#avr | Logs for 2012-10-22

Back
[01:07:20] <skorket> anyone still up?
[01:07:30] <Casper> o/
[01:07:45] <skorket> I'm trying to do an ADC conversion on two different lines
[01:07:59] <Casper> which is weird, since I woke up early and did 10 hours of construction... (hard wood floor)
[01:08:40] <skorket> after I read the first value, can I just set ADMUX to the other one, then just do an 'ADCSRA |= (1<<ADSC);'?
[01:08:43] <skorket> It's in free running mode
[01:09:42] <Casper> free running mode can cause issue, as if you change channel too late, it may misread the port
[01:10:10] <skorket> What's the approved way?
[01:10:19] <Casper> unsure
[01:12:48] <skorket> by the time I read it, the 'start conversion' flag is down....
[01:14:12] <skorket> and also by the time I'm fiddling with the ADMUX, the 'start conversion' flag is down...
[01:15:49] <Casper> it seems like if speed isn't important, using a single conversion can be better.... set, convert, read, repeat
[01:17:07] <skorket> but isn't that what I'm doing?
[01:18:44] <Casper> free running is different...
[01:18:58] <Casper> I'm talking about single conversion, not free running
[01:18:59] <skorket> ah, ok, how do I not put it into free running mode
[01:19:10] <Casper> and use the conversion complete interrupt
[01:19:22] <Casper> do not set the free running bit
[01:20:12] <skorket> ok, so I don't know, but I found that what I thought to be 'free running' wasn't the actual definition and, if that's true, I think you might be falling into the same trap
[01:20:16] <skorket> does '
[01:20:44] <skorket> whoops, sorry... does 'free running' mean 'keep converting' or does it mean 'independent of timer interrupts'?
[01:23:39] <Casper> it mean it convert in loop, once you start it, it retrigger itself
[01:24:22] <skorket> the examples I've seen not only use the fact that ADSC goes to zero but explicitly set it to one when they want to do subsequent conversions
[01:27:43] <e2580> Anyone interested in a sneak peek at a hardware based security device we are about ready to release?
[01:29:36] <Casper> skorket: that's single conversion...
[01:30:06] <skorket> Casper, but they don't use an interrupt. They set it to 'free running' mode...
[01:30:36] <Casper> interrupt can be used in both mode
[01:32:23] <skorket> I'm happy to hear that this guy got it wrong, but this is the example code I was basing my stuff off of: http://www.embedds.com/adc-on-atmega328-part-1/
[01:35:12] <skorket> Sorry, Casper, I have to run. I'll be on later
[01:35:32] <Casper> ok, going bed soon
[01:44:05] <OndraSter> I think I will murder somebody
[01:44:14] <OndraSter> my internet is down AGAIN
[01:44:28] <OndraSter> I am pretty sure that last time it fixed itself by another fault...
[01:44:41] <OndraSter> because I spoke with neighbour where the local AP is located
[01:44:57] <OndraSter> and I know now what they think about us (part of the community)
[01:45:38] <OndraSter> "there is only 30 people... that is nothing if we cut them off."
[01:46:00] <OndraSter> also today I had a buzzer that woke me up but I fell asleep again :D
[01:46:39] <e2580> Anyone interested in a sneak peek at a hardware based security device we are about ready to release?
[01:46:53] <OndraSter> mmm?
[01:47:05] <OndraSter> naked pictures of PCBs are always hot
[01:47:45] <e2580> well, then check out our site, im sure you will like it www.cryptx2.com
[01:48:02] <e2580> we have bare populated pcbs ready for your viewing pleasure lol
[01:48:42] <Richard_Cavell> OndraSter: I put those 4700 ohm resistors in
[01:48:51] <Richard_Cavell> Now there's no chance of too-high current
[01:49:00] <Richard_Cavell> it's impossible for anything over about 1 milliAmp to go through those thigs
[01:49:01] <Richard_Cavell> things
[01:49:56] <OndraSter> e2580: nice
[01:50:15] <OndraSter> what micro does it use?
[01:50:25] <OndraSter> also, does anything block me from taking out the microSDs? :P
[01:50:34] <e2580> at32uc3a3256s
[01:50:39] <OndraSter> ah
[01:50:45] <e2580> nope, you have 2 micro sd cards in the device, you can swap them
[01:50:49] <OndraSter> I almost thought that nobody uses avr32 :)
[01:51:07] <e2580> well, i am beinging to understand why, its a pain to code for
[01:51:08] <OndraSter> so I can just take out the SD and put it into any reader and... get around the safety?
[01:51:09] <specing> "Due to export regulations we cant ship the micro-controller that is used in the CryptX2 outside the USA. We can ship CryptX2 units that are assembled without the micro controller (not included). You would then be able to add your own micro controller. We are working on other solutions to this issue, check back soon for a better solution."
[01:51:15] <specing> Seriously?
[01:51:21] <OndraSter> specing: yes
[01:51:31] <OndraSter> that is the US law
[01:51:38] <OndraSter> anything that has security module can not be exported
[01:51:38] * specing signs up to voluntarily work for the German nuclear program
[01:51:42] <OndraSter> or requires some paperwork
[01:51:42] <e2580> yes, USA has export laws on crypto, which means the atmel micro we use since it has crypto accelerator on it
[01:51:53] <OndraSter> but it is not completely true
[01:51:54] <OndraSter> you can export them
[01:51:58] <OndraSter> I have bought atxmegas
[01:52:01] <OndraSter> I had to sign some papers
[01:52:06] <OndraSter> that I am not a terrorist
[01:52:14] <specing> I so wish Germany had nukes
[01:52:25] <specing> the world would be so much better
[01:52:32] <OndraSter> I so wish there would be peace around the world.
[01:52:32] <e2580> we cant sell them.... but since this is open source, you can expect very soon someone will have them availabel outside usa
[01:53:43] <OndraSter> are you sure? It just requires few papers and the guy has to sign that he is not in "banned" country
[01:53:44] <e2580> you can take the sd cards out and put them into a normal card reader, and you will be able to make a dd or image of the data, which is aes256 encrypted. so you wont be able to read it
[01:53:53] <OndraSter> ah
[01:53:57] <OndraSter> that's what I wanted to hear :)
[01:54:09] <specing> I don't get it... How is crypt2x supposed to be more secure than a LUKS container on the uSD?
[01:54:23] <OndraSter> hardware vs sofware
[01:54:26] <OndraSter> software*
[01:54:27] <specing> Also the LUKS method costs $0
[01:54:56] <e2580> LUKS/dm-crypt, truecrypt etc are fine, except there are a bunch of ways to get the data
[01:55:01] <OndraSter> with this you need zero extra software on your PC
[01:55:38] <e2580> ie, keyloggers, evil maid attacks, trojans, etc.. the encryption key is in memory on the pc
[01:56:13] <specing> How do you input the key on the crypt2x?
[01:56:26] <e2580> there are buttons onthe cryptx2
[01:56:32] <e2580> have you seen the pictures o the website ?
[01:57:59] <specing> 5 buttons
[01:58:08] <e2580> specing, i am not saying dm-crypt is bad, i am just saying it has methods to exploit it, and under some situations it may nt fit the users needs
[01:58:08] <specing> I guess it would take ages to input anything
[01:58:20] <e2580> you can use 4-128 buttons for the passkey
[01:58:33] <e2580> it is salted and then hashed to create the key for aes
[01:58:49] <e2580> after 10 bad attempts, it wipes the salt, so the data becomes useless
[02:01:58] <OndraSter> it just hit me - my internet is down and there are 6 or 7 TV shows I want to download! lol
[02:02:09] <OndraSter> e2580: and what will be the pricing?
[02:02:26] <e2580> $65-75 USD without sd cards
[02:02:35] <OndraSter> hmm
[02:02:37] <e2580> all updates are free
[02:02:45] <OndraSter> of course
[02:02:58] <OndraSter> also, are they for now hand assembled or already in some factory?
[02:03:37] <e2580> first batches are done in house, we have smd production equipment
[02:03:46] <OndraSter> wow, nice
[02:04:21] <OndraSter> I wish I had smd production equipment at my hosue too :P
[02:04:22] <e2580> small scale :)
[02:04:23] <OndraSter> house*
[02:04:36] <e2580> we dont have full automation or anything crazy like that
[02:04:37] <OndraSter> still more than me
[02:05:10] <OndraSter> are the boards loaded from both sides?
[02:05:32] <e2580> both sides are populated, it is a double sided pcb
[02:05:44] <OndraSter> and how are you going to in house reflow those?
[02:05:57] <e2580> each build requires 3 steps
[02:05:58] <OndraSter> I am asking because I have got few boards for home reflow double loaded too
[02:06:23] <e2580> we paste/populate top side then put in oven
[02:06:35] <e2580> then cool, paste / populate backside and put in oven
[02:06:48] <e2580> then we cool and hand solder the buttons
[02:07:00] <e2580> buttons are not oven friendly :)
[02:07:07] <OndraSter> the tension keeps the parts on the top (bottom) side without falling off?
[02:07:13] <e2580> yes
[02:07:20] <e2580> the solder holds small parts
[02:07:26] <OndraSter> that's assuring, thanks :D
[02:07:31] <e2580> large parts (which we dont have) require glue
[02:07:40] <e2580> they sell red glue for this purpose
[02:08:33] <OndraSter> 3v3 ldo could keep it without falling off..
[02:08:47] <OndraSter> it has got large pads and the tab on the "top"
[02:08:59] <OndraSter> so first reflow the bottom and then the top with xmega
[02:14:56] <OndraSter> Dexter, Mythbusters, The Walking Dead, all waiting in the RSS. But MY INTERNET IS DOWN!
[02:16:06] <OndraSter> meh, time to get up and go to school in a few minutes. Bb
[02:35:10] <RikusW> my toy -> http://tinyurl.com/9vs8ajc
[02:46:25] <jadew> that's awesome
[02:46:30] <jadew> how much did you pay for it?
[02:46:39] <jadew> I wanted to get one for quite some time
[02:47:01] <RikusW> R1500 ~~ $200
[02:47:09] <jadew> nice
[02:47:21] <RikusW> its got 4x 10x 40x objectives and 10x and 16x eyepieces
[02:47:40] <RikusW> and mine gor variable intensity light too
[02:48:02] <RikusW> and a white eyepiece tube and metallic, not black objectives
[02:48:06] <jadew> really cool, I was considering one of those USB microscopes, but the resolution on those kinda sucks
[02:48:15] <RikusW> apart from that nearly identical to the photo
[02:48:36] <RikusW> I'll like to get a 100x oil immersion objective sometime
[02:48:54] <RikusW> and maybe a eyepiece camera
[02:49:11] <RikusW> looking through that too long messes with your eye's focus
[02:49:32] <RikusW> in particular the eye not looking through the eyepiece, weird...
