[01:46] <bjsnider> nvidia is now saying the 195 unix blobs are affected by the gpu fan issue in the 196 windows drivers. unfortunately, this is the nvidia-current driver in lucid
[01:48] <BUGabundo> I noticed
[01:49] <BUGabundo> my fan is always at max
[01:49] <BUGabundo> the clock is almost always at 400MHz too
[01:50] <bjsnider> i don't think it would affect integrated hardware since there's no gpu fan there
[01:53] <Sarvatt> he doesnt have integrated, and that explains why the battery life is sucking on that machine my wifes using
[01:55] <bjsnider> who doesn't have integrated?
[01:55] <BUGabundo> I think I do have
[01:55] <Sarvatt> BUGabundo had the same 8400M GS I did last I checked?
[01:55] <BUGabundo> its a 8400 GM
[01:55] <bjsnider> it's a laptop
[01:55] <BUGabundo> or better 8400m G
[01:55] <BUGabundo> right
[01:55] <BUGabundo> its _just_ a chip there
[01:55] <bjsnider> unless laptops are using pci-express cards now he's got integrated
[01:56] <Sarvatt> ah yeah thats got lower clocks than GS but the same thing basically
[01:56] <BUGabundo> I had it clean, and applied new temp mass a few weeks ago
[01:56] <Sarvatt> yep they are wired up over PCI-E internally on the board
[01:56] <BUGabundo> but I do notice much more fan, latelly
[01:57] <BUGabundo> then again we keep getting new gnome-power-managers in lucid
[01:57] <BUGabundo> and new kernels
[01:57] <BUGabundo> so its hard to pin point
[01:57] <Sarvatt> ya cant adjust the profile BUGabundo?
[01:57] <Sarvatt> theres a module option you can use to force it but i dont have it off the top of my head
[01:57] <BUGabundo> Sarvatt: nvidia-settings gives me two choices
[01:58] <BUGabundo> adaptive and performance
[01:58] <Sarvatt> oh the RegistryDwords option doesnt work anymore since they added it to nvidia-settings :(
[02:03] <Sarvatt> yay mesa_7.9.0~git20100305.67277a6d-0ubuntu0sarvatt_source.changes -- up to 7.9 already
[02:04] <Sarvatt> ugh there's already a bug against apw's -16 kernel where r100 KMS is crashing
[02:04] <libv> bjsnider: about the openchrome/gnome-shell guy: ohloh doesn't lie: https://www.ohloh.net/p/gnome-shell/contributors/116060753925724
[02:17] <bjsnider> libv, you mean jon nettlet?
[02:26] <bjsnider> he does a lot of testing with owen taylor. jon doesn't always do the commit after a problem is fixed
[02:28] <libv> gnome uses git, author and committer are separate, ohloh checks author
[02:28] <bjsnider> jon's not here to defend himself
[02:28] <bjsnider> what would he tell me about you if i were to ask him?
[02:28] <libv> bjsnider: ask him when he is around, nickname is jnettlet
[02:29] <bjsnider> yes, i know
[02:29] <bjsnider> he's not in the gnome-shell channel right now
[02:31] <libv> bjsnider: but try to match talk to code, it tells a very clear story here, but does require you to look at actual code
[02:32] <bjsnider> well, any contribution is better than none at all, which would be the case with me
[02:33] <libv> sure, but the point i was trying to make about the openchrome people is that they're mostly into talking
[02:33] <libv> (the fact that i am saying instead of coding is slightly ironic, yes)
[02:33] <bjsnider> jon and who else?
[02:34] <libv> xavier bachelot and ivor hewitt
[02:34] <libv> the guy who doesn't talk in the openchrome deal, he's the coder.
[02:34] <libv> or at least the bugfixer.
[02:34] <libv> bartosz (gang65)
[02:35] <bjsnider> the original reason i mentioned this is that jon told me he had to go off and fix that driver so it would work with gnome-shell
[02:35] <bjsnider> or work on it anyway
[02:35] <bjsnider> maybe they're caucusing on how to fix it
[02:36] <bjsnider> but i woudln't actually use the hardware if someone had a .44 magnum to my head
[02:37] <libv> bjsnider: but did he fix the driver so it would work with gnome-shell?
