**** BEGIN LOGGING AT Tue Feb 26 02:59:58 2013 Feb 26 03:08:22 wait Feb 26 03:08:32 someone made a smartphone called a 'keon'? Feb 26 03:11:35 ok genius folks Feb 26 03:54:18 woot! pancake tube here... Feb 26 03:59:01 that sounded dirty. Feb 26 03:59:57 emeb_mac: it is like FPGAs to you Feb 26 04:00:26 FPGAs don't sound dirty. Feb 26 04:15:51 gah, darktable upload to g+ is slow Feb 26 04:17:02 ds2 : fuck the BIOS Feb 26 04:27:35 anyone here know much about heat sinks Feb 26 04:27:45 we don't like those here Feb 26 05:09:02 sakoman! Feb 26 05:09:35 hey mranostay! Feb 26 05:16:03 sakoman rises! Feb 26 05:17:08 from the dead? Feb 26 05:27:11 hey guys, i have a question if anyone is available. I'm cross-compiling a program using angstrom-2011.03-i686-linux-armv7a-linux-gnueabi-toolchain for my xM rev C, but when I try to execute it I get a 'cannot execute binary file' Feb 26 05:27:52 uname -m gives arm7l, and it's ubuntu 12.04 on the xM and on my laptop Feb 26 05:28:55 I tried readelf -h on a program compiled locally and a cross compiled program, and the only difference I notice is Version5 EABI vs Version4 EABI. any ideas what the issue is? Feb 26 05:33:04 is that the compiler from the angstrom you are running on the xm? Feb 26 05:36:44 the xM is running Ubuntu 12.04, and g++ --version gives g++ (Ubuntu/Linaro 4.6.3-lubuntu5) 4.6.3 Feb 26 05:37:52 the laptop is running Mint and arm-angstrom-linux-gnueabi-g++ --version gives "arm-angstrom-linux-gnueabi-g++ (GCC) 4.3.3" Feb 26 05:44:37 why not use the linaro gcc? Feb 26 05:45:20 you mean why not compile locally on the xM, or why not cross-compile using the linaro compiler? Feb 26 05:46:40 if the first, then because it's too slow. if the second, I would definitely try that if it is available somewhere for cross-compilation Feb 26 05:54:06 why don't you just install the ubuntu arm cross compiler? Feb 26 05:54:10 that should match Feb 26 05:54:54 http://packages.ubuntu.com/search?keywords=gcc-arm Feb 26 05:57:07 ok, i'm on it - thanks Feb 26 06:15:35 <_av500_> ohai Feb 26 06:16:40 _av500_: hey big boy Feb 26 06:36:57 koen: ping Feb 26 06:37:02 okay, I installed arm-linux-gnueabihf-g++ and recompiled everything Feb 26 06:38:02 I was able to get a 'hello world' program to execute Feb 26 06:38:56 however, I cross-compiled my actual program (which links to a shared library that I also cross-compiled), and I still get the 'cannot execute ' Feb 26 06:42:50 the output of readelf -h for the hello world program, the actual program, and the shared library all give Version5 EABI and ELF32 ARM and Unix System V Feb 26 06:52:16 hayk: missing libraries? try strace? Feb 26 06:52:53 gm Feb 26 06:54:08 can you give an example command? I have a .so library and an executable Feb 26 06:54:15 i thought missing libraries would be printed out Feb 26 06:58:55 try 'ldd' on the binary? Feb 26 07:00:53 ldd works on the hello world program but terminates 1 on the other Feb 26 07:00:57 no output Feb 26 07:02:10 strace has a bunch of output for both but also exits 1 for my program Feb 26 07:06:38 is MMC support working for the beagle bone in v3.8? Feb 26 08:48:17 test Feb 26 08:48:25 is there anybody here??? Feb 26 08:49:11 no Feb 26 08:49:42 lol Feb 26 08:49:43 :D Feb 26 08:58:54 <_av500_> we all float down here Feb 26 09:29:56 anyone here knows how to deal with the "port 2 disabled by hub (EMI?), re-enabling " problem? Every now and then my ethernet disconnects and doesn't come back without restart Feb 26 09:30:18 I tried putting a cron task checking for connectivity and trying to reload the kernel module but that doesn't help Feb 26 09:45:48 hello, i'm having some trouble with my beaglebone console access Feb 26 09:58:11 hi, i'm having some trouble with my textbook on general relativity Feb 26 09:59:21 <_av500_> KotH: you mean, you cannot measure the speed and the mass of the book at the same time? Feb 26 10:01:28 yes, the book seems to be smeared over the whole wall after i tried to accelerate it to half of the speed of light Feb 26 10:12:59 funny guys? Feb 26 10:14:25 nevermind... i just discovered this SD card doesn't show me the uboot console and just boots right to armstrong, very quietly. or at least, that's my guess... i'm compiling u-boot right now... Feb 26 10:15:01 ezequiel: well... you stated you have a problem, not telling us what the problem is... but you still expect a usefull answer? Feb 26 10:15:27 ezequiel: welll... 42, then... or 137 if you are rather into physics Feb 26 10:15:55 yes, i know. perhaps Feb 26 10:16:15 i wasn't asking the smart way Feb 26 10:17:08 just to confirm: is the only way to boot a bone the SD card? there's no NAND or SPI? Feb 26 10:17:20 you werent asking at all Feb 26 10:17:46 * KotH points at the board manual and the TRM Feb 26 10:18:24 yeah, i looked at those... i'd just want a human confirmation, it at all possible Feb 26 10:18:30 s/it/if Feb 26 10:19:53 * KotH doesnt know the answer Feb 26 10:20:15 * KotH doesnt even have a beagle* Feb 26 10:21:09 i see, thanks anyway Feb 26 10:21:42 <_av500_> no nand Feb 26 10:21:44 <_av500_> no spi Feb 26 10:21:47 <_av500_> you can boot from SD Feb 26 10:21:53 <_av500_> or from serial/usb Feb 26 10:23:44 ok, thanks Feb 26 10:24:24 another question: has anyone tried the suspend/resume cycle? it seems broken here... and i've heard there are some issues with it Feb 26 10:33:17 mmm.. i just had to remove the "quiet" from uEnv.txt :S Feb 26 10:39:10 ezequiel: the cheese troll is messing with you Feb 26 10:40:06 no problem: i got my question answered, and actually i already solved my issue! i'm troll-proof buitl Feb 26 10:40:11 s/buitl/built Feb 26 10:40:35 the beer troll will be asleep Feb 26 10:40:41 moaning Feb 26 10:40:50 and the hobbits are almost goign to bed now Feb 26 10:42:58 the hobbit has to work on creating a fridge Feb 26 10:43:00 :-( Feb 26 10:45:10 <_av500_> buy lots of ice Feb 26 10:51:57 sigh. Feb 26 10:52:03 hugely annoying problem Feb 26 10:52:13 customer wants a totally sealed box Feb 26 10:52:18 for our electroincs Feb 26 10:52:43 problme is i have about 40W of heat to loose Feb 26 10:52:53 it will heat up! Feb 26 10:53:27 does "totally sealed" include thermally? Feb 26 10:53:42 and what about information? Feb 26 10:54:42 if the thing does anything observable at all, you'll be leaking entropy Feb 26 10:54:55 your only choice is to build it from dark matter Feb 26 10:55:21 and if it's totally sealed, no electrons should be going in or out Feb 26 10:55:29 or put a black hole in the box too Feb 26 10:55:38 dm8tbr: then things will still go _in_ Feb 26 10:55:51 dm8tbr: then things will still go _out_ Feb 26 10:56:05 (at least gamma ray radiations) Feb 26 10:58:17 or just steal steve's reality-distortion-field-generatorâ„¢ - after all he won't need it anymore, will he? Feb 26 11:00:50 only beer and cheese will be flowing Feb 26 11:01:33 "sealed" as in air tight Feb 26 11:02:09 oh, then it's easy Feb 26 11:02:16 just slap a huge heatsink on it Feb 26 11:02:27 one does never build air tight cases ... for these will be filled with water Feb 26 11:03:10 not if you fill it with dry nitrogent Feb 26 11:03:14 :-) Feb 26 11:03:23 fill it with epoxy Feb 26 11:03:32 mrpackethead: does not work Feb 26 11:04:00 makes rice last for ages Feb 26 11:04:03 :-) Feb 26 11:04:09 so say the prepers Feb 26 11:04:10 mru: we do that with electronics that we put into an atoclave Feb 26 11:04:38 mrpackethead: i guess the device will be used outdors? Feb 26 11:04:45 yup, nasty environmetn Feb 26 11:04:57 ambinet temps of up to 40C, near the sea Feb 26 11:05:09 then it's easy Feb 26 11:05:09 high humidity Feb 26 11:05:29 do tell Feb 26 11:05:31 but nearly constant temperature Feb 26 11:05:38 humid air conducts heat better Feb 26 11:05:45 so easier for you Feb 26 11:05:50 no, not constant at all over the year Feb 26 11:06:22 constant as in "doesnt change from -40°C to +50 in 2h" Feb 26 11:06:31 unlikely Feb 26 11:07:10 20-40C over the course of a year Feb 26 11:07:19 as i said, constant temperature Feb 26 11:07:35 southern coast of USA Feb 26 11:07:55 so, whats the easy solution sir? Feb 26 11:07:58 oh, hurricanes could be a problem there Feb 26 11:08:13 yes, a few do pass through Feb 26 11:08:57 metal case Feb 26 11:09:01 yes. Feb 26 11:09:11 the rest follows automatically Feb 26 11:09:31 would need to be a big case Feb 26 11:09:38 to get enough surface area Feb 26 11:09:43 to keep thigns in check Feb 26 11:09:58 space isnt that forthcoming Feb 26 11:10:01 12:02 < mru> just slap a huge heatsink on it Feb 26 11:10:10 mmm.. Feb 26 11:10:24 dude! do you even engineer? Feb 26 11:10:50 dude, do you Feb 26 11:11:07 no, i'm reading lwn Feb 26 11:12:07 if necessary, use heat pipes or some kind of active heat transfer Feb 26 11:12:39 40W isnt that much Feb 26 11:12:40 e.g. if the surface area of the box is too small to conduct the produced heat passively Feb 26 11:12:55 he didn't specify the size of the container Feb 26 11:13:05 yeah. Feb 26 11:13:07 40W in a 4 mm^3 box is a lot Feb 26 11:13:19 at 1K/W that's a mere 40°C over ambient Feb 26 11:13:20 bit bigger than that Feb 26 11:13:21 :) Feb 26 11:13:38 and getting to 1K/W is easy Feb 26 11:14:32 i'm not so keen to get 40over ambinet Feb 26 11:14:34 mru: uhmm.. IBM had a couple of years back (2-3.. not sure) a CPU dummy running at 2kW Feb 26 11:15:06 mru: IIRC they kept it's temp below 100°C using pressurised water trough the chip itself Feb 26 11:15:06 i'd really like to stay no higher than 60C actual Feb 26 11:15:11 there's always _some_ limit where passive transfer becomes insufficient Feb 26 11:15:30 and yes, I recall reading about that thing from ibm Feb 26 11:15:50 mru: do you also remember their goal: 7kW :) Feb 26 11:15:52 unisys had a mainframe that did the same thing Feb 26 11:16:13 mrpackethead: mainframes are in a different class Feb 26 11:16:16 liquid-cooled computers are nothing novel or exciting Feb 26 11:16:22 there's one under my desk Feb 26 11:16:24 mrpackethead: these produce the heat over many dies not one single Feb 26 11:16:27 1970's vintage Feb 26 11:17:30 oh.. then it's probably half-discrete build from precessors to the 74's Feb 26 11:17:57 lots of wire-wrapped boards Feb 26 11:18:10 * KotH doubts that Feb 26 11:18:27 dobuts what? Feb 26 11:19:33 i doubt they shipped anything to a customer made of wire-wrapped boards Feb 26 11:19:54 lol Feb 26 11:19:55 http://users.monash.edu.au/~ralphk/imgp3266-burroughs-bx350-board.jpg Feb 26 11:20:27 http://users.monash.edu.au/~ralphk/thumb.imgp2977-b7800-backplane.jpg Feb 26 11:23:49 it's not clear from the second pic whether that is wirewrap Feb 26 11:24:14 http://users.monash.edu.au/~ralphk/thumb.imgp2948-b7800-pcb.jpg Feb 26 11:24:29 it is Feb 26 11:24:49 borrughs b7800 Feb 26 11:25:02 thankfully i did'nt work on these Feb 26 11:25:24 * KotH wtfs Feb 26 11:25:26 it was just a pile of parts no one had the heart to throw out. Feb 26 11:25:51 how did they manage to get even one computer working? Feb 26 11:25:59 who says they did? Feb 26 11:26:05 right Feb 26 11:26:13 they did Feb 26 11:27:01 memory was hand threaded with tiny ferrite beads Feb 26 11:27:23 anyway.. Feb 26 11:27:37 AGC had solid state storage as well Feb 26 11:27:38 yes, but these were usually soldered Feb 26 11:27:39 closed box, passive cooling is unlikely to be sufficent. Feb 26 11:27:47 wirewrap is not soldered Feb 26 11:27:55 so stick a fan on the box Feb 26 11:28:12 what generates the heat? Feb 26 11:28:18 and how big is the box? Feb 26 11:28:22 yup. and let all the salt air! Feb 26 11:28:26 in the box Feb 26 11:28:30 no, no Feb 26 11:28:43 what generates the heat? please answer Feb 26 11:28:47 AC/DC powersupply Feb 26 11:28:50 mru: 1000s of 74xx logic chips Feb 26 11:28:58 jacekowski: the core memory i've seen sofar all were soldered at the ends Feb 26 11:29:16 jacekowski: and i have seen wire wrap boards that were additionally soldered (for obvious reasons) Feb 26 11:29:38 mrpackethead: PSU, nothing else? Feb 26 11:30:10 mrpackethead: with 40W of disipated power? what do you delvier? 