[02:49:47] <RikusW> it lasts a few minutes after leaving the scope...
[02:50:12] <RikusW> it would be much more comfortable viewing on a PC screen
[02:50:39] <RikusW> jadew: then there are widefield scopes too, it would be perfect for assembing 0402 and 0201 :)
[02:51:02] <RikusW> those are around $100
[02:51:15] <RikusW> at least it was in 2006
[02:51:37] <RikusW> and a similar scope without the mechanical table was about $100 too
[02:54:02] <jadew> yeah, looking trough the eyepice for too long will screw up your eyesight, especially if it's not focused properly
[02:54:52] <RikusW> yep
[03:01:56] <RikusW> have to run
[04:02:06] <jadew> alright, OLSFront works under WINE now
[04:02:16] <jadew> crippled, but it works
[04:14:34] <jadew> Horologium, if you ever get an up to date WINE, I would appreciate it if you could try my program with the logic shrimp
[04:28:24] <jadew> anyone with a bus pirate feels like trying out my client? I'm curious if it works on other hardware than the OLS
[04:53:16] <Kevin`> jadew:
[04:53:30] <Kevin`> jadew: I can try it, but it'll be a little while
[04:53:44] <jadew> sure, no rush, thank you
[04:54:19] <jadew> I feel kinda bad today anyway, so I'll be mostly afk
[05:02:43] <tld> jadew: what client?
[05:02:53] <tld> jadew: and get well and stuff.
[05:03:00] <jadew> thanks
[05:03:10] <jadew> http://dumb.ro/lafront/
[05:03:25] <jadew> it's a bit buggy under wine tho
[05:08:27] <tld> you know of sigrok, right? (just checking)
[05:08:52] <jadew> yes
[05:10:59] <tld> curious… why'd you go with javascript? why not something like lua?
[05:11:11] <jadew> because everyone knows javascript :)
[05:11:11] <tld> (not complaining, just curious)
[05:11:20] <tld> figured that might be it.
[05:11:51] <tld> this actually looks fairly cool.
[05:12:20] <jadew> it is pretty cool, still in alpha tho
[05:12:38] <jadew> trying to get full GUI functionality under wine
[05:12:51] <jadew> a pain in the ass to debug and work around stuff
[05:12:59] <tld> I can imagine.
[05:13:12] <tld> I'm wondering how other projects do cross-platform, like the Saleae stuff.
[05:13:36] <jadew> probably by using some library like qt
[05:14:40] <jadew> but I'd rather have native support on one OS and use a library on the other ones
[05:14:54] <jadew> cuz that's what wine basically is
[05:15:04] <jadew> with a few exceptions
[05:20:23] <jadew> alright, just fixed a major bug for wine
[05:35:34] <jadew> ok, so now it should have full functionality under wine as well, expect for the layout saving and toolbar moving thing
[06:34:10] <tld> Hmm.
[06:34:22] <tld> First delivery through china post in 3 or 4 weeks, and it's like the *last* thing I ordered. :/
[06:34:34] <tld> I'm starting to wonder if they burned the rest for fuel or something.
[06:48:18] <jadew> yeah, I'm still waiting for a usb logic sniffer that I ordered more than a month ago
[06:49:06] <Roklobsta> i got a logic sniffer today and i never even ordered it
[06:49:27] <jadew> it must be mine
[06:50:07] <Roklobsta> yes yes thanks!
[06:50:15] <jadew> you're welcome :P
[07:25:11] <Roklobsta> hack your rigol for 100mhz yet?
[07:26:49] <jadew> no, I'm still tempted, but.. not yet
[07:39:39] <tld> jadew: how many arms and legs do you have to give up for a usb logic sniffer?
[07:40:32] <Roklobsta> nor i. avr doesnt need 100mhz
[07:57:20] <jadew> tld, oh, I didn't realize that's what came out lol, I meant a cheap logic sniffer that connects on usb hehe
[07:57:39] <tld> ahh
[07:58:05] <jadew> as for usb logic sniffers, I don't know why they cost that much, will look into it, doubt it's hard to make one
[07:59:11] <Kevin`> it's not hard to make one, but you have to have both a competant hardware designer and a competant software designer
[07:59:39] <Kevin`> and most of your target market will pay anything for a name
[07:59:40] <tld> I'm betting USB3 and all speeds would be considered non-trivial.
[08:10:54] <RikusW> jadew: http://www.lasec.co.za/images/stories/Microscope_Online.pdf nice toys ;)
[08:11:33] <jadew> tld, yeah, I think usb3 is where the hardware part gets tricky
[08:11:41] <jadew> RikusW, checking
[08:11:48] <RikusW> the two at bottome would be nice for smd work
[08:12:49] <RikusW> seems the one at bottom is intended for electronics :)
[08:13:41] <jadew> yeah
[08:13:58] <RikusW> that would make 0201 look big :)
[08:14:43] <jadew> the fact that you need magnification is why I don't get into smd :)
[08:15:52] <jadew> and the fact that I finally have a good variety of components and would suck to start replacing them for smd
[08:15:57] <jadew> I have to use them!
[08:16:01] <RikusW> hmm
[08:16:11] <RikusW> soldering 0805 is easy enough
[08:16:26] <RikusW> using a single magnifiying lens helps a little
[08:16:37] <RikusW> like those on "helping hands"
[08:16:52] <jadew> I have a big one on my lamp
[08:17:08] <RikusW> even better
[08:17:46] <jadew> and those hoods with various magnifying glasses but I don't know...
[08:17:56] <jadew> seems more like for production than for what I'm doing
[08:18:00] <jadew> which is basically learning
[08:18:25] <jadew> half a year ago I wasn't understanding current properly
[08:18:30] <RikusW> ooh
[08:19:05] <RikusW> I did my Industrial electronics N4 and Logic Systems N4 back in 1999
[08:19:22] <jadew> nice, so you're a veteran
[08:19:25] <RikusW> grade 12 == N3
[08:19:50] <RikusW> haven't used my electronics knowledge as much as I would have liked
[08:20:16] <RikusW> I wish I knew about microcontrollers back in 2000
[08:20:44] <RikusW> I got pics back in 2005 or so
[08:21:00] <RikusW> and tried getting a 8051 to work in 2004, but never finished that project....
[08:21:09] <RikusW> 90% and stalled :-P
[08:21:10] <jadew> I used to dissassemble crap as a kid, made radios by following other people schmatics and stuff, then I dropped the hobby in favor of programming and I decided to start again, properly this time
[08:21:39] <RikusW> now I've made 100 U2S boards and need to sell them ;)
[08:21:42] <jadew> oh, I knew about microcontrollers 12 years ago
[08:22:10] <RikusW> converted to AVR in ~~2008
[08:22:12] <jadew> I would have known how to code them as well, since I was doing a lot of asm back then
[08:22:16] <jadew> but 1) I didn't have the money
[08:22:24] <jadew> and 2) I didn't have the necesary electronics background
[08:22:44] <jadew> and when I say I didn't have the money, I actually mean I didn't have the money to buy a freaking PIC
[08:22:47] <RikusW> or the time ?
[08:22:51] <jadew> cuz... I was in highschool
[08:23:37] <jadew> time I had a lot, I was a bad student, so I didn't care much about school
[08:23:46] <jadew> was doing whatever I enjoyed
[08:24:11] <jadew> I guess time only becomes a problem after you get a job
[08:24:58] <RikusW> yep
[08:25:02] <RikusW> or a wife :-P
[08:26:34] <jadew> I think we're more of a problem for them than they are for us
[08:27:52] <jadew> RikusW, you're doing this as a hobby, isn't it?
[08:28:01] <jadew> I mean, it's unrelated to your job?
[08:28:22] <RikusW> I work and live on my father's farm
[08:28:30] <jadew> nice
[08:28:34] <RikusW> but electronics might just become a job too :)
[08:28:45] <RikusW> with the tracker I'm building now
[08:28:51] <DrLuke> Does anybody know if Peter Fleury's UART library has been updated to include the atmega1284p yet?
[08:29:24] <jadew> what do you need a library for?
[08:29:35] <RikusW> DrLuke: if 164, 324 or 644 is supported it should be trivial to add 1284 support
[08:29:36] <jadew> it's 5 lines of code
[08:29:46] <DrLuke> ah ok
[08:31:32] <jadew> RikusW, the reason I asked is because I think the most vocal users are most likely in for the fun
[08:31:56] <RikusW> yep its fun too :)
[08:32:20] <RikusW> unlike some people who thinks programming is difficult...
[08:32:35] * RikusW can't imagine what so difficult
[08:33:04] <RikusW> unless it involves massive api's like WIN32 :-P
[08:33:17] <jadew> well programming is my second nature and it got kinda borring to be honest, that's why I'm really happy I found this hobby
[08:34:23] <RikusW> for us its second nature yes :)
[08:35:29] <RikusW> seems that once the basic idea is understood any language can be used with a bit of effort
[08:36:24] <jadew> yeah, there's not much that can be added
[08:36:33] <jadew> the only difference is that some have stupid syntax
[08:36:58] <jadew> retarded ideas built in, or dumb designers
[08:37:35] <RikusW> yep
[08:41:14] <jadew> btw, have you made any more progress with the debugWire?
[08:41:49] <jadew> I was going to look into it as well, but I guess I got side tracked
[08:42:08] <jadew> and I can't really test stuff since I don't have a device that actually supports it
[08:42:34] <jadew> *a programmer
[08:44:30] <jadew> it would be really cool if usbtiny would gain debugging capabilities
[08:45:21] <jadew> actually, it could be a device of it's own, it wouldn't matter as long as it could do proper debugging
[09:05:32] <RikusW> I'll probably work on dW again
[09:05:54] <jadew> could bring joy to many people :)
[09:06:02] <RikusW> will finish the autobaud first, then the PC sw which will take considerable effort....
[09:06:46] <RikusW> jadew: my ISP programmer already disables it automatically
[09:06:55] <RikusW> much less headaches
[09:07:22] <jadew> I know, I think I've read everything you wrote on the subject when I started looking into it
[09:07:50] <jadew> including log files since the begining of time :P
[09:07:57] <RikusW> nice
[09:08:18] <RikusW> if you have a serial port you can easily start messing around
[09:08:32] <RikusW> as explained in that doc on rue's site
[09:08:41] <jadew> I know, did that
[09:08:43] <RikusW> the untested part was tested and works
[09:08:51] <RikusW> and got it working ?
[09:09:01] <jadew> I managed to sync
[09:09:11] <RikusW> nice, and disable dW again ?
