ATI Catalyst 8.8 drivers not compatable

Catalyst 8.8 drivers released, brings Hybrid Graphics support to XP ... nothing but pain with the SL viewer crashing and closing as soon as I enter world.

They work with a lot of things but SL's old Visual C++ or something old in the code is not working with the new drivers.

I see a lot of complaints out there so I know i'm not the only one with the problem.

Beware if your thinking of updating your drivers for ATI cards.

Ooh, good thing I haven't updated them yet. Just downloaded, but didn't install the drivers.

better read this first.............

http://jira.secondlife.com/browse/VWR-8014

I'm having no problems with Catalyst 8.8; it's been more stable than 8.7 was, I had some mysterious crashes with that version. But I've got an ancient integrated graphics chipset (Radeon XPress 200M), so the problems in the new driver might not affect me.

4850 and 4870 cards crashing Client

Still no answer from ATI and AMD on the cause or the fix or if there is a way to remove all the driver and install the old one back..will keep you posted

Strangely enough, the 8.8 drivers fixed my crashing. I would crash on 8.6 and 8.7 in Vista x64 Ultimate + 4870 as detailed here: /111/82/268338/1.html#post2120433

Same problem here.... 3870 card on Vista x64 ultimate and catalyst 8.8, always crash.
Just done a rollback to 8.7....

Wait, is this affecting only the newer cards? I'm running an X1650.. so I guess I'll give it a shot.

Hi Folks,

I've been experianced strange behaviour like Textures mismatch, very slow rendering during Texture load and total SL Viewer crash, right after login with a ATI 4850 card, using the latest ATI Driver v8.8.
I've opened a JIRA issue, right here: http://jira.secondlife.com/browse/VWR-8717

Please login to JIRA, and VOTE for this issue, so it will become more & more importand to be fixed ASAP. This is the only way to get rid of this delayed attention of Linden Lab, to support ATI Cards PROPERLY in the SL Viewer!

This urgently need to be debugged and fixed by the LL developer team.
Hey Guys, since the ATI 4850 Card is so rockin' cheap while it bring's a tremendous boost to your PC, it'll be spread on the market rapidly during the next month. So this has to be paid attention...

If you VOTE for it, also WATCH this issue on JIRA, so LL will notice there's a lot Users out there, using a ATI Card and struggling with the SL Viewer (wich is btw the ONLY application/game nowadays, that has such problems with a ATI Card).

Thanks for your attention & voting for this, to help a lot of SL Residents, playing SL with a ATI Card, BETTER.

Cheers,
Dandare

I would think it's the other way around. ATI develop a driver that is compatable with OpenGL rendering so programs like Second Life will work better. Problem lies in the driver, not the program that uses the driver. Second Life works for the vast majority of users.........users with different video cards/accellerators than ATI. Even many ATI cards have no or few issues with SL.

Go to ATI and submit a JIRA.............that would be the more logical thing to do.

Can you explain the issues with this a bit more? I am having a hecuva time with texture issues right now, clothing textures that result in this: http://www.flickr.com/photos/dinalyadawes/2814334521/ even though the clothing item is perfectly fine and these extra bits of material shouldnt be showing. Not sure I am running the same thing everyone else is (darn it, the computer geek hubby is at the gym and I'm rather lost when it comes to 'what' I use lol) and this issue started last night. Its not the designers, its the extra bits of texture showing regardless of the slider positioning, not blending etc. Just curious if its the same type of issue, might have multiple or another cause.

Is it this problem: https://jira.secondlife.com/browse/VWR-7957 ?

Thank you! *saves for her tech support to see (ie, the hubby)* Maybe reading through it will find a fix. Its driving me batty >.<

Although it's more logical, it's not the most likely way to get the problem solved. This will not be the first time that Linden developers have had to tutor the ATI OpenGL experts. (They've had to do it for Apple, too: same cards, differently b0rked drivers.)

So now Linden Lab is supposed to go to every driver manufacturer and teach them now to develop drivers? I think LL has enough learning to do on their own with SL.

Besides, aren't all LL techs, idiots, morons and grade school dropouts?

It's ATI's problem..........ATI needs to fix their own problems. No ATI tech is going to walk into LL technical lab and "teach" them how to develop a game/platform that complies with ATI's driver standards.

You were kidding, weren't you?

Dear Peggy,

really, it's not a ATI specific problem (only). Of course, graphic card producers also fix minor things in theier drivers, to make them compatible for single games. You can read this in any driver release note.
But You know, why iam state this as a matter of fact that the SL Viewer is responsable for this OpenGL bases issues? Because i tried maybe 20 different games (Also OpenGL based btw) on this machine, and really ANY perform perfect, but the only that makes urgent trouble, is..unfortunately SL (the most importand app for me btw). I wont go into detail, you can read the issues i have, in my jira report.
The LL Development need to feedback and check the OpenGL routines, maybe together with ATI. But afaik, really it's just NOT a ATI fault.

Greetings,
Dan

I really wasn't kidding. It's just the way of the world: there's no better way to find bugs in any software than to use it for something that really exercises the edge cases of the API--and the WindMark folks aren't noted for constraining themselves to the least-common-denominator functionality of OpenGL.

The VWR-8717 jira has already been assigned, so I don't think voting for it would have much additional impact. Scanning through the comments (also on the VWR-8014 predecessor), it's not evident that any resident has looked at the viewer source or found a bug there. It's certainly possible that such a bug exists, but the fact that the problem seems to be specific to a particular generation of card certainly suggests that it originates with ATI's driver implementation for that hardware.

(FWIW, I have a newish ATI card that's back in its box and on a shelf, replaced by the nVidia equivalent, because ATI's *Linux* OpenGL drivers for that card are just unusably bad. Although AMD has some serious OpenGL experts on-staff, they just can't seem to produce a driver that works for all their hardware and all operating systems at the same time.)