[00:17] Hi, is anybody else having trouble updating python-gobject? [00:19] Yes, everybody. [00:20] There's some (presumably temporary) archive skew; nothing to see here. [00:27] Ah, cool. [00:27] RAOF: Thanks [00:40] Is there a package available that has libgtk2.0-dev and all its dependencies in it? [00:42] theunixgeek: Yes. libgtk2.0-dev [00:43] I'm obviously confused as to what you actually want. [00:43] RAOF: but then it has to download all the dependencies' packages with it [00:43] I'm looking for an all-in-one [00:43] because I don't have internet on my Linux computer :( [00:43] I want to wget the package from my mac [00:44] Right. So, that's not the way we work, so there isn't such a package. [00:44] On the other hand, there are tools to create a script which will download the package and all its dependencies. [00:45] RAOF: I know but some of the downloaded packages didn't install correctly :/ [00:45] namely libc6 [00:45] I'm having trouble with libc6 too [00:45] You were using apt-zip? [00:45] RAOF: synaptic [00:45] Oh, that includes the apropriate functionality? Awesome. [00:46] Anyway, so we've moved on to your _acutal_ problem then; libc6 hasn't installed correctly. [00:46] Got the dpkg output for trying to install libc6? [00:47] RAOF: yes - that it installs correctly :) [00:47] Ok. So, how about the _failed_ dpkg log? :) [00:48] RAOF: nothing :) [00:49] Ah. It now works, then. Great. [00:53] I have to go now, good night [01:30] the python-gobject problem has been fixed [01:37] in 8.10 final, will audio finally not fuckup? [01:38] so i can listen rhythmbox and watch youtube? [01:39] look into libflashsupport, and bug adobe about their bugs? :) [01:39] isnt this a problem of pulseaudio? [01:40] its not only adobe, i cant use rhythmbox and eSpeak or Festival together at same time either [01:40] flash works great when you dont use rhythmbox, and rhythmbox works great when you dont use flash [01:41] its not flash problem, its integration problem [01:41] both apps work fine alone, just not together [01:43] looks like your soundcard doesn't support hardware mixing, so you'll always have some amount of trouble [01:44] works great in windows [01:44] now the issue is if pulseaudio is active, all apps using pulseaudio will run fine but others won't be able to use the soundcard [01:44] you can play flash and rhythmbox at same time? [01:44] if pulseaudio is on standby a single other app works [01:44] windows drivers always do software mixing [01:44] then linux should do software mixing too [01:44] you can also set up alsa to do software mixing of course [01:46] doggymenz: pulseaudio _does_ do software mixing :) [01:46] yes, for those apps that support it which is probably most now. there's also a wrapper for oss available, padsp [01:47] obviously not, else rhythmbox would be able to work together with flash, eSpeak, Festival, etc, it dont [01:47] There are some integration problems; it'd be nice to turn on ALSA->pulse by default, too. [01:48] flash is just a piece of garbage software. they've quite literally "outsourced" the problem by introducing support for this liblfashsupport thingie [01:48] yeah, flash on linux works garbage [01:48] Flash 10 is much better about not abusing the ALSA API, too. [01:48] i don't know those other applications [01:48] poor performance too [01:48] i wonder why ALSA dont do software mixing [01:48] i just booted the intrepid disk [01:48] goes to a black screen [01:49] doggymenz: because open source devs are lazy [01:49] it _can_ do, but it nomally doesn't since hardware mixing is better [01:49] when i try to shutdown my computer it shows the ubuntu progress screen for shutting down ;) [01:49] and why is software mixing needed? doesnt all audio since 10 years do hardware mixing? [01:49] doggymenz: It does. But pulseaudio doesn't use it, because it introduces unpredictable latency. [01:49] yours doesn't? [01:49] all the older versions for the last few years have worked fine on my hardware [01:49] doggymenz: Audio hardware is moving to the CPU, now that they're infinitely fast. [01:49] doggymenz: alot of hardware doesnt