1kW? Feb 26 11:30:23 some low power control electronics, ( .2A @ 5V ) Feb 26 11:30:30 if you are soldering wire wrap - you are doing it wrong Feb 26 11:30:32 this sounds very, very wrong Feb 26 11:30:56 40W is the the heat from the conversion of AC/DC Feb 26 11:30:56 indeed it sounds wrong Feb 26 11:31:02 jacekowski: yes, wire-wrap uses tension in the wires to maintain contact Feb 26 11:31:03 the rest of the energy leaves the ox Feb 26 11:31:07 mrpackethead: sounds wrong Feb 26 11:31:12 jacekowski: i have not seen anyone doing wirewrap at all Feb 26 11:31:16 what sounds wrong? Feb 26 11:31:26 AC->DC Switchmode Feb 26 11:31:27 mrpackethead: 40W losses in any kind of conversion like that Feb 26 11:31:29 mrpackethead: all of what you just said Feb 26 11:31:30 mrpackethead: yes Feb 26 11:31:37 jacekowski: only seen old boards nobody could tell whether they still worked or not ;) Feb 26 11:31:39 mrpackethead: switchmode PSUs are 90%+ efficient Feb 26 11:31:40 mrpackethead: what is the electrical power output of your PSU? Feb 26 11:31:58 jacekowski: not at 5V they are not Feb 26 11:32:03 mru: otherwise most countries have laws agains selling them Feb 26 11:32:08 mrpackethead: they are Feb 26 11:32:16 mrpackethead: otherwise most countries have laws agains selling them Feb 26 11:32:24 sigh. Feb 26 11:32:46 tell us the fucking specs of your psu already! Feb 26 11:32:56 mrpackethead: for reference: we build earlier this year a PSU from 3-phase 400V down to 24V with an efficiency of >92% from 10-5000mA Feb 26 11:32:59 go and pull up mouser or digikey, and serach on AC/DC convertsrs Feb 26 11:33:02 mrpackethead: power supply can't be legally sold in europe if it doesn't have at least 80% efficiency Feb 26 11:33:07 do you have a part number? Feb 26 11:33:28 @24V i belive you Feb 26 11:33:31 jacekowski: only consumer equipment Feb 26 11:33:33 try doing it at 5V Feb 26 11:33:42 mrpackethead: still, easy Feb 26 11:33:56 mrpackethead: 24V->5V you said? Feb 26 11:33:59 mrpackethead: if you are running low power elecronics with just 200mA @ 5V and have an input power of 40W, then you are doing it very wrong Feb 26 11:34:09 indeed Feb 26 11:34:12 koth, i din't say that Feb 26 11:34:18 mrpackethead: you said that above Feb 26 11:34:21 i said, i have 40W of conversion loss in teh psu Feb 26 11:34:28 and the rest of the energy is leaving the box! Feb 26 11:34:43 you're making no sense at all Feb 26 11:34:57 if you have 40W _loss_ your output should be something like 500W Feb 26 11:34:59 at least Feb 26 11:35:04 mrpackethead: then i ask again: what is the power rating of your PSU? 1kW? Feb 26 11:35:05 your right Feb 26 11:35:13 listen: Feb 26 11:35:21 theres a small load in the box Feb 26 11:35:27 and a large load outside of the box Feb 26 11:35:45 how large Feb 26 11:35:48 so there are _two_ boxes now? Feb 26 11:35:52 no.. Feb 26 11:35:55 for 40% of losses you would need at least 1-2kW PSU Feb 26 11:35:57 40W* Feb 26 11:36:01 mru: they multiply... like rabbits! Feb 26 11:36:25 40W of 2kw, would imply a 98% efficent psu Feb 26 11:36:32 i'd like to buy one... Feb 26 11:36:47 i can sell you one Feb 26 11:37:00 for $60 Feb 26 11:37:03 lol. Feb 26 11:37:26 's füfi, s'weggli, s'usegeld und die hübsch verchäuferi Feb 26 11:37:39 so this box contains a bigass PSU, some low-power circuitry, and a cable carrying power out to some external user? Feb 26 11:37:48 yes. Feb 26 11:37:54 and what's the problem? Feb 26 11:38:01 no real problem Feb 26 11:38:08 *facepalm* Feb 26 11:38:18 other htan getting the heat out of the box, thats due to the conversion loss of the PSU Feb 26 11:38:40 how large is the box? Feb 26 11:38:42 i gues that is a real problme Feb 26 11:38:50 260 x 160 x 90mm Feb 26 11:38:53 easy Feb 26 11:39:00 4mm diecast allumimum Feb 26 11:39:06 dream Feb 26 11:39:21 just bolt the heat-producing components to the inside of the case Feb 26 11:39:36 yes. Feb 26 11:39:51 or if layout precludes that, use heatpipes Feb 26 11:40:02 combination of the two Feb 26 11:40:08 woudl probably work Feb 26 11:42:14 assuming i can achieve a conversion efficencly of something approaching what is commonly commerically availble Feb 26 11:42:22 for a 5V ouput Feb 26 11:42:50 (AC 120V/60Hz in, 5V out ) Feb 26 11:43:08 i'm looking at around 80% Feb 26 11:43:13 efficency Feb 26 11:43:36 that's still 200W of power out Feb 26 11:43:43 ( 30A @ 5V ) / .8 Feb 26 11:43:48 40A Feb 26 11:44:05 187.5W Feb 26 11:44:13 150W used 'elsewhere' Feb 26 11:44:22 37W to get rid of Feb 26 11:44:25 in heat Feb 26 11:44:33 is that a one-off project? Feb 26 11:44:49 one off, but x 300 or so boxes Feb 26 11:44:54 all the same Feb 26 11:45:14 i could potentially change the enclosure Feb 26 11:45:20 but not ideally wanting to do Feb 26 11:45:27 it was all good Feb 26 11:45:37 untill the client said.. " this is not for inside " Feb 26 11:45:39 that's not one-off, that's a small series production Feb 26 11:45:49 one of is when you build one device only Feb 26 11:45:50 one-off project Feb 26 11:46:02 then we do only one off projects here Feb 26 11:46:10 even those that are produced in millions per year Feb 26 11:46:51 we would have typically just taken off the shelf SMPS Feb 26 11:46:56 put in box. Feb 26 11:47:03 attached fan to box Feb 26 11:47:09 and vented Feb 26 11:47:14 problem gone. Feb 26 11:47:34 now... Feb 26 11:47:54 i have potentially 40C outdoor temps Feb 26 11:48:17 lucky this stuff only needs to run at night Feb 26 11:48:36 a PSU should be fine running at 80C Feb 26 11:49:36 some are rated Feb 26 11:49:42 at that temp Feb 26 11:50:05 im not so keen to do so Feb 26 11:50:17 hi Feb 26 11:50:22 you can't get components rated for less than 70C Feb 26 11:50:34 higher-rated ones are easy to get Feb 26 11:51:03 i think the reliablity stats are somethign like double the failure for every 10 degrees ? Feb 26 11:52:01 mru, finding a better of the shelf PSU ( better efficency ) Feb 26 11:52:06 woudl be highly desireable Feb 26 11:52:28 the only question is how much you are willing to pay Feb 26 11:52:42 well, actively cooling it, is going to be expenisve Feb 26 11:53:01 you can have a fan for $2 Feb 26 11:53:05 yes.. Feb 26 11:53:22 but a fan on the outside is not great! Feb 26 11:54:06 submerge it in water then Feb 26 11:54:16 mmm, nice idea Feb 26 11:54:24 uniklely to be considered practical Feb 26 11:55:12 i was reading earlier about peltier devices Feb 26 11:55:23 direct to air Feb 26 11:55:27 those can't help you here Feb 26 11:55:37 Can I use the old beaglebone to exchange "The Next-Gen BeagleBone"?? Feb 26 11:56:01 peltier elements are useful for forcing heat flow through a small surface Feb 26 11:56:12 you still need something on the other side to dissipate it Feb 26 11:56:12 yup.. Feb 26 11:56:28 ie a large heat sink Feb 26 11:56:42 but your box is plenty large enough Feb 26 11:57:06 i might tommorrow then sick a psu in the box Feb 26 11:57:13 and stick a dummy load on it Feb 26 11:57:15 and see what happens Feb 26 11:58:06 the box woudl only be big enough if the parts are direclty attached to teh box Feb 26 11:58:08 ? Feb 26 11:58:12 sorry thats a quesiton Feb 26 11:58:16 so do that Feb 26 11:58:24 if you don't do that, nothing you do on the outside will help Feb 26 11:58:35 short of dunking it liquid nitrogen Feb 26 11:58:39 lol. Feb 26 11:58:48 liquid nitrogen is fun Feb 26 11:58:52 that it is Feb 26 11:58:57 not so practical though Feb 26 11:59:07 i do a soem 'science' magic at the museum with it Feb 26 11:59:27 kids think its funny when i use a banana to put nails in bits of wood Feb 26 11:59:31 :-) Feb 26 12:00:07 anyway, your psu probably has a few components attached to a heatsink Feb 26 12:00:28 psu comes with a fan built in Feb 26 12:00:35 to a enclosed case Feb 26 12:00:42 if that goes in teh box Feb 26 12:00:58 then we fill teh box with warm air Feb 26 12:01:16 you could modify the psu Feb 26 12:01:23 or use an open frame psu Feb 26 12:01:50 for a small run, doing some mods to a standard part is acceptable Feb 26 12:01:58 if it's otherwise the best choice Feb 26 12:02:09 yes. Feb 26 12:02:25 I'd remove any casing from the psu and lose the fan Feb 26 12:02:36 if i tool a 'stock' psu, that was normally forced air cooled Feb 26 12:02:43 then run a strip of aluminium from its heatsink to the external casse Feb 26 12:02:53 and had it sitting inside a box Feb 26 12:03:01 we would end up with a box of warm air Feb 26 12:03:04 effectively Feb 26 12:03:21 its not going to cool sufficently doing that Feb 26 12:03:23 ? Feb 26 12:03:24 _av500_: did you get EC running on your lunchbox? Feb 26 12:04:17 on of the other issues, we face is that we have to use UL listed PSU's Feb 26 12:04:23 or the paperwork is insane Feb 26 12:04:57 hmm, would that be invalidated by tinkering with it? Feb 26 12:05:42 almost certianly Feb 26 12:05:43 :-) Feb 26 12:06:15 then you'd better find one that already has an exposed heatsink Feb 26 12:10:01 http://www.meanwell.com/search/usp-225/default.htm Feb 26 12:10:10 that might just work Feb 26 12:32:12 thanks mru Feb 26 12:32:16 :-) Feb 26 12:35:37 <_av500_> mdp: compiled it, but have no lunchbox here with me Feb 26 12:38:41 so you'll be hunnnngrrrrrey Feb 26 12:39:03 brrrraiiinsss Feb 26 12:40:25 zombie trolls from outer space Feb 26 12:42:08 good thing bradfa is familiar with this channel... Feb 26 12:42:23 * bradfa is not here... Feb 26 12:42:25 koen: you seen panto lately? Feb 26 12:42:47 define lately Feb 26 12:42:55 * mru saw him friday... Feb 26 12:43:01 since friday Feb 26 12:43:08 on IRC Feb 26 12:43:40 he wrote some helpful code over the weekend and we all know no good deed goes unpunished. Feb 26 12:43:54 ah, that is true Feb 26 12:44:00 time for support Feb 26 12:44:12 <_av500_> last post was from Munich airport Feb 26 12:44:14 on g+ he posted something about MUC airport Feb 26 12:44:15 you'll probably find him shackled to the wall down in the dungeons Feb 26 12:44:32 yup dungeons, muc, that matches... Feb 26 12:45:13 or strapped inside a tin can hurtling through the air Feb 26 12:47:14 alluminum or composite can.. not many ( any? ) sky hurtlng cans make from tin i'd have though Feb 26 12:47:42 but as the cheese troll demonstrated, just becuase you dont' know about it, it doe'snt exisit. Feb 26 12:47:59 tin in the british generic sense Feb 26 12:48:03 meaning any thin metal Feb 26 12:48:15 lol. Feb 26 12:48:18 http://oxforddictionaries.com/definition/english/tin%2Bcan Feb 26 12:48:52 #beagle is so strictly typed except for when its not Feb 26 12:55:39 bradfa: btw: i got some numbers on cr-5000 and cr-8000: single floating license is in the range of 70-90kEUR Feb 26 12:57:47 bradfa: but that'd be a complete license with bells and whistles Feb 26 13:02:45 <_av500_> cow bells? Feb 26 13:03:42 more Feb 26 13:04:43 probably a fuurin, given that the company is japanese Feb 26 13:11:09 KotH, gah! Feb 26 13:11:30 holy guacamole that's some real money Feb 26 13:12:04 KotH, you should write a blog post comparing Eagle and CR-8000 Feb 26 13:12:17 <_av500_> http://www.dubsounds.com/image_files/CR-8000-BIG.jpg Feb 26 13:12:27 <_av500_> looks nice this cr 8000 Feb 26 13:12:45 +1 Feb 26 13:14:16 KotH, guess we'll stick with Altium Feb 26 13:15:51 Hi! Anyone using S/PDIF with am335x? Feb 26 13:16:25 <_av500_> spdif? Feb 26 13:17:14 _av500_: http://en.wikipedia.org/wiki/Spdif Feb 26 13:17:42 _av500_: meant as protocol and not physical link Feb 26 13:18:04 _av500_: over one of the McASP data pins Feb 26 13:18:19 _av500_: DIT mode Feb 26 13:20:14 I've tried to get help over official channel (e2e.ti.com forum) http://e2e.ti.com/support/arm/sitara_arm/f/791/p/247447/866743.aspx#866743, but no luck so far Feb 26 13:21:09 The davinci-mcasp.c driver implementation seems to differ from technical reference manual Feb 26 13:22:51 bradfa: i would, unofrunately i dont have a blog, dont use eagle and dont have the monies for cr-8000 Feb 26 13:22:59 I'm using sii9022 HDMI transmitter and still cannot see, what's wrong: weather mcasp configuration of sii9022 configuration. Though sii9022 configuration means configuring two bits, the rest will be derived automatically from the incoming data stream Feb 26 13:24:33 bradfa: we are currently looking at cadstar to see whether it would be worth to replace altium with it Feb 26 13:24:56 bradfa: according to what the guy said, it would cost us about as much as the next upgrade round of altium would cost us Feb 26 13:25:37 bradfa: but considering that we could have floating license (which altium doesnt properly yet) and very cheap schematic entry licenses (~1000eur each), it might be worth it¨ Feb 26 13:25:58 bradfa: oh.. and small versions of cadstar cost nearly half as much as the full license Feb 26 13:27:39 KotH, does altium's online license thing not work well enough for you? Feb 26 13:27:52 that's what we have but I don't know the details Feb 26 13:28:10 we use CADSTAR here with no issues Feb 26 13:33:05 bradfa: it requires internet :) Feb 26 13:33:17 bradfa: we dont like to rely on something that needs a network connection Feb 26 13:33:43 bradfa: our isp here is kind of sucky... and unfortunately we are stuck with him Feb 26 13:34:07 jackmitchell: for what kind of projects do you use it? Feb 26 13:34:15 jackmitchell: how many people? what additional packages? Feb 26 13:34:26 jackmitchell: is there something you dont like about cadstar? Feb 26 13:36:00 KotH: I'm not the one who uses it so I can't really comment but I would say our projects aren't huge, but definitely not small and we only have the once licence Feb 26 13:36:11 one* licence Feb 26 13:36:23 <_av500_> yegorich: i know what spdif is :) I was not aware the thing had SPDIF output... Feb 26 13:36:30 jackmitchell: could you ask the guy who uses it for some comments? Feb 26 13:36:35 does the "once licence" let you run the program one time? Feb 26 13:36:49 mru: think skynet ;) Feb 26 13:37:02 KotH: I'll ask if he has any gripes with it Feb 26 13:37:39 or like the old photoshop eval licence: as long as you didn't close the app, it would run forever Feb 26 13:37:48 don't know if they ever changed that Feb 26 13:37:57 * KotH doesnt need that Feb 26 13:38:02 * KotH uses student licenses Feb 26 13:38:02 now combine that with a virtual machine snapshot... Feb 26 13:39:04 virtualize everything! Feb 26 13:39:06 KotH: he says he doesn't have any issues with it, we don't use it to the full extent so it's quite an advanced package for us Feb 26 13:39:37 KotH: licensing isn't an issue, we have a maintenance licence which is updated once a year with the new release Feb 26 13:40:05 _av500_: I've found that feature almost to late :-( Feb 26 13:40:14 KotH: you do have to use a usb dongle for the licensing, but you don't need an internet connection Feb 26 13:43:09 jackmitchell: juup, i got the licenisng terms from the vendor Feb 26 13:43:18 jackmitchell: and lots of links to further documentation Feb 26 13:43:30 jackmitchell: but it's hard to evaluate something w/o having it :) Feb 26 13:43:42 jackmitchell: may i ask what kind of electronics you usually do? Feb 26 13:44:07 a mixture, google dbbroadcast Feb 26 13:46:07 jackmitchell: looks like some uC/uP with slow interfaces and some RF ontop of it? Feb 26 13:46:47 yeah almost, FPGA+uC+RF Feb 26 13:47:03 jackmitchell: QFP or BGA/FBGA? Feb 26 13:47:33 BGA/FBGA Feb 26 13:47:41 cooll... Feb 26 13:47:43 apparently ;) Feb 26 13:47:53 that's not so apparent ;) Feb 26 13:48:21 but hat sounds mostly of the same complexity of the stuff we are doing Feb 26 13:48:24 thanks a lot Feb 26 13:48:43 I'm just the code monkey; our FPGA guy says thats pretty much what we use Feb 26 14:06:45 ls Feb 26 14:06:53 morning :) Feb 26 14:07:34 no such file or directory Feb 26 14:07:39 afternoon :-) Feb 26 14:08:09 jackmitchell: dont worry about that, i'm the local janitor ;) Feb 26 14:35:17 hi everyone, how's it going? Feb 26 14:35:28 fast Feb 26 14:35:47 #beagle is on internet time Feb 26 14:36:11 * bradfa remembers internet time fondly Feb 26 14:36:25 * bradfa wishes everyone just used UTC, stupid time zones and daylight savings Feb 26 14:37:21 bradfa: UGT! Feb 26 14:37:42 I am having some problems on my bbxm, as when i want to run a benchmark compiled using arm-angstrom-linux-gnueabi-gcc (4.5.4, both native on bbxm and toolchain) when running the program stops saying: "*** glibc detected *** /home/root/h264dec/JM/bin/ldecod.dbg.exe: free(): invalid pointer: 0x4204a008 *** " Feb 26 14:38:12 starklombardi, don't free memory you didn't alloc Feb 26 14:38:21 valgrind is your friend Feb 26 14:38:43 i didn't touch the code, which runs fine on x86 Feb 26 14:39:11 koenkooi, I learn something new every day! Feb 26 14:39:13 * bradfa is new to the internet Feb 26 14:39:26 starklombardi, well, that doesn't matter Feb 26 14:39:47 just cause code runs somewhere doesn't mean it does it correctly Feb 26 14:39:55 valgrind will save you Feb 26 14:40:04 maybe Feb 26 14:40:09 well, it'll at least tell you where your code is being less than stellar Feb 26 14:40:16 or at least point to where the error is, fixing it will be another matter Feb 26 14:40:17 there are _many_ ways code can be broken that valgrind doesn't detect Feb 26 14:40:23 mru, shhhh Feb 26 14:40:29 * bradfa needs big board points Feb 26 14:40:34 on the technical reference manual, address 0x40250000 - 0x47FFFFFF is reserved, can that be the problem? Feb 26 14:40:53 why is there an .exe in the filename; is that normal? Feb 26 14:41:04 starklombardi, are you looking at physical or virtual addresses? Feb 26 14:41:09 is there a way to memory map at compiling/linking? Feb 26 14:41:28 at the very least, code should be tested with valgrind, clang address sanitiser, clang undefined behaviour checker, and if relevant some thread checker Feb 26 14:41:40 /dev/mem fixes everything Feb 26 14:41:47 you can never throw too many tools at bugs Feb 26 14:41:53 the .exe is automatically generated by the makefile, it is still an arm binary Feb 26 14:41:56 mru, got any recommendations on thread chekers? valgrind does this to some extent, no? Feb 26 14:42:12 valgrind has some such tools, yes Feb 26 14:43:09 starklombardi, are you running with MALLOC_CHECK_? Feb 26 14:43:39 no. do i add it on the make? Feb 26 14:43:45 http://www.gnu.org/software/libc/manual/html_node/Heap-Consistency-Checking.html Feb 26 14:43:51 anyone built Angstrom using EC2? I'm interested in starting a faster instance for my builds and wonder if anyone knows how to configure it to build fast. Feb 26 14:43:58 thread checkers tend to be rather inaccurate though Feb 26 14:44:12 My current instance takes 24 hours+ to build the cloud9 image Feb 26 14:44:24 jkridner___, have you tried the high mem and using tmpfs? Feb 26 14:44:45 High-Memory Quadruple Extra Large Instance should have enough ram for builds, no? Feb 26 14:45:08 yeah, and the I/O didn't seem to go up like it was really RAM. Feb 26 14:45:39 I can try one of those, but I'm concerned it'll be super expensive to leave running over the next month while I'm doing repeated builds. Feb 26 14:45:49 jkridner___, just spin up for the build then spin down Feb 26 14:45:56 if you've tried it and it is fast, then I'd go for it. Feb 26 14:45:57 High I/O Quadruple Extra Large Instance <- about $3 / hour Feb 26 14:46:06 jkridner___, no, haven't tried, just thought about Feb 26 14:46:28 jkridner___, get employer to buy you bigger machine :) Feb 26 14:46:50 I can use one of the disks that is persistent between boots. Feb 26 14:46:59 a decent-size build server isn't _that_ expensive Feb 26 14:47:04 I want to be able to get to it anywhere. Feb 26 14:47:07 ssh Feb 26 14:47:17 if I put it at home, getting through the cable company can be a pain. Feb 26 14:47:24 jkridner___, gotomypc? Feb 26 14:47:28 or similar Feb 26 14:47:31 8-core with 32G ram should be easily affordable Feb 26 14:47:35 plus, shouldn't I be able to get a faster machine using a cloud service? Feb 26 14:47:38 and I don't know about the US but in the UK upload speeds sucks donkeys Feb 26 14:47:41 not neccessarly Feb 26 14:47:51 interesting. Feb 26 14:47:53 colocate your build box Feb 26 14:48:05 most colo facilities have good internet Feb 26 14:48:13 free power, too Feb 26 14:48:34 jackmitchell: stuck on adsl too? Feb 26 14:48:36 I've got an 8-core Xeon Mac that I could up the RAM on. is it worth it or should I just get a new machine. Feb 26 14:48:41 it is about 2 years old. Feb 26 14:48:57 what kind of cpu is that? Feb 26 14:48:59 http://prgmr.com/xen/ Feb 26 14:49:00 i7 based? Feb 26 14:49:07 mru: no; I'm living the high life on Virgin Feb 26 14:49:18 whoops, that's xen, this one http://prgmr.com/san-jose-co-location.html Feb 26 14:49:32 jackmitchell: how's that working out? Feb 26 14:49:36 no, predates i7. 2 x 2.8GHz Quad-Core Intel Xeon Feb 26 14:49:47 currently 4GB 800MHz DDR2 Feb 26 14:49:48 :( Feb 26 14:49:52 mru: very well, very fast; I don't appreciate the fair use policy though Feb 26 14:49:53 jkridner___, get a new machine Feb 26 14:49:59 jkridner___: don't put money into that machine Feb 26 14:50:00 $2k buys a lot of Dell Server these days Feb 26 14:51:06 <_av500_> get a cluster of r-pis Feb 26 14:51:17 <_av500_> arm based multiserver is the future Feb 26 14:51:31 <_av500_> hadoop all the things! Feb 26 14:51:33 +1 for _av500_ Feb 26 14:51:42 http://www.southampton.ac.uk/~sjc/raspberrypi/ Feb 26 14:51:48 <_av500_> map, reduce, rinse, repeat Feb 26 14:51:52 <_av500_> 4) profit Feb 26 14:52:01 jackmitchell, +1 for python on rasppi Feb 26 14:52:12 in a cluster! Feb 26 14:52:21 #exactsteps Feb 26 14:52:33 jackmitchell, wrong channel, use /join before channel name :) Feb 26 14:52:54 <_av500_> jkridner___: you could also make every beagle buyer dedicate 5% of his CPU time for a good cause Feb 26 14:53:02 sorry, looks like I needed exact steps to visit #exactsteps Feb 26 14:53:03 hi all, I have one doubt, the reload address is at zreladdr-y := 0x80008000, here why we have a offset of 8000 around 32k Feb 26 14:53:11 jackmitchell, ? Feb 26 14:53:41 bradfa: just having fun; the rapsberry pi page has exact steps and I was hashtagging it to be cool Feb 26 14:53:41 <_av500_> bradfa: meta Feb 26 14:53:51 * bradfa gets +1 Feb 26 14:56:32 * bradfa gets free second breakfast from NXP, bbl Feb 26 15:03:34 ship dates from dell seem to start on March 7. :( Feb 26 15:06:27 * jkridner___ checks out http://www.openembedded.org/wiki/EC2CloudBuild Feb 26 15:06:39 <_av500_> to Dell there is no world before March 7? Feb 26 15:06:50 jkridner___, go to dell store, no? Feb 26 15:06:51 <_av500_> it needs