[09:09:15] <jadew> bricked an attiny13 (in the idea that I'll disable dw)
[09:09:26] <RikusW> keep it
[09:09:30] <jadew> but failed and I think my attention shifted to something else
[09:09:38] <jadew> of course :)
[09:09:56] <jadew> yeah, I think that's about when I decided to design a power source
[09:10:06] <RikusW> I'm thinking of adding some dW bootloading sw on the AVR
[09:10:15] <RikusW> it would greatly speed up fw upload
[09:10:16] <jadew> that would be nice
[09:10:26] <jadew> deffinitely
[09:10:27] <RikusW> via dW that is
[09:10:50] <RikusW> 10kb/s :)
[09:10:53] <jadew> and it would make the reset pin a bit less useless
[09:11:14] <RikusW> so instead of using a bootloader use dW :)
[09:11:24] <RikusW> single pin uart
[09:11:26] <jadew> yeah, but can it actually be done?
[09:11:32] <RikusW> of course
[09:11:36] <jadew> as a bootloader I mean
[09:11:45] <RikusW> thats what AS does to load the fw
[09:12:03] <jadew> interresting
[09:12:06] <RikusW> having dW on increases current consumption
[09:12:16] <RikusW> so only a bootloader for prototyping
[09:12:41] <RikusW> on rue's site its actually documented how to flash via dW too
[09:13:25] <jadew> nice
[09:13:42] <RikusW> basically using code like a bootloader
[09:13:56] <RikusW> all dW stuff works by executing instructions on the AVR
[09:14:09] <jadew> yeah
[09:14:31] <jadew> I guess the instruction words differ based on the chip, isn't it?
[09:16:13] <RikusW> all mega and even xmega instructions are the same
[09:16:25] <jadew> interresting
[09:16:27] <RikusW> though some AVRs support a few more instructions
[09:16:39] <RikusW> xmega support DES for example
[09:16:48] <jadew> what's that?
[09:16:48] <RikusW> and not all AVRs support MUL
[09:16:54] <RikusW> encryption
[09:17:05] <jadew> ah, thought it's an opcode hehe
[09:17:15] <jadew> or is it?
[09:17:16] <RikusW> it is
[09:17:27] <RikusW> both
[09:17:27] <jadew> ah, neat
[09:17:37] <RikusW> AES comes as a seperate module
[09:17:47] <RikusW> don't know why DES was done that way...
[09:18:13] <jadew> don't know the way they work, but I guess DES is straight forward
[09:18:24] <jadew> AES might look back in the stream or something
[09:19:06] <RikusW> don't know, don't care to much right now
[09:19:19] <RikusW> I do care about import/export nastyness :S
[09:19:56] <jadew> any idea if the firmware on the original programmers that support dw, can be read?
[09:20:28] <jadew> I suppose no, otherwise the whole dW mistery would probably be solved by now, eh?
[09:20:51] <RikusW> dragon fw is AES (probably) encrypted :(
[09:20:56] <RikusW> I already had a look
[09:21:10] <jadew> hmm
[09:21:19] <RikusW> Steffanx even connected a jtagice to the dragon jtag ports underneath
[09:21:28] <RikusW> the m128 ans m2560 are locked :(
[09:21:32] <RikusW> *and
[09:21:52] <jadew> wait, did you guys managed to get any data out?
[09:22:00] <RikusW> lets go to the die level :-P
[09:22:12] <RikusW> no data, lockbits was set :-/
[09:22:23] <jadew> you said it's AES encrypted
[09:22:36] <jadew> how did you figure out?
[09:22:39] <RikusW> probably, and decrypted by the bootloader on the dragon
[09:22:50] <jadew> hmm
[09:23:03] <RikusW> Atmel have two appnotes, for DES and AES bootloaders
[09:23:12] <RikusW> so ofcourse they used that....
[09:23:23] <jadew> can you run stuff from outside the program memory? as in.. from the sram?
[09:23:30] <RikusW> only on AVR32
[09:23:32] <jadew> ah, got it
[09:23:50] <jadew> yeah, so it's most likely crypted, however if you would read the flash
[09:23:56] <jadew> you'd gain access to the key as well
[09:24:12] <jadew> so AES or not, it would be just a minor set back
[09:24:28] <RikusW> yep
[09:24:35] <jadew> let's think about it
[09:24:45] <RikusW> but you'll need to disassemble the actual ic's on the dragon
[09:24:56] <jadew> there's no way of resetting the lock bits, right?
[09:25:06] <RikusW> chip erase ofcourse :-P
[09:25:13] <jadew> ah ha
[09:25:17] <jadew> so it is possible
[09:25:29] <jadew> I thought it's a physical thing that gets switched
[09:25:40] <jadew> like... a one time fuse
[09:25:52] <RikusW> no
[09:26:05] <RikusW> AVR "fuses" are reprogrammable
[09:26:06] <jadew> no known hacks to only switch the lock bits tho, right?
[09:26:15] <RikusW> open the chip ;)
[09:26:23] <RikusW> thats if you have the tools
[09:26:40] <jadew> yeah, I would think that's out of our hands
[09:26:41] <RikusW> tool$$$$$
[09:26:45] <jadew> hehe yeah
[09:27:11] <RikusW> Focused Ion Beam FIB and so on
[09:27:31] <jadew> there must be a bug we can exploit :P
[09:27:54] <jadew> like... something to kick the dw mode on
[09:28:41] <RikusW> lockbits disable that
[09:28:49] <jadew> I figured
[09:29:33] <jadew> and I don't think it's a just a switch that needs to be turned on
[09:29:45] <jadew> it's deffinitely a complex module, since you said it eats more power
[09:30:03] <jadew> so there's more to it than just an on off switch at base of the pin
[09:30:55] <jadew> wait, doesn't it support firmware upgrade?
[09:30:58] <jadew> *update
[09:32:48] <jadew> I just realized this (I never used an original programmer), but all this would mean that they never issue firmware updates for their programmers
[09:32:51] <jadew> is that true?
[09:34:07] <jadew> "The flexible and secure firmware upgrade feature allows AVR Studio to easily upgrade the AVR Dragon to support new devices."
[09:34:13] <jadew> http://support.atmel.no/knowledgebase/avrstudiohelp/mergedProjects/AVRDragon/whnjs.htm
[09:34:17] <RikusW> they do give new fw with each new studio version
[09:34:28] <RikusW> jadew: nice scale -> http://www.lasec.co.za/index.php?page=shop.getfile&file_id=980&product_id=1079&option=com_virtuemart&Itemid=79
[09:34:36] <RikusW> 5g in 1ug increments
[09:35:08] <jadew> nice, but what is it useful for?
[09:35:20] <RikusW> chemicals
[09:35:33] <RikusW> labs use it to measure contamination on air filters
[09:35:33] <jadew> ah, right
[09:36:19] <RikusW> I'd use it to say measure 1mg of CuSO4, to be eaten ;)
[09:36:44] <RikusW> actually 5mg of Cu can be taken each day without harm
[09:36:45] <jadew> seems a bit extreme only to measure stuff for your etcher :)
[09:36:58] <RikusW> not etch, EAT
[09:37:14] <RikusW> like mineral supplements
[09:37:36] <jadew> there must be safer ways :)
[09:37:40] <RikusW> though I guess that one is EXPENSIVE
[09:38:09] <RikusW> CuSO4 isn't that bad, though the copper taste remains for a whole day :-P
[09:38:25] <RikusW> 1g of it will be BAD indeed....
[09:39:11] <jadew> you might be able to get to the oz only from mineral water
[09:39:13] <RikusW> al long as its pure I don't see what harm it can do in small quantities
[09:39:25] <RikusW> *as
[09:39:48] <RikusW> Ca and Mg sort of prevent absorbsion of Cu
[09:39:58] <jadew> hold on, I don't know if oz is an abbreviation from my language
[09:40:08] <jadew> oz would be the daily recommended quantity
[09:40:27] <RikusW> its also a antidote for Cu poisoning (dolomite)
[09:40:37] <RikusW> RDA....
[09:40:48] <RikusW> RDA for Cu is a few mg
[09:42:14] <jadew> hmm, I'm not sure, but isn't CuSO4 a good poison?
[09:42:59] <RikusW> in large quantities yes
[09:43:00] <jadew> I looked it up on wiki and this cought my eye: "bluestone"
[09:43:08] <RikusW> yes its bluestone
[09:43:35] <jadew> I'll ask my g/f if it's the poison I'm thinking about
[09:43:50] <jadew> she's the chemist of the family
[09:43:56] <RikusW> I used it on cancerous growths on horses too, it simple falls of after a while
[09:44:07] <RikusW> I got that advice in an OLD veterinary book\
[09:44:24] <RikusW> it actually works better and much cheaper than surgery
[09:44:40] <jadew> low tech chemo therapy
[09:44:44] <RikusW> yep
[09:44:51] <RikusW> but maybe better than hightech
[09:45:04] <RikusW> simple cheap, works fast too
[09:45:08] <RikusW> sometimes in a week
[09:45:51] <jadew> probably more dangerous as well
[09:45:59] <RikusW> there is actually a book on the subject of copper -> soil grass and cancer by Andre voisin
[09:46:02] <jadew> how were the horses feeling?
[09:46:14] <RikusW> didn't notice any problems
[09:46:24] <RikusW> and now you can't even see any scars
[09:47:56] <RikusW> I feed them a mixture of 8p CuSO4 8p sulfur 8p kelp 48p dolomite
[09:48:01] <RikusW> pound
[09:48:16] <RikusW> and usually adding 8p of fine salt too
[09:48:43] <RikusW> kelp == bamboo == seaweed
[09:49:04] <jadew> I see
[09:49:14] <RikusW> it was recommended in a book
[09:49:34] * RikusW took a little bite too
[09:49:39] <RikusW> don't taste that bad
[09:49:50] <jadew> heh
[09:50:06] <RikusW> book -> "Natural Horse Care" by Pat Coleby
[09:50:22] <RikusW> can be found on acresuse.com iirc
[09:50:30] <RikusW> *acresusa
[09:50:32] <jadew> wish I lived on a farm
[09:50:53] <RikusW> much more relaxed..
[09:51:00] <jadew> I bet
[09:51:22] <RikusW> but that will probably change with this tracker project and so on...
[09:52:08] <jadew> do you have a deadline or it's a personal project?
[09:52:28] <RikusW> it should be working asap...
[09:52:34] <RikusW> not personal
[09:53:23] <RikusW> the PA6H does geofencing for me, so one less thing to worry about :)
[09:53:30] <RikusW> and the SIM900 does TCP/IP
[09:54:01] * RikusW didn't really feel like implementing all that, especially in limited time
[09:54:29] <jadew> what chip are you using?