do hardware mxing [01:49] all the AC97 stuff doesnt [01:49] and that is probably about the most common audio hardware now [01:49] yeah, and AC'97 has only been around for like, 10 years [01:50] The reason this problem continues to exist even after all this time is laziness, pure and simple [01:50] also with software mixing you can set volume levels and outputs stream by stream [01:50] you cant do that with hardware mixing [01:50] thats already, with pulseaudio [01:51] later who knows what kind of effects/filters youll be able to do stream by stream [01:51] yeah, effects/filters would be awesome [01:51] Well, there's already positional effects, but that's more for sound notifications than anything else. [01:51] like if i use a microphone, it make me sound like a girl lool [01:51] pulseaudio is quite nice, though i usually use it just to redirect audio to another room easily [01:52] doggymenz: they have that with direct audio on windows ,but it would be cool to have weird effects like that on linux too [01:52] yeah [01:52] didnt know they have that [01:52] ive heard people using it on ventrilo [01:52] which runs on linux ;) [01:52] oh cool [01:52] they sound like girl? [01:52] they sound like a chipmunk more like [01:52] hehe [01:52] also like a scary low voice hehe [01:52] you can do weird effects [01:53] wonder if it can make me sound scottish [01:53] doubtful [01:53] hehe [01:53] I'd like to clear the misconception that ALSA doesn't do "software mixing" by default. It does, and in fact, has since 1.0.9b. [01:53] i had the dmix .. thing [01:53] since for a long time [01:53] it does software mixing [01:53] yeah, but with bad latency? [01:53] i havent noticed much latency issue with it [01:53] no, with tolerable and mostly negligible latency. [01:54] like maybe 200ms at most? i dont know [01:54] oh [01:54] all the "horrible latency" FUD is just that. FUD. [01:54] how could pulseaudio have better latency than alsa? is that possible? :S [01:54] pulse is pretty damn badass though [01:54] crimsun: ah. didn't know that. might be why i'm rarely encountering issues on non-hardwaremixing boards these days :) [01:54] pulse even has a new method they are going to use thats going to be super awesome [01:54] it does weird stuff with buffering [01:54] there will be like zero latency [01:54] i forget what they were calling it [01:54] x1250: glitch-free in git has smarter semantics. [01:54] it may be out by now.. oh ya glitch free [01:55] one disadvantage of pulseaudio is noticeable cpu overhead [01:55] im loving pulse [01:55] it can do amazing things. like this listen.. i have a server with NO AUDIO HARDWARE [01:55] i can run pulse, i can run audio streams to 'null streams' and then use that null stream as a recording source [01:55] void^: by default, yes, because speex-3 is used. speex-1 has virtually unnoticable difference in resampling (to the user's ears) with lower CPU usage. [01:56] and then broadcast xmms audio over ventrilo on a machine with no soundcard [01:56] its pretty sweet [01:56] you can do tons of other amazing things with it [01:56] you could sorta do it with .. arts? i think before. but it was a tremendous pain in the ass to setup [01:56] and you had to do it EVERYTIME you wanted to run it [01:56] and also it required tons of utils [01:56] there's so much confusion and FUD surrounding Linux audio that I'm giving a talk on it at OhioLinuxFest. [01:56] now i configure it once, and it remembers the settings and works, with no more than the default installed tools [01:56] crimsun: right, thanks. i'll take a look on their site and read up before talking more garbage :] [01:57] i was skeptical of pulse audio at first but i am totally loving it [01:57] and it works with all the other sound library.. things [01:57] like oss apps can use it with padsp [01:57] crimsun, does Vista or Mac OS X have better audio than Linux? [01:57] and alsa apps can output to pulse instead of soundcard [01:58] doggymenz: that's a contextless and thus, unanswerable, question. [01:58] ive never seen features like pulse has in windows [01:58] windows has other features