**** BEGIN LOGGING AT Fri Jun 29 02:59:59 2012 Jun 29 04:45:10 * _av500_ just found an Osborne out in the rain... Jun 29 05:13:33 _av500_: ossie? Jun 29 05:38:27 <_av500_> Osborne 1 from 1981 Jun 29 05:38:44 <_av500_> got an angstrom build for it? Jun 29 05:39:37 mehhhhhhhhhhhh Jun 29 06:11:23 hi Jun 29 06:12:14 i have one question which is best bluetooth gps receiver Jun 29 06:12:40 i want to interface Bluetooth gps receiver to beagleboard Jun 29 06:15:39 i have one question which is best bluetooth gps receiver, i want to interface Bluetooth gps receiver to beagleboard Jun 29 06:19:13 i think they are pretty much the same Jun 29 06:48:02 i have one question which is best bluetooth gps receiver, i want to interface Bluetooth gps receiver to beagleboard Jun 29 06:48:47 why don't you fix your gpsd first Jun 29 06:49:27 yeah i fixed gpsd Jun 29 06:50:27 bt gps should be as easy as: pair it, configure a rfcom port, use it in gpsd Jun 29 06:50:42 Now i want to interface Bluetooth gps receiver it can work in linux,angstron OS Jun 29 06:52:27 * mranostay wants to shoot people that repeat questions on irc Jun 29 06:53:01 When we purchase Bluetooth GPS receivers they will give cd it will work fine in windows ,beagle board does not any cd rom ,my doubt is can any bluetooth gps receiver works for beagleboard or not? Jun 29 06:53:06 mranostay: submit a RFC for transmission of lethal force over IP Jun 29 06:54:35 dm8tbr: i'd be serving multiple life sentences if that was possible :) Jun 29 06:58:55 hi sundar Jun 29 07:01:08 mranostay: only if you wouldn't use a proper anonymizer ;) Jun 29 07:02:19 hi Jun 29 07:30:31 I have a beagleboard-xm rev C, and it hangs on boot with a fresh build from narcissus. Is there a fix for this, or do I need to use the Angstrom demo image? Jun 29 07:31:21 it hangs where? Jun 29 07:32:11 First it hangs with only the beagleboard logo showing up. I then added a Uenv file, where it now hangs when it find the computer mouse Jun 29 07:33:00 oh, you didn't hook up a serial cable... Jun 29 07:34:00 nope, generating build and running it alone with monitor, keyboard and mouse Jun 29 08:24:07 is this blog entry up to date? http://beaglexmc.blogspot.no/ Jun 29 08:33:44 hmm, wonders what a university that is where students are porting ported distributions. Jun 29 08:34:24 aholler_: one that does not care much about the semantic correctness of their project descriptions :) Jun 29 08:35:48 maybe they just have invented the wheel. Jun 29 08:36:51 aholler: https://plus.google.com/u/0/photos/116338049603737180343/albums/5758648266194395265/5758648265798503074 Jun 29 08:41:00 what mean porting angstrom to beagleboard ? O_o Jun 29 08:43:10 http://en.wikipedia.org/wiki/Plenken Jun 29 08:43:56 * stuk_gen solved the problem of dhcp bb xM...simply his dhcp server is full -.- Jun 29 08:48:18 hi Jun 29 09:13:19 hi everyone Jun 29 09:26:50 moin! Jun 29 09:27:32 excellent the post from maddy to the list : "Sorry, if you think that i'm miss behaving" Jun 29 09:27:55 had a good laugh :) Jun 29 09:36:39 q simpel question what is the difference from using a sdk from texas and a build from angstrom oe? what is the best stable? Jun 29 09:37:37 angstrom is supported here more or less Jun 29 09:37:43 ti sdk is supported by TI Jun 29 09:38:41 av500: ok...but what is the best solution for an company? Jun 29 09:38:55 http://radar.oreilly.com/2012/06/why-learn-c.html Jun 29 09:39:06 I've read the comments and now I want to die Jun 29 09:39:30 av500: i think texas sdk is stable but old...vs....angstrom image is update...but maybe is not more stable? Jun 29 09:42:48 stuk_gen - well, I personally like the angstrom solution (for a company), because you can build your own image with only the tools you need and no more bloat Jun 29 09:43:12 guyzmo: ok and this is a great option...but about stability? Jun 29 09:44:05 and, as we use continuous integration, when the build of our own code is stable we can trigger an image build with the last stable version to always have an image ready for production Jun 29 09:44:12 ++agility :) Jun 29 09:44:26 stuk_gen - well, what you mean by stability ? Jun 29 09:45:31 stuk_gen - angstrom definitely has bugs, but most of them are minor and can easily be patched that you can write and keep in your own repository Jun 29 09:45:37 panto: I dont see any comments Jun 29 09:45:38 guyzmo: i mean not the stability of 'our' code...but the oe code...i mean today building an image is different that build an image tomorrow..because oe have a lot of update every day Jun 29 09:45:52 err, av500 comment in the lwn page Jun 29 09:45:58 linky? Jun 29 09:45:59 http://lwn.net/Articles/504023/ Jun 29 09:46:26 stuk_gen - well, it's only been a few months I'm working with the beaglebone Jun 29 09:47:03 stuk_gen: well, then dont update Jun 29 09:47:09 and I have seen no major regression Jun 29 09:47:14 guyzmo: i think the problem is you have to test if the image really works or not Jun 29 09:47:19 pick a release/git hash and stick with it Jun 29 09:47:31 make sure the features you need work Jun 29 09:47:34 av500: yes this i think is the only solution Jun 29 09:47:36 and then stabilize your SW Jun 29 09:47:43 stuk_gen - at each git pull/update of the tools, you'll need to test the image for regression Jun 29 09:47:46 this is regardless of ti sdk or oe or anything Jun 29 09:47:56 av500: no Jun 29 09:48:06 enlighten me Jun 29 09:48:14 me too, please. Jun 29 09:48:21 av500: because i think when ti give out a new sdk release...i think they have tested yet Jun 29 09:48:31 tested *what* Jun 29 09:48:35 av500: and the release is not every day Jun 29 09:48:41 stuk_gen - you think, but you don't know Jun 29 09:48:44 that all they give to you works Jun 29 09:48:49 and you'll have