to be created first? Feb 26 15:07:04 _av500_: yes, starting march 1 Feb 26 15:07:09 takes 7 days Feb 26 15:07:39 hi all, I have one doubt, the reload address is at zreladdr-y := 0x80008000, here why we have a offset of 8000 around 32k Feb 26 15:07:53 bradfa: are there any Dell stores? Feb 26 15:08:03 best buy? Feb 26 15:08:17 staples? Feb 26 15:08:20 dunno Feb 26 15:08:31 Didn't gateway2000 used to have stores? Feb 26 15:08:43 those don't seem likely to carry server-grade hw Feb 26 15:08:51 google says dell has stores Feb 26 15:08:56 dell website slow... Feb 26 15:09:15 http://www.dell.com/content/topics/segtopic.aspx/stores/dell-store?c=us&l=en&cs=19 says they are all closed Feb 26 15:09:29 <_av500_> gateway2000 still exists? Feb 26 15:09:42 jkridner___, wait, they closed like 2 weeks ago? Feb 26 15:09:50 they must have known you needed a new computer Feb 26 15:10:06 er, like last week? Feb 26 15:10:10 go go dell Feb 26 15:10:22 so apparently dell doesn't exist between feb 20 and march 7 Feb 26 15:10:36 maybe they are rebranding? Feb 26 15:11:53 How can i add a new boot logo in the latest kernel Feb 26 15:12:11 can anyone point me to a existing logo implementation Feb 26 15:12:28 jkridner___, http://unixsurplus.com/product/supermicro-x8dtu-f-192gb-1x1tb-sata-wvmware-xen-virtualization-host-rail-kit Feb 26 15:12:44 that's a decent build box Feb 26 15:13:20 in stock Feb 26 15:13:23 jkridner___: what's wrong about buying a mainboard, a cpu, some ram and be done with it? Feb 26 15:13:32 jkridner___: you can get that in 10min in any computer shop Feb 26 15:13:44 KotH: you probably want a PSU as well... Feb 26 15:14:02 http://unixsurplus.com/product/supermicro-x8dtu-f-single-xeon-l5520-24gb-ddr3-w-rail-kit Feb 26 15:14:06 that's not bad for $500 Feb 26 15:14:40 mru: that goes under "small parts you have lying around anyways" :) Feb 26 15:15:03 until the caps blow up Feb 26 15:15:42 * KotH usually buys brand psus Feb 26 15:15:45 which reminds me, I should either fix or get rid of that dead psu in the corner Feb 26 15:15:51 bradfa: add another 500 for shipping ;) Feb 26 15:15:52 what brand? Feb 26 15:16:06 KotH, if you want overnight, only like $85 ground to me Feb 26 15:16:43 mru: mostly enermax, some be quiet... Feb 26 15:16:49 bradfa: oh..kay.. Feb 26 15:16:58 bradfa: BUY! BUY! BUY! ;) Feb 26 15:17:07 * bradfa BUY BUY BUYs! Feb 26 15:17:45 bradfa: imho that be worth to try wether you can use it for anything for 600 bucks Feb 26 15:17:54 http://unixsurplus.com/product/sun-ultra-40-nvidia-sli-powered-multi-monitor-workstation Feb 26 15:17:59 KotH: I have some PSUs that have lasted forever, some have died after various amounts of time Feb 26 15:18:12 bradfa: just be aware that those 1U supermicros are superloud Feb 26 15:18:28 KotH, puttem in a shed Feb 26 15:18:35 anything 1U is loud Feb 26 15:18:39 2U too Feb 26 15:18:46 mru: i've so far only killed one psu... and that with repeated current surges from mains Feb 26 15:19:28 mru: hakuoro is so silent, i couldnt tell whether it was on or not werent it for the leds... Feb 26 15:19:39 mru: and i tried it in our otherwise quite silent office Feb 26 15:19:58 KotH: the dead psu in the corner here is a tagan Feb 26 15:20:24 mru: tagan... never heard of them Feb 26 15:20:41 mru: they require a flash plugin on their homepage.. can't be that good Feb 26 15:20:51 clearly they're not Feb 26 15:20:53 it broke Feb 26 15:21:14 I have some unmarked ones that are very old and still working Feb 26 15:21:43 oh.. and i just let the magic smoke of one lab psu this saturday Feb 26 15:21:58 buy just switching it on and cranking up the output voltage to maximum Feb 26 15:22:06 ... didnt even need to connect anything ^^' Feb 26 15:22:16 s/buy/by/ Feb 26 15:23:04 bradfa, might as well get a real powerhouse box instead, http://goo.gl/w4zqJ Feb 26 15:24:01 +1 mdp Feb 26 15:24:32 mdp: i'd rather use a door stopper than a pizza box: http://sturgeon.css.psu.edu/~mloewen/Oldtech/Sun/IPX-1L.jpg Feb 26 15:24:36 it's the recommended machine for reading comp.os.linux Feb 26 15:24:49 huh? Feb 26 15:25:01 linux is damn slow on these ss5's Feb 26 15:25:10 ;) Feb 26 15:25:56 (ok, compared to linux running on a intel box with same clock speed, these things are rocket fast... but compared to solaris... no) Feb 26 15:26:05 KotH: you should see linux on an ipx...makes the ss5 look lightning fast Feb 26 15:26:12 KotH: but it didn't seem slow back in the day Feb 26 15:26:44 mdp: as i just said, compared to intel machines these were damn fast... but linux was simply the wrong os on them Feb 26 15:27:00 mdp: and you make me feel old ^^' Feb 26 15:27:30 KotH: right or wrong...it was about having fun ;) Feb 26 15:27:45 mdp: hehe.. yeah! Feb 26 15:28:00 mdp: had a damn lot of fun with my personal pile of ss2's and ss5's Feb 26 15:28:03 I know we've lost that magic a bit...but it's ok to do stuff for hack value ;) Feb 26 15:28:19 mdp: lost? rather dont have the time Feb 26 15:28:25 :-( Feb 26 15:28:30 yeah :( Feb 26 15:29:11 mdp: apropos... you, being an old fart... have you ever played with avalance transistors or transistors run deliberately in avalanche mode? Feb 26 15:29:36 mru: at least the make fancy looking psus.... http://www.overclock3d.net/gfx/articles/2007/03/30120913985l.jpg Feb 26 15:29:59 oh no Feb 26 15:30:02 blue leds Feb 26 15:30:03 again Feb 26 15:30:09 mine doesn't have that Feb 26 15:30:17 KotH: only remember talking about that concept in class Feb 26 15:30:28 not every single piece of electrical/electronic hardware has to have superbright blue LED Feb 26 15:30:32 mdp: meh... Feb 26 15:30:39 and that PSU has 20 of those by the looks of it Feb 26 15:30:57 mdp: i tried to get an BC547 to work as ultra short impulse source using it in avalanche mode this weekend... Feb 26 15:31:03 mdp: but i couldnt get it working properly Feb 26 15:31:20 my workstation has one of these: http://www.silverstonetek.com/product_legacy.php?area=en&model=ST45NF Feb 26 15:31:51 KotH: ahhh, that class is right about at 20 years stale for me now :-/ Feb 26 15:32:08 mdp: at least you had that in some class... i never did :-( Feb 26 15:32:41 my sedra-smith microelectronics book is staring at me on the bookshelf now Feb 26 15:32:45 * KotH curses the eidgenösisch theoretische hochschule Feb 26 15:33:59 * mru has a book covering the topic Feb 26 15:34:25 if briefly Feb 26 15:34:39 mru: does it cover the topic as in "this or that happens" or does it cover the topic as in "you have to do this and that to get it working"? :) Feb 26 15:34:58 to answer that I'd have to _read_ it Feb 26 15:35:12 *sigh*...bench is so messy I've now lost the logic analyzer Feb 26 15:35:12 which I may or may not have done ~12 years ago Feb 26 15:35:14 mdp: i just bought a new edition of the tietze schenk... and had a EE-gasm about all the fancy stuff they added in the last 10 years Feb 26 15:35:27 KotH: LOL Feb 26 15:35:43 mdp: the book is really cool Feb 26 15:36:40 mdp: it was a great book when i first found it in my high school lib (5th ed), it was a superb book when i bought it in university (10th ed), now(14th ed) it's a spectacular book Feb 26 15:36:42 cool there is a picture of the next beagle bone : https://plus.google.com/102344295024422039483/posts Feb 26 15:37:03 mdp: the introductory chapters cover as many pages as the whole 10th edition was ^^' Feb 26 15:37:47 jet: it's already on preorder on mouser or farnell or someoneoneoftheusualsuspects Feb 26 15:37:49 omg, that beaglebone is completely unzipped! Feb 26 15:38:08 mdp: dont worry, it will be sold in europe, where people are not afraid of naked nipples Feb 26 15:38:15 +1 Feb 26 15:38:27 KotH: yes, but that's the first picture of it Feb 26 15:38:32 (that I found) Feb 26 15:38:46 so exactly the same size, and still one usb port Feb 26 15:38:54 I think it is the first one that I shared Feb 26 15:39:16 mdp: i always find it disturbing how afraid usians are of naked skin Feb 26 15:39:24 mdp: i mean.. it's something everyone has Feb 26 15:39:24 2 USB ports, one host and one client Feb 26 15:39:40 would be nice if it was hacked to make the client a dual-role port. Feb 26 15:40:08 yes, but I meant one usb port to plug wifi/keyboard/mouse Feb 26 15:40:12 mdp: and it's not like i'm killing someone by ranning around naked... Feb 26 15:40:15 KotH: cultural differences, in general, are very interesting Feb 26 15:41:14 mdp: indeed Feb 26 15:41:42 mdp: what i find most funny is, that muslim countries are often a lot more relaxed in that regard than the us Feb 26 15:41:43 damn, I have totally managed to destroy the ubuntu installer :-D Feb 26 15:41:56 nice work Feb 26 15:42:00 XorA: good, now you can install a real os Feb 26 15:42:06 I suggest linux Feb 26 15:42:33 it does not like LVM and luks at all ;-) Feb 26 15:42:37 mdp: on turkish beaches, half of the women are toples... and most of the pre-school kids ran around completely naked.... similar things can be seen in a few arabian countries as well... Feb 26 15:43:18 and parts of europe Feb 26 15:43:57 parts? most of europe, i'd say... or rather, i'm not aware of any place in europe where topless women would cause any uproar Feb 26 15:44:36 ok, maybe vatican ... but that place is special ;) Feb 26 15:44:50 they don't have any beaches Feb 26 15:45:15 and you definitly wouldnt want your children roaming around there.. especially not naked Feb 26 15:45:18 ;-) Feb 26 15:45:30 they'd be raped in an instant Feb 26 15:45:44 KotH: in my experience on greek beaches...only the people I don't want to see topless are as such ;) Feb 26 15:45:53 mdp: lol Feb 26 15:46:04 mdp: that's why you dont go to greece, but to turkey Feb 26 15:46:10 :) Feb 26 15:46:10 mdp: much nicer boops ;-) Feb 26 15:46:32 but dont tell that panto ;) Feb 26 15:47:10 I was quite fond of beaches around Nice! Feb 26 15:47:23 certainly an improvement to being in TI office there Feb 26 15:48:10 XorA: I think any beach is generally an improvement over an engineering office Feb 26 15:48:40 beaches are horrible places, full of sand and water Feb 26 15:48:52 mdp: cocktails and scenery, :-D Feb 26 15:49:39 * KotH would agree Feb 26 15:49:51 though, i havent been at a beach for ages :-/ Feb 26 15:49:56 last one was at the pacific... Feb 26 15:49:59 and that was 4y ago Feb 26 15:50:03 now if one could get the scenery without the sand and water... Feb 26 15:50:15 design better gadgets :-D Feb 26 15:50:18 most of them have that horribly salty water...hurray for freshwater beaches Feb 26 15:50:21 mru: wear a space suit Feb 26 15:50:54 mdp: I once had the "pleasure" of repairing a studio flash that had been used on some beach in greece Feb 26 15:51:01 mdp: drink the cocktails, not the water ;-) Feb 26 15:51:09 the circuit boards had a thin layer of salt on them Feb 26 15:51:31 mru, nice...it is a challenging environment for everything Feb 26 15:51:53 repair was as simple as washing the boards Feb 26 15:52:03 once the cause was determined Feb 26 15:52:22 none of the usual fault finding methods worked Feb 26 15:52:47 mru: haven't you learned to first check the basics? Feb 26 15:53:06 mru: like... check power, input signal and any dirt covering the circuit boards Feb 26 15:53:11 the usual fault was blown-up diode Feb 26 15:53:53 hmm... Feb 26 15:53:56 or blown-up capacitor Feb 26 15:54:01 both easy to find Feb 26 15:54:02 mru: what did you use to trigger the flash? Feb 26 15:54:14 high voltage transistor? Feb 26 15:54:17 triac? Feb 26 15:55:01 the flash capacitors are charged to 1kV and connected directly across a xenon tube Feb 