**** BEGIN LOGGING AT Thu Jun 14 02:59:58 2012
Jun 14 03:48:05 <codinho> Hi guys
Jun 14 03:48:33 <codinho> is gstreamer works with ducati in 12.04 ?
Jun 14 03:50:23 <twb> Your car is running ubuntu?
Jun 14 03:51:18 <twb> Er, motorcycle
Jun 14 03:53:24 <codinho> twb I mean omap4 ducati
Jun 14 03:53:31 <twb> Never heard of it
Jun 14 03:53:45 <twb> But AFAIK Ubuntu works well on omap4 generally
Jun 14 03:54:32 <codinho> Ducati is hardware accelerator for image processing and video/audio decode/encode
Jun 14 03:55:03 <twb> Oh.  Well I don't know, sorry :-)
Jun 14 03:55:09 <heathkid> anyone here runnning ubuntu on an ARM with a USB wifi adapter that works 100%?
Jun 14 03:55:34 <twb> heathkid: define "100%
Jun 14 03:55:47 <heathkid> as in it actually works and is reliable
Jun 14 03:56:04 <heathkid> doesn't work for a few hours or a day and then stop working...
Jun 14 06:05:33 <codinho> is it possible to install ubuntu-omap-extras-multimedia from trunk ppa on 12.04?
Jun 14 14:46:45 <ppisati> if anyone has an idle panda board and an empy sd card, could you try to reproduce lp1013204? thanks
Jun 14 14:46:51 <ppisati> bug 1013204
Jun 14 14:46:52 <ubot2> Launchpad bug 1013204 in linux-ti-omap4 "Serial losing characters" [Undecided,New] https://launchpad.net/bugs/1013204
Jun 14 14:51:31 <mythos> ppisati, erm, so you loose chars over a serial-console
Jun 14 14:51:55 <mythos> which char is missing?
Jun 14 14:52:14 <mythos> i mean, in your bugreport
Jun 14 14:54:15 <mythos> and you know, that with the default settings (http://www.omappedia.com/wiki/Serial_Port_Settings) it's not unusual, that bits are not transfered right?
Jun 14 14:57:57 <ppisati> any kind of characters
Jun 14 14:58:12 <ppisati> just type and you'll see that some won't get through
Jun 14 14:58:22 <ppisati> some times it works, some times it won't
Jun 14 14:58:28 <ppisati> on armel
Jun 14 14:58:32 <ppisati> on armhf everyhing is ok
Jun 14 14:59:46 <mythos> maybe they use another baudrate?
Jun 14 14:59:59 <ppisati> identical
Jun 14 15:00:05 <ppisati> 115200n8
Jun 14 15:00:34 <mythos> so... which ubuntu should i use?
Jun 14 15:00:51 <mythos> 11.10?
Jun 14 15:01:45 <mythos> oh, i see 12.04 armel
Jun 14 15:02:01 <mythos> i'm going to try it later
Jun 14 15:02:03 <ppisati> the simplest way to reproduce it
Jun 14 15:02:10 <ppisati> get the O omap4 preinstall
Jun 14 15:02:30 <ppisati> once you finished with the installation, let it reboot till the serial console
Jun 14 15:02:44 <ppisati> then paste a long line of text and see what happens
Jun 14 15:03:12 <mythos> i'm going to try it in an hour or so
Jun 14 15:08:04 <mythos> but first i have to go home ;)
Jun 14 15:10:16 <ppisati> mythos: i updated the bug with a comment about reproducing the problem
Jun 14 15:12:24 <ogra_> and make sure the insulation of your cable has no cracks, might be that the chars are dripping out through it :P
Jun 14 15:12:40 <ogra_> what a weird bug :)
Jun 14 15:12:49 <ppisati> yes it is
Jun 14 15:12:54 <ppisati> i swapped hw
Jun 14 15:12:56 <ppisati> cables
Jun 14 15:12:59 <ppisati> sd cards
Jun 14 15:13:05 <ppisati> distributions
Jun 14 15:13:11 <ppisati> kernels
Jun 14 15:13:15 <ogra_> serial terminal app ?
Jun 14 15:13:20 <ppisati> now i want someone else to reproduce it
Jun 14 15:13:23 <ppisati> nope
Jun 14 15:13:26 <ppisati> i'm using cu
Jun 14 15:13:28 <ppisati> well
Jun 14 15:13:36 <ppisati> it works with my amrhf installation
Jun 14 15:13:41 <ogra_> well, did you try screen or minicom ?
Jun 14 15:13:44 <ppisati> so should work with armel
Jun 14 15:13:46 <ogra_> oh, ok
Jun 14 15:13:47 <ppisati> nope
Jun 14 15:13:56 <ogra_> well, who cares ... its armel :P
Jun 14 15:14:02 <ppisati> :)
Jun 14 15:15:27 <ppisati> ogra_: if you can devote and sd card to the cause, i would love it
Jun 14 15:15:40 <ppisati> i want someone else to say "yes, i've it too"
Jun 14 18:37:41 <ppisati> my bad
Jun 14 18:38:02 <ppisati> the sd that i used was actually a preinstalled system were i used to mess on
Jun 14 18:38:05 <ppisati> :P
Jun 14 18:38:20 <ppisati> mythos: ^
Jun 14 18:38:23 <ppisati> ogra_: ^
Jun 14 18:38:56 <mythos> ppisati, so i don't have to? :O
Jun 14 18:39:01 <ppisati> nope
Jun 14 18:39:04 <ppisati> forget about it
Jun 14 18:39:08 <mythos> hurray \o/
Jun 14 18:39:21 * ppisati feels really stupid now
Jun 15 00:11:15 <GrueMaster> http://www.phoronix.com/scan.php?page=article&item=phoronix_effimass_cluster
Jun 15 00:42:22 <lilstevie> GrueMaster: wow, 6 pandaboard ES clustered was heaps more efficient than the atom, that is unexpected
Jun 15 00:44:42 <infinity> lilstevie: It's an Atom 330, that's not a useful or fair comparison.
