You are not logged in.
Vi0L0 wrote:catalyst 11.3 is already on amd's server. Downloading...
well... and maybe good support of gnome 3
Yesplz!
And uh... *goes to check AUR*
With catalyst-test gnome 3 from [gnome-unstable] seems to work fine.
Moving windows through desktop isn't as fast as it should be, but other than this - it's ok, no artifacts etc.
edit: aahh... i said it too early - it's still freaking out after launching some kde apps inside... but still there is a progress
i'am using kernel26-pf (2.6.38-pf2) and when version 11.2 worked ok, now on 11.3 i get :
Fatal error on inserting (/lib/modules/2.6.38-pf/video/fglrx.ko) no such device
I have to boot with option acpi=off , because i have ati mobility radeon 3650 and it have problems to work with active acpi.
As far as i understand on 2.6.28 this version of driver don't load with acpi=off can anyone confirm it, or have any workaournd ?
I tested new catalyst-test and i got the same problem...
i'm no expert here, i was never using acpi=off, but i'm not even able to boot [testing] kernel with this option, it stops on: Starting udev... done
Last edited by Vi0L0 (2011-03-30 17:16:46)
Offline
Hi guys. I'm trying to uninstall catalyst (hook & utils) with pacman -Rd but it totally refuses to do that unlike before. Is there's any way around this?
Offline
Hi guys. I'm trying to uninstall catalyst (hook & utils) with pacman -Rd but it totally refuses to do that unlike before. Is there's any way around this?
It's new functionality of pacman 3.5, use pacman -Rdd
Offline
Falstaff wrote:Hi guys. I'm trying to uninstall catalyst (hook & utils) with pacman -Rd but it totally refuses to do that unlike before. Is there's any way around this?
It's new functionality of pacman 3.5, use pacman -Rdd
I wasn't aware of the new functionality. Thanks.
Offline
I've tried running gnome3 with catalyst on my HD 3850.. it works well, except the compositing effects don't work right on my system.. they just look digital and scramble the screen until the effect is over
is the new catalyst an improvement for anyone for running gnome3?
If I have the gift of prophecy and can fathom all mysteries and all knowledge, and if I have a faith that can move mountains, but have not love, I am nothing. 1 Corinthians 13:2
Offline
Gnome3 is really buggy with the catalyst-test, but I'm getting a bug error for some reason.
irc.freenode.net @ #koush,#android-dev,#archlinux64
Offline
Gnome3 is really buggy with the catalyst-test, but I'm getting a bug error for some reason.
and for open drivers?
Offline
i tested with the open driver on 1xRadeon HD 4XXX and 2xRadeon HD 5XXX
works just like you expect. Smooth, fast, without graphical bugs or things like that (or let us say very rare )
Offline
Waiting for support anything new is so typical in catalyst...
Gnome3 is really buggy with the catalyst-test, but I'm getting a bug error for some reason.
$ aticonfig --initial
Fail to link to fglrx-libglx.so, please check whether driver is installed correctly
Yes, i found that this "bug" touches also ubuntu.
More info is here: https://bugs.launchpad.net/ubuntu/+sour … bug/709505, look aspecially at comment #38.
I have tried to do some copying/symlinks but neither of this helped - but i could make some mistake, i don't have PowerExpress to test it better.
The good thing is that this "bug" isn't affecting anything, it is more like aticonfig got some funny problems, but it seams to work fine.
And as soon as your X starts, you don't see errors in /var/log/Xorg.0.log, and your fglrxinfo is telling true - everything is working ok.
$ aticonfig --initial
aticonfig: Writing to '/etc/X11/xorg.conf' failed. Permission denied.
as for this - run aticonfig as root
Offline
GNOME 3 users:
you probably noticed that GNOME 3 is in [testing] right now...
On "[testing] repo forum" wonder presented patched gnome-shell & mutter packages for nvidia users, which i had installed and it helped me with my kde apps... who knows maybe it can also help you in some/other cases
Last edited by Vi0L0 (2011-04-09 20:34:52)
Offline
kernel 2.6.38 in [core], maybe it is necessary to update catalyst dependency array in repo package:
:: catalyst: requires kernel26<2.6.38
AFAIK, catalyst 11.3 is working with 2.6.38 using a patch, right Vi0l0?
Offline
Interesting analysis on open source ATI drivers over time:
http://www.phoronix.com/scan.php?page=a … q111&num=1
KUDOS to the developement team!
Offline
kernel 2.6.38 in [core], maybe it is necessary to update catalyst dependency array in repo package:
:: catalyst: requires kernel26<2.6.38
AFAIK, catalyst 11.3 is working with 2.6.38 using a patch, right Vi0l0?
Indeed, we got 2.6.38 patches since it's very beggining.
catalyst was rebuild, it should work well. Thanks.
BTW i was expecting that with kernel 2.6.38 also xorg-server 1.10 will land in [extra], just like it was mentioned on [arch-dev-public] ML some time ago, but oh well... if it will land you know what to do - just to remind - either:
- add xorg-server-* and xf86-input-* packages which you installed into pacman.conf's IgnorePkg list
OR
- use [xorg19] repo - read wiki
OR
- install catalyst-test package which have xorg-server 1.10 support, but know that it lacks vaapi support
Last edited by Vi0L0 (2011-04-11 14:26:23)
Offline
snack wrote:kernel 2.6.38 in [core], maybe it is necessary to update catalyst dependency array in repo package:
:: catalyst: requires kernel26<2.6.38
AFAIK, catalyst 11.3 is working with 2.6.38 using a patch, right Vi0l0?