thatl inux doesnt have yet though [01:58] but i think the windows features are more of things youd never use. and dont care about [01:58] someone should make comparison [01:58] intangir: padsp suffers the same drawbacks that every preloadable library for Linux audio does. [01:58] and the linux features are things youd never think of, but once you have them holy shit are they ever badass [01:58] crimsun: what drawbacks? [01:59] intangir: OSS relies on mixing semantics that cannot be done reliably from userspace. [01:59] that is why OSS has always done mixing in kernelspace. [01:59] im not sure what that means really, im using it though and it seems fine [01:59] havent had any problems with it [01:59] actually it works better than OSS did [02:00] intangir: try using padsp with multiple OSS apps that rely on mmap access. [02:01] i'd say vista's userspace sound system is better than pulse right now [02:01] can you apply effect to a stream? like if i play piano, it will sound like piano is inside draculas 1800 year old stone castle? [02:02] bsnider: I disagree w/ that assessment under the most common Vista uses. [02:02] because all volume control is on one easy pane even for apps that don't normally have volume controls but do use sound [02:02] crimsun: im using it with wine [02:02] i use it on ventrilo and world of warcraft [02:02] they both share it fine [02:02] without any issues, with real OSS and also with alsa i had tons of lock ups, crashes, and it would just stop working without any obvious error.. other than no audio [02:02] intangir: now try it on a different audio chipset that doesn't support it. [02:02] but with pulse ive had 0 issues [02:03] im using it on a soundblaster live, and an AC97 thing [02:03] it can even play to both at once [02:03] intangir: of course, those both are mmapable. And they're both AC'97, BTW. [02:03] or to be pedantic, they both implement some AC'97 spec revision. [02:04] bsnider: I certainly do not disagree that UI is much better integrated in Vista. [02:04] you should keep an eye on what Fedora 10 offers in terms of integration. [02:04] pulse does this too, but not so conveniently [02:05] a pulse plugin for alsamix that integrates volume controls for apps would be nice [02:05] does it exist those voice alter machines that the murderer guy uses when he calls up the victims on the phone in movies? [02:05] they exist in real? [02:06] vocoders? Of course. [02:06] oh [02:06] bsnider: if I understand what you're alluding to, that's being tackled at both the app and the PA layers. [02:07] can you have a virtual software vocoder in linux? [02:07] currently - at least in GNOME - the mixer controls are a mess. Per-app volumes are utterly discombobulated. [02:07] vista not only has volume controls for apps that don't normally have them, like msn messenger, the control is persistent even whent he app isn't actually using sound, which is superior to what pulse does [02:07] bsnider: as I mentioned, that's being addressed in PA and in GNOME apps. [02:07] alls i know is pulse audio is kicking ass for me [02:08] also intrepid wont boot for me! [02:08] thats why i came in here [02:08] the volume control in pulse for pidgin will come and go too quickly to be useful [02:08] i hope fedora does fix it [02:08] it shows loading, then goes toa ablack screen, till i close it [02:08] till i hit power i mean [02:08] anyone else here still use xmms ;) [02:09] i havent found another player than can be used for an AUDIO and VIDEO playlist [02:09] why would i? [02:09] banshee [02:09] a searchable playlist too [02:09] uses mplayer plugin to play movies [02:09] i like mplayer out of all the movie players best [02:10] can't kmplayer do that? [02:10] or kplayer? [02:10] or smplayer? [02:10] i havent tried any of those [02:10] plays audio, and video, with a playlist, and its searchable? [02:10] bsnider: that, too, is being addressed. [02:10] and hotkeyable [02:10] * crimsun chuckles [02:10] all these problems are known and WIP [02:11] crimsun, not to be a buttinski here, but how do you know? [02:11] intangir: where does the boot sequence