[09:55:05] <jadew> proper tcp/ip is not that easy to implement, I doubt the sim900 is actually implementing it properly
[09:55:28] <RikusW> I does implement sockets afaik
[09:55:36] <jadew> yeah, I don't doubt that
[09:56:01] <RikusW> as long as I can connect to a server, its all fine
[09:56:04] <jadew> the problem is with the fact that TCP/IP has to ensure that no matter the order in which packets come in, they are served in the right order to the client (your app)
[09:56:09] <jadew> however, that means a lot of ram
[09:56:25] <jadew> which is not abundent in embeded projects
[09:56:32] <RikusW> there is an ARM on the SIM900 afaik
[09:57:26] <jadew> I'm sure it's fine for the ammount of data you'll pass trough :)
[09:57:50] <RikusW> yeah, like 1MB a month :-P
[10:02:14] <jadew> hmm I think you need to ack tcp/ip packets...
[10:02:23] <jadew> in which case 64k should be enough
[10:02:47] <RikusW> right
[10:02:52] <jadew> if that's the case you could reject out of sync packets until you get the one you're expecting
[10:03:19] <RikusW> I'm not doing it that low level yet :-P
[10:03:23] <RikusW> fortunately
[10:03:43] <jadew> yeah, it's deffinitely nice to have the hardware do it for you :)
[10:03:47] <RikusW> and packets will be very small anyways
[10:04:10] <RikusW> less than 1kb
[10:04:33] <jadew> yeah, I'm sure it will work just fine, was just thinking out loud :P
[10:05:15] <jadew> now I got curious again about this dW thing
[10:05:39] <RikusW> you can actually mess with it if you wnat
[10:05:43] <jadew> especially about the part in which the programmers get new firmware with each AVR Studio :)
[10:06:07] <RikusW> thats just Atmel nonsense
[10:06:23] <RikusW> I hacked AS4 to accept newer fw instead of forcing a downgrade
[10:06:31] <RikusW> and AS5 fw works just fine in AS4
[10:06:41] <RikusW> I guess its all just quality control
[10:06:48] <jadew> yeah, but the point is that there's FW flying around
[10:06:48] <RikusW> and avoidance of support headaches
[10:06:55] <RikusW> it is indeed
[10:06:58] <jadew> we could learn something from it
[10:06:59] <RikusW> encrypted
[10:07:18] <RikusW> I do have the decrypted STK500 and JTAGICE mki fw
[10:07:33] <jadew> do any of those support dw?
[10:07:37] <RikusW> its decrypted by avrprog before being sent to the board
[10:07:39] <RikusW> no
[10:07:58] <jadew> heh, nice
[10:08:01] <RikusW> https://sites.google.com/site/megau2s/home/jtag
[10:08:09] <RikusW> there you can look at the jtag fw
[10:08:16] <RikusW> its a bit nasty though....
[10:08:22] <RikusW> optimizing C compiler :(
[10:08:42] <RikusW> and 8000 lines or so after commenting it
[10:09:24] <RikusW> my clone is much more readable
[10:09:30] <jadew> if it doesn't do dW it wouldn't make me much good to read it
[10:09:42] <RikusW> get a jtag AVR
[10:09:53] <RikusW> m16 162 32 64 128
[10:10:14] <RikusW> my AS4 supports programming newer AVRs too but not debugging
[10:10:20] <RikusW> though thats AS4's fault
[10:10:44] <RikusW> I modified my clone to support new AVRs too
[10:11:10] <jadew> jtag or avrisp?
[10:11:26] <RikusW> jtag
[10:11:49] <jadew> never used jtag
[10:11:59] <RikusW> stk500 supports about all know AVR's, so does my clone
[10:12:07] <RikusW> *known
[10:12:22] <jadew> yeah, it's because it has that isp multi command
[10:12:33] <RikusW> Actually AS don't use that
[10:12:39] <RikusW> avrdude abuses it :S
[10:12:43] <RikusW> breaking my bootloader
[10:12:52] <jadew> I only used avrdude :)
[10:13:00] <RikusW> so I had to add HVPP mode for use with avrdude
[10:13:10] <jadew> hehe
[10:13:12] <RikusW> so my bootloader understands both ISP and HVPP commands
[10:13:30] <RikusW> and USB CDC in 2k flash
[10:13:39] <RikusW> and the module code too...
[10:13:53] <jadew> neat
[10:13:53] <RikusW> there is literally not a single word left...
[10:14:17] <RikusW> USB is about 800 bytes
[10:14:33] <megal0maniac> Dejavu :)
[10:14:36] <RikusW> hi
[10:14:36] <tobbor> hi RikusW.
[10:14:47] <jadew> hi
[10:14:47] <tobbor> jadew! like, totally tell us about the project!
[10:14:52] <RikusW> lol
[10:14:53] <jadew> haha
[10:15:05] <jadew> now I know it's a bot
[10:15:23] <RikusW> robbot...
[10:15:34] <jadew> oh..
[10:15:35] <jadew> haha
[10:15:41] <RikusW> megal0maniac: look in the scrollup for the microscope link
[10:16:16] <RikusW> http://www.lasec.co.za/images/stories/Microscope_Online.pdf
[10:16:55] <megal0maniac> LEAVES!
[10:16:56] <megal0maniac> :)
[10:22:14] <jadew> RikusW, this is the only presentation page for the board? https://sites.google.com/site/megau2s/home
[10:25:09] <jadew> you at least need this, in the title: "U2S ATmega32U2 development board"
[10:27:09] <RikusW> ok
[10:27:14] <RikusW> will change that
[10:27:48] <RikusW> already on the first line
[10:28:12] <jadew> also, if you can remove the scripts in the body, before <div id="sites-chrome-everything-scrollbar"> and move them somewhere else (at the end of the body - if possible) it would be good
[10:28:36] <jadew> and change <h2><a id="sites-chrome-userheader-title" dir="ltr" href="https://sites.google.com/site/megau2s/">U2S</a></h2>
[10:28:38] <jadew> to h1
[10:28:39] <RikusW> I don't have control over that...
[10:28:45] <jadew> oh...
[10:28:50] <RikusW> its google sites...
[10:29:58] <jadew> well, I know that U2S by itself at the top looks cool, but maybe it could be more descriptive, it would help with the search rating
[10:31:14] <RikusW> hmm
[10:31:21] <RikusW> suggestions ?
[10:31:28] <jadew> a new page
[10:31:31] <jadew> :)
[10:31:39] <jadew> that you can do whatever you want with
[10:32:04] <jadew> so you can make it flashier and make sure all the important words get the highest weight
[10:32:29] <RikusW> the purpose of that site was to host support sw
[10:32:47] <RikusW> and that it does well enough
[10:32:57] <jadew> well, you can still keep the site
[10:33:23] <jadew> but if you want to start sell those things you need more visibility
[10:33:23] <RikusW> host it on kickstart too maybe
[10:33:36] <RikusW> kickstarter
[10:33:39] <jadew> *start selling (sorry, I'm very tired)
[10:33:52] * RikusW is gmt+2, you ?
[10:33:58] <jadew> yes
[10:34:12] <RikusW> europe ?
[10:34:24] <jadew> yep, but screwed up sleep schedule
[10:35:13] <jadew> was pretty sick and I still feel a bit weird, sleept most of the weekend
[10:35:31] <jadew> woke up last night at 12
[10:35:36] <RikusW> heh
[10:36:53] <jadew> you should deffinitely try to make a better page, at least the presentation part
[10:37:15] <RikusW> yep, that wasn't intended for marketing
[10:37:19] <jadew> otherwise people won't know you're selling something
[10:39:25] <jadew> I guess your best bet of being found is by "ATmega32U2 development board"
[10:39:35] <RikusW> yeah
[10:40:16] <RikusW> google -> megau2s :)
[10:40:46] <jadew> :)
[10:41:17] <jadew> ATMEGA32U2.NET <- not taken :P
[10:41:42] <RikusW> mine is on the 4th page...
[10:41:44] <jadew> .com is free as well (wonder if it would be legal to take them tho)
[10:41:57] <jadew> RikusW, for "ATmega32U2 development board" ?
[10:42:26] <RikusW> yes
[10:42:59] <RikusW> without quotes
[10:43:14] <jadew> yeah, I couldn't find it tho
[10:43:22] <RikusW> with on quotes on page one
[10:44:11] <jadew> for me it doesn't show up on the 4th page either
[10:44:20] <jadew> I went trough 12 pages and it didn't show up
[10:44:34] <RikusW> seems like google is not the same for everyone...
[10:44:39] <jadew> yeah
[10:45:03] <RikusW> google -> megau2s ?
[10:45:09] <jadew> first result
[10:45:33] <RikusW> nice :)
[10:48:18] <OndraSterver> WHAT
[10:48:19] <OndraSterver> A
[10:48:20] <OndraSterver> FUCKING
[10:48:21] <OndraSterver> DAY
[10:48:53] <jadew> you got laid twice? with 3 chicks?
[10:49:00] <RikusW> what went wrong ? or shouldn't I ask ? :-P
[10:49:17] <OndraSterver> :D
[10:49:20] <OndraSterver> jadew, I wish
[10:49:27] <OndraSterver> there were chicks involved alright
[10:49:34] <OndraSterver> well, as I was walking to the school I got a call
[10:49:41] <OndraSterver> I was still in the subway (metro) so I couldn't pick it up
[10:50:07] <OndraSterver> when transferring from line B to A I called her back - it is one of the employees of one company that I am helping with IT in
[10:50:13] <OndraSterver> "the internet does not work"
[10:50:40] <OndraSterver> "I am like 10 minutes away from you, I could stop by for 10 minutes, then I have to go to the school"
[10:50:44] <OndraSterver> "okay, we will be waiting"
[10:50:47] <OndraSterver> I come there
[10:50:48] <jadew> "I will do anything, if you can make it work"
[10:50:48] <OndraSterver> internet dead
[10:50:50] <OndraSterver> VoIP dead
[10:50:53] <OndraSterver> haha
[10:50:57] <jadew> "_ANYTHING_"
[10:50:59] <OndraSterver> yes, she is very, very pretty :D
[10:51:14] <OndraSterver> four out of five chick employees I would have offers for :D
[10:51:23] <OndraSterver> two out of those are redheads.
[10:51:24] <OndraSterver> anyway
[10:51:32] <OndraSterver> I didn't find anything wrong on the first look
[10:51:34] <OndraSterver> so I went to the school
[10:51:39] <OndraSterver> did my presentation in 5 minutes and ran back
[10:51:41] <OndraSterver> mkay
[10:51:48] <OndraSterver> so I started pinging and trying stuff
[10:51:50] <OndraSterver> rebooting switches
[10:52:00] <jadew> someone tripped over a wire
[10:52:03] <OndraSterver> ah-ha! that main ISP's switch, when rebooted, would work for like 10 seconds
[10:52:07] <OndraSterver> VoIP phones would come up
[10:52:16] <OndraSterver> they even managed to get a call in that time window :D
[10:52:19] <OndraSterver> then it would crash again
[10:52:20] <OndraSterver> reboot
[10:52:25] <OndraSterver> and diag LED started blinking
[10:52:30] <OndraSterver> yet the switch LEDs were blinking as well
[10:52:41] <OndraSterver> okay, I called the ISP after 1.5 more hours of trying (dis)conencting stuff
[10:52:51] <OndraSterver> they said "okay, we will sent employee to replace that switch"
[10:52:58] <OndraSterver> after 1 hour I am calling them "so where is he?"