to test anyway Jun 29 09:49:28 why would one ever release a second version of a sdk if everything was tested and fully functional in the already released one? q.e.d? Jun 29 09:49:39 I think the best solution is to update on a regular basis the whole setup-tools and sources, generate a new image, test it for regression, patch if you find out regressions and stick with it until next update Jun 29 09:50:03 LetoThe2nd: maybe because there improvement Jun 29 09:50:04 yes, if you are a linux distro Jun 29 09:50:13 not if you make a PRODUCT Jun 29 09:50:28 you have your set of features, once they work, why upgrade anything? Jun 29 09:50:35 maybe for the next product Jun 29 09:50:38 stuk_gen: what might improve if everything was 100% functional and tested? reach 100% functionality? Jun 29 09:50:44 111% Jun 29 09:50:47 110%, even. Jun 29 09:50:48 av500 - security and continous availability of the tools Jun 29 09:50:59 yeah, security Jun 29 09:51:14 security and improvement Jun 29 09:51:28 yes, "security" is a nice buzzword Jun 29 09:51:32 now lets define what it means Jun 29 09:51:37 stuk_gen: you still fail to specify "improvement." Jun 29 09:51:44 and your product is aimed at evolve, and may need new dependancies Jun 29 09:52:03 improvement...for example...qt 4.6 and 4.8 is a big change for graphics Jun 29 09:52:06 and if a dependancy needs a full chain of new versions of all the tools you're using, if you don't update Jun 29 09:52:06 its too speed Jun 29 09:52:06 if you have the means to regression test everything, sure upgrade all you want Jun 29 09:52:10 you're doomed Jun 29 09:52:28 av500 - that's why we use continuous integration :) Jun 29 09:52:46 stuk_gen: why should one care for that if you already claimed that your last revision was fully tested and functional? Jun 29 09:52:53 we use continous bug fixing Jun 29 09:52:55 I'm even looking out to do CI on microcontroller code Jun 29 09:53:05 CI? Jun 29 09:53:10 continuous integration Jun 29 09:53:14 ah Jun 29 09:53:44 and no word about tests? Jun 29 09:53:52 stuk_gen: you just have to accept that no sdk/basesystem/whatever will ever have guarantees for buzzwords like "stability, tested, secure" Jun 29 09:54:25 LetoThe2nd: for a product yes...have no sense...but for example i made an application that use opengl...today works but its slow..maybe tomorrow an update make speed up...so i want use that Jun 29 09:54:26 aholler: i don't always test my code, but when i du, i do it in production. Jun 29 09:54:34 and I think that one of the best ways to avoid security problems, is to keep the bloat out Jun 29 09:54:45 keep *only* what you need for your product to work, and no more Jun 29 09:54:54 security depends a lot on what product you make Jun 29 09:55:07 stuk_gen: at the same moment admitting that your last product was not 100% functional, because it was too slow. Jun 29 09:55:08 how it is connected etc etc Jun 29 09:55:12 av500 - of course, but once something is connected to a network, security matters Jun 29 09:55:20 LetoThe2nd: no Jun 29 09:55:55 you don't want to run a service that is not useful and that can be easily exploited by a 10y/r Jun 29 09:56:01 y/o* Jun 29 09:56:01 LetoThe2nd: this is not true! for now works but slow. But i can do nothing this is what today i can get. this is the 100% today Jun 29 09:56:01 LetoThe2nd: using the user as tester isn't an option for every product Jun 29 09:56:12 stuk_gen: then you are claiming that speed does not count of part of functionality and can therefore totally be disregarded anyways? Jun 29 09:56:33 aholler, LetoThe2nd - that's what does microsoft since 1978! :) Jun 29 09:56:44 aholler: was not meant totally serious ;) Jun 29 09:56:52 LetoThe2nd: the speed is important but if this is the maximus speed that i can use today i have nothing todo...this is my product...no? Jun 29 09:57:33 stuk_gen: or are you rather saying that you define your 100% new everyday, just hoping that you might archive your randomly defined aims one day by accident? Jun 29 09:58:25 sounds like a whole lot of contradiction in itself. yay for buzzwords, therefore. Jun 29 09:58:28 LetoThe2nd: but its software :) how can you said that if can or not have improvement? Jun 29 09:59:03 stuk_gen: no. i claim that no software is perfect. hence, i can never have 100% tested and functional. you however did that. Jun 29 09:59:09 LetoThe2nd - got one more buzzword for you: I chose angstrom because it helps us be agile on our product development Jun 29 09:59:12 :p Jun 29 09:59:16 if you use a 3rd party library like qt...you can think future release have more functionality and maybe more speed Jun 29 09:59:35 and that future release will break your code Jun 29 09:59:57 and have new bugs Jun 29 10:00:15 continues integration means continuesly new bugs ;) Jun 29 10:00:26 aholler - yup, but new, better and faster bugs! :p Jun 29 10:00:28 this is true Jun 29 10:00:29 stuk_gen: and like i said - if on the other hand you are free to redefine your 100% anytime the way you like it, the number itself is not worth anything. Jun 29 10:01:02 aholler - nope, it just means that you won't break your code when you'll patch it in live Jun 29 10:01:22 ok..100% is a number...i mean only today you have your product that run...but maybe tomorrow you can run your product with improvement Jun 29 10:01:42 stuk_gen: so you admit you are actually just buzzwording :) Jun 29 10:01:49 great, NEXT TICKET! :) Jun 29 10:02:12 the start is angstrom oe VS sdk ti Jun 29 10:02:28 i'm only tell about this not generate a war XD Jun 29 10:03:06 and a release from ti i think is 'testes' vs every release of oe angstrm Jun 29 10:03:08 aonly this Jun 29 10:03:11 only* Jun 29 10:04:31 so the conclusion is...using oe angstrom you need