26 15:55:37 a pulse of high voltage to a thin wire running around the tube ionises it enough to trigger the discharge Feb 26 15:56:00 hmm... Feb 26 15:56:01 ok... Feb 26 15:56:11 damn.. no techinque i could use for short pulses Feb 26 15:56:50 does anyone have an idea how i could generate 100ps pulses with a high repetition rate (at least 100MHz, better 700MHz) Feb 26 15:57:02 * mru does not Feb 26 15:57:28 and please no uses of unobtainable stuff like SDR diodes Feb 26 15:57:44 these things did ~1ms pulses with ~1s repetition rate Feb 26 15:57:44 SRD* Feb 26 15:57:52 at full power Feb 26 15:58:00 faster at lower power Feb 26 15:58:12 oh.. i dont need that much power.. just around 5-10V, 30mA Feb 26 15:58:12 or energy, more accurately Feb 26 15:58:35 megawatts are more fun Feb 26 15:58:43 even if they only last a millisecond Feb 26 15:59:03 well.. more energy and the laser diode burns Feb 26 15:59:41 if i would want to have more power per pulse, i'd use a q-switched solid state laser Feb 26 15:59:55 it was a long time ago, and only for a summer, but at least I had an actual EE job once... Feb 26 16:00:01 Hi Feb 26 16:00:34 Can some one help me HDMI output related stuff for beagle board Feb 26 16:00:42 * KotH cannot Feb 26 16:00:50 only if you state your problem Feb 26 16:00:54 but if you'd have questions about high power lasers, i might be of help Feb 26 16:01:02 :) Feb 26 16:01:07 Here is my problem Feb 26 16:01:08 KotH: you should make a cape for that Feb 26 16:01:30 I did n't get you Feb 26 16:01:40 I am the first time user. Feb 26 16:01:45 please never mind.. Feb 26 16:01:46 mru: the end goal is to have a 2 laser system (possibly also a 4 or 6 laser system) that usess a beagle as controler :) Feb 26 16:01:46 for extra credit you make a reusable shieldCapeLure of that Feb 26 16:02:29 but first i have to figure out how i generate two laser beams that are locked to each other and 6.9GHz apart Feb 26 16:02:29 dogs with frickin' lasers Feb 26 16:03:35 one in each shark eyeball? Feb 26 16:04:03 XorA: silence! Feb 26 16:04:10 XorA: these are serious scientific questions! Feb 26 16:04:11 Problem: I have Dell Monitor connected to BB RevC. I have build my own linux 3.2 version kernel.I could see the beagle board logo after that the monitor goes it to sleep mode. Feb 26 16:04:15 ;) Feb 26 16:05:16 sorry Feb 26 16:05:28 i have tried all verity of resolution combinations. My boot args are proper.. Feb 26 16:05:36 XorA: dont worry, i wont kill you....yet Feb 26 16:05:47 but no luck Feb 26 16:05:51 murali: my BB RevC will not drive my BenQ monitor no matter the mode settings Feb 26 16:06:24 murali: I suspect there is some borderline case involved Feb 26 16:06:47 but Amstrong pre-build images are able to give proper display Feb 26 16:07:21 murali: compare the dmesg output of both Feb 26 16:07:24 murali: ah, on mine Angstrom images have same behaviour Feb 26 16:07:36 I have to walk across room to use TV instead :-D Feb 26 16:08:11 Any thing to do with DSS Feb 26 16:08:15 driver> Feb 26 16:08:41 hello @ all Feb 26 16:09:14 yet another one who need help.... Feb 26 16:09:21 time to leave the battle field Feb 26 16:09:26 bye boys and girls Feb 26 16:09:35 oh no.. i have a short question.. Feb 26 16:09:47 any ideas pls Feb 26 16:09:48 that's just a troll anyway Feb 26 16:10:54 could someone help me? i have an node.js script with an interrupt (button) as service. Feb 26 16:11:17 if i start the service manually (systemctl start myNode.js) everything is ok Feb 26 16:11:58 but if i set the service to autostart (systemctl enable myNode.js) ne interrupt doesn't work.. Feb 26 16:12:05 anyone an idea? Feb 26 16:12:37 what does systemctl status tell you? Feb 26 16:13:15 also with enabled by autostart the script starts (another led is blinking) Feb 26 16:13:34 so the script should have started and worked.. Feb 26 16:13:41 bb_needHelp: ^^^ Feb 26 16:13:48 only the interrupt doesn't works.. Feb 26 16:15:43 systemctl says: Active: active (running) Feb 26 16:17:52 no idea anyone? Feb 26 16:18:38 bb_needHelp: have you got plenty of debug statements in your code to make sure it's doing what its supposed to be doing? Feb 26 16:18:46 if it says it is running, then it is likely that the process is running and there is a bug in your code. Feb 26 16:18:59 if you have any debug statements, they should show up with 'journalctl' Feb 26 16:19:05 bb_needHelp: i.e. just some prints to ensure that it is starting properly and running Feb 26 16:20:03 the code is starting. In the loop is blinking a LED.. Feb 26 16:20:21 I've started using winston for most of my node.js code and that seems to work nicely. Feb 26 16:20:22 this is my debug statement :) Feb 26 16:20:35 then, starting the code isn't your problem. Feb 26 16:21:05 just interrupts don't seem to work when you invoke from boot-up, eh? Feb 26 16:21:15 yes Feb 26 16:21:15 pastebin your code. Feb 26 16:21:24 have you got some checks to see if reading the interrupt is giving you an error? Feb 26 16:21:30 is there a chance that something comes in and reconfigures your pin? Feb 26 16:22:41 http://pastebin.com/uweF7r9g Feb 26 16:23:14 i dont think so because the other pin(the onboard led) is working Feb 26 16:23:39 did you look at 'journalctl' for your current errors? Feb 26 16:24:33 I don't know much about nodeJS but having a realtive path in the exec() call may be causing a problem Feb 26 16:24:49 that was one of my first thoughts as well... Feb 26 16:24:50 mom Feb 26 16:24:56 but, it comes after the LED write Feb 26 16:25:49 the console.log should occur as well before the exec, though I don't see what clears it after it is set the first time. Feb 26 16:26:13 nothing in the interrupt works.. Feb 26 16:26:39 and i don't see any console.log if it starts automaticaly Feb 26 16:26:45 I see that x.value is used to set the LED, so ignore the last statement. Feb 26 16:27:08 did you type 'journalctl'? (last time I'm going to say that) Feb 26 16:27:44 processes invoked by systemctl have their output sent to the journal Feb 26 16:27:54 er, invoked by systemd Feb 26 16:27:58 yes.. i didn't see any error but i while pastebin it in 1 minit Feb 26 16:28:37 we should see that a process was invoked to monitor the input pin. Feb 26 16:29:17 btw, you should move the pinMode(ledPIN, OUTPUT) *before* the digitalWrite(ledPin, LOW) Feb 26 16:29:25 but, that shouldn't cause your issue. Feb 26 16:30:08 you need to add something similar for ledPin2 as well. Feb 26 16:30:15 *before* the digitalWrite Feb 26 16:32:14 ok.. Feb 26 16:32:22 this is the journalctl Feb 26 16:32:22 http://imageshack.us/photo/my-images/405/journalctl.gif/ Feb 26 16:32:42 now i changed the path and restart the bb Feb 26 16:34:43 a pastebin of the entire journalctl output is desired. that image is useless Feb 26 16:35:38 no change with absolute path Feb 26 16:35:46 no the journalctl pastebin.. Feb 26 16:36:18 try journalctl -f and then press your switch a few times while it is running Feb 26 16:36:52 nothing happens Feb 26 16:36:56 or 'journalctl --no-pager > log.txt' Feb 26 16:37:00 and then post your log.txt Feb 26 16:40:36 to big.. Feb 26 16:41:22 try it with imagehack Feb 26 16:41:36 no Feb 26 16:41:47 journalctl -u=YOUR_UNIT Feb 26 16:42:15 what does YOUR_UNIT mean? Feb 26 16:43:51 is it my .service? Feb 26 16:44:20 yes. try 'journalctl -b -u YOUR_UNIT' without the .service Feb 26 16:44:31 example: journalctl -b -u cloud9 Feb 26 16:45:10 invalid option -- 'u' Feb 26 16:46:04 hmm.. journalctl --version Feb 26 16:46:18 mine likes it. systemd 195 Feb 26 16:46:38 systemd 189 Feb 26 16:47:09 does it show up under 'journalctl --help'? Feb 26 16:47:27 yes Feb 26 16:47:35 what did we need? Feb 26 16:47:58 does 'journalctl -b -u cloud9' or 'journalctl -b -u dbus' work? Feb 26 16:48:20 nothing with a "-u" works Feb 26 16:48:39 how about --unit=dbus Feb 26 16:48:41 the point is to simply strip down the output. guess you can just use grep Feb 26 16:49:14 https://wiki.archlinux.org/index.php/Systemd#Using_units Feb 26 16:49:18 unrecognized option '--unit=dbus' Feb 26 16:50:07 maybe the journalctl is too old. Feb 26 16:50:16 how can i update it? Feb 26 16:50:20 anyway, try using grep. Feb 26 16:50:28 grep? Feb 26 16:51:02 journalctl | grep YOUR_UNIT > text.txt Feb 26 16:51:11 how about using a real system that actually work instead? Feb 26 16:51:22 jackmitchell: it might show 'node' as the process Feb 26 16:51:44 so, you might need to replace YOUR_UNIT with node Feb 26 16:52:28 mru: are you insulting Angstrom or simply suggesting starting with a newer image? Feb 26 16:52:51 I'm suggesting using something based on sysvinint and shell scripts Feb 26 16:53:02 if you take that as an insult.... Feb 26 16:53:11 i'm using angstrom Feb 26 16:54:11 bb_needHelp: did the grep work? Feb 26 16:54:30 yes.. now I have a file Feb 26 16:54:32 a log, a log, my kingdom for a log Feb 26 16:54:38 heh Feb 26 16:54:49 well, that journal contraption screwed you on that one Feb 26 16:54:50 trade you a horse for your log :P Feb 26 16:54:52 ^^ Feb 26 16:55:05 but it is to short i think Feb 26 16:55:10 please share. Feb 26 16:55:12 pastebin Feb 26 16:55:13 * mru chops up a device tree Feb 26 16:55:15 there, have a log Feb 26 16:55:43 oh, mru I found a small leak in omapfbplay Feb 26 16:55:46 right, sorry to jump but my day is over here in blightly so good luck and I'll catch y'all tomorrow! Feb 26 16:56:00 meanwhile, it seems my gentoo doesn't want to build python-exec-0.2 Feb 26 16:56:19 thurbad: what did you expect from code hacked up a couple of weeks for a one-off demo? Feb 26 16:56:20 thanks jackmitchell Feb 26 16:56:37 not a show stopper unless you use it the way I do :) Feb 26 16:57:05 I'm a little worried that someone refers to 'using' that code Feb 26 16:57:12 with journalctl | grep node > node.txt Feb 26 16:57:13 http://pastebin.com/VVRwDgBC Feb 26 16:57:17 ll Feb 26 16:57:18 * jkridner___ looks forward to 'boneplay' Feb 26 16:58:09 next time use journalctl -b to not include logs from old boots Feb 26 16:58:17 jkridner___: I believe I am overdue for another stunt... Feb 26 16:58:38 * jkridner___ found bb_needHelp's problem Feb 26 16:58:47 maybe it was a one off that took a couple of weeks but it's leaps and bounds better than ffplay Feb 26 16:58:50 your app isn't finding the bonescript library Feb 26 16:59:01 thurbad: it's different Feb 26 16:59:16 and it was polished quite a bit after those weeks Feb 26 16:59:41 you need to have a shell script that performs 'cd /var/lib/cloud9' ahead of invoking 'node your_script.js' --or-- do 'npm install -g bonescript' to make it global. Feb 26 16:59:49 though, I've never tried the latter. Feb 26 17:00:03 only works in theory (which likely means it doesn't work) Feb 26 17:00:33 * mru switches to theoretical engineering Feb 26 17:00:40 mh.. ok.. Feb 26 17:01:13 but why its fine if start the service after booting manually with systemctl start BMA020.service? Feb 26 17:01:39 perhaps you start it from within /var/lib/cloud9? Feb 26 17:01:47 not sure what the directory would be Feb 26 17:01:58 and why is the rest working except the interrupt? Feb 26 17:02:21 also the other pins.. which also need the bonescript.. Feb 26 17:04:55 dunno, but I can see the error Feb 26 17:05:10 this is the log with -b Feb 26 17:05:11 http://pastebin.com/dwhy9xMf Feb 26 17:08:34 looks better Feb 26 17:09:22 yes.. but i can't see any error.. Feb 26 17:09:38 did you update the setup function per my recommendations? Feb 26 17:09:48 which? Feb 26 17:10:58 setup the pins before doing the