While AMD has confirmed a bug in the Catalyst drivers that results in crashes, it claims it only affects a "small number" of users.
AMD has publicly confirmed reports of crashes with its ATI Radeon HD 5000-series graphics processors, which result in users systems displaying a series of grey or coloured lines.
As reported over on
Tom's Hardware, the company has advised customers experiencing the bug that the issue isn't a hardware bug but a software one exacerbated by an update issued by Microsoft for Windows 7.
Despite this explanation, users running other versions of Windows are complaining of the same issue - and in the release notes for the latest version of theATI drivers, Catalyst 10.1, the grey screen crash is noted as an issue which affects all versions of Windows.
Company spokesman Dave Erskine has stated that "
only a small number of ATI Radeon HD 5800 series and ATI Radeon HD 5700 series cards are exhibiting the behaviour" and promises that "
the [Radeon] team is testing a driver hotfix and initial tests indicate that it resolves the issue."
When pressed on a firm release date, Erskine said only that "
we need to test it further but we expect to make the hotfix available shortly."
The driver bug isn't the first software flaw to vex users of AMD's graphics hardware: late last year the PC release of
The Saboteur was marred by
incompatibilities withATI Radeon graphics cards which resulted in the game crashing on multi-core systems equipped with AMD's graphics processors. While this was later
fixed in a patch from Electronic Arts, it left a bad taste in many gamers' mouths - especially as
The Saboteur was part of rival graphics chip manufacturerNvidia's "
The Way It's Meant To Be Played" programme.
Are you pleased to hear that AMD is fixing they grey screen crash, or are you shocked that such a major software bug ever made it into production drivers? Share your thoughts over in
the forums.
36 Comments
Discuss in the forums Replyimportant news is important news regardless the source!
on sabatur it was EAs fault as it was their update that fixed the issue not AMDs driver or hotfix
Nevertheless, glad I stayed on 9.12 :-)
Another sign that AMD rushed out the recent software updates without any thought. This company has it's head completely up it's own ass!
Give us an honest explanation and we'll give you the benefit of the doubt. However, AMD cannot do this and have to point fingers and cry that everyones picking on them. Well I'm telling my mommy!
AMD said an update for Win 7 made it worse, not that the problem only affected Win 7 or that it was MS's fault.
I wish people would actually read.
when i was on xp i installed a new cpu e7600 and new gpu 4670 and installed latest drivers then i could play the game no problem then i fresh installed x64 win7 and installed drivers and it didnt work it was a wierd but so i just ran in it win xp mode and it worked flawlessly till they released patch
Despite this explanation, users running other versions of Windows are complaining of the same issue - and in the release notes for the latest version of theATI drivers, Catalyst 10.1, the grey screen crash is noted as an issue which affects all versions of Windows."
Read it yourself! AMD said it was an update on Windows7 then their own release notes say it affect ALL versions of Windows.
Anything else or is your defence dead?
http://www.thefreedictionary.com/exacerbated
Holy cow, seriously, read what you post CAREFULLY before you post it. The first part is the quote from AMD. It states "the issue isn't a hardware bug but a software one exacerbated by..." EXACERBATED by means made worse. The SECOND part states "Despite this explanation, users running other versions of Windows are complaining of the same issue..." This is where the writer at Tom's hardware pops in and is adding his own observations and interpretations. The problem is that you're interpreting his addition of the problems on other OS's to say somehow that AMD was originally saying that it didn't affect those other OS's. BUT THAT WAS NEVER STATED BY AMD. They said that the Windows 7 update made it WORSE. The bug was already there and on other systems. For the love of God man, learn to read above a grammar school level. :(:(:(:(:(
edit: I should have said "the writer from Bit-Tech" and not "the writer from Tom's hardware" so sorry about that inaccuracy.
Here Here!
Can only assume Crazyceo is an AMD fanboy! :P *Wooosh Sparks Fly!*
AMD and Nvidia have both had their issues with drivers but after using both types of graphics cards il be keeping my money with AMD for now.
Just wish my order of Vapor-X 5870 would hurry up and be in stock again! :D
Here Here!
Can only assume Crazyceo is an AMD fanboy! :P *Wooosh Sparks Fly!*
AMD and Nvidia have both had their issues with drivers but after using both types of graphics cards il be keeping my money with AMD for now.
Just wish my order of Vapor-X 5870 would hurry up and be in stock again! :D[/QUOTE]
Taros, crazyceo is more of an anti-AMD than of an AMD fanboy. Well, I guess misinterpretation is getting more common this days
But it did have something to do with AMD did it not? They even admit to it.
psst, actually Farfalho, he put that in there to see if he could get a rise out of crazyceo. Sarcasm f t w . :D
At least they're admitting it, if it's out in the public domain then there is a better chance they'll do something about it.
They dropped the ball bigtime and are scurrying as quickly as their little feet can carry them.
With that said...how exactly did they get away with this?
so it has to be a user error ;)
absolute bollocks
[SIZE=7]UNLEASH THE CAPSLOCK FURY! ;) j/k
Hope this doesn't affect their mobile/laptop gpu's.
Tell me!
I spend my days chained to rack full of gaming PCs so I've seen a few GSODs.
Many failures due to an OC will be put down to the cards as a GSOD may also occur on 124 (AKA insufficient CPU voltage) and memory related errors. So check you event log first 116 or 117 is Graphics. In my experience this is far more likely to be the problem even in stock clocked systems poor RAM/settings can cause crashes.
I have a system under testing at the moment that will run Prime95 and Furmark simultaneously but has intermittent crashes (Display Device Stopped Working...) on the desktop and in Direct3D. i.e. the CPU and Graphics are rock solid but the error still occurs. Unigene Heaven is a good tool to use as it can test in OpenGL and DX9,10 & 11 which can help isolate the problem. This is the first system I have seen with the exact issue so it is fairly rare. Changing the 5850 for a different brand did not solve it. I changed the RAM from GSKill to OCZ and it hasn't GSODed yet we'll see if it's still running in the morning.
Forcing Windows Vista/7 to run in Basic mode can stop the desktop freezes.
Something else I've noticed is that the systems that end up doing this often get the wrong Windows Index Score typically 6.0.
Sometimes the 9.12 driver will eliminate the issue, sometimes 10.1 will.
The problem is similar to one seen with the 4870x2 and 3870x2 shortly after launch.
8800 GTX/Ultra nvdklm problem has a similar failure mode in 3DMark06 was sometimes fixed by changing memory brand/type despite Memetest/Prime/IntelBurn stable memory.
I can only assume that it is either specific combinations of software and/or hardware or something that will occur randomly at some point in the indeterminate future...
Note: The PSU replacement was planned anyway for a crossfire setup I'm planning, (CM Storm Scout, too small for a 58xx or 5970.) It was just rushed because of this.
So despite the memory passing Memtest and Prime95 Blend testing the memory appears to have caused the problem.
I'll try and repeat the fix, for verification, next time the bug crops up.
http://support.amd.com/us/kbarticles/Pages/Grey-screen-and-vertical-line-corruptions.aspx
A fix for the 5770 and others should be coming out soon.