fail? [02:11] bsnider: I formerly maintained Ubuntu's audio infrastructure. [02:11] is your first name lennart? [02:12] bsnider: no, Daniel. [02:12] crimsun: once it should be opening X [02:12] bsnider: however, you can follow Lennart's work [02:12] who mantains it now? [02:12] bsnider: Luke Yelavich [02:12] oh, that's right [02:12] i'm using his ppa pulse 9.11 packages right now [02:13] they are starting to replace all of that "sink" crap with the actual sound card names [02:13] that's good [02:13] * RAOF should really file a bug about those; glitch-free seems to interact badly with hda_intel [02:13] RAOF: that's our fault; we're not using git. [02:13] (rather, a git snapshot) [02:14] crimsun: I thought we'd pulled in the alsa patches needed to make it work? Am I wrong? [02:14] namely, we need to use alsa-lib 1.0.18 + 2 git patches and PA git. [02:14] (that should be 1.0.18rc, obviously) [02:15] i use hda intel and pulse 9.11 and it works fine here [02:15] the roof hasn't crashed in [02:15] bsnider: "hda intel" doesn't mean a thing. [02:15] bsnider: You don't get annoying crackles with mini-xruns? [02:15] bsnider: different HDA codecs use different buffering techniques. [02:16] i think Intel HDA means that it conforms to the Intel Azalia spec [02:16] how would you define a mini-xrun? [02:16] http://en.wikipedia.org/wiki/Intel_High_Definition_Audio [02:16] k i gotta go. laters [02:16] A buffer underrun that's small enough to result in a 'crackle' or 'pop' in the music. [02:16] doggymenz: no, it means it implements - in some fashion - Azalia. There is no such thing as "conforming" to it, which is why there are hundreds of quirk entries in the sound driver code. [02:17] i haven't heard anything like that [02:17] oh [02:17] audio preview doesn't work, but i'm sure that's something else [02:17] RAOF: (I suppose you may be able to convince Luke to work at alsa-lib git and PA git, but I don't make any claim on his time) [02:18] crimsun: It's not in Intrepid yet; I should file the bug to either prevent it entering or ensure it enters with newer alsa-lib. [02:18] RAOF: _some_ of those symptoms can be worked around by tweaking fragment parameters for PA, but that's hardly scalable across audio hardware. [02:18] And Luke's on holiday at the moment, I think. [02:19] x is broken in intrepid? why my xchat and firefox have broken rendering? [02:19] doggymenz: Not here. [02:19] he deserves it. Picking up what he has been doing is hardly a breeze :) [02:24] i thought the latest alsa was already int he kernel [02:25] The drivers in the kernel tend to be slightly behind the ones in the external package. For the most part though, the ones in the kernel are the way to go [03:29] anyone know if X.Org 7.4 / X Server 1.5 is going to be in intrepid? [03:30] yes [03:30] it will [03:30] nice, and also DRI2 and KMS? [03:30] and therefore GEM too? [03:30] no, not DRI2 because it was delayed because Intel switched from TTM to GEM [03:30] KMS i dont know what is [03:30] kernel mode setting [03:31] dont know about that [03:31] No. None of those things are in drm master. [03:32] bummer [03:32] But Xorg 7.4 is in, we've got 1.5rc6 at the moment. [03:32] http://www.phoronix.com/scan.php?page=news_item&px=NjYzNw -- Xorg 7.4 have no DRI2 [03:33] but xorg 7.4 should have GEM support for intel though yes? [03:33] No. [03:33] ugh [03:33] GEM doesn't actually have much to do with Xorg, anyway. [03:34] It's for the drm. [03:34] RAOF, it's just a feature of the intel driver correct? [03:34] I mean as of now [03:34] Not really. It's partial GPU API. [03:34] s/GPU/GPU memory/ [03:35] right, it replaces TTM [03:35] Again, not really. [03:35] :) [03:35] but I thought all the TTM stuff was taken out of the current intel driver? [03:35] Yes. [03:35] But that's because their hardware is crap. [03:35] For cards with real resources (actual onboard VRAM, for example), TTM still makes some sense. [03:36] radeon is currently having a GEM+TTM makeover, I believe. And nouveau has some of that, too. [03:37] but why did you say GEM won't be supported by 7.4? [03:37] Because GEM isn't something that Xorg needs