digitalWrites Feb 26 17:13:09 this way? Feb 26 17:13:09 http://pastebin.com/rPKGHGkQ Feb 26 17:13:51 yes Feb 26 17:14:18 restarting.. Feb 26 17:14:48 nope.. Feb 26 17:14:53 doesn't work.. Feb 26 17:19:35 hmmm A1 doesn't boot with mainline u-boot and don't have a FTDI cable.. Feb 26 17:20:30 jkridner___: no weird erratas like using DDR2 instead of DDR3 on it right? :) Feb 26 17:20:38 is MMC support working for the beagle bone in v3.8? Feb 26 17:21:05 (beagle bone white, rev A6a) Feb 26 17:21:07 mranostay: it has a 256MB DDR3. Feb 26 17:21:23 mranostay: use one of those maxim chips you seem to be so fond of Feb 26 17:21:40 heh Feb 26 17:22:24 I have a few converters with max-something chips Feb 26 17:22:35 work fine for connecting to 3.3V uarts Feb 26 17:22:43 mturquette: it works for me. Feb 26 17:23:01 yo mturquette Feb 26 17:23:10 the closest one has a max202 Feb 26 17:23:38 mturquette: sorry, I mean 3.8 with http://github.com/beagleboard/kernel/tree/3.8 patches Feb 26 17:23:54 so nevertheless thank you for your help :) Feb 26 17:24:50 mturquette: yes it is Feb 26 17:24:58 mranostay: btw, please tell me you didn't find an actual tizen in my photos Feb 26 17:25:12 mturquette: not upstream Feb 26 17:25:27 mturquette: probably 3.10 Feb 26 17:31:22 XorA: howdy Feb 26 17:31:38 mdp: :-( Feb 26 17:32:03 mdp: so i have to pull from some random tree for mmc to work? can you provide a link? Feb 26 17:34:06 http://github.com/beagleboard/kernel/tree/3.8 Feb 26 17:34:27 like jkridner___ mentioned Feb 26 17:34:28 or apply http://linux.omap.com/pipermail/davinci-linux-open-source/2013-February/026265.htm Feb 26 17:35:22 hi Feb 26 17:35:36 mturquette: the reasons why have something to do with those "darn maintainers" as we were discussing about another topic at ELC. Feb 26 17:37:49 koenkooi: ah, i see where jkridner___ mentioned earlier. my nickname highlighting appears to have been turned off... Feb 26 17:38:19 excuses Feb 26 17:38:59 mturquette: you'll find a lot of other useful things in that tree as well..if you are trying to do more than just boot the board...you pretty much need that tree Feb 26 17:39:04 or you'll recreate it yourself Feb 26 17:39:49 panto: internets hating you today? Feb 26 17:39:50 bb_needHelp: sorry not to get you over the hump. Feb 26 17:40:07 you might want to update the bonescript lib as well and enable more debug statements. Feb 26 17:40:13 you might also look at the state of the pin Feb 26 17:40:35 mru, oh yeah Feb 26 17:41:11 panto, going to starbucks again? Feb 26 17:42:21 panto! Feb 26 17:42:29 mdp: that IP belongs to "Hellas On Line S.A." Feb 26 17:42:30 doh Feb 26 17:42:34 gone already Feb 26 17:42:41 must have seen me Feb 26 17:43:16 jkridner___, not really :) Feb 26 17:44:11 hi panto. any chance you can send me a working .dts? my attempts to hack have failed miserably. Feb 26 17:44:19 guess not Feb 26 17:49:33 I'm still here. ;-) Feb 26 17:49:48 panto: hide! Feb 26 17:50:01 jkridner___, spartans don't run... :P Feb 26 17:50:39 no friggin clue why the internets are so finicky today Feb 26 17:51:14 well, you seem to be connected to some kind of greek AOL Feb 26 17:51:21 if the name is any indicator... Feb 26 17:52:06 mru, fortunately the name has nothing to do with the original AOL Feb 26 17:52:08 (I hope) Feb 26 17:52:28 surely they got the idea somewhere Feb 26 17:53:09 back in the beginning of time when AOL was something you'd aspire to be, rather the dried up husk of senior citizen connectivity that is today Feb 26 17:53:13 panto: do I need to update the main board .dts/.dtsi? Feb 26 17:53:18 or just create an overlay? Feb 26 17:53:25 jkridner___, just use an overlay Feb 26 17:53:37 k, my overlay syntax is clearly screwed. :) Feb 26 17:54:10 it was aol that gave us eternal september Feb 26 17:54:42 remember, remember eternal september Feb 26 17:54:54 http://pastebin.com/bfrY1tAq Feb 26 17:55:10 panto: http://pastebin.com/bfrY1tAq Feb 26 17:55:48 jkridner___, you don't have a target at the second fragment Feb 26 17:56:02 something like target = <&ocp>; should work Feb 26 17:57:17 still get "echo: write error: Invalid argument" Feb 26 17:57:42 that pretty much describes panto right now Feb 26 18:00:40 is the address-cells/size-cells necessary? Feb 26 18:09:12 panto: address-cells/size-cells didn't seem to have any impact on the error. Feb 26 18:09:30 jkridner___, missed a bunch of things Feb 26 18:09:36 care to paste the scrollback? Feb 26 18:10:09 http://www.beagleboard.org/irclogs/index.php?date=2013-02-26#T18:09:36 Feb 26 18:10:21 not much there really Feb 26 18:10:48 http://pastebin.com/RCgPn53y doesn't work... gets "echo: write error: Invalid argument" Feb 26 18:11:08 http://pastebin.com/DQQHyxKS is my test script Feb 26 18:11:31 jkridner___, address cells is just to keep dtc from whining Feb 26 18:11:33 it dies at "echo pinctrl-test-7 > $SLOTS" with "echo: write error: Invalid argument" Feb 26 18:12:40 dmesg tells me: "Failed to resolve tree" Feb 26 18:12:46 jkridner___, what is the name of the dtbo file? Feb 26 18:12:48 couldn't find symbol pinctrl_test_7_pins Feb 26 18:13:05 it is in the script Feb 26 18:13:17 pinctrl-test-7-00A0.dtbo Feb 26 18:13:42 jkridner___, the name of the pinctrl label is 'test' not pinctrl_test_7_pins Feb 26 18:13:59 test: pinctrl_test_7_pins -> pinctrl_test_7: pinctrl_test_7_pins Feb 26 18:14:45 * jkridner___ is very confused by the 2-label system Feb 26 18:15:01 it is progress Feb 26 18:15:17 jkridner___ is talktive today Feb 26 18:15:18 the first is the label Feb 26 18:15:21 the other is the node name Feb 26 18:15:22 it seems to have worked the first time, but not unload Feb 26 18:15:35 panto: what was that pin helper patch for exactly? Feb 26 18:15:53 to enable pinmux control settings Feb 26 18:15:59 from userspace (kinda) Feb 26 18:16:00 can't you configure pinmux already? Feb 26 18:16:09 it's a long story Feb 26 18:16:20 i seem to recall a pinctrl-single way in /sys Feb 26 18:17:33 panto: after the first load, the next one gives an error due to the pin already being requested Feb 26 18:18:23 jkridner___, in that case pinmux is not released... Feb 26 18:18:33 hmm, ok, let me take a look at it Feb 26 18:18:53 it is possibly that the devm pinmux doesn't get released properly automatically Feb 26 18:19:24 <_av500_> hmm, panto Feb 26 18:20:38 of_overlay_revert_one: revert on node /ocp/helper Failed! Feb 26 18:21:22 hmm, ok, that's weird Feb 26 18:21:37 let me take a look on a real board Feb 26 18:21:54 want me to send you links to a build kernel? Feb 26 18:22:05 I cranked out the patch while waiting for my pickup to the airport Feb 26 18:23:07 * panto points out that today is jet-lag day #1 Feb 26 18:29:36 panto: no, what you're having is net-lag Feb 26 18:52:29 jkridner___, got the email...looking later today Feb 26 18:52:40 k, thanks Feb 26 19:09:00 hrm, sigrok could really use a clock freq/period measurement "decoder" Feb 26 19:32:03 <_av500_> \o/ #define I2C_BITBANG_DELAY_US5 Feb 26 19:34:21 :) Feb 26 19:34:46 but what's the value? Feb 26 19:37:17 <_av500_> 5, there isa space missing Feb 26 19:37:25 <_av500_> there too Feb 26 19:40:06 _av500_, need to share this with gregkh, he clearly loves bitbanging too Feb 26 19:40:26 <_av500_> its from the ec code Feb 26 19:41:30 one thing I noticed was the have to manually operate their h/w chip select. it seems no SPI IP can handle all the required permutations of keep it asserted or deasserted between transfers Feb 26 19:41:38 s/the have/they have/ Feb 26 19:41:54 I wonder if any spi ip bother to try and handle everything in h/w Feb 26 19:41:56 <_av500_> for lm4? or stm32? Feb 26 19:42:09 lm4, actually Feb 26 19:42:34 same thing most linux drivers do Feb 26 19:43:25 <_av500_> just learned the lm4 has an LPC interface Feb 26 19:43:41 the lm4f120 in particular? Feb 26 19:43:44 <_av500_> somebody wanted to sell into intel mainboard :) Feb 26 19:43:47 or that family does somewhere? Feb 26 19:43:50 <_av500_> mdp: the onethey use Feb 26 19:43:59 ok, interesting, yes Feb 26 19:44:01 <_av500_> in the retina chromebook Feb 26 19:44:09 <_av500_> aka pixel Feb 26 19:44:20 hehe...retina Feb 26 19:44:33 * _av500_ has retina eyes Feb 26 19:45:03 <_av500_> mdp: I will "port" ec to launchpad and tell phoronix Feb 26 19:45:08 I can't afford retina or pixel eyes Feb 26 19:45:27 _av500_: fame shall come to you Feb 26 19:45:31 <_av500_> sakoman can make a pixel cape for it then.... Feb 26 19:45:41 <_av500_> adding the x86 etc... Feb 26 19:45:53 definitely need a pixelLure Feb 26 19:46:29 <_av500_> I thouhgt msp430 lp was stopped, but there is a G2 even Feb 26 19:48:04 it occurred to me today to question why minnowboard has a proprietary connector instead of doing something PCI/104-Express...then I remembered it's Intel Feb 26 19:48:16 <_av500_> :) Feb 26 19:48:32 <_av500_> will there bye a PCIe breadboard lure? Feb 26 19:48:37 <_av500_> be a* Feb 26 19:49:20 no doubt in my mind that we'll see a miniPCIslotLure Feb 26 19:49:31 <_av500_> meh Feb 26 19:49:40 <_av500_> ppl want to handle arduino over PCIe Feb 26 19:49:46 <_av500_> no? Feb 26 19:49:47 wah? Feb 26 19:49:50 but what a horribly limited selection miniPCIe has Feb 26 19:50:13 definitely need a FC HBA ;) Feb 26 19:51:13 <_av500_> found your spi cs via gpio place :) Feb 26 19:51:55 jkridner___, bug confirmed - capes don't get unloaded... Feb 26 19:52:09 panto: thanks! Feb 26 19:52:52 * mdp wonders if learning python will be painful Feb 26 19:53:05 _looking_ at it is painful Feb 26 19:53:32 that's what I think Feb 26 19:53:43 but I need this silly protocol decoder Feb 26 19:53:51 so write one in C Feb 26 19:53:52 <_av500_> use perl Feb 26 19:53:57 perl I can do! Feb 26 19:53:59 yes, or perl Feb 26 19:54:25 well, the sigrok linkage stuff for decoders is all in python..so I have to understand a little in any case Feb 26 19:54:26 text or binary protocol? Feb 26 19:54:55 binary..it's actually really trivial .. just measuring a clock period Feb 26 19:55:09 if java is the new cobol, python is the new c++ Feb 26 19:55:18 mru, +1 Feb 26 19:55:21 all the overhead is figuring out all the sigrok decoder api Feb 26 19:55:26 heh Feb 26 19:55:47 I really need that intern ... stat Feb 26 19:55:54 and ruby is objective-c Feb 26 19:56:49 panto: we can drop the ADC from our list for the time being, it's not critical in the tester Feb 26 19:57:06 * koenkooi covers jkridner___s ears Feb 26 19:57:14 koenkooi, I see Feb 26 19:57:46 idiots at mistral switched back to 3.2.18 Feb 26 19:58:11 they will need some talking to Feb 26 19:58:13 *again* Feb 26 19:58:25 people: never hire mistral for anything, just set your money on fire Feb 26 19:58:32 much more fun Feb 26 19:58:54 * koenkooi goes back to working on the eMMC image Feb 26 19:59:18 can banknotes be used as a base for cellulose nitrate? Feb 26 19:59:52 koenkooi: tell us how you really feel...please don't hold back. Feb 26 20:05:23 dinner Feb 26 20:09:05 mru: can't we make javascript the new objective-c? Feb 26 20:11:51 jkridner___: your love of javascript disturbs me Feb 26 20:12:08 it's everywhere man! Feb 26 20:12:16 dig it Feb 26 20:12:42 * jkridner___ prefers {} languages to the evil tabbing of python Feb 26 20:18:23 <_av500_> real mean prefer () Feb 26 20:18:31 <_av500_> men* Feb 26 20:18:49 * _av500_ loves his new watch running Android Feb 26 20:18:50 http://xkcd.com/297/ Feb 26 20:20:28 <_av500_> now I need to get google+ onto it Feb 26 20:20:58 that's just a galaxy note strapped to your pinky Feb 26 20:21:13 <_av500_> shhh Feb 26 20:21:25 there are real men