build image and test. When you have an image that works and is teste. Copy the rootfs,the bootloader. And this is your product 1.0 Jun 29 10:05:15 if i lost the 'image' i need know the hash to git from repository and rebuild all Jun 29 10:05:29 and have a copy of my program :P Jun 29 10:05:41 this is the correctly roadmap...ok? Jun 29 10:05:57 you are forgetting that you need a copy of the build-environment too Jun 29 10:06:08 and all sources, etc. Jun 29 10:06:48 * LetoThe2nd prefers ptxdist, but thats only a sidenote for now. Jun 29 10:07:17 LetoThe2nd - ptxdist ? Jun 29 10:07:34 guyzmo: ptxdist.org Jun 29 10:08:14 hm.. Jun 29 10:08:20 yup, I was already on it Jun 29 10:08:27 guyzmo: not exactly the same as oe, but comparable in some points. Jun 29 10:08:54 but one thing I like about oe or angstrom, is that you can remote patch the systems Jun 29 10:09:11 ptxdist looks like another buildroot Jun 29 10:09:12 if you're managing your feed Jun 29 10:11:29 you can dothe same with wget, tar and a bit of bash-scripting Jun 29 10:11:39 aholler: yeah they have a lot in common now. when they both started, it was like uclibc<->glibc Jun 29 10:15:28 except that the source ptx wasn't availableto the public ;) Jun 29 10:15:45 ? Jun 29 10:16:05 stuk_gen: unless you know exactly what TI "tests" in SDK, I would not trust that "testing" Jun 29 10:16:11 LetoThe2nd: when I'velast had a look at it, no source was available Jun 29 10:16:24 who needs source, there is always XDA Jun 29 10:16:25 av500: would you trust it if you knew? Jun 29 10:16:33 aholler: for ptxdist?!? i seriously doubt that. Jun 29 10:17:05 it was the case, only by request for customers or such. Jun 29 10:17:46 aholler: ah not... thats only for bsps they developed for paying customers Jun 29 10:20:17 aholler: ptxdist itself as well as toolchain and some more or less generic bsps are free, and certainly have been for at least 6 years Jun 29 10:38:03 av500: ok...but there a TI behind...not a hobbies people :D Jun 29 10:38:19 ti is a hw company Jun 29 10:40:05 I followed this guide, but it hangs at boot Jun 29 10:40:06 http://beaglexmc.blogspot.no/ Jun 29 10:40:16 INIT: no more processes left in this runlevel Jun 29 10:40:39 and then it slowly says INIT: Id "1" respawning too fast: disabled for 5 minutes Jun 29 10:41:55 also chown: invalid user 'gdm:gmd' Jun 29 10:43:14 stuk_gen: wouldyou describe yourself as hobbyist or pro? Jun 29 10:46:37 stuk_gen: so? do you have an idea what they test? is there a test coverage document? Jun 29 10:46:50 if not, assume the test is "it boots and does not explode" Jun 29 10:47:30 ok...if you call it Jun 29 10:49:11 av500: what if it HCFs? Jun 29 11:14:59 * bradfa cheers for Friday! Jun 29 11:20:00 \o/ Jun 29 11:20:21 do I want to know what that is? Jun 29 11:24:17 f*ck yeah it's f*ckin' friday! Jun 29 11:26:47 bradfa: you don't know what friday is? Jun 29 11:26:59 mru, apparently not... Jun 29 11:27:03 :) Jun 29 11:27:06 dude, you're missing out Jun 29 11:27:12 oooo! do tell! Jun 29 11:27:17 what am I missing?!?! Jun 29 11:27:29 crazy on the mailing list - check Jun 29 11:27:38 black coffee - check Jun 29 11:27:47 ? - check? Jun 29 11:30:47 yesterday afternoon, rookie provided me with a kernel patch to enable poweroff of the tps65217 after a 'shutdown -h', I proceeded to hack it into half its length (which then relies on u-boot doing the right thing) and where it'll have to try to turn off 2 times if the first attempt's time of day seconds end in :x9 Jun 29 11:31:01 if anyone else is interested, please let rookie or myself know Jun 29 11:31:41 rookie's version was a little more elegant than mine Jun 29 11:32:59 and apparently I didn't burn down the building yesterday with my lithium battery charging, which is always good Jun 29 11:34:21 party? Jun 29 11:37:10 pooper Jun 29 11:45:21 av500, that's what I thought... Jun 29 11:56:55 \o/ new boot loader: https://plus.google.com/115547683951727699051/posts/gDfHgApPma1 Jun 29 11:57:03 http://freedesktop.org/wiki/Software/gummiboot Jun 29 11:59:24 heh Jun 29 11:59:35 i wasnt aware that kay works for RH now Jun 29 12:02:25 av500: without RTFA, does it at least support CWM? *duck* *run* Jun 29 12:06:16 dm8tbr: I guess xda can binary patch that in Jun 29 12:10:55 av500: +1 Jun 29 12:32:55 av500: I think you overestimate the mad h4ck3r skillz of the average xda member Jun 29 12:38:08 he's not overestimating their _est1m4t3_ of their skillz Jun 29 12:39:52 jannau: btw, JB on gslate, when? Jun 29 12:41:36 I have a kernel/multicast problem. Does anybody know a mailing list/newsgroup/forum for linux/networking problems? I don't get an answer at comp.os.linux.networking (low traffic there). Jun 29 12:44:07 LKML? Jun 29 12:45:22 Aren't they only for patches and bug reports? Jun 29 12:45:37 av500: as soon as you leak the pdk Jun 29 13:01:13 hallo Jun 29 13:02:14 is external debugger required for beagle-xm Jun 29 13:02:31 for debugging linux and HAL Jun 29 13:02:41 what hal? Jun 29 13:02:55 hardware abstraction layer Jun 29 13:02:56 HAL 9000 Jun 29 13:03:16 bradfa: please, when experimenting with lithium battery charging...do engage g+ party mode! Jun 29 13:03:20 paggy: same answer as yesterday Jun 29 13:03:44 what Jun 29 13:04:01 mdp, party mode, ENGAGED! Jun 29 13:04:17 been up for 50 minutes so far, just idling printing the date every 10 seconds Jun 29 13:04:27 [14:31:01] can anybody tell me is there need of external debugger for beagle board- xm Jun 29 13:04:28 [14:32:21] no Jun 29 13:05:09 bradfa, ....fire extinguisher - check! ;) Jun 29 13:05:21 for HAL debugging Jun 29 13:05:28 av500 let bet he will join under new name and asking the same Jun 29 13:05:30 mdp, I asked about fire extinguisher yesterday. answer wasn't reassuring Jun 29 