IRC Logs for #crux-devel Wednesday, 2007-08-15

*** jaeger has quit IRC00:55
clbUpdate from opt: 15 Aug 07:52 - sqlite3: update to 3.4.203:00
*** treach has joined #crux-devel04:03
treachRomster: fwiw, I'm using an ancient 32MB radeon 9000, which gives "5238 frames in 5.0 seconds = 1047.591 FPS", with xorg drivers04:07
treachcpu running @700MHz04:08
prologic21361 frames in 5.0 seconds = 4272.064 FPS04:11
prologiccpu @2400Mhz04:11
treachI presume that's some indefininitely more modern nvidia card, with proprietary drivers.04:12
prologicyes :)04:14
treachthe point I was trying to convey was that this card does everything I need, like watching movies, pdfs etc without problems.04:14
prologicGF 7600 GS with 512Mb DDR2 ram04:14
treachand he should be able to find it in a dumpster, if need be.04:14
prologicon the card that is04:14
prologicdesktop has 2G of ram04:14
treachI don't think you'll find that kind of hw in a dumpster.04:15
prologicbut yes I get your point04:15
prologicwhat disturbs me is that his cpu maxes out04:15
treachbad drivers.04:15
prologicthis really shouldn't happen at the best of times even with old hardware04:15
prologicmy cpu hardly sees 2% usage04:15
prologicoccasionally my programming will spin it out :)04:16
treachif he's using the open nvidia driver, it has viritually no hw acceleration iirc.04:16
prologicin which case he should be using the propiatary drivers04:17
treachthen again, it's Romster we're talking about, no way of knowing what's wrong then. :>04:17
treachif it's old... there are no drivers.04:17
prologicthat too04:17
prologicit's probably a legacy card04:18
prologiche should just throw the PC out and buy a new one :)04:18
treachif it works that's just stupid.04:18
prologicwell he's the one that wants to upgrade it :)04:18
prologicit's hard to upgrade old hw04:18
prologicI used to have 3 old P2's for servers04:18
prologickinda hard to upgrade them now :)04:18
treachI only noticed he wanted to upgrade the video part.04:19
prologicnow I just have AMD64's04:19
treachwhich was my point that he could get sufficent performance from a dumpster.04:19
prologicyeah well04:19
prologicbut I fail to see how that'll improve his cpu performance04:19
prologicunless the card doesn't do any04:19
treachbetter drivers, less cpu-04:19
prologicbetter card, less cpu too04:20
prologicbut yeah04:20
treachwhy waste money on "better" hw, when all that is wrong is the drivers?04:20
prologicperhaps he should write some ? :)04:20
treachthere is the noveau(sp?) project.04:21
treachno idea how far they've come, if anywhere04:21
prologicmust write a driver one day :)04:22
prologicI don't suppose it's really that hard is it ?04:22
prologicand the kernel supports user space drivers now04:22
prologicwith a user space api04:22
prologicwith a minimal kernel-space driver shell04:22
Romstertreach, prologic i'm using the propiatary driver and its watching dvd's and playing runescape really canes my cpu04:35
prologicI seriously doubt playing a dvd would chew up your cpu04:36
prologicI'd say the game is badly written04:36
prologicand being java - well java is just horrible04:36
Romstermore ram on video card less work load on the CPU04:36
prologicwell not necessarily - depends on how much work the GPU is doing in the first place04:36
Romstereven PDFs take time to render if its huge and i have it zoomed in and move it04:36
Romsterplaying a dvd is jsut margional..04:37
Romstertrue.. i got a Geforce4 MX420 64MB and looking at a Geforce FX520004:37
Romsteri'd like to build a new machine but pfft fuck all money still.04:37
treachRomster: all of those things works just fine here.05:24
treacheven eclipse is quite useable..05:24
Romsterhmmz then why is my pc slow...05:25
