AMD 8.42.3 Driver Released -- The Baby Is Born!

[mahler]

Does anyone know why 8.42.3 isnt already released on the amd driver page?
Maybe they are ashamed to release such a crap.

They probably want to make an inventory of known-bugs first. And then decide how to announce it.

On the other hand, I have the theory that whenever AMD/ATI releases a driver which has many bugs, they do this at the end of the month. So that the next release will be relatively sooner and people get their fixes. So I am curious if 8.43 will indeed be sooner rather than later.

[Stormking]

Some progress but still not the great leap I'm waiting for

8.42.3 is a bit better than 8.41. Here are my experiences:

Positive:

- XVideo-on-TV-Out is fixed. There are still problems, switching to fullscreen gives me a blank screen. But finally I am able to use the oxine media player (a xine frontend), again.

- Unlike 8.41, the latest wine works, even on the second screen.

- No more screen corruption in Maya when the Composite Extension is enabled

Negative:

- NO AIXGL for me :-( Xorg.log says it's enabled but compiz can't find GLX_EXT_texture_from_pixmap. The workarounds in this thread do not work for me. With LIBGL_ALWAYS_INDIRECT=true, X is killed immediately after starting compiz.

- I have a dual head setup and every OpenGL app gives me the "*** glibc detected *** glxinfo: double free or corruption (!prev): 0x08063060 ***" error.

- Civilization IV under Cedega is extremely slow. I used to have playable 15-18 fps. Now, I only get 4-5.

[chabo]

Downgradeted to xorg-server-1.3 and compiz-fusion works with LIBGL_ALWAYS_INDIRECT=true.

glxgears:
61529 frames in 5.0 seconds = 12305.760 FPS

fgl_glxgears:
15601 frames in 5.0 seconds = 3120.200 FPS

Linux pass 2.6.22-gentoo-r8 #1 SMP PREEMPT Wed Oct 17 11:12:00 CEST 2007 x86_64 Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz GenuineIntel GNU/Linux

ATI 1950XTX

I don't have problem with slow scrolling in Firefox but XVideo and all 3D application flick...

Without AIGLX all work good, except switching video modes in applications, but its only x86_64 problem i think

[Wojtek]

There still no sign of the new drivers on ATI's website. Who knows, maybe we are treated as beta testers and fglrx 8.43 will be released soon, fixing most of the bugs listed on this forum? I cannot think of any other reason why ATI's website still shows only 8.40?
Cheers,

Wojtek

PS This is my first post on this forum so hello everyone!

[LucasG15]

For those that are getting flickering while watching a video, do you have compiz enabled? If so, try enabling vertical refresh in the ATI catalyst control panel.
Under 3D -> More Settings -> Vertical Rerfrsh: I set the slider all the way to the right (Quality).

[Roger]

I just installed the fglrx 8.42.3 driver yesterday in a Ubuntu Gutsy AMD64 (yes, the one with the problems). I have a Radeon 9600XT.

I used the script meant before in this thread to install, and a common tutorial to install fglrx from the source.

My impressions are:
1- AIGLX is running fine. I was even surprised as it started compiz without any complaining or white screens. Speed for Compiz seems fine to me. I was using the opensource drivers, so now compiz seems better or the same. (And to think that Michael told that hadrdly compiz would run in a 9600XT)
2- Scrolling may be a little slower, I noticed something like that, but nothing really abysmal, like some posted here.
3- Everything else seems OK. I hadn't got much time to test it yet, like the new control panel, but so far it seems fine.

I saw people calling this driver crap, well, you should stop complaining and simply look to the great step ATI/AMD took with this driver. Now the refining begin.

[aoanla]

I just installed the fglrx 8.42.3 driver yesterday in a Ubuntu Gutsy AMD64 (yes, the one with the problems). I have a Radeon 9600XT.

I used the script meant before in this thread to install, and a common tutorial to install fglrx from the source.

My impressions are:
1- AIGLX is running fine. I was even surprised as it started compiz without any complaining or white screens. Speed for Compiz seems fine to me. I was using the opensource drivers, so now compiz seems better or the same. (And to think that Michael told that hadrdly compiz would run in a 9600XT)
2- Scrolling may be a little slower, I noticed something like that, but nothing really abysmal, like some posted here.
3- Everything else seems OK. I hadn't got much time to test it yet, like the new control panel, but so far it seems fine.

I saw people calling this driver crap, well, you should stop complaining and simply look to the great step ATI/AMD took with this driver. Now the refining begin.

But, of course, they're complaining because their experience hasn't been nearly as good as yours...

[kir?]

I just installed the fglrx 8.42.3 driver yesterday in a Ubuntu Gutsy AMD64 (yes, the one with the problems). I have a Radeon 9600XT.

I used the script meant before in this thread to install, and a common tutorial to install fglrx from the source.

My impressions are:
1- AIGLX is running fine. I was even surprised as it started compiz without any complaining or white screens. Speed for Compiz seems fine to me. I was using the opensource drivers, so now compiz seems better or the same. (And to think that Michael told that hadrdly compiz would run in a 9600XT)
2- Scrolling may be a little slower, I noticed something like that, but nothing really abysmal, like some posted here.
3- Everything else seems OK. I hadn't got much time to test it yet, like the new control panel, but so far it seems fine.

I saw people calling this driver crap, well, you should stop complaining and simply look to the great step ATI/AMD took with this driver. Now the refining begin.

What BIG step? radeon open source works with things like compiz and xmame-opengl better than 8.42 and I don't remember I waited for radeon open source driver mornings and nights, 22 days. It exists since my first experience with linux, 1 year ago!

[Euskafreez]

I'm waiting this driver for years

[enzobelmont]

mine was working fine until i upgraded to xserver 1.4, since then my results are like yours.

i will downgrade to xserver 1.3 and see what happen.

by the way i'm usin gentoo.

yeap, downgrading to xorg-server 1.3 fixed segfault problem.