13:05:32 hi ynezz Jun 29 13:05:36 we have one but it hasn't been inspected in a while Jun 29 13:05:43 I'm sorry dave, but I can't let you do that. Jun 29 13:05:45 and kodak (who owns the building) wants to remove them... Jun 29 13:05:54 luckily we have lots of asbestos Jun 29 13:06:04 bradfa hm kodak is dead Jun 29 13:06:15 bradfa good for cancer Jun 29 13:06:26 .hallo any body knows that Jun 29 13:06:26 woglinde, queue monty python sketch Jun 29 13:06:27 bradfa, "we have this one with the tag from 1982!" Jun 29 13:06:37 mdp, almost! Jun 29 13:06:58 woglinde, they're not dead yet, coal fired power plant on site is still running, I can see it out the window Jun 29 13:07:04 and power is still on Jun 29 13:07:10 granted, bone is on battery, just in case :) Jun 29 13:07:18 paggy: the answer is NO Jun 29 13:08:01 we could always sell him some lauterbach... the commission on this stuff must be awesome. Jun 29 13:09:21 paggy use the serial cable for debug Jun 29 13:09:34 coal power plant on kodak site: http://goo.gl/maps/TDgO still running! Jun 29 13:09:35 dm8tbr *g* Jun 29 13:09:56 I think they shut the nuclear plant down though :( Jun 29 13:10:03 it was just a little one Jun 29 13:11:46 bradfa, you *may* need to run on battery 24/7 then Jun 29 13:12:27 bradfa, enjoy off mode Jun 29 13:13:09 mdp, ya Jun 29 13:13:34 bradfa hm ah rochester Jun 29 13:14:09 woglinde, it's not snowing today! :) Jun 29 13:14:15 lol Jun 29 13:14:21 we have summer now Jun 29 13:14:34 after the rain the last 2 weeks Jun 29 13:14:44 but I have some cold Jun 29 13:16:08 bradfa, hot there yesterday? Jun 29 13:16:16 it's been warm lately Jun 29 13:17:24 yeah, we hit 38c yesterday which is pretty much our limit here Jun 29 13:17:38 or an average day when I lived in phx Jun 29 13:18:03 my sister-in-law is moving to AZ from Rochester, it's warmer there in January than it is here in July Jun 29 13:18:30 bradfa, it gets cold at night in the winter..despite the heat island effect...but yeah Jun 29 13:18:55 we liked it there too Jun 29 13:28:57 is there a way to set default setting for alsamixer? ever reboot i have to re set up everything Jun 29 13:29:42 look at alsa state Jun 29 13:29:59 woglinde: ? Jun 29 13:32:18 google it Jun 29 13:32:19 alsactl Jun 29 13:36:18 Hi! I'm looking for a kernel repo with functioning framebuffer driver Jun 29 13:37:04 so far I'm using https://github.com/koenkooi/linux.git for am3359 Jun 29 13:40:05 master branch Jun 29 13:40:43 the first problem I encoutered was that this driver doesn't suppor FB_BLANK_BLANK Jun 29 13:40:56 is there a trick to force fsck on mmc-partitions mounted by udev? Jun 29 13:41:07 yegorich: da8xx-fb, you mean? Jun 29 13:41:07 koen: any idea? Jun 29 13:41:21 mdp: yes da8xx-fb Jun 29 13:43:52 mdp: is there any good arago repo you can recommend? Jun 29 13:44:55 * mdp senses that this could be a trick question Jun 29 13:45:42 yegorich, all I know of is that we have the arago psp am335x tree that's the latest TI kernel code... Jun 29 13:46:13 then there's koen's tree with additional features and tweaking that's actually used for the official builds...and the one I use for beaglebone tinkering Jun 29 13:46:24 stuk_gen there is a package alsa-state Jun 29 13:46:27 no idea about da8xx-fb stuff at all Jun 29 13:46:32 mdp: as I started with arago's kernel (about 4 month ago) I couldn't even boot beaglebone Jun 29 13:46:56 I've never used it so I can't comment Jun 29 13:47:08 mdp: that's why I switched to Koen's tree Jun 29 13:47:56 yegorich implement FB_BLANK yourself? Jun 29 13:48:21 it is a child of the psp arago tree though, keep that in mind Jun 29 13:49:08 woglinde: made, but then I have another issue: (EE) FBDEV(0): internal error: miCreateDefColormap failed in FBDevScreenInit() Jun 29 13:50:16 and? Jun 29 13:50:26 debug it with kernel debugger Jun 29 13:50:43 who else should help you Jun 29 13:52:44 koen often states he is nor programmer Jun 29 13:53:00 I can confirm he's a NOR programmer Jun 29 13:53:31 *g* Jun 29 13:54:05 ouch... I missed product management 101! Jun 29 13:54:35 yegorich: totally unrelated...your e2e i2c/sil question...I can't believe that hasn't been solved in countless android tables with omaps....if you know what I mean Jun 29 13:55:27 mdp: that's a point. The question is if they use this chip sii9022a Jun 29 13:55:41 dwery? Jun 29 13:56:57 yegorich: well, sii has had previous transmitters...I worked briefly with one in 2006...typically they keep the same design..unless they acquired some IP. Jun 29 13:57:03 mdp: it can be that they don't ask for edid info and just try one of the resolutions Jun 29 13:58:01 possibly...my use case back then didn't use EDID at the time Jun 29 13:59:00 but it's a popular design choice as are parts with the omap-i2c block..so... Jun 29 13:59:05 woglinde: it seems there is nothing left to do but debug myself ;-) Nevetherless there are LCD capes for BeagleBone, so there should be a driver working with X. Or is my assumption wrong? Jun 29 13:59:50 mdp. what I find funny is that none got the driver mainlined Jun 29 14:00:11 *ahem* that's because of open android ;) Jun 29 14:00:46 mdp: do you know any of the system I could get the kernel source using this chip? Jun 29 14:01:08 it never hurts to search linaro tilt and omapzoom trees for working code Jun 29 14:01:18 yegorich: nope Jun 29 14:01:55 looks like you had to find the driver in an mxc tree? Jun 29 14:02:52 yegorich hw doesnt meant its fully supported Jun 29 14:03:48 oh and working xorg-driver is another case Jun 29 14:03:59 be lucky if fb works Jun 29 14:04:01 mdp: yes I have found one and talked to the developer, but he couldn't tell me exactly how they solved the problem with EDID Jun 29 14:04:44 woglinde: X fb driver is working very