Jun 15 00:45:34 <lilstevie> infinity: yeah just realized, old as atom
Jun 15 00:45:44 <infinity> lilstevie: Oldest Atom core there is, yeah.
Jun 15 00:46:27 <lilstevie> the z530 atom was also less efficient too though
Jun 15 00:47:15 <infinity> lilstevie: z530 is the same vintage.
Jun 15 00:47:29 <infinity> lilstevie: Both are 45mn non-SoC cores from 2008.
Jun 15 00:47:42 <infinity> lilstevie: A comparison with something from this year would be much more interesting.
Jun 15 00:47:43 <lilstevie> ah
Jun 15 00:47:51 <lilstevie> I didn't know where the z530 fit
Jun 15 00:48:23 <lilstevie> the medfield atom would be interesting
Jun 15 00:48:35 <infinity> Indeed.
Jun 15 00:50:10 <lilstevie> also would be interesting to see performance of something like the S4 in the same situation
Jun 15 00:51:14 <infinity> Well, it would also be interesting to make that an OpenCL cluster, and let the AMD Fusion actually use its GPU.
Jun 15 00:51:25 <infinity> Perf/W could skew a bit there.
Jun 15 00:51:50 <infinity> Cause it's powered up, despite it not being used.
Jun 15 00:52:03 <lilstevie> yeah
Jun 15 00:52:16 <infinity> (Though this is true of the PVR on the OMAP4 as well, there's no way it can come close to delivering the performance of the on-die Radeon on the Fusion)
Jun 15 00:52:28 <lilstevie> agreed
Jun 15 00:52:45 <lilstevie> but then use the tegra powered openCL rig and see how that one goes :p
Jun 15 00:53:12 <infinity> Has anyone mangled OpenCL for the Tegra yet?
Jun 15 00:53:13 <lilstevie> that isn't an on-die gpu though
Jun 15 00:54:31 <infinity> "Tegra does not currently support CUDA or OpenCL. NVIDIA believes in and invests heavily in GPGPU computing so this is an area we're investigating but do not have any announcements for Tegra at this time."
Jun 15 00:54:46 <lilstevie> yeah
Jun 15 00:54:52 <lilstevie> not likely to happen
Jun 15 00:54:53 <lilstevie> :p
Jun 15 00:55:06 <infinity> Oh, right, I remember researching this before.  There was some internal corporate fear that supporting CUDA and OpenCL on the Tegra parts would cut into other business units.
Jun 15 00:55:30 <infinity> Cause one could ditch the x86/GeForce clusters and build Tegra clusters instead.
Jun 15 00:55:45 <infinity> To which, I say, "let 'em", cause they'd still hit two different HPC markets.
Jun 15 00:56:01 <infinity> Given the workload-per-core characteristics would be dissimilar.
Jun 15 00:56:09 <infinity> But, arguing with management over money issues is never easy.
Jun 15 00:57:00 <lilstevie> heh
Jun 15 00:58:40 <infinity> That actually puts AMD in a unique situation.  I wonder if they realise this and have tried selling into that market more heavily.
Jun 15 00:59:09 <infinity> Cause, one would think, bang-for-buck, Fusion would beat Atom, i7, and any ARM part for an OpenCL solution.
Jun 15 00:59:24 <lilstevie> yeah
Jun 15 01:00:48 <lilstevie> imagine how killer they could be with tegra if they put their gpu tech to work though, full OpenGL rather than just the embedded standard
Jun 15 01:01:21 <infinity> Yeah, it could be shiny.
Jun 15 01:01:55 <lilstevie> I can't see why the ULP GeForce would be ruined with GL
Jun 15 01:01:56 <infinity> I'd also love to see an AMD ARM part, but I suspect they don't have the spare cash flying around internally to hedge their bets on forking their CPU division.
Jun 15 01:02:04 <lilstevie> yeah
Jun 15 01:02:52 <infinity> And no, there's no reason whatsoever that the GeForce on the Tegra couldn't support full OpenGL, it was a conscious decision not to.
Jun 15 01:03:17 <infinity> Partially because of the CUDA/shader/CL concerns and, of course, because GLES is the mobile standard.
Jun 15 01:03:53 <lilstevie> yeah, well GLES being the mobile standard isn't so much of a problem
Jun 15 01:04:01 <lilstevie> you could handle compat in driver really
Jun 15 01:04:06 <infinity> But, from what I understand, it's just a "normal" GeForce under the hood.
Jun 15 01:04:15 <infinity> Like, electrically.
Jun 15 01:04:17 <lilstevie> I suspect you may be right with CUDA/shader/CL
Jun 15 01:04:21 <lilstevie> yeah
Jun 15 01:04:27 <lilstevie> thats what I have heard too
Jun 15 01:05:44 <lilstevie> it will be better if the open driver ever makes it too
Jun 15 01:05:50 <lilstevie> the nvidia ones are pretty bad
Jun 15 01:14:28 <infinity> lilstevie: Pretty bad is an understatement.
Jun 15 01:18:34 <lilstevie> infinity: well I have a tegra3 device and it is pretty bad, but the earlier drivers were way worse on the trimslice
Jun 15 01:18:43 <lilstevie> :p
Jun 15 01:28:44 <prpplague> ho ho hum
**** ENDING LOGGING AT Fri Jun 15 02:59:59 2012