from the dumpster
[13:58:12] <amee2k> _abc_: wtf is that bluelight thing??
[13:58:18] <_abc_> amee2k: what?
[13:58:25] <amee2k> your link
[13:58:42] <_abc_> amee2k: Chinese Alpha knockoff
[13:58:56] <amee2k> it looks very ... chinese >_<
[13:59:07] <_abc_> Yeah but it is used in a supercomputer :)
[13:59:11] <_abc_> By the bucket
[13:59:17] <amee2k> lol
[13:59:34] <amee2k> i want to make a supercomputer from a huge stack of Z80s
[13:59:36] <eruif_> "The Bluelight super has 8,704 SW1600 processors, which are an offshoot of RISC processors designed by the Chinese military for its own use, and presumably licensed from Digital shortly before or after it was eaten by Compaq in 1998."
[13:59:43] <_abc_> 141 GFLOPs is very decent amee2k
[14:00:19] <_abc_> 68GB/s memory bandwidth
[14:00:30] <amee2k> nice
[14:00:47] <amee2k> i bet i could beat it with 5000000x Z80s!
[14:01:01] <_abc_> amee2k: make no mistake , that cpu has sixty-four cores. 10 times more than any x86 has on chip now.
[14:01:17] <amee2k> okay 50000000x Z80s then >_>
[14:01:42] <_abc_> 5000000 Z80s would require 5000000*3W = 15,000,000 W or 15MW
[14:01:52] <_abc_> Do you own a power plant amee2k ?
[14:01:53] <amee2k> sounds fine to me >_>
[14:01:55] <OndraSter> :D
[14:02:00] <eruif_> "the power requirements of the new Chinese supercomputer were relatively modest � about one megawatt"
[14:02:01] <OndraSter> I want small nuclear power plant in my basement
[14:02:13] <amee2k> _abc_: i'd make one with 5000 diesel generators to match
[14:02:44] <amee2k> distributed designs are the new shitz
[14:04:25] <_abc_> amee2k: Anyway apparently HP still makes/owns the Alphas and they support them for their servers.
[14:04:45] <_abc_> http://shopper.cnet.com/processors/dec-alpha-21264-1/4014-3086_9-31298776.html $6100 and some change, includes Unix license for CPU
[14:05:19] <OndraSter> Tom_itx, where to get syringe for the soldering paste?
[14:17:55] <dirty_d> ebay
[14:18:54] <OndraSter> "paste syringe"?
[14:19:29] <specing> $6000 for a 1.25 Ghz CPU? Not a chance
[14:19:37] <dirty_d> whats that?
[14:20:01] <dirty_d> oh
[14:20:02] <dirty_d> ouch
[14:22:08] <OndraSter> I have got here some syringe for refilling, it has cut off the tip
[15:08:03] <OndraSter> Hi,
[15:08:04] <OndraSter> We have submitted your design to the factory. We will contact you by email if there are any issues regarding the design that may stop production.
[15:08:07] <OndraSter> Regards
[15:08:09] <OndraSter> ITead Studio
[15:08:11] <OndraSter> <3
[15:08:13] <OndraSter> many hours ago
[15:10:10] <amee2k> nice :)
[15:26:31] <vectory> OndraSter: in czech?
[15:27:04] <OndraSter> vectory, what should I czech?
[15:27:18] <vectory> the factory
[15:27:30] <OndraSter> no
[15:27:32] <OndraSter> iteadstudio
[15:27:42] <OndraSter> czech = 20 times higher prices almost :(
[15:28:43] <amee2k> <not in china> is always 20 times the price
[15:29:45] <amee2k> can't beat a horde of chinese kinds with sharpies
[15:29:58] <amee2k> well, technically you can, but that wasn't the point >_<
[15:31:25] <vectory> OndraSter: czeched olimex in bulgaria?
[15:31:30] <vectory> vulgaria xD
[15:31:31] <OndraSter> ye
[15:31:35] <vectory> k
[15:49:53] <Tom_itx> OndraSter, i don't use a syringe
[15:50:02] <Tom_itx> i have stencils made
[15:50:03] <OndraSter> oh
[15:50:17] <OndraSter> that's better indeed
[15:50:46] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/boards/USBTiny_Mkii/Stencils/stencil_form4.jpg
[15:53:01] <OndraSter> yay
[15:53:04] <OndraSter> sexy
[15:53:39] <Tom_itx> i did the plastic on my cnc
[15:54:25] <OndraSter> is that kepton?
[15:54:32] <Tom_itx> yes
[15:54:33] <Tom_itx> kapton
[15:54:41] <Tom_itx> good stuff
[15:54:50] <OndraSter> "on my cnc" = your homemade cnc?
[15:54:58] <Tom_itx> sherline
[15:55:01] <OndraSter> oh
[15:55:07] <OndraSter> I almost said "may I marry you?"
[15:55:19] <Tom_itx> http://www.ohararp.com/Stencils.html
[15:55:22] <Tom_itx> does the stencils
[15:55:39] <amee2k> wheeee free 2.1 sound system ftw!
[15:59:14] <amee2k> got it for free because some crackhat ripped out one of the satellite speaker cords lol
[16:11:10] <Valen> Tom_itx: have you tried cutting your own stencils?
[16:12:16] <Tom_itx> no
[16:12:40] <Tom_itx> well i tried once but gave up
[16:12:53] <Tom_itx> inflex does with a vinal cutter
[16:52:24] <Jan-> hihi :)
[16:53:10] <asteve> hihihihihihi
[16:53:14] <asteve> :(
[16:53:21] <Jan-> uhoh asteve
[16:53:23] <Jan-> what's wrong?
[16:53:27] <Jan-> was anothersteve mean to you?
[16:56:02] <OndraSter> probably bsteve
[16:56:05] <OndraSter> or thesteve
[16:56:19] <stevieB> there can only be one steve
[16:56:29] <stevieB> he happens to have many name change abilities
[16:56:29] <Jan-> The only stevie I know is a girl.