[02:37] <bjsnider> hahahaa
[02:37] <bjsnider> you've got me there
[02:37] <bjsnider> this was two weeks ago
[02:38] <bjsnider> in that time he has not fixed it
[02:38] <libv> bjsnider: via might be an almost dead company, but it is still the 4th x86 graphics hw maker out there
[02:38] <libv> with a market percentage smaller than the error margin, but still :)
[02:39] <bjsnider> yeah, it's the most popular, right after all of the others
[02:40] <libv> VIA was doing something amazing in 2003, when they, almost as an error i think, handed alan cox the code to their xfree86 driver
[02:41] <bjsnider> is this 2003?
[02:41] <libv> they claimed that they opensourced their graphics driver, in marketing, 3 times after that, but at that point, marketing totally overlooked that
[02:41] <bjsnider> libv, why are you working on this driver?
[02:41] <libv> bjsnider: because i bought this hw in 2003
[02:42] <bjsnider> on a board or in a laptop?
[02:42] <libv> board
[02:42] <bjsnider> i know how you can save yourself a lot of work
[02:42] <libv> heh.
[02:42] <bjsnider> you know those massive jet engines on a 747?
[02:42] <bjsnider> go near one and ask the pilot to crank it up to full throttle
[02:43] <bjsnider> toss the board into the engine
[02:43] <bjsnider> problem solved
[02:43] <libv> *shrug*
[02:44] <libv> so what do i do with those 24 other boards and laptops/netbooks?
[02:44] <bjsnider> the pilot will wait until they have all been vaporized
[02:45] <bjsnider> then buy something made by the nvidia corporation
[02:48] <bjsnider> it's interesting that nobody's bought via, if they're the 4th largest gpu manufacturer. they must be an almost owrthless property, with only obligations to their customers left at this point
[02:52] <libv> maybe because it is largely owned and run by one and the same family?
[03:10] <bjsnider> mac folks cannot play 24/96 flac audio files without vlc
[03:10] <bjsnider> good thing apple doesn't leech off the open source community
[07:02] <tjaalton> bryceh: bh.org down again?
[10:04] <Duke`> Sarvatt, will you package mesa 7.9-devel for Karmic, or stay with 7.8?
[16:39] <Sarvatt> ok i managed to get plymouth working in the screwed up manner that sends sigquit to X when you press enter on one of my machines to troubleshoot it some more
[16:41] <BUGabundo> woot
[16:41] <Sarvatt> and the isig flag IS set on tty7 *sometimes* which is for sure whats sending sigquit when you press enter, I can see it with sudo stty -F /dev/tty7
[16:42] <Sarvatt> and sudo stty -F /dev/tty7 -isig from another VT fixes it
[16:43] <Sarvatt> i think this is more prevalent when you aren't using the drm plymouth renderer which is why its hitting nvidia blob drivers more
[16:53] <Sarvatt> where the heck is the ubuntu plymouth bzr branch at? I just see lp:plymouth
[17:30] <Sarvatt> Duke`: probably wont do 7.9 for karmic honestly unless someone else decides to do it, at least I know 7.8 will work under karmic for the near future and thats already a huge jump from 7.6.0. why dont you upgrade to lucid already if you like the crack that much? it's not far from release :)
[17:32] <Sarvatt> things are just so different in lucid vs karmic and i'm already spending most of my free time keeping the ppa up to date
[17:32] <tjaalton> yeah drop it already
[17:41] <Sarvatt> Duke`: if you want to help keep karmic up to date in edgers it'd be *much* appreciated
[17:42] <Sarvatt> i live in development releases from the day they open, though i plan on keeping lucid going for a long time in there
[17:44] <Sarvatt> i think readding the tty-device-added requirements to the plymouth upstart job might help with this sigquit problem since it screws with the flags when the tty isnt ready at the start, have to test that out when i get some time
[18:49] <Duke`> well I'm not sure I'll have the time for that, unless a script can do it all in one or two commands (build & upload or something like this)
=== radoe_ is now known as radoe
[22:51] <chrisccoulson> does anyone know what would cause XF86VidModeSetGamma to throw a BadValue error?
[22:51] <chrisccoulson> (there's not much documentation for that)
[22:52] <chrisccoulson> i'm trying to debug a screensaver crash (http://launchpadlibrarian.net/40255454/xtrace-mdeslaur.log) and can't see anything obvious that the screensaver is doing wrong