[10:53:00] <OndraSterver> "he just left"
[10:53:10] <OndraSterver> 1 more hour till he arrived (the company is like 10 minutes by car from the ISP)
[10:53:21] <OndraSterver> in 3 minutes he replaced the switch
[10:53:28] <OndraSterver> in 20 minutes I rebooted the server
[10:53:30] <OndraSterver> and it all worked.
[10:53:43] <RikusW> augh
[10:53:43] <OndraSterver> they have got bad circuit breakers
[10:53:45] <jadew> in xx minutes you banged one of the readheads
[10:53:46] <OndraSterver> 10A and 10A
[10:53:49] <OndraSterver> yet they trip too much
[10:53:50] <jadew> 20 minutes back home
[10:53:53] <OndraSterver> haha
[10:53:58] <OndraSterver> no, I have got it about an hour home
[10:54:03] <OndraSterver> that's why I didn't leave
[10:55:18] <jadew> trying to remember a readhead I like...
[10:56:37] <OndraSterver> jadew, http://sphotos-b.ak.fbcdn.net/hphotos-ak-ash4/334846_431613313565402_575066497_o.jpg
[10:56:53] <OndraSterver> http://sphotos-c.ak.fbcdn.net/hphotos-ak-ash3/539374_430236243703109_294063854_n.jpg
[10:57:03] <jadew> one of the chicks?
[10:57:16] <OndraSterver> haha
[10:57:22] <OndraSterver> <jadew> trying to remember a readhead I like...
[10:57:25] <OndraSterver> I picked two random
[10:57:29] <OndraSterver> I saw on facebook
[10:57:32] <OndraSterver> (I don't know them)
[10:57:37] <OndraSterver> facebook group Redheads
[10:57:39] <jadew> ah, I already found mine, I'm trying to find a good picture now :P
[10:57:44] <OndraSterver> :D
[10:58:52] <OndraSterver> this is the new employee: http://m3.licdn.com/media/p/4/000/157/283/2a8561b.jpg
[10:59:40] <jadew> annoying, can't find one that does her justice
[10:59:50] <jadew> it's anna faris in my super ex-g/f
[10:59:55] <jadew> I like her blonde too
[11:00:11] <OndraSterver> eh
[11:00:51] <jadew> http://2.bp.blogspot.com/_8KdM_8NUdfg/TA4BxSYYkFI/AAAAAAAAACc/Yl6mlISLQww/s1600/anna-faris.jpg this is the kind of readhead I like hehe
[11:01:22] <OndraSterver> one question: are you colourblind? :P
[11:01:40] <OndraSterver> Linkedin, Y U NO AGE!
[11:01:47] <jadew> if a chick can do that, the color of her hair doesn't matter :P
[11:01:54] <OndraSterver> :D
[11:02:01] <OndraSterver> I can do that too.
[11:02:31] <OndraSterver> and because I am awesome I can do much more!
[11:02:42] <OndraSterver> and I should probably at least partially shave, I look too much like some homeless guy.
[11:03:26] <jadew> same here, I'm not sure if that's a standard geek look or if it comes with enjoying electronics
[11:04:07] <RikusW> or laziness to shave :-P
[11:04:15] <jadew> or that hehe
[11:04:20] <RikusW> or avoiding the ithcing :S
[11:04:22] <OndraSterver> i vote RikusW 's answer
[11:04:35] <RikusW> OndraSterver: 1 2 or both ?
[11:04:51] <RikusW> clipper once a month is much easier :)
[11:05:00] <OndraSterver> <RikusW> or laziness to shave :-P
[11:05:12] <OndraSterver> I haven't shaved since... the end of july :D
[11:05:16] <OndraSterver> well, I did
[11:05:32] * RikusW haven't shaved since 2003
[11:05:36] <RikusW> clipped yes
[11:05:49] <jadew> I got a cool shaving brush so I tend to shave more often now
[11:06:05] <OndraSterver> heh
[11:06:08] <OndraSterver> I just use electric one
[11:06:42] <jadew> I like the foam
[11:06:45] <RikusW> electric might be better
[11:06:54] <jadew> and the feeling you get when you use the brush
[11:06:56] * RikusW don't like blades...
[11:08:21] <jadew> I have very sensitive skin, that's why I decided to go for brush + razor
[11:08:45] <jadew> to desensitise it
[11:10:05] <jadew> maybe it's just me, but people that shave often seem to have a much healthier skin
[11:10:12] <RikusW> I'd rather not scratch where it doesn't itch
[11:10:18] <OndraSterver> :D
[11:10:32] * RikusW applies that comment to electronics too
[11:10:35] <jadew> well it itches
[11:10:38] <OndraSterver> I don't like how everything (water) keeps stuck in the beard
[11:10:42] <RikusW> like the inside of a HDD :-P
[11:10:50] <jadew> cuz I get iritated easily
[11:16:50] <RikusW> jadew: http://imgur.com/Euxbw
[11:16:54] <RikusW> bloodsmear
[11:17:58] <jadew> you just took that?
[11:18:05] <RikusW> using 640x and 2MP cellphone camera
[11:18:09] <RikusW> a while ago
[11:18:13] <RikusW> no staining
[11:18:18] <jadew> nice
[11:18:25] <RikusW> camera made it look blueish
[11:18:44] <jadew> ok, I want a microscop too
[11:18:46] * RikusW goes looking for the uveprom images
[11:19:05] <jadew> *microscope
[11:19:09] <jadew> damn I'm tired
[11:19:37] <RikusW> those rough cells are probably an artifact, moisture on the slide
[11:20:03] <jadew> speaking of pictures, I was meaning to ask about this, took the screenshot this weekend: http://imagebin.org/232845
[11:20:12] <RikusW> its called an echinocyte iirc
[11:21:11] <jadew> I was trying to get 0 and 1 out of a photoresistor sensitive to IR and was shooting a remote control at it, had it in series with another resistor to get a voltage divider so I can look at the signal
[11:21:39] <jadew> wth is with the lower amplitude waves? is that normal for IR? sending dimmer signals?
[11:21:54] <jadew> do they have a purpose?
[11:24:59] <jadew> if I was using a different resistor in series with the photoresistor, I could have gotten full amplitude waves instead of the smaller ones so what's their purpose cuz they seem to prevent me from getting a signal I can work with
[11:26:15] <RikusW> afk
[11:27:01] <RikusW> not worked too much with ir
[11:27:21] <RikusW> though 12k pullups on the photo diode worked just fine
[11:27:22] <jadew> my first idea was to feed that trough a capacitor to an opamp and get a clear 0 and 1 no matter the distance, but with that thing in there, I'm not sure how to use the signal
[11:27:45] <RikusW> but then again it was against the led with a greycode wheel inbetween, no ambient light entering
[11:27:49] <JyZyXEL> what could cause AVR to execute code too slow?
[11:28:01] <Richard_Cavell> JyZyXEL: Are you using the right clock settings?
[11:28:03] <RikusW> JyZyXEL: slow clock
[11:28:03] <jadew> JyZyXEL, a slow clock
[11:28:27] <JyZyXEL> could the internal oscillator be an order of magnitude off?
[11:28:27] <RikusW> jadew: use a schmitt trigger and opamp
[11:28:47] <RikusW> actually an opamp can be used as a trigger too
[11:29:05] <RikusW> so a 8pin dip with 2 opamps should be fine
[11:29:06] <JyZyXEL> the led flasher is running way way way too slow
[11:29:18] <RikusW> wrong clock setting in the makefile...
[11:29:20] <jadew> looking into it
[11:29:22] <jadew> thanks
[11:29:32] <JyZyXEL> im using _delay_ms(20); yet in the real world its more like 200ms
[11:29:58] <jadew> JyZyXEL, 3 variables here
[11:30:11] <jadew> 1) the clock setting in the makefile, as RikusW said
[11:30:24] <jadew> 2) the actual clock you're using
[11:30:31] <jadew> 3) CKDIV8
[11:30:56] <JyZyXEL> :O
[11:31:00] <JyZyXEL> that must be it!
[11:31:19] <RikusW> change the PLL register
[11:31:49] <JyZyXEL> i think i have CKDIV8 enabled
[11:31:59] <RikusW> PLL register can override that
[11:32:35] <JyZyXEL> hmm?
[11:33:17] <RikusW> JyZyXEL: use the CLKPR register
[11:33:30] <RikusW> SetPrescaler:
[11:33:30] <RikusW> ldi r17,(1<<CLKPCE)
[11:33:30] <RikusW> sts CLKPR,r17
[11:33:30] <RikusW> sts CLKPR,r16
[11:33:31] <RikusW> ret
[11:33:56] <RikusW> CLKPR = (1<<CLKPCE);
[11:34:02] <RikusW> CLKPR = value;
[11:34:45] <jadew> I'm gonna take a nap
[11:34:47] <jadew> ttyl
[11:36:32] <RikusW> mega128 http://imagebin.org/232851
[11:38:53] <RikusW> eeprom -> http://imagebin.org/232852
[11:39:49] <RikusW> http://imagebin.org/232853
[11:40:34] <JyZyXEL> judging from attiny85 manual i think: CLKPR = (1<<CLKPCE); alone is sufficient
[11:40:37] <OndraSterver> now read the data!
[11:40:45] <RikusW> http://imagebin.org/232854
[11:40:59] <JyZyXEL> i think you need to wait for four cycles though
[11:41:07] <OndraSterver> no
[11:41:17] <OndraSterver> IIRC clkpr changes = twice in <4 cycles apart
[11:41:17] <RikusW> JyZyXEL: I think you need to add =0; too
[11:41:25] <OndraSterver> or was it another register
[11:41:52] <RikusW> OndraSterver: above is full resolution eeprom photos
[11:41:54] <JyZyXEL> The CLKPCE bit must be written to logic one to enable change of the CLKPS bits. The CLKPCE
[11:41:56] <JyZyXEL> bit is only updated when the other bits in CLKPR are simultaniosly written to zero. CLKPCE is
[11:41:58] <JyZyXEL> cleared by hardware four cycles after it is written or when the CLKPS bits are written. Rewriting
[11:42:00] <JyZyXEL> the CLKPCE bit within this time-out period does neither extend the time-out period, nor clear the
[11:42:02] <JyZyXEL> CLKPCE bit.