good with another SoC am3517 Jun 29 14:05:01 woglinde: but it has another LCD controller IP Jun 29 14:07:14 I know only little about lcd stuff Jun 29 14:07:34 I as gald to hack my msm7x27 phone a bit Jun 29 14:11:08 yegorich, er, you meant it doesn't implement FB_BLANK_NORMAL? Jun 29 14:11:29 I'm curious as to what exactly you debugged and determined is missing for X.org Jun 29 14:11:39 mdp: yes, compared to am3517 DSS Jun 29 14:13:03 mdp: I'm right at the beginning. Found this thread, though not 100% related, but nevertheless interesting: http://e2e.ti.com/support/dsp/sitara_arm174_microprocessors/f/791/t/189862.aspx Jun 29 14:13:13 it's probably best to wait for koen, he works with those lcd capes regularly and should know any limitations Jun 29 14:13:24 mdp: everyone seems to work with Android ;-) Jun 29 14:13:57 android makes $$$$ Jun 29 14:14:37 yegorich: not on our parts ;) meaning "this TI" ... android is almost irrelevant tbh Jun 29 14:14:40 what is the maximum allowded current for beaglebone? Is 5V, 2A fine for beaglebone or it may harm the HW? Jun 29 14:15:02 angs I read up to 3a Jun 29 14:15:09 ups 3A is okay Jun 29 14:15:15 woglinde: thank you Jun 29 14:15:31 so 2A is okay Jun 29 14:15:42 sure, thanks Jun 29 14:15:42 and good for usb devices Jun 29 14:15:47 yegorich, that's a typical integration issue with non-cache coherent processors and the DMA API...wrt the e2e link Jun 29 14:16:11 mdp: O.K. Jun 29 14:17:27 mdp: I'll dig further Jun 29 14:19:50 yegorich, keep in mind that the driver was only hacked with support for the displays on the reference boards...so if you depart from constraints set by those displays, you are likely to run into coding opportunities. Jun 29 14:20:36 mdp: I understand Jun 29 14:20:44 mdp: :-) Jun 29 14:21:03 yegorich...also consider the history of that driver/IP...it comes from "underpowered" arm926 socs.....so imagine what displays were ever driven by it before... Jun 29 14:21:22 mdp: for our am3517 based system I submitted the patch with our screen :-) Jun 29 14:22:21 mdp: what about a Sitara SoC based on OMAP6? Jun 29 14:22:54 what does based on omap6 mean to you? Jun 29 14:23:12 omap6 is just a testbed for omap7.... Jun 29 14:23:15 mdp: so that I have built-in HDMI support ;-) Jun 29 14:24:17 sitara are broadmarket parts...if there was some broad demand for that...it would be in there Jun 29 14:24:26 OMAP is built for only one thing atm ;) Jun 29 14:24:46 mdp: I would also prefer that TI chooses one consistent set of IP blocks (CAN, EMAC etc.) So that if SoC is OMAP, then it uses the same IP between various SoC's. am3517 is OMAP3 and am3359 is OMAP3, but they have almost nothing in common Jun 29 14:25:35 yegorich, you can't choose EMAC and meet new requirements Jun 29 14:26:34 and the dcan block is totally standard...even fit into an existing driver as an incremental step (see mainline acceptance of that) so I don't see the issue Jun 29 14:27:05 in any case, I have no influence on silicon design so this discussion is pointless ;) Jun 29 14:27:42 mdp: that was just a soul cry of a developer working with two SoC's at a time ;-) Jun 29 14:28:13 yegorich, understand that sometimes people choose something for cost/market reasons...and DSS is expensive. Jun 29 14:30:04 mdp: I see :-) Jun 29 14:31:38 tell me why you didn't choose am37xx? Jun 29 14:31:57 no CAN? Jun 29 14:32:13 CAN? who said CAN? Jun 29 14:32:28 dwery: you CAN, we CAN, everybody CAN! Jun 29 14:32:41 mdp: I was not in a postion to choose the SoC :-( Jun 29 14:32:47 mdp: May I introducte you to Mr. Bosch? :D Jun 29 14:32:56 yegorich, ahh, so we have much in common Jun 29 14:33:38 yegorich, also, I wouldn't exactly call am3517 and am3359 "omap3"...that's way off Jun 29 14:35:02 am335x, in particular, departs from omap architecture in huge ways...having nothing to do with the actual IP blocks chosen. Jun 29 14:36:48 mdp: yup! Jun 29 14:38:42 Mr Bosch is dead? Jun 29 14:39:10 his IP is well alive... Jun 29 14:39:30 the Bosch Bosch? Jun 29 14:39:40 zuendkerze Jun 29 14:39:44 yeaj Jun 29 14:39:47 ups yeah Jun 29 14:39:49 that one is not dead Jun 29 14:39:58 hm Jun 29 14:40:02 let me check wiki Jun 29 14:40:25 there's only one Bosch.. and one CAN :D Jun 29 14:41:04 robert bosch is was born in the 19th century should be dead by now Jun 29 14:41:15 true Jun 29 14:41:22 http://en.wikipedia.org/wiki/Gottlob_Honold Jun 29 14:41:25 "should" be the operative word Jun 29 14:41:28 but his legacy lives on, I have a fridge in his name Jun 29 14:41:50 bosch is big in asia too I think Jun 29 14:41:54 av500, I have some ceremonial spark plugs in his honor as well! Jun 29 14:42:00 saw documentation some weeks agao Jun 29 14:42:09 mdp: how kind of you Jun 29 14:42:40 mdp: I have glow plugs :) Jun 29 14:42:44 av500: made by Siemens Jun 29 14:42:53 jannau: of course Jun 29 14:42:59 I wanted a good one :) Jun 29 14:43:03 av500, I used to have those too ;) Jun 29 14:43:10 but the bosch equivalent was cheaper Jun 29 14:43:48 mdp you can spell diesel in the us? Jun 29 14:43:54 * woglinde runs Jun 29 14:43:59 hehe Jun 29 14:44:19 lpg needs sparkys too Jun 29 14:44:20 alas, I sent my jetta tdi to a new home Jun 29 14:44:31 jetta woot Jun 29 14:44:38 thats really memorial Jun 29 14:44:44 woglinde: only here Jun 29 14:44:49 jetta was big outside DE Jun 29 14:44:51 woglinde: we do still have a kubota diesel tractor...incredible machine Jun 29 14:44:52 av500 I know Jun 29 14:44:57 for countries that require a trunk Jun 29 14:45:14 av500, we need the trunk to lock our rifles in Jun 29 14:45:19 yes Jun 29 14:45:29 hence the term, "trunk gun" Jun 29 14:45:46 mdp: no rifle rack in your pickup cab? Jun 29 14:45:52 no comment Jun 29 14:46:25 you are more likely to see me with a crossbow