[16:56:48] <OndraSter> lol
[16:56:48] <OndraSter> I know Stevie Leigh from Twitter :P
[16:56:58] <stevieB> you clearly don't hang out with 70's R&B musicians
[16:57:00] <OndraSter> and then we have Stewie from Family Guy :P
[16:57:03] <OndraSter> no
[16:57:33] <OndraLappy> if I listen, it is either dnb, dubstep, trance, house...
[16:57:58] <asteve> all of the best stuff in the world
[16:58:44] * Jan- is currently listening to Zoe Poledouris, which nobody will have heard of
[16:58:48] <OndraLappy> even "dance" music
[16:58:49] <OndraLappy> nope
[16:58:54] <OndraLappy> never heard of have we
[16:58:58] <Jan-> seen "Starship Troopers"?
[16:59:02] <OndraLappy> "Yoda has spoken"
[16:59:04] <OndraLappy> maybe
[16:59:05] <OndraLappy> maybe not
[16:59:06] <Jan-> "All is well, but I have not been to paradise..."
[16:59:14] <OndraLappy> hmm nope
[16:59:18] <OndraLappy> doesn't ring any bell
[16:59:25] <asteve> i like that song
[16:59:45] <Jan-> asteve: have you heard the full track, or just what's used in the movie?
[17:00:23] <Jan-> Here: http://www.youtube.com/watch?v=-YnVMPfjweU
[17:01:19] <asteve> how intensely 80s in the beginning
[17:02:11] <Jan-> I can't help but think
[17:02:21] <Jan-> the 80s may be as high tech as we ever get :/
[17:02:39] <asteve> the sliding power riff guitar really ruins the song
[17:02:44] <asteve> the starship troopers version is better
[17:02:45] <Jan-> I think the movie used all the best parts
[17:03:16] <Jan-> I think that is the version they used, tho
[17:03:53] <Jan-> hang on
[17:03:55] * Jan- cues up
[17:04:31] <Jan-> hmm maybe not
[17:05:00] <Jan-> maybe this one: http://www.youtube.com/watch?v=E3kb8mlmlXY
[17:05:41] <asteve> ya, that's a better mix
[17:05:55] <Jan-> it's a fricken wierd mix
[17:06:19] <Jan-> it's kinda hilarious, that movie
[17:06:23] <Jan-> and that song works so well
[17:06:39] <Jan-> "Everything's great, everything's cool, someone just got his guts blown out, everything's fine..."
[17:23:30] <pepsi> hello
[17:23:35] <OndraLappy> ello
[17:23:56] <pepsi> who's lookin for work in the Midwest, USA?
[17:24:12] <OndraLappy> I am looking for work in middle europe :/
[17:24:22] <pepsi> move here
[17:24:26] <OndraLappy> I wish
[17:24:30] <OndraLappy> what kind of job?
[17:24:34] <pepsi> firmware
[17:25:33] * Jan- tugs on pepsi's ringpull
[17:26:01] * Jan- would like a job doing firmware, if she knew how :/
[17:26:03] <pepsi> PIC18 asm for now, moving to a new platform.. ARM? your pick? dunno.. we need C eventually
[17:26:32] <Jan-> C
[17:26:33] <Jan-> aargh
[17:26:41] <Jan-> is it perverse of me to be more scared of C than assembly
[17:26:46] <pepsi> yes
[17:26:46] <OndraLappy> Jan-: "she"? :P
[17:26:50] <OndraLappy> nope Jan-
[17:27:00] <Jan-> OndraLappy: The impersonal feminine pronoun, yes
[17:27:07] <_abc_> What? An unlicensed, illegal, mpaa violation?
[17:27:09] <OndraLappy> how many women are here? :P
[17:27:16] <Jan-> how the hell should I know
[17:27:23] <Tom_itx> pepsi wtf!
[17:27:25] <OndraLappy> that was question for the people here
[17:27:29] <_abc_> OndraLappy: About 3 billion...
[17:27:41] <pepsi> Tom_itx, yeah that never happened, did it?
[17:27:44] <_abc_> OndraLappy: but you will not mate in this life since you are a geek
[17:27:49] <OndraLappy> :(
[17:27:52] <_abc_> :)
[17:27:54] <OndraLappy> I know one nerdy and geeky chick though
[17:27:57] <OndraLappy> from Hungary
[17:28:14] <Jan-> I'm not sure if I'm both a nerd AND a geek
[17:28:15] <_abc_> She must be booked full until age 125
[17:28:17] <Jan-> what's the difference?
[17:28:23] <Tom_itx> i don't wanna move to chicago or program pics or i'd put my name in the hat
[17:28:34] <_abc_> ha
[17:29:13] <_abc_> Chicago is a good place for electronics now? I was under the impression it is really fsckd up now economically.
[17:30:03] <Jan-> Hey, at least the housing'll be cheap
[17:30:15] <_abc_> Yeah and you get 10,000 free housemates.
[17:30:22] <_abc_> They'll be crawling all over you
[17:30:42] * _abc_ really leaves now
[17:33:06] <pepsi> chicago's not any more fucked than the rest of the US
[17:33:48] <pepsi> there aren't many electronics jobs, but there seem to be even fewer electronics professionals
[17:38:59] <OndraLappy> gn
[17:53:08] <Jan-> I'm not sure how you become an electronics professional
[17:53:16] <Jan-> other than a related degree, what piece of paper do you need?
[17:56:15] <pepsi> a brain
[17:56:45] <spybert> In some countries, you need a professional license
[17:56:57] <pepsi> no...
[17:57:10] <pepsi> you need a license to be an engineer :P
[17:57:10] <Jan-> what, in case you over cook someone's microwave macaroni?
[17:57:31] <spybert> Jan-: In case you design a faulty medical device and somebody dies
[17:57:42] <Jan-> well sheesh, I'd expect extra paperwork for THAT
[17:57:51] <Jan-> but perhaps not for less critical stuff?
[17:58:14] <spybert> Well, most electrical power work requires a license even in the US
[17:59:16] <spybert> There was a big push to get all EEs licensed but industry kind of killed off the idea when they found out it would cost them a lotof money
[17:59:17] <Jan-> here too
[17:59:25] <ftc> so i am looking at a interrupt in a tutorial that goes ISR(TIMER1_COMPA_vect) but the datasheet seems to only specify "TIMER1 COMPA" is that just a standard transformation for creating a ISR vector or am I missing something? I am curious because if I want to create a ISR for something else like "TIMER1 OFV" would it be "TIMER1_OFV_vect"?