[11:42:10] <OndraSterver> RikusW, ah
[11:42:12] <RikusW> JyZyXEL: you need TWO assignments
[11:42:13] <OndraSterver> (just came from the kitchen
[11:42:29] <JyZyXEL> it says simultaniosly
[11:42:32] <RikusW> iirc thats at 100x
[11:43:16] <JyZyXEL> wait i want to turn it off
[11:43:27] <JyZyXEL> or wait what the hell do i want it off or on
[11:43:35] <RikusW> JyZyXEL: still you need CLKPR = (1<<CLKPCE); CLKPR = 0;
[11:43:52] <JyZyXEL> why do i want it on in the first place?
[11:45:16] <JyZyXEL> meh im really confused ill just take your word for it
[11:45:22] <RikusW> you want it off...
[11:45:27] <RikusW> /1
[11:47:26] <JyZyXEL> ohh ok se were actually setting the value to /1 and its not just a on/off thingy
[11:48:15] <RikusW> yep
[11:48:27] <RikusW> there are a few different settings
[11:49:45] <JyZyXEL> haha yay now my leds blink fast
[11:50:24] <RikusW> http://imagebin.org/232857 10um micrometer slide
[12:07:24] <OndraSterver> one has to love Robot Chicken
[12:30:41] <RikusW> http://www.theregister.co.uk/2012/09/14/internet_antidote/
[12:30:51] <OndraSterver> cyanide?
[12:33:19] <RikusW> nah H2O
[12:35:04] <OndraSterver> raw H2O?
[12:35:10] <OndraSterver> maybe add some CH
[12:35:18] <OndraSterver> (ethanol is **CH right?)
[12:46:15] <RikusW> its H CH2 COOH
[12:46:22] <RikusW> err formic acid
[12:46:35] <RikusW> err acetic...
[12:46:45] <RikusW> H COOH == formic
[12:46:52] <OndraSterver> alcohol
[12:46:54] <OndraSterver> vodka, tequilla, anything
[12:47:00] <OndraSterver> is better than H2O
[12:47:07] <RikusW> H CH2 CH2 OH
[12:47:08] <OndraSterver> as an internet antidote.
[12:47:24] <RikusW> digging a well is more fun :)
[12:47:42] <RikusW> especially when using heavy tools
[12:47:51] <OndraSterver> well
[12:47:59] <OndraSterver> I always wanted to drive a mown lawer
[12:48:00] <OndraSterver> ..
[12:48:15] <RikusW> methanol is a permanent antidote H CH2 OH
[12:48:47] <OndraSterver> I know what methanol is
[12:48:50] <RikusW> lawnmower == lightweight tools :-P
[12:48:55] <OndraSterver> search the internet for "czech republic methanol"
[12:49:20] <RikusW> methanol is also used to fix blood smears...
[12:49:31] * RikusW did some research on lab stuff
[12:49:48] <RikusW> fix == dehydrate
[13:01:42] <megal0maniac> Ba Co N
[13:01:44] <megal0maniac> :)
[13:03:19] <megal0maniac> The lawnmower they use for crops is NOT lightweight
[13:08:46] <RikusW> thats now called a lawnmower... :-P
[13:08:52] <RikusW> *not
[13:09:07] <RikusW> more like a slasher
[13:10:11] <megal0maniac> Lawnmower, cropmower, smasher. Meh :)
[13:14:17] <RikusW> harvester
[13:14:30] <RikusW> plough, ripper
[13:14:32] <RikusW> disc
[13:15:35] <RikusW> megal0maniac: http://images.google.com/search?num=10&hl=en&site=&tbm=isch&source=hp&biw=1024&bih=686&q=agricultural+disc&oq=agricultural+disc&gs_l=img.3..0j0i24l9.2199.7778.0.8032.19.13.2.4.6.1.427.2440.1j4j4j0j2.11.0...0.0...1ac.1.jca3BToIwMM
[13:31:01] <megal0maniac> Dit sal jou kop afsny :D
[13:32:13] <OndraSterver> NO.
[13:32:32] <tld> Any thoughts on using 0805 SMD parts on a 0.1" protoboard? (plated holes type)… would it work well to simply dump it between two of the plated holes?
[13:32:47] <OndraSterver> no thoughts
[13:32:51] <prpplague> tld: generally works fine
[13:32:51] <OndraSterver> IF IT FITS, I SITS!
[13:32:56] <tld> thanks. :)
[13:33:04] <prpplague> tld: just make sure you don't get the proto's that are pre-wired
[13:33:40] <tld> yeah, have the board already, just wondering starting to use SMD parts instead (some are hard to find as throughhole, and I'm guessing I'll be moving over to SMD anyways, so better start stocking up)
[13:33:41] <tld> brb
[13:34:59] <prpplague> tld: i use these extensively for prototyping - http://beldynsys.com/
[13:35:20] <prpplague> tld: they are available directly from the website or from many distributors such as jameco and digikey
[13:36:08] <RikusW> megal0maniac: if a disc goes over you there won't be much left....
[13:36:42] <RikusW> we have one of them
[13:37:08] <RikusW> its very very heavy
[13:51:14] <tld> prpplague: hmm, thanks for the tip.
[13:51:28] <tld> prpplague: I've been using futurelec for that kind of thing, which is cheaper but not one bit reliable.
[13:51:29] <tld> :/
[13:53:48] <megal0maniac> tld: I did exactly that
[13:53:55] <megal0maniac> (0805 on protoboard)
[13:54:31] <tld> megal0maniac: worked well?
[13:54:37] <megal0maniac> http://i.imgur.com/Z1DEV.jpg
[13:54:51] <megal0maniac> http://i.imgur.com/4MTgA.jpg
[13:55:02] <megal0maniac> Seemed to work best between the holes
[13:55:07] <megal0maniac> SOT23 on there too :)
[13:55:19] <tld> hmm
[13:55:23] <tld> doesn't look too bad.
[13:55:30] <tld> size-wise it seems like a good fit.
[13:55:42] <megal0maniac> It does fit nicely
[13:56:01] <megal0maniac> Mind you, I didn't do such a great job. Finest soldering I've ever had to do. But if I can do it...
[13:56:12] <tld> (reason I'm considering it btw, is that higher quality ceramics, in higher capacities, seems to only be easily available as 0805 (an by easier, I mean eBay))
[13:56:38] <megal0maniac> I say go for it
[13:56:42] <tld> some of the joints doesn't look entirely perfect, but you get so much solder with strips… I'll be using strip-less boards, with just the holes, which I *think* should be easier.
[13:56:44] <Thetawaves> megal0maniac, why don't you just get some pcbs made?
[13:56:47] <tld> thanks guys. :)
[13:57:08] <megal0maniac> Thetawaves: Was a once off, and a personal challenge :)
[13:57:15] <megal0maniac> Also it's really expensive here
[13:57:26] <megal0maniac> And I suck at designing PCBs
[13:57:27] <Thetawaves> where? the internet?
[13:57:29] <tld> cost seems to be a common factor.
[13:57:38] <tld> for my sake, it's also a lot about speed.
[13:57:47] <tld> PCBs at okay prices usually takes about 6 weeks.
[13:57:58] <megal0maniac> Thetawaves: I already had the veroboard. Didn't need to design or print or fetch or pay
[13:58:02] <tld> most of what I do now, is smaller stuff, so I can begin and finish a project in about a week or even weekend.
[13:58:14] <Thetawaves> my designs would take a thousand hours to wire by hand
[13:58:19] <Thetawaves> fuck that
[13:58:39] <megal0maniac> Thetawaves: Have you counted the number of components on that board? :P
[13:58:40] <Thetawaves> pcbs save tons of time
[13:59:03] <megal0maniac> Not with that
[13:59:13] <tld> Thetawaves: I hope I'll end with that myself, just not there yet. ;)
[13:59:29] <megal0maniac> I started at 20H00, working on and off, and finished by 01H00.
[13:59:51] <megal0maniac> Can't even place an order in that time :P
[14:00:12] <tld> 1206 should be even easier than 0805, right? (and still not overshoot size of two 0.1" spaced holes?)
[14:00:34] <megal0maniac> tld: I don't see a reason not to use 0805
[14:00:41] <megal0maniac> But 1206 *should* work
[14:00:53] <Thetawaves> 0805 is too big
[14:00:57] <Thetawaves> 0603!
[14:01:01] <megal0maniac> 0402?
[14:01:03] <megal0maniac> No
[14:01:08] <megal0maniac> 1206 is fine :)
[14:01:29] <tld> the capacitor I'm looking for is only available as 10V at 0805, but 16V as 1206.
[14:01:43] <megal0maniac> Then 1206
[14:01:45] <megal0maniac> Boom
[14:01:56] <Thetawaves> find a different capacitor?
[14:01:57] <tld> $2.70 for 50pcs, including shipping… why am I even worried about this?
[14:01:58] <tld> thanks. :)
[14:02:18] <megal0maniac> Gotta love China
[14:03:33] <megal0maniac> Home time, yay
[14:03:36] <megal0maniac> 'Night
[14:03:50] <tld> 'night
[14:03:56] <tld> (and thanks again)
[15:27:56] <megal0maniac_afk> Got my chinese stuff. Accelerometer and 2x nrf24l01+
[15:29:10] <megal0maniac> All looks good. Wonder if it works...
[15:30:13] * RikusaWay got 4x nrf boards
[15:30:50] <RikusW> will have to write some code for that sometime, I'll pass it along ;)
[15:31:36] <Tom_itx> nice scope RikusW
[15:31:45] <RikusW> thanks
[15:32:12] <megal0maniac> RikusW: That would be cool. I still have no idea what I'm doing :P
[15:32:29] <Tom_itx> RikusW, is it for soldering?
[15:32:31] <megal0maniac> But I like to portray the image that I know some things...
[15:32:57] <Tom_itx> :)
[15:32:58] <megal0maniac> Tom_itx: The flyer says it's for looking at leaves :)
[15:32:58] <RikusW> Tom_itx: the widefield ? I don't have that, but would it too :)
[15:33:12] <Tom_itx> i didn't read the specs
[15:33:28] <Tom_itx> i like mine
[15:33:35] <RikusW> the one on google images is the one I have
[15:33:36] <Tom_itx> it's a 7-45 zoom
[15:33:39] <RikusW> the tinyurl link
[15:34:12] <RikusW> somewhere in todays log
[15:34:15] <megal0maniac> Tom_itx: I have a 10x loupe, which gets in the way, and I'm always scared I'm going to touch it with my iron
[15:34:38] <megal0maniac> Also, my tweezers have yet to arrive. But I hear that this is DX tradition
[15:34:40] <Tom_itx> i have one of those as well but if i need that close work, i get the scope out
[15:34:49] <RikusW> Tom_itx: its too high magnification and to narrow to be of use for soldering
[15:34:52] <Tom_itx> megal0maniac, about 4 weeks usually
[15:34:59] <Tom_itx> oh
[15:35:09] <Tom_itx> i just figured that's why you got it
[15:35:40] <RikusW> its like 4 10 40x objectives and 10 & 16 eyepieces
[15:35:51] <RikusW> so max 640x now
[15:36:12] <RikusW> the 40x needs to be <1mm from the object to focus
[15:37:08] <RikusW> and can see like 250um wide
[15:37:51] * RikusW wants a 100x oil objective, but thats like ZAR500....