hanging in the truck at certain times of the year Jun 29 14:46:28 I thought the trunk was for bait and tackle Jun 29 14:46:56 that has something to do with the deers of the other day, right? Jun 29 14:47:10 dwery, yes, there's a connection there I guess Jun 29 14:47:54 gotta love this channel! Jun 29 14:49:23 dwery, I don't think "love" is why were here...it's more about watching a train wreck on a daily basis Jun 29 14:50:17 uhm... Jun 29 14:50:58 train spotting Jun 29 14:51:02 yes Jun 29 16:02:20 anyone know who's manufacturing the google Nexus Q? Jun 29 16:06:25 * gurgalof bought a beaglebone yesterday... Jun 29 16:08:20 bradfa, I was wondering that myself Jun 29 16:08:57 maybe a tear-down will answer it Jun 29 16:09:14 Keebler elfs Jun 29 16:09:37 av500: why are you sending me linkedin invite spam? Jun 29 16:11:09 * bradfa didn't get any linkedin spam, feels left out Jun 29 16:11:46 bradfa: I got 3 from av500, want one? Jun 29 16:11:59 mru, sure! :) Jun 29 16:12:20 currently on hour number 4 with lipo battery powered bone Jun 29 16:12:27 lasting longer than I expected Jun 29 16:12:50 2.4Ah pack Jun 29 16:13:45 I got around 3h using a bb c4 with a sanyo kbc-l2 and an attached 7" mimo Jun 29 16:14:22 smplman, hmm, I always thought of the Keebler elves as being outside the US Jun 29 16:14:51 aholler, I'm running lipo through battery header P6 on bone, no boost, but then no screen either Jun 29 16:15:13 linuxjacques: they are everywhere, EVERYWHERE i tell ya Jun 29 16:15:19 heh Jun 29 16:15:22 hmm, the linkedin web page only lists two of the invites Jun 29 16:15:29 wonder what happens if I accept both Jun 29 16:15:55 mru, you become extra connected! Jun 29 16:17:00 on social media https://sphotos.xx.fbcdn.net/hphotos-ash4/402080_2622304685082_1013522940_n.jpg Jun 29 16:29:06 who ever was asking about the manufacturing of the Nexus Q --> http://tech.slashdot.org/story/12/06/29/1224237/google-on-shores-manufacturing-of-the-nexus-q Jun 29 16:30:35 smplman, I read through most of that and didn't see what company is actually doing the manufacturing Jun 29 16:30:48 yea and the article is pay walled Jun 29 16:30:59 search google for it to get around paywall Jun 29 16:31:04 nice Jun 29 16:32:17 mru: sorry, I am new to this game Jun 29 16:32:42 and I thought it would find out by itself if there were 2 identical ones from different adress books Jun 29 16:32:47 seems not Jun 29 16:33:01 Correct me if I am wrong, but I thought that *Google* was making the Nexus Q? Jun 29 16:33:05 Or was that just the design? Jun 29 16:33:06 yes Jun 29 16:33:10 in their spare time Jun 29 16:33:21 every google employee takes home a few pieces after work Jun 29 16:33:24 and assembles them Jun 29 16:33:32 made in USA! Jun 29 16:33:39 it does appear to be google doing it http://www.cnn.com/2012/06/28/tech/mobile/google-nexus-q-usa/index.html Jun 29 16:33:57 google is paying somebody to do it Jun 29 16:34:09 its like NASA building rocketd Jun 29 16:34:13 rockets* Jun 29 16:34:22 /etc/init.d/rocketd start Jun 29 16:34:46 * bradfa 's rocket didn't start :( Jun 29 16:34:59 something about dependencies Jun 29 16:35:22 smplman, they're outsourcing the manufacturing, google doesn't own (any?) manufacturing facilities Jun 29 16:35:45 maybe they do - secretly Jun 29 16:35:55 are we sure they don't "own" it? Jun 29 16:36:02 smplman, pretty sure Jun 29 16:36:19 it'd be silly for Google to own a manufacturing site Jun 29 16:36:22 i still like its all USA Jun 29 16:36:29 smplman, yes Jun 29 16:36:53 that could solve the fragmentation issue, just let google manufacture their devices Jun 29 16:37:04 smplman, what fragmentation issue? Jun 29 16:37:16 of the android OS Jun 29 16:37:21 they bought motorola, and I assume somebody knows where and how to manufacture such Jun 29 16:37:32 oh I want Jelly Bean, but my phone won't get it Jun 29 16:37:35 smplman, I think there's more to android fragmentation than just manufacturing Jun 29 16:37:37 root it Jun 29 16:37:49 hardware is the main issue in fragmentation Jun 29 16:37:55 aholler, they did, but did moto actually own manufacturing? I dunno Jun 29 16:38:05 smplman, you'd be surprised Jun 29 16:38:39 separate companies, separate hardware, separate code Jun 29 16:38:55 but that's what google wants Jun 29 16:39:04 they don't want what apple does, one company, one hardware, one code Jun 29 16:39:15 manufactured in the us could mean somebody in the us placed the chinese board into the chinese case Jun 29 16:39:43 aholler, the article says the case is mfg in usa along with a few other components and then final assy is done in USA, too Jun 29 16:39:48 so possibly Jun 29 16:39:55 it's OMAP4, so where's those built? Jun 29 16:40:05 i was about to ask the same Jun 29 16:40:16 I'd assume at TSMC or the like Jun 29 16:40:57 although globalfoundaries has a nice new plan in NY state now Jun 29 16:41:04 saratoga I think Jun 29 16:42:55 the way intel does it, the dies are mostly manufactured in the usa then shipped overseas to get placed in the packages and boxed up Jun 29 16:44:16 everyone does such. Jun 29 16:48:10 av500: codec wrangler, eh? :) Jun 29 16:49:30 mranostay, chief codec wrangler! Jun 29 16:51:25 hello there, I'm wondering if anybody has been used the beaglebone with real time patches (xenomai, rtai, preempt-rt) Jun 29 16:52:40 I am sure someone has. Jun 29 16:52:59 i'm sure someone has been bored and insane enough to try Jun 29 16:55:48 no rtc, no real time Jun 29 16:56:22 well pop a RTC clock on :) Jun 29 16:56:44 Are they not using the same management IC on the bone as on the xM? Jun 29 16:56:48 That has an RTC in it. Jun 29 16:57:06 No battery populated, but easy enough to add that and enable the module. Jun 29 16:57:25 no Jun 29 16:57:50 not