Printable View
No you look, then don't put OS's up then, not unless people have tested against all of them. Its utterly pointless, the way it is a simple yes or no would be just as accurate. The sarcasm is undeserved so climb down from your horse its not helpful.Quote:
Originally Posted by Biker
I've never had any issues but maybe cause I've been using the new BIOS right from the start.
Well if you have nothing useful to offer and / or are not suffering these issues as a lot of us are then please go and find a different thread to troll in.Quote:
Originally Posted by fornowagain
Edit:
10.2 Drivers now available from here
Edit 2
10.2 Windows XP 32 bit @ Megaupload
10.2 Windows XP 32 bit @ Mediafire
Edit 3:
10.2 Vista & Win 7 x64 @ Megaupload
10.2 Vista & Win 7 x64 @ Mediafire
I don't troll. If you bothered reading my comments, I already have asked a pertinent question. I also have two of these cards overclocked from the day of release with early bios' and no problems. You're way too defensive. It's just as important to determine hardware, settings etc from working systems as it is from those with problems.Quote:
Originally Posted by Biker
Both cards tested on 2 different ( 100% stable though, also tried @ stock to be 100% sure ) systems...Quote:
Originally Posted by fornowagain
System 1 - Gigabyte GA-EX58-UD5 + i7 920 - 6GB Chaintech DDR3 - Corsair HX620
System 2 - eVGA X58 Classified + i7 950 - 6GB A-data DDR3 - modified OCZ ProXStream 1000W
Tested both cards with stock BIOS & MSI's unlocked BIOS from Garrett Wu.
And both card mucked around on both systems? Exactly the same issues?
For reference, both mine are on the Asus bios, originally on Sapphire. On a Gigibyte x58 Extreme, pretty much the UD5 board. PCP&C psu.
Same issues ( GSOD @ 3D, and sporadic driver recoveries ).
They're out of control, and cannot be replicated as easily as someone would think.
The funniest part of this crazy issue is that I can get a GSOD at stock clocks easier and more often than at overclocked frequencies :D
I have yet to experience any 2D corruption issues at all, but some other members and people I know have :D
What boggles me is how this can be a driver issue since all of us use the same drivers and the same OSs yet some people are fine and some are badly screwed...
Programming device support software (drivers) is probably the worst job any programmer could have, because there are an infinite number of variables involved (Or so my professor told me). I believe this situation has happened before where some where affected by a problem, and some weren't, although they were using the same drivers.
A recent example would be the corrupted Intel firmware, where some people saw their entire SSD formatted, whilst others could run it without any problem.
I am having some fun with 3D corruption now :)
Tested with 9.12, 10.1 beta, 10.2 and 8.703 whatever that is ;)
Running @ stock clocks / volts:
3D Mark 01:
I am getting half the framerate for most of Nature (round about 420) which only jumps back to 999 briefly as the camera pans over the lake... anyone else getting this?
3D mark 05:
Corrupted smoke when the large canister drops on the minigun survivor at the end of Proxycon.
The smoke effect is replaced by very bright largish white squares which look a lot like the corrupted cursor that I also have had twice.
Also getting corruption during cpu tests 1 and 2:
In cpu test 1 as the camera pulls back I quickly start getting thn grey / brown lines appearing and dissapearing rapidly they are all displayed at isometric angles.
In cpu test 2 I also get the lines but these are joined by thousands of small black blotches that flicker on and off rapidly.
3D mark 06:
The smoke is also corrupted in proxycon in a very similar way to '05 except it is prettier / shinier :)
Both CPU tests are corrupted in exactly the same way as '05.
As an experiment I overclocked the card to 900 / 1300 and ran '06 again.
Unsurprisingly the corruption was worse:
In Proxycon wall textures were dropping in and out and the whole screen would flash.
In cpu 1 in addition to the '05 mess I also got for the first time vertical black / grey "prison cell" bars that many others have also been reporting.
In cpu test 2 I nearly had the whole screen flashing with large grey polygons in addition to all the rest ... I really thought I was going to get my first GSOD :D
This time around Canyon Flight was also not happy with the sea corrupted at the beginning, followed by the canyon walls then the monster and ship underbelly had that crappy mottled shadow effect texture very similar to the rubbish shadow effects in GTA IV.
After this I went back and repeated all tests 3 times and found that in my case the corruptions were easily repeatable.
Conclusion:
After lots of testing I am fairly cartain that no amount of updated drivers or hotfixes are going to get this card running well.
:yepp:
...
What I find hard to believe is that ATI basically has the solution at hand, but doesn't really use it.
First off, they release a hotfix for 9.12 which gets rid of the increased cursor. Works perfectly. But then, with 10.1, the problems return. Then they release a hotfix for 10.1 which gets rid of most of the GSOD's, which also works amazingly well. But when 10.2 is released, the amount of GSOD's start to increase again.
ATI, try combining those 2 hotfixes. You'll please a lot of people ;).
I have had lockup issues on the 5850 with the card overclocked with the MSI Unlock Bios when overclocked playing L4D 2 only. I don't know why LFD 2. Last time I played the card was screaming. There is no fan, since I watercool. The card itself was making noise. If I drop the OC on the card L4D 2 seems to play well w/o issues. Hopefully this will be fixed. All my other games run fine overclocked to 1300v, 1030clock, and 1200 ram.
I have the same problem. My 5850 gives off a high pitch noise when I'm using Google SketchUp :S. I'm pretty sure it's not the fan, but the card itself.
temp related issue maybe with VRM's or MEM IO...what are the MEM IO temps reading when errors happen..?
Voted "Win 7 x64" since i don't use anything else. I wonder if ATi can fix their cards with drivers or we need to wait for the new revised cards.
The high pitch noise is a choke or coil "resonating" on the card. It most often occurs in 3D programs with very high FPS (dark screen at a game intro and such).Quote:
Originally Posted by Alexandr0s
No issues with GSOD's with my 5870. It's a XFX card running win 7 x64. I've even been running under 100% load for over a week crunching Milkyway@home.
I get GSOD with catalyst 9.11 and 10.1 hotfix, (Win7-64)
and BSOD with 10.2 beta and 10.2 final. BSOD indicate a dxgmms1.sys corruption
I remembered that these problem appear after I updated my GA-EX58-UD3 bios to F8, F9 also.
Since I am back in the BIOS F6 (2 days), everything works perfectly!
=>BenchZowner: what bios you use for you Gigabyte GA-EX58-UD5? if it's F2 or F3, I suggest you to try to return to F1 for test...
Bios F2 & F3 provide same new feature that F8 & F9 for GA-EX58-UD3R (Add support for GulfTown and "Enhanced Memory capability ").
Interesting article posted over @ Toms covering 2D corruption & performance issues on the 5xxx series now opdated with responses from ATI / AMD...
Thankfully now also republished in English (Mein Deutsch ist nicht so gut, ja?) :D
Edit:
After a lot of reading and some fiddling around I have finally managed to get rid of 95% of my 2D corruption.
How you ask?
By disabling all cursor and advanced drawing graphics acceleration!
In XP: go to display properties > advanced > troubleshoot and then set hardware acceleration two notches away from full....
This is a great relief as trying to read a forum was nearly impossible as every time I scrolled up or down I instantly got half a page of unreadable garbage!
Note this does not get rid of it completely, to do that i need to completely disable hardware acceleration!
I haven't experienced it on either XP Pro 32bit or Vista Home Premium 64bit.