FREQUENT Crashing with nVidia GeForce 6800 GT

I've been experiencing frequent and random crashes with my GeForce 6800 GT. The crashes don't seem to have any triggers, they do however only occur during game play. The crashes occur randomly, regardless of what game I'm playing, it could be fifteen minutes into a game or an hour. When the crash occurs the screen will freeze with a random color mosaic pattern. Sometimes the audio may loop, or continue playing. The crash is unrecoverable, and forces me to reboot the computer.

I have tried just about everything I can possibly think of to fix this, but nothing seems to work. I've updated all drivers, tried old drivers, turned FastWrites off in BIOS, ran a memory test, etc. As far as cooling is concerned I don't believe that is the issue as the problem has just recently shown up, and I have made no changes to my cooling layout. The card runs idle at about 60 C to 65 C, and while gaming it can get into the 80's. If any extra info is needed, DXDiag info etc. I'll be more than happy to provide it. I really appreciate any input anybody can give me.

Do you have the latest bios for your motherboard?

Also, can you give us the info on your PSU. Make/model etc.

(P.S. your video card getting up into the 80s.... high 80s?)

Don't have the latest, but I have an updated version. The only updates available currently only add support for new processors, doesn't seem to be any other fixes. During game play it seems to average in the higher 70's and lower 80's, I've never seen it break 90. From what I've read on the card this should be acceptable, only the mid to higher 90's seem to cause issues, the threshold is 120 C. As far as the PSU, I'm sorry, but I'm not familiar with the term I'm a bit of a newbie to all of this.

I have a 6800GT as well, and mine now is idle at 55c, and has never reached 75 at full load just so you have another reference. But you're right, it doesn't sound like you are getting too hot if the max you are hitting is mid 80s.

Sorry about being a geek and using abreviations. lol.

PSU is power supply unit.

It hit me about as soon as your replied what PSU meant, sorry. I have an Antec NeoHE 430 ATX12V 430W Power Supply, you can get the hardcore specs on it here if you need them, http://www.newegg.com/Product/Product.asp?Item=N82E16817103939.

Alright then, if it's an Antec then that is probably not the problem.

You have tested your ram, no errors.
You have the latest drivers for the video card
You have the latest bios
Your power supply is a good one (triple 12v rails... wow)

...... I am running out of ideas on this one.

Take a look in your event viewer for any hints. Right click on my computer, then manage, then event viewer. Look around for errors.

Same, that's why it's killing me. As a last resort I even went so far as to reformat and do a clean install of Windows in the off chance there was something software-wise I just wasn't finding. It's really weird though because of how random it seems. I played Fable for about two hours straight last night, no troubles, I thought the reformat helped. Then I pick up F.E.A.R. this morning and I don't even make it fifteen minutes before a crash hits. Now the crashes have a occured while playing Fable before as well, they do however seems to happen more frequently in F.E.A.R.

Have you tried updating the chipset drivers?

How long did you run memtest? I'd let it do 6 or 7 passes.

I'd probably go ahead and install that latest bios as well. Can't hurt.

And in case you didn't see my late edit in my above post, try poking around in your event viewer for some hints.

Tried the Event Viewer, no dice. Cleared the logs and then started the game up and played until a crash hit. When I got back in the logs didn't show anything out of the norm. My chipset (VIA) drivers are up to date, I'll run Memtest a bit longer and try to update the BIOS too. Thanks for the info.

You say when it crashes you are getting a jumbled color screen right?

I'm starting to think that your card may actually be..... nah. I won't say it. I hate saying those words. Do you have another card you can test in your machine? One from a friend for example?

Yep, jumbled color. I don't really have another card that I could switch out. I do however (it escaped my memory earlier) have a friend that mentioned he's been having an issue with the exact same problem. He has the exact same card, he mentioned updating to the newest drivers from nVidia seemed to help, but that he hadn't really tested it out much enough to be sure. I think though if it is a driver issue, it would probably be a known issue by now as the drivers have been out for about a month now.

Tried everything, no go, still crashing. Going to try the new nVidia beta drivers, looks like it listed more than a few fixes for the 6800.

Well that didn't work either. Called BFG (the card maker) and they're going to do a RMA, only thing they came up with is that my sound card is on the same IRQ so there might be a conflict. I disabled the sound card though and ran a game with just the 6800 on that IRQ and the crash still came up. So, I'll have to live with my old 9700 Pro for a week or two

Thanks for the help though, really appreciate it.

Good luck with that. I hope it was a faulty card, and you don't have the same problems with the new one!!

But if you do, just post back. We're here.

i had the same freezing problem with my 6800 ultra.

find out if it came from the factory over clocked. if so this might work for you.

" ok here is the fix. your card is probably over clocked. so you need to get a tweek program to turn down the settings.

i used Riva Tuner. get it here
http://downloads.guru3d.com/download.php?id=13

after you install the program go to \main\driver settings(forceware detected)\customize\system settings

bring the Core clock down to 400 and the memory clock down to 1100.

i havent had a freeze up since."

I'm having the same exact problem. It only happens on FEAR for me. I'm going to try the fix listed above, but tell us how the RMA goes when you get your card back.

Surfsand- those are ultra base specs, The GT runs at 350/1000. They normally overclock very easily to 400/1100 which is where my two XFX 6800GT's run during games.

Hypnosan-Those temps, IMHO, are way too high. My GPU's start artifacting in th low 80's and by the mid to high 80's you can expect a crash. I overclock both may 6800GT's to 400/1100 speeds but had to replace the stck coolers with the Arctic Cooling NV5's to do it. SLI rigs are big heat generators though and with and O/Ced CPU and two O/Ced GPU's running at full bore the case got hot.

I rarely see temps above 72c now and that occurs only when the ambient room temp approaches 78f-80f (due mostly to PC waste heat).

I am going to guess that you will continue to have problems after receiveing the new cards until you get your GPU temps below 75c under load.

I have a similar problem. My video card is a Leadtek Winfast 6800GT TDH.

It will run Windows XP and small stuff (console apps, internet explorer) run just fine, but when I try to play a game it may work for a minute or two and then restarts the computer.

My computer specs:
Thermaltake 500W Purepower PS
Intel D915GAV motherboard
Intel P4 3.0GHz LGA775 processor
1 GB of Corsair Value Select PC3200 DDR 400 RAM
Creative Labs Sound Blaster Audigy 2 ZS

I had problems before when I built this rig. The original was using a gigabyte board and the PS died after 5 days. Got the rig home from the computer repair center and upon 2 hours of use the PS died again. It turned out to be a short in the PCI Express slot that was frying the PS's, but did nothing to the graphics card?

So I sent the Thermaltake PS back and received it yesterday, gigabyte board was RMA'd back to Newegg and I took the Intel board over a replacement gigabyte.

I know the PS is good. I know the memory is good. The motherboard has been flashed with the latest bios.

Is it time to RMA the video card?

Still waiting on the RMA, I'll look into the heating issues though. I just don't think that's my main problem as the issue just recently started, if it was the heat I probably would have noticed it earlier (especially during the Summer).

Just got the new card in, and it's just that...new. It seems to have a different design than the older card I sent in. I'm assuming that since my original card was purchased near the release of BFG's 6800 GT OC that they've made some improvements to the design of the card. Right now the card is idling around 55-57 C, which is about 10 degrees cooler than before. Going to run some games and see what I can get out of it in heavy gameplay, I'll be sure to keep you guys posted so if anybody else runs into this problem they have an idea of what to do.