rtc, rt... deterministic output Jun 29 16:58:25 antonio__: it was a joke Jun 29 16:58:26 haven't found anything or anyone working on it yet Jun 29 16:59:15 I don't want to start from scratch :S Jun 29 16:59:53 antonio__: Should not really be any different on the bone as on any other ARM embedded board. Jun 29 17:00:10 many of the rt-stuff already went into the mainline kernel Jun 29 17:00:32 s/many/much/ Jun 29 17:01:29 so first try if you really need rt-patches. Jun 29 17:05:03 yes, the patch from Igno is quite good, and yes, I really need a rt-system. Right now, with the BB I can control a motor about 2 khz and my target is 40 khz Jun 29 17:06:06 I want to know if the BB can make it Jun 29 17:06:41 a lot of work to do Jun 29 17:06:43 I thought that was what this PRUSS thing in the bbone was for Jun 29 17:08:42 there's always the option to not use linux. I wouldn't control motors directly by linux Jun 29 17:09:19 Yeah, if you nreally want RT, use an RTOS. (Actually, if you really want RT, use a PLC, but that is neither here nor there). Jun 29 17:14:21 but that all depends what your box should do besides controlling the motors. if you don't use usb or network, there isn't much which could disturb your motor control ;) Jun 29 17:15:42 Also, I have to wonder at your phrasing. Are you trying to change the state of a motor 40 thousand times a second, or are you trying to use 40kHz as a PWM base? Jun 29 17:17:30 sounds like bitbanging pwc Jun 29 17:17:33 I need to sent the pulse (square) signal at 40 Khz, not PWM Jun 29 17:18:13 s/pwc/pwm/ Jun 29 17:18:51 antonio__: filter it? Jun 29 17:19:34 you could bit bang GPIO lines Jun 29 17:19:57 antonio__: What kind of control method are you using with a square wave that is *not* PWM? Jun 29 17:20:00 PPM? Jun 29 17:20:07 Stepping? Jun 29 17:20:15 yes Jun 29 17:20:16 the bone has hw-pwm, so just use a 50% duty cycle Jun 29 17:20:19 stepper motors Jun 29 17:20:26 Ah. Jun 29 17:20:30 sorry Jun 29 17:20:51 So, you are trying to run software-controlled 40kHz square waves. What is blocking you from doing this? Jun 29 17:21:11 Also, as aholler says, use the onboard hardware PWM generation. Jun 29 17:22:33 I use the GPIO directly and I can only get ~2.6 Khz for the pulse sign Jun 29 17:24:49 I like the idea of use PWM, it's now on the to-do list Jun 29 17:27:20 I did try with the arduino first, it can handle as far as 66 Khz but it can not make the necessary maths for the trajectory Jun 29 17:29:42 antonio__: So use both. Jun 29 17:30:08 antonio__: A low-level micro (or a stepper controller!) for the stepper, and the bone for high-level math and processing., Jun 29 17:38:04 I guest that solution will be not well see. Need to check that PRUSS thing, didn't know about it Jun 29 17:58:10 hm I should linkedin to av500 too Jun 29 17:58:13 args Jun 29 17:59:44 antonio__: Have you looked at my solution at https://github.com/modmaker/BeBoPr ? Jun 29 18:03:16 ncbas your makefile is not in beste state Jun 29 18:05:13 oh, what's wrong? Jun 29 18:06:11 hm does overriding CROSS_COMPILE works? Jun 29 18:07:04 I've had no complaints thus far, but I welcome any improvements. Jun 29 18:08:09 I think it should be := Jun 29 18:08:31 ncbas: that's nice Jun 29 18:10:37 antonio__: What is it that you want to control? Jun 29 18:11:49 a lens polisher, two arms, 9 steppers in total Jun 29 18:12:08 antonio telescope? Jun 29 18:12:31 no, eye glasses Jun 29 18:12:58 telescope lenses are more difficult Jun 29 18:13:45 esp in place of eye glasses Jun 29 18:13:50 antonio__: To polish as in clean, or polish as in take a rouch-cut lens and work it smooth? Jun 29 18:14:00 *rough Jun 29 18:14:45 The PRUSS has only a limited number of outputs, so 9 steppers isn't obvious. But you could multiplex as the PRUSS is very fast. Jun 29 18:15:44 to make it smooth :) Jun 29 18:15:45 Or, use an external stepper controller. Jun 29 18:15:58 antonio__: Ah, so you need high precision as well as speed. Jun 29 18:16:31 ncbas: PRUSS seems very interesting, I just grab a cup of coffe to check it out Jun 29 18:18:01 agmlego: not really, we use some rubbers and relatively low speed, it have not good precision Jun 29 18:18:57 agmlego: but that's the key of the machine Jun 29 18:20:03 antonio__: YOu are polishing down lenses with rubber? Generally, that is done with like a mild abrasive and cloth pads. Jun 29 18:21:18 woglinde: The Makefile seems to work with both '=' and ':=' . Can you explain the difference? Jun 29 18:25:58 agmlego: yes, we do that, but the tool is based in soft neopreno for it can change the its form enough without deform the lenses curve. I have seen it from many manufacturers Jun 29 18:26:15 Ah, OK. Jun 29 18:26:37 antonio__: So, why are you using the bone insytead of the controller that came with the machine? Jun 29 18:28:22 ncbas: how goes the cape? Jun 29 18:28:45 agmlego: eventually price cut, new forms of control.. in a few word for testing purposes Jun 29 18:29:38 antonio__: Makes sense. Jun 29 18:31:07 agmlego: that project have like 10 years old and like 8 of them dormant, everything is obsolete right now Jun 29 18:32:21 I can imagine. Jun 29 18:34:19 ncbas: the BeBoPr is your project? Jun 29 19:11:08 re Jun 29 19:17:10 antonio__: yes, BeBoPr is mine, both hardware and software. Jun 29 19:18:23 koen: some capes are being assembled right now. How's the 'verhuizing' going ; ) Jun 29 19:20:39 is guessing that is dutch for moving Jun 29 19:23:38 I installed ubuntu-server image then installed ubuntu-desktop image on beagleboard-xm rev C. I can not get any video output, the board is connected to the monitor via hdmi to dvi-d. do I need to configure the video output or do I suppose to see the video output as I reboot the board? Jun 29 19:29:12 Beagleboard XM