[17:59:48] <Jan-> so how do I become (theoretically) a licensed electronics engineer?
[18:00:03] <Tom_itx> pay a university some money
[18:00:11] <Jan-> oh dear :/
[18:00:22] <Jan-> but I assume it's more than a degree
[18:00:31] <spybert> You need (in the US) to pass the EIT test, then work under a licensed professional engineer for x number of years
[18:00:54] <Tom_itx> but how did he get his?
[18:01:55] <spybert> http://en.wikipedia.org/wiki/Fundamentals_of_Engineering_exam
[18:02:07] * Jan- is still trying to get her computer to say Tom_itx
[18:03:41] <spybert> http://www.pels.ca.gov/
[18:04:03] <spybert> most engineers in the US dont have a license. I never had one :))
[18:04:24] <spybert> it`s really only usefull for certain types of work
[18:04:32] <Jan-> I was gonna say
[18:04:36] <jacekowski> you don't need a license in most places
[18:04:36] <Jan-> safety critical stuff
[18:04:38] <jacekowski> Jan-: nope
[18:04:54] <jacekowski> nothing like that is legally required
[18:05:19] <jacekowski> it's between your client and you and your insurance company
[18:05:44] <Jan-> ain't that always the rub
[18:06:04] <Jan-> the only person I know who has professional insurance is a doctor
[18:06:14] <jacekowski> you have those all organisations like EIT or IET and other
[18:06:46] <jacekowski> Jan-: engineers are covered by their employers insurance
[18:07:03] <jacekowski> doctors are very often self employed
[18:07:25] <Jan-> well, she works for the NHS most of the time
[18:07:26] <Jan-> but she does jobs on the side
[18:07:44] <Jan-> she's really into motor racing, so she sometimes works at race days
[18:08:13] <Jan-> it wasn't that expensive because british people aren't very litigious :)
[18:08:16] <spybert> I`ve heard people say that the US is now getting worse than Europe as far as regulations and insurance requirements go
[18:08:42] <Jan-> the US can be quite red-tapey
[18:08:43] <jacekowski> spybert: in europe only insurance you need is third party insurance for your car
[18:08:47] <Jan-> unionised labor is very strong in the US
[18:08:50] <jacekowski> spybert: everything else is pretty much voluntary
[18:08:54] <Jan-> that can cause people a lot of trouble
[18:09:36] <spybert> jacekowski: Not here. Just about anything you do requires some form of liability insurance.
[18:09:50] <Jan-> plus the US does everything either twice or fifty-plus times, if you want to do certain activities in either one state while keeping the feds happy, or worse if you want to do it in any state
[18:09:55] <jacekowski> spybert: no legal requirement for that
[18:10:05] <jacekowski> spybert: it's your client that ussualy requires that
[18:10:15] <spybert> right, in many cases no legal requirement
[18:10:30] <Jan-> I was gonna say
[18:10:32] <Jan-> "requires"
[18:10:33] * Tom_itx throws an avr in the conversation just for the hell of it
[18:10:42] <Jan-> basically Americans really like to sue one another
[18:10:48] <Jan-> it's less common here
[18:10:48] <Jan-> by far
[18:11:14] <spybert> ebedded systems designers using AVRs can frequently run into trouble with FCC regs, UL requirements, etc
[18:11:31] <Jan-> sure but that's just compliance
[18:11:39] <spybert> Suing is the national pastime here :))
[18:11:40] <Jan-> thatdoesn't require you to be a licensed FCC compliance engineer, does it?
[18:11:52] <jacekowski> no
[18:12:04] <jacekowski> i'm doing SIL 3 stuff
[18:12:20] <Jan-> hey, we're watching Starship Troopers
[18:12:20] <jacekowski> and i've got no qualifications in relation to that
[18:12:27] <spybert> Jan-: But frequently you end up using the services of one
[18:12:35] <jacekowski> no formal qualifications
[18:12:35] <Jan-> and we just got to the super-massively jarheaded gung-ho bit where they assault the alien homeworld
[18:12:39] <Jan-> GO MOBILE INFANTRY!
[18:12:44] <spybert> hehe
[18:13:17] <spybert> oh man, the class of '45 was something else
[18:13:24] <karlp> spybert: what's the fcc problems you speak of?
[18:13:38] <Jan-> class of 45?
[18:13:52] <karlp> 2045!
[18:14:00] <karlp> are you watching starship troopers realllly?
[18:14:03] <karlp> or just saying you are?
[18:14:04] <Jan-> Yup!
[18:14:13] <spybert> karlp: Well, for one thing people dont realize that you arent supposed to sell stuff that isnt compliant. Usually though you get away with it
[18:14:25] <Jan-> "Hey johnny, wake up, fish boy! You're dead! Killed in action! Three more days and you're out, good as new. See you later, bye!"
[18:14:28] <Jan-> that's the bit we're at.
[18:14:34] <spybert> Jan-: People who fought in WWII
[18:14:43] <karlp> spybert: I think I just missed the bit where you said that designers that use avrs frequently run into trouble...
[18:15:10] <karlp> what makes designers using avrs run into trouble more than anyone else?
[18:15:43] <spybert> karlp: Actually, they are no less likely than anyone else to have problems, I was just mentioning it so people would realize it is on-topic :-)
[18:15:49] <Tom_itx> china does
[18:16:09] <karlp> sounded more like troll bait :) would have been appropriate without the "using avr"
[18:16:13] <ziph> AVR users tend to be hobbyists with no analog or RF knowledge.
[18:16:26] <Valen> fake CE tick stickers are cheap
[18:16:36] <Jan-> Mobile infantry body armor seems to be complete crap
[18:16:45] <ziph> That said, even the things most AVR hobbyists sell are excluded from FCC EMC requirements.