[15:38:00] <RikusW> maybe I'll get it anyways sometime
[15:38:19] <RikusW> the proper lenses are more like ZAR10 000
[15:39:12] <RikusW> Tom_itx: I might get a widefield on sometime
[15:39:27] <RikusW> don't really have a use rigth now
[15:40:03] <RikusW> I did use the 4x10 to identify some smd components, but had to supply light on the top
[15:40:52] <Tom_itx> i still use the ring light from my desk magnifier for light for mine
[15:41:01] <Tom_itx> works out ok since it's all on the desk anyway
[15:41:49] <RikusW> I don't think the lenses are going to like the flux smoke too much... (aka smux :-P )
[15:43:30] <Tom_itx> i put a filter on the end of mine so i can remove it and keep it clean
[15:44:03] <Tom_itx> so the lenses are protected
[15:48:30] <JViz> !seen abcminiuser
[15:48:30] <tobbor> abcminiuser was last seen in #avr on Oct 17 13:24 2012
[15:49:05] <Tom_itx> he's a lonely aussie in a foreign land now
[15:49:23] <megal0maniac> A busy one, though
[15:49:33] <megal0maniac> (Or so it seems)
[15:49:43] <Tom_itx> his isp sucks there
[15:50:08] <megal0maniac> From what I could gather, his ISP is his neighbour's wireless :P
[15:50:17] <Tom_itx> heh could be
[15:51:25] <megal0maniac> Except his "neighbour" is probably his landlord, and he pays for it
[15:52:45] <Tom_itx> that's the good thing about this neighborhood, if my router quits i can always connect to someone else's
[15:53:09] <Tom_itx> i had to get a better wifi because of all the traffic
[15:53:26] <megal0maniac> Heh.
[15:53:33] <Tom_itx> unifi
[15:53:51] <Tom_itx> i don't use it on my desk stuff
[15:53:56] <Tom_itx> but all the laptops are
[15:54:20] <megal0maniac> We used to have 2 seperate networks in the house, but now I've set them up as redundant connections for each other. Makes things like network printers less of a mission
[15:54:34] <megal0maniac> And only half the network needs to be up at any given time
[15:55:48] <JViz> why does your network go down so much you need two of them?
[15:56:24] <megal0maniac> Because *someone* has a tendancy to switch the one router off
[15:56:46] <JViz> move the router?
[15:56:48] <megal0maniac> I don't know why, but if I don't know about it, it won't annoy me
[15:57:05] <megal0maniac> The other router is the primary, and it's in my room :)
[15:57:43] <JViz> so then why does your network go down so often you need two of them?
[15:59:35] <megal0maniac> Because human error
[16:02:15] <RikusaWay> Tom_itx: good idea to protect the lenses
[16:02:33] <Tom_itx> it was a UV or some clear lense
[16:04:02] <OndraSterver> Tom_itx, do you ESD suppress only to gnd, but also to vcc and gnd-vcc too?
[16:04:17] <OndraSterver> I am now penny pinching because I need to order 25 xmegas to get usable price
[16:04:20] <OndraSterver> 25 * 3.89 EUR
[16:04:21] <OndraSterver> + 20% tax
[16:04:23] <OndraSterver> :(
[16:04:23] <Tom_itx> gnd only
[16:04:30] <OndraSterver> mkay
[16:04:40] <megal0maniac> Which xmegas?
[16:04:43] <megal0maniac> 256a3u
[16:04:44] <megal0maniac> ?
[16:04:44] <Tom_itx> never heard of any problems
[16:04:54] <OndraSterver> megal0maniac, yes
[16:05:08] <OndraSterver> + 25* 32a4u aka 25 * 1,89 eur + 20% tax
[16:05:24] <megal0maniac> 20%?
[16:05:26] <megal0maniac> Shame!
[16:05:31] <OndraSterver> heh
[16:05:32] <Tom_itx> megal0maniac, OndraSterver is an xmaniac
[16:05:33] <OndraSterver> is it not enough?
[16:05:40] <megal0maniac> Americans think that our 14% is high
[16:05:43] <Tom_itx> get a whole tray of them
[16:05:49] <OndraSterver> lol
[16:05:56] <Tom_itx> better prices
[16:06:00] <OndraSterver> Tom_itx, did you get them off digikey or from atmel themselves?
[16:06:08] <Tom_itx> DK i think
[16:06:13] <OndraSterver> ah
[16:06:20] <OndraSterver> because IIRC digikey and mouser have prices only upto 100
[16:06:26] <OndraSterver> no 250/500 options anymore :P
[16:06:33] <Tom_itx> really?
[16:06:49] <Tom_itx> i need to get a few things one of these days
[16:07:44] <OndraSterver> I would go for kickstarter but..
[16:07:49] <OndraSterver> NO KICKSTARTER PROJECTS OUTSIDE THE US!
[16:07:57] <OndraSterver> http://www.digikey.com/product-detail/en/ATXMEGA256A3U-MH/ATXMEGA256A3U-MH-ND/3046478
[16:08:01] <OndraSterver> digikey 1/25/100
[16:08:04] <OndraSterver> mouser 1/10/25/100
[16:08:05] <Tom_itx> move
[16:08:17] <OndraSterver> if I were to move I would probably choose the UK
[16:08:20] <creep> h
[16:08:24] <creep> sup dudes
[16:08:33] <Tom_itx> i actually might too
[16:08:59] <OndraSterver> hmm?
[16:09:01] <OndraSterver> to the UK?
[16:09:27] <Tom_itx> if i ever did
[16:09:36] <OndraSterver> oh
[16:09:42] <OndraSterver> any reasons why US > UK?
[16:09:45] <OndraSterver> other than kickstarter
[16:10:02] <Tom_itx> it's probably no better there
[16:13:47] <OndraSterver> indiegogo and rockethub should work outside the US
[18:41:31] <JViz> !seen abcminiuser
[19:53:06] <JoeLlama> hi ninwa :)
[19:53:19] <JoeLlama> does anyone here know about atmel processors? :P
[19:53:55] <JoeLlama> it's been quiet ninwa
[19:55:54] <Horologium> yes.
[19:55:57] <Horologium> many of us do.
[19:56:05] <Horologium> what atmel processors specifically?
[20:06:10] <Casper> JoeLlama: and you mean microcontroller? or microprocessor?
[20:10:43] <JoeLlama> both of course :) sure why not
[20:11:03] <JoeLlama> I think ninwa fell asleep :(
[20:52:04] <creep> can somebody give a ATX 12V 2.3 standard pdf ?
[21:03:38] <Tom_itx> that made no sense
[21:05:57] <_nin> JoeLlama, hehe sorry I didn't even notice
[21:06:40] <_nin> I was wondering if anyone knew if it were possible to reporgram the avr butterfly just using the onboard 3v battery, a guide I saw mentions an external source. The chip is programmable through serial alone, I can't figure out how though. The guide shows atmel studio 4 which is no longer available as far as I can tell
[21:08:00] <_nin> heres the guide showing atmel studio 4: http://www.siwawi.arubi.uni-kl.de/avr_projects/pbfi.pdf -- here is the atmel user manual: http://www.atmel.com/Images/doc4271.pdf
[21:08:26] <_nin> in 3.7 of the user manual it describs how to pin the serial to its on board leveler, was wondering if anyone has used this with atmel studio 6 before
[22:07:41] <SuperMiguel> so i have a gyro that gives me x and y... how do i get the angle from it??
[22:21:15] <Casper> SuperMiguel: check the datasheet to see what is the output
[22:21:57] <Casper> it vary from "-max is max angle and +max is the other side max angle" to "calibrate it for every angle"
[22:21:59] <SuperMiguel> Casper, ya the thing barely has a datasheet
[22:22:25] <SuperMiguel> http://support.robotis.com/en/product/auxdevice/sensor/dxl_gyro.htm
[22:22:41] <SuperMiguel> so it says 455 Angular velocity 300+
[22:24:04] <Casper> dosen'T giro tell you not the angle but the speed of rotation?
[22:24:15] <SuperMiguel> correct
[22:24:24] <SuperMiguel> but is there a way to conver that into angle?
[22:24:34] <Casper> yes and no
[22:24:46] <Casper> there is some drift
[22:25:20] <SuperMiguel> ω = θ / t can we use that?
[22:26:14] <Casper> from what I understand, 1.23V is "there is no change", and 2.23V is "it rotate by 300 degree per second" I can't say if it's left or right
[22:27:02] <Casper> but basically, it would be: read the speed of rotation, calculate the time, do the math, add that to the current angle
[22:27:36] <SuperMiguel> ok read speed... then find out what time>
[22:29:19] <Casper> but here is the issue: since it's analog and low quality, there can be quite a big error
[22:29:32] <Casper> if you read 150 degree/sec, it can really be 140 degree
[22:29:43] <SuperMiguel> so how people use this things?
[22:30:05] <Casper> probably in conjunction with an accelerometer
[22:30:26] <SuperMiguel> what will an accelerometer give me that this doesnt?
[22:31:13] <Casper> an accelerometer give you the gravity direction, which, if there is no speed change, can give you the exact vector where the gravity point
[22:31:38] <Casper> integrate that with a gyro and you can calculate more preciselly the angle
[22:31:38] <Casper> also
[22:31:42] <Casper> think RC airplane
[22:32:06] <Casper> they use gyro to help against suddent rotation (wind gust for example)
[22:32:26] <Casper> it basically do nothing unless there is a big change
[22:32:54] <SuperMiguel> big change for a bit of time
[22:32:57] <Casper> also, since the remote control do have a setting for calibrating the rudders, it somewhat kill the error
[22:33:58] <Casper> let's say for example that your gyro output 1.24V instead of it's 1.23V for no change. the remote could output 1.22V, which get mixed with the gyro, and you get 1.23V, or no change
[22:34:23] <SuperMiguel> gotcha
[22:34:25] <Casper> and the rudders get straight, and the plane stay level, and the pilot do the tiny correction from time to time
[22:34:50] <SuperMiguel> i was trying to use one on a robot, to know how flat compared to the ground it was
[22:35:03] <Casper> acceletometer would be a better choice
[22:35:32] <Casper> it tell you the G force in the 3 axes, some math and you get where the G-force is
[22:35:55] <Casper> that is actually what the wiimote and the playstation wang use
[22:36:59] <Casper> too bad that nintendo used a cheapie on the wiimote V1... it was easy to max it... causing massive issues in some game
[22:37:10] <Casper> because the programmers was too stupid to autocorrect
[22:37:28] <SuperMiguel> lol
[22:38:02] <Casper> one game was quite fun, a sword one
[22:38:25] <Casper> it was tracking the wiimote movement to make it match in the game
[22:38:26] <SuperMiguel> there are a few fun games in it
[22:38:39] <Casper> sadly, when you max the value, there is some desync
[22:39:07] <Casper> and then you go back to "home" position and the sword in the game point.... somewhere...