Romstersome dma not on forvideo or something?05:25
treachno idea.05:25
treachbut I'd blame your video drivers.05:26
prologichave you looked at what process is chewing you cpu cycles ?05:26
prologicso would i :)05:26
tilmanif rendering pdfs is slow, then you can safely blame XRender05:26
tilmanresp your fucked up driver's render accel05:26
treachah right. I forgot to mention the possibility that you've misconfigured X.05:27
tilmanbtw, iirc "gtkperf"  can be used to measure gtk's performance05:28
tilmanwhich is heavily dependent on xrender too these days05:28
treachtilman: new poppler (, in my repo.) :p05:37
tilmanatm i have it in /usr/local05:37
tilmanbecause needs gtk-doc (no way, it needs docbook)05:37
treachok. :)05:38
tilmanyour README claims you included qt3 support. did you really?05:50
tilmani don't have qt3 and i see no footprint mismatch :P05:50
treachheh, it's not my readme, it's stens..05:55
treachI guess I should remove it.05:55
tilmanwhat the fuck is wrong with the poppler people wrt api docs?06:14
tilmanor am i too blind or stupid to find them?06:14
treach"rtfs" :p06:15
tilmanyes, but those hyperlinks and colors etc are pretty nice06:16
*** maro_ has joined #crux-devel06:20
tilmani think epdfview can handle pgup/pgdown these days06:20
tilmanthat's 50% of the reasons why i wrote raktpdf o_O06:21
tilmaniirc ;)06:21
treachheh, what about speed?06:21
treachit's unbearable slow here, and easily flips out if you scroll too much..06:22
tilmanwhen you scroll using the mouse it's really jerky06:22
*** jjpk has joined #crux-devel06:22
treachjerky? It goes completely crazy, I think it might even crash.06:22
tilmani click the middle of the document, and hold the mouse button06:23
tilmanthen i move the mouse up06:23
tilmani expect to move the document up, but it also goes down a little bit every s06:23
tilmanit's kind of zig-zagging ;)06:23
treachnice :D06:23
tilmanbloody annoying06:24
tilmanjjpk: what pdf viewer to you use? ;)06:24
treachIt's like that damned "smooth" scrolling.06:24
treachbloody useless.06:24
tilmani never understood that06:24
tilmanit was advertised as the greatest thing since web browsing06:25
tilmanoh well06:25
clbUpdate from opt: 15 Aug 11:26 - uw-imap: fixed url06:31
*** maro_ has quit IRC07:36
*** jaeger has joined #crux-devel08:10
j^2hey guys08:33
clbUpdate from core: 15 Aug 13:30 - [notify] openssl: added patch for CVE-2007-310808:53
*** clb has joined #crux-devel09:06
tilmantreach: haha, i think epdfview now doesn't segfault on pgup/pgdown anymore because glib relaxed its rules for passing NULL to v_value_unset() :)11:53
tilman2.12: sefault, 2.4: big-ass warning11:53
treachhm.. I can't say that surprises me. :/11:55
*** pitillo has joined #crux-devel12:03
*** maro has joined #crux-devel12:08
*** pitillo has quit IRC13:12
*** pitillo has joined #crux-devel13:12
*** maro has quit IRC15:25
*** tilman has quit IRC15:25
*** jjpk has quit IRC15:25
*** Romster has quit IRC15:25
*** pedja has quit IRC15:25
*** jaeger has quit IRC15:25
*** j^2 has quit IRC15:25
*** pitillo has quit IRC15:25
*** prologic has quit IRC15:25
*** tilman has joined #crux-devel15:26
*** pitillo has joined #crux-devel15:26
*** maro has joined #crux-devel15:26
*** jaeger has joined #crux-devel15:26
*** jjpk has joined #crux-devel15:26
*** Romster has joined #crux-devel15:26
*** pedja has joined #crux-devel15:26
*** prologic has joined #crux-devel15:26
*** j^2 has joined #crux-devel15:26
*** sets mode: +o tilman15:26
*** pitillo has quit IRC15:54
*** jaeger has quit IRC16:33
*** jaeger has joined #crux-devel17:06
*** jjpk has quit IRC17:48
*** jaeger has quit IRC23:44

Generated by 2.11.0 by Marius Gedminas - find it at!