[18:16:53] <Jan-> it isn't even proof against insectoid stabbing weapons
[18:16:54] <spybert> ziph: It is used in industry, too. I`ve seen recruiting ads looking for EEs with AVR knowledge
[18:16:58] <Jan-> I hate to think how it'd work against bullets
[18:17:18] <karlp> what, you think atmel only makes money off hobbyists?!
[18:17:21] <Jan-> I thought AVR was mainly a hobbyist thing
[18:17:26] <karlp> and dataflash?
[18:17:35] <Jan-> don't they sell the IP for the cores to put on other people's chips too?
[18:17:40] <karlp> that would be arm.
[18:17:51] <karlp> two out of three letters :)
[18:17:56] <Jan-> I thought they did that with AVRs too
[18:17:58] <Jan-> hm whatever
[18:18:17] <karlp> not too my knowledge, but I could be wrong of course.
[18:18:28] <Jan-> what are they used in?
[18:18:29] <Jan-> all the usual stuff? DVD players?
[18:19:06] <Tom_itx> everything
[18:19:30] <karlp> avrs in ALL THE THINGS!
[18:19:41] <Jan-> multiple things of DOOM
[18:20:06] <Jan-> (by the way, with regard to the movie: TANKERRRRRRR!)
[18:20:12] <Tom_itx> it would be nice to think they have multiple facilities just for me
[18:20:41] <Jan-> I'm annoyed.
[18:20:50] <ziph> I didn't say it wasn't used in industry.
[18:20:54] <Jan-> I bought a whole RS-232 card and it's immediately been lost by other people.
[18:20:58] * Jan- kicks a can
[18:21:09] <Tom_itx> blame the blameless
[18:21:58] <ziph> Just that P(hobbyist | project uses AVR) > P(hobbyist | project uses e.g. renesys mcu)
[18:23:55] <Jan-> say that again
[18:24:10] <Jan-> this time in a modern language
[18:24:50] <district> it's more likely that hobbiests use avrs :P
[18:25:04] <Jan-> why d'you suppose that is
[18:25:14] <Tom_itx> free tools
[18:25:26] <Tom_itx> we're cheapasses
[18:25:26] <district> arduino, lots of publicity, breadboardable packages
[18:25:39] <district> well, the avr chips themselves are kind of expensive nowadays
[18:26:02] <Tom_itx> before that 68hc11 was
[18:26:34] <Jan-> AARGH
[18:26:34] <Jan-> how many fricken USB to RS232 devices are there?!
[18:26:39] * Jan- waves her arms
[18:27:12] <atmega> fricken ?
[18:27:28] <district> Jan-: most of them have the same internals
[18:28:25] <Tom_itx> ever since i installed as5 ie has been fucked
[18:28:34] <Tom_itx> studio doesn't play nice with it
[18:31:49] <dirty_d> have you guys tried the 32bit avrs?
[18:40:48] <district> nope but some of them are pretty nice
[18:50:29] <dirty_d> hmm yea, probably a lot more $ for the programmer for them
[18:51:44] <Tom_itx> jtag
[18:53:01] <dirty_d> have you used one?
[18:53:05] <Tom_itx> no
[18:54:53] <dirty_d> theyre pretty cheap too
[18:55:08] <dirty_d> i wonder if theyre any harder to use than 8bit avrs
[18:55:21] <Tom_itx> more complex
[18:55:23] <dirty_d> i looked for a tutorial on them, but found nothing
[18:55:38] <Tom_itx> join ##avr32
[18:55:55] <dirty_d> oh snap
[18:55:59] <Tom_itx> err #avr32
[18:56:36] <ftc> could someone quickly review my code and make sure i am doing this right? http://pastie.org/3299388
[18:56:58] <ftc> it just blinks an led but it is using hardware interrupts on timers which is a new concept for me
[18:57:52] <ftc> it appears to work but its hard to tell if it is actually functioning the way i think it is
[18:58:03] <Tom_itx> ftc, http://tom-itx.dyndns.org:81/~webpage/how_to/atmega168/mega168_led_blink_int_index.php
[18:59:22] <ftc> Tom_itx: thank you but i understand that part, i am curious about how i am handling the disable/enable of the interrupts to not cause a conflict by using msecondCount while it is being modified by the ISR
[18:59:39] <Jan-> the mobile infantry just won
[18:59:42] <Jan-> yaaaaaaaaay
[19:00:34] <Tom_itx> why not leave it running if all you're doing is updating?
[19:00:48] <Tom_itx> either way there's a chance to miss something
[19:01:42] <ftc> isn't there a possibility that the interrupt would occur in the middle of reading msecondCount causing unpredictable behavior?
[19:01:58] <Tom_itx> the instruction will complete
[19:02:14] * Jan- is a bit scared of interrupts
[19:02:19] <Tom_itx> what if the interrupt occurred during the copy?
[19:02:30] <Tom_itx> msecondCountCopy = msecondCount;
[19:02:48] <Tom_itx> interrupt is frantically waving it's hand that is tied
[19:03:10] <ftc> doesn't it run after the program hits "sei();"?
[19:03:18] <Tom_itx> yes
[19:03:32] <Tom_itx> but what if it needed service before that?
[19:04:16] <ftc> well in this case the interrupt should be triggered every millisecond and i can't imagine the copy operation takes even a fraction of a millisecond to complete right?
[19:04:16] <Jan-> is one absolutely required to service interrupts, or could you just design the logic of your program to work in a loop and check things
[19:04:23] <Tom_itx> there are at least 2 or 3 asm instructions in msecondCountCopy = msecondCount;
[19:06:46] <Tom_itx> and you don't know if it's gonna happen just before the interrupt wanted to fire or not
[19:07:18] <Tom_itx> test it
[19:07:32] <Tom_itx> see what happens if you comment out sei() and cli()
[19:07:36] <ftc> ok
[19:08:31] <ftc> it stopped blinking
[19:08:40] <Smashcat> Hi, I'm programming an ATTiny2313 using an UNO as a programmer. I'm using the Arduino IDE. I'm working with the ports directly, and have set up the timers etc ready to call an interrupt (timer1). However I don't know the syntax to specify the interrupt routine itself. I don't want to use the Arduino interrupt methods as they add too much overhead.