dv-D connector does not support the full HDMI interface and is used to Jun 29 19:29:12 provide the DVI-D interface portion only. Jun 29 19:29:59 works with most computer monitors with DVI-D input and some tv's that have DVI-D Jun 29 19:31:06 reference page 25 of BBSxMSRM docs Jun 29 19:31:21 BBxMSRM Jun 29 19:50:22 djlewis: thank you. I am using the same cable in the figure 57, but I do not get any video output as I power the board. Jun 29 19:51:21 which brand? Jun 29 19:51:46 or hmm, with BBxm you need to pass the resolution manually anyway Jun 29 19:52:11 do you mean the monitor brand? Jun 29 19:52:17 it is dell Jun 29 19:52:18 yes Jun 29 19:52:37 I think it is square 17 inch screen Jun 29 19:53:05 you might need to set the dvi output resolution manually Jun 29 19:53:13 how can I do it? Jun 29 19:53:32 is there any document that I can read or use? Jun 29 19:53:39 haven't done that in a while. someone else probably has better pointers Jun 29 19:54:54 http://www.sakoman.com/OMAP/changing-the-display-resolution.html Jun 29 19:55:04 see! :D Jun 29 19:55:16 forget the nand part Jun 29 19:55:23 try printenv Jun 29 19:55:25 before Jun 29 19:55:30 to see if something changed Jun 29 19:55:34 with dvimode Jun 29 19:55:40 variable Jun 29 19:55:46 dm8tbr ay Jun 29 19:55:58 dm8tbr, woglinde: thank you, I will try it Jun 29 19:56:31 hm 64 ram computer would cost me 790 euros Jun 29 19:56:41 +gig Jun 29 19:57:03 RAM _is_ cheap :-O Jun 29 19:57:04 I could use a 50 gig ramdisk for oe Jun 29 19:57:11 eehhhh Jun 29 19:57:15 that would be cool Jun 29 19:57:34 or 60 gig Jun 29 19:57:36 yes, ramdisk for building stuff is quite popular recently Jun 29 19:57:36 speed it up :) Jun 29 19:58:02 okay be 800-900 power and case Jun 29 19:58:08 +would Jun 29 19:58:28 high performance OBS installations also run the 'worker' instances completely in RAM Jun 29 20:00:27 obs? Jun 29 20:00:34 guess why I maxed out the ram in my machine Jun 29 20:00:39 woglinde: opensuse build something Jun 29 20:00:49 ieehks Jun 29 20:01:01 opensuse searches it way now Jun 29 20:01:51 nowadays only 'open' Jun 29 20:01:56 open build server Jun 29 20:02:13 it's a neat concept, but very resource intensive Jun 29 20:03:08 we had pbuilder before Jun 29 20:10:38 its a cool 106F today :) Jun 29 20:11:13 +g* Jun 29 20:14:54 woglinde: the monitor shows "self test check" screen after env setting, however the I can not get any another output when I power the board. Is there anything else that I can do? this is the output of the boot procedure http://pastebin.com/9Tjridi4 Jun 29 20:17:13 angs gets the monitor orange when u-boot is loaded? Jun 29 20:17:25 otherwise something is wrong with the cables Jun 29 20:17:37 or you need some edid stuff Jun 29 20:18:02 whats the resolution of the monitor? Jun 29 20:18:02 I have this screen all the time, http://support.dell.com/support/edocs/monitors/E228WFP/En/ug/graphics/stfcD.gif Jun 29 20:18:19 it is a square 17 inch monitor Jun 29 20:18:28 how did you connect it now? Jun 29 20:18:36 via hdmi-dvi? Jun 29 20:19:37 woglinde: yes hdmi-dvi Jun 29 20:20:52 beagleboardXM HDMI connector to monitor DVI-D ? Jun 29 20:21:01 yes Jun 29 20:21:26 I will try it on an another board Jun 29 20:21:27 is this a straight cable or something with adapters inline? Jun 29 20:22:02 straight cable Jun 29 20:22:21 I have an extra cable, I will try it with a new cable Jun 29 20:23:01 hm some tv's needed some tweaking inside kernel Jun 29 20:23:14 but I dont remember monitors Jun 29 20:23:25 besides some 7inch Jun 29 20:24:37 sh.t the cable was broken :S I see the output now Jun 29 20:24:50 ah I spent my almost whole day on it :( Jun 29 20:24:53 cool Jun 29 20:24:59 that it is working Jun 29 20:25:05 thank you very much for helping me Jun 29 20:25:20 hm yes Jun 29 20:25:28 first you should consider the hw broken Jun 29 20:25:35 i would like to buy a vowel Jun 29 20:25:40 I will go and burn the cable :) Jun 29 20:25:48 and sell the copper Jun 29 20:28:25 can't do that anymore Jun 29 20:31:01 mranostay, and so goes away a reason to enjoy the pretty colors of burning insulation Jun 29 20:32:42 and killing the ozone layer? :) Jun 29 20:37:01 i could lay it out on the asphalt parking lot and melt it off. Jun 29 20:37:06 ka6sox: is there any effort to put webOS on Angstrom? Jun 29 20:37:34 seems to me it could really take Angstrom to another level, especially with Cloud9 IDE as a development engine. Jun 29 20:37:58 ka6sox: I just learned you are involved in the webOS community edition release. Jun 29 20:39:35 jkrinder *g* Jun 29 20:40:47 woglinde: are you involved as well? Jun 29 21:00:36 the silent treatment Jun 29 21:01:08 No, just IRC. Jun 29 21:01:43 * djlewis was just kidding about jkridner's last question Jun 29 21:02:07 jkridner no, bu I posted the article from heise.de here Jun 29 21:02:25 where ka6sox was mentioned Jun 29 21:04:25 ncbas: it's really nice project, do you have more info about pruss? Jun 29 21:06:14 hm intressting the thunderstorm has shutdown our tv cable net Jun 29 21:40:48 * djlewis wants some thunderstorms. its getting brittle here Jun 30 00:01:14 Can somebody give me a hand with UART, I have hooked a gs407 gps with u-blox and I'm trying to get info from it Jun 30 00:05:23 er how is that a beagle thing? Jun 30 00:07:26 I am using Beaglebone Jun 30 00:07:46 want to hook it up and i think i have the cables right Jun 30 00:08:01 but still don't know where to look for the coordinates Jun 30 00:08:13 break out the datasheet? Jun 30 00:40:10 is something like xev showing me raw inputs or what the drivers are translating it to? Jun 30 00:42:50 ???? Jun 30 00:42:55 translating? Jun 30 00:45:00 * mranostay digs aroudn his desk for his scotch **** ENDING LOGGING AT Sat Jun 30 02:59:58 2012