in here? Feb 26 20:21:40 <_av500_> I'm more of a bignum Feb 26 20:21:55 mranostay: it's hard for you to imagine a real man, I know... Feb 26 20:22:22 * mranostay smacks mru Feb 26 20:22:46 see, he can't even do that properly Feb 26 20:23:00 * _av500_ steps in and separates the brawling kids Feb 26 20:23:15 * mru slaps mranostay around with a large trout Feb 26 20:23:15 <_av500_> no pudding tonight! Feb 26 20:23:24 ^^ that's how it's done right Feb 26 20:23:36 with no pudding? Feb 26 20:23:56 <_av500_> pudding trout, I guess the brits eat that too Feb 26 20:23:56 trout, no pudding Feb 26 20:25:03 if you don't eat yer meat… Feb 26 20:25:14 <_av500_> yer meat will eat you Feb 26 20:25:24 no, you can't have any pudding! Feb 26 20:25:45 <_av500_> with fries? Feb 26 20:25:57 Wrong! Do it again! Feb 26 20:26:06 * mru had burger with "fries" for dinner Feb 26 20:27:01 freedom fries Feb 26 20:27:04 "fries" Feb 26 20:27:08 horse fries Feb 26 20:27:12 mmmm Feb 26 20:27:13 what are "fries" Feb 26 20:27:24 we call 'em chips here Feb 26 20:27:38 you have your flaws, yes Feb 26 20:28:20 like misspelling a few words like color, labor, and etc Feb 26 20:28:29 we need something different for "crisps" here Feb 26 20:28:44 someone trolled my irc client, it's stuck on en-gb spell checking Feb 26 20:28:55 ugh Feb 26 20:29:05 you'll waste bandwidth! Feb 26 20:29:07 heh Feb 26 20:29:13 actually not Feb 26 20:29:32 are you going to confuse us with the facts now? Feb 26 20:29:46 <_av500_> confuse US Feb 26 20:29:49 <_av500_> indeed Feb 26 20:30:18 british english has a larger vocabulary with more precise words Feb 26 20:30:49 so fewer words are needed to express the same thing Feb 26 20:31:22 like hootenanny? Feb 26 20:31:45 +1 Feb 26 20:31:51 that's an american word Feb 26 20:31:56 scottish Feb 26 20:32:05 <_av500_> outlandish Feb 26 20:32:49 that's not what the dictionary says: http://oxforddictionaries.com/definition/english/hootenanny Feb 26 20:35:44 always looking to confuse us with facts Feb 26 20:37:44 would you rather I confuse you with lies? Feb 26 20:37:56 or perhaps some statistics? Feb 26 20:38:33 as an american, I prefer my statistics to be presented in the most misleading way possible Feb 26 20:38:42 and I don't think you are prepared to offer that Feb 26 20:40:27 a google news search for US stories with the keyword "percent" might help though Feb 26 20:43:11 * mdp thanks becky for helping to make hootenanny a community word of choice Feb 26 20:43:25 o.O Feb 26 20:43:50 sleep-attempt-#1 Feb 26 20:43:56 later Feb 26 20:48:35 <_av500_> good lurk Feb 26 21:42:23 jkridner___: uploading the image, will take a few hours Feb 26 21:42:35 thanks Feb 26 21:42:45 koenkooi: woot!!!!1eleven Feb 26 21:45:37 <_av500_> mranostay: please. some people are trying to sleep here Feb 26 21:56:05 * mdp wakes up Feb 26 21:56:58 mranostay: not fri2, bone secret colour Feb 26 21:57:47 Jacmet: the 96 versus 192 MHz ref clk is interesting..as you notice 192 is correct...but even when calculating a dsor divider of half what's needed, things still work here Feb 26 21:59:06 Jacmet: I've confirmed an expected 25 MHz SD Clk with the proper 192 ref clk, but some unexpected results at 96 mhz ref clk (trying to grok how it could even work as by inspection it should be generating a 48 MHz SD Clk for a card that can't handle it.. Feb 26 21:59:45 Jacmet: also, I went back and look at our assortment of vendor trees, rowboat, etc...too many to list Feb 26 21:59:47 mdp: ok, sounds odd indeed Feb 26 22:00:06 mdp: :/ Feb 26 22:00:28 Jacmet: it appears wrong in all of them...as there's an mmc.h which includes that correct 192 value from mmc_host_def.h then redefines it as 96 ;) Feb 26 22:01:00 mdp: heh Feb 26 22:01:33 Jacmet: *sigh*...so we'll make it correct and move on with our lives Feb 26 22:02:13 btw, one of our guys posted a patch to fix a related issue in the kernel driver too Feb 26 22:02:29 as it was doing stupid things with magic values for lower ref clks Feb 26 22:02:31 mdp: something like that indeed. I've just had a day off and Monday I had to work on other stuff, so I haven't made any progress on the 816x stuff since last week yet Feb 26 22:02:37 ok Feb 26 22:02:57 I'm going to go back to the semantical fixup stuff that requires no thinking now Feb 26 22:03:25 * Jacmet is going to push out another buildroot -rc Feb 26 22:03:43 time is running out for actually getting 2013.02 out in February ;) Feb 26 22:03:49 koenkooi, you around? Feb 26 22:03:49 I'm going to have to believe the logic analyzer trace over everything else Feb 26 22:03:56 ka6sox: yes Feb 26 22:04:13 did you push out a later version that rc7? Feb 26 22:04:17 Jacmet: hehe Feb 26 22:04:27 ka6sox: push out what? Feb 26 22:04:33 oh, wait. Feb 26 22:04:53 Jacmet, I've got to post v2 and then rework some kernel crap before -rc1 Feb 26 22:07:28 life would be easier if we just booted the kernel with the kernel Feb 26 22:07:46 felipebalbi: ^^^^ Feb 26 22:08:23 mdp: I used linux-as-bootloader on my xscale PDAs in 2005 or so Feb 26 22:09:18 heh minix 3.2.1 release Feb 26 22:09:30 koenkooi: I had a board in the linuxppc tree that used arch/ppc/boot/* stuff as the bootloader ;) Feb 26 22:15:06 seen bjrosas Feb 26 22:15:13 ? Feb 26 22:15:21 lastseen bjrosas Feb 26 22:16:12 no ibot here Feb 26 22:17:34 koenkooi: that was the era when ARM had 572.3 bootloaders Feb 26 22:18:56 anyone know bjrosas then, of the beagleScan project http://processors.wiki.ti.com/index.php/Beagle_Board_Challenge:_Beagle_Scan Feb 26 22:19:57 well that page hasn't been updated for 2 years Feb 26 22:20:57 mdp: awesome :-) Feb 26 22:20:59 koen, why I never trust vendor symbols or packages Feb 26 22:21:07 mdp: it is possible though with configuration header Feb 26 22:21:46 mdp: imagine how easy it would be to support different boot interfaces :-p Feb 26 22:23:44 Russ: well I hang out in lots of places online for >2 years Feb 26 22:23:50 felipebalbi: heresy! Feb 26 22:24:15 Anyone out there up for trying to answer a bone boot question for me? Feb 26 22:24:19 mdp: hah Feb 26 22:24:36 search for an email in the source commits Feb 26 22:26:24 Oh please dispense some wisdom bone masters... Feb 26 22:27:04 ask the damn question Feb 26 22:27:06 my crystal ball says one must ask question to get answer Feb 26 22:27:23 oh djlewis, no hints plz Feb 26 22:28:49 how do people actually get work done in offices...? Feb 26 22:29:06 by delegating Feb 26 22:29:15 lots of planning Feb 26 22:30:00 first vacate the offoce of all people then move in. work is done after move in. Feb 26 22:30:09 s/offoce/office Feb 26 22:30:11 okay cool. I'm trying to boot a bone that has a cape which contains a MPU and I want to use a UART on the bone with flow control. It appears that every bone UART that contains flow control is mapped right over the mode select pins (P8 pins 31-48) Feb 26 22:30:59 my cape MPU goes tri-state-ish... 40K pullups when held in reset. Feb 26 22:31:24 are you aware of pinmux? Feb 26 22:31:30 so the bone doesn't boot if I use those pins on the expansion header because the boot mode is hosed Feb 26 22:31:40 ah Feb 26 22:32:03 can't you wait to turn those pullups on until the bone is up? Feb 26 22:32:15 er, they are in the mpu Feb 26 22:32:20 yep Feb 26 22:32:28 ok, can't you wait to turn on the mpu until after the bone is up? Feb 26 22:32:29 NXP JN5148 is the part Feb 26 22:33:19 if by turn on you mean release the reset, that doesn't work because the MPU places 40K pullups on all digital pins Feb 26 22:33:23 while in reset Feb 26 22:33:34 if it isn't on, it can't pull anything up Feb 26 22:33:47 just be sure not to drive any of its inputs high Feb 26 22:34:08 your other option is just putting mosfets on the problem pins instead of on the mpu power Feb 26 22:34:49 the cape gets powwered by the bone so it's always on when the bone is on. I don't think I have a way to keep the MPU off. I can hold it in reset, but the 40K pullups in reset defeat the 40K pulldowns on the bone for mode Feb 26 22:35:09 yeah, I thought about the fet solution actually Feb 26 22:35:23 I'm talking about the design of the cape, connecting the mpu's power through fets instead of directly Feb 26 22:35:43 bowzer, like Russ says, you want to design to control the power to the cape from the bone Feb 26 22:35:44 ah, gotcha Feb 26 22:36:14 or, you could just put a pulldown resistor in Feb 26 22:36:25 4.7k or so Feb 26 22:36:48 on the 6502 cape I had a similar issue at first..but only needed proper pulldowns Feb 26 22:36:57 right, defeat the 40K pullups on selected pins Feb 26 22:37:17 you can also look at other cape designs that use modesel pins..like the lcd capes Feb 26 22:37:30 how small of a resistor can the bone drive positive signals on? Feb 26 22:37:41 check the datasheet :) Feb 26 22:37:54 I would approach it from how large I resistor I could use Feb 26 22:38:32 well, yeah, how large so it boots and works. Feb 26 22:38:48 reliably... Feb 26 22:38:59 okay thanks guys, great ideas. Feb 26 22:38:59 again again, the datasheet :) Feb 26 22:39:05 resistance is irrelevant Feb 26 22:39:12 currents and voltages are what matter Feb 26 22:39:13 futile mru, futile Feb 26 22:39:14 yep.. I'm there. Feb 26 22:39:19 that too Feb 26 22:39:25 who ended up winning the resistor packs at elc anyway? Feb 26 22:40:28 heh Feb 26 22:40:31 Russ: hehe Feb 26 22:40:49 Russ: was someone from the Maker group, i have his business card, hehe Feb 26 22:41:02 good fit Feb 26 22:41:08 Nexus 7 to resistor packs.. nice dropoff :) Feb 26 22:41:09 indeed Feb 26 22:41:13 I never win anything useful like that :) Feb 26 22:41:13 makers love resistors Feb 26 22:41:25 mdp, I got a really nice poker set at scale Feb 26 22:41:41 they'll be installed in a makeamatic 1000 manual cnc machine on hackaday Feb 26 22:42:14 Russ: Maker's love THT just love it Feb 26 22:42:18 Russ, I kid..I won a nice Calao at91 board kit at ELC grenoble Feb 26 22:42:27 http://www.amazon.com/Suited-Design-composite-lockable-silver/dp/B000BF6USU Feb 26 22:42:28 anti-makers love TNT Feb 26 22:42:32 Russ, all I had to do was explain how lowsy Android is Feb 26 22:42:39 or lousy even Feb 26 22:43:13 mdp: hehe you don't need to win it, as i recall you got some "useful stuff", hehe Feb 26 22:43:16 well that an tangram fury Feb 26 22:44:15 *sigh* http://thread.gmane.org/gmane.linux.network/260062 Feb 26 22:45:12 prpplague: I got a coveted stack of beaglebone unzipped flyers ;) Feb 26 22:45:24 mdp: hehe Feb 26 22:45:27 Russ: bugs like that are why you only enable the protocols etc you actually need Feb 26 22:45:30 prpplague: I want for nothing else Feb 26 22:45:48 prpplague: showed my wife that..first thing was, "sounds like an HR violation" :) Feb 26 22:46:21 mdp: yea i was going to ask jessica about how she came up with that, but i figured i might get in trouble Feb 26 22:46:25 heh Feb 26 22:46:34 prpplague, now you're in the clear Feb 26 22:46:48 prpplague: between the color of the board, the existing name, the unzipped..the whole thing is way too dangerous Feb 26 22:46:59 Russ: thursday after 5pm Feb 26 22:47:02 I feel guilty being associated with such a thing Feb 26 22:47:03 oversensitive americans Feb 26 22:47:06 mru, :) Feb 26 22:47:27 mru, HR lurks around every corner Feb 26 22:48:04 in europe nobody would bat an eyelid at that Feb 26 22:50:20 most of europe has less to feel guilty for Feb 26 22:54:37 mdp, so its ok for a board to be white, but not to be black? Is that what you're saying? Feb 26 22:57:00 no no! I'm not saying anything! Feb 26 22:57:02 * mdp