Indeed, we got 2.6.38 patches since it's very beggining.
catalyst was rebuild, it should work well. Thanks.BTW i was expecting that with kernel 2.6.38 also xorg-server 1.10 will land in [extra], just like it was mentioned on [arch-dev-public] ML some time ago, but oh well... if it will land you know what to do - just to remind - either:
- add xorg-server-* and xf86-input-* packages which you installed into pacman.conf's IgnorePkg list
OR
- use [xorg19] repo - read wiki
OR
- install catalyst-test package which have xorg-server 1.10 support, but know that it lacks vaapi support
Would it be enough to just put xorg and xorg-drivers group in IgnoreGroup list ?
IgnoreGroup = xorg xorg-drivers
I believe in a world I can and do understand. A rational universe, explained through rational means.
Offline
@LeCrayonVert:
this could be good idea. Although i don't know how will package xorg-server-common behave.
This package isn't related with any group, but xorg-server package got dependency set on it.
This dependency is set only on name, not on version.
So will then package xorg-server-common be updated to 1.10 version?
Last edited by Vi0L0 (2011-04-12 00:18:19)
Offline
Is catalyst-testing fairly stable? I'm considering switching with my desktop (4870x2) so I can update to the latest kernel and xorg-server. I don't use vaapi because there are some bugs that cause graphical artifacting in playback with certain codecs/containers. mkv with H264 IIRC.
Offline
edit: forget it.
But catalyst-test seems to work good, yes.
Last edited by Cdh (2011-04-12 07:55:45)
฿ 18PRsqbZCrwPUrVnJe1BZvza7bwSDbpxZz
Offline
Argh, wonder removed his packages which seemed to help quite a bit. Currently running gnome3 and gnome-shell from [testing] with catalyst-test and xorg-1.10, and experiencing quite some artefacts and more notably slowdowns on most animations.
Offline
edit: forget it.
But catalyst-test seems to work good, yes.
I agree, I just installed today and it runs quite well. So far it's at worst equal to catalyst-daemon 11.3.
I still experience the annoying bug where smplayer's overlay (while in full screen) fails to display properly and causes temporary video artifacts on certain MKV H264 videos.
With compositing disabled the overlay will display when I hover to the bottom of the screen with the cursor, but it will still cause video artifacts. This began around 10.10 or so. Not a big deal, but annoying nonetheless.
I guess the last pacman update change the way dependencies are handled, since I had to install both catalyst-test and catalyst-utils-test in the same command in order to prevent the 'XXXX is required by YYYY' message.....even with the -Rd switch. I suppose this is good for preventing the user from removing required packages.
Last edited by DarksideEE7 (2011-04-14 00:44:19)
Offline
have you tried vlc 2.0 git ?! It is quite stable and . . . waouw : ultra smooth videos (even 1080p mkv) with pratically no cpu usage ! really better than mplayer-vaapi !
Dell Studio XPS 1647 | i5-560m | 4go ddr3 | ATI radeon mobility 4670HD 1go gddr3 | Seagate Momentus 7200.4 G-force | wled 1600x900
Offline
have you tried vlc 2.0 git ?! It is quite stable and . . . waouw : ultra smooth videos (even 1080p mkv) with pratically no cpu usage ! really better than mplayer-vaapi !
I've stuck with smplayer for some time now, but I often switch back and forth. To be honest I had high hopes for vlc because I've had the best luck with it streaming DTS and AC3 via SPDIF, however the bitstream would simply 'crash' after 15 minutes or so. I couldn't get it to work at all with smplayer.
With that criteria eliminated I chose smplayer because:
It automatically saves the last position of each video, even after reboot. I think this is done by database, but I'm not sure.
I prefer the overlay (not really a big deal, but it counts)
A few other reasons that I can't recall off the top of my head.
I am certainly willing to give it a shot though. I haven't tried the git package recently, and I noticed that there is a phonon-vlc-git. I wonder if that will help with my DTS/AC3 problem?
My sound card is an Auzentech Prelude X-Fi.
Thanks for the heads-up, downloading now
Offline
With that criteria eliminated I chose smplayer because:
It automatically saves the last position of each video, even after reboot. I think this is done by database, but I'm not sure.
I prefer the overlay (not really a big deal, but it counts)
A few other reasons that I can't recall off the top of my head.
I also prefer mplayer but watching 1080p movies on Linux with vlc git is so much better than switching on Windows every time !
Dell Studio XPS 1647 | i5-560m | 4go ddr3 | ATI radeon mobility 4670HD 1go gddr3 | Seagate Momentus 7200.4 G-force | wled 1600x900
Offline
DarksideEE7 wrote:With that criteria eliminated I chose smplayer because:
It automatically saves the last position of each video, even after reboot. I think this is done by database, but I'm not sure.
I prefer the overlay (not really a big deal, but it counts)
A few other reasons that I can't recall off the top of my head.I also prefer mplayer but watching 1080p movies on Linux with vlc git is so much better than switching on Windows every time !
When I get my ASUS UL30A back from RMA (screen problem, loose cable) I will need to test and compare with smplayer. CPU usage isn't much of an issue on my desktop, since I have a Q6600 OCed to 3.76 GHz, however the laptop could use a little boost on higher bitrate 1080P MKVs.
I also noticed the vlc-git has changed the audio menu a bit. It has more detail alsa device descriptions. Hopefully that means that the audio backend has been improved. I also installed phonon-vlc-git.
Offline
Hello guys, is it normal that xf86-video-vesa runs far better then any xf86-video-ati & ati-git drivers ? o.O
With vesa my laptop doesn't overheat and runs as all the desktop-effect even smoother o.O
( ati radeon M4830 )
Offline
I'm using 2.6.38 and x server 1.10, and catalyst-test 11.4, but I can't run into gnome, the log located at :
http://pastebin.com/iPqA0PGJ
It said no enough video memory, my brain crashed.
mine is desktop pc, i5 2300 + dataland6870 + ausuP8H67
Offline