to 'support'. [03:37] seems like (as I said) it's more of a driver/mesa implementation [03:37] It's actually lower level than that. [03:37] wtf... (vboxgtk:16242): libglade-WARNING **: could not find glade file '/home/dana/vboxgtk.glade' [03:37] Neither X nor mesa really need to care about GEM. [03:37] oh ok... semantics... the intel driver and mesa that ships with intrepid will have GEM [03:38] lol, well RAOF I'm too dense for this I guess... thanks for trying to get me to understand though [03:38] No, it won't. Because that requires git snapshots of various branches. [03:38] self.base_path = os.getcwd() + '/' [03:38] Is it just me.... or is that way wrong? [03:39] Wrong, as in... I sure don't install packages there! [03:39] Seems pretty unreasonable, yes. [03:39] I guess I can always build mesa and intel and a KMS kernel too :) [03:39] If you run "git clone" against drm, mesa, and xf86-video-intel, you won't get any GEM code. [03:39] (You additionally need to checkout the various branches) [03:40] looks like they merged the gem branch to main a couple weeks ago [03:40] http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-intel.git;a=commit;h=08326827fdb8abecbff20c7b051537dbad433c40 [03:42] Oh, cool. [03:42] I don't follow intel too closely. My interest is with nouveau. [03:43] i have gf8600, but nvidia sucks, they only can make proprietary device driver, and dont want make open source, so i think they are assholes, and i dont like them [03:43] next time, i dont buy nvidia [03:43] ... [03:43] Intel had open source for GMA chipset, and ATI has open source for Radeon [03:44] now even VIA has make some open source, i read on slashdot [03:44] only nvidia dont want [03:44] ... [03:44] * DanaG ♥ ATI. [03:45] I just hope nvidia releases legacy drivers soon [03:45] legacy driver? [03:45] that work with the latest xorgf [03:45] *xorg [03:45] 96.x series and 71.x [03:45] my 177 driver work on gf8600 on xorg in [03:45] oh, they old [03:45] If I were them I'd be taking my time too [03:46] yeah, they must update those [03:46] yeah, they want take time be lazy to make you buy new [03:46] ... [03:46] if it was open source, it could be compiled for new xorg long time ago [03:46] doggymenz: Dude, there is a world outside your mums basement. Go out into it, please. [03:46] ??? [03:46] why? [03:46] you are nvidia fanboy [03:46] ? [03:47] or maybe you think its good to wait and take time and not update legacy driver? well, i think its being an asshole who is mean to customer [03:48] * JontheEchidna just sighs and goes off to bed [03:50] and mesa added some level of GEM support to main too [03:51] so mesa 7.1 and git driver-intel should = GEM I think... [03:54] My next laptop will have an ATI Mobility Radeon HD3650 (or Mobility FireGL V5700). [03:55] ryanpg: But GEM doesn't get you anything interesting. All you really want is DRI2 [03:58] what is GEM good for? [03:58] what is DRI2 good for? [03:59] gem manages memory [03:59] what is good for lol [03:59] GEM is good for implementing DRI2. DRI2 is good for fixing 3d + Composite [03:59] but i already have compiz [03:59] why should i care about GEM or DRI2? [03:59] using which driver? [03:59] well, dri2 is something I look forward to [03:59] GEM helps with EXA performance [03:59] i use nvidia-177 [04:00] what EXA good for? [04:00] then you already have a much better 3d/opengl than intel or ati will ever have [04:00] accelerating fun the stuff [04:00] nvidia has their own stack for _everything_ [04:00] oh [04:00] they even replace the glx module in xserver-xorg-coer [04:00] core* [04:00] too bad they wont share :( [04:00] I'd rather have open-source, myself. [04:01] nvidia implements opengl 2.1.2 on linux, nobody else comes close to that [04:01] even apple doesn't have 2.1 [04:01] apple has 2.0 [04:01] oh [04:02] but now is opengl 3.0 released [04:02] ati and intel dont have opengl 2.1? [04:02] no [04:02] they have 1.3 [04:02] oh, wow that sucks [04:02] 1.3 is like what? 15 years old? [04:02] yeah, tell me about it [04:02] with no memory manager (before gem) [04:02] meh, I stopped caring about OpenGL years ago [04:02] i dont know what memory manager is good for [04:02] using your graphics card's ram [04:02] nvidia also has this habit of breaking my consoles if you VT-switch too much. [04:04] the only way to get opengl 3.0 right now is to use windows, which is strange because microsoft isn't a member of the chronos group and doesn't know anything about opengl [04:04] oh [04:04] chronos's leading member, apple, doesn't even properly implement it [04:04] opengl 3.0 doesn't really offer anything over 2.0 [04:05] they cut all the cool stuff [04:05] the two upcoming updates will [04:06] but i'm sure tungsten will get around to implementing opengl 3.xx in gallium3d, which will then lead to open source drivers for ati and intel...some time in the 24th century [04:07] oh [04:07] what gallium3d good for? [04:07] what is memory manager good for? [04:07] gallium is for making mesa a useful platform for a modern 3d driver [04:08] oh [04:09] your graphics card has ram all of its own. there needs to be a memory manager to deal with that ram and use it. currently, there isn't one for linux [04:09] tungsten graphics website say "a group of graphics experts with an unrelenting desire to pursue state-of-the-art graphics technologies." but it still only has opengl 1.3 15 year old tech? not so state-of-the-art [04:09] that's what gem is [04:09] ok, so there isnt a memory manager, but i can still use compiz and play portal and hl2 in wine, so what is memory manager good for? [04:10] your nvidia driver does have a memory manager [04:10] but it's a proprietary nvidia memory manager [04:10] it's not available to other graphics cards and can't be included int he linux kernel [04:11] oh [04:11] seems like nvidia has done some amazing work that via, intel, ati, tungsten and nobody else has done - shame that they wont share it [04:12] meh [04:12] what's a shame is htat these issues have beena round for so long and nobody has fixed them [04:12] but lately things are picking up and moving? arent they? [04:12] yes [04:13] doggymenz: they probably really can't (code licensing and patents) and it wouldn't really be helpful as it's all nvidia-specific [04:13] oh [04:13] but obviously tungsten can do it [04:14] i'll believe it when i see it [04:14] right now gallium is vaporware [04:14] i hope they do it, but i'm not holding my breath [04:17] doggymenz: they're writing fresh code [04:20] Anybody else unable to type into this thing? http://bestbuy.shoplocal.com/bestbuy/default.aspx?action=entryflash&adref=header&h=504 [04:22] i can type zipcode in the flash [04:24] not that it's any help in my case but same :P [04:48] Eeh, I had to copy and paste. [04:59] monitor power saving doesnt work. help. [06:59] Ok, that's a pretty screwy behaviour. When I run emacs, it opens on the desktop _below_ my current one. Thus neatly making it appear that it doesn't start at all! [07:25] laughtear: So, what's your problem? [07:26] (Note that some drivers are broken and beyond our control, care of crappy binary-blobness) [07:26] RAOF: i was using hardy heron, and upgraded to ibex within it, now my screen resolution is so low, and nvidia drivers stopped working [07:27] laughtear: What's your card? [07:27] RAOF: probably i must uninstall the old drivers and re-install them, but i'm not quite sure how to do it.. [07:28] laughtear: Were you using the Ubuntu-supplied nvidia drivers before you upgraded? And did you upgrade with update-manager? [07:28] RAOF: i have an nvidia 6600gt with 128mb memory, and a samsung 226bw monitor which has default 1680x1050 resolution. [07:28] i upgraded with upgrade manager by pressing alt-f2 and typing -d. [07:29] RAOF: nvidia drivers were the restricted ones probably [07:29] Hm, OK. I believe it should have got everything working for you. If you were using the Ubuntu-supplied drivers, this would be good to report as a bug against upgrade-manager. [07:29] So, can you pastebin the output of "aptitude search nvidia-glx"? [07:29] RAOF: sure, gimme a sec [07:31] RAOF: here you are: http://paste.ubuntu.com/42958/ [07:32] Hm. That looks OK; you've got the right driver installed. [07:34]