[22:39:14] <Casper> then you have to press a button to reset
[22:39:14] <SuperMiguel> haha
[22:39:22] <Casper> so the game goes like...
[22:39:49] <Casper> slash slash slash slash fuck home button hit slash slash fuck home hit button...
[22:40:01] <hackvana> barrell roll!
[22:40:15] <Casper> http://www.rigolna.com/products/waveform-generators/dg1000/dg1022/ ← I want that
[22:41:28] <SuperMiguel> 379 thats not bad
[22:41:34] <jadew> casper, saw that on e-bay for a bit cheapper
[22:41:57] <Casper> jadew: yeah I saw too, but sadly, the customs here are theif
[22:42:08] <jadew> ah
[22:42:25] <Casper> and I bet that they would also charge the eco fees and the highest custom fee possible
[22:42:30] <SuperMiguel> where
[22:42:36] <jadew> so far they didn't charge me for anything that I bought from china, kinda strage
[22:42:40] <Casper> the eco fee seems to be the mafia...
[22:43:01] <jadew> the eco fee in here is like... $1 or less
[22:43:03] <Casper> yeah china usually declare them as gift, or very low value
[22:43:05] <Casper> SuperMiguel: canada
[22:44:42] <jadew> was actually thinking earlier about making another function generator
[22:44:59] <jadew> wonder what't he most powerful cpu I could use
[22:45:07] <jadew> *what's the
[22:46:53] <jadew> it would be interresting to try something with a nano-itx board
[22:47:40] <Casper> actually, I don't think that it need a fast cpu at all
[22:48:08] <Casper> you just need a fast DAC that you can connect to a RAM and tell it what range to play at what speed
[22:48:11] <jadew> well, unless you have a device made by the factory to do exactly that, I don't know how else you can do it
[22:48:26] <jadew> hmm
[22:49:14] <Casper> then when you change the waveform, you only have to fill the ram buffer, then issue the start command
[22:49:34] <jadew> yeah, that sounds doable
[22:49:46] <jadew> but what dacs can work with ram and do that?
[22:50:26] <Casper> that I do not know, but those for VGA do it
[22:50:59] <jadew> hmm, so I might actually have one around eh?
[22:51:06] <Casper> hmmm I wonder... RAMDAC?
[22:51:12] * Xark notes you can't really buy RAMDACs anymore (that I've found)...
[22:51:19] <Casper> that tells me something... isn't it what we want?
[22:52:05] <jadew> oh man, I need to dig trough this stuff
[22:52:21] <jadew> *trough my stuff
[22:52:31] <jadew> but what speed would one of those things have?
[22:52:47] <jadew> let's see... 1024 * 768
[22:52:59] <Casper> I remember that it was saying 350MHz RAMDAC back in the old day of dos
[22:53:13] <Casper> and remember, it also have 3 channels
[22:53:22] <jadew> 1024 * 768 * 60 = 47 185 920
[22:53:47] <Casper> most can go in the 25xx by whatever
[22:54:50] <jadew> yeah, you probably need a lot more than 47Mhz to get something awesome
[22:55:19] <jadew> you probably need at least 10 points for the edge of a sine wave
[22:56:02] <Xark> From my list 640x480@60Hz uses 25.175Mhz pixel clock. 1024x768@75Hz is 78.8Mhz pixel clock. 1600x1200@85Hz is 202.5Mhz pixel clock.
[22:56:05] <Casper> "As of 2006, the DAC of a modern graphics card runs at a clock rate of 400 MHz."
[22:56:48] <jadew> interresting
[22:56:55] <Casper> but I do not think it would work well... because... I think it's a fixed size
[22:57:05] <Casper> you can't select how many sample to use
[22:57:15] <Xark> Casper: That doesn't matter I don't think.
[22:57:21] <Casper> but I bet it's how they actually do
[22:57:47] <Casper> Xark: it does matter that it's fixed length...
[22:57:55] <Xark> Casper: Why?
[22:58:09] <Casper> and VGA one could possibly blank during the normal beam return time
[22:58:15] <Xark> Casper: You mean 256 "samples"?
[22:58:59] <Casper> Xark: let's simplify it, let's say it have like you say 256 samples buffer... and your waveform actually have one cycle per 200 samples... what will happend?
[22:59:03] <Xark> Casper: Well, if you are discussing a RAMDAC, I think it is like a RAM chip that outputs an analog voltage (it doesn't control timing AFAIK - other than having an upper limit of speed).
[22:59:47] <Casper> you would normally tell the DAC to "reset" at the 200th sample, so it goes back to the begening
[22:59:54] <Casper> I do not think that it can do that
[23:00:06] <jadew> ah, it should be able to do it
[23:00:14] <jadew> since differnet video modes use up different memory space
[23:00:15] <Casper> so you would have 56 samples of nothing or garbage
[23:00:21] <Xark> Casper: You double up samples to make it an even multiple. Same way I can have many different horizontal resolutions with the same RAMDAC.
[23:00:49] <Casper> Xark: but you can't always fit the sample in a frame...
[23:01:01] <Xark> Casper: Think of combinations of "colors" on a scanline as the same as a waveform per scan line.
[23:02:02] <Casper> really, it all depend how they did it internally, if you can vary the sampling speed to any one, then it's doable
[23:02:08] <Casper> if not then it won't work
[23:02:20] <Xark> Casper: I am pretty sure you can (think of all the graphics modes - each a different speed).
[23:03:09] <Xark> However, I looked I couldn't find any RAMDACs available (I have been building various software VGA generators of late and though one would be handy for a "palette").
[23:03:12] <Casper> depend how the ramdac is made
[23:03:32] <Casper> if it have it's own frequency generator or need an external one
[23:04:25] <Xark> Casper: A traditional RAMDAC is like a RAM chip with a DAC tacked on the output. So just as RAM chip doesn't include a clock, I don't think RAMDACs do either.
[23:05:23] <Casper> I don't think it does too, but who knows, with such specialised chip, they could have made it to simplify the life of the engineers
[23:05:52] <Casper> and I actually beleive it have it's own clock, because when the gpu freeze, it still generate a valid image
[23:05:59] <Xark> Casper: Well, this is why they are not available. Nowadays, they make a "specialized" GPU that includes all this (and why lone RAMDACs are extinct).
[23:06:32] <jadew> so how can we build an awesome function generator that does 100Mhz+ ? :D
[23:06:44] <Xark> Casper: A RAMDAC is not a GPU, it is an individual component that (used to) be used to make graphics cards.
[23:07:44] <jadew> you know what's confusing me
[23:07:45] <Xark> jadew: Well, a quick google shows IDT DAC chips that (e.g.) can do 16-bits, 2 channels at 1250Msps (but I bet they are not cheap).
[23:07:48] <jadew> this specs: 100MHz sine and 62.5MHz square waves
[23:07:56] <jadew> why slower freq. square waves?
[23:08:01] <Casper> Xark: I'm not stupid, and that ain'T what I said
[23:08:45] <Casper> I said that when a GPU crash, the DAC still spit out the valid data
[23:08:48] <Xark> Casper: I certainly don't mean to imply that, but I just haven't heard of a RAMDAC that did its own clocking. No offense intended.
[23:09:16] * Xark notes he is not a graphics hardware engineer either (but has been looking into VGA generation recently).
[23:09:17] <Casper> which make me beleive that it could have it's own clock source, that or the GPU clock never crash
[23:11:41] <Xark> Casper: Ahh I see your point , but I would suspect that parts of the GPU keep going after higher level parts crash (and "scanout" is pretty low level).
[23:12:24] <Xark> jadew: In nature there are only sine waves. :)
[23:12:51] <jadew> Xark, yeah, but I was expecting them to use a DAC
[23:12:55] <Casper> hmmm there is also some audio RAMDAC btw
[23:12:59] <Casper> but slow
[23:13:02] <jadew> that would suggest that they are indeed generating sine waves and modulating them
[23:13:31] <jadew> http://www.alibaba.com/product-gs/500593591/NXP_ORIGINAL_IC_DAC1627D1G25.html
[23:14:19] <jadew> what's the price on that, cuz I don't know what that notation means
[23:14:23] <jadew> is it 0.1 / 100?
[23:14:34] <jadew> that must be a mistake
[23:14:45] <jadew> I mean 0.1 when buying more
[23:14:51] <jadew> but it's still way too cheap to be right
[23:14:53] <Xark> http://www.idt.com/products/data-converters/high-speed-dac
[23:15:20] <Casper> jadew: probably 0.1 in 100 quantity
[23:15:32] <jadew> that's damn cheap
[23:15:46] <jadew> so for $10 you get 100 of those...
[23:18:02] * Xark notes that NXP sold its DAC division to IDT (he got to his link starting at NXP)...
[23:19:57] <jadew> The LVDS DDR receiver interface allows a high data bandwidth (312.5 MS/sec) at the input.
[23:20:02] <jadew> so you're limited by that as well
[23:20:51] <jadew> that would be good for what... a 15Mhz sine wave?
[23:21:19] <Casper> crap... I bought an inverter...
[23:21:31] <Casper> ... to find that it have a soft power switch on it...
[23:21:38] <Xark> jadew: Isn't that 312.5 mega-samples per second?
[23:21:41] <Casper> I wonder if I'm gonna keep it or not
[23:22:01] <jadew> Xark, yeah but you need lots of samples to create one edge of a signal
[23:22:26] <jadew> Casper, can't you replace it?
[23:23:00] <Casper> no, I wanted no power switch, or an hardware one...
[23:23:17] <Casper> so I could mount the inverter in the trunk...
[23:23:21] <Xark> jadew: Right. That is pretty fast though (every waveform looks like sine at that speed).
[23:25:17] <jadew> guess I'd better think of ways to improve the one I already have
[23:27:37] <Casper> the more I read about the eco fees, the more I seems to find a loophole...
[23:27:56] <Casper> computer should have a 7.50$ ecofee on it
[23:28:25] <Casper> but it's the computer conceptor that must collect it
[23:28:35] <Casper> the one that put the computer on the market
[23:28:54] <Casper> we, at our store, assemble a computer from different parts
[23:29:09] <hackvana> Usually a local distributor, which doesn't make much sense in a global online marketplace.
[23:29:18] <Casper> I think that there is no ecofee on clone pc
[23:29:30] * hackvana wants e-coffee
[23:30:34] * Casper wants a working dc ups
[23:30:59] <Casper> with hope, there will be almost no work today at work, so I'll be able to do some proper research