runs Feb 26 22:57:09 aieee Feb 26 22:57:34 * _av500_ throws creeper cards around Feb 26 22:57:38 <_av500_> and scooby snacks Feb 26 22:57:52 hr drone got a new meaning? Feb 26 22:58:03 we need to chris tucker in here and his 500 dollar chips Feb 26 23:01:05 <_av500_> Z1 has powerful Dual-core processor and CPU Frequency 280+460MHZ design. Feb 26 23:01:17 <_av500_> 460MHZ is for processing high-class level multi-media application and 280MHz is for basis application. Feb 26 23:01:24 <_av500_> right Feb 26 23:01:31 what are you reading? Feb 26 23:01:36 something from 1995? Feb 26 23:01:42 cpufreq? Feb 26 23:02:01 <_av500_> mru: my awesome android watch Feb 26 23:02:07 http://www.neuwatch.com/Watch%20Phone-Android%20watch-Sports%20watch-ProductDetails/Android%20Watch%20Phone-Z1.htm Feb 26 23:02:55 <_av500_> indeed Feb 26 23:02:55 tell me more about this "audio-visual enjoyment is infinite updating" Feb 26 23:03:07 <_av500_> Russ: it's beyond words Feb 26 23:03:34 I'm pretty excited about sharing my happiness in time via bluetooth Feb 26 23:03:47 are you truly the social master? Feb 26 23:04:17 <_av500_> now that I own it, yes Feb 26 23:09:40 hmmm Feb 26 23:16:22 <_av500_> ...nand... Feb 26 23:20:01 heh http://www.flickr.com/photos/13825348@N03/8508473453/ Feb 26 23:20:37 <_av500_> heh, and there is the kid too Feb 26 23:29:25 prpplague is smiling! http://www.flickr.com/photos/13825348@N03/8508477491/in/photostream/ Feb 26 23:31:30 a very nefarious smile Feb 26 23:31:51 strange, prpplague looks taller in irc ;) Feb 26 23:32:32 <_av500_> he is smiling because despite all the stickers his laptop runs Windows Feb 26 23:33:18 mru: psu just arrived, going to go and fashion up some kind of mount now Feb 26 23:33:27 <_av500_> POIDNH Feb 26 23:38:04 it's an evil vendor tree grin Feb 26 23:38:09 I've seen koen with that grin Feb 26 23:38:35 anyone know a work-around to see uSD card when you boot off of eMMC? Feb 26 23:39:22 jkridner___: emmc muxed with the uSD slot? Feb 26 23:39:34 no, on different host controllers Feb 26 23:39:59 that's just a weird question on a surface Feb 26 23:40:04 s/a/the/ Feb 26 23:40:17 is something broken on our part? Feb 26 23:40:19 seems odd that it doesn't work. Feb 26 23:40:31 looks that way. Feb 26 23:40:53 the eMMC becomes /dev/mmcblk0 and the uSD doesn't get dev nodes created. Feb 26 23:41:13 I wouldn't have been surprised if it was reversed, but it is just missing. Feb 26 23:41:19 dmesg didn't show anything interesting. Feb 26 23:41:27 if you don't have /dev/mmcblkM and /dev/mmcblkN and both devices are instantiated with media then it's broken Feb 26 23:41:59 the /sys/devices entries have mmc.3 and mmc.4 with mmc.4 being the eMMC Feb 26 23:42:23 then it's broken. :) Feb 26 23:42:51 that's my professional opinion Feb 26 23:44:55 jkridner___: dmesg so both mmc devices? Feb 26 23:45:10 dmesg only shows the eMMC Feb 26 23:45:15 never mentions the uSD Feb 26 23:53:47 * _av500_ blames ds2 Feb 26 23:55:57 yes blame NAND-boy over there :P Feb 26 23:59:39 _av500_: should have used ferrite core memory Feb 27 00:00:29 now there is a cable idea Feb 27 00:00:32 *cape Feb 27 00:00:52 <_av500_> core cape Feb 27 00:00:55 <_av500_> cape core Feb 27 00:01:17 if Russ though my RS232 to TTL cape was overkill :) Feb 27 00:01:32 mranostay, indiegogo or kickstarter? Feb 27 00:01:47 <_av500_> mranostay: your RS232 cape lacked the proper 25pin connector anyway.... Feb 27 00:01:57 +1 Feb 27 00:02:13 mranostay, write something up about OpenCL and OpenCV support to go with it Feb 27 00:04:15 IIRC, there was a bug in how devices get instantiated Feb 27 00:04:26 but the last time I looked at that was before 2.6.30 so... Feb 27 00:04:52 so? nobody ever fixes bugs Feb 27 00:06:26 <_av500_> documented bugs become features Feb 27 00:06:48 and certainly nobody fixes features Feb 27 00:07:25 it's easier to get the author tag for features...bugs are hard Feb 27 00:11:52 jkridner___: i think you got everyone with the stache cam Feb 27 00:12:04 hehe. got quite a few. Feb 27 00:12:34 everyone in what sense? Feb 27 00:13:06 the intersection of #beagle and ELC? Feb 27 00:13:06 mru: https://twitter.com/BeagleStache/status/304326221158817793/photo/1 Feb 27 00:24:26 jkridner___: ewww, i look like i am saying something very sarcastic Feb 27 00:25:06 see i still have my signal11 sticker Feb 27 00:25:15 unlike some people i didn't cut the sticker Feb 27 00:27:01 damm Feb 27 00:27:18 man alan_o really makes my baldness stand out - http://www.flickr.com/photos/13825348@N03/8508478311/in/photostream/lightbox/ Feb 27 00:27:59 going to have to tease alan_o about his hair next time - http://www.flickr.com/photos/13825348@N03/8508478673/in/photostream/ Feb 27 00:29:23 mdp: http://www.flickr.com/photos/13825348@N03/8509587476/in/photostream/lightbox/ Feb 27 00:29:33 mdp: you can see all the same emails from GM Feb 27 00:31:32 hehe...my new customer Feb 27 00:31:38 there's one where alan_o is really animated Feb 27 00:31:50 alan_o is always animated! Feb 27 00:32:02 better animated than decimated Feb 27 00:32:27 hehe "omg that's an omap5 board in that picture!" Feb 27 00:32:55 hey they got some good pictures this time Feb 27 00:33:40 mr.rainman - http://www.flickr.com/photos/13825348@N03/8508473741/in/photostream/lightbox/ Feb 27 00:34:35 guys, this is what animated looks like: https://lh4.googleusercontent.com/-7td6p0Kq3Xk/T8cwLpelxuI/AAAAAAAABQ0/oI0IrD4--L0/s982/IMG_1296.JPG Feb 27 00:34:36 alan_o, you're going down Feb 27 00:36:20 they posted those quick Feb 27 00:36:40 they aren't done yet, it only goes through the morning of the 21st Feb 27 00:37:35 mru: no that is what ADD, four double expresso, and a monster drink Feb 27 00:37:48 mru: look like Feb 27 00:37:49 prpplague, did he mention that the nordstroms group came over because they thought he was a girl? Feb 27 00:38:09 Russ: alan? Feb 27 00:38:11 well, because they thought he was a girl Feb 27 00:38:11 ya Feb 27 00:38:17 hehehe Feb 27 00:38:17 nordstroms? Feb 27 00:38:33 * prpplague wonders what embedded group that is Feb 27 00:38:48 another group at the parc 55 bar Feb 27 00:39:02 Russ: hehe Feb 27 00:39:16 I'm sure you were aware of them, just not who they were Feb 27 00:39:21 * prpplague can't believe he is trolling around looking for a used atom dev kit Feb 27 00:39:38 prpplague: to kick across the room? Feb 27 00:39:50 Russ: i only have eyes for my bajan beauty, or else she's cut them out Feb 27 00:40:21 its not that, they were a bit disruptive Feb 27 00:41:59 love random people expressions in the photos Feb 27 00:42:28 I'd love to see some of the ones they decided weren't worth posting Feb 27 00:43:00 when shooting at an event like that you end up with hundreds of crap photos Feb 27 00:43:05 they're not funny, just crap Feb 27 00:43:22 bits are cheap Feb 27 00:43:28 sure Feb 27 00:43:56 there's just no value in keeping or seeing them Feb 27 00:44:17 enough of the SpaceX guy Feb 27 00:44:24 you'll have lots of shots where someone stepped in front of the camera just as you pressed the trigger Feb 27 00:44:29 or was that the Google guy Feb 27 00:44:42 spacex Feb 27 00:44:51 didn't catch that one Feb 27 00:45:24 so many macbooks Feb 27 00:46:26 and still I kept getting accosted by random people querying about my laptop Feb 27 00:47:09 are macs not that stylish after all? Feb 27 00:47:15 * wmat noticed the macbooks too Feb 27 00:47:51 http://i.imgur.com/dsLMq.jpg Feb 27 00:48:07 nice Feb 27 00:48:44 I think the acer zenbook ux31a prime is a pretty comparative piece of hardware (not pictured) Feb 27 00:48:55 samsung series 9, bummer about the sd size Feb 27 00:49:23 er, ssd Feb 27 00:49:47 there's this: http://www.dell.com/ca/enterprise/p/xps-13-linux/pd Feb 27 00:49:50 hehe i still have a good laugh about the spacex deal Feb 27 00:49:58 what did I miss? Feb 27 00:50:02 * prpplague puts that one down for his grandchildren story Feb 27 00:50:25 Russ: mru was trolling Robert Ross(spacex) and i happened to pass by Feb 27 00:51:05 Russ: gave him one of business cards, and told him that the next time they had a really difficult hardware/software interface problem to debug, feel free to call me Feb 27 00:51:07 hehe Feb 27 00:51:10 prpplague: won't CCO give you a minnow? Feb 27 00:51:54 Olof's talk was packed I see Feb 27 00:51:56 does anyone know how to make the beagleboard xM boot from usb instead of the sd/mmc Feb 27 00:52:13 ??? Feb 27 00:52:21 angelr91: with anger and yelling Feb 27 00:52:30 angelr91: maybe some chickens blood and tanus root Feb 27 00:52:42 hahaha Feb 27 00:52:55 well I need to make the beagleboard boot from usb since Feb 27 00:53:03 prpplague: you and alan_o are easy to spot in pics Feb 27 00:53:04 i have ubuntu and it runs really slow Feb 27 00:53:09 for exact opposite reasons Feb 27 00:53:28 angelr91: booting from usb and having the root file system on usb are two different things Feb 27 00:53:31 mranostay: indeed Feb 27 00:53:49 opposite ends of the same reason Feb 27 00:54:03 prpplague could you explain Feb 27 00:54:14 wmat, no 1080p Feb 27 00:54:28 er, nm, it does now Feb 27 00:54:31 didn't before Feb 27 00:54:48 without 1080p it is worthless Feb 27 00:55:00 ah, about 2 weeks ago Feb 27 00:55:14 I installed ubuntu on an sd and managed to make the BB xM boot into ubuntu but it runs really slow and I need to see how I can make the BB xM boot from usb into ubuntu Feb 27 00:55:20 that is my problem currently Feb 27 00:55:36 mranostay, I was walking between yours and olof's talk quite a bit Feb 27 00:55:43 and 8GB ram, nice Feb 27 00:56:44 angelr91: http://www.omappedia.com/wiki/Ubuntu_on_OMAP_FAQ#I_want_to_install_Ubuntu_on_external_USB_hard_disk_instead_of_sluggish_SD_card Feb 27 00:56:46 Russ: they just upgraded it, I thought to full 1080p Feb 27 00:57:05 yes, that's what I said Feb 27 00:57:09 * wmat runs to play with kids Feb 27 00:57:16 cool I will check this one out Feb 27 00:57:20 8GB ram is not "nice", it's the bare minimum Feb 27 00:57:39 well you don't get many options in the soldered on form factor Feb 27 00:57:51 getting more is mostly impossible Feb 27 00:57:57 but getting less is stupid Feb 27 00:58:07 oh wait i thought you said 8mb Feb 27 00:58:07 time for hot air rework? Feb 27 00:58:10 google says 4GB is the bare maximum Feb 27 00:58:10 hehe Feb 27 00:58:14 koenkooi: is also easy to spot :P Feb 27 01:00:49 heh my room had more people that i remember Feb 27 01:01:48 what was the seating capacity of those rooms? Feb 27 01:02:55 cool my overkill PCB has been sent to fab Feb 27 01:03:32 did you decide to stick with dip? Feb 27 01:13:16 yeah Feb 27 01:13:34 next you'll be hacking z80 Feb 27 01:18:35 this is your brain: O Feb 27 01:18:46 this is your brain on x86 dev: #$%#$%#$%!@@$!@#$!@#$!@#$! Feb 27 01:20:54 prpplague: hope they are paying you well :) Feb 27 01:21:27 you're halfway there once you go little endian Feb 27 01:21:28 there would be an intel multipler for any job i'd take :) Feb 27 01:22:05 mranostay: kind of like the "multiplier" i use for moving to the bay area? Feb 27 01:22:46 touche Feb 27 01:23:58 multiplication values: 1.no way jose 2. well maybe if hell is chilly 3. what kind of crack are you smokin? 4. that kind of sounds illegal 5. oh yea getting a tesla with a custom C64 installed in the dash board Feb 27 01:25:59 * prpplague is reminded he needs to send out some tutorial links Feb 27 01:31:12 heh 2 is about what you can expect Feb 27 02:09:51 mru: 13.7degrees Feb 27 02:09:56 im happy Feb 27 02:09:58 :-) **** ENDING LOGGING AT Wed Feb 27 02:59:58 2013