[19:09:09] <Tom_itx> Smashcat, what interrupt?
[19:09:32] <Jan-> do I have the option to just not sevice interrupts, and just check on the things I'm interested in periodically?
[19:09:32] <ftc> seems like it shouldn't just stop though
[19:09:54] <Smashcat> Tom_itx: I've set Timer1 up with CTC etc, and I just need it to call an interrupt handler once it fires.
[19:10:49] <ftc> smashcat: this document details timer interrupts at the bottom: http://www.avrfreaks.net/index.php?name=PNphpBB2&file=viewtopic&t=50106
[19:10:54] <Tom_itx> compare or overflow interrupt?
[19:11:06] <Smashcat> the problem is I haven't Tom_itx: Compare (CTC)
[19:11:24] <Tom_itx> #define TIMER1_COMPA_vect _VECTOR(4)
[19:11:24] <Tom_itx> #define SIG_OUTPUT_COMPARE1A _VECTOR(4)
[19:11:24] <Tom_itx> #define SIG_TIMER1_COMPA _VECTOR(4)
[19:11:36] <Tom_itx> TIMER1_COMPA_vect
[19:11:40] <Tom_itx> that's the one
[19:11:54] <Smashcat> Tom_itx: Ah! Thanks!
[19:12:14] <Tom_itx> if you include the device header that is
[19:12:35] <ftc> i should probably learn to use my jtag debugger and poke at it with that
[19:13:08] <Smashcat> Tom_itx: Yes, I just get an error that __vector_4 was not declared. Have to see where the header is.
[19:13:19] <Tom_itx> #include <avr/io.h>
[19:13:28] <Tom_itx> are you using avrgcc?
[19:13:44] <Tom_itx> then define the device in your makefile
[19:13:50] <Smashcat> Tom_itx: Yes, I'm just using the normal Ardiuno IDE
[19:13:53] <ftc> i think you also need <avr/interrupt.h>?
[19:14:45] <Tom_itx> i don't use the arduino crap
[19:15:02] <Tom_itx> you'll have to figure out where they put the device headers
[19:15:12] <Smashcat> Tom_itx: Well this is my first project outside of the Arduino. I also have WinAVR studio but haven't used it yet
[19:15:45] <Tom_itx> winavr or avrstudio?
[19:16:02] <Smashcat> Yeah, the io.h and interrupt.h includes made no difference. I have AVR Studio
[19:16:13] <Tom_itx> get winavr
[19:16:27] <dirty_d> is winavr gcc-avr for windows?
[19:16:32] <Tom_itx> and throw arduino in the river
[19:16:37] <Tom_itx> yes
[19:16:40] <dirty_d> ok
[19:16:57] <Smashcat> Well I got AVR Studio as I have a Dragon programmer/emulator too, and apparently it works well with that.
[19:17:39] <Tom_itx> that's fine but i'd still use avr gcc
[19:17:53] <Tom_itx> use studio to program the device if you want
[19:17:55] <Tom_itx> i do
[19:18:15] <Tom_itx> but i don't use it to write code or compile
[19:18:57] <pepsi> what do you use?
[19:19:06] <Tom_itx> pn
[19:19:38] <Tom_itx> it's bundled in winavr
[19:19:58] <Tom_itx> but you already knew that
[19:20:06] <pepsi> i did?
[19:20:15] <pepsi> oh ya
[19:20:19] <pepsi> i hated that thing
[19:20:22] <pepsi> i just used studio
[19:20:23] <pepsi> heh
[19:20:31] <Tom_itx> i'm impartial
[19:20:38] <Tom_itx> i just got used to it
[19:20:44] <pepsi> ya, whatever works
[19:20:53] <pepsi> MPLAB sucks a big fat one, but i use that everyday at work
[19:22:31] <Tom_itx> i gotta finish here or the irs may be up my ass
[19:22:34] <Tom_itx> later.
[19:25:25] <Smashcat> Tom_itx: All working now, thanks for the help - also thanks ftc - yes I needed to include the interrupts.h as well
[19:33:52] <Tom_itx> pepsi
[19:34:04] * Jan- tugs on pepsi's ringpull
[19:35:09] <Tom_itx> pepsi, take a look at this and tell me why cli() and sei() are needed during the var copy in the middle http://pastie.org/3299388
[19:37:02] <Tom_itx> inflex!
[19:47:07] <Jan-> inflex?
[19:47:14] <Jan-> oh! right, yes. Inflex.
[19:51:34] <molavy> hi
[19:51:57] <molavy> how can i install avr programmer on linux
[19:52:05] <karlp> sudo apt-get install avrdude
[19:52:12] * Jan- snickers
[19:52:42] <molavy> i use this , but it can't find my programmer typw
[19:53:01] <molavy> my programmer is hattel pocket programmer
[19:53:01] <dirty_d> what is it?
[19:53:35] <Tom_itx> molavy where did you get it?
[19:54:00] <karlp> and, did it come with any documentation or software of it's own?
[19:54:08] <molavy> mahtab bazar , saddi st ,mashhad city
[19:54:12] <Tom_itx> that's one i've never heard of
[19:54:17] <Tom_itx> go figure
[19:54:36] <Tom_itx> maybe you should ask them
[19:54:39] <Jan-> Iran
[19:54:45] <Jan-> second city thereof
[19:54:46] <molavy> no, it has help just for windows and software for windows that work well on win
[19:54:53] <Tom_itx> i got that Jan-
[19:55:02] <karlp> pocket programmer I've heard before, is it just a clone fo something else?
[19:55:31] <molavy> maybe?
[19:55:43] <karlp> does it have a usb connector?
[19:55:48] <karlp> try -c usbtiny
[19:55:50] <Jan-> I just love the way you can now buy microcontroller equipment in a persian souk
[19:56:03] <molavy> lsusb command return this:Bus 002 Device 005: ID 0403:6001 Future Technology Devices International, Ltd FT232 USB-Serial (UART) IC
[19:56:06] <Jan-> that's excellent :)
[19:58:52] <karlp> what avrdude version do you have?
[19:58:57] <vectory_> +derhad a look at their website, its all in arabic >_<
[19:59:03] <vectory_> -+der
[19:59:05] <karlp> quite a few more were added in 5.11, vs 5.10.
[19:59:15] <karlp> and 5.10 is the standard in debian/ubuntu
[19:59:46] <vectory_> as far as i can tell, pocket programmer can be used with avr studio, so shouldnt it be possible in avrdude, too?
[20:00:30] <karlp> if pocket programmer is this: http://www.sparkfun.com/products/9825 or http://www.sparkfun.com/products/9231
[20:00:33] <karlp> then yes.
[20:00:41] <karlp> if it's something completely different, then who knows?
[20:00:52] <molavy> Jan-,if you see lots of 8 years old children buy chips and microcontroller that may you more happy
[20:00:54] <molavy> avrdude: Version 5.10, compiled on Mar 2 2011 at 21:55:09
[20:01:13] <Jan-> molavy: well, that's cool, they'll learn - right?
[20:02:44] <molavy> yes,of course ,they were there because of robotic course.
[20:03:16] <Jan-> for eight year olds?
[20:03:58] <vectory_> karlp: doesnt look like it
[20:04:05] <vectory_> it has a zif
[20:04:12] <molavy> ya,really, maybe i should take some picture ,maybe you believe
[20:05:10] <Jan-> that would be amazing :)
[20:05:28] <Jan-> This is why the west is getting its ass kicked, guys. Microcontroller robotics courses for eight year olds.
[20:05:36] <Jan-> I'm 33, and I can't do microcontrollers.
[20:07:28] <vectory_> im fairly certain you could, there are enough beginner tutorials and documentation. you just need the stamina to read and digest it all
[20:08:02] <vectory_> 33 :/ bit old for me
[20:08:19] <Jan-> gee thanks :/
[20:09:39] <Tom_itx> send pic, we'll talk
[20:09:40] <Valen> 's ok Jan- i'd still have yer ;->
[20:09:45] <Valen> lol snap
[20:09:45] <Tom_itx> kidding of course...
[20:09:53] <Jan-> Valen: No, you wouldn't.
[20:10:07] <Valen> bah some kind of pesky husband i'll wager
[20:10:14] <Jan-> Tom_itx: try http://emeraldbox.dyndns.org/old/jan/IMG_8537.jpg
[20:10:22] <Jan-> Valen: Near as dammit yes
[20:10:46] <Valen> i bet he's all chiseled good looks and successful to get him some of that pic ;->
[20:11:06] <Valen> my missus would probably rather i didn't too
[20:11:09] <Valen> she's fussy like that
[20:11:15] <Tom_itx> iirc you got into this because of him
[20:11:27] <Jan-> He's really tall and sandy haired
[20:11:49] * inflex wakes up
[20:11:59] <Jan-> Tom_itx: well, originally
[20:12:02] <Jan-> they sent that out in the end
[20:12:06] <Jan-> I'm too stupid :(
[20:12:24] <Tom_itx> i forgot what 'that' was now
[20:12:30] <Tom_itx> some kind of timer thing
[20:12:44] <vectory_> Jan-: nevermind, you look younger :>
[20:12:52] <Jan-> yeah, it was kind of a big deal
[20:13:02] <vectory_> "that" serial card?
[20:13:03] <Jan-> vectory_: I get carded going into bars, which means people fear I'm under 18.
[20:13:16] <Tom_itx> hold on to that as long as you can
[20:13:25] <ziph> Or the bouncer just wants to know where you live.
[20:13:32] <vectory_> :O
[20:13:35] <Valen> one of my friends was getting that at 28, she was getting pissed, she was over the whole :yay i'm young thing lol
[20:13:42] <Valen> crap
[20:13:44] <Valen> i'm 28
[20:13:48] <Valen> crap crap
[20:13:53] <Valen> my birthday is in march
[20:13:55] <Tom_itx> i'm not
[20:14:09] <Valen> and abcminiuser shows all of us up :-<
[20:14:12] <Jan-> Valen: No kidding
[20:14:25] <ziph> I can't believe people will have their drivers licence scanned and fingerprints and a photograph taken just to get in to a bar or nightclub.
[20:14:27] <Jan-> Just before xmas, I was standing at a pedestrian crossing with a load of schoolkids around me, with their teacher
[20:14:41] <Jan-> and the teacher waits for them all to start waking, shoves me, and says "come along, young lady - OMIGODIMSORY"
[20:14:43] <Valen> ziph: fingerprinted for a bar?
[20:14:52] <ziph> Valen: Yes.
[20:14:56] <Valen> Jan-: roflcopter
[20:15:02] <vectory_> wtf? im out
[20:15:20] * Jan- idly stabs Valen
[20:15:49] * Valen chooses Jan-s weapon to be something black and nasty purchased from an adult store
[20:15:59] * Jan- is ferocious
[20:16:06] * Valen enjoys it
[20:16:32] * Jan- will tell Valen's other half
[20:16:42] <Valen> 's ok, she knows i'm a dirty old man
[20:16:56] <Jan-> also, what d'you mean chiseled?
[20:17:02] <Tom_itx> Valen we all knew that
[20:17:04] <molavy> ok, i can't find pic on web for 8 years old child, but may this link help http://robotic.sico-s.com/
[20:17:07] <Jan-> I'd send you a pic, but phil hates photos of himself
[20:17:17] <Jan-> there are some but I couldn't find them
[20:18:23] <Valen> http://www4.pictures.zimbio.com/img/0d94/abman/2m.jpg
[20:18:47] <molavy> and this news about learning robotic at school : http://www.farsnews.com/newstext.php?nn=9004010453
[20:18:47] <Jan-> I can't see pics, remember
[20:19:03] <Valen> duh sorry i forgot its been months since i spoke to you last lol
[20:19:12] <Valen> ripped perhaps?
[20:19:16] <Valen> muscular and lean
[20:19:30] <Jan-> every time you forget, it means we're both doing something right :)
[20:19:51] <Jan-> Actually he's working on his gut right now
[20:19:52] <Jan-> it sags
[20:19:55] <Jan-> or at least it used to
[20:19:55] <Valen> you must have the text to speech speed set to 11 to keep up on IRC
[20:19:57] <Jan-> it sags a bit less now
[20:20:02] <Jan-> oh, you want to hear?
[20:20:06] <Valen> tis ok
[20:20:15] <Valen> i normally have sound turned off
[20:20:19] <Valen> and i'm late for breakfast
[20:20:28] <Valen> its 1 PM here
[20:20:46] <Jan-> Tom_itx: try http://emeraldbox.dyndns.org/old/jan/jawsdemo.mp3 when you can
[20:20:51] <Jan-> er no I mean Valen
[20:21:14] <Valen> just thaught of more work i shopuld be doing
[20:21:38] <Valen> nasty nasty
[20:22:03] <Valen> well that'll learn you for using yahoo ;-P
[20:22:38] <Jan-> people could use easier navigation paradigms
[20:22:48] <Jan-> but I'm in a minority of bugger all so why should anyone care
[20:22:55] <Valen> how does www.vapourforge.com sound to you?
[20:23:31] <Jan-> vaporforge specialises in improving business use of information technology through unerstanding our clients business and providing solutions that work the way they do.
[20:23:39] <Jan-> Oooh, corporate speak. Gimme!
[20:23:49] <Valen> i mean more the navigation and such
[20:24:07] <Jan-> Are they convergently solutionizing a bleeding-edge workflow via paradigm enrichment?
[20:24:10] <Valen> I have been told i can spin a good line of BS
[20:24:20] <Jan-> You wrote that?
[20:24:20] <Valen> i think i'm missing paradigm
[20:24:23] <Valen> yeah
[20:24:25] * Jan- stands ready for more stabbing
[20:24:25] <Valen> my company lol
[20:24:32] * Valen gets some lube
[20:24:40] <Jan-> you do realise that anyone worth doing business with can see straight through that sort of stuff.
[20:24:43] <Jan-> Write it REAL\
[20:25:05] <Valen> it is real actually
[20:25:17] <Jan-> Well it's only got eleven links on it
[20:25:19] <Jan-> which is a nice change
[20:25:23] <Jan-> and the most important ones are first
[20:26:05] <Jan-> you could do accesskeys, but so few places do it that I hardly ever bother trying it.
[20:26:19] <Valen> never heard of em
[20:26:23] <Valen> i shall look into it
[20:26:26] <Jan-> <a href="index.html" accesskey="h">Home</a>
[20:26:44] <Jan-> then I can alt+h and jump to the link
[20:26:57] <Valen> In the summer of 2002, a Canadian Web Accessibility[1] consultancy did an informal survey to see if implementing accesskeys caused issues for users of adaptive technology, especially screen reading technology used by blind and low vision users. These users require numerous keyboard shortcuts to access web pages, as “pointing and clicking” a mouse is not an option for them. Their research showed that most key stroke combinations did in
[20:27:04] <Jan-> (in chrome, other browsers differ, but that's my problem)
[20:27:27] <molavy> avrdude -c usbtiny -B 1 -p m32 -U flash:w:main.hex
[20:27:27] <molavy> avrdude: Error: Could not find USBtiny device (0x1781/0xc9f)
[20:27:27] <Jan-> Hm. I uhoh.
[20:27:43] <Jan-> Like I say, I never really use them, I was parroting what I'd read :)
[20:27:47] <Valen> but thats cool i haven't heard of them before
[20:28:02] <Valen> i can see applications for them in "programs" i do
[20:28:15] <Valen> where people are using web forms for text entry and such
[20:28:36] <Jan-> the web is such a broken thing really
[20:28:45] <Jan-> it's used for so much it was never intended for
[20:28:47] <Valen> its like a sand castle
[20:29:06] <Valen> built out of the smashed up shells of ancient creatures
[20:29:30] <Jan-> that said, the wiki entry is right
[20:29:39] <Jan-> when I'm surfing the web, people assume I'm typing an article up
[20:31:16] <Jan-> anyway I should at least try to sleep
[20:31:39] * Valen is tempted to make a bad joke about "shuteye" and decides not to
[20:31:45] * Valen is sore from all the stabbing
[20:31:51] <Jan-> errr
[20:31:53] <Jan-> what joke
[20:31:59] <Jan-> you got me interested now!
[20:32:12] <Valen> never heard of sleeping as getting some shuteye?
[20:32:59] <Jan-> yes, but I'm not quite sure how you were going to make that into a joke.
[20:33:09] <Valen> that was the joke
[20:33:18] <Valen> i did mention it was bad right?
[20:33:20] * Jan- *blink*
[20:33:27] <Jan-> It's not bad, I just don't get it
[20:34:13] * Jan- buys limoncello online
[20:34:29] <Valen> saying shuteye to a vision impaired person, seems rather like the classic Hollywood inadvertent blunder wouldn't you say?
[20:34:38] <Jan-> mmmmnnnn
[20:34:40] <Jan-> no
[20:34:45] <Jan-> I shut my eyes when I go to sleep :)
[20:34:58] <Valen> well i know that, you know that, but most people are idiots
[20:35:05] <Jan-> If you cut me, do I not bleed
[20:35:21] * Jan- can't remember any more shakespeare
[20:35:25] <Valen> yeah but its probably green and eats through metal knowing you
[20:36:03] <Jan-> ha!
[20:36:31] <Valen> LUNCH!
[20:36:44] <Valen> does your reader convey all caps to you somehow?
[20:37:05] <Jan-> not usually, but sometimes it decides that short words in caps are initialisms to be read out character by character
[20:37:10] <Jan-> AVR, for instance
[20:37:22] <Jan-> earlier on it read "ACR" as "acre"
[20:37:25] <Valen> I always say each letter in that
[20:37:47] <Valen> for your reference my lunch was in all caps, smiley face
[20:38:00] <Valen> or should i say, wink face
[20:38:01] <Jan-> mmhmm
[20:38:08] <Jan-> well you could just say ";)"
[20:38:18] <Jan-> that's what everyone else does
[20:38:41] <Valen> i figured it'd be nicer to expand it rather than throwing semi colons around the place ;-P
[20:39:16] <Jan-> dude if I waited for everyone to modify their behavior on the basis of some odd idea they'd got of what would help me, I'd sit at home doing nothing all day
[20:39:51] <Tom_itx> ok bak
[20:40:01] <Tom_itx> Jan-, what was the last link for?
[20:40:07] <Jan-> my screen reader
[20:40:13] <Jan-> sorry, it was intended for Valen
[20:40:19] <Tom_itx> o
[20:41:06] <Valen> wasn't trying to help, trying to be nice before getting some lunch
[20:41:16] <Jan-> if you want to be nice you can get ME some lunch :)
[20:41:37] <Valen> sure, chorisos on toast with cheese, served in ~15 minutes
[20:41:39] <Valen> come and get it
[20:41:48] <Jan-> dammit
[20:41:51] <Jan-> anyone got a lightspeed drive
[20:41:56] <Valen> if your not here, i'll assume you don't want them anymore and eat them myself
[20:42:14] <Valen> i had one, but it was in a pair of pants that i lost
[20:42:21] <Jan-> nooooo!
[20:42:23] <Tom_itx> yes
[20:42:23] <Jan-> oh well
[20:42:25] <Jan-> I should sleep
[20:42:26] <Tom_itx> i have 2
[20:42:28] <Tom_itx> err no
[20:42:31] <Tom_itx> lightscribe
[20:42:32] <Tom_itx> sry
[20:42:39] <Jan-> nini all
[20:42:52] <Tom_itx> i forgot where you were
[20:43:12] <Jan-> near London
[20:43:13] <Tom_itx> must be E of me quite a ways
[20:43:18] <Tom_itx> ahh
[20:43:25] <Valen> catchya
[21:16:37] <molavy> i use this 2 command with no success:
[21:16:57] <molavy> avrdude -c avr910 -p m32 -P /dev/ttyUSB0 -F
[21:17:08] <molavy> return : avrdude: avr910_recv(): programmer is not responding
[21:17:24] <molavy> avrdude -c usbtiny -B 1 -p m32 -P /dev/ttyUSB0 -F
[21:17:38] <molavy> return : avrdude: Error: Could not find USBtiny device (0x1781/0xc9f)
[21:19:14] <Casper> what kind of programmer?
[21:20:50] <molavy> hattel pocket programmer
[21:21:03] <molavy> http://translate.google.com/translate?sl=auto&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Fwww.hattel.com%2FPocket%2FProgrammer.htm
[21:21:18] <molavy> i found this too:
[21:21:19] <molavy> http://www.iranled.com/forum/thread-20664-page-2.html
[21:23:30] <Casper> is it supported by avrdude?
[21:25:54] <molavy> i don't know
[21:26:21] <Casper> try to figure it out
[21:52:15] <Tom_itx> yay
[21:52:23] <Tom_itx> new rev programmer works great
[22:15:28] <Landon> ok, I really screwed up that solder o.O
[22:16:11] <Landon> any tips on getting a tqfp32 soldered?
[22:16:24] <Valen> solder paste + frying pan
[22:16:25] <Tom_itx> it's not so hard
[22:16:34] <Tom_itx> you have a fine tip?
[22:17:04] <Tom_itx> i did all my programmers manually until i made my toaster oven
[22:17:41] <Valen> solder one pin on, do the diagonally opposite pin then solder all the things?
[22:17:49] <Landon> not necessarily a fine tip, but a chisel tip that goes to a relatively fine point
[22:18:08] <Valen> most people drag solder, but i never liked it
[22:18:09] <Landon> right now I'm looking at a chip tacked onto a board with 2 solder bridges >.>
[22:18:12] <Tom_itx> yeah, do 2 opposing pins first
[22:18:17] <Landon> that's exactly what I tried, valen :P
[22:18:19] <Tom_itx> then drag the tip across the rest
[22:18:20] <Valen> Landon: get some solder wick
[22:18:27] <Tom_itx> use liquid flux too
[22:18:31] <Tom_itx> that helps a bunch
[22:19:15] <Landon> none on hand :(
[22:19:36] <Landon> flux that is
[22:21:05] <Tom_itx> it makes a big difference
[22:21:08] <Tom_itx> even paste flux
[22:23:01] <Landon> good thing I got two sets, although I'll try to salvage this mess before bed
[22:23:35] <keenerd> Solder paste + sparkfun's $10 hot air gun. (The only decently priced thing they sell.)
[22:23:55] <Valen> anybody have some suggestions for an introduction to doing some kind of textual menu system for an AVR or should i not bother and make some sort of pyusb communication thing
[22:24:12] <keenerd> I got it on a lark to see if it could do SMT, and it does rather well for not being adjustable.
[22:25:38] <keenerd> Valen: Have you ever used telnet style text menues? That works okay over a serial connection.
[22:25:54] <Valen> i've used them plenty of time
[22:26:06] <Valen> i have no less than 8 ssh windows open at the moment ;->
[22:26:14] <Valen> i've never written one in C however
[22:26:58] <keenerd> It is pretty simple. Print a list of numbered stuff, wait for input. Standard fare if you did BASIC in the '80s.
[22:27:17] <Valen> yeah, but basic didn't use stdio
[22:27:30] <Valen> text manipulation seems to not be the strong point of C
[22:27:53] <keenerd> Single digit numbers work alright ;-)
[22:28:06] <Valen> not helping ;-P
[22:43:38] <ziph> Valen: Just send binary control packets.
[22:43:56] <ziph> Valen: Then the PC software can be nice and easy to use.
[22:44:20] <Valen> yeah probably the way to do it
[22:44:27] <ziph> Valen: And if you move your project to USB later you can wrap those same packets in USB requests.
[22:44:38] <Valen> its going to be usb to start with
[22:44:40] <Valen> lufa ftw
[23:39:28] <skorket> I have an ISR on timer 0 overflow for the ATTiny13. I have it set to increment every clock tick which means it should be called once every 256 cycles. My ISR takes about 100 clock cycles to execute. Is this too much? Is